Coumarin derivatives protect against ischemic brain injury in rats

Article abstract of DOI:10.1016/j.ejmech.2013.04.015

Neuroprotection strategies are of great importance in the treatment of ischemic brain injury. Screening of our in-stock coumarin derivatives identified compound 1 as exhibiting neuroprotective activity. Subsequently, a structural optimization was carried out, which led to the discovery of the potent compound 20. This compound signi ficantly attenuated the damage in a cell line derived from a pheochromocytoma of the rat adrenal medulla induced by oxygen-glucose deprivation in vitro. Furthermore, compound 20 exhibited clear neuroprotection in middle cerebral artery occlusion rats by reducing infarct size and brain-water content, improving neurological function, and suppressing neuronal loss and neuropathological changes in the cortex and hippocampus. Pharmacokinetic evaluation indicated that compound 20 could penetrate the blood-brain barrier of rats.

Full text of DOI:10.1016/j.ejmech.2013.04.015

European Journal of Medicinal Chemistry 67 (2013) 39e53
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Coumarin derivatives protect against ischemic brain injury in rats
,
,
Mingna Sun a 1, Jinfeng Hu a 1, Xiuyun Song a, Donghui Wu a, Linglei Kong a, Yupeng Sun a,
Dongmei Wang a, Yan Wang a , Naihong Chen
, Gang Liu
,
a,
a,b,c,
*
**
***
a State Key Laboratory of Bioactive Substance and Function of Natural Medicines, Institute of Materia Medica, Chinese Academy of Medical Sciences
and Peking Union Medical College, 2 Nanwei Rd., Xicheng Dist., Beijing 100050, PR China
b Tsinghua-Peking Center for Life Sciences, School of Medicine, Tsinghua University, Haidian Dist., Beijing 100084, PR China
c Department of Pharmacology and Pharmaceutical Sciences, School of Medicine, Tsinghua University, Haidian Dist., Beijing 100084, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Neuroprotection strategies are of great importance in the treatment of ischemic brain injury. Screening of
our in-stock coumarin derivatives identified compound 1 as exhibiting neuroprotective activity. Subse-
quently, a structural optimization was carried out, which led to the discovery of the potent compound 20.
This compound significantly attenuated the damage in a cell line derived from a pheochromocytoma of
the rat adrenal medulla induced by oxygeneglucose deprivation in vitro. Furthermore, compound 20
exhibited clear neuroprotection in middle cerebral artery occlusion rats by reducing infarct size and
brain-water content, improving neurological function, and suppressing neuronal loss and neuropatho-
logical changes in the cortex and hippocampus. Pharmacokinetic evaluation indicated that compound 20
could penetrate the bloodebrain barrier of rats.
Received 26 February 2013
Received in revised form
2 April 2013
Accepted 6 April 2013
Available online 9 May 2013
Keywords:
Neuroprotective agents
Coumarin
Ischemic brain injury
Piperazine
Ó 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction
excitotoxicity, apoptosis and inflammation by triggering various
critical cellular signal transduction pathways [5].
Ischemic brain injury is considered one of the leading causes of
death and adult disability in many countries because of its high
mortality rate. Ischemic stroke, which results from occlusion of an
artery in the brain, accounts for more than 80% of all strokes. A large
body of evidence suggests that a transient or permanent reduction
of cerebral blood flow often initiates brain ischemia and usually
leads to neuronal cell death in the central core and penumbra.
Transient ischemia was found to produce large amounts of free
radicals and neurotoxicity in the penumbral cortex during reper-
fusion [1e3]. In particular, oxidative stress is a major factor in
cerebral ischemic/reperfusion (I/R) damage because the brain
consumes a large quantity of oxygen. The reactive oxygen and ni-
trogen species can lead to lipid peroxidation, protein oxidation and
DNA damage in brain I/R injury [4]. Moreover, oxidative damage
caused by reactive oxygen radicals has complex interactions with
Antioxidant strategies have led to many results in numerous
experimental stroke studies and clinical therapy (Fig. 1). Ebselen is
a selenium compound that possesses glutathione peroxidase-like
activity [6]. NXY-059, a nitrone-derived free radical trapping
agent, protects against I/R injury by preventing the decline of
mitochondrial functions, blocking cytochrome c efflux and subse-
quent caspase activation [7e9]. Tirilazad also possesses free-
radical-scavenging activity [10,11]. Edaravone inhibits lipid perox-
idation by scavenging free radicals, such as superoxide anion
radical (Oꢀ2 $), nitric oxide radical (NO$) and peroxynitrite anion
(ONOOe), and has been used in the treatment of cerebral in-
farctions [12e17]. Hydroxyl substituted coumarins are widely
distributed in plants. Recently, it has been reported that many
coumarin derivatives exhibit a neuroprotective effect by their
antioxidant and anti-inflammatory activities, and the structuree
activity relationships (SARs) of coumarin derivatives have been
extensively investigated [18]. For example, esculetin shows neu-
roprotective effects on cerebral I/R injury in mice and exerts anti-
* Corresponding author. Department of Pharmacology and Pharmaceutical Sci-
ences, School of Medicine, Tsinghua University, Haidian Dist., Beijing 100084, PR
China. Tel./fax: þ86 10 63167165.
apoptotic activity by upregulating the expression of
B cell
** Corresponding author. Tel./fax: þ86 10 63165238.
lymphoma/leukemia-2 (Bcl-2) and downregulating the expres-
sion of Bcl-2-associated X protein (Bax) [19].
In previous studies, we developed a series of 3-piperazine-
substituted coumarin derivatives and discovered compound 1,
*** Corresponding author. Tel./fax: þ86 10 63165177.
E-mail addresses: wangyan@imm.ac.cn (Y. Wang), chennh@imm.ac.cn (N. Chen),
gangliu27@tsinghua.edu.cn, gliu@imm.ac.cn, gangliu27@gmail.com (G. Liu).
1
These authors contributed equally to this work.
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.04.015

40
M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
between the piperazinyl group and the coumarin core, was pre-
pared from phloroglucinol via a Pechmann reaction with 9 using
PBr3 as a catalyst, and then nucleophilic substitution with N-
methylpiperazine in the presence of K2CO3 (Scheme 2A) [21]. The
alkyl carbonyl compounds 17a and 17b were also designed and
synthesized (Scheme 2B). Starting material 12 reacted with ethyl
bromoacetate/propionate in the presence of sodium ethoxide
(EtONa) to give the intermediates 14a and 14b, which subsequently
reacted with phloroglucinol to be transformed into intermediates
15a and 15b, respectively. After hydrolysis and condensation with
N-methylpiperazine catalyzed by benzotriazole-1-yloxy-tris-
(dimethylamino)-phosphonium hexafluorophosphate (BOP), the
target compounds 17a and 17b were obtained, respectively [22,23].
To synthesize the target compounds 20 and 21, the Pechmann
reaction of 18 with 19 catalyzed by boron trifluoride diethyl ether
(BF3$Et2O) and direct methylation of compound 1 by dimethyl
sulfate (Me2SO4) were carried out (Scheme 3A). The former route
resulted in compounds 20 and 21 with an equal amount of each, as
analyzed by liquid chromatographyemass spectrometry (LCeMS)
and thin-layer chromatography detection under ultraviolet 254 nm
wavelength radiation, and it was too difficult to separate them by
silica gel column chromatography. Interestingly, the direct
methylation of compound 1 produced compound 22 but no C5-/C7-
hydroxyl methylated products. This observation let us successfully
develop the third synthetic route (Scheme 3B). Compound 1 was
protected by p-toluenesulfonyl chloride (TsCl) in pyridine to give
intermediate 23. The anticipated compound 20 was prepared from
23 through five steps, involving benzylation of the N-methyl-
piperazine by benzyl bromide (BnBr), selective expulsion of the
tosyl group by tetrabutylammonium fluoride (TBAF) to form the
C5-hydroxyl group, methylation of the C5-hydroxyl group, removal
of the protective tosyl group of the C7-hydroxyl group and
debenzylation of N-methylpiperazine. The anticipated compound
21 was also prepared from 23 through five steps, involving selective
expulsion of the tosyl group by TBAF to form the C5-hydroxyl
group, benzylation of the N-methylpiperazine and the C5-
hydroxyl group by BnBr, removal of the protective tosyl group of
the C7-hydroxyl group, methylation of the C7-hydroxyl group and
debenzylation of N-methylpiperazine.
Fig. 1. Chemical structures of literature neuroprotective agents.
which exhibits anti-inflammatory activity as an antagonist of
chemokine-like factor 1 (Fig. 2) [20]. Considering the potential
antioxidant activity of the analogs with phenolic hydroxyl groups
such as compound 1, we screened 37 coumarin derivatives (see
supplementary content, Table S1) in our in-stock library at a single
dose through the assay of PC12 cells (a cell line derived from a
pheochromocytoma of the rat adrenal medulla) damaged by oxy-
geneglucose deprivation (OGD). A preliminary review of the data
and structures suggested that appropriate substituents on the C5-
and C7-positions (e.g., hydroxyls) and the C3-position (e.g., com-
pounds 1ꢀ4, Fig. 2) of coumarin might contribute to the neuro-
protective activity of these compounds. This paper reports our
optimization of compound 1 and the corresponding biological
evaluation of neuroprotective activity in vitro and in vivo. With the
aim of improving the neuroprotective activity, the optimization
focused on the substituents of the piperazinyl group (compounds
6e8), the linker between the piperazinyl group and the coumarin
scaffold (compounds 11 and 17) and the selective modification of
the C5- and C7-hydroxyl groups (compounds 20, 21, 33, 35 and 37).
2. Results and discussion
Intermediates 32aev were prepared from the key intermediate
28 via a Mitsunobu reaction with appropriate alcohols in the
presence of diethyl azodicarboxylate (DEAD) and triphenylphos-
phine (PPh3). Removal of the protective tosyl group catalyzed by
TBAF afforded the target compounds 33aev (Scheme 4). After
removing the tert-butoxycarbonyl group of 32u and 32v by
CF3COOH in dichloromethane (DCM), the intermediates 34a and
2.1. Chemistry
Compounds 6e8 were prepared from the intermediate 5
through reaction with various acid anhydrides, isocyanates and
isothiocyanates (Scheme 1). Compound 11, with an ethylene linker
Fig. 2. Chemical structures of some in-stock coumarins with neuroprotective activity.

M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
41
Scheme 1. Synthesis of coumarin analogs 6e8. Reagents and conditions: (a) (R1CO)2O, triethylamine (TEA), tetrahydrofuran (THF), rt; (b) R2NCO, THF/dimethylformamide (DMF),
50 ꢁC; (c) R3NCS, THF/DMF, 50 ꢁC.
34b were obtained, respectively, which further afforded the target
compounds 35a and 35b by removing the tosyl group. Treatment of
intermediate 34a with methanesulfonyl chloride and TsCl produced
intermediates 36a and 36b, respectively, which were subsequently
deprotected to achieve the anticipated sulfamide products 37a and
37b (Scheme 5).
selective methylation of the C7-hydroxyl group of compound 1
produced inactivity (22). (Table 1).
Further modification of compound 1 included elongation of the
carbon chain length on the C5-position, introduction of other atoms
into the carbon chain of the C5-position, and introduction of cyclic
alkyls, aryls and heterocycles into the C5-position directly or with
an interval of one or two carbons; unfortunately, none of the above
methods resulted in satisfactory compounds. Investigation of rat
pheochromocytoma PC12 cell viability indicated that many com-
pounds (33a, 33b, 33d, 33e, 33g, 33h, 33jel, 33n, 33p, 33q, 33sev
and 37b) were cell proliferative activators for an unknown reason.
Two compounds (33c and 33m) exhibited significant cytotoxicity.
The remaining compounds (33f, 33i, 33o, 33r, 35a, 35b and 37a) did
not exhibit obvious protection in the tested assays (Table 2). All of
the above syntheses and analysis proved that compound 20 was a
potent new agent that effectively protected against damage of rat
pheochromocytoma PC12 cells in vitro.
2.2. Neuroprotective activity in vitro
All synthesized compounds were evaluated for neuroprotective
activity on damaged PC12 cells caused by OGD and proliferative
effects on normal PC12 cells at a single dose of 10 mM.
The initial screening of the stock coumarins (Table S1) showed
that nitrogen heterocycles at the C3-position could elicit neuro-
protective activity and that free hydroxyl groups at the C5- or C7-
position were helpful for the activity. While none of the newly
synthesized compounds with modified piperazinyl groups (6a, 6b,
7, 8a and 8b) improved the neuroprotective activity. Those com-
pounds (11, 17a and 17b) whose molecular flexibility was increased
by adding carbon atoms between the coumarin nucleus and the
piperazinyl group were also inactive. Selective methylation of the
C5- or C7-hydroxyl group of compound 1 resulted in different po-
tency. Selective methylation of the C5-hydroxyl group of compound
1 generated significant neuroprotective activity (20). However,
2.3. Neuroprotective activity in vivo
Being the most neuroprotective agent in vitro and without cell
proliferative effect, compound 20 was selected to treat the middle
cerebral artery occlusion (MCAO) rats to investigate the in vivo
neuroprotective activity.
A
B
Scheme 2. Synthesis of coumarin analogs 11 and 17. Reagents and conditions: (a) PBr3, benzene, reflux; (b) 1-methylpiperazine, K2CO3, KI, butanone, reflux; (c) EtONa, ethanol
(EtOH), rt / reflux; (d) phloroglucinol, HCl/methanol (MeOH), rt; (e) K2CO3, EtOH/H2O, reflux; (f) 1-methylpiperazine, BOP, TEA, THF, rt.

42
M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
A
B
Scheme 3. Synthesis of coumarin analogs 20 and 21. Reagents and conditions: (a) BF3$Et2O, EtOH, reflux; (b) Me2SO4, K2CO3, acetone, rt; (c) TsCl, pyridine, rt; (d) BnBr, K2CO3,
acetone, rt; (e) TBAF, THF, 0 ꢁC; (f) Me2SO4, K2CO3, acetone, reflux; (g) TBAF, THF, rt; (h) Pd/C, HCOONH4, THF, reflux.
2.3.1. Protection by compound 20 against rat ischemic brain injury
Injection of compound 20 using ethanol as cosolvent improved
neurological function and reduced infarct size of rats at 24 h after
reperfusion in a dose-dependent manner (Fig. 3). The neurological
score of animals treated with compound 20 at doses of 1.5, 3.0 and
6.0 mg/kg was significantly lower than that of the vehicle control
group (2.44 ꢂ 0.81, 2.00 ꢂ 0.47 and 1.84 ꢂ 0.80, respectively, vs
3.00 ꢂ 0.87, Fig. 3C). Compound 20 at doses of 3.0 and 6.0 mg/kg
significantly reduced the percentage of the infarct area within the
ipsilateral hemispheres compared with the vehicle control group
(20.36 ꢂ 7.9% and 18.89 ꢂ 5.83%, respectively). The same result was
observed for Edaravone at 6.0 mg/kg (Fig. 3A and B), with which the
significant reduction of brain-water content also occurred (Fig. 3D).
that neurons were clear and moderately sized with normal
microstructure in untreated rats (Fig. 4a1 and c1). In contrast, the
brain regions in MCAO rats were observed with significant
neuronal shrinkage and dark staining (Fig. 4a2 and c2). These
pathological changes were prominently suppressed in rats treated
with compound 20 in a dose-dependent manner in the hippo-
campus and cortex (Fig. 4a3, a4, a5 and c3, c4, c5). Terminal
deoxynucleotidyl transferase-mediated dUTP nick end-labeling
(TUNEL), which is based on the specific binding of terminal deox-
ynucleotidyl transferase to the 30-OH ends of fragmented DNA, has
become the most widely used in situ test for the study of apoptosis.
Fig. 4(B and D) shows the results of TUNEL staining; the number of
TUNEL-positive cells increased significantly in the hippocampus
(Fig. 4b2) and cortex (Fig. 4d2) of MCAO rats showing apoptosis.
A remarkable decrease in TUNEL-positive cells was observed in
20-treated rat groups (Fig. 4b3, b4, b5 and d3, d4, d5). An ultra-
structural study also provided strong evidence of neuronal damage
(Fig. 5B). A high-electron-concentration neuronal nucleus was
observed in MCAO rats, and the morphology was improved in
20-treated groups (Fig. 5C, D and E).
2.3.2. Effect of compound 20 on neuronal loss in the cortex and
hippocampus
Typical neuropathological changes were observed in the cortex
and the CA1 area of the hippocampus after MCAO. Fig. 4(A and C)
shows representative photographs of Nissl staining from the hip-
pocampus and cerebral cortex. The results of Nissl staining showed

M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
43
Scheme 4. Synthesis of target compounds 33. Reagents and conditions: (a) ROH, PPh3, DEAD, THF, rt; (b) TBAF, THF, rt.
2.3.3. Effect of compound 20 on the cell apoptosis-related proteins
To explore the mechanism of protection against neuronal loss of
compound 20 in MCAO rats, the cell apoptosis-related proteins were
investigated. It is well known that apoptosis is actively regulated by
several factors, including the decrease of antiapoptotic proteins such
as Bcl-2, the increase of proapoptotic proteins such as Bax [24] and
the activation of caspase-3, which is a major executioner of apoptotic
signals [25,26]. The results (Fig. 6) showed that Bcl-2 expression
Scheme 5. Synthesis of target compounds 35 and 37. Reagents and conditions: (a) CF3COOH, DCM, 40 ꢁC; (b) TBAF, THF, 50 ꢁC; (c) RSO2Cl, TEA, DCM, rt.

44
M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
Table 1
compound 20 exhibits anti-ischemia activity by blocking the
apoptosis pathway. Further studies on the antiapoptotic effect and its
mechanism will be reported in detail in another journal.
Neuroprotection of coumarin derivatives at a single dose (10
pheochromocytoma PC12 cells.
m
M) in damaged rat
OGD
Compound
Proliferationa
Controlb
Modelc
1
6a
6b
7
8a
8b
11
100.0 ꢂ 3.4
e
100.0 ꢂ 5.9
44.2 ꢂ 0.7##
52.7 ꢂ 3.7*
48.4 ꢂ 0.5*
49.3 ꢂ 1.9*
46.3 ꢂ 0.8
48.9 ꢂ 1.2*
48.8 ꢂ 0.9*
45.9 ꢂ 0.9
46.7 ꢂ 2.1
46.1 ꢂ 0.4
66.0 ꢂ 2.8**
46.1 ꢂ 0.4
2.4. Pharmacokinetic evaluation of compound 20
107.4 ꢂ 1.2
98.1 ꢂ 1.7
97.4 ꢂ 1.5
104.1 ꢂ 3.2
103.3 ꢂ 4.0
101.1 ꢂ 1.8
102.3 ꢂ 1.4
97.7 ꢂ 2.1
97.5 ꢂ 3.7
100.7 ꢂ 5.4
99.8 ꢂ 3.2
To obtain information on the in vivo behavior of compound 20,
its pharmacokinetic properties were studied. After intravenous
administration of compound 20 to male Sprague Dawley rats
(6 mg/kg) (n ¼ 4), plasma and cortex samples were collected at
intervals of 0, 2, 3.5, 3.75, 7.5, 15, 30 and 60 min. Samples were
analyzed by LCeMS. The plasma and cortex concentrationetime
profiles of compound 20 are presented in Fig. 7. All of the phar-
macokinetic parameters are listed in Table 3. The results showed
that the Cmax values were 13.309 and 0.122 mg/L in plasma and in
the cortex at 2 min, respectively. The concentration of compound
20 in the plasma or the cortex was very low at 60 min. These
results indicated that compound 20 was capable of crossing the
bloodꢀbrain barrier (BBB) of Sprague Dawley rats.
17a
17b
20
21
*p < 0.05 vs model group.
**p < 0.01 vs model group.
##p < 0.01 vs control group (n ¼ 3).
a
Cell proliferation was performed in a full-culture medium and 10 mM tested
compounds.
b
c
A full-culture medium was used for the control group.
OGD assay: the cells were treated with low-glucose medium and 5 mM sodium
3. Conclusions
dithionite for 24 h.
In summary, 36 new coumarin derivatives were designed and
synthesized in this study. A SARs study of these compounds was
carried out, and a potent neuroprotective compound was discov-
ered. Compound 20 significantly attenuated the PC12 cell damage
induced by OGD in vitro, and exhibited clear neuroprotection in
MCAO rats by reducing infarct size and brain-water content,
improving neurological function, and suppressing neuronal loss
and neuropathological changes in the cortex and hippocampus.
Pharmacokinetic evaluation indicated that compound 20 was able
to penetrate the BBB of rats. Mechanistic studies indicated that
compound 20 acted against the apoptosis of nerve cells induced by
oxygen and glucose deprivation/reoxygenation.
decreased significantly compared with the control in both the cortex
and the hippocampus, while compound 20 promoted the expression
of Bcl-2. Remarkable increases in Bax and cleaved caspase-3 were
also observed in MCAO rats; however, compound 20 inhibited the
expression of Bax and cleaved caspase-3. These results indicate that
Table 2
Neuroprotection of 5-hydroxyl group derivatives of compound 1 at a single dose
(10 mM) in damaged rat pheochromocytoma PC12 cells.
Compound
Controlb
Proliferationa
OGD
100.0 ꢂ 3.4
100.0 ꢂ 2.5
Modelc
20
e
44.2 ꢂ 0.7##
66.0 ꢂ 2.8**
33.9 ꢂ 5.7
46.2 ꢂ 3.9
15.6 ꢂ 1.2**
51.8 ꢂ 1.6
36.3 ꢂ 4.9
61.5 ꢂ 4.8**
33.8 ꢂ 6.3
36.1 ꢂ 4.7
52.0 ꢂ 2.0
19.9 ꢂ 3.7**
51.5 ꢂ 2.8
37.7 ꢂ 3.3
27.3 ꢂ 2.2*
44.2 ꢂ 5.9
54.7 ꢂ 2.8*
48.2 ꢂ 2.8
24.9 ꢂ 4.2*
56.9 ꢂ 1.1*
39.1 ꢂ 4.7
48.2 ꢂ 1.5
23.4 ꢂ 2.6*
15.8 ꢂ 0.3**
49.2 ꢂ 1.2
40.5 ꢂ 4.2
40.3 ꢂ 0.6
45.1 ꢂ 3.3
4. Experimental protocols
100.7 ꢂ 5.4
167.1 ꢂ 2.9**
197.2 ꢂ 7.7**
59.4 ꢂ 3.0**
119.1 ꢂ 3.4**
196.9 ꢂ 8.9**
103.9 ꢂ 5.3
181.1 ꢂ 4.3**
176.8 ꢂ 0.9**
106.9 ꢂ 3.2
138.7 ꢂ 7.8**
137.3 ꢂ 7.8**
184.4 ꢂ 6.0**
51.6 ꢂ 2.8**
115.5 ꢂ 4.5**
108.1 ꢂ 4.9*
134.7 ꢂ 5.6**
144.1 ꢂ 5.6**
107.5 ꢂ 2.3
177.9 ꢂ 8.4**
152.1 ꢂ 2.4**
131.8 ꢂ 4.8**
153.3 ꢂ 2.8**
103.6 ꢂ 5.5
107.4 ꢂ 4.4
103.0 ꢂ 2.9
140.1 ꢂ 3.8**
33a
33b
33c
33d
33e
33f
33g
33h
33i
4.1. Chemistry
All materials used were of commercial grade without purifica-
tion unless otherwise specified. All nuclear magnetic resonance
(NMR) experiments were carried out on a Varian Mercury 300 or
400 or 500 or 600 MHz spectrometer or a Bruker Avance 400 MHz
spectrometer using dimethylsulfoxide-d6 (DMSO-d6), methanol-d4
and acetone-d6 as the solvent. Chemical shifts were reported in
33j
33k
33l
ppm (d) relative to the solvent, and coupling constants (J) were
reported in Hz. Melting points were determined without correction
with a Yanaco micromelting point apparatus. HPLC-MS analysis
was performed on a Thermo Finnigan LCQ Advantage mass spec-
trometer equipped with an Agilent pump, an Agilent detector, an
Agilent liquid handler, and a fluent splitter. The column used was a
33m
33n
33o
33p
33q
33r
33s
33t
33u
33v
35a
35b
37a
37b
Kromasil C18 column (4.6
mm, 4.6 mm ꢃ 50 mm) from DIKMA for
analysis. The eluent was a mixture of acetonitrile and water con-
taining 0.05% HCOOH with a linear gradient from 5:95 (v/v) to 95:5
(v/v) of acetonitrileeH2O within 5 min at a 1.0 mL/min flow rate for
analysis. The UV detection was carried out at a UV wavelength of
254 nm. The 5% of the eluent was split into the MS system. Mass
spectra were recorded in either positive or negative ion mode using
electrospray ionization (ESI). High resolution LC-MS (HRMS) was
carried out by Agilent LC/MSD TOF using a column of Agilent
*p < 0.05 vs model group.
**p < 0.01 vs model group.
##p < 0.01 vs control group (n ¼ 3).
ZORBAX SBeC18 (rapid resolution, 3.5
m
m, 2.1 mm ꢃ 30 mm) at a
a
Cell proliferation was performed in a full-culture medium and 10 mM tested
flow of 0.40 mL/min. The solvent is methanol/water ¼ 75:25 (v/v)
containing 5 mmol/L ammonium formate. The ion source is elec-
trospray ionization (ESI). Flash column chromatography was per-
formed with silica gel 60 (200e300 mesh) from Qingdao Haiyang
compounds.
b
c
A full-culture medium was used for the control group.
OGD assay: the cells were treated with low-glucose medium and 5 mM sodium
dithionite for 24 h.

M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
45
Fig. 3. Compound 20 effectively protected against rat ischemic brain injury. (A and B) Brain infarct size, (C) Neurological score and (D) brain-water content at 24 h after ischemia.
After the rats underwent MCAO and were treated with compound 20, edaravone or saline, they were scored by Longa’s 5-point scale, and then the rat brains were stained with TTC.
Finally, the brain-water content was determined. Data are expressed as the mean ꢂ SD: (*) p < 0.05 vs vehicle group; (**) p < 0.01 vs vehicle group; (##) p < 0.01 vs normal group
(n ¼ 12).
Fig. 4. Compound 20 decreased the neuronal loss in both the cerebral cortex and the hippocampus. After the rats underwent MCAO and were treated with compound 20, Edaravone
or saline, the rat brains were perfused and fixed with 4% buffered paraformaldehyde. Neuronal loss was observed by Nissl staining and TUNEL staining. (A) Hippocampus by Nissl
staining; (B) hippocampus by TUNEL staining; (C) cerebral cortex by Nissl staining; (D) cerebral cortex by TUNEL staining. Parts a1, b1, c1 and d1 were from the untreated group.
Parts a2, b2, c2 and d2 were from the vehicle control group. Parts a3, b3, c3 and d3 were from the 1.5 mg/kg 20 group. Parts a4, b4, c4 and d4 were from the 3.0 mg/kg 20 group.
Parts a5, b5, c5 and d5 were from the 6.0 mg/kg 20 group. Parts a6, b6, c6 and d6 were from the 6.0 mg/kg Edaravone group (magnification 200ꢃ).

46
M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
Fig. 5. The ultrastructural neuropathological changes were significantly suppressed by compound 20, as observed by electron microscopy. After the rats underwent MCAO and were
treated with compound 20, Edaravone or saline, the rat brains were perfused, fixed and embedded in Epon resin. Ultrathin sections were stained with uranyl acetate and lead
citrate, and the ultrastructure was observed using electron microscopy. (A) Untreated group; (B) vehicle control group; (C) compound 20 (1.5 mg/kg)-treated group; (D) compound
20 (3.0 mg/kg)-treated group; (E) compound 20 (6.0 mg/kg)-treated group; (F) Edaravone (6.0 mg/kg)-treated group (magnification 12000ꢃ).
Fig. 6. Effect of compound 20 on Bcl-2, Bax and cleaved caspase-3 expression in the cerebral cortex and hippocampus. After the rats underwent MCAO and were treated with
compound 20, Edaravone or saline, the cerebral cortex and hippocampus were separated. The expression of Bcl-2, Bax and cleaved caspase-3 was detected by Western blot analysis.
(A) Hippocampus; (B) Cerebral cortex. Quantitative analysis of the ratio of Bcl-2/Bax, the expression of cleaved caspase-3 by Science Lab 2005 Image Gauge software. (n ¼ 4), (*)
p < 0.05 vs vehicle group; (**) p < 0.01 vs vehicle group; (##) p < 0.01 vs normal group).

M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
47
Fig. 7. Mean plasma (A) and mean cortex (B) concentrationetime curves after a single intravenous dose of compound 20 (6 mg/kg, n ¼ 4).
Chemical Factory. The tested compounds were purified until the
purity was ꢄ95%, detected by HPLC under UV 254 nm wavelength,
NMR, melting point, and HRMS.
was stirred at 50 ꢁC. After the reaction was completed, the solvent
was evaporated in vacuo. The final product 7 was obtained as white
powder after purification by silica gel column chromatography
eluting with DCMeCH3OH (20:1, v/v). Yield 51%; mp 210e213 ꢁC. 1H
4.1.1. General procedure for the synthesis of compounds 6a and 6b
The key intermediate 5 was prepared according to the literature
[20]. To a solution of 5 (0.1 mmol) in 5 mL of dry THF was added
various acid anhydrides (0.1 mmol). The reaction mixture was
stirred at room temperature for 30 min and then got clear. After the
reaction was completed monitored by LC-MS analysis, the solvent
was evaporated in vacuo. The final products were characterized
after purification by silica gel column chromatography.
NMR (300 MHz, DMSO-d6) d 10.27 (s, 2H), 8.58 (s, 1H), 7.48 (dd,
J ¼ 8.7 Hz, 4.8 Hz, 2H), 7.07 (dd, J ¼ 8.7 Hz, 4.8 Hz, 2H), 6.27 (d,
J ¼ 1.5 Hz, 1H), 6.13 (d, J ¼ 2.1 Hz, 1H), 4.09e2.80 (m, 8H), 2.69 (s,
3H). 13C NMR (100 MHz, DMSO-d6)
d 160.2,158.0,157.9,155.0,154.5,
150.0, 136.9, 136.9, 128.2, 121.3, 121.2, 114.9, 114.6, 102.3, 99.3, 94.1,
49.1, 44.5, 17.0. HRMS calcd for C21H21FN3O5 (M þ Hþ) 414.1465;
found. 414.1457.
4.1.3. General procedure for the synthesis of compounds 8a and 8b
To a solution of 5 (0.1 mmol) in 5 mL of dry THF was added
isothiocyanate (0.12 mmol). The reaction mixture was stirred at
50 ꢁC. After the reaction was completed, the solvent was evapo-
rated in vacuo. The final product was obtained after purification by
silica gel column chromatography.
4.1.1.1. 3-(4-Acetylpiperazin-1-yl)-5,7-dihydroxy-4-methyl-2H-chro-
men-2-one (6a). This compound was obtained as green-yellow
powder in 85％ yield. mp > 300 ꢁC. 1H NMR (400 MHz, DMSO-
d6)
d
10.41 (s, 1H), 10.18 (s, 1H), 6.25 (d, J ¼ 1.6 Hz, 1H), 6.13 (d,
J ¼ 1.6 Hz, 1H), 3.8e2.6 (m, 8H), 2.65 (s, 3H), 2.01 (s, 3H). 13C NMR
(125 MHz, DMSO-d6) d 168.10, 160.17, 157.95, 157.88, 154.50, 149.92,
128.11, 102.20, 99.30, 94.09, 49.43, 48.98, 46.44, 41.51, 21.22, 16.95.
4.1.3.1. N-butyl-4-(5,7-dihydroxy-4-methyl-2-oxo-2H-chromen-3-yl)
piperazine-1-carbothioamide (8a). This compound was obtained as
off-white powder in 48％ yield; mp 245e248 ꢁC. 1H NMR
HRMS calcd for C16H19N2O5 (M þ Hþ) 319.1294; found 319.1289.
4.1.1.2. 5,7-Dihydroxy-4-methyl-3-(4-(2,2,2-trifluoroacetyl)piper-
azin-1-yl)-2H-chromen-2-one (6b). This compound was obtained
as pale-yellow powder in 88％ yield. mp 274e275 ꢁC. 1H NMR
(300 MHz, DMSO-d6)
d
10.42 (s, 1H), 10.19 (s, 1H) 7.64 (t, J ¼ 5.1 Hz,
1H), 6.25 (d, J ¼ 2.4 Hz, 1H), 6.12 (d, J ¼ 2.4 Hz, 1H), 4.5e2.5 (m, 8H),
3.52e3.46 (m, 2H), 2.66 (s, 3H), 1.56e1.47 (m, 2H),1.34e1.10 (m,
(300 MHz, DMSO-d6)
1H), 6.13 (d, J ¼ 2.1 Hz, 1H), 4.23e3.57 (m, 4H), 3.15e2.80 (m, 4H),
2.66 (s, 3H). 13C NMR (125 MHz, DMSO-d6)
160.3, 158.0, 157.9,
d
10.46 (s, 1H), 10.22 (s, 1H), 6.26 (d, J ¼ 2.1 Hz,
2H), 0.88 (t, J ¼ 7.2 Hz, 3H). 13C NMR (100 MHz, DMSO-d6)
d 181.22,
161.80, 160.68, 158.14, 155.26, 151.14, 126.51, 102.06, 99.87, 92.93,
48.82, 47.98, 45.37, 30.57, 19.62, 17.02, 14.18. HRMS calcd for
d
154.5, 150.3, 127.6, 117.5, 115.2, 102.1, 99.4, 94.1, 49.441, 48.7, 46.2,
43.7, 17.00. HRMS calcd for C16H16F3N2O5 (M þ Hþ) 373.1011; found
373.1012.
C
19H26N3O4S (M þ Hþ) 392.1644; found. 392.1632.
4.1.3.2. 4-(5,7-Dihydroxy-4-methyl-2-oxo-2H-chromen-3-yl)-N-(3-
fluorophenyl)piperazine-1-carbothioamide (8b). This compounds
was obtained as pale-yellow powder in 38％ yield; mp 282e284 ꢁC.
4.1.2. Synthesis of 4-(5,7-dihydroxy-4-methyl-2-oxo-2H-chromen-
3-yl)-N-(4-fluorophenyl)piperazine-1-carboxamide (7)
1H NMR (400 MHz, DMSO-d6)
d 10.45 (s, 1H), 10.21 (s, 1H), 9.45 (s,
To a solution of 5 (0.1 mmol) in dry THF 5 mL was added
1-fluoro-4-isocyanatobenzene (0.12 mmol). The reaction mixture
1H), 7.35e7.24 (m, 2H), 7.16 (d, J ¼ 8.0 Hz,1H), 6.91 (t, J ¼ 7.8 Hz,1H),
6.28 (d, J ¼ 2.1 Hz, 1H), 6.14 (d, J ¼ 2.1 Hz, 1H), 2.70 (s, 3H). 13C NMR
(100 MHz, DMSO-d6)
d
181.39, 162.09 (d, J ¼ 240 Hz, 1C), 161.07,
159.08, 158.48, 155.10, 150.96, 143.42 (d, J ¼ 11 Hz, 1C), 129.84 (d,
J ¼ 10 Hz, 1C), 127.93, 120.75 (d, J ¼ 2 Hz, 1C), 111.88 (d, J ¼ 24 Hz,
1C),110.90 (d, J ¼ 20 Hz,1C),102.67, 99.94, 94.33, 49.49, 49.37,17.53.
Table 3
Obtained pharmacokinetic parameters of compound 20 after a single intravenous
dose (2 ꢃ 10ꢀ5 mol/kg) in the rat plasma and cortex.
HRMS calcd for
430.12199.
C
21H20FN3O4S (M
þ
Hþ) 430.12161; found.
Parametersa
Cmax
tmax
t1/2 (min) AUC (min mg/L) Vd (L)
CL
(mg/L) (min)
(L/min)
4.1.4. Synthesis of 5,7-dihydroxy-4-methyl-3-(2-(4-methylpiperazin-
1-yl)ethyl)-2H-chromen-2-one (11)
In plasma 13.309 2.0
6.514
15.526
172.147
2.544
0.084 0.009
12.273 0.558
In cortex
0.122 2.0
To a stirred solution of pholoroglucinol (2 mmol) in benzene
(5 mL), 9 (2 mmol) and PBr3 (6 mmol) were added at room tem-
perature. The reaction mixture refluxed for 1 h and some precipi-
tation was formed. Analyzed by LC-MS the solid was demonstrated
a
Cmax: maximum measured concentration; tmax: time to reach the maximum
measured concentration; t1/2: half-life of the terminal slope of the concentratione
time curve; AUC: area under the concentrationetime curve; Vd: volume of distri-
bution; CL: clearance.

48
M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
as the target intermediate 10 (460 mg, in 77% yield). mp > 252 ꢁC
decomposed. 1H NMR (300 MHz, DMSO-d6)
10.53 (s, 1H), 10.25 (s,
1H), 6.13 (d, J ¼ 2.4 Hz, 1H), 3.58 (s, 3H), 2.74 (t, J ¼ 8.4 Hz, 2H), 2.52
(s, 3H), 2.41 (t, J ¼ 8.4 Hz, 2H).
d
1H), 9.45 (s, 1H), 6.28 (d, J ¼ 2.4 Hz, 1H), 6.15 (d, J ¼ 2.4 Hz, 1H), 3.53
The intermediate 16b was obtained as white powder in 57%
(t, J ¼ 7.5 Hz, 2H), 3.04 (t, J ¼ 7.5 Hz, 2H), 2.55 (s, 3H). 13C NMR
yield; 1H NMR (300 MHz, DMSO-d6)
d
6.21 (d, J ¼ 1.8 Hz, 1H), 5.98
(100 MHz, DMSO-d6)
d
160.6, 160.5, 157.8, 154.9, 151.2, 116.2, 102.4,
(d, J ¼ 2.1 Hz, 1H), 2.63 (t, J ¼ 8.1 Hz, 2H), 2.52 (s, 3H), 2.06 (t,
99.5, 94.3, 31.5, 30.1, 19.1.
J ¼ 8.1 Hz, 2H). 13C NMR (100 MHz, DMSO-d6)
d 175.7, 161.3, 161.2,
Intermediate 10 (1 mmol) was dissolved in 2-butanone (15 mL).
K2CO3 (5 mmol), KI (1 mmol) and 1-methylpiperazine (5 mmol)
were added to the above solution and the reaction mixture was
refluxed with stirring for 1.5 h. After the reaction was completed,
the solvent was evaporated in vacuo. The final product 11 was
obtained as yellow powder after purification by silica gel column
chromatography eluting with DCMeCH3OH (10:1, v/v). Yield 75%;
154.9, 149.9, 142.5, 117.7, 102.5, 100.1, 92.8, 36.1, 23.7, 18.5.
The final product 17b was obtained as off-white powder. Yield
79%; mp > 250 ꢁC decomposed. 1H NMR (300 MHz, DMSO-d6)
d
10.29 (brs, 2H), 6.23 (d, J ¼ 2.4 Hz, 1H), 6.11 (d, J ¼ 2.1 Hz, 1H),
3.39e3.38 (m, 4H), 2.66 (t, J ¼ 7.2 Hz, 2H), 2.49 (s, 3H), 2.37 (t,
J ¼ 7.2 Hz, 2H), 2.20e2.17 (m, 4H), 2.11 (s, 3H). 13C NMR (100 MHz,
DMSO-d6)
d 169.9, 157.5, 157.1, 154.8, 154.1, 149.6, 118.2, 102.2, 99.4,
mp > 225 ꢁC decomposed. 1H NMR (400 MHz, DMSO-d6)
d
10.45 (s,
94.2, 54.5, 54.1, 45.3, 44.6, 40.7, 31.1, 22.7, 18.7. HRMS calcd for
1H),10.18 (s,1H), 6.27 (d, J ¼ 1.8 Hz, 1H), 6.14 (d, J ¼ 1.7 Hz,1H), 3.2e
C
18H23N2O5 (M þ Hþ) 347.1067; found. 347.1596.
2.2 (m, 8H), 2.67 (t, J ¼ 6.9 Hz, 2H), 2.54 (s, 3H), 2.44 (s, 3H), 2.44(t,
J ¼ 6.9 Hz, 2H). 13C NMR (100 MHz, DMSO-d6)
d
169.9, 160.1, 157.5,
4.1.7. Synthesis of 7-hydroxy-5-methoxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (20)
154.8, 149.8, 117.0, 102.5, 99.4, 94.2, 55.9, 53.8, 51.4, 44.5, 23.8, 18.9.
HRMS calcd for C17H23N2O4 (M þ Hþ) 319.1658; found. 319.1646.
The intermediate compound 1 was prepared in a similar manner
to the synthesis of compound 5, substituting piperazine with 1-
methylpiperazine. To a solution of 1 (20 mmol) in pyridine
(20 mL) at 0 ꢁC was added TsCl (50 mmol) slowly. The mixture was
stirred at room temperature for 2 h. When the reaction was
completed as monitored by LC-MS, the pyridine was evaporated in
vacuo and water (50 mL) was added to the residue followed by
extraction with ethyl acetate (3 ꢃ 30 mL). The organic layers were
combined, dried over anhydrous MgSO4, and evaporated in vacuo.
Intermediate 23 was obtained as yellow powder after chromatog-
raphy by silica gel eluting with DCMeCH3OH (20:1, v/v). Yield 89%;
4.1.5. Synthesis of 5,7-dihydroxy-4-methyl-3-(2-(4-
methylpiperazin-1-yl)-2-oxoethyl)-2H-chromen-2-one (17a)
To a solution of 12 (3 mmol) in absolute ethanol (5 mL) was
added ethylate sodium (2 mmol) in absolute ethanol (2 mL). The
mixture was heated to 60 ꢁC slowly followed by the addition of 13a
(2 mmol) in ethanol (2 mL). The reaction was stirred at room
temperature for 2 d and then was refluxed for 2 h. Diethyl ether
(Et2O) was added to the mixture when cooled to room temperature.
After filtration and concentration, brown oil 14a was obtained
without purification. 10 mL of methanol saturated with dry HCl gas
was added to the mixture of pholoroglucinol (2 mmol) and 14a. The
mixture was stirred at room temperature until pholoroglucinol was
completely disappeared monitored by LC-MS. The mixture was
concentrated in vacuo. Intermediate 15a was obtained as brown
powder after recrystallization in ethanol. Yield 37% for two steps;
1H NMR (400 MHz, DMSO-d6)
J ¼ 8.3 Hz, 2H), 7.48 (d, J ¼ 8.2 Hz, 2H), 7.44 (d, J ¼ 8.2 Hz, 2H), 7.08
(d, J ¼ 2.4 Hz, 1H), 6.69 (d, J ¼ 2.4 Hz, 1H), 2.88 (s, 4H), 2.40 (s, 6H),
2.34 (s, 4H), 2.32 (s, 3H), 2.17 (s, 3H). 13C NMR (100 MHz, DMSO-d6)
d
7.75 (d, J ¼ 8.3 Hz, 2H), 7.62 (d,
d
156.2, 151.6, 147.9, 146.7, 146.4, 145.6, 139.1, 135.6, 130.6, 130.6,
130.4, 128.3, 128.2, 114.9, 112.6, 109.2, 55.1, 49.1, 46.0, 21.2, 21.1, 17.2.
To a solution of compound 23 (15 mmol) in acetone (100 mL)
was added K2CO3 (22.5 mmol) and BnBr (15 mmol) with stirring at
room temperature. When the reaction was completed as monitored
by LC-MS, the insoluble substrate was filtered off and the filtrate
was evaporated in vacuo. The intermediate 24 was obtained as
light-yellow powder after chromatography by silica gel eluting
with DCMeCH3OH (20:1, v/v). Yield 87%; 1H NMR (300 MHz,
1H NMR (400 MHz, DMSO-d6)
d 10.55 (s, 1H), 10.27 (s, 1H), 6.29 (d,
J ¼ 2.3,1H), 6.17 (d, J ¼ 2.3,1H), 3.61 (s, 3H), 3.59 (s, 2H), 2.49 (s, 3H).
Intermediate 15a (0.5 mmol) and K2CO3 (1 mmol) were dis-
solved in 10 mL of ethanol/H2O (1:1, v/v). The mixture was refluxed
for 3 h, diluted by H2O (5 mL) followed by neutralization with 1 N
HCl until pH 3e4, and then left overnight. Intermediate 16a was
obtained as pale-yellow needle crystal after filtration. Yield 68%; 1H
DMSO-d6)
d 7.78e7.71 (m, 4H), 7.65e7.58 (m, 2H), 7.58e7.47 (m,
NMR (300 MHz, DMSO-d6) d 12.34 (s,1H),10.54 (s,1H),10.26 (s,1H),
7H), 7.20 (d, J ¼ 2.4, 1H), 6.69 (d, J ¼ 2.4, 1H), 4.74 (s, 2H), 3.67 (t,
J ¼ 10.8, 2H), 3.52 (t, J ¼ 12.5, 2H), 3.41e3.37 (m, 2H), 3.05 (s, 3H),
3.02 (brs, 2H), 2.48 (s, 3H), 2.45 (s, 3H), 2.44 (s, 3H).
6.29 (d, J ¼ 1.8 Hz, 1H), 6.16 (d, J ¼ 1.8 Hz, 1H), 3.49 (s, 3H), 2.47
(s, 2H).
To a solution of 16a (0.2 mmol), BOP (0.24 mmol) and TEA
(0.3 mmol) in THF (5 mL) was added 1-methylpiperazine
(0.4 mmol). The mixture was stirred at room temperature for
1.5 h and some precipitate was formed. After filtration compound
17a was obtained as white powder. Yield 78%; mp > 300 ꢁC
Intermediate 24 (12 mmol) was dissolved THF (100 mL). The
mixture was added with TBAF (12 mmol) in ice bath and kept at
0 ꢁC for 4 h with stirring. Then the solvent was evaporated in vacuo
and the intermediate 25 was obtained as yellow powder after
chromatography by silica gel eluting with DCMeCH3OH (15:1, v/v).
decomposed. 1H NMR (300 MHz, DMSO-d6)
d 10.48 (s, 1H), 10.23 (s,
Yield 87%; 1H NMR (300 MHz, DMSO-d6)
7.62e7.53 (m, 5H), 7.42 (d, J ¼ 8.1, 2H), 6.17 (d, J ¼ 1.5, 1H), 6.01 (d,
J ¼ 1.9, 1H), 4.71 (s, 2H), 3.65 (brs, 4H), 3.39 (brs, 4H), 2.99 (s, 3H),
2.72 (s, 3H), 2.37 (s, 3H).
d
7.74 (d, J ¼ 8.3, 2H),
1H), 6.29 (d, J ¼ 1.8 Hz, 1H), 6.15 (d, J ¼ 2.1 Hz, 1H), 3.57 (s, 2H), 3.57
(brs, 2H), 3.42 (brs, 2H), 2.39 (s, 3H), 2.34 (brs, 2H), 2.24 (brs, 2H),
2.19 (s, 3H). 13C NMR (100 MHz, DMSO-d6)
d 167.62, 161.01, 160.19,
157.60, 154.89, 150.89, 114.75, 102.54, 99.45, 94.21, 54.81, 54.36,
45.63, 44.79, 41.30, 30.91, 19.28. HRMS calcd for C17H21N2O5
(M þ Hþ) 333.1450; found. 333.1439.
To a solution of 25 (9 mmol) in acetone (70 mL) was added K2CO3
(15 mmol) and Me2SO4 (15 mmol) successively with stirring at room
temperature. 5 min later, the mixture was refluxed for 2 h. When the
reaction was completed as monitored by LC-MS, the insoluble sub-
strate was filtered off and the filtrate was evaporated in vacuo. The
intermediate 26 was obtained as yellow powder after chromatog-
raphy by silica gel eluting with DCMeCH3OH (15:1, v/v). Yield 82%;
4.1.6. Synthesis of 5,7-dihydroxy-4-methyl-3-(3-(4-
methylpiperazin-1-yl)-3-oxopropyl)-2H-chromen-2-one (17b)
Compound 17b was prepared in a similar manner to the syn-
thesis of compound 17a, substituting 13a with 13b. The interme-
diate 15b was obtained as pale-yellow in 33% yield; 1H NMR
1H NMR (300 MHz, DMSO-d6)
d
7.82 (d, J ¼ 8.3, 2H), 7.65e7.46 (m,
7H), 6.68 (d, J ¼ 2.3,1H), 6.62 (d, J ¼ 2.3,1H), 4.74 (s, 2H), 3.76 (s, 3H),
(300 MHz, DMSO-d6)
d
10.45 (s, 1H), 10.18 (s, 1H), 6.25 (d, J ¼ 2.1 Hz,
3.67e3.37 (m, 8H), 3.00 (s, 3H), 2.60 (s, 3H), 2.42 (s, 3H).

M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
49
TBAF (12 mmol) was added to the solution of 26 (6 mmol) in THF
(30 mL) and the mixture was stirred at room temperature. When the
reaction was completed as monitored by LC-MS, the solvent was
evaporated in vacuo. The intermediate 27 was obtained as yellow
powder after chromatography by silica gel eluting with DCMeCH3OH
mp > 218 ꢁC decomposed. 1H NMR (400 MHz, DMSO-d6)
d
10.57
(s, 1H), 6.37 (d, J ¼ 1.6 Hz, 1H), 6.31 (d, J ¼ 1.6 Hz, 1H), 4.06 (q,
J ¼ 6.8 Hz, 2H), 2.76 (s, 3H), 2.64 (s, 3H), 1.40 (t, J ¼ 6.8 Hz, 3H). 13
C
NMR (150 MHz, DMSO-d6)
d 160.4, 158.4, 157.6, 154.2, 148.7, 129.2,
103.1, 96.6, 94.9, 64.4, 55.3, 48.8, 46.0, 17.2, 14.3. HRMS calcd for
(15:1, v/v). Yield 85%; 1H NMR (400 MHz, DMSO-d6)
d
7.59e7.52 (m,
C
17H23N2O4 (M þ Hþ) 319.1658; found 319.1639.
5H), 6.02 (s, 1H), 5.85 (s, 1H), 4.67 (s, 2H), 3.71 (s, 3H), 3.68 (brs, 4H),
3.32e3.30 (m, 2H), 2.97 (s, 3H), 2.74e2.72 (m, 2H), 2.56 (s, 3H).
To the solution of 27 (4 mmol) in CH3OH (20 mL) was added
Pd/C (8 mmol) and HCOONH4 (20 mmol). The reaction mixture was
refluxed for 40 min. After filtration, the filtrate was concentrated in
vacuo and the final product 20 was obtained as pale-yellow powder
after chromatography by silica gel eluting with DCMeCH3OH (20:1,
v/v). Yield 91%; mp 199e201 ꢁC. 1H NMR (300 Hz, DMSO-d6)
4.1.9.2. 5-Butoxy-7-hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-
2H-chromen-2-one (33b). 33b was white powder. Yield 85%;
mp > 220 ꢁC decomposed. 1H NMR (400 MHz, DMSO-d6)
d 10.41 (s,
1H), 6.33 (s, 1H), 6.26 (s, 1H), 3.98 (s, 2H), 2.62 (s, 3H), 2.37 (s, 3H),
1.75 (s, 2H), 1.46 (dd, J ¼ 14.1, 7.2 Hz, 2H), 0.94 (t, J ¼ 6.9 Hz, 3H). 13
C
NMR (75 MHz, CD3OD)
d 162.52, 160.68, 160.57, 156.13, 153.16,
128.98, 104.70, 97.56, 96.18, 70.04, 55.70, 48.49, 44.29, 32.28, 20.61,
18.26, 14.19. HRMS calcd for C19H27N2O4 (M þ Hþ) 347.1971; found
347.1965.
d
10.428 (s, 1H), 6.347 (d, J ¼ 2.1 Hz, 1H), 6.266 (d, J ¼ 1.8 Hz, 1H),
3.812 (s, 3H), 3.60e2.20 (m, 8H), 2.586 (s, 3H), 2.204 (s, 3H). 13C
NMR (100 MHz, DMSO-d6) 160.5, 159.2, 157.6, 154.2, 148.6, 129.2,
d
103.1, 96.2, 95.0, 55.9, 55.3, 48.9, 46.1, 17.0. HRMS calcd for
C
4.1.9.3. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-(octy-
16H21N2O4 (M þ Hþ) 305.1501; found. 305.1503.
loxy)-2H-chromen-2-one (33c). 33c was pale-yellow powder. Yield
67%; mp 85e87 ꢁC. 1H NMR (400 MHz, DMSO-d6)
d 10.49 (s, 1H),
4.1.8. Synthesis of 5-hydroxy-7-methoxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (21)
6.35 (d, J ¼ 2.0 Hz, 1H), 6.28 (d, J ¼ 2.0 Hz, 1H), 3.99 (t, J ¼ 6.0 Hz,
2H), 2.90 (brs, 8H), 2.63 (s, 3H), 2.56 (s, 3H), 1.80e1.74 (m, 2H),
1.49e1.41 (m, 2H), 1.29e1.23 (m, 8H), 0.88e0.85 (m, 2H). 13C NMR
The intermediate 28 was prepared from 23 in a similar manner
to the synthesis of 25 from 24 as pale-yellow powder after chro-
matography through silica gel eluting with DCMeCH3OH (20:1,
(100 MHz, DMSO-d6) d 160.68,158.67,157.75,154.34,130.72,128.39,
102.91, 96.84, 95.08, 68.66, 53.86, 47.52, 43.82, 31.19, 28.65, 28.58,
28.48, 25.77, 22.05, 17.34, 13.94. HRMS calcd for C23H35N2O4
(M þ Hþ) 403.2597; found 403.2597.
v/v). Yield 84%; 1H NMR (400 Hz, DMSO-d6)
d
7.76 (d, J ¼ 4.0 Hz,
2H), 7.45 (d, J ¼ 4.0 Hz, 2H), 6.45 (s, 1H), 6.33 (s, 1H), 2.93 (brs, 4H),
2.60 (s, 3H), 2.44 (brs, 4H), 2.23 (s, 3H).
The intermediate 29 was prepared as yellow solid in a similar
manner to the synthesis of 24 from 23. Yield 89%; 1H NMR
4.1.9.4. 7-Hydroxy-5-isopropoxy-4-methyl-3-(4-methylpiperazin-1-
yl)-2H-chromen-2-one (33d). 33d was yellow powder. Yield 85%;
(300 MHz, DMSO-d6)
d
7.79 (d, J ¼ 7.6, 2H), 7.65e7.32 (m, 12H), 6.81
mp 73e75 ꢁC. 1H NMR (400 MHz, DMSO-d6)
d 10.37 (s, 1H), 6.35 (d,
(s,1H), 6.69 (s, 1H), 5.09 (s, 2H), 4.67 (s, 2H), 3.72e3.34 (m, 8H), 2.97
(s, 3H), 2.56 (s, 3H), 2.42 (s, 3H).
J ¼ 2.0 Hz, 1H), 6.24 (d, J ¼ 2.1 Hz, 1H), 4.70e4.61 (m, 1H), 2.62 (s,
3H), 2.33 (s, 3H), 1.34 (s, 2H), 1.32 (s, 2H). 13C NMR (150 MHz,
The intermediate 30 was prepared as light yellow solid in a
DMSO-d6) d 160.4, 157.6, 157.2, 154.4, 148.9, 129.1, 103.6, 97.4, 94.8,
similar manner to the synthesis of 27 from 26. Yield 92%; 1H NMR
70.7, 55.1, 48.6, 45.7, 21.6,17.5. HRMS calcd for C18H25N2O4 (M þ Hþ)
(300 MHz, DMSO-d6)
d
10.58 (s, 1H), 7.68e7.31 (m, 10H), 6.51 (d,
333.1814; found 333.1816.
J ¼ 1.9, 1H), 6.34 (d, J ¼ 1.9, 1H), 5.17 (s, 2H), 4.69 (s, 2H), 3.64 (brs,
4H), 3.33 (brs, 4H), 2.98 (s, 3H), 2.59 (s, 3H).
4.1.9.5. 5-(3-Chloropropoxy)-7-hydroxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (33e). 33e was pale-
yellow powder. Yield 89%; mp > 270 ꢁC decomposed. 1H NMR
The intermediate 31 was prepared as light yellow solid in a
similar manner to the synthesis of 26 from 25. Yield 84%; 1H NMR
(300 MHz, DMSO-d6)
d
7.71e7.30 (m, 10H), 6.65 (d, J ¼ 2.1, 1H), 6.58
(400 MHz, DMSO-d6)
d
10.51 (s, 1H), 6.36 (d, J ¼ 1.6 Hz, 1H), 6.28 (d,
(d, J ¼ 2.0, 1H), 5.21 (s, 2H), 4.80 (s, 2H), 3.83 (s, 3H), 3.71 (brs, 4H),
J ¼ 1.5 Hz, 1H), 4.11 (t, J ¼ 5.8 Hz, 2H), 3.82 (t, J ¼ 6.5 Hz, 2H), 3.24e
3.39 (brs, 4H), 3.03 (s, 3H), 2.60 (s, 3H).
2.66 (m, 8H), 2.61 (s, 2H), 2.30 (s, 3H), 2.27e2.21 (m, 2H). 13C NMR
The final compound 21 was obtained through the similar
operation as 20. Yield 95%; mp 184e186 ꢁC. 1H NMR (300 Hz,
(150 MHz, DMSO-d6) d 160.47, 158.17, 157.61, 154.24, 148.65, 129.04,
103.06, 96.64, 95.20, 65.64, 54.86, 48.34, 45.26, 42.15, 31.47,
17.33.HRMS calcd for C18H24ClN2O4 (M þ Hþ) 367.1425; found
367.1423.
DMSO-d6)
1H), 3.76 (s, 3H), 2.86 (brs, 4H), 2.64 (s, 3H), 2.41 (brs, 4H), 2.21 (s,
3H). 13C NMR (100 MHz, DMSO-d6)
160.3, 158.1, 157.7, 154.4, 149.1,
d
10.67 (s, 1H), 6.36 (d, J ¼ 2.4 Hz, 1H), 6.32 (d, J ¼ 2.4 Hz,
d
129.2, 103.6, 98.2, 92.0, 55.4, 48.9, 46.1, 16.9. HRMS calcd for
C
4.1.9.6. 5-(2-(Dimethylamino)ethoxy)-7-hydroxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (33f). 33f was yellow
powder. Yield 89%; mp > 225 ꢁC decomposed. 1H NMR (400 MHz,
16H21N2O4 (M þ Hþ) 305.1501; found. 305.1498.
4.1.9. General procedure for the synthesis of compounds 33aev
To a solution of 28 (1 mmol), PPh3 (2.5 mmol) and various
commercially available alcohol (2.5 mmol) in dry THF (1 mL) was
added DEAD (2.5 mmol) dropwise with intensive stirring at room
temperature. 20 min later, the reaction mixture was concentrated
in vacuo and the corresponding intermediates 32aev were ob-
tained after chromatography by silica gel eluting with DCMe
CH3OH. The final products 33aev were prepared in a similar
manner to the synthesis of compound 27 from 26 and characterized
after purification by silica gel chromatography eluting with DCMe
CH3OH.
DMSO-d6)
1H), 4.37 (brs, 2H), 3.8e2.8 (m, 8H), 3.43 (brs, 2H), 2.73 (s, 6H), 2.72
(s, 3H), 2.66 (s, 3H). 13C NMR (100 MHz, CD3OD)
162.75, 160.86,
160.10, 156.33, 152.74, 129.98, 105.20, 98.32, 96.91, 67.74, 58.95,
56.33, 45.91, 45.42, 18.38. HRMS calcd for C19H28N3O4 (M þ Hþ)
362.2080; found 362.2060.
d
10.78 (s, 1H), 6.45 (d, J ¼ 1.9 Hz, 1H), 6.40 (d, J ¼ 2.0 Hz,
d
4.1.9.7. 5-(Cyclopentyloxy)-7-hydroxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (33g). 33g was yellow
powder. Yield 70%; mp > 240 ꢁC decomposed. 1H NMR (400 MHz,
DMSO-d6)
d 10.52 (s, 1H), 6.37 (s, 1H), 6.28 (s, 1H), 4.87 (brs, 1H),
3.15e2.67 (brs, 8H), 2.61 (s, 3H), 2.57 (s, 3H), 1.92e1.91 (m, 2H),
4.1.9.1. 5-Ethoxy-7-hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-
1.83e1.79 (m, 2H), 1.76e1.59 (m, 4H). 13C NMR (150 MHz, DMSO-d6)
2H-chromen-2-one (33a). 33a was gray powder. Yield 78.6%;
d 160.5, 157.7, 157.3, 154.4, 149.3, 128.3, 103.2, 97.4, 94.7, 80.0, 54.0,

50
M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
47.5, 43.8, 32.1, 23.6, 17.4. HRMS calcd for C20H27N2O4 (M þ Hþ)
4.1.9.14. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-(pyr-
idin-2-ylmethoxy)-2H-chromen-2-one (33n). 33n was white pow-
der. Yield 81%; mp 166e168 ꢁC. 1H NMR (600 MHz, DMSO-d6)
359.1971; found 359.1969.
4.1. 9. 8. 5-(Cyclohexyloxy)-7-hydroxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (33h). 33h was yellow
powder. Yield 73%; mp 80e82 ꢁC. 1H NMR (400 MHz, DMSO-d6)
d
10.522 (s, 1H), 8.59 (d, J ¼ 4.8 Hz, 1H), 7.85 (td, J ¼ 7.8, 1.8 Hz, 1H),
7.56 (d, J ¼ 8.4 Hz, 1H), 7.37 (dd, J ¼ 6.6, 4.8 Hz, 1H), 6.44 (d,
J ¼ 2.4 Hz, 1H), 6.31 (d, J ¼ 2.4 Hz, 1H), 5.22 (s, 2H), 3.18e2.65 (m,
d
10.47 (s, 1H), 6.41 (d, J ¼ 1.9 Hz, 1H), 6.26 (d, J ¼ 2.0 Hz, 1H), 4.46e
8H), 2.62 (s, 3H), 2.49 (s, 3H). 13C NMR (150 MHz, DMSO-d6)
d 160.5,
4.42 (m, 1H), 3.66e2.74 (m, 8H), 2.65 (s, 6H), 1.95e1.94 (m, 2H),
157.9, 157.7, 155.7, 154.3, 149.2, 137.0, 123.2, 122.2, 103.3, 97.5, 95.5,
71.6, 54.39, 47.7, 45.1, 17.5. HRMS calcd for C21H24N3O4 (M þ Hþ)
382.1767; found 382.1756.
1.71e1.69 (m, 2H), 1.58e1.28 (m, 6H). 13C NMR (100 MHz, DMSO-d6)
d
160.71, 157.80, 157.21, 154.55, 129.08, 128.16, 103.34, 97.28, 94.83,
75.44, 53.87, 47.10, 43.83, 30.78, 25.06, 22.68, 17.88. HRMS calcd for
C
21H29N2O4 (M þ Hþ) 373.2127; found 373.2109.
4.1.9.15. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-(thia-
zol-2-ylmethoxy)-2H-chromen-2-one (33o). 33o was yellow pow-
der. Yield 88%; mp > 230 ꢁC decomposed. 1H NMR (400 MHz,
4.1.9.9. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-((tetra-
hydrofuran-3-yl)oxy)-2H-chromen-2-one (33i). 33i was pale-yellow
powder. Yield 77%; mp 163e165 ꢁC. 1H NMR (400 MHz, DMSO-d6)
DMSO-d6)
1H), 6.41 (d, J ¼ 1.8 Hz, 1H), 5.50 (s, 2H), 3.38 (s, 2H), 2.89 (s, 8H),
2.59 (s, 3H), 2.56 (s, 3H). 13C NMR (150 MHz, DMSO-d6)
160.56,
d
10.71 (s, 1H), 9.22 (s, 1H), 8.11 (s, 1H), 6.61 (d, J ¼ 1.7 Hz,
d
d
10.55 (s,1H), 6.34 (d, J ¼ 2.0 Hz,1H), 6.30 (d, J ¼ 2.0 Hz,1H), 5.09 (s,
157.63, 157.35, 155.71, 154.20, 143.27, 133.09, 128.77, 103.18, 97.59,
95.66, 62.78, 54.10, 47.55, 43.78, 17.55. HRMS calcd for C19H22N3O4S
(M þ Hþ) 388.1331; found 388.1318.
1H), 3.92e3.78 (m, 4H), 2.83 (brs, 8H), 2.60 (s, 3H), 2.51 (s, 3H),
2.29e2.20 (m, 1H), 2.09e2.01 (m, 1H). 13C NMR (150 MHz, DMSO-
d6)
d 160.4, 157.6, 156.8, 154.4, 149.0, 128.7, 103.4, 97.6, 95.2, 78.3,
92.0, 66.4, 54.3, 47.7, 44.3, 32.2, 17.3. HRMS calcd for C19H25N2O5
(M þ Hþ) 361.1763; found 361.1750.
4.1.9.16. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-(2-(pyr-
rolidin-1-yl)ethoxy)-2H-chromen-2-one (33p). 33p was pale-yellow
powder. Yield 91%; mp 207e209 ꢁC. 1H NMR (400 MHz, DMSO-d6)
4.1.9.10. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-((1-
methylpiperidin-4-yl)oxy)-2H-chromen-2-one (33j). 33j was pale-
yellow powder. Yield 45%; mp 127e129 ꢁC. 1H NMR (600 MHz,
d
10.63 (s, 1H), 6.42 (d, J ¼ 1.6 Hz, 1H), 6.34 (d, J ¼ 1.6 Hz, 1H), 4.30
(brs, 2H), 3.02 (brs, 12H), 2.66 (s, 3H), 2.60 (s, 3H), 1.86 (brs, 4H). 13
NMR (100 MHz, CD3OD) 161.34, 159.24, 158.32, 154.75, 150.88,
C
d
DMSO-d6)
d
10.31 (s, 1H), 6.35 (d, J ¼ 2.1 Hz, 1H), 6.22 (d, J ¼ 2.2 Hz,
128.59, 103.64, 97.04, 95.61, 66.56, 56.65, 54.94, 54.36, 54.06, 53.87,
48.10, 44.22, 22.84, 22.53, 16.90. HRMS calcd for C21H30N3O4
(M þ Hþ) 388.2236; found 388.2218.
1H), 4.47e4.40 (m, 1H), 2.63 (s, 3H), 2.58e2.51 (m, 2H), 2.28e2.20
(m, 2H), 2.19 (s, 3H), 2.17 (s, 3H), 1.99e1.92 (m, 2H), 1.77e1.66 (m,
2H). 13C NMR (150 MHz, DMSO-d6)
d 160.3,157.5,156.9,154.4,148.6,
129.3, 103.5, 97.2, 94.9, 55.4, 52.1, 48.9, 46.1, 45.8, 30.1, 17.8. HRMS
4.1.9.17. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-(2-
(piperidin-1-yl)ethoxy)-2H-chromen-2-one (33q). 33q was yellow
powder. Yield 86%; mp 110e112 ꢁC. 1H NMR (300 MHz, DMSO-d6)
calcd for C21H30N3O4 (M þ Hþ) 388.2236; found 388.2226.
4.1.9.11. Tert-butyl 4-((7-hydroxy-4-methyl-3-(4-methylpiperazin-1-
d
10.49 (s, 1H), 6.35 (d, J ¼ 1.1 Hz, 1H), 6.27 (d, J ¼ 1.1 Hz, 1H), 4.08 (t,
yl)-2-oxo-2H-chromen-5-yl)oxy)piperidine-1-carboxylate
33k was pale-yellow powder. Yield 77%; mp 107e109 ꢁC. 1H NMR
(600 MHz, DMSO-d6)
(33k).
J ¼ 4.6 Hz, 2H), 2.76 (t, J ¼ 4.6 Hz, 2H), 2.62 (s, 3H), 2.33 (s, 3H),
1.50e1.49 (m, 4H),1.44e1.27 (m, 2H). 13C NMR (150 MHz, DMSO-d6)
d
6.47 (d, J ¼ 1.8 Hz, 1H), 6.31 (d, J ¼ 2.0 Hz,
d
160.5, 158.3, 157.6, 154.2, 149.0, 128.9, 103.1, 96.9, 95.1, 66.3, 56.8,
1H), 4.64e2.61 (m, 1H), 3.65e3.54 (m, 2H), 3.26 (m, 2H), 3.18e2.65
(m, 8H), 2.62 (s, 3H), 2.59 (s, 3H), 1.99e1.90 (m, 2H), 1.61e1.58 (m,
54.8, 54.1, 48.3, 45.2, 25.3, 23.6, 17.2. HRMS calcd for C22H32N3O4
(M þ Hþ) 402.2393; found 402.2380.
2H), 1.39 (s, 9H). 13C NMR (100 MHz, DMSO-d6)
d 160.76, 157.71,
156.73, 154.42, 153.86, 149.17, 128.38, 103.35, 97.47, 95.12, 78.78,
73.03, 53.82, 47.27, 43.58, 29.97, 28.04, 17.83. HRMS calcd for
C
4.1.9.18. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-(2-
morpholinoethoxy)-2H-chromen-2-one (33r). 33r was yellow pow-
der. Yield 87%; mp > 215 ꢁC decomposed. 1H NMR (400 MHz,
25H36N3O6 (M þ Hþ) 474.2604; found 474.2597.
DMSO-d6)
d
10.60 (s, 1H), 6.40 (d, J ¼ 1.7 Hz, 1H), 6.32 (d, J ¼ 1.8 Hz,
4.1.9.12. 5-(Cyclopropylmethoxy)-7-hydroxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (33l). 33l was pale-
yellow powder. Yield 89%; mp > 238 ꢁC decomposed. 1H NMR
1H), 4.11 (t, J ¼ 4.8 Hz, 2H), 3.59 (t, J ¼ 4.0 Hz, 4H), 3.3e2.8 (m, 8H),
2.77 (t, J ¼ 4.0 Hz, 2H), 2.70 (s, 3H), 2.67 (s, 3H), 2.48 (brs, 4H). 13
C
NMR (150 MHz, DMSO-d6)
d 160.78, 158.46, 157.78, 154.25, 149.67,
(400 MHz, DMSO-d6)
d
10.48 (s, 1H), 6.31 (d, J ¼ 2.1 Hz, 1H), 6.28 (d,
128.17, 103.00, 96.96, 95.11, 66.18, 66.06, 56.70, 53.57, 53.38, 46.96,
43.11, 17.36. HRMS calcd for C21H30N3O5 (M þ Hþ) 404.2185; found
404.2164.
J ¼ 2.1 Hz, 1H), 3.86 (d, J ¼ 7.5 Hz, 2H), 3.5e2.5 (m, 8H), 2.69 (s, 3H),
2.63 (s, 3H), 1.33e1.23 (m, 1H), 0.62e0.57 (m, 2H), 0.39e0.35 (m,
2H). 13C NMR (100 MHz, CD3OD)
d 162.63, 160.92, 160.64, 156.26,
153.36, 129.42, 105.02, 97.88, 96.34, 75.02, 56.09, 48.94, 44.88,
18.33, 11.11, 3.92. HRMS calcd for C19H25N2O4 (M þ Hþ) 345.1814;
found 345.1797.
4.1.9.19. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-(2-(thi-
ophen-2-yl)ethoxy)-2H-chromen-2-one (33s). 33s was pale-yellow
powder. Yield 68%; mp 103e105 ꢁC. 1H NMR (400 MHz, DMSO-
d6)
d
10.45 (s, 1H), 7.37 (dd, J ¼ 4.3, 1.8 Hz, 1H), 6.99 (d, J ¼ 4.2 Hz,
4.1.9.13. 5-((2,4-Dimethylbenzyl)oxy)-7-hydroxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (33m). 33m was pale-
yellow powder. Yield 81%; mp 155e157 ꢁC. 1H NMR (400 MHz,
2H), 6.38 (d, J ¼ 1.8 Hz,1H), 6.28 (d, J ¼ 1.9 Hz,1H), 4.28 (t, J ¼ 5.9 Hz,
2H), 3.34 (t, J ¼ 5.9 Hz, 2H), 2.51 (s, 6H). 13C NMR (100 MHz, CD3OD)
d
161.17, 159.25, 158.76, 154.80, 151.85, 140.42, 127.67, 126.53,
DMSO-d6)
1H), 6.31 (d, J ¼ 2.1 Hz, 1H), 5.06 (s, 2H), 2.89 (brs, 8H), 2.58 (s, 3H),
2.53 (s, 3H), 2.29 (s, 6H). 13C NMR (150 MHz, DMSO-d6)
160.5,
d
10.53 (s, 1H), 7.09 (s, 2H), 6.99 (s, 1H), 6.46 (d, J ¼ 2.1 Hz,
125.33, 123.50, 103.45, 96.46, 95.19, 69.53, 54.29, 47.07, 42.85,
29.20, 16.92. HRMS calcd for C21H25N2O4S (M þ Hþ) 401.1535;
found 401.1537.
d
158.2, 157.7, 154.3, 137.5, 136.0, 129.4, 128.6, 125.6, 103.1, 97.3, 95.2,
70.7, 54.1, 47.5, 44.2, 20.9, 17.5. HRMS calcd for C24H29N2O4
(M þ Hþ) 409.2127; found 409.2122.
4.1.9.20. 7-Hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-5-(2-(pyr-
idin-2-yl)ethoxy)-2H-chromen-2-one (33t). 33t was white powder.

M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
51
Yield 86%; mp 90e92 ꢁC. 1H NMR (400 MHz, DMSO-d6)
d
10.39 (s,
4.1.11.1. N-(2-((7-hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-2-
oxo-2H-chromen-5-yl)oxy)ethyl)methanesulfonamide (37a).
The final product 37a was pale-yellow powder. Yield 62%; mp 189e
191 ꢁC. 1H NMR (400 MHz, DMSO-d6)
1H), 8.51 (d, J ¼ 4.2 Hz, 1H), 7.74 (t, J ¼ 7.4 Hz, 1H), 7.37 (d, J ¼ 7.7 Hz,
1H), 7.26e7.23 (m, 1H), 6.39 (d, J ¼ 1.5 Hz, 1H), 6.25 (d, J ¼ 1.6 Hz,
1H), 4.43 (t, J ¼ 6.0 Hz, 2H), 3.25 (t, J ¼ 5.9 Hz, 2H), 2.31 (s, 3H), 2.19
d
7.28 (t, J ¼ 5.6 Hz, 1H), 6.36
(s, 3H). 13C NMR (150 MHz, DMSO-d6)
d
160.3, 158.3, 158.2, 157.5,
(d, J ¼ 2.2 Hz, 1H), 6.31 (d, J ¼ 2.1 Hz, 1H), 4.08 (t, J ¼ 5.2 Hz, 2H),
154.2, 149.1, 148.6, 136.5, 129.2, 123.5, 121.7, 103.0, 96.6, 95.0, 67.8,
55.3, 48.8, 46.1, 36.7, 16.9. HRMS calcd for C22H26N3O4 (M þ Hþ)
396.1923; found 396.1915.
3.42e3.38 (m, 2H), 2.95 (s, 3H), 2.67 (s, 3H), 2.61 (s, 3H). 13C NMR
(100 MHz, CD3OD)
d 162.47, 160.74, 159.91, 156.10, 152.81, 129.76,
105.05, 97.97, 96.70, 69.31, 56.19, 49.43, 45.30, 43.27, 39.92, 18.40.
HRMS calcd for C18H26N3O6S (M þ Hþ) 412.1542; found 412.1533.
4.1.9.21. Tert-butyl (2-((7-hydroxy-4-methyl-3-(4-methylpiperazin-
1-yl)-2-oxo-2H-chromen-5-yl)oxy)ethyl)carbamate (33u). 33u was
pale-yellow powder. Yield 63%; mp 114e116 ꢁC. 1H NMR (400 MHz,
4.1.11.2. N-(2-((7-hydroxy-4-methyl-3-(4-methylpiperazin-1-yl)-2-
oxo-2H-chromen-5-yl)oxy)ethyl)-4-methylbenzenesulfonamide
(37b). The final product 37b was white solid. Yield 57%;
DMSO-d6)
(s, 2H), 3.39 (s, 2H), 2.60 (s, 3H), 2.20 (s, 3H), 1.39 (s, 9H). 13C NMR
(100 MHz, CD3OD) 162.4, 160.7, 160.1, 158.4, 156.1, 152.7, 129.8,
d 10.45 (s, 1H), 7.02 (s, 1H), 6.31 (s, 1H), 6.27 (s, 1H), 3.99
mp > 250 ꢁC. 1H NMR (400 MHz, DMSO-d6)
d 10.57 (s, 1H), 7.85 (t,
d
J ¼ 5.7 Hz, 1H), 7.69 (d, J ¼ 8.1 Hz, 2H), 7.37 (d, J ¼ 8.0 Hz, 2H), 6.30
(d, J ¼ 2.1 Hz, 1H), 6.29 (d, J ¼ 2.1 Hz, 1H), 4.01 (t, J ¼ 4.9 Hz, 2H),
3.8e2.8 (m, 8H), 3.19e3.15 (m, 2H), 2.79 (s, 3H), 2.62 (s, 3H), 2.37 (s,
105.1, 97.8, 96.6, 80.3, 68.7, 56.3, 49.1, 45.4, 40.8, 28.8, 18.3. HRMS
calcd for C22H32N3O6 (M þ Hþ) 434.2291; found 434.2286.
3H). 13C NMR (150 MHz, DMSO-d6)
d 160.7, 158.1, 157.8, 154.2, 142.6,
4.1.9.22. Tert-butyl (3-((7-hydroxy-4-methyl-3-(4-methylpiperazin-
1-yl)-2-oxo-2H-chromen-5-yl)oxy)propyl)carbamate (33v). 33v was
pale-yellow powder. Yield 59%; mp 164e166 ꢁC. 1H NMR (400 MHz,
137.4, 129.6, 128.0, 126.4, 103.0, 96.7, 95.2, 67.3, 53.4, 46.7, 42.7, 41.9,
20.9, 17.5. HRMS calcd for C24H30N3O6S (M þ Hþ) 488.1855; found
488.1845.
DMSO-d6)
(s, 2H), 2.68 (s, 3H), 2.64 (s, 3H), 1.90 (s, 2H), 1.37 (s, 9H). 13C NMR
(100 MHz, CD3OD) 152.92, 151.20, 150.86, 149.01, 146.59, 143.34,
d 10.56 (s, 1H), 6.36 (s, 1H), 6.30 (s, 1H), 4.00 (s, 2H), 3.14
4.2. Biology
d
117.44, 95.37, 88.14, 86.81, 70.47, 58.28, 46.59, 39.57, 35.57, 29.06,
21.15, 19.28, 8.85. HRMS calcd for C23H34N3O6 (M þ Hþ) 448.2448;
found 448.2445.
Cell culture and 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenytetra-
zolium bromide (MTT) assay. PC12 cells were purchased from the
American Type Culture Collection (ATCC). The cells were cultured in
Dulbecco’s modified eagle medium (DMEM) supplemented with 5%
4.1.10. Procedure for the synthesis of compounds 35a and 35b
To a solution of intermediate 32u or 32v (0.3 mmol) in 10 mL
DCM was added CH3COOH (2 mL). The mixture was stirred at 40 ꢁC
for 1 h. The corresponding intermediate 34a or 34b was obtained
after purification by silica gel chromatography eluting with DCMe
CH3OH. Then to the solution of intermediate 34a or 34b in THF
(5 mL) was added TBAF (0.6 mmol) under stirring at 50 ꢁC. The
corresponding product 35a or 35b was obtained after purification
by C18 chromatography eluting with CH3CNeH2O.
fatal bovine serum (FBS) and 5% equine serum, and L-glutamine
(2 mM). Cultures were maintained at 37 ꢁC in 5% CO2 in a humid-
ified incubator. For cell viability assay, PC12 cells were incubated in
96-well plates at a density of 5 ꢃ 103 cells per well for 24 h. Then
compounds at dose of 10 mM and damage factor were added in the
cells. OGD assay: the cells were treated by low glucose medium and
5 mM sodium dithionite for 24 h. Then 5 mg/mL MTT was added
and maintained for 4 h. Absorbance was measured using an
Ultramark microplate reader at a wavelength of 570 nm. The cell
viability was expressed as a percentage of the absorbance density
value of the control cultures.
4.1.10.1. 5-(2-Aminoethoxy)-7-hydroxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-on (35a). The final product
35a was pale-yellow powder. Yield 53%; mp 148e150 ꢁC. 1H NMR
4.2.1. Animals
(300 MHz, DMSO-d6)
d
10.79 (s, 1H), 8.20 (s, 2H), 6.43 (d, J ¼ 2.0 Hz,
Male SpragueeDawley rats (age, 7 weeks; weighting, 260e280 g)
were purchased from Experimental Animal Center of Chinese
Academy of Medical Sciences (Beijing China). All animals were
handled in accordance with the standards established in the Guide
for the Care and Use of Laboratory Animals published by the Institute
of Laboratory Animal Resources of the National Research Council
(United States) and approved by the Animal Care and Use Committee
of the Peking Union Medical College and the Chinese Academy of
Medical Sciences.
1H), 6.35 (d, J ¼ 2.1 Hz, 1H), 4.24 (t, J ¼ 4.8 Hz, 2H), 3.43 (s, 6H), 3.32
(t, J ¼ 4.8 Hz, 3H), 3.17 (s, 2H), 2.85 (s, 3H), 2.66 (s, 3H). 13C NMR
(100 MHz, DMSO-d6) d 160.9, 158.6, 158.3, 157.8, 154.3, 128.0, 103.2,
97.2, 95.7, 65.7, 53.4, 46.7, 42.6, 38.1, 17.4. HRMS calcd for
C
17H24N3O4 (M þ Hþ) 334.1767; found 334.1785.
4.1.10.2. 5-(3-Aminopropoxy)-7-hydroxy-4-methyl-3-(4-
methylpiperazin-1-yl)-2H-chromen-2-one (35b). The final product
35b was yellow powder. Yield 45%; mp > 245 ꢁC decomposed. 1H
NMR (400 MHz, DMSO-d6)
d
6.34 (d, J ¼ 2.2 Hz, 1H), 6.2 (d,
4.2.2. Focal brain ischemia
J ¼ 1.7 Hz, 1H), 4.07 (t, J ¼ 5.6 Hz, 2H), 2.91 (t, J ¼ 7.0 Hz, 2H), 2.62 (s,
Transient middle cerebral artery occlusion (TMCAO) was
performed as previously described with some modifications.
Briefly, under 10% chloral hydrate (4 mL/kg, intraperitoneal injec-
tion), a 4e0 nylon thread, the tip of which was burned (diameter
0.36 mm), was inserted into the right common carotid artery and
advanced until the origin of the right MCA was occluded. After
60 min of the occlusion, the thread was removed to allow reper-
fusion, and then the rats (n ¼ 12) were returned to the chamber.
To investigate the effect of 20 on MCAO, saline or 20 at doses of
1.5, 3.0, 6.0 mg/kg or edaravone at dose of 6.0 mg/kg was intrave-
nous injection at the same time of reperfusion. The rats were
assessed for neurologic deficits at 24 h after reperfusion according
to longa’s 5-point scale. A score of 0: no neurological deficit; a score
3H), 2.20 (s, 3H), 2.05e1.99 (m, 2H). 13C NMR (100 MHz, CD3OD)
d
162.78, 160.14, 156.30, 153.28, 129.87, 126.91, 104.81, 98.10, 96.73,
67.09, 55.55, 48.36, 43.92, 38.25, 28.32, 18.32. HRMS calcd for
C
18H26N3O4 (M þ Hþ) 348.1923; found 348.1911.
4.1.11. Procedure for the synthesis of compounds 37a and 37b
To a solution of 34a (0.2 mmol) in 10 mL DCM was added TEA
(0.4 mmol) and various acyl chlorides (0.3 mmol) with stirring at
room temperature. The intermediate 36a or 36b was obtained after
purification by silica gel chromatography eluting with DCMe
CH3OH. The final product 37a or 37b was synthesized in a similar
manner to the synthesis of 35a from 34a.

52
M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
of 1: failure to extend left forepaw fully; a score of 2: circling to the
left; a score of 3: falling to the left; and a score of 4: did not walk
spontaneously and had a depressed level of consciousness. The
animals without symptoms of neurological impairment or dying
after the surgery were rejected and other rats were recruited.
The animals were killed at 24 h after reperfusion. Brains of the
animals were removed and cut into 2 mm-thick slices, fora total of six
slices per animal. The slices were immersed in a 1% solution of 2,3,5-
triphenyltetrazolium chloride (TTC) in phosphate buffered saline
(PBS) at 37 ꢁC for 30 min and fixed in 4% phosphate buffered formalin.
Images of the slices were obtained with a scanner and a computer.
The infarct area and the total area were calculated by tracing the areas
on the computer screen with Image J. The percentage of the infarct
area was expressed as the infarct area to total area.
detected with the Enhanced Chemiluminescence (ECL) plus
detection system (Molecular Device, Lmax). The density of each
band was quantified using image analysis software (Science Lab
2005 Image Guage; Fuji Film Co. Ltd, Tokyo, Japan).
4.2.6. Statistic analysis
Results were expressed as mean ꢂ SD. Statistical evaluation was
performed using one-way analysis of variance. Significant differ-
ence was further performed in conjunction with the Studente
NewmaneKeuls method. Statistical significance was accepted at
P < 0.05 (n ¼ 12).
4.3. General procedure for pharmacokinetic investigation
To assess the brain water content, brain samples were dried in
an oven at 110 ꢁC for 24 h, and water content of these samples was
then measured by the (weight-dry)/weight.
For histological analysis, the rats were anesthetized with 10%
chloral hydrate and then perfused with PBS followed by 4% buffered
paraformaldehyde 24 h. The brain of each rat was then removed,
Male SpragueeDawley rats (260e280 g) were intravenously
injected with compound 20 at dose of 6 mg/kg. Then the rats were
anesthetize by ether and blood samples (approximately 1.5 mL)
were collected from the jugular vein into the tubes containing so-
dium heparin at 3.5, 7, 15, 30, 60 and 120 min. Plasma was prepared
by centrifugation of blood at 4000 rpm for 5 min, and then stored
at ꢀ70 ꢁC until analysis. The rats were perfused with saline after the
blood samples were collected, and the cortex and hippocampus
were separated and homogenated with 2 volumes (v/w) of Milli-Q
water. The samples were centrifuged at 13,000 rpm for 10 min, and
the supernatant was collected and stored at ꢀ70 ꢁC until analysis.
postfixed, and embedded in paraffin. Coronal sections of 8 mM were
stained with 0.5% cresyl fast violet.
4.2.3. TUNNEL staining
Detection of cell death insituDNA fragmentationisone of thewide
accepted markers by which apoptotic cells are recognized. To detect
DNA fragmentation of cell nuclei, TUNEL reaction was applied to the
paraffin sections using the In Situ Cell Death Detection Kit, peroxidase
(POD) (Roche Applied Science, Germany). The brain sections were
incubated with TUNEL reaction according to the manufacturer’s
protocol. After dewaxing and rehydration, brain slices were treated
Test tubes containing rat plasma or cortex (200
standard (100 L of 1 g/mL) were subjected to protein precipita-
tion with acetonitrile (200 L) using a vortex mixer for 1 min and
mL) and the internal
m
m
m
centrifuged at 10,000 rpm for 10 min. Supernatant was evaporated
to dryness under a gentle stream of nitrogen. The residue was
reconstituted in an aliquot of the mobile phase (200
using 0.45 M Teflon filters before LC-MS analysis. An Alltima C18
column (150 mm ꢃ 4.6 mm, 5 m) with a C18 guard column were
mL) and filtered
with protease K (20 mg/mL) for 15 min at room temperature. Then
m
slices were incubated with reaction buffer containing TdT enzyme
(37 ꢁC, 1 h). After washing with stop/wash buffer, slices were treated
with anti-digoxigenin conjugate for 30 min at room temperature, and
subsequently developed color in alkaline phosphatase substrate.
m
applied, using a mixture of acetonitrile and 0.02 M ammonium
formate with a gradient elution (0e15 min, acetonitrile of 10%e
80%). The flow rate was set at 0.8 mL$minꢀ1 and the detective
wavelength fixed at 332 nm. An Agilent 1200 HPLC system coupled
to Agilent 6130 mass spectrometer (Aglient, USA) equipped with
ESI were used for the study. A positive mode was applied with
selected ion monitoring mode (SIM); the selected ions for com-
pound 20 and the internal standard (IS), were 304.9 ([M þ H]) and
240.8 [M þ H] respectively. Quantitation was achieved by con-
structing a calibration curve by weighted linear regression of the
ratio of the analyte peak area to that of the added internal standard.
The standard calibration curve for compound 20 was linearly pro-
4.2.4. Electron microscopy
Processing and electron microscope (EM) were conducted as
previously described (Jasinska et al., 2006). Briefly, anesthetized
animals were perfused with 0.1 M PBS, followed by 3.75% acrolein
in buffered 2% paraformaldehyde (PF). Then the brains were
removed, postfixed with 2.5% glutaraldehyde for 2 h, washed with
0.1 M PBS, and then exposed to 1% osmium tetraoxide for 2 h. After
several subsequent washes with water, the tissues were dehy-
drated with gradient alcohol and embedded in EPON resin.
Randomly selected ultrathin sections were stained with uranyl
acetate and lead citrate and examined using a transmission elec-
tron microscope (H-7650, HITACHI, Tokyo, Japan).
portional to the concentration from 0.1
mg/mL to 30 mg/mL in
plasma and from 1 ng/mL to 2000 ng/mL in cortex, respectively.
All the detection limits of compound 20 were 0.05 ng.
Acknowledgments
4.2.5. Western blot analysis
The animals were anesthetized with 10% chloral hydrate and
then perfused with saline at 24 h after reperfusion, and the cortex
and hippocampus samples from the rats were homogenized thor-
oughly and then lysed in a RIPA lysis buffer (50 mmol$Lꢀ1 Tris (pH
7.4), 150 mmol Lꢀ1 NaCl, 1% NP40, 0.5% sodium deoxycholate and
0.1% sodium dodecyl sulfonate (SDS)). Protein concentrations were
measured with a bicinchoninic acid (BCA) kit. The lysates were
solubilized in SDS sample and separated by SDS-PAGE, then
transferred to polyvinyl difluoride (PVDF) membrane (Millipore,
Billerica, MA, USA). The membrane was blocked by 3% bull serum
This research is supported financially by National Natural Sci-
ence Foundation of China (No. 90713045) and National 863 Pro-
gram of China (No. 2012AA020303).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2013.04.015.
References
albumin (BSA) and incubated with anti-Bcl-2, anti-Bax, anti-b-actin
(Santa Cruz, CA, California, USA), anti-Caspase-3 (Cell Signaling,
Beverly, MA, USA), followed by horseradish peroxides (HRP)-con-
jugated secondary antibody (Santa Cruz, CA, California, USA), and
[1] K.P. Doyle, P.P. Simon, M.P. Stenzel-Poore, Neuropharmacology 55 (2008)
310e318.
[2] P.J. Crack, J.M. Taylor, Free Radical Biol. Med. 38 (2005) 1433e1444.

M. Sun et al. / European Journal of Medicinal Chemistry 67 (2013) 39e53
53
[3] T. Wang, J. Gu, P.F. Wu, F. Wang, Z. Xiong, Y.J. Yang, W.N. Wu, L.D. Dong,
[14] H. Shichinohe, S. Kuroda, H. Yasuda, T. Ishikawa, M. Iwai, M. Horiuchi,
Y. Iwasaki, Brain Res. 1029 (2004) 200e206.
[15] N. Zhang, M. Komine-Kobayashi, R. Tanaka, M.Z. Liu, Y. Mizuno, T. Urabe,
Stroke 36 (2005) 2220e2225.
[16] S. Amemiya, T. Kamiya, C. Nito, T. Inaba, K. Kato, M. Ueda, K. Shimazaki,
Y. Katayama, Eur. J. Pharmacol. 516 (2005) 125e130.
[17] Anon, Cerebrovasc. Dis. 15 (2003) 222e229.
J.G. Chen, Free Radical Biol. Med. 47 (2009) 229e240.
ꢀ
[4] D.S. Warner, H. Sheng, I. Batinic-Haberle, J. Exp. Biol. 207 (2004) 3221e3231.
[5] S.A. Andrabi, I. Sayeed, D. Siemen, G. Wolf, T.F. Horn, FASEB J. 18 (2004)
869e871.
[6] N. Noguchi, Y. Yoshida, H. Kaneda, Y. Yamamoto, E. Niki, Biochem. Pharmacol.
44 (1992) 39e44.
[7] A.R. Green, T. Ashwood, Curr. Drug Targets CNS Neurol. Disord. 4 (2005) 109e118.
[8] T. Yoshimoto, P. Kanakaraj, M.J. Ying, M.S. Cheng, I. Kerr, L. Malaiyandi,
J.A. Watson, B.K. Siesjö, K.R. Maples, Brain Res. 947 (2002) 191e198.
[9] T. Yoshimoto, T. Kristian, B. Hu, Y.B. Ouyang, B.K. Siesjö, Brain Res. 932 (2002)
99e109.
[10] R. Schmid-Elsaesser, S. Zausinger, E. Hungerhuber, A. Baethmann, H.J. Reulen,
Neurosurgery 44 (1999) 163e171.
[11] R. Schmid-Elsaesser, S. Zausinger, E. Hungerhuber, A. Baethmann, H.J. Reulen,
Exp. Brain Res. 122 (1998) 121e127.
[12] C. Zhu, X. Wang, Z. Huang, L. Qiu, F. Xu, N. Vabsen, M. Nilsson, P.S. Eriksson,
H. Hagberg, C. Culmsee, N. Plesnila, G. Kroemer, K. Blomgren, Cell Death Differ.
14 (2007) 775e784.
[13] J.I. Noor, T. Ikeda, K. Mishima, N. Aoo, S. Ohta, N. Egashira, K. Iwasaki,
M. Fujiwara, T. Ikenoue, Stroke 36 (2005) 2468e2474.
[18] S.Y. Kang, Y.C. Kim, Arch. Pharm. Res. 30 (2007) 1368e1373.
[19] C. Wang, A.J. Pei, J. Chen, H.L. Yu, M.L. Sun, C.F. Liu, X.S. Xu, J. Neurochem. 121
(2012) 1007e1013.
[20] G. Li, D.M. Wang, M.N. Sun, G.Y. Li, J.F. Hu, Y. Zhang, Y.H. Yuan, H.J. Ji,
N.H. Chen, G. Liu, J. Med. Chem. 53 (2010) 1741e1754.
[21] M.D. Cowart, Y.L. Bennani, R. Faghih, L.A. Black, U.S. Patent 20020183309A1, 2002.
[22] K.E. Fahrenholtz, M. Lurie, R.W. Kierstead, J. Am. Chem. Soc. 89 (1967)
5934e5941.
[23] S.P. Findlay, G. Dougherty, J. Org. Chem. 13 (1948) 560e569.
[24] T.A. Kosten, M.P. Galloway, R.S. Duman, D.S. Russell, C. D’Sa, Neuro-
psychopharmacology 33 (2008) 1545e1558.
[25] B. Natasa, W. Christian, Brain Res. 80 (2009) 251e267.
[26] Z.H. Cheung, M.C. Leung, H.K. Yip, W. Wu, F.K. Siu, K.F. So, Cell. Mol. Neurobiol.
28 (2008) 137e155.