New 5-HT1A receptor agonists possessing 1,4-benzoxazepine scaffold exhibit highly potent anti-ischemic effects.

Article abstract of DOI:10.1016/S0960-894X(01)00008-7

A series of new 3-substituted-4-(4-aminobutyl)-1,4-benzoxazepin-5(4H)-one derivatives (1-5) which showed a very high affinity for 5-HT1A receptor with good selectivity over dopamine D2 receptor was synthesized. Among these compounds, 3-chloro-4-[4-[4-(2-p

Full text of DOI:10.1016/S0960-894X(01)00008-7

Bioorganic & Medicinal Chemistry Letters 11 (2001) 595±598
New 5-HT Receptor Agonists Possessing 1,4-Benzoxazepine
Sca€old Exhibit Highly Potent Anti-Ischemic E€ects
1A
a,
a
a
a
Katsuhide Kamei, * Noriko Maeda, Ryoko Ogino, Makoto Koyama,
a
b
a
b
Mika Nakajima, Toshio Tatsuoka, Tomochika Ohno and Teruyoshi Inoue
aSuntory Biomedical Research Limited, 1-1-1, Wakayama-dai, Shimamoto-cho, Mishima-gun, Osaka 618-8503, Japan
bSuntory Limited, Mori Bldg. No. 31, 5-7-2, Kojimachi, Chiyoda-ku, Tokyo 102-8530, Japan
Received 9 November 2000;accepted 28 December 2000
AbstractÐA series of new 3-substituted-4-(4-aminobutyl)-1,4-benzoxazepin-5(4H)-one derivatives (1±5) which showed a very high
anity for 5-HT1A receptor with good selectivity over dopamine D
2
receptor was synthesized. Among these compounds, 3-chloro-
4
-[4-[4-(2-pyridinyl)-1,2,3,6-tetrahydropyridin-1-yl]butyl]-1,4-benzoxazepin-5(4H)-one (5: SUN N4057) exhibited remarkable neu-
roprotective activity in a transient middle cerebral artery occlusion (t-MCAO) model. # 2001 Elsevier Science Ltd. All rights
reserved.
Acute cerebral ischemia is one of major causes of death
and the pharmacological treatments have been focused
on limiting neuronal damage, so called neuroprotec-
over dopamine D receptor. The neuroprotective e€ect
2
in an in vivo t-MCAO model is also presented.
1
tion. Serotonin (5-HT) and its receptors are known to
play important roles in various physiological and
Chemistry
2
,3
pathophysiological processes. Of these receptors, the
-HT1A receptor subtype is generally accepted to be
5
involved in psychiatric disorders such as depression,
BZO derivatives 1±519 were prepared by the pathway
shown in Scheme 1. 3-H-BZO 1 was prepared from sal-
icylamide 6. Compound 6 was selectively O-alkylated
with 2-bromomethyl-1,3-dioxolane in the presence of
K CO and followed by cyclization with 10% HCl and
4
5
6
anxiety, and psychosis. It has been reported that 5-
HT1A receptor agonists have protective e€ects on the
7À12
brain in cerebral ischemic conditions,
3
due to hyper-
polarization of cell membrane and glutamate release
2
3
1
dehydration with methanesulfonyl chloride and
triethylamine, giving 7. Compound 1 was obtained by
the alkylation of 7 with 1-bromo-4-chlorobutane and
1
4
inhibition.
Buspirone (Fig. 1) has developed as a 5-HT1A agonist
and it has been useful in the treatment of anxiety and
1
5,16
depression.
This compound, however, is not optimal
in terms of selectivity versus dopamine D receptor. It
2
has been said that dopamine D antagonists might cause
2
1
7
undesirable side e€ects such as prolactin stimulation
1
8
and extrapyramidal symptoms. In order to obtain the
high anity and selectivity for the 5-HT1A receptor, we
have attempted to modify the liposoluble unit and the
amine part of Buspirone. In this paper, we describe the
synthesis of novel 1,4-benzoxazepine (BZO) compounds
(1±5) which bind to 5-HT1A receptor with good selectivity
*
6
Corresponding author. Tel.: +81-75-962-1663;fax: +81-75-962-
448;e-mail: katsuhide_kamei@suntory.co.jp
Figure 1. Buspirone and 3-substituted-4-(4-aminobutyl)-1,4-benzox-
azepine-5(4H)-one derivatives (1±5).
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00008-7

596
K. Kamei et al. / Bioorg. Med. Chem. Lett. 11 (2001) 595±598
Scheme 1. Reagents: (a) K
imidinyl)piperazine;(f) K
2
CO
3
, 2-bromomethyl-1,3-dioxolane;(b) 10% HCl;(c) MsCl, Et
CO , BrCH COMe;(g) p-TsOH;(h) K CO , BrCH CO Et;(i) NaOEt;(j) K
3
N;(d) NaH, Br(CH
2
)
4
Cl;(e) NaI, Et
Cl;(k) POCl
3
N, 1-(2-pyr-
, PhNEt ;(l)
2
3
2
2
3
2
2
2
CO , Br(CH
3
2
)
4
3
2
0
NaI, Et
3
N, HNRR .
the subsequent amination with 1-(2-pyrimidinyl)piper-
azine. 3-Me-BZO 2 was prepared in the same way as 3-
H-BZO 1, by employing bromoacetone instead of 2-
bromomethyl-1,3-dioxolane.
otonergic receptor:21 [3H]8-OH-DPAT (8-hydroxy-2-
(N,N-di-n-propylamino)tetralin), rat hippocampus mem-
branes;(b) dopamine D 2 receptor: [ H]Raclopride, rat
striatum membranes.
2
2 3
3
-Cl-BZO 3±5 were synthesized through the inter-
We investigated the e€ects of various groups at the 3-
position of the BZO derivatives (see 1±3). The methyl-
ene chain length at the 4-position of the BZO and the
amino moiety were the same as Buspirone. As regards
the 5-HT1A receptor, 3-Cl-BZO 3 showed higher
binding anity than 3-H-BZO 1 and 3-Me-BZO 2.
3-Cl-BZO 3 showed stronger anity and more selectiv-
ity for 5-HT1A receptor than Buspirone. The amine
moiety of the side chain was of great importance for
both 5-HT1A receptor binding anity and selectivity
mediate 1,4-benzoxazepine-3,5-dione 9. Compound 9
was prepared by O-alkylation of 6 with ethyl bromo-
acetate and subsequent cyclization by treatment with
sodium ethoxide. Compounds 3±5 were prepared by N-
alkylation of 9 with 1-bromo-4-chlorobutane and fol-
lowed by chlorination with POCl in the presence of
3
N,N-diethylaniline and the subsequent amination with
0
2
the corresponding amine.
(see 3±5). Replacement to 1-(2-pyridinyl)piperazinyl
group 4 from 1-(2-pyrimidinyl)piperazinyl group 3
Results and Discussion
showed more potent binding to 5-HT1A receptor but
poorer selectivity over dopamine D receptor. The 4-(2-
Compounds 1±5 in Table 1 were evaluated for their
binding anity to 5-HT1A and dopamine D receptor
by radioligand binding assays. The speci®c ligands and
tissue sources were used as follows: (a) 5-HT1A ser-
2
pyridinyl)-1,2,3,6-tetrahydropyridinyl derivative 5,
however, exhibited the highest anity and selectivity for
2
2
3
5-HT1A receptor.

K. Kamei et al. / Bioorg. Med. Chem. Lett. 11 (2001) 595±598
597
Table 1. Structures and their receptor binding data of 3-substituted-
4(4-aminobutyl)-1,4-benzoxazepin-5(4H)-one derivatives (1±5)
increase in PTBBS levels by 21, 32* and 63%**,
respectively (*p<0.05 vs vehicle, **p<0.01 vs vehicle).
In this model, rectal temperature was found to increase
ꢀ
during ischemia to above 38.5 C, but compound 5
reduced the ischemic hyperthermia at the neuroprotec-
tive doses. It has been reported that 5-HT agonists
1
A
2
8
possess a hypothermic e€ect. In contrast, 4-dimethyl-
aminoantipyrine, an antipyretic drug, at a dose of
IC50 (nM)
200 mg/kg ip immediately after t-MCAO did not a€ect
Compda
R
1
NRR
0
5-HT1A
D
IC50 ratio
2
PTBBS levels by only 21% inhibition, although it
caused hypothermia to the same degree as 5 (1 mg/kg
sc). These results indicate that pharmacological e€ects
in addition to the hypothermic e€ect are involved in the
mechanism of neuroprotective e€ect of compound 5.
D
2
/5-HT1A
1b
H
Me
Cl
18.0
NT
1100
199
51
Ð
2b
10.2
1.59
0.77
0.47
11.0
108
125
66
In conclusion, we described the synthesis and biological
evaluation of a novel class of 1,4-benzoxazepine
derivatives 1±5 that show not only highly potent anity
for 5-HT1A receptor but also low anity for dopamine
D receptor. Since compound 5 is a potent and selective
2
5-HT1A receptor agonist compared with Buspirone and
has a desirable neuroprotective e€ect in vivo, it might be
more suitable for a therapeutic agent for ischemic neu-
ronal damage. The SARs of this series of compounds
will be reported elsewhere. Compound 5 (SUN N4057)
is currently being developed for treatment of acute
phase of cerebral infarction.
3b
4b
Cl
5c
Cl
84
179
Buspirone
55
5
aAll compounds had analytical results within 0.4% of the theoretical
values.
Acknowledgements
b
Fumarate.
c2HCl 2H
.
O.
2
The authors wish to thank Dr. Y. Hayashi for useful
suggestions for the binding assays. Thanks are also due
to Drs. F. Satoh, T. Tanaka and H. Annoura for their
encouragement throughout this work and Dr. A.
Mizuno for helpful advice during the preparation of this
manuscript.
Table 2. Neuronal protective e€ects of compounds against ischemic
brain damage in rat t-MCAO modela
Compd
(SUN N4057)
Buspirone
% inhibition
5
63%**
53%**
21%
Dimethylaminoantipyrine
References and Notes
aSee ref 27 for details. **p<0.01 versus vehicle (one-way ANOVA
followed by Dunnett's multiple comparison test).
1. Keyser, J. D.;Sulter, G.;Luiten, P. G. Trends Neurosci.
1
2
3
4
2
5
6
999, 22, 535.
. Saudou, F.;Hen, R. Med. Chem. Res. 1994, 4, 16.
. Zifa, E.;Fillion, G. Pharmacol. Rev. 1992, 44, 401.
. Blier, P.;Montigny, C. Trends Pharmacol. Sci. 1994, 15,
20.
. Hamon, M. Trends Pharmacol. Sci. 1994, 15, 36.
. Central Serotonin Receptors and Psychotropic Drugs;
Compound 5 is a 5-HT1A receptor agonist because it
inhibits forskolin-stimulated adenylate cyclase activity
in plasma membrane prepared from the rat hippo-
campus (IC =2.67 Æ 0.74 nM). Next, we investigated
5
0
in vivo neuroprotective e€ect of 3-Cl-BZO 5 in a rat
model of transient focal cerebral ischemia (Table 2).
Male Wistar rats were subjected to t-MCAO using the
Marsden, C. A., Heal, D. J., Ed.;Blackwell Scienti®c Pub-
lications: Oxford, 1992.
7. Bielenberg, G. W.;Burkhardt, M. Stroke (Supplement IV)
1990, 21, 161.
8. Prehn, J. H. M.;Welsch, M.;Backhauû, C.;Nuglisch, J.;
Ausmeier, F.;Karkoutly, C.;Krieglstein, J. Brain Res. 1993,
2
4
intraluminal suture method of Koizumi et al. The tested
compounds and vehicle (saline) were subcutaneously
administered immediately after the occlusion. The mea-
surement of peripheral type benzodiazepine binding
sites (PTBBS) in ipsilateral cortical and striatal homo-
genates was carried out as an index for quanti®cation of
6
9
30, 10.
. Globus, M. Y. T.;Wester, P.;Busto, R.;Dietrich, W. D.
Stroke 1992, 23, 1595.
0. Nakata, N.;Kogure, K.;Itoyama, Y.;Kato, H.;Ikeda,
Y.;Tanaka, Y.;Izumi, J.;Suda, H. Behav. Brain Res. 1997, 83,
1
2
5,26
neuronal damage 10 days after recirculation.
pound 5 and Buspirone reduced the increase in PTBBS
Com-
2
1
17.
1. Tracy, M.;Prous, J.;Castaner, J. Drugs Fut. 1997, 22, 225.
levels at a dose of 1 mg/kg sc.27 Single administration of
5 at doses of 0.1, 0.3 and 1 mg/kg sc immediately after t-
MCAO exerted a dose-dependent reduction of the
12. Semkova, I.;Wolz, P.;Krieglstein, J. Eur. J. Pharmacol.
1998, 359, 251.

598
K. Kamei et al. / Bioorg. Med. Chem. Lett. 11 (2001) 595±598
13. Beck, S. G.;Choi, K. C.;List, T. J. J. Pharmacol. Exp.
Ther. 1992, 263, 350.
21. Peroutka, S. J. Neurochem. 1986, 47, 529.
22. Kohler, C.;Hall, H.;Ogren, S. O.;Gawell, L.
Biochem.
14. Matsuyama, S.;Nei,K.;Tanaka, C. Brain Res. 1996, 728, 175.
15. Levy, A. D.;Van de Kar, L. D. Life Sci. 1992, 51, 83.
16. Taylor, D. P.;Moon, S. L. Neuropeptides 1991, 19, 15.
17. Awad, A. G.;Lapierre, Y. D.;Jostell, K. G. Prog. Neu-
Pharmacol. 1985, 34, 2251.
23. The binding anities to other receptors of compound 5
showed as follows;5-HT 2A: IC >1000 nM, dopamine D :
50
1
IC >1000 nM, a -adrenergic: IC =128 nM, a -adrenergic:
5
0
1
50
2
ropsychopharmacol. Biol. Psychiatry 1990, 14, 769.
8. For reviews: Montastruc, J. L.;Llau, M. E.;Rascol, O.;
Senard, J. M. Fundam. Clin. Pharmacol. 1994, 8, 293.
9. All compounds were fully characterized by spectral meth-
IC50=228 nM.
1
24. Koizumi, J.;Yoshida, Y.;Nakagawa, T.;Oneda, G. Jpn.
J. Stroke 1986, 8, 1.
25. Benavides, J.;Fage, D.;Carter, C.;Scatton, B. Brain Res.
1987, 421, 167.
1
ods, and their purity checked by elemental analysis. Repre-
sentative data of the most potent derivative: compound 5
(
26. Junck, L.;Jewett, D. M.;Kibourn, M. R.;Young, A. B.;
Kuhl, D. E. Ann. Neurol. 1990, 40(Suppl. 1), 265.
27. [ H]PK 11195 was used as a radio-labeled ligand for
.
ꢀ
2HCl 2H O);mp 133±134 C; H NMR (400 MHz, DMSO-
2
1
3
d
3
4
6
) ꢀ 1.70±1.76 (m, 2H), 1.84±1.86 (m, 2H), 2.91±2.93 (m, 2H),
.21±3.25 (m, 3H), 3.64±3.67 (m, 1H), 3.84±3.87 (m, 3H),
.03±4.08 (m, 1H), 6.74±6.76 (m, 1H), 7.14±7.17 (m, 2H), 7.34
PTBBS and its amount binding to the crude of cerebral cell
membrane was measured. The PTBBS levels of rats subjected
to t-MCAO were 4- to 6-fold higher than that of sham-oper-
(
t, 1H, J=8 Hz), 7.42±7.44 (m, 1H), 7.60 (t, 1H, J=8 Hz),
3
7
1
3
.70±7.72 (m, 1H), 7.79 (dd, 1H, J=2 and 8 Hz), 7.95±7.97 (m,
À1
ated group. Experimental data of 5: [ H]PK 11195 binding
H), 8.61±8.63 (m, 1H), 10.74 (br s, 1H);IR (KBr) cm : 3320,
015, 2600, 1644, 1612, 1513, 1455;FAB-MS m/z: 410
(pmol/g protein);saline treated ( n=14): 402 Æ 35;compound
5 treated (n=11): 197 Æ 20;sham ( n=11): 78 Æ 7;Each data
represents mean Æ S.E.M. 63% inhibition.
+
(
3 3 2 2
M +1). Anal. calcd for C23H26Cl N O 2H O: C, 53.24;H,
5
2
.83;N, 8.10. Found: C, 53.56;H, 6.04;N, 7.92.
0. 4-(2-Pyridinyl)-1,2,3,6-tetrahydropyridine was prepared as
28. Green, A. R.;Goodwin, G. M. In Brain 5-HT1A Recep-
tors;Dourish, C. T., Huston, P. H., Ahlenius, S., Ed.;Ellis
Horwood: Chichester, 1987;pp 161±176.
described in Sari, W. S., et al. J. Med. Chem. 1984, 27, 1182.