Mutagenicity of isomeric alkanediazotates, precursors for ultimate alkylating species of carcinogenic N-nitroso compounds

Article abstract of DOI:10.1016/S1383-5718(97)00180-0

Alkanediazohydroxides are common key intermediates in carcinogenesis and mutagenesis of N-nitroso compounds, which are widely found in human environment. Mutagenicity of (E)- and (Z)-potassium alkanediazotates, as precursors of corresponding alkanediazohydroxides were evaluated to investigate the effect of geometric isomerism and also the effect of alkyl groups on their biological activity. Mutagenicity of N-nitroso-N-alkylureas which spontaneously produce alkanediazohydroxides after non-enzymatic hydrolysis were also tested in comparison to that of the corresponding diazotates and other activated chemical species of N-nitrosamines. When the mutagenicity was assayed in three microbial strains, Salmonella typhimurium TA1535, and Escherichia coli WP2 and WP2 uvrA, the order of mutagenic potency of the compounds with the same alkyl group was as follows, (E)-diazotates > (Z)-diazotates > nitrosoureas. The effect of alkyl groups on the mutagenic potency was different in Salmonella strain and in E. coli strains, and this result could be explained by the efficiency of O⁶-alkylguanine-DNA alkyltransferase. In each bacterial strain, this effect of alkyl groups was similar in mutagenicity induced by (E)- and (Z)-diazotates, N-nitroso-N-alkylureas and other activated N-nitrosodialkylamines such as α-hydroxy nitrosamines. The geometrical isomerism affected the mutagenicity of (E)- and (Z)-potassium alkanediazotates, and the result suggested that alkanediazohydroxides react through diazonium ions in a cage rather than through free alkyldiazonium ions which have no geometrical isomerism. Our results confirmed that (E)-potassium alkanediazotates, (Z)-potassium alkanediazotates and N-nitroso-N-alkylureas all decomposed through diazohydroxides, and that alkanediazohydroxides are the active alkylating species of N-nitroso compounds, and also that the geometrical isomerism is important for carcinogenic N-nitroso compounds to show their biological activity.

Full text of DOI:10.1016/S1383-5718(97)00180-0

Ž
.
Mutation Research 412 1998 99–107
Mutagenicity of isomeric alkanediazotates, precursors for ultimate
1
alkylating species of carcinogenic N-nitroso compounds
)
Satoko Ukawa-Ishikawa, Atsuko Sawada, Keiko Kasuya, Masataka Mochizuki
Kyoritsu College of Pharmacy, Shibakoen 1-5-30, Minato-ku, Tokyo 105, Japan
Received 18 January 1997; accepted 15 September 1997
Abstract
Alkanediazohydroxides are common key intermediates in carcinogenesis and mutagenesis of N-nitroso compounds,
Ž
.
Ž .
which are widely found in human environment. Mutagenicity of E - and Z -potassium alkanediazotates, as precursors of
corresponding alkanediazohydroxides were evaluated to investigate the effect of geometric isomerism and also the effect of
alkyl groups on their biological activity. Mutagenicity of N-nitroso-N-alkylureas which spontaneously produce alkanediazo-
hydroxides after non-enzymatic hydrolysis were also tested in comparison to that of the corresponding diazotates and other
activated chemical species of N-nitrosamines. When the mutagenicity was assayed in three microbial strains, Salmonella
typhimurium TA1535, and Escherichia coli WP2 and WP2 uÕrA, the order of mutagenic potency of the compounds with
Ž
.
Ž .
the same alkyl group was as follows; E -diazotates) Z -diazotates)nitrosoureas. The effect of alkyl groups on the
mutagenic potency was different in Salmonella strain and in E. coli strains, and this result could be explained by the
efficiency of O6-alkylguanine-DNA alkyltransferase. In each bacterial strain, this effect of alkyl groups was similar in
Ž
.
Ž .
mutagenicity induced by E - and Z -diazotates, N-nitroso-N-alkylureas and other activated N-nitrosodialkylamines such
Ž
.
Ž .
as a-hydroxy nitrosamines. The geometrical isomerism affected the mutagenicity of E - and Z -potassium alkanediazo-
tates, and the result suggested that alkanediazohydroxides react through diazonium ions in a cage rather than through free
Ž
.
alkyldiazonium ions which have no geometrical isomerism. Our results confirmed that E -potassium alkanediazotates,
Ž .
Z -potassium alkanediazotates and N-nitroso-N-alkylureas all decomposed through diazohydroxides, and that alkanediazo-
hydroxides are the active alkylating species of N-nitroso compounds, and also that the geometrical isomerism is important
for carcinogenic N-nitroso compounds to show their biological activity. q 1998 Elsevier Science B.V.
Keywords: Alkanediazotate; Ultimate alkylating species; N-Nitroso compound; Mutagenicity; Geometrical isomerism
w x
1. Introduction
foods, drugs, or tobacco smoke 2 . They are also
formed in vivo under acidic conditions in stomach
N-Nitroso compounds are strong carcinogens for
animals, and are found in our environment such as
by the reaction of amines in foods and drugs, with
nitrite arising from nitrate in human saliva 3,4 , or
under neutral conditions by the action of activated
w
x
w
x
macrophages on amines 5,6 . Thus N-nitroso com-
pounds are considered to be a possible cause of
)
Corresponding author. Tel.: q81 3 54002665; Fax: q81 3
54002695; E-mail: mochizuki-mt@kyoritsu-ph.ac.jp
1 This work was presented partly at the tenth International
w x
human cancer 7 .
w x
N-Nitroso compounds are classified into two
Symposium on N-Nitroso Compounds 1 .
1383-5718r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII S1383-5718 97 00180-0

(
)
100
S. Ukawa-Ishikawa et al.rMutation Research 412 1998 99–107
groups; N-nitrosamines and N-nitrosamides. N-
mers, E and Z, due to a double bond character of
w
x
Nitrosamines need metabolic activation to show any
N5N bond 16 . Although diazohydroxides are quite
unstable to isolate, potassium alkanediazotates of
w x
biological activity 8 . N-Nitrosodialkylamines, one
w
x
major group of N-nitrosamines, are metabolically
hydroxylated by cytochrome P450 2E1 at a carbon,
forming active metabolites, a-hydroxy nitrosamines.
The a-hydroxy nitrosamines decompose to alkanedi-
azohydroxides, and then to alkyldiazonium ions,
which alkylate DNA or other biological molecules
each isomer can be synthesized 17,18 . The synthe-
sis of alkanediazotates was reported in 1910 by
Thiele 17 , then studies about their structure and
chemistry were reported 16,19,20 , Recently Fish-
bein and co-workers reported some studies on de-
composition of simple alkanediazotates in aqueous
w
x
w
x
Ž
. w
x
w
x
Fig. 1 8,9 . Since a-hydroxy nitrosamines are
solution 21–23 . Since the alkanediazohydroxides
may be common key intermediates in carcinogenesis
or mutagenesis of N-nitroso compounds, it is impor-
tant to investigate the relation between their geome-
try and biological responses. Because alkanediazo-
tates are easily converted to the corresponding alka-
nediazohydroxides in aqueous solution, they are use-
ful precursors for investigation of behavior of alka-
quite unstable in aqueous solution, their acetyl esters,
a-acetoxy nitrosamines, have been used as precur-
sors for a-hydroxy nitrosamines in research for
chemical and biological activities of N-nitrosamines
w
x
10 . a-Hydroperoxy nitrosamines were synthesized
by nucleophilic substitution of the corresponding
a-acetoxy nitrosamines with hydrogen peroxide in
w
x
w x
acetic acid 11 , and also by oxygenation of dialkyl-
nediazohydroxides 24 . However, few studies about
biological activity related to its chemical reactivity
were reported so far, except the mutagenicity of
w
x
nitrosamines in the presence of a strong base 12 .
The a-hydroperoxy nitrosamines also showed strong
w
x
Ž .
mutagenicity in bacterial strains 13 . The a-hydroxy
nitrosamines were isolated by deoxygenation of a-
Z -alkanediazotates reported in 1983 by Hecker et
w
x
al. 24 .
This paper describes the mutagenicity of the two
geometrical isomers of diazotates, E - and Z -
potassium alkanediazotates, concerning the effect of
geometry and their alkyl groups on biological activ-
ity. The effect of alkyl groups was compared with
w
x
hydroperoxy nitrosamines 13–15 , and their chem-
w
x
Ž .
Ž .
istry and mutagenicity have been elucidated 15 .
Alkanediazohydroxides have been thought to be
important intermediates in alkylation of carcinogenic
N-nitroso compounds. They have geometrical iso-
Fig. 1. Activation of carcinogenic N-nitroso compounds.

(
)
S. Ukawa-Ishikawa et al.rMutation Research 412 1998 99–107
101
that of the corresponding N-nitroso-N-alkylureas or
activated N-nitrosodialkylamines to confirm the role
of alkanediazohydroxides in mutagenesis or carcino-
genesis of N-nitroso compounds.
scribed above except using N-nitroso-N-alkylhydra-
zines as starting materials. N-Nitroso-N-alkylhydra-
Ž
.
zines alkylsethyl, propyl and butyl were synthe-
sized by nitrosation of the corresponding alkylhydra-
w
x
zine in acetic acid 17,25,26 . N-Nitroso-N-al-
kylhydrazine prepared was dissolved in anhydrous
ether, and was added by etherial solution of potas-
sium ethoxide and alcoholic solution of ethyl nitrite.
The mixture was stirred for 2 h at 08C, and the
resulting precipitates were filtered, and washed with
anhydrous ether. White solid obtained was used
without further purification. Purity was calculated by
integration of 1H-NMR data adding tri-t-butylben-
2. Materials and methods
2.1. Chemicals
Most reagents used were purchased from Wako
Ž
.
Pure Chemical Osaka, Japan as the purest grade
available. Methylhydrazine was purchased from
Ž
.
Ž .
Aldrich Chemical Milwaukee, WI, USA . Ethylhy-
drazine oxalate, propylhydrazine oxalate and butyl-
hydrazine oxalate were purchased from Fluka Chemie
zene as an internal standard. E -Potassium ethane-
Ž
.
diazotate CAS 92078-92-7 . Purity: 32.8%. Yield:
. Ž
23.2%. 1H-NMR DMSO-d6
. Ž .
ppm : 0.98 3H,
Ž
.
Ž
Ž
.
.
Ž
Buchs, Switzerland . N-Nitroso-N-methyl-p-
triplet , 3.38 2H, quartet . E -Potassium propanedi-
Ž
.
toluenesulfonamide and N-butyl-p-toluenesulfona-
mide were purchased from Tokyo Kasei Kogyo
azotate CAS 98114-63-7 . Purity: 76.2%. Yield:
. Ž
21.4%. 1H-NMR DMSO-d6
ppm : 0.84 3H,
Ž . Ž .
Ž
.
Ž
Ž
.
.
Ž
.
Tokyo, Japan , and N-ethyl-p-toluenesulfonamide
triplet , 1.42 2H, sextet , 3.34 2H, triplet . E -
Ž
Ž
.
was purchased from Eastman Kodak Rochester, NY,
Potassium butanediazotate CAS 98114-64-8 . Pu-
rity: 55.9%. Yield: 32.2%. 1H-NMR DMSO-d6
.
Ž
.
USA . N-Nitroso-N-alkylureas having methyl, ethyl,
Ž
.
Ž
.
Ž
.
Ž
propyl and butyl groups were kindly provided by
ppm : 0.87 3H, triplet , 1.28 2H, sextet , 1.40 2H,
. w
Ž
.
.
Ž
x
Toshin Chemical Industry Tokyo, Japan . Ethanol
and dimethyl sulfoxide were freshly distilled from
calcium hydride before use. Bacto agar and nutrient
quintet , 3.38 2H, triplet 23 .
( )
2.3. Z -Potassium alkanediazotates
Ž
broth were purchased from Difco Laboratories De-
.
troit, MI, USA , and sodium ammonium hydrogen-
Ž
Ž . .
Ž
phosphate was purchased from Merck Rahway, NJ,
Z -Potassium methanediazotate CAS 3058-37-5
.
w
x
USA .
was synthesized by the methods reported 18 . N-
Nitroso-N-alkyl-p-toluenesulfonamide having ethyl,
propyl and butyl groups was prepared from the
( )
2.2. E -Potassium alkanediazotates
Ž
corresponding N-alkyl-p-toluenesulfonamide ethyl
Ž .
Ž
.
E -Potassium methanediazotate CAS 98114-62-
and butyl by nitrosation, or nitrosation after alkyla-
.
w x
Ž
. w x
6 was synthesized by the methods reported 17 .
Methylhydrazine was dissolved in anhydrous ether,
and was added by freshly prepared etherial solution
of potassium ethoxide, then was added by alcoholic
solution of ethyl nitrite. The mixture was stirred for
2 h at 08C, and the resulting precipitate was filtered
under nitrogen atmosphere, and washed with anhy-
drous ether. White solid obtained was used without
further purification. Purity was calculated by integra-
tion of 1H-NMR data adding tri-t-butylbenzene as an
tion of p-toluenesulfonamide propyl 27 . N-
Nitroso-N-alkyl-p-toluenesulfonamide was dissolved
in anhydrous ether, and was added by freshly pre-
pared etherial solution of potassium ethoxide. The
mixture was stirred for 2 h at 08, and the resulting
precipitate was filtered, and washed with anhydrous
ether. White solid obtained was used without further
purification. Purity was calculated by integration of
1H-NMR data adding tri-t-butylbenzene as an inter-
Ž .
nal standard. Z -Potassium methanediazotate. Pu-
1
rity: 17.8%. Yield: 17.4%. 1H-NMR DMSO-d6
Ž
.
internal standard. Purity: 70.3%. Yield: 15.6%. H-
Ž
. Ž
.
Ž
. w x
Ž
.
Ž
. w x Ž .
NMR DMSO-d6 ppm : 3.10 3H, singlet 21 .
ppm : 2.63 3H, singlet 18 . Z -Potassium ethane-
diazotate CAS 94610-15-8 . Purity: 31.7%. Yield:
6.7%. H-NMR DMSO-d6 ppm : 1.02 3H, triplet ,
Ž .
Ž
.
Other E -diazotates having ethyl, propyl and
butyl groups were synthesized by the methods de-
1
Ž
. Ž
.
Ž
.

(
)
102
S. Ukawa-Ishikawa et al.rMutation Research 412 1998 99–107
Ž
. Ž .
3.07 2H, quartet . Z -Potassium propanediazotate
agar plates contained 30 ml of the minimal medium
containing 1.5% Bacto agar.
The bacteria were grown in 5 ml of NB for 15.5 h
at 378C. Test compounds were diluted in dimethyl
Ž
.
CAS 98114-60-4 . Purity: 38.1%. Yield: 13.8%.
H-NMR DMSO-d6 ppm : 0.86 3H, triplet , 1.47
2H, sextet , 2.99 2H, triplet 24 . Z -Potassium
1
Ž
. Ž
.
Ž
.
Ž
.
Ž
. w x Ž .
Ž
.
Ž .
sulfoxide DMSO . To a tube containing 0.5 ml of
butanediazotate CAS 98114-61-5 . Purity: 46.5%.
. Ž
Yield: 31.4%. 1H-NMR DMSO-d6 ppm : 0.86
0.1 M sodium phosphate buffer pH7.4 and 0.1 ml
of a culture of bacterial tester strain, 0.1 ml of
sample solution was added, then the mixture was
shaken immediately, and was mixed with 2 ml of top
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
3H, triplet , 1.47 2H, sextet , 1.62 2H, quintet ,
Ž
.
3.05 2H, triplet .
Ž
.
2.4. Bacterial strains
agar 0.6% agar in 0.6% NaCl at 458C and was
poured on a minimal agar plate. After incubation at
378C for 2 days, the colonies on the plate were
counted manually, and the number of spontaneous
revertant colonies was subtracted from those of the
induced revertants. Concentration of the test com-
pounds were expressed as mmolrplate. All data
reported are representative of at least three experi-
ments using duplicate plates per each dose level.
A culture of Salmonella typhimurium TA1535
was kindly provided by Dr. B.N. Ames, University
of California, Berkeley, USA. Escherichia coli WP2
and WP2 uÕrA were kindly donated by Dr. S. Iwa-
hara, Food and Drug Safety Center, Hadano, Japan.
2.5. Mutation assay
The medium used for overnight culture of bacte-
Ž
.
rial strains was a nutrient broth medium NB con-
taining 6 g of nutrient broth and 5 g of NaCl per
liter. The minimal media used for mutation assays
were as follows: S. typhimurium strain, the medium
3. Results
w
x
Ž .
Ž .
reported by Maron and Ames 28 ; E. coli strains,
modified Vogel–Bonner E medium supplemented
with 20 ml of NB per liter and 0.4% glucose. The
The mutagenicity of E - and Z -potassium alka-
nediazotates and N-nitroso-N-alkylureas was assayed
in S. typhimurium TA1535, and E. coli WP2 and
Ž
.
Fig. 2. Mutagenicity of E -potassium alkanediazotates in three different microbial strains: S. typhimurium TA1535, E. coli WP2 and WP2
uÕrA. `, Methanediazotate; v, ethanediazotate; B, propanediazotate; ', butanediazotate.

(
)
S. Ukawa-Ishikawa et al.rMutation Research 412 1998 99–107
103
Ž
.
Fig. 3. Mutagenicity of Z -potassium alkanediazotates in three different microbial strains: S. typhimurium TA1535, E. coli WP2 and WP2
uÕrA. `, Methanediazotate; v, ethanediazotate; B, propanediazotate; ', butanediazotate.
WP2 uÕrA. All compounds were mutagenic, and the
mutagenic potency was linearly related to the con-
method. Fig. 5 shows the specific mutagenicity for
three series of chemicals in three different microbial
Ž
.
Ž .
centration of chemicals Figs. 2–4 . A specific muta-
genicity per mmol of chemicals was defined by the
slope of the linear part in the dose–mutagenicity
relation and was calculated by the least squares
strains. In all strains, E -diazotates were most muta-
genic, and the activity was ten times stronger than
that of the corresponding Z -diazotates. N-Nitroso-
N-alkylureas were less mutagenic than diazotates. In
Ž .
Fig. 4. Mutagenicity of N-nitroso-N-alkylureas in three different microbial strains: S. typhimurium TA1535, E. coli WP2 and WP2 uÕrA.
`, Nitrosomethylurea; v, nitrosoethylurea; B, nitrosopropylurea; ', nitrosobutylurea.

(
)
104
S. Ukawa-Ishikawa et al.rMutation Research 412 1998 99–107
nitrosodialkylamines. The strains used by Hecker et
al. were S. typhimurium TA1535 and TA100, which
w
x
can detect base-pair change mutagenicity 28 . Al-
though strain TA100 are commonly used in many
investigations, we selected strain TA1535 to exclude
the effect of SOS response due to R-factor plasmid
pKM101. The two E. coli strains, WP2 and WP2
uÕrA, can also detect base-pair change mutagenicity
w
x
29 , and they were used together with TA1535 strain
for the comparison to the mutagenicity of the series
of activated N-nitrosamines already reported
w
x
10,13,15 . Both S. typhimurium TA1535 and E. coli
WP2 uÕrA are deficient in excision repair system
w
x
28,29 .
Ž .
Ž .
E - and Z -potassium alkanediazotates and N-
nitroso-N-alkylureas were all mutagenic in three mi-
crobial strains, and the difference in alkyl group
Ž .
affected their mutagenicity. The activity of Z -
ethanediazotate was higher than that of reported
w
x
result 24 , since our determination of purity of
diazotates by using internal standard allowed the
accurate calculation of molar concentration for muta-
tion assay. When the alkyl group was changed, the
mutagenic activity was decreased with an increase of
alkyl chain length in S. typhimurium TA1535. The
effect of changing alkyl groups on relative muta-
genicity described above was similar in two series of
diazotates and nitrosoureas, and also in other acti-
vated N-nitrosodialkylamines, such as a-hydroxy
Ž
.
Fig. 5. Relative mutagenicity of E -potassium alkanediazotates
Ž
.
Ž
.
Ž
Ž .
v , Z -potassium alkanediazotates ' and N-nitroso-N-al-
.
kylureas ` in S. typhimurium TA1535, E. coli WP2 and WP2
uÕrA strains. Revertants per mmol were calculated by the least-
squares method from the linear part in the dose–mutagenicity
relationship.
S. typhimurium TA1535 strain, increase of alkyl
chain length decreased the mutagenic potency. In E.
coli strains, ethyl homologues were more mutagenic
than methyl homologues, while in WP2 uÕrA strain,
butyl homologues were more mutagenic than propyl
w
x
w x
nitrosamines 13 , a-acetoxy nitrosamines 10 and
w
x
a-hydroperoxy nitrosamines 13 . The effects were
explained by reactivity and selectivity of alkyl groups
by nucleophiles. The rate of alkylation of de-
oxyguanosine by a-hydroxy nitrosamine was de-
creased by the following order: methyl)ethyl)
Ž
.
homologues Fig. 5 .
w
x
4. Discussion
propyl)butyl 13 . Furthermore, in the hard and
Ž
.
w x
soft acids and bases HSAB principle 30 , alkyl
Hecker et al. already reported the mutagenicity of
cations with more carbon branches were harder acids
7
Ž .
w
x
Z -potassium alkanediazotates having methyl, ethyl
31,32 , and the N -position of guanine was a harder
base, while the O6-position of guanine was a softer
and propyl groups as precursors of activated ni-
w
x
w
x
trosamines 24 . In the present study, we systemati-
cally and quantitatively evaluated the mutagenicity
base 19 . Thus, a softer methyl group could more
effectively alkylate the O6-position of guanine to
make methanediazotate the most mutagenic.
Ž .
Ž .
of both geometrical isomers, E - and Z -potassium
alkanediazotates, having methyl, ethyl, propyl and
butyl groups with the purity determined by NMR
spectroscopy, together with N-nitroso-N-alkylureas
which decompose to alkanediazohydroxides as com-
mon intermediates in the metabolism of N-
While, in E. coli, ethyl homologues were more
Ž
.
mutagenic than methyl homologues Fig. 5 . The
difference between the response in Salmonella strain
and that in E. coli strains was explained by the
efficiency of O6-alkylguanine-DNA alkyltransferase.

(
)
S. Ukawa-Ishikawa et al.rMutation Research 412 1998 99–107
105
.
The enzyme activity for the repair in E. coli was
their a-hydroxy derivatives also have geometrical
isomers due to N–N5O bond. The actual structure
of N–N5O is a resonance hybrid among two reso-
nance structures with a single N–N bond and a
double N5N bond, then the N–N bond is realized as
a partial double bond. It is unknown whether the
geometry of N-nitrosamines is retained when they
convert to alkanediazohydroxides, however, our re-
sults suggested that the geometry of the reactive
intermediate possibly affected the biological activity
of carcinogenic nitrosamines. In a-acetoxy ni-
trosamines, the ratio of Z-isomer to E-isomer in-
w
x
inducible in contrast to that in S. typhimurium 33 .
Therefore, the repair enzyme of E. coli was effec-
tively induced in the conditions of the present study,
so the methyl homologue was less mutagenic than
ethyl homologue.
N-Nitrosamides, such as N-nitroso-N-methylurea
or N-methyl-NX-nitro-N-nitrosoguanidine, are chemi-
cally more reactive than N-nitrosamines, and do not
require any metabolic activation. They decompose
spontaneously through alkanediazohydroxides under
physiological conditions, and subsequently alkylate
w
x Ž
.
w
x
nucleophiles in vivo 34 Fig. 1 . There are several
reports about mutagenicity of N-nitroso-N-methyl-
urea or simple N-nitrosoureas having larger alkyl
creased by substitution of bulky alkyl group 38 , and
the higher the ratio of Z-isomer in a-acetoxy ni-
trosamine, the faster was their rate of decomposition
w
w
x
w x
groups such as ethyl, propyl and butyl groups 35–
39 . Farrelly et al. 40 reported that the difference
x
Ž .
Ž .
37 . Our results that N-nitroso-N-methylurea was
in the rate of metabolism of E - and Z -isomers
for N-nitroso-N-methyl-2-oxopropylamine, and N-
nitroso-N-methyl-N-n-pentylamine behaved similarly
most mutagenic in S. typhimurium TA1535, and that
N-nitroso-N-ethylurea was most mutagenic in E. coli
strains, were identical with the reported results. In
this paper, the mutagenicity of N-nitroso-N-al-
kylureas was examined under the same conditions
for comparison with those of diazotates, and the
effect of alkyl groups on mutagenicity was similar to
that of E - and Z -diazotates. These results sup-
ported that N-nitrosoureas decomposed to the corre-
sponding alkanediazohydroxides, and then showed
their alkylating activity.
w
x Ž .
Ž .
41 . E - and Z -diazotates cannot easily isomerize
w
x
to each other 19 , so it is of interest to determine
whether N-nitroso compounds decompose to alkane-
diazohydroxides retaining or inverting their geomet-
ric form. Since few studies about the relation of
geometrical isomerism of N-nitroso compounds and
their related biological properties were reported, the
present results revealed that geometrical isomerism
of alkanediazohydroxides greatly affected on their
biological activities.
Ž .
Ž .
Ž .
E -Potassium alkanediazotates were more muta-
genic than Z -diazotates. The result suggested that
geometrical isomerism affected biological activity
such as mutagenicity or carcinogenicity. The reactiv-
Ž .
Ž .
Ž .
The results that mutagenicities of E - and Z -di-
azotates were different suggested that alkanediazohy-
droxides may not react through free alkyldiazonium
ions which have no geometrical isomerism but
through diazonium ions in a cage, and a possible
reactive form in protic solvent is an ion triplet
Ž .
ity toward nucleophiles is higher in Z -diazotates
Ž .
Ž .
than in E -diazotates, because Z -diazotates with
softer carbon react with the O6-position of guanine
Ž .
w
x
more easily than E -diazotates with harder carbon
proposed by Moss 16 , which retains the nature of
Ž .
Ž .
estimated by HSAB principle and ab initio calcula-
tion 19 . In spite of the greater reactivity of Z -di-
steric structure of E - or Z -isomer.
w
x
Ž .
Although the mutagenic potency of a series of
N-nitrosamines, N-nitroso-N-alkylureas, or alkanedi-
azotates were different quantitatively, the effect of
alkyl groups on their mutagenicity in three microbial
strains was identical. Present results using alkanedia-
zotates as precursors for alkanediazohydroxides fur-
ther supported that carcinogenic N-nitroso com-
pounds decompose through alkanediazohydroxides,
and that alkanediazohydroxides are the common ac-
tive alkylating species of N-nitroso compounds. In
addition, the present result that geometrical iso-
Ž .
azotates, most Z -diazotates have shorter lifetimes
Ž .
than E -diazotates in aqueous media used as solvent
in mutation assay, because the anti-periplanar ar-
rangement of lone pair of electrons on nitrogen of
Z -diazotate assists easy elimination of the hydroxyl
group from Z -diazohydroxide, conjugated acid of
Ž .
Ž .
x
w
diazotate 23 . Thus, the weaker mutagenicity of
Ž .
Z -diazotates could be explained by their instability.
Besides alkanediazohydroxides or alkanediazo-
Ž
tates, non-symmetrical N-nitrosodialkylamines or

(
)
106
S. Ukawa-Ishikawa et al.rMutation Research 412 1998 99–107
Ž
.
a-acetoxyalkyl nitrosamines, models for metabolically acti-
merism of alkanediazotates affected on bacterial mu-
tagenicity suggested that alkanediazohydroxide or
Ž
.
vated N,N-dialkylnitrosamines, Gann 70 1979 663–670.
w
w
w
x
11 M. Mochizuki, T. Anjo, Y. Wakabayashi, T. Sone, M.
w
x
alkyldiazonium ion in a cage 16 is a possible
ultimate alkylating species rather than free alkyldia-
zonium ion, and that the geometry of alkanediazohy-
droxides might determine biological behaviors of
their parent N-nitroso compounds.
Ž
.
Okada, Formation of N-alkyl-N- 1-hydroperoxyalkyl -
nitrosamines from N-alkyl-N- 1-acetoxyalkyl nitrosamines,
Tetrahedron Lett. 21 1980 1761–1764.
Ž
.
Ž
.
x
12 M. Mochizuki, T. Sone, T. Anjo, M. Okada, Synthesis of
Ž
.
N-alkyl-N- 1-hydroperoxyalkyl nitrosamines by oxygenation
of lithiated dialkylnitrosamines, Tetrahedron Lett. 21 1980
1765–1766.
Ž
.
x
13 M. Okada, M. Mochizuki, T. Anjo, T. Sone, Y. Wak-
abayashi, E. Suzuki, Formation, deoxygenation and muta-
genicity of a-hydroperoxydialkylnitrosamines, in: E.A.
Acknowledgements
This work was supported in part by a Grant-in-aid
for Cancer Research from the Ministry of Education,
Science, Sports and Culture, Japan, and a Grants
from the Foundation for Promotion of Cancer Re-
search, Japan.
Ž
.
Walker, M. Castegnaro, L. Griciute, M. Borzsonyi Eds. ,
¨
¨
N-Nitroso Compounds: Analysis, Formation and Occurrence,
IARC Scientific Publications No. 31, International Agency
for Research on Cancer, Lyon, 1980 pp. 71–82.
w
w
x
14 M. Mochizuki, T. Anjo, M. Okada, Isolation and characteri-
Ž
.
zation of N-alkyl-N- hydroxymethyl nitrosamines from N-
alkyl-N- hydroperoxymethyl nitrosamines by deoxygenation,
Ž
.
Ž
.
Tetrahedron Lett. 21 1980 3693–3696.
x
References
15 M. Mochizuki, T. Anjo, K. Takeda, E. Suzuki, N. Sekiguchi,
G.F. Huang, M. Okada, Chemistry and mutagenicity of
a-hydroxy nitrosamines, in: H. Bartsch, I.K. O’Neill, M.
w x
1
S. Ukawa, M. Mochizuki, Biological and chemical properties
of alkanediazotates as active species of N-nitroso com-
Ž
.
Castegnaro, M. Okada Eds. , N-Nitroso Compounds: Occur-
rence and Biological Effects, IARC Scientific Publications
No. 41, International Agency for Research on Cancer, Lyon,
1982 pp. 553–559.
Ž
.
pounds, in: I.K. O’Neill, J. Chen, H. Bartsch Eds. , Rele-
vance to Human Cancer of N-Nitroso Compounds, Tobacco
Smoke and Mycotoxins, IARC Scientific Publications No.
105, International Agency for Research on Cancer, Lyon,
1991 pp. 404–406.
w
w
x
16 R.A. Moss, Chemistry of some alkanediazotates, Acc. Chem.
Ž
.
Res. 7 1974 421–427.
¨
x
w x
2
R. Preussmann, G. Eisenbrand, N-Nitroso carcinogens in the
17 J. Thiele, Uber Nitrosohydrazine. Isoazotate und
Ž
.
environment, in: C.E. Searle Ed. , Chemical Carcinogens,
2nd ed., ACS Monograph No. 182, American Chemical
Society, Washington, DC, 1984 pp. 829–868.
Azoverbindungen der Fettreihe, Justus Liebigs Ann. Chem.
Ž
.
376 1910 239–268.
w
x
18 E. Muller, W. Hoppe, H. Hagenmaier, H. Haiss, R. Huber,
¨
w x
3
¨
S.S. Hecht, J.B. Morrison, A sensitive method for detecting
in vivo formation of N-nitrosomorpholine and its application
to rats given low doses of morpholine and sodium nitrite,
W. Rundel, H. Suhr, Uber Diazoverbindungen,
XX. Struktur isomerer Diazotate: das Hantzschsche Methyl-
Ž
.
diazotat, Chem. Ber. 96 1963 1712–1719.
Ž
.
Cancer Res. 44 1984 2873–2877.
w
x
19 J.W. Lown, S.M.S. Chauhan, R.R. Koganty, A.-M. Sapse,
Alkyldinitrogen species implicated in the carcinogenic, muta-
genic, and anticancer activities of N-nitroso compounds:
Characterization by 15 N NMR of 15 N-enriched compounds
and analysis of DNA base situselectivity by ab initio calcula-
w x
4
W. Lijinsky, Induction of tumours in rats by feeding nitrosat-
able amines together with sodium nitrite, Food Chem. Toxi-
Ž
.
col. 22 1984 715–720.
w x
5
M.A. Marletta, Mammalian synthesis of nitrite, nitrate, nitric
oxide, and N-nitrosating agents, Chem. Res. Toxicol. 1
Ž
.
tions, J. Am. Chem. Soc. 106 1984 6401–6408.
Ž
.
1988 249–257.
w
x
20 A.-M. Sapse, E.B. Allen, J.W. Lown, Quantum chemical
w x
6
M. Miwa, D.J. Stuehr, M.A. Marletta, J.S. Wishnok, S.R.
Tannenbaum, Nitrosation of amines by stimulated
Ž
studies of the products of decomposition of anticancer 2-
haloethyl nitrosoureas under physiological conditions, J. Am.
.
Ž
.
macrophages, Carcinogenesis 8 1987 955–958.
Ž
.
Chem. Soc. 110 1988 5671–5675.
w x
7
H. Bartsch, R. Montesano, Relevance of nitrosamines to
human cancer, Carcinogenesis 5 1984 1381–1393.
w
w
x
21 J. Hovinen, J.C. Fishbein, Rate constants for the decomposi-
Ž
.
tion of a simple alkanediazoate at physiological pH, J. Am.
w x
8
R. Preussmann, B.W. Stewart, N-Nitroso carcinogens, in:
Ž
.
Chem. Soc. 114 1992 366–367.
Ž
.
C.E. Searle Ed. , Chemical Carcinogens, 2nd ed., ACS
Monograph No. 182, American Chemical Society, Washing-
ton, DC, 1984 pp. 643–828.
x
22 J. Hovinen, J.I. Finneman, S.N. Satapathy, J. Ho, J.C. Fish-
Ž
.
bein, Mechanism of decomposition of E -methanediazoate
in aqueous solutions, J. Am. Chem. Soc. 114 1992 10321–
10328.
Ž
.
w x
9
W. Lijinsky, Species differences in nitrosamine carcino-
Ž
.
genesis, J. Cancer Res. Clin. Oncol. 108 1984 46–55.
w
x
23 J. Ho, J.C. Fishbein, Rate-limiting formation of diazonium
w
x
10 M. Mochizuki, E. Suzuki, T. Anjo, Y. Wakabayashi, M.
ions in the aqueous decomposition of primary alkanedia-
Okada, Mutagenic and DNA-damaging effects of N-alkyl-N-
Ž
.
zoates, J. Am. Chem. Soc. 116 1994 6611–6621.

(
)
S. Ukawa-Ishikawa et al.rMutation Research 412 1998 99–107
107
w
w
w
w
x
w
x
24 L.I. Hecker, J.E. Saavedra, J.G. Farrelly, A.W. Andrews,
35 J. McCann, E. Choi, E. Yamasaki, B.N. Ames, Detection of
carcinogens as mutagens in the Salmonellarmicrosome test:
Assay of 300 chemicals, Proc. Natl. Acad. Sci. USA 72
Mutagenicity of potassium alkanediazotates and their use as
model compounds for activated nitrosamines, Cancer Res. 43
Ž
.
Ž
.
1983 4078–4082.
1975 5135–5139.
x
w
x
25 G.A. Lanovaya, E.D. Korniets, O.I. Sidoraova, N.I. Pavlenko,
A.I. Rubailo, V.M. Makul’kin, PMR and IR spectral charac-
teristics of N-alkyl-N-nitrosohydrazines, J. Org. Chem. USSR
36 W. Lijinsky, A.W. Andrews, The mutagenicity of ni-
trosamides in Salmonella typhimurium, Mutation Res. 68
Ž
.
1979 1–8.
Ž
.
w
x
19 1983 237–240.
37 R.C. Garner, C. Pickering, C.N. Martin, Mutagenicity of
methyl-, ethyl-, propyl- and butylnitrosourea towards Es-
cherichia coli WP2 strains with varying DNA repair capabil-
x
26 S.N. Danilov, G.A. Lanovaya, S.Ya. Lazarev, V.I. Romanov,
V.A. Fedorova, V.I. Beresneva, T.N. Timofeeva, Study of
monosubstituted nitrosohydrazines by methods of UV and
Ž
.
ities, Chem.-Biol. Interact. 26 1979 197–205.
Ž
.
w
x
PMR spectroscopy, J. Gen. Chem. USSR 47 1977 190–193.
38 M. Mochizuki, T. Anjo, M. Okada, Syntheses of N-alkyl-N-
x
Ž
.
27 D.H. Hey, Th.J. De Boer, Thermal decomposition of higher
a-acetoxyalkyl nitrosamine, model compounds for metabol-
Ž
.
sulphonylalkylnitrosamides, Recl. Trav. Chim. 73 1954
ically activated N,N-dialkylnitrosamines, Chem. Pharm. Bull.
Ž
.
686–694.
26 1978 3905–3908.
w
w
w
w
w
x
w
x
28 D.M. Maron, B.N. Ames, Revised methods for the Salmonella
39 M. Mochizuki, T. Anjo, N. Sekiguchi, A. Ikarashi, Y. Wak-
Ž
.
Ž
mutagenicity test, Mutation Res. 113 1983 173–215.
abayashi, M. Okada, Solvolysis of N-nitroso-N- 1-
acetoxyalkyl alkylamines in phosphate buffer: Characteriza-
tio n an d m u tag en icity o f N -n itro so -N - 1-
.
phosphonooxyalkyl alkylamines, Chem. Pharm. Bull. 34
29 M.H.L. Green, W.J. Muriel, Mutagen testing using trpq
x
.
Ž
.
Ž
reversion in Escherichia coli, Mutation Res. 38 1976 3–32.
30 T.-L. Ho, The hard soft acids bases HSAB . Principle and
organic chemistry, Chem. Rev. 75 1975 1–20.
31 R.G. Pearson, Hard and soft acids and bases, J. Am. Chem.
x
Ž
.
Ž
.
Ž
.
1986 3956–3959.
40 J.G. Farrelly, M.L. Stewart, D.W. Farnsworth, J.E. Saavedra,
Metabolism of the Z and E isomers of N-nitroso-N-methyl-
2-oxopropyl amine by rat hepatocytes, Cancer Res. 48
1988 3347–3349.
41 C. He, J.G. Farrelly, Microsomal metabolism of the Z and E
x
w
w
x
Ž
.
Soc. 85 1963 3533–3539.
x
Ž
Ž
.
32 T.-L. Ho, H.C. Ho, L.D. Hamilton, Biochemical significance
.
of the hard and soft acids and bases principle, Chem.-Biol.
Interact. 23 1978 65–84.
33 J.B. Guttenplan, N-Nitrosamines: Bacterial mutagenesis and
in vitro metabolism, Mutation Res. 186 1987 81–134.
34 G.A. Digenis, C.H. Issidorides, Some biochemical aspects of
Ž
.
x
w
w
x
isomers of N-nitroso-N-methyl-N-n-pentylamine, Cancer
Ž .
Lett. 45 1989 189–194.
Ž
.
x
Ž
.
N-nitroso compounds, Bioorg. Chem. 8 1979 97–137.