Full text of DOI:10.1093/toxsci/62.1.176

TOXICOLOGICAL SCIENCES 62, 176–182 (2001)
Copyright © 2001 by the Society of Toxicology
Association of Quinone-Induced Platelet
Anti-Aggregation with Cytotoxicity
Se-Ryun Kim, Joo-Young Lee, Moo-Yeol Lee, Seung-Min Chung, Ok-Nam Bae, and Jin-Ho Chung1
College of Pharmacy, Seoul National University, Shinrim-dong San 56–1, Seoul 151–742, Korea
Received January 9, 2001; accepted April 3, 2001
and prostacyclin (PGI2). If these controlling factors are dis-
Various anti-platelet drugs, including quinones, are being inves-
rupted by certain diseases or chemical exposures, excessive
tigated as potential treatments for cardiovascular disease because
platelet aggregation can occur, leading ultimately to cardiovas-
of their ability to prevent excessive platelet aggregation. In the
cular disease.
present investigation
3 naphthoquinones (2,3-dimethoxy-1,4-
To date, several potential mechanisms for inhibition of
platelet aggregation have been suggested. These include inhi-
bition of adenosine receptors and thromboxane synthase (Mar-
tinez et al., 1992; Vittori et al., 1996), increased cGMP (Bas-
senge and Stewart, 1988), increased NO-releasing compounds
(Civelli et al., 1994), and synthesis of prostacyclin analogues
(Katsube et al., 1993). Depletion of glutathione (GSH) in
platelets has been correlated with inhibition of platelet aggre-
gation by several compounds. These include helaniline (a
derivative of sesquiterpene lactone) and 1-chloro-2,4-dinitro-
benzene (CDNB), which deplete GSH (Bosia et al., 1985;
Schroder et al., 1990), diamide, which oxidizes GSH (Matsuda
et al., 1979; Patscheke and Worner, 1978), and N-ethylmale-
naphthoquinone [DMNQ], menadione, and 1,4-naphthoquinone
[4-NQ]) were compared for their abilities to inhibit platelet aggre-
gation, deplete glutathione (GSH) and protein thiols, and cause
cytotoxicity. Platelet-rich plasma, isolated from Sprague-Dawley
rats, was used for all experiments. The relative potency of the 3
quinones to inhibit platelet aggregation, deplete intracellular GSH
and protein thiols, and cause cytotoxicity was 1,4-NQ > mena-
dione >> DMNQ. Experiments using 2 thiol-modifying agents,
dithiothreitol (DTT) and 1-chloro-2,4-dintrobenzene (CDNB),
confirmed the key roles for GSH in quinone-induced platelet
anti-aggregation and for protein thiols in quinone-induced cyto-
toxicity. Furthermore, the anti-aggregative effects of a group of 12
additional quinone derivatives were positively correlated with
their ability to cause platelet cytotoxicity. Quinones that had a
weak anti-aggregative effect did not induce cytotoxicity (measured imide (NEM), which alkylates GSH (Hill et al., 1989).
as LDH leakage), whereas quinones that had a potent anti-aggre-
gative effect resulted in significant LDH leakage (84–96%). In one
instance, however, p-chloranil demonstrated a potent anti-aggre-
gative effect, but did not induce significant LDH leakage. This can
be explained by the inability of p-chloranil to deplete protein
thiols, even though intracellular GSH levels decreased rapidly.
These results suggest that quinones that deplete GSH in platelets
demonstrate a marked anti-aggregative effect. If this anti-aggre-
gative effect is subsequently followed by depletion of protein thiols,
cytotoxicity results.
Quinones are widely distributed in nature and are used
clinically as antitumor drugs, components of multivitamin for-
mulations, and anti-allergic agents (Mandel and Cohn, 1996;
Monks et al., 1992; Smith, 1985). Recently, quinones have
been shown to be platelet anti-aggregative agents by several
different mechanisms. Platelet aggregation was reduced by
2-chloro-3-methyl-1,4-naphthoquinone via inhibition of phos-
phoinositide breakdown (Ko et al., 1990). Alternatively, men-
adione and vitamin K analogues were suggested to interfere
with the mobilization and/or utilization of intracellular
Ca2ϩ (Blackwell et al., 1985). In addition, 2-[(4-cyanophe-
nyl)amino]-3-chloro-1,4-naphthalenedione inhibited throm-
boxane A2 (TXA2) synthase (Chang et al., 1997), and 2-me-
thoxy-5-methyl-1,4-benzoquinone inhibited TXA2 receptor
binding (Lauer and Anke, 1991). Other quinone substances,
such as derivatives of benzoquinone (Suzuki et al., 1997) and
naphthoquinone (Lien et al., 1996), also prevent platelet
aggregation, but the mechanism(s) of action have not been
clarified.
Key Words: quinones; platelet anti-aggregation; cytotoxicity;
protein thiols; glutathione.
Blood platelets play an important role in hemostasis, throm-
bosis, and initiation of various cardiovascular diseases. After
exposure to the vessel wall by physical and chemical stress,
platelets become rapidly activated and platelet aggregation and
adhesion occur (Mustard and Packman, 1979; Stormorken,
1984). The aggregation of platelets is normally controlled by
endogenous anti-aggregating factors such as nitric oxide (NO)
It is well known that quinones rapidly deplete intracellular
GSH in various tissues including hepatocytes, kidney cells, and
platelets (Brown et al., 1991; Seung et al., 1998; Stone et al.,
1996). Likewise, several quinones inhibit platelet aggregation
1 To whom correspondence should be addressed. Fax: (822) 885-4157.
E-mail: jhc302@plaza.snu.ac.kr.
176

QUINONE-INDUCED CYTOTOXICITY IN PLATELETS
177
TABLE 1
Inhibition of Platelet Aggregation by 3 Quinone Compounds
within a short period of time as described above. However, the
relationship between inhibition of platelet aggregation and
depletion of GSH by quinones has not been investigated. Our
laboratory is interested in examining the effects of quinones on
GSH depletion as a mechanism of platelet anti-aggregation.
Following rapid depletion of GSH, quinones can subse-
quently deplete protein thiols. We previously demonstrated
that protein thiol depletion was closely related with cytotoxic-
ity (Cho et al., 1997). Because both the anti-aggregative effect
and the depletion of GSH occur within a few minutes following
quinone exposure, whereas depletion of protein thiols and
DMNQ
Menadione
1,4-NQ
(␮M)
(␮M)
(␮M)
Agonist
IC50
IC100
IC50
IC100
IC50
IC100
Thrombin
ADP
Collagen
NDa
ND
ND
ND
ND
ND
137 Ϯ 5
125 Ϯ 12
23 Ϯ 2
255 Ϯ 8
286 Ϯ 28
59 Ϯ 12
42 Ϯ 3
23 Ϯ 5
5 Ϯ 1
76 Ϯ 15
48 Ϯ 2
12 Ϯ 3
Note. Platelet-rich plasma was preincubated with 2,3-dimethoxy-1,4-naph-
overt cytotoxicity occur at later time points, we postulated that thoquinone (DMNQ; 100–500 ␮M), menadione (5–300 ␮M), or 1,4-naphtho-
quinone (1,4-NQ; 1–100 ␮M) for 3 min at 37°C and then each agonist was
added to activate platelets. The concentrations of agonists were 1.6–1.8 U/ml,
6.4–8.0 ␮M, or 6.0–8.0 ␮g/ml for thrombin, collagen, or ADP, respectively.
The values of IC50 and IC100 were determined based upon the concentration-
anti-aggregation was correlated to depletion of GSH and that
this event precedes protein thiol depletion and cytotoxicity. To
further investigate this hypothesis, we examined the relation-
ships among platelet anti-aggregation, depletion of GSH, pro-
response curve using a computer program (PHARM/PCS). Changes in light
tein thiol depletion, and cytotoxicity using a large number of transmission were detected by a Lumi-aggregometer. Data represent means Ϯ
SEM from 3 experiments.
quinone substances in platelets.
aND, not determined; the highest concentration tested was 500 ␮M.
MATERALS AND METHODS
PRP (1 ϫ 109 platelets/ml) were centrifuged at 10,000 ϫ g for 20 s at room
Drugs and chemicals. The following chemicals and purified enzymes
temperature to obtain a platelet pellet. The pellets were resuspended with 0.6
were purchased from Sigma Chemical Co. (St. Louis, MO): menadione,
ml of 0.125 M perchloric acid containing 0.4 mM EDTA and centrifuged at
1-chloro-2,4-dinitro-benzene (CDNB), dithiothreitol (DTT), duroquinone,
10,000 ϫ g for 2 min to obtain the acid soluble fraction (supernatant). The
aloe-emodin, glutathione reduced form (GSH), thrombin, ADP, oxidized glu-
supernatant was mixed with 2 M KOH containing 0.3 M MOPS to remove
tathione (GSSG), triton X-100, trisodium citrate, pyruvic acid, DMSO, 5,5Ј-
excess perchloric acid and to adjust the pH to 7.0. A 0.2 ml aliquot of this
dithiobis (2-nitrobenzoic acid) (DTNB), N-ethylmaleimide (NEM), NADPH,
preparation was then assayed for glutathione. Total glutathione levels were
and NADH. The following chemicals were purchased from Aldrich Chemical
determined by the enzymatic recycling method described by Griffith (1980),
Co. (Gillingham, Dorset, U.K.): 1,4-benzoquinone, 1,4-naphthoquinone, and
with the following modification—a higher activity of glutathione reductase
2,3-dichloro-1,4-naphthoquinone. 2- or 6- (1-Oxopentyl)-5,8-dimethoxy-1,4-
(1.5 kU/ml in assay buffer) was used in order to optimize conditions for
naphthoquinone was provided by Dr. Byung-Zun Ahn (Chungnam National
assaying platelets.
University, Korea) and quinonelinedione derivatives were provided by Dr.
Chung-Kyu Ryu (Ewha Womans University, Korea). All other chemicals were
obtained from standard commercial sources.
Lactate dehydrogenase leakage. Leakage of lactate dehydrogenase (LDH)
from platelets was measured as described by Bergmeyer et al. (1965). LDH
activity was measured in both the incubation medium and platelets (lysed with
0.3% Triton X-100). LDH leakage was expressed as % of total enzyme activity.
Animals. Female Sprague-Dawley rats (Laboratory Animal Center of
Seoul National University, Korea) weighing 200–250 g were used throughout
all experiments. Prior to experiments, animals were acclimated for 1 week in
the laboratory animal facility maintained at constant temperature and humidity
Protein thiol levels. Protein thiol concentrations were measured using a
modification of the colorimetric method of Di Monte et al. (1984). One ml of
quinone-treated PRP was spun at 10,000 ϫ g for 20 s, and the supernatant was
with a 12-h light/dark cycle. Food and water were provided ad libitum.
discarded. The pellet was washed once with 5% perchloric acid and then
Preparations of platelets. All procedures were conducted at room temper-
ature, and the use of glass containers and pipettes was avoided. Blood was
resuspended in 2.5 ml of Tris-EDTA buffer (0.5 M Tris, 5 mM EDTA, pH 7.6).
DTNB (250 ␮M final concentration) was added and, after 20 min, the absor-
collected from the abdominal aorta of ether-anesthetized rats using a syringe
bance was measured at 412 nm. Protein thiol levels were calculated on the
containing sodium citrate (3.8%, 9:1), and then centrifuged for 15 min at
basis of a glutathione calibration curve.
150 ϫ g at room temperature. Platelet-rich plasma (PRP) was obtained from
the supernatant resulting from this relatively low g-force centrifugation.
Statistical analysis. The means and standard errors of means were calcu-
lated for all treatment groups. The data were subjected to one-way analysis of
Platelet-poor plasma (PPP) was obtained from the supernatant of a 20 min,
variance (ANOVA) followed by Duncan’s multiple range test to determine
1500 ϫ g centrifugation of the blood cell residue resulting from the first spin.
which means were significantly different from each other or control. In all
cases, a p-value of Ͻ 0.05 was used to determine significance.
Throughout all experiments, the platelet number was adjusted to 3 ϫ 108
platelets/ml by diluting PRP with PPP.
Measurement of platelet aggregation. Platelet aggregation was measured
RESULTS
by light transmission, with 100% calibrated as the absorbance of PPP and 0%
calibrated as the absorbance of PRP. PRP suspension in a silicon-coated
cuvette was stirred at 1200 rpm for 1 min before addition of quinones.
Dimethylsulfoxide (DMSO) was used as the vehicle for quinones, such that the
final concentration of DMSO in the incubation medium was 0.5%. This
concentration was shown to have no effect on either platelet aggregation
induced by agonists or platelet lysis. Changes in light transmission were
detected by a Lumi-aggregometer (Chrono-log Corp., Havertown, PA).
The capacity for 3 naphthoquinone compounds, 2,3-di-
methoxy-1,4-naphthoquinone (DMNQ), menadione, and 1,4-
naphthoquinone (1,4-NQ), to prevent platelet aggregation was
investigated. PRP was pretreated with 1 of 3 quinones for 3
min and then thrombin, ADP, or collagen was added to activate
platelets. Table 1 summarizes the relative inhibitory capacity
Intracellular glutathione levels. Sample preparation was accomplished by
incubating PRP with 100 ␮M quinones. One ml-aliquots of quinone-treated of the 3 quinones. Quinone concentrations that resulted in 50%

178
KIM ET AL.
platelet aggregation by menadione, while pretreatment with 0.2
mM CDNB significantly potentiated the inhibitory effect of
menadione on platelet aggregation (Fig. 2A). A similar pattern
was observed for 1,4-NQ’s effects (Fig. 2B).
Since previous studies demonstrated that quinones can de-
crease protein thiol levels following rapid depletion of GSH in
platelets (Cho et al., 1997), total protein thiol levels were
measured following addition of 250 ␮M DMNQ, menadione,
or 1,4-NQ (Fig. 3). No effect on protein thiols was observed
with DMNQ treatment. However, menadione and 1,4-NQ de-
pleted protein thiols in a time-dependent manner, with 1,4-NQ
pretreatment more rapidly depleting protein thiol levels than
menadione. The order of the 3 quinones at 250 ␮M to deplete
protein thiols was 1,4-NQ Ͼ menadione ϾϾ DMNQ.
To investigate the role of protein thiols in the cytotoxic
effects of these quinone compounds on platelets, the ability of
DMNQ, menadione, or 1,4-NQ to induce platelet cytotoxicity
was assessed at various time points by measuring LDH leakage
(Fig. 4). At equimolar concentrations of 250 ␮M, DMNQ did
not induce LDH leakage, while menadione and 1,4-NQ in-
duced LDH leakage in a time-dependent manner. 1,4-NQ ap-
peared more effective than menadione, since LDH leakage was
70% of total LDH activity by 30 min and 100% of total by 60
min. The order of the 3 quinines at 250 ␮M to cause cytotox-
icity was also 1,4-NQ Ͼ menadione ϾϾ DMNQ (Fig. 4), a
similar observation to that describing effects on platelet aggre-
gation, glutathione depletion, and protein thiols depletion.
In addition, the ability of DTT and CDNB to modulate
FIG. 1. Effect of quinones on total GSH levels in platelets. Platelet rich
plasma (PRP) suspensions were incubated in the presence of 100 ␮M quinone
for 3 or 10 min. Total GSH levels in platelets were determined as described in
Methods section. Values represent mean Ϯ SEM from 3 experiments. *Rep-
resent significant differences from control (p Ͻ 0.05).
inhibition of platelet aggregation (IC50) or 100% inhibition
of aggregation (IC100) were determined. Concentrations of quinone-induced cytotoxicity was investigated (Fig. 5). PRP
DMNQ as great as 500 ␮M failed to inhibit aggregation was treated with menadione or 1,4-NQ following exposure to
induced by all 3 agonists. On the other hand, menadione and CDNB and DTT. Neither CDNB nor DTT alone resulted in
1,4-NQ consistently inhibited platelet aggregation, with IC100s LDH leakage (data not shown). Similar to the results shown
being 2 times greater than the IC50s, irrespective of the specific above with GSH (Fig. 2), pretreatment with 1 mM DTT almost
agonist. The relative potency of quinones to inhibit platelet completely inhibited menadione- and 1,4-NQ-induced cytotox-
aggregation was 1,4-NQ Ͼ menadione ϾϾ DMNQ.
To determine if inhibition of platelet aggregation by qui-
nones was related to the depletion of glutathione (GSH), total
GSH levels in PRP were determined following exposure to 100
␮M DMNQ, menadione, or 1,4-NQ (Fig. 1). DMNQ did not
alter intracellular total GSH levels compared to control. In
contrast, menadione treatment resulted in depletion of intracel-
lular GSH by 45% by 3 min and complete depletion by 10 min.
1,4-NQ completely depleted intracellular GSH within 3 min.
These data were consistent with quinone potency to inhibit
aggregation, as summarized in Table 1. To confirm the possible
role of GSH in platelet aggregation, the effects of a thiol-
donating agent (DTT) and a GSH-depleting agent (CDNB) on
the inhibition of thrombin-induced platelet aggregation by
FIG. 2. Modification of anti-aggregative effects of menadione and 1,4-
menadione and 1,4-NQ were investigated (Fig. 2). After PRP naphthoquinone by DTT and CDNB. PRP suspensions were pretreated with 1
mM DTT for 5 min, or 0.2 mM CDNB for 10 min prior to thrombin addition.
(A) 140 ␮M menadione or (B) 40 ␮M 1,4-naphthoquinone was added to PRP
suspension 3 min before thrombin addition. Changes in light transmission were
detected by a Lumi-aggregometer. DTT or CDNB alone did not affect throm-
was pretreated with CDNB or DTT for the indicated time, PRP
was treated with approximately IC50s of menadione (140 ␮M)
or 1,4-NQ (40 ␮M). DTT or CDNB treatment alone did not
affect aggregation in platelets (data not shown). Pretreatment
with 1 mM DTT almost completely prevented the inhibition of pendent experiments.
bin-induced platelet aggregation. Data are representative tracings of 3 inde-

QUINONE-INDUCED CYTOTOXICITY IN PLATELETS
179
FIG. 3. Effect of quinones on protein thiol levels in platelets. PRP sus-
pensions were incubated with 250 ␮M quinone and total protein thiol levels
were determined at the indicated time points. Total protein thiol level at time
0 was 78 Ϯ 6 nmol/mg protein. Values represent mean Ϯ SEM from 3
experiments. *Represent significant differences from control (p Ͻ 0.05).
FIG. 4. Effect of quinones on LDH release from platelets. All treatments
were the same as described in Fig. 3 and release of LDH was determined as
described in Methods section. Values represent mean Ϯ SEM from 3 experi-
ments. *Represent significant differences from control (p Ͻ 0.05).
modest LDH leakage (20% of total activity). Since we ob-
served that protein thiols depletion was associated with qui-
none cytotoxicity (Figs. 3 and 4), we postulated that the rela-
tively low cytotoxicity of p-chloranil was associated with a
minimal capacity to deplete protein thiols. The effects of
p-chloranil on total soluble glutathione levels (Fig. 6A) and
protein thiol levels (Fig. 6B) in platelets were studied. GSH
icity, while pretreatment with 0.2 mM CDNB greatly potenti-
ated the effect of the 2 quinones (Fig. 5).
To determine if the apparent association between anti-aggrega-
tive effects and cytotoxicity observed with DMNQ, menadione, or
1,4-NQ was applicable to a broad class of quinones, the same 2
parameters (anti-aggregative effect and cytotoxicity) were mea-
sured for a number of additional quinone derivatives (Table 2).
Quinones demonstrating a weak anti-aggregative effect (i.e.,
IC50 Ͼ 250 ␮M), such as duroquinone, 2-(1-oxopentyl)-5,8-di-
methoxy-1,4-naphthoquinone (2-oxoDMNQ), 6-(2,3,4-trifluoro-
benzyl)-anilino-5,8-quinolinedione (FAQ 1), 6-(3-fluorobenzyl)-
anilino-5,8-quinolinedione (FAQ 2), or anthraquinones, did not
induce LDH leakage. Conversely, quinones demonstrating a po-
tent anti-aggregative effect (i.e., IC50 ϭ 7.0 ␮Mϳ137 ␮M), such
as 1,4-benzoquinone, 2,3-dichloro-1,4-naphthoquinone (DCNQ),
6-(1-oxopentyl)-5,8-dimethoxy-1,4-naphthoquinone (6-oxoDMNQ),
6-(2,4-difluorobenzyl)-anilino-5,8-quinolinedionylchloride (FAQ
4), or menadione, induced significant LDH leakage (84–96%). In
addition, the most potent anti-aggregative quinone, 2,3-dichloro-
1,4-naphthoquinone (i.e., IC50 ϭ 7.0 ␮M), also most rapidly
induced LDH leakage (69.3 Ϯ 2.5% of the total LDH level within
FIG. 5. Modification of cytotoxicity of menadione and 1,4-naphthoqui-
30 min, data not shown). These results indicate that, in general, none by DTT and CDNB. PRP suspensions were pretreated with 1 mM DTT
for 5 min, or 0.2 mM CDNB for 30 min prior to addition of (A) 250 ␮M
menadione or (B) 200 ␮M 1,4-naphthoquinone (1,4-NQ). LDH release was
determined at the indicated time point. DTT or CDNB alone did not induce any
significant cytotoxicity in platelets. Values represent mean Ϯ SEM from 3
the more potent the anti-aggregative effect of a quinone substance,
the more cytotoxic that quinone substance will be.
An exception to this general rule, however, was p-chloranil.
Among the 15 quinones tested, only p-chloranil demonstrated
a potent anti-aggregative effect (IC50 ϭ 7.5 ␮M) with relatively (p Ͻ 0.05).
experiments. *Represent significant differences from corresponding control

180
KIM ET AL.
TABLE 2
Correlation between Anti-Aggregative Effect (IC50) and Cytotoxicity (LDH) by Various Quinones
Compounds
IC50 (␮M)
LDH (%)
Control (DMSO)
Benzoquinones
—
7.0 Ϯ 0.0
2,3,5,6-Tetramethyl-1,4-benzoquinone (Duroquinone)
p-Chloranil
1,4-Benzoquinone
Ն 250
7.5 Ϯ 1.2
106 Ϯ 2
8.1 Ϯ 1.7
20.1 Ϯ 1.5
93.0 Ϯ 1.5
Naphthoquinones
2,3-Dimethoxy-1,4-naphthoquinone
Menadione
2,3-Dichloro-1,4-naphthoquinone (DCNQ)
1,4-Naphthoquinone
Ն 250
137 Ϯ 5
7.0 Ϯ 2.2
42 Ϯ 3
11.0 Ϯ 1.6
87.0 Ϯ 3.2
95.3 Ϯ 2.0
96.0 Ϯ 3.4
(1-Oxoalkyl)-5,8-dimethoxy-1,4-naphthoquinones
2-(1-Oxopentyl)-5,8-dimethoxy-1,4-naphthoquinone (2-oxoDMNQ)
6-(1-Oxopentyl)-5,8-dimethoxy-1,4-naphthoquinone (6-oxoDMNQ)
Quinolinediones
Ն 250
11.0 Ϯ 0.4
8.6 Ϯ 1.7
91.0 Ϯ 3.0
6-(2,3,4-Trifluorobenzyl)-anilino-5,8-quinolinedione (FAQ 1)
6-(3-Fluorobenzyl)-anilino-5,8-quinolinedione (FAQ 2)
6-(2,3,4-Trifluorobenzyl)-anilino-5,8-quinolinedionylchloride (FAQ 3)
6-(2,4-Difluorobenzyl)-anilino-5,8-quinolinedionylchloride (FAQ 4)
Anthraquinones
Ն 250
Ն 250
186 Ϯ 13
63 Ϯ 5
8.0 Ϯ 1.6
10.9 Ϯ 3.5
25.0 Ϯ 3.4
84.0 Ϯ 6.2
2-(1-Hydroxyalkyl)-1,4,9,10-anthradiquinone
1,8-Dihydroxy-3-(hydroxymethyl)-anthraquinone (Aloe-emodine)
Ն 250
Ն 250
9.2 Ϯ 1.5
12.9 Ϯ 3.7
Note. The concentration resulting in 50% inhibition of platelet aggregation (IC50) for each of 15 quinone substances was calculated to compare their relative
potency. LDH leakage was measured following 120 min incubation to each quinone at 250 ␮M to compare relative cytotoxicity. Values represent mean Ϯ SEM
from 3 experiments.
was rapidly depleted by 100 ␮M p-chloranil, while concentra- the compounds that inhibit platelet aggregation with IC50s Ͻ
tions as high as 250 ␮M p-chloranil failed to induce significant 250 ␮M (Table 2) also significantly deplete cellular GSH
depletion of protein thiols with even much longer durations of levels within 10 min (data not shown), suggesting that the
exposure.
anti-aggregative effect of quinones is related to GSH depletion.
To date, several mechanisms have been proposed for inhi-
bition of platelet aggregation by GSH depletion, but the exact
DISCUSSION
Our current research in platelets can be summarized with the
following observations: (1) quinones that deplete GSH in plate-
lets demonstrate a marked anti-aggregative effect; the more
potent the ability to deplete GSH, the more potent the anti-
aggregative effect; (2) quinones that subsequently deplete pro-
tein thiols eventually lead to cytotoxicity. These observations
lead us to conclude that if GSH depletion by quinones is
subsequently followed by depletion of protein thiols, the anti-
aggregative effect of the quinones may represent an early event
of cytotoxicity.
It has been previously reported that GSH plays an important
role in platelet aggregation, and that GSH-depleting chemicals
inhibit platelet aggregation significantly (Bosia et al., 1985;
Heptinstall et al., 1987; Schroder et al., 1990). However, the
FIG. 6. Depletion of GSH and protein thiols by p-chloranil. PRP suspen-
association between the depletion of GSH and inhibition of sions were incubated in the presence of 100 ␮M p-chloranil (for total GSH) or
250 ␮M p-chloranil (for protein thiols). Total GSH and protein thiol levels
were determined at the indicated time point, respectively. (A) Total GSH,
(circle) Control; (triangle) 100 ␮M p-chloranil. (B) Protein thiols, (circle)
Control; (triangle) 250 ␮M p-chloranil. Values represent mean Ϯ SEM from
aggregation in platelets has not been reported for quinones.
Our data in Table 1 and Figure 1 demonstrate that the potency
of GSH depletion for 3 quinone model compounds (DMNQ,
menadione, and 1,4-NQ) correlates well with their potency to
inhibit aggregation. In addition, among all the quinones tested, (p Ͻ 0.05).
3 experiments. *Represent significant differences from corresponding control

QUINONE-INDUCED CYTOTOXICITY IN PLATELETS
181
mechanism remains unclear. Previous research (Schroder et protein thiols also play a critical role in platelet cytotoxicity as
al., 1990) has shown that GSH depletion by some substances well.
other than quinones was associated with inhibition of platelet
p-Chloranil, an exceptional quinone among 15 quinones
aggregation and suggested that this possibly caused cytoskel- tested, has a potent anti-aggregative effect but it induces a
etal protein alterations. Our research has shown that GSH weak cytotoxicity. p-Chloranil was effective at depleting in-
depletion by quinones also caused inhibition of platelet aggre- tracellular GSH, but did not significantly deplete protein thiols
gation. It is possible that cytoskeletal alterations may be the (Fig. 6). This modification of intracellular GSH but not protein
key mechanism for our observations. This premise is supported thiols by p-chloranil resulted in significant anti-aggregative
by other studies (Bellomo et al., 1990; Thor et al., 1988) in effect in the absence of significant cytotoxicity. These results
isolated hepatocytes that showed that quinones induce cy- suggest that some quinones, such as p-chloranil, which selec-
toskeletal damage mediated by depletion of glutathione and tively deplete GSH without affecting protein thiol levels, may
protein thiols, leading to plasma membrane blebbing and ulti- be potentially effective anti-platelet drugs. However, caution is
mately cell death.
necessary in interpreting the potential use of quinone com-
The relative potency (IC50s and IC100s) for the 3 model pounds as anti-platelet drugs because quinones also deplete
naphthoquinones to inhibit platelet aggregation was 1,4-NQ Ͼ intracellular GSH in tissues other than platelets and this deple-
menadione ϾϾ DMNQ, regardless of which of 3 agonists tion has itself been shown to be toxic to cells in these tissues.
(collagen, ADP, thrombin) was used (Table 1). Both menadi- For example, p-hydroxybenzoate ester-induced cytotoxicity in
one and 1,4-NQ were most effective at inhibiting platelet hepatocytes (Nakagawa and Moldeus, 1998), 2-amino-5-chlo-
aggregation induced by collagen, as compared to aggregation rophenol-induced toxicity in renal cortical slices (Valentovic et
induced by the other 2 agonists. There is currently no known al., 1999), and glutamate-induced cytotoxicity in brain cells
explanation for why collagen’s action is most easily affected (Pereira and Oliveira, 1997) are primarily dependent upon the
by quinones, but this observation has been reported in several depletion of intracellular GSH.
studies using quinones (Chang et al., 1997; Lien et al., 1996).
In summary, the depletion of GSH by quinones results in the
The effects of DTT and CDNB on the inhibition of platelet inhibition of platelet aggregation, and the anti-aggregative ef-
aggregation by quinones were obtained when platelet aggrega- fect of the quinones may be an early event in cytotoxicity, as
tion was induced by thrombin (Fig. 2). Similar effects, how- represented by protein thiol depletion. As a consequence, al-
ever, were observed using ADP and collagen agonists (data not though much research has focused on the possibility that qui-
shown), suggesting that GSH-depletion by quinones in plate- none substances could be developed as anti-platelet drug ther-
lets plays a key role in modulating platelet aggregation induced apy, our results suggest that caution must be applied, since it
by any of the 3 agonists.
appears that quinone inhibition of platelet aggregation precedes
In general, quinones rapidly deplete GSH and subsequently cytotoxicity.
deplete protein thiols in various tissues such as liver, kidney,
and heart (Brown et al., 1991; Di Monte et al., 1984; Tzeng et
al., 1994). In platelets, menadione also rapidly depletes GSH
ACKNOWLEDGMENTS
within 10 min and subsequently depletes protein thiols for up
to 2 h, finally leading to cytotoxicity (Cho et al., 1997). In an
attempt to confirm a role for GSH and protein thiols in platelet
aggregation and cytotoxicity, we pre-exposed cells to CDNB
and DTT, 2 thiol modifying agents, to determine their ability to
modulate thrombin-induced platelet aggregation and quinone-
induced cytotoxicity. DTT was used as a donator of intracel-
lular thiols based upon the report that DTT donates sulfhydryl
groups to GSH and protein thiols (Sandy et al., 1988). CDNB
was used to deplete intracellular thiols. CDNB alone did not
affect thrombin-induced platelet aggregation. However, pre-
treatment with CDNB potentiated the inhibitory effects of
1,4-NQ and menadione on platelet aggregation (Fig. 2). The
major target of CDNB in platelets appears to be GSH (Bosia et
al., 1985), but in our experimental system, CDNB may also
induce alteration of protein thiol levels, since it subsequently
depletes protein thiols by 20% throughout 2-h incubation (data
not shown). We have demonstrated that pretreatment with
CDNB potentiates quinone-induced cytotoxicity in platelets,
whereas DTT completely protects (Fig 5), suggesting that
This work was supported by National Research Laboratory (NRL) Program
of the Korean Ministry of Science and Technology and by 2000 BK21 project
for Medicine, Dentistry and Pharmacy. We thank Dr. David Thompson for all
his help to review this manuscript.
REFERENCES
Bassenge, E., and Stewart, D. J. (1988). Interdependence of pharmacologi-
cally-induced and endothelium-mediated coronary vasodilation in antiangi-
nal therapy. Cardiovasc. Drugs Ther. 2, 27–34.
Bellomo, G., Mirabelli, F., Richelmi, P., Malorni, W., and Orrenius, S. (1990).
The cytoskeleton as a target in quinone toxicity. Free Radic. Res. Commun.
8, 391–399.
Bergmeyer, H. U., Dernt, E., and Hess, S. (1965). Lactate dehydrogenase. In
Methods in Enzymatic Analysis (H. U. Bergmeyer, Ed.), p. 736, Acacemic
Press, New York.
Blackwell, G. J., Radomski, M., and Moncada, S. (1985). Inhibition of human
platelet aggregation by vitamin K. Thromb. Res. 37, 103–114.
Bosia, A., Spangenberg, P., Ghigo, D., Heller, R., Losche, W., Pescarmona,
G. P., and Till, U. (1985). Effect of GSH depletion by 1-chloro-2,4-
dinitrobenzene on human platelet aggregation, arachidonic acid oxidative
metabolism and cytoskeletal proteins. Thromb. Res. 37, 423–434.

182
KIM ET AL.
Brown, P. C., Dulik, D. M., and Jones, T. W. (1991). The toxicity of
menadione (2-methyl-1,4-naphthoquinone) and two thioether conjugates
studied with isolated renal epithelial cells. Arch. Biochem. Biophys. 285,
187–196.
(1992). Contemporary issues in toxicology: Quinone chemistry and toxicity.
Toxicol. Appl. Pharmacol. 112, 2–16.
Mustard, J. F., and Packman, M. A. (1979). Factors influencing platelet
function: Adhesion, release, and aggression. Pharmacol. Rev. 22, 97–187.
Chang, T. S., Kim, H. M., Lee, K. S., Khil, L. Y., Mar, W. C., Ryu, C. K., and
Moon, C. K. (1997). Thromboxane A2 synthase inhibition and thromboxane
A2 receptor blockade by 2-[(4-cyanophenyl) amino]-3-chloro-1.4-naphtho-
quinone (NQ-Y15) in rat platelets. Biochem. Pharmacol. 54, 259–268.
Nakagawa, Y., and Moldeus, P. (1998). Mechanism of p-hydroxybenzoate
ester-induced mitochondrial dysfunction and cytotoxicity in isolated rat
hepatocytes. Biochem. Pharmacol. 55, 1907–1914.
Patscheke, H., and Worner, P. (1978). Sequential effects of the thiol-oxidizing
agent, diamide, on human platelets. Thromb. Res. 12, 609–618.
Cho, Y. S., Kim, M. J., Lee, J. Y., and Chung, J. H. (1997). The role of thiols
in protecting against simultaneous toxicity of menadione to platelet plasma
and intracellular membranes. J. Pharmacol. Exp. Ther. 280, 1335–1340.
Pereira, C. M., and Oliveira, C. R. (1997). Glutamate toxicity on a PC 12 cell
line involves glutathione (GSH) depletion and oxidative stress. Free Radic.
Biol. Med. 23, 637–647.
Civelli, M., Caruso, P., Giossi, M., Bergmaschi, M., Razzetti, R., Bongrani, S.,
and Gasco, A. (1994). CHF 2206, a new potent vasodilating and anti-
aggregating drug as potential nitric oxide donor. Eur. J. Pharmacol. 255,
17–24.
Sandy, M. S., Di Monte, D., and Smith, M. T. (1988). Relationships between
intracellular vitamin E, lipid peroxidation, and chemical toxicity in hepato-
cytes. Toxicol. Appl. Pharmacol. 93, 288–297.
Di Monte, D., Ross, D., Bellomo, G., Eklow, L., and Orrenius, S. (1984).
Alterations in intracellular thiol homeostasis during the metabolism of
menadione by isolated rat hepatocytes. Arch. Biochem. Biophys. 235, 334–
342.
Seung, S. A., Lee, J. Y., Lee, M. Y., Park, J. S., and Chung, J. H. (1998). The
relative importance of oxidative stress versus arylation in the mechanism of
quinone-induced cytotoxicity to platelets. Chem. -Biol. Interact. 113, 133–
144.
Griffith, O. W. (1980). Determination of glutathione and glutathione disulfide
using glutathione reductase and 2-vinylpyridine. Anal. Biochem. 106, 207–
212.
Schroder, H., Losche, W., Strobach, H., Leven, W., Willuhn, G., Till, U., and
Schror, K. (1990). Helenalin and 11alpha, 13-dihydrohelenalin, two constit-
uents from Arnica montana L., inhibit human platelet function via thiol-
dependent pathways. Thromb. Res. 57, 839–845.
Heptinstall, S., Groenewegen, W. A., Spangenberg, P., and Losche, W. (1987).
Extracts of feverfew may inhibit platelet behaviour via neutralization of
sulphydryl groups. J. Pharm. Pharmacol. 39, 459–465.
Smith, M. T. (1985). Quinones as mutagens, carcinogens, and anticancer
agents: Introduction and overview. J. Toxicol. Environ. Health 16, 665–672.
Hill, T. W., White, J. G., and Rao, G. H. R. (1989). The influence of
glutathione depleting agents on human platelet function. Thromb. Res. 53,
457–465.
Stone, V., Coleman, R., and Chipman, J. K. (1996). Comparison of the effects
of redox cycling and arylating quinones on hepatobiliary function and
glutathione homeostasis in rat hepatocyte couplets. Toxicol. Appl. Pharma-
col. 138, 195–200.
Katsube, N., Sakaguchi, K., Fujitani, B., and Aishita, H. (1993). Anti-platelet
and anti-thrombotic effects of OP-41483, alpha-CD, a prostacyclin ana-
logue, in experimental animals. Prostaglandins Leukot. Essent. Fatty Acids
49, 795–804.
Stormorken, H. (1984). Platelets in hemostasis and thrombosis. In Platelet
Responses and Metabolism (H. Holmsen, Ed.), pp. 3–32. CRC Press, Boca
Raton, FL.
Ko, F. N., Sheu, S. J., Liu, Y. M., Huang, T. F., and Teng, C. M. (1990).
Inhibition of rabbit platelet aggregation by 1,4-naphthoquinone. Thromb.
Res. 57, 453–463.
Suzuki, K., Tatsuoka, T., Ishihara, T., Ogino, R., Miyazaki, T., Satoh, F.,
Miyano, S., and Sumoto, K. (1997). 2-Arylmethyl-1,4-benzoquinones. II.
Novel inhibitors of platelet aggregation: Synthesis and pharmacological
evaluation. Chem. Pharm. Bull. 45, 668–674.
Lauer, U., and Anke, T. (1991). 2-Methoxy-5-methyl-1,4-benzoquinone, a
thromboxane A2 receptor antagonist. J. Antibiotics 44, 59–65.
Lien, J. C., Huang, L. J., Wang, J. P., Teng, C. M., Arese, P., Lee, K. H., and Thor, H., Mirabelli, F., Salis, A., Cohen, G. M., Bellomo, G., and Orrenius, S.
Kuo, S. C. (1996). Synthesis and anti-platelet, anti-inflammatory and anti-
allergic activities of 2,3-disubstituted 1,4-naphthoquinones. Chem. Pharm.
Bull. 44, 1181–1187.
(1988). Alterations in hepatocyte cytoskeleton caused by redox cycling and
alkylating quinones. Arch. Biochem. Biophys. 266, 397–407.
Tzeng, W. F., Chiou, T. J., Wang, C. P., Lee, J. L., and Chen, Y. H. (1994).
Cellular thiol as a determinant of responsiveness to menadione in cardio-
myocytes. J. Mol. Cell Cardiol. 26, 889–897.
Mandel, H. G., and Cohn, V. H. (1996). Vitamin K. In The Pharmacological
Basis of Therapeutics (J. G Hardman and L. E. Limbird, Eds.), pp. 1582–
1585. McGraw-Hill, New York.
Valentovic, M., Meadows, M. K., Harmon, R. C., Ball, J. G., Hong, S. K., and
Rankin, G. O. (1999). 2-Amino-5-chlorophenol toxicity in renal cortical
slices from Fischer 344 rats: Effect of antioxidants and sulfhydryl agents.
Toxicol. Appl. Pharmacol. 161, 1–9.
Martinez, G. R., Walker, K. A., Hirschfeld, D. R., Bruno, J. J., Yang, D. S., and
Maloney, P. J. (1992). 3,4-Dihydroquinolin-2 (1H)-ones as combined inhib-
itors of thromboxane A2 synthase and cAMP phosphodiesterase. J. Med.
Chem. 35, 620–628.
Vittori, S., Camaioni, E., Di Francesco, E., Volpini, R., Monopoli, A., Dioni-
sotti, S., Ongini, E., and Cristalli, G. (1996). 2-Alkenyl and 2-alkyl deriv-
atives of adenosine and adenosine-5Ј-N-ethyluronamide: Different affinity
and selectivity of E- and Z-diastereomers at A2A adenosine recepters.
J. Med. Chem. 39, 4211–4217.
Matsuda, S., Ikeda, Y., Aoki, M., Toyama, K., Watanabe, K., and Ando, Y.
(1979). Role of reduced glutathione on platelet function. Thrombos. Hae-
mostas. 42, 1324–1331.
Monks, T. J., Hanzlik, R. P., Cohen, G. M., Ross, D., and Graham, D. G.