Biotransformation of vinclozolin in rat precision-cut liver slices: Comparison with in vivo metabolic pattern

Article abstract of DOI:10.1021/jf0728045

Vinclozolin is a dicarboxymide fungicide that presents antiandrogenic properties through its two hydrolysis products M1 and M2, which bind to the androgen receptor. Because of the lack of data on the biotransformation of vinclozolin, its metabolism was investigated in vitro in precision-cut rat liver slices and in vivo in male rat using [¹⁴C]-vinclozolin. Incubations were performed using different concentrations of substrate, and the kinetics of formation of the major metabolites were studied. Three male Wistar rats were fed by gavage with [¹⁴C]-VZ. Urine was collected for 24 h and analyzed by radio-HPLC for metabolic profiling. Metabolite identification was carried out on a LCQ ion trap mass spectrometer. In rat liver slices and in vivo, the major primary metabolite has been identified as 3′,5prime;-dichloro-2,3,4- trihydroxy-2-methylbutyranilide (M5) and was mainly present as glucurono-conjugates. M5 is produced by dihydroxylation of the vinyl group of M2. Other metabolites have been identified as 3-(3,5-dichlorophenyl)-5-methyl-5- (1,2-dihydroxyethyl)-1,3-oxazolidine-2,4-dione (M4), a dihydroxylated metabolite of vinclozolin, which undergoes further conjugation to glucuronic acid, and 2-[[(3,5-dichlorophenyl)-carbamoyl]oxy]-2-methyl-3,4-dihydroxy-butanoic acid (M6), a dihydroxylated metabolite of M1.

Full text of DOI:10.1021/jf0728045

4832 J. Agric. Food Chem. 2008, 56, 4832–4839
Biotransformation of Vinclozolin in Rat Precision-Cut
Liver Slices: Comparison with in Vivo Metabolic
Pattern
JULIAN BURSZTYKA, LAURENT DEBRAUWER, ELISABETH PERDU,
ISABELLE JOUANIN, JEAN-PHILIPPE JAEG, AND JEAN-PIERRE CRAVEDI*
UMR1089 Xe´nobiotiques, INRA, F-31931 Toulouse, France
Vinclozolin is a dicarboxymide fungicide that presents antiandrogenic properties through its two
hydrolysis products M1 and M2, which bind to the androgen receptor. Because of the lack of data on
the biotransformation of vinclozolin, its metabolism was investigated in vitro in precision-cut rat liver
slices and in vivo in male rat using [14C]-vinclozolin. Incubations were performed using different
concentrations of substrate, and the kinetics of formation of the major metabolites were studied.
Three male Wistar rats were fed by gavage with [14C]-VZ. Urine was collected for 24 h and analyzed
by radio-HPLC for metabolic profiling. Metabolite identification was carried out on a LCQ ion trap
mass spectrometer. In rat liver slices and in vivo, the major primary metabolite has been identified
as 3′,5′-dichloro-2,3,4-trihydroxy-2-methylbutyranilide (M5) and was mainly present as glucurono-
conjugates. M5 is produced by dihydroxylation of the vinyl group of M2. Other metabolites have been
identified as 3-(3,5-dichlorophenyl)-5-methyl-5-(1,2-dihydroxyethyl)-1,3-oxazolidine-2,4-dione (M4),
a dihydroxylated metabolite of vinclozolin, which undergoes further conjugation to glucuronic acid,
and 2-[[(3,5-dichlorophenyl)-carbamoyl]oxy]-2-methyl-3,4-dihydroxy-butanoic acid (M6), a dihydroxy-
lated metabolite of M1.
KEYWORDS: Vinclozolin; in vitro and in vivo metabolism; toxicology; endocrine disruptor
INTRODUCTION
observed that neonatal injection of vinclozolin (200 mg/kg/day)
demasculinized aggressive play behavior in male rats at 35 days
of age, indicating that it altered CNS sexual differentiation in
an antiandrogenic manner. Moreover, VZ can exert antiandro-
genic effects in fish and mollusks, where this pesticide has been
shown to cause a significant reduction of ejaculated sperm cells,
smaller testes, and disrupted male courtship behavior in guppies
(Poecilia reticulate) (9), and reduced penis length and accessory
male sex organs in two prosobranch snails species, Marisa
cornuarietis and Nucella lapillus (10). It has recently been
demonstrated that embryonic exposure to VZ at the time of
gonadal sex determination caused a transgenerational effect on
male rat fertility and testis function (11). Adult animals from
F1 to F4 generations developed a number of disease states or
tissue abnormalities including prostate disease, kidney disease,
immune system abnormalities, spermatogenesis abnormalities,
breast tumor development, and blood abnormalities including
hypercholesterolemia (12). The incidence or prevalence of these
transgenerational disease states was high, from 10% for tumors
to more than 50% for prostatic lesions in all generations,
indicating that transgenerational effects were transmitted through
epigenetic alterations (involving apparent permanent DNA
methylation) in the male germline (i.e., sperm) (13).
Vinclozolin[3-(3,5-dichlorophenyl)-5-methyl-5-vinyl-1,3-ox-
azolidine-2,4-dione] is a nonsystemic dicarboxymide fungicide
efficient in controlling plants or fruit diseases caused by Botrytis
spp., Monilia spp., or Sclerotinia spp. Vinclozolin (VZ)
undergoes spontaneous hydrolysis in aqueous media, giving
three products, 2-[[(3,5-dichlorophenyl)-carbamoyl]oxy]-2-
methyl-3-butenoic acid (M1), 3′,5′-dichloro-2-hydroxy-2-me-
thylbut-3-enanilide (M2), and 3,5-dichloroaniline (M3) (1). M1
and M2 are responsible for the antiandrogenic effects attributable
to vinclozolin since they competitively inhibit the binding of
androgens to human androgen receptor (hAR) and inhibit
androgen-induced gene expression in Vitro and in ViVo (2, 3).
When administered during sexual differentiation, this fungicide
has been shown to demasculinize and feminize the male rat
offspring who display female-like anogenital distance at birth,
retain nipples, a blind vaginal pouch, hypospadias, suprainguinal
ectopic testes, and small to absent sex accessory glands (4–6).
It has been demonstrated that VZ inhibits testosterone induced
growth of androgen-dependent tissues (ventral prostate, seminal
vesicles, and levator ani plus bulbocavernosus muscles) in a
dose-additive manner in the Hershberger assay using castrated
immature testosterone-treated rats (7). Hotchkiss et al. (8) also
Knowing the metabolic fate of VZ and its primary degradation
products M1 and M2 in rat is of considerable interest to better
understand the processes involved in the toxicological effects
* To whom correspondence should be addressed. Phone: +33 561
285 002. Fax: +33 561 285 244. E-mail: jcravedi@toulouse.inra.fr.
10.1021/jf0728045 CCC: $40.75  2008 American Chemical Society
Published on Web 06/04/2008

Biotransformation of Vinclozolin
J. Agric. Food Chem., Vol. 56, No. 12, 2008 4833
Figure 2. Typical radio-HPLC profile of [14C]-vinclozolin metabolites
present in rat liver slice culture medium (DMEM). Incubation and HPLC
conditions were as described in the Materials and Methods section.
Substrate final concentration was 10 µM. (a) Incubation without liver slices
(6 h); (b) incubation with liver slices (6 h).
Figure 1. Structures of vinclozolin, M1, and M2.
of this molecule. Since the sequential metabolic steps in
precision-cut liver slices are similar to those occurring in ViVo
(14) and because the retention of heterogeneous cell population
and liver architecture is close to the physiological situation in
ViVo (15), we used this in Vitro model to study the metabolism
of VZ in Wistar rats. Our investigations also include kinetic
data. The in ViVo metabolic profile of VZ was also examined
in the urine of rats for comparison with the biotransformation
pattern obtained in Vitro.
exsanguination. Rat livers were perfused with 40 mL of ice-cold (4
°C) oxygenated Krebs-Henseleit buffer to remove blood. Liver cores
were obtained using a stainless steel tube and drill press and im-
mediately placed in ice-cold buffer. A Krumdieck tissue slicer (Alabama
Research and Development Corp., Munford, AL) was used to produce
precision-cut slices of 8 mm diameter and 0.2 mm thickness (17). An
equilibration period of 45 min allowed for the sloughing of damaged
cells and for recovery of Ca2+ and K+ homeostasis in the slices. At
the end of incubations, the mean wet weight of slices was 20 mg,
containing about 2 mg of protein. Slices were incubated in 12-well
plastic tissue culture plates (18) using DMEM medium. The incubation
volume was 1 mL. Plates were maintained at 37 °C in a rotary shaker,
set at 90 rpm and saturated with a 95% O2/5% CO2 atmosphere. For
each experiment, incubations were performed in triplicate using liver
slices from three different rats. VZ was introduced dissolved into 2 µL
of DMSO. In a first experiment, 6 incubation times (1, 2, 3, 4, 5, and
6 h) with one concentration (10 µM) of labeled VZ (3150 Bq/
incubation) were tested. In a second experiment, 5 concentrations (5,
10, 50, 100, and 250 µM) of labeled VZ (3150 Bq/incubation) were
incubated for 4 h. Blank samples were carried out by incubating 10
µM radiolabeled VZ (3150 Bq/incubation) in DMEM without rat liver
slices in the conditions described above for 1 to 6 h. In each case, at
the end of incubation, media samples were immediately harvested and
stored at -20 °C until analysis.
Liver Slice Viability. 7-Ethoxycoumarin (7-EC) was used as a probe
substrate for oxidative and conjugative metabolism, to assess the drug-
metabolizing capacity of the rat liver slices. Control incubations were
performed in triplicate using 14C-labeled 7-EC 25 µM during 4 h for
the rat liver slices.
In ViWo Metabolic Studies. Three male rats were individually
housed in metabolic cages. Animals were individually fed by gavage
with 1 mg/kg [14C]-VZ dissolved in DMSO and adjusted with unlabeled
VZ to obtain a final activity of 0.481 MBq [14C]-VZ/kg b.w. The final
concentration of VZ in DMSO was 1 mg/mL. Urine was collected for
24 h and stored at -20 °C until analysis.
MATERIALS AND METHODS
Chemicals. [Ring-U-14C]-VZ was purchased from IZOTOP (Budap-
est, Hungary). Its specific activity was 71.17 MBq/mmol and its
radiopurity g99%, as controlled by radio-HPLC. Unlabeled VZ was
extracted from Ronilan (BASF France, Levallois-Perret) as described
by Molina-Molina (16). Briefly, VZ was extracted with dichlo-
romethane, and then the extract was evaporated to dryness, dissolved
in methanol, and recrystallized. Solid vinclozolin was then ground to
powder, and its purity was checked by melting point measurement on
a Bu¨chi 510 apparatus (108 °C corresponding to the value given in the
literature) and was analyzed by HPLC coupled to a diode array detector
(Spectra, Thermo Electron, Courtaboeuf, France). Its chemical purity
was found to be greater than 97%. [14C]-7-Ethoxycoumarin (7-EC) was
purchased from Amersham (Buckinghamshire, UK). 7-EC, standard
3,5-dichloroaniline PESTANAL, Krebs-Henseleit buffer, and Dubel-
cco’s modified eagle medium (DMEM) were from Sigma-Aldrich
Chimie (Saint Quentin Fallavier, France).
Standard dihydroxylated M1, M2, and VZ on the vinyl group were
chemically produced by catalytic amounts of osmium tetroxide with
N-methylmorpholine N-oxide. The purified glycols were then analyzed
by mass spectrometry, and their retention times were controlled by UV
detection at 260 nm for comparison with metabolites.
All chemicals, solvents, and reagents for the preparation of buffers
and HPLC eluents were of the highest grade commercially available
from Merck (Nogent-sur-Marne, France). Ultrapure water from Milli-Q
system (Millipore, Saint Quentin en Yvelines, France) was used for in
Vitro preparations and HPLC mobile phases.
Animals. Male Wistar rats weighing about 250 g were purchased
from Iffa Credo (L’arbresle, France). Upon arrival at the animal facility,
animals were allowed at least 1 week of acclimation prior to
experimentation. Animals were given free access to laboratory diet
(UAR 210; Villemoisson sur Orge, France) and tap water.
Analytical Procedure. All samples were individually thawed
immediately before analysis. The radioactivity of each incubation
medium was measured from aliquots (50 µL) by liquid scintillation
counting in a Packard Tricarb 4430 counter with Ultima Gold (Packard
Instruments Co., Downers Grove, IL) as the scintillation cocktail.
Aliquots (100 µL) of media samples from liver slice incubation or
from rat urine were centrifuged for 5 min at 8600g, and the supernatant
Liver Slice Preparation and Incubation. In each experiment, the
animals were killed by cervical dislocation, followed by immediate

4834 J. Agric. Food Chem., Vol. 56, No. 12, 2008
Bursztyka et al.
Figure 3. ESI-MS/MS mass spectra of M5 (a), M5 glucuronides (b), and M4 glucuronide (c).
was diluted with 400 µL of the corresponding HPLC system mobile
phase A. HPLC coupled to online radioactivity detection (Flo-One
A-500 instrument with Flo-scint II as scintillation cocktail, Packard
Instrument Co.) was used for metabolite profiling. HPLC separations
were performed on a Kromasil C18 column (250 × 4.6 mm, 5 µm)
coupled to a Nucleodur C8 column (250 × 4.6 mm, 5 µm). The C18
column was protected by a Kromasil C18 guard precolumn (18 × 4.6
mm, 5 µm). Collection of radioactive peaks for MS analysis was carried
out with a Gilson Model 202 fraction collector (Gilson, Villiers-Le-
Bel, France). Briefly, 5 fractions per minute were obtained. For each
of the main peaks, only the fraction in the middle of the peak was
purified for further analysis.
gradient from 40% to 90% B, 44 to 49 min linear gradient from 90%
to 100% B, 49 to 66 min linear gradient from 100% B, and 66 to 81
min linear gradient from 100% B to 100% A. Metabolites were
quantified by integrating the area under the radioactivity detected peaks.
Biotransformation of [3-14C]-7-EC to radiolabeled 7-hydroxycou-
marin, 7-hydroxycoumarin sulfate, and 7-hydroxycoumarin glucuronide
was analyzed by radio-HPLC as previously described (19).
Metabolite Identification. Analyses were carried out on a quadru-
pole ion trap mass spectrometer (Finnigan LCQ, Thermo Finnigan, Les
Ulis, France) fitted with an electrospray ionization source operating in
thenegativeionmode.Samples(typically0.3-1ng/µLinmethanol-water
(50-50, v/v)) were directly introduced into the ionization source via a
syringe pump operating at a flow rate of 5 µL/min. The heated capillary
was maintained at 200 °C. Typical operating parameters for production
and transmission of electrospray produced ions into the ion trap analyzer
were needle voltage 4.8-5.3 kV, nebulizing gas flow rate 40 arb-units,
The mobile phases consisted of ammonium acetate buffer (20 mM,
pH 3.5) and acetonitrile 80:20 v/v in A and 30:70 v/v in B. The flow
rate was 0.7 mL, and the temperature was 35 °C. The gradient used
was 0-41 min linear gradient from 0% to 40% B, 41 to 44 min linear

Biotransformation of Vinclozolin
J. Agric. Food Chem., Vol. 56, No. 12, 2008 4835
MS/MS and by comparison with data from the literature (1, 22).
The major one (RT ) 53.8 min) gave an [M - H]- ion at m/z
302 and characteristic fragment ions at m/z 258, 175, and 115,
corresponding to M1. The NI-ESI-MS spectrum of the second
one (RT ) 58 min) displayed an [M - H]- ion at m/z 258 whose
MS/MS analysis gave a fragment ion at m/z 160 diagnostic of
M2. No other degradation products were observed.
Biotransformation of VZ by rat liver slices was extensive
(Figure 2). The radio-HPLC profiles obtained in our experi-
mental conditions were qualitatively similar whatever the
concentration tested or the duration of incubation. Collection
of the radioactive peaks was carried out prior to MS analysis
of metabolites. The identification of metabolites was performed
only on 100 µM incubations.
Peaks M5a and M5b (RT ) 42.6 and 43 min) were
incompletely separated and gave the same [M - H]- ion at
m/z 292 accompanied by the isotopic contribution of the 37Cl
atoms giving an ion at m/z 294. After isolation and fragmentation
of the m/z 292 precursor ion into the ion trap device, the MS/
MS spectrum presented in Figure 3a was obtained. Diagnostic
fragment ions were observed at m/z 103, 131, 160, and 232.
The complementary fragment ions at m/z 160 and 131 cor-
respond to the cleavage of the amide bond of M5 with negative
charge retention being possible on both sides of the molecule.
R-Cleavages of the tertiary alcohol explain the occurrence of
the fragment ions at m/z 103 and 232, respectively. On the basis
of these data, peaks M5a and M5b could be identified as two
different forms (likely diastereoisomers) of 3′,5′-dichloro-2,3,4-
trihydroxy-2-methylbutyranilide (M5). In order to confirm this
identification, a chemical hydroxylation of M2 was carried out
using osmium tetroxyde. The resulting compound displayed a
RT of 42.6 min. when injected in the same HPLC system. The
corresponding MS and MS/MS spectra were super imposable
with those obtained for M5a.
Figure 4. Time course study of the hydrolysis/biotransformation of
vinclozolin. (a) Spontaneous hydrolysis of vinclozolin in DMEM at 37 °C.
(b) Metabolism of vinclozolin in precision-cut rat liver slices. The
concentration of vinclozolin in the incubation medium (DMEM) was 10
µM. The values are the means of three rats, and the error bars correspond
to SD.
The negative ESI mass spectrum of peaks M5G1, M5G2 and
M5G3 (RT ) 27.6, 29.2 and 30.3 min respectively) gave an
intense signal at m/z 468 with the characteristic isotopic pattern
of a doubly chlorinated molecule (i.e., m/z 470 present at ca.
65% intensity relative to m/z 468). The MS/MS spectrum of
the m/z 468 precursor ion of the first peak is presented in Figure
3b. It displays characteristic fragment ions at m/z 193, 232, 235,
and 292, which could be attributed to the fragmentation pattern
presented in Figure 3b. These characteristic fragment ions
allowed us to identify M5G1 as a glucuronic acid conjugate of
M5. The m/z 235 fragment ion allowed us to rule out the
glucuronidation on the tertiary alcohol group of M5. However,
on the basis of the structural information available on the MS/
MS spectrum, we were not able to determine the position of
glucuronidation, which may concern either the primary or the
secondary hydroxyl function of M5. The MS/MS spectra
obtained from the m/z 468 ion of M5G2 and M5G3 were
identical to that of M5G1. Thus, these peaks also corresponded
to M5 glucuronides. The presence of several peaks for M5
glucuronides can be attributed to (i) the two possible positions
of glucuronidation evidenced by MS/MS analysis and (ii) the
occurrence of diastereoisomeric forms of M5 glucuronide, which
may be separated by reversed phase HPLC.
transfer capillary voltage -30 to +10 V, and tube lens offset -35 to
+10 V. Other parameters for MS/MS experiments were as follows:
isolation width 1 amu and excitation time 30 ms. Excitation voltage
was adjusted for each compound in order to get maximum structural
information for the compound of interest (typically 0.8-1.2 Vp-p). All
analyses were achieved under automatic gain control conditions using
helium as damping as well as collision gas for MS/MS experiments.
Data Analysis. Data analysis was conducted using an unpaired
Student’s t-test using Instat (Graph Pad software). Differences were
considered significant if P < 0.05. Km and Vmax values were determined
by nonlinear regression analysis as indicated by Loft and Poulsen (20).
RESULTS
Metabolism of Vinclozolin in Rat Liver Slices. After
incubation of radiolabeled VZ with the precision-cut liver slices,
about 80% of the initial radioactivity was found in the incubation
media for all the concentrations tested. No further analysis was
undertaken on the radiolabeled material remaining in the slices.
The qualitative and quantitative results obtained with incuba-
tion of 7-EC with rat liver slices were in accordance with those
reported previously (21) with more than 50% of 7-EC metabo-
lized in 4 h into 7-hydroxycoumarine, 7-hydroxycoumarine-
glucuronide, and 7-hydroxycoumarine-sulfate, indicating a good
metabolic capacity of the rat liver slices used.
Identification of Metabolites. The radio-HPLC analysis of
the media control incubation revealed that about 85% of VZ
was hydrolyzed after only 1 h in DMEM at 37 °C, giving two
breakdown products; after 6 h, the parent compound was totally
hydrolyzed (Figure 2). These products were identified by ESI-
The isotopic pattern of a chlorinated metabolite was located
at m/z 494/496 on the NI-ESI mass spectrum of peak M4G (RT
) 34 min). For getting structural information, the m/z 494 ion
was selected and submitted to resonant excitation in an MS/
MS experiment (Figure 3c). The main resulting fragment ions
were produced at m/z 235 and 307. A weak signal was also
observed at m/z 175, suggesting the occurrence of a glucuronic

4836 J. Agric. Food Chem., Vol. 56, No. 12, 2008
Bursztyka et al.
Figure 5. Michaelis-Menten plots of vinclozolin biotransformation in precision-cut rat liver slices. Slices were incubated for 4 h with various concentrations
of vinclozolin. The values are the means of three rats, and the error bars correspond to SD. (Left) Formation of M1, M2, and M5. (Right) Formation of
M5 glucuronides (total glucuronides of M5 were measured for calculations).
Metabolite formation over the concentration range is shown
in Figure 5. Because of an incomplete solubilization of VZ at
250 µM, the results corresponding to this concentration were
not represented. M1 and M5 formations were linear up to 100
µM in precision-cut liver slices (Figure 5, left). M2 appeared
to be produced under saturation kinetics with apparent Vmax
)
1.22 ( 0.92 nmol/h/mg of protein. However, M1 and M2 are
produced by nonenzymatic hydrolysis of VZ; thus, the satura-
tion-look of M2 production is due to its biotransformation into
M5 and M5 glucuronides, while its production is limited by
the VZ hydrolysis into M1. M5 conjugation to glucuronide acid
over the concentration range is represented in Figure 5, right.
Since separation of the M5G1, M5G2, and M5G3 peaks did
not allow a precise quantification of these metabolites, the kinetic
data were calculated by using the sum of the glucuronide
conjugates produced from M5 (M5Gs) at each concentration.
Total M5Gs were produced under saturation kinetics with
Figure 6. Zero to 24 h urinary radio-HPLC profile from rats force-fed a
single dose of [14C]-vinclozolin (1 mg/kg bw).
acid conjugate. The m/z 235 fragment ion could reasonably
present the same structure as that indicated for M5 glucuronide
(see Figure 3). The formation of the m/z 307 ion can result in
the cleavage of the five-membered heterocycle considering that
no metabolic ring opening occurred on VZ in this case. Thus,
this peak was identified as the glucuronic acid conjugate of
3-(3,5-dichlorophenyl)-5-methyl-5-(1,2-dihydroxyethyl)-1,3-ox-
azolidine-2,4-dione (M4).
M3 coeluted with standard 3,5-dichloroaniline (RT ) 56.2
min) monitored by UV detection at 212 nm.
Kinetics of VZ Hydrolysis and Metabolism in Rat Liver
Slices. The hydrolysis time course of VZ in DMEM at 37 °C is
shown in Figure 4a. VZ is almost totally hydrolyzed in 2 h,
giving M1 (about 80%) and M2 (about 20%).
apparent Vmax ) 0.54 ( 0.14 nmol/h/mg of protein and Km
29.14 ( 20.76 µM.
)
Identification of Urinary Metabolites. An average of 24%
of the administered radioactivity was found in 0-24 h urine
samples. The radio-HPLC profile corresponding to pooled urine
samples of male rat urine 24 h after being force fed is shown
in Figure 6. The major metabolites were identified by ESI/MS-
MS as M5G1, M5G2, M5G3, M4G, M5, and M1. Several
additional peaks corresponding to polar compounds were
detected at elution times between 8 and 14 min. Probably
because of an incomplete separation of several minor metabo-
lites, we failed to identify these compounds. Another additional
peak was detected at RT ) 46 min. Its analysis by negative
ESI-MS gave a signal corresponding to the [M - H]- ion of
two dichlorinated species at m/z 318/320 and m/z 336/338. By
MS/MS analysis, the m/z 318 ion yielded main characteristic
fragment ions at m/z 160, 258, and 274. The first one
corresponded to the deprotonated form of dichloroaniline,
whereas the m/z 258 and 274 fragment ions resulted in the
elimination of HO-CHdCH-OH and CO2, respectively, from
the m/z 318 [M - H]- precursor ion. These data supported the
identification of this peak as 3-(3,5-dichlorophenyl)-5-methyl-
5-(1,2-dihydroxyethyl)-1,3-oxazolidine-2,4-dione (M4). This
was also confirmed by LC and MS/MS analysis of the glycol
resulting from the chemical hydroxylation of VZ by osmium
tetroxyde, which gave both a RT of 46 min and the same
fragment ions as M4 after MS/MS analysis of the m/z 318
precursor ion. Besides, the MS/MS spectrum of the m/z 336
ion displayed a unique fragment ion at m/z 149. The chemical
hydroxlation of M1 by osmium tetroxyde gave a compound with
a RT of 46 min, also displaying an [M - H]- isotopic cluster
at m/z 336/338. The MS/MS analysis of the m/z 336 ion also
yielded a unique fragment ion at m/z 149, corresponding to the
The kinetics of formation of [14C]-VZ metabolites by rat liver
slices is shown in Figure 4b. The amount of VZ in incubation
media rapidly decreased and represented about 15% of total
radioactivity after 1 h, about 6% after 2 h, and less than 2%
after 4 h. The amounts of M1 and M2 reached the maximum
values after about 1 h, accounting for more than 34% of total
radioactivity for M1 and 4% for M2, whereas for M5, the
maximum values were reached after about 2 h, corresponding
to about 36% of the total radioactivity. After reaching the
maximum values, a steady decrease of the concentrations of
these three metabolites was observed, M1 accounting for 12%
of the total radioactivity at 6 h of incubation, M5 still
representing more than 21% at the same sampling time, whereas
only traces of M2 were detected. After 2 h, M5 was the major
metabolite at all time points. M5-glucuronides (M5G1, M5G2,
and M5G3) increased after about 1 h to represent 33% of the
total radioactivity after 6 h. M4G increased to represent about
6% of the total radioactivity after 3 h of incubation. 3,5-
Dichloroaniline represented between 2 and 3% of the total
radioactivity present in incubation media at all time points (data
not shown).

Biotransformation of Vinclozolin
J. Agric. Food Chem., Vol. 56, No. 12, 2008 4837
Figure 7. Proposed metabolic pathways of the in vitro (precision-cut rat liver slices) and in vivo biotransformation of vinclozolin. (*) Only present in rat
urine.
cleavage of the ester function of this compound with charge
retention on the nonaromatic part of the molecule. On the basis
of the comparison of both retention time and MS/MS spectra,
this metabolite was identified as 2-[[(3,5-dichlorophenyl)-
carbamoyl]oxy]-2-methyl-3,4-dihydroxy-butanoic acid (M6).
From 24 h after the force-feeding, M5 and its conjugates to
glucuronic acid were the major metabolites present in urine with
M1.
a maximum after 1 h and decreased gradually thereafter. We
observed that a dihydroxyladed metabolite of M1 (M6) can be
produced. However, no phase II metabolites originating from
M1 were identified, and the amount of M6 produced could not
explain the decrease in M1 amount. This result could be
explained by the recyclization of M6 to give M4, which can
undergo hydrolysis to give M5 (data not shown). It is also
possible that M6 undergoes further biotransformation into
unidentified metabolites. Moreover, the rapid biotransformation
of M2 could result in a concomitant hydrolysis of VZ to M2,
favoring the subsequent conversion of M1 to VZ. At the same
time, M2 is always present in very small amounts because it is
readily metabolized, giving M5 and M5 glucuronides. After 6 h,
VZ and M2 were present only in trace amounts, while
glucuronides of M5 and M4 represented about 33% of the total
radioactivity.
DISCUSSION
The present study characterizing VZ metabolites produced
in Vitro in liver slices and in ViVo provides evidence that VZ
and corresponding breakdown products are efficiently biotrans-
formed by rats.
Spontaneous hydrolysis of VZ occurred in the incubation
medium without slices. This result was expected since the
hydrolysis of VZ in various aqueous buffers was described by
Szeto et al. (1). These authors demonstrated that the reaction
was base-catalyzed and that the rate of hydrolysis was propor-
tional to pH. At 35 °C and pH 7.0, 10% of VZ (40 µM)
hydrolyzed after 1 h to yield M1 (8.5%) and M2 (1.5%), whereas
at pH 8.0, 47% of VZ was converted to M1 (39%) and M2
(8%) after 1 h. In our experimental conditions (incubations in
DMEM without slices) and after the same period of time, M1
and M2 accounted for 72 and 13% of total radioactivity,
respectively, suggesting that the conversion of VZ to M1 and
M2 could also be partly due to DMEM catalysis of the reaction.
M3 could not be observed because the incubation times were
too short (1).
Although the chemical conversion of VZ to M1 is known to
be reversible, the formation of VZ from M1 by recyclization
only occurs at a significant extent at acidic pH (1). Taking into
consideration that our incubations were undertaken at pH 7.4,
it was likely that only very minute amounts of M1, if any, could
be converted back to VZ. When incubations were performed
with liver slices, the concentration of M1 increased steadily to
The major metabolic pathways of VZ are summarized in
Figure 7. The same metabolites involving phase I and phase II
enzymes are characterized from rat urine and from rat liver
slices. The first step is the hydrolytic opening of the oxazolidine
ring, involving nonenzymatic cleavage of the 2-, 3-, or 3,4-
nitrogen-carbon bonds, to give M2 and M1 compounds,
respectively. Cleavage of the 2,3-nitrogen-carbon bond is
followed by decarboxylation and is not reversible, in contrast
to the cleavage of the 3,4-nitrogen-carbon bond (1). The
enzymatic phase-I step includes the dihydroxylation of the vinyl
group, presumably via an epoxyde intermediate, resulting in
metabolite M4 from VZ, metabolite M5 from M2, and M6 from
M1. Phase-II pathways include extensive conjugation to glu-
curonic acid of the dihydroxylated M4 and M5 compounds,
giving three glucuronoconjugates. These pathways are in
agreement with those reported previously in unpublished reports
(23). M5 has also been reported as a fungal metabolite of VZ
(24). M5 can be obtained also from M4 by hydrolytic cleavage
of the 2,3-nitrogen-carbon bond. Moreover, the same hydrolysis
could happen with M4 conjugates, giving M5 conjugates,

4838 J. Agric. Food Chem., Vol. 56, No. 12, 2008
Bursztyka et al.
(5) Shono, T.; Suita, S.; Kai, H.; Yamaguchi, Y. The effect of a
prenatal androgen disruptor, vinclozolin, on gubernacular
migration and testicular descent in rats. J. Pediatr. Surg. 2004,
39, 213–6.
explaining that the major metabolites identified in rat liver slice
incubation media are M5 and M5 conjugates. On the basis of
the RT of standard 3,5-dichloroaniline, M3 seemed to be present
only in precision-cut rat liver slice incubation media. M3 is a
well described end product of VZ degradation (1, 25) but was
not detected in rat urine, probably because 3,5-dichloroaniline
undergoes further biotransformation, such as hydroxylation and
conjugation to glucuronic acid (26).
In Vitro and in ViVo, the major metabolites identified are M1
(hydrolysis product), M5 (dihydroxylated metabolite from M2),
and M5 glucurono-conjugates, and to a lesser extent M4 and
its glucuronoconjugate. VZ antiandrogenic activity is described
to be mainly due to M2 and M1 since they compete with [17
R-methyl]-methyltrienolone (R1881) for the androgen receptor
with a Ki of 9.7 and 92 µM, respectively, that is about 100-
and 10-fold more effective than VZ itself (3).
Data on the fate of VZ in humans are also scarce. The
presence of VZ residues in human serum and urine was
investigated recently (27–31). In some cases, only VZ was
determined, whereas in others, the sum VZ + metabolites was
estimated, on the basis of the quantification of 3,5-dichloroa-
niline, determined after basic or acidic hydrolysis of the samples.
No data on the metabolic pathways of VZ in humans were
identified.
This is the first published study on the metabolic fate of VZ
in rats in ViVo and in Vitro, including identification of the major
metabolites and kinetics of biotransformation. It demonstrates
that M1 and M2, which are mainly responsible of the antian-
drogenic effects attributable to VZ, are not the end products of
the biotransformation of VZ in rats. VZ and M2 are quickly
biotransformed by dihydroxylation of the vinyl group and by
further conjugation to glucuronic acid. The activity of these
metabolites toward the androgen receptor remains to be
definitively established.
(6) Shono, T.; Suita, S.; Kai, H.; Yamaguchi, Y. Short-time exposure
to vinclozolin in utero induces testicular maldescent associate with
a spinal nucleus alteration of the genitofemoral nerve in rats.
J. Pediatr. Surg. 2004, 39, 217–9, discussion 217-9.
(7) Kang, I. H.; Kim, H. S.; Shin, J. H.; Kim, T. S.; Moon, H. J.;
Kim, I. Y.; Choi, K. S.; Kil, K. S.; Park, Y. I.; Dong, M. S.; Han,
S. Y. Comparison of anti-androgenic activity of flutamide,
vinclozolin, procymidone, linuron, and p,p′-DDE in rodent 10-
day Hershberger assay. Toxicology 2004, 199, 145–59.
(8) Hotchkiss, A. K.; Ostby, J. S.; Vandenbergh, J. G.; Gray, L. E.,
Jr. An environmental antiandrogen, vinclozolin, alters the orga-
nization of play behavior. Physiol. BehaV. 2003, 79, 151–6.
(9) Baatrup, E.; Junge, M. Antiandrogenic pesticides disrupt sexual
characteristics in the adult male guppy Poecilia reticulata.
EnViron. Health. Perspect. 2001, 109, 1063–70.
(10) Tillmann, M.; Schulte-Oehlmann, U.; Duft, M.; Markert, B.;
Oehlmann, J. Effects of endocrine disruptors on prosobranch snails
Mollusca: Gastropoda) in the laboratory. Part III: Cyproterone
acetate and vinclozolin as antiandrogens. Ecotoxicology 2001, 10,
373–88.
(11) Anway, M. D.; Cupp, A. S.; Uzumcu, M.; Skinner, M. K.
Epigenetic transgenerational actions of endocrine disruptors and
male fertility. Science 2005, 308, 1466–9.
(12) Anway, M. D.; Leathers, C.; Skinner, M. K. Endocrine disruptor
vinclozolin induced epigenetic transgenerational adult-onset dis-
ease. Endocrinology 2006, 147, 5515–23.
(13) Chang, H. S.; Anway, M. D.; Rekow, S. S.; Skinner, M. K.
Transgenerational epigenetic imprinting of the male germline by
endocrine disruptor exposure during gonadal sex determination.
Endocrinology 2006, 147, 5524–41.
(14) Parrish, A. R.; Gandolfi, A. J.; Brendel, K. Precision-cut tissue
slices: applications in pharmacology and toxicology. Life Sci.
1995, 57, 1887–901.
(15) Ekins, S.; Williams, J. A.; Murray, G. I.; Burke, M. D.; Marchant,
N. C.; Engeset, J.; Hawksworth, G. M. Xenobiotic metabolism
in rat, dog, and human precision-cut liver slices, freshly isolated
hepatocytes, and vitrified precision-cut liver slices. Drug. Metab.
Dispos. 1996, 24, 990–5.
(16) Molina-Molina, J. M.; Hillenweck, A.; Jouanin, I.; Zalko, D.;
Cravedi, J. P.; Fernandez, M. F.; Pillon, A.; Nicolas, J. C.; Olea,
N.; Balaguer, P. Steroid receptor profiling of vinclozolin and its
primary metabolites. Toxicol. Appl. Pharmacol. 2006, 216, 44–
54.
ABBREVIATIONS USED
AR, androgen receptor; CNS, central nervous system; DMEM,
Dubelcco’s modified eagle medium; DMSO, dimethyl sulfoxide;
7-EC, 7-ethoxycoumarin; ESI, electrospray ionization; HPLC,
high-performance liquid chromatography; LC-MS, combined
high-performance liquid chromatography/mass spectrometry;
MS, mass spectrometry; RT, retention time; UV, ultraviolet; VZ,
vinclozolin.
(17) Krumdieck, C. L.; dos Santos, J. E.; Ho, K. J. A new instrument
for the rapid preparation of tissue slices. Anal. Biochem. 1980,
104, 118–23.
ACKNOWLEDGMENT
We thank Florence Blas y Estrada and Raymond Gazel for
animal care.
(18) Dogterom, P. Development of a simple incubation system for
metabolism studies with precision-cut liver slices. Drug. Metab.
Dispos. 1993, 21, 699–704.
(19) Zalko, D.; Perdu-Durand, E.; Debrauwer, L.; Bec-Ferte, M. P.;
Tulliez, J. Comparative metabolism of clenbuterol by rat and
bovine liver microsomes and slices. Drug. Metab. Dispos. 1998,
26, 28–35.
(20) Loft, S.; Poulsen, H. E. Metabolism of metronidazole and
antipyrine in isolated rat hepatocytes. Influence of sex and enzyme
induction and inhibition. Biochem. Pharmacol. 1989, 38, 1125–
36.
(21) Ekins, S.; Murray, G. I.; Burke, M. D.; Williams, J. A.; Marchant,
N. C.; Hawksworth, G. M. Quantitative differences in phase I
and II metabolism between rat precision-cut liver slices and
isolated hepatocytes. Drug. Metab. Dispos. 1995, 23, 1274–9.
(22) Dhananjeyan, M. R.; Erhardt, P. W.; Corbitt, C. Simultaneous
determination of vinclozolin and detection of its degradation
products in mouse plasma, serum and urine, and from rabbit bile,
by high-performance liquid chromatography. J. Chromatogr., A
2006, 1115, 8–18.
LITERATURE CITED
(1) Szeto, S. Y.; Burlinson, N. E.; Rahe, J. E.; Oloffs, P. C. Kinetics
of hydrolysis of the dicarboximide fungicide vinclozolin. J. Agric.
Food Chem. 1989, 37, 523–529.
(2) Kelce, W.; Monosson, R.; Gamcsik, E.; Laws, M. P.; Gray, S. C.;
L. E. Jr. Environmental hormone disruptors: evidence that
vinclozolin developmental toxicity is mediated by antiandrogenic
metabolites. Toxicol. Appl. Pharmacol. 1994, 126, 276–85.
(3) Kelce, W. R.; Lambright, C. R.; Gray, L. E., Jr.; Roberts, K. P.
Vinclozolin and p,p′-DDE alter androgen-dependent gene expres-
sion: in vivo confirmation of an androgen receptor-mediated
mechanism. Toxicol. Appl. Pharmacol. 1997, 142, 192–200.
(4) Gray, L. E., Jr.; Ostby, J. S.; Kelce, W. R. Developmental effects
of an environmental antiandrogen: the fungicide vinclozolin alters
sex differentiation of the male rat. Toxicol. Appl. Pharmacol. 1994,
129, 46–52.

Biotransformation of Vinclozolin
J. Agric. Food Chem., Vol. 56, No. 12, 2008 4839
(23) Watson, W. Vinclozolin Pesticide Residues in Food: 1995
Evaluations. Part II: Toxicological & Environmental. IPCS
INCHEM, 1995. http://www.inchem.org/documents/jmpr/jmpmono/
v95pr18.htm.
mass spectrometry. Rapid Commun. Mass Spectrom. 2007, 21,
536–42.
(29) Moreno Frias, M.; Garrido Frenich, A.; Martinez Vidal, J. L.;
Mateu Sanchez, M.; Olea, F.; Olea, N. Analyses of lindane,
vinclozolin, aldrin, p,p′-DDE, o,p′-DDT and p,p′-DDT in human
serum using gas chromatography with electron capture detection
and tandem mass spectrometry. J. Chromatogr., B 2001, 760, 1–
15.
(30) Turci, R.; Barisano, A.; Balducci, C.; Colosio, C.; Minoia, C.
Determination of dichloroanilines in human urine by gas chro-
matography/mass spectrometry: validation protocol and establish-
ment of Reference values in a population group living in central
Italy. Rapid Commun. Mass Spectrom. 2006, 20, 2621–5.
(31) Wittke, K.; Hajimiragha, H.; Dunemann, L.; Begerow, J. Deter-
mination of dichloroanilines in human urine by GC-MS, GC-MS-
MS, and GC-ECD as markers of low-level pesticide exposure.
J. Chromatogr., B 2001, 755, 215–28.
(24) Pothuluri, J. V.; Freeman, J. P.; Heinze, T. M.; Beger, R. D.;
Cerniglia, C. E. J. Agric. Food Chem. 2000, 48, 6138–6148.
(25) Cabras, P.; Diana, P.; Meloni, M.; Pirisi, F. M.; Pirisi, R. Reversed-
phase high-performance liquid chromatography of pesticides. VII.
Analysis of vinclozolin, iprodione, procymidone, dichlozolinate
and their degradation product 3,5-dichloroaniline on white must
and wine extracts. J. Chromatogr. 1983, 256, 176–81.
(26) Valentovic, M. A.; Lo, H. H.; Brown, P. I.; Rankin, G. O. 3,5-
Dichloroaniline toxicity in Fischer 344 rats pretreated with
inhibitors and inducers of cytochrome P450. Toxicol. Lett. 1995,
78, 207–14.
(27) Carlucci, G.; Pasquale, D. D.; Ruggieri, F.; Mazzeo, P. Determi-
nation and validation of a simple high-performance liquid
chromatographic method for simultaneous assay of iprodione and
vinclozolin in human urine. J. Chromatogr., B 2005, 828, 108–
12.
Received for review September 20, 2007. Revised manuscript received
April 14, 2008. Accepted April 15, 2008. This study has been partly
supported by the European Union network CASCADE (FOOD-CT-
2003-506319).
(28) Lindh, C. H.; Littorin, M.; Amilon, A.; Jonsson, B. A. Analysis
of 3,5-dichloroaniline as a biomarker of vinclozolin and iprodione
in human urine using liquid chromatography/triple quadrupole
JF0728045