Evaluation of anticonvulsant and analgesic effects of benzyl- and benzhydryl ureides

Article abstract of DOI:10.1016/j.ejphar.2006.12.002

The anticonvulsant effects of benzyl- and benzhydryl ureides in mice models of seizures (maximal electroshock seizure test, pentylenetetrazol test, picrotoxin-induced seizure test) and the influence on spontaneous locomotor activity has been assessed. Fur

Full text of DOI:10.1016/j.ejphar.2006.12.002

European Journal of Pharmacology 559 (2007) 138–149
www.elsevier.com/locate/ejphar
Evaluation of anticonvulsant and analgesic effects
of benzyl- and benzhydryl ureides
a
b
b
Tadeusz Librowski a, , Monika Kubacka , Manuela Meusel , Silvia Scolari ,
⁎
Christa E. Müller b, Michael Gütschow b,
⁎
a
Department of Pharmacodynamics, Medical College, Jagiellonian University, Medyczna 9, 30-688 Kraków, Poland
b
Pharmaceutical Institute, Pharmaceutical Chemistry I, University of Bonn, An der Immenburg 4, 53121 Bonn, Germany
Received 6 June 2006; received in revised form 30 November 2006; accepted 6 December 2006
Available online 12 December 2006
Abstract
The anticonvulsant effects of benzyl- and benzhydryl ureides in mice models of seizures (maximal electroshock seizure test, pentylenetetrazol test,
picrotoxin-induced seizure test) and the influence on spontaneous locomotor activity has been assessed. Furthermore, the analgesic effect of ureide
derivatives was studied in the hot-plate test in mice. Selected compounds were investigated for their in vitro interaction with adenosine receptors as
well as the benzodiazepine binding site of GABAA receptors. This study demonstrated the strong anticonvulsant activity of several ureides in
electrically or chemically induced seizure models, and structure-activity relationships were discussed. 1-Benzyl-3-butyrylurea (9) was found to be
equipotent to ethosuximide in the pentylenetetrazol test with regard to the number of attacks as well as the time of the onset of seizures. The
ureide 9 also revealed the highest protective activity against seizures in the other models, maximal electroshock seizure and picrotoxin test. Moreover,
1-benzyl-3-butyrylurea was not neurotoxic at doses up to 200 mg/kg. Benzylureides 8–10 showed affinity to the adenosine A1 receptors at low
micromolar concentrations. However, the apparent anticonvulsant activity in different seizure models does not appear to result from direct activation
of adenosine A1 receptors or GABAA receptors, respectively. In the hot-plate test, the majority of investigated compounds exhibited analgesic activity.
Again, compound 9 was superior to the other substances investigated, suggesting a potential therapeutic value of that ureide derivative.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Ureides; Anticonvulsant activity; Analgesic effect; GABA; Adenosine receptors; Benzodiazepine binding site
1. Introduction
(Rogawski and Löscher, 2004a), the involvement of the various
adenosine receptor subtypes and their ligands in seizure
disorders is still an object of discussion (Drabczyñska et al.,
2004). Available antiepileptic drugs are effective in only 60–
80% of epileptic patients and a number of limitations of the
antiepileptic drug therapy continue to exist (Baulac, 2003).
Therefore, the search for new anticonvulsant drugs is an
ongoing area of investigation in medicinal chemistry (Beghi and
Perucca, 1995). The present study is part of our common efforts
(Mendyk et al., 2001; Jastrzębska-Więsek et al., 2003; Kiec-
Kononowicz et al., 2002, 2004) in the development and
synthesis of new potent anticonvulsants. The structure-activity
relationships of the established antiepileptic drugs such as
phenytoin, carbamazepine or phenobarbital led to the proposal
of a general model for anticonvulsant activity comprising two
aromatic rings usually bound to an ureide ring system. The
ureide (N-acylurea) motif is a common structure in acyclic and
Epilepsy is one of the most frequent neurological disorders.
The basic cause of seizures is not clearly understood and the
different biochemical aspects of epileptic seizures still remain
unknown. Antiepileptic drugs exert their action by different
mechanisms. Mainly they can influence inhibitory (GABA) or
excitatory neurotransmitter (glutamic acid) systems and thus ion
transports across cell membranes. While interactions of
antiepileptic drugs with the different allosteric binding sites of
the GABAA receptor and the resulting enhancement of
inhibitory GABAergic effects is a long known mechanism
⁎
Corresponding authors. Librowski is to be contacted at Tel./fax: +48 12
6588219. Gütschow, Tel.: +49 228 732317; fax: +49 228 732567.
E-mail addresses: mflibrow@cyf-kr.edu.pl (T. Librowski),
guetschow@uni-bonn.de (M. Gütschow).
0014-2999/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejphar.2006.12.002

T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
139
heterocyclic compounds, such as barbituric acids and hydan-
toins. There are a number of publications describing the
anticonvulsant activities of N-, N,N- and N,N′-substituted
acylureas (Spielman et al., 1948; Frommel and Radouco-
Thomas, 1953; Beasley et al., 1961; Kulev and Dobychina,
1963; Polezhaeva, 1966; Carraz and Emin, 1967; Zirvi and
Fakouhi, 1982; Khalil and Weaver, 1990; Sobol et al., 2004).
Beside their anticonvulsant activity, many of those substances
show antiarrhythmic effects (Meusel and Gütschow, 2004) as
well as central depressant (Mrongovius, 1975), sedative (Weiss,
1971; Mrongovius et al., 1976), antianxiety (Goldstein and
Pfeiffer, 1971), local anesthetic (Chiti, 1960), or antimycotic
activities (Baraldi et al., 1989). The acylurea moiety can also be
found in several other drugs modulating their pharmacodynam-
ic and -kinetic properties (e.g. azlocillin (Graber et al., 1981)).
Moreover, an analgesic effect was described for some acylureas
(Arbuzov et al., 1989). Therefore, the analgesic potential of the
compounds of this study was also investigated. Herein we
report the results of the examination of a series of partly novel
benzyl- and benzhydryl ureides for their analgesic and
anticonvulsant activities in vivo. Anticonvulsant activity was
determined in the following mice models: picrotoxin-induced
convulsions, pentylenetetrazol-induced convulsions and maxi-
mal electroshock model. In addition, effects on spontaneous
locomotor activity and neurotoxicity were evaluated.
et al., 2002). Voltage-gated calcium channel blockers (gaba-
pentin and ethosuximide) and voltage-gated sodium channel
blockers (phenytoin and carbamazepine) have been shown to be
potent local analgesics to ameliorate acute thermal nociception.
The activity of locally applied phenytoin and carbamazepine
was substantially higher than that of the sodium channel-
blocking local anaesthetic lidocain (Todorovic et al., 2003).
Both sodium channel-blocking antiepileptic drugs are effective
by virtue of the same selective block of high-frequency action-
potential firing that accounts for their protective activity against
seizures (Rogawski and Loscher, 2004b).
2. Materials and methods
2.1. Animals
The studies were carried out on male Albino Swiss mice
(18–24 g). The animals were kept in groups of 15 mice in type
III-1290 cages (26.5×42.0×15.0 cm) at a room temperature of
22 2 °C, under 12/12 h light/dark cycle (light on from 7 a.m. to
7 p.m.), and had free access to water and food (standard
laboratory pellets; Bacutil, Motycz, Poland) before the experi-
ments. Each experimental group consisted of 10 animals/dose
(unless stated otherwise), and all the animals were used only
once. The experiments were performed between 8 a.m. and 3 p.m.
Treatment of laboratory animals in the present study was in
full accordance with the respective Polish and European regula-
tions. All procedures were conducted according to Animal Care
and Use Committee guidelines, and approved by the Ethical
Committee of Jagiellonian University, Kraków.
Furthermore, to characterize their anticonvulsant properties,
we investigated selected compounds in vitro for their interaction
with potential molecular targets, i.e. adenosine receptors and
GABAA receptors. Such an investigation might also elucidate
the analgesic activity of the compounds, as it has been
documented that both GABAA (Mui et al., 1997) and adenosine
receptors (Sawynok and Liu, 2003; Abo-Salem et al., 2004)
participate in antinociception.
2.2. Compounds
A link between antiepileptic activity and antinociceptive
effects was proven for many established anticonvulsant drugs
which have long been used in pain management, particularly in
chronic neuropathic pain. For example, carbamazepine, phe-
nytoin, lamotrigine and felbamate are clinically effective in the
treatment of trigeminal neuralgia, valproate and topiramate are
useful for migraine prophylaxis (Rogawski and Loscher, 2004b;
Rogawski, 2006a), and gabapentin and carbamazepine as well
as phenytoin are used in the treatment of post-herpetic neuralgia
and diabetic neuropathy, respectively (Wiffen et al., 2005;
Taylor, 1996; Collins et al., 2000). The mechanisms underlying
the antinocicepitive action of antiepileptic drugs might include
an enhancement of the GABAergic neurotransmission or effects
on neuronal voltage-gated sodium and/or calcium channels
(Rogawski and Loscher, 2004b). Whereas antiepileptic drugs
are efficacious in neuropathic pain, their activity in models of
acute pain is less clear. Only negligible effects of lamotrigine,
felbamate and gabapentin against an acute noxious stimulus
were determined in the rat tail flick test (Hunter et al., 1997).
Retigabine did not show antinociceptive effects in this test
(Blackburn-Munro and Jensen, 2003). Tiagabine, a GABA
uptake inhibitor, produced dose-dependent antinociception in
the hot-plate test in mice (Giardina et al., 1998) where
lamotrigine and gabapentin were not efficacious (Laughlin
The synthetic procedures of the ureides 1–13 (Fig. 1) are
given in the Supplementary Material. Ureides 1–13 were
suspended in 0.5% methylcellulose (Loba-Chemie, Germany).
Picrotoxin (Fluka, Germany), pentylenetetrazol (Cefarm,
Poland), and ethosuximide (Ronton, ICN-Polfa Rzeszów,
Poland) were dissolved in 0.9% sodium chloride (Rhone-
Poulenc Rorer, France). Diazepam (Valium, Roche, France) was
used as a 1% aqueous solution. Phenytoin (Phenhydan, Desitin
Pharma, Switzerland) was used as a 5% solution. Appropriate
amounts of the corresponding vehicles were given to the control
animals. Compounds 1–13 and reference compounds (ethosux-
imide, diazepam and phenytoin) were given intraperitoneally
(i.p.). Control animals received appropriate amounts of the
vehicle i.p.
All chemical reagents for the synthesis were obtained from
Merck (Darmstadt, Germany), Fluka (Taufkirchen, Germany),
Aldrich (Steinheim, Germany) or Acros (Geel, Belgium). For
detailed information on further materials and methods see
Supplementary Material.
2.3. Spontaneous locomotor activity
The spontaneous locomotor activity including different types
of movements such as locomotions, sniffing and grooming of a

140
T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
Table 1
injected i.p. into mice 45 min after the tested compound had
been administered. The animals were placed individually in
cages and observed for the next 30 min. The absence of seizures
showed the ability of the compound to elevate the seizure
threshold. The time of the onset of seizures, the number of
attacks and the mortality were assessed (Vogel and Vogel,
1997). Ethosuximide (130 mg/kg) was used as a reference drug.
Results and effects of the compounds in the maximal electroshock seizure test
Compound
Dose
(mg/kg)
Tonic seizures
Mortality
X/Y
X (%)
Z/Y
Z (%)
Methylcellulose
Phenytoin
1
–
5
10/10
0/10
10/10
10/10
8/10
9/10
8/10
6/10
9/10
9/10
9/10
9/10
9/10
4/10
6/10
1/10
6/10
1/10
5/10
0/10
5/10
0/10
10/10
9/10
9/10
4/10
9/10
10/10
100
0
0/10
0/10
3/10
4/10
1/10
0/10
0/10
1/10
2/10
1/10
2/10
1/10
0/10
0/10
3/10
0/10
2/10
0/10
0/10
0/10
1/10
0/10
3/10
1/10
0/10
0/10
1/10
0/10
0
0
30
40
10
0
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
100
80
90
80
60
90
90
90
90
90
40
60
10
60
10
50
0
2
2.6. Maximal electroshock seizure model
3
0
Electroconvulsions were produced by means of an alternat-
ing current (0.2 s stimulus duration, 50 Hz and 150 mA)
delivered via ear-clip electrodes by a COTM stimulator (Rodent
Shocker, type GE-01, Bialystok, Poland). Mice were divided
into groups of ten. A drop of 0.9% saline solution was poured
into each ear prior to placing the electrodes. Tested compounds
(1–13, 100 and 200 mg/kg) were injected i.p. 60 min before the
test, respectively. The animals are observed closely for 2 min.
Disappearance of the hindleg extensor tonic convulsions, the
percentage of inhibition of seizure relative to control were
10
20
10
20
10
0
0
30
0
20
0
0
0
10
0
4
5
6
7
8
9
10
11
12
13
50
0
Table 2
Results and effects of the compounds in the pentylenetetrazol test
100
90
90
40
90
100
30
10
0
0
10
0
Compound
Dose
Latency of first Number of attacks Mortality
(mg/kg) the attack (%)a per group (%)b
(%)
Ethosuximide 130
+946c
+349c
+832c
+103
+159
+2
−87d
−55
−75e
−20
−36
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
X: Number of mice with tonic seizures.
Y: Total number of tested mice in the experiment.
Z: Number of dead mice.
2
3
+187
−30
−20
−50e
+80
single mouse was measured in circular photoresistor actometers
(30 cm in diameter, provided with two photocells, and connected
to the impulse electronic counter), in 30 min sessions. The
investigated compounds 1–13 were administered i.p. at doses
100 and 200 mg/kg. Thirty minutes after the injection of the
investigated compounds, mice were placed in the actometers.
Each crossing of the light beam was recorded automatically. The
amount of impulses was noted after 30 and 60 min.
4
+563e
+482e
+538d
+389d
+248d
+847c
+187f
+505c
+188f
+533c
+293e
+1119c
+194
5
−40f
0
6
−75e
−30
−80c
−30
−84c
−45
−100c
−36f
−100c
+20
7
8
9
2.4. Picrotoxin-induced convulsions
10
11
12
13
Groups of 10 mice were treated i.p. with a suspension of the
tested compounds (in the range between 100 and 200 mg/kg) in
0.5% methylcellulose. After 60 min 3.2 mg/kg picrotoxin were
injected subcutaneously (s.c.). This dose produced convulsions
in 100% of non-pretreated animals, but did not cause death.
During the next 90 min the animals were observed for clonic
and tonic seizures. The times of the onset of seizures and the
number of attacks have been recorded (Vogel and Vogel, 1997).
Diazepam (1.5 mg/kg) was used as a reference drug.
+655c
+162c
+320d
+168f
+502c
+354e
+270d
−30
−20
−84c
−55e
−45d
Values in comparison to a control group which received only pentylenetetrazol
and 0.5% methylcellulose.
a
The first attack in the control groups was in the range of 113–288 s (see
Supplementary Material).
b
Number of attacks within a period of 30 min. The number of attacks in the
2.5. Pentylenetetrazol-induced convulsions
control groups was in the range of 1.0–1.25 (see Supplementary Material).
c
Statistical significance: Pb0.001.
Pb0.02.
Pb0.01.
Pb0.05.
Groups of 10 mice were treated i.p. with a suspension of the
tested compounds (100 and 200 mg/kg) in 0.5% methylcellu-
lose. A convulsant dose of pentylenetetrazol (70 mg/kg) was
d
e
f

T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
141
Table 3
measured. Mice were placed individually on the metal plate
heated to 52 0.4 °C and covered with a glass cylinder (25 cm
high, 15 cm in diameter). The time (s) elapsing to the first pain
response (licking of the forepaws or jumping) was determined
by a stop-watch and then recorded. The experiments were
conducted 30 min following the i.p. and p.o. administration of
the investigated compounds.
Results and effects of the compounds in the picrotoxin test
Compound
Dose
(mg/kg)
Latency of the
Number of attacks
per group (%)b
first attack (%)a
Diazepam
1
1.5
+450c
+171d
+191e
+14
+76
+53
+16
+41
+68
+8
−100c
+20e
−47
+40
−10
−26
−18
−23
−23
+117
+83
+9
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
2
2.9. Statistical analysis and calculations
3
4
The data are expressed as means S.E.M. Student's t-test
was used to determine the significance of differences between
mean values of control and treatment groups. Differences were
considered significant when Pb0.05.
The latency of the first attack in both the pentylenetetrazol
and the picrotoxin test was quantified, and percentage values in
comparison with the control group were calculated as % value=
[100×(latency of the first attack in the treatment group)/(latency
of the first attack in the control group)]−100. The number of
attacks in both the pentylenetetrazol and the picrotoxin test were
5
+13
+47
6
+140d
+104d
+253c
+174c
+137f
+209f
+337c
+138d
+186c
+37
0
7
−51e
−71f
−70f
−56d
−75
−73f
−47d
−40
+69
+43
+10
−20
−50
−60e
8
9
10
11
12
13
Table 4
Results and effects of the compounds on the spontaneous locomotor activity
+65
+84
Compound
Dose (mg/kg)
(i.p.)
Locomotor activity
(%)a after 60 minb
Locomotor activity
(%)a after 90 minb
+144d
+122d
+141e
1
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
−9
+3
+4
+6
2
−12
−46d
−32
−30c
−7
−39c
−68e
−49d
−57d
+16
Values in comparison to a control group, which only received picrotoxin and
0.5% methylcellulose.
3
a
The first attack in the control groups was in the range of 868–1597 s (see
Supplementary Material).
4
b
Number of attacks within a period of 90 min. The number of attacks in the
−10
−21
−69e
−24
−49
−42f
−62e
−63e
−72e
−27
−68d
−10
−47d
−21
−21
+19
+16
control groups was in the range of 1.0–1.5 (see Supplementary Material).
5
−16
c
Statistical significance: Pb0.001.
Pb0.05.
Pb0.02.
Pb0.01.
−29f
−20
d
6
e
−61d
−43f
−74e
−66d
−86e
−40d
−63d
−29
f
7
recorded and calculated (Vogel and Vogel, 1997). Phenytoin
(5 mg/kg) was used as a reference drug.
8
9
2.7. Neurotoxicity
10
11
12
13
−72e
−23
Neurotoxicity of the investigated compounds was assessed
by means of the chimney test. This test was carried out 60 min
after i.p. administration of the tested compounds (1–13, 100 and
200 mg/kg). The animals were previously trained and selected.
Then they were placed in a 25 cm long and 3 cm in diameter
horizontally located tube which was reversed in such a way that
the mice were able to leave it only climbing backward up as soon
as they reached another end. The ability of the mice to leave the
tube within 1 min was accounted for the lack of neurotoxic
properties of the investigated compounds (Boissier et al., 1960).
+15
+13
−49d
−14
−12
−70d
−36d
−52d
Values in comparison to a control group which only received a solution of
NaCl (0.9%).
a
Movements were counted within a period of 30 min.
Movements were counted 60 and 90 min after compound administration,
b
respectively. The locomotor activity of the control groups after 60 min was in
the range of 110.8–144.8 (see Supplementary Material). The locomotor
activity of the control groups after 90 min was in the range of 356.6–485.2 (see
Supplementary Material).
2.8. Analgesic effects
c
Statistical significance: Pb0.05.
Pb0.01.
Pb0.001.
Pb0.02.
Analgesic activity of the compounds 1–13 (100 and 200 mg/
kg) was examined by the hot-plate test (Eddy and Leimbach,
1953). In this test, reaction of mice to painful stimulus was
d
e
f

142
T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
Table 5
determination of nonspecific binding 2-chloroadenosine
(10 μM) was used in the A1 assay and N-ethylcarboxamidoa-
denosine (50 μM) in case of the A2A assay. Incubation times were
90 min (A1) and 30 min (A2A) at 23 °C, respectively. Initially, a
single high concentration of each compound (250 μM) was tested
in three independent experiments. For potent compounds which
showed greater than 50% inhibition of radioligand binding at the
test concentration, curves were determined using 6–7 different
concentrations of tested compound spanning 3 orders of
magnitude. At least three separate experiments were performed
each in triplicate. Data were analyzed using the PRISM program
version 3.0 (GraphPad, San Diego, CA, USA).
Adenosine A3 receptor binding assays were carried out at
membranes of Chinese hamster ovary (CHO) cells stably
transfected with the human adenosine A3 receptors. The
radioligand used for this experiment was (R)-2-phenyl-8-ethyl-
4-methyl-4,5,7,8-tetrahydro-1H-imidazo[2,1-i]purine-5-one
([3H]PSB-11, specific activity 53 Ci/mmol, concentration
0.5 nM), and for the nonspecific binding (R)-phenylisopropyl-
adenosine (100 μM) was used. The incubation was performed at
23 °C for 60 min.
Neurotoxicity of the compounds in the chimney test
Compound
Dose (mg/kg)
Mice impaired (%)
Methylcellulose
1
–
0
0
0
11.1
20.0
10.0
10.0
0
0
0
0
0
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
10
2
3
4
5
6
0
7
10.0
37.5
0
50.0
0
0
0
30.0
0
0
30.0
30.0
10.0
10.0
50.0
8
9
10
11
For two potent compounds (8, 10) at adenosine A1 receptors,
a GTP shift experiment was performed. Inhibition of binding of
12
13
Table 6
Analgesic activity of the compounds in the hot-plate test
Diazepam
Compound
Dose
(mg/kg)
Time of reaction to pain
stimulus (s) S.E.M.
Analgesic
activity (%)
The results are expressed in percentage of animals that failed to perform the
chimney test.
The experimental groups consisted of 10 mice.
Route
p.o.
i.p.
p.o.
i.p.
Control
1
–
11.3 0.7
24.9 4.6
25.4 2.5
13.2 1.7
21.7 3.2
17.1 1.6
25.3 3.2
22.6 4.6
25.0 2.4
16.3 2.3
33.5 6.8
29.4 4.5
22.6 2.0
11.2 1.2
19.0 3.3
15.2 2.0
20.3 3.3
19.3 2.7
34.4 4.8
16.6 2.4
13.2 2.7
19.7 1.8
31.2 4.3
14.5 2.3
21.9 2.5
21.0 3.8
25.7 2.8
11.0 1.1
23.9 2.2
19.0 2.0
24.8 3.2
19.9 2.8
23.2 2.6
27.7 4.3
24.4 3.5
19.5 3.1
20.7 2.5
26.1 4.1
22.9 1.7
23.0 2.7
16.8 3.2
15.6 2.4
12.4 3.6
19.5 3.2
29.5 3.0
31.8 5.1
13.9 2.5
15.5 3.1
22.5 2.5
21.4 2.5
21.3 2.5
20.9 2.7
32.0 5.6
23.0 3.5
–
–
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
100
200
121a
125b
17
116 b
72b
125b
81c
quantified as % value=[100×(number of attacks per treatment
group)/(number of attacks per control group)]−100. The loco-
motor activity was quantified as % value=[100×(movements
of the treatment group within a period of 30 min)/(movements
of the control group within a period of 30 min)]−100. Analgesic
activity was quantified as % value=[100×(time of reaction to
pain stimulus of the treatment group)/(time of reaction to pain
stimulus of the control group)]−100.
2
93b
52b
124b
100a
122b
44a
198c
161b
101b
0
3
111 b
151b
121b
76b
88b
137b
107b
108b
52a
4
5
6
2.10. Radioligand binding studies at adenosine receptors and
GTP shift experiments
7
69c
35a
80c
71c
206b
47a
17
41
8
12
77c
Radioligand binding assays were performed essentially as
previously described (Müller et al., 2000; Hayallah et al., 2002)
using rat brain cortical membrane preparations as a source for
adenosine A1 receptors, and rat brain striatal membrane
preparations as a source of adenosine A2A receptors. Frozen rat
brains were obtained from Pel-Freez®, Rogers, Arkansas, USA.
The test compounds were dissolved in a mixture of aqueous
Tris–HCl buffer and dimethyl sulfoxide (DMSO) (9:1); the final
DMSO concentration was 2.5%. [3H]2-Chloro-N6-cyclopentyla-
denosine ([3H]CCPA, 1 nM) was used as the selective A1
radioligand (Klotz et al., 1989), [3H]3-(3-hydroxypropyl)-7-
methyl-8-(m-methoxystyryl)-1-propargylxanthine ([3H]MSX-2,
1 nM) as the A2A ligand (Müller et al., 2000). For the
9
167b
188b
26
10
11
12
13
41
74b
177b
29
104
94b
93b
89b
190b
108b
95b
86c
128b
Route: p.o.=oral, i.p.=intraperitoneal.
a
Statistical significance: Pb0.05.
Pb0.01.
b
c
Pb0.02.

T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
143
the A1 antagonist radioligand [3H]1,3-dipropyl-N6-cyclopentyl-
xanthine ([3H]DPCPX, 0.4 nM) by the compounds was
measured in the absence and in the presence of GTP
(100 μM). For the nonspecific binding unlabelled DPCPX
(10 μM) was used, and as positive control curves for the full
adenosine A1 receptor agonist N6-cyclopentyladenosine (CPA,
100 μM) were determined under the same conditions. The
incubation was terminated after 90 min at 23 °C, by rapid
filtration with a Brandel 48-channel cell harvester through GF/B
glass fiber filters. The filters were rinsed three times with ice-cold
Tris–HCl buffer, 50 mM, pH 7.4.
were dissolved in DMSO, and the final DMSO concentration in
the assays was 1%. In a final volume of 1 ml, each test tube
contained 790 μl of Tris–HCl buffer (50 mM, pH 7.4), 10 μl of
compound solution in DMSO, 100 μl of rat cerebral cortical
membrane preparation with a protein concentration of ca.
100 μg per tube, and 100 μl of [3H]diazepam solution, to give a
final concentration of 1 nM. Incubations were performed on an
ice bath for 1 h and were terminated by rapid filtration through
GF/B glass-fiber filters (Whatman, USA) using a Brandel 48-
channel cell harvester (Brandel, Gaithersburg, MD, USA).
Three 5 ml washes with ice-cold Tris–HCl buffer 50 mM, pH
7.4 were performed. Unlabelled diazepam (5 μM) was used to
define non-specific binding. All compounds were initially
tested in a single concentration of 1 mM in at least three
independent experiments each in triplicate. For the more potent
compounds full inhibition curves were determined in 4–5
separate experiments each in triplicate. Ki values were
calculated from IC50 values, determined by the non-linear
regression program PRISM version 3.0 (GraphPad, San Diego,
2.11. Radioligand binding studies at benzodiazepine binding
sites of GABAA receptors
Compounds 1 and 8–10 were investigated in radioligand
binding assays at rat brain cortical membranes for their affinity
to the benzodiazepine binding site of the GABAA receptor
essentially as described (Geis et al., 1996). The compounds
Table 7
Affinities of selected urea derivatives to adenosine receptors and the benzodiazepine binding site of GABAA receptors
Compound
Structure
Ki [μM] (n=3)
A1 rat brain cortical
membranes vs.
[3H]CCPA
A2A rat brain striatal
membranes vs.
[3H]MSX-2
A3 human
recombinant vs.
[3H]PSB-11
Benzodiazepine rat brain
cortical membranes vs.
[3H]diazepam
1
N250 (18 7)a
N100 (15 7)b
N100 (25 6)b
N1000 (−6 2)c
5
N100 (18 3)b
N100 (19 4)b
N100 (38 1)b
n.d. d
8
26.4 1.5
N100 (27 7)b
N100 (28 5)b
N100 (29 6)b
N100 (17 3)b
15.6 2.7
661 30
621 30
922 31
n.d. d
9
18.9 1.2
65.1 5.9
10
17.2 3.8
81.5 3.5
11
N100 (17 4)b
N100 (26 6)b
a
Percent inhibition of radioligand binding S.E.M. at 250 μM.
Percent inhibition of radioligand binding S.E.M. at 100 μM.
Percent inhibition of radioligand binding S.E.M. at 1000 μM.
Not determined.
b
c
d

144
T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
CA, USA), using the Cheng–Prusoff equation and a KD value
of 4 nM for diazepam.
Depending of the given concentration, the time of the onset of
the first epileptic seizure was increased when 1, 6 and 9 were
administered. With N-benzyl-N′-butyrylurea (9) at 200 mg/kg
the latency was increased compared to the reference drug
ethosuximide at 130 mg/kg. Compounds 1 and 6 (at 200 mg/kg)
delayed the onset of the first attack nearly as effective as
ethosuximide. For compounds 2 and 3 no significant anticon-
vulsant effect could be found in this test. Referring to the
number of attacks, compounds 9 and 10 were proven to be very
active (Table 2), leading to a complete reduction at 200 mg/kg,
thus being more potent than ethosuximide at 130 mg/kg. A
reduction of the number of attacks, that was comparable to that
of ethosuximide, could also be produced with compounds 1, 6–
8 and 12 (at a dose of 200 mg/kg). With 6 and 9 an increased
activity could be observed by increasing the doses given (see
Supplementary Material).
Binding of selected compounds to the GABA binding site
and activation of GABAA receptors was tested indirectly by
measuring the allosteric enhancement by GABA and GABAA
agonists of [3H]diazepam binding to the benzodiazepine
binding site. GABA (100 μM) was used as a standard for
comparison. The tested compounds were used in concentrations
of 1, 10, 100 and 1000 μM.
3. Results
3.1. Maximal electroshock seizure test
Compounds 6, 7, 8 and 12 were able to protect more than 50%
of the animals from tonic seizures at a dose of 200 mg/kg. The
test compounds 9 and 10 were active in both doses (Table 1) and
even caused a complete reduction and full protection from
convulsions at 200 mg/kg. However, phenytoin completely
prevented tonic seizures in a 40-fold lower dose. After
administration of compounds 6, 9 and 12 no mouse died from
seizures (Table 1).
3.3. Picrotoxin test
The test compounds delayed the onset of seizures to different
extents (Table 3). Compounds 2–5 and 11 did not or only
slightly delay the occurrence of the first epileptic attack. In
particular, ureides 7 and 9 showed a significant prolongation of
latency at a dose of 200 mg/kg. Concerning the number of
attacks (Table 3), compounds 5 and 11 turned out to strengthen
the activity of the proconvulsant substance picrotoxin, similar to
the results obtained in the pentylenetetrazol test (see compound
5 at a dose of 100 mg/kg, Table 2). Best reduction of the seizures
3.2. Pentylenetetrazol test
Nearly all tested compounds showed an anticonvulsant effect
due to the considerably delayed onset of seizures (Table 2).
Several substances reduced the number of attacks significantly.
Fig. 1. Chemical structures of the ureides.

T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
145
was found with compounds 7, 8 and 9 (about 50–75%
reduction). Nearly no effect was observed with compounds 2,
6 and 12 (Table 3).
In this test, none of the ureides reached the inhibition level
achieved with diazepam, irrespective of the much lower dose of
diazepam.
8, 9, and 10 exhibited affinity for the rat brain A1 receptor with
Ki values of 17–26 μM (Table 7), while the derivatives 1, 5 and
11 showed only weak affinity. At A3 receptors, the most potent
compound was 8 (Ki =15.6 μM). Compounds 9 and 10 were
several-fold less potent (9, Ki =65.1 μM, 10, Ki =81.5 μM), and
again 1, 5 and 11 were virtually inactive.
In order to investigate whether the compounds were agonists
or antagonists at adenosine A1 receptors, GTP shift assays were
performed using the antagonist radioligand [3H]8-cyclopentyl-
1,3-dipropylxanthine ([3H]DPCPX). Curves for agonists are
shifted to the right due to an allosteric effect of GTP binding to
the Gi protein and thereby shifting the high affinity conforma-
tion of the receptor protein to a low affinity conformation for
agonists. Antagonist curves are not affected by GTP addition.
Fig. 2 shows the curves for the standard agonist N6-
cyclopentyladenosine (CPA, A), compounds 8 (B) and 10 (C).
While a clear shift of the curve for the agonist CPAwas observed
3.4. Spontaneous locomotor activity
Compounds 1, 4 and 11 had no influence on the spontaneous
locomotor activity of the mice (Table 4). Substance 12 increased
the locomotor activity at a dose of 100 mg/kg and decreased it
significantly, when 200 mg/kg were given. All other tested
compounds showed a decreased spontaneous locomotor
activity, that was significant in many cases (Table 4). A higher
dose of these substances caused a clear reduction of the activity.
The lowest locomotor activity was found, when compound
8 was given (Table 4).
3.5. Neurotoxicity
Compounds 1, 4–6, 9 and 11 were proven to be not
neurotoxic at the investigated doses (Table 5). All animals were
able to leave the tube within 1 min. Ureides 2, 3 and 13 were
slightly neurotoxic. At the dose of 200 mg/kg compounds 7, 8,
10 and 12 showed an increased neurotoxic potential.
3.6. Analgesic effects
All tested compounds showed a clear analgesic effect in the
hot-plate test (except 7 and 10 which possessed an only weak
activity, Table 6). After p.o. administration, the compounds 1,
3–5, 9, 11 and 13 showed the highest, statistically significant
analgesic activity at a dose of 200 mg/kg. Compounds 3, 5 and 9
caused highest analgesia at 200 mg/kg when given intraper-
itoneally. Dose-dependency was observed after oral as well as
intraperitoneal administration of compounds 3, 5 and 9. Among
all compounds examined in the hot-plate test, the ureide 9 was
observed to be most potent (Table 6).
3.7. Adenosine receptors
Selected compounds (1, 5, 8–11) were investigated in vitro
for their interaction with adenosine receptors (Table 7).
Radioligand binding studies were performed at rat brain cortical
membranes as a source for adenosine A1 receptors, and at rat
brain striatal membranes as a source for adenosine A2A
receptors. Due to the low density of A3 receptors in rat brain
tissues, membrane preparations of Chinese hamster ovary cells
containing recombinant human A3 receptors were used. [3H]2-
Chloro-N6-cyclopentyladenosine ([3H]CCPA), was used as A1,
[3H]3-(3-hydroxypropyl)-7-methyl-8-(m-methoxystyryl)-1-
propargylxanthine ([3H]MSX-2) as A2A, and (R)-2-phenyl-8-
ethyl-4-methyl-4,5,7,8-tetrahydro-1H-imidazo[2,1-i]purine-5-
one ([3H]PSB-11) as A3 receptor radioligand. None of the
compounds showed appreciable affinity for the adenosine A2A
receptor subtype. In contrast, the benzyl-substituted compounds
Fig. 2. GTP shift experiments at adenosine A1 receptors. Radioligand binding
curves were determined at rat brain cortical membranes in the absence and in the
presence of GTP (100 mM) for the full agonist N6-cyclopentyladenosine (CPA,
A), and for urea derivatives 8 (B) and 10 (C) using the antagonist radioligand
DPCPX. Data represent means S.E.M. of 6 (CPA) or 3 (compound 8 and 10)
separate experiments performed in triplicate.

146
T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
antiepileptic drugs offer better tolerability and improved
pharmacokinetic characteristics (Rogawski, 2006b). Many
urea derivatives have been discovered as potent anticonvulsant
drugs (Wong et al., 1986; Pavia et al., 1990) and the structure-
activity relationship studies have been reported, especially the
analysis of the interaction of these compounds with the
benzodiazepine binding site of the GABAA receptor complex
(Palluotto et al., 1996; Marder et al., 2001; Verli et al., 2002;
Hemmateenejad et al., 2004). The GABAA/benzodiazepine
receptor recognizes a large spectrum of compounds from
different chemical classes that represent benzodiazepine
receptor ligands. Benzodiazepines are a well-known class of
anticonvulsants. On the other hand, a consequence of an
intensive search for GABA modulatory agents with a non-
benzodiazepine structure was the evaluation of the antiepileptic
profile of such compounds (Teuber et al., 1999). A number of
GABA analogues have been developed and are therapeutically
used (Bryans and Wustrow, 1999).
In the present study, the anticonvulsant activity of thirteen
newly synthesized benzyl- and benzhydryl ureides in different
models of seizures in mice was investigated. Antiepileptic drug
research has for several decades focused on identifying new
potential drugs based on their anticonvulsant activity against
single acute seizures induced by various stimulators, usually in
mice. The most widely used models have been the maximal
electroshock seizure test, pentylenetetrazol test or picrotoxin-
induced seizures test. All established antiepileptic drugs have
anticonvulsant activity in at least one of these models (Löscher
and Schmidt, 1994; Rogawski, 2006a). These tests may in
some way distinguish the potential utility of compounds
against different seizure types. Compounds effective in the
maximal electroshock seizure screen may be more useful in
partial and generalized tonic — clonic seizures, whereas those
effective in the pentylenetetrazol test may have activity against
absence seizures. Pentylenetetrazol and picrotoxin block the
GABAA receptor in a similar way, but different characteristics
of pentylenetetrazol inhibition compared with picrotoxin
inhibition indicated that the domains of their interaction are
not identical (Huang et al., 2001). In general, picrotoxin has
been reported to display inhibitory interactions to various
anion-selective, ligand-gated channels besides GABAA recep-
tors, i.e. GABAC receptors, glycine receptors, and glutamate-
gated chloride channels (Qian et al., 2005). In future
investigations to characterize the anticonvulsant effects of the
ureides, glutamatergic or cholinergic agonists could be used to
induce seizures in mice (Weaver et al., 1997; Yamashita et al.,
2004; Jaworska-Feil et al., 2001; Turski et al., 1987).
Fig. 3. Effect of GABA (100 μM) and various concentrations of compound 9
(1–1000 μM) on [3H]diazepam binding (1 nM) to rat brain cortical membranes.
Results are means S.E.M. of 3 separate experiments performed in triplicate.
Percent increase (GABA) or decrease (compound 9) of specific binding is given
in comparison to controls without compound added.
in the presence of GTP, no change was seen for the two urea
derivatives investigated indicating that they are antagonists.
While GTP caused a statistically highly significant, 31-fold
increase in the IC50 value of the agonist CPA, virtually identical
IC50 values were determined for 8 and 9, whose IC50 values
with and without GTP added were statistically not significantly
different.
3.8. GABAA receptors
Compounds 1, and 8–10, that showed good activities in
electrically or chemically induced seizure models, were
investigated in radioligand binding studies for their interaction
with the benzodiazepine binding site of GABAA receptors
(Table 7). Inhibition of [3H]diazepam binding by the com-
pounds in rat brain cortical membrane preparations was
determined. Compound 1 had no affinity for the benzodiazepine
binding site even at high concentrations (1 mM), while
compounds 8–10 exhibited low affinities (Ki =620–920 μM).
While benzodiazepines are allosteric modulators (enhan-
cers) of the binding of GABA to GABAA receptors, GABA
conversely enhances the binding of diazepam to its allosteric
site at the receptors. Fig. 3 shows this effect: GABA (100 μM)
enhanced the binding of 1 nM of [3H]diazepam by 38%. In
order to investigate whether the anticonvulsant urea derivative
9 can also activate GABAA receptors by binding to the
orthosteric GABA binding site, the allosteric effect of different
concentrations of 9 on [3H]diazepam binding was investigat-
ed. However, in contrast to GABA, [3H]diazepam binding was
not enhanced, but only reduced by 9 at high concentrations
indicating that it is not a direct GABA agonist.
From the results of this study, the following structure-activity
relationships could be derived. On the one hand, the substitution
pattern at position 1 of the ureas was compared: hydrogen (1) vs.
benzyl (2, 4) or benzhydryl groups (3, 5). On the other hand, the
influence of benzoyl (4, 5) and trichloroacetyl (1–3) at position 3
was determined. The compounds 1–5 did practically not reduce
the number of tonic seizures in the maximal electroshock seizure
test. Trichloroacetylurea (1) behaved differently in electrically
and chemically induced seizure models. Showing no effect in the
maximal electroshock seizure test, 1 clearly reduced the number
4. Discussion
In past years the discovery and development of anticonvul-
sant drugs has been one of the noticeable research fields.
However, most antiepileptic drugs currently used in therapy of
epilepsy, exhibit dose related toxicity and many undesirable
side effects. The search for new compounds combining strong
anticonvulsant activity with lower toxicity and restricted side
effects is in progress, and some of the new generation

T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
147
of attacks in the pentylenetetrazol test and delayed the onset of
the first seizure in pentylenetetrazol as well as in picrotoxin test.
However, compound 1 was not chosen as starting compound for
further structural modifications as it led to a relatively high
mortality of the mice in the maximal electroshock seizure test.
Whereas 3-benzoyl-1-benzylurea (4) had no influence on the
spontaneous locomotor activity, an undesired reduction was
observed with compounds 2, 3 and 5. Furthermore 4 was not
neurotoxic. From these results benzyl was selected as the fixed
substituent at position 1 whereas different acyl residues were
introduced at the position 3 of the urea core. Two subgroups of
new compounds were generated, the first one including the
acetyl derivates 6, 7 and 11, and the second one the alkanoyl
derivatives 7–10, 12–13.
1000 mg/kg, 5- to 10-fold higher than those used in this
investigation. Compound 9 itself had not been included in those
experiments.
Our results with benzyl ureides revealed an optimal anti-
convulsant activity if a butyryl substituent was introduced. With
longer or shorter alkanoyl chain lenghts the activity of the com-
pounds decreased. The branched 1-benzyl-3-isobutyrylurea (10)
was equipotent in some experiments compared to its isomer 9.
The results from the hot-plate test indicate that the best
anticonvulsant compound 9 had remarkable analgesic activity
according to the strongly prolonged reaction time of the mice to
the pain stimulus. It should be noted, that the reduced sponta-
neous locomotor activity caused by 9 could partially contribute
to this effect.
Within the first subgroup 3-acetyl-1-benzylurea (7) caused
the strongest reduction of the number of seizures in the maximal
electroshock seizure test as well as in the pentylenetetrazol and
picrotoxin test at 200 mg/kg compared with 1-benzyl-3-
cyanoacetylurea (6) and 1-benzyl-3-phenacetylurea (11). Com-
pound 7 had the highest anticonvulsant potential in this
subgroup, but it also showed the most pronounced side effects,
i.e. neurotoxicity and reduction of spontaneous locomotor
activity.
To possibly explain both the anticonvulsant and analgesic
effects, adenosine receptors were considered as potential
molecular targets from the following reasons. Adenosine has
been shown to exhibit anticonvulsant activityin vivo by activating
adenosine receptors and was thus considered as an endogenous
anticonvulsant (Dunwiddie, 1985; During and Spencer, 1992).
The A1 receptor subtype appears to be responsible for the
anticonvulsant activity of adenosine since A1-selective agonists
exhibited the same activity (Adami et al., 1995; De Sarro et al.,
1999; Zhang et al., 1994). This anticonvulsant effect could result
from hyperpolarization mediated by opening of potassium
channels. Some adenosine A1 receptor agonists are effective in
different types of seizures, e.g. induced by picrotoxine. Because
selective A1 and A2A receptor agonists showed dose-dependent
activity in the pentylenetetrazol test, both receptor-subtypes may
be involved in the protection against seizures. On the other hand,
A1 adenosine receptor antagonists may induce epileptic seizures
(De Sarro et al., 1999).
Endogenous adenosine is known to regulate pain transmis-
sion. Spinal analgesia by adenosine is mainly mediated by
adenosine A1 receptors and to a lesser extent through A2
receptors. In the periphery, adenosine exerts a variety of effects
on pain signalling by action on different receptors and cellular
targets (Sawynok and Liu, 2003). Pain transmission could by
modulated by a number of agents, particularly acting as ligands
of adenosine A1 receptors. The response of A1 receptor knock-
out mice to thermal pain was faster than that of wild-type mice
(Johansson et al., 2001). Thus, adenosine was evaluated as an
analgesic substance, and the A1 receptor was suggested to be a
target for the development of antinociceptive drugs. On the other
hand, reduced response to acute pain was found in A2A receptor
knock-out mice compared to A2A receptor wild-type mice. This
may be attributed to the peripheral lacking of A2A receptors in
those mice. Interaction with adenosine receptors and their
complex effects on pain signalling may also explain why
caffeine – an adenosine receptor antagonist – has analgesic
effects against some types of pain (Fredholm et al., 2005).
Taken this into account, we investigated selected compounds
for their interaction with adenosine receptors. In fact, the
benzylurea derivatives investigated (8–10) showed relatively
high affinity for adenosine A1 and A3 receptors with Ki values
in the low micromolar range, but were virtually inactive at A2A
receptors. Subsequently, compounds 8 and 10 were functionally
To further improve the pharmacological profile of 7, the acetyl
group was replaced by other alkanoyl residues. This led to the
second subgroup of compounds (7–10, 12–13) to investigate the
influence of the chain length and chain branching of the residues
on the antiepileptic activity. It was found in a study with
monosubstituted acylureas (Mrongovius, 1975), that ureides with
seven carbon atoms showed high central depressant activity that
was increased in isomeric derivatives with branched chains. The
reduced spontaneous locomotor activity of 7–10, 12–13 could be
explained as a result of a central depression. On the basis of the
data at 200 mg/kg the hexanoyl derivative 13 showed the lowest
reduction of spontaneous locomotor activity whereas the
propionyl compound 8 reduced it most strikingly. These findings
differed from Mrongovius' results with NH2-substituted ureides.
The ability to reduce the number of tonic seizures in the
maximal electroshock seizure test increased in the following
order (i) hexanoyl (13)bpentanoyl (12)bacetyl (7)≅propionyl
(8)b butyryl (9)≅isobutyryl (10). In the pentylenetetrazol and
picrotoxin tests (data from a dose of 200 mg/kg) the following
rank for the reduction of the number of seizure could be found (ii)
hexanoyl (13)bpentanoyl (12)≅propionyl (8)≅acetyl (7)b
isobutyryl (10)≅butyryl (9), (iii) pentanoyl (12)bisobutyryl
(10)bhexanoyl (13)≅propionyl (8)bacetyl (7)≅butyryl (9).
From these comparisons the excellent antiepileptic activity of
1-benzyl-3-butyrylurea (9) could be recognized. This ureide
prolonged the onsets of the first epileptic attack in the chemically
induced tests dramatically and in a dose-dependent manner.
Moreover, compound 9 was not neurotoxic and did not cause
mortality. Compound 9 was superior to the other derivatives in
this study and turned out to be equipotent to ethosuximide in the
pentylenetetrazol test.
A generally good activity of butyrylureas in the pentylene-
tetrazol test had already been reported by Zirvi and Fakouhi
(Zirvi and Fakouhi, 1982), however with administered doses of

148
T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
Adami, M., Bertorelli, R., Ferri, N., Foddi, M.C., Ongini, E., 1995. Effects of
repeated administration of selective adenosine A1 and A2A receptor agonists
on pentylenetetrazole-induced convulsions in the rat. Eur. J. Pharmacol. 294,
383–389.
Arbuzov, B.A., Fedotova, N.R., Zobova, N.N., Nazyrova, A.Z., Anan'ev, E.V.,
Gorbunov, S.M., 1989. Synthesis, antiinflammatory and analgesic activities
of N-acylureas. Khim.-Farm. Zh. 23, 682–683.
Baraldi, P.G., Guarneri, M., Manfredini, S., Simoni, D., Zanirato, V., 1989.
Synthesis and evaluation of the antifungal activity of a new series of N-
acylureas. Farmaco 44, 391–398.
Baulac, M., 2003. Rational conversion from antiepileptic polytherapy to
monotherapy. Epileptic Disord. 5, 125–132.
characterized by GTP shift assays indicating that these ureides
behaved as antagonists at adenosine A1 receptors. Thus, the
observed anticonvulsant activity could not be due to the
activation of adenosine A1 receptors. In contrast, this effect may
even reduce their apparent anticonvulsant activity. An effect on
the benzodiazepine binding site of GABAA receptors could also
be excluded since very high concentrations of the investigated
compounds (1, 8–10) were required to compete with [3H]
diazepam binding. A direct action on the GABA binding site of
the GABAA receptor is also not likely to be the molecular
mechanism of the anticonvulsant action, since the most potent
compound of the present series (9) – in contrast to GABA – did
not allosterically enhance [3H]diazepam binding. The molecu-
lar mechanism of action of the new acylurea derivatives appears
to be non-adenosinergic and probably non-GABAergic.
1-Benzyl-3-butyrylurea (9), also the best analgesic compound
of the series, exhibited affinity to the adenosine A1 receptors
(Ki =18.9 μM) in the same range as caffeine (Ki =18.8 μM)
(Bulicz et al., 2006), but in contrast to caffeine exhibited no A2A
affinity. However, action at the adenosine receptors did most
likely not contribute to the analgesic properties of 9. From the
results of the adenosine and GABA receptor binding studies as
well as the seizure and hot-plate tests, it may be concluded that
voltage-gated ion channels are likely to be the molecular target
for this ureide. This assumption has to be proven in further
experiments to assess its potential therapeutic value.
In summary, novel ureides were evaluated as anticonvulsants
in representative seizure models in mice. Benzyl derivatives 8–
10 showed strong anticonvulsant activity and relatively high
affinity for the adenosine A1 receptors. The anticonvulsant
activity of ureide derivatives is probably not related to their
influence on adenosine A1 receptors since compounds 8 and 10
turned out to be antagonists at this receptor subtype. Although it
affected the spontaneous locomotor activity, 1-benzyl-3-butyr-
ylurea (9) was superior to all other ureides investigated,
combining strong anticonvulsant and antinociceptive effects
with a lack of neurotoxicity.
Beasley, Y.M., Petrow, V., Stephenson, O., 1961. Some 1-Alkyl-3-aroyl ureas.
J. Pharm. Pharmacol. 13, 694–697.
Beghi, E., Perucca, E., 1995. The management of epilepsy in the 1990s. Acqui-
sitions, uncertainties and priorities for future research. Drugs 49, 680–694.
Blackburn-Munro, G., Jensen, B.S., 2003. The anticonvulsant retigabine
attenuates nociceptive behaviours in rat models of persistent and neuropathic
pain. Eur. J. Pharmacol. 460, 109–116.
Boissier, J.R., Tardy, J., Diverres, J.C., 1960. Une nouvelle méthode simple pour
explorer l'action ‘tranquillisante’: le test de la cheminée. Med. Exp. 3,
81–84.
Bryans, J.S., Wustrow, D.J., 1999. 3-Substituted GABA analogs with central
nervous system activity: a review. Med. Res. Rev. 19, 149–177.
Bulicz, J., Bertarelli, D.C., Baumert, D., Fülle, F., Müller, C.E., Heber, D., 2006.
Synthesis and pharmacology of pyrido[2,3-d]pyrimidinediones bearing
polar substituents as adenosine receptor antagonists. Bioorg. Med. Chem.
14, 2837–2849.
Carraz, G., Emin, N., 1967. Anticonvulsive action of di-n-propylacetic acid
monoureide and of hydantoin derivatives of this acid. Thérapie 22, 641–652.
Chiti, W., 1960. New ureides with local anesthetic action. Farmaco Sci. 15, 29–43.
Collins, S.L., Moore, R.A., McQuay, H.J., Wiffen, P., 2000. Antidepressants and
anticonvulsants for diabetic neuropathy and postherpetic neuralgia: a
quantitative systematic review. J. Pain Symp. Manage. 20, 449–458.
De Sarro, G., De Sarro, A., Donato Di Paola, E., Bertorelli, R., 1999. Effects of
adenosine receptor agonists and antagonists on audiogenic seizure-sensible
DBA/2 mice. Eur. J. Pharmacol. 371, 137–145.
Drabczyńska, A., Müller, C.E., Schumacher, B., Hinz, S., Karolak-Wojcie-
chowska, J., Michalak, B., Pekala, E., Kiec-Kononowicz, K., 2004. Tricyclic
oxazolo[2,3-f]purinediones: potency as adenosine receptor ligands and
anticonvulsants. Bioorg. Med. Chem. 12, 4895–4908.
During, M.J., Spencer, D.D., 1992. Adenosine: a potential mediator of seizure
arrest and postictal refractoriness. Ann. Neurol. 32, 618–624.
Dunwiddie, T.V., 1985. The physiological role of adenosine in the central
nervous system. Int. Rev. Neurobiol. 27, 63–139.
Eddy, N.B., Leimbach, D., 1953. Synthetic analgesics. II. Dithienylbutenyl- and
dithienylbutylamines. J. Pharmacol. Exp. Ther. 107, 385–393.
Fredholm, B.B., Chen, J.F., Masino, S.A., Vaugeois, J.M., 2005. Action of
adenosine at its receptors in the CNS: insights from knockouts and drugs.
Annu. Rev. Pharmacol. Toxicol. 45, 385–412.
Frommel, E., Radouco-Thomas, C., 1953. Effects of phenobarbital, acylureas
and diphenylhydantoin on coma during electric epileptic crisis in guinea pig.
Helv. Physiol. Acta 11, 270–273.
Geis, U., Kiec-Kononowicz, K., Müller, C.E., 1996. Benzylidene-substituted
imidazo-thiazole, -thiazine and -thiazepine derivatives. A new class of
ligands for the benzodiazepine binding site of GABAA receptors. Sci.
Pharm. 64, 383–390.
Giardina, W.J., Decker, M.W., Porsolt, R.D., Roux, S., Collins, S.D., Kim, D.J.,
Bannon, A.W., 1998. An evaluation of the GABA uptake blocker tiagabine
on animal models of neuropathic and nociceptive pain. Drug Dev. Res. 44,
106–113.
Acknowledgements
M.M., C.E.M., and M.G. want to thank the DFG
Graduiertenkolleg 804 “analysis of cell functions by combina-
torial chemistry and biochemistry” for financial support. S.S.
was supported by the European Commission (ERASMUS
program). The authors dedicate this work to Professor Dr. Kurt
Eger, Leipzig, on the occasion of his 65th birthday.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at doi:10.1016/j.ejphar.2006.12.002.
Goldstein, L., Pfeiffer, C.C., 1971. Quantitative EEG analysis of single-dose
effect relationships in normal volunteers of Pacinox® (capuride), a new
antianxiety drug. Biol. Psychiatry 3, 165–172.
References
Graber, H., Árr, M., Csiba, A., Ludwig, E., Magyar, T., 1981. Azlocillin and
mezlocillin: new penicillins and their value. Orv. Hetil. 122, 749–756.
Hayallah, A.M., Sandoval-Ramirez, J., Reith, U., Schobert, U., Preiss, B.,
Schumacher, B., Daly, J.W., Müller, C.E., 2002. 1,8-disubstituted xanthine
Abo-Salem, O.M., Hayallah, A.M., Bilkei-Gorzo, A., Filipek, B., Zimmer, A.,
Müller, C.E., 2004. Antinociceptive effects of novel A2B adenosine receptor
antagonists. J. Pharmacol. Exp. Ther. 308, 358–366.

T. Librowski et al. / European Journal of Pharmacology 559 (2007) 138–149
149
derivatives: synthesis of potent A2B-selective adenosine receptor antago-
nists. J. Med. Chem. 45, 1500–1510.
striatal membranes — a new, selective antagonist radioligand for A(2A)
adenosine receptors. Eur. J. Pharm. Sci. 10, 259–265.
Hemmateenejad, B., Miri, R., Tabarzad, M., Jafarpour, M., Zand, F., 2004.
Molecular modeling and QSAR analysis of the anticonvulsant activity of
some N-phenyl-N′-(4-pyridinyl)-urea derivatives. J. Mol. Struct., Theochem
684, 43–49.
Palluotto, F., Carotti, A., Casini, G., Campagna, F., Genchi, G., Rizzo, M., De
Sarro, G.B., 1996. Structure-activity relationships of 2-aryl-2,5-dihydropyr-
idazino[4,3-b]indol-3(3H)-ones at the benzodiazepine receptor. Bioorg.
Med. Chem. 4, 2091–2104.
Huang, R.Q., Bell-Horner, C.L., Dibas, M.I., Covey, D.F., Drewe, J.A., Dillon,
G.H., 2001. Pentylenetetrazole-induced inhibition of recombinant γ-
aminobutyric acid type A (GABAA) receptors: mechanism and site of
action. J. Pharmacol. Exp. Ther. 298, 986–995.
Hunter, J.C., Gogas, K.R., Hedley, L.R., Jacobson, L.O., Kassotakis, L.,
Thompson, J., Fontana, D.J., 1997. The effect of novel anti-epileptic drugs
in rat experimental models of acute and chronic pain. Eur. J. Pharmacol. 324,
153–160.
Pavia, M.R., Lobbestael, S.J., Taylor, C.P., Hershenson, F.M., Miskell, D.L.,
1990. N-phenyl-N′-pyridinylureas as anticonvulsant agents. J. Med. Chem.
33, 854–861.
Polezhaeva, A.I., 1966. A comparative investigation of the phenylacetyl urea
and phenoxyacetyl urea anticonvulsant activity. Farmakol. Toksikol. 29,
278–281.
Qian, H., Pan, Y., Zhu, Y., Khalili, P., 2005. Picrotoxin accelerates relaxation of
GABAC receptors. Mol. Pharmacol. 67, 470–479.
Jastrzębska-Więsek, M., Librowski, T., Czarnecki, R., Marona, H., Nowak, G.,
2003. Central activity of new xanthone derivatives with chiral center in some
pharmacological tests in mice. Pol. J. Pharmacol. 55, 461–465.
Jaworska-Feil, L., Kajta, M., Budziszewska, B., Leskiewicz, M., Lason, W.,
2001. Protective effects of TRH and its stable analogue, RGH-2202, on
kainate-induced seizures and neurotoxicity in rodents. Epilepsy Res. 43,
67–73.
Johansson, B., Halldner, L., Dunwiddie, T.V., Masino, S.A., Poelchen, W.,
Gimenez-Llort, L., Escorihuela, R.M., Fernandez-Teruel, A., Wiesenfeld-
Hallin, Z., Xu, X.J., Hardemark, A., Betsholtz, C., Herlenius, E., Fredholm,
B.B., 2001. Hyperalgesia, anxiety, and decreased hypoxic neuroprotection in
mice lacking the adenosine A1 receptor. Proc. Natl. Acad. Sci. 98, 9407–9412.
Khalil, M., Weaver, D.F., 1990. A pattern recognition study of acyclic ureide
anticonvulsants. J. Pharm. Pharmacol. 42, 349–351.
Rogawski, M.A., 2006a. Molecular targets versus models for new antiepileptic
drug discovery. Epilepsy Res. 68, 22–28.
Rogawski, M.A., 2006b. Diverse mechanisms of antiepileptic drugs in the
development pipeline. Epilepsy Res. 69, 273–294.
Rogawski, M.A., Löscher, W., 2004a. The neurobiology of antiepileptic drugs.
Nat. Rev., Neurosci. 5, 553–564.
Rogawski, M.A., Loscher, W., 2004b. The neurobiology of antiepileptic drugs
for the treatment of nonepileptic conditions. Nat. Med. 10, 685–692.
Sawynok, J., Liu, X.J., 2003. Adenosine in the spinal cord and periphery: release
and regulation of pain. Prog. Neurobiol. 69, 313–340.
Sobol, E., Bialer, M., Yagen, B., 2004. Tetramethylcyclopropyl analogue of a
leading antiepileptic drug, valproic acid. Synthesis and evaluation of
anticonvulsant activity of its amide derivatives. J. Med. Chem. 47,
4316–4326.
Kieć-Kononowicz, K., Müller, C.E., Pekala, E., Karolak-Wojciechowska, J.,
Handzlik, J., Łażewska, D., 2002. Imidazo[2,1-b]thiazoles, imidazo[2,1-b]
imidazoles and pyrrolo[1,2-c]imidazoles. Synthesis, structures and evalua-
tion of benzodiazepine receptor binding. J. Heterocycl. Chem. 39, 243–253.
Kieć-Kononowicz, K., Karolak-Wojciechowska, J., Michalak, B., Pękala, E.,
Schumacher, B., Müller, C.E., 2004. Imidazo[2,1-b]thiazepines. Synthesis,
structures and evaluation of benzodiazepine receptor binding. Eur. J. Med.
Chem. 39, 205–218.
Klotz, K.N., Lohse, M.J., Schwabe, U., Cristalli, G., Vittori, S., Grifantini, M.,
1989. 2-Chloro-N6-[3H]cyclopentynadenosine ([3H]CCPA) — a high
affinity agonist radioligand for A1 adenosine receptors. Naunyn-Schmiede-
berg's Arch. Pharmacol. 340, 679–683.
Kulev, L.P., Dobychina, N.S., 1963. N,N′-Benzylacyl derivatives of urea. Zh.
Vses. Khim. Obshch. Mendeleeva 8, 701–702.
Laughlin, T.M., Tram, K.V., Wilcox, G.L., Birnbaum, A.K., 2002. Comparison
of antiepileptic drugs tiagabine, lamotrigine, and gabapentin in mouse
models of acute, prolonged, and chronic nociception. J. Pharmacol. Exp.
Ther. 302, 1168–1175.
Spielman, M.A., Geiszler, A.O., Close, W.J., 1948. Anticonvulsant drugs. II.
Some acylureas. J. Am. Chem. Soc. 70, 4189–4191.
Taylor, C.P., 1996. Voltage-gated Na+ channels as targets for anticonvulsant,
analgesic and neuroprotective drugs. Curr. Pharm. Des. 2, 375–388.
Teuber, L., Watjens, F., Jensen, L.H., 1999. Ligands for the benzodiazepine
binding site — a survey. Curr. Pharm. Des. 5, 317–343.
Todorovic, S.M., Rastogi, A.J., Jevtovic-Todorovic, V., 2003. Potent analgesic
effects of anticonvulsants on peripheral thermal nociception in rats. Br. J.
Pharmacol. 140, 255–260.
Turski, L., Cavalheiro, E.A., Czuczwar, S.J., Turski, W.A., Kleinrok, Z., 1987.
The seizures induced by pilocarpine: behavioral, electroencephalographic
and neuropathological studies in rodents. Pol. J. Pharmacol. Pharm. 39,
545–555.
Verli, H., Albuquerque, M.G., Bicca de Alencastro, R., Barreiro, E.J., 2002.
Local intersection volume: a new 3D descriptor applied to develop a 3D-
QSAR pharmacophore model for benzodiazepine receptor ligands. Eur. J.
Med. Chem. 37, 219–229.
Vogel, H.G., Vogel, W.H., 1997. Drug Discovery and Evaluation, Pharmaco-
logical Assays. Springer, Berlin, pp. 260–261.
Weaver Jr., C.E., Marek, P., Park-Chung, M., Tam, S.W., Farb, D.H., 1997.
Neuroprotective activity of a new class of steroidal inhibitors of the N-methyl-
D-aspartate receptor. Proc. Natl. Acad. Sci. U. S. A. 94, 10450–10454.
Weiss, H., 1971. Capuride for sleep in rehabilitation patients: a controlled trial.
Arch. Phys. Med. Rehabil. 52, 68–72.
Wiffen, P., Collins, S., McQuay, H., Carroll, D., Jadad, A., Moore, A., 2005.
Anticonvulsant drugs for acute and chronic pain. Cochrane Database Syst.
Rev. Issue 3, CD001133.
Wong, M.G., Defina, J.A., Andrews, P.R., 1986. Conformational analysis of
clinically active anticonvulsant drugs. J. Med. Chem. 29, 562–572.
Yamashita, H., Ohno, K., Amada, Y., Hattori, H., Ozawa-Funatsu, Y., Toya, T.,
Inami, H., Shishikura, J., Sakamoto, S., Okada, M., Yamaguchi, T., 2004.
Effects of 2-[N-(4-chlorophenyl)-N-methylamino]-4H-pyrido[3.2-e]-1,3-
thiazin-4-one (YM928), an orally active alpha-amino-3-hydroxy-5-methyl-
4-isoxazolepropionic acid receptor antagonist, in models of generalized
epileptic seizure in mice and rats. J. Pharmacol. Exp. Ther. 308, 127–133.
Zhang, G., Franklin, P.H., Murray, T.F., 1994. Activation of adenosine A1
receptors underlies anticonvulsant effect of CGS 21680. Eur. J. Pharmacol.
255, 239–243.
Löscher, W., Schmidt, D., 1994. Strategies in antiepileptic drug development: is
rational drug design superior to random screening and structural variation?
Epilepsy Res. 17, 95–134.
Marder, M., Estiu, G., Blanch, L.B., Viola, H., Wasowski, C., Medina, J.H.,
Paladini, A.C., 2001. Molecular modeling and QSAR analysis of the
interaction of flavone derivatives with the benzodiazepine binding site of the
GABA(A) receptor complex. Bioorg. Med. Chem. 9, 323–335.
Mendyk, A., Sałat, K., Librowski, T., Czarnecki, R., Malawska, B., 2001.
Influence of new γ-aminobutyric acid amide, derivatives and its phthalimide
precursors on the central nervous system activity in mice. Pol. J. Pharmacol.
53, 689–693.
Meusel, M., Gütschow, M., 2004. Recent developments in hydantoin chemistry.
A review. Org. Prep. Proced. Int. 36, 391–443.
Mrongovius, R.I., 1975. Structure-activity correlations for central depressant
acylureas and alkylureas. Eur. J. Med. Chem. Chem. Ther. 10, 474–479.
Mrongovius, R.I., Gaff, G.A., Rand, M.J., 1976. Comparison of the bromureide
sedative-hypnotic drugs, bromvaletone (bromisoval) and carbromal, and
their chloro analogues in mice. Clin. Exp. Pharmacol. Physiol. 3, 443–447.
Mui, W.C., Zbuzek, V.K., Wu, W.H., 1997. GABA(A) antagonist and nicotin-
induced antinociception. Life Sci. 61, PL 221–PL 225.
Müller, C.E., Maurinsh, J., Sauer, R., 2000. Binding of [3H]MSX-2 (3-(3-
hydroxypropyl)-7-methyl-8-(m-methoxystyryl)-1-propargylxanthine) to rat
Zirvi, K.A., Fakouhi, T., 1982. Synthesis and neuropharmacology of
butyrylureas. Farmaco Sci. 37, 335–342.