Structure - Activity relationships in a series of 2(1H)-Quinolones bearing different acidic function in the 3-Position: 6,7-dichloro-2(1H)-oxoquinoline-3-phosphonic acid, a new potent and selective AMPA/kainate antagonist with neuroprotective properties

Article abstract of DOI:10.1021/jm950323j

Recently, we reported the synthesis of 3-(sulfonylamino)-2(1H)-quinolones, a new series of α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA)/kainate and N-methyl-D-aspartic acid (NMDA)/glycine antagonists. By exploring the structure - activity r

Full text of DOI:10.1021/jm950323j

J . Med. Chem. 1996, 39, 197-206
197
Str u ctu r e-Activity Rela tion sh ip s in a Ser ies of 2(1H)-Qu in olon es Bea r in g
Differ en t Acid ic F u n ction in th e 3-P osition :
6,7-Dich lor o-2(1H)-oxoqu in olin e-3-p h osp h on ic Acid , a New P oten t a n d
Selective AMP A/Ka in a te An ta gon ist w ith Neu r op r otective P r op er ties
Patrice Desos, J ean M. Lepagnol, Philippe Morain, Pierre Lestage, and Alex A. Cordi*
Institut de Recherches Servier, 11, rue des Moulineaux, 92150 Suresnes, France
Received May 3, 1995X
Recently, we reported the synthesis of 3-(sulfonylamino)-2(1H)-quinolones, a new series of
R-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA)/kainate and N-methyl-D-aspartic
acid (NMDA)/glycine antagonists. By exploring the structure-activity relationships (SAR) in
this series, we were able to identify the 6,7-dinitro derivative 6 as a potent and balanced
antagonist at both receptors. Unfortunately, compound 6 was devoid of in vivo activity in
mice anticonvulsant testing. To overcome this critical limitation, new compounds bearing
various acidic moieties at the 3-position of the quinolone skeleton were synthesized and
evaluated. The SAR of these new analogues indicated that not all acidic groups are acceptable
at the 3-position: A rank order of potency going from carboxylic ≈ phosphonic > tetrazole >
mercaptoacetic > hydroxamic . other heterocyclic acids was defined. In addition, the selectivity
between the AMPA/kainate and NMDA/glycine sites is dependent on the nature of the
substitution (nitro > chloro for AMPA selectivity), its position (5,7- > 6,7-pattern for glycine
selectivity), and the distance between the quinolone moiety and the heteroatom bearing the
acidic hydrogen (the longer the distance the more AMPA selective the compound). Among
these new AMPA antagonists, we have identified 6,7-dichloro-2(1H)-oxoquinoline-3-phosphonic
acid (24c) as a water soluble and selective compound endowed with an appealing pharmacologi-
cal profile. Compared with the reference AMPA antagonist NBQX, the phosphonic acid 24c is
much less potent in vitro but almost equipotent in vivo in the audiogenic seizures model after
intraperitoneal administration. Moreover, unlike NBQX, compound 24c is also active after
oral administration. In the gerbil global ischemia model, compound 24c shows a neuroprotective
effect at 10 mg/kg/ip, equivalent to the reference NBQX.
In tr od u ction
The most potent selective AMPA/kainate antagonists
belong to the quinoxalinedione class of compounds, such
as NBQX (1), the first selective AMPA/kainate antago-
nist as reported by Sheardown,9 and YM90K (2), dis-
closed more recently by Yamanouchi10 (Figure 1). The
structurally close isatine oxime NS 22911 (3) is described
as the first compound showing AMPA antagonism after
oral administration. Recently, scientists at Merck
disclosed a series of 3-nitro-3,4-dihydro-2(1H)-quinolo-
nes 4 as potent combined AMPA/kainate and NMDA/
glycine antagonists,12 and a group at Parke-Davis
claimed N-sulfonyl derivatives of 3,4-dihydro-3-oxoqui-
noxalinecarboxylate 5 as antagonists at the NMDA-
associated glycine site and AMPA receptor.13
Recently, we reported14 the synthesis of 3-(sulfony-
lamino)-2(1H)-quinolones, a new series of AMPA/kain-
ate and NMDA/glycine antagonists. In the respective
publication, we speculated that the hydroxyl of one of
the tautomeric forms of the quinoxalinedione could be
mimicked by an acidic [(trifluoromethyl)sulfonyl]amino
function at the 3-position of a quinolone. By exploring
the structure-activity relationships (SAR) in this series,
we were able to identify 6,7-dinitro derivative 6 (S
16678) as a potent and balanced antagonist at the two
receptor types. However, the compound was devoid of
any in vivo activity in a convulsion model in mice.
L-Glutamic acid and L-aspartic acid are the major
excitatory amino acid (EAA) neurotransmitters in the
central nervous system.1 Different receptors for glutam-
ic acid have been characterized;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 more well-
defined receptors are the ionotropic channels activated
by either N-methyl-D-aspartic acid (NMDA) or R-amino-
3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA)
and kainic acid and the metabotropic receptors activated
by trans-1-aminocyclopentane-1,3-carboxylic acid (ACPD).
Since 1982,4 numerous competitive and noncompeti-
tive NMDA antagonists have been discovered, and some
were brought into the clinic to assess their potential as
therapeutic agents in the acute treatment of stroke and
head trauma. Unfortunately, these compounds ap-
peared to be liable to strong psychotomimetic side effects
which may impede their development.5,6
Therefore, there has been considerable, recent interest
in selective antagonists of the AMPA receptor because
of their relevance as therapeutic agents for stroke and
other ischemic conditions where excessive EAA release
has excitotoxic action on neurons.7 The pharmacology
of AMPA/kainate antagonists and their role in cerebral
ischemia has been reviewed.8
At that time, the structures of the known, selective
AMPA antagonists 1-3 (Figure 1) were rather homo-
geneous, so the construction of a meaningful receptor
model or the generation of predictive SAR could not be
X Abstract published in Advance ACS Abstracts, November 1, 1995.
0022-2623/96/1839-0197$12.00/0 © 1996 American Chemical Society

198 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Desos et al.
Sch em e 2a
a
Reagents: (a) 3 N NaOH; (b) N,N′-carbonyldiimidazole (CDI),
NH2OH HCl, THF/DMF; (c) NH3, MeOH/DMF; (d) NH2NH2,
MeOH; (e) CDI, Et3N, THF/DMF; (f) CDI, 5-aminotetrazole, THF;
(g) CDI, magnesium enolate monoethyl ester malonic acid, THF/
DMF; (h) NH2NH2, MeOH.
F igu r e 1.
10a was purified by crystallization in DMF and trans-
formed into the tetrazole 11a through the action of tri-
n-butyltin azide in refluxing toluene. The dinitro
tetrazole 11b was obtained through nitration of 11a at
50 °C in a mixture of HNO3/H2SO4.
Sch em e 1a
Additional acidic compounds were prepared from
ester 8a and acid 9a (Scheme 2). For example, the ester
was substituted with ammonia and hydrazine to give
the amide 13 and the hydrazide 14, respectively.
Condensation of 14 with N,N′-carbonyldiimidazole (CDI)
afforded the hydroxyoxadiazole 15 in reasonable yield.
In addition, the acid 9a when activated by CDI in
refluxing THF reacted with a variety of N and C
nucleophiles such as hydroxylamine, 5-aminotetrazole,
and the magnesium enolate of malonic acid monoethyl
ester,15 giving hydroxamate 12, amidotetrazole 18, and
â-keto acid 16, respectively. The latter compound
afforded the pyrazole 17 by condensation with hydra-
zine.
Another key intermediate was generated from the
acids 9a -e. An unfruitful attempt of aromatic bromi-
nation of 9a by bromine in pyridine led to the recovery
of the 3-bromo derivative 19a in essentially quantitative
yield (Scheme 3). As shown in Scheme 4, the mecha-
nism of this reaction could proceed through the addition
of bromonium to the 3,4-double bond followed by depro-
tonation and decarboxylation-aromatization. Alterna-
tively, as suggested by one referee, a possibility could
be the bromination of the 3,4-double bond followed by
a decarboxylation due to the presence of the â-carbonyl
group. The double bond is then reformed by a base-
catalyzed dehydrobromination which would proceed to
give the 3-bromo product because of the greater acidity
of the proton at the 3-position.
To our knowledge, there is no literature precedent of
such halodecarboxylation reactions on a quinolone
nucleus. Moreover, there are only a few literature
reports on the synthesis of 3-bromo-2(1H)-quinolones.
Earlier approaches started from the little accessible
3-unsubstituted-2(1H)-quinolone and proceeded by bro-
mination of the heterocycle leading to a mixture of 6-
and 3-bromo derivatives in moderate yields.16-18 When
the bromination is carried out on the quinoline N-oxide,
prior to its rearrangement to 2(1H)-quinolone, again
nonselective 6- and 3-bromination is observed,19 and
moreover, the subsequent transformation into the 2(1H)-
a
Reagents: (a) MeONa, MeOH; (b) 1 N HCl; (c) n-Bu3SnN3,
toluene; (d) HNO3, H2SO4; (e) 5 N HCl or 3 N NaOH.
considered. Meanwhile the critical limitation of lack of
in vivo activity needed to be overcome, and it was
suspected that the physicochemical properties (solubility
and lipophilicity) of the compounds were more to blame
for this failure rather than their intrinsic activity at the
level of the AMPA receptor. Therefore the hypothesis
stated above was explored through the synthesis and
biological evaluation of compounds bearing various
acidic groups at the 3-position of the quinolone skeleton
aiming at the discovery of compounds endowed with
improved physicochemical properties. The SAR of these
new analogs of 6 are described below.
Syn th esis
The key intermediate carboxylic esters 8a -e were
readily prepared by the condensation of dimethyl ma-
lonate with variously substituted 2-aminobenzaldehydes
7a -e (Scheme 1). The reaction proceeds smoothly at
room temperature in methanol and in the presence of 2
equiv of sodium methoxide followed by acidic workup.
Acid or base hydrolysis of 8a -e gave carboxylic acids
9a -e. Dinitro compound 8b was obtained by nitration
of 8a at 80 °C in a mixture of HNO3/H2SO4.
Similarly, the condensation of 2-aminobenzaldehyde
7a with methyl cyanoacetate provided the nitrile 10a
plus a small quantity (16% by H NMR) of the methyl
(E)-2-amino-4-nitro-R-cyanocinamic ester isomeric in-
termediate 10b which is unable to cyclize. The nitrile
1

SAR of Quinolones with Different Acidic Functions
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 199
Sch em e 3a
Sch em e 5a
a
Reagents: (a) 1 N NaOH; (b) (i) PPh3, diisopropyl azodicar-
boxylate (DIAD), cyclopentanemethanol, THF, (ii) TMSBr, (iii) 3
N HCl; (c) 3 N HCl; (d) (i) oxalyl chloride, DMF (catalytic), CH2Cl2,
(ii) MeMgCl, THF; (e) (i) TMSBr, (ii) 3 N HCl.
a
Reagents: (a) Br2, pyridine; (b) ethyl 2-mercaptoacetate,
K2CO3, acetone; (c) LiOH, THF/H2O; (d) POCl3; (e) MeONa, MeOH;
(f) B(OEt)3, BuLi, THF; (g) HCl; (h) HPO(OEt)2, Pd[PPh3]4, Et3N,
THF; (i) TMSBr, CH3CN; (j) 3 N HCl, (k) HNO3, H2SO4.
thus providing the monoesters 25c,d . The lactams
could be specifically deprotected by acid treatment
affording 27. In addition, the protected monoester 25c
reacted with cyclopentanemethanol under Mistunobu
conditions22 leading to the monocyclopentane methyl
ester 26 after hydrolysis. Reaction of 25d with oxalyl
chloride in dichloromethane with DMF catalysis23 gave
the arylphosphonyl chloride intermediate which reacted
with methylmagnesium chloride in THF leading to the
corresponding arylmethylphosphinic acid 29.
Sch em e 4
Resu lts a n d Discu ssion
The electrophysiological data on kainate- and NMDA/
glycine-induced currents of 3-substituted-2(1H)-quino-
lones are presented in Tables 1 and 2. Our previous
investigations on quinoxalinedione derivatives24 had
shown that for Xenopus oocytes injected with rat
cerebral cortex mRNA, the antagonism of AMPA and
kainate responses is fairly parallel, indicating that both
agonists activate the same non-NMDA type receptor in
this preparation. Therefore, we used kainate-induced
current in oocytes and membrane [3H]AMPA binding
as an initial evaluation of the AMPA/kainate antagonist
potency of our compounds.
In the tables, the compounds have been organized so
as to keep the substitution pattern constant (nitro vs
chloro) while varying the 3-acidic functions. Initially
considering those nine compounds bearing only a 7-nitro
substituent, a rank order of activity regarding the
kainate challenge could be deduced wherein 9a ≈ 24a
> 11a > 20 > 12 . 17, 18, 15, 13. The equiactivity of
9a and 24a indicates that when attached to the 2(1H)-
quinolone skeleton, carboxylic and phosphonic acid
functions are bioisosteric with regards to the activity
at the AMPA/kainate receptor and differ by 1 order of
magnitude with regards to their efficacy at the level of
the NMDA/glycine receptor, the phosphonate 24a being
more AMPA/kainate selective than the carboxylate 9a .
Tetrazole 11a , hydroxamic acid 12, and mercaptoacetic
acid 20 appear to be marginally less active, while the
other acidic heterocycles and the nonacidic amide 13 are
devoid of any activity. The relative potency of the
mercaptoacetic acid 20 as an antagonist at the AMPA/
kainate site is interesting because in this structure the
quinolone, mediated by various acylation agents, results
with many side products.20
In our laboratory, we took advantage of this simple
and unexpected accessibility of 3-bromo-2(1H)-quinolo-
nes to extend our SAR, introducing various acidic groups
at the 3-position. Initially we used the bromine atom
in position 3 as leaving group, allowing the introduction
of a carboxylic side chain by direct nucleophilic substi-
tution with the anion of ethyl 2-mercaptoacetate, pro-
viding 20 after saponification. Later, we used this
functionality as a nucleophile via its metalation with
BuLi or Pd(0). Initial attempts to generate in situ the
lithio intermediate by addition of BuLi to a solution of
21d , followed by addition of triethyl borate, failed to give
the boronate 22 but led exclusively to the 3-dehaloge-
nated 2(1H)-quinolone, thus indicating the instability
of the organometallic species which must be quenched
by proton abstraction from the solvent (THF). On the
other hand, when BuLi was added to a THF solution of
21d concomitant with triethyl borate, the boronate 22
was obtained with moderate yield after acid workup.
In contrast, palladium(0)-catalyzed coupling reaction
of the 3-bromo derivatives with diethyl phosphite in the
presence of triethylamine21 (Heck conditions) allowed
smooth preparation of the corresponding phosphonates
23c,d . The latter were hydrolyzed in high yields by
successive treatment with trimethylsilyl bromide and
hydrochloric acid.
The phosphonate esters 23c,d were subjected to
selective, basic hydrolysis of one ethyl ester (Scheme 5),

200 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Desos et al.
Ta ble 1. Influence of the Acidic Isosters in the 3-Position on in
Vitro Pharmacological Data of the Nitro-Substituted
2(1H)-Quinolonesa
Ta ble 2. Influence of the Acidic Isosters in the 3-Position on in
Vitro Pharmacological Data of the Halogeno-Substituted
2(1H)-Quinolonesa
Xenopus oocytes
IC50 (µM)
selectivity
NMDA/ NMDA/Gly-
compd R5 R6 R7
X
kainate
Gly
kainate
9c
24c
26
H
H
H
Cl Cl CO2H
Cl Cl PO3H2
Cl Cl
10
2.1
15
0.81
83.5
74
0.08
39.8
5
27
24e
9d
22
24d Cl
H
H
Cl
Cl
Cl Cl PO3HEt
15
65
8
40
111
0.65
2,6
1.71
0.08
F
F
PO3H2
H
H
H
H
Cl CO2H
Cl B(OH)2
Cl PO3H2
Cl PO2MeH
>100
>300
4
30
51
7.5
<0.5
29
Cl
>100
a
In the Xenopus oocyte assays, IC50 values were derived from
inhibition curves constructed with data from four to eight experi-
ments at each drug concentration. In all cases, enough experi-
ments (4 e n e 7) were performed to calculate the corresponding
standard deviation which never exceeded 20% of the mean values.
could indicate that the steric hindrance around position
3 of the 2(1H)-quinolone is less well defined in the
AMPA/kainate receptor than at the glycine site of the
NMDA receptor. Moreover, the replacement of AMPA-
favoring nitro substituents25 by glycine-favoring chloro
substituents26 improves the IC50 of the carboxylic acid
derivatives for the glycine site, while it has no effect,
or the opposite effect, on the phosphonic acid-bearing
compounds. This selectivity trend could be interpreted
as reflecting the existence, at the level of the AMPA
receptor, of a positively charged environment close to
the positions 6 and 7 of the quinolone which tolerate
sterically crowded polar groups such as NO2, CN, or
1-imidazolyl.
a
In the Xenopus oocyte assays, IC50 values were derived from
inhibition curves constructed with data from four to eight experi-
ments at each drug concentration. In all cases, enough experi-
ments (4 e n e 7) were performed to calculate the corresponding
standard deviation which never exceeded 20% of the mean values.
Another way to consider the selectivity results is by
analyzing the distance between the quinolone moiety
and the heteroatom bearing the acidic hydrogen. In the
6,7-dinitro-substituted entities, the compound 6 (3-
sulfonylamino) which has the shorter distance is not
selective at all, while 24b (3-phosphonic acid) which has
the longest distance is the more AMPA/kainate selective
compound. Thus, the increasing order of selectivity 6
< 9b < 11b < 24b could be correlated with the distance
increase between carbon 3 of the quinolone and the
heteroatom bearing the negative charge in the conju-
gated base. These distances measured on Dreiding
models gave the order: 15 < 24 < 25 < 25.5 nm,
respectively. This trend is confirmed in the 7-mononi-
tro-substituted series with the mercaptoacetic acid 20
as well as with the hydroxamic acid 12, where the
proton attached to the oxygen is considered the ionizable
one.27
Attempts to expand these SARs to include other acidic
groups such as the boronic (22) and phosphinic (29)
acids led to inactive compounds. Phosphonic acid mo-
noesters are expected to be better mimics of carboxylic
acid functions as both are monobasic acids, while the
phosphonic acids are dibasic acids. Hence, the synthesis
of the monoesters of 24c was undertaken, but the
resulting products 26 and 27, where the overall lipo-
acidic function is substantially removed from the qui-
nolone, indicating that the indispensable acidic function
in that part of the molecule is interacting with as
hypothetical cationic center on this receptor, which can
not be very demanding in terms of distances.
An additional nitro group in the 6-position, leading
to compounds 24b > 9b > 11b, confirms the validity of
the phosphonic and carboxylic acid replacement hypoth-
esis, and the efficacy of the three most potent com-
pounds is increased as expected25 from 4- to 10-fold.
When the same pattern of 6,7-disubstitution is ap-
plied with chloro substituents in place of nitro groups
(Table 2), the same rank order 24c > 9c is obtained but
the activity is lowered by more than 1 order of magni-
tude. Fluoro substitution (analogue 24e) decreased the
potency by 2 orders of magnitude.
In addition, the isomeric 5,7-dichloro pattern (Table
2) increases the affinity for the NMDA/glycine site while
keeping the affinity for the AMPA/kainate receptor (see
compounds 9d and 24d ) more or less constant. By
considering the ratio of IC50 between the kainate and
NMDA/glycine assays, selectivity of the compounds can
be defined (Tables 1 and 2). The phosphonic acid
derivatives 24a -d are more AMPA/kainate selective
than their carboxylic acid counterparts 9a -d . This

SAR of Quinolones with Different Acidic Functions
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 201
Compared with the reference AMPA antagonist NBQX,
the phosphonic acid 24c is much less potent in vitro but
nearly equipotent in vivo in the audiogenic seizures
model after intraperitoneal administration. Moreover,
unlike NBQX, compound 24c is also active when
administered orally. Surprisingly, its isomer 24d ,
resulting only from the shift of the 6-chloro substitution
to the 5-position, is devoid of any in vivo activity. There
is no explanation for this discrepancy, and one can only
speculate that this may be due to either an unexpected
better bioavailability of 24c compared to 24d or the
above-mentioned difference in selectivity of AMPA
versus glycine manifested by these two compounds.
In the gerbil global ischemia model, compound 24c
shows a neuroprotective effect at 10 mg/kg ip equivalent
to the reference NBQX, and at 30 mg/kg ip a complete
hippocampal neuronal protective effect was observed.
In conclusion, among these new AMPA antagonists,
we have identified 24c (S 17625) as a selective and
water soluble compound which is endowed with an
impressive pharmacological profile. More in vivo ex-
periments on ischemia models are now in progress to
validate the use of S 17625 as a potential therapeutic
agent in the acute treatment of stroke. Moreover, at
neuroprotective doses, the available AMPA/kainate
receptor antagonists induce a number of side effects
(depression of central glucose utilization, potentiation
of anesthesia, ataxia, and/or nephrotoxicity) that may
limit their clinical usefulness. Hence the side effect
profile of 24c will have to be clearly established so as
to fully define its therapeutic potential, safety margin,
and risk/benefit ratio.
F igu r e 2. Tentative fitting of compound 24c into the receptor
models proposed in ref 12 for the AMPA (A) and NMDA/glycine
(B) receptors, respectively.
philicity of the molecule was tentatively increased, were
only marginaly active.
In the course of this work, a literature report was
published28 which described a close analogue of 24c,
7-chloro-2(1H)-oxoquinoline-4-phosphonic acid, where
the phosphonic acid function has been shifted from the
3- to the 4-position with a complete loss of activity. In
the same publication, the homologue 5,7-dichloro-2(1H)-
oxoquinoline-4-methylenephosphonic acid is described
as a mild but selective AMPA/kainate antagonist.
Overall, this result is difficult to reconciliate with the
model of the AMPA receptor published recently,12 where
bulk intolerance at the 4-position is claimed to endow
glycine selectivity to 4-substituted compounds. In
contrast, 24c would fit perfectly well in this proposed
model although it will fit equally well in the proposed
model of the NMDA/glycine receptor (Figure 2), there-
fore lacking the discriminative differences needed to
explain the selectivity of the compound.
In addition to preliminary biological evaluations, our
experience gained in the course of the chemical synthe-
sis indicated that the phosphonic acid derivatives were
surprisingly endowed with the better physicochemical
properties. Indeed, in contrast to the very insoluble
carboxylic acid derivatives 9a -d and tetrazole deriva-
tives 11a ,b, which could only be purified by recrystal-
lization in DMF, the phosphonic acid derivatives 24a -d
were easily purified by recrystallization in ethanol or
water. Moreover, the in vitro evaluation was dramati-
cally facilitated through the ease of dissolving these
phosphonate derivatives in aqueous solutions (disodium
salt: 13 mg/mL), in contrast to the situations prevailing
with the carboxylic acids (monosodium salt: <1 mg/mL)
and, above all, with the tetrazoles, where sophisticated
dissolution protocols employing cosolvents were neces-
sary. These qualities are also in contrast with the
solubility properties of the reference compound NBQX.
Indeed, the in vivo evaluations and even the clinical
experiments have been impeded by the poor water
solubility and the bright yellow color which prevents
blinding of the studies.29
Exp er im en ta l Section
Biology. Xen opu s Oocyte Exp er im en ts. Poly(A+) mes-
senger RNA isolated from rat cerebral cortex was injected in
Xenopus oocytes,30 and the oocytes were incubated for 2-3
days at 18 °C to allow expression. They were then stored at
6-8 °C until use (typically 1-2 weeks). Kainate-induced
currents were recorded in “OR2” medium31 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 amplifier. Glass microelectrodes (1-3 MΩ)
were filled with 3 M KCl. The holding potential was adjusted
to -60 mV.
For the initial evaluation of the inhibitory potency of the
compounds, the agonists were applied for 30 s at the following
concentrations: kainate (100 µM) and glycine (3 µM)/NMDA
(30 µM). The compounds were bath applied for 45 s before
and 30 s after the application of agonists. Agonist responses
were evaluated at the peak of the inward current. The IC50
values were calculated from inhibition curves using the
relationship:
nH
I ) Imax/(1 + [antagonist]/IC50
)
(where nH represents the Hill slope coefficient) by a nonlinear
curve-fitting program.
Bin d in g Exp er im en ts. Mem br a n e P r ep a r a tion . Rat
cerebral cortices were homogenized in ice cold 0.32 M sucrose
buffer (AMPA binding) or 50 mM Tris-acetate (glycine binding)
and centrifuged at 1000g for 10 min. The resulting superna-
tant was centrifuged at 37000g for 25 min or at 40000g for 15
min, respectively. For AMPA binding the pellet was then
washed in 30 mM Tris-HCl containing 2.5 mM CaCl2, centri-
fuged as described above, lysed in distilled water, and centri-
fuged again. The resulting pellet was then incubated with
The results obtained in the oocyte electrophysiology
assay with 24c,d were corroborated by binding experi-
ments (Table 3) which confirms and expanded the
difference in selectivity between the two compounds. An
AMPA selective antagonist was at hand with 24c and
a balanced AMPA/kainate-NMDA/glycine antagonist
with 24d .

202 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Desos et al.
Ta ble 3. Comparison of in Vitro and in Vivo Pharmacological Data of Selected 2(1H)-Quinolones and NBQXa
transient global ischemia
in gerbil protection (%)
rat cortical membrane
audiogenic seizures in
binding, IC50 (µM)
DBA/2 mice, ED50 (mg/kg)
10 mg/kg ip
(n ) 10)
30 mg/kg ip
(n ) 6)
compd
R
X
[3H]AMPA
[3H]Gly
ip
po
6
9b
24d
24c
NBQX
6,7-di-NO2
6,7-di-NO2
5,7-di-Cl
NHSO2CF3
CO2H
PO3H2
0.09
0.15
3.5
0.9
0.06
1.38
1.8
5.1
>50
>50
>50
19 (13-26)
8 (4-17)
nt
nt
nt
nt
nt
nt
40
40
nt
nt
nt
100
50
6,7-di-Cl
PO3H2
111
>500
45 (21-97)
na
a
In the binding assays, the IC50 values were determined from inhibition curves fitting five to six points, each point representing the
mean value of duplicates. Enough experiments (n g 5) were conducted to calculate the corresponding standard deviations which were in
the (8% range of the mean values. In the DBA/2 mice audiogenic seizure assay, the ED50s were determined from dose-effect curves
fitting four points (10 mice/dose). Values in parentheses are 95% confidence intervals. In the gerbil transient forebrain ischemia study,
the protection index was determined through comparison of the mean injury score of treated animals (n ) 10) with the score, measured
in the same experiment, of sham-operated gerbils (n ) 5, score ) 0) and untreated ischemic gerbils (n ) 10, score ) 4), respectively. nt
) not tested. na ) not active.
Triton TX-100 for 30 min at 25 °C, centrifuged, and rinsed
twice with Tris buffer and then stored at -80 °C. For glycine
binding, the pellet was washed three times with Tris-acetate
buffer, suspended in 0.32 M sucrose buffer, and frozen at -80
°C until further use.
Ch em istr y. Reagents were commercially available and of
synthetic 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 emulsion. 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 calculated values within the (0.4% range.
Melting points were obtained on a Reichert hot stage micro-
scope and are uncorrected. Silica gel 60, Merck 230-400
mesh, was used for both flash and medium pressure chroma-
tography. TLC were performed on precoated 5 × 10 cm, Merck
silica gel 60 F254 plates (layer thickness 0.25 mm).
7-Nitr o-2(1H)-oxoqu in olin e-3-ca r boxylic Acid Meth yl
Ester (8a ). To a solution of 4-nitro-2-aminobenzaldehyde (7a )
(500 mg, 3.01 mmol) and dimethyl malonate (0.51 mL, 4.5
mmol) in methanol (50 mL) was added a solution of 5.25 M
sodium methoxide in methanol (2.6 mL, 11.37 mmol). The
mixture was stirred at room temperature overnight. The
yellow precipitate was filtered, washed with methanol, and
stirred in 1 N HCl (75 mL). The white solid was filtered and
washed with ether to yield 8a (390 mg, 49%): mp >260 °C;
1H NMR (DMSO-d6) δ 12.4-12.5 (bs, 1H), 8.65 (s, 1H), 8.1 (d,
1H), 8.05 (d, 1H), 8.0 (dd, 1H), 3.85 (s, 3H). Anal. (C11H8N2O5)
C, H, N.
Bin d in g Assa y.32 The thawed membrane suspension (200
µL) was incubated in the presence of 25 µL of the ligand [3H]-
AMPA (4 °C, 30 min, 20 nM) or [3H]glycine (4 °C, 60 min, 50
nM) and 25 µL of a known solution of the investigated
compound. Nonspecific binding was defined in the presence
of 100 µM quisqualic acid or 100 µM glycine. The specific
binding was assayed over the range 0.1-300 nM in the absence
(control) or presence of the drug to be tested. The membranes
were collected by filtration using a Skatron cell harvester. Data
were analyzed using Lundon software.
Au d iogen ic Seizu r es in DBA/2 Mice. Audiogenic sei-
zures were performed on 22-25 day old DBA/2 mice according
to literature procedures.33,34 Compounds were administered,
to groups of 10 mice/dose, intraperitoneally (ip) 30 min before
or orally (po) 1 h before an auditory stimulation (14 kHz, 100
dB, 60 s) generated by an amplifier (Brueel and Kjaer 2706)
delivered through loud speakers in a sound chamber. Mice
were placed singly in the sound chamber and observed before,
during, and after the auditory simulation. The criterion for a
positive response was the occurrence of a running response
within 20 s of the auditory stimulation. Doses which protected
50% of animals from clonic seizure (ED50 value with 95%
confidence limits) were calculated by the method of Lichtfield
and Wilcoxon.35
Tr a n sien t F or ebr a in Isch em ia in Ger bils. Transient
complete forebrain ischemia was produced in adult Mongolian
gerbils (60-80 g) anesthetized with 2% isoflurane according
to the method of Kirino.36 Cerebral ischemia was induced by
bilateral occlusion of the carotid arteries for 5 min. Before
and during ischemia, body temperature was monitored and
maintained at an average of 37 °C using a rectal probe and
thermostatically controlled heating blanket. After ischemia,
temperature control was accomplished with the aid of a
heating lamp and kept at around 37 °C until thermal homeo-
stasis was restored. After 4 days, the animals were killed and
the brain was removed and fixed. Paraffined coronal sections
(7 µm) were cut at the hippocampal level and stained with
luxol fast blue. The CA1 subfield was assessed for neuronal
death on a scale ranging from 0 for a normal hippocampus to
4 for an almost total loss of neuronal cells (>90%). This
evaluation was carried out by two independent technicians
who were blind to treatment given to the specific sample
(sham, control, or drug treated). Compounds were adminis-
tered ip at doses of 10 and 30 mg/kg at 60 min before and 30
and 90 min after the onset of occlusion. In each experiment,
controls (n ) 10) received the compound vehicle only following
the same time schedule.
6,7-Din itr o-2(1H)-oxoqu in olin e-3-car boxylic Acid Meth -
yl Ester (8b). To a solution of the above compound (330 mg,
1.33 mmol) in 96% H2SO4 (0.5 mL) was added 87% HNO3 (0.5
mL) dropwise, and the reaction mixture was stirred at 80 °C
for 4 h. The mixture was cooled to 0 °C, ice added, and the
precipitate filtered, washed with water, and dried under
vacuum to yield 8b as a yellow solid (300 mg, 77%): mp >260
°C; 1H NMR (DMSO-d6) δ 12.9 (bs, 1H), 8.9 (s, 1H), 8.7 (s,
1H), 7.8 (s, 1H), 3.85 (s, 3H).
6,7-Din itr o-2(1H)-oxoqu in olin e-3-ca r boxylic Acid (9b).
A suspension of ester 8b (200 mg, 0.68 mmol) was refluxed in
5 N HCl (20 mL) for 4 h. The reaction was cooled to room
temperature, and the solid was filtered, washed with water,
and crystallized from DMF to yield 9b as a yellow solid (90
1
mg, 47%): mp 298 °C; H NMR (DMSO-d6) δ 13 (bs, 2H), 9.0
(s, 1H), 8.9 (s, 1H), 7.9 (s, 1H). Anal. (C10H5N3O7) C, H, N.
7-Nitr o-2(1H)-oxoqu in olin e-3-ca r boxylic Acid (9a ). A
suspension of ester 8a (2.0 g, 8.05 mmol) in 5 N HCl (40 mL)
was stirred at reflux for 6 h. The precipitate was filtered,
washed with water, and recrystallized from DMF to yield 9a
1
(1.83 g, 97%): mp >260 °C; H NMR (DMSO-d6) δ 12.5-15.0
(bs, 2H), 9.0 (s, 1H), 8.25 (d, 1H), 8.20 (d, 1H), 8.1 (dd, 1H).
Anal. (C10H6N2O5) C, H, N.
6,7-Dich lor o-2(1H)-oxoqu in olin e-3-car boxylic Acid (9c).
9c was obtained from the condensation of 4,5-dichloro-2-
aminobenzaldehyde14 and dimethyl malonate as described for

SAR of Quinolones with Different Acidic Functions
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 203
8a , followed by hydrolysis as described for 9a , and recrystal-
lized from DMF: 89% overall yield; mp 230 °C; 1H NMR
(DMSO-d6) δ 12.0-16.0 (bs, 2H), 8.9 (s, 1H), 8.4 (s, 1H), 7.6
(s, 1H). Anal. (C10H5Cl2NO3) C, H, N, Cl.
5,7-Dich lor o-2(1H)-oxoqu in olin e-3-car boxylic Acid (9d).
9d was obtained from the condensation of 4,6-dichloro-2-
aminobenzaldehyde14 and dimethyl malonate as described for
8a , followed by hydrolysis as described for 9a , and recrystal-
lized from DMF: 78% overall yield; mp >260 °C; 1H NMR
(DMSO-d6) δ 12.0-15.5 (bs, 2H), 8.8 (s, 1H), 7.7 (d, 1H), 7.45
(d, 1H). Anal. (C10H5Cl2NO3) C, H, N, Cl.
6,7-Diflu or o-2(1H)-oxoqu in olin e-3-car boxylic Acid (9e).
9e was obtained from the condensation of 4,5-difluoro-2-
aminobenzaldehyde37 and dimethyl malonate as described for
8a , followed by hydrolysis as described for 9a : 31% overall
yield; mp >300 °C; 1H NMR (DMSO-d6) δ 12.0-16.0 (bs, 2H),
8.95 (s, 1H), 8.2 (dd, 1H), 7.4 (dd, 1H).
washed with MeOH, and recrystallized from DMF to yield 14
as a yellow solid (233 mg, 77%): mp >300 °C; 1H NMR (DMSO-
d6) δ 13.5-12.0 (s, 1H), 10.5 (s, 1H), 8.9 (s, 1H), 8.25 (d, 1H),
8.20 (d, 1H), 8.05 (dd, 1H), 4.5-5.2 (s, 1H). Anal. (C10H8N4O4)
C, H, N.
7-Nit r o-3-(2-h yd r oxy-1,3,4-oxa d ia zol-5-yl)-2(1H )-q u i-
n olon e (15). CDI (260 mg, 1.57 mmol) and triethylamine (290
µL, 2.09 mmol) were added to a suspension of 14 (260 mg, 1.05
mmol) in a mixture of THF (10 mL) and DMF (1 mL). The
reaction mixture was heated to reflux, with stirring, for 20 h.
After cooling to room temperature, 1 N HCl (10 mL) was
added. The yellow precipitate was collected, washed with
MeOH, and crystallized from DMF to yield 15 as a yellow solid
(70 mg, 16%): mp >300 °C; 1H NMR (DMSO-d6) δ 12.6 (s, 1H),
8.65 (s, 1H), 8.10 (m, 2H), 8.0 (d, 1H). Anal. (C11H6N4O5) C,
H, N.
7-Nitr o-2(1H)-oxoqu in olin e-3-ca r boxa m id e (13). A sus-
pension of methyl ester 8a (500 mg, 2.02 mmol) in a mixture
of methanol (30 mL) and DMF (30 mL) was saturated with
ammonia gas and the mixture stirred at 60 °C overnight. The
solvents were evaporated under vacuum, and the residue was
recrystallized in DMF to yield 13 as a yellow solid (373 mg,
79%): mp >300 °C; 1H NMR (DMSO-d6) δ 13.0-12.5 (bs, 1H),
9.0-8.9 (bs, 2H), 8.95 (s, 1H), 8.2 (d, 1H), 8.0 (dd, 1H), 7.95
(d, 1H). Anal. (C10H7N3O4) C, H, N.
3-Cya n o-7-n itr o-2(1H)-qu in olon e (10a ). To a solution of
4-nitro-2-aminobenzaldehyde (7a ) (1.0 g, 6.0 mmol) in metha-
nol (50 mL) were added methyl cyanoacetate (0.634 mL, 7.2
mmol) and a solution of 5.25 M sodium methoxide in methanol
(1.64 mL, 8.61 mmol). The mixture was stirred at room
temperature overnight. The yellow precipitate was collected
and washed with methanol before being stirred in 1 N HCl
(75 mL). The solid was filtered and recrystallized from DMF
1
to yield 10a (750 mg, 58%): mp >260 °C; H NMR (DMSO-
3-(1-Car boxy-2-oxo-2-eth yl)-7-n itr o-2(1H)-qu in olon e (16).
CDI (2.1 g, 12.95 mmol) was added to a suspension of acid 9a
(2.0 g, 8.54 mmol) in a mixture of THF (200 mL) and DMF
(20 mL) and the mixture heated to reflux, with stirring, for 1
h. The reaction mixture was cooled to room temperature, and
the magnesium enolate of monoethyl ester malonic acid15 (1.98
g, 12.8 mmol) was added. The reaction mixture was heated
to reflux for 1.5 h and then cooled to room temperature. HCl
(1 N) was added and the precipitate collected to yield 16 (1.75
d6) δ 12.0-14 (bs, 1H), 8.95 (s, 1H), 7.9-8.15 (m, 3H). Anal.
(C10H5N3O3) C, H, N.
7-Nitr o-3(1H-tetr a zol-5-yl)-2(1H)-qu in olon e (11a ). Tri-
n-butyltin azide (1.53 g, 4.64 mmol), 3-cyano-7-nitro-2(1H)-
quinolone (10a ) (500 mg, 2.32 mmol), and toluene (50 mL) were
combined and refluxed for 12 h. The solvent was removed in
vacuo, and the residue was stirred in 1 N NaOH for 30 min.
The medium was acidified with 4 N HCl. The solid which
precipitated was collected and after crystallization in DMF
gave 180 mg (30%) of 11a : mp >260 °C; 1H NMR (DMSO-d6)
δ 11.5-14 (bs, 1H), 9.1 (s, 1H), 8.25 (d, 1H), 8.2 (bs, 1H). Anal.
(C10H6N6O3) C, H, N.
1
g, 67%): mp >320 °C; H NMR (DMSO-d6) δ (keto form) 8.7
(s, 1H), 8.3-7.9 (m, 3H), 4.15 (s, 2H), 4.1 (q, 2H), 1.15 (t, 3H),
(enol form) 12.4 (s, 1H), 9.0 (s, 1H), 8.3-7.9 (m, 3H), 6.8 (s,
1H), 4.25 (q, 2H), 1.3 (t, 3H). Anal. (C11H7N7O4) C, H, N.
3-(5-H yd r oxy-1H -p yr a zol-3-yl)-7-n it r o-2(1H )-q u in o-
lon e (17). Hydrazine monohydrate (98%, 30 µL, 0.64 mmol)
was added to a suspension of 16 (100 mg, 0.32 mmol) in
ethanol (10 mL) and the reaction mixture heated to reflux,
with stirring, for 12 h. The reaction mixture was cooled to
room temperature; the precipitate was collected and washed
with ethanol to yield 17 (58 mg, 67%): mp >300 °C; 1H NMR
(DMSO-d6) δ 12.5 (s, 1H), 12.0 (s, 1H), 9.8 (s, 1H), 8.5 (s, 1H),
8.2 (s, 1H), 8.05 (d, 1H), 7.95 (d, 1H), 6.3 (s, 1H). Anal.
(C12H8N4O4) C, H, N.
6,7-Din it r o-3-(N-(1H )-t et r a zol-5-yl)-2(1H )-q u in olon e
(11b). The above compound (1 g, 3.87 mmol) was stirred at
50 °C for 1.5 h in a mixture of 96% H2SO4 (2 mL) and 87%
HNO3 (2 mL). The reaction mixture was cooled to 0 °C, ice
added, and the precipitate collected, washed with water, dried,
and recrystallized from DMF to give 11b as a yellow solid (736
1
mg, 63%): mp >320 °C; H NMR (DMSO-d6) δ 9.15 (s, 1H),
9.0 (s, 1H), 7.95 (s, 1H). Anal. (C10H5N7O5) C, H, N.
7-Nit r o-3-(N-(1H )-t et r a zol-5-ylca r b a m oyl)-2(1H )-q u i-
n olon e (18). To a suspension of acid 9a (600 mg, 2.56 mmol)
in THF (40 mL) was added CDI (456 mg, 2.81 mmol), and the
mixture was heated to reflux for 1 h. The reaction mixture
was cooled to room temperature, and 5-aminotetrazole (264
mg, 2.56 mmol) was added. The mixture was stirred at reflux
for 3 h and then cooled to room temperature and 1 N HCl (20
mL) added. The precipitate was collected, washed with 1 N
HCl, and crystallized from DMF to yield 18 as a yellow solid
3-Br om o-6,7-d ich lor o-2(1H)-qu in olon e (19c). Bromine
(3.6 mL, 69.8 mmol) was added dropwise to a suspension of
acid 9c (9.0 g, 34.9 mmol) in pyridine (90 mL) at 0 °C, and the
mixture was stirred at 100-110 °C for 1.5 h. The dark brown
solution was cooled to room temperature and poured into 1 N
HCl (400 mL). The precipitate was collected, washed with 1
N HCl followed by water, and then dried to yield 19c (8.5 g,
1
1
83%): mp > 260 °C; H NMR (DMSO-d6) δ 8.05 (s, 1H), 7.65
(420 mg, 54%): mp >320 °C; H NMR (DMSO-d6) δ 13.25-
(s, 1H), 7.3 (s, 1H). Anal. (C9H4BrCl2NO) C, H, N. Br.
3-Br om o-5,7-d ich lor o-2(1H)-qu in olon e (19d ). 19d was
prepared from 9d according to the method described for 19c:
98% yield; mp >260 °C; 1H NMR (DMSO-d6) δ 8.5 (s, 1H), 7.5
(s, 1H), 7.35 (s, 1H). Anal. (C9H4BrCl2NO) C, H, N, Br.
3-Br om o-7-n itr o-2(1H)-qu in olin e (19a ). 19a was pre-
pared from 9a according to the method described for 19c and
13.0 (bs, s, 2H), 9.05 (s, 1H), 8.35 (d, 1H), 8.25 (s, 1H), 8.10 (d,
1H). Anal. (C11H7N7O4) C, H, N.
7-Nitr o-2(1H)-qu in olon e-3-h yd r oxa m ic Acid (12). To a
solution of acid 9a (400 mg, 1.71 mmol) in a mixture of THF
(40 mL) and DMF (3 mL) was added CDI (416 mg, 2.56 mmol),
and the mixture was heated to reflux for 1 h. The reaction
mixture was cooled to room temperature and hydroxylamine
hydrochloride (238 mg, 3.42 mmol) added. The reaction mix-
ture was heated to reflux for 3 h and then cooled to room
temperature and 3 N HCl (40 mL) added. The yellow precipi-
tate was collected, washed with 1 N HCl, and crystallized from
DMF to yield 12 as a yellow solid (177 mg, 42%): mp >300
1
recrystallized from DMF: 64% yield; mp >260 °C; H NMR
(DMSO-d6) δ 12.6 (bs, 1H), 8.7 (s, 1H), 8.10 (s, 1H), 8.0 (d,
1H), 7.9 (d, 1H). Anal. (C9H5BrN2O3) C, H, N, Br.
3-Br om o-6,7-d iflu or o-2(1H)-qu in olon e (19e). 19e was
prepared from 9e according to the method described for 19c
and recrystallized from DMF: 87% yield; mp 265-271 °C; 1H
NMR (DMSO-d6) δ 12.5-12.2 (bs, 1H), 8.5 (s, 1H), 7.8 (dd, 1H),
7.25 (dd, 1H). Anal. (C9H4BrF2NO) C, H, N, Br.
3-Br om o-6,7-d ich lor o-2-m eth oxyqu in olin e (21c). A so-
lution of 19c (8.5 g, 29 mmol) in phosphorus oxychloride (100
mL) was stirred at reflux overnight. Most of the excess of
phosphorus oxychloride was removed in vacuo, and the residue
was carefully hydrolyzed at 0 °C by adding crushed ice. The
1
°C; H NMR (DMSO-d6) δ 12.8 (s, 1H), 11.45 (s, 1H), 9.55 (s,
1H), 8.9 (s, 1H), 8.25 (d, s, 2H). Anal. (C11H7N7O4) C, H, N.
7-Nitr o-2(1H)-qu in olon e-3-h yd r a zin eca r boxylic Acid
(14). Hydrazine monohydrate (98%, 300 µL, 6.04 mmol) was
added to a suspension of methyl ester 8a (300 mg, 1.21 mmol)
in methanol (30 mL) and the reaction mixture heated to reflux,
with stirring, for 24 h. The reaction mixture was cooled to
room temperature, and the yellow precipitate was collected,

204 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Desos et al.
1
precipitate was collected, washed several times with water,
and then dried at 80 °C under vacuum to give 7.9 g (88%) of
the chloroquinoline: mp 184 °C; 1H NMR (DMSO-d6) δ 8.9 (s,
1H), 8.35 (s, 1H), 8.25. Anal. (C9H3BrCl3N) C, H, N.
To a suspension of the chloro intermediate (4.0 g, 12.8 mmol)
in methanol (100 mL) was added a solution of 5.25 M sodium
methoxide in methanol (2.8 mL, 14.7 mmol). The mixture was
stirred at 65 °C for 10 h. After cooling, the precipitate was
collected to yield 21c (3.60 g, 92%): mp 176-179 °C; 1H NMR
(DMSO-d6) 8.65 (s, 1H), 8.2 (s, 1H), 8.0 (s, 1H), 4.10 (s, 3H).
Anal. (C10H6BrCl2NO) C, H, N.
3-Br om o-5,7-d ich lor o-2-m eth oxyqu in olin e (21d ). 21d
was prepared from 19d according to the method described for
21c: 77% overall yield; mp 134-138 °C; 1H NMR (DMSO-d6)
δ 8.65 (s, 1H), 7.85 (s, 1H), 7.75 (s, 1H), 4.05 (s, 3H). Anal.
(C10H6BrCl2NO) C, H, N.
24b as a light yellow solid (100 mg, 57%): mp >260 °C; H
NMR (DMSO-d6) δ 9.15 (s, 1H), 9.0 (s, 1H), 7.95 (s, 1H). Anal.
(C10H5N7O5) C, H, N.
6,7-Diflu or o-(1H)-oxoqu in olin e-3-ph osph on ic Acid (24e).
24e was prepared from the 19e according to the method
described for 23c, followed by treatment with bromotrimeth-
ylsilane as for the preparation of 24c, and recrystallized from
ethanol: 18% overall yield; mp 291-294 °C; 1H NMR (DMSO-
d6) δ 8.35 (d, 1H), 8.0 (dd, 1H), 7.25 (dd, 1H). Anal. (C9H6F2-
NO4P) C, H, N.
3-[(Ca r boxym eth yl)th io]-7-n itr o-2(1H)-qu in olon e (20).
A suspension of 19a (800 mg, 2.97 mmol), potassium carbonate
(452 mg, 3.27 mmol), and ethyl 2-mercaptoacetate (0.36 mL,
3.27 mmol) in a mixture of acetone (25 mL) and DMF (2 mL)
was refluxed under nitrogen for 2 h. After cooling to room
temperature, the mixture was diluted with water and the
suspension filtered to give 3-[(carbethoxymethyl)thio]-7-nitro-
2(1H)-quinolone (800 mg, 87%): mp 254-257 °C; 1H NMR
(DMSO-d6) δ 12.5 (bs, 1H), 8.15 (s, 1H), 8.0 (d, 1H), 7.8 (m,
2H), 4.15 (qd, 2H), 4.05 (s, 2H), 1.2 (t, 2H).
To a suspension of this compound (720 mg, 2.4 mmol) in a
mixture of THF (10 mL) and water (10 mL) was added lithium
hydroxide monohydrate (300 mg, 7.2 mmol), and the mixture
was refluxed for 4 h. After cooling to room temperature, the
reaction mixture was acidified with 1 N HCl and the solid
collected, washed several times with water, and then dried at
80 °C under vacuum to give 20 (585 mg, 83%): mp 290-294
°C; 1H NMR (DMSO-d6) δ 13.05 (bs, 1H), 12.45 (bs, 1H), 8.15
(s, 1H), 8.0 (d, 1H), 7.8 (d, s, 2H), 3.95 (s, 2H). Anal.
(C11H8N2O5S) C, H, N, S.
5,7-Dich lor o-2(1H)-oxoqu in olin e-3-bor on ic Acid (22).
Triethyl borate (2.24 mL, 13.2 mmol) followed by 1.6 M
N-butyllithium in hexane (4.95 mL, 7.9mmol) was added
dropwise at -70 °C via syringe to a solution of 21d (2.0 g, 6.6
mmol) in THF (40 mL). The solution was stirred for 45 min
at -70 °C and allowed to warm to room temperature over 3 h;
1 N HCl (50 mL) was added and the solution extracted with
ethyl acetate (150 mL). The organic extract was washed with
brine, dried over magnesium sulfate, filtered, and concen-
trated. The residue was taken up in methylene chloride (30
mL) and the precipitated product was collected to give 5,7-
dichloro-2-methoxyquinoline-3-boronic acid (650 mg, 36%): 1H
NMR (DMSO-d6) δ 8.2-8.4 (bs, 2H), 8.5 (s, 1H), 7.8 (d, 2H),
7.7 (d, 2H), 4.0 (s, 3H).
3-Br om o-7-n itr o-2-m eth oxyqu in olin e (21a ). 21a was
prepared from 19a according to the method described for
21c: 69% overall yield; mp 163-170 °C; 1H NMR (DMSO-d6)
δ 8.9 (s, 1H), 8.55 (s, 1H), 8.25 (d, 1H), 8.10 (d, 1H), 4.10 (s,
3H). Anal. (C10H7BrN2O3) C, H, N.
6,7-Dich lor o-2-m et h oxyqu in olin e-3-p h osp h on ic Acid
Dieth yl Ester (23c). Tetrakis(triphenylphosphine)palladium
(1.85 g, 1.6 mmol) and 21c (10.0 g, 32.1 mmol) were added
under nitrogen atmosphere to a stirred mixture of diethyl
phosphite (8.30 mL, 64.2 mmol) and triethylamine (8.95 mL,
64.2 mmol) in THF (15 mL). The resultant mixture was
stirred at reflux for 12 h. After cooling to room temperature,
the mixture was taken up in ethyl acetate (450 mL) and
washed with 1 N HCl and brine. After evaporation in vacuo,
the residue was purified by chromatography on silica gel
(cyclohexane/ethyl acetate, 50/50, and then ethyl acetate) to
give the diethyl phosphonate 23c (7.3 g, 62%): mp 119-120
°C; 1H NMR (DMSO-d6) δ 8.75 (d, 1H), 8.5 (s, 1H), 8.1 (s, 1H),
4.15 (m, 4H), 4.1 (s, 3H), 1.3 (m, 6H). Anal. (C14H16Cl2NO4P)
C, H, N, Cl.
5,7-Dich lor o-2-m et h oxyq u in olin e-3-p h osp h on ic Acid
Dieth yl Ester (23d ). 23d was prepared from 21d according
to the method described for 23c: 49% yield; mp 52-54 °C; 1H
NMR (DMSO-d6) δ 8.85 (d, 1H), 7.7 (d, 1H), 7.4 (d, 1H), 4.15
(m, 4H), 4.1 (s, 3H), 1.3 (t, 6H). Anal. (C14H16Cl2NO4P) C, H,
N, Cl.
7-Nitr o-2-m eth oxyqu in olin e-3-p h osp h on ic Acid Dieth -
yl Ester (23a ). 23a was prepared from 21a according to the
method described for 23c: 39% yield; 83-85 °C; 1H NMR
(DMSO-d6) δ 8.75 (d, 1H), 9.0 (d, 1H), 8.6 (d, 1H), 8.35 (dd,
1H), 8.3 (dd, 1H), 4.3 (q, s, 7H), 1.3 (m, 6H). Anal.
(C14H17N2O6P) C, H, N.
This compound (300 mg, 1.10 mmol) was refluxed in 6 N
HCl for 4 h, the suspension filtered, and the solid recrystallized
from DMF to give 22 (248 mg, 88%): mp >300 °C; H NMR
(DMSO-d6) δ 12.4 (bs, 1H), 8.7-8.9 (bs, 2H), 8.55 (s, 1H), 7.5
(d, 1H), 7.35 (d, 1H). Anal. (C9H6BCl2NO3) C, H, N, Cl.
6,7-Dich lor o-2-m et h oxyq u in olin e-3-p h osp h on ic Acid
Mon oeth yl Ester (25c). A suspension of diethyl ester 23c
(200 mg, 0.55 mmol) was stirred at 100 °C in 3 N NaOH (2
mL) for 3 h. The reaction mixture was acidified with 1 N HCl
and extracted with ethyl acetate to give an oil which crystal-
1
6,7-Dich lor o-2(1H )-oxoq u in olin e-3-p h osp h on ic Acid
(24c). Bromotrimethylsilane (40 mL) was added to a suspen-
sion of phosphonate diester 23c (17.1g, 47 mmol) in acetonitrile
(300 mL) and the mixture stirred at 80 °C for 1.5 h. The
reaction mixture was evaporated to dryness under vacuum and
the resulting residue stirred at 80 °C in 3 N HCl (100 mL) for
3 h. The precipitate was collected and recrystallized from
ethanol to give 24c (11.5 g, 83%): mp >260 °C; 1H NMR
(DMSO-d6) δ 11.5-12.5 (m, 3H), 8.35 (d, 1H), 8.2 (s, 1H), 7.5
(s, 1H). Anal. (C9H6Cl2NO4P) C, H, N, Cl.
1
lized (180 mg, 97%): mp 166-169 °C; H NMR (DMSO-d6) δ
8.7 (d, 1H), 8.45 (s, 1H), 8.05 (s, 1H), 4.05 (s, 3H), 4.0 (m, 2H),
1.25 (t, 3H). Anal. (C12H12Cl2NO4P) C, H, N, Cl.
6,7-Dich lor o-2(1H)-oxoqu in olin e-3-ph osph on ic Acid Mo-
n oeth yl Ester (27c). The above compound (100 mg, 0.29
mmol) was stirred at 80 °C for 1 h in 3 N HCl (1 mL), and the
suspension was collected by filtration to give pure 27 (88 mg,
92%): mp 268-276 °C; H NMR (DMSO-d6) δ 11.5-12.5 (m,
1H) 8.45 (d, 1H), 8.25 (s, 1H), 7.5 (s, 1H), 4.05 (q, 2H), 1.2 (t,
3H). Anal. (C11H10Cl2NO4P) C, H, N, Cl.
5,7-Dich lor o-2-m et h oxyq u in olin e-3-p h osp h on ic Acid
Mon oeth yl Ester (25d ). 25d was prepared from 23d ac-
cording to the method described for 25c: 92% yield; mp 153-
157 °C; 1H NMR (DMSO-d6) δ 8.75 (d, 1H), 7.9 (d, 1H), 7.7 (d,
1H), 4.05 (s, 3H), 4.0 (q, 2H), 1.25 (t, 3H).
5,7-Dich lor o-2(1H )-oxoq u in olin e-3-p h osp h on ic Acid
(24d ). 24d was prepared from 23d according to the method
described for 24c and recrystallized from ethanol: 62% yield;
mp >260 °C; 1H NMR (DMSO-d6) δ 11.5-13.0 (m, 3H), 8.4 (d,
1H), 7.5 (s, 1H), 7.3 (s, 1H). Anal. (C9H6Cl2NO4P) C, H, N,
Cl.
7-Nitr o-2(1H)-oxoqu in olin e-3-p h osp h on ic Acid (24a ).
24a was prepared from 23a according to the method described
for 24c and recrystallized from water: 59% yield; mp >260
°C; 1H NMR (DMSO-d6) δ 11.5-13.0 (m, 3H), 8.5 (d, 1H),
8.1 (d, 1H), 8.1 (d, 1H), 7.95 (dd, 1H). Anal. (C9H7N2O6P) C,
H, N.
1
6,7-Din it r o-2(1H )-oxoq u in olin e-3-p h osp h on ic
Acid
5,7-Dich lor o-2(1H)-oxoqu in olin e-3-ph osph on ic Acid Mo-
n oeth yl Ester (27d ). 27d was prepared from 25d according
to the method described for 27c: 89% yield; mp 225-230 °C;
1H NMR (DMSO-d6) δ 13.0-12.0 (bs, 1H), 8.5 (d, 1H), 7.55 (d,
1H), 7.35 (d, 1H), 4.05 (q, 2H), 1.2 (t, 3H). Anal. (C11H10Cl2-
NO4P) C, H, N, Cl.
(24b). The above compound (150 mg, 0.55 mmol) was stirred
at 80 °C for 2.5 h in a mixture of 96% H2SO4 (0.3 mL) and
87% HNO3 (0.3 mL). The reaction mixture was cooled to 0 °C
and ice added. The precipitate was stirred for 30 min, filtered,
washed with water, and recrystallized from ethanol to yield

SAR of Quinolones with Different Acidic Functions
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 205
(10) Sakamoto, S.; Ohmori, J .; Shimizu-Sasamata, M.; Okada, M.;
Kawasaki, S.; Yatsugi, S.; Hidaka, K.; Togami, J .; Tada, S.;
Usuda, S.; Murase, K. Imidazolylquinoxaline-2,3-diones: Novel
and Potent Antagonists of R-Amino-3-hydroxy-5-methylisox-
azole-4-propionate (AMPA) Excitatory Amino Acid Receptor.
Presented at the XIIth International Symposium on Medicinal
Chemistry, Basel, 1992; P-022A. Ohmori, J .; Sakamoto, S.;
Kubota, H.; Shimizu-Sasamata, M.; Okada, M.; Kawasaki, S.;
Hidaka, K.; Togami, J .; Furuya, T.; Murase, K. 6-(1H)-Imidazol-
1-yl)-7-nitro-2,3(1H,4H)-quinoxalinedione Hydrochloride (YM90K)
and Related Compounds: Structure-Activity Relationships for
the AMPA-Type non-NMDA Receptor. J . Med. Chem. 1994, 37,
467-475.
5,7-Dich lor o-2-m eth oxyqu in olin e-3-m eth ylp h osp h in -
ic Acid Eth yl Ester (28). A solution of oxalyl chloride (0.155
mL, 1.77 mmol) in anhydrous methylene chloride was added
dropwise to the monoester 25d (500 mg, 1.49 mmol) dissolved
in a mixture of anhydrous methylene chloride (3 mL) and DMF
(20 µL). Upon complete addition of the oxalyl chloride, the
solution was heated to reflux for 1 h. The solvent was removed
in vacuo and the residue taken up in anhydrous THF (25 mL).
Methylmagnesium chloride (3 M solution in THF, 0.58 mL,
1.74 mmol) was added dropwise, and the reaction mixture was
stirred at room temperature for 1.5 h; 1 N HCl (1 mL) was
added dropwise and the reaction mixture diluted with ethyl
acetate (150 mL) and washed with 1 N HCl and 1 N NaOH.
After evaporation, the residue was purified by chromatography
on silica gel (methylene chloride/methanol, 98/2) to give pure
28 (240 mg, 53%): 1H NMR (CDCl3) δ 9.1 (d, 1H), 7.8 (d, 1H),
7.5 (d, 1H), 4.15 (s, 3H), 4.1 (m, 1H), 3.85 (m, 1H), 1.85 (d,
3H), 1.3 (t, 3H).
(11) Wa¨tjen, F.; Nielsen, E. O.; Drejer, J .; J ensen, L. H. Isatin Oximes
-
A Novel Series of Bioavailable Non-NMDA Antagonists.
Bioorg. Med. Chem. Lett. 1993, 3, 105-106.
(12) Carling, R. W.; Leeson, P. D.; Moore, K. W.; Smith, J . D.; Moyes,
C. R.; Mawer, I. M.; Thomas, S.; Chan, T.; Baker, R.; Foster, A.
C.; Grimwood, S.; Kemp, J . A.; Marshall, G. R.; Tricklebank, M.
D.; Saywell, K. L. 3-Nitro-3,4-dihydro-2(1H)-quinolones. Exci-
tatory Amino Acid Antagonists Acting at Glycine-site NMDA and
(RS)-R-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic Acid Re-
ceptors. J . Med. Chem. 1993, 36, 3397-3408.
5,7-Dich lor o-2(1H )-oxoq u in olin e-3-m et h ylp h osp h in -
ic Acid (29). The above compound was hydrolyzed according
to the method described from 24c: mp >300 °C; 1H NMR
(DMSO-d6) δ 12.0-12.5 (bs, 1H), 8.55 (d, 1H), 7.5 (d, 1H), 7.35
(d, 1H), 1.65 (d, 3H). Anal. (C10H8Cl2NO3P) C, H, N, Cl.
(13) Hays, S. J .; Boxer, P. A.; Taylor, C. P.; Vartarian, M. G.;
Robichaud, L. J .; Nielsen, E. O. N-Sulfonyl Derivatives of 6,7-
Dichloro-3,4-dihydro-3-oxo-quinoxalinecarboxylate as Glycine-
site NMDA and AMPA Antagonists. Bioorg. Med. Chem. Lett.
1993, 3, 77-80.
6,7-Dich lor o-2(1H )-oxoq u in olin e-3-p h osp h on ic Acid
Mon ocyclop en tylm eth yl Ester (26). Cyclopentanemetha-
nol (49 µL, 0.42 mmol) and triphenylphosphine (109 mg, 0.42
mmol) were added to a solution of monoester 25c (100 mg,
0.28 mmol) in anhydrous THF (2 mL). The solution was
stirred for 10 min at room temperature, and diisopropyl
azodicarboxylate was added. Upon completion of the conden-
sation reaction, bromotrimethylsilane (92 µL, 0.7 mmol) was
added and stirring continued for an additional 1 h. The
reaction mixture was evaporated to dryness, 3 N HCl (4
mL) was added, and the mixture was stirred at 90 °C for 5 h.
After cooling to room temperature, ethyl acetate (25 mL) was
added. The biphasic system was stirred vigorously, and the
white solid which precipitated was collected to give 26 (89
(14) Cordi, A. A.; Desos, P.; Randle, J . C. R.; Lepagnol, J . Structure-
Activity Relationships in a Series of 3-Sulfonylamino-2-(1H)-
Quinolones as New AMPA/kainate and Glycine Antagonists.
Bioorg. Med. Chem. 1995, 3, 129-141.
(15) Bram, G.; Vilkas, M. Nouvelle Synthe`se de â-ce´toesters du type
RCOCH2CO2Et a` partir du malonate acide d’e´thyle. (New
Synthesis, from Monoethylmalonate, of â-Keto Esters of the
RCOCH2CO2Et Type.) Bull. Soc. Chim. 1964, 954-961.
(16) Kauffmann, T.; Schulz, J . Hetarynes VI. Intermediary Appear-
ance of 3,4-Dehydrocarbostyril and 1-Methyl-3,4-dehydrocar-
bostyril. Chem. Ber. 1966, 99, 1837-1842.
(17) Linda, P.; Gianlorenzo, M. Uncatalyzed Halogenation of Car-
bostyril in Acetic Acid Solution. Ric. Sci., Rend. Ses. 1964, 7,
309-314; Chem. Abstr. 1965, 63:5602d.
(18) Robert, D. A.; Campell, S. F. Quinolone Inotropic Agents. EP
148623 A2, J uly 17, 1985.
1
mg, 85%): mp 286-288 °C; H NMR (DMSO-d6) δ 12.5-11.5
(bs, 1H), 8.4 (d, 1H), 8.25 (s, 1H), 7.5 (s, 1H), 3.8 (t, 2H), 2.15
(m, 1H), 1.75-1.15 (m, 8H). Anal. (C15H16NO4Cl2P) C, H, N,
Cl.
(19) Claret, P. A. Quinolines. In Comprehensive Organic Chemistry;
Barton, D., Ollis, W. D., Sammes, P. G., Eds.; Pergamon Press:
Oxford, 1979; Vol. 4, pp 155-203.
(20) Miura, Y.; Takaku, S.; Nawata, Y.; Hamana, M. Reactions of
3-Substituted Quinoline 1-Oxides with Acylating Agents. Het-
erocycles 1991, 32, 1579-1586.
Ack n ow led gm en t. The authors thank the expert
technical assistance of Elisabeth Ruano, Christine Bo-
bichon, and Pascal Vrillaud as well as Angela D. Morris
for reviewing the manuscript.
(21) Hirao, T.; Masunaga, T.; Ohshiro, Y.; Agawa, T. A. Novel
Synthesis of Dialkyl Arenephosphonates. Synthesis 1981, 56-
57.
(22) Campbell, D. A. The Synthesis of Phosphonate Esters, an
Extension of the Mitsunobu Reaction. J . Org. Chem. 1992, 57,
6331-6335.
(23) Rogers, R. S. A General Synthesis of Phosphonic Acid Dichlorides
Using Oxalyl Chloride and DMF Catalysis. Tetrahedron Lett.
1992, 33, 7473-7474.
(24) Randle, J . C. R.; Guet, T.; Bobichon, C.; Moreau, C.; Curutchet,
P.; Lambolez, B.; Prado De Carvalho, L.; Cordi, A.; Lepagnol, J .
M. Quinoxaline Derivatives: Structure-Activity Relationships
and Physiological Implications of Inhibition of N-Methyl-D-
aspartate and Non-N-methyl-D-aspartate Receptor-Mediated
Currents and Synaptic Potentials. Mol. Pharmacol. 1992, 41,
337-345.
(25) Randle, J . C. R.; Guet, T. L.; Cordi, A.; Lepagnol, J . M.
Competitive Inhibition by NBQX of Kainate/AMPA Receptor
Currents and Excitatory Synaptic Potentials: Importance of
6-Nitro Substitution. Eur. J . Pharmacol. 1992, 215, 237-
244.
(26) Leeson, P. D.; Baker, R.; Carling, R. W.; Curtis, N. R.; Moore,
K. W.; Williams, B. J .; Foster, A. C.; Donald, A. E.; Kemp, J . A.;
Marshall, G. R. Kynurenic Acid Derivatives. Structure-Activity
Relationshps for Excitatory Amino Acid Antagonism and Iden-
tification of Potent and Selective Antagonists at the Glycine Site
on the N-Methyl-D-aspartate Receptor. J . Med. Chem. 1991,
34, 1243-1252.
Refer en ces
(1) Krogsgaard-Larsen, P. Amino Acid Receptors. In Comprehensive
Medicinal Chemistry; Hansh, C., Ed.; Pergamon Press: Oxford,
1990; Vol. III, pp 521-537.
(2) Hollmann, M.; O’Shea-Greenfield, A.; Rogers, S. W.; Heinemann,
S. Cloning by Functional Expression of a Member of the
Glutamate Receptor Family. Nature 1989, 342, 643-648.
(3) Moriyoshi, K.; Masu, M.; Ishii, T.; Shigemoto, R.; Mizuno, N.;
Nakanishi, S. Molecular Cloning and Characterization of the Rat
NMDA receptor. Nature 1991, 354, 31-37.
(4) Croucher, M. J .; Collins, J . F.; Meldrum, B. S. Anticonvulsant
Action of Excitatory Amino Acid Antagonists. Science 1982, 216,
899-901.
(5) Herting, P. L. Competitive NMDA Antagonists: Results from
Animal Models and First Results in Humans. J . Neurochem.
1993, 61 (Suppl.), S283C.
(6) Clark, W. M.; Coull, B. M. Randomized Trial of CGS 19775, a
Glutamate Antagonist, in Acute Ischemic Stoke Treatment.
Neurology 1994, 44 (Suppl. 2), A270.
(7) Sheardown, M. J .; Suzdak, P. D.; Nordholm, L. AMPA but not
NMDA, Receptor Antagonism is Neuroprotective in Gerbil
Global Ischaemia, even when Delayed 24 h. Eur. J . Pharmacol.
1993, 236, 347.
(8) Gill, R. The Pharmacology of R-Amino-3-hydroxy-5-methyl-4-
isoxazole propionate (AMPA)/Kainate Antagonists and their Role
in Cerebral Ischaemia. Cerebrovasc. Brain Metab. Rev. 1994,
6, 225-256.
(9) Sheardown, M. S.; Nielsen, E. O.; Hansen, A. J .; J acobsen, P.;
Honore´, T. 2,3-Dihydro-6-nitro-7-sulfamoyl-benzo(F)quinoxa-
line: a Neuroprotectant for Cerebral Ischemia. Science 1990,
247, 571-574.
(27) Challis, B. C.; Challis, J . A. Amides and Related Compounds.
In Comprehensive Organic Chemistry, Barton, D., Ollis, W. D.,
Eds.; Pergamon Press: Oxford, 1979; Vol. 2, pp 1036-1045.
(28) Hamilton, G. S.; Bednar, D.; Borosky, S. A.; Huang, Z.; Zubrows-
ki, R.; Ferkany, J . W.; Karbon, E. W. Synthesis and Glutamate
Antagonist Activity of 4-Phosphonoalkylquinoline Derivatives:
a Novel Class of Non-NMDA Antagonist. Bioorg. Med. Chem.
Lett. 1994, 4, 2035-2040.
(29) Muir, K. W.; Lees, K. R. Clinical Experience with Excitatory
Amino Acid Antagonist Drugs. Stroke 1995, 26, 503-513.

206 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Desos et al.
(30) Verdoorn, T. A.; Dingledine, R. Excitatory Amino Acid Receptors
Expressed in Xenopus Oocytes: Agonist Pharmacology. Mol.
Pharmacol. 1988, 34, 298-307.
Including Reference Compounds PD50 Values. Br. J . Pharma-
col. 1969, 37, 140-150.
(34) Loscher, W.; Meldrum, B. S. Evaluation of Anticonvulsant Drugs
in Genetic Animal Models of Epilepsy. Fed. Proc. 1984, 43,
276-284.
(35) Lichtfield, J .; Wilcoxon, F. A Simplified Method of Evaluating
Dose-effect Experiments. J . Pharmacol. Exp. Ther. 1949, 96,
99-113.
(36) Kirino, T. Delayed Neuronal Death in the Gerbil Hippocampus
Following Ischemia. Brain Res. 1982, 239, 57-69.
(37) Nanuet, P. B.; Tappan, Y.-I. L. Process for Preparing 2-Chloro-
4,5-Difluorobenzoic Acid. U.S. Patent 4,833,270, 1989.
(31) Wallace, R. A.; J ared, D. W.; Dumont, J . N.; Sega, M. W. Protein
Incorporation by Isolated Amphibian Oocytes III. Optimum
Incubation Conditions. J . Exp. Zool. 1973, 184, 321-334.
(32) Hood, W. F.; Gray, N. M.; Dappen, M. S.; Watson, G. B.;
Compton, R. P.; Cordi, A. A.; Lanthorn, T. H.; Monahan, J . B.
Characterization of Indole-2-carboxylate Derivatives as Antago-
nists of N-Methyl-D-aspartate Receptor Activity at the Gly-
cine Recognition Site. J . Pharmacol. Exp. Ther. 1992, 262, 654-
660.
(33) Collins, A. J .; Horlington, M. A Sequential Screening Test Based
on the Running Component of Audiogenic Seizures in Mice,
J M950323J