Synthesis of 4,4-Disubstituted L-Glutamic Acids by Enzymatic Transamination

Article abstract of DOI:10.1002/1615-4169(200108)343:6/7<692::AID-ADSC692>3.0.CO;2-6

The syntheses of optically pure 4,4-dimethyl-L-glutamic acid as well as (2S,4R)- and (2S,4S)-4-hydroxy-4-methylglutamic acids have been achieved by transamination of the corresponding 2-oxo-4,4-dimethyl- and rac-2-oxo-4-hydroxy-4-methylglutaric acids using glutamic oxalacetic transaminase (GOT).

Full text of DOI:10.1002/1615-4169(200108)343:6/7<692::AID-ADSC692>3.0.CO;2-6

FULL PAPER
Synthesis of 4,4-Disubstituted l-Glutamic Acids by
Enzymatic Transamination
Virgil HeÂlaine, JoeÈl Rossi, Thierry Gefflaut, SeÂbastien Alaux, Jean Bolte*
Universite Blaise Pascal, Laboratoire SEESIB (UMR CNRS 6504) DeÂpartement de Chimie, 63177 AubieÁre Cedex, France
Fax: (+33) 473-407-717, e-mail: jbolte@chisg1.univ-bpclermont.fr
Received May 14, 2001; Accepted July 7, 2001
Abstract: The syntheses of optically pure 4,4-di- Keywords: amino acids; biotransformations; en-
methyl-l-glutamic acid as well as (2S,4R)- and zyme catalysis; glutamate analogues; transamina-
(2S,4S)-4-hydroxy-4-methylglutamic acids have tion; transferases
been achieved by transamination of the correspond-
ing 2-oxo-4,4-dimethyl- and rac-2-oxo-4-hydroxy-4-
methylglutaric acids using glutamic oxalacetic
transaminase (GOT).
Moreover, since Glu or pyroglutamic acid are impor-
tant chiral synthons,[4] analogues and specially analo-
gues bearing an additional asymmetric center are of
real interest.
Introduction
Both synthetic and non-proteinogenic natural amino
acids have been the subject of numerous studies for
their enzyme inhibitory and antimetabolite proper-
ties. They have also been often incorporated into syn-
thetic peptides.[1] Glutamic acid (Glu) analogues are
of special interest because of their interaction with
glutamate receptors in the central nervous system.[2]
Even small changes in the structure of Glu can
strongly influence the binding selectivity towards
Glu receptors. For example, (4S)-4-methyl-l-gluta-
mic acid interacts specifically with the metabotropic
receptor Glu R1 as an agonist, whereas the (4R)-iso-
mer is specific for the KA ionotropic receptor.[3]
For these reasons, numerous studies have been
dedicated to the synthesis of natural and non-natural
substituted glutamic acids in the last few years.[5]
However, very few syntheses of 4,4-disubstituted glu-
tamic acids have been published.[6]
Our interest for enzymes as tools in organic syn-
thesis led us to consider the potential of aminotrans-
ferases for Glu analogues syntheses. Aminotrans-
ferases catalyze the reversible stereospecific transfer
of the amino group from a donor amino acid to an ac-
ceptor keto acid (therefore providing a new amino
acid and a new keto acid). Since Glu is a substrate of
Scheme 1. Glutamic oxaloacetic transaminase (GOT)-catalyzed syntheses of glutamic acid analogues.
692
Ó WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2001
1615-4150/01/34306±7-692±697 $ 17.50-.50/0
Adv. Synth. Catal. 2001, 343, No. 6±7

FULL PAPER
most transaminases, these enzymes appear appropri- glutaric acid via the formation of the a-diazo ester,
ate for the synthesis of Glu analogues provided that a as suggested by the work of Taber et al.[9] The syn-
strategy can be defined to shift the equilibrium by re- thesis was achieved according to Scheme 2.
moving the produced a-keto acid.[7]
2,2-Dimethylglutaric acid was esterified in metha-
We have already published the synthesis of 4-alkyl- nol in the presence of thionyl chloride to give 5 in
and 4-hydroxyglutamic acids catalyzed by glutamic 95% yield. Claisen reaction between 5 and methyl
oxalacetic transaminase (GOT).[8] In these studies, benzoate in the presence of NaH gave 6 (65% yield)
cysteinesulfinic acid (CSA) was used as the amino which, in the presence of DBU, reacted with p-nitro-
acid donor for the transamination. This aspartic acid benzenesulfonyl azide[10] providing the a-diazo ester
analogue is a good substrate for GOT and the keto 7 with a nearly quantitative yield. The diazoester was
acid produced, spontaneously decomposes into pyru- oxidized by m-chloroperbenzoic acid,[11] leading to
vic acid and SO2 which allows the equilibrium to be the a-keto diester 8 in 53% yield. Its hydrolysis was
shifted (Scheme 1).
carried out in 93% yield by addition of the stoichio-
In this paper, we report the first enzymatic syn- metric amount of LiOH in water to give 3. In spite of
thesis of 4,4-disubstituted l-Glu, namely 4,4-di- the modest yield observed in the oxidation step, the
methyl- and 4-hydroxy-4-methyl-Glu (1 and 2), by lithium salt 3 was prepared with an overall yield of
transamination of the corresponding 4,4-disubsti- 31% from commercial 2,2-dimethylglutaric acid.
tuted a-ketoglutaric acid, catalyzed by GOT
The activity of GOT for 3 was measured in phos-
phate buffer (pH 7.5) (Scheme 1) using aspartic acid
as the donor amino acid. Oxalacetic acid produced in
the transamination was reduced by NADH in the pre-
sence of malate dehydrogenase. The decrease in ab-
sorbance at 340 nm due to the oxidation of NADH al-
lowed us to monitor the reaction. Compound 3 was
found to be a substrate for GOT with a Km of 15 mM
and a kcat of 4% relative to the natural substrate a-ke-
(Scheme 1).
Results and Discussion
4,4-Dimethyl-l-glutamic Acid (1)
To the best of our knowledge, 1 has never been iso-
lated from natural sources. It was obtained by Koski-
nen and coworkers. as a by-product during the alky-
lation of protected l-glutamic acid,[6a] and by Bisset
et al.[6b] in a protected form.
Table 1. Activity of glutamic oxaloacetic aminotransferase
(GOT) towards a-ketoglutaric acid analogues.
2-Oxo-4,4-dimethylglutaric acid, the required sub-
strate for transamination, has not been described
before. Having already prepared 2-oxo-4-methylglu-
taric acid by methylation of dimethyl 2,2-dimethoxy-
glutarate, we tried to achieve the dimethylation by
using an excess of alkylating agent. However, even
starting from the monomethyl intermediate, the yield
of the desired dimethylated product was never higher
than 4%. Since the dimethylation of a 2-oxoglutaric
acid derivative appeared difficult, we turned our at-
tention to the oxidation of commercial 2,2-dimethyl-
Scheme 2. Synthesis of lithium 2-oxo-4,4-dimethylglutarate
Adv. Synth. Catal. 2001, 343, 692±697
693

asc.wiley-vch.de
toglutaric acid (Table 1). These values can be com-
pared with those observed for the (4S)- and (4R)-2-
oxo-4-methylglutaric acids, respectively. The methyl
group in the R position does not affect the catalysis
by GOT, whereas for the S-isomer the affinity for the
enzyme site as well as the reactivity are decreased.
When both positions are substituted, the kinetic con-
stants are, of course, in the same range as those of the
S-isomer.
The transamination was carried out on a preparative
scale in the presence of an excess of cysteinesulfinic
acid (CSA) as the donor amino acid. 4,4-Dimethyl-Glu
formed is easily purified by selective absorption on a
Dowex H+ resin. Excess of cysteinesulfinic acid is
eluted first, and 1, eluted with 1 N NH4OH was ob-
tained with a 70% yield and a purity higher than 95%
according to NMR spectra and HPLC analysis.
Scheme 3. Synthesis of potassium 2-oxo-4-hydroxy-4-
methylglutarate.
8% relative to the natural substrate. These values
can be compared to those of the C-4 monosubstitued
oxoglutarates reported in Table 1. We already ob-
served that GOT does not discriminate between the
enantiomers of 4-hydroxy-2-oxoglutaric acid since
(2S,4R)- and (2S,4S)-4-hydroxy-Glu are formed in
equal amounts during the synthesis.[8a,8b] The kcat
and Km values reported in entry 6 are for the racemic
substrate. On the other hand, GOT prefers the (4R)-4-
methyl isomer (entry 2), to the (4S)-4-methyl (en-
try 3),[8d] so that one can expect that the position of
the hydroxy group is not determinant for activity and
that the enantiomer of 4 that is the best substrate will
have the methyl group in the same configuration as
that in the (4R)-4-methyl compound. Therefore,
GOT should present some enantioselectivity towards
4-hydro-4-methyl-2-oxoglutaric acid in favor of the
(4S)-enantiomer.
The measured kinetic constant of the transamina-
tion reaction shows that GOT would be an efficient
catalyst for synthetic purposes. The reaction was then
achieved on a preparative scale with CSA as the donor
amino acid. CSA was first added in deficit so that the
conversion rate would be under 50%. In the condi-
tions used (0.4 equivalent of CSA), only one isomer
was produced. After purification by ion exchange
chromatography, 2a was obtained in 84% yield. When
the transamination was carried out with an excess of
CSA, the second isomer 2b appeared, but it never ex-
ceeded 20% of the mixture. Careful purification of the
mixture by ion exchange chromatography allowed us
to obtain a pure sample of 2b. The configurations of
2a and 2b were determined by comparison with
1H NMR spectra of authentic samples previously
synthesized by us.[16] The spectra of the two isomers
in diluted NaOD solution are quite different, specially
the protons at C3 with d values of 1.74 and 2.31 ppm
for the 2S,4S-configuration (2a) and 1.94 and
2.10 ppm for the 2S,4R-configuration (2b). Moreover,
having in hands the four possible isomers of 2, we can
confirm the assignment of the configuration 2S,4S for
2a and 2S,4R for 2b by chiral gas chromatography
after derivatization.[19] No traces of the 2R-isomers
were detected, as expected due to the stereospecifi-
city of the transaminase.
4-Hydroxy-4-methyl-l-glutamic Acid (2)
(4S)- and (4R)-4-hydroxy-4-methyl-l-glutamic acids
are natural compounds, mainly occurring in plants.
The (4R)-isomer is found in Ledenbergia roseoae-
na,[12] the (4S)-isomer in Pandanus veichii,[13] while
both isomers are present in Phyllitis scolopen-
drium.[14] These compounds can be obtained by re-
duction of properly substituted isoxazolines[15] and
we recently published a synthesis of the four stereo-
isomers (of the d- and l-series) based on this meth-
od.[16]
2-Oxo-4-hydroxy-4-methylglutaric acid (4) is
formed by aldol condensation of pyruvic acid and is
often present as impurities in commercial pyruvic
acid or sodium pyruvate samples.[17] Recently, both
stereoisomers of diethyl 2-oxo-4-hydroxy-4-methyl-
glutarate were synthesized in 70 to 95% enantiomeric
excess by catalytic asymmetric homo-aldol reaction
of ethyl pyruvate.[18] Racemic 4 has been synthesized
by base-catalyzed aldol condensation of pyruvic acid
at pH 12 and purified by chromatography on a Sepha-
dex G-15 column. However the yield was not indi-
cated in this report.[17] We repeated this experiment
with a few modifications: an aqueous solution of
pyruvic acid was maintained at pH 12 for 15 minutes,
then the solution was neutralized by addition of Dow-
ex H+ resin, and concentrated to give a solid. The so-
lid was dissolved in a methanol/water mixture and
precipitated by addition of diethyl ether, giving race-
mic potassium 2-oxo-4-hydroxy-4-methylglutarate
(4) in 64% yield (Scheme 3). Analysis by NMR spec-
troscopy showed that 4 is about 90% pure. Attempts
to purify this compound were unsuccessful, probably
due to the easy retro-aldol reaction.
In spite of this low purity, we decided to study the
transamination in the presence of GOT in the same
Conclusion
conditions as described for 2-oxo-4,4-dimethylgluta- GOT-catalyzed transamination constitutes a good
ric acid. We found a Km value of 18 mM and a kcat of method to synthesize 4,4-disubstituted l-glutamic
694
Adv. Synth. Catal. 2001, 343, 692±697

FULL PAPER
2´CH3), 2.37 (d, 2H, J = 5 Hz, CH2), 3.57 and 3.66 (2´s, 2´3H,
2´CO2CH3), 4.50 (t, 1H, J = 5 Hz, CH), 7.49 (dd, 2H, J = 9 Hz,
2´CH Ar meta), 7.62 (dd, 1H, J = 9 Hz, CH Ar para), 8.02 (d,
2H, J = 9 Hz, 2´CH Ar ortho); 13C NMR (CDCl3): d = 25.3 and
25.5 (2´4-CH3), 38.6 (C3), 41.8 (C4), 50.6 (C2), 51.9 and 52.7
(2´CO2CH3), 128.8 (4´CH Ar ortho and meta), 133.6 (CH Ar
para), 135.9 (C Ar), 170.3 (C1), 177.4 (C5), 194.8 (2-CO); IR
(neat): m = 2952.6, 1731.6, 1688.3, 1448.2, 1293.4, 1138.9,
acids, such as 4,4-dimethyl-Glu and (4S)-4-hydroxy-
4-methyl-Glu. The corresponding ketoglutaric acid
analogues are quite easy to prepare. These results
confirm that the proR hydrogen on position 4 of a-ke-
toglutaric acid can be substituted by an apolar methyl
group without loss of the enzymatic activity, whereas
the substitution of the proS hydrogen by a methyl or a
hydroxy group results in a decrease in both the affi-
nity for the enzymatic site (higher Km) and the transa-
mination rate (Vmax).
690.0 cm±1
.
Dimethyl 2-Diazo-4,4-dimethylglutarate (7)
To a solution of 6 (244 mg, 0.83 mmol) in 5 mL of dry CH2Cl2
under argon at 0 °C was added DBU (260 lL, 1.75 mmol)
dropwise. The colorless solution turned to yellow. Then p-
nitrobenzenesulfonyl azide (400 mg, 1.75 mmol) in 4 mL of
dry CH2Cl2 was added and the mixture allowed to warm to
room temperature. After stirring during 30 min, 4 mL of
phosphate buffer (pH 7, 0.5 N) and then 4 mL of water were
added. The solution was extracted with CH2Cl2, the organic
phase dried over MgSO4 and the solvent evaporated under
vacuum. Chromatography of the residue eluting with di-
chloromethane/acetone (97:3) gave a yellow liquid; yield:
177 mg (99%); 1H NMR (CDCl3): d = 1.24 (s, 6H, 2´4-CH3),
2.55 (s, 2H, CH2), 3.68 and 3.76 (2´s, 2´3H, 2´CO2CH3);
13C NMR (CDCl3): d = 24.7 and 24.8 (2´4-CH3), 33.7 (C3),
43.4 (C4), 52.0 (2´CO2CH3), 177.3 (C1 and C5); MS: m/z (re-
lative intensity %) = 186 (31, M+ ± N2), 171 (29), 127 (22), 123
(21), 105 (12), 95 (41), 73 (69), 59 (100, CO2Me), 55 (37), 41
(72), 29 (36), 15 (29); IR (neat): m = 2953.9, 2085.4, 1735.1,
Experimental Section
General
Chemicals were purchased from Aldrich or Fluka and en-
zymes are from Sigma. Chromatography were carried out
on 220 ± 400 mesh silica gel using the `flash' methodology.
Thin-layer chromatography was performed on Merck pre-
coated silica gel 60 F254 plates and spots were visualized with
vanillin [vanillin (1 g) is dissolved in MeOH (60 mL) and
conc. H2SO4 (0.6 mL)]. 1H (400 MHz), 13C (100 MHz) experi-
ments were performed on a Bruker AC 400 spectrometer.
CPV chromatography were performed on an apparatus with
a chiral capillary column CHROMPACK (25 m ´ 0.25 mm)
coated with Chirasil-l-valine.
1696.7, 1453.3, 1310.5, 1144.1 cm±1
.
Dimethyl 2,2-Dimethylglutarate (5)
2,2-Dimethylglutaric acid (25.54 g, 0.16 mol) was dissolved
in 150 mL of methanol. Thionyl chloride (25.5 mL,
0.35 mol) was added dropwise under stirring and the solu-
tion was refluxed during 2 h. Methanol was partially evapo-
rated under vacuum and ethyl acetate was poured onto the
residue. The organic phase was washed with saturated so-
dium bicarbonate solution until no more carbon dioxide
evolved and then with brine. The organic phase was dried
over MgSO4 and concentrated under vacuum to give an oil;
yield: 28.7 g (96%); 1H NMR (CDCl3): d = 1.18 (s, 6H, 2´CH3),
1.88 (m, 2H, CH2), 2.29 (m, 2H, CH2), 3.67 (s, 6H, 2´CO2CH3);
13C NMR (CDCl3): d = 25.0 (2´2-CH3), 30.1 (C3), 35.2 (C4),
41.8 (C2), 51.8 and 52 (2´CO2CH3), 173.9 (C11), 177.8 (C5)
ppm.
Dimethyl 2-Oxo-4,4-dimethylglutarate (8)
To a solution of 7 (130 mg, 0.61 mmol) in 5 mL of CH2Cl2 was
added in portions m-chloroperbenzoic acid (175 mg,
0.61 mmol). After one hour of stirring at room temperature,
the solvent was evaporated under vacuum. Pentane was
then added to the residue and the m-chlorobenzoic acid
formed was filtered. The filtrate was washed with diluted bi-
carbonate solution, dried over MgSO4 and evaporated under
reduced pressure. The liquid thus obtained was chromato-
graphed on silica gel (eluent: cyclohexane/ethyl acetate,
9:1) to afford an oil; yield: 65 mg (53%); 1H NMR (CDCl3):
d = 1.27 (s, 6H, 2´4-CH3), 3.12 (s, 2H, CH2), 3.65 (s, 3H, 5-
CO2CH3), 3.87 (s, 3H, 1-CO2CH3); 13C NMR (CDCl3): d = 25.6
and 25.7 (2´4-CH3), 40.7 (C4), 48.2 (C3), 52.2 and 53.1
(2´CO2CH3), 161.2 (C1), 177.0 (C5), 191.8 (C22); MS: m/z (re-
lative intensity %) = 203 (4, M+ + H), 171 (5), 143 (27), 115
(20), 83 (25), 73 (100), 59 (31), 41 (21), 29 (12), 15 (11); IR
(neat): m = 2957.5, 2874.5, 1758.1, 1731.9, 1434.3, 1388.8,
Dimethyl 2-Benzoyl-4,4-dimethylglutarate (6)
NaH (60% in mineral oil; 8.51 g, 210 mmol) was washed three
times under argon with dry hexane and then suspended in
100 mL of dry 1,2-dimethoxyethane (DME). The mixture
was cooled to 0 °C and 5 (10 g, 53 mmol) in 20 mL of DME
was added dropwise under argon. The mixture was stirred
at 0 °C during 10 min. Methyl benzoate (10 mL, 80 mmol) in
20 mL of dry DME was then added, followed by 10 drops of
methanol. The reaction mixture was refluxed overnight. Ex-
cess of NaH was destroyed adding 1 N HCl until pH 4. Then
the solution was extracted with ethyl acetate, the organic
phase was dried over MgSO4 and concentrated under va-
cuum. Chromatography of the residue eluting with cyclo-
hexane/ethyl acetate (8:2) afforded a yellow liquid; yield:
1257.4, 1166.4, 1060.3, 969.3 cm±1
.
2-Oxo-4,4-dimethylglutaric Acid Dilithium Salt (3)
Compound 8 (3.1 g, 15 mmol) was dissolved in 30 mL of
water and the solution was cooled to 4 °C. Solid lithium hy-
droxide (74 mg, 30 mmol) was added in portions while
maintaining a pH below 12. Acetone was then poured in un-
til precipitation of the product occurred. The precipitate was
filtered, washed with acetone and then with diethyl ether to
obtain a white solid; yield: 2.6 g (93%); 1H NMR (D2O): d =
1
9.2 g (65%); H NMR (CDCl3): d = 1.18 and 1.25 (2´s, 2´3H,
Adv. Synth. Catal. 2001, 343, 692±697
695

asc.wiley-vch.de
1
1.22 (s, 6H, 2´4-CH3), 3.07 (s, 2H, CH2); 13C NMR (D2O): d =
28.2 (2´4-CH3), 44.0 (C4), 52.2 (C3), 173.3 (C1), 189.1 (C5),
(C4), 177.0 (C1), 184.0 (C5) ppm; the H NMR spectral data
recorded in NaOD are in agreement with those already pub-
lished by us.[16] 1H NMR (NaOD): d = 1.49 (s, 3H, CH3), 1.89
(dd, 1H, J = 11, 15 Hz, part of CH2), 2.43 (dd, J = 2, 15 Hz, part
of CH2), 3.55 (dd, 1H, J =2, 11 Hz, CH).
207.4 (C2); IR (neat): m = 2976.5, 1722.6, 1653.1, 1576.3 cm±1
.
2-Oxo-4-hydroxy-4-methylglutaric Acid
Dipotassium Salt (4)
(2S,4R)-4-Hydroxy-4-methyl-l-glutamic Acid (2b)
A solution of pyruvic acid (4 g, 45.4 mmol) in 50 mL of water
was adjusted to pH 12 with a 6 N KOH solution. After 30 min
of stirring, the pH is then adjusted to the neutrality by addi-
tion of a strong acidic ion exchange resin (Dowex 50WX8,
100 ± 200 mesh) and the resin removed by filtration. The
keto acid is dissolved in a hydro-methanolic solution (9:1,
methanol/water) and then purified by two precipitations in
diethyl ether to give a white solid; yield: 3.7 g (64%), that
was used for transamination without further purification;
spectroscopic data are in agreement with those reported by
Margolis et al.[17]
The transamination was carried out in the presence of
1.1 equivalent of CSA, and 2a and 2b were isolated with an
overall yield of 74%. Analysis of the NMR spectrum of the
mixture showed that the ratio 2a/2b is around 80:20. Separa-
tion of these two diastereoisomers was realized as pre-
viously described for a mixture of the four isomers.[20]
300 mg of the 2a/2b mixture were applied to a column
(50 cm long and 1 cm diameter) of a strong basic resin
(Dowex 2X8, 200 ± 400 mesh, acetate form). The column
was eluted with 0.5 M acetic acid and fractions of 5 mL col-
lected. 2b was eluted first in fractions 80 to 90, 2a appeared
in the fractions 91 to 105. The fractions were neutralized
with NaOH and the amino acids isolated after the usual
work-up. (application to a strong acidic resin and elution
with 1 M NH4OH). A pure sample of 2b was obtained and
characterized by comparison with an authentic sample al-
ready prepared by us.[16] [a]2D0: ±7.8° (c 0.6, 0.5 N NaOH);
1H NMR (D2O): d = 1.51 (s, 3H, CH3), 2.15 (dd, 1H, J = 9,
15 Hz, part of CH2), 2.39 (d, J = 15 Hz, part of CH2), 3.99 (d,
1H, J = 9 Hz, CH); 1H NMR (NaOD): d = 1.50 (s, 3H, CH3),
2.04 (dd, 1H, J =8, 15 Hz, part of CH2), 2.23 (dd, J =5, 15 Hz,
part of CH2), 3.65 (dd, 1H, J = 5, 8 Hz, CH) ppm.
General Procedure for the Transamination of
a-Keto Acids using Glutamic Oxalacetic
Transaminase (GOT)
To a 1 mmol solution of the a-keto acid dilithium salt in
50 mL of distilled water was added 1 mmol of cysteinesulfi-
nic acid monohydrate (CSA) and the pH was adjusted to 7.5
by addition of a 1 M NaOH solution. GOT (50 lL, 100 U) was
then added and the mixture was stirred during 24 h at 30 °C.
The product was purified through a 20-mL strong acidic re-
sin column (Dowex 50WX8, 16 ± 40 mesh, H+ form) and
eluted with 40 mL of a 1 M NH4OH solution. The alkaline
fractions were evaporated under reduced pressure and be-
low 40 °C (to avoid corresponding pyroglutamic acid forma-
tion) affording the amino acid as a zwitterion. The remain-
ing starting material can be recycled by evaporating the
acidic fractions from the resin up to 50 mL, adjusting the
pH to 7.5 with a 1 M NaOH solution and starting a new trans-
amination by adding the required quantity of enzyme.
References
[1] R. O. Duthaler, Tetrahedron 1994, 50, 1539±1650.
[2] G. Johnson, Bioorg. Med. Chem. Lett. 1993, 3, 9.
[3] (a) Z.-Q. Gu, D. P. Hesson, J. C. Pelletier, M. L. Mac-
caecchini, J. Med. Chem. 1995, 38, 2518±2519; (b) N.
Todeschi, J. Gharbi-Benarous, F. Acher, V. Larue, J.-P.
Pin, J. Bockaert, R. Azerad, J.-P. Girault, Bioorg. Med.
Chem. 1997, 5, 335±352.
4,4-Dimethyl-l-glutamic Acid (1)
The transamination was carried out as described above.
Compound 1 was isolated in 70% yield after one recycling.
[a]D20 : +33.3° (c1, 0.1 N HCl); 1H NMR (D2O): d = 1.26 and
1.30 (2´s, 2´3H, 2´CH3), 1.97 (dd, 1H, J = 2, 15 Hz, part of
CH2), 2.12 (dd, 1H, J = 9, 15 Hz, part of CH2), 3.79 (dd, 1H,
J = 2, 9 Hz, CH); 13C NMR (D2O): d = 26.1 and 31.1 (2´4-
CH3), 43.8 (C3), 45.9 (C4), 55.1 (C2), 178.5 (C1), 189.3 (C5);
IR (neat): m = 3500 ± 2500, 1584.1, 1521.6, 1404.9 cm±1; HRMS
(electrospray): calcd. for C7H14NO4 (M + H): 176.0923;
found: 176.0923.
[4] C. Najera, M. Yus, Tetrahedron: Asymmetry 1999, 10,
2245±2303.
[5] (a) M. Del Bosco, A. N. C. Johnstone, G. Bazza, S. Lo-
patriello, M. North, Tetrahedron 1995, 51, 8545±8554;
(b) S. Hanessian, R.-Y. Yang, Tetrahedron Lett. 1996,
50, 8997±9000; (c) E. Coudert, F. Acher, R. Azerad,
Synthesis 1997, 863±865; (d) A. Escribano, J. Ezquerra,
C. Pedregal, A. Rubio, B. Yruretagoyena, S. R. Baker,
R. A. Wright, B. G. Johnson, D. D. Schoepp, Bioorg.
Med. Chem. Lett. 1998, 8, 765±770.
[6] (a) A. M. P. Koskinen, H. Rapoport, J. Org. Chem. 1989,
54, 1859±1866; (b) V. Bavetsias, A. L. Jackman, J. H.
Marriot, R. Kimbell, W. Gibson, F. T. Boyle, G. M. F.
Bisset, J. Med. Chem. 1997, 40, 1495±1510.
[7] P. P. Taylor, D. P. Pantaleone, R. F. Senkpeil, I. G.
Fotheringham, Trends Biotechnol. 1998, 16, 412±418.
[8] (a) N. Passerat, J. Bolte, Tetrahedron Lett. 1987, 28,
1277±1280; (b) F. Echalier, O. Constant, J. Bolte, J.
Org. Chem. 1993, 58, 2747±2750; (c) V. Helaine, J. Ros-
si, J. Bolte, Tetrahedron Lett. 1999, 40, 6577±6580; (d)
(2S,4S)-4-Hydroxy-4-methyl-l-glutamic Acid (2a)
The transamination was carried out with only 0.4 equiva-
lents of CSA and 2a was isolated in 84% yield. [a]D20: ±3.7° (c
0.8, 0.5 N NaOH); HRMS (electrospray): calcd. for C6H12NO5:
178.0715; found: 178.0717; 1H NMR (D2O): d = 1.50 (s, 3H,
CH3), 2.05 (dd, 1H, J = 11, 15 Hz, part of CH2), 2.56 (dd, J =
2, 15 Hz, part of CH2), 3.71 (dd, 1H, J = 2, 11 Hz, CH);
13C NMR (D2O): d = 29.5 (4-CH3), 42.1 (C3), 55.2 (C2), 79.1
696
Adv. Synth. Catal. 2001, 343, 692±697

FULL PAPER
V. Helaine, J. Bolte, Eur. J. Org. Chem. 1999, 3403±
3406.
[9] D. F. Taber, K. You, Y. Song¹ J. Org. Chem. 1995, 60,
1093±1094.
[10] M. T. Reagan, A. Nickon, J. Am. Chem. Soc. 1968, 90,
4096±4105.
[11] E. D. Thorset, Tetrahedron Lett. 1982, 23, 1875±1878.
[12] J. Jadot, J. Casimir, A. Loffet, Biochim. Biophys. Acta
1967, 136, 79±88.
[15] V. JaÈger, H. Grund, V. Buss, W. Schwab, I. MuÈ ller, R.
Schohe, R. Franz, R. Ehrler, Bull. Soc. Chim. Belg.
1983, 92, 1039±1053.
[16] V. Helaine, J. Bolte, Tetrahedron : Asymmetry 1998, 9,
3855±3861.
[17] S. A.Margolis, B. Coxon, Anal. Chem. 1986, 58, 2504±
2510.
[18] K. Juhl, N. Gathergoo, K. A. Jorgensen, Chem. Com-
mun. 2000, 2211±2212.
[13] F. Alderweireldt, J. Jadot, J. Casimir, A. Loffet, Bio-
chim. Biophys. Acta 1967, 136, 89±94.
[14] L. K. Meier, J. Sorensen, Phytochemistry 1979, 18,
1173±1175.
[19] M. Maurs, C. Ducrocq, A. Righini-Tapie, R. Azerad, J.
Chromatogr. 1985, 325, 444±449.
[20] E. P. Kristensen, L. M. Larsen, O. Olsen, H. Sorensen
Acta Chem. Scand. B 1980, 34, 497±504.
Adv. Synth. Catal. 2001, 343, 692±697
697