In vitro and in silico evaluation of chromene based aroyl hydrazones as anticonvulsant agents

Article abstract of DOI:10.1007/s00044-017-1902-1

Series of aroyl hydrazones of 2H-chromene and coumarin carbaldehydes were synthesized and evaluated for their anticonvulsant activity and neurotoxicity. Further docking study on gamma-aminobutyric acid receptor was performed to elucidate their mechanisms of action. The highest protection was demonstrated by 2-furyl substituted 2H-chromene 8b in the maximal electroshock test (ED₅₀ = 12.51 mg kg^?1, PI MES > 23.98) and the subcutaneous pentylenetetrazole tests (ED₅₀ = 127.10 mg kg^?1). Furyl-substituted derivative 4b (ED₅₀ = 68.66 mg kg^?1) was the most active in the maximal electroshock test while methoxyphenyl-substituted derivate 4c was the most active in the 6-Hz test (ED₅₀ = 94.34 mg kg^?1). None of the compounds displayed neurotoxicity in the rota-rod test. In silico assessment of their blood-brain barrier permeability indicated them as central nervous system active agents. The results suggest that coumarin/2H-chromene aroyl hydrazones scaffold deserve further evaluation in models of epilepsy and derivatization.

Full text of DOI:10.1007/s00044-017-1902-1

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-017-1902-1
ORIGINAL RESEARCH
In vitro and in silico evaluation of chromene based aroyl
hydrazones as anticonvulsant agents
Violina T. Angelova1 Yulian Voynikov1 Pavlina Andreeva-Gateva2,3
●
●
●
Slavina Surcheva2 Nikolay Vassilev4 Tania Pencheva5 Jana Tchekalarova6
●
●
●
Received: 3 February 2017 / Accepted: 13 April 2017
© Springer Science+Business Media New York 2017
Abstract Series of aroyl hydrazones of 2H-chromene and
coumarin carbaldehydes were synthesized and evaluated for
their anticonvulsant activity and neurotoxicity. Further
docking study on gamma-aminobutyric acid receptor was
performed to elucidate their mechanisms of action. The
highest protection was demonstrated by 2-furyl substituted
indicated them as central nervous system active agents. The
results suggest that coumarin/2H-chromene aroyl hydra-
zones scaffold deserve further evaluation in models of
epilepsy and derivatization.
●
●
●
●
Keywords Coumarin Docking Hydrazones 6-Hz
MES PTZ
●
2H-chromene 8b in the maximal electroshock test (ED50
=
12.51 mg kg−1, PI MES > 23.98) and the subcutaneous
pentylenetetrazole tests (ED50 = 127.10 mg kg−1). Furyl-
substituted derivative 4b (ED50 = 68.66 mg kg−1) was the
most active in the maximal electroshock test while
methoxyphenyl-substituted derivate 4c was the most active
in the 6-Hz test (ED50 = 94.34 mg kg−1). None of the
compounds displayed neurotoxicity in the rota-rod test. In
silico assessment of their blood-brain barrier permeability
Introduction
Epilepsy is a chronic neurologic disease with high sus-
ceptibility to seizure generation and propagation, and with a
predisposition to neurobiological, cognitive, psychological,
and social disturbances (Fisher et al. 2005). Epileptic sei-
zures result from hyper synchronized discharge of neurons
and their clinical manifestation depends mainly on their
brain localization and spread (Terrone et al. 2016).
According to a recent meta-analysis, the median lifetime
epilepsy ranges from 5.8 per 1000 for developed countries,
to 15.4 per 1000 for developing countries (Ngugi et al.
2010). Although more than 25 new antiepileptic drugs
(AEDs) had been registered by 2015, it is well-known that
they neither prevent, nor cure completely this brain dis-
order, and are used mostly to suppress its symptoms
(Picaud et al. 2016). Moreover, it is known that up to 40%
of patients with epilepsy are drug-resistant (Mohanraj and
Brodie 2005). Most of AEDs exhibit adverse effects and
because of the need of chronic treatment the quality
of life of epilepsy patients can be aggravated. Therefore,
nowadays a lot of research is focused on discovering new
pharmacophores characterized with high activity and less
neurotoxicity.
* Violina T. Angelova
violina_stoyanova@abv.bg
violina@pharmfac.net
* Jana Tchekalarova
janetchekalarova@gmail.com
1
Department of Chemistry, Faculty of Pharmacy, Medical
University of Sofia, Sofia, Bulgaria
2
Department of Pharmacology and Toxicology, Medical University
of Sofia, Sofia, Bulgaria
3
Department of Pharmacology, Faculty of Medicine, Sofia
University “St. Kliment Ohridsky”, Sofia, Bulgaria
4
Institute of Organic Chemistry with Centre of Phytochemistry,
Bulgarian Academy of Sciences, Sofia, Bulgaria
5
Department of QSAR and Molecular Modelling, Institute of
Biophysics and Biomedical Engineering, Bulgarian Academy of
Sciences, Sofia, Bulgaria
6
Institute of Neurobiology, Bulgarian Academy of Sciences, Sofia,
Bulgaria

Med Chem Res
Over the last few years the molecular hybridization
strategy has emerged as a novel approach involving the
conglomeration of two or more pharmacophores in one
molecule in order to develop new hybrid multifunctional
molecules. The latter could potentially possess combined
pharmacological activities, a modified selectivity profile,
different or dual modes of action and/or reduced side
effects. Conjugation of coumarin with various pharmaco-
phores has yielded many novel hybrid molecules with a
wide spectrum of biological activities (Sandhu et al. 2014;
Costa et al. 2016; Medina et al. 2015). It is well known that
2H-chromene and coumarin derivatives are potential drug
candidates for treatment of multifactorial diseases such as
cancer, metabolic syndromes, acquired immune deficiency
syndrome, malaria, cardiovascular, and Alzheimer’s dis-
eases. Certain coumarin drugs, particularly of anticoagulant
and anti-neurodegenerative germane, have been success-
fully applied in clinic (Jameel et al. 2016; de Souza et al.
2016; Pisani et al. 2016). For years modified coumarin
derivatives have been synthesized as a privilege scaffold for
the design and development of anticonvulsants (Skalicka-
Woźniak et al. 2016; Arshad et al. 2014; Bhat and Al-Omar
2011), (Fig. 1). Additionaly, the coumarins as imperatorin,
osthole, and esculetin as well as the furanocoumarins (Zhu
et al. 2014; Singhuber et al. 2011) are known to exert
anticonvulsant activity comparable to that of valproate and
have been shown to interact with the benzodiazepine site of
the gamma-aminobutyric acid (GABAA) receptor which is a
target for the old class of AEDs, the benzodiazepines and
phenobarbital, and in part newer AEDs, such as gabapentin
and topiramate. In general, most AEDs affect the disturbed
balance of classical excitatory/inhibitory synapses via either
an enhancement of GABA-ergic inhibitory neurotransmis-
sion or attenuation of glutamatergic excitatory neuro-
transmission (directly or via inhibition of voltage-dependent
sodium and calcium channels) (Landmark 2006; Meldrum
and Rogawski 2007; Sucher and Carles 2015).
On the other hand, hydrazide-hydrazone derivatives have
also attracted much attention due to their remarkable
structural diversity and relevant biological activities (Verma
et al. 2014; Blair and Sperry 2013; Angelova et al. 2016a;
Dimmock et al. 2000). After pharmacophoric mapping, and
three-dimensional structure analysis, various aryl acid
hydrazones have been developed as anticonvulsant agent
(Ulloora et al. 2013; Jain et al. 2011; Cakir et al. 2001;
Ragavendran et al. 2007). Pandeya et al. (Sridhar et al.
2002; Pandeya et al. 2000) and Tripathi et al. (Tripathi et al.
2011) (PCH 6, Fig. 1) proposed that the following features
are important for anticonvulsant activity: (i) hydrophobic
aryl ring (Ar); (ii) a hydrogen bonding domain (HBD)—a
hydrogen donor (HD) and hydrogen acceptor (HA) units;
(iii) an electron-donor group (D), and (iv) another distal
hydrophobic site.
It is known that the variability in anticonvulsant drug
activity are due to numerous mechanisms of action (Mac-
donald and McLean 1985). Thus, docking study results
Fig. 1 Structures of well-known anticonvulsants and designed com-
pounds—coumarin (Bhat and Al-Omar 2011) and hydrazide deriva-
tives (Tripathi et al. 2011). The essential structural features were a
hydrophobic HP unit (R), an electron donor (D) group, and a hydrogen
donor HD and hydrogen acceptor HA unit

Med Chem Res
(Tripathi et al. 2011) showed that the pyridine-4-
carbohydrazides exhibited good binding properties with
glutamate, GABA (A) delta and GABA (A) alpha-1
receptor and hydrazones reported by Kumar et al. (Kumar
et al. 2012) exhibited good binding properties with gluta-
mate, GABA (A) alpha-1, GABA (A) delta receptor, and
Na/H exchanger. The aryl acid hydrazones, synthesized by
Sinha et al. (Sinha et al. 2011) showed excellent binding
properties with Lys 329 residue of gamma amino butyrate
amino transferase (GABA-AT). The hydrazide compounds
reported by Raj et al. (Raj Kumar and Kumar 2015)
exhibited good binding properties with established epilepsy
receptor namely Na/H exchanger and. Additionally, pro-
mising results were obtained for different hydrazone deri-
vatives (Kulandasamy et al. 2009, Tripathi et al. 2011; Jain
et al. 2011) in the 6 Hz psychomotor seizure test.
The literature survey revealed that hydrazide-hydrazones
and the coumarin derivatives could be considered as suc-
cessful pharmacophores in the design of novel antic-
onvulsant drugs. These observations encouraged us to
evaluate the anticonvulsant and toxicological profiles of a
series of coumarin and 2H-chromene based hydrazide-
hydrazones.
2H-chromene-3-carbaldehyde (7) was synthesised from
salicylaldehyde 5 and acrolein 6 under base-promoted
conditions as described previously (Azizmohammadi et al.
2013). Yield 62%, yellow solid, m.p. 35–3 °C (lit. m.p.
34–35 °C).
General synthetic procedure for compounds (4a–c) and
(8a–d)
The compounds were prepared by an established procedure
(Angelova et al. 2016b, 2017). An appropriate hydrazide
3a–d (2 mmol) was added to a stirred solution of
4-chlorocoumarin-3-carbaldehyde 2 or 2H-chromene-3-
carbaldehydes (2 mmol) 7: in 10–20 mL anhydr. ethanol.
The reaction mixture was vigorously stirred at room tem-
perature for 15 min. After 15 min to 1 h the clusters of
yellow crystalline compounds 4a–c and 8a–d deposited
from the solution. Upon filtration the crystals were washed
with aqueous ethanol to give TLC pure crystals.
4-chloro-N′-[(E)-(4-chloro-2-oxo-2H-chromen-3-yl)
methylidene]benzohydrazide (4a)
Yellow crystals, yield 0.400 g, 55%; m.p. 190–192 °C;
FTIR (ATR) νmax 3244, 1713, 1661, 1598 cm−1; H NMR
1
Materials and methods
Chemistry
(DMSO-d6, 600 MHz): 1:0.19 mixture of conformers; sig-
nals for major E-imine isomer syn-amide rotamer δ = 7.442
(d, J = 8.3 Hz, 1H, H-8), 7.469 (t, J = 7.9 Hz, 1H, H-6),
7.570 (d, J = 8.2 Hz, 2H, H-3′ and H-5′), 7.719 (t, J = 7.8
Hz, 1H, H-7), 7.899 (d, J = 8.2 Hz, 2H, H-2′ and H-6′),
7.997 (d, J = 7.8 Hz, 1H, H-5), 8.558 (bs, 1H, C-9), 12.229
(bs, 1H, NH); resolved signals for minor E-imine isomer
anti-amide bond rotamer δ = 12.012 (bs, 1H, NH) ppm;
13C-NMR (DMSO-d6, 150 MHz): δ = 116.62 (C-8), 118.41
(C-4a), 119.38 (C-3), 125.95 (C-6), 126.16 (C-5), 128.71
(C-3′ and C-5′), 129.74 (C-2′ and C-6′), 131.73 (C-1′),
133.77 (C-7), 136.92 (C-4′), 141.93 (C-9), 145.03 (C-4),
151.37 (C-8a), 157.74 (C-2), 162.06 (C-10) ppm; HRE-
SIMS m/z: found [M + H]+ 361.01392 (calcd: [M + H]+
361.01412).
The melting points were determined using a Buchi 535
apparatus. The FTIR spectra were recorded on a Nicolet
IS10 FT-IR Spectrometer (ThermoScientific, USA) using
ATR technique. All NMR experiments were carried out on
a BrukerAvance spectrometer II + 600 MHz at 20 °C in
DMSO-d6 as a solvent, using tetramethylsilane (TMS) as an
internal standard. The assignment of the H and 13C NMR
1
spectra was accomplished by measurement of 2D homo-
nuclear correlation (COSY), DEPT-135 and 2D inverse
detected heteronuclear (C–H) correlations (HMQC and
HSQC). Mass spectra were recorded on an LTQ Orbitrap
Discovery® spectrometer (ThermoFisher, Germany)
equipped with electrospray ionization module Ion Max®
(ThermoScientific, USA) operating in positive mode. All
chemicals and solvents were either purchased as puriss p.a.
from commercial suppliers or prepared by published
methods.
N′-[(E)-(4-chloro-2-oxo-2H-chromen-3-yl)methylidene]
furan-2-carbohydrazide (4b)
Yellow crystals, yield 0.140 g, 22%; m.p. 218–220 °C;
1
FTIR (ATR) νmax 3196, 3042, 1716, 1660, 1603 cm−1; H
The preparation of 4-chlorocoumarin-3-carbaldehyde
(2) by Vilsmeier–Haack reaction of 4 hydroxycoumarin 1
was described elsewhere (Heber et al. 1995). For 4-
chlorocoumarin-3-carbaldehyde 2: yield 84%, yellow
solid; m.p. 120–122 °C (lit. m.p. 120–122 °C) (Heber et al.
1995). HRESIMS m/z: found [M + H]+ 209.00002 (calcd.:
[M + H]+ 209.00000).
NMR (DMSO-d6, 600 MHz): 1:0.72 mixture of conformers;
signals for major E-imine isomer syn-amide rotamer δ =
6.725 (dd, J = 1.7, 3.5 Hz, 1H, H-4′), 7.360 (bs, 1H, H-3′),
7.516 (d, J = 8.0 Hz, 1H, H-8), 7.510 (dt, J = 1.1, 6.5 Hz,
1H, H-6), 7.766 (dt, J = 1.4, 7.8 Hz, 1H, H-7), 7.984 (dd,
J = 0.8, 1.7 Hz, 1H, H-5′), 8.027 (dd, J = 1.1, 8.4 Hz, 1H,
H-5), 8.625 (bs, 1H, C-9), 12.164 (bs, 1H, NH); resolved

Med Chem Res
signals for minor E-imine isomer anti-amide bond rotamer
δ = 11.944 (bs, 1H, NH) ppm; 13C-NMR (DMSO-d6, 150
MHz): δ = 113.15 (C-4a), 112.22 (C-4′), 115.51 (C-3′),
116.60 (C-8), 122.93 (C-3), 125.37 (C-6), 126.12 (C-5),
133.84 (C-7), 141.56 (C-9), 146.31 (C-5′), 146.78 (C-2′),
151.34 (C-4), 153.17 (C-8a), 158.52 (C-2), 159.08 (C-10)
ppm; HRESIMS m/z: found [M + H]+ 317.03236 (calcd.
[M + H]+ 317.03236).
1576; 1H NMR (DMSO-d6, 600 MHz, ppm) 1:0.12 mixture
of conformers; signals for major E-imine isomer syn-amide
rotamer δ = 5.034 (s, 2H, H-2), 6.670 (dd, J = 1.7, 3.5 Hz,
1H, H-4′); 6.826 (d, J = 8.3 Hz, 1H, H-8), 6.922 (dt, J =
1.0, 7.4 Hz, 1H, H-6), 6.929 (s, 1H, H-4), 7.184 (dt, J = 1.6,
8.3 Hz, 1H, H-7), 7.192 (d, J = 7.4 Hz, 1H, H-5), 7.257 (d,
J = 3.0 Hz, 1H, H-3′), 7.856 (bs, 1H, H-5′), 8.110 (s, 1H, H-
9), 11.959 (s, 1H, NH); resolved signals for minor E-imine
isomer anti-amide bond rotamer δ = 11.58 (s, 1H, NH);
13C-NMR (DMSO-d6, 150 MHz, ppm) δ = 64.33 (C-2),
112.88 (C-4′), 116.11 (C-3′), 116.21 (C-8), 122.29 (C-4a),
122.44 (C-6), 128.47 (C-7), 129.50 (C-3), 129.74 (C-4),
131.26 (C-5), 146.62 (C-2′), 146.66 (C-9), 146.62 (C-5′),
154.61 (C-8a), 155.11 (C-10); HRESIMS m/z: found
269.09198[M + H]+ (calcd. 269.092069 [M + H]+).
N′-[(E)-(4-chloro-2-oxo-2H-chromen-3-yl)methylidene]-4-
methoxybenzohydrazide (4c)
Yellow crystals, yield: 0.200 g, 30%; m.p. 229–230 °C.
1
FTIR (ATR) νmax 3200, 3031, 1716, 1651, 1602 cm−1; H
NMR (DMSO-d6, 600 MHz) 1:0.19 mixture of conformers;
signals for major E-imine isomer syn-amide rotamer δ =
3.841 (s, 3H, OCH3), 7.076 (bd, J = 7.0 Hz, 2H, H-3′ and
H-5′), 7.497 (t, J = 7.3 Hz, 1H, H-6),7.513 (d, J = 7.8 Hz,
1H, H-8), 7.756 (t, J = 7.7 Hz, 1H, H-7), 7.9552 (bd, J =
7.1 Hz, 2H, H-2′ and H-6′), 8.030 (d, J = 7.9 Hz, 1H, H-5),
8.630 (bs, 1H, C-9), 12.033 (bs, 1H, NH); resolved signals
for minor E-imine isomer anti-amide bond rotamer δ =
11.89 (bs, 1H, NH) ppm; 13C-NMR (DMSO-d6, 150 MHz):
δ = 55.49 (OCH3), 113.84 (C-3′ and C-5′), 116.60 (C-8),
118.43 (C-4a), 125.12 (C-1′), 125.37 (C-6), 126.09 (C-5),
129.78 (C-2′ and C-6′), 133.64 (C-7), 135.52 (C-4), 140.87
(C-9), 151.29 (C-8a), 157.77 (C-2), 162.29 (C-4′), 162.53
(C-10) ppm; HRESIMS m/z: found [M + H]+ 357.06357
(calcd. [M + H]+ 357.06366).
N′-[(E)-2H-Chromen-3-ylmethylidene]-4-
methoxybenzohydrazide (8c)
Yellow crystals, yield 62%, 0.382 g; m.p. 204–206 °C;
FTIR (ATR) cm−1 3219, 3070, 2836, 1632, 1602, 1580; 1H
NMR (DMSO-d6, 600 MHz, ppm) δ = 3.790 (s, 3H,
CH3O), 5.051 (s, 2H, H-2), 6.823 (d, J = 8.1 Hz, 1H, H-8),
6.910 (s, 1H, H-4), 6.930 (dt, J = 0.9, 7.5 Hz, 1H, H-6),
7.025 (d, J = 8.8 Hz, 2H, H-3′ and H-5′), 7.16–7.19 (m, 2H,
H-7 and H-5), 7.840 (d, J = 8.8 Hz, 2H, H-2′ and H-6′),
8.104 (s, 1H, H-9), 11.986 (s, 1H, NH); 13C-NMR (DMSO-
d6, 150 MHz, ppm) δ = 56.00 (CH3O), 64.40 (C-2), 114.42
(C-3′ and C-5′), 116.19 (C-8), 122.36 (C-4a), 122.41 (C-6),
125.42 (C-1′), 128.39 (C-5), 129.28 (C-4), 129.69 (C-3),
130.18 (C-2′ and C-6′), 131.14 (C-7), 146.81 (C-9), 154.58
(C-8a), 162.75 (C-4′), 163.70 (C-10); HRESIMS m/z: found
309.12335 [M + H]+ (calcd. 309.123369 [M + H]+).
4-Chloro-N′-[(E)-2H-chromen-3-ylmethylidene]
benzohydrazide (8a)
Yellow crystals, yield 68%, 0.424 g; m.p. 190–192 °C;
FTIR (ATR) cm−1 3378, 3195, 3031, 1653, 1623, 1592,
N′-[(E)-2H-Chromen-3-ylmethylidene]thiophene-2-
carbohydrazide (8d)
1565 cm−1 1H NMR (DMSO-d6, 600 MHz, ppm) δ =
;
5.055 (s, 2H, H-2), 6.835 (d, J = 8.3 Hz, 1H, H-8), 6.929
(dt, J = 0.9, 7.4 Hz, 1H, H-6), 6.955 (s, 1H,H-4), 7.18–7.21
(m, 2H, H-7 and H-5), 7.569 (d, J = 8.5 Hz, 2H, H-3′ and
H-5′), 7.857 (d, J = 8.5 Hz, 2H, H-2′ and H-6′), 8.110 (s,
1H, H-9), 11.970 (s, 1H, NH); 13C-NMR (DMSO-d6, 150
MHz, ppm) δ = 64.35 (C-2), 116.23 (C-8), 122.29 (C-4a),
122.50 (C-6), 128.50 (C-5), 129.30 (C-2′ and C-6′), 129.50
(C-3), 129.89 (C-4), 130.10 (C-3′ and C-5′), 131.31 (C-7),
132.23 (C-1′),137.47 (C-4′), 147.70 (C-9), 154.63 (C-8a),
163.26 (C-10). HRESIMS m/z: found 313.07384 [M +
H]+:(calcd.: 313.073832 [M + H]+).
Yellow crystals, yield 75%, 0.426 g; m.p. 233–234 °C;
FTIR (ATR) cm−1 3290, 3029, 2932, 1650, 1630, 1604,
1590; 1H NMR (DMSO-d6, 600 MHz, ppm) 1:0.96 mixture
of conformers; signals for major E-imine isomer syn-amide
rotamer δ = 5.041 (s, 2H, H-2), 6.831 (d, J = 8.3 Hz, 1H,
H-8), 6.926 (dt, J = 1.1, 7.5 Hz, 1H, H-6), 6.945 (s, 1H, H-
4), 7.17–7.20 (m, 3H, H-7 and H-5 and H-4′), 7.843 (s, 2H,
H-9 and H-5′), 7.825 (d, J = 5.0 Hz, 1H, H-3′), 11.963 (s,
1H, NH); resolved signals for minor E-imine isomer anti-
amide bond rotamer: δ = 5.148 (s, 2H, H-2), 7.906 (d, J =
4.6 Hz, 1H, H-5′), 7.980 (d, J = 2.6 Hz, 1H, H-3′), 8.098 (s,
1H, H-9), 11.737 (s, 1H, NH); 13C-NMR (DMSO-d6, 150
MHz, ppm) signals for major synperiplanar conformer
around the amide bond δ = 64.33 (C-2), 116.20 (C-8),
122.31 (C-4a), 122.44 (C-6), 128.46 (C-7), 129.22 (C-3),
129.84 (C-4), 131.24 (C-5), 135.79 (C-3′), 135.65 (C-5′),
N′-[(E)-2H-Chromen-3-ylmethylidene]furan-2-
carbohydrazide (8b)
Yellow crystals, yield 72%, 0.386 g; m.p. 204–205 °C;
FTIR (ATR) cm−1 3246, 3030, 2850, 1652, 1632, 1602,

Med Chem Res
138.07 (C-2′), 147.07 (C-9), 154.39 (C-8a), 163.37 (C-10);
resolved signals for minor antiperiplanar conformer around
the amide bond δ = 64.74 (C-2), 116.20 (C-8), 122.18 (C-
4a). HRESIMS m/z: found 285.06915[M + H]+ (calcd.
285.069224 [M + H]+).
6-Hz psychomotor seizure test
Like the MES test, the 6-Hz seizure test (Barton et al. 2001;
Toman et al. 1951) was used to assess the efficacy of the
compounds against electrically induced seizures but at a
lower frequency (6 Hz) and longer duration of stimulation
(3 s). Mice (n = 8) were challenged with a sufficient current
delivered through corneal electrodes to elicit a psychomotor
seizure in 97% of animals (32 mA for 3 s) at testing time of
0.5 and 4.0 h, respectively. The seizure in the controls is
characterized by a minimal clonic phase followed by ste-
reotyped, automatistic behaviors, including twitching of the
vibrissae, and Straub-tail. If the animal resumed its normal
exploratory behavior within 10 s after stimulation this was
considered as a protection. The derivatives were investi-
gated quantitatively in the 6-Hz test and ED50 was reported.
Anticonvulsant screening
Animals
Male albino ICR mice weighing 25–30 g were used as
experimental animals. The animals were maintained at an
ambient temperature 22 1 °C, in groups of six per cage
under standard laboratory conditions and allowed free
access to food and water. All procedures were performed in
agreement with the European Communities Council
Directive 2010/63/EU. The experimental design was
approved by the Institutional Ethics Committee at the
Institute of Neurobiology.
Neurotoxicity screening
The minimal motor impairment in mice was measured using
the rota-rod test. The mice were trained to stay on an
accelerating rota-rod of diameter 3.2 cm rotating at 10 rpm.
Neurotoxicity was indicated by the inability of the animal to
maintain equilibration on the rod for at least 1 min in each
of the three trials. The dose at which 50% of the animals
were unable to balance themselves and fell off the rotating
rod was determined as toxic.
Drugs and dosage
Control group (treated with vehicle) and experimental
groups (treated with either referent drugs phenytoin sodium
and diazepam or tested compounds) were injected intra-
peritoneally (i.p.) at different doses and were dissolved in
10% DMSO, at a volume of 10 mL kg−1. The antic-
onvulsant activity was assessed at two time points 0.5 and
4 h, respectively.
Quantification studies
Maximal electroshock test (MES test)
ED50 and TD50, which are the concentrations that give a
response in 50% subjects in seizure and toxicity tests, were
determined as best-fit values. Protective index (PI) was
calculated as a ratio of TD50/ED50. The dose of drug
required to produce the desired end point in 50% of animals
in each test, the 95% confidence interval, the slope of the
regression line and standard error of the slope were then
calculated by Prism 6.07 software. Both dose-response and
toxicity data were fitted using non-linear sigmoidal dose-
response model with variable slope: Y = Bottom +
A drop of a local anesthetic was applied to the eyes of each
animal and corneal electrodes were applied followed by an
electric stimulus of 50 mA, 60 Hz delivered for 0.2 s
(Constant Current Shock Generator). Abolition of the hind
limb extensor component following drug treatment was
taken as an end point of the test. The tonic component was
considered abolished, if the hind limb extension did not
exceed a 90o with the plane of the body. Abolition of hind
limb tonic extensor component of the seizure in half or
more of the animals was defined as protection.
̂
(Top-Bottom)/(1 + 10((LogED50−X)*HillSlope)), where
Bottom, Top, HillSlope, and ED50 were the fitting para-
meters, X was the logarithm of drug concentration, and Y
was the response parameter.
Subcutaneous pentylenetetrazole (scPTZ) seizure test
For the scPTZ test the compounds that raise seizure
threshold were primarily identified. A dose of 85 mg kg−1
utilized at the anticipated time of testing (0.5 and 4.0 h)
during the scPTZ test, produced clonic seizures lasting for a
period of at least 5 s in 97% (CD97) of animals tested. The
mice were further observed during a period of 20 min. The
protection against clonic seizures in half or more of the
animals was defined as an anticonvulsant activity.
Molecular docking studies
Molecular docking of synthesized compounds was performed
into a human GABAA receptor using Molecular Operating
Environment (MOE, version 2016.08) of Chemical Comput-
ing Group (https://www.chemcomp.com/MOE-Molecular_
Operating_Environment.htm). The crystal structure of
GABAA protein–ligand complex was downloaded from

Med Chem Res
Protein Data Bank (http://www.rcsb.org/, PDB ID 4COF). In
this crystal structure, GABAA receptor was crystallized bound
to a previously unknown agonist, benzamidine (BEN) (Miller
and Aricescu 2014). Before processed to docking the protein
structure was assigned the correct ionization states and the
missing hydrogen atoms were positioned with the tool “Pro-
tonate 3D” of MOE. The molecular docking was performed in
“Docking” module in MOE. No changes are made in the
default settings of docking procedure. The top 30 poses as
ranked by London dG are kept and minimized using
MMFF94x within a rigid receptor. The resulting poses are
then scored using the new GBVI/WSA dG scoring function
described previously and 5 best poses are recorded.
(Angelova et al. 2016b). The synthesis of hydrazide-
hydrazones 4a–c and 8a–d was performed as follows:
equimolar
concentration
of
4-chlorocoumarin-3-
carbaldehyde 2 or 2H-chromene-3-carbaldehyde 7 reacted
with appropriate hydrazides 3a–d in abs. ethanol, at room
temperature. The synthetic scheme is illustrated in Scheme 1.
Synthesized compounds 4a–c and 8a–d were character-
ized by FTIR (ATR), 1H NMR, 13C NMR, 2D NMR
(1H–1H COSY, 1H–13C COSY, HMBC, and NOESY) and
1
HRMS spectral techniques. The H NMR spectra of the
compounds showed the characteristic NH proton at δ
12.22–11.96 ppm, and proton of –N=C–H at δ 8.63–8.10
1
ppm. The two sets of resonances observed in the H NMR
spectra of the studied hydrazones in DMSO-d6 are due to
the hindered rotation in the CO–NH group. According to
the observed NOE effect between the HC=N proton and the
NH proton the hydrazone derivatives are a single E geo-
metrical imine isomer syn-amide rotamer. All the synthe-
sized compounds comprised the essential pharmacophoric
elements (Scheme 1) necessary for good anticonvulsant
activity (Tripathi et al. 2011).
In silico calculation of physicochemical parameters
The physicochemical properties of the synthesized
hydrazide-hydrazone derivatives and controls were calcu-
lated using the Marvin 16.2.8.0 software (http://www.chema
xon.com). The calculated properties are logarithm of the
partition coefficient between n-octanol/water (LogP), polar
surface area (PSA), hydrogen bond acceptors count (nON),
hydrogen bond donors count (nOHNH), molecular weight
and rotatable bonds count (n-rotb) and the logarithm of brain/
plasma concentration (logBB). To estimate the blood-brain
barrier (BBB) permeability, we implemented the prediction
model described by Clark 1999, using the equation: logBB
= −0.0148*PSA + 0.152* clogP + 0.139 (Clark 1999).
Pharmacology
Following guidelines from the Epilepsy Therapy Screening
Program (ETSP) at the National Institute of Neurological
Disorders and Stroke (https://panache.ninds.nih.gov), we
conducted tests in mice after i.p. injection at initial doses of
30, 100, and 300 mg/kg for primary evaluation of the syn-
thesized compounds. In parallel, acute neurotoxicity was
determined via the test for minimal motor impairment (rota-
rod). Phenytoin and diazepam were used as reference drugs.
Initial screening evaluation was executed at two different
time intervals, namely 0.5 and 4 h. The results are shown in
Table 1. Additionally, quantitative assessment for
Results and discussion
Synthesis
For the synthesis of the novel compounds we used classical
condensation methods described in a previous work
Scheme 1 Synthesis of compounds 4a–c and 8a–d. Reaction conditions: a) POCl3, DMF, 60 °C; b) abs. EtOH, r.t., 15 min

Med Chem Res
anticonvulsant potency of the most promising compounds
was conducted at a time peak of their pharmacological
activity (Tables 2 and 3).
tonic-clonic seizures at the highest dose of 300 mg kg−1
after 0.5 h. Moreover, two coumarin derivatives, 4a and 4c,
exhibited anti-seizures activity in the MES at doses of 100
and 300 mg kg−1, respectively, at a time point of 4 h.
In the scPTZ test, only two of the chromene derivatives
—8a and 8b, exhibited a capability to raise the seizure
threshold at the highest dose of 300 mg.kg−1 at 0.5 h. It is
known that PTZ induce seizures by inhibiting GABA
neurotransmission (Okada et al. 1989). The inhibition of
GABAergic neurotransmission or activity has been shown
to promote and facilitate seizures, (Krall et al. 1978; Okada
et al. 1989) while the enhancement of GABAergic neuro-
transmission is known to inhibit or attenuate seizures. The
finding that compounds 8a and 8b were inhibitors of sei-
zures induced by PTZ suggests that chromene derivatives
might function by enhancing GABAergic neurotransmis-
sion. However, further studies are needed to ascertain the
precise anticonvulsant activity mechanism of these com-
pounds. In the rota-rod test, both 2H-chromene and
coumarin-based hydrazide/hydrazones were devoid of
neurotoxicity at the maximum dose administered (300 mg
kg−1) compared to the referent drugs phenytoin (100 mg
kg−1) and diazepam (30 mg kg−1) at two time points
(2 and 4 h).
Some clinically useful AEDs are ineffective in the
standard MES and scPTZ tests but can demonstrate antic-
onvulsant activities in the 6-Hz seizure test. The 6-Hz test is
considered as a model of drug-resistant epilepsy (Potschka
2012) and seems to be predominantly resistant to sodium
channel modulators. Drugs with other mechanisms, parti-
cularly GABAergic compounds, are quite effective. Thus,
this test is certainly not a model of drug-refractory seizures
but may help to discriminate drugs during development.
Noteworthy is that when using this test, it is important to
note that the genetic background of mice strongly affects
drug resistance (Löscher 2016). Our results from screening
the compounds at 0.5 and 4 h time points are summarized in
Table 4. As observed compound 4c displayed better activity
Except for the inactive 2H-chromene derivatives 8c and
8d, all other tested 2H-chromene and coumarin-based
hydrazones demonstrated potency in the MES test. As
shown in Table 1, the most active compound was furyl-
substituted 2H-chromene derivative 8b, which revealed a
protection in the MES test at the lowest dose administrated
at a 0.5-h interval. The effect was comparable to that of the
referent drug phenytoin. In the MES test, two coumarin
derivatives, 4a and 4b, were potent at a dose of 100 mg
kg−1, while the 4-methoxybenzohydrazide derivative with
coumarin scaffold 4c and the 4-chlorobenzohydrazide
derivative with 2H-chromene scaffold 8a suppressed
Table 1 Anticonvulsant activity and neurotoxicity of the tested
compounds (mice, i.p.)
Compd.
MES
0.5 h
scPTZ
0.5 h
Rota-roda
0.5 h
4.0 h
4.0 h
4.0 h
4a
100
100
300
300
30
–
100
–
–
–
–
–
–
–
–
–
4b
4c
300
–
–
–
–
–
8a
300
300
–
–
–
–
8b
–
–
–
–
8c
–
–
–
–
8d
–
–
–
–
–
–
Phenytoinb
Diazepamb
30
–
30
–
–
–
100
–
100
–
30
30
The data in the table indicate the minimum dose, whereby bioactivity
or neurotoxicity was demonstrated in at least 50% of treated animals.
A dash indicates the absence of activity or neurotoxicity at the
maximum dose administrated (300 mg kg−1
)
MES maximal electroshock test, scPTZ subcutaneous pentylenetetra-
zole test
a
Test for neurotoxicity (rota-rod)
b
Data from Ref. (Dimmock et al. 1995)
Table 2 Quantitative anticonvulsant data in mice (test drug administered i.p.)
Compd.
Test
TPE(h)
ED50
95%confidence
interval
Hill slope
Standard errors
of the slope
TD50
PI
(mg kg−1
)
(mg kg−1
)
4a
4b
4c
8a
MES
MES
MES
MES
PTZ
0.5
0.5
0.5
0.5
0.5
0.5
0.5
99.71
68.66
81.29
87.63
218.50
12.51
127.10
(91.87–108.20)
(28.62–164.70)
(32.00–201.50)
(18.51–414.80)
(144.70–329.80)
(5.46–28.7)
−3.22
1.05
0.24
0.28
0.39
0.25
0.51
0.25
1.36
>300
>300
>300
>300
>300
>300
>300
>3.01
>4.37
>3.69
>3.42
>1.37
>23.98
>2.36
−1.20
0.64
−2.30
1.12
8b
MES
PTZ
(51.00–316.60)
−2.82
The values represented are in the 95% confidence interval
TPE time to peak effect, ED50 median effective doses, TD50 median minimal neurotoxic doses, PI protective index (rotarod TD50/ED50
)

Med Chem Res
Table 3 Quantitative screening of selected compounds by 6-Hz seizure test (current 32 mA) in mice
Compd.
Test
TPE(h)
ED50
95% confidence
interval
Hill
slope
Standard errors
of the slope
TD50
PI
(mg kg−1
)
(mg kg−1
)
4b
4c
8c
6-Hz
6-Hz
6-Hz
0.5
0.5
0.5
137.3
94.34
198.1
(85.92–324.70)
(57.55–154.70)
(92.11–326)
0.76
2.27
1.08
0.28
0.24
0.39
>300
>300
>300
>2.18
>3.18
>1.51
TPE time to peak effect, ED50 median effective doses, TD50 median minimal neurotoxic doses, PI protective index (rotarod TD50/ED50
)
compounds 4c and 4b showed higher potency than the
chromene analogue 8c (Table 3).
Table 4 Results of anticonvulsant activity by 6-Hz psychomotor
seizure test
The PI values were in the range of 3.01–23.98 in the
MES test and in the range of 1.37–2.36 in the scPTZ test.
The PI value is a ratio of the TD50 and ED50 values, which
indicates the safety of the compounds tested in the animal
models. In the MES test, the PI of 8b was superior to those
of other tested compounds (>23.98) and higher than the PI
reported earlier for the reference drug phenytoin (6.9)
(Löscher 2016). For other tested compounds 4a, 4b, 4c, and
8a the PI was >3. However, regarding the scPTZ test, the PI
of 8a and 8b, respectively, was lower than that of the
reference drug diazepam (Sandhu et al. 2014) (>1.37 and
>2.36 vs. 12.00). In summary, among all tested compounds,
compound 8b, carrying a furyl substituent, exhibited the
strongest potency with higher ED50 and PI compared to the
reference drug phenytoin in the MES test. The PI values of
compounds 4b, 4c, and 8c indicated good margin of safety
in the 6-Hz test and tolerability between anticonvulsant
doses.
Compd.
0.5 h
4.0 h
4a
–
300
100
–
–
–
4b
4c
–
8a
–
8b
–
–
8c
300
–
–
8d
–
Phenytoin
10
10
compared to the other compounds. In particular the results
of the 6-Hz test suggest that the compounds 4b, 4c, and 8c
have useful pharmacological properties and efficiency in
partial epilepsy as well as refractory seizures in humans.
Quantification studies
The structure–activity relationship of the compounds 4a–
c and 8a–d was analyzed using their activity in the MES test
as follow:
On the basis of the above preliminary screening data in
mice, except for the inactive compound 8c and 8d, all 2H-
chromene and coumarin-based hydrazones we evaluated
against the pharmacological parameters (ED50 and TD50) in
the MES and scPTZ test in mice after i.p. administration
(Table 2).
The quantitative data were examined at a previously
estimated time to peak effect (TPE). In the MES test, furyl-
substituted chromene derivative 8b being as potent as
phenytoin (Robertson et al. 1988) and more potent than
valproate (Löscher 2016) exhibited the most beneficial
activity (ED50 = 12.51 mg kg−1). Coumarin-based hydra-
zide/hydrazones 4a–c and 2H-chromene analogue 8a
demonstrated comparable efficacy for abolishing tonic-
clonic seizures in the MES test at median doses 99.71,
68.66, 81.29, and 87.63 mg kg−1, respectively. In the scPTZ
test a 2-furyl-substituted 2H-chromene derivative 8b
i) the presence of a coumarin scaffold in derivatives 4a–
c displayed good results in the MES test irrespective
of the substituents present in the molecule;
ii) the long-term efficacy of coumarin derivatives 4a and
4c compared to that of 2H-chromene derivatives 8a
and 8b in the MES test might be due to the presence
of the coumarin scaffold.
iii) the variety of the substituted group on the phenyl ring
and changing with a furyl or thienyl moiety appeared
to greatly influence the anticonvulsant activity.
Thus, replacement of the furyl moiety in compounds
4b, 8b with a 4-chlorophenyl moiety in compounds
4a and 8a resulted in a decreased anti-seizure efficacy
of the coumarin molecule 8a only. However, 4-
chlorophenyl substituted coumarin 4a and 2H-chro-
mene derivative 8a displayed better activity in
comparison with that of methoxyphenyl substituted
derivatives 4c and 8c as well as with 2-thienyl
derivative 8d.
showed higher activity compared to that of 8a (ED50
=
127.10 and 218.50 mg kg−1, respectively). However, in the
6-Hz seizure test, the 2H-chromene derivatives were less
active than coumarin derivatives. Thus, coumarin

Med Chem Res
In the scPTZ test, from the two selected 2H-chromene
compounds 8a and 8b, again the 2-furyl-substituted chro-
mene derivative 8b showed higher activity than that of 8a
(ED50 = 127.10 and 218.50 mg kg−1, respectively). Thus
compounds 8a and 8b with 2H-chromene scaffold dis-
played a broad spectrum of activity in several models and
were likely to have several mechanisms of action (including
inhibiting GABAergic activity and voltage-gated ion
channels). In the 6-Hz seizure test, the coumarin scaffold
seems to have a stronger impact suggesting that the cou-
marin scaffold is more crucial for the anticonvulsant activity
than the methoxyphenyl group in this test.
The results obtained clearly indicate that the
hydrazide–hydrazone chain is necessary for the optimal
anticonvulsive effect but variations of the substitution in the
aromatic rings and the chromene moiety lead to remarkable
changes in the anticonvulsant activity.
protein. The main interactions include hydrogen bonds
(HBs), salt bridges, hydrophobic and cation-π interactions,
and solvent exposure. Close, but non-bonded residues
(approaching the ligand, but not having any strong inter-
actions, i.e. HBs) are also reported in the ligand interaction
diagram. Figure 2 presents the protein–ligand interactions
(PLI) diagram of the co-crystalized ligand of GABAA in the
ligand-binding domain, while Fig. 3 demonstrates the PLI
diagrams of the synthesized compounds, docked in the
GABAA ligand-binding domain.
Molecular docking studies of the synthesized compounds
with the GABAA receptor protein displayed good docking
scores (Table 5). The most active derivative 8b as well as
4b in the in vivo study demonstrated best results in mole-
cular docking studies also. As seen from Fig. 2, three
residues are involved in interactions, namely Glu155,
Ser156, and Tyr157, while residues Thr202, Tyr205, Tyr97,
and Phe200 are approaching the ligand, but not having any
strong interactions. Compound 4b (Fig. 3a), presenting the
best pose, shows an interaction with Glu155 as in Fig. 2)
and additionally demonstrates two newly appeared inter-
actions with Tyr202 and Phe200. Residues Tyr97, Tyr157,
Tyr205 remain too close, but still in a non-binding distance.
Compound 8b (Fig. 3b), presenting the top score, shows an
interaction with Tyr157 as in Fig. 2) and additionally
demonstrates a newly appeared interaction with Tyr202.
Residues Tyr97, Glu155, Tyr157 again remain too close,
but still in a non-binding distance.
Molecular docking study
Two docking solutions can be regarded as “standard” in a
typical docking analysis, namely: (1) the top score—the
pose with the highest rank; and, (2) the best pose—the pose
with lowest root mean square deviation (RMSD) to the
reference ligand from the experimentally solved structure.
Table 5 lists the essential of docking results for the
synthesized compounds, based on RMSD and S-value,
equal to the E_score2 —the final energy score from the
rescoring stage 2 (in kcal.mol−1).
As seen from the results obtained, the lowest RMSD
(bolded in Table 5) was obtained for the compound 4b,
while the highest rank was recorded to the compound 8b
(bolded in Table 5, meanwhile with the “second best”
RMSD). Further analysis of the molecular docking studies
is going to be performed based on protein–ligand interac-
tions (PLI), applying the MOE tool “Ligand Interactions”
for visualization of the active site of a complex in a diagram
form. The tool aims at identification and visualization of the
interactions between the ligand and the receptor-interacting
entities, solvent molecules and ions in the active site of the
Docking study results shows that the compounds 4b and
8b exhibited good binding properties with GABA (A)
Table 5 Docking results for the synthesized compounds
Compound
RMSD
S-value
4a
4b
4c
8a
8b
8c
1.88
0.92
1.60
1.89
1.38
1.92
−4.42
−4.25
−4.46
−4.27
−4.84
−4.30
RMSD root mean square deviation; S-value, equal to E_score2 - the
final energy score from the rescoring stage 2 (in kcal mol−1
Fig. 2 Interaction diagram of the ligand-binding domain of GABAA
with benzamidine (PDB ID 4COF).
)

Med Chem Res
Fig. 3 Interaction diagram of the ligand-binding domain of GABAA
with synthesized compounds (the legend is equal to the presented
above in Fig. 2). a - compound 4b, presenting the best pose—the pose
with lowest RMSD. b - compound 8b, presenting the top score—the
pose with the highest rank (lowest energy)
receptor. However, other important anticonvulsant targets
such as glutamate receptors and calcium channels are not
evaluated here to ascertain the precise mechanism of action
of anticonvulsant activity of these compounds. That leads to
differences in the relative relations between the model and
experimental activities. In addition, any discrepancy
between the docking data and the actual activity of the
tested compounds could be ascribed to pharmacokinetic and
biopharmaceutical properties, which make a given com-
pound better accessible by receptor site and condition
superior activity even when the binding interaction is a less
prominent one.
good permeability across cell membrane and thus could
influence the bioavailability of the drugs and presumably
the good pharmacokinetic behavior. Molecular weight of
hydrazone derivatives was found to be lower than 400. Thus
these molecules are anticipated to be transported, diffused,
and absorbed easier compared to large molecules. The
number of hydrogen bond acceptors (O and N atoms),
hydrogen bond donors (NH and OH) and rotatable
bonds in the synthesized compounds are in accordance with
the Lipinski’s rule-of-five i.e. less than 3, less than 5 and
less than 10, respectively. Additionally, the molecular
polar surface area (PSA) is a very useful parameter
for the prediction of drug transport properties. PSA
is the sum of surfaces of polar atoms (usually oxygen,
nitrogen and attached hydrogen) in a molecule. Van de
Waterbeemd et al. suggested that compounds showing CNS
activity could be found within the ranges of PSA within 0
and 90 (van de Waterbeemd et al. 1998) and that with
decreasing PSA the likelihood of CNS activity appears to
increase.
In silico calculation of physicochemical parameters and
prediction of blood-brain barrier (BBB) penetration
Intestinal absorption and central nervous system (CNS)
penetration have been shown to be influenced by
physicochemical properties. The development of many
novel CNS-acting drug candidates has been hampered
by low BBB permeability, resulting in failure in the
clinical trials. Nowadays, this potential drawback can
be circumvented by the use of in silico models that allow
pre-screening of compound sets prior to the synthesis.
As stated by the famous Lipinski’s rule-of-five, for oral
absorption, a drug should be characterized by MW ≤ 500,
logP ≤ 5, number of hydrogen bond donors (HBD) ≤ 5
and number of hydrogen bond acceptors (HBA) ≤ 10
(Lipinski 2004).
Prediction of BBB penetration
Substances cross the blood-brain barrier (BBB) by a variety
of mechanisms as trans-membrane diffusion, saturable
transporters, adsorptive endocytosis, and the extracellular
pathways. Considering that trans-membrane diffusion is
universal for membranes, it has been suggested that the
factors governing a compound’s intestinal absorption could
be used to model its BBB permeability (Banks 2009). The
calculation of straightforward physicochemical descriptors
is the only requirement for the prediction of the logBB of
novel compounds, which can be conveniently applied in
All synthesized compounds are in agreement with the
formal Lipinski’s rule-of-five. As evident from the calcu-
lated logP values which are less than 5 (Table 6), the
compounds have acceptable water solubility, suggesting

Med Chem Res
Table 6 Physicochemical parameters for the synthesized compounds and standards
Compd.
logP
PSA
natoms
nO,N
nOH,NH
nrotb
MW
logBBa
Pred. CNS activity
4a
2.18
0.61
1.41
3.32
1.75
2.55
2.62
2.28
3.01
67.76
80.90
76.99
50.69
63.83
59.92
50.69
58.20
32.67
34
31
38
35
39
32
32
31
33
3
3
4
3
3
4
3
2
2
1
1
1
1
1
1
1
2
0
3
3
4
3
3
4
3
2
1
361.1780
316.6970
356.7620
312.7530
268.2720
308.3370
284.3330
252.2730
284.7430
−0.53
−0.97
−0.79
−0.11
−0.54
−0.36
−0.21
−0.38
0.11
Active
Active
Active
Active
Active
Active
Active
Active
Active
4b
4c
8a
8b
8c
8d
Phenytoin
Diazepam
logP Logarithm of partition coefficient between n-octanol/water, PSA Polar surface area, nON Hydrogen bond acceptors count, nOHNH Hydrogen
bond donors count, nrotb Rotatable bonds count
a
Logarithm of brain/plasma concentration (logBB). The parameters were calculated by Marvin 16.2.8.0 software (http://www.chemaxon.com)
conjunction with drug design and virtual screenings. The
polar surface area (PSA), which represents a compound’s
capacity to form hydrogen bonds, and logP have been
shown to correlate with passive membrane transport in the
intestines and were further suggested to be useful in mod-
eling BBB permeability. The logBB, defined as the loga-
rithm of the ratio of the concentration of a drug in the brain
and in the blood, measured at equilibrium, is an index of
BBB permeability.
For estimating the BBB permeability, we have used the
prediction model described by Clark 1999, using the
equation: logBB = −0.0148*PSA + 0.152*clogP + 0.139
(Clark 1999). The calculated logBB values range from
−0.97 to −0.11 (Table 6). Published values of logBB range
from −2.00 to + 1.00 (Ahangar et al. 2011) and according
to Abraham et al. compounds with logBB > 0.3 cross the
BBB readily, while those with logBB < −1.0 are poorly
distributed to the brain (Abraham et al. 1997; Vilar
et al. 2010). Therefore, according to the calculated data
(Table 6), all synthesized compounds possess moderate to
good BBB permeability. Most of the compounds possess
comparable logBB values to the referent phenytoin.
Nevertheless, the prediction model of Clarkis valid for
passive diffusion processes across the BBB and does not
account for other active transport systems for both influx
and efflux in the brain or binding of drugs by plasma protein
(Clark 1999).
capability to inhibit seizure spread which was further con-
firmed by the docking simulations, as a GABAA receptor
agonist. The two 2H-chromene derivatives 8a and 8b were
active in the scPTZ test, which is used to identify com-
pounds that elevate the seizure threshold. In the 6-Hz psy-
chomotor seizure test, coumarin scaffold seems to play an
important role for the anticonvulsant activity of newly
synthetized derivatives. In silico assessment of their BBB
permeability indicated the hydrazide–hydrazone derivatives
as CNS active agents making them potentially promising
agents for epilepsy therapy. Moreover, favorable in silico
performance was observed for the synthesized derivatives
with excellent correlation to CNS related parameters, which
in turn support anticonvulsant efficacy, as observed in the
in vivo studies. Finally, our results suggest that the cou-
marin/2H-chromene aroyl hydrazones scaffold deserve
further evaluation in models of epilepsy as well as further
derivatization.
Acknowledgements Our work was supported financially by the
Council of Medical Sciences of the Medical University of Sofia,
Bulgaria—Grant No. 7/2016.
Compliance with ethical standards All procedures performed in
studies involving animals were in accordance with the ethical stan-
dards of the institution or practice at which the studies were calculated.
Conflict of interest The authors declare that they have no competing
interests.
Conclusion
References
The present work reports the synthesis and anticonvulsant
activities of series of hydrazide–hydrazone derivatives 4a–c
and 8a–d. The 2-furyl-substituted 2H-chromene and cou-
marin derivatives 8b and 4b, respectively have been found
to be the most active compounds among the tested chro-
mene based aroyl hydrazones in the MES test suggesting a
Abraham MH, Takács‐Novák K, Mitchell RC (1997) On the partition
of ampholytes: application to blood–brain distribution. J Pharm
Sci 86:310–315
Ahangar N, Ayati A, Alipour E, Pashapour A, Foroumadi A, Emami S
(2011) 1‐[(2‐Arylthiazol‐4‐yl) methyl] azoles as a new class of
anticonvulsants: design, synthesis, in vivo screening, and in silico
drug‐like properties. Chem Biol Drug Des 78:844–852

Med Chem Res
Angelova V, Karabeliov V, Andreeva‐Gateva PA, Tchekalarova J
(2016a) Recent developments of hydrazide/hydrazone derivatives
and their analogs as anticonvulsant agents in animal models.
Drug Dev Res 77:379–392
Angelova VT, Valcheva V, Vassilev NG, Buyukliev R, Momekov G,
Dimitrov I, Saso L, Djukic M, Shivachev B (2017) Anti-
mycobacterial activity of novel hydrazide-hydrazone derivatives
with 2H-chromene and coumarin scaffold. Bioorg Med Chem
Lett 27:223–227
Angelova VT, Vassilev NG, Nikolova-Mladenova B, Vitas J, Malbaša
R, Momekov G, Djukic M, Saso L (2016b) Antiproliferative and
antioxidative effects of novel hydrazone derivatives bearing
coumarin and chromene moiety. Med Chem Res 25:2082–2092
Arshad MF, Siddiqui N, Elkerdasy A, Al Rohaimi AH, Khan SA
(2014) Anticonvulsant and neurotoxicity evaluation of some
newly synthesized thiazolyl coumarin derivatives. Am J Phar-
macol Toxicol 9:132
Krall R, Penry J, White B, Kupferberg H, Swinyard E (1978) Anti-
epileptic drug development: II. Anticonvulsant drug screening.
Epilepsia 19:409–428
Kulandasamy R, Adhikari AV, Stables JP (2009) A new class of
anticonvulsants possessing 6Hz activity: 3, 4-Dialkyloxy
thiophene bishydrazones. Eur J Med Chem 44:4376–4384
Kumar P, Shrivastava B, Pandeya SN, Tripathi L, Stables JP (2012)
Design, synthesis, and anticonvulsant evaluation of some novel 1,
3 benzothiazol-2-yl hydrazones/acetohydrazones. Med Chem Res
21:2428–2442
Landmark CJ (2006) Targets for antiepileptic drugs in the synapse.
Med Sci Monitor 13:RA1–RA7
Lipinski CA (2004) Lead-and drug-like compounds: the rule-of-five
revolution. Drug Discovery Today: Technol 1:337–341
Löscher W (2016) Fit for purpose application of currently existing
animal models in the discovery of novel epilepsy therapies.
Epilepsy Res 126:157–184
Azizmohammadi M, Khoobi M, Ramazani A, Emami S, Zarrin A,
Firuzi O, Miri R, Shafiee A (2013) 2H-chromene derivatives
bearing thiazolidine-2, 4-dione, rhodanine or hydantoin moieties
as potential anticancer agents. Eur J Med Chem 59:15–22
Banks WA (2009) Characteristics of compounds that cross the blood-
brain barrier. BMC Neurol 9:S3
Macdonald R, McLean M (1985) Anticonvulsant drugs: mechanisms
of action. Adv Neurol 44:713–736
Medina FG, Marrero JG, Macías-Alonso M, González MC, Córdova-
Guerrero I, García AGT, Osegueda-Robles S (2015) Coumarin
heterocyclic derivatives: chemical synthesis and biological
activity. Nat Prod Rep 32:1472–1507
Barton ME, Klein BD, Wolf HH, White HS (2001) Pharmacological
characterization of the 6 Hz psychomotor seizure model of partial
epilepsy. Epilepsy Res 47:217–227
Bhat MA, Al-Omar MA (2011) Coumarin incorporated triazoles: a
new class of anticonvulsants. Acta Pol Pharm 68:889–895
Meldrum BS, Rogawski MA (2007) Molecular targets for antiepileptic
drug development. Neurotherapeutics 4:18–61
Miller PS, Aricescu AR (2014) Crystal structure of a human GABAA
receptor. Nature 512:270
Mohanraj R, Brodie MJ (2005) Outcomes in newly diagnosed
localization-related epilepsies. Seizure 14:318–323
Blair LM, Sperry
J (2013) Natural products containing a
nitrogen–nitrogen bond. J Nat Prod 76:794–812
Ngugi AK, Bottomley C, Kleinschmidt I, Sander JW, Newton CR
(2010) Estimation of the burden of active and life‐time epilepsy: a
meta‐analytic approach. Epilepsia 51:883–890
Okada R, Negishi N, Nagaya H (1989) The role of the nigrotegmental
GABAergic pathway in the propagation of pentylenetetrazol-
induced seizures. Brain Res 480:383–387
Pandeya SN, Yogeeswari P, Stables JP (2000) Synthesis and antic-
onvulsant activity of 4-bromophenyl substituted aryl semi-
carbazones. Eur J Med Chem 35:879–886
Cakir B, Dag O, Yildirim E, Erol K, Sahin M (2001) Synthesis and
anticonvulsant activity of some hydrazones of 2 [(3H)-oxo-
benzoxazolin-3-yl-aceto] hydrazide. Eczacilik Fakultesi Dergisi-
Gazi Universitesi 18:99–106
Clark DE (1999) Rapid calculation of polar molecular surface area and
its application to the prediction of transport phenomena. 2. Pre-
diction of blood–brain barrier penetration.
88:815–821
J Pharm Sci
Costa M, Dias TA, Brito A, Proença F (2016) Biological importance
of structurally diversified chromenes. Eur J Med Chem 123:
487–507
Picaud S, Sahel JA, Firas J, Coriat C, Simonutti M (2016) Antic-
onvulsive pharmaceutical compositions. Google Patents.
Pisani L, Farina R, Catto M, Iacobazzi RM, Nicolotti O, Cellamare S,
Mangiatordi GF, Denora N, Soto-Otero R, Siragusa L (2016)
Exploring basic tail modifications of coumarin-based dual
acetylcholinesterase-monoamine Oxidase B inhibitors: identifi-
cation of water-soluble, brain-permeant neuroprotective multi-
target agents. J Med Chem 59:6791–6806
de Souza LG, Rennó MN, Figueroa-Villar JD (2016) Coumarins
as cholinesterase inhibitors:
a review. Chem Biol Interact
254:11–23
Dimmock J, Pandeya S, Quail J, Pugazhenthi U, Allen T, Kao G,
Balzarini J, DeClercq E (1995) Evaluation of the semicarbazones,
thiosemicarbazones and bis-carbohydrazones of some aryl ali-
cycylic ketones for anticonvulsant and other biological propertie.
Eur J Med Chem 30:303–314
Potschka H (2012) Animal models of drug-resistant epilepsy. Epileptic
Disord 14:226–234
Ragavendran JV, Sriram D, Patel SK, Reddy IV, Bharathwajan N,
Dimmock JR, Vashishtha SC, Stables JP (2000) Anticonvulsant
properties of various acetylhydrazones, oxamoylhydrazones and
semicarbazones derived from aromatic and unsaturated carbonyl
compounds. Eur J Med Chem 35:241–248
Stables J, Yogeeswari
anticonvulsants from
anilide and hydrazone pharmacophore. Eur
P
a
(2007) Design and synthesis of
combined phthalimide–GABA–
J Med Chem
42:146–151
Fisher RS, van E. Boas W, Blume W, Elger C, Genton P, Lee P et al.
(2005) Epileptic seizures and epilepsy: definitions proposed by
the international league against epilepsy (ILAE) and the inter-
national bureau for epilepsy (IBE). Epilepsia 46:470–472
Heber D, Ivanov I, Karagiosov S (1995) The vilsmeier reaction in the
synthesis of 3‐substituted [1] benzopyrano [4, 3‐b] pyridin‐5‐
ones. An unusual pyridine ring closure. J Heterocycl Chem
32:505–509
Jain J, Kumar Y, Sinha R, Kumar R, Stables J (2011) Menthone aryl acid
hydrazones: a new class of anticonvulsants. Med Chem 7:56–61
Jameel E, Umar T, Kumar J, Hoda N (2016) Coumarin: a privileged
scaffold for the design and development of antineurodegenerative
agents. Chem Biol Drug Des 87:21–38
Raj V, Kumar P, Kumar A (2015) Design and synthesis of 4-(4-ethyl-
phenyl)-4-oxo-butyric acid (3-oxo-1, 3-diphenyl-propylidene)/[1-
(4-chloro-phenyl)-ethylidene]-hydrazide as potential antic-
onvulsant agents. Med Chem Res 24:603–610
Robertson DW, Lawson RR, Rathbun RC, Leander JD (1988) Phar-
macology of LY201409, a potent benzamide anticonvulsant.
Epilepsia 29:760–769
Sandhu S, Bansal Y, Silakari O, Bansal G (2014) Coumarin hybrids as
novel therapeutic agents. Bioorg Med Chem 22:3806–3814
Singhuber J, Baburin I, Ecker GF, Kopp B, Hering S (2011) Insights
into structure–activity relationship of GABA A receptor mod-
ulating coumarins and furanocoumarins. Eur
J Pharmacol
668:57–64

Med Chem Res
Sinha R, Sara US, Khosa R, Stables J, Jain J (2011) Nicotinic acid
hydrazones: a novel anticonvulsant pharmacophore. Med Chem
Res 20:1499–1504
Ulloora S, Shabaraya R, Ranganathan R, Adhikari AV (2013)
Synthesis, anticonvulsant and anti-inflammatory studies of new 1,
4-dihydropyridin-4-yl-phenoxyacetohydrazones. Eur
J
Med
Skalicka-Woźniak K, Orhan IE, Cordell GA, Nabavi SM, Budzyńska
B (2016) Implication of coumarins towards central nervous sys-
tem disorders. Pharmacol Res 103:188–203
Sridhar SK, Pandeya SN, Stables JP, Ramesh A (2002) Anticonvulsant
activity of hydrazones, Schiff and Mannich bases of isatin deri-
vatives. Euro J Pharm Sci 16:129–132
Chem 70:341–349
van de Waterbeemd H, Camenisch G, Folkers G, Chretien JR,
Raevsky OA (1998) Estimation of blood-brain barrier crossing of
drugs using molecular size and shape, and H-bonding descriptors.
J Drug Target 6:151–165
Verma G, Marella A, Shaquiquzzaman M, Akhtar M, Ali MR, Alam
MM (2014) A review exploring biological activities of hydra-
zones. J Pharm Bioallied Sci 6:69
Sucher NJ, Carles MC (2015) A pharmacological basis of herbal
medicines for epilepsy. Epilepsy Behav 52:308–318
Terrone G, Pauletti A, Pascente R, Vezzani A (2016) Preventing
epileptogenesis: a realistic goal? Pharma Res 110:96–100
Toman J, Everett G, Richards R (1951) The search for new drugs
against epilepsy. Tex Rep Biol Med 10:96–104
Vilar S, Chakrabarti M, Costanzi S (2010) Prediction of passive
blood–brain partitioning: straightforward and effective classifi-
cation models based on in silico derived physicochemical
descriptors. J Mol Graph Model 28:899–903
Tripathi L, Singh R, Stables JP (2011) Design & synthesis of N
′-[substituted] pyridine-4-carbohydrazides as potential antic-
onvulsant agents. Eur J Med Chem 46:509–518
Zhu HL, Wan JB, Wang YT, Li BC, Xiang C, He J, Li P (2014)
Medicinal compounds with antiepileptic/anticonvulsant activities.
Epilepsia 55:3–16