Structure-activity relationships of 4-hydroxy-3-nitroquinolin-2(1H)- ones as novel antagonists at the glycine site of N-methyl-D-aspartate receptors

Article abstract of DOI:10.1021/jm960520y

A series of 4-hydroxy-3-nitroquinolin-2(1H)-ones (HNQs) was synthesized by nitration of the corresponding 2,4-quinolinediols. The HNQs were evaluated as antagonists at the glycine site of NMDA receptors by inhibition of [³H]DCKA binding to rat brain membranes. Selected HNQs were also tested for functional antagonism by electrophysiological assays in Xenopus oocytes expressing either 1a/2C subunits of NMDA receptors or rat brain AMPA receptors. The structure-activity relationships (SAR) of HNQs showed that substitutions in the 5-, 6-, and 7-positions in general increase potency while substitutions in the 8-position cause a sharp reduction in potency. Among the HNQs tested, 5,6,7-trichloro HNQ (8i) was the most potent antagonist with an IC₅₀ of 220 nM in [³H]DCKA binding assay and a K(b) of 79 nM from electrophysiological assays. Measured under steady-state conditions HNQ 8i is 240-fold selective for NMDA over AMPA receptors. The SAR of HNQs was compared with those of 1,4-dihydroquinoxaline-2,3-diones (QXs) and 1,2,3,4-tetrahydroquinoline-2,3,4-trione 3-oximes (QTOs). In general, HNQs have similar potencies to QXs with the same benzene ring substitution pattern but are about 10 times less active than the corresponding QTOs. HNQs are more selective for NMDA receptors than the corresponding QXs and QTOs. The similarity of the SAR of HNQs, QXs, and QTOs suggested that these three classes of antagonists might bind to the glycine site in a similar manner. With appropriate substitutions, HNQs represent a new class of potent and highly selective NMDA receptor glycine site antagonists.

Full text of DOI:10.1021/jm960520y

4682
J . Med. Chem. 1996, 39, 4682-4686
Str u ctu r e-Activity Rela tion sh ip s of 4-Hyd r oxy-3-n itr oqu in olin -2(1H)-on es a s
Novel An ta gon ists a t th e Glycin e Site of N-Meth yl-D-a sp a r ta te Recep tor s
Sui Xiong Cai,*,† Zhang-Lin Zhou,‡ J in-Cheng Huang,† Edward R. Whittemore,† Zizi O. Egbuwoku,†
J on E. Hawkinson,† Richard M. Woodward,† Eckard Weber,† and J ohn F. W. Keana‡
CoCensys, Inc., 213 Technology Drive, Irvine, California 92618, and Department of Chemistry, University of Oregon,
Eugene, Oregon 97403
Received J uly 18, 1996X
A series of 4-hydroxy-3-nitroquinolin-2(1H)-ones (HNQs) was synthesized by nitration of the
corresponding 2,4-quinolinediols. The HNQs were evaluated as antagonists at the glycine site
of NMDA receptors by inhibition of [3H]DCKA binding to rat brain membranes. Selected HNQs
were also tested for functional antagonism by electrophysiological assays in Xenopus oocytes
expressing either 1a/2C subunits of NMDA receptors or rat brain AMPA receptors. The
structure-activity relationships (SAR) of HNQs showed that substitutions in the 5-, 6-, and
7-positions in general increase potency while substitutions in the 8-position cause a sharp
reduction in potency. Among the HNQs tested, 5,6,7-trichloro HNQ (8i) was the most potent
antagonist with an IC50 of 220 nM in [3H]DCKA binding assay and a Kb of 79 nM from
electrophysiological assays. Measured under steady-state conditions HNQ 8i is 240-fold
selective for NMDA over AMPA receptors. The SAR of HNQs was compared with those of
1,4-dihydroquinoxaline-2,3-diones (QXs) and 1,2,3,4-tetrahydroquinoline-2,3,4-trione 3-oximes
(QTOs). In general, HNQs have similar potencies to QXs with the same benzene ring
substitution pattern but are about 10 times less active than the corresponding QTOs. HNQs
are more selective for NMDA receptors than the corresponding QXs and QTOs. The similarity
of the SAR of HNQs, QXs, and QTOs suggested that these three classes of antagonists might
bind to the glycine site in a similar manner. With appropriate substitutions, HNQs represent
a new class of potent and highly selective NMDA receptor glycine site antagonists.
Ch a r t 1
In tr od u ction
Excessive stimulation by excitatory amino acids such
as glutamate is known to cause degeneration and death
of neurons, a process termed excitotoxicity.1 It is now
generally accepted that N-methyl-D-aspartate (NMDA)
receptors, a subclass of ionotropic glutamate receptors,
play a major role in the excitotoxicity of glutamate.2 The
discovery that glycine is a necessary coagonist for
NMDA receptor activation3 has prompted a search for
glycine antagonists4 as potential central nervous system
(CNS) therapeutic agents with anticonvulsant and
neuroprotective properties.5 Examples include 4-hy-
droxy-3-phenylquinolin-2(1H)-ones such as 1,6 4-hy-
droxy-3-(3-phenoxyphenyl)quinolin-2(1H)-ones such as
2,7 and 3-acyl-4-hydroxyquinolin-2(1H)-ones such as 38
which are selective glycine antagonists. In addition, 3,4-
dihydro-3-nitroquinolin-2(1H)-ones such as 49 (Chart 1)-
are reported to have essentially equal potency as
antagonists at NMDA receptor glycine sites and at
R-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
(AMPA) receptors, a subtype of the non-NMDA class of
ionotropic glutamate receptors.
functional groups. These changes led to the discovery
of 1,2,3,4-tetrahydroquinoline-2,3,4-trione 3-oximes
(QTOs) such as 6, which is a highly potent glycine site
antagonist.11 Herein we report that 4-hydroxy-3-nitro-
quinolin-2(1H)-ones (HNQs) 8 constitute a novel class
of potent and selective NMDA receptor glycine site
antagonists.
We have recently reported on the structure-activity
relationship (SAR) of substituted 1,4-dihydroquinoxa-
line-2,3-diones (QXs) leading to ACEA 1021 (5), a highly
potent and systemically active glycine site antagonist.10
In search of other heterocyclic glycine site antagonists,
we modified the QX heterocycle ring by preserving the
NH and CO in the 1- and 2-positions while replacing
the CO and NH in the 3- and 4-positions with other
Ch em istr y
The HNQs 8 were synthesized by nitration (HNO3/
CH3CO2H) of the corresponding 2,4-quinolinediols 7.12
6,7-Dichloro HNQ (8g) was obtained from nitration of
a mixture of 6,7- and 5,6-dichloro-2,4-quinolinediols.11
† CoCensys, Inc.
‡ University of Oregon.
X Abstract published in Advance ACS Abstracts, October 15, 1996.
S0022-2623(96)00520-1 CCC: $12.00 © 1996 American Chemical Society

Notes
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23 4683
Sch em e 1a
Ta ble 1. Structure-Activity Relationship of HNQs 8 vs QXs 9
and QTOs 10
[3H]DCKA IC50 (µM)
a
(a) 1, EtO2CCH2COCl or (EtO2C)2CH2, 2, NaOH, 3, PPA; (b)
1, MeOH, 2, (CH3CO)2O, 3, KN(SiMe3)2; (c) HNO3/CH3CO2H; (d)
KNO3/CF3CO2H.
no.
R5
R6
R7
R8
8a
9b
10b
a
b
c
d
e
f
g
h
i
H
H
H
Cl
H
Cl
H
Me
Cl
H
H
H
Cl
H
H
H
Cl
H
Cl
F
H
Cl
H
H
H
Cl
Cl
Me
Cl
Cl
F
H
H
H
H
Cl
H
H
H
H
F
34 ( 3
9.8
1.8
1.4
0.053
3.4
1.5 ( 0.1
>100
Nitration of 5,6,7- and 6,7,8-trichloro-2,4-quinolinediols
(7i,l) was carried out in CF3CO2H instead of CH3CO2H
owing to the poor solubility of the starting materials in
CH3CO2H. 2,4-Quinolinediols 7b-l were prepared from
the corresponding anilines or 2-aminobenzoic acids as
reported previously (Scheme 1).11
d
61 ( 2
>100
NDc
ND
0.28
0.13
2.1
0.030
0.63
0.73
d
4.4
>100
0.033
0.012
0.037
0.007
0.91
3.3
0.22 ( 0.05
0.40 ( 0.02
0.29 ( 0.03
0.22 ( 0.04
40 ( 5
j
k
l
F
H
F
Cl
F
Cl
>100
>100
P h a r m a cology
Cl
ND
The affinity of HNQs for the NMDA receptor glycine
site was measured by inhibition of [3H]-5,7-dichlo-
rokynurenic acid ([3H]DCKA) binding to rat brain
cortical membranes.11 In addition, apparent antagonist
dissociation constants (Kb) of selected HNQs were
determined by electrophysiological techniques using
either cloned 1a/2C rat NMDA receptors expressed in
Xenopus oocytes or AMPA receptors expressed in oocytes
using whole rat brain poly(A)+ RNA. Kb values were
estimated by assuming competitive inhibition, as indi-
cated by the binding assays, and measuring suppression
of membrane current responses elicited by fixed con-
centrations of agonist: 1 µM glycine and 100 µM
glutamate for NMDA receptors; 10 µM AMPA for AMPA
receptors. The apparent Kb value of HNQ 8i at the
glycine site was also calculated from the rightward shift
of the glycine concentration-response curves induced
by fixed concentrations of antagonist.
a
Affinity is expressed as IC50 estimates for inhibition of
[3H]DCKA binding to rat brain membranes. Values are means (
b
SEMs of at least three independent experiments. Data from ref
11. c ND, not determined. Due to the symmetrical structure of
d
QXs, 9c,i are identical with 9b,l, respectively.
Remarkably, 7-chloro-6,8-difluoro HNQ (8j) has very
low affinity. 5,6,7,8-Tetrafluoro HNQ (8k ) and 6,7,8-
trichloro HNQ (8l) are inactive. The comparison of 8i
to 8l is most striking. By moving the chlorine atom in
the 5-position (8i) to the 8-position (8l), a 220 nM
antagonist was converted into an inactive ligand, i.e.,
>400-fold reduction in potency. We interpret this as
an indication that the NH in the 1-position of the HNQ
is critical for binding of the ligand to the receptor,
possibly by hydrogen-bonding, and substitution in the
8-position interferes with this critical interaction.
A comparison of the SAR of HNQs 8 with the SAR of
QXs 9 shows that for the chloro-substituted compounds
b, f, and g, HNQs and the corresponding QXs have
similar affinity. 5,6,7-Trichloro HNQ (8i) is an excep-
tion, being about 7 times less active than the corre-
sponding QX 9i.
We note that 5,7-dimethyl HNQ (8h ) is about 7-fold
more potent than the corresponding 5,7-dimethyl QX
(9h ). This is mainly due to the lower potency of 5,7-
dimethyl QX (9h ) vs 5,7-dichloro QX (9f). Interestingly,
5,7-dimethyl QTO (10h ) is also much more potent than
the corresponding QX 9h . Apparently electron-with-
drawing groups in the 5- and 7-positions may not be
critical for the potency of HNQs and QTOs, while
electron-withdrawing groups in those positions are
important for the potency of QXs.
It is also noteworthy that 6,7,8-trisubstituted (8j) and
5,6,7,8-tetrasubstituted (8k ) HNQs are much less active
than the corresponding 6,7,8-trisubstituted (9j) and
5,6,7,8-tetrasubstituted (9k ) QXs. Two factors could
contribute to the observed difference in potency. First,
the Cl and F in 9j,k serve as electron-withdrawing
groups, increasing the potency of the QXs. Second, QXs
are symmetrical with respect to the two amide NHs,
both being potentially available for hydrogen-bonding
depending on the orientation of the ligand in the
receptor. In contrast, the F atom in the 8-position of
8j,k might serve to reduce the potency of HNQs, such
Resu lts a n d Discu ssion
The SAR of HNQs as antagonists at the glycine site
of NMDA receptors is shown in Table 1. Unsubstituted
HNQ (8a ) has low micromolar affinity for the glycine
site, which is comparable to that of the unsubstituted
QX (9a ). Among the monochloro-substituted HNQs,
7-chloro HNQ (8b) is the most potent antagonist,
consistent with the SAR of other classes of glycine
antagonists.11,13 The other monochloro-substituted HNQs
either show very low potency (8d ) or are devoid of
activity (8c,e).
Two or more chlorine atoms in the 5-, 6-, and 7-posi-
tions enhanced the affinity of the HNQs 85-150-fold
relative to the unsubstituted HNQ (8a ). 5,6,7-Trichloro
HNQ (8i) is among the most potent HNQs with an IC50
of 220 nM in the [3H]DCKA binding assay. 6,7-Dichloro
HNQ (8g) and 5,7-dichloro HNQ (8f) also have compa-
rable potencies.
It is interesting to note that 5,7-dimethyl HNQ (8h )
is about as potent as 5,7-dichloro HNQ (8f), indicating
that the presence of an electron-withdrawing group is
not critical for high potency in the HNQ series. This is
different from most of the known glycine antagonists
in which the presence of a polar or electron-withdrawing
group in the 7-position is required for high potency.14

4684 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23
Notes
sponding 5,6,7-trichloro QX (9i). This is reasonable
given the similar SAR of HNQs and QXs. The moderate
differences observed between IC50 values from [3H]-
DCKA binding and Kb values from electrophysiology
could be due to several factors including differences in
affinity between cloned 1a/2C NMDA receptors and the
heterogeneous native NMDA receptor population present
in rat brain cortical membranes.
In a second series of electrophysiological experiments,
we assayed effects of 8i on glycine concentration-
response curves in oocytes expressing the NR 1a/2C
receptors. As shown in Figure 2, 1 µM 8i shifted the
affinity for glycine from 170 nM to 3.0 µM, an 18-fold
increase. The Kb for 8i at the NMDA receptor glycine
site was assessed by simultaneous fits of glycine inhibi-
tion curves alone and in the presence of 1 µM 8i.
Analysis of these data gives a Kb value for 8i of 0.058
(0.056-0.059) µM, which is close to that estimated from
the inhibition curve (0.079 µM). Statistical analysis of
the curve shift induced by 1 µM 8i indicates that
F igu r e 1. Glycine site binding model for HNQs in comparison
with QXs and QTOs. A represents a receptor hydrogen bond
acceptor and D-H represent a receptor hydrogen bond donor;
+ represents a positively charged receptor group.
inhibition conforms to the competitive model (F1,38
0.30).
)
as by interfering with the critical hydrogen-bonding of
the NH in the 1-position.
Although 10 µM 8i shifts the glycine curve further to
the right, inhibition at this concentration of antagonist
is not fully surmountable. This suggests that high
concentrations of 8i may induce inhibition that is not
completely due to glycine site antagonism. No further
experiments were performed to investigate this effect.
Due to the insurmountable inhibition at 10 µM 8i,
statistical analysis performed using the control, 1 and
10 µM 8i curves did not conform to the competitive
model (F2,57 ) 21.7).
The three HNQs tested all have low affinity for AMPA
receptors (Table 2). Interestingly, AMPA receptor
potency almost directly parallels glycine site potency,
so there is only a 2-fold difference in selectivity for these
three HNQs. HNQ 8i is the most selective ligand with
a selectivity of 240-fold favoring the NMDA receptor
glycine site. The other two HNQs also have a selectivity
of >100-fold, suggesting that HNQs generally are highly
selective for the glycine site. By comparison, QX 9i
shows a 50-fold selectivity, and only the most potent
compounds in the QX series such as ACEA 1021 (5)
have a selectivity of >100-fold.10 The most selective
compound among the QTOs (10i) has a selectivity of
80-fold.11
The most striking structural difference between HNQs
and the broad spectrum antagonist 4 is the 4-OH group.
The low AMPA activity observed for the HNQs is
consistent with the pharmacophore model proposed by
Carling et al. for AMPA receptors.9 They propose that
AMPA receptors exhibit a bulk intolerance adjacent to
the 4-position, while for the glycine site a hydrogen-
bond-accepting group adjacent to the 4-position is
obligatory for good potency.9 Therefore, the presence
of the 4-OH group in the HNQs appears to favor the
glycine site of NMDA receptors and disfavor the
glutamate site of AMPA receptors, thus accounting for
their high selectivity.
Comparing the SAR of HNQs 8 with that of QTOs 10
reveals that for compounds with the same substituents,
HNQs are generally about 10 times less potent than
QTOs. Interestingly, both 8e and 10e are inactive,
lending further support to the notion that an unhin-
dered NH in the 1-position of the HNQs and QTOs is
important for interaction with the receptor.
The similarity among the SARs of HNQs 8, QXs 9,
and QTOs 10 suggests that these three classes of ligands
may bind to the glycine site of NMDA receptors in a
similar way. We propose the following model for HNQs
binding to the glycine site of NMDA receptors which is
similar to the models proposed for QXs and QTOs11
(Figure 1) and also for 3-acyl-4-hydroxyquinolin-2(1H)-
ones.8 The model consists of a hydrogen bond donor
(NH) in the 1-position, a hydrogen bond acceptor (CdO)
in the 2-position, a charge-charge interaction in the 3-
to 4-positions, a hydrogen bond acceptor in the 4-posi-
tion, and hydrophobic interactions involving the benzene
ring portion of the molecules. Since HNQs are more
acidic than QXs,15 presumably owing to ionization of the
4-OH group, electron-withdrawing groups in the ben-
zene ring are not critical for the potency of HNQs,
though they are very important for the potency of QXs.
Similarly, electron-withdrawing groups in the benzene
ring are not critical for potency of QTOs.11 HNQs in
general are less active than QTOs, possibly because
there is better charge-charge interaction between QTOs
and the receptor than that of HNQs. As seen from the
model, the negative charge in HNQs is distributed over
several atoms, while the negative charge in QTOs is
likely to be more directional and is located mainly on
the N-O oxygen.
Functional antagonism of NMDA and AMPA recep-
tors by selected HNQs is summarized in Table 2. The
glycine Kb values obtained from electrophysiology for
8f,h are about 2- and 3-fold higher (i.e., less potent) than
the corresponding IC50 values obtained from [3H]DCKA
binding, while the Kb value for 8i indicates that this
compound is about 3-fold more potent in the functional
assay. Based on the electrophysiology assays, 5,6,7-
trichloro HNQ (8i) is about 5-fold more potent than 5,7-
dichloro HNQ (8f) and about equipotent to the corre-
Con clu sion s
In conclusion, a series of HNQs has been synthesized
and found to be a novel class of potent and selective
antagonists for the glycine site of NMDA receptors.
Substitution in the benzene portion of HNQ by chloro

Notes
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23 4685
Ta ble 2. Functional Antagonism by HNQs of NR 1a/2C NMDA Receptors and Rat Brain AMPA Receptors Expressed in Xenopus
Oocytes
8
9e
AMPA Kb
(µM)b
selectivity
for NMDAc
glycine
Kb (µM)
AMPA Kb
(µM)
no.
glycine Kb (µM)a
nd
f
h
i
0.41 (0.39-0.44)f
0.91 (0.86-0.95)
0.079 (0.072-0.087)
37 (33-41)
150 (140-170)
19 (17-20)
118
164
240
3, 3
3, 3
3, 3
0.73
9.3
ND
4.2
NDg
0.082
a
Inhibition of NMDA receptors was measured in oocytes expressing the cloned rat receptor subunit combination NR 1a/2C. Kb values
for glycine binding sites were estimated, assuming simple competitive antagonism, from inhibition of currents elicited by 1 µM glycine
b
and 100 µM glutamate. Inhibition of AMPA receptors was measured in oocytes expressing rat cerebral cortex poly(A)+ RNA. Kb values
for glutamate binding sites were estimated from inhibition of currents elicited by 10 µM AMPA. HNQs were initially dissolved in DMSO,
and the final aqueous solution contained 0.1-1% DMSO. c The steady-state selectivity index for inhibition at NMDA receptor glycine
d
sites was estimated by dividing Kb AMPA by Kb glycine. Indicates the number of independent experiments (cells examined); numbers
refer to NMDA and AMPA, respectively. e Data from ref 10 for QXs. f Numbers in parentheses are 95% confidence intervals adjusted to
g
the linear scale. ND, not determined.
residue were added MeOH (50 mL) and 1 N NaOH (20 mL).
The resulting mixture was refluxed for 3 h, cooled to room
temperature, and acidified to pH ) 2 with 2 N HCl. The
mixture was evaporated, and the residue was extracted with
CHCl3 (3 × 30 mL). The CHCl3 solutions were combined,
washed with brine, dried (Na2SO4), and evaporated to give 525
mg (37%) of the acid as a yellowish powder: mp 160-161 °C;
1H NMR (DMSO-d6) 3.50 (s, 2H), 7.64 (d, J ) 9, 1H), 7.87 (d,
J ) 9, 1H), 10.06 (s, 1H).
A mixture of the acid (490 mg, 1.74 mmol) with polyphos-
phoric acid (20 g) was stirred at 170 °C for 3 h and then cooled
to room temperature. To the resulting mixture were added
slowly 150 mL of 1 N HCl and then 20% of aqueous NaOH to
adjust the pH to 4. The precipitate was filtered, washed with
H2O (5 × 5 mL), and dried to give 270 mg of crude product.
F igu r e 2. Inhibition of NR 1a/2C receptors in Xenopus oocytes
by 8i. Inhibition is associated with a rightward shift in the
glycine concentration-response relationship. Curves were first
measured under control conditions and then repeated in two
The crude product was treated with DMSO (20 mL) and
filtered. To the filtrate was added water (100 mL), and the
precipitate was filtered, washed with water (3 × 5 mL), and
dried to give 230 mg (50%) of 7l as a cream powder: mp 325
concentrations of 8i. A fixed concentration of 100 µM glutamate
1
°C dec; H NMR (DMSO-d6) 5.83 (s, 1H), 7.91 (s, 1H), 10.77
was used throughout. The EC50 and slope values for the three
(s, 1H), 11.98 (s, 1H).
curves measured independently were as follows: control, 0.17
µM and 1.4; 1 µM 8i, 3.0 µM and 1.4; 10 µM 8i, 40 µM and
1.0. Amplitudes of maximal responses ranged between 61 and
5,7-Dich lor o-4-h yd r oxy-3-n itr oqu in olin -2(1H)-on e (8f).
To a mixture of 5,7-dichloro-1,4-quinolinediol (7f) (534 mg, 2.32
mmol) in glacial acetic acid (3 mL) was added HNO3 (69-71%,
0.7 mL), and the mixture was heated at 90 °C for 30 min. The
535 nA (mean ) 280 ( 173 (n ) 3)).
mixture was cooled to room temperature, filtered, washed with
and methyl, especially in the 5-, 6-, and 7-positions, was
found to increase the potency of the HNQs. 5,6,7-
Trichloro HNQ (8i) is the most potent antagonist with
an IC50 of 220 nM in DCKA binding and a Kb of 79 nM
in electrophysiological assays. HNQ 8i is 240-fold
selective for NMDA vs AMPA receptors. Comparison
of the SAR of HNQ with the SAR of QX and QTO
suggests that these three classes of antagonists most
probably bind to the glycine site in a similar manner.
water, and dried to leave 562 mg (88%) of 8f as a yellow solid:
mp 170-171 °C dec; H NMR (CDCl3 + DMSO-d6) 6.91 (d, J
) 1.8, 1H), 7.00 (d, J ) 1.8, 1H), 11.99 (sb, 1H); HRMS calcd
for C9H435Cl2N2O4 273.9544, found 273.9551. Anal. (C9H4-
Cl2N2O4) C, H; N: calcd, 10.18; found, 9.75.
HNQs 8a -e,h ,j,k were prepared from the corresponding
2,4-quinolinediols using the same procedures.
4-Hyd r oxy-3-n itr oqu in olin -2(1H)-on e (8a ): mp 216 °C
dec (lit.12 mp 216 °C); 1H NMR (CDCl3 + DMSO-d6) 7.01 (t, J
) 7.5, 1H), 7.11 (d, J ) 7.5, 1H), 7.40 (t, J ) 7.2, 1H), 7.88 (d,
J ) 7.8, 1H), 11.72 (mb, 1H).
1
7-Ch lor o-4-h ydr oxy-3-n itr oqu in olin -2(1H)-on e (8b): mp
220-222 °C dec; 1H NMR (DMSO-d6) 7.21 (m, 2H), 7.94 (d, J
) 7.4, 1H), 11.69 (s, 1H); HRMS calcd for C9H535ClN2O4
239.9938, found 239.9946. Anal. (C9H5ClN2O4) C, H, N.
6-Ch lor o-4-h ydr oxy-3-n itr oqu in olin -2(1H)-on e (8c): mp
Exp er im en ta l Section
Ch em istr y. Melting points were determined in open
capillary tubes on a Mel-Temp apparatus and are uncorrected.
The 1H NMR spectra were recorded at 300 MHz. Chemical
shifts are reported in ppm (δ), and J coupling constants are
reported in Hz. Elemental analyses were performed by Desert
Analytics, Tucson, AZ. Mass spectra (MS) were obtained with
a VG 12-250 or a VG ZAB-2FHF mass spectrometer. The
preparation of 2,4-quinolinediols 7b-k has been reported.11
6,7,8-Tr ich lor o-2,4-qu in olin ed iol (7l). To a solution of
2,3,4-trichloroaniline (985 mg, 5.0 mmol) and Et3N (714 mg,
7.05 mmol) in anhydrous ether (80 mL) was added dropwise
a solution of ethyl malonyl chloride (900 mg, 6.0 mmol) in dry
ether (15 mL) at room temperature. The resulting mixture
was stirred at room temperature under N2 for 20 h. The
reaction mixture was diluted with H2O (50 mL) and then
acidified to pH ) 2 with 2 N HCl. The ether layer was
separated, and the aqueous phase was extracted with ether
(2 × 25 mL). The ether solutions were combined, washed with
brine, dried (Na2SO4), and evaporated to dryness. To the
1
223-225 °C dec (lit.12 mp 200 °C); H NMR (DMSO-d6) 7.21
(d, J ) 8.7, 1H), 7.63 (d, J ) 8.7, 1H), 7.96 (s, 1H), 11.80 (s,
1H); HRMS calcd for C9H535ClN2O4 239.9938, found 239.9929.
5-Ch lor o-4-h ydr oxy-3-n itr oqu in olin -2(1H)-on e (8d): mp
1
180-182 °C dec; H NMR (DMSO-d6) 7.20 (m, 2H), 7.46 (m,
1H), 11.73 (s, 1H); HRMS calcd for C9H535ClN2O4 239.9938,
found 239.9929. Anal. (C9H5ClN2O4) C, H, N.
8-Ch lor o-4-h ydr oxy-3-n itr oqu in olin -2(1H)-on e (8e): mp
205-206 °C (lit.12 mp 204-205 °C); 1H NMR (DMSO-d6) 7.21
(dd, J ) 8.1, 7.5, 1H), 7.71 (d, J ) 7.5, 1H), 7.95 (d, J ) 8.1,
1H), 10.87 (bs, 1H); HRMS calcd for C9H535ClN2O4 239.9938,
found 239.9934.
5,7-Dim e t h yl-4-h yd r oxy-3-n it r oq u in olin -2(1H )-on e
(8h ): mp 188-190 °C (lit.12 mp 226 °C); 1H NMR (DMSO-d6)
2.28 (s, 3H), 2.66 (s, 3H), 6.61 (s, 1H), 6.81 (s, 1H), 11.52 (s,
1H); HRMS calcd for C11H10N2O4 234.0639, found 234.0651.

4686 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23
Notes
1991, 34, 1283-1292. (c) Swartz, K. J .; Koroshetz, W. J .; Rees,
A. H.; Huettner, J . E. Competitive Antagonism of Glutamate
Receptor Channels by Substituted Benzazepines in Cultured
Cortical Neurons. Mol. Pharmacol. 1992, 41, 1130-1141. (d)
Nagata, R.; Tanno, N.; Kodo, T.; Ae, N.; Yamaguchi, H.;
Nishimura, T.; Antoku, F.; Tatsuno, T.; Kato, T.; Tanaka, Y.;
Nakamura, M. Tricyclic Quinoxalinediones: 5,6-Dihydro-1H-
pyrrolo[1,2,3-de]quinoxaline-2,3-diones and 6,7-Dihydro-1H,5H-
pyrido[1,2,3-de]quinoxaline-2,3-diones as Potent Antagonists for
the Glycine Binding Site of the NMDA Receptor. J . Med. Chem.
1994, 37, 3956-3968.
7-Ch lor o-6,8-diflu or o-4-h ydr oxy-3-n itr oqu in olin -2(1H)-
on e (8j): mp 200-201 °C dec; 1H NMR (DMSO-d6) 7.66 (d, J
) 9.3, 1H), 11.07 (mb, 1H); HRMS calcd for C9H335ClF2N2O4
275.9746, found 275.9741. Anal. (C9H3ClF2N2O4) C, H; N:
calcd, 10.13; found, 9.67.
5,6,7,8-Tet r a flu or o-4-h yd r oxy-3-n it r oq u in olin -2(1H )-
on e (8k ): mp 176-177 °C dec; 1H NMR (DMSO-d6) 10.80 (mb,
1); HRMS calcd for C9H2F4N2O4 277.9948, found 277.9970.
Anal. (C9H2F4N2O4) C, H, N.
6,7-Dich lor o-4-h yd r oxy-3-n itr oqu in olin -2(1H)-on e (8g).
To a mixture of 6,7-dichloro-2,4-quinolinediol and 5,6-dichloro-
2,4-quinolinediol (1:1)11 (1.25 g, 5.43 mmol) in glacial acetic
acid (8 mL) was added HNO3 (69-71%, 1.5 mL), and the
mixture was heated at 90 °C for 2 h. The mixture was cooled
to room temperature, filtered, washed with water, and dried
to leave 280 mg (19%) of 8g as a yellow solid: mp 235-236 °C
(5) (a) Kemp, J . A.; Leeson, P. D. The Glycine Site of the NMDA
Receptor - Five Years on. Trends Pharmacol. Sci. 1993, 14, 20-
25. (b) Leeson, P. D.; Iversen, L. L. The Glycine Site on the
NMDA Receptor: Structure-Activity Relationships and Thera-
peutic Potential. J . Med. Chem. 1994, 37, 4053-4067.
(6) McQuaid, L. A.; Smith, E. C. R.; Lodge, D.; Pralong, E.; Wikel,
J . H.; Calligaro, D. O.; O’Malley, P. J . 3-Phenyl-4-hydroxy-
quinoline-2(1H)-ones: Potent and Selective Antagonists at the
Strychnine-Insensitive Glycine Site on the N-Methyl-D-aspartate
Receptor Complex. J . Med. Chem. 1992, 35, 3423-3425.
(7) Kulagowski, J . J .; Baker, R.; Curtis, N. R.; Leeson, P. D.; Mawer,
I. M.; Moseley, A. M.; Ridgill, M. P.; Rowley, M.; Stansfield, I.;
Foster, A. C.; Grimwood, S.; Hill, R. G.; Kemp, J . A.; Marshall,
G. R.; Saywell, K. L.; Tricklebank, M. D. 3’-(Arylmethyl)- and
3’-(Aryloxy)-3-phenyl-4-hydroxyquinoline-2(1H)-ones: Orally Ac-
tive Antagonists of the Glycine Site on the NMDA Receptor. J .
Med. Chem. 1994, 37, 1402-1405.
(8) Rowley, M.; Leeson, P. D.; Stevenson, G. I.; Moseley, A. M.;
Stansfield, I.; Sanderson, I.; Robinson, L.; Baker, R.; Kemp, J .
A.; Marshall, G. R.; Foster, A. C.; Grimwood, S.; Tricklebank,
M. D.; Saywell, K. L. 3-Acyl-4-hydroxyquinolin-2(1H)-ones.
Systemically Active Anticonvulsants Acting by Antagonism at
the Glycine Site of the N-Methyl-D-aspartate Receptor Complex.
J . Med. Chem. 1993, 36, 3386-3396.
(9) 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, Kay L. 3-Nitro-3,4-dihydro-2(1H)-quinolones. Ex-
citatory Amino Acid Antagonists Acting at Glycine-Site NMDA
and (RS)-R-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic Acid
Receptors. J . Med. Chem. 1993, 36, 3397-3408.
(10) Keana, J . F. W.; Kher, S. M.; Cai, S. X.; Dinsmore, C. M.; Glenn,
A. G.; Guastella, J .; Huang, J . C.; Lu, Y.; Ilyin, V.; Lu, Y.;
Mouser, P. L.; Woodward, R. M.; Weber, E. Synthesis and
Structure-Activity Relationships of Substituted 1,4-Dihydroqui-
noxaline-2,3-diones (QXs): Antagonists of the Glycine Site on
the NMDA Receptor and of Non-NMDA Glutamate Receptors.
J . Med. Chem. 1995, 38, 4367-4379.
(11) Cai, S. X.; Zhou, Z.-L.; Huang, J .-C.; Whittemore, E. R.; Egbu-
woku, Z. O.; Lu, Y.; Hawkinson, J . E.; Woodward, R. M.; Weber,
E.; Keana, J . F. W. Synthesis and Structure-Activity Relation-
ships of 1,2,3,4-Tetrahydroquinoline-2,3,4-trione-3-oximes: Novel
and Highly Potent Antagonists for NMDA Receptor Glycine Site.
J . Med. Chem. 1996, 39, 3248-3255.
(12) Buckle, D. R.; Cantello, B. C. C.; Smith, H.; Spicer, B. A.
4-Hydroxy-3-nitro-2-quinolones and Related Compounds as In-
hibitors of Allergic Reactions. J . Med. Chem. 1975, 18, 726-
732.
(13) 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
Relationships 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.
(14) Bigge, C. F.; Malone, T. C.; Boxer, P. A.; Nelson, C. B.; Ortwine,
D. F.; Schelkun, R. M.; Retz, D. M.; Lescosky, L. J .; Borosky, S.
A.; Vartanian, M. G.; Schwarz, R. D.; Campbell, G. W.; Ro-
bichaud, L. J .; Watjen, F. Synthesis of 1,4,7,8,9,10-Hexahydro-
9-methyl-6-nitropyrido[3,4-f]-quinoxaline-2,3-dione and Related
Quinoxalinediones: Characterization of R-Amino-3-hydroxy-5-
methyl-4-isoxazolepropionic Acid (and N-Methyl-D-aspartate)
Receptor and Anticonvulsant Activity. J . Med. Chem. 1995, 38,
3720-3740.
1
dec; H NMR (CDCl3 + DMSO-d6) 7.36 (s, 1H), 8.04 (s, 1H),
11.37 (s, 1H); HRMS calcd for C9H435Cl2N2O4 273.9544, found
273.9548. Anal. (C9H4Cl2N2O4) C, H; N: calcd, 10.18; found,
9.76.
5,6,7-Tr ich lor o-4-h yd r oxy-3-n it r oq u in olin -2(1H )-on e
(8i). A mixture of 5,6,7-trichloro-2,4-quinolinediol (7i) (264
mg, 1.0 mmol) and KNO3 (120 mg, 1.2 mmol) in trifluoroacetic
acid (5 mL) was refluxed for 24 h, cooled to room temperature,
and diluted with water (20 mL). The precipitate was filtered,
washed with water (3 × 2 mL) and ethanol (2 × 2 mL), and
dried to give 120 mg (39%) of 8i as a yellow powder: mp 225
°C dec; 1H NMR (DMSO-d6) 7.54 (s, 1H), 10.93 (bs, 1H). Anal.
(C9H3Cl3N2O4‚0.2H2O) C, H, N.
6,7,8-Tr ich lor o-4-h yd r oxy-3-n it r oq u in olin -2(1H )-on e
(8l). HNQ 8l was prepared from 7l in a manner similar to 8i
as an orange-yellow powder: mp 222-224 °C; 1H NMR
(DMSO-d6) 8.01 (s, 1H), 9.96 (bs, 1H). Anal. (C9H3Cl3N2O4‚
0.2H2O) C, H; N: calcd, 8.94; found, 8.22.
P h a r m a cology. DCKA Bin d in g Assa y. The [3H]DCKA
binding assay was performed according to the method reported
in ref 11 for QTOs.
Electr op h ysiology. Electrophysiological assay of selected
HNQs on NMDA and AMPA receptors was carried out accord-
ing to the method reported in ref 11.
Ack n ow led gm en t. We thank Dr. Yan Ni and Dr.
Ricardo Miledi (University of California, Irvine) for the
generous gift of rat cerebral cortex poly(A)+ RNA used
in this study, Dr. P. H. Seeburg (Heidelburg University,
Heidelburg, Germany) for the cDNAs encoding rat NR1
and NR2 subunits, and Dr. J ohn Guastella (CoCensys,
Inc.) for preparation of cRNAs encoding NMDA receptor
subunits. We also thank Dr. Yixin Lu of the University
of Oregon for preparation of HNQ 8h . Financial support
was provided in part by CoCensys, Inc., and by the
National Institute of Drug Abuse (DA-06726).
Refer en ces
(1) Doble, A. Excitatory Amino Acid Receptors and Neurodegen-
eration. Therapie 1995, 50, 319-337.
(2) Rothman, S. M.; Olney, J . W. Excitotoxicity and the NMDA
Receptor - Still Lethal after Eight Years. Trends Neurosci. 1995,
18, 57-58.
(3) J ohnson, J . W.; Ascher, P. Glycine Potentiates the NMDA
Response in Cultured Mouse Brain Neurons. Nature (London)
1987, 325, 529-531.
(4) (a) 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
Relationships 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. (b) Gray, N. M.; Dappen, M. S.; Cheng, B. K.; Cordi,
A. A.; Biesterfeldt, J . P.; Hood, W. F.; Monahan, J . B. Novel
Indole-2-carboxylates as Ligands for the Strychnine-Insensitive
N-Methyl-D-aspartate-Linked Glycine Receptor. J . Med. Chem.
(15) The pKa of QX (9a ) was estimated by UV spectroscopy to be 9.5
(9.7 as reported by Krishnamurthy, M.; Iyer, K. A.; Dogra, S. K.
Electronic Structure of Quinoxaline-2,3(1H,4H)dione and Its
Prototropic Species in the Ground and Excited Singlet States.
J . Photochem. 1987, 38, 277). The pKa of HNQ (8a ) was
estimated to be 8.5.
J M960520Y