Novel quinolinone-phosphonic acid AMPA antagonists devoid of nephrotoxicity.

Article abstract of DOI:10.1016/S0014-827X(02)01281-8

We reported previously the synthesis and structure-activity relationships (SAR) in a series of 2-(1H)-oxoquinolines bearing different acidic functions in the 3-position. Exploiting these SAR, we were able to identify 6,7-dichloro-2-(1H)-oxoquinoline-3-phosphonic acid compound 3 (S 17625) as a potent, in vivo active AMPA antagonist. Unfortunately, during the course of the development, nephrotoxicity was manifest at therapeutically effective doses. Considering that some similitude exists between S 17625 and probenecid, a compound known to protect against the nephrotoxicity and/or slow the clearance of different drugs, we decided to synthesise some new analogues of S 17625 incorporating some of the salient features of probenecid. Replacement of the chlorine in position 6 by a sulfonylamine led to very potent AMPA antagonists endowed with good in vivo activity and lacking nephrotoxicity potential. Amongst the compounds evaluated, derivatives 7a and 7s appear to be the most promising and are currently evaluated in therapeutically relevant stroke models.

Full text of DOI:10.1016/S0014-827X(02)01281-8

Il Farmaco 57 (2002) 787ꢀ802
/
www.elsevier.com/locate/farmac
Novel quinolinone-phosphonic acid AMPA antagonists devoid of
nephrotoxicity
Alex A. Cordi a,*, Patrice Desos a, Elisabeth Ruano a, Hashim Al-Badri b,1
,
Claude Fugier c, Astrid G. Chapman d, Brian S. Meldrum d, Jean-Yves Thomas e,
Anita Roger e, Pierre Lestage e
a Institut de Recherches Servier, 11, rue des Moulineaux, F-92150 Suresnes, France
b IRCOF, 76821 Mont-Saint-Aignan Cedex, France
c ORIL Industrie, 13, rue Auguste Desgenetais, BP 17, 76210 Bolbec, France
d Department of Neurology, Institute of Psychiatry, De Crespigny Park, London SE5 8AF, United Kingdom
e Institut de Recherches Servier, 125, chemin de Ronde, 78290 Croissy, France
Received 29 March 2002; accepted 10 May 2002
Abstract
We reported previously the synthesis and structureꢀ
/activity relationships (SAR) in a series of 2-(1H)-oxoquinolines bearing
different acidic functions in the 3-position. Exploiting these SAR, we were able to identify 6,7-dichloro-2-(1H)-oxoquinoline-3-
phosphonic acid compound 3 (S 17625) as a potent, in vivo active AMPA antagonist. Unfortunately, during the course of the
development, nephrotoxicity was manifest at therapeutically effective doses. Considering that some similitude exists between S
17625 and probenecid, a compound known to protect against the nephrotoxicity and/or slow the clearance of different drugs, we
decided to synthesise some new analogues of S 17625 incorporating some of the salient features of probenecid. Replacement of the
chlorine in position 6 by a sulfonylamine led to very potent AMPA antagonists endowed with good in vivo activity and lacking
nephrotoxicity potential. Amongst the compounds evaluated, derivatives 7a and 7s appear to be the most promising and are
currently evaluated in therapeutically relevant stroke models.
´
# 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
Keywords: AMPA antagonist; Anticonvulsant; Nephrotoxicity; Phosphonic acid; 2-(1H)-Oxoquinoline
1. Introduction
-Glutamic acid and
receptors which are subdivided in three groups activated
by quisqualate (group I), 2R,4R-APDC (group II) and
-AP4 (group III), respectively [4].
L
L-aspartic acid are the major
L
excitatory amino acid (EAA) neurotransmitters in the
central nervous system [1]. Different receptors for
glutamic acid have been characterised [2,3], while some
were shown to be associated with cationic channels,
others were found to act through G protein coupled
secondary messenger systems. Pharmacologically, the
best defined receptors are the ionotropic channels
Since 1982 [5], numerous competitive and non-com-
petitive NMDA antagonists have been disclosed and
brought into clinical trials to assess their potential as
therapeutic agents in the acute treatment of stroke and
head trauma or the chronic treatment of epileptic seizure
[6]. Unfortunately, strong psychotomimetic side effects
have impeded their development [7,8].
activated by either N-methyl- -aspartic acid (NMDA)
or by a-amino-3-hydroxy-5-methylisoxazole-4-propio-
nic acid (AMPA) and kainic acid, and the metabotropic
D
Thereafter, selective antagonists of the AMPA recep-
tor were evaluated because of their relevance as ther-
apeutic agents for stroke and other ischaemic conditions
where excessive EAA release has excitotoxic action on
neurones [9]. The pharmacology of AMPA/kainate
antagonists and their role in cerebral ischaemia [10,11],
and epilepsy [12] has been reviewed.
* Corresponding author
E-mail address: alex.cordi@fr.netgrs.com (A.A. Cordi).
1
Present address: LCMT, UMR CNRS 6507, IMSRA, 6, Bd du
Mare´chal Juin, 14050 Caen Cedex, France.
´
0014-827X/02/$ - see front matter # 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
PII: S 0 0 1 4 - 8 2 7 X ( 0 2 ) 0 1 2 8 1 - 8

788
A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ802
/
The most potent selective AMPA/kainate antagonists
incorporating some of the salient features of probenecid.
We initially synthesised the 6-unsubstituted analogue of
S 17625 (4), followed by the 6-nitro derivative (5), and
the 6-pyrrolo (6), leading to the 6-sulfonyl amines
belong to the quinoxalinedione class of compounds,
such as NBQX 1, the first selective AMPA/kainate
antagonist as reported by Sheardown [13] and YM90K 2
disclosed by Yamamouchi [14] (Fig. 1). We reported the
synthesis of 2-(1H)-quinolone-3-phosphonic acids [15],
a series of AMPA/kainate antagonists. In a previous
publication [16], we speculated that the hydroxyl of one
of the tautomeric forms of the quinoxalinedione could
be mimicked by an acidic function at the 3 position of a
compounds (7aꢀ
probenecid. Introduction of a 6-aryl substitution is
represented by 8aꢀl, changing the 7-substitution from
chlorine to trifluoromethyl (10aꢀh) or fluorine (11aꢀb)
/u) where 7l is the closest analogue of
/
/
/
was also undertaken. Finally, inversion of the 6 and 7
substitution for the more representative molecules, was
realised by the synthesis of 12aꢀe.
/
quinolone. By exploring the structureꢀactivity relation-
/
ships (SAR) in these series, we were able to identify 6,7-
dichloro-2-(1H)-oxoquinoline-3-phosphonic acid com-
pound 3 (S 17625) as a potent, in vivo active AMPA
antagonist. However, in the course of the extensive
evaluation of the compound in different in vivo models
[17], nephrotoxicity was manifest at therapeutically
effective doses (data not shown). It is known that
probenecid [18] can protect against the nephrotoxicity
[19] and/or slow the clearance [20] of different drugs.
Therefore, coadministration of probenecid and S 17625
was tested and shown to attenuate the nephrotoxicity
(data not shown). The use of probenecid in association
with any EAA antagonist is claimed in a Warner
Lambert patent [21] and specifically, the same group
reports in the literature [22] the advantage of using the
association of probenecid and NBQX. Considering the
structural similarity between S 17625 and probenecid,
we decided to synthesise some new analogues of S 17625
2. Chemistry
Synthesis of all compounds described started with
anthranaldehydes 18aꢀc, some have already been des-
/
cribed [16] (18a and c) or have been synthesised
according to Scheme 1. Thus N-Boc-3-trifluoromethy-
laniline 14 was readily prepared by reaction of appro-
priate aniline 13 with di-tert-butyl dicarbonate in
NaOH/THF. Regioselective introduction of the formyl
function was allowed by the strong ortho metalation
directing NHBoc group [23] to give aldehyde 15 along
with starting compound 14 (60:40). Acidic deprotection
of amine led to polymerisation of the desired an-
thranaldehyde. Therefore, aldehyde function in 15 was
first reduced by sodium borohydride to alcohol 16, then
the amine was deprotected by HCl/dioxane and even-
tually the aldehyde function was regenerated in neutral
conditions by oxidation with activated manganese
dioxide in THF to give anthranaldehyde 18b.
According to Scheme 2, the different anthranalde-
hydes 18aꢀ
cetylchloride to give the amide intermediates 19aꢀ
avoid intramolecular HornerꢀEmmons olefination be-
/c were condensed with diethyl phosphonoa-
/c. To
/
tween the aldehyde and the phosphonoester functiona-
lities, the use of a strong base such as sodium methoxide
was prohibited. Therefore, cyclisations to quinolones
20aꢀc were achieved in the presence of catalytic
/
amounts of piperidine in toluene at reflux. Hydrolysis
of the diethylphosphono ester function proceeds
through successive treatment with trimethylsilylbromide
and methanol affording the phosphonic acids 4 and 10a
Fig. 1. Reference compound structures.
from the esters 20a and 20b. Nitration of 20aꢀ
nitric acid in sulfuric acid occurred selectively at the 6
position, leading to the esters 21aꢀc which were hydro-
lysed as usual to the acids 5 and 10b. Reduction of the
nitro group of 21aꢀc was accomplished without any
dehalogenation by ironꢀamonium chloride [24] as
reducing agent to give in high yields the corresponding
amines 22aꢀc. Amine 22a was sulfonylated by reaction
/c with
/
/
/
/
with methylsulfonyl chloride in pyridine to provide
compound 9 after standard hydrolysis of the diethyl
phosphono ester function. Introduction of the chloro-
sulfonyl group was achieved through diazotation of
Scheme 1. Reagents: (a) (BOC)2O, NaOH×H2O, THF; (b) (i) sec-
BuLi, TMEDA, THF; (ii) DMF; (c) NaBH4, MeOH; (d) HCl,
dioxane; (e) MnO2, THF.
/

A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ
/
802
789
Scheme 2. Reagents: (a) pyridine, toluene; (b) piperidine cat., toluene; (c) HNO3/H2SO4; (d) Fe NH4Cl, EtOH/H2O; (e) (i) NaNO2, AcOH/HCl/H2O,
(ii) SO2, CuCl2× c, 7k, 10d, 10e, 11a, pyridine/CH2Cl2 for 7d,j,l,m, 10g, CH3CN for 7e,g,h, 10f; (g)
2H2O cat., AcOH/H2O; (f) R1R2NH, H2O for 7aꢀ
/
/
(i) TMSBr, CH3CN, (ii) MeOH; (h) ArCO2H, N,N?-carbonyldimidazol, DBU, THF; (i) RCONHNH2, CH2Cl2, pyridine; (j) O-benzylhydrox-
ylamine hydrochloride or O-methylhydroxylamine hydrochloride, pyridine, CH3CN/H2O; (k) BBr3, CH2Cl2; (l) CH3SO2Cl, pyridine.
amine 22aꢀ
/
c with sodium nitrite, followed by reaction
with sulfur dioxide in acetic acid/water in presence of
The amines 22aꢀ
key intermediates in the synthesis of the 6-aryl sub-
stituted quinoleinones (8aꢀi, 10g and 11b) via a
/
c described in Scheme 2 were also
CuCl2 [25] to give the corresponding sulfonylchloride
/
23aꢀ
different amines leading, after hydrolysis, to the corres-
ponding sulfonyl amides 7aꢀm, 10dꢀg and 11a. The
/
c. These sulfonyl chlorides were reacted with
modified Suzuki reaction [26,27] as shown in Scheme
3. Therefore, diazotation with sodium nitrite of the
/
/
aminosulfonyl diethylphosphono ester corresponding to
7a is further acylated by benzoic acid or 4-[(dipropyla-
mino)sulfonyl] benzoic acid in the presence of carbonyl
diimidazole and DBU providing the benzoylaminosul-
fonyl derivatives 7p and 7q, respectively, after hydrolysis
of the diethyl phosphono ester function. The sulfonyl
chloride 23a was also condensed with acyl hydrazides in
the presence of pyridine, affording, after hydrolysis,
the corresponding acyl hydrazinosulfonyl derivatives
7rꢀu. Two sulfohydroxamates 7n and 7o were prepared
/
from 23a by condensation with either O-methyl-
hydroxylamine or O-benzylhydroxylamine. For this
latter, O-debenzylation was conducted with boron
tribromide in dichloromethane to give, after hydrolysis,
compound 7o.
Scheme 3. Reagents: (a) NaNO2, HBF4 H2O; (b) aryl boronic acid,
Pd(OAc)2, MeOH/dioxane; (c) (i) TMSBr, CH3CN; (ii) MeOH; (d) 2,5-
dimethoxytetrahydrofuran, AcOH, H2O, 1,2-dichlorethane.

790
A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ802
/
Scheme 4. Reagents: (a) SO2Cl2, AcOH; (b) (i) NaOH 1N, EtOH, (ii) H3Oꢁ; (c) (ii) CDI, THF, (ii) NaBH4, H2O; (d) MnO2, CH2Cl2; (e) pyridine,
toluene; (f) piperidine cat., toluene; (g) (i) TMSBr, CH3CN, (ii) MeOH; (h) Fe, NH4Cl, EtOH/H2O; (i) (i) NaNO2, AcOH/HCl/H2O, (ii) SO2, CuCl2×
/
2H2O cat.; AcOH/H2O; (j) NH3, H2O for 12c, (n-C3H7)2NH, pyridine/CH2Cl2 for 12d; (k) NaNO2, HBF4 H2O; (l) phenyl boronic acid, Pd(OAc)2,
MeOH/dioxane; (m) 2,5-dimethoxytetrahydrofuran, AcOH, H2O, 1,2-dichlorethane.
amine function in aqueous tetrafluoroboric acid led to
the precipitation of the corresponding arenediazonium
tetrafluoroborate salt which was isolated by simple
filtration and could be stored for an extended period
of time. Cross coupling reaction of this former with
various arylboronic acids was achieved at room tem-
perature with 10 mol% Pd(OAc)2 in anhydrous condi-
tions in the absence of any added base and stabilising
phosphine ligand to produce, after the usual hydrolysis
molecule impedes its further development. Aiming at
avoiding this problem, we focus first on different
modifications of the 6 position of the 2-(1H)-oxoquino-
line. Replacement of the chlorine in this position by
hydrogen (4), nitro (5), pyrrole (6), aminosulfonyl (7a)
and phenyl (8a) yields better (5 and 7a) or equiactive (6
and 8a) compounds except for the hydrogen substitution
(4) which affords a much less active (ten times)
compound (Table 1). Our previous experiences in the
field have led us to keep clear of the nitro substitution as
it has always endowed our compounds with poor
bioavailability properties. So, we embarked in a search
of SARs dedicated to the aminosulfonyl as well as the
aromatic moiety in the 6 position of 7-chloro-2-(1H)-
oxoquinoline-3-phosphonic acid.
step, the corresponding compounds 8aꢀl, 10h and 11b.
/
When condensed with 2,5-dimethoxytetrahydro-
furane in a mixture of aqueous acetic acid and 1,2-
dichloroethane, amines 22a and 22b provided after
hydrolysis, the corresponding pyrroles 6 and 10c,
respectively.
When applied to 18d which was obtained from the
corresponding 5-nitro anthranilic acid methyl ester 25
through chlorination (26), selective reduction (28) and
managed oxidation, the same set of reactions described
In vitro results presented in Table 2 indicate that the
inversion of the aminosulfonyl substitution (7a) to a
sulfonylamino substitution (9) leads to a 40-fold loss of
activity. Additionally, monosubstitution of the nitrogen
in Schemes 2 and 3, afford the compounds 12aꢀd
/
atom by linear alkyl groups (7bꢀd and g) induces a
/
following the identical procedures as shown in
Scheme 4.
monotonous decrease in activity. Ramification of the
alkyl chain makes things even worse with perhaps a
slight advantage for the cyclised forms compared to the
open one (7eꢀ/f). Aryl, aralkyl and amino alkyl subs-
titution (7hꢀj) does not afford any improvement.
/
3. Discussion
Disubstitution of the nitrogen atom gives marginally
less potent molecules as can be seen by comparing the
activity of the isomeric 7c and 7k. Additional results of
in vivo assay are presented in Table 2 and are considered
Results accumulated with S 17625 (3) have convinced
us of the advantages lying in this specific skeleton.
Unfortunately, the nephrotoxicity manifested by this

A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ
/
802
791
Table 1
In vitro AMPA antagonistic activity of compounds 1ꢀ
Table 2
In vitro and in vivo AMPA antagonistic activity of compounds 7aꢀ
and 9
/
8a
/
m
In the Xenopus oocyte assays, IC50 values were derived from inhibition
curves constructed with data from two to four experiments at each
drug concentration. In all cases, enough experiments were performed
to calculate the corresponding standard deviation.
as a useful indication of brain bioavailability of the
compounds. As the ease of crossing the blood brain
barrier follows a different SAR from the one which
governs the interaction with the receptor, the correlation
between the in vitro activity and the in vivo activity is
not straightforward. Indeed, the monomethyl derivative
(7b) is 7 times less active in vitro than the unsubstituted
compound (7a) but only three times less active in vivo.
Even more contrasting, the ethyl derivative (7c) is losing
activity in vitro compared to the methyl substituted 7b
but it is gaining activity in vivo. The cyclopropyl
derivative 7f which is marginally more potent in vitro
than the isopropyl derivative 7e is definitively more
potent in vivo than this compound. In conclusion, from
analysis of the results of Table 2, two contrasting trends
have been detected in vitro and in vivo as a small alkyl
substituent is favoured at the receptor level while larger,
usually more lipophilic, substituents are, as expected,
favoured for crossing the blood brain barrier.
The results presented in Table 3 are partly in contra-
diction with these conclusions. Indeed, comparison of
the in vitro and in vivo activity of compounds 7n and 7o
indicate that at fairly the same level of potency in vitro,
corresponds a two to one ratio for in vivo activity in
favour of the more lipophilic methylsubstituted com-
pound (0.7 unit difference in calculated Log P and one
H-bond donor moiety less). In contrast, the more potent
compound in vitro and in vivo from this Table 3 is the
bulky benzoylhydrazide (7s) which is 1000 times more
potent than its more lipophilic deaza analogue 7p (0.4
unit difference in calculated Log P and one additional
a
ND : not determined. . In the Xenopus oocyte assays, IC50 values
were derived from inhibition curves constructed with data from two to
four experiments at each drug concentration. In all cases, enough
experiments were performed to calculate the corresponding standard
b
deviation. In the mice anticonvulsant testing, HED50 values for
inhibition against sound induced clonic seizures in DBA/2 mice, are
calculated with 95% confidence limits (between brackets), according to
the method of Litchfield and Wilcoxon.
H-bond donor and acceptor entity). Additionally, the
acetylhydrazide (7r) which is six times less potent in
vitro than 7s, is only 1.5 times less potent in vivo while it
is definitively less lipophilic (1.2 unit difference in
calculated Log P and the same number of H-bond
donor and acceptor entities). Phenyl substitution (7t
and u) does not affect in vitro activity but has a dramatic
effect on in vivo activity. Results from the measure of
the activity of these compounds on the other ion channel
where NMDA and glycine act as coagonists, demon-
strate the excellent selectivity for the AMPA receptor, of
the described compounds. Additional negative results at
the level of this receptor were generated for all active
compounds from this report (data not shown).

792
A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ802
/
Table 3
Table 4
In vitro and in vivo AMPA antagonistic activity of compounds 8aꢀl
In vitro and in vivo AMPA antagonistic activity of compounds 7nꢀ
/
u
/
a
na : not active, ND : not determined. In the Xenopus oocyte assays,
IC50 values were derived from inhibition curves constructed with data
from two to four experiments at each drug concentration. In all cases
except for 7p, enough experiments were performed to calculate the
corresponding standard deviation. b In the mice anticonvulsant testing,
HED50 values for inhibition against sound induced clonic seizures in
DBA/2 mice, are calculated with 95% confidence limits (between
brackets), according to the method of Litchfield and Wilcoxon.
ND : not determined. a In the Xenopus oocyte assays, IC50 values were
derived from inhibition curves constructed with data from two to four
experiments at each drug concentration. In all cases except for 8k,
enough experiments were performed to calculate the corresponding
standard deviation. b In the mice anticonvulsant testing, HED50 values
for inhibition against sound induced clonic seizures in DBA/2 mice, are
calculated with 95% confidence limits (between brackets), according to
the method of Litchfield and Wilcoxon.
The discovery that bulky benzoylhydrazide such as in
7s can be accommodated by the AMPA receptor lets us
investigate the outcome of aryl substitutions in position
6 of the 2-(1H)-oxoquinoline (Table 4). Such a substitu-
tion has already been disclosed in a recent Kyorin patent
[28] describing some 1,2-dihydroquinoline-3-carboxylic
acid as AMPA antagonists. Comparing 8a, 8b and 8c,
and analysing the in vitro activity first, it is apparent
that para substitution is favoured over meta substitu-
tion or even over no substitution. The electronic nature
of the substituent has little or no effect (see for instance
8d and 8e, with same IC50 but with strongly electro-
attracting or electrodonating substituents, respectively)
pointing to the hypothesis that, as usual, the interaction
must be van der Waals in nature. A further demonstra-
tion of this character is brought by the activity of 8i, 8j
and 8k which fairly precisely delineate the shape of the
lipophilic pocket. Replacement of the phenyl ring by a
pyrrole (6) or a thiophene (8l) is also tolerated by the
receptor. Unfortunately, meanwhile all the compounds
of this series have very favourable lipophilic character
(1B
/
calculated Log PB3); for some unknown reasons,
/
they all are devoid of any in vivo activity.
So far, we have only considered compounds with a
chlorine substituent in the 7 position. In Table 5,
variation in this position is introduced for the more
representative compounds according to their in vitro
activity. With the exception of the 6-aminosulfonyl
derivatives (8a, 10d and 11a) where the chlorine has a

A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ
/
802
793
definitive advantage over the trifluoromethyl or fluorine
substituent, replacement of chlorine with one of these
two other substituents has only limited influence on the
in vitro activity. Moreover, it has no influence at all on
the in vivo activity, except for the methylaminosulfonyl
compounds 7b and 10e where the trifluoromethyl
substituent endows the molecule with a significant
over two-fold increase in potency in this test.
Table 6
Influence of the 6,7 substitution versus the 7,6 substitution on the in
vitro AMPA antagonistic activity of compounds 12aꢀ
/e
Permutation between the substituents in position 6
and 7 leads to the compounds reported in Table 6. It can
be seen at first glance that in each couple of compounds
described, the more active compound is the one having
the chlorine atom in position 7 and the other substituent
in position 6.
As stated in the introduction, this research was
undertaken with the aim of discovering new AMPA
antagonists endowed with less liability of being nephro-
toxic. This potential side effect was assessed, in the rat,
through monitoring of urea and creatinine plasmatic
Table 5
Influence of the nature of the 7 substitution on the in vitro and in vivo
ND : not determined. a In the Xenopus oocyte assays, IC50 values were
derived from inhibition curves constructed unless otherwise stated,
with data three experiments at each drug concentration. In all cases
except for 12b, 12d and 12e, enough experiments were performed to
calculate the corresponding standard deviation.
AMPA antagonistic activity of compounds 10aꢀf
/
concentrations increase after i.v. administration of
different doses of the compounds. In this test, S 17625
induces a significant increase of urea and creatinine
plasma concentration at the dose of 10 mg kgꢂ1 while
7a is without any effect up to the dose of 30 mg kgꢂ1
(Fig. 2). In general, the aminosulfonyl derivatives 7aꢀu
/
are free from nephrotoxicity up to the dose of 30 mg
kgꢂ1, with the exception of de dimethylamino 7k and
the n-butylamino 7g compounds which exhibit nephro-
toxicity from the dose of 10 mg kgꢂ1. The 6-aryl
substituted compounds 6 and 8aꢀl are even less safe as
/
some of them, such as the pyrrole 6, the thiophene 8l
and the para-cyanophenyl 8f, are endowed with ne-
phrotoxicity from the dose of 3 mg kgꢂ1 (data not
shown). Considering the ratio between the maximal dose
tolerated in the rat, without elevation of urea and
creatinine levels and the DE50 in the audiogenic seizures
in DBA/2 mice as an approximation of the therapeutic
index, it is clearly apparent that we have significantly
increased this ratio as it is calculated as 0.036 for S
ND : not determined. a In the Xenopus oocyte assays, IC50 values were
derived from inhibition curves constructed with data from three to
four experiments at each drug concentration. In all cases, enough
experiments were performed to calculate the corresponding standard
17625 and ꢀ/9 for 7a which represents a 200-fold
improvement.
b
deviation. In the mice anticonvulsant testing, HED50 values for
In conclusion, we have discovered new AMPA
antagonists endowed with potent in vitro and in vivo
activity. Among the products described, the aminosul-
inhibition against sound induced clonic seizures in DBA/2 mice, are
calculated with 95% confidence limits (between brackets), according to
the method of Litchfield and Wilcoxon.

794
A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ802
/
fonyl derivatives 7a and 7s appear to be the most
promising and are currently being evaluated in thera-
peutically relevant stroke models such as the rat MCA.
4.1.2. Audiogenic seizures in DBA/2 mice
DBA/2 mice (Institute of Psychiatry colony, or
purchased from Harlan and Olac), 21ꢀ28 days old,
/
weighing between 7 and 13 g were used. The animals
were housed on a 12 h/12 h light/dark cycle, and allowed
free access to food and water. Test compounds or
vehicle, were administered i.p. (0.1 ml/10 g body weight).
Subsequently body temperature was monitored and
4. Experimental section
maintained at 36ꢀ38 8C. Mice were observed for
/
abnormal behaviour. Anticonvulsant testing [31,32]
was carried out on individual mice under a perspex
dome (58 cm diameter) fitted with an electric door bell
at the apex, generating a sound stimulus of 109 db for a
period of 60 s or until the onset of clonic convulsions.
The sound stimulus produced a sequential seizure
response, consisting of a ‘‘wild running’’ phase latency
4.1. Biology
4.1.1. Xenopus oocyte experiments
Poly(Aꢁ) messenger RNA isolated from rat cerebral
cortex was injected in Xenopus oocytes [29] and the
1ꢀ
latency 10ꢀ
latency 20ꢀ40 s. The occurrence and latency of the
seizures phases was scored. Mice were tested at nꢃ10,
per group. Doses which protected 50% of animals from
clonic seizure (ED50 value with 95% confidence limits)
were calculated by the method of Litchfield and
Wilcoxon [33].
/
4 s, clonic seizure latency 4ꢀ
/
15 s, tonic extension
oocytes were incubated for 2ꢀ
expression. They were then stored at 6ꢀ
(typically 1ꢀ2 weeks). (R,S)-AMPA-induced currents
were recorded in ‘‘OR2’’ medium [30] of composition (in
mM): NaCl 82.5; KCl 2.5; CaCl2 1; MgCl2 1; NaH2PO4
1; HEPES 5; pH 7.4. For the recording of NMDA/
glycine currents, MgCl2 was omitted from the medium
and the concentration of CaCl2 was raised to 2 mM.
Experiments were performed by means of a two
electrode voltage clamp using an Axoclamp 2A ampli-
/
3 days at 18 8C to allow
/30 s, and commonly respiratory arrest,
/
8 8C until use
/
/
/
4.1.3. Evaluation of nephrotoxicity (renal function) in
the anaesthetised adult Fischer 344 rat
Male adult Fischer 344 rats (Iffa Credo, l’Arbresle,
fier. Glass microelectrodes (1ꢀ
M KCl. The holding potential was adjusted to ꢂ
/
3 MV) were filled with 3
/60 mV.
For the initial evaluation of the inhibitory potency of
the compounds, the agonists were applied for 30 s at the
following concentrations (R,S)-AMPA (10 mM) and
glycine (3 mM)/NMDA (30 mM). The compounds were
both applied for 45 s before and 30 s after the
application of agonists. Agonist responses were evalu-
ated at the peak of the inward current. The IC50 values
were calculated from inhibition curves using the rela-
tionship:
France) weighing 200ꢀ240 g, were used. Animals were
/
anaesthetised with pentobarbital (60 mg kgꢂ1 i.p.).
Ninety minutes later, rats were treated with an i.v. bolus
(penis vein) of compounds under study or vehicle
(saline, 2 ml kgꢂ1). Twenty four hours after drug
administration, animals were decapitated and blood
samples were collected. Renal function was determined
by uraemia and creatinaemia measurements using a
COBAS FARA diagnostic automatic analyser. Results
I ꢃImax=(1ꢁ[antagonist]=IC50)nH
(where nH representstheHillslope coefficient)
were expressed as means9SEM of plasmatic urea (nmol
/
lꢂ1) and creatinine (nmol lꢂ1) concentrations. Signifi-
cant differences between control and treated groups
by a non-linear curve fittingprogram:
Fig. 2. Effect of compounds 3 and 7a on creatinine and urea plasmatic concentrations after i.v. administration to the rat.

A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ
/
802
795
using ANOVA and Newmanꢀ
assessed.
/Keuls test have been
tion: 1H NMR (CDCl3) d 1.55 (s, 9H), 7.40 (d, 1H), 7.80
(d, 1H), 8.80 (s, 1H), 10.00 (s, 1H), 10.5 (bs, 1H).
4.2. Chemistry
4.2.3. tert-Butyl 2-(hydroxymethyl)-5-
(trifluoromethyl)phenylcarbamate (16)
Reagents were commercially available and of syn-
thetic grade. 1H NMR spectra were recorded on Bruker
200 or 400 MHz spectrometers and are given in ppm
relative to TMS. Infrared spectra were recorded on a
Bruker Fourier transform spectrometer as nujol emul-
sion. All new substances were monospot by TLC and
exhibited spectroscopic data consistent with the assigned
structures. Elemental analyses (C, H, N) were performed
on a Carlo Erba 1108 instrument and agree with the
To a solution of the above mixture (9.6 g) in MeOH
(35 ml) was added a solution of NaBH4 (1.26 g, 33.2
mmol) in H2O (25 ml) and the reaction was stirred at
room temperature for 3 h. The solvent was removed in
vacuo and the residue was stirred in water and extracted
with ethyl ether. The organic phase was washed with
H2O, brine and dried (MgSO4) to give, after evaporation
of the solvent, an oil (9.2 g) corresponding to the title
compound 16 along with 14 (7:3). This mixture was used
1
calculated values within the 9
points were obtained on a Reichert hot stage microscope
and are uncorrected. Silica gel 60, Merck 230ꢀ400 mesh,
was used for both flash and medium pressure chroma-
tography. TLC were performed on pre-coated 5ꢄ10
/0.4% range. Melting
directly in the next step without further purification: H
NMR (DMSO-d6) d 1.55 (s, 9H), 5.65 (t, 1H), 4.60 (d,
2H), 7.40 (d, 1H), 7.55 (d, 1H), 7.95 (s, 1H), 8.8 (s, 1H).
/
/
4.2.4. [2-Amino-4-(trifluoromethyl)phenyl]methanol
(17)
cm, Merck silica gel 60 F254 plates (layer thickness 0.25
mm).
A 4 M solution of HCl in dioxane (65 ml) was added
to a solution of the above mixture (9.0 g) and the
reaction was stirred at room temperature for 2 h. After
evaporation in vacuo of the solvent, the residue was
taken up in water, 1N NaOH was added and the mixture
was extracted with ethyl acetate. The organic extract
was washed with H2O, brine, and dried (MgSO4) to give,
after evaporation of the solvent the title compound
along with 4 (7:3) as a white solid (4.53 g). This mixture
was used directly in the next step without further
4.2.1. tert-Butyl 3-(trifluoromethyl)phenylcarbamate
(14)
A mixture of 3-(trifluoromethyl)aniline (13, 51.6 g,
0.320 mmol), di(tert-butyl) dicarbonate (81.8 g, 0.375
mmol), 1N NaOH (360 ml) and THF (140 ml) was
stirred at room temperature overnight. The reaction
mixture was diluted with ethyl acetate (200 ml), washed
with 1N HCl, H2O, brine successively and dried
(MgSO4). The solvent was removed in vacuo and the
crude reaction mixture was treated with charcoal in
ethyl acetate. After filtration and evaporation of the
solvent in vacuo, the residue was triturated in a mixture
of heptane/cyclohexane to yield the title compound as a
1
purification: H NMR (CDCl3) d 1.65 (m, 1H), 4.35
(bs, 2H), 4.75 (s, 2H), 6.95 (m, 2H), 7.15 (d, 1H).
4.2.5. 2-Amino-4-(trifluoromethyl)benzaldehyde (18b)
A mixture of alcohol 17 (0.8 g, 4.18 mmol) and
activated MnO2 (0.73 g, 8.36 mmol) in THF (10 ml) was
stirred at reflux for 4 h. The suspension was filtered off
and washed several times with THF. The filtrate was
evaporated to dryness and the residue was purified by
chromatography on silica gel (cyclohexane/ethyl acetate
80/20) to give the aldehyde 18b as a yellow solid (0.25 g,
1
white solid (49.6 g, 59%): H NMR (DMSO-d6) d 1.45
(s, 9H), 7.30 (m, 1H), 7.50 (m, 1H), 7.65 (m, 1H), 7.95
(m, 1H).
4.2.2. tert-Butyl 2-formyl-5-
(trifluoromethyl)phenylcarbamate (15)
A 1.3 M solution of n-butyllithium in hexanes (100
ml, 130 mmol) was added dropwise via syringe to a
solution of TMEDA (15.1 g, 130 mmol) in THF (180
31% yield): m.p. 41ꢀ
/
43 8C; 1H NMR (CDCl3) d 6.5ꢀ
7.0
/
(bs, 2H), 6.9 (d, 1H), 6.95 (dd, 1H), 7.60 (d, 1H), 9.95 (s,
1H).
ml) at ꢂ
(14.8 g, 56.5 mmol) in THF (75 ml) was added and the
reaction was stirred at ꢂ65 8C for 30 min. DMF (3.8
/
65 8C. A solution of the above carbamate 14
4.2.6. Diethyl 2-[(5-chloro-2-formylphenyl)amino]-2-
oxoethylphosphonate (19a)
/
ml, 73.5 mmol) was added and the solution was allowed
to warm to room temperature over 3 h. The reaction
mixture was acidified to pH 5 with 2N HCl at 0 8C and
the mixture was extracted with ethyl ether, washed with
H2O, and dried (MgSO4). The solvent was removed in
vacuo to yield 17.2 g of an oil corresponding to a 6:4
mixture (1H NMR) of the title compound 15 and the
non-formylated starting material 14. This mixture was
used directly in the next step without further purifica-
To a solution of 2-amino-4-chlorobenzaldehyde (18a,
7.91 g, 50.85 mmol) and pyridine (4.73 ml, 58.48 mmol)
in toluene (220 ml) was added dropwise under nitrogen
atmosphere a solution of diethyl 2-chloro-2-oxoethyl-
phosphonate (58.48 mmol) in toluene (20 ml) in keeping
the temperature below 30 8C. After the addition was
complete, the mixture was stirred at room temperature
for 45 min. The mixture was washed with H2O and brine
and dried (MgSO4). The solvent was removed in vacuo

796
A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ802
/
to yield an oil which was purified by chromatography on
silica gel eluted with cyclohexane/ethyl acetate 20/80.
The title compound 19a was obtained (8.04 g, 53% yield)
as a colourless oil: 1H NMR (CDCl3) d 1.35 (t, 6H), 3.1
(d, 2H), 4.25 (m, 4H), 7.25 (dd, 1H), 7.60 (d, 1H), 8.80
(s, 1H), 9.9 (s, 1H), 11.4 (bs, 1H). Anal.
(C13H17ClNO5P) C, H, N, Cl.
4.2.10. Diethyl 6-amino-7-chloro-2-oxo-1,2-
dihydroquinolin-3-ylphosphonate (22a)
A suspension of 21a (6.0 g, 16.63 mmol), iron powder
(9.27 g, 166 mmol), ammonium chloride (8.88 g, 166
mmol) in MeOH (230 ml) and water (75 ml) was stirred
vigorously to reflux for 1 h. The still warm suspension
was filtered through celite and washed several times with
MeOH. The filtrate was concentrated under vacuum,
water was added and the yellow precipitate was filtered
off to give amine 22a (4.0 g, 73%): 1H NMR (DMSO-d6)
d 1.25 (t, 6H), 4.1 (m, 4H), 5.3 (bs, 2H), 7.15 (s, 1H),
7.25 (s, 1H), 8.25 (d, 1H). Anal. (C13H16ClN2O4P) C, H,
N.
4.2.7. Diethyl 7-chloro-2-oxo-1,2-dihydroquinolin-3-
ylphosphonate (20a)
In a flask equipped with a DeanꢀStark apparatus
/
were placed compound 19a (15.0 g, 50.29 mmol),
piperidine (300 ml) and toluene (330 ml). The mixture
was heated to reflux for 4 h, piperidine (100 ml) was
added and heating was continuing for 4 h. The reaction
was left overnight at room temperature. The yellow
precipitate which separated was filtered off, washed with
ethyl ether to give the title compound 20a (11.55 g,
4.2.11. 7-Chloro-6-[(methylsulfonyl)amino]-2-oxo-1,2-
dihydroquinolin-3-ylphosphonic acid (9)
A solution of amine 21a (1.0 g, 3.02 mmol) and
methylsulfonyl chloride (281 ml, 3.62 mmol) in pyridine
(5 ml) was stirred at room temperature overnight. The
solvent was removed in vacuo and the residue was taken
in 1N HCl and extracted with ethyl acetate. The organic
extracts were washed several times with 1N HCl then
with brine and were dried over MgSO4. After evapora-
tion of the solvent in vacuo the residue was triturated in
ethyl ether/ethyl acetate and the solid was filtered off to
give diethyl 7-ch1oro-6-[(methylsulfonyl)amino]-2-oxo-
1,2-dihydroquinolin-3-ylphosphonate (320 mg, 26%):
1H NMR (DMSO-d6) d 1.25 (t, 6H), 3.05 (s, 4H), 4.10
(m, 4H), 7.40 (s, 1H), 8.0 (s, 1H), 8.50 (d, 1H), 9.6 (m,
1H), 12.1 (d, 1H). Anal. (C14H18ClN2O6PS) C, H, N, S,
Cl. The hydrolysis of the diethyl phosphonoester func-
tions was carried out as for the preparation of com-
pound 4 to give title compound 9: 1H NMR (DMSO-d6)
d 3.05 (s, 3H), 7.45 (s, 1H), 7.95 (s, 1H), 8.40 (d, 1H),
9.55 (bs, 1H), 12.0 (bs, 1H). Yield, elemental analysis
and m.p. are given in Table 8.
1
73%): H NMR (CDCl3) d 1.45 (t, 6H), 4.35 (m, 4H),
7.25 (dd, 1H), 7.45 (s, 1H), 7.60 (d, 1H), 8.6 (d, 1H), 12.5
(bs, 1H). Anal. (C13H15ClNO4P) C, H, N, Cl.
4.2.8. 7-Chloro-2-oxo-1,2-dihydroquinolin-3-
ylphosphonic acid (4)
Bromotrimethylsilane (1.67 ml, 12.64 mmol) was
added to a suspension of the above phosphonoester
20a (500 mg, 1.58 mmol) in CH3CN (20 ml) and the
mixture was stirred to reflux for 1 h. The resulting
solution was evaporated to dryness in vacuo and the
residue was taken up in MeOH. The resulting precipitate
was stirred for 10 min and was collected to yield
phosphonic acid 4 as a white solid (307 mg, 75%):
1
m.p.ꢀ
/
300 8C; H NMR (CDCl3) d 4.5 (bs, 2H), 7.25
(dd, 1H), 7.35 (d, 1H), 7.85 (d, 1H), 8.35 (d, 1H), 12.0
(bs, 1H). Anal. (C9H7ClNO4P) C, H, N, Cl.
4.2.12. Diethyl 7-chloro-6-(chlorosulfonyl)-2-oxo-1,2-
dihydroquinolin-3-ylphosphonate (23a)
A solution of sodium nitrite (767 mg, 11.11 mmol) in
water (5 ml) was added dropwise to a stirred solution of
amino 22a (3.34 g, 10.1 mmol) in glacial acetic acid (10
ml) and concentrated HCl (17 ml) cooled to 0 8C. After
addition was complete, the mixture was stirred at 5 8C
for an additional 30 min. This solution was added in
4.2.9. Diethyl 7-chloro-6-nitro-2-oxo-1,2-
dihydroquinolin-3-ylphosphonate (21a)
To a solution of H2SO4 (95%, 13.3 ml) cooled to 5 8C
in an ice/water bath was added dropwise HNO3 (86%,
13.5 ml). Compound 20a (8.83 g, 27.97 mmol) was then
added in several portions to this solution, keeping the
temperature below 5 8C. Upon complete addition the
reaction was stirred at 5 8C for 15 min and allowed to
warm to room temperature. The solution was poured
onto crushed ice and the mixture was stirred until a
precipitate separated. The solid was filtered off, washed
several times with water, dried and recrystallised from
portions to a saturated cold (0ꢀ5 8C) solution of sulfur
/
dioxide in glacial acetic (13.4 ml) and water (2.25 ml), in
the presence of cupric chloride dihydrate (689 mmol,
4.04 mmol). After addition was complete the solution
was stirred at 0ꢀ5 8C for 1 h and then allowed to warm
/
to room temperature for 3 h. The reaction was poured
onto ice and the precipitate was filtered, washed with
water and dried to give the sulfonyl chloride 23a (3.81 g,
91% yield) as a white solid: 1H NMR (DMSO-d6) d 1.30
(t, 6H), 4.15 (m, 4H), 7.30 (s, 1H), 8.30 (s, 1H), 8.65 (d,
1H), 12.1 (bs, 1H). Anal. (C13H14Cl2NO6PS) C, H, N, S.
1
ethanol to give nitro 21a (8.1 g, 80% yield): H NMR
(CDCl3) d 4.5 (bs, 2H), 7.25 (dd, 1H), 7.35 (d, 1H), 7.85
(d, 1H), 8.35 (d, 1H), 12.0 (bs, 1H). Anal. (C13H14ClN2-
O6P) C, H, N, Cl.

A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ
/
802
797
Table 7
Preparation methods for sulfonylamine 7aꢀ
/
u
4.2.13. 6-(Aminosulfonyl)-7-chloro-2-oxo-1,2-
chloro-2-oxo-1,2-dihydroquinolin-3-ylphosphonate (448
1
mg, 78% yield): H NMR (DMSO-d6) d 1.39 (t, 6H),
dihydroquinolin-3-ylphosphonic acid (7a)
To a 28% aqueous ammonia solution (9 ml) was
added in several portions sulfonyl chloride 23a (600 mg,
1.45 mmol) and water (2 ml). The resulting solution was
stirred at room temperature for 1 h. The reaction was
cooled to 5 8C in an ice/water bath and carefully
acidified upon stirring with 3N HCl. Precipitation was
initiated in adding ethyl acetate (10 ml) and the solid
was filtered off to yield diethyl 6-(aminosulfonyl)-7-
4.15 (m, 4H), 7.45 (s, 1H), 8.55 (s, 1H), 8.60 (d, 1H),
7.2ꢀ8.2 (bs, 2H). Anal. (C13H16ClN2O6PS) C, H, N, S.
/
Hydrolysis of the phophonoester functions was achieved
according to the conditions used in the preparation of
phosphonic acid 4 to give title compound 7a (yield,
elemental analysis and m.p. given in Table 7): 1H NMR
(DMSO-d6) d 7.40 (s, 1H), 7.6 (bs, 2H), 8.45 (s, 1H),
8.45 (d, 1H), 12.2 (bs, 1H).

798
A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ802
/
Compounds 7bꢀ
analogous manner from diethyl 7-substituted-6-(chlo-
rosulfonyl)-2-oxo-1,2-dihydroquinolin-3-ylphosphonate
/
m, 10dꢀ
/
g, 11a, were prepared in an
4.2.16. 7-Chloro-6-[(hydroxyamino)sulfonyl]-2-oxo-
1,2-dihydroquinolin-3-ylphosphonic acid (7o)
To a suspension of 7v (310 mg, 0.655 mmol) in
(23aꢀ
/
c) using the appropriate amine and the conditions
CH2Cl2 (30 ml) at ꢂ
BBr3 solution in CH2Cl2 (1.35 ml). The reaction was
allowed to warm to ꢂ20 8C in about 30 min and poured
/
60 8C was added dropwise 1 M
mentioned in Scheme 2. Analytical data and yields are
given in Tables 7 and 9.
/
onto crushed ice. The gelatinous white precipitate was
stirred vigorously in the biphasic mixture and filtered off
to give, after washing several times with water and ethyl
ether diethyl 7-chloro-6-[(hydroxyamino)sulfonyl]-2-
4.2.14. 6-[(Benzoylamino)sulfonyl]-7-chloro-2-oxo-1,2-
dihydroquinolin-3-ylphosphonic acid (7p)
A solution of benzoic acid (111 mg, 0.91 mmol) and
N,N?carbonyldiimidazol (173 mg, 1.06 mmol) in THF
(3 ml) was stirred to reflux for 45 min. A suspension of
diethyl 6-(aminosulfonyl)-7-chloro-2-oxo-1,2-dihydro-
quinolin-3-ylphosphonate (300 mg, 0.76 mmol) and
DBU (125 ml, 0.83 mmol) in THF (3 ml) was stirred at
room temperature for 15 min and to this was added
dropwise the previous solution of activated benzoic acid.
The mixture was stirred at 60 8C for 3 h. The reaction
oxo-1,2-dihydroquinolin-3-ylphosphoiiate (173 mg,
1
62% yield): m.p. 215ꢀ
/
219 8C; H NMR (DMSO-d6) d
1.30 (t, 6H), 4.15 (qt, 4H), 7.45 (s, 1H), 8.60 (s, 1H), 8.65
(d, 1H), 9.70ꢁ9.75 (dꢁd, 1Hꢁ1H), 12.4 (d, 1H). Anal.
/
/
/
(C13H16ClN2O7PS) C, H, N, S. The hydrolysis of the
diethyl phosphonoester functions was carried out as for
the preparation of compound 4 to give title compound
7o: 1H NMR (DMSO-d6) d 7.45 (s, 1H), 8.45 (d,
1H), 8.48 (s, 1H), 9.70 (thin d, 1H), 9.75 (thin d, 1H),
12.25 (s, 1H). Yield, elemental analysis and m.p. given in
Table 7.
Compound 7n was prepared in an analogous manner
from chlorosulfonyl and O-methylhydroxylamine hy-
drochloride. Analytical data and yield are given in
Table 7.
was then cooled to 0ꢀ5 8C in an ice bath, diluted with
/
water, acidified with 1N HCl and extracted with ethyl
acetate. The organic extract was washed with H2O,
brine, and dried (MgSO4). After evaporation of the
solvent the residue was purified by chromatography on
silica gel eluted with a gradient CH2Cl2/MeOH 95/5 to
90/10 to yield diethyl 6-[(benzoylamino)sulfonyl]-7-chlo-
ro-2-oxo-1,2-dihydroquinolin-3-ylphosphonate as
white solid: (305 mg, 80% yield): m.p.ꢀ
300 8C; 1H
NMR (DMSO-d6) d 1.30 (m, 6H), 4.15 (m, 4H), 7.25ꢀ
a
/
4.2.17. 6-[(2-Acetylhydrazino)sulfonyl]-7-chloro-2-oxo-
1,2-dihydroquinolin-3-ylphosphonic acid (7r)
/
7.40 (m, 4H), 7.35 (bs, 1H), 7.90 (m, 2H), 8.45 (s, 1H),
8.55 (d, 1H), 12.15 (bs, 1H). Hydrolysis of the phopho-
noester functions to give 7p was achieved according to
the conditions used in the preparation of phosphonic
acid 4 (yield, elemental analysis and m.p. are given in
A solution of chlorosulfonyl 23a (400 mg, 0.966
mmol) and acetic hydrazide (215 mg, 2.90 mmol) in
CH2Cl2 (10 ml) was stirred at room temperature over-
night. The resulting white precipitate was filtered off
and purified by chromatography on silica gel eluted with
CH2Cl2/MeOH 90/10 to give diethyl 6-[(2-acetylhydra-
zino)sulfonyl]-7-chloro-2-oxo-1,2-dihydroquinolin-3-yl-
1
Table 7). H NMR (DMSO-d6) d 7.40 (s, 1H), 7.50 (m,
2H), 7.65 (m, 1H), 7.90 (m, 2H), 8.60 (d, 1H), 8.75 (s,
1H), 11.5ꢀ
/
14.0 (bs, 1H).
phosphonate as a white solid (208 mg, 48% yield) m.p.
1
Compound 7q was prepared in an analogous manner
from diethyl 6-(aminosulfonyl)-7-chloro-2-oxo-1,2-dihy-
droquinolin-3-ylphosphonate and 4-[(dipropylami-
no)sulfonyl]benzoic acid. Yield, elemental analysis and
m.p. are given in Table 7.
202ꢀ
/
204 8C; H NMR (DMSO-d6) d 1.35 (t, 6H), 1.85
(s, 3H), 4.15 (m, 4H), 7.45 (s, 1H), 8.50 (s, 1H), 8.65 (d,
1H), 9.65 (s, 1H), 10.05 (s, 1H), 12.35 (bs, 1H). The
hydrolysis of the diethyl phosphonoester functions was
carried out as for the preparation of compound 4 to give
title compound 7r: 1H NMR (DMSO-d6) d 1.70 (s, 3H),
7.40 (s, 1H), 8.40 (s, 1H), 8.45 (d, 1H), 9.5 (bs, 1H), 13.0
(bs, 1H), presence of conformers at room temperature.
Anal. (C11H11ClN3O7PS) C, H, N, S, Cl.
4.2.15. Diethyl 6-{[(benzyloxy)amino]sulfonyl}-7-
chloro-2-oxo-1,2-dihydroquinolin-3-ylphosphonate (7v)
A solution of chlorosulfonyl 23a (500 mg, 1.21 mmol),
O-benzylhydroxylamine hydrochloride (231 mg, 1.45
mmol), pyridine (205 ml, 2.54 mmol) in CH3CN (40 ml)
and H2O (0.25 ml) was stirred at room temperature
overnight. The solvent was evaporated in vacuo and the
residue was taken in 1N HCl and ethyl acetate. The
resulting precipitate was filtered off washed with water
Compounds 7sꢀu were prepared in an analogous
/
manner from diethyl 7-chloro-6-(chlorosulfonyl)-2-oxo-
1,2-dihydroquinolin-3-ylphosphonate (23a) using the
appropriate hydrazide. Analytical data and yield are
given in Table 7.
and ethyl ether to give compound 7v as a white solid:
1
4.2.18. 7-Chloro-3-(diethoxyphosphoryl)-2-oxo-1,2-
dihydroquinoline-6-diazonium tetrafluoroborate (24a)
To a 24% tetrafluoroboric acid aqueous solution (4
ml) was added in small portions amino 22a (1.0 g, 3.02
mmol). The suspension was stirred at room temperature
m.p. 233ꢀ
/
236 8C; H NMR (DMSO-d6) d 1.30 (t, 6H),
4.15 (qt, 4H), 4.85 (s, 2H), 7.31 (m, 5H), 7.48 (s, 1H),
8.65 (s, 1H), 8.65 (d, 1H), 10.65 (s, 1H), 12.36 (thin d,
1H). Anal. (C20H22ClN2O7PS) C, H, N, S.

A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ
/
802
799
for 15 min and then cooled to 0 8C in an ice/water bath.
Dropwise addition of sodium nitrite (208 mg, 8.02
mmol) in water (1 ml) resulted in a solution and after
a few minutes a very thick precipitate appeared. Water
was added to allow stirring for further 30 min and the
precipitate was filtered oft washed with a minimum of
water and then with ether to give the diazonium salt 24a
(1.21 g, 93% yield): 1H NMR (DMSO-d6) d 1.25 (t, 6H),
4.15 (q, 4H), 7.65 (s, 1H), 8.70 (d, 1H), 9.3 (s, 1H), 13.1
(bs, 1H).
4.2.19. 7-Chloro-2-oxo-6-phenyl-1,2-dihydroquinolin-3-
ylphosphonic acid (8a)
A suspension of diazonium 24a (500 mg, 1.16 mmol),
phenyl boronic acid (170 mg, 1.39 mmol), Pd(OAc)2 (26
mg, 0.11 mmol) in a mixture of dioxane (50 ml) and
MeOH (5 ml) was stirred at room temperature for 1 h.
The reaction was diluted with ethyl acetate and filtered
off. The filtrate was washed with water and brine, after
evaporation of the solvent in vacuo, the residue was
purified by chromatography on silica gel eluted with
Table 8
Preparation methods for the 6-aryl quinoleinone 8aꢀl
/

800
A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ802
/
Table 9
Preparation methods for 6-substituted-7-trifuoromethyl or 7-fluoro-quinoleinones 10aꢀf
/
CH2Cl2/MeOH (96/4) to yield diethyl 7-chloro-2-oxo-6-
phenyl-1,2-dihydroquinolin-3-ylphosphonate (207 mg,
1H), 8.55 (d, 1H), 12.25 (s, 1H). Anal. (C17H18ClN2O4P)
C, H, N, Cl. Phosphonate ester functions were hydro-
lysed in the same conditions used for preparation of
compound 4 to give phosphonic acid 6: 1H NMR
(DMSO-d6) d 6.25 (m, 1H), 7.00 (m, 1H), 7.50 (s, 1H),
8.00 (s, 1H), 8.40 (1H, d), 12.1 (bs, 1H). Analytical data
and yield are given in Table 8.
1
46% yield) as a white solid: H NMR (DMSO-d6) d
1.30 (t, 6H), 4.10 (m, 4H), 7.45 (m, 6H), 7.95 (s, 1H),
8.50 (d, 1H), 12.1 (bs, 1H). Anal. (C19H19ClNO4P) C, H,
N, Cl. Phosphonate ester functions were hydrolysed in
the same conditions used for preparation of compound
1
4: H NMR (DMSO-d6) d 7.45 (m, 6H), 7.90 (s, 1H),
8.40 (d, 1H), 12.0 (bs, 1H). Analytical data and yield are
given in Table 8.
4.2.21. Methyl 2-amino-5-chloro-4-nitrobenzoate (26)
To a suspension of methyl 2-amino-4-nitrobenzoate
(25, 23.4 g, 0.119 mol) in glacial acetic acid (1200 ml)
was added dropwise a solution of sulfuryl dichloride (9.6
ml, 119 mmol) in glacial acetic acid (20 ml). The
suspension was stirred at 30 8C until complete dissolu-
tion of the reaction mixture. The solvent was removed in
vacuo and the residue was purified by chromatography
on silica gel eluted with CH2Cl2 to give the title
compound 26 (10.1 g, 37% yield) as an orange solid:
1H NMR (CDCl3) d 3.9 (s, 3H), 7.1 (s, 1H), 8.0 (s, 1H).
Compounds 8aꢀl, 10g, 11b were prepared in an
/
analogous manner from 7-substituted-3-(diethoxypho-
sphoryl)-2-oxo-1,2-dihydroquinoline-6-diazonium tetra-
fluoroborate (24aꢀc) using the appropriate aryl boronic
/
acid. Analytical data and yields are given in Tables 8
and 9.
4.2.20. 7-Chloro-2-oxo-6-(1H-pyrrol-1-yl)-l,2-
dihydroquinolin-3-ylphosphonic acid (6)
A suspension of amine 22a (2.43 g, 7.34 mmol) and
2,5-dimethoxy-tetrahydrofurane (1.33 ml, 10.29 mmol)
in a mixture of water (24 ml), acetic acid (12 ml) and 1,2-
dichloroethane (36 ml) was stirred at 75 8C for 1 h 30
min. The reaction was extracted (CH2Cl2), washed with
water, brine, dried (MgSO4) to give an oil which was
purified by chromatography on silica gel eluted with
CH2Cl2/MeOH (95/5). Diethyl 7-chloro-2-oxo-6-(1H-
pyrrol-1-yl)-1,2-dihydroquinolin-3-ylphosphonate crys-
4.2.22. 2-Amino-5-chloro-4-nitrobenzoic acid (27)
Ester 26 (10.1 g, 4308 mmol) was hydrolysed in a
mixture of 1N NaOH (250 ml) and ethanol (30 ml) at
reflux for 2 h. After acidification of the reaction with 1N
HCl, the precipitate was filtered off to give acid 27 (9.5
1
g, 100% yield) as a yellow solid: m.p. 160 8C; H NMR
(CDCl3) d 7.35 (s, 1H), 7.85 (s, 1H).
tallised from isopropyl ether (1.68 g, 58% yield): m.p.
1
4.2.23. (2-Amino-5-chloro-4-nitrophenyl)methanol (28)
A solution of acid 27 (1.74 g, 8.03 mmol) and N,N?-
carbonyldiimidazol (1.95 g, 12.04 mmol) in THF (50 ml)
193ꢀ
/
196 8C; H NMR (DMS0-d6) d 1.25 (m, 6H), 4.15
(m, 4H), 6.25 (m, 2H), 7.00 (m, 2H), 7.50 (s, 1H), 8.1 (s,

A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ
/
802
801
Table 10
Preparation methods for 7-substituted quinoleinone 12aꢀ
/
f
was stirred at room temperature for 3 h. This solution
was then added dropwise to a solution of NaBH4 (1.52
g, 40.15 mmol) in water (35 ml). The reaction was stirred
at room temperature for 30 min. Hydride excess was
hydrolysed with 3N HCl and the medium was alkali-
nised with 1N NaOH and extracted with ethyl acetate.
The organic phase was washed with water and brine,
dried over MgSO4 and the solvent was evaporated in
vacuo. The residue was taken up in water and filtered off
to give after drying, alcohol 28 (1.41 g, 86% yield) as a
Noel Collignon and Jean-Charles Quirion, and Profes-
sor Jacques Reisse for fruitful discussions.
References
[1] P. Krogsgaard-Larsen, Amino acid receptors, in: C. Hansh (Ed.),
Comprehensive Medicinal Chemistry, vol. III, Pergamon Press,
Oxford, 1990, pp. 521ꢀ537.
/
[2] M. Hollmann, A. O’Shea-Greenfield, S.W. Rogers, S. Heine-
niann, Cloning by functional expression of a member of the
yellow solid: m.p. 123ꢀ
/
130 8C; 1H NMR (CDCl3ꢁ
/
glutamate receptor family, Nature 342 (1989) 643ꢀ648.
/
DMSO-d6) d 2.85 (m, 1H), 4.50 (s, 2H), 5.0 (m, 2H),
7.2 (s, 2H).
[3] K. Moriyoshi, M. Masu, T. Ishii, R. Shigemoto, N. Mizuno, S.
Nakanishi, Molecular cloning and characterization of the rat
NMDA receptor, Nature 354 (1991) 31ꢀ37.
/
[4] P.J. Conn, J.-P. Pin, Pharmacology and functions of metabo-
tropic glutamate receptors, Annu. Rev. Pharmacol. Toxicol. Sci.
4.2.24. 2-Amino-5-chloro-4-nitrobenzaldehyde (18d)
A suspension of alcohol 28 (1.4 g, 6.9 mmol) and
activated MnO2 (3.33 g, 34.5 mmol) in a mixture of
CH2Cl2 (50 ml) and THF (10 ml) was stirred at room
temperature for 2 h. The suspension was filtered off and
washed several times with THF. The filtrate was
evaporated to dryness and the residue was taken up in
heptane and filtered off to give the title compound 18d
37 (1997) 205ꢀ237.
/
[5] M.J. Croucher, J.F. Collins, B.S. Meldrum, Anticonvulsant
action of excitatory amino acid antagonists, Science 216 (1982)
899ꢀ901.
/
[6] A.S. Troupin, J.R. Mendius, F. Cheng, M.W. Risinger, MK-801,
in: B.S. Meldrum, R.J. Porter (Eds.), Current Problems in
Epilepsy 4: New Anticonvulsant Drugs, John Libbey, London,
1986, pp. 191ꢀ201.
/
[7] P.L. Herrling, Competitive NMDA antagonists: results from
animal models and first results in humans, J. Neurochem. 61
(Suppl.) (1993) S283C.
(1.12 g, 81% yield) as an orange solid: m.p. 160ꢀ162 8C;
/
1H NMR (CDCl3) 36.35 (m, 2H), 7.05 (s, 1H), 7.06 (s,
1H), 9.95 (s, 1H).
[8] W.M. Clark, B.M. Coull, Randomized trial of CGS 19755, a
glutamate antagonist, in acute ischemic stoke treatment, Neurol-
ogy 44 (Suppl. 2) (1994) A270.
Compounds 12aꢀe (Scheme 4) were obtained from
/
18d in following the same procedure described for
compounds of Schemes 2 and 3. Yields, elemental
analysis and melting points are given in Table 10.
[9] M.J. Sheardown, P.D. Suzdak, L. Nordholm, AMPA but not
NMDA, receptor antagonism is neuroprotective in gerbil global
ischaemia, even when delayed 24 h, Eur. J. Pharmacol. 236 (1993)
347.
[10] R. Gill, The pharmacology of a-amino-3-hydroxy-5-methyl-4-
isoxazole propionate (AMPA)/kainate antagonists and their role
in cerebral ischaemia, Cerebrovasc. Brain Metabol. Rev. 6 (1994)
Acknowledgements
225ꢀ
[11] A. Chimirri, R. Gotto, M. Zappala`, AMPA receptors antagonists,
Exp. Opin. Ther. Patents 9 (1999) 557ꢀ570.
[12] Y. Auberson, Competitive AMPA antagonism: a novel mechan-
ism for antiepileptic drugs?, Drugs Future 26 (2001) 463ꢀ471.
/256.
The authors thank the expert technical assistance of
Pascal Vrillaud, the help of Catherine Rochat in setting
up the collaboration, as well as the input of Professors
/
/

802
A.A. Cordi et al. / Il Farmaco 57 (2002) 787ꢀ802
/
[13] M.S. Sheardown, E.O. Nielsen, A.J. Hansen, P. Jacobsen, T.
Honore´, 2,3-Dihydro-6-nitro-7-sulfanioyl-benzo(F)quinoxaline: a
on serum levels, renal excretion and tolerance of intravenous
doses of cefoxitin in humans: comparison with cephalothin,
neuroprotectant for cerebral ischemia, Science 247 (1990) 571ꢀ
/
Antimicrob. Agents Chemother. 6 (1974) 338ꢀ346;
/
574.
(b) G. Aherne, E. Piall, V. Marks, W. Mould, W.F. White,
[14] (a) S. Sakamoto, J. Omori, M. Shimizu-Sasamata, M. Okada, S.
Kawasaki, S. Yatsugi, K. Hidaka, K. Togami, S. Tada, S. Usuda,
K. Murase, Imidazolylquinoxaline-2,3-diones: novel and potent
antagonists of a-amino-3-hydroxy-5-niethylisoxazole-4-propio-
nate (AMPA) excitatory amino acid receptor. XIIth International
Symposium on Medicinal Chemistry, Basel, 1992, P-022A.;
(b) J. Ohmori, S. Sakamoto, H. Kubota, M. Shimizu-Sasamata,
M. Okada, S. Kawasaki, K. Hidaka, K. Togami, T. Furuya, K.
Murase, 6-(1H-Imidazoli-1-yl)-7-nitro-2,3(1H,4H)-quinoxaline-
dione hydrochloride (YM90K) and related compounds:
Prolongation and enhancement of serum methotrexate concentra-
tions by probenecid, Br. Med. J. 1 (1978) 1097ꢀ1099;
/
(c) W.C. Zamboni, P.J. Houghton, R.K. Johnson, J.L. Hulstein,
W.R. Crom, P.J. Cheshire, S.K. Hanna, L.B. Richmond, X. Luo,
C.F. Stewart, Probenecid alters topotecan systemic and renal
disposition by inhibiting renal tubular secretion, J. Pharmacol.
Exp. Ther. 284 (1998) 89ꢀ94.
/
[21] C.F. Bigge, G. Johnson, C.P. Taylor, D.F. Welty, Pharmaceutical
Preparation Containing an Uricosuric and an Excitatory Amino
Acid Antagonist. W09304688.
structureꢀ
NMDA receptor, J. Med. Chem. 37 (1994) 467ꢀ
[15] P. Desos, J.M. Lepagnol, P. Morain, P. Lestage, A.A. Cordi,
Structureꢀactivity relationships in a series of 2(1H)-quinolones
/activity relationships for the AMPA-type non-
[22] C.P. Taylor, M.G. Vartanian, Probenecid pretreatment enhances
anticonvulsant action of NBQX in mice, Eur. J. Pharmacol. 213
/475.
(1992) 151ꢀ
[23] J.C. Carretero, J.L. Garcia-Ruano, M. Vicioso, A practical route
to C-8 substituted fluoroquinolones, Tetrahedron 48 (1992) 7373ꢀ
7382.
[24] K. Ramadas, N. Srinivasan, Ironꢀ
venient and inexpensive reductant, Synth. Commun. 22 (1992)
3189ꢀ3195.
/
153.
/
bearing different acidic function in the 3-position: 6,7-dichloro-
2(1H)-quinolone-3-phosphonic acid, a new potent and selective
AMPA/kainate antagonist with neuroprotective properties, J.
/
/
ammonium chloride*a con-
/
Med. Chem. 39 (1996) 197ꢀ
[16] A.A. Cordi, P. Desos, J.C.R. Randle, J. Lepagnol, Structureꢀ
activity relationships in series of 3-sulfonylamino-2-(1H)-
quinolones as new AMPA/kainate and glycine antagonists,
Bioorg. Med. Chem. 3 (1995) 129ꢀ141.
/206.
/
/
a
[25] W.S. Saari, J.E. Schwering, A convenient synthesis of nitro-
substituted 1,2-benzisothiazol-3(2H)-one 1,1-dioxides (nitrosac-
/
charins), J. Heterocyclic Chem. 23 (1986) 1253ꢀ1255.
/
[17] (a) J.M. Lepagnol, A.M. Roger, A.A. Cordi, P.J. Lestage, Post-
ischemic treatment with S 17625-2, a new AMPA receptor
antagonist, protects the rat brain after permanent focal ischemia,
J. Cereb. Blood Flow Metab. 15 (Suppl. 1) (1995) S373;
[26] S. Darses, T. Jeffery, J.-P. Genet, J.-L. Brayer, J.-P. Demoute,
Cross-coupling of arenediazonium tetrafluoroborates with aryl-
boronic acids catalysed by palladium, Tetrahedron Lett. 37 (1996)
3857ꢀ3860.
/
(b) P. Morain, S 17625, a selective and competitive non-NMDA
antagonist: cerebral anti-ischemic effects in rodents, Br. J.
Pharmacol. 114 (Suppl.) (1995) Abst 330P;
[27] S. Sengupta, S. Bhattacharyya, Palladium-catalyzed cross-cou-
pling of arenediazonium salts with arylboronic acids, J. Org.
Chem. 62 (1997) 3405ꢀ3406.
/
(c) J. Lepagnol, A. Roger, J.-M. Rivet, P. Lestage, Delayed
intravenous treatment with S 17625-2 a competitive AMPA
receptor antagonist, reduces infarction after permanent focal
ischemia in rats, Soc. Neurosci. Abst. 21 (1995) Abst. 145.2.
[18] J.V. Moller, M.I. Sheikh, Renal organic anion transport system:
pharmacological, physiological and biochemical aspects, Pharma-
[28] Y. Takano, 6-Arylquinoline carboxylate derivatives and their
addition salts and their preparation method. JP 2000 169450.
[29] T.A. Verdoorn, R. Dingledine, Excitatory amino acid receptors
expressed in Xenopus oocytes: agonist pharmacology, Mol.
Pharmacol. 34 (1988) 298ꢀ307.
/
[30] R.A. Wallace, D.W. Jared, J.N. Dumont, M.W. Sega, Protein
col. Rev. 34 (1983) 315ꢀ358.
[19] (a) L.M. Fowler, J.R. Foster, E.A. Lock, Effect of ascorbic acid,
acivicin and probenecid on the nephrotoxicity of 4-aminophenol
/
incorporation by isolated amphibian oocytes III. Optimum
incubation conditions, J. Exp. Zool. 184 (1973) 321ꢀ334.
/
[31] A.J. Collins, M. Horlington, A sequential screening test based on
the running component of audiogenic seizures in mice, including
reference compounds PD50 values, Br. J. Pharmacol. 37 (1969)
in the Fisher 344 rat, Arch. Toxicol. 67 (1993) 613ꢀ621;
(b) L.M. Fowler, J.R. Foster, E.A. Lock, Nephrotoxicity of 4-
amino-3-S-gluthathionylphenol and its modulation by metabo-
/
140ꢀ
[32] W. Loscher, B.S. Meldrum, Evaluation of anticonvulsant drugs in
genetic animal models of epilepsy, Fed. Proc. 43 (1984) 276ꢀ284.
[33] J. Litchfield, F. Wilcoxon, A simplified method of evaluating
dose-effect experiments, J. Pharmacol. Exp. Ther. 96 (1949) 99ꢀ
113.
/
150.
lism or transport inhibitors, Arch. Toxicol. 68 (1994) 15ꢀ23;
(c) L. Trumper, L.A. Monasterolo, M.M. Ellas, Probenecid
protects against in vivo acetaminophen-induced nephrotoxicity
/
/
in male Wistar rats, J. Pharmacol. Exp. Ther. 284 (1998) 606ꢀ610.
/
/
[20] (a) C. Stewart Goodwin, E.B. Raftery, A.D. Goldberg, H. Skeggs,
A.E. Till, C.M. Martin, Effects of rate of infusion and probenecid