Design, synthesis, biological evaluation, homology modeling and docking studies of (E)-3-(benzo[d][1,3]dioxol-5-ylmethylene) pyrrolidin-2-one derivatives as potent anticonvulsant agents

Article abstract of DOI:10.1016/j.bmcl.2018.03.015

A series of (E)-3-(benzo[d][1,3]dioxol-5-ylmethylene)pyrrolidin-2-one derivatives were designed, synthesized, and evaluated for their anticonvulsant activities. In the preliminary screening, compounds 5, 6a–6f and 6h–6i showed promising anticonvulsant activities in MES model, while 6f and 6g represented protection against seizures at doses of 100 mg/kg and 0.5 h in scPTZ model. The most active compound 6d had a high-degree protection against the MES-induced seizures with ED₅₀ value of 4.3 mg/kg and TD₅₀ value of 160.9 mg/kg after intraperitoneal (i.p.) injection in mice, which provided 6d in a high protective index (TD₅₀/ED₅₀) of 37.4 comparable to the reference drugs. Beyond that, 6d has been selected and evaluated in vitro experiment to estimate the activation impact. Apparently, 6d clearly inhibits the Na_v1.1 channel. Our preliminary results provide new insights for the development of small-molecule activators targeting specifically Na_v1.1 channels to design potential drugs for treating epilepsy. The computational parameters, such as homology modeling, docking study, and ADME prediction, were made to exploit the results.

Full text of DOI:10.1016/j.bmcl.2018.03.015

Accepted Manuscript
Design, Synthesis, Biological Evaluation, Homology Modeling and Docking
Studies of (E)-3-(benzo[d][1,3]dioxol-5-ylmethylene) Pyrrolidin-2-one Deriv-
atives as Potent Anticonvulsant Agents
Tiantian Wang, Shiyang Dong, Xiaodong Chen, Kun Qian, Huayu Wang, Hexiu
Quan, Zhongli Zhang, Yueming Zuo, Liping Huang, Dongxun Li, Ming Yang,
Shilin Yang, Yi Jin, Zengtao Wang
PII:
S0960-894X(18)30194-X
https://doi.org/10.1016/j.bmcl.2018.03.015
BMCL 25669
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
7 January 2018
18 February 2018
5 March 2018
Please cite this article as: Wang, T., Dong, S., Chen, X., Qian, K., Wang, H., Quan, H., Zhang, Z., Zuo, Y., Huang,
L., Li, D., Yang, M., Yang, S., Jin, Y., Wang, Z., Design, Synthesis, Biological Evaluation, Homology Modeling
and Docking Studies of (E)-3-(benzo[d][1,3]dioxol-5-ylmethylene) Pyrrolidin-2-one Derivatives as Potent
Anticonvulsant Agents, Bioorganic & Medicinal Chemistry Letters (2018), doi: https://doi.org/10.1016/j.bmcl.
2
018.03.015
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Design, Synthesis, Biological Evaluation, Homology Modeling and Docking
Studies of (E)-3-(benzo[d][1,3]dioxol-5-ylmethylene) Pyrrolidin-2-one
Derivatives as Potent Anticonvulsant Agents
a, b
a, e
a
a
a
b, c
Tiantian Wang , Shiyang Dong , Xiaodong Chen , Kun Qian , Huayu Wang , Hexiu Quan
,
a
a
a
b
a, d
b
Zhongli Zhang , Yueming Zuo , Liping Huang , Dongxun Li , Ming Yang , Shilin Yang , Yi
a, b
a,
Jin , Zengtao Wang *
a
College of Pharmacy, Jiangxi University of Traditional Chinese Medicine; Nanchang 330004, China.
The National Pharmaceutical Engineering Center for Solid Preparation in Chinese Herbal Medicine;
Nanchang 330006, China.
b
c
Basic Medical College, Jiangxi University of Traditional Chinese Medicine, Nanchang, Jiangxi, China.
Key Laboratory of Modern Preparation of TCM, Ministry of Education, Jiangxi University of
Traditional Chinese Medicine, Nanchang, Jiangxi, China.
d
e
College of Pharmacy, Hubei University of Science and Technology; Xianning 437100, China.
*
Corresponding author. Address: College of Pharmacy, Jiangxi University of Traditional Chinese
Medicine; Nanchang 330004, China.
E-mail: zengtaowang@126.com (Zengtao Wang)
Abstract:
A series of (E)-3-(benzo[d][1,3]dioxol-5-ylmethylene)pyrrolidin-2-one derivatives were designed,
synthesized, and evaluated for their anticonvulsant activities. In the preliminary screening,
compounds 5, 6a-6f and 6h-6i showed promising anticonvulsant activities in MES model, while
6f and 6g represented protection against seizures at doses of 100 mg/kg and 0.5 h in scPTZ model.
The most active compound 6d had a high-degree protection against the MES-induced seizures
with ED50 value of 4.3 mg/kg and TD50 value of 160.9 mg/kg after intraperitoneal (i.p.) injection
in mice, which provided 6d in a high protective index (TD50/ED50) of 37.4 comparable to the

reference drugs. Beyond that, 6d has been selected and evaluated in vitro experiment to estimate
the activation impact. Apparently, 6d clearly inhibits the Na 1.1 channel. Our preliminary results
v
provide new insights for the development of small-molecule activators targeting specifically
Nav1.1 channels to design potential drugs for treating epilepsy. The computational parameters,
such as homology modeling, docking study, and ADME prediction, were made to exploit the
results.
Keywords: (E)-3-(benzo[d][1,3]dioxol-5-ylmethylene)pyrrolidin-2-one, design; synthesis,
anticonvulsant; homology modeling; docking study
Nowadays, it is urgent to discover new chemical entities as antiepileptic drugs
(
AEDs), since about one-third of patients undergoing epilepsy is still insufficiently
1
treated . Besides, currently available AEDs tend to cause multiple serious side-effects,
such as drowsiness, ataxia, gastrointestinal disturbances, gingival hyperplasia,
2
-4
hirsutism, and megaloblastic anaemia .
Due to incomplete information on the pathogenesis of human epilepsy and the
complex mechanisms of actions of AEDs, it is challenging to use rational
5
methodologies to discover new AEDs . Therefore, the ligand-based pharmacophore
6
-8
approach is an important strategy for designing novel anticonvulsant agents . For
example, Malik S. et. al. reported a series of 3-(benzo[d]isoxazol-3-yl)-N-substituted
pyrrolidine-2, 5-diones, which were designed dependent upon the existing biological
data from old drugs, new drugs, and some other anticonvulsant active components. In
this regard, we examined the structural characteristics of typical anticonvulsant agents,
9
10
11
12
such as ethosuximide , levetiracetam , brivaracetam , and seletracetam (Figure 1).
The inspection showed that these drugs shared a common pyrrolidinone moiety in
their molecules. Inspired by this observation, we assumed that the compound
containing pyrrolidinone moiety in a single molecule could be favorable to

anticonvulsant activity.
Figure 1. Anticonvulsant agents bearing pyrrolidinone fragments
1
3
An earlier structure-activity relationship studies of stiripentol revealed the
remarkable anticonvulsant activity for the skeleton of benzo[d][1,3]dioxole. Aboul-
1
3
enein et. al. attached semicarbazone moiety to the backbone of stiripentol (strategy a)
and cyclization of the semicarbazone (strategy b) to afford the semicarbazone S and
racemic pyrazoline S (Figure 2), respectively. S showed an excellent ED50 value of
7 mg/kg in the MES model, while S had a good activity against scPTZ-induced
seizures (ED50 =110 mg/kg).
1
2
1
8
2
Figure 2. Stiripentol and designed anticonvulsant agents
In this study, we implemented the strategy c. To be specific, a five-membered
pyrrolidinone ring was used to modify neopentyl alcohol, which aimed to generate a

synergetic effect in the treatment of epilepsy. The structures of designed targeting
compounds along with stiripentol were presented in Figure 2.
Generally speaking, the benzyl-based derivatives were beneficial and important
1
4-16
for anticonvulsant activity, which had been depicted in numerous studies
including the marketed drugs Lacosamide, Rufinamide, Safinamide, and Retigabine.
Herein, the influences of N-terminal benzyl substitutions on the anticonvulsant
profiles of 5 were further explored via synthesis and pharmacological evaluation of
6
a-6i.
Scheme 1. The synthesis route of targeting compounds 6a-6i.
The synthetic route of targeting compounds was illustrated in Scheme 1. 3, 4-
Dihydroxy benzaldehyde 1 was firstly reacted with CH Cl in presence of DMF
2
2
1
7
according to the published method . Besides, the resulting 2 was used as an active
scaffold for the synthesis of E-configuration 5 carrying Knoevenagel condensation
reaction. However, due to the weak acidity of pyrrolidone 3 α-hydrogens,
benzo[d][1,3]dioxole-5-carbaldehyde 2 could not be directly condensed with 3.

Instead, compound 3 was initially reacted with acetic anhydride to produce 4. Herein,
the N-acetyl group is an electron-withdrawing group which facilitates the
1
8
condensation reaction through enhancing the acidity of the α-hydrogens . Then, an
excess of base was used during the condensation and the activating group was easily
removed from the product 5. Furthermore, the resulting 5 was modified with different
substituted benzyl chlorides and 6a-6i was obtained.
Figure 4. Computational geometries of the hypothetical Z- and E- configurations of
5
obtained by using MMFF94 with 5000 iterations and minimum RMS gradient of
.10 (Discovery Studio 4.5 version).
0
Compound 5 did not form crystals suitable for X-ray diffraction (XRD) analysis.
1
As a result, the further structural elucidations were limited to the H-NMR
experiments. From the two hypothetical conformations of 5 (Figure 4), the E-
configuration was proposed as the more feasible one. Besides, a long-range spin-spin
coupling of H-4, H′-4 and H-Me was observed in the NMR spectrum of 5.
Additionally, the signal of protons H-4 and H′-4 appeared as a triplet of doublets with
two vicinal coupling constants of JH4, H′4- H5, H′5 = 6.6 and J H4, H′4- -HMe = 3.0 (in Hz). It
is worth mentioning that the NOE experiment showed no correlations between the H-
Me and H-4, H′-4 signals, suggesting that H-Me and H-4 H′-4 were on the opposite

sides of the double bond, which was only possible for the E-configuration 5 (See
1
Supplementary Figure S2 H-NMR and NOE spectrums in Supplementary data).
The preclinical discovery and development of novel chemical entities to treat
epilepsy heavily rely on using animal models of seizures. The maximal electroshock
(
MES) and subcutaneous pentylenetetrazole (scPTZ) screening are two important and
1
9, 20
routine animal models for the anticonvulsant studies.
Almost all AEDs are
2
1
protective in at least one of these two models. Therefore, these two kinds of
anticonvulsant tests are significant for the clinical prediction of the anticonvulsant
drug candidates. Besides, the acute neurological toxicity (NT) was required and
2
2
determined by the rotarod test. The screening results had been summarized in Table
1
.
Table 1 The primary anticonvulsant data in mice of compounds 5 and 6a–6i (i.p.).
a
Intraperitoneal injection in mice
b
c
d
Compounds
MES screening
scPTZ screening
NT screening
0
.5h
4h
100
100
-
0.5h
4h
-
0.5h
300
300
300
300
300
300
300
300
300
300
100
4h
5
30
100
30
30
30
-
-
300
300
300
300
300
300
300
300
300
300
100
6
6
a
-
-
b
-
-
6c
30
30
30
30
-
-
-
6d
-
-
6e
-
100
100
-
-
6f
-
-
6
6
g
-
-
h
30
100
30
30
100
30
-
6i
-
-
Phenytoin
-
-
a
Number of animal used = 3, Doses of 30, 100 and 300 mg/kg were administered. The figure in
the table indicates the minimum dose whereby bioactivity was demonstrated in two or three mice.
After injection was made, the animals were examined 0.5 and 4.0 h. The dash (-) indicates the
b
c
absence of activity at maximum dose administered (300 mg/kg). Maximal electroshock test.
d
Subcutaneous pentylenetetrazole test. Neurotoxocity screening (rotarod test).

As shown in Table 1, compounds 5, 6a-6f and 6h-6i exhibited considerable
anticonvulsant activities in the MES test. An instilling into these outcomes showed
that 6c, 6d and 6h were more effective in a dose of 30 mg/kg at the both reported time
intervals (0.5 and 4.0 h), which depicted a quick onset and prolonged anticonvulsant
potential of these derivatives at the minimum dose. Among the entire compounds in
scPTZ test, 6f and 6g were found to be significantly active at a dose of 100 mg/kg
after 0.5 h, but not continue to show activity after 4.0 h. The rotarod test indicated that
all compounds were less toxic than phenytoin and displayed motor impairment at the
dose of 300 mg/kg at 0.5 and 4.0 h.
As a result of the primary screening, the selected highly active compounds (5, 6c,
6
d and 6h) were further subjected to secondary trials to quantify their preliminary
anticonvulsant activity in mice. As displayed in Table 2, 6d was the most active and
promising compound in this study. With an ED50 of 4.3 mg/kg and TD50 value of
1
60.9 mg/kg, it can cause a PI value of 37.4. The order of anticonvulsant activities in
the MES test as judged from ED50 values is 6d > 6h > 6c > 5. As shown by the
pharmacological results, these compounds were active against the MES test, which
could be recommended to develop as anticonvulsant candidates in treating grand mal
seizures.
Table 2 The secondary quantitative anticonvulsant evaluation in MES model in mice
(
i.p.).
TPE
b
c
d
Compounds
ED50
TD50
PI
a
(
h)
1
1
1
1
2
1
e
5
17.3(12.1-27.7)
6.3(1.4-13.3)
4.3(2.3-7.3)
179.4 (142.1-223.1)
167.2 (123.7-195.3)
160.9 (115.3-224.4)
238.5(220.0--272.0)
65.5(52.5-72.9)
10.4
26.5
37.4
42.6
6.9
6c
6
6
d
h
5.6(1.0-11.3)
9.5 (8.1-10.4)
21.8(21.825.5)
Phenytoin
Phenobarbital
69(62.8-72.9)
3.2

a
b
Time to peak effect; ED50: median effective dose affording anticonvulsant protection in 50% of
c
animals, the dose is measured in mg/kg; TD50: median toxic dose eliciting minimal neurological
d
e
toxicity in 50% of animals, the dose is measured in mg/kg; PI: protective index (TD50/ED50); 95%
confidence intervals given in parentheses.
2
3-26
A wealth of evidence
shows that compounds effective in the MES model
are thought to be sodium channel blockers and they can block seizure spread (e.g.,
phenytoin, carbamazepine), while those effective in the scPTZ model are often
GABA
A
agonists. To explain the possible mechanism of anticonvulsant action, for
+
the chosen active 6d, its influence on voltage-dependent Na channel was tested in
vitro. The selected Na
V
1.1 (encoded by the SCN1A gene), as a test channel, is the
most frequent target of mutations, which is responsible for genetic epilepsy
2
7, 28
syndromes with a wide range of severity
.
+
Figure 3. State- and use-dependent block of Na 1.1 voltage-gated Na channel
V
by 6d. (A): state-dependency (TP1); (B): use-dependency (TP2).
Small molecule sodium channel modulators frequently possess state- and use-
dependency. An ideal sodium channel blocker has much higher affinity to the
inactivated (use-dependency) than the resting conformation of the sodium channel
2
9-30
protein (state-dependency)
use-dependent 6d block of Na currents carried by Na
CHO-K1 cells, and the result is summarized in Table 3 and illustrated in Figure 3.
. In the present study, we compared the state- and
+
V
1.1 isoform expressed in

The state-dependent block of Na
inhibition ratio. By contrast, the use-dependent block of Na
d during repetitive pulses was strong at 10 µM, up to 99.9% inhibition ratio for
V
1.1 channel by 6d was weak, with a 19.5%
V
1.1 channel isoform by
6
Nav1.1. As indicated by this result, 6d was highly selective for use-dependent block
+
of Na
v
1.1 voltage-gated Na channel, which would have little toxic and side-effects
on the sodium channel of normal resting state.
Table 3. Electrophysiology studies for 6d. Anticonvulsant drugs on voltage dependent
sodium currents in CHO-K1 cells.
Current (pA)
Dependency
Inhibition (%)
Control
6d (10µM)
-969.6 ± 78.4
3.1 ± 6.6
-
1205.8 ± 117.0
state-dependency
use-dependency
19.5 ± 1.3
99.9 ± 0.1
-
702.7 ± 196.5
In order to illustrate the possible binding mode of the title compound 6d
towards the Na 1.1 channel, we have planned to perform molecular docking
simulations. However, because of the unavailability of the crystal structure of Na 1.1,
the 3D structure of Na 1.1 was constructed based on the homology modeling using
v
v
v
the crystal structure of the open-pore Electric eel Nav1.4 as the template (Supporting
Information, Figure S3, S4 and S5). The homology modeling algorithm generated
ten Na
Verify Protein (Profiles-3D)” protocol in Discovery Studio 4.5 version (DS).
Beyond that, the best model regarding the least Discrete Optimized Protein Energy
DOPE) score was chosen. Validation of the model was performed by analyzing
v v
1.1 models, and the quality of generated Na 1.1 models were assessed using
“
(
Ramachandran plot (Supporting Information, Figure S6.). The resulting
Ramachandran plot showed more than 98% residues distributed in the allowed
regions, indicating that the generated model is reasonable.

Figure 5. The 3D binding mode of 6d in our human Na 1.1 open-pore conformation
v
homology model was shown in the cartoon form. The helix was shown in red, and the
sheet strands were yellow. The green region was a space filling representation of
atoms lining the active site.
The 2D/3D binding model of 6d in our human Na 1.1 open-pore homology model
v
was shown in Figure 5 and 6. Besides, the sodium channel-blocking antiepileptic drug
phenytoin was performed by a molecular docking to compare the binding modes with
6
d in Nav1.1 (Figure 6). The most important residues in the binding mode of Na 1.1
v
include TRP668, ILE1067, TYR1068, PHE1071, IEU705, LEU708, ALA667,
LEU724, VAL1160 and LEU1159. The benzo[d][1,3]dioxole fragment of compound
6
d occupies the large hydrophobic pocket directed toward TRP668, ALA667,
LEU724, and LEU1159. Specifically, one of the ether oxygen groups at the
benzo[d][1,3]dioxole ring interacts with TRP668 of the Na 1.1 active domain via the
formation of a medium-strong hydrogen bond (3.33Å). This would help to stabilize
the open form and generate tighter binding to the catalytic sites of Na 1.1.
v
v
Furthermore, the interactions were also stabilized by a π-π stacked interactions with
TRP668 (3.64Å), and three alkyl interactions with ALA667 (4.21Å), LEU724 (5.39Å)
and LEU1159 (6.56Å). In addition to the above interactions, two alkyl interactions
with TRP668 (5.03Å) and ALA667 (4.84Å) are observed in the pyrrolidinone ring. It

is worth mentioning that the benzyl moiety plays an important role in the binding, as
the aromatic moiety is involved in a π-π stacked interaction with TYR1068 (5.83Å) as
well as six alkyl interactions with ILE1067 (6.57Å), TYR1068 (4.27Å), PHE1071
(
4.75Å), IEU705 (5.52, 6.08Å) and LEU708 (6.18Å). More importantly, the fluorine
atom of benzyl group presented in 6d showed a halogen bonding interaction with
ILE1067 (4.71Å) as well as a C-H bonding interaction with TYR1068 (3.50Å).
Figure 6. The 2D binding mode of 6d and Phenytoin in our Na 1.1 open-pore
v
conformation homology model
For comparison, phenytoin not only entirely occupied the hydrophobic pocket
but also formed an H-bond between the carbonyl oxygen atom and the residue
SER1152 (4.45Å). Apparently, two benzene rings of phenytoin are responsible for
two π-π stacked interactions as well as five alkyl interactions with the hydrophobic
residues of the inner cavity, such as TYR1068, ALA704, PHE1071, LEU705,
LEU708 and ILE1067 as the active pocket consisted of 13 amino acid residues in
Figure 6b. These docking study results demonstrated a minimal difference between
the binding modes of 6d and phenytoin. However, they still share four critical
residues (PHE1071, LEU705, LEU708 and ILE1067) in the active site of Na
v
1.1
receptor.

A computational study to predict the absorption, distribution, metabolism and
excretion (ADME) properties of the compounds was performed. Topological polar
surface area (TPSA) is a descriptor that is strongly correlated with passive molecular
transport through membranes, and accordingly predicts the transport properties of
3
1
drugs in the intestines and blood brain barrier crossing . Other included parameters
are the number of rotatable bonds, molecular volume, topological polar surface area,
percentage absorption, and in vitro plasma protein binding. These parameters allow
ascertaining oral absorption, or membrane permeability that occurs when evaluated
molecules obey Lipinski’s rule-of-five. The results were calculated for 5, 6a-6i and
were shown in Table 4. From the data, it could be observed that the title compounds 5,
and 6b-6i obeyed Lipinski's rule of five with good drug-like properties and membrane
permeability (log P≤5, MW<500, HBD<5 and HBA<10). In vitro plasma protein
binding (%) for the compounds 5, 6a-6i was given in Table 4. Compounds 5, 6a-6i
presented less than 92% in vitro plasma protein binding, thus suggesting weak plasma
3
2
protein binding of these compounds .
a
Table 4. Structural and pharmacokinetic properties of compounds 5 and 6a–6i
Comp.
Rule
5
MW
cLog P
≤5
HBA
HBD
nVio
≤1
0
n-ROTB
Volume
-
TPSA
-
% ABS
-
iPPB
-
<500
<10
4
<5
1
-
217.07
363.18
325.11
321.14
375.11
325.11
321.14
341.08
325.11
343.10
1.715
5.325
3.642
3.998
4.382
3.642
3.948
4.212
3.642
3.785
1
4
3
3
4
3
3
3
3
3
189.54
344.32
283.06
294.69
309.43
283.06
294.69
291.67
283.06
287.99
47.57
38.78
38.78
38.78
38.78
38.78
38.78
38.78
38.78
38.78
93.28
95.62
95.62
95.62
95.62
95.62
95.62
95.62
95.62
95.62
51.3
91.0
89.5
88.0
92.0
89.5
88.1
90.3
88.9
89.7
6a
4
0
1
6b
4
0
0
6c
4
0
0
6d
4
0
0
6e
4
0
0
6f
4
0
0
6g
4
0
0
6h
4
0
0
6i
4
0
0
a
cLog P, calculated partition co-efficient; MW, molecular weight; HBA, the number of hydrogen
bond acceptors; HBD, the number of hydrogen bond donors; nVio, the number of violations from
Lipinski’s rule-of-five; n-ROTB, the number of rotatable bonds; volume, molecular volume; iPPB,

in vitro plasma protein binding (%); TPSA, topological polar surface area; %ABS, absorption
percentage.
In conclusion, by using both the MES and PTZ tests, a series of 5-substituted
benzo[d][1,3]dioxole derivatives were designed, synthesized, and evaluated for their
anticonvulsant activities. Among them, 6d was found to be the most potent with ED50
value of 4.3 mg/kg and protective index (PI = TD50/ED50) value of 37.4. The in vivo
experiments demonstrated that it could block the Nav1.1 channel. Moreover, the study
results may promote the rational design of additional novel anticonvulsants.
Acknowledgments
This work was supported by the Jiangxi Provincial Natural Science Foundation
(
No. 20171BAB215072), the Traditional Chinese Medicine Research Project of
Jiangxi Provincial Health and Family Planning Commission (No.2017A304) as well
as Science and Technology Project of Jiangxi Provincial Health and Family Planning
Commission (No. 20185522).
References
1
2
3
4
5
6
7
.
.
.
.
.
.
.
Shekh-Ahmad, T., M. Bialer, E. Yavin, Epilepsy Res. 2012, 98, 238.
Lin, Z., P.K. Kadaba, Med. Res. Rev. 1997, 17, 537.
Wagner, M.L., Am. J. Hosp. Pharm. 1994, 51, 1657.
Zaccara, G., D. Franciotta, E. Perucca, Epilepsia 2007, 48, 1223.
Barkmeier D. T., Loeb J. A., Clin. EEG Neurosci. 2009, 40, 234.
Malik S., Ahuja P., Sahu K., Khan S.A., Eur. J. Med. Chem. 2014, 84, 42.
Ulloora S., Shabaraya R., Aamir S., Adhikari A.V., Bioorg. Med. Chem. Lett. 2013, 23,
1502.
8
9
.
.
Farghaly A.R., Esmail S., Abdel-Zaher A., Abdel-Hafez A., El-Kashef H., Bioorg. Med.
Chem. 2014, 22, 2166.
Wallace, S.J., Epilepsy Res. 1998, 29, 147.

1
0.
Alvarez, V., J.M. Januel, B. Burnand, A.O. Rossetti, Epilepsia 2011, 52, 1292
Moseley BD, Sperling MR, Asadi-Pooya AA, et al. Epilepsy research. 2016;127, 179.
Pollard J. R. Curr Opin Investig Drugs. 2008, 9, 101.
11.
1
1
2.
3.
Aboul-Enein MN, El-Azzouny AA, Attia MI, Maklad YA, Amin KM, Abdel-Rehim M,
El-Behairy MF. Eur. J. Med. Chem. 2012, 47, 360.
14.
Braga FG, Coimbra ES, de Oliveira Matos M, Lino Carmo AM, Cancio MD, da Silva AD.
Eur. J. Med. Chem. 2007, 42, 530.
1
1
5.
6.
Piao ZT, Guan LP, Zhao LM, Piao HR, Quan ZS. Eur. J. Med. Chem. 2008, 43, 1216.
Xie ZF, Chai KY, Piao HR, Kwak KC, Quan ZS. Bioorg. Med. Chem. Lett. 2005, 15,
4803.
1
2
1
1
2
2
2
2
7.
Li DD, Fang F, Li JR, Du QR, Sun J, Gong HB, Zhu HL. Bioorg. Med. Chem. Lett. 2012,
2, 5870.
8.
9.
0.
1.
2.
3.
Zimmer H, Armbruster DC, Trauth LJ. Heterocycl. Chem. 1965, 2, 171.
Krall RL, Penry JK, White BG, Kupferberg HJ, Swinyard EA. Epilepsia 1978, 19, 409.
Porter RJ, Cereghino JJ, Gladding GD, et al. Cleve. Clin. Q. 1984, 51, 293.
Dawidowski M, Herold F, Chodkowski A, Kleps J. Eur. J. Med. Chem. 2012, 48, 347.
Dunham NW, Miya TS. J. Am. Pharm. Assoc. 1957, 46, 208.
Pevarello P, Traquandi G, Bonsignori A, McArthur RA, Maj R, Caccia C, Salvati P,
Varasi M. Bioorg. Med. Chem. Lett. 1999, 9: 1783.
4. Pevarello P, Bonsignori A, Caccia C, Amici R, McArthur RA, Fariello RG, Salvati P,
2
Varasi M. Bioorg. Med. Chem. Lett. 1999, 9: 2521.
5. Obniska J, Byrtus H, Kamiński K, Pawłowski M, Szczesio M, Karolak-Wojciechowska J.
Bioorg. Med. Chem. 2010, 18, 6134.
2
2
2
2
2
6.
Zuliani V, Fantini M, Nigam A, Stables JP, Patel MK, Rivara M. Bioorg. Med. Chem.
010, 18, 7957.
7.
8.
Catterall WA, Kalume F, Oakley JC. J. Physiol. 2010, 588, 1849.
Volkers L, Kahlig KM, Verbeek NE, Das JH, van Kempen MJ, Stroink H, Augustijn P,
van Nieuwenhuizen O, Lindhout D, George AL Jr, Koeleman BP, Rook MB. Euro. J. neurosci.
2
2
3
011, 34, 1268.
9.
0.
Courtney KR, Kendig JJ, Cohen EN. J. Pharmacol. Exp. Ther. 1978, 207:594.
Chernoff DM. Biophys. J. 1990, 58: 53.

3
3
1.
2.
Ertl, P., Rohde, B., Selzer, P. J. Med. Chem. 2000, 43, 3714.
Selick HE, Beresford AP, Tarbit MH. Drug discovery today. 2002, 7, 109.

Graphical Abstract