Discovery of a novel potent Na+/Ca2+ exchanger inhibitor: Design, synthesis and structure-activity relationships of 3,4-dihydro-2(1H)-quinazolinone derivatives

Article abstract of DOI:10.1016/S0960-894X(03)00744-3

Design, synthesis and structure-activity relationships for 3,4-dihydro-2(1H)-quinazolinone derivatives with the inhibitory activities of the Na⁺/Ca²⁺ exchanger are discussed. These studies based on lead compound 1a lead to the discovery of a structurally novel and highly potent inhibitor against the Na⁺/Ca²⁺ exchanger 4f (SM-15811), which directly inhibited the Na⁺-dependent Ca ²⁺ influx via the Na⁺/Ca²⁺ exchanger in cardiomyocytes with a high potency.

Full text of DOI:10.1016/S0960-894X(03)00744-3

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3471–3475
Discovery of a Novel Potent Na+/Ca2+ Exchanger Inhibitor:
Design, Synthesis and Structure–Activity Relationships of
3,4-Dihydro-2(1H)-quinazolinone Derivatives
Hirohiko Hasegawa,a,* Masami Muraoka,a Kazuki Matsuia and Atsuyuki Kojimab
aResearch Center, Sumitomo Pharmaceuticals Co., Ltd., 1-98, Kasugadenaka 3-Chome, Konohana-ku, Osaka 554-0022, Japan
bTakarazuka Organic Synthesis Department, Sumika Technoservice Co., Ltd., 2-1, Takatsukasa 4-Chome,
Takarazuka City, Hyogo 665-0051, Japan
Received 27 March 2003; accepted 14 July 2003
Abstract—Design, synthesis and structure–activity relationships for 3,4-dihydro-2(1H)-quinazolinone derivatives with the inhibi-
tory activities of the Na+/Ca2+ exchanger are discussed. These studies based on lead compound 1a lead to the discovery of a
structurally novel and highly potent inhibitor against the Na+/Ca2+ exchanger 4f (SM–15811), which directly inhibited the
Na+-dependent Ca2+ influx via the Na+/Ca2+ exchanger in cardiomyocytes with a high potency.
# 2003Elsevier Ltd. All rights reserved.
Introduction
value of 3.4 mM has been reported which is a modified
amiloride derivative like dimethylamiloride.6 Further-
more, KB-R79437 and SEA04008 have also been repor-
ted as non-peptidic inhibitors. SEA0400 shows more
than 80% inhibition at 1 mM against the Na+/Ca2+
exchanger, which is a 10-fold higher potency than that
of KB-R7943( Fig. 1).
Coronary reperfusion by thrombolytic therapy or per-
cutaneous transluminal angioplasty has emerged as a
fundamental strategy in the management of ischemic
heart disease. Nonetheless, it has been suggested that
sometimes early restitution of blood flow after a period
of hypoxia results in the deteriorous effects called
reperfusion injury.1 Intracellular Ca2+ overload via
activation of the Na+/Ca2+ exchanger after ischemic
reperfusion has been indicated as a potential cause of
this, which induces post-ischemic cardiac injury.2ꢀ4
Thus, we anticipated that Na+/Ca2+ exchanger inhibi-
tors could become a new approach for the treatment of
ischemic reperfusion injury. Our research therefore
focused upon the discovery of an inhibitor of the Na+/
Ca2+ exchanger with high potency and selectivity.
To date, although a number of inhibitors against the
Na+/Ca2+ exchanger have been reported, most are
weak and non-selective inhibitors.3 Recently, Val-Met-
Arg-Met-NH2 with IC50 value of 1.5 mM has been
reported, which is a peptidic inhibitor.5 As a non-pepti-
dic inhibitor, an aroylguanidine derivative with IC50
*Corresponding author. Tel.: +81-6-6466-5220; fax: +81-6-6466-
5287; e-mail: hhasegaw@sumitomopharm.co.jp
Figure 1. Chemical structures of non-peptidic inhibitor of Na+/Ca2+
exchanger.
0960-894X/$ - see front matter # 2003Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00744-3

3472
H. Hasegawa et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3471–3475
In our research, we identified that 3,4-dihydro-2(1H)-
quinazolinone derivative 1a in our chemical library,
through a random screening, had a moderately potent
inhibitory activity against the Na+/Ca2+ exchanger.
Since 1a has a non-peptidic, structurally novel, and
quite different skeleton from other inhibitors, we there-
fore used 1a as a lead compound and synthesized deri-
vatives to find a more potent compound as well as to
evaluate the structure–activity relationships (SARs) in
three parts systemically (boxed regions in Fig. 2). These
studies lead to the discovery of a structurally novel and
highly potent inhibitor against the Na+/Ca2+ exchan-
ger 4f (SM-15811). Herein, we wish to report these
studies.
Scheme 3. Reagents and conditions: (e) NaH, MeI, DMF, 94%.
Scheme 4. Reagents and conditions: (f) P2S5, xylene, reflux, 36%.
carbonyl analogue 3 was prepared by treating 1b with
P2S5 in xylene (Scheme 4).
Chemistry
Compounds 1a–l, 1n–q were prepared using Yamamo-
to’s procedure.9,10 Trichloroacetylation of 2-amino-
benzophenone derivatives 5a and 5b with trichloroacetyl
chloride, followed by treatment with primary amines in
DMSO gave directly cyclized products 7, with accom-
panying rearrangement of the trichloromethyl group.
Removal of trichloromethyl group leading to the 3,4-
dihydro-2(1H)-quinazolinone 1a–l, 1n–q was effected by
NaBH4 in DMF (Scheme 1).
In order to synthesize 4-phenyl-3-(4-piperidinyl)-3,4-
dihydro-2(1H)- quinazolinone derivatives having cyclic
amines 4a–g, we developed a more versatile procedure.
Treatment of trichloroacetylamide 6b with 4-amino-1-
benzylpiperidine in DMSO gave imine 8. Reduction of
the imine and removal of the trichloroacetyl moiety
with NaBH4 gave diamine 9. Treatment of diamine 9
with 1,10-carbonyldiimidazole lead to cyclization to
afford 3-[4-(1-benzyl)piperidinyl]-4-phenyl-3,4-dihydro-
2(1H)-quinazolinone 4f. Debenzylation of 4f gave 4a,
which was subsequently alkylated or reductively
alkylated to afford 4b–e, 4g (Scheme 5).11
Primary amine 1m was prepared by debenzylation of 1l
with ammonium formate in the presence of Pd/C in
MeOH (Scheme 2). 2 was prepared from 1a by reaction
with NaH followed by iodomethane (Scheme 3). Thio-
Pharmacological Results and Discussion
In cardiomyocytes, fura 2 fluorescene ratio (an index of
[Ca2+]i) increases by Na+-dependent Ca2+ influx via the
Na+/Ca2+ exchanger after Na+-free treatment. The lead
compound 1a concentration-dependently attenuated the
Figure 2. Chemical structure of 3,4-dihydro-2(1H)-quinazolinone
derivative 1a.
Scheme 1. Reagents and conditions: (a) CCl3COCl, Et3N, CH2Cl2,
71–99%; (b) Z-NH2, DMSO, 12–94%; (c) NaBH4, DMF, 11–96%.
Scheme 5. Reagents and conditions: (g) 4-amino-1-benzyl piperidine,
DMSO, 66%; (h) NaBH4, MeOH, 86%; (i) 1,10-carbonyldiimidazole,
THF, reflux, 81%; (j) HCO2NH4, Pd/C, MeOH, reflux, 84%; (k) alde-
hyde, NaBH3CN, HCl, MeOH or R3-X, K2CO3, DMF (X=Br, I).
Scheme 2. Reagents and conditions: (d) HCO2NH4, Pd/C, MeOH,
reflux, 84%.

H. Hasegawa et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3471–3475
3473
increase in the Na+-free-induced fura 2 fluorescene
ratio (Fig. 3).12
cyclic thiourea (3 vs 1b). Therefore these results sug-
gested that the 3,4-dihydro-2(1H)-quinazolinone
skeleton was essential for activity.
It is well known that an increase in the [Ca2+]i is correl-
ated with the development of cardiac contractile force.
Therefore, we evaluated the inhibitory activity against
Na+- and K+-free contracture after 30 min of K+ free
incubation in isolated left atria from guinea pigs.13 The
inhibitory activities were calculated as IC30 values.
Next we replaced the phenyl group at the 4-position
(Table 1). Replacement of the phenyl group with het-
eroaryl groups such as 2-furyl 1c, 2-thienyl 1d and 2-
pyridyl 1e, which are comparable to the phenyl ring in
bulkiness, reduced the activity slightly. Introduction of
a methyl group 1f resulted in a drop in inhibitory activ-
ity by almost 10-fold, although the cyclohexyl 1g had
only a slightly reduced activity. A larger group such as
benzyl 1h abolished the activity. These results suggested
that the hydrophobic pocket at the 4-position was not
so large. The phenyl group was the most suitable for
high activity.
The lead compound 1a showed inhibitory activity with
IC30 value of 0.46 mM. We first examined the 3,4-dihy-
dro-2(1H)-quinazolinone skeleton of 1a (Table 1).
Removal of the chlorine atom at the 6-position did not
affect the activity (1b vs 1a). But introduction of a
methyl group at the 1-position (2 vs 1a) reduced the
activity markedly as did conversion of the cyclic urea to
For further exploration of the SARs, modifications of
side chain at the 3-position were carried out (Table 2).
Replacement of a nitrogen atom with a carbon atom (1i
vs 1b) or oxygen atom (1j vs 1o) resulted in complete
loss in activity, showing that the nitrogen atom is
essential. Furthermore, the amide derivative also had
diminished activity (1k vs 1o). Hence a basic nitrogen
atom was essential for activity. Actually, the calculated
pKa values of the amino group of the side chain in 1k
and 1o were ꢀ0.58 and 10.23, respectively, which means
that 3,4-dihydro-2(1H)-quinazolinones show activities
in protonated form.14
The secondary amino group showed stronger activity
than the primary amino group, but the tertiary amino
group showed the strongest activity (1o vs 1n and 1m),
suggesting that the tertiary amino group was preferred.
In terms of the length between nitrogen atom in the side
chain and nitrogen atom at the 3-position of the 3,4-
dihydro-2(1H)-quinazolinone, two carbon atoms 1o and
three carbon atoms 1p showed strong activity, but four
carbon atoms 1q reduced the activity, suggesting either
two or three carbon atoms were preferred. These results
suggested that amine at the 3-position was very impor-
tant and tertiary amine with either a length of two or
three carbons were preferred.
Figure 3. Effect of 1a on the Na+/Ca2+ exchange activity in rat car-
diomyocytes. The Na+/Ca2+ exchange activity was estimated as the
increase in fura 2 fluorescene ratio induced by exposing to the Na+
-
free HEPES-based buffer using a Ca2+ sensitive fluorescent indicator
fura 2. Each point represents the meanꢁSE of five experiments.
Table 1. Inhibitory activity of 2,4-dihydro-2(1H)-quinazolinones
against the Na+/Ca2+ exchanger
Table 2. Inhibitory activity of 3,4-dihydro-2(1H)-quinazolinones
against the Na+/Ca2+ exchangera
Compda
X
R1
W
R2
IC30 (mM)
Dimethylamiloride
30
Val-Met-Arg-Phe-NH2
1ab
1b
2
Cl
H
Cl
H
Cl
H
H
H
H
H
Ph
Ph
Ph
Ph
O
O
O
S
O
O
O
O
O
O
H
H
Me
H
H
H
H
H
H
H
10
0.46
0.35
>10
3.8
Compd
Z
IC30 (mM)
3c
1i
1j
(CH2)3CH(Me)2
(CH2)2OEt
>10
>10
>10
>10
0.58
0.069
0.04
1.9
1c
1d
1e
1f
1g
1h
2-Furyl
2-Thienyl
2-Pyridyl
Me
Cyclohexy
Benzyl
0.69
0.72
0.51
5.3
1k
1m
1n
1o
1p
1q
CH2CONEt2
(CH2)2NH2
(CH2)2NHEt
(CH2)2NEt2
(CH2)3NEt2
(CH2)4NEt2
0.57
aAll the compounds tested were racemic.
bCompound tested as the citrate.
cCompound tested as the hydrochloride.
aAll the compounds tested were racemic.

3474
H. Hasegawa et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3471–3475
Table 3. Inhibitory activity of 3,4-dihydro-2(1H)-quinazolinones
against the Na+/Ca2+ exchanger
Compda
R3
IC30 (mM)
4a
4bd
H
Me
Et
nPr
1.1
0.087
0.053
0.093
4.0
0.017
3
4cc,d
4dc,e
4ec,f
Cyclohexylmethyl
PhCH2–
Ph(CH2)2–
SM-15811 (4f)b
4gc,f
.6
aAll the compounds tested were racemic.
bCompound tested as the citrate.
cCompound tested as the hydrochloride.
dCorresponding alkyl iodes were used for N-alkylation.
enPrBr was used for N-alkylation.
Figure 4. Effect of SM–15811 (4f) on the Na+/Ca2+ exchange activity
in rat cardiomyocytes. The Na+/Ca2+ exchange activity was esti-
mated as the increase in fura 2 fluorescene ratio induced by exposing
to the Na+-free HEPES-based buffer using a Ca2+ sensitive fluor-
escent indicator fura 2. Each point represents the meanꢁSE of five
experiments.
fCorresponsing aldehydes were used for N-alkylation.
Based on these results, we therefore focused our atten-
tion on the introduction of a cyclic amine into the side
chain at the 3-position of this derivative so as to
increase the inhibitory activity. Table 3 summarizes our
results in introducing the 4-piperidinyl as a cyclic amine.
We synthesized the unsubstituted 4a and various sub-
stituted piperidine derivatives 4b–4g. Among them, non-
substituted 4a was weaker than 4b. But introduction of
chain alkyl groups enhanced the activities (4b–4d). Fur-
ther introduction of a bulky alkyl group such as cyclo-
hexylmethyl 4e reduced the inhibitory activity. In
contrast to cyclohexylmethyl 4e, the benzyl group 4f
(SM-15811) showed strong activity, with IC30 value of
0.017 mM, suggesting in this case there may be a p-p
interaction between the inhibitor and the binding site.
However, phenylethyl group 4g decreased the activity
drastically, suggesting location of the phenyl ring was
important for activity (SM-15811 vs 2g).
Conclusion
In summary, we found that 3,4-dihydro-2(1H)-quina-
zolinone derivative 1a in our library concentration-
dependently attenuated the increase in Na+-free-
induced fura 2 fluorescene ratio and exhibited inhibitory
activity against the Na+/Ca2+ exchanger with IC30
value of 0.46 mM. We disclosed the SARs of lead com-
pound 1a by systemic modification in three parts. The
results of the SARs studies are as follows: (1) Removal
of chlorine atom at the 6-position of 3,4-dihydro-2(1H)-
quinazolinone did not affect the activity. (2) 3,4-Dihy-
dro-2(1H)-quinazolinone skeleton was essential for
activity. (3) Phenyl group at the 4-position of 3,4-dihy-
dro-2(1H)-quinazolinone was preferred. (4) Amine at
the 3-position of 3,4-dihydro-2(1H)-quinazolinone was
very important and tertiary amine with either 2 or 3
carbon chain length were preferred. (5) Introduction of
N-alkyl-4-piperidine at the 3-position of 3,4-dihydro-
2(1H)-quinazolinone further enhanced the activity. (6)
Introduction of a benzyl group that interacted in a p–p
manner at the 1-position on the 4-piperidine improved
the activity and gave the highly potent inhibitor of
Na+/Ca2+ exchanger SM-15811 with IC30 value of
0.017 mM. (7) The location of phenyl ring is important
for activity.
We found that SM-15811 had strong inhibitory activity
against Na+- and K+-free contracture after 30 min of
K+ free incubation in isolated left atria from guinea
pigs. Moreover, we evaluated SM-15811 by fura 2
fluorescene ratio [an index of (Ca2+)i] increased by
Na+-dependent Ca2+ influx via Na+/Ca2+ exchanger
after Na+-free treatment in cardiomyocytes. Figure 4
shows the results. SM-15811 concentration-dependently
attenuated the increase in Na+-free induced fura 2
fluorescene ratio, indicating SM-15811 discovered by
our systemic modification of 3,4-dihydro-2(1H)-quina-
zolinone derivative directly inhibited the Na+-dependent
Ca2+ influx via Na+/Ca2+ exchanger after Na+-free
treatment in cardiomyocytes. The inhibitory activity of
SM-15811 was almost 2 orders stronger than the lead
compound 1a (Figs. 3 vs 4). These results suggested that
SM-15811 found by our systemic SARs studies showed
strong inhibitory activity against the Na+/Ca2+
exchanger.
SM-15811 discovered by this SARs study concen-
tration-dependently attenuated the increase in Na+-free
induced fura 2 fluorescene ratio, indicating SM-15811
directly inhibited the Na+-dependent Ca2+ influx via
Na+/Ca2+ exchanger in cardiomyocytes, whose activity
was almost 2 orders stronger than the lead compound 1a.
Thus we discovered the structually novel 3,4-dihydro-
2(1H)-quinazolinone derivative as the highly potent
inhibitor of Na+/Ca2+ exchanger. In particular, SM-
15811 could have therapeutic potential in the treatment

H. Hasegawa et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3471–3475
3475
of ischemic reperfusion injury as well as a powerful tool
for further studies on the role of the Na+/Ca2+
exchanger in the heart. Further pharmacological eval-
uation is currently in progress in our laboratory.
ref 15. The starting material of 1d, 1g, and 1h were synthesized
according to ref 16. The starting material of 1e was synthe-
sized according to ref 17. The starting materials (5a and 5b) of
other 3,4-dihydro-2(1H)-quinazolinone derivatives 1 and 4
were commercially available.
11. Alkyl halides and aldehydes used for N-alkylation are lis-
ted in Table 3.
12. Yamamoto, S.; Matsui, K.; Kitano, M.; Ohashi, N. J.
Cardiovasc. Pharm. 2000, 35, 855 The method of the eval-
uation system is cited therein.
Acknowledgements
We wish to thank Ms. Hitomi Sakanaka for evaluating
the inhibitory activity of compounds against the Na+/
Ca2+ exchanger. We also wish to thank Ms. Mikiko
Ohmori for evaluating the effect of 1a and SM-15811 on
the Na+/Ca2+ exchange activity in cultured cardio-
myocytes.
13. To test the inhibitory actions of several compounds on
Na+/Ca2+ exchange, we examined the effects of compounds
on Na+- and K+-free contracture after 30 min of K+ free
incubation in isolated left atria from guinea pigs according to
a modification of a technique previously described.18 Briefly,
male Hartley guinea pigs (about 400 g) were sacrified and the
left atria were quickly excised. Left atrial preparations were
mounted in 25-mL organ baths containing normal HEPES
solution maintained at 32 ꢂC and bubbled with 100% O2. The
contractile response of the preparation was measured iso-
metrically with a force–displacement transducer and recorded
on a linearly recording thermostylus oscillograph. Prepara-
tions were equilibrated for 30 min in normal HEPES solution
under 0.5 g resting tension before initiating the following
experimental procedures. Contracture was produced by expo-
sure of preparations to Na+- and K+-free HEPES-buffered
medium in the presence of Ca2+ antagonist, verapamil (10
mM), after 30 min of Na+/K+ pump inhibition in K+-free
buffered medium. Contracture was monitored in the presence
of different concentration of test compounds. Test compounds
were preincubated for 15 min Na+- and K+-free HEPES
buffered medium and then incubated in Na+- and K+-free
HEPES buffered medium. Inhibitory activity was assessed
relative to an untreated control contracture at 30 s after
exposure to Na+- and K+-free HEPES buffered medium.
14. The pKa values were calculated with ACD/pKa ver. 6.0.
Advanced Chemistry Development: Toronto, Ontario,
Canada.
References and Notes
1. Nayler, W. G.; Elz, J. S. Circulation 1986, 74, 215.
2. (a) Tani, M. Annu. Rev. Physiol. 1990, 52, 543. (b) Opie,
L. H. Cardiovasc. Drugs Ther. 1991, 5, 237.
3. Recent review on Na+/Ca2+ exchanger: Blaustein, M. P.;
Lederer, W. J. Physiol. Rev. 1999, 79, 763.
4. (a) Grinwald, P. M. J. Mol. Cell. Cardiol. 1982, 14, 359. (b)
Renlund, D. G.; Gerstenblith, G.; Lakatta, E. G.; Jacobs,
W. E.; Kallman, C. H.; Weisfeldt, M. L. J. Mol. Cell. Cardiol.
1984, 16, 795. (c) Tani, M.; Neely, J. R. Circ. Res. 1989, 65,
1045. (d) Murphy, J. G.; Smith, T. W.; Marsh, J. D. Am. J.
Physiol. 1988, 254, H1133.
5. Khananshvili, D.; Price, D. C.; Greenberg, M. J.; Sarne, Y.
J. Biol. Chem. 1993, 268, 200.
6. (a) Brown, L.; Cragoe, E. J., Jr.; Abel, K. C.; Manley,
S. W.; Bourke, J. R. Arch. Pharmacol. 1991, 344, 220. (b)
Rogister, F.; Laekmann, D.; Plasman, P.; Eylen, V.; Ghyoot,
M.; Maggetto, C.; Liegeois, J.; Geczy, J.; Herchuelz, A.;
Delarge, J.; Masereel, B. Eur. J. Med. Chem. 2001, 36, 597.
7. (a) Iwamoto, T.; Watano, T.; Shigekawa, M. J. Biol. Chem.
1996, 271, 22391. (b) Watano, T.; Kimura, J.; Morita, T.;
Nakanishi, H. Br. J. Pharmcol 1996, 119, 555.
8. (a) Matsuda, T.; Arakawa, N.; Takuma, K.; Kishida, Y.;
Kawasaki, Y; Sakue, M.; Takahasi, K.; Takahashi, T.; Suzuki,
T.; Ota, T.; Takahasi, A. H.; Onishi, M.; Tanaka, Y.; Kameo,
K.; Baba, A. J. Pharmacol. Exp. Ther. 2001, 298, 249. (b)
Tanaka, H.; Nishimaru, K.; Aikawa, T.; Hirayama, W.;
Tanaka, Y.; Shigenobu, K. Br. J. Pharmcol. 2002, 135, 1096.
9. Yamamoto, M.; Yamamoto, H. Chem. Pharm. Bull. 1981,
29, 2135.
15. Berger, L.; Stempel, A.; Sternbach, L. H.; Wenis, E.;
Fryer, R. I.; Schmidt, R. A.; Belg. Patent 619,101, 1962;
Chem. Abstr. 1963, 59, 10092.
16. Hunziker, F.; Fischer, R.; Kipfer, P.; Schmutz, J.; Burki,
H. R.; Eichenberger, E.; White, T. G. Eur. J. Med. Chem.
1981, 16, 391.
17. (a) Goldberg, N. N.; Barkley, L. B.; Levine, R. J. Am.
Chem. Soc. 1951, 73, 4301. (b) Yamamoto, H.; Saito, C.;
Inaba, S.; Awata, H.; Yamamoto, M.; Sakai, Y.; Komatsu, T.
Arzneim.-Forsch. 1973, 23, 1266. (c) Ockenden, D. W.; Scho-
field, K. J. Chem. Soc 1953, 3440.
18. Chapman, R. A.; Coray, A.; Mcguigan, J. A. S. J. Physiol.
(London) 1983, 343, 253.
10. The starting material of 1c was synthesized according to