3-Hydroxy-1H-quinazoline-2,4-dione derivatives as new antagonists at ionotropic glutamate receptors: Molecular modeling and pharmacological studies

Article abstract of DOI:10.1016/j.ejmech.2012.05.036

Based on our 3-hydroxy-7-chloroquinazoline-2,4-dione derivatives, previously reported as antagonists at ionotropic glutamate receptors, we synthesized new 3-hydroxyquinazoline-2,4-diones bearing a trifluoromethyl group at the 7-position and different grou

Full text of DOI:10.1016/j.ejmech.2012.05.036

European Journal of Medicinal Chemistry 54 (2012) 470e482
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
3-Hydroxy-1H-quinazoline-2,4-dione derivatives as new antagonists at ionotropic
glutamate receptors: Molecular modeling and pharmacological studies
,
a
a
a
a
Vittoria Colotta a , Ombretta Lenzi , Daniela Catarzi , Flavia Varano , Lucia Squarcialupi ,
Chiara Costagli b, Alessandro Galli b, Carla Ghelardini b, Anna Maria Pugliese b, Giovanna Maraula b,
Elisabetta Coppi b, Domenico E. Pellegrini-Giampietro b, Felicita Pedata b, Davide Sabbadin c,
Stefano Moro c
*
a Dipartimento di Scienze Farmaceutiche, Laboratorio di Progettazione, Sintesi e Studio di Eterocicli Biologicamente Attivi, Universita’ di Firenze, Polo Scientifico, Via Ugo Schiff 6,
50019 Sesto Fiorentino, FI, Italy
b Dipartimento di Farmacologia Preclinica e Clinica, Università degli Studi di Firenze, Viale Pieraccini 6, 50139 Firenze, Italy
c Molecular Modeling Section (MMS), Dipartimento di Scienze del Farmaco, Università degli Studi di Padova, via Marzolo 5, 35131 Padova, Italy
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
< Heterocyclic rings at the 6-position
gave good AMPA receptor affinity.
< Derivative 8 (R6 ¼ 1,2,4-triazol-4-yl)
showed high AMPA receptor affinity.
< Derivative
3
showed the best
affinity and selectivity for the kai-
nate receptor.
< Derivatives 3 and 8 were effective
neuroprotective agents in rat
models of cerebral ischemia.
< Compound
8 showed anticonvul-
sant effect in pentylenetetrazole-
induced convulsions.
a r t i c l e i n f o
a b s t r a c t
Article history:
Based on our 3-hydroxy-7-chloroquinazoline-2,4-dione derivatives, previously reported as antagonists at
ionotropic glutamate receptors, we synthesized new 3-hydroxyquinazoline-2,4-diones bearing a tri-
fluoromethyl group at the 7-position and different groups at position 6. Glycine/NMDA, AMPA and kainate
receptor binding data showed that the 7-trifluoromethyl residue increased AMPA and kainate receptor
affinity and selectivity, with respect to the 7-chlorine atom. Among the probed 6-substituents, the 6-(1,2,4-
triazol-4-yl) group (compound 8) was the most advantageous for AMPA receptor affinity and selectivity.
Derivative 8 demonstrated to be effective in decreasing neuronal damage produced by oxygen and glucose
deprivation in organotypic rat hippocampal slices and also showed anticonvulsant effects in
pentylenetetrazole-induced convulsions. The previously reported kainate receptor antagonist 6-(2-
carboxybenzoyl)-amino-7-chloro-3-hydroxyquinazoline-2,4-dione 3 prevented the failure of neurotrans-
mission induced by oxygen and glucose deprivation in the CA1 region of rat hippocampal slices.
Ó 2012 Elsevier Masson SAS. All rights reserved.
Received 20 December 2011
Received in revised form
21 May 2012
Accepted 23 May 2012
Available online 4 June 2012
Keywords:
Ionotropic glutamate receptors
3-Hydroxyquinazoline-2,4-diones
AMPA receptor antagonists
Kainate receptor antagonists
Neuroprotective agents
Ligandereceptor modeling studies
Abbreviations: Glu, glutamate; iGluRs, ionotropic glutamate receptors; NMDA, N-methyl-D-aspartate; AMPA, (S)-2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic
acid; KA, kainate; DCKA, 5,7-dichlorokynurenic acid; NBQX, 2,3-dihydroxy-6-nitro-7-sulphamoyl-benzo[f]quinoxaline; MK-801, (þ)-5-methyl-10,11-dihydro-5H-benzo[a,d]
cyclohepten-5,10-iminemaleate; PTZ, pentylenetetrazole; OGD, oxygen and glucose deprivation; PI, propidium iodide; fepsp, field excitatory postsynaptic potential; AD,
anoxic depolarization; aCSF, artificial cerebrospinal fluid.
* Corresponding author. Tel.: þ39 055 4573731; fax: þ39 055 4573780.
E-mail address: vittoria.colotta@unifi.it (V. Colotta).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2012.05.036

V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
471
1. Introduction
COOH
3 R6=
H
N
N
R6=
2
N
Glutamate (Glu) is the major excitatory neurotransmitter in
the mammalian central nervous system and plays key roles in
regulating many physiological processes through activation of
metabotropic (mGluRs) and ionotropic receptors (iGluRs) [1,2].
N
KA selectivity O
AMPA selectivity
O
N
1 R6= H
Gly/NMDA selectivity
The iGluRs include three families classified as N-methyl-D-aspar-
OH
O
R6
tate (NMDA), (S)-2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)
propionic acid (AMPA) and kainate (KA) receptors. The NMDA
receptor complex possesses different binding sites including the
glutamate co-agonist glycine binding site (Gly/NMDA) [2]. To date,
seven NMDA receptor subunits (GluN1, GluN2A-2D, and GluN3A
and GluN3B), four AMPA receptor subunits (GluA1-GluA4) and
five KA receptor subunits (GluK1-GluK5) have been cloned and
characterized [2]. It is well known that an overstimulation of
iGluRs induce an increase of intracellular Ca2þ concentrations,
release of Kþ into the extracellular space and cell swelling due to
the passive movement of water with Naþ influx [3]. The massive
increase of intracellular Ca2þ triggers numerous deleterious
processes, including free radical formation and membrane
degradation, mitochondrial dysfunction, inflammation, DNA-
damage and apoptosis. These processes lead to neuronal death
and are implicated in many diseases [1] such as Alzheimer’s [4]
and Parkinson’s diseases [5], amyotrophic lateral sclerosis [6],
epilepsy [1,7] and cerebral ischemia [8].
Due to the iGluR-mediated excitotoxicity, several compounds
acting as iGluR antagonists demonstrated beneficial effects against
the above cited pathologies [1,4,5].
It has to be noted that AMPA and KA receptor antagonists seem
to possess greater therapeutic potential than NMDA antagonists
because these latter often show adverse effects such as hallucina-
tions, agitation, ataxia and catatonia [9].
In the last decade, a part of our research has been focused on the
study of competitive and noncompetitive iGluR antagonists,
belonging to different heteroaromatic systems [10e21]. Among
them, the 3-hydroxyquinazoline-2,4-dione was disclosed as a useful
scaffold to obtain selective iGluR antagonists [15,18], since many
derivatives of this series showed affinity for AMPA and KA receptors
or for the Gly/NMDA site. The 3-hydroxyquinazolin-2,4-diones
possess the most important structural requirements for Gly/NMDA
and AMPA or KA receptor recognition [22,23]: a flat hydrophobic
area represented by the fused benzo ring; a NH hydrogen-bond
Cl
N
H
Fig. 1. Previously reported 7-chloro-3-hydroxyquinazoline-2,4-dione derivatives.
which inprevious studies turned out tobe profitable for the AMPA or
KA receptor affinity and selectivity, i.e. heterocyclic rings or the 2-
carboxybenzoylammino group, respectively.
2. Chemistry
The target compounds 4e11 were synthesized as described
in Schemes
1 and 2. The 7-trifluoromethyl-1,2-dihydro-3,1-
benzoxazine-2,4-dione 12 [21] was reacted with O-benzylhydrox-
ylamine in refluxing ethanol to give the 2-amino-N-benzyloxy-4-
trifluoromethylbenzamide 13. Cyclization of 13 with triphosgene
afforded the 3-benzyloxy-7-trifluoromethylquinazoline-2,4-dione
14 which was debenzylated with 48% HBr in glacial acetic acid to
give the desired 3-hydroxy derivative 4. This latter was transformed
into the corresponding 3-acetate 15 by refluxing in acetic anhy-
dride. Nitration of 15 with 90% HNO3 afforded the corresponding
6-nitro compound 16 that was transformed into the 6-amino
derivative 17. Reaction of 17 with phthalic anhydride afforded the
desired 3-hydroxy-6-(2-carboxybenzoylamino)-quinazoline-2,4-
dione 7. The 3-hydroxy-6-nitroquinazoline-2,4-dione 5, its corre-
sponding 6-amino derivative 6 and the 6-heteroaryl substituted
compounds 8e11 were synthesized as shown in Scheme 2. Deriv-
ative 12 was reacted with O-methylhydroxylamine to yield the
2-amino-3-methoxy-4-trifluoromethylbenzamide 18 which was
cyclized with triphosgene to give derivative 19. Nitration of 19
afforded 6-nitro-3-methoxy derivative 20 which was desmethy-
lated with 48% hydrobromic acid in glacial acetic acid to give the
3-hydroxy-6-nitroquinazoline-2,4-dione 5. Catalytic reduction (Pd/
C) of compound 5 gave the 6-amino derivative 6.
donor that binds a proton acceptor of the receptor; a d-negatively
charged moiety, represented by both the 2-carbonyl group and the
3-hydroxy substituent, able to form a hydrogen bond with a cationic
hydrogen bond donor receptor site. All the previously described 3-
hydroxyquinazoline-2,4-dione derivatives (Fig. 1) are substituted
at the 7-position with a chlorine atom that is combined with
different substituents at the 6-position. Structureeaffinity rela-
tionship (SAR) studies suggested that the nature of the 6-
substituents was critical for the selectivity towards the different
iGluRs. In particular, R6 ¼ H gave a selective Gly/NMDA antagonist
(compound 1), while heterocyclic rings gave AMPA receptor antag-
onists, the most advantageous was the 6-(1,2,4-triazol-4-yl) moiety
(derivative 2) [15]. Instead, the presence of a 6-(2-carboxybenzoyl)-
amino group on the 3-hydroxyquinazoline-2,4-dione scaffold
afforded a selective KA receptor antagonist (compound 3) [18]. On
the basis of these results, we decided to continue the study of this
class of compounds to further investigate the SARs and to obtain
more potent AMPA and/or KA receptor antagonists. Thus, we plan-
ned the synthesis of new 6-substituted 3-hydroxyquinazoline-2,4-
dione derivatives (Fig. 2, compounds 4e11) in which the 7-
chlorine atom was replaced with a 7-trifluoromethyl residue that
was thought to increase affinity towards AMPA and KA receptors
[12,16]. The 6-position of the new derivatives bears the substituents
To prepare the 7-trifluoromethyl-3-hydroxy-6-(1,2,4-triazol-4-
yl)-quinazoline-2,4-dione 8, compound 20 was reduced to the
corresponding 6-amino compound 21 that was transformed into 22
by reaction with diformylhydrazine. Different attempts to turn the
3-O-methyl derivative 22 into the corresponding 3-hydroxy deriv-
ative 8 failed. In fact, treatment of 22 with boron tribromide at room
temperature for three to four days left compound 22 unmodified,
while reaction with 48% aqueous hydrobromic acid in glacial acetic
acid at 100 ꢀC afforded the 6-amino compound 6. Thus, the desired
derivative
8
was obtained by reacting compound 6 with
O
R6
OH
O
N
F3C
N
H
4-11
R6= H, NO2, NH2, Heterocycle,
2-COOH-C6H4-CONH
Fig. 2. Currently reported 3-hydroxyquinazoline-2,4-dione derivatives.

472
V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
O
O
O
O
O
O
a
O
N
b
c
N
N
H
NH2
F3C
N
H
O
F3C
N
H
F3C
13
14
12
O
O
COOH
HN
OR
O
O
O2N
F3C
OCOCH3
O
N
e
g
OH
O
N
F3C
N
H
N
H
O
F3C
N
H
4
R= H
16 R= NO2
17 R= NH2
f
d
15 R= COCH3
7
Scheme 1. Reagents and conditions. (a) O-Benzylhydroxylamine hydrochloride, NEt3, EtOH, reflux; (b) triphosgene, NEt3, r.t.; c) 48% HBr, glacial AcOH, reflux; d) Ac2O, reflux; e) 90%
HNO3, from ꢁ10 to 0 ꢀC; f) H2, Pd/C, EtOH; g) (i) phthalic anhydride, NaOAc, 60 ꢀC; (ii) 0.25 M NaOH, r.t.; (iii) 6 N HCl.
diformylhydrazine. Reaction of 6 with 2,5-diethoxytetrahydrofuran
or 2,5-dimethoxytetrahydrofuran-3-carbaldehyde in glacial acetic
acid at 90 ꢀC gave, respectively, the 6-(pyrrol-1-yl)- derivative 9 and
the 6-(3-formylpyrrol-1-yl)- substituted compound 10. This latter
was oxidized with potassium permanganate in water/acetone to
give the 6-(3-carboxypyrrol-1-yl)- derivative 11.
were evaluated for their binding at AMPA, Gly/NMDA and high-
affinity KA receptors in rat cortical membranes. The binding data
are shown in Table 1, together with those of the previously reported
7-chloroquinazolines 1e3, 23e25 [15,18], included as reference
compounds. Table
1
also displays the affinities of 5,7-
dichlorokyunurenic acid (DCKA) and of 2,3-dihydroxy-6-nitro-7-
sulphamoyl-benzo[f]quinoxaline (NBQX), well-known antagonists
of the Gly/NMDA and AMPA receptors, respectively. The obtained
results were coherent with our expectations. Replacement of the
7-chlorine atom with the 7-trifluoromethyl group positively affected
the binding at the AMPA receptor. In fact, derivatives 4, 5, 8,10 and 11
possess an increased AMPA affinity with respect to those of the
corresponding 7-cloro derivatives 1, 23, 2, 24 and 25 [15,18].
Compound 4, lacking the 6-substituent, has a good affinity for the
3. Results and discussion
3.1. Structureeaffinity relationship
The newlysynthesized 7-trifluoromethyl-3-hydroxyquinazoline-
2,4-dione derivatives 4e11 and the synthetic intermediate 22, 3-O-
methyl analog of the 6-(1,2,4-triazolyl)-substituted derivative 8,
Scheme 2. Reagents and conditions. (a) O-Methylhydroxylamine hydrochloride, NEt3, EtOH, reflux; (b) triphosgene, NEt3, r.t.; c) 90% HNO3, from ꢁ10 to 0 ꢀC; d) H2, Pd/C, EtOH; e)
diformylhydrazine, Me3SiCl, anhydrous pyridine, 100 ꢀC; f) 48% HBr, glacial AcOH, reflux; g) 2,5-diethoxytetrahydrofuran, glacial AcOH; h) 2,5-diethoxytetrahydrofuran-3-
carbaldehyde, glacial AcOH; (i) KMnO4, H2O/acetone (1:1), 0 ꢀC; (ii) 38% sodium hydrogen sulfite; (iii) 6 N HCl.

V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
473
Table 1
Binding affinity at AMPA, Gly/NMDA and high-affinity KA receptors.
N
O
O
N
R6
OH
O
N
OCH3
O
N
N
R7
N
H
F3C
N
H
1-11, 23-25
22
M)a or I%b
IC50
(m
M)c or I%b
log BBpred
e
compd
R6
R7
Ki (m
[3H]AMPA
[3H]glycine
[3H]KA
1d
2d
H
Cl
Cl
11.6 ꢂ 1.9
0.24 ꢂ 0.02
140 ꢂ 15
ꢁ0.08
ꢁ0.74
N
N
0.25 ꢂ 0.04
47%
7.0 ꢂ 0.5
N
COOH
CONH
3d
Cl
Cl
Cl
16.5 ꢂ 1.2
1.3 ꢂ 0.1
4.2 ꢂ 0.1
40%
0.62 ꢂ 0.02
19.1 ꢂ 1
ꢁ0.86
ꢁ0.11
ꢁ0.45
23d
24d
NO2
1.1 ꢂ 0.1
22.7 ꢂ 2.7
OHC
6.2 ꢂ 0.4
N
HOOC
25d
Cl
1.3 ꢂ 0.2
44%
13 ꢂ 0.6
ꢁ1.19
N
4
5
6
H
NO2
NH2
CF3
CF3
CF3
4.8 ꢂ 0.4
0.26 ꢂ 0.02
6.1 ꢂ 6.2
0.85 ꢂ 0.09
1.3 ꢂ 0.06
6.1 ꢂ 6.5
37.6 ꢂ 7.0
7.0 ꢂ 0.5
32 ꢂ 4
ꢁ0.03
ꢁ0.02
ꢁ0.07
COOH
CONH
7
CF3
8%
10%
25 ꢂ 4
ꢁ0.74
N
N
8
CF3
CF3
CF3
0.08 ꢂ 0.007
126 ꢂ 9
6.5 ꢂ 0.8
32%
3.0 ꢂ 0.3
10%
ꢁ0.57
ꢁ0.20
ꢁ0.31
N
N
9
OHC
10
0.72 ꢂ 0.13
0.8 ꢂ 0.1
4.2 ꢂ 0.3
N
HOOC
11
CF3
0.46 ꢂ 0.04
5.1 ꢂ 0.9
17.3 ꢂ 3
ꢁ0.98
N
22
DCKA
NBQX
20 ꢂ 2.5
5%
0.07 ꢂ 0.06
21%
0%
8%
7.0 ꢂ 1.1
ꢁ0.38
ꢁ0.10
ꢁ0.11
0.08 ꢂ 0.02
3%
a
Ki values are means ꢂ SEM of three or four separate determinations in triplicate.
b
c
Percentage of inhibition (I%) of specific binding at 100 mM concentration.
IC50 values were means ꢂ SEM of three or four separate determinations in triplicate.
d
e
References [15,18].
Predicted bloodebrain barrier permeation (log BBpred ¼ log [Brain]/[Blood]).
Gly/NMDA receptor (Ki ¼ 0.85
m
M), while it binds the AMPA and KA
of selectivity, the most interesting being the 6-(1,2,4-triazolo-1-yl)-
substituted derivative 8 which displayed the highest AMPA receptor
receptors with lower affinities (Ki ¼ 4.8
m
M and 37.6 M, respec-
m
tively). Introduction of substituents at the 6-position of compound 4
differently affected receptor affinity and selectivity. The presence of
the 6-NO2 group (derivative 5) shifted affinity toward the AMPA
affinity (Ki ¼ 0.08
mM) and selectivity versus both Gly/NMDA and KA
receptors. These results confirm that the 1,2,4-triazol-4-yl moiety is
one of the most profitable heterocyclic rings to obtain potent and
selective AMPA receptor antagonists. Good AMPA receptor affinity
was also shown by the 6-(3-formyl-1H-pyrrol-1-yl)- substituted
receptor (Ki ¼ 0.26
mM), while the 6-NH2 substituent afforded
a nonselective Gly/NMDA versus AMPA receptor antagonist (deriva-
tive 6). The 6-heterocycle-substituted compounds 8, 10 and 11
possess good to high AMPA receptor affinities and different degrees
derivative 10 (Ki ¼ 0.72
m
M) and the 6-(3-carboxypyrrol-1-yl)-
substituted compound 11 (Ki ¼ 0.46
mM), even though they are less

474
V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
Fig. 3. Comparison between AMPA and Gly/NMDA receptor surface properties at the top of the binding pocket with one of the new quinazoline-2,4-dione derivatives. Molecular
surface of receptors in contact with ligands has been highlighted. Hydrophobic area is colored in yellow and polar areas of the binding pocket are depicted in blue. The peculiar
hydrogen bonding network involving Arg 96, Thr 91 and Pro 89 of the AMPA receptor (Panel A), and the homologous Arg 131, Thr 126, and Pro 124 of the Gly/NMDA receptor (Panel
B), with the reported compounds (compound 4, panel A and compound 1, panel B) is shown in orange. An additional
pep stacking interaction between Tyr 61 side chain at the
AMPA (Phe 92 at the Gly/NMDA) receptor and the 3-hydroxyquinazoline-2,4-dione scaffold is also common to all related derivatives. (For interpretation of the references to colour
in this figure legend, the reader is referred to the web version of this article.)
AMPA selective than compound 8. It is interesting to note that both
the3-formyl andthe3-carboxygroup onthe6-pyrrolering playa key
role for the binding at all three receptors since their removal signif-
icantly reduced the affinity (compare compounds 10 and 11 with
the 6-pyrrol-1-yl-substituted compound 9). Introduction of the
2-carboxybenzoylamino group at position 6 of the parent compound
4 annulled both AMPA and Gly/NMDA receptor affinity and main-
tained the ability to bind the KA receptor (derivative 7). Actually,
compound 7 was expected to possess a higher KA receptor affinity,
with respect to the corresponding 7-chloro derivative 3. In fact,
replacement of the chlorine atom with the trifluoromethyl group
ameliorated not only the affinity for the AMPA receptor but also for
the KA receptor (compare new compounds 4, 5, 8, 10, 11 to the cor-
responding 7-chloro derivatives 1, 23, 2, 24 and 25). The 3-O-meth-
ylation of compound 4 caused a drastic reduction of the binding
affinity at the three receptors (compound 22), thus confirming the
importance of the free 3-hydroxy group in this class of derivatives.
the AMPA receptor (Fig. 3A), and the homologous Arg 131, Thr 126,
and Pro 124 at the Gly/NMDA receptor (Fig. 3B). An additional pep
stacking interaction, between the aromatic amino acid side chain
located at the top of the binding pocket (Tyr 61 at the AMPA, and
the homologous Phe 92 at the Gly/NMDA receptor), with the qui-
nazoline-2,4-dione scaffold suggested that the binding motif of
these newly synthesized antagonists could be compatible with the
poses shown in Fig. 4A. This hypothesis is also confirmed by the
recently published quinazolinedione sulfonamide derivative bound
to human GluR2 [24].
Docking results at the AMPA receptor site showed that the NH
group of compound 8 acts as a proton donor toward the backbone
C]O of Pro 89, while the 2-carbonyl group accepts a hydrogen bond
from the backbone NH of Thr 91 (Fig. 4A). Moreover, both 2-carbonyl
and 3-hydroxy groups of compound 8 interact with the side chain of
Arg 96. The triazole ring at the R6-position accepts a hydrogen bond
from the Thr 174 side chain. This specific polar interaction can
explain the difference between the AMPA receptor affinities of
compounds 8 and 9. In fact, compound 9 completely lacks interac-
tionwith Thr 174, as shown in Fig. 4A. Interestingly, the 1,2,4-triazole
ring of derivative 8 is directed toward Thr 174 in the AMPA receptor,
which is mutated to Ala 206 in the Gly/NMDA receptor (Fig. 4B). This
mutation is located in one of the less conserved regions between
iGluRs and has the direct consequence of weakening the
ligandereceptor interactions network, thus justifying the selectivity
profile of the 6-(1,2,4-triazole) substituted derivative 8 (Table 1).
In the Gly/NMDA receptor site, the NH group of compounds 1
(Fig. 3B) and 4 acts as a proton donor toward the backbone C]O of
Pro 124, while the 2-carbonyl group accepts a hydrogen bond from
the backbone NH of Thr 126. Both the 2-carbonyl and 3-hydroxy
3.2. Molecular modeling studies
Molecular modeling studies were carried out to rationalize the
observed SARs, thus suggesting the hypothetical binding mode of
these new antagonists at the Gly/NMDA and AMPA receptors.
The synthesized compounds were docked, using a genetic
algorithm, into the putative binding sites of each receptor single
subunit which structure has been determined by X-ray crystallog-
raphy (See Section 5.2 for further details). Analysis of docking
results highlighted a peculiar ligandereceptor interaction motif
represented by a hydrogen bonding network involving the highly
conserved residue triad among iGluRs: Arg 96, Thr 91 and Pro 89 at
Fig. 4. Panel A: Predicted binding mode of compounds 8 (in white) and 9 (in gray) into the binding pocket of the AMPA receptor. Panel B: The R6 substituent is directed towards the
less conserved region between iGluRs, where Thr 174 (AMPA) - Ala 206 (Gly/NMDA e depicted in yellow) mutation is present. AMPA receptor ribbon, colored in brown, is shown. A
part of Gly/NMDA receptor ribbon representation, where Ala 206 is located, is colored in white and is shown in figure. Receptoreligand hydrogen-bonding interactions are depicted
in orange. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
475
Table 3
groups of compounds 1 and 4 were found to interact with the Arg
Anticonvulsant effect against PTZ-induced convulsions.
131 side chain. Compound 4 (R7 ¼ CF3) exhibited an increased
binding affinity to AMPA receptors compared to compound 1
(R7 ¼ Cl) and, intriguingly, this trend was reversed on the Gly/
NMDA receptor (Table 1). The peculiar effect observed around
position 7 of the quinazoline scaffold can be associated to the
important changes of the receptor surface properties in this specific
region of the receptor cavities. In fact, the highly polar residues
located at the top of the AMPA receptor binding pocket, such as Glu
13 and Tyr 61, are replaced with the more hydrophobic residues Gln
13 and Phe 92, in the Gly/NMDA receptor (Fig. 3). Following our
modeling results, the change of the selectivity profiles of
compounds 1 and 4 can be explained by the modification of the
electrostatic potential surface properties of the two receptor areas
around position 7 of the quinazoline scaffold.
Finally, blood brain barrier (BBB) penetration, plasma protein
binding (PPB) and human intestinal absorption (HIA) were
computed using StarDrop [25]. The predicted BBB penetration data
(Table 1) were promising for several compounds (drugs with a poor
ability to cross BBB typically have a log BB value ꢃ ꢁ1.0), and in
particular for the most interesting compounds of the series, i.e. 4
and 8, showing the highest affinity and selectivity for the Gly/
NMDA and AMPA receptor, respectively. Finally, the PPB and HIA
predictions were encouraging for all the novel antagonists
(unpublished data) based on the binary prediction models imple-
mented by StarDrop software (HIA threshold ¼ 30% absorbed; PPB
threshold ¼ 80% absorbed) [25].
Treatment
Dose mg/kg p.o.
n. of mice
% Protection on
PTZ-induced convulsionsa
Vehicle
Compound 8
Diazepam
e
41
24
30
0
30
1.0
65.5
93.3
a
Pentylenetetrazole (PTZ) 90 mg/kg i.p.
preparations [14]. The results, reported in Table 2, showed that
compounds 8, 10 and 11 inhibited AMPA and NMDA responses with
potencies which match their binding affinities. In fact, compounds
8 and 11 were the most active in inhibiting, respectively, AMPA-
and NMDA-evoked responses, coherently with their AMPA and Gly/
NMDA receptor affinities. It has to be noted that compound 8
potency is similar to that of NBQX in counteracting AMPA-induced
depolarization.
3.3.2. Anticonvulsant studies
In our previous studies some 7-chloro-3-hydroxyquinazoline-
2,4-diones showed anticonvulsant properties [20], therefore we
decided to test compound 8 to evaluate its protective effect from
convulsions in mice using pentylentetrazole (PTZ) as a chemical
convulsant agent. Compound 8, administered per os, exerted
a significant anticonvulsant effect at 30 mg/kg (Table 3), thus
showing both a good oral absorption and a good ability to penetrate
the BBB, in line with the predicted BBB permeation (Table 1).
3.3.3. Oxygeneglucose deprivation (OGD) in organotypic rat
hippocampal slice cultures
3.3. Pharmacological studies
Compound 8, which showed the highest AMPA receptor affinity
and selectivity among the herein reported compounds, was tested
in an in vitro model of cerebral ischemia to assess its efficacy in
decreasing neuronal damage produced by oxygen and glucose
deprivation (OGD) in organotypic rat hippocampal slices. It has
been previously reported that OGD-neurotoxicity is significantly
attenuated by adding either NMDA or AMPA receptor antagonists in
the incubation medium during the 30 min insult and the subse-
quent 24 h recovery period [26,27]. Neuronal cell death was
quantified using cellular uptake of the fluorescent dye propidium
iodide (PI). When rat hippocampal slices incubated with PI were
exposed to glutamate, fluorescence levels increased dramatically in
virtually all neuronal populations. Maximal damage was achieved
in this system by exposing the slices to 10 mM glutamate for 24 h.
Exposure to OGD for 30 min caused a fluorescence pattern in which
PI staining was more intense in the CA1 area of hippocampus [26].
In our experiments, quantitative analysis, carried out in the
hippocampal CA1 sub region 24 h after the insult, revealed that
30 min OGD produced PI fluorescence levels that were 57 ꢂ 8% of
those observed when neuronal injury was maximal (Fig. 5).
3.3.1. Functional antagonism studies
The antagonistic effect of some selected derivatives was
evaluated both at the NMDA and AMPA receptors. Functional
antagonism at the NMDA receptor-ion channel complex of
compounds 4 and 8 was evaluated by measuring their ability to
inhibit the stimulated binding of the NMDA receptor channel
blocking agent [3H](þ) MK-801 [19]. The results are reported in
Table 2, together with that of DCKA included as reference
antagonist at the glycine site. Derivatives 4 and 8 demonstrated
an antagonistic behavior with potencies correlated with their
Gly/NMDA affinities.
Derivatives 8, 10 and 11, together with NBQX and DCKA as
reference compounds, were assayed for their ability to inhibit
depolarization induced by 5 mM AMPA and NMDA in cortical wedge
Table 2
Inhibition of stimulated [3H] (þ)MK-801 binding and functional antagonism at
NMDA and AMPA sites.
compd
[3H] (þ)MK-801
AMPA and NMDA induced
depolarizationd
bindinga
Compound 8 (1
m
M) significantly reduced (29 ꢂ 9%) the extent of
IC50
(m
M)b
IC50 (m
M)b
OGD-induced neuronal death when added to the incubation
medium during the 30 min insult and the subsequent 24 h recovery
period (Fig. 5). Under the same experimental conditions, the AMPA
receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo[f]
NMDA
Ntc
AMPA
Ntc
4
8
10
11
7.2 ꢂ 0.9
14 ꢂ 0.9
Ntc
24 ꢂ 3
7.0 ꢂ 1.0
>50
0.17 ꢂ 0.02
1.5 ꢂ 0.2
2.9 ꢂ 0.35
52 ꢂ 11
quinoxaline-2,3-dione (NBQX), at 30 mM concentration, dramati-
Ntc
cally reduced CA1 PI fluorescence levels (20 ꢂ 4%). Fig. 5 also shows
DCKA
NBQX
0.74 ꢂ 0.09
4.7 ꢂ 0.9
Ntc
(*)e
0.20 ꢂ 0.02
that compound 8 alone, at the concentration of 10 mM, was devoid
a
of toxicity in slice cultures.
Inhibition of stimulated [3H] (þ)MK-801 binding. All assays were carried out in
the presence of 10 mM glutamate and 0.1 mM glycine.
b
IC50 values are means ꢂ SEM of three or four separate determinations in
3.3.4. Electrophysiological studies in the CA1 region of rat acute
hippocampal slices
Physiological functions of KA receptors are not fully understood
yet and much attention has been paid to their role in ischemic brain
injury. It is well known that KA receptors exert a variety of functions
triplicate.
c
Not tested.
d
Inhibition of depolarization induced by S-AMPA or NMDA in mouse cortical
wedge preparation.
e
At 10 mM concentration the inhibition was not significant.

476
V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
in the regulation of synaptic activity [28]. A considerable amount of
125
100
75
50
25
0
data indicates that they are located both at the pre- and post-
synaptic site and play a special role in regulating transmission and
controlling short- and long-term plasticity in all subfields in the
hippocampus [29e32]. On this basis, we decided to test at KA
receptors the most active compound so far identified among the
3-hydroxyquinazoline-2,4-dione series, i.e. derivative 3, to evaluate
its effect on synaptic transmission in the CA1 region of acute iso-
lated rat hippocampal slices, both under basal conditions and
during OGD. Compound 3 possesses good affinity for high-affinity
and low-affinity native KA receptors (Ki ¼ 0.62
mM and 1.6 mM,
respectively) (Table 1 and Ref. [18]). High-affinity and low-affinity
native KA receptors consist of a combination of GluK4-5 [33e36]
and GluK1-3 [35e37] subunits, respectively. Compound 3 also
*
binds AMPA receptor with a Ki value of 16.5 mM.
**
In the first series of experiments we evaluated the efficacy of
compound 3 in antagonizing the effect of kainate on field excitatory
postsynaptic potential (fepsp) in the hippocampal CA1 region. In
agreement with previous results [38e41], bath application of KA
reversibly suppressed fepsp amplitude (Fig. 6A), thus inhibiting
synaptic neurotransmission. It is known that this inhibitory effect
of KA on fepsp is due to the stimulation of KA receptors located at
presynaptic level in the CA1 region, thus inducing a decrease in
glutamate release [39,42]. Kamiya and Ozawa [39] and Dargan et al.
GLU
10 µM
Cmp 8
CRL
1 µM
Cmp 8
30 µM
NBQX
OGD
[42] have demonstrated that at low concentrations (0.5e1 mM), like
Fig. 5. Effect of compound (Cmp) 8 on OGD-induced neuronal damage in organotypic
rat hippocampal slices. Hippocampal slices were exposed to 30 min OGD and 24 h later
CA1 damage was assessed by measuring the intensity of PI fluorescence. Data are
expressed as percent of maximal neuronal death, as measured following application of
10 mM glutamate (GLU) for 24 h. Background fluorescence was determined in control
sister cultures not exposed to OGD and was subtracted from all experimental values.
Each bar represents the mean ꢂ S.E.M. of four experiments. *P < 0.05 vs OGD (t-
test þ ANOVA); **P < 0.01 vs OGD (t-test þ ANOVA).
those utilized in the present study, the fepsp suppression induced
by KA, possibly by GLUK1 receptors, is not related i) to the activa-
tion of KA receptors located at postsynaptic levels; ii) to the acti-
vation of NMDA or AMPA receptors; or iii) to the effects of KA on
inhibitory GABAergic neurotransmission. The inhibitory effect of
KA on neurotransmission was antagonized by compound 3. In fact,
the KA-induced fepsp suppression was reduced by 1 mM compound
Fig. 6. Effect of compound 3 on KA-mediated suppression of excitatory transmission in the CA1 hippocampal region. Graph A shows the time course of fepsp amplitude recorded in
a typical experiment before, during and after KA application (500 nM), in the absence or in the presence of 1 M compound 3. Data are expressed as percent of pre-drug. Closed bars
indicate KA application; open bar indicates compound 3 application. Graph B displays the summary of results obtained in the presence of 500 nM KA, applied alone or in the
presence of 1 M compound 3. Error bars are SEM; *p < 0.05 paired Student’s t-test vs respective pre-drug. Graph C shows the summary of the effects of compound 3, applied at two
m
m
different concentrations, on synaptic potentials under basal conditions. Error bars are SEM; *p < 0.05 paired Student’s t-test vs respective pre-drug; xp < 0.05 vs 1
mM compound 3,
One-way ANOVA followed by NewmaneKeuls Multiple Comparison Test. The number of experiments are shown in parentheses.

V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
477
3 (Fig. 6A). In particular, KA (500 nM, 10 min) suppressed fepsp
amplitude to 52.1 ꢂ 15.3% of the respective pre-drug (p < 0.05,
suppress excitatory neurotransmission in this brain area where
AMPA receptors are particularly abundant [43]. Our results, there-
fore, suggest that compound 3, at the concentrations tested, also
blocks native AMPA receptors.
Subsequently, the effect of compound 3 on neurotransmission
was evaluated under ischemic conditions obtained, in vitro, by OGD
application. In these experimental conditions, a severe, 7-min
period of OGD always produced an irreversible and complete loss
of synaptic transmission (Fig. 7A) and the appearance of anoxic
depolarization (AD) in all slices examined (Fig. 7D). The loss of
synaptic transmission is attributable to the massive release of
n ¼ 5) when applied alone, while in the presence of 1
mM
compound 3 it suppressed fepsps to 85.8 ꢂ 6.4% of the respective
pre-drug level (n ¼ 5) (Fig. 6B). Interestingly, compound 3 affected
per se CA1 basal excitatory transmission (Fig. 6A and C). At
a concentration of 1
(p < 0.05, n ¼ 6) and at 5
m
M, it reduced fepsp amplitude by 19.2 ꢂ 3.2%
m
M by 39.6 ꢂ 5.9% (p < 0.05, n ¼ 6), in
comparison to the respective controls (Fig. 6C). It is known that
excitatory transmission in the CA1 region is due in large part to
stimulation of AMPA receptors and that AMPA receptor antagonists
Fig. 7. Effect of compound 3 on the appearance of AD and the irreversible loss of neurotransmission induced by 7-min OGD in CA1 hippocampal slices (AeC). Each graph shows the
time course of fepsp amplitude before, during 7-min OGD and after reperfusion in oxygenated aCSF in the absence (A) and in the presence of 1 M compound 3 (B) or 5
m
mM
compound 3 (C). Data are expressed as percent of pre-drug application (mean ꢂ SEM). The number of slices are shown in parentheses. (DeF) AD was recorded as a negative voltage
shift in response to 7-min OGD in the absence (D, n ¼ 15) or in the presence of 1
m
M compound 3 (E, n ¼ 3) or 5 M compound 3 (F, n ¼ 6). The open bars show the application time
m
of each drug. Closed bars indicate OGD time duration.

478
V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
excitatory neurotransmitters [3]. AD, measured as negative direct
current shift, is secondary to the strong depolarization induced by
the OGD, and represents an important parameter of brain damage
[44e48]. The direct current shift showed a mean peak latency of
6.45 ꢂ 0.2 min (n ¼ 15, calculated from the beginning of OGD) and
a mean peak amplitude of 7.1 ꢂ 0.5 mV (n ¼ 15). The amplitude of
fepsp did not recover (1.5 ꢂ 1.8%, n ¼ 15) when the slices were
reperfused with oxygenated, glucose-containing artificial cerebro-
spinal fluid (aCSF) (Fig. 7A). Seven minutes of OGD was then applied
in the presence of compound 3 which was superfused at two
cmꢁ1. NMR spectra were recorded on a Brucker Avance 400 spec-
trometer (400 MHz for 1H NMR,100 MHz for 13C NMR). The chemical
shifts are reported in d (ppm) and are relative to the central peak of
the solvent which was always DMSO-d6. The following abbreviations
are used: s ¼ singlet, d ¼ doublet, t ¼ triplet, m ¼ multiplet,
br ¼ broad, and ar ¼ aromatic protons. 13C NMR signals are singlets,
unless otherwise stated.
5.1.1. 2-Amino-N-(benzyloxy)-4-trifluoromethylbenzamide (13)
Triethylamine (4.3 mmol) was added to a suspension of
O-benzylhydroxylamine hydrochloride (4.3 mmol) in ethanol
(15 mL). The 7-trifluoromethyl-1,2-dihydro-3,1-benzoxazine-2,4-
dione 13 [21] (4.3 mmol) was added to the solution and the
mixture was heated at reflux for 3 h. Cooling at room temperature
and dilution with water (about 50 mL) gave a solid that was
collected, washed with water and recrystallized. Yield: 90%; mp
146e148 ꢀC (cyclohexane) 1H NMR: 4.88 (s, 2H, CH2), 6.55 (br s, 2H,
NH2) 6.75 (d,1H, ar, J ¼ 7.7 Hz), 7.00 (s,1H, ar), 7.33e7.39 (m, 6H, ar),
11.70 (s, 1H, NH) Anal. Calcd for: C15H13F3N2O2: C, 58.07; H, 4.22; N,
9.03. Found: C, 57.96; H, 4.41; N, 9.31.
different concentrations (1 and 5
tion. At 1 M concentration it was unable to prevent the appear-
ance of AD and the loss of neurotransmission induced by 7-min
M concentration,
mM), 15 min before OGD induc-
m
OGD (n ¼ 3, Fig. 7B and E). Conversely, at 5
m
compound 3 prevented the appearance of AD in all slices tested
(n ¼ 6, Fig. 7F), allowing an almost complete recovery of fepsp
amplitude (92.2
ꢂ
4.4%; Fig. 7C). As previously described,
compound 3 reduced fepsp amplitude under basal conditions
(Fig. 7B and C, see also Fig. 6C).
In summary, compound 3, applied at the concentration of 5 mM
before, during and after a severe period of OGD allows a significant
recovery of fepsp amplitude and prevents the appearance of AD.
This demonstrates that compound 3 is able to limit, during
OGD, an excess of excitatory transmission which leads to excito-
toxicity. We cannot exclude that the preventive effects of
compound 3 on AD appearance and loss of neurotransmission
induced by OGD could be related to the AMPA receptor blockade
that was demonstrated to counteract AD appearance during OGD
in hippocampal CA1 region [49]. Interestingly, specific antagonists
of the KA receptor GluK2 subunit and of AMPA receptor are
protective in cerebral ischemia models, both in vivo and in vitro
[50e54].
5.1.2. 3-Benzyloxy-7-trifluoromethyl-1H-quinazoline-2,4-dione
(14)
Triethylamine (7.7 mol) was dropwise added to a solution of
compound 13 (3.2 mmol) and triphosgene (1.28 mmol) in anhy-
drous tetrahydrofuran (30 mL). The mixture was stirred at room
temperature for about 1 h then diluted with water (80 mL) to give
a solid which was collected, washed with water and recrystallized.
Yield: 92%; mp 228e230 ꢀC (cyclohexane) 1H NMR: 5.09 (s, 2H,
CH2), 7.39e7.56 (m, 7H, ar), 8.15 (d, 1H, ar, J ¼ 8.4 Hz), 11.93 (s, 1H,
NH). Anal. Calcd for C16H11F3N2O3: C, 57.15; H, 3.30; N, 8.33. Found:
C, 57.38; H, 3.01; N, 8.62.
4. Conclusion
5.1.3. 7-Trifluoromethyl-3-hydroxy-1H-quinazoline-2,4-dione (4)
A suspension of compound 14 (3.3 mmol) in 48% hydrobromic
acid (6 mL) and glacial acetic acid (6 mL) was refluxed for about 3 h.
The resulting solution was cooled at room temperature and diluted
with water (50 mL), the solid was collected, washed with water and
recrystallized. Yield: 95%; mp 256e257 ꢀC (cyclohexane/EtOAc). 1H
NMR: 7.46 (s, 1H, ar), 7.51 (d, 1H, ar, J ¼ 8.0 Hz), 8.12 (d, 1H, ar,
J ¼ 8.0 Hz), 10.80 (s, 1H, OH), 11.80 (s, 1H, NH). 13C NMR: 112.74,
117.74, 118.91, 125.15 (q, CF3), 129.30 (C5), 134.41 (q, C7), 139.15,
148.97 (C2), 159.04 (C4). IR 1670, 1750, 3480. Anal. Calcd for
C9H5F3N2O3: C, 43.92; H, 2.05; N, 11.38. Found: C, 43.78; H, 1.95;
N, 11.02.
The studies described in the present work have clarified the
SARs of the 3-hydroxyquinazolin-2-4-dione series as iGluRs
antagonists. Molecular modeling investigation has permitted
prediction of the hypothetical binding mode of the new derivatives
at the Gly/NMDA and AMPA receptors and rationalization of affinity
and selectivity profiles of some derivatives.
Some new Gly/NMDA and AMPA receptor antagonists were
identified and the most potent at the AMPA receptor was the 6-
(1,2,4-triazol-yl)-substituted compound 8. This derivative demon-
strated efficacy both in counteracting neuronal damage in
organotypic rat hippocampal slices exposed to OGD and as anti-
convulsant agent. Compound 3, possessing the highest KA receptor
affinity and selectivity among the 3-hydroxyquinazolin-2,4-dione
series, showed neuroprotective properties in an in vitro electro-
physiological model of cerebral ischemia. In fact, it was effective in
preventing both the failure of synaptic activity and the appearance
of AD induced by OGD.
5.1.4. 3-Acetoxy-7-trifluoromethyl-1H-quinazoline-2,4-dione (15)
A suspension of derivative 4 (1.4 mmol) in acetic anhydride
(15 mL) was refluxed for 2 h. Evaporation of the excess of acetic
anhydride at reduced pressure gave a residue that was stirred with
water (40 mL). The solid obtained was collected, washed with
water and recrystallized. Yield: 85%; mp 245e246 ꢀC (cyclohexane/
EtOAc). 1H NMR: 2.38 (s, 3H, CH3), 7.50 (s, 1H, ar), 7.58 (d, 1H, ar,
J ¼ 8.0 Hz), 8.15 (d, 1H, ar, J ¼ 8.0 Hz), 12.20 (br s, 1H, NH). 13C NMR:
17.79 (CH3), 113.28, 117.23, 119.59, 123.58 (q, CF3), 129.45 (C5),
135.21 (q, C7), 139.60, 146.85 (C2), 157.31 (C4), 167.36 (acetoxy CO).
Anal. Calcd for C11H7F3N2O4: C, 45.85; H, 2.45; N, 9.72. Found: C,
45.66; H, 2.23; N, 9.63.
5. Experimental section
5.1. Chemistry
Silica gel plates (Merck F254) and silica gel 60 (Merck,
70-230 mesh) were used for analytical and column chromatography,
respectively. All melting points were determined on a Gallenkamp
melting point apparatus and are uncorrected. Microanalyses were
performed with a PerkineElmer 260 elemental analyzer for C, H, N,
and the results were within ꢂ0.4% of the theoretical values, unless
otherwise stated. The IR spectra were recorded with a PerkineElmer
Spectrum RX I spectrometer in Nujol mulls and are expressed in
5.1.5. 3-Acetoxy-7-trifluoromethyl-6-nitro-1H-quinazoline-2,4-
dione (16)
Compound 15 (4.0 mmol) was portionwise added to cooled
(ꢁ10 ꢀC) HNO3 (90%, 14 mL). When the addition was completed
(1 h), the mixture was stirred for about 30 min at 0 ꢀC, then the
solution was poured onto ice (about 100 g) and the solid collected,

V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
479
washed with abundant water and recrystallized. Yield: 82%; mp
5.1.10. 3-Methoxy-7-trifluoromethyl-6-nitro-1H-quinazoline-2,4-
dione (20)
252e254 ꢀC (EtOH). 1H NMR: 2.42 (s, 3H, CH3), 7.73 (s, 1H, ar), 8.67
(s, 1H, ar), 12.72 (br s, 1H, NH). Anal. Calcd for C11H6F3N3O6: C,
39.65; H, 1.82; N, 12.61. Found: C, 41.56; H, 1.99; N, 10.43.
Compound 18 (4.6 mmol) was portionwise added to cooled
(ꢁ10 ꢀC) HNO3 (90%, 15 mL). After the addition was completed
(1 h), the mixture was stirred for about 1 h and 30 min, then the
solution was poured onto ice (about 100 g). The mixture was
extracted with EtOAc (60 mL for three times), the organic layers
were dried (Na2SO4) and the solvent evaporated at reduced
pressure to yield a solid that was collected by filtration and
recrystallized. Yield: 80%; mp 220e222 ꢀC (EtOH). 1H NMR: 3.91
5.1.6. 6-Amino-3-acetoxy-7-trifluoromethyl-1H-quinazoline-2,4-
dione (17)
20% Pd/C was added to a solution of compound 16 (1.7 mmol) in
ethanol (200 mL) and the mixture was hydrogenated in a Parr
apparatus at 40 psi for 24 h. The catalyst was filtered off and the
solvent evaporated at reduced pressure to yield a solid that was
recrystallized. Yield: 80%; mp 266e267 ꢀC (EtOAc). 1H NMR: 2.37
(s, 3H, CH3), 5.73 (s, 2H, NH2), 7.27 (s, 1H, ar), 7.45 (s, 1H, ar), 12.00
(br s, 1H, NH). Anal. Calcd for C11H8F3N3O4: C, 43.58; H, 2.66; N,
13.86. Found: C, 43.71; H, 1.95; N, 14.59.
(s, 3H, CH3), 7.67 (s, 1H, ar), 8.64 (s, 1H, ar), 12.36 (br s, 1H, NH). 13
C
NMR: 64.03 (OCH3), 116.36 (C8), 117.82, 121.9 (q, CF3), 126.64 (C5),
127.40 (q, C7), 141.28, 142.44, 148.15 (C2), 157.69 (C4). Anal. Calcd
for C10H6F3N3O5: C, 39.36; H, 1.98; N, 13.77. Found: C, 39.54; H,
1.67; N, 13.94.
5.1.7. 6-(2-Carboxybenzoylamino)-7-trifluoromethyl-3-hydroxy-
1H-quinazoline-2,4-dione (7)
5.1.11. 6-Amino-3-methoxy-7-trifluoromethyl-1H-quinazoline-2,4-
dione (21)
A mixture of the 6-amino derivative 17 (1.1 mmol), phthalic
anhydride (3.3 mmol) and sodium acetate (1.1 mmol) in glacial
acetic acid (5 mL) was heated at 60 ꢀC for about 36 h. After cooling
at room temperature, the suspension was diluted with water
(30 mL) and acidified to pH 4 with 6 N HCl. The solid which
precipitated was collected and suspended in 0.25 M NaOH solu-
tion (4 mL). The mixture was stirred at room temperature for
30 min, then the solution acidified to pH ¼ 1 with 6 N HCl, the
solid collected, washed with water and recrystallized. Yield: 35%;
mp >300 ꢀC (EtOH); 1H NMR: 7.44 (d, 1H, ar, J ¼ 7.4 Hz), 7.54 (s,
1H, ar), 7.57e7.71 (m, 3H, ar), 8.14 (d, 1H, ar, J ¼ 7.59), 10.10 (s, 1H,
OH or NH), 11.39 (s, 1H, OH or NH), 11.58 (s, 1H, NH or OH), 13.24
(br s, 1H, COOH); IR 1698, 1728, 2500e3300. Anal. Calcd for
C17H10F3N3O6: C, 49.89; H, 2.46; N, 10.27. Found: C, 49.61; H, 2.25;
N, 10.39.
A mixture of compound 20 (3.4 mmol) and Pd/C (10%, 0.2 g) in
EtOH (100 mL) was hydrogenated at 40 psi in a Parr apparatus for
14 h. The catalyst was filtered off and the solvent evaporated at
reduced pressure to give a solid that was collected and recrystal-
lized. Yield: 82%; mp 250e252 ꢀC (EtOH). 1H NMR: 3.86 (s, 3H, CH3),
5.65 (br s, 2H, NH2), 7.23 (s, 1H, ar), 7.47 (s, 1H, ar), 11.36 (s, 1H, NH).
IR 1697, 1718, 1747, 3151, 3185. Anal. Calcd for C10H8F3N3O3: C,
43.65; H, 2.93; N, 15.27. Found: C, 43.89; H, 2.68; N, 15.01.
5.1.12. 3-Methoxy-7-trifluoromethyl-6-(1,2,4-triazol-4-yl)-1H-
quinazoline-2,4-dione (22)
Diformylhydrazine (3.3 mmol) and then, drop by drop, trime-
thylsilylchloride (16.5 mmol) were added to
a solution of
compound 21 (1.1 mmol) in anhydrous pyridine (5 mL). The
mixture was heated at 100 ꢀC for 24 h, then the solvent evaporated
at reduced pressure. The residue was treated with water (5e10 mL)
and extracted with EtOAc (20 mL for three times). Evaporation of
the dried (Na2SO4) organic layers afforded a solid which was
collected and recrystallized. Yield: 90%; mp >300 ꢀC (MeOH). 1H
NMR: 3.91 (s, 3H, CH3), 7.67 (s, 1H, ar), 8.20 (s, 1H, ar), 8.79 (s, 2H,
triazole protons), 12.15 (br s, 1H, NH). IR 1697, 1745, 3121. Anal.
Calcd for C12H8F3N5O3: C, 44.05; H, 2.46; N, 21.40. Found: C, 44.24;
H, 2.39; N, 21.25.
5.1.8. 2-Amino-N-methoxy-4-trifluoromethylbenzamide (18)
Triethylamine (2.16 mmol) was added to a suspension of
O-methylhydroxylamine hydrochloride (2.16 mmol) in ethanol
(10 mL). The 7-trifluoromethyl-1,2-dihydro-3,1-benzoxazine-2,4-
dione 12 [21] (2.16 mmol) was added to the solution and the
mixture was heated at reflux for about 2 h. The solution was
cooled at room temperature and concentrated to small volume by
evaporation of the solvent at reduced pressure. Dilution with
water (about 80 mL) gave a solid that was collected, washed with
water and recrystallized. Yield: 85%; mp 165e167 ꢀC (cyclo-
hexane/EtOAc) 1H NMR: 3.69 (s, 3H, CH3), 6.58 (br s, 2H, NH2)
6.78 (d, 1H, ar, J ¼ 8.1 Hz), 7.06 (s, 1H, ar), 7.47 (d, 1H, ar,
J ¼ 8.1 Hz), 11.63 (s, 1H, NH). IR 1648, 3197, 3382, 3489. Anal.
Calcd for C9H9F3N2O2: C, 46.16; H, 3.87; N, 11.96. Found: C, 46.01;
H, 4.09; N, 10.89.
5.1.13. 7-Trifluoromethyl-3-hydroxy-6-nitro-1H-quinazoline-2,4-
dione (5)
A solution of compound 20 (2.9 mmol) in 48% HBr (18 mL) and
glacial acetic acid (18 mL) was refluxed for about 48 h. The solution
was concentrated to small volume, diluted with water (about
60 mL) and extracted with EtOAc (50 mL for three times). Evapo-
ration of the dried (Na2SO4) organic layers afforded a solid which
was collected and recrystallized. Yield: 53%; mp 229e231 ꢀC
(EtOH). 1H NMR: 7.64 (s, 1H, ar), 8.60 (s, 1H, ar), 11.02 (br s, 1H,
OH), 12.30 (br s, 1H, NH). 13C NMR: 116.5 (C8), 117.36, 121.91 (q, CF3),
126.58 (C5), 127.10 (q, C7), 141.03, 142.36, 148.73 (C2), 157.90 (C4).
Anal. Calcd for C9H4F3N3O5: C, 37.13; H, 1.38; N, 14.43. Found: C,
38.86; H, 1.59; N, 16.33.
5.1.9. 3-Methoxy-7-trifluoromethyl-1H-quinazoline-2,4-dione (19)
Triethylamine (5.1 mol) was dropwise added to a solution of
compound 18 (2.1 mmol) and triphosgene (0.85 mmol) in anhy-
drous tetrahydrofuran (30 mL). The mixture was stirred at room
temperature for about 2 h then diluted with water (100 mL) and
extracted with EtOAc (60 mL for three times). The organic layers
were anhydrified (Na2SO4) and the solvent evaporated at reduced
pressure. The obtained oil was treated with diethyl ether to give
a solid that was collected and recrystallized. Yield: 65%; mp
220e223 ꢀC (EtOH) 1H NMR: 3.89 (s, 3H, CH3), 7.48 (s, 1H, ar), 7.55
(d, 1H, ar, J ¼ 8. 1 Hz), 8.15 (d, 1H, ar, J ¼ 8.1 Hz), 11.86 (s, 1H, NH). 13C
NMR: 64.08 (OCH3), 112.73, 117.97, 118.80 (C8), 123.64 (q, CF3),
129.24 (C5), 134.63 (q, C7), 139.40, 147.92, 158.31. IR 1726, 1759,
3567. Anal. Calcd for C10H7F3N2O3: C, 46.16; H, 2.71; N,10.77. Found:
C, 46.38; H, 2.92; N, 10.54.
5.1.14. 6-Amino-7-trifluoromethyl-3-hydroxy-1H-quinazoline-2,4-
dione (6)
20% Pd/C was added to a solution of compound 5 (1.7 mmol) in
ethanol (200 mL) and the mixture was hydrogenated in a Parr
apparatus at 40 psi for 24 h. The catalyst was filtered off and the
solvent evaporated at reduce pressure to yield a solid that was
recrystallized. Yield: 85%; mp 290e292 ꢀC (EtOAc). 1H NMR: 5.59
(s, 2H, NH2), 7.22 (s, 1H, ar), 7.43 (s, 1H, ar), 10.60 (br s, 1H, OH or
NH), 11.20 (br s, 1H, NH or OH). IR 1670, 1735, 3390e3420. Anal.

480
V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
Calcd for C9H6F3N3O3: C, 41.39; H, 2.32; N, 16.09. Found: C, 42.91; H,
2.06; N, 16.29.
5.2. Molecular modeling studies
5.2.1. Target structures
5.1.15. 7-Trifluoromethyl-3-hydroxy-6-(1,2,4-triazol-4-yl)-1H-
quinazoline-2,4-dione (8)
The crystal structures of AMPA and Glycine/NMDA receptor
subunit were retrieved from the PDB (PDB code: 3R7X and 1PBQ,
respectively) [24,55]. Hydrogen atoms were added using standard
geometries to the protein structure with the Molecular Operation
Environment (MOE, version 2010.11) program [56]. To appropri-
ately assign ionization states and hydrogen positions, the
“Protonate-3D” tool implemented by MOE [57,58] was used. To
minimize contacts among hydrogens, the structures were sub-
jected to Amber94 [5] force field minimization until the rms of
conjugate gradient was <0.05 kcal/mol/Å, keeping the heavy atoms
fixed at their crystallographic positions.
The title compound was obtained by reacting derivative 6
(3.3 mmol) with diformylhydrazine (3.3 mmol) in the same
conditions described above to prepare compound 22 from 21. Yield:
85%; mp >300 ꢀC (EtOH). 1H NMR: 7.65 (s, 1H, ar), 8.13 (s, 1H, ar),
8.76 (s, 2H, triazole protons), 10.70 (br s, 1H, OH or NH), 12.00 (br s,
1H, NH or OH). 13C NMR: 115.20 (C8), 117.88, 122.45 (q, CF3), 125.31,
129.68 (C5), 130.65 (C7), 139.83, 144.42 (C triazole), 148.87 (C2),
158.29 (C4). IR 1683, 1734, 3506. Anal. Calcd for C11H6F3N5O3: C,
42.18; H, 1.93; N, 22.36. Found: C, 42.24; H, 2.09; N, 22.65.
5.1.16. 7-Trifluoromethyl-3-hydroxy-6-(pyrrol-1-yl)-1H-
quinazoline-2,4-dione (9)
5.2.2. Molecular docking protocol
All docked ligands were built using the “Builder” module of MOE
[55]. Ligands were docked into the putative binding sites using
GOLD 5.0 suite [59]. Charges for ligands were imported from the
MOPAC program output files using PM3/ESP semiempirical
Hamiltonian [60,61]. The genetic algorithm implemented by GOLD
performs 25 independent docking runs and writes the resulting
conformations and their energies in a molecular database file. The
top ranking pose for each compound was selected according to
the scaffold orientation of a quinazolinedione sulfonamide deriv-
ative, which has been recently co-crystallized by M. Koller and
co-workers [24].
A solution of 2,5-diethoxytetrahydrofuran (3.6 mmol) in glacial
acetic acid (3 mL) was dropwise added to a hot (90 ꢀC) suspension
of the 6-amino-derivative 6 (1.2 mmol) in glacial acetic acid (8 mL).
The solution was heated at 90 ꢀC for 25 min, then the solvent was
distilled off at reduced pressure. The tarry residue was treated with
water and the solid collected and recrystallized. Yield: 84%; mp
280e282 ꢀC (EtOH); 1H NMR: 6.24 (s, 2H, pyrrole protons), 6.92
(s, 2H, pyrrole protons), 7.61 (s, 1H, ar), 7.77 (s, 1H, ar), 11.54 (br s,
1H, NH or OH), 11.70 (br s, 1H, NH or OH). IR 1690, 1708, 1734, 3178.
Anal. Calcd for C13H8F3N3O3: C, 50.17; H, 2.59; N, 13.50. Found: C,
52.29; H, 2.38; N, 13.79.
5.3. Pharmacology
5.1.17. 7-Trifluoromethyl-3-hydroxy-6-(3-formylpyrrol-1-yl)-1H-
quinazoline-2,4-dione (10)
All animal procedures were conducted according to the Italian
Guidelines for Animal Care, DL 116/92, application of the European
Communities Council Directive (86/609/EEC).
A
solution of 2,5-dimethoxytetrahydrofuran-3-carbaldehyde
(2.8 mmol) in glacial acetic acid (7 mL) was dropwise added to
a hot (90 ꢀC) suspension of the 6-amino-derivative 6 (1.9 mmol) in
glacial acetic acid (8 mL). After the addition was completed, the
mixture was stirred at room temperature for 30 min. Evaporation of
the solvent at reduced pressure yielded a solid which was sus-
pended in water (10e20 mL), collected and recrystallized. Yield:
83%; mp 280e282 ꢀC (EtOH); 1H NMR: 6.65 (s, 1H, pyrrole protons),
7.11 (s, 1H, pyrrole proton), 7.65 (s, 1H, ar), 7.87 (s, 1H, pyrrole
proton), 7.98 (s, 1H, ar), 9.77 (s, 1H, CHO), 10.94 (br s, 1H, OH or NH),
12.05 (br s, 1H, NH or OH); IR 1656, 1687, 1736, 3125, 3250. Anal.
Calcd for C14H8F3N3O4: C, 49.57; H, 2.38; N, 12.39. Found: C, 49.81;
H, 2.26; N, 12.53.
5.3.1. Binding assay
Rat cortical synaptic membrane preparation [3H]glycine (53 Ci/
mmol), [3H]AMPA (66 Ci/mmol), and [3H]-(þ)-MK-801 (22.5 Ci/
mmol) binding experiments were performed following the proce-
dure reported in Refs. [12,62] and [19] respectively. The high
affinity [3H]kainate (58 Ci/mmol), binding assays were performed
on rat cortical membrane, according to a previously described
method [16].
5.3.2. Functional antagonism studies at AMPA and NMDA sites
The mouse cortical wedge preparation described by Mannaioni
et al. [63] was used, while the electrophysiological assays were
performed following the procedures described in Ref. [14].
5.1.18. 7-Trifluoromethyl-3-hydroxy-6-(3-carboxypyrrol-1-yl)-1H-
quinazoline-2,4-dione (11)
Potassium permanganate (0.67 g) was portionwise added to
a cooled (0 ꢀC) suspension of compound 10 (1.54 mmol) in a 1:1
acetone/water mixture (10 mL). Each small addition was made after
the disappearance of the violet color of the oxidizing agent. After
the additions were completed, the mixture was stirred at room
temperature for 1 h and 30 min, then the excess of potassium
permanganate was quenched with a 38% solution of sodium
hydrogen sulfite and the solution acidified with 6 N HCl. The solid
was collected, washed with water and recrystallized. Yield: 75%;
mp 210e212 ꢀC (Cyclohexane/EtOAc); 1H NMR: 6.57 (s, 1H, pyrrole
proton), 7.02 (s, 1H, pyrrole proton), 7.52 (s, 1H, ar), 7.65 (s, 1H,
pyrrole proton), 7.94 (s, 1H, ar), 10.92 (s, 1H, OH or NH), 12.02 (s, 1H,
NH or OH), 12.07 (br s, 1H, COOH). 13C NMR: 110.68 (C pyrrole),
115,09 (C8), 117.79, 118.02, 122.80 (q CF3), 125.79 (C pyrrole), 129.07
(C5 and C pyrrole), 130.69 (q C7), 131.61, 138.96, 148.88 (C2), 158.38
(C4), 165.60 (COOH). IR 1680, 1740, 3100e3200. Anal. Calcd for
C14H8F3N3O5: C, 47.34; H, 2.27; N, 11.83. Found: C, 47.07; H, 2.02;
N, 11.61.
5.3.3. Sample preparation and result calculation
A stock 1 mM solution of the tested compound was prepared in
50% DMSO and the subsequent dilution was accomplished in buffer.
The IC50 values were calculated from three or four displacement
curves based on four to six scalar concentrations of the tested
compound in triplicate using the ALLFIT computer program [64]. Ki
values were calculated according to the ChengePrusoff equation
[65]. Under our experimental conditions, the dissociation constant
(KD) for [3H]glycine and [3H]-DL-AMPA were 75 ꢂ 6 and 28 ꢂ 3 nM,
respectively.
5.3.4. Oxygeneglucose deprivation (OGD) in rat organotypic
hippocampal slice cultures
Organotypic hippocampal slice cultures were prepared from
seven-day-old Wistar rats, used at 14 DIV (days in vitro) and
exposed to OGD as previously described [26]. Briefly, the slices
were preincubated for 30 min in serum-free medium and then

V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
481
subjected to OGD by exposing them to a serum-free medium
devoid of glucose and previously saturated with 95% N2/5% CO2.
Following 30 min incubation at 37 ꢀC in an airtight anoxic chamber,
the cultures were transferred to oxygenated serum-free medium
All drugs were stored at ꢁ20 ꢀC in stock solutions at
1000e10,000 times the desired final concentration, dissolved in
aCSF and applied to the slice by superfusion. Compound 3 was
dissolved in dimethylsulfoxide (DMSO) and stock solutions were
made to obtain concentrations of DMSO lower than 0.001% in the
superfusing aCSF. Control experiments, carried out in parallel,
showed that this concentration of DMSO did not affect CA1
hippocampal neurotransmission. Changes in superfusing solutions
reached the preparation in 60 s and this delay was taken into
account in our calculations.
containing 5 mg/mL glucose and 5 mg/mL propidium iodide (PI) and
returned to the incubator under normoxic conditions until
neuronal injury was evaluated 24 h later. In this system, 30-min
OGD induced a CA1 pyramidal cell damage that was approxi-
mately 65% of the maximal damage achieved by exposing the slices
for 24 h to 10 mM glutamate. Test compounds were added to
cultures during OGD and maintained for the subsequent 24 h. A
stock 1 mM solution of compound 8 was prepared in 50% DMSO
and the subsequent dilution was accomplished in buffer.
Cell injury in the CA1 hippocampal region, was assessed
according to Ref. [26], using PI, a polar dye which enters the cells
only if the membrane is damaged and becomes fluorescent upon
binding to DNA. Background PI fluorescence was determined in
control cultures not exposed to OGD and was subtracted to all
experimental values. There was a linear correlation between CA1 PI
fluorescence intensity and the number of injured CA1 pyramidal
cells as detected by morphological criteria.
5.3.5.3. Data analysis. Data were analyzed using WinLTP 1.11
reanalysis program and the software package GRAPHPAD PRISM
(version 4.0; GraphPad Software, San Diego, CA, USA). All numerical
data are expressed as mean ꢂ SEM. Statistical significance was
evaluated using paired two-tailed Student’s t-test and One-way
ANOVA. Differences were considered significant at p < 0.05.
5.3.6. Anticonvulsant studies
Male Swiss albino mice (23e30 g) from Harlan (Varese)
breeding farm were used. Fifteen mice were housed per cage. The
cages were placed in the experimental room 24 h before the test for
acclimatization. The animals were kept at 23 ꢂ 1 ꢀC with a 12 h
light/dark cycle, light at 7 a.m., with food and water ad libitum.
Drugs: Diazepam (Valium 10 e Roche), Pentylenetetrazole (PTZ)
(Sigma). PTZ was dissolved in isotonic (0.9% NaCl) saline solution
and injected by the s.c. route. Compound 8 was dispersed in 1%
sodium carboxymethylcellulose (CMC) immediately before use.
Drug concentrations were prepared in such a way that the neces-
sary dose could be administered in a volume of 10 ml kgꢁ1 by the
p.o route. PTZ (90 mg/kg s.c.) was injected 30 min after the
administration of drugs. The frequency of the occurrence of clonic
generalized convulsions was noted over a period of 30 min.
5.3.5. Oxygeneglucose deprivation (OGD) in rat acute hippocampal
slices
5.3.5.1. Electrophysiology. Experiments were carried out on acute
hippocampal slices as previously described [44]. Briefly, male
Wistar rats (Harlan, Udine, Italy, 1-2 month-old) were killed under
isoflurane-induced anesthesia and their hippocampi were rapidly
removed and placed in ice-cold oxygenated (95% O2e5% CO2)
artificial cerebrospinal fluid (aCSF). Rat brains and slices were
maintained in aCSF containing (in mM): NaCl 125, KCl 3, NaH2PO4
1.25, MgSO4 1, CaCl2 2, NaHCO3 25 and
D-glucose 10. Parasagittal rat
hippocampal slices (400 m thick) were prepared using a McIlwain
m
tissue chopper (The Mickle Lab. Engineering, Co. Ltd., Gomshall,
UK) and kept in oxygenated aCSF for at least 1 h at room temper-
ature. A single slice was then placed on a nylon mesh in a small
chamber (0.8 mL), completely submerged and superfused with
oxygenated aCSF (31e32 ꢀC) at a constant flow rate of 1.5e1.8 mL/
Acknowledgements
The synthetic work was supported by a grant of the University of
Florence and the Italian Ministry for University and Research, Italy.
The molecular modeling work coordinated by S.M. has been carried
out with financial support from the University of Padova, Italy, and
the Italian Ministry for University and Research, Rome, Italy.
min. Test pulses (100 ms, 0.066 Hz) were delivered through a bipolar
nichrome electrode positioned in the CA1 stratum radiatum in
order to stimulate Schaffer collateral/commissural fibers. Evoked
extracellular fepsps were recorded with glass microelectrodes
(2e10 MU, Harvard Apparatus LTD, Edenbridge, UK) filled with
References
150 mM NaCl and placed in the stratum radiatum of the CA1
hippocampal region. Responses were amplified (BM 622, Mangoni,
Pisa, Italy), digitized (sample rate, 33.33 kHz), and stored for later
analysis with LTP (version 2.30D) program [66].
[1] H. Bräuner-Osborne, J. Egebjerg, E.Ø. Nielsen, U. Madsen, P. Krogsgaard-Larsen,
J. Med. Chem. 43 (2000) 2609e2654.
[2] S.F. Traynelis, L.P. Wollmuth, C.J. McBain, F.S. Menniti, K.M. Vance, K.K. Odgen,
K.B. Hansen, H. Yuan, S.J. Myers, R. Dingledine, Pharmacol. Rev. 62 (2010)
405e496.
Stimuluseresponse curves were obtained by gradually
increasing stimulus strength at the beginning of each experiment,
until a stable baseline of evoked response was reached. The test
stimulus pulse was then adjusted to produce a fepsp, whose slope
and amplitude was 40e50% of the maximum and was kept constant
throughout the experiment. The fepsp amplitude was routinely
measured, and expressed as the percentage of the average ampli-
tude of the potentials measured during the 5-min preceding
exposure of the hippocampal slices to drug treatment (pre-drug) or
OGD (pre-ischemic).
[3] G.G. Somjen, Physiol. Rev. 81 (2001) 1065e1096.
[4] M.R. Hynd, H.L. Scott, P.R. Dodd, Neurochem. Int. 45 (2004) 583e595.
[5] W.M. Caudle, J. Zhang, Exp. Neurol. 220 (2009) 230e233.
[6] J.C. Corona, L.B. Tovar-y-Romo, R. Tapia, Expert Opin. Ther. Targets 11 (2007)
1415e1428.
[7] M.A. Rogawski, Epilepsy Currents 11 (2011) 56e63.
[8] N.G. Walgren, Int. Rev. Neurobiol. 40 (1997) 337e363.
[9] K.W. Muir, Curr. Opin. Pharmacol. 6 (2006) 53e60.
[10] V. Colotta, D. Catarzi, F. Varano, L. Cecchi, G. Filacchioni, A. Galli, C. Costagli,
Arch. Pharm. Pharm. Med. Chem. 330 (1997) 129e134.
[11] D. Catarzi, V. Colotta, F. Varano, L. Cecchi, G. Filacchioni, A. Galli, C. Costagli,
J. Med. Chem. 42 (1999) 2478e2484.
[12] D. Catarzi, V. Colotta, F. Varano, L. Cecchi, G. Filacchioni, A. Galli, C. Costagli,
V. Carlà, J. Med. Chem. 43 (2000) 3824e3826.
[13] D. Catarzi, V. Colotta, F. Varano, L. Cecchi, G. Filacchioni, A. Galli, C. Costagli,
V. Carlà, J. Med. Chem. 44 (2001) 3157e3165.
[14] F. Varano, D. Catarzi, V. Colotta, G. Filacchioni, A. Galli, C. Costagli, V. Carlà,
J. Med. Chem. 44 (2002) 1035e1044.
[15] V. Colotta, D. Catarzi, F. Varano, F.R. Calabri, G. Filacchioni, C. Costagli, A. Galli,
Bioorg. Med. Chem. Lett. 14 (2004) 2345e2349.
5.3.5.2. Application of OGD and drugs in rat acute hippocampal
slices. OGD was obtained by superfusing the slice with aCSF
without glucose and gassed with nitrogen (95% N2 -5% CO2) [67].
This caused a drop in pO2 in the recording chamber from w500 mm
Hg (normoxia) to a range of 35e75 mm Hg (after 7-min OGD) [44].
At the end of the ischemic period, the slice was again superfused
with normal, glucose containing, oxygenated aCSF.
[16] D. Catarzi, V. Colotta, F. Varano, F.R. Calabri, G. Filacchioni, A. Galli, C. Costagli,
V. Carlà, J. Med. Chem. 47 (2004) 262e272.

482
V. Colotta et al. / European Journal of Medicinal Chemistry 54 (2012) 470e482
[17] F. Varano, D. Catarzi, V. Colotta, F.R. Calabri, O. Lenzi, G. Filacchioni, A. Galli,
C. Costagli, F. Deflorian, S. Moro, Bioorg. Med. Chem. 13 (2005) 5536e5549.
[18] V. Colotta, D. Catarzi, F. Varano, O. Lenzi, G. Filacchioni, C. Costagli, A. Galli,
C. Ghelardini, N. Galeotti, P. Gratteri, J. Sgrignani, F. Deflorian, S. Moro, J. Med.
Chem. 49 (2006) 6015e6026.
[19] F. Varano, D. Catarzi, V. Colotta, L. Cecchi, G. Filacchioni, A. Galli, C. Costagli,
Eur. J. Med. Chem. 36 (2001) 203e209.
[20] D. Catarzi, O. Lenzi, V. Colotta, F. Varano, D. Poli, G. Filacchioni, K. Lingenhöhl,
S. Ofner, Chem. Pharm. Bull. 58 (2010) 908e911.
[42] S.L. Dargan, V.R. Clarke, G.M. Alushin, J.L. Sherwood, R. Nisticò, Z.A. Bortolotto,
A.M. Ogden, D. Bleakman, A.J. Doherty, D. Lodge, M.L. Mayer, S.M. Fitzjohn,
D.E. Jane, G.L. Collingridge, Neuropharmacology 56 (2009) 121e130.
[43] J.F. Blake, M.W. Brown, G.L. Collingridge, Br. J. Pharmacol. 95 (1988) 291e299.
[44] A.M. Pugliese, E. Coppi, G. Spalluto, R. Corradetti, F. Pedata, Br. J. Pharmacol.
147 (2006) 524e532.
[45] A.M. Pugliese, E. Coppi, R. Volpini, G. Cristalli, R. Corradetti, L.S. Jeong,
K.A. Jacobson, F. Pedata, Biochem. Pharmacol. 74 (2007) 768e779.
[46] C. Traini, F. Pedata, S. Cipriani, T. Mello, A. Galli, M.G. Giovannini, F. Cerbai,
R. Volpini, G. Cristalli, A.M. Pugliese, Eur. J. Neurosci. 33 (2011) 2203e2215.
[47] V. Colotta, D. Catarzi, F. Varano, O. Lenzi, G. Filacchioni, C. Martini,
L. Trincavelli, O. Ciampi, C. Traini, A.M. Pugliese, F. Pedata, E. Morizzo, S. Moro,
Bioorg. Med. Chem. 16 (2008) 6086e6102.
[21] F.R. Calabri, V. Colotta, D. Catarzi, F. Varano, O. Lenzi, G. Filacchioni, C. Costagli,
A. Galli, Eur. J. Med. Chem. 40 (2005) 897e907.
[22] P.D. Leeson, L.L. Iversen, J. Med. Chem. 37 (1994) 4053e4067.
[23] C.F. Bigge, S.S. Nikam, Exp. Opin. Ther. Pat. 7 (1997) 1099e1114.
[24] M. Koller, K. Lingenhoehl, M. Schmutz, I.-T. Vranesic, J. Kallen, Y.P. Auberson,
D.A. Carcache, H. Mattes, S. Ofner, D. Orain, S. Urwyler, Bioorg. Med. Chem.
Lett. 21 (2011) 3358e3361.
[48] V. Colotta, O. Lenzi, D. Catarzi, F. Varano, G. Filacchioni, C. Martini,
L. Trincavelli, O. Ciampi, A.M. Pugliese, C. Traini, F. Pedata, E. Morizzo, S. Moro,
J. Med. Chem. 52 (2009) 2407e2419.
[25] StarDrop. Optibrium Ltd., 7226 IQ Cambridge, Beach Drive, Cambridge, CB25
9TL, UK. http://www.optibrium.com/.
[49] S. Yamamoto, E. Tanaka, Y. Shoji, Y. Kudo, H. Inokuchi, H. Higashi,
J. Neurophysiol. 78 (1997) 903e911.
[26] D.E. Pellegrini-Giampietro, A. Cozzi, F. Peruginelli, P. Leonardi, E. Meli,
R. Pellicciari, F. Moroni, Eur. J. Neurosci. 11 (1999) 3637e3647.
[27] M. Montero, M. Nielsen, L.C.B. Rønn, A. Møller, J. Noraberg, J. Zimmer, Brain
Res. 1177 (2007) 124e135.
[50] D.S. Pei, Q.H. Guan, Y.F. Sun, Q.X. Zhang, T.L. Xu, G.Y. Zhang, J. Neurosci. Res. 82
(2005) 642e649.
[51] D.S. Pei, X.T. Wang, Y. Liu, Y.F. Sun, Q.H. Guan, W. Wang, J.Z. Yan, Y.Y. Zong,
T.L. Xu, G.Y. Zhang, Brain 129 (2006) 465e479.
[28] G.L. Collingridge, R.W. Olsen, J. Peters, M. Spedding, Neuropharmacology 56
(2009) 2e5.
[29] R. Chittajallu, M. Vignes, K.K. Dev, J.M. Barnes, G.L. Collingridge, J.M. Henley,
Nature 379 (1996) 78e81.
[52] A.M. Buchan, H. Li, S. Cho, W.A. Pulsinelli, Neurosci. Lett. 132 (1991) 255e258.
[53] E. Le Peillet, B. Arvin, C. Moncada, B.S. Meldrum, Brain Res. 571 (1992) 115e120.
[54] B. Nellgard, T. Wieloch, J. Cereb. Blood Flow Metab. 12 (1992) 2e11.
[55] H. Furukawa, E. Gouaux, EMBO J. 22 (2003) 2873e2885.
[30] J. Lerma, Nat. Rev. Neurosci. 4 (2003) 481e495.
[31] D.E. Jane, D. Lodge, G.L. Collingridge, Neuropharmacology 56 (2009) 90e113.
[32] J. Lerma, Nat. Neurosci. 14 (2011) 808e810.
[33] H. Herb, N. Burnashev, P. Werner, B. Sakman, W. Wisden, P.H. Seeburg,
Neuron 8 (1992) 775e785.
[34] P. Werner, M. Voight, K. Keinänem, W. Wisden, P.H. Seeburg, Nature 351 (1991)
742e744.
[56] MOE (Molecular Operating Environment), version 2008.10; software available
from Chemical Computing Group Inc. (1010 Sherbrooke Street West, Suite
910, Montreal, Quebec, Canada H3A 2R7); http://www.chemcomp.com.
[57] M. Wojciechowski, B. Lesyng, J. Phys. Chem. B 108 (2004) 18368e18376.
[58] W.D.C. Cornell, C.I. Bayly, I. Gould, K.M. Merz, D.M. Ferguson, D.C. Spellmeyer,
T. Fox, J.W. Caldwell, P.A. Kollman, J. Am. Chem. Soc. 117 (1995) 5179e5196.
[59] C.A. Baxter, C.W. Murray, D.E. Clark, D.R. Westhead, M.D. Eldridge, Proteins
Struct. Funct. Genet. 33 (1998) 367e382.
[35] B. Bettler, J. Egebjerg, G. Sharma, G. Pecht, I. Hermans-Borgmeyer, C. Moll,
C.F. Stevens, S. Heinemann, Neuron 8 (1992) 257e265.
[60] T. Halgren, J. Comput. Chem. 17 (1996) 490e519.
[36] B. Sommer, N. Burnashev, T.A. Verdoon, K. Keinanem, B. Sakmann,
P.H. Seeburg, EMBO J. 11 (1992) 1651e1656.
[37] T.H. Johansen, J. Drejer, F. Wätjen, E.Ø. Nielsen, Eur. J. Pharmacol. 246 (1993)
195e204.
[38] G.L. Collingridge, S.J. Kehl, H. McLennan, J. Physiol. 334 (1983) 33e46.
[39] H. Kamiya, S. Ozawa, J. Physiol. 509 (1998) 833e845.
[40] V.R. Clarke, G.L. Collingridge, Neuropharmacology 42 (2002) 889e902.
[41] V.R. Clarke, G.L. Collingridge, Neuropharmacology 47 (2004) 363e372.
[61] J.J.P. Stewart, MOPAC 7, Fujitsu Limited, Tokyo, Japan, 1993.
[62] E.O. Nielsen, U. Madsen, K. Schaumburg, L. Brehm, P. Krogsgaard-Larsen, Eur. J.
Chem. Chim. Ther. 21 (1986) 433e437.
[63] G. Mannaioni, V. Carlà, F. Moroni, Br. J. Pharmacol. 118 (1996) 1530e1536.
[64] A. De Lean, P.J. Munson, D. Rodbared, Am. J. Physiol. 235 (1978) E97e102.
[65] Y.C. Cheng, W.H. Prusoff, Biochem. Pharmacol. 22 (1973) 3099e3108.
[66] W.W. Anderson, G.L. Collingridge, J. Neurosci. Methods 108 (2001) 71e83.
[67] F. Pedata, S. Latini, A.M. Pugliese, G. Pepeu, J. Neurochem. 61 (1993) 284e289.