Design and synthesis of orally active benzamide derivatives as potent serotonin 4 receptor agonist

Article abstract of DOI:10.1016/S0968-0896(03)00412-7

A series of 4-amino-5-chloro-2-methoxy-N-(piperidin-4-ylmethyl)benzamide derivatives bearing an aralkylamino, alkylamino, benzoyl or phenylsulfonyl group at its side chain part at the 1-position on the piperidine ring was synthesized. They were evaluated

Full text of DOI:10.1016/S0968-0896(03)00412-7

Bioorganic & Medicinal Chemistry 11 (2003) 4225–4234
Design and Synthesis of Orally Active Benzamide Derivatives as
Potent Serotonin 4 Receptor Agonist
Shuji Sonda,a,* Toshio Kawahara,a Takahiro Murozono,b Noriko Sato,d
Kiyoshi Asanoc and Keiichiro Hagae
aPharmaceutical Development Laboratories, Technology & Production Division, Mitsubishi Pharma Corporation,
955, Koiwai, Yoshitomi-cho, Chikujo-gun Fukuoka 871-8550, Japan
bPharmacokinetics Laboratory, Research & Development Division, Mitsubishi Pharma Corporation, 1-1-1, Kazusa-kamatari,
Kisarazu, Chiba 292-0818, Japan
cResearch Laboratory I, Research & Development Division, Mitsubishi Pharma Corporation, 1000, Kamoshida-cho, Aoba-ku,
Yokahama 227-0033, Japan
dResearch Laboratory III, Research & Development Division, Mitsubishi Pharma Corporation, 1000, Kamoshida-cho, Aoba-ku,
Yokahama 227-0033, Japan
eProtein Research Laboratory, Research & Development Division, Mitsubishi Pharma Corporation, 2-25-1, Shodai-ohtani, Hirakata,
Osaka 573-1153, Japan
Received 23 January 2003; revised 24 March 2003; accepted 19 June 2003
Abstract—A series of 4-amino-5-chloro-2-methoxy-N-(piperidin-4-ylmethyl)benzamide derivatives bearing an aralkylamino, alkyl-
amino, benzoyl or phenylsulfonyl group at its side chain part at the 1-position on the piperidine ring was synthesized. They were
evaluated for serotonin 4 (5-HT4) receptor agonist activity by testing their ability to contract the isolated guinea-pig ascending
colon. 4-Amino-5-chloro-2-methoxy-N-[1-[5-(1-methylindol-3-ylcarbonylamino)pentyl]piperidin-4-ylmethyl]benzamide (1a, Y-
34959) and its related compounds possessed favorable pharmacological profiles for gastrointestinal motility. Unfortunately, the
compound 1a showed low bioavailability when given orally presumably due to its poor intestinal absorption rate. Replacement of
the 1-methylindol-3-yl carbonylamino moiety of 1a with an aralkylamino (or alkylamino) group did not improve the intestinal
absorption rate. Replacement of the 1-methylindol-3-ylcarbonylamino moiety with a benzoyl or phenylsulfonyl group increased the
intestinal absorption rate compared with 1a. These compounds revealed good pharmacological profiles for gastrointestinal motility
and were superior to 1a in oral bioavailability.
# 2003Elsevier Ltd. All rights reserved.
Introduction
pseudoobstruction, and characterized as the 5-HT4
receptor agonist. However, these benzamides have the
binding affinities for not only the 5-HT4 receptor but
also the dopamine D2, serotonin 5-HT2 and 5-HT3
receptors.6 Antagonisms against the dopamine D2, sero-
tonin 5-HT2 and 5-HT3 receptors would reduce the
prokinetic effect and cause unfavorable side effects.7
Therefore, we have made an effort to find out selective
and potent 5-HT4 receptor agonists without other
receptor binding affinities.7,8
Serotonin 4 (5-HT4) receptor, which was discovered by
Dumuis et al. in 1989,1 regulates cAMP concentration
in neuron synapse, and contracts intestinal smooth
muscle.2,3 It is known that the stimulation of this
receptor exerts activation of gastric motility, ileal moti-
lity and colonic transit.
Benzamides (metoclopramide,4 cisapride,5 etc.) have
been used clinically for the treatment of gastric disorder,
gastroesophageal reflux disease (GERD) and intestinal
In our previous papers, we reported a new type of ben-
zamides as the 5-HT4 receptor agonists.7,8 The benza-
mide derivative, 4-amino-5-chloro-2-methoxy-N-[1-[5-(1-
methylindol-3-yl carbonylamino)pentyl]piperidin-4-ylme-
thyl]benzamide (1a, Y-34959) and its related compounds
*Corresponding author. Tel.: +81-979-23-8972; fax: +81-979-24-
2769; e-mail: sonda.shuuji@ma.m-pharma.co.jp
0968-0896/03/$ - see front matter # 2003Elsevier Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00412-7

4226
S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
demonstrated good pharmacological profiles. Those
compounds also showed very high and selective binding
affinity for the 5-HT4 receptor, and had potent agonistic
activities. Accordingly, we proposed that 1a was a novel
gastrointestinal motility stimulant which can enhance
both upper and lower gastrointestinal motility with few
side effects.8
position on the piperidine ring as shown in Figure 1. On
the other hand, cisapride has a phenoxy frame at the
corresponding part. Therefore, modification of the 1-
methylindol-3-ylcarbonylamino moiety (A part) would
improve the oral absorption although we described the
importance of the 1-methylindol-3-ylcarbonylamino for
5-HT4 receptor agonistic activity in previous paper.8
Unfortunately, compound 1a showed low oral bio-
availability (5.1%) in dogs.9 Cisapride reveals higher
oral bioavailability in dogs (53%).6 We assumed that
compound 1a revealed poor oral bioavailability because
it has a 1-methylindol-3-ylcarbonylamino frame (A
part) on the terminal of the straight alkyl chain at the 1-
Here, we replaced the 1-methylindol-3-ylcarbonylamino
moiety with the aralkylamino (or alkylamino) group,
benzoyl or phenylsulfonyl group for improvement of
oral bioavailability. (It is known that alkylamine and
carbony groups are useful as a bioisostere of an amide
bond.10,11) As a result we discovered derivatives with
Figure 1. Design of orally active benzamine derivatives.
Scheme 1. Reagents: (a) benzaldehyde, toluene, reflux; (b) (Boc)2O; (c) KHSO4, H2O, rt, 12 h (92% in three steps); (d) EDC, HOBT, Et3N, 25 ꢀC, 24
h (92%); (e) HCl/dioxane, 25 ꢀC, 4 h (82%); (f) K2CO3, DMF, 70–75 ꢀC; (g) NH2NH2–H2O, EtOH, reflux, 6 h; (h) R1CHO,NaBH3CN; (i) R2CHO,
NaBH3CN, EtOH or NaBH4, EtOH.

S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
4227
Scheme 2. Reagents: (a) AlCl3, CH2Cl2, 25 ꢀC; (b) POCl3, CHCl3, reflux (38%); (c) K2CO3, DMF, 50–60 ꢀC; (d) m-CPBA or H2O2, HCOOH; (e)
K2CO3, DMF, 70–75 ꢀC, 5–12 h.
Table 1. Pharmacological data of 1a and 1b
Compd
R1
Binding affinitiesa
Potencyb
5-HT4, Ki (nM)
5-HT3, IC50 (nM)
EC50 (nM)c
1.2
% Absorptiond
1a (Y-34959)
0.3
1.7
>1000
6.9
1b
>1000
3.7
29
18.9
97.0
Cisapride
70
200
aEach value is the mean from triplicate assay in a single experiment.
b5-HT4 receptor agonistic activities; contractile effects in guinea pigs ascending colon.
cEC50 values were determined by linear regression.
dIntestinal absorption rate.
good oral bioavailability, very high and selective bind-
ing affinity for the 5-HT4 receptor, and potent 5-HT4
receptor agonistic activity. This paper outlines the pre-
paration of aralkylamino (or alkylamino), benzoyl, and
phenylsulfonyl derivatives, as well as pharmacological
profiles for gastrointestinal motility and pharmaco-
kinetic properties of compounds.
di-tert-butyldicarbonate followed by aqueous KHSO4
to deprotect the benzylidene group to give 4-amino-
methyl-1-(tert-butoxycarbonyl) piperidine (2).11 The
compound 2 was condensed with 4-amino-5-chloro-2-
methoxybenzoic acid using 1-ethyl-3-[3-(dimethylamino)-
propyl]carbodiimide hydrochloride (EDC) and 1-
hydroxybenzotriazole (HOBT). The resulting 3 was
treated with HCl-dioxane to afford 4-amino-5-chloro-2-
methoxy-N-(piperidin-4-ylmethyl)benzamide
hydro-
chloride (4). The compound 4 was converted into 6 by
treating with 5. Compound 7 was obtained from com-
pound 6 by treating with hydrazine hydrate for the
removal of the phthaloyl group. The secondary (or ter-
tiary) amine derivatives 8a–k were prepared by reductive
amination of the primary amine derivative 7 with the
corresponding aldehydes using NaBH4 or NaBH3CN.
Chemistry
The preparation of all compounds described in this
paper is illustrated in Schemes 1 and 2. 4-(Amino-
methyl)piperidine was treated with benzaldehyde for
protection of the primary amino group at the 4-position
to give a benzylidene derivative, which was reacted with

4228
S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
Table 2. Pharmacological data of aralkylamino and alkylamino derivatives 8a–k
Compd
n
R1
R2
Binding affinitiesa
Potencyb
% Absorptiond
5-HT4, Ki (nM)
5-HT3, IC50 (nM)
EC50 (nM)c
8a
8b
8c
8d
8e
8f
8g
8h
8i
4
5
5
5
5
5
5
5
5
5
5
3,4-Dichlorobenzyl
Benzyl
Benzyl
Et
H
Me
Propyl
Et
H
Ethyl
Et
Et
H
Benzyl
0.62
1.7
3.5
0.84
1.2
2.8
2.2
2.7
1.6
0.51
1.1
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
No effect
3.7
29.9
12.0
9.2
22.39.3
62.5
54.0
6.7
9.0
47.5
26.5
24.8
33.6
8.3
Benzyl
4-Methoxybenzyl
2-Thienylmethyl
2-Thienylmethyl
1-Naphthylmethyl
2-Naphthylmethyl
Cylohexylmethyl
Cylohexylmethyl
57.9
28.3
21.6
NTe
NTe
8j
8k
No effect
aEach value is the mean from triplicate assay in a single experiment.
b5-HT4 receptor agonistic activities; contractile effects in guinea pigs ascending colon.
cEC50 values were determined by linear regression.
dIntestinal absorption rate.
eNT, not tested.
Table 3. Pharmacological data of benzoyl and phenylsulfonyl derivatives
Compd
n
R1
Binding affinitiesa
5-HT4, Ki (nM) 5-HT3, IC50 (nM)
Potencyb
EC50(nM)c
9.1
% Absorptiond
11a
11b
11c
11d
12a
12b
12c
4
5
6
5
4
5
6
4.0
2.4
1.5
0.34
3.0
1.4
3.4
>1000
>1000
>1000
>1000
>1000
>1000
>1000
87.0
10.0
84.9
61.6
6.5
4.326.0
12.0
71.6
45.1
6.360.4
6.5
aEach value is the mean from triplicate assay in a single experiment.
b5-HT4 receptor agonistic activities; contractile effects in guinea pigs ascending colon.
cEC50 values were determined by linear regression.
dIntestinal absorption rate.

S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
4229
Result and discussion
Synthesized compounds were evaluated for 5-HT4
receptor binding affinity by [3H]GR-113808 binding
assay in guinea-pig striatum membranes12 and for in
vitro 5-HT4 receptor agonistic activities (EC50 values )
by their ability to contract isolated guinea-pig ascending
colon.13 The compounds were evaluated for 5-HT3
receptor binding affinity in rat cerebrocortical mem-
branes by [3H] Granisetron binding.13 The bioavail-
abilities were determined using male dogs.9,14 The
intestinal absorption rate (% absorption) was evaluated
using the in situ rat loop method in which the absorp-
tion rate of compounds was measured 2 h after the
injection. 1-Methylindol-3-yl carbonylamino derivative
1a showed low oral bioavailability (5.1%) in dogs.9 In
our previous report, when compound 1a was given
intravenously to dogs at 0.01 mg/kg, the motility of
gastric antrum and ascending colon was clearly
enhanced,8 suggesting that compound 1a was rela-
tively stable in plasma. The intestinal absorption rate
of 1a (6.9%) was much lower than that of cisapride
(97.0%) as shown in Table 1. Therefore, low oral
bioavailability of 1a would be due to its poor intest-
inal absorption rate rather than susceptibility to
metabolism. Accordingly, the pharmocokinetic eval-
uation of compounds started with measuring intestinal
Figure 2. Relationship between logP and % absorption of com-
pounds. Physical properties and spectral data of the compounds are
given in Tables 4–6.
The benzoyl type compounds 11a–d and phenylsulfonyl
type compounds 12a–c were synthesized by coupling
reaction of compound 4 with compounds 9a–d and 10a–
c in K2CO3–DMF, respectively. The compounds 9a–c
were prepared by Friedel–Craft reaction. The indole
derivative 9d were prepared using phosphorus oxy-
chloride by the Vilsmeier–Haack reaction of 1-methyl-
indole with 6-bromoheptyl-N,N-dimethylamide. The
compounds 10a–c were synthesized by coupling of ben-
zenethiol and bromochloroalkanes followed by oxidation
with m-CPBA or H2O2/HCOOH.
Table 4. Physical data of compounds
Compd
Formula
Elemental analysis (%) found (calculated)
C
H
N
MP (ꢀC)
201–202
.
8a
C27H37N4O2Cl3
51.56
(51.47
64.98
(64.78
63.47
(63.75
66.15
(66.45
54.06
(54.35
56.14
(56.45
50.66
(50.45
65.237.87
(65.45
68.97
(68.837.91
53.18
(53.21
67.61
(67.49
63.58
(63.56
65.06
(65.16
63.88
(63.77
65.98
(65.88
53.57
(53.43
55.56
(53.38
51.96
(51.74
5.94
5.9
8.00
8.00)
11.60
11.62)
11.01
11.01)
10.67
10.69)
7.39
.
3/2C2H2O4 1/2H2O
.
C26H37N4O2Cl 1/2H2O
8b
7.77
7.95
7.92
8.20
8.48
8.46
6.88
6.77
7.2310.76
7.6
6.28
6.49
9.65
8.07
7.81
10.03
7.16
7.29
8.69
8.75
7.10
7.17
7.19
7.32
7.49
7.63
7.20
7.1310.60)
6.39
6.35
6.61
6.63
6.836.88
7.01
Amorphous
Amorphous
Amorphous
Amorphous
.
C27H39N4O2Cl 6/5H2O
8c
.
8d
C29H43N4O2Cl
1/2H2O
C29H43N4O3Cl
.
8e
.
2C2H2O4 H2O
C24H35N4O2SCl
7/4H2O
7.68)
.
.
8f
67–69
10.97)
7.93104–108
7.84)
8g
C26H39N4O2SCl
.
2C2H2O4 3/2H2O
.
C32H43N4O2Cl 2H2O
8h
Amorphous
9.54)
10.04
8i
C32H43N4O2Cl2/5H2O
Amorphous
125–129
)
.
8j
C26H43N4O2Cl
8.26
8.27)
9.44
9.54)
8.84
8.89)
9.05
8.77)
8.29
.
2C2H2O4 H2O
.
8k
11a
11b
11c
11d
12a
12b
12c
C33H49N4O2Cl
H2O
C25H32N3O3Cl
4/5H2O
C26H34N3O3Cl
2/5H2O
C27H36N3O3Cl
5/4H2O
Amorphous
100–102
.
.
.
.
138–141
128–130
8.26)
10.56
C29H37N4O3Cl
1/5H2O
C24H32N3O4SCl
125–127
.
.
.
7.71
7.79)
7.51
7.47)
201–202
.
HCl 1/2H2O
C25H34N3O4SCl
.
HCl H2O
C26H36N3O4SCl
114–117
90–92
.
HCl 5/2H2O
6.96)

4230
S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
Table 5. 1H NMR spectra.l data for alkylamino and aralkylamino type compounds 8a–k
Compd
No.
d (ppm)
8a
(DMSO-d6)
1.07 (3H, t, J=7.3 Hz), 1.35–1.90 (9H, m), 2.55–3.50 (10H, m), 3.83 (3H, s), 5.75–6.05 (2H, broad), 6.50 (1H, s),
7.42 (1H, d, J=7.9 Hz), 7.62–7.68 (2H, m), 7.67 (1H, s), 8.00–8.02 (1H, m).
8b
(CDCl3)
1.25–2.00 (14H, m), 2.30 (2H, t, J=7.7 Hz,), 2.62 (2H, t, J=7.3 Hz), 2.85–2.98 (2H, m), 3.32 (2H, t, J=5.7 Hz),
3.78 (2H, s), 3.88 (3H, s), 4.44 (2H, brS), 6.29 (1H, s), 7.20–7.36 (5H, m), 7.73–7.77 (1H, m), 8.09 (1H, s).
8c
(CDCl3)
1.24–2.05 (12H, m), 2.17 (2H, s), 2.25–2.40 (4H, m), 2.90–3.01 (2H, t, J=5.7 Hz), 3.32 (2H, t, J=6.1 Hz),
3.47 (4H, d, J=3.3 Hz), 3.89 (3H, s), 4.38 (2H, brs), 6.29 (1H, s), 7.19–7.74 (5H, m), 7.72–7.74 (1H, m), 8.10 (1H, s).
8d
(CDCl3)
0.85 (3H, t, J=3.5 Hz), 1.20–1.75 (13H, m), 1.85–1.96 (2H, m), 2.24–2.43 (6H, m), 2.88–2.96 (2H, m),
3.31 (2H, t, J=6.6 Hz), 3.53 (2H, s), 3.87 (3H, s), 4.51 (2H, s), 6.30 (1H, s), 7.16–7.35 (5H, m), 7.71–7.79 (1H, m),
8.09 (1H, s).
8e
(DMSO-d6)
1.06 (3H, t, J=6.9 Hz), 1.15–1.90 (11H, m), 2.70–3.50 (12H, m), 3.78 (3H, s), 3.83 (3H, s), 4.17 (2H, s),
5.70–6.15 (2H, broad), 6.50 (1H, s), 6.99 (2H, d, J=8.6 Hz), 7.45 (2H, d, J=8.6 Hz), 7.66 (1H, s), 8.00–8.03(1H, m).
8f
(CDCl3)
1.25 (11H, m), 2.01–2.16 (2H, m), 2.35–2.45 (2H, m), 2.65 (2H, t, J=6.9 Hz), 2.85–3.10 (2H, m),
3.32 (2H, t, J=6.2 Hz), 3.89 (3H, s), 3.97 (3H, s), 4.45 (2H, brs), 6.31 (1H, s), 6.90–7.30 (3H, m), 7.77–7.87 (1H, m),
8.09 (1H, s).
8g
(CDCl3)
1.16 (3H, t, J=7.3 Hz), 1.20–1.90 (11H, m), 2.70–3.50 (12H, m), 3.82 (3H, s), 4.29 (2H, s), 5.80–6.10 (2H, broad),
6.49 (1H, s), 7.07–7.09 (1H, m), 7.24–7.25 (1H, m), 7.60–7.62 (1H, m), 7.66 (1H, s), 8.00–8.02 (1H, m).
8h
(CDCl3)
1.05 (3H, t, J=7.3Hz,), 1.18–2.02 (13H, m), 2.22 (2H, t, J=7.9 Hz), 2.45 (2H, t, J=7.9 Hz), 2.55 (2H, q, J=7.2 Hz),
2.81–2.94 (2H, m), 3.31 (2H, t, J=6.3 Hz), 3.87 (3H, s), 3.95 (2H, s), 4.39 (2H, brs), 6.27 (1H, s), 7.38–7.55 (4H, m),
7.70–7.87 (3H, m), 8.10 (1H, s), 8.30–8.34 (1H, m).
8i
(CDCl3)
1.05 (3H, t, J=7.0 Hz,), 1.19–1.95 (15H, m), 2.26 (2H, t, J=7.9 Hz), 2.45 (2H, t, J=7.9 Hz), 2.55 (2H, q, J=7.2 Hz),
2.81–2.94 (2H, m), 3.31 (2H, t, J=6.3 Hz), 3.87 (3H, s), 4.39 (2H, brs), 6.27 (1H, s), 7.42–7.52 (3H, m),
7.70–7.84 (5H, m), 8.11 (1H, s).
8j
(DMSO-d6)
0.86–1.90 (21H, m), 2.60–3.50 (14H, m), 3.82 (3H, s), 5.92 (2H, brs), 6.48 (1H, s), 7.66 (1H, s), 7.96–8.01 (1H, m).
8k
(CDCl3)
0.72–0.84 (2H, m), 1.09–1.80 (20H, m), 2.01–2.13(2H, m), 2.16 (2H, d, J=7.3Hz), 2.32 (2H, t, J=7.3Hz),
2.36–2.44 (2H, m), 2.99–3.07 (2H, m), 3.32 (2H, t, J=5.9 Hz), 3.48 (2H, s), 3.90 (3H, s), 4.40 (2H, s), 6.30 (1H, s),
7.17–7.33 (5H, m), 7.73–7.80 (1H, m), 8.10 (1H, s).
absorption rate of analogues. [The compound having a
benzene group instead of the 1-methylindole, 1b8
showed low intestinal absorption rate (18.9%) as shown
in Table 1].
show 5-HT4 receptor agonistic activity. The tertiary
amine type, ethyl(2-thienylmethyl)amino derivative 8g
possessed better intestinal absorption rate (57.9%) than
the secondary amine type, (2-thienylmethyl)amino
derivative 8f (9.3%).
The aralkylamine and alkylamine type compounds 8a–k
had high and selective 5-HT4 receptor binding affinity as
shown in Table 2. The secondary amine type, benzyl-
amino derivative 8b revealed potent 5-HT4 receptor
agonistic activity (EC50=3.7 nM), although it showed
poor intestinal absorption rate (26.5%). The tertiary
amine type, benzyl(methyl)amino derivative 8c and
benzyl(propyl)amino derivative 8d also showed poor
intestinal absorption rate (24.8 and 33.6%, respec-
tively). A decrease in intestinal absorption rate was
observed when a methoxy group was at the 4-position
of the benzene ring (compound 8e, 8.3%). Introduction
of chlorine atoms into the 3,4-positions of the benzene
ring resulted in a moderately increased intestinal
absorption rate (8a, 47.5%). However, compound 8a
did not show 5-HT4 receptor agonistic activity. The
compound 8k with benzyl(cyclohexylmethyl)amino
group showed no 5-HT4 receptor agonistic activity. On
the other hand, the benzylamino derivative 8b and
cyclohexylmethylamino derivative 8j possessed 5-HT4
receptor agonistic activities. These results indicated that
the compounds with a bulky substituent group (such as
(3,4-dichlorobenzyl)ethylamino or benzyl(cyclohexyl-
methyl)amino) on the terminal of the straight alkyl
chain at the 1-position on the piperidine ring did not
A correlation between the logP value and the %
absorption of aralkylamine derivatives was low as
shown in Figure 2. Although naphthylmethyl deriva-
tives 8h (logP=1.8) and 8i (logP=1.4) were more lipo-
philic than 8a–g (logP=À0.9–1.1, octanol/pH 6.8
buffer), a modest increase in lipophilicity was not
sufficient to confer improved intestinal absorption rate
(Fig. 2). In the aralkylamine and alkylamine derivatives
series 8a–k, the potency (EC50=3.7–62.5 nM) of tested
compounds was weaker than that of compounds 1a
(EC50=1.2 nM) and 1b (EC50=3.7 nM). Furthermore,
the intestinal absorption rates of tested compounds
were also as poor as those of compound 1a and 1b
(Table 1).
The benzoyl type compounds 11a–d and phenylsulfonyl
type compounds 12a–c also showed selective and high 5-
HT4 receptor binding affinity as depicted in Table 3.
Compounds 11a–d and 12a–c revealed potent 5-HT4
receptor agonistic activities. These results suggested that
the benzoyl and phenylsulfonyl moiety at its side-chain
part at the 1-position on the piperidine ring played an
important role in enhancing 5-HT4 receptor agonistic
activity as well as the 1-methylindol-3-yl carbonylamino

S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
Table 6. 1H NMR for benzoyl and phenylsulfonyl type compounds 11a–d and 12a–c
4231
Compd No.
d (ppm)
11a
(DMSO-d6)
1.05–1.22 (2H, m), 1.37–1.70 (7H, m), 1.72–1.83 (2H, m), 2..25–2.38 (2H, m), 2.82–2.90 (2H, m), 3.02 (2H, t, J=7.3Hz),
3.14 (2H, t, J=6.2 Hz), 3.82 (3H, s), 5.90 (2H, brs), 6.47 (1H, s), 7.52-7.54 (2H, m), 7.60–7.64 (2H, m), 7.84–7.88 (1H, m),
7.95–7.98 (2H, m).
11b
(CDCl3)
1.30–1.94 (13H, m), 2.27–2.34 (2H, m), 2.90–2.99 (4H, m), 3.32 (2H, t, J=6.3 Hz), 3.89 (3H, s), 4.38 (2H, brs), 6.29 (1H, s),
7.46–7.58 (3H, m), 7.65–7.80 (1H, m), 7.93–7.95 (2H, m), 8.11 (1H, s).
11c
(CDCl3)
1.28–1.81 (13H, m), 1.90–2.10 (2H, m), 2.25–2.41 (2H, m), 2.88–3.02 (4H, m), 3.32 (2H, t, J=6.3 Hz), 3.89 (3H, s),
4.38 (2H, brs), 6.29 (1H, s), 7.42–7.54 (3H, m), 7.70–7.81 (1H, m), 7.95 (2H, dd, J=6.6 Hz, 2.0 Hz), 8.10 (1H, s).
11d
(DMSO-d6)
1.25–1.91 (15H, m), 2.25–2.40 (2H, m), 2.89–2.97 (4H, m), 3.32 (2H, t, J=6.6 Hz), 3.85 (3H, s), 3.89 (3H, s), 4.35 (2H, s),
6.28 (1H, s), 7.24–7.35 (3H, m), 7.71 (1H, s), 7.77–7.79 (1H, m), 8.11 (1H, s), 8.37–8.39 (1H, m).
12a
(DMSO-d6)
1.39–2.00 (13H, m), 2.67–2.89 (2H, m), 2.94-3.00 (2H, m), 3.01–3.11 (2H, m), 3.82 (3H, s), 5.93 (2H, brs), 6.48 (1H, s),
7.60–7.67 (3H, m), 7.76–7.78 (1H, m), 7.87–7.89 (2H, m), 7.90–7.99 (1H, m), 9.98–10.03 (1H, m).
12b
(CDCl3)
1.30–1.85 (11H, m), 2.75–350 (10H, m), 3.82 (3H, s), 5.93 (2H, bs), 6.48 (1H, s), 7.64 (1H, s), 7.65–7.94 (5H,m),
7.95–8.03(1H, m), 9.65–9.90 (1H, broad).
12c
(DMSO-d6)
1.14–1.99 (13H, m), 2.77–2.99 (4H, m), 3.16–3.24 (2H, m), 3.25–350 (4H, m), 3.82 (3H, s), 3.90–4.01 (2H, brs), 6.50 (1H, s),
7.64 (1H, s) 7.68 (2H, d, J=7.7 Hz), 7.73–7.75 (1H, m), 7.90 (2H, d, J=7.7 Hz), 8.00 (1Hm).
moiety of compound 1a.8 These compounds except for
11d showed better intestinal absorption rate than 1a, 1b
and aralkyl (or alkyl) amine derivative 8a–k. Conver-
sion of the benzoyl moiety in 11b to the 1-methylindole
carbonyl (11d) resulted in considerably increased 5-HT4
receptor binding affinity (Ki=0.34 nM) and agonistic
activity (EC50=4.3nM) but decreased the intestinal
absorption rate (26.0%). This result suggested that the
compounds with 1-methylindole moiety showed lower
intestinal absorption rate than the benzene derivatives
(1a vs 1b; 11d vs 11b, Tables 1 and 3). The benzoyl type
compounds possessed better intestinal absorption rate
than phenylsulfonyl derivatives (11a vs 12a; 11b vs 12b;
11c vs 12c, Table 3). As exemplified by compounds 11a–
c, the length of alkyl linker (n=4, 5 and 6) appeared to
contribute to intestinal absorption rates (87.0, 84.9 and
61.6%, respectively).
Conclusion
We described the preparation, biological evaluation and
pharmacokinetic profile of a series of 4-amino-5-chloro-
2-methoxy-N-(piperidin-4-ylmethyl)benzamides as selec-
tive 5-HT4 receptor agonists. Among them, 4-amino-5-
chloro-2-methoxy-N-[1-(6-oxo-6-phenylhexyl)piperidin-
4-yl methyl]benzamide (11b) and its related compounds
with a benzoyl or phenylsulfonyl moiety at its side chain
part at the 1-position on the piperidine ring showed
improved intestinal absorption rate compared to com-
pound 1a. In our studies, replacement of the 1-methyl-
indol-3-yl carbonylamino moiety of compound 1a with
a benzoyl or phenylsulfonyl moiety significantly
improved oral bioavailability.
Experimental
In also the phenylsulfonyl type derivatives such as 12a–
c, a variety of the length of alkyl linker (n=4–6) affected
to intestinal absorption rates (71.6, 60.4 and 45.1%,
respectively). In benzoyl derivatives and phenylsulfonyl
derivatives, lipophilicity could be increased by manipu-
lation of the straight alkyl side chain at the 1-position
on the piperidine ring. The relationship with intestinal
absorption rate and lipophilicity of compounds was
illustrated in Figure 2. In benzoyl derivatives 11a–c,
intestinal absorption rate correlated with logP.
Melting points were determined in open capillaries and
are uncorrected. Proton nuclear magnetic resonance (1H
NMR) spectra were recorded on JEOL JNM-EX270
spectrometer. Coupling constants are reported in hertz
(Hz) and chemical shifts were expressed in ppm down-
field from tetramethylsilane as an internal standard.
Mass spectra (MS) were obtained by a JMS-O1SG
spectrometer. Elementary analysis was performed for C,
H and N by our laboratory.
The carbonyl and sulfonyl groups were good bioiso-
steres of the amide bond in order to improve intestinal
absorption rate and to show potent 5-HT4 receptor
agonistic effect. Especially, 4-amino-5-chloro-2-methoxy-
N-[1-(6-oxo-6-phenylhexyl)piperidin-4-ylmethyl]benz-
amide (11b) possessed good intestinal absorption rate
(84.9%). As might be expected, compound 11b showed
sufficient oral bioavailability in dogs (54%) because of
its good intestinal absorption rate. Compound 11b also
showed high and selective 5-HT4 receptor binding affi-
nity (Ki=2.4 nM), and potent agonistic activity
(EC50=10 nM).
Chemistry
4-Aminomethyl-1-(tert-butoxycarbony) piperidine (2). 4-
(Aminomethyl)piperidine (137 g, 1.20 mol) was refluxed
with benzaldehyde (127 g, 1.20 mol) in toluene (1.0 L)
with azeotropic removal of water for 12 h. To the reac-
tion mixture was added di-tert-butyldicarbonate (288 g,
1.32 mol) below 10 ꢀC, then stirring continued at room
temperature for 6 h. The reaction mixture was reacted
with aqueous KHSO4 at room temperature for 12 h. It
was washed with diisopropylether and extracted with

4232
S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
chloroform. The extracts were washed with brine and
dried over MgSO4. The solvent was removed in vacuo
to give 2 (273g, 92%).
mL) in EtOH (250 mL) for 6 h. After cooling, insoluble
compound was removed by filtration and solution was
evaporated in vacuo. The residue was purified by silica
gel column chromatography (CHCl3/MeOH/NH4OH)
1H NMR (CDCl3) d: 1.01–1.30 (5H, m), 1.45 (9H, s),
1.65–1.75 (2H, m), 2.58 (2H, d, J=6.6 Hz), 2.60–2.85
(2H, m), 4.07–4.16 (2H, m); MS EI-MS m/z: 214 (M+).
to give compound 6 (17.0 g, 88%). H NMR(CDCl3) d
1
1.23–1.74 (9H, m), 1.76–1.90 (2H, m), 1.95–2.10 (2H,
m), 2.30–2.41 (2H, m), 2.72 (2H, d, J=7.3Hz), 2.90–
3.05 (2H, m), 3.31 (2H, t, J=6.6 Hz), 3.91 (3H, s), 4.42
(2H, s), 4.69–4.85 (2H, m), 6.37 (1H, s), 7.84–7.97 (1H,
m), 8.00 (1H, s); EI–MS m/z: 382 (M+). Anal. calcd for
4-Amino-N-[1-(tert-butoxycarbonyl)piperidin-4-ylmethyl]-
5-chloro-2-methoxybenzamide (3). To the mixture of 2
(100 g, 467 mmol), 4-amino-5-chloro-2-methoxybenzoic
acid (94.1 g, 467 mmol) and Et3N (67.9 mL, 467 mmol)
in DMF (1000 mL) were added 1-ethyl-3-[3-(dimethyl-
amino)propyl]carbodiimide hydrochloride (EDC) (93.9
g, 467 mmol) and 1-hydroxybenzotriazole (HOBT)
(66.2 g, 467 mmol) at 5 ꢀC. The reaction mixture was
stirred at 25 ꢀC for 24 h and concentrated in vacuo. The
resulting residue was extracted with AcOEt. The com-
bined organic layers were washed with aqueous K2CO3
and dried over MgSO4. The solvent was removed in
vacuo to give 3 (170 g, 92%).
.
C19H31N4O2Cl H2O/C, 56.92; H, 8.30; N, 13.97. Found
C, 56.84; H, 8.42; N, 13.78.
General procedure for preparation of amine type benz-
amide derivatives (8a–k). The compound was stirred
with aldehyde at 60–65 ꢀC in EtOH for 2 h. To the
reaction mixture which was cooled to 5 ꢀC was added
NaBH4 and stirred at 25 ꢀC for 2–4 h. The reaction
mixture was poured into ice water, extracted with
CHCl3 and washed with saturated brine. The organic
layer was dried over MgSO4 and concentrated in vac-
cuo. The residue was purified by silica gel column chro-
matography (CHCl3/MeOH). The resulting oil was
transformed into oxalate and recrystallized from AcOEt
to give colorless crystals.
1H NMR (CDCl3) d 1.01–1.27 (5H, m), 1.45 (9H, s),
1.65–1.85 (2H, m), 2.62–2.75 (2H, m), 3.27–3.37 (2H,
m), 3.88 (3H, s), 4.61 (2H, s), 6.34 (1H, s), 7.72–7.82
(1H, m), 8.07 (1H, s); EI-MS m/z: 397 (M+).
General procedure for preparation of derivatives (9a–c)
4-Amino-5-chloro-2-methoxy-N-(piperidin-4-ylmethyl)-
benzamide hydrochloride (4). To the mixture of 3 (170
g, 430 mmol) in dioxane was added 10%HCl/dioxane
portionwise at 5 ꢀC. The reaction mixture was stirred at
25 ꢀC for 4 h and evaporated in vacuo to afford pale
yellow solid. The resulting solid was recrystallized from
A cooled (5 ꢀC) solution of benzene and 6-chlorohex-
anoyl chloride in CH2Cl2 was treated with AlCl3 and
stirred at 25 ꢀC for 2–4 h. The reaction mixture was
poured into ice water and extracted with CH2Cl2. The
organic layer was concentrated in vaccuo and purified
by silica gel column chromatography (CHCl3/MeOH)
to give a colorless oil.
1
IPE-EtOH to give 4 (118.5 g, 82%). H NMR (DMSO-
d6) d 1.30–1.50 (2H, m), 1.65–1.90 (3H, m), 2.60–2.90
(2H, m), 3.12–3.29 (4H, m), 3.83 (3H, s), 4.94 (3H,
broad), 6.53(1H, s), 7.66 (1H, s), 7.96–8.03(1H, m),
6-Bromo-1-(1-methyl-1H-indol-3-yl)-1-hexanone (9d). A
mixture of 1-methyl-1H-indole (1.5 g, 11 mmol), N,N-
dimethyl-5-bromopentylamide (2.5 g, 11 mmol), POCl3
(1.2 mL, 13mmol) in CHCl 3 (20 mL) was refluxed for 7
h. The reaction mixture was poured into ice and was
extracted with CHCl3. The extract was washed with
brine and was dried over MgSO4. After evaporation, the
residue was chromatographed on silica gel (CHCl3) to
give 9d (1.3g, 38%); 1H NMR (CDCl3) d 1.48–1.60
(2H, m), 1.75–2.00 (4H, m), 2.83(2H, t, J=7.7 Hz), 3.50
(2H, t, J=7.7 Hz), 3.82 (3H, s), 7.27–7.32 (3H, m), 7.69
(1H, s), 8.35–8.39 (1H, m), EI-MS m/z: 309 (M++2),
307 (M+).
8.80–9.00 (1H, broad), 9.10–9.25 (1H, broad); EI–MS
+
.
.
m/z: 297 (M ). Anal. calcd for C14H20N3O2Cl 2HCl 3/
4H2O: C, 43.77; H, 6.16 ; N, 10.94. Found C, 43.73; H,
6.23; N, 10.94.
4-Amino-5-chloro-N-[1-[5-(2,3-dihydro-1,3-dioxo-1H-iso-
indole-2-yl)pentyl]piperidin-4-yl methyl]-2-methoxybenz-
amide (6). The compound 4 (67.7 g, 203mmol) was
stirred with N-(5-bromopentyl)phthalimide (60.0 g, 203
mmol) at 70–75 ꢀC in K2CO3 (84.0 g, 609 mmol)/DMF
(700 mL) for 12 h. The reaction mixture was evaporated
and extracted with AcOEt. The extracts were washed
with aqueous K2CO3 and dried over MgSO4 and con-
centrated in vacuo. The mixture was purified by silica
gel column chromatography (CHCl3/MeOH, 10/1) to
General procedure for preparation of derivatives (10a–c)
1
give 6 (39.0 g, 37%). H NMR (CDCl3) d 1.23–1.76
A solution of benzenethiol and bromochloroalkane in
DMF was stirred at 50–60 ꢀC for 3–4 h in the presence
of K2CO3. The reaction mixture was poured into ice
water and extracted with AcOEt. The combined organic
extracts were concentrated in vacuo and purified by
silica gel column chromatography (CHCl3/MeOH) to
give thioether compound. H2O2 was added to the
thioether compound in HCOOH, and the reaction mix-
ture was stirred at 25 ꢀC for 12 h. The reaction mixture
was poured into ice water and filtrate to afford white
solid.
(11H, m), 1.83–1.94 (2H, m), 2.30 (2H, t, J=7.3Hz),
2.87–2.95 (2H, m), 3.31 (2H, t, J=6.6 Hz), 3.68 (2H, t,
J=7.3 Hz), 3.89 (3H, s), 4.42 (2H, s), 6.30 (1H, s), 7.67–
7.86 (5H, m), 8.10 (1H, s); EI-MS m/z: 512 (M+). Anal.
.
calcd for C27H33N4O4Cl 1/2H2O: C, 62.12; H, 6.56; N,
10.73. Found C, 62.19; H, 6.63; N, 10.63.
4-Amino-N-[1-(5-aminopentyl)piperidin-4-ylmethyl]-5-
chloro-2-methoxybenzamide (7). The compound 5 (26.0
g, 50.7 mmol) was refluxed with hydrazine hydrate (3.5

S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
4233
General procedure for preparation of benzamide deriva-
tives (11a–d, 12a–c)
had been presoaked in 0.1% polyethyleneimie. Filters
were washed immediately with 3Â3mL of ice-cold
Tris–HCl buffer (50 mM, pH 7.4). ICS 205930 (100
mmmol/L) was used for the determination of nonspecific
binding.
The compound was stirred with halides (9a–d, 10a–c) at
70–75 ꢀC in K2CO3/DMF for 5–12 h. The reaction
mixture was evaporated and extracted with AcOEt. The
extracts were washed with aqueous K2CO3. The organic
layer was dried over MgSO4 and concentrated in vac-
cuo. The residue was purified by silica gel column chro-
matography (CHCl3/MeOH) and recrystallized to give
colorless crystals.
5-HT4 receptor agonistic activities in vitro contraction of
isolated guinea-pig ascending colon
Male Hartley guinea pigs (Japan SLC, Ltd., Shizuoka,
Japan) were killed by cervical dislocation and the
ascending colon (a 10-cm segment starting 1 cm from
the caecum) was removed. The longitudinal muscle
layer was separated from the underlying circular mus-
cle and divided into four segments of about 2.5 cm.
Four muscle strip preparations were individually
mounted vertically for isotonic measurement into a
tissue bath containing 10 mL Tyrode solution. Only 5-
HT was tested in the Tyrode solution with containing
methysergide (1 mmol/L) and granisetron (1 mmol/L) to
inhibit responses mediated by 5-HT2 and 5-HT1-like
and 5-HT3 receptors, respectively. This solution was
kept at 37 ꢀC and gassed with 95% O2, 5% CO2. The
strips were subjected to a preload of 1 g and allowed
to stabilize for 20 min. After stabilization, the response
of the longitudinal muscle to 30 mmol/L methacholine
was measured. Agonist concentration–effect curves
were constructed using sequential dosing, leaving 15
min between doses. A 15-min dosing cycle was
required to prevent desensitization. The agonist was
left in contact with a preparation until the response
had reached a maximum, the preparation was washed.
40 min was left between the determination of concen-
tration–effect curves. GR113808 (10 nmol/L) were incu-
bated for 10 min before repeating agonist concentration–
effect curves. After each determination of concentration–
effect curve, 30 mmol/L of methacholine was added to the
tissue bath again. All responses were expressed as a per-
centage of the mean of the two contractions induced by
30 mmol/L methacholine. The EC50 value, the concen-
tration causing 50% of the maximal response, was
determined by linear regression analysis.
Pharmacology
5-HT4 receptor binding assay. Male Hartley guinea pigs
(Japan SLC, Ltd., Shizuoka, Japan) were sacrificed by
cervical dislocation and the striatum was separated
from each brain. The striatum was homogenized in 15
volume of 50 mmol/L ice-cold HEPES buffer (pH 7.4)
with Polytron PT-10 and then centrifuged at 35,000g for
20 min. The resulting pellet was resuspended in the
HEPES buffer and finally diluted to the appropriate
concentration for assay (6 mg wet weight per assay
tube). This suspension was used as the tissue prepara-
tion. Assay tube contained 50 mL of HEPES buffer or a
solution of the test agents, 50 mL solution of
[3H]GR113808 (Amersham International, UK) to give a
final concentration of 0.1 nmol/L and 900 mL of tissue
preparation. Each tube was incubated for 30 min at
37 ꢀC and the reaction was terminated by rapid fil-
tration through a Whatmann GF/B filter (presoaked in
0.01% v/v polyethyleneimine) followed by washing with
1Â4 mL of ice-cold HEPES buffer. Then the filter was
placed in 3mL of scintillator and the radioactivity was
determined by scintillation counting in a Beckman
model LS3801 scintillation counter. Non specific bind-
ing was defined in the presence of unlabelled GR113808
to give a final concentration of 1 mmol/L. The IC50
value was determined by non-linear regression of the
displacement curve, and the Ki value was calculated
according to the formula [Ki=IC50/(1+L/Kd)], where L
is the concentration of radioligand and Kd is the
dissociation constant of the radioligand.
Acknowledgements
5-HT3 receptor binding assay
[3H]Granisetron binding assay was performed according
to the method of Nelson and Thomas.15 Male Wistar
rat (Japan SLC, Ltd., Shizuoka, Japan) cerebral cortex
was homogenized in 20 volumes of 0.32 mol/L sucrose
and the centrifuged at 1000g for 10 min. The super-
natant was centrifuged at 40,000g for 15 min. The pellet
was suspended in 20 volumes of HEPES buffer (50
mmol/L, pH 7.4) and suspension was incubated at 37 ꢀC
for 10 min, was centrifuged at 40,000g for 15 min. The
pellet was washed and centrifuged (40,000g for 15 min).
The final pellet was resuspended in 30 volumes of
HEPES buffer and used as tissue homogenate. The
binding assay consisted of 50 mmmol/L of [3H]Granise-
tron, 50 mL of displacing drugs and 900 mL of tissue
homogenate. Following a 30-min incubation at 25 ꢀC,
the assay mixture was rapidly filtered under reduced
pressure through Whatman GF/B glass filters which
We thank Mr. K. Katayama, Mrs. F. Matsugaki, Mrs.
M. Miyoshi and Mrs. Y. Hattori for some of the biolo-
gical results. We also thank Mr. K. Adachi and Mr. T.
Ikebe for helpful discussion.
References and Notes
1. Dumuis, A.; Sebbon, M.; Bockaert, J. Naunyn Schmiede-
bergs Arch. Pharmacol. 1989, 340, 403.
2. Craig, D. A.; Clarke, D. E. J. Pharmcol. Exp. Ther. 1990,
252, 1378.
3. Elswood, C. J.; Bunce, K. T.; Hunphrey, P. A. Eur. J.
Pharmcol. 1993, 196, 618.
4. Harrington, R. A.; Hamilton, C. W.; Brogden, R. N.; Lin-
kewich, J. A.; Romankiewicz, J. A.; Heel, R. C. Drugs 1983,
25, 451.
5. Schuurkes, J. A.; Van Nueten, J. M.; Van Daele, P. G. H.;

4234
S. Sonda et al. / Bioorg. Med. Chem. 11 (2003) 4225–4234
Reyntjens, A. J.; Janssen, P. A. J. J. Pharmacol. Exp. Ther.
1985, 234, 775.
6. Taniyama, K.; Nakayama, S.; Takeda, K.; Matsuyama, S.;
Shinakawa, J.; Sano, I.; Tanaka, C J. Parmacol. Exp. Ther.
1991, 258, 1098.
7. Itoh, K.; Kanzaki, K.; Ikebe, T.; Kuroita, T.; Tomozane,
H.; Sonda, S.; Sato, N.; Haga, K.; Kawakita, T. Eur. J. Med.
Chem. 1999, 34, 977.
8. Itoh, K.; Tomozane, H.; Hakira, H.; Sonda, S.; Asano, K.;
Fujimura, M.; Sato, N.; Haga, K.; Kawakita, T. Eur. J. Med.
Chem. 1999, 34, 1101.
10. Szelke, M.; Leckie, B.; Hallett, A.; Jones, D. M.; Sueiras,
J.; Atrash, B.; Lever, A. F. Nature 1982, 299, 555.
11. Almquist, R. G.; Chao, W. R.; Elliss, M. F.; Johnson,
H. L. J. Med. Chem. 1980, 23, 1392.
12. Katayama, K.; Morio, Y.; Haga, K.; Hukuda, T. Folia
Pharmacol. Jpn. 1995, 105, 461.
13. Nippon Yakurigaku Zasshi 1995 Jun;105(6):461-8 Related
Articles, Books [Cisapride, a gastroprokinetic agent, binds to
5-HT4 receptors].
14. Dogs were administered with 11b at intravenous and oral
doses of 10 mg/kg. After oral administration of 11b, the
unchanged compound concentrations reached the Cmax of 0.46
9. Dogs were administered with Y-34959 at intravenous and
oral doses of 5.0 mg/kg. After oral administration of Y-34959,
the unchanged compound concentrations reached the Cmax of
.
mg/mL at 0.7 h. The AUC0À24 was 1.48 mg h/mL and the
h
bioavailability was calculated to be 54%.
15. Nelson, D. R.; Thomas, D. R. Biochem. Pharmacol. 1989,
38, 1693.
.
26 ng/mL at 0.5 h. The AUC0À24 h was 52 ng h/mL and the
bioavailability was calculated to be 5.1%.