Chemoenzymatic synthesis of a series of 4-substituted glutamate analogues and pharmacological characterization at human glutamate transporters subtypes 1-3

Article abstract of DOI:10.1021/jm050597z

A series of nine L-2,4-s;yrc-4-alkylglutamic acid analogues (1a-i) were synthesized in high yield and high enantiomeric excess (>99% ee) from their corresponding 4-substituted ketoglutaric acids (2a-i), using the enzyme aspartate aminotransferase (AAT) from pig heart or E. coli. The synthesized compounds were evaluated as potential ligands for the glutamate transporters EAAT1, EAAT2, and EAAT3 (excitatory amino acid transporter, subtypes 1-3) in the FLIPR membrane potential (FMP) assay. We found a distinct change in the pharmacological profile when the 4-methyl group (compound 1a, an EAAT1 substrate and EAAT2,3 inhibitor) was extended to a 4-ethyl group, compound 1b, as this analogue is an inhibitor at all three subtypes, EAAT1-3. Furthermore, we conclude that both large and bulky hydrophobic substituents in the 4-position of L-2,4-syn Glu are allowed by all three glutamate transporter subtypes EAAT1-3 while maintaining inhibitory activity.

Full text of DOI:10.1021/jm050597z

7980
J. Med. Chem. 2005, 48, 7980-7992
Chemoenzymatic Synthesis of a Series of 4-Substituted Glutamate Analogues
and Pharmacological Characterization at Human Glutamate Transporters
Subtypes 1-3
Sebastien Alaux,† Mie Kusk,‡ Emanuelle Sagot,† Jean Bolte,† Anders A. Jensen,‡ Hans Bra¨uner-Osborne,‡
Thierry Gefflaut,*,† and Lennart Bunch*,‡
Department of Medicinal Chemistry, The Danish University of Pharmaceutical Sciences, Universitetsparken 2,
DK-2100 Copenhagen, Denmark, and De´partement de Chimie, Universite´ Blaise Pascal, 63177 Aubie`re Cedex, France
Received June 23, 2005
A series of nine L-2,4-syn-4-alkylglutamic acid analogues (1a-i) were synthesized in high yield
and high enantiomeric excess (>99% ee) from their corresponding 4-substituted ketoglutaric
acids (2a-i), using the enzyme aspartate aminotransferase (AAT) from pig heart or E. coli.
The synthesized compounds were evaluated as potential ligands for the glutamate transporters
EAAT1, EAAT2, and EAAT3 (excitatory amino acid transporter, subtypes 1-3) in the FLIPR
membrane potential (FMP) assay. We found a distinct change in the pharmacological profile
when the 4-methyl group (compound 1a, an EAAT1 substrate and EAAT2,3 inhibitor) was
extended to a 4-ethyl group, compound 1b, as this analogue is an inhibitor at all three subtypes,
EAAT1-3. Furthermore, we conclude that both large and bulky hydrophobic substituents in
the 4-position of L-2,4-syn Glu are allowed by all three glutamate transporter subtypes
EAAT1-3 while maintaining inhibitory activity.
Introduction
erative diseases such as Alzheimer’s disease, dementia,
Huntington’s disease, amyotrophic lateral sclerosis
(ALS), epilepsy, and cerebral stroke may be directly
related to disordered glutamatergic neurotransmission
originating from dysfunction of either the Glu receptors
(iGluR or mGluR) or the Glu reuptake system (EAATs).1
In the mammalian central nervous system (CNS),
L-glutamate (Glu) functions as the major excitatory
neurotransmitter. Once released from the presynaptic
neuron into the glutamatergic synapse, Glu activates a
number of pre- and postsynaptic glutamate receptors.
On the basis of pharmacological profile and ligand
selectivity, these receptors have been grouped in two
classes: the fast acting ionotropic receptors (the iGluRs,
which are further divided into N-methyl-D-aspartate
(NMDA) receptors, R-amino-3-hydroxy-5-methyl-4-isox-
azolepropionate (AMPA) receptors, and kainate recep-
tors) and the metabotropic receptors (mGluR1-8), which
are G-protein-coupled and thus produce a much slower
signal transduction through second messenger sys-
tems.1,2 The glutamatergic neurotransmission is termi-
nated by reuptake of Glu from the synaptic cleft by the
action of sodium-dependent excitatory amino acid trans-
porters (EAATs).3,4 To this date, five subtypes have been
identified of which four, EAAT1-4, are present in the
mammalian CNS, while EAAT5 is located exclusively
in the retina. EAAT1-3 are localized on neurons and/
or astroglial cells and exhibit high capacity for trans-
porting Glu across the membrane, whereas EAAT4 and
EAAT5 function predominately as chloride channels. In
the healthy CNS, activation of Glu receptors is involved
in important neurophysiological processes such as
memory and learning, motor functions, and neural
plasticity, and development.1 However, under conditions
of metabolic stress and oxygen deprivation, Glu is a
neurotoxic agent. Thus, it is believed that neurodegen-
Considering the essential role played by the EAATs
in the maintenance of synaptic Glu concentrations below
neurotoxic levels, surprisingly little attention has been
paid to the transporters as potential drug targets in the
treatment of the plethora of neurotoxic and neurode-
generative disorders mentioned above.5,6 Recently, it has
been suggested that the EAATs may exhibit reverse
transport of Glu under ischemic conditions and that
selective inhibitors may prevent this.7 At present, only
a limited amount of EAAT inhibitors and substrates
have been described, all which, except for dihydrokainic
acid (DHK), do not show high EAAT subtype selectivity.
In a previous study, Vandenberg et al.8 characterized
the pharmacological profiles of L-2,4-syn/anti-4-methyl
and L-2,4-syn/anti-4-hydroxy Glu analogues at EAAT1
and EAAT2 (Figure 1, Table 1). In the present study,
we have explored the SAR of EAAT1-3 through the
design and synthesis of a series of L-2,4-syn-4-alkyl Glu
analogues 1b-i (Figure 1).
Chemistry
The stereoselective synthesis of 4-alkylated Glu ana-
logues 1a-i was considered using a transamination-
based chemoenzymatic approach (Scheme 1). Ami-
notransferases catalyze the reversible transfer of the
amino group from a donor amino acid to an acceptor
R-keto acid with an equilibrium constant near unity.
These broad substrate spectrum enzymes have proven
useful for the synthesis of unnatural amino acids from
the corresponding R-keto acids, provided that an equi-
* To whom correspondence should be addressed. For T.G., regarding
synthetic work: phone, +33 473407866; fax, +33 473407717; e-mail,
Thierry.GEFFLAUT@univ-bpclermont.fr. For L.B.: phone, +45 3530-
6244; fax, +45 35306040; e-mail, lebu@dfuni.dk.
† Universite´ Blaise Pascal
‡ The Danish University of Pharmaceutical Sciences.
10.1021/jm050597z CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/10/2005

Glutamate Analogues
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7981
Scheme 2. Retrosynthesis of KG Analogues 2 through
Claisen-Johnson Rearrangement
Figure 1. Chemical structures of novel and previously
reported Glu analogues 1a-l.
equilibrium shift is thus obtained because pyruvate is
not a substrate for AAT.
Table 1. Reported Pharmacology of Glu Analogues 1a, 1j, 1k,
and 1l at EAAT1,2a
A few synthetic approaches to alkyl-substituted KGs
2 have been described so far. The Claisen condensation
of the alkyl succinic ester onto diethyl oxalate gives a
mixture of 3- and 4-alkyl-KGs and requires the previous
synthesis of the corresponding 2-alkyl succinic ester.
The direct alkylation at position 4 of protected R-keto-
glutaric acid was also tried;23 however, this method is
efficient only for monomethylation. Substituted KGs 2
have also been obtained by a new route in only a few
steps and in good yields through a Claisen-Johnson
rearrangement, as depicted in the retro synthetic Scheme
2.21
In the present study, we extended this methodology
to the synthesis of nine substituted R-ketoglutaric acids
2a-i and their stereoselective transamination into the
corresponding L-2,4-syn-4-alkylglutamic acids 1a-i,
catalyzed by isolated pig heart or E. coli AAT.
4-substitution
EAAT1
EAAT2
Glu
1a
1j
1k
1l
H
20 ( 3
18 ( 3
2,4-syn-Me
2,4-anti-OH
2,4-syn-OH
2,4-anti-Me
54 ( 17b
61 ( 14c
>1000
3.4 ( 0.2
48 ( 5c
>1000
no activity
no activity
a Km values for substrates (in italics) and Ki values for nonsub-
strate inhibitors are given in µM. Data are taken from ref 8. b The
Imax value of 1a at EAAT1 was 80 ( 5% of the Imax for (S)-Glu.
c The Imax values of 1j at EAAT1 and EAAT2 were 78 ( 6% and
90 ( 4%, respectively, of the Imax values for (S)-Glu.
Scheme 1. AAT-Catalyzed Synthesis of 4-Substituted
L-Glu from the Corresponding KG Analogues
Results and Discussions
Synthesis of 4-Alkyl-2-oxoglutaric Acids 2a-i.
The syntheses of 4-alkyl-KG analogues 2a-i were
carried out according to Scheme 3. The key step is the
acid-catalyzed reaction of the allylic alcohol 5 and ortho
esters 4a-i in refluxing toluene. After transesterifica-
tion and elimination of methanol, a ketene ketal is
formed that irreversibly undergoes a Claisen-Johnson
rearrangement,24 leading to dimethyl 2-ethylidene-4-
alkylglutarate 6a-i in 70-95% yields. In every case, a
mixture of Z and E isomers was obtained with a ratio
librium shift is performed.9-18 Because glutamic acid
is a preferred substrate for most naturally occurring
aminotransferases, these enzyme are of particular
interest for the synthesis of L-Glu analogues from the
corresponding substituted ketoglutaric acids (KG). As-
partate aminotransferase (AAT, also called glutamic
oxaloacetic transaminase) from pig heart was shown to
be active toward 4-substituted ketoglutaric acids 2 and
was used to prepare both isomers of 4-hydroxy-L-Glu19
1j and 1k, 4-methyl, 4-ethyl,20 and 4-propyl derivatives21
1a, 1b, 1c and 1l, as well as the disubstituted 4-OH,4-
Me and 4,4-diMe analogues.22 In the case of the mono-
substituted 4-alkyl derivatives, AAT showed a marked
enantioselectivity in favor of the 4R isomer, allowing
the kinetic resolution of the racemic keto acid precursors
and providing a stereoselective route to the L-2,4-syn
stereomers. These syntheses were carried out according
to Scheme 1. Cysteine sulfinic acid (CSA), a close
analogue of aspartic acid, was used as the amino donor.
The ketoacid produced from CSA is the very unstable
pyruvylsulfinic acid, which undergoes a rapid decom-
position into pyruvate and sulfurous anhydride. An
1
close to 4/1 except 6i (3/2) as determined by H NMR.
As previously shown, ozonolysis of the C-C double bond
can provide the 2-ketoglutarates 7a-c.21 Alternatively,
the oxidative cleavage was accomplished with sodium
periodate in the presence of a catalytic amount of
ruthenium tetroxide.25,26 The R-keto esters 7a-i were
thus isolated in 70-90% yields. Ester hydrolysis was
performed with a stoechiometric amount of LiOH and
afforded the desired substituted lithium 2-ketoglu-
tarates 2a-i in quantitative yields. Methyl 2-(1-hy-
droxyethyl)acrylate 5 was easily prepared by the Baylis-
Hillman reaction between methyl acrylate and acet-
aldehyde in the presence of DABCO.27,28 Propanoic,
butanoic, and pentanoic ortho esters were commercially
available. Ortho esters 4d-i were prepared from the
corresponding nitriles using the Pinner method.29 Ni-
triles 3d-i were synthesized by standard, described

7982 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Alaux et al.
Scheme 3. Synthesis of 4-Alkyl-2-oxoglutaric Acids
2a-ia
nase. The optical density decrease consecutive to NADH
oxidation was measured at 340 nm and allowed to
monitor the transamination reaction. Rate measure-
ments for various substrate concentrations allowed us
to estimate the apparent Michaelis constants (Km) and
maximum velocity (Vm ) kcat(E)) in the defined experi-
mental conditions on the basis of the Michaelis-Menten
model for enzyme-catalyzed reactions. Results are re-
ported in Table 2.
The activity of pig heart AAT for both enantiomers
of 2a was already described.23 The R isomer is a very
good substrate, with Km and kcat values in the same
range as those measured for the natural substrate. The
S isomer was characterized by higher Km and lower kcat
,
resulting in a lower activity, which is still valuable for
synthetic purposes. This AAT enantioselectivity (E >
600) allowed the kinetic resolution of rac-2a, 2b,20 and
2c.21 We have now observed that when larger substit-
uents are introduced in position 4 of KG, the Km values
suffer no great changes. Except for the branched sub-
stituted 2f showing lower affinity and, to a lesser extent,
2g, 4-alkyl analogues are bound equally well as KG
itself, even the bulky pentyl, isopentyl, or benzyl deriva-
tives. Furthermore, rate measurements were performed
with the racemic derivatives and we can assume that
the determined Km values correspond to the best
recognized 4R isomer (4S for 2f), as was observed for
the 4-methyl analogue 2a. This hypothesis was con-
firmed by the structure of the transamination products
(vide infra). The Km values determined from the racemic
concentration should thus be divided by a factor of 2
because only one enantiomer is a good substrate.
Comparison with the natural substrate then indicates
that hydrophobic substituents at position 4 can actually
increase the enzyme-substrate affinity. The kcat values
also remain close to that of KG, so the kcat/Km value,
indicating the overall enzyme efficiency for the tested
reaction, is in the same order of magnitude as for the
natural substrate with the exception of 2f. Transami-
nation of KG analogues 2a-i catalyzed by E. coli AAT
a Reagents and conditions: (a) (i) MeOH, HCl, 0 °C; (ii) MeOH,
cyclohexane, reflux; (b) EtCO2H, PhCH3, reflux; (c) NaIO4, RuO4,
H2O, CCl4, CH3CN, room temp; (d) LiOH, H2O, THF, room temp.
methods. This synthesis offers a very general access to
the 4-alkyl-2-ketoglutaric acids. It should be noted that
it is applicable to 4,4-disubstituted derivatives as well
as 3-substituted analogues provided that the corre-
sponding ortho ester or the 3-substituted acrylate is
available. Moreover, the concerted nature of the Claisen
rearrangement could allow the development of a ste-
reoselective version using enantiomerically pure allylic
alcohol 5.
Aspartate Aminotransferase Activity toward
2-Ketoglutaric Acid Analogues. Kinetic constants for
the transamination reaction of the KG analogues 2a-i
were estimated using aspartic acid as the donor amino
acid in quasi-saturating concentration (40 mM). The
oxalacetate produced from Asp by transamination was
reduced by NADH in the presence of malic dehydroge-
Table 2. Kinetic Parameters for Pig Heart AAT Transamination of 2a-ia
a Values and standard errors were calculated from the Hanes-Woolf plot according to the least-squares method and Gauss’ error
c
propagation law. b Parameters for already described enantiomers of 2a23 are included for comparison. kcat and kcat/Km are expressed as
percent relative to the natural unsubstituted KG substrate.

Glutamate Analogues
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7983
was also studied. The Km and relative kcat values
measured with the bacterial enzyme are very similar
to those reported in Table 1 for the pig heart enzyme.
This high activity of AAT toward the 4-substituted
analogues is somewhat surprising; however, Kirsch and
co-workers already pointed out that the presence of a
long aliphatic chain on substrates does not decrease the
E. coli enzyme-ligand complex stabilization.30 Numer-
ous X-ray structures of AAT complexed with substrates
or inhibitors have been published.31-34 A careful inspec-
tion of a chicken mitochondrial enzyme-substrate
structure35 (PDB code 1MAQ) allowed us to observe that
the 4-proR hydrogen is pointed to the outside of the
active site and is surrounded by hydrophobic side chains
creating a favorable environment for alkyl group sta-
bilization in that position.
we have been looking for additional evidence of the
stereochemical assignments. 2a-e and 2i were submit-
ted to reductive amination by reaction with ammonia
and reduction of the imine with NaBH4. The four-
stereomer mixture thus obtained for each compound was
analyzed by chiral HPLC on a Crownpack column.
Under the used experimental conditions and for all the
studied amino acids, D-isomers were always eluted first.
The chromatogram obtained with the 4-methylglutamic
acids 1a showed four signals with 8, 9, 18, and 28 min
retention times. Peaks observed at 18 and 28 min were
attributed to the L-2,4-syn and L-2,4-anti isomers,
respectively, by comparison with authentic samples
previously prepared with AAT. For each mixture stud-
ied, conditions were found to resolve the four stereomers
and similar chromatographic profiles were observed.
Analyses of the samples obtained by enzymatic trans-
amination showed only one signal always corresponding
to the third eluted isomer of the corresponding mixture
obtained by chemical amination. Since the order of
elution cannot be considered as absolute evidence for
the absolute configuration, a complementary experiment
was performed in the case of the bulky 4-pentyl-L-
glutamic acid 1e. This compound was converted into
4-pentyl-L-pyroglutamic acid ethyl ester 8 by reaction
with thionyl chloride and EtOH. This constrained
derivative was then studied by NMR spectroscopy.
Comparison of the observed coupling constants with the
theoretical values estimated by molecular modeling for
the expected (3R,5S)-8, as well as experimental NOE,
is reported in Figure 2.
The measured coupling constants are in good agree-
ment with the calculated values for the expected pre-
ferred conformation II of (3R,5S)-8. Analysis of the NOE
is also consistent with the proposed structure. In
contrast, the experimental data cannot account for the
proposed structure of the (3S,5S)-stereomer and espe-
cially for the expected antiperiplanar position of H3
toward H1 and H4 in structure III. These experiments
confirmed previous findings and were considered to be
demonstrative of the general stereoselectivity of the
AAT catalyzed transamination providing L-2,4-syn-4-
alkylglutamic acids.
Pharmacology and Modeling. All synthesized 4-sub-
stituted Glu analogues 1a-k were characterized at the
glutamate transporter subtypes EAAT1, EAAT2, and
EAAT3, using a FLIPR membrane potential (FMP)
assay (see Experimental Section for details). The assay
was performed essentially as described previously,36 and
the results are summarized in Table 3. Compound 1a
was found to be a substrate at EAAT1, whereas it was
an inhibitor at the two other EAAT subtypes. The
4-hydroxy Glu analogues 1j and 1k were substrates at
all three EAAT subtypes (EAAT1-3), the L-2,4-anti
diastereomer 1j being considerably more potent than
the L-2,4-syn diastereomer 1k (Table 3). The pharma-
cological properties of 1a, 1j, and 1k in the FMP assay
were in good agreement with data previously obtained
for these Glu analogues in electrophysiological record-
ings on EAAT1 and EAAT2 expressed in Xenopus
oocytes (Table 1).8 The eight novel 4-substituted Glu
analogues presented in this study (compounds 1b-h)
were all inhibitors at all three EAAT subtypes, where
they displayed Ki values in the mid-to-high micromolor
Synthesis of Glutamic Acid Analogues. The en-
zymatic transaminations were performed according to
Scheme 1 as described above. In early experiments, we
observed a loss of enzyme activity as the reaction
proceeded. This was not considered before as a major
obstacle for small-scale syntheses; however, we tried in
this study to minimize enzyme inhibition. It was at-
tributed to the use of CSA as the amino donor substrate.
Indeed, sulfur dioxide produced from CSA in the reac-
tion medium is hydrated into sulfites and the strongly
-
nucleophilic HSO3 anions interact with the pyridoxal
coenzyme bound at the active site as several imine
intermediates. E. coli enzyme proved to be less sensitive
to inhibition than pig heart AAT. Moreover, we have
found that this inactivation can be limited by the
addition of acetaldehyde to the reaction mixture; the
aldehyde acts as a hydrogenosulfite scavenger through
bisulfitic adduct formation and allows faster reactions
and yield improvements. The 4-alkyl-KG derivatives
2a-i were submitted to the action of pig heart or E.
coli AAT in the presence of CSA and acetaldehyde. The
initial concentration was 20 mM in both substrates and
40 mM in acetaldehyde. The reaction was monitored by
enzymatic assay of pyruvic acid produced from CSA,
using lactic acid dehydrogenase and NADH. The reac-
tion was stopped when a conversion rate between 40%
and 50% was reached, to achieve a good kinetic resolu-
tion of the racemic substrates. The reaction mixture was
then poured onto a short column of a sulfonic resin (H+
form). The column was washed with water to elute the
unreacted KG analogue and then CSA, which was only
weakly adsorbed because of the strong acidity of the
sulfinic moiety. Glu analogue 1 was then eluted with
diluted ammonia. The amino acid containing fractions
were concentrated, and the desired Glu analogues were
thus isolated in 32-42% yields from racemic KGs and
with purities over 90%. The purity was further in-
creased to more than 98% by a second chromatography
on anion-exchange resin (AcO- form) using an acetic
acid gradient for elution. For each synthesis one stere-
omer only was obtained with diastereomeric excess over
1
98% as determined by H NMR and HPLC analyses.
Since AAT is known to be highly stereoselective toward
amino acids of the L series, an S configuration was
assigned to carbon 2. Carbon 4 configuration was
expected to be R in any case except for 1f by analogy
with the well described 4-methyl and 4-ethyl analogues
already obtained by AAT transamination.20 However,

7984 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Alaux et al.
Figure 2. Conformational and NMR analyses of compound 8.
Table 3. Pharmacological Characteristics of (S)-Glu and Its 4-Substituted Analogues 1a-k at Human EAAT1, EAAT2, and EAAT3
in the FMP Assaya
4-substitution
EAAT1
EAAT2
EAAT3
(S)-Glu
1a
1b
1c
1d
1e
1f
1g
1h
1i
H
9.2 [5.04 ( 0.05]
13 [4.92 ( 0.11]b
23 [4.64 ( 0.01]
13 [4.99 ( 0.20]
50 [4.32 ( 0.09]
120 [3.93 ( 0.04]
35 [4.54 ( 0.15]
55 [4.39 ( 0.20]
29 [4.54 ( 0.03]
13 [4.94 ( 0.12]
140 [4.06 ( 0.28]c
∼1000 [∼3.0]
21 [4.69 ( 0.05]
13 [4.93 ( 0.16]
14 [4.97 ( 0.23]
22 [4.68 ( 0.12]
42 [4.42 ( 0.12]
77 [4.17 ( 0.16]
19 [4.79 ( 0.17]
22 [4.66 ( 0.07]
26 [4.66 ( 0.19]
36 [4.49 ( 0.12]
67 [4.21 ( 0.13]c
∼1,000 [∼3.0]
25 [4.61 ( 0.06]
6.6 [5.27 ( 0.20]
38 [4.43 ( 0.07]
32 [4.50 ( 0.07]
80 [4.15 ( 0.14]
98 [4.06 ( 0.14]
51 [4.30 ( 0.06]
300 [3.53 ( 0.06]
81 [4.18 ( 0.20]
66 [4.21 ( 0.11]
81 [4.16 ( 0.13]c
∼1000 [∼3.0]
L-2,4-syn-Me
L-2,4-syn-Et
L-2,4-syn-Pr
L-2,4-syn-Bu
L-2,4-syn-Pn
L-2,4-syn-iPr
L-2,4-syn-iBu
L-2,4-syn-iPn
L-2,4-syn-Bn
L-2,4-anti-OH
L-2,4-syn-OH
1j
1k
a Km values for substrates (in italics) and Ki values for nonsubstrate inhibitors are given in µM with pKm ( SEM and pKi ( SEM
values in brackets. b The Rmax value of 1a at EAAT1 was 48 ( 8% of the Rmax for (S)-Glu. c The Rmax values of 1j at EAAT1, EAAT2, and
EAAT3 were 82 ( 3%, 68 ( 9%, and 87 ( 2%, respectively, of the Rmax values for (S)-Glu.
ranges (Table 3). Compound 1g displayed a slightly
higher Ki value at EAAT3 than at the two other
transporters, but overall the eight ligands were nonse-
lective among the three EAATs.
formation at the iGluRs by X-ray crystallographic
studies39 and rational ligand design studies.40
To identify biologically relevant conformational min-
ima, we submitted 4-alkyl Glu analogues 1a-h to a
stochastic conformational search in the modeling soft-
ware MOE (see Experimental Section for details). For
each compound except 1f and 1i, we located three
important low-energy conformations: one folded and
two extended conformations (Table 4). On the other
hand, the stochastic conformational search of com-
pounds 1f and 1i returned only two low-energy confor-
mation: one folded and one extended (denoted A). For
all nine Glu analogues, the folded conformation is
favored by an average of 2.1 kcal/mol over the extended
A conformation and by an average 3.4 kcal/mol (com-
pounds 1i and 1f not included) over the extended B
conformation (Table 4). It is noteworthy that the ex-
tended Glu conformation, denoted A, is similar to the
conformation of Glu found when crystallized together
with the extracellular domain of the metabotropic Glu
receptor mGluR1.41
Two major observations can be extracted from the
characterization of the 11 4-substituted Glu analogues
1a-k at EAAT1-3. First, extension of the L-2,4-syn-4-
alkyl substituent from a methyl group, compound 1a,
to an ethyl group, compound 1b, marks a clear change
in the pharmacological profile from substrate to inhibi-
tor at the EAAT1 subtype. This, with the inhibitory
activity at EAAT2 and EAAT3, is maintained. Second,
this study shows that both large and bulky substituents
of a hydrophobic nature in the 4-position, compounds
1b-h, are readily accommodated by all three EAAT
subtypes investigated here (EAAT1-3). Thus, we pre-
dict that EAAT subtype selectivity is likely not obtained
by introduction of another alkyl substituent in the L-2,4-
syn-4-position of Glu.
Structure-Activity-Relation Study. It has been
proposed by other research groups that substrates and
inhibitors bind in two distinct ways at the EAATs.5,37,38
While substrates are believed to bind in an extended
Glu conformation, the binding conformation of inhibitors
is assumed to be equivalent to the folded Glu conforma-
tion previously identified as the agonist binding con-
EAAT Inhibitor Pharmacophore. To clarify the
size and bulkiness of the 4-alkyl substituent allowed
while maintaining inhibitory activity at EAAT1-3, the
low-energy folded conformation of compounds 1a-i,
were superimposed by fitting the ammonium and the

Glutamate Analogues
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7985
Table 4. Biologically Important Low-Energy Conformations of
4-Alkyl Glu Analogues 1a-i
total free conformational energy (kcal/mol)a
∆G-
∆G-
∆G-
b
c
(folded) (extended A) (extended B) ∆∆GA
∆∆GB
Figure 4. Superimposition by fitting the ammonium group
and the two carboxylate groups of calculated low-energy folded
conformations (up to +1 kcal/mol) of compounds 1a-i. The two
conformations of the phenyl ring in compound 1i are high-
lighted: lowest energy conformation (yellow); +0.66 kcal/mol
(green).
1a -90.20
1b -89.34
-87.50
-86.95
-86.76
-86.81
-86.72
-80.65
-80.96
-81.16
-73.20
-86.53
-85.73
-85.74
-85.37
-85.30
cnf d
-79.56
-79.50
cnf d
-2.70
-3.67
-3.61
-3.03
-3.04
-3.11
-2.39
-2.01
-1.87
-1.69
1c
-88.77
1d -88.68
1e -88.41
1f
-83.11
-2.46 <-7.0
1g -83.04
1h -83.46
-2.08
-2.30
-3.48
-3.96
1i
-75.03
-1.83 <-7.0
a For details, see Experimental Section. b ∆∆GA calculated
from ∆G(folded)
-
∆G(extended A). c ∆∆GB calculated from
∆G(folded) - ∆G(extended B). d cnf: conformation not found.
Figure 5. Superimposition by fitting the ammonium group
and the two carboxylate groups of calculated low-energy folded
conformations (up to +1 kcal/mol) of compounds 1a-i and
;-threo-â-benzyloxyaspartate (TBOA) (low-energy conforma-
tion, yellow; +0.93 kcal/mol, purple).
Figure 3. Superimposition by fitting the ammonium group
and the two carboxylate groups of calculated low-energy folded
conformation of 1a-i. Two conformations of the phenyl ring
in compound 1i are highlighted: lowest energy conformation
(yellow); +0.66 kcal/mol (green).
on the brink of our 4-alkyl substituents (Figure 5). This
supports our hypothesis that the 4-alkyl substituents
(compounds 1a-i) are most likely directed into a larger
area in space rather than into a pocket of the EAAT
protein.
two carboxylate groups (Figure 3). In eight of the nine
compounds, the 4-alkyl substituents arrange themselves
in a flat planar fashion away from the Glu backbone,
except for the 4-benzyl Glu analogue 1i (benzyl group
in yellow, Figure 3), of which the 4-benzyl group is
oriented perpendicular. To fit the orientation of the
4-benzyl group to the other 4-alkyl substituents, a
slightly higher energy conformation (+0.66 kcal/mol) of
1i is used (benzyl group in green, Figure 3). On the other
hand, if a +1 kcal/mol margin of error is allowed, the
number of conformations found for the 4-alkyl substit-
uents is highly increased. Superimposition of all of these
conformations suggests a 4-alkyl region, which is now
best characterized as spherical (Figure 4). On the basis
of these modeling considerations and the fact that we
do not see significant EAAT subtype selectivity, we put
forward that these 4-alkyl substituents are likely ori-
ented out in space rather than into a pocket of the EAAT
protein. Thus, we suggest that the observed decrease
in potency as the alkyl group is grown longer (com-
pounds 1a-e) is primarily due to loss of entropic energy.
Recently, some aryl analogues of L-threo-â-benzyloxy-
aspartate (TBOA) were published as being highly
potent, yet not subtype selective, EAAT inhibitors.42 We
therefore decided to probe if the large arylic substituents
of these novel TBOA analogues were likely to occupy
the same area in space as our 4-alkyl substituents.
Superimposition of low-energy conformations of TBOA
with low-energy conformations of 1a-i (up to +1 kcal/
mol) suggests that the phenyl ring in TBOA is oriented
EAAT Substrate Pharmacophore. It is important
to stress that any compound that can be accommodated
by the EAAT substrate pocket is to be characterized as
a substrate, even though its folded conformation (inhibi-
tory conformer) is energetically favored. An example is
the L-2,4-syn-4-methyl analogue 1a, which is a substrate
at EAAT1, while it is an inhibitor at EAAT2 and EAAT3
(Tables 3 and 4). For an improved understanding of the
pharmacophore of the EAAT substrate binding pocket,
it is essential to determine which extended conformer,
extended A or extended-B, is the biologically active one.
On the basis of total free energy calculations, this
question cannot be answered because the difference
amounts to only <1 kcal/mol (Table 4). We therefore
turned to investigate the low-energy conformations of
L-2,4-anti-4-methyl Glu (1l), which is completely inac-
tive at all EAAT subtypes (Table 1). When this com-
pound is submitted to a stochastic conformational
search, only one low-energy extended conformation is
found (denoted extended AA). Superimposition of this
conformer with the extended A conformer of 1a and the
two well-described rigid EAAT substrates t-PDC and
L-2,4-MPDC shows that the two diastereomeric 4-meth-
yl substituents occupy the same area in space (Figure
6). Since 1l is inactive at EAAT1-3, the extended-A/
extended-AA Glu conformations are likely not the
bioactive substrate conformers, and we suggest that the

7986 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Alaux et al.
Table 5. Biological Relevant Low-Energy Conformations of
1j and 1k
Figure 6. Pharmacophore model A of substrate binding
conformation. Superimposition was done by fitting the am-
monium group and the two carboxylate groups of calculated
low-energy conformations of L-2,4-MPDC (green), t-PDC (yel-
low), extended A 1a (purple), extended AA 1l (type code).
total free conformational energy (kcal/mol)a
∆G- ∆G-
(folded) (extended A) (extended B) ∆∆GA
∆G-
b
c
∆∆GB
1j
-70.33
-64.23
-69.00
-70.75
-56.56
-6.10
+4.26
+0.42
-8.18
1k -64.74
a For details, see Experimental Section. b ∆∆GA calculated
from ∆G(folded) - ∆G(extended A). c ∆∆GB calculated from
∆G(folded) - ∆G(extended B).
Figure 7. Pharmacophore model B of substrate binding
conformation. Superimposition was done by fitting the am-
monium group and the two carboxylate groups of calculated
low-energy conformations of L-2,4-MPDC (green), t-PDC (yel-
low), extended B 1a (purple), extended B 1l (type code).
area in space occupied by these two methyl groups is
most likely forbidden in the EAAT substrate pharma-
cophore.
Also, the conformer of 1l, which resembles the ex-
tended B conformer of 1a, can only be considered at the
theoretical level because of the imposed highly disfa-
vored interaction between the L-2,4-anti-4-methyl group
and the ammonium group (Figure 7). Thus, we suggest
that the extended B conformer of 1a (Figure 7) best
describes the substrate binding conformation at the
EAATs (pharmacophore model B) and that the origin
of the observed difference in pharmacology of com-
pounds 1a and 1b is due to differentiated structural
features of the substrate binding pocket in each EAAT
subtype. It remains to be identified which exact amino
acids in the EAAT proteins are involved.
We then turned to investigate the origin of difference
in pharmacology of the two diastereomeric 4-OH Glu
analogues 1j and 1k. First, we wanted to address why
1j is a substrate despite the fact that its 4-methyl
replacement analogue 1l is completely inactive. Second,
we wanted to investigate why 1k is a low potent
substrate even though its 4-methyl replacement ana-
logue 1a is indeed a potent substrate at EAAT1. Thus,
amino acid alcohols 1j and 1k were submitted to a
stochastic conformational search in MOE, which re-
vealed three low-energy conformations for each com-
pound (Table 5). The conformational energies calculated
for 1j suggest the folded and extended B conformations
to exist in about 1:1 equilibrium, while the extended A
conformation is unfavored by +6.52 kcal/mol (relative
to the extended B conformation). Because the extended
B conformation is believed to be the active substrate
binding conformation, these free energy calculations are
thus in line with the observed substrate character of 1j
at EAAT1-3. The conformational energies calculated
for 1k show a clear preference for the extended A
conformation by -4.26 kcal/mol (relative to the folded
conformation) and as much as -12.44 kcal/mol over the
extended B conformation. The highly favorable hydro-
Figure 8. Pharmacophore model of substrate binding con-
formation, showing three-point superimposition of calculated
low-energy conformations of L-2,4-MPDC (green), t-PDC (yel-
low), extended B 1j (purple), extended A 1k (type code).
gen bond interaction between the ammonium group and
the alcohol in the extended A conformation is largely
accountable for this observation. However, 1k is a low
potent substrate at EAAT1-3, which may be explained
from a superimposition study of the known rigid sub-
strates L-2,4-MPDC and t-PDC with extended B 1j and
extended A 1k (Figure 8).
It is evident that the 4-hydrogen in the extended A
conformer of 1k is found in the area previously identi-
fied as forbidden (see above and Figure 6). We thus
suggest that this observation and the fact that the
extended B conformer of 1k is not energetically favored
are the two major reasons for the observed low substrate
potency of 1k at EAAT1-3.
Conclusion
In conclusion, nine Glu analogues were synthesized
in a stereoselective fashion taking advantage of a
chemoenzymatic strategy. Aspartate aminotransferase
proved to be useful for the preparation of L-2,4-syn-4-
alkylglutamic acid analogues from the corresponding
racemic ketoglutarates. These latter compounds were
prepared by a simple and general procedure. The 11
compounds were then characterized at human EAAT1,
EAAT2, and EAAT3 in the FMP assay. A distinct
change in the pharmacological profile was observed
when the 4-methyl group (compound 1a, an EAAT1
substrate and EAAT2,3 inhibitor) was extended to a
4-ethyl group, compound 1b, as this analogue is an
inhibitor at all three subtypes EAAT1-3. We have
shown that quite large and bulky hydrophobic substit-
uents in the 4-position of Glu are accommodated by

Glutamate Analogues
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7987
1.50 (2H, q, J ) 7.5 Hz), 0.89 (6H, d, J ) 7.0 Hz); 13C NMR
(100 MHz, CDCl3) δ 119.8, 33.6, 27.1, 21.6, 14.9.
EAAT1-3, maintaining the inhibitory activity. Our
modeling study presents new insight on the origin of
the diverse pharmacological profiles of structurally close
Glu analogues 1a, 1j, 1k, and 1l. Furthermore, our two
suggested binding conformations, one for substrates and
one for inhibitors, clarify the EAAT1-3 structure-
activity relations and are in agreement with results
from earlier reports.
5-Methylhexanenitrile 3h. Sodium (23 g, 1 mol) was
added to absolute EtOH (500 mL). After complete reaction,
ethyl cyanoacetate (200 g, 1.77 mol) was added slowly before
1-bromo-3-methylbutane (151 g, 1 mol). The mixture was
stirred under reflux for 10 h until neutralization of the reaction
mixture was indicated. After cooling, the mixture was filtered
on Celite and ethanol was evaporated under reduced pressure.
The residue was diluted with water (500 mL) and extracted
with Et2O (3 × 100 mL). The organic layer was dried over
MgSO4 and concentrated under reduced pressure. The residue
was distilled under reduced pressure to give 115 g of ethyl
2-cyano-5-methylhexanoate (88-91 °C, 10 mmHg). To a solu-
tion of this product (50 g, 0.27 mol) in DMSO (250 mL) were
added water (10.7 mL, 0.6 mol) and NaCl (5.26 g, 0.09 mol).
The mixture was heated under reflux for 2 h, then diluted with
water (100 mL) and extracted with Et2O (3 × 100 mL). The
organic layer was washed with brine (2 × 50 mL), dried over
MgSO4, and concentrated under reduced pressure. The residue
was distilled under reduced pressure (75-80 °C, 30 mmHg)
to give 3h (24.7 g, 52% overall yield) as a colorless liquid: IR
(neat film) 2244 cm-1; 1H NMR (400 MHz, CDCl3) δ 2.29 (2H,
t, J ) 7.0 Hz), 1.65-1.48 (3H, m), 1.28 (2H, m), 0.86 (6H, d,
J ) 7.0 Hz); 13C NMR (100 MHz, CDCl3) δ 119.7, 37.7, 27.3,
23.2, 22.2, 17.2.
General Procedure for the Synthesis of Trimethyl-
ortho Esters 4d-i. To a solution of nitrile (0.2 mol) and
MeOH (9.7 mL, 0.24 mol) in anhydrous Et2O (200 mL) was
added gaseous HCl (0.6 mol) dried by passing through con-
centrated H2SO4 and generated in a separated vessel by
dropwise addition of concentrated HCl (50 mL) into concen-
trated H2SO4 (75 mL). The solution obtained was kept at 4 °C
for 2-5 days to complete the imidate hydrochloride crystal-
lization. The solid was isolated by filtration under N2, washed
with anhydrous Et2O, and dried under reduced pressure in
the presence of KOH pellets. MeOH (41.1 mL, 0.6 mol) was
added to a suspension of the imidate hydrochloride in anhy-
drous hexane (400 mL). The mixture was stirred at room
temperature for 48 h. The precipitated ammonium chloride
was discarded by filtration, and K2CO3 (50 mg, 0.36 mmol)
was added to the filtrate before concentration under reduced
pressure. The residue was distilled under reduced pressure
to give the ortho ester isolated as a colorless liquid.
Experimental Section
Chemistry. Melting points were determined on a Reichert
hot-stage apparatus and are uncorrected. IR spectra were
recorded on a Perkin-Elmer 801 spectrophotometer. 1H 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 experiments were per-
formed at the Centre Regional de Mesures Physiques de
l’Universite´ de Rennes, France, using an MS/MS ZABspec TOF
micromass spectrometer. Optical rotations were determined
with a JASCO DIP 370 polarimeter and are reported at the
sodium D line (589 nm). Elemental analyses were performed
at the Service Central d’Analyse du CNRS, Solaize, France.
Silica gel 60 (Merck, 40-63 µm) and precoated F254 plates were
used for column and TLC chromatographies. All solvents were
purified by distillation following the usual procedures. HPLC
analyses of 1a-e and 1i were performed at 220 nm using a
Crownpack column (Daicel, 4.0 mm × 150 mm) with the
following conditions: 1a, HClO4, pH 1, 0.3 mL min-1, 25 °C;
1b, HClO4, pH 1, 0.5 mL min-1, 25 °C; 1c-d, HClO4, pH 1,
15% MeOH (v/v), 1 mL min-1, 25 °C; 1e, HClO4, pH 1, 15%
MeOH (v/v), 1.6 mL min-1, 43 °C; 1i, HClO4, pH 2, 15% MeOH
(v/v), 1.3 mL min-1, 25 °C. Racemic reference samples were
prepared from 2a-i using a described procedure for reductive
amination.43
Nitrile 3d and ortho esters 4a-c were purchased from
Aldrich. 3-Phenylpropanenitrile 3i was prepared from cinna-
monitrile following a described procedure.44 Cysteine sulfinic
acid was prepared from cystine following a described proce-
dure.45 Pig heart aspartate aminotransferase, bovine heart
malic dehydrogenase, and rabbit muscle lactic dehydrogenase
were purchased from Sigma and Roche. E. coli aspartate
aminotransferase was produced and purified following a
described procedure from an overexpressing E. coli strain
TY103 transformed with pUC19-aspC.46,47 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 IU), and malic
dehydrogenase (2 IU) in a total volume of 1 mL. Rates were
Trimethyl Orthohexanoate 4d. Yield 24.8 g, 70%; bp 64-
1
68 °C (16 mmHg); H NMR (400 MHz, CDCl3) δ 3.20 (9H, s),
1.68 (2H, m), 1.28 (6H, m), 0.88 (3H, t, J ) 7.0 Hz); 13C NMR
(100 MHz, CDCl3) δ 115.7, 49.1, 31.8, 30.2, 22.5, 22.4, 13.8.
Trimethyl Orthoheptanoate 4e. Yield 16.4 g, 43%; bp
75-80 °C (18 mmHg); 1H NMR (400 MHz, CDCl3) δ 3.18 (9H,
s), 1.67 (2H, m), 1.25 (8H, m), 0.85 (3H, t, J ) 7.0 Hz); 13C
NMR (100 MHz, CDCl3) δ 115.8, 49.2, 31.7, 30.3, 29.1, 22.7,
22.5, 14.0.
calculated from the OD linear decay at 340 nm using ꢀNADH
)
6220 cm-1 M-1
.
General Procedure for the Syntheses of Nitriles 3e-
g. A mixture of finely powdered KCN (48.8 g, 0.375 mol), water
(13.4 mL, 0.375 mol), bromoalcane (0.5 mol), and Aliquat 336
(4.6 mL, 10 mmol) was heated at 85 °C. The reaction was
monitored by TLC for 4-48 h. After addition of Et2O (100 mL),
the mixture was filtered on Florisil, and the filtrate was dried
over MgSO4 and concentrated under reduced pressure. The
residue was distilled under reduced pressure to give the
desired nitrile isolated as a colorless liquid.
Trimethyl 3-Methylorthobutanoate 4f. Yield 17.5 g,
1
54%; bp 46-48 °C (12 mmHg); H NMR (400 MHz, CDCl3) δ
3.20 (9H, s), 1.77 (1H, non., J ) 7.0 Hz), 1.60 (2H, d, J ) 7.0
Hz), 0.95 (6H, d, J ) 7.0 Hz); 13C NMR (100 MHz, CDCl3) δ
115.7, 49.1, 38.1, 23.4, 23.2.
Trimethyl 4-Methylorthopentanoate 4g. Yield 12.7 g,
1
38%; bp 65-67 °C (15 mmHg); H NMR (400 MHz, CDCl3) δ
3.20 (9H, s), 1.69 (2H, m), 1.50 (1H, non., J ) 7.0 Hz), 1.18
(2H, m), 0.88 (6H, d, J ) 7.0 Hz); 13C NMR (100 MHz, CDCl3)
δ 116.0, 49.3, 31.6, 28.1, 27.9, 22.5.
Heptanenitrile 3e. Yield 50.5 g, 91%; bp 82-85 °C (30
mmHg); IR (neat film) 2247 cm-1; 1H NMR (400 MHz, CDCl3)
δ 2.34 (2H, t, J ) 7.0 Hz), 1.66 (2H, quint, J ) 7.0 Hz), 1.45
(2H, m), 1.32 (4H, m), 0.90 (3H, t, J ) 7.0 Hz); 13C NMR (100
MHz, CDCl3) δ 119.6, 30.7, 28.1, 25.1, 22.1, 16.8, 13.7.
Trimethyl 5-Methylorthohexanoate 4h. Yield 19.0 g,
1
50%; bp 80-84 °C (12 mmHg); H NMR (400 MHz, CDCl3) δ
3.22 (9H, s), 1.69 (2H, m), 1.53 (1H, non., J ) 6.5 Hz), 1.32
(2H, m), 1.18 (2H, m), 0.87 (6H, d, J ) 6.5 Hz); 13C NMR (100
MHz, CDCl3) δ 115.9, 49.4, 38.9, 30.7, 28.0, 22.6, 20.6.
3-Methylbutanenitrile 3f. Yield 30.8 g, 74%; bp 56-58
1
°C (16 mmHg); IR (neat film) 2247 cm-1; H NMR (400 MHz,
CDCl3) δ 2.20 (2H, d, J ) 7.0 Hz), 1.95 (1H, non., J ) 7.0 Hz),
1.00 (6H, d, J ) 7.0 Hz); 13C NMR (100 MHz, CDCl3) δ 118.7,
25.8, 25.7, 21.5.
Trimethyl 3-Phenylorthopropanoate 4i. Yield 19.7 g,
1
47%; bp 114-120 °C (10 mmHg); H NMR (400 MHz, CDCl3)
4-Methylpentanenitrile 3g. Yield 37.8 g, 78%; bp 70-75
δ 7.34-7.20 (5H, m), 3.31 (9H, s), 2.71 (2H, m), 2.09 (2H, m);
13C NMR (100 MHz, CDCl3) δ 141.6, 128.1, 128.0, 125.8, 115.3,
50.4, 32.1, 29.2.
1
°C (16 mmHg); IR (neat film) 2251 cm-1; H NMR (400 MHz,
CDCl3) δ 2.30 (2H, t, J ) 7.5 Hz), 1.68 (1H, non., J ) 7.0 Hz),

7988 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Alaux et al.
General Procedure for the Synthesis of Dialkyl 2-Eth-
ylidene-4-alkylglutarate 6a-i. To a solution in anhydrous
toluene (25 mL) of methyl 2-(1-hydroxyethyl)acrylate 5 (3.25
g, 25 mmol) prepared following a described procedure48 were
added the ortho ester (50 mmol) and propionic acid (0.18 mL,
2.5 mmol). The mixture was heated under reflux for 1 h before
methanol was slowly removed by azeotropic distillation. The
solution was concentrated under reduced pressure. Flash
chromatography (eluent, cyclohexane-AcOEt, 9:1, v/v) af-
forded 6a-i isolated as colorless liquids and consisting of a
mixture of Z and E isomers in an approximately 4/1 ratio
except for 6i (0.64/0.36).
(400 MHz, CDCl3) δ 6.92 and 6.05 (1H, q, J ) 7.0 Hz, E and
Z isomers), 3.74 and 3.73 (3H, s, Z and E isomers), 3.64 and
3.60 (3H, s, E and Z isomers), 2.75-2.25 (3H, m), 1.96 and
1.80 (3H, d, J ) 7.0 Hz, Z and E isomers), 1.54 (2H, m), 1.25
(1H, m), 0.88 and 0.86 (6H, d, J ) 7.0 Hz, E and Z isomers);
13C NMR (100 MHz, CDCl3) δ 176.4, 176.3, 167.8, 139.8, 139.5,
130.4, 129.8, 51.6, 51.4, 51.3, 51.2, 43.3, 42.9, 41.6, 41.4, 37.8,
29.7, 26.2, 26.1, 22.2, 22.1, 15.9, 14.4.
Dimethyl 2-Ethylidene-4-isopentylglutarate 6h. Yield
5.0 g, 78%; IR (neat film) 1737, 1723, 1647 cm-1; 1H NMR (400
MHz, CDCl3) δ 6.94 and 6.08 (1H, q, J ) 7.0 Hz, E and Z
isomers), 3.76 and 3.74 (3H, s, Z and E isomers), 3.64 and 3.63
(3H, s, Z and E isomers), 2.70-2.25 (3H, m), 1.98 and 1.82
(3H, d, J ) 7.0 Hz, Z and E isomers), 1.70-1.10 (5H, m), 0.88
and 0.87 (6H, d, J ) 7.0 Hz, E and Z isomers); 13C NMR (100
MHz, CDCl3) δ 176.1, 176.0, 167.8, 139.6, 139.3, 130.4, 129.8,
51.5, 51.24, 51.18, 51.1, 45.3, 45.0, 38.3, 36.3, 36.2, 30.2, 30.0,
29.3, 27.9, 22.5, 22.3, 15.7, 14.3.
Methyl, Ethyl 2-Ethylidene-4-methylglutarate 6a. Yield
1
4.18 g, 78%; IR (neat film) 1731, 1714, 1647 cm-1; H NMR
(400 MHz, CDCl3) δ 6.95 and 6.07 (1H, q, J ) 7.0 Hz, E and
Z isomers), 4.09 (2H, q, J ) 7.0 Hz), 3.74 and 3.73 (3H, s, Z
and E isomers), 2.71-2.30 (3H, m), 1.97 and 1.82 (3H, d, J )
7.0 Hz, Z and E isomers), 1.20 (3H, t, J ) 7.0 Hz), 1.12 and
1.10 (3H, d, J ) 7.0 Hz, E and Z isomers); 13C NMR (100 MHz,
CDCl3) δ 176.0, 167.8, 139.7, 139.5, 130.4, 129.8, 60.1, 51.6,
51.1, 38.9, 38.7, 30.2, 16.8, 15.7, 14.5, 14.2.
Dimethyl 2-Ethylidene-4-benzylglutarate 6i. Yield 6.53
1
g, 95%; IR (neat film) 1735, 1720, 1648, 1604 cm-1; H NMR
(400 MHz, CDCl3) δ 7.35-7.10 (5H, m), 6.95 and 6.07 (1H, q,
J ) 7.0 Hz, E and Z isomers), 3.73 and 3.72 (3H, s, Z and E
isomers), 3.54 and 3.53 (3H, s, Z and E isomers), 3.00-2.45
(5H, m), 1.98 and 1.76 (3H, d, J ) 7.0 Hz, Z and E isomers);
13C NMR (100 MHz, CDCl3) δ 175.3, 175.2, 167.5, 140.2, 139.8,
139.0, 130.7, 129.5, 128.8, 128.3, 126.3, 51.6, 51.33, 51.25, 51.2,
47.1, 46.8, 38.4, 38.2, 37.3, 29.2, 15.9, 14.3.
Dimethyl 2-Ethylidene-4-ethylglutarate 6b. Yield 3.92
1
g, 73%; IR (neat film) 1736, 1719, 1647 cm-1; H NMR (400
MHz, CDCl3) δ 6.87 and 6.02 (1H, q, J ) 7.0 Hz, E and Z
isomers), 3.69 and 3.68 (3H, s, Z and E isomers), 3.58 and 3.57
(3H, s, Z and E isomers), 2.60-2.40 (3H, m), 1.92 and 1.75
(3H, d, J ) 7.0 Hz, Z and E isomers), 1.60-1.40 (2H, m), 0.85
and 0.83 (3H, t, J ) 7.0 Hz, E and Z isomers); 13C NMR (100
MHz, CDCl3) δ 176.0, 175.9, 167.8, 139.6, 139.4, 130.5, 129.9,
51.6, 51.3, 51.2, 51.1, 46.7, 46.5, 37.1, 29.0, 25.3, 25.2 15.8,
14.4, 11.7, 11.6.
General Procedure for the Synthesis of Dimethyl
4-Alkyl-2-oxoglutarates 7a-i. To a solution of 6a-i (20
mmol) in a mixture of CCl4 (45 mL) and CH3CN (45 mL) were
added a solution of NaIO4 (17.1 g, 80 mmol) in water (60 mL)
and RuO2 (270 mg, 2 mmol). The mixture was stirred at room
temperature for 24 h and then filtered on Celite. The organic
layer was isolated, and the aqueous phase was extracted with
CH2Cl2 (2 × 25 mL). The combined organic layers were washed
with brine (50 mL), dried over MgSO4, and concentrated under
reduced pressure. Flash chromatography (eluent, cyclohex-
ane-AcOEt, 8:2, v/v) afforded 7a-i isolated as colorless
liquids.
Dimethyl 2-Ethylidene-4-propylglutarate 6c. Yield 4.10
1
g, 72%; IR (neat film) 1735, 1718, 1647 cm-1; H NMR (400
MHz, CDCl3) δ 6.90 and 6.04 (1H, q, J ) 7.0 Hz, E and Z
isomers), 3.73 and 3.71 (3H, s, Z and E isomers), 3.60 and 3.59
(3H, s, Z and E isomer), 2.60-2.40 (3H, m), 1.94 and 1.77 (3H,
d, J ) 7.0 Hz, Z and E isomers), 1.60-1.20 (4H, m), 0.86 and
0.85 (3H, t, J ) 7.0 Hz, E and Z isomers); 13C NMR (100 MHz,
CDCl3) δ 176.3, 176.2, 167.9, 139.7, 139.5, 130.5, 129.9, 51.7,
51.33, 51.26, 51.2, 45.0, 44.7, 37.5, 34.5, 34.4, 29.3, 20.6, 20.5,
15.9, 14.4, 13.9.
Methyl, Ethyl 4-Methyl-2-oxoglutarate 7a. Yield 3.24 g,
80%; IR (neat film) 1730 cm-1 (broad); 1H NMR (400 MHz,
CDCl3) δ 4.12 (2H, q, J ) 7.0 Hz), 3.86 (3H, s), 3.31 (1H, dd,
J ) 8.0 and 18.0 Hz), 2.99 (1H, m), 2.86 (1H, dd, J ) 5.0 and
Dimethyl 2-Ethylidene-4-butylglutarate 6d. Yield 4.10
1
g, 72%; IR (neat film) 1737, 1725, 1647 cm-1; H NMR (400
MHz, CDCl3) δ 6.92 and 6.05 (1H, q, J ) 7.0 Hz, E and Z
isomers), 3.75 and 3.72 (3H, s, Z and E isomers), 3.62 and 3.61
(3H, s, Z and E isomer), 2.65-2.40 (3H, m), 1.96 and 1.80 (3H,
d, J ) 7.0 Hz, Z and E isomers), 1.65-1.15 (6H, m), 0.86 (3H,
t, J ) 7.0 Hz); 13C NMR (100 MHz, CDCl3) δ 176.0, 175.9,
167.6, 139.7, 139.2, 130.4, 129.8, 51.7, 51.4, 51.1, 51.0, 45.0,
44.7, 37.4, 34.2, 31.8, 31.2, 29.4, 29.3, 22.4, 22.2, 15.6, 14.2,
13.7.
18.0 Hz), 1.23 (3H, d, J ) 8.0 Hz), 1.23 (3H, t, J ) 7.0 Hz); 13
C
NMR (100 MHz, CDCl3) δ 192.1, 174.8, 161.0, 60.8, 52.9, 42.3,
34.8, 16.9, 14.0.
Dimethyl 4-Ethyl-2-oxoglutarate 7b. Yield 2.90 g, 71%;
1
IR (neat film) 1732 cm-1 (broad); H NMR (400 MHz, CDCl3)
δ 3.83 (3H, s), 3.64 (3H, s), 3.26 (1H, dd, J ) 10.0 and 19.5
Hz), 2.86 (1H, dd, J ) 4.5 and 19.5 Hz), 2.85 (1H, m), 1.70-
1.50 (2H, m), 0.89 (3H, t, J ) 7.0 Hz); 13C NMR (100 MHz,
CDCl3) δ 192.3, 174.9, 161.0, 53.0, 51.8, 41.2, 40.2, 24.8, 11.2.
Dimethyl 2-Ethylidene-4-pentylglutarate 6e. Yield 5.89
1
g, 92%; IR (neat film) 1737, 1722, 1647 cm-1; H NMR (400
Dimethyl 4-Propyl-2-oxoglutarate 7c. Yield 2.90 g, 70%;
MHz, CDCl3) δ 6.91 and 6.05 (1H, q, J ) 7.0 Hz, E and Z
isomers), 3.72 and 3.70 (3H, s, Z and E isomers), 3.60 and 3.59
(3H, s, Z and E isomer), 2.65-2.30 (3H, m), 1.95 and 1.78 (3H,
d, J ) 7.0 Hz, Z and E isomers), 1.65-1.15 (8H, m), 0.85 (3H,
t, J ) 7.0 Hz); 13C NMR (100 MHz, CDCl3) δ 176.0, 175.8,
167.5, 139.6, 139.2, 130.4, 129.7, 51.4, 51.04, 50.96, 50.9, 45.0,
44.7, 37.4, 32.2, 32.0, 31.5, 29.2, 26.8, 26.7, 22.3, 15.6, 14.2,
13.7.
1
IR (neat film) 1733 cm-1 (broad); H NMR (400 MHz, CDCl3)
δ 3.85 (3H, s), 3.66 (3H, s), 3.23 (1H, dd, J ) 8.0 and 17.5 Hz),
2.93 (1H, m), 2.88 (1H, dd, J ) 5.0 and 17.5 Hz), 1.65 (1H, m),
1.50 (1H, m), 1.34 (2H, hex., J ) 7.0 Hz), 0.91 (3H, t, J ) 7.0
Hz); 13C NMR (100 MHz, CDCl3) δ 192.2, 175.0, 160.9, 52.9,
51.8, 40.6, 39.5, 33.8, 20.0, 13.7.
Dimethyl 4-Butyl-2-oxoglutarate 7d. Yield 3.65 g, 79%;
1
IR (neat film) 1732 cm-1 (broad); H NMR (400 MHz, CDCl3)
Dimethyl 2-Ethylidene-4-isopropylglutarate 6f. Yield
1
4.28 g, 75%; IR (neat film) 1737, 1724, 1647 cm-1; H NMR
δ 3.87 (3H, s), 3.67 (3H, s), 3.30 (1H, dd, J ) 10.5 and 19.5
Hz), 2.92 (1H, m), 2.90 (1H, dd, J ) 4.5 and 19.5 Hz), 1.67
(1H, m), 1.55 (1H, m), 1.30 (4H, m), 0.89 (3H, t, J ) 7.0 Hz);
13C NMR (100 MHz, CDCl3) δ 192.3, 175.1, 161.0, 53.1, 52.0,
40.7, 39.8, 31.4, 29.1, 22.5, 13.9.
(400 MHz, CDCl3) δ 6.90 and 6.05 (1H, q, J ) 7.0 Hz, E and
Z isomers), 3.74 and 3.71 (3H, s, Z and E isomers), 3.60 and
3.58 (3H, s, Z and E isomers), 2.65-2.30 (3H, m), 1.95 and
1.80 (3H, d, J ) 7.0 Hz, Z and E isomers), 1.85 (1H, m), 1.00
and 0.95 (3H, d, J ) 7.0 Hz, E and Z isomers), 0.94 and 0.90
(3H, d, J ) 7.0 Hz, E and Z isomers); 13C NMR (100 MHz,
CDCl3) δ 175.6, 175.4, 167.8, 139.7, 139.4, 130.7, 130.1, 52.2,
51.7, 51.6, 51.2, 51.1, 34.9, 30.8, 30.6, 26.9, 20.4, 20.3, 20.1,
15.9, 14.3.
Dimethyl 4-Pentyl-2-oxoglutarate 7e. Yield 3.02 g, 62%;
1
IR (neat film) 1731 cm-1 (broad); H NMR (400 MHz, CDCl3)
δ 3.86 (3H, s), 3.68 (3H, s), 3.30 (1H, dd, J ) 10.5 and 20.0
Hz), 2.93 (1H, m), 2.90 (1H, dd, J ) 4.5 and 20.0 Hz), 1.65
(1H, m), 1.55 (1H, m), 1.28 (6H, m), 0.87 (3H, t, J ) 7.0 Hz);
13C NMR (100 MHz, CDCl3) δ 192.4, 175.2, 161.1, 53.1, 52.0,
40.8, 39.9, 31.7, 31.6, 26.6, 22.5, 14.0.
Dimethyl 2-Ethylidene-4-isobutylglutarate 6g. Yield
1
4.98 g, 82%; IR (neat film) 1737, 1723, 1647 cm-1; H NMR

Glutamate Analogues
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7989
Dimethyl 4-Isopropyl-2-oxoglutarate 7f. Yield 3.55 g,
82%; IR (neat film) 1732 cm-1 (broad); 1H NMR (400 MHz,
CDCl3) δ 3.88 (3H, s), 3.68 (3H, s), 3.34 (1H, dd, J ) 11.5 and
19.5 Hz), 2.85 (1H, dd, J ) 3.5 and 19.5 Hz), 2.84 (1H, m),
2.07 (1H, m), 0.95 (3H, d, J ) 7.0 Hz), 0.92 (3H, d, J ) 7.0
Hz); 13C NMR (100 MHz, CDCl3) δ 192.7, 174.4, 161.0, 53.0,
51.8, 46.0, 37.6, 29.8, 20.1, 19.3.
Dimethyl 4-Isobutyl-2-oxoglutarate 7g. Yield 3.36 g,
73%; IR (neat film) 1732 cm-1 (broad); 1H NMR (400 MHz,
CDCl3) δ 3.88 (3H, s), 3.68 (3H, s), 3.26 (1H, dd, J ) 9.0 and
18.0 Hz), 2.99 (1H, m), 2.92 (1H, dd, J ) 4.5 and 18.0 Hz),
1.60 (2H, m), 1.32 (1H, m), 0.94 (3H, d, J ) 6.5 Hz), 0.89 (3H,
d, J ) 6.5 Hz); 13C NMR (100 MHz, CDCl3) δ 192.3, 175.5,
161.0, 53.1, 52.0, 41.2, 41.0, 38.1, 25.8, 22.4, 22.3.
Dimethyl 4-Isopentyl-2-oxoglutarate 7h. Yield 3.32 g,
68%; IR (neat film) 1732 cm-1 (broad); 1H NMR (400 MHz,
CDCl3) δ 3.87 (3H, s), 3.68 (3H, s), 3.30 (1H, dd, J ) 10.5 and
19.5 Hz), 2.90 (2H, m), 1.75-1.50 (3H, m), 1.20 (2H, m), 0.87
(3H, d, J ) 6.5 Hz), 0.86 (3H, d, J ) 6.5 Hz); 13C NMR (100
MHz, CDCl3) δ 192.4, 175.1, 161.0, 53.0, 51.9, 40.7, 40.1, 35.9,
29.6, 27.9, 22.44, 22.40.
Dilithium 4-Isopentyl-2-oxoglutarate 2h. 1H NMR (400
MHz, D2O) δ 3.09 (1H, dd, J ) 8.5 and 18.0 Hz), 2.93 (1H, dd,
J ) 6.0 and 18.0 Hz), 2.69 (1H, m), 1.57 (3H, m), 1.25 (2H, m),
0.93 (6H, d, J ) 7.0 Hz); 13C NMR (100 MHz, D2O) δ 208.1,
186.8, 172.9, 46.1, 45.1, 38.6, 32.8, 30.0, 24.7, 24.5.
Dilithium 4-Benzyl-2-oxoglutarate 2i. 1H NMR (400
MHz, D2O) δ 7.50-7.30 (5H, m), 3.15-2.75 (5H, m); 13C NMR
(100 MHz, D2O) δ 207.5, 185.5, 172.5, 142.6, 131.9, 131.3,
129.2, 48.0, 44.4, 40.9.
General Procedure for the Synthesis of (2S,4R)-4-
Alkylglutamic Acids 1a-i. To a solution of racemic 2a-i
(2.5 mmol) in water (125 mL) were added cysteine sulfinic acid
(2.5 mmol) and acetaldehyde (0.3 mL, 5 mmol). The pH of the
solution was adjusted to 7.6 with 1 M NaOH before the
addition of pig heart or E. coli AAT (20-100 units). The
mixture was stirred slowly at room temperature and monitored
by titration of pyruvate in the following way: 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
rabbit muscle lactate dehydrogenase (1 unit). Pyruvate con-
centration was calculated from 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 50X8 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 was adjusted to 7.0 with 1 M
NaOH before adsorption of the product on a column of Dowex
2X8 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 (0.1-0.5 M). The ninhydrin positive frac-
tions were combined and dried under reduced pressure to
afford 1a-i as white solids.
Dimethyl 4-Benzyl-2-oxoglutarate 7i. Yield 3.17 g, 60%;
1
IR (neat film) 1732 cm-1 (broad); H NMR (400 MHz, CDCl3)
δ 7.35-7.10 (5H, m), 3.84 (3H, s), 3.67 (3H, s), 3.35-3.25 (2H,
m), 3.10 (1H, dd, J ) 6.0 and 13.5 Hz), 2.90-275 (2H, m); 13
C
NMR (100 MHz, CDCl3) δ 192.2, 174.3, 160.9, 137.9, 129.0,
128.7, 126.9, 53.1, 52.1, 42.0, 39.8, 37.5.
General Procedure for the Synthesis of Dilithium
4-Alkyl-2-oxoglutarate 2a-i. To a solution of 7a-i (10
mmol) in THF or MeOH (50 mL) was added dropwise a 0.4 M
solution of LiOH (52.5 mL, 21 mmol). The mixture was stirred
at room temperature for 3-4 h until completion. After
evaporation of the organic solvent, the pH of the aqueous
solution was adjusted to 7.6 by addition of Dowex 50X8 resin
(H+ form). The resin was removed by filtration before evapora-
tion of the water. 2a-i were isolated in quantitative yields as
white solids.
Dilithium 4-Methyl-2-oxoglutarate 2a. 1H NMR (400
MHz, D2O) δ 2.99 (1H, dd, J ) 8.0 and 17.5 Hz), 2.70 (1H, dd,
J ) 6.0 and 17.5 Hz), 2.62 (1H, m), 1.05 (3H, d, J ) 7.0 Hz);
13C NMR (100 MHz, D2O) δ 205.1, 184.8, 170.2, 43.8, 37.5, 17.4.
Dilithium 4-Ethyl-2-oxoglutarate 2b. 1H NMR (400 MHz,
D2O) δ 3.07 (1H, dd, J ) 8.5 and 18.0 Hz), 2.88 (1H, dd, J )
5.5 and 18.5 Hz), 2.63 (1H, m), 1.56 (2H, quint, J ) 7.0 Hz),
0.93 (3H, t, J ) 7.0 Hz); 13C NMR (100 MHz, D2O) δ 205.3,
183.8, 170.2, 44.6, 42.0, 25.3, 11.1.
Dilithium 4-Propyl-2-oxoglutarate 2c. 1H NMR (400
MHz, D2O) δ 3.06 (1H, dd, J ) 8.5 and 18.5 Hz), 2.90 (1H, dd,
J ) 5.5 and 18.0 Hz), 2.70 (1H, m), 1.50 (2H, m), 1.35 (2H, m),
0.94 (3H, t, J ) 7.0 Hz); 13C NMR (100 MHz, D2O) δ 208.2,
186.8, 172.9, 45.8, 45.1, 37.3, 22.8, 16.1.
Dilithium 4-Butyl-2-oxoglutarate 2d. 1H NMR (400
MHz, D2O) δ 3.05 (1H, dd, J ) 8.5 and 18.5 Hz), 2.89 (1H, dd,
J ) 5.5 and 18.5 Hz), 2.67 (1H, m), 1.52 (2H, m), 1.32 (4H, m),
0.90 (3H, t, J ) 7.0 Hz); 13C NMR (100 MHz, D2O) δ 207.9,
186.8, 172.8, 45.9, 45.0, 34.5, 31.7, 24.7, 15.9.
Dilithium 4-Pentyl-2-oxoglutarate 2e. 1H NMR (400
MHz, D2O) δ 3.09 (1H, dd, J ) 8.5 and 18.0 Hz), 2.93 (1H, dd,
J ) 6.0 and 18.0 Hz), 2.72 (1H, m), 1.56 (2H, m), 1.36 (6H, m),
0.94 (3H, t, J ) 7.0 Hz); 13C NMR (100 MHz, D2O) δ 208.1,
186.8, 173.0, 46.0, 45.1, 34.9, 33.8, 29.0, 24.5, 16.1.
(2S,4R)-4-Methylglutamic Acid 1a. Yield 165 mg, 41%;
mp 178 °C (lit.49 mp 177 °C); [R]25 +24.0° (c 1.3, 6 N HCl)
D
(lit.50 [R]25D +23.2° (c 0.7, 6 N HCl)); 1H NMR (400 MHz, D2O)
δ 3.83 (1H, dd, J ) 4.5 and 8.5 Hz), 2.58 (1H, m, J ) 5.0, 7.0,
8.5), 2.23 (1H, ddd, J ) 5.0, 8.5, and 14.0 Hz), 1.96 (1H, ddd,
J ) 5.0, 8.5, and 13.5 Hz), 1.25 (3H, d, J ) 7.0 Hz); 13C NMR
(100 MHz, D2O) δ 185.3, 178.5, 53.9, 39.5, 36.9, 17.8. Anal.
(C6H11NO4) C, H, N.
(2S,4R)-4-Ethylglutamic Acid 1b. Yield 174 mg, 40%; mp
175 °C (lit.20 mp 179 °C); [R]25 +31.0° (c 1.3, 6 N HCl); 1H
D
NMR (400 MHz, D2O) δ 3.80 (1H, dd, J ) 4.0 and 9.0 Hz),
2.55 (1H, m, J ) 4.0, 7.0, and 10.0 Hz), 2.25 (1H, ddd, J ) 4.0,
10.0, and 14.0 Hz), 1.95 (1H, ddd, J ) 4.0, 9.0, and 14.0 Hz),
1.65 (2H, quint, J ) 7.0 Hz), 0.90 (3H, t, J ) 7.0 Hz); 13C NMR
(100 MHz, D2O) δ 186.7, 177.8, 56.3, 46.5, 35.0, 28.4, 13.6.
Anal. (C7H13NO4) C, H, N.
(2S,4R)-4-Propylglutamic Acid 1c. Yield 198 mg, 42%;
mp 164 °C; [R]25D +34.5° (c 1.0, 6 N HCl); 1H NMR (400 MHz,
D2O) δ 3.79 (1H, dd, J ) 4.0 and 9.0 Hz), 2.52 (1H, m), 2.24
(1H, ddd, J ) 4.0, 10.0, and 14.5 Hz), 1.98 (1H, ddd, J ) 4.5,
9.0, and 14.0 Hz), 1.70-1.50 (2H, m), 1.39 (2H, hex., J ) 7.0
Hz), 0.98 (3H, t, J ) 7.0 Hz); 13C NMR (100 MHz, D2O) δ 186.3,
177.1, 55.8, 47.3, 36.9, 35.6, 22.5, 15.7. Anal. (C8H15NO4) C,
H, N.
(2S,4R)-4-Butylglutamic Acid 1d. Yield 163 mg, 32%; mp
162 °C; [R]25D +29.7° (c 1.2, 6 N HCl); 1H NMR (400 MHz, D2O)
δ 3.79 (1H, dd, J ) 4.5 and 9.0 Hz), 2.55 (1H, m), 2.26 (1H,
ddd, J ) 4.5, 9.5, and 14.5 Hz), 1.99 (1H, ddd, J ) 4.5, 9.0,
and 14.0 Hz), 1.75-1.55 (2H, m), 1.45-1.30 (4H, m), 0.96 (3H,
t, J ) 7.0 Hz); 13C NMR (100 MHz, D2O) δ 185.0, 177.2, 56.1,
46.6, 35.6, 34.7, 31.5, 24.7, 15.9; HRMS (EI, positive ion) m/z
204.1235 ([MH]+, C9H18NO4 requires 204.1236). Anal. (C9H17-
NO4‚0.33H2O) C, H, N.
Dilithium 4-Isopropyl-2-oxoglutarate 2f. 1H NMR (400
MHz, D2O) δ 3.07 (1H, dd, J ) 9.5 and 18.5 Hz), 2.97 (1H, dd,
J ) 4.5 and 18.5 Hz), 2.52 (1H, m), 1.91 (1H, oct, J ) 7.0 Hz),
0.99 (3H, d, J ) 7.0 Hz), 0.95 (3H, d, J ) 7.0 Hz); 13C NMR
(100 MHz, D2O) δ 208.3, 186.1, 172.8, 52.5, 42.2, 32.8, 22.9,
21.8.
1
Dilithium 4-Isobutyl-2-oxoglutarate 2g. H NMR (400
(2S,4R)-4-Pentylglutamic Acid 1e. Yield 195 mg, 36%;
mp 157 °C (lit.51 mp 160-1 °C); [R]25D +18.6° (c 1.1, 6 N HCl);
1H NMR (400 MHz, D2O) δ 3.78 (1H, dd, J ) 4.0 and 9.0 Hz),
2.51 (1H, m), 2.25 (1H, ddd, J ) 4.5, 9.5, and 14.5 Hz), 1.98
(1H, ddd, J ) 4.5, 9.0, and 14.0 Hz), 1.70-1.50 (2H, m), 1.45-
1.30 (6H, m), 0.94 (3H, t, J ) 7.0 Hz); 13C NMR (100 MHz,
MHz, D2O) δ 3.04 (1H, dd, J ) 8.5 and 18.0 Hz), 2.90 (1H, dd,
J ) 5.5 and 18.0 Hz), 2.79 (1H, m), 1.65-1.45 (2H, m), 1.31
(1H, m), 0.96 (3H, d, J ) 7.0 Hz), 0.92 (3H, d, J ) 7.0 Hz); 13
C
NMR (100 MHz, D2O) δ 207.8, 187.1, 172.8, 45.4, 44.4, 44.3,
28.3, 24.8, 24.4.

7990 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Alaux et al.
D2O) δ 183.3, 176.8, 56.0, 45.3, 35.3, 34.7, 33.6, 28.6, 24.5, 16.0.
HRMS (FAB, positive ion) m/z 262.1034 ([M - H + 2Na]+,
C10H18NO4Na2 requires 262.1031). Anal. (C10H19NO4‚0.33H2O)
C, H, N.
was added to each well (in the characterization of nonsubstrate
inhibitors, the inhibitors were added to this buffer). An amount
of 50 µL of Krebs buffer supplemented with FMP assay dye
was then added to each well, and the plate was incubated at
37 °C for 30 min. The plate was assayed at room temperature
in a NOVOstar plate reader measuring emission at 560 nm
caused by excitation at 530 nm before and up to 1 min after
addition of 25 µL of substrate solution. The experiments were
performed in duplicate at least three times for each compound.
For the characterization of nonsubstrate inhibitors 30 µM Glu
was used as substrate concentration. Concentration-response
curves were constructed on the basis of the maximal responses
obtained for seven different concentrations of each compound.
IC50 values were converted to Ki values by the use of the
Cheng-Prusoff equation.53
Molecular Modeling Study of 4-Substituted Glu Ana-
logues 1a-1l and Other Selected Ligands. The modeling
study was performed using the software package MOE (Mo-
lecular Operating Environment, version 2004.03, Chemical
Computing Group, 2004) using the built-in mmff94x force field
and the GB/SA continuum solvent model. Each compound was
submitted to a stochastic conformational search and with
respect to its global minimum returned (∆G in kcal/mol);
conformations above +7 kcal/mol were discarded. For Glu
analogues 1a-i,l, the γ-carboxylate group was protonated prior
to execution of the conformational search because this gave a
larger and thus more reliable number of output conformations.
Superimpositions of ligands were carried out using the built-
in function in MOE, by fitting the ammonium group and the
two carboxylate groups.
(2S,4S)-4-Isopropylglutamic Acid 1f. Yield 195 mg, 41%;
mp 178 °C (lit.52 mp 175-177 °C); [R]25 +31.0° (c 1.0, 6 N
D
1
HCl); H NMR (400 MHz, D2O) δ 3.81 (1H, dd, J ) 4.5 and
9.0 Hz), 2.48 (1H, ddd, J ) 3.0, 6.0, and 9.0 Hz), 2.35 (1H,
ddd, J ) 5.0, 11.0, and 15.0 Hz), 2.14-2.02 (2H, m), 1.06 (3H,
d, J ) 7.0 Hz), 1.04 (3H, d, J ) 7.0 Hz); 13C NMR (100 MHz,
D2O) δ 185.5, 177.4, 56.4, 54.6, 32.9, 32.8, 23.1, 21.7. HRMS
(FAB, positive ion) m/z 190.1080 ([M + H]+, C8H16NO4 requires
190.1079). Anal. (C8H15NO4) C, H, N.
(2S,4R)-4-Isobutylglutamic Acid 1g. Yield 213 mg, 42%;
mp 165 °C; [R]25D +26.0° (c 1.1, 6 N HCl); 1H NMR (400 MHz,
D2O) δ 3.75 (1H, dd, J ) 5.5 and 8.5 Hz), 2.66 (1H, m), 2.22
(1H, ddd, J ) 5.0, 9.5, and 15.0 Hz), 1.92 (1H, ddd, J ) 4.5,
8.5, and 15.0 Hz), 1.58 (2H, m), 1.38 (1H, m), 0.90 (3H, d, J )
7.0 Hz), 0.88 (3H, d, J ) 7.0 Hz); 13C NMR (100 MHz, D2O) δ
186.4, 177.3, 56.2, 45.8, 44.4, 36.3, 29.4, 25.0, 24.3. HRMS
(FAB, positive ion) m/z 204.1233 ([M + H]+, C9H18NO4 requires
204.1236). Anal. (C9H17NO4‚0.5H2O) C, H, N.
(2S,4R)-4-Isopentylglutamic Acid 1h. Yield 185 mg, 34%;
mp 150 °C; [R]25D +26.4° (c 1.1, 6 N HCl); 1H NMR (400 MHz,
D2O) δ 3.80 (1H, dd, J ) 4.0 and 9.0 Hz), 2.58 (1H, m), 2.30
(1H, ddd, J ) 5.5, 9.0, and 15.0 Hz), 1.92 (1H, ddd, J ) 4.0,
8.0, and 15.0 Hz), 1.66 (3H, m), 1.29 (2H, m), 0.98 (6H, d, J )
7.0 Hz); 13C NMR (100 MHz, D2O) δ 184.1, 177.0, 56.1, 46.2,
38.3, 35.5, 32.8, 30.0, 24.6, 24.4. HRMS (FAB, positive ion)
m/z 218.1399 ([M + H]+, C10H20NO4 requires 218.1392). Anal.
(C10H19NO4‚0.5H2O) C, H, N.
(2S,4R)-4-Benzylglutamic Acid 1i. Yield 250 mg, 42%; mp
161 °C; [R]25D +36.7° (c 1.0, 6 N HCl); 1H NMR (400 MHz, D2O)
δ 7.45 (5H, m), 3.82 (1H, dd, J ) 4.5 and 9.0 Hz), 3.03 (1H,
dd, J ) 8.5 and 13.5 Hz), 2.94 (1H, dd, J ) 6.5 and 13.5 Hz),
2.82 (1H, m), 2.29 (1H, ddd, J ) 4.5, 10.0, and 14.5 Hz), 2.01
(1H, ddd, J ) 4.5, 9.0, and 15.0 Hz); 13C NMR (100 MHz, D2O)
δ 184.3, 177.1, 142.2, 131.8, 131.3, 129.3, 56.2, 49.2, 41.2, 35.4;
HRMS (EI, positive ion) m/z 238.1083 ([M + H]+, C12H16NO4
requires 238.1079). Anal. (C12H15NO4‚0.33H2O) C, H, N.
Acknowledgment. We thank The Danish Medical
Research Council, the French National Center for
Scientific Research (CNRS), the Lundbeck Foundation,
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 strains.
Ethyl (2S,4R)-4-Pentylpyroglutamate 8. A solution of 1e
(45 mg, 0.2 mmol) in water (15 mL) was heated under reflux
for 48 h until complete disappearance of 1e. After evaporation
of water under reduced pressure, the solid residue was diluted
with absolute EtOH (10 mL). SOCl2 (100 µL, 1.4 mmol) was
added. The mixture was stirred at room temperature for 20 h
before concentration under reduced pressure. The residue was
chromatographed on silica gel to give 8 (29 mg, 64%) isolated
Supporting Information Available: Combustion analy-
sis data of compounds 1a-e,g-i and 1H and 13C NMR spectra
of compound 1f. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
as a colorless oil. [R]25 + 9.3° (c 1.5, MeOH); IR (neat film)
D
3235, 1742, 1708 cm-1; 1H NMR (400 MHz, C6D6) δ 7.85 (1H,
s), 3.88 (2H, q, J ) 7.0 Hz), 3.70 (1H, dd, J ) 2.5 and 9.0 Hz),
2.30 (1H, m), 2.05 (1H, ddd, J ) 2.5, 9.0, and 13.0 Hz), 1.82
(1H, m), 1.55 (1H, td, J ) 9.0 and 13.0 Hz), 1.12 (7H, m), 0.92
(3H, t, J ) 7.0 Hz), 0.81 (3H, t, J ) 7.0 Hz); 13C NMR (100
MHz, C6D6) δ 180.2, 172.5, 61.1, 54.1, 39.9, 32.0, 31.6, 31.2,
27.1, 22.8, 14.2, 14.0; HRMS (EI, positive ion) m/z 228.1611
([M + H]+, C12H22NO3 requires 228.1600).
Pharmacological Characterization of 4-Substituted
Glu Analogues 1a-k at Human EAATs. The pharmacologi-
cal properties of 4-substituted Glu analogues at human
EAAT1, EAAT2, and EAAT3 were determined in the FLIPR
membrane potential (FMP) assay. The construction of human
embryonic kidney 293 (HEK293) cell lines stably expressing
human EAAT1, EAAT2, and EAAT3 has been reported previ-
ously, and the pharmacological characterization of the 4-sub-
stituted Glu analogues was performed essentially as described
here.36 Briefly, cells were split into poly-D-lysine-coated black
walled clear-bottom 96-well plates in Dulbecco’s modified Eagle
medium supplemented with penicillin (100 U/mL), strepto-
mycin (100 µg/mL), 10% dialyzed fetal bovine serum, and 1
mg/mL G-418. After 16-24 h, the medium was aspirated, and
the cells were washed with 100 µL of Krebs buffer [140 mM
NaCl/4.7 mM KCl/2.5 mM CaCl2/1.2 mM MgCl2/11 mM
HEPES/10 mM D-glucose, pH 7.4]. Then 50 µL of Krebs buffer
(1) Bra¨uner-Osborne, H.; Egebjerg, J.; Nielsen, E. Ø.; Madsen, U.;
Krogsgaard-Larsen, P. Ligands for glutamate receptors: Design
and therapeutic prospects. J. Med. Chem. 2000, 43 (14), 2609-
2645.
(2) Dingledine, R.; Borges, K.; Bowie, D.; Traynelis, S. F. The
glutamate receptor ion channels. Pharmacol. Rev. 1999, 51 (1),
7-61.
(3) Campiani, G.; Fattorusso, C.; De Angelis, M.; Catalanotti, B.;
Butini, S.; Fattorusso, R.; Fiorini, I.; Nacci, V.; Novellino, E.
Neuronal high-affinity sodium-dependent glutamate transport-
ers (EAATs): Targets for the development of novel therapeutics
against neurodegenerative diseases. Curr. Pharm. Des. 2003, 9
(8), 599-625.
(4) Seal, R. P.; Amara, S. G. Excitatory amino acid transporters: A
family in flux. Annu. Rev. Pharmacol. Toxicol. 1999, 39, 431-
456.
(5) Mennini, T.; Fumagalli, E.; Gobbi, M.; Fattorusso, C.; Catalan-
otti, B.; Campiani, G. Substrate inhibitors and blockers of
excitatory amino acid transporters in the treatment of neuro-
degeneration: critical considerations. Eur. J. Pharmacol. 2003,
479 (1-3), 291-296.
(6) Shigeri, Y.; Seal, R. P.; Shimamoto, K. Molecular pharmacology
of glutamate transporters, EAATs and VGLUTs. Brain Res. Rev.
2004, 45 (3), 250-265.
(7) O’Shea, R. D.; Fodera, M. V.; Aprico, K.; Dehnes, Y.; Danbolt,
N. C.; Crawford, D.; Beart, P. M. Evaluation of drugs acting at
glutamate transporters in organotypic hippocampal cultures:
New evidence on substrates and blockers in excitotoxicity.
Neurochem. Res. 2002, 27 (1-2), 5-13.

Glutamate Analogues
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7991
(8) Vandenberg, R. J.; Mitrovic, A. D.; Chebib, M.; Balcar, V. J.;
Johnston, G. A. R. Contrasting modes of action of methyl-
glutamate derivatives on the excitatory amino acid transporters,
EAAT1 and EAAT2. Mol. Pharmacol. 1997, 51 (5), 809-815.
(9) Stewart, J. D. Dehydrogenases and transaminases in asym-
metric synthesis. Curr. Opin. Chem. Biol. 2001, 5 (2), 120-129.
(10) Bartsch, K.; Schneider, R.; Schulz, A. Stereospecific production
of the herbicide phosphinothricin (glufosinate): Purification of
aspartate transaminase from Bacillus stearothermophilus, clon-
ing of the corresponding gene, aspC, and application in a coupled
transaminase process. Appl. Environ. Microbiol. 1996, 62 (10),
3794-3799.
(11) Meiwes, J.; Schudok, M.; Kretzschmar, G. Asymmetric synthesis
of L-thienylalanines. Tetrahedron: Asymmetry 1997, 8 (4), 527-
536.
(12) Hanzawa, S.; Oe, S.; Tokuhisa, K.; Kawano, K.; Kobayashi, T.;
Kudo, T.; Kakidani, H. Chemo-enzymatic synthesis of 3-(2-
naphthyl)-L-alanine by an aminotransferase from the extreme
thermophile, Thermococcus profundus. Biotechnol. Lett. 2001,
23 (8), 589-591.
(13) Cameron, M.; Cohen, D.; Cottrell, I. F.; Kennedy, D. J.; Roberge,
C.; Chartrain, M. The highly stereospecific enzyme catalysed
transamination of 4-fluorophenylglyoxylic acid to 4-(S)-fluo-
rophenylglycine. J. Mol. Catal. B: Enzym. 2001, 14 (1-3), 1-5.
(14) Ager, D. J.; Li, T.; Pantaleone, D. P.; Senkpeil, R. F.; Taylor, P.
P.; Fotheringham, I. G. Novel biosynthetic routes to non-
proteinogenic amino acids as chiral pharmaceutical intermedi-
ates. J. Mol. Catal. B:Enzym. 2001, 11 (4-6), 199-205.
(15) Shin, J. S.; Kim, B. G. Comparison of the omega-transaminases
from different microorganisms and application to production of
chiral amines. Biosci. Biotechno. Biochem. 2001, 65 (8), 1782-
1788.
(16) Fotheringham, I. G.; Grinter, N.; Pantaleone, D. P.; Senkpeil,
R. F.; Taylor, P. P. Engineering of a novel biochemical pathway
for the biosynthesis of L-2-aminobutyric acid in Escherichia coli
K12. Bioorg. Med. Chem. 1999, 7 (10), 2209-2213.
(17) Ager, D. J.; Fotheringham, I. G.; Li, T.; Pantaleone, D. P.;
Senkpeil, R. F. The large scale synthesis of “unnatural” amino
acids. Enantiomer 2000, 5 (3-4), 235-243.
(18) Taylor, P. P.; Pantaleone, D. P.; Senkpeil, R. F.; Fotheringham,
I. G. Novel biosynthetic approaches to the production of un-
natural amino acids using transaminases. Trends Biotechnol.
1998, 16 (10), 412-418.
(19) Bessis, A. S.; Bolte, J.; Pin, J. P.; Acher, F. New probes of the
agonist binding site of metabotropic glutamate receptors. Bioorg.
Med. Chem. Lett. 2001, 11 (12), 1569-1572.
(20) Echalier, F.; Constant, O.; Bolte, J. Enzymatic-synthesis of
4-methyl-L-glutamic and 4-ethyl-L-glutamic acids. J. Org. Chem.
1993, 58 (10), 2747-2750.
(21) Helaine, V.; Rossi, J.; Bolte, J. A new access to alkyl-alpha-
ketoglutaric acids, precursors of glutamic acid analogues by
enzymatic transamination. Application to the synthesis of
(2S,4R)-4-propyl-glutamic acid. Tetrahedron Lett. 1999, 40 (36),
6577-6580.
(22) Helaine, V.; Rossi, J.; Gefflaut, T.; Alaux, S.; Bolte, J. Synthesis
of 4,4-disubstituted L-glutamic acids by enzymatic transamina-
tion. Adv. Synth. Catal. 2001, 343 (6-7), 692-697.
(23) Helaine, V.; Bolte, J. Chemoenzymatic synthesis of (4S)- and
(4R)-4-methyl-2-oxoglutaric acids, precursors of glutamic acid
analogues. Eur. J. Org. Chem. 1999, 3403-3406.
(24) 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 (3), 741.
(25) Mehta, G.; Murthy, A. N. A General stereocontrolled approach
to the 5-8 fused ring-system. Application to the total synthesis
of marine natural product (()-precapnelladiene. J. Org. Chem.
1987, 52 (13), 2875-2881.
(31) Jager, J.; Moser, M.; Sauder, U.; Jansonius, J. N. Crystal-
structures of Escherichia-coli aspartate-aminotransferase in 2
conformations. Comparison of an unliganded open and 2 ligand-
ed closed forms. J. Mol. Biol. 1994, 239 (2), 285-305.
(32) McPhalen, C. A.; Vincent, M. G.; Picot, D.; Jansonius, J. N.; Lesk,
A. M.; Chothia, C. Domain closure in mitochondrial aspartate-
aminotransferase. J. Mol. Biol. 1992, 227 (1), 197-213.
(33) Malashkevich, V. N.; Jager, J.; Ziak, M.; Sauder, U.; Gehring,
H.; Christen, P.; Jansonius, J. N. Structural basis for the
catalytic activity of aspartate-aminotransferase K258h lacking
the pyridoxal 5′-phosphate-binding lysine residue. Biochemistry
1995, 34 (2), 405-414.
(34) McPhalen, C. A.; Vincent, M. G.; Jansonius, J. N. X-ray structure
refinement and comparison of 3 forms of mitochondrial aspar-
tate-aminotransferase. J. Mol. Biol.1992, 225 (2), 495-517.
(35) Malashkevich, V. N.; Toney, M. D.; Jansonius, J. N. Crystal-
structures of true enzymatic-reaction intermediates. Aspartate
and glutamate ketimines in aspartate-aminotransferase. Bio-
chemistry 1993, 32 (49), 13451-13462.
(36) Jensen, A. A.; Bra¨uner-Osborne, H. Pharmacological character-
ization of human excitatory amino acid transporters EAAT1,
EAAT2 and EAAT3 in a fluorescence-based membrane potential
assay. Biochem. Pharmacol. 2004, 67 (11), 2115-2127.
(37) Campiani, G.; De Angelis, M.; Armaroli, S.; Fattorusso, C.;
Catalanotti, B.; Ramunno, A.; Nacci, V.; Novellino, E.; Grewer,
C.; Ionescu, D.; Rauen, T.; Griffiths, R.; Sinclair, C.; Fumagalli,
E.; Mennini, T. A rational approach to the design of selective
substrates and potent nontransportable inhibitors of the exci-
tatory amino acid transporter EAAC1 (EAAT3). New glutamate
and aspartate analogues as potential neuroprotective agents. J.
Med. Chem. 2001, 44 (16), 2507-2510.
(38) Koch, H. P.; Kavanaugh, M. P.; Esslinger, C. S.; Zerangue, N.;
Humphrey, J. M.; Amara, S. G.; Chamberlin, A. R.; Bridges, R.
J. Differentiation of substrate and nonsubstrate inhibitors of the
high-affinity, sodium-dependent glutamate transporters. Mol.
Pharmacol. 1999, 56 (6), 1095-1104.
(39) Hogner, A.; Kastrup, J. S.; Jin, R.; Liljefors, T.; Mayer, M. L.;
Egebjerg, J.; Larsen, I. K.; Gouaux, E. Structural basis for AMPA
receptor activation and ligand selectivity: Crystal structures of
five agonist complexes with the GluR2 ligand-binding core. J.
Mol. Biol.2002, 322 (1), 93-109.
(40) Bunch, L.; Liljefors, T.; Greenwood, J. R.; Frydenvang, K.;
Bra¨uner-Osborne, H.; Krogsgaard-Larsen, P.; Madsen, U. Ra-
tional design, synthesis, and pharmacological evaluation of
2-azanorbornane-3-exo-5-endo-dicarboxylic acid: A novel con-
formationally restricted glutamic acid analogue. J. Org. Chem.
2002, 68 (4), 1489-1495.
(41) Kunishima, N.; Shimada, Y.; Tsuji, Y.; Sato, T.; Yamamoto, M.;
Kumasaka, T.; Nakanishi, S.; Jingami, H.; Morikawa, K. Struc-
tural basis of glutamate recognition by a dimeric metabotropic
glutamate receptor. Nature 2000, 407 (6807), 971-977.
(42) Shimamoto, K.; Sakai, R.; Takaoka, K.; Yumoto, N.; Nakajima,
T.; Amara, S. G.; Shigeri, Y. Characterization of novel L-threo-
beta-benzyloxyaspartate derivatives, potent blockers of the
glutamate transporters. Mol. Pharmacol. 2004, 65 (4), 1008-
1015.
(43) Shi, G. Q.; Cao, Z. Y.; Zhang, X. B. Ethyl 3-fluoro-3-(tributyl-
stannyl)-2-methoxyacrylate. Preparation and palladium/copper-
cocatalyzed cross-coupling reactions as a novel route to â-fluoro-
R-keto acid-derivatives. J. Org. Chem. 1995, 60 (20), 6608-6611.
(44) Bellamy, A. J.; Kerr, J. B.; McGregor, C. J.; Mackirdy, I. S.
Cyanomethylation versus reductive saturation and hydrodimer-
ization during the electroreduction of alpha-beta-unsaturated
nitriles in acetonitrile. J. Chem. Soc., Perkin Trans. 2 1982, 161-
167.
(45) Emiliozzi, R.; Pichat, L. A simple method for the preparation of
cysteinesulfinic acid. Bull. Soc. Chim. Fr. 1959, 1887-1888.
(46) Yagi, T.; Kagamiyama, H.; Motosugi, K.; Nozaki, M.; Soda, K.
Crystallization and properties of aspartate-aminotransferase
from Escherichia-coli-B. FEBS Lett. 1979, 100 (1), 81-84.
(47) 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 characterization of aspartate-aminotransferase from Es-
cherichia-coli. J. Biochem. 1987, 101 (3), 813-816.
(48) Drewes, S. E.; Emslie, N. D.; Karodia, N.; Loizou, G. Synthesis
of useful intermediates from the Claisen-rearrangement of
alpha-hydroxy alkyl acrylate precursors. Synth. Commun. 1990,
20 (10), 1437-1443.
(49) Belokon, Y.; Bulychev, A. G.; Ryzhov, M. G.; Vitt, S. V.; Batsanov,
A. S.; Struchkov, Y.; Bakhmutov, V. I.; Belikov, V. M. Synthesis
of enantio- and diastereo-isomerically pure b- and g-substituted
glutamic acids via glycine condensation with activated olefins.
J. Chem. Soc., Perkin Trans. 1 1986, 1865-1872.
(50) Righini-Tapie, A.; Azerad, R. An enzymatic synthesis of 4-meth-
yl-L-glutamic acid diastereoisomers. Journal of Applied Bio-
chemistry 1984, 6 (5-6), 361-366.
(26) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B. A
Greatly improved procedure for ruthenium tetraoxide catalyzed
oxidations of organic-compounds. J. Org. Chem. 1981, 46 (19),
3936-3938.
(27) Basavaiah, D.; Rao, P. D.; Hyma, R. S. The Baylis-Hillman
reaction: A novel carbon-carbon bond forming reaction. Tetra-
hedron 1996, 52 (24), 8001-8062.
(28) Drewes, S. E.; Roos, G. H. P. Synthetic potential of the tertiary-
amine-catalysed reaction of activated vinyl carbanions with
aldehydes. Tetrahedron 1988, 44 (15), 4653-4670.
(29) Casy, G.; Patterson, J. W.; Taylor, R. J. K. Methyl 7-hydroxyhept-
5-ynoate (5-heptynoic acid, 7-hydroxy acid, methyl-ester). Org.
Synth. 1989, 67, 193-201.
(30) Onuffer, J. J.; Ton, B. T.; Klement, I.; Kirsch, J. F. The use of
natural and unnatural amino-acid substrates to define the
substrate-specificity differences of Escherichia-coli aspartate and
tyrosine aminotransferases. Protein Sci. 1995, 4 (9), 1743-1749.

7992 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Alaux et al.
(51) Pedregal, C.; Collado, I.; Escribano, A.; Ezquerra, J.; Dominguez,
C.; Mateo, A. I.; Rubio, A.; Baker, S. R.; Goldsworthy, J.; Kamboj,
R. K.; Ballyk, B. A.; Hoo, K.; Bleakman, D. 4-Alkyl- and
4-cinnamylglutamic acid analogues are potent GluR5 kainate
receptor agonists. J. Med. Chem. 2000, 43 (10), 1958-1968.
(52) Ezquerra, J.; Pedregal, C.; Yruretagoyena, B.; Rubio, A.; Carreno,
M. C.; Escribano, A.; Ruano, J. L. G. Synthesis of enantiomeri-
cally pure 4-substituted glutamic acids and prolines. General
aldol reaction of pyroglutamate lactam lithium enolate mediated
by Et2O‚BF3. J. Org. Chem. 1995, 60 (9), 2925-2930.
(53) Cheng, Y. C.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes
50% inhibition (I50) of an enzymatic reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
JM050597Z