Analogs of 3-hydroxy-1H-1-benzazepine-2,5-dione: Structure-activity relationship at N-methyl-D-aspartate receptor glycine sites

Article abstract of DOI:10.1021/jm960479z

A series of aromatic and azepine ring-modified analogs of 3-hydroxy-1H- 1-benzazepine-2,5-dione (HBAD) were synthesized and evaluated as antagonists at NMDA receptor glycine sites. Aromatic ring-modified HBADs were generally prepared via a Schmidt reaction with substituted 2-methoxynaphthalene-1,4- diones followed by demethylation. Electrophilic aromatic substitution of benzazepine 3-methyl ethers gave 7-substituted analogs. The preparation of multiply substituted 2-methoxynaphthalene-1,4-diones was effected via Diels- Alder methodology utilizing substituted butadienes with 2- methoxybenzoquinones followed by aromatization. Structural modifications, such as elimination of the aromatic ring, removal of the 3-hydroxyl group, and transfer of the hydroxyl group from C-3 to C-4, were also studied. An initial evaluation of NMDA antagonism was performed using a [³H]MK801 binding assay. HBADs demonstrating NMDA antagonist activity as indicated by inhibition of [³H]MK801 binding were further evaluated employing a [³H]- 5,7-dichlorokynurenic acid (DCKA) glycine site binding assay. Selected HBADs were characterized for functional antagonism of NMDA and AMPA receptors using electrophysiological assays in Xenopus oocytes and cultured rat cortical neurons. Antagonist potency of HBADs showed good correlation between the different assay systems. HBADs substituted at the 8-position possessed the highest potency with the 8-methyl (5), 8-chloro (6), and 8-bromo (7) analogs being the most active. For HBAD 6, the IC₅₀ in [³H]-DCKA binding assays was 0.013 μM and the K(b) values for antagonism of NMDA receptors in oocytes (NR1a/2C) and cortical neurons were 0.026 and 0.048 μM, respectively. HBADs also antagonized AMPA-preferring non-NMDA receptors expressed in oocytes but at a lower potency than corresponding inhibition of NMDA receptors. HBADs demonstrating a high potency for NMDA glycine sites showed the highest steady-state selectivity index relative to AMPA receptors. Substitution at the 6-, 7-, and 9-positions generally reduced or eliminated glycine site affinity. Moving the hydroxyl group from C-3 to C-4 reduced receptor affinity, and potency was eliminated by the removal of the aromatic ring or the hydroxyl group. These data indicate that the HBAD series has specific structural requirements for high receptor affinity. With the exception of substitution at C-8, modified HBADs generally have a lower affinity at NMDA receptor glycine sites than the parent compound 3. Mouse maximum electroshock-induced seizure studies show that the three HBADs selected for testing have in vivo potency with the 6,8-dimethyl analog (52) being the most potent (ED₅₀ = 3.9 mg/kg, iv).

Full text of DOI:10.1021/jm960479z

J . Med. Chem. 1996, 39, 4643-4653
4643
An a logs of 3-Hyd r oxy-1H-1-ben za zep in e-2,5-d ion e: Str u ctu r e-Activity
Rela tion sh ip a t N-Meth yl-D-a sp a r ta te Recep tor Glycin e Sites
Anthony P. Guzikowski,† Sui Xiong Cai,‡ Stephen A. Espitia,‡ J on E. Hawkinson,‡ J ames E. Huettner,§
Daniel F. Nogales,† Minhtam Tran,‡ Richard M. Woodward,‡ Eckard Weber,*,‡ and J ohn F. W. Keana*,†
CoCensys, Inc., 213 Technology Drive, Irvine, California 92618, Department of Cell Biology and Physiology,
Washington University School of Medicine, St. Louis, Missouri 63110, and Department of Chemistry,
University of Oregon, Eugene, Oregon 97403
Received J uly 3, 1996X
A series of aromatic and azepine ring-modified analogs of 3-hydroxy-1H-1-benzazepine-2,5-
dione (HBAD) were synthesized and evaluated as antagonists at NMDA receptor glycine sites.
Aromatic ring-modified HBADs were generally prepared via a Schmidt reaction with substituted
2-methoxynaphthalene-1,4-diones followed by demethylation. Electrophilic aromatic substitu-
tion of benzazepine 3-methyl ethers gave 7-substituted analogs. The preparation of multiply
substituted 2-methoxynaphthalene-1,4-diones was effected via Diels-Alder methodology
utilizing substituted butadienes with 2-methoxybenzoquinones followed by aromatization.
Structural modifications, such as elimination of the aromatic ring, removal of the 3-hydroxyl
group, and transfer of the hydroxyl group from C-3 to C-4, were also studied. An initial
evaluation of NMDA antagonism was performed using a [3H]MK801 binding assay. HBADs
demonstrating NMDA antagonist activity as indicated by inhibition of [3H]MK801 binding were
further evaluated employing a [3H]-5,7-dichlorokynurenic acid (DCKA) glycine site binding
assay. Selected HBADs were characterized for functional antagonism of NMDA and AMPA
receptors using electrophysiological assays in Xenopus oocytes and cultured rat cortical neurons.
Antagonist potency of HBADs showed good correlation between the different assay systems.
HBADs substituted at the 8-position possessed the highest potency with the 8-methyl (5),
8-chloro (6), and 8-bromo (7) analogs being the most active. For HBAD 6, the IC50 in [3H]-
DCKA binding assays was 0.013 µM and the Kb values for antagonism of NMDA receptors in
oocytes (NR1a/2C) and cortical neurons were 0.026 and 0.048 µM, respectively. HBADs also
antagonized AMPA-preferring non-NMDA receptors expressed in oocytes but at a lower potency
than corresponding inhibition of NMDA receptors. HBADs demonstrating a high potency for
NMDA glycine sites showed the highest steady-state selectivity index relative to AMPA
receptors. Substitution at the 6-, 7-, and 9-positions generally reduced or eliminated glycine
site affinity. Moving the hydroxyl group from C-3 to C-4 reduced receptor affinity, and potency
was eliminated by the removal of the aromatic ring or the hydroxyl group. These data indicate
that the HBAD series has specific structural requirements for high receptor affinity. With
the exception of substitution at C-8, modified HBADs generally have a lower affinity at NMDA
receptor glycine sites than the parent compound 3. Mouse maximum electroshock-induced
seizure studies show that the three HBADs selected for testing have in vivo potency with the
6,8-dimethyl analog (52) being the most potent (ED50 ) 3.9 mg/kg, iv).
In tr od u ction
activation of NMDA receptors.4 Therefore, compounds
such as 3′-(3-thienyloxy)-3-phenyl-4-hydroxy-1H-quino-
lin-2-one (1),5 ACEA 1021 (2),6 and 8-chloro-3-hydroxy-
1H-1-benzazepine-2,5-dione (6),7 which act as antago-
nists at NMDA receptor glycine sites, may offer a means
for the treatment of neurodegenerative disorders as-
sociated with excitotoxicity.8
N-Methyl-D-aspartate (NMDA), kainate, and R-amino-
3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA)
receptors are the three major types of ionotropic exci-
tatory amino acid receptors found in the mammalian
central nervous system.1 The NMDA receptor in par-
ticular has been associated with a variety of acute and
chronic neurological disorders including the brain dam-
age resulting from stroke, epilepsy, Alzheimer’s disease,
and AIDS-related dementia.2 Receptor overstimulation,
caused by excessive glutamate in the synaptic region,
is believed to contribute to the symptomatology of these
disorders. The overstimulation leads to pathological
levels of intracellular Ca2+ triggering a cascade of events
leading to neuronal death.3 In addition to glutamate,
the amino acid glycine is a necessary coagonist for
Our interest in the 3-hydroxy-1H-1-benzazepine-2,5-
dione (HBAD) series arose from a report that described
NMDA receptor glycine site antagonism by HBADs 3-5,
with 5 possessing submicromolar potency.9 Subse-
quently, the synthesis of 6-8 and several other halo-
substituted HBADs was described in the patent litera-
ture,7 but the pharmacological characterization of these
compounds was very limited. In order to establish a
more detailed structure-activity relationship (SAR), a
series of structurally modified HBADs was synthesized.
NMDA antagonist activity was determined initially by
glycine dependent inhibition of [3H]MK801 binding.
Affinity at NMDA receptor glycine sites was determined
for the active compounds using a [3H]-5,7-dichloro-
* Authors to whom correspondence should be addressed.
† University of Oregon.
‡ CoCensys, Inc.
§ Washington University School of Medicine.
X Abstract published in Advance ACS Abstracts, October 1, 1996.
S0022-2623(96)00479-7 CCC: $12.00 © 1996 American Chemical Society

4644 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23
Guzikowski et al.
Sch em e 1a
Sch em e 2a
a
(a) NaN3, H2SO4; (b) KNO3, H2SO4; (c) NBS, H2SO4; (d) H2SO4,
H2O/EtOH, ∆.
a
(a) H2, quinoline, MeOH/CH2Cl2; (b) Ac2O, MeOH/CH2Cl2; (c)
TFAA, EtOAc; (d) NaNO2, HCl, H2O then NaN3; (e) H2SO4, H2O/
EtOH, ∆.
kynurenic acid (DCKA) assay. Functional antagonism
of NMDA and AMPA receptors was determined using
electrophysiological assays. In vivo potency was mea-
sured using a mouse maximum electroshock-induced
seizure (MES) model.
Sch em e 3a
Ch em istr y
Ar om a tic Rin g-Su bstitu ted HBADs. Aromatic
ring-modified HBADs were generally synthesized via a
Schmidt reaction of appropriate 2-methoxynaphthlene-
1,4-diones as described by Birchall and Rees10 with
subsequent modifications described by Chapdelaine and
McLaren7 followed by demethylation by either aqueous
acid hydrolysis or with BBr3 in CH2Cl2. HBADs 3 and
5-8 were prepared as previously reported.7,9,10
The 7-substituted HBADs 13 and 14 were obtained
by direct electrophilic nitration or bromination of 10
(formed in situ by a Schmidt reaction on 911) followed
by hydrolysis of the intermediate methyl ethers 11 and
12 (Scheme 1). The regiochemistry of the Schmidt
reaction and of the nitration reaction was confirmed by
single-crystal X-ray analysis of 11. The regiochemistry
of the bromination reaction was assigned by analogy
with the nitration reaction and by NMR comparison of
14 with 7.
Catalytic hydrogenation of 11 employing a quinoline-
poisoned catalyst to inhibit reduction of the 3,4-double
bond gave amine 15 (Scheme 2). Derivatizations of the
amino group of 15 gave the methyl ethers 16-18.
Aqueous acid hydrolysis gave HBADs 19-21.
The synthesis of multiply substituted HBADs em-
ployed Diels-Alder methodology for the preparation of
the requisite intermediate 2-methoxynaphthalene-1,4-
diones (Scheme 3).12 Allowing methyl-substituted buta-
dienes 22-25 to react with 2613 yielded the correspond-
ing crude adducts (not shown). The reactions of dienes
22 and 23 were approximately 90% regioselective as
determined by 1H NMR. Bohlman et al.12 reported that
the major regioisomer obtained from the reaction of 22
a
(a) Toluene, hydroquinone, ∆; (b) O2, Et3N, MeOH; (c) DDQ,
toluene, ∆; (d) NaN3, H2SO4 or CF3SO3H; (e) BBr3, CH2Cl2.
with 26 had the methyl group in the 8-position of the
hydronaphthalene ring system. The regiochemical as-
signment for the major product of this reaction was
tentatively based on this report. Confirmation of this
assignment was made by single-crystal X-ray analysis
of the corresponding HBAD (vide infra). The regio-
chemical assignment for the major adduct derived from
23 was based on analogy with the reaction of 22. The
crude adducts derived from 22-24 were directly con-
verted into quinone derivatives 28-30,12,14 respectively,
via a base-catalyzed oxidation followed by purification.
In the case of the adduct derived from 25, this oxidation
stopped at the 5,8-dihydronaphthalene-1,4-dione 27.
Treatment of 27 with DDQ in refluxing toluene gave
the desired 31. Quinones 28-31 were carried on to the

Analogs of 3-Hydroxy-1H-1-benzazepine-2,5-dione
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23 4645
Ta ble 1. Physical Data, NMR Spectroscopic Data, and Methods of Preparation for Substituted HBADs
compd
no.
method/
R6
H
R7
NO2
R8 R9 yield (%)a
mp (°C)
250 decd
formulab
C10H6N2O5
1H NMRc
13
14
19
H
H
H
H
H
H
A/58
A/46
A/70
6.45, 7.61 (d, J ) 9.0), 8.39 (dd, J ) 9.0, 2.7),
8.79 (d, J ) 2.7), 11.14 (b), 12.04
H
H
Br
284-286 dec
C
10H6BrNO3‚0.16H2O 6.42, 7.41 (d, J ) 9.0), 7.80 (dd, J ) 8.7, 2.4),
8.10 (d, J ) 2.1), 10.86 (b), 11.71
AcNH
310 decd
C12H10N2O4‚0.40H2O
2.03 (3H), 6.41, 7.39 (d, J ) 8.7), 7.86 (dd,
J ) 8.7, 2.1), 8.20 (d, J ) 1.8), 10.15,
10.60 (b), 11.59
20
21
36
37
38
39
51
52
53
H
TFAcNH
H
H
H
A/74
330 decd
C12H7F3N2O4‚0.20H2O 6.43, 7.49 (d, J ) 8.7), 7.90 (dd, J ) 8.7, 2.1),
8.36 (d, J ) 2.1), 10.73 (b), 11.45, 11.69
H
N3
H
H
A/18e
190 decd
C10H6N4O3
6.42, 7.39 (dd, J ) 8.7, 2.7). 7.51 (d, J ) 8.7),
7.66 (d, J ) 2.7), 10.78 (b), 11.70
2.46 (3H), 6.28, 7.19 (t, J ) 7.5), 7.48 (d,
J ) 7.2), 7.67 (d, J ) 7.8), 9.94, 10.85 (b)
2.28 (3H), 2.42 (3H), 6.27, 7.31, 7.47,
9.90, 10.74 (b)
H
H
Me B/52
Me B/69
184-186 dec
211-213 dec
C
C
11H9NO3‚0.50H2O
H
Me
Me
H
H
12H11NO3
H
Me
H
H
B/77
302-304 decf C12H11NO3
2.23 (3H), 2.24 (3H), 6.39, 7.24, 7.80,
10.46, 11.49
Me
Me
Me
Me B/12
183-185 dec
188-189 dec
189-191 dec
206-208 dec
C
C
C
C
12H11NO3
11H9NO3
12H11NO3
12H11NO3
2.24 (3H), 2.35 (3H), 6.20, 7.03 (d, J ) 7.8),
7.25 (d, J ) 7.8), 9.98, 10.53 (b)
2.34 (3H), 6.30, 7.09 (d, J ) 7.2), 7.21 (d,
J ) 8.1), 7.37 (t, J ) 7.8), 10.39, 11.20
2.26 (3H), 2.31 (3H), 6.28, 6.92, 7.01,
10.29, 11.12
H
H
H
H
H
B/45
B/49
B/55
H
Me
H
Me Me
2.16 (3H), 2.23 (3H), 6.30, 7.09 (d, J ) 8.1),
7.28 (d, J ) 8.4), 10.31 (b), 11.08
2.17 (3H), 2.23 (3H), 6.34, 7.23, 10.72, 11.40
1.18 (t, J ) 7.5, 3H), 2.64 (q, J ) 7.5, 2H), 6.40,
7.13 (d, J ) 8.4), 7.31, 7.97 (d,
55
60
Me NO2
H
Me
Et
H
H
B/43
g
230-231 dec C12H10N2O5
250-251 dec 12H11NO3
H
C
J ) 8.1), 10.51, 11.51
a
For detailed methods, see the Experimental Section. Method A is H2SO4 in H2O/EtOH, ∆. Method B is BBr3 in CH2Cl2 at 25 °C.
Analyses for C, H, and N are within (0.40% of theoretical values. c Unless otherwise noted, values listed are for one-proton singlets.
b
Abbreviations: b ) broad singlet, d ) doublet, dd ) doublet doublet, t ) triplet, q ) quartet, and m ) multiplet. Splitting values are
reported in hertz (Hz). Decomposition without melting. e Total yield for the reaction sequences starting from 15. f Lit. mp 288 °C dec;
d
g
ref 10. HBAD 60 was generously supplied by Prof. Alun H. Rees and prepared according to the procedures detailed in ref 10.
benzazepine methyl ethers 32-35 as described above.
Demethylation employing BBr3 in CH2Cl2 gave HBADs
36-39. The structure of 36 was established by single-
crystal X-ray analysis thus confirming the regioselec-
tivity of the Schmidt reaction of 28 and of the reported
Diels-Alder reaction of 22 with 26.12
Alder reaction of 2-bromo-5-methoxybenzoquinone (42)16
with a series of 1,3-dienes (Scheme 4). Dienophile 42
was prepared by the bromination of hydroquinone 4017
followed by periodate oxidation of bromohydroquinone
41. The adduct (not shown) derived from the reaction
of 22 with 42 formed readily but turned to a black tar
upon isolation. Therefore, this adduct and the adducts
derived from 23 and 45 were not isolated but im-
mediately treated with Et3N to yield the corresponding
5,8-dihydronaphthalene-1,4-diones (not shown) by de-
hydrobromination. Subsequent oxidation with MnO2 in
refluxing CHCl3 yielded 43, 44, and a mixture of 46 and
47, respectively. Compounds 43 and 44 were regio-
chemically pure by 1H NMR. The regiochemistry of
these compounds was assigned by NMR comparison
with 28 and 29. Quinones 46 and 47 were obtained as
a 3:7 mixture (by NMR), respectively. Isolation of 47
was effected by crystallization, and the regiochemistry
was tentatively based on analogy with 43 and 44.
Conversion of 43, 44, and 47 to HBADs 51-53 was
performed as described above.
The regiochemical assignments of HBAD 37 and its
1
precursor 29 were based on the H NMR comparison of
methyl ether 33 with methyl ethers 32, 34, and 35.
Chemical shift values for the amide proton of methyl
ethers 32 and 35, where a peri-methyl group is present,
are δ 9.45 and 9.96, respectively. Conversely, the
chemical shift of the amide proton of methyl ether 34,
which does not have a peri-methyl group, is δ 11.19. The
chemical shift of the amide proton of methyl ether 33
is δ 9.90, thus providing confirmation for the regio-
chemical assignments.
In order to prepare HBADs analogous to 36 and 37
but with the opposite regiochemistry in the aromatic
ring portion, the regioselectivity of the Diels-Alder
reaction required alteration. Tegmo-Larsson et al.15
reported that 2,5-dimethoxybenzoquinone underwent a
Diels-Alder reaction with 22 to give an adduct which,
upon loss of MeOH, gave the 5,8-dihydronaphthalene
with the methyl group in the desired 5-position. In our
hands, this reaction proceeded slowly and was impracti-
cal for our needs. Therefore, we investigated the Diels-
The regiochemistry of 50, and thus that of 47 and 53,
was confirmed by 1H NMR analysis. The chemical shift
value for the amide proton of 50 (δ 10.88) is consistent
with the values measured for the amide protons of 48
and 49 (δ 11.00 and 10.94, respectively) and inconsistent
with values measured for HBAD methyl ethers that

4646 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23
Guzikowski et al.
Sch em e 4a
Ta ble 2. Radioligand Binding Data at NMDA Receptors for
HBAD Analogs
a
(a) Br2, AcOH; (b) NaIO4, H2O/CHCl3; (c) toluene, ∆; (d) Et3N,
toluene/MeOH; (e) MnO2, CHCl3, ∆; (f) NaN3, H2SO4; (g) BBr3,
CH2Cl2.
Sch em e 5a
a
KNO3, H2SO4; (b)BBr3, CH2Cl2.
possess a peri-methyl group (vide supra). Also, irradia-
tion of the amide proton of 50 gave the expected NOE
enhancement (22%) of the H-9 doublet.
a
IC50 values are mean ( SEM and are the result of a minimum
of three determinations.
Nitration of 49 (formed in situ via a Schmidt reaction
of 44) gave the nitromethyl ether 54 (Scheme 5).
Demethylation gave the 7-nitro HBAD 55. The regio-
chemistry of the nitration reaction was assigned by
analogy with the nitration of 10 and by a NOE experi-
ment on 54. Irradiation of the amide proton resulted
in a 23% enhancement of the H-9 singlet.
MK801 binding to rat brain cortical membranes by
methods previously described.21 HBADs inhibiting [3H]-
MK801 binding with an IC50 < 300 µM were considered
to be potentially active as antagonists at NMDA recep-
tors (Table 2). Active compounds were evaluated fur-
ther using the selective glycine site ligand [3H]DCKA.22
Oth er Modified 1H-1-Ben zazepin e-2,5-dion es. The
syntheses of 56-58 were performed according to known
literature procedures.18-20 Oxime 59 was prepared by
the reaction of 3 with NaNO2 in aqueous base followed
by treatment with H2SO4.
Electr op h ysiology in Xen opu s Oocytes
To confirm that HBADs are functional antagonists at
NMDA receptors, seven compounds were assayed for
inhibition of NMDA responses in Xenopus oocytes
expressing the cloned rat NMDA receptor subunit
combination NR1a/2C.23-25 Potency of antagonism was
assessed by determination of apparent antagonist dis-
sociation constants (Kb values) calculated from partial
(3-5 point) concentration-inhibition curves (Table
3).23,26 Calculation of Kb values assumes that inhibition
of membrane current responses is by competitive an-
tagonism at glycine coagonist sites.
Ra d ioliga n d Bin d in g Stu d ies
All HBADs were initially screened for NMDA antago-
nism by measuring inhibition of glycine dependent [3H]-
Like various other classes of glycine site antagonist,
e.g., quinoxalinediones (QXs),6 tetrahydroquinolin-3-

Analogs of 3-Hydroxy-1H-1-benzazepine-2,5-dione
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23 4647
Ta ble 3. Functional Antagonism of Rat Brain NMDA and AMPA Receptors by HBADs
Kb (µM)
Kb AMPA
compd no.
NMDA (glycine)a
AMPA (glutamate)b
Kb NMDAc
nd
Oocyte Recordings
3
5
6
7
8
1.3e (1.2-1.5)f
42 (39-46)
32
40
77
100
25
44
4, 3
3, 4
7, 5
4, 6
3, 4
4, 4
2, 2
0.25 (0.24-0.26)
0.026 (0.025-0.028)
0.026 (0.024-0.029)
0.21 (0.18-0.24)
0.086 (0.080-0.094)
9.9 (8.7-11)
2.0 (1.7-2.3)
2.7 (2.3-3.3)
5.3 (4.8-5.8)
3.8 (3.5-4.2)
38
56
>55
>240
Neuron Recordingsg
j
3h
5
6
3.0i (2.7-3.4)
0.47 (0.41-0.54)
0.048 (0.038-0.061)
0.061 (0.050-0.073)
65 (53-80)
6.4 (5.5-7.5)
10 (9.2-11.7)
22
14
210
2, 2
3, 3
5, 8
6, -
7
a
Inhibition of NMDA receptors was measured in oocytes expressing cloned rat brain receptor subunits (NR1a/2C).24 Affinity of HBADs
b
for glycine binding sites was estimated from inhibition of currents elicited by 1 µM glycine and 100 µM glutamate. Inhibition of AMPA-
preferring non-NMDA receptors was measured in oocytes expressing rat cerebral cortex poly(A)+ RNA. Affinity of HBADs for AMPA
receptors was estimated from inhibition of currents elicited by 10 µM AMPA. c Steady-state selectivity index was estimated by dividing
Kb AMPA by Kb NMDA. Number of cells examined for NMDA and AMPA, respectively. e Kb values given to two significant figures.
d
f 95% confidence intervals adjusted to the linear scale. Recordings were made using cultured rat cortical neurons. Values taken from
g
h
i
ref 9. Kb values for glycine sites were estimated from the rightward displacement of glycine concentration-response curves by fixed
j
concentrations of HBADs (NMDA ) 0.1 or 1 mM). Kb values for AMPA receptor glutamate sites were estimated from the displacement
of concentration response curves for kainate.
oximes,27 kynurenic acids (KAs),28 etc., HBADs also
antagonize AMPA-preferring non-NMDA receptors.9 In
the present study, antagonism of AMPA receptors was
measured in oocytes expressing rat cerebral cortex poly-
(A)+ RNA.23,29 As described for NMDA receptors, po-
tency of inhibition was estimated from partial concen-
tration-inhibition curves on currents elicited, in this
case, by 10 µM AMPA (Table 3).
Electr op h ysiology in Ma m m a lia n Neu r on s
F igu r e 1. Inhibition of NMDA responses by HBADs 6 and 7
Effects of HBADs on neuronal NMDA and AMPA
receptors were measured using whole cell recordings
from cultured rat cortical neurons. Potency of HBADs
6 and 7 was estimated from the rightward displacement
of steady-state agonist concentration-response curves.9,23
Agonists were glycine for NMDA receptors and kainate
for AMPA receptors.
in cultured rat cortical neurons. Steady-state membrane
current responses are expressed as a fraction of the maximum
response and plotted as the mean ( SEM. Smooth curves are
the best individual fits of the logistic equation to the data.
Simultaneous fits are not shown. EC50 and slope values are,
respectively: 0.18 µM and 1.1 for the control curve, 3.9 µM
and 1.5 in 1 µM HBAD 6, 3.2 µM and 1.6 in 1 µM HBAD 7. Kb
values calculated from the simultaneous fits of these data are
given in Table 3.
Under control conditions the EC50 and slope values
for activation of NMDA receptors by glycine were 0.18
µM and 1.1, respectively. Inhibition induced by HBADs
6 and 7 was associated with a rightward displacement
of the glycine concentration-response curve and was
fully surmountable when the glycine concentration was
increased to 100 µM (Figure 1). Kb values calculated
from the shift in apparent glycine affinity are given in
Table 3. These experiments confirm that antagonism
is predominantly due to competitive inhibition at glycine
sites. Antagonism did, however, deviate from the strict
competitive model (F1,45 ) 6.3 for 6, F1,50 ) 12 for 7).
For both HBADs, deviation from the model was due to
slightly increased slopes in the presence of drug.
Collectively, the neuronal recordings confirm that 6
and 7 are potent antagonists of NMDA receptors, that
the major mechanism of antagonism is competitive
inhibition at glycine sites, and that 6 shows a higher
degree of steady-state selectivity for NMDA receptors
than do previously described HBADs.9
In Vivo Mea su r em en ts
Systemic bioavailability and anticonvulsant effects of
three HBADs were measured using a mouse MES model
(3 determinations/HBAD).29,30 For each HBAD, the
onset of seizure protection peaked approximately 2 min
after iv administration and declined with a biological
half-time of approximately 10 min. The ED50 values
measured at the peak of protection for 6, 38, and 52
were 13 (11-17), 6.3 (5.1-7.7), and 3.9 (2.6-5.8) mg/
kg, respectively.
Under control conditions, the EC50 and slope values
for kainate as an agonist of AMPA receptors were 160
µM and 1.3, respectively. As described for glycine,
inhibition of AMPA receptors by HBAD 6 was due to a
rightward shift in the kainate concentration-response
relationship (not illustrated). With 50 µM HBAD 6, the
EC50 for kainate was 930 µM and the slope was 1.2. In
this case, inhibition conformed with the competitive
model (F1,96 ) 0.65). The Kb value calculated from this
pair of curves is given in Table 3.
Discu ssion
Str u ctu r e-Activity Rela tion sh ip a t NMDA Re-
cep tor Glycin e Sites. The [3H]MK801 assays show
that all three 9-substituted HBADs (36, 37, and 39) are

4648 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23
Guzikowski et al.
F igu r e 2. Putative pharmacophore for HBAD NMDA receptor
glycine site antagonists (adapted from ref 8b).
F igu r e 3. Correlation of HBAD antagonist potency for NMDA
receptors as determined by radioligand binding and electro-
physiological assays. For radioligand binding studies, IC50
values were determined by measuring displacement of [3H]-
DCKA from rat brain cortical membranes. For electrophysi-
ological assays, Kb values were estimated from inhibition of
NR1a/2C subunit combinations of cloned rat NMDA receptor
in Xenopus oocytes. In selected cases, Kb values were also
estimated from inhibition of NMDA responses in cultured rat
cortical neurons; > indicates an indefinite minimum value.
inactive as NMDA receptor antagonists. It seems likely
that 9-substitution inhibits hydrogen bonding between
the amide proton and a putative hydrogen-bonding site
on the glycine receptor8b thereby obviating binding
(Figure 2). Similar loss or reduction of affinity has been
noted for 5,6,7,8-tetrasubstituted QXs6 and 8-substitut-
ed KAs.28
In the 7-monosubstituted series, HBADs 14, 20, and
21 are inactive as NMDA receptor antagonists in the
[3H]MK801 assay. HBADs 13 and 19 have at best weak
affinity as determined by [3H]DCKA binding. A similar
reduction of potency in the case of 4 versus 3 has been
reported.9 The loss of affinity in this series may be the
result of an unfavorable steric interaction of 7-substit-
uents with the binding pocket since, in every HBAD
examined, the 7-substituted analog is significantly less
potent than the 7-unsubstituted counterpart.
The introduction of a 6-methyl group (51) reduces
binding affinity by over 1 order of magnitude relative
to 3. This is in contrast to the SAR observed for QXs
and KAs where the introduction of Me or Cl at the
5-position (assumed to correspond to the 6-position in
the HBAD series) improves the binding affinity.28
These data suggest that benzazepine type antagonists
have a different orientation in the receptor pocket than
do other antagonist series. The binding of the 7-mem-
bered azepine ring within the hydrophilic portion of the
receptor pocket may place C-6 and C-7 in a confined
region of the hydrophobic binding pocket (Figure 2).
The most active HBADs incorporate a single substitu-
ent at the 8-position. The introduction of a methyl (5),
chloro (6), or bromo (7) substituent to this position
increases the binding affinity relative to 3, with 6 and
7 having the highest affinities. These data confirm
previous findings7,9 and are also consistent with the
observed SAR for various antagonists at NMDA receptor
glycine sites such as 7-substituted QXs,6 KAs,28 and 2,3-
dihydro-KAs.31 This suggests that the 8-position in the
HBAD series and the 7-position of 6,6-fused ring glycine
site antagonists may occupy the same site within the
receptor pocket. Interestingly, the 8-F-HBAD 8 has
considerably less affinity than 6 or 7. The modest 2-fold
improvement in affinity of 8 compared to 3 likely stems
from a positive steric contribution and an unfavorable
electronic interaction between the receptor pocket sur-
face and the highly electronegative fluorine atom. A
similar trend is observed in the QX series.6 An ethyl
group in the 8-position has a negative effect on the
binding affinity with HBAD 60 being 6-fold less potent
than 3. This indicates a size limitation at the 8-position
with substituents such as bromo, chloro, and methyl
being allowed but larger groups being unfavored. This
effect is observed in the KA series.28
We have recently shown that multiple substitution
in the QX series greatly enhances antagonist potency
at NMDA receptor glycine sites.6 QXs that are trisub-
stituted in the 5-, 6-, and 7-positions are found to be
the most potent. We initially thought that this trend
might also apply to the HBAD series. The present
study, however, demonstrates that multiple substitution
generally results in a reduction of affinity. Dimethyl-
substituted HBADs 38 and 52 have approximately one-
half the affinity of 5. In light of the poor affinity of 51,
the affinity of 52 is better than expected. The 6,7-
dimethyl HBAD 53 is inactive as a NMDA receptor
antagonist. This is not surprising since both 6- and
7-substituents individually have a negative effect on
glycine site binding affinity. The poor affinity of trisub-
stituted HBAD 55 is consistent with these trends.
The binding affinities of analogs with major structural
modifications were also determined. The reduced af-
finity of 4-hydroxy analog 56 relative to 3 suggests that
the hydroxyl group of 56 is unable to effectively bind to
the proposed polar region of the putative glycine site
pharmacophore (Figure 2). Differences in ionizing
ability of the two isomers is not likely to be a major
factor. The inactivity observed for deshydroxy analog
57 and azepine 58 underscores the binding requirement
in the polar and hydrophobic regions, respectively.32,33
Oxime 59 is also inactive. It is of note that none of the
intermediate HBAD methyl ethers act as NMDA recep-
tor antagonists.34
Selectivity: NMDA Recep tor Glycin e Sites ver -
su s NMDA Glu ta m a te Sites. There is good agree-
ment between potencies of HBADs estimated by elec-
trophysiological assays and IC50 values determined by
the [3H]DCKA binding studies (Figure 3). In all cases
studied, there is <3-fold difference between the two
assay systems. There is no systematic trend within this
limited level of discrepancy. The unsubstituted 4-hy-
droxy analog 56 is essentially inactive in both types of
assays. These experiments confirm that HBADs are
functional antagonists of NMDA receptors and that

Analogs of 3-Hydroxy-1H-1-benzazepine-2,5-dione
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23 4649
potency of inhibition correlates well with [3H]DCKA
binding affinity at glycine sites. A previous study
indicates that some HBADs are also low-potency an-
tagonists of NMDA receptor glutamate sites.9 In the
present experiments, the close correlation between
glycine site affinity and functional potency implies that
inhibition of NMDA responses is predominantly due to
effects at glycine sites.
is reduced or eliminated by substitution in the 6-, 7-, or
9-positions, or any combinations thereof, and by repo-
sitioning or eliminating the hydroxyl group. Removing
the aromatic ring also eliminates affinity. The struc-
tural requirements in the size-limited hydrophobic
region are more specific in the HBAD series than for
other series of NMDA receptor glycine site antagonists
such as QXs and KAs. This may be a result of
constraints imposed by the 7-membered azepine ring
binding in the polar region of these receptors. HBADs
are systemically bioavailable with the 6,8-dimethyl-
substituted analog 52 possessing the greatest in vivo
potency.
Selectivity: NMDA Glycin e Sites ver su s Non -
NMDA Glu ta m a te Recep tor s. As observed for qui-
noxaline derivatives,35 HBAD potencies at AMPA re-
ceptors generally correlate with affinities at NMDA
receptor glycine sites. In particular, HBADs 6 and 7,
which display the highest affinity as glycine site an-
tagonists, are also the most potent AMPA receptor
antagonists. Even with this trend, however, the HBADs
with highest affinity for glycine sites still show the
highest steady-state selectivity index for NMDA recep-
tors relative to AMPA receptors (Table 3). Levels of
selectivity for the less potent HBADs drop to between
20- and 40-fold. As described for quinoxalinediones,23,29
the functional selectivity of HBADs for glycine and
AMPA in vivo will depend on additional factors, includ-
ing the steady-state levels of extracellular glycine and
glutamate (relative to their respective EC50s) and the
kinetics of receptor equilibration between the agonists
and the antagonist. For in vivo work, therefore, the
moderate potency HBADs are probably best thought of
as broad spectrum ionotropic glutamate receptor an-
tagonists.
In Vivo Activities of Selected HBADs. Various
examples of potent NMDA receptor glycine site antago-
nists, such as KAs,28 2-carboxytetrahydroquinolines,31
and many indolecarboxylic acids,36 demonstrate no
significant in vivo potency upon systemic administra-
tion. It is believed that this lack of bioavailability
results from poor compound penetration of the blood-
brain barrier. In the present study, mouse MES studies
show that HBADs 6, 38, and 52 are bioavailable by
demonstrating moderately potent anticonvulsant effects
after iv administration. Interestingly, the relative in
vivo potencies of the tested HBADs (52 > 38 > 6) are
the reverse of the inhibitory potencies measured for
these compounds at NMDA and, where tested, AMPA
receptors (Tables 2 and 3). This suggests that, within
a given compound series, in vivo potency is determined
by factors other than potency at the receptor.
Exp er im en ta l Section
Ch em istr y. Compounds 3, 5-8, and 56-58 were prepared
as previously reported.7,9,10,18-20 Reagents were used as re-
ceived unless otherwise noted. Melting points were measured
on a Thomas Hoover or Mel-Temp melting point apparatus
and are uncorrected. For compounds melting above 260 °C, a
preheated block was employed. CH2Cl2 was distilled from
CaH2 immediately prior to use. Solvent removal was routinely
performed on a rotoevaporator at 30-40 °C. All reactions were
performed under N2 unless otherwise noted. TLC analyses
were performed on plastic-backed F-254 silica gel plates. 1H
and 13C NMR spectra were recorded on a General Electric QE-
300 spectrometer. Chemical shifts are reported in δ units
referenced to the residual proton or the carbon signal of the
deuterated solvent (CHCl3, δ 7.26 or 77.23; CD3SOCD2H, δ
2.49 or 39.51). Infrared spectra were obtained on a Nicolet
Magna-IR 550 spectrophotometer. MS were recorded on a VG
ZAB-2-HF mass spectrometer with a VG-11-250 data system
in the electron ionization mode (70 eV). Microanalyses were
performed by Desert Analytics, Tuscon, AZ.
3-Meth oxy-7-n itr o-1H-1-ben za zep in e-2,5-d ion e (11). To
stirred, ice bath cold, concentrated H2SO4 (66 mL) was added
2-methoxynaphthalene-1,4-dione11 (9; 10.0 g, 53.1 mmol) in
portions to give a deep red solution. NaN3 (3.80 g, 58.4 mmol)
was added in portions. The ice bath was removed, and the
reaction mixture was allowed to stir at 25 °C for 20 h. The
reaction mixture was recooled in an ice bath, and KNO3 (5.90
g, 58.4 mmol) was added in small portions. Foaming was
noted during this addition. After foaming subsided, the cooling
bath was removed and the reaction mixture was allowed to
stir at 25 °C for 3.5 h. The reaction mixture was slowly added
to crushed ice (200 g) with additional ice being added as needed
(final volume 600 mL). The resulting yellow precipitate was
collected by filtration to yield a paste. The paste was sus-
pended in water (200 mL) and carefully neutralized with solid
NaHCO3. A solution of 30% MeOH/70% CHCl3 (1 L) was
added with stirring to dissolve the solid product. The phases
were separated, and the organic portion was washed with 30%
MeOH/70% H2O (2 × 500 mL). The organic portion was
filtered through cotton, and the solvent was removed. The
resulting yellow solid was crystallized from EtOH to yield 11
as near-colorless needles (5.2 g, 40%): mp 252-253 °C; 1H
NMR (DMSO-d6) δ 3.83 (s, 3 H), 6.42 (s, 1 H), 7.55 (d, J ) 9.0
Hz, 1 H), 8.40 (dd, J ) 9.0, 2.4 Hz, 1 H), 8.70 (d, J ) 2.7 Hz,
1 H), 11.84 (s, 1 H); 13C NMR (DMSO-d6) δ 56.76, 112.4, 122.0,
125.0, 125.7, 127.9, 141.2, 142.5, 156.5, 158.9, 183.0.
In the QX series,6,23,29,37 substituent type and position
on the size-limited hydrophobic region are determining
factors for bioavailability. Methyl-substituted QXs are
generally more bioavailable than the corresponding
halogen-substituted analogs. Also, compounds substi-
tuted at positions 5 and 7 generally have higher in vivo
potency than analogs substituted at positions 6 and 7.
The present experiments suggest that similar trends are
found in the HBAD series, though a larger number of
molecules require in vivo testing to confirm these
relationships.
7-Br om o-3-m eth oxy-1H-1-ben za zep in e-2,5-d ion e (12).
To stirred, ice bath cold, concentrated H2SO4 (6.6 mL) was
added 9 (1.00 g, 5.31 mmol) in portions to give a deep red
solution. NaN3 (690 mg, 10.6 mmol) was added in portions.
The ice bath was removed, and the reaction mixture was
allowed to stir at 25 °C for 20 h. NBS (1.03 g, 5.80 mmol)
was added in one portion, and the reaction mixture was
allowed to stir at 25 °C for 48 h. A second portion of NBS
(500 mg, 2.81 mmol) was added, and the reaction mixture was
allowed to stir for an additional 24 h. The reaction mixture
was added to 50 mL of crushed ice, and a solution of 30%
MeOH/70% CHCl3 (100 mL) was stirred in. The layers were
separated, and the aqueous portion was extracted with 30%
Su m m a r y
HBADs 5-7, which possess an 8-methyl, -chloro, and
-bromo substituent, respectively, are the highest affinity
benzazepine antagonists for the NMDA receptor glycine
sites tested. Replacing these substituents with fluoro
or ethyl (8 or 60) reduces the binding affinity. Affinity

4650 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23
Guzikowski et al.
MeOH/70% CHCl3 (3 × 35 mL). The combined organic portion
was washed with 30% MeOH/70% H2O (3 × 50 mL), filtered
through a cotton plug, and dried over anhydrous Na2SO4. The
solvent was removed to yield a yellow solid (900 mg). The solid
was heated to boiling in 10% MeOH/90% EtOAc (30 mL) to
yield a solid suspended in an orange solution. The solid was
collected by filtration, washed with EtOAc, and dried in vacuo
to yield a yellow powder (120 mg). The powder was crystal-
lized from EtOH to yield 12 as gold plates (94 mg, 6%): mp
additional 80 mg of crude product. The combined brown
powder consisted of 18, 21, and side products as determined
by TLC (57% 2-propanol, 20% dioxane, 11.5% water, 11.5%
NH4
OH). The side products were removed by dissolving the
powder in boiling EtOH (75 mL), concentrating the resulting
solution to 25 mL, and allowing the concentrate to cool to 25
°C. A precipitate formed which was collected by filtration to
yield 91 mg of a brown powder. The powder was dissolved in
boiling EtOH (100 mL). The resulting solution was allowed
to stir for 1 min at 25 °C; then a solution of 10% H2SO4/90%
water (10 mL) was added. The reaction mixture was allowed
to stir for 5 min at 25 °C; then an additional portion of the
acid solution (90 mL) was added. A precipitate formed after
a few minutes. After 12 min of total reaction time, the mixture
was added to crushed ice (50 mL) and the EtOH was removed.
The solid was collected and washed with water (10 × 2 mL).
The damp filter cake was crystallized from EtOH to yield 21
as an orange crystalline solid (55 mg, 18%): mp 190 °C dec
(without melting); 1H NMR (DMSO-d6) δ 6.42 (s, 1 H), 7.39
(dd, J ) 8.7, 2.7 Hz, 1 H), 7.51 (d, J ) 8.7 Hz, 1 H), 7.66 (d, J
) 2.7 Hz, 1 H), 10.78 (bs, 1 H), 11.70 (s, 1 H); IR (KBr) 2116
cm-1 (N3). Anal. (C10H6N4O3) C, H, N.
1
290-291 °C dec; H NMR (DMSO-d6) δ 3.80 (s, 3 H), 6.36 (s,
1 H), 7.35 (d, J ) 8.7 Hz, 1 H), 7.78 (dd, J ) 8.7, 2.1 Hz, 1 H),
7.99 (d, J ) 2.1 Hz, 1 H), 11.45 (s, 1 H). Anal. (C11H8BrNO3)
C, H, N.
7-Am in o-3-m eth oxy-1H-1-ben za zep in e-2,5-d ion e (15). A
mixture of 11 (250 mg, 1.01 mmol), quinoline (20 mg), and Pd/C
(10%, 100 mg) in CH2Cl2/MeOH (1:1, 250 mL) was hydroge-
nated (Parr) at 5 psig for 1 h at 25 °C. The catalyst was
removed by filtration (Celite), and the solvent was removed
to yield 15 as an orange powder. This was combined with an
identically prepared portion of 15 and was crystallized from
MeOH/EtOAc (1:1) to yield 15 as an orange solid (170 mg,
1
39%): mp 247-249 °C dec; H NMR (DMSO-d6) δ 3.76 (s, 3
H), 5.31 (s, 2 H), 6.28 (s, 1 H), 6.85 (dd, J ) 8.7, 2.7 Hz, 1 H),
7.03-7.15 (m, 2 H), 11.01 (s, 1 H). Anal. (C11H10N2O3) C, H,
N.
Gen er a l Met h od for t h e P r ep a r a t ion of 2-Met h oxy-
n a p h th a len e-1,4-d ion es via a Diels-Ald er Rea ction of
2-Meth oxyben zoqu in on e (26): Syn th esis of 2-Meth oxy-
6,8-d im eth yln a p h th a len e-1,4-d ion e (29). A suspension of
2613 (4.90 g, 35.5 mmol), 1,3-dimethylbuta-1,3-diene (23; 25
mL, technical grade), and hydroquinone (500 mg) in toluene
(50 mL) was stirred at 60 °C for 20 h. The resulting pale
orange, near-homogeneous solution was allowed to cool to 25
°C. The reaction mixture was filtered, and the solvent was
removed to give an orange solid (9.3 g). Without purification,
the solid was dissolved in MeOH (a gummy colorless solid that
failed to dissolve was removed by filtration). Et3N (3 mL) was
added, and the resulting dark solution was allowed to stir
vigorously under O2 for 2.5 h at 25 °C. The reaction mixture
was acidified with 10% HCl (100 mL) and diluted with water
(200 mL). The MeOH was removed to give an orange suspen-
sion. The suspension was extracted with CHCl3 (3 × 100 mL).
The extract was washed with saturated NaHCO3 (3 × 75 mL)
and water (1 × 200 mL) and filtered through cotton. The
solvent was removed to give an orange solid (5.9 g). Crystal-
lization from EtOH yielded 29 as yellow needles (2.79 g,
7-Acet a m id o-3-m et h oxy-1H -1-b en za zep in e-2,5-d ion e
(16). Hydrogenation of 11 (400 mg, 1.61 mmol) was performed
as described for 15. After filtration of the catalyst, acetic
anhydride (3 mL) was added to the filtrate and the resulting
solution was allowed to stir for 1 h at 25 °C to yield a yellow,
cloudy mixture. The mixture was filtered, and the solvent was
removed from the filtrate to yield 16 as a yellow powder (386
1
mg, 92%; ∼90% pure by H NMR and suitable for use in the
next reaction without further purification). An analytical
sample was prepared by crystallization from EtOH: mp 304-
1
306 °C dec; H NMR (DMSO-d6) δ 2.03 (s, 3 H), 3.79 (s, 3 H),
6.34 (s, 1 H), 7.33 (d, J ) 8.7 Hz, 1 H), 7.80 (dd, J ) 8.7, 2.1
Hz, 1 H), 8.13 (d, J ) 1.8 Hz, 1 H), 10.13 (s, 1 H), 11.28 (s, 1
H). Anal. (C13H12N2O4) C, H, N.
7-(Tr iflu or oa ceta m id o)-3-m eth oxy-1H-1-ben za zep in e-
2,5-d ion e (17). Hydrogenation of 11 (400 mg, 1.61 mmol) was
performed as described for 15. After filtration of the catalyst,
the solvent was removed from the filtrate to yield an orange
solid. This solid was dissolved in boiling EtOAc (700 mL),
filtered while hot, and allowed to cool to 25 °C to give a clear,
orange solution. TFAA (0.5 mL) was added, and the reaction
mixture was allowed to stir at 25 °C for 1 h, at which point a
yellow precipitate was present. The solvent was removed to
yield a yellow solid. Crystallization from EtOH yielded 17 as
yellow plates (260 mg, 52%): 1H NMR (DMSO-d6) δ 3.80 (s, 3
H), 6.37 (s, 1 H), 7.42 (d, J ) 8.7 Hz, 1 H), 7.87 (dd, J ) 8.7,
2.4 Hz, 1 H), 8.28 (d, J ) 2.1 Hz, 1 H), 11.41 (s, 1 H), 11.44 (s,
1 H). A portion was recrystallized from EtOH: mp 327-328
°C dec. Anal. (C13H9F3N2O4) C, H, N.
7-Azido-3-h ydr oxy-1H-1-ben zazepin e-2,5-dion e (21). Hy-
drogenation of 11 (400 mg, 1.61 mmol) was performed as
described for 15. After filtration of the catalyst, the solvent
was removed from the filtrate to yield an orange solid. This
solid was dissolved in boiling EtOAc (700 mL), filtered while
hot, and allowed to cool to 25 °C to give a clear, orange solution.
The solution was made acidic by the addition of anhydrous
HCl at 25 °C. Solvent removal yielded an orange/brown
powder (382 mg) which was utilized without further purifica-
tion. A portion (125 mg, 491 µmol) was suspended in water
(19 mL) and cooled in an ice bath. Concentrated HCl (9 mL)
was added while maintaining the temperature between 0 and
5 °C. A solution of NaNO2 (37 mg, 540 µmol) in water (1 mL)
was added in one portion. The reaction mixture was allowed
to stir between 0 and 5 °C for 1 h to yield a near-homogeneous
yellow solution. Solid NaN3 (35 mg, 540 µmol) was added in
one portion. The ice bath was removed, and the reaction
mixture was allowed to warm with stirring to 20 °C over a 2
h period. The resulting precipitate was collected, washed with
water (6 × 2 mL), and dried in vacuo to yield a brown powder
(50 mg). A second portion of amine hydrochloride (231 mg,
909 µmol) was treated in a similar manner to yield an
1
36%): mp 176-177 °C; H NMR (CDCl3) δ 2.44 (s, 3 H), 2.72
(s, 3 H), 3.88 (s, 3 H), 6.10 (s, 1 H), 7.30 (s, 1 H), 7.82 (s, 1 H).
5,8-Dih yd r o-2-m eth oxy-5,8-d im eth yln a p h th a len e-1,4-
d ion e (27). Dione 27 was prepared in a manner similar to
that described for 29 with the crude product being purified
chromatographically (silica gel/CHCl3) prior to crystallization
to yield 27 as orange plates (2.22 g, 46%): mp 110-112 °C;
1H NMR (CDCl3) δ 1.19 (s, 3 H), 1.21 (s, 3 H), 3.39-3.42 (m,
2 H), 3.81 (s, 3 H), 5.75-5.85 (m, 2 H), 5.88 (s, 1 H).
2-Meth oxy-5,8-d im eth yln a p h th a len e-1,4-d ion e (31). A
suspension of 27 (500 mg, 2.29 mmol) and DDQ (1.30 g, 5.72
mmol, freshly crystallized from toluene) in toluene (12 mL)
was heated at reflux for 2.6 h. The reaction mixture was
allowed to cool to 25 °C. The suspended material was removed
by filtration and washed with fresh toluene (4 × 3 mL). The
toluene was removed, and the residue was dissolved in CHCl3
(60 mL). This solution was washed with 25% saturated
NaHCO3 (5 × 40 mL) and 50% saturated brine (1 × 40 mL)
and filtered through cotton. The solvent was removed to give
a dark brown solid (442 mg, 89% crude yield). A total of 980
mg of crude material was passed through a silica gel column
(2.5 × 25 cm) with CHCl3 elution to give a green solid (906
mg). Crystallization from EtOH yielded 31 as a greenish-
yellow crystalline solid (612 mg, 62% from crude material):
1
mp 145-146 °C; H NMR (CDCl3) δ 2.71 (s, 6 H), 3.86 (s, 3
H), 6.06 (s, 1 H), 7.35 (d, J ) 8.1 Hz, 1 H), 7.39 (d, J ) 8.1 Hz,
1 H).
2-Br om o-5-m eth oxyh yd r oqu in on e (41). To a stirred,
water bath-cooled (15-20 °C) solution of 2-methoxyhydro-
quinone17 (40; 20.0 g, 143 mmol) in glacial AcOH (400 mL)
was added Br2 (7.50 mL, 23.3 g, 146 mmol) dropwise over 20
min. The color of the reaction mixture darkened during the

Analogs of 3-Hydroxy-1H-1-benzazepine-2,5-dione
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23 4651
addition. TLC (5% EtOAc/95% CHCl3) indicated the formation
of one major higher Rf spot and one minor higher Rf spot
relative to 40. The solvent was removed to give a dark brown
solid. The solid was dissolved in hot CHCl3 (400 mL) and
filtered through a pad of silica gel (8 × 8 cm). The solvent
was removed to give a tan solid. Crystallization from benzene
yielded 41 as brown to purple needles (24.5 g, 78%): mp 127-
(4 × 25 mL) to give a green organic solution. This was washed
with water (3 × 50 mL) and filtered through cotton. The
solvent was removed to give a purple solid (500 mg). The
product was purified by chromatography on silica gel (2.5 ×
25 cm) with CHCl3 elution give an orange solid (367 mg).
Crystallization from EtOH yielded 32 as beige needles (245
mg, 47%). A portion was recrystallized from EtOH to yield
32 as near-colorless needles: mp 188-189 °C; 1H NMR
(DMSO-d6) δ 2.41 (s, 3 H), 3.75 (s, 3 H), 6.26 (s, 1 H), 7.18 (t,
J ) 7.8 Hz, 1 H), 7.46 (d, J ) 7.5 Hz, 1 H), 7.52 (d, J ) 8.1 Hz,
1 H), 9.45 (s, 1 H); MS m/z 217 (M+, 100).
1
128 °C; H NMR (CDCl3) δ 3.86 (s, 3 H), 5.12 (bs, 1 H), 5.23
(bs, 1 H), 6.60 (s, 1 H), 7.00 (s, 1 H).
2-Br om o-5-m eth oxyben zoqu in on e (42). A solution of 41
(4.00 g, 18.3 mmol) in CHCl3 (150 mL, prepared with warming)
was allowed to vigorously stir with a solution of NaIO4 (7.81
g, 36.5 mmol) in water (200 mL) at 25 °C. The reaction
mixture immediately turned dark and then became orange
over a period of time. After 30 min, the layers were separated
and the aqueous portion was extracted with CHCl3 (1 × 30
mL). The combined CHCl3 portion was washed with brine (1
× 30 mL) and filtered through cotton. The solvent was
removed to yield 42 as a yellow solid which was of sufficient
purity for further reaction (4.00 g, 100%): mp 184-186 °C.
Crystallization of a portion from MeOH yielded 42 as orange
needles: mp 186-187 °C (lit.16 mp 190-191 °C); 1H NMR
(CDCl3) δ 3.86 (s, 3 H), 6.12 (s, 1 H), 7.24 (s, 1 H).
3-Meth oxy-6,8-d im eth yl-7-n itr o-1H-1-ben za zep in e-2,5-
d ion e (54). Methyl ether 54 was prepared from 49 (prepared
in situ from 44, 1.00 g, 4.63 mmol) employing methods
described for the preparation of 11 to yield 54 as an orange
1
solid (668 mg, 52%): mp 262-266 °C dec; H NMR (DMSO-
d6) δ 2.17 (3 H), 2.23 (3 H), 3.73 (3 H), 6.38 (s, 1 H), 7.18 (s, 1
H), 11.25 (s, 1 H); 13C NMR (DMSO-d6) δ 14.88, 16.85, 56.35,
113.2, 120.7, 128.6, 128.8, 132.2, 136.0, 148.9, 154.4, 160.1,
189.6. A portion recrystallized from EtOH yielded 54 as a
beige powder: mp 268-271 °C dec.
Gen er a l Meth od A for th e Dem eth yla tion of 3-Meth -
oxy-1H -1-b en za zep in e-2,5-d ion es: Syn t h esis of 3-H y-
d r oxy-7-n itr o-1H-1-ben za zep in e-2,5-d ion e (13). A solution
of 11 (225 mg, 907 µmol) in 80 mL of boiling EtOH was
prepared. The solution was stirred without heating for 1 min;
then a solution of 10% aqueous H2SO4 (10 mL) was added.
The reaction mixture was allowed to stir for 5 min; then
additional acid solution (70 mL) was added. Further stirring
for 10 min yielded a near-colorless precipitate. The reaction
mixture was added to crushed ice, and the ethanol was
removed. The precipitate was collected and washed with water
(5 × 5 mL). The damp filter cake was crystallized from EtOH
to yield 13 as colorless needles (124 mg, 58%): mp 250 °C dec
(without melting); 1H NMR (DMSO-d6) δ 6.45 (s, 1 H), 7.61
(d, J ) 9.0 Hz, 1 H), 8.39 (dd, J ) 9.0, 2.7 Hz, 1 H), 8.79 (d, J
) 2.7 Hz, 1 H), 11.14 (bs, 1 H), 12.04 (s, 1 H). Anal.
(C10H6N2O5) C, H, N.
Gen er a l Met h od for t h e P r ep a r a t ion of 2-Met h oxy-
n a p h th a len e-1,4-d ion es via a Diels-Ald er Rea ction of
42: Syn th esis of 2-Meth oxy-5-m eth yln a p h th a len e-1,4-
d ion e (43). A suspension of 42 (4.0 g, 18 mmol) and trans-
piperylene (22; 5.5 mL, 3.7 g, 47 mmol, technical grade) in
toluene (100 mL) was allowed to stir at 60-70 °C for 4 days.
A homogeneous solution was present. The status of the
1
reaction was determined by H NMR. Since the adduct was
not stable once the reaction solvent was evaporated, an aliquot
was diluted in half with MeOH and treated with an excess of
Et3N. After 1 min, dilute HCl was added and the mixture
extracted with CHCl3. The extract was washed with water
and filtered through cotton. The solvent was removed. The
1H NMR was determined in CDCl3. The absence of a reso-
nance at 7.24 ppm indicated that the starting quinone was
totally consumed. The remaining reaction mixture was al-
lowed to cool to 25 °C and diluted with MeOH (100 mL). The
reaction mixture was purged with N2 for 5 min with stirring,
and Et3N (3.69 g, 36.5 mmol) was added. The reaction mixture
darkened. The reaction mixture was allowed to stir for 5 min.
Dilute HCl (150 mL) was added with stirring to give a lighter
yellow/orange mixture. The layers were separated, and the
aqueous portion was extracted with toluene (1 × 30 mL). The
combined toluene portion was washed with water and brine
(30 mL each) and filtered through cotton. The solvent was
removed to give an orange solid (4.2 g). Crystallization from
EtOH yielded the 5,8-dihydro analog of 43 as an orange
crystalline solid (2.87 g, 77%): mp 116-117 °C (lit.15 mp not
reported); 1H NMR (CDCl3) δ 1.17 (d, J ) 6.9 Hz, 3 H), 2.86-
3.01 (m, 1 H), 3.09-3.24 (m, 1 H), 3.36-3.50 (m, 1 H), 3.80 (s,
3 H), 5.75-5.82 (m, 2 H), 5.88 (s, 1 H).
Gen er a l Meth od B for th e Dem eth yla tion of 3-Meth -
oxy-1H -1-b en za zep in e-2,5-d ion es: Syn t h esis of 3-H y-
d r oxy-9-m eth yl-1H-1-ben za zep in e-2,5-d ion e (36). To a
stirred suspension of 32 (200 mg, 921 µmol) in CH2Cl2 (2 mL)
was added a solution of BBr3 in CH2Cl2 (2 mL, 1 M) in one
portion at 25 °C. The reaction mixture instantaneously
became homogeneous and yellow; then a yellow precipitate
formed after a few seconds. The reaction mixture was allowed
to stir at 25 °C for 45 min. The reaction mixture was added
to saturated NaHCO3 (15 mL), and the resulting beige suspen-
sion was allowed to stir for 15 min to give a cloudy, orange
solution. The solution was filtered. The filtrate was acidified
with concentrated HCl to a pH of 2. The resulting precipitate
was collected by filtration and washed with water (5 × 2 mL).
The damp filter cake was crystallized from EtOH to yield 36
as brown needles (98 mg, 52%): mp 184-186 °C dec; 1H NMR
(DMSO-d6) δ 2.46 (s, 3 H), 6.28 (s, 1 H), 7.19 (t, J ) 7.5 Hz, 1
H), 7.48 (d, J ) 7.2 Hz, 1 H), 7.67 (d, J ) 7.8 Hz, 1 H), 9.94 (s,
1 H), 10.85 (bs, 1 H). Anal. (C11H9NO3‚0.50 H2O) C, H, N.
A solution of the above material (100 mg, 490 µmol) in
CHCl3 (5 mL) was refluxed in the presence of MnO2 (213 mg,
2.45 mmol) for 45 min (the progress of the reaction was
monitored by 1H NMR). The reaction mixture was allowed to
cool to 25 °C and filtered through a pad of Celite. Solvent
removal yielded 43 as a yellow powder (99 mg, 100%): mp
1H-1-Ben za zep in e-2,3,4,5-tetr on e 4-Oxim e (59). A mix-
ture of 3 (188 mg, 1.00 mmol) and NaNO2 (182 mg, 2.64 mmol)
in 0.1 M aqueous NaOH (6 mL) was allowed to stir at room
temperature for 1 h. The resulting mixture was treated with
2 M aqueous H2SO4 (2 mL) and allowed to stir for 3 h at room
temperature. The reaction mixture was treated with a solu-
tion of NaNO2 (181 mg, 2.62 mmol) in water (2 mL) and
allowed to stir overnight at room temperature. The reaction
mixture was treated every 24 h with a solution of NaNO2 (100
mg, 1.45 mmol) in water (0.5 mL) for 7 days. After the final
addition, the reaction mixture was allowed to stir for an
additional 24 h to yield a suspension. The solid was collected
by filtration, washed with water, and dried in vacuo to yield
1
139-140 °C; H NMR (CDCl3) δ 2.76 (s, 3 H), 3.88 (s, 3 H),
6.11 (s, 1 H), 7.50-7.59 (m, 2 H), 8.07 (dd, J ) 6.6, 1.8 Hz, 1
H). A portion crystallized from MeOH yielded 43 as orange
needles: mp 140-141 °C.
Gen er a l Meth od for th e Sch m id t Rea ction of 2-Meth -
oxyn a p h th a len e-1,4-d ion es: Syn th esis of 3-Meth oxy-9-
m et h yl-1H-1-ben za zep in e-2,5-d ion e (32). To stirred, ice
bath cold CF3SO3H (3.0 mL) was added 2-methoxy-8-methyl-
naphthalene-2,5-dione (28; 483 mg, 2.39 mmol) in portions. A
deep red solution resulted. To this cold solution was added
NaN3 (186 mg, 2.87 mmol) in portions. The reaction mixture
was allowed to slowly warm to 25 °C with stirring under N2.
Gas evolution was noted. After 24 h, the reaction was judged
complete by TLC (10% EtOAc/90% CHCl3). The reaction
mixture was added to stirred ice water (30 mL) to give a green
suspension. This was extracted with 30% MeOH/70% CHCl3
1
59 as a pale yellow solid (149 mg, 68%): mp 196-197 °C; H
NMR (DMSO-d6) δ 7.16-7.23 (m, 2 H), 7.55 (t, J ) 7.5 Hz, 1
H), 7.96 (d, J ) 7.8 Hz, 1 H), 10.49 (s, 1 H), 12.00 (s, 1 H).
Anal. (C10H6N2O4‚1.10H2O) C, H, N.

4652 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23
Guzikowski et al.
[3H]MK801 Assa y. This assay was performed with rat
brain cortical membranes employing methods previously
described.21
J .F.W.K.). We thank Dr. J ohn Guastella (CoCensys) for
the preparation of NMDA receptor (NR1a/2C) cRNA,
Sylvia Loyola (University of California, Irvine) for the
[3H]MK801 binding determinations, Prof. Ricardo Mile-
di and Dr. Yan Ni (University of California, Irvine) for
the gift of rat brain poly(A)+ RNA, Prof. Alun H. Rees
(Trent University, Peterborough, Ontario, Canada) for
providing compound 60, Dr. Peter H. Seeburg (Univer-
sity of Heidelberg, Heidelberg, Germany) for the NMDA
receptor clones, and Dr. Timothy Weakley (University
of Oregon) for the X-ray crystallographic determina-
tions.
[3H]DCKA Assa y. This assay was performed as previously
described.22 Briefly, [3H]DCKA (15-20 nM) was incubated
with well-washed rat brain cortical membranes and nine
concentrations (up to 100 µM) of HBAD in 50 mM HEPES-
KOH, pH 7.5, buffer. Following a 30 min incubation on ice,
the assays were terminated by filtration and the filter-bound
radioactivity was determined by liquid scintillation counting.
The IC50 (concentration of test compound producing 50%
inhibition of specific binding) was estimated using the sigmoi-
dal equation (Prism Graph Pad).
Electr op h ysiology a n d Da ta An a lysis. Methods for
RNA preparation, dissection and microinjection of Xenopus
oocytes, two-electrode voltage-clamp recordings, and prepara-
tion of HBAD stock solutions were as described in Woodward
et al.23 Methods for culturing rat cortical neurons and for
whole cell clamp recordings were as described in Huettner and
Baughman.38
In oocyte experiments, concentration-response data for
glycine and AMPA and concentration-inhibition curves for
HBADs were all fit with logistic equations.23 The EC50 and
slope values for glycine at NR1a/2C receptors were 0.174 µM
and 1.45 (n ) 6). The EC50 and slope values for AMPA were
5.89 µM and 1.96 (n ) 5). These values were used in all
subsequent calculations of HBAD antagonist potency. Kb
values were calculated from concentration-inhibition curves
using a generalized form of the Cheng-Prusoff equation:26
Su p p or tin g In for m a tion Ava ila ble: Accounts of struc-
tural analyses and listings of crystallographic properties and
numerical details of refinement, atomic coordinates, bond
lengths and angles, anisotropic thermal parameters, contact
distances, torsion angles, and mean planes, together with
ORTEP diagrams for compounds 11 and 36 and tables of
physical data, NMR spectroscopic data and methods of prepa-
ration for substituted 2-methoxynaphthalene-1,4-diones and
HBAD 3-methyl ethers (30 pages); observed and calculated
structure factors (27 pages). Ordering information is given
on any current masthead page.
Refer en ces
IC50
(1) Watkins, J . C. The NMDA Receptor Concept: Origins and
Development. In The NMDA Receptor, 2nd ed.; Collingridge, G.
L., Watkins, J . C., Eds.; Oxford University Press: Oxford, 1994;
pp 1-30.
Kb )
{2 + ([agonist]f/EC50)n}1/n - 1
where IC50 is the concentration of HBAD that reduces control
response by 50%, [agonist]f is the concentration of agonist used
for the control response, EC50 is the concentration of agonist
(glycine or AMPA) that elicits a half-maximal response, and
n is the slope factor of the agonist concentration-response
relation. Fixed agonist concentrations were 1 µM glycine plus
100 µM glutamate for NR1a/2C receptors. The control mem-
brane current range in these experiments was 71-589 nA, and
the mean response was 140 ( 20 nA (n ) 27). The fixed
agonist concentration was 10 µM AMPA for AMPA receptors.
The current range in these experiments was 99-200 nA, and
the mean response was 150 ( 4 (n ) 28). The holding
potential for all measurements was -70 mV.
(2) (a) McCulloch, J . Excitatory Amino Acid Antagonists and Their
Potential for the Treatment of Ischemic Brain Damage in Man.
Br. J . Clin. Pharmacol. 1992, 34, 106-114. (b) Dodd, P. R.; Scott,
H. L.; Westphalen, R. I. Excitotoxic Mechanisms in the Patho-
genesis of Dementia. Neurochem. Int. 1994, 25, 203-219. (c)
Lipton, S. A. Models of Neuronal Injury in AIDS: Another Role
for the NMDA Receptor? Trends Neurosci. 1992, 15, 75-79.
(3) (a) Choi, D. W.; Rothman, S. M. The Role of Glutamate
Neurotoxicity in Hypoxic-Ischemic Neuronal Death. Annu. Rev.
Neurosci. 1990, 13, 171-182. (b) Meldrum, B.; Garthwaite, J .
Excitatory Amino Acid Neurotoxicity and Neurodegenerative
Disease. Trends Pharmacol. Sci. 1990, 11, 379-387.
(4) (a) J ohnson, J . W.; Ascher, P. Glycine Potentiates the NMDA
Response in Cultured Mouse Brain Neurons. Nature (London)
1987, 325, 529-531. (b) Kleckner, N. W.; Dingledine, R. Re-
quirement for Glycine in Activation of NMDA-Receptors Ex-
pressed in Xenopus Oocytes. Science (Washington, D.C.) 1988,
241, 835-837.
(5) 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.
(6) Keana, J . F. W.; Kher, S. M.; Cai, S. X.; Dinsmore, C. M.; Glenn,
A. G.; Guastella, J .; Huang, J .-C.; Ilyin, V.; Lu¨, Y.; Mouser, P.
L.; Woodward, R. M.; Weber, E. Synthesis and Structure-Activity
Relationships of Substituted 1,4-Dihydroquinoxaline-2,3-di-
ones: Antagonists of N-Methyl-D-Aspartate (NMDA) Receptor
Glycine Sites and of Non-NMDA Glutamate Receptors. J . Med.
Chem. 1995, 38, 4367-4379.
Kb values in the neuronal recordings were determined from
simultaneous fit of concentration-response data under control
conditions and in the presence of a fixed concentration of
HBAD.9 Current ranges for these experiments were 323-1530
pA for glycine/NMDA (100 µM), and 205-1800 pA for kainate.
Mean responses were 729 ( 150 pA (n ) 15) and 534 ( 108
pA (n ) 18), respectively. The holding potential for all
experiments was -70 mV. Statistical conformity to the
competitive model was determined by F-test.9 F values are
given in the form F(df -df ),df , where df1 is the degree of freedom
2
1
in the individual fits, and 1df2 is the degree of freedom in the
simultaneous fit. Other data are given as mean ( SEM.
Mou se MES Stu d ies. General methods for MES studies
were performed as previously reported.29,30 Briefly, seizures
were induced in male Swiss Webster mice (body weight 23-
27 g, housed with ad libitum food and water) via corneal
electrodes (ECT 7801, Ugo Basile). Rectangular pulses (50
mA, 60-75 Hz, 0.8 ms width, 0.2 s train length) were
employed. Seizure occurrence was recorded as a tonic hind-
limb extension after electroshock stimulus. HBADs were
formulated for iv administration as 5 mg/mL solutions in 0.2
M tris(hydroxymethyl)aminomethane. The vehicle alone in-
duced no detectable levels of protection. ED50 values and 95%
confidence intervals were calculated by Litchfield and Wilcoxon
analysis (MicroComputer Specialists, Philadelphia, PA).
(7) Chapdelaine, M. J .; McLaren, C. D. Therapeutic Benzazarine
Compounds. U.S. Patent 5,254,683, Oct. 19, 1993.
(8) (a) Leeson, P. D. Glycine-Site N-Methyl-D-Aspartate Receptor
Antagonists. In Drug Design for Neuroscience; Kozikowski, A.
P., Ed.; Raven Press: New York, 1993; pp 339-381. (b) Leeson
P. D.; Iversen, L. L. The Glycine Site on the NMDA Receptor:
Structure-Activity Relationships and Therapeutic Potential. J .
Med. Chem. 1994, 37, 4053-4067. (c) Iversen, L. L.; Kemp, J .
A. Non-Competitive NMDA Antagonists as Drugs. In The NMDA
Receptor, 2nd ed.; Collingridge, G. L., Watkins, J . C., Eds.;
Oxford University Press: Oxford, 1994; pp 469-486.
(9) 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.
Ack n ow led gm en t. Financial support was provided
by CoCensys, Inc. and by National Institutes of Health
Grants NS 30888 (J .E.H.) and DA 06726 (E.W.,

Analogs of 3-Hydroxy-1H-1-benzazepine-2,5-dione
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 23 4653
(10) Birchall, G. R.; Rees, A. H. Some Derivatives of 1-Benzazepine.
Part II. Can. J . Chem. 1974, 52, 610-615.
(28) 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.
(29) Woodward, R. M.; Huettner, J . E.; Tran, M.; Guastella, J .; Keana,
J . F. W.; Weber, E. Pharmacology of 5-Chloro-7-trifluoro-
methyl-1,4-dihydro-2,3-quinoxalinedione: A Novel Systemically
Active Ionotropic Glutamate Receptor Antagonist. J . Pharmacol.
Exp. Ther. 1995, 275, 1209-1218.
(30) Swinyard, E. A. Assay of Antiepileptic Drug Activity in Experi-
mental Animals: Standard Tests. In Anticonvulsant Drugs.
International Encyclopedia of Pharmacology and Therapeutics;
Mercier, J ., Ed.; Pergamon Press: Oxford, 1972; pp 47-65.
(31) Carling, R. W.; Leeson, P. D.; Moseley, A. M.; Baker, R.; Foster,
A. C.; Grimwood, S.; Kemp, J . A.; Marshall, G. R. 2-Carboxytet-
rahydroquinolines. Conformational and Stereochemical Require-
ments for Antagonism of the Glycine Site on the NMDA
Receptor. J . Med. Chem. 1992, 35, 1942-1953.
(11) Fieser, L. F. Alkylation of Hydroxynaphthoquinones I. Ortho-
Esters. J . Am. Chem. Soc. 1926, 48, 2922-2937.
(12) Bohlmann, F.; Mathar, W.; Schwarz, H. U¨ ber die Regioselek-
tivita¨t von Diensynthesen Substituierter Chinone. (Regioselec-
tivity of diene syntheses of substituted quinones.) Chem. Ber.
1977, 110, 2028-2045.
(13) Adler, E.; Magnusson, R. Periodate Oxidation of Phenols I.
Monoethers of Pyrocatechol and Hydroquinone. Acta Chem.
Scand. 1959, 13, 505-519.
(14) Trave, R.; Garanti, L.; Pavan, M. The Chemical Constitution of
the Venom of the Myriapod Archiulus Sabulosus. Chim. Ind.
(Milan) 1959, 41, 19-29.
(15) Tegmo-Larsson, I.-M.; Rozeboom, M. D.; Houk, K. N. Regiose-
lectivities of Diels-Alder Cycloadditions to Methoxy-Substituted
Quinones. Tetrahedron Lett. 1981, 22, 2043-2046.
(16) Blatchly, J . M.; Green, R. J . S.; McOmie, J . F. W.; Searle, J . B.
Thiele Acetylation of Quinones. Part III. p-Benzoquinones with
Bromo- and Methoxy-Substituents. J . Chem. Soc. C 1969, 1353-
1358.
(17) Logan, C. D.; Husband, R. M.; Purves, C. B. Studies in the
Polyoxyphenol Series VIII. The Oxidation of Substances Related
to Vanillin with Sodium Chlorite and Chlorine Dioxide. Can. J .
Chem. 1955, 33, 82-96.
(32) In
a recent report, we demonstrated receptor tolerance to
nonplanarity in the size-limited hydrophobic region for the
HBAD series: Guzikowski, A. P.; Hawkinson, J .; Weber, E.;
Keana, J . F. W. 6,7,8,9-Tetrahydro-3-hydroxy-1H-1-benzazepine-
2,5-diones Via a Diels-Alder Reaction: Antagonists with a Non-
Planar Hydrophobic Region for NMDA Receptor Glycine Sites.
Bioorg. Med. Chem. Lett. 1995, 5, 2747-2748.
(18) Moore, H. W.; Shelden, H. R.; Weyler, W., J r. Rearrangement
of Azidoquinones IV. Ring Expansion of 4-Azido-1,2-naphtho-
quinone. Tetrahedron Lett. 1969, 16, 1243-1246.
(19) Moriconi, E. J .; Maniscalco, I. A. π-Equivalent Heterocyclic
Congeners of Tropone. Azatropones. J . Org. Chem. 1972, 37,
208-215.
(33) The preparation of 4-aryl derivatives of 6 for use as antagonists
at NMDA receptor glycine sites has been described. However,
no direct comparison was made between the binding affinity of
6 and the binding affinities of the newly prepared compounds.
J ackson, P. F.; Davenport, T. W.; Garcia, L.; McKinney, J . A.;
Melville, M. G.; Harris, G. G.; Chapdelaine, M. J .; Damewood,
J . R.; Pullan, L. M.; Goldstein, J . M. Synthesis and Biological
Activity of a Series of 4-Aryl Substituted Benz[b]azepines:
Antagonists at the Strychnine-Insensitive Glycine Site. Bioorg.
Med. Chem. Lett. 1995, 5, 3097-3100.
(34) Such HBAD methyl ethers are reported to be neuroprotective
in a gerbil ischemia model. It is hypothesized that such ethers
act as prodrugs for the active HBADs; see ref 7.
(35) 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.
(36) 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.
1991, 34, 1283-1292.
(37) Ilyin, V. I.; Whittenmore, E. R.; Tran, M.; Shen, K.-Z.; Cai, S.
X.; Kher, S. M.; Keana, J . F. W.; Weber, E.; Woodward, R. M.
Pharmacology of Acea-1416: A Potent Systemically Active
N-Methyl-D-Aspartate Receptor Glycine Site Antagonist. Eur.
J . Pharmacol., in press.
(20) Hughes, C. G.; Lewars, E. G.; Rees, A. H. Azatropolones. Part
II. The Schmidt Reaction of 2-Methoxy-5-methylbenzoquinone.
Can. J . Chem. 1974, 52, 3327-3330.
(21) Kaneko, S.; Sugimura, M.; Inoue, T.; Satoh, M. Effects of Several
Cerebroprotective Drugs on NMDA Channel Function: Evalu-
ation Using Xenopus Oocytes and [3H]MK-801 Binding. Eur. J .
Pharmacol., Mol. Pharmacol. Sect. 1991, 207, 119-128.
(22) Canton, T.; Doble, A.; Miquet, J . M.; J imonet, P.; Blanchard, J .
C. A Rapid Filtration Assay for the Glycine Binding Site on the
NMDA Receptor in Rat Cortical Membranes using [3H]Dichlo-
rokynurenic Acid. J . Pharm. Pharmacol. 1992, 44, 812-816.
(23) Woodward, R. M.; Huettner, J . E.; Guastella, J .; Keana, J . F.
W.; Weber, E. In Vitro Pharmacology of ACEA-1021 and
ACEA-1031: Systemically Active Quinoxalinediones with High
Affinity and Selectivity for N-Methyl-D-Aspartate Receptor
Glycine Sites. Mol. Pharmacol. 1995, 47, 568-581.
(24) Monyer, H.; Sprengel, R.; Schoepfer, R.; Herb, A.; Higuchi, M.;
Lomeli, H.; Burnashev, N.; Sakmann, B.; Seeburg, P. H. Het-
eromeric NMDA Receptors: Molecular and Functional Distinc-
tion of Subtypes. Science (Washington, D.C.) 1992, 256, 1217-
1221.
(25) Sugihara, H.; Moriyoshi, K.; Ishii, T.; Masu, M.; Nakanishi, S.
Structures and Properties of Seven Isoforms of the NMDA
Receptor Generated by Alternative Splicing. Biochem. Biophys.
Res. Commun. 1992, 185, 826-832.
(26) Leff, P.; Dougall, I. G. Further Concerns Over Cheng-Prusoff
Analysis. Trends Pharmacol. Sci. 1993, 14, 110-112.
(27) Cai, S. X.; Zhou, Z.-L.; Huang, J .-C.; Whittenmore, E. R.;
Egbuwoku, Z. O.; Lu¨, Y.; Hawkinson, J . E.; Woodward, R. M.;
Weber, E.; Keana, J . F. W. Synthesis and Structure-Activity
Relationships of 1,2,3,4-Tetrahydroquinoline-2,3,4-trione-3-
oximes: Novel and Highly Potent Antagonists for NMDA
Receptor Glycine Sites. J . Med. Chem. 1996, 39, 3248-3255.
(38) Huettner, J . E.; Baughman, R. W. Primary Culture of Identified
Neurons from the Visual Cortex of Postnatal Rats. J . Neurosci.
1986, 6, 3044-3060.
J M960479Z