Covalent protein adducts of hydroquinone in tissues from rats: Identification and quantitation of sulfhydryl-bound forms

Article abstract of DOI:10.1021/tx000037x

The Michael-type addition of sulfhydryl groups to benzoquinone (BQ) or substituted benzoquinones is proposed as the primary mechanism by which these electrophilic intermediates react with either cellular glutathione or protein sulfhydryls. This reaction constitutes a reductive alkylation with a substituted hydroquinone (HQ) derivative resulting from the addition. In the case of HQ, oxidative conversion of the parent material to BQ followed by conjugation with glutathione leads to metabolic activation, producing intermediates which are nephrotoxic as well as having other proposed biological activities. Chemically, BQ may react with more than 1 equiv of glutathione (or other sulfhydryl reagents) to produce HQ derivatives substituted with up to four sulfhydryl groups. Similarly, multiply substituted protein - S adducts of HQ were anticipated to occur in vivo following administration of this material. In the current studies, sulfhydryl-bound HQ protein adducts were detected and quantitated in protein isolated from rats using a modification of the alkaline permethylation procedure of Slaughter and Hanzlik [(1993)Anal. Biochem. 208, 288-295]. In particular, total protein - S adducts to HQ in kidney or blood reached a level of 420 or 80 pmol/mg of protein, respectively, 6 h following a single gavage dose of 100 mg/kg HQ. Measured half-lives of protein - S adducts in kidney and blood were 23.9 and 36.0 h, respectively. The applicability of protein - S adducts as a tissue dosimeter for HQ is discussed.

Full text of DOI:10.1021/tx000037x

Chem. Res. Toxicol. 2000, 13, 853-860
853
Cova len t P r otein Ad d u cts of Hyd r oqu in on e in Tissu es
fr om Ra ts: Id en tifica tion a n d Qu a n tita tion of
Su lfh yd r yl-Bou n d F or m s†
Rodney J . Boatman,* J . Caroline English, Louise G. Perry, and Laurie A. Fiorica
Health and Environment Laboratories, Eastman Kodak Company, Rochester, New York 14652-6272
Received February 17, 2000
The Michael-type addition of sulfhydryl groups to benzoquinone (BQ) or substituted
benzoquinones is proposed as the primary mechanism by which these electrophilic intermediates
react with either cellular glutathione or protein sulfhydryls. This reaction constitutes a reductive
alkylation with a substituted hydroquinone (HQ) derivative resulting from the addition. In
the case of HQ, oxidative conversion of the parent material to BQ followed by conjugation
with glutathione leads to metabolic activation, producing intermediates which are nephrotoxic
as well as having other proposed biological activities. Chemically, BQ may react with more
than 1 equiv of glutathione (or other sulfhydryl reagents) to produce HQ derivatives substituted
with up to four sulfhydryl groups. Similarly, multiply substituted protein-S adducts of HQ
were anticipated to occur in vivo following administration of this material. In the current
studies, sulfhydryl-bound HQ protein adducts were detected and quantitated in protein isolated
from rats using a modification of the alkaline permethylation procedure of Slaughter and
Hanzlik [(1993) Anal. Biochem. 208, 288-295]. In particular, total protein-S adducts to HQ
in kidney or blood reached a level of 420 or 80 pmol/mg of protein, respectively, 6 h following
a single gavage dose of 100 mg/kg HQ. Measured half-lives of protein-S adducts in kidney
and blood were 23.9 and 36.0 h, respectively. The applicability of protein-S adducts as a tissue
dosimeter for HQ is discussed.
have recently reported that radiolabeled 2,3,5-(triGSyl)-
HQ was more strongly retained by the in situ perfused
rat kidney than was the monosubstituted homologue,
suggesting that the more highly substituted conjugate
or metabolites derived from it are more reactive and thus
potentially more cytotoxic.
In tr od u ction
A growing body of experimental evidence suggests that
the conversion of chemicals such as p-aminophenol,
hydroquinone (HQ),1 and 2-bromo-HQ to their respective
glutathione S conjugates is an obligatory first step in the
bioactivation of these materials to nephrotoxic metabo-
lites (1). In the case of hydroquinone, initial oxidation to
BQ and subsequent reaction with reduced glutathione
produce a monosubstituted conjugate which, under in
vivo conditions, may be further oxidized and conjugated
with additional glutathione. Sequential oxidation fol-
lowed by conjugation may yield hydroquinone substituted
with up to 4 equiv of glutathione (see Figure 1) (2). In
confirmation of this mechanism, Hill et al. (3) have
identified the mono-, di-, and tri(glutathion-S-yl)hydro-
quinone conjugates in bile from male Sprague-Dawley
(SD) rats treated ip with HQ (1.8 mmol/kg) and pre-
treated with AT-125 to inhibit γ-glutamyl transpeptidase
(γ-GT) activity. When administered iv to SD rats, all the
glutathione conjugates (with the exception of the tetra-
substituted conjugate which was inactive) induced renal
toxicity similar to that observed with the parent material
but at considerably lower dose levels (2). In this series,
2,3,5-(triglutathion-S-yl)HQ [2,3,5-(triGSyl)HQ] was the
most active nephrotoxicant. In addition, Hill et al. (4)
Though still a matter of debate (5), it has been
proposed that protein binding of reactive quinone or
quinone thioethers is necessary for the cytotoxicity of
these materials. By this mechanism, a target protein
within a tissue may react with a quinone or a quinone
metabolite to form a protein-S adduct completely analo-
gous to the corresponding reaction with reduced glu-
tathione. It has been shown recently by Hanzlik et al.
(6) that in the model reaction of ribonuclease A with BQ,
covalent binding to protein sulfhydryls occurs exclusively
with no significant binding to other nucleophilic protein
side chains. These same investigators further suggest
that under physiologically relevant conditions (i.e., low
concentrations of reactive metabolites and short reaction
times), sulfhydryl arylation by BQ would be the only type
of reaction observed.
Quinone protein-S adducts are sufficiently unstable
such that traditional isolation techniques involving hy-
drolysis followed by amino acid analysis are not suitable
(7). Several methods, however, do exist for the detection
and/or quantitation of sulfhydryl-bound HQ and substi-
tuted HQ derivatives. Thus, the method of McDonald et
al. (8), employing Raney nickel reduction of protease
hydrolysates of serum and bone marrow protein, gener-
ates HQ. This is subsequently derivatized and analyzed
by gas chromatography/mass spectrometry (GC/MS) or
† Presented in part at the Society of Toxicology meeting, Dallas, TX,
March 13-17, 1994.
* Address all correspondence to this author. Phone: (716) 588-5961.
Fax: (716) 722-7561.
Abbreviations: HQ, hydroquinone; BQ, benzoquinone; 2,3,5-(triG-
Syl)HQ, 2,3,5-(triglutathion-S-yl)hydroquinone; TCA, trichloroacetic
1
acid.
10.1021/tx000037x CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/11/2000

854 Chem. Res. Toxicol., Vol. 13, No. 9, 2000
Boatman et al.
F igu r e 1. (A) Product of the reaction of benzoquinone with reduced glutathione. (B) Substitutions of one, two, three, or four glutathion-
S-yl residues on the HQ nucleus.
gas chromatography/electron capture detection (GC/
ECD). Although this method has sufficient sensitivity to
detect background levels of protein-bound HQ in bone
marrow and blood, the reduction reaction itself presum-
ably cleaves all HQ-thiol bonds, thus providing no
information concerning the positions or extent of sulfhy-
dryl substitution. More recently, Rombach and Hanzlik
(9) have described a polyclonal antibody specific for the
thiol-bound S-(2,5-dihydroxyphenyl) moiety. This “anti-
HQ” antibody exhibits a high specificity for protein-HQ
adducts to liver proteins in rats following incubation of
bromobenzene with rat liver microsomes. The specificity
of this antibody for more highly conjugated forms of HQ,
however, is not reported. As an alternative, we have
employed the alkaline “permethylation” procedure de-
scribed originally by Slaughter and Hanzlik (7) and more
completely in a subsequent publication by Slaughter et
al. (10). In this procedure, basic hydrolysis of tissue
proteins results in a chemical fragmentation of sulfhy-
dryl-substituted HQ intermediates which, in the presence
F igu r e 2. Methylthio-substituted products resulting from the
of excess methyl iodide, are rapidly converted to the
chemically stable methylthio derivatives as shown in
Figure 2. A GC/MS procedure was developed which
permethylation of either glutathione- or protein-derived sulf-
hydryl adducts of hydroquinone.
allowed for the simultaneous quantitation of all six of
the possible sulfhydryl-substituted HQ products. To
An im a ls. Specific details, including the source and age of
animals used in these studies, can be found in the following
calibrate this procedure, the glutathione conjugates of
HQ were synthesized as standards and subjected to the
permethylation conditions used to derivatize protein
samples.
paper (11).
Dose P r ep a r a tion , Ad m in istr a tion , a n d An a lysis. Stock
solutions of HQ were prepared in degassed, distilled water or
degassed, sterile saline at concentrations not exceeding 50 mg/
mL. Rats were given HQ by oral gavage in degassed, distilled
water or by intraperitoneal injection in degassed, sterile saline
at one or more dose levels as described in the following paper
(11). The concentration of HQ in treatment solutions was
determined by HPLC immediately prior to use.
Exp er im en ta l P r oced u r es
Ma ter ia ls. Hydroquinone (HQ) and benzoquinone (BQ) were
obtained from the Eastman Kodak Co. (Rochester, NY). The
chemical purities of these were determined to be >99% by
HPLC. Reduced glutathione (96%) was obtained from the
Aldrich Chemical Co. (Milwaukee, WI). Methyl iodide and
resorcinol were obtained from the Aldrich Chemical Co. and
were reagent grade. Pentane (Aldrich Chemical Co.) was HPLC-
grade. Heptane (glass-distilled) was obtained from EM Sciences,
Inc. (Cherry Hill, NJ ) and was chromatographic grade. All other
chemicals were reagent grade unless otherwise noted.
Syn t h esis of HQ Glu t a t h ion e Con ju ga t es. The glu-
tathione conjugates of hydroquinone were prepared as standards
by a modification of the procedure of Lau et al. (2) and purified
by preparative HPLC. In a typical procedure, 0.5022 g (4.646
mmol) of BQ was placed in a 50 mL Erlenmeyer flask and
dissolved in 10-11 mL of methanol. With stirring and under a
continuous nitrogen purge, a solution of 1.2517 g (4.073 mmol)
of reduced glutathione in 10-11 mL of distilled, deionized water

Quantitation of HQ Protein Adducts
Chem. Res. Toxicol., Vol. 13, No. 9, 2000 855
(25%), 168 (13%).
was added dropwise to the BQ solution. A thick precipitate
formed which required the addition of 30 mL of nitrogen-purged
distilled, deionized water to dissolve. Following a brief (<15 min)
additional period, the stirred reaction solution was poured into
a separatory funnel and washed four times with 50 mL portions
of ether. The resulting aqueous phase was then transferred to
a 250 mL round-bottomed flask and the majority of methanol
removed by rotary evaporation at reduced pressure. The addi-
tion of acetone (equal volume) produced a voluminous precipitate
from which the supernatant was discarded. The resulting solid
product was dried first with a gentle stream of nitrogen followed
by drying under reduced pressure (<0.01 Torr). This procedure
yielded a mixture of the crude conjugates (0.9428 g) which were
dissolved in degassed, distilled water (300 mg/mL) and purified
by preparative HPLC by repeatedly injecting 200 µL aliquots
onto a 10 mm × 50 cm Whatman Partisil M20 10/25 ODS-3
preparative column (Clifton, NJ ). The column was eluted with
a mobile phase consisting of methanol, water, and acetic acid
(9:90:1, v:v:v) at a flow rate of 3 mL/min. The column effluent
was monitored by UV detection at 280 nm, and fractions of the
column effluent were collected. The individual conjugates were
found to elute from the preparative column in the order reported
by Lau et al. (2). Fractions corresponding to individual compo-
nents were combined and excess methanol removed by brief
rotary evaporation. The remaining solutions were then frozen
in dry ice and acetone and lyophilized to produce off-white solids.
The various glutathione conjugates were identified by order of
elution from the chromatographic column and by comparison
with published UV spectrophotometric and NMR spectrometric
data as well as by conversion to the permethylated derivatives
(Figure 2).
(3) 2,5-(Diglu ta th ion -S-yl)h yd r oqu in on e: UV λmax 315 nm;
1H NMR δ 2.02 (acetate impurity), 2.08 (dd, 4H, Glu-â), 2.42 (t,
4H, Glu-γ), 3.18 and 3.36 (ddd, 4H, Cys-â), 3.73 (m, 2H, Glu-R),
3.78 (s, 4H, Gly-R), 4.43 (dd, 2H, Cys-R), 6.96 (s, 2H, aromatic);
GC/MS (of the permethylated derivative) EI (relative intensity)
m/z 232 [11%, (M + 2)+], 230 (100%, M+), 215 (95%), 200 (26%),
181 (43%).
(4) 2,6-(Diglu ta th ion -S-yl)h yd r oqu in on e: UV λmax 266,
1
318 nm; H NMR δ 2.05 (acetate impurity), 2.08 (dd, 4H, Glu-
â), 2.42 (t, 4H, Glu-γ), 3.18 and 3.36 (ddd, 4H, Cys-â), 3.73 (m,
2H, Glu-R), 3.76 (s, 4H, Gly-R), 4.43 (dd, 2H, Cys-R), 6.96 (s,
2H, aromatic); GC/MS (of the permethylated derivative) EI
(relative intensity) m/z 232 [11%, (M + 2)+], 230 (100%, M+),
215 (42%).
(5) 2,3,5-(Tr iglu ta th ion -S-yl)h yd r oqu in on e: UV λmax 337
nm; 1H NMR δ 2.02 (acetate impurity), 2.06 (m, 6H, Glu-â), 2.42
(m, 6H, Glu-γ), 3.18 and 3.38 (m, 6H, Cys-â), 3.7-3.8 (m, 9H,
Glu-R and Gly-R), 4.26-4.45 (m, 3H, Cys-R), 7.04 (s, 1H,
aromatic); GC/MS (of the permethylated derivative) EI (relative
intensity) m/z 278 [14%, (M + 2)+], 276 (100%, M+), 261 (34%),
246 (62%).
(6) 2,3,5,6-(Tetr a glu ta th ion -S-yl)h yd r oqu in on e: UV λmax
1
356 nm; H NMR δ 2.05 (acetate impurity), 2.12 (dd, 8H, Glu-
â), 2.51 (m, 8H, Glu-γ), 3.19 and 3.42 (ddd, 8H, Cys-â), 3.72-
3.86 (m, 12H, Glu-R and Gly-R), 4.28 (dd, 4H, Cys-R); GC/MS
(of the permethylated derivative) EI (relative intensity) m/z 324
[20%, (M + 2)+], 322 (100%, M+), 307 (8%), 292 (13%), 279 (27%),
264 (24%).
Tissu e Collection . At the termination of each study, rats
were anesthetized (CO2) and exsanguinated via the inferior vena
cava. Blood was collected in tubes containing EDTA. The livers,
kidneys, and spleens were removed and placed immediately into
plastic bags or glass bottles stored over ice. Tissue samples
which were not further processed on the day of collection were
stored frozen at -70 °C until they were analyzed.
P u r ity An a lysis of Glu ta th ion e Con ju ga tes. The pre-
paratively purified glutathione conjugates of HQ were analyzed
for purity by HPLC using a Hewlett-Packard 1090 HPLC system
equipped with a 4.6 mm × 25 cm Zorbax ODS reversed-phase
analytical column (Dupont Instruments, Wilmington, DE).
Samples were dissolved in degassed, distilled water (1-3 mg/
mL) and 20 µL aliquots analyzed immediately. The column was
eluted with 0.1 M ammonium acetate buffer (pH 3.9) containing
1% (by volume) acetonitrile at a flow rate of 1.0 mL/min.
Detection was carried out with a diode array (Hewlett-Packard),
allowing the collection of full UV spectra. The purities of all
glutathione conjugates were determined to be >95% by analyti-
cal HPLC.
P r ep a r a tion of P r otein Sa m p les. Dissected livers, kidneys,
and spleens were washed in phosphate-buffered saline [0.1 M
potassium phosphate, 1 mM EDTA, and 1.12% KCl (pH 7.4)],
blotted dry, weighed, and homogenized using approximately 5
mL of the same buffer per gram of tissue. Equal volumes of 40%
TCA were added to the resulting suspensions to precipitate
proteins. Protein precipitates were collected by centrifugation
and were washed sequentially with acetone (two times), 80%
methanol/H2O (four times), acetone (two times), and ethyl ether
(two times). The resulting protein samples were then dried by
either brief rotary evaporation or application of a gentle nitrogen
stream followed by application of high vacuum (>24 h) to
produce free-flowing, off-white solids. Samples of whole blood
were treated with equal volumes of 40% TCA, and the protein
precipitates washed sequentially with acetone (six times), 80%
methanol/H2O (four times), acetone (four times), and ethyl ether
(four times). Solid blood protein samples were dried initially
with a gentle stream of nitrogen followed by application of high
vacuum (>24 h) to produce free-flowing solids.
Alk a lin e P er m eth yla tion . Methylthio-substituted deriva-
tives of 1,4-dimethoxybenzene, resulting from the alkaline
permethylation of tissue, urine, or blood protein, were obtained
using a modification of a published procedure (7, 10). In the case
of tissue proteins, approximately 125 mg samples were placed
into PTFE-lined, screw-capped culture tubes (16 mm × 100 mm)
followed by 1.0 mL of methyl iodide. Resorcinol as an internal
standard was added at this point (50 µL of a 0.1 mg/g of solution
in distilled water containing 0.5 mg/g of ascorbic acid). Dry
nitrogen was gently bubbled through the mixture, and 1.0 mL
of nitrogen-purged, 4 N NaOH (purged a minimum of 10 min
with nitrogen immediately prior to use) was then added. Tubes
were flushed for an additional 30 s with nitrogen, capped tightly,
and heated in thermostatically controlled heated blocks at 80
°C for 4 h. Following this, the tubes were cooled in ice and
opened. A 0.75 mL aliquot of 14 N NaOH was cautiously added
to each tube followed by 4 mL of methanol and acetone (1:1,
Str u ctu r a l Ch a r a cter iza tion of th e Glu ta th ion e Con -
ju ga tes. The 1H NMR spectra of the purified glutathione
conjugates of HQ (dissolved in degassed D2O) were recorded on
a 300 MHz Varian model VXR300S spectrometer. Structures
were confirmed by comparison with published spectra (2). In
this report, assignment of these isomers is based on the
similarity of spectral parameters to those published by Lau et
al. (2). It should be noted that Eckert et al. (12) reverse the
assignment of the 2,5- and 2,6-bis-glutathione-substituted
conjugates. In this regard, the nomenclature used in the present
report for these gluathione conjugates is also that used by Lau
et al. (2). The permethylated derivatives correspond to the
structures shown in Figure 2.
(1) 2-(Glu ta th ion -S-yl)h yd r oqu in on e: UV λmax 303 nm; 1H
NMR δ 2.05 (acetate impurity), 2.07 (dd, 2H, Glu-â), 2.41 (t,
2H, Glu-γ), 3.19 and 3.33 (ddd, 2H, Cys-â), 3.72 (m, 1H, Glu-R),
3.80 (s, 2H, Gly-R), 4.43 (dd, 1H, Cys-R), 6.72-6.96 (m, 3H,
aromatic); GC/MS (of the permethylated derivative) EI (relative
intensity) m/z 186 [5%, (M + 2)+], 184 (100%, M+), 169 (64%),
154 (32%), 137 (14%), 123 (42%).
(2) 2,3-(Diglu ta th ion -S-yl)h yd r oqu in on e. Impure material
was obtained following semipreparative HPLC, indicating con-
tamination primarily with the tetraglutathionyl-substituted
material: UV λmax 322 nm; 1H NMR δ 2.12 (m, 4.6H, Glu-â),
2.50 (m, 4.5H, Glu-γ), 3.2-3.6 (m, 4.52H, Cys-â), 3.75 (m, 6.7H,
Gly-R), 4.32 (m, 2H, Cys-R), 6.97 (s, 1.8H, aromatic); GC/MS (of
the permethylated derivative) EI (relative intensity) m/z 232
[11%, (M + 2)+], 230 (100%, M+), 215 (23%), 200 (88%), 185

856 Chem. Res. Toxicol., Vol. 13, No. 9, 2000
Boatman et al.
film thickness), 5 psi head pressure (He); oven program, 100
°C initial (hold 3 min), 10 °C/min to 300 °C (hold 5 min); injector
at 300 °C, split flow of 12 mL/min; MSD used in a single-ion
mode (150 ms dwell times) with ions monitored at m/z 138.1
(3-8 min, resorcinol), 184.1 (8-12 min, mono), 230.0 [12-16
min; di(2,3), di(2,5), and di(2,6)], 276.0 (16-18 min, tri), and
322.1 (18-20 min, tetra).
Sta tistica l An a lyses. The results for individual and total
protein-S adduct levels were analyzed statistically using the
computer program SAS (version 6, SAS Institute, Cary, NC).
Resu lts a n d Discu ssion
BQ and other oxidized metabolites of HQ containing
the quinone moiety are capable of reacting with tissue
protein, producing protein-S-bound adducts. Thus, co-
valently bound HQ (as well as bromo-HQ) has been
identified in liver protein from rats treated with bro-
mobenzene (7), in hemoglobin and bone marrow protein
from rats treated with benzene (13), and has been
tentatively identified in hemoglobin and liver protein
from mice treated with acetaminophen (14). In a recent
publication, Hill et al. (4) demonstrated that protein
binding occurs within the in situ perfused rat kidney
following administration of the nephrotoxic glutathione
conjugates of HQ. In these studies, the extents of binding
of radiolabeled substrates and excretion of the brush
border membrane-bound enzyme, γ-GT, were greatest for
the triglutathionyl conjugate, 2,3,5-(triglutathion-S-yl)-
hydroquinone. The goal of the current studies was to
determine the extent and nature of the covalent binding
of HQ in rats following either oral or ip administration
and to determine the relevance of this tissue dosimeter
to the known acute and chronic toxicities of HQ.
Id en tifica tion a n d Qu a n tita tion of HQ Th iol Ad -
d u cts to Ra t Liver P r otein . As shown in Figure 4, all
six of the methylthiol-substituted adducts of HQ were
identified by GC/MS in liver and kidney proteins from a
male F344 rat treated ip with HQ at 100 mg/kg (11). The
specific structural assignments for all of the substituted
methylthiol derivatives were determined by retention
time and mass spectral comparisons to the corresponding
permethylated glutathionyl-substituted standards. Mass
spectra of permethylated components from tissue samples
were nearly identical with spectra derived from perm-
ethylated standards. In all cases, the total ion mass
spectra of these methylthiol derivatives indicated that
the molecular (M+) ions are the most abundant ions in
the spectra. Fragmentation with a successive loss of CH3
groups leads to the observed spectra containing signifi-
cant (M - 15)+, (M - 30)+, and, in some cases, (M - 45)+
ions. This distinctive fragmentation pattern, character-
ized by successive loss of methyl groups, was common to
all the methylthiol derivatives. As further confirmation
of these structural assignments, the contribution of the
34S isotope to the intensity of the (M + 2)+ ions in these
spectra was clearly seen to increase from approximately
5% in the mono derivative to 20% in the tetra derivative.
Calibration curves generated from permethylated stan-
dards allowed accurate quantitation of individual meth-
ylthio derivatives in the range of 1-50 pmol/sample
(based on a 125 mg protein sample).
F igu r e 3. Concentrator apparatus used for the permethylation
of protein samples, which was constructed by joining 7 mm (o.d.)
glass tubing to 12 mm (o.d.) glass tubing. It was designed to
contain e5 mL of liquid when filled.
v:v). After being cooled in ice for approximately 5 min, a 4 mL
aliquot of pentane was added to each tube, and the tubes were
vortically mixed and the phases allowed to separate. The organic
pentane extracts were separated. The mixtures were extracted
a total of three times with pentane and pooled pentane extracts
washed four times with small volumes of water and finally dried
over granular, anhydrous sodium sulfate. Initially, the pentane
extracts were concentrated by placing them into pear-shaped
flasks (50 mL) equipped with small (10-15 cm in length)
Vigreux columns and heating these in a 50 °C water bath to
evaporate the majority of solvent. When the volume had reduced
to approximately 0.5-1 mL, the residues were transferred into
specially designed concentrator units as shown in Figure 3. A
small boiling chip was added, and the end of each concentrator
was immersed in a 50 °C water bath. The solvent was allowed
to evaporate from each unit until all residual pentane had
completely disappeared, at which point the concentrators were
removed from the water bath. A 20 µL aliquot of n-heptane was
added to each concentrator, and the concentrators were returned
to the heated bath for an additional 1 h to ensure removal of
all residual pentane. The concentrators were then scored and
broken as indicated in Figure 3 and the residues (20-50 µL)
analyzed by GC/MS. Concentrated, stock solutions of the
glutathione conjugates of HQ (prepared in degassed, distilled
water containing 0.5% ascorbic acid to inhibit oxidation) were
added to tubes (by weight) and subjected to identical permethyl-
ation conditions.
GC/MS An a lysis Con d ition s. Quantitative determinations
of the various methylthio-substituted derivatives were made by
comparisons of integrated ion intensities from sample analyses
(single-ion mode) to the results obtained by permethylation of
synthetically prepared HQ glutathione conjugates used as
standards (resorcinol as an internal standard). Calibration
standards were prepared over the range necessary to encompass
levels observed in permethylated tissue protein (1-50 nmol/
sample). Separate calibration curves for individual permethyl-
ated components (Figure 2) were obtained in which relative area
responses (ratio of calibration standard to internal standard)
were plotted versus amount ratios. The calibration curves
obtained were linear over the ranges analyzed with correlation
coefficients (r2) generally exceeding 0.98. Insufficient amounts
of pure 2,3-(diglutathion-S-yl)hydroquinone were obtained to use
as a GC/MS standard. In this latter case, the calibrated response
for the permethylated 2,5-(diglutathion-S-yl)hydroquinone stan-
dard was used to quantitate amounts of the 2,3-isomer. The
following chromatographic conditions were employed for these
analyses: instrument, Hewlett-Packard HP 5890 gas chromato-
graph with a HP 5970 mass selective detector (MSD); column,
DB-5 (J & W Scientific, Folsom, CA), 30 m × 0.25 mm (0.1 µm
Sta bility Stu d y. The in vivo stability of the protein-
bound forms of hydroquinone was determined in male
F344 rats. Tissues chosen for examination in this study
included blood and kidney. Both total protein-S adduct
levels as well as individual adduct levels were measured

Quantitation of HQ Protein Adducts
Chem. Res. Toxicol., Vol. 13, No. 9, 2000 857
Ta ble 2. Levels of In d ivid u a l HQ P r otein -S Ad d u cts
ver su s Tim e of Collection for Blood P r otein fr om Gr ou p s
of F ive Ma le F 344 Ra ts Tr ea ted Or a lly w ith HQ
(100 m g/k g)a
adduct level (pmol/mg of protein)
individual
adduct
control
6 h
24 h
48 h
mono
2,3-di
2,5-di
2,6-di
tri
6.30 ( 1.99 38.0 ( 5.70 21.9 ( 4.08 16.0 ( 3.31
NDb
ND
6.55 ( 0.977 5.02 ( 0.874 3.74 ( 2.13
13.1 ( 1.66 8.14 ( 1.42 4.96 ( 2.80
1.69 ( 0.757 13.3 ( 3.79 6.67 ( 1.04 5.25 ( 0.666
ND
ND
8.91 ( 1.19 6.07 ( 0.847 5.14 ( 0.755
ND ND ND
tetra
a
b
Error expressed as means ( 1 SD. ND, none detected.
Ta ble 3. Levels of In d ivid u a l HQ P r otein -S Ad d u cts
ver su s Tim e of Collection for Kid n ey P r otein fr om
Gr ou p s of F ive Ma le F 344 Ra ts Tr ea ted Or a lly w ith HQ
(100 m g/k g)a
adduct level (pmol/mg of protein)
individual
adduct
control
6 h
24 h
48 h
mono
2,3-di
2,5-di
2,6-di
tri
5.24 ( 0.943 76.7 ( 7.03 42.6 ( 7.81 21.2 ( 6.85
3.03 ( 2.82 18.5 ( 3.41 12.6 ( 2.62 7.15 ( 1.98
NDb
28.3 ( 2.45 9.91 ( 1.12 1.77 ( 2.60
0.691 ( 0.661 28.6 ( 3.69 13.4 ( 1.58 5.31 ( 4.45
2.71 ( 0.577 85.0 ( 9.48 57.2 ( 7.09 30.2 ( 9.54
tetra
2.53 ( 2.58
183 ( 33.2 120 ( 25.8 58.7 ( 19.4
a
b
Error expressed as means ( 1 SD. ND, none detected.
sampling are shown for blood in Table 2 and for kidney
in Table 3. In either tissue protein, adduct levels (both
total and individual) were found to decrease steadily after
6 h, with calculated, mean half-lives of 23.9 h for kidney
and 36.0 h for blood protein-S adducts.
Decreases in HQ protein-S adduct levels within a
tissue are determined not only by the chemical stability
of the initially formed adducts but also by the inherent
protein turnover rate for the particular tissue. In this
regard, adduct stability in protein from whole blood
(primarily hemoglobin) versus that from kidney provides
a useful comparison for these two dissimilar tissues. In
particular, the turnover rate or lifespan of hemoglobin
(Hb) is unique in that it is equivalent to the lifespan of
the erythrocyte, which for the rat is approximately 60
days (15). Similarly, half-lives for chemically stable
hemoglobin adducts are generally long, with the lifetimes
of the adducts determined primarily by that of the
erythrocyte life span (15, 16). The relatively short (36 h)
half-life of HQ blood protein-S adducts in this study
suggests that under in vivo conditions these adducts are
not stable but undergo further oxidative transformations
and are, as a consequence, no longer detected by perm-
ethylation. Oxidation within the red blood cell by hemo-
globin itself provides a likely explanation for this de-
creased stability. In this regard, hemoglobin has been
shown to display both monooxygenase-like and peroxy-
genase-like activities toward carcinogenic primary aryl-
amines (17). On the basis of the criteria of Skipper and
Tannenbaum (16), the relatively short half-life of the
blood protein-S adducts of HQ makes them a less useful
(sensitive) tissue dosimeter for total HQ exposure.
The fate of HQ protein-S adducts in the kidney
provides a significant contrast to that in blood. Although
adducts to kidney protein have a half-life similar to that
measured in blood (24 vs 36 h), the rate of protein
turnover within this tissue is far greater than that of
hemoglobin, and this more rapid turnover provides an
explanation for the half-life of the HQ kidney protein-S
F igu r e 4. GC/MS chromatographic profiles for the methylthiol-
substituted HQ derivatives: 2-methylthio-1,4-dimethoxyben-
zene, retention time of 10.5 min; 2,3-bis(methylthio)-1,4-
dimethoxybenzene, retention time of 14.2 min; 2,5-bis(methyl-
thio)-1,4-dimethoxybenzene, retention time of 14.3 min; 2,6-bis-
(methylthio)-1,4-dimethoxybenzene, retention time of 14.8 min;
2,3,5-tris(methylthio)-1,4-dimethoxybenzene, retention time of
17.5 min; and 2,3,5,6-tetrakis(methylthio)-1,4-dimethoxyben-
zene, retention time of 18.9 min. (A) Representative chromato-
gram of permethylated glutathionyl-HQ standards. (B) Perm-
ethylated liver protein from a male F344 rat treated ip with
HQ at 100 mg/kg (11). (C) Permethylated kidney protein from
a male F344 rat treated ip with HQ at 100 mg/kg (11).
Ta ble 1. Levels of Tota l P r otein -S Ad d u cts in Blood a n d
Kid n eys fr om Gr ou p s of F ive Ma le F 344 Ra ts befor e
Tr ea tm en t (0 h ) a n d a fter a Sin gle Or a l Dose of HQ (100
m g/k g)a
total protein-S adduct level
(pmol/mg of protein)
collection time (h)
blood
kidney
0 (control)
7.06 ( 3.44b
79.9 ( 11.0
47.8 ( 7.88
35.0 ( 9.38
14.2 ( 4.19
420 ( 51.9
255 ( 41.3
124 ( 42.8
6
24
48
a
Tissues from groups of rats were collected before treatment
b
(0 h) and 6, 24, and 48 h after treatment. Results are expressed
as means ( 1 SD.
in blood and kidney proteins obtained from groups of rats
(n ) 5) sacrificed at 6, 24, or 48 h following a single
gavage administration of 100 mg/kg HQ. Total protein-S
adduct levels found in the tissues at the various collection
times are shown in Table 1. Individual adduct concentra-
tions (picomoles per milligram of protein) versus time of

858 Chem. Res. Toxicol., Vol. 13, No. 9, 2000
Boatman et al.
F igu r e 5. Proposed scheme for the formation of multiply substituted glutathione- and protein-sulfhydryl adducts of HQ.
adducts seen in this study. Thus, Goldspink and Kelly
(18) report relatively consistent half-life values for mixed
kidney proteins from male CD rats ranging from 2 days
(at 8 weeks of age) to 3 days (at 105 weeks of age). Also,
it can be anticipated that the rate of protein turnover
within the metabolically active proximal tubular epithe-
lial cells will be similar to or greater than that for mixed
kidney protein. In support of this, Tsao et al. (19) report
an in vitro half-life of 32 h (turnover rate of 2.18%/h) for
protein in freshly isolated proximal tubules from male
SD rats. Thus, results from the current study are
consistent with protein turnover being the primary
determinant of the measured half-life of HQ kidney
protein-S adducts in the rat.
The permethylation procedure used in the current
studies provides a rapid and sensitive method for the
quantitation of quinone-derived protein thiol adducts.
However, the degradative nature of the derivitization
precludes the possibility of direct identification of the
protein precursors. Thus, for methylthio-HQ derivatives
containing two or more methylthio groups, it is not
possible to know which or how many of the methylthio
groups derive from a protein sulfhydryl in the initial
adduct. Also, the extent of intra- and interprotein cross-

Quantitation of HQ Protein Adducts
Chem. Res. Toxicol., Vol. 13, No. 9, 2000 859
Ta ble 4. P r otein -S-Bou n d F or m s of HQ Exp r essed a s a
Con clu sion s
P er cen ta ge of th e Ad m in ister ed Dosea
Sulfhydryl-bound forms of HQ were found in tissue
proteins from rats following either gavage or ip admin-
istration. Permethylation allowed identification and quan-
titation of all six of the possible protein-S-bound forms
of HQ in which the HQ nucleus is substituted with one,
two, three, or four methylthio substituents. In addition,
this work provides the first conclusive evidence for the
existence of the more highly substituted protein-S-bound
forms of HQ (containing either three or four methylthio
substituents). The following paper (11) will explore
protein-S-bound adduct levels in tissues of rats following
either single gavage or ip administration of HQ or
following 6 weeks of repeated gavage administrations of
HQ (11).
protein-S-bound HQ expressed as a
percentage of the administered dose
male F344 rats
female F344 rats
study (dose)
liver
0.25
kidney
liver
kidney
currentb (25 mg/kg)
0.015
0.010
0.39
0.23
0.031
0.020
[
14C]HQ, oralc (25 mg/kg) 0.075
a
Values from the current study are compared with residual
b
tissue radioactivity in rats treated orally with [14C]HQ. See ref
11. c Male and female F344 rats, 7-9 weeks of age, were treated
orally with [14C]HQ (25 mg/kg). In the case of male rats, tissues
were collected 72 h after treatment. For females, tissues were
collected 48 h after treatment.
linking cannot be determined by the procedure described.
As a simplifying assumption, it is proposed that the
majority of multiply S-substituted methylthio derivatives
of HQ identified in the current studies are derived from
protein precursors in which the HQ moiety is linked
through a single protein thiol group. This assumption is
based on the fact that oxidized quinone precursors will
preferentially react with glutathione, which is the pre-
dominant thiol-containing material in nearly all mam-
malian cells, present at millimolar concentrations (20),
and consequently, reaction with protein sulfhydryls will
be far less favored. Additional support for this contention
comes from the work of Hanzlik et al. (6), who report little
or no intermolecular cross-linking of ribonuclease in vitro
after treatment with BQ at levels as high as 64 mol of
BQ/mol of protein. A simplified scheme for the formation
of multiply S-substituted protein sulfhydryl adducts of
HQ is shown in Figure 5.
Ack n ow led gm en t . We are grateful for the advice
and suggestions provided by Prof. Robert P. Hanzlik of
the University of Kansas (Lawrence, KS). Financial
support was provided by Eastman Chemical Co., Mitsui
Petrochemicals (America), Ltd., and Rhodia, Inc.
Refer en ces
(1) Dekant, W. (1993) Bioactivation of nephrotoxins and renal
carcinogens by glutathione S-conjugate formation. Toxicol. Lett.
67, 151-160.
(2) Lau, S. S., Hill, B. A., Highet, R. J ., and Monks, T. J . (1988)
Sequential oxidation and glutathione addition to 1,4-benzo-
quinone: Correlation of toxicity with increased glutathione
substitution. Mol. Pharmacol. 34, 829-836.
(3) Hill, B. A., Kleiner, H. E., Ryan, E. A., Dulik, D. M., Monks, T.
J ., and Lau, S. S. (1993) Identification of multi-S-substituted
conjugates of hydroquinone by HPLC-coulometric electrode array
analysis and mass spectrometry. Chem. Res. Toxicol. 6, 459-469.
(4) Hill, B. A., Davison, K. L., Dulik, D. M., Monks, T. J ., and Lau,
S. S. (1994) Metabolism of 2-(glutathion-S-yl)hydroquinone and
2,3,5-(triglutathion-S-yl)hydroquinone in the in situ perfused rat
kidney: Relationship to nephrotoxicity. Toxicol. Appl. Pharmacol.
129, 121-132.
The protein-S-bound adducts identified in the current
studies represent less than 1% of the administered dose
in the liver and less than 0.1% of the administered dose
in the kidneys. Using published values (21-23) for the
protein contents of liver and kidneys of rats (milligrams
of protein per gram of tissue wet weight), adduct con-
centrations (picomoles per milligram of protein) were
converted to total adduct equivalents in the respective
tissues. In the case of liver, the highest adduct levels that
were obtained corresponded to 0.89% and 0.54% of the
administered doses following repeat oral and single ip
administrations, respectively (11). The corresponding
highest values for kidney adducts were 0.07% of the
administered dose following repeat oral administration
and 0.09% of the administered dose following single ip
administration. Thus, protein arylation by HQ and
reactive HQ metabolites represents only a small percent-
age of the administered dose of HQ.
(5) O’Brien, P. J . (1991) Molecular mechanisms of quinone cytotox-
icity. Chem.-Biol. Interact. 80, 1-41.
(6) Hanzlik, R. P., Harriman, S. P., and Frauenhoff, M. M. (1994)
Covalent binding of benzoquinone to reduced ribonuclease. Adduct
structures and stoichiometry. Chem. Res. Toxicol. 7, 177-184.
(7) Slaughter, D. E., and Hanzlik, R. P. (1991) Identification of
epoxide- and quinone-derived bromobenzene adducts to protein
sulfur nucleophiles. Chem. Res. Toxicol. 4, 349-359.
(8) McDonald, T. A., Waidyanatha, S., and Rappaport, S. M. (1993)
Measurement of adducts of benzoquinone with hemoglobin and
albumin. Carcinogenesis 14, 1927-1932.
(9) Rombach, E. M., and Hanzlik, R. P. (1997) Detection of benzo-
quinone adducts to rat liver protein sulfhydryl groups using
specific antibodies. Chem. Res. Toxicol. 10, 1407-1411.
(10) Slaughter, D. E., Zheng, J ., Harriman, S., and Hanzlik, R. P.
(1993) Identification of covalent adducts to protein sulfur nucleo-
philes by alkaline permethylation. Anal. Biochem. 208, 288-295.
(11) Boatman, R. J ., English, J . C., Perry, L. G., and Fiorica, L. A.
(2000) Covalent protein adducts of hydroquinone in tissues from
rats: Quantitation of sulfhydryl-bound forms following single
gavage or intraperitoneal administration or repetitive gavage
administration. Chem. Res. Toxicol. 13, 861-872.
(12) Eckert, K.-G., Eyer, P., Sonnenbichler, J ., and Zetl, I. (1990)
Activation and detoxification of aminophenols. III. Synthesis and
structural elucidation of various glutathione addition products
to 1,4-benzoquinone. Xenobiotica 20, 351-361.
(13) McDonald, T. A., Waidyanatha, S., and Rappaport, S. M. (1993)
Production of benzoquinone adducts with hemoglobin and bone-
marrow proteins following administration of [13C6]benzene to rats.
Carcinogenesis 14, 1921-1925.
(14) Axworthy, D. B., Hoffman, K.-J ., Streeter, A. J ., Calleman, C. J .,
Pascoe, G. A., and Baillie, T. A. (1988) Covalent binding of
acetaminophen to mouse hemoglobin. Identification of major and
minor adducts formed in vivo and implications for the nature of
the arylating metabolites. Chem.-Biol. Interact. 68, 99-116.
(15) Cheever, K. L., DeBord, D. G., Swearengin, T. F., and Booth-
J ones, A. D. (1992) ortho-Toluidine blood protein adducts: HPLC
It has been previously suggested that under physi-
ologically relevant conditions, sulfhydryl-bound forms of
HQ bound to protein should predominate (6). In Table 4,
protein-S adducts in livers and kidneys of male and
female F344 rats determined in the current studies
(expressed as a percentage of the administered dose) are
compared with the residual radioactivity in tissues from
animals treated orally with [14C]HQ (unpublished results
from this laboratory). Although these values are uncor-
rected for protein turnover, the agreement is remarkably
good and provides convincing evidence that the sulfhy-
dryl-bound forms of HQ are, in fact, the primary adducts
present in these tissues.

860 Chem. Res. Toxicol., Vol. 13, No. 9, 2000
Boatman et al.
analysis with fluorescence detection after a single dose in the
adult male rat. Fundam. Appl. Toxicol. 18, 522-531.
(16) Skipper, P. L., and Tannenbaum, S. R. (1990) Protein adducts in
the molecular dosimetry of chemical carcinogens. Carcinogensis
11 (4), 507-518.
(17) Hlavica, P., Golly, P., Lehnerer, M., and Schulze, J . (1997)
Primary aromatic amines: Their N-oxidative bioactivation. Hum.
Exp. Toxicol. 16, 441-448.
(21) Landes, D. R., and Bough, W. A. (1976) Effects of chitosansa
coagulating agent for food processing wastessin the diets of rats
on growth and liver and blood composition. Bull. Environ.
Contam. Toxicol. 15 (5), 555-563.
(22) Nath, K. A., and Salahudeen, A. K. (1990) Induction of renal
growth and injury in the intact rat kidney by dietary deficiency
of antioxidants. J . Clin. Invest. 86, 1179-1192.
(23) Beyer, R. E., Morales-Corral, P. G., Ramp, B. J ., Kreitman, K.
R., Falzon, M. J ., Rhee, S. Y. S., Kuhn, T. W., Stein, M.,
Rosenwasser, M. S., and Cartwright, K. J . (1984) Elevation of
tissue coenzyme Q (ubiquinone) and cytochrome c concentrations
by endurance exercise in the rat. Arch. Biochem. Biophys. 234
(2), 223-229.
(18) Goldspink, D. F., and Kelly, F. J . (1984) Protein turnover and
growth in the whole body, liver, and kidney of the rat from foetus
to senility. Biochem. J . 217, 507-516.
(19) Tsao, T. C., Shi, J . D., Mortimore, G. E., Cragoe, E. J ., and Rabkin,
R. (1990) Modulation of kidney cell protein degradation by insulin.
J . Lab. Clin. Med. 116 (3), 369-376.
(20) Meister, A., and Anderson, M. E. (1983) Glutathione. Annu. Rev.
Biochem. 52, 711-760.
TX000037X