Effects of oligofluorine substitution on the mutagenicity of quinoline: A study with twelve fluoroquinoline derivatives

Article abstract of DOI:10.1016/S1383-5718(98)00188-0

A total of 12 variously fluorinated derivatives of quinoline (Q) were tested for their mutagenicity in Salmonella typhimurium TA100 in the presence of S9 mix to investigate the structure-mutagenicity relationship in oligofluorinated quinolines. Nine of th

Full text of DOI:10.1016/S1383-5718(98)00188-0

Ž
.
Mutation Research 439 1999 149–157
Effects of oligofluorine substitution on the mutagenicity of
quinoline: a study with twelve fluoroquinoline derivatives
a
a
b,1
Taka-aki Kato , Ken-ichi Saeki a,), Yutaka Kawazoe , Atsushi Hakura
a
Faculty of Pharmaceutical Sciences, Nagoya City UniÕersity, Tanabedori, Mizuho-ku, Nagoya 467-8603, Japan
b
Drug Safety Research Laboratories, Eisai, 1-3 Tokodai, 5-chome, Tsukuba 300-2635, Japan
Received 10 August 1998; revised 8 December 1998; accepted 9 December 1998
Abstract
A total of 12 variously fluorinated derivatives of quinoline
Ž .
Q
were tested for their mutagenicity in Salmonella
typhimurium TA100 in the presence of S9 mix to investigate the structure–mutagenicity relationship in oligofluorinated
quinolines. Nine of them, 3,7-di-, 5,6-di-, 6,7-di-, 6,8-di-, 7,8-di-, 3,5,7-tri-, 5,6,8-tri-, 6,7,8-tri-, and 5,6,7,8-tetrafluoro-
Ž
.
quinolines FQs , were newly synthesized for this purpose. Those fluorinated at position 3 were all non-mutagenic.
Ž
Mutagenicity was enhanced by fluorine-substitution at position 5 or 7, but not in 3-FQs i.e., 3,5-di-, 3,7-di-, and
.
3,5,7-triFQs . Some of the 6-fluorinated derivatives showed less maximum induced-revertants with more mutagenic
potencies in terms of induced-revertants per dose than quinoline. No marked change occurred by fluorine-substitution at
position 8. These results show that the effect of di- and trifluoro-substitution on mutagenicity is generally additive, while that
of tetrafluorination approaches the deactivating effect of perfluorination. Our study suggests that 3-fluorine-substitution in
the pyridine moiety may be a useful means of antimutagenic structural modification in pyridine-fused aromatic chemicals for
medicinal and agricultural use. q 1999 Elsevier Science B.V. All rights reserved.
Keywords: Quinoline; Fluoroquinoline; Mutagenicity; Ames test; Salmonella typhimurium
Ž
1. Introduction
sible for the mutagenic modification of DNA Chart
. w
x
2–7 . This is supported by the fact that muta-
1
Ž .
Quinoline Q is a potent mutagen in Salmonella
typhimurium TA100 in the presence of rat liver
genicity was eliminated by fluorine-substitution at
position 3 and enhanced by the same at position 5
2,3 . Thus, the former fluorine must inhibit the
oxidative activation in the pyridine moiety due to its
electron-withdrawing nature 8–11 and the latter
fluorine would inhibit the oxidative detoxification at
position 5 to result in a higher chance of allowing
quinoline to be activated. It is worth mentioning that
3-fluoroquinoline lacks the potency to induce both
the appearance of liver GST-P-positive foci in rats
and the in vivo mutation in the liver of the transgenic
w x
microsomal enzymes 1 . As has previously been
w
x
proposed, metabolic oxidation takes place in the
pyridine moiety of quinoline to produce the activated
form, the 1,4-hydrated 2,3-epoxide, which is respon-
w
x
)
Corresponding author. Tel.: q81-52-836-3410; Fax: q81-52-
834-9309; E-mail: saeki@phar.nagoya-cu.ac.jp
1 Also corresponding author. Fax: q81-298-47-2037; E-mail:
a1-hakura@eisai.co.jp
1383-5718r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S1383-5718 98 00188-0

(
)
150
T.-a. Kato et al.rMutation Research 439 1999 149–157
Ž
.
MutaeMouse, whereas 5-fluoroquinoline was as car-
cinogenic and mutagenic as quinoline in both the
residue 982 mg was heated in 20 ml of c-HCl and
20 ml of MeOH at 1008C for 3 h. The reaction
mixture was poured into water, neutralized with
Na2CO3, and extracted with CHCl3. The CHCl3
layer was dried over anhydrous MgSO4 and evapo-
rated. Purification of extract by column chromatogra-
w
x
above in vivo assay systems 12,13 . It was, there-
fore, expected that fluorine-substitution at an appro-
priate position could be an anticarcinogenic and an-
timutagenic chemical modification of genotoxic N-
containing heteroaromatic compounds such as quino-
line. In the present study, some positional isomers of
mono-, di-, and trifluorinated quinolines were pre-
pared and tested for mutagenicity in order to obtain a
more comprehensive view of fluorine-substitution
effects on the mutagenicity of quinoline.
Ž
.
phy aluminum oxide, CHCl3 yielded 3-Br-7-FQ in
Ž
.
65% yield from BrCN . mp 72-748C. MS mrz: 225
q
q
Ž
Ž
Ž
.
M
, 227 M q2 . 1H-NMR CDCl3 d: 7.38
Ž . Ž .
.
Ž
.
Ž
.
ddd H-6 , 7.74 m H-5 and H-8 , 8.31 d H-4 , 8.91
.
d H-2 ; J2 – 4 s2.4, J5– 6 s8.5, J6 – 7F s10.7 Hz.
Anal. Calcd for C9 H5BrFN: C, 47.82; H, 2.23; N,
6.20. Found: C, 47.52; H, 2.29; N, 6.11.
Ž
.
3-Br-7-FQ 497 mg was allowed to react with 3
Ž
.
2. Materials and methods
ml of c-NH3 aq. and CuSO4 P5H2O 55 mg in a
sealed tube at 1608C for 20 h. The reaction mixture
was poured into 50 ml of H2O and extracted with
CHCl3. The organic layer was washed with water,
2.1. Materials
Ž
Fluorinated anilines were purchased from Aldrich.
Tetraethylammonium fluoride-5.5 mol eq. of hydro-
gen fluoride was from Morita Chemical Industries,
Osaka, Japan. Melting points were determined with a
Yamato MP-500D micro melting point apparatus
without correction. Mass spectra were measured with
a JEOL AX505HA spectrometer. 1H-NMR spectra
were recorded with a JEOL JNM-EX 270 or a
JNM-GSX 400 spectrometer in CDCl3 or DMSO-d6
using tetramethylsilane as an internal standard.
dried over anhydrous MgSO4 and evaporated 280
.
mg, crude 3-NH2-7-FQ . The crude 3-NH2-7-FQ
195 mg was suspended in 3 ml of tetraethyl-
ammonium fluoride-5.5 mol eq. of hydrogen fluo-
ride. Sodium nitrite 111 mg was added to the
solution, and was kept stirring for 30 min at 08C.
Then the solution was kept stirring for 1 h at 958C.
The reaction mixture was poured into 100 ml of
water and neutralized with sodium carbonate and
extracted with CHCl3. The organic layer was dried
over anhydrous MgSO4 and evaporated. Purification
of extract by column chromatography aluminum
oxide, CHCl3 yielded 3,7-diFQ in 67% yield. mp
Ž . Ž . Ž
858C sublimated . H-NMR CDCl3 d: 7.40 ddd
H-6 , 7.78 m H-4, H-5, and H-8 , 8.83 d H-2 ;
J2 – 4 s2.9 Hz. HR-MS mrz: 165.039, Calcd for
C9 H5F2 N: 165.039.
Ž
.
Ž
.
Ž
.
Quinoline CAS Registry No. 91-22-05 was pur-
chased from Sigma, MO, USA. 7-F, 5,7-diF, and
Ž
Ž
.
5,8-difluoroquinoline Registry Nos. 396-32-7,
1
.
34522-72-0, and 16650-32-1, respectively were syn-
thesized according to the reported methods 14,15 .
w
x
.
Ž
.
Ž
.
(
)
2.1.1. 3,7-Difluoroquinoline 3,7-diFQ
3,7-Difluoroquinoline was synthesized from 7-flu-
oroquinoline via 3-amino-7-fluoroquinoline, which
was synthesized by the same method as for the
(
)
2.1.2. 5,6-Difluoroquinoline 5,6-diFQ
Ž
.
3,4-Difluoroaniline 1.07 g was allowed to react
w
x
Ž
.
synthesis of 3-aminoquinoline from quinoline 16 .
with Br2 0.51 ml, 1.2 eq in 50 ml of MeOH at r.t.
for 4 h. The reaction mixture was poured into water
and neutralized with K2CO3 and extracted with
CHCl3. The organic layer was dried over anhydrous
MgSO4 and evaporated. Purification of extract by
Ž
.
7-Fluoroquinoline 1.01 g was allowed to react with
Ž
.
BrCN 560 mg, 0.78 eq of 7-FQ in 20 ml of MeOH
at room temperature for 4 days. N-Bromoacetamide
Ž
.
821 mg 1.1 eq of BrCN was added to the reaction
Ž
mixture at r.t. After stirring at r.t. for 24 h, the
reaction mixture was poured into ice water. The
resulting mixture was extracted with CHCl3. The
organic layer was washed with 1 N HCl, dried over
column chromatography aluminum oxide, hexane:
benzenes1: 1™benzene™benzene: methanols
.
19:1 yielded 2-bromo-4,5-difluoroaniline in 42%
1
Ž
.
Ž
yield. mp 41-448C. H-NMR CDCl3 d: 4.00 s
. Ž . Ž .
NH2 , 6.57 dd H-6 , 7.25 d H-3 ; J4F – 6 s7.1,
Ž
.
anhydrous MgSO4 and evaporated 1.27 g . The

(
)
T.-a. Kato et al.rMutation Research 439 1999 149–157
151
J5F – 6 s11.7 Hz. Anal. Calcd for C6 H4 BrF2 N: C,
34.65; H, 1.94; N, 6.73. Found: C, 34.26; H, 1.91;
N, 6.24.
10.3, J5– 7F s8.4, J6F – 8 s7.7, J7F – 8 s11.2 Hz.
Anal. Calcd for C9 H5F2 N: C, 65.46; H, 3.05; N,
8.48. Found: C, 65.12; H, 3.16; N, 8.48.
Ž
.
2-Bromo-4,5-difluoroaniline 304 mg , glycerol
Ž
Ž
.
.
(
)
0.33 ml, 3 eq and sodium m-nitrobenzenesulfonate
978 mg, 3 eq were dissolved in 80% H2 SO4 10
ml , and the mixture was stirred at 1408C for 4 h.
The reaction mixture was poured into ice water 300
2.1.4. 6,8-Difluoroquinoline 6,8-diFQ
Ž
Ž . Ž
.
.
2,4-Difluoroaniline 2.0 g , glycerol 3.4 ml, 3 eq
.
Ž
and sodium m-nitrobenzenesulfonate 10.5 g, 3 eq
Ž
Ž
.
were dissolved in 80% H2 SO4 20 ml , and the
mixture was stirred at 1408C for 4 h. The reaction
mixture was poured into ice water 300 ml . The
filtrate was neutralized with aqueous NH3 and ex-
tracted with CHCl3. The CHCl3 layer was dried over
anhydrous MgSO4 and evaporated. Purification of
.
ml . The filtrate was neutralized with aqueous NH3
Ž
.
and extracted with CHCl3. The CHCl3 layer was
dried over anhydrous MgSO4 and evaporated. Purifi-
cation of extract by column chromatography
Ž
.
aluminum oxide, benzene yielded 8-bromo-5,6-
diFQ in 89% yield. mp 105-1068C. MS mrz: 243
Ž
extract by column chromatography aluminum oxide,
benzene yielded 6,8-diFQ in 39% yield. mp 64–
1
q
q
Ž
.
Ž
Ž
.
.
Ž
.
Ž
.
M
, 245 M q2 . H-NMR CDCl3 d: 7.58 dd
1
q
.
Ž
.
Ž
.
Ž
.
Ž
.
H-3 , 7.98 dd H-7 , 8.47 dd H-4 , 9.07 dd H-2 ;
J2 – 3 s 4.1, J2 – 4 s 1.4, J3 – 4 s 8.5, J5F – 7 s 7.8,
J6F – 7 s9.7 Hz. Anal. Calcd for C9 H4 BrF2 N: C,
44.30; H, 1.65; N, 5.74. Found: C, 44.44; H, 1.65;
N, 5.61.
688C. MS mrz: 165 M
.
H-NMR CDCl3 d:
Ž
.
Ž
.
Ž
7.23-7.30 m H-5 and H-7 , 7.50 dd H-3 , 8.14 d
H-4 , 8.94 d H-2 ; J2 – 3 s4.2, J3 – 4 s8.4, J4 – 8F
1.3 Hz. HR-MS mrz: 165.039, Calcd for C9 H5F2 N:
.
Ž
.
s
165.039.
Ž
.
8-bromo-5,6-difluoroquinoline 48.5 mg was hy-
drogenated with 5% Pd-C in ca. 1 M NaOH in
MeOH solution for 1 h. The suspended Pd-C was
filtered off and washed with MeOH. Purification of
(
)
2.1.5. 7,8-Difluoroquinoline 7,8-diFQ
Ž
.
Ž
2,3-Difluoroaniline 1.04 g , glycerol 1.76 ml, 3
eq and sodium m-nitrobenzenesulfonate 5.50 g, 3
eq were dissolved in 80% H2 SO4 10 ml , and the
mixture was stirred at 1408C for 4 h. The reaction
mixture was poured into ice water 300 ml . The
filtrate was neutralized with aqueous NH3 and ex-
tracted with CHCl3. The CHCl3 layer was dried over
anhydrous MgSO4 and evaporated. Purification of
.
.
Ž
Ž
Ž
.
filtrate by column chromatography silica gel,
CHCl3 yielded 5,6-diFQ in 93% yield. mp 62-668C
MS mrz: 165 M . H-NMR CDCl3 d: 7.49 dd
H-3 , 7.56 m H-8 , 7.91 m H-7 , 8.43 d H-4 , 8.93
dd H-2 ; J2 – 3 s4.1, J2 – 4 s1.4, J3 – 4 s8.5, J5F – 7
s7.8, J6F – 7 s9.7 Hz. Anal. Calcd for C9 H5F2 N:
C, 65.46; H, 3.05; N, 8.48. Found: C, 65.65; H, 3.34;
N, 8.32.
.
1
q
Ž
.
Ž
.
.
Ž
Ž
Ž
.
.
Ž
.
Ž
.
Ž
.
Ž
extract by column chromatography aluminum oxide,
.
benzene yielded 7,8-diFQ in 86% yield. mp 65-668C.
q
Ž
MS mrz: 165
M
1H-NMR CDCl3 d: 7.44
. Ž .
.
(
)
Ž
. Ž . Ž . Ž
2.1.3. 6,7-Difluoroquinoline 6,7-diFQ
3,4-Difluoroaniline 5.04 g , glycerol 8.55 ml, 3
eq and sodium m-nitrobenzenesulfonate 26.0 g, 3
eq were dissolved in 80% H2 SO4 50 ml , and the
mixture was stirred at 1408C for 4 h. The reaction
mixture was poured into ice water 300 ml . The
filtrate was neutralized with aqueous NH3 and ex-
tracted with CHCl3. The CHCl3 layer was dried over
anhydrous MgSO4 and evaporated. Purification of
extract by column chromatography aluminum oxide,
benzene and recrystallization from EtOH yielded
6,7-diFQ in 56% yield. mp 114-1158C. MS mrz:
165 M . H-NMR CDCl3 d: 7.41 dd H-3 , 7.55
ddd H-6 , 7.46 dd H-3 , 7.61 ddd H-5 , 8.18 ddd
Ž
.
Ž
.
Ž
.
H-4 9.00 dd H-2 ; J2 – 3 s4.3, J2 – 4 s1.7, J3 – 4
s
s
.
.
Ž
.
8.3, J4 – 8F s1.7, J5– 6 s8.9, J5– 7F s5.3, J5– 8F
Ž
2.0, J6 – 7F s9.2, J6 – 8F s6.9 Hz. Anal. Calcd for
C9 H5F2 N: C, 65.46; H, 3.05; N, 8.48. Found: C,
65.34; H, 3.11; N, 8.58.
Ž
.
(
)
2.1.6. 3,5,7-Trifluoroquinoline 3,5,7-triFQ
3,5,7-Trifluoroquinoline was synthesized from
5,7-difluoroquinoline via 3-amino-5,7-difluoro-
quinoline by the same method as for the synthesis of
Ž
.
Ž .
3,7-diFQ from 7-FQ. 5,7-Difluoroquinoline 1.45 g
1
q
Ž
.
Ž
.
Ž
.
Ž
was allowed to react with BrCN 560 mg, 0.6 eq of
Ž
.
Ž
.
Ž
.
Ž
.
5,7-diFQ in 20 ml of MeOH at room temperature
dd H-5 , 7.86 dd H-8 , 8.10 dd H-4 8.90 dd
.
Ž
for 4 days, N-Bromoacetamide 814.6 mg 1.1 eq of
H-2 ; J2 – 3 s4.2, J2 – 4 s1.5, J3 – 4 s8.4, J5– 6F
s

(
)
152
T.-a. Kato et al.rMutation Research 439 1999 149–157
.
Ž
. Ž . Ž .
BrCN was added to the reaction mixture at r.t. After
stirring at r.t. for 24 h, the reaction mixture was
poured into ice water. The resulting mixture was
extracted with CHCl3. The organic layer was washed
with 1 N HCl, dried over anhydrous MgSO4 and
d H-8 , 7.99 dd H-4 , 8.85 d H-2 ; J2 – 4 s2.7,
J3F – 4 s9.5 J7F – 8 s9.5 Hz. HR-MS mrz: 183.030,
Calcd for C9 H4 F3N: 183.030.
(
)
2.1.7. 5,6,8-Trifluoroquinoline 5,6,8-triFQ
Ž
.
Ž
.
Ž . Ž
2,4,5-Trifluoroaniline 496.8 mg , glycerol 0.8
ml, 3 eq and sodium m-nitrobenzenesulfonate 2.34
evaporated 1.12 g . This residue 923.0 mg was
heated in 20 ml of c-HCl and 20 ml of MeOH at
1008C for 3 h. The reaction mixture was poured into
water, and neutralized with Na2CO3, and extracted
with CHCl3. The CHCl3 layer was dried over anhy-
drous MgSO4 and evaporated. Purification of extract
.
Ž
.
Ž
.
g, 3 eq were dissolved in 80% H2 SO4 10 ml , and
the mixture was stirred at 1408C for 4 h. The reac-
Ž
.
tion mixture was poured into ice water 300 ml . The
filtrate was neutralized with aqueous NH3 and ex-
tracted with CHCl3. The CHCl3 layer was dried over
anhydrous MgSO4 and evaporated. Purification of
Ž
by column chromatography aluminum oxide,
CHCl3 yielded 3-Br-5,7-diFQ in 55% yield of
.
Ž
q
.
Ž
.
Ž
BrCN . mp 75–778C. MS mrz: 243
M
247
extract by column chromatography aluminum oxide,
M q2 . 1H-NMR CDCl3 d: 7.10 ddd H-6 ,
benzene yielded 5,6,8-triFQ in 64% yield. mp 89–
H-NMR CDCl3 d:
7.36 ddd H-7 , 7.59 dd H-3 , 8.45 ddd H-4 , 8.99
q
Ž
.
Ž
.
Ž
.
.
1
q
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
7.56 m H-8 , 8.32 d H-4 , 8.94 d H-2 ; J2 – 4 s2.2,
J6-8 s2.4 Hz. Anal. Calcd for C9 H4 BrF2 N: C,
44.30; H, 1.65; N, 5.74. Found: C, 44.04; H, 1.70;
N, 5.56.
908C. MS mrz: 183 M
.
Ž
.
Ž
.
Ž
.
Ž
.
dd H-2 ; J2 – 3 s4.4, J2 – 4 s1.5, J3 – 4 s8.8, J4 – 8F
s1.5 Hz. Anal. Calcd for C9 H4 F3N: C, 59.03; H,
Ž
.
3-Br-5,7-diFQ 499.6 mg was allowed to react
with 3 ml of c-NH3 aq and CuSO4 P5H2O 55 mg
in a sealed tube at 1608C for 20 h. The reaction
mixture was poured into 50 ml of H2O and extracted
with CHCl3. The organic layer was washed with
water, dried over anhydrous MgSO4 and evaporated
296.7 mg, crude 3-NH2-5,7-diFQ . The crude 3-
NH2-5,7-diFQ 202.8 mg was suspended in 3 ml of
tetraethylammonium fluoride-5.5 mol eq. hydrogen
fluoride. Sodium nitrite 126.0 mg was added to the
solution, and was kept stirring for 30 min at 08C.
Then the solution was kept stirring for 1 h at 958C.
The reaction mixture was added 100 ml of water and
neutralized with sodium carbonate and extracted with
CHCl3. The organic layer was dried over anhydrous
MgSO4 and evaporated. Purification of extract by
column chromatography aluminum oxide, CHCl3
yielded 3,5,7-triFQ in 22% yield. mp 708C sub-
limated . H-NMR CDCl3 d: 7.13 ddd H-6 , 7.60
2.20; N, 7.65. Found: C, 58.86; H, 2.44; N, 7.58.
Ž
.
( )
2.1.8. 6,7,8-Trifluoroquinoline 6,7,8-triFQ
Ž
.
Ž
2,3,4-Trifluoroaniline 1.07 g , glycerol 1.58 ml,
.
Ž
3 eq and sodium m-nitrobenzenesulfonate 4.96 g, 3
eq were dissolved in 80% H2 SO4 10 ml , and the
.
Ž
.
Ž
.
mixture was stirred at 1408C for 4 h. The reaction
Ž
.
Ž
.
mixture was poured into ice water 300 ml . The
filtrate was neutralized with aqueous NH3 and ex-
tracted with CHCl3. The CHCl3 layer was dried over
anhydrous MgSO4 and evaporated. Purification of
Ž
.
Ž
extract by column chromatography aluminum oxide,
.
benzene yielded 6,7,8-triFQ in 77% yield. mp 105-
1068C. MS mrz: 183 M . H-NMR CDCl3 d:
7.40 ddd H-5 , 7.50 dd H-3 , 8.13 dd H-4 , 8.97
dd H-2 ; J2 – 3 s4.3, J2 – 4 s1.3, J3 – 4 s8.4, J5– 6F
s9.9, J5– 7F s7.6, J5– 8F s2.3 Hz. Anal. Calcd for
C9 H4 F3N: C, 59.03; H, 2.20; N, 7.65. Found: C,
58.79; H, 2.29; N, 7.63.
1
q
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž
1
.
Ž
.
Ž
.
Fig. 1. Proposed mutagenic mechanism of quinoline.

(
)
T.-a. Kato et al.rMutation Research 439 1999 149–157
153
prepared by incubation for 10 h at 378C with rotary
shaking at 180 rpm following the transfer of 20 ml of
the stock cell culture to a flask containing 20 ml of
the medium Oxoid Nutrient broth No. 2. The chemi-
cal dissolved in 0.1 ml of dimethyl sulfoxide was
added to a test tube containing a mixture of 0.5 ml of
S9 mix consisting of 0.45 ml of co-factors and 0.05
ml of S9 fraction, and 0.1 ml of the cell culture in
the early stationary phase of the tester strain. After
Ž
incubation for 20 min at 378C with shaking 120
.
strokesrmin , 2.0 ml of 0.05 mM L-histidine-0.05
mM biotin molten top agar was added to the test
tube. The contents were mixed immediately and
layered on minimal glucose–agar plates. The plate
was incubated for 48 h at 378C, and then revertant
colonies were counted. In this study, at least 3
independent experiments were performed.
Fig. 2. Mutagenicity of monofluorinated quinolines in S. ty-
phimurium TA100 in the presence of S9-mix. The symbols shown
indicate the means of at least 3 independent experiments. The data
w x
of 3-F, 5-F, 6-F, and 8-Fq are from our previous report 2 . 3-FQ
was non-mutagenic and non-cytotoxic at up to 8 mmolrplate.
3. Results
(
)
2.1.9. 5,6,7,8-Tetrafluoroquinoline 5,6,7,8-tetraFQ
Ž
.
2,3,4,5-Tetrafluoroaniline 303.6 mg , glycerol
Mutagenicity was tested in S. typhimurium TA100
in the presence of an S9 mix according to the
Ž
Ž
.
.
0.40 ml, 3 eq and sodium m-nitrobenzenesulfonate
1.23 g, 3 eq were dissolved in 80% H2 SO4 10
ml , and the mixture was stirred at 1408C for 4 h.
The reaction mixture was poured into ice water 300
Ž
w
x
procedure of the Ames test 17–19 . In the present
paper, the mutagenic potency of fluorinated quino-
lines is compared to the parent quinoline in the lower
and higher dose ranges. Derivatives inducing less
than twice the number of revertants in the back-
ground were considered non-mutagenic. The dose-re-
.
Ž
.
ml . The filtrate was neutralized with aqueous NH3
and extracted with CHCl3. The CHCl3 layer was
dried over anhydrous MgSO4 and evaporated. Purifi-
cation of extract by column chromatography
Ž
.
aluminum oxide, benzene yielded 5,6,7,8-triFQ in
1
q
Ž
.
.
75% yield. mp 95–978C. MS mrz: 201 M . H-
NMR CDCl3 d: 7.57 dd H-3 , 8.43 dd H-4 , 9.02
Ž
.
Ž
.
Ž
Ž
.
dd H-2 ; J2 – 3 s4.3, J2 – 4 s1.3, J3 – 4 s8.6 Hz.
Anal. Calcd for C9 H4 F3N: C, 53.75; H, 1.50; N,
6.96. Found: C, 53.46; H, 1.62; N, 6.90.
2.2. Mutation assay
Chemicals were tested for mutagenicity as previ-
w
x
ously reported 17–19 , using S. typhimurium TA100
in the presence of S9 mix, consisting of the S9
Ž
.
fraction 50 ml obtained from the male Sprague–
Dawley rat liver pretreated with phenobarbital and
Ž
.
5,6-benzoflavone Oriental Yeast, Tokyo , and co-
Fig. 3. Mutagenicity of 5-fluorinated quinolines in S. typhimurium
TA100 in the presence of S9-mix. The symbols shown indicate
the means of at least 3 independent experiments. The data of 5-FQ
Ž
.
factors Oriental Yeast, Tokyo . Assays were carried
out after preincubation of the test chemical with S9
mix at 378C for 20 min. Briefly, the cell culture was
w x
are from out previous report 2 .

(
)
154
T.-a. Kato et al.rMutation Research 439 1999 149–157
deprive the mutagenic quinolines of mutagenicity.
As shown in Fig. 2 and Table 1, all the 3-F-deriva-
tives, i.e., 3-FQ, 3,5-diFQ, 3,7-diFQ, and 3,5,7-triFQ,
were devoid of mutagenicity in the whole dose range
Ž
.
examined up to 10 mmolrplate , although at higher
doses some of them were slightly toxic to the tester
bacteria. These results afford further support to the
enamine epoxide theory that we have proposed for
the activation mechanism of the genotoxicity of
w
x
quinoline 2–6 and its benzene- or pyridine-fused
w x
heteroaromatics such as benzo f quinoline,
w x
w x
benzo h quinoline, and 1,7-phenanthroline 7 .
3.3. Mutagenicities of 5-fluorinated quinolines
Fig. 4. Mutagenicity of 7-fluorinated quinolines in S. typhimurium
TA100 in the presence of S9-mix. The symbols shown indicate
the means of at least 3 independent experiments.
As shown in Fig. 3 and Table 1, all the 5-fluorin-
ated quinolines, except for the ones bearing an F
atom at position 3, were more mutagenic than quino-
line in the lower dose range regardless of the number
of substituent F atoms or their location:
sponse curves are shown in Figs. 1–5. Table 1 lists
all the mutagenic derivatives with their mutagenici-
Ž
ties expressed as revertantsrnmolrplate calculated
5-FQf5,6-diFQf5,7-diFQ
f5,8-diFQf5,6,8-triFQ
.
from the linear portion of each curve near the origin
and maximum induced-revertantsrplate obtained
Ž
)quinoline43,5-diFQ, 3,5,7-triFQ
.
from the assay up to 10 mmolrplate .
= non-mutagenic
Ž
.
3.1. Mutagenicities of monofluoroquinolines
In the higher dose range, however, the mutagenic
potencies of 5,6-diFQ and 5,6,8-triFQ declined be-
low that of quinoline. They seem to be cytotoxic at
Mutagenicities of monofluoroquinolines are
shown in Fig. 1 and Table 1. The data were taken
w x
from our previous report 2 except for 7-FQ. Similar
w
x
results were also reported by LaVoie et al. 20 .
Monofluoroquinolines can be arranged in the de-
creasing order of their mutagenic potencies as fol-
lows:
7-FQf5-FQ)quinoline
f8-FQ)6-FQ
43-FQ non-mutagenic
Ž
.
3-FQ was non-mutagenic and non-cytotoxic in the
Ž
.
whole dose range examined up to 8 mmolrplate as
w x
previously reported 2 . 6-FQ was only weakly muta-
genic, but seemed to be most cytotoxic.
3.2. Mutagenicities of 3-fluorinated quinolines
Fig. 5. Mutagenicity of 6-fluorinated quinolines in S. typhimurium
TA100 in the presence of S9-mix. The symbols shown indicate
the means of at least 3 independent experiments. The data of 6-FQ
Potently mutagenic derivatives, 5-FQ, 7-FQ and
5,7-diFQ, were further fluorinated at position 3, be-
cause fluorination at this position was expected to
w x
are from our previous report 2 .

(
)
T.-a. Kato et al.rMutation Research 439 1999 149–157
155
Table 1
atom at position 3, were more mutagenic than quino-
line in the lower dose range just as seen in the series
of 5-fluorinated quinolines:
Ž
.
and seventeen fluoroquino-
Mutagenic potency of quinoline
Q
Ž
.
lines FQs toward S. typhimurium TA100 in the presence of
S9-mix
7-FQf5,7-diFQf7,8-FQ
Chemicals
Induced-revertants
per plate per nmola
Maximum induced-
revertants per plateb
f6,7-diFQf6,7,8-triFQ
)quinoline43,7-FQ, 3,5,7-triFQ
Ž
.
Quinoline Q
1.0
606
= non-mutagenic
Ž
.
Monofluoroquinolinesc
In the higher dose range, however, the mutagenic
potencies of 6,7-diFQ and 6,7,8-triFQ, both of which
bear an F atom at position 6, declined below that of
quinoline just as seen in the 5-fluorinated series.
Whenever F is substituted at position 7, such
derivatives were as mutagenic as those of the 5-fluo-
rinated quinoline series. It was true again that muta-
genic potency was lost by introduction of another F
atom to position 3.
3-FQ
5-FQ
6-FQ
7-FQ
8-FQ
Negative
1.3
0.2
1.5
0.4
1788
250
1629
479
Difluoroquinolinesd
3,5-diFQ
3,7-diFQ
5,6-diFQ
5,7-diFQ
5,8-diFQ
6,7-diFQ
6,8-diFQ
7,8-diFQ
Negative
Negative
3.0
1.9
1.9
0.9
0.4
1.3
395
2298
1313
301
136
2060
3.5. Mutagenicities of 6-fluorinated quinolines
Ž .
6-Fluorinated quinolines without an F-atom s at
position 5 andror position 7, i.e., 6-FQ and 6,8-diFQ,
were less mutagenic than quinoline in the whole
dose range, as shown in Fig. 5 and Table 1. As
already mentioned, all of the six 6-fluorinated quino-
lines including the above two were less mutagenic
than quinoline at doses higher than ca. 1.0
mmolrplate.
Trifluoroquinolines
3,5,7-triFQ
5,6,8-triFQ
Negative
2.3
1.3
369
342
6,7,8-triFQ
Tetrafluoroquinoline
5,6,7,8-tetraFQ 0.2
144
3.6. Mutagenicities of 8-fluorinated quinolines
aCalculated from the slope of the linear portion of each curve near
the origin. The numbers indicate the means of at least 3 indepen-
dent experiments.
Introduction of an F atom to position 8 dose not
seem to produce any remarkable substituent effect on
b The number indicated is subtracted by the number of sponta-
neous revertants and obtained from the assay up to 10 mmolrplate.
In this study, the mean number of spontaneous revertants per plate
was 130 with standard deviation of 55. The numbers indicate the
means of at least 3 independent experiments.
c The data of 3-F, 5-F, 6-F, and 8-FQ are from our previous report
w x
2 .
d
w x
The data of 3,5-diFQ are from our previous report 7 .
the higher doses. A structural feature for this declin-
ing tendency is the presence of an F atom at position
6, if one takes into account the same tendency found
Ž
.
in the series of 7-fluorinated quinolines see below .
3.4. Mutagenicities of 7-fluorinated quinolines
Fig. 6. Mutagenicity of 5,6,7,8-tetraFQ in S. typhimurium TA100
in the presence of S9-mix. The symbols shown indicate the means
of at least 3 independent experiments.
As shown in Fig. 4 and Table 1, all the 7-fluorin-
ated quinolines, except for the ones bearing an F

(
)
156
T.-a. Kato et al.rMutation Research 439 1999 149–157
the mutagenicity, when one compares mutagenicity
in the following pairs of compounds: quinoline and
completely inhibited by 5-F, 6-F, or 7-F-substitution,
and that 8-FQ was readily metabolized to its 5,6-diol
Ž
.
Ž
.
Ž
.
8-FQ Fig. 1 ; 5-FQ and 5,8-diFQ Fig. 3 ; 6-FQ and
data not shown . This finding suggests that the
Ž
.
Ž
.
6,8-diFQ Fig. 5 ; 7-FQ and 7,8-diFQ Fig. 4 ; 5,6-
diFQ and 5,6,8-triFQ Figs. 3 and 5 ; 6,7-diFQ and
6,7,8-triFQ Figs. 4 and 5 Table 1 .
enhancement effect on mutagenicity by fluorine-sub-
stitution at the positions-5 and -7 may mainly be due
to the inhibition of the major detoxification pathway
of quinoline. However, a major question still remains
to be answered why 6-F-substitution shows different
mutagenicrcytotoxic effects from 5- and 7-F-sub-
stitutions with quinolines. Further investigation is
now underway.
Ž
.
Ž
. Ž
.
3.7. Mutagenicity of a tetrafluoroquinoline
5,6,7,8-tetraFQ
(
)
There seems to be an additivity in the substituent
effect of F atoms at various locations on the muta-
genicity among di- and trifluoroquinolines. However,
5,6,7,8-tetraFQ, in spite of the presence of F atoms
at the positions 5 and 7, was only marginally muta-
Acknowledgements
Ž
.
genic Fig. 6 and Table 1 . Since perfluorinated
compounds, such as perfluorodecaline for the artifi-
cial blood, freons, and perfluoropolyethylene, are
generally inert in any chemicalrbiochemical reac-
tions, an increases in the number of F-substituents
must make chemicals much less mutagenic. It is,
therefore, likely that the electronic effect of an F
atom should be regarded as additive only among the
derivatives carrying less than four F atoms in the
molecule.
We wish to thank Dr. Kunitaka Momota of Morita
Chemical Industries, Osaka, for kindly supplying us
with the fluorination reagent.
References
w x
1
M. Nagao, T. Yahagi, Y. Seino, T. Sugimura, N. Ito, Muta-
genicities of quinoline and its derivatives, Mutat. Res. 42
Ž
.
1977 335–342.
w x
2
K. Takahashi, M. Kamiya, Y. Sengoku, K. Kohda, Y. Kawa-
zoe, Deprivation of the mutagenic property of quinoline:
Inhibition of mutagenic metabolism by fluorine substitution,
Ž
.
4. Discussion
Chem. Pharm. Bull. 36 1988 4630–4633.
w x
3
M. Kamiya, Y. Sengoku, K. Takahashi, K. Kohda, Y. Kawa-
zoe, Antimutagenic structure modification of quinoline: fluo-
rine-substitution at position-3, in: Y. Kuroda, D.M. Shankel,
M.D. Waters Eds. , Antimutagenesis and Anticarcinogene-
sis, Vol. II, Plenum, New York, 1990, pp. 441–446.
M. Tada, K. Takahashi, Y. Kawazoe, Binding of quinoline to
nucleic acid in a subcellular microsomal system, Chem.-Biol.
Interactions 29 1980 257–266.
As we previously reported, quinoline and 5-FQ
were deprived of mutagenicity by fluorine-substitu-
Ž
.
w
x
tion at position 3 2,3 . The same result was shown
in the present study, i.e., a second or third fluorine-
substitution at position 3 also deprived the potently
mutagenic 7-FQ and 5,7-diFQ of their mutagenici-
ties. These results suggest that 3-fluorine-substitution
in the pyridine moiety would be a good means of
antimutagenic structural modification in pyridine-
fused aromatic compounds. The effects of fluorine-
substitution in the benzene moiety among mono-,
di-, and tri-F-quinolines on the mutagenicity can be
summarized as follows; suppression at position 3,
enhancement at position 5 or 7, decrease in muta-
genicity at position 6, and no effect at position 8.
Recently, our preliminary study on the metabolism
of these fluoroquinolines with the rat microsomal
enzymes indicated that the major detoxification
metabolism of quinoline, i.e., 5,6-epoxidation, was
w x
4
Ž
.
w x
5
M. Tada, K. Takahashi, Y. Kawazoe, Metabolites of quino-
line, a hepatocarcinogen, in a subcellular microsomal system,
Ž
.
Chem. Pharm. Bull. 30 1982 3834–3837.
w x
6
K. Saeki, K. Takahashi, Y. Kawazoe, Metabolism of muta-
genicity-deprived 3-fluoroquinoline: comparison with muta-
Ž
.
genic quinoline, Biol. Pharm. Bull. 16 1993 232–234.
K. Saeki, H. Kawai, Y. Kawazoe, A. Hakura, Dual stimula-
tory and Inhibitory effects of fluorine-substitution on muta-
genicity: an extension of the enamine epoxide theory for
activation of the quinoline nucleus, Biol. Pharm. Bull. 20
w x
7
Ž
.
1997 646–650.
w x
8
E.J. LaVoie, L. Tulley-Freiler, V. Bedenko, D. Hoffmann,
Mutagenicity of substituted phenanthrenes in Salmonella ty-
phimurium, Mutat. Res. 116 1983 91–102.
S.S. Hecht, E.J. LaVoie, V. Bedenko, L. Pingaro, S.
Ž
.
w x
9
Katayama, D. Hoffmann, D.J. Sardella, E. Boger, R.E. Lehr,

(
)
T.-a. Kato et al.rMutation Research 439 1999 149–157
157
w
x
Reduction of tumorigenicity and of dihydrodiol formation by
15 H. Gerson, M.W. McNeil, R. Parmegiani, P.K. Godfrey,
Secondary mechanisms of antifungal action of substituted
8-quinolinols.3: 5,7,8-substituted quinolines, J. Med. Chem.
w
x
fluorine substitution in the angular rings of dibenzo a,i -
Ž
.
pyrene, Cancer Res. 41 1981 4341–4345.
w
w
x
Ž .
10 E. Hubenman, T.J. Slag, Mutagenicity and tumor-initiating
15 1972 105–106.
w
x
activity of fluorinated derivatives of 7,12-dimethyl-
16 Y. Hamada, M. Sugiura, Studies on nitrogen-containing hete-
w x
Ž
.
benz a anthracene, Cancer Res. 39 1979 411–414.
rocyclic compounds. XXVII: Syntheses of 3-bromo meth-
Ž .
ylquinolines, Yakugaku Zasshi 98 1978 1–10.
x
11 L. Diamond, K. Chain, R.G. Harvey, J. DiGiovanni, Muta-
genic activity of methyl- and fluoro-substituted derivatives of
polycyclic aromatic hydrocarbons in a human hepatoma
w
x
17 B.N. Ames, J. McCann, E. Yamazaki, Methods for detecting
carcinogens and mutagens with the Salmonellar mamma-
lian-microsome mutagenicity test, Mutat. Res. 31 1975
Ž
.
Ž
.
cell-mediated assay, Mutat. Res. 136 1984 65–72.
w
x
12 K. Saeki, M. Kadoi, Y. Kawazoe, M. Futakuchi, D. Ti-
wawech, T. Shirai, Modification of the carcinogenic property
of quinoline, a hepatocarcinogen, by fluorine atom substitu-
tion: evaluation of carcinogenicity by a medium-term assay,
347–363.
w
w
w
x
18 K. Takahashi, T. Kaiya, Y. Kawazoe, Structure-mutagenicity
relationship among aminoquinolines, aza-analogues of
naphthylamine, and their N-acetyl derivatives, Mutat. Res.
Ž
.
Ž .
Biol. Pharm. Bull. 20 1997 40–43.
187 1987 191–197.
w
x
x
13 Y. Miyata, K. Saeki, Y. Kawazoe, M. Hayashi, T. Sofuni, T.
Suzuki, Antimutagenic structural modification of quinoline
assessed by an in vivo mutagenesis assay using lacZ trans-
19 A. Hakura, Y. Tsutsui, H. Mochida, Y. Sugihara, T. Mikami,
F. Sagami, Mutagenicity of dihydroxybenzenes and dihy-
droxynaphthalenes for Ames Salmonella tester strains, Mutat.
Res. 371 1996 293–299.
20 E.J. LaVoie, J. Defauw, M. Fealy, B.M. Way, C.A. Mc-
Ž
.
Ž .
genic mice, Mutat. Res. 414 1998 165–169.
w
x
x
14 M.H. Palmer, The Skraup reaction: formation of 5- and
Ž
.
7-substituted quinolines, J. Chem. Soc. 1962 pp. 3645–
3652.
Queen, Genotoxicity of fluoroquinolines and methylquino-
lines, Carcinogenesis 12 1991 217–220.
Ž
.