1,2-Ethane bis-1-amino-4-benzamidine is active against several brain insult and seizure challenges through anti-NMDA mechanisms targeting the 3H-TCP binding site and antioxidant action

Article abstract of DOI:10.1016/j.ejmech.2010.03.044

Five bis-benzamidines were screened towards murine magnesium deficiency-dependent audiogenic seizures, unravelling two compounds with efficacious doses 50 (ED₅₀) less than 10 mg/kg. They were also screened against maximal electroshock and subcutaneous pentylenetetrazole- induced seizures, and explored for superoxide -scavenging activity. 1,2-Ethane bis-1-amino-4-benzamidine (EBAB) was selected and evaluated in 6 Hz seizure test (ED₅₀= 49 mg/kg) and at 4 μg/kg in focal cerebral ibotenate poisoning in pups (sizes of both white and grey matter wounds were halved). EBAB was further tested on NMDA-induced seizures in mice (ED₅₀=6 mg/kg) and on 3H-TC -binding to a rodent cerebral preparation (IC₅₀=1.4 μM). Taken as a whole, present data emphasise the suitability of bis-benzamidines as templates for designing brain protective compounds.

Full text of DOI:10.1016/j.ejmech.2010.03.044

European Journal of Medicinal Chemistry 45 (2010) 3101e3110
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
1,2-Ethane bis-1-amino-4-benzamidine is active against several brain insult and
seizure challenges through anti-NMDA mechanisms targeting the 3H-TCP binding
site and antioxidant action
,
b
b,
Joseph Vamecq a , Pierre Maurois , Nicole Pages c, Pierre Bac b, James P. Stables d, Pierre Gressens e,
*
Dimitri Stanicki f, Jean Jacques Vanden Eynde f
a Inserm Univ 045131, Faculty of Medicine, Research Branch, University of Lille 2, France and Département de Biochimie et de Biologie Moléculaire, Laboratoire d’Endocrinologie,
Métabolisme-Nutrition & Oncologie, Centre de Biologie et Pathologie Pierre Marie Degand, Centre Hospitalier Régional Universitaire de Lille, 59037 Lille, France
b Faculté de Pharmacie, Université Paris Sud 11, Châtenay-Malabry & CNRS UMR 8162, IFR 13, Centre Chirurgical Marie Lannelongue, Le Plessis Robinson, France
c Laboratoire de Toxicologie, Faculté de Pharmacie, Université Louis Pasteur, Illkirch, France
d Preclinical Pharmacological Section, Epilepsy Branch, DCDND, NINDS, National Institutes of Health, Neuroscience Center, Rockville, MD 20892-9523, USA
e Inserm U676, Neurosciences, Neuropédiatrie, Hôpital Robert-Debré & Université Paris 7, Faculté de Médecine Denis Diderot, IFR02 and IFR25, Paris, France
f Laboratory of Organic Chemistry, University of Mons-UMons, Mons, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
Five bis-benzamidines were screened towards murine magnesium deficiency-dependent audiogenic
seizures, unravelling two compounds with efficacious doses 50 (ED50) less than 10 mg/kg. They were also
screened against maximal electroshock and subcutaneous pentylenetetrazole-induced seizures, and
explored for superoxide -scavenging activity. 1,2-Ethane bis-1-amino-4-benzamidine (EBAB) was
Received 11 November 2009
Received in revised form
24 March 2010
Accepted 29 March 2010
Available online 3 April 2010
selected and evaluated in 6 Hz seizure test (ED50 ¼ 49 mg/kg) and at 4
mg/kg in focal cerebral ibotenate
poisoning in pups (sizes of both white and grey matter wounds were halved). EBAB was further tested on
NMDA-induced seizures in mice (ED50 ¼ 6 mg/kg) and on 3H-TC -binding to a rodent cerebral preparation
Keywords:
Bis-benzamidines
1,2-Ethane bis-1-amino-4-benzamidine
(EBAB)
(IC50 ¼ 1.4
mM). Taken as a whole, present data emphasise the suitability of bis-benzamidines as
templates for designing brain protective compounds.
Ó 2010 Elsevier Masson SAS. All rights reserved.
Ibotenate challenge
seizure tests
3TCP-binding site
Magnesium deficiency-dependent
audiogenic seizures (MDDAS)
1. Introduction
rate to yield intracerebral concentrations compatible with effica-
cious pharmacological activity. In this respect, animal models of
Alzheimer’s and Parkinson’s diseases, multiple sclerosis and stroke
may rest on experimental conditions in which the disruption of the
BBB is more rapid and/or severe than that developing in the cor-
responding human disorder. Obviously getting information on the
pathogenesis, these animal models might be less suitable for
evaluation of drugs in a way translatable to human patients. To
overcome this issue, some laboratories have developed in vitro
cellular models combining endothelial and astrocytic cells aimed at
mimicking the BBB in a relatively predictive way [3]. An alternative
approach is search for an animal model which would combine its
suitability for neuroprotective evaluation along with a relative
integrity of its BBB. In fact, the brain neurotoxicity challenge should
be such that BBB would not be altered. This is not easy to obtain in
practice because BBB injury is precisely an expected component of
brain injury conditions.
Though search for new compounds exhibiting brain anticon-
vulsant and neuroprotective properties has literally exploded in the
last decade, many promising neuroprotectants have seen their
pharmacological development compromized [1,2]. Importantly, the
compound needs to cross bloodebrain barrier (BBB) at a sufficient
Abbreviations: MES, maximal electroshock seizures; scPtz, subcutaneous
pentylenetetrazol; EBAB, 1,2-Ethane bis-1-amino-4-benzamidine; ED50, efficacious
doses 50 (dose protecting 50% animals); IC50
(concentration inhibiting 50% of the binding); MDDAS, magnesium deficiency-
dependent audiogenic seizures; BBB, bloodebrain barrier; NMDA, N-methyl-
, inhibitory concentration 50
D
-
aspartate; TCP, the thienyl analogue of phencyclidine, namely N-[1-(2-thienyl)
cyclohexyl]-3,4-piperidine).
* Corresponding author. Tel.: þ33 676 06 10 85; fax: þ 33 320 44 56 93.
E-mail address: joseph.vamecq@inserm.fr (J. Vamecq).
0223-5234/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.03.044

3102
J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
Nutritional magnesium deprivation provides mice with brain
7.5 mg/kg and 6 mg/kg, respectively. In the 50% unprotected mice
submitted to these doses, duration of the latency plus wild run-
nning as well as duration of seizures were reduced whereas dura-
tion of the recovery phase was not significantly affected (Table 1).
These changes in the durations of phases distinguished clearly from
those of reference compounds including GABAergic and sodium
blocker prototype drugs previously tested in the audiogenic seizure
model, suggesting distinct protective mechanisms for compounds 1
and 2 (EBAB).
toxicity including central hyperactivity and susceptibility to
audiogenic seizures which develop into four successive phases
(latency, wild running, convulsion and recovery phases) [4]. This
magnesium deficiency-dependent audiogenic seizure (MDDAS)
model is the resultant of two factors: susceptibility to audiogenic
seizures (linked to the generalized oxidative/inflammatory state of
magnesium deficiency) and development of seizures (triggered by
an acoustic stimulus acting as a pro-convulsant factor). Influencing
only one of these two factors is sufficient to prevent seizures;
susceptibility to and development of which respond to neuro-
protective (notably antioxidant) and anticonvulsant compounds,
respectively, all acting here as anti-seizure compounds. Elucidation
of underlying anticonvulsant or neuroprotective mechanisms
requires additional compound testing which may include
modulation of the audiogenic seizure phases mentioned above,
specific anticonvulsant evaluation in classic animal seizure tests,
and specific neuroprotective evaluation in well standardized brain
injury models. This overall strategy previously emphasized
original neuroprotective potentials of WEB2170 (an anti-PAF
3.1.2. MES, scPtz and rotarod tests
The series of five bis-benzamidines were further evaluated for
anticonvulsant properties in classic animal seizure models
including MES and scPtz tests (this subsection) and the 6 Hz seizure
test (next subsection). Table 2 provides screening data for the
activity of compounds in animals undergoing seizures induced by
electroshock (MES test) or pentylenetetrazol (scPtz test), their
minimal acute neurotoxicity being determined in the rotorod test.
Compounds were administered to mice by the intraperitoneal
route 30 min or 4 h before testing. Activity at these time points
prompted further studies on administration performed 15 min and
1 h before anticonvulsant or toxic evaluation. At doses tested (30,
100 and/or 300 mg/kg), compounds were inactive in the MES test at
the 30 min and 4 h time points and were toxic at the 30 min time
point. At 30 mg/kg dose, compound 2 (EBAB) was not toxic, con-
trasting with toxicity observed for other compounds at that dose at
the 15 min time point. The toxicity aggravated with increasing dose
from 30 to 100 and 300 mg/kg. At the 4 h time point and admin-
istering the 30 mg/kg dose, minimal acute neurotoxicity was not,
however, observed for compounds 2 (EBAB), 3, 4 and 5 despite that
toxicity was expressed by compound 1. The former compounds
were also not toxic at 100 mg/kg at this time point. Some activity
was recorded in the scPtz test for compounds 2 (EBAB) and 3
(active at 300 mg/kg) and 5 (active at 100 mg/kg) at 30 min but not
4 h. Additional evaluation in the scPtz test at the 15 min and 1 h
time points indicated that 100 mg/kg compound 2 (EBAB) was fully
active at these two time points while 100 mg/kg compound 3 was
active moderately at 1 h and inactive at the other studied time
points.
receptor antagonist) [4], FMOC-L-leucine (a PPARg agonist) [5]
and 1-methyl-2-[3-trifluoromethylphenyl]-4-mercapto-imidazole
(a synthetic ovothiol analogue) [6], and confirmed neuroprotective
potentialities of melatonin [4], ebselen and magnesium [7].
Importantly, compounds described as neuroprotective but inactive
in the MDDAS model, rosiglitazone [5] and fenofibrate (the authors,
unpublished data), represent compounds which do not cross the
intact BBB at a substantial rate [8,9]. Integrity of the BBB in the
MDDAS model was suggested by failure of systemic folic acid to
induce seizures [4]. All of this emphasizes that MDDAS might be
one of these in vivo models ensuring this rare combination of brain
neurotoxicity and relative integrity of BBB mentioned above and,
for this reason, might be helpful in detecting brain protective
compounds potentially suitable for human neurodegenerative
disorders. The present work was aimed at exploring whether one
might include among such compounds those issued from a short
series of five bis-benzamidines including compound 2, 1,2-ethane
bis-1-amino-4-benzamidine, referred throughout the text to as
EBAB.
2. Chemistry
3.1.3. The 6 Hz seizure test
Compounds 2 (EBAB) and 5 were selected for further evaluation
in the 6 Hz seizure test. This selection especially of compound 2
(EBAB) took into account the potent activity expressed in the
MDDAS test, which like the 6 Hz seizure test, is known also to
respond to a wide range of anticonvulsant compounds. In contrast
to compound 5 which was inactive at non toxic doses, compound 2
(EBAB) displayed a good activity in the 6 Hz seizure test, being here
endowed with a ED50 of 48.95 mg/kg.
Compounds 2 to 5 were synthesised as indicated in Scheme 1. A
first step resulted in the synthesis of bis-benzonitrile intermediates
obtained by reaction of a diamine with 4-fluorobenzonitrile in
boiling N,N-dimethylacetamide in the presence of a base. The
subsequent two steps allowed the conversion of the nitrile func-
tions into amidines, involving either the Pinner’s reaction for asy-
metric compounds 4 and 5 (imidate synthesis from an alcohol in
the presence of hydrochloric acid and further reaction with an
amine) or formation followed by reduction of an amidoxime for the
symetric compounds 2 and 3. The synthesis route followed for
compounds 4 and 5 was dependent on the solubility of the dinitrile
in the presence of hydrochloric acid. The chemical synthesis of
compound 1 was previously described [10,11].
3.2. Antioxidant and neuroprotective evaluation of compounds
The differences observed between ED50 values of EBAB in the
MDDAS test (6 mg/kg) and in the 6 Hz seizure test (49 mg/kg)
suggested that in the former test part of anti-seizure activity might
be accounted for by mechanisms uncovered by classic anti-seizure
tests, suggesting alternative protective mechanisms. So, in this
context, EBAB was evaluated for neuroprotective properties. In this
respect, antioxidant and so-called neuroprotective mechanisms of
EBAB were directly assessed in specific tests. These tests included
in vivo evaluation of protection given by EBAB in the focal intra-
cerebral ibotenate poisoning of rat immature brain and by in vitro
measurement of antioxidant activities of compounds.
3. Results
3.1. Anticonvulsant and neurotoxic evaluation of compounds
3.1.1. Audiogenic seizure test
The five compounds were evaluated in the MDDAS test. Only
compounds 1 and 2 (EBAB) were active at non toxic doses. The
efficacious doses 50 (ED50) of these compounds (doses providing
50% animals with protection against audiogenic seizures) were

J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
3103
Scheme 1. Synthesis pathway for compounds 2 to 5. A first step, common to the synthesis of compounds 2 to 5, resulted in the formation of dinitriles (bis-benzonitriles). The second
and third steps converted bis-benzonitriles to final compounds, involving either Pinner’s reaction (imidate formation [left portion of the chemical pathway]) for compounds 4 and 5,
or amidoxime reduction [right intermediate] for compounds 2 and 3. Compound 1 was synthesised as described in the literature (see Refs [10,11] in the text). The chemical names of
depicted compounds are: piperazine-N1,N4-bis-benzamidine (1), 1,2-ethane bis-1-amino-4-benzamidine (namely, 4,40-(1,2-ethanediyldiimino)bis-(N-hydroxybenzenecarbox-
imidamide)) (2) (EBAB), 1,2-ethane bis-1-methylamino-4-benzamidine (namely, 4,40-[1,2-ethanediyl (N,N0-dimethyl bis-nitrilo)]bis-(N-hydroxy benzenecarboximidamide)) (3), 1-
(1-amino-4-benzamidine)-2-(1-methylamino-4-benzamidine)-ethane (namely, 4,40-(N-methyl 1,2-ethanediyldiimino)bis-(benzene-carboxymidamide) dihydrochloride) (4), 1-
(1-amino-4-benzamidine)-2-(1-ethylamino-4-benzamidine)-ethane (namely, 4,40-(N-ethyl 1,2-ethanediyldiimino)bis-(benzene-carboximidamide) dihydrochloride) (5).
3.2.1. Antioxidant evaluation
neuroprotective compounds, namely the ibotenate-induced focal
cerebral injury model using pups. In this evaluation model, 4 g/kg
Compounds were evaluated for their ability to scavenge the
superoxide anion radical by determining their capacity to inhibit
rates of pyrogallol autooxidation, a reaction involving generation of
the superoxide anion radical. Table 3 gives an account for this
antioxidant evaluation and indicates that all five compounds were
endowed with substantial scavenging properties towards super-
oxide with potencies of 1 >3 > 2 EBAB > 5 > 4. Interestingly,
4-aminobenzamidine, a motif recovered in the chemical formula of
compounds, was quasi-devoid of the superoxide-scavenging
activity.
m
EBAB reduced significantly the size of lesions in both the cortical
plate (i.e. gray matter) and white matter by 49% and 69%, respec-
tively (Fig. 1). The protective effects of EBAB were of the same order
of magnitude as those afforded by 0.1 mg/kg leptin, a well accepted
neuroprotectant [12,13], the former being a little more active on
white matter and the latter on grey matter (Fig. 1).
3.3. Molecular modelling studies
Molecular modelling of compounds 1 to 5 was carried out to
visualize the tridimensional structure of their low energy confor-
mation (Fig. 2). These conformations were compared with that of
spermine, a compound which also exhibits protonated nitrogen
ends (namely, eNHþ3 ). The distance between the two protonated
3.2.2. Neuroprotective evaluation
EBAB was selected for direct evaluation of its protective activity
towards brain wounds. It was evaluated at the dose of 4
mg/kg in
a
well standardized model designed for the evaluation of

3104
J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
Table 1
Modulation by compounds 1 and 2 (EBAB) ED50 doses of durations of phases in the MDDAS test.
Duration of latency
plus wild running periods
Duration of
convulsions
Duration of recovery
Controls
Compound 1 (7.5 mg/kg)
Compound 2 (EBAB) (6 mg/kg)
7.2 ꢀ 1.5
1.6 ꢀ 0.4
50.3 ꢀ 5.6
48.0 ꢀ 3.2
50.3 ꢀ 3.5
6.05 ꢀ 1.45*
4.75 ꢀ 0.25**
0.79 ꢀ 0.24**
1.4 ꢀ 0.4*
Qualitative modifications by ED50 of
neuroprotective compounds devoid
of intrinsic anticonvulsant properties
anticonvulsant compounds acting on
voltage-gated sodium channels
GABAergic anticonvulsant compounds
ethosuximide
Increased
Increased
Increased
Decreased
Unaffected or decreased
Decreased
Increased
Unaffected
Decreased
Unaffected
Decreased
Unaffected
Adult OF1 mice receiving a magnesium-deficient diet (35-40 ppm for 27 days) were utilized. Drugs were given 1 hr before testing via the intraperitoneal route at a dose
corresponding to their previously determined ED50 in the MDDAS testing (for additional experimental details and rationale, see Ref. [4]). Phase durations are expressed as
seconds (mean values ꢀ SEM, n ¼ 5) in convulsing animals at drug dosing ensuring anticonvulsant protection in 50% and no protection in the other 50% of tested animals (those
concerned by this table). Controls refer to untreated magnesium-deficient mice (mice given neither vehicle nor drug), the administration of the compound vehicle being
without effect on the control values. The first three lanes of results are those experimentally obtained in the scope of this work. The data concerning the qualitative modi-
fication of durations of phases refer to previous works [4e6] and are given to allow a rapid comparison of phase modulation properties of compounds 1 and 2 (EBAB) with
those of reference prototype antiepileptic drugs.
*p < 0.05 and ** p < 0.01.
nitrogen ends of bis-benzamidine (namely, ¼NHþ2 ) was measured
for each compound, and compared with the distance between
protonated nitrogen ends of spermine. This distance was compa-
rable for 1, EBAB and spermine (15.55, 16.09 and 16.13 Å, respec-
tively), three compounds for which conformation was extended. By
contrast, this distance was lower for compounds 3, 4 and 5 (11.76,
12.70 and 12.21 Å, respectively) for which conformation was
partially folded. Though bioactive conformations are not neces-
sarily low energy conformations, EBAB might be, however,
considered to represent a spermine analogue and in this respect
was evaluated for its ability to interfere with the NMDA receptor
function.
towards which EBAB shares in common some structural analogies
(see above) influences also the activity of the NMDA receptor [16].
For all these reasons, EBAB was evaluated for its ability to control
features directly and indirectly related to the NMDA receptor.
3.4.1. Testing EBAB against NMDA-induced seizures
Whether EBAB might act, like polyamines, on the NMDA
receptor calcium channel complex was studied by evaluating the
protection given by EBAB against NMDA-induced seizures in mice.
EBAB at 49 mg/kg was found to increase the threshold for NMDA
seizures by more than 10%, and this finding prompted to evaluate
the protection given by this compound against seizures induced in
normally fed OF1 mice by 137 mg NMDA. In the absence of EBAB,
100% of mice given intraperitoneal NMDA (137 mg/kg) developed
fatal seizures. Administration of EBAB, 1 hr prior to giving 137 mg/
kg NMDA to animals, resulted in a protection against seizures, this
protective activity being referred to the absence of seizure devel-
opment in animals observed for a period of 4 hours following
NMDA administration and the absence of death within 48 hours.
3.4. Direct and indirect exploration of the activity of EBAB towards
the NMDA receptor
Both the magnesium deficiency [7] and the focal intracerebral
ibotenate-poisoning challenge [14,15] are characterized, among
other features, by an activation of the NMDA receptor. Polyamines
Table 2
Anticonvulsant and neurotoxicity screening data in mice dosed intraperitoneally with compounds.
Compounds
MES test
ScPtz test
Toxicity
30 min
4 h
30 min
15 min
1 h
4 h
30 min
15 min
1 h
4 h
1 30 mg
ꢁ
ꢁ
ꢁ
ꢁ
þþþ
þþ
2 (EBAB)
30 mg
100 mg
300 mg
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
þþþþ
ꢁ
þþþþ
þ
þþ
þþþ
þþþ
þþþþ
þþ
ꢁ
þþþþ
þþþþ
þþ
3 30 mg
100 mg
300 mg
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
þ
ꢁ
ꢁ
ꢁ
ꢁ
þþ
þþþþ
þþþþ
4 30 mg
100 mg
300 mg
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
þ
ꢁ
ꢁ
þþþþ
þþþþ
5 3 mg
10 mg
30 mg
100 mg
300 mg
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
þþþ
ꢁ
þþ
ꢁ
þþþþ
þþþþ
The anticonvulsant (MES and scPtz tests) and neurotoxicity activities were determined 30 min and 4 h (occasionally 15 min and 1 h) after the intraperitoneal administration of
compounds to mice. Symbols are as follows: þþþþ, activity in 75e100% of administered animals; þþþ, in 50e75% of animals; þþ, in 25e50% of animals; þ, 0e25% of
animals; and ꢁ, no activity or toxicity. Toxicity was determined by the rotorod test. Abbreviations are as follows: MES, maximal electroshock seizures; scPtz, subcutaneous
pentylenetetrazol.

J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
3105
Table 3
Effect of compounds on pyrogallol autooxidation rates.
Experimental conditions
Inhibition of pyrogallol
autooxidation rates
Vehicle
0%
Compound 1
Compound 2 (EBAB)
Compound 3
Compound 4
Compound 5
69.4%
36.3%
51.0%
17.5%
28.8%
<5%
4-aminobenzamidine
Pyrogallol autoxidation rates were measured at 420 nm as indicated in the exper-
imental procedure. The assay medium contained pH 7.4 TRIS saline buffer (5 mM
TRIS and 50 mM NaCl), 10 mL DMSO or 10 mL compound 50 mM stock solution
prepared in DMSO (final concentration of compound in the cuve being 0.5 mM).
Reaction was started by the addition of 0.75 mM pyrogallol. 4-Aminobenzamidine,
a chemical motif present in each compound formula was also tested.
Each result corresponds to the mean of three separate experimental determinations.
Inhibition of pyrogallol autooxidation was calculated using rates monitored during
the first two minutes of the reaction and are expressed as percentage values in the
table.
Fig. 2. Molecular modelling studies of low energy conformations of spermine and bis-
benzamidine compounds
1 to 5. Low energy conformations of compounds were
determined using the Chem 3D Ultra version 9.0.1 de Cambridgesoft. Carbon, nitrogen
and hydrogen atoms are represented by spheres coloured in grey, blue and white,
respectively. Red spheres refer to protonation of nitrogen atoms, and distances
between protonated nitrogens are indicated by a red line annotated in red characters.
Note that compound 2 (EBAB) and spermine adopt both a low energy conformation
which is linear and in which the distances between the protonated nitrogens placed at
each extremity of the molecules are very close, 16.09 and 16.13 Å, respectively. The two
compounds along with compound 1 adopt an extended conformation in contrast to
compounds 3, 4 and 5 which exhibit a folded conformation in which the distance
between the protonated nitrogens is lower (For interpretation of the references to
color in this figure legend, the reader is referred to the web version of this article.).
The ED50 exhibited by EBAB in the 137 mg/kg NMDA-induced
seizure test was found to be 6 mg/kg. Compound 4 for which no
activity was obtained in MDDAS, MES and scPtz tests, failed to
provide animals with protection against NMDA-induced seizures at
doses up to 50 mg/kg.
3.4.2. Effect of EBAB on locomotor activity and body temperature
Indirect evidence for a depression of the brain excitatory path-
ways was explored by determining the effect of EBAB on the
locomotor activity and body temperature. For locomotor activity,
mice were transferred individually in the actimeter and were
allowed to explore for a 2 min period. Their spontaneous locomotor
activity was then measured during 3 minutes. The number of
crossings of the photocell activity meter was 121.8 ꢀ 28.2 for
control (untreated) mice (n ¼ 10), and 13.3 ꢀ 2.2, 90.5 ꢀ10.7 or
103.3 ꢀ 20.4 for mice given 30, 10 and 5 mg/kg EBAB (1 hr before
testing), respectively. Mean rectal temperature was lowered from
38 ꢂC to 35.7 ꢂC (30 mg/kg EBAB). The EBAB 30 mg/kg dose
reduced spontaneous (actimeter test) but not forced (rotarod test)
locomotor activity, a differential effect lost by a small increase in the
administered dose (from 30 to 35 mg/kg) which altered both types
of locomotor activities (data not shown).
phencyclidine site of NMDA receptor and not that of sigma recep-
tors) to rat brain structures. The IC50 value determined for EBAB
was 1.4 10ꢁ6 M. The corresponding competition curve is shown in
Fig. 3.
3.4.4. Evaluation of EBAB in the hot plate test
The effect of EBAB on the phencyclidine site of NMDA receptor
might be accounted for by an effect on the polyamine site of this
receptor. Action on the last site (namely, antagonism of stimulatory
effects of polyamines on the NMDA receptor) is known for other
compounds to convey potent antinociceptive activity (for details,
see Discussion). So EBAB, given 45 min before testing, was evalu-
ated in the hot plate test. To correct for individual differences in
base-line latencies, the nociceptive data (latencies) were converted
to percentage maximum possible effect (% MPE) in accordance with
3.4.3. In vitro pharmacology of EBAB on the phencyclidine site of
the NMDA receptor
100
75
50
25
0
EBAB was evaluated for its ability to displace the binding of 3H-
TCP (a thienyl analogue of phencyclidine which targets the
1000
Cortical plate
(grey matter)
750
***
White matter
500
250
0
***
*
**
Ibo
leptin
Ibo
leptin
EBAB
EBAB
-25
-9
-8
-7
-6
-5
-4
-3
-2
-10
Fig. 1. Effect of EBAB on lesions induced by focal intracerebral injection of ibotenate to
pups. EBAB was given to CNS ibotenate-treated pups at the dose of 4 g/kg. Ibo
represents the reference conditions (animals given ibotenate plus the vehicle) and
leptin (0.1 mg/kg) is standard neuroprotective compound. *p < 0.05, **p < 0.01,
***p < 0.001 (ANOVA, Bonferroni test).
m
Log [EBAB] (M)
a
Fig. 3. Curve of competition by EBAB of binding of 3H-TCP to a rat cerebral prepara-
tion. EBAB IC50 ¼ 1.4 ꢃ10ꢁ6 M (nH ¼ 1.2).

3106
J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
Brady and Holtzman [17] (for calculation of %MPE values, see the
Materials and methods section). EBAB provided protection against
nociception in mice undergoing the hot plate test with % MPE at
30 mg and 60 mg/kg amounting to 31.17 ꢀ 14.52% and
70.33 ꢀ 28.91% (6 EBAB-treated versus 6 untreated mice for each
drug dosing group), respectively.
the residual discrepancy between the two evaluations has led us to
determine whether it could not be attributed to alternative
mechanisms, so-called neuroprotective properties adding to the
classic anti-seizure hits (NIH testing) observed for the selected
compound. For this reason, EBAB was evaluated (with success) for
its activity in a standard test of cerebral injury (ibotenate-poisoned
brain) classically employed for detecting neuroprotectants as well
as in antioxidant assays. As a general rule, neuroprotectants and/or
antioxidants even if devoid of intrinsic anti-seizure activity in
classic tests appear to be active against audiogenic seizures asso-
ciated with magnesium deficiency. Fig. 4 provides the reader with
an illustrated account of this particular facet of the MDDAS model
which combines neurotoxicity and seizure development.
The anti-NMDAergic activity of bis-benzamidines, here clearly
shown for the selected compound EBAB, might account for most
central nervous system (CNS) effects here reported. Magnesium-
deficiency has been recently shown to be associated with a lowered
threshold to NMDA-induced seizures [7], being consistent with
a sensitivity of audiogenic seizures to EBAB. Ibotenate is a cyclic
analogue of glutamate acting on glutamate receptors including
metabotropic and NMDA receptors but neither alpha-3-amino-
hydroxy-5-methyl-4-isoxazole (AMPA) nor kainate receptors [28].
Blunting the effect of ibotenate on the NMDA receptor by an anti-
NMDA activity might explain the protection offered by EBAB
against ibotenate poisoning. The activity of EBAB in the 6 Hz
seizure test is also logical in view of the large spectrum of anti-
convulsant mechanisms to which this model responds; anti-NMDA
activity (such as that conveyed by EBAB) representing a known
anticonvulsant mechanism as for instance inferred from the
previous use of MK-801 [29e31]. Finally, indirect evidence for
impact of EBAB on the NMDA receptor may have been also given by
a reduction in locomotor activity and body temperature, and in
pain. This antinociceptive effect of EBAB might be linked to
a putative action at polyamine sites of the NMDA receptor,
consistent with modulation of the binding at the phencyclidine site,
as previously incriminated for drugs such as ifenprodil which
antagonizes polyamine stimulatory sites located on NR2B-con-
taining NMDA receptors and binding of 3H-TCP to rat cerebral
membranes and which exhibits antinociceptive properties [32,33
and references therein].
4. Discussion
The amidine (eC(eNH2)]NH2þ) and benzamidine (-phenyl-C
(eNH2)]NHþ2 ) moieties represent the pharmacophore of several
pharmaceutical specialities such as propamidine (used to cure
infectious corneal ulcers caused by Acanthamoeba keratitis), pent-
amidine (clinically prescribed in the treatment of Pneumocystis
pneumonia, human african trypanosomiasis cause of sleeping
sickness, human and animal leishmaniasis), hexamidine, an
approved antiseptic found in many topical, ocular, and oral prep-
arations [18,19] and a prodrug of furamidine which is in phase III
development for the treatment of african trypanosomiasis [20].
Treatment of infectious diseases, including recent developments
[21e23], represents thus a major indication of amidines and ben-
zamidines. The present short series of bis-benzamidines were in
this line, displaying also anti-infectious properties. On the basis of
neuroprotective and anticonvulsant properties here documented,
medical indications of benzamidines might be potentially extended
to counteract diverse nutrition, chemical or disease-driven brain
injuries.
At a mechanistic point of view, pentamidine analogues were
previously shown to interact with the NMDA receptor calcium
channel complex [24]. More precisely, pentamidine analogues were
shown to interact with the spermine ligand binding site of this
receptor channel complex [24], acting via a mechanism referred to
as an inverse agonistic activity [25,26]. This is for instance the case
of polyamines flanked with aromatic substituents that provide
compounds with a more potent blocker activity towards NMDA
receptor calcium channel complex [26]. In addition to these works,
our data report in vivo efficiency of this class of compounds (ary-
lamidines) towards the NMDA receptor. Therefore, a breakthrough
of the present work is to demonstrate clearly that this type of
compounds is effective in vivo, i.e. on the whole animals,
indicating that compounds (or metabolites) are likely to permeate
bloodebrain barrier at substantial rates, a original feature which
should promote their future pharmacological development in the
field of disease-driven brain toxicity.
Exploration of NMDA receptor modulating properties of bis-
benzamidines, though a posteriori logical, has been, however, the
result of a methodical approach providing the MDDAS test with
a central place in the strategy developed to characterise brain
protective properties of our short series of compounds. In this
respect, our data have been presented grosso modo in the chrono-
logical order in which they were collected. In agreement with the
overall strategy appearing in the Introduction, MDDAS testing
(protection against audiogenic seizures) was used to detect
compounds of interest for drug development in the field of disease-
driven brain injuries, guiding the pursuit of evaluation to assess the
compound mode of action via additional experiments. They
included a previously described alternative use of the MDDAS
model (modulation by compound ED50 of phase durations) [4] and
the testing in classic animal seizure tests (MES and scPtz tests).
Because of the discrepancy between the potencies of compounds in
MDDAS and other (MES,scPtz) tests, the selected compound, EBAB,
was further evaluated in the 6 Hz test, a seizure test responsive to
a wide spectrum of anticonvulsant activities [27]. Potency of EBAB
in this test (ED50 value of 49 mg/kg) was here less in discrepancy
with the evaluation in the MDDAS test (ED50 ¼ 6 mg/kg). However,
5. Conclusion
EBAB is a novel compound active against various brain neuro-
insult models. Though other mechanisms may not be ruled out,
EBAB antagonizes the NMDA receptor/calcium channel complex,
via antagonism possibly of polyamine stimulatory sites and effec-
tively of 3H-TCP-binding site. This view is consistent with most of
the CNS effects reported here for this compound.
6. Materials and methods
6.1. Materials
Reagents and solvents were purchased from SigmaeAldrich
(Bornem, Belgium).
6.2. Analytical procedures
Melting points (uncorrected) were determined in open capillary
tubes on an Electrothermal apparatus. Carbons of the phenyl ring
on the linker nitrogen bearing substituent R1 (see Scheme 1) were
numbered from 1 to 6, carbon 1 bearing nitrile, amidine or ami-
doxime group. The 10 to 60 numbering was used similarly for the
other phenyl ring. IR spectra were recorded in the ATR mode on

J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
3107
Fig. 4. Current understanding of magnesium deficiency in mice as a model emphasizing neurotoxicity aspects of magnesium-deficient brain and their suitability for evaluation of
brain protective drugs. A. Normal diet supply in magnesium. B. Deficient diet supply in magnesium. a. The susceptibility to develop seizures is induced by magnesium deprivation
and is likely to result from the pro-oxidant and pro-inflammatory state associated with magnesium deficiency (note that disturbances of neurotransmission are concomitantly
induced). By removing this inflammatory/oxidative stress-driven susceptibility to audiogenic seizures, compounds endowed with antioxidant properties even if devoid of intrinsic
anti-seizure activity are capable of protecting magnesium-deficient mice against audiogenic seizures. b. The development of audiogenic seizures is triggered by a well defined
acoustic stimulus which here acts as a physicial convulsant stimulus like the chemical convulsant pentylenetetratzol and the electrical convulsant electroshock do in their respective
tests. Anti-seizure drugs protect mice against audiogenic seizures through classic inhibition of seizures. c e Compounds combining anti-seizure and antioxidant properties are
capable of interfering with both susceptibility to and development of audiogenic seizures.
a Perkin-Elmer Spectrum 100 spectrophotometer. NMR spectra
were measured at 25 ꢂC on compounds in solution in DMSO-d6 at
300 MHz for 1H on a BRUKER AMX-300 apparatus. Chemical shifts
are expressed relative to the resonance of (CD3)2SO at & 2.49 for 1H
NMR. All compounds had IR and 1H NMR spectra consistent with
their assigned structure. Their high resolution mass spectra were
recorded on a Waters QtoF 2 spectrometer.
6.3.1.2. 4,40-(N-methyl 1,2-ethanediyldiimino)bis-benzonitrile. Synth-
esis was performed with 1-amino-2-methylamino-ethane as the
diamine reagent; yield 71%; mp 150e152 ꢂC; 1H RMN (DMSO d6):
0
0
d
(ppm): 7,5 (2H, d, J ¼ 8 Hz, H2 (2 ) and H6 (6 )); 7,4 (2H, d, J ¼ 8 Hz,
0
0
0
H
2 (2 ) and0 H6 (6 )); 6,8 (1H, s large, NeH); 6,8 (2H, d, J ¼ 8 Hz, H3 (3 )
0
(30)
and H5 (5 )); 6,6 (2H, d, J ¼ 8 Hz, H3
and H5 (5 )); 3,6 (2H, t,
J ¼ 6 Hz, NHeCH2); 3,3 (2H, t, J ¼ 6 Hz, Met-N-CH2); 3,0 (3H, s,
CH3eNeCH2); IR (KBr):
n
(cmꢁ1): 3388 (NeH), 2214 (C^N), 1603,
6.3. Chemistry, intermediary and final compounds
1521, 1386, 1181; HRMS (ESI-ToF) MHþ C17H17N4: exp.: m/z
277,1453; Calc.: m/z 277,1453.
Compounds were synthesized according to the chemical
pathway depicted in Scheme 1. General comments of this chemical
pathway were given in the Chemistry section. Detailed aspects are
the following.
6.3.1.3. 4,40-[1,2-ethanediyl
(N,N0-dimethylbis-nitrilo)]bis-benzon-
itrile. Synthesis was performed with 1, 2-dimethylamino-ethane as
the diamine reagent; yield 68%; mp 204e206 ꢂC; 1H0 RMN (DMSO
0
d6):
d
(ppm): 7,5 (4H, d, J ¼ 8 Hz, H3, H5, H3 and H5 ); 6,7 (4H, d,
0
0
6.3.1. Synthesis of the bis-benzonitriles (products of the first step of
the chemical synthesis pathway, Scheme 1)
J ¼ 8 Hz, H2, H6, H2 and H6 ); 3,6 (4H, s, CH3eNeCH2); 2,9 (6H, s,
CH3eN); IR (KBr):
n
(cmꢁ1): 2208 (C^N), 1604, 1520, 1387, 1177;
A mixture of the appropriate diamine (25 mmol), 4-fluo-
robenzonitrile (55 mmol), and potassium carbonate (50 mmol) in
N,N-dimethylacetamide (25 ml) was heated under reflux for 9 h.
After cooling, the reaction mixture was poured into cold water, the
solid was filtered and successively washed with water and ethanol.
HRMS (ESI-ToF) MHþ C18H19N4: exp.: m/z 291,1613; Calc.: m/z
291,1610.
6.3.1.4. 4,40-(1,2-ethanediyldiimino)bis-(benzonitrile). Synthesis was
performed with 1, 2-dimethylamino-ethane as the diamine reagent;
yield 70%; mp 207e210 ꢂC; 1H RMN (DMSO d6):
d (ppm): 7,5 (4H, d,
0
0
6.3.1.1. 4,40-(N-ethyl 1,2-ethanediyldiimino)bis-benzonitrile. Synthesis
was performed with 1-amino-2-ethylamino-ethane as the diamine
J ¼ 9 Hz, H2, H6, H2 and H6 ); 6,8 (2H, s large, NeH); 6,6 (4H, d,
0
0
J ¼ 9 Hz, H3, H5, H3 and H5 ); 3,3 (4H, s, CH2eNH); IR (KBr):
n
(cmꢁ1):
reagent; yield 23%; mp 144e146 ꢂC; 1H RMN (DMSO d6):
d
(ppm): 7,6
336þ8 (NeH); 2211 (C^N); 1604, 1528, 1303, 1283; HRMS (ESI-ToF)
0
0
0
(2H, d, J ¼ 9 Hz, H2 (2 ) and H6 (6 )); 7,5 (2H, d, J ¼ 9 Hz, H2 (2 ) and H6
MH C16H15N4: exp.: m/z 263,1296; Calc.: m/z 263,1297.
0
0
0
(6 )); 6,8 (1H, s large, N-H); 6,8 (2H, d, J ¼ 9 Hz, H3 (3 ) and H5(5 )); 6,5
0
0
(2H, d, J ¼ 9 Hz, H3 (3 ) and H5 (5 )); 3,5 (2H, t, J ¼ 6 Hz, NH-CH2); 3,4
(2H, q, J ¼ 7 Hz, NeCH2eCH3); 3,3 (2H, t, J ¼ 6 Hz, Et-N-CH2); 1,1 (3H,
6.3.2. Synthesis of diamidines via the Pinner’s reaction (second step
of the chemical synthesis pathway relative to the asymmetric
diamidine molecules, Scheme 1 and its legend)
A mixture of the appropriate bis-benzonitrile (3.4 mmol) and
methanol (20 ml) in dichloromethane (200 ml) was saturated with
t, J ¼ 7 Hz, CH3eCH2); IR (KBr):
n
(cmþꢁ1): 3380 (NeH), 2211 (C^N),
1602, 1522, 1178; HRMS (ESI-ToF) MH C18H19N4: experimental: m/z
291,1600; calculated: m/z 291,1610.

3108
J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
gaseous hydrochloric acid. After 4 days at room temperature, the
solvents were evaporated under reduced pressure and the solid
was washed with ether. The imidate was then treated with
a methanolic solution of ammonia (7 N, 15 ml) and the mixture was
heated under reflux for 1 h. After cooling the precipitate was
filtered and washed with alcohol.
4,40-(1,2-ethanediyldiimino)bis-(N-hydroxybenzenecarboximida-
mide) e Synthesis was performed with 4,40-(1,2-ethanediyldiimino)
bis-(benzonitrile) as the bis-benzonitrile reagent; yield 87%; mp
206e209 ꢂC; 1H RMN (DMSO d6) : 9,3 (2H, s, OeH) ; 7,5 (4H,0 d,
0
0
0
J ¼ 9 Hz, H2, H6, H2 and H6 ) ; 6,6 (4H, d, J ¼ 9 Hz, H3, H5, H3 and H5 ) ;
5,9 (2H, s, NeH) ; 5,6 (4H, s, eNH2) ; 3,2 (4H, s, NHeCH2); IR (KBr) :
n
(cmþꢁ1) : 3489 (NH2), 3379 (NH2), 1661, 1608, 1529; HRMS (ESI-ToF)
6.3.2.1. 4,40-(N-ethyl1,2-ethanediyldiimino)bis-(benzene-carboximi-
damide) dihydrochloride. Synthesis was performed with 4,40-(N-
ethyl 1,2-ethanediyldiimino) bis-benzonitrile as the bis-benzonitrile
MH C16H21N6O2 : exp.: m/z 329,1728; Calc.: m/z 329,1732.
6.3.3.4. Reduction of amidoximes (diamidine formation). 4,40-[1,2-
ethanediyl(N,N0-dimethylbis-nitrilo)] bis-(benzenecarboximidamide)
dihydrochloride e Synthesis was performed by reduction of 4,40-[1,2-
ethanediyl (N,N0-dimethyl bis-nitrilo)]bis-(N-hydroxy benzene car-
boximidamide) as the amidoxime reagent; yield 68%; mp > 300 ꢂC; 1H
reagent; yield 33%; mp > 300 ꢂC; 1H RMN (DMSO d6):
d (ppm): 9,1
(2H, s, NeH amidine); 8,9 (2H, s, NeH amidine); 8,8 (2H, s, NeH
0
amidine); 8,7 (2H, s, NeH amidine); 7,8 (2H, d, J ¼ 9 Hz, H2 (2 ) and H6
(60)
2 (20)
6 (60)
); 7,7 (2H, d, J ¼ 9 Hz, H
and H
); 7,1 (1H, t large, N-H); 6,8
0
0
0
(2H, d, J ¼ 9 Hz, H3 (3 ) and H5 (5 )); 6,7 (2H, d, J ¼ 9 Hz, H5 (5 ) and H3
RMN (DMSO d6) : d (ppm) : 8,9 (2H, s, N-H amidine) ; 8,6 (4H, s, N-H
0
0
0
(3 )); 3,6 (2H, t, J ¼ 8 Hz, CH2eNH); 3,5 (2H, q, J ¼ 7 Hz, CH3eCH2eN);
amidine) ; 7,7 (4H, d, J ¼ 8 Hz, H2, H6, H2 and H6 ) ; 6,8 (4H, d, J ¼ 8 Hz,
0
0
3,4 (2H, t, J ¼ 8 Hz, Et-N-CH2); 1,1 (3H, t, J ¼ 7 Hz, CH3eCH2); IR (KBr):
H3, H5, H3 and H5 ) ; 3,7 (4H, s, CH3-N-CH2) ; 3,0 (6H, s, CH3eNeCH2);
n
(cmꢁ1): 3317, 3136, 1663, 1605, 1494; HRMS (ESI-ToF) MHþ
IR (KBr) : n
(cmꢁ1) : 3307, 3137, 1664, 1609, 1485; HRMS (ESI-ToF) MHþ
C18H25N6 exp.: m/z 325,2136; Calc.: m/z 325,2141.
C18H25N6 : exp./m/z 325,2133; Calc.: m/z 325,2141.
4,40-(1,2-ethanediyldiimino)bis-benzenecarboximidamide dihy-
drochloride e Synthesis was performed by reduction of 4,40-(1,2-
ethanediyldiimino)bis-(N-hydroxybenzenecarboximidamide) as
the amidoxime reagent; yield 65%; mp > 300 ꢂC; 1H RMN (DMSO
6.3.2.2. 4,40-(N-methyl-1,2-ethanediyldiimino)bis-(benzene-carbox-
imidamide) dihydrochloride. Synthesis was performed with 4,40-(N-
methyl 1,2-ethanediyldiimino)bis-benzonitrile as the bis-benzoni-
trile reagent; yield 36%; mp > 300 ꢂC; 1H RMN (DMSO d6) :
d
(ppm)
d6) : d (ppm) : 8,9 (2H, s, NeH amidine) ; 8,7 (4H, s, NeH ami-
0
0
: 9,1 (2H, s, NeH amidine) ; 8,9 (2H, s, NeH amidine) ; 8,8 (2H, s, N-
dine) ; 7,7 (4H, d, J ¼ 8 Hz, H2, H6, H2 0and H6 ) ; 7,1 (2H, s, NeH) ;
0
0
H amidine) ; 8,7 (2H, s, N-H amidine) ; 7,8 (2H, d, J ¼ 9 Hz, H2 (2 ) and
6,8 (4H, d, J ¼ 8 Hz, H3, H5, H3 and H5 ) ; 3,4 (4H, s, CH3eNeCH2);
0
0
0
H
6 (6 )) ; 7,7 (2H, d, J ¼ 9 Hz, H2 (2 ) and H6 (6 )) ; 7,1 (1H, t large, N-H) ;
IR (KBr) : n
(cmꢁ1) : 3296, 3211, 3122, 1657,1609, 1490, 1460;
0
0
0
6,8 (2H, d, J ¼ 9 Hz, H3 (3 ) and H5 (5 )) ; 6,7 (2H, d, J ¼ 9 Hz, H5 (5 ) and
HRMS (ESI-ToF) MHþ C16H21N6 : exp.: m/z 297,1829 ; Calc.: m/z
297,1828.
0
H
3 (3 )) ; 3,6 (2H, t, J ¼ 5 Hz, CH2eNH) ; 3,4 (2H, t, J ¼ 5 Hz, Et-N-CH2)
; 3,0 (3H, s, CH3eNeCH2); IR (KBr) :
n
(cmꢁ1) : 3329, 3143, 1661,
1608, 1495; HRMS (ESI-ToF) MHþ C17H23N6 : exp.: m/z 311,1970;
Calc.: m/z 311,1984.
6.4. In vivo and in vitro experiments
6.4.1. Preparation and mode of administration of compounds
All tested compounds were administered via the intraperitoneal
route. Compounds studied in the audiogenic seizure model were
6.3.3. Synthesis of diamidines via the formation followed by
reduction of amidoximes (second step of the chemical synthesis
pathway relative to the symmetric diamidine molecules, Scheme 1
and its legend)
6.3.3.1. Formation of amidoximes. A mixture of hydroxylamine
hydrochloride (40 mmol) and a methanolic solution of sodium
methoxide (30%, 40 mmol) in methanol (25 ml) was heated under
reflux for 30 min. After cooling the precipitate was filtered and the
filtrate was transferred into a round-bottom flask. The appropriate
bis-benzontrile (4 mmol) was added and the mixture was heated
under reflux for 8 h. After cooling the precipitate was filtered and
washed with methanol.
dissolved in a 10 ml dimethylsufoxide (DMSO)/10 ml polyethylene
glycol 300 mixture and administered 1 h before testing. For the
ibotenate poisoning experiments and in vitro assays, compounds
were dissolved in DMSO. Compounds were dissolved in methyl-
cellulose for evaluation in the MES, scPtz and 6 Hz seizure test at
different time-points ranging from 15 min to 4 h.
6.4.2. In vivo evaluations on the animals
6.4.2.1. Audiogenic seizure susceptibility. This was induced in OF1
mice by magnesium deprivation in the diet as described previously
[4], except that 35 ppm instead of 50 ppm magnesium content of
the diet was used in order to induce the susceptibility in 100%
animals in 28 days instead of 42 days in this mouse strain. The
magnesium deficiency dependent audiogenic seizure (MDDAS) was
triggered by an acoustic stimulus. It was studied as an animal
model for yes/no anticonvulsant protection screening, for deter-
mination of the efficacious dose ED50 (compound dose preventing
seizures in 50% tested animals) and its subsequent impact on the
duration of the phases of audiogenic seizures in convulsing mice
(latency plus wild running, seizure and recovery). Essentially for
the audiogenic seizure testing [4], individual animals were placed
in a 9-dm3 volume test chamber (30, 20 and 15 cm for length, width
and height, respectively) and exposed for 15 s to an auditory signal
of 10 ꢀ 0.1 kHz frequency and 100 ꢀ 1 db intensity. This acoustic
signal was produced by a signal generator and projected via a high
frequency speaker mounted on the roof of the chamber. The noise
level was measured close to the animal’s ear by an external decibel-
meter probe. Each audiogenic test used only one test chamber and
each animal was subjected to only one audiogenic test during the
experimentation. Tested animals were OF1 mice [3-month-old
6.3.3.2. Reduction of amidoximes (diamidine formation). The so-
obtained diamidoxime (2.2 mmol) was dissolved in acetic acid
(15 ml). Ammonium formate (25 mmol) and Pd/C 10% (0.2 g) were
successively added. The mixture was heated under reflux for 2 h.
After cooling the mixture was filtered on CeliteÒ and the filtrate was
concentrated under reduced pressure. The solid was separated,
poured into dichloromethane and precipitated as the dihydro-
chloride salt by addition of gaseous hydrochloric acid.
6.3.3.3. Formation of amidoximes. 4,40-[1,2-ethanediyl (N,N0-
dimethyl bis-nitrilo)]bis-(N-hydroxy benzenecarboximidamide)
e
Synthesis was performed with 4,40-[1,2-ethanediyl (N,N0-dimethyl
bis-nitrilo)]bis-benzonitrile as the bis-benzonitrile reagent; yield
90%; mp 198-202 ꢂC; 1H RMN (DMSO d6): 9,3 (2H, s, OeH); 7,6 0(4H,
0
0
d, J ¼ 9 Hz, H2, H6, H2 and H6 ); 6,7 (4H, d, J ¼ 9 Hz, H3, H5, H3 and
0
H5 ); 5,7 (4H, s, eNH2); 3,6 (4H, s, CH3eNeCH2); 2,9 (6H, s, CH3eN);
IR (KBr) :
n
(cmꢁ1) : 3496 (NH2), 3374 (NH2), 1665, 1609, 1527;
HRMS (ESI-ToF) MHþ C18H25N6O2 : exp.: m/z 357,2033; calc. : m/z
357,2039.

J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
3109
(adult) mice] obtained from Janvier (Le Genest Saint Isle,France)
and were randomly divided into groups of 20 mice per cage.
individual differences in base-line latencies, the nociceptive data
(latencies) were converted to percentage maximum possible
effect (% MPE) using the following formula [17]: % MPE ¼ 100 x
[(postdrug latency - predrug latency) divided by (maximum
latency -predrug latency)], maximum latencies (100%) being equal
to the maximal time of observance, i.e. 30 s.
6.4.2.2. Focal intracerebral injection of ibotenate. In Swiss mouse
pups and determination of the subsequent lesions in white and
grey matters were performed with or without concomitant drug
administration as previously described [14,15,28]. In these studies,
Swiss mouse pups were anesthetized by isoflurane inhalation for
intracerebral and intraperitoneal injections. Essentially for the
6.4.2.7. Ethics. These experimental protocols and procedures
complied with the European Communities Council Directives of 24
November 1996 (86/609/EEC).
stereotaxic intracerebral injections, 10
mg of ibotenate (Sigma, St.
Louis, MO) were administered with a 26-gauge needle on a 50
ml
Hamilton syringe mounted on a calibrated microdispenser. The
needle was inserted 2 mm under the external surface of scalp skin
in the frontoparietal area of the right hemisphere, 2 mm from the
midline in the lateral-medial plane and 3 mm (in the rostro-caudal
plane) from the junction between the sagittal and lambdoid
6.4.3. In vitro evaluations
6.4.3.1. Superoxide dismutase-like activity of compounds. refers to
their ability to scavenge the superoxide anion radical and was
measured as the capacity of compounds to inhibit pyrogallol
autooxidation, a process which involves superoxide anion radical
as an intermediate. Pyrogallol autooxidation rates were monitored
at 420 nm according to the procedure of Marklund and Marklund
[36] as described previously [6]. Incubation mixture included
saline pH 7.4 TRIS buffer (5 mM TRIS and 50 mM NaCl in final
concentrations) with the vehicule DMSO (blank conditions) or
with the compound dissolved in this vehicule (test conditions)
before the reaction was started by the addition of 0.75 mM
pyrogallol.
sutures. Two 1
was left in place for an additional 30 s. Compound 2 (EBAB) powder
were diluted in DMSO and injected i.p. at the dose of 4 g/kg.
ml boluses were injected at 30-s intervals. The needle
m
Controls received i.p. DMSO. Five days later, the surviving pups
were killed by decapitation and the brains were fixed in formalin.
Paraffin normal serial sections, 15 mm thick, were cut and every
third section was stained with cresyl-violet. These serial sections of
the entire brain were made in the coronal plane. As explained
previously, this permitted an accurate and reproducible determi-
nation of the maximal diameter of the lesion in the sagittal fronto-
occipital axis (which is equal to the number of sections where the
6.4.3.2. In vitro pharmacology: binding assay. The binding assay
using 3H-TCP as a ligand and rat cerebral cortex as a receptor/site
source was adapted from the procedure described by Vignon et al.
[37] and was performed by Cerep, an independent contract labo-
ratory (Cerep, Bois l’Evêque, France), using 5 nM tritiated ligand
and a 60 min incubation at 22 ꢂC, in the absence or in the presence
of various concentrations of the selected benzamidine (EBAB). The
analysis and expression of the binding assay results were as follows.
The specific ligand binding to the receptors was defined as the
difference between the total binding and the nonspecific binding
determined in the presence of an excess of unlabelled ligand. The
results were expressed as a percent of control specific binding
lesion was present multiplied by 15
mm). We used this measure as
an index of the volume of the lesion.
6.4.2.3. Electrical and chemical seizure tests. This included the
classic MES (50 Hz) and scPtz tests of the NIH [34] as we reported
previously [4]. The induction of seizures by low frequency electrical
stimulation, using the 6 Hz seizure model, was performed as
described by Kaminski et al. [27]. NMDA-induced seizures were
induced as previously indicated [7].
6.4.2.4. Minimal acute neurotoxicity. In adult mice was determined
by the rotarod procedure in untreated animals and in treated
animals at different time points after the tested compound was
administered. The mouse was placed on a 1 inch diameter knurled
plastic rod rotating at 6 rpm. Normally, unimpaired mice can easily
remain on a rod rotating at this speed. Minimal acute neurotoxicity
refers to neurological deficit (e.g. ataxia, sedation, hyperexcitability)
indicated by the inability of the animal to maintain equilibrium on
the rod for at least 1 min, in each of three concurrent trials.
((measured specific binding/control specific binding)
obtained in the presence of EBAB.
x 100)
The IC50 values (concentration causing a half-maximal inhibi-
tion of control specific binding) and Hill coefficients (nH) were
determined by non-linear regression analysis of the competition
curves generated with mean replicate values using Hill equation
curve fitting (Y ¼ D þ [(A ꢁ D)/(1 þ (C/C50
)
nH)], where Y ¼ specific
binding, D ¼ minimum specific binding, A ¼ maximum specific
binding, C ¼ compound concentration, C50 ¼ IC50, and nH ¼ slope
factor). This analysis was performed using a software developed at
Cerep (Hill software) and validated by comparison with data
generated by the commercial software SigmaPlotÒ 4.0 for Win-
dowsÒ (Ó 1997 by SPSS Inc.). The inhibition constants (Ki) were
calculated using the Cheng Prusoff equation (Ki ¼ IC50/(1 þ (L/KD)),
where L ¼ concentration of radioligand in the assay, and
KD ¼ affinity of the radioligand for the receptor).
6.4.2.5. Locomotor activity. Mice were transferred individually in
an Apelex type 01-1668B actimeter (Bagneux, France). They were
allowed to explore for a 2 min period. Their spontaneous locomotor
activity was measured for another 3 minutes by the crossing of the
photocell activity meter and automatically recorded. The experi-
ment was carried out in a sound proof room between 9:00 and
13:00 to reduce the confounding influence of diurnal variation in
motility.
6.4.4. Molecular modelling studies
Low energy conformations of compounds were determined
using the Chem 3D Ultra version 9.0.1 of Cambridgesoft.
6.4.2.6. Hot plate test. The degree of antinociception induced by
EBAB (administered i.p. 45 min before testing) was quantified
using the hot plate latency test [35]. The animals (OF1 mice) were
placed on a hot plate maintained at a temperature 55 ꢀ1 ꢂC. The
latency for the appearance of pain reflexes (vocalization, paw lick,
hindpaw flick or jump) in response to the thermal stimulus, were
measured and the animals immediately removed from the hot
plate, the maximal time of observance being fixed to 30 sec in
order to prevent tissue damage to the mouse’s paws. To correct for
6.5. Statistical analysis
Quantitative data were expressed as the means ꢀ S.E.Ms for
each treatment group. Means were compared using Student’s t test
or ANOVA with Dunnett’s or NewmaneKeul’s multiple comparison
of the means test (GraphPad Prism version 3.03 for Windows;

3110
J. Vamecq et al. / European Journal of Medicinal Chemistry 45 (2010) 3101e3110
[11] B. Tao, T.L. Huang, Q. Zhang, L. Jackson, S.F. Queener, I.O. Donkor, Eur. J. Med.
Chem. 34 (1999) 531e538.
[12] E. Dicou, S. Attoub, P. Gressens, Neuroreport 12 (2001) 3947e3951.
[13] A.P. Signore, F. Zhang, Z. Weng, Y.Q. Gao, J. Chen, J. Neurochem. 106 (2008)
1977e1990.
[14] S. Marret, R. Mukendi, J.F. Gadisseux, P. Gressens, P. Evrard, J. Neuropathol,
Exp. Neurol. 54 (1995) 358e370.
[15] P. Gressens, S. Marret, J.M. Hill, D.E. Brenneman, I. Gozes, I. Fridkin, P. Evrard, J.
Clin. Invest. 100 (1997) 390e397.
GraphPad Software Inc, San Diego, CA). For hot plate experiments,
results were means ꢀ SD calculated using Microsoft ExcelÒ.
Acknowledgments
DS was sponsored by the Belgian F.R.I.A. (Fonds pour la forma-
tion de la Recherche dans l’Industrie et l’Agriculture).
[16] M. Benveniste, M.L. Mayer, J. Physiol. 464 (1993) 313e363.
[17] L.S. Brady, S.G. Holtzman, J. Pharmacol. Exp. Ther. 222 (1982) 190e197.
[18] M.N.C. Soeiro, E.M. De Souza, C.E. Stephens, D.W. Boykin, Expert Opin.
Investig. Drugs 14 (2005) 957e972.
[19] K. Werbovetz, Curr. Opin. Invest. Drugs 7 (2006) 147e157.
[20] W.D. Wilson, B. Nguyen, F.A. Tanious, A. Mathis, J.E. Hall, C.E. Stephens, D.
W. Boykin, Curr. Med. Chem. 5 (2005) 389e408.
[21] M.T. Cushion, P.D. Walzer, A. Ashbaugh, S. Rebholz, R. Brubaker, J.J. Vanden
Eynde, A. Mayence, T.L. Huang, Antimicrob. Agents Chemother. 50 (2006)
2337e2343.
[22] T.L. Huang, C.J. Bacchi, N.R. Kode, Q. Zhang, G. Wang, N. Yartlet, D. Rattendi,
I. Londono, L. Mazumder, J.J. Vanden Eynde, A. Mayence, I.O. Donkor, J. Anti-
microb. Agents 30 (2007) 555e561.
No conflict of interest was associated with the present study.
DS and JJVE contributed to the design and chemical synthesis of
compounds. PM, NP, PB and JPS contributed to the evaluation of
compounds in the various animal seizure tests. PM and NP also
performed the measurements relative to the locomotor activity,
body temperature and hot plate test. PG investigated the effects of
EBAB on a neonatal model of focally induced ibotenate poisoning of
immature brain. JV performed biochemical assays. JV and JJVE
contributed to the supervision and drafting of the manuscript.
[23] T.L. Huang, J.J. Vanden Eynde, A. Mayence, I.O. Donkor, S.I. Khan, B.L. Tekwani,
J. Pharmacy Pharmacol. 58 (2006) 1033e1042.
References
[24] I.O. Donkora, M.L. Berger, Bioorg. Med. Chem. Lett. 11 (1997) 1455e1460.
[25] M.L. Berger, C.R. Noe, in: U. Ramchandran (Ed.), Current Topics in Medicinal
Chemistry; Research Trends: Poojapura, vol. 3, 2003, pp. 51e64.
[26] M.L. Berger, A.Y. Bitar, M.J. Waitner, P. Rebernika, M.C. O’Sullivan, Bioorg. Med.
Chem. Lett. 16 (2006) 2837e2841.
[27] R.M. Kaminski, M.R. Livingood, M. Rogawski, Epilepsia 45 (2004) 864e867.
[28] P. Gressens, L. Schwendimann, I. Husson, G. Sarkosy, E. Mocaer, J. Vamecq,
M. Spedding, Eur. J. Pharmacol. (2008). doi:10.1016/j.ejphar.2008.04.016.
[29] G.G.C. Hwa, M. Avoli, Neurosci. Lett. 98 (1989) 189e193.
[30] J.O. McNamara, R.D. Russell, D.W. Bonhaus, Neuropharmacology 27 (1988)
563e568.
[1] J. De Keyser, G. Sulter, P.G. Luiten, Trends Neurosci. 22 (1999) 535e540.
[2] A.R. Young, C. Ali, A. Duretête, D. Vivien, J. Neurochem. 103 (2007)
1302e1309.
[3] R. Cecchelli, B. Dehouck, L. Descamps, L. Fenart, V. Buée-Scherrer, C. Duhem,
S. Lundquist, M. Rentfel, G. Torpier, M.P. Dehouck, Adv. Drug Delivery Rev. 36
(1999) 165e178.
[4] P. Bac, P. Maurois, C. Dupont, N. Pages, J.P. Stables, P. Gressens, P. Evrard,
J. Vamecq, J. Neurosci. 18 (1998) 4363e4373.
[5] P. Maurois, S. Rocchi, N. Pages, P. Bac, J.P. Stables, P. Gressens, J. Vamecq,
Biomed. Pharmacother. 62 (2008) 259e263.
[31] K. Sato, K. Morimoto, M. Okamoto, Brain Res. 463 (1988) 12e20.
[32] M. Bernardi, A. Bertolini, K. Szczawinska, S. Genedani, Eur. J. Pharmacol. 298
(1996) 51e55.
[33] L.L. Coughenour, J.J. Cordon, J. Pharmacol. Exp. Ther. 280 (1997) 584e592.
[34] J.P. Stables, H.J. Kupferberg, in: G. Avanzini, G. Regestra, P. Tanganelli, M. Avoli
(Eds.), Molecular and Cellular Targets for Anti-Epileptic Drugs, John Libbey
Eurotext, London, 1997, pp. pp191epp198 Chapter 16.
[6] J. Vamecq, P. Maurois, P. Bac, F. Bailly, J.L. Bernier, J.P. Stables, I. Husson,
P. Gressens, Eur. J. Neurosci. 18 (2003) 1110e1120.
[7] P. Maurois, N. Pages, P. Bac, M. German-Fattal, G. Agnani, B. Delplanque,
J. Durlach, J. Poupaert, J. Vamecq, Br. J. Nutr. 101 (2009) 317e321.
[8] W.T. Festuccia, S. Oztezcan, M. Laplante, M. Berthiaume, C. Michel, S. Dohgu, R.
G. Denis, M.N. Brito, N.A. Brito, D.S. Miller, W.A. Banks, T.J. Bartness, D. Richard,
Y. Deshaies, Endocrinology 149 (2008) 2121e2130.
[35] N.B. Eddy, D. Leimbach, Exp. Ther. 107 (1953) 385e393.
[36] S. Marklund, G. Marklund, Eur. J. Biochem. 47 (1974) 469e474.
[37] J. Vignon, A. Privat, I. Chaudieu, A. Thierry, J.M. Kamenka, R. Chicheportiche,
Brain Res. 378 (1986) 133e141.
[9] S. Fourcade, S. Savary, S. Albet, D. Gauthé, C. Gondcaille, T. Pineau, J. Bellenger,
M. Bentejac, A. Holzinger, J. Berger, M. Bugaut, Eur. J. Biochem. 268 (2001)
3490e3500.
[10] S.S. Berg, J. Chem. Soc. (1960) 5172e5176. doi:10.1039/JR960000 5172.