N-methyl-D-aspartic acid receptor agonists: Resolution, absolute stereochemistry, and pharmacology of the enantiomers of 2-amino-2-(3-hydroxy-5-methyl-4-isoxazolyl)acetic acid

Article abstract of DOI:10.1021/jm950393q

(R,S)-2-Amino-2-(3-hydroxy-5-methyl-4-isoxazolyl)acetic acid [(R,S)-AMAA, 4] is a potent and selective agonist at the N-methyl-D-aspartic acid (NMDA) subtype of excitatory amino acid receptors. Using the Ugi "four-component condensation" method, the two d

Full text of DOI:10.1021/jm950393q

J . Med. Chem. 1996, 39, 183-190
183
N-Meth yl-D-a sp a r tic Acid Recep tor Agon ists: Resolu tion , Absolu te
Ster eoch em istr y, a n d P h a r m a cology of th e En a n tiom er s of
2-Am in o-2-(3-h yd r oxy-5-m eth yl-4-isoxa zolyl)a cetic Acid
Ulf Madsen,* Karla Frydenvang, Bjarke Ebert, Tommy N. J ohansen, Lotte Brehm, and Povl Krogsgaard-Larsen
PharmaBiotec Research Center, Department of Medicinal Chemistry, The Royal Danish School of Pharmacy,
DK-2100 Copenhagen, Denmark
Received May 26, 1995X
(R,S)-2-Amino-2-(3-hydroxy-5-methyl-4-isoxazolyl)acetic acid [(R,S)-AMAA, 4] is a potent and
selective agonist at the N-methyl-D-aspartic acid (NMDA) subtype of excitatory amino acid
receptors. Using the Ugi “four-component condensation” method, the two diastereomers (2R)-
and(2S)-2-[3-(benzyloxy)-5-methyl-4-isoxazolyl]-N-tert-butyl-2-[N-[(S)-1-phenylethyl]benzamido]-
acetamide (16 and 17, respectively) were synthesized and separated chromatographically. The
absolute stereochemistry of 16 was confirmed by an X-ray analysis. Deprotection of these
intermediates did, however, provide (R)- (8) and (S)- (9) AMAA, respectively, in extensively
racemized forms. N-BOC-protected (R,S)-AMAA (21) was successfully resolved via diastereo-
meric salt formation using cinchonidine. The stereochemical purity and stability of 8 and 9
obtained via this resolution were determined using chiral HPLC. (R)-AMAA (8) showed weak
affinity for [3H]AMPA receptor sites (IC50 ) 72 ( 13 µM) and was shown to be a more potent
inhibitor of [3H]CPP binding (IC50 ) 3.7 ( 1.5 µM) than (S)-AMAA (9) (IC50 ) 61 ( 6.4 µM).
Neither enantiomer of AMAA affected [3H]kainic acid receptor binding significantly. In
electrophysiological studies using rat brain tissue, 8 (EC50 ) 7.3 ( 0.3 µM) was 1 order of
magnitude more potent than 9 (EC50 ) 75 ( 9 µM) as an NMDA receptor agonist.
In tr od u ction
(S)-Glutamic acid (1) is the major excitatory amino
acid (EAA) neurotransmitter in the central nervous
system. Multiple EAA receptor subtypes have been
identified, including both ionotropic and G protein-
coupled (metabotropic) receptors.1-4 The naturally oc-
curring analog of glutamic acid, ibotenic acid (2) (Figure
1), is a potent excitotoxin and an agonist at all classes
of EAA receptors, notably the N-methyl-D-aspartic acid
(NMDA) receptors and subtypes of the metabotropic
receptors.5,6 Compound 2 has been extensively used as
an experimental neurotoxic agent and as a lead struc-
ture for the design of a variety of EAA receptor ligands,
e.g., the specific AMPA receptor agonist (S)-2-amino-3-
(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid [(S)-
AMPA, 3]5,7 and the selective NMDA receptor agonist
(R,S)-2-amino-2-(3-hydroxy-5-methyl-4-isoxazolyl)ace-
tic acid [(R,S)-AMAA, 4].8 The nonselectivity of ibotenic
acid (2) and other EAA agonists seems to indicate that
the recognition sites at EAA receptors share some
F igu r e 1. Structures of a number of excitatory amino acids
structural characteristics. On the othe r hand, the
including the enantiomers of AMAA.
distinctly different pharmacology of the two agonists,
AMPA and AMAA, show that even minor structural
changes can have dramatic effects on affinity for and
potency in spite of the N-methyl substituent. N-
Methylation of other NMDA agonists normally results
in a marked decrease in activity, as observed for
agonistic effects at different EAA receptors.
N-methyl-glutamic acid9 and N-methyl-(R,S)-AMAA.8 A
(S)-Glutamic acid is the endogenous ligand at NMDA
limited number of stereostructure-activity studies on
receptors, whereas the standard agonist NMDA (5) has
NMDA agonists have disclosed a relatively low degree
R-configuration (Figure 1) and is more potent than the
S-enantiomer. This is opposite to what is observed for
the enantiomers of glutamic acid and aspartic acid.
of stereoselectivity. Some agonists are most active in
the R-form, whereas the S-form of other agonists is
selectively recognized by the NMDA receptors.10,11
Another unusual feature concerning NMDA is its high
Thus, (1R,3R)-1-aminocyclopentane-1,3-dicarboxylic acid12
(6) and (2S,3R,4S)-(2-carboxycyclopropyl)glycine13 (7),
* Address correspondence to this author at PharmaBiotec Research
Center, Department of Medicinal Chemistry, The Royal Danish School
of Pharmacy, 2 Universitetsparken, DK-2100 Copenhagen, Denmark.
having opposite configuration at the R-amino acid
carbon, have been shown to be the more active stereo-
isomers of these amino acids at NMDA receptors.
Although a low stereoselectivity is normally observed
Phone: (+45) 35370850, extension 243. Fax: (+45) 35372209. e-mail:
ulf@medchem.dfh.dk.
X Abstract published in Advance ACS Abstracts, November 1, 1995.
0022-2623/96/1839-0183$12.00/0 © 1996 American Chemical Society

184 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Madsen et al.
Sch em e 1a
at the NMDA receptors, the orientation of functional
groups as well as the molecular volume is of importance
for receptor activation. In relation to the latter aspect,
an agonist pocket, being able to accommodate extra
volume of NMDA agonists, has been suggested,11 and
this may be a determining factor for the stereoselectivity
observed for the various agonists. With a very few
exceptions, the NMDA receptor affinity of chiral com-
petitive antagonists resides in the enantiomer having
R-configuration at the R-amino acid carbon.14,15
The NMDA agonist (R,S)-AMAA (4) has previously
been shown to be slightly more potent than NMDA in
in vitro electrophysiological experiments.8 Further-
more, (R,S)-AMAA has recently been shown to possess
substituting effects in a drug discrimination animal
model toward NMDA, being more potent than NMDA
itself.16 This activity indicates that (R,S)-AMAA, at
least to some extent, can penetrate the blood-brain
barrier after systemic administration. In light of these
observations, we decided to synthesize and pharmaco-
logically characterize the enantiomers of (R,S)-AMAA
(4).
Resu lts
Asym m etr ic Syn th esis. The synthesis of the stereo-
isomers of (R,S)-AMAA (4) was attempted using the Ugi
“four-component condensation” reaction.17 This method
for asymmetric synthesis of R-amino acids affords both
enantiomers through the formation of diastereomeric
intermediates, using (R)- or (S)-1-phenylethylamine as
the chiral auxiliary. 3-(Benzyloxy)-5-methyl-4-isox-
azolecarbaldehyde (12) was synthesized from the car-
boxylic acid 10,18 via the acid chloride 11, which by
reduction with lithium tri-tert-butoxyaluminum hydride
afforded the desired aldehyde 12 (Scheme 1). The “four-
component condensation” of aldehyde 12, (S)-1-phenyl-
ethylamine (13), tert-butylisonitrile (14), and benzoic
acid (15) afforded the two diastereomeric intermediates
16 and 17, which were easily separated on a silica gel
column. The assignment of the absolute stereochem-
a
Reagents: (i) SOCl2; (ii) LiAl[OC(CH3)3]3H; (iii) HCOOH; (iv)
6 M HCl.
meric salt formation using cinchonidine as the resolving
base. For this purpose, a more facile synthesis of (R,S)-
AMAA21 (4) was developed. Ethyl (R,S)-2-amino-2-(3-
methoxy-5-methyl-4-isoxazolyl)acetate hydrobromide22
(20) was deprotected by reflux in concentrated hydro-
bromic acid to give (R,S)-AMAA (4), which could be
isolated in the zwitterionic form (Scheme 2).
Solutions of N-BOC-(R,S)-AMAA (21) and cinchoni-
dine in ethyl acetate were mixed, and a salt primarily
consisting of 22 precipitated. After three recrystalliza-
tions, no change in optical rotation values of 22 was
observed. Evaporation of the mother liquor from the
first precipitation of 22 followed by three recrystalliza-
tions gave 23, again with no change in optical rotation
values after the third recrystallization. The two dia-
stereomeric salts 22 and 23 were deprotected to give
(R)-AMAA (8) and (S)-AMAA (9), respectively.
Ch ir a l HP LC An a lysis. The stereochemical purity
of 8 and 9 was investigated by chiral HPLC analyses
using an (S)-proline-derivatized and Cu2+-chelated col-
umn. Initial experiments showed very poor resolution
of (R,S)-AMAA (4), whereas a satisfactory resolution of
the enantiomers of N-BOC-protected AMAA (21) was
observed. Thus, 8 and 9 were N-BOC-protected, and
the chiral HPLC analyses disclosed enantiomeric excess
(ee) values of 95.0% and 96.0%, respectively.
1
istry of 16 and 17 was based on H NMR chemical shift
values of especially the tert-butyl groups. Empirically,
the chemical shift values for the tert-butyl groups of R,S-
and S,R-forms of this class of compounds are in the
range of δ 1.3-1.4, whereas the corresponding values
for the respective R,R- and S,S-diastereomers are ca. δ
1.1.19,20 Deprotection of the two diastereomeric inter-
mediates was carried out either by a two-step procedure
or in one step (Scheme 1). Following the two-step
procedure, 16 or 17 was N-debenzylated by treatment
with formic acid to give compound 18 or 19, respectively.
Subsequent deprotection in 6 M HCl under reflux
afforded (R)- (8) and (S)-(9) AMAA, respectively. The
deprotection could also be carried out in one step by
reflux of 16 or 17 in 6 M HCl to give 8 or 9, respectively.
In both cases, 8 or 9 was isolated as the zwitterion after
recrystallization from water. The optical rotation mea-
surements for the final products from different depro-
tection experiments gave varying and generally very low
values. These results were interpreted as being indica-
tive of racemization during the deprotection reactions,
and chiral HPLC analyses confirmed that the final
products were extensively racemized.
Resolu tion by Dia ster eom er ic Sa lt F or m a tion .
Resolution was subsequently carried out by diastereo-
Due to the stereochemical instability of the enanti-
omers of AMAA observed during the deprotection

NMDA Receptor Agonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 185
Sch em e 2a
a
Reagents: (i) HBr/H2O; (ii) BOC-O-BOC; (iii) cinchonidine;
(iv) AcOH; (v) TFA; (vi) TEA.
experiments on the products from the Ugi synthesis (16
and 17), a stability experiment on a sample of 8,
obtained from the diastereomeric salt resolution, was
carried out. The stereochemical stability of (R)-AMAA
(8) was investigated in the Krebs buffer used for
electrophysiological studies of the compounds (see In
Vitro Pharmacology). Compound 8 was dissolved in the
buffer, and samples were taken after 0, 2, and 24 h. The
samples were N-BOC-protected and analyzed by chiral
HPLC as described above. These experiments showed
slight racemization after 2 h (ee value decreasing from
93.5% to 93.0% of the sample analyzed) and increased
racemization after 24 h (ee value decreasing from 93.0%
to 89.5%).
X-r a y Cr ysta llogr a p h ic An a lysis of Com p ou n d
16. An X-ray crystallographic analysis established the
assignment of the absolute configuration of compound
16, which was found to have the R,S-configuration. The
chiral center with S-configuration originates from (S)-
1-phenylethylamine, used for the synthesis of compound
16. The configuration of the protected R-amino acid
moiety was determined relative to this and shown to
be R. The deprotection experiments performed on
compound 16 (see Asymmetric Synthesis) afforded
partially racemized AMAA showing negative optical
rotation. This qualitatively corresponds to the rotation
observed for compound 8, obtained via diastereomeric
salt formation. Thus, the absolute configuration of 8
was established to be R. This absolute stereochemistry
confirms the stereochemistry predicted from the 1H
NMR experiments performed on the Ugi products 16
and 17. In agreement with this, the N-BOC-protected
form of 8 was shown to elute first from the HPLC
column, thus showing lower affinity than the S-form (9)
for the (S)-proline-derivatized column. This order of
elution is analogous to what is normally seen for
underivatized R-amino acids on the same type of ligand
exchange columns.23
F igu r e 2. Perspective drawing of the two independent
molecules of 16 (ORTEP24), A and B. The crystallographic atom
labeling is shown on A. Displacement ellipsoids represent 50%
probability; H atoms are drawn as circles of arbitrary radius.
pendent molecules of 16 in the asymmetric unit are the
same within three standard deviations, except for C41-
N42 [A, 1.469(3) Å; B, 1.485(3) Å] and N42-C43 [A,
1.501(3) Å; B, 1.483(4) Å]. Larger deviations are found,
however, for the corresponding bond angles (Table 1).
The greatest deviations were observed for the noncyclic
parts of the molecules, which may be due to the
differences in the conformation of the two independent
molecules. In Figure 2, the isoxazole rings of the two
molecules (A and B) are oriented in the same way in
order to illustrate the conformational differences. The
orientations of the three different phenyl groups are
different for molecules A and B. For two of the phenyl
groups, the changes (10-15°) are of minor importance
for the overall conformations of the two molecules, while
for the third one (C71-C76), the orientation is changed
by 99° (Figure 2). For molecule B, the two phenyl
groups are almost parallel, but no stacking effect is
observed (Figure 3). The phenyl groups of the benzoyl
moieties (C81-C86) of molecules A and B have large
displacement parameters. The groups are disordered
and can adopt slightly different orientations. The
orientations of the carbonyl groups of the amide moieties
of molecules A and B, C46dO46, differ by 30°, and this
affects the spatial position of the tert-butyl moieties. The
orientation of the carbonyl group in molecule A results
in a favorable intermolecular close contact (C-H‚‚‚OdC)
(Table 1).
The asymmetric unit of the crystal structure of
compound 16 consists of two independent molecules of
16 and one molecule of diethyl ether. Perspective
drawings of the two independent molecules of 16
(labeled A and B), the former with the crystallographic
atom-labeling scheme, are shown in Figure 2. The
corresponding bond lengths (Table 1) of the two inde-
The crystal packing (Figure 3) can be described as a
collection of columns, packed along the a-axis. The
alternating columns, consisting of either A or B mol-
ecules, interact with one another through van der Waals

186 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Madsen et al.
Ta ble 1. Selected Bond Lengths (Å), Bond Angles (Deg), Torsion Angles (Deg), and Hydrogen Bond Dimensions (Å, Deg) for
Compound 16a
molecule
A
B
molecule
N42-C45
A
B
O1-C5
1.361(4)
1.417(3)
1.307(4)
1.347(3)
1.424(4)
1.354(4)
1.512(4)
1.483(4)
1.464(4)
1.492(4)
1.469(3)
1.542(4)
1.350(4)
1.423(3)
1.303(4)
1.350(3)
1.427(4)
1.362(4)
1.500(4)
1.483(4)
1.460(4)
1.494(4)
1.485(3)
1.544(4)
1.349(4)
1.501(3)
1.529(4)
1.524(4)
1.238(3)
1.501(3)
1.221(3)
1.336(4)
1.481(4)
1.522(4)
1.518(5)
1.525(5)
1.359(4)
1.483(4)
1.525(4)
1.527(4)
1.237(4)
1.493(3)
1.225(3)
1.340(4)
1.489(4)
1.524(5)
1.524(4)
1.528(5)
O1-N2
N42-C43
C43-C431
C43-C71
C45-O45
C45-C81
C46-O46
C46-N47
N47-C48
C48-C91
C48-C92
C48-C93
N2-C3
C3-O31
C3-C4
C4-C5
C4-C41
C5-C51
O31-C32
C32-C61
C41-N42
C41-C46
Bond Angles
C5-O1-N2
109.2(2)
109.5(2)
N42-C43-C431
N42-C43-C71
C71-C43-C431
C76-C71-C43
C72-C71-C43
O45-C45-N42
O45-C45-C81
N42-C45-C81
C82-C81-C45
C86-C81-C45
O46-C46-N47
O46-C46-C41
N47-C46-C41
C46-N47-C48
N47-C48-C91
N47-C48-C92
N47-C48-C93
C92-C48-C91
C92-C48-C93
C91-C48-C93
111.9(2)
111.1(3)
C3-N2-O1
103.6(2)
122.2(3)
114.2(3)
123.6(3)
102.9(2)
132.7(3)
124.3(2)
110.1(3)
135.5(3)
114.4(3)
113.4(2)
108.8(3)
122.2(3)
119.5(3)
110.4(2)
110.8(2)
115.8(2)
121.7(2)
120.9(2)
117.4(2)
103.3(2)
122.9(3)
114.6(3)
122.5(3)
102.3(3)
133.2(3)
124.5(3)
110.3(3)
135.0(3)
114.7(3)
114.4(2)
107.3(3)
120.2(3)
121.6(3)
111.9(2)
110.6(2)
113.1(2)
121.8(2)
118.9(2)
119.3(2)
115.2(2)
112.4(2)
120.1(3)
122.4(3)
120.8(3)
118.2(2)
120.9(2)
121.9(2)
117.6(2)
124.5(3)
120.3(3)
115.1(2)
124.6(2)
106.5(2)
110.2(3)
110.6(3)
109.6(3)
110.9(3)
108.9(3)
113.0(2)
113.6(3)
119.4(3)
122.6(3)
121.3(3)
119.7(3)
119.0(2)
121.0(2)
118.9(2)
125.3(3)
120.7(3)
114.0(2)
124.0(3)
106.0(3)
109.8(3)
109.9(3)
109.4(3)
111.4(3)
110.2(3)
N2-C3-O31
N2-C3-C4
O31-C3-C4
C5-C4-C3
C5-C4-C41
C3-C4-C41
C4-C5-O1
C4-C5-C51
O1-C5-C51
C3-O31-C32
O31-C32-C61
C66-C61-C32
C62-C61-C32
N42-C41-C4
N42-C41-C46
C4-C41-C46
C45-N42-C41
C45-N42-C43
C41-N42-C43
Torsion Angles
O1-N2-C3-O31
N2-C3-O31-C32
C3-O31-C32-C61
O3-C32-C61-C62
N2-O1-C5-C51
N2-C3-C4-C41
C3-C4-C41-N42
C3-C4-C41-C46
C4-C41-N42-C43
C4-C41-N42-C45
C41-N42-C43-C71
178.6(3)
-5.9(4)
-176.3(3)
115.2(3)
-178.1(3)
-178.3(3)
68.7(3)
177.5(3)
-0.1(5)
-173.6(3)
102.5(4)
-177.9(3)
-178.3(3)
79.7(3)
N42-C43-C71-C72
C41-N42-C43-C431
C41-N42-C45-O45
C41-N42-C45-C81
N42-C45-C81-C82
C4-C41-C46-O46
C4-C41-C46-N47
C41-C46-N47-C48
C46-N47-C48-C91
C46-N47-C48-C92
C46-N47-C48-C93
-83.7(4)
124.9(3)
170.6(3)
-10.2(4)
-59.1(3)
-123.8(3)
59.0(3)
174.7(3)
-179.4(3)
-60.6(4)
62.4(4)
15.7(4)
119.7(3)
173.9(3)
-6.1(4)
-71.0(3)
-93.9(3)
86.4(3)
178.5(3)
-172.8(3)
-54.7(4)
68.1(4)
-164.3(3)
-154.7(3)
59.3(3)
52.8(3)
-120.2(3)
-105.1(3)
-128.4(3)
-111.3(3)
Hydrogen Bond Dimensions and Close Contact
D-H‚‚‚X
(symmetry code)
(x-0.5,0.5-y,-z)
(x-0.5,1.5-y,-z)
(x-0.5,0.5-y,-z)
D‚‚‚X
H‚‚‚X
<DHX
A, N47-H47---O45
B, N47-H47‚‚‚O45
A, C62-H62‚‚‚O46
2.792(3)
2.941(3)
3.269(4)
1.944(3)
2.064(3)
2.378(4)
161.3(1)
173.9(1)
156.0(1)
a
Estimated standard deviations are given in parentheses.
forces. Molecules within each column, consisting of
either A or B molecules, are linked to each other by
N-H‚‚‚OdC hydrogen bonds (Table 1). As mentioned
above, another close contact (C-H‚‚‚OdC), which may
be classified as a hydrogen bond,25 is observed within
the columns containing A molecules. The diethyl ether
molecules are situated between the columns with no
favorable contacts to the molecules of 16. The large
displacement parameters observed for the atoms of the
diethyl ether molecule suggest that the molecule is
disordered.
In Vitr o P h a r m a cology. The affinities of 8 and 9
for EAA receptors were investigated in receptor-binding
assays using [3H]-(R,S)-[3-(2-carboxy-4-piperazinyl)-
prop-1-yl]phosphonic acid ([3H]CPP), [3H]AMPA, and
[3H]kainic acid as ligands for NMDA, AMPA, and kainic
acid receptors, respectively.26-28 The results obtained
are listed in Table 2 together with similar data obtained
for (R,S)-AMAA (4), (R)-AMPA 2, (S)-AMPA (3), and
NMDA (5). (R)-AMAA (8) showed high affinity for the
[3H]CPP-binding site, slightly higher than (R,S)-AMAA
(4), whereas the S-form, 9, was markedly weaker.
Compound 8 showed very veak affinity for the AMPA-
binding site, whereas neither enantiomer showed af-
finity for kainic acid receptor sites.
In vitro electrophysiological experiments were per-
formed using the rat cortical slice model (Table 2).29 (R)-
AMAA (8) showed an EC50 value of 7.3 µM compared to
12 µM for (R,S)-AMAA (4), whereas (S)-AMAA (9) was
much weaker (EC50 ) 75 µM). Thus, the R-form, 8, is
ca. 10 times more potent than the S-form, 9. The
activity of both enantiomers could be fully antagonized
by the NMDA antagonist CPP. Dose-response curves
for (R)- and (S)-AMAA are shown in Figure 4. The
enantiomeric excess for (S)-AMAA was determined to
be 96.0% (see Chiral HPLC Analysis), meaning that it

NMDA Receptor Agonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 187
F igu r e 3. Stereoscopic view of the molecular packing of compound 16, seen in the direction of the a-axis, with horisontal b- and
vertical c-axes. Displacement ellipsoids represent 50% probability for non-hydrogen atoms. For clarity only hydrogen atoms
connected to N atoms are shown. Hydrogen bonds are indicated by thin lines.
Ta ble 2. Receptor-Binding and Electrophysiological Data
(Mean ( SEM, n ) 3-6)
cortical
IC50 (µM)
slice,
compound
[3H]CPP [3H]AMPA [3H]kainic acid EC50 (µM)
(R,S)-AMAAa (4) 4.5
>100
72 ( 13
>100
>100
76
>100
>100
>100
>100
>100
>100
12
(R)-AMAA (8)
(S)-AMAA (9)
NMDAa (5)
3.7 ( 1.5
7.3 ( 0.3
75 ( 9
15
3.8
580
61 ( 6.4
35
(R)-AMPAb 2
(S)-AMPAb (3)
>100
>100
0.02
a
b
Reference 8. Reference 30.
contains 2.0% of (R)-AMAA. This content of impurity
of the more potent enantiomer 8 is too low to explain
the excitatory activity measured for (S)-AMAA (9). The
estimated activity (EC50 ) 75 µM) thus seems to reflect
a true activity of (S)-AMAA (9), considering the degree
of stereochemical stability determined in the test buffer
(see Chiral HPLC Analysis).
F igu r e 4. Dose-response curves for (R)- (8) and (S)- (9)
AMAA obtained from the cortical slice model. The curves
represent one experiment, which has been repeated three
times.
Discu ssion
Stereostructure-activity studies on NMDA receptor
ligands have not yet produced a coherent picture of the
stereochemical requirements for activation or blockade
of this class of EAA receptors. For a majority of
competitive NMDA receptor antagonists, the R-form
shows markedly greater effect than the S-form.10 In
contrast, a low degree of stereoselectivity for NMDA
receptor agonists is normally observed, and whereas the
S-form of some NMDA agonists show higher activity
than the R-form, the opposite relative potency of such
compounds is frequently observed.11
Attempts to produce enantiomerically pure (R)-AMAA
(8) and (S)-AMAA (9) by Ugi reaction were unsuccessful
due to extensive racemization during the final depro-
tection. This stereochemical lability of the AMAA
enantiomers is similar to, but less pronounced than, that
observed for the analog ibotenic acid18 (Figure 1).
Successful resolution was carried out by diastereomeric
salt formation, which did not involve treatment with
strong acid or base.
potent as an NMDA agonist than (S)-AMAA (9). The
enantiomeric impurity of (S)-AMAA was determined to
be 2.0%. On the basis of this and the result of stability
experiments, it is concluded that the observed activity
of (S)-AMAA (9) primarily is reflecting its own activity
rather than the activity of the small impurity of (R)-
AMAA (8). The stereochemical preference of NMDA
receptors for (R)-AMAA is analogous to what has been
observed for compounds like NMDA (5) and (1R,3R)-1-
aminocyclopentane-1,3-dicarboxylic acid (6),12 whereas
(S)-glutamic acid9 and (2S,3R,4S)-(2-carboxycyclopro-
pyl)glycine13 (7) have been shown to be the more potent
enantiomers. Molecular modeling studies on a number
of NMDA agonists, including AMAA, have predicted (R)-
AMAA to be the more active enantiomer.11 The 10-fold
higher potency determined for (R)-AMAA (8) as com-
pared to the S-enantiomer 9 is relatively low compared
to the degree of stereoselectivity exhibited by the AMPA
receptor subtype of EAA receptors. Thus (S)-AMPA (3)
shows a 3800 times higher affinity than the enantiomer
The electrophysiological studies showed (R)-AMAA (8)
to be the more potent enantiomer, ca. 10 times more

188 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Madsen et al.
(tol-petroleum ether): mp 170-171 °C. [R]21D ) -4.9°, [R]21
(R)-AMPA 2 in [3H]AMPA-binding experiments (Table
2). Glutamic acid also shows high stereoselective af-
finity for the [3H]AMPA-binding site, the S-form having
an IC50 value of 0.50 µM compared to 47 µM for the
R-form.31 A similar degree of stereoselectivity is ob-
served for the binding of (S)- and (R)-glutamic acid to
NMDA receptor sites, whereas the enantiomers of
aspartic acid show equal affinity for NMDA receptors.32
In conclusion, the enantiomers of the NMDA agonist
AMAA have been synthesized, their absolute stereo-
chemistry and stereochemical stability have been de-
termined, and the R-form (8) has been shown to be 1
order of magnitude more active than (S)-AMAA (9).
365
) +8.8 ° (c )1.0, MeOH); 1H NMR (90 MHz, CDCl3) δ 7.5-7.1
(15H, m), 5.8 (1H, s), 5.1 (1H, d, J ) 12 Hz), 4.9 (1H, d, J ) 12
Hz), 4.8 (1H, br), 4.7 (1H, s), 2.2 (3H, s), 1.8 (3H, d, J ) 7 Hz),
1.3 (9H, s); IR 3250 (m), 3050 (m), 3020 (w), 2955 (m), 2920
(w), 1680 (s), 1615 (s) cm-1. Anal. [C32H35N3O4‚1/2C7H8(tol)]
C, H, N.
Further elution afforded after recrystallization 17 (205 mg,
24%) (tol-petroleum ether): mp 116.5-117.5 °C; [R]21D ) -2.5
°, [R]21365 ) -5.8 ° (c ) 1.0, MeOH); 1H NMR (90 MHz, CDCl3)
δ 7.6-7.2 (15H, m), 6.0 (1H, s), 5.3 (2H, s), 5.0 (1H, q, J ) 7
Hz), 4.8 (1H, s), 2.4 (3H, s), 1.5 (3H, d, J ) 7 Hz), 1.1 (9H, s);
IR 3250 (w), 3050 (w), 3020 (w), 2970 (m), 2920 (w), 1670 (s),
1635 (s), 1625 (s) cm-1. Anal. (C32H35N3O4) C, H, N.
(R)-2-Ben za m id o-2-[3-(b en zyloxy)-5-m et h yl-4-isoxa z-
olyl]-N-ter t-bu tyla ceta m id e (18). A solution of compound
16 (100 mg, 0.2 mmol) in concentrated formic acid (95%, 5 mL)
was stirred for 30 min at room temperature and then at 60 °C
for 30 min. Evaporation followed by CC [CH2Cl2-EtOAc (9:
1)] afforded after recrystallization 18 (55 mg, 68%) (tol-
Exp er im en ta l Section
Ch em istr y. Gen er a l P r oced u r es. Melting points were
determined in capillary tubes and are uncorrected. Optical
rotations were determined on a Perkin-Elmer 141 polarimeter.
Elemental analyses were performed by Mr. G. Cornali, Mi-
croanalytical Laboratory, LEO Pharmaceutical Products, Den-
mark, or Mr. P. Hansen, Department of General and Organic
Chemistry, University of Copenhagen, Denmark, and are
within (0.4% of the calculated values unless otherwise stated.
1H NMR spectra were recorded on a Varian EM 360L (60
MHz), a J eol FX 90Q (90 MHz), or a Bruker AC-200F (200
MHz) instrument using TMS or 1,4-dioxane as internal
standard for spectra recorded in CDCl3 or D2O, respectively.
IR spectra were recorded from KBr disks or for oils as liquid
sandwiches between NaCl disks on a Perkin-Elmer 781 grating
infrared spectrophotometer. Column chromatography (CC)
was performed on Merck silica gel 60 (0.060-0.200 mm).
Analytical thin-layer chromatography (TLC) was carried out
using Merck silica gel 60 F254 plates. Compounds containing
the 3-hydroxyisoxazole moiety were visualized on TLC plates
using UV light and a FeCl3 spraying reagent (yellow color).
Compounds containing amino groups were visualized using a
ninhydrin spraying reagent, and all compounds under study
were also detected on TLC plates using a KMnO4 spraying
reagent. Evaporations were performed under vacuum on a
rotary evaporator at 15 mmHg.
petroleum ether): mp 83-84 °C; [R]24
) -251 ° (c ) 0.6,
365
1
MeOH); H NMR (90 MHz, CDCl3) δ 7.5-7.3 (10 H, m), 5.7
(1H, br), 5.35 (2H, br), 5.3 (2H, s), 2.45 (3H, s), 1.25 (9H, s);
IR 3280 (s), 3050 (m), 3015 (w), 2960 (m), 2920 (w), 1645 (s),
1630 (s) cm-1
found, 66.92.
. Anal. (C24H27N3O4) H, N; C: calcd, 68.36;
(S)-2-Ben za m id o-2-[3-(b en zyloxy)-5-m et h yl-4-isoxa -
zolyl]-N-ter t-bu tyla ceta m id e (19). Compound 19 (42 mg,
52%) was obtained from 17 (100 mg, 0.19 mmol) by a method
similar to that described for the synthesis of 18. IR and 1H
NMR spectra were virtually identical with those of compound
18. Compound 19 did not crystallize and could not be
completely purified due to coelution with the starting material
when chromatographed [Rf ) 0.25, CH2Cl2-EtOAc (9:1)].
(R)- a n d (S)-2-Am in o-2-(3-h yd r oxy-5-m et h yl-4-isox-
a zolyl)a cetic Acid [(R)-AMAA (8) a n d (S)-AMAA (9)].
Meth od A: A suspension of compound 18 (47 mg, 0.11 mmol)
was refluxed in 6 M HCl (3 mL) for 2 h. The reaction mixture
was evaporated and the residue dissolved in H2O (5 mL) and
extracted with CH2Cl2 (2 × 5 mL). After evaporation of the
aqueous phase, recrystallization (H2O) afforded 8 (8.2 mg,
39%): [R]24
) -15.8° (c ) 0.20, 0.16 M HCl). Anal.
365
(C6H8N2O4,H2O) C, H, N. Compound 9 (6.5 mg, 41%) was
synthesized by the same method from 19 (35 mg, 0.083
3-(Ben zyloxy)-5-m eth yl-4-isoxa zoleca r bon yl Ch lor id e
(11). To a suspension of 3-(benzyloxy)-5-methyl-4-isoxazole-
carboxylic acid18 (10) (1.0 g, 4.3 mmol) in thionyl chloride (5
mL) was added DMF (1 drop), and the mixture was refluxed
for 15 min. After evaporation and re-evaporation from CH2-
Cl2, Kugelrohr distillation (0.2 mmHg, 200-225 °C) afforded
11 (990 mg, 92%) as a colorless oil, which crystallized when
stored at 5 °C: IR 3070 (w), 3030 (w), 2960 (w), 2930 (m), 1750
(s), 1590 (s) cm-1. Anal. (C12H10ClNO3) C, H, N; Cl: calcd,
14.12; found, 13.24.
mmol): [R]24 ) +12.9° (c ) 0.13, 0.16 M HCl).
365
Meth od B: Compound 8 (10.3 mg, 27%) was synthesized
as described in method A using compound 16 (105 mg, 0.20
mmol) as the starting material and a reflux time of 4 h: [R]24
365
) -2.8° (c ) 0.16, 0.16 M HCl). Similarly was compound 9
(12 mg, 33%) synthesized from compound 17 (100 mg, 0.19
mmol): [R]24 ) +3.4° (c ) 0.17, 0.16 M HCl). For further
365
characterization of 8 and 9, see the later synthesis of these
enantiomers from compounds 22 and 23, respectively.
(R,S)-2-Am in o-2-(3-h yd r oxy-5-m eth yl-4-isoxa zolyl)a ce-
tic Acid [(R,S)-AMAA (4)] Mon oh yd r a te. A solution of 2022
(430 mg, 1.5 mmol) in concentrated HBr (7 mL, 48%) was
refluxed for 30 min and evaporated. The residue was re-
evaporated twice from H2O and recrystallized (H2O) to give
(R,S)-AMAA (4) monohydrate (211 mg, 76%): mp 184-186 °C
3-(Ben zyloxy)-5-m eth yl-4-isoxa zoleca r ba ld eh yd e (12).
To a solution of compound 11 (400 mg, 1.6 mmol) in dry THF
(7 mL) at -78 °C was added, during a period of 5 min, a
solution of lithium tri-tert-butoxyaluminum hydride (450 mg,
1.8 mmol) in dry THF (5 mL). After stirring at -78 °C for 30
min and slow return to room temperature, H2O (10 mL) was
added. The mixture was acidified with AcOH and extracted
with Et2O (3 × 15 mL). The combined extracts were dried
and evaporated, and the residue was subjected to CC [tol-
EtOAc (9:1)]. Recrystallization (tol-petroleum ether) afforded
12 (190 mg, 55%): mp 68-69 °C; 1H NMR (60 MHz, CDCl3) δ
9.8 (1H, s), 7.4 (5H, s), 5.3 (2H, s), 2.3 (3H, s); IR 3015 (w),
1
dec; H NMR (90 MHz, D2O) δ 4.64 (1H, s), 2.34 (3H, s); IR
3260 (s), 3200-2200 (multiple, w-m), 1645 (s), 1620 (s), 1585
(m), 1530 (s), 1510 (s) cm-1. Anal. (C6H8N2O4‚H2O) C, H, N.
(R,S)-2-[(ter t-Bu toxyca r bon yl)a m in o]-2-(3-h yd r oxy-5-
m et h yl-4-isoxa zolyl)a cet ic Acid (21). To a solution of 4
(380 mg, 2 mmol) in triethylamine (1 mL, 7 mmol) and H2O
(10 mL) was added di-tert-butyl dicarbonate (0.5 mL, 22 mmol)
dissolved in THF (10 mL). The reaction mixture was stirred
at room temperature for 2.5 h, after which the THF was
evaporated. After acidification with 4 M HCl, the aqueous
phase was quickly extracted with EtOAc (3 × 10 mL), and the
organic phases were dried and evaporated. Recrystallization
(CH2Cl2) afforded 21 (465 mg, 85%) as colorless crystals: mp
2915 (w), 2850 (w), 2760 (w), 1675 (s), 1610 (s), 1510 (s) cm-1
Anal. (C12H11NO3) C, H, N.
.
(2R)- a n d (2S)-2-[3-(Ben zyloxy)-5-m eth yl-4-isoxa zolyl]-
N-ter t-b u t yl-2-[N-[(S)-1-p h en ylet h yl]b en za m id o]a cet a -
m id e (16 a n d 17). A solution of 12 (350 mg, 1.6 mmol) and
(S)-1-phenylethylamine (200 µL, 1.6 mmol) in MeOH (4 mL)
was refluxed for 30 min. After cooling to room temperature,
benzoic acid (200 mg, 1.6 mmol) and tert-butylisonitrile (180
µL, 1.6 mmol) were added, and the reaction mixture was
stirred over night. Evaporation followed by CC [CH2Cl2-
EtOAc (9:1)] afforded after recrystallization 16 (380 mg, 45%)
1
99-104 °C; H NMR (200 MHz, CD3OD) δ 5.07 (1H, s), 2.32
(3H, s), 1.45 (9H, s); IR 3500-2400 (multiple bands, m-s), 3260
(s), 1650 (s), 1620 (s), 1585 (s), 1530 (s), 1515 (s), 1505 (s) cm-1
.
Anal. (C11H16N2O6‚1/4H2O) C, H, N.

NMDA Receptor Agonists
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 189
Cin ch on id in iu m (R)-2-[(ter t-Bu toxyca r bon yl)a m in o]-
2-(3-h yd r oxy-5-m eth yl-4-isoxa zolyl)a ceta te (22) a n d Cin -
ch on id in iu m (S)-2-[(ter t-Bu t oxyca r b on yl)a m in o]-2-(3-
h yd r oxy-5-m eth yl-4-isoxa zolyl)a ceta te (23). A solution of
21 (500 mg, 1.8 mmol) in EtOAc (10 mL) was added to a
solution of cinchonidine (540 mg, 1.8 mmol) in EtOAc (50 mL),
both dissolved by heating. Upon standing at room tempera-
ture and then at 5 °C, a precipitate, primarily consisting of
22 (435 mg), was collected; [R]29365 ) -332.4° (c ) 0.5, MeOH).
These crystals were recrystallized three times from EtOAc (the
optical rotation did not change significantly for the crystals
obtained from second and third recrystallizations) to give 22
Pharmacology). Samples were taken out at 0, 2, and 24 h and
then treated with di-tert-butyl dicarbonate to obtain the
N-BOC-protected product, which was analyzed by chiral
HPLC, as described above. The ee of the sample taken out at
the beginning (0 h) was determined to be 93.5%; after 2 and
24 h in the buffer solution, the ee was reduced to 93.0% and
89.5%, respectively.
X-r a y Cr ysta llogr a p h ic An a lysis of (2R)-2-[3-(Ben zyl-
o xy)-5-m e t h y l-4-isox a zolyl]-N -t er t -b u t y l-2-[N -[(S )-1-
p h en yleth yl]ben za m id o]a ceta m id e (16). Crystal data:
2(C32H35N3O4), CH3CH2OCH2CH3, Mr ) 1125.4, colorless
needles; recrystallized from Et2O, mp 168-170 °C [161-163
°C (rearrangment)]; orthorhombic, space group P212121 (no.
19); a ) 12.443(3) Å, b ) 21.161(1) Å, c ) 23.271(2) Å, V )
6127(1) Å3, Z ) 4, Dc ) 1.220 Mg m-3; F(000) ) 2408, µ(Cu
KR) ) 0.649 mm-1, T ) 122 K, crystal dimensions ) 0.07 ×
0.24 × 0.38 mm3.
(94 mg, 18%): mp 186.5-187.0 °C dec; [R]25 ) -399.0° (c )
365
1
0.6, MeOH); H NMR (200 MHz, CDCl3) (aromatic region) δ
8.80 (1H, d, J ) 5 Hz), 7.98 (1H, d, J ) 8 Hz), 7.89 (1H, d, J
) 8 Hz), 7.67 (1H, d, J ) 5 Hz), 7.55 (1H, t, J ) 8 Hz), 7.38
(1H, t, J ) 8 Hz); IR 3320 (m, br), 3290 (m), 2980 (m), 2940
(w), 2560 (w, br), 1715 (s), 1615 (s) cm-1. Anal. (C30H28N4O7)
C, H, N.
Da ta collection a n d p r ocessin g: Diffraction data were
collected on an Enraf-Nonius CAD-4 diffractometer, using
graphite monochromated Cu KR radiation (λ ) 1.5418 Å). The
crystal was cooled to 122 ( 0.5 K in a stream of nitrogen gas.
Unit cell dimensions were determined by least-squares refine-
ment of 20 reflections with θ values in the range 36.42-38.52°.
Two octants of data were collected in the ω/2θ scan mode (θ <
75°: 0 e h e 15; 0 e k e 26; -29 e l e 29). Data were reduced
by a procedure including detailed peak profile analysis using
the programs of Blessing (DREADD).34,35 Intensities of five
reflections were monitored every 104 s, and they showed a
decrease in intensity of 4.6%. Appropriate scaling was per-
formed using the program SCALE3.34,35 Absorption corrections
were applied using the program ABSORB36 (Tmin ) 0.830; Tmax
The mother liquor from the first precipitation of 22 was
evaporated. The residue consisted primarily of 23 (605 mg);
[R]29 ) -233.8° (c ) 0.5, MeOH). Three consecutive recrys-
365
tallizations from EtOAc (the optical rotation did not change
significantly for the crystals obtained from second and third
recrystallizations) afforded 23 (102 mg, 20%): mp 189.5-190.5
°C dec; [R]25365 ) -173.4° (c ) 0.5, MeOH); 1H NMR (200 MHz,
CDCl3) (aromatic region) δ 8.79 (1H, d, J ) 5 Hz), 7.96 (2H,
m), 7.65-7.45 (3H, m); IR 3420 (m), 3250 (m, br), 2970 (m),
2940 (w), 2560 (w, br), 1715 (s), 1615 (s) cm-1
(C30H28N4O7) C, H, N.
.
Anal.
(R)-(-)-2-Am in o-2-(3-h yd r oxy-5-m et h yl-4-isoxa zolyl)-
a cetic Acid [(R)-AMAA (8)] Mon oh yd r a te. Compound 22
(85 mg, 0.15 mmol) was dissolved in H2O (5 mL) containing
AcOH (5 drops), and the aqueous phase was extracted with
EtOAc (3 × 5 mL). The organic phases were dried and
evaporated to give crude N-BOC-protected 8. The crude
product was treated with trifluoroacetic acid (0.5 mL) in H2O
(2 mL) for 1 h at room temperature. After evaporation, the
residue was dissolved in H2O (2 drops) and EtOH (1 mL) and
pH adjusted to 4.5 by addition of triethylamine to precipitate
crude 8. The crude product was recrystallized from H2O to
) 0.957). A total of 14 368 reflections were averaged (Rint
)
0.011 on Fo2) according to the point group symmetry 222,
resulting in 12 621 unique reflections.
Str u ctu r e d eter m in a tion a n d r efin em en t: The struc-
ture was solved by direct methods using the program SHELXS-
86.37,38 Full-matrix least-squares refinements (SHELXL-9339
)
2
were performed on F2, minimizing Σw(Fo - Fc2)2 with aniso-
tropic displacement parameters for non-hydrogen atoms. The
positions of some of the hydrogen atoms were observed in ∆F
maps, but all hydrogen atoms were included in idealized
positions (riding atoms) with Uiso set to 1.2Ueq of the parent
atom (1.5Ueq for H atoms of methyl groups). The phenyl
groups (C81-C86) of the two (crystallographically) indepen-
dent molecules were during the refinements treated as rigid
groups, in order to keep bond lengths and bond angles of equal
magnitude. These phenyl groups are slightly disordered and
were observed with large displacement parameters. Refine-
ment was carried out on 12 617 unique reflections. Four
reflections were suppressed as the intensities were observed
negative. The refinement (736 parameters, 12 617 reflections)
give 8 (14.6 mg, 51%): mp 180.5-182.5 °C; [R]25 ) -357.0°
365
1
(c ) 0.26, MeOH); H NMR spectrum was virtually identical
with that of the racemate 4; IR 3380 (m), 3300-2200 (multiple,
w-m), 1670 (s), 1610 (m, br), 1585 (m, br), 1495 (s) cm-1
.
(S)-(+)-2-Am in o-2-(3-h yd r oxy-5-m et h yl-4-isoxa zolyl)-
a cetic Acid [(S)-AMAA (9)] Mon oh yd r a te. (S)-AMAA (9)
(15.0 mg, 47%) was obtained from compound 23 (95 mg, 0.17
mmol) by a method similar to that used for compound 8: mp
180.5-182.5 °C; [R]25365 ) +328.0 ° (c ) 0.22, MeOH); 1H NMR
was virtually identical with that of the racemate 4 and IR with
that of the enantiomer 8.
with the molecule 16 having the R,S-configuration converged
2
Deter m in a tion of En a n tiom er ic P u r ity. Chiral HPLC
was performed on a Waters HPLC Model 510, Waters UK6
injector, LKB 2155 HPLC column oven, Waters UV detector
Model 480 (210 nm), Hitachi D-2000 Chromato-Integrator, and
an analytical column (length, 120 mm; i.d., 4.6 mm) containing
a silica-based packing material with (S)-proline chemically
bound via a 3-glycidoxypropyl spacer and Cu2+ as a complexing
agent.33 Chromatography was performed with the column at
50 °C and isocratic elution with an aqueous buffer (50 mM
KH2PO4, pH ) 4.6, 1 mL/min) as the mobile phase. The
enantiomeric composition was determined from peak areas.
Very poor separation of the enantiomers (R)- (8) and (S)-
(9) AMAA was observed on the chiral column, whereas good
enantiomeric separation was obtained when analyzing the
N-BOC-protected derivative 21. Thus, an analytical sample
of the obtained crystals of (R)- (8) and (S)- (9) AMAA was
N-BOC-protected with di-tert-butyl dicarbonate, by a method
similar to that described for the synthesis of 21, and the
products were analyzed by chiral HPLC. For (R)-AMAA (8)
the ee, after N-BOC-protection, was determined to 95.0% and
for (S)-AMAA (9), 96.0%.
at RF ) 0.060, RwF ) 0.128 [w-1 ) (σ2(Fo2) + (0.0521P)2
+
2
3.6171P, where P ) (Fo + 2Fc2)/3, GOOF ) 1.06, 10 092
reflections with Fo g 4σ(Fo)]. In the final refinement, the
largest parameter shift/esd ) -0.047, and the residual electron
density varied between 0.50 and -0.66 e Å-3 (observed in the
area in which the diethyl ether molecule is located). The
S-configuration of the N-(1-phenylethyl) moiety was known
from the asymmetric synthesis, and the R-configuration of the
protected R-amino acid moiety was determined relative to this.
The absolute configuration was supported by refinement of the
Flack absolute structure factor ꢀ in the final structure factor
calculation, ꢀ ) -0.09(21).40 The large standard deviation of
ꢀ is due to the atoms present in this compound (C, H, N, O),
which produce minor anomalous scattering. Complex atomic
scattering factors (C, H, N, O) were used.41
Recep tor Bin d in g. Affinity for NMDA, AMPA, and kain-
ate receptors were determined using the ligands [3H]CPP, [3H]-
AMPA, and [3H]kainic acid, respectively.26-28 The membrane
preparations used in all the receptor-binding experiments were
prepared according to the method of Ransom and Stec.42
In Vitr o Electr op h ysiology. A rat cortical slice prepara-
tion for determination of EAA-evoked depolarizations de-
scribed by Harrison and Simmonds29 was used in a slightly
modified version. Wedges (500 µm thick) of rat brain, contain-
Deter m in a tion of En a n tiom er ic Sta bility. An analyti-
cal sample of (R)-AMAA (8) was dissolved in the Krebs buffer
used for the in vitro electrophysiological studies (see In Vitro

190 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Madsen et al.
(17) Gokel, G.; Lu¨dke, G.; Ugi, I. Four-component condensations and
related reactions. In Isonitrile Chemistry; Ugi, I., Ed.; Academic
Press: New York, 1971; pp 145-199.
(18) Madsen, U.; Brehm, L.; Krogsgaard-Larsen, P. Excitatory amino
acids. Improved synthesis of ibotenic acid and X-ray analysis of
an unexpected reaction product, (RS)-N-[2-(3-benzyloxyisoxazol-
5-yl)-1-phenylethyl]-3-oxobutyramide. J . Chem. Soc., Perkin
Trans. I 1988, 359-364.
(19) Marquarding, D.; Hoffman, P.; Heitzer, H.; Ugi, I. Stereoselective
four-component condensations of R-ferrocenylethylamine and its
absolute configuration. J . Am. Chem. Soc. 1970, 92, 1969-1971.
(20) Semple, J . E.; Wang, P. C.; Lysenko, Z.; J oullie´, M. M. Total
synthesis of (+)-furanomycin and stereoisomers. J . Am. Chem.
Soc. 1980, 102, 7505-7510.
(21) Christensen, S. B.; Krogsgaard-Larsen, P. Structural analogues
of ibotenic acid. Synthesis of (()-R-amino-3-hydroxy-5-methyl-
4-isoxazoleacetic acid and derivatives thereof. Acta Chem. Scand.
1978, B32, 27-30.
(22) J ohansen, T. N.; Frydenvang, K.; Ebert, B.; Krogsgaard-Larsen,
P.; Madsen, U. Synthesis and structure-activity studies on acidic
amino acids and related diacids as NMDA receptor ligands. J .
Med. Chem. 1994, 37, 3252-3262.
(23) Gu¨bitz, G. Direct separation of enantiomers by high performance
ligand exchange chromatography on chemically bonded chiral
phases. J . Liq. Chromatogr. 1986, 9, 519-535.
(24) J ohnson, C. K. ORTEP II. Report ORNL-5138; Oak Ridge
National Laboratory: Oak Ridge, TN, 1976.
ing cerebral cortex and corpus callosum, were placed through
a grease barrier for electrical isolation with each part in
contact with an Ag/AgCl pellet electrode. The cortex and
corpus callosum parts were constantly superfused with a Mg2+
-
free oxygenated Krebs buffer at room temperature, the test
compounds were added to the cortex superfusion medium, and
the potential difference between the electrodes was recorded
on a chart recorder.
Ack n ow led gm en t. This work was supported by
grants from the Lundbeck and Alfred Benzon Founda-
tions, EEC (BIO2-CT93-0243), The Danish Natural
Science Research Council, and the Danish State Bio-
technology Programme (1991-1995). The assistance of
Mr. F. Hansen with the X-ray data collection is grate-
fully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Tables listing final
atomic coordinates for compound 16 and diethyl ether, equiva-
lent isotropic displacement parameters, anisotropic displace-
ment parameters for the non-hydrogen atoms, and a full list
of bond lengths and bond angles (8 pages); a list of structure
factors (31 pages). Ordering information is given on any
current masthead page.
(25) J effrey, G. A.; Saenger, W. Hydrogen bonding in biological
structures; Springer-Verlag: Berlin, 1991; pp 29-30.
(26) Murphy, D. E.; Schneider, J .; Boehm, C.; Lehmann, J .; Williams,
M. Binding of [3H]3-(2-carboxypiperazin-4-yl)propyl-1-phospho-
nic acid to rat brain membranes. A selective high-affinity ligand
for N-methyl-D-aspartate receptors. J . Pharmacol. Exp. Ther.
1987, 240, 778-784.
Refer en ces
(1) Monaghan, D. T.; Bridges, R. J .; Cotman, C. W. The excitatory
amino acid receptors. Annu. Rev. Pharmacol. Toxicol. 1989, 29,
365-402.
(2) Watkins, J . C.; Krogsgaard-Larsen, P.; Honore´, T. Structure-
activity relationships in the development of excitatory amino acid
receptor agonists and competitive antagonists. Trends Pharma-
col. Sci. 1990, 11, 25-33.
(3) Simon, R. P., Ed. Excitatory Amino Acids; Thieme: New York,
1992.
(4) Hollmann, M.; Heinemann, S. Cloned Glutamate Receptors.
Annu. Rev. Neurosci. 1994, 17, 31-108.
(5) Krogsgaard-Larsen, P.; Honore´, T.; Hansen, J . J .; Curtis, D. R.;
Lodge, D. New class of glutamate agonist structurally related
to ibotenic acid. Nature (London) 1980, 284, 64-66.
(6) Schoepp, D. D.; J ohnson, B. G. Excitatory amino acid agonist-
antagonist interactions at 2-amino-4-phosphonoburytic acid-
sensitive quisqualate receptors coupled to phosphonoinositide
hydrolysis in slices of rat hippocampus. J . Neurochem. 1988, 50,
1605-1613.
(7) Hansen, J . J .; Lauridsen, J .; Nielsen, E.; Krogsgaard-Larsen,
P. Enzymic resolution and binding to rat brain membranes of
the glutamic acid agonist R-amino-3-hydroxy-5-methyl-4-isox-
azolepropionic acid. J . Med. Chem. 1983, 26, 901-903.
(8) Madsen, U.; Ferkany, J . W.; J ones, B. E.; Ebert, B.; J ohansen,
T. N.; Holm, T.; Krogsgaard-Larsen, P. NMDA receptor agonists
derived from ibotenic acid. Preparation, neuroexcitation and
neurotoxicity. Eur. J . Pharmacol., Mol. Pharmacol. Sect. 1990,
189, 381-391.
(9) Watkins, J . C.; Olverman, H. J . In Exctiatory Amino Acids in
Health and Disease; Lodge, D., Ed.; J . Wiley & Sons: Chichester,
England, 1988; pp 13-45.
(10) Watkins, J . C., Collingridge, G. L., Eds. The NMDA receptor;
Oxford University Press: Oxford, U.K., 1989.
(11) Kyle, D. J .; Patch, R. J .; Karbon, E. W.; Ferkany, J . W. NMDA
receptors: heterogeneity and agonism. In Excitatory Amino Acid
Receptors: Design of Agonists and Antagonists; Krogsgaard-
Larsen, P., Hansen, J . J ., Eds.; Ellis Horwood: Chichester,
England, 1992; pp 121-161.
(12) Curry, K.; Peet, M. J .; Magnuson, D. S. K.; McLennan, H.
Synthesis, resolution, and absolute configuration of the isomers
of the neuronal excitant 1-amino-1,3-cyclopentanedicarboxylic
acid. J . Med. Chem. 1988, 31, 864-867.
(13) Shinozaki, H.; Ishida, M.; Shimamoto, K.; Ohfune, Y. A confor-
mationally restricted analogue of L-glutamate, the (2S,3R,4S)
isomer of L-R-(carboxycyclopropyl)glycine, activates the NMDA-
type receptor more markedly than NMDA in the isolated rat
spinal cord. Brain Res. 1989, 480, 355-359.
(14) Watkins, J . C. In Excitatory Amino Acid Antagonists; Meldrum,
B. S., Ed.; Blackwell Science Publishers: London, 1991; pp 84-
100.
(27) Honore´, T.; Nielsen, M. Complex structure of quisqualate-
sensitive glutamate receptors in rat cortex. Neurosci. Lett. 1985,
54, 27-32.
(28) Braitman, D. J .; Coyle, J . T. Inhibition of [3H]Kainic acid receptor
binding by divalent cations correlates with ion affinity for the
calcium channel. Neuropharmacology 1987, 26, 1247-1251.
(29) 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.
(30) Nielsen, B.; Fisker, H.; Ebert, B.; Madsen, U.; Curtis, D. R.;
Krogsgaard-Larsen, P.; Hansen, J . J . Enzymatic resolution of
AMPA by use of R-chymotrypsin. Bioorg. Med. Chem. Lett. 1993,
3, 107-114.
(31) Hansen, J . J .; Nielsen, B.; Krogsgaard-Larsen, P.; Brehm, L.;
Nielsen, E. Ø.; Curtis, D. R. Excitatory amino acid agonists.
Enzymic resolution, X-ray structure and enantioselective activi-
ties of (R)- and (S)-bromohomoibotenic acid. J . Med. Chem. 1989,
32, 2254-2260.
(32) Fagg, G. E.; Baud, J . Characterization of NMDA receptor
ionophore complexes in the brain. In Excitatory Amino Acids in
Health and Disease; Lodge, D., Ed.; J . Wiley & Sons: Chichester,
England, 1988; pp 63-90.
(33) Gu¨bitz, G.; J ellenz, W.; Santi, W. Resolution of the optical
isomers of underivatized amino acids on chemically bonded
chiral phases by ligand exchange chromatography. J . Liq.
Chromatogr. 1981, 4, 701-712.
(34) Blessing, R. H. Data reduction and error analysis for accurate
single crystal diffraction intensities. Crystallogr. Rev. 1987, 1,
3-58.
(35) Blessing, R. H. DREADD - Data reduction and error analysis
for single-crystal diffractometer data. J . Appl. Crystallogr. 1989,
22, 396-397.
(36) DeTitta, G. T. ABSORB: an absorption correction program for
crystals enclosed in capillaries with trapped mother liquor. J .
Appl. Crystallogr. 1985, 18, 75-79.
(37) Sheldrick, G. M. SHELXS-86. Program for the Solution of Crystal
Structure; University of Go¨ttingen: Go¨ttingen, Germany, 1986.
(38) Sheldrick, G. M. Phase annealing in SHELX-90: Direct methods
for larger structures. Acta Crystallogr. 1990, A46, 467-473.
(39) Sheldrick, G. M. SHELXL-93. Program for Crystal Structure
Refinement; University of Go¨ttingen: Go¨ttingen, Germany, 1993.
(40) Flack, H. D. On enantiomorph-polarity estimation. Acta Crys-
tallogr. 1983, A39, 876-881.
(41) International Tables for Crystallography; Kluwer Academic
Publishers: Dordrecht, The Netherlands, 1992; Vol. C, Tables
4.2.6.8 and 6.1.1.4.
(15) Ornstein, P. L.; Klimkowski, V. J . Competitive NMDA receptor
antagonists. In Excitatory Amino Acid Receptors: Design of
Agonists and Antagonists; Krogsgaard-Larsen, P., Hansen, J . J .,
Eds.; Ellis Horwood: Chichester, England, 1992; pp 183-200.
(16) Arnt, J .; Sa´nches, C.; Lenz, S. M.; Madsen, U.; Krogsgaard-
Larsen, P. Differentiation of in vivo effects of AMPA and NMDA
receptor ligands using drug discrimination methods and con-
vulsant/anticonvulsant activity. Eur. J . Pharmacol. 1995, in
press.
(42) Ransom, R. W.; Stec, N. L. Cooperative modulation of [3H]MK-
801 to the N-methyl-D-aspartate receptor ion channel complex
by L-glutamate, glycine and polyamines. J . Neurochem. 1988,
51, 830-836.
J M950393Q