A focus on the asymmetric synthesis of a novel threo-β-benzyl-β- hydroxy aspartate analogue

Article abstract of DOI:10.1016/j.tetasy.2012.01.004

Considering the biological activity of β-alkyl-β-hydroxyaspartate derivatives as potent blockers of glutamate transporters (EAATs) impacting on glutamatergic synapses activity, we have developed a concise, asymmetric synthesis of enantiomerically pure thr

Full text of DOI:10.1016/j.tetasy.2012.01.004

Tetrahedron: Asymmetry 23 (2012) 94–99
Contents lists available at SciVerse ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
A focus on the asymmetric synthesis of a novel threo-b-benzyl-b-hydroxy
aspartate analogue
Sofiane Mekki a,b, Salima Bellahouel b, Nicolas Vanthuyne d, Marc Rolland c, Aicha Derdour b, Jean Martinez a,
Michel Vignes a, Valérie Rolland a,
⇑
a IBMM-UMR 5247, CNRS, UM2, UM1, Place E. Bataillon, 34095 Montpellier Cédex 5, France
b Laboratoire de Synthèse Organique Appliquée LSOA, Université d’Oran, Département de Chimie, Bp 1524 El M’Naouer, Oran 31000, Algeria
c IEM-UMR 5635, CNRS, UM2-ENSCM, Place E. Bataillon, 34095 Montpellier Cédex 5, France
d Université Paul Cézanne-UMR 6263, Chirosciences, Avenue Escadrille Normandie Niémen, 13397 Marseille Cédex 20, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Considering the biological activity of b-alkyl-b-hydroxyaspartate derivatives as potent blockers of gluta-
mate transporters (EAATs) impacting on glutamatergic synapses activity, we have developed a concise,
asymmetric synthesis of enantiomerically pure threo-b-benzyl-b-hydroxyaspartates. The key step is a
regiospecific and stereoselective Sharpless asymmetric aminohydroxylation (SAA) on previously synthe-
sized benzyl fumarate.
Received 16 December 2011
Accepted 6 January 2012
Available online 1 February 2012
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
EAAT2 (Ki = 18 lM) (Fig. 1). Our group described the enantioselec-
tive synthesis of T3MG in four steps by an asymmetric Michael
addition with ethyl crotonate using (1R,2R,5R)-2-hydroxypinan-
3-one as a chiral auxiliary. The results concerning biological activ-
ity were evaluated by recording spontaneous excitatory transmis-
sion and showed that (2S,3R)-3-methyl glutamic acid exhibited an
important inhibitory effect on glutamate transport.8
As the predominant excitatory neurotransmitter, glutamate po-
tently modulates the function of almost all neuronal circuits in the
central nervous system. To limit glutamate receptor activation
which can trigger excitotoxic mechanisms and cell death when
overstimulated, extracellular concentrations of excitatory aminoac-
ids are tightly controlled by both neuron and glial cells. Excitatory
amino acid transporters (EAATs) have been so far involved in a wide
array of functions ranging from glutamate recycling to synaptic ac-
tions such as signal termination, synaptic spillover prevention but
also excitotoxicity protection.1 A substantial number of pharmaco-
logical agents have been shown to inhibit glutamate transport but
most of them act as competitive substrates, leading to reversed
We have previously reported a concise, asymmetric synthesis of
enantiomerically pure b-substituted b-hydroxy aspartates via an
aldol reaction between
oxazinone intermediate used as facial discriminator and
various
-keto esters with excellent diastereoisomeric excess.9
a glycine enolate derived from an
a
Enantiomerically pure b-methyl-b-hydroxyaspartates, b-isopro-
pyl-b-hydroxyaspartates, and b-phenylethyl-b-hydroxyaspartate
were prepared and characterized. These derivatives exhibited
moderate inhibitory effect on glutamate transport as evaluated
by electrophysiological measurements.
transport, such as L
-trans-pyrrolidine 2,3-dicarboxylate.2 In view
of this non-transportable blockers are more interesting tools for
the investigation of the physiological roles of glutamate transport-
ers than transportable ones.3,4 It was reported that some derivatives
of DL-threo-b-hydroxyaspartate act as blockers: DL-threo-b-benzyl-
oxyaspartate (DL-TBOA) being the most potent blocker for human
EAAT1 and EAAT2.5,6 The same group synthesized optically pure iso-
Following our interest in non-proteinogenic amino acid
syntheses and glutamatergic synapse studies, we propose here a
short, regiospecific, and stereoselective synthesis of
a new
L-b-threo-benzyl-b-hydroxyaspartate 6a (Fig. 2). The biological
mers and found that
L
-TBOA was the most efficient blocker for the
activity was first evaluated on cultured hippocampal neurons.
This new target would mimic both the inhibitory EAATs effect
human EAAT1–3, while D-TBOA revealed a difference in the phar-
macophores between EAAT1 and EAAT3.7
of L-b-threo-OH-Asp and of L-b-threo-benzyl-Asp that are the most
Glutamate analogue derivatives were also studied as EAAT
blockers. threo-3-Methyl glutamate (T3MG) as a racemic mixture
of (2S,3R)- and (2R,3S)-isomers blocked glutamate transport by
potent synthetic EAATs blockers known to date.3
2. Results and discussion
To the best of our knowledge, few research groups have devel-
oped a synthetic strategy that relies on the Sharpless asymmetric
⇑
Corresponding author. Tel.: +33 04 67 14 49 36; fax: +33 04 67 14 48 66.
E-mail address: Valerie.rolland@univ-montp2.fr (V. Rolland).
0957-4166/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2012.01.004

S. Mekki et al. / Tetrahedron: Asymmetry 23 (2012) 94–99
the most potent EAATs blockers:
NH2
95
NH2
NH2
NH2
CH3
COOH
Ph
COOH
COOH
HOOC
HOOC
HOOC
HOOC
COOH
OH
O
T3MG
(2S,3R) and (2R,3S)
Bn
L-βthreo-OH-Asp
L-β-threo-benzyl-Asp
DL-TBOA
moderate transportable EAATs blockers:
H3C
COOH
COOH
COOH
COOH
N
H
N
H
cis-3-CH3-L-trans-2,3-PDC
L-trans-2,3-PDC
Figure 1. Some EAATs blockers.
Our approach began by an esterification of commercially avail-
able (S)-malic acid followed by an alkylation of the corresponding
dibenzyl ester 1 to give dibenzyl 2-benzyl-3-hydroxysuccinate 2
with total stereospecificity (ee >99%) and better yield than de-
scribed by Lee et al.13 The (2R,3S)-configuration of compound 2
NH2
COOH
HOOC
Bn
HO
(2S,3S)-6-a
has been confirmed by [a]D measurements and NMR spectroscopic
analyses compared with those of the literature.13 Regiospecificity
and stereospecificity on C3 were induced by the C2 high steric hin-
drance and (S)-configuration of malic acid.
Figure 2.
L
-b-threo-Benzyl-b-hydroxyaspartic acid 6a.
Then Mitsunobu reaction afforded (E)-dibenzyl 2-benzylfuma-
rate 3 in moderate yield via a b-elimination reaction.14 This
dehydration can be carried out on racemic dibenzyl 2-benzyl-3-
hydroxysuccinate. The formation of dibenzyl 2-benzylidene
succinate regioisomer in 20% yield was due to uncontrolled double
bond isomerization. Even if the reaction rate was optimized, the
Mitsunobu dehydration reached a limit and compound 2 was still
present in the crude mixture beside (E) dibenzyl 2-benzylfumarate
3 and benzylidene succinate by-product.
As mentioned before, the key step of our strategy was a Sharp-
less asymmetrica (SAA), never described in the literature for the
preparation of b-alkylated-b-hydroxyaspartate derivatives.10,15
Lohray et al. had previously reported an investigation on diol for-
mation (SAD) during the SAA procedure; this competitive reaction
aminohydroxylation (SAA) reaction as the key step. This is most
likely due to the difficulty in controlling the regioselectivity and
enantioselectivity of the SAA reaction for complex substrates as
discussed in a recent review on this reaction.10
Concerning aspartate derivative synthesis, the SAA reaction was
present in very few syntheses: recently Bionda et al. reported a
threo-hydroxyaspartate synthesis in four steps applying Khalaf’s
strategy.11,12 In this synthesis, a standard SAA protocol on trans-
ethyl cinnamate was followed by a transformation of the phenyl
group in the carboxyl moiety in 45% overall yield and high regio
and stereoselectivities.
As presented in the introduction our target is L-b-benzyl-b-
hydroxyaspartate: C–C bond formation and Mitsunobu dehydra-
tion occurred before our study of the SAA reaction (Scheme 1).
OH
COOBn
1
BnOOC
1. LiHMDS, HMPA, THF, -78ºC,1h.
54%
2. BnBr, THF,-78ºC,1h.
OH
COOBn
COOBn
Ph
BnOOC
COOBn
BnOOC
20%
+
BnOOC
Bn
(E)-3; 50%
PPh3, DIAD,
Bn
phtalimide, RT, 3h
(2R,3S)-2
K2OsO2(OH)4 4mol%, (DHQD)2PHAL
5mol%, ACN/H2O (2 :1), Chloramine T,
RT, 48h
chiral separation (Chiralpak IB column)
SAA
75%
SAD
25%
OH
NHTs
COOBn
BnOOC
COOBn
BnOOC
HO
Bn
Bn
(2S,3S)-4
HO
(2S,3S)-5
Scheme 1. General plan of enantiomerically pure
L
-threo-b-benzyl b-hydroxy aspartate synthesis from (S)-malic acid.

96
S. Mekki et al. / Tetrahedron: Asymmetry 23 (2012) 94–99
fumarate 3 in 2 days. Activation under microwaves after the sec-
ond addition has been tested to reduce SAA reaction time, but
led to 100% diol 5 formation with the total disappearance of (E)-
benzylfumarate 3 in 90 min at 50 °C.
The SAA reaction on benzylfumarate 3 (E) should be totally reg-
iospecific. The same procedure carried out on a,b-unsaturated car-
boxamide olefins and on Baylis–Hillman olefins led to both
regioisomers in each case.17,18 This total regioselectivity was
confirmed by NMR and X-ray crystal structure determination of
the dimer of (2S,3S)-4a and (2R,3R)-4b as described by Mekki
et al. (Fig. 3).19
However, with this optimized SAA procedure with the very effi-
cient (DHQD)2PHAL chiral ligand, the enantiomeric ratio of the
compound (2S,3S)-4a was only 61:39 (ee = 22%) and revealed a
low asymmetric induction. The regiospecificity and stereoselectiv-
ity were totally lost when the SAA was carried out without
(DHQ)2PHAL or (DHQD)2PHAL ligands. With the aim to ameliorate
SAA enantioselectivity and despite
a slower reaction rate,
Chloramine M has been used as a nitrogen source instead of Chlo-
ramine T.20 The N-mesyl derivative of 4, prepared with chloramine
M presented unfortunately exactly the same ee than with Chlora-
mine T. This low stereoselectivity was due to the drastic steric
effect of trisubstituted inactivated olefin 3 (E).
Figure 3. An Ortep view of (2S,3S)-4a and (2R,3R)-4b dimer crystals.19
Prediction of the absolute configuration of the major enantiomer
(2S,3S)-4a was based on Sharpless’ mnemonic device,21 assuming a
similar mode of action of (DHQ)2PHAL and (DHQD)2PHAL ligands in
the asymmetric dihydroxylation (SAD) and asymmetric amin-
ohydroxylation (SAA) processes (Scheme 2).22 The (DHQD)2PHAL
directs addition to the b-face of trisubstituted olefin (E)-3 and
between nitrogen and oxygen sources was frequently observed.16
Effectively the reaction of phenyl cinnamate with stoichiometric
osmium tetroxyde and chloramine salts afforded a near quantita-
tive yield of diol and no aminoalcohol. In contrast, reaction with
10 mol % osmium catalyst, afforded a 14:86 ratio of diol to amino-
alcohol with exactly the same stereoinduction and with high levels
of enantioselectivity.
(DHQ)2PHAL directs addition to the
a-face. Even with low
stereoselectivity, the Sharpless’ mnemonic device allowed us to
predict which enantiomer of 2-benzyl-2-hydroxy-3-(4-meth-
ylphenylsulfonamido)succinate was obtained preferentially.
In parallel (2S,3S)-4a can be enriched by trituration in an EtOAc/
cyclohexane solvent system. The racemic compound crystallized
(Fig. 3) as a different crystalline lattice and the liquid was progres-
sively enriched affording (2S,3S)-4a in 87% ee.
Separation of both isomers has also been accomplished by
semi-preparative chiral HPLC on a Chiralpak IB column to obtain
and characterized separately (2S,3S)-4a and (2R,3R)-4b. (2S,3S)-
4a crystallized as very thin needles and the determination of
(2S,3S)-4a absolute configuration was impossible.
The standard SAA procedure with 10 mol % osmium catalyst16
applied to olefin (E)-3 was yielded in 100% dibenzyl-2-benzyl-
2,3-dihydroxysuccinate 5 via a competitive SAD. The SAA was first
optimized by increasing the equivalents of Chloramine T, diminish-
ing T°C, ACN proportion and equivalents of osmium catalyst in the
reaction mixture (from 10 mol % to 4 mol %).10 This procedure
afforded a 45:55 ratio of diol 5 to aminoalcohol 4 and (E)-benzyl-
fumarate 3 was still present in the mixture after 2 days. Finally,
only 10 mol % of (E)-benzylfumarate 3 was engaged in the first cat-
alytic cycle of the reaction. After 30 min of stirring, the remaining
90 mol % of (E)-benzylfumarate 3 was added and supplied the sec-
ond catalytic cycle. This new procedure afforded a 25:75 ratio of
diol 5 to aminoalcohol 4 with the total disappearance of benzyl
Finally saponification of (2S,3S)-4a with LiOH refluxing THF
over 7 h, followed by tosyl deprotection by HBr/AcOH 33% mixture
N
N
N
N
OH
Bn
O
O
CO2Bn
H
Dihydroquinidinyl Phthalazine
BnO2C
(2S,3S)-4a
(DHQD)2PHAL (L*)
MeO
OMe
NHTs
β-face
NW
NE
N
N
Bn
BnO2C
CO2Bn
(E)-3
H
N
N
N
N
SW
SE
Bn
HO
O
O
H
CO2Bn
NHTs
α-face
H
BnO2C
MeO
OMe
Dihydroquininyl Phthalazine
(DHQ)2PHAL (L*)
(2R,3R)-4b
N
N
Scheme 2. Sharpless’ mnemonic device for (E)-3: stereoinduction by (DHQ)2PHAL and (DHQD)2PHAL ligands.

S. Mekki et al. / Tetrahedron: Asymmetry 23 (2012) 94–99
97
Gel (Kieselgel 60, 230–400 mesh) as the stationary phase. Thin
layer chromatography was carried out on aluminum plates pre-
coated with Merck Silicagel 60F254 and the plates were visualized
by quenching of Ultra-Violet fluorescence, by iodine vapor or by
ninhydrin spray. Analytic HPLC was performed on a Waters appa-
ratus 717 plus autosampler with Millenium32 program on Symme-
NH2
HO
NHTs
LiOH refluxing THF,
microwaves, 2h
COOH
Bn
COOBn
Bn
HOOC
BnOOC
then refluxing HBr 33%
/AcOH, PhOH, 80%.
HO
(2S,3S)-4a
(2S,3S)-6a
ee>99%
tryShield™ RP18 3.5
gradient of ACN in H2O with 0.1% TFA in 5 min with 3 mL/min flow.
l
m 2.1 ꢀ 20 mm column and using a linear
Scheme 3. Final deprotection of (2S,3S)-4a.
Preparative HPLC was performed on a Waters Delta 4000 appara-
tus equipped with
a
Delta-Pack C18 column (15 lm,
40 ꢀ 100 nm) and a UV detector, with a linear gradient of ACN in
H2O with 0.1% TFA with 50 mL/min flow. Analytical chiral HPLC
experiments were performed on a unit composed of a Merck D-
7000 system manager, Merck-Lachrom L-7100 pump, Merck-La-
chrom L-7360 oven, Merck-Lachrom L-7400 UV-detector and Jasco
OR-1590 polarimeter. Chiralpak IB from Chiral Technology Europa
(Illkirch, France) was the chiral stationary phase used. Semi-pre-
parative separations were performed by successive injections on
a Knauer unit composed of a Smartline 1000 pump, a Smartline
3900 autosampler, a Smartline 2500 UV-detector and a valve to
collect separately the different isomers. Optical rotation values
were measured on a Perkin–Elmer 341 (20 °C, sodium ray). The
reactions under microwaves were carried out on an apparatus Bio-
tage Initiator Microwave Synthesizer. THF was distilled from so-
dium/benzophenone ketyl. Reagents were supplied from
commercial sources (ALDRICH, FLUKA).
with phenol as
a Br scavenger afforded (2S,3S)-b-benzyl-
b-hydroxyaspartic acid hydrochloride 6a in an 80% yield
(Scheme 3).15 The LiOH reaction has been carried out under
lwaves activation, thus diminishing the reaction time. Even with
these harsh conditions, no epimerization of 6a was detected, as
confirmed by chiral chromatography on a CROWNPACK CR(+) col-
umn. The final product (2S,3S)-b-benzyl-b-hydroxyaspartic acid
hydrochloride 6a was obtained with high enantiomeric purity (ee
>99%).
The biological activity of (2S,3S)-6a and (2R,3R)-6b, b-benzyl-
b-hydroxyaspartate analogues obtained separately was evaluated
by measuring the accumulation of synaptic-release in the presence
of these compounds with patch-clamp recordings of spontaneous
excitatory synaptic transmission in cultured hippocampal neurons.
Glutamate uptake impairment results in the occurrence of an in-
ward current due to the stimulation of NMDA receptors associated
with an increase of spontaneous transmission frequency.23 At a
4.1. (2R,3S)-Dibenzyl 2-benzyl-3-hydroxysuccinate 2
low concentration (20 lM), (2S,3S)-6a induced a moderate inward
current (20 pA range; n = 3), while the (2R,3R)-6b had no effect at
that concentration.
A
mixture of (S)-dibenzyl 2-hydroxysuccinate 1 (6.5 g,
20,7 mmol) in anhydrous THF (28 mL) was stirred under argon
and cooled at ꢁ78 °C. A solution of LiHMDS 1 M in anhydrous
THF (41 mL, 41 mmol, 2 eqts) followed by HMPA solution in anhy-
drous THF (7.1 mL, 41 mmol, 2 eqts) were added dropwise under
argon. After 1 h of stirring, a solution of benzylbromide (2.4 mL,
20.6 mmol,) in anhydrous THF (14 mL) was added dropwise. After
1 h at ꢁ78 °C, the reaction was stopped by slow addition of satu-
rated NH4Cl solution at rt. The solution was extracted three times
with Et2O. The combined organic layers were washed with water
and brine, dried over MgSO4, concentrated under reduced pressure.
The crude product was purified by chromatography (silica gel,
EtOAc/cyclohexane, 1:4) and yielded in (2R,3S)-2 (4.5 g, 54%) as a
3. Conclusion
A concise, asymmetric synthesis of enantiomerically pure threo-
b-benzyl-b-hydroxyaspartate as new blocker of glutamate trans-
porters was reported herein. The key step is a regiospecific and
stereoselective Sharpless asymmetric aminohydroxylation (SAA)
on substituted and deactivated benzylfumarate. The optimized
procedure described allowed us to reduce diol formation and in-
crease enantioselectivity by using different chiral ligands and chi-
ral separations; total regiospecificity was observed.
Biological activity was evaluated by measuring spontaneous
synaptic transmission and showed that (2S,3S)-3-amino-2-ben-
zyl-2-hydroxysuccinic acid 6 exhibited a moderate inhibitory ef-
fect on glutamate transport.
white powder. Mp 65–66 °C. ½a D20
¼ ꢁ17 (c 1.0, CH3OH) lit.:
ꢂ
[a
]
D = ꢁ17.2 (c 1.0 CH3OH).13 MS (ES+) m/z 405.2 (M+H)+, 427.2
(M+Na)+. 1H NMR (300 MHz, CDCl3): d (ppm) 2.93 (m, 1H), 3.14
(m, 3H), 4.07 (m, 1H, –CHOH), 4.94 (s, 4H, 2 ꢀ PhCH2O–), 7.10–
7.29 (m, 15H, 3 ꢀ Bn-ArH). 13C NMR (75 MHz, CDCl3) dC (ppm)
34.04, 50.36, 67.10, 67.77, 69.91, 126.86, 128.60, 128.71, 128.80,
129.43, 135.14, 138.39, 172.07, 173.56.
This approach may be considered a promising new strategy to
develop new target blockers of glutamate transport as new tools
to evaluate the impact of glutamate uptake on glutamatergic trans-
mission in the CNS.
4.2. (E)-Dibenzyl 2-benzylfumarate 3
4. Experimental
To a mixture of (2R,3S)-dibenzyl 2-benzyl-3-hydroxysuccinate 2
(2.95 g, 7.31 mmol, 1 eqt) in anhydrous THF (33 mL), phthalimide
(1.08 g, 7.31 mmol, 1 eqt), triphenylphosphine (1.92 g, 7.31 mmol,
1 eqt) and DIAD (1.49 g, 7.31 mmol, 1 eqt) were added, previously
desolved in THF. The mixture was stirred for 3 h at rt until the reac-
tion rate stopped. The THF was evaporated under reduced pressure
and the crude product was purified by chromatography (silica gel,
DCM) and dibenzyl 2-benzylfumarate 3 was yielded in 50% as a
colorless oil. MS (ES+) m/z 387.3 (M+H)+, 409.3 (M+Na)+. 1H NMR
(300 MHz, CDCl3) d (ppm) 4.15 (s, 2H, PhCH2–), 5.06 (s, 2H,
PhCH2O–), 5.15 (s, 2H, PhCH2O–), 6.85 (s, 1H, –CH@C–) 7.14–7.29
(m, 15H, 3 ꢀ Bn-ArH). 13C NMR (75 MHz, CDCl3) dC (ppm) 33.10,
66.72, 67.38, 126.37, 127.10, 128.21, 128.37, 128.39, 128.44,
Melting points were obtained using a Büchi 510 capillary appa-
ratus and were uncorrected. 1H NMR spectra were recorded at
200 MHz and 300 MHz using a Brüker AC200 and AC300 instru-
ments, Chemical shifts are quoted in parts per million and were
referenced to the residual solvent peak. The following abbrevia-
tions are used: s, singlet; d, doublet; t, triplet; q, quartet; m, mul-
tiplet. Coupling constants were reported in Hertz (Hz). Low
resolution mass spectra were recorded on micromass electrospray
instrument with only molecular ion and other major peaks being
reported. LC–MS identification by electrospray on HPLC Waters
Alliance 2690. Optical rotations were determined on a Perkin–El-
mer 241 polarimeter at room temperature in the appropriate sol-
vent. Flash chromatography was carried out using E-Merck Silica

98
S. Mekki et al. / Tetrahedron: Asymmetry 23 (2012) 94–99
128.44, 128.53, 128.61, 128.88, 135.16, 135.34, 137.93, 146.17,
165.41, 166.27.
(Mꢁ2ꢀCO2H)+. 1H NMR (300 MHz, D2O)
d (ppm) 3.2 (dd,
J2 = 13.8 Hz, 2H, PhCH2–), 4.47 (s, 1H, –CHNH2), 7.22–7.32 (m,
5H, Bn-ArH); 13C NMR (75 MHz, DMSO) dC (ppm) 40.55, 58.29,
76.27, 127.36, 128.48, 130.81, 135.39, 168.41, 173.06.
4.3. (2S,3S)-Dibenzyl 2-benzyl-2-hydroxy-3-(4-methyl-
phenylsulfonamido)succinate 4a
4.5. Electrophysiology
At first, (DHQD)2PHAL (13 mg, 0.017 mmol, 5 mol %) and (E)-3
(13 mg, 10 mol %) were dissolved in ACN (2 mL) at rt. After addi-
tion of a solution of K2[OsO2(OH)4] (5 mg, 0.014 mmol, 4 mol %)
in water (2 mL), chloramine-T (285 mg, 3 eqts) was added and
the mixture was stirred for 30 min at rt. The second portion of
(E)-3 previously solved in ACN (117 mg, 90 mol %, 1 mL) was added
subsequently. The reaction was monitored by HPLC and was com-
plete after 48 h. The reaction mixture was filtrated to eliminate
diol 5 (25%) and washed with a solution of ACN/H2O (1/1)
(10 mL). Dibenzyl 2-benzyl-2-hydroxy-3-(4-methylphenylsulfo-
namido)succinate 4 was obtained as a white powder (144 mg,
75% yield, 22% ee). Mixture can be enriched in the (2S,3S)-4a enan-
tiomer by trituration in EtOAc/cyclohexane (38.4 mg, 20% yield,
87% ee, mp 123–124 °C). Semi-preparative Chiral HPLC on
Chiralpak IB column in hexane/chloroform/ethanol (75/20/5) as
eluent, 5.0 mL/min, 254 nm, allows the separation of both
enantiomers of racemic or enriched dibenzyl 2-benzyl-2-hydro-
xy-3-(4-methylphenylsulfonamido)succinates.
Experiments were performed on cultured hippocampal neurons
grown on coverslips as previously described.24 On the day of the
experiment, a coverslip was transferred to the recording chamber
of an upright microscope (DMLFS, Leica). Cells were perfused (flow
rate ꢃ5 ml minꢁ1) with the extracellular solution containing:
124 mM NaCl, 3 mM KCl, 26 mM NaHCO3, 1.25 mM NaH2PO4,
1.5 mM CaCl2, 1 mM MgSO4, 10 mM
CO2: 95/5), 10 M glycine, and 50
D
-glucose (bubbled with O2/
l
lM picrotoxin. Spontaneous
excitatory postsynaptic currents (sEPSCs) were measured using
whole-cell recording by holding the voltage at ꢁ60 mV with glass
microelectrodes (4–5 MX resistance) filled with a solution com-
prising: 120 mM CsMeSO3, 1 mM NaCl, 1 mM MgCl2, 1 mM EGTA,
5 mM N-(2,6-dimethyl-phenylcarbamoylmethyl)-triethylammo-
nium bromide (QX-314), 5 mM HEPES (pH 7.3), and 4 mM
Mg-ATP. In order to isolate NMDA receptor mediated sEPSCs, NBQX
(5 l
M), an AMPA receptor antagonist, was applied in an Mg2+-free
extracellular solution.25 Signals were measured with patch-clamp
amplifier (Axopatch 200 B, Axon Instruments, USA) and digitized
(Digidata 1200 Interface, Axon Intruments, USA). They were
filtered at 1 kHz and sampled at 10 kHz. Continuous recording
and analysis of sEPSCs were performed with John Dempster’s soft-
wares ‘WinEDR’ (Strathclyde University). Experiments were per-
formed at 31 °C, by warming up the perfusing medium.
(2S,3S)-4a: tR = 6.18 min ½a D20
ꢂ
¼ ꢁ102:0 (c 0.5, DCM), ee >99%.
Mp = 131–133 °C.
(2R,3R)-4b: tR = 4.29 min, ½a D20
Mp = 131–133 °C.
ꢂ ¼ þ102:0 (c 0.5, DCM), ee >99%.
HRMS calcd for C32H31NO7S m/z (M+H)+ 574.1899; found
574.1896. MS (ES+) m/z: 574.2 (M+H)+, 596.2 (M+Na)+. 1H NMR
(200 MHz, CDCl3) d (ppm) 2.38 (s, 3H, CH3), 2.95 (dd, 2H, PhCH2–,
2J = 13.7 Hz), 3.5 (s, 1H, OH), 4.42 (d, 1H, –NH–CH–, 3J = 11.1 Hz),
4.83 (s, 2H, PhCH2O–), 4.99 (dd, 2H, PhCH2O–, 2J = 11.8 Hz), 5.57
(d, 1H, –NH-CH–, 3J = 11.1 Hz), 6.9 (d, 2H, CH3–C6H4–, 3J = 6.1 Hz),
7.09–7.41 (m, 15H, 3 ꢀ Bn-ArH), 7.66 (d, 2H, –C6H4-SO2–,
3J = 8.2 Hz). 13C NMR (75 MHz, CDCl3) dC (ppm) 21.8, 41.79,
61.27, 67.87, 69.16, 79.91, 127.42, 127.6, 128.43, 128.53, 128.83,
128.91, 129.09, 129.64, 129.85, 130.27, 134.31, 134.46, 134.71,
136.72, 144.01, 168.12, 171.88.
Acknowledgment
This work was supported by Erasmus Mundus Averroes
program.
References
1. Danbolt, N. C. Prog. Neurobiol. 2001, 65, 1–105.
2. Volterra, A.; Bezzi, P.; Rizzini, B. L.; Trotti, D.; Ullensvang, K.; Danbolt, N. C.;
Racagni, G. Eur. J. Neurosci. 1996, 8, 2019–2028.
3. Esslinger, C. S.; Agarwal, S.; Gerdes, J.; Wilson, P. A.; Davis, E. S.; Awes, A. N.;
O’Brien, E.; Mavencamp, T.; Koch, H. P.; Poulsen, D. J.; Rhoderick, J. F.;
Chamberlin, A. R.; Kavanaugh, M. P.; Bridges, R. J. Neuropharmacology 2005, 49,
850–861.
4. Maragakis, N. J.; Rothstein, J. D. Neurobiol. Dis. 2004, 15, 461–473.
5. Lebrun, B.; Sakaitani, M.; Shimamoto, K.; Yasuda-Kamatani, Y.; Nakajima, T. J.
Biol. Chem. 1997, 272, 20336–20340.
6. Shimamoto, K.; Lebrun, B.; Yasuda-Kamatani, Y.; Sakaitani, M.; Shigeri, Y.;
Yumoto, N.; Nakajima, T. Mol. Pharmacol. 1998, 53, 195–197.
7. Shimamoto, K.; Shigeri, Y.; Yasuda-Kamatani, Y.; Lebrun, B.; Yumoto, N.;
Nakajima, T. Bioorg. Med. Chem. Lett. 2000, 10, 2407–2410.
8. Wehbe, J.; Rolland, V.; Roumestant, M. L.; Martinez, J. Tetrahedron: Asymmetry
2003, 14, 1123–1126.
9. Wehbe, J.; Kassem, T.; Rolland, V.; Rolland, M.; Tabcheh, M.; Roumestant, M. L.;
Martinez, J. Org. Biomol. Chem. 2003, 1, 1938–1942.
10. Bodkin, J. A.; McLeod, M. D. J. Chem. Soc., Perkin Trans. 1 2002, 2733–2746.
11. Bionda, N.; Cudic, M.; Barisic, L.; Stawikowski, M.; Stawikovska, R.; Binetti, D.;
Cudi, P. Amino Acids 2010, online first 16 November (doi:10.1007/s00726-010-
0806-x).
12. Khalaf, J. K.; Datta, A. Amino Acids 2008, 35, 5007–5010.
13. Lee, S. S.; Li, Z. H.; Lee, D. H.; Kim, D. H. J. Chem. Soc., Perkin Trans. 1 1995, 2877–
2882.
14. Wada, M.; Sano, T.; Mitsunobu, O. Bull. Chem. Soc. Jpn. 1973, 46, 2833–
2835.
4.4. (2S,3S)-3-Amino-2-benzyl-2-hydroxysuccinic acid 6a
The enantiomerically pure succinate 4a (150 mg, 0.261 mmol,
ee >99%) and an aqueous solution of LiOH (132 mg, 3.14 mmol,
1,6 mL) were refluxed at 60 °C for 7 h. Activation under micro-
waves decreased the saponification reaction time which was com-
plete in only 2 h. The mixture was washed with EtOAc to eliminate
benzyl alcohol. Then the aqueous layer was acidified with aqueous
HCl (3 M) and extracted three times with EtOAc. The combined or-
ganic layers were washed with water, dried on MgSO4, and concen-
trated under reduce pressure. Then phenol (50.6 mg, 0.538 mmol,
3 eqts) and a solution of 33% HBr in AcOH (2 mL) were added
and refluxed for 21 h. Water (2 mL) was then added and the mix-
ture was stirred for 20 min at 50 °C. After concentration in vacuo,
the residue was purified by preparative HPLC applying isocratic
conditions (11% ACN in Water) with 50 mL/min flow to give
(2S,3S)-6a as a white solid (Y = 80% ; ee >99%). Chiral HPLC: Crown-
pak CR(+) column (150 ꢀ 4 mm), mobile phase: HClO4 (pH 1)/
methanol (90/10), flow: 0.5 mL/min, 200 nm, T = 8 °C.
15. Li, G.; Chang, H. T.; Sharpless, K. B. Angew. Chem., Int. Ed. 1996, 35, 451–454.
16. Lohray, B. B.; Bhushan, V.; Reddy, G. J.; Reddy, A. S. Indian J. Chem., Sect. B 2002,
41, 161.
17. Rubin, A. E.; Sharpless, K. B. Angew. Chem., Int. Ed. 1997, 36, 2637–2640.
18. Pringle, W.; Sharpless, K. B. Tetrahedron Lett. 1999, 40, 5151–5154.
19. Mekki, S.; Rolland, V.; Bellahouel, S.; Van der Lee, A.; Rolland, M. Acta
Crystallogr., Sect. C 2011, 67, o301–o305.
(2S,3S)-6a: Chiral HPLC: tR = 23.9 min
½
a 2D0
ꢂ
¼ ꢁ20:0 (c 0.3,
DMSO).
(2R,3R)-6b: Chiral HPLC: tR = 28.49 min ½a D20
ꢂ ¼ þ20:0 (c 0.3,
DMSO).
HRMS calcd for C11H13NO5 m/z (M+H)+ 240.0872; found
240.0867; MS (ES+) m/z: 240.1(M+H)+, 194.3 (MꢁCO2H)+, 150.1
20. Rudolph, J.; Sennhenn, P. C.; Vlaar, C. P.; Sharpless, K. B. Angew. Chem., Int. Ed.
1996, 35, 2810–2817.

S. Mekki et al. / Tetrahedron: Asymmetry 23 (2012) 94–99
99
21. Wang, Z.-H.; Zhang, X.-L.; Sharpless, K. B. Tetrahedron Lett. 1992, 33, 6407–
6410.
24. Crouzin, N.; de Jesus Ferreira, M. C.; Cohen-Solal, C.; Aimar, R. F.; Vignes, M.;
Guiramand, J. Free Radical Biol. Med. 2007, 42, 1326–1337.
22. Andersson, M. A.; Epple, R.; Fokin, W.; Sharpless, K. B. Angew. Chem., Int. Ed.
2002, 41, 472–475.
23. Jabaudon, D.; Shimamoto, K.; Yasuda-Kamatani, Y.; Scanziani, M.; Gähwiler, B.
H.; Gerber, U. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 8733–8738.
25. Lenda, F.; Crouzin, N.; Cavalier, M.; Guiramand, J.; Lanté, F.; Barbanel, G.;
Cohen-Solal, C.; Martinez, J.; Guenoun, F.; Lamaty, F.; Vignes, M. Amino Acids
2011, 40, 913–922.