Characterization of the anticonvulsant profile of valpromide derivatives

Article abstract of DOI:10.1016/j.bmc.2004.05.003

The antiepileptic activity of nine derivatives of valpromide is discussed. They comply with a pharmacophore model that establishes the essential structural and electronic features responsible for the protection against the MES test. The model results from

Full text of DOI:10.1016/j.bmc.2004.05.003

Bioorganic & Medicinal Chemistry 12 (2004) 3857–3869
Characterization of the anticonvulsant profile of valpromide
derivatives
a
Silvina M. Tasso, Sung Ch. Moon, Luis E. Bruno-Blanch and Guillermina L. Estiu
a
a
ꢀb,*
aMedicinal Chemistry, Department of Biological Sciences, Facultad de Ciencias Exactas,
Universidad Nacional de La Plata, B1900AVV La Plata, Argentina
bDepartment of Chemistry, The Pennsylvania State University, 152 Davey Laboratory, University Park, PA 16801-6300, USA
Received 23December 2003; revised 2 May 2004; accepted 4 May 2004
Available online 7 June 2004
Abstract—The antiepileptic activity of nine derivatives of valpromide is discussed. They comply with a pharmacophore model that
establishes the essential structural and electronic features responsible for the protection against the MES test. The model results
from the comparison of 17 structures, using density functional methodologies combined with an active analog approach. The
derivatives of valpromide have been tested for anticonvulsant activity in mice. These compounds displayed a phenytoin-like profile,
being active in the MES test and inactive in the PTZ test. 4-(Valproylamido)benzenesulfonamide is the most active compound, with
an ED50 of 53 lmol/kg and no neurotoxicity at doses up to 1000 lmol/kg. The pharmacological behavior of the drugs points to a
sodium channel blocking effect as one of the associated mechanisms. This mechanism was tested positive for N-ethylvalpromide
through its competition with the binding of [3H]batrachotoxin-A-20a-benzoate to the voltage-dependent sodium channels from rat
brain synaptosomes.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
channels, modulation of GABA synthesis, or degrada-
tion, inhibition of cellular GABA uptake, modulation of
GABAA receptors, modulation of various excitatory
amino acid receptors, and modulation of adenosine
metabolism.3;4;9 For this reason, several seizure models
have been developed to evaluate the anticonvulsant
(AC) activity during the research process. Among them,
the maximal electroshock (MES) and the subcutaneous
pentylenetetrazol (PTZ) tests are the most widely used.2
These models respond well to neuronal voltage-gated
sodium channel (VGSC) inhibitors and gabaergic com-
pounds, respectively.5;7
Among the different neurological disorders that affect
human condition, epilepsy has been largely studied
during the last century,1–4 becoming a dynamic research
field in recent years.4–13 It describes disorders charac-
terized by recurrent seizure attacks due to synchronous
neuronal firing. The main drawback related to its
therapy is associated with the fact that antiepileptic
drugs (AEDs) fail to control seizures in 20–25% of
patients.9–11;14;15 Moreover, even the new generation of
AEDs causes sizeable side effects, which include ataxia,
diplopia, mental dulling, rash, blood dyscrasias, and
hepatotoxicity.7 Thus, new AEDs with better safety, less
toxicity, and higher efficacy in difficult-to-control
patients are urgently needed.8
A large number of ligands that are presently marketed
(or in the process of being marketed) as AEDs exhibit a
similar pharmacological profile, being quite potent in
the MES test and inactive in the PTZ test. This is the
case of the widely used phenytoin (PHE), as well as
carbamazepine (CZ), topiramate (TOP), lamotrigine
(LAM), zonisamide (ZON), ralitoline (RAL), and ox-
carbazepine (OCZ).4;5;9;15–17 The drugs sharing the PHE-
like profile of activity are accepted today to prevent
seizures because they block Naþ channels in the
brain.7;15 Naþ channel blockers are front-runners in the
treatment of epilepsy. This mechanism is not, however,
unique for the above mentioned drugs. As an example,
The broad spectra of mechanisms, which may occur
during drug action in the epileptic patient, include,
among others, voltage-dependent blockade of Naþ
Keywords: Anticonvulsant activity; Pharmacophore; Valpromide
derivatives; Antiepileptic drugs.
* Corresponding author. Tel.: +1-814-8637591; fax: +1-814-8619300;
e-mail: gle10@psu.edu
0968-0896/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.05.003

3858
S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
the variety of molecular actions displayed by valproic
acid (VPA) includes Naþ channel blockade, increase of
GABA levels in the brain by glutamate decarboxylase
activation and GABA transaminase inhibition, increase
of postsynaptic GABA responses, direct membrane
effects on the neurons, and blockade of Caþþ chan-
nels.9;15;16;18
In spite of the large amount of attempts of postulating a
general pharmacophore for the antiepileptic activity,
only one has been reported, to our knowledge, for the
Naþ channel blocking activity of dissimilar antiepileptic
compounds.19 The model reported is based on the
structural comparison of CZ, PHE, LAM, ZON, and
rufinamide (RUF), and identifies three groups of dif-
ferent characteristics that should be present, at well-
defined distances, in order for the structures to be active.
These different groups involve an aryl ring, an electron
donor atom and a second donor atom which, close to an
NH group, defines a hydrogen bond acceptor/donor
unit. On the other hand, several model pharmacophores
have been developed from the comparison of complexes
that belong to a given family. Some of them bear some
similarities in relation to the position of H-bond donor/
acceptor and lipophilic groups. In this way, from the
comparative analysis of compounds structurally related
to N-benzylamides of c-hydroxybutyric acid (GMB),20
the requirements for activity have been associated with
the presence of an N-benzylamide fragment, an aryl
ring, and a H-bond donor group located between them
(see Fig. 1).
Figure 2. (a) Schematic representation of the ureylenes and suggested
pharmacophore models for the anticonvulsant activity.21 (b) Schematic
representation and proposed pharmacophore model for the anticon-
vulsant activity of the semicarbazones.21 (1) Hydrophobic binding
areas, (2) hydrogen bonding site, (3) electron donor group.
On the basis of this pattern, we have considered a set of
AEDs, structurally related to VPA. They bear a high
degree of flexibility and we speculate this might help to
accommodate the receptor pocket more easily. Showing
a PHE-like profile, they manifest higher anti-MES po-
tency than the parent compound. As a further test of the
relation between the PHE-like activity and the Naþ
channel blocking capability, in vitro tests have been
performed for a representative member of the newly
designed set, which meets the requirements of potency
and solubility.
For the case of ureylenes and semicarbazones, a model
pharmacophore has been proposed, which points to the
existence of an H-bonding site located between two
hydrophobic areas (Fig. 2). The H-bonding site has been
further disclosed in H-bonding and electron donor
groups.21
We present, in the first part of this article, a validation of
the previous pharmacophore, using a training set that
extends the previous one, including AEDs developed in
previous years. Some of them derive from the already
marketed PHE and LAM,22;24;25 but also structures as
diverse as c-butyrolactams, semicarbazides, and 3-
aminopyrrole are included.19;26–30 The second part of the
article deals with the synthesis and biological evaluation
of the VPD-related AEDs, thoroughly discussing their
pharmacological profile.
Computational analysis has also contributed in other
cases to design novel antiepileptic ligands, but the
studies were limited to compounds of a given family.22
We have recently analyzed several AEDs, which are
recognized to bind to Naþ channels.23 On the basis of
this analysis, we have proposed a pharmacophore that,
showing some similarities with the one previously
described,19 mainly differs in the necessity of an electron
donor group. According to our analysis, based on a
larger set of structures, the presence of an acceptor/
donor unit comprising two closely located H-donor–
acceptor groups, together with a lipophilic portion,
defines enough conditions to attain activity.
2. Results and discussion
2.1. Pharmacophore identification
The activity data for the compounds that define the
training set (Fig. 3, Table 1) quantify their response to
Figure 1. (a) Schematic representation of the compounds related to N-benzylamides of c-hydroxybutyric acid derivatives (GMB) considered in Ref.
20. (b) Pharmacophore model proposed for the anticonvulsant activity of substituted GMB.

S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
3859
1
8
NH2
N
O
2
H
N
11N8
3
3
3
2
9
3
1
H2N
N
O
O1
O
NH2
5
2
10
4
3
4
O
N
2
4
1
H2N
2
N
HO
O
6
9 H
5
4
5
1
6
5
O
O
Cl
Cl
6
7
7
6
H2N
4
7
6
7
5
7
CZ
PHE
VPA
LAM
FLB
NH2
NH2
NH2
3
1
9
O
4
S2
O
O
9
1
3
1
8
O
3
O 2S O
H
3
2
4
Cl
2
5
6
N
N
HN4
7
1
5
O
6
2
O
O
1
O
O
NH
5
6
8
O
3
6
5
4
4
O
S
7
5
7
O
7
6
O
N
O
RMC
ETH
TOP
ZON
RAL
1
3
2
9
O
H2N
8
3
4
5
1 O
NH2
3
O 1
2
2
O
4
N
N
6
4 N
5
N
5
N
10
7
6
6
8
9
F
F
7
7
N
O
RUF
OCZ
VIN
Figure 3. Schematic representation of the compounds that define the training set.
Table 1. Biological data of the compounds that constitute the training set and those new ones selected for their Naþ channel activity
Compound
Log P
ED50 (lmol/kg)
MES test (mice)
Ref.
ED50 (lmol/kg)
PTZ test (mice)
Ref.
Naþ channel binding Ref.
CZ
2.72
2.26
2.6
37
38
32
32
32
1
NE
NE
1033
621
NE
NE
NE
NE
NE
921
1,32
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
NA
Yes
Yes
Yes
Yes
Yes
Yes
Yes
5,26,31,33,34
5,7,27,31,34,35
27,31
PHE
VPA
FLB
32
32
1
1886
210
7
0.78
2.66
2.66
5.16
0.64
1.85
0.55
2.42
0.65
1.66
6.86
2.99
1.03
3,9
5,10,26,34
LAM
RMC
TOP
ZON
RAL
ETH
VIN
1,5
1,4
1,5
1
1
292
112
92
1
4
5
1
5
9
1,31,33,34,37
1
32
5
>7000
77
32
38
31
5
RUF
OCZ
SSDG
3BEP
AWD
21–71
79
3,9
1
5
NE
193
89
1
27
27
33
36
29
30
10,27
341
7
28,29
30
10,27
PNU-151774E 2.45
26
The main references that report their activities are given. The partition coefficients (log P) of the compounds were calculated using the program
HyperChem QSAR Properties, version 7.01 Hypercube, Inc.39
the MES and PTZ tests. The majority of them show a
PHE-like profile, being active against the MES test
and inactive against the PTZ test. Some of them
(felbamate––FLB, remacemide––RMC, ethosuximide––
ETH, vinpocetina––VIN) are also active in the PTZ
test. They have been selected on the basis of their
common capability of blocking the VGSC, as either
established or newer drugs that have overcome the
development stage.
The majority of the active structures bear the carboxylic
acid or amide (sulfonamide) function that may be either
substituted or not. ETH has been included as an inactive
analog, because it is one of the major antiepileptics used

3860
S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
the active conformation (rigid analog). Whereas PHE
determines the orientation of the polar moiety, that is
present in all the structures, CZ does the same for the
lipophilic portion. They comply, on the other hand, with
the requirements of high antiepileptic potency and Naþ
channel binding efficiency.
From the structure shown in Figure 5, the constrains
have been defined as the nonbonded distances in the
PHE–CZ overlapping structure, considering only three
consecutive atoms of the aryl ring on the basis of the
activity and the chain length of VPA.
Three main steps have been followed in the computa-
tions:
(i) The initial optimization of each structure by means
of DFT calculations. It allows the identification of
the most stable conformer, associated with the
absolute minima in the potential energy hypersur-
face. The optimized conformers represent the most
probable structures when far enough from the
receptor. It is assumed that the conformational
change in the event of the interaction only modifies
the torsion angles, keeping the bond lengths and
planar angles equal to the DFT optimized values.
(ii) The alignment of the resultant structures with the
template, defined by CZ and PHE. This allows
one to identify the conformation capable of inter-
acting with the active site (bioactive conformation).
The aligned structures are shown in Figure 4.
(iii) The refinement of the geometry of the bioactive
conformers by means of constrained optimizations,
where the dihedral angles defined by overlapping
atoms were kept frozen to the pharmacophore val-
ues. The resulting conformations, associated with
relative minima in the potential energy hypersur-
face, have been used to evaluate the electronic char-
acteristics of the ligands.
Figure 4. Superposition of the molecules that define the training set.
Color pattern as follows: nitrogen atoms blue, oxygen atoms red,
sulfur atoms yellow, carbon atoms light blue. H atoms are not shown
for clarity. The dissimilarity of the structures becomes evident. Phar-
macophoric points (green atoms), defined as those that are common to
all the overlapping structures.
in absence seizures.31 It does not show activity in the
sodium channels40;41 and also has a different pharmaco-
logical behavior, being inactive against the MES test.42
The structures of the training set have been aligned (Fig.
4) after a search of the conformational space through
torsional modifications of the geometries initially opti-
mized at a density functional (DFT) level. The align-
ment procedure followed an active analog approach
43
(
steps.
SYBYL
)
where the torsion angles were varied in 5°
CZ and PHE were used as starting structures. Because
of their rigidity and activity, the consideration of their
overlapping portions (Fig. 5) becomes enough to define
Density functional calculations have been used for the
first and third steps, maintaining the same level of the-
ory. The last step is oriented to find a stable structure,
close to the active one, optimized at a level of theory
that allows the evaluation of the electronic properties. In
the active analog approach, used for the second step,
root mean square (rms) values were lower than 0.5 but
for remacemide (RMC). For the latter, the superposi-
tion of the polar moiety involves an additional atom
linking the positive and negative ends (Fig. 3), rising the
rms to 0.7. The conformational flexibility of some
ligands, as VPA and FLB, increases the likelihood of
finding numerous conformations that match the dis-
tance requirements. To reduce the size of the confor-
mational space to be explored, rotable bonds have been
only defined for the portion that should adapt to the
pharmacophore requirements, allowing the rest of the
molecule to maintain the DFT optimized geometry.
In the superimposed structures, the overlapping
bonds are highlighted in green (Fig. 4). This common
pattern allows us to propose certain requirements for
the structures to manifest AC activity. The requirements
Figure 5. Superposition of CZ and PHE. The overlapping atoms are
highlighted in green and define the rigid analog. Atom colors as in
Figure 2.

S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
3861
Table 2. Local density charges (Qi) on the atomic centers of the polar
moiety derived from a fitting to calculated electrostatic potentials
within B3LYP/6-31þG** calculations
The overlapping region comprises strictly the green
portion (Fig. 4) when all the molecules of the training set
are considered. However, Figure 4 shows a lipophilic
portion larger than this, which extends a region in the
space pointing opposite to the polar moiety. Although
the atoms of different ligands do not comply with a
regular pattern in this region, it renders information on
the possible size of the lipophilic moiety. One of the
negative ends of the polar end, free from steric hin-
drance, is more easily accessible than the other for
receptor interaction.
Q1
Q2
Q3
CZ
À0.634
À0.524
À0.627
À0.646
À0.734
À0.568
À0.535
À0.589
À0.586
À0.558
À0.551
À0.620
À0.606
1.065
0.650
0.868
1.032
0.557
0.521
1.433
1.419
0.838
0.595
0.797
0.746
1.028
À1.048
À0.669
À0.712
À0.971
À0.768
À1.011
À0.551
À0.608
À0.690
À0.615
À0.473
À0.926
À1.136
PHE
VPA
FLB
LAM
RMC
TOP
ZON
RAL
ETH
VIN
RUF
OCZ
The previous discussion allows one to postulate the
position of complementary sites in the receptor, associ-
ated with H-donor–acceptor groups and, at least, one
lipophilic pocket, at well defined relative orientations.
This orientation is imposed by the bioactive conforma-
tion of the ligands, shown in Figure 5. Pair wise dis-
tances between the overlapping atoms (see Fig. 6) helps
in the definition of the shape of the proposed pharma-
cophore.
The index i denotes the atom number in Figure 3.
can be summarized in: (1) a hydrophobic chain (atoms
5–7, Fig. 3), (2) a polar moiety (atoms 1–3, Fig. 3).
The polar nature of the second portion is shown by the
local density charges depicted in Table 2, which have
been derived from a fitting to calculated electrostatic
potentials at the DFT level. The electronic description
corresponds to the active conformation of the deriva-
tives.
The lipophilic pocket should be wide enough to allow
the interaction of ligands as vinpocetine (VIN).
Neuronal VGSCs have emerged as useful targets for
anticonvulsant therapy in the last decade, and a large
number of families of promising ligands have been
designed, which are still under development. It is inter-
esting to analyze how these newly designed ligands fit
the requirements of the proposed pharmacophore. For
this purpose, we have chosen four ligands from different
families, which are shown in Figure 7.
In all the cases included in the present study, the nega-
tive end is defined by N or O atoms, whereas C or S
atoms are comprised in the positive one (only in the case
of RMC the negative center in position 3is placed over a
carbon atom). The electronic distribution is always
compatible with the polar characteristics of this moiety.
The charges on atoms 5, 6, and 7, on the other hand,
remain close to zero (data not shown), defining the
hydrophobic region. For this region, a minimum lipo-
philicity seems to be required, as it is demonstrated by
ETH which, bearing the polar portion is not active.
Although the lipophilic moiety is aromatic in a large
number of compounds of the training set, this condition
does not define a requirement for the activity, as it is
mainly demonstrated by VPA. The hydrophobic
requirements cannot be easily quantified as the elec-
tronic ones. Whereas some compounds show two
hydrophilic regions, like RMC, PHE, and even VPA,
only one hydrophobic tail is present in ligands as FLB,
RAL, ZON. The calculated log P values (see Table 1)
seem to indicate that, whenever a minimum lipophilicity
is attained, the other requirements dictate the AE
potency. Nevertheless, this minimum value cannot be
easily quantified. ETH can be considered to somehow
define the limiting region. The AC activity of VPA and
related acids has been largely studied.44;45 Whereas no
correlation with calculated log P values has been found
when different hydrocarbon chains were considered, a
requirement for activity has been associated with the
existence of ramifications.45 Nevertheless, the activity of
carboxylic acids related to VPA is originated in several
different mechanisms,46 and no simple interpretation of
the different anticonvulsant potencies has been yet pro-
vided.
3BEP27 belongs to a series of 2-piperidinones, synthe-
sized as noncompetitive GABAA antagonists, and was
recently found to modulate sodium channels in a use-
dependent manner.28 PNU-151774E is a propanamide
derivative developed from milacemide,26;27 whereas
AWD-140-190 is a 3-aminopyrrole derivative.29 SSDG
is the more active member of a family of conforma-
tionally constrained analogs of N,N-diaryl-guanidines,
locked into the SSþ conformation.43 All of them share
the Naþ channel blocking activity as one of their related
mechanisms of action. The superposition of the active
conformation of each of these structures with the CZ–
PHE template is shown in Figure 8. The local density
Figure 6. Distance map for the pharmacophore proposed for the MES
antiepileptic activity. Hydrogen atoms not shown. Color pattern as
follows: nitrogen atoms blue, oxygen atoms red, sulfur atoms yellow,
carbon atoms light blue.

3862
S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
7
requirements, usually obtained from modifications of a
7
1
6
+
member of the training set. VPA is an ideal test com-
pound for this objective, as a large number of modifi-
cations can be designed, which can lead to either
structural or functional derivatives.18;48–50 We have
simultaneously considered the AE activity and the side
effects related to VPA therapy: sedation, hepatotoxicity,
teratogenicity, and microsomal epoxide hydrolase
(mEH) inhibition.47 The propyl chains have been re-
tained on the consideration that VPA possess the opti-
mal chemical structure with regard to the margin
between its anticonvulsant activity and its sedative and
hypnotic side effects.51–53 Functionalization to the
amide, valpromide (VPD), appears as promising, as a
higher anticonvulsant potency and lower teratogenicity
are attained in mice after this transformation.54–56
However, VPD is actually a prodrug of VPA when
administered in human.56 This condition is not neces-
sarily retained after structural modifications on the
amide moiety, that will certainly modify the kinetics of
biotransformation to VPA. This hypothesis is sustained
by the results of pharmacokinetic studies of various N-
mono and N,N-disubstituted VPD derivatives,50;57;58
which has demonstrated that they act as intact entities,
following a very low hydrolysis to VPA. For Valroce-
mide (N-valproylglycinamide-TV1901), a N-substituted
VPD that has reached phase IIb of clinical trails,4 it is
well-known that only a 4% fraction of the drug is
metabolized to valproic acid in humans.4 Another
advantage of N-substituted derivatives of VPD, over
VPA and VPD, is that they probably do not inhibit the
human mEH, an enzyme responsible for detoxifying
reactive epoxide intermediates formed by oxidative
metabolism of xenobiotics.59
6
5
5
12 NH2
O
2
HN 3
4
N 3
N4
3BEP
SSDG
O
1 O
3
NH2
2
4
HN5
6
7
Cl
7
F
N
5
6
3
O
4
2
O
N
H
O
1
AWD
PNU
Figure 7. Schematic representation of the structure of some newly
designed anticonvulsant compounds associated with a Naþ channel
blocking mechanism of action.
The advantages associated with N-substitution of val-
promide can be considered together with the high anti-
epileptic potency of RAL and LAM, which are
N-substituted in the polar end. To this end we have
analyzed nine N-substituted VPD derivatives in order to
validate the proposed pharmacophore and to evaluate
the importance of a lipophilic substituent in the polar
moiety. In this way, we are simultaneously evaluating
the agreement with the pharmacophore previously
proposed by Stables and co-workers20 (Fig. 1), which
locates the H-bond donor/acceptor group between two
lipophilic regions. The structures (Fig. 9) comply with
the pharmacophore previously identified.
Figure 8. Superposition of SSDG, 3BEP, AWD, and PNU with the
pharmacophore template (in green), defined by CZ and PHE. For the
case of PNU, two different conformations (C1, C2) are considered.
The superposition of the valpramides with the
pharmacophore template is shown in Figure 10. The
capability of interaction with the Naþ channels has
been tested for ETVPD, which, being more potent than
VPD, has the same pharmacological profile, and
matches the solubility requirements for the in vitro
assay.
charges on the polar moieties are shown in Table 2 for 3-
BEP. For the case of PNU-151774E, its interaction with
the lipophilic pocket may imply a conformational
change (Fig. 8).
2.2. Design and synthesis of valpromide derivatives
2.2.1. Chemistry. The general procedure implies the
reaction of the amine with valproyl chloride in anhy-
drous conditions. Particular details for each reaction are
given in Section 4.
The step that follows the identification of a pharmaco-
phore is its further validation by means of the consid-
eration of other ligands, which comply with its

S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
3863
3
NH
3
HN
1
1
3
2
2
1
NH
2
4
O
H2NO2S
O
O
4
4
6
6
7
7
6
5
5
7
5
SUVPD-1
CHVPD-2
IPVPD-3
3
N
3
HN
1
O
3
1
2
2
2
HO2C
HN
O
O
4
4
6
4
6
6
7
7
7
5
5
5
DEVPD-6
BUVPD-4
CPVPD-5
3
1
3
HN
3
1
2
2
2
HN
N
O
O
5
O
4
4
6
6
7
4
6
7
3
7
5
5
ETVPD-8
DMVPD-9
BZVPD-7
3
1
1
2
4
2
4
HO
H2N
O
O
6
7
6
7
5
5
VPA
VPD
Figure 9. Schematic representation of the structures of the valpromide derivatives.
With exception of CPVPD-5, the VPD’s derivatives
show protection against the MES test, but not against
the PTZ test (Table 3). The MES test is a valuable tool
to identify drugs that are active against generalized tonic
clonic and focal seizures.2;67 The PHE-like profile,
characteristic of the majority of the VPD’s, indicates an
effectiveness against generalized seizures and is consid-
ered as indicative of a Naþ channel blocking activity.
In some cases (SUVPD-1, CHVPD-2, ETVPD-8,
DMVPD-9), there is an important increase in the AE
potency, and in the protective index (PI), relative to
VPA (Table 3). SUVPD-1 is the most promising deriv-
ative, being 19 times more potent than VPA and almost
seven times more than VPD, when considering the anti-
MES activity. It is also remarkably safer (more than 11
times) than the parent drugs VPA and VPD.
Intraperitoneal testing becomes difficult in several cases,
due to the low solubility of the drugs. This is the case of
DEVPD-6 and BZVPD-7, for which protection per-
centage is reported.
Figure 10. Superposition of the valpromide derivatives with the
pharmacophore template (in green), defined by CZ and PHE.
SUVPD-1 opens a promising family of substituted val-
promides, showing an ED50 value close to that of PHE
or CZ. We have previously analyzed the AC of sulfon-
amides, including acetazolamide (1) and two analogs
(compounds 2 and 3, Fig. 11), testing also their ability to
inhibit carbonic anhydrase (CA).68 The compounds
exerted a PHE-like profile, being active in the MES test
and inactive against PTZ-induced convulsions. At that
time, it was noted that bulky substituents in position 5
(Fig. 11) increases the anti-MES potency.
2.2.2. Pharmacology. The N-substituted valpromides
were tested for anticonvulsant activity by their ability to
suppress experimentally induced convulsions in labora-
tory animals. The tests used were the Maximal Elec-
troShock Seizure test (MES test), related to electrical
induction, and the pentylenetetrazol test (PTZ test),
based on chemical induction of the seizure episode. The
RotoRod test was used to determine the possible neu-
rotoxic effects. The results obtained are listed in Table 3.

3864
S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
Table 3. Biological data determined (mice) for the valpromide derivatives designed in this research
Compound
TPE (h)
ED50 (lmol/kg)
MES test
ED50 (lmol/kg)
PTZ test
TPEN (h)
TD50 (lmol/kg)
RotoRod test
PIa
Pot.b
Cat.c
SUVPD-1
CHVPD-2
IPVPD-3
BUVPD-4
CPVPD-5
DEVPD-6
BZVPD-7
ETVPD-8
DMVPD-9
VPA
4
53 (33–87)
61 (44–85)
0% at 1000
0% at 65
0% at 250
––
––
0% at 1000
0% at 65
0% at 450
>19
19
17
1
2
>1.1
1
1
384 (300–492)
487 (412–576)
1225 (1107–1355)
50% at 300e
––
>1.2
>1.2
>1.2
>1.0
––
2.6
2.1
0.8
3.4
––
1
1
0% at 600
1817 (1669–1978)
0% at 375e
0% at 200e
0% at 800
0.50
––
40% at 600
0% at 1500
0% at 300e
0% at 200e
632 (550–727)
0% at 700
1
2
2
1
––
––
––
––
1
2
0% at 200e
0.5
0.5
0.25
1
200 (148–272)
347 (266–452)
1008 (792–1282)
353 (267–466)
0.5
––
3.2
>2.0
1.1–1.7d
1.3–1.4d
5.0
2.9
1
0% at 1000
1261 (1155–1377)
490 (447–537)
1
0.25
2
17% at 1700
25% at 600
2
VPD
2.9
1
TPE: time of peak effect, TPEN: Neurotoxic time of peak effect.
a PI (protective index) ¼ TD50/ED50, calculated for the MES test.
b Pot. (Relative Potency) ¼ ED50VPA/ED50drug, calculated for the MES test.
c Category according to the classification of Phase 1 of the Anticonvulsant Screening Project of the ADD program.
d According to bibliography.52;60–66
e Not soluble at higher concentration.
1
1
O
O
5
5
7
2
7
S 4
6
2
4
S
NH2
S
H2N
6
S
NH2
CH3COHN
O
N
N
3
O
N
N
3
Acetazolamine - AZ 1
Hats 2
1
1
O
O
5
5
2
7
7
2
t
S 4
N
6
6
S
4
S
NH2
S
H N
Bu OCOHN
NH2
n
O
O
N
O3
N
N
3
VPD-AZ - 4
B-H2ats- 3
Figure 11. Acetazolamide (1) and derivatives previously analyzed (2, 3, Ref. 68); (4), Ref. 67.
That research led to the design of SUVPD-1, whose
anticonvulsant activity has already been discussed by
us.69 The same drug has been considered two years later
by Masereel et al.67 These authors have combined val-
proic acid derivatives and sulfonamide functionalities,
thoroughly analyzing the resulting compounds in their
inhibitory activity of CA in the brain, and further
relating it to the AE activity.68 For the drugs showing a
higher performance in the in vitro CA inhibition, among
which SUVPD-1 was not included, the in vivo anti-MES
AE activity has been determined. In this way, they have
succeeded in synthesizing other sulfonamide-related
AEDs, which showed protection comparable to that of
TOP67 and SUVPD-169 (Table 3). Among them, 5-
valproylamido-1,3,4-thiadiazole-2-sulfonamide (VPD-
AZ 4, Fig. 11) demonstrated to be highly effective in
protecting mice from MES over a long period of time
(3h) We have also found protection over a long period
of time for SUVPD-1. It is a remarkable that this drug
develops its maximum activity 4 h after its administra-
tion, in contrast to VPD, whose activity is higher after
an hour. At no time was protection against PTZ noted.
This fact cancels out the possible hydrolysis of SUVPD-
1 to either VPD or valproic acid, as both of them shows
activity in both animal models. The outstanding pro-
tection of SUVPD-1 against MES convulsions may be
not only restricted to a Naþ channel blocking mecha-
nism. Its functionality resembles TOP, whose multiple
mechanisms of action include the inhibition of the
generation of action potentials in neurons antagonizing
the activation of Naþ channels, the positive modulation
of some types of GABAA receptors, and the antagonism
of kainate/AMPA receptors.9;16;70 The capability to
interact with CA67 is not enough (Ki ¼ 22; 500 nM vs
250 nM for TOP), to justify the remarkable activity of
SUVPD-1. Thus, in addition to the testing of CA inhi-
bition, it would be certainly interesting to monitor other
possible mechanisms in the development of this class
of compounds over the next few years.
The pharmacological profile of ETVPD-3is also
attractive, as it almost duplicates the protection of VPD.
In order to further test the capability of the PHE-like
drugs to block the Naþ channels, ETVPD-3has been
tested in its competition with the binding of [3H]batra-
chotoxinin A-20a-benzoate to the VGSCs from rat
brain synaptosomes. It was found to block the Naþ
channels in a dose dependent manner. Due to its limited
solubility, an IC50 could not be calculated. It would be
interesting to do the same test for SUVPD-1, as it is the

S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
3865
most active drug of the set. However, it does not meet
the solubility requirements that are necessary to perform
the binding assay.
valpromides, resembles the second lipophilic moiety that
has been proposed within the pharmacophore previ-
ously proposed for semicarbazides and urelylene anti-
convulsants.21 According to our analysis, based on a
broader set of dissimilar structures, the lipophilic por-
tion (Fig. 2) does not define a requirement, but certainly
helps in improving the activity.
2.2.3. Structure–activity relationships. The valpromide
derivatives are shown in Figure 10, aligned with the CZ–
PHE template. The modifications of the parent VPD
structure do not influence the interaction with the lipo-
philic pocket, as they involve substitution in the N-atom
of the polar moiety.
3. Conclusions
The protection against MES increases from VPD to
ETVPD-8. Two carbon atoms in the N-alkyl chain
seems to define the optimum size for this kind of sub-
stituents, as its increase to three or four centers (IPVPD-
3, BUVPD-4) shows a negative effect. An inverse cor-
relation with the length of the alkyl chain seems to hold
for N-substituted VPD’s when no additional function is
present in the substituent (VPD, ETVPD-8, IPVPD-3,
BUVPD-4). However, an alkyl bulky substituent
(CHVPD-2) surprisingly increases the antiepileptic
activity. It may be inferred that a lipophilic effect helps
in the delivery of the drug, whereas a linear increase of
the chain counterbalance this effect due to steric repul-
sions. We have synthesized N-aryl substituted VPD
derivatives in order to analyze the effect of this sub-
stituent on the activity, but no definite conclusions have
The pharmacological profile of a set of valpromide
derivatives has been extensively analyzed. The pharma-
cological behavior resembles PHE being active against
MES and inactive against PTZ. The derivatives comply
with the requirements imposed by a pharmacophore
proposed for the anti-MES activity. The pharmaco-
phoric pattern, on the other hand, has been shown to
represent the structural and electronic features com-
patible with a Naþ channel blocking mechanism.
SUVPD-1 appears as the most potent anti-MES drug of
the set. The efficacy of SUVPD-1 to prevent electrically
induced convulsions, and the demonstrated resistance of
N-substituted valpromides to biotransformation, point
to this set of structures as promising candidates for the
development of new drugs efficient for antiepileptic
therapy.
been extracted.
A sulfonamidophenyl substituent,
bearing a –SO2NH2 function, increases the MES pro-
tection, approaching the ED50 value of PHE (Tables 1
and 3).
On the basis of the largely accepted assumption that the
drugs with a PHE-like profile block the VGSC as one of
their related mechanisms, we have tested this behavior
by means of in vitro tests. This mechanism has been
tested positive for ETVPD-8, which is the most potent
derivative, among those that meet the solubility
requirements for the binding assay.
Surprisingly, substitution of the –SO2NH2 group by
–COOH (CPVPD-5) markedly reduces the activity,
changing, at the same time, the pharmacological profile
of the drug, which turns out to show protection against
both MES and PTZ tests. The pharmacological re-
sponse of CPVPD-5 closely resembles that of VPA,
being 40% more active against MES than against PTZ
test (Table 3). This fact leads to consider the possible
hydrolysis of 5 to VPA. In order to confirm this infer-
ence, the hydrolysis has been analyzed in vitro, follow-
ing the procedure described in Ref. 78. VPA and
p-aminobenzoic acid were obtained from the reaction. On
the other hand, diaryl substitution leads to a derivative
(BZVPD-11) not soluble enough to allow the evaluation
of its response in the animal models.
4. Experimental
4.1. Chemistry
1H and 13C NMR spectra were recorded on a Bruker
AC-200 spectrometer with tetramethylsilane as an
internal standard. Melting points were determined on an
Electrothermal IA6304 apparatus and are uncorrected.
Elemental analyses were measured on a Carlo Erba EA
1108. Solvents were purified and dried by standard
procedures. The evolution of the reaction was moni-
tored by TLC (thin layer chromatography). All TLC
data were determined using aluminum sheets with silica
gel 60 F254 (Merck-Darmstadt, Germany). Cyclohexyl-
amine, benzhydryl amine, ethyl amine, and valproic acid
were from Aldrich (Milwaukee, USA). N-butyl amine,
isopropyl amine, and diethyl amine were from Eastman
Kodak. 4-aminobenzoic acid and sulfanilamide were
from Merck, whereas dimethylamine was from Carlo
Erba. The amines were distilled previous use. Valpro-
mide has been prepared from valproic acid by standard
procedures.
The consideration of N-valproylglycinamide (TV1901),3
opens an optimistic insight on the set of compounds
presented in this paper. TV1901 is less active than VPD,
but more than VPA, when administered to rodents.
However, its resistance to hydrolysis makes this drug
more effective than VPD in humans, as the latter bio-
transforms to VPA. N-Substituted VPD’s, with
improved performance in animal models become, then,
highly promising for the development of new drugs for
antiepileptic therapy, based on the grounds of the higher
resistance to hydrolysis induced by N-substitution.
The lipophilic substitution of the polar moiety, which is
present in RAL and LAM, as well as in the substituted

3866
S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
4.2. General procedure for the preparation of compounds
11.35%; N, 7.57%. Found: C, 71.63%; H, 11.40%; N,
7.61%.
4.2.1. Compounds 1–7. Method A. A solution of valproyl
chloride (33 mmol) in anhydrous acetone (10 mL, com-
pounds 1 and 5) or anhydrous benzene (10 mL, com-
pounds 2–4, 6, and 7) was added dropwise to a solution
of 30 mmol of the reactant in anhydrous acetone (or
benzene), containing anhydrous pyridine (70.0 mmol).
(Reactant: 4- sulfanilamide: compound 1; cyclohexyl-
amine: compound 2; isopropyl amine: compound 3;
n-butyl amine: compound 4; 4-aminobenzoic acid: com-
pound 5; diethyl amine: compound 6; benzhydryl amine:
compound 7). The reaction mixture was stirred for a
period of 24–48 h at room temperature and concentrated
in vacuo. The residue was purified by recrystallization
(1–5 and 7) or distillation under reduced pressure (6).
4.2.6. N-Butylvalpromide (4). Yield 58% (acetone), mp:
1
70.0–71.0 °C. H NMR (chloroform-d3) d 0.82–0.93(m,
9H, 3Me), 1.15–1.69 (m, 12H, 6CH2), 1.94–2.05 (m, 1H,
CH), 3.17–3.27 (m, 2H, CH2N), 5.72 (br, 1H, NH); 13C
NMR (chloroform-d3) d 13.6 (butyl CH3), 14.0 (CH3),
20.0 (butyl CH2), 20.7 (CH2), 31.8 (butyl CH2), 35.3
(CH2), 38.9 (CNH), 47.6 (CH), 175.8 (COO). Anal.
Calcd for C12H25NO: C, 72.28%; H, 12.66%; N, 7.02%.
Found: C, 72.53%; H, 12.55%; N, 6.99%.
4.2.7. 4-(Valproylamido)benzoic acid (5). Yield 70%
(acetone), mp: 233.0–235.0 °C. TLC Rf 0.59 (chloro-
form/ethanol/water/ammonia, 18:20:2:1). IR (KBr):
1
3301 cmÀ1 (NH), 1688 cmÀ1 (COO). H NMR (chloro-
4.2.2. Compounds 8 and 9. Method B. The amine was
bubbled into a solution of valproyl chloride (3mL,
18.2 mmol) in anhydrous benzene (10 mL). The mixture
was stirred at 0–4 °C for 1 h and at room temperature
for 48 h. The solid was filtered and washed twice with
3mL of benzene. The filtrate was evaporated under
reduced pressure to give a white solid and the residue
was purified by recrystallization (8) or distillation under
reduced pressure (9).
form-d3) d 0.89–0.96 (m, 6H, 2Me), 1.25–1.75 (m, 8H,
4CH2), 2.38–2.51 (m, 1H, CH), 4.88 (br, 1H, NH), 7.67–
7.73(m, 2H, ar), 7.94–8.01 (m, 2H, ar); 13C NMR
(chloroform-d3) d 16.7 (CH3), 24.1 (CH2), 38.8 (CH2),
51.1 (CH), 122.7, 129.3, 134.1, 146.5 (C6H4), 171.8
(CONH), 180.4 (COOH). Anal. Calcd for C15H21NO3:
C, 68.38%; H, 7.98%; N, 5.32%. Found: C, 68.22%; H,
7.94%; N, 5.36%.
4.2.8. N,N-Diethylvalpromide (6). Yield (54%), bp 92 °C
1.8 mmHg. TLC Rf 0.73(dichloromethane/methanol,
20:1). 1H NMR (chloroform-d3) d 0.84–0.93(m, 6H,
2Me), 1.06–1.70 (m, 14H, 4CH2 and ethyl 2CH3), 2,50–
2.61 (m, 1H, CH), 3.28–3.43 (m, 4H, CH2 NCH2); 13C
NMR (chloroform-d3) d 13.0 (ethyl CH3), 14.8 (ethyl
CH3), 13.9 (CH3), 14.2 (CH3), 20.5 (CH2), 20.8 (CH2),
34.4 (CH2), 35.4 (CH2), 40.4 (NCH2), 45.0 (NCH2), 41.8
(CH), 175.6 (CONH). Anal. Calcd for C12H25NO: C,
72.36%; H, 12.56%; N, 7.04%. Found: C, 72.60%; H,
12.55%; N, 7.07%.
4.2.3. 4-(Valproylamido)benzenesulfonamide (1). Yield
65% (acetone), mp: 278–279 °C.59 TLC Rf 0.5 (dichloro-
methane/acetone, 5:1). 1H NMR (acetone-d6) d 0.89–
0.93 (t, 6H, 2Me), 1.33–1.70 (m, 8H, 2CH2CH2), 2,45–
2.49 (m, 1H, CH), 6.44 (s, 2H, SO2NH2), 7.82–7.87 (m,
4H, C6H4), 9.43(s, 1H, CONH); 13C NMR (acetone-d6)
d 13.9 (CH3), 20.9 (CH2), 35.5 (CH2), 47.8 (CH), 119.2,
127.4, 138.9, and 142.9 (C6H4), 175.3(CONH). Anal.
Calcd for C14H22N2O3S: C, 56.38%; H, 7.38%; N,
9.40%; S, 10.74%. Found: C, 56.56%; H, 7.32%; N,
9.40%; S, 10.79%.
4.2.9. N-Benzhydrylvalpromide (7). Yield 30% (MeOH),
mp: 174–176 °C. TLC Rf 0.77 (dichloromethane). 1H
NMR (chloroform-d3) d 0.85–0.92 (t, 6H, J ¼ 7:0,
2Me), 1.20–1.70 (m, 8H, 4CH2), 2.09–2.16 (m, 1H, CH),
5.99–6.04 (d, 1H, J ¼ 7:8, NH), 6.29–6.33 (d, 1H,
J ¼ 8:0, benzhydryl CH), 7.21–7.39 (m, 10H, benzhy-
dryl 5CH2); 13C NMR (chloroform-d3) d 14.0 (CH3),
20.7 (CH2), 35.3 (CH2), 47.6 (CH), 56.6 (CHNH), 127.3,
127.4, 128.5, 141.7 (C6H5), 175.0 (COO). Anal. Calcd
for C21 H27 NO: C, 81.55%; H, 8.74%; N, 4.53%. Found:
C, 81.57%; H, 8.77%; N, 4.53%.
4.2.4. N-Cyclohexylvalpromide (2). Yield 72% (MeOH),
1
mp: 172.0–172.5 °C. H NMR (chloroform-d3) d 0.85–
0.96 (m, 6H, 2Me), 1.03–1.19 (m, 18H, 9CH2), 1.97–2.03
(m, 1H, CH), 3.71–3.89 (m, 1H, cyclohexyl CH), 5.41
(br, 1H, NH); 13C NMR (chloroform-d3) d 14.1 (2CH3),
20.7 (CH2), 24.8, 25.5, 33.3 (cyclohexyl) 35.3 (CH2), 47.8
(CH), 174.8 (COO). Anal. Calcd for C14H27NOC,
74.66%; H, 12.00%; N, 6.22%. Found: C, 74.82%; H,
12.02%; N, 6.29%.
4.2.5. N-Isopropylvalpromide (3). Yield 88% (acetone),
mp 131–132 °C. TLC Rf 0.69 (dichloromethane/metha-
4.2.10. N-Ethylvalpromide (8). Yield 52% (MeOH), mp
93–95 °C. TLC Rf 0.65 (dichloromethane/methanol,
20:1). 1H NMR (chloroform-d3) d 0.86–0.97 (m, 6H,
2Me), 1.10–1.17 (t, 3H, J ¼ 7:26 ethyl CH3), 1.21–1.71
(m, 8H, 4CH2), 1.92–2.04 (m, 1H, CH), 3.24–3.37 (m,
2H, CH2N), 5.42 (br, 1H, NH); 13C NMR (chloroform-
d3) d 14.0 (CH3), 14.9 (ethyl CH3), 20.7 (CH2), 34.0
(ethyl CH2), 35.2 (CH2), 47.5 (CH),175.8 (CO). Anal.
1
nol, 15:1). H NMR (chloroform-d3) d 0.86–0.93(t, 6H,
J ¼ 7:0, 2Me), 1.13and 1.16 (2s, 6H, isopropyl 2CH ),
3
1.21–1.65 (m, 8H, 4CH2), 1.90–1.96 (m, 1H, CH), 4.06–
4.17 (m, 1H, isopropyl CH), 5.22 (br, 1H, NH); 13C
NMR (chloroform-d3) d 14.0 (CH3), 20.7 (CH2), 22.8
(isopropyl CH3), 35.3 (CH2), 40.9 (CHNH), 47.6 (CH),
174.9 (COO). Anal. Calcd for C11H23NO: C, 71.35%; H,

S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
3867
Calcd for C10H21NO: C, 70.17%; H, 12.28%; N, 8.19%.
Found: C, 70.37%; H, 12.26%, N, 8.16%.
The PTZ tests identify substances that raise the seizure
threshold. The freshly made solution of PTZ (1.7% in
0.9% saline solution) is administered subcutaneously (sc)
into a loose fold of skin in the midline of the neck in a
volume of 5 mL/kg body weight. This amount of PTZ
induces convulsions in more than 97% of mice. Animals
are observed for at least 30 min after sc injection of PTZ
for the presence or absence of a convulsive episode
persisting for at least 5 s. Absence of a clonic seizure
indicates protection.42
4.2.11. N,N-Dimethylvalpromide (9). Yield (30%), bp 75–
6 °C 1.2 mmHg. TLC Rf 0.67 (dichloromethane/metha-
1
nol, 20:1). H NMR (chloroform-d3) d 0.85–0.92 (t, 6H,
J ¼ 7:0, 2Me), 1.21–1.72 (m, 8H, 4CH2), 2.62–2.75 (m,
1H, CH), 2.96 and 3.06 (2s, 6H, CH3NCH3); 13C NMR
(chloroform-d3) d 14.1 (CH3), 20.8 (CH2), 31.1 (NCH3),
35.2 (NCH3), 35.5 (CH2), 47.7 (CH), 174.3(CONH).
Anal. Calcd for C10H21NO: C, 70.17%; H, 12.28%; N,
8.19%. Found: C, 70.06%; H, 12.31%; N, 8.16%.
4.3.4. Determination of the time of peak effect (TPE). The
test compounds were administered to groups of three–
five animals each that were tested at different times
(usually 30 min, 1, 2, and 4 h, or when the maximum
effect has been passed). The percentages of protection or
neurotoxicity are recorded and plotted against time and
the TPEs are determined by visual inspection of the
graphs.
4.2.12. Log P calculations. The log P calculated values
were estimated using the HyperChem QSAR Properties,
version 7.01 Hypercube, Inc.39
4.3. Pharmacological methods
4.3.5. Estimation of the ED50s. Groups of eight animals
per dose are injected via ip at least four doses, between
those that induce no protection (0% of the animals) and
total protection (100% of the animals), were assayed at
the TPE determined as previously described. The per-
centages of protection at each dose (converted to probit)
were plotted against log-doses. These data were then
subjected to statistical analysis and the ED50s with the
95% confidence intervals, slopes of the regression lines,
and standard errors of the slopes were estimated by the
method of Litchfield and Wilcoxon.71
4.3.1. In vivo experiments. The pharmacological tests
were performed according to standard procedures pro-
vided by the Antiepileptic Drug Development (ADD)
Program of the National Institute of Neurological and
Communicative Disorders and Stroke (NINCDS).60
Albino mice BALB/cN weighing between 25 and 30 g at
time of testing, were used as experimental animals. Mice
are housed in colony cages on a 12 h light/dark cycle,
and allowed free access to both commercial rodent chow
and water, except when they are removed from their
cages and placed on individual ones for the experimental
testing.
4.3.6. Neurotoxicity tests. The RotoRod test is used
exclusively in mice to assess minimal neurotoxicity. A
normal mouse can maintain its equilibrium on a rotating
rod (6 rpm) for long periods of time. Neurological deficit
is indicated by failure to maintain balance on a rotating
rod in each of three trials of 1 min each.
4.3.2. Preparation and injection of the testing compounds.
The compounds were dissolved in 30% PEG 400.51
A
volume of 10 mL/kg of the freshly made preparations
was injected intraperitoneally (ip). When higher doses of
the compounds were tested, compromising the solubil-
ity, the proportion of PEG 400 was increased, main-
taining the same maximum volume of 3mL/kg of PEG
400 and reducing the total volume to less than 10 mL/kg.
4.3.7. Estimation of the neurotoxic effects. Groups of
eight animals per dose were injected and tested with the
RotoRod test. The percentages of animals showing
minimal neurotoxicity were recorded and the higher
dose producing neurotoxic effects is reported. TD50
estimations could not be completed due to solubility
problems with high doses.
4.3.3. Anticonvulsant tests. Maximal electroshock sei-
zures were elicited in mice by delivering a 60 Hz/50 mA
electrical stimulus for 0.2 s via ear clip electrodes. A
drop of saline solution applied on each ear before
placing the electrodes ensures adequate electrical con-
tact. In these conditions, maximal seizures are produced
in virtually all normal mice. The maximal seizure typi-
cally consists of a short period of tonic flexion followed
by a longer period of tonic extension of the hind limbs
and a final clonic episode. Blockade of the hind limbs
tonic extensor component due to the drug treatment is
taken as the end point. The tonic component is con-
sidered abolished if the hindleg tonic extension does not
exceed a 90° angle with the trunk.42
4.4. In vitro experiments
4.4.1. Sodium channel binding assay. The assay was
performed in Novascreen Biosciences Co. (7170 Stan-
dard Drive, Hanover, Maryland 21076-1334) accord-
ing to reported methods72;73 with modifications.
[3H]Batrachotoxin (KD ¼ 32 nM) with a specific activity
of 30–60 Ci/mmol was used in a final concentration of
2.0 nM. Rat forebrain membranes were the source of
receptors (Bmax ¼ 52 fmol/mg tissue as wet weight).
The experiments were performed in triplicate. The

3868
S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
10. Pevarello, P.; Bonsignori, A.; Caccia, C.; Amici, R.;
McArthur, R.; Fariello, R.; Salvati, P.; Varasi, M. Bioorg.
Med. Chem. Lett. 1999, 9, 2521.
11. Villetti, G.; Bregola, G.; Bassani, F.; Bergamaschi, M.;
Rondelli, I.; Pietra, C.; Simonato, M. Neuropharmacology
2001, 40, 866.
12. (a) Andurkar, S.; Beguin, C.; Stables, J.; Kohn, H. J. Med.
Chem. 2001, 44; (b) Grimm, J. B.; Stables, J. P.; Milton,
L. B. Bioorg. Med. Chem. 2003, 11, 4133.
13. Treiman, D. M. Epilepsia 2001, 42(Suppl. 3), 8.
14. Brodie, M. J.; Dichter, M. A. N. Engl. J. Med. Chem.
1996, 334, 1583.
nonspecific determinant was aconitine in a concentra-
tion of 1.0 mM. Reactions were carried out in 50 nM
HEPES (pH 7.4) containing 130 mM choline chloride at
37 °C for 60 min. The reaction was terminated by rapid
vacuum filtration of the reaction contents onto glass fi-
ber filters. Radioactivity trapped onto the filters was
determined and compared to control values in order to
ascertain any interactions of test compound with the
sodium channel, site 2 binding site.
15. (a) White, H. S. Epilepsia 1999, 40(Suppl. 5), S2; (b)
White, H. S. Epilepsia 2003, 44(Suppl. 7), 2.
16. Ragsdale, D. S.; Avoli, M. Brain Res. Rev. 1998, 26, 16.
17. Pugsley, M. K.; Yu, E. J.; McLean, T. H.; Goldin, A. L.
Proc. West. Pharmacol. Soc. 1999, 42, 105.
18. Netzeva, T.; Doytchinova, I.; Natcheva, R. Pharm. Res.
2000, 17(6), 727.
19. Unverferth, K.; Engel, J.; Hofgen, N.; Rostock, A.;
Gunther, R.; Lankau, H.; Menzer, M.; Rolfs, A.; Lieb-
scher, J.; Muller, B.; Hofmann, H. J. Med. Chem. 1998,
41, 63 .
20. Malawska, B.; Kulig, K.; S’Piewaka, A.; Stables, J.
Bioorg. Med. Chem. 2004, 12, 625.
21. Pandeya, S. N.; Raja, A. S. J. Pharm. Sci. 2002, 266.
22. Cox, B.; Nobbs, M. S.; Shah, G. P.; Edney, D. D.; Loft,
M. S. Patent WO9838174, 1998.
4.4.2. Computational procedure. The structures of all the
ligands have been geometry optimized with no con-
strains at the same level of theory. Density functional
calculations have been performed using GAUSSIAN 98
(G98)74 and the Becke’s three-parameter hybrid func-
tional75 with LYP correlation functional (B3LYP).76
Basis were of double zeta split valence plus polarization
quality (6-31G**) The electronic characteristics have
been derived from a fitting to applied electrostatic
potentials (CHELPG),77 using the same level of theory
and the same program package. The superposition
43
analysis has been performed using the SYBYL software
version 6.6 on a Silicon Graphics Iris Octane MIPS
R12000 computer.
23. Tasso, S. M.; Bruno-Blanch, L. E.; Estiu, G. L. J. Mol.
Model. 2001, 7, 231.
24. Brown, M. L.; Zha, C. C.; Van Dyke, C. C.; Brown, G. B.;
Brouillette, W. J. J. Med. Chem. 1999, 42, 1537.
25. Cox, B.; Healy, M. P.; Nobbs, M. S.; Shah, G. P. Patent
WO9932462, 1999.
26. Salvati, P.; Maj, R.; Caccia, C.; Cervini, M. A.; Forna-
retto, M.; Lamberti, E.; Pevarello, P.; Skeen, G. A.; White,
H. S.; Wolf, H. H.; Faravelli, L.; Mazzanti, M.; Manci-
nelli, E.; Varasi, M.; Fariello, R. J. Pharmacol. Exp. Ther.
1999, 288, 1151.
27. Fariello, R.; McArthur, R. A.; Bonsignori, A.; Cervini, M.
A.; Maj, R.; Pevarello, P.; Wolf, H. H.; Woodhead, J. W.;
White, H. S.; Varasi, M.; Salvati, P.; Post, C. J. Pharma-
col. Exp. Ther. 1998, 285, 397.
28. Reddy, P. A.; Hsiang, B. C.; Latifi, T. N.; Hill, M. W.;
Woodward, K. E.; Rothman, S. M.; Ferendelli, J. A.;
Covey, D. F. J. Med. Chem. 1996, 39, 1898.
Acknowledgements
G.L.E. has performed this research as a member of the
ꢀ
Consejo Nacional de Investigaciones Cientıficas y
ꢀ
ꢀ
Tecnicas de la Republica Argentina (CONICET),
L.E.B.-B. and S.Ch.M. are researchers of the Facultad
de Ciencias Exactas, Universidad Nacional de La Plata.
S.M.T. is a fellowship holder of CONICET. This work
was supported in part through grants from Agencia de
ꢀ
ꢀ
ꢀ
Promocion Cientıfica y Tecnologica (PICT 98-06-03237)
CONICET (PIP No 02207 Res 1478) and Universidad
Nacional de La Plata, Argentina.
29. Hill, M. W.; Reddy, P. A.; Covey, D. F.; Rothman, S. M.
J. Pharmacol. Exp. Ther. 1998, 285, 1303.
30. Rundfelt, C. Epilepsy Res. 1999, 34, 57.
References and notes
31. Ferrendelli, J. A.; Holland, K. D. In Antiepileptic Drugs,
3rd ed.; Levy, R., Mattson, R., Meldrum, B., Penry, J. K.,
Dreifuss, F. E., Eds.; Raven: New York, 1989; Vol. 1.
32. Fisher, R. S. Brain Res. Rev. 1989, 14, 245.
33. Benes, J.; Parada, A.; Figueirredo, A. A.; Alves, P. C.;
Freitas, A. P.; Learmmonth, D. A.; Cunha, R. A.; Garrett,
J.; Soares-da-Silva, P. J. Med. Chem. 1999, 42, 2582.
34. Taylor, C. P.; Meldrum, B. S. TiPS 1995, 16, 309.
35. Vamecq, J.; Lambert, D.; Poupaert, J.; Masereel, B.;
Stables, J. P. J. Med. Chem. 1998, 41, 3307.
36. Maillard, M. C.; Perlman, M. E.; Amitay, O.; Baxter, D.;
Berlove, D.; Connaughton, S.; Fischer, J. B.; Guo, J.; Hu,
L.; McBurney, R. N.; Nagy, P. I.; Subbarao, K.; Yost, E.
A.; Zhang, L.; Durant, G. J. J. Med. Chem. 1998, 41, 3048.
37. Fischer, W.; Bodewel, R.; Satzinger, G. Arch. Pharmacol.
1992, 346, 442.
1. Rogawski, M.; Porter, R. Pharmacol. Rev. 1990, 49(3),
122.
2. Loscher, W.; Schmidt, D. Epilepsy Res. 1994, 17, 95.
3. (a) Bialer, M.; Johannessen, S. I.; Kupferberg, H. K.;
Levy, R. H.; Loiseau, P.; Perucca, E. Epilepsy Res. 1996,
25, 299; (b) Bialer, M.; Johannessen, S. I.; Kupferberg, H.
K.; Levy, R. H.; Loiseau, P.; Perucca, E. Epilepsy Res.
2002, 51, 3 1.
4. Bialer, M.; Johannessen, S. I.; Kupferberg, H. K.;
Levy, R. H.; Loiseau, P.; Perucca, E. Epilepsy Res. 2001,
43, 11.
5. Anger, T.; Madge, D.; Mulla, M.; Riddall, D. J. Med.
Chem. 2001, 44, 115.
6. Moldrich, R. X.; Beart, P. M.; Jane, D. E.; Chapman,
A. G.; Meldrum, B. S. Neuropharmacology 2001, 40, 732.
7. Vamecq, J.; Bac, P.; Herrenknecht, C.; Maurois, P.;
Delcourt, P.; Stables, J. J. Med. Chem. 2000, 43, 1311.
8. Loscher, W.; Leppik, I. E. Epilepsy Res. 2002, 50, 17.
9. Jain, K. K. Exp. Opin. Invest. Drugs 2000, 9(4), 1.
ꢀ
38. Molnar, P.; Erdo, S. L. Eur. J. Pharmacol. 1995, 273, 303.
39. QSAR Properties Version 7.00. HyperChem Release 7.01
ꢀ
for Windows Molecular Modeling System 2002. Hypecube
Inc. 1115 NW 4th Street, Gainesville, Florida 32601, USA.

S. M. Tasso et al. / Bioorg. Med. Chem. 12 (2004) 3857–3869
3869
40. Willow, M.; Catterall, W. A. Mol. Pharmacol. 1982, 22,
627.
R., Meldrum, B., Penry, J. K., Dreifuss, F. E., Eds.;
Raven: New York, 1989; pp 85–102.
41. Willow, M.; Kuenzel, E. A.; Catterall, W. A. Mol.
Pharmacol. 1984, 25, 228.
64. Scriba, G. K. E. Arch. Pharm. Pharm. Med. Chem. 1996,
329, 554.
42. Chen, A.; Weston, J. K.; Bratton, A. C. Epilepsia 1963, 4,
66.
43. SYBYL is a product Tripos, Inc., San Diego CA.
44. Loscher, W.; Nau, H. Neuropharmacology 1985, 24(5), 427.
45. Abbott, F. S.; Acheampong, A. A. Neuropharmacology
1988, 27(3), 287.
65. Tantisira, B.; Tantisira, M. H.; Patarapanich, C.; Sook-
sawate, T.; Chunngam, T. Res. Commun. Mol. Pathol.
Pharmacol. 1997, 97, 151.
66. Bialer, M.; Haj-Yehia, A.; Badir, K.; Hadad, S. Pharmacy
World & Science 1994, 16, 2.
67. Masereel, B.; Rolin, S.; Abbate, F.; Scozzafava, A.;
Supuran, C. J. Med. Chem. 2002, 45(2), 312.
46. Elmazar, M. M.; Hauck, R. S.; Nau, H. J. Pharm. Sci.
1993, 82(12), 1255.
ꢀ
68. Chufan, E. E.; Pedregosa, J. C.; Baldini, O. N.; Bruno-
47. Perucca, E. CNS Drugs 2002, 16(10), 695.
48. Hadad, S.; Vree, T.; Van der Kleijn, E.; Bialer, M.
J. Pharm. Sci. 1992, 81, 1047.
49. Rekatas, G. V.; Tani, E.; Demopoulos, V.; Kourounakis,
P. Arch. Pharm. Pharm. Med. Chem. 1996, 329, 393.
50. Levi, M.; Yagen, B.; Bialer, M. Pharm. Res. 1997, 14(2),
213.
51. Keane, P. E.; Simiand, J.; Mendes, E.; Santucci, V.
Neuropharmacology 1983, 22, 875.
52. Loscher, W.; Nau, H. Neuropharmacology 1985, 24,
427.41.
Blanch, L. Il Fꢀarmaco 1999, 54, 838.
69. Tasso, S.; Bruno-Blanch, L.; Moon, S. Ch.; Estiu, G.
J. Mol. Struct. (THEOCHEM) 2000, 504, 229.
70. Maryanoff, B.; Costanzo, M.; Nortey, S.; Greco, M.;
Shank, R.; Schupsky, J.; Ortegon, M.; Vaught, J. J. Med.
Chem. 1998, 41, 1315.
71. Litchfield, J. T.; Wilcoxon, F. J. Pharmacol. Exp. Ther.
1949, 96, 99.
72. Trainer, V. L.; Moreau, E. J. Biol. Chem. 1993, 268(23),
17114.
73. Creveling, C. R. Mol. Pharmacol. 1983, 23, 350.
74. Frisch M. J.; Trucks, G. W.; Schlegel, H. B; Scuseria, G.
E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, J. A.; Stratman, R. E.; Burant, J. C.;
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.;
Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, C.; Liashenko,
A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R.
L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P.
M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres,
J. L.; Gonzales, C.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. GAUSSIAN 98; Gaussian: Pittsburgh PA;
1998.
53. Abbott, F. S.; Acheampong, A. A. Neuropharmacology
1988, 27, 287.
54. Bialer, M.; Hadad, S.; Kadry, B.; Abdul-Hai, A.; Haj-
Yehia, A.; Sterling, J.; Herzig, Y.; Yagen, B. Pharm. Res.
1996, 13, 284; Bialer, M. Clin. Pharmacokinet. 1991, 20, 114.
55. Haj-Yehia, A.; Bialer, M. Pharm. Res. 1989, 6, 682.
56. Haj-Yehia, A.; Hadad, S.; Bialer, M. Pharm. Res. 1992, 9,
1058.
57. Bialer, M.; Kadry, B.; Abdul-Hai, A.; Haj-Yehia, A.;
Sterling, J.; Herzig, Y.; Shirvan, M. Biopharm. Drug
Dispos. 1996, 17, 565.
58. Hadad, S.; Bialer, M. Pharm. Res. 1995, 12, 905.
59. Spiegelstein, O.; Kroetz, D. L.; Levy, R. H.; Yagen, B.;
Hurst, S. I.; Levi, M.; Haj-Yehia, A.; Bialer, M. Pharm.
Res. 2000, 17, 216.
60. Porter, M.; Cereghino, M.; Gladding, R.; Hessie, B.;
Kupferberg, D.; Scoville, M.; White, D. Clev. Clin. Q.
1984, 51, 293.
61. Edafiogho, I.; Scott, K. In Burger’s Medicinal Chemistry
and Drug Discovery, 5th ed.; Wolff, M., Ed.; John Wiley
and Sons, 1996.
62. Gladding, G. D.; Kupferberg, H. J.; Swinyard, E. A. In
Handbook Exp. Pharm. 1985; Vol. 74, pp 341–347.
63. Swinyard, E. A.; Woodhead, J. H.; White, H. S.; Franklin,
M. R. In Antiepileptic Drugs, 3rd ed.; Levy, R., Mattson,
75. Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
76. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, B37,
785.
77. Breneman, C. N.; Wiberg, K. B. J. Comp. Chem. 1990,
11, 361; Breneman, C. N.; Wiberg, K. B. Science 1994, 16,
2.
78. Greenwal, R. B.; Zhao, H.; Yang, K.; Reddy, P.;
Martinez, A. J. Med. Chem. 2004, 47, 726.