Metabolism of fungicide diethofencarb in grape (Vitis vinifera L.): Definitive identification of thiolactic acid conjugated metabolites

Article abstract of DOI:10.1021/jf034367+

The metabolic fate of diethofencarb (isopropyl 3,4-diethoxycarbanilate) separately labeled with ¹⁴C at the phenyl ring and 2-position of the isopropyl moiety was studied in grape (Vitis vinifera L.). The acetonitrile solution of ¹⁴C-diethofencarb at a rate of 500 g a.i. ha ^-1 was once applied topically to fruits or leaves at the maturity stage of fruits (PHI 35 days), and the plants were grown in the greenhouse until harvest. In the grape plants, diethofencarb was scarcely translocated to the untreated portion and was degraded more in the fruit as compared to the leaf. For the fruit, diethofencarb primary underwent O-deethylation at the 4-position of the phenyl ring to form the phenolic derivative, isopropyl 3-ethoxy-4-hydroxycarbanilate (0.9% of the total radioactive residue, TRR). This metabolite was successively transformed via conjugation with glucose at the phenolic hydroxy group (8.1-18.1% TRR) or with thiolactic acid at the 5-position of the phenyl ring (1.5-1.7% TRR). The thiolactic acid conjugate was further metabolized mainly to two different types of glucose conjugates at the 4-position of the phenyl ring (8.7-13.5% TRR) and the hydroxy group in the thiolactic acid moiety (6.4-7.3% TRR), as evidenced by ¹H NMR and atmospheric pressure chemical ionization-liquid chromatography-mass spectrometry together with cochromatographies with synthetic standards.

Full text of DOI:10.1021/jf034367+

J. Agric. Food Chem. 2003, 51, 5329−5336 5329
Metabolism of Fungicide Diethofencarb in Grape (Vitis vinifera
L.): Definitive Identification of Thiolactic Acid Conjugated
Metabolites
TAKUO FUJISAWA,*,† KEIKO ICHISE-SHIBUYA,† TOSHIYUKI KATAGI,†
†
LUIS O. RUZO,§ AND YOSHIYUKI TAKIMOTO
Environmental Health Science Laboratory, Sumitomo Chemical Co., Ltd., 2-1,
Takatsukasa 4-Chome, Takarazuka 665-8555, Japan, and PTRL West, Inc.,
625-B Alfred Nobel Drive, Hercules, California 94547
The metabolic fate of diethofencarb (isopropyl 3,4-diethoxycarbanilate) separately labeled with 14C
at the phenyl ring and 2-position of the isopropyl moiety was studied in grape (Vitis vinifera L.). The
acetonitrile solution of 14C-diethofencarb at a rate of 500 g a.i. ha-1 was once applied topically to
fruits or leaves at the maturity stage of fruits (PHI 35 days), and the plants were grown in the
greenhouse until harvest. In the grape plants, diethofencarb was scarcely translocated to the untreated
portion and was degraded more in the fruit as compared to the leaf. For the fruit, diethofencarb primary
underwent O-deethylation at the 4-position of the phenyl ring to form the phenolic derivative, isopropyl
3-ethoxy-4-hydroxycarbanilate (0.9% of the total radioactive residue, TRR). This metabolite was
successively transformed via conjugation with glucose at the phenolic hydroxy group (8.1-18.1%
TRR) or with thiolactic acid at the 5-position of the phenyl ring (1.5-1.7% TRR). The thiolactic acid
conjugate was further metabolized mainly to two different types of glucose conjugates at the 4-position
of the phenyl ring (8.7-13.5% TRR) and the hydroxy group in the thiolactic acid moiety (6.4-7.3%
1
TRR), as evidenced by H NMR and atmospheric pressure chemical ionization-liquid chromatog-
raphy-mass spectrometry together with cochromatographies with synthetic standards.
KEYWORDS: Diethofencarb; metabolism; Vitis vinifera L.; conjugate; thiolactic acid
INTRODUCTION
Still et al. have found in oat metabolism of chlorpropham that
cysteine is attached to the phenyl ring of 4-OH-chlorpropham
by the C-S linkage (9). Propham has been shown to be
metabolized via ring hydroxylation in several plant species. The
only aglycone identified in soybean when treated with â-glu-
cosidase was isopropyl 2-hydroxycarbanilate (2-OH-propham),
produced by roots and shoots (11). In alfalfa, isopropyl
4-hydroxycarbanilate (4-OH-propham) and 2-OH-propham were
identified as major and minor aglycones, respectively, in the
shoot and the root (12). Zurqiyah et al. have shown that
hydroxylation and conjugation are important transformation
routes in the metabolism of propham in alfalfa (13). In the
shoots, hydroxylation on the isopropyl moiety occurred to an
equal extent with hydroxylation on the 2- and 4-positions of
the phenyl ring.
The carbamate fungicide diethofencarb (1) (isopropyl 3,4-
diethoxycarbanilate) is very effective for the control of various
fungal species, especially Botrytis sp., Cercospora sp., and
Venturia sp., which are resistant to benzimidazole fungicide
(1-3). The metabolic fate of N-aryl carbamates in various plants
has been extensively studied especially for herbicides chlor-
propham (isopropyl 3-chlorocarbanilate) and propham (isopropyl
carbanilate). James and Prendeville (4) have reported that
chlorpropham treated on leaf is extensively converted to water
soluble metabolites in smartweed, pigweed, and tomato, which
are susceptible to chlorpropham. These water soluble fractions
were supposed to consist of â-glucose derivatives conjugated
at the 1-hydroxyprop-2-yl moiety of the primary metabolite. In
hydroponically grown soybean, polar metabolites were detected
in both roots and shoots (5). Isopropyl 5-chloro-2-hydroxycar-
banilate (2-OH-chlorpropham) was identified as an aglycone
in both roots (6) and shoots (7), and isopropyl 3-chloro-4-
hydroxycarbanilate (4-OH-chlorpropham) was in the shoot only
(7). In cucumber, potato, and oat, chlorpropham was found to
be metabolized only to 4-OH-chlorpropham (8-10). Especially,
Diethofencarb can be classified into the carbanilate pesticides
and possesses the electron-donating ethoxy groups in the phenyl
ring different from chlorpropham and propham; hence, the
different types of metabolism could be supposed in plants. From
this standpoint, we have conducted metabolism of 1 in grape
grown in a greenhouse and attempted to identify the metabo-
lites by extensive spectrometric analysis of plant extracts
using liquid chromatography-mass spectrometry (LC-MS) and
* Corresponding author (e-mail: fujisawat1@sc.sumitomo-chem.co.jp).
† Sumitomo Chemical Co., Ltd..
§ PTRL West, Inc..
10.1021/jf034367+ CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/25/2003

5330 J. Agric. Food Chem., Vol. 51, No. 18, 2003
Fujisawa et al.
Figure 1. Synthetic routes of authentic standards 3−6.
(5.0 equiv) to prepare the tetraacetyl-1-bromoglucose derivative, which
was subsequently treated with Me2NCHS in the similar manners of
route 1. The obtained glucose derivative 10 was coupled with 8 to
synthesize 11, which was hydrolyzed to 5 (total yield, 13.6%). The
chemical structures of the authentic standards 2-6 were confirmed by
1H NMR and LC-MS, as listed in Table 1, and their high-performance
liquid chromatography (HPLC) retention times and thin-layer chroma-
tography (TLC) Rf values are listed in Table 2.
Compound 1 uniformly labeled with 14C at the phenyl ring [14C-
phenyl] or at the 2-position of the isopropyl moiety [14C-isopropyl]
was prepared in our laboratory and had a specific activity of 8.88 and
7.64 MBq mg-1, respectively. Their radiochemical purity was more
than 98% as determined by HPLC. Cellulase (EC 3.2.1.4) purchased
from Wako Pure Chemical Industries Ltd. (Osaka) were used for
enzymatic hydrolysis of sugar conjugates. Other reagents were of the
purest grade commercially available.
Spectroscopies. 1H NMR spectra were measured with a Varian Unity
300 FT-NMR spectrometer operating at 299.94 MHz with 4-Nucleus
auto NMR probe, using trimethylsilylpropionate-2,2,3,3-d4 as an internal
standard (δ ) 0.0 ppm). LC-atmospheric pressure chemical ioniza-
tion-MS (LC-APCI-MS) in positive and negative ion modes was
performed using a Hitachi M-1000 spectrometer equipped with a Hitachi
6000 series liquid chromatography. Samples dissolved in methanol were
manually injected into an APCI source at 260 °C through an ODS
column with a flow rate of 1.0 mL min-1 using a gradient system with
acetonitrile (solvent A) and 0.1% formic acid in water (solvent B).
The composition of the mobile phase was changed stepwise as
follows: 0 min, %A-%B, 25-75; 60 min, %A-%B, 50-50 (method
1); 0 min, %A-%B, 50-50; 30 min, %A-%B, 100-0 (method 2).
Radioassay. Radioactivity in the liquid extracts and the surface rinse
from plants was determined by mixing each aliquot with 10 mL of
Packard Scintillator Plus and analyzed by liquid scintillation counting
(LSC) with Packard model 1600TR and 2000CA spectrometers
equipped with an automatic external standard. The background level
of radioactivity in LSC was 30 dpm in an average, which was subtracted
from the dpm value of a measured sample. Radioactivity in the
unextractable residues from the treated plants was measured by using
a Packard model 306 Sample Oxidizer. Unextractable residues were
air-dried at room temperature overnight and weighed with a Mettler
model AE240, and each aliquot was subjected to combustion. 14CO2
produced was absorbed into 9 mL of Packard Carb-CO2 absorber and
mixed with 15 mL of Packard Permafluor scintillator, and the
radioactivity in it was quantified by LSC. The efficiency of combustion
was determined to be greater than 90%.
nuclear magnetic resonance (NMR) spectrometry in conjunction
with direct chromatographic comparison with the synthetic
standards.
MATERIALS AND METHODS
Chemicals. The nonradiolabeled synthetic standards, 1, its O-
deethylated derivative at the 4-position of the phenyl ring (isopropyl
3-ethoxy-4-hydroxycarbanilate, 2), the glucose conjugate of 2 (isopropyl
3-ethoxy-4-â-glucopyranosyloxycarbanilate, 3), the thiolactic acid
conjugate of 2 (3-{3-ethoxy-2-hydroxy-5-[(isopropoxycarbonyl)amino]-
phenylthio}-2-hydroxypropionic acid, 4), and the two glucose conju-
gates of 4 (3-{3-ethoxy-2-hydroxy-5-[(isopropoxycarbonyl)amino]-
phenylthio}-2-â-glucopyranosyloxypropionic acid, 5, and 3-{3-ethoxy-
2-â-glucopyranosyloxy-5-[(isopropoxycarbonyl)amino]phenylthio-2-
hydroxypropionic acid}, 6) were synthesized in our laboratory. The
synthetic routes of 3-6 are briefly shown in Figure 1. All intermediates
were purified by silica gel column chromatography except for the
unstable quinoneimine derivative 8. 3,4-Diethoxynitrobenzene prepared
by reaction of 4-nitrocatechol in dioxane with 3.0 equiv of ethyl iodide
in the presence of 1.05 equiv of sodium hydride was hydrolyzed by
potassium hydroxide in aqueous 2-methoxy ethanol under reflux to give
3-ethoxy-4-hydroxy-nitrobenzene (14). The catalytic reduction of this
nitrobenzene derivative by 10% Pd-C under hydrogen atmosphere
produced the corresponding aniline, which was subsequently coupled
with isopropyl chloroformate (1.5 equiv) in dimethyl formamide in the
presence of 0.2 equiv N,N-dimethylaniline leading to formation of 2
(total yield, ∼30%) (2). The trichloromethylimidate derivative 7 was
separately prepared from tetraacetylglucose, which was prepared by
treating tetraacetyl-1-bromo-glucose and AgCO3 with 1,1,1-trichlo-
romethyl cyanide (10 equiv) in dichloromethane. By coupling 2 and 7
in dichloromethane (15), the tetraacetylglucose derivative of 2 was
obtained and hydrolyzed by sodium hydroxide to afford 3 (total yield
67.0%, route 1). Along with the route 2 (Figure 1), ethyl bromolactate
was first prepared by reduction of ethyl bromopyruvate with 0.3 equiv
of NaBH4 in ethanol. Its hydroxy group was subsequently acetylated
in the usual manner, followed by reaction with N,N-dimethylthiofor-
mamide (Me2NCHS) to give the corresponding ethyl mercaptolactate
(16). Compound 2 in dry dichloromethane was oxidized by lead acetate
to form the corresponding quinoneimide derivative 8 (17) and then
coupled with ethyl mercaptolactate to synthesize 9 (18). A portion of
9 was hydrolyzed with sodium hydroxide to yield 4. Another portion
of 9 was then reacted with 7 in dichloromethane and subsequently
hydrolyzed by sodium hydroxide to 6 (total yield, 45.4%). To synthesize
5 via route 3, the pentaacetylglucose was first coupled with ethyl
bromolactate in the presence of trifluoroborane-diethyl ether complex
Chromatographies. HPLC was carried out by using a Hitachi
L-6200 Pump linked in series with L-4000 UV detector and Packard

Metabolism of Fungicide Diethofencarb in Grape
J. Agric. Food Chem., Vol. 51, No. 18, 2003 5331
%A-%B, 0-100; 80 min, %A-%B, 0-100 (method 4); YMC-Pack
AM-324; 1/33M phosphate buffer (solvent A) and acetonitrile (solvent
B); 0 min, %A-%B, 85-15; 25 min, %A-%B, 85-15; 25.1 min,
%A-%B, 50-50; 45 min, %A-%B, 50-50 (method 5).
TLC was conducted using silica gel 60 F254 thin-layer chromatoplates
(20 cm × 20 cm, 0.25 mm thickness, E. Merck). The following solvent
systems were used for development: toluene/ethyl formate/formic acid,
5/7/1 (v/v/v); n-butanol/acetic acid/water, 6/1/1 (v/v/v). The nonradio-
labeled reference standards were detected by exposing TLC plates to
ultraviolet light. Autoradiograms were prepared by exposing TLC plates
to BAS-IIIs Fuji Imaging Plate for several hours. The radioactivity on
an imaging plate was detected by using a Fuji Bio-Imaging Analyzer
BAS-1500.
Table 1. NMR and MS Data of Authentic Standards
1H NMR (ppm)
MS (m/z)
2
3
4
1.28 (d, J ) 6 Hz, 6H; (CH ) CH),
240 (M + H)+,
226, 198
3 2
1.43 (t, J ) 6 Hz, 3H; CH CH ),
3
2
4.11 (q, J ) 6 Hz, 2H; CH CH ),
3
2
4.99 (m, J ) 6 Hz, 1H; (CH ) CH),
3 2
5.42 (s, 1H; OH), 6.40 (br-s, 1H; NH),
6.45−7.00 (m, 3H; aromatic-H)
1.29 (d, J ) 6 Hz, 6H; (CH ) CH),
400 (M − H)-,
238
3 2
1.42 (t, J ) 6 Hz, 3H; CH CH ),
3
2
4.11 (q, J ) 6 Hz, 2H; CH CH ),
3
2
4.99 (m, J ) 6 Hz, 1H; (CH ) CH),
3 2
Plant Material, Maintenance, and Treatment. For investigation
of metabolic profiles, grape vines (Vitis Vinifera L., cv. Kyoho) with
fruits setting at maturity stages were purchased from Takii & Co., Ltd.
The grape plants were grown in pots in a greenhouse with the
temperature maintained at 25 °C for day and 20 °C for night. The plant
was appropriately watered until harvest for approximately 1.5 months.
[14C]-1 in the following solutions was topically applied at a rate of
500 g a.i. ha-1 to grape bunches (fruits) and leaves once with a
preharvest interval (PHI) of 35 days. The acetonitrile solutions of [14C-
phenyl] and [14C-isopropyl]-1 isotopically diluted by 10-15-fold with
nonradiolabeled 1 were prepared at the concentrations of 2.85-4.00
mg mL-1 for fruit and leaf treatments. The surface area of the grape
bunch was conveniently calculated to be 114 cm2 by assuming its shape
as a circular cone (radius of the bottom circle, 5 cm; height, 12 cm).
The surface area of the leaf was geometrically estimated to be 78.5
cm2 on average. [14C]-1 was topically applied to the surface of fruits
on each bunch and leaves using a microsyringe. Six bunches and seven
leaves per pot were treated with each label. The application was
conducted when the color of the skin began to change (verasion).
For the purpose of collecting enough conjugated metabolites
subjected to spectrometric analyses (MS, NMR), Pinot Noir grapes
under the actual field conditions were used. The field phase of this
study was carried out at Valley Farm Management, 37500 Foothill
Drive, Soledad, CA. [14C-Phenyl]-1 in the following solutions was
applied to the grape bunches with a hand sprayer at a rate of 1000 g
a.i. ha-1, three times with an approximately 40 days interval. For
preparation of dose solution, [14C-phenyl]-1 was isotopically diluted
by 2-fold with nonradiolabeled 1. For each application, the grape
bunches of five vines (130 bunches) were sprayed with 175 mg of [14C-
phenyl]-1 (4.4 MBq) dissolved in 75 mL of methanol/water (1/1, v/v).
The surface area of the grape bunch (cm2) before application was
approximately calculated as 3 in. × 6 in., which was equal to an area
of 18 square inches per bunch. The first, second, and third applications
were conducted at the stage of late fruit setting, beginning of fruit
verasion, and fruit verasion, respectively. PHI was set as 14 days. There
was no irrigation carried out during the actual field phase of this study.
Extraction and Isolation of Metabolites. For investigation of the
metabolic profiles, the grape bunches applied with [14C]-1 were sampled
at 35 days after the treatment. The harvested samples were weighed
and stored in a freezer (< -20 °C) until analysis was conducted. After
the surface of bunches was rinsed by methanol (100 mL per bunch),
grape fruits were removed from the bunches by using scissors,
homogenized in a homogenizer (Nissei, AM-8) at 10 000 rpm for 10
min, and centrifuged (Tomy, CX250) at 5000 rpm for 10 min to separate
supernatant (juice) and pulp. The supernatant was filtered through a
filter paper (pore size; 1 µm) to remove the precipitate. The pulp was
further extracted in the homogenizer (10 000 rpm, 10 min) by acetone/
water (4/1, v/v) at the ratio of 2 mL per 1 g of pulp. The mixture was
filtered, and the residue was extracted two additional times in the same
way. The aliquots of surface rinse, juice, and pulp extracts were
individually radioassayed with LSC and analyzed by HPLC and TLC
cochromatographies with authentic standards. Pulp residues were air-
dried in open vessels at room temperature overnight, and subsamples
of the dried residues were subjected to combustion analysis to determine
the remaining radioactivity. The untreated leaves were also sampled
to examine the extent of translocation of radioactivities. They were
chopped, air-dried in a chamber at room temperature for 1 week, and
subjected to combustion.
3.10−4.60 (m, 7H; sugar-H),
6.86−7.20 (m, 3H; aromatic-H)
1.30 (d, J ) 6 Hz, 6H; (CH ) CH),
358 (M − H)-,
313
3 2
1.44 (t, J ) 6 Hz, 3H; CH CH ),
3
2
3.20 (d, J ) 6 Hz, 2H; CH CH),
2
4.13 (q, J ) 6 Hz, 2H; CH CH ),
3
2
4.32 (t, J ) 3 Hz, 1H; CH CH),
2
4.97 (m, 1H; (CH ) CH),
3 2
6.95−7.05 (s, 2H; aromatic-H)
5
6
1.28 (d, J ) 6 Hz, 6H; (CH ) CH),
520 (M − H)-,
340, 270
3 2
1.43 (t, J ) 6 Hz, 3H; CH CH ),
3
2
4.11 (q, J ) 6 Hz, 2H; CH CH ),
3
2
4.30 (t, J ) 6 Hz, 1H; CH CH),
2
4.98 (m, J ) 6 Hz, 1H; (CH ) CH),
3 2
3.10−4.60 (m, 7H; sugar-H
and d, 2H; CH CH),
2
7.00−7.10 (s, 2H; aromatic-H)
1.28 (d, J ) 6 Hz, 6H; (CH ) CH),
520 (M − H)-
3 2
1.42 (t, J ) 6 Hz, 3H; CH CH ),
3
2
4.12 (q, J ) 6 Hz, 2H; CH CH ),
3
2
4.34 (t, J ) 6 Hz, 1H; CH CH),
2
4.99 (m, J ) 6 Hz, 1H; (CH ) CH),
3 2
3.10−4.60 (m, 7H; sugar-H
and d, 2H; CH CH),
2
7.00−7.10 (s, 2H; aromatic-H)
Table 2. HPLC Retention Times and TLC R Values of Synthetic
f
Standards
retention
time (min)
method 1
R value
f
compd
system 1a
system 2b
1
2
3
4
5
6
66.6
50.3
35.1
46.0
39.7
34.7
0.63
0.57
0.07
0.28
0.03
0.02
0.73
0.72
0.55
0.53
0.32
0.28
a Toluene/ethyl formate/formic acid ) 5/7/1 (v/v/v). b Butanol/acetic acid/water
) 6/1/1 (v/v/v).
Flow-one/Beta A-120 radio detector equipped with a 500 µL liquid
cell where Ultima-Flo AP (Packard) was utilized as a scintillator. A
Sumipax ODS A-212 column (150 mm × 6 mm i.d., 5 µm, SCAS
Co., Ltd.) was employed for both analytical and preparative purposes
at a flow rate of 1 mL min-1, and a YMC-Pack AM-324 S-5 120A
ODS column (300 mm × 10 mm i.d., 5 µm, YMC Co., Ltd.) was used
for a preparative purpose at a flow rate of 2 mL min-1. The following
gradient system was used for typical analysis of the metabolites with
0.01% trifluoroacetic acid (solvent A) and acetonitrile (solvent B): 0
min, %A-%B, 95-5; 60 min, %A-%B, 50-50; 60.1 min, %A-%B,
0-100; 80 min, %A-%B, 0-100 (method 3). The following HPLC
methods (methods 4 and 5) and/or their slightly modified ones were
used for separation and purification of metabolites: YMC-Pack AM-
324; 0.01% trifluoroacetic acid (solvent A) and acetonitrile (solvent
B); 0 min, %A-%B, 75-25; 60 min, %A-%B, 50-50; 60.1 min,

5332 J. Agric. Food Chem., Vol. 51, No. 18, 2003
Fujisawa et al.
Grape leaves applied with [14C]-1 were also sampled at 35 days after
the treatment. The grape leaves were surface rinsed by methanol (100
mL) to remove external radioactivity, cut into small pieces, and similarly
homogenized as fruits at the ratio of 20 mL per 1 g of the plant material.
The mixture was filtered through a filter paper, and the residue was
extracted two additional times in the same way. The extract and surface
rinse were radioassayed and then individually concentrated to dryness,
redissolved in a small volume of organic solvent, and each aliquot was
subjected to HPLC and TLC analyses. Unextractable residues and
untreated leaves were air-dried and combusted with the sample oxidizer
prior to LSC analysis.
Table 3. Distribution of Radioactivities and Metabolites of [14C]-1 with
Fruit Treatment
[14C-phenyl]-1
[14C-isopropyl]-1
a
a
%TRR
ppm
%TRR
ppm
surface rinse
0.811
1
50.71
ND
54.98
0.57
1.038
0.011
others
ND
juice
ND
ND
1
2
ND
ND
1.61
ND
0.030
ND
For the purpose of collecting enough amount of conjugated
metabolites, grape fruits applied with [14C-phenyl]-1 in the field were
sampled at 14 days after the last treatment. The harvested samples were
weighed and stored in a freezer (< -20 °C) until analysis was
conducted. While still frozen, bunches were rinsed with acetone
followed by acetone/water (9/1, v/v). Bunches were allowed to thaw,
and grape fruits were removed from bunches to crush by hand. The
grape juice and pulp were separated through a funnel and glass wool
plug. All of the juice was combined (1.5 L) and subjected to preparative
purification by the solid phase absorbent Porapack Q. Approximately
300 mL of juice was loaded on the Porapack Q column packed in a
glass Econo-column (50 cm × 5 cm) to a height of 30 cm, and the
column was successively washed with 900 mL of water, 5 and 50%
acetonitrile in water, and acetonitrile. These procedures were repeated
five times, and each of four fractions (4.5 L) was separately combined
for LSC and HPLC analysis, indicating that most of the unknown
metabolites was eluted in the third fraction. The third fraction
concentrated to remove acetonitrile was further fractionated on the
Porapack Q column by successive elutions with 750 mL of aqueous
acetonitrile with more gradual solvent gradient (0, 1, 5, 10, 15, 25, 35,
50, and 100% acetonitrile in water). The fifth and sixth fractions (15-
25% acetonitrile) were combined and subjected to the repeated
purification using HPLC by methods 3-5.
3
4
5
6
13.46
ND
3.29
9.76
2.73
0.215
ND
0.053
0.156
0.044
5.41
ND
3.97
6.51
7.04b
0.102
ND
0.075
0.123
0.133b
others
pulp extract
0.033
1
2
3
4
5
6
2.04
0.90
4.60
1.51
3.10
3.72
3.57c
0.61
4.28
ND
0.081
ND
0.014
0.074
0.024
0.050
2.70
1.68
3.28
2.20
1.07d
4.68
0.051
0.032
0.062
0.042
0.020d
0.088
0.060
others
unextractable
0.057c
0.010
untreated leaf
0.001
<0.01
<0.01
0.001
1.888
total
1.599
100.00
100.00
a ppm of 1 equivalent. b Consisted of more than four metabolites, each metabolite
below 3.70% TRR (0.070 ppm). c Consisted of more than eight metabolites, each
metabolite below 0.80% TRR (0.013 ppm). d Consisted of more than two
metabolites, each metabolite below 0.57% TRR (0.010 ppm).
Identification of Metabolites. The major unknown metabolites were
subjected to enzymatic hydrolysis (cellulase, 37 °C, overnight) in 10
mM phosphate buffer at pH 5. Separately, they were isolated from grape
fruits and subjected to LC-APCI-MS and NMR analyses. Acetylation
was conducted by treating the isolated metabolites with anhydrous acetic
acid in the trace amount of dry pyridine, and methylation was conducted
by putting droplets of trimethylsilyldiazomethane into metabolite
dissolved in methanol/dichloromethane (1/1, v/v), according to the usual
manners. Acetylated and methylated unknown metabolites were also
subjected to LC-APCI-MS and NMR analyses. Finally, the identity of
the unknown metabolites was confirmed by HPLC and TLC cochro-
matographies with nonradiolabeled reference standards.
Table 4. Distribution of Radioactivities and Metabolites of [14C]-1 with
Leaf Treatment
[14C-phenyl]-1
[14C-isopropyl]-1
a
a
%TRR
ppm
%TRR
ppm
surface rinse
75.741
1
94.90
ND
95.07
0.55
61.180
0.349
others
ND
extract
0.516
ND
1
2
0.65
ND
0.73
ND
0.466
ND
3
4
0.49
ND
0.391
ND
ND
ND
ND
ND
RESULTS
5
6
0.57
ND
0.458
ND
0.43
ND
0.273
ND
Distribution of Metabolites. In the case of fruit and leaf
treatments at a rate of 500 g a.i. ha-1, the recovered radioactivity
was 73.0-77.2 and 77.3-78.7% of the applied 14C, respectively.
The radioactive residues in untreated leaves with the fruit
treatment were less than 0.01% of the applied 14C (1 ppb), and
those in untreated leaves and fruits were less than 0.01% (0.01
ppm) when the leaf was treated. These results indicated the
insignificant translocation of radioactivity from treated portions,
and the unrecovered 14C at harvest was likely to be lost by
vaporization.
Distribution of radioactivity and metabolites in the surface
rinse, juice, and pulp in fruit treated with [14C]-1 at a rate of
500 g a.i. ha-1 is summarized in Table 3. No typical difference
in distribution was observed between the two radiolabels. Most
of the radioactivity in the surface rinse remained as intact 1
(50.7-55.0% total radioactive residue, TRR). Metabolites such
as 2-6 were mainly detected from juice and pulp extracts.
Within the whole grape, 1 remained as the major residue (52.8-
60.9% TRR). Major metabolites detected were 3 (8.1-18.1%
TRR, 0.153-0.289 ppm), 5 (6.4-7.3% TRR, 0.102-0.137
others
unextractable
2.33a
1.06
1.856a
0.847
2.53b
0.69
1.620b
0.444
untreated leaf
<0.010
<0.01
ND
<0.01
ND
<0.010
ND
untreated fruit
ND
total
79.809
100.00
100.00
64.332
a ppm of 1 equivalent. b Consisted of more than 17 metabolites, each metabolite
below 0.46% TRR (0.367 ppm). c Consisted of more than eight metabolites, each
metabolite below 0.55% TRR (0.348 ppm).
ppm), and 6 (8.7-13.5% TRR, 0.165-0.216 ppm). Minor
metabolites 2 and 4 were both detected below 1.7% TRR.
Meanwhile, the distribution of radioactivity from the treated
leaves was different from that of the fruits (Table 4). Most of
the radioactivity in the surface rinse remained as intact 1 (94.9-
95.1% TRR). Metabolites such as 3 and 5 were mainly detected
from the extract. Within the whole leaf, 1 was major residue

Metabolism of Fungicide Diethofencarb in Grape
J. Agric. Food Chem., Vol. 51, No. 18, 2003 5333
(95.6-95.8% TRR) and the amounts of metabolites 3 and 5
were less than 0.6% TRR. These results indicated more
penetration of 1 followed by metabolic degradation in fruits as
compared with leaves.
Identification of Metabolites 2 and 3. The chemical structure
of 2 was easily confirmed by HPLC and TLC cochromatogra-
phies of the isolated metabolite with the synthetic standard. In
the case of 3, the corresponding HPLC fraction of extracts was
collected and divided into two portions. One portion was
incubated with cellulase, and a new peak corresponding to 2
appeared via concomitant decrease of the peak due to 3. The
formation of 2 as an aglycone was also confirmed by two-
dimensional TLC cochromatography with the authentic standard.
To further investigate the chemical nature of 3, the remaining
portion was subjected to LC-APCI-MS analysis in a negative
ion mode. Metabolite 3 was shown to have a molecular weight
of 401, as evidenced from the MS peak at m/z 400 ([M - H]-).
These results strongly suggest glucose, or equivalent, as the
conjugation moiety. Definitive identification of 3 was achieved
by synthesizing the glucose conjugate of 2 and carrying out
direct comparison of the isolated 3 with this synthetic standard
by HPLC and TLC cochromatographies.
Identification of Metabolites 4 and 5. Metabolite 5 was
incubated with cellulase, and a new peak corresponding to 4
appeared via concomitant decrease of the peak due to 5. The
formation of 4 was also confirmed by two-dimensional TLC
cochromatography with the authentic standards. LC-APCI-MS
analysis in a negative ion mode showed that the molecular
weight of 5 isolated from the extracts was most likely to be
521, as demonstrated by the MS peak at m/z 520 [M - H]-.
The methylation of 5 was attempted by using trimethylsilyl-
diazomethane, and the resultant product was subjected to LC-
APCI-MS in a negative ion mode. The introduction of the two
methyl groups was shown by the MS peak at m/z 548 [M -
H]-. When 5 was acetylated by the usual manners, the MS peak
at m/z 730 [M - H]- was observed, indicating the introduction
of five acetyl groups into the molecule. To estimate the
functional groups of 5 in detail, the successive derivatization
(methylation followed by acetylation) was conducted. LC-APCI-
MS in a negative ion mode showed the MS peak at m/z 716 [M
- H]-; hence, the two methyl and four acetyl groups were
formed to be added to the molecule. Assuming that the aglycone
of 5 is 2, one methyl group may be introduced via methylation
of the activated phenol at the 4-position, while the other one
may be introduced from methylation of a carboxylic acid
functionally associated with a conjugating moiety. The acetyl
groups introduced are probably due to hydroxy functionalities
of a monosaccharide, which is also consistent with observed
molecular weight. The NMR spectrum of the methylated
derivative of 5 was measured as shown in Figure 2. Only two
aromatic protons were observed at 6.95 and 7.05 ppm with the
coupling constant of 2.0 Hz, indicating the presence of two
aromatic protons in the meta-position; hence, some metabolic
conversion at the 5-position of the phenyl ring was most likely.
By the way, the isopropyl and one ethoxy groups were found
to remain intact. The methyl groups appeared as a doublet at
1.25 ppm with an accompanying isopropyl proton at 5.00 ppm,
and the ethoxy protons at 1.40 (triplet) and 4.05 ppm (quartet)
were also present. A large number of coupled resonance possibly
due to glucopyranosyl protons appeared at 3.10-4.60 ppm,
secondary hydroxyl groups, implying that 5 was a sugar
conjugate. The two sharp singlets appeared at 3.75 and 3.80
ppm due to methylation of 5, possibly implying the presence
of the methyl ester and methoxy groups and, hence, the for-
Figure 2. 1H NMR spectrum of methylated 5 (3-{3-ethoxy-2-hydroxy-5-
[(isopropoxycarbonyl)amino]phenylthio}-2-â-glucopyranosyloxypropionic acid).
Metabolite 5 was isolated from grape fruits.
mation of a more complex conjugate than a simple sugar deriv-
ative. Taking into account all of the spectrometric evidence to-
gether with the conjugation form of other pesticides (19-22),
the chemical structure of 5 was proposed. Finally, the chemical
identity of 5 as 3-{3-ethoxy-2-hydroxy-5-[(isopropoxycarbonyl)-
amino]phenylthio}-2-â-glucopyranosyloxypropionic acid was
definitely confirmed by both HPLC and TLC cochromatogra-
phies with the reference standard as prepared by the route 3 in
Figure 1.
Considering the metabolic pathway, 4 was assumed to exist
in grape as an intermediate metabolite of 5. Therefore, the
synthetic standard of 4 was used to confirm its existence in the
extract of grape by HPLC and TLC cochromatographies. As a
result, the metabolite corresponding to 4 was detected in grape
fruit.
Identification of Metabolite 6. Metabolite 6 was incubated
with cellulase; however, no degradation of 6 was observed. The
LC-APCI-MS spectrum of the purified 6 in a negative ion mode
showed the same molecular weight as 5 (m/z 520 [M - H]-),
indicating isomeric structure. LC-MS analysis in a negative ion
mode showed the MS peak of the methylated 6 at m/z 534 ([M
- H]-) and that of methylated and acetylated derivative of 6 at
m/z 744 ([M - H]-). The molecular weight of the latter
derivative indicated the addition of five acetyl unites to the
methylated molecule. These results together with the data on
metabolite 5 implied that 6 was a monosaccharide derivative,
but only four acetyl groups were possibly to be introduced.
Therefore, the metabolite should have an additional center
sensitive to acetylation (hydroxy group of thiolactic acid attached
to benzene ring via C-S bond). Considering all of the above
spectrometric evidence and taking the chemical structure of 5
into account, the possible chemical structure of 6 was 3-{3-
ethoxy-2-â-glucopyranosyloxy-5-[(isopropoxycarbonyl)amino]-
phenylthio-2-hydroxypropionic acid}], which was definitely
confirmed by both HPLC and TLC cochromatographies with

5334 J. Agric. Food Chem., Vol. 51, No. 18, 2003
Fujisawa et al.
Figure 3. Proposed metabolic pathway of 1 in the grape plants. F, fruit treatment; L, leaf treatment.
the synthetic standard. On the basis of these results, the meta-
bolic pathway of 1 in grape plants is proposed in Figure 3.
had discussed the specificity of reacting positions on benzene
ring of N,N′-quinonediimine against a nucleophilic substitution.
The regioselectivity of a nucleophilic addition to quinonediimine
was examined by Hu¨ckel molecular orbital (HMO) theory, but
a clear explanation could not be obtained due to the limitations
of HMO calculations dealing only with π-electrons. However,
assuming that all resonance integrals in a quinonediimine and
its intermediates in nucleophilic addition were equal to the
benzene resonance integral, it was proposed that the 1-, 3-, and
5-positions of N,N′-quinonediimine ring were highly susceptible
to nucleophilic substitution. In our study, the fully optimized
molecular geometries of 2 and its quinoneimine form in lower
energy states were estimated by the AM1 method (31) in
WINMOPAC program (version 2.0, Fujitsu, Ltd.). The distribu-
tion of the lowest unoccupied molecular orbital (LUMO) of 2
at the 5-position (12.1%) vs the 2- and 6-positions (15.9 and
2.2%, respectively) implied the higher reactivity to the nucleo-
philic attack at this position, while the examination of the
LUMO of the quinoneimine form did not show any particular
preference at the 5-position (23.4%) over the other ones (18.8
and 26.3% for the 2- and 6-positions, respectively). This might
be elucidated by indicating the steric imposition on a substrate
in the neighborhood of a reactive site of enzyme, which affects
and/or determines the selectivity of reacting position on the
phenyl ring of 2. Actually, it has been reported in the metabolism
of acetaminophen (32-35) that the 3- and 5-positions of
quinoneimine are highly reactive to the nucleophilic attack by
glutathione (GSH).
DISCUSSION
A great difference in the metabolic pattern of 1 was observed
between the fruit and the leaf treatments. The test compound
(1) was less degraded in the leaf while it was extremely
metabolized mainly to conjugated products in the fruit. This
may be, at least in part, due to the difference in the amounts
and the kinds of waxes existing on the surface of plant, which
affect the penetration of compound to the inner cell where the
various metabolism by enzymes actually occurs. Furthermore,
the difference in distribution of metabolic enzymes may exist
as fruit and leaf are compared.
The mechanism on formation of the thiolactic acid conjugate
was postulated from the various works conducted with acetami-
nophen (N-acetyl-p-aminophenol). Acetaminophen is a widely
used analgesic medicine, which has p-aminophenol as a
fundamental chemical structure. The p-aminophenol moiety is
considered to undergo two electron oxidation, and it is easily
transformed to a highly reactive benzoquinoneimine. The in vitro
oxidation of acetaminophen has been extensively investigated
using various enzymes such as cytochrome P-450, prostaglandin
H synthase (PHS), and peroxidase (23-29), and the three
possible routes to generate benzoquinoneimine from p-ami-
nophenol by enzymatic oxidation was postulated. First, ac-
etaminophen is oxidized with peroxidase or PHS via one
electron oxidation to produce N-acetyl-p-benzosemiquinone-
imine and the resulting semiquinoneimine is disproportionated
to give acetaminophen and p-benzoquinoneimine. Second,
acetaminophen is oxidized with P-450 and PHS directly via two
electron oxidation to form benzoquinoneimine. Third, acetami-
nophen is oxidized with P-450 to form hydroxylated intermedi-
ate (N-acetyl-N-hydroxy-p-aminophenol), subsequently followed
by dehydration to form benzoquinoneimine. The main metabo-
lite of 1 at the early stage of metabolism is most likely to be 2,
possessing the common p-aminophenol structure to acetami-
nophen. Therefore, 2 would be similarly oxidized to form the
corresponding benzoquinoneimine as an intermediate. Actually,
2 could be chemically oxidized by lead tetraacetate very rapidly
to yield the benzoquinoneimine. Additionally, Finley et al. (30)
In summary, 1 primarily underwent O-deethylation at the
4-position of the phenyl ring to form 2. Successively, 2 was
likely to be converted via two electron oxidation to the corre-
sponding benzoquinoneimine whose 5-position of the ring was
attached by cysteine. Cysteine is the most probable nucleophile,
since it is known to be formed by catabolism of the correspond-
ing conjugates of GSH, a major ingredient in grape fruit (36).
Then, the amino group of cysteine was considered to be
transformed to the pyruvate structure as an intermediate by
transaminase and finally to the thiolactic acid (Figure 3).
Incidentally, Lamoureux et al. first reported the existence of
a thiolactic acid conjugate in various plants (19-22). They
detected thiolactic acid conjugates in peanut, corn, cotton,

Metabolism of Fungicide Diethofencarb in Grape
J. Agric. Food Chem., Vol. 51, No. 18, 2003 5335
(17) Fernando, R. C.; Calder, I. C.; Ham, K. N. Studies on the
mechanism of toxicity of acetaminophen. Synthesis and reactions
of N-acetyl-2,6-dimethyl- and N-acetyl-3,5-dimethyl-p-benzo-
quinone imines. J. Med. Chem. 1980, 23, 1153-1158.
(18) Lee, H. H.; Palmer, B. D.; Denny, W. A. Reactivity of quinone
imine and quinone diimine metabolites of the antitumor drug
amsacrine and related compounds to nucleophiles. J. Org. Chem.
1988, 53, 6042-6047.
(19) Lamoureux, G. L.; Gouot, J. M.; Davis, D. G.; Rusness, D. G.
Pentachloronitrobenzene metabolism in peanut. 3. Metabolism
in peanut cell suspension cultures. J. Agric. Food Chem. 1981,
29, 996-1002.
soybean, and spruce using mainly cell suspension culture. For
example, EPTC, propachlor, and metolachlor were metabolized
in cotton and corn to the corresponding (O-malonyl)thiolactic
acid conjugates. However, the identification of the conjugate
was only performed from the results of mass spectrometric
analysis; thus, ambiguity remains for their chemical identities.
Therefore, our study gives the first concrete evidence of the
thiolactic acid conjugate by the definite identification by HPLC
and TLC cochromatographies with the synthetic standards,
together with LC-APCI-MS and NMR analyses. Additionally,
as shown with the chemical structure of 6, this is a new type of
conjugation form in metabolites that has not been reported; that
is, two natural products are conjugated within a different position
of the same aglycone.
(20) Lamoureux, G. L.; Rusness, D. G. EPTC metabolism in corn,
cotton, and soybean: Identification of a novel metabolite derived
from the metabolism of a glutathione conjugate. J. Agric. Food
Chem. 1987, 35, 1-7.
(21) Lamoureux, G. L.; Rusness, D. G. Propachlor metabolism in
soybean plants, excised soybean tissues, and soil. Pestic.
Biochem. Physiol. 1989, 34, 187-204.
(22) Lamoureux, G. L.; Rusness, D. G.; Schro¨der, P.; Rennenberg,
H. Diphenyl ether herbicide metabolism in a spruce cell
suspension culture: The identification of two novel metabolites
derived from a glutathione conjugate. Pestic. Biochem. Physiol.
1991, 36, 291-301.
(23) Hinson, J. A.; Monks, T. J.; Hong, M.; Highet, R. J.; Pohl, L.
R. 3-(Glutathion-S-yl)acetaminophen: a biliary metabolite of
acetaminophen. Drug Metab. Dispos. 1982, 10, 47-50.
(24) Morgan, E. T.; Koop, D. R.; Coon, M. J. Comparison of six
rabbit liver cytochrome P-450 isozymes in formation of a reactive
metabolite of acetaminophen. Biochem. Biophys. Res. Commun.
1983, 112, 8-13.
(25) Albano, E.; Rundgren, M.; Harvison, P. J.; Nelson, S. D.;
Molde´us, P. Mechanism of N-acetyl-p-benzoquinone imine
cytotoxicity. Mol. Pharmacol. 1985, 28, 306-311.
(26) Potter, D. W.; Miller, D. W.; Hinson, J. A. Horseradish
peroxidase-catalyzed oxidation of acetaminophen to intermediates
that form polymers or conjugate with glutathione. Mol. Phar-
macol. 1986, 29, 155-162.
(27) Potter, D. W.; Hinson, J. A. Reactions of N-acetyl-p-benzo-
quinone imine with reduced glutathione, acetaminophen, and
NADPH. Mol. Pharmacol. 1986, 30, 33-41.
(28) Potter, D. W.; Hinson, J. A. Mechanisms of acetaminophen
oxidation to N-acetyl-p-benzoquinone imine by horseradish
peroxidase and cytochrom P-450. J. Biol. Chem. 1987, 262, 966-
973.
(29) Potter, D. W.; Hinson, J. A. The 1- and 2-electron oxidation of
acetaminophen catalyzed by prostaglandin H synthase. J. Biol.
Chem. 1987, 262, 974-980.
(30) Finley, T. K.; Tong, J. L. K. Quinonediimine and related
compounds. In The Chemistry of the Carbon-Nitrogen Double
Bond; Patai, S., Ed.; Interscience Publishers, London, 1970; pp
663-729.
(31) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P.
AM1: A new general purpose quantum mechanical molecular
model. J. Am. Chem. Soc. 1985, 107, 3902-3909.
(32) Gemborys, M. W.; Gribble, G. W.; Mudge, G. H. Synthesis of
N-hydroxyacetaminophen, a postulated toxic metabolite of
acetaminophen, and its phenolic sulfate conjugate. J. Med. Chem.
1978, 21, 649-652.
(33) Miner, D. J.; Kissinger, P. T. Evidence for involvement of
N-acetyl-p-quinoneimine in acetaminophen metabolism. Bio-
chem. Pharmacol. 1979, 28, 3285-3290.
(34) Buckpitt, A. R.; Rollins, D. E.; Nelson, S. D.; Franklin, R. B.;
Mitchell, J. R. Quantitative determination of the glutathione,
cycteine, and N-acetyl cysteine conjugates of acetaminophen by
high-pressure liquid chromatography. Anal. Biochem. 1977, 83,
168-177.
LITERATURE CITED
(1) Takahashi, J.; Nakamura, S.; Noguchi, H.; Kato, T.; Kamoshita,
K. Fungicidal activity of N-phenylcarbamate against benzimi-
dazole resistant fungi. J. Pestic. Sci. 1988, 13, 63-69.
(2) Takahashi, J.; Kirino, O.; Takayama, C.; Nakamura, S.; Noguchi,
H.; Kato, T.; Kamoshita, K. Quantitative structure-activity
relationships of fungicidal N-(3,4-diethoxyphenyl)carbamates. J.
Pestic. Sci. 1988, 13, 587-593.
(3) Takahashi, J.; Kirino, O.; Takayama, C.; Nakamura, S.; Noguchi,
H.; Kato, T.; Kamoshita, K. Quantitative structure-activity
relationships of the fungicidal methyl N-phenylcarbamate. Pestic.
Biochem. Physiol. 1988, 30, 262-271.
(4) James, C. S.; Prendeville, G. N. Metabolism of chlorpropham
(isopropyl m-chlorocarbanilate) in various plant species. J. Agric.
Food Chem. 1969, 17, 1257-1260.
(5) Still, G. G.; Mansager, E. R. Metabolism of isopropyl 3-
chlorocarbanilate by soybean plants. J. Agric. Food Chem. 1971,
19, 879-884.
(6) Still, G. G.; Mansager, E. R. Aryl hydroxylation of isopropyl-
3-chlorocarbanilate by soybean plants. Phytochemistry 1972, 11,
515-520.
(7) Still, G. G.; Mansager, E. R. Soybean shoot metabolism of
isopropyl-3-chlorocarbanilate: Ortho and para aryl hydroxyla-
tion. Pestic. Biochem. Physiol. 1973, 3, 87-95.
(8) Still, G. G.; Mansager, E. R. Metabolism of isopropyl-3-
chlorocarbanilate by cucumber plants. J. Agric. Food Chem.
1973, 21, 787-791.
(9) Still, G. G.; Rusness, D. G. S-cysteinyl-hydroxychlorpropham:
Formation of the S-cysteinyl conjugate of isopropyl-3′-chloro-
4′-hydroxycarbanilate in oat (AVena satiVa L.). Pestic. Biochem.
Physiol. 1977, 7, 210-219.
(10) Heikes, D. L. Mass spectral identification of a metabolite of
chlorpropham in potatoes. J. Agric. Food Chem. 1985, 33, 246-
249.
(11) Still, G. G.; Mansager, E. R. Metabolism of isopropyl carbanilate
by soybean plants. Pestic. Biochem. Physiol. 1973, 3, 289-299.
(12) Still, G. G.; Mansager, E. R. Alfalfa metabolism of propham.
Pestic. Biochem. Physiol. 1975, 5, 515-522.
(13) Zurqiyah, A. A.; Jordan, L. S.; Jolliffe, V. A. Metabolism of
isopropyl carbanilate (propham) in alfalfa grown in nutrient
solution. Pestic. Biochem. Physiol. 1976, 6, 35-45.
(14) Castello, A.; Marquet, J.; Moreno-Manas, M.; Sirera, X. The
nucleophilic aromatic photosubstitution in photoaffinity labeling.
A model study of a cycloheximide derivative. Tetrahedron Lett.
1985, 26, 2489-2492.
(15) Schmidt, R. R.; Zimmermann, P. Synthesis of glycosphingolipids
and psychosines. Angew. Chem., Int. Ed. Engl. 1986, 25, 725-
726.
(16) Hattori, K.; Takido, T.; Itabashi, K. The synthesis of thiols by
use of N,N-dimethylthioformamide. Nippon Kagaku Kaishi 1979,
101-105.

5336 J. Agric. Food Chem., Vol. 51, No. 18, 2003
Fujisawa et al.
(35) Pascoe, G. A.; Calleman, C. J.; Baille, T. A. Identification of
S-(2,5-dihydroxyphenyl)-cysteine and S-(2,5-dihydroxyphenyl)-
N-acetyl-cysteine as urinary metabolites of acetaminophen in the
mouse. Evidence for p-benzoquinone as a reactive intermediate
in acetaminophen metabolism. Chem.-Biol. Intract. 1988, 68,
85-98.
juice and wine using pre-column derivatization with fluorescence
detection. Food Chem. 2000, 68, 475-480.
Received for review April 11, 2003. Revised manuscript received June
11, 2003. Accepted June 17, 2003.
(36) Park, S. K.; Boulton, R. B.; Noble, A. C. Automated HPLC
analysis of glutathione and thiol-containing compounds in grape
JF034367+