Role of glutathione S-transferases A1-1, M1-1, and P1-1 in the detoxification of 2-phenylpropenal, a reactive felbamate metabolite

Article abstract of DOI:10.1021/tx000141e

Felbamate has proven to be an effective therapy for treating refractory epilepsy. However, felbamate therapy has been limited due to the associated reports of hepatotoxicity and aplastic anemia. Previous research from our laboratory has proposed 2-phenylpropenal as the reactive metabolite in felbamate bioactivation and identified its mercapturates in the urine of rats and patients undergoing felbamate therapy. While the reaction between 2-phenylpropenal and GSH has been shown to occur spontaneously under physiological conditions, the potential catalysis by glutathione transferases (GST) has remained unknown. The work presented here demonstrates a role for GST in the detoxification of 2-phenylpropenal. The kinetic data show that 2-phenylpropenal is a substrate for all three isoforms tested, with a k_cat/K_m of 0.275 ± 0.035 μM^-1 s^-1 for GSTM1-1, 0.164 ± 0.005 μM^-1 s^-1 for GSTP1-1, and 0.042 ± 0.005 μM^-1 s^-1 for GSTA1-1. Given that electrophilic substrates such as 2-propenal have been shown to inhibit GSTs, we also examined the inhibition of GSTM1-1, GSTP1-1 and GSTA1-1 by 2-phenylpropenal. The enzyme inhibition studies demonstrate that 2-phenylpropenal inhibits GSTP1-1 and GSTM1-1. The inhibition of GSTP1-1 was completely reversible upon filtration and reconstitution in buffer containing 10 mM GSH. However, 2-phenylpropenal inhibition of GSTM1-1 was irreversible under the same conditions. The irreversible inhibition of GSTM1-1 may be important in understanding the toxicities associated with felbamate. Given that 2-phenylpropenal is both a substrate and irreversible inhibitor for GSTM1-1, GSTM1-1 represents a potential target for 2-phenylpropenal haptenization in vivo, which may in turn mediate the observed idiosyncratic reactions.

Full text of DOI:10.1021/tx000141e

Chem. Res. Toxicol. 2001, 14, 511-516
511
Role of Glu ta th ion e S-Tr a n sfer a ses A1-1, M1-1, a n d P 1-1
in th e Detoxifica tion of 2-P h en ylp r op en a l, a Rea ctive
F elba m a te Meta bolite
Christine M. Dieckhaus,† Shane G. Roller,† Webster L. Santos,†
R. Duane Sofia,‡ and Timothy L. Macdonald*,†
Department of Chemistry, University of Virginia, Charlottesville, Virginia 22901, and
Research and Development, Wallace Laboratories, Cranbury, New J ersey 08512
Received J uly 11, 2000
Felbamate has proven to be an effective therapy for treating refractory epilepsy. However,
felbamate therapy has been limited due to the associated reports of hepatotoxicity and aplastic
anemia. Previous research from our laboratory has proposed 2-phenylpropenal as the reactive
metabolite in felbamate bioactivation and identified its mercapturates in the urine of rats and
patients undergoing felbamate therapy. While the reaction between 2-phenylpropenal and GSH
has been shown to occur spontaneously under physiological conditions, the potential catalysis
by glutathione transferases (GST) has remained unknown. The work presented here demon-
strates a role for GST in the detoxification of 2-phenylpropenal. The kinetic data show that
2-phenylpropenal is a substrate for all three isoforms tested, with a kcat/Km of 0.275 ( 0.035
µM-1 s-1 for GSTM1-1, 0.164 ( 0.005 µM-1 s-1 for GSTP1-1, and 0.042 ( 0.005 µM-1 s-1 for
GSTA1-1. Given that electrophilic substrates such as 2-propenal have been shown to inhibit
GSTs, we also examined the inhibition of GSTM1-1, GSTP1-1 and GSTA1-1 by 2-phenylpro-
penal. The enzyme inhibition studies demonstrate that 2-phenylpropenal inhibits GSTP1-1
and GSTM1-1. The inhibition of GSTP1-1 was completely reversible upon filtration and
reconstitution in buffer containing 10 mM GSH. However, 2-phenylpropenal inhibition of
GSTM1-1 was irreversible under the same conditions. The irreversible inhibition of GSTM1-1
may be important in understanding the toxicities associated with felbamate. Given that
2-phenylpropenal is both a substrate and irreversible inhibitor for GSTM1-1, GSTM1-1
represents a potential target for 2-phenylpropenal haptenization in vivo, which may in turn
mediate the observed idiosyncratic reactions.
phobic electrophilic species (3). The binding of GSH in
the G-site lowers the pKa of the cysteine sulfhydryl from
9 to 6.5 such that bound GSH is thought to exist as the
thiolate anion under physiological conditions (3). GSTs
are expressed ubiquitously with isoform specific tissue
localization: GSTA1-1 in the liver, kidney, intestine, and
lung, GSTM1-1 in liver, muscle, testis, and brain, and
GSTP1-1 in kidney and neoplastic tissue with notable
absence in the liver (3). Regulation of enzyme expression
has been correlated with numerous response factors and
shown to be able to be induced by polycyclic aromatic
hydrocarbons, phenolic antioxidants, Michael reaction
acceptors, reactive oxygen species, organic isothiocyan-
ates, barbiturates, and synthetic glucocorticoids (3).
In tr od u ction
Previous research has demonstrated the bioactivation
of the antiepileptic drug felbamate to the highly reactive
R,â-unsaturated aldehyde, 2-phenylpropenal (1), which
is thought to mediate felbamate-associated toxicities,
namely, hepatotoxicity and aplastic anemia. The detoxi-
fication of 2-phenylpropenal by GSH has been shown to
occur in vivo by identification of the corresponding
mercapturates in the urine of both rats and patients
being treated with felbamate (1). In addition, the conju-
gation of 2-phenylpropenal with GSH was shown to occur
spontaneously under physiological conditions with an
extremely short half-life (2). However, the potential role
for glutathione transferases in mediating this reaction
has remained unexplored.
Several R,â-unsaturated aldehydes, including the reac-
tive cylcophosphamide metabolite 2-propenal (acrolein),
nucleotide base propenals, and 4-hydroxyalkenals, have
been characterized as substrates for GSTs (4) and led us
to consider the possible role of GSTs in the detoxification
of 2-phenylpropenal. Given that the spontaneous reaction
between GSH and 2-phenylpropenal occurs rapidly under
normal physiological concentrations of GSH, the roles of
GSTs may become important in GSH-depleted individu-
als or in understanding the metabolic dispositions in
polymorphic patients. In addition, irreversible inhibition
of GST by R,â-unsaturated aldehydes has also been
detected (5, 6) and may occur in the case of 2-phenylpro-
GSTs1 are a family of enzymes (GST-A, GST-M, GST-
P, and GST-T) characterized by their ability to catalyze
the reaction between GSH and hydrophobic electrophiles.
The active site of GSH has two binding sites, the G-site,
which binds GSH, and the H-site, which binds hydro-
* To whom correspondence should be addressed: Department of
Chemistry, University of Virginia, McCormick Road, Charlottesville,
VA 22901. Phone: (804) 924-7718. E-mail: tlm@virginia.edu.
† University of Virginia.
‡ Wallace Laboratories.
1
Abbreviations: GST, glutathione S-transferase; CDNB, 1-chloro-
2,4-dinitrobenzene.
10.1021/tx000141e CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/14/2001

512 Chem. Res. Toxicol., Vol. 14, No. 5, 2001
Dieckhaus et al.
The kinetic analysis of the control and enzyme-catalyzed
reactions were performed at 37 °C in 500 µL of 100 mM
potassium phosphate buffer (pH 6.5) containing 500 µM reduced
GSH, N-acetyl-S-(2-phenylpropan-3-ol)-L-cysteine as an internal
standard, and 2.5 µg of the respective enzyme. The reactions
were initiated by the addition of 2-phenylpropenal to a final
concentration of 25, 50, 75, 100, or 150 µM and stopped with
the addition of NaBH4 (∼1 mg). The zero time point was
obtained by adding NaBH4 to the reaction mixture immediately
before the addition of 2-phenylpropenal. The quenched reactions
were neutralized by the addition of 50 µL of 20% acetic acid
and quantified by HPLC using UV detection and integration
as described above. A calibration curve demonstrated a linear
absorbance response for N-acetyl-S-(2-phenylpropan-3-ol)-L-
cysteine and S-(2-phenylpropan-3-ol)-γ-GSH to 200 µM, the
highest concentration that was examined. The linear relation-
ship allowed for the absolute quantification of S-(2-phenylpro-
pan-3-ol)-γ-GSH. Each of the time points and concentrations was
analyzed in triplicate.
penal. Demonstrating the irreversible inhibition of GST
by 2-phenylpropenal under GSH-depleted conditions
would identify a candidate protein for in vivo alkylation.
Identification of the hapten that mediates felbamate-
associated toxicity remains an important goal in the
mechanistic understanding of felbamate idiosyncratic
reactions.
Exp er im en ta l P r oced u r es
Ch em ica ls a n d In str u m en ts. The chemicals were pur-
chased from either Sigma (St. Louis, MO) or Aldrich (Milwau-
kee, WI) and were of the highest grade available. The enzymes
were purchased from Panvera (Madison, WI). Panvera reported
the GST specific activities for the reaction of CDNB (1-chloro-
2,4-dinitrobenzene) with 1 mM reduced GSH in 100 mM
potassium phosphate buffer (pH 6.5) at room temperature as
follows: 95 µmol min-1 mg-1 for GSTP1-1, 207 µmol min-1 mg-1
for GSTM1-1, and 57 µmol min-1 mg-1 for GSTA1-1. The HPLC
analysis was conducted on a Waters 2690 Separations Module
using a Waters Symmetry C8 (2.1 mm × 150 mm) column. The
separation was achieved isocratically with a 40% MeOH/60%
(0.1%) formic acid mixture and a 200 µL/min flow rate. The flow
was directed into a Waters 486 tunable absorbance detector
equipped with a microbore flow cell and set at λ ) 214 nm. The
HPLC/UV quantification was achieved using a Hewlett-Packard
3394A integrator. The absorbance assays were performed on a
Perkin-Elmer UV/Vis Lambda 20 spectrometer set at λ ) 340
nm. The mass spectra were obtained using a Finnigan LCQ ion
trap electrospray ionization mass spectrometer, and the NMR
spectra were obtained on a 300 MHz General Electric QE300
spectrometer
Syn t h esis of N-Acet yl-S-(2-p h en ylp r op a n -3-ol)-L-cys-
tein e an d S-(2-P h en ylpr opan -3-ol)-GSH. N-Acetyl-S-(2-phen-
ylpropan-3-ol)-L-cysteine was synthesized as previously pub-
lished (1) except that the alcohol was generated in situ before
product purification. The GSH adduct, S-(2-phenylpropan-3-ol)-
GSH, was obtained in an analogous manner by substituting
N-acetyl-L-cysteine with GSH. LC/ESI-MS for N-acetyl-S-(2-
phenylpropan-3-ol)-L-cysteine: tR ) 7.5 min, [M + 1H]+ m/z 298.
LC/ESI-MS for S-(2-phenylpropan-3-ol)-GSH: tR ) 3.3 min, [M
+ 1H]+ m/z 442. Both products were obtained in quantitative
yield and determined to be >95% pure by HPLC.
Syn th esis of 2-P h en ylp r op en a l. 2-Phenyl-1,3-propanediol
monocarbamate was synthesized as previously published (1).
To a solution of 2-phenyl-1,3-propanediol monocarbamate (0.55
g, 3.1 mmol) in CH2Cl2 in a round-bottom flask (protected from
light) was added Dess-Martin Periodinane (1.50 g, 3.5 mmol)
in one portion. After 2 h, triethylamine (0.471 mL, 3.4 mmol)
was added to the reaction mixture and the solution stirred for
an additional 1 h, at which time the reaction was complete. The
crude material was purified by flash chromatography on silica
(1:1 ether/petroleum ether mixture), affording the product as
solid white crystals (0.253 g, 62%): thin-layer chromatography
Rf ) 0.75 (1:1 ether/petroleum ether mixture). The compound
was characterized by 1H NMR and 13C NMR, and the results
are in agreement with the previously published data (2). The
purity of the compound was determined to be >99% by HPLC
(λ ) 214 nm).
Kin etic An a lysis of th e Rea ction betw een 2-P h en ylp r o-
p en a l a n d GSH. Given that the uncatalyzed reaction between
2-phenylpropenal and GSH proceeds on the second time scale,
the kinetic experiments required a method that could instan-
taneously quench the reaction. Sodium borohydride (NaBH4)
was selected to quench the reactions because it could both
inactivate the enzyme and reduce 2-phenylpropenal to the
corresponding alcohol. Control experiments demonstrated that
when 2-phenylpropenal (to 150 µM) was added to a reaction
mixture containing 500 µM GSH, 2.5 µg of the respective
enzyme, and a spatula tip of NaBH4 (∼1 mg) in 500 µL of 100
mM potassium phosphate buffer (pH 6.5) at 37 °C, product
formation was not observed.
Without the use of stopped-flow kinetics, which was not
available to us, the determination of the kinetic parameters (Km,
Vmax, and kcat/Km) was difficult given the reaction half-lives (38
s for no enzyme, 18 s for GSTP1-1, 11 s for GSTM1-1, and 26 s
for GSTA1-1). Therefore, the initial rates were obtained by
plotting the amount of product formed versus time for several
time points and fitting the data to a second-degree equation,
e.g., y ) a + bt + ct2, where b is the initial rate. Lineweaver-
Burk plots of each data set were constructed to determine Km
and Vmax. Comparative analysis utilizing the Eadie-Hofstee plot
produced similar values for Km and Vmax, validating the use of
the Lineweaver-Burk analysis. The kcat was calculated in the
typical manner where kcat ) (voKm)/([substrate][enzyme]). The
half-life data are reported for reaction mixtures containing 50
µM 2-phenylpropenal, 500 µM GSH, and 2.5 µg of the respective
enzyme.
In h ibition of GSTP 1-1, GSTM1-1, or GSTA1-1 by 2-P h en -
ylp r op en a l. The method for monitoring enzyme inhibition has
been adapted from similar experiments (5, 7), and the method
for testing enzyme activity is described below. Briefly, the
enzyme (1 µM) was incubated in 200 mM potassium phosphate
buffer (pH 7.4) containing 200 µM EDTA at 37 °C. The enzyme
activity at time zero was measured just prior to the addition of
2-phenylpropenal. The inhibition was initiated by the addition
of 2-phenylpropenal from a stock solution in CH3CN to a final
concentration of 25 µM. The concentration of CH3CN was 0.3%
(v/v) of the incubation buffer, and the CH3CN was added to the
control. At the appropriate time points, aliquots were removed
from the reaction mixture and tested for enzyme activity. The
incubations with 2-phenylpropenal and controls were performed
in triplicate, and the percent inhibition is reported as the
enzyme activity with 2-phenylpropenal relative to the control.
GST Activity w ith CDNB a n d GSH. GST activity was
determined at 25 °C by measuring the rate of CDNB-GSH
product formation at λ ) 340 nm (5, 7). Briefly, 10 µL of the
enzyme reaction mixture described above was added to 240 µL
of 100 mM potassium phosphate buffer (pH 6.5) containing 200
mM EDTA, 1 mM GSH, and 100 µM CDNB. The reaction was
monitored at λ ) 340 nm for 1 min, and the amount of product
formed was calculated using the difference extinction coefficient
[∆ꢀ ) 9.6 mM-1 cm-1 (7)].
Rever sible In h ibition of GSTP 1-1 a n d GSTM1-1 by
2-P h en ylp r op en a l. The enzyme reaction mixture (125 µL),
incubated as described above for the inhibition of GSTP1-1 and
GSTM1-1 by 2-phenylpropenal, was incubated for 30 min at 37
°C. After the 30 min incubation, the enzyme activity was
determined as described above (reported as time zero) and the
remaining enzyme reaction mixture was filtered through a
Microcon-10 apparatus (Amicon, Beverly, MA) at 10 000 rpm
for 15 min. The Microcon-10 apparatus had been prewashed
with 200 µL of 200 mM potassium phosphate buffer (pH 7.4).
After filtration, the enzyme was reconstituted in 125 µL of 200
mM potassium phosphate buffer (pH 7.4) containing 200 mM

Role of GSTs in 2-Phenylpropenal Detoxification
Chem. Res. Toxicol., Vol. 14, No. 5, 2001 513
F igu r e 2. Lineweaver-Burk plot for GSTM1-1.
F igu r e 1. Graph demonstrating a decrease in the half-life of
the reaction between 2-phenylpropenal and GSH upon incuba-
tion with GSTM1-1, GSTP1-1, or GSTA1-1.
Ta ble 1
half-life of 2-phenylpropenal (s)
no enzyme
GSTM1-1
GSTP1-1
GSTA1-1
37.8 ( 6.8
11.3 ( 1.2
17.7 ( 0.4
26.4 ( 2.4
F igu r e 3. Lineweaver-Burk plot for GSTA1-1.
EDTA and 10 mM GSH and incubated at 37 °C. At each time
point, 10 µL of the enzyme incubation mixture was tested for
activity using CDNB as described above. The experiments with
2-phenylpropenal and control experiments were performed in
triplicate, and the percent activity is reported as the activity
for incubations containing 2-phenylpropenal relative to the
controls.
Resu lts
R ea ct ion Or d er . The order of the non-enzyme-
catalyzed reaction between 2-phenylpropenal and GSH
was determined by holding the concentration of one
reactant constant while varying the concentration of the
other reactant and plotting the log of the initial velocity
versus the log of the initial concentration. As expected
for a bimolecular addition reaction, the reaction was first-
order with respect to 2-phenylpropenal, first-order with
respect to GSH, and second-order overall.
Ha lf-Life of 2-P h en ylp r op en a l w ith a n d w ith ou t
GSTA1-1, GSTP 1-1, a n d GSTM1-1. To determine if
GSTA1-1, GSTP1-1, or GSTM1-1 catalyzes the reaction
between 2-phenylpropenal and GSH, we measured the
half-life of 2-phenylpropenal with and without enzyme.
The results are summarized in Figure 1 and Table 1, and
show that all three GST isoforms catalyze the reaction.
In all cases, a decease in the amount of 2-phenylpropenal
corresponded to an increase in the amount of S-(2-
phenylpropan-3-al)-GSH formed.
F igu r e 4. Lineweaver-Burk plot for GSTP1-1.
Ta ble 2
enzyme
Km (µM)
Vmax (µM s-1
)
kcat/Km (µM-1 s-1
)
GSTM1-1
GSTP1-1
GSTA1-1
153 ( 57
55 ( 10
217 ( 107
10.0 ( 2.2
2.6 ( 0.3
2.1 ( 0.7
0.275 ( 0.035
0.164 ( 0.005
0.042 ( 0.005
and Vmax. The results of triplicate experiments are
summarized in Table 2. The reaction between GSH and
2-phenylpropenal is best catalyzed by GSTM1-1 (kcat/Km
) 0.275 ( 0.035 µM-1 s-1) followed by GSTP1-1 (kcat/Km
) 0.164 ( 0.005 µM-1 s-1) and then GSTA1-1 (kcat/Km
)
0.042 ( 0.005 µM-1 s-1). Correcting for the amount of
product formed in the non-enzyme-catalyzed reaction did
not alter the kcat/Km. Therefore, the data are reported
without modification.
En zym e Kin etic Ch a r a cter iza tion . To determine
which GST isoforms may be important for the in vivo
detoxification of 2-phenylpropenal, we determined the
Km, Vmax, and kcat/Km values for GSP1-1, GSTM1-1, and
GSTA1-1. The kinetic parameters were obtained for each
isozyme by varying the concentration of 2-phenypropenal,
obtaining the initial rates, and plotting the Lineweaver-
Burk plots (Figures 2-4). Comparative analysis using
the Eadie-Hofstee plot provided similar values for Km
In h ibition of GSTM1-1, GSTP 1-1, or GSTA1-1 by
2-P h en ylp r op en a l. We sought to determine if 2-
phenylpropenal could inhibit GSTM1-1, GSTP1-1, or
GSTA1-1 under GSH-depleted conditions. To determine
if 2-phenylpropenal could inhibit GSTM1-1, GSTP1-1, or
GSTA1-1, we incubated the respective enzyme with 25
µM 2-phenylpropenal at 37 °C and determined the
enzyme activity over the course of 2 h. The results for
GSTM1-1 and GSTP1-1 are reported as percent activity

514 Chem. Res. Toxicol., Vol. 14, No. 5, 2001
Dieckhaus et al.
by 2-phenylpropenal is reversible as soon as the enzyme
is incubated with an excess of GSH. However, the
inhibition of GSTM1-1 was not reversible when the
enzyme was incubated with an excess of GSH, which
suggests that 2-phenylpropenal irreversibly inhibits
GSTM1-1.
Discu ssion
The role of glutathione transferases in the detoxifica-
tion of 2-phenylpropenal may be important, particularly
in GSH-depleted patients. Even though the spontaneous
reaction between GSH and 2-phenylpropenal occurs
rapidly, the reaction exhibits second-order kinetics and
should drop precipitously under GSH-depleted conditions.
Under our assay conditions, the rate of the reaction is
increased 3.1-fold with GSTM1-1, 2.1-fold with GSTP1-
1, and 1.4-fold with GSTA1-1. Given that the enzyme
concentrations in our reaction conditions are much lower
than the physiological levels in the liver (8), we suspect
that the reaction between 2-phenylpropenal and GSH is
catalyzed in vivo. Likewise, it is important to note that
the concentration of GSTA1-1 is greater than the con-
centration of GSTM1-1 in liver (8, 9) such that GSTA1-1
may contribute to the in vivo detoxification of 2-phenyl-
propenal. The differences in reaction rates between the
enzyme-catalyzed and non-enzyme-catalyzed reaction
suggest that a patient deficient in GST may be more
susceptible to felbamate toxicity.
F igu r e 5. Irreversible inhibition of GSTM1-1 with 25 µM
2-phenylpropenal.
Polymorphisms in GSTM1-1 and GSTP1-1 have been
identified (3). Polymorphisms in GSTM1-1 have been
shown to exist in 50% of the population (10). One study
was able to demonstrate that human lymphocytes defi-
cient in GSTM1-1 were more prone to epoxide-induced
toxicity (3). Therefore, patients polymorphic in GSTM1-1
may be more susceptible to felbamate-associated toxicity.
Due to its lack of commercial availability, we did not
characterize the role of GSTT1-1 in the conjugation
between GSH and 2-phenylpropenal. GSTT1-1 catalyzes
the conjugation reaction between GSH and compounds
such as 4-nitrophenyl bromide, methyl bromide, dichlo-
romethane, and ethylene oxide (3). Given typical sub-
strate characteristics, we do not expect GSTT1-1 to
catalyze the reaction between GSH and 2-phenylprope-
nal. However, it is worth mentioning that GSTT1-1 is
expressed polymorphically in red cells in ∼15% of the
population (11).
GSTP1-1 is thought to mediate the GSH conjugation
of the majority of small molecule R,â-unsaturated alde-
hydes (3). Given the reported literature values for ac-
rolein, we expected GSTP1-1 to be the main isoform for
catalyzing the reaction between GSH and 2-phenylpro-
penal. However, a thorough kinetic analysis has shown
that GSTM1-1 has the highest rate of enzyme turnover
for the reaction between GSH and 2-phenylpropenal. A
comparison of our kinetic data with the reported data
for other R,â-unsaturated aldehydes is presented in Table
3. The data in Table 3 show an increase in substrate
specificity for GSTM1-1 over GSTP1-1 with an increase
in alkyl chain length. A comparison of 2-phenylpropenal
to acrolein and the base propenals suggests that the
phenyl ring substitution at the C2 position increases
selectivity for GSTM1-1 over GSTP1-1. Structural knowl-
edge about GST isoform specificity may be useful in the
prediction of other R,â-unsaturated aldehydes and in
generating computer-based predictive software.
F igu r e 6. Reversible inhibition of GSTP1-1 with 25 µM
2-phenylpropenal.
as compared to controls without 2-phenylpropenal and
are summarized in Figures 5 and 6. The results show
that at 25 µM, 2-phenylpropenal is able to inhibit the
activity of GSTM1-1 and GSTP1-1; however, at this
concentration, 2-phenylpropenal did not inhibit the activ-
ity of GSTA1-1. The IC50 for the inhibition of both
GSTP1-1 and GSTM1-1 by 2-phenylpropenal is ∼20 µM.
The inactivation of GSTP1-1 occurs rapidly within the
first 15 min of incubation. The activity of GSTM1-1 is
reduced to 26% of the activity of the control by 15 min
and steadily decreases over the course of 2 h.
Rever sibility of th e In h ibition of GSTM1-1 a n d
GSTP 1-1 by 2-P h en ylp r op en a l. Given the inhibition
of GSTP1-1 and GSTM1-1 by 2-phenylpropenal, we
examined the reversibility of the inhibition by removing
unbound 2-phenylpropenal and incubating the inacti-
vated enzyme in 200 mM potassium phosphate buffer
containing 10 mM GSH. Aliquots of the enzyme were
tested for activity over the course of 2 h, and the results
are shown in Figures 5 and 6. The inhibition of GSTP1-1

Role of GSTs in 2-Phenylpropenal Detoxification
Chem. Res. Toxicol., Vol. 14, No. 5, 2001 515
Ta ble 3a
a
Values for acrolein are from ref 20; values for crotonaldehyde, base propenals, and 4-vinylpyridine are from ref 4, and values for
4-hydroxyalkenals are from ref 21.
Substrates for GST are often found to inhibit the
enzyme (3, 5). It has been proposed that GSTs react with
electrophilic compounds to protect from intracellular
damage (3). Not surprisingly, a number of R,â-unsatur-
ated aldehydes inhibit GSTs, particularly GSTP1-1.
GSTP1-1 contains a reactive cysteine, Cys-47, which has
an estimated pKa of 4.2 (12) and has been shown to
become alkylated by a number of electrophiles, including
ethacrynic acid (5). The reversibility of the inhibition is
compound-dependent. We have shown that 2-phenylpro-
penal reversibly inhibits GSTP1-1. In addition, R,â-
unsaturated aldehydes have also been shown to inhibit
GSTM1-1 (13), although details about the mechanism of
inhibition are not as well understood. We have shown
that 2-phenylpropenal irreversibly inhibits GSTM1-1
after filtration and reconstitution in 10 mM GSH. The
irreversible inhibition of GSTM1-1 represents the iden-
tification of a candidate protein for in vivo alkylation,
which may contribute to our understanding of felbamate-
associated toxicities. Often the proteins implicated in
mediating toxic drug-induced reactions are the metabo-
lizing enzymes themselves (14). Under GSH-depleted

516 Chem. Res. Toxicol., Vol. 14, No. 5, 2001
Dieckhaus et al.
(2) Thompson, C. D., Kinter, M. T., and Macdonald, T. L. (1996)
Synthesis and in vitro reactivity of 3-carbamoyl-2-phenylprope-
nal: putative reactive metabolite of felbamate. Chem. Res. Toxicol.
9, 1225-1229.
(3) Hayes, J . D., and Pulford, D. J . (1995) The Glutathione S-
Transferase Supergene Family: Regulation of GST and the
contribution of the isoenzymes to cancer chemoprotection and
drug resistance. Crit. Rev. Biochem. Mol. Biol. 30, 445-600.
(4) Berhane, K., Widersten, M., Engstrom, A., Kozarich, J . W., and
Mannervick, B. (1994) Detoxication of base propenals and other
R,â-unsaturated aldehyde products of radical reactions and lipid
peroxidation by human glutathione transferases. Proc. Natl. Acad.
Sci. U.S.A. 91, 1480-1484.
(5) van Iersel, M. L., Ploemen, J . H., Lo Bello, M., Federici, G., and
van Bladeren, P. J . (1997) Interactions of R,â-unsaturated alde-
hydes and ketones with human glutathione S-transferase P1-1.
Chem.-Biol. Interact. 108, 67-78.
conditions, 2-phenylpropenal may also alkylate GSTP1-1
and thereby mediate the observed toxicities.
The potential deleterious effects of 2-phenylpropenal
inhibiting GSTP1-1 and GSTM1-1 are not limited their
inability to detoxify atropaldehyde but extend to implica-
tions concerning the general redox status of a cell. The
delicate balance between normal cellular metabolism
generating reactive oxygen species and lipid oxidation
products, such as 4-hydroxynonenal, and detoxification
of those molecules remains critical to cell survival. A
reduction in GSH concentrations and the level of inhibi-
tion of GST may lead to a cumulative overload of reactive
oxygen species and its products, resulting in cellular
death and toxicity.
(6) van Iersel, M. L., Ploemen, J . H., Struik, I., van Amersfoort, C.,
Keyzer, A. E., Schefferlie, J . G., and van Bladeren, P. J . (1996)
Inhibition of glutathione S-transferase activity in human mela-
noma cells by R,â-unsaturated carbonyl derivatives. Effects of
acrolein, cinnamaldehyde, citral, crotonaldehyde, curcumin,
ethacrynic acid, and trans-2-hexenal. Chem.-Biol. Interact. 102,
117-132.
In addition to the role of GST, the aldehyde dehydro-
genase (ALDH) that oxidizes 3-carbamoyl-2-phenylpro-
panal, to the corresponding acid, 3-carbamoyl-2-phenyl-
propionic acid must also be considered and is the focus
of future research. Previous studies have demonstrated
the formation of 3-carbamoyl-2-phenylpropionic acid from
its metabolic precursor, 2-phenyl-1,3-propanediol mono-
carbamate, upon incubation with human liver S9 frac-
tions (15). Additional studies with human liver S9
fractions were able to demonstrate that atropaldehyde
inhibits ALDH and conversion of 3-carbamoyl-2-phenypro-
panal to 3-carbamoyl-2-phenylpropionic acid (16). Studies
with the structurally similar cyclophosphamide metabo-
lite, aldophosphamide, have demonstrated substrate
specificity for ALDH1 and ALDH3 in the generation of
the nontoxic metabolite carboxycyclophosphamide (17).
Like 3-carbamoyl-2-phenypropanal, aldophosphamide
undergoes â-elimination affording 2-propenal that is
thought to inhibit erythrocyte ALDH1, affecting the
overall pharmacokinetics of 4-hyrdoxycyclophosphamide
(18). Given the similarities between cyclophosphamide
metabolism and felbamate metabolism, we propose that
ALDH1 may be important in the detoxification of 2-phen-
ylpropenal and may be the enzyme inhibited by 2-phen-
ylpropenal in the human liver S9 fractions. Polymor-
phisms in several ALDH enzymes, including ALDH1, are
well-characterized (19) and may contribute to the predic-
tion of patients at risk for felbamate toxicity.
In summary, all of our data at present support the
hypothesis that 2-phenylpropenal is the reactive metabo-
lite that mediates felbamate-associated hepatotoxicity
and aplastic anemia. We further propose that maintain-
ing normal GSH levels remains the key determinate in
preventing felbamate toxicity. Our results suggest that
GSTs catalyze the reaction between 2-phenylpropenal
and GSH in vivo. Particularly, in patients who become
GSH-depleted, the role of GST may become increasingly
important and any GST polymorphisms resulting in a
loss of activity may further promote felbamate toxicity.
In addition, polymorphisms in ALDH activity may result
in a relative increase in the amount of atropaldehyde that
is formed. An increase in the amount of atropaldehyde
would promote GSH depletion.
(7) Habig, W. H., Pabst, M. J ., and J akoby, W. B. (1974) Glutathione
S-transferases. J . Biol. Chem. 249, 7130-7139.
(8) van Ommen, B., Bogaards, J . J . P., Peters, W. H. M., Blaauboer,
B., and van Bladeren, P. J . (1990) Quantification of human
glutathione S-transferases. Biochem. J . 269, 609-613.
(9) Mulder, T. P. J ., Court, D., and Peters, W. H. M. (1999) Variability
of glutathione S-transferase R in human liver and plasma. Clin.
Chem. 45, 355-359.
(10) Zhao, L., Alldersea, J ., Fryer, A., Tighe, A., Ollier, B., Thomson,
W., J ones, P., and Strange, R. (1994) Polymorphism at the
glutathione S-transferase GSTM1 locus: a study of the frequen-
cies of the GSTM1 A, B, A/B and null phenotypes in Nigerians.
Clin. Chim. Acta 225, 85-88.
(11) Kempkes, M., Wiebel, F. A., Golka, K., Heitmann, P., and Bolt,
H. M. (1996) Comparative genotyping and phenotyping of glu-
tathione S-transferase GSTT1. Arch. Toxicol. 70, 306-309.
(12) LoBello, M., Parker, M. W., Desideri, A., Polticelli, F., Falconi,
M., Del Boccio, G., Pennelli, A., Federici, G., and Ricci, G. (1993)
Peculiar spectroscopic and kinetic properteis of Cys-47 in human
placental glutathione transferase. J . Biol. Chem. 268, 19033-
19038.
(13) Chien, C., Kirollos, K. S., Linderman, R. J ., and Dauterman, W.
C. (1994) R,â-Unsaturated carbonyl compounds: inhibition of rat
liver glutathione S-transferase isozymes and chemical reaction
with reduced glutathione. Biochim. Biophys. Acta 1204, 175-180.
(14) Bourdi, M., Chen, W., Peter, R. M., Martin, J . L., Buters, J . T.,
Nelson, S. D., and Pohl, L. R. (1996) Human Cytochrome P450
2E1 is a major autoantigen associated with halothane hepatitis.
Chem. Res. Toxicol. 9, 1159-1166.
(15) Kapetanovic, I. M., Torchin, C. D., Thompson, C. D., Miller, T.
A., McNeilly, P. J ., Macdonald, T. L., Kupferberg, H. J ., Perhach,
J . L., Sofia, R. D., and Strong, J . M. (1998) A potentially reactive
cyclic carbamate metabolite of the anti-epileptic drug felbamate
produced by human liver tissue in vitro. Drug Metab. Dispos. 26,
1089-1085.
(16) Kapetanovic, I. M. (2001) unpublished data.
(17) Sladek, N. E. (1999) Aldehyde dehydrogenase-mediated cellular
relative insensitivity to the oxazaphosphorines. Curr. Pharm. Des.
5, 607-625.
(18) Ren, S., Kalhorn, T. F., and Slattery, J . T. (1999) Inhibition of
human aldehyde dehydrogenase 1 by the 4-hydroxycyclophos-
phamide degradation product acrolein. Drug Metab. Dispos. 27,
133-137.
(19) Yoshida, A., Rzhetsky, A., Hsu, L. C., and Chang, C. (1998)
Human aldehyde dehydrogenase gene family. Eur. J . Biochem.
251, 549-557.
(20) Berhane, K., and Mannervik, B. (1990) Inactivation of the
genotoxic aldehyde acrolein by human glutathione transferases
of classes alpha, pi, and mu. Mol. Pharmacol. 37, 251-254.
Ack n ow led gm en t. Wallace Laboratories are grate-
fully acknowledged for funding this project.
(21) Danielson, U. H., Esterbauer, H., and Mannervik, B. (1987)
Structure-activity relationships of 4-hydroxyalkenals in the
conjugation catalysed by mammalian glutathione transferases.
Biochem. J . 247, 707-713.
Refer en ces
(1) Thompson, C. D., Gulden, P. H., and Macdonald, T. L. (1997)
Identification of modified atropaldehyde mercapturic acids in rat
and human urine after felbamate administration. Chem. Res.
Toxicol. 10, 457-462.
TX000141E