Toxicity of a trivalent organic arsenic compound, dimethylarsinous glutathione in a rat liver cell line (TRL 1215)

Article abstract of DOI:10.1038/sj.bjp.0706899

Background and purpose:Although inorganic arsenite (As^III) is toxic in humans, it has recently emerged as an effective chemotherapeutic agent for acute promyelocytic leukemia (APL). In humans and most animals, As ^III is enzymatically methylated in the liver to weakly toxic dimethylarsinic acid (DMAs^V) that is a major pentavalent methylarsenic metabolite. Recent reports have indicated that trivalent methylarsenicals are produced through methylation of As^III and participate in arsenic poisoning. Trivalent methylarsenicals may be generated as arsenical-glutathione conjugates, such as dimethylarsinous glutathione (DMAs^IIIG), during the methylation process. However, less information is available on the cytotoxicity of DMAs^IIIG.Experimental approach:We synthesized and purified DMAs^IIIG using high performance TLC (HPTLC) methods and measured its cytotoxicity in rat liver cell line (TRL 1215 cells).Key results:DMAs^IIIG was highly cytotoxic in TRL 1215 cells with a LC₅₀ of 160 nM. We also found that DMAs^IIIG molecule itself was not transported efficiently into the cells and was not cytotoxic; however it readily became strongly cytotoxic by dissociating into trivalent dimethylarsenicals and glutathione (GSH). The addition of GSH in micromolar physiological concentrations prevented the breakdown of DMAs ^IIIG, and the DMAs^IIIG-induced cytotoxicity. Physiological concentrations of normal human serum (HS), human serum albumin (HSA), and human red blood cells (HRBC) also reduced both the cytotoxicity and cellular arsenic uptake induced by exposure to DMAs^IIIG.Conclusions and implications:These findings suggest that the significant cytotoxicity induced by DMAs^IIIG may not be seen in healthy humans, even if DMAs ^IIIG is formed in the body from As^III.

Full text of DOI:10.1038/sj.bjp.0706899

British Journal of Pharmacology (2006) 149, 888–897
2006 Nature Publishing Group All rights reserved 0007–1188/06 30.00
&
$
www.brjpharmacol.org
RESEARCH PAPER
Toxicity of a trivalent organic arsenic compound,
dimethylarsinous glutathione in a rat liver cell line
(TRL 1215)
T Sakurai1, C Kojima1, Y Kobayashi2, S Hirano2, MH Sakurai3, MP Waalkes4 and S Himeno1
1Laboratory of Molecular Nutrition and Toxicology, Faculty of Pharmaceutical Sciences, Tokushima Bunri University, Yamashiro-cho,
Tokushima, Japan; 2Environmental Health Sciences Division, National Institute for Environmental Studies, Tsukuba, Ibaraki, Japan;
3Laboratory of Environmental Hygiene, Department of Environmental Health, Azabu University, Sagamihara, Kanagawa, Japan and
4Laboratory of Comparative Carcinogenesis, Inorganic Carcinogenesis Section, National Cancer Institute at National Institute of
Environmental Health Sciences, National Institute of Health, Research Triangle Park, NC, USA
Background and purpose: Although inorganic arsenite (AsIII) is toxic in humans, it has recently emerged as an effective
chemotherapeutic agent for acute promyelocytic leukemia (APL). In humans and most animals, AsIII is enzymatically
methylated in the liver to weakly toxic dimethylarsinic acid (DMAsV) that is a major pentavalent methylarsenic metabolite.
Recent reports have indicated that trivalent methylarsenicals are produced through methylation of AsIII and participate in
arsenic poisoning. Trivalent methylarsenicals may be generated as arsenical–glutathione conjugates, such as dimethylarsinous
glutathione (DMAsIIIG), during the methylation process. However, less information is available on the cytotoxicity of DMAsIIIG.
Experimental approach: We synthesized and purified DMAsIIIG using high performance TLC (HPTLC) methods and measured
its cytotoxicity in rat liver cell line (TRL 1215 cells).
Key results: DMAsIIIG was highly cytotoxic in TRL 1215 cells with a LC50 of 160 nM. We also found that DMAsIIIG molecule
itself was not transported efficiently into the cells and was not cytotoxic; however it readily became strongly cytotoxic by
dissociating into trivalent dimethylarsenicals and glutathione (GSH). The addition of GSH in micromolar physiological
concentrations prevented the breakdown of DMAsIIIG, and the DMAsIIIG-induced cytotoxicity. Physiological concentrations
of normal human serum (HS), human serum albumin (HSA), and human red blood cells (HRBC) also reduced both the
cytotoxicity and cellular arsenic uptake induced by exposure to DMAsIIIG.
Conclusions and implications: These findings suggest that the significant cytotoxicity induced by DMAsIIIG may not be seen
in healthy humans, even if DMAsIIIG is formed in the body from AsIII.
British Journal of Pharmacology (2006) 149, 888–897. doi:10.1038/sj.bjp.0706899; published online 16 October 2006
Keywords: arsenic; arsenite; dimethylarsenic; dimethylarsinous glutathione; dimethylarsinic acid; trivalent methylarsenic;
trivalent dimethylarsenic; GSH; methylation; acute promyelocytic leukemia
Abbreviations: AAS, atomic absorption spectrophotometry; APL, acute promyelocytic leukemia; AsIII, inorganic arsenite; AsV,
inorganic arsenate; DMAsIII þ , trivalent dimethylarsenite ion; DMAsV, dimethylarsinic acid; DMAsIIIG,
dimethylarsinous glutathione; DMAsIIII, iododimethylarsine; DMAsIIIOH, dimethylarsinous acid; FAB, fast atom
bombardment; FBS, fetal bovine serum; GC, gas chromatography; GSH, glutathione; GSSG, oxidized form of
glutathione; HPTLC, high-performance TLC; HRBC, human red blood cells; HS, human serum; HSA, human
serum albumin; ICP, inductively coupled argon plasma; LC50, in vitro lethal concentration in 50% of a population;
MMAsV, monomethylarsonic acid; MS, mass spectrometry; NAC, N-acetyl-L-cysteine; Rf, relative mobility
Introduction
Arsenic is a metalloid that is widely distributed in the
environment in inorganic trivalent (inorganic arsenite; AsIII
or pentavalent (inorganic arsenate; AsV) forms (Morton and
Dunnette, 1994), and its high toxicity is well known.
Epidemiological studies have provided clear evidence that
inorganic arsenicals are human carcinogens (NRC, 1999);
however, AsIII has recently emerged as an outstanding
chemotherapeutic agent with remarkable efficacy for certain
human cancers such as acute promyelocytic leukemia (APL)
(Chen et al., 1997; Shen et al., 1997). In humans and
numerous experimental animals, AsIII is enzymatically
methylated in the liver to organic arsenicals such as
)
Correspondence: Dr
T Sakurai, Laboratory of Molecular Nutrition and
Toxicology, Faculty of Pharmaceutical Sciences, Tokushima Bunri University,
Yamashiro-cho, Tokushima 770-8514, Japan.
E-mail: teruaki@ph.bunri-u.ac.jp
Received 12 May 2006; revised 30 June 2006; accepted 15 August 2006;
published online 16 October 2006

Cytotoxicity of dimethylarsinous glutathione
T Sakurai et al
889
monomethylarsonic acid (MMAsV) and dimethylarsinic acid
(DMAsV) (Yamauchi and Yamamura, 1979; Buchet et al.,
1980). MMAsV and DMAsV are the major organic pentavalent
arsenic metabolites found in human urine after exposure to
inorganic arsenicals. It is believed that the methylation of
inorganic arsenicals results in a reduction in general toxicity,
as indicated by increases in the values for LD50 and the
that included 50 mM Tris-HCl (pH 7.5), 10 mM potassium
chloride, 5 mM magnesium chloride, 1 mM dithiothreitol and
1 mM phenylmethanesulfonyl fluoride, and sonicated on ice
using MICROSON ultrasonic homogenizer (Model XL2007,
Misonix Inc., Farmingdale, NY, USA). The degree of cell
rupture was checked by phase-contrast microscopy and 95%
cell rupture was sufficient. The suspension of ruptured cells
was centrifuged at 2000 g for 10 min, and divided into the
supernatant (cytoplasm) and the pellet (cell membrane and
nucleus). The pellet was suspended again in 1 ml of ice-cold
0.25 M sucrose solution, placed on ice-cold 1 M sucrose and
centrifuged at 1500 g for 10 min. The cell membrane at the
0.25 M sucrose/1 M sucrose interface was collected with a
pipette, rinsed with 0.25 M sucrose solution and resuspended
in 1 ml of distilled water. Three milliliters of nitric acid and
1 ml of sulfuric acid were then added to the solutions
of cytoplasm or cell membrane, heated at 2401C until sulfur
trioxide was visible, and the digested solutions were
neutralized with ammonium hydroxide.
in vitro lethal concentration in 50% of a population (LC50
)
(Sakurai et al., 1998, 2002). However, recent studies have
increasingly suggested that the methylation of inorganic
arsenicals is not a universal detoxification metabolic reac-
tion. Wang et al. (2004) reported that trivalent dimethyl-
arsenicals were found in the urine collected from APL patients
undergoing AsIII treatment. The synthetic trivalent dimethyl-
arsenicals, such as iododimethylarsine (DMAsIIII), were
more cytotoxic in vitro than inorganic arsenicals and
pentavalent methylarsenicals (Styblo et al., 2000; Mass
et al., 2001). Recent evidence has suggested that trivalent
methylarsenicals may be generated as arsenical–glutathione
(GSH) conjugates, such as dimethylarsinous glutathione
(DMAsIIIG), in the human body (Hayakawa et al., 2005);
however, due to its instability, little information is available
on the cytotoxicity of DMAsIIIG (Styblo et al., 2000;
Hayakawa et al., 2005).
AAS. The aqueous solutions containing arsenicals prepared
from HPTLC samples or TRL 1215 cells exposed to arsenicals
were made in a volume of 8 ml with distilled water, and
1 ml of hydrochloric acid, 0.5 ml of 20% ascorbic acid
and 0.5 ml of 20% potassium iodide were added to the
solutions. Arsenic in the solutions was analyzed by hydride
generation coupled with AAS (Ohta et al., 2004; Sakurai
et al., 2005) using SpestraAA-220 (Varian Australia Pty Ltd,
Mulgrave, Victoria, Australia). The cellular arsenic content
was expressed as nanograms of arsenicals per milligram
of cellular protein, as determined by BCA protein assay
(Pierce Co., Rockford, IL, USA), or expressed as nM calcu-
lated using total cell numbers and estimated cell density
In this study, we synthesized and purified DMAsIIIG using
a high-performance TLC (HPTLC) method and observed the
cytotoxicity of the synthesized DMAsIIIG using rat liver cells.
The liver is a major site of arsenic methylation. We found
that DMAsIIIG itself was not transported efficiently into the
cells and was not cytotoxic; however, it readily became
highly cytotoxic in
a neutral solution by a chemical
conversion to the trivalent dimethylarsenite ion (DMAsIII þ
)
and/or dimethylarsinous acid (DMAsIIIOH). The addition of
GSH in micromolar physiological concentrations inhibited
this chemical change and prevented the cytotoxicity of
trivalent dimethylarsenicals.
(2.4 Â 10À9 cm3/cell) as measured using
a
hemacyto-
meter counting chamber (Mizoguchi and Hara, 1996;
Sakurai et al., 2005).
HPLC-ICP MS. Arsenicals extracted from TLC were also
analyzed by HPLC-ICP MS and FAB-MS. For HPLC-ICP MS
analysis (Shraim et al., 2001; Hayakawa et al., 2005; Sakurai
et al., 2005), 10 ml of a sample solution was applied to a
reversed-phase C18 column (Inertsil ODS; GL Sciences Inc.,
Tokyo, Japan) connected to an L-6200 Inteligent pump
(Hitachi Ltd, Ibaraki, Japan) with a mobile phase of 5 mM
tetrabutylammonium hydroxide, 3 mM malonic acid and 5%
methanol at flow rate of 1 ml minÀ1. The outlet of the HPLC
system was coupled directly to the inlet of the ICP MS (ELAN
6000, Perkin-Elmer Co., Norwalk, CT, USA), and signals at
m/z 75 and 77 (corresponding to arsenicals and ArCl,
respectively) were monitored.
Methods
Cell culture
The TRL 1215 cell line is a rat epithelial liver cell line
originally derived from the liver of 10-day-old Fisher F344
rats (Idoine et al., 1976). TRL 1215 cells were cultured in
William’s E medium (Sigma) supplemented with 10% heat-
inactivated fetal bovine serum (FBS), 2 mM glutamine,
100 U mlÀ1 penicillin G and 100 mg mlÀ1 streptomycin under
a humidified atmosphere of 5% CO2/95% air at 371C.
Arsenic analysis
Preparation of arsenic samples. Arsenicals in samples from
HPTLC or cellular arsenicals in TRL 1215 cells exposed to
arsenicals were analyzed by atomic absorption spectro-
photometry (AAS), HPLC-inductively coupled argon plasma
(ICP) MS (HPLC-ICP MS) or fast atom bombardment MS
(FAB-MS). TRL 1215 cells which were grown in flat-bottomed
75-cm2 tissue culture flask to confluence (8 Â 107 cells/flask)
and incubated with arsenicals were isolated by trypsinization
and resuspended in 1 ml of ice-cold 0.25 M sucrose solution
Assay for cytotoxicity
TRL 1215 cells were plated on flat-bottomed 96-well tissue
culture plates (1 Â 104 cells per 200 ml per well) and allowed
to adhere to the plate for 24 h at 371C, at which time the
medium was removed and replaced with fresh medium
containing the various test compounds, including arsenicals.
Cells were then incubated with test compounds for an
British Journal of Pharmacology (2006) 149 888–897

Cytotoxicity of dimethylarsinous glutathione
890
T Sakurai et al
additional 1–48 h at 371C. After incubation, cells were
washed twice with warmed phosphate-buffered saline (pH
7.4) to remove nonadherent dead cells, and cell viability was
determined by the AlamarBlue assay. The AlamarBlue assay is
similar to 3-(4,5-dimethylthiazoyl-2-yl)-2,5-diphenyl tetra-
zolium bromide (MTT assay and measures the metabolic
integrity of cells (Ohta et al., 2004; Sakurai et al., 2005).
Briefly, after incubations with test samples and replacement
with 200 ml per well fresh media, 20 ml per well AlamarBlue
solution (Iwaki Grass Co., Chiba, Japan) was added directly to the
96-well plates, incubated for 4 h at 371C, and the absorbance at
570nm (referenced to 630nm) was measured by a microplate
reader model 550 (Bio-Rado Laboratories, Hercules, CA, USA).
Data are expressed as percent metabolic integrity using the values
from control cells as 100%. Similar results were obtained when
cytotoxicity of arsenicals was determined by the counting of
viable cells under microscopy (data not shown).
Results
Preparation of DMAsIIIG by HPTLC
It has been reported that GSH nonenzymatically reduces
DMAsV to DMAsIIIG in water (Scott et al., 1993; Kala et al.,
2000). In order to determine the likelihood of DMAsIIIG
production, 10 mM DMAsV was incubated with or without
10–50 mM GSH in distilled water for 1 h at 371C. After
incubation, these mixtures were applied to an HPTLC plate
and developed using ethyl acetate:acetic acid:water (3:2:1).
Separated compounds were detected with iodine vapor. As
shown in Figure 1a, GSH (lane 1, relative mobility (Rf) ¼ 0.33)
and GSSG (lane 2, Rf ¼ 0.06) spots were detected with iodine
vapor; however, DMAsV was not detected under these
experimental conditions (lane 3). A spot of the putative
DMAs-GSH conjugate DMAsIIIG was detected with iodine
vapor at a different position from the GSH and GSSG spots
(Figure 1a, lanes 4–12, Rf ¼ 0.49) after incubating DMAsV
with GSH. Moreover, arsenical was only detected from
this putative DMAsIIIG spot on the HPTLC plate as
determined by AAS (Figure 1b). The spot density of the
putative DMAsIIIG was dependent on the mixed GSH
concentrations (lanes 4–12), and the spot density of
the remaining GSH markedly increased when DMAsV
was incubated with GSH at GSH:DMAsV molar ratios
greater than 3 (Figures 1a and c). Similar results were
observed when DMAsV and GSH were reacted in phosphate
buffer (pH ¼ 7.4) for 1 h at 371C (data not shown). A putative
DMAsIIIG was extracted from an HPTLC sample (Figure 1a,
lanes 4–12) by using distilled water, applied to another
HPTLC plate, and separated. As shown in Figure 1d, a spot of
the extracted DMAsIIIG (lane 5) appeared at the same
position as that of the putative DMAsIIIG (lane 4) produced
by the mixture of DMAsV and GSH. A GSSG spot was also
detected when the extracted DMAsIIIG was separated on
the HPTLC plate (lane 5).
Statistics
The data represent the mean7s.e.m. of three or more
determinations. Statistical evaluations where appropriate
were carried out with analysis of variance (ANOVA) followed
by Dunnett’s multiple comparison test or Student’s t-test.
A value of Po0.05 was considered significant in all cases.
Reagents
Sodium arsenite (AsIII) and sodium arsenate (AsV) were
purchased from Sigma Chemical Co. (St Louis, MO, USA).
DMAsV sodium salt was purchased from Calbiochem Bio-
sciences Inc. (La Jolla, CA, USA). MMAsV was purchased from
Tri Chemical Co. (Yamanashi, Japan). These purchased
arsenicals were recrystallized twice, and their purities were
499.9% as determined by gas chromatography/mass
spectrometry (GC/MS) (Sakurai et al., 2004a). Endotoxin
contamination of these arsenicals was not detected
(o0.0000003%, wt/wt) using the endotoxin-specific limulus
test (Seikagaku Co., Tokyo, Japan). GSH, oxidized form of
GSH (GSSG), cysteine, N-acetyl-L-cysteine (NAC), human
serum albumin (HSA; 499.0%) were purchased from Sigma.
Bovine serum albumin (BSA) (498.0%) was purchased from
Wako Pure Chemical Industries Ltd (Osaka, Japan). Human
serum (HS) was purchased from Chemicon International
Inc. (Temecula, CA, USA). FBS was purchased from Thermo
Electron Co. (Melbourne, Australia).
We subsequently identified the chemical species of the
putative DMAsIIIG by HPTLC-ICP MS (Figure 2a) and FAB-MS
(Figure 2b). As shown in Figure 2a, when 10mM DMAsV was
incubated with 30 mM GSH for 1 h at 01C, the DMAsV was
unchanged; however, when DMAsV was incubated with GSH
for 1 h at 371C, more than 85% of the DMAsV was converted to
an unknown arsenic compound that is possibly DMAsIIIG.
This reaction was very similar to those observed using the
HPTLC method (Figure 1). We also analyzed the chemical form
of the putative DMAsIIIG extracted from the HPTLC (Figure 1d,
lane 4) by FAB-MS. As shown in Figure 2b, this was confirmed
to be DMAsIIIG. These data imply that DMAsV combines with
GSH under physiological conditions (pH ¼ 7.4, 371C) at molar
ratios of DMAsV:GSH ¼ 1:3 and is converted to DMAsIIIG that
can be detected by iodine vapor on HPTLC plates.
DMAsIIIG was synthesized from DMAsV and GSH by
incubating in distilled water for 1 h at 371C (Scott et al.,
1993; Kala et al., 2000), and separated by HPTLC method (see
details in Results section). HPTLC was performed on 0.1 mm
precoated silica gel HPTLC plates (Merck KgaA, Darmstadt,
Germany) with a developing solvent of ethyl acetate:acetic
acid:water (3:2:1), and iodine vapor was used for the visual
detection of DMAsIIIG (Sakurai et al., 2002). Separated
DMAsIIIG was collected from the HPTLC plate with silica
gel and stored at À851C. DMAsIIIG was extracted by ice-cold
distilled water before experimental use and centrifuged at
20 000 g for 5 min at 41C to remove silica gel. Supernatants
were then filtered through 0.20 mm filters and used as an
aqueous solution of DMAsIIIG.
Cytotoxicity of purified DMAsIIIG in TRL 1215 cells
The cytotoxicity of the purified DMAsIIIG extracted from an
HPTLC spot was compared with that of inorganic trivalent
AsIII and pentavalent DMAsV using rat liver TRL 1215 cells.
It was found that DMAsIIIG showed strong cytotoxicity at
submicromolar levels; its LC50 value after exposure of TRL
1215 cells for 48 h was 160 nM (Figure 3a). When cells were
exposed to 200 nM DMAsIIIG, a significant number of cells died
British Journal of Pharmacology (2006) 149 888–897

Cytotoxicity of dimethylarsinous glutathione
T Sakurai et al
891
a
b
8
8
7
6
5
4
3
7
6
5
4
3
2
1
0
2
1
0
1
4
6 7 8 9
101112
2 3
5
0
1
2
Arsenic (µg per area)
c
160
140
120
100
80
d
60
40
20
0
1 2 3 4 5
1 2 3 4 5
1
2
3
GSH/DMAsV Ratio
4
5
Figure 1 DMAsV readily combines with GSH in water. (a) DMAsV (10 mM) was incubated with or without 10–50 mM GSH in distilled water for
1 h at 371C. After incubation, aliquots (5 ml) of these mixtures were spotted on a HPTLC plate and developed using ethyl acetate: acetic acid:
water (3:2:1). Lane 1, 10 mM GSH; lane 2, 10 mM GSSG; lane 3, 10 mM DMAsV; lanes 4–12, 10 mM DMAsV plus 10 (lane 4), 15 (lane 5), 20
(lane 6), 25 (lane 7), 30 (lane 8), 35 (lane 9), 40 (lane 10), 45 (lane 11) or 50 mM (lane 12) GSH. Separated compounds were visually detected
with iodine vapor. (b) The compounds separated in (a, lane 8) were extracted at every 1 cm length (area) from the development starting point
by distilled water, and the arsenic content in each area was measured using AAS. Results are expressed as the arithmetic mean7s.e.m. of three
separate experiments. (c) Changes in visible spot density for the remaining GSH in (a, lanes 4–12) are expressed as percentages of the spot
density of 10 mM GSH alone (100%) (a, lane 1). (d) Purification of DMAsIIIG. Putative DMAsIIIG separated in (a, lane 8) was extracted from the
HPTLC plates using distilled water and purified. Five *microliters of the 10 mM purified putative DMAsIIIG (arsenic content was measured by
AAS) was then spotted onto a new HPTLC plate (d, lane 5); 5 ml of 10 mM GSH (d, lane 1), 10 mM GSSG (d, lane 2), 10 mM DMAsV (d, lane 3)
and 10 mM DMAsV incubated with 30 mM GSH for 1 h at 371C (d, lane 4) were also spotted. These compounds were developed using ethyl
acetate:acetic acid:water (3:2:1) and visually detected with iodine vapor.
after only 24 h exposure (Figure 3b). In contrast, AsIII and
DMAsV exhibited no cytotoxicity up to concentrations of 1mM
and up to 48 h exposure (Figure 3a). The uptake of DMAsIIIG
into TRL 1215 cells was also markedly higher than that of
other inorganic and pentavalent methyl arsenicals such as
AsIII, AsV, MMAsV, or DMAsV. As shown in Figure 3c, when
cells were exposed to 200nM DMAsIIIG for 48 h at 371C, the
cellular arsenic content increased quickly from 10 min
and reached a peak at 12 h exposure, when cell death was
still low (Figure 3b), of 82.379.0 ng mgÀ1 cellular protein
(69.177.3 nM; n ¼ 3), as determined by AAS. This high cellular
arsenic content was retained until 48 h. However, when cells
were exposed to 200 nM AsIII or DMAsV for 48 h at 371C, the
cellular arsenic content was only 4.970.4 or 1.370.1 ng mgÀ1
cellular protein (4.170.4 or 1.170.1 nM; n ¼ 3), respectively.
Preincubation for 1 h at 371C or more significantly reduced,
and a 48 h preincubation completely abolished, the cyto-
toxicity induced by DMAsIIIG. As shown in Figure 4b,
DMAsIIIG-induced cytotoxicity was significantly reduced by
preincubation in water at a temperature greater than 41C;
however, the cytotoxicity was retained after a preincubation
at À851C. Figure 5 shows that preincubation of DMAsIIIG in
water for 24 or 48 h at 371C also reduced its cellular uptake
into TRL 1215 cells. Most arsenicals accumulated in the
cytoplasm of the cells. As shown in Figure 5b, a DMAsIIIG
HPTLC spot also disappeared after preincubation in water at
371C in a time-dependent manner. However, the addition of
GSH prevented this and a visible DMAsIIIG HPTLC spot was
retained in a GSH concentration-dependent manner even
after a 48 h preincubation at 371C (lanes 12–15; similar
results are also shown in Figure 8, lanes 7 and 8).
DMAsIIIG is an unstable chemical
The purified DMAsIIIG extracted from HPTLC was preincu-
bated in water for 1–48 h at 371C, and its cytotoxicity was
then observed. As shown in Figure 4a, preincubation reduced
the cytotoxicity of DMAsIIIG in a time-dependent manner.
Exogenous GSH prevented DMAsIIIG-induced cytotoxicity by
reducing cellular uptake of DMAsIIIG
Figure 6 shows that exogenous GSH prevented DMAsIIIG-
induced cytotoxicity in TRL 1215 cells, either at millimolar
British Journal of Pharmacology (2006) 149 888–897

Cytotoxicity of dimethylarsinous glutathione
892
T Sakurai et al
AsIII
a
DMAsV
MMAsV
Standard
AsV
2
3
4
5
6
7
8
9
10
0
1
DMAsV + GSH 0°C
2
3
4
5
6
7
8
9
10
0
1
DMAsV + GSH 37°C
2
3
4
5
6
7
8
9
10
0
1
Elution Time (min)
b
369.2
10
5
434.1
412.1
461.3
391.2
450.0
0
370
380
390
400
410
420
430
m/z
440
450
460
470
480
490
Figure 2 Arsenic analysis. (a) HPLC-ICP MS chromatograms of standard arsenic solutions (upper figure; arsenic concentrations are 10 ppb);
10 mM DMAsV incubated with 30 mM GSH in distilled water for 1 h at 01C (middle figure) or 371C (lower figure). (b) FAB-MS spectrum of
purified DMAsIIIG. DMAsV (10 mM) was incubated with 30 mM GSH in distilled water for 1 h at 371C. After incubation, aliquots of this mixture
were spotted onto an HPTLC plate and developed using ethyl acetate:acetic acid:water (3:2:1). Separated DMAsIIIG was extracted using
distilled water and purified. The FAB-MS of purified DMAsIIIG employed positive-ion mode and showed signals at m/z 412.1 [M þ H]þ , m/z
434.1 [M þ Na]þ and m/z 450.0 [M þ K]þ . Other signals originating from the glycerol matrix were observed at m/z 369.2 [4 M þ H]þ , m/z
391.2 [4 M þ Na]þ and m/z 461.3 [5 M þ H]þ
.
concentrations (Figure 6a) or at 20 mM, GSH significantly
inhibited cytotoxicity induced by 100nM DMAsIIIG
(Figure 6b). Similar effects were observed with other thiol
reagents, such as cysteine and N-acetyl-L-cysteine (NAC)
(Figure 6a), but was not observed with GSSG (data not shown).
Figure 7 shows that exogenous GSH also inhibited the uptake
of DMAsIIIG into TRL 1215 cells. When cells were incubated
with 200 nM DMAsIIIG for 48 h at 371C, exogenous GSH
decreased the cellular uptake of DMAsIIIG in a dose-dependent
manner, and 5 mM GSH resulted in almost complete inhibi-
tion (Figures 7a and b). In contrast, exogenous GSH did not
affect the cellular uptake of other arsenic compounds, such as
AsIII and DMAsV, at any concentration (Figure 7a).
Serum, serum albumin and red blood cells prevented DMAsIIIG-
induced cytotoxicity by reducing the cellular uptake of DMAsIIIG
Table 1 shows the effects of the addition of HS, HSA, or
human red blood cells (HRBC) on DMAsIIIG-induced cyto-
toxicity in TRL 1215 cells. HS, HSA and HRBC significantly
inhibited DMAsIIIG-induced cytotoxicity at concentrations
similar to those found in vivo in normal human blood. These
blood components also decreased the cellular uptake of
DMAsIIIG into TRL 1215 cells. As shown in Figure 8, an
HPTLC spot of DMAsIIIG disappeared following preincuba-
tion with either HS or HSA at 371C for more than 6 h (lanes
9–14). Similar results were obtained by the addition of FBS
and BSA (data not shown).
British Journal of Pharmacology (2006) 149 888–897

Cytotoxicity of dimethylarsinous glutathione
T Sakurai et al
893
120
100
80
00hh
11hh
33hh
a
120
a
*
*
100
80
*
*
*
*
*
6 h
6h
*
12h
12 h
*
60
*
*
*
*
24h
24 h
4488hh
*
*
*
60
40
20
DMAsIIIG
AsIII
40
*
*
*
*
*
DMAsV
20
*
*
0
0
**
1
0
10
10
100
100
1000
Arsenic Concentration (nM)
Arsenic Concentration (nM)
0
0
0
1000
120
100
b
Arsenic Concentration (nM)
120
b
80
60
40
20
0
– 85 °C
4 °C
22 °C
37 °C
*
*
*
100
80
40
20
10
0
*
*
*
*
*
DMAsIIIG
*
*
*
*
0
10
20
30
40
50
*
*
Incubation time (h)
**
*
***
*
*
c
100
100
80
*
*
80
80
0
10
10
20
0
30
30
40
40
50
50
*
60
40
20
0
*
*
Incubation Time (h)
*
60
60
DMAsIIIG
AsIII
Figure 4 DMAsIIIG is an unstable chemical; DMAsIIIG-induced
cytotoxicity was reduced by preincubation. (a) DMAsIIIG alone was
preincubated in distilled water for 0–48 h at 371C. TRL 1215 cells
were exposed to various concentrations of these preincubated
DMAsIIIG for 48 h at 371C, and cellular viability was then assessed by
the AlamarBlue assay. (b) DMAsIIIG was preincubated in distilled
water for 0–48 h at À85, 4, 22 or 371C. TRL 1215 cells were exposed
to 500 nM of these preincubated DMAsIIIG for 48 h at 371C, and
cellular viability was then assessed by the AlamarBlue assay. All
results are expressed as the arithmetic mean7s.e.m. of three
separate experiments each performed in triplicate (n ¼ 9).
*Po0.001, in comparison to the cells exposed to the same
concentrations of DMAsIIIG without preincubation, **Po0.01,
***Po0.05.
40
40
AsV
*
MMAsV
20
20
DMAsV
*
0
0
10
20
20
30
30
40
50
50
0
0
10
40
Incubation time (h)
Figure 3 Cytotoxicity and cellular uptake of purified DMAsIIIG in
TRL 1215 cells. (a) TRL 1215 cells were exposed to various
concentrations of DMAsIIIG, AsIII or DMAsV for 48 h at 371C, and
cellular viability was then assessed by the AlamarBlue assay. (b) TRL
1215 cells were exposed to 200 nM DMAsIIIG for 0-48 h at 371C, and
cellular viability was assessed by the AlamarBlue assay. (c) TRL 1215
cells were exposed to 200 nM AsIII, AsV, MMAsV, DMAsV or DMAsIIIG
for 0–48 h at 371C, and time course of cellular arsenic contents were
measured using AAS. Cellular arsenic contents are expressed as ng
per mg cellular protein or nanomolar. All results are expressed as the
arithmetic mean7s.e.m. of three separate experiments each
performed in triplicate (n ¼ 9). *Po0.001, in comparison to the
control cells incubated with the medium alone.
1999); therefore, the accumulation of the metabolites of
inorganic arsenicals is likely to occur. It was recently reported
that toxic trivalent dimethylarsenic compounds might be
produced through the methylation of inorganic arsenicals
(Wang et al., 2004), and that the cytotoxicity of a synthetic
trivalent dimethylarsenic compound – DMAsIIII – was higher
than that of inorganic arsenicals (Styblo et al., 2000; Mass
et al., 2001). Trivalent dimethylarsenical is thought to be
generated as an arsenical-GSH conjugate – DMAsIIIG – in the
human body. Hayakawa et al. (2005) recently reported that
inorganic arsenicals were metabolized to DMAsIIIG in the
cells by a putative arsenic methyltransferase – Cyt 19
(AS3MT) – and might be excreted from the cells in the
Discussion
Recently, an inorganic arsenical – AsIII – was found to be
effective in inducing complete remission in patients with
APL (Chen et al., 1997; Shen et al., 1997). Multiple AsIII
injections (10 mg dayÀ1) for at least 28 consecutive days are
required to induce complete remission (Shen et al., 1997).
Methylated arsenicals are formed from inorganic arsenicals
by enzymatic methylation (Hindmarsh and McCurdy, 1986)
and accumulate in the human body during AsIII treatment as
a chemotherapeutic agent (Wang et al., 2004; Fukai et al.,
2006). Arsenic intoxication also occurs widely through the
consumption of contaminated well water or foods contain-
ing inorganic arsenicals (Morton and Dunnette, 1994; NRC,
human body as DMAsIIIG. However, the toxicity of DMAsIII
G
has not been established. Here, we studied in detail the
in vitro cytotoxicity of synthetic DMAsIIIG compared with
those of AsIII and pentavalent DMAsV using rat liver TRL
1215 cells.
Scott et al. (1993) reported that DMAsV might combine
with GSH in a molar ratio of DMAsV:GSH ¼ 1:3 and that
British Journal of Pharmacology (2006) 149 888–897

Cytotoxicity of dimethylarsinous glutathione
894
T Sakurai et al
DMAsIII
+ GSH (5 mM)
G
a
60
cytoplasm
cell membrane
GSH (5 mM)
60
40
+ Cys (5 mM)
CySH (5 mM)
40
N+ANCA(5Cm(5Mm) M)
120
100
80
a
20
0
20
0
**
**
60
40
20
*
*
Incubation Time (h);
0
24
48
*
*
*
b
*
*
*
*
0
10
0
100
100
1000
1000
Arsenic Concentration (nM)
Arsenic Concentration (nM)
B b
120
100
80
†
†
†
†
†
†
†
†
†
†
†
†
†
†
60
40
20
†
†
1
2
3
4
5
6 7 8 9 10 11 12 13 14 15
DMAsIIIG 100 nM
††
Figure 5 DMAsIIIG is an unstable chemical. (a) Cellular uptake of
DMAsIIIG was reduced by preincubation. DMAsIIIG was preincubated
in distilled water for 24 or 48 h at 371C. TRL 1215 cells were exposed
to 200 nM of these preincubated DMAsIIIG for 48 h at 371C. After
exposure, cellular arsenic contents were measured using AAS and are
expressed as ng per mg cellular protein or nanomolar. Results are
expressed as the arithmetic mean7s.e.m of three separate experi-
ments each performed in triplicate (n ¼ 9). (b) The chemical form
of DMAsIIIG was altered by preincubation. 10 mM DMAsIIIG was
preincubated with or without 1–30 mM GSH in distilled water for 0–
48 h at 371C. After preincubation, aliquots (5 ml) of these mixtures
were spotted on an HPTLC plate that was developed using ethyl
acetate:acetic acid:water (3:2:1), and the separated spots were
detected with iodine vapor. Lane 1, 10 mM GSH; lane 2, 10 mM
GSSG; lane 3, 10 mM DMAsV; lane 4, 10 mM DMAsV incubated with
30 mM GSH for 1 h at 371C; lane 5, 10 mM DMAsIIIG; lanes 6–11,
10 mM DMAsIIIG alone preincubated in distilled water for 1 (lane 6),
3 (lane 7), 6 (lane 8), 12 (lane 9), 24 (lane 10) or 48 h (lane 11) at
371C; lanes 12–15, 10 mM DMAsIIIG preincubated in distilled water
with 1 (lane 12), 3 (lane 13), 10 (lane 14) or 30 mM (lane 15) GSH
for 48 h at 371C.
DMAsIIIG 1 ꢀM
0
0.1
0.1
1
1
10
10
0
0.02
0.02
GSH concentration (mM)
GSH Concentration (mM)
Figure 6 Effects of GSH, cysteine or NAC on DMAsIIIG-induced
cytotoxicity. (a) TRL 1215 cells were exposed to various concentra-
tions of DMAsIIIG for 48 h at 371C in the presence or absence of
5 mM GSH, Cys or NAC, and cellular viability was then assessed by
the AlamarBlue assay. (b) TRL 1215 cells were exposed to 100 nM or
1 mM DMAsIIIG for 48 h at 371C in the presence or absence of various
concentrations of GSH, and cellular viability was then assessed. All
results are expressed as the arithmetic mean7s.e.m. of three
separate experiments each performed in triplicate (n ¼ 9).
*Po0.001 in comparison to the control cells incubated with the
medium alone, **Po0.05. wPo0.001 compared with the cells
exposed to the same concentrations of DMAsIIIG alone, wwPo0.01.
oxidized to the weakly toxic pentavalent dimethylarsenic
compound DMAsV (Figure 9).
The submicromolar LC50 values of DMAsIIIG determined
in this study are very important in the elucidation of the
actual role of the metabolic methylation of arsenicals in the
human body, because other major human arsenic metabo-
lites, including inorganic arsenicals, do not show cytotoxi-
city in mammalian cells at the submicromolar level (Sakurai
et al., 1998, 2002). Additionally, it has been reported that the
mean blood concentration of total arsenical was in the
submicromolar range in relapsed APL patients, in whom
complete remission had been induced by AsIII treatment
(Fukai et al., 2006); similar levels were found in the blood of
patients with chronic arsenic poisoning in Inner Mongolia,
China and northeastern Taiwan who continued to drink well
water containing high concentrations of inorganic arsenicals
(Pi et al., 2000; Wu et al., 2003). Interestingly, exogenous
thiol agents, such as GSH, cysteine and NAC, greatly
decreased both the cytotoxicity and cellular arsenic contents
induced by DMAsIIIG exposure. The addition of GSH
inhibited the preincubation-induced disappearance of the
visible DMAsIIIG spot on HPTLC plates. It is generally
DMAsIIIG was formed in vitro (Figure 9). In this study, we
confirmed the formation of the DMAs-GSH conjugate
DMAsIIIG during the incubation of 10 mM DMAsV with
30 mM GSH for 1 h at 371C using HPLC-ICP MS and FAB-MS
analysis. We readily purified this chemical by an HPTLC
method and observed its in vitro cytotoxicity using rat liver
TRL 1215 cells. Purified DMAsIIIG was much more cytotoxic
than either AsIII or DMAsv; and its LC50 value was 160 nM.
However, this action of purified DMAsIIIG was readily
reduced by preincubation in water at temperatures greater
than 41C. The high cellular arsenic uptake of DMAsIIIG and
a visible DMAsIIIG HPTLC spot were also greatly reduced
by preincubation. Under our experimental conditions, only
GSH and/or GSH-conjugated chemicals could be visually
detected by iodine vapor on an HPTLC plate; however, a
DMAsV spot could not be detected. Thus, the GSH molecule
might readily be removed from DMAsIIIG during the
preincubation period. The released DMAsIII þ might sub-
sequently be converted in water to DMAsIIIOH and finally be
British Journal of Pharmacology (2006) 149 888–897

Cytotoxicity of dimethylarsinous glutathione
T Sakurai et al
895
a
cytoplasm
60
cell membrane
60
60
40
20
4400
20
20
*
*
*
0
*
0
0
GSH (mM);
0
1
5
0
1
5
0
1
5
AsIII
DMAsV
DMAsIIIG
80
80
b
60
60
*
1
2
3
4
5
6
7
8
9 10 11 12 13 14
40
40
*
Figure 8 Effects of GSH, HS or HSA on the chemical form of
DMAsIIIG. DMAsIIIG (10 mM) was incubated with or without 10 mM
GSH, 50% HS, or 50 g lÀ1 HSA for 6 or 48 h at 371C. After incubation,
aliquots (5 ml) of these mixtures were spotted onto an HPTLC plate
that was developed using ethyl acetate:acetic acid:water (3:2:1), and
the separated spots were detected with iodine vapor. Lane 1, 10 mM
GSH; lane 2, 10 mM GSSG; lane 3, 10 mM DMAsV; lane 4, 10 mM
DMAsIIIG; lanes 5–6, 10 mM DMAsIIIG preincubated in distilled water
for 6 (lane 5) or 48 h (lane 6) at 371C; lanes 7–8, 10 mM DMAsIIIG
incubated with 10 mM GSH for 6 h (lane 7) or 48 h (lane 8) at 371C;
lane 9, 50% HS; lanes 10–11, 10 mM DMAsIIIG incubated with 50%
HS for 6 h (lane 10) or 48 h (lane 11) at 371C; lane 12, 50 g lÀ1 HSA;
lanes 13–14, 10 mM DMAsIIIG incubated with 50 g lÀ1 HSA for 6 h
(lane 13) or 48 h (lane 14) at 371C.
20
20
*
*
*
0
0
0
20
100
500
1000
5000
GSH Concentration (ꢀM)
Figure 7 Effect of GSH on cellular uptake of arsenicals. (a) TRL 1215
cells were exposed to 500 nM AsIII, 500 nM DMAsV or 200 nM
DMAsIIIG for 48 h at 371C in the presence or absence of 1 or 5 mM
GSH, and the cellular arsenic contents were then measured by AAS.
(b) TRL 1215 cells were exposed to 200 nM DMAsIIIG for 48 h at 371C
in the presence or absence of various concentrations of GSH, and
cellular arsenic contents were then measured by AAS. Cellular arsenic
contents are expressed as ng per mg cellular protein (a) or
nanomolar (a and b). All results are expressed as the arithmetic
mean7s.e.m. of three separate experiments each performed in
triplicate (n ¼ 9). *Po0.001 in comparison to the cells exposed to
200 nM DMAsIIIG alone.
3GSH
2H2O + GSSG
O
H3C
H3C
H3C
H3C
AsIII
AsV
SG
OH
Dimethylarsinic Acid
(DMAsV)
Dimethylarsinous Glutathione
(DMAsIIIG)
Table 1 Effects of HS, HSA, and HRBC on the cytotoxicity (LC50) and
cellular uptake of DMAsIIIG
LC50 (nM)
Cellular arsenic contents (nM)
DMAsIIIG
2H2O + O2
100 nM
200 nM
O2
2GSSG
H3C
H3C
DMAsIIIG alone
DMAsIIIG þ HS
DMAsIIIG þ HSA
DMAsIIIG þ HRBC
160.3712.3
710.0775.1*
41000*
42.770.6
19.470.5*
18.170.3*
8.070.2*
62.771.2
45.870.7*
38.670.4*
11.470.8*
AsIII+
OH
965.0712.2*
Dimethylarsinous Acid
(DMAsIII
)
Abbreviations: DMAsIIIG, dimethylarsinous glutathione; HS, human serum;
HSA, human serum albumin; HRBC, human red blood cells.
Figure 9 Putative chemical reactions of dimethylarsenic com-
pounds with GSH.
TRL1215 cells were exposed to DMAsIIIG in the presence or absence of HS
(50%), HSA (50 g lÀ1) or HRBC (5 Â 1012 cells lÀ1) for 48 h at 371C. Cellular
viability and cellular arsenic contents were then measured. Results are
expressed as the arithmetic mean7s.e.m. of three separate experiments each
performed in triplicate (n ¼ 9).
so that DMAsIIIG itself cannot subsequently be transported
into cells. The significant cytotoxicity induced by DMAsIIIG
could be the result of the dissociation of the conjugate into
trivalent dimethylarsenic compounds and GSH, the former
probably becoming DMAsIII þ and/or DMAsIIIOH, before
being transported into cells (Thompson, 1993; Dopp et al.,
2004, 2005) (Figure 9). We showed in this study that a GSSG
*Po0.001, in comparison to the cells exposed to DMAsIIIG alone.
believed that the large GSH molecule is not transported
efficiently into cells (De Flora et al., 2001). Therefore, these
thiol agents may maintain the molecular form of DMAsIIIG
British Journal of Pharmacology (2006) 149 888–897

Cytotoxicity of dimethylarsinous glutathione
896
T Sakurai et al
spot appeared when purified DMAsIIIG was applied and
developed again on a new HPTLC plate, indicating that
the GSH molecule readily dissociates from DMAsIIIG and is
converted to oxidized GSH. The plasma GSH concentrations
remain at micromolar levels in healthy humans (Donner
et al., 1998; Kokcam and Naziroglu, 1999), and we showed
in this study that exogenous GSH at micromolar levels
prevented submicromolar DMAsIIIG-induced cytotoxicity. In
addition, the physiological concentrations of normal HS,
HSA and HRBC significantly reduced both the cytotoxicity
and cellular arsenic contents induced by DMAsIIIG exposure.
These findings suggest that the significant cytotoxicity of
DMAsIIIG may never be manifest in humans even if
DMAsIIIG is enzymatically formed from inorganic arsenicals
in the human body. HS and HSA could not prevent the
disappearance of a DMAsIIIG spot during the preincubation
on HPTLC plates. The reason why serum and serum albumin
inhibit the cytotoxicity of DMAsIIIG is unclear; however, it
is suggested that they specifically reduce the cytotoxicity
by combining protein thiols with the DMAsIII þ ion or by
facilitating the oxidation of DMAsIII þ to DMAsV.
References
Buchet JP, Lauwerys R, Roels H (1980). Comparison of the urinary
excretion of arsenic metabolites after a single oral dose of sodium
arsenite, monomethylarsonate, or dimethylarsinate in man. Int
Arch Occup Environ Health 48: 71–79.
Chen GQ, Shi XG, Tang W, Xiong SM, Zhu J, Cai X et al. (1997). Use
of arsenic trioxide (As2O3) in the treatment of acute promyelocytic
leukemia (APL): I. As2O3 exerts dose-dependent dual effects on
APL cells. Blood 89: 3345–3353.
De Flora S, Izzotti A, D’Agostini F, Balansky RM (2001). Mechanisms
of N-acetylcysteine in the prevention of DNA damage and cancer,
with special reference to smoking-related end-points. Carcinogen-
esis 22: 999–1013.
Donner MG, Klein GK, Mathes PB, Schwandt P, Richter WO (1998).
Plasma total homocysteine levels in patients with early-onset
coronary heart disease and
Metaboism 47: 273–279.
a low cardiovascular risk profile.
Dopp E, Hartmann LM, Florea AM, von Recklinghauesn U, Pieper R,
Shokouhi et al. (2004). Uptake of inorganic and organic
B
derivatives of arsenic associated with induced cytotoxic and
genotoxic effects in Chinese hamster ovary (CHO) cells. Toxicol
Appl Pharmacol 201: 156–165.
Dopp E, Hartmann LM, von Recklinghauesn U, Florea AM, Rabieh S,
Zimmermann U et al. (2005). Forced uptake of trivalent and
pentavalent methylated and inorganic arsenic and its cyto-/
genotoxicity in fibroblasts and hepatoma cells. Toxicol Sci 87:
46–56.
Fukai Y, Hirata M, Ueno M, Ichikawa N, Kobayashi H, Saitoh H
et al. (2006). Clinical pharmacokinetic study of arsenic trioxide
in an acute promyelocytic leukemia (APL) patient: speciation
of arsenic metabolites in serum and urine. Biol Pharm Bull 29:
1022–1027.
Hayakawa T, Kobayashi Y, Cui X, Hirano S (2005). A new metabolic
pathway of arsenite: arsenic-glutathione complexes are substrates
for human arsenic methyltransferase Cyt19. Arch Toxicol 79:
183–191.
GSH may be a key molecule in preventing chronic arsenic
toxicity. Clearly, if GSH levels are reduced, inorganic
arsenicals are more toxic and induce complete necrosis in
mammalian cells (Sakurai et al., 1998, 2002). GSH depletion
causes an essentially minor, nontoxic, human monomethyl-
arsenic metabolite – MMAsV – to become toxic (Sakurai et al.,
2004a, 2005). DMAsV appears unable to induce apoptosis
after GSH depletion (Sakurai et al., 1998, 2002, 2004b) that
allows the survival of damaged abnormal cells. Furthermore,
the strongly toxic DMAsIIIG may exert its cytotoxicity by
dissociating into DMAsIII þ and GSÀ and by being trans-
ported into cells after depletion of blood GSH levels. GSH
may be also a cofactor in the enzymatic methylation of
inorganic arsenicals in humans (Zakharyan et al., 2001;
Hayakawa et al., 2005). These findings suggest that GSH
depletion may trigger the disruption of the arsenic detoxify-
ing system and induce various arsenic toxicities in humans.
In fact, reduced nonprotein sulfhydryl levels in peripheral
blood, often seen as an indicator of GSH levels, were
observed in chronic arsenic poisoning patients in Inner
Mongolia, China (Pi et al., 2000). Further research will
be required in order to determine the role of metabolic
methylation and GSH in arsenic toxicity in APL patients who
are injected with AsIII as a chemotherapeutic agent and/or in
chronic arsenic poisoning in those who regularly ingest
inorganic arsenic-contaminated well water.
Hindmarsh JT, McCurdy
F (1986). Clinical and environmental
aspects of arsenic toxicity. Crit Rev Clin Lab Sci 23: 315–347.
Idoine JB, Elliott JM, Wilson MJ, Weisburger EK (1976). Rat liver cells
in culture: effect of storage, long-term culture, and transformation
on some enzyme levels. In Vitro 12: 541–553.
Kala SV, Neely MW, Kala G, Prater CI, Atwood DW, Rice JS et al.
(2000). The MRP2/cMOAT transporter and arsenic-glutathione
complex formation are required for biliary excretion of arsenic.
J Biol Chem 275: 33404–33408.
Kokcam I, Naziroglu M (1999). Antioxidants and lipid peroxidation
status in the blood of patients with psoriasis. Clin Chim Acta 289:
23–31.
Mass MJ, Tennant A, Roop BC, Cullen WR, Styblo M, Thomas DJ et al.
(2001). Methylated trivalent arsenic species are genotoxic. Chem
Res Toxicol 14: 355–361.
Mizoguchi H, Hara
S (1996). Effect of fatty acid saturation in
membrane lipid bilayers on simple diffusion in the presence of
ethanol at high concentrations. J Ferment Bioeng 81: 406–411.
Morton WE, Dunnette DA (1994). Health effects of environmental
arsenic. In: Nriagu JO (ed). Arsenic in the Environment Vol. II. John
Wiley & Sons, New York, pp 17–34.
NRC (1999). Arsenic in the Drinking Water. National Research Council,
National Academy Press, Washington, DC.
Ohta T, Sakurai T, Fujiwata K (2004). Effects of arsenobetaine, a major
organic arsenic compound in seafood, on the maturation and
functions of human peripheral blood monocytes, macrophages
and dendritic cells. Appl Organomet Chem 18: 431–437.
Pi J, Kumagai Y, Sun G, Yamauchi H, Yoshida T, Iso H et al. (2000).
Decreased serum concentrations of nitric oxide metabolites
among Chinese in an endemic area of chronic arsenic poisoning
in inner Mongolia. Free Radic Biol Med 28: 1137–1142.
Acknowledgements
We express our thanks to Ms Chihiro Kawata, Mr Kouichirou
Matsuda, Ms Tomoe Sakoda and Mr Hiroki Soejima (Faculty
of Pharmaceutical Sciences, Tokushima Bunri University) for
their valuable technical assistance.
Sakurai T, Kaise T, Matsubara C (1998). Inorganic and methylated
arsenic compounds induce cell death in murine macrophages via
different mechanisms. Chem Res Toxicol 11: 273–283.
Sakurai T, Kojima C, Ochiai M, Ohta T, Sakurai MH, Waalkes MP et al.
(2004a). Cellular glutathione prevents cytolethality of mono-
methylarsonic acid. Toxicol Appl Pharmacol 195: 129–141.
Conflict of interest
The authors state no conflict of interest.
British Journal of Pharmacology (2006) 149 888–897

Cytotoxicity of dimethylarsinous glutathione
T Sakurai et al
897
Sakurai T, Ochiai M, Kojima C, Ohta T, Sakurai MH, Takada NO et al.
(2004b). Role of glutathione in dimethylarsinic acid-induced
apoptosis. Toxicol Appl Pharmacol 198: 354–365.
Sakurai T, Ochiai M, Kojima C, Ohta T, Sakurai MH, Takada NO et al.
(2005). Preventive mechanism of cellular glutathione in mono-
methylarsonic acid-induced cytolethality. Toxicol Appl Pharmacol
206: 54–65.
Sakurai T, Qu W, Sakurai MH, Waalkes MP (2002). A major human
arsenic metabolite, dimethylarsinic acid, requires reduced glu-
tathione to induce apoptosis. Chem Res Toxicol 15: 629–637.
Scott N, Hatlelid KM, Mackenzie NE, Carter DE (1993). Reactions of
arsenic (III) and arsenic (V) species with glutathione. Chem Res
Toxicol 6: 102–106.
Styblo M, Del Razo LM, Vega L, Germolec DR, Lecluyse EL, Hamilton
GA et al. (2000). Comparative toxicity of trivalent and pentavalent
inorganic and methylated arsenicals in rats and human cells. Arch
Toxicol 74: 289–299.
Thompson DJ (1993). A chemical hypothesis for arsenic methylation
in mammals. Chem Biol Interact 88: 89–114.
Wang Z, Zhou J, Lu X, Gong Z, Le XC (2004). Arsenic speciation in
urine from acute promyelocytic leukemia patients undergoing
arsenic trioxide treatment. Chem Res Toxicol 17: 95–103.
Wu MM, Chiou HY, Ho IC, Chen CJ, Lee TC (2003). Gene expression
of inflammatory molecules in circulating lymphocytes from
arsenic-exposed human subjects. Environ Health Perspect 111:
1429–1438.
Shen ZX, Chen GQ, Ni JH, Li XS, Xiong SM, Qiu QY et al. (1997). Use
of arsenic trioxide (As2O3) in the treatment of acute promyelocytic
leukemia (APL): II. Clinical efficacy and pharmacokinetics in
relapsed patients. Blood 89: 3354–3360.
Shraim A, Hirano S, Yamauchi H (2001). Extraction and speciation
of arsenic in hair using HPLC-ICP MS. Anal Sci 17 (Suppl):
i1729–i1732.
Yamauchi H, Yamamura Y (1979). Dynamic change of inorganic
arsenic and methylarsenic compounds in human urine after oral
intake as arsenic trioxide. Ind Health 17: 79–83.
Zakharyan RA, Sampayo-Reyes A, Healy SM, Tsaprallis G, Board PG,
Liebler DC et al. (2001). Human monomethylarsonic acid (MMAV)
reductase is a member of the glutathione-S-transferase super-
family. Chem Res Toxicol 14: 1051–1057.
British Journal of Pharmacology (2006) 149 888–897