Toxicology and Applied Pharmacology 174, 146 –152 (2001)
doi:10.1006/taap.2001.9207, available online at http://www.idealibrary.com on
The Synergistic Upregulation of Phase II Detoxification Enzymes by
Glucosinolate Breakdown Products in Cruciferous Vegetables
Chu Won Nho and Elizabeth Jeffery 1
Division of Nutritional Sciences, University of Illinois at Urbana–Champaign, Urbana, Illinois 61801
Received January 25, 2001; accepted April 20, 2001
The Synergistic Upregulation of Phase II Detoxification En-
zymes by Glucosinolate Breakdown Products in Cruciferous Veg-
etables. Nho, C. W., and Jeffery, E. (2001). Toxicol. Appl. Phar-
macol. 174, 146 –152.
Cruciferous vegetables contain secondary metabolites termed
glucosinolates that break down to products that upregulate he-
patic detoxification enzymes. We have previously shown that a
mixture of four major glucosinolate breakdown products from
Brussels sprouts interact to produce synergistic induction of phase
II detoxification enzymes. Here we tested the hypothesis that this
synergism is at the level of transcription and is due to the inter-
action between the oral bifunctional inducer, indole-3-carbinol
(I3C), and monofunctional inducer, crambene (1-cyano 2-hydroxy
3-butene). Adult male rats were treated by gavage with either corn
oil (vehicle); crambene (50 mg/kg), I3C (56 mg/kg), or a mix of
crambene and I3C at the doses shown. Given orally, I3C alone and
crambene with I3C caused significant induction of CYP1A activity
and CYP1A1 mRNA levels, whereas crambene alone had no
significant effect on CYP1A activity or mRNA levels. Crambene
and I3C individually caused induction of glutathione S-trans-
ferase (GST) and quinone reductase (QR) activity. The mixture of
crambene and I3C caused induction of GST and QR that was
significantly greater than the sum of the induction by individual
treatments. Upregulation of total GST activity was not as great as
that of QR, possibly because some subunits did not show this
effect. GST Ya2 mRNA showed a synergistic upregulation by
crambene and I3C, while Yc1 and Yc2 showed only an additive
response. We speculate that this different regulation is partly due
to differences in gene sequences within the antioxidant response
element and xenobiotic response element in the regulatory region
of GST Ya2 compared to those within the regulatory region of the
Yc1/Yc2 subunits. © 2001 Academic Press
Key Words: detoxification enzymes; glucosinolates; crambene;
indole-3-carbinol; synergism; quinone reductase; glutathione S-
transferases.
Numerous epidemiological studies have shown that crucif-
erous vegetables have a role in dietary prevention of cancers
1
To whom correspondence should be addressed at Division of Nutritional
Sciences, National Soybean Research Center, 1101 W. Peabody Dr., Urbana,
Illinois 61801. Fax: (217) 333-8046; E-mail: ejeffery@uiuc.edu.
0041-008X/01 $35.00
Copyright © 2001 by Academic Press
All rights of reproduction in any form reserved.
(Steinmetz and Potter, 1991). This protective effect may be due
to their glucosinolate content. Brassicas, including all types of
cabbages, broccoli, cauliflower, and Brussels sprouts, are a
genus of the family Crucifeacea and have relatively high glu-
cosinolate content. Glucosinolates can be hydrolyzed by the
enzyme myrosinase (thioglucoside glucohydrolase; EC 3.2.3.1;
Verhoeven et al., 1997). Myrosinase is present in plant cells in
a separate compartment from glucosinolates. When the plant
cells are damaged, e.g., by cutting or chewing, the myrosinase
comes into contact with the glucosinolates and hydrolysis
occurs. Upon hydrolysis of glucosinolates, three major classes
of breakdown products can be formed: isothiocyanates, nitriles,
and thiocyanates. Many of these products are bioactive, caus-
ing upregulation of detoxification enzymes. The bioactive
agents have been divided into two groups, termed mono- and
bifunctional inducers. Monofunctional inducers upregulate a
number of phase II detoxification enzymes, including quinone
reductase (QR) and glutathione S-transferases (GSTs). Bifunc-
tional inducers upregulate a similar array of phase II enzymes,
but in addition they upregulate a few phase I enzymes, includ-
ing CYP1A1. Because CYP1A1 is known to bioactivate poly-
cyclic hydrocarbon carcinogens, bifunctional inducers are con-
sidered less closely associated with cancer prevention than are
monofunctional inducers (Prochaska and Talalay, 1988). In
addition, some monofunctional inducers have been found to act
as competitive inhibitors of cytochrome P450 activation of
carcinogens (Stresser et al., 1995; Teel and Huynh, 1998).
GSTs are a major group of phase II detoxification enzymes,
mediating the conjugation between GSH and a variety of
xenobiotics. The cytosolic GSTs consist of at least five classes,
including ␣⫺, ␮⫺, ␲⫺, ␴⫺, and ␶⫺GST. Each class contains
a number of subunits and active GST consists of homo- and
heterodimers within the class. In liver, many of the GST
isozymes are expressed constitutively; many more are induc-
ible (Hayes and Pulford, 1995). The mechanism of induction of
␣⫺GST subunit Ya2 has been studied extensively; GST Ya2 is
induced by many compounds, including monofunctional in-
ducers via a DNA-regulatory region termed the antioxidant
response element (ARE) (Nguyen and Pickett, 1992). The
␣⫺GST subunit Yc1 is reported to be both constitutively
expressed in rat liver and inducible when rats are fed mono-
146
 SYNERGISTIC UPREGULATION OF DETOXIFICATION ENZYMES
functional and bifunctional inducers. The GST subunit Yc2 is
only found constitutively in the liver of the rat fetus and young
rats but, like Yc1, is induced by monofunctional and bifunc-
tional inducers in the adult liver (Hayes and Pulford, 1995).
Crambene (1-cyano-2-hydroxy-3-butene), formed through
the hydrolysis of progoitrin, is an aliphatic nitrile occurring
naturally in cruciferous vegetables including Brussels sprouts,
broccoli, and cauliflower. Seeds from the plant Crambe abys-
sinica are the richest source of crambene (Fenwick et al.,
1983). Crambene is known to upregulate GST synthesis in
liver and other organs (March et al., 1998).
Indole-3-carbinol (I3C) is formed through the hydrolysis of
glucobrassicin. I3C and its acid condensation products formed
in the stomach following ingestion are known to have anticar-
cinogenic properties (Takahashi et al., 1995b). The mechanism
may be by induction of phase I P450-dependent and/or phase
II detoxification enzymes (Wortelboer et al., 1992; Chen et al.,
1996) and is thought to involve a direct effect of I3C acid
condensates as ligand bound to the Ah receptor interacting with
a sequence in the regulatory region of these detoxification
genes, the xenobiotic response element (XRE) (Jellinck et al.,
1993).
Recently, the individual and collective effects of specific
glucosinolate breakdown products on induction of detoxifica-
tion enzymes were examined (Staack et al., 1998). Crambene,
I3C, iberin, and phenethylisothiocyanate (PEITC), a mixture
formulated in the relative proportions found in Brussels
sprouts, or the four individual compounds, were fed to rats for
7 days. At the doses used, only crambene and I3C caused
significant induction of the phase II enzymes QR and GST
when given individually. The group receiving the mix showed
significantly greater activity than the sum of the individual
treatments. Since only crambene and I3C caused significant
induction individually, we hypothesized that these two com-
pounds were responsible for the synergistic induction. The
dose of crambene was chosen based on previous work showing
that 30 –100 mg/kg caused a dose-related response and that
GST was stable at 7 days (Wallig et al., 1992; March et al.,
1998).
Here we report that a mix of I3C and crambene, at the same
doses as used in the previous study (Staack et al., 1998),
caused a synergistic increase in activity for both QR and GST.
When upregulation of GST was studied in detail, we found
that, while induction of all three subunits studied was at the
transcriptional level, transcription was only synergistically up-
regulated for one of three subunits evaluated.
MATERIALS AND METHODS
Chemicals. Crambene was isolated and purified from the seeds of Crambe
abyssinica (Wallig et al., 1992). I3C was purchased from Sigma (St. Louis,
MO).
Animals, diet, and treatments. Adult male CDF 344 (crl/BR) rats (Charles
River Laboratories, Wilmington, MA), weighing 150 –200 g, were used in all
147
experiments. Animals were acclimated for a week on a regular chow diet and
for 3 days on a AIN76A powdered diet purchased from Teklad (Madison, WI),
which was continued through the experiment. Rats (four animals/group) re-
ceived either crambene (50 mg/kg), I3C (56 mg/kg), or a mixture of crambene
and I3C at these same doses, given by gavage daily for 7 days. Two groups
received vehicle (corn oil) alone: a control group that was fed ad libitum, and
a second group that was pair fed to the group receiving the crambene ⫹ I3C
mixture.
Preparation of hepatic cytosol and microsomes. Rats were killed by
carbon dioxide asphyxiation followed by cervical dislocation. The livers were
perfused with 1.15% KCl and homogenized in 4 vol of buffer (0.15 M KCl,
0.25 M K 2HPO 4/KH 2O 4, pH 7.25). Homogenates were centrifuged at 10,000g
for 20 min at 4°C. The supernatant was centrifuged at 105,000g for 1 h at 4°C
to separate microsomal and cytosolic fractions. Microsomal fractions were
resuspended in 0.1 M freezing buffer (Na 2HPO 4/NaH 2PO 4 buffer, pH 7.4)
containing 1.0 mM dithiothreitol and 0.25 M sucrose. Aliquots of cytosol and
microsomes were snap frozen in liquid nitrogen and stored at ⫺80°C until
needed.
Measurement of enzyme activities. Ethoxyresorufin O-deethylase
(EROD) activity was determined essentially as published (Pohl and Fouts,
1980). A 0.2-ml microsomal suspension (2 mg of protein/ml) was incubated
with 10 ␮l ethoxyresorufin (4.2 ␮M in methanol) for 4 min and the reaction
was stopped by adding 2 ml methanol. Resorufin formation was measured by
fluorimetry using an excitation wavelength of 550 nm and an emission wave-
length of 580 nm. Resorufin produced was compared to a standard curve of
known concentration of resorufin (Sigma). Total GST activity was measured
spectrophotometrically using 1-chloro-2,4-dinitrobenzene (CDNB) as a sub-
strate (Habig et al., 1974). Cytosolic samples were diluted 1:100, which gave
rates in the linear range, and were measured in triplicate. The reaction was
started by adding CDNB, and initial velocity was measured at 340 nm for 90 s.
Units of specific activity are reported as nmol CDNB conjugate formed per mg
cytosolic protein per min. QR activity was measured spectrophotometrically as
the dicumorol-sensitive reduction of 2,6-dichlorophenolindophenol (DCPP)
(Benson et al., 1980). The reaction was initiated by adding 10 ␮l of DCPP (12
mM) to 300 ␮l of cytosol in 2.65 ml reaction buffer. The initial velocity of the
reaction was measured at 600 nm for 90 s. Assays were performed in triplicate
with and without dicumorol. The units of specific activity are reported as nmol
DCPP reduced/mg cytosolic protein/min. All assays were performed at room
temperature.
Northern blotting. Total RNA was extracted from frozen livers according
to the manufacture’s instructions using Trizol reagent (Life Technologies). The
concentration and purity of RNA were determined spectrophotometrically at
260 nm and by the 260/280 nm ratio, respectively. Fifteen or 20 ␮g of total
RNA was separated on a 1.5% agarose/formaldehyde gel, blotted onto a
nitrocellulose membrane, and hybridized with a 32P-random-labeled cDNA
probe at 65°C for 2 h in a hybridization oven. After hybridization, blots were
washed twice at room temperature with wash buffer [2⫻ SSC and 0.1%
sodium dodecyl sulfate (SDS)] for 5 min each. The blots were exposed to
X-ray films overnight at ⫺80°C. Autoradiographs were scanned and quanti-
tatively analyzed by densitometry. The cDNA probe for rat GST Ya2 was a gift
from Dr. Cecil Pickett, Schering–Plough Research Institute (Kenilworth, NJ)
and the cDNA probe for human CYP1A1 was a gift from Dr. Roland Wolf,
University of Dundee (Dundee, UK).
Western blot analysis. Cytosolic proteins (20 ␮g) were separated on a
12% SDS–polyacrylamide gel and transferred onto polyvinylidene difluoride
membranes for 1 h at room temperature with 0.25– 0.3 A. After blocking with
5% bovine serum albumin (BSA), the membrane was incubated with primary
antibodies (monoclonal Yc1 or Yc2 antibody) for 1 h at room temperature in
buffer containing 1% BSA. Goat anti-rabbit alkaline phophatase-conjugated
IgG was used as secondary antibody (Sigma). The blots were developed using
a color substrate mix (5-bromo-4-chloro-3-indoyl phosphate/nitro blue tetra-
zolium; Sigma) for alkaline phosphatase. The monoclonal anti-rat GST Yc1
and Yc2 antibodies (Buetler et al., 1995) were generous gifts from Dr. J.Hayes,
Ninewells Hospital (Dundee, UK).
148 NHO AND JEFFERY

Statistical analysis. Data were analyzed for treatment effects across

groups using one-way analysis of variance (ANOVA). Where a significant
effect was found, Fisher’s least significant difference test for multiple com-
parisons was used to identify differences between groups. p ⬍ 0.05 was
considered to be significant. To determine synergism, we used the method of
contrasts, which compares treatments 2 and 3 vs treatment 4, where crambene
alone was treatment 2, I3C alone was treatment 3, and the mixed treatment
crambene ⫹ I3C was treatment 4. The method of contrasts was run from a SAS
program, version 8.0, using different variances and a p ⬍ 0.05 was considered
to be significant.

RESULTS

Induction of Detoxification Enzyme Activities in Vivo

Hepatic GST enzyme activity was induced 1.4- and 1.5-fold
by crambene and I3C, respectively. The mixture of crambene
⫹ I3C caused a 2.1-fold induction in GST activity, which was
significantly greater than additive (p ⬍ 0.05; Fig. 1A). Hepatic
QR enzyme activity was induced 1.8- and 2.1-fold by cram-
bene and I3C, respectively. The mixture of crambene ⫹ I3C
caused a 4-fold induction in QR activity, which was signifi-
cantly greater than additive (p ⬍ 0.05; Fig. 1B). Hepatic
EROD activity was not significantly increased by crambene but
was significantly increased 8-fold by I3C. The mixture of
crambene ⫹ I3C also caused an 8-fold increase in EROD
activity, which was not significantly greater than I3C alone
(Fig. 1C). In agreement with the measurement of activity,
CYP1A1 mRNA was not significantly increased by crambene
alone. In contrast, I3C alone and the mixture of crambene ⫹
I3C caused 32-fold and 30-fold increases, respectively, in the
level of mRNA encoding for CYP1A1 (Fig. 2).

Induction of mRNAs Encoding GST-␣ Class Subunits

The mRNA encoding GST subunit Ya2 was induced 2-fold
both by crambene alone and by I3C alone. The mixture of cram-
bene ⫹ I3C caused a 5-fold induction in mRNA encoding GST
subunit Ya2, which was significantly greater than additive (p ⬍
0.05; Fig. 3). Crambene, I3C, and the crambene ⫹ I3C mix
induced GST Yc1 mRNA approximately 3-, 3-, and 5.5-fold,
respectively (Fig. 4). Similarly, the GST Yc2 mRNA was induced FIG. 1. Determination of hepatic phase I and phase II enzyme activities in
2.5-, 2.4, and 4.5-fold by crambene, I3C, and crambene ⫹ I3C, rats. Effect of crambene and indole-3-carbinol on rat hepatic GST activity (A),
respectively (Fig. 5). There was no synergistic induction found for QR activity (B), and EROD activity (C). Rats were treated daily for 7 days
either GST Yc1 or Yc2 subunit, unlike GST Ya2. with crambene (CB, 50 mg/kg), I3C (56 mg/kg), or both together (MIX) and
killed on the 8th day. Control animals received corn oil and were fed ad libitum
(CO) or pair fed to the group receiving the mix (PF). GST and QR activity was
Induction of GST Yc Proteins measured in the hepatic cytosolic fraction and EROD activity in the hepatic
microsomal fraction as described under Materials and Methods. Values are
Since the nucleotide sequence of the cDNAs encoding GST
means ⫾ SD (n ⫽ 4); values bearing different superscripts differ significantly
Yc1 and Yc2 shows approximately 93% homology (Hayes et (p ⬍ 0.05). Actual mean value is shown above each bar; a dashed line across
al., 1994), cross-reaction may occur between Yc1 and Yc2 the graph identifies the control, baseline value.
cDNA probes and the corresponding mRNAs. In order to
confirm the mRNA induction for each subunit, a Western blot
was performed using rat GST Yc1 and Yc2 monoclonal anti- DISCUSSION
bodies. The induction pattern of GST Yc1 and Yc2 proteins by
different treatments closely matched the mRNA induction pro- In this study, we have shown that there is a synergistic
files (Fig. 6). induction of phase II enzymes, including GST and QR activ-
 SYNERGISTIC UPREGULATION OF DETOXIFICATION ENZYMES
FIG. 2. Effects of crambene and I3C on the expression of mRNA encod-
ing CYP1A1. (A) mRNA levels were measured by Northern blot, as described
under Materials and Methods. Fifteen micrograms of total RNA was loaded per
well and hybridized with a human CYP1A1 cDNA probe. Each lane represents
RNA of one individual animal. (B) Densitometry readings of P450 1A1 mRNA
were normalized to 28S ribosomal RNA and expressed as percentage of
control mRNA. Rats were treated daily for 7 days with crambene (CB, 50
mg/kg), I3C (56 mg/kg), or both together (MIX) and killed on the 8th day.
Control animals received corn oil and were fed ad libitum (CO) or pair fed to
the group receiving the mix (PF). Values are means ⫾ SD (n ⫽ 4); values
bearing different superscripts differ significantly (p ⬍ 0.05). Actual mean
value is shown above each bar; a dashed line across the graph identifies the
control, baseline value.
ities, in the presence of two components found in cruciferous
vegetables, crambene and I3C. In contrast, there was no syn-
ergistic induction observed in the phase I enzyme measured.
These results confirm the previous study in which there was a
synergistic phase II enzyme induction by a mixture of four
different glucosinolate breakdown products (Staack et al.,
1998). Furthermore, these results show that crambene and I3C
alone can produce the synergism.
In this study, the oral bifunctional inducer I3C showed
significant induction of P4501A activity and P4501A1 mRNA
levels in agreement with previous findings (Takahashi et al.,
1995a). I3C has been reported as a protective agent against
AFB 1 carcinogenecity in long-term studies in vivo. Increased
P4501A levels may be the reason for this, since P4501A is
responsible for metabolism of AFB 1 to less toxic and less
carcinogenic metabolites (Takahashi et al., 1995b). Such a
mechanism, however, cannot be extended to explain protection
from other carcinogens, such as polyarylhydrocarbons, which
are bioactivated by P4501A1 (Whitlock, 1999). No induction
was found by crambene alone for P4501A activity or mRNA
levels. While some glucosinolate hydrolysis products have
been reported to inhibit cytochrome P450 activity (Stresser et
149
al., 1995; Teel and Huynh, 1998), no inhibition was observed
under these dosing conditions. In the group that received both
I3C and crambene, induction of P4501A activity and P4501A1
mRNA levels was no greater than that seen with I3C alone.
This suggests that P4501A induction was solely due to I3C and
that crambene was not able to have a synergistic or enhancing
effect on P4501A induction.
While QR and GST activates both exhibited synergism in
the rats receiving both crambene and I3C, the increase was far
greater for QR than for GST. With respect to induction of GST,
the fold increase in GST activity was much less than the fold
increase in GST Ya2 mRNA level in rats receiving either I3C
or crambene. Similarly, the synergistic increase was far greater
for GST Ya2 mRNA than for total GST activity. This may be
explained by the complex make-up of cytosolic GST, which
consists of 13 subunits across five different families including
␣⫺, ␮⫺, ␲⫺, ␴⫺, and ␶⫺GST (Hayes and Pulford, 1995).
Since all 13 subunits do not respond to the same extent, when
measuring total GST activity, a synergistic response in synthe-
sis of one subunit may be dampened by activity of subunits
where synthesis is not synergistically induced. Among these
GST subunits, rat GST Ya2 has been studied in the greatest
detail and is known to be one of the major players in induction
of GST activity. Thus, the exaggerated visualization of the
synergistic effect when induction of GST Ya2 mRNA is eval-
FIG. 3. Induction of mRNA encoding GST Ya2 subunit by crambene,
I3C, or both. (A) mRNA levels were measured by Northern blot, as described
under Materials and Methods. Fifteen micrograms of total RNA was loaded per
well and hybridized with a rat GST Ya2 cDNA probe. Each lane represents
RNA of one individual animal. (B) Densitometry readings of GST Ya2 mRNA
were normalized to 28S ribosomal RNA and expressed as percentage of
control mRNA. Rats were treated daily for 7 days with crambene (CB, 50
mg/kg), I3C (56 mg/kg), or both together (MIX) and killed on the 8th day.
Control animals received corn oil and were fed ad libitum (CO) or pair fed to
the group receiving the mix (PF). Values are means ⫾ SD (n ⫽ 4); values
bearing different superscripts differ significantly (p ⬍ 0.05). Actual mean
value is shown above each bar; a dashed line across the graph identifies the
control, baseline value.
150 NHO AND JEFFERY
FIG. 4. Induction of mRNA expression of GST Yc1 by crambene, I3C, or
both. (A) mRNA levels were measured by Northern blot, as described under
Materials and Methods. Fifteen micrograms of total RNA was loaded per well
and hybridized with a rat GST Yc1 specific 400-bp cDNA probe. Each lane
represents RNA of one individual animal. (B) Densitometry readings of GST
Yc1 mRNA were normalized to 28S ribosomal RNA and expressed as per-
centage of control mRNA. Rats were treated daily for 7 days with crambene
(CB, 50 mg/kg), I3C (56 mg/kg), or both together (MIX) and killed on the 8th
day. Control animals received corn oil and were fed ad libitum (CO) or pair fed
to the group receiving the mix (PF). The shaded portion of each bar represents
control levels of enzyme activity and the open portion of each bar represents
the enzyme levels induced over control by each treatment group. Values are
means ⫾ SD (n ⫽ 4); values bearing different superscripts differ significantly
(p ⬍ 0.05). Actual mean value is shown above each bar; a dashed line across
the graph identifies the control, baseline value.
uated alone suggests that the GSTYa2 subunit is responsible
for a substantial portion of the increase in total GST activity.
We also examined induction of GST Yc1 and Yc2 subunits
by Northern and Western blots. Since GST Yc1 and Yc2 show
about 90% homology in gene sequences, we needed to confirm
the mRNA induction by evaluating protein levels using specific
monoclonal antibodies. In adult rat liver GST Yc1 is constitu-
tively expressed, while GST Yc2 is expressed constitutively
only in young rats up to 8 weeks of age (Hayes et al., 1994).
Both GSTs are inducible in young and old rats. Since the
animals used in our study were 7 weeks old, we observed
notable basal levels of both GST Yc1 and Yc2 mRNA and
protein. We also confirmed that I3C was able to significantly
induce GST Yc2 protein, as reported in previous studies (Man-
son et al. 1997, 1998).
Among the three different subunits examined, the only syn-
ergistic response that we observed was in transcription of the
GST Ya2 subunit. Although GST Yc1 and Yc2 are in the
␣⫺GST class with Ya2, neither Northern nor Western blot
FIG. 5. Induction of expression of GST Yc2 mRNA by crambene, I3C, or
both. (A) mRNA levels were measured by Northern blot, as described under
Materials and Methods. Fifteen micrograms of total RNA was loaded per well
and hybridized with a rat GST Yc2 specific 430-bp cDNA probe. Each lane
represents RNA of one individual animal. (B) Densitometry readings of GST
Yc2 mRNA were normalized to 28S ribosomal RNA and expressed as per-
centage of control mRNA. Rats were treated daily for 7 days with crambene
(CB, 50 mg/kg), I3C (56 mg/kg), or both together (MIX) and killed on the 8th
day. Control animals received corn oil and were fed ad libitum (CO) or pair fed
to the group receiving the mix (PF). The shaded portion of each bar represents
control levels of enzyme activity and the open portion of each bar represents
the enzyme levels induced over control by each treatment group. Values are
means ⫾ SD (n ⫽ 4); values bearing different superscripts differ significantly
(p ⬍ 0.05). Actual mean value is shown above each bar; a dashed line across
the graph identifies the control, baseline value.
showed synergism in the rats receiving I3C and crambene
together. This further suggests that the subunit Ya2 may be
responsible for synergism in GST activity by crambene and
FIG. 6. Western blot analysis of cytosolic proteins of livers from rats
treated with crambene, I3C, or both. Rats were treated daily for 7 days with
crambene (CB, 50 mg/kg), I3C (56 mg/kg), or both together (MIX) and killed
on the 8th day. Control animals received corn oil and were fed ad libitum (CO)
or pair fed to the group receiving the mix (PF). Twenty-five micrograms of
cytosolic proteins was loaded in each lane. Each lane represents one cytosolic
sample from each group. The Yc1 and Yc2 proteins were detected using
monoclonal anti-rat GST Yc1 or Yc2 antibodies. Samples were loaded onto
SDS–PAGE as follows: lane 1, 25 ␮g of control liver; lane 2, 25 ␮g of
crambene-treated liver; lane 3, 25 ␮g of I3C-treated liver; lane 4, 25 ␮g of
crambene ⫹ I3C-treated liver. The molecular weights of GST Yc1 and Yc2
proteins are 2.8 and 2.5 kDa, respectively.
 SYNERGISTIC UPREGULATION OF DETOXIFICATION ENZYMES
I3C. Further investigation is needed across a large number of
GST subunits to confirm this.
The presence of an ARE is obligatory in the regulatory
region of detoxification enzymes for a response to monofunc-
tional inducers to occur (Rushmore and Pickett, 1990). The
ARE has been found in the regulatory region for ␥-glutamyl
cysteinyl synthase and several phase II detoxification enzymes,
including several GSTs, QR, epoxide hydrolase, and UDP-
glucuronosyltransferase. This battery of enzymes is thought to
protect cells against the toxic and neoplastic action of carcin-
ogens. We suggest that crambene upregulates phase II enzymes
via an ARE, since crambene effects are consistent with those of
a monofunctional inducer. I3C has been classified as a bifunc-
tional inducer of phase I and phase II detoxification enzymes
(Loub et al., 1975; Wattenberg, 1975). Induction of these
enzymes has been shown to depend on activation of the XRE
found in the regulatory region of the gene sequence of these
enzymes (Chen et al., 1996; Jellinck et al., 1993). Our data are
consistent with the possibility that the synergism observed at
the transcriptional level is due to concurrent stimulation at both
the ARE and XRE by crambene and I3C, respectively. For
example, the stimulation of both might prolong binding at the
ARE, resulting in cooperativity. Furthermore, that the syner-
gistic induction occurred for the GST Ya2 subunit, but not in
the GST Yc1/Yc2 subunit further supports this hypothesis.
Although GST Ya2 and Yc1/Yc2 contain AREs and XREs in
common, the XRE motifs of GST Yc1/Yc2 are not in the
5⬘-regulatory region as they are in GST Ya2 (Pulford and
Hayes, 1996).
In conclusion, this study shows that a mixture of crambene
and I3C cause synergistic induction of phase II detoxification
enzymes, including GST and QR enzyme activity. Of the GST
subunits tested, only GST Ya2 showed a synergistic induction
by this mixture of I3C ⫹ crambene. The induction could be
traced to the level of transcription. The molecular mechanisms
of synergistic induction of phase II enzymes by an I3C and
crambene mixture are under investigation.
ACKNOWLEDGMENTS
This research was supported by grants from the USDA (National Research
Initiative 99 –35503-7010) and Standard Process, Inc., Palmyra, Wisconsin.
REFERENCES
Benson, A. M., Hunkeler, M. J., and Talalay, P. (1980). Increase of NAD(P)H:
quinone reductase by dietary antioxidants: Possible role in protection against
carcinogenesis and toxicity. Proc. Natl. Acad. Sci. USA 77, 5216 –5220.
Buetler, T. M., Gallagher, E. P., Wang, C., Stahl, D. L., Hayes, J. D., and
Eaton, D. L. (1995). Induction of phase I and phase II drug-metabolizing
enzyme mRNA, protein, and activity by BHA, ethoxyquin, and oltipraz.
Toxicol. Appl. Pharmacol. 135, 45–57.
Chen, I., Safe, S., and Bjeldanes, L. (1996). Indole-3-carbinol and diindolyl-
methane as aryl hydrocarbon (Ah) receptor agonists and antagonists in
T47D human breast cancer cells. Biochem. Pharmacol. 51, 1069 –1076.
151
Fenwick, G. R., Heaney, R. K., and Mullin, W. J. (1983). Glucosinolates and
their breakdown products in food and food plants. Crit. Rev. Food Sci. Nutr.
18, 123–201.
Habig, W. H., Pabst, M. J., and Jakoby, W. B. (1974). Glutathione S-trans-
ferases. The first enzymatic step in mercapturic acid formation. J. Biol.
Chem. 249, 7130 –7139.
Hayes, J. D., Nguyen, T., Judah, D. J., Petersson, D. G., and Neal, G. E. (1994).
Cloning of cDNAs from fetal rat liver encoding glutathione S-transferase Yc
polypeptides. The Yc2 subunit is expressed in adult rat liver resistant to the
hepatocarcinogen aflatoxin B 1. J. Biol. Chem. 269, 20707–20717.
Hayes, J. D., and Pulford, D. J. (1995). The glutathione S-transferase super-
gene family: Regulation of GST and the contribution of the isoenzymes to
cancer chemoprotection and drug resistance. Crit. Rev. Biochem. Mol. Biol.
30, 445– 600.
Jellinck, P. H., Makin, H. L., Sepkovic, D. W., and Bradlow, H. L. (1993).
Influence of indole carbinols and growth hormone on the metabolism of 4-
androstenedione by rat liver microsomes. J. Steroid Biochem. Mol. Biol. 46,
791–798.
Loub, W. D., Wattenberg, L. W., and Davis, D. W. (1975). Aryl hydrocarbon
hydroxylase induction in rat tissues by naturally occurring indoles of cru-
ciferous plants. J. Natl. Cancer Inst. 54, 985–988.
Manson, M. M., Ball, H. W., Barrett, M. C., Clark, H. L., Judah, D. J.,
Williamson, G., and Neal, G. E. (1997). Mechanism of action of dietary
chemoprotective agents in rat liver: Induction of phase I and II drug
metabolizing enzymes and aflatoxin B 1 metabolism. Carcinogenesis 18,
1729 –1738.
Manson, M. M., Hudson, E. A., Ball, H. W., Barrett, M. C., Clark, H. L.,
Judah, D. J., Verschoyle, R. D., and Neal, G. E. (1998). Chemoprevention
of aflatoxin B1-induced carcinogenesis by indole-3-carbinol in rat liver-
predicting the outcome using early biomarkers. Carcinogenesis 19, 1829 –
1836.
March, T. H., Jeffery, E. H., and Wallig, M. A. (1998). The cruciferous nitrile,
crambene, induces rat hepatic and pancreatic glutathione S-transferases.
Toxicol. Sci. 42, 82–90.
Nguyen, T., and Pickett, C. B. (1992). Regulation of rat glutathione S-
transferase Ya subunit gene expression. DNA–protein interaction at the
antioxidant responsive element. J. Biol. Chem. 267, 13535–13539.
Pohl, R. J., and Fouts, J. R. (1980). A rapid method for assaying the metab-
olism of 7-ethoxyresorufin by microsomal subcellular fractions. Anal. Bio-
chem. 107, 150 –155.
Prochaska, H. J., and Talalay, P. (1988). Regulatory mechanisms of mono-
functional and bifunctional anticarcinogenic enzyme inducers in murine
liver. Cancer Res. 48, 4776 – 4782.
Pulford, D. J., and Hayes, J. D. (1996). Characterization of the rat glutathione
S-transferase Yc2 subunit gene, GSTA5: Identification of a putative antiox-
idant-responsive element in the 5⬘-flanking region of rat GSTA5 that may
mediate chemoprotection against aflatoxin B 1. Biochem. J. 318, 75– 84.
Rushmore, T. H., and Pickett, C. B. (1990). Transcriptional regulation of the
rat glutathione S-transferase Ya subunit gene. Characterization of a xeno-
biotic-responsive element controlling inducible expression by phenolic an-
tioxidants. J. Biol. Chem. 265, 14648 –14653.
Staack, R. F., Kingstone, S., Wallig, M., and Jeffery, E. H. (1998). A com-
parison of the individual and collective effects of four glucosinolate break-
down products from brussels sprouts on induction of detoxification en-
zymes. Toxicol. Appl. Pharmacol. 149, 17–23.
Steinmetz, K. A., and Potter, J. D. (1991). Vegetables, fruit, and cancer. II.
Mechanisms. Cancer Causes Control 2, 427– 442.
Stresser, D. M., Bjeldanes L. F., Bailey G. S., and Williams D. E. (1995). The
anticarcinogen 3,3⬘-diindolylmethane is an inhibitor of cytochrome P450.
J. Biochem. Toxicol. 10, 191–201.
Takahashi, N., Dashwood, R. H., Bjeldanes, L. F., Bailey, G. S., and Williams,
152 NHO AND JEFFERY
D. E. (1995a). Regulation of hepatic cytochrome P4501A by indole-3-
carbinol: Transient induction with continuous feeding in rainbow trout.
Food Chem. Toxicol. 33, 111–120.
Takahashi, N., Dashwood, R. H., Bjeldanes, L. F., Williams, D. E., and Bailey,
G. S. (1995b). Mechanisms of indole-3-carbinol (I3C) anticarcinogenesis:
Inhibition of aflatoxin B 1–DNA adduction and mutagenesis by I3C acid
condensation products. Food Chem. Toxicol. 33, 851– 857.
Teel, R. W., and Huynh H. (1998). Modulation by phytochemicals of cyto-
chrome P450-linked enzyme activity. Cancer Lett. 27, 135–141.
Verhoeven, D. T., Verhagen, H., Goldbohm, R. A., van den Brandt, P. A., and
van Poppel, G. (1997). A review of mechanisms underlying anticarcinoge-
nicity by brassica vegetables. Chem.–Biol. Interact. 103, 79 –129.
Wallig, M. A., Kore, A. M., Crawshaw, J., and Jeffery, E. H. (1992). Sepa-
ration of the toxic and glutathione-enhancing effects of the naturally occur-
ring nitrile, cyanohydroxybutene. Fundam. Appl. Toxicol. 19, 598 – 606.
Wattenberg, L. W. (1975). Effects of dietary constituents on the metabolism of
chemical carcinogens. Cancer Res. 35, 3326 –3331.
Whitlock, J. P., Jr. (1999). Induction of cytochrome P4501A1. Annu. Rev.
Pharmacol. Toxicol. 39, 103–125.
Wortelboer, H. M., van der Linden, E. C., de Kruif, C. A., Noordhoek, J.,
Blaauboer, B. J., van Bladeren, P. J., and Falke, H. E. (1992). Effects of
indole-3-carbinol on biotransformation enzymes in the rat: In vivo changes
in liver and small intestinal mucosa in comparison with primary hepatocyte
cultures. Food Chem. Toxicol. 30, 589 –599.