Chemo-enzymatic synthesis of a series of 2,4-syn-functionalized (S)-glutamate analogues: New insight into the structure-activity relation of ionotropic glutamate receptor subtypes 5, 6, and 7

Article abstract of DOI:10.1021/jm800092x

(S)-Glutamic acid (Glu) is the major excitatory neurotransmitter in the central nervous system (CNS) activating the plethora of ionotropic Glu receptors (iGluRs) and metabotropic Glu receptors (mGluRs). In this paper, we present a chemo-enzymatic strategy for the enantioselective synthesis of five new Glu analogues 2a-f (2d is exempt) holding a functionalized substituent in the 4-position. Nine Glu analogues 2a-j are characterized pharmacologically at native 2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid (AMPA), kainic acid (KA), and N-methyl-D-aspartic acid (NMDA) receptors in rat synaptosomes as well as in binding assays at cloned rat iGluR5-7 subtypes. A detailed in silico study address as to why 2h is a high-affinity ligand at iGluR57 (K_i = 3.81, 123, 57.3 nM, respectively), while 2e is only a high affinity ligand at iGluR5 (K_i = 42.8 nM). Furthermore, a small series of commercially available iGluR ligands are characterized in iGluR5-7 binding.

Full text of DOI:10.1021/jm800092x

J. Med. Chem. 2008, 51, 4093–4103
4093
Chemo-Enzymatic Synthesis of a Series of 2,4-Syn-Functionalized (S)-Glutamate Analogues:
New Insight into the Structure-Activity Relation of Ionotropic Glutamate Receptor Subtypes 5,
6, and 7
Emanuelle Sagot,‡ Darryl S. Pickering,§ Xiaosui Pu,| Michelle Umberti,| Tine B. Stensbøl, Birgitte Nielsen,† Marion Chapelet,‡
Jean Bolte,‡ Thierry Gefflaut,*,‡ and Lennart Bunch*,†
Department of Medicinal Chemistry, The Faculty of Pharmaceutical Sciences, UniVersity of Copenhagen, UniVersitetsparken 2, DK-2100
Copenhagen, Denmark, De´partement de Chimie, UniVersite´ Blaise Pascal, 63177 Aubie`re Cedex, France, Department of Pharmacology and
Pharmacotherapy, The Faculty of Pharmaceutical Sciences, UniVersity of Copenhagen, UniVersitetsparken 2, DK-2100 Copenhagen, Denmark,
Biological Research, H. Lundbeck A/S, 215 College Road, Paramus, New Jersey 07652, Department of Neurobiology, H. Lundbeck A/S,
OttiliaVej 9, DK-2500 Valby, Denmark
ReceiVed January 30, 2008
(S)-Glutamic acid (Glu) is the major excitatory neurotransmitter in the central nervous system (CNS) activating
the plethora of ionotropic Glu receptors (iGluRs) and metabotropic Glu receptors (mGluRs). In this paper,
we present a chemo-enzymatic strategy for the enantioselective synthesis of five new Glu analogues 2a-f
(2d is exempt) holding a functionalized substituent in the 4-position. Nine Glu analogues 2a-j are
characterized pharmacologically at native 2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid
(AMPA), kainic acid (KA), and N-methyl-D-aspartic acid (NMDA) receptors in rat synaptosomes as well as
in binding assays at cloned rat iGluR5-7 subtypes. A detailed in silico study address as to why 2h is a
high-affinity ligand at iGluR5-7 (Ki ) 3.81, 123, 57.3 nM, respectively), while 2e is only a high affinity
ligand at iGluR5 (Ki ) 42.8 nM). Furthermore, a small series of commercially available iGluR ligands are
characterized in iGluR5-7 binding.
Introduction
mGluR1-8, which are grouped with respect to the second
messenger system involved with receptor signaling, pharmacol-
ogy, and molecular biology (group I: mGluR1,5; group II:
mGluR2,3; group III: mGluR4,6-8).5
To study a specific iGluR subtype, when functioning in its
intrinsic biological environment, the employment of iGluR
subtype selective ligands (agonists, partial agonists, and an-
tagonists) is a frequently applied strategy. In this paper, we
present the synthetic strategy which allowed the preparation of
Glu analogues 2a-f (2d is exempt), and the pharmacological
evaluation of the nine 2,4-syn-4-substituted Glu analogues 2a-j
at iGluRs. New insight into the structure-activity relation for
KA receptor subtypes iGluR5-7 is presented.
(S)-Glutamic acid (Glu) is the major excitatory neurotrans-
mitter in the central nervous system (CNS) activating the
plethora of ionotropic Glu receptors (iGluRs)a and metabotropic
Glu receptors (mGluRs).1,2 While the iGluRs are ion channels
and thus mediate a fast excitatory response (Na+, K+, Ca2+
flux), the mGluRs are classified as G-protein coupled receptors
and produce a slower signal transduction through second
messenger systems. Termination of the excitatory signal is
controlled by uptake of Glu from the synaptic cleft by the
excitatory amino acid transporters (EAATs).3 On the basis of
pharmacological studies, the iGluRs are further divided into:
2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid
(AMPA) receptors (homo- or heteromeric receptors comprising
the subunits iGluR1-4), kainic acid (KA) receptors (homo- or
heteromeric receptors comprising the subunits iGluR5-7 and
KA1,2),4 and the N-methyl-D-aspartic acid (NMDA) receptors
(heteromeric receptors comprising the subunits NR1, 2A-D, 3A-
C). The mGluRs are divided into eight homodimeric subtypes,
Chemistry
Glu analogues 2a-j were all prepared following a chemoen-
zymatic route involving a stereoselective transamination reaction
as the key step (Scheme 1). Aspartate aminotransferase (AAT)
from Escherichia coli was shown to be a very effective catalyst
for the conversion of a variety of substituted R-ketoglutarates
(KG) into L-Glu analogues. A very high stereoselectivity was
observed in favor of the L-2,4-syn Glu analogues obtained
through the kinetic resolution of racemic KGs.6
Synthesis of 2-Oxoglutaric Acids. Two synthetic approaches
were developed for the preparation of the oxoglutarates 1a-j.
Substrates 1g-j bearing the amide functionality were prepared
via a short three-step procedure from the readily available
dimethyl 4-oxocyclohexane-1,3-dicarboxylate.7 However, this
method was not feasible for introduction of the functionalities
present in 1a-f.
* Author to whom correspondence should be addressed. For L.B.: phone,
+45 35336244; fax, +45 35336040; E-mail, lebu@farma.ku.dk. For T.G.
(for correspondence regarding the synthetic work): phone, +33 473407866;
fax, +33 473407717; E-mail, thierry.gefflaut@univ-bpclermont.fr.
‡
De´partement de Chimie, Universite´ Blaise Pascal.
Department of Pharmacology and Pharmacotherapy, The Faculty of
§
Pharmaceutical Sciences, University of Copenhagen.
|
Biological Research, H. Lundbeck A/S.
Department of Neurobiology, H. Lundbeck A/S.
Department of Medicinal Chemistry, The Faculty of Pharmaceutical
†
Sciences, University of Copenhagen.
a Abbreviations: iGluR, ionotropic Glu receptors; mGluR, metabotropic
Glu receptors; EAAT, excitatory amino acid transporter; KG, R-ketoglut-
arates; AAT, aspartate aminotransferase; DHK, dihydrokainic acid; TBOA,
threo-benzyloxyaspartate; AMPA, 2-amino-3-(3-hydroxy-5-methyl-4-isox-
azolyl)propionic acid; KA, kainic acid; NMDA, N-methyl-D-aspartic acid;
FMP, FLIPR membrane potential; DOMO, domoric acid; QUIS, quisqualic
acid.
As shown in Scheme 2, the syntheses of 1a-c were
performed by following a recently described procedure:8 the
glutarate carbon skeleton was obtained by a Michael reaction
between methyl acetoacetate and an R-substituted acrylate,
10.1021/jm800092x CCC: $40.75
2008 American Chemical Society
Published on Web 06/25/2008

4094 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
Sagot et al.
Scheme 1. AAT Catalyzed Synthesis of 4-Substituted Glu
Scheme 3. Synthesis of 4-Hydroxymethyl KG 1a
Analogues
Scheme 4. Synthesis of 1d and 1e
Scheme 2. Synthesis of 4-Substituted KG 1a-c
Scheme 5. Synthesis of Acrylates 3d and 3e
whereas ozonolysis of an enol acetate intermediate afforded the
R-keto ester moiety. This strategy, already used for the synthesis
of 3-alkyl KGs,8 appeared very general and compatible with
the functionalities of 1a-c.
increased to 43%. The benzyl group of 6b was cleaved by
hydrogenolysis to furnish 6a in quantitative yield. Finally, the
lithium salts 1a-c were prepared quantitatively from 6a-c by
basic hydrolysis using a stoichiometric amount of LiOH.
Interestingly, NMR analyses indicated that both compounds 6a
and 1a exist in solution as a 1:1 mixture of hemiketal isomers,
coming from the addition of the hydroxyl group on the carbonyle
at position 2 (Scheme 3).
The synthesis of ketoglutarates 1d and 1e bearing an ester
functionality was performed following a slightly different
procedure (Scheme 4): the Michael condensation of 3d,e with
benzylacetoacetate was done in DMSO in order to avoid the
transesterification reaction occurring when MeOH was used as
the solvent. The ketoesters 6d and 6e were thus obtained from
3d and 3e in 40% overall yield. Finally, the methyl ester
functionality of 1d and 1e could be preserved through the
selective hydrogenolysis of the benzyl esters of 6c and 6d in
the presence of Pd/C.
2-(Benzyloxymethyl)acrylate (3b) was prepared from methyl
(bromomethyl)acrylate9 by reaction with benzyl alcohol in the
presence of base. Among the several bases tested, 1,4-
diazabicyclo[2.2.2]octane (DABCO) gave the best results for
the preparation of 3b, which was isolated in 80% yield. The
Michael condensations of 3b,c with methyl acetoacetate were
done in mild conditions using KF as the base in MeOH. The
Michael products 4b and 4c were both isolated as 1:1 mixtures
of diastereomers and in 95% and 78% yields, respectively. These
ꢀ-keto esters were then treated with acetyl chloride and pyridine
to form the enol acetates 5b and 5c isolated after flash
chromatography in 64% and 86% yields, respectively. Finally,
these intermediates were oxidized by ozone to form the
oxoglutarates 6b and 6c. A yield of 72% was obtained for 6c.
Unfortunately, using the same ozonolysis procedure, the oxo-
glutarate 6b was isolated with a low yield of 17%. This result
was mainly due to an unexpected oxidation of the benzyl ether
group to a benzoate isolated as the major byproduct in 45%
yield. The oxidative cleavage of 5b was tried using NaIO4 and
catalytic RuO2 in biphasic conditions using a CHCl3/CH3CN/
H2O mixture. However, hydrolysis of the enol acetate 5b was
observed in these conditions and the yield of 6b could only be
As shown in Scheme 5, the unsymmetrical diester 3d was
obtained from itaconic anhydride: regioselective addition of
MeOH on the nonconjugated carbonyl of the anhydride allowed
the formation of itaconic acid monoester at position 1. Consecu-
tiveesterificationwithbenzylbromideand1,8-diazabicyclo[5.4.0]undec-
7-ene (DBU) afforded 1-methyl-4-benzyl itaconate 3d isolated
in 95% overall yield. The methyl ester formation was highly

Synthesis of 2,4-Syn-Functionalized (S)-Glutamate Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4095
Scheme 6. Instant Cyclization of 2d into Lactam 7
Table 1. Kinetic Parameters of AAT Catalyzed Transaminations with
KG Analogues 1a-ea
and synthesis of 1b behaving as a very good substrate of AAT.
The same strategy was applied for the tricarboxylic acid 2c:
after the transamination reaction involving the ester 1d, the crude
product was refluxed with 6N HCl to hydrolyze the ester group
of 2d. The Glu analogue 2c was then purified by ion exchange
chromatography as described for 2b and isolated in 41% yield
from rac-1d. All attempts to isolate 2d were unsuccessful due
to its rapid base catalyzed cyclization into lactame 7. NMR
analysis of the cyclic derivative 7 allowed the confirmation of
the pseudoaxial position of H4 and H6, i.e., a 2,4-syn configu-
ration in 2c and 2d, which is in agreement with the previously
observed stereoselectivity of AAT toward 4-substituted Glu
analogues (Scheme 6).
The tricarboxylic Glu analogue 2f was prepared from 2e
following the procedure described above for 2c: HCl catalyzed
hydrolysis of 2e before purification by ion exchange chroma-
tography afforded 2f in 40% yield. Contrary to 2d, the Glu
analogue 2e bearing the methyl ester functionality could be
isolated because the cyclization giving a seven-membered ring
lactame was not favored in this case. 2e was isolated in 39%
yield after purification on the cationic dowex 1 resin. In this
case, the first chromatographic step on dowex 50 was omitted
in order to avoid the ester hydrolysis during elution with aqueous
ammonia.
a Values and standard errors were calculated from the Hanes-Woolf
plot according to the least-squares method and Gauss’s error propagation
law. b The absolute kcat value measured in our experimental conditions was
39.6 ( 1.2 s-1
.
regioselective, giving a 97:3 mixture of 3d and 1-benzyl-4-
methyl itaconate. Acrylate 3e was prepared in 73% yield from
benzyl R-(hydroxymethyl)acrylate10 via the Claisen-Johnson
rearrangement of the ketene ketal formed in the presence of
trimethyl orthoacetate.11
Transamination Reactions. The substituted KGs 1a-e were
evaluated as substrate for E. coli AAT on the basis of the
Michaelis-Menten model: rate measurement of the AAT
catalyzed transaminations between aspartate (40 mM) and 1a-e
in variable concentrations allowed estimation of the kinetic
parameters reported in Table 1.
These results offer a new highlight on the substrate specificity
of AAT: the parameters measured for 1a and, even more, for
1c clearly indicate that the introduction of a polar group at
position 4 of KG results in a dramatic drop in affinity as well
as in kcat value. The low activity of AAT toward 1a could be
the result of the cyclization mentioned above, hiding the
carbonyl group of this substrate. The low affinity of AAT toward
substrate, which contain polar functional groups, was also
noticed with the amide derivatives precursors of 2 g-j.7
However, when the alcohol and carboxylic acid functionalities
are masked as benzyl ether or methyl ester in compounds 1b,
1d, and 1e, the activity is restored. These results are consistent
with the previous finding that the AAT active site accommodates
various hydrophobic groups substituting the pro-R hydrogen
atom at position 4 of KG.6
Preparative scale transaminations were easily carried out in
the case of the 3 protected derivatives 1b, 1d, and 1e following
the procedure described previously:6,8 cysteine sulfinic acid
(CSA) was used as an irreversible amino donor, and the reaction
was stopped near 40% conversion. Therefore, the kinetic
resolution of the racemic substrates 1b,d,e was performed,
resulting in the stereoselective formation of L-2,4-syn Glu
analogues 2b,d,e. The Glu analogue 2b was isolated after
selective adsorption on sulfonic dowex 50 resin (H+ form) and
elution with aqueous ammonia. It was further purified by ion
exchange chromatography on cationic dowex 1 resin (AcO-
form): elution with an AcOH gradient afforded 2b isolated in
43% yield and with a diastereomeric excess over 98%.
Hydrogenation of 2b in the presence of Pd/C gave 2a in
quantitative yield. In this way, the low activity of AAT toward
1a could be bypassed through “substrate tailoring” by the design
Pharmacology. The pharmacological properties of the nine
2,4-syn-4-substituted Glu analogues 2a-j were first evaluated
for binding affinities at native AMPA, KA, and NMDA
receptor,12 and results are summarized in Table 2. In [3H]AMPA
binding, all nine Glu analogues showed mid-to-high range
micromolar affinities (IC50 ) 25 to >100 µM). The same trend
was observed in [3H]KA binding experiments, except for 2a,
which was found to exhibit high nanomolar affinity (IC50
)
0.24 µM). Binding affinities for NMDA receptors were deter-
mined by means of the radioactive ligand [3H]CGP39653. Again
we observed mid-to-high micromolar affinities (Ki ) 15 to >100
µM) for the nine compounds.
We subsequently investigated the binding affinities of 2a-j
at recombinant homomeric rat iGluR5, iGluR6, and iGluR7
expressed in Sf9 cell membranes (Table 2). Compounds 2b,c,f
were found to be low affinity ligands (low-high micromolar
range) at all three subtypes, iGluR5-7. However, methylester
analogue 2e and amide analogue 2g displayed high binding
affinity for the iGluR5 subtype (42.8 and 7.45 nM, respectively),
whereas they exhibited low affinity for the two other subtypes,
iGluR6,7 (1-10 µM range). Unexpectedly, this subtype affinity
profile was discontinued for the methylamide analogue, 2h,
which showed high affinity for all three subtypes, iGluR5-7
(3.81, 123, 57.3 nM, respectively). Furthermore, upon extending
the N-alkyl substituent of 2h to an N-n-propyl group, 2i, or a

4096 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
Sagot et al.
Table 2. Binding Affinities at Native iGluRs (Rat Synaptosomes) and at Cloned Rat iGluR5-7 Subtypesa
a Data are given as mean [mean pIC50 or mean pKi(M) ( S.E.M.] of at least three independent experiments. b Hill-values are close to unity for all
compounds. c The compound displayed weak binding affinity to this iGluR subtype (low-high micro-molar range). d Radio ligand: [3H]CGP 39653. n.t. )
not tested.
benzyl group, 2j, the preference for subtype iGluR5 over
iGluR6,7 was observed again. However, with lower affinities
for iGluR5 (641 and 138 nM, respectively), as compared with
2e,g,h.
The four Glu analogues 2e,g,h,i (Figure 1) were further
characterized in a functional assay at subtypes mGluR1,5 (group
I), mGluR2 (group II), and mGluR4 (group III). None of the
analogues displayed agonist, antagonist, nor modulatory activi-
ties at these mGluR subtypes (EC50 >100 µM). Moreover, these
four compounds exhibited low affinity (Ki >1000 µM) for
subtypes mGluR2,3 (group II) in binding assays.
Discussion
The orthosteric binding pocket of the KA receptor subtypes
KA1,2 and iGluR5-7 are highly confined compartments. The
key amino acid residues involved in the molecular recognition
of Glu are highly conserved and can be identified from receptor
protein crystallographic studies13,14 or receptor subtype homol-
ogy models built from such data (Table 3). In detail (numbering
Figure 1. Commercially available standard GluR ligands: NMDA, (S)-
AMPA, KA, DHK, DOMO, and (S)-QUIS.
for iGluR5 used, Figure 2), the R-carboxylate group of Glu
forms hydrogen bonds to the guanidinium side chain of Arg538
and backbone amide-NH of Thr533 and Ser704. The R-am-
monium group of Glu hydrogen bonds to the γ-carboxylate
group of Glu738 and amide-carbonyl of backbone residue
Pro531, whereas the γ-carboxylate group of Glu hydrogen bonds
to the ꢀ-OH and amide-NH of Thr705. Residue Tyr488 does
not form specific interactions with Glu but is significant in the
sense that it participates strongly in the occluding of the binding
pocket (Table 3).
In contrast to the above-described structural similarities of
the KA receptor binding pockets, the most prominent residues
ofdifferentiationare:Thr706(KA1),Thr705(KA2),Ser736(iGluR5),
Asn721(iGluR6), and Asn723(iGluR7). The residues are found

Synthesis of 2,4-Syn-Functionalized (S)-Glutamate Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4097
Table 3. Distinct Amino Acid Residues in the Binding Pocket of KAa Preferring iGluRs (Subtypes KA1,2 and iGluR5-7)
conserved ligand-amino acid residue interactions (CI)c
conserved
residues of
+
R-COO-
R-NH3
γ-COO-
residues(CR) differentiation (RD)
KA1b
KA2b
Arg507 (γ-GH+) Thr502 (A-NH) Ser674 (A-NH) Glu723 (γ-COO-) Gly500 (A-CO) Ser675 (ꢀ-OH/A-NH) Tyr473 Glu425
Thr706
Thr705
Ser736
Asn721
Asn723
Arg506 (γ-GH+) Thr501 (A-NH) Ser673 (A-NH) Glu722 (γ-COO-) Ala500 (A-CO) Thr674 (ꢀ-OH/A-NH) Tyr472 Glu424
iGluR5 Arg538 (γ-GH+) Thr533 (A-NH)Ser704 (A-NH) Glu753 (γ-COO-) Pro531 (A-CO) Thr705 (ꢀ-OH/A-NH) Tyr504 Glu456
iGluR6b Arg523 (γ-GH+) Ala518 (A-NH) Ala689 (A-NH) Glu738 (γ-COO-) Pro516 (A-CO) Thr690 (ꢀ-OH/A-NH) Tyr488 Glu440
iGluR7b Arg525 (γ-GH+) Thr520 (A-NH) Ala691 (A-NH) Glu740 (γ-COO-) Pro518 (A-CO) Thr692 (ꢀ-OH/A-NH) Tyr491 Glu443
a The amino acid residues within a distance of 4.5 Å from the Glu ligand occludes the binding pocket. b On the basis of homology modeling starting from
iGluR5-S1S2 construct crystallized with Glu (PDB code: 1txf). c (GH+) ) guanidinium; A-CO ) amide-CO; A-NH ) amide-NH.
Figure 2. Superimposition of the binding pockets of rat iGluR5 (amino
acid residues in green; Glu ligand in pink; PDB code: 1txf) and rat
iGluR6 (amino acid residues in type code; Glu ligand in type code;
PDB code: 1s50). See Experimental Section for details. Amino acid
residue numbering according to iGluR5 unless otherwise noted.
Figure 3. Docking of methylamide 2h (purple) into full agonist state
of the iGluR6 receptor (PDB code:1s50, cocrystallized Glu ligand, in
in the same area of space (see Figure 2 for illustrative example
green). The favorable hydrogen bond interaction (1.72 Å) between the
with iGluR5,6) and form a hydrogen bond with the γ-carboxy-
amide-NH of 2h and the γ-carboxylate group of residue Glu440, is
late group of a Glu amino acid residue (iGluR5:Glu456; iGluR6:
Glu440, for clarity, not shown in Figure 2). Intrinsically, residue
indicated by the double-headed black arrow.
Ser736(iGluR5) is less voluminous as compared with residues
can participate in a favorable hydrogen bond to the γ-carboxylate
Thr706(KA1),Thr705(KA2),Asn721(iGluR6),andAsn723(iGluR7).
Thus, subtype iGluR5 may accommodate larger substituents in
the (2S)-2,4-syn position of Glu, compared to KA1, KA2,
iGluR6, and iGluR7.
group of residue Glu440 (1.72 Å, Figure 3). For methylester
2e, such a ligand pose would induce disfavored electron-electron
repulsion between the ester-oxygen: γ-carboxylate of Glu440.
On the other hand, a disadvantage to this binding mode of 2h
is its amide-NH close encounter with Tyr488 (approximately
1.36 Å). However, we are not concerned by this relatively short
distance as in silico docking experiments are static representa-
tions of dynamic processes and the mobility of residue Tyr488
is well-documented (see Figure 4). In summary, we thus uphold
this amide-NH-γ-carboxylate hydrogen bond interaction to play
a crucial role as to why methylamide 2h shows unexpected
nanomolar binding affinity to iGluR6 (and iGluR7, as amino
acid residues are identical with iGluR6).
While the amide-NH2 of analogue 2g is indeed also capable
of hydrogen bonding to the γ-carboxylate group of Glu440, it
is nevertheless experimentally shown to be a low affinity ligand
at iGluR6,7 (Table 2). The structural difference of 2g, as
compared with 2h, is solely the N-methyl group. We have
investigated this issue in silico by looking at specific favored
and disfavored interactions as well as the influence of this group
on the low-energy conformation of the Glu analogues 2g,h.
Regrettably, we are not able to put forward conclusive remarks
at this point and we believe X-ray crystallographic studies are
the method of choice to verify our hypothesis for the high
affinity binding of 2h at iGluR6,7 and eventually guide us to
understand why amide 2g is yet a low affinity ligand at
iGluR6,7. This work is ongoing in our laboratories.
To address as to what extent residues Ser736(iGluR5),
Asn721(iGluR6), and Asn723(iGluR7) play a role in the
observed altering of the iGluR5/6/7 subtype selectivity profile
of structurally very close Glu analogues 2e,h, we conducted an
in silico study using the built-in docking procedure in the
comprehensive modeling software package MOE (see Experi-
mental Section for details). We first performed an automated
docking (see Experimental Section for details) of methylester
2e, amide 2g, methylamide 2h, and n-propylamide 2i to the
full-agonist state of iGluR5 (PDB code: 1txf). The Glu
analogues: methylester 2e, amide 2g, and methylamide 2h are
all well accommodated by the iGluR5 subtype, whereas n-
propylamide analogue 2i induces severe steric clashes with the
iGluR5 protein. These findings correlate well with our experi-
mentally determined binding affinities (Table 2).
To investigate the structural origin underlying the unexpected
high binding affinity of methylamide 2h to iGluR6 (and iGluR7),
we performed an automated docking of Glu analogues 2e,h,i
into the full-agonist state of the iGluR6 subtype. We were not
able to identify successful Glu binding poses for either methy-
lester 2e or n-propylamide 2i, a result that is in agreement with
the experimental data (Table 2). In contrast, docking of
methylamide 2h proved to be successful. This is because of
the increased steric hindrance (residue Asn721) in iGluR6,
forcing the N-methylamide side chain of 2h upward, which then
We decided to also characterize a small series of standard
commercially available iGluR ligands at iGluR7 and for direct

4098 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
Sagot et al.
a ligand at iGluR7 with affinity comparable with values found
for iGluR5,6.
Experimental Section
Chemistry. Melting points were determined on a Reichert hot-
stage apparatus and are uncorrected. IR spectra were recorded on
1
a Perkin-Elmer 801 spectrophotometer. H and 13C NMR spectra
were recorded on a Bruker Avance 400 MHz spectrometer.
Chemical shifts are reported in ppm (δ) relative to TMS as internal
standard. HRMS were recorded on a q-tof Micromass spectrometer.
Optical rotations were determined with a JASCO DIP 370 pola-
rimeter and are reported at the sodium D line (589 nm). Elemental
analyses were performed at the Service Central d’Analyze du
CNRS, Solaize, France. Silicagel 60 (Merck, 40-63 µm) and
precoated F254 plates were used for column and TLC chromatog-
raphies. All solvents were purified by distillation following usual
procedures. Cysteine sulfinic acid was prepared from cystine
following a described procedure.19 Bovine heart malic dehydro-
genase and rabbit muscle lactic dehydrogenase were purchased from
Sigma. E. coli AAT was produced and purified following described
procedures from overexpressing E. coli strains JM103 transformed
with pUC119-aspC (AAT).20 Enzyme kinetic measurements were
performed at 25 °C in 0.1 M potassium phosphate buffer, pH 7.6,
Asp (40 mM), NADH (0.2 mM), ketoacid substrate (0.1-10 mM),
AAT (0.05 UI), and malic dehydrogenase (2 UI) in a total volume
of 1 mL. Rates were calculated from the OD linear decay at 340
Figure 4. Superimposition of the binding pockets of iGluR6 crystal-
lized with Glu (amino acid residues in type code; Glu ligand in green;
PDB code: 1s50) and crystallized with DOMO (amino acid residues
in orange; DOMO ligand in pink; PDB code: 1yae).15 The carboxylate
group of the side chain of DOMO hydrogen bond (pink line) to the
backbone NH of residue Tyr488, thus stabilizing a more voluminous
state of the iGluR6 binding pocket (pink arrow: Tyr488 is moved 1.85
Å up-and-backward).
nm using ꢀNADH ) 6220 cm-1 ·M-1
.
The synthesis of 2 g-j was described previously by us.7
Methyl 2-(benzyloxymethyl)acrylate 3b. To a solution of
methyl 2-(bromomethyl)acrylate9 (5 g, 27.9 mmol) and benzyl
alcohol (5.4 mL, 51.8 mmol) in THF (5 mL) cooled at 0 °C was
added dropwise a solution of DABCO (4.37 g, 39 mmol) in THF
(20 mL). The suspension of white solid was heated at 70 °C for
24 h. After filtration, the solution was washed with an aqueous 1%
HCl solution (2 × 20 mL), an aqueous saturated NaHCO3 solution
(20 mL), with brine (20 mL), dried over MgSO4 and concentrated
under reduced pressure. Flash-chromatography (eluent, CH2Cl2)
afforded 3b as a colorless liquid (4.6 g, 80%). IR (neat film) 1720,
comparison also at iGluR5,6 (Table 4 and Figure 1). The binding
affinity of KA at iGluR7 (Ki ) 32.8 nM) is comparable with
the affinities for iGluR5,6 and KA1,2. In contrast, DHK shows
high nanomolar affinity for iGluR7 (Ki ) 376 nM). This is a
20-30 fold increase in affinity as compared with iGluR5,6 (Ki
) 8-12 µM), which indicates that the iGluR7 binding pocket
is more voluminous. The sterically more bulky KA like ligand,
domoric acid (DOMO), exhibits nanomolar binding affinity for
iGluR7 (Ki ) 3.84 nM) comparable with values found for
iGluR5,6 (Ki ) 1-6 nM). Quisqualic acid (QUIS) is, however,
10-fold less prone to bind to iGluR7 as compared with iGluR5,6.
Finally, the binding affinity for the endogenous ligand of the
iGluR7 subtype, Glu, was determined (Ki ) 494 nM), a value
comparable to that of iGluR5,6 (Ki ) 140-331 nM). In general,
the profile of our pharmacological data for this small series of
commercially available iGluR ligands is in agreement with data
reported earlier using [3H]KA as the radioligand at cloned
iGluR5-7 expressed in HEK293 cells.16
1637 cm-1 1H NMR (400 MHz, CDCl3) δ 7.24 (5H, m), 6.26
.
(1H, d, J ) 0.8 Hz), 5.87 (1H, d, J ) 0.8 Hz), 4.52 (2H,s), 4.17
(3H, s), 3.69 (3H, s). 13C NMR (100 MHz, CDCl3) δ 166.2, 138.2,
137.6, 128.3, 127.9, 127.5, 125.6, 72.8, 68.5, 51.5.
1-Benzyl-4-methyl Itaconate 3d. A solution of itaconic anhy-
dride (6 g, 53.5 mmol) in MeOH (27 mL) was refluxed for 30 h.
MeOH was evaporated, and the white residual solid was dissolved
intoluene(40mL).Tothissolutionwasadded1,8diazabicyclo[5.4.0]undec-
7-ene (DBU, 9.0 mL, 53.5 mmol) and then, dropwise, benzylbro-
mide (6.35 mL, 53.5 mmol). The reaction mixture was stirred at
room temperature for 70 min. The solution was then washed with
an aqueous saturated NaHCO3 solution (20 mL), with brine (20
mL), dried over MgSO4, and concentrated under reduced pressure
to give 3d isolated as a colorless liquid (11.8 g, 95%). IR (neat
1
film) 1732, 1641 cm-1. H NMR (400 MHz, CDCl3) δ 7.35 (5H,
m), 6.39 (1H, d, J ) 0.8 Hz),5.73 (1H, td, J ) 0.8 and 1,2 Hz),
5.21 (2H,s), 3.64 (3H, s), 3.36 (2H, d, J ) 1.2 Hz). 13C NMR (100
MHz, CDCl3) δ 171.1, 165.9, 135.8, 133.7, 128.9, 128.5, 128.2,
128.1, 66.8, 52.0, 37.6. HRMS (ES+) m/z 257.0784 ([M + Na]+,
C13H14NaO4 requires 257.0790). Anal. (C13H14O4): C, H, N.
1-Benzyl-5-methyl 2-methylideneglutarate 3e. To a solution
of benzyl 2-(hydroxymethyl)acrylate10 (5 g, 26.0 mmol) in toluene
(60 mL) were added trimethyl orthoacetate (6.6 mL, 52.0 mmol)
and propionic acid (0.2 mL, 2.6 mmol). The reaction mixture was
heated at 100 °C for 1.5 h before methanol was slowly removed
by azeotropic distillation. The solution was concentrated under
reduced pressure. Flash-chromatography (eluent, cyclohexane-
AcOEt, 8:2, v/v) afforded 3e isolated as colorless liquid. (4.71 g,
73%). IR (neat film) 1738, 1720, 1632 cm-1. 1H NMR (400 MHz,
CDCl3) δ 7.38 (5H, m), 6.27 (1H, s), 5.65 (1H, s), 5.23 (2H, s),
3.69 (3H, s), 2.69 (2H, t, J ) 7.5 Hz), 2.54 (2H, t, J ) 7.5 Hz).
13C NMR (100 MHz, CDCl3) δ 173.1, 166.4, 138.8, 135.9, 128.6,
128.2, 128.1, 126.3, 66.5, 51.6, 32.9, 27.3. HRMS (ES+) m/z
Conclusion
In conclusion, we have presented the chemo-enzymatic
enantioselective synthesis of five (2S)-2,4-syn-4-substituted Glu
analogues 2a-c,e,f and the subsequent pharmacological evalu-
ation of nine Glu analogues 2a-j as potential AMPA, KA, and
NMDA ligands in rat synaptosomal binding assays. Furthermore,
2b-j were characterized in binding assays at cloned rat KA
receptor subtypes iGluR5-7. Methylester 2e was shown to have
high binding affinity (low nanomolar) at iGluR5 with 120- and
25-fold selectivity over iGluR6,7, respectively. What was quite
unexpected was that fact that methylamide 2h proved to be a
high affinity ligand at all three subtypes iGluR5-7. We
performed an in silico study to address this finding and have
put forward a hypothesis as to why this is. In addition, we have
presented iGluR7 binding data for a selected series of com-
mercially available iGluR ligands and shown that Glu itself is

Synthesis of 2,4-Syn-Functionalized (S)-Glutamate Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4099
Table 4. Binding Affinities of Commercially Available Standard iGluR Ligands: KA, DHK, DOMO, and QUIS at Cloned Homomeric Rat KA1,2 and
Homomeric Rat iGluR5-7 Subtypesa
compound
KA
rat KA1 Ki (nM)
4.7b
rat KA2 Ki (nM)
15c
rat iGluR5 Ki (nM)
rat iGluR6 Ki (nM)
rat iGluR7 Ki (nM)
75.9
12.7
32.8
[7.15 ( 0.08]
[7.91 ( 0.06]
[8.42 ( 0.04]
DHK
nd
40b
nd
nd
nd
nd
nd
nd
nd
nd
8180
12300
376
[5.09 ( 0.02]
1.11
[8.97 ( 0.09]
171
[6.77 ( 0.01]
140
[6.86 ( 0.01]
0.663
[4.92 ( 0.05]
6.04
[8.24 ( 0.08]
134
[6.88 ( 0.05]
331
[6.48 ( 0.02]
17.0
[6.44 ( 0.07]
3.84
[8.42 ( 0.04]
1670
[5.86 ( 0.13]
494
[6.31 ( 0.03]
5.69
DOMO
QUIS
Glu
(2S,4R)-Me-Glu
[9.18 ( 0.02]
[7.83 ( 0.09]
[8.33 ( 0.10]
a Mean value from at least three experiments conducted in triplicate is shown [pKi (M) ( S.E.M.]. nd ) no data available. b Value taken from ref 17.
c Value taken from reference 18.
271.0959 ([M + Na]+, C14H16NaO4 requires 271.0946). Anal.
(C14H16O4): C, H, N.
aqueous saturated solution of CuSO4 (2 × 50 mL), with water (50
mL), dried over MgSO4 and concentrated under reduced pressure.
Flash chromatography (eluent, cyclohexane-AcOEt, 8:2, v/v)
afforded 5b-e isolated as colorless liquids and as a single
stereomer.
General Procedure for the Synthesis of 4b-e. To a solution
of acrylate 3b-e (20 mmol) in MeOH (3b and 3c) or DMSO (3d
and 3e) (40 mL) were added KF (3.5 g, 60 mmol) and methyl or
benzyl acetoacetate (30 mmol). The reaction mixture was heated
at 65 °C for 1-16 h until the disappearance of acrylate 3. The
reaction mixture was then diluted with brine (40 mL) and extracted
with AcOEt (4 × 50 mL). The combined organic layers were dried
over MgSO4 and concentrated under reduced pressure. Flash
Dimethyl 2-(1-Acetoxyethylidene)-4-(benzyloxymethyl) Glu-
1
tarate 5b. Yield 64%. IR (neat film) 1762, 1723, 1653 cm-1. H
NMR (400 MHz, CDCl3) δ 7.31 (5H, m), 4.47 (2H, m), 3.73 (3H,
s), 3.66 (3H, s), 3.60 (1H,dd, J ) 7.2 and 9.0 Hz), 3.52 (1H, dd,
J ) 5.50 and 9.2 Hz), 2.81 (1H, m), 2.57 (2H, m), 2.28 (3H, s),
2.10 (3H, s). 13C NMR (100 MHz, CDCl3) δ 174.1, 168.1, 167.5,
158.7, 138.1, 128.3, 127.5, 119.2, 73.0, 70.1, 51.8, 45.0, 26.4, 20.8,
19.6. HRMS (ES+) m/z 387.1427 ([M + Na]+, C19H24NaO7
requires 387.1420).
Trimethyl 5-Acetoxyhex-4-ene-1,2,4-tricarboxylate 5c. Yield
86%. IR (neat film) 1760, 1732, 1651 cm-1. 1H NMR (400 MHz,
CDCl3) δ 3.73 (3H, s), 3.64 (3H, s), 3.62 (3H, s), 2.92 (1H, m),
2.62 (2H, m), 2.43 (2H, m), 2.26 (3H, s), 2.16 (3H, s). 13C NMR
(100 MHz, CDCl3) δ 174.6, 172.1, 167.9, 167.2, 159.2, 118.6, 51.9,
51.8, 51.7, 40.2, 35.0, 29.1, 20.8, 19.6. HRMS (ES+) m/z 339.1057
([M + Na]+, C14H20NaO8 requires 339.1056).
chromatography
(eluent,
cyclohexane-AcOEt,
8:2
or
CH2Cl2-MeOH, 99:1 for 4b, v/v) afforded 4b-e as colorless
liquids and as 1:1 mixtures of diastereomers.
Dimethyl 2-Acetyl-4-(benzyloxymethyl)glutarate 4b. Yield
1
95%. IR (neat film) 1738 cm-1. H NMR (400 MHz, CDCl3) δ
7.33 (2 × 5H, m), 4.49 (2 × 2H, m), 3.73 (3H, s), 3.70 (3H, s),
3.69 (2 × 3H, s), 3.76-3.57 (2 × 3H, m), 2.69 (2 × 1H, m), 2.24
(3H, s), 2.22 (3H, s), 2.30-2.05 (2 × 2H, m). 13C NMR (100 MHz,
CDCl3) δ 202.5, 173.7, 169.5, 137.8, 128.4, 127.7, 127.6, 73.1,
70.7, 70.3, 57.2, 56.8, 52.5, 51.9, 43.6, 43.5, 29.5, 29.0, 26.9, 26.8.
HRMS (ES+) m/z 345.1306 ([M + Na]+, C17H22NaO6 requires
345.1314).
2,4-Dibenzyl-1-methyl 5-acetoxyhex-4-ene-1,2,4-tricarboxy-
late 5d. Yield 86%. IR (neat film) 1760, 1732, 1650 cm-1 1H
.
Trimethyl 5-Oxohexane-1,2,4-tricarboxylate 4c. Yield 78%.
IR (neat film) 1737 cm-1 1H NMR (400 MHz, CDCl3) δ 3.72
.
NMR (400 MHz, CDCl3) δ 7.32 (10H, m), 5.18 (2H, s), 5.08 (2H,
s), 3.58 (3H, s), 3.01 (1H, m), 2.68 (2H, m), 2.46 (2H, m), 2.26
(3H, s), 2.14 (3H, s). 13C NMR (100 MHz, CDCl3) δ 174.5, 173.9,
168.0, 166.6, 159.5, 135.9, 135.6, 128.6, 128.5, 128.3, 128.1, 118.8,
66.8, 66.6, 51.7, 40.4, 35.1, 29.3, 20.8, 19.7. HRMS (ES+) m/z
491.1692 ([M + Na]+, C26H28NaO8 requires 491.1682).
3,5-Dibenzyl-1-methyl 6-acetoxyhept-5-ene-1,3,5-tricarboxy-
late 5e. Yield 91%. IR (neat film) 1758, 1736, 1651 cm-1. 1H NMR
(400 MHz, CDCl3) δ 7.32 (10H, m), 5.17 (2H, m), 5.05 (2H, s),
3.62 (3H, s), 2.68 (1H, dd, J ) 8.2 and 13.1 Hz), 2.59 (1H, m),
2.43 (1H, dd, J ) 6.1 and 13.1 Hz), 2.26 (3H, s), 2.22 (2H, m),
2.15 (3H, s), 1.95-1.75 (2H, m). 13C NMR (100 MHz, CDCl3) δ
174.6, 173.1, 168.0, 166.7, 159.0, 135.9, 135.7, 128.6, 128.5, 128.3,
128.2, 119.2, 66.7, 66.4, 51.6, 43.9, 31.6, 30.0, 27.0, 20.8, 19.7.
HRMS (ES+) m/z 505.1847 ([M + Na]+, C27H30NaO8 requires
505.1838).
(3H, s), 3.69 (3H, s), 3.66 (3H, s), 3.63 (2 × 3H, s), 3.63-3.57 (2
× 1H, m), 2.75 (2 × 1H, m), 2.75-2.35 (2 × 2H, m), 2.24 (3H,
s), 2.21 (3H, s), 2.15-2.00 (2 × 2H, m). 13C NMR (100 MHz,
CDCl3) δ 202.0, 201.8, 174.2, 171.7, 169.4, 169.3, 57.3, 56.6, 52.6,
52.0, 51.8, 51.7, 39.2, 39.0, 36.3, 36.0, 29.8, 29.7, 29.5, 29.0. HRMS
(ES+) m/z 297.0935 ([M + Na]+, C12H18NaO7 requires 297.0950).
2,4-Dibenzyl-1-methyl 5-oxohexane-1,2,4-tricarboxylate 4d.
Yield 66%. IR (neat film) 1745, 1712 cm-1. 1H NMR (400 MHz,
CDCl3) δ 7.35 (4 × 5H, m), 5.12 (4 × 2H, m), 3.62 (2 × 1H, m),
3.60 (2 × 3H, s), 2.84 (2 × 1H, m), 2.72 (2 × 1H, m), 2.48 (2 ×
1H, m), 2.13 (3H, s), 2.10 (3H, s), 2.20-2.00 (2 × 2H, m). 13C
NMR (100 MHz, CDCl3) δ 201.9, 201.7, 173.6, 171.6, 168.8, 168.6,
135.6, 135.1, 128.7, 128.6, 128.4, 128.2, 128.1, 128.0, 67.4, 66.8,
57.4, 56.6, 51.8, 39.3, 39.1, 36.5, 36.1, 29.7, 29.6, 29.5, 29.0. HRMS
(ES+) m/z 449.1566 ([M + Na]+, C24H26NaO7 requires 449.1576).
3,5-Dibenzyl-1-methyl 6-oxoheptane-1,3,5-tricarboxylate 4e.
Yield 75%. IR (neat film) 1737 cm-1. 1H NMR (400 MHz, CDCl3)
δ 7.35 (4 × 5H, m), 5.05 (4 × 2H, m), 3.56 (2 × 3H, s), 3.43 (2
× 1H, m), 2.34 (2 × 1H, m), 2.02 (2 × 3H, s), 2.25-1.70 (2 ×
6H, m). 13C NMR (100 MHz, CDCl3) δ 202.0, 201.8, 174.3, 173.0,
168.9, 168.7, 135.7, 135.2, 135.1, 128.6, 128.5, 128.4, 67.31, 67.28,
66.6, 57.4, 56.6, 51.6, 42.5, 42.3, 31.4, 31.3, 29.8, 29.0, 27.5, 27.1.
HRMS (ES+) m/z 463.1724 ([M + Na]+, C25H28NaO7 requires
463.1733).
General Procedure for the Synthesis of 5b-e. To a solution
of 4b-e (10 mmol) in anhydrous pyridine (20 mL) was added acetyl
chloride (1.1 mL, 15 mmol). The mixture was stirred at room
temperature for 3-48 h until the complete disappearance of 4b-e.
The solution was then diluted with Et2O (50 mL), washed with an
General Ozonolysis Procedure for the Synthesis of 6c-e. A
solution of enol acetate 5c-e (5 mmol) in anhydrous CH2Cl2 (30
mL) was treated at -70 °C with a mixture of O2 and O3 at a rate
of 10 L/h until saturation (blue coloration of the solution). After
30 min, the excess ozone was eliminated by oxygen bubbling.
Dimethyl sulfide (0.55 mL, 7.5 mmol) was added, and the reaction
mixture was allowed to warm to room temperature. The solution
was washed with water (20 mL), brine (20 mL), dried over MgSO4
and concentrated under reduced pressure. Flash-chromatography
(eluent, cyclohexane-AcOEt, 8:3, v/v or AcOEt for 6c) afforded
6c-e as colorless liquids.
Trimethyl 4-Oxobutane-1,2,4-tricarboxylate 6c. Yield 72%;
IR (neat film) 1732 cm-1 1H NMR (400 MHz, CDCl3) δ 3.84
.
(3H, s), 3.65 (2 × 3H, s), 3.35 (2H, m) 3.02 (1H, m), 2.73 (1H,

4100 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
dd, J ) 5.8 and 16.8 Hz), 2.60 (1H, dd, J ) 6.8 and 16.8 Hz). 13
Sagot et al.
C
176.9, 176.7, 104.5, 104.1, 71.3, 71.2, 46.3, 45.3, 40.5, 40.1. HRMS
NMR (100 MHz, CDCl3) δ 191.2, 173.2, 171.6, 160.7, 53.1, 52.4,
51.9, 39.9, 36.2, 34.8. HRMS (ES+) m/z 269.0634 ([M + Na]+,
C10H14NaO7 requires 269.0637). Anal. (C10H14O7): C, H, N.
2,4-Dibenzyl-1-methyl 4-oxobutane-1,2,4-tricarboxylate 6d.
Yield 73%; IR (neat film) 1736 cm-1. 1H NMR (400 MHz, CDCl3)
δ 7.34 (10H, m), 5.27 (2H, s), 5.12 (2H, m), 3.63 (3H, s), 3.42
(2H, m) 3.09 (1H, m), 2.78 (1H, dd, J ) 5.6 and 16.8 Hz), 2.64
(1H, dd, J ) 6.4 and 16.8 Hz). 13C NMR (100 MHz, CDCl3) δ
191.5, 172.7, 171.6, 160.1, 135.4, 134.4, 128.8, 128.7, 128.6, 128.3,
128.2, 68.1, 67.1, 51.9, 40.0, 36.4, 34.9. HRMS (ES+) m/z 421.1270
([M + Na]+, C22H22NaO7 requires 421.1263). Anal. (C22H22O7):
C, H, N.
1,3-Dibenzyl-5-methyl 1-oxopentane-1,3,5-tricarboxylate 6e.
Yield 59%; IR (neat film) 1736 cm-1. 1H NMR (400 MHz, CDCl3)
δ 7.31 (10H, m), 5.24 (2H, s), 5.08 (2H, s), 3.62 (3H, s), 3.30 (2H,
dd, J ) 8.4 and 18.1 Hz), 2.98 (1H, m), 2.93 (1H, dd, J ) 4.7 and
18.1 Hz), 2.33 (2H, m), 2.05-1.80 (2H, m). 13C NMR (100 MHz,
CDCl3) δ 191.7, 173.6, 172.9, 160.1, 135.5, 134.3, 128.8, 128.7,
128.6, 128.3, 128.2, 68.1, 66.9, 51.7, 40.8, 39.2, 31.3, 26.6. HRMS
(ES+) m/z 435.1423 ([M + Na]+, C23H24NaO7 requires 435.1420).
Anal. (C23H24O7): C, H, N.
(ES-) m/z 175.0236 ([M - 2Li + H]-, C6H7O6 requires 175.0243).
1
Dilithium 2-(Benzyloxymethyl)-4-oxoglutarate 1b. H NMR
(400 MHz, D2O) δ 7.38 (5H, m), 4.52 (2H, s), 3.68 (1H, dd, J )
6.0 and 9.7 Hz), 3.59 (1H, dd, J ) 6.7 and 9.7 Hz), 3.05 (1H, dd,
J ) 7.2 and 17.1 Hz), 2.96 (1H, m), 2.84 (1H, dd, J ) 4.8 and
17.2 Hz). 13C NMR (100 MHz, D2O) δ 204.1, 180.6, 169.4, 137.4,
128.7, 128.4, 128.1, 72.6, 71.1, 43.6, 39.3. HRMS (ES-) m/z
271.0795 ([M - Li]-, C13H12LiO6 requires 271.0794).
Trilithium 4-Oxobutane-1,2,4-tricarboxylate 1c. 1H NMR (400
MHz, D2O) δ 3.00 (1H, dd, J ) 5.9 and 16.4 Hz), 2.94 (1H, m),
2.84 (1H, dd, J ) 5.0 and 16.4 Hz), 2.56 (1H, dd, J ) 5.5 and
15.3 Hz), 2.28 (1H, dd, J ) 8.6 and 15.3 Hz). 13C NMR (100
MHz, D2O) δ 202.6, 181.6, 179.6, 168.4, 41.1, 40.6, 40.1. HRMS
(ES-) m/z 203.0196 ([M - 2Li + H]-, C7H7O7 requires 203.0192).
General Procedure for the Synthesis of Lithium Oxoglut-
arates 1d and 1e. To a solution of 6d or 6e (2 mmol) in MeOH
(20 mL) was added 10% Pd/C (50 mg). The solution was degassed
under reduced pressure and then flushed with hydrogen using a
balloon. The suspension was stirred at room temperature for 30
min before filtration on a 0.2 µm membrane and concentration under
reduced pressure. The residue was dissolved in H2O (10 mL) and
the solution brought to pH 7.6 by addition of an aqueous 0.4 M
solution of LiOH. Water was evaporated under reduced pressure
to give 1d or 1e isolated in quantitative yields as white solids.
Dilithium 2-(2-Methoxy-2-oxoethyl)-4-oxoglutarate 1d. 1H
NMR (400 MHz, D2O) δ 3.66 (3H, s), 3.10 (1H, dd, J ) 7.1 and
18.0 Hz), 3.05 (1H, m), 2.92 (1H, dd, J ) 7.2 and 17.6 Hz), 2.61
(1H, dd, J ) 7.8 and 16.4 Hz), 2.52 (1H, dd, J ) 6.2 and 16.3
Hz). 13C NMR (100 MHz, D2O) δ 203.8, 181.4, 175.4, 169.4, 52.2,
41.4, 39.3, 36.3. HRMS (ES-) m/z 217.0351 ([M - 2Li + H]-,
C8H9O7 requires 217.0348).
Dimethyl 2-(Benzyloxymethyl)-4-oxoglutarate 6b. To a solu-
tion of 5b (0.83 g, 2.3 mmol) in CHCl3 (8 mL) were added CH3CN
(8 mL), H2O (15 mL), NaIO4 (1.95 g, 9.1 mmol), and RuO2 (60
mg, 0.45 mmol). The mixture was stirred vigorously at room
temperature for 4 h. After filtration through celite, the organic layer
was isolated and the aqueous layer extracted with CHCl3 (10 mL).
The combined organic layers were dried over MgSO4 and concen-
trated under reduced pressure. Flash chromatography (eluent,
cyclohexane-AcOEt, 7:3, v/v) afforded 6b as a colorless liquid
(0.29 g, 43%). IR (neat film) 1736 cm-1 1H NMR (400 MHz,
.
CDCl3) δ 7.31 (5H, m), 4.50 (2H, s), 3.86 (3H, s), 3.72 (3H, s),
3.71 (2H, m), 3.44 (1H, dd, J ) 8.1 and 18.1 Hz), 3.31 (1H, m),
3.10 (2H, dd, J ) 5.3 and 18.1 Hz). 13C NMR (100 MHz, CDCl3)
δ 191.9, 172.5, 160.9, 137.6, 128.4, 127.8, 127.6, 73.1, 69.4, 53.0,
52.2, 41.2, 38.1. HRMS (ES+) m/z 317.1008 ([M + Na]+,
C15H18NaO6 requires 317.1001). Anal. (C15H18O6): C, H, N.
Dimethyl 2-(Hydroxymethyl)-4-oxoglutarate 6a. To a solution
of 5b (0.14 g, 0.47 mmol) in MeOH (8 mL) was added 10% Pd/C
(20 mg). The solution was degassed under reduced pressure and
then flushed with hydrogen using a balloon. The suspension was
stirred at room temperature for 1 h before filtration on a 0.2 µm
membrane and concentration under reduced pressure. Flash chro-
matography (eluent, cyclohexane-AcOEt, 5:5, v/v) afforded 6a as
a colorless liquid (93 mg, 97%). As shown by NMR, 6a exists in
solution as a 1:1 mixture of hemiacetals. IR (neat film) 3458, 1737
1
Dilithium 2-(3-Methoxy-3-oxopropyl)-4-oxoglutarate 1e. H
NMR (400 MHz, D2O) δ 3.69 (3H, s), 3.04 (1H, dd, J ) 7.8 and
17.5 Hz), 2.86 (1H, dd, J ) 6.1 and 17.6 Hz), 2.65 (1H, m), 2.42
(2H, t, J ) 7.8 Hz), 1.79 (2H, m).
13C NMR (100 MHz, D2O) δ
204.8, 182.1, 177.1, 170.2, 52.8, 42.4, 42.1, 32.1, 27.0. HRMS
(ES-) m/z 231.0514 ([M - 2Li + H]-, C9H11O7 requires
231.0505).
(2S,4S)-4-(Benzyloxymethyl)glutamic Acid 2b. To a solution
of racemic 1b (154 mg, 0.55 mmol) in water (25 mL) was added
cysteine sulfinic acid (84 mg, 0.55 mmol). The pH of the solution
was adjusted to 7.6 with 1 M NaOH before the addition of E. coli
AAT (0.4 mg, 20 Units). The reaction was stirred slowly at room
temperature and monitored by titration of pyruvate: 5 µL aliquots
of the reaction mixture were added to 995 µL of 0.1 M potassium
phosphate buffer, pH 7.6, containing NADH (0.2 mM) and lactate
dehydrogenase (1 Unit). Pyruvate concentration was calculated from
1
cm-1. H NMR (400 MHz, CDCl3) δ 4.44 (1H, s), 4.30 (1H, t, J
) 8.4 Hz), 4.29 (1H, dd, J ) 6.3 and 8.7 Hz), 4.17 (1H, t, J ) 8.6
Hz), 4.15 (1H, t, J ) 8.5 Hz), 4.12 (1H, s), 3.81 (2 × 3H, s), 3.73
(3H, s), 3.71 (3H, s), 3.43 (1H, m), 3.29 (1H, m), 2.71 (1H, dd, J
) 9.4 and 13.8 Hz), 2.68 (1H, dd, J ) 9.2 and 13.9 Hz), 2.42 (1H,
the ∆OD measured at 340 nm using ꢀNADH ) 6220 M-1 ·cm-1
.
When a conversion rate of 40% was reached, the reaction mixture
was rapidly passed through a column of dowex 50WX8 resin (H+
form, 25 mL). The column was then washed with water (100 mL)
until complete elution of CSA and then eluted with 1 M NH4OH.
The ninhydrin positive fractions were combined and concentrated
to dryness under reduced pressure. The residue was diluted in water
(5 mL) and, if necessary, the pH adjusted to 7.0 with 1 M NaOH
before adsorption of the product on a column of dowex 1 × 8 resin
(200-400 Mesh, AcO- form, 1.5 cm × 12 cm). The column was
washed with water (50 mL) and then eluted with an AcOH gradient
(0.1-1 M). The ninhydrin positive fractions were combined and
concentrated under reduced pressure. The residue was dissolved
in water (4 mL) and the solution was lyophilized to afford 2b (64
dd, J ) 5.7 and 13.7 Hz), 2.32 (1H, dd, J ) 8.4 and 13.1 Hz). 13
C
NMR (100 MHz, CDCl3) δ 173.4, 172.7, 170.5, 170.2, 102.6, 102.2,
70.4, 53.3, 53.2, 52.5, 52.3, 43.6, 42.7, 38.3. HRMS (ES+) m/z
227.0536 ([M + Na]+, C8H12NaO6 requires 227.0532). Anal.
(C8H12O6): C, H, N.
General Procedure for the Synthesis of Lithium Oxoglut-
arates 1a, 1b, and 1c. To a solution of 6a, 6b, or 6c (1 mmol) in
MeOH (5 or 7.5 mL for 6c) was added dropwise a 0.4 M solution
of LiOH (5.25 mL, 2.1 mmol or 7.75 mL, 3.1 mmol for 6c). The
mixture was stirred at room temperature for 24 h. After evaporation
of MeOH, the pH of the aqueous solution was adjusted to 7.6 by
addition of dowex 50WX8 resin (H+ form). The resin was removed
by filtration before evaporation of the water under reduced pressure.
1a, 1b, and 1c were isolated in quantitative yields as white solids.
Dilithium 2-(Hydroxymethyl)-4-oxoglutarate 1a. 1H NMR
(400 MHz, D2O) δ 4.40 (1H, t, J ) 8.6 Hz), 4.34 (1H, t, J ) 8.3
Hz), 4.24 (1H, dd, J ) 6.5 and 8.4 Hz), 4.13 (1H, t, J ) 8.7 Hz),
3.43 (1H, m), 3.30 (1H, m), 2.55 (1H, dd, J ) 9.3 and 13.5 Hz),
2.39 (1H+2 × 1H, m). 13C NMR (100 MHz, D2O) δ 182.1, 181.5,
mg, 43%) as a white solid; mp 144 °C; [R]25 ) +31.0 (c 1.0, 6
D
1
N HCl). H NMR (400 MHz, D2O) δ 7.41 (5H, m), 4.58 (2H, s),
3.98 (1H, t, J ) 6.8 Hz), 3.81 (1H, dd, J ) 5.1 and 10.0 Hz), 3.76
(1H, dd, J ) 5.8 and 10.0 Hz), 2.97 (1H, m), 2.38 (1H, m), 2.04
(1H, m). 13C NMR (100 MHz, D2O) δ 180.1, 174.2, 137.4, 128.7,
128.4, 128.2, 72.8, 71.4, 53.2, 44.8, 30.0. HRMS (ES-) m/z
266.1033 ([M - H]-, C13H16NO5 requires 266.1028). Anal.
(C13H17NO5 ·0.1H2O) C, H, N.

Synthesis of 2,4-Syn-Functionalized (S)-Glutamate Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4101
(2S,4R)-4-(3-Methoxy-3-oxopropyl)glutamic Acid 2e. It was
prepared following the procedure already described for 2b, except
for the first chromatography step on dowex 50WX8 that was
Systems and Insect Cell Culture Techniques” (Life Technologies,
Paisley, UK) and “Baculovirus Expression Vector System:
Procedures and Methods Manual”, 2nd ed., (Pharmingen). [3H]-
SYM2081 ([3H]-(2S,4R)-4-methylglutamic acid) (47.9 Ci/mmol;
ARC Inc., St. Louis, MO) microtiter plate binding assays were
performed in 250 µL assay buffer (50 mM Tris-HCl, pH 7.1 at
4 °C) at a protein concentration from 10-50 µg/mL, incubating
for 1-2 h at 4 °C. Bound and free radiolabels were separated
by cold filtration through GF/B glass fiber filters (UniFilter-96,
Perkin-Elmer, Waltham MA) on a Perkin-Elmer FilterMate
manifold using two washes with cold assay buffer. Nonspecific
binding was determined in the presence of 1 mM L-Glu.
Competition studies were performed using 1-5 nM radiolabel
in the presence of 10 nM to 1 mM cold ligand. Filters were
dried 1 h at 50 °C and then 50 µL/well Microscint 20 was added.
Radioactivity was detected as DPM using a TopCounter. The
affinity of the radiolabel for the kainate receptors was determined
from saturation binding experiments, Kd (mean ( SEM):
iGluR5(Q)1b, 0.663 ( 0.035 nM; iGluR6(V,C,R)a, 17.0 ( 3.0
nM; iGluR7a, 5.69 ( 1.12 nM. Competition data were analyzed
using Grafit v3.00 (Erithacus Software Ltd., Horley, UK) and
fit to equations as previously described for the determination of
Ki.21
mGluR Assays. Mammalian Cell Tissue Culture and Trans-
fection. CHO cells were grown in T150 flasks in Dulbecco’s
modified Eagle medium (DMEM) (Invitrogen, Carlsbad, CA) with
supplements (10% bovine calf serum, 4 mM glutamine, 100 units/
mL penicillin 100 µg/mL streptomycin) at 37 °C, 5% CO2. Stock
plates of CHO cells were trypsinized and split 1:6 every 3-4 days.
COS-7 cells were grown in 225 cm2 flasks in Dulbecco’s modified
Eagle medium (DMEM) with supplements (10% bovine calf serum,
4 mM glutamine, 100 units/mL penicillin/100 µg/mL streptomycin)
at 37 °C, 5% CO2. Stock plates of COS-7 cells were trypsinized
and split 1:6 every 3-4 days.
The rat metabotropic glutamate receptors 2 and 3 were cloned
as follows; the coding regions of mGluR2 and mGluR3 were
amplified by PCR from rat brain cDNA in 3 fragments that were
ligated using internal restriction sites. A Kozak consensus sequence
was included in the first PCR primer. The full-length receptor
cDNAs were subcloned into mammalian expression vectors suitable
for stable and transient expression. The chimera, C elegans GR Qi3,
cDNA was cloned by a PCR approach and sequenced by Lundbeck
Research, USA, as previously described.22,23 Both cDNAs were
transiently transfected into CHO cells in T150 flasks by the
Lipofectamine Plus (Invitrogen Carlsbad, California) method using
a total of 10 µg of DNA/ ∼ 7 × 106 cells. The standard cDNA
transfection ratio was 8:2 (8 µg GPCR cDNA and 2 µg chimera
GR cDNA). The cDNAs for rat metabotropic glutamate receptors
1 and 5 were generous gifts from Prof. S. Nakanishi (Kyoto
University, Kyoto, Japan). These cDNAs were transiently trans-
fected into HEK293 cells using the same method.
omitted. 2e was isolated as a white solid (45 mg, 39%); mp 135
1
°C; [R]25 ) +29.5 (c 1.0, 6 N HCl). H NMR (400 MHz, D2O)
D
δ 3.75 (1H, dd, J ) 5.4 and 8.6 Hz), 3.70 (3H, s), 2.58 (1H, m),
2.47 (2H, m), 2.25 (1H, ddd, J ) 5.2, 9.6 and 14.8 Hz), 1.92 (3H,
m). 13C NMR (100 MHz, D2O) δ 179.2, 176.2, 173.8, 53.1, 52.2,
41.8, 32.5, 31.2, 27.0. HRMS (ES+) m/z 234.0982 ([M + H]+,
C9H16NO6 requires 234.0978). Anal. (C9H15NO6 · 0.66H2O) C,
H, N.
(2S,4S)-4-(Hydroxymethyl)glutamic Acid 2a. To a solution of
2b (48 mg, 0.18 mmol) in H2O (5 mL) was added 10% Pd/C (10
mg). The solution was degassed under reduced pressure and then
flushed with hydrogen using a balloon. The suspension was stirred
at room temperature for 4 h before filtration on a 0.2 µm membrane
and concentration under reduced pressure. 2a was purified by
chromatography on dowex 1 × 8 as described for 2b and isolated
as a white solid (31 mg, 97%); mp 145 °C; [R]25D ) +26.0 (c 1.1,
1
6 N HCl). H NMR (400 MHz, D2O) δ 3.81 (3H, m), 2.76 (1H,
m), 2.30 (1H, ddd, J ) 5.4, 9.4 and 14.8 Hz), 1.99 (1H, ddd, J )
4.6, 8.0 and 14.8 Hz). 13C NMR (100 MHz, D2O) δ 180.6, 174.4,
63.3, 53.3, 47.0, 29.8. HRMS (ES+) m/z 178.0719 ([M + H]+,
C6H12NO5 requires 178.0715). Anal. (C6H11NO5 · 0.33H2O) C,
H, N.
(2S,4S)-4-(2-Hydroxy-2-oxoethyl)glutamic Acid 2c. The ke-
toacid 1c (170 mg, 0.74 mmol) was used in a transamination
reaction as described for 2b. When a conversion rate close to 40%
was reached, an aqueous 6N HCl solution (25 mL) was added to
the reaction mixture, which was then heated at 100 °C for 4 h before
concentration under reduced pressure. The residue was dissolved
in water (10 mL) and the pH adjusted to 7 before the two-step
purification done as described for 2b. 2c was thus isolated as a
white solid (63 mg, 41%); mp 145 °C; [R]25 ) +8.0 (c 1.0, 6 N
D
1
HCl). H NMR (400 MHz, D2O) δ 3.88 (1H, dd, J ) 5.4 and 8.1
Hz), 2.68 (1H, m), 2.98 (1H, m), 2.74 (2H, m), 2.34 (1H, ddd, J )
5.7, 10.0, 15.0 Hz). 13C NMR (100 MHz, D2O) δ 177.1, 175.4,
171.3, 51.2, 37.5, 35.6, 31.1. HRMS (ES+) m/z 206.0672 ([M +
H]+, C7H12NO6 requires 206.0665). Anal. (C7H11NO6 ·1.5H2O) C,
H, N.
(2S,4R)-4-(3-Hydroxy-3-oxopropyl)glutamic Acid 2f. It was
prepared from 1e (250 mg, 1.0 mmol) following the procedure
described for 2c. 2f was isolated as a white solid (90 mg, 41%);
1
mp 95 °C; [R]25 ) +6.4 (c 1.1, 6 N HCl). H NMR (400 MHz,
D
D2O) δ 3.77 (1H, dd, J ) 5.6 and 8.2 Hz), 2.61 (1H, m), 2.41 (2H,
m), 2.25 (1H, ddd, J ) 5.6, 9.7 and 15.0 Hz), 1.89 (3H, m). 13C
NMR (100 MHz, D2O) δ 178.5, 177.6, 173.4, 52.8, 41.2, 32.2,
31.2, 26.9. HRMS (ES+) m/z 220.0815 ([M + H]+, C8H15NO6
requires 220.0821). Anal. (C8H14NO6 ·1.2H2O) C, H, N.
6-Oxopiperidine-2,4-dicarboxylic Acid 7. The ketoacid 1d (250
mg, 1.08 mmol) was used in a transamination reaction as described
for 2b. After elution of the dowex 50WX8 resin with aqueous 1 M
ammonia, the fractions containing 7 were concentrated under
Membrane Preparation and [32H]LY43915 Binding. CHO
cells were transiently transfected as described above. They were
harvested 48 h after transfection and washed 3 times with cold
PBS. The cells were centrifuged at 1200 RPM, 4 °C, for 10
min, and the pellet was resuspended in wash buffer A (20 mM
HEPES, 10 mM EDTA, pH 7.4). The pellet was sonicated on
the Virsonic 300 model, 4 times in 30 s intervals on ice,
centrifuged for 10 min at 1200 RPM, 4 °C. The supernatant was
collected and centrifuged for 20 min at 17000 RPM. The pellet
was resuspended in wash buffer B (20 mM HEPES, 0.1 mM
EDTA, pH 7.4) and homogenized for 10 s. Protein concentration
was determined by Bradford method.24 The membrane was
stored at -80 °C prior to use.
reduced pressure. The lactame 7 was isolated as a white solid (70
1
mg, 35%). mp 225 °C; [R]25 ) -25.4 (c 0.8, H2O). H NMR
D
(400 MHz, D2O) δ 4.20 (1H, dd, J ) 5.4 and 10.2 Hz), 3.03 (1H,
tdd, J ) 3.6, 6.0, 9.6 Hz), 2.67-2.50 (3H, m), 2.74 (2H, m), 1.92
(1H, td, J ) 10.0 and 13.6 Hz). 13C NMR (100 MHz, D2O) δ
177.2, 175.8, 173.5, 54.4, 36.9, 32.0, 27.3. HRMS (ES+) m/z
253.0674 ([M + Na]+, C10H14NaO6 requires 253.0688).
AMPA, KA, and NMDA Binding Assays. Glu analogues 2a-j
were evaluated for AMPA, KA, NMDA (CGP 39653) binding
affinity in native rat synaptosomes, in accordance with previously
described experimental procedures.12
iGluR5-7 Binding Assay. Rat iGluR5(Q)1b, iGluR6(V,C,R)a,
and iGluR7a were inserted into recombinant baculoviruses,
receptors expressed by infection of Sf9 insect cells and infected
Sf9 cell membranes utilized for radioligand binding assays. Cells
were maintained in BaculoGold Max-XP serum-free medium
(BD Biosciences-Pharmingen, San Diego, CA) according to
standard protocols in “Guide to Baculovirus Expression Vector
Binding assays were performed as described with slight
modifications.25 Briefly, after thawing, the membrane homoge-
nates resuspended in 50 mM Tris-HCl, 2 mM MgCl2, and 2 mM
CaKCl2 binding buffer at pH 7.0 to a final assay concentration
of 10 µg protein/well for [3H] LY341495 scintillation proximity
assay (SPA) binding. Incubations included 2 nM [3H] LY341495,
1.0 mg of WGA-SPA beads (GE Health, Buckinghamshire,UK).

4102 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
Sagot et al.
Each well contained either buffer or varying concentrations of
compound. Nonspecific binding was defined with 10 µM
LY341495. After 45 min incubation on ice, the 96-well SPA
microplate was further incubated at 4 °C overnight and counted
with Wallac MicroBeta TriLux. At the Kd value, the nonspecific
binding for [3H]LY341495 was approximately 5%. IC50 values
were derived from the inhibition curve and Ki values were
calculated according to Cheng and Prusoff equation Ki ) IC50/
(1+ [L]/Kd) (Cheng and Prusoff 1973), where [L] is the
concentration of radioligand and Kd is its dissociation constant
at the receptor, derived from the saturation isotherm.
Figures 2-4 as PDB and MOE file formats. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Meldrum, B. S. Glutamate as a neurotransmitter in the brain: Review
of physiology and pathology. J. Nutr. 2000, 130, 1007S–1015S.
(2) Bra¨uner-Osborne, H.; Egebjerg, J.; Nielsen, E. O.; Madsen, U.;
Krogsgaard-Larsen, P. Ligands for glutamate receptors: Design and
therapeutic prospects. J. Med. Chem. 2000, 43, 2609–2645.
(3) Beart, P. M.; O’Shea, R. D. Transporters for L-glutamate: An update
on their molecular pharmacology and pathological involvement. Br. J.
Pharmacol. 2007, 150, 5–17.
Calcium Mobilization Assay. Twenty-four hours after transient
transfection, cells were seeded into 384-well black wall microtiter
plates coated with poly-D-lysine for assay the following day.
Just prior to assay, media was aspirated and cells were dye-
loaded (25 µL/well) with 3 µM Fluo-4/0.01% pluronic acid in
assay buffer (HBSS ) Hank’s buffered saline-check: 150 mM
NaCl, 5 mM KCl, 1 mM CaCl2, 1 mM MgCl2, plus 20 mM
HEPES, pH 7.4, 0.1% BSA and 2.5 mM probenicid) for 1 h in
5% CO2 at 37 °C. After excess dye was discarded, cells were
washed in assay buffer and layered with a final volume equal to
30 µL/well. Basal fluorescence was monitored in a fluorometric
imaging plate reader (FLIPR, Molecular Devices) with an
excitation wavelength of 488 nm and an emission range of 500
to 560 nm. Laser excitation energy was adjusted so that basal
fluorescence readings were approximately 10000 relative fluo-
rescent units. Cells were stimulated with an EC80 or EC10
concentration of glutamate in the presence of a test compound,
both diluted in assay buffer, and relative fluorescent units were
measured at defined intervals (exposure ) 0.6 s) over a 3 min
period at room temperature. Basal readings derived from negative
controls were subtracted from all samples. Maximum change in
fluorescence was calculated for each well. Concentration-
response curves derived from the maximum change in fluores-
cence were analyzed by nonlinear regression (Hill equation).
In Silico Studies. The modeling software MOE (version 2006.08,
Chemical Computing Group) was used with the built-in mmff94x force
field and GB/SA solvation model. iGluR X-ray crystal structures
(Brookhaven Protein Data Bank (PDB) data files) were submitted to
the following preparative procedure: addition of hydrogen atoms, fixing
heavy atoms, energy minimization, unfixing heavy atoms. Docking
studies were carried out with parameter setup: explicit water molecules
were excluded, ligand conformational database (LCD); placement
methodology: alpha triangle with PH4-pharmacophore filtering (R-
ammonium group, 1.6 vol) and pocket atom as restrains; London-dG
scoring-function (readout: free energy of binding in kcal/mol). LCDs
were generated by introducing two dihedral angle restrains to the Glu
analogue to impose a folded conformation Glu scaffold, then subse-
quently performing a stochastic conformational search with standard
parameter setup. (Dihedrals: +NH3CH-CH2CHR: -79.2 ( 5°, weight
100.000; CHCH2-CHRCOO-: -66.6 ( 5°, weight 100.000. Dihedral
angle values adapted from Glu when crystallized with iGluR6, PDB
code 1s50). Homology models: receptor protein sequences KA1(Q01812;
956aa), KA2(NP_113696; 979aa), iGluR5(P22756; 949aa), iGluR6-
(P42260; 908aa), iGluR7(P42264; 919aa) were obtained from the
Brookhaven Protein Data Bank and homology models built onto
iGluR5 (PDB code: 1txf) using the built-in alignment algorithm
(standard parameter setup).
(4) Wisden, W.; Seeburg, P. H. A Complex Mosaic of High-Affinity
Kainate Receptors in Rat Brain. J. Neurosci. 1993, 13, 3582–3598.
(5) Ferraguti, F.; Shigemoto, R. Metabotropic glutamate receptors. Cell
Tissue Res. 2006, 326, 483–504.
(6) Alaux, S.; Kusk, M.; Sagot, E.; Bolte, J.; Jensen, A. A.; Brauner-
Osborne, H.; Gefflaut, T.; Bunch, L. Chemoenzymatic synthesis of a
series of 4-substituted glutamate analogues and pharmacological
characterization at human glutamate transporters subtypes 1-3. J. Med.
Chem. 2005, 48, 7980–7992.
(7) Sagot, E.; Jensen, A. A.; Pickering, D. S.; Pu, X.; Umberti, M.;
Stensbøl, T. B.; Nielsen, B.; Assaf, Z.; Aboab, B.; Bolte, J.; Gefflaut,
T.; Bunch, L. Chemo-Enzymatic Synthesis of (2S,4R)-2-Amino-4-(3-
(2,2-diphenylethylamino)-3-oxopropyl)pentanedioic Acid: A Novel
Selective Inhibitor of Human Excitatory Amino Acid Transporter
Subtype 2. J. Med. Chem. 2008, 51, 4085–4092.
(8) Xian, M.; Alaux, S.; Sagot, E.; Gefflaut, T. Chemoenzymatic Synthesis
of Glutamic Acid Analogues: Substrate Specificity and Synthetic
Applications of Branched Chain Aminotransferase from Escherichia
coli. J. Org. Chem. 2007, 72, 7560–7566.
(9) Borrell, J. I.; Teixido, J.; Martinez-Teipel, B.; Matallana, J. L.; Copete,
M. T.; Llimargas, A.; Garcia, E. Synthesis and Biological Activity of
4-Amino-7-oxo-Substituted Analogues of 5-Deaza-5,6,7,8-tetrahydro-
folic Acid and 5,10-Dideaza-5,6,7,8-tetrahydrofolic Acid. J. Med.
Chem. 1998, 41, 3539–3545.
(10) O’Leary, B. M.; Szabo, T.; Svenstrup, N.; Schalley, C. A.; Lutzen,
A.; Schafer, M.; Rebek, J. “Flexiball” Toolkit: A Modular Approach
to Self-Assembling Capsules. J. Am. Chem. Soc. 2001, 123, 11519–
11533.
(11) Johnson, W. S.; Werthema, L.; Bartlett, W. R.; Brocksom, T. J.; Li,
T. T.; Faulkner, D. J.; Petersen, M. R. A Simple Stereoselective
Version Of Claisen Rearrangement Leading To trans-Trisubstituted
Olefinic Bonds. Synthesis Of Squalene. J. Am. Chem. Soc. 1970, 92,
741–743.
(12) Hermit, M. B.; Nielsen, B.; Greenwood, J. R.; Bunch, L.; Jorgensen,
C. G.; Vestergaard, H. T.; Stensbol, T. B.; Sanchez, C.; Krogsgaard-
Larsen, P.; Madsen, U.; Bra¨uner-Osborne, H. Ibotenic acid and
thioibotenic acid: a remarkable difference in activity at group III
metabotropic glutamate receptors. Eur. J. Pharmacol. 2004, 486, 241–
250.
(13) Naur, P.; Vestergaard, B.; Skov, L. K.; Egebjerg, J.; Gajhede, M.;
Kastrup, J. S. Crystal structure of the kainate receptor GluR5 ligand-
binding core in complex with (S)-glutamate. FEBS Lett. 2005, 579,
1154–1160.
(14) Mayer, M. L. Crystal structures of the GluR5 and GluR6 ligand binding
cores: molecular mechanisms underlying kainate receptor selectivity.
Neuron 2005, 45, 539–552.
(15) Nanao, M. H.; Green, T.; Stern-Bach, Y.; Heinemann, S. F.; Choe, S.
Structure of the kainate receptor subunit GluR6 agonist-binding domain
complexed with domoic acid. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
1708–1713.
(16) Bettler, B.; Egebjerg, J.; Sharma, G.; Pecht, G.; Hermansborgmeyer,
I.; Moll, C.; Stevens, C. F.; Heinemann, S. Cloning of a putative
glutamate receptor: a low affinity kainate-binding subunit. Neuron
1992, 8, 257–265.
(17) Werner, P.; Voigt, M.; Keinanen, K.; Wisden, W.; Seeburg, P. H.
Cloning of A Putative High-Affinity Kainate Receptor Expressed
Predominantly in Hippocampal CA3 Cells. Nature 1991, 351, 742–
744.
(18) Herb, A.; Burnashev, N.; Werner, P.; Sakmann, B.; Wisden, W.;
Seeburg, P. H. The KA-2 subunit of excitatory amino acid receptors
shows widespread expression in brain and forms ion channels with
distantly related subunits. Neuron 1992, 8, 775–785.
(19) Emiliozzi, R.; Pichat, L. A simple method for the preparation of
cysteinesulfinic acid. Bull. Soc. Chim. Fr. 1959, 1887–1888.
(20) Kamitori, S.; Hirotsu, K.; Higuchi, T.; Kondo, K.; Inoue, K.;
Kuramitsu, S.; Kagamiyama, H.; Higuchi, Y.; Yasuoka, N.; Kusunoki,
M.; Matsuura, Y. Overproduction And Preliminary-X-Ray Charac-
terization Of Aspartate Aminotransferase From Escherichia Coli.
J. Biochem. 1987, 101, 813–816.
Acknowledgment. We thank The Carlsberg Foundation, the
Lundbeck Foundation, The Danish Medical Research Council,
the French National Center for Scientific Research (CNRS), and
the Novo Nordisk Foundation for funding. We are grateful to
Pr. Kagamiyama’s group (Osaka Medical College) for providing
us with the AAT overexpressing E. coli strain. Dr. Susan Amara
is thanked for providing us with the EAAT cDNAs.
Supporting Information Available: Combustion analysis data
for 2a-f and 6a-e, 1H and 13C NMR spectra of compounds 2a-f.

Synthesis of 2,4-Syn-Functionalized (S)-Glutamate Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4103
(21) Nielsen, B. S.; Banke, T. G.; Schousboe, A.; Pickering, D. S.
Pharmacological properties of homomeric and heteromeric
GluR1(o) and GluR3(o) receptors. Eur. J. Pharmacol. 1998, 360,
227–238.
(22) Walker, M. W.; Jones, K. A.; Tamm, J.; Zhong, H. L.; Smith, K. E.;
Gerald, C.; Vaysse, P.; Branchek, T. A. Use of Caenorhabditis elegans
G alpha(q) chimeras to detect G-protein-coupled receptor signals.
J. Biomol. Screening 2005, 10, 127–136.
(24) Bradford, M. M. Rapid and sensitive method for quantitation of
microgram quantities of protein utilizing principle of protein-dye
binding. Anal. Biochem. 1976, 72, 248–254.
(25) Johnson, M. P.; Barda, D.; Britton, T. C.; Emkey, R.; Hornback, W. J.;
Jagdmann, G. E.; McKinzie, D. L.; Nisenbaum, E. S.; Tizzano, J. P.;
Schoepp, D. D. Metabotropic glutamate 2 receptor potentiators:
receptor modulation, frequency-dependent synaptic activity, and
efficacy in preclinical anxiety and psychosis model(s). Psychophar-
macology 2005, 179, 271–283.
(23) Coward, P.; Chan, S. D. H.; Wada, H. G.; Humphries, G. M.; Conklin,
B. R. Chimeric G proteins allow a high-throughput signaling assay of
G(i)-coupled receptors. Anal. Biochem. 1999, 270, 242–248.
JM800092X