Evaluation of canthinone alkaloids as cerebral protective agents

Article abstract of DOI:10.1016/j.bmcl.2016.09.006

Considerable attention has been paid to cerebral protective drugs as a potential therapy for dementia. Screening of a natural compound library here resulted in identification of five canthinone alkaloids, viz., picrasidine L (1), picrasidine O (2), eurycomine E (3), 3-ethyl-canthin-5,6-dione (4), and 3-ethyl-4-methoxy-canthin-5,6-dione (5), as novel cerebral protective agents. The structure–activity relationship indicated that C-4, C-9, and N-3 substitutions greatly affected their cerebral protective effect. Among these, compound 2 exhibited a cerebral protective effect through suppressing neuronal hyperexcitability due to an increase in the excitatory neurotransmitter glutamic acid. Furthermore, compound 2 did not affect heart rate and mean systolic blood pressure. This investigation suggests that compound 2 has potential for further development as a cerebral protective drug.

Full text of DOI:10.1016/j.bmcl.2016.09.006

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Evaluation of canthinone alkaloids as cerebral protective agents
y
⇑
⇑
Tatsunori Sasaki, Wei Li , Taichi Ohmoto , Kazuo Koike
Faculty of Pharmaceutical Sciences, Toho University, Miyama 2-2-1, Funabashi, Chiba 274-8510, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Considerable attention has been paid to cerebral protective drugs as a potential therapy for dementia.
Screening of a natural compound library here resulted in identification of five canthinone alkaloids,
viz., picrasidine L (1), picrasidine O (2), eurycomine E (3), 3-ethyl-canthin-5,6-dione (4), and 3-ethyl-4-
methoxy-canthin-5,6-dione (5), as novel cerebral protective agents. The structure–activity relationship
indicated that C-4, C-9, and N-3 substitutions greatly affected their cerebral protective effect. Among
these, compound 2 exhibited a cerebral protective effect through suppressing neuronal hyperexcitability
due to an increase in the excitatory neurotransmitter glutamic acid. Furthermore, compound 2 did not
affect heart rate and mean systolic blood pressure. This investigation suggests that compound 2 has
potential for further development as a cerebral protective drug.
Received 28 April 2016
Revised 14 August 2016
Accepted 2 September 2016
Available online xxxx
Keywords:
Alkaloid
Canthinone
Cerebral protective effect
Dementia
Ó 2016 Elsevier Ltd. All rights reserved.
Picrasidine O
Dementia is a clinical syndrome characterized by a cluster of
symptoms and signs manifested by memory difficulties, language
disturbances, psychological and psychiatric changes, and impair-
ment in the activities of daily living. Alzheimer’s disease is the
most common type of dementia, followed by vascular dementia,
and Lewy body dementia.1 Clinical drugs for the treatment of
dementia include acetylcholinesterase inhibitors, such as donepe-
protective effects. The experimental scheme is summarized in
Figure 2.4
Cerebral ischemia was induced with bilateral carotid ligation in
Mongolian gerbils (Meriones unguiculatus).5 Gerbils were divided
into groups: (1) Normal group, a group of gerbils that underwent
no treatment; (2) Control group, a group of gerbils in which cere-
bral ischemia was induced, but which was given no compound;
(3) Compound groups, groups of gerbils that underwent cerebral
ischemia and were given compounds 1–22, individually; and (4)
the vinpocetine group, a group of gerbils that underwent cerebral
ischemia and were given vinpocetine.6
A step-down passive avoidance test is widely used as a standard
test for evaluation of learning/memory in gerbils. In this test, elec-
trical stimulation was provided when the gerbils stepped downed
from the platform. The step-down latency time, which was defined
as the length of time that gerbils stayed on the platform, was used
as a parameter for accessing learning and memory ability.7
Cerebral ischemia led to selective necrosis of neurons in specific
brain regions. The CA1 subfield of the hippocampus is a brain
region that is particularly sensitive to ischemia.8 Thus, in this
study, the density of surviving neurons in the CA1 subfield of the
hippocampus was measured to examine the cerebral protective
effect.
zil, galantamine, rivastigmine, and an N-methyl-D-aspartate
(NMDA) receptor antagonist, memantine.2
Picrasma quassioides and other plants in the Picrasma genus of
the Simaroubaceae family have been used as bitter stomachics
for gastritis, loss of appetite, and indigestion in Chinese and Japa-
nese traditional medicine. From these plants, a number of b-carbo-
line and canthinone type alkaloids have been isolated, which have
been reported to exert a variety of biological activities, including
PTP1B-inhibition, anti-inflammatory activity, 30,50-cyclic adenosine
monophosphate phosphodiesterase inhibition, and cytotoxicity.3
In the present study, we report five canthinone alkaloids (1–5) that
are potential new cerebral protective agents (Fig. 1).
Natural or chemical synthetic canthinone alkaloids (1–16) and
b-carboline alkaloids (17–22) were screened for their cerebral pro-
tective effects at 100 mg/kg (p.o.) in ischemic animal, which pro-
vided a disease model of dementia. Measurement of the latency
time in a step-down passive avoidance test and the density of sur-
viving neurons of these animals were used to assess the cerebral
We identified five canthin-5,6-dione alkaloids, namely, picra-
sidine L (1), picrasidine O (2), eurycomine E (3), 3-ethyl-canthin-
5,6-dione (4), and 3-ethyl-4-methoxy-canthin-5,6-dione (5), which
resulted in a longer step-down latency time and greater density of
surviving neurons than in the control animals (Figs. 3 and 4). Nota-
bly, picrasidine L (1) and picrasidine O (2) treatment resulted in
virtually the same results as the normal group (Figs. 3 and 4). How-
⇑
Corresponding authors. Tel.: +81 47 4721161; fax: +81 47 4721404.
E-mail addresses: liwei@phar.toho-u.ac.jp (W. Li), koike@phar.toho-u.ac.jp
(K. Koike).
y
Deceased.
http://dx.doi.org/10.1016/j.bmcl.2016.09.006
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Sasaki, T.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.09.006

2
T. Sasaki et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
11
8
1
2
10
9
N
N
N
N
6
CH3
N
N
CH3
H3CO
N
CH3
3
7
O
4
O
O
OCH3
O
5
O
O
O
2
1
3
H
N
N
N
N
CH2CH3
CH2CH3
N
O
OCH3
H3CH2COOC
CH2CH3
Vinpocetine
O
O
5
4
Figure 1. Five canthinone alkaloids and vinpocetine.
Figure 3. Step-down latency time (s) in ischemic gerbils (n = 10) treated with
vehicle or with either of five canthinone alkaloids or vinpocetine administered
before bilateral carotid ligation. Results are presented as the mean standard error
of the mean. Normal: Normal group; Control: Control group; 1, 2, 3, 4, 5: groups
administered compound 1, 2, 3, 4, or 5; Vinpocetine: group administered
ever, the other 11 canthinone alkaloids (6–16) and six b-carboline
alkaloids (17–22) showed no cerebral protective effects in either
measurements (Supporting information). Vinpocetine, a struc-
turally related carboline alkaloid, which is currently prescribed
for the treatment of disorders arising from cerebrovascular and
cerebral neurodegenerative diseases that ultimately lead to
dementia in the elderly,6 showed a much weaker effect than com-
pounds 1–5 in this assay model.
vinpocetine. One asterisk (*) indicates
a p-value smaller than 0.05 (p <0.05),
compared to the control group, three asterisks (***) indicate a p-value smaller than
0.001 (p <0.001), compared to the control group.
In terms of step-down latency time, compounds containing an
N-3-methyl moiety were approximately twice as potent as those
containing an N-3-ethyl moiety (1 vs 4, and 2 vs 5). Compounds
containing a C-9-methoxy moiety showed a remarkably reduced
activity (3 vs 1) (Fig. 3). In terms of the measurement of density
of surviving neurons, a C-4-methoxy moiety decreased the cerebral
protective activity [1 vs 2 (p <0.05), and 4 vs 5] (Fig. 4).
The mechanisms underlying the cerebral protective effects of
compound 2 were investigated further. Benzodiazepines have been
reported to have a cerebral protective effect.9 b-carbolines, such as
b-carboline-carboxyl-ethylester and harmane, were reported to
bind to rat brain benzodiazepine receptor.10 Compound 2 had b-
carboline backbone skeleton in the molecular, and the cerebral
protective activity of 2 was predicted to show by binding with ben-
zodiazepine receptors. Thus, the ability of compound 2 to bind to
the benzodiazepine receptor was investigated using an in silico
docking study.11 However, compound 2 showed much weaker
Figure 4. The density (/mm) of surviving neurons in the CA1 subfields of the
hippocampus in ischemic gerbils (n = 10) treated with vehicle or either of five
canthinone alkaloids or vinpocetine, administered before bilateral carotid ligation.
Results are presented as the mean standard error of the mean. Normal: Normal
group; Control: Control group; 1, 2, 3, 4, 5: groups administered compound 1, 2, 3, 4,
or 5; Vinpocetine: group administered vinpocetine. One asterisk (*) indicates a p-
value smaller than 0.05 (p <0.05), compared to the control group, three asterisks
(***) indicate a p-value smaller than 0.001 (p <0.001), compared to the control
group.
Step-down passive avoidance test
7 days
p.o. of sample or 0.5% CMC
30 min
Cerebral ischemia (5 min)
3 days
binding (ꢀ11.90 kcal/mol) than diazepam (binding energy: ꢀ
72.64 kcal/mol), suggesting that it exerts its cerebral protective
effect by other mechanisms.
Glutamic acid is the principal excitatory neurotransmitter in the
brain. Endogenous glutamic acid may contribute to acute brain
damage occurring after status epilepticus, cerebral ischemia, or
Step-down passive avoidance test
1 day
Measurement of latency time
traumatic brain injury, by activating NMDA,
a-amino-3-hydroxy-
5-methyl-4-isoxazolepropionic acid, or metabotropic glutamate
receptor 1 receptors.12 Canthinone alkaloids might exert a cerebral
protective effect through suppression of neuronal cell death due to
the hyperexcitability caused by a cerebral ischemia-induced eleva-
tion in the glutamic acid concentration. To test this mechanism,
kainic acid was administered peripherally to gerbils to provoke
over-excitement and neuronal cell death, and the gerbils were then
given compound 2 orally.
3 days
Fixation of brain
10 days
Measurement of neuronal density
Figure 2. Experimental scheme of measurement of step-down latency time and
neuronal density.
Please cite this article in press as: Sasaki, T.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.09.006

T. Sasaki et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
3
Figure 7. The density (/mm) of survival neurons in the CA1 subfields of the
hippocampus in ischemic gerbils (n = 10) treated with vehicle or compound 2
administered after bilateral carotid ligation. Results are presented as the
mean standard error of the mean. Normal: Normal group; Control: Control group;
A: Orally administered compound 2 (at 30 mg/kg) 3 h after the cerebral ischemia; B:
Orally administered compound 2 (at 30 mg/kg) 3 h and 6 h after cerebral ischemia.
Three asterisks (***) indicate a p-value smaller than 0.001 (p <0.001), compared to
the control group.
Figure 5. The density (/mm) of survival neurons in the CA1 subfields of
hippocampus in ischemic gerbils (n = 10) treated vehicle or compound 2 admin-
istered after neuronal cell death induced by kainic acid injection. Results are
presented as the mean standard error of the mean. Normal: Normal group;
Control: Control group; 2: group administered compound 2. Three asterisks (***)
indicate a p-value smaller than 0.001 (p < 0.001), compared to the control group.
Kainic acid is a potent agonist of the glutamate receptor. In ger-
bils, injections of kainic acid results in recurrent seizures, behav-
ioral changes, and subsequent degeneration of selective
populations of neurons in the brain.13 Thus, administration of kai-
nic acid14 has been widely used as a model for studying the mech-
anisms underlying neurodegenerative pathways induced by
excitatory neurotransmitters. In this study, kainic acid-induced
neuronal necrosis was significantly improved in the compound
2-treated group compared to the control group (Fig. 5). This sug-
gested that the mechanism underlying the cerebral protective
effect of compound 2 involved the suppression of over-excitation
of neuronal cells, caused by abnormal glutamatergic signaling.
As mentioned above, cerebral ischemia significantly increases
extracellular glutamate levels. Activation of the glutamate receptor
enhances the influx of calcium ions into the cell, resulting in dam-
age to nerve cells.15 Thus, the effects of compound 2 on tissue dam-
age outside the hippocampus were investigated by 45Ca-
autoradiography.16 In the control group, after subjecting the ani-
mals to cerebral ischemia for 15 min, accumulation of 45Ca2+ was
measured in the cerebral cortex, striatum, and optic vesicle, in
addition to the hippocampus. The levels of 45Ca2+ accumulation
were reduced in the group that received compound 2 at 30 mg/
kg orally, suggesting that compound 2 resulted in significant
improvement in extent of brain damage (Fig. 6).
To investigate the administration schedule for using compound
2 as an acute-phase cerebral protection drug candidate, the effect
of compound 2, orally administered at 3 h and/or 6 h after cerebral
ischemia, on the density of surviving neuronal cells was examined.
Neuronal cell death was reduced more when compound 2 (at
30 mg/kg) was administered twice, at 3 h and 6 h after ischemia,
than when only administered once at 3 h (Fig. 7).
The use of antihypertensive drugs in acute ischemic stroke must
be considered carefully, as the drugs decrease cerebral blood flow
and can exacerbate ischemic injuries.18 Effects of compound 2 on
heart rate and mean systolic blood pressure were therefore tested;
neither was affected by compound 2, even at a dose 3-fold higher
than that required for pharmacological effect (30 and 100 mg/kg).
These results indicated that brain damage would not be aggravated
by administration of compound 2. Furthermore, compound 2
showed no effect on body temperature, and showed a weak loco-
motion-reducing effect.
In conclusion, the main effects of compound 2 were (i) improv-
ing learning and memory performance, (ii) inhibiting delayed neu-
ronal cell death induced by cerebral ischemia, (iii) reducing
neuronal cell death induced by the excitatory neurotransmitter,
glutamic acid and the excitotoxin, kainic acid, and (iv) decreasing
the extent of brain tissue damage. Moreover, compound 2 was
shown to have no effect on heart rate or mean blood pressure.
These effects suggest that compound 2 has a cerebral protective
effect, preventing the progression of various intractable neurolog-
ical diseases that are accompanied by neuronal cell death. More
detailed investigations are required for development of picrasidine
O (2) as a cerebral protective drug candidate.
Acknowledgements
The authors thank the Exploratory Research Laboratories II, Dai-
ichi Pharmaceutical Co., Ltd. (currently Daiichi Sankyo Co., Ltd.) for
technical support to this study.
Supplementary data
Figure 6. The volume of 45Ca accumulation (mm3) in gerbils (n = 10). Results are
presented as the mean standard error of the mean. Control: Control group; 2:
group administered compound 2. One asterisk (*) indicates a p-value smaller than
0.05 (p <0.05), compared to the control group.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.09.
006.
Please cite this article in press as: Sasaki, T.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.09.006

4
T. Sasaki et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
11. Molecular docking simulation: Molecular docking simulation was performed
References and notes
against the GABA(A)-receptor carried out using the CDOCKER docking protocol
of Discovery studio 4.5 (Dassault Systemes Biovia K.K., Vélizy-Villacoublay,
France). CDOCKER (CHARMm-based DOCKER) is a molecular dynamics-based
docking algorithm. It uses the CHARMm family of force fields and offers full
flexibility for ligands, including dihedrals, angles, and bonds. Docking helps to
predict the best binding compounds based on various scoring functions. The
high-resolution crystal structure of the GABA(A)-receptor-associated protein,
GABARAP was obtained from the Protein Data Bank (PDB-ID: 1kjt) (http://
www.rcsb.org). Missing amino acid sequences were complemented, and all co-
crystalized ligands and water data were removed. The starting 3D
1. Burns, A.; Iliffe, S. BMJ 2009, 338, b75.
2. Qaseem, A.; Snow, V.; Cross, J. T.; Forciea, M. A.; Hopkins, R.; Shekelle, P.;
Adelman, A.; Mehr, D.; Schellhase, K.; Campos-Outcalt, D.; Santaguida, P.;
Owens, D. K. Ann. Int. Med. 2008, 148, 370.
3. (a) Sung, Y.; Koike, K.; Nikaido, T.; Ohmoto, T.; Sankawa, U. Chem. Pharm. Bull.
1984, 32, 1872; (b) Ohmoto, T.; Nikaido, T.; Koike, K.; Kohda, K.; Sankawa, U.
Chem. Pharm. Bull. 1988, 36, 4588; (c) Koike, K.; Yoshino, H.; Nikaido, T.
Heterocycles 1999, 51, 315; (d) Yoshino, H.; Koike, K.; Nikaido, T. Heterocycles
1999, 51, 281; (e) Sasaki, T.; Li, W.; Higai, K.; Koike, K. Bioorg. Med. Chem. Lett.
2015, 25, 1979.
conformations of compound
2 and diazepam were prepared with Dock
Ligands Fit. The calculated binding score of all docking positions were
evaluated. The best ranked position from each of the binding sites was
determined by the lowest calculated binding energy.
4. The protective effect of compounds on delayed neuronal death after transient
cerebral ischemia was assessed. Animals: Male Mongolian gerbils were
purchased from Seiwa Experimental Animals Ltd. (Oita Japan). They were
maintained with ad libitum access to water and chow diet (Funabashi Farm-F2,
Chiba, Japan) on a 12-h light–dark cycle, in a temperature- and humidity-
controlled room, until the experiment, at which time they weighed 53–85 g.
Assessment of cerebral protective effect of compounds: Five to 11 animals
were used for each experimental group, and the effect of 22 alkaloids and
vinpocetine were assessed. First, the animals underwent a step-down passive
avoidance test for about 1 week, and were then orally administered 100 mg/kg
per 3.0 mL of samples dissolved or suspended in 0.5% carboxymethyl cellulose
solution. Thirty minutes after administration, the cervical region was opened
via a midline incision under anesthesia with 2.5% halothane/N 20(70)/02(30),
and the common carotid arteries were revealed. After cessation of anesthesia,
cerebral ischemia was induced in animals for 5 min by bilateral ligation of the
common carotid arteries, using Yasargil temporal clips (AESCULAP Co. Ltd.,
Tuttlingen Germany), after which the blood flow was resumed. Three days
thereafter, the animals underwent the step-down passive avoidance test. After
1 day, the step-down latency time was measured. After an additional 3 days,
the animals were again anesthetized by intraperitoneal injection of 50 mg/kg
of sodium pentobarbital (Abott Co. Ltd., Newark, UK). Their brains were
perfusion-fixed using 10% formalin neutral buffer solution (Wako Pure
Chemical Industries Ltd., Osaka Japan). The brains were removed from the
12. Meldrum, B. S. J. Nutr. 2000, 130, 1007S.
13. Zhang, X. M.; Zhu, J. Curr. Neuropharmacol. 2011, 9, 388.
14. Kainic acid injection: Kainic acid injection was performed according to the
method of Kirino et al.8 Male Mongolian gerbils (n = 10) were deprived of food
overnight. On the following morning, they were injected with atropine
sulphate (5.0 lg/animal) and anesthetized with pentobarbital (40 mg/kg, ip).
They were then fixed on a stereotaxic apparatus. A small burr hole was made in
the skull at 1.6 mm posterior to the bregma and 3.0 mm right of the midline.
An unbeveled stainless steel needle (external caliber = 100 lm) was lowered
1.5 mm from the dural contact. The tip of the needle was thus located in the
CA3 sector of the right hippocampus. Kainic acid (Sigma-Aldrich, St Louis, MO,
USA) was dissolved in saline and its pH adjusted to 7.4. Using a microinfusion
pump, 0.2 lL of the kainate solution (1.0 lg/lL) was injected into this sector
over a period of 4.0 min. The needle was left undisturbed for 10 min and then
withdrawn. Four days after this procedure, animals were perfusion-fixed. Five
animals were fixed with 3.5% formaldehyde in 0.1 M phosphate buffer. These
specimens were silver impregnated for localization of degenerating presynap-
tic terminals. The remaining animals were perfusion-fixed and processed for
electron microscopy as described above. For comparison, male Mongolian
gerbils were perfusion-fixed 1 month after ischemia, without kainate injection,
and the specimens were processed using the silver impregnation method.
15. Kristián, T.; Siesjö, B. K. Stroke 1998, 29, 705.
skulls, embedded in paraffin, and the coronal sections were sliced into 5.0-lm-
thick sections. The density of surviving neurons in the CA1 subfields of the
hippocampus was determined by microscopy after hematoxylin–eosin
staining. Statistical analysis: Experimental data were statistically assessed
using Student’s t-test.
16. Ca autoradiography: 45Ca autoradiography was performed according to the
method of Sakamoto et al.17 Seven days after ischemia, the animals were
anesthetized with 2% halothane in a mixture of 30% O2 and 70% N2O. A bolus of
100 l
Ci of 45CaC12 (Amersham, Little Chalfont, UK; spec. act. 10–40 mCi/mg)
5. Kirino, T. Brain Res. 1982, 239, 57.
6. Nyakas, C.; Felszeghy, K.; Szabó, R.; Keijser, J. N.; Luiten, P. G.; Szombathelyi, Z.;
Tihanyi, K. CNS Neurosci. Ther. 2009, 15, 89.
7. Kameyama, T.; Nabeshima, T.; Kozawa, T. J. Pharmacol. Methods 1986, 16, 39.
8. Kirino, T.; Tamura, A.; Sano, K. Brain Res. 1990, 510, 17.
9. Schwartz, R. D.; Yu, X.; Katzman, M. R.; Hayden-Hixson, D. M.; Perry, J. M. J.
Neurosci. 1995, 15, 529.
10. (a) Braestrup, C.; Nielsen, M.; Olsen, C. E. Proc. Natl. Acad. Sci. U.S.A. 1980, 77,
2288; (b) Rommelspacher, H.; Nanz, C.; Borbe, H. O.; Fehske, K. J.; Müller, W. E.;
Wollert, U. Naunyn Schmiedebergs Arch. Pharmacol. 1980, 314, 97.
was injected intravenously (iv). The animals were anesthetized with ether at
5 h after the injection of CaCl2; their brains were then rapidly removed and
frozen in powdered dry ice. Serial coronal sections, 20-lm-thick, were cut from
the frozen brains in a cryostat at ꢀ20 °C, and mounted and dried at 60 °C on
cover glass. For the 45Ca autoradiographic study, the sections were placed in
contact with Kodak NMC-film and exposed for 2 weeks.
17. Sakamoto, N.; Kogure, K.; Kato, H.; Ohtomo, H. Brain Res. 1986, 364, 372.
18. Osaki, Y.; Matsubayashi, K.; Yamasaki, M.; Okumiya, K.; Yoshimura, K.;
Yoshimura, K.; Hamashige, N.; Doi, Y. Hypertens. Res. 1998, 21, 169.
Please cite this article in press as: Sasaki, T.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.09.006