Isolation and identification of a new 2-phenylbenzotriazole-type mutagen (PBTA-3) in the Nikko River in Aichi, Japan

Article abstract of DOI:10.1021/tx0000264

We have previously determined the chemical structures of two 2- phenylbenzotriazole mutagens (PBTA-1 and PBTA-2) in blue cotton-adsorbed material from the Nishitakase River in Kyoto, Japan. In the present study, further analysis of mutagenic substances in the Nikko River, which flows through Aichi Prefecture in Japan, allowed the isolation of a new mutagen. Material (2.2 g) adsorbed on blue cotton (3 kg) at a site below the sewage plant on the Nikko River was purified by various column chromatographies, and a mutagen (120 μg) accounting for 11% of the total mutagenicity was isolated. On the basis of data from UV, mass, and ¹H NMR spectra of the mutagen, the compound was deduced to be a PBTA-1 analogue. As with PBTA-1, the mutagen was able to be synthesized from the azo dye 2-[(2-bromo-4,6- dinitrophenyl)azo]-4-methoxy-5-[(2-hydroxyethyl)amino]acetanilide by reduction and chlorination. Since all spectra of the mutagen isolated from the river water were the same as those of the synthesized form, the structure was concluded to be 2-[2-(acetylamino)-4-[(2-hydroxyethyl) amino]-5- methoxyphenyl]-5-amino-7-bromo-4-chloro-2H-benzotriazole (PBTA-3). PBTA-3 is a potent mutagen, inducing 81 000 and 3 000 000 revertants per microgram of Salmonella typhimurium TA 98 and YG1024 respectively, in the presence of an S9 mix. In addition to its detection in the water of the Nikko River, PBTA-3 was detected in water samples from three other rivers flowing through regions where dyeing industries have been developed. Like PBTA-1 and PBTA-2, PBTA-3 might have also been produced from azo dyes during industrial processes in dyeing factories and/or through treatment at sewage plants.

Full text of DOI:10.1021/tx0000264

Chem. Res. Toxicol. 2000, 13, 535-540
535
Articles
Isola tion a n d Id en tifica tion of a New
2-P h en ylben zotr ia zole-Typ e Mu ta gen (P BTA-3) in th e
Nik k o River in Aich i, J a p a n
Tatsushi Shiozawa,† Atsuko Tada,‡ Haruo Nukaya,§ Tetsushi Watanabe,|
Yoshifumi Takahashi,| Masaharu Asanoma, Takeshi Ohe,@
Hiroyuki Sawanishi,X Taka Katsuhara,∇ Takashi Sugimura,‡
Keiji Wakabayashi,*,‡ and Yoshiyasu Terao†
Graduate School of Nutritional and Environmental Sciences and School of Pharmaceutical Science,
University of Shizuoka, 52-1, Yada, Shizuoka 422-8526, Cancer Prevention Division,
National Cancer Center Research Institute, 1-1 Tsukiji 5-chome, Chuo-ku, Tokyo 104-0045,
Kyoto Pharmaceutical University, 5 Nakauchi-cho, Misasagi, Yamashina-ku, Kyoto 607-8414,
Nagoya City Public Health Research Institute, 1-11 Hagiyama-cho, Mizuho-ku, Nagoya-shi,
Aichi 467-8615, Department of Food and Nutrition Science, Kyoto Women’s University,
Kitahiyoshi-cho, Imakumano, Higashiyama-ku, Kyoto 605-8501, Faculty of Pharmaceutical
Sciences, Hokuriku University, Kanagawa-machi, Kanazawa 920-1181, and Central Research
Laboratories, Tsumura & Co., 3586 Yoshiwara, Ami-machi, Inashiki-gun, Ibaraki 300-1155, J apan
Received February 11, 2000
We have previously determined the chemical structures of two 2-phenylbenzotriazole
mutagens (PBTA-1 and PBTA-2) in blue cotton-adsorbed material from the Nishitakase River
in Kyoto, J apan. In the present study, further analysis of mutagenic substances in the Nikko
River, which flows through Aichi Prefecture in J apan, allowed the isolation of a new mutagen.
Material (2.2 g) adsorbed on blue cotton (3 kg) at a site below the sewage plant on the Nikko
River was purified by various column chromatographies, and a mutagen (120 µg) accounting
1
for 11% of the total mutagenicity was isolated. On the basis of data from UV, mass, and H
NMR spectra of the mutagen, the compound was deduced to be a PBTA-1 analogue. As with
PBTA-1, the mutagen was able to be synthesized from the azo dye 2-[(2-bromo-4,6-dinitro-
phenyl)azo]-4-methoxy-5-[(2-hydroxyethyl)amino]acetanilide by reduction and chlorination.
Since all spectra of the mutagen isolated from the river water were the same as those of the
synthesized form, the structure was concluded to be 2-[2-(acetylamino)-4-[(2-hydroxyethyl)-
amino]-5-methoxyphenyl]-5-amino-7-bromo-4-chloro-2H-benzotriazole (PBTA-3). PBTA-3 is a
potent mutagen, inducing 81 000 and 3 000 000 revertants per microgram of Salmonella
typhimurium TA 98 and YG1024 respectively, in the presence of an S9 mix. In addition to its
detection in the water of the Nikko River, PBTA-3 was detected in water samples from three
other rivers flowing through regions where dyeing industries have been developed. Like PBTA-1
and PBTA-2, PBTA-3 might have also been produced from azo dyes during industrial processes
in dyeing factories and/or through treatment at sewage plants.
been detected in river water, drinking water, and indus-
trial/domestic wastewater (1-5), many sources of geno-
toxic contamination still remain to be clarified.
In tr od u ction
Since environmental genotoxic compounds may play
some role in the development of various diseases such
as cancer, their identification is an important step to our
understanding of their significance for human health.
Although various kinds of genotoxic compounds have
Recently, we isolated five mutagens in water samples
from the Nishitakase River below sewage plants in Kyoto
with 100-600 times the mutagenicity of river water in
Tokyo via adsorption onto blue rayon or blue cotton (6)
using Salmonella typhimurium YG1024 with an S9 mix
(7). Among the isolated mutagens, two were identified
as 2-phenylbenzotriazole derivatives, 2-[2-(acetylamino)-
4-[bis(2-methoxyethyl)amino]-5-methoxyphenyl]-5-amino-
7-bromo-4-chloro-2H-benzotriazole (PBTA-1) and 2-[2-
(acetylamino)-4-[N-(2-cyanoethyl)ethylamino]-5-meth-
oxyphenyl]-5-amino-7-bromo-4-chloro-2H-benzotriazole
(PBTA-2) (6, 8). Furthermore, on the basis of synthesis
* Corresponding author.
† Graduate School of Nutritional and Environmental Sciences,
University of Shizuoka.
‡ National Cancer Center Research Institute.
§ School of Pharmaceutical Science, University of Shizuoka.
| Kyoto Pharmaceutical University.
Nagoya City Public Health Research Institute.
@ Kyoto Women's University.
X Hokuriku University.
∇ Tsumura & Co.
10.1021/tx0000264 CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/03/2000

536 Chem. Res. Toxicol., Vol. 13, No. 7, 2000
Shiozawa et al.
following gradient system of methanol in distilled water: 0-40
min, 75%; 40-50 min, a linear gradient of 75-85%; 50-104 min,
85%, at a flow rate of 2 mL/min. An aliquot of each 2 min
fraction was tested for mutagenicity. The above HPLC procedure
was repeated four more times.
Fractions with a retention time of 13-20 min were combined
and evaporated to dryness. Residues were dissolved in 40%
acetonitrile in 25 mM phosphate buffer (pH 2.0) and injected
into a YMC-Pack ODS-A 303 column (5 µm particle size, 4.6 ×
250 mm; YMC Co.) with a mobile phase of 40% acetonitrile in
25 mM phosphate buffer (pH 2.0) at a flow rate of 1 mL/min.
Fractions with a retention time of 16-17 min were purified on
a CAPCELL PAK C18 column (UG80, 5 µm particle size, 4.6 ×
250 mm; Shiseido Co., Tokyo, J apan) with a mobile phase of
30% acetonitrile in 25 mM Tris-HCl buffer (pH 8.5) at a flow
rate of 1 mL/min. A mutagenic compound was isolated from the
fraction with a retention time of 30-32 min. Its purity was
confirmed on the second YMC-Pack ODS-A 303 column with a
mobile phase of 38% acetonitrile in 25 mM phosphate buffer
(pH 2.0) at a flow rate of 0.5 mL/min.
All HPLC procedures were carried out at ambient tempera-
ture, and eluates were monitored for absorbance at 260 nm.
P r ep a r a tion of a La r ge Qu a n tity of th e Mu ta gen . A large
amount of the mutagen, required for structural analysis, was
isolated as follows, using the original compound described above
as a standard marker. A total of 3 kg of blue cotton, prepared
as described previously (14), was divided equally into 22 net
bags, which were hung for 24 h in the Nikko River, then
combined, and washed twice with distilled water. After under-
going dehydration in a spin-dryer, the adsorbed materials were
extracted by being stirred once in 18 L of methanol/ammonia
water (50:1, v/v) for 3 h and twice more in 12 L of the same
solution. The extract was then evaporated to dryness, and the
residue (2.2 g) was dissolved in 30 mL of methanol, filtered
through a glass filter, and applied to a Sephadex LH-20 column
(50 × 870 mm, Pharmacia, Uppsala, Sweden). The materials
were eluted with methanol, and fractions of 85.5 mL each were
collected. Fractions at elution volumes of 2300-2560 mL were
combined and evaporated. The residue (44.2 mg) was dissolved
in 2 mL of methanol and applied again to a Sephadex LH-20
column (23 × 1360 mm) with methanol as a mobile phase, and
fractions of 3.4 mL each were collected. Fractions containing
the mutagen, eluting at elution volumes of 442-479 mL, were
combined and evaporated. The residue, dissolved in 300 µL of
methanol, was finally purified by HPLC on a semipreparative
F igu r e 1. Geographic locations of the water sampling site and
the sewage plant on the Nikko River in Aichi Prefecture.
studies, it was suggested that both mutagens can be
formed from dinitrophenylazo dyes via reduction with
sodium hydrosulfite and subsequent chlorination with
sodium hypochlorite (9). Thus, it is postulated that these
mutagens are produced from azo dyes during industrial
processes in dyeing factories and through treatment at
sewage plants and are then discharged into the river.
Actually, PBTA-1 and PBTA-2 were also detected at
many other points near dyeing factories in the Yodo river
system, in J apan (10).
In addition, the structure-activity relationships, based
on results for PBTA-1, its congeners, and five related
2-phenylbenzotriazoles, suggested that a primary amino
group, along with the planarity of the 2-phenylbenzo-
triazole ring and halogen groups, plays an essential role
in the mutagenic activity (11).
During the assessment of mutagenicity of river water
in J apan, we found that blue rayon-adsorbed material
taken at a site below the sewage plant of the Nikko River
in Aichi Prefecture showed strong mutagenicity toward
S. typhimurium YG1024 in the presence of an S9 mix,
and we attempted to isolate the mutagens in the water
sample. In this paper, isolation and chemical synthesis
from an azo dye of the third PBTA-type mutagen is
described. Detection of this novel mutagen in other rivers
is also described along with possible formation routes,
on the basis of chemical synthesis studies.
YMC-Pack ODS-AM 324 column, followed by HPLC on
a
semipreparative TSK gel ODS-120A column (10 µm particle size,
7.8 × 300 mm, Toso Corp). The mobile phases of 63 and 65%
methanol were pumped in isocratically for the YMC-Pack ODS-
AM 324 column and the TSK gel ODS-120A column, respec-
tively, at a flow rate of 2 mL/min. The mutagen was found in
the peak fractions with retention times of 28-30 and 26-28
min, respectively. Using the above processes, we obtained about
120 µg of the mutagen.
Detection of th e Mu ta gen in Wa ter Sa m p les fr om
River s Oth er Th a n th e Nik k o River . Sample collection was
carried out at the Asuwa River in Fukui in March 1998 and at
the Katsura and Nishitakase rivers in Kyoto in May 1998. At
each sampling site, 15 g of blue rayon was hung for 24 h in the
river and then collected and treated as described above. Blue
rayon extracts were separated by HPLC on a semipreparative
YMC-Pack ODS-AM 324 column with a mobile phase of 75%
methanol at a flow rate of 1.6 mL/min. Fractions with a
retention time of 19-20 min were evaporated to dryness,
dissolved in 400 µL of 50% methanol, and further applied to a
CAPCELL PAK C18 column. A mobile phase of 35% acetonitrile
in 25 mM Tris-HCl buffer (pH 6.5) was pumped in at a flow
rate of 1 mL/min. Fractions with a retention time of 20-21 min
were evaporated to dryness, and residues were dissolved in 400
µL of 50% methanol and applied to a YMC-Pack ODS-A 303
column with a mobile phase of 35% acetonitrile in 25 mM
phosphate buffer (pH 2.0) at a flow rate of 1.0 mL/min. Peaks
Exp er im en ta l P r oced u r es
Isola tion of th e Mu ta gen fr om Wa ter of th e Nik k o
River . Five net bags of blue rayon (5 g per bag; Funakoshi Co.,
Tokyo, J apan) were attached to a floating board and were hung
in the Nikko River for 24 h in April 1997 at a site downstream
from a sewage plant (Figure 1). The blue rayon from different
bags was combined and washed twice with 500 mL of water.
Adsorbed materials were extracted by shaking the blue rayon
in 1 L of methanol/ammonia water (50:1, v/v) three times for 1
h (12, 13). The combined extracts were evaporated to dryness,
and the residue originating from the 25 g of blue rayon was
dissolved in 6 mL of methanol. One-fifth of the solution was
applied to a semipreparative YMC-Pack ODS-AM 324 column
(5 µm particle size, 10 × 300 mm; YMC Co., Kyoto, J apan) for
HPLC (Tosoh Corp., Tokyo, J apan) and then eluted with the

Identification of a River Mutagen, PBTA-3
Chem. Res. Toxicol., Vol. 13, No. 7, 2000 537
due to the same mutagen as that isolated from the Nikko River
were detected at 19.5 min.
removal of the solvent, the oily residue was subjected to column
chromatography on silica gel (eluent, AcOEt) to yield colorless
crystals (670 mg, 15%). Mp: 133-134 °C. MS m/z: 224 (M+).
1H NMR (DMSO-d6): δ 1.98 (3H, s, COCH3), 3.07 (2H, q, J )
5.8 Hz, CH2), 3.62 (2H, q, J ) 5.8 Hz, CH2), 3.75 (3H, s, OCH3),
4.75 (1H, t, J ) 5.8 Hz, OH or NH), 4.77 (1H, t, J ) 5.8 Hz, OH
or NH), 6.67 (1H, d, J ) 8.7 Hz, ArH), 6.79 (1H, br, ArH), 6.80
(1H, dd, J ) 2.4, 8.7 Hz, ArH), 9.54 (1H, br, NH). 13C NMR
(DMSO-d6): δ 23.84, 45.32, 55.36, 59.32, 101.66, 106.32, 109.38,
133.23, 137.96, 142.48, 167.45.
All HPLC procedures were carried out at ambient tempera-
ture, and the eluates were monitored for absorbance at 260 nm
using a Shimadzu SPD-10AV spectrometric detector (Kyoto,
J apan).
Sp ectr a l Mea su r em en t. UV absorption spectra were mea-
sured with a Tosoh PD-8020 photodiode array detector and a
Beckman DU 640 spectrophotometer. 1H NMR and 13C NMR
spectra were taken of solutions in chloroform-d and DMSO-d6
with a J EOL J NM-GSX270 (1H, 270 MHz; 13C, 67.5 MHz) or a
J NM-GSX500 (1H, 500 MHz,; 13C, 125 MHz) Fourier transform
spectrometer. Chemical shifts are shown in parts per million
using tetramethylsilane as an internal standard. High-resolu-
tion mass spectra were measured using a J EOL J MS-DX300
or a J EOL J MS-AX505w mass spectrometer, equipped with a
direct inlet system. Fast atom bombardment mass spectra (FAB-
MS, matrix, m-nitrobenzyl alcohol) were measured with a J EOL
J MS-SX103 mass spectrometer. All melting points were ob-
tained on a Yazawa micro melting point apparatus (5Y-1) and
are given as uncorrected values.
Syn th esis of 2-[(2-Br om o-4,6-d in itr oph en yl)a zo]-4-m eth -
oxy-5-[(2-h yd r oxyeth yl)a m in o]a ceta n ilid e (AZO DYE-3).
The azo coupling reaction of 2-bromo-4,6-dinitrobenzenediazo-
nium sulfate [prepared from 2-bromo-4,6-dinitroaniline (2.9 g,
11 nmol; Aldrich Co., Milwaukee, WI), sodium nitrite (760 mg,
11 mmol), and concentrated sulfuric acid (20 mL)] with 3-[(2-
hydroxyethyl)amino]-4-methoxyacetanilide (2.2 g, 10 mmol) was
carried out as described previously (6) to yield a green powder
(4.2 g, 85%). Mp: 254-256 °C. HRMS for C17H17BrN6O7 m/z:
found, 496.0332; calcd, 496.0341. 1H NMR (DMSO-d6): δ 2.20
(3H, s, COCH3), 3.39 (2H, q, J ) 5.7 Hz, CH2), 3.65 (2H, q, J )
5.7 Hz, CH2), 4.95 (1H, t, J ) 5.7 Hz, OH), 7.16 (1H, s, ArH),
7.70 (1H, t, J ) 5.7 Hz, NH), 7.78 (1H, s, ArH), 8.64 (1H, d, J
) 2.0 Hz, ArH), 8.65 (1H, d, J ) 2.0 Hz, ArH)., 9.1 (1H, br,
CONH). 13C NMR (DMSO-d6): δ 24.60, 45.29, 55.44, 58.66,
97.41, 115.40, 119.53, 121.24, 128.18, 130.11, 132.14, 140.11,
142.50, 144.26, 147.04, 149.85, 168.29.
Syn th esis of 3-[(2-Hyd r oxyeth yl)a m in o]-4-m eth oxyn i-
tr oben zen e. A solution of 3-amino-4-methoxynitrobenzene
(15.6 g, 0.1 mol; Tokyo Kasei Co., Tokyo, J apan), benzaldehyde
(15.9 g, 0.15 mol), and p-toluenesulfonic acid (0.1 g) in benzene
(200 mL) was heated under reflux in a Dean-Stark apparatus
for 3 h with azeotropic distillation of water. The solvent was
removed under reduced pressure to yield a crude product,
N-benzylidene-2-methoxy-5-nitroaniline, to which was added
2-iodoethanol (25.8 g, 0.15 mol; Tokyo Kasei Co.) . The mixture
was stirred at 80 °C for 5 h. After the mixture was cooled, 100
mL of water was added, and the mixture was heated at 100 °C
with stirring for 1 h. The aqueous layer was separated, and the
organic layer was extracted with 2 N HCl. To remove the
resultant benzaldehyde, we extracted the combined aqueous
layer three times with 50 mL of ether, alkalinized the aqueous
layer with sodium hydroxide, and then extracted it with
dichloromethane. The organic solution was dried over magne-
sium sulfate and concentrated under reduced pressure to yield
an oily residue, which was subjected to column chromatography
on silica gel (E. Merck Co., Darmstadt, Germany) (eluent, 2%
methanol in chloroform) to yield a solid (9.54 g, 45%). MS m/z:
212 (M+). 1H NMR (CDCl3): δ 1.82 (1H, t, J ) 5.2 Hz, OH),
3.39 (2H, q, J ) 5.2 Hz, CH2), 3.92 (2H, q, J ) 5.2 Hz, CH2),
3.95 (3H, s, OCH3), 4.77 (1H, br, t, NH), 6.77 (1H, d, J ) 8.8
Hz, ArH), 7.41 (1H, d, J ) 2.9 Hz, ArH), 7.65 (1H, dd, J ) 2.9,
8.8 Hz, ArH). 13C NMR (CDCl3): δ 45.44, 56.05, 60.98, 103.90,
108.07, 113.59, 138.36, 142.42, 151.86.
Syn t h esis of 2-[2-(Acet yla m in o)-4-[(2-h yd r oxyet h yl)-
a m in o]-5-m eth oxyp h en yl]- 6-a m in o-4-br om o-2H-ben zotr i-
a zole (n on -ClP BTA-3). Sodium hydrosulfite (10 g; Kanto
Chemical Co., Tokyo, J apan) was added in small portions to a
vigorously stirred solution of AZO DYE-3 (5 g, 10 mmol) in 500
mL of mixed solvent (3:1:1 tetrahydrofuran/methanol/water) at
room temperature. The mixture was stirred for an additional
hour, insoluble material was filtered off, and the resulting
mixture was washed with tetrahydrofuran. The filtrate was
concentrated to 1/3 of its original volume under reduced pressure,
and the solution was extracted with dichloromethane. The
organic solution was washed with brine and dried over mag-
nesium sulfate. Removal of the solvent under reduced pressure
yielded a dark-colored oil, which was subjected to column
chromatography on silica gel (eluent, 2% methanol in chloro-
form) and Sephadex LH-20 (eluent, methanol) to yield a yellow
powder (230 mg, 5.3%). Mp: 172-174 °C. HRMS for C17H19
-
BrN6O3 m/z: found, 434.0704; calcd, 434.0701. UV max
(MeOH): 220, 264, 393 nm. 1H NMR (DMSO-d6): δ 2.03 (3H,
s, COCH3), 3.19 (2H, q, J ) 5.5 Hz, CH2), 3.68 (2H, q, J ) 5.5
Hz, CH2), 3.89 (3H, s, OCH3), 4.83 (1H, t, J ) 5.5 Hz, OH or
NH), 5.27 (1H, t, J ) 5.5 Hz, OH or NH), 5.64 (2H, s, NH2),
6.72 (1H, d, J ) 1.7 Hz, ArH), 7.23 (1H, d, J ) 1.7 Hz, ArH),
7.28 (1H, s, ArH), 7.36 (1H, s, ArH), 10.22 (1H, br, CONH). 13C
NMR (DMSO-d6): δ 24.13, 45.10, 55.73, 59.13, 92.34, 103.70,
105.03, 109.11, 119.26, 123.25, 125.22, 137.66, 138.77, 142.70,
145.33, 148.61, 167.86.
Syn th esis of 3-[(2-Hyd r oxyeth yl)a m in o]-4-m eth oxya c-
eta n ilid e. A solution of trifluoroacetic anhydride (12.6, 60
mmol; Wako Pure Chemical Industries Co., Osaka, J apan) in
dichloromethane (10 mL) was added dropwise to a stirred
mixture of 3-[(2-hydroxyethyl)amino]-4-methoxynitrobenzene
(4.2 g, 20 mmol) and potassium carbonate (6 g) in dichlo-
romethane (100 mL) at room temperature. After the mixture
was stirred for 1 h, insoluble material was filtered off. The
filtrate was washed with cooled brine and dried over magnesium
sulfate. After removal of the solvent under reduced pressure,
an oily residue including the N,O-bistrifluoroacetylated ni-
trobenzene derivative was used in the next step without further
purification.
Syn t h esis of 2-[2-(Acet yla m in o)-4-[(2-h yd r oxyet h yl)-
a m in o]- 5-m eth oxyp h en yl]-5-a m in o-7-br om o-4-ch lor o-2H-
ben zotr ia zole (P BTA-3). An aqueous solution of sodium
hypochlorite (1 mL, available chlorine minimum of 1%; Wako
Co.) was added dropwise to a stirred solution of non-ClPBTA-3
(130 mg, 0.3 mmol) in dichloromethane (30 mL) at room
temperature. After the solution was stirred for 5 min, the
organic layer was separated, washed with brine, and dried over
magnesium sulfate. The solvent was removed under reduced
pressure to yield a pale yellow solid, which was subjected to
medium-pressure liquid chromatography (Nippon Seimitsu Ka-
gaku Co., Tokyo, J apan) with a silica gel column (eluent, 2%
methanol in chloroform) and then with an ODS silica gel column
(eluent, 70% methanol) to yield a yellow powder (105 mg, 75%).
Mp: 231-234 °C. UV max (MeOH): 219, 264 (sh), 381 nm.
FABMS m/z: 468, 470, 472 (M+, 74:100:29 relative abundance).
1H NMR (DMSO-d6): δ 1.99 (3H, s, COCH3), 3.19 (2H, q, J )
5.6 Hz, CH2), 3.65 (2H, t, J ) 5.6 Hz, CH2), 3.89 (3H, s, OCH3),
A mixture of the crude product described above and 5%
palladium on charcoal (330 mg; Wako Co.) in ethyl acetate (100
mL) was stirred overnight under a hydrogen atmosphere. The
catalyst was removed by filtration. Potassium carbonate (1 g)
was added to the filtrate, and then acetic anhydride (3 g, 30
mmol) was added dropwise to the stirred mixture in an ice bath.
After being stirred for an additional hour, the solvent was
removed under reduced pressure to yield an oily residue
including the N,O-bistrifluoroacetylated acetanilide derivative,
to which methanol (20 mL) and potassium carbonate (0.5 g) were
added. The mixture was stirred for 2 h in an ice bath. After

538 Chem. Res. Toxicol., Vol. 13, No. 7, 2000
Shiozawa et al.
F igu r e 3. UV absorption spectrum of the mutagen on the
second YMC-Pack ODS-A 303 column with a photodiode array
detector. The material was eluted with 38% acetonitrile in 25
mM phosphate buffer (pH 2.0).
F igu r e 2. Purification of a new PBTA-type mutagen by HPLC.
Mutagenic fractions from a YMC-Pack ODS-A 303 column with
retention times of 16-17 min were purified on a CAPCELL PAK
C18 ODS column. The mutagen was obtained at a retention time
of 31 min. The UV absorbance and mutagenicity are shown by
the upper line and lower bar, respectively.
4.87 (1H, br, OH or NH), 5.33 (1H, t, J ) 5.6 Hz, OH or NH),
5.93 (2H, s, NH2), 7.13 (1H, s, ArH), 7.32 (1H, s, ArH),7.40 (1H,
s, ArH), 10.03 (1H, br, CONH). 13C NMR (DMSO-d6): δ 23.81,
45.05, 55.79, 59.10, 96.16, 104.23, 105.34, 108.00, 119.74, 122.66,
125.41, 137.55, 139.10, 142.87, 142.90, 144.09, 167.97.
Mu ta gen esis Assa y. All test samples were dissolved in 100
µL aliquots of 50% dimethyl sulfoxide, and mutagenicity was
examined by the preincubation method (15) using S. typhimu-
rium YG1024 in the presence of an S9 mix. S. typhimurium
YG1024, produced by introducing plasmids containing the
acetyltransferase gene from TA1538 into TA98 (16) was kindly
provided by T. Nohmi from the National Institute of Health
Sciences, Tokyo. The S9 mix contained 5 µL of S9 (25 mg of
protein/mL), prepared from livers of male Sprague-Dawley rats
intraperitoneally administered phenobarbital and 5,6-benzofla-
vone, at a total volume of 500 µL. Mutagenic activities of test
samples were calculated from the linear portions of the dose-
response curves, obtained with four or five doses, and duplicate
plates, in two independent experiments.
F igu r e 4. 1H NMR spectrum of the mutagen in DMSO-d6.
the blue rayon-adsorbed material. Figure 3 shows the UV
absorption spectrum of this mutagen, obtained on the
second YMC-Pack ODS-A 303 column with a photodiode
array detector. UV absorption maxima were observed at
219, 264 (sh), and 381 nm. Several other mutagenic
fractions were also observed at the second HPLC step
on a YMC-Pack ODS-A 303 column; however, each of
their contributions to the total mutagenicity was much
smaller than that of the above isolated compound.
With the new mutagen isolated from the adsorbate to
blue rayon as a marker, about 120 µg of the mutagen
could be isolated by Sephadex LH-20 column chroma-
tography and HPLC on semipreparative ODS columns
from materials adsorbed to 3 kg of blue cotton.
Resu lts a n d Discu ssion
Isola tion of a Mu ta gen fr om Wa ter of th e Nik k o
River . The materials adsorbed to 1 g of blue rayon from
the water of the Nikko River were found to induce
3 900 000 revertants of S. typhimurium YG1024 in the
presence of an S9 mix. Mutagens extracted from 25 g of
blue rayon were separated by HPLC on a semiprepara-
tive YMC-Pack ODS-AM 324 column using a gradient
solvent system. Many fractions showed mutagenicity, and
the fractions recovered at this HPLC step showed 85%
of the total mutagenicity. Among them, the most potent
mutagenic activity was observed in the fraction with
retention times of 13-20 min, and this fraction accounted
for 34% of the total mutagenicity of the blue rayon-
adsorbed material. Mutagenic activity in each fraction
with other retention times was much lower than that in
the fraction with retention times of 13-20 min. Two
minor mutagenic fractions were also observed at reten-
tion times of 24-28 and 20-24 min, which corresponded
to those of PBTA-1 and PBTA-2, accounting for 1.2 and
4.5% of the total mutagenicity, respectively.
St r u ct u r a l An a lysis of t h e Mu t a gen . The UV
absorption spectrum of the new mutagen shown in Figure
3 is similar to those of PBTA-1 and PBTA-2 (6, 8).
Analysis using high-resolution mass spectrometry indi-
cated the molecular formula to be C17H18BrClN6O3 (m/z:
found, 468.0320; calcd, 468.0312). Compared with the
structure of PBTA-1, the mutagen has four fewer carbon
atoms, eight fewer protons, and one fewer oxygen atom
but the same number of nitrogen atoms.
1
Figure 4 shows the H NMR spectrum of the mutagen
in DMSO-d6, indicating the presence of 18 protons in the
molecule, which is consistent with the results of high-
resolution mass spectrometry. Aromatic protons and
several other protons were observed at chemical shifts
similar to those of PBTA-1 as shown in Figure 5,
suggesting the presence of similar substituents at the
same positions. For example, broad singlets at 5.93 (2H)
and 10.03 ppm (1H) indicated the presence of exchange-
able protons due to a primary amino and an amido group,
respectively. Two singlets at 1.99 (3H) and 3.89 ppm (3H)
suggested acetyl and methoxy groups, respectively. The
A mutagenic fraction with retention times of 13-20
min was successively purified by HPLC on a YMC-Pack
ODS-A 303 column and a CAPCELL PAK C18 ODS
column. A single UV absorption peak exhibiting mutage-
nicity was observed at a retention time of 31 min on the
CAPCELL PAK C18 ODS column (Figure 2), and this
mutagen accounted for 11% of the total mutagenicity of

Identification of a River Mutagen, PBTA-3
Chem. Res. Toxicol., Vol. 13, No. 7, 2000 539
Sch em e 2. Syn th esis of P BTA-3
NMR (DMSO-d6) spectrum, aromatic protons were ob-
served as two meta doublets (J ) 1.7 Hz) at 6.72 and
7.23 ppm of the benzotriazole ring protons and two
singlets at 7.28 and 7.36 ppm of the 2-phenyl moiety. Two
quartets (J ) 5.5 Hz) due to ethylene protons were
observed at 3.19 and 3.68 ppm. Protons of acetamide and
primary amine were noted at 10.22 (br) and 5.64 ppm
(s), respectively. Two triplets (J ) 5.5 Hz) of 5.27 and
4.83 ppm were assigned to protons of secondary amino
and hydroxy groups. Two singlets at 2.03 and 3.89 ppm
were due to methyls of acetyl and methoxy groups,
respectively. On the basis of these findings and the 13C
NMR spectrum, this compound was determined to be
non-ClPBTA-3.
F igu r e 5. Chemical structures of PBTA-1 and PBTA-3. 1H
NMR chemical shifts in DMSO-d6 are presented as ppm. s )
singlet, t ) triplet, q ) quartet, br ) broad.
Sch em e 1. Syn th esis of AZO DYE-3
The monochlorinated product, PBTA-3, was produced
at a yield of 75% from a nonchlorinated derivative of
PBTA-3 (non-ClPBTA-3) by treatment with sodium hy-
pochlorite. The UV, MS, and 1H NMR spectra of the
prepared compound matched those of the mutagen,
PBTA-3, isolated from water samples from the Nikko
River. The 13C NMR spectrum was also consistent. Thus,
we concluded that the third PBTA-type mutagen isolated
from the Nikko River is PBTA-3.
absorbance pattern due to aromatic protons at 7.13, 7.32,
and 7.40 ppm was similar to that of PBTA-1. On the other
hand, the triplets at 5.33 (1H) and 4.87 ppm (1H) and
quartets at 3.19 (2H) and 3.65 ppm (2H) predicted the
presence of a hydroxyethylamine moiety (NHCH2CH2-
OH), where both hydroxy and amine protons display
coupling with the adjacent methylene protons instead of
with N(CH2CH2OCH3)2 for PBTA-1.
Detection of P BTA-3 in Wa ter fr om River s Oth er
Th a n th e Nik k o River . To investigate the distribution
of PBTA-3 in water of other rivers in J apan, we collected
water samples as blue rayon-adsorbed products from the
Asuwa River flowing through Fukui Prefecture and the
Katsura and Nishitakase rivers in Kyoto Prefecture.
These rivers flow through the regions where dyeing
industries have been developed. The blue rayon extracts
were separated by HPLC on a YMC-Pack ODS-AM 324
column. Fractions with retention times of 19-20 min
were further analyzed with CAPCELL PAK C18 and
YMC-Pack ODS-A 303 columns. The retention times of
single peaks observed with both columns were identical
to those of PBTA-3, and the UV-spectra also matched.
The concentrations of PBTA-3 detected in the Asuwa,
Katura, and Nishitakase rivers were 38, 35, and 22 ng/g
of blue rayon, respectively. Using the same method, we
found the level of PBTA-3 in the Nikko River to be 140
ng/g of blue rayon.
In the present study, we identified a new mutagen,
PBTA-3, as a major PBTA-type mutagen in water of the
Nikko River. PBTA-3 induced 81 000 and 3 000 000
revertants/µg of S. typhimurium TA98 and YG1024,
respectively; it is a strain originating from TA98 with a
high O-acetyltransferase activity in the presence but not
in the absence of an S9 mix. This mutagenic potency is
comparable to those of PBTA-1 and PBTA-2 (6, 8).
Furthermore, we successfully synthesized PBTA-3 from
the azo compound, AZO DYE-3, through non-ClPBTA-3.
Although AZO DYE-3 is not listed as an industrial
material, as are AZO DYE-1 (9) and AZO DYE-2 (8), it
On the basis of results of the spectral analyses de-
scribed above, the chemical structure of the mutagen was
deduced to be PBTA-3 (Figure 5). To further confirm its
structure, we conducted synthesis from 2-[(2-bromo-4,6-
dinitrophenyl)azo]-4-methoxy-5-[(2-hydroxyethyl)amino]-
acetanilide (AZO DYE-3) by the method used for PBTA-1
(9). AZO DYE-3 was prepared by the coupling reaction
of 2-bromo-4,6-dinitrobenzenediazonium sulfonate with
3-[(2-hydroxyethyl)amino]-4-methoxyacetanilide prepared
as shown in Scheme 1. The molecular formula, C17H17
-
BrN6O7, of synthesized AZO DYE-3 was determined by
high-resolution mass spectrometry (m/z: found, 496.0332;
1
calcd, 496.0341). In the H NMR (DMSO-d6) spectrum,
peaks due to four aromatic protons were observed as two
singlets at 7.16 and 7.78 ppm and two meta doublets
(J ) 2.0 Hz) at 8.64 and 8.65 ppm, suggesting that azo
coupling occurred at the 6 position of the acetanilide
derivative. Two methylene protons appeared as A2B2X
system quartets at 3.39 and 3.65 ppm. Peaks due to
protons of acetamide, amino, and hydroxyl groups were
observed at 9.1 (br), 7.70 (t), and 4.95 ppm (t), respec-
tively.
As shown in Scheme 2, sodium hydrosulfite reduction
of AZO DYE-3 yielded a 2-phenylbenzotriazole skeleton
product, with a UV absorption pattern similar to that of
PBTA-1, at a yield of 5.3%. The high-resolution mass
spectrum of this product indicated a molecular formula
of C17H19BrN6O3 (m/z: found, 434.0704; calcd, 434.0701),
which corresponded to a compound in which a chlorine
1
atom of PBTA-3 is replaced with a hydrogen. In the H

540 Chem. Res. Toxicol., Vol. 13, No. 7, 2000
Shiozawa et al.
bates to
a
copper-phthalocyanine derivative recovered from
is possible that it is formed by dehydroxyethylation of
the azo dye, 2-[(2-bromo-4,6-dinitrophenyl)azo]-4-meth-
oxy-5-[bis(2-acetoxyethyl)amino]acetanilide (Color Index
Name, Disperse Blue 79:1), which is very commonly used
in dyeing factories. Similar dehydroxyethylation was
reported to occur for tri-, di-, and monoethanolamines by
treatment with hypochlorous acid or chlorine dioxide (17).
We also detected a small amount of the acetic acid ester
of AZO DYE-3 in commercially available Disperse Blue
79:1, suggesting contamination of AZO DYE-3 or its ester
in the process of industrial production.
The reduction agent sodium hydrosulfite is generally
used in dyeing factories for discharge printing and
bleaching, while the chlorination reagent sodium hy-
pochlorite is employed in sewage plants for disinfection.
AZO DYE-3 is thought to be converted to the correspond-
ing 2-phenylbenzotriazole derivative, non-ClPBTA-3, with
reducing reagents such as hydrosulfite during industrial
dyeing processes. Therefore, PBTA-3 could be produced
from non-ClPBTA-3 by subsequent chlorination during
disinfection in sewage plants before being discharged into
rivers. Since various dinitrophenylazo compounds are
known to be used as industrial materials, it is plausible
that other PBTA-type mutagens are also formed from azo
dyes and contaminate the river water near dyeing
factories. Quantification of the levels of these mutagens
in the water of various rivers and determination of their
carcinogenic effects as well as genotoxic activities in
organisms other than Salmonella strains are necessary
to fully estimate the impact of PBTA compounds on
human health. Moreover, it is important to design new
methods that prevent the formation of PBTA compounds
and enhance its degradation.
municipal river water. Mutat. Res. 300, 207-213.
(5) Kreijl, C. F., and Slooff, W. (1985) Mutagenic activity in Dutch
river water and its biological significance for fish. In Mutagenicity
Testing in Environmental Pollution Control (Zimmermann, F. K.,
and Taylor-Mayer, R. E., Eds.) pp 86-104, Ellis Horwood, West
Sussex. U.K.
(6) Nukaya, H., Yamashita, J ., Tsuji, K., Terao, Y., Ohe, T., Sawan-
ishi, H., Katsuhara, T., Kiyokawa, K., Tezuka, M., Oguri, A.,
Sugimura, T., and Wakabayashi, K. (1997) Isolation and chemical-
structural determination of a novel aromatic amine mutagen in
water from the Nishitakase River in Kyoto. Chem. Res. Toxicol.
10, 1061-1066.
(7) Kusamran, W. R., Wakabayashi, K., Oguri, A., Anong, T., Nagao,
M., and Sugimura, T. (1994) Mutagenicities of Bangkok and
Tokyo river waters. Mutat. Res. 32, 99-104.
(8) Oguri, A., Shiozawa, T., Terao, T., Nukaya, H., Yamashita, J .,
Ohe, T., Sawanishi, H., Katsuhara, T., Sugimura, T., and Waka-
bayashi, K. (1998) Identification of
(PBTA)-type mutagen, PBTA-2, in water from the Nishitakase
River in Kyoto. Chem. Res. Toxicol. 11, 1195-1200.
a 2-phenylbenzotriazole
(9) Shiozawa, T., Muraoka, K., Nukaya, H., Ohe, T., Sawanishi, H.,
Oguri, A., Wakabayashi, K., Sugimura, T., and Terao, Y. (1998)
Chemical synthesis of a novel aromatic amine mutagen isolated
from water of the Nishitakase River in Kyoto, and possible route
of its formation. Chem. Res. Toxicol. 11, 375-380.
(10) Ohe, T., Takeuchi, N., Watanabe, T., Tada, A., Nukaya, H., Terao,
Y., Sawanishi, H., Hirayama, T., Sugimura, T., and Wakabayashi,
K. (1999) Quantification of two aromatic amine mutagens,
PBTA-1 and PBTA-2, in the yodo river system. Environ. Health
Perspect. 107, 701-704.
(11) Shiozawa, T., Suyama, K., Nakano, K., Nukaya, H., Sawanishi,
H., Oguri, A., Wakabayashi, K., and Terao, Y. (1999) Mutagenic
activity of 2-phenylbenzotriazole derivatives related to a mutagen,
PBTA-1, in river water. Mutat. Res. 442, 105-111.
(12) Takahashi, M., Wakabayashi, K., Nagao, M., Yamamoto, M.,
Masui, T., Goto, T., Kinae, N., Tomita, I., and Sugimura, T. (1985)
Quantification of 2-amino-3-methylimidazo[4,5-f]quinoline (IQ)
and 2-amino-3,8-dimethylimidazo[4,5-f]-quinoxaline (MeIQx) in
beef extracts by liquid chromatography with electrochemical
detection (LCEC). Carcinogenesis 6, 1195-1199.
Ack n ow led gm en t. This study was supported by
grants-in-aid for cancer research from the Ministry of
Health and Welfare of J apan and was funded under a
contract with the Environment Agency of J apan.
(13) Kira, S., Hayatsu, H., and Ogata, M. (1989) Detection of mutage-
nicity in mussels and their ambient water. Bull. Environ. Contam.
Toxicol. 43, 583-589.
(14) Hayatsu, H., Oka, T., Wakata, A., Ohara, Y., Hayatsu, T.,
Kobayashi, H., and Arimoto, S. (1983) Adsorption of mutagens
to cotton bearing covalently bound trisulfo-copper-phthalocyanine.
Mutat. Res. 119, 233-238.
Refer en ces
(15) Yahagi, T., Nagao, M., Seino, Y., Matsushima, T., Sugimura, T.,
and Okada, M. (1977) Mutagenicities of N-nitrosamines on
Salmonella. Mutat. Res. 48, 121-130.
(16) Watanabe, M., Ishidate, M., J r., and Nohmi, T. (1990) Sensitive
method for the detection of mutagenic nitroarenes and aromatic
amines: new derivatives of Salmonella typhimurium tester
strains possessing elevated O-acetyltransferase levels. Mutat. Res.
234, 337-348.
(1) International Agency for Research on Cancer (1991) Chlorinated
drinking-water. In IARC Monographs on the Evaluation of
Carcinogenic Risks to Humans, Vol. 52, pp 45-141, IARC, Lyon,
France.
(2) Holmbom, B. R., Voss, R. H., Mortimer, R. D., and Wong, A. (1981)
Isolation and identification of an Ames-mutagenic compound
present in kraft chlorination effluents. Tappi 64, 172-174.
(3) Hemming, J ., Holmbom, B., Reunanen, M., and Kronberg, L.
(1986) Determination of the strong mutagen 3-chloro-4-(dichlo-
romethyl)-5-hydroxy-2(5H)-furanone in chlorinated drinking and
humic waters. Chemosphere 15, 549-556.
(17) Dennis, W. H., J r., Hull, L. A., and Rosenblatt, D. H. (1967)
Oxidation of amines. IV. Oxidative fragmentation, J . Org. Chem.
32, 3783-3787.
(4) Sayato, Y., Nakamuro, K., Ueno, H., and Goto, R. (1993) Identi-
fication of polycyclic aromatic hydrocarbons in mutagenic adsor-
TX0000264