Optimization of (4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepine-5-ylidene)acetamide derivatives as arginine vasopressin V2 receptor agonists and discussion of their binding modes

Article abstract of DOI:10.1016/j.bmc.2009.10.038

A series of (4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepine-5-ylidene)acetamide derivatives were optimized to achieve potent agonistic activity, both in vitro and in vivo, for the arginine vasopressin V₂ receptor, resulting in the eventual di

Full text of DOI:10.1016/j.bmc.2009.10.038

Bioorganic & Medicinal Chemistry 17 (2009) 8161–8167
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Optimization of (4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepine-5-ylidene)
acetamide derivatives as arginine vasopressin V2 receptor agonists and
discussion of their binding modes
a
a
a
a
Issei Tsukamoto a, , Hiroyuki Koshio , Masaya Orita , Chikashi Saitoh , Hiroko Yanai-Inamura ,
*
Chika Kitada-Nozawa a, Eisaku Yamamoto a, Takeyuki Yatsu a, Shuichi Sakamoto b, Shin-ichi Tsukamoto a
a Drug Discovery Research, Astellas Pharma Inc., 21 Miyukigaoka, Tsukuba-shi, Ibaraki 305-8585, Japan
b Technology, Astellas Pharma Inc., 3-17-1 Hasune, Itabashi, Tokyo 174-8612, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A
series of (4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepine-5-ylidene)acetamide derivatives were
Received 12 September 2009
Revised 19 October 2009
Accepted 20 October 2009
Available online 23 October 2009
optimized to achieve potent agonistic activity, both in vitro and in vivo, for the arginine vasopressin
V2 receptor, resulting in the eventual discovery of compound 1g. Molecular modeling of compound 1g
with V2 receptor was also examined to evaluate the binding mode of this series of compounds.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
(4,4-Difluoro-1,2,3,4-tetrahydro-5H-1-
benzazepin-5-ylidene)acetamide
derivatives
Vasopressin V2 receptor
Anti-diuretic activity
Homology modeling
1. Introduction
binding and agonistic activity. In particular, the presence of a polar
hetero atom at a certain position often increased the compound’s
Arginine vasopressin (AVP) is a cyclic nonapeptide that is pro-
duced and secreted by the hypothalamo-neurohypophysial sys-
tem. Stimulation of the V2 receptor with AVP induces water
reabsorption in the kidneys by increasing cAMP with subsequent
activation of the aquaporin-2 water channel, resulting in a reduc-
tion in urine volume.1 This implies that a V2 receptor agonist
may be used to treat diseases such as central diabetes insipidus
and nocturia.2 Well-known compounds which exhibit this mecha-
nism of action include desmopressin (dDAVP),3 OPC-51803,4 and
VNA9325 (Fig. 1). More recently, the utility of novel benzylurea
derivatives has been reported.6 In a previous paper, we reported
on the primary exploration of (4,4-difluoro-1,2,3,4-tetrahydro-
5H-1-benzazepin-5-ylidene)acetamide derivatives as novel V2
receptor agonists (Fig. 2).7
affinity for the V2 receptor (Fig. 2). For example, 2-hydroxyethyl
group in the head position is preferable.
Here, in order to develop a derivative with potent in vitro and
in vivo V2 agonistic activity, we describe the structural optimization
of this series of compounds by combining above-mentioned head
and tail structures to discover (2Z)-2-(4,4-difluoro-1-{4-[(3S)-3-
methylpyrrolidin-1-yl]-2-(trifluoromethyl)benzoyl}-1,2,3,4-tetra-
hydro-5H-1-benzazepin-5-ylidene)-N-(2-hydroxyethyl)acetamide
(1g), which showed potent V2 agonistic activity both in vitro and in
vivo. We also discuss the binding of compound 1g with the V2
receptor, using the molecular modeling method.
2. Chemistry
In our previous study, we showed that the structure of the tail
moiety was crucial in achieving expression of agonistic activity.
Specifically, the tail had to be a less sterically hindered hydropho-
bic moiety, such as 3-methylpyrrazolyl or 3-methylpyrrolidinyl. In
contrast, the head moiety was found to have wide latitude for V2
The synthetic route for the compounds evaluated for their V2
activity is shown in Scheme 1. Treatment of methyl 2-chloro-4-flu-
orobenzoate (2) or methyl 4-fluoro-2-(trifluoromethyl)benzoate
(3) with cyclic amines gave 4b–4g. Hydrolysis of the ester groups
and subsequent amidations with methyl (4,4-difluoro-1,2,3,4-tet-
rahydro-5H-1-benzazepin-5-ylidene)acetate8 using corresponding
acid chlorides resulted in compounds 5b–5g. The esters in the head
moiety were converted to the carboxylic acids and then reacted
* Corresponding author. Tel.: +81 29 863 6682; fax: +81 29 852 5387.
E-mail address: issei.tsukamoto@jp.astellas.com (I. Tsukamoto).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.10.038

8162
I. Tsukamoto et al. / Bioorg. Med. Chem. 17 (2009) 8161–8167
NH2
O
O
O
NH2
O
O
H
N
NH2
NH
H
N
NH2
NH
HN
HN
NH2
NH2
N
N
H
O
HO
O
HN
H
O
O
HO
O
HN
O
S
NH
O
S
NH
O
O
H
N
N
O
H
N
N
O
NH
S
NH
S
NH2
NH2
O
O
N
O
N
H
O
O
H
O
NH2
O
AVP
Desmopressin
N
H
N
N
N
N
N
O
Cl
O
O
N
H
N
N
O
N
O
N
Cl
OPC-51803
Benzylurea derivative
VNA932
Figure 1. Chemical structures of AVP and well-known arginine vasopressin V2 receptor agonists.
"head"-moiety
O
O
O
OH
N
H
N
N
H
N
N
H
F
F
F
F
F
F
N
N
N
O
O
O
O
N
"tail"-moiety
N
N
Cl
N
H
N
Cl
1a
1b
YM-35278
V2 receptor antagonist
V2 receptor agonist
Figure 2. An outline of our previous study regarding (4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepine-5-ylidene)acetamide derivatives as arginine vasopressin V2 receptor
agonists.
with 2-hydroxyethylamine in the presence of N-ethyl-N0-(3-
dimethylaminopropyl)carbodiimide hydrochloride and 1-hydroxy-
benzotriazole to yield the target derivatives 1b–1g (Scheme 1).
study, the (S)-3-methylpyrrolidine derivative exhibited more po-
tent anti-diuretic activity in rats, while both of the (R)-3-methyl-
pyrrolidine and (S)-3-methylpyrrolidine had nearly identical
in vitro profiles, possibly due to the fact that the (S)-3-methylpyr-
rolidine derivative is more stably metabolized in rats. As a collat-
eral evidence, the pharmacodynamics of compound 1e and 1g
are illustrated in Figure 3, in which compound 1g showed more
consistent suppression effect of urinary excretion rate than 1e.
Thus we successfully obtained the selective, potent, orally
active V2 receptor agonist 1g after optimizing several moieties
(see Table 1).
3. Results and discussion
To optimize the derivatives, 3-methylpyrazole, (R)-3-methyl-
pyrrolidine, and (S)-3-methylpyrrolidine were introduced into
the tail moiety, while the head moiety was fixed with 2-hydroxy-
ethyl amide. At the same time, introduction of either chloro- or tri-
fluoromethyl group as R2 substitution was also examined.
A structural model of the human V2 receptor-1g complex is
shown in Figure 4. A homology model for the human V2 receptor
was constructed using the recently identified bovine rhodopsin
crystal structure9 as a template, and a docking study of 1g was per-
formed using GOLD ver. 3.1.1 (Cambridge Crystallographic Data Cen-
ter, Cambridge, UK).10
Results from the docking study indicated that the binding site
for 1g is located in a pocket surrounded by the human V2 recep-
tor transmembrane helices III, V, VI, and VII. The tail moiety of 1g
occupies the bottom of the pocket where hydrophobic structures
Most compounds (1b, 1d–1g) showed potent binding affinity to
the V2 receptor, with Ki values of approximately 10 nM, though
only 1c was found to reduce the affinity. With regard to the substi-
tution of the benzene ring, replacing the chloro group with a tri-
fluoromethyl group resulted in decreased binding affinity to the
V1a receptor in every case. Attaching 3-methylpyrrolidine ring as
tail moiety led to significantly more potent intrinsic efficacy com-
pared to the 3-methylpyrazole derivative. Further, 1e and 1g
exhibited excellent in vivo anti-diuretic activity, with respective
ED50 values of 0.040 mg/kg po and 0.012 mg/kg po. In the present

I. Tsukamoto et al. / Bioorg. Med. Chem. 17 (2009) 8161–8167
8163
O
CO2Me
OH
N
H
F
F
F
F
CO2Me
R2
CO2Me
R2
b, c
d, e
a
N
N
F
R1
Me
O
O
N
N
2: R2=Cl
3: R2=CF3
N
N
N
R1
R2
, R2=Cl
, R2=CF3
, R2=Cl
4b : R1=
R1
R2
N
N
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
N
N
N
N
N
N
4c : R1=
4d : R1=
4e : R1=
4f: R1=
, R2=Cl
, R2=CF3
, R2=Cl
, R2=Cl
1b : R1=
5b: R1=
N
N
Me
Me
Me
Me
Me
, R2=CF3
, R2=Cl
1c : R1=
1d : R1=
1e : R1=
1f: R1=
5c: R1=
5d: R1=
5e: R1=
5f: R1=
N , R2=CF3
N
, R2=Cl
N , R2=CF
N , R2=CF3
3
N
, R2=CF3
N
N
N
4g: R1=
, R2=Cl
, R2=Cl
N
, R2=CF3
, R2=CF3
1g: R1=
5g: R1=
Scheme 1. Reagents and conditions: (a) amine, K2CO3, NMP; (b) aq NaOH, MeOH; (c) SOCl2, catalytic DMF, THF then methyl (4,4-difluoro-1,2,3,4-tetrahydro-5H-1-
benzazepin-5-ylidene)acetate,8 pyridine; (d) aq NaOH, MeOH; (e) 2-hydroxyethylamine, N-ethyl-N0-(3-dimethylaminopropyl)carbodiimide hydrochloride, 1-hydroxyben-
zotriazole, DMF.
4. Conclusion
100
80
Here, to develop a derivative with potent in vitro and in vivo V2
agonistic activity, a series of (4,4-difluoro-1,2,3,4-tetrahydro-5H-1-
benzazepine-5-ylidene)acetamide derivatives were optimized by
structural combination of head and tail moiety in addition to benzene
ring substitutions. These attempts led to the eventual discovery of
(2Z)-2-(4, 4-difluoro-1-{4-[(3S)-3-methylpyrrolidin-1-yl]-2-(trifluo-
romethyl)benzoyl}-1,2,3,4-tetrahydro-5H-1-benzazepin-5-ylidene)-
N-(2-hydroxyethyl)acetamide (1g), which showed extremely potent
in vivo anti-diuretic activity, with an ED50 value of 0.012 mg/kg po
in rats. Further, the binding mode of this series of derivatives was
determined in a molecular modeling study between compound 1g
and the human V2 receptor. The docking study well agreed to the ac-
tual SARs of this series of derivatives.
Vehicle
1e
60
40
1g
20
0
0
1
2
3
4
Time (h)
Figure 3. Pharmacodynamics of compounds 1e and 1g, orally-administered
0.03 mg/kg, on urinary excretion rate in water-loaded rats. The urinary excretion
rate is the ratio of urine volume to the volume of loaded water. All assays were
performed in n = 3–6.
5. Experiment
5.1. Chemistry
are preferred, and the head moiety extends into the solvent, as
this moiety is able to easily tolerate a range of chemical conver-
sion. These observations may help explain our previous findings.
Observations that substituting polar groups such as N-methyl
piperazine and dimethylamino pyrrolidine in the tail moiety
led to markedly reduced binding affinity than hydrophobic sub-
stitutions7a were well supported by the structure model of the
tail moiety. Further, with regard to the head moiety, structure
of the binding pocket revealed in the docking study showed that
this region had wide latitude for structural conversion and was
thus a good place to tune physiochemical properties as well as
pharmacological activities. The interaction between the 2-
hydroxyethyl group in the head moiety and Gln180 was also
considered helpful in achieving more effective binding between
this compound and the V2 receptor, indicating that the introduc-
tion of a polar atom into the appropriate position in the head
moiety (such as a 2-hydroxyethyl group or pyridin-2-yl-methyl
group7b) often increases binding potency.
In general, reagents and solvents were used as purchased with-
out further purification. Melting points were determined with a
Yanaco MP-500D melting point apparatus and left uncorrected.
1H NMR spectra were recorded on a JEOL JNM-LA300 or a JEOL
JNM-EX400 spectrometer. Chemical shifts were expressed in d
(ppm) values with tetramethylsilane as an internal standard
(NMR descriptions; s, singlet; d, doublet; t, triplet; q, quartet; dt,
double triplet; m, multiplet, and br, broad peak). Mass spectra
were recorded on a JEOL JMS-LX2000 spectrometer. The elemental
analyses were performed with a Yanaco MT-5 microanalyzer (C, H,
N) and Yokogawa IC-7000S ion chromatographic analyzer (halo-
gens) and were within 0.4% of theoretical values.
(2Z)-2-{1-[2-Chloro-4-(3-methyl-1H-pyrazol-1-yl)benzoyl]-4,4-di-
fluoro-1,2,3,4-tetrahydro-5H-1-benzazepin-5-ylidene}-N-(2-hydroxy-

8164
I. Tsukamoto et al. / Bioorg. Med. Chem. 17 (2009) 8161–8167
Table 1
Binding affinity, cAMP accumulation activity and in vivo activity of compounds 1b–1g
O
OH
N
H
F
F
N
O
R1
R2
R1
=
R2
Cl
=
Binding affinity: Kia (nM)
cAMP accumulation
Anti-diuretic activity
ED50 (mg/kg po)
d
IAc (%)
b
V2 (nM)
V1a (nM)
V1a/V2
EC50 (nM)
1b
1c
9.7
78
28
170
2.9
2.2
6.3
2.4
93.9
115
0.14
0.20
N
CF3
N
N
N
1d
1e
Cl
CF3
13
16
13
39
1.0
2.4
0.99
1.0
118
114
—
0.040
1f
1g
Cl
CF3
14
14
17
43
1.2
3.1
0.98
1.0
109
111
—
0.012
a
Binding affinity for human V2 and V1a receptors. Receptors expressed on CHO cells were used. All assays were performed in triplicate.
EC50 values were determined as the concentration of the test compound required to increase the cAMP level to 50% of the maximum response to AVP. All assays were
b
performed in triplicate.
c
Intrinsic activity (IA) was calculated as the percentage (%) of the maximum response to the test compound compared to the maximum response (100%) to AVP. All assays
were performed in triplicate.
d
Effects of oral administration of test compounds on urinary excretion rate in water-loaded rats. The ED50 value represents the dose of the test compounds required to
decrease the urinary excretion rate by 50%.
Figure 4. Structural model of the human V2 receptor-1g complex. The stick-and-ball representation shows the compound 1g.

I. Tsukamoto et al. / Bioorg. Med. Chem. 17 (2009) 8161–8167
8165
ethyl)acetamide (1b) was prepared from 2-chloro-4-fluorobenzoic acid
perature and subsequently poured into aqueous hydrochloric acid.
Resulting precipitates were collected by filtration and dried in va-
cuo to give 4-(3-methyl-1H-pyrazol-1-yl)-2-(trifluoromethyl) ben-
zoic acid (1.00 g, 97%) as white solid. A mixture of 4-(3-methyl-1H-
pyrazol-1-yl)-2-(trifluoromethyl) benzoic acid (1.00 g, 3.70 mmol),
DMF (several drops) and oxalyl chloride (0.80 mL) in tetrahydrofu-
ran (10 mL) was stirred at room temperature for 3 h. The reaction
mixture was evaporated and azeotroped with toluene. To a solu-
tion of this residue in pyridine (10 mL) was added methyl (2Z)-
(4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepin-5-ylidene)ace-
tate8 (850 mg, 3.36 mmol), and the mixture was stirred overnight
at room temperature. The solvent was evaporated, and the residue
was partitioned between water and chloroform. The organic phase
was washed with water and brine, dried over sodium sulfate, fil-
tered, and concentrated in vacuo. The crude product was purified
by silica gel column chromatography (n-hexane/ethyl ace-
tate = 4:1) to give the title compound 5c (1.70 g, quantitative yield)
as colorless amorphous solid. 1H NMR (400 MHz, DMSO-d6) d 2.23
(3H, s), 2.40–2.60 (1H, m), 3.15–3.30 (1H, m), 3.50–3.75 (1H, m),
3.79 (3H, s), 4.60–4.96 (1H, br), 6.37 (1H, d, J = 2.4 Hz), 6.74 (1H,
s), 6.84 (1H, d, J = 7.8 Hz), 6.86–6.97 (1H, m), 7.21 (1H, dt, J = 1.4,
7.8 Hz), 7.28 (1H, t, J = 7.8 Hz), 7.40 (1H, d, J = 7.3 Hz), 7.87 (1H,
d, J = 8.8 Hz), 8.09 (1H, s), 8.48 (1H, d, J = 2.5 Hz). MS (FAB) m/z
506 [M+H]+.
according to the procedure described in our previous paper.7
5.1.1. Methyl 4-(3-methyl-1H-pyrazol-1-yl)-2-(trifluoromethyl)
benzoate (4c)
A mixture of methyl 4-fluoro-2-(trifluoromethyl)benzoate (3;
20.8 g, 93.5 mmol), 3-methylpyrazole (10.8 g, 131 mmol) and
potassium carbonate (25.8 g, 187 mmol) in dimethylformamide
(DMF) (150 mL) was heated overnight at 100 °C. The reaction
was then allowed to cool to room temperature and partitioned be-
tween water and ethyl acetate. The organic phase was washed
with water and brine, dried over MgSO4, filtered, and concentrated
in vacuo. The crude product was purified by silica gel column chro-
matography (n-hexane/ethyl acetate = 15:1) to give the title com-
pound 4c (15.2 g, 57%) as white solid. 1H NMR (400 MHz, DMSO-
d6) d 2.30 (3H, s), 3.88 (3H, s), 6.45 (1H, d, J = 2.4 Hz), 8.00 (1H, d,
J = 8.4 Hz), 8.20 (1H, d, J = 9.6 Hz), 8.25 (1H, s), 8.64 (1H, d,
J = 2.0 Hz).
5.1.2. (R)-Methyl 2-chloro-4-(3-methylpyrrolidin-1-yl)benzoate
(4d)
Compound 4d was prepared according to the procedure de-
scribed for 4c. The title compound 4d (830 mg, 33%) was obtained
as white solid. 1H NMR (400 MHz, DMSO-d6) d 1.07 (3H, d,
J = 6.9 Hz), 1.53–1.64 (1H, m), 2.05–2.14 (1H, m), 2.29–2.40 (1H,
m), 2.82–2.89 (1H, m), 3.25–3.50 (3H, m), 3.75 (3H, s), 6.50 (1H,
dd, J = 8.8, 2.5 Hz), 6.54 (1H, d, J = 2.5 Hz), 7.75 (1H, d, J = 8.8 Hz).
MS (FAB) m/z 254 [M+1]+.
5.1.7. Methyl (2Z)-2-(1-{2-chloro-4-[(3R)-3-methylpyrrolidin-
1-yl]benzoyl}-4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepin-
5-ylidene)acetate (5d)
Compound 5d was prepared according to the procedure de-
scribed for 5c. The title compound 5d (330 mg, 22% from 4d) was
obtained as colorless amorphous solid. 1H NMR (400 MHz,
DMSO-d6) d 1.03 (3H, d, J = 6.4 Hz), 1.48–1.58 (1H, m), 1.97–2.09
(1H, m), 2.25–2.55 (3H, m), 2.64–2.77 (1H, m), 3.10–3.40 (4H,
m), 3.76 (3H, s), 4.40–5.05 (1H, br), 6.13–6.25 (1H, m), 6.34–6.40
(1H, m), 6.51 (1H, s), 6.60–6.67 (1H, m), 6.88–6.98 (1H, m), 7.18–
7.36 (3H, m). MS (FAB) m/z 475 [M+H]+.
5.1.3. (R)-Methyl 4-(3-methylpyrrolidin-1-yl)-2-(trifluoro
methyl)benzoate (4e)
Compound 4e was prepared according to the procedure de-
scribed for 4c. The title compound 4e (790 mg, 42%) was obtained
as white solid. 1H NMR (300 MHz, CDCl3) d 1.15 (3H, d, J = 6.6 Hz),
1.61–1.73 (1H, m), 2.12–2.23 (1H, m), 2.36–2.50 (1H, m), 2.89–2.97
(1H, m), 3.31–3.55 (3H, m), 3.86 (3H, s), 6.56 (1H, dd, J = 8.8,
2.6 Hz), 6.81 (1H, d, J = 2.4 Hz), 7.85 (1H, d, J = 8.8 Hz). MS (FAB)
m/z 288 [M+1]+.
5.1.8. Methyl (2Z)-2-(4,4-difluoro-1-{4-[(3R)-3-methyl pyrroli
din-1-yl]-2-(trifluoromethyl)benzoyl}-1,2,3,4-tetrahydro -5H-1-
benzazepin-5-ylidene)acetate (5e)
Compound 5e was prepared according to the procedure de-
scribed for 5c. The title compound 5e (1.07 g, 77% from 4e) was ob-
tained as colorless amorphous solid. 1H NMR (300 MHz, CDCl3) d
1.10 (3H, d, J = 6.2 Hz), 1.53–1.62 (3H, m), 2.05–2.83 (5H, m),
3.15–3.44 (3H, m), 3.84 (3H, s), 6.14–6.21 (2H, m), 6.60–6.68
(2H, m), 6.82–6.89 (1H, m), 7.06–7.33 (3H, m). MS (FAB) m/z 509
[M+H]+.
5.1.4. (S)-Methyl 2-chloro-4-(3-methylpyrrolidin-1-yl)benzoate
(4f)
Compound 4f was prepared according to the procedure de-
scribed for 4c. The title compound 4f (546 mg, 58%) was obtained
as white solid. 1H NMR (400 MHz, DMSO-d6) d 1.07 (3H, d,
J = 6.4 Hz), 1.53–1.64 (1H, m), 2.05–2.14 (1H, m), 2.29–2.40 (1H,
m), 2.82–2.89 (1H, m), 3.24–3.49 (3H, m), 3.75 (3H, s), 6.49 (1H,
dd, J = 8.8, 2.5 Hz), 6.53 (1H, d, J = 2.4 Hz), 7.75 (1H, d, J = 8.8 Hz).
MS (FAB) m/z 254 [M+1]+.
5.1.9. Methyl (2Z)-2-(1-{2-chloro-4-[(3S)-3-methylpyrrolidin-1-
yl]benzoyl}-4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepin-5-
ylidene)acetate (5f)
5.1.5. (S)-Methyl 4-(3-methylpyrrolidin-1-yl)-2-(trifluoro
methyl)benzoate (4g)
Compound 4g was prepared according to the procedure de-
scribed for 4c. The title compound 4g (710 mg, 60%) was obtained
as white solid. 1H NMR (400 MHz, DMSO-d6) d 1.08 (3H, d,
J = 6.4 Hz), 1.55–1.66 (1H, m), 2.07–2.16 (1H, m), 2.31–2.42 (1H,
m), 2.88–2.93 (1H, m), 3.28–3.55 (3H, m), 3.77 (3H, s), 6.74 (1H,
dd, J = 8.8, 2.4 Hz), 6.80 (1H, d, J = 2.5 Hz), 7.80 (1H, d, J = 8.8 Hz).
MS (FAB) m/z 288 [M+1]+.
Compound 5f was prepared according to the procedure de-
scribed for 5c. The title compound 5f (433 mg, 45% from 4f) was
obtained as colorless amorphous solid. 1H NMR (400 MHz,
DMSO-d6) d 1.03 (3H, d, J = 6.4 Hz), 1.47–1.58 (1H, m), 1.98–2.09
(1H, m), 2.23–2.55 (3H, m), 2.65–2.76 (1H, m), 3.10–3.40 (4H,
m), 3.76 (3H, s), 4.50–5.05 (1H, br), 6.13–6.25 (1H, m), 6.33–6.41
(1H, m), 6.50 (1H, s), 6.60–6.69 (1H, m), 6.88–6.98 (1H, m), 7.18–
7.36 (3H, m). MS (FAB) m/z 475 [M+H]+.
5.1.6. Methyl (2Z)-2-{4,4-difluoro-1-[4-(3-methyl-1H-pyrazol-
1-yl)-2-(trifluoromethyl)benzoyl]-1,2,3,4-tetrahydro-5H-1-
benzazepin-5-ylidene}acetate (5c)
A mixture of methyl 4-(3-methyl-1H-pyrazol-1-yl)-2-(trifluoro-
methyl)benzoate (4c; 1.09 g, 3.83 mmol) and aqueous sodium
hydroxide (5 M, 4.0 mL, 20 mmol) in methanol (20 mL) was re-
fluxed for 1 h. The reaction was then allowed to cool to room tem-
5.1.10. Methyl (2Z)-2-(4,4-difluoro-1-{4-[(3S)-3-methylpyrroli
din-1-yl]-2-(trifluoromethyl)benzoyl}-1,2,3,4-tetrahydro-5H-1-
benzazepin-5-ylidene)acetate (5g)
Compound 5g was prepared according to the procedure de-
scribed for 5c. The title compound 5g (840 mg, 68% from 4g) was
obtained as white solid. 1H NMR (400 MHz, DMSO-d6) d 1.03 (3H,

8166
I. Tsukamoto et al. / Bioorg. Med. Chem. 17 (2009) 8161–8167
d, J = 6.3 Hz), 1.48–1.62 (1H, m), 2.00–2.09 (1H, m), 2.27–2.50 (3H,
m), 2.71–2.79 (1H, m), 3.13–3.39 (4H, m), 3.77 (3H, s), 4.58–4.95
(1H, br), 6.38–6.81 (5H, m), 7.15–7.40 (3H, m). MS (FAB) m/z 509
[M+H]+.
Calcd for C27H28F5N3O3: C, 60.33; H, 5.25; N, 7.82; F, 17.67. Found:
C, 60.12; H, 5.10; N, 7.93; F, 17.86.
5.1.14. (2Z)-2-(1-{2-Chloro-4-[(3S)-3-methylpyrrolidin-1-yl]
benzoyl}-4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepin-5-
ylidene)-N-(2-hydroxyethyl)acetamide (1f)
5.1.11. (2Z)-2-{4,4-Difluoro-1-[4-(3-methyl-1H-pyrazol-1-yl)-2-
(trifluoromethyl)benzoyl]-1,2,3,4-tetrahydro-5H-1-benzazepin-
5-ylidene}-N-(2-hydroxyethyl)acetamide (1c)
Compound 1f was prepared according to the procedure de-
scribed for 1c. The title compound 1f (103 mg, 92% in hydrolysis
step, 79% in amidation) was obtained as colorless crystals. Mp:
Methyl (2Z)-2-{4,4-difluoro-1-[4-(3-methyl-1H-pyrazol-1-yl)-
2-(trifluoromethyl)benzoyl]-1,2,3,4-tetrahydro-5H-1-benzazepin-
5-ylidene}acetate (5c; 1.69 g, 3.36 mmol) in methanol (20 mL) was
treated with sodium hydroxide (1.0 M, 6.0 mL, 6.0 mmol) at room
temperature for 14 h. The reaction was partitioned between aque-
ous hydrochloride and chloroform. The organic phase was washed
with water and brine, dried over Na2SO4, filtered, and concentrated
in vacuo to give the corresponding carboxylic acid (1.78 g, quanti-
tative yield). A mixture of the foregoing carboxylic acid (350 mg,
0.71 mmol), 2-aminoethanol (0.060 mL, 1.07 mmol), WSCD
(274 mg, 1.07 mmol), and HOBt (106 mg, 1.07 mmol) in DMF
(10 mL) was stirred overnight at room temperature. The reaction
mixture was partitioned between water and ethyl acetate, then
the organic phase was washed with water and brine, dried over
Na2SO4, filtered, and concentrated in vacuo. The residue was puri-
fied by silica gel column chromatography (CHCl3/methanol = 35:1)
and subsequently crystallized from n-hexane/2-propanol = 4:1 to
give the title compound 1c (295 mg, 78%) as colorless crystals.
Mp: 125–128 °C. 1H NMR (400 MHz, DMSO-d6): d 2.24 (3H, s),
2.37–2.45 (1H, m), 2.71–2.87 (1H, m), 3.08–3.29 (4H, m), 3.49
(2H, t, J = 6.4 Hz), 4.70–4.92 (1H, br), 6.36 (1H, d, J = 2.5 Hz), 6.48
(1H, s), 6.77 (1H, d, J = 7.8 Hz), 7.03 (1H, d, J = 8.8 Hz), 7.15 (1H,
dt, J = 1.5, 7.8 Hz), 7.25 (1H, dt, J = 1.5, 7.8 Hz), 7.34 (1H, dd,
J = 1.5, 7.8 Hz), 7.84 (1H, dd, J = 1.5, 8.8 Hz), 8.09 (1H, d,
J = 1.5 Hz), 8.47 (1H, d, J = 2.5 Hz), 8.51 (1H, t, J = 5.3 Hz). MS
(FAB) m/z 535 [M+1]+. Anal. Calcd for C26H23F5N4O3ꢀ0.5H2O: C,
57.46; H, 4.45; N, 10.31; F, 17.48. Found: C, 57.53; H, 4.35; N,
10.10; F, 17.69.
124–126 °C. 1H NMR (400 MHz, DMSO-d6):
d 1.02 (3H, d,
J = 5.9 Hz), 1.45–1.57 (1H, m), 1.98–2.07 (1H, m), 2.20–2.34 (1H,
m), 2.65–2.75 (1H, m), 3.08–3.31 (6H, m), 3.43–3.49 (2H, m),
4.34 (1H, d, J = 3.9 Hz), 4.72 (1H, t, J = 5.4 Hz), 4.75–4.90 (1H, br),
6.13–6.18 (1H, m), 6.27 (1H, s), 6.36–6.40 (1H, m), 6.62–6.70
(1H, m), 6.81–6.90 (1H, m), 7.13–7.32 (3H, m), 8.32–8.41 (1H,
m). MS (FAB) m/z 504 [M+1]+. Anal. Calcd for C26H28ClF2N3O3ꢀ-
C3H8O: C, 61.75; H, 6.43; N, 7.45; Cl, 6.29; F, 6.74. Found: C,
62.04; H, 6.39; N, 7.51; Cl, 6.20; F, 6.76.
5.1.15. (2Z)-2-(4,4-Difluoro-1-{4-[(3S)-3-methylpyrrolidin-1-yl]-
2-(trifluoromethyl)benzoyl}-1,2,3,4-tetrahydro-5H-1-benz
azepin-5-ylidene)-N-(2-hydroxyethyl)acetamide (1g)
Compound 1g was prepared according to the procedure de-
scribed for 1c. The title compound 1g (157 mg, 93% in hydrolysis
step, 60% in amidation) was obtained as colorless crystals. Mp:
178–179 °C. 1H NMR (400 MHz, DMSO-d6):
d 1.03 (3H, d,
J = 6.3 Hz), 1.47–1.59 (1H, m), 2.00–2.10 (1H, m), 2.25–2.40 (2H,
m), 2.65–2.81 (2H, m), 3.14–3.50 (8H, m), 4.72 (1H, t, J = 5.4 Hz),
4.76–4.90 (1H, br), 6.35–6.42 (2H, m), 6.62 (1H, s), 6.65–6.75
(2H, m), 7.16 (1H, t, J = 7.3 Hz), 7.22 (1H, t, J = 7.3 Hz), 7.31 (1H,
d, J = 6.8 Hz), 8.43 (1H, s). MS (FAB) m/z 538 [M+1]+. Anal. Calcd
for C27H28F5N3O3: C, 60.33; H, 5.25; N, 7.82; F, 17.67. Found: C,
60.32; H, 5.01; N, 7.84; F, 17.38.
5.2. Docking study
The program, MOE (CCG, Montreal, Canada)11 was used to build
and geometry optimize the ligand structure of 1g. The energy min-
imizations were carried out with the MMFF94x force field. A
homology model for the human V2 receptor was constructed using
the recently determined the bovine rhodopsin crystal structure9 as
a template, using the program, MOE. Docking calculation of 1g was
performed with the program, GOLD ver.3.1.1 (CCDC, Cambridge,
UK).10
5.1.12. (2Z)-2-(1-{2-Chloro-4-[(3R)-3-methylpyrrolidin-1-yl]be
nzoyl}-4,4-difluoro-1,2,3,4-tetrahydro-5H-1-benzazepin-5-
ylidene)-N-(2-hydroxyethyl)acetamide (1d)
Compound 1d was prepared according to the procedure de-
scribed for 1c. The title compound 1d (126 mg, 94% in hydrolysis
step, 83% in amidation) was obtained as colorless crystals. Mp:
127–129 °C. 1H NMR (400 MHz, DMSO-d6):
d 1.02 (3H, d,
J = 5.4 Hz), 1.45–1.55 (1H, m), 1.98–2.05 (1H, m), 2.20–2.35 (1H,
m), 2.65–2.75 (1H, m), 3.08–3.31 (6H, m), 3.43–3.49 (2H, m),
4.33 (1H, d, J = 3.9 Hz), 4.71 (1H, t, J = 5.4 Hz), 4.75–4.90 (1H, br),
6.13–6.18 (1H, m), 6.27 (1H, s), 6.36–6.40 (1H, m), 6.62–6.69
(1H, m), 6.81–6.90 (1H, m), 7.13–7.32 (3H, m), 8.32–8.41
(1H, m). MS (FAB) m/z 504 [M+1]+. Anal. Calcd for C26H28ClF2
N3O3ꢀC3H8O: C, 61.75; H, 6.43; N, 7.45; Cl, 6.29; F, 6.74. Found: C,
61.39; H, 6.28; N, 7.54; Cl, 6.13; F, 6.62.
5.3. Biology
5.3.1. Binding assay for human V2 receptor
Chinese hamster ovary (CHO) cells stably expressing human V2
receptors, which were established by Tahara et al.12 were used.
Cells were washed with phosphate buffered saline, and then col-
lected in ice-cold hypotonic buffer (10 mmol/L Tris–HCl, 5 mmol/
L EDTA, pH 7.4). Subsequently, cells were collected using a cell
scraper and then homogenized using POLYTRONÒ followed by cen-
trifugation (1000g, 10 min) at 4 °C. The supernatant was centri-
fuged (35,000g, 30 min) at 4 °C, and the pellet was suspended in
Tris buffer. Membrane fractions were stored at ꢁ80 °C until used
for binding assay. The concentration of membrane protein was
determined by the Coomassie blue method using BSA as a
standard.
The affinities of test compounds for human V2 receptor were
evaluated by the radioligand binding study. For the competitive
binding study, 50
(final concentration of 0.91 nmol/L) were mixed with 150
membrane suspension in 50 mmol/L Tris–HCl (pH 7.4) buffer con-
5.1.13. (2Z)-2-(4,4-Difluoro-1-{4-[(3R)-3-methylpyrrolidin-1-yl]-
2-(trifluoromethyl)benzoyl}-1,2,3,4-tetrahydro-5H-1-benza
zepin-5-ylidene)-N-(2-hydroxyethyl)acetamide (1e)
Compound 1e was prepared according to the procedure de-
scribed for 1c. The title compound 1e (167 mg, quantitative yield
in hydrolysis step, 74% in amidation) was obtained as colorless
crystals. Mp: 174–177 °C. 1H NMR (400 MHz, DMSO-d6): d 1.03
(3H, d, J = 6.8 Hz), 1.47–1.60 (1H, m), 2.00–2.10 (1H, m), 2.25–
2.40 (2H, m), 2.65–2.81 (2H, m), 3.14–3.50 (8H, m), 4.72 (1H, t,
J = 5.4 Hz), 4.76–4.91 (1H, br), 6.35–6.43 (2H, m), 6.62 (1H, s),
6.65–6.74 (2H, m), 7.16 (1H, t, J = 7.3 Hz), 7.22 (1H, t, J = 7.3 Hz),
7.31 (1H, d, J = 6.8 Hz), 8.43 (1H, s). MS (FAB) m/z 538 [M+1]+. Anal.
lL of drug solution and 50
l
L of [3H]vasopressin
L of
l
taining 10 mmol/L MgCl2 and 0.1% bovine serum albumin in a final

I. Tsukamoto et al. / Bioorg. Med. Chem. 17 (2009) 8161–8167
8167
volume of 250
lL. This mixture was incubated at room tempera-
5.3.4. Anti-diuretic effect in water-loaded rats
ture for 60 min. Reactions were terminated by filtration through
UniFilterÒ GF/B (Perkin–Elmer) using a MicroMate Cell Harvester
(Packard Instrument Company, Meriden, CT, USA) and the filter
was washed with ice-cold Tris buffer. The radioactivity retained
on the filter was counted by TopCountTM microplate scintillation
counter (Perkin–Elmer) using the scintillation cocktail (MicroScin-
ti-40TM, Perkin–Elmer). Nonspecific binding or total binding were
A test to determine which rats would be selected for use was
performed at least a week before the beginning of the study. In this
test, male Wistar rats (SLC, 200–300 g) were given distilled water
(30 mL/kg) orally. Afterwards, the animals were kept in metabolic
cages, and urine was collected for 4 h after water loading. Animals
whose urinary excretion rate was at least 70% of the volume of
water-loaded (which was regarded as 100%) were used. While
the animals were deprived of feed and water, they orally received
the test compounds without anesthesia. Distilled water (30 mL/kg,
po) was loaded 15 min after administration, and the animals were
kept in a metabolic cage. Urine was collected for 4 h after water
loading. (n = 3–6) The urinary excretion rate (%) was calculated
by regarding the volume of loaded water as 100%. Linear regression
was performed to obtain the doses of the test compounds required
to decrease the urinary excretion rate to 50% (ED50). All analyses
were performed using SAS.
determined by including 1 lmol/L AVP or without test compounds
in the reaction mixture, respectively. The number of concentra-
tions of compounds was 11, appropriately chosen from 1 ꢂ 10ꢁ11
to 1 ꢂ 10ꢁ5 mol/L, using a common ratio of approximately three.
We also performed the saturation binding study to yield the disso-
ciation constants (Kd values) of [3H]vasopressin for each human V2
receptors. A membrane suspension was incubated with various
concentrations of [3H]vasopressin (0.1–3.2 nmol/L) in the absence
or presence of 1 lmol/L AVP. Assay conditions were the same as
those described for the competitive binding assay.
All values were determined by four separate experiments per-
formed in triplicate and represented as the mean SEM. Statistical
analysis was performed using a SAS software (SAS Institute, USA).
Specific binding was calculated as total binding minus nonspecific
binding. The concentration of each compound required to reduce
specific binding of [3H]vasopressin by 50% (IC50 value) was ob-
tained by non-linear regression analysis. A Kd value of [3H]vaso-
pressin for each vasopressin receptor was yielded by Scatchard
plot analysis. The affinity constants (Ki values) were calculated
from the following equation,13 using the Kd values yielded from
each separate experiment. Ki = IC50/(1 + [[3H]vasopressin concen-
tration]/Kd).
Acknowledgments
The authors deeply acknowledge Dr. Mitsuaki Ohta and Dr.
Minoru Okada for their helpful support in this study. The authors
are also grateful to the staff of the Division of Analytical Science
Laboratories for conducting the elemental analysis and spectral
measurements.
References and notes
1. (a) Hardman, J. G.; Limbird, L. E. Goodman & Gilman’s the Pharmacological Basis
of Therapeutics, 10th ed.; McGraw-Hill: New York, 2001; (b) Guyton, A. C.; Hall,
J. E. Textbook of Medical Physiology, 9th ed.; W.B. Saunders: Medina, 1999; (c)
Michel, R. H.; Kirk, C. J.; Billah, M. M. Biochem. Soc. Trans. 1979, 7, 861; (d)
Thibonnier, M.; Roberts, J. M. J. Clin. Invest. 1985, 76, 1857; (e) Laszlo, F. A.;
Laszlo, F.; De Weid, D. Pharmacol. Rev. 1991, 43, 73.
2. (a) Asplund, R.; Sundberg, B.; Bengtsson, P. BJU Int. 1999, 83, 591; (b) Weiss, J.
P.; Blaivas, J. G. J. Urol. 2000, 163, 5.
3. (a) Dimson, S. B. Lancet 1977, 1, 1260; (b) Tuvemo, T. Acta Pediatr. Scand. 1978,
67, 753.
5.3.2. Binding assay for human V1a receptor
The binding assay for human V1a receptor was performed in a
manner similar to that for human V2 receptor.
4. (a) Kondo, K.; Ogawa, H.; Shinohara, T.; Kurimura, M.; Tanada, Y.; Kan, K.;
Yamashita, H.; Nakamura, S.; Hirano, T.; Yamamura, Y.; Mori, T.; Tominaga, M.;
Itai, A. J. Med. Chem. 2000, 43, 4388; (b) Kondo, K.; Kan, K.; Tanada, Y.; Bando,
M.; Shinohara, T.; Kurimura, M.; Ogawa, H.; Nakamura, S.; Hirano, T.;
Yamamura, Y.; Kido, M.; Mori, T.; Tominaga, M. J. Med. Chem. 2002, 45, 3805.
5. (a) Venkatesan, A. M.; Grosu, G. T.; Failli, A. A.; Chan, P. S.; Coupet, J.; Saunders,
T.; Mazandarani, H.; Ru, X. Bioorg. Med. Chem. Lett. 2005, 15, 5003; (b) Failli, A.
A.; Shumsky, J. S.; Steffan, R. J.; Caggiano, T. J.; Williams, D. K.; Trybulsky, E. J.;
Ning, X.; Lock, Y.; Tanikella, T.; Hartmann, D.; Chan, P. S.; Park, C. H. Bioorg. Med.
Chem. Lett. 2006, 16, 954.
6. Christopher, M. Y.; Christine, E. A.; Doreen, M. A.; James, B.; Andy, J. B.; Janice,
D. B.; Richard, J. F.; Sally, L. H.; Peter, H.; John, A. H.; Paul, D. J.; Andy, M. P.;
Gary, R. W. P.; Pierre, R.; Peter, A. R.; David, P. R.; Graeme, S.; Andy, S.; Robert,
M. H.; Michael, B. R. J. Med. Chem. 2008, 51, 8124.
7. (a) Tsukamoto, I.; Koshio, H.; Kuramochi, T.; Akamatsu, S.; Saitoh, C.; Yanai-
Inamura, H.; Kitada, C.; Yamamoto, E.; Yatsu, T.; Sakamoto, S.; Tsukamoto, S.
Bioorg. Med. Chem. 2008, 16, 9524;; (b) Tsukamoto, I.; Koshio, H.; Kuramochi, T.;
Saitoh, C.; Yanai-Inamura, H.; Kitada-Nozawa, C.; Yamamoto, E.; Yatsu, T.;
Shimada, Y.; Sakamoto, S.; Tsukamoto, S. Bioorg. Med. Chem. 2009, 17, 3130.
8. Shimada, Y.; Akane, H.; Taniguchi, N.; Matsuhisa, A.; Kawano, N.; Tanaka, A.
Chem. Pharm. Bull. 2005, 53, 589.
5.3.3. Stimulatory effect on the production of intracellular
cAMP in human vasopressin V2 receptor
CHO cells stably expressing human V2 receptors, prepared by
Tahara et al. were used.12 The cells were incubated in
a-MEM, con-
taining 10% fetal bovine serum (FBS, Invitrogen Japan K.K.), 1% pen-
icillin/streptomycin (Invitrogen Japan K.K.), and 0.1% amethopterin
(dihydrofolate reductase inhibitor), in the absence of nucleic acid,
at 37 °C, in an atmosphere of 95% air/5% CO2.
CHO cells expressing human V2 receptors were grown to
subconfluence on a 96-well plate, and then incubated in serum-
free medium for 24 h before assay. The medium was replaced with
a-MEM containing 1 mmol/L 3-isobutyl-1-methylxantine (IBMX,
Sigma) and 0.1% bovine serum albumin (BSA, Sigma) then the test
compound was added and incubated at 37 °C for 10 min in order to
induce a reaction. The cells were then dissolved in phosphatebuf-
fered saline (PBS, Invitrogen Japan K.K.) containing 0.2% triton
X-100. The cAMP level in the cell lysate was determined using
the homogenous time resolved fluorescence (HTRF) assay with a
cyclic AMP kit (Nihon Schering K.K.).14
Intrinsic activity was calculated as the percentage (%) of the
maximum response to the test compound compared to the maxi-
mum response (100%) to AVP. All data analyses were performed
using SAS. The activities of test compounds for cAMP production
were calculated by logistic regression as EC50 values.
9. Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Motoshima, H.; Fox, B. A.;
Le Trong, I.; Teller, D. C.; Okada, T.; Stenkamp, R. E.; Yamamoto, M.; Miyano, M.
Science 2000, 289, 739–745.
10. Jones, G.; Willett, P.; Glen, R. C. J. Mol. Biol. 1995, 245, 43–53.
11. MOE Revision 2007.0902; Chemical Computing Group: Montreal, 2007.
12. Tahara, A.; Saito, M.; Sugimoto, T.; Tomura, Y.; Wada, K.; Kusayama, T.;
Tsukada, J.; Ishii, N.; Yatsu, T.; Uchida, W.; Tanaka, A. Br. J. Pharmacol. 1998, 125,
1463.
13. Lazareno, S.; Bridsall, N. Br. J. Pharmacol. 1993, 109, 1110.
14. Takasaki, J.; Kamohara, M.; Saito, T.; Matsumoto, M.; Matsumoto, S.; Ohishi, T.;
Soga, T.; Matsushime, H.; Furuichi, K. Mol. Pharmacol. 2001, 60, 432.