Characterization of new bisphenol A metabolites produced by CD1 mice liver microsomes and S9 fractions

Article abstract of DOI:10.1021/jf049762u

Bisphenol A [2,2-bis(4-hydroxyphenyl)propane] (BPA) is a widely used industrial chemical resulting in occupational and consumer exposure. BPA possesses weak estrogenomimetic activity and can be cytotoxic, though the underlying mechanisms of its toxicity toward cells are not completely understood. The metabolism of BPA by CD1 mice liver microsomal and S9 fractions was investigated. Nine metabolites were isolated and characterized using HPLC and mass spectrometry. Many of these metabolites were characterized for the first time in mammals, namely isopropyl-hydroxyphenol (produced by the cleavage of BPA), a bisphenol A glutathione conjugate, glutathionyl-phenol, glutathionyl 4-isopropylphenol, and BPA dimers. Most of these metabolites apparently share a common metabolic pathway, for which considerable evidence supports the hypothesis of the production of a reactive intermediate, and also helps explain BPA cytotoxicity.

Full text of DOI:10.1021/jf049762u

J. Agric. Food Chem. 2004, 52, 4935−4942 4935
Characterization of New Bisphenol A Metabolites Produced by
CD1 Mice Liver Microsomes and S9 Fractions
JEAN PHILIPPE JAEG,* ELISABETH PERDU, LAURENCE DOLO,
LAURENT DEBRAUWER, JEAN-PIERRE CRAVEDI, AND DANIEL ZALKO
Institut National de la Recherche Agronomique, Unite´ Mixte de Recherche Xe´nobiotiques,
Toulouse, France
Bisphenol A [2,2-bis(4-hydroxyphenyl)propane] (BPA) is a widely used industrial chemical resulting
in occupational and consumer exposure. BPA possesses weak estrogenomimetic activity and can
be cytotoxic, though the underlying mechanisms of its toxicity toward cells are not completely
understood. The metabolism of BPA by CD1 mice liver microsomal and S9 fractions was investigated.
Nine metabolites were isolated and characterized using HPLC and mass spectrometry. Many of these
metabolites were characterized for the first time in mammals, namely isopropyl-hydroxyphenol
(produced by the cleavage of BPA), a bisphenol A glutathione conjugate, glutathionyl-phenol,
glutathionyl 4-isopropylphenol, and BPA dimers. Most of these metabolites apparently share a common
metabolic pathway, for which considerable evidence supports the hypothesis of the production of a
reactive intermediate, and also helps explain BPA cytotoxicity.
KEYWORDS: BPA; in vitro metabolism; mouse; endocrine disruptor
INTRODUCTION
attempting to assess its toxicity at the cellular level, as well as
when assessing its pharmacological properties.
Bisphenol A [2,2-bis(4-hydroxyphenyl)propane, CAS Reg-
istry No. 80-05-7] (BPA) is used as a monomer in the production
of epoxy-phenolic resins, polycarbonate plastics, and as an
antioxidant in PVC plastics. BPA-based polymers are widely
used in industry for the packaging of food (lacquered coating
of food cans) and drinks (polycarbonate bottles). These polymers
can be a source of contamination of human foodstuff (1, 2). It
has been demonstrated that BPA monomers can be released from
packaging material and then migrate into foods and beverages
(3, 4). The rate of migration is enhanced by treatments such as
heat processing (5, 6). Human exposure to BPA is expected to
be relatively low but nevertheless daily, and the presence of
BPA in human biological fluids has already been demonstrated
(7-9).
BPA can induce oxidative stress (10, 11) and can be
chemically converted into reactive intermediates, presumably
via a cytochrome P450 (CYP) pathway (12, 13). These
intermediates have been shown to bind DNA in vitro and in
vivo (12, 13). BPA was also reported to be a synthetic estrogenic
agent (14). BPA estrogenic activity is weak compared with 17-
â-oestradiol, but this chemical has been shown to efficiently
bind and activate human estrogen receptors (5, 15). According
to Yoshihara (16), the estrogenicity of BPA is enhanced through
its biotransformation by rat liver S9 fractions. Thus, understand-
ing the metabolism of BPA may be of great importance when
Several studies on the biotransformations of BPA have been
carried out in vivo (17-21) and in vitro (16, 22-27). Bisphenol
A glucuronide was characterized as the major metabolite of
BPA. Other metabolites such as BPA sulfate conjugate, BPA
diglucuronide, 5-hydroxybisphenol A, and the corresponding
sulfate conjugate were also identified. Recently, an extensive
study of BPA metabolism in pregnant CD1 mice demonstrated
the formation of additional metabolites (28) showing that the
metabolism of BPA was even more complex. To elucidate the
mechanism of its toxicity, more extensive knowledge of BPA
biotransformation is essential.
In the present study, the biotransformation of radiolabeled
BPA was investigated in vitro using liver microsomes and S9
systems prepared from female and male CD1 mice. Metabolites
were separated, isolated by high-performance liquid chroma-
tography (HPLC), and analyzed by mass spectrometry (MS),
MSn, and/or LC-MS.
MATERIALS AND METHODS
Chemicals. 3H-labeled bisphenol A [([3H]BPA); 2,2-bis(4-hydroxy-
phenyl)propane], with a specific activity of 185 GBq/mmol, was
purchased from Moravek Biochemicals (Mercury Lane, Brea, CA) and
stored in ethanol at -20 °C. In these conditions only limited tritium
exchange occurs (< 3%/6 months). Prior to the experiments, the solution
was evaporated to dryness under a nitrogen stream and was redissolved
in ethanol. Its purity was verified by radio-HPLC and was greater than
99.9%.
* Address correspondence to this author at INRA, UMR 1089 Xe´no-
biotiques, 180 Chemin de Tournefeuille, B.P. 3, 31931 Toulouse cedex 9,
France [telephone (33) 561 285 485; fax (33) 561 285 244; e-mail
jp.jaeg@envt.fr].
Methanol and acetonitrile were purchased from Scharlau Chemie
S.A. (Barcelona, Spain); ethanol and acetic acid were purchased from
10.1021/jf049762u CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/18/2004

4936 J. Agric. Food Chem., Vol. 52, No. 15, 2004
Jaeg et al.
Merck (Briare-Le-Canal, France). Unlabeled BPA and all biochemical
reagents were purchased from Sigma Aldrich (L’Isle d’Abeau Chesnes,
France).
infused at a flow rate of 3 µL/min into the ionization source. Hydrogen/
Deuterium exchange experiments were carried out using CH3OD and
D2O as solvents, using the same operating conditions.
Animals. A total of 10 female and 10 male CD1 mice (16-20 g)
were obtained from Harlan France SARL (Gannat, France). The animals
were housed at 21 ( 1 °C with a relative humidity of 60-70% under
a 12 h light/dark cycle. They were fed a standard commercial diet
(UAR, Villemoisson-Sur-Orge, France) and were allowed free access
to water.
Preparation of Mice Liver Subcellular Fractions. Mice were killed
by cervical dislocation followed by exsanguination and their livers were
perfused immediately using 0.1 M sodium phosphate buffer (pH 7.4).
Livers from two animals were pooled and weighed after homogenization
by a Potter-Elvehjem glass Teflon homogenizer in 4 volumes/g of ice-
cold phosphate buffer. S9 fractions were obtained by centrifugation at
10000g for 20 min at 5 °C. Cytosol and microsomal pellets were
obtained by centrifugation of S9 fractions at 105000g for 60 min.
Microsomes were resuspended with gentle homogenization in 1
volume/g of 0.1 M phosphate buffer containing 20% glycerol. S9
fractions, microsomes, and cytosols were stored at -80 °C until use.
The protein content of subcellular fractions was determined by the
method of Lowry (29).
Metabolism of [3H]BPA by Mice Liver S9 and Microsomes. In
vitro biotransformations of BPA were studied by incubating for 2 h at
37 °C under shaking radiolabeled [3H]BPA (8400 Bq) fortified with
unlabeled BPA at different concentrations (20, 50, 100, 200, and 500
µM), with 2 mg of microsomal protein or 8 mg of S9 protein (100 mg
liver equivalent) in a final volume of 1 mL of 0.1 M sodium/potassium
buffer, 5 mM MgCl2 (pH 7.4) containing an NADPH generating system
(NADP 1.3 mM, glucose-6-phosphate 5 mM, glucose-6-phosphate
dehydrogenase 2 IU). All incubations were carried out in triplicate.
Blank samples without an NADPH generating system were used as
controls for non-NADPH-mediated BPA transformation.
HPLC Analysis and Metabolite Quantification. The HPLC system
consisted of a Zorbax C18 column (250 × 4.6 mm, 5 µm) (Interchim,
Montluc¸on, France) coupled to a Kromasil C18 guard precolumn (18
× 4.6 mm, 5 µm) (Interchim). Mobile phases were A: ammonium
acetate buffer (20 mM, pH 3.5)/acetonitrile (95:5; v/v) and B:
ammonium acetate buffer (20 mM, pH 3.5)/acetonitrile (10:90; v/v).
The flow rate was 1 mL/min and the temperature was maintained at
30 °C. A 5-step gradient was used as follows: 0-4 min 100% A; 4-6
min linear gradient from 100% A to A:B 85:15 v/v, 6-16 min A:B
85:15 v/v; 16-18 min linear gradient from 15% B to 25% B; 18-28
min A:B 75:25 v/v; 28-30 min linear gradient from 25% B to 30% B;
30-37 min A:B 70:30 v/v; 37-39 min linear gradient from 30% B to
70% B; 39-50 min A:B 30:70 v/v; 50-52 min linear gradient leading
to 100% B; 52-62 min 100% B.
[3H]-BPA and related metabolites were monitored by online
radioactivity detection (Flo-one â A500, Packard) and quantified by
integrating the area under detected peaks.
Metabolite Isolation. Additional incubations of 50 µM BPA (1 700
Bq) were carried out in the same conditions as described above with
liver microsomes and liver S9 fractions, respectively, in order to obtain
enough material for metabolite production.
After the addition of 3 volumes of methanol and centrifugation,
supernatants were individually concentrated under a nitrogen stream
and the metabolites were separated by HPLC using the analytical system
described above. The collected fractions corresponding to the different
BPA metabolites were individually evaporated to dryness under a
nitrogen stream, then dissolved in 5 mL ammonium acetate buffer (50
mM, pH 3.2) and extracted using an Oasis HLB 0.2 g LP cartridge
(Waters, Milford, MA) previously washed with methanol (5 mL) and
equilibrated with the ammonium acetate buffer (5 mL). Elution was
performed successively with 5 mL of the buffer and with 5 mL of
methanol (3 times). The methanol eluate was evaporated to dryness
under a nitrogen stream and subjected to structural analysis.
Reactions were stopped by adding 3 volumes of methanol and
samples were centrifuged for 10 min at 6000g. Supernatant samples
(500 µL) were concentrated under a nitrogen stream and analyzed by
HPLC coupled to online radioactivity detection.
Combined High Performance Liquid Chromatography-Mass
Spectrometry (LC-MS). For LC-MS experiments, the LCQ mass
spectrometer was fitted with an atmospheric pressure chemical ioniza-
tion (APCI) source operated in the negative ion mode. Liquid
chromatography was carried out using a Thermo Separation P4000
pump (Thermo Finnigan, Les Ullis, France) and a Thermo Separation
AS3000 autosampler was used for sample injection. The LC column
used was a Zorbax SB-C18 column (250 × 4.6 mn, 5 µm) from Agilent
Technologies (Les Ullis, France). Separation was achieved at a flow
rate of 1 mL/min using the following gradient elution: 100% A from
0 to 4 min, 100% A to 100% B from 4 to 39 min, and then 100% B
from 39 to 50 min, with A, water-acetonitrile (95:5; v/v) and B, water-
acetonitrile (10:90, v/v).
Operating conditions of the APCI source were as follows: vaporizer
temperature, 450 °C, capillary temperature, 150 °C, capillary voltage,
-20 V. Ion isolation and collision conditions were adjusted separately
for each metabolite in order to gain maximum structural information.
Statistical Analysis. Mean values were compared by two-tailed
Student’s t test and the differences were considered to be significant
when P < 0.05.
â Glucuronidase Hydrolysis Assays. For the confirmation of BPA
glucuronide structure, extracts of S9 incubation media were incubated
in 0.1 M sodium acetate buffer (pH 5) for 4 h at 37 °C, with 500 IU
of bovine liver â glucuronidase (controlled without sulfatase activity
using 4-nitrophenyl sulfate as substrate) prior to analysis by radio-
HPLC.
In Vitro Synthesis of [3H]BPA Glucuronide and [3H]BPA Sulfate.
[3H]BPA glucuronidation was performed by incubating 2 mg microso-
mal protein with labeled BPA (100 µM; 8400 Bq) and UDPGA (1
mM) in 1 mL 0.5 M Tris buffer (pH 7.4) containing 10 mM MgCl2,
for 2 h at 37 °C.
[3H]BPA sulfate was produced by incubating 2 mg of cytosolic
protein with [3H]BPA (100 µM; 400 Bq) and 3′-phosphoadenosine-5-
phosphosulfate (PAPS) 40 µM in a final volume of 1 mL of Tris buffer
50 mM (pH 7.5), 5 mM MgCl2, for 2 h at 37 °C.
Synthesized [3H]BPA glucuronide and BPA sulfate were analyzed
by HPLC, characterized by MS and used as standards.
Apparatus. Samples were analyzed by HPLC on an HP 1050
apparatus (Hewlett-Packard, Waldbronn, Germany) equipped with a
Rheodyne model 7125 injector (Rheodyne, Cotati, CA) connected for
radioactivity detection to a Radiomatic flo-one/â A 500 instrument
(Radiomatic, La-Queue-Lez-Yvelines, France) using Flow-scint II as
the scintillation cocktail (Packard Instrument Co., Rungis, France) in
order to establish metabolic profiles, or to an HP 1050 UV detector
set at 270 nm and a Gilson model 201/202 fraction collector (Gilson
France, Villiers-Le-Bel, France) for metabolite isolation.
Mass spectrometry analyses were performed using a Finnigan LCQ
ion trap mass spectrometer (Thermo Finnigan, Les Ullis, France)
equipped with an electrospray ionization (ESI) source. Typical condi-
tions for recording mass spectra were as follows: spray voltage, 4 kV;
capillary voltage, -22 V; capillary temperature, 200 °C. All mass
spectra were acquired using automatic gain control conditions and the
helium buffer gas was used as the collision gas for MSn studies. Sample
solutions (typically 1 ng/µL in methanol-water (50:50, v/v) were
RESULTS
Metabolism of BPA by Mice Liver Microsomes. A
representative radio-chromatogram corresponding to the analysis
of microsomal incubation supernatants is given in Figure 1,
showing nine different radioactive peaks.
The radioactive peak corresponding to unchanged BPA (RT:
42.80 min) accounted for 60% to 80% of the radioactivity
incubated. In our HPLC system, seven compounds eluted before
BPA (RT: 10.50, 12.9, 13.80, 20.80, 22.60, 23.40, and 36.20
min, respectively), whereas two metabolites (RT: 45.00 and
46.30 min, respectively) exhibited a lower polarity than BPA.

Characterization of New BPA Metabolites
J. Agric. Food Chem., Vol. 52, No. 15, 2004 4937
m/z 440 (Figure 2c) and was fragmented in MS2 to m/z 306
(depronated glutathione). In MS3, this fragment produced an
[M - H]- ion at m/z 272 (Table 1). Thus, this metabolite was
identified as glutathionyl 4-isopropylphenol.
The three metabolites, with respective retention times of 13.8,
22.6, and 36.2 min were identified as isopropyl-hydroxyphenol,
2,2-bis(4-hydroxyphenyl)-1-propanol, and 5-hydroxybisphenol
A (BPA catechol) on the basis of MS analyses (Table 1).
The metabolite with a RT of 20.80 min could not be identified
by MS analysis because of insufficient material, but this
metabolite coeluted with authentic phenol.
The metabolites eluting in our chromatographic system after
unchanged BPA (peaks 8 and 9, Figure 1), collected and
analyzed by LC/ESI-MS, gave similar results (Figure 3).
In both cases, two [M - H]- quasimolecular ions were
observed: a major one at m/z 453 and another at m/z 451. The
fragmentation of m/z 453 ion in MS2 produced fragment ions
at m/z 359 and 319 (see Figure 3). In MS3, the m/z 359 ion
gave a fragment ion at m/z 265 (corresponding to the loss of
the phenol group), and the m/z 319 ion gave a major fragment
ion at m/z 225 (also corresponding to the loss of a phenol group).
These results support the hypothesis of the formation of BPA
dimers by linkage of two BPA molecules, thus possessing a
molecular weight of 454 and detected as [M - H]- quasi-
molecular ions at m/z 453. These metabolites are fragmented
in MS2, either by the loss of an isopropyl-phenol moiety (m/z
319), or by the loss of a phenol moiety (m/z 359), as shown in
Figure 4.
Figure 1. Representative HPLC-radiochromatogram obtained after incuba-
tion of BPA (100 µM, 2 h, 37 ˚C) with CDI mice liver microsomes.
Both resulting fragment ions can be further fragmented in
MS3, as detailed above. Further structural characterization of
metabolites 8 and 9 by NMR analysis was not possible due to
the limited amount of metabolites available. However, these
metabolites were analyzed by APCI-MS, and similar results
were obtained (data not shown). When infused into the ESI
source in CH3OD/D2O, the quasimolecular species was detected
at m/z 456, corresponding to an [Md4-D]- ion, thus indicating
the presence of 4 exchangeable hydrogen atoms in the dimeric
structure. These results indicate that for metabolite 8 as well as
for metabolite 9, the dimerization of BPA does not involve the
phenolic groups via ether bonds. Consequently, metabolites 8
and 9 should correspond to the two possible isomers resulting
from the carbon-carbon linkage of two molecules of BPA, as
proposed in Figure 4. When metabolites 8 and 9 were analyzed
by ESI-MS, we observed an additional [M - H]- quasi-
molecular ion at m/z 451, whose fragmentation in MS2 led to a
major fragment ion at m/z 436 (corresponding to the loss of a
methyl radical), m/z 357 (corresponding to the loss of a phenol
group) and to two minor fragment ions at m/z 225 (likely due
to the cleavage of the two ether bonds of such dimeric parent
ions). Thus, m/z 451 was tentatively identified as a cyclic dimer
of BPA, which would be coeluted from the HPLC column with
metabolites 8 and 9. Three isomers could theoretically be
formed, one of which is shown in Figure 4.
2
Figure 2. ESI-MS analysis of (a) glutathione BPA conjugate (peak 6),
(b) glutathionyl-phenol (peak 1), and (c) glutathionyl 4-isopropylphenol
(peak 2).
Polar metabolites (peaks 1-7) were isolated as described in
the materials and methods section and their structures were
investigated by mass spectrometry using electrospray in the
negative ionization mode (Table 1). The ESI analysis of the
metabolite corresponding to peak 6 gave an [M - H]- ion at
m/z 532 (Figure 2a).
The fragmentation of this ion (MS2) led to the formation of
the m/z 272 ion resulting from the cleavage of the glutathione
thioether bond, and the m/z 403 ion resulting from the loss of
glutamic acid. Thus, this metabolite was characterized as a
glutathione conjugate of bisphenol A. The mass spectrometric
analysis of the metabolite corresponding to peak 1 (Figure 2b)
gave an [M - H]- ion at m/z 398, which yielded m/z 272 and
m/z 269 fragment ions in MS2. These fragments corresponded
to the same decomposition pathways as observed for peak 6,
and thus metabolite 1 was identified as glutathionyl phenol. The
metabolite obtained from peak 2 exhibited an [M - H]- ion at
When BPA was incubated with female and male mice liver
microsomes in triplicate for 2 h at 37 °C in the presence of
NADPH, no sex-related difference was observed. In addition
no metabolites were formed when an NADPH generating system
was omitted from the incubation mixtures.
Metabolite formation rates were investigated over a substrate
concentration range of 20-500 µM. The formation of the two
major hydroxylated metabolites, namely isopropyl-hydroxy-
phenol and 5-hydroxybisphenol A (respectively peaks 3 and 7,
Figure 1) as a function of BPA concentration is shown in
Figure 5 and followed Michaelis-Menten kinetics (the apparent

4938 J. Agric. Food Chem., Vol. 52, No. 15, 2004
Jaeg et al.
Table 1. Mass Spectrometric Data and Structure of BPA Metabolites 1−7
Km ) 84.7 ( 31.7 µM and the apparent Vm ) 1.96 ( 0.4 nmol/
mg/h for isopropyl-hydroxyphenol, the apparent Km ) 837 (
310 µM and the apparent Vm ) 3.27 ( 0.67 nmol/mg/h for
5-hydroxybisphenol A). Whereas the production of the two
metabolites eluted after BPA (respectively peaks 8 and 9, Figure
1) did not follow Michaelis-Menten kinetics but was linear up
to 500 µM BPA (Figure 6).
different S9 fractions. No significant difference was observed
between females and males (data not shown). Radio-HPLC
analyses of the supernatants of S9 BPA incubations (Figure 7)
showed high metabolization rates, compared with microsomes;
unchanged BPA accounting for 10 to 50% of the total
radioactivity. Metabolites eluted at respective RT of 10.5, 12.9,
13.8, 23.4, and 24.2 min were detected.
Metabolism of BPA by Mice Liver S9 Fractions. BPA
incubations (20-500 µM) were carried out with S9 fractions
of female and male mice livers (2 h, 37 °C) in the presence of
NADPH. Each incubation was carried out in triplicate, with three
All metabolites were collected and analyzed by mass spec-
trometry using negative electrospray ionization mode, some of
them were found to be similar to metabolites already character-
ized from the supernatants of microsomal incubations: gluta-

Characterization of New BPA Metabolites
J. Agric. Food Chem., Vol. 52, No. 15, 2004 4939
2
Figure 3. LC/ESI-MS analysis of BPA dimers (peaks 8 and 9).
Figure 4. Proposed molecular structures for BPA dimers: (a) linear dimer
isomer (ortho/ortho isomer; other possible structures; meta/ortho or meta/
meta); (b) cyclic dimer (ortho/ortho isomer; other possible structures; meta/
ortho or meta/meta).
Figure 5. Microsomal production of isopropyl-hydroxyphenol ([) and
5-hydroxylbisphenol A (9) as a function of substrate concentration. Data
are expressed as mean ± SD (n ) 6).
thionyl-phenol (RT 10.5 min), glutathionyl 4-isopropylphenol
(RT 12.9 min), glutathione bisphenol A conjugate (RT 23.4) and
the two dimers (RT 45 and 46.30 min, respectively). The
structure of BPA glucuronide (RT 24.20 min) was confirmed
by cochromatography with the authentic standard synthesized
in vitro and by mass spectrometry. Whatever the concentration
of BPA incubated, the detected amount of BPA-glucuronide
was always lower than the detected amount of glutathione BPA
conjugate (RT 23.4) (Figure 8).
NADPH was required for glutathione conjugate synthesis and
when 100 µM glutathione reduced was added to the incubation

4940 J. Agric. Food Chem., Vol. 52, No. 15, 2004
Jaeg et al.
events and the possible associated health hazards for humans
due to chronic exposure to BPA, it is necessary to investigate
the metabolic fate of this xenobiotic, as well as the conditions
in which human exposure occurs. To explore BPA metabolism,
we used two complementary in vitro systems: CD1 mice liver
microsomes and S9 fractions.
No major differences in the metabolism of BPA were
observed between females and males, either using microsomes
or S9 fractions. These observations are consistent with the results
obtained via in vitro studies in humans and reported by Elsby
(26), as well as with in vivo studies carried out in humans (21).
However, gender-related differences in the metabolic pathways
of BPA have previously been demonstrated in cynomolgus
monkeys (20).
Figure 6. Microsomal production of BPA dimers [([) peak 8; (9) peak
9] as a function of substrate concentration. Data are expressed as mean
± SD (n ) 6).
We isolated and characterized nine metabolites, including two
metabolites exhibiting, in our system, a lower polarity than BPA.
Some of them, BPA glucuronide and 5-hydroxybisphenol A,
had already been described in different in vitro studies, but many
metabolites are characterized here for the first time in mam-
mals: glutathione conjugates, metabolites produced by the
cleavage of BPA, and BPA dimers.
Glutathione BPA conjugates were not produced without
NADPH, even when glutathione was added to microsomal and
S9 incubations, suggesting the formation of a reactive interme-
diate. Two hypotheses explaining the formation of glutathione
conjugates can be proposed. BPA could first be oxidized into
an arene epoxide and subsequently conjugated to glutathione,
this reaction being catalyzed by a glutathione S transferase
(GST). Classically, when glutathione adds to an arene epoxide,
the net result is a dihydroxyl-S-glutathionyl metabolite. How-
ever, in our study, we characterized a glutathionyl BPA that
could be produced by the rearomatization of the ring and the
loss of H2O (30, 31). The glutathione conjugate metabolites are
quantitatively minor ones when BPA is incubated with CD1
liver microsomes, each peak accounting for less than 2% of
the total radioactivity in 20 µM BPA incubations. However,
they are quantitatively major metabolites when BPA is incubated
with CD1 liver S9 fractions, each representing 15-20% of the
total amount of radioactivity detected. These microsomal
glutathione conjugate metabolites may be produced either by
microsomal GST (mGST), with poor efficiency, or by the
cytosolic GST (cGST) present in microsomal fractions (32).
Microsomal and S9 fractions contain different GSTs and mGST
that are distinct from cGST, are present in small amounts and
possess specific substrates (33, 34).
Figure 7. Representative HPLC-radiochromatogram obtained after incuba-
tion of BPA (100 µM, 2 h, 37 ˚C) with CD1 mice liver S9 fraction.
The second mechanism that can be hypothesized to explain
the formation of these glutathione conjugates is a first-step CYP-
mediated oxidation of BPA into 5-hydroxybisphenol A (catechol
BPA, identified in our incubations), this intermediate possibly
being converted into bisphenol-o-quinone. o-Quinones are
electrophile-reactive species that can covalently bind to nucleo-
philic sites (35) and glutathione, a nucleophile compound, may
react directly with bisphenol-o-quinone to produce glutathione
BPA conjugates. This hypothesis is supported by the fact that
S9 fractions, which contain more glutathione than microsomes,
produced more glutathione conjugates, and that BPA conjugate
production increased when glutathione was added to the
incubation medium (data not shown). In addition, Atkinson and
Roy (12, 13) have already provided evidence that bisphenol-
o-quinone can be formed in vitro and that this reactive
intermediate is able to covalently bind to DNA.
Figure 8. S9 fraction production of glutathionyl BPA ([) and BPA
glucuronide (9) as a function of substrate concentration. Data are
expressed as mean ± SD (n ) 6).
medium, production of glutathione conjugates increased whereas
production of dimers did not.
When PAPS was added to S9 fraction incubations, BPA
sulfate was detected at an RT of 27.50 min (data not shown),
but this metabolite was never observed in vitro without PAPS.
DISCUSSION
BPA possesses an estrogenic activity (15, 16), can induce
oxidative stress (10, 11), and produces nucleotide adducts (12,
13). However, the molecular mechanisms of BPA toxicity are
not fully understood. To better understand these toxicological
Metabolites resulting from the cleavage of BPA: isopropyl-
hydroxyphenol, glutathionyl-phenol, and glutathionyl 4-isopro-
pylphenol were produced in vitro but the biochemical pathway

Characterization of New BPA Metabolites
J. Agric. Food Chem., Vol. 52, No. 15, 2004 4941
explaining the formation of these metabolites is not fully
understood. No other studies have, to date, reported the
occurrence of these metabolites in mammals. However a Gram-
negative aerobic bacterium (strain MV1) has been shown to
degrade BPA into many metabolites including 4-hydroxybenz-
aldehyde, 4-hydroxybenzoic acid, and 4-hydroxyacetophenone
(36). None of these compounds was characterized in our study.
Hirano (37) showed that a manganese peroxidase produced by
a fungus (Pleurotus ostreatus) can degrade BPA into phenol,
isopropyl phenol, and isopropenyl phenol, and that H2O2 is
necessary to cleave the C-C bond between the central carbon
and one of the aromatic rings of the molecule. Our experiments
indicate that the cleavage of BPA only occurs in the presence
of NADPH. In addition, if H2O2 and soybean peroxidase are
added to either microsomal or S9 incubations, an increase in
the production of metabolites resulting from the cleavage of
BPA is observed, while 5-hydroxybisphenol A (catechol BPA)
is no longer detected (data not shown). Consequently, BPA
cleavage may be achieved by a peroxidase, after a first-step
CYP oxidation of BPA into catechol BPA.
We identified several new BPA metabolites: glutathione BPA
conjugates, metabolites resulting from BPA cleavage, and
dimers. A reactive intermediate is necessary to produce these
different metabolites. There is indirect evidence suggesting that
the bisphenol A-ortho-quinone, produced from 5-hydroxy
bisphenol A (catechol), may be the reactive intermediate.
Additional experiments are necessary to explore the mechanisms
involved in these pathways and to identify the enzymes that
play a role in the formation of these intermediates.
ABBREVIATIONS USED
APCI, atmospheric pressure chemical ionization; BPA,
bisphenol A; CYP, cytochrome P 450; ESI, electrospray
ionization; GST, glutathione S transferase; HPLC, high-
performance liquid chromatography; LC-MS, combined high-
performance liquid chromatography/mass spectrometry; MS,
mass spectrometry; NADPH, nicotinamide adenine dinucleotide
phosphate reduced; PAPS, 3′-phosphoadenosine-5-phosphosul-
fate; RT, retention time.
The characterized BPA dimers, which exhibited lower polarity
than BPA in our HPLC system, were produced in higher
quantities with liver microsomes than when incubated with S9
fractions. An LC-MS system was developed in order to further
study these metabolites and led us to the tentative characteriza-
tion of aliphatic dimers (two isomers) and of a cyclic dimer.
The increase in the amount of dimers was proportional to the
concentration of BPA incubated and even at 500 µM BPA,
saturation was not reached. In addition, these dimers were not
formed in the absence of NADPH. We hypothesize that BPA
dimers formation may be a two-step metabolic pathway initiated
by the enzymatic oxidation of BPA into an unidentified reactive
intermediate, followed by a nonenzymatic reaction between the
intermediate reactive compound and unchanged BPA. A poly-
phenoloxidase may be implicated in dimer production, as
suggested by a study by Guyot (38), demonstrating that
polyphenoloxidase catalyses the dimerization of catechin.
ACKNOWLEDGMENT
We thank Florence Blas y Estrada and Raymond Gazel for
animal care.
LITERATURE CITED
(1) Feldman, D. Krishnan, A. Estrogens in unexpected places:
possible implications for researchers and consumers. EnViron.
Health. Perspect. 1995, 103 (Suppl. 7), 129-133.
(2) Goodson, A.; Summerfield, W.; Cooper, I. Survey of bisphenol
A and bisphenol F in canned foods. Food Addit. Contam. 2002,
19, 796-802.
(3) Brotons, J. A.; Olea-Serrano, M. F.; Villalobos, M.; Pedraza,
V.; Olea, N. Xenoestrogens released from lacquer coatings in
food cans. EnViron. Health. Perspect. 1995, 103, 608-612.
(4) Biles, J. E.; McNeal, T. P.; Begley, T. H. Determination of
bisphenol A migrating from epoxy can coatings to infant formula
liquid concentrates. J. Agric. Food Chem. 1997, 45, 4697-
4700.
(5) Krishnan, A. V.; Stathis, P.; Permuth, S. F.; Tokes, L.; Feldman,
D. Bisphenol A: an estrogenic substance is released from
polycarbonate flasks during autoclaving. Endocrinology 1993,
132, 2279-2286.
(6) Munguia-Lopez, E.; M.Soto-Valdez, H. Effect of heat processing
and storage time on migration of bisphenol A (BPA) and
bisphenol A-diglycidyl ether (BADGE) to aqueous food simulant
from Mexican can coatings. J. Agric. Food Chem. 2001, 49,
3666-3671.
(7) Schonfelder, G.; Wittfoht, W.; Hopp, H.; Talsness, C. E.; Paul,
M.; Chahoud, I. Parent bisphenol A accumulation in the human
maternal-fetal-placental unit. EnViron. Health Perspect. 2002,
110, A703-A707.
(8) Yamada, H.; Furuta, I.; Kato, E. H.; Kataoka, S.; Usuki, Y.;
Kobashi, G.; Sata, F.; Kishi, R.; Fujimoto, S. Maternal serum
and amniotic fluid bisphenol A concentrations in the early second
trimester. Reprod. Toxicol. 2002, 16, 735-739.
(9) Matsumoto, A.; Kunugita, N.; Kitagawa, K.; Isse, T.; Oyama,
T.; Foureman, G. L.; Morita, M.; Kawamoto, T. Bisphenol A
levels in human urine. EnViron. Health Perspect. 2003, 111,
101-104.
(10) Bindhumol, V.; Chitra, K. C.; Mathur, P. P. Bisphenol A induces
reactive oxygen species generation in the liver of male rats.
Toxicology 2003, 188, 117-124.
Snyder (19), studying the metabolism of BPA by HepG2 cells,
observed an HPLC peak corresponding to a compound with a
low polarity, which however could not be characterized.
Metabolites exhibiting a similar polarity were described by
Yoshihara (16), when incubating BPA with rat liver S9 fractions,
but not with microsomal or cytosolic incubations alone. These
metabolites were not characterized, but Yoshihara et al. (26)
found that one of them exhibited higher estrogenicity than BPA.
Several BPA metabolites have already been tested to character-
ize their estrogenic potency: BPA glucuronide was shown to
be far less estrogenic than BPA; 5-hydroxy BPA was also
demonstrated to be a weak estrogenomimetic compound (25,
26), as was BPA sulfate (39). None of these metabolites is the
estrogenic compound isolated by Yoshihara (16), which, on the
basis of its weak polarity, might have been one of the dimers
we characterized. However, Yoshihara (16) did not observe the
formation of this metabolite in BPA incubations with mi-
crosomes, while, in our study, microsomes produced more
dimers than S9 fractions. Based on the mass spectral data
provided by these authors, its appears that the nonpolar
metabolite they isolated was possibly a dimer of propenylphenol.
Thus, the in vitro metabolization of BPA may lead to the
formation of both BPA dimers (metabolites 8 and 9) and BPA
metabolites (isopropylphenol and/or isopropyl-hydroxyphenol)
dimers. In this study, the amount of BPA dimers we were able
to isolate was too small to test their estrogenic activity.
Additional studies will be necessary to clarify this point.
(11) Chitra, K. C.; Latchoumycandane, C.; Mathur, P. P. Induction
of oxidative stress by bisphenol A in the epididymal sperm of
rats. Toxicology 2003, 185, 119-127.

4942 J. Agric. Food Chem., Vol. 52, No. 15, 2004
Jaeg et al.
(12) Atkinson, A.; Roy, D. In Vitro conversion of environmental
estrogenic chemical bisphenol A to DNA binding metabolite(s).
Biochem. Biophys. Res. Commun. 1995, 210, 424-433.
(13) Atkinson, A.; Roy, D. In vivo DNA adduct formation by
bisphenol A. EnViron. Mol. Mutagen. 1995, 26, 60-66.
(14) Dodds, E. C.; Lawson, W. Synthetic oestrogenic agents without
the phenanthrene nucleus. Nature 1936, 137, 996.
(15) Gaido, K. W.; Leonard, L. S.; Lovell, S.; Gould, J. C.; Babai,
D.; Portier, C. J.; McDonnell, D. P. Evaluation of chemicals with
endocrine modulating activity in a yeast-based steroid hormone
receptor gene transcription assay. Toxicol. Appl. Pharmacol.
1997, 143, 205-212.
(16) Yoshihara, S.; Makishima, M.; Suzuki, N.; Ohta, S. Metabolic
activation of bisphenol A by rat liver S9 fraction. Toxicol. Sci.
2001, 62, 221-227.
(17) Knaak, J. B.; Sullivan, L. J. Metabolism of bisphenol A in the
rat. Toxicol. Appl. Pharmacol. 1966, 8, 175-184.
(18) Pottenger, L. H.; Domoradzki, J. Y.; Markham, D. A.; Hansen,
S. C.; Cagen, S. Z.; Waechter, J. M., Jr. The relative bioavail-
ability and metabolism of bisphenol A in rats is dependent upon
the route of administration. Toxicol. Sci. 2000, 54, 3-18.
(19) Snyder, R. W.; Maness, S. C.; Gaido, K. W.; Welsch, F.; Sumner,
S. C.; Fennell, T. R. Metabolism and disposition of bisphenol
A in female rats. Toxicol. Appl. Pharmacol. 2000, 168, 225-
234.
(20) Kurebayashi, H.; Harada, R.; Stewart, R. K.; Numata, H.; Ohno,
Y. Disposition of a low dose of bisphenol a in male and female
cynomolgus monkeys. Toxicol. Sci. 2002, 68, 32-42.
(21) Volkel, W.; Colnot, T.; Csanady, G. A.; Filser, J. G.; Dekant,
W. Metabolism and kinetics of bisphenol A in humans at low
doses following oral administration. Chem. Res. Toxicol. 2002,
15, 1281-1287.
(22) Yokota, H.; Iwano, H.; Endo, M.; Kobayashi, T.; Inoue, H.;
Ikushiro, S.; Yuasa, A. Glucuronidation of the environmental
oestrogen bisphenol A by an isoform of UDP-glucuronosyl-
transferase, UGT2B1, in the rat liver. Biochem. J. 1999, 340
(Part 2), 405-409.
(23) Nakagawa, Y.; Tayama, S. Metabolism and cytotoxicity of
bisphenol A and other bisphenols in isolated rat hepatocytes.
Arch. Toxicol. 2000, 74, 99-105.
(28) Zalko, D.; Soto, A. M.; Dolo, L.; Dorio, C.; Rathahao, E.;
Debrauwer, L.; Faure, R.; Cravedi, J. P. Biotransformations of
Bisphenol A in a Mammalian Model: Answers and New
Questions Raised by Low-Dose Metabolic Fate Studies in
Pregnant CD1 Mice. EnViron. Health Perspect. 2003, 111, 309-
320.
(29) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J.
Protein measurement with the Folin phenol reagent. J. Biol.
Chem. 1951, 193, 265-275.
(30) Preston, B. D.; Miller, J. A.; Miller, E. C. Reactions of 2,2′,5,5′-
tetrachlorobiphenyl 3,4-oxide with methionine, cysteine and
glutathione in relation to the formation of methylthio-metabolites
of 2,2′,5,5′-tetrachlorobiphenyl in the rat and mouse. Chem.-
Biol. Interact. 1984, 50, 289-312.
(31) Yergey, J. A.; Trimble, L. A.; Silva, J.; Chauret, N.; Li, C.;
Therien, M.; Grimm, E.; Nicoll-Griffith, D. A. In vitro metabo-
lism of the COX-2 inhibitor DFU, including a novel glutathione
adduct rearomatization. Drug Metab. Dispos. 2001, 29, 638-
644.
(32) Cribb, A. E.; Griffin, G. P.; Spielberg, S. P.; Rushmore, T. H.;
Chauret, N.; Nicoll-Griffith, D. A. Increased cosedimentation
of cytosolic glutathione S-transferases with microsomal fractions
prepared from frozen rat liver. Drug Metab. Dispos. 1994, 22,
969-972.
(33) Lee, C. Y.; McKinney, J. D. Identity of microsomal glutathione
S-transferases. Mol. Cell. Biochem. 1982, 48, 91-96.
(34) Morgenstern, R.; Guthenberg, C.; Depierre, J. W. Microsomal
glutathione S-transferase. Purification, initial characterization and
demonstration that it is not identical to the cytosolic glutathione
S-transferases A, B and C. Eur. J. Biochem. 1982, 128, 243-
248.
(35) Bolton, J. L.; Trush, M. A.; Penning, T. M.; Dryhurst, G.; Monks,
T. J. Role of quinones in toxicology. Chem. Res. Toxicol. 2000,
13, 135-160.
(36) Spivack, J.; Leib, T. K.; Lobos, J. H. Novel pathway for bacterial
metabolism of bisphenol A. Rearrangements and stilbene cleav-
age in bisphenol A metabolism. J. Biol. Chem. 1984, 269, 7323-
7329.
(37) Hirano, T.; Honday, Y.; Watanabe, T.; Kuwahara, M. Degrada-
tion of bisphenol A by the lignin-degrading enzyme, manganese
peroxidase, produced by the white-rot basidiomycete, Pleurotus
ostreatus. Biosci., Biotechnol., Biochem. 2000, 64, 1958-1962.
(38) Guyot, S.; Vercauteren, J.; V. C. Structural determination of
colourless and yellow dimers resulting from (+)-catechin
coupling catalused by grape polyphenoloxidase. Phytochemistry
1996, 42, 1279-1288.
(39) Shimizu, M.; Ohta, K.; Matsumoto, Y.; Fukuoka, M.; Ohno, Y.;
Ozawa, S. Sulfation of bisphenol A abolished its estrogenicity
based on proliferation and gene expression in human breast
cancer MCF-7 cells. Toxicol. in Vitro 2002, 16, 549-546.
(24) Inoue, H.; Yokota, H.; Makino, T.; Yuasa, A.; Kato, S. Bisphenol
a glucuronide, a major metabolite in rat bile after liver perfusion.
Drug Metab. Dispos. 2001, 29, 1084-1087.
(25) Nakagawa, Y.; Suzuki, T. Metabolism of bisphenol A in isolated
rat hepatocytes and oestrogenic activity of a hydroxylated
metabolite in MCF-7 human breast cancer cells. Xenobiotica
2001, 31, 113-123.
(26) Elsby, R.; Maggs, J. L.; Ashby, J.; Paton, D.; Sumpter, J. P.;
Park, B. K. Assessment of the effects of metabolism on the
estrogenic activity of xenoestrogens: a two-stage approach
coupling human liver microsomes and a yeast estrogenicity assay.
J. Pharmacol. Exp. Ther. 2001, 296, 329-337.
(27) Pritchett, J. J.; Kuester, R. K.; Sipes, I. G. Metabolism of
bisphenol A in primary cultured hepatocytes from mice, rats,
and humans. Drug Metab. Dispos. 2002, 30, 1180-1185.
Received for review February 12, 2004. Revised manuscript received
May 13, 2004. Accepted May 13, 2004.
JF049762U