The role of glutathione in DNA damage by potassium bromate in vitro

Article abstract of DOI:10.1093/mutage/15.4.311

We have investigated the role of reduced glutathione (GSH) in the genetic toxicity of the rodent renal carcinogen potassium bromate (KBrO₃). A statistically significant increase in the concentration of 8-oxodeoxyguanosine (8-oxodG) relative to deoxyguanosine was measured following incubation of calf thymus DNA with KBrO₃ and GSH or N-acetylcysteine (NACys). This was dependent on these thiols and was associated with the loss of GSH and production of oxidized glutathione. A short-lived (<6 min) intermediate was apparent which did not react with the spin trap dimethylpyrroline N-oxide. DNA oxidation was not evident when potassium chlorate (KCIO₃) or potassium iodate (KIO₃) were used instead of KBrO₃, though GSH depletion also occurred with KIO₃, but not with KCIO₃. Other reductants and thiols in combination with KBrO₃ did not cause a significant increase in DNA oxidation. DNA strand breakage was also induced by KBrO₃ in human white blood cells (5 mM) and rat kidney epithelial cells (NRK-52E, 1.5 mM). This was associated with an apparent small depletion of thiols in NRK-52E cells at 15 min and with an elevation of 8-oxodG at a delayed time of 24 h. Depletion of intra-cellular GSH by diethylmaleate in human lymphocytes decreased the amount of strand breakage induced by KBrO₃. Extracellular GSH, however, protected against DNA strand breakage by KBrO₃, possibly due to the inability of the reactive product to enter the cell. In contrast, membrane-permeant NACys enhanced KBrO₃-induced DNA strand breakage in these cells. DNA damage by KBrO₃ is therefore largely dependent on access to intracellular GSH.

Full text of DOI:10.1093/mutage/15.4.311

Mutagenesis vol.15 no.4 pp.311–316, 2000
The role of glutathione in DNA damage by potassium bromate
in vitro
J.L.Parsons and J.K.Chipman1
This was also observed by DeAngelo et al. (1998), who also
found renal cell tumours in male B6C3F1 mice (although the
treatment was not dose related). The carcinogenicity of KBrO3
in rodents is believed to be due to its ability to oxidize DNA.
This was suggested by Kasai et al. (1987) and Sai et al. (1991,
1992), who observed increased levels of 8-oxodeoxyguanosine
(8-oxodG) relative to deoxyguanosine (dG) in the kidney DNA
of male treated rats. The oxidation of DNA was found to be
inhibited by various antioxidants, in particular by reduced
glutathione (GSH) (Sai et al., 1992). These results suggest a
protective role of GSH against KBrO3-induced DNA oxida-
tion in vivo.
In contrast, an in vitro study by Ballmaier and Epe (1995)
suggests that GSH can activate KBrO3 to a reactive species
that is capable of oxidizing DNA. It was found that KBrO3
(only in the presence of GSH) produced DNA strand breaks
and modifications sensitive to formamidopyrimidine-DNA gly-
cosylase protein (Fpg protein; a repair enzyme recognizing
particularly 8-oxodG). This involved an apparent short-lived
intermediate that was revealed not to share the characteristics
of the hydroxyl radical, based on a comparison of DNA
damage profiles.
The mechanism whereby KBrO3 can oxidize DNA is there-
fore not clear, particularly regarding a direct or indirect
interaction. GSH therefore has a potential activation and
protection role in KBrO3-mediated DNA oxidation. We have
now further tested the role of GSH (and other thiols and
reductants) in DNA oxidation and strand breakage by KBrO3
in vitro using isolated calf thymus DNA, human white blood
cells and cultured rat kidney epithelial cells (NRK-52E). We
have also tested if DNA damage can occur from other halates.
School of Biosciences, The University of Birmingham, Edgbaston,
Birmingham B15 2TT, UK
We have investigated the role of reduced glutathione
(GSH) in the genetic toxicity of the rodent renal carcinogen
potassium bromate (KBrO3). A statistically significant
increase in the concentration of 8-oxodeoxyguanosine
(8-oxodG) relative to deoxyguanosine was measured follow-
ing incubation of calf thymus DNA with KBrO3 and GSH
or N-acetylcysteine (NACys). This was dependent on these
thiols and was associated with the loss of GSH and
production of oxidized glutathione. A short-lived (Ͻ6 min)
intermediate was apparent which did not react with the
spin trap dimethylpyrroline N-oxide. DNA oxidation was
not evident when potassium chlorate (KClO3) or potassium
iodate (KIO3) were used instead of KBrO3, though GSH
depletion also occurred with KIO3, but not with KClO3.
Other reductants and thiols in combination with KBrO3
did not cause a significant increase in DNA oxidation. DNA
strand breakage was also induced by KBrO3 in human
white blood cells (5 mM) and rat kidney epithelial cells
(NRK-52E, 1.5 mM). This was associated with an apparent
small depletion of thiols in NRK-52E cells at 15 min and
with an elevation of 8-oxodG at a delayed time of 24 h.
Depletion of intra-cellular GSH by diethylmaleate in human
lymphocytes decreased the amount of strand breakage
induced by KBrO3. Extracellular GSH, however, protected
against DNA strand breakage by KBrO3, possibly due to
the inability of the reactive product to enter the cell. In
contrast, membrane-permeant NACys enhanced KBrO3-
induced DNA strand breakage in these cells. DNA damage
by KBrO3 is therefore largely dependent on access to
intracellular GSH.
Materials and methods
Chemicals
All chemicals, unless otherwise stated, were obtained from Sigma-Aldrich
(Poole, UK). The standard sample of 8-oxodG was synthesized by the
Udenfriend hydroxylation system (Kasai and Nishimura, 1984).
Introduction
Potassium bromate (KBrO3) has been used as a food additive
in the treatment of flour and as a constituent of the neutralizer
in cold-wave hair solutions (Norris, 1965; FAO/WHO, 1979).
It is also formed as a by-product of reactions such as the
disinfection of water by ozonation (Fielding and Hutchinson,
1993). The provisional guideline limit value set by the WHO
for KBrO3 in drinking water is 25 µg/l, although a concentration
limit of 10 µg/l has been set in the USA and is indicated in
the 1998 EC Drinking Water Directive. These limits are based
on the practical quantitation limit. KBrO3 has been found to
be mutagenic in bacterial mutation assays (Ishidate et al.,
1984; Kurokawa et al., 1990), in chromosome aberration tests
(Ishidate and Yoshikawa, 1980) and in micronucleus assays
(Hayashi et al., 1988). It also caused renal cell tumours,
mesotheliomas of the peritoneum and follicular cell tumours
of the thyroid in F344 rats (Kurokawa et al., 1986, 1990).
DNA incubations
Calf thymus DNA [0.5 mg/ml in Ca2ϩ-free phosphate-buffered saline, pH 7.4
(PBS)] was incubated in a total volume of 1 ml at 37°C. Hydrogen peroxide
with iron(II) sulphate (FeSO4) (both 250 µM) was used as a positive control.
DNA was precipitated with 4 ml ice-cold ethanol plus 100 µl 0.3 M sodium
acetate and stored at –20°C overnight. The DNA precipitate was spun at
2000 g for 10 min and the ethanol decanted. Finally, the DNA was resuspended
in 100 µl SSC buffer (5 mM sodium citrate, 20 mM sodium chloride, pH 6.5).
DNA extraction from cultured cells
Rat kidney epithelial cells (NRK-52E, obtained from the European Collection
of Cell Cultures, Centre for Applied Microbiology and Research, Salisbury,
UK) were cultured in Dulbecco’s modified Eagle’s medium supplemented
with 2 mM L-glutamine, 10% fetal bovine serum and 1% non-essential amino
acids at 37°C in 5% CO2. DNA was extracted using a Puregene DNA isolation
kit (Flowgen plc, Lichfield, UK) and throughout the extraction procedure the
samples were incubated in the presence of 1% dimethylsulphoxide and gassed
with nitrogen as practically possible to avoid artifactual DNA oxidation. The
medium from cell culture flasks (75 cm2) containing ~1ϫ107 rat kidney
1To whom correspondence should be addressed. Tel: ϩ44 121 414 5422; Fax: ϩ44 121 414 3982; Email: j.k.chipman@bham.ac.uk
© UK Environmental Mutagen Society/Oxford University Press 2000
311

J.L.Parsons and J.K.Chipman
epithelial cells was removed and the cells washed with Ca2ϩ-free PBS. Cell
lysis solution (3 ml) was added to the cells and incubated at 37°C for 15 min,
after which the lysed cells were transferred to a 4.5 ml polypropylene tube.
RNase solution (4 mg/ml, 15 µl) was added to the lysate, the tube inverted
~25 times and incubated for a further 15–60 min at 37°C. The samples were
cooled on ice for 5 min and, after the addition of 1 ml protein precipitation
solution, were vortexed vigorously for 20 s prior to incubation on ice for a
further 5 min. The samples were spun at 2000 g for 10 min at 4°C and the
supernatant transferred to a 15 ml centrifuge tube containing 2 ml sodium
iodide solution (13.5 M NaI, 20 mM Na2EDTA, 40 mM Tris–HCl, pH 8.0).
The tube was inverted several times and the DNA precipitated by the addition
of 4.5 ml ethanol and inverted ~50 times until the DNA had formed a visible
clump. After centrifuging at 2000 g for 3 min, the supernatant was discarded
and the DNA washed with 4.5 ml 70% ethanol. The DNA was centrifuged at
2000 g for 1 min, the supernatant discarded carefully and the pellet air dried
for 10 min. The DNA was resuspended in 100 µl SSC buffer and then
transferred to a clean Eppendorf tube.
of 0.5% normal melting point agarose in PBS, which had already been allowed
to solidify on a fully frosted microscope slide. A third layer of LMPA was
added on top of the cells and, following solidification, the slides were lowered
into fresh lysing solution (2.5 M NaCl, 100 mM EDTA, 10 mM Tris, 1%
lauroyl sarcosinate, 10% dimethylsulphoxide, 1% Triton X-100, pH 10.0) for
1 h at 4°C in the dark. The DNA was allowed to unwind by placing the slides
in 75 mM NaOH and 1 mM EDTA and, following electrophoresis at 25 V,
300 mA for 20 min, the slides were washed with 0.4 M Tris, pH 7.5, drained
and then stained with ethidium bromide (20 µg/ml) before analysis under a
fluorescence microscope (Zeiss Axiovert inverted fluorescence microscope,
ϫ200 magnification, with a 515–560 nm excitation filter and a 590 nm barrier
filter). Fifty cells per slide were automatically analysed using the Komet
Image Analysis System (Kinetic Imaging Limited, Liverpool, UK). The mean
per cent of the DNA in the comet tails was calculated and also the per cent
tail DNA grouped and classified as a modification of that used by Anderson
et al. (1994). Cell viability was determined by measurement of the leakage
of lactate dehydrogenase (Moldeus et al., 1978).
Statistical analyses
DNA hydrolysis and analysis of 8-oxodG by HPLC
Analyses were made by a two-sample t-test, both with and without log
transformation of data from Comet analyses. ANOVA was also used for
thiol changes.
The method used for hydrolysis of DNA was based on that of Adachi et al.
(1995) and throughout the hydrolysis procedure the samples were incubated
in the presence of 1% dimethylsulphoxide and gassed with nitrogen. The
DNA, in SSC buffer, was heated to 95°C for 4 min and, after cooling on ice,
95 µl 20 mM sodium acetate buffer, pH 4.8, containing 0.1 mM ZnCl2 was
added followed by 5 U nuclease P1 and this was then incubated for 30 min
at 45°C. After this, 90 µl 50 mM Tris–HCl, pH 7.4, and 3 U alkaline
phosphatase (Type VII-S) were added and incubation continued for 1 h at
37°C. Samples were centrifuged at 2000 g for 10 min and the supernatants
stored at –70°C until required.
The method for reverse phase HPLC of the DNA hydrosylates was modified
from the procedure of Floyd et al. (1986). The column was a silica C18 Ultra
Techsphere (HPLC Technology, Macclesfield, UK) and was used with a flow
rate of 1.0 ml/min with an RP-18 5 µm guard column. The mobile phase
consisted of 12.5 mM citric acid, 25 mM sodium acetate, 30 mM sodium
hydroxide, 10 mM acetic acid and 1 mM EDTA, pH 5.1, with 5% methanol
(v/v). Deoxyguanosine was detected by UV absorption (254 nm) and 8-oxodG
using a model LC-4B electrochemical detector attached to a temperature
controller (at 37°C). A glassy carbon electrode (potential ϩ0.6 V) was used.
Authentic standards for HPLC were used for quantitation as previously
described (Faux et al., 1992).
Results
KBrO3 alone did not increase the levels of 8-oxodG in calf
thymus DNA at concentrations up to 1500 µM (Table I).
However, there was a dose-dependent increase in 8-oxodG in
the presence of both KBrO3 (0–1500 µM) and GSH (500 µM),
which was highly significant (P Ͻ 0.01) at 1500 µM KBrO3
[0.008 Ϯ 0.001 and 0.091 Ϯ 0.019% ratio of 8-oxodG/dG
(mean Ϯ SD, n ϭ 4) at 0 and 1500 µM KBrO3, respectively].
The positive control (250 µM H2O2 with 250 µM FeSO4) gave
a value of 0.076 Ϯ 0.004% (mean Ϯ SD, n ϭ 2). When GSH
was substituted by N-acetylcysteine (NACys) (500 µM), a
similar concentration-dependent profile was observed (Table
I). The use of other reducing agents (including other thiols)
in combination with KBrO3 (Table I) did not significantly
elevate the levels of 8-oxodG, although there was a significant
increase in the presence of FeSO4 at 1500 µM KBrO3. When
KBrO3 and NACys were reacted for various times prior to
addition of DNA, oxidation of the latter did not occur if
preincubation exceeded 6 min. There was, however, no evid-
ence for a free radical species able to be trapped by DMPO
from high concentrations of KBrO3 and GSH, compared with
reactive oxygen produced by FeSO4 and H2O2 at the same
concentrations used to oxidize DNA (Figure 1).
GSH and oxidized glutathione (GSSG) analyses
The method was based on that of Hissin and Hilf (1976). For both analyses,
conditions of incubation were identical to those used for studying the effects
on calf thymus DNA, except that the latter was omitted. Levels of intracellular
thiols in rat kidney epithelial cells (NRK-52E) were measured as in the GSH
procedure, after lysing the cells by repetitive freeze–thawing in Ca2ϩ-free PBS
GSH. The final assay mixture (2.0 ml) contained 100 µl of the incubates,
1.8 ml phosphate–EDTA buffer, pH 8.0, and 100 µl 1 mg/ml o-phthalaldehyde
(OPT) in methanol. This was left for 15 min at room temperature prior to
reading the fluorescence at 420 nm with excitation at 350 nm and quantification
from a calibration curve constructed using GSH (0–20 µg/ml).
GSSG. To 100 µl of the incubates was added 5 µl 100 mM dithiothreitol,
10% (v/v) Triton X-100. The samples were vortexed and allowed to stand at
room temperature for 30 min. The volume was then made up to 1.9 ml with
phosphate–EDTA buffer, pH 8.0, followed by incubation with 100 µl OPT
for 15 min and the fluorescence read as described above. The GSH value
calculated as above was then subtracted from the total glutathione thus
obtained, resulting in a value for GSSG.
Substitution of KBrO3 with KClO3 or KIO3, with and
without GSH, at the optimal concentration of 1500 µM (Table
II) did not elevate the levels of 8-oxodG. In a complementary
study, the levels of GSH were found to be depleted by a
15 min incubation with KBrO3 or KIO3, but not with KClO3
(Table III), in accord with their relative oxidative potential.
The subsequent formation of GSSG was detected on incubation
of KBrO3 at 1.5 mM with GSH. However, at a higher
concentration of KBrO3 (15 mM), both GSH and GSSG
were lowered.
DNA strand breakage measured by the Comet assay showed
an increase in the per cent tail DNA in human white blood
cells treated with KBrO3 (5 mM) for 15 min compared with
the control (Figure 2). This effect was reduced by pretreatment
with 2 mM diethylmaleate (DEM), a glutathione depletor
(Plummer et al., 1981), for 15 min prior to KBrO3 administra-
tion and was significantly different (P Ͻ 0.002) compared
with treatment with KBrO3 alone. Co-addition of KBrO3 with
extracellular GSH (5 mM) showed a marked reduction in DNA
strand breakage (P Ͻ 0.005 compared with KBrO3 alone),
whereas treatment with KBrO3 and NACys (5 mM) greatly
Electron spin resonance spectroscopy
Spectra were acquired at 298K using a Bruker ESP 300 continuous wave
X-band spectrometer, operating in the TE104 mode (Bruker Analytische
Messtechnik GmbH, Karlsruhe, Germany). The modulation frequency was
100 kHz and modulation amplitude was 4 G. The spin trap agent 5,5Ј-
dimethyl-l-pyrroline N-oxide (DMPO) was used for this study.
Alkaline single cell gel electrophoresis (Comet) assay
Human white blood cells (in 5 µl of blood), freshly obtained, were incubated
in a total volume of 1 ml Ca2ϩ-free PBS at 37°C for 15 min. Rat kidney
epithelial cells were cultured in 25 cm2 tissue culture flasks at 37°C (5%
CO2) and then removed by trypsinization (0.05% trypsin, 0.02% EDTA). The
method of Singh et al. (1988) was then followed, in which the cells were
precipitated by centrifugation (2000 r.p.m., 5 min) and the pellet carefully
mixed with 100 µl of 0.5% low melting point agarose (LMPA) in PBS (rat
kidney epithelial cells after centrifugation were suspended in 100 µl PBS and
10 µl of the cell suspension added to the LMPA). Cells were kept under
subdued yellow light throughout. The cells were then applied on top of 130 µl
312

Glutathione and KBrO3-induced DNA damage
Table I. DNA oxidation by potassium bromate in combination with reducing agents
Ratio 8-oxodG/dG (%)a
KBrO3 alone
GSH (500 µM)
NACys (500 µM) FeSO4 (250 µM) Dithiothreitol
(500 µM)
Ascorbate (500 µM) Mercaptoethanol
(500 µM)
Control
KBrO3 (15 µM)
0.014 Ϯ 0.004
0.009 Ϯ 0.003
0.008 Ϯ 0.001
0.035 Ϯ 0.024
0.049 Ϯ 0.035
0.091 Ϯ 0.033c
0.016 Ϯ 0.003
0.014 Ϯ 0.001
0.030 Ϯ 0.003b
0.076 Ϯ 0.004b
0.049 Ϯ 0.007
0.052 Ϯ 0.008
0.056 Ϯ 0.009
0.102 Ϯ 0.013b
0.031 Ϯ 0.025
0.009 Ϯ 0.002
0.010 Ϯ 0.001
0.011 Ϯ 0.002
0.020 Ϯ 0.001
0.026 Ϯ 0.007
0.019 Ϯ 0.004
0.028 Ϯ 0.004
0.015 Ϯ 0.001
0.013 Ϯ 0.000
0.014 Ϯ 0.003
0.021 Ϯ 0.011
(150 µM) 0.009 Ϯ 0.003
(1500 µM) 0.010 Ϯ 0.003
Calf thymus DNA (0.5 mg/ml) was incubated with KBrO3 for 15 min in the presence of the various reductants shown.
aMean per cent ratio of 8-oxodG to dG Ϯ SD (n ϭ 2–4).
c
b,cStatistically significant compared with the control: bP Ͻ 0.001, P Ͻ 0.01, n ϭ 4.
Table III. GSH measurements (nmol) after incubation with KBrO3, KClO3
and KIO3
KBrO3
KClO3
KIO3
GSH concentration
GSH onlya
432.0 Ϯ 34.2 555.8 Ϯ 85.8 625.1 Ϯ 56.9
GSH ϩ halate (15 µM) 394.9 Ϯ 70.0 671.3 Ϯ 37.1 458.3 Ϯ 45.0
(150 µM) 381.6 Ϯ 44.5 622.4 Ϯ 54.5
34.6 Ϯ 7.7
0.2 Ϯ 0.3
0.4 Ϯ 0.7
(1.5 mM) 59.9 Ϯ 13.6 555.7 Ϯ 85.6
(15 mM)
GSSG concentration
0.4 Ϯ 0.3
684.9 Ϯ 108.1
GSH ϩ halate (15 µM) Ͻ10
n.d.
n.d.
n.d.
(150 µM) Ͻ10
n.d.
n.d.
n.d.
(1.5 mM) 224.8 Ϯ 20.8 n.d.
(15 mM) Ͻ10
n.d.
Mean GSH and GSSG Ϯ SD after incubation (15 min, nmol; n ϭ 4–5).
Values of GSSG are provided after subtraction of the background level seen
in control incubations of GSH alone.
aInitial concentrations of GSH (~500 nmol) are given for each experiment
prior to halate addition.
n.d., not determined.
Fig. 1. Detection of free radicals using electron spin resonance spectroscopy
and the spin trap DMPO. KBrO3 (0.15 M) was incubated with GSH
(0.05 M) and the spin trap DMPO (0.1 M) prior to identification of free
radical presence using ESR (B). This was compared to a positive control
(A) derived from reaction of the spin trap with H2O2 (250 µM) and FeSO4
(250 µM). The background signal seen in both traces is characteristic of the
presence of manganese.
Table II. DNA oxidation induced by KBrO3, KClO3 and KIO3 in the
presence or absence of glutathione
Fig. 2. The influence of GSH and NACys on DNA strand breakage in
human white blood cells by KBrO3 as measured by the Comet assay.
Intracellular GSH was depleted by pretreatment with DEM (2 mM) for 15
min prior to treatment with KBrO3 (5 mM), H2O2 (200 µM), NACys (5
mM) or extracellular GSH (5 mM) for 15 min. Results are expressed as
mean per cent DNA in the comet tail Ϯ SD after subtraction of the control
value (11.60 Ϯ 2.30%; n ϭ 4–18, 50 cells analysed per experiment). *P Ͻ
0.005, **P Ͻ 0.001 compared with control (untreated). †P Ͻ 0.005, ††P Ͻ
0.002 compared with treatment with KBrO3 alone.
Ratio 8-oxodG/dG (%)a
Without GSH
With GSH (500 µM)
Control
KBrO3
KClO3
KIO3
0.012 Ϯ 0.001
0.010 Ϯ 0.003
0.011 Ϯ 0.005
0.010 Ϯ 0.002
0.008 Ϯ 0.001
0.080 Ϯ 0.031b
0.010 Ϯ 0.001
0.012 Ϯ 0.003
Calf thymus DNA (0.5 mg/ml) was incubated for 15 min in the presence of
the various halates (1.5 mM) shown.
enhanced the per cent tail DNA (P Ͻ 0.005 compared with
KBrO3 alone).
A study of the time course of DNA damage in rat kidney
aMean per cent ratio of 8-oxodG to dG Ϯ SD (n ϭ 2–6).
bStatistically significant compared with the control: P Ͻ 0.002, n ϭ 6.
313

J.L.Parsons and J.K.Chipman
Fig. 3. Time course profile of DNA strand breakage as measured by the
Comet assay, after exposure of rat kidney epithelial cells (NRK-52E) to
KBrO3 and H2O2. Cells were harvested at day 4 after seeding (for all
incubations including controls), with or without treatment with KBrO3
(1.5 mM) or H2O2 (200 µM) for the times indicated during this period. The
per cent tail DNA was arranged in ascending order and then classified
according to Anderson et al. (1994) with modifications (minimal, Ͻ20%;
medium, 20–40%; high, Ͼ40%). Values are given as a percentage of total
cells (n ϭ 3–6, 50 cells per experiment). Mean per cent tail DNA ϭ
15.27 Ϯ 0.75, 30.16 Ϯ 1.93 and 35.61 Ϯ 1.40 for control and KBrO3 (15
min and 24 h), respectively. P Ͻ 0.001.
Fig. 5. Levels of intracellular thiols after incubation of rat kidney epithelial
cells (NRK-52E) with KBrO3. The cells were harvested at day 4 after
seeding (for all incubations including controls), with or without treatment
with KBrO3 (1.5 mM) for the times indicated during this period, and were
then analysed for thiols by the method of Hissin and Hilf (1976). Results
are expressed as mean thiol (nmol/mg protein) Ϯ SEM (n ϭ 4–5
experiments, 2–3 samples averaged/experiment). *No significant differences
from control by ANOVA but P Ͻ 0.05 by t-test.
Fig. 6. Viability of rat kidney epithelial cells (NRK-52E) treated with
KBrO3 (1.5 mM). Cells were analysed for the release of lactate
dehydrogenase activity into the medium on day 4 after seeding with a
pretreatment time as indicated. Values are means Ϯ SD (n ϭ 8–12). No loss
of viability was indicated.
Fig. 4. Formation of 8-oxodG in rat kidney epithelial cells (NRK-52E) by
KBrO3. Cells were harvested at day 4 after seeding (for all incubations
including controls), with or without treatment with KBrO3 (1.5 mM) for the
times indicated during this period. DNA was extracted using a Puregene kit
in association with chaotropic sodium iodide. Following digestion with
nuclease P1 and alkaline phosphatase, the DNA was analysed by HPLC
with electrochemical detection. Values are given as per cent ratio of
8-oxodG/deoxyguanosine Ϯ SD (n ϭ 4–5). *P Ͻ 0.001.
washing was also investigated (data not shown). The level of
DNA strand breaks was found to return to control levels
between 4 and 24 h post-exposure and cell wash. The use of
H2O2 as a positive control gave a similar profile. KBrO3 was
not cytotoxic at 1.5 mM as measured by lactate dehydrogenase
leakage at all time points investigated (Figure 6). In preliminary
studies (data not shown) we also found no evidence for
elevation of apoptosis based on Hoechst 33258 nuclear staining
and microscopy.
epithelial cells (Figure 3) revealed that DNA strand breakage
by KBrO3 (1.5 mM) peaked at 15 min, with a decrease thereon
up to 4 h. However, at 24 h DNA strand breakage was again
enhanced, in contrast to H2O2, which gave a greater response
at 15 min than at 24 h. There was a small but not significant
elevation in 8-oxodG (0.0034 Ϯ 0.0007% ratio of 8-oxodG/
dG; mean Ϯ SD, n ϭ 5) compared with the control (0.0028
Ϯ 0.0003%; n ϭ 4) after 15 min incubation of rat kidney
epithelial cells (NRK-52E) with KBrO3 (1.5mM). However, a
statistically significant elevation of 8-oxodG (0.0102 Ϯ
0.0005%; n ϭ 5) was observed after 24 h treatment with
KBrO3 (Figure 4). The DNA strand breakage seen at 15 min
was associated with an apparent small lowering of intracellular
thiols (Figure 5). The persistence of DNA strand breakage
after a 4 h treatment of rat kidney cells and subsequent cell
Discussion
Studies herein aimed to expand upon initial observations
(Ballmaier and Epe, 1995; Chipman et al., 1998), which
suggested that GSH plays a role in DNA damage by KBrO3
in vitro. KBrO3 was found to produce 8-oxodG in isolated
calf thymus DNA in the presence of either GSH or NACys.
This suggests the formation of similar reactive species, which
are equipotent, derived from both thiols. The lifetime of the
DNA-reactive species was found to be Ͻ6 min, in agreement
with that estimated by Ballmaier and Epe (1995). Other
314

Glutathione and KBrO3-induced DNA damage
reductants (including thiols) in combination with KBrO3 were
not effective in causing DNA oxidation, although an increase
was observed at high concentrations of KBrO3 in the presence
of FeSO4, suggesting the possibility of a Fenton-type reaction
under these conditions. This, however, needs to be clarified.
The apparent importance of a thiol interaction with KBrO3
has also been suggested to play a role in the activity of KBrO3
as a flour ‘improver’ in bread making (Dupuis, 1997). However,
a specificity of GSH and NACys thiols was suggested in
relation to DNA damage by the observation that the thiols
dithiothreitol and β-mercaptoethanol were not effective in
catalysing 8-oxodG formation. The coincident formation of
GSSG suggests that GSH is oxidized in formation of the
reactive product but, at high concentrations of KBrO3, GSSG
is also degraded. The identity of the reactive species is not
reactants to the cell. The fact that, in contrast, addition of
NACys resulted in an increase in DNA strand breakage
suggests that either KBrO3 reacts intracellularly with this
membrane-permeable thiol (Traber et al., 1992) or that reactants
formed extracellularly by NACys have access to intracellular
DNA.
There is, therefore, a predicted dual role for GSH in the
genotoxicity of KBrO3. Our results in mammalian cells and
in isolated DNA suggest an activating role, with a protective
effect of extracellular GSH. There is also an apparent protective
effect in vivo, as evidenced by enhanced toxicity with GSH
depletion (Sai et al., 1992). Thus, activation of KBrO3 by
GSH may occur in vivo largely at a site distant from the
kidney, which is a major target organ. A further anomaly is
that significant increases in 8-oxodG in vivo were only observed
24 h after oral (Kasai et al., 1987) or i.p. (Sai et al., 1991;
Cho et al., 1993) administration. This is not consistent with a
rapid interaction with thiols and the production of a short-
lived intermediate. Since there is also a coincident formation
of lipid peroxides at high doses of KBrO3 (Sai et al., 1991,
1992; Chipman et al., 1998), a potential secondary mechanism
of DNA oxidation by lipid peroxides (Hruszkewycz and
Bertold, 1988) is therefore a possibility in vivo. Previously,
we observed no increase in 8-oxodG in kidney tissue up to
4 h following infusion with KBrO3 at 5 mM and, in the present
study, an increase in 8-oxodG in cells treated with KBrO3
occurred at 24 h but not at 15 min. 8-Oxodeoxyguanosine was
reported to be elevated in V79 Chinese hamster cells within
1 h, but only at 100 mM KBrO3, which was also reported to
be cytotoxic (Speit et al., 1999). The concentrations able to
produce strand breaks in our study (1.5 mM) and in the study
by Speit (1 mM) are, in contrast, not cytotoxic in either cell
line. In this context, it is interesting that chlorate also produced
strand breakage in kidney cells but did not produce 8-oxodG
in calf thymus DNA when incubated with GSH. Taken together,
these findings suggest a lack of correlation between strand
breakage by halates and 8-oxodG formation.
`
yet known. Lafleur and Retel (1993) have reviewed data which
indicates that although thiols are effective antioxidants, they
also possess potential genotoxic consequences, such as through
the formation of thiyl peroxy radicals and other reactants from
interaction with singlet oxygen. GSH appears also to form
superoxide radical (O2·–) in the presence of iron salt (Rowley
and Halliwell, 1982). However, the relationship between these
reactions and the synergistic effect of GSH and KBrO3 is not
known. Although Ballmaier and Epe (1995) suggest various
potential free radical products, we found no evidence for a
radical species using the trapping agent DMPO. Using repair
endonucleases, Ballmaier and Epe (1995) found that the DNA
damage profile produced by KBrO3 in the presence of GSH
was not characteristic of hydroxyl, superoxide or thiyl radicals.
It was assumed that bromine radicals (Br·) or oxides (BrO·
and BrO2·) were likely to be the species responsible for the
cellular and cell-free DNA damage observed. Mutagenicity
tests (Kurokawa et al., 1990) using Salmonella typhimurium
TA102 and TA104 (strains sensitive to oxygen radicals; Levin
et al., 1982) suggested that KBrO3 was only mutagenic in the
presence of metabolic activation (S9). We now propose that
the S9 fraction provides a source of thiols which react
with KBrO3.
Although 8-oxodG formation in rat kidney in vivo is
associated with high bolus doses of KBrO3 and lipid peroxida-
tion, it has also been detected in male rats after 7 days of
inclusion of KBrO3 in the drinking water at 500 p.p.m.
(Umemura et al., 1998). We have also measured an increase
in 8-oxodG in kidney following exposure at this concentration
in the absence of lipid peroxidation (J.E.Davies, J.L.Parsons
and J.K.Chipman, unpublished results). Umemura et al. (1998)
showed that the elevation in 8-oxodG was concurrent with
kidney cell proliferation and proposed that the elevation was
a consequence of this proliferative activity. This was based on
the study of Adachi et al. (1994), which gave evidence
for increased susceptibility to oxidative DNA damage in
regenerating liver. It may also be that repair of 8-oxodG is
impaired in dividing cells. It can therefore be proposed that
the elevation of 8-oxodG by KBrO3 will occur at dose levels
able to cause kidney cell proliferation (e.g. 500 p.p.m. in the
drinking water) or with high bolus doses able to cause
lipid peroxidation. The mechanism responsible for 8-oxodG
elevation may differ for each of these two conditions. It is
still not known if 8-oxodG plays a causative role in KBrO3-
induced kidney tumours. However, if it does, then a non-linear
dose–response relationship is likely based on the discussion
above.
No effect on 8-oxodG was observed with KClO3 or KIO3
in the presence or absence of GSH. GSH was depleted by
KBrO3 and KIO3, but not by KClO3. Therefore the damage to
DNA appears to be more specific for KBrO3 and GSH and is
not a feature of other halates. Since we have found all three
halates to produce DNA strand breaks in NRK-52E cells
(Parsons and Chipman, 1999a), there may be an additional,
alternative mechanism of DNA damage in cells.
KBrO3 induced DNA strand breakage in both human white
blood cells and a rat kidney epithelial cell line (NRK-52E).
Strand breakage in these cells by KBrO3 was induced within
15 min, indicating a rapid response. The damage then declined
up to 4 h, although a further elevation of DNA strand breakage
occurred at 24 h. This was associated with an increase in
8-oxodG at this time point (but not at 15 min). We have also
noted the formation of lipid peroxides at 24 h (Parsons and
Chipman, 1999b) and are currently investigating the possibility
that the delayed DNA damage may be mediated by lipid
peroxidation products. A role for GSH in the production of
strand breaks was also supported by the observation that
depletion of intracellular GSH with DEM in human white
blood cells reduced the amount of strand breakage. In contrast,
the addition of extracellular GSH with KBrO3 also reduced
the amount of strand breakage. This protective influence of
extracellular GSH may relate to restricted access of the
DNA damage in vivo and in vitro through lipid peroxides
is currently being investigated further by observing etheno- and
315

J.L.Parsons and J.K.Chipman
bromate in F344 rats after long-term oral administration. J. Natl Cancer
Inst., 77, 977–982.
Kurokawa,Y., Maekawa,A., Takahashi,M. and Hayashi,Y. (1990) Toxicity and
carcinogenicity of potassium bromate—a new renal carcinogen. Environ.
Health Perspect., 87, 309–335.
Lafleur,M.V.M. and Rete`l,J. (1993) Contrasting effects of SH-compounds on
oxidative DNA damage: repair and increase of damage. Mutat. Res., 295,
1–10.
malondialdehyde-adduct formation to determine the relative
importance of these products versus a direct thiol-mediated
intermediate in DNA damage. Resolution of the mechanism
of DNA damage in vivo is important in allowing a better
understanding of the dose–response relationship in KBrO3
carcinogenesis and the target organ specificity.
Levin,D.E., Hollstein,M.C., Christman,M.F., Schwiers,E.A. and Ames,B.N.
(1982) A new Salmonella tester strain (TA102) with A:T base pairs at the
site of mutation detects oxidative mutagens. Proc. Natl Acad. Sci. USA, 79,
7445–7449.
Moldeus,P., Hogberg,J. and Orrenius,S. (1978) Isolation and use of liver cells.
Methods Enzymol., 52, 60–71.
Acknowledgements
This work was partially supported by the UKWIR and we thank W.Matthews,
R.Mitchell, and J.Fawell of WRc (Medmenham, UK) for useful discussions.
We also gratefully acknowledge the help of T.L.Green and P.A.Anderson from
the School of Chemistry, The University of Birmingham, in the ESR study.
Norris,J.A. (1965) Toxicity of home permanent waving and neutralizer
solutions. Food Cosmet. Toxicol., 3, 93–97.
Parsons,J.L. and Chipman,J.K. (1999a) Comparison of DNA damage induced
by potassium chlorate (KClO3), potassium bromate (KBrO3) and potassium
iodate (KIO3) and their interaction with glutathione (GSH). Toxicol. Lett.,
109 (suppl. 1), 48.
Parsons,J.L. and Chipman,J.K. (1999b) Time dependency of potassium bromate
induced DNA strand breakage in vitro. Hum. Exp. Toxicol., 18, 535.
Plummer,J.L., Smith,B.R., Sies,H. and Bend,J.R. (1981) Chemical depletion
of glutathione in vivo. Methods Enzymol., 77, 50–63.
References
Adachi,S., Kawamura,K. and Takemoto,K. (1994) Increased susceptibility to
oxidative DNA damage in regenerating liver. Carcinogenesis, 15, 539–543.
Adachi,S., Zeisig,M. and Moller,L. (1995) Improvements in the analytical
method for 8-OH-deoxyguanosine in nuclear DNA. Carcinogenesis, 16,
253–258.
Anderson,D., Yu,T.-W., Phillips,B.J. and Schmezer,P. (1994) The effect of
various antioxidants and other modifying agents on oxygen-radical-
generated DNA damage in human lymphocytes in the COMET assay.
Mutat. Res., 307, 261–271.
Rowley,D.A. and Halliwell,B. (1982) Superoxide-dependent formation of
hydroxyl radicals in the presence of thiol compounds. FEBS Lett., 138,
33–36.
Ballmaier,D. and Epe,B. (1995) Oxidative damage induced by potassium
bromate under cell free conditions and in mammalian cells. Carcinogenesis,
16, 335–342.
Chipman,J.K., Davies,J.E., Parsons,J.L., Nair,J., O’Neill,G. and Fawell,J.K.
(1998) DNA oxidation by potassium bromate; a direct mechanism or linked
to lipid peroxidation? Toxicology, 126, 93–102.
Cho,D.H., Hong,J.T., Chin,K., Cho,T.S. and Lee,B.M. (1993) Organotropic
formation and disappearance of 8-hydroxydeoxyguanosine in the kidney of
Sprague-Dawley rats exposed to adriamycin and KBrO3. Cancer Lett., 74,
141–145.
DeAngelo,A.B., George,M.H., Kilburn,S.R., Moore,T.M. and Wolf,D.C.
(1998) Carcinogenicity of potassium bromate in the drinking water to male
B6C3F1 mice and F344/N rats. Toxicol. Pathol., 26, 587–594.
Dupuis,B. (1997) The chemistry and toxicology of potassium bromate. Cereal
Foods World, 42, 171–183.
FAO/WHO (1979) Food and Agriculture Organisation/World Health
Organisation. Guide to the Safe Use of Food Additives, 2nd Series. World
Health Organisation, Geneva, Switzerland, p. 60.
Faux,S.P., Francis,J.E., Smith,A.G. and Chipman,J.K. (1992) Induction of
8-hydroxydeoxyguanosine in Ah-responsive mouse liver by iron and Aroclor
1254. Carcinogenesis, 13, 247–250.
Sai,K., Takagi,A., Umemura,T., Hasegawa,R. and Kurokawa,Y. (1991)
Relation of 8-hydroxydeoxyguanosine formation in rat kidney to lipid
peroxidation, glutathione level and relative organ weight after a single
administration of potassium bromate. Jpn J. Cancer Res., 82, 165–169.
Sai,K., Umemura,T., Takagi,A., Hasegawa,R. and Kurokawa,Y. (1992) The
protective role of glutathione, cysteine and vitamin C against oxidative
DNA damage induced in rat kidney by potassium bromate. Jpn J. Cancer
Res., 83, 45–51.
Singh,N.P., McCoy,M.T., Tice,R.R. and Schneider,E.L. (1988) A simple
technique for quantitations of low levels of DNA damage in individual
cells. Exp. Cell Res., 175, 184–191.
Speit,S., Haupter,S., Schutz,P. and Kreis,P. (1999) Comparative evaluation of
¨
the genotoxic properties of potassium bromate and potassium superoxide
in V79 Chinese hamster cells. Mutat. Res., 439, 213–221.
Traber,J., Suter,M., Walter,P. and Richter,C. (1992) In vivo modulation of total
and mitochondrial glutathione in rat liver; depletion by phorane and rescue
by N-acetylcysteine. Biochem. Pharmacol., 43, 961–964.
Umemura,T., Takagi,A., Sai,K., Hasegawa,R. and Kurokawa,Y. (1998)
Oxidative DNA damage and cell proliferation in kidneys of male and
female rats during 13-weeks exposure to potassium bromate. Arch. Toxicol.,
72, 264–269.
Fielding,M. and Hutchinson,J. (1993) Formation of bromate and other
ozonation by-products in water treatment. International Workshop, Bromate
and Water Treatment, AIDE-IWSA, Paris, France.
Received on September 16, 1999; accepted on February 25, 2000
Floyd,R.A., Watson,J.J., Wong,P.K., Altmiller,D.H. and Rickard,R.C. (1986)
Hydroxyl free radical adduct of deoxyguanosine: sensitive detection and
mechanisms of formation. Free Radic. Res. Commun., 1, 163–172.
Hayashi,M., Kishi,M., Sofuni,T. and Ishidate,M.Jr (1988) Micronucleus tests
with mice on 39 food additives and 8 miscellaneous chemical substances.
Food Chem. Toxicol., 26, 487–500.
Hissin,P.J. and Hilf,R. (1976) A fluorometric method for determination of
oxidised and reduced glutathione in tissues. Anal. Biochem., 74, 214–225.
Hruszkewycz,A.M. and Bergtold,D.S. (1988) Oxygen radicals, lipid
peroxidation and DNA damage in mitochondria. Basic Life Sci., 49, 449–456.
Ishidate,M.Jr and Yoshikawa,K. (1980) Chromosome aberration test with
Chinese hamster cells in vitro with and without metabolic activation—a
comparative study on mutagens and carcinogens. Arch. Toxicol., 4 (suppl.),
41–44.
Ishidate,M.Jr, Sofuni,T., Yoshikawa,K., Hayashi,M., Nohmi,T., Sawada,M.
and Matsuoka,A. (1984) Primary mutagenicity screening of food additives
currently used in Japan. Food Chem. Toxicol., 22, 623–636.
Kasai,H. and Nishimura,S. (1984) Hydroxylation of deoxyguanosine at the
C-8 position by ascorbic acid and other reducing agents. Nucleic Acids
Res., 12, 2137–2145.
Kasai,H., Nishimura,S., Kurokawa,Y. and Hayashi,Y. (1987) Oral
administration of rat renal carcinogen, potassium bromate, specifically
produces 8-hydroxydeoxyguanosine in rat target organ DNA.
Carcinogenesis, 8, 1959–1961.
Kurokawa,Y., Aoki,S., Matsushima,Y., Takamura,N., Imazawa,T. and
Hayashi,Y. (1986) Dose-response studies on the carcinogenicity of potassium
316