Cyclobutane quisqualic acid analogues as selective mGluR5a metabotropic glutamic acid receptor ligands

Article abstract of DOI:10.1021/jm9806897

The conformationally constrained cyclobutane analogues of quisqualic acid (Z)- and (E)-1-amino-3-[2'-(3',5'-dioxo-1',2',4'- oxadiazolidinyl)]cyclobutane-1-carboxylic acid, compounds 2 and 3, respectively, were synthesized. Both 2 and 3 stimulated phosphoi

Full text of DOI:10.1021/jm9806897

J . Med. Chem. 1999, 42, 1639-1647
1639
Cyclobu ta n e Qu isqu a lic Acid An a logu es a s Selective m Glu R5a Meta botr op ic
Glu ta m ic Acid Recep tor Liga n d s
Louis Littman,† Christopher Tokar,‡ Shankar Venkatraman,‡ Robert J . Roon,§ J ames F. Koerner,§
Michael B. Robinson,† and Rodney L. J ohnson*,‡
Departments of Medicinal Chemistry and Biochemistry, University of Minnesota, Minneapolis, Minnesota 55455, and
The Children’s Seashore House, Children’s Hospital of Philadelphia, and Departments of Pediatrics and Pharmacology,
University of Pennsylvania, Philadelphia, Pennsylvania 19104
Received December 8, 1998
The conformationally constrained cyclobutane analogues of quisqualic acid (Z)- and (E)-1-amino-
3-[2′-(3′,5′-dioxo-1′,2′,4′-oxadiazolidinyl)]cyclobutane-1-carboxylic acid, compounds 2 and 3,
respectively, were synthesized. Both 2 and 3 stimulated phosphoinositide (PI) hydrolysis in
the hippocampus with EC50 values of 18 ( 6 and 53 ( 19 µM, respectively. Neither analogue
stimulated PI hydrolysis in the cerebellum. The effects of 2 and 3 were also examined in BHK
cells which expressed either mGluR1a or mGluR5a receptors. Compounds 2 and 3 stimulated
PI hydrolysis in cells expressing mGluR5a but not in those cells expressing mGluR1a. The
EC50 value for 2 was 11 ( 4 µM, while that for 3 was 49 ( 25 µM. Both 2 and 3 did not show
any significant effect on cells expressing the mGluR2 and mGluR4a receptors. In addition,
neither compound blocked [3H]glutamic acid uptake into synaptosomal membranes, and neither
compound was able to produce the QUIS effect as does quisqualic acid. This pharmacological
profile indicates that 2 and 3 are selective ligands for the mGluR5a metabotropic glutamic
acid receptor.
In tr od u ction
4-phosphonobutanoic acid (L-AP4).10 The mGluRs have
been implicated in mediating development of the ner-
vous system,11,12 long-term potentiation,13,14 modulation
of neuronal activity,15,16 and regulation of glutamic acid
release.17,18
The acidic amino acids L-glutamic acid and L-aspartic
acid are the major excitatory neurotransmitters in the
mammalian central nervous system (CNS). These ex-
citatory amino acids (EAAs) activate both receptors
coupled to ion channels (ionotropic) and receptors
coupled to second-messenger systems (metabotropic).1
The ionotropic glutamic acid receptors have been di-
vided into the following subtypes based on agonist
specificity: N-methyl-D-aspartate (NMDA), kainic acid,
and R-amino-3-hydroxy-5-methyl-4-isoxazolepropionic
acid (AMPA) receptors. These receptors have been
implicated in mediating fast synaptic events, are in-
volved in long-term potentiation, and can cause neu-
ronal degeneration upon excessive activation.1-3
To date, eight distinct metabotropic glutamic acid
receptors (mGluRs) have been cloned (mGluR1-mGluR8)
with multiple splice variants existing for several of these
receptor subtypes.1,4,5 The mGluRs have been catego-
rized into three groups based on agonist specificity
and second-messenger coupling. Group 1 receptors
(mGluR1a-d, mGluR5a,b) are coupled to phospho-
inositide (PI) hydrolysis and are stimulated by the EAA
analogues quisqualic acid (1) and (1S,3R)-1-amino-1,3-
cyclopentanedicarboxylic acid [(1S,3R)-ACPD].6,7 Group
2 receptors (mGluR2, mGluR3) are coupled to inhibition
of cyclic AMP (cAMP) formation and are stimulated by
(+)-2-aminobicyclo[3.1.0]hexane-2,6-dicarboxylic acid
(LY354740)8 and (2S,2′R,3′R)-2-(2′3′-dicarboxycyclopro-
pyl)glycine (DCG-IV).9 Group 3 receptors (mGluR4a,b,
mGluR6, mGluR7, mGluR8) are also coupled to inhibi-
tion of cAMP formation but are stimulated by L-2-amino-
Great success, in the past, has been achieved in the
synthesis of selective metabotropic receptor ligands by
synthesizing conformationally constricted analogues of
glutamic acid and L-AP4. Since quisqualic acid is one
of the most potent agonists for group 1 mGluRs, the
synthesis of conformationally constrained analogues of
1 has been undertaken in hopes of developing high-
affinity ligands which can be used to define the func-
tions of these receptors in the mammalian CNS. In the
present study, we describe the synthesis and selective
mGluR5a receptor agonist activity of the cyclobutyl
analogues of quisqualic acid, compounds 2 and 3.
Ch em istr y
The synthesis of 2 and 3 is outlined in Scheme 1. The
synthetic strategy for the two cyclobutyl quisqualic acid
analogues was centered around the construction of the
oxadiazolidinedione ring on a suitable cyclobutane pre-
cursor and employed methodology similar to that used
in the synthesis of L-quisqualic acid.19,20 In our ap-
proach, condensation of cyclobutanone 421 with O-ben-
zylhydroxylamine in the presence of sodium cyanoboro-
hydride gave the N-benzyloxyamino derivative as a
‡ Department of Medicinal Chemistry, University of Minnesota.
§ Department of Biochemistry, University of Minnesota.
† Children’s Hospital of Philadelphia and University of Pennsylva-
nia.
10.1021/jm9806897 CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/21/1999

1640 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Littman et al.
F igu r e 1. NOEs observed for 6b.
Sch em e 1
F igu r e 2. Molecular structure of one of the two independent
molecules of 7a with atomic numbering scheme.
pendent molecules plus one THF molecule per asym-
metric unit. Both independent molecules showed a Z
relationship between the tert-butoxycarbonylamino moi-
ety and the N-(benzyloxy)-N-(ethoxycarbonyl) group.
The molecular structure of one independent molecule
of 7a is shown in Figure 2. On the basis of these results,
the geometrical configuration of cyclobutyl quisqualic
acid analogues 2 and 3 was assigned to be Z and E,
respectively.
Biologica l Resu lts
The two cyclobutyl quisqualic acid analogues 2 and
3, along with L-glutamic acid, (1SR,3RS)-ACPD, and
quisqualic acid, were tested for their ability to stimulate
PI hydrolysis in either neonatal rat hippocampus or
cerebellum (Figure 3). Quisqualic acid (100 µM), L-
glutamic acid (1 mM), and (1SR,3RS)-ACPD (300 µM)
stimulated PI hydrolysis in both hippocampus (p <
0.001, p < 0.01, and p < 0.01 for difference from base-
line, respectively) and cerebellum (p < 0.01, p < 0.001,
and p < 0.01 for difference from baseline, respectively).
Interestingly, compounds 2 (100 µM) and 3 (100 µM)
stimulated PI hydrolysis in hippocampus (p < 0.001 and
p < 0.01 for difference from baseline, respectively) but
not in cerebellum.
The effects of increasing concentrations of 2 and 3 on
PI hydrolysis in hippocampus were examined (Figure
4). In hippocampus, we previously demonstrated that
data for the stimulation of PI hydrolysis by L-glutamic
acid or (1SR,3RS)-ACPD were best fit to curves having
a Hill coefficient of 1 with EC50 values of 450 ( 120 and
51 ( 5 µM, respectively, while data for quisqualic acid
(see Figure 4) were best fit to curves having two
components (p < 0.01) with the high-affinity component
(EC50 ) 0.43 µM) accounting for 70% of the total effect
and the low-affinity component (EC50 ) 44 µM) account-
ing for 30% of the total effect.22,23 In the current study,
the EC50 values for 2 and 3 were 18 ( 6 and 53 ( 19
µM, respectively. The maximal effects (in DPM/100 000
DPM incorporated) were 30 200 ( 7 900 for 2 and
31 800 ( 8 400 for 3. The maximal effects of these com-
pounds were comparable, in parallel assays, to the effect
caused by a concentration of (1SR,3RS)-ACPD (300 µM)
that induces a maximal effect in the hippocampus.
Of the metabotropic receptor subtypes that are coupled
to PI hydrolysis (mGluR1 and mGluR5), mGluR1a is
expressed at high levels in both cerebellum and hippo-
campus while mGluR5a is expressed at high levels in
hippocampus but at low levels in cerebellum.24 The
mixture of Z (5a ) and E (5b) isomers in a 1:1 ratio. The
isomers were separated by HPLC, and each isomer then
was acylated with ethoxycarbonyl isocyanate to yield
6a and 6b, respectively. Removal of the benzyl protect-
ing group from each isomer of 6 by hydrogenolysis gave
the N-hydroxy esters 7a and 7b. Treatment of each of
these under basic conditions led to formation of the
oxadiazolidinedione derivatives 8a and 8b. Removal of
the tert-butoxycarbonyl protecting group with acid
provided 2 and 3 as their HCl salts.
Assignment of the geometrical configuration of 2 and
3 was accomplished through a combination of NOE
experiments and X-ray crystallography. Initially, an
NOE experiment was carried out on 6b involving
selective irradiation of the two individual cyclobutane
methylene envelopes. By virtue of the symmetry within
6b, one envelope was assumed to be the axial-like pro-
tons on C-2 and C-4 and the other to be the equatorial-
like protons. As illustrated in Figure 1, enhancement
in the benzyl protons was observed upon irradiation of
the cyclobutane ring protons at 2.94-3.01 ppm, while
enhancements in the carbamate NH and methine 3-H
protons were observed upon irradiation of the cyclobu-
tane ring protons at 2.31-2.42 ppm. These results were
consistent with a geometry in which the tert-butoxycar-
bonylamino and the N-(benzyloxy)-N-(ethoxycarbonyl)
groups were in an E relationship to one another.
Subsequently, a crystal structure of 7a was achieved.
This compound was found to crystallize with two inde-

Cyclobutane Quisqualic Acid Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1641
F igu r e 4. Stimulation of PI hydrolysis by quisqualic acid
analogues in neonatal rat hippocampus. The effects of increas-
ing concentrations of 2 (O) or 3 (4) on PI hydrolysis were
measured. Data for 2 or 3 were best fit to curves having a Hill
coefficient of 1. The effect of a maximal concentration of
(1SR,3RS)-ACPD (300 µM), measured in parallel assays, is
shown for comparison. These data represent the mean ( SEM
of four experiments. Basal levels of PI hydrolysis in these
experiments were 12 270 ( 800 DPM/100 000 DPM incorpo-
rated. The effect of quisqualic acid (dotted line), measured in
previous experiments,22 is shown for comparison. Data for
quisqualic acid were best fit to curves having two components.
In two experiments on the effects of increasing concentrations
of quisqualic acid performed in parallel with the current
studies (data not shown), data were also best fit to two
components with EC50 values of 0.14 µM (82% of the total
response) and 24 µM (18% of the total response); the maximal
effect was 19 760 DPM/100 000 DPM incorporated.
100 000 DPM incorporated) were 26 700 ( 3 700 for 2
and 40 200 ( 11 600 for 3. At concentrations up to 1
mM, neither compound 2 nor 3 stimulated PI hydrolysis
in cells expressing mGluR1a (Figure 5D,E, open circles).
The effect of increasing concentrations of (RS)-2-
chloro-5-hydroxyphenylglycine (CHPG), a compound
that has been identified previously as a specific agonist
of mGluR5a relative to mGluR1a,25 also was examined
in cells expressing mGluR5a and was compared with
the results obtained for 2 and 3 (Figure 6). The EC50
value for CHPG was 400 ( 70 µM, and the maximal
effect was 22 000 ( 2 700 DPM/100 000 DPM incorpo-
rated.
To further explore the specificity of these compounds
for mGluR5a, the effects of 2 and 3 on the inhibition of
cAMP production stimulated by forskolin were exam-
ined in BHK 570 cells stably expressing either a group
2 (mGluR2) or group 3 (mGluR4a) metabotropic receptor
subtype. In addition, since several EAA analogues,
including quisqualic acid, block L-[3H]glutamic acid
uptake,26,27 the effects of 2 and 3 on L-[3H]glutamic acid
uptake into synaptosomal membrane preparations were
measured. The results of these studies are depicted in
Figure 7. L-Glutamic acid (100 µM) and (1SR,3RS)-
ACPD (100 µM) blocked cAMP formation in cells ex-
pressing mGluR2 (p < 0.001 for difference from control).
L-AP4 (10 µM), an agonist of group 3 metabotropic
receptors, blocked cAMP formation in cells expressing
mGluR4a (p < 0.001 for difference from control). Com-
pounds 2 and 3, however, were without significant effect
in either of these systems at concentrations up to 100
µM. Interestingly, CHPG (1 mM) reduced forskolin-
stimulated cAMP formation in cells expressing mGluR2
to 58 ( 8% of control (p < 0.01 for difference from
F igu r e 3. Effects of quisqualic acid analogues on PI hydroly-
sis in neonatal rat hippocampus and cerebellum. The effects
of L-glutamic acid, (1SR,3RS)-ACPD, quisqualic acid, 2, or 3
on PI hydrolysis were examined at the concentrations indi-
cated in neonatal rat hippocampus (top panel) or cerebellum
(bottom panel). Data represent the mean ( SEM of three ex-
periments performed in parallel. Basal levels of PI hydrolysis
(in DPM/100 000 DPM incorporated) were 6890 ( 490 in hip-
pocampus and 7170 ( 1310 in cerebellum. **p < 0.01 for dif-
ference from control; ***p < 0.001 for difference from control.
observation that compounds 2 and 3 activate PI hy-
drolysis in hippocampus but not in cerebellum suggests
that these compounds activate mGluR5. To further
explore this possibility, the effects of these compounds
on PI hydrolysis in BHK cells which express either
mGluR1a or mGluR5a were examined. In BHK cells
expressing mGluR1a (Figure 5A-C, open circles), EC50
values were 1.0 ( 0.4 µM for quisqualic acid, 5.9 ( 2.4
µM for L-glutamic acid, and 60 ( 20 µM for (1SR,3RS)-
ACPD; the maximal effects (in DPM/100 000 DPM
incorporated) were 14 200 ( 4 660 for quisqualic acid,
15 200 ( 2 730 for L-glutamic acid, and 12 170 ( 1 900
for (1SR,3RS)-ACPD (data not shown, n g 3). In BHK
cells expressing mGluR5a (Figure 5A-C, solid circles),
EC50 values were 0.03 ( 0.004 µM for quisqualic acid,
1.3 ( 0.3 µM for L-glutamic acid, and 12.5 ( 1.6 µM for
(1SR,3RS)-ACPD; the maximal effects (in DPM/100 000
DPM incorporated) were 36 830 ( 4 540 for quisqualic
acid, 31 800 ( 3 330 for L-glutamic acid, and 32 900 (
2 460 for (1SR,3RS)-ACPD. Compounds 2 and 3 stimu-
lated PI hydrolysis in cells expressing mGluR5a (Figure
5D,E, solid circles) with EC50 values of 11 ( 4 µM for 2
and 49 ( 25 µM for 3; the maximal effects (in DPM/

1642 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Littman et al.
F igu r e 5. Effects of quisqualic acid analogues on PI hydrolysis in BHK cells which express either mGluR1a or mGluR5a. The
effects of increasing concentrations of quisqualic acid (A), L-glutamic acid (B), (1SR,3RS)-ACPD (C), 2 (D), or 3 (E) on PI hydrolysis
were examined in cells which express either mGluR1a (open circles, right axis) or mGluR5a (filled circles, left axis). These data
represent the mean ( SEM of four to six experiments unless otherwise indicated. Due to the limited supply of 2 and 3, the effects
of a high concentration (1000 µM) of these compounds on cells expressing mGluR1a were only examined in one experiment. Basal
levels of PI hydrolysis (in DPM/100 000 DPM incorporated) were 7840 ( 2280 in cells expressing mGluR1a and 6090 ( 1030 in
cells expressing mGluR5a. The EC50 values (mean ( SEM) for each compound are given.
control, data not shown). In addition, compounds 2 and
3 did not block L-[3H]glutamic acid uptake in synapto-
somal membranes.
widely distributed in the brain, results in a sensitization
of neurons to depolarization by other excitatory amino
acid analogues, in particular amino acid phosphonates,
after exposure of the neurons to quisqualic acid.28-30 As
shown in Table 1, neither 2 nor 3 was able to induce
the QUIS effect.
Since quisqualic acid is also a prototypic agonist for
the ionotropic glutamate receptors now termed AMPA
receptors, analogues 2 and 3 were examined for their
effects on these receptors. As shown in Table 1, ana-
logues 2 and 3 are 10 and 58 times less potent,
respectively, than 1 in depolarizing rat hippocampal
CA1 pyramidal neurons, an effect which is presumably
mediated by their interaction with AMPA receptors.
Analogues 2 and 3 also were examined for their ability
to produce the QUIS effect.28-30 This effect, which is
Discu ssion
Quisqualic acid (1) affects a number of systems within
the CNS that are associated with excitatory amino acid
neurotransmission. It is a potent agonist at several
excitatory amino acid receptor subtypes including the
kainate,31 AMPA,32 and metabotropic receptors.33,34 It

Cyclobutane Quisqualic Acid Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1643
F igu r e 6. Stimulation of PI hydrolysis by quisqualic acid
analogues in BHK cells which express mGluR5a. The effects
of increasing concentrations of 2 (O), 3 (4), and (RS)-2-chloro-
5-hydroxyphenylglycine (0) on PI hydrolysis were measured.
These data represent the mean ( SEM of three to five
experiments. Basal levels of PI hydrolysis in these experiments
were 6880 ( 1450 DPM/100 000 DPM incorporated. The effect
of (1SR,3RS)-ACPD at 300 µM is shown for comparison.
also inhibits the enzyme N-acetyl-R-linked acidic dipep-
tidase which hydrolyzes the brain dipeptide N-acetyl-
L-aspartyl-L-glutamic acid35 and the Ca2+/Cl--dependent
glutamic acid uptake system in brain synaptic plasma
membrane preparations.36 Furthermore, quisqualic acid
produces a phenomenon known as the QUIS effect
which entails the sensitization of neurons to depolar-
ization by amino acid phosphonates such as AP4 after
exposure of the neurons to quisqualic acid.28-30
The present study demonstrates that appending the
three essential functionalities of quisqualic acid (the
amino and carboxy groups and the oxadiazolidinedione
ring) to a cyclobutane ring results in analogues of
quisqualic acid that show selective pharmacological
effects. Introduction of the cyclobutane constraint yielded
analogues of quisqualic acid, compounds 2 and 3, that
were found to stimulate PI hydrolysis in the hippocam-
pus but not in the cerebellum. In contrast, quisqualic
acid stimulated PI hydrolysis in both areas. Interest-
ingly, data for the stimulation of PI hydrolysis by
quisqualic acid were best fit to two components in
hippocampus but to only a single low-affinity component
in cerebellum. Since mGluR1a is expressed in both the
cerebellum and the hippocampus while mGluR5a is
expressed in the hippocampus but not in the cerebel-
lum,24 these results indicate that 2 and 3 are selective
for the group 1 metabotropic mGluR5a subtype of
receptor. This result was further substantiated by the
studies that were carried out with BHK cells which
expressed either mGluR1a or mGluR5a metabotropic
receptors. In these systems, quisqualic acid, L-glutamic
acid, and (1SR,3RS)-ACPD were 33-, 4.5-, and 4.8-fold
more potent for the stimulation of PI hydrolysis in cells
expressing mGluR5a than for the stimulation of PI
hydrolysis in cells expressing mGluR1a. Although 2 and
3 stimulated PI hydrolysis in cells expressing mGluR5a
with EC50 values of 11 and 49 µM, respectively, neither
2 nor 3, at concentrations up to 1000 µM, stimulated
PI hydrolysis in cells expressing mGluR1a. Due to the
limited supply of 2 and 3, the effects of higher concen-
trations of these compounds were not examined, but the
data from the current study suggests that the selectivi-
ties of these analogues for stimulation of mGluR5a
F igu r e 7. Effect of quisqualic acid analogues on cAMP
accumulation stimulated by forskolin and L-[3H]glutamic acid
uptake. The effects of L-glutamic acid, (1SR,3RS)-ACPD,
quisqualic acid, 2, 3, and L-2-amino-4-phosphonobutanoic acid
(L-AP4) on cAMP production stimulated by forskolin at the
concentrations indicated were examined in BHK cells which
expressed either mGluR2 (top panel) or mGluR4 (middle
panel). None of the quisqualic acid analogues stimulated cAMP
accumulation in the absence of forskolin (data not shown). The
effects of L-trans-pyrolidine-2,4-dicarboxylic acid (trans-PDC),
(1SR,3RS)-ACPD, quisqualic acid, 2, and 3 on L-[3H]glutamic
acid uptake in synaptosomes (bottom panel) were also exam-
ined at the concentrations indicated. These data represent the
mean ( SEM of three to four experiments. ***p < 0.001 for
difference from control.
compared to mGluR1a are at least 270-fold for 2 and
60-fold for 3. Compounds 2 and 3, therefore, show
greater selectivity than quisqualic acid in these systems.
In addition, both 2 and 3 were substantially more potent
than (RS)-2-chloro-5-hydroxyphenylglycine, a compound
which had previously been identified as a specific
agonist for mGluR5a metabotropic receptors.
The selectivity of 2 and 3 for the group 1 mGluR5a
metabotropic receptor was also demonstrated by the fact
that neither compound showed significant activity at the
group 2 mGluR2 metabotropic receptor and the group
3 mGluR4a metabotropic receptor. Although quisqualic
acid shows some effect in blocking L-[3H]glutamic acid
uptake into synaptosomal membranes and produces the
QUIS effect, 2 and 3 did not show any significant
activity in either of these systems.

1644 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Littman et al.
Ta ble 1. AMPA Receptor-Mediated Depolarization by Quisqualic Acid (1) and Analogues 2 and 3 and Sensitization of CA1 Pyramidal
Neurons to Depolarization by L-AP6a after Exposure to 1-3
IC50 for L-AP6 (mM) ( SD
QUIS effect conditions of exposure
depolarization
before
exposure
after
after reversal
compd
IC50 (µM)
concn (µM)
time (min)
exposure
with L-R-AAb
1
2
3
7.0 ( 0.8
69 ( 3
390 ( 140
16
200
400
4
20
20
>10
>10
>10
0.04 ( 0.008
7.0 ( 0.8
>10
1.9 ( 0.1
>10
nd
a
b
L-AP6, L-2-amino-6-phosphonohexanoic acid. Treatment with 2 mM L-R-aminoadipic acid (L-R-AA) for 10 min.
[3H]-myo-Inositol (15.4 Ci/mmol) was obtained from DuPont
Co. (Boston, MA). (1SR,3RS)-1-Amino-1,3-cyclopentanedicar-
boxylic acid, quisqualic acid, and L-2-amino-4-phosphonobu-
tanoic acid were obtained from Tocris Cookson (Bristol,
England). L-Glutamic acid was obtained from Sigma Chemical
Co. (St. Louis, MO).
Structurally, 2 and 3 can be viewed as analogues of
(S)-homoquisqualic acid (9) in which a methylene bridge
has been placed between the R- and γ-carbon atoms.
Previously, 9 was found to be a full agonist at group 1
metabotropic receptors with an EC50 of 50.2 ( 1.6 µM
in rat pup cerebrocortical slices.37 In comparison, in our
system 2 and 3 exhibited IC50 values of 18 ( 6 and 53
( 19 µM, respectively. It is also interesting to compare
(Z)- a n d (E)-Eth yl 1-[(ter t-Bu toxyca r bon yl)a m in o]-3-
[(b en zyloxy)a m in o]cyclob u t a n e-1-ca r b oxyla t e (5a a n d
5b). O-Benzylhydroxylamine (1.5 g, 10.0 mmol), ethyl 1-[(tert-
butoxycarbonyl)amino]-3-oxocyclobutane-1-carboxylate21 (4; 1.0
g, 3.89 mmol), and sodium cyanoborohydride (0.326 g, 5.18
mmol) were dissolved in 25 mL of anhydrous methanol. The
pH of the solution was brought to 5 by addition of HCl in
dioxane, and the reaction was stirred for 48 h at room
temperature. The pH of the solution then was brought to 8 by
addition of 1 M NaHCO3. The solvents were removed under
aspirator pressure. The crude product was partitioned between
EtOAc (50 mL) and H2O (25 mL). The organic layer was
washed with water and brine and dried. The product was
purified using flash column chromatography with hexanes/
EtOAc (3:1) as an eluant. The product was obtained as an oil
in a yield of 0.80 g (50%). Anal. (C19H28N2O5) C, H, N. HPLC
analysis (Waters Novapak silica cartridge column, 8 × 100
mm, hexanes/EtOAc (2:1), 4 mL/min) indicated the product
was a mixture of Z and E isomers in the ratio of 1:1 with
retention times of 18 and 20 min, respectively. The two isomers
were separated by HPLC under the condition described above.
Z Isom er 5a : 1H NMR (CDCl3) δ 7.35 (br s, 5 H, Ar-H),
5.77 (br s, 1 H, HNOBn), 5.29 (br s, 1 H, HNBoc), 4.71 (s, 2 H,
CH2Ph), 4.24 (q, J ) 6 Hz, 2 H, OCH2), 3.78 (dd, J ) 6 and 9
Hz, 1 H, 3-CH), 2.71-2.78 (m, 2 H, 2- and 4-CH2), 2.30-2.39
(m, 2 H, 2- and 4-CH2), 1.42 (s, 9 H, C(CH3)3) 1.26 (t, J ) 6
Hz, CH3); 13C NMR (CDCl3) δ 14.8 (CH3), 28.9 (C(CH3)3), 37.9
(2- and 4-C), 49.8 (CNOBn), 53.6 (1-C), 62.2 (OCH2), 77.0
(OCPh), 80.5 (OC(CH3)3), 128.5, 129.1, 138.5 (Ar-C), 155.3
(NCdO), 174.7 (CO2Et); FAB MS m/z 365 (100%) [M + H]+;
FAB HRMS m/z 365.2081 (C19H28N2O5 + H+ requires 365.1998).
E Isom er 5b: 1H NMR (CDCl3) δ 7.34 (br s, 5 H, Ar-H),
5.77 (br s, 1 H, HNOBn), 5.22 (br s, 1 H, HNBoc), 4.69 (s, 2 H,
CH2Ph), 4.17 (q, J ) 6 Hz, 2 H, OCH2), 3.83 (m, 1 H, 3-CH),
2.50-2.58 (m, 2 H, 2-and 4- CH2), 2.30-2.45 (m, 2 H, 2-and
4-CH2), 1.42 (s, 9 H, C(CH3)3) 1.21 (t, J ) 6 Hz, CH3); 13C NMR
the activity seen with 2 and 3 to that reported recently
with L-R-methylquisqualic acid (10).38 In contrast to 2
and 3, 10 was found to be a weak antagonist at the
group 1 metabotropic receptors. The IC50 values of 10
for mGluR1a and mGluR5a receptors were 500 and
1000 µM, respectively. Although 2 and 3, like 10, are
substituted at the R-carbon, the nature of this substitu-
ent clearly has a major impact on whether the com-
pound will be an agonist or antagonist.
Although both 2 and 3 possessed the ability to act as
agonists at the mGluR5a receptor, the isomer with the
amino and oxadiazolidinedione ring system in a syn
relationship was found to be the most potent, suggesting
that this conformationally constrained analogue comes
closest to mimicking the bioactive conformation needed
for the selective stimulation of the mGluR5a metabo-
tropic receptor.
Exp er im en ta l Section
(CDCl3) δ 14.8 (CH3), 28.9 (C(CH3) ), 37.1 (2- and 4-C), 50.8
3
(CNOBn), 54.6 (1-C), 62.2 (OCH2), 77.0 (CH2Ph), 80.7 (C(CH3)3),
128.5, 129.0, 138.5 (Ar-C), 155.8 (NCdO), 174.1 (CO2Et); FAB
MS m/z 365 [M + H]+.
Gen er a l. Melting points were determined on a Thomas-
Hoover Unimelt melting point apparatus model 6406-K and
are uncorrected. Elemental analyses were performed by M-
H-W Laboratories, Phoenix, AZ. Unless otherwise indicated,
all analytical results were within (0.4% of the theoretical
values. 1H NMR spectra were recorded on either a Bruker 200-
MHz or a GE 300-MHz spectrometer. The chemical shifts are
reported in parts per million (ppm) relative to tetramethylsi-
lane (TMS) in either CDCl3, MeOH-d4, or acetone-d6 and to
TSP in D2O. 13C NMR spectroscopy was performed on either
a Bruker 200-MHz or a GE 300-MHz spectrometer at 50 and
75 MHz, respectively. FAB mass spectra were obtained on a
Finnigan 4000 spectrometer. Column chromatography was
performed on silica gel (Merck, grade 60, 240-400 mesh, 60
Å) from Aldrich Chemical Co. Thin-layer chromatography
(TLC) was carried out on Analtech 250-mm silica gel GHLF
Uniplates. Visualization was achieved with either UV, I2, or
ninhydrin spray. THF was freshly distilled from sodium and
benzophenone ketyl. CH2Cl2 was distilled from sodium hy-
dride.
(Z)- a n d (E)-Eth yl 1-[(ter t-Bu toxyca r bon yl)a m in o]-3-
[N-(ben zyloxy)-N-(eth oxyca r bon yl)u r eid o]cyclobu ta n e-
1-ca r boxyla te (6a a n d 6b). Each isomer of 5 (0.10 g, 0.28
mmol) was dissolved in 5 mL of dry THF under N2, and
ethoxycarbonyl isocyanate (0.03 g, 0.28 mmol) was added after
which the reaction was stirred for 5 min. An additional 1 equiv
of ethoxycarbonyl isocyanate was added, and the reaction was
stirred for an additional 10 min. The solvent was removed on
a rotatory evaporator under aspirator pressure. The crude
product was purified using flash chromatography with hex-
anes/EtOAc (1:1) as the eluant.
Z Isom er 6a : obtained as an oil in a yield of 0.215 g (100%);
1H NMR (CDCl3) δ 7.79 (br s, 1 H, CONHCO), 7.41 (br s, 5 H,
Ar-H), 5.29 (br s, 1 H, HNBoc), 4.89 (s, 2 H, PhCH2O), 4.81
(t, 1 H, 3-CH), 4.07-4.26 (m, 4 H, OCH2), 2.78-2.84 (m, 4 H,
2- and 4-CH2), 1.42 (s, 9 H, C(CH3)3), 1.26 (t, 6 H, CH3); 13C
NMR (CDCl3) δ 14.8 and 15.2 (CH3), 28.9 (C(CH3)3), 37.6 (2-

Cyclobutane Quisqualic Acid Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1645
and 4-C), 49.0 (CNOBn), 53.3 (1-C), 61.7 and 62.4 (OCH2), 62.8
(OCH2Ph), 80.7 (OC(CH3)3), 129.8, 130.3, 134.5, 145.5 (Ar-
C), 155.3, 153.8, 151.4 (NCOO, NCON, and NCO2Et), 174.1
(CO2Et); FAB MS m/z 478 [M - H]-. Anal. (C23H33N3O8) C, H,
N.
E Isom er 6b: obtained as an oil in a yield of 0.115 g (100%);
1H NMR (CDCl3) δ 7.80 (br s, 1 H, CONHCO), 7.41 (br s, 5 H,
Ar-H), 5.05 (br s, 1 H, HNBoc), 5.01 (t, J ) 8.4 Hz, 1 H, 3-H),
4.89 (s, 2 H, PhCH2O), 4.13-4.26 (m, 4 H, OCH2), 2.94-3.01
(m, 2 H, 2- and 4-CH2), 2.31-2.44 (m, 2 H, 2- and 4-CH2), 1.43
(s, 9 H, C(CH3)3), 1.26 (t, 6 H, CH3); 13C NMR (CDCl3) δ 14.8
(CH3), 28.9 (C(CH3)3), 36.0 (2- and 4-C), 49.6 (CNOBn), 54.1
(1-C), 62.3 and 62.8 (OCH2), 63.5 (OCH2Ph), 81.0 (OC(CH3)3),
129.8, 130.2, 134.4 (Ar-C), 151.3, 153.9, 156.0 (NCOO, NCON,
and NCO2Et), 173.0 (CO2Et); FAB MS m/z 478 [M - H]-. Anal.
(C23H33N3O8) C, H, N.
(Z)- a n d (E)-Eth yl 1-[(ter t-Bu toxyca r bon yl)a m in o]-3-
[N-(h yd r oxy)-N-(eth oxyca r bon yl)u r eid o]cyclobu ta n e-1-
ca r boxyla te (7a a n d 7b). Each isomer of 6 (0.10 g, 0.22
mmol) was dissolved in 20 mL of EtOH, and this solution was
mixed with 0.05 g of 10% Pd/C. The solution was hydrogenated
at 30 psi for 3 h. The solution was then filtered through Celite,
and the solvent was removed on a rotatory evaporator under
aspirator pressure. The crude compound was purified by flash
chromatography with CH2Cl2/MeOH (20:1) as the eluant.
Z Isom er 7a : obtained as a solid in a yield of 0.17 g (90%);
1H NMR (CDCl3) δ 8.65 (s, 1 H, NOH), 8.13 (s, 1 H, CONHCO),
5.82 (s, 1 H, BocNH) 4.96-5.01 (m, 1 H, 3-CH), 4.19-4.31 (m,
4 H, OCH2), 3.02-3.16 (br m, 2 H, 2-and 4-CH2), 2.82-2.92
(m, 2 H, 2-and 4-CH2), 1.44 (s, 9 H, C(CH3)3), 1.27-1.35 (m, 6
H, CH3); 13C NMR (CDCl3) δ 14.1 and 14.2 (CH3), 28.2
(C(CH3)3), 36.2 (2-C and 4-C), 45.7 (CNOBn), 52.3 (1-C), 61.6
and 61.8 (OCH2), 79.9 (OC(CH3)3), 151.8, 153.1, 155.0 (NCOO,
NCON, and NCO2Et), 173.7 (CO2Et); FAB MS m/z 342 [M -
EtOH]+. Anal. (C16H27N3O8) C, H; N: calcd, 10.79; found, 10.31.
E Isom er 7b: obtained as a solid in a yield of 0.08 g (100%);
1H NMR (CDCl3) δ 9.04 (br s, 1 H, NOH), 8.63 (s, 1 H,
CONHCO), 5.62 (s, 1 H, BocNH), 4.96-5.10 (m, 1 H, 3-CH),
4.11-4.27 (m, 4 H, OCH2), 2.93-3.07 (m, 2 H, 2- and 4-CH2),
2.25-2.38 (m, 2 H, 2- and 4-CH2), 1.41 (s, 9 H, C(CH3)3), 1.21-
1.29 (m, 6 H, CH3); 13C NMR (CDCl3) δ 14.1 and 14.2 (CH3),
28.2 (C(CH3)3), 34.8 (2- and 4-C), 46.8 (3-C), 53.2 (1-C), 61.5
and 61.9 (OCH2), 80.1 (OC(CH3)3), 151.8, 153.5, 155.6 (NCOO,
NCON, and NCO2Et), 172.9 (CO2Et); FAB MS m/z 390 [M +
H]+. Anal. (C16H27N3O8) C, H, N.
The solvent and HCl were removed under vaccum. Water (5
mL) was added to the residue, and the solution was lyophilized.
Z Isom er 2‚HCl: obtained in an 80% yield; 1H NMR (D2O)
δ 4.94 (dd, J ) 7.5 and 9 Hz, 1 H, 3-CH), 3.60-3.79 (m, 2 H,
2- and 4-CH2), 2.87-3.03 (m, 2 H, 2- and 4-CH2); 13C NMR
(D2O) δ 43.1 (2-C and 4-C), 46.3 (3-C), 53.1 (1-C), 145.9, 157.2
(NCON, NCOO), 174.2 (CO2H); FAB MS m/z 216 [M + H]+.
Anal. (C7H10N3O5Cl) C, H, N.
E Isom er 3‚HCl: obtained in an 88% yield; 1H NMR (D2O)
δ 4.97 (t, J ) 6 Hz, 1 H, 3-CH), 3.19-3.27 (m, 2 H, 2- and
4-CH2), 2.65-2.74 (m, 2 H, 2- and 4-CH2); 13C NMR (D2O) δ
44.0 (2- and 4-C); 47.1 (3-C), 54.0 (1-C), 145.5, 156.8 (NCON,
NCOO), 174.1 (CO2H); FAB MS m/z 216 [M + H]+. Anal.
(C7H10N3O5Cl) C, H, N.
X-r a y Diffr a ction . Colorless platelike crystals of 7a were
grown by vapor diffusion from THF/pentane. All measure-
ments were made on the Siemens SMART platform CCD
diffractometer with graphite monochromated Mo KR radiation
(λ ) 0.71073 Å) using the hemisphere collector technique at
193(2) K. The structure was solved by direct methods and
refined by full matrix least-squares using SHELXTL-V5.0.39
All non-hydrogen atoms were refined with anisotropically
displacement parameters unless otherwise stated. All hydro-
gen atoms were placed in ideal positions and refined as riding
atoms with individual (or group if appropriate) isotropic
displacement parameters. The compound crystallized with two
independent molecules plus one THF molecule per asymmetric
unit. One showed normal crystallographic behavior, while the
other possessed disorder with rotation about both carbonyl
carbon to OEt bonds. Crystallographic data are provided as
Supporting Information.40
Cell Cu ltu r es. BHK570 cell lines stably expressing mGluR1a
and mGluR5a were generously provided by Novo Nordisk A/S
(Malov, Denmark).41 A BHK570 cell line stably expressing
mGluR2 was prepared by the methods of Chen and Okayama42
using a pZEM219b expression vector that was generously
donated by Zymogenetics, Inc. (Seattle, WA). BHK570 cells
stably expressing mGluR4a was generously provided by Zy-
mogenetics, Inc. BHK570 cells were grown in DMEM supple-
mented with 5% heat-inactivated fetal bovine serum, 1 µM
sodium pyruvate, 2 mM glutamine, 50 mg/mL penicillin, and
50 µg/mL streptomycin, in an atmosphere of 5% CO2. Stable
expression of metabotropic receptor subtypes was maintained
by inclusion of appropriate selection media: 1 µM methotrex-
ate and 0.5 mg/mL G418 (Geneticin) for mGluR1a or mGluR4a;
0.5 mg/mL G418 (Geneticin) for mGluR2; or 2 µM methotrex-
ate for mGluR5.
P h osp h oin ositid e Hyd r olysis. Phosphoinositide (PI) hy-
drolysis was measured in neonatal (6-11-day-old) rat hippo-
campal or cerebellar slices as previously described.43 Adult
Sprague-Dawley rats obtained from Charles River (Wilming-
ton, MA) were bred at Children’s Hospital of Philadelphia. For
measurement of PI hydrolysis in cell cultures, BHK cells were
grown to 60-90% confluence in 12-well tissue culture plates
as described above. Cells were treated for 16-20 h with
DMEM containing 1 µCi/mL [3H]-myo-inositol. PI hydrolysis
in cell cultures was measured using the method of Thomsen
et al.44 Measurements of cAMP formation, l-[3H]glutamic acid
uptake, and electrophysiological assays were performed as
previously described.27,43,45
The results from PI hydrolysis experiments were normalized
to 100 000 DPM of the total amount of radiolabel incorporated
(the amount of radiolabel in the lipid fraction plus the amount
of radiolabel in the soluble IP fraction) under basal conditions.
For adenylate cyclase experiments, data are presented as
fractions of forskolin control. Concentration-response curves
were analyzed using nonlinear regression analysis. Curves
were fit to the data from each experiment, and the EC50 values
and maximal stimulatory effects from each experiment were
used to determine the average EC50 values and maximal
stimulatory effect. All data presented are the mean ( SEM.
Statistical significance was determined using either Student’s
t-test or ANOVA with a Fisher PLSD post-hoc test. A p < 0.05
was considered significant.
(Z)- a n d (E)-1-[(ter t-Bu toxyca r bon yl)a m in o]-3-[2′-(3′,5′-
d ioxo-1′,2′,4′-oxa d ia zolid in yl)]cyclob u t a n e-1-ca r b oxyl-
ic Acid (8a a n d 8b). Each isomer of 7 (0.05 g, 0.12 mmol)
was dissolved in a mixture of 5 mL of THF and 5 mL of water.
Sodium hydroxide (0.014 g, 0.25 mmol) was added, and the
reaction was stirred at room temperature for 3 h. It then was
quenched by the addition of 3 mL of 10% citric acid solution.
The product was extracted with 3 × 25 mL of EtOAc. The
organic layer was dried over MgSO4 and the solvent then
removed under aspirator pressure to give product.
1
Z Isom er 8a : obtained in a 62% yield; H NMR (MeOH-
d4) δ 4.67-4.72 (m, 1 H, 3-H), 2.79-2.91 (m, 2 H, 2- and
4-CH2), 2.56-2.63 (m, 2 H, 2- and 4-CH2), 1.46 (s, 9 H,
C(CH3)3); 13C NMR (MeOH-d4) δ 27.8 (C(CH3)3), 37.1 (2- and
4-C), 43.1 (CNOBn), 52.5 (1-C), 79.9 (OC(CH3)3), 152.9 (NCO),
157.2 (NCON), 172.9 (CO2H); FAB MS m/z 480 [M + MNBA
matrix]+. Anal. (C12H17N3O7) C, H, N.
E Isom er 8b: obtained in a yield of 0.03 g (85%); 1H NMR
(acetone-d6) δ 4.67-4.72 (m, 1 H, 3-CH), 2.86-3.00 (m, 2 H,
2- and 4-CH2), 2.45-2.52 (m, 2 H, 2- and 4-CH2), 1.42 (s, 9 H,
C(CH3)3); 13C NMR (acetone-d6) δ 28.3 (C(CH3)3), 35.8 (2- and
4-C), 43.0 (3-C), 49.6 (1-C), 79.9 (OC(CH3)3), 152.1 (NCO),
156.2, 156.6 (NCON, NCOO), 171.6 (CO2H); FAB MS m/z 314
[M - H]-. Anal. (C12H17N3O7) C, H, N.
(Z)- a n d (E)-1-Am in o-3-[2′-(3′,5′-d ioxo-1′,2′,4′-oxa d ia zo-
lid in yl)]cyclo-bu ta n e-1-ca r boxylic Acid Hyd r och lor id e
(2•HCl a n d 3•HCl). Each isomer of 8 (0.075 g, 0.24 mmol)
was dissolved in 2 mL of THF under Ar, and 0.1 mL of 4 N
HCl in dioxane was added. The solution was stirred for 1 h.

1646 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Littman et al.
(15) Koh, J .-Y.; Palmer, E.; Cotman, C. W. Activation of the Metabo-
tropic Glutamate Receptor Attenuates N-Methyl-D-aspartate
Neurotoxicity in Cortical Cultures. Proc. Natl. Acad. Sci. U.S.A.
1991, 88, 9431-9435.
(16) Pizzi, M.; Fallacara, C.; Arrighi, V.; Memo, M.; Spano, P.
Attenuation of Excitatory Amino Acid Toxicity by Metabotropic
Glutamate Receptor Agonists and Aniracetam in Primary
Cultures of Cerebellar Granule Cells. J . Neurochem. 1993, 61,
683-689.
(17) Herrero, I.; Miras-Portugal, M.; Sanchez-Prieto, J . Positive
Feedback of Glutamate Exocytosis by Metabotropic Presynaptic
Receptor Stimulation. Nature 1992, 360, 163-166.
(18) Collins, G. G. S. Actions of Agonists of Metabotropic Glutamate
Receptors on Synaptic Transmission and Transmitter Release
in the Olfactory Cortex. Br. J . Pharmacol. 1993, 108, 422-430.
(19) Baldwin, J . E.; Adlington, R. M.; Birch, D. J . Synthesis of
L-Quisqualic Acid: A General Method for Enanatio-efficient
Synthesis of â-Aminoalanine Derivatives. J . Chem. Soc., Chem.
Commun. 1985, 256-257.
(20) Bycroft, B. W.; Chhabra, S. R.; Grout, R. J .; Crowley, P. J . A
Convenient Synthesis of the Neuroexcitatory Amino Acid
Quisqualic Acid and Its Analogues. J . Chem. Soc., Chem.
Commun. 1984, 1156-1157.
(21) Allan, D. R.; Hanrahan, R. J .; Hambley, T. W.; J ohnston, G. A.
R.; Mewett, K. N.; Mitrovic, A. D. Synthesis and Activity of a
Potent N-Methyl-D-Aspartic Acid Agonist, trans-1-Aminocy-
clobutane-1,3-dicarboxylic Acid, and Related Phosphonic and
Carboxylic Acids. J . Med. Chem. 1990, 33, 2905-2915.
(22) Littman, L.; Munir, M.; Flagg, S. D.; Robinson, M. B. Multiple
Mechanisms for Inhibition of Excitatory Amino Acid Receptors
Coupled to Phosphoinositide Hydrolysis. J . Neurochem. 1992,
59, 1893-1904.
AMP A Recep tor -Med ia ted Dep ola r iza tion a n d QUIS
Effect. The procedures for AMPA receptor-mediated depolar-
ization and its QUIS effect assay were conducted on rat
hippocampal slices in the same manner as reported previ-
ously.46
Ack n ow led gm en t. This work was supported by the
following grants from the National Institutes of
Health: NS35608 (R.L.J .), NS29868 (M.B.R.), and
GM34781 (M.B.R.). The authors thank Andy Goldfine
for technical assistance and Victor G. Young, J r., for the
X-ray crystallographic analysis. The authors also thank
Betty Haldeman of Zymogenetics for cell lines express-
ing mGluR1a, mGluR2, and mGluR4a and Christian
Thomsen of Novo Nordisk for the cell line expressing
mGluR5.
Su p p or tin g In for m a tion Ava ila ble: X-ray crystallo-
graphic data including tables of positional parameters, bond
distances, and bond angles for 7a . This information is available
free of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Monaghan, D. T.; Bridges, R. J .; Cotman, C. W. The Excitatory
Amino Acid Receptors: Their Classes, Pharmacology, and
Distinct Properties in the Function of the Central Nervous
System. Annu. Rev. Pharmacol. Toxicol. 1989, 29, 365-402.
(2) Choi, D. W. Glutamate Neurotoxicity and Diseases of the
Nervous System. Neuron 1988, 1, 623-634.
(23) Littman, L.; Robinson, M. B. The Effects of L-Glutamate and
trans-(()-1-Amino-1,3-cyclopentanedicarboxylate on Phosphoi-
nositide Hydrolysis Can be Pharmacologically Differentiated. J .
Neurochem. 1994, 63, 1291-1302.
(3) McDonald, J . W.; J ohnston, M. V. Physiological and Pathophysi-
ological Roles of Excitatory Amino Acid During Central Nervous
Systems Development. Brain Res. Rev. 1990, 15, 41-70.
(4) Schoepp, D.; Bockaert, J .; Sladeczek, F. Pharmacological and
Functional Characteristics of Metabotropic Excitatory Amino
Acid Receptors. Trends Pharmacol. Sci. 1990, 11, 508-515.
(5) Nakanishi, S. Molecular Diversity of Glutamate Receptors and
Implications for Brain Function. Science 1992, 258, 597-603.
(6) Abe, T.; Sugihara, H.; Nawa, H.; Shigemoto, R.; Mizuno, N.;
Nakanishi, S. Molecular Characterization of a Novel Metabo-
tropic Glutamate Receptor mGluR5 Coupled to Inositol Phosphate/
Ca2+ Signal Transduction. J . Biol. Chem. 1992, 267, 13361-
13368.
(7) Thomsen, C.; Mulvihill, E. R.; Haldeman, B.; Pickering, D. S.;
Hampson, D. R.; Suzdak, P. D. A Pharmacological Characteriza-
tion of the mGluR1a Subtype of the Metabotropic Glutamate
Receptor Expressed in a Cloned Baby Hamster Kidney Cell Line.
Brain Res. 1993, 619, 22-28.
(8) Monn, J . A.; Valli, M. J .; Massey, S. M.; Wright, R. A.; Salhoff,
C. R.; J ohnson, B. G.; Howe, T.; Alt, C. A.; Rhodes, G. A.; Robey,
R. L.; Griffey, K. R.; Tizzano, J . P.; Kallman, M. J .; Helton, D.
R.; Schoepp, D. D. Design, Synthesis, and Pharmacological
Characterization of (+)-2-Aminobicyclo[3.1.0]hexane-2,6-dicar-
boxylic Acid (LY354740): A Potent, Selective, and Orally Active
Group 2 Metabotropic Glutamate Receptor Agonist Possessing
Anticonvulsant and Anxiolytic Properties. J . Med. Chem. 1997,
40, 428-537.
(9) Ishida, M.; Saitoh, T.; Shimamoto, K.; Ohfune, Y.; Shinozaki,
H. A Novel Metabotropic Glutamate Receptor Agonist: Marked
Depression of Monosynaptic Excitation in the Newborn Rat
Isolated Spinal Cord. Br. J . Pharmacol. 1993, 109, 1169-1177.
(10) Thomsen, C.; Kristensen, P.; Mulvihill, E.; Haldeman, B.;
Suzdak, P. D. L-2-Amino-4-phosphonobutyrate (L-AP4) is an
Agonist at the Type IV Metabotropic Glutamate Receptor which
is Negatively Coupled to Adenylate Cyclase. Eur. J . Pharmacol.
1992, 227, 361-362.
(11) Nicoletti, F.; Iadarola, M. J .; Wroblewski, J . T.; Costa, E.
Excitatory Amino Acid Recognition Sites Coupled with Inositol
Phospholipid Metabolism: Developmental Changes and Interac-
tion with R1-Adrenoceptors. Proc. Natl. Acad. Sci. U.S.A. 1986,
83, 1931-1935.
(24) Casabona, G.; Knopfel, T.; Kuhn, R.; Gasparini, F.; Baumann,
P.; Sortino, M. A.; Copani, A.; Nicoletti, F. Expression and
Coupling to Polyphosphoinositide Hydrolysis of Group I Me-
tabotropic Glutamate Receptors in Early Postnatal and Adult
Rat Brain. Eur. J . Neurosci. 1997, 9, 12-17.
(25) Doherty, A. J .; Palmer, M. J .; Henley, J . M.; Collingridge, G. L.;
J ane, D. E. (RS)-2-Chloro-5-hydroxyphenylglycine (CHPG) Ac-
tivates mGluR5, but not mGluR1 Receptors Expressed in CHO
Cells and Potentiates NMDA Responses in the Hippocampus.
Neuropharmacology 1997, 36, 265-267.
(26) Roberts, P. J .; Watkins, J . C. Structural Requirements for the
Inhibition of L-Glutamate Uptake by Glia and Nerve Endings.
Brain Res. 1975, 85, 120-125.
(27) Robinson, M. B.; Sinor, J . D.; Dowd, L. A.; Kerwin, J . F., J r.
Subtypes of Sodium-dependent High-affinity L-[3H]Glutamate
Transport Activity: Pharmacologic Specificity and Regulation
by Sodium and Potassium. J . Neurochem. 1993, 60, 167-179.
(28) Robinson, M. B.; Whittemore, E. R.; Marks, R. L.; Koerner, J .
F. Exposure of Hippocampal Slices to Quisqualate Sensitizes
Synaptic Responses to Phosphonate-Containing Analogues of
Glutamate. Brain Res. 1986, 381, 187-190.
(29) Whittemore, E. R.; Koerner, J . F. Novel Recognition Site for
L-Quisqualate Sensitizes Neurons to Depolarization by L-2-
Amino-4-phosphonobutanoate (L-AP4). Brain Res. 1989, 489,
146-156.
(30) Schulte, M. K.; Roon, R. J .; Koerner, J . F. Quisqualic Acid
Induced Sensitization and the Active Uptake of L-Quisqualic
Acid by Hippocampal Slices. Brain Res. 1993, 605, 85-92.
(31) London, E. D.; Coyle, J . T. Specific Binding of 3H-Kainic Acid to
Receptor Sites in Rat Brain. Mol. Pharmacol. 1979, 15, 492-
505.
(32) Honore, T.; Lauridsen, J .; Krogsgaard-Larsen, P. The Binding
of [3H]AMPA, a Structural Analogue of Glutamic Acid, to Rat
Brain Membranes. J . Neurochem. 1982, 38, 173-178.
(33) Schoepp, D. D.; J ohnson, B. G. Excitatory Amino Acid Agonist-
Antagonist Interactions at 2-Amino-4-phosphonobutyric Acid-
Sensitive Quisqualate Receptors Coupled to Phosphoinositide
Hydrolysis in Slices of Rat Hippocampus. J . Neurochem. 1988,
50, 1605-1613.
(34) Recasens, M.; Guiramand, J .; Nourigat, A.; Sassetti, I.; Devil-
liers, G. A New Quisqualate Receptor Subtype (sAA2) Respon-
sible for the Glutamate-induced Inositol Phosphate Formation
in Rat Brain Synaptoneurosomes. Neurochem. Int. 1988, 13,
463-467.
(35) Robinson, M. B.; Blakely, R. D.; Couto, R.; Coyle, J . T. Hydrolysis
of the Brain Dipeptide N-Acetyl-L-aspartyl-L-glutamate. J . Biol.
Chem. 1987, 262, 14498-14506.
(36) Zaczek, R.; Arlis, S.; Markl, A.; Murphy, T.; Drucker, H.; Coyle,
J . T. Characteristics of Chloride-Dependent Incorporation of
Glutamate into Brain Membranes Argue Against a Receptor
Binding Site. Neuropharmacology 1987, 26, 281-287.
(12) Palmer, E.; Nangel-Taylor, K.; Krause, J . D.; Roxes, A.; Cotman,
C. W. Changes in Excitatory Amino Acid Modulation of Phos-
phoinositide Metabolism During Development. Dev. Brain Res.
1990, 51, 132-134.
(13) Izumi, Y.; Clifford, D. B.; Zorumski, C. F. 2-Amino-3-phospho-
nopropionate Blocks the Induction and Maintenance of Long-
term Potentiation in Rat Hippocampal Slices. Neurosci. Lett.
1991, 122, 187-190.
(14) Zheng, F.; Gallagher, J . P. Metabotropic Glutamate Receptors
are Required for the Induction of Long-term Potentiation.
Neuron 1992, 9, 163-172.

Cyclobutane Quisqualic Acid Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1647
(37) Porter, R. H. P.; Roberts, P. J .; J ane, D. E.; Watkins, J . C. (S)-
Homoquisqualate: a Potent Agonist at the Glutamate Metabo-
tropic Receptor. Br. J . Pharmacol. 1992, 106, 509-510.
(38) Kozikowski, A. P.; Steensma, D.; Varasi, M.; Pshenichkin, S.;
Surina, E.; Wroblewski, J . T. R-Substituted Quisqualic Acid Ana-
logues: New Metabotropic Glutamate Receptor Group II Selec-
tive Antagonists. Bioorg. Med. Chem. Lett. 1998, 8, 447-452.
(39) SHELXTL-Plus V5.0, Siemens Industrial Automation, Inc.,
Madison, WI.
(40) The authors have deposited atomic coordinates for this structure
with the Cambridge Crystallographic Data Centre. The structure
has been given the deposition number CCDC 116965. The
coordinates can be obtained, upon request, from the Director,
Cambridge Crystallographic Data Centre, 12 Union Rd, Cam-
bridge CB2 1EZ, U.K.
(43) Littman, L.; Chase, L. A.; Renzi, M.; Garlin, A. B.; Koerner, J .
F.; J ohnson, R. L.; Robinson, M. B. Effects of Quisqualic Acid
Analogues on Metabotropic Glutamate Receptors Coupled to
Phosphoinositide Hydrolysis in Rat Hippocampus. Neurophar-
macology 1995, 34, 829-841.
(44) Thomsen, C.; Boel, E.; Suzdak, P. D. Actions of Phenylglycine
Analogues at Subtypes of the Metabotropic Glutamate Receptor
Family. Eur. J . Pharmacol. 1994, 267, 77-84.
(45) J ohansen, P. A.; Chase, L. A.; Sinor, A. D.; Koerner, J . F.;
J ohnson, R. L.; Robinson, M. B. Type 4a Metabotropic Glutamate
Receptor: Identification of New Potent Agonists and Differentia-
tion from the L-(+)-2-Amino-4-phosphonobutanoic Acid-Sensitive
Receptor in the Lateral Perforant Pathway in Rats. Mol.
Pharmacol. 1995, 48, 140-149.
(46) Venkatraman, S.; Roon, R. J .; Schulte, M. K.; Koerner, J . F.;
J ohnson, R. L. Synthesis of Oxadiazolidinedione Derivaties as
Quisqualic Acid Analogues and Their Evaluation at a Quisqualate
Sensitized Site in the Rat Hippocampus. J . Med. Chem. 1994,
37, 3939-3946.
(41) Pellicciari, R.; Raimondo, M.; Marinozzi, M.; Natalini, B.;
Costantino, G.; Thomsen, C. (S)-(+)-2-(3′-Carboxybicyclo[1.1.1]-
pentyl)glycine,
a Structurally New Group I Metabotropic
Glutamate Receptor Antagonist. J . Med. Chem. 1996, 39, 2874-
2876.
(42) Chen, C. A.; Okayama, H. Calcium Phosphate-mediated Gene
Transfer: A Highly Efficient Transfection System for Stably
Transforming Cells with Plasmid DNA. Biotechniques 1988, 6,
623-638.
J M9806897