7-Azabicyclo[2.2.1]heptane as a structural motif to block mutagenicity of nitrosamines

Article abstract of DOI:10.1016/j.bmc.2011.02.049

Nitrosamines are potent carcinogens and toxicants in the rat and potential genotoxins in humans. They are metabolically activated by hydroxylation at an α-carbon atom with respect to the nitrosoamino group, catalyzed by cytochrome P450. However, there has been little systematic investigation of the structure-mutagenic activity relationship of N-nitrosamines. Herein, we evaluated the mutagenicity of a series of 7-azabicyclo[2.2.1]heptane N-nitrosamines and related monocyclic nitrosamines by using the Ames assay. Our results show that the N-nitrosamine functionality embedded in the bicyclic 7-azabicylo[2.2.1]heptane structure lacks mutagenicity, that is, it is inert to α-hydroxylation, which is the trigger of mutagenic events. Further, the calculated α-C-H bond dissociation energies of the bicyclic nitrosamines are larger in magnitude than those of the corresponding monocyclic nitrosamines and N-nitrosodimethylamine by as much as 20-30 kcal/mol. These results are consistent with lower α-C-H bond reactivity of the bicyclic nitrosamines. Thus, the 7-azabicyclo[2.2.1]heptane structural motif may be useful for the design of nongenotoxic nitrosamine compounds with potential biological/medicinal applications.

Full text of DOI:10.1016/j.bmc.2011.02.049

Bioorganic & Medicinal Chemistry 19 (2011) 2726–2741
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
7-Azabicyclo[2.2.1]heptane as a structural motif to block mutagenicity
of nitrosamines
Tomohiko Ohwada a, , Satoko Ishikawa b, Yusuke Mine c, Keiko Inami c, Takahiro Yanagimoto a,
⇑
Fumika Karaki a, Yoji Kabasawa a, Yuko Otani a, Masataka Mochizuki c,
⇑
a Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
b Faculty of Pharmacy, Keio University, 1-5-30 Shibakoen, Minato-ku, Tokyo 105-8512, Japan
c Faculty of Pharmaceutical Sciences, Tokyo University of Science, Yamazaki 2641, Noda-shi, Chiba 278-8510, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Nitrosamines are potent carcinogens and toxicants in the rat and potential genotoxins in humans. They
Received 27 January 2011
Revised 23 February 2011
Accepted 24 February 2011
Available online 2 March 2011
are metabolically activated by hydroxylation at an
a-carbon atom with respect to the nitrosoamino
group, catalyzed by cytochrome P450. However, there has been little systematic investigation of the
structure–mutagenic activity relationship of N-nitrosamines. Herein, we evaluated the mutagenicity of
a series of 7-azabicyclo[2.2.1]heptane N-nitrosamines and related monocyclic nitrosamines by using
the Ames assay. Our results show that the N-nitrosamine functionality embedded in the bicyclic
7-azabicylo[2.2.1]heptane structure lacks mutagenicity, that is, it is inert to
the trigger of mutagenic events. Further, the calculated -C–H bond dissociation energies of the bicyclic
nitrosamines are larger in magnitude than those of the corresponding monocyclic nitrosamines and
N-nitrosodimethylamine by as much as 20–30 kcal/mol. These results are consistent with lower -C–H
Keywords:
a-hydroxylation, which is
Nitrosamines
Mutagenicity
Ames test
a
a
Bond dissociation energy
bond reactivity of the bicyclic nitrosamines. Thus, the 7-azabicyclo[2.2.1]heptane structural motif may
be useful for the design of nongenotoxic nitrosamine compounds with potential biological/medicinal
applications.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
and its formation has also been correlated with the cytotoxicity
of methylating agents in rapidly dividing cells.9 The metabolism
Nitrosamines, of which N-nitrosodimethylamine (NDMA) is a
representative simple example, are well-known as potent carcino-
gens and toxicants in rats1,2 and as potential genotoxins in
humans.3 Like many toxicants/mutagens, NDMA requires meta-
bolic activation to manifest its toxicity, and hydroxylation at an
of N-nitrosodialkylamines by rat liver microsomes has also been
reported,4,5 and short-lived radicals have been detected in the
presence of rat liver microsomes.10 On the other hand, there has
been little systematic investigation of the structural requirements
for mutagenic activity of N-nitrosamines. If we acquire this knowl-
edge,11,12 it may be possible to block the genotoxic activity of
a
-carbon atom with respect to the nitrosoamino group, catalyzed
by cytochrome P450 2E1, has been proposed to be the major met-
abolic activation pathway (Fig. 1).4,5 The formation of
-hydroxy-
NDMA (B, Fig. 1) is thought to involve intermediacy of an -carbon
radical or -carbocationic species (A, Fig. 1), which is captured by a
hydroxyl functionality.6 The resultant
-hydroxy-NDMA is a kind
a
a
H3C
H3C
H3C
H2C
H3C
H2C
a
P450
H3C
or
N NO
N NO
N NO
a
of hemiaminal, which can undergo fragmentation reaction to form
formaldehyde and an alkanediazohydroxide intermediate (C,
Fig. 1), and this is transformed to methanediazonium ion (D,
Fig. 1) upon elimination of water. The latter reactive species (D)
can react with DNA to form alkylated adducts, such as O6-methyl-
guanine (O6-MeG).7,8 O6-MeG is classed as a promutagenic lesion,
A
H3C
+
α-hydroxylation
O
N NO
N N
H
H
OH
H2C
OH
C
B
H3C
H+
- H2O
DNA
DNA Adduct
(such as O6-Me-G)
H3C N N
N N
⇑
Corresponding authors. Tel.: +81 3 5841 4730; fax: +81 3 5841 4735 (T.O.);
tel./fax: +81 4 7121 3641 (M.M.).
OH
C
D
E-mail addresses: ohwada@mol.f.u-tokyo.ac.jp (T. Ohwada), mochizuk@rs.noda.
tus.ac.jp (M. Mochizuki).
Figure 1. Metabolic oxidative activation of N-nitrosodimethylamine.
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.02.049

T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
2727
NO
N
NO
N
NO
N
NO
N
O
N
CH3
O
O
CF3
O
CF3
N
N
O
2
O
4
1
3
CH3
NO
N
NO
N
NO
N
NO
N
O
N
CO2Me
CO2Me
O
CO2Me
8
5
6
7
CO2Et
NO
N
NO
N
NO
N
NO
N
OH
H3CO
H3CO
O2N
O2N
O2N
9
12
10
11
NO
N
NO
NO
NO
N
N
N
CH2OMe
OH
CO2CH3
CO2CH3
CO2CH3
13
14
15
16
NO
N
CH2O
N NO
N NO
CO2CH3
17
18
19
Scheme 1. Nitrosamines studied in this work.
nitrosamines, which would enable the nitrosamine functionality to
be utilized in the molecular design of active pharmaceutical ingre-
dients (API) in medicinal chemistry,3 as well as bioactive molecules
in biology.
nitrosamines in the Ames assay, evaluated the structure–toxicity
relationship, and calculated the -C–H bond dissociation energies.
Our results indicate that the 7-azabicyclo[2.2.1]heptane structural
motif is effective to block the metabolic activation of nitrosamine
to the proximate mutagenic species.
a
7-Azabicyclo[2.2.1]heptane N-nitrosamines (1–17, Scheme 1)
are N-nitroso derivatives of the bicyclic amine, 7-azabicyclo[2.2.1]-
heptane. These nitrosamines constitute a new type of nitric oxide
(NO) donors, which do not produce NO directly at neutral pH,
but it can transnitrosate to thiols to generate S-nitrosothiols.13
While these nitrosamines have potential applications in biochem-
istry and medicinal chemistry, via modulation of protein functions
through S-nitrosation of cysteine residue(s),14 there is great con-
cern about their possible mutagenicity. Thus, we have evaluated
the mutagenicity of these bicyclic nitrosamines, as well as related
monocyclic nitrosamines, by using the Ames assay, which has been
used widely for detecting DNA-damaging compounds.15–17 In this
assay, S9 mix, which consists of cytochrome P450s and an
NADPH-regenerating system, is used for activating promutagens
to active forms.
2. Results and discussion
2.1. Ames assay of bicyclic and monocyclic N-nitrosamines
2.1.1. Bicyclic nitrosamines
N-Nitrosamines of 7-azabicyclo[2.2.1]heptane were assayed
for mutagenicity in the absence or presence of rat liver S9 mix in
two strains, Salmonella typhimurium TA100 and TA98, which are
used for detecting DNA-base substitution and frame-shift muta-
tion, respectively.15–17 Among 17 bicyclic nitrosamines (1–17,
Scheme 1), 14 compounds were negative in the Ames assay
(Figs. 2A–C), but three bearing aromatic dimethoxy groups (9) or
an aromatic nitro group (11 and 12) showed apparent mutagenic-
ity in the absence and presence of rat liver S9 mix in both
TA100 and TA98 (Fig. 3A). In addition, nitrosamine (3) showed
Herein, we determined the mutagenicity of 7-azabicyclo[2.2.1]-
heptane N-nitrosamines and the corresponding monocyclic

2728
T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
Figure 2A. Mutagenicity of compounds 1 (d), 2 (s), 3 (N), and 4 (4) in S. typhimurium TA98 and TA100 in the presence of S9 mix or absence of S9 mix.
cytotoxicity in the absence of S9 mix in the case of the TA98 strain,
while in the presence of S9 mix, no cytotoxicity or mutagenicity
was detected in either strain.18 This type of phenomenon, that is,
and 12 can be detected regardless of metabolic activation, that
is, in the absence or presence of S9 mix. The mutagenicity of both
11 and 12 in the absence of the S9 mix was stronger than that in
the presence of the S9 mix. The olefin (11) and hydroxyl (12)
functionalities are chemically interconvertible and may also be
interconvertible in Salmonella.23 Aromatic dimethoxy derivative
9 is also mutagenic, regardless of metabolic activation (Fig. 3A).
The mutagenicity of 9 is also stronger in the absence of the S9
mix than that in the presence of the S9 mix in both TA100 and
TA98.
a
cytotoxicity-antagonizing effect of S9 mix has often been
observed, and is considered to be due to the reaction of test
compounds or their metabolites with proteins in S9 mix.17
2.1.2. Monocyclic nitrosamines
On the other hand, monocyclic nitrosamines (18 and 19)
showed apparent mutagenicity only upon metabolic activation in
the presence of rat liver S9 mix; in the absence of S9 mix, they
were not mutagenic (Fig. 3B). Intriguingly, the mutagenic poten-
cies of these monocyclic nitrosamines are different: the benzo
derivative 18 showed strong activities in both TA100 and TA98,
while the non-benzo derivative 19 showed significantly reduced
activity in TA100 and little activity in TA98. The mutagenic activi-
ties of both 18 and 19 in TA98 were much weaker than those in
TA100 (Fig. 3B). Thus, the mutagenicity of these nitrosamines
arises from DNA-base substitution rather than DNA frame-shift
mutation. This conclusion is consistent with a previous report.19
While the mutagenicity of the bicyclic nitrosamines 11 and 12
can be detected, the nitrosamine (10) bearing an aromatic nitro
group showed no activity in the Ames assay with either TA98 or
TA100. This is consistent with previous reports that nitrobenzene
is negative in the Ames assay.20–22 Also the mutagenicity of 11
These above results suggest that the N-nitrosamine functional-
ity embedded in the bicyclic 7-azabicylo[2.2.1]heptane structure
lacks mutagenicity, that is, the structure may be inert to metabolic
a-hydroxylation, which is a trigger of mutagenic events in the case
of generally carcinogenic N-nitrosodialkylamines. This is in strong
contrast to the observed mutagenicity of the monocyclic nitrosa-
mines (18, 19 and others), which are susceptible to enzymatic
a-hydroxylation. Because three bicyclic N-nitrosamines (9, 11
and 12) showed mutagenicity even in the absence of S9 mix, their
mutagenicity is considered to be independent of the N-nitroso
structure. At present the origin of their mutagenicity is unclear.
Therefore, it appears that the 7-azabicyclo[2.2.1]heptane struc-
tural motif can be useful for structural design of non-genotoxic
nitrosamine compounds with potential for biological/medicinal
applications.

T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
2729
Figure 2B. Mutagenicity of compounds 5 (j), 6 (d), 7 (s), 8 (N), and 10 (4) in S. typhimurium TA98 in the presence of S9 mix or absence of S9 mix.
2.2. Evaluation of ease of
C–H bond dissociation energies
a
-hydroxylation based on computed
have shown that these more sophisticated methods may still have
serious problems with some BDEs.28 On the other hand, density
functional theory (DFT) calculations usually do not show serious
spin-contamination and, therefore, have been proposed to be reli-
able for open-shell systems.29 At present, the most reliable way to
calculate BDEs is to use composite ab initio methods, such as G3
and Complete Basis Set (CBS)-Q, which can provide predicted BDEs
within 1–2 kcal/mol of the experimental values,30–32 although
these methods are very expensive in terms of CPU time and mem-
ory requirement. CBS-Q is a composite ab initio method, which
starts with HF/6-31G(d) geometry optimization and frequency
calculation, followed by MP2(FC)/6-31G(d) optimization. Then the
single-point energy is calculated at MP2/6-311+G(3d2f,2df,2p),
MP4(SDQ)/6-31+G(d(f),p), and QCISD(T)/6-31+G(d) levels. These
energies are then extrapolated to the complete basis set limit.
Herein, we used the DFT methods to evaluate C–H BDE of these
bicyclic and monocyclic nitrosamines (Table 1). The geometries of
the nitrosamines and the corresponding carbon radical species
were fully optimized at the Restricted (R) and Unrestricted (U)
B3LYP/6-31G(d) level (method A), and the energies were estimated
on the basis of the R(U)B3LYP/6-31G(d)-optimized structures by
two DFT methods, R(U)B3LYP/6-311++G(d,p) (method B) and
R(U)B3P86/6-311++G(d,p) (method C), at 273.15 K. These DFT
methods were reported to show a good accuracy in calculation of
In order to rationalize the Ames assay results, we computation-
ally evaluated the homolytic C–H bond dissociation energy (BDE)
at the
a-position of the nitrogen atom of the nitrosamine group.
Previous studies have suggested the involvement of the carbon
radical/carbocationic intermediate in the metabolic
a-hydroxyl-
ation of N-nitrosamines (see Fig. 1).1,2,7 Thus, the homolytic BDE
of the relevant C–H bond can be used as a measure of the strength
of the C–H bond and thus of the likelihood of homolytic bond
cleavage, which leads to the putative
a-carbon radical intermedi-
ates for -hydroxylation (Fig. 4). The larger the magnitude of the
a
BDE of the relevant C–H bond, the harder it is to form the carbon
radical/carbocation intermediate,24 and this corresponds to
increasing resistance to metabolic
ness in the Ames assay.
a-hydroxylation, that is, inert-
BDE is experimentally defined as the enthalpy change at 298 K
and 1 atm for the reaction A–B (g)?Aꢀ(g) + Bꢀ(g). The commonly
used Hartree-Fock or perturbation (e.g., MP2, etc.) methods are
not suitable for BDE calculations due to spin-contamination prob-
lems in dealing with open-shell systems.25,26 More sophisticated
ab initio methods such as QCISD and CCSD may perform better than
HF and MP2 for the BDE calculations.27 However, recent studies

2730
T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
Figure 2C. Mutagenicity of compounds 13 (s), 14 (d), 15 (N), 16 (4) and 17 (j) in S. typhimurium TA98 and TA100 in the presence of S9 mix or absence of S9 mix.
the C–H BDE, and in particular, the latter is comparable to the CBS-
Q method.29 Some selected simple molecules were also evaluated
with the CBS-Q method (method D) at 293.15 K. Similar general
trends in the order of magnitude of the BDEs of the relevant C–H
bonds were obtained with both the DFT and CBS-Q calculation
methods.
8b (Table 1 and Fig. 6). The BDEs of 8a are 106.1 [108.9] kcal/mol
and 106.7 [109.4] kcal/mol (method B and [method C]) for the
formation of 8a-anti-Rad and 8a-syn-Rad, respectively. The BDEs
of 8b are 106.0 [108.6] kcal/mol and 106.4 [109.1] kcal/mol
(method B and [method C]) for the formation of 8b-anti-Rad and
8b-syn-Rad, respectively (Table 1 and Fig. 6). The corresponding
BDE values obtained by method
D (CBS-Q) from 8b are
2.2.1. Bicyclic nitrosamines
119.3 kcal/mol (anti-radical) and 119.5 kcal/mol (syn-radical) at
298.15 K, respectively. Thus, the differences in BDEs between the
two isomeric bonds, that is, anti and syn C–H bonds with respect
to the nitrosoamino group, were small, and there is no significant
effect of the initial ground-state conformation of the nitrosamine
on the BDE values (Fig. 6). For the non-benzo derivative 14, the
BDE values were in a similar range of magnitude to those of the
benzo-derivative 8: for the formation of 14-anti-Rad and
14-syn-Rad, the BDEs are 105.3 [107.9] kcal/mol and 105.7
[108.3] kcal/mol (method B and method C), respectively (Table
1). The corresponding value obtained by method D for the least-
energy process to generate 14-anti-Rad is 110.6 kcal/mol at
298.15 K. These BDE values of 8 and 14 are comparable to or rather
larger than those of the C–H bond of simple alkanes, such as cyclo-
pentane (93.1 kcal/mol (method C) and 96.5 kcal/mol (method D)),
respectively.29 This is consistent with the notion that the resultant
bridgehead carbon radicals (8-Rad and 14-Rad, Fig. 4) are not
In the case of the bicyclic nitrosamines, nitrogen pyramidaliza-
tion and cis/trans-isomerism of the N–NO group take place.33 Thus,
four possible ground-state structures at most need to be consid-
ered in the cases of asymmetrical nitrosamines (see Fig. 9). Upon
the putative homolytic cleavage of the C–H bond, there are two
possible
with respect to the N@O group (Fig. 5, and see also Fig. 6). We com-
pared the BDEs of these two unequivalent -C–H bonds.
a-radical formations, that is, anti-radical and syn-radical
a
The C–H BDEs of two simple bicyclic nitrosamines, the benzo
derivative 8 (Fig. 6) and the non-benzo derivative 14 (Fig. 7) were
calculated. For the benzo derivative 8, we calculated two ground-
state conformers 8a (exo-NO, bending of the nitrosamine group
opposite to the aromatic ring) and 8b (endo-NO, bending of the
nitrosamine group toward the aromatic ring; 8a is more stable
than 8b, though the energy difference is less than 1 kcal/mol), as
well as four possible carbon radical species derived from 8a and

T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
2731
Figure 3A. Mutagenicity of compounds 9, 11 and 12 in S. typhimurium strains in the presence or absence of S9 mix.
subject to conjugative stabilization by the adjacent nitrogen
atom of the N–(NO) group and the fused benzene ring, probably
because of the structural features of the present bicyclic system
(see Fig. 8).
bond of the monocyclic nitrosamines and NDMA, which show po-
sitive Ames assay responses. It is noteworthy that the BDE of the
benzo derivative 18 is smaller than that of the non-benzo deriva-
tive 19 by 10 kcal/mol. This indicates that the benzene ring stabi-
lizes the resultant carbon radical in the case of 18 (see Fig. 8),
which is consistent with the strongly positive Ames assay result
of 18 as compared with the weakly positive response of 19
(Fig. 3B).
2.2.2. Monocyclic nitrosamines
In the series of monocyclic nitrosamines, optimized structures
showed that the nitrogen atom of the N–(NO) group of benzo 18
is planar, while non-benzo 19 has a slightly pyramidalized nitro-
gen atom (see Fig. 7 and see Section 2.3). The BDEs of the benzo
derivative 18 are 74.2 [76.9] and 74.6 [77.4] kcal/mol (method B
and [method C]) for generation of 18-anti-Rad and 18-syn-Rad,
respectively. The corresponding value obtained by method D
(CBS-Q) for the least-energy process to generate 18-anti-Rad is
78.7 kcal/mol at 298.15 K (Table 1). For the non-benzo derivative
19, the BDEs are 81.7 [85.9] and 83.4 [84.7] kcal/mol for generation
of 19-anti-Rad and 19-syn-Rad (method B and [method C]), respec-
tively. The corresponding value obtained by method D (CBS-Q) for
generation of 19-anti-Rad is 89.0 kcal/mol at 298.15 K. This value
is comparable in magnitude to that of NDMA (89.2 kcal/mol, meth-
od D) for the generation of NDMA-syn-Rad (Table 1). It was evident
that the BDE values of the monocyclic nitrosamines (18 and 19)
and NDMA are smaller in magnitude than those of the correspond-
ing bicyclic nitrosamines (8 and 14) by about 20–30 kcal/mol.
2.3. Structural difference of carbon radicals derived from bicyclic
and monocyclic nitrosamines
2.3.1. Monocyclic nitrosamines
Nitrosamine 19 has sp3-hybridized carbons (CH2) at the
a-position of the nitrosamino group. The nitrosamine nitrogen
atom of 19 is only very slightly pyramidalized: h1 (the summation
of the three bond angles of the nitrogen atom) is 359.2° (360.0°
when the nitrogen atom is planar) (Fig. 7). Upon radical formation,
the monocyclic radical 19-Rad takes a planar structure with
respect to both the radical carbon center and the nitrosoamino
nitrogen atom: 19-anti-Rad: h1: 360.0°; h2 (the summation of the
three bond angles of the radical carbon atom) is 359.8°; 19-syn-
Rad: h1: 360.0°; h2: 360.0°. The N–C(radical) bond length is also
shortened as compared with the corresponding N–C bond of the
starting nitrosamine 19: 19-anti-Rad: d1: 1.349 Å versus 1.465 Å
These results are consistent with the high reactivity of the a-C–H

2732
T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
Figure 3B. Mutagenicity of compounds 18 and 19 in S. typhimurium strains in the presence (d) or absence of S9 mix (s).
NO
NO
N
2.3.2. Bicyclic nitrosamines
N
On the other hand, the geometrical changes upon formation of
the corresponding carbon radical were significantly attenuated in
the cases of the bicyclic nitrosamines (Fig. 7). The nitrosoamino
nitrogen atom was pyramidalized (i.e., the summation of the three
bond angles of the nitrogen atom is less than 360°) in the starting
bicyclic nitrosamine 14: h1 is 339.9°. Upon radical formation, the
bridgehead radicals (14-anti-Rad and 14-syn-Rad) maintain non-
planar structures with respect to the nitrosoamino nitrogen atom
and also the carbon radical center: 14-anti-Rad: h1: 341.3°, h2:
320.4°; 14-syn-Rad: h1: 340.0°; h2: 319.4°. The shortening of the
N–C(radical) bond length as compared with the corresponding
N–C bond of the starting nitrosamine 14 is rather small:
14-anti-Rad: d1: 1.458 Å versus 1.483 Å (14); 14-syn-Rad: d2:
1.467 Å versus 1.475 Å (14). These structural features are consis-
tent with the postulate that the contribution of the resonance
structure 14-Rad-II (Fig. 8) is small and the formation of the carbon
radical at the bridgehead position is difficult.
H
+
+
H
H
8
H
8-Rad
N
NO
N
NO
18-Rad
18
NO
H
NO
N
N
+
+
H
H
14
H
14-Rad
N NO
N NO
19
19-Rad
2.4. C–H Bond dissociation energies of the mutagenic bicyclic
nitrosamines
Figure 4. Bond dissociation energy of relevant C–H bonds.
(19); 19-syn-Rad: d2: 1.339 Å versus 1.469 Å (19) (Fig. 7). These
structural changes upon carbon radical formation are consistent
with a large contribution of the resonance structure, 19-Rad-II
(Fig. 8), that is, an increase of N–C double bond character.
We also calculated the BDEs of the
genic bicyclic nitrosamines (9, 11 and 12) for comparison with
those of the inert bicyclic nitrosamines (e.g., 10) (Table 1). In the
a-C–H bonds of the muta-

T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
2733
Table 1
Calculated bond dissociation energies (kcal/mol)
Carbon radical speciese
Geometry of NOf
Method A R(U)B3LYPa,b
6-31G(d)
Method B R(U)B3LYPb,c
6-311G++(d,p)
Method C R(U)B3P86b,c
6-311++G(d,p)
Method D CBS-Qd
Bicyclic nitrosamines
8a-anti-Rad
8a-syn-Rad
8b-anti-Rad
8b-syn-Rad
9-anti-Rad
g
g
exo
exo
endo
endo
exo-cis
exo-cis
exo-cis
exo-cis
exo-trans
exo-trans
exo-trans
exo-trans
—
106.2
106.8
106.1
106.6
105.7
106.3
106.8
107.6
107.6
108.0
106.9
106.8
105.6
105.9
106.1
106.7
106.0
106.4
105.4
106.2
106.7
107.1
107.7
108.1
106.9
107.2
105.3
105.7
108.9
109.4
108.6
109.1
108.1
108.8
109.3
109.7
110.3
110.7
109.6
109.8
107.9
108.3
—
—
119.3
119.5
h
—
—
—
—
—
—
—
—
h
9-syn-Rad
h
10-anti-Rad
10-syn-Rad
11-anti-Rad
11-syn-Rad
12-anti-Rad
12-syn-Rad
14-anti-Rad
14-syn-Rad
h
h
h
h
h
110.6
h
—
—
Monocyclic nitrosamines and NDMA
18-anti-Rad
18-syn-Rad
19-anti-Rad
19-syn-Rad
NDMA-anti-Rad
NDMA-syn-Rad
—
—
—
—
—
—
74.6
75.4
83.1
83.7
87.9
86.1
74.2
74.6
81.7
83.4
86.2
84.3
76.9
77.4
85.9
84.7
89.3
87.4
78.7
—
h
89.0
h
—
91.3
89.2
a
b
c
d
e
f
Geometries were fully optimized.
Zero-point energies (obtained by B3LYP/6-31G(d)) were corrected.
Structures optimized on B3LYP/6-31G(d) were used.
At 298.15 K.
Anti and syn-radicals (Rad) are
a-carbon radicals with the radical center anti and syn with respect to the (N)–NO group, respectively. See Figure 5.
Exo-NO and endo-NO refer to the bending of the nitrosoamine group with respect to the aromatic ring; cis-NO and trans-NO refer to the orientation of (N)–NO group with
respect to the aromatic substituent. See Figure 9.
g
The calculations were not convergent.
Not determined.
h
cases of the asymmetrical nitrosamines 11 and 12, there are four
possible ground-state conformers (a: exo-cis-NO, b: exo-trans-NO,
c: endo-cis-NO, d: endo-trans-NO; exo-NO and endo-NO define the
direction of bending of the nitrosoamine group, see also Figure 6;
cis-NO and trans-NO refer to the orientation of the (N)–NO group
with respect to the aromatic substituent) (for 11a–d, Fig. 9). The
relative energy differences among the four conformers were small,
that is, within 0.3–0.7 kcal/mol. The magnitudes of the C–H BDE
were rather constant, regardless of the initial conformation, in
the case of 8 (Table 1 and Fig. 6). Thus, we calculated the C–H
BDE values of 9–12 on the basis of the least-energy ground-state
conformers (Table 1). The BDEs (obtained by method C) of muta-
genic 9 (conformer 9a (exo-cis-NO): 108.1 and 108.8 kcal/mol),
11 (conformer 11b (exo-trans-NO): 110.3 and 110.7 kcal/mol) and
12 (conformer 12b (exo-trans-NO): 109.6 and 109.8 kcal/mol) for
the anti-C–H bond and the syn-C–H bonds are similar in magnitude
to those of Ames-inert 10 (conformer 10a (exo-cis-NO): 109.3 and
109.7 kcal/mol), respectively. These similar magnitudes of the
BDEs are consistent with the idea that all these bicyclic nitrosa-
3. Conclusion
While nitrosamines are generally regarded as carcinogenic, the
N-nitrosamines of 7-azabicyclo[2.2.1]heptane are essentially nega-
tive in the Ames assay with the TA100 and TA98 strains. Some
bicyclic nitrosamines show a positive Ames assay response with-
out enzymatic activation, implying that the activity is independent
of the nitrosamine functionality. The calculated BDEs of the a-C–H
bonds of the present bicyclic nitrosamines (8–12, and 14) are lar-
ger in magnitude than those of the corresponding monocyclic
nitrosamines (18 and 19) and NDMA, and the difference is as large
as 20–30 kcal/mol. These results are consistent with low
bond reactivity of the present bicyclic nitrosamines, and conse-
quently with low mutagenicity in the Ames assay, whereas the
monocyclic nitrosamines and NDMA have higher a-C–H bond reac-
tivity, which is consistent with positive Ames assay responses.
These findings indicate that 7-azabicyclo[2.2.1]heptane is available
as a structural motif for the design of non-genotoxic nitrosamine
compounds.
a-C–H
mines have similar chemical inertness to
a-C–H bond cleavage.
That is, the observed mutagenicity of the bicyclic nitrosamines
4. Experimental section
(9, 11 and 12) is unrelated to
a
-hydroxylation.
4.1. Materials, and reagents for biological assay
Sodium ammonium hydrogen phosphate tetrahydrate was
purchased from Merck (Darmstadt, Germany). Bacto agar and
bacto nutrient broth were obtained from Becton Dickinson Micro-
biology Systems (Sparks, USA). Other reagents used were pur-
chased from Wako Pure Chemical Industries (Osaka, Japan). S9/
cofactor A set for the Ames test was purchased from Oriental
Yeast Co., Ltd (Tokyo, Japan). Dr. T. Nohmi (National Institute of
Health Sciences, Tokyo, Japan) kindly provided S. typhimurium
TA100 and TA98.
O
O
N
N
O
N
N
N
N
H
H
H
H
- H
or
anti-radical
syn-radical
Figure 5. Two isomeric C–H bonds at the a-position of nitrosamine.

2734
T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
NO
N
NO
H
H
N
N NO
19-Rad-II
N NO
19-Rad-I
14-Rad-II
14-Rad-I
Figure 8. Resonance structures to stabilize radicals.
Figure 9. Four possible ground-state conformers of 11a–d. Relative energies (by
method C) are shown in parentheses. Exo-NO and endo-NO refer to the bending of
the nitrosoamine group with respect to the aromatic ring (see also Fig. 6); Cis-NO
and trans-NO refer to the orientation of (N)–NO group with respect to the aromatic
substituent.
Figure 6. Ground-state conformers (8a and 8b) and four possible carbon radical
species of benzonitrosamine 8. Two conformers, 8a (exo-NO) and 8b (endo-NO)
were calculated. Exo-NO and endo-NO refer to the bending of the nitrosamine
group with respect to the aromatic ring. The energy difference obtained by method
C is shown in parentheses. (b) The CAH BDEs from conformer 8a (method C) are
shown in brackets. (c) The CAH BDEs from conformer 8b (method C) are shown in
brackets.
Several tester strains have been employed for assays of muta-
genesis involving N-nitrosamines in the presence of rat liver S9
mix. The his G46 Salmonella tester strains TA1535 and TA100 are
sensitive mainly to damage through base substitution mutation.34
Furthermore some N-nitrosamines having a bulky group in the
structure, for example, N-nitrosophenylurea35 and N-nitrosometh-
ylbenzylamine,36 are weak frame-shift mutagens in S. typhimurium
Figure 7. Structural features of nitrosamines and the corresponding carbon radical species. h1 and h2 are the summation of the three bond angles of the nitrogen and radical
carbon atoms, respectively. d1 and d2 are the bond lengths between the nitrogen and the relevant carbon atoms, respectively.

T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
2735
TA98. Since the presence in Salmonella of the R factor, pKM101,
which is believed to enhance error-prone DNA repair37 enhances
the number of revertants induced by N-nitrosodialkylamines,38,39
the mutagenicity of bicyclic nitrosamines was determined with
the University of Tokyo. Compounds 1–3, 6–8, 14, 15, 18 and 19
were synthesized as previously described.13,33
4.3.2. Synthesis of 2
O
Ph
N Me
O
LiAlH4
N
Ph
Ph
O
THF-MeOH
N
N
-40°C 30 min
CH2Cl2
0°C 3.5 hr
84 %
O
0°C 30 min
N
Me
O
61 %
O
22
21
20
1) NBS
Dioxane-H2O
r.t. 24 hr
N
N
O
2) NaNO2
O
AcOH-H2O
N
O
0°C 2.5 hr
66 %
Me
2
S. typhimurium TA100 and TA98 for detection of base substitution
N-Benzylisoindole (21): To
a
suspension of LiAlH4 (3.61 g,
and frame-shift mutation, respectively.
94.9 mmol) in 75 mL of dry THF under Ar, a solution of 6.5 mL of
methanol in 75 mL of THF was added over 30 min at 0 °C, followed
by cooling of the whole to ꢁ78 °C (in an acetone-dry ice bath). To
this suspension, N-benzylphthalimide 20 (7.23 g, 30.5 mmol) was
added in portions at ꢁ78 °C, and the mixture was stirred at this
temperature for 30 min. Then the mixture was stirred at 0 °C for
another 30 min. A solution of sodium sulfate (1.0 g) in 9 mL of
water was added, and the resultant inorganic salt was filtered by
suction, and the precipitate was washed with acetone. The com-
bined organic layer was dried over sodium sulfate, and the organic
solvent was evaporated below 40 °C. To the obtained residue,
20 mL of ethanol was added, and the resultant suspension was al-
lowed to stand in refrigerator for overnight. The precipitate was
collected and washed with cold ethanol, affording 3.86 g
(18.6 mmol) of N-benzylisoindole 21 (61% yield). 1H NMR
(400 MHz, CDCl3): d 7.52 (2H, m), 7.31 (3H, m), 7.14 (4H, m),
6.92 (2H, m), 5.36 (2H, m).
Compound 22: To a solution of N-benzylisoindole 21 (1.12 g,
5.31 mmol) in 15 mL of CH2Cl2, a solution of N-phenylmaleimide
(613 mg, 5.52 mmol) in 3 mL of CH2Cl2 was added over 5 min at
0 °C. The whole was stirred at 0 °C for 2.5 h. The residue, obtained
after evaporation of the solvent, was column-chromatographed
(n-hexane/ethyl acetate = 3:1) to give the adduct 22 (1.42 g, yield
84%). MS ([M+H]+): 319. 1H NMR (400 MHz, CDCl3): d 7.33–7.18
(9H, m), 4.57 (2H, m), 3.69 (2H, m), 3.38 (2H, m), 2.26 (3H, s).
4.2. Ames assay
The bacterial mutation assay was carried out according to the
pre-incubation method reported by Ames.15 Test compounds were
diluted to the concentrations of 0.03, 0.06, 0.13, 0.25, 0.5 mg/50 lL
DMSO. The DMSO solution at each concentration was put into a
test tube, and then 0.5 mL of S9 mix or 0.1 M sodium phosphate
buffer (pH 7.4) was added, followed by 0.1 mL of a culture of the
tester strain. The mixture was incubated for 30 min at 37 °C with
shaking (60 strokes/min), and then top agar (2 mL) was added.
The mixture was poured onto a minimal-glucose agar plate. After
incubation at 37 °C for 44 h, colonies were counted. All plates were
prepared in duplicate and the experiments were repeated at least
twice. Data are presented as the mean of duplicate determinations.
The results were considered positive if the assay produced repro-
ducible, dose-related increases in the number of revertants, while
a mutagen was considered weakly positive if it produced a repro-
ducible, dose-related increase in the number of revertants, but the
number of revertants was less than double the background number
of revertants.15
4.3. Synthesis of compounds
4.3.1. General methods
Nitrosamine 2: To a solution of the adduct 22 (1.42 g,
All the melting points were measured with a Yanaco Micro
Melting Point Apparatus and are uncorrected. Proton (400 MHz)
NMR spectra were measured on a Bruker Avance 400 NMR spec-
trometer with TMS as an internal reference in CDCl3 as the solvent,
unless otherwise specified. Chemical shifts d are shown in ppm.
Coupling constants are given in hertz. Low-resolution electron-im-
pact (EI) mass spectra (MS, EI+) and high-resolution EI mass spec-
tra (HRMS, EI+) were recorded on a JEOL JMS-O1SG-2 spectrometer
and on a JEOL JMS-SX 102A. Low-resolution FAB mass spectra (MS,
FAB+) and high-resolution FAB mass spectra (HRMS, EI+) were re-
corded on a JEOL MStation JMS-700 spectrometer and on a JEOL
JMS-SX 102A. Electron-spray ionization time-of-flight mass spectra
(ESI-TOF MS, ESI+) were recorded on a Bruker Daltonics, micro-
TOF05. The combustion analyses were carried out in the microan-
alytical laboratory of Graduate School of Pharmaceutical Sciences,
4.46 mmol) in 20 mL of dioxane, a solution of NBS (957 mg,
5.35 mmol) in 8 mL of H2O was added over 2 min at ambient tem-
perature. The mixture was stirred at room temperature for 18 h.
The residue, obtained after evaporation of the solvents, was dis-
solved in 8 mL of acetic acid and 18 mL of water. To this solution,
an aqueous solution of NaNO2 (415 mg, 6.01 mmol) in 10 mL of
water was added over 5 min at 0 °C. The whole was stirred at
0 °C for 2.5 h, then extracted with CHCl3, and the organic layer
was washed with saturated aqueous Na2CO3, brine and dried over
Na2SO4. The organic solvent was evaporated to give a residue,
which was column-chromatographed (n-hexane/AcOEt = 2:1) to
give 2 (758 mg, 2.94 mmol, 55% yield). Mp; 139–140 °C (yellow
powder, recrystallized from n-hexane/CH2Cl2). HRMS ([M+H]+):
Calcd for C13H10N3O4þ: 257.0873. Found: 258.0875. 1H NMR
(400 MHz, CDCl3): d 7.44–7.24 (4H, m), 6.44–6.12 (2H, m),

2736
T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
3.84–3.52 (2H, m), 2.33 (3H, m). Anal. Calcd for C13H11N215N1O3
15NO): C, 60.46; H, 4.29; N, 16.66. Found: C, 60.63; H, 4.45; N,
16.40.
2 N aqueous NaOH and brine, dried over sodium sulfate. The organ-
ic solvent was evaporated to give 25 as a colorless oil (412 mg,
2.09 mmol, 43% yield). 1H NMR (400 MHz, CDCl3): d 6.77 (2H, s),
4.13 (2H, q, J = 7.2 Hz), 3.82 (2H, t, J = 7.2 Hz), 2.63 (2H, t,
J = 7.2 Hz), 1.25 (3H, t, J = 7.2 Hz).
(
4.3.3. Synthesis of 4
O
1) NBS
Ph
N
N
O
N Me
Dioxane-H2O
r.t. 24 hr
O
O
Ph
N
O
N
N Me
N Me
2) NaNO2
AcOH-H2O
0°C 2.5 hr
toluene
reflux in seal tube
8 hr
O
O
21
23
4
Compound 23: The preparation of 23 was carried out in a similar
manner to 22. White powder (1.38 g, 52%). MS ([M+H]+): 319. 1H
NMR (400 MHz, CDCl3): d 7.36–7.34 (2H, m), 7.28–7.21 (5H, m),
7.04–7.01 (2H, m), 4.50 (2H, s), 3.27 (2H, s), 3.10 (3H, s), 2.82 (2H, s).
Nitrosamine 4: The preparation of 4 was carried out in a similar
manner to 2. Mp: 163–167 °C (yellow powder, recrystallized from
Compound 26: The preparation of 26 was carried out
similar manner to 22. MS m/z: 405 ([M+H]+). 1H
NMR (400 MHz, CDCl3): d 7.33–7.18 (9H, m), 4.58 (2H, br s),
4.04 (2H, q, J = 7.2 Hz), 3.70 (2H, br s), 3.36 (2H, br s),
3.18 (2H, t, J = 8.0 Hz), 1.62 (2H, t, J = 8.0 Hz), 1.20 (3H, t,
J = 7.2 Hz).
in
a
O
CO2Et
Ph
N
N
O
N
1) NBS
N
Dioxane-H2O
O
25
Ph
r.t. 17 hr
N
O
N
O
N
CH2Cl2
0°C 3.5 hr
2) NaNO2
AcOH-H2O
5
26
21
O
O
0°C 1.5 hr
CO2Et
CO2Et
methanol). MS ([M+H]+): 258. Anal. Calcd for C13H11N3O3: C, 60.70;
H, 4.31; N, 16.33. Found: C, 60.55; H, 4.43; N, 16.12. 1H NMR
(400 MHz, CDCl3): d 7.51–7.33 (2H, m), 7.30–7.27 (2H, m), 6.38
(1H, br s), 6.18 (1H, br s), 3.18 (1H, br s), 3.04 (1H, br s), 2.96
(3H, m).
Nitrosamine 5: The preparation of 5 was carried out in a similar
manner to 2. Mp: 59.3–61.0 °C (yellow plates, recrystallized from
n-hexane/CH2Cl2). Anal. Calcd for C17H17N3O5: C, 59.47; H, 4.99;
N, 12.24. Found: C, 59.40; H, 5.01; N, 12.08. 1H NMR (400 MHz,
CDCl3): d 7.41–7.26 (4H, m), 6.40–6.13 (2H, m), 4.06 (2H, q,
J = 7.2 Hz), 3.84–3.61 (2H, m), 3.22 (2H, t, J = 8.0 Hz), 1.67 (2H, t,
J = 8.0 Hz), 1.21 (3H, t, J = 7.2 Hz).
4.3.4. Synthesis of 5
4.3.5. Synthesis of 8
O
O
β−Alanine ethyl ester hydrochloride
CO2Et
CH3CO2H
O
N
reflux, 10 hr
43 %
O
O
24
25
Boc
N
CO2H
CH3CN-CH2Cl2
+
+
N Boc
ONO
r.t. to reflux
2.5 hr
NH2
27
N-(Ethyl propionlate)imide 25: The synthesis of 25 was described
previously.40 To a solution of maleic anhydride 24 (480 mg,
4.89 mmol) in 8 mL of glacial acetic acid, b-alanine ethyl ester
hydrochloride (800 mg, 5.21 mmol) was added in portions at ambi-
ent temperature and the mixture was refluxed for 10 h. The whole
was extracted with CHCl3, and the organic layer was washed with
Compound 27: To a solution of N-Boc pyrrole (1.50 g, 9.0 mmol)
in 60 mL of CH3CN was added a solution of anthranilic acid (1.03 g,
7.5 mmol) in 35 mL of CH2Cl2 and a solution of isoamyl nitrite
(1.76 g, 15 mmol) in 30 mL of CH3CN simultaneously at 25 °C over
30 min. The mixture was refluxed for 2 h, then the whole was
cooled to rt. The organic solvents were evaporated to give a reddish

T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
2737
brown oil, which was column-chromatographed (Et2O/n-hex-
ane = 1:3 to 1:2) to give 27 (924 mg, 41%) as a brown oil. 1H
NMR (400 MHz, CDCl3): d 7.26–7.25 (2H, br m), 7.03–6.93 (4H, br
m), 5.45 (2H, br s), 1.38 (9H, s).
Compound 30: The preparation of 30 was carried out in a similar
manner to 27. Brown oil (1.58 g, yield 50%). 1H NMR (400 MHz,
CDCl3): d 7.11–6.91 (4H, m), 5.52–5.39 (2H, m), 3.84 (6H, s), 1.38
(9H, s).
Boc
Boc
H
H2, Pd/C
MeOH
HCOOH
MeOH
N
N
N
r.t. 6 hr
r.t. 8 hr
27
28
29
NO
NaNO2
CH3CO2H-H2O
N
0°C, 2 hr
8
Compound 28: The alkene derivative 27 (1.33 g, 5.47 mmol)
was hydrogenated over 10% Pd–C (148 mg) in methanol for 6 h
at ambient temperature. The reaction mixture was filtrated and
the solvent was evaporated to give 28 as a pale yellow oil.
(1.22 g, 91% yield). MS m/z: 246 ([M+H]+). 1H NMR (400 MHz,
CDCl3): d 7.24–7.21 (2H, d,d, J = 3.2 Hz, 5.3 Hz), 7.14–7.11 (2H,
d,d, J = 3.2 Hz, 5.3 Hz), 5.10 (2H, s), 2.16–2.02 (2H, m), 1.39 (9H,
s), 1.30–1.26 (2H, m).
31: The preparation of 31 was carried out in a similar manner to
28. Yield 81%. MS (M+H)+: 306. 1H NMR (400 MHz, CDCl3): d 6.86
(2H, s), 5.05 (2H, s), 3.86 (6H, s), 2.07 (2H, m), 1.40 (9H, s), 1.23
(2H, m).
Nitrosamine 9: The preparation of 9 was carried out in a similar
manner to 8. Mp: 161–164 °C (pale yellow plates, recrystallized
from THF/n-hexane). Yield 38%. MS (FAB, [(M+H)+]): 235.1. Anal.
Calcd for C12H14N2O3: C, 61.53; H, 6.02; N, 11.96. Found: C,
61.61; H, 6.03; N, 11.94. 1H NMR (400 MHz, CDCl3): d 6.96–6.87
(2H, m), 5.87–5.85 (2H, m), 3.89 (3H, s), 3.87 (3H, s), 2.25–2.17
(1H, m), 2.05–1.98 (1H, m), 1.56–1.51 (1H, m), 1.37–1.33 (1H, m).
Nitrosamine 8: Compound 28 (1.52 g, 6.21 mmol) was dissolved
in 10 mL of formic acid and the mixture was stirred for 8 h at room
temperature under argon. The solvent was evaporated and the res-
idue was washed with 10% aqueous Na2CO3 and the whole was ex-
tracted with CH2Cl2. The organic layer was dried over sodium
sulfate, and evaporated to give the crude of amine derivative 29
(839.6 mg, 5.79 mmol). To a solution of the amine derivative 29
(839.6 mg, 5.79 mmol) in a mixture of 3.8 mL of acetic acid and
5.0 mL of water, an aqueous solution of NaNO2 (500 mg,
7.24 mmol) in 8.0 mL of H2O was added over 5 min at 0 °C (in an
ice-water bath), and the whole was stirred for 2 h. The whole
was extracted with CHCl3, and the organic layer was washed with
brine, and dried over sodium sulfate. The organic solvent was evap-
orated to give a residue, which was column-chromatographed
(AcOEt/n-hexane = 2:5) to give a yellow solid 8 (517 mg, 51%).
Mp; 59.3–61.0 °C (yellow plates, recrystallized from n-hexane/
CH2Cl2). MS (M+H)+: 175.1. Anal. Calcd for C10H10N2O: C, 68.95;
H, 5.79; N, 16.08. Found: C, 69.18; H, 5.63; N, 16.03. 1H NMR
(CDCl3/TMS): d 7.37–7.21 (4H, m), 5.94–5.91 (2H, m), 2.29–2.01
(2H, m), 1.61–1.35 (2H, m).
4.3.7. Synthesis of 10
HCl, EtOH
O2N
CO2H
NH2
O2N
CO2H
isoamyl nitrite
0°C, 1 hr
N2+ Cl-
33
Compound 33: To a solution of 5.0 g (27.5 mmol) of 5-nitro-
anthranilic acid in ethanol (75 mL) was added slowly 2.75 mL of
concentrated hydrochloric acid at 0 °C. The resulting clear solution
was kept below at 0 °C and then 5.5 mL of isoamyl nitrite was
added dropwise (15 min). After the solution was stirred at 0 °C
for 1 h, 85 mL of ether was added, and the mixture was stirred
for another 1 h at 0 °C, during which an orange precipitate was
formed. The product was collected and washed with ether and
dried under vacuum at rt, yield 3.24 g (51%).
4.3.6. Synthesis of 9
Boc
N
MeO
MeO
CO2H
NH2
CH3CN-CH2Cl2
+
+
N
Boc
MeO
MeO
ONO
r.t. to reflux
3 hr
30
NO
N
Boc
H
NaNO2
H2, Pd/C
MeOH
r.t. 7 hr
HCOOH
MeOH
r.t. 6 hr
N
N
CH3CO2H-H2O
MeO
MeO
MeO
MeO
MeO
MeO
0°C, 2 hr
32
9
31

2738
T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
Nitrosamine 10: The preparation of 10 was carried out in a sim-
N Boc
Boc
ilar manner to 8.
Yellow solid (112 mg, 28.5% (from 35)). HRMS ([M+H]+): Calcd
for C10H10N3O3: 220.0723. Found: 220.0718. 1H NMR (400 MHz,
CDCl3): d 8.68–8.63 (1H, m), 8.34–8.37 (1H, m), 7.66–7.64 (1H,
m), 5.97–5.88 (2H, m), 2.33–2.11 (1H, m), 2.11–2.00 (1H, m),
1.66–1.54 (1H, m), 1.45–1.34 (1H, m).
N
O2N
CO2H
Propylene oxide
(CH2Cl)2
r.t. to reflux, 3 hr
N2+ Cl-
O2N
34
33
4.3.8. Synthesis of 11
NO
H
Boc
NaNO2
N
N
HCOOH
N
CH3CO2H-H2O
MeOH
r.t. 8 hr
0°C, 2 hr
O2N
O2N
O2N
11
34
37
Compound 34: To
a
solution of N-Boc-pyrrole (2.72 g,
Compound 34 (443 mg, 1.54 mmol) was dissolved in formic
16.3 mmol) in ethylene dichloride (20 mL), a solution of 2-car-
boxy-4-nitro-benzenediazonium chloride 33 (3.74 g, 16.3 mmol)
in 20 mL of ethylene dichloride was added at ambient tempera-
ture. To this stirred solution 3 mL of propyleneoxide in 10 mL of
ethylene dichloride was added and then the solution was gradu-
ally warmed. As gas evolution commenced, the temperature was
controlled to prevent vigorous gas foaming. After 10 min of gas
evolution the solution became clear. The whole was heated at
reflux for 3 h, the organic solvent was removed under vacuum,
and the brown liquid residue was taken up in ether, washed
with aqueous sodium hydroxide and water, and dried over so-
dium sulfate. The solvent was removed at reduced pressure
and the residue was column-chromatographed (Et2O/n-hex-
ane = 1:3 to 1:2) to give 34 (871 mg, 3.02 mmol, 19%) as a red-
dish-brawn oil. HRMS ([M+Na]+): Calcd for C15H16N2O4Na:
311.1008. Found: 311.1005. 1H NMR (400 MHz, CDCl3): d 8.06
(1H, d, J = 2.02 Hz), 7.97–7.94 (1H, d, d, J = 7.88 Hz, 2.02 Hz),
7.37 (1H, d, J = 7.88 Hz), 7.01 (2H, d, J = 16.3 Hz), 5.58 (2H, s),
1.38 (9H, s).
acid (8 mL). The reaction mixture was stirred for 8 h at rt (under
Ar). The solvent was then evaporated and the residue was washed
with 10% aqueous Na2CO3 and extracted with CH2Cl2. The organic
phase was dried over Na2SO4 and evaporated to afford crude 37
(185 mg, 0.99 mmol). To a solution of the amine derivative 37
(185 mg, 0.99 mmol) in a mixture of 1.2 mL of acetic acid and
1.5 mL of water, an aqueous solution of NaNO2 (152 mg,
2.20 mmol) in 1.2 mL of H2O was added over 2 min at 0 °C (in
an ice-water bath), and the whole was stirred for 1 h. The whole
was extracted with CHCl3, and the organic layer was washed with
brine, and dried over sodium sulfate. The organic solvent was
evaporated to give
a residue, which was column-chromato-
graphed (AcOEt/n-hexane = 2:5 to 2:3, gradient elution) to give
11 (134.6 mg, 40% from 34) as an oil. MS (FAB, [(M+H)+]):
218.1. 1H NMR (400 MHz, CDCl3): d 8.32–8.24 (1H, m), 8.06–
8.00 (1H, m), 7.76–7.65 (1H, s), 7.39–7.65 (2H, m), 6.94 (1H, s),
6.48 (1H, m).
4.3.9. Synthesis of 12
1) 9-BBN
THF
Boc
Boc
H
HCOOH
MeOH
N
N
N
r.t. 24 hr
OH
OH
r.t. 8 hr
2) H2O2
2N NaOH
r.t. 6 hr
O2N
NO
O2N
O2N
38
34
NaNO2
39
N
CH3CO2H-H2O
OH
0°C, 2 hr
O2N
12
Compound 38: The preparation of 38 was carried out in a sim-
Boc
Boc
H
ilar manner to 40. Pale orange powder (yield 50%). HRMS
([M+Na]+): Calcd for C15H18N2NaO5: 329.1108. Found: 329.1113.
1H NMR (400 MHz, CDCl3): d 8.15–8.14 (1H, m), 8.12–8.10 (1H,
m), 7.37–7.35 (1H, m), 5.22–5.21 (1H, m), 5.11 (1H, br s), 1.97–
1.87 (1H, m), 1.39 (9H, s), 1.26–1.21 (1H, m).
Nitrosamine 12: The preparation of 12 was carried out in a sim-
ilar manner to 13. Yield 45%. Mp: 151–153 °C (black powder,
recrystallized from Et2O/n-hexane). Anal. Calcd for C10H9N3O4: C,
51.07; H, 3.86; N, 17.87. Found: C, 50.96; H, 4.01; N, 17.64. 1H
NMR (400 MHz, DMSO): d 8.41–8.11 (2H, m), 7.82–7.61 (1H, m),
N
N
H2, Pd/C
MeOH
r.t. 7 hr
HCOOH
MeOH
r.t. 8 hr
O2N
O2N
O2N
35
34
NO
NaNO2
CH3CO2H-H2O
N
N
0°C, 2 hr
O2N
10
36

T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
2739
6.33–5.63 (3H, m), 4.11–3.95 (1H, m), 2.10–1.81 (1H, m), 1.60–1.54
(1H, m).
Nitrosamine 16: Compound 42 (86.8 mg, 0.2900 mmol)42 was
dissolved in TFA 1 mL at 0 °C and the reaction mixture was stirred
for 45 min at this temperature. Then TFA was evaporated to give
the crude amine. To a solution of the resultant amine derivative
4.3.10. Synthesis of 13
Boc
1) 9-BBN
THF
Boc
H
N
HCOOH
MeOH
N
N
r.t. 24 hr
OH
OH
2) H2O2
2N NaOH
r.t. 6 hr
r.t. 8 hr
27
40
41
NO
NaNO2
CH3CO2H-H2O
N
OH
0°C, 2 hr
13
Compound 40: 9-BBN solution in THF (0.4 M, 32.3 mL,
12.9 mmol) was added to alkene 27 (2.61 g, 10.8 mmol) over
5 min at rt under argon. The mixture was stirred for 24 h at rt,
then cooled to 0 °C. H2O2 (13.5 mL of a 35% aqueous solution)
was added dropwise, followed by NaOH (7.8 mL of a 2 N aqueous
solution). The whole was allowed to warm to 25 °C and stirred for
6 h. The aqueous layer was saturated with K2CO3, and the organic
layer was extracted with Et2O, dried over sodium sulfate. The sol-
vent was evaporated to give a residue, which was column-chro-
matographed (Et2O/n-hexane = 1:1 to 2:1) to give 40 as pale
yellow oil (2.08 g, yield 74%). MS (M+H)+: 262. 1H NMR
in 1 mL of acetic acid, an aqueous solution of NaNO2 (74.7 mg,
1.083 mmol) in 2 mL of H2O was added over 1 min at 0 °C, and
the whole was stirred for 2.5 h at this temperature. The reaction
mixture was poured into water (50 mL) and extracted with Et2O
(25 mL ꢂ 3). The organic layer was washed with water and brine
(5 mL). The organic layer was dried over Na2SO4 and evaporated.
Purification by column-chromatography (n-hexane/AcOEt = 3:1)
gave compound 16 (48.6 mg, 0.2129 mmol, 73% yield) as a yellow
oil. 1H NMR (CDCl3): d = 5.21–5.18 (1H, m), 4.12 (1H, d, J = 10.4 Hz),
4.09 (1H, d, J = 10.8 Hz), 3.73 (3H, s), 3.48 (3H, s), 3.03–2.99 (1H,
m), 2.25–1.63 (6H, m). 13C NMR (CDCl3): d =171.76, 70.26, 69.53,
59.65, 54.40, 52.28, 43.15, 35.09, 31.89, 23.02. HRMS (ESI-TOF,
[M+Na]+): Calcd for C10H16N2NaO4þ, 251.1002. Found: 251.1001.
Anal. Calcd for C10H16N2O4: C, 52.62; H, 7.07; N, 12.27. Found: C,
52.62; H, 6.91; N, 12.27.
(400 MHz,CDCl3):
d
7.27–7.23 (2H, m), 7.19–7.17 (2H, m),
5.11 (1H, s), 5.01 (1H, s), 4.04–4.00 (1H, m), 3.20 (1H, br s),
1.87–1.84 (2H, m), 1.38 (9H, s).
Nitrosamine 13: Compound 40 (1.24 g, 4.74 mmol) was dis-
solved in a mixture of formic acid (10 mL) and methanol (5 mL)
and the reaction mixture was stirred for 8 h at room temperature
under argon atmosphere. Then the solvent was evaporated and
the whole was washed with 10% aqueous Na2CO3 and the residue
was extracted with CH2Cl2. The solvent was dried over sodium sul-
fate and evaporated to give crude 41 (735 mg, 4.57 mmol). To a
solution of the amine derivative 41 (735 mg, 4.57 mmol) in a mix-
ture of 3.0 mL of acetic acid and 8.0 mL of H2O, an aqueous solution
of NaNO2 (434 mg, 6.29 mmol) in 4.0 mL of H2O was added over
5 min at 0 °C, and the whole was stirred for 2 h. The reaction mix-
ture was extracted with CHCl3, and the organic layer was washed
with brine, and dried over sodium sulfate. The organic solvent
was evaporated to give a residue which was chromatographed
(AcOEt/n-hexane = 1:2 to 1:1) to give 13 (535 mg, yield 59% from
40). Mp; 74.0–77.0 °C (pale black plates, recrystallized from n-hex-
ane/Et2O). HRMS (ESI, [M+H]+): Calcd for C10H11N2O2: 191.0821.
Found: 191.0822. Anal. Calcd for C10H10N2O2: C, 63.15; H, 5.30;
N, 14.73. Found: C, 63.25; H, 5.30; N, 14.62. 1H NMR (400 MHz,
CDCl3): d 7.42–7.17 (4H, m), 5.96–5.75 (2H, m), 4.26–4.15 (1H,
m), 2.18–2.10 (1H, m), 2.00–1.95 (1H, m), 1.79–1.75 (1H, m).
4.3.12. Synthesis of 17
Boc
CH2OBn
NO
CH2OBn
N
1) TFA, 0 C, 40min
N
2) NaNO2, AcOH-H2O
0 C, 2 hrs,
90% from 43
CO2Me
CO2Me
43
17
Nitosamine 17: Compound 43 (74.7 mg, 0.1990 mmol)41 was
dissolved in TFA 1.0 mL at 0 °C and the reaction mixture was stir-
red for 40 min at this temperature. Then TFA was evaporated and
to give the crude amine. To a solution of the resultant amine deriv-
ative in 0.5 mL of acetic acid, an aqueous solution of NaNO2
(60.0 mg, 0.8696 mmol) in 1.0 mL of H2O was added over 1 min
at 0 °C, and the whole was stirred for 2 h at this temperature.
The reaction mixture was poured into water (40 mL) and extracted
with Et2O (30 mL ꢂ 3). The organic layer was washed with brine
(5 mL). The organic layer was dried over Na2SO4 and evaporated.
Purification by column-chromatography (n-hexane/AcOEt = 3:1)
gave compound 17 (54.5 mg, 0.1791 mmol, 90% from 43) as a yel-
low oil. 1H NMR (CDCl3): d = 7.36–7.30 (5H, m), 5.21 (1H, m), 4.66
(2H, s), 4.20 (1H, d, J = 10.8 Hz), 4.16 (1H, d, J = 10.4 Hz), 3.73 (3H,
s), 3.02–2.99 (1H, m), 2.28–1.62 (6H, m). 13C NMR (CDCl3): d
= 171.68, 137.74, 128.32, 127.67, 127.55, 73.58, 69.52, 67.71,
54.35, 52.20, 43.09, 35.15, 31.97, 22.96. HRMS (ESI-TOF,
[M+Na]+): Calcd for C16H20N2NaO4þ, 327.1321. Found: 327.1309.
Anal. Calcd for C16H20N2O4: C, 63.14; H, 6.62; N, 9.20. Found: C,
63.00; H, 6.60; N, 9.12.
4.3.11. Synthesis of 16
Boc
CH2OMe
NO
CH2OMe
N
N
1) TFA, 0 °C, 45 min
2) NaNO2, AcOH-H2O
0 °C, 2.5 hrs,
CO2Me
CO2Me
42
16
73% from 42

2740
T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
4.3.13. Synthesis of N-nitrosoisoindoline 18
The synthesis of 18 was carried out as described previously
(Ref. 33).
ergy conformation with respect to the two methoxy groups was
used for the DFT calculations.
Benzylamine
Et3N, CHCl3
H2, Pd/C
Ph
Br
Br
MeOH
N
NH
reflux, 10 hr
r.t. 8 hr
45
44
NaNO2
CH3CO2H-H2O
N NO
0°C, 2 hr
18
Isoindoline 45: To a solution of
a,a-dibromo-o-xylene (5.23 g,
Acknowledgments
19.9 mmol) in CHCl3 (15 mL), a solution of triethylamine (6 mL)
in CHCl3 (10 mL) was added at 0 °C over 5 min. To this solution,
benzylamine (2.21 g, 20.6 mmol) in 10 mL of CHCl3 was added
at 0 °C over 5 min. The whole was heated at reflux for 10 h and
the solvent was evaporated to give a residue (44). The N-benzyl
derivative 44 was hydrogenated over 10% Pd–C (420 mg) in meth-
anol (15 mL) for 8 h at ambient temperature. The reaction mix-
ture was filtrated to remove Pd–C and the solvent was
evaporated to give amine derivative 45. 1H NMR (400 MHz,
CDCl3): d 7.43–7.13 (9H, m), 3.91 (4H, s), 3.88 (2H, s). MS
([M+H]+): 210.
N-Nitrosoisoindoline 18: To a solution of isoindoline 45 in a mix-
ture of 10 mL of acetic acid and 20 mL of water, an aqueous solu-
tion of NaNO2 (1.53 g, 22.2 mmol) in 10 mL of water was added
at 0 °C. The whole was stirred for 2 h, then extracted with CH2Cl2,
and the organic layer was washed with 2 N aqueous NaOH, 0.5 N
aqueous HCl, brine, and dried over sodium sulfate. The solvent
was evaporated to give a residue, which was chromatographed
(hexane: AcOEt = 1: 4 to 1: 3) to give 18 as an oil (1.32 g,
This work was supported by a Grant-in-Aid for Scientific
Research from the Japan Society for the Promotion of Science.
Some of the calculations were carried out at the Computer
Center, Institute for Molecular Science at Okazaki, Japan. We
thank the computational facility for generous allotments of
computer time.
References and notes
1. Lee, V. M.; Keefer, L. K.; Archer, M. C. Chem. Res. Toxicol. 1996, 9, 1319.
2. Inami, K.; Ishikawa, S.; Mochizuki, M. Genes Environ. 2009, 31, 97.
3. Mueller, L.; Mauthe, R. J.; Riley, C. M.; Andino, M. M.; De Antonis, D.; Beels, C.;
DeGeorge, J.; De Knaep, A. G. M.; Ellison, D.; Fagerland, J. A.; Frank, R.; Fritschel,
B.; Galloway, S.; Harpur, E.; Humfrey, C. D. N.; Jacks, A. S.; Jagota, N.;
Mackinnon, J.; Mohan, G.; Ness, D. K.; O’Donovan, M. R.; Smith, M. D.;
Vudathala, G.; Yotti, L. Regul. Toxicol. Pharmacol. 2006, 44, 198.
4. Yang, C. S.; Yoo, J. S. H.; Ishizaki, H.; Hong, J. Y. Drug Metab. Rev. 1990, 22, 147.
5. Phillipson, C. E.; Ioannides, C. Carcinogenesis 1984, 5, 1091.
6. Mochizuki, M.; Anjo, T.; Sekiguchi, N.; Ikarashi, A.; Suzuki, A.; Wakabayashi, Y.;
Okada, M. Chem. Pharm. Bull. 1986, 34, 3956.
7. Jarman, M.; Manson, D. Carcinogenesis 1986, 7, 559.
8. Domoradzki, J.; Pegg, A. E.; Dolan, M. E.; Maher, V. M.; McCormick, J. J.
Carcinogenesis 1984, 5, 1641.
9. Kaina, B.; Fritz, G.; Mitra, S.; Coquerelle, T. Carcinogenesis 1991, 12, 1857.
10. Floyd, R. A.; Soong, L. M.; Stuart, M. A.; Reigh, D. L. Photochem. Photobiol. 1978,
28, 857.
8.91 mmol, 45% from
a,a-dibromo-o-xylene). Mp: 85–89 °C
(recrystallized from MeOH, grayish needles). MS ([M+H]+): 149.
1H NMR (400 MHz, CDCl3): d 7.40–7.31 (4H, m), 5.63 (2H, s), 4.90
(2H, s).
N-Nitrosopyrrolidine (19): N-nitrosopyrrolidine (19) was syn-
thesized by nitrosation of pyrrolidine and purified by column-
chromatography as described previously.33 Further purification
was carried out in the present study: the synthesized N-nitroso-
pyrrolidine (19) was dissolved in methanol saturated with
sodium hydroxide and the whole solution was stirred
overnight at room temperature, then purified by extraction
(see Ref. 18d).
11. Voutchkova, A. M.; Ferris, L. A.; Zimmerman, J. B.; Anastas, P. T. Tetrahedron
2010, 66, 1031.
12. Voutchkova, A. M.; Osimitz, T. G.; Anastas, P. T. Chem. Rev. 2010, 110, 5845.
13. Yanagimoto, T.; Toyota, T.; Matsuki, N.; Makino, Y.; Uchiyama, S.; Ohwada, T. J.
Am. Chem. Soc. 2007, 129, 736.
14. Hess, D. T.; Matsumoto, A.; Kim, S. O.; Marshall, H. E.; Stamler, J. S. Nat. Rev. Mol.
Cell Biol. 2005, 6, 150.
15. McCann, J.; Choi, E.; Yamasaki, E.; Ames, B. N. Proc. Natl. Acad. Sci. U.S.A. 1975,
72, 5135.
16. Ames, B. N.; McCann, J.; Yamasaki, E. Mutat. Res. 1975, 31, 347.
17. Mortelmans, K.; Zeiger, E. Mutat. Res., Fund. Mol. Mech. Mutagen. 2000, 455,
29.
18. In the metabolism of NDMA, a denitrosation pathway has been also proposed
and this pathway can account for a part of the NDMA metabolism: (a) Kier, L.
E.; Brusick, D. J.; Auletta, A. E.; Vonhalle, E. S.; Brown, M. M.; Simmon, V. F.;
Dunkel, V.; McCann, J.; Mortelmans, K.; Prival, M.; Rao, T. K.; Ray, V. Mutat. Res.
1986, 168, 69; (b) Keefer, L. K.; Anjo, T.; Wade, D.; Wang, T. Y.; Yang, C. S. Cancer
Res. 1987, 47, 447; (c) Streeter, A. J.; Nims, R. W.; Sheffels, P. R.; Heur, Y. H.;
Yang, C. S.; Mico, B. A.; Gombar, C. T.; Keefer, L. K. Cancer Res. 1990, 50, 1144;
(d) Tsutsumi, N.; Inami, K.; Mochizuki, M. Bioorg. Med. Chem. 2010, 18, 8284;
(e) Wink, D. A.; Kasprzak, K. S.; Maragos, C. M.; Elespuru, R. K.; Misra, M.;
Dunams, T. M.; Cebula, T. A.; Koch, W. H.; Andrews, A. W.; Allen, J. S.; Keefer, L.
K. Science 1991, 254, 1001. In this pathway, the formation of nitric oxide (NO),
which may cause lethal cell injury, is possible. While some involvement of this
alternative metabolic pathway of N-nitrosamines cannot be ruled out, the
present observation that most of these bicyclic nitrosamines showed no
cytotoxicity in the presence of S9 mix may exclude a significant contribution of
denitrosation in the present cases.
4.4. Calculations
All conformations of each molecule were optimized in the
Gaussian 03 program42 at the B3LYP/6-31G(d) (method A) or
CBS-Q level (method D) of theory to find the global minimum.
The number of imaginary frequencies confirmed ground states.
Zero-point energy corrections were applied without scaling.
Single-point energies at 273.15 K were obtained by two DFT
methods, R(U)B3LYP/6-311++G(d,p) (method B) and R(U)B3P86/
6-311++G(d,p) (method C) on the basis of the R(U)B3LYP/
6-31G(d)-optimized structures. The lowest-energy structure of
9 with respect to two methoxy groups was obtained with Mac-
roModel 8.1 programs; Monte Carlo (MC) conformational search
was performed using the OPLS-AA force field in the gas phase.
One thousand starting structures were generated. The lowest-en-
19. Rao, T. K.; Cox, J. T.; Allen, B. E.; Epler, J. L.; Lijinsky, W. Mutat. Res. 1981, 89, 35.
20. Haworth, S.; Lawlor, T.; Mortelmans, K.; Speck, W.; Zeiger, E. Environ. Mutagen.
1983, 5, 3.
21. Dellarco, V. L.; Prival, M. J. Environ. Mol. Mutagen. 1989, 13, 116.

T. Ohwada et al. / Bioorg. Med. Chem. 19 (2011) 2726–2741
2741
22. Assmann, N.; Emmrich, M.; Kampf, G.; Kaiser, M. Mutat. Res., Genet. Toxicol.
Environ. Mutagen. 1997, 395, 139.
37. McCann, J.; Spingarn, N. E.; Koboriand, J.; Ames, B. N. Proc. Natl. Acad. Sci. U.S.A.
1975, 72, 979.
23. 2-Nitronapthalene was reported to be positive in the Ames test (Ref. 21). It
may be assumed that the olefin derivative 11 can undergo homoallylic
conjugation, which may lead to electronically equivalent 2-nitronaphthalene.
24. Meshram, G. P.; Malini, R. P.; Rao, K. M. Mutat. Res. 1992, 279, 275.
25. Parkinson, C. J.; Mayer, P. M.; Radom, L. J. Chem. Soc., Perkin Trans. 2 1999, 2305.
26. Parkinson, C. J.; Mayer, P. M.; Radom, L. Theor. Chem. Acc. 1999, 102, 92.
27. Chuang, Y. Y.; Coitino, E. L.; Truhlar, D. G. J. Phys. Chem. A 2000, 104, 446.
28. Song, K. S.; Cheng, Y. H.; Fu, Y.; Liu, L.; Li, X. S.; Guo, Q. X. J. Phys. Chem. A 2002,
106, 6651.
29. Feng, Y.; Liu, L.; Wang, J. T.; Huang, H.; Guo, Q. X. J. Chem. Inf. Comput. Sci. 2003,
43, 2005.
30. Ochterski, J. W.; Petersson, G. A.; Montgomery, J. A. J. Chem. Phys. 1996, 104,
2598.
31. Mayer, P. M.; Parkinson, C. J.; Smith, D. M.; Radom, L. J. Chem. Phys. 1998, 108, 604.
32. Henry, D. J.; Parkinson, C. J.; Mayer, P. M.; Radom, L. J. Phys. Chem. A 2001, 105,
6750.
33. Ohwada, T.; Miura, M.; Tanaka, H.; Sakamoto, S.; Yamaguchi, K.; Ikeda, H.;
Inagaki, S. J. Am. Chem. Soc. 2001, 123, 10164.
34. Guttenplan, J. B. Mutat. Res. 1978, 186, 81.
35. Lijinsky, W.; Andrews, A. W. Mutat. Res. 1979, 68, 1.
36. Wiessler, M.; Romruen, K.; Pool, B. L. Carcinogenesis 1983, 4, 867.
38. Perry, K. L.; Walkler, G. C. Nature 1982, 300, 278.
39. Zielinska, M.; Guttenplan, J. B. Environ. Mutagen. 1986, 8, 95.
40. Rich, D. H.; Gesellchen, P. D.; Tong, A.; Cheung, A.; Buckner, C. K. J. Med. Chem.
1975, 18, 1004.
41. Hosoya, M.; Otani, Y.; Kawahata, M.; Yamaguchi, K.; Ohwada, T. J. Am. Chem.
Soc. 2010, 132, 14780.
42. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montogomery, J. A., Jr.; Kudin, K. N.; Burant, J. C.; Millam, J.
M.; Lyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.;
Rega, N.; Peterson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda,
R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene,
M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, G.; Dapprich, S.; Daniels, A. K.; Foresman, J. B.;
Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu,
G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-
Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. GAUSSIAN 03;
Gaussian: Pittsburgh, PA, 2003.