Synthesis and anticonvulsant activity of new 2,3-benzodiazepines as AMPA receptor antagonists

Article abstract of DOI:10.1016/S0014-827X(99)00022-1

Novel 1-aryl-3,5-dihydro-7,8-methylenedioxy-4H-2,3-benzodiazepin-4-ones (12a-j) were prepared and their anticonvulsant effects were evaluated by using various models of experimental epilepsy. The seizures were evoked both by means of auditory stimulation

Full text of DOI:10.1016/S0014-827X(99)00022-1

Il Farmaco 54 (1999) 178–187
Synthesis and anticonvulsant activity of new 2,3-benzodiazepines
as AMPA receptor antagonists
Angela De Sarro a, Giovambattista De Sarro b, Rosaria Gitto c, Silvana Grasso c,*,
c
Nicola Micale c, Maria Zappala`
a Cattedra di Chemioterapia, Istituto di Farmacologia, Facolta` di Medicina, Uni6ersita` di Messina, piazza XX Settembre, I-98100 Messina, Italy
b Cattedra di Farmacologia, Dipartimento di Medicina Sperimentale e Clinica, Uni6ersita` di Catanzaro, 6ia T. Campanella,
I-88100 Catanzaro, Italy
c Dipartimento Farmaco-Chimico, Uni6ersita` di Messina, 6iale Annunziata, I-98168 Messina, Italy
Received 15 November 1998; received in revised form 10 December 1998; accepted 25 January 1999
Abstract
Novel 1-aryl-3,5-dihydro-7,8-methylenedioxy-4H-2,3-benzodiazepin-4-ones (12a–j) were prepared and their anticonvulsant
effects were evaluated by using various models of experimental epilepsy. The seizures were evoked both by means of auditory
stimulation in DBA/2 mice and by pentylenetetrazole or maximal electroshock in Swiss mice. Some of these compounds possess
marked anticonvulsant properties in all tests employed. Compounds 12 antagonise seizures induced by AMPA in analogy to the
structurally-related 1-(4%-aminophenyl)-4-methyl-7,8-methylenedioxy-5H-2,3-benzodiazepine (1) (GYKI 52466), a well-known
non-competitive AMPA-receptor antagonist. On the other hand, these novel 2,3-benzodiazepines exhibit anticonvulsant properties
that are not affected by flumazenil, but are reversed by aniracetam. In addition, when compared to model compound 1,
compounds 12 show a longer-lasting anticonvulsant activity and a lower toxicity. A structure–activity relationship study carried
out on compounds 12 as well as analogous 7,8-dimethoxy derivatives 2 offers an approach for designing more potent agents.
© 1999 Elsevier Science S.A. All rights reserved.
Keywords: 7,8-Methylenedioxy-4H-2,3-benzodiazepin-4-ones; Anticonvulsant activity; AMPA-receptor antagonists
1. Introduction
tic acid (NMDA) receptor and two non-NMDA
receptors named (R,S)-2-amino-3-(3-hydroxy-5-methyl-
isoxazol-4-yl)propionic acid (AMPA) and kainic acid
(KA) receptors [3,4]. It is now widely accepted that in
ischemic or hypoxic conditions such as stroke, trauma
and epilepsy both NMDA and non-NMDA receptors
are overstimulated by an increased amount of endoge-
nous Glu [5,6]. Therefore, the identification of competi-
tive and non-competitive NMDA as well as non-NMDA
receptor antagonists, which prevent the opening of the
ion channels associated with these iGluRs, could be of
potential clinical value both in acute (stroke and trauma)
and chronic (Alzheimer’s disease, epilepsy) neurological
disorders [7]. During the past 15 years numerous NMDA
antagonists have been the subject of an intense research
program [8,9] but preliminary results were not encourag-
ing since strong side-effects e.g. psychotomimetic action
[10,11] and impairment of learning and memory [12]
hampered their clinical development.
(S)-Glutamic acid (Glu), the most abundant excita-
tory neurotransmitter of the central nervous system
(CNS), plays a key role in the physiological function of
the CNS [1]. Two main classes of Glu receptors have
been characterised: the metabotropic (mGluRs) and
ionotropic (iGluRs) receptors. The mGluRs regulate the
activity of ionic channels or enzymes by producing
second messengers via GTP-coupled proteins [2]. The
iGluRs are Glu-gated cationic channels directly respon-
sible for the fast depolarisation of postsynaptic cells.
They are constituted of different subunits and classified
into three heterogeneous types based on their pharma-
cology and functional properties: the N-methyl-D-aspar-
* Corresponding author. Tel.: +39-090-676 6470; fax: +39-090-
355 613.
E-mail address: grasso@pharma.unime.it (S. Grasso)
0014-827X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0014-827X(99)00022-1

A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
179
The general interest seems now to be focused on
agents acting selectively on AMPA receptors because
of their relevance in the treatment of epilepsy [13,14]
and cerebral ischaemia [15,16]. At this time, 6-nitro-7-
sulfamoylbenzo[ f ]quinoxaline-2,3-dione (NBQX) [15]
and 6-(1H-imidazol-1-yl)-7-nitro-2,3(1H,4H)-quinoxa-
linedione (YM90K) [17], both belonging to the class of
the quinoxaline-2,3-diones, are among the most potent
and selective competitive AMPA-antagonists.
1-(4%-Aminophenyl)-4-methyl-7,8-methylenedioxy-
5H-2,3-benzodiazepine (1) (GYKI 52466) (Fig. 1) has
been identified as a potent and selective non-competi-
tive AMPA-receptor antagonist that appears to act via
a novel allosteric site on the receptor complex [18,19].
It shows anticonvulsant properties [20–23] and behaves
as a neuroprotective agent in focal and global is-
chaemia [24,25]. Compound 1, in spite of its chemical
similarity to 1,4-benzodiazepines, does not bind to the
benzodiazepine receptor (BZR) complex and, therefore,
it is devoid of any sedative–hypnotic effect [26].
As part of a program aimed at identifying potent
and selective AMPA receptor antagonists, we have re-
cently reported on the synthesis of a series of 1-aryl-3,5-
dihydro-7,8-dimethoxy-4H-2,3-benzodiazepin-4-ones
(2) (Fig. 1) [27,28]. Some of these compounds proved
to be more potent than 1 as anticonvulsants in various
seizure models. Furthermore, when compared to the
model compound 1, they showed a longer-lasting activ-
ity and a lower toxicity. Electrophysiological assays
suggest that the above-mentioned 2,3-benzodiazepines
2, in analogy to the lead compound 1, act via a
non-competitive blocking mechanism of the AMPA
receptor complex [28].
Fig. 1. Structures of compounds 1 (GYKI 52466) and 2a–f.
compounds 12 for the benzodiazepine receptor was
evaluated through the inhibition of [3H]flumazenil (Ro
15-1788) binding in a cortical membrane preparation.
In addition the potency of compounds 12 for the BZR
was assessed in vivo by the ability of flumazenil to
reverse their anticonvulsant activity. The activity of
12a and 12e on the AMPA-receptor complex was eval-
uated through their ability to inhibit the AMPA-in-
duced seizures. Furthermore, aniracetam (Ro 13-5057),
a compound which potentiates the effects of AMPA
[31], antagonises the anticonvulsant properties of com-
pounds 12a and 12e. Finally, we assessed the propen-
sity of title compounds to induce neurological
impairment by using the rotarod test. A preliminary
account of the present results has been reported
[32].
If we compare the structure of 2,3-benzodiazepines 1
and 2, we note two structural modifications, i.e. the
replacement of the dioxole nucleus with two methoxy
groups on the fused aromatic ring and the introduction
of a carbonyl moiety in position 4 of the heptatomic
nucleus. In order to test the role of the two methoxy
groups of derivatives 2 on the biological activity, we
prepared and tested some dioxole analogues as
anticonvulsants.
This paper reports the synthesis of novel 1-aryl-3,5-
dihydro-7,8-methylenedioxy-4H-2,3-benzodiazepin-4-
ones (12a–j) and the evaluation of their anticonvulsant
properties in DBA/2 mice, a strain genetically suscepti-
ble to sound-induced seizures, which has been consid-
ered an excellent animal model for generalised epilepsy
and for screening new anticonvulsant drugs [29,30].
Some derivatives were also examined in Swiss mice
both against pentylenetetrazole (PTZ) and maximal
electroshock (MES)-induced seizures and the time
course of anticonvulsant activity was also studied. In
addition, we attempted to correlate the anticonvulsant
properties of the new 2,3-benzodiazepines 12 with their
affinity for the BZR or AMPA receptor. The affinity of
2. Chemistry
At first, we tackled the synthesis of compounds 12
through the procedure already successfully applied to
the preparation of derivatives 2 [28]. This strategy
requires a Jones oxidation of 1-aryl-6,7-methylene-
dioxyisochromans 5 to yield ketoacids 6 followed by a
cyclocondensation with hydrazine (Scheme 1). This
procedure proved to be unsatisfactory when applied to
isochromans
5 because the condensation of 3,4-
methylenedioxyphenethyl alcohol (3) with an aromatic
aldehyde 4 gave the expected products 5 in poor yields.
Furthermore, the oxidation of compounds 5 with Jones
reagent (method b) led to 2-aroyl-4,5-methylenedioxy-
benzoic acid (7) as the only isolable product, whereas, a
mixture of 2-aroyl-4,5-methylenedioxyphenylacetalde-
hyde (8), the main product, and a small amount of acid
7 was obtained by using ruthenium dioxide/sodium
periodate (method c) [33]. In both cases the expected
2-aroyl-4,5-methylenedioxyphenylacetic acid (6) was
never detected. Therefore, we developed an alternative
strategy as shown in Scheme 2. Ketoesters 11 were

180
A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
Scheme 1. Reagents: (a) HCl(g) saturated dioxane, reflux, 1 h; (b) 35% H2SO4, CrO3, acetone; (c) NaIO4, RuO2, CCl4/CH3CN/H2O, room
temperature, 20 h.
easily prepared via Friedel–Craft acylation of methyl
3,4-methylenedioxyphenylacetate (9) with the appropri-
ate aroyl chloride 10 by using tin(IV) chloride as the
catalyst. The subsequent treatment of ketoesters 11
with hydrazine or N-methylhydrazine gave 1-aryl-3,5-
dihydro-7,8-methylenedioxy-4H-2,3-benzodiazepin-4-
ones (12) in good yields. When an amino or hydroxy
group is present on the phenyl ring, i.e. 12d–g, the
Friedel–Craft acylation step was carried out on tri-
fluoroacetamido- (10d–e) or acetoxy- (10f) benzoyl
chlorides, respectively. The treatment of the resulting
ketoesters 11d–f with an excess of hydrazine or N-
methylhydrazine gave both cyclisation and complete
deprotection of the amino and hydroxy functionalities
to yield directly final derivatives 12d–g. N-Acetyl
derivative 12c was obtained by reacting 12a with an
excess of acetic anhydride in the presence of triethyl-
amine. Physical and spectral data (1H NMR) of the
synthesised compounds are in agreement with the pro-
posed structures.
3. Results and discussion
The novel 7,8-methylenedioxy-4H-2,3-benzodiaze-
pines (12a–j) were tested for anticonvulsant activity
against audiogenic seizures in DBA/2 mice and the
results were compared with those previously reported
[27,28] for 7,8-dimethoxy-4H-2,3-benzodiazepine ana-
logues (2a–f) and for GYKI 52466 (1), chosen as the
reference compound. The compounds were intraperi-
toneally (i.p.) injected in various doses (dose range:
3.3–200 mmol/kg) in DBA/2 mice and their anticonvul-
sant properties were evaluated 30 min after administra-
tion; the measured ED50 values are reported in Table 1.
Among the synthesised compounds, derivatives 12d
and 12e are the most active ones showing anticonvul-
sant potency comparable to 2d and 2e and higher than
that of GYKI 52466.
A survey of the present results put in evidence that
the replacement of the two methoxy groups on the
benzene fused ring with a dioxole nucleus seems to have
Scheme 2. Reagents: (a) SnCl4/CH2Cl2, 0–5°C, 1 h; room temperature, 12 h; (b) NH2NH2 · H2O or MeNHNH2, EtOH, reflux, 6–7 h; (c) Ac2O,
Et3N, CHCl3, room temperature, 1–2 h.

A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
181
Table 1
mmol/kg for 12c versus ED50=40.6 mmol/kg for 12a in
contrast to what was previously observed for deriva-
tives 2 (Table 1). Further work is in progress in order to
clarify the role played by the substituent at N-3.
The time course of anticonvulsant activity was also
studied in order to discover new compounds with a
longer time course than 1. Following i.p. administration
of some active compounds such as 2a and 12a (50
mmol/kg), maximum protection was observed after 45–
90 min with subsequent return to control seizure re-
sponse at 180 min. In the same experimental conditions,
GYKI 52466 displayed the maximum protection in
5–15 min followed by a gradual return to control
seizure response in 30–90 min (Fig. 2).
Derivatives 12d and 12e, the most active ones of the
series, as well as the unsubstituted parent compound
12a were selected for further evaluation against MES-
and PTZ-induced seizures in Swiss mice. As shown in
Table 2, the tonic extension and the clonic phase of the
seizures induced by MES and PTZ, respectively were
significantly reduced at 45 min after i.p. administration
of the tested compounds.
In order to correlate the anticonvulsant activity of
novel compounds 12 with affinity for AMPA receptor,
an additional test against AMPA-induced seizures in
DBA/2 mice was performed (Table 2). As shown in Fig.
3, the clonic and tonic phases of the seizures induced by
intracerebroventricular (i.c.v.) administration of AMPA
were significantly reduced 30 min after i.p. administra-
tion of 12a, 12d, 12e and 1, analogously to what has
already been reported [28] for 2a, and 2e. In the present
study we also demonstrated that aniracetam, a potenti-
ator of the AMPA effects [31], markedly antagonised
the anticonvulsant effects of 12e in DBA/2 mice (Table
2) and shifted to the right the dose–response curves
(Fig. 4) with a pattern of activity similar to that of 1
and 2e [28]. On the basis of electrophysiological experi-
Anticonvulsant activity of compounds 1, 2a–fa and 12a–j against
audiogenic seizures in DBA/2 mice and TD50 values on locomotion
assessed by rotarod test
b
Comp.ED50 (mmol/kg)
TD50 (mmol/kg)
TIc
Clonic phase
Tonic phase
Locomotor deficit
2a
12a
2b
33.9 (26.0–44.2) 31.8 (24.8–40.6) 142 (87.3–231)
43.3 (34.4–54.6) 40.6 (30.1–54.9) 159 (82.6–306)
37.8 (23.7–60.1) 26.7 (14.7–48.2) 154 (104–227)
4.2
3.7
4.1
12b
2c
12c
2d
82.5 (58.6–116) 70.4 (58.8–84.3)
101 (52.0–194) 72.1 (47.6–109) 181 (106–309)
36.6 (28.5–47.1) 32.6 (25.5–42.1) 110 (79.3–151)
\200
N.D.d
1.8
3.0
19.3 (16.9–22.0) 18.3 (16.0–20.8) 51.5 (34.1–77.8) 2.7
18.0 (10.0–32.5) 12.7 (6.13–26.2) 101 (52.0–194) 5.6
15.0 (9.01–24.0) 12.6 (8.01–19.0) 56.8 (39.3–82.1) 3.8
21.8 (13.2–36.0) 10.9 (4.60–26.4) 99.1 (72.4–135) 4.5
12d
2e
12e
2f
12f
12g
12h
12i
12j
1
50.2 (34.6–73.0) 43.7 (31.3–61.0) 159 (95.0–268)
3.2
41.1 (30.8–55.0) 29.5 (18.0–48.5) 80.1 (55.3–116) 2.0
52.3 (44.5–61.4) 49.6 (41.0–60.1) 151 (103–221)
2.9
\120
89.8 (55.2–146) 63.8 (45.4–89.7)
\120 \120
35.8 (24.4–52.4) 25.3 (16–40.0)
\120
\200
\200
\200
N.D.
N.D.
N.D.
76.1 (47.5–122) 2.1
a Data taken from Ref. [28].
b ED50 values with 95% confidence limits were calculated according
to the method of Litchfield and Wilcoxon [46].
c TI=ratio between TD50 and ED50 (from the clonic phase of the
audiogenic seizures).
d N.D., not determined.
little influence on anticonvulsant activity. As a matter
of fact, 7,8-methylenedioxy derivatives 12a, 12d and 12e
display a potency comparable to that of the corre-
sponding compounds 2a, 2d and 2e.
When compared to the unsubstituted compound 12a,
the introduction of a fluorine atom, a methoxy or a
trifluoromethyl group at C-4% drastically reduces or
suppresses the activity. On the contrary, the presence of
an amino group, in position 3% or 4%, increases the
anticonvulsant activity as previously observed [28] for
derivatives 2d and 2e. An increase in activity was also
obtained by the introduction of a hydroxy group at
C-4% even if its effect is minor than that of the amino
group.
As a matter of fact, derivative 12e, the most active
term of the series, is roughly three times more potent
than GYKI 52466 (1), i.e. ED50 10.9 mmol/kg for 12e
versus 31.8 mmol/kg for 1. Since the structures of 1 and
12e differ solely in the replacement of the azomethine
group of 1 with a lactam moiety, it can be deduced that
such a modification is fruitful and allows a better fitting
of the new structure with the complementary receptor
subsites of the AMPA-receptor complex.
The introduction of a methyl group at N-3 decreases
the activity i.e. ED50=29.5 mmol/kg for 12f versus
ED50=10.9 mmol/kg for 12e, whereas an opposite ef-
fect was observed with the acetyl group i.e. ED50=32.6
Fig. 2. Anticonvulsant effects of 2a, 12a, and 1 (50 mmol/kg i.p.)
against audiogenic seizures in DBA/2 mice. The ordinate shows
seizure score, the abscissa shows the time after i.p. administration of
drug in min. Ten animals were used for the determination of each
point.

182
A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
Table 2
Anticonvulsant activity of some benzodiazepines against the MES-, PTZ- and AMPA-induced seizures and against audiogenic seizures after
pretreatment with aniracetam
a
Comp. ED50 (mmol/kg)
MES (Swiss mice) PTZ (Swiss mice) AMPAb (DBA/2 mice)
Pretreatment with aniracetamc (DBA/2 mice)
Tonic phase
Clonic phase
Clonic phase
Tonic phase
Clonic phase
Tonic phase
2ad
12a
12d
2ed
12e
1
35.8 (28.6–44.7)
42.6 (26.4–68.8)
19.3 (16.9–22.0)
15.9 (7.3–33.5)
32.1 (23.2–44.3)
35.7 (29.3–43.4)
68.2 (54.6–85.2)
78.0 (46.0–132)
40.5 (22.9–71.7)
22.6 (11.7–43.8)
71.8 (53.2–96.9)
68.3 (56.2–83.1)
66.0 (45.9–94.9)
70.3 (58.7–84.1)
29.2 (16.9–50.4)
32.1 (23.2–44.3)
37.9 (27.3–52.8)
57.5 (43.5–76.0)
42.6 (26.4–68.8)
56.9 (40.9–79.2)
23.8 (14.2–39.9)
25.0 (16.5–30.0)
28.5 (20.6–39.4)
40.5 (26.3–60.8)
215 (142–324)*
N.D.e
157 (114–217)*
N.D.
N.D.
N.D.
65.4 (44.5–96.2)*
62.6 (44.7–87.7)
134 (88.8–203)*
58.2 (43.4–77.9)*
39.6 (22.9–68.7)
100 (63.4–158)*
a ED50 values with 95% confidence limits were calculated according to the method of Litchfield and Wilcoxon [46].
b AMPA was administered i.c.v. at the CD97 for either clonus (9.7 nmol) or forelimb tonic extension (11.7 nmol) 30 min after injection of each
compound.
c Significant differences between ED50 values of group treated with aniracetam+2,3-benzodiazepine and group treated with 2,3-benzodiazepine
alone (Table 1) are denoted: * PB0.01.
d Data taken from Ref. [28].
e N.D., not determined.
The anticonvulsant activity of the title compounds 12
was evident at doses which generally did not cause
sedation and ataxia. It has long been known that
ments carried out on derivatives 1 and 2e [28], along
with the observation that the anticonvulsant properties
of 12e were significantly reduced (from 2.9 to 3.6 times)
by aniracetam, a positive allosteric modulator of
AMPA receptors [31], we suggest that the present com-
pounds 12 antagonise the AMPA receptor-mediated
responses via an allosteric blocking mechanism.
To rule out a possible involvement of BZR on phar-
macological effects of the present compounds, deriva-
tives 1, 12a, and 12e were administered concomitantly
with flumazenil, but no significant modification of the
antiseizure effects of these derivatives was observed
(Table 3). On the other hand, compounds 12 were
assessed for the ability to displace [3H]flumazenil from
membranes of cortex. No inhibition was observed up to
concentration of 10 mM (IC50\10 mM).
Furthermore, by means of crude cortical synaptic
membranes, derivatives 12a and 12e (100 mM) failed
to displace [3H]spiperone from dopamine and 5-HT1
receptors; [3H]ketanserin from 5-HT2 receptors; [125I]-
pindolol from b-adrenergic receptors; [3H](R,S)-3-(2-
carboxypiperazin-4-yl)-propyl-1-phosphonic acid ([3H]-
CPP) and [3H]dizocilpine from NMDA receptors; [3H]-
5,7-dichlorokynurenic acid ([3H]5,7-DCKA) from gly-
cine site on NMDA receptors; [3H]AMPA and [3H]6-
cyano-7-nitroquinoxaline-2,3-dione ([3H]CNQX) from
AMPA/kainate receptors; mixture of [3H](1S,3R)ACPD
and [3H](1R,3S)ACPD from metabotropic glutamate
receptors. Despite the lack of activity of the present
2,3-benzodiazepines at dopamine, serotonin, nora-
drenaline, NMDA, glycine site on NMDA and metabo-
tropic glutamate binding sites an interaction at other
sites involved in the generation or expression of seizures
cannot presently be ruled out.
Fig. 3. Anticonvulsant effects of 12a, 12d, 12e and 1 against seizures
induced by AMPA in DBA/2 mice. The ordinate shows % of re-
sponse of clonic (a) or tonic (b) seizures, abscissa shows the dose in
mmol/kg i.p. For the determination of each point ten animals were
used.

A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
183
side effects [19,34–36]. There are, however, studies
showing that potent antagonists at the AMPA/kainate
receptor possess anticonvulsant effects at doses below
those impairing learning and behaviour [37–39]. Table
1 shows the doses which induced motor toxicity in 50%
of mice (TD50 values) obtained 30 min following the i.p.
administration of 2,3-benzodiazepines 12a–j. It is note-
worthy that the present compounds 12 cause signifi-
cantly less motor impairment in the rotarod test with
respect to 1 and some of them have therapeutic index
(TI) values approximately twice that of 1 (Table 1).
4. Conclusions
The novel 7,8-methylenedioxy-4H-2,3-benzodiazepin-
4-ones (12) reported in this study, analogously to 7,8-
dimethoxy-4H-2,3-benzodiazepin-4-ones (2), possess a
marked anticonvulsant activity both in the audiogenic
seizure test in DBA/2 mice and against MES and PTZ
induced seizures in Swiss mice. In particular, derivatives
12a and 12c–f show a potency comparable to or higher
than that of 1, the reference compound. They also
exhibit longer-lasting anticonvulsant action and lower
toxicity than compound 1. Moreover, they display anti-
convulsant effects against seizures induced by AMPA
in agreement with an involvement of AMPA receptors.
The structure–activity relationships in this series were
examined by three types of structural changes: (i) a
number of substituents at the C-4% position; (ii) the
insertion of an amino group in different positions on
the phenyl ring at C-1; (iii) the introduction of a methyl
or an acetyl group at N-3. In the light of these findings,
we can conclude that two methoxy groups or a dioxole
nucleus on the benzene ring play a similar role in
determining anticonvulsant effectiveness of these com-
pounds, whereas the replacement of the azomethine
moiety of GYKI 52466 with the lactam moiety plays a
role of utmost importance in the interaction with the
AMPA-receptor complex. Such a structural modifica-
tion is responsible for the improvement in the pharma-
cological profile observed in this new class of
2,3-benzodiazepines which promise to become useful
tools in the mapping of the AMPA receptors.
Fig. 4. Antagonism by aniracetam (50 nmol i.c.v.) of the anticonvul-
sant effects of 12e against audiogenic seizures in DBA/2 mice. The
ordinate shows % of response of clonic (a) or tonic (b) seizures. The
abscissa shows the dose in mmol/kg i.p. For the determination of
each point ten animals were used.
NMDA antagonists, especially those which block ion
channels (e.g. dizocilpine), can induce cognitive deficits
and a variety of other neurological and behavioural
Table 3
Anticonvulsant activity against audiogenic seizures after concomitant
treatment with flumazenil in DBA/2 mice
a
Comp. Flumazenil
ED50 (mmol/kg)
(mmol)
Clonic phase
Tonic phase
2ab
8.24
24.72
8.24
24.72
8.24
24.72
8.24
24.72
8.24
38.2 (28.5–51.2)
50.9 (36.9–70.1)
41.2 (36.0–47.2)
36.8 (28.3–47.7)
14.1 (10.1–19.9)
12.0 (6.82–21.0)
21.0 (16.2–27.2)
19.5 (16.7–22.8)
37.8 (23.7–60.1)
39.5 (29.6–53.7)
35.8 (26.2–49.0)
41.3 (28.0–60.7)
40.6 (27.4–60.2)
37.6 (31.2–45.3)
12.0 (6.82–21.0)
10.2 (7.61–13.7)
10.2 (7.61–13.7)
9.1 (5.70–14.5)
26.7 (14.7–48.2)
29.7 (20.9–42.1)
12a
2eb
5. Experimental
5.1. Chemistry
12e
Melting points were determined on a Kofler hot stage
apparatus and are uncorrected. Elemental analyses
were carried out on a Carlo Erba 1106 elemental ana-
lyzer for C, H, and N, and the results are within
90.40% of the theoretical values. Merck silica gel 60
F254 plates were used for analytical TLC; column chro-
1
24.72
a ED50 values with 95% confidence limits were calculated according
to the method of Litchfield and Wilcoxon [46].
b Data taken from Ref. [28].

184
A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
Table 4
Physical and spectral data of compounds 11 and 12
Yield
(%)
Comp.
11a
M.p. (°C)a
80–82 (A)
1H NMR l (ppm, J [Hz])b
81
3.59 (s, 3H, CH3), 3.80 (s, 2H, CH2), 5.98 (s, 2H, O–CH2–O), 6.82 and 6.87 (2s, 2H, H-3 and H-6),
7.42–7.78 (m, 5H, Ar)
11d
150–152 (B)
107–109 (B)
72
3.62 (s, 3H, CH3), 3.78 (s, 2H, CH2), 6.02 (s, 2H, O–CH2–O), 6.79 and 6.82 (2s, 2H, H-3 and H-6),
7.66 and 7.76 (2d, 4H, J=8.4, Ar), 8.76 (br s, 1H, NH)
3.62 (s, 3H, CH3), 3.82 (s, 2H, CH2), 6.06 (s, 2H, O–CH2–O), 6.83 and 6.87 (2s, 2H, H-3 and H-6),
7.51 (dd, 1H, J=7.9 and 8.1, H-5%), 7.64 (d, 1H, J=7.9, H-6%), 7.77 (s, 1H, H-2%), 8.00 (d, 1H,
J=8.1, H-4%), 8.18 (br s, 1H, NH)
11e
59
11g
11h
11i
150–152 (B)
99–101 (C)
129–131 (C)
157–159 (C)
184–186 (A)
170–172 (A)
188–190 (A)
246–248 (B)
83
89
50
53
74
58
46
42
2.31 (s, 3H, CH3CO), 3.60 (s, 3H, CH3), 3.78 (s, 2H, CH2), 6.03 (s, 2H, O–CH2–O), 6.84 and 6.87
(2s, 2H, H-3 and H-6), 7.14 and 7.82 (2d, 4H, J=8.8, Ar)
3.60 (s, 3H, CH3), 3.76 (s, 2H, CH2), 3.88 (s, 3H, CH3O-4%), 6.04 (s, 2H, O–CH2–O), 6.84 and 6.87
(2s, 2H, H-3 and H-6), 6.94 and 7.79 (2d, 4H, J=8.9, Ar)
3.61 (s, 3H, CH3), 3.79 (s, 2H, CH2), 6.04 (s, 2H, O–CH2–O), 6.83 and 6.85 (2s, 2H, H-3 and H-6),
7.13 (dd, 2H, J=8.8 and JH–F=8.5, H-3%,5%), 7.81 (dd, 2H, J=8.8 and JH–F=5.5, H-2%,6%)
3.62 (s, 3H, CH3), 3.85 (s, 2H, CH2), 6.05 (s, 2H, O–CH2–O), 6.83 and 6.84 (2s, 2H, H-3 and H-6),
7.73 and 7.88 (2d, 4H, J=8.0, Ar)
3.46 (s, 2H, CH2-5), 6.03 (s, 2H, OCH2O), 6.63 (s, 1H, H-9), 6.85 (s, 1H, H-6), 7.38–7.61 (m, 5H,
Ar), 8.56 (br s, 1H, NH)
3.44 (s, 5H, CH2-5 and CH3), 6.03 (s, 2H, OCH2O), 6.63 (s, 1H, H-9), 6.86 (s, 1H, H-6), 7.42–7.64
(m, 5H, Ar)
2.58 (s, 3H, CH3), 3.56 (s, 2H, CH2-5), 6.06 (s, 2H, OCH2O), 6.67 (s, 1H, H-9), 6.90 (s, 1H, H-6),
7.42–7.71 (m, 5H, Ar)
3.45 (s, 2H, CH2-5), 3.78 (br s, 2H, NH2), 6.03 (s, 2H, OCH2O), 6.68 (s, 1H, H-9), 6.78 (dd, 1H,
J=7.6 and 1.5, H-6%), 6.82 (s, 1H, H-6), 6.89 (dd, 1H, J=7.6 and 1.5, H-4%), 6.96 (t, 1H, J=1.9,
H-2%), 7.19 (t, 1H, J=7.6, H-5%), 8.50 (br s, 1H, NH)
11j
12a
12b
12c
12d
12e
12f
12g
12h
12i
12j
228–230 (B)
236–238 (B)
308–310 (B)
205–207 (C)
206–208 (C)
265–267 (C)
51
38
64
68
56
53
3.42 (s, 2H, CH2-5), 3.82 (bs s, 2H, NH2), 6.01 (s, 2H, OCH2O), 6.70 (s, 1H, H-9), 6.81 (s, 1H, H-6),
6.67 and 7.40 (2d, 4H, J=8.5, Ar), 8.57 (br s, 1H, NH)
3.37 (s, 5H, CH2-5 and CH3), 3.88 (br s, 2H, NH2), 6.03 (s, 2H, OCH2O), 6.70 (s, 1H, H-9), 6.83
(s, 1H, H-6), 6.68 and 7.43 (2d, 4H, J=8.7, Ar)
3.40 (s, 2H, CH2-5), 6.09 (s, 2H, OCH2O), 6.60 (s, 1H, H-9), 7.06 (s, 1H, H-6), 6.81 and 7.36
(2d, 4H, J=8.2, Ar), 9.88 (br s, 1H, NH), 10.74 (br s, 1H, OH)
3.45 (s, 2H, CH2-5), 3.87 (s, 3H, OCH3), 6.04 (s, 2H, OCH2O), 6.66 (s, 1H, H-9), 6.83 (s, 1H, H-6),
6.94 and 7.55 (2d, 4H, J=8.9, Ar), 8.32 (br s, 1H, NH)
3.46 (s, 2H, CH2-5), 6.04 (s, 2H, OCH2O), 6.61 (s, 1H, H-9), 6.84 (s, 1H, H-6), 7.11 (dd, 2H, J=8.8
and JH–F=8.5, H-3%,5%), 7.60 (dd, 2H, J=8.8 and JH–F=5.5, H-2%,6%), 8.48 (br s, 1H, NH)
3.48 (s, 2H, CH2-5), 6.05 (s, 2H, OCH2O), 6.58 (s, 1H, H-9), 6.85 (s, 1H, H-6), 7.68 and 7.74
(2d, 4H, J=8.3, Ar), 8.60 (br s, 1H, NH)
a Crystallisation solvent: A, ethanol; B, ethyl acetate; C, methanol.
b 1H NMR spectra were recorded in CDCl3, and in DMSO-d6 for 12g.
matography was performed on Merck silica gel 60
pressure and the residue was dissolved in dry CH2Cl2
(20 ml) and used immediately in the following step.
(70–230 mesh or 230–240 mesh for compounds 11d–e
1
and 12d–f). H NMR spectra were recorded using a
Varian Gemini-300 spectrometer in the indicated sol-
vents. Chemical shifts were expressed in ppm and cou-
pling constants (J) in Hz. All exchangeable protons
were confirmed by addition of D2O. The data are
collected in Table 4. The preparation of 3%- and 4%-tri-
fluoroacetamidobenzoic acid was accomplished follow-
ing a reported procedure [40]. The starting acyl chlo-
rides 10a and 10h–j are commercially available whereas
10d, 10e, and 10g were prepared as follows.
5.1.2. General procedure for methyl 2-aroyl-4,5-methyl-
enedioxyphenylacetate (11a, 11d–e, 11g–j)
To a cooled (0–5°C) and stirred solution of methyl
3,4-methylenedioxyphenylacetate (9) (1.94 g, 10 mmol)
in CH2Cl2 (20 ml) were added 16.5 ml of 0.1 M tin(IV)
chloride (16.5 mmol) in CH2Cl2 and the appropriate
acyl chloride 10 (13 mmol) in the same solvent (20 ml).
The ice-bath was removed and the mixture was stirred
at 20°C overnight, then poured into water and the
product was isolated in ether and dried over Na2SO4.
The solvent was removed under reduced pressure and
the residue was subjected to chromatography eluting
with diethyl ether–light petroleum (6:4).
5.1.1. General procedure for acyl chlorides 10d–e, 10g
The appropriate benzoic acid (13 mmol) was heated
with thionyl chloride (11 ml, 150 mmol) under reflux
for 2–3 h. Excess SOCl2 was evaporated under reduced

A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
185
5.1.3. General procedure for 1-aryl-3,5-dihydro-7,8-
methylenedioxy-4H-2,3-benzodiazepin-4-ones (12a–b,
12d–j)
5.2.2. Maximal electroshock seizure test in Swiss mice
Electrical stimuli were applied via ear-clip electrodes
to Swiss mice (rectangular constant current impulses,
amplitude 50 mA, width 20 ms, frequency 35 Hz,
duration 400 ms) according to the method of Swinyard
et al. [42]. Abolition of tonic hindlimb extension after
drug treatment was considered as the endpoint of pro-
tection. In general, the dose–response curves were esti-
mated by testing four to five doses using eight to ten
mice for each dose.
The appropriate methyl 2-aroyl-4,5-methylenedioxy-
phenylacetate (11) (3 mmol) was refluxed for 6–7 h
with hydrazine hydrate (10 or 30 mmol for 12d, 12e,
and 12g) in EtOH (50 ml) or monomethylhydrazine (12
or 36 mmol for 12f) in dry toluene (50 ml). After
cooling, the solvent was removed under reduced
pressure and the residue purified by column
chromatography using EtOAc–light petroleum (70:30)
for 12d–g and diethyl ether/light petroleum (80/20) for
12a,b and 12h–j as eluant.
5.2.3. PTZ-induced seizures test in Swiss mice
Male Swiss mice (20–26 g, 42–48 days old) were
purchased from Charles River (Calco, Como, Italy) and
were pretreated with vehicle or drug 45 min before the
subcutaneous (s.c.) administration of pentylenetetra-
zole. For systemic injections, all 2,3-benzodiazepines
were given i.p. (0.1 ml/10 g of body weight of the
mouse) as a freshly prepared solution in 50% DMSO
and 50% sterile saline (0.9% NaCl). The convulsive
dose 97 (CD97) of pentylenetetrazole (85 mg/kg) was
applied and the animals observed for 30 min. A
threshold convulsion was an episode of clonic spasms
lasting for at least 5 s. Absence of this threshold
convulsion over 30 min indicated that the tested sub-
stance had the ability to elevate pentylenetetrazole
seizure threshold [43].
5.1.4. 3-Acetyl-3,5-dihydro-7,8-methylenedioxy-
1-phenyl-4H-2,3-benzodiazepin-4-one (12c)
To a solution of 12a (0.28 g, 1 mmol) in CHCl3 (30
ml) and Et3N (5 ml) which was cooled while stirring,
Ac2O (5 ml) was added dropwise. The mixture reaction
was stirred at room temperature for 1–2 h, then poured
into water, extracted with CHCl3, and dried over
Na2SO4. Evaporation of the solvent under reduced
pressure afforded a crude product, which was purified
by treatment with EtOH–light petroleum.
5.2. Pharmacology
5.2.1. Test of anticon6ulsant acti6ity against audiogenic
seizures in DBA/2 mice
5.2.4. AMPA-induced seizures in DBA/2 mice
Seizures were also induced by i.c.v. injection of
AMPA. The CD50 of AMPA for clonus was 1.76
(1.06–3.07) whilst that for tonus was 2.90 (1.83–4.58)
nmol. The CD97 (the dose which induced convulsions in
97% of the mice) values of 9.7 nmol for all-limb clonic
seizures (latency 1.390.3 min) and 11.7 nmol for fore-
limb tonic seizures (latency 1.9 9 0.4 min), respectively
were used to assess the effects of some 2,3-benzodi-
azepines on the convulsant properties of AMPA. For
i.c.v. injection, mice were anaesthetised with diethyl
ether and injections were made in the left or right
lateral ventricle (coordinates: 1 mm posterior and 1 mm
lateral to the bregma; depth 2.4 mm) using a 10 ml
Hamilton microsyringe (type 701N) fitted with a nylon
cuff on the needle as previously described [44]. Injec-
tions of drugs by this procedure led to an uniform
distribution throughout the ventricular system within
10 min. The animals were placed singly in a 30×30×
30 cm box and the observation time was 30 min after
the administration of AMPA.
Experiments were performed with DBA/2 mice which
are genetically susceptible to sound-induced seizures
[41]. DBA/2 mice (8–12 g, 22–25 days old) were pur-
chased from Charles River (Calco, Como, Italy).
Groups of ten mice of either sex were exposed to
auditory stimulation 30 min following administration
of vehicle or drugs. The compounds were given i.p. (0.1
ml/10 g body weight of the mouse) as a freshly pre-
pared solution in 50% DMSO and 50% sterile saline
(0.9% NaCl). Individual mice were placed under a
hemispheric perspex dome (diameter 58 cm) and 60 s
was allowed for habituation and assessment of locomo-
tor activity. Auditory stimulation (12–16 kHz, 109 dB)
was applied for 60 s or until tonic extension occurred,
and induced a sequential seizure response in control
DBA/2 mice, consisting of an early wild running phase,
followed by generalised myoclonus and tonic flexion
and extension sometimes followed by respiratory arrest.
The control (vehicle-treated) and drug-treated mice
were scored for latency to and incidence of the different
phases of the seizures [27]. The time course of the
anticonvulsant action of 2a, 12a, and 1 was determined
following the administration of 50 mmol/kg of 2,3-ben-
zodiazepines to groups of 10 mice for each time. The
animals were tested for sound-induced seizure responses
at 5 to 240 min after drug administration.
5.2.5. Pretreatment with aniracetam
The i.c.v. microinjection of aniracetam was per-
formed according to experimental procedures previ-
ously described for AMPA microinjection [44]. The
dose of aniracetam (50 nmol i.c.v.) was administered 60

186
A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
min before auditory stimulation or 30 min before
References
2,3-benzodiazepines in DBA/2 mice.
[1] T.P. Hicks, The role of glutamate and its receptors in synaptic
transmission, in: T.W. Stone (Ed.), CNS Neurotransmitters and
5.2.6. Treatment with flumazenil
Neuromodulators: Glutamate, CRC Press, New York, 1995, pp.
201–215.
[2] P. Conn, J.-P. Pin, Pharmacology and functions of metabotropic
glutamate receptors, Annu. Rev. Pharmacol. Toxicol. 37 (1997)
205.
[3] P. Krogsgaard-Larsen, J.J. Hansen (Eds.), Excitatory Amino
Acid Receptors: Design of Agonists and Antagonists, Ellis Hor-
wood, Chichester, UK, 1992.
[4] D.T. Monaghan, R.J. Wenthold (Eds.), The Ionotropic Gluta-
mate Receptors, Humana Press, New Jersey, 1997.
[5] C. Ikonomidou, L. Turski, Glutamate in neurodegenerative dis-
orders, in: T.W. Stone (Ed.), CNS Neurotransmitters and Neu-
romodulators: Glutamate, CRC Press, New York, 1995, pp.
253–266.
[6] W. Danysz, C.G. Parsons, I. Bresink, G. Quack, Glutamate in
CNS disorders, Drug News Perspect. 8 (1995) 261–277.
[7] P. Krogsgaard-Larsen, B. Ebert, T.M. Lund, H. Bra¨uner-Os-
borne, K.A. Slok, T.N. Johansen, L. Brehn, U. Madsen, Design
of excitatory amino acid receptor agonists, partial agonists and
antagonists: ibotenic acid as a key lead structure, Eur. J. Med.
Chem. 31 (1996) 515–537.
As previously demonstrated [27], flumazenil,
administered i.p. at 8.24 or 24.72 mmol/kg, is not
itself convulsant and did not significantly modify the
phases of the audiogenic seizure response in DBA/2
mice, but antagonised the anticonvulsant action of
classical 1,4-benzodiazepines. Flumazenil was admin-
istered i.p. concomitantly to 2,3-benzodiazepine deriv-
atives 12a, 12e and 1.
5.2.7. Membrane preparation and [3H]flumazenil and
other binding studies
The binding affinities were obtained by the meth-
ods previously described by us [28].
5.2.8. Effects on motor mo6ements
Male Swiss mice (20–26 g, 48–54 days old) were
purchased from Charles River (Calco, Como, Italy).
Groups of ten mice were trained to do coordinated
motor movements continuously for 2 min on a ro-
tarod, 3 cm diameter, at 8 r.p.m. (U. Basile, Come-
rio, Varese, Italy). Impairment of coordinated motor
movements was defined as inability of the mice to
remain on the rotarod for a 2 min test period [45].
The ability of the mice to remain on the rotarod was
tested 30 min after administration of various 2,3-ben-
zodiazepines.
[8] B.S. Meldrum (Ed.), Excitatory Amino Acid Antagonists, Black-
well, London, 1991.
[9] M.A. Rogawski, The NMDA receptor, NMDA antagonists and
epilepsy therapy, Drugs 44 (1992) 279–292.
[10] W. Koek, J.H. Woods, G.D. Winger, MK-801, a proposed
non-competitive antagonist of excitatory amino acid neurotrans-
mission, produces phencyclidine-like behavioral effects in pi-
geons, rats and rhesus monkeys, J. Pharmacol. Exp. Ther. 245
(1988) 969–974.
[11] P.L. Herting, Competitive NMDA antagonists: results from
animal models and first results in human, J. Neurochem. (Suppl.)
61 (1993) S283C.
[12] R.G.M. Morris, E. Anderson, G.S. Lynch, M. Baudry, Selective
impairment of learning and blockade of long-term potentiation
by an N-methyl-D-aspartate receptor antagonist, AP5, Nature
319 (1986) 774–776.
5.2.9. Statistical analysis
[13] P. Desos, J.M. Lepagnol, P. Morain, P. Lestage, A.A. Cordi,
Structure–activity relationships in a series of 2(1H)-quinolones
bearing different function in the 3-position: 6,7-dichloro-2(1H)-
oxoquinoline-3-phosphonic acid, a new potent and selective
AMPA/kainate antagonist with neuroprotective properties, J.
Med. Chem. 39 (1996) 197–206.
[14] Z. Lin, P.K. Kadaba, Molecular targets for the rational design
of antiepileptic drugs and related neuroprotective agents, Med.
Res. Rev. 17 (1997) 537–572.
[15] M.G. Sheardown, E.O. Nielsen, A.J. Hansen, P. Jacobsen, T.
Honore, 2,3-Dihydroxy-6-nitro-7-sulfamoylbenzo[ f ]quinoxaline:
a neuroprotectant for cerebral ischemia, Science 247 (1990)
571–574.
[16] A.M. Buchan, H. Lesiuk, K.A. Barnes, H. Li, Z.G. Huang, K.E.
Smith, D. Xue, AMPA antagonists: do they hold more promise
for clinical stroke trials than NMDA antagonists?, Stroke
(Suppl. I) 24 (1993) 148–152.
Statistical comparisons between groups of control
and drug-treated animals were made using Fisher’s
exact probability test (incidence of the seizure phases)
or ANOVA followed by post-hoc Dunnett’s t-test
(rectal temperatures). The ED50 values of each phase
of the audiogenic seizure or seizures induced by MES,
PTZ or AMPA were determined for each dose of
compound administered and dose–response curves
were fitted using a computer program by the method
of Litchfield and Wilcoxon [46]. The TD50 values
were estimated using the method of Litchfield and
Wilcoxon [46]. For the binding experiments IC50 val-
ues for the [3H]flumazenil or other ligands displace-
ment were determined by the non-linear curve-fitting
program based on LIGAND [47].
[17] J. Ohmori, S. Sakamoto, H. Kubota, M. Shimizu-Sasamata, M.
Okada, S. Kawasaki, K. Nikada, J. Togami, T. Furuya, K.
Murase, 6-(1H-Imidazol-1-yl)-7-nitro-2,3(1H,4H)-quinoxaline-
dione hydrochloride (YM90K) and related compounds: struc-
ture–activity relationships for the AMPA-type non-NMDA
receptor, J. Med. Chem. 37 (1994) 467–475.
Acknowledgements
[18] S.D. Donevan, M.A. Rogawski, GYKI 52466, a 2,3-benzodi-
azepine, is a highly selective, non-competitive antagonist of
AMPA/kainate receptor responses, Neuron 10 (1993) 51–59.
This work was supported by CNR, 97.02578.CT03
(Rome, Italy).

A. De Sarro et al. / Il Farmaco 54 (1999) 178–187
187
[19] M.A. Rogawski, Therapeutic potential of excitatory amino acid
antagonists: channel blockers and 2,3-benzodiazepines, Trends
Pharmacol. Sci. 14 (1993) 325–331.
for the oxidative removal of benzyl ethers: the oxidation of
benzyl ethers to benzoate by ruthenium tetraoxide, Tetrahedron
Lett. 24 (1983) 3829–3830.
[20] A.G. Chapman, S.E. Smith, B.S. Meldrum, The anticonvulsant
effect of the non-NMDA antagonists, NBQX and GYKI 52466
in mice, Epilepsy Res. 9 (1991) 92–96.
[34] A.G. Chapman, B.S. Meldrum, Non-competitive N-methyl-D-as-
partate antagonists protect against sound-induced seizures in
DBA/2 mice, Eur. J. Pharmacol. 166 (1989) 201–211.
[35] G.B. De Sarro, A. De Sarro, Anticonvulsant properties of
non-competitive antagonists of NMDA receptor in genetically
epilepsy prone rats. Comparison with CPPene, Neuropharmacol-
ogy 32 (1993) 51–58.
[21] S.E. Smith, N. Durmuller, B.S. Meldrum, The non-N-methyl-D-
aspartate receptor antagonists, GYKI 52466 and NBQX are
anticonvulsant in two animal models of reflex epilepsy, Eur. J.
Pharmacol. 201 (1991) 179–183.
[22] A.G. Chapman, Z. Al-Zubaidy, B.S. Meldrum, Aniracetam re-
verses the anticonvulsant action of NBQX and GYKI 52466 in
DBA/ 2 mice, Eur. J. Pharmacol. 231 (1993) 301–303.
[23] S.D. Donevan, S.I. Yamaguchi, M.A. Rogawski, Non-N-
[36] W.C. McEntee, T.H. Crook, Glutamate, its role in learning,
memory, and the aging brain, Psychopharmacology 111 (1993)
391–401.
[37] C. Chiamulera, S. Costa, A. Reggiani, Effect of NMDA- and
strychnine-insensitive glycine site antagonists on NMDA-medi-
ated convulsions and learning, Psychopharmacology 102 (1990)
551–552.
[38] T.V.P. Bliss, G.L. Collingridge, A synaptic model of memory:
long-term potentiation in the hippocampus, Nature 361 (1993)
31–39.
[39] A.V. Paternain, M. Morales, J. Lerma, Selective antagonism of
AMPA receptors unmasks kainate receptor–mediate responses
in hippocampal neurones, Neuron 14 (1995) 185–189.
[40] J.W. Daly, W.L. Padgett, M.T. Shamin, Analogues of 1,3-
dipropyl-8-phenylxanthines: enhancement of selectivity at A1-
adenosine receptors by aryl substituents, J. Med. Chem. 29
(1986) 1520–1524.
[41] R.L. Collins, Audiogenic seizures, in: P. Purpura, J.K. Penry, D.
Tower, D.M. Woodbury, R. Walter (Eds.), Experimental Models
of Epilepsy, Raven Press, New York, 1972, pp. 347–372.
[42] E.A. Swinyard, W.C. Brown, L.S. Goodman, Comparative as-
says of antiepileptic in mice and rats, J. Pharmacol. Exp. Ther.
106 (1952) 319–330.
[43] E.A. Swinyard, J.M. Woodhead, Experimental detection, quan-
tification and evaluation of anticonvulsants, in: D.M. Wood-
bury, J.K. Penry, C.E. Pippenger (Eds.), Epileptic Drugs, Raven
Press, New York, 1982, pp. 111–126.
[44] G.B. De Sarro, E. Ongini, R. Bertorelli, U. Aguglia, A. De
Sarro, Excitatory amino acid neurotransmission through both
NMDA and non-NMDA receptors is involved in the anticonvul-
sant activity of felbamate in DBA/2 mice, Eur. J. Pharmacol. 262
(1994) 11–19.
methyl-D-aspartate receptor antagonism by 3-N-substituted 2,3-
benzodiazepines: relationship to anticonvulsant activity, J.
Pharmac. Exp. Ther. 271 (1994) 25–29.
[24] S.E. Smith, B.S. Meldrum, Cerebroprotective effect of a non-N-
methyl-D-aspartate antagonist, GYKI 52466, after focal ischemia
in the rat, Stroke 23 (1992) 861–864.
[25] B. Arvin, D. Lekieffre, J.L. Graham, C. Moncada, A.G. Chap-
man, B.S. Meldrum, Effect of the non-NMDA receptor antago-
nist GYKI 52466 on the microdialyzate and tissue
concentrations of amino acids following transient forebrain is-
chemia, J. Neurochem. 62 (1994) 1458–1467.
[26] I. Tarnawa, S. Farkas, P. Berzsenyi, A. Pataki, F. Andrasi,
Electrophysiological studies with a 2,3-benzodiazepine muscle
relaxant: GYKI 52466, Eur. J. Pharmacol. 67 (1989) 193–199.
[27] G. De Sarro, A Chimirri, A. De Sarro, R. Gitto, S. Grasso, P.
Giusti, A.G. Chapman, GYKI 52466 and related 2,3-benzodi-
azepines as anticonvulsant agents in DBA/2 mice, Eur. J. Phar-
macol. 294 (1995) 411–422.
[28] A. Chimirri, G. De Sarro, A. De Sarro, R. Gitto, S. Grasso, S.
Quartarone, M. Zappala`, P. Giusti, V. Libri, A. Constanti, A.G.
Chapman,
1-Aryl-3,5-dihydro-4H-2,3-benzodiazepin-4-ones:
novel AMPA receptor antagonists, J. Med. Chem. 40 (1997)
1258–1269.
[29] A.G. Chapman, M.J. Croucher, B.S. Meldrum, Evaluation of
anticonvulsant drugs in DBA/2 mice with sound-induced
seizures, Arzneim. Forsch. 34 (1984) 1261–1264.
[30] F.L. Engstrom, D.M. Woodbury, Seizure susceptibility in DBA/
2 and C57 mice: the effects of various convulsants, Epilepsia 29
(1988) 389–395.
[45] N.W. Dunham, T.S. Miya, A note on a simple apparatus for
detecting neurological deficit in rats and mice, J. Am. Pharm.
Ass. 46 (1957) 208–209.
[31] I. Ito, S. Tanabe, A. Kohda, H. Sugiyama, Allosteric potentia-
tion of quisqualate receptors by nootropic drug aniracetam, J.
Physiol. (London) 424 (1990) 533–543.
[46] J.T. Litchfield, F. Wilcoxon, A simplified method of evaluating
dose–effects experiments, J. Pharmac. Exp. Ther. 96 (1949)
99–113.
[47] G.I. McPherson, LIGAND (release 2.0), Elsevier Biosoft, Cam-
bridge, 1987.
[32] A. De Sarro, G. De Sarro, R. Gitto, S. Grasso, N. Micale, S.
Quartarone, M. Zappala`, 7,8-Methylenedioxy-4H-2,3-benzodi-
azepin-4-ones as novel AMPA receptor antagonists, Bioorg.
Med. Chem. Lett. 8 (1998) 971–976.
[33] P.F. Schuda, M.B. Cichowicz, M.R. Heimann, A facile method
.