Synthesis of diaminobutane derivatives as potent Ca2+-permeable AMPA receptor antagonists

Article abstract of DOI:10.1016/S0960-894X(01)00530-3

We synthesized diaminobutane derivatives as potent Ca²⁺-permeable AMPA receptor antagonists with non-hypotensive activity. Compound 10c showed selective Ca²⁺-permeable AMPA receptor antagonist activity and neuroprotective effects in transient global ischemia models in gerbils.

Full text of DOI:10.1016/S0960-894X(01)00530-3

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2663–2666
Synthesis of Diaminobutane Derivatives as Potent
Ca2+-Permeable AMPA Receptor Antagonists
Yoshiyuki Yoneda,* Shinichi Kawajiri, Masunobu Sugimura, Ken Osanai,
Fusako Kito, Emi Ota and Tetuya Mimura
Medicinal Chemistry Research Laboratory, Daiichi Pharmaceutical Co., Ltd., 16-13, Kitakasai 1-Chome,
Edogawa-ku, Tokyo 134-8630, Japan
Received 15 May 2001; accepted 28 July 2001
Abstract—We synthesized diaminobutane derivatives as potent Ca2+-permeable AMPA receptor antagonists with non-hypotensive
activity. Compound 10c showed selective Ca2+-permeable AMPA receptor antagonist activity and neuroprotective effects in tran-
sient global ischemia models in gerbils. # 2001Elsevier Science Ltd. All rights reserved.
Introduction
result, we found that diamine compounds showed potent
Ca2+-permeable AMPA receptor antagonist activity
with non-hypotensive activity, and that one of the
diamine compounds exhibited neuroprotective effects in
an in vivo ischemia model after ip administration. In
this paper, we describe the synthesis and biological
activity of a novel series of diaminobutane derivatives.
l-Glutamate is a major excitatory neurotransmitter in
the mammalian central nervous system.1 Excessive
amounts of glutamate, which are released in pathophy-
siological conditions such as cerebral ischemia and epi-
lepsy, overstimulate glutamate receptors, leading to the
delayed neuronal cell death after global ischemia
through an increase in Ca2+ influx. We focused on
Ca2+-permeable AMPA receptors, one of the glutamate
receptors modulating Ca2+ influx,2 and studied
antagonists against the receptors.
Chemistry
The synthetic pathways of compounds 9, 10a–c, 13 and
15a–d are shown in Scheme 1. Commercially available
4-aminomethylpiperidine 4 was converted to phthali-
mide 5, which was coupled with 1-naphthylacetic acid
and then treated with hydrazine to give amine 6. Amine
6 was alkylated with 4-bromobutylphthalimide and then
protected with a tert-butoxycarbonyl (Boc) group to
give 7, the phthaloyl group of which was removed to
yield amine 8. Amine 8 was converted to compound 9
by treatment with concd HCl. Compounds 10a–c were
prepared by reductive alkylation using the correspond-
ing aldehydes and successive deprotection of the Boc
group with concd HCl. Alcohol 12 was synthesized
from amine 6 by alkylation with 4-bromobutyl acetate
followed by protection with a Boc group and alkali
hydrolysis. Alcohol 12 was mesylated with mesyl chlo-
ride, coupled with piperidine and then treated with
concd HCl to afford compound 13. Swern oxidation of
alcohol 12 gave aldehyde 14, which was converted to
compounds 15a–d by reductive alkylation using the
corresponding amines and treatment with concd HCl.
Spider toxin JSTX-3 (1) and its analogue NAS (2) are
well known Ca2+-permeable AMPA receptor antago-
nists (Fig. 1).3 However, compound 2 and its deriva-
tives5 show strong hypotensive activity which may
deteriorate cerebral blood flow in patients with cerebral
ischemia.4 In a previous paper, we investigated non-
hypotensive Ca2+-permeable AMPA receptor antago-
nists and found a series of polyamine derivatives
including 3 (Fig. 1).5 However, compound 3, as well as 1
and 2, did not exhibit neuroprotective effects in tran-
sient global ischemia models in gerbils after intraper-
itoneal (ip) or intravenous (iv) administration. We
therefore focused on an increase in lipophilicity of
newly synthesized compounds, expecting that they show
neuroprotective effects in in vivo ischemia model. As a
*Corresponding author. Tel.: +81-3-3680-0151; fax: +81-3-5696-
8344; e-mail: yoned10z@diichipharm.co.jp
0960-894X/01/$ - see front matter # 2001Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00530-3

2664
Y. Yoneda et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2663–2666
Results and Discussion
Antagonist activity of test compounds against Ca2+
-
permeable AMPA receptors (IC50) was measured using
a two-electrode voltage clamp method:6 kainate (KA)
was used as an agonist for the receptors expressed in
Xenopus oocytes by injection of rat brain mRNA.
However, since KA non-selectively induces inward cur-
rents via stimulation of KA receptors, Ca2+-permeable
and Ca2+-impermeable receptors, antagonist activity of
test compounds for Ca2+-permeable AMPA receptors
was calculated by offsetting the inward currents due to
blockade of the other two receptors.7 Hypotensive
activity was evaluated after iv administration to Wistar
rats. These activities are shown in Table 1.
Figure 1.
Compounds 21 and 25 were synthesized as shown in
Scheme 2. Ethyl isonipecotate 16 was protected with a
benzyloxycarbonyl (Z) group, and the resulting com-
pound was converted to aldehyde 18 by reduction of its
ester group and successive Swern oxidation. Wittig
reaction of 18 with 4-carboxybutyltriphenylphos-
phonium bromide followed by esterification, catalytic
reduction and condensation with 1-naphthylacetic acid
afforded 19, which was converted to aldehyde 20 in 3
steps. Aldehyde 20 was reductively alkylated with cyclo-
hexylmethylamine and then treated with concd HCl to
give compound 21. Compound 25 was synthesized from
cyclohexylaldehyde 22 in 8 steps in a similar procedure.
In this paper, we planned to modify compound 3 to
match those with higher lipophilicity. The design was
carried out based on the speculation that a decrease in
the number of basic nitrogen atoms of compound 3
would result in an increase in lipophilicity. At first, we
evaluated whether simple diaminobutane derivative 9
and monoamine derivatives 21 and 25 had Ca2+
-
permeable AMPA receptor antagonist activity. Among
these compounds, diaminobutane 9 and monoamine 21
showed considerable activity, but not monoamine 25.
Based on these results, we selected the diaminobutane
skeleton for further modification because of its easiness
and higher possibility in modification.
Alkylation of the terminal amino group of compound 9
caused improvement of potency. Cyclohexylmethylamino
derivative 10c exhibited the highest potency in the com-
pounds synthesized here. On the other hand, introduc-
tion of an oxygen atom into the terminal alkyl group
Scheme 2. Synthesis of compounds 21 and 25. Reagents and condi-
tions: (a) (i) Z-Cl, aq NaHCO3, CH2Cl2, (ii) LiBH4, EtOH (2 steps
85%); (b) Swern oxidation (quant.); (c) (i) 4-carboxybutyl triphenyl-
phosphonium bromide, NaHMDS, THF, (ii) c H2SO4, MeOH (2 steps
Scheme 1. Synthesis of compounds 9, 10a–c, 13 and 15a–d. Reagents
and conditions: (a) phthalic anhydride, Á (61%); (b) (i) (1-naphthyl)-
.
acetic acid, EDC HCl, TEA, CH2Cl2 (69%), (ii) NH2NH2 H2O, EtOH
.
.
46%), (iii) H2, Pd–C, MeOH, (iv) (1-naphthyl)acetic acid, EDC HCl,
(quant.); (c) (i) 4-bromobutylphthalimide, KF–Celite, MeCN, (ii)
.
Boc2O, CH2Cl2; (d) NH2NH2 H2O, EtOH (3 steps 55%); (e) c HCl,
HOBt, NMM, CH2Cl2 (2 steps 79%); (d) (i) LiOH, MeOH, (ii), BH3–
THF, THF (2 steps 34%), (iii) Swern oxidation (quant.); (e) (i) cyclo-
hexanemethylamine, NaBH4, MeOH, (ii) Boc2O, CH2Cl2 (2 steps
58%), (iii) c HCl, EtOH (quant); (f) (i) 4-carboxybutyl triphenylpho-
sphonium bromide, NaHMDS, (ii) c H2SO4, MeOH (2 steps 83%),
(iii) H2, Pd–C, MeOH (quant.); (g) (i) LAH, THF (85%), (ii) Swern
oxidation (quant.); (h) (i) 6, NaBH4, MeOH, (ii) Boc2O, CH2Cl2 (2
steps 89%), (iii) c HCl, EtOH (quant.).
EtOH (62%); (f) (i) aldehyde, NaBH4, MeOH, (ii) Boc2O, CH2Cl2,
(iii) c HCl, EtOH (3 steps 25–57%); (g) (i) 4-bromobutyl acetate, KF–
Celite, MeCN, (ii) Boc2O, CH2Cl2; (h) K2CO3, MeOH (3 steps 48%);
(i) (i) MsCl, Py, (ii) piperidine, KF–Celite, MeCN, (iii) c HCl, EtOH (3
steps 58%); (j) Swern oxidation (71%); (k) (i) amine, NaBH4, MeOH,
(ii) Boc2O, CH2Cl2, (iii) c HCl, EtOH (3 steps 24–50%).

Y. Yoneda et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2663–2666
2665
resulted in a slight decrease in potency (10b and 15b),
suggesting that a polar group, such as an oxygen atom,
in this position is unfavorable for interactions with the
receptor. Dimethylamino analogue 10a seemed to have
somewhat higher potency compared to those of sec-
ondary amino derivatives 15a,c,d. Piperidino deriva-
tive 13 showed potency almost comparable to that of
10c.
With respect to hypotensive activity, compound 9 with a
primary amino group showed potent activity. Mono- or
di-alkylation of the terminal amino group of 9 elimi-
nated the hypotensive activity (10a,c, 13 and 15a–d),
except for compound 10b. Contrary to non-hypotensive
compound 10c, monoamine derivatives 21 and 25 with a
similar cyclohexylalkylamino group showed strong
hypotensive activity. It is uncertain why compounds 9,
10b, 21 and 25 have hypotensive activity and also
whether all the compounds interact with the same
receptors mediating hypotensive activity.
Table 1. Inhibitory effect on the Ca2+-permeable AMPA receptor
and effect on blood pressure
Compd
X
Y
IC50
(mM)a
ÁSBP
(mmHg)c
Predicted
LogPf
b
9
54%
À53
2.44
3.21
We selected compounds 10c and 13, which showed both
potent Ca2+-permeable AMPA receptor antagonist
activity in vitro and non-hypotensive activity in vivo,
for evaluation of neuroprotective effects using a tran-
sient global ischemia models in gerbils.8 Compound 10c
was effective in the models, but not 13. A 4-time repe-
ated dose of 2.5, 5.0 or 10 mg/kg of 10c was intraper-
itoneally administered at 0, 2, 4 and 6 h after 5-min
ischemia: compound 10c significantly protected neurons
from cell death at a repeated dose of 10 mg/kg (Fig. 2).
Compound 10c had higher predicted log P than com-
pound 3 or 13, suggesting that the high lipophilicity of
10c may contribute to the in vivo neuroprotective
effects. In addition, compound 10c would not show
hypotensive effects in the gerbil transient global ische-
mia models, since the compound did not reduce blood
pressure in rats up to at least 10 mg/kg iv administra-
tion.
10a
0.53
35%b
0.25
12
10b
10c
13
À22
3.45
5.32
4.22
3.89
2.47
3.91
4.18
7.51
7.51
11 (17d)
0.39
4
9
15a
15b
15c
15d
21
0.74
44%b
0.73
À6
10
Selectivity of 10c between Ca2+-permeable and Ca2+
-
impermeable AMPA receptors was determined by using
recombinant AMPA receptors according to the descri-
bed method (Table 2).9 Ca2+-Permeable AMPA recep-
tors are assembled from only GluR3 subunits, while
Ca2+-impermeable AMPA receptors are assembled
from GluR3 and GluR2 subunits.10 Compound 10c as
0.79
7
42%b
1%b
À35
À50
well as JSTX-311 showed selective inhibition for Ca2+
permeable AMPA receptors.
-
25
3
0.19
0.69
5
3.72
1.74
NAS (2)
—
—
À55e
aIC50 (mM) was defined as the concentration of a compound that
reduces the inward current induced by kainate by 50%. The com-
pound was co-administered with kainate to Xenopus oocytes and the
inward current was measured by the two-electrode voltage clamp
method. The values represent the means of results from 6 Xenopus
oocytes.
bBecause some compounds could not exert 50% inhibition, the
amount of inhibition (%) exerted at a concentration of 1 mM is shown.
cThe systolic blood pressure (mmHg) of rats after intravenous (iv)
administration of a 3 mg/kg dose was compared with the systolic
blood pressure before administration (n=1–2).
Figure 2. Effect of compound 10c on number of surviving neurons in
the hippocampal CA1area in gerbils 5 days after global ischemia.
Values are meansÆSEM of neuronal density in individual hippocampi
in each group (n=6–20). Compound 10c was intraperitoneally admin-
istered 4 times after ischemia. *: p<0.01versus vehicle.
dThe dose of 10 mg/kg was used.
eThe dose of 1mg/kg was used.
fThe values were calculated by PROLOGP(ver. 3.0).12

2666
Y. Yoneda et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2663–2666
Table 2. Blocking effects of JSTX-3 and 10c on Ca2+-permeable
(GluR3 only) or Ca2+-impermeable AMPA receptors (GluR3+
GluR2)
J. A.; Grooms, S.; Kessler, J. A.; Bennett, M. L.; Zukin, R. S.;
Rosenbaum, D. M. Proc. Natl. Acad. Sci. U.S.A. 1998, 95,
7115. (c) Oguro, K.; Oguro, N.; Kojima, T.; Grooms, S. Y.;
Calderone, A.; Zheng, X.; Bennett, M. V.; Zukin, R. S. J.
Neurosci. 1999, 19, 9218.
3. (a) Iino, M.; Koike, M.; Isa, T.; Ozawa, S. J. Physiol. 1996,
496, 2 431. (b) Savide, J. R.; Bristow, D. R. Eur. J. Pharmacol.
1998, 351, 131. (c) Koike, M.; Iino, M.; Ozawa, S. Neurosci.
Res. 1997, 29, 27.
Sample
%inhibitiona
(GluR3 only)
%inhibitiona
(GluR3+GluR2)
JSTX-3 (1)
10c
79.7
65.8
4.3
4.4
aXenopous oocytes that had been injected with GluR3cRNA (50 ng)
have Ca2+-permeable AMPA receptors. On the other hand, Xenopous
oocytes that had been injected with both GluR3cRNA and
GluR2cRNA (1:9, total 100 ng) had Ca2+-impermeable AMPA
receptors. Both of these compounds strongly decreased the inward
current via Ca2+-permeable AMPA receptors but had little effect on
the current via Ca2+-impermeable AMPA receptors. JSTX-3 or 10c
was administered at the dose of 3 mM with 300 mM KA. Each value is
the mean of results from 2–3 Xenopous oocytes.
4. Ahmde, N.; Nasman, P.; Wahlgren, N. G. Stroke 2000, 31,
1250.
5. Yoneda, Y.; Kawajiri, S.; Hasegawa, A.; Kito, F.; Katano,
S.; Takano, E.; Mimura, T. Bioorg. Med. Chem. Lett. 2001,
11, 1261.
6. Debon, M. W.; Le Guren, J.; Canton, T.; Doble, A.; Pra-
dier, L. Eur. J. Pharmacol. 1993, 235, 283.
7. After measurement of inward currents induced by applica-
tion of 300 mM KA (current A), co-application of 300 mM KA
with 0.01–3.0 mM test compounds (current B), and co-appli-
cation of 300 mM KA with 3.0 mM JSTX-3 (current C), inhi-
bition (%) was calculated by a numerical formula of (current
AÀcurrent B)/(current AÀcurrent C)Â100.
Consequently, diamine derivative 10c showed selective
antagonist activity for Ca2+-permeable AMPA recep-
tors, non-hypotensive effects in rats and neuroprotective
effects in transient global ischemia models in gerbils.
8. (a) Kirino, T. Brain Res. 1982, 239, 57. (b) Suzuki, R.;
Yamaguchi, T.; Kirino, T. Acta. Neuropathol. 1983, 60, 207.
9. Blaschke, M.; Keller, B. U.; Rivosecchi, R.; Hollmann, M.;
Heinemann, S.; Konnerth, A. Proc. Natl. Acad. Sci. U.S.A.
1993, 90, 6528.
References and Notes
10. Hollmann, M.; Hartley, M.; Heinemann, S. Science 1991,
252, 851.
11. Blaschke, M.; Keller, B. U.; Rivosecchi, R.; Hollmann,
M.; Heinemann, S.; Konnerth, A. Proc. Natl. Acad. Sci.
U.S.A. 1993, 90, 6528.
1. Monaghan, D. T.; Bridges, R. J.; Cotoman, C. W. Annu.
Rev. Pharmacol. Toxicol. 1989, 29, 365.
2. (a) Gorter, J. A.; Petrozzino, J. J.; Aronica, E. M.; Rosen-
baum, D. M.; Opitz, T.; Bennett, M. V.; Connor, J. A.; Zukin,
R. S. J. Neurosci. 1997, 17, 6179. (b) Aronica, E. M.; Gorter,
12. CompuDrug International, Inc. (Hungary).