Structure-activity studies of 6-substituted decahydroisoquinoline-3- carboxylic acid AMPA receptor antagonists. 2. Effects of distal acid bioisosteric substitution, absolute stereochemical preferences, and in vivo activity

Article abstract of DOI:10.1021/jm950913h

We have explored the excitatory amino acid antagonist activity in a series of decahydroisoquinoline-3-carboxyic acids, and within this series found the potent and selective AMPA antagonist (3SR,4aRS,6RS,8aRS)-6-(2-(1H-tetrazol- 5-yl)ethyl)decahydroisoquin

Full text of DOI:10.1021/jm950913h

2
232
J . Med. Chem. 1996, 39, 2232-2244
Str u ctu r e-Activity Stu d ies of 6-Su bstitu ted Deca h yd r oisoqu in olin e-3-ca r boxylic
Acid AMP A Recep tor An ta gon ists. 2. Effects of Dista l Acid Bioisoster ic
Su bstitu tion , Absolu te Ster eoch em ica l P r efer en ces, a n d in Vivo Activity
,
†
†
†
†
†
Paul L. Ornstein,* M. Brian Arnold, Nancy K. Allen, Thomas Bleisch, Peter S. Borromeo,
†
†
‡
†
Charles W. Lugar, J . David Leander, David Lodge, and Darryle D. Schoepp
Lilly Research Laboratories, A Division of Eli Lilly and Company, Lilly Corporate Center, Indianapolis, Indiana 46285, and
Lilly Research Centre, Limited, Windlesham, Surrey GU20 6PH, England
X
Received December 12, 1995
We have explored the excitatory amino acid antagonist activity in a series of decahydroiso-
quinoline-3-carboxyic acids, and within this series found the potent and selective AMPA
antagonist (3SR,4aRS,6RS,8aRS)-6-(2-(1H-tetrazol-5-yl)ethyl)decahydroisoquinoline-3-carboxy-
lic acid (1). In this and the preceding paper, we looked at the structure-activity relationships
for AMPA antagonist activity in this series of compounds. We have already shown that 1 had
the optimal stereochemical array and that AMPA antagonist activity was maximized for a
two-carbon spacer separating a tetrazole from the bicyclic nucleus. In this paper, we explored
the effects of varying the distal acid and the absolute stereochemical preferences of many of
these analogs. We looked at a variety of different acid bioisosteres, including 5-membered
hetereocyclic acids such as tetrazole, 1,2,4-triazole, and 3-isoxazolone; carboxylic, phosphonic,
and sulfonic acid; and acyl sulfonamides. Compounds were evaluated in rat cortical tissue for
3
their ability to inhibit the binding of radioligands selective for AMPA ([ H]AMPA), NMDA
3
3
(
[ H]CGS 19755), and kainic acid ([ H]kainic acid) receptors and for their ability to inhibit
depolarizations induced by AMPA (40 µM), NMDA (40 µM), and kainic acid (10 µM). A number
of compounds from this and the preceding paper were also evaluated in mice for their ability
to block maximal electroshock-induced convulsions and ATPA-induced rigidity in mice.
In tr od u ction
a methyl or a phenyl group) was tolerated, but did not
enhance affinity. In this paper, we report on the effects
of varying the distal acid bioisostere from tetrazole to
other 5-membered heterocyclic ring acids and phospho-
nic, sulfonic, and carboxylic acid. We also look at the
absolute stereochemical requirements for AMPA an-
tagonist activity in this series. A number of the more
potent compounds from this SAR (from both papers)
were evaluated in some behavioral assays in mice.
Our medicinal chemistry research efforts are targeted
toward the identification of compounds which are potent
and selective antagonists of excitatory amino acid (EAA)
1
receptors, with the goal of developing these compounds
as novel therapeutic agents for the treatment of a
variety of central nervous system disorders. Within a
series of decahydroisoquinoline-3-carboxylic acids, we
found that the 6-tetrazolylethyl-substituted compound
2
1
is a selective antagonist of neurotransmission medi-
Ch em istr y
ated through the 2-amino-3-(5-methyl-3-hydroxyisox-
azol-4-yl)propanoic acid (AMPA) subclass of excitatory
amino acid (EAA) receptors.3 Structure-activity stud-
ies (SAR) on this class of EAA antagonists have revealed
features which enhance potency and affinity of these
In the previous study, we established that the ster-
eochemical array for the decahydroisoquinoline nucleus
corresponding to 1 was optimal for activity; it was also
best to have the distal acid bioisostere (e.g., tetrazole)
attached to the bicyclic nucleus through a two-atom
spacer. Holding these values constant, we focused in
,4
5
,6
compounds for either NMDA or AMPA receptors. In
the preceding paper, we reported data on the effect of
7
varying stereochemistry in the decahydroisoquinoline
ring nucleus, varying the length of the tether connecting
a tetrazole to the decahydroisoquinoline and substitu-
tion with a nitrogen or oxygen atom on this tether in
the postion adjacent to the bicyclic ring, and the effect
of substitution with a methyl or phenyl group on the
tether. What we found was that activity was optimized
for the stereoisomer corresponding to 1 and that a two-
atom all-carbon chain was optimal for AMPA antagonist
activity. Substitution on the connecting chain (e.g., with
this study on compounds with a variety of different acid
isosteres in the distal acid position and on the resolution
of a number of analogs to identify the optimal absolute
stereochemistry for AMPA antagonist activity. Even
though the C-6 stereochemistry of tetrazole-substituted
compounds was optimal as in 1, a number of these
compounds with different bioisosteric substitutions were
prepared in both C-6 epimeric forms. We thus allowed
*
Address correspondence to: this author at: Lilly Research Labo-
ratories, Lilly Corporate Center DC 0510, Indianapolis, IN 46285.
Phone: 317-276-3226. FAX: 317-277-1125. E-mail: Ornstein_Paul@
Lilly.Com.
†
Lilly Research Laboratories.
Lilly Research Centre, Limited.
Abstract published in Advance ACS Abstracts, May 1, 1996.
‡
X
S0022-2623(95)00913-7 CCC: $12.00 © 1996 American Chemical Society

Hydroisoquinoline AMPA Antagonist SAR. 2
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2233
Sch em e 1a
a
(a) Et2O3PCH2CO2Et, NaH, THF, 0 °C; H2, 5% Pd/C, EtOH, 60 psi, room temperature; (b) 6N HCl, reflux; Dowex 50-X8, 10% pyridine/
water; (c) CH3SO2NH2 (for 8a ) or C6H5SO2NH2 (for 8b), DCC, DBU, THF, reflux; (d) 1 N NaOH, EtOH, room temperature; Me3SiI,
CHCl3, reflux; Dowex 50-X8, 10% pyridine/water; (e) (CHN4)NH2, DCC, THF, reflux.
for any unexpected changes in recognition of these
ligands by the receptor protein that might occur from a
change in the distal acid.
for 3 and 4 to the unsaturated benzyl ester 13, which
upon hydrogenation gave the acid 14 as a 2:1 mixture
of diastereomers (ratio by H NMR integration of the
1
The ω-carboxy analogues (Scheme 1) were prepared
from the aldehyde 2, obtained as a 2:1 mixture of
diastereomers from acidic hydrolysis of the correspond-
C-3 proton). Selective reduction of the acid with borane
methyl sulfide afforded a mixture of alcohols that were
subsequently reacted with triphenylphosphine dibro-
mide to give bromides 15 and 16 (chromatographically
6
ing enol ether. Horner-Emmons condensation with
the sodium salt of triethyl phosphonoacetate followed
by catalytic hydrogenation of the resulting enoate
afforded a seperable mixture of diesters 3 and 4,
epimeric at C-6. Exhaustive hydrolysis of 3 then af-
forded amino acid 5, while hydrolysis of 4 afforded 6,
respectively. The ω-acyl sulfonamide analogs were
prepared by a similar route; however, the sodium salt
of benzyl (diethylphosphono)acetate was used in the
Horner-Emmons condensation; selective deprotection
of the ω-acid could then be achieved during catalytic
separable; 16, the higher R C-6 epimer is the minor
f
product; 15, the lower Rf C-6 epimer is the major
product). Arbuzov reaction of 15 with triethyl phosphite
afforded phosphonate 17, which upon exhaustive hy-
rolysis afforded 18. The same sequence of reactions
applied to bromide 16 afforded 19 and ultimately 20.
Reaction of either bromide 15 or 16 with sodium sulfite
in aqueous ethanol followed by exhaustive hydrolysis
afforded the ω-sulfonic acid derivatives 21 or 22,
respectively.
7
hydrogenation of the olefin. The acid 7 was reacted
We also prepared a series of thiotetrazoles, where the
thio group was incorporated either as a constituent atom
in the two-atom connecting chain or as a substituent
on the tetrazole ring (Scheme 3). Reaction of bromides
with methane- or benzenesulfonamide in the presence
of DCC to afford the desired acyl sulfonamides 8a (R )
methyl) or 8b (R ) phenyl). Basic hydrolysis of the
ester followed by treatment with excess iodotrimethyl-
silane cleaved the carbamate under conditions that
precluded hydrolysis of the acyl sulfonamide, leading
to 9a (R ) methyl) and 9b (R ) phenyl). The acid 7
was also condensed with 5-aminotetrazole to afford
amide 10 which was deprotected as for the acyl sul-
fonamides to afford 11.
9
9
2
3 and 24 with 5-thiotetrazole and triethylamine in
acetonitrile affored tetrazoles 25 and 26, respectively.
Only S-alkylation was seen in this reaction. Exhaustive
hydrolysis of 25 and 26 afforded the desired amino acids
27 and 28, respectively. Alternatively, bromide 23 was
reacted with 5-(S-phenacylthio)tetrazole (29) to give a
separable mixture of the N-1 and N-2 alkylated products
30 and 31, respectively. After photolytic removal of the
phenacyl group, either 30 or 31 was exhaustively
The phosphono and sulfono amino acids were pre-
pared through reaction of a bromoethyl intermediate
8
such as 15 (Scheme 2). Ketone 12 was converted as

2
234 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Ornstein et al.
Sch em e 2a
a
(
a) Bn2O3PCH2CO2Et, NaH, THF, room temperature; (b) H2, 5% Pd/C, EtOAc, 60 psi, room temperature; (c) BH3‚SMe2, THF, 0 °C to
room temperature; Ph3P, Br2, pyridine, CH2Cl2, 0 °C; (d) (EtO)3P, 150 °C; (e) 6 N HCl, reflux; propylene oxide, EtOH; (f) Na2SO3, EtOH,
H2O, reflux; 6 N HCl, reflux; Bio-Rad AG1-X8, 3 N HOAc.
hydrolyzed to afford the desired ω-tetrazolethiols 32 or
bromide in 39 with methoxide did not occur. Sodium
borohydride reduction of the mixed anhydride formed
from 41 and isobutyl chloroformate gave the desired 42.
Bromination of 42 gave 43, which upon displacement
with triethyl phosphite afforded 44. Condensation of
the sodium salt of 44 with the above-mentioned 2:1
mixture of aldehydes 2 followed by catalytic hydrogena-
tion of the olefin gave a separable mixture of C-6
epimers 45 and 46. Exhaustive hydrolysis of 45 then
afforded amino acid 47, while hydrolysis of epimer 46
afforded amino acid 48.
3
3, respectively.
As for the above tetrazoles, we prepared thiotriazole-
substituted amino acids where the sulfur atom was
incorporated as a constituent of the two-atom connecting
chain (Scheme 4). Reaction of bromide 23 with 3-thio-
1
,2,4-triazole afforded 34, which was then exhaustively
hydrolyzed to the sulfide 35. Alternatively, 34 was first
oxidized to the sulfone with m-CPBA and then exhaus-
tively hydrolyzed to afford amino acid 36. All attempts
to oxidize the sulfide of 25 to the corresponding sulfone
were unsuccessful. Either no reaction occurred, or when
starting material was consumed, no product was ob-
tained that could be characterized as the sulfone.
Scheme 4 also shows the preparation of an amino-
tetrzole compound, whereas for the thiotetrazoles and
triazoles, the amine is part of the two-atom connecting
chain. Alkylation of 23 with 5-aminotetrzole afforded
For the preparation of the above compounds in
nonracemic form, the chemistry thus described was
carried out using the 3S,4aR,8aR- or 3R,4aS,8aS-
8
isomers of ketone 12. Table 1 shows analytical data
and melting points for the novel intermediates and
products from this SAR. Also given in Table 1 is the
optical rotations ([R]D) for nonracemic amino acids.
3
3
7, which after exhaustive hydrolysis yielded amino acid
8.
Resu lts a n d Discu ssion
In order to prepare amino acids substituted with an
isoxazole, as is found in AMPA, we required the novel
All of the compounds that we prepared were evalu-
ated in selective radioligand binding assays for affinity
3
isoxazole substituted Horner-Emmons reagent 44
at NMDA, AMPA, and kainic acid receptors using [ H]-
1
2
3
13
3
14
(
Scheme 5). Conversion of 3-bromo-5-(hydroxymethyl)-
CGS 19755, [ H]AMPA, and [ H]kainic acid, re-
spectively; and for functional activity at NMDA, AMPA,
and kainic acid receptors using a cortical slice prepara-
tion (cortical wedge) by measuring their ability to inhibit
depolarizations induced by 40 µM NMDA, 40 µM AMPA
(or in a few cases, 40 µM quisqualic acid (QUIS)), and
1
0
isoxazole (39, which was contaminated with about 10%
of 3-bromo-4-(hydroxymethyl)isoxazole; this was re-
moved upon purification of 42) to the corresponding
3
-methoxy compound 42 was achieved only when the
1
1
alcohol was first oxidzed to the corresponding acid 40.
1
5
Without oxidation to the acid, displacement of the
10 µM kainic acid, respectively. While the data is not

Hydroisoquinoline AMPA Antagonist SAR. 2
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2235
Sch em e 3a
a
(
a) (CHN4)SH, Et3N, CH3CN; (b) 6N HCl, reflux; Dowex 50-X8, 10% pyridine/water (for 28 only); (c) K2CO3, DMF, 90 °C; EtOH, hν,
ambient temperature; (d) 6 N HCl, reflux.
Sch em e 4a
a
(a) (C2H2N3)SH, DMF, 100 °C; (b) 6 N HCl, reflux; Dowex 50-X8, 10% pyridine/water; (c) m-CPBA, CH2Cl2, room temperature; 6 N
HCl, reflux; Dowex 50-X8, 10% pyridine/water; (d) K2CO3, (CHN4)NH2, CH3CN, reflux; 6 N HCl, reflux; Dowex 50-X8, 10% pyridine/
water.
shown, none of the compounds showed significant
agonist activity when tested alone in the cortical slice
preparation. The data for these novel compounds are
shown in Tables 2 and 3, with data for the AMPA
electroshock-induced convulsions in mice;17 to block the
characteristic rigidity induced in mice by a 30 mg/kg
(iv) dose of the systemically active AMPA agonist
2-amino-3-(3-hydroxy-5-tert-butylisoxazol-4-yl)pro-
3
,4
3
18
antagonist 1 and its corresponding C-6 epimer 49
panoic acid (ATPA); and to produce neurological
1
7
included for comparison. Selected compounds were
evaluated in mice (Table 4) for NMDA and AMPA
antagonist activity by examining their ability to protect
mice from lethality induced by a 200 mg/kg (ip) dose of
NMDA (data not shown in Table 4);16 to block maximal
impairment in a horizontal screen assay. Compounds
were administered intraperitoneally (ip) in mice.
Having previously established the optimal stereo-
chemistry and length of tether that joins the requisite
distal acidic moiety to the bicyclic ring nucleus, we next

2
236 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Ornstein et al.
Sch em e 5a
a
CrO3, H2SO4, H2O; (b) KOH, MeOH, H2O, reflux; (c) ClCO2-i-Bu, Et3N, THF, 0 °C; NaBH4, H2O, room temperature; (d) Ph3P, Br2,
pyridine, CH2Cl2, 0 °C; (e) (EtO)3P, toluene, 120 °C; (f) 44, NaN(SiMe3)2, THF, -17 °C, then 2, room temperature; H2, 5% Pd/C, EtOAc,
6
0 psi, room temperature; (g) 48% aqueous HBr, reflux; Dowex 50-X8, 10% pyridine/water.
Ta ble 1. Analytical Data, Melting Points, and Optical Rotations for Novel Compounds
compd
formula
analysis
mp (°C)
oil
190-191
270-271
oil
208-215
oil
215-218
oil
223-225
oil
oil
oil
oil
oil
157-160
>265
>265
oil
[R]Da
4
5
6
C19H31NO6
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C13H21NO4‚H2O
C13H21NO4
C18H30N2O7S
C14H24N2O5S
C23H32N2O7S
8
9
8
9
1
1
1
1
1
1
1
2
2
2
2
2
a
a
b
b
0
1
3
5
6
8
9
0
1
2
5
7
C19H26N2O5S‚H2O
C18H28N6O5
C14H22N6O3‚2.5H2O
C23H29NO6‚0.05CHCl3
C16H26BrNO4
C16H26BrNO4
C12H22NO5P‚0.75H2O
C20H36NO7P
C12H22NO5P‚1.25H2O
C12H21NO5S‚0.25H2O
C12H21NO5S‚0.75H2O
C16H25N5O4S‚0.25C2H4O2
C12H19N5O2S‚HCl
C12H19N5O2S
C12H19N5O2S‚0.5H2O‚0.1C5H5N
C16H25N5O4S‚0.1C2H4O2‚0.1CH2N4S
C12H19N5O2S
C12H19N5O2S‚0.25H2O
C12H19N5O2S‚0.25H2O
C16H25N5O4S‚0.5H2O
C16H25N5O4S
C12H19N5O2S‚HCl
C12H19N5O2S‚HCl
C13H20N4O2S‚2.0H2O
C13H20N4O4S‚H2O
C13H20N4O4S‚0.6H2O
C13H20N4O4S‚0.75H2O
C16H26N6O4‚0.1C4H8O‚0.2C6H14
C12H20N6O2‚1.5H2O
C5H6NO2Br
285
(
-)-27
199-207
165-172
oil
257
246
275-276
oil
oil
279-280
250-251
156-160
215-221
286-287
267-270
oil
197
oil
oil
oil
-53.2
+49.3
(+)-27
2
2
6
8
(
(
+)-28
+34.6
-35.6
-)-28
b
3
3
3
3
3
3
0
1
2
3
5
6
C,H; N
C,H; Nc
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
(
(
-)-36
-39.4
+33.8
+)-36
3
3
4
4
4
4
4
4
7
8
3
4
5
6
7
8
C9H16NO5P‚0.5H2O
C20H36N2O6‚0.25H2O
C20H36N2O6
C15H22N2O4‚1.0H2O
C15H22N2O4‚2.0H2O
C13H21N5O2‚0.75H2O
C13H21N5O2‚0.5H2O
oil
256
244
201-209
250-252
(
+)-49
+20.4
-22.0
(-)-49
a
All optical rotations were determined in 1 N HCl at a concentration, c ) 1. b Anal. C, H. N; calcd, 17.84; found, 17.07. c Anal. C, H.
N: calcd, 18.26; found, 19.07.

Hydroisoquinoline AMPA Antagonist SAR. 2
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2237
Ta ble 2. Effects of Heteroatoms Substitution Adjacent to the Tetrazole in the Two-Atom Linker of 6-Substituted
Decahydroisoquinoline-3-carboxylic Acids and Resolution of Selected Compounds
IC50 (µM) versus radioligand binding
IC50 (µM) versus agonist-induced
depolarizations in a cortical slice preparation
NMDA AMPA kainic acid
a,b
c
at ionotropic excitatory amino acid receptors
C-6
compound
Y
epimer
NMDA
AMPA
kainic acid
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
()-1d
CH2
A
A
A
B
B
B
A
A
A
B
B
B
A
A
A
26.4 ( 2.0
12.1 ( 2.0
>100
4.8 ( 1.2
247 ( 8
28.1 ( 1.7
>100
180 ( 22
87 ( 5
>100
61 ( 3 6.0 ( 1.0
31.7 ( 4.4
60% at 100 µM
>100
d
60% at 100 µMe 1.8 ( 0.2
e
-)-3S,4aR,6R,8aR-1 CH2
14. ( 0.1
>100
d
+)-3R,4aS,6S,8aS-1
()-49
CH2
CH2
>100
>100
60.6 ( 24.8
59.6 ( 4.3
31.5 ( 7.3
>100
0.9 ( 0.1
0.6 ( 0.03
22.1 ( 1.8
4.9 ( 0.4
17.8g
>100
<100f
>100
g
+)-3S,4aR,6S,8aR-49 CH2
-)-3S,4aR,6S,8aR-49 CH2
99.2
76.9
>100
21.9 ( 3.3
>100
28.9 ( 9.2
>100
g
>100
()-27
S
S
S
S
S
S
29.4 ( 6.8
18.1 ( 1.7
>10
30.1 ( 1.6
15.7 ( 2.1
>100
>100
19.8 ( 2.2
1.4 ( 0.4
19.8 ( 2.2
6.1 ( 1.5
9.0 ( 0.8
>100
>31.6
-)-3S,4aR,6S,8aR-27
+)-3R,4aS,6R,8aS-27
()-28
22.9 ( 3.8
>100
>31.6
>31.6
47.9 ( 3.9
>100
37.3 ( 2.3
>100
>100
+)-3S,4aR,6R,8aR-28
-)-3R,4aS,6S,8aR-28
()-32
()-33
()-38
27.9g
>100
29.1 ( 4.3
>100
29.8 ( 7.1
>10
70.5 ( 8.5
>100
62.2 ( 7.6
g
16.7g
12.9
agonist
>10
>100
57.9 at 100 µMh
>100
>100
>100
>100
>100
55% at 100 µMh >100
NH
>100
>100
a
3
Affinity at NMDA receptors was determined using [ H]CGS 19755; see ref 12. Affinity at AMPA receptors was determined using
[3
H]AMPA; see ref 13. Affinity at kainic acid receptors was determined using [ H]kainic acid; see ref 14. All assays for affinity were run
3
b
c
d
e
in triplicate, unless otherwise indicated. See ref 15. Data from refs 3 and 4. Percent inhibition of agonist-induced depolarizations at
the concentration shown. f Tested versus quisqualic acid instead of AMPA. g IC50 was the result of a single assay. h Percent inhibition of
radioligand binding at the concentration shown.
examined effects of bioisosteric replacement of the
tetrazole ring and substitution with a heteroatom in the
connecting chain adjacent to the acidic moiety. Table
We prepared compounds in which the tetrazole was
replaced with either acidic functional groups (e.g.,
carboxylate, phosphonate, and sulfonate) or an acidic
5-membered ring heterocyclic acid. For most com-
pounds, we prepared both of the C-6 epimers. The
carboxylic acid analog (5) corresponding to 1 was 5-fold
less potent in affinity at AMPA sites and 7-fold less
potent as an AMPA antagonist. The C-6 epimer 6 was
inactive. Converison of 5 to the corresponding acyl-
methane- and benzenesulfonamides 9a and 9b, respec-
tively, afforded compounds which showed low or no
AMPA receptor affinity or antagonist activity. The
acylamidotetrazole 11, while being 12-fold less potent
than 1 in terms of affinity at AMPA receptors, nonethe-
less retained some functional AMPA antagonist activity,
with an IC50 of 19 µM in the cortical slice preparation.
Both C-6 epimers 18 and 20, which have a phosphonic
acid replacing the tetrazole, were inactive. However,
the sulfonic acid-substituted compound 21 was only
2-fold less potent in affinity at AMPA sites relative to
1 and 5-fold less potent as an antagonist in the cortical
slice. The corresponding C-6 epimeric sulfonic acid 22
was inactive.
2
shows the effects of substitution of a sulfur or nitrogen
atom in the position adjacent to the tetrazole. Com-
pounds 27 and 28 are epimeric with each other at C-6,
as in 1 and 49. Each contains a sulfur atom adjacent
to the tetrazole ring in the connecting chain and shows
a 5- and 12-fold increase in affinity for AMPA receptors
over their all-carbon counterparts, 1 and 49, respec-
tively. No change in affinity for NMDA receptors is
observed so that this sulfur atom imparts slightly higher
selectivity for AMPA over NMDA. This increase in
affinity for AMPA receptors is also manifest in an
increase in AMPA antagonist activity in the cortical slice
preparation, with a more robust increase in antagonist
activity for the C-6 epimeric thiotetrazole 28. The
corresponding compound, 38, with a nitrogen atom
adjacent to the tetrazole in the two-atom tether, is
devoid of affinity or activity at ionotropic EAA receptors.
The exact role that sulfur plays in this increase in
affinity is not clear. One might envision conformational
and electronic effects from this change. If the thiotet-
razole moiety is bound to the molecule in a way that
maintains the atom count between the acidic site and
the bicylic nucleus, as in 32 and 33, we saw much lower
affinity for AMPA receptors, and while 33 was a weak
antagonist, 32 actually showed some agonist activity in
the cortical slice preparation.
Amino acid 47, in which the tetrazole of 1 was
replaced with the 3-isoxazolone moiety found in AMPA,
was 2-fold less potent in binding to AMPA recepetors,
and 3-fold less potent as an AMPA antagonist in the
cortical slice preparation. The corresponding C-6 epimer,
48, was inactive. We also prepared some triazole-

2
238 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Ornstein et al.
Ta ble 3. Effects of Bioisosteric Substitution for a Series of 6-Substituted Decahydroisoquinoline-3-carboxylic Acids with a Two-Atom
Connecting Chain and Resolution of Selected Compounds
IC50 (µM) versus radioligand binding at
ionotropic excitatory amino acid receptors
IC50 (µM) versus agonist-iinduced
depolarizations in a cortical slice preparation
a,b
c
linker
C-6
compound
Y-Z
epimer
NMDA
AMPA
kainic acid
NMDA
AMPA
kainic acid
X ) CO2H
5
CH2CH2
CH2CH2
A
B
43.9 ( 4.1
27.8 ( 0.6
>100
>100
57 ( 4.4
42 ( 6.5d
>100
>100
6
>100
>100
>100
>100
X ) methanesulfonamide
8
a
CH2CH2
CH2CH2
CH2CH2
A
A
A
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
X ) benzenesulfonamide
8
b
>100
X ) amidotetrazole
1
1
72.6 ( 13.6
84 ( 19
19 ( 6
71 ( 28
X ) PO3H2
1
2
8
0
CH2CH2
CH2CH2
A
B
>10
>10
>100
>10
NTe
NTe
>100
>100
>100
>100
>100
>100
X ) SO3H
2
2
1
2
CH2CH2
CH2CH2
A
B
61.8f
>100
11.6 ( 1.5
>100
>100
>100
>100
>100
31.4 ( 1.7
>100
100
>100
X ) 3-isoxazolone
4
4
6
7
CH2CH2
CH2CH2
A
B
>100
>100
13.8 ( 1.2
>100
>100
>100
>100
16.5 ( 2.7
>100
X ) 1,2,4-triazole
3
3
5
6
SCH2
SO2CH2
SO2CH2
SO2CH2
A
A
A
A
>10
>100
>100
>10
>10
>100
>100
>100
>100
>100e
NT
68.8 ( 12.3
2e.16 ( 0.79
>100
g
1.7 ( 0.1
10
f
e
(
-)-3S,4aR,6S,8aR-36
0.6
NT
NTe
NT
NT
60.9f
23.6f
NTe
e
(+)-3R,4aS,6R,8aS-36
a
3
Affinity at NMDA receptors was determined using [ H]CGS 19755; see ref 12. Affinity at AMPA receptors was determined using
[3
H]AMPA; see ref 13. Affinity at kainic acid receptors was determined using [ H]kainic acid; see ref 14. All assays for affinity were run
3
b
in triplicate, unless otherwise indicated. See ref 15. Tested versus quisqualic acid instead of AMPA. e NT ) not tested. f IC50 was the
c
d
g
result of a single assay. Approximately 50% inhibition of kainic acid-induced depolarization at 10 µM.
substituted amino acids, which incorporated a sulfide
or sulfone in the two-atom connecting chain adjacent
to the triazole ring. The sulfide 35 was inactive;
however, the sulfonyltriazole 36 was 3-fold more potent
in both affinity and antagonist activity when compared
to 1. Both the 3-isoxazolone-substituted amino acid 47
and the sulfonyltriazole 36 were more selectve for
AMPA over NMDA and kainic acid receptors than the
tetrazole 1.
The tetrazole-, thiotetrazole-, and sulfonyltriazole-
substituted compounds 1, 49, 27, 28, and 36 were
prepared in nonracemic form, and each enantiomer was
tested for affinity and antagonist activity at AMPA,
NMDA, and kainic acid receptors (Tables 2 and 3). We
found that for both C-6 epimers it was the compound
having the S-absolute stereochemistry at the amino acid
carbon (C-3) that had affinity at the AMPA receptor and
carried the functional AMPA antagonist activity. This
is distinct from what we found in an earlier structure-
activity study of hydroisoquinoline amino acids which
yielded NMDA antagonists.6 For example, we found
NMDA antagonist activity in the 3S-isomer of the
tetrazole-substituted compound 50, but for the corre-
sponding C-6 epimer 51, NMDA antagonist activity was
found in the 3R-isomer. Thus, how one might envision
6
an overlap of the two C-6 epimers 1 and 49, which are
AMPA antagonists, or 50 and 51, which are NMDA
antagonists, is distinctly different; we have attempted
to show these overlaps in Figure 1. Very different
pictures emerge of the pharmacophore for 6-substituted
hydroisoquinoline NMDA and AMPA antagonists, and
this will be discussed in more detail in a forthcoming
paper covering a series of aryl-spaced decahydroiso-
1
9
quinoline amino acids.
SAR Su m m a r y
In this and the preceding paper we have detailed the
many different aspects that we explored in the struc-
ture-activity studies of this class of 6-substituted
decahydroisoquinoline-3-carboxylic acids. Optimal ac-
tivity was observed for the stereoisomer corresponding
to 1. AMPA antagonist activity was optimized when
the length of the tether connecting an acidic functional-
ity to the bicyclic nucleus was two atoms. Substitution
with a nitrogen or oxygen atom in the tether adjacent
to the bicyclic ring led to lower potency, as did substitu-

Hydroisoquinoline AMPA Antagonist SAR. 2
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2239
Ta ble 4. Activity of AMPA Antagonists in Mice versus Maximal Electroshock-Induced Convulsions, ATPA-Induced Rigidity, and
Activity on the Horizontal Screen
a
ED50 [95% CI] (mg/kg, ip) for effects in assays in mice
3
0 mg/kg (iv)
maximal electroshock-
induced convulsionsc
impairment on the
horizontal screen
b
c
compd
ATPA-induced rigidity
d
NTe
5
2
1
3
7
6
7
5
4
44.5
3.6 [1.8-6.3]
24.1
9.0 [6.8-11.8]
>100
f
19.6 [15.3-25.6]
>100
5
2
3
4
>100
8e.1 [5.3-13.9]
NT
8.5 [6.2-12.5]
9.5 [7.0-11.9]
31.1 [17.6-52.9]
>100
20.1 [14.9-27.4]
25.2 [20.1-33.8]
>100
8.8 [5.0-15.5]
>100
>100
21.0 [17.8-26.6]
5
30.6 [19.2-56.0]
24.7 [16.1-40.2]
a
All compounds administered intraperitoneally 30 min prior to testing in mice. See references 18 and 20. c See ref 17. dSee refe ?.
b
d
e
f
Data from ref 20. NT ) not tested. Data from refs 3 and 4.
tion with a nitrogen atom adjacent to the tetrazole ring.
Substitution with sulfur adjacent to the tetrzole, how-
ever, led to an increase in potency and selectivity.
Bioisosteric replacement of the tetrazole with a car-
boxylate, phosphonate, or sulfonate led to a derease in
activity, and while the isoxazolone showed only a slight
decrease in affinity, a triazole-substituted compound in
which a sulfonyl group was incorporated in the tether
adjacent to the triazole ring led to a significant increase
in potency and selectivity as an AMPA antagonist.
The change in activity observed following small
changes in structure is a striking feature of this class
of 6-substituted decahydroisoquinoline-3-carboxylic ac-
tetrazole is directly bound to the bicyclic nucleus, is both
an AMPA and an NMDA antagonist. Separate the two
functionalities by one methylene, and the compound (51)
is a selective NMDA antagonist; increase the length of
the tether to two methylenes, and the compound is a
selective AMPA antagonist; compounds with a three-
or four-carbon atom tether are only weakly active.
In Vivo Activity
The true utility of an AMPA antagonist can only be
realized if the compounds are active in animals following
systemic administration. A series of compounds pre-
7
sented in this and the preceding paper were selected
2
0
6
3,4
ids. For a series of compounds, 52, 51, and 1
Figure 2), where the relative and absolute stereochem-
for evaluation in a number of assays in mice, and these
are shown in Table 4. All of the compounds have the
same stereochemistry in the bicyclic ring system and
all are racemic. They were all administered by the
intraperitoneal route 30 min prior to testing. Two of
the compounds, 52 and 53, are variants of 1 which have
(
istry of the bicyclic ring is the same, and for each the
distal acid is a tetrazole, we see a distinct change in
activity by changing the length of the all-carbon tether
connecting the two groups. Amino acid 52, where the

2
240 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Ornstein et al.
F igu r e 1. Comparison of AMPA and NMDA receptor pharmacophores as exemplified by compounds from a series of 6-substituted
decahydroisoquinoline-3-carboxylic acids. For the AMPA receptor, the two nonracemic amino acids (-)-1 and (+)-49 are used;
they are epimeric at C-6. For the NMDA receptor, the two nonracemic amino acids (-)-50 and (+)-51 are used; they are also
epimeric at C-6.
blocking NMDA-induced lethality at a dose of 160 mg/
kg.
ATPA, the tert-butyl analog of AMPA, is a selective
AMPA agonist that shows activity following systemic
administration. Intravenous administration of ATPA
in mice produces a characteristic muscle rigidity that
2
0
is blocked selectively by AMPA antagonists.
Com-
pounds from this series that potently displaced AMPA
binding and were antagonists in the cortical slice
preparation dose-dependently prevented ATPA-induced
rigidity. Amino acid 1 for example, was effective in this
assay with an ED50 of 3.6 mg/kg (ip); the lower homolog
5
2, which is an AMPA and NMDA antagonist, was
much less potent (ED50 of 44.5 mg/kg), while the higher
homolog 53 was inactive at doses up to 100 mg/kg.
Amino acids 27, 47, and 54, which are bioisosteric
analogs of 1, also blocked ATPA-induced rigidity in mice,
with ED50s of 8.1, 8.8, and 30.6 mg/kg, respectively; the
carboxy analog 5 was inactive at doses up to 100 mg/
kg.
F igu r e 2. Comparison of structure versus EAA receptor
antagonist activity for a series of 6-substituted decahydroiso-
quinoline-3-carboxylic acids that differ only in the length of
an all-carbon tether that connects a tetrazole to the bicyclic
ring nucleus. Amino acid 54, where the tetrazole is directly
bound to the ring, is both an AMPA and an NMDA antagonist.
Amino acid 51, with a one-carbon tether, is a selective NMDA
antagonist; 1, with a two-carbon tether, is a selective AMPA
antagonist.
These amino acids were also evaluated for their
ability to block maximal electroshock (MES)-induced
convulsions in mice. While this assay is not specific for
AMPA antagonists (NMDA antagonists and conven-
tional anticonvulsant compounds are also active in this
assay), it serves as another measure of the in vivo
activity of this class of compounds and also serves to
demonstrate their potential as anticonvulsant agents.
Comparable potency was seen in this assay for amino
acids 1, 27, and 36 (ED50s of 9.0, 8.5, and 9.5 mg/kg,
respectively), with 52, 47, and 54 being 3-4-fold less
potent (ED s of 24.1, 31.1, and 24.7 mg/kg, respec-
different tether lengths; the others are compounds that
differ in the distal acid bioisostere (5, 36, and 47) and/
or have a heteroatom as a component of the tether (27,
3
6, and 51). While the data are not shown in Table 4,
all of these compounds were evaluated for their ability
to inhibit lethality induced by a 200 mg/kg (ip) dose of
NMDA. Amino acid 52 blocked this lethality with an
ED50 of 19.4 mg/kg,20 while the oxa analog 54 blocked
5
0
2
1
this effect at a minimum effective dose of 80 mg/kg.
All of the other compounds tested were ineffective at
tively); 53 and 5 were inactive.
Finally, we examined the potential for these com-

Hydroisoquinoline AMPA Antagonist SAR. 2
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2241
pounds to produce neurological impairments in mice
using the horizontal screen assay.22 Before the mice
(5 mL rinse). After an additional half hour at room temper-
ature, workup (water/3× ether) afforded 3.9 g of an oil. This
mixture was hydrogenated with 1.0 g of 5% Pd/C in 96 mL of
ethanol at room temperature and 60 psi for 3 h and then
filtered through diatomaceous earth and concentrated in
vacuo. The residue was purified by medium-pressure liquid
chromatography on a LOBAR C column, eluting with 25%
ethyl acetate/hexane at 12 mL/min, to give 0.6 g (15%) of 4,
were tested for MES-induced convulsions, they were
placed on a horizontal screen. Under control conditions
where the mouse is unimpaired, when the screen is
turned upside down, the mouse will climb back to the
top within 2 min. If the mouse is impaired by the test
compound, it will cling to the bottom or fall off. The
best amino acids from this SAR impaired performance
on the horizontal screen at test doses 2-3-fold higher
than their ED50s in the other assays, e.g., 1, 27, 36, and
1
.5 g (38%) of 3, and 1.1 g of mixed fractions for a total yield
of 3.2 g (80%).
3S R ,4a R S ,6R S ,8a R S )-6-(2-C a r b o x y e t h y l)-1,2,3,4,-
a,5,6,7,8,8a-decah ydr oisoqu in olin e-3-car boxylic Acid (5).
(
4
A 1.4 g (3.8 mmol) portion of 3 was heated to reflux overnight
in 50 mL of 6 N hydrochloric acid, cooled, and concentrated in
vacuo. Cation exchange chromatography afforded 0.6 g (65%)
of 5.
4
7 (ED50s of 19.6, 20.1, 25.2, and >100 mg/kg). The two
compounds with both NMDA and AMPA antagonist
activity, 52 and 54, impaired performance on the
horizontal screen test at doses similar to those that
produced protective effects in the other assays.
(
3S R ,4a R S ,6S R ,8a R S )-6-(2-C a r b o x y e t h y l)-1,2,3,4,-
4
a,5,6,7,8,8a-decah ydr oisoqu in olin e-3-car boxylic Acid (6).
As for 5, 0.6 g (1.5 mmol) of 4 gave 0.2 g (46%) of 6.
Eth yl (3SR,4a RS,6RS,8a RS)-6-(2-(N-(m eth ylsu lfon yl)-
ca r ba m oyl)eth yl)-2-(m eth oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,-
8a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (8a ). A solution
of 1.9 g (11.7 mmoL) of 1,1′-carbonyldiimidazole and 4.0 g (11.7
mmol) of 77 in 50 mL of THF was heated to reflux for 1 h,
then cooled to room temperature, and treated with 1.1 g (11.7
mmol) of methanesulfonamide. After 10 min, this solution was
treated with 1.8 g (11.7 mmol) of 1,8-diazabicyclo[5.4.0]undec-
Con clu sion s
From a series of 6-substituted decahyroisoquinoline-
3
-carboxylic acids, we have identified potent excitatory
amino acid antagonists. From this SAR we have real-
5
,6
ized compounds that are selective NMDA antagonists
and selective AMPA antagonists,3
,4,7
as well as two
compounds which possess both NMDA and AMPA
antagonist activity within a single entity.20 We have
demonstrated that these compounds are effective in
displacing EAA receptor ligand binding and that they
antagonize depolarizations due to selective EAA ago-
nists. We have also shown that many of the more
potent AMPA antagonists from this SAR are active
following systemic administration in mice. Amino acid
7
-ene in 10 mL of THF and then stirred overnight at room
temperature. Workup (1 N hydrochloric acid/3× ether/3×
saturated sodium bicarbonate) gave 4.4 g (89%) of 8a .
(
3SR,4aRS,6RS,8aRS)-6-(2-(N-(Meth ylsu lfon yl)ca r ba m -
oyl)eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-
ca r boxylic Acid (9a ). A solution of 4.3 g (10.3 mmol) of 8a
and 22.6 mL (22.6 mmol) of 1 N sodium hydroxide in 45 mL
of ethanol was stirred overnight at room temperature and then
concentrated in vacuo. The residue was dissolved in water
and extracted with ethyl acetate, and then the aqueous layer
was acidified with 5 N HCl and extracted three times with
ethyl acetate. The ethyl acetate extracts were combined, dried
over sodium sulfate, filtered, and concentrated in vacuo to give
1
was evaluated and found to be efficacious in models
2
2
23
of ischemia in rats and cats. Thus, these compounds
may serve as therapeutic agents in the treatment of
acute and chronic neurodegenerative diseases.
3
4
.7 g (92%) of the corresponding acid. This was dissolved in
0 mL of chloroform, treated with 11.4 g (57.0 mmol) of
Exp er im en ta l Section
iodotrimethylsilane, heated to reflux for 2 h, and then con-
centrated in vacuo. The residue was dissolved in water,
extracted with ether, and concentrated in vacuo. Cation
exchange chromatography afforded 2.8 g (87%) of 9a .
Gen er a l Exp er im en ta l. See this section in ref 7. “Anion
exchange chromatography” refers to anion exchange with Bio-
Rad AG1-X8 anion exchange resin (hydroxide form). The resin
(
obtained in acetate form) was prepared by washing (in a
Eth yl (3SR,4a RS,6RS,8a RS)-6-(2-(N-(P h en ylsu lfon yl)-
ca r ba m oyl)eth yl)-2-(m eth oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,-
8a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (8b). As for 8a ,
1.9 g (11.7 mmoL) of 1,1′-carbonyldiimidazole, 4.0 g (11.7
mmol) of 7, 1.1 g (11.7 mmol) of benzenesulfonamide, and 1.8
g (11.7 mmol) of 1,8-diazabicyclo[5.4.0]undec-7-ene in 60 mL
coarse porosity sintered glass funnel) with water, methanol,
water, twice with 1 N sodium hydroxide (converts to the
hydroxide form), and then water until pH 7. The resin was
packed into a glass column in water, and the compound (at
pH g 9) was slowly eluted on with water, and then the column
was washed with water, 50% aqueous THF, and then water
until the pH was neutral. The compound was eluted off of
the column with 3 N aqueous acetic acid, and product-
containing fractions (which were detected with ninhydrin stain
on a TLC plate) were combined and concentrated in vacuo.
7
of THF gave 3.8 g (67%) of 8b.
(3SR,4aRS,6RS,8aRS)-6-(2-(N-(P h en ylsu lfon yl)car bam -
oyl)eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-
ca r boxylic Acid (9b). As for 9a , 3.6 g (7.6 mmol) of 8b and
16.7 mL (16.7 mmol) of 1 N sodium hydroxide in 30 mL of
ethanol gave 3.3 g (96%) of the corresponding acid. This was
treated as for 9a in 30 mL of chloroform with 8.8 g (43.8 mmol)
of iodotrimethylsilane to afford 2.2 g (73%) of 9b.
2
:1 Mixtu r e of Eth yl (3SR,4a RS,6SR,8a RS)- a n d (3SR,-
4
4
a RS,6RS,8a RS)-6-F or m yl-2-(m eth oxyca r bon yl)-1,2,3,4,-
a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (2). A
solution of 3.3 g (10.7 mmol) of ethyl (3SR,4aRS,8aRS)-2-
methoxycarbonyl)-6-(methoxymethylene)-1,2,3,4,4a,5,6,7,8,8a-
Eth yl (3SR,4a RS,6RS,8a RS)-6-(2-(N-(1H-Tetr a zol-5-yl)-
ca r ba m oyl)eth yl)-2-(m eth oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,-
8a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (10). A solution
of 1.9 g (11.7 mmoL) of 1,1′-carbonyldiimidazole and 4.0 g (11.7
(
9
decahydroisoquinoline-3-carboxylate in 125 mL of THF and
60 mL of 1 N hydrochloric acid was stirred for 4 h at room
1
7
temperature. Workup (water/3× dichloromethane/saturated
sodium bicarbonate) afforded 3.2 g of a 2:1 mixture of 2, used
without purification.
mmol) of 7 in 50 mL of THF was heated to reflux for 1 h and
then treated with 1.0 g (11.7 mmol) of 5-amino-1H-tetrazole,
and the solution was refluxed overnight. The cooled solution
was treated with 1 N hydrochloric acid and extracted three
times with ether. The organic extracts were combined and
washed twice with saturated sodium bicarbonate. The aque-
ous bicarbonate washes were combined, acidified with 5 N
hydrochloric acid, and extracted three times with ether. The
ether extracts were combined, dried over sodium sulfate,
filtered, and concentrated in vacuo to give 4.6 g (96%) of 10.
(3SR,4aRS,6RS,8aRS)-6-(2-(N-(1H-Tetr azol-5-yl)car bam -
oyl)eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-
Eth yl (3SR,4a RS,6RS,8a RS)-6-(2-ca r beth oxyeth yl)-2-
(
m eth oxycar bon yl)-1,2,3,4,4a,5,6,7,8,8a-decah ydr oisoqu in -
olin e-3-ca r boxyla te (3) a n d Eth yl (3SR,4a RS,6SR,8a RS)-
6
5
-(2-ca r b et h oxyet h yl)-2-(m et h oxyca r b on yl)-1,2,3,4,4a ,-
,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (4). To a
suspension of 0.6 g (5.0 mmol) of sodium hydride in 20 mL of
THF was added 3.4 g (15.0 mmol) of triethyl phosphonoacetate.
After 30 min at room temperature, this mixture was treated
with 3.2 g (10.7 mmol) of a 2:1 mixture of 2 in 10 mL of THF

2
242 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
Ornstein et al.
ca r boxylic Acid (11). A solution of 4.4 g (10.8 mmol) of 10
and 23.7 mL (23.7 mmol) of 1 N sodium hydroxide in 45 mL
of ethanol was stirred overnight at room temperature and then
concentrated in vacuo. Workup (5 N hydrochloric acid/3×
ethyl acetate) afforded 3.9 g (95%) of the corresponding acid.
This material was treated as for 9a with 8.7 mL (12.3 g, 61.5
mmol) of iodotrimethylsilane in 45 mL of chloroform to give
(20). A 1.1 g (2.5 mmol) portion of 18 and 50 mL of 6 N
hydrochloric acid was heated to reflux overnight, then cooled,
and concentrated in vacuo. The residue was dissolved in 10
mL of ethanol and treated with 1.0 mL of propylene oxide (pH
2-3). The resulting precipitate was filtered, washing with
ethanol, acetone, and ether. The resulting solid was sus-
pended in water, heated to reflux, cooled to room temperature,
filtered, washed with acetone and ether, and dried in vacuo
at 60 °C to afford 0.2 g (33%) of 20.
3
.1 g (84%) of 11.
Eth yl (E)- a n d (Z)-(3SR,4a RS,8a RS)-6-((Ben zyloxyca r -
bon yl)m eth ylen e)-2-(m eth oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,-
a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (13). To a sus-
(3SR,4aRS,6SR,8aRS)-6-(2-Su lfoeth yl)-1,2,3,4,4a,5,6,7,8,-
8a -d eca h yd r oisoqu in olin e-3-ca r boxylic Acid (21). A so-
lution of 1.8 g (4.7 mmol) of 15 and 0.6 g (5.1 mmol) of sodium
sulfite in 11 mL of ethanol and 18 mL of water was heated to
reflux overnight, and then an additional 0.6 g (5.1 mmol) of
sodium sulfite (0.59 g) was added. The mixture was again
heated overnight at reflux and then concentrated in vacuo.
The residue was partitioned between ether and water, the
ether layer was separated and washed again with water, and
then the combined aqueous washes were extracted with ether
and concentrated in vacuo to afford an oil. This material was
heated to reflux overnight in 80 mL of 6 N hydrochloric acid,
cooled to room temperature, and concentrated in vacuo. Anion
exchange chromatography afforded 0.9 g (65%) of 21.
(3SR,4aRS,6RS,8aRS)-6-(2-Su lfoeth yl)-1,2,3,4,4a,5,6,7,8,-
8a -d eca h yd r oisoqu in olin e-3-ca r boxylic Acid (22). As for
21, 1.8 g of 16 afforded 0.7 g (50%) of 22.
8
pension of 3.1 g (77.7 mmol) of sodium hydride in 200 mL of
THF was added 22.2 g (77.7 mmol) of benzyl (diethylphospho-
no)acetate in 50 mL of THF. After 30 min, the resulting clear
solution was treated with 20.0 g (70.6 mmol) of 12 in 80 mL
of THF and then stirred for 5 h at room temperature. Workup
(
brine/3× ether) and Prep HPLC (hexane to 50% ethyl acetate/
hexane) gave 26.3 g of 13.
Eth yl (3SR,4aRS,6SR,8aRS)- an d (3SR,4aRS,6RS,8aRS)-
6
8
-(Car boxym eth yl)-2-(m eth oxycar bon yl)-1,2,3,4,4a,5,6,7,8,-
a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (14). A solution
of 26.2 g (62.9) of 13 in 270 mL of ethyl acetate was
hydrogenated with 5.0 g of 5% Pd/C at 60 psi and room
temperature for 4 h. Filtration through diatomaceous earth
and concentration in vacuo afforded 20.5 g (99%) of 14, which
was used in the next step without further purification.
Eth yl (3SR,4a RS,6SR,8a RS)-6-(2-Br om oeth yl)-2-(m eth -
oxycar bon yl)-1,2,3,4,4a,5,6,7,8,8a-decah ydr oisoqu in olin e-
Eth yl (3SR,4a RS,6SR,8a RS)-6-(2-(1H-Tetr a zol-5-yl)-2-
t h ia e t h yl)-2-(m e t h oxyca r b on yl)-1,2,3,4,4a ,5,6,7,8,8a -
d eca h yd r oisoqu in olin e-3-ca r boxyla te (25). A solution of
3
-ca r boxyla te (15) a n d Eth yl (3SR,4a RS,6RS,8a RS)-6-(2-
9
Br om oet h yl)-2-(m et h oxyca r b on yl)-1,2,3,4,4a ,5,6,7,8,8a -
d eca h yd r oisoqu in olin e-3-ca r boxyla te (16). To a 0 °C
solution of 20.5 g (61.1 mmol) of 14 in 200 mL of THF was
added 61 mL (122 mmol) of a 2 M solution of borane-methyl
sulfide in THF. After 4 h, workup (saturated aqueous sodium
bicarbonate/3× ether) gave 19.3 g of a diastereomeric mixture
of ethyl 6-(2-hydroxyethyl)-2-(methoxycarbonyl)decahydroiso-
quinoline-3-carboxylates. A solution of this compound in 225
mL of dichloromethane and 10 mL (9.8 g, 123.4 mmol) of
pyridine was added dropwise to a 0 °C suspension of tri-
phenylphosphine dibromide [prepared from 24.3 g (92.6 mmol)
of triphenylphosphine and 4.7 mL (14.9 g, 92.6 mmol) of
bromine] in 300 mL of dichloromethane. After 15 min at 0
8.3 g (23.0 mmol) of 23, 2.6 g (25.3 mmol) of thiotetrazole (2.58
g), and 6.4 mL (4.7 g, 46.0 mmol) of triethylamine (4.65 g) in
70 mL of acetonitrile was heated to 80 °C for 20 h and then
cooled to room temperature. Workup (10% sodium bisulfate/
3× ethyl acetate) and chromatography (4/36/60 acetic acid/
ethyl acetate/toluene) afforded 8.9 g (99%) of 25.
Eth yl (3SR,4a RS,6RS,8a RS)-6-(2-(1H-Tetr a zol-5-yl)-2-
t h ia e t h yl)-2-(m e t h oxyca r b on yl)-1,2,3,4,4a ,5,6,7,8,8a -
d eca h yd r oisoqu in olin e-3-ca r boxyla te (26). As for 25, 1.9
9
g (5.1 mmol) of 24, 0.6 g of thiotetrazole, and 1.4 mL (1.0 g,
10.2 mmol) of triethylamine in 20 mL of acetonitrile afforded
1.8 g (91%) of 26.
(3SR,4a RS,6SR,8a RS)-6-(2-(1H -Tet r a zol-5-yl)-2-t h ia -
eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r -
boxylic Acid (27). A solution of 8.9 g (23.2 mmol) of 25 in
100 mL of 6 N hydrochloric acid was heated to 90 °C for 3 h,
then cooled to room temperature, and filtered, and the filter
cake was washed with water and acetone. The solid was dried
in vacuo at room temperature to give 5.4 g (69%) of 27.
(3SR,4a RS,6RS,8a RS)-6-(2-(1H -Tet r a zol-5-yl)-2-t h ia -
eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r -
boxylic Acid (28). A solution of 1.7 g (4.4 mmol) of 26 in 25
mL of 6 N hydrochloric acid was heated to reflux overnight,
then cooled to room temperature, and concentrated in vacuo.
Cation exchange chromatography afforded 0.7 g (51%) of 28.
Eth yl (3SR,4a RS,6SR,8a RS)-6-((5-(P h en a cylth io))-1H-
t e t r a zo l-1-y l)m e t h y l)-2-(m e t h o x y c a r b o n y l)-1,2,3,4,-
4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (30a )
a n d Eth yl (3SR,4a RS,6SR,8a RS)-6-((5-(P h en a cylth io))-
2H -t et r a zol-2-yl)m et h yl)-2-(m et h oxyca r b on yl)-1,2,3,4,-
4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (31a ).
°
C, workup (2× 10% aqueous sodium bisulfate/2× ether)
afforded a solid, which was suspended in ether and filtered to
removed triphenylphosphine oxide. The filtrate was concen-
trated in vacuo, and this procedure was repeated twice.
Preparative HPLC (hexane to 30% ethyl acetate/hexane) gave
3
.7 g (16%) of 16 (first eluted) and 5.3 g (23%) of 15 (second
eluted).
Eth yl (3SR,4a RS,6SR,8a RS)-6-(2-(Dieth ylp h osp h on o)-
eth yl)-2-(m eth oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h y-
d r oisoqu in olin e-3-ca r boxyla te (17). A solution of 1.7 g (4.6
mmol) of 15 and 5 mL of triethyl phosphite was heated to 150
°
C overnight, at which time an additional 1.6 mL of triethyl
phosphite was added. The mixture was heated for 4 h more,
then cooled, and concentrated in vacuo. Chromatography (115
g of silica gel; 2.5% ethanol/ethyl acetate) afforded 1.4 g (69%)
of 17.
Eth yl (3SR,4a RS,6RS,8a RS)-6-(2-(Dieth ylp h osp h on o)-
eth yl)-2-(m eth oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h y-
d r oisoqu in olin e-3-ca r boxyla te (18). As for 17, a solution
of 1.5 g (4.0 mmol) of 16 and 5 mL of triethyl phosphite gave
9
A solution of 4.0 g (11.1 mmol) of 23, 2.7 g (12.2 mmol) of 29,
and 2.7 g (26.7 mmol) of potassium carbonate in 16 mL of
dimethylformamide was heated to 90 °C for 5 h and then cooled
to room temperature. Workup (water/3× dichloromethane; 1×
ether/2× water; 1× brine) and chromatography (350 g of silica
gel; linear gradient of 35% to 50% ethyl acetate/hexane) gave
3.1 g (55%) of 31a and 1.1 g of a mixture of 30a and 31a . The
latter was rechromatographed using radial chromatography,
eluting wtih 50% ethyl acetate/hexane, to afford 0.6 g (11%)
of 30a .
1
.2 g (67%) of 18.
3SR,4a RS,6SR,8a RS)-6-(2-P h osp h on oet h yl)-1,2,3,4,-
a ,5,6,7,8,8a -d eca h yd r oisoq u in olin e-3-ca r b oxylic Acid
(
4
(
19). A 1.3 g (3.0 mmol) portion of 17 and 50 mL of 6 N
hydrochloric acid were heated to reflux overnight, then cooled,
and concentrated in vacuo. The residue was dissolved in 10
mL of ethanol and treated with 1.0 mL of propylene oxide (pH
2
-3). The resulting precipitate was filtered, washing with
ethanol, acetone, and ether. This solid was suspended in
acetone and heated to reflux, then cooled to room temperature,
filtered, washed with acetone and ether, and dried in vacuo
at 60 °C to afford 0.5 g (61%) of 19.
Eth yl (3SR,4a RS,6SR,8a RS)-6-((5-Mer ca p to-2H-tetr a -
zol-2-yl)m et h yl)-2-(m et h oxyca r b on yl)-1,2,3,4,4a ,5,6,7,8,-
8a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (31b). A de-
gassed solution of 1.8 g (3.5 mmol) of 31a in 250 mL of ethanol
was photolyzed for 3 h using a 450 W medium-pressure
mercury lamp and then concentrated in vacuo. The residue
(
3SR,4a RS,6RS,8a RS)-6-(2-P h osp h on oet h yl)-1,2,3,4,-
4
a ,5,6,7,8,8a -d eca h yd r oisoq u in olin e-3-ca r b oxylic Acid

Hydroisoquinoline AMPA Antagonist SAR. 2
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11 2243
was partitioned between dichloromethane and water, adjusting
the pH of the aqueous layer to pH 8-9 using 1 N sodium
hydroxide. The organic portion was washed 3× more with
water, then the combined aqueous washes were acidified to
pH 2-3 with 1 N hydrochloric acid, and the aqueous portion
was extracted four times with dichloromethane. The combined
ylic Acid (38). A 0.7 g (1.9 mmol) portion of 37 and 50 mL of
6 N hydrochloric acid were heated to reflux overnight and then
concentrated in vacuo. Cation exchange chromatography
afforded 0.3 g (58%) of 38.
3-Br om oisoxa zole-5-ca r boxylic Acid a n d 3-Br om oisox-
a zole-4-ca r boxylic Acid (40). A room temperature solution
of 35.2 g (198.0 mmol) of a mixture of 3-bromo-5-(hydroxy-
4
organics were dried (MgSO ), filtered, and concentrated in
vacuo to afford 1.2 g (87%) of 31b.
1
0
methyl)isoxazole and 3-bromo-4-(hydroxymethyl)isoxazole (39)
Eth yl (3SR,4a RS,6SR,8a RS)-6-((5-Mer ca p to-1H-tetr a -
zol-1-yl)m et h yl)-2-(m et h oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,-
in 500 mL of acetone was treated dropwise with 950 mL of
J ones reagent, stirred for 6 h at room temperature, and then
treated with 1000 mL of 2-propanol. The resulting mixture
was filtered through diatomaceous earth, and the filtrate was
concentrated in vacuo. Workup (water/3× ether) gave 34.1 g
(90%) of 40 (mixture of regioisomers).
8
3
a -d eca h yd r oisoqu in olin e-3-ca r boxyla te (30b). As for
1b, 0.6 g (1.2 mmol) of 30a in 225 mL of ethanol gave 0.3 g
(54%) of 30b.
(
3SR,4a RS,6SR,8a RS)-6-((5-Mer ca p t o-2H -t et r a zol-2-
yl)m eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-
ca r boxylic Acid (33). A solution of 1.8 g (4.7 mmol) of 31b
and 50 mL of 6 N hydrochloric acid was heated to reflux
overnight and then cooled to room temperature, whereupon a
precipitate formed. The solid was filtered; washed with water,
acetone, and ether; and then dried in vacuo at room temper-
ature to afford 1.1 g (72%) of 33.
3-Meth oxyisoxa zole-5-ca r boxylic Acid a n d 3-Meth oxy-
isoxa zole-4-ca r boxylic Acid (41). A solution of 34.1 g (178.0
mmol) of 40 (mixture of regioisomers) and 169 g (3.0 mol) of
potassium hydroxide in 580 mL of methanol and 103 mL of
water was heated to reflux for 4 h, then cooled to room
temperature, and treated with 450 mL of concentrated hydro-
chloric acid and 350 mL of water. The resulting solution was
extracted six times with ether, and then the ether extracts
were combined, dried over magnesium sulfate, filtered, and
concentrated in vacuo. The residue was diluted with toluene
and methanol and then concentrated in vacuo to give 20.2 g
(79%) of a mixture of 41 (mixture of regioisomers).
(
3SR,4a RS,6SR,8a RS)-6-((5-Mer ca p t o-1H -t et r a zol-1-
yl)m eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-
ca r boxylic Acid (32). As for 33, 0.3 g (0.7 mmol) of 30b gave
0
.2 g (70%) of 32.
E t h yl (3SR,4a RS,6SR,8a RS)-6-(2-(1H -1,2,4-Tr ia zol-5-
yl)-2-th ia eth yl)-2-(m eth oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,8a -
3-Met h oxy-5-(h yd r oxym et h yl)isoxa zole (42) a n d 3-
Meth oxy-4-(h yd r oxym eth yl)isoxa zole. To a 0 °C solution
of 20.2 g (141.0 mmol) of 41 (mixture of regioisomers) and 14.3
g (141.0 mmol) of triethylamine in 210 mL of THF was added
a solution of 19.3 g (141.0 mmol) of isobutyl chloroformate in
35 mL of THF. After 1.25 h, the precipitate was removed by
filtration and washed with THF. The filtrate was carefully
added to a room temperature solution of 13.4 g (353.0 mmol)
of sodium borohydride in 140 mL of water (cooled as necessary
to avoid a strong exotherm). Workup (1 N hydrochloric acid/
3× ether) and preparative HPLC (hexane to 35% ethyl acetate/
hexane) gave 1.4 g (7.5%) of 3-methoxy-4-(hydroxymethyl)-
isoxazole and 9.1 g (50%) of 42.
d eca h yd r oisoqu in olin e-3-ca r boxyla te (34). A solution of
9
.2 g (25.5 mmol) of 239 and 3.1 g (30.5 mmol) of 1H-1,2,4-
triazole-3-thiol in 92 mL of anhydrous dimethylformamide was
heated to 100 °C overnight and then cooled to room temper-
ature. Workup (10% sodium bisulfate/3× 1/1 chloroform/ethyl
acetate, 1× ether) and preparative HPLC (linear gradient of
3
5% ethyl acetate/hexane to ethyl acetate) afforded an oil
1
which by H NMR showed the presence of dimethylformamide.
The residue was dissolved in ethyl acetate and extracted with
1
N hydrochloric acid and brine. The organic layer was dried
over magnesium sulfate, filtered, and concentrated in vacuo
to give 9.4 g (96%) of 34.
(
3S R ,4a R S ,6S R ,8a R S )-6-(2-(1H -1,2,4-Tr ia zol-5-yl)-2-
3-Meth oxy-5-(br om om eth yl)isoxa zole (43). A solution
of 9.1 g (70.5 mmol) of 42 in 11 mL of dichloromethane and
11.4 mL (11.2 g, 141.0 mmol) of pyridine was added to a 0 °C
suspension of triphenylphosphine dibromide [prepared from
27.7 g (105.8 mmol) of triphenylphosphine and 5.4 mL (16.9
g, 105.8 mmol) of bromine] in 425 mL of dichloromethane.
Workup (2× 10% sodium bisulfate/2× dichloromethane, 1×
ether) and chromatography (450 g of silica gel; step gradient
of 10% ethyl acetate/hexane followed by 20% ethyl acetate/
hexane) afforded 10.8 g (80%) of 43.
th ia eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-
ca r boxylic Acid (35). A mixture of 1.8 g (5.5 mmol) of 34 in
0 mL of 6 N hydrochloric acid was heated to 100 °C overnight
and then concentrated in vacuo. Cation exchange chromatog-
raphy afforded a solid that was suspended in acetone, heated
to reflux for 1 h, filtered, washed with acetone and ether, and
then dried in vacuo at 40 °C to give 0.4 g (22%) of 35.
1
(
3SR,4a RS,6SR,8a RS)-6-((1H -1,2,4-Tr ia zol-5-ylsu lfon -
yl)m eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-
ca r boxylic Acid (36). A room temperature solution of 9.4 g
Dieth yl ((3-Meth oxyisoxa zol-5-yl)m eth yl)p h osp h on a te
(44). A solution of 10.8 g (56.2 mmol) of 43 and 19.3 mL (18.7
g, 112.5 mmol) of triethyl phosphite in 150 mL of toluene was
heated to 120 °C overnight and then concentrated in vacuo.
Preparative HPLC (ethyl acetate to 5% ethanol/ethyl acetate)
of the residue afforded 11.8 g (84%) of 44 (hygroscopic oil).
Eth yl (3SR,4a RS,6RS,8a RS)-6-(2-(3-Meth oxyisoxa zol-
5-yl)e t h yl)-2-(m e t h oxyca r b on yl)-1,2,3,4,4a ,5,6,7,8,8a -
d eca h yd r oisoq u in olin e-3-ca r b oxyla t e (45) a n d E t h yl
(3SR,4aRS,6SR,8aRS)-6-(2-(3-Meth oxyisoxazol-5-yl)eth yl)-
2-(m eth oxyca r bon yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oiso-
qu in olin e-3-ca r boxyla te (46). A -17 °C solution of 3.7 g
(14.9 mmol) of 44 in 10 mL of THF was treated with 14.9 mL
(14.9 mmol) of a 1 M solution of sodium bis(trimethylsilyl)-
amide in THF. After 30 min, a solution of 3.2 g (10.6 mmol)
(24.5 mmol) of 34 in 105 mL of dichloromethane was treated
with 13.3 g (61.2 mmol, 80% by weight) of 3-chloroperoxyben-
zoic acid in three portions over a period of 30 min, then stirred
overnight, and concentrated in vacuo. Chromatography (250
g of silica gel; step gradient of 50% ethyl acetate/hexane (500
mL) followed by ethyl acetate (2000 mL)) of the residue gave
9
2
.0 g of a clear oil. This was treated at reflux overnight with
50 mL of 6 N hydrochloric acid, then cooled, and concentrated
in vacuo. The residue was dissolved in 100 mL of water and
extracted with ether, and then the aqueous phase was
concentrated in vacuo. Ion exchange chromatography afforded
a solid that was refluxed for 1 h in acetone, filtered, washed
with acetone and ether, and dried in vacuo at 60 °C overnight
to give 3.4 g (44%) of 36.
9
Eth yl (3SR,4aRS,6SR,8aRS)-6-(1H-Tetr azol-5-ylam in o)-
m eth yl)-2-(m eth oxycar bon yl)-1,2,3,4,4a,5,6,7,8,8a-decah y-
d r oisoqu in olin e-3-ca r boxyla te (37). A solution of 3.0 g (8.3
of 2 in 10 mL of THF was added, and the susequent mixture
was stirred for 1.5 h while warming to room temperature.
Workup (water/3× ether) and chromatography (250 g of silica
gel; 15% ethyl acetate/toluene) gave 3.0 g (71%) of correspond-
9
mmol) of 23, 2.9 g (20.7 mmol) of powdered potassium
carbonate, and 1.4 g (16.6 mmol) of 5-amino-1H-tetrazole in
ing unsaturated analogs as a mixture of diastereomers. A
2
25 mL of acetonitrile was heated to reflux for 2 days and then
solution of 2.3 g (5.9 mmol) of this mixture in 50 mL of ethyl
acetate was hydrogenated with 2.3 g of 5% Pd/C for 6 h at 60
psi. The reaction mixture was filtered through diatomaceous
earth and then concentrated in vacuo. Chromatography (200
g of silica gel; linear gradient of 5% ethyl acetate/toluene to
15% ethyl acetate/toluene) yielded 0.3 g (14%) of 46 and 1.0 g
(43%) of 45.
concentrated in vacuo. Workup (water/3× ethyl acetate) and
chromatography (200 g of silica gel; 25% ethyl acetate/hexane)
gave 0.7 g (23%, 60% based on recovered starting material) of
3
7 and 1.9 g (61%) of starting bromide.
3SR,4a RS,6SR,8a RS)-6-(1H-Tetr a zol-5-yla m in o)m eth -
yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r box-
(

2
244 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 11
3SR,4a RS,6RS,8a RS)-6-(2-(3-H yd r oxyisoxa zol-5-yl)-
Ornstein et al.
(
(9) Ornstein, P. L.; Augenstein, N. K.; Arnold, M. B. Stereoselective
synthesis of 6-substituted decahyroisoquinoline-3-carboxy-
lates: Intermediates for the preparation of conformationally
constrained acidic amino acids. J . Org. Chem. 1994, 59, 7862-
eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r -
boxylic Acid (47). A mixture of 0.9 g (2.3 mmol) of 45 and
1
for 3 h, then cooled, and concentrated in vacuo. The residue
was diluted with water and concentrated in vacuo. This
residue was diluted with water, the precipitate was filtered,
washing with acetone, and then the filtrate was concentrated
in vacuo. Cation exchange chromatography of this residue
then afforded 0.1 g (17%) of 47.
0 mL of 48% aqueous hydrobromic acid was heated to reflux
7
869.
(
10) Chiarino, D.; Napoletano, M.; Sala, A. 1.3-Dipolar cycloaddition
synthesis of 3-bromo-5-substituted isoxazoles, useful intermedi-
ates for the preparation of pharmacologically active compounds.
J . Heterocycl. Chem. 1987, 24, 43-46.
(11) Bowden, K.; Crank, G.; Ross, W. J . The synthesis of pantherine
and related compounds. J . Chem. Soc. C 1968, 172-185.
(12) Murphy, D. E.; Hutchinson, A. J .; Hurt, S. D.; Williams, M.; Sills,
(
3SR,4a RS,6SR,8a RS)-6-(2-(3-H yd r oxyisoxa zol-5-yl)-
3
M. A. Characterization of the binding of [ H]-CGS-19755: a novel
N-methyl-D-aspartate antagonist with nanomolar affinity in rat
eth yl)-1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin e-3-ca r -
boxylic Acid (48). As for 47, 0.3 g (0.8 mmol) of 46 and 2.5
mL of 48% aqueous hydrobromic acid were heated to reflux
for 3 h, then cooled, and concentrated in vacuo. Cation
exchange chromatography of this residue afforded 0.1 g (62%)
of 48.
brain. Br. J . Pharmacol. 1988, 95, 932-938.
(
13) Nielsen, E. Ø.; Madsen, U.; Schaumburg, K.; Krogsgaard-Larsen,
P. Studies on receptor-active conformations of excitatory amino
acid agonists and antagonists. Eur. J . Med. Chem. Chim. Ther.
1
986, 21, 433-437.
3
(
14) Simon, J . R.; Contrera, J . F.; Kuhar, M. J . Binding of [ H]kainic
Refer en ces
acid, an analogue of L-glutamate. J . Neurochem. 1976, 26, 141-
1
47.
(
(
(
1) Collingridge, G. L.; Lester, R. A. Excitatory amino acid receptors
(
15) Harrison, N. L.; Simmonds, M. A. Quantitative studies on some
antagonists of N-methyl-D-aspartate in slices of rat cerebral
cortex. Br. J . Pharmacol. 1985, 84, 381-391.
(16) Leander, J . D.; Lawson, R. R.; Ornstein, P. L.; Zimmerman, D.
M. N-methyl-D-aspartic acid-induced lethality in mice: selective
antagonism by phencyclidine-like drugs. Brain Res. 1988, 448,
115-120.
17) Leander, J . D.; Rathbun, R. C.; Zimmerman, D. M. Anticonvul-
sant effects of phencyclidine-like drugs: Relation to N-methyl-
D-aspartic acid antagonism. Brain Res. 1988, 454, 368-372.
18) Turski, L.; J acobsen, P.; Honor e´ , T.; Stephens, D. N. Relief of
experimental spasticity and anxiolytic/anticonvulsant actions of
the alpha-amino-3-hydroxy-5-methyl-4-isoxazoleprorpionate an-
tagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(f)quinoxaline.
J . Pharmacol. Exp. Ther. 1992, 260, 742-747.
19) Ornstein, P. L. Aryl-spaced decahydroisoquinoline-3-carboxylic
acids as excitatory amino acid receptor antagonists. U.S. Patent
5,446,501, 1995.
(20) Ornstein, P. L.; Arnold, M. B.; Allen, N. K.; Leander, J . D.;
Tizzano, J . P., Lodge, D.; Schoepp, D. D. 3SR,4aRS,6SR,8aRS-
6-(1H-Tetrazol-5-yl)decahydroisoquinoline-3-carboxylic acid, a
novel, competitive, systemically active NMDA and AMPA recep-
tor antagonist. J . Med. Chem. 1995, 38, 4885-4890.
21) The minimum effective dose is the lowest dose that showed
greater than 50% protection from the insult amongs all of the
mice tested.
22) Gill, R.; Lodge, D. The neuroprotective effects of the decahy-
droisoquinoline, LY 215490; a novel AMPA antagonist in focal
ischemia. Neuropharmacology 1994, 33, 1529-1536.
in the vertebrate central nervous system. Pharmacol. Rev. 1989,
4
0, 143-210.
2) Cunningham, M. D.; Ferkany, J . W.; Enna, S. J . Excitatory
amino acid receptors: a gallery of new targets for pharmacologi-
cal intervention. Life Sci. 1994, 54, 135-148.
3) Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Lodge, D.;
Leander, J . D.; Schoepp, D. D. 3SR,4aRS,6RS,8aRS-6-(2-(1H-
Tetrazol-5-yl)ethyl)-decahydroisoquinoline-3-carboxylic acid: A
structurally novel, systemically active, competitive AMPA recep-
tor antagonist. J . Med. Chem. 1993, 36, 2046-2048.
(
(
4) Schoepp, D. D.; Lodge, D.; Bleakman, D.; Leander, J . D.; Tizzano,
J . P.; Wright, R. A.; Palmer, A. J .; Salhoff, C. R.; Ornstein, P. L.
In vitro and in vivo antagonism of AMPA receptor activation by
(
3
S,4aR,6R,8aR-6-(2-(1H-tetrazol-5-yl)ethyl)decahydroisoquino-
line-3-carboxylic acid. Neuropharmacology 1995, 34, 1159-1168.
5) Ornstein, P. L.; Schoepp, D. D.; Arnold, M. B.; Augenstein, N.
K.; Lodge, D.; Millar, J . D.; Chambers, J . W.; Campbell, J .;
Paschal, J . W.; Zimmerman, D. M.; Leander, J . D. 6-Substituted
decahydroisoquinoline-3-carboxylic acids as potent and selective
conformationally constrained NMDA receptor antagonists. J .
Med. Chem. 1992, 35, 3547-3560.
(
(
(
6) Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Deeter, J . B.;
Leander, J . D.; Lodge, D.; Calligaro, D. O.; Schoepp, D. D.
Unusual stereochemical preferences of decahydroisoquinoline-
3
-carboxylic acid competitive NMDA antagonists. BioMed. Chem.
(
Lett. 1993, 3, 2067-2072.
(
7) Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Borromeo, P.
S.; Lugar, C. W.; Leander, J . D.; Lodge, D.; Schoepp, D. D.
Structure-activity studies of 6-(tetrazolylalkyl)-substituted
decahydroisoquinoline-3-carboxylic acid AMPA receptor antago-
nists. 1. Effects of stereochemistry, chain length, and chain
substitution. J . Med. Chem. 1996, 39, 2219-2231.
(
(23) Bullock, R.; Graham, D. I.; Swanson, S.; McCulloch, J . Neuro-
protective effect of the AMPA receptor antagonist LY293558 in
focal cerebral ischemia in the cat. J . Cerebr. Blood Flow Metab.
1994, 14, 466-471.
(
8) Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Paschal, J .
W. Syntheses of 6-oxodecahydroisoquinoline-3-carboxylates. Use-
ful intermediates for the preparation of conformationally defined
excitatory amino acid antagonists. J . Org. Chem. 1991, 56,
4
388-4392.
J M950913H