(CANCER RESEARCH 50, 2747-2752. May I. 1990)
Selectivity of Rat and Human Glutathione S-Transferases in Activation of Ethylene
Dibromide by Glutathione Conjugation and DNA Binding and Induction of
Unscheduled DNA Synthesis in Human Hepatocytes1
Joan L. Cmarik, Philip B. Inskeep,2 Michael J. Meredith,3 David J. Meyer, Brian Ketterer, and
F. Peter Guengerich4
Department of Biochemistry and Center in Molecular Toxicology, Vanderbiit university School of Medicine, Nashville, Tennessee 37232 [J. L. C., P. B. I., M. J. M.,
F. P. G.I, and CRC Molecular Toxicology Research Group, Middlesex School of Medicine, Cleveland St., London, H IP6DB. L'nited Kingdom ¡D.J. M., B. KJ
ABSTRACT
The major DNA adduct formed by the carcinogen ethylene dibromide
(EDB) is S-|2-(A"-guanyl)ethyI|glutathione. This adduct results from
the glutathione 5-transferase (GST)-catalyzed conjugation of EDB with
glutathione (GSH), which generates an episulfonium ion capable of
reacting with cellular nucleophiles. Purified rat and human GST enzymes
were compared for their ability to conjugate EDB with GSH and dis
played high selectivity. Of the six forms of rat GST tested, conjugation
was catalyzed by the a class enzyme 2-2 and, to a lesser extent, by the u
class enzyme 3-3. Of the three classes of cytosolic human GST, EDB
conjugation was catalyzed by the a class enzymes. Three dimers of the
human a class (a,-a„<».,-«,,
and ay-ay) were separated by chromatofocusing. The »,-<*,
preparation demonstrated the highest specific activity.
Rat microsomal GST had negligible activity for the conjugation of EDB
with GSH. The levels of EDB-DNA adducts formed in rat and human
hepatocytes were compared. DNA was isolated from both rat and human
hepatocytes incubated with 0.5 HIMEDB, and the level of DNA adduct
formation in the human samples was about 40% of that in the rat
hepatocytes. EDB concentration-dependent unscheduled DNA synthesis
was demonstrated in isolated human hepatocytes. Concurrent treatment
of the hepatocytes with diethylmaleate to deplete intracellular GSH
inhibited EDB-induced unscheduled DNA synthesis. These results indi
cate that EDB alkylates DNA in human hepatocytes and that enzymatic
repair of adducts may occur. The results of experiments done in rat and
human systems using both purified GST enzymes and intact hepatocytes
imply that the genotoxic pathway of EDB metabolism in rats and humans
is similar.
INTRODUCTION
GST5 is generally recognized as an important
detoxifying
enzyme, but occasionally it generates harmful reactive inter
mediates (1). One such instance is the GST-catalyzed conjuga
tion of GSH with EDB (2). EDB, formerly used as a fumigant
and as an anti-knock additive in gasoline, is a mutagen (3-6)
and a carcinogen in experimental animals (7-10). Little is
known concerning the carcinogenicity of EDB in humans; two
epidemiology studies on industrial workers exposed to relatively
high levels of EDB were inconclusive (11, 12).
EDB binds covalently to DNA in vitro and in vivo. The major
DNA adduct formed by EDB is S-[2-(N 7-guanyl)ethyl]GSH
(13-15). Formation of this adduct results from conjugation of
Received 9/7/89; revised 2/1/90.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1Supported in part by USPHS Grants CA 44353, ES 00267. ES 01590, and
ES 03272.
J Present address: Pfizer Central Research, Drug Metabolism Department,
Eastern Point Rd.. Croton. CT 06340.
3 Present address: Department of Biochemistry, School of Dentistry. Oregon
Health Sciences University. Portland, OR 97201.
4 To whom requests for reprints should be addressed.
5The abbreviations used are: GST. glutathione 5-transferase: GSH, glutathi
one; EDB. ethylene dibromide (1.2-dibromoethane); HPLC, high performance
liquid chromatography; CDNB, l-chloro-2.4-dinitrobenzene; KH, Krebs-Henseleit (buffer) [25 mM /V-(2-hydroxyethyl)piperazine-A"-(2-ethanesulfonic
acid)
buffer (pH 7.4) containing 118 mM Nad. 4 m.M KC1, 1 mM KH2PO4, 1.2 mM
MgSO«.3.4 mM CaCIj. and 24 mM NaHCOj).
EDB and GSH by GST to form a sulfur half-mustard. A second
bromide ion is then displaced to form an episulfonium ion
intermediate, the activated form of EDB which ultimately reacts
with DNA (16) (Fig. 1).
GSTs are dimeric enzymes composed of M, 23,000-26,000
subunits. The various subunits differ in their substrate specific
ities. Four different classes of GST have been identified: cyto
solic forms «,n, and ITand a microsomal form (17). Within
each cytosolic class different enzymes have been isolated, and
the subunits may associate to form homo- or heterodimers.
Binding of EDB to DNA through the action of GST has been
demonstrated in vivo in rats (18) as well as in rat hepatocytes
(13). Because of the potential carcinogenicity of EDB to hu
mans, it is of interest to determine if such binding occurs in a
human system and which GST enzymes are responsible for this
conjugation. In this paper, we assess the abilities of six purified
rat GST enzymes, five purified human GST enzymes, and rat
microsomal GST to conjugate GSH and EDB. The separation
of human a class enzymes ax-a„a,-ay, and ay-ay in active form
is described. We also report a comparison of binding of EDB
to DNA in rat and human hepatocytes and the induction of
unscheduled DNA synthesis by EDB in human hepatocytes.
MATERIALS
AND METHODS
Chemicals. GSH was purchased from Sigma Chemical Co. (St. Louis,
MO) and used without further purification (dissolved immediately prior
to use). [3H]Thymidine was purchased from Amersham-Searle (Arling
ton Heights, IL). [1,2-'4C]EDB was purchased from Amersham-Searle
or New England Nuclear Corp. (Boston, MA). The purity of EDB
(after dilution in (CH3)2SO or propylene glycol) was confirmed by
HPLC using an updated Flo-One Model HS Radioactive Flow Detector
(Radiomatic Instruments and Chemical Co., Inc., Tampa, FL) with
Flo-Scint II scintillation cocktail (Radiomatic Instruments and Chem
ical Co., Inc.) at a flow rate of 3 mi/min. The HPLC system consisted
of a 5-^m 4.6- x 250-mm Zorbax octadecylsilyl (C18)column (DuPont,
Wilmington, DE) with a mobile phase flow rate of 1 ml/min. A 35-min
gradient increasing from 14 to 57% CH3OH in H2O was followed by a
50-min gradient increasing from 57 to 95% CH3OH, and the retention
time of EDB was 35 min. The radiochemical purity of freshly prepared
solutions of [1,2-14C]EDB used in experiments was >97%. The purity
of [1,2-I4C]EDB solutions in (CH3)2SO stored at -20°C was observed
to decrease with time. EDB (unlabeled) was from Aldrich Chemical
Co. (Milwaukee, WI) (Gold Label grade) and was dissolved in
(CH3)2SO. All other reagents used were of highest commercial purity
available.
Tissue Sources. Male Sprague-Dawley rats (200-250 g) were pur
chased from HarÃ-anIndustries (Indianapolis, IN) and housed under
conventional procedures without further treatment before experiments
unless otherwise noted.
Human liver samples were obtained through the Nashville Regional
Organ Procurement Agency or from Dr. Philippe H. Beaune, INSERM,
Paris. Livers (obtained from accident victims) were perfused with cold
0.15 M NaCl within 20 min of clinical death and kept at 0°Cuntil cells
were prepared (4-10 hours after clinical death); small portions (<4
cm') were frozen by immersion in liquid nitrogen and stored at —70°C
for preparation of microsomes.
2747
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
KDB BIOACTIVATION
ir-
Enzymes. Rat liver GST (EC 2.5.1.18), a mixture of GST enzymes,
and collagenase (type 4) were purchased from Sigma. Liver cytosolic
and microsomal fractions were prepared as described elsewhere (19)
and stored at -70°C. [Microsomes were prepared from rats induced
with pregnenolone 16«-carbonitrile (20)]. Human liver microsomes
were prepared as described elsewhere (21) and stored at -70°C.Láclate
A
B
dehydrogenase assays (22) indicated that the microsomal preparations
contained <1% cytosolic contamination.
GST 1-1, 2-2, 3-3, and 4-4 enzymes were prepared from rat liver
essentially as described by Beale et al. (23). GST 7-7 was prepared from
rat kidney as described by Meyer et al. (24). [The nomenclature used is
that of Jakoby et al. (25).) GST 6-9 [previously known as "6-6" (26)] is
the major acidic isoenzyme present in rat testis (27).
Human GSTs ir, p, and a mixture of a class enzymes were prepared
according to the method of Ostlund Farrants et al. (26). Individual
enzymes of the a class, namely ax-a„afay, and a,.-ay [which correspond
to B,B,, B,B2, and B2B2,respectively, as described by Stockman et al.
(28)], were prepared by fast protein liquid chromatography chromatofocusing as follows. The a class enzymes from two livers were each
transferred by gel filtration into 30 niM diethanolamine-HCl buffer (pH
9.5) and applied to a Mono P column (Pharmacia, Uppsala, Sweden)
equilibrated with this buffer. The bound enzymes were eluted with
Polybuffer 96-HC1 (pH 6.0), and the catalytically active fractions were
transferred into 20 m.Msodium phosphate buffer (pH 6.7) and analyzed
by reverse phase HPLC (29). One liver (FH 83) yielded a major peak
of «,-a,and a minor peak of «,-«,.
(Fig. 2, A and B); the other (FH 61)
yielded major peaks of ax-af and a,.-«,.(Fig. 2, C and D). Similar
analyses of other human liver samples have consistently given major
peaks corresponding to the three enzymes described above, as well as
minor forms which have not yet been fully characterized.
Assays of GST Activity. GSH-CDNB conjugation was monitored
spectrophotometrically (Ano) at 25°Cas described by Habig et al. (30).
The GSH concentration was 1 mM, except in microsomal incubations,
where it was 3 mM. The extinction coefficient used was 9.6 HIM' cm '.
In the metabolism of EDB to various products by GST, including
DNA adducts (18), only the first step (formation of the GSH halfmustard) is GST catalyzed. The parameters selected for measurement
of this process included total non-CH2C!2-extracted products and 5,5'ethylenebis(GSH), the product of reaction of the episulfonium ion with
(a second molecule of) GSH (18). GSH has been shown to react nearly
completely with an excess of 5-(2-chloroethyl)GSH (31), an interme
diate analogous to the 5-(2-bromoethyl)GSH formed enzymatically in
this study. All in vitro reactions contained 2 mM [1,2-I4C]EDB (specific
activity, 1-16 mCi/mmol) and 5 mM GSH in 100 mM Tris-HCl buffer
(pH 7.7) in a total volume of 0.2 ml (under an argon atmosphere) and
were incubated in a gyrorotary shaking water bath at 37°C.Samples
were diluted to 0.55 ml and extracted four times with 5 ml of CH2C12
to remove unreacted EDB.
Total conjugation of EDB with GSH in in vitro incubations was
estimated by determining the amount of radiolabel remaining in the
aqueous phase after removal of unreacted [l^-'^CJEDB with CH2C12
(corrected for radioactivity in control samples incubated without GST).
Reaction products were separated by HPLC on a S-^m 4.6- x 250-mm
Zorbax C,g column with a mobile phase consisting of 0.15% CF.,CO2H
(in H2O) containing either 3.6 or 5.4% CH3CN (v/v), at a flow rate of
1 ml/min, and were monitored by UV absorbance at 214 nm. 5,5'ethylenebis(GSH) was identified by coelution with authentic material
D
pH
IV
61-
10
15
Elution volume (ml)
20
20
30
40
Time (min)
50
Fig. 2. Separation of human GSTs of the n class. Mixtures of n class enzymes
prepared from two human livers were separated by fast protein liquid chromatog
raphy chromatofocusing (A, C) as described in the text and analyzed by Ama
(
), activity towards CDNB (not shown), and pH (
). Major peaks of
activity (I, II, III. and IV) were desalted and samples were analyzed by reverse
phase HPLC monitoring <42i4(A,D). Peaks I and IV were concluded to contain
the homodimeric enzymes «,-««,
and ay-ay, respectively, while peaks II and III
both contained the heterodimer a„-ar
(synthesized by reaction of EDB with excess GSH sodium thiolate in
CHjOH and characterized by 'H nuclear magnetic resonance and fast
atom bombardment mass spectrometry; data not shown) and was
quantitated using an updated Flo-One Model HS Radioactive Flow
Detector with Flo-Scint III scintillation cocktail at a flow rate of 3.5
ml/min. Representative chromatograms are shown in Fig. 3.
For assays with the purified rat and human GST enzymes described
in Table 1, samples were prepared in duplicate or triplicate and con
tained 0.04-0.10 mg of each purified rat or human GST enzyme per
ml, depending on the amount and concentration of each enzyme avail
able. Samples and duplicate control reactions containing no GST were
incubated for 15 min.
In the cytosolic and microsomal assays, duplicate reaction mixtures
containing 2 mg of cytosolic or microsomal protein per ml and 0.2%
Lubrol PX (w/v) and control reactions containing no GST were incu
bated for 15 min. Protein concentrations were determined using the
Pierce bicinchoninic acid procedure (Pierce Chemical Co., Rockford,
IL).
Preparation of Rat and Human Hepatocytes. Hepatocytes were pre
pared from rat or human liver by an adaptation of a collagenase/neutral
protease digestion procedure (32). Approximately 50 g of liver (pooled
from 4-6 animals in the case of rats) were cut into cubes of approxi
mately 1 cm' and washed five times with cold KH buffer. The tissue
was minced with scissors and washed five more times with KH buffer.
The minced tissue was divided into three equal portions, placed into
2748
Fig. 1. Bioaclivation of EDB by GST and the formation of its major DNA
adduci.
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
F.DB BIOACTIVATION
in 17 mM Tris-HCl buffer (pH 7.2) containing 140 mivi NH..C1 and 15
mM NaCl in order to lyse RBC. After a 2-min incubation, the cells
were centrifuged and resuspended in Williams' or Fischer's medium
1600-
containing 5% (v/v) calf serum and 5 mM glucose. Cell viability was
estimated on the basis of trypan blue exclusion. In general. ~8 x IO7
cells were recovered from each digestion, with viabilities of 80-95%
observed for human hepatocytes and 75-90% for rat hepatocytes.
Incubations of Rat or Human Hepatocytes with |1,2-"C]EDB. Hepa
tocytes were suspended in Fischer's medium containing 5% (v/v) calf
serum and 5 m,Mglucose at a density of approximately 2 x IO6 cells/
ml and incubated at 37°Cwith 95% CO2/5% O2 in an orbital shaking
water bath. A portion of the cells was removed and frozen for subse
quent GSH and GST assays. Cells were incubated with 0.5 mM [1,214C]EDB (5 mCi/mmol) for 2 h. Cells were isolated by centrifugation
Ë
a
C
800
and washed twice with KH buffer to remove extracellular radioactivity
in the medium. The cells were suspended in 20 ml of 1 mM sodium
phosphate buffer (pH 6.8) containing 2 M NaCl and 5 M urea and
homogenized with a motorized glass-Teflon homogenizer. DNA from
the homogenate was isolated using the procedure described elsewhere
(18). For estimation of CDNB activity, 1 ml of hepatocyte suspension
was first homogenized in 2 ml of H2O and centrifuged at IO4 x g for
3
10
20
O
10
20
Time (min)
Fig. 3. Formation of S,S"-ethylenebis(GSH)
by GST. HPLC radioactivity
profiles of (A) a control incubation with no enzyme and (B) an incubation with
rat GST 2-2. Arrow, tK of bis conjugate (from A,,4 measurements). The material
eluting at ÕR
approximately 5 min is primarily GSCH2CH2OH and other products
(13).
Table 1 Conjugation ofEDB with GSH by purified GST enzymes
Purified GST enzymes (between 0.017 and 0.10 mg/ml) were incubated with
[1,2-WC]EDB (2 mm) and GSH (5 HIM)for 15 min at 37°C.The total amounts
of EDB-GSH conjugates formed were calculated from the amount of radiolabel
remaining after extraction of the reaction mixture with CH2CI2. S.S'-Ethylenebis(GSH) was separated by HPLC and quantified using a radioactive flow detec
tor. Conjugation of CDNB was measured spectrophotometrically.
90 min, and the supernatant was filtered to remove lipid.
GSH Assays. GSH was estimated using 5,5'-dithiobis(2-nitrobenzoic
acid). To a 1-ml aliquot of cellular suspension, 0.5 ml of a mixture of
4 M HC1O4 and 8 mM EDTA was added, and, after centrifugation, 1
ml of a 2 M KOH/300 mM 7V-(2-hydroxyethyl)piperazine-A"-(2-ethanesulfonic acid) mixture was added to the supernatant to remove KC1O4.
Estimates of GSH (nonprotein sulfhydryls) were then made spectro
photometrically (34). With some samples, aliquots were also assayed
for GSH using o-phthalaldehyde (35). Both methods resulted in ap
proximately equivalent estimates of GSH.
Unscheduled DNA Synthesis in Human Hepatocytes. Human hepa
tocytes were prepared as described above. Following centrifugation and
resuspension in fresh Williams' Medium E, hepatocytes were counted
with a hemocytometer. Cells used for the unscheduled DNA repair
experiment exhibited a viability of 85%.
Monolayer hepatocyte cultures were prepared by placing 3 x IO6
cells in 50-mm collagen-coated culture dishes containing 3 ml Williams'
Medium E with 10% fetal calf serum, 1 ^M insulin, and 0.1 ^M
corticosterone. Cultures were maintained at 37°Cin a humidified
EDB conjugation activity
incubator under an atmosphere of 5% CO2. After a 2-h incubation,
(nmol product formed/mg/min)
unattached cells were removed and fresh medium containing test con
EnzymeRat
conjugates11"117'70"21"103«43«141108"63GSCH2CH2SG46836104111816231(nmol/mg/min)3338381311-17'4072'419255
centrations of EDB was added. EDB was added from serial dilutions
1-1Rat
of a 30 mM stock solution prepared in (CH3)2SO. Diethylmaleate was
2-2Rat
added from a 450 mM stock solution (dissolved in (CH3)2SO). Diethyl
3-3Rat
maleate treatment was done during EDB treatment or after removal of
4-4Rat
6-9Rat
EDB. Carrier control incubations contained 0.36% (CH3)2SO during
7-7Human
the EDB treatment period to control for solvent effects on DNA
synthesis. Following a 2-h incubation period with EDB, fresh medium
uHuman
containing 0.1 mM ['Hjthymidine (450 mCi/mmol) was added. Incor
ITHuman
a«,«*«„
a.a,—
-a,Total
" Mean of duplicate incubations.
* Mean of triplicate incubations.
c Values from earlier preparations.
poration of thymidine into hepatocyte DNA was measured after 10 h.
Plates were washed twice with 5 ml of warm 10 mM potassium phos
phate buffer (pH 7.4) containing 0.9% NaCl (w/v), and monolayers
were scraped into 2 ml of 10% CC13CO2H (w/v). DNA was trapped on
glass fiber filters using a vacuum manifold and washed with an addi
tional 15 ml of 10% CC13CO2H. Filters were dried in place and
radioactivity was determined by scintillation counting in 15 ml ACS
cocktail (Amersham-Searle, Arlington Heights, IL).
150-mI Erlenmeyer flasks, and incubated with gentle orbital agitation
in 75 ml of KH buffer containing 5 x IO4 units collagenase, 20 mg
DNase I. 20 mg soybean trypsin inhibitor, 2 g dispase, 10 m\i CaCl2,
and 5 mM glucose at 37°C.Incubation was done under an atmosphere
of 95% O2/5% COjwith shaking for 30 min. At 25-30-min intervals,
cells that were released from the tissue were harvested by centrifugation
at 300 x g (the first harvest of cells, which contained damaged tissue
from the mincing process, was always discarded). The cell pellets were
suspended in 5 ml of calf serum (obtained from Grand Island Biological
Co., Grand Island, NY) and added to 25 ml of Williams' Medium E
(33) or Fischer's medium for leukemic mice (obtained from Grand
RESULTS
Selectivity of Rat and Human GSTs. The EDB conjugation
activity of GSTs was assessed by incubation of the enzymes
with [I,2-UC]EDB and GSH at 37°C.The total rate of conju
gation was estimated by the amount of radiolabel remaining in
the aqueous phase after removal of unreacted EDB with
CH2C12. The rate of formation of the specific conjugate 5,5'-
ethylenebis(GSH) [which results from nonenzymatic reaction
Island Biological Co.). The supernatant, containing active enzymes,
of the GSH half-mustard with a second molecule of GSH (18,
was added to the minced liver tissue to continue releasing liver cells.
After 3 or 4 harvests, the pooled cells were centrifuged and resuspended
31)] was also determined after separation of this conjugate by
2749
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
EDB BIOACTIVATION
HPLC, monitored using a radioactive flow detector. Repre
sentative chromatograms for a control sample and an active
GST enzyme are shown in Fig. 3. The amount of 5,5"-ethylenebis(GSH) formed increased linearly with both increasing
incubation time and increasing GST concentration (results not
shown).
Six rat and five human GST enzymes were tested for their
abilities to catalyze the conjugation of EDB with GSH. The
rates of total conjugate formation and production of S,S'ethylenebis(GSH) for each GST tested are presented in Table
1. The difference in observed cpm between replicate samples
was less than 10% in all cases. The ratio of 5,5"-ethylenebis(GSH) to total conjugates formed averaged 0.46 ±0.10 for
the 11 enzymes tested. Of the rat enzymes, the a class enzyme
2-2 and the M class enzyme 3-3 were clearly most active.
Danielson and Mannervik (36) have shown that the subunits of
GST dimers function independently; therefore, it may be in
ferred that both rat subunit 6 and rat subunit 9 have low
catalytic activity for EDB. Among the human enzymes, those
of the a class displayed the most activity. From data on the
separated forms of the human «class enzymes it may be inferred
that the a, subunit is more active than the «,.subunit.
The possibility that differences in EDB-conjugating ability
were the result of inactivation of the enzymes during prepara
tion procedures was tested by measuring the CDNB-conjugating activity of all enzyme preparations (except rat 6-9 and 7-7
and human TT).A comparison of the activity profiles for EDB
and CDNB for the various enzyme preparations indicates that
EDB-conjugating ability does not follow the pattern of CDNB
activity (Table 1). Therefore, the observed selectivity of the
enzymes for EDB does not appear to be an artifact related to
individual enzyme preparations.
Activity of Microsomal GST. To determine if microsomal
GST catalyzes the conjugation of EDB with GSH, rat liver
microsomes and cytosol (a positive control) were incubated
with [1,2-UC]EDB and GSH, and the amounts of total conju
gates and 5,5"-ethylenebis(GSH) were determined, as described
for the purified cytosolic enzymes. The rates of total conjuga
tion and production of 5,5'-ethylenebis(GSH)
for rat cytosol
and microsomes are presented in Table 2, along with GSHCDNB conjugation activities. The measured CDNB activities
are similar to those reported in the literature (37). The micro
somal fraction catalyzed very low levels of EDB conjugation.
The content of GST in total microsomal proteins is reported
to be 3% (38); assuming this value, the activity of the rat
microsomes per mg GST is calculated to be 1.5 nmol product
formed/mg GST protein/min, a level comparable with the least
active cytosolic GST enzymes (in Table 2, activity is reported
as nmol product formed/mg total protein/min).
In an experiment comparing conjugation catalyzed by rat
enzyme 2-2 with and without the addition of rat microsomes (2
mg total protein/ml), approximately 26% of the EDB-GSH
Table 2 Conjugation of EDB with GSH by hepatic cytosol and microsomes
Rat cytosol or microsomes (2 mg/ml) were incubated with [1.2-I4C|EDB (2
mM) and GSH (5 mm) for 15 min at 37'C. The total amounts of EDB-GSH
conjugates formed were calculated from the amount of radiolabel remaining after
extraction of the reaction mixture with CH;CI2. S.S'-Elhylene-bis(GSH) was
separated by HPLC and quantified using a radioactive flow detector. Conjugation
of CDNB was measured spectrophotometrically. Each value is the mean of
duplicate incubations.
Activity (nmol product formed/mg/min)
Enzyme
CDNB
Total EDB conjugates
GSCH2CH2SG
Rat cytosol
Rat microsomes
1730
75
0.044
5.2
<0.01
Table 3 EDB binding to DNA, GSH levels, and GST activities in isolated rat and
human hepatocyles
Each value represents the mean ±SD (the number of samples for each
determination is shown in parentheses). Each sample of human hepatocytes was
from one liver, and each rat hepatocyte sample was from a pool of five rat livers.
DNA binding
GSH
(nmol EDB/mg DNA) (nmol/10'cells)
Rat
Human
0.45 ±0.08 (3)
0.16 ±0.05 (5)
16 ±3(3)
21 ±13(4)
GST activity
(nmol CDNB
conjugated/min/10'cells)
300 ±70 (3)
130 ±130(4)
conjugates formed were bound to protein.6 The true rates of
conjugation are therefore slightly higher than those reported in
Table 2. In an analysis of a single human liver microsomal
fraction (results not presented) the catalytic activity for EDB
conjugation was also much lower than the human liver GSTs
examined (Table 1). The average ratio of 5,S'-ethylenebis(GSH) formation to total conjugate formation for experi
ments with cytosol and microsomes was 0.59 ±0.06, which is
consistent with that reported above for the purified enzymes.
Binding of EDB to DNA in Isolated Hepatocytes. Binding of
EDB to DNA in isolated rat hepatocytes has been demonstrated
previously (13, 18). In order to compare DNA binding of EDB
by rat hepatocytes with DNA binding of EDB by human hepa
tocytes, rat hepatocytes were prepared from liver by a mincing/
enzymatic digestion procedure identical to that used for human
hepatocytes. GSH content and GST activities of the hepato
cytes were measured and are reported in Table 3. During a 2-h
incubation with 0.5 mivi [1,2-MC]EDB, both rat and human
hepatocytes formed EDB-DNA adducts (Table 3). In separate
experiments in rat hepatocytes (data not presented), >80% of
the GSH was depleted at the end of similar 2-h incubations
with unlabeled EDB (GST activity was unaffected).
Unscheduled DNA Synthesis. Human hepatocytes were found
to readily adhere to culture dishes, and by 2 h the cells had
begun to assume the polygonal shape typical of hepatocytes in
culture. No visually detectable differences were observed be
tween the culture morphology of EDB-treated and untreated
cells. Cultures treated only with (CH.,)2SO exhibited a very low
level of [3H]thymidine incorporation through the 10-h synthesis
period, as shown in Fig. 4. All cultures receiving EDB showed
enhanced unscheduled DNA synthesis, which increased with
the EDB concentration. Those plates treated with 10 and 100
^M EDB showed approximately 6-fold higher levels of incor
poration than plates treated only with (CH3)2SO. Those cultures
which were treated simultaneously with EDB and diethylmaleate (i.e., GSH levels were depleted during EDB treatment)
showed markedly less thymidine incorporation. Thymidine in
corporation in cultures in which diethylmaleate treatment was
done subsequent to EDB treatment was only about 15% less
than in cases where no diethylmaleate was added.
DISCUSSION
The bioactivation pathway of EDB involves the action of
GST to conjugate GSH with EDB (Fig. 1) (2). Our studies with
purified rat GST enzymes indicate the a class enzyme 2-2, and
to a lesser extent the Mclass enzyme 3-3, catalyze this conju
gation (Table 1). These results are consistent with previous
studies which indicated that rat subunit Yc (equivalent to subunit 2) resulted in the highest level of binding of EDB to DNA
among the limited number of forms of GST tested ( 14). Studies
6 To determine the amount of binding to protein in assays with microsomes.
reaction mixtures containing rat subunit 2-2 were incubated with and without the
addition of 2 mg/ml microsomal protein. The rate of formation of total nonextractable products was 61 nmol product/min/mg GST without microsomes and
45 nmol product/min/mg GST with microsomes.
2750
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
EDB BIOACTIVATION
11200
e
o
i
A
J
800
Õ
—
[E.
:g 400
I
E
«-
O
0.01
0.1
1
10
[EDB] (pM)
100
1000
Fig. 4. EDB-mediated unscheduled DNA synthesis in human hepatocytes.
Human hepatocytes were exposed for 2 h to EDB with or without thiol depletion
by diethylmaleate, and [3H]thymidine incorporation was measured after a K) h
synthesis period. Each value is the mean of duplicate assays. Incubations included
(CH3)2SO only (•),EDB (A), 150 UMdiethylmaleate present only during the 10h synthesis period following EDB treatment (A), and concurrent EDB and 150
;/M EDB treatment (•).
with other halogenated compounds, including a-bromoisovaleric acid, a-bromoisovalerylurea, 2-iodooctane, 2-bromooctane,
and phenylethyl chloride, have also shown that conjugation
with GSH is dependent on the action of a or n class GST
enzymes (most often rat subunits 2 and/or 3) (39-41).
The results of experiments with purified human GST en
zymes are also consistent with the results using rat enzymes.
The a class of the human GSTs has significantly greater activity
than the ¿t
or ir classes (Table 1). We were able to separate three
active dimers of the a class, ax-otx, ax-ctr, and ay-ay, using the
technique of fast protein liquid chromatography chromatofocusing. Although both a* and a,, subunits conjugate EDB with
GSH, the specific activity of the as subunit for the reaction is
higher than that of a,..
The microsomal form of rat GST has no significant conju
gation activity with EDB as a substrate, as demonstrated in
incubations with liver microsomes (Table 2). These results
confirm those of earlier studies in which the addition of microsomes to cytosol did not significantly increase the binding of
EDB to DNA (14), and binding of ethylene dichloride to DNA
was very low in the presence of microsomes and GSH (42).
Previously we demonstrated that the stability of the EDBGSH conjugate is sufficient for it to exit rat hepatocytes and
alkylate extracellular DNA (18). Presumably, binding to intracellular DNA could be attributed to the ability of conjugates
formed in the cytosol to reach the nucleus. However, the GSTs
we report to be active in EDB metabolism have been found in
rat liver nuclei (43). The GSH concentration in the nuclei
appears to be similar to that of cytosol in rat kidney cells (44).
It is thus possible that EDB could be directly activated within
cell nuclei, increasing the likelihood of its reaction with DNA
to form adducts.
The levels and forms of several enzymes involved in the
metabolism of xenobiotics vary from individual to individual,
and it has been postulated that such variations may place certain
individuals at greater risk from exposure to various procarcinogens (45). There are reports of interindividual variations in
levels of both GSTs (46, 47) and cytochrome P-450s (48). The
human livers used in the present study for purification of human
class a GSTs differed greatly in the levels of ax and a,, subunits.
The levels of GST a class enzymes could directly affect the rate
of conjugation of EDB with GSH and ultimately affect the level
of DNA adducts formed. The most abundant metabolites of
EDB are formed as a result of microsomal oxidation (49);
therefore individuals with decreased levels of P-450s might
utilize the EDB-GSH conjugation pathway to a greater extent,
placing them at greater risk. It has been demonstrated that
concurrent administration of disulfiram (a P-450 inhibitor) and
EDB to rats results in an increase in the number of tumors and
tumor sites over those resulting from EDB alone (9).
Similarly, differences in tissue distribution of GSTs may
contribute to the sites of tumors associated with exposure to
EDB. Tumor sites in laboratory animals vary depending on the
mode of administration but include the skin, lungs, stomach,
nasal cavity, liver, and mammary glands (7, 9, 10). The expres
sion of GST enzymes is known to vary according to tissue type
in both rats and humans (50).
The composition of GST isozymes in cultured hepatocytes
changes with time, but the composition in freshly prepared
hepatocytes is almost the same as in whole liver (51). Since the
hepatocytes in our DNA binding studies were used shortly after
preparation, the composition of GST isozymes should reflect
that of the liver in vivo. The level of binding of EDB to DNA
found in studies with rat hepatocytes was 0.45 nmol adduci/
mg DNA, only one-tenth the level found in earlier studies (13,
18). This discrepancy is most likely due to the different tech
niques used to prepare the hepatocytes. A collagenase perfusion
method was used in the earlier studies, but due to the impracticality of perfusing human liver, a mincing/enzymatic digestion
procedure was used in this study for both human and rat
hepatocytes to allow their direct comparison. The level of
binding of EDB to DNA in the human hepatocytes is only 40%
of that in the rat hepatocytes (Table 2). This difference is not
due to GSH concentration but could be the result of different
levels of the relevant GST enzymes in the two preparations.
Significant binding does occur in both rat and human cells,
suggesting the possibility of similar bioactivation pathways in
rats and humans.
Human hepatocyte studies strongly implicate EDB-GSH
conjugation as the cause of DNA damage responsible for the
induction of unscheduled DNA synthesis. This conclusion can
be drawn because depletion of GSH levels during EDB exposure
causes a decrease in unscheduled DNA synthesis, but depletion
of GSH after EDB exposure causes little change in unscheduled
DNA synthesis. The induction of unscheduled DNA synthesis
suggests that EDB-DNA adducts undergo some enzymatic re
pair. EDB also causes sister chromatid exchange in human
lymphocytes (52), another indication that DNA damage occurs
as a result of EDB exposure in a human system. Unscheduled
DNA synthesis has been observed in rat hepatocytes and spermatocytes treated with EDB (53). The agreement of rat and
human unscheduled DNA synthesis studies is a further indica
tion that the genotoxic pathway may be similar in rats and
humans.
The validity of extrapolating results of animal studies to
humans is a very important concern which is not easily ad
dressed. The correlation of results in rat and human systems in
this study does seem to imply that in the case of EDB, the
genotoxic pathway in both organisms is similar. Studies are
currently under way to determine if the major DNA adduct
formed as a result of EDB-GSH conjugation is actually the
mutagenic lesion resulting in cellular transformation.
REFERENCES
2751
1. Anders, M. W.. Lash, L., Dekant, W., Elfarra, A. A., and Dohn, D. R.
Biosynthesis and biotransformation of glutathione S-conjugates. CRC Crit.
Rev. Toxicol., 18: 311-341, 1988.
2. Guengerich, F. P., Peterson, L. A., Cmarik, J. L., Koga, N., and Inskeep. P.
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
EIJB BIOACTIVATION
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
B. Activation of dihaloalkancs by glutathione conjugation and formation of
DNA adducts. Environ. Health Perspect., 76: 15-18, 1987.
Rannug, U. Genotoxic effects of 1,2-dibromoethane and 1,2-dichloroethane.
Mutât.Res. 76: 269-295. 1980.
Tan, E-L., and Hsie, A. W. Mutagenicity and cytotoxicity of haloethanes as
studied in the CHO/HGPRT system. Mutât.Res.. 90: 18.1-191. 1981.
Buijs, W., van der Gen. A.. Mohn. G. R., and Breimer. D. D. The direct
mutagenic activity of <>.u>-dihalogenoalkanes in Salmonella typhimurium:
strong correlation between chemical properties and mutagenic activity. Mu
tât.Res.. 141: 11-14, 1984.
Crespi. C. L., Seixas. G. M., Turner. T. R.. Ryan. C. G., and Penman. B. W.
Mutagenicity of 1,2-dichlorocthane and 1,2-dibromoethane in two human
lymphoblastoid cell lines. Mutât.Res.. 142: 133-140. 1985.
Weisburger. E. K. Carcinogenicity studies on halogenated hydrocarbons.
Environ. Health Perspect., 21: 7-16, 1977.
National Toxicology Program Carcinogenesis Bioassay of 1,2-Dibromoethanc in F344 Rats and B6C3F, Mice (Inhalation Study). Washington. DC:
United States Department of Health and Human Services, Technical Report
No. 210, 1982.
Wong, L. C. K., Winston. J. M., Hong, C. B., and Plotnick. H. Carcinogen
icity and toxicity of 1,2-dibromoethane in the rat. Toxicol. Appi. Pharmacol.,
63:155-165, 1982.
Huff. J. E. 1.2-dibromoethane (ethylene dibromide). Environ. Health Per
spect., 47: 359-363, 1983.
Ramsey, J. C., Park, C. N., Ott, M. G., and Gehring. P. J. Carcinogenic risk
assessment: ethylene dibromide. Toxicol. Appi. Pharmacol.. 47: 411-414,
1979.
Ott, M. G.. Scharnweber, H. C., and Langer. R. R. Mortality experience of
161 employees exposed to ethylene dibromide in two production units. Br.
J. Ind. Med.. 37: 163-168. 1980.
Ozawa, N.. and Guengerich, F. P. Evidence for formation of an 5'-[2-(A'7guanyl)ethyl|glutathione adduci in glutathione-mediatcd binding of 1,2-di
bromoethane to DNA. Proc. Nati. Acad. Sci. USA, 80: 5266-5270. 1983.
Inskeep, P. B.. and Guengerich. F. P. Glutathione-mediated binding of
dibromoalkanes to DNA: specificity of rat glutathione 5-transferases and
dibromoalkane structure. Carcinogenesis (Lond.). 5: 805-808, 1984.
Koga. N.. Inskeep. P. B.. Harris. T. M.. and Guengerich, F. P. 5-|2-(A"Guanyl)ethyl]glutathione, the major DNA adduci formed from 1.2-dibrom
oethane. Biochemistry. 25: 2192-2198. 1986.
Peterson, L. A.. Harris, T. M., and Guengerich. F. P. Mechanism of forma
tion of .S-[2-(.V-guanyl)ethyl]glutathione DNA adducts and the role of an
episulfonium ion intermediate. J. Am. Chem. Soc.. 110: 3284-3291, 1988.
Mannervik, B.. Alin, P., Guthenberg, C., Jensson, H., Tahir, M. K.. Warholm, M., and Jornvall. H. Identification of three classes of cytosolic gluta
thione transferase common to several mammalian species: correlation be
tween structural data and enzymatic properties. Proc. Nati. Acad. Sci. USA,
82: 7202-7206. 1985.
Inskeep. P. B., Koga, N., Cmarik, J. L., and Guengerich. F. P. Covalent
binding of 1,2-dihaloalkanes to DNA and stability of the major DNA adduci.
S-[2-(Ã-V7-guanyl)ethyl|glutath¡one.Cancer Res.. 46: 2839-2844. 1986.
Guengerich, F. P. Studies on the activation of a model furan compoundtoxicity and covalent binding of 2-(.V-ethylcarbamoylhydroxymethyl)furan.
Biochem. Pharmacol., 26: 1909-1915, 1977.
Guengerich, F. P., Wang. P.. and Davidson, N. K. Estimation of isozymes
of microsomal cytochrome P-450 in rats, rabbits, and humans using immunochemical staining coupled with sodium dodecyl sulfate-polyacrylamide gel
electrophoresis. Biochemistry. 21: 1698-1706. 1982.
Wang, P. P., Beaune, P., Kaminsky, L. S.. Dannan, G. A.. Kadlubar. F. F.,
Larrey, D., and Guengerich, F. P. Purification and characterization of six
cytochrome P-450 isozymes from human liver microsomes. Biochemistry,
22:5375-5383, 1983.
Láclate dehydrogenase. In: Worthinglon Manual: Enzymes and Related
Biochemicals. pp. 199-202. Freehold, NJ: Worthinglon Biochemical Corp..
1988.
Beale, D., Meyer, D. J., Taylor, J. B., and Ketterer. B. Evidence that the Yb
MIlum n-,of hepatic glutathione transferascs represent two different but related
families of polypeptides. Eur. J. Biochem., 137: 125-129, 1983.
Meyer, D. J.. Beale. D.. Tan. K. H., Coles. B. and Ketterer. B. Glutathione
transferases in primary rat hepatomas: the isolation of a form with GSH
peroxidase activity. FEBS Lett.. 184: 139-143. 1985.
Jakoby. W. B., Ketterer, B., and Mannervik, B. Glutathione transferases:
nomenclature. Biochem. Pharmacol.. 33: 2539-2540. 1984.
Ostlund Farrants, A-K., Meyer, D. J.. Coles. B.. Southan, C., Aitken, A.,
Johnson. P. J.. and Kctterer. B. The separation of glutathione transferase
subunits by using reverse-phase high-pressure liquid chromatography.
Biochem. J., 245: 423-428, 1987.
Meyer, D. J., Christodoulides, L. G., Nyan. O., Schuster Bruce. R.. and
Ketterer. B. A comparison of the glutathione transferases (GSTs) of three
extrahepatic organs with different functions-lhe adrenal, the lactating mam
mary gland and the male reproductive system. In: J. Rydstrom. J. Montelius.
and M. Bengtsson (eds.), Extrahepatic Drug Metabolism and C'hemical
Carcinogenesis. pp. 189-190. Amsterdam: Elsevier/North-Holland Bioméd
ical Press, 1983.
Stockman. P. K., McLellan. L.. and Hayes, J. D. Characterization of the
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
basic glutathione S-transferase B, and B2 subunits from human liver.
Biochem. J., 244: 55-61. 1987.
Meyer. D. J., Lalor, E., Coles, B., Kispert. A.. Alin, P., Mannervik, B.. and
Ketterer. B. Single-step purification and h.p.l.c. analysis of glutathione trans
ferase 8-8 in rat tissues. Biochem. J., 260: 785-788. 1989.
Habig. W. H.. Pabst, M. J., and Jakoby, W. B. Glutathione S-transferases.
The first enzymatic step in mercapturic acid formation. J. Biol. Chem., 249:
7130-7139. 1974.
Jean, P. A., and Reed, D. J. In vitro dipeptide, nucleoside. and glutathione
alkylation by 5-(2-chloroethyl)glutathione and 5-(2-chloroethyl)-L-cysteine.
Chem. Res. Toxicol., 2:455-460. 1989.
Maekubo. H., Ozaki. S.. Mitmaker. B., and Kalant. N. Preparation of human
hepatocytes for primary culture. In Vitro (Rockville), /*: 483-491. 1982.
Laishes. B. A., and Williams. G. M. Conditions affecting primary cell cultures
of functional adult rat hepatocytes. II. Dexamethasone enhanced longevity
and maintenance of morphology. In Vitro (Rockville), 12: 821-832, 1976.
van Bladeren. P. J.. Breimer. D. D.. Rotteveel-Smijs, G. M. T., De Jong, R.
A.. Buijs. W.. van der Gen. A., and Mohn. G. R. The role of glutathione
conjugation in the mutagenicity of 1,2-dibromoethane. Biochem. Pharmacol..
29:2975-2982, 1980.
Akerboom, T. P. M., and Sies, H. Assay of glutalhione, glutathione disulfide,
and glutathione mixed disulfides in biological samples. Methods Enzymol.,
77:373-382.1981.
Danielson, U. H.. and Mannervik. B. Kinetic independence of the subunits
of cytosolic glutathione transferase from the rat. Bioehem. J., 231: 263-267.
1985.
Morgenstern, R., DePierre, J. W., and Ernster, L. Activation of microsomal
glutathione 5-transferase activity by sulfhydryl reagents. Biochem. Biophys.
Res. Commun.. 87: 657-663, 1979.
Morgenstern. R.. Guthenberg. C., and DePierre, J. W. Microsomal glutathi
one S-transferase: purification, initial characterization and demonstration
that it is not identical to the cytosolic glutathione 5-transferases A, B and C.
Eur. J. Biochem., 72«:243-248, 1982.
te Koppele, J. M., Coles. B.. Ketterer. B., and Mulder, G. J. Stereoselectivity
of rat liver glutathione transferase isoenzymes for a-bromoisovaleric acid
and n-bromoisovalerylurea enantiomers. Biochem. J.. 252: 137-142. 1988.
Ridgewell. R. E., and Abdel-Monem, M. M. Stereochemical aspects of the
glutathione S-transferase-catalyzed conjugations of alkyl halides. Drug Metab. Dispos.. 15: 82-90, 1987.
Mangold, J. B.. and Abdel-Monem, M. M. Stereochemical aspects of con
jugation reactions catalyzed by rat liver glutathione 5-transferase isozymes.
J. Med. Chem.. 26: 66-71. 1983.
Guengerich. F. P., Crawford, W. M., Jr.. Domoradzki. J. Y.. Macdonald, T.
L. and Watanabe. P. G. In vitro activation of 1,2-dichloroethane by micro
somal and cytosolic enzymes. Toxicol. Appi. Pharmacol.,55:303-317,
1980.
Tan, K. H., Meyer, D. J., Gillies. N., and Kelterer, B. Detoxification of DNA
hydroperoxide by glutathione transferases and the purification and charac
terization of glutathione transferases of the rat liver nucleus. Biochem. J..
254: 841-845. 1988.
Tirmenslein. M. A., and Reed, D. J. The glutathione status of rat kidney
nuclei following administration of buthionine sulfoximine. Biochem. Bio
phys. Res. Commun.. ISS: 956-961. 1988.
Guengerich. F. P. Inter-individual metabolic variation in humans: mecha
nisms, methods of assessment, and consequences. In: Banbury Report 31:
Carcinogen Risk Assessment: New Directions in the Qualitative and Quan
titative Aspects, pp. 71-79. Cold Spring Harbor, NY: Cold Spring Harbor
Press, 1988.
Hayes, J. D.. McLellan. L. I., Stockman, P. K.. Howie. A. F., Hussey, A. J.,
and Beckett, G. J. Human glutathione 5-transferases; a polymorphic group
of detoxification enzymes. In: T. J. Mantle, C. B. Picket!, and J. D. Hayes
(eds.), Glutathione 5-Transferases and Carcinogenesis, pp. 3-18. London:
Taylor and Francis Ltd.. 1987.
Seidegà rd, J., DePierre. J. W.. and Pero, R. W. Hereditary inter-individual
differences in the glutathione transferase activity towards frani-stilbene oxide
in resting human mononuclear leukocytes are due to a particular isozyme(s).
Carcinogenesis (Lond.). 6: 1211-1216, 1985.
Guengerich, F. P. Polymorphism of cytochrome P-450 in humans. Trends
Pharmacol. Sci., 10: 107-109, 1989.
van Bladeren, P. J.. Breimer, D. D., van Huijgevoort, J. A. T. C. M.,
Vermeulen, N. P. E.. and van der Gen. A. The metabolic formation of A7acetyl-S-2-hydroxyethyl-L-cysteine
from tetradeutero-1,2-dibromoethane.
Relative importance of oxidation and glutathione conjugation in vivo.
Biochem. Pharmacol., 30: 2499-2502, 1981.
Ketterer, B., Meyer, D. J., and Clark, A. G. Soluble glutathione transferase
isozymes. In: H. Sies and B. Ketterer (eds.), Glutathione Conjugation:
Mechanisms and Biological Significance, pp. 73-135. San Diego. CA: Aca
demic Press, 1988.
Vandenberghc. Y'., Glaise, D., Meyer, D. J., Guillouzo, A., and Ketterer, B.
Glutathione transferase isoenzymes in cultured rat hepatocytes. Biochem.
Pharmacol., 37: 2482-2485, 1988.
52. Tucker, J. D., \u, J., Stewart, J., and Ong, T. Detection of sister-chromatid
exchanges in human peripheral lymphocytes induced by ethylene dibromide
vapor. Mutât.Res.. 138: 93-98. 1984.
53. Working. P. K., Smith-Oliver, T., White, R. D., and Butlerworth, B. E.
Induction of DNA repair in rat spermatocytes and hepatocytes by 1,2dibromoethane: the role of glutathione conjugation. Carcinogenesis (Lond.).
7:467-472. 1986.
2752
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.
Selectivity of Rat and Human Glutathione S-Transferases in
Activation of Ethylene Dibromide by Glutathione Conjugation
and DNA Binding and Induction of Unscheduled DNA Synthesis
in Human Hepatocytes
Joan L. Cmarik, Philip B. Inskeep, Michael J. Meredith, et al.
Cancer Res 1990;50:2747-2752.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/50/9/2747
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1990 American Association for Cancer Research.