Design and synthesis of orally bioavailable inhibitors of inducible nitric oxide synthase. Synthesis and biological evaluation of dihydropyridin-2(1H)-imines and 1,5,6,7-tetrahydro-2H-azepin-2-imines

Article abstract of DOI:10.1016/S0968-0896(02)00540-0

The process of discovery and biological evaluation of α,β-unsaturated cyclic amidines, as selective inhibitors of inducible nitric oxide synthase (iNOS), is reported. Dihydropyridin-2(1H)-imines and 1,5,6,7-tetrahydro-2H-azepin-2-imines were synthesized a

Full text of DOI:10.1016/S0968-0896(02)00540-0

Bioorganic & Medicinal Chemistry 11 (2003) 689–702
Design and Synthesis of Orally Bioavailable Inhibitors of
Inducible Nitric Oxide Synthase. Synthesis and Biological
Evaluation of Dihydropyridin-2(1H)-imines and
1,5,6,7-Tetrahydro-2H-azepin-2-imines
Yasufumi Kawanaka,a,* Kaoru Kobayashi,b Shinya Kusuda,b Tadashi Tatsumi,b
Masayuki Murota,b Toshihiko Nishiyama,b Katsuya Hisaichi,b Atsuko Fujii,b
Keisuke Hirai,b Minoru Nishizaki,b Masao Naka,b Masaharu Komeno,b
Hisao Nakaib and Masaaki Todab
aFukui Research Institute, Ono Pharmaceutical Co., Ltd., Technoport, Yamagishi, Mikuni, Sakai, Fukui 913-8538, Japan
bMinase Research Institute, Ono Pharmaceutical Co., Ltd., Shimamoto, Mishima, Osaka 618-8585, Japan
Received 13 September 2002; accepted 17 October 2002
Abstract—The process of discovery and biological evaluation of a,b-unsaturated cyclic amidines, as selective inhibitors of inducible
nitric oxide synthase (iNOS), is reported. Dihydropyridin-2(1H)-imines and 1,5,6,7-tetrahydro-2H-azepin-2-imines were synthesized
and biologically evaluated both in vitro and in vivo using a nitric oxide synthase inhibition assay. Compounds 1, 5, 6, 8–12 and 16
exhibited potent inhibition of iNOS. Among these, compounds 6, 7, 10, 11 and 16 showed 5- to 19-fold isoform selectivity. Com-
pounds 1, 6, 10, 11 and 16 also showed potent inhibitory activity in the NOx accumulation assay in mice. Compounds 1 and 6
showed excellent bioavailability (BA) in rats when administered orally. Full details are presented here, including the structure–
activity relationship (SAR) studies, the chemistry of these compounds, and the pharmacokinetic data and the computer-aided
docking study of 10 with hiNOS.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
blood pressure regulation and control of platelet aggre-
gation. In contrast, the iNOS is found in activated
macrophages and other tissues, and it seems to play a
role in the pathogenesis of various diseases. Over-
expression of NO, which damages tissues during acute
and chronic inflammation, is caused by iNOS. Identifi-
cation of potent and selective iNOS inhibitors has been
a subject of intense interest, because of their therapeutic
potential for the treatment of diseases mediated by
overexpression of NO.4,5 The natural substrate for
NOS, arginine, has been the obvious basis for the
molecular design of NOS inhibitors.
In recent years, nitric oxide (NO) has emerged as one of
the important biological mediators that plays a sig-
nificant role in various physiological processes.1,2 NO is
produced by nitric oxide synthase (NOS), which cata-
lyzes the oxidation of l-arginine to l-citrulline, and is
synthesized by two major classes of NOS, termed con-
stitutive NOS (cNOS) and inducible NOS (iNOS).3 The
cNOS is essential to maintain homeostasis and com-
prises both neuronal NOS (nNOS) and endothelial NOS
(eNOS). The nNOS is believed that it appears to func-
tion as a neurotransmitter, which regulates neuronal
transmission. The eNOS is found predominantly in the
vascular endothelium and it continuously produces low
concentrations of NO, which have an important role in
Structural analogues of l-arginine, such as NG-methyl-l-
arginine (l-NMA),6,7 NG-nitro-l-arginine,8 N-imi-
noethyl-l-ornithine (l-NIO),9 and l-thiocitrulline,10 have
been shown to inhibit the various forms of NOS, but lack
selectivity for iNOS. Abundant pharmacological and
biological data have been obtained utilizing some of the
early nonspecific arginine analogue inhibitors. A number
*Corresponding author. Tel.: +81-776-82-6161; fax: +81-776-82-
8420; e-mail: kawanaka@ono.co.jp
0968-0896/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00540-0

690
Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
of non-amino acid inhibitors of iNOS, such as amino-
guanidine,11 isothioureas,12 2-amino-5,6-dihydro-6-
Chemistry
methyl-4H-1,3-thiazine,13
2-iminopiperidine14
and
Synthesis of the compounds listed in Tables 1 and 2 is
outlined in Schemes 1–5b. As shown in Scheme 1, com-
pounds 1 and 3 were synthesized from 20 by Birch
reduction, followed by the appropriate work up proce-
dure, at a 66% yield (Method A) and a 64% yield
(Method B), respectively.17 Application of this partial
reduction procedure to 2215 and 2318 resulted in 2 and 4
at a yield of 79 and 75%, respectively.
2-aminopyridine,15 have also been described, and we
recently reported on 5,6-dihydropyridin-2(1H)-imine.16
As shown in Chart 1, our molecular design process started
with structural hybridization of 2-amino-4-methylpyr-
idine 2015 and (4R)-4-methylpiperidin-2(1H)-imine 21.14
This modification led us to the discovery of 4-methyl-5,6-
dihydropyridin-2(1H)-imine 1,16 which is structurally new
and possesses no asymmetric center. Further chemical
modification was continued. Ring expansion of 1, fol-
lowed by introduction of alkyl group at position-7, gave
16, which showed increased iNOS inhibition and better
isoform selectivity (Chart 1). Two series of cyclic amidines
were synthesized and biologically evaluated in an attempt
to identify a new selective inhibitor of human iNOS
(hiNOS). Here we report on a series of 5,6-dihydropyr-
idin-2(1H)-imines and 1,5,6,7-tetrahydro-2H-azepin-2-
imines, which are structurally new iNOS inhibitors.
As described in Scheme 2a, the common intermediates
[substituted 5,6-dihydropyridin-2(1H)-ones 27a–i] were
prepared from the corresponding cyclopentenones 24a–i
by conventional Beckmann rearrangement. Conversion
of 24a and 24e–i to their corresponding oximes 25a and
25e–i, followed by the O-tosylation, provided 26a and
26e–i. Then Beckmann rearrangement of these com-
pounds was successfully carried out to give 27a and
27e–i (Method C: 22–48% yield), respectively. Direct
Beckmann rearrangement of 24b–d also resulted in 27b–
d (Method D: 44–60% yield), respectively.
Table 1. Inhibitory activities of 1–13 against hiNOS, heNOS and
their isoform selectivityc
Structure
Compd
hiNOS heNOS
Selectivity
IC50 (mM)IC50 (mM)heNOS/hiNOSa
1 R=Me
2 R=Et
0.20
1.2
0.23
1.5
1.2
1.3
3 R=H
0.70
2.7
0.44
4.8
0.6
1.8
4 R=Me
Chart 1. Molecular design of 5,6-dihydropyridin-2(1H)-imine (1 and
9–11) and 1,5,6,7-tetrahydro-2H-azepin-2-imine (14 and 16).
5
0.31
0.32
1.0
Table 2. Inhibitory activities of 14–19 against hiNOS, heNOS and
their isoform selectivityc
6 R=Me
7 R=H
0.42
2.8
2.1
5.0
Structure
Compd
hiNOS
heNOS
Selectivity
IC50 (mM) IC50 (mM) heNOS/hiNOSa
31.0
11.1
14 R=H
15 R=Meb
1.1
0.68
4.6
0.42
1.0
4.2
0.6
19.2
16 R=n-Prb 0.052
8b
0.51
0.52
1.0
17b
0.82
2.1
2.6
9 R=Meb
0.18
0.06
0.24
0.58
0.36
0.51
7.4
1.3
9.7
7.2
3.0
5.3
10 R=n-Prb
11 R=Allylb 0.05
18 R=H
19 R=n-Prb 1.6
1.4
24.6
17.4
17.6
10.9
12 n=1b
13 n=2b
0.17
1.4
aSelectivity was evaluated as the rate of the IC50 values for heNOS and
hiNOS.
bSynthesized as a mixture of racemates.
cAll compounds were prepared as their hydrochlorides.
aSelectivity was evaluated as the rate of the IC50 values for heNOS and
hiNOS.
bSynthesized as a mixture of racemates.
cAll compounds were prepared as their hydrochlorides.

Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
691
Among the cyclopentenones used in the above-men-
tioned synthesis, 24e,19a 24f and 24g19b were prepared
by alkylation of 24a (28–55% yield), as described in
Scheme 2b. Compounds 24a and 24b are commercially
available. The methods for synthesis of 24c–d19c,19d
24h19e and 24i19f are reported. As described in Scheme 3,
another dihydropyridin-2(1H)-one 30 was prepared by
oxidative dehydrogenation of 2920a (70% yield), which
was prepared from 2820b by direct Beckmann rearran-
gement (52% yield).
Synthesis of 5,6-dihydropyridin-2(1H)-imines 1 and 5–
13 is described in Scheme 4. Dihydropyridin-2(1H)-ones
27a–i and 30 were converted to the corresponding com-
pounds 31a–i and 32, after which aminolysis and treat-
ment with hydrogen chloride provided 1 and 5–13 (8–
70% yield), respectively.
Scheme 1. Synthesis of 5,6-dihydropyridin-2(1H)-imines (1 and 2) and
3,6-dihydropyridin-2(1H)-imines (3 and 4). Reagents: (a) Li, liquid
NH3, THF/EtOH, ꢀ78ꢁC; (b) NH4Cl; (c) H2O (64–79%).
ꢁ
.
Scheme 2. a. Synthesis of 5,6-dihydropyridin-2(1H)-ones (27a–i). Method C: (a) NH2OH HCl, NaOAc, MeOH, reflux; (b) p-TsCl, Pyridine, 0 C; (c)
concd HCl, MeOH, reflux (27a and 27e–i: 22–48% in three steps). Method D: (d) NaN3, TFA, reflux (27b–d: 44–60%). b. Synthesis of substituted 2-
cyclopenten-1-ones (24e–g). Reagents: (a) LDA. RX (RX=Mel or n-PrI or CH2=CHCH2Br), THF-DMPU, ꢀ78ꢁC (28–55%).

692
Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
Scheme 3. Synthesis of 5,5-dimethyl-5,6-dihydropyridin-2-one (30). Reagents: (a) NH2OSO3H, HCO2H, reflux (52%); (b) DDQ, BSTFA, dioxane,
reflux (70%).
Scheme 4. Synthesis of 5,6-dihydropyridin-2(1H)-imines (1 and 5–13). Reagents: (a) Et3O+BF4ꢀ, CH2Cl2; (b) NH3, EtOH; (c) 4 M HCl dioxane (8–
70% in three steps).
Synthesis of 1,5,6,7-tetrahydro-2H-azepin-2-imines 14–
19 is outlined in Scheme 5a. Cyclohexenone 33a–f were
converted to their corresponding oximes 34a–f, and
Beckmann rearrangement of 34a–f afforded 35a–f at a
39–48% yield, respectively. Formation of imidates using
35a–f resulted in 36a–f, aminolysis of which gave 14–19
(10–26% yield). Among the cyclohexenones used in the
above-described synthesis, 33b–c21 and 33f were pre-
pared by alkylation of 33a or 33e (23–82% yield), as
described in Scheme 5b. Compounds 33a, 33d and 33e
are commercially available.
bitory activity and still no isoform selectivity. Shifting the
double bond of compound 1 from position 3,4 to 4,5
resulted in compound 3, with reduced inhibitory activity
for both isoforms as well as reduced selectivity. Intro-
duction of another methyl group at position 5 of com-
pound 3 afforded compound 4, which showed a nearly
4-fold reduction of inhibitory activity against hiNOS
and a 10-fold reduction of activity against heNOS as
compared with 3. However, the isoform selectivity
(heNOS/hiNOS) of compound 4 was nearly 3 times
greater than that of compound 3. Based on this infor-
mation, our chemical modification was carried out by
fixing the double bond at position 3,4, as illustrated in
compounds 5–13. Introduction of another methyl group
at position 3 of compound 1 gave compound 5, with
little change of inhibitory activity or isoform selectivity.
Introduction of a gem-dimethyl group at position 5 of
compound 1 provided compound 6, which showed a
2-fold reduction of inhibitory activity for hiNOS and a
10-fold reduction of heNOS inhibition. As a result, its
isoform selectivity was 5 times greater than that of 1.
Removal of the 4-methyl group from 6 afforded com-
pound 7, showing a 14-fold reduction of hiNOS inhibi-
tion and a 135-fold reduction of heNOS inhibition. As a
Results and Discussion
As shown in Tables 1 and 2, compounds 1–19 were
evaluated for their ability to inhibit two isoforms of
human NOS, that is human iNOS (hiNOS) and human
eNOS (heNOS). As expected, 4-methyl-5,6-dihydropyr-
idin-2(1H)-imine 1 demonstrated potent inhibition of
both hiNOS and heNOS without any selectivity.
Replacement of the 4-methyl group with a 4-ethyl group
gave compound 2, with a nearly 6-fold reduction of inhi-

Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
693
ꢁ
.
Scheme 5. a. Synthesis of 1,5,6,7-tetrahydro-2H-azepin-2-imines (14–19). Reagents: (a) NH2OH HCl, NaOAc, MeOH, reflux; (b) PPA 120 C (39–
48% in two steps); (c) Et3O+BF4ꢀ, CH2Cl2; (d) NH3, EtOH; (e) 4 M HCl dioxane (10–26% in three steps). b. Synthesis of substituted 2-cyclohexen-
1-ones (33b, 33c and 33f). Reagents: (a) LDA, RX (RX=Mel or n-PrI), THF-DMPU, ꢀ78ꢁC (23–82%).
result, its isoform selectivity (heNOS/hiNOS) was
nearly 10 times greater than that of 1. Introduction of
another methyl group at position-5 of 1 afforded com-
pound 8, which showed a nearly 2-fold reduction of
both hiNOS and heNOS inhibition, but retained the
same isoform selectivity. Introduction of another
methyl group at position 6 of 1 gave compound 9, which
retained inhibitory activity for both isoforms as well as
selectivity. Replacement of the 6-methyl group of 9 with
a 6-propyl group produced compound 10, with a 3-fold
increase of inhibitory activity against hiNOS and a 2.4-
fold reduction of activity against heNOS, resulting in
7.5-fold higher isoform selectivity than that of 9. Nearly
the same result was obtained with the corresponding
allyl derivative, compound 11.
both isoforms indicated that it was much less potent
than 12.
The biological activity profile of 1,5,6,7-tetrahydro-2H-
azepin-2-imines 14–19 is described in Table 2. Ring
expansion of 1 provided a seven-membered amidine
derivative 14, which showed reduced inhibitory activity.
Introduction of an alkyl group at position 7 of the
seven-membered compound 14 was effective for
increasing the inhibition of both isoforms. Introduction
of a methyl group at position 7 of 14 afforded com-
pound 15, which showed a slight increase, 10-fold
increase, and 7-fold reduction of hiNOS inhibition,
heNOS inhibition, and isoform selectivity, respectively.
Introduction of a n-propyl group at position 7 of 14
resulted in compound 16, which showed a 20-fold
increase and a 4.6-fold increase of hiNOS inhibition and
heNOS inhibition, respectively. It was also interesting
that a 5-fold increase of isoform selectivity was obtained
by this chemical modification. Analogous results were
observed in the biological evaluation of compounds 9–
10. Introduction of a methyl group at position-6 of 14
provided compound 17, with slight increases of hiNOS
Two cis-fused bicyclic analogues were prepared and
evaluated as iNOS inhibitors. The cis-fused 5,6-bicyclic
analogue 12 demonstrated nearly the same hiNOS inhi-
bition and 2.2-fold less heNOS inhibition when com-
pared with 1, while the isoform selectivity of 12 was
better than that of 1. This was also true for the cis-fused
6,6-bicyclic analogue 13, although its IC50 values for

694
Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
and heNOS inhibition, but less isoform selectivity.
Removal of the 4-methyl group of 14, followed by
introduction of a 5,5-dimethyl moiety, gave 18, which
retained hiNOS inhibition and showed nearly 5-fold less
heNOS inhibition, resulting in nearly 4-fold more
isoform selectivity. Introduction of an n-propyl group at
position 7 of 18 did not increase either its iNOS inhibi-
tory activity or isoform selectivity.
MTD was determined for each compound. As shown in
Table 3, the MTD was 20mg/kg for both compound 1 and
compound 6, when a single intravenous (iv) dose was given
to normal mice. The MTD values of five compounds 1, 6,
9, 10 and 11 were lower than that of l-NMMA (3000 mg/
kg), although the MTD/ID50 ratio for NOx accumula-
tion in mice was 130 for 1, 71 for 6, 330 for 9, 14 for 10
and 31 for 11, respectively. The MTD of 16 was not
evaluated, because it was unexpectedly less potent in vivo.
Compound 14 also showed unexpectedly less potent ID50
value in NOx accumulation assay for its potent IC50 value
in mouse. 1,5,6,7-Tetrahydro-2H-azepin-2-imines exhib-
ited relatively less potency than dihydropyridin-2(1H)-
imines in both the in vitro and in vivo assays.
As a result, introduction of a n-propyl group at the car-
bon adjacent to the nitrogen was found to be effective for
improving the iNOS inhibitory activity and isoform
selectivity of these two series of inhibitors, as illustrated
in compounds 10 and 16, although this was not always
true for 5,5-dimethyl-1,5,6,7-tetrahydro-2H-azepin-2-
imine derivatives, as illustrated in compound 19.
As shown in Table 4, compounds 1 and 6 had a promising
pharmacokinetic profile in rats and these compounds
(2 mg/kg, iv; 10mg/kg, po) achieved good plasma levels.
Intravenous administration of compounds 1 and 6 yielded
the following values for mean pharmacokinetic half-life
(t1/2) (1: 1.08ꢂ0.40h and 6: 1.93ꢂ0.21h, respectively),
plasma clearance (CL) (1: 3.084ꢂ0.573 L/h/kg and 6:
2.127ꢂ0.447 L/h/kg, respectively) and steady-state
volume of distribution (Vss) (1: 3.305ꢂ1.258 L/kg and 6:
4.589ꢂ0.116 L/kg, respectively). After oral administra-
tion, the mean maximum plasma concentration of
unchanged drug in plasma (Cmax), time of maximum
concentration (tmax), pharmacokinetic half-life (t1/2) and
oral bioavailability (BA) were 1.082ꢂ0.156 mg/mL,
0.67ꢂ0.29h, 2.18ꢂ1.18h and 88% for compound 1 ver-
sus 1.312ꢂ0.225 mg/mL, 1.000ꢂ0.000 h, 2.41ꢂ0.26h and
84% for compound 6, respectively.
Among the compounds described in Tables 1 and 2,
compound 1 is structurally representative of this series
and shows potent iNOS inhibitory activity. It was
selected together with compounds 3, 5–6, 8–12 and 14–
17 for further biological and pharmacodynamic eval-
uation in animal models, on the basis of their in vitro
potency and isoform selectivity, as well as the acute
toxicity in mice (maximum tolerated dose) and the
structural features. Compounds 10, 11 and 16 exhibited
greater potency and more isoform selectivity than the
above-mentioned compounds 1 and 6. As shown in
Table 3, compounds 10 and 11 demonstrated a higher
acute toxicity than 1 and 6, as evaluated by the ratio
between the maximum tolerated dose (MTD) and the
ID50 value for NOx22 accumulation (MTD/NOx).
In order to assess compounds 1, 3, 5–6, 8–12 and 14–17
for their ability to inhibit iNOS in vivo, mice were given
a single compound subcutaneously (sc) at 3 h after
lipopolysaccharide (LPS) injection. Then the plasma
NOx accumulation from 3 to 6 h after LPS injection was
determined. As shown in Table 3, all of the tested com-
pounds exhibited more potency in NOx accumulation
assay in mice. Of these, compounds 1, 6, 9, 10, 11 and
16 inhibited NOx accumulation in plasma and their
ID50 values were 0.16, 0.28, 0.09, 0.36, 0.16 and
0.80 mg/kg, sc, respectively. To assess acute toxicity, the
Figure 1 demonstrates a stereo view of the docking
study of 10 with hiNOS (PDB code 2NSI). Clear-cut
interaction between the basic amidine moiety with the
acidic Glu377 residue and insertion of the 4-methyl
group of 10 into one of the small pockets of the enzyme
were observed. Additionally, 6-n-propyl moiety, which
played a role in an increased hiNOS inhibition and iso-
form selectivity as illustrated in 10 and 16, was found to
Table 4. Rat pharmacokinetic data for compounds 1 and 6a (N b=3)
Route (Dose)
Parameters
1
6
MeanꢂSD MeanꢂSD
Table 3. Pharmacological evaluation of 1, 3, 5–6, 8–12 and 14–17
iv (2 mg/kg)
t1/2c (1 or 2 ! 4) (h)
1.08ꢂ0.40
1.93ꢂ0.21
Compd
mouse iNOS
Nox
MTD
MTD/NOx
AUCd (0 ! 1) (mg h/mL) 0.665ꢂ0.137 0.969ꢂ0.207
CLe (L/h/kg)
Vssf (L/kg)
3.084ꢂ0.573 2.127ꢂ0.447
3.305ꢂ1.258 4.589ꢂ0.116
1.082ꢂ0.156 1.312ꢂ0.225
0.67ꢂ0.29 1.000ꢂ0.000
IC50 (mM)
ID50 (mg/kg, sc) mg/kg, iv
po (10 mg/kg)
Cmaxg (mg/mL)
tmaxh (h)
1
3
5
6
8
9
10
11
12
14
15
0.022
0.066
0.12
0.10
0.067
0.14
0.10
0.025
0.038
0.10
0.17
0.20
0.22
3.5
0.16
0.38
0.30
0.28
0.30
0.09
0.36
0.16
20
50
30
20
30
30
5
5
10
30
130
130
100
71
100
330
14
31
56
<10
12
NTa
36
t1/2c (1 or 2 ! 8) (h)
2.18ꢂ1.18
2.41ꢂ0.26
AUCd (0 ! 1) (mg h/mL) 2.927ꢂ0.760 4.067ꢂ0.218
BAi
88.0ꢂ22.9
83.9ꢂ4.5
aThe mean values are reported. Standard deviations are reported.
bN is the number of animals used in the study.
cPharmacokinetic half-life.
dIntegrated area under plasma concentration versus time curve from
time 0 to time infinity.
0.18
3 mg/kg (47.9%)
0.83
0.80
0.84
26
10
ePlasma clearance.
fSteady-state volume of distribution.
16
17
NTa
30
gMaximum plasma concentration of unchanged drug in plasma.
hTime of maximum concentration.
l-NMMA
3000
120
aNot tested.
iOral bioavailability.

Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
695
occupy the space close to the entrance of the hole in
which the inhibitor is inserted.
the assigned structures. All 1H NMRspectra were
obtained on a Varian Gemini-200, VXR-200s spectro-
meter. Fast atom bombardment mass spectra (FAB-
MS) and electron ionization (EI) were obtained on a
JEOL JMS-DX303HF or PerSeptive Voyager Elite
spectrometer. Atmospheric pressure chemical ionization
(APCI) was determined on a Hitachi M1200H spectro-
meter. Matrix assisted laser desorption ionization-time
of flight high-resolution mass spectra (MALDI-TOF
HRMS) were obtained on a PerSeptive Voyager Elite
spectrometer. IRspectra were measured on a Perkin–
Elmer FT-IR1760X or Jasco FT/IR-430 spectrometer.
Melting points (mp) were determined by Yanaco micro
melting point apparatus MP-500D and are uncorrected.
Column chromatography was carried out on silica gel
[Merck silica gel 60 (0.063–0.200 mm) or Wako Gel
C200 or Fuji Silysia FL60D]. Thin layer chromato-
graphy was performed on silica gel (Merck TLC plate,
silica gel 60 F254).
Summary
In summary, we explored the SARfor a series of sub-
stituted 5,6-dihydropyridin-2(1H)-imines and substituted
1,5,6,7-tetrahydro-2H-azepin-2-imines and obtained a
significant increase of both isoform selectivity and iNOS
inhibition. 4-Methyl and 4-methyl-5,5-dimethyl sub-
stitution gave compound, 1 and 6, which were potent and/
or selective inhibitors of iNOS. The oral bioavailability of
1 and 6 in rats was excellent (1:88% and 6:84%).
Compounds 10 and 11 exhibited greater inhibitory
activity and isoform selectivity than 1 and 6, but their
acute toxicity in mice was also relatively higher. Com-
pound 16 exhibited potent hiNOS inhibitory activity and
excellent isoform selectivity relative to other inhibitors,
but it showed unexpectedly week miNOS inhibition.
Method A (Birch reduction)
The computer-aided docking study of 10 with hiNOS
demonstrates that the 4-methyl moiety is inserted into
one of the small pockets and the 6-n-propyl moiety
occupies the space at the entrance of the hole in which
the inhibitor is inserted (Fig. 1).
4-Methyl-5,6-dihydropyridin-2(1H)-imine hydrochloride
salt (1). To a stirred solution of 2-amino-4-methylpyr-
idine (20) (2 g, 18.5 mmol) in THF (6.2 mL) and EtOH
(2.2 mL) was added liquid NH3 (60 mL) at ꢀ78 ꢁC.
Small pieces of lithium metal (320 mg, 46.2 mmol) were
added to the reaction mixture and the solution stirred
for an additional 1.5 h at ꢀ78 ꢁC before quenched with
solid ammonium chloride (75 g). The reaction mixture
was slowly warmed up to room temperature removing
the ammonia with a stream of nitrogen and then treated
with aqueous ammonium chloride. The reaction mix-
ture was extracted with CHCl3. The combined organic
layers were dried over anhydrous sodium sulfate and
evaporated under reduced pressure. The residue was
converted to its hydrochloride with 4 M HCl/EtOAc–
EtOH at 0 ꢁC and further purified by column chroma-
tography on silica gel (CHCl3/MeOH, 5/1) to afford 1
as a pale yellow powder (1.8 g, 66%). TLC Rf 0.38
Based on the currently available data, pharmacological
evaluation of some of these compounds suggested that
they might become clinically useful drugs with a good
safety margin.
Experimental
General directions
Analytical samples were homogeneous as confirmed by
TLC, and afforded spectroscopic results consistent with
Figure 1. Stereo view of a docking model of compound 10 with human iNOS (PDB code 2NSI).

696
Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
(CHCl3/MeOH/AcOH, 10/1/1); IR(KBr) 3327, 3138,
1692, 1639, 1614, 1528, 1443, 1383, 1370, 1344, 1178,
from 20. White powder; TLC Rf 0.42 (CHCl3/MeOH/
AcOH, 10/1/1); IR(KBr) 3137, 3004, 2941, 2880, 1721,
1052, 987, 849, 785, 677 cmꢀ1
;
1H NMR(200 MHz,
1687, 1621, 1546, 1440, 1417, 1389, 1363, 1326, 1141,
1
CDCl3) d 9.30–9.15 (brs, 1H), 9.15–8.95 (brs, 1H), 8.65–
8.50 (brs, 1H), 6.24 (s, 1H), 3.50 (t, J=7.5 Hz, 2H), 2.43
(t, J=7.6 Hz, 2H), 2.05 (s, 3H); mp: 157–158 ꢁC; MS
and HR-MS analysis showed only the ion from
C6H10N2 corresponding to the loss of HCl, MS (APCI,
Pos.) m/z 111 (M+H)+; HR-MS (MALDI-TOF, Pos.)
calcd for C6H10N2+ H+: 111.0922; found: 111.0949.
839, 585 cmꢀ1; H NMR(200 MHz, CDCl ) d 10.20–
3
10.00 (brs 1H), 8.90–8.70 (brs, 1H), 8.60–8.50 (brs, 1H),
3.90–3.75 (m, 2H), 3.30–3.05 (m, 2H), 1.72 (s, 3H),
1.70(s, 3H); mp: 68–69 ꢁC; MS and HR-MS analysis
showed only the ion from C7H12N2 corresponding to
the loss of HCl, MS (APCI, Pos.) m/z 125 (M+H)+;
HR-MS (MALDI-TOF, Pos.) calcd for C7H12N2+
H+: 125.1079; found: 125.1061.
4-Ethyl-5,6-dihydropyridin-2(1H)-imine
hydrochloride
salt (2). Compound 2 was prepared from 4-ethyl-2-ami-
nopyridine (22)15 in 79% yield according to the same
procedure as described for the preparation of 1 from 20.
White powder; TLC Rf 0.45 (CHCl3/MeOH/AcOH, 10/
1/1); IR(KBr) 3100, 1688, 1649, 1607, 1425, 1380, 1338,
1249, 1212, 1164, 1054, 992, 976, 867, 765, 707,
680 cmꢀ1; 1H NMR(200 MHz, CDCl 3) d 9.48–9.18 (brs,
1H), 8.98–8.75 (brs, 1H), 8.74–8.52 (brs, 1H), 6.30 (s,
1H), 3.49 (t, J=7.2 Hz, 2H), 2.42 (t, J=7.2 Hz, 2H),
2.32 (q, J=7.4 Hz, 2H), 1.14 (t, J=7.4 Hz, 3H); mp:
113–114 ꢁC; MS and HR-MS analysis showed only the
ion from C7H12N2 corresponding to the loss of HCl,
MS (APCI, Pos.) m/z 125 (M+H)+; HR-MS (MALDI-
TOF, Pos.) calcd for C7H12N2+ H+: 125.1079; found:
125.1080.
Method C (Beckmann rearrangement)
4-Methyl-5,6-dihydropyridin-2(1H)-one (27a). To a stir-
red solution of 3-methyl-2-cyclopenten-1-one (24a)
(10.3 g, 107 mmol) in MeOH (200 mL) were added
.
NaOAc (21.2 g, 258 mmol), NH2OH HCl (8.7 g,
125 mmol) and water (40 mL). The reaction mixture was
heated at reflux temperature for 2 h. After completing
the reaction, the reaction mixture was cooled to room
temperature and evaporated under reduced pressure.
The residue was treated with water and extracted with
CHCl3. The combined organic layers were washed with
saturated aqueous sodium bicarbonate, brine, dried
over anhydrous magnesium sulfate and evaporated to
afford 25a (8.43 g), which was used for the subsequent
reaction without further purification.
Method B (Birch reduction)
To a stirred solution of the compound obtained above
in pyridine (57 mL) was added p-TsCl (17.3 g,
90.7 mmol) at 0 ꢁC under an argon atmosphere. After
stirring for 1 h at room temperature, the reaction mix-
ture was treated with saturated aqueous sodium bicar-
bonate and evaporated under reduced pressure. The
residue was treated with water and extracted with
EtOAc. The combined organic layers were washed with
2 M HCl, water and brine, dried over anhydrous mag-
nesium sulfate, and evaporated to afford 26a (14.5 g),
which was used for the subsequent reaction without
further purification.
4-Methyl-3,6-dihydropyridin-2(1H)-imine hydrochloride
salt (3). To a stirred solution of 2-amino-4-methylpyr-
idine (20) (1.0 g, 9.2 mmol) in THF (3.1 mL) and EtOH
(1.1 mL) was added liquid NH3 (30 mL) at ꢀ78 ꢁC.
After the addition of small pieces of lithium metal
(210 mg, 30.3 mmol), the reaction mixture was stirred
for an additional 1.5 h at ꢀ78 ꢁC before quenched with a
small amount of water (4 mL). The reaction mixture was
slowly warmed up to room temperature removing the
ammonia with a stream of nitrogen and then treated
with additional water (20 mL). The reaction mixture
was extracted with CHCl3. The combined organic layers
were dried over anhydrous sodium sulfate and evapo-
rated under reduced pressure. The residue was treated
with 4 M HCl/EtOAc–EtOH at 0 ꢁC and further pur-
ified by column chromatography on silica gel (CHCl3/
MeOH, 5/1) to afford 3 as a pale yellow powder
(860 mg, 64%). TLC Rf 0.38 (CHCl3/MeOH/AcOH, 10/
1/1); IR(KBr) 3292, 3137, 2979, 2944, 2878, 1718, 1682,
1544, 1417, 1356, 1305, 1154, 1016, 946, 866, 812,
The residue was dissolved in MeOH (412 mL) and trea-
ted with conc HCl (28 mL). The reaction mixture was
stirred at room temperature for 12 h, then at reflux
temperature for 12 h, and was concentrated under
reduced pressure. The residue was purified by column
chromatography on silica gel (CHCl3/MeOH, 20/1) to
afford 27a as a white powder (1.3 g, 11% in three steps).
TLC Rf 0.38 (CHCl3/MeOH, 10/1); IR(KBr) 3418,
600 cmꢀ1
;
1H NMR(200 MHz, CDCl 3) d 10.15–9.95
1671, 1615, 1484, 1455, 1380, 1339, 1157, 1109, 996,
1
(brs, 1H), 9.05–8.95 (brs, 1H), 8.90–8.65 (brs, 1H), 5.55
(brs, 1H), 4.05–3.90 (m, 2H), 3.22 (t, J=5.1 Hz, 2H),
1.80 (s, 3H); mp: 158–159 ꢁC; MS and HR-MS analysis
showed only the ion from C6H10N2 corresponding to
the loss of HCl, MS (APCI, Pos.) m/z 111 (M+H)+;
HR-MS (MALDI-TOF, Pos.) calcd for C6H10N2+
H+: 111.0922; found: 111.0896.
857, 672, 506, 470 cmꢀ1; H NMR(200 MHz, CDCl ) d
3
5.90 (brs, 1H), 5.73–5.69 (m, 1H), 3.42 (dt, J=7.0,
2.6 Hz, 2H), 2.30 (t, J=7.1 Hz, 2H), 1.94 (d, J=1.4 Hz,
3H); mp: 58–59 ꢁC; MS (APCI, Pos.) m/z 112 (M+H)+,
92, 83; HR-MS (MALDI-TOF, Pos.) calcd for
C6H9N1O1+ H+: 112.0762; found: 112.0775.
Method D (Beckmann rearrangement)
4,5-Dimethyl-3,6-dihydropyridin-2(1H)-imine hydrochlo-
ride salt (4). Compound 4 was prepared from 4,5-dime-
thyl-2-aminopyridine (23)18 in 75% yield according to
the same procedure as described for the preparation of 3
3,4-Dimethyl-5,6-dihydropyridin-2(1H)-one (27b). To a
stirred solution of 2,3-dimethyl-2-cyclopenten-1-one
(24b) (1.0 g, 9.1 mmol) in TFA (27 mL) was added NaN3

Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
697
(0.89 g, 13.7 mmol) and the reaction mixture was heated
at reflux temperature for 15 h. After completing the
reaction, the reaction mixture was cooled to room tem-
perature and evaporated under reduced pressure. The
resulting mixture was treated with water and extracted
with CHCl3. The combined organic layers were washed
with saturated aqueous sodium bicarbonate, water,
brine, dried over anhydrous magnesium sulfate and
evaporated under reduced pressure. The residue was
purified by column chromatography on silica gel
(EtOAc) to afford 27b as a white powder (682 mg,
60%). TLC Rf 0.40 (CHCl3/MeOH, 10/1); IR(KBr)
3202, 3066, 2929, 1667, 1632, 1479, 1415, 1375, 1344,
1248, 1228, 1158, 1091, 1003, 903, 838, 796, 669, 539,
505, 456 cmꢀ1; 1H NMR(200 MHz, CDCl 3) d 5.76 (brs,
1H), 3.34 (dt, J=7.0, 2.8 Hz, 2H), 2.31 (t, J=7.0 Hz,
2H), 1.89 (q, J=1.0 Hz, 3H), 1.87–1.83 (m, 3H); mp:
75–76 ꢁC; MS (APCI, Pos.) m/z 126 (M+H)+; HR-MS
(MALDI-TOF, Pos.) calcd for C7H11N1O1+ H+:
126.0919; found: 126.0931.
LDA (2 M in heptane/THF/ethylbenzene, 5.5 mL,
11 mmol) at ꢀ78 ꢁC under an argon atmosphere. The
reaction mixture was warmed up to room temperature
over 2 h and stirred for an additional 30 min at this
temperature. After completing the reaction, the reaction
mixture was treated with saturated aqueous ammonium
chloride and extracted with EtOAc. The combined
organic layers were washed with brine, dried over
anhydrous magnesium sulfate and evaporated under
reduced pressure. The residue was purified by column
chromatography on silica gel (n-hexane/EtOAc, 10/1) to
afford 24f as a pale yellow oil (483 mg, 35%). TLC Rf
0.42 (n-hexane/EtOAc, 5/1); IR(neat) 3481, 2958, 2930,
2871, 1736, 1697, 1625, 1432, 1379, 1323, 1281, 1178,
1
925, 838, 532, 438 cmꢀ1; H NMR(200 MHz, CDCl ) d
3
5.91 (s, 1H), 2.74 (dd, J=18.4, 6.6 Hz, 1H), 2.64–2.32
(m, 1H), 2.25 (d, J=18.4 Hz, 1H), 2.12 (s, 3H), 1.90–
1.66 (m, 1H), 1.50–1.10 (m, 3H), 0.93 (t, J=6.6 Hz, 3H);
MS (APCI, Pos.) m/z 139 (M+H)+; HR-MS (MALDI-
TOF, Pos.) calcd for C9H14O1+ H+: 139.1123; found:
139.1106.
4,5,5 - Trimethyl - 5,6 - dihydropyridin - 2(1H) - one (27c).
Compound 27c was prepared from 3,4,4-trimethyl-2-
cyclopenten-1-one (24c)19c in 47% yield according to the
same procedure as described for the preparation of 27b
from 24b. White powder; TLC Rf 0.36 (CHCl3/MeOH,
4-Methyl-6-propyl-5,6-dihydropyridin-2(1H)-one (27f).
Compound 27f was prepared from 24f in 24% yield
according to the same procedure as described for the
preparation of 27a from 24a. Brown oil; TLC Rf 0.38
(CHCl3/MeOH, 20/1); IR(neat) 3235, 2960, 2874, 2354,
1
10/1); H NMR(200 MHz, CDCl 3) d 5.88 (brs, 1H),
5.63–5.60 (m, 1H), 3.14 (d, J=2.8 Hz, 2H), 1.86 (d,
J=1.4 Hz, 3H), 1.10 (s, 6H); MS (APCI, Pos.) m/z 140
(M+H)+, 124, 111, 98, 83; HR-MS (MALDI-TOF,
Pos.) calcd for C8H13N1O1+ H+: 140.1075; found:
140.1049.
1661, 1602, 1464, 1382, 1344, 1160, 1004, 882, 492 cmꢀ1
;
1H NMR(200 MHz, CDCl ) d 5.69 (s, 1H), 5.50 (br,
3
1H), 3.70–3.45 (m, 1H), 2.35–2.00 (m, 2H), 1.92 (s, 3H),
1.63–1.20 (m, 4H), 0.95 (t, J=7.0 Hz, 3H); MS (APCI,
Pos.) m/z 154 (M+H)+; HR-MS (MALDI-TOF, Pos.)
calcd for C9H15N1O1+ H+: 154.1232; found: 1541266.
4,5-Dimethyl-5,6-dihydropyridin-2(1H)-one (27d). Com-
pound 27d was prepared from 3,4-dimethyl-2-cyclo-
penten-1-one (24d)19d in 44% yield according to the
same procedure as described for the preparation of 27b
from 24b. White powder; TLC Rf 0.36 (CHCl3/MeOH,
6-Allyl-4-methyl-5,6-dihydropyridin-2(1H)-one (27g).
Compound 27g was prepared from 5-allyl-3-methyl-2-
cyclopenten-1-one (24g)19b in 48% yield according to
the same procedure as described for the preparation of
27a from 24a. Brown oil; TLC Rf 0.50 (CHCl3/MeOH,
10/1); IR(neat) 3353, 2970, 2930, 2357, 1697, 1621,
1
10/1); H NMR(200 MHz, CDCl 3) d 6.17 (brs, 1H),
5.66 (s, 1H), 3.54 (ddd, J=12.2, 5.3, 1.8 Hz, 1H), 3.1
(ddd, J=12.2, 4.9, 3.7 Hz, 1H), 2.41–2.23 (m, 1H), 1.92
(d, J=1.4 Hz, 3H), 1.15 (d, J=7.0 Hz, 3H); MS (APCI,
Pos.) m/z 126 (M+H)+; HR-MS (MALDI-TOF, Pos.)
calcd for C7H11N1O1+ H+: 126.0919; found: 126.0929.
1444, 1381, 1328, 1098, 1054, 932, 887, 825, 775, 720,
1
455 cmꢀ1; H NMR(200 MHz, CDCl ) d 5.90–5.55 (m,
3
3H), 5.21 (br, 1H), 5.15–5.10 (m, 1H), 3.70–3.55 (m,
1H), 3.00–1.70 (m, 4H), 1.92 (s, 3H); MS (APCI, Pos.)
m/z 152 (M+H)+; HR-MS (MALDI-TOF, Pos.) calcd
for C9H13N1O1+ H+: 152.1075; found: 152.1062.
4,6-Dimethyl-5,6-dihydropyridin-2(1H)-one (27e). Com-
pound 27e was prepared from 3,5-dimethyl-2-cyclo-
penten-1-one (24e)19a in 34% yield according to the
same procedure as described for the preparation of 27a
from 24a. White powder; TLC Rf 0.65 (CHCl3/MeOH,
10/1); IR(KBr) 3239, 2974, 2928, 1672, 1623, 1440,
1340, 1188, 1156, 1074, 879, 847, 769, 627, 532, 491,
(dl)-(4aR,7aR)-4-Methyl-1,4a,5,6,7,7a-hexahydro-2H-
cyclopenta[b]pyridin-2-one (27h). Compound 27h was
prepared from (dl)-(3aS,6aR)-3-methyl-4,5,6,6a-tetra-
hydropentalen-1(3aH)-one (24h)19e in 33% yield
according to the same procedure as described for the
preparation of 27a from 24a. White powder; TLC Rf
0.48 (CHCl3/MeOH, 10/1); 1H NMR(200 MHz,
CDCl3) d 5.79 (brs, 1H), 5.62–5.57 (m, 1H), 4.08–3.98
(m, 1H), 2.43 (q, J=7.8 Hz, 1H), 2.20–1.52 (m, 6H),
1.92 (s, 3H); MS (APCI, Pos.) m/z 152 (M+H)+; HR -
MS (MALDI-TOF, Pos.) calcd for C9H13N1O1+ H+:
152.1075; found: 152.1073.
1
436 cmꢀ1; H NMR(200 MHz, CDCl3) d 5.70 (s, 1H),
5.68–5.35 (br, 1H), 3.83–3.60 (m, 1H), 2.33–2.00 (m,
2H), 1.91 (s, 3H), 1.08 (s, 3H), 1.24 (d, J=6.6 Hz, 3H);
mp: 41–43 ꢁC; MS (APCI, Pos.) m/z 126 (M+H)+; HR -
MS (MALDI-TOF, Pos.) calcd for C7H11N1O1+ H+:
126.0919; found: 126.0913.
3-Methyl-5-propyl-2-cyclopenten-1-one (24f). To a stir-
red solution of 3-methyl-2-cyclopenten-1-one (24a)
(962 mg, 10 mmol) in THF (15 mL), DMPU (2 mL) and
n-propyl iodide (2.0 mL, 20 mmol) was slowly added
(dl)-(4aR,8aR)-4-Methyl-4a,5,6,7,8,8a-hexahydroquino-
lin-2(1H)-one (27i). Compound 27i was prepared from
(dl)-(3aS,7aR)-3-methyl-3a,4,5,6,7,7a-hexahydro-1H-

698
Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
inden-1-one (24i)19f in 22% yield according to the same
procedure as described for the preparation of 27a from
24a. White powder; TLC Rf 0.48 (CHCl3/MeOH, 10/1);
A solution of 34a obtained above in PPA (5 mL) was
heated at 120 ꢁC for 4 h under stirring. The reaction
mixture was cooled up to room temperature and diluted
with water and extracted with CHCl3. The combined
organic layers were washed with brine, dried over
anhydrous magnesium sulfate and evaporated under
reduced pressure. The residue was purified by column
chromatography on silica gel (CHCl3/MeOH, 20/1) to
afford 35a as a brown powder (158 mg, 42% in two
steps). TLC Rf 0.16 (CHCl3/MeOH, 10/1); IR(KBr)
3236, 2930, 1661, 1614, 1479, 1339, 1278, 1142, 1056,
1H NMR(200 MHz, CDCl ) d 5.69–5.63 (m, 1H), 5.23
3
(brs, 1H), 3.77–3.72 (m, 1H), 2.07–1.95 (m, 1H), 1.92 (d,
J=1.6 Hz, 3H), 1.80–1.11 (m, 8H); MS (APCI, Pos.) m/
z 166 (M+H)+; HR-MS (MALDI-TOF, Pos.) calcd for
C10H15N1O1+ H+: 166.1232; found: 166.1223.
5,5-Dimethylpiperidin-2-one (29).20a To a stirred solu-
tion of 3,3-dimethylcyclopentanone (28)20b (8.5 g,
75.8 mmol) in HCO2H (76 mL) was added NH2OSO3H
(12.8 g, 113.2 mmol) and the reaction mixture was
heated at reflux temperature for 12 h. After cooling at
room temperature, the reaction mixture was diluted
with water, then treated with 5 M aqueous sodium
hydroxide and extracted with CHCl3. The organic layer
was washed with brine, dried over anhydrous magne-
sium sulfate and evaporated under reduced pressure.
The residue was purified by recrystallization (EtOAc) to
1
1016, 889, 841, 770, 651, 502, 434, 417 cmꢀ1; H NMR
(200 MHz, CDCl3) d 6.60–6.40 (brs, 1H), 5.75 (s, 1H),
3.28–3.10 (m, 2H), 2.65–2.42 (m, 1H), 2.18–1.90 (m,
2H), 1.78–1.60 (m, 1H), 1.14 (d, J=7.0 Hꢁz, 3H); MS
(APCI, Pos.) m/z 126 (M+H)+; mp: 74–75 C; HR-MS
(MALDI-TOF, Pos.) calcd for C7H11N1O1+ H+:
126.0919; found: 126.0923.
4,7-Dimethyl-1,5,6,7-tetrahydro-2H-azepin-2-one (35b).
Compound 35b was prepared from 3,5-dimethy-2-
cyclohexen-1-one (33b)21 in 39% yield according to the
same procedure as described for the preparation of 35a
from 33a. Pale yellow powder; TLC Rf 0.49 (CHCl3/
MeOH, 15/1); IR(KBr) 3419, 2971, 2939, 1656, 1609,
afford 29 as a white powder (5.0 g, 52%). TLC Rf 0.54
1
(CHCl3/MeOH, 10/1); H NMR(200 MHz, CDCl ) d
3
5.88 (brs, 1H), 3.02 (d, J=2.5 Hz, 2H), 2.38 (t,
J=7.0 Hz, 2H), 1.61 (t, J=7.0 Hz, 2H), 1.05 (s, 6H);
MS (APCI, Pos.) m/z 128 (M+H)+.
1458, 1320, 1006, 866, 806, 646, 578, 528, 442, 416 cmꢀ1
;
5,5-Dimethyl-5,6-dihydropyridin-2(1H)-one (30). To a
stirred solution of 29 (2.5 g, 19.7 mmol) in 1,4-dioxane
(65 mL) were added DDQ (4.7 g, 20.6 mmol) and
BSTFA (Bis(trimethylsilyl)trifluoroacetamide) (21 mL,
79.1 mmol) and the reaction mixture was heated at
reflux temperature for 16 h. After cooling up to room
temperature, the reaction mixture was diluted with
CH2Cl2 (100 mL) and treated with in aqueous sodium
bicarbonate carefully. After stirring for 15 min, inso-
luble substances were removed by filtration. The organic
layer was separated and washed with 2 M HCl, brine,
dried over anhydrous magnesium sulfate and evapo-
rated under reduced pressure. The residue was purified
by column chromatography on silica gel (EtOAc to
CHCl3/MeOH, 10/1) to afford 30 as a reddish solid
1H NMR(200 MHz, CDCl ) d 5.98 (brs, 1H), 5.78 (s,
3
1H), 3.50 (m, 1H), 2.41 (m, 1H), 2.30 (m, 1H), 1.90 (s,
3H), 1.85 (m, 2H), 1.23 (d, J=6.6 Hz, 3H); mp: 53–
54 ꢁC; MS (APCI, Pos.) m/z 140 (M+H)+; HR-MS
(MALDI-TOF, Pos.) calcd for C8H13N1O1+ H+:
140.1075; found: 140.1103.
3-Methyl-6-propyl-2-cyclohexen-1-one (33c). Compound
33c was prepared from 3-methyl-2-cyclohexen-1-one
(33a) in 38% yield according to the same procedure as
described for the preparation of 24f from 24a. Pale yel-
low oil; TLC Rf 0.72 (n-hexane/EtOAc, 2/1); IR(neat)
3852, 3749, 3648, 3393, 2957, 2930, 2871, 2365, 1710,
1667, 1541, 1456, 1379, 1200, 889 cmꢀ1 1H NMR
;
(200 MHz, CDCl3) d 5.83 (brs, 1H), 2.29 (t, J=6.2 Hz,
2H), 2.26–2.00 (m, 1H), 1.93 (s, 3H), 1.88–1.64 (m, 2H),
1.50–1.25 (m, 4H), 0.92 (t, J=6.8 Hz, 3H); MS (APCI,
Pos.) m/z 153 (M+H)+; HR-MS (MALDI-TOF, Pos.)
calcd for C10H16O1+ H+: 153.1279; found: 153.1289.
1
(1.72 g, 70%). TLC Rf 0.36 (CHCl3/MeOH, 10/1); H
NMR(200 MHz, CDCl ) d 6.37 (d, J=10.0 Hz, 1H),
3
5.76 (dd, J=10.0, 1.8 Hz, 1H), 5.58 (br, 1H), 3.22–3.17
(m, 2H), 1.14 (s, 6H); MS (APCI, Pos.) m/z 126
(M+H)+; HR-MS (MALDI-TOF, Pos.) calcd for
C7H11N1O1+ H+: 126.0919; found: 126.0892.
4-Methyl-7-propyl-1,5,6,7-tetrahydro-2H-azepin-2-one
(35c). Compound 35c was prepared from 33c in 45%
yield according to the same procedure as described for
the preparation of 35a from 33a. Brown oil; TLC Rf
0.29 (CHCl3/MeOH, 20/1); IR(neat) 3226, 2958, 2930,
4-Methyl-1,5,6,7-tetrahydro-2H-azepin-2-one (35a). To a
stirred solution of 3-methyl-2-cyclohexen-1-one (33a)
(331 mg, 3 mmol) in MeOH (6 mL) were added
1
NaOAc (310 mg, 3.8 mmol), NH2OH HCl (260 mg,
2872, 1660, 1439, 1381, 1192, 1110, 744, 536 cmꢀ1; H
.
3.7 mmol), water (1 mL) and the reaction mixture
was heated at reflux temperature for 2 h. After
completing the reaction, the reaction mixture was
cooled to room temperature and evaporated under
reduced pressure. The reaction mixture was treated with
water and extracted with CHCl3. The combined organic
layers were washed with saturated aqueous sodium
bicarbonate, brine, dried over anhydrous magnesium
sulfate and evaporated to afford 34a (307 mg), which
was used for the subsequent reaction without further
purification.
NMR(200 MHz, CDCl ) d 5.78 (s, 1H), 5.65 (brs, 1H),
3
3.32 (m, 1H), 2.45 (m, 1H), 2.23 (m, 1H), 1.90 (s, 3H),
2.00–1.70 (m, 2H), 1.55–1.25 (m, 4H), 0.93 (t,
J=7.2 Hz, 3H); MS (APCI, Pos.) m/z 168 (M+H)+;
HR-MS (MALDI-TOF, Pos.) calcd for C10H17N1O1+
H+: 168.1388; found: 168.1343.
4,6-Dimethyl-1,5,6,7-tetrahydro-2H-azepin-2-one (35d).
Compound 35d was prepared from 3,5-dimethyl-2-
cyclohexen-1-one (33d) in 48% yield according to the
same procedure as described for the preparation of 35a

Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
699
from 33a. Pale yellow powder; TLC Rf 0.41 (CHCl3/
MeOH, 10/1); IR(KBr) 3419, 2958, 1660, 1617, 1457,
1379, 1355, 1310, 1284, 1118, 942, 905, 842, 688, 538,
421 cmꢀ1; 1H NMR(200 MHz, CDCl 3) d 6.95–6.65 (brs,
1H), 5.76 (s, 1H), 3.17 (ddd, J=14.2, 6.6, 4.0 Hz, 1H),
2.94 (ddd, J=14.2, 7.4, 5.0 Hz, 1H), 2.42 (dd, J=15.8,
5.5 Hz, 1H), 2.35–2.11 (m, 1H), 2.01 (dd, J=15.8,
7.8 Hz, 1H), 1.92 (s, 3H), 0.97 (d, J=6.7 Hz, 3H); mp:
52–54 ꢁC; MS (APCI, Pos.) m/z 140 (M+H)+; HR-MS
(MALDI-TOF, Pos.) calcd for C8H13N1O1+ H+:
140.1075; found: 140.1059.
an argon atmosphere, and the reaction mixture was
stirred at room temperature for 5 h. Concentration of
the reaction mixture under reduced pressure gave the
31a, which was used for the subsequent reaction without
further purification.
To a stirred solution of 31a in EtOH (5 mL) was added
saturated ethanolic ammonia (20 mL) under an argon
atmosphere at room temperature and stirring was con-
tinued for an additional 24 h. The reaction mixture was
diluted with CHCl3 and the resulting precipitates were
removed by filtration. The filtrate was concentrated
under reduced pressure. The reaction mixture was trea-
ted with 2 M aqueous sodium hydroxide and extracted
with CHCl3. The combined organic layers were dried
over anhydrous sodium sulfate and evaporated under
reduced pressure. The residue was converted to its
hydrochloride with 4 M HCl/EtOAc–EtOH at 0 ꢁC and
further purified by column chromatography on silica gel
(CHCl3/MeOH, 5/1) to afford 1 as a pale yellow powder
(126 mg, 8%).
5,5-Dimethyl-1,5,6,7-tetrahydro-2H-azepin-2-one (35e).
Compound 35e was prepared from 4,4-dimethyl-2-
cyclohexen-1-one (33e) in 43% yield according to the
same procedure as described for the preparation of 35a
from 33a. Pale yellow powder; TLC Rf 0.71 (CHCl3/
MeOH, 10/1); IR(KBr) 3195, 3036, 2961, 1657, 1618,
1476, 1427, 1394, 1360, 1337, 1308, 1269, 1249, 1205,
1
1156, 1111, 985, 904, 825, 695, 666, 602, 516 cmꢀ1; H
NMR(200 MHz, CDCl 3) d 6.50–6.18 (brs, 1H), 5.94 (d,
J=12.8 Hz, 1H), 5.70 (dd, J=12.8, 2.2 Hz, 1H), 3.24 (q,
J=5.2 Hz, 2H), 1.83 (t, J=4.7 Hz, 2H), 1.11 (s, 6H);
mp: 102–103 ꢁC; MS (APCI, Pos.) m/z 140 (M+H)+;
HR-MS (MALDI-TOF, Pos.) calcd for C8H13N1O1+
H+: 140.1075; found: 140.1081.
The spectral data of 1 (prepared by Method C) were
identical with that of 1 prepared by Method A.
3,4-Dimethyl-5,6-dihydropyridin-2(1H)-imine hydrochlo-
ride salt (5). Compound 5 was prepared from 27b in
72% yield according to the same procedure as described
for the preparation of 1 from 27a. White powder; TLC
Rf 0.51 (CHCl3/MeOH/AcOH, 10/2/1); IR(KBr) 3312,
1679, 1633, 1599, 1525, 1427, 1362, 1241, 1165, 1141,
4,4-Dimethyl-6-propyl-2-cyclohexen-1-one (33f). Com-
pound 33f was prepared from 4,4-dimethyl-2-cyclo-
hexen-1-one (33e) in 23% yield according to the same
procedure as described for the preparation of 24f from
24a. Pale yellow oil; TLC Rf 0.45 (n-hexane/EtOAc, 4/
1); IR(neat) 3901, 3852, 3838, 3734, 3689, 3675, 3649,
3566, 2958, 2871, 2369, 1681, 1558, 1541, 1507, 1457,
1
1107, 904, 603 cmꢀ1; H NMR(200 MHz, DMSO- d6) d
9.41 (brs, 1H), 8.90 (brs, 1H), 8.44 (brs, 1H), 3.29 (dt,
J=7.6, 2.8 Hz, 2H), 2.38 (t, J=7.6 Hz, 2H), 1.97 (s,
3H), 1.86 (s, 3H); mp: 72–74 ꢁC; MS and HR-MS ana-
lysis showed only the ion from C7H12N2 corresponding
to the loss of HCl, MS (EI, Pos.) m/z 124 (M)+, 109
(MꢀCH3)+, 94 (Mꢀ2CH3)+; HR-MS (MALDI-TOF,
Pos.) calcd for C7H12N2+ H+: 125.1079; found:
125.1067.
1
1376, 1237, 1199, 1121, 910, 813, 419 cmꢀ1; H NMR
(200 MHz, CDCl3) d 6.57 (dd, J=10.0, 2.2 Hz, 1H), 5.81
(d, J=10.0 Hz, 1H), 2.50–2.30 (m, 1H), 2.00–1.78 (m,
2H), 1.70–1.10 (m, 4H), 1.18 (s, 3H), 1.14 (s, 3H), 0.93
(t, J=6.6 Hz, 3H); MS (APCI, Pos.) m/z 167 (M+H)+;
HR-MS (MALDI-TOF, Pos.) calcd for C11H18O1+
H+: 167.1436; found: 167.1454.
4,5,5-Trimethyl-5,6-dihydropyridin-2(1H)-imine hydro-
chloride salt (6). Compound 6 was prepared from 27c
in 68% yield according to the same procedure as
described for the preparation of 1 from 27a. Pale pink
powder; TLC Rf 0.43 (CHCl3/MeOH/AcOH, 10/2/1);
IR(KBr) 3421, 3208, 2968, 2053, 1690, 1521, 1473,
1442, 1379, 1366, 1348, 1277, 1159, 1072, 1026, 1015,
5,5-Dimethyl-7-propyl-1,5,6,7-tetrahydro-2H-azepin-2-
one (35f). Compound 35f was prepared from 33f in 46%
yield according to the same procedure as described for
the preparation of 35a from 33a. Brown powder; TLC
Rf 0.50 (CHCl3/MeOH, 10/1); IR(KBr) 3175, 2964,
2929, 1663, 1616, 1435, 1389, 1258, 1202, 824, 742, 679,
1
538 cmꢀ1
;
1H NMR(200 MHz, CDCl 3) d 5.95 (d,
854, 812, 716, 642, 515, 446 cmꢀ1; H NMR(200 MHz,
J=10.4 Hz, 1H), 5.90–5.70 (brs, 1H), 5.73 (dd, J=10.4,
0.8 Hz, 1H), 3.50–3.32 (m, 1H), 1.90–1.50 (m, 6H), 1.10
(s, 3H), 1.08 (s, 3H), 0.95 (t, J=6.6 Hz, 3H); mp: 63–
64 ꢁC; MS (APCI, Pos.) m/z 182 (M+H)+; HR-MS
(MALDI-TOF, Pos.) calcd for C11H19N1O1+ H+:
182.1545; found: 182.1553.
DMSO-d6) d 9.30 (brs, 1H), 8.98 (brs, 1H), 8.60 (brs,
1H), 5.95–5.90 (m, 1H), 3.15 (d, J=3.0 Hz, 2H), 1.96 (d,
J=1.2 Hz, 3H), 1.05 (s, 6H); mp: 127–128 ꢁC; MS and
HR-MS analysis showed only the ion from C8H14N2
corresponding to the loss of HCl, MS (APCI, Pos.) m/z
139 (M+H)+, 124 (MꢀCH3+H)+; HR-MS (MALDI-
TOF, Pos.) calcd for C8H14N2+ H+: 139.1235; found:
139.1216.
General procedure for preparation of compounds 1 and
5–19
5,5-Dimethyl-5,6-dihydropyridin-2(1H)-imine hydrochlo-
ride salt (7). Compound 7 was prepared from 30 in
21% yield according to the same procedure as described
for the preparation of 1 from 27a. White powder; TLC
Rf 0.44 (CHCl3/MeOH/AcOH, 10/1/1); IR(KBr) 3408,
4-Methyl-5,6-dihydropyridin-2(1H)-imine hydrochloride
salt (1). To a stirred solution of 27a (1.2 g, 10.8 mmol)
in CH2Cl2 (10 mL) was added triethyloxonium tetra-
fluoroborate (2 M in CH2Cl2, 6.0 mL, 12 mmol) under

700
Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
3258, 3086, 2970, 2875, 1688, 1600, 1523, 1463, 1395,
1
TLC Rf 0.50 (CHCl3/MeOH/AcOH, 15/2/1); IR(neat)
3079, 1683, 1643, 1605, 1532, 1440, 1353, 1174, 1001,
926 cmꢀ1; 1H NMR(200 MHz, CDCl 3) d 9.40–9.05 (brs,
2H), 8.70–8.50 (brs, 1H), 6.37 (s, 1H), 5.89–5.68 (m,
1H), 5.24–5.15 (m, 2H), 3.76–3.63 (m, 1H), 2.54–2.30
(m, 4H), 2.03 (s, 3H); MS and HR-MS analysis showed
only the ion from C9H14N2 corresponding to the loss of
HCl, MS (APCI, Pos.) m/z 151 (M+H)+, 109; HR-MS
(MALDI-TOF, Pos.) calcd for C9H14N2+ H+:
151.1235; found: 151.1240.
1366, 1336, 1168, 796, 667 cmꢀ1; H NMR(200 MHz,
DMSO-d6) d 9.28 (brs, 1H), 9.12 (brs, 1H), 8.73 (brs,
1H), 6.80 (d, J=9.8 Hz, 1H), 6.11 (dd, J=9.8, 2.0 Hz,
1H), 3.19 (d, J=2.6 Hz, 2H), 1.07 (s, 6H); mp: 131–
133 ꢁC; MS and HR-MS analysis showed only the ion
from C7H12N2 corresponding to the loss of HCl, MS
(EI, Pos.) m/z 124 (M)+, 123 (MꢀH)+, 109 (MꢀCH3)+,
94 (Mꢀ2CH3)+; HR-MS (MALDI-TOF, Pos.) calcd for
C7H12N2+ H+: 125.1079; found: 125.1055.
4,5-Dimethyl-5,6-dihydropyridin-2(1H)-imine hydrochlo-
ride salt (8). Compound 8 was prepared from 27d in
21% yield according to the same procedure as described
for the preparation of 1 from 27a. White powder; TLC
Rf 0.50 (CHCl3/MeOH/AcOH, 10/2/1); IR(KBr) 3387,
1686, 1607, 1533, 1457, 1436, 1386, 1360, 1332, 1294,
(dl)-(4aR,7aR)-4-Methyl-1,4a,5,6,7,7a-hexahydro-2H-cy-
clopenta[b]pyridin-2-imine hydrochloride salt (12). Com-
pound 12 was prepared from 27h in 70% yield
according to the same procedure as described for the
preparation of 1 from 27a. White powder; TLC Rf 0.44
(CHCl3/MeOH/AcOH, 10/1/1); IR(KBr) 3398, 3122,
2965, 1687, 1645, 1620, 1536, 1443, 1391, 1365, 1312,
1
1181, 1019, 966, 849, 636 cmꢀ1; H NMR(200 MHz,
1
DMSO-d6) d 9.13 (brs, 1H), 8.92 (brs, 1H), 8.53 (brs,
1H), 5.97–5.92 (m, 1H), 3.43 (ddd, J=13.2, 6.0, 1.8 Hz,
1H), 3.17 (ddd, J=13.2, 5.0, 4.4 Hz, 1H), 2.51–2.40 (m,
1H), 2.01 (s, 3H), 1.05 (d, J=7.2 Hz, 3H); mp: 109–
110 ꢁC; MS and HR-MS analysis showed only the ion
from C7H12N2 corresponding to the loss of HCl, MS
(EI, Pos.) m/z 124 (M)+, 109 (MꢀCH3)+, 94
(Mꢀ2CH3)+; HR-MS (MALDI-TOF, Pos.) calcd for
C7H12N2+ H+: 125.1079; found: 125.1053.
1196, 1051, 851, 780, 672 cmꢀ1; H NMR(200 MHz,
DMSO-d6) d 9.41 (brs, 1H), 9.0 (brs, 1H), 8.37 (brs,
1H), 5.92 (s, 1H), 4.05–3.92 (m, 1H), 2.70 (q, J=8.6 Hz,
1H), 2.25–1.90 (m, 2H), 2.01 (s, 3H), 1.87–1.34 (m, 4H);
mp: 121–122 ꢁC; MS and HR-MS analysis showed only
the ion from C9H14N2 corresponding to the loss of HCl,
MS (FAB, Pos.) m/z 337 (2M+HCl+H)+, 151
(M+H)+, 122, 109; HR-MS (MALDI-TOF, Pos.)
calcd for C9H14N2+ H+: 151.1235; found: 151.1249.
4,6-Dimethyl-5,6-dihydropyridin-2(1H)-imine hydrochlo-
ride salt (9). Compound 9 was prepared from 27e in 34%
yield according to the same procedure as described for the
preparation of 1 from 27a. Beige amorphous; TLC Rf 0.15
(CHCl3/MeOH/AcOH, 10/1/1); IR(KBr) 3373, 1680,
1631, 1543, 1417, 1371, 1336, 1060, 881, 840, 762, 689,
620 cmꢀ1; 1H NMR(200Mz, DMSO -d6) d 8.98–8.82 (brs,
1H), 8.80–8.64 (brs, 1H), 8.08–7.92 (brs, 1H), 5.91 (s,
1H), 3.80–3.58 (m, 1H), 2.58–2.40 (m, 1H), 2.18 (dd,
J=18.0, 10.0 Hz, 1H), 1.98 (s, 3H), 1.18 (d, J=6.6 Hz,
3H); MS and HR-MS analysis showed only the ion from
C7H12N2 corresponding to the loss of HCl, MS (APCI,
Pos.) m/z 125 (M+H)+; HR-MS (MALDI-TOF, Pos.)
calcd for C7H12N2+ H+: 125.1079; found: 125.1085.
(dl)-(4aR,8aR)-4-Methyl-4a,5,6,7,8,8a-hexahydroquino-
lin-2(1H)-imine hydrochloride salt (13). Compound 13
was prepared from 27i in 44% yield according to the
same procedure as described for the preparation of 1
from 27a. White powder; TLC Rf 0.36 (CHCl3/MeOH/
AcOH, 10/1/1); IR(KBr) 3447, 3248, 2940, 2861, 1689,
1620, 1594, 1532, 1450, 1437, 1367, 1332, 1315, 1286, 1208,
1167, 1104, 1065, 1031, 992, 960, 917, 893, 856, 728, 640,
565, 509, 480 cmꢀ1; 1H NMR(200 MHz, DMSO -d6) ꢀ 9.26
(brs, 1H), 8.90 (brs, 1H), 8.13 (brs, 1H), 5.96–5.92 (m, 1H),
3.70–3.61 (m, 1H), 2.35–2.21 (m, 1H), 2.02 (s, 3H), 1.95–
0.95 (m, 8H); mp: 76–77 ꢁC; MS and HR-MS analysis
showed only the ion from C10H16N2 corresponding to
the loss of HCl, MS (EI, Pos.) m/z 164 (M)+, 149 (M-
CH3)+, 135, 121, 109; HR-MS (MALDI-TOF, Pos.)
calcd for C10H16N2+ H+: 165.1392; found: 165.1379.
4 - Methyl - 6 - propyl - 5,6 - dihydropyridin - 2(1H) - imine
hydrochloride salt (10). Compound 10 was prepared
from 27f in 67% yield according to the same procedure
as described for the preparation of 1 from 27a. Brown
oil; TLC Rf 0.30 (CHCl3/MeOH/AcOH, 20/1/1); IR
(neat) 3393, 2961, 1685, 1638, 1605, 1536, 1439, 1382,
4-Methyl-1,5,6,7-tetrahydro-2H-azepin-2-imine hydro-
chloride salt (14). Compound 14 was prepared from
35a in 22% yield according to the same procedure as
described for the preparation of 1 from 27a. Beige vis-
cous amorphous; TLC Rf 0.25 (CHCl3/MeOH/AcOH,
10/1/1); IR(KBr) 3115, 1672, 1616, 1438, 1386, 1084,
1351, 1177, 835, 654 cmꢀ1
;
1H NMR(200 MHz,
DMSO-d6) d 9.38 (brs, 1H), 8.98 (brs, 1H), 8.42 (brs,
1H), 5.97 (s, 1H), 3.68–3.45 (m, 1H), 2.60–2.40 (m, 1H),
2.21 (dd, J=18.0, 10.0 Hz, 1H), 1.98 (s, 3H), 1.70–1.10
(m, 4H), 0.88 (t, J=7.0 Hz, 3H); MS and HR-MS ana-
lysis showed only the ion from C9H16N2 corresponding
to the loss of HCl, MS (EI, Pos.) m/z 152 (M)+, 109, 92;
HR-MS (MALDI-TOF, Pos.) calcd for C9H16N2+
H+: 153.1392; found: 153.1364.
1
694 cmꢀ1; H NMR(200 MHz, DMSO- d6) d 9.48–9.32
(brs, 1H), 8.67–8.46 (brs, 1H), 8.26–8.13 (brs, 1H), 5.88
(s, 1H), 3.32–3.24 (m, 2H), 2.52–2.43 (m, 2H), 1.98 (s,
3H), 1.95–1.78 (m, 2H); MS and HR-MS analysis
showed only the ion from C7H12N2 corresponding to
the loss of HCl, MS (FAB, Pos.) m/z 125 (M+H)+;
HR-MS (MALDI-TOF, Pos.) calcd for C7H12N2+
H+: 125.1079; found: 125.1077.
6-Allyl-4-methyl-5,6-dihydropyridin-2(1H)-imine hydro-
chloride salt (11). Compound 11 was prepared from
27g in 51% yield according to the same procedure as
described for the preparation of 1 from 27a. Brown oil;
4,7 - Dimethyl - 1,5,6,7 - tetrahydro - 2H - azepin - 2 - imine
hydrochloride salt (15). Compound 15 was prepared
from 35b in 44% yield according to the same procedure

Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
701
as described for the preparation of 1 from 27a. Yellow
amorphous; TLC Rf 0.37 (CHCl3/MeOH/AcOH, 15/2/
1); IR(KBr) 3292, 1676, 1616, 1540, 1457, 1387,
(neat) 3243, 3099, 2961, 2873, 1681, 1608, 1539, 1471,
1455, 1049, 817, 667 cmꢀ1; 1H NMR(200 MHz, CDCl 3)
d 9.65–9.20 (brs, 1H), 8.75–8.43 (brs, 1H), 8.40–8.00
(brs, 1H), 6.33 (d, J=12.8 Hz, 1H), 6.08 (d, J=12.8 Hz,
1H), 3.52–3.33 (m, 1H), 2.00–1.38 (m, 6H), 1.14 (s, 3H),
1.13 (s, 3H), 0.97 (d, J=6.4 Hz, 3H); MS and HR-MS
analysis showed only the ion from C11H20N2 corre-
sponding to the loss of HCl, MS (APCI, Pos.) m/z 181
(M+H)+; HR-MS (MALDI-TOF, Pos.) calcd for
C11H20N2+ H+: 181.1705; found: 181.1685.
1321 cmꢀ1 1H NMR(200 MHz, CDCl 3) d 9.71 (brs,
;
1H), 8.53 (brs, 1H), 8.35 (brs, 1H), 6.17 (s, 1H), 3.63 (m,
1H), 2.49 (m, 2H), 2.04 (s, 3H), 1.99–1.91 (m, 2H), 1.38
(d, J=6.8 Hz, 3H); MS and HR-MS analysis showed
only the ion from C8H14N2 corresponding to the loss of
HCl, MS (APCI, Pos.) m/z 139 (M+H)+; HR-MS
(MALDI-TOF, Pos.) calcd for C8H14N2+ H+:
139.1235; found: 139.1213.
Preparation of partially purified enzyme and
determination of Ki values
4-Methyl-7-propyl-1,5,6,7-tetrahydro-2H-azepin-2-imine
hydrochloride salt (16). Compound 16 was prepared
from 35c in 19% yield according to the same procedure
as described for the preparation of 1 from 27a. White
powder; TLC Rf 0.58 (CHCl3/MeOH/AcOH, 15/2/1);
IR(KBr) 3234, 3117, 1678, 1640, 1609, 1534, 1438,
646 cmꢀ1; 1H NMR(200 MHz, CDCl 3) d 9.82 (brs, 1H),
8.91 (brs, 1H), 7.82 (brs, 1H), 6.08 (s, 1H), 3.44 (m, 1H),
2.48 (m, 2H), 2.04 (s, 3H), 2.10–1.40 (m, 6H), 0.95 (t,
J=7.0 Hz, 3H); mp: 103–104 ꢁC; MS and HR-MS ana-
lysis showed only the ion from C10H18N2 corresponding
to the loss of HCl, MS (APCI, Pos.) m/z 167 (M+H)+;
HR-MS (MALDI-TOF, Pos.) calcd for C10H18N2+
H+: 167.1548; found: 167.1552.
Human eNOS was overexpressed in Sf-21 cells, by
infecting the cells with baculovirus carrying heNOS
cDNA. The hiNOS was overexpressed in A549 by sti-
mulation with LPS (10 mg/mL) plus cytokines (10 ng/mL
TNF-a, 5 ng/mL IL-1 b and 100 ng/mL interferon-g).
Human eNOS and iNOS were partially purified by
chromatography on 20, 50-ADP-Sepharose gels. NOS
activity was determined by the method for the conver-
sion of [14C]-l-arginine to l-citrulline with a minor
modification. The conversion rates for various con-
centrations of the test compounds and l-arginine were
measured. Dixon and Lineweaver–Burk plots were con-
structed to determine the Ki values and the mode of
inhibition.
4,6 - Dimethyl - 1,5,6,7 - tetrahydro - 2H - azepin - 2 - imine
hydrochloride salt (17). Compound 17 was prepared
from 35d in 34% yield according to the same procedure
as described for the preparation of 1 from 27a. Orange
viscous oil; TLC Rf 0.38 (CHCl3/MeOH/AcOH, 10/1/
1); IR(neat) 3292, 3152, 2964, 1678, 1641, 1615, 1537,
1454 cmꢀ1; 1H NMR(200 MHz, DMSO- d6) d 9.64–9.54
(brs, 1H), 8.78–8.63 (brs, 1H), 8.49–8.37 (brs, 1H), 5.91
(s, 1H), 3.40–2.94 (m, 2H), 2.62–2.42 (m, 1H), 2.25–2.06
(m, 2H), 1.98 (s. 3H), 0.91 (d, J=6.6 Hz, 3H); MS and
HR-MS analysis showed only the ion from C8H14N2
corresponding to the loss of HCl, MS (APCI, Pos.) m/z
139 (M+H)+; HR-MS (MALDI-TOF, Pos.) calcd for
C8H14N2+ H+: 139.1235; found: 139.1227.
Enzyme assay with recombinant mouse iNOS
Recombinant mouse iNOS was purchased from Cay-
man Chemical (Cat. No. 60862) and the inhibitory
activity of each test compound was evaluated by mea-
suring the conversion rate from [14C]-l-arginine to
[14C]-l-citrulline. Then the IC50 values of the com-
pounds were determined.
The ID50 values were determined from log-logit trans-
formation of the dose–response curves (1, 3, 5–6, 8–12
and 14–17; 0.1, 0.3, 1 mg/kg, sc, l-NMMA; 10, 30 and
100 mg/kg, sc). The ID50 value was defined as the dose
of a test compound that produced 50% inhibition in the
NOx accumulation induced by LPS treatment alone.
The MTD was defined as the maximum dose at which
no death was observed within 24 h after an intravenous
injection administration. The doses used were 5, 10, 20,
30, 40, and 50 mg/kg for 1, 3, 5–6, 8–12 and 14–17,
while 1000, 2000, 3000, 4000, and 5000 mg/kg were used
for l-NMMA.
5,5 - Dimethyl - 1,5,6,7 - tetrahydro - 2H - azepin - 2 - imine
hydrochloride salt (18). Compound 18 was prepared
from 35e in 16% yield according to the same procedure
as described for the preparation of 1 from 27a. Yellow
powder; TLC Rf 0.40 (CHCl3/MeOH/AcOH, 10/1/1);
IR(KBr) 3401, 1681, 1624, 1534, 1474, 1385, 1350,
1309, 1253, 1222, 1182, 1125, 1085 cmꢀ1 1H NMR
;
(200 MHz, DMSO-d6) d 10.07–9.89 (brs, 1H), 9.13–8.94
(brs, 1H), 8.70–8.53 (brs, 1H), 6.45 (d, J=12.8 Hz, 1H),
5.90 (d, J=12.8 Hz, 1H), 3.38–3.20 (m, 2H), 1.77–1.72
(m, 2H), 1.07 (s, 6H); mp: 98–100 ꢁC; MS and HR-MS
analysis showed only the ion from C8H14N2 corre-
sponding to the loss of HCl, MS (APCI, Pos.) m/z 139
(M+H)+; HR-MS (MALDI-TOF, Pos.) calcd for
C8H14N2+ H+: 139.1235; found: 139.1220.
Inhibition of NOx accumulation and the maximum
tolerated dose (MTD) in mice
The test compounds or saline were administered sub-
cutaneously at 3 h after LPS (10 mg/kg, iv) injection into
7 weeks old Balb/c mice (Charles River Japan, Inc.).
Blood was collected by venipuncture from the abdom-
inal aorta under light anesthesia at 6 h after LPS treat-
ment. Plasma was obtained by centrifugation and the
concentration of accumulated NOx over 3 h was deter-
mined by the method described below. To evaluate the
acute toxicity, the MTD (iv maximum dose where no
5,5-Dimethyl-7-propyl-1,5,6,7-tetrahydro-2H-azepin-2-
imine hydrochloride salt (19). Compound 19 was pre-
pared from 35f in 10% yield according to the same
procedure as described for the preparation of 1 from
27a. Brown oil; TLC Rf 0.25 (CHCl3/MeOH, 10/1); IR

702
Y. Kawanaka et al. / Bioorg. Med. Chem. 11 (2003) 689–702
death was observed within 24 h after the administration)
of the test compound was determined.
References and Notes
1. Moncada, S.; Palmer, R. M. J.; Higgs, E. A. Pharmacol.
Rev. 1991, 43, 109.
Measurement of nitrite/nitrate
2. (a) Nathan, C.; Xie, Q.-W. Cell 1994, 78, 915. (b) Kerwin,
J. F., Jr.; Lancaster, J. R., Jr.; Feldman, P. L. J. Med. Chem.
1995, 38, 4343.
3. (a) Stuehr, D. J.; Griffith, O. W. Adv. Enzymol. Rev. Mol.
Biol. 1992, 65, 287. (b) Marletta, M. A. J. Biol. Chem. 1993,
268, 12231.
4. Kerwin, J. F., Jr.; Heller, M. Med. Res. Rev. 1994, 14, 23.
5. Marletta, M. A. J. Med. Chem. 1994, 37, 1899.
6. Olken, N. M.; Marletta, M. A. Biochemistry 1993, 32, 9677.
7. Feldman, P. L.; Griffith, O. W.; Hong, H.; Stuehr, D. J. J.
Med. Chem. 1993, 36, 491.
Nitrite and nitrate, the oxidized form of nitric oxide that
accumulated in the culture medium and plasma were
determined by the use of nitrite/nitrate colorimetric
assay kit (Cayman Chemical, Cat. No. 780001). Basi-
cally, the nitrate in the sample was reduced to nitrite
with a nitrate reductase contained in the assay kit;
nitrite levels were then determined spectro-
photometrically as the total NOx concentration.
Pharmacokinetic (PK) studies
8. Furfine, E. S.; Harmon, M. F.; Paith, J. E.; Garvey, E. P.
Biochemistry 1993, 32, 8512.
Pharmacokinetic parameters were determined in rats
after intravenous (2 mg/kg, iv) or oral (10 mg/kg, po)
administration. Rats (N=3) were given the test com-
pounds intravenously or orally in physiological saline
(1 mL/kg, iv or 2.5 mL/kg, po). Blood samples (250 mL)
were collected from the jugular veins at 0.25, 0.5, 1, 2, 3,
4, 6 and 8 h after dosing.
9. Rees, D. D.; Palmer, R. M. J.; Schulz, R.; Hodson, H. F.;
Moncada, S. Br. J. Pharmacol. 1990, 101, 746.
10. Narayanan, K.; Griffith, O. W. J. Med. Chem. 1994, 37,
885.
11. Misko, T. P.; Moore, W. M.; Kasten, T. P.; Nickols,
G. A.; Corbett, J. A.; Tilton, R. G.; McDaniel, M. L.; Wil-
liamson, J. R.; Currie, M. G. Eur. J. Pharmacol. 1993, 233,
119.
12. Garvey, E. P.; Oplinger, J. A.; Tanoury, G. J.; Sherman,
P. A.; Fowler, M.; Marshall, S.; Harmon, M. F.; Paith, J. E.;
Furfine, E. S. J. Biol. Chem. 1994, 269, 26669.
13. Nakane, M.; Klinghofer, V.; Kuk, J. E.; Donnelly, J. L.;
Budzik, G. P.; Pollock, J. S.; Basha, F.; Carter, G. W. Mol.
Pharmacol. 1995, 47, 831.
14. Webber, R. K.; Metz, S.; Moore, W. M.; Connor, J. R.;
Currie, M. G.; Fok, K. F.; Hagen, T. J.; Hansen, D. W., Jr.;
Jerome, G. M.; Manning, P. T.; Pitzele, B. S.; Toth, M. V.;
Trivedi, M.; Zupec, M. E.; Tjoeng, F. S. J. Med. Chem. 1998,
41, 96.
Plasma samples were analyzed after extraction of the
test compounds by simple liquid–liquid extraction. LC/
MS/MS analysis was performed with a Quattro II
(Micromass Co., Ltd.) mass spectrometer consisting of
a Perkin–Elmer series 200 solvent delivery system, a
Perkin–Elmer ISS 200 autoinjector, and a Waters Sym-
metry octyl mini-bore column (2.1ꢃ50 mm).
Docking study
15. Hagmann, W. K.; Galdwell, C. G.; Chen, P.; Durette,
P. L.; Esser, C. K.; Lanza, T. J.; Kopka, I. E.; Guthikonda,
R.; Shah, S. K.; MacCoss, M.; Chabin, R. M.; Fletcher, D.;
Grant, S. K.; Green, B. G.; Humes, J. L.; Kelly, T. M.; Luell,
S.; Meurer, R.; Moore, V.; Pacholok, S. G.; Pavia, T.; Wil-
liams, H. R.; Wong, K. K. Bioorg. Med. Chem. Lett. 2000, 10,
1975.
16. Kawanaka, Y.; Kobayashi, K.; Kusuda, S.; Tatsumi, T.;
Murota, M.; Nishiyama, T.; Hisaichi, K.; Fujii, A.; Hirai, K.;
Naka, M.; Komeno, M.; Nakai, H.; Toda, M. Bioorg. Med.
Chem. Lett. 2002, 12, 2291.
Docking study was performed using InsightII/Discover
molecular modeling package (Accelrys, San Diego, CA)
with CVFF force field on a SGI Octane2 workstation
with R12000 processors. Simulated-annealing procedure
in gas phase was used to search the complex model of
the lowest-energy. Initial protein structure of hiNOS
(PDB code 2NSI) was obtained from the Protein Data
Bank. All the hydrogens were added and their positions
were energetically optimized. During simulation, all the
heavy atoms were fixed (dielectric constant: e=4r).
17. Kawanaka, Y.; Kobayashi, K.; Kusuda, S.; Tatsumi, T.;
Murota, M.; Nishiyama, T.; Hisaichi, K.; Fujii, A.; Hirai, K.;
Naka, M.; Komeno, M.; Nakai, H.; Toda, M. Synlettt 2002, 8,
1233.
First of all, a complex model of low-energy, consisting
of 10 with hiNOS was constructed, in which 10 was put
near the Glu377 residue of the enzyme to form a salt
bridge with its amidine group. A complex model was
first equilibrated by running dynamics (dielectric con-
stant: e=4r) for 6 ps while increasing the temperature
from 50 to 1200 K by time steps of 1 fs, after which the
resulting conformations were sampled every 5000 steps
over a span of 300 ps at 1200 K to yield 60 snapshots.
Each snapshot was then equilibrated for 4 ps while
decreasing the temperature from 1200 to 200 K, fol-
lowed by 200 K simulation for 4 ps. Each annealed
model was energetically optimized using the steepest
descent method followed by the conjugate-gradient
method to an energy difference of 0.001 kcal/mol
between successive iterations.
18. Siegl, W. O. J. Heterocycl. Chem. 1981, 18, 1613.
19. (a) Toder, B. H.; Branca, J. S.; Dieter, R. K.; Smith, A. B.,
III. Synth. Commun. 1975, 5, 435. (b) Smith, A. B., III.; Wex-
ler, B. A.; Slade, J. Tetrahedron Lett. 1982, 23, 1631. (c)
Motoyoshiya, J.; Yazaki, T.; Hayashi, S. J. Org. Chem. 1991,
56, 735. (d) Smith, A. B., III.; Toder, B. H.; Branca, J. S.;
Dieter, R. K. J. Amer. Chem. Soc. 1981, 103, 1996. (e) Mag-
nus, P.; Quagliato, D. J. Org. Chem. 1985, 50, 1621. (f) Den-
mark, S. E.; Habermas, K. L.; Hite, G. A. Helv. Chim. Acta
1988, 71, 168.
20. (a) Shirota, F. N.; Nagasawa, H. T.; Elberling, J. A. J.
Med. Chem. 1977, 20, 1176. (b) Eilbracht, P.; Gersmeier, A.;
Lennartz, D.; Huber, T. Synthesis 1995, 3, 330.
21. Piers, E.; Wai, J. S. M. Can. J. Chem. 1994, 72, 146.
22. Mixture of nitric oxides.