Interaction of γ-glutamyltranspeptidase with clofibryl-S-acyl-glutathione in vitro and in vivo in rat

Article abstract of DOI:10.1021/tx010039x

Clofibric acid (CA) is metabolized to chemically reactive acylating products that can transacylate glutathione to form clofibryl-S-acyl-glutathione (CA-SG) in vitro and in vivo. We investigated the first step in the degradation of CA-SG to the mercapturic acid conjugate, clofibryl-S-acyl-N-acetylcysteine (CA-SNAC), which is catalyzed by γ-glutamyltranspeptidase (γ-GT). After γ-GT mediated cleavage of glutamate from CA-SG, the product clofibryl-S-acylcysteinylglycine (CA-S-CG) should undergo an intramolecular rearrangement reaction [Tate, S. S. (1975) FEBS Lett. 54, 319-322] to form clofibryl-N-acyl-cysteinylglycine (CA-N-CG). We performed in vitro studies incubating CA-SG with γ-GT to determine the products formed, and in vivo studies examining the products excreted in urine after dosing rats with CA-SG or CA. Thus, CA-SG (0.1 mM) was incubated with γ-GT (0.1 unit/mL) in buffer (pH 7.4, 25 °C) and analyzed for products formed by reversed-phase HPLC and electrospray mass spectrometry (ESI/MS). Results showed that CA-SG is degraded completely after 6 h of incubation leading to the formation of two products, CA-N-CG and its disulfide, with no detection of CA-S-CG thioester. After 36 h of incubation, only the disulfide remained in the incubation. Treatment of the disulfide with dithiothreitol led to the reappearance of CA-N-CG. ESI/LC/MS analysis of urine (16 h) extracts of CA-SG-dosed rats (200 mg/kg, iv) showed that CA-SG is degraded to CA-N-CG, CA-N-acyl-cysteine (CA-N-C) and their respective S-methylated products. The mercapturic acid conjugate (CA-SNAC) was found as a minor product. dAnalysis of urine extracts from CA-dosed rats (200 mg/kg, ip) resulted in the detection of clofibryl-N-acyl-cysteine (CAN-C), but no evidence for the formation of CA-SNAC was obtained. These in vitro and in vivo experiments indicate that γ-GT mediated degradation of clofibryl-S-acyl-glutathione leads primarily to the formation and excretion of clofibryl-N-acyl-cysteine products rather than the S-acyl-NAC conjugate.

Full text of DOI:10.1021/tx010039x

Chem. Res. Toxicol. 2001, 14, 1033-1040
1033
In ter a ction of γ-Glu ta m yltr a n sp ep tid a se w ith
Clofibr yl-S-a cyl-glu ta th ion e in Vitr o a n d in Vivo in Ra t
Mark P. Grillo† and Leslie Z. Benet*
Department of Biopharmaceutical Sciences, University of California,
San Francisco, California 94143-0446
Received February 22, 2001
Clofibric acid (CA) is metabolized to chemically reactive acylating products that can
transacylate glutathione to form clofibryl-S-acyl-glutathione (CA-SG) in vitro and in vivo. We
investigated the first step in the degradation of CA-SG to the mercapturic acid conjugate,
clofibryl-S-acyl-N-acetylcysteine (CA-SNAC), which is catalyzed by γ-glutamyltranspeptidase
(γ-GT). After γ-GT mediated cleavage of glutamate from CA-SG, the product clofibryl-S-acyl-
cysteinylglycine (CA-S-CG) should undergo an intramolecular rearrangement reaction [Tate,
S. S. (1975) FEBS Lett. 54, 319-322] to form clofibryl-N-acyl-cysteinylglycine (CA-N-CG). We
performed in vitro studies incubating CA-SG with γ-GT to determine the products formed,
and in vivo studies examining the products excreted in urine after dosing rats with CA-SG or
CA. Thus, CA-SG (0.1 mM) was incubated with γ-GT (0.1 unit/mL) in buffer (pH 7.4, 25 °C)
and analyzed for products formed by reversed-phase HPLC and electrospray mass spectrometry
(ESI/MS). Results showed that CA-SG is degraded completely after 6 h of incubation leading
to the formation of two products, CA-N-CG and its disulfide, with no detection of CA-S-CG
thioester. After 36 h of incubation, only the disulfide remained in the incubation. Treatment
of the disulfide with dithiothreitol led to the reappearance of CA-N-CG. ESI/LC/MS analysis
of urine (16 h) extracts of CA-SG-dosed rats (200 mg/kg, iv) showed that CA-SG is degraded
to CA-N-CG, CA-N-acyl-cysteine (CA-N-C) and their respective S-methylated products. The
mercapturic acid conjugate (CA-SNAC) was found as a minor product. Analysis of urine extracts
from CA-dosed rats (200 mg/kg, ip) resulted in the detection of clofibryl-N-acyl-cysteine (CA-
N-C), but no evidence for the formation of CA-SNAC was obtained. These in vitro and in vivo
experiments indicate that γ-GT mediated degradation of clofibryl-S-acyl-glutathione leads
primarily to the formation and excretion of clofibryl-N-acyl-cysteine products rather than the
S-acyl-NAC conjugate.
acyl-linked metabolites to protein has been proposed to
be a mechanism for the hypersensitivity reactions as-
sociated with the use of CA (3, 8-11). Transacylation of
the cysteinyl thiol of GSH by reactive clofibric acid
metabolites is likely to provide a detoxification mecha-
nism that leads to the formation of CA-SG. Evidence for
the formation of this glutathione conjugate in vivo comes
from studies where it has been detected in the bile of CA-
dosed rats (1), and, after conversion to the mercapturic
acid derivative clofibryl-S-acyl-N-acetylcysteine (CA-
SNAC), detected in the urine of CA-dosed patients (2).
In tr od u ction
Clofibric acid [2-(4-chlorophenoxyl)-2-methylpropionic
acid, CA],1 a lipid lowering drug, is metabolized to
reactive acylating derivatives (Scheme 1) that have been
shown to transacylate GSH forming clofibryl-S-acyl-
glutathione (CA-SG) in vivo and in vitro (1, 2). Metabo-
lism of CA by acyl glucuronidation, leading to clofibryl-
1-O-acyl glucuronide (1-O-CA-G), has been proposed to
be a route whereby CA becomes metabolically activated
to chemically reactive species that are able to bind
irreversibly to protein (3, 4) and endogenous nucleophiles
such as GSH (1). Another metabolic route by which CA
is converted to reactive acylating species is conversion
to clofibryl-S-acyl-CoA [CA-SCoA (5-7)], an intermediary
metabolite of CA which has also been shown to trans-
acylate GSH in vitro (6). Covalent binding of electrophilic
For those glutathione conjugates that are not excreted
into the bile, they must first be degraded in the kidneys
before excretion into the urine as the mercapturate. The
first step in the mercapturic acid pathway is catalyzed
by the enzyme γ-glutamyltranspeptidase (γ-GT), whereby
the enzyme cleaves off the γ-glutamyl portion of the
conjugate (12) resulting in cysteinylglycine S-derivatives.
Abundant levels of γ-GT are located in the proximal
tubular cells of the kidney. The conjugates then are
further processed to the corresponding cysteine S-deriva-
tives by cysteinylglycine dipeptidase (13) or aminopep-
tidase-M (14) by cleavage of the glycine moiety. Finally,
the cysteine S-derivatives are acetylated by cysteine
S-conjugate N-acetyltransferase using acetyl-CoA as a
cosubstrate to form the N-acetylcysteine conjugate (15).
* To whom correspondence should be addressed. Phone: (415) 476-
3853. Fax: (415) 476-8887. E-mail: benet@itsa.ucsf.edu.
† Present address: Pharmacia Corporation, Global Metabolism and
Investigative Sciences, Kalamazoo, MI 49007.
1
Abbreviations: CA, clofibric acid; CA-SG, clofibryl-S-acyl-gluta-
thione; γ-GT, γ-glutamyltranspeptidase; CA-SNAC, clofibryl-S-acyl-
N-acetylcysteine; CA-N-CG, clofibryl-N-acyl-cysteinylglycine; CA-S-CG,
clofibryl-S-acyl-cysteinylglycine; CA-N-C, clofibryl-N-acyl-cysteine; CA-
S-C, clofibryl-S-acyl-cysteine; CA-SCoA, clofibryl-S-acyl-CoA; 1-O-CA-
G, 1-O-acyl-clofibryl glucuronide; DTT, dithiothreitol; ECF, ethyl
chloroformate; THF, tetrahydrofuran; ESI, electrospray ionization;
SIM, selected-ion monitoring; NAC, N-acetylcysteine.
10.1021/tx010039x CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/13/2001

1034 Chem. Res. Toxicol., Vol. 14, No. 8, 2001
Grillo and Benet
Sch em e 1. P r op osed Sch em e for th e (A) Meta bolic Activa tion of Clofibr ic Acid F ollow ed by Rea ction w ith
GSH a n d (B) th e P a th w a ys in th e Degr a d a tion of CA-SG in Vivo, Wh ich In clu d es th e P r op osed S to N
Rea r r a n gem en t Rea ction th a t Occu r s a fter γ-Glu ta m yltr a n sp ep tid a se-Med ia ted Degr a d a tion of
S-Acyl-Lin k ed Glu ta th ion e Con ju ga tes (16). In ter m ed ia tes w ith Str u ctu r es Sh ow n in Br a ck ets Ha ve Not
Been Id en tified a
a
Structures of disulfide products, formed ex vivo, are not shown.
The interaction of γ-GT with S-acyl-linked glutathione
conjugates of lactic acid, acetic acid, and benzoic acid
leads to S-acylated cysteinylglycine products that have
been shown to undergo an intramolecular rearrangement
reaction from the S-acyl-derivative to the N-acyl-deriva-
tive in in vitro experiments with purified γ-GT enzyme
(16). The high reactivity of the thioester bond (17) of the
S-acylated cysteinylglycine product allows for the cys-
teinylamine to react in an intramolecular fashion with
the carbonyl carbon of the thioester linkage resulting in
the formation of a more stable N-acyl-cysteinylglycine
amide product. If such a rearrangement occurs in vivo
for γ-GT-mediated products of S-acyl-linked glutathione
conjugates, then the respective mercapturic acid deriva-
tives would not be expected to be the only degradation
products of the S-acyl-glutathione derivatives excreted
into the urine. Depending on the extent of rearrangement
of the S-acyl-cysteinylglycine products formed by degra-
dation of the S-acyl-glutathione conjugates by γ-GT in
the kidney, the relative amount of excreted mercaptur-
ates versus rearrangement products would vary. To our
knowledge, N-acylated cysteinylglycine conjugates have
not yet been detected as urinary products of S-acylated
glutathione conjugates. The formation of CA-SNAC in
humans represents the only example of an S-acyl-linked
mercapturate of an acidic drug found in man (2). Similar

Interaction of Clofibryl-Glutathione with γ-GT
Chem. Res. Toxicol., Vol. 14, No. 8, 2001 1035
reaction was terminated by the addition of concentrated HCl
(8 drops). The THF then was removed by evaporation under
reduced pressure and the remaining aqueous phase extracted
with diethyl ether (4 × 50 mL). After the first evaporation step
and subsequent extractions with diethyl ether, a white precipi-
tate, the S-acyl-glutathione conjugate, was formed. This pre-
cipitate was washed with distilled water (4 × 50 mL), to remove
remaining GSH, followed by acetone (4 × 50 mL), to wash away
remaining CA and water. Finally, the precipitate was dried
under a stream of nitrogen gas (25 °C, 1 h) to afford 161 mg of
a white solid (20% yield) of CA-SG. The poor yield of CA-SG
probably resulted from the washing procedure at the acetone-
wash step. Unlike most glutathione conjugates (20), CA-SG has
a slight solubility in acetone. HPLC analysis with UV detection
at 226 nm showed the synthetic CA-SG to be 100% pure. ESI
mass spectrometric analysis was performed for CA-SG, m/z (%):
504 (100%, MH+), 429 ([M + H - Gly]+, 4%), 375 ([M + H -
pyroglutamic acid]+, 9%), 307 ([GSH]+, 2%), 272 ([p-Cl-C6H4-
OC(CH3)2COSCH2-CHdNH2]+, 2%), 169 ([p-Cl-C6H4OC(CH3)2]+,
3%). These fragments are characteristic of those found during
the ESI/MS analysis of glutathione conjugates (21). 1H NMR
(2H6-DMSO): δ 1.42 (s, 6H, -C(CH3)2-), 2.28 (m, 2H, Glu-â,â′),
2.5 (m, 2H, Glu-γ,γ′), 2.95 (m, 2H, Cys-â,â′), 3.67 (t, 2H, J ) 6.4
Hz, Glu-R), 3.72 (s, 2H, Gly-R,R′), 4.46 (m, 1H, Cys-R), 6.9-7.4
(m, 4H, Ar).
experiments performed in our laboratory could not find
in human urine (unpublished observations)2 S-acyl-linked
mercapturates of ibuprofen and tolmetin, two carboxylic-
acid-containing drugs known to form reactive acyl glu-
curonides (18, 19). Therefore, we hypothesized that the
lack of detection of the respective S-acylated mercaptu-
rates may have resulted from an extensive rearrange-
ment reaction of the γ-GT mediated S-acyl-cysteinyl-
glycine products to the N-acylated cysteinylglycine de-
rivatives of these drugs.
In the present work, we examined the degradation of
CA-SG by γ-GT in vitro with purified enzyme to deter-
mine the extent of intramolecular rearrangement of the
S-acylated cysteinylglycine intermediate. Second, the
degradation of synthetic CA-SG in vivo in CA-SG-dosed
rats was studied by looking particularly for the urinary
S-linked mercapturate and N-acylated cysteinylglycine
rearrangement products. Finally, we examined the urine
extracts from CA-dosed rats for derivatives related to the
γ-GT mediated degradation of biologically formed CA-
SG.
Exp er im en ta l Section
Syn th esis of CA-SNAC Th ioester . CA-SNAC was obtained
by reacting CA-SG (1 mM) with NAC (20 mM) in buffer (pH
7.4, 37 °C) in a volume of 10 mL until the reaction was complete,
as indicated by HPLC analysis of the reaction mixture. The
incubation then was acidified (pH 2.5) by the addition of HCl
(1 N), followed by extraction of the CA-SNAC conjugate with
ethyl acetate (1 × 10 mL). The ethyl acetate layer was dried
(anhydrous MgSO4) and evaporated (N2 gas, 25 °C) to afford
2.7 mg (75% yield) of product as a clear colorless oil. HPLC
analysis of the CA-SNAC product showed it to have less than
1% impurity of CA. ESI/MS for CA-SNAC, m/z (%): 360 (MH+,
11%), 382 (MNa+, 318 (MH+-CH3CO, 3%), 169 ([p-Cl-C6H4-
OC(CH3)2]+, 100%).
Ch em ica ls. CA, GSH, dithiothreitol (DTT), N-acetylcysteine
(NAC), cysteinylglycine, and bovine kidney γ-glutamyltrans-
peptidase (γ-GT) were purchased from Sigma Chemical Co. (St.
Louis, MO). Triethylamine, ethyl chloroformate, monobasic
potassium phosphate, potassium bicarbonate, methyl iodide, and
THF (anhydrous) were purchased from Aldrich Chemical Co.
(Milwaukee, WI). CA-SG and CA-SNAC were synthesized as
described below. All solvents used for HPLC were of chroma-
tography grade.
In str u m en ta tion a n d An a lytica l Meth od s. HPLC was
carried out on a Shimadzu LC-600 isocratic system coupled to
a Shimadzu SPD-6AV UV-Vis detector. All isocratic HPLC
analyses were performed on a reversed-phase column (Beckman
C8, 15 cm, 5 µm, 1 mL/min). Electrospray mass spectrometry
of synthetic standards and in vitro biological extracts was
performed on a Hewlett-Packard HP 1100 LC/MSD benchtop
electrospray mass spectrometer. Analysis was conducted in the
positive ion mode at a fragmentor voltage of 80 and by direct
infusion of the sample [dissolved in methanol/1% acetic acid in
distilled water (50/50)] into the ion source. ESI/LC/MS analysis
of reconstituted urine extracts was conducted by gradient
elution (from 40 to 70% methanol in 1% aqueous acetic acid over
30 min) on a Beckman C8 reverse-phase column (15 cm × 4.6
mm, 5 µm, 1 mL/min), which functioned to separate the
compounds-of-interest. 1H NMR spectra were recorded on a
General Electric QE300 spectrometer operating at 300 MHz.
Chemical shifts are reported in parts per million as referenced
to the residual solvent peak (2.49 ppm for 2H6-DMSO).
Syn th esis of CA-SG. Synthesis of CA-SG was performed by
conventional procedures employing ethyl chloroformate (ECF)
(20). Briefly, to CA (342 mg, 1.6 mmol) dissolved in anhydrous
THF (25 mL) was added, at room temperature and while
stirring, triethylamine (220 µL, 1.6 mmol) followed by ECF (100
µL, 1.6 mmol). After 30 min of continued stirring, the precipitate
which formed (triethylamine hydrochloride) was removed by
passing through a glass funnel, fitted with a glass wool plug,
and added directly into a solution containing GSH (500 mg, 1.6
mmol), KHCO3 (100 mg, 1.6 mmol), distilled water (10 mL), and
THF (15 mL). The solution was stirred continuously under
nitrogen gas at room temperature for 2 h, after which the
In Vitr o Exp er im en ts With γ-GT a n d CA-SG. Synthetic
CA-SG (0.1 mM) was incubated with γ-glutamyltranspeptidase
(type II from bovine kidney, 0.1 unit/mL) in potassium phos-
phate buffer (0.05 M, pH 7.4, total volume of 4 mL, 25 °C,
duplicate incubations) in screw-capped glass vials in a shaking
incubator. One unit of the transpeptidase will liberate 1 mmol
of p-nitroaniline from L-γ-glutamyl-p-nitroaniline/min at pH
8.5 and 25 °C, as reported by the supplier. The γ-GT prepara-
tion, from the supplier, was reported to have less than 0.5%
creatine phosphokinase, glutamic-oxalacetic transaminase,
glutamic pyruvic transaminase activity, but with no mention
of any dipeptidase activity. The γ-GT preparation was used
without further purification. Aliquots (50 µL) of the incubation
were taken at times 0, 1, 2, 3, 4, 5, 6, 7, 8, and 36 h and were
analyzed directly by isocratic reversed-phase HPLC (50%
methanol in 0.05 M potassium phosphate, pH 4.2) for the loss
of CA-SG and the respective formation of CA-S-CG, CA-N-CG
or its disulfide. CA-SG and degradation products were detected
by UV absorbance at 226 nm. Quantitative measurements were
made using a standard curve generated from absolute peak
areas of CA-SG. In the present experiments, it was assumed
that the UV characteristics of CA-SG and its degradation
products are the same. At the 5 and 36 h time points of
incubation, aliquots (1 mL) of the incubations were acidified to
pH 2.5 with HCl (1 N) and then extracted with ethyl acetate (1
× 5 mL). The extracts were then dried (MgSO4) and evaporated
to dryness under a stream of N2 gas. Identification of the
products formed was obtained by ESI/MS analysis of the
reconstituted (50% methanol in distilled water with 1% acetic
acid, 0.5 mL) by direct infusion of the sample into the ion source
(25 µL/min) using a syringe pump (Applied Biosystems, Foster
City, CA).
2
Preliminary in vivo experiments were performed with ibuprofen
and tolmetin, where, after an oral dose of 600 or 400 mg to volunteers,
respectively, 8 h urines were collected, processed, and analyzed for
S-acyl-mercapturic acid conjugates by reported procedures (2). ESI/
LC/MS and analysis of the urine extracts did not show the presence
of S-acyl-mercapturates relative to authentic standards.
In Vivo Exp er im en ts in Ra ts. Male Sprague-Dawley rats
(200-220 g) were given doses of CA-SG (200 mg/kg, iv, tail vein)

1036 Chem. Res. Toxicol., Vol. 14, No. 8, 2001
Grillo and Benet
F igu r e 2. ESI/MS positive ion mass spectrum of in vitro
formed CA-N-CG obtained from extracts of the incubation of
CA-SG (0.1 mM, pH 7.4, 25 °C) with γ-GT (0.1 unit/mL) after 1
h of incubation. The origins of the characteristic ions are as
shown.
F igu r e 1. Representative reversed-phase isocratic HPLC chro-
matograms of 50 µL of the in vitro reaction mixture containing
CA-SG (0.1 mM, pH 7.4, 25 °C, retention time of 5 min) and
γ-GT (0.1 unit/mL) after 0, 1, and 36 h of incubation. The bottom
HPLC chromatogram was obtained 30 min after the addition
of DTT (1 mM final concentration) to the incubation at the 36
h time-point. DTT functioned to reduce the CA-N-CG disulfide
(retention time of 53 min) to the free cysteinyl sulfhydryl form
(retention time of 7 min).
containing clofibric acid show the characteristic 35Cl and
37Cl isotope cluster pattern. The mass spectrum of the
peak that eluted at 5 min was consistent with the
degradation product having lost glutamate and giving a
protonated molecular (MH+) ion at m/z 375 (less 129
AMU) and the corresponding sodium and potassium
adduct ions at m/z 397 and 413, respectively. The mass
spectrum also shows major ion fragments at m/z 300 (loss
of glycine), 272, 247 and 169, which are consistent with
its structure (Figure 2). The spectrum cannot differenti-
ate between the N-acyl- or S-acyl-cysteinylglycine prod-
ucts, although the data showing that the product readily
oxidizes to the disulfide (Figure 1) provides strong
evidence for the formation of the rearranged N-acyl
product.
or CA (200 mg/kg, ip), both dissolved in phosphate-buffered
saline (0.5 mL, pH 7.0). Sixteen hours postadministration,
during which time the animals (one rat used for each treatment)
were kept unrestrained in metabolic cages, total urines (∼15
mL) were collected, acidified (pH 2.5, 1 N HCl), and extracted
with ethyl acetate (3 × 15 mL). The combined extracts were
dried (MgSO4) and evaporated to dryness with N2 gas at room
temperature. Residues were dissolved in 50% methanol in
potassium phosphate buffer (0.05 M, pH 4.2, 4 mL) and vortex
mixed. Portions (100 µL) of dissolved extracts were analyzed
by reversed-phase gradient ESI/LC/MS with in-line UV detec-
tion (226 nm).
Analysis of extracts of the incubation mixture after 36
h of incubation by ESI/MS gave a protonated molecular
ion (MH+) at m/z 747, as well as the corresponding
sodium (m/z 769) and potassium (m/z 785) adduct ions,
which are consistent with the molecular weight of the
CA-N-CG disulfide (Figure 3). The disulfide is more
stable than the reduced product under the ESI/MS
conditions used as indicated by the lack of intense
fragment ions produced during the ESI/MS analysis.
Resu lts
HP LC An a lysis of th e P r od u cts F or m ed in In cu -
ba tion s of CA-SG w ith γ-GT. Incubation of CA-SG (0.1
mM) with γ-GT (0.1 unit/mL) led to the formation of two
products, the rearrangement product clofibryl-N-acyl-
cysteinylglycine (CA-N-CG) and its disulfide. Reversed-
phase isocratic HPLC analysis of 50 µL of the incubation
mixture (Figure 1) shows that CA-SG elutes before CA-
N-CG, having retention times of 5 and 7 min, respec-
tively. The disulfide of CA-N-CG elutes much later as a
broad peak at approximately 53 min. Treatment of the
incubation mixture, at the 36 h time point, with dithio-
threitol (DTT, final concentration 1 mM, 30 min at 25
°C) led to the complete disappearance of the disulfide
peak and the reappearance of the CA-N-CG free cysteinyl
sulfhydryl product. Electrospray mass spectrometric
analysis of an extract of the reaction mixture at the 1 h
time point gave a mass spectrum as shown in Figure 2.
Mass spectra for all degradation products and fragments
The time course of the degradation of CA-SG, with
aliquots of the incubation mixture taken every hour for
24 h, showed the glutathione derivative to be completely
degraded after 6 h of incubation (Figure 4). The re-
arranged free sulfhydryl N-acyl-cysteinylglycine product
reached its maximum concentration (63 µM) at 6 h, but
then slowly oxidized to the disulfide after about 24 h of
incubation.
In Vivo Exp er im en ts w ith CA-SG. Sixteen hours
after dosing rats with CA-SG (200 mg/kg, iv, tail vein),
total urine (0-16 h, ∼15 mL) was collected and the
extract analyzed by reversed-phase ESI/LC/MS with in-
line UV detection. Analysis of the urine extract at a UV

Interaction of Clofibryl-Glutathione with γ-GT
Chem. Res. Toxicol., Vol. 14, No. 8, 2001 1037
F igu r e 5. Representative reversed-phase gradient HPLC
chromatogram with UV detection of urine extracts from CA-
SG dosed rats. An aliquot of the reconstituted urine extract (100
µL) was injected onto a Beckman C8 column and analyzed by
in-line UV (A226) and ESI/MS detection. The peaks-of-interest
are labeled 1 (S-methyl-CA-N-CG), 2 (CA), 4 (S-methyl-CA-N-
C), 5 (CA-N-CG disulfide), 6 (CA-N-CG/CA-N-C mixed disulfide),
and 7 (CA-N-C disulfide). The label 3 (CA-SNAC) indicates the
approximate retention time of the authentic mercapturate (11.2
min). The biologically formed CA-SNAC, as shown, was not
detected in the extract by UV analysis.
F igu r e 3. ESI/MS positive ion mass spectrum of in vitro
formed CA-N-CG disulfide obtained from extracts of the incuba-
tion of CA-SG (0.1 mM, pH 7.4, 25 °C) with γ-GT (0.1 unit/mL)
after 36 h of incubation. The origins of the characteristic ions
are as shown.
(not shown) of peak 1 (retention time 10.3 min) is
consistent with the structure being S-methyl-clofibryl-
N-acyl-cysteinylglycine (S-methyl-CA-N-CG, Figure 6A,
MH+ at m/z 389, with fragments at m/z of 169, 261, 186,
and 314). A synthetic S-methyl-CA-N-CG conjugate
standard was synthesized by reacting CA-N-CG free
sulfhydryl, obtained by reducing CA-N-CG disulfide with
DTT, with methyl iodide in buffer at pH 7.4. Confirma-
tion of the structure of this product was obtained by
comparing its mass spectrum with that of the synthetic
derivative, which were identical, and by having the same
HPLC retention times (data not shown).
The mass spectrum of peak 2 (retention time 11.0 min)
showed it to be identical to that of standard CA (Figure
6B). CA proved to be poorly suited to analysis by ESI/
MS by giving a weak mass spectrum (not shown) with
no MH+ ion detected. The mass spectra of CA did provide
major fragment ions at m/z 169 (loss of CO2), which is a
characteristic fragment for clofibric acid derivatives, and
m/z 129 (loss of dimethylacetic acid).
Authentic mercapturic acid conjugate, CA-SNAC, elutes
at a retention time of ∼11.2 min (just after CA and
approximately where the label 3 is placed on the chro-
matogram in Figure 5), but the biologically formed
product was not detected by UV analysis of the CA-SG-
dosed rat urine extract (Figure 5). Equal amounts of CA-
SNAC and CA give the same peak areas when analyzed
by HPLC at 226 nm (data not shown), indicating the
relatively low amount of CA-SNAC formed in vivo upon
administration of CA-SG. The mercapturate was detected
in the extract by ESI/MS detection, in that baseline
subtraction of the ESI/LC/MS chromatogram around the
11.2 min retention time provided a mass spectrum of CA-
F igu r e 4. Time course of the degradation of CA-SG (0.1 mM)
by γ-GT (0.1 unit/mL) in buffer (0.05 M potassium phosphate,
pH 7.4, 25 °C). Values are expressed as the average of duplicate
experiments.
absorbance of 226 nm showed the formation of substances
eluting after a retention time of 10 min, which were the
focus of the present study (Figure 5). The mass spectrum

1038 Chem. Res. Toxicol., Vol. 14, No. 8, 2001
Grillo and Benet
F igu r e 7. Representative reversed-phase gradient HPLC
chromatogram with UV detection of urine extracts from CA-
dosed rats. An aliquot of the reconstituted urine extract (100
µL) was injected onto a Beckman C8 column and analyzed by
in-line UV (A226) and ESI/MS detection. Peak 7 (labeled on the
chromatogram) refers to the retention time of the CA-N-C
disulfide product (shown as a very small peak).
(not shown) indicative for the disulfide of CA-N-C (Figure
6F) by providing masses at m/z 633 (MH+) and m/z 505
(MH+-p-chlorophenol).
In Vivo Exp er im en ts w ith CA. Sixteen hours post-
administration of CA (200 mg/kg, ip), total urine (∼15
mL) was collected, processed, and analyzed for gluta-
thione related metabolites. Analysis of a portion (100 µL)
of the reconstituted extract by ESI/LC/MS, with in-line
UV detection, showed the presence (by UV detection) of
a very small peak at 17.0 min (Figure 7). This substance
has the same HPLC retention time as the CA-N-C
disulfide derivative (peak 7, Figure 5) found in the CA-
SG-dosed rat urine extract. The ESI/MS mass spectrum
(not shown) of this in vivo metabolite of CA was identical
to the mass spectrum of the in vivo product detected in
the CA-SG-dosed rat urine extract (peak 7, Figures 5 and
6F).
Analysis of the urine extracts from CA-SG and CA
treated rats by selected-ion monitoring (SIM) gave the
ESI/LC/MS chromatogram shown in Figure 8. Selected-
ions monitored were for the protonated molecular ions
of peaks 1, 3, 4, 5, 6, and 7, and were as follows: m/z
389 (1, S-methyl-CA-N-CG, Figure 6A), 360 (3, CA-
SNAC, Figure 6C), 332 (4, S-methyl-CA-N-C, Figure 6D),
747 (5, CA-N-CG disulfide, Figure 3), 690 (6, mixed
disulfide of CA-N-CG and CA-N-C, Figure 6E), and 633
(7, CA-N-C disulfide, Figure 6F). SIM analysis of the CA-
SG urine extract gave a chromatogram with a similar
profile of eluting substances as detected by UV analysis
at 226 nm (Figure 5). The chromatogram also showed
the presence of CA-SNAC (3) in the extract eluting at
11.0 min, the same retention time of the CA-SNAC
synthetic standard when analyzed during the same SIM
experiment (data not shown). Analysis of the CA-dosed
rat urine extract showed peak 7 (CA-N-C disulfide) as
the most abundant ion. Other peaks in this SIM chro-
F igu r e 6. Structures and proposed ESI/MS fragmentation of
the in vivo formed products 1, 2, 3, 4, 6, and 7, isolated from
the urine of a CA-SG-dosed rat.
SNAC (not shown) that was identical to the synthetic
standard. The mass spectrum of the biologically formed
CA-SNAC (3) exhibited an MH+ ion at m/z 360 and one
major fragment ion at m/z 169, the characteristic CA
fragment (Figure 6C).
The mass spectrum (not shown) of peak 4 eluting at
12.4 min (Figure 5) provided evidence for the formation
of S-methyl-clofibryl-N-acyl-cysteine (S-methyl-CA-N-C)
by showing the MH+ ion at m/z 332 and fragment ions
at m/z 204 (loss of p-chlorophenol), and m/z 169 (Figure
6D).
The disulfide of CA-N-CG (peak 5) eluted at 15.0 min
as a minor peak (Figure 5), which is the same HPLC
retention time of the disulfide formed upon in vitro
degradation of CA-SG by γ-GT (data not shown). The
mass spectrum (not shown) for this disulfide, provided
ions at m/z 747 (MH+), 769 and 785 (sodium and
potassium adduct ions), and 672 (MH+-75, loss of glycine),
which was identical to the spectrum shown for the in
vitro formed product (Figure 3).
The substance labeled peak 6 eluting at 15.8 min
(Figure 5) gave an ESI mass spectrum (not shown),
containing the MH+ ion at m/z 690, and a fragment ion
at m/z 615 (MH+-75, loss of glycine). The mass spectrum
is consistent with the substance being the mixed disulfide
of CA-N-CG and clofibryl-N-acyl-cysteine (CA-N-C) (Fig-
ure 6E).
Finally, the major glutathione related product (peak
7) eluting at 17.0 min (Figure 5) gave a mass spectrum

Interaction of Clofibryl-Glutathione with γ-GT
Chem. Res. Toxicol., Vol. 14, No. 8, 2001 1039
reported to be only 0.1% of the amount of 1-O-CA-G
excreted in bile of CA-dosed rats (1). Unlike the observa-
tions reported for CA, we have searched unsuccessfully
for mercapturic acid conjugates of carboxylic-acid-con-
taining drugs as a marker for metabolism to reactive acyl
glucuronides or acyl-CoA thioester derivatives in vivo.
A report by Tate (16) offered an explanation for our
finding by showing that γ-glutamyltranspeptidase (γ-GT),
the enzyme that catalyzes the first step in the degrada-
tion of glutathione conjugates to mercapturic acid deriva-
tives, interacts with S-acyl derivatives of glutathione to
yield N-acyl-cysteinylglycine conjugates. The degradation
of the S-acyl-derivative by γ-GT, via cleavage of the
γ-glutamyl group, leads to S-acyl-cysteinylglycine prod-
ucts that quickly and quantitatively convert to the
N-acyl-dipeptide conjugate (Scheme 1) (16). S-Acyl-gluta-
thione conjugates used by Tate (16), where colorimetric
techniques were employed to detect the free sulfhydryl
formed in the S to N transfer, included S-acyl-glutathione
conjugates of acetic, lactic, and benzoic acid. We propose
that the rearrangement reaction that occurs in vitro for
these S-acyl-thioester derivatives via the transpeptidase
may also occur in vivo for glutathione thioester detoxi-
fication products of reactive acidic drug metabolites such
as CA-1-O-G or CA-SCoA. We propose that rearrange-
ment of dipeptide S-acyl-cysteinylglycine conjugates to
N-acyl-cysteinylglycine derivatives would preclude fur-
ther degradation to the corresponding mercapturates and
explain their lack of detection.
F igu r e 8. Representative reverse-phase gradient ESI/LC/MS
SIM chromatogram of (top) CA-SG- and (bottom) CA-dosed rat
urine extracts. The labeled peaks-of-interest are 1 (S-methyl-
CA-N-CG), 3 (CA-SNAC), 4 (S-methyl-CA-N-C), 5 (CA-N-CG
disulfide), 6 (CA-N-CG/CA-N-C mixed disulfide), and 7 (CA-N-C
disulfide). Selected-ions used corresponded to the protonated
molecular ions of substances 1, 3, 4, 5, 6, and 7 (m/z 389, 360,
332, 747, 690, and 633, respectively). CA (2) was not measured
during the SIM ESI/LC/MS analysis of the urine extracts.
In the present studies, we focused on the γ-GT medi-
ated degradation of the GSH thioester conjugate of
clofibric acid in vitro and in vivo. We found that CA-SG
was degraded by γ-GT in vitro to give the rearranged CA-
N-CG product rather than the S-acylated CA-S-CG
dipeptide. This conclusion was based on ESI/MS data
showing the protonated molecular ion (m/z 375) having
lost 129 amu (loss of γ-glutamic acid) and that the
rearrangement product readily oxidizes to its disulfide,
which indicates the presence of a free cysteinyl sulf-
hydryl. The reaction that occurs is an intramolecular
transacylation reaction between the cysteinylamine and
the carbonyl-carbon of the thioester linkage forming the
more stable amide. It has been proposed that γ-GT can
facilitate the S to N rearrangement reaction (16), al-
though we found that the intermediate S-acyl-thioester
can rearrange rapidly in the absence of the γ-GT enzyme.
We reached this conclusion since attempts to synthesize
CA-S-CG thioester, by reacting clofibryl acyl-chloride
with cysteinylglycine at pH 7.4, resulted only in the
formation of the rearrangement product, CA-N-CG,
(unpublished observations).3 All together, these results
are in agreement with the literature report by Tate (16)
indicating that the rearranged N-acyl-cysteinylglycine
conjugate is formed in vitro by γ-GT-mediated degrada-
tion of CA-SG.
matogram were not different than analysis performed of
control rat urine extracts (data not shown).
Discu ssion
It has been proposed that CA is metabolized to reactive
metabolites that can acylate proteins and lead to the
associated hepatotoxic effects (1-4). Greatest attention
has focused on the acyl glucuronide metabolite of the
drug, 1-O-CA-G, in view of reports that this metabolite
transacylates the nucleophilic cysteinyl-thiol of GSH, in
vitro in buffer, to form CA-SG (Scheme 1) (1). In addition,
studies have been performed showing that the corre-
sponding mercapturic acid conjugate, CA-SNAC, is ex-
creted from the urine of CA-dosed volunteers (2). The
proposed metabolic activation of CA by acyl glucuronida-
tion is illustrated in Scheme 1. Also shown in this scheme
is another proposed, and understudied, bioactivation
mechanism, namely acyl-CoA formation, resulting in CA-
SCoA (5, 6). We reported that this thioester metabolite
is also reactive with GSH forming CA-SG (6). According
to Scheme 1, CA, after conversion to the acyl glucuronide
or acyl-CoA thioester, reacts with GSH to form the
glutathione thioester CA-SG. This thioester glutathione
derivative, a proposed detoxification product (2), then is
either excreted unchanged in bile (1) or undergoes
sequential enzyme catalyzed degradation steps to the
mercapturic acid conjugate, prior to excretion in the urine
[Scheme 1 (2)]. Although of mechanistic importance in
understanding chemical reactivity of CA metabolites, the
formation of CA-SG is only of minor quantitative impor-
tance, in that the amount of CA-SG excreted in bile was
To directly examine the degradation of CA-SG in vivo,
and to bypass the bioactivation steps (Scheme 1), we
administered CA-SG (iv) to rats and looked for the
degradation products excreted in urine. ESI/LC/MS
analysis with in-line UV detection (Figure 5) showed that
CA-SG, in addition to being hydrolyzed to CA, is degraded
3 An attempt to synthesize CA-S-CG thioester was made by reacting
clofibryl acyl-chloride (1 mM) (obtained quantitatively by conventional
procedures employing oxalyl chloride) with cysteinylglycine (1 mM)
in phosphate buffer (0.1 M, pH 7.4 and 25 °C).

1040 Chem. Res. Toxicol., Vol. 14, No. 8, 2001
Grillo and Benet
by γ-GT to products that are consistent with the re-
arrangement of the unstable S-acyl-cysteinylglycine in-
termediary product. As shown in Scheme 1, CA-SG is
cleaved by γ-GT forming the unstable, and unidentified,
S-acylated derivative. This derivative then undergoes an
S to N transfer forming the N-acylated dipeptide, CA-
N-CG. The subsequent formation of clofibryl-N-acyl-
cysteine (CA-N-C) could happen by two routes, including
dipeptidase mediated cleavage of glycine from CA-N-CG,
or from CA-S-CG followed by an S to N rearrangement
of CA-S-C thioester (Scheme 1). Since such rearrange-
ment reactions occur rapidly (16), we believe that CA-
N-CG serves as the actual substrate of the dipeptidase
enzymes, although we have no knowledge of N-acetylated
derivatives of cysteinylglycine being degraded by dipep-
tidases. The clofibryl-N-acylated products were also found
to be metabolized to their corresponding S-methylated
derivatives, S-methyl-CA-N-CG (1) and S-methyl-CA-N-C
(4). S-Methylation occurs in general for degradation
products of glutathione conjugates and is mediated by
S-methyltransferase enzymes. The free sulfhydryl forms
of CA-N-CG and CA-N-C were not found in the urine
extracts as such, but were detected as disulfide products.
These disulfides were most likely formed ex vivo in the
urine collection tubes that were left open to air and
untreated, but not formed in vivo. The classic urinary
mercapturic acid conjugate, CA-SNAC, was found only
as a minor metabolite of CA-SG. The major degradation
product was clofibryl-N-acyl-cysteine disulfide and, there-
fore, should be the metabolite-of-interest to detect in the
urine extracts of rats treated with CA.
Refer en ces
(1) Shore, L. J ., Fenselau, C., King, A. R., and Dickinson, R. G. (1995)
Characterization and formation of glutathione conjugate of
clofibric acid. Drug Metab. Dispos. 23, 119-123.
(2) Stogniew, M., and Fenselau, C. (1982) Electrophilic reactions of
acyl-linked glucuronides. Drug Metab. Dispos. 10, 609-613.
(3) Sallustio, B. C., Knights, K. M., Roberts, B. J ., and Zacest, R.
(1991) In vivo covalent binding of clofibric acid to human plasma
proteins and rat liver proteins. Biochem. Pharmacol. 42, 1421-
1425.
(4) Bailey, M. J ., and Dickinson, R. G. (1996) Chemical and im-
munological comparison of protein adduct formation of four
carboxylate drugs in rat liver and plasma. Chem. Res. Toxicol. 9,
659-666.
(5) Lygre, T., Aarsaether, N., Stensland, E., Aarsland, A., and Berge,
R. K. (1986) Separation and measurement of clofibryl coenzyme
A and clofibric acid in rat liver after clofibrate administration by
reverse-phase high-performance liquid chromatography with pho-
todiode array detection. J . Chromatogr. 381, 95-105.
(6) Grillo, M. P., and Benet, L. Z. (1995) In vitro studies on the
reactivity of clofibryl-S-acyl-CoA thioester with glutathione.
Pharm. Res. 13 (Suppl.), S-401.
(7) Caldwell, J . (1984) Xenobiotic acyl-coenzymes A: critical inter-
mediates in the biochemical pharmacology and toxicology of
carboxylic acids. Biochem. Soc. Trans. 12, 9-11.
(8) Spahn-Langguth, H., and Benet, L. Z. (1992) Acyl glucuronides
revisited: Is the glucuronidation process a toxification as well as
a detoxification mechanism? Drug Metab. Rev. 24, 5-47.
(9) Benet, L. Z., Spahn-Langguth, H., Iwakawa, S., Volland, C.,
Mizuma, T., Mayer, S., Mutschler, E., and Lin, E. T. (1993)
Predictability of covalent binding of acidic drugs in man. Life Sci.
53, PL141-146.
(10) Faed, E. M. (1984) Properties of acyl glucuronides: Implications
for studies of the pharmacokinetics and metabolism of acidic
drugs. Drug Metab. Rev. 15, 1213-1249.
(11) Grubb, N., Weil, A., and Caldwell, J . (1993) Studies on the
reactivity of clofibryl and fenofibryl glucuronides. Biochem. Phar-
macol. 46, 357-364.
(12) Tate, S. S., and Meister, A. (1985) γ-Glutamyltranspeptidase from
kidney. Methods Enzymol. 113, 400-419.
(13) Hirota, T., Nishikawa, Y., Komai, T., Igarashi, T., and Kitagawa,
H. (1987) Role of dehydropeptidase-I in the metabolism of
glutathione and its conjugates in the rat kidney. Res. Commun.
Chem. Pathol. Pharmacol. 56, 235-242.
(14) Kozak, E. M., and Tate, S. S. (1982) Glutathione-degrading
enzymes of microvillus membranes. J . Biol. Chem. 257, 6311-
6377.
Finally, during the ESI/LC/MS analysis of urine ex-
tracts from CA-dosed rats for the presence of CA-SG
degradation products, no mercapturic acid conjugate was
detected (Figure 8). As predicted by observations from
studies showing the urinary products from CA-SG-dosed
rats, the disulfide (7) of the rearranged γ-GT/dipeptidase
product, CA-N-C, was the only observed glutathione-
derived conjugate detected in urine. This is the first
example in our laboratory of a carboxylic acid-containing
drug being excreted as a glutathione-related derivative,
and, we believe, the first example of an N-acyl-cysteine
amide metabolite of an acidic drug.
(15) Greene, R. M., and Elce, J . S. (1975) Acetylation of S-substituted
cysteines by a rat liver and kidney microsomal N-acetyl trans-
ferase. Biochem. J . 147, 283-289.
In summary, clofibric acid S-acyl-glutathione has been
shown to undergo a γ-GT mediated degradation in vitro
and in vivo to S-acylated products that are unstable and
rearrange from S- to N-acyl-cysteine amide derivatives.
The formation of the mercapturic acid conjugate of CA
occurs only to a minor extent, and therefore may not be
the metabolite that should be searched for in urine
extracts as a marker for reactive metabolite formation
in vivo. From the results of the present studies on the
degradation of CA-SG thioester by γ-GT, we propose that
drug-N-acyl-cysteine derivatives be used as markers of
S-acyl-glutathione conjugate formation occurring in vivo,
especially when studying urine extracts from patients
dosed with carboxylic acid-containing drugs.
(16) Tate, S. S. (1975) Interaction of γ-glutamyltranspeptidase with
S-acyl derivatives of glutathione. FEBS Lett. 54, 319-322.
(17) Huxtable, R. (1986) Thiols, disulfides and thioesters. In Biochem-
istry of Sulfur (Frieden, E., Ed.) pp 230-245, Plenum Press, New
York.
(18) Castillo, M., and Smith, P. C. (1995) Disposition and reactivity
of ibuprofen and ibufenac acyl glucuronides in vivo in the rhesus
monkey and in vitro with human serum albumin. Drug Metab.
Dispos. 23, 566-572.
(19) Ojingwa, J . C., Spahn-Langguth, H., and Benet, L. Z. (1994)
Irreversible binding of tolmetin to macromolecules via its glu-
curonide: binding to blood constituents, tissue homogenates and
subcellular fractions in vitro. Xenobiotica 24, 495-506.
(20) Stadtman, E. R. (1957) Preparation and assay of acyl coenzyme
A and other thiol esters; use of hydroxylamine. Methods Enzymol.
3, 931-946.
(21) Baillie, T. A., and Davis, M. R. (1993) Mass spectrometry in the
analysis of glutathione conjugates. Biol. Mass Spectrom. 22,
319-325.
Ack n ow led gm en t. We thank Mrs. Milagros Han for
assistance in performing HPLC. This work was supported
in part by National Institute of Health Grant GM36633.
TX010039X