Efficient synthesis of optically active α-substituted glutamate analogs possessing α-hydroxymethyl and α-alkoxymethyl groups

Article abstract of DOI:10.1016/S0040-4039(02)02810-1

Highly enantioselective synthesis of (2R)-α-(hydroxymethyl)glutamate (1), a selective agonist of mGluR2 and 3, was achieved in short steps using an asymmetric version of the Strecker synthesis. This was converted into its α-methoxymethyl- and α-benzyloxymethyl derivatives 2 and 3, possible ligands as tools to investigate glutamate receptors, via protection of the sterically hindered amino group by means of phase transfer catalyst.

Full text of DOI:10.1016/S0040-4039(02)02810-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1235–1238
Efficient synthesis of optically active a-substituted glutamate
analogs possessing a-hydroxymethyl and a-alkoxymethyl groups
Masanori Kawasaki, Kosuke Namba, Hidekazu Tsujishima, Tetsuro Shinada and Yasufumi Ohfune*
Graduate School of Science, Osaka City University, Sugimoto, Osaka 558-8585, Japan
Received 5 November 2002; revised 5 December 2002; accepted 6 December 2002
Abstract—Highly enantioselective synthesis of (2R)-a-(hydroxymethyl)glutamate (1), a selective agonist of mGluR2 and 3, was
achieved in short steps using an asymmetric version of the Strecker synthesis. This was converted into its a-methoxymethyl- and
a-benzyloxymethyl derivatives 2 and 3, possible ligands as tools to investigate glutamate receptors, via protection of the sterically
hindered amino group by means of phase transfer catalyst. © 2003 Elsevier Science Ltd. All rights reserved.
L
-Glutamic acid functions at many synapses in mam-
CCG-I has shown to exhibit antagonistic action to this
group.3 Recent efforts for the synthesis of a-substituted
glutamate analogs have revealed that (2R)-a-(hydroxy-
methyl)glutamate (1) acts as a selective agonist of the
group II receptors.8,9 Among these analogs, our atten-
tion was focused on the a-hydroxymethyl derivative 1
which is not only a useful ligand for glutamate recep-
tors but also can be viewed as a synthetic precursor
common to various types of ether-linked derivatives
which enable the preparation of an isotope-labelled or
resin-bounded glutamate analog. Described herein are a
five-step synthesis of optically active 1 and its conver-
sion to a-(methoxymethyl)- and a-(benzyloxymethyl)-
glutamate (2) and (3) as representative examples of the
a-(alkoxymethyl)glutamate analogs.
malian central nervous systems as an excitatory neuro-
transmitter and is implicated in the construction of
memory and learning as well as in the pathogenesis of
neuron damage to cause various neuronal diseases.1
Glutamate receptors have been classified into two
types: ionotropic (iGluRs) and metabotropic (mGluRs)
types. The mGluRs are coupled to intracellular second-
messenger systems and are sub-divided into the follow-
ing three groups by the sequence similarity of the
protein, transduction mechanisms, and pharmacologi-
cal profile to agonists and antagonists: group I
(mGluR1 and 5), group II (mGluR2 and 3), and Group
III (mGluR4 and 6ꢀ8).2 It has been reported that
group I receptors activate PLC resulting in stimulation
of PI hydrolysis and inhibition of different types of K+
ion channels which enhance neuronal toxicity and
accelerate neuronal cell death, while the other groups
reduce cAMP accumulation resulting in neuroprotect-
ing effect and are closely linked to construction of
memory and learning.2 Several types of agonists and
antagonists for the group II receptors have been devel-
oped.3 Representative agonists are
L
-CCG-I,4 DCG-
IV,5 and LY-3547406 whose structures fix the
conformation of glutamate to an extended form (anti–
anti conformation).5,7 On the other hand, a-substituted
The key to the synthesis is enantioselective construction
of the quarternary carbon center of (2R)-1. Our plan is
the use of an asymmetric version of the Strecker synthe-
sis which has proven to be an efficient method for the
synthesis of optically active b-hydroxy a,a-disubstituted
a-amino acid from the corresponding a-acyloxy ketone
(Eq. (1)).10 According to this method, the stereochem-
istry of the a-amino nitrile intermediate is controlled by
the configuration of the acyloxy amino chiral center,
Keywords: a-substituted glutamate analogs; mGluRs agonist; asym-
metric Strecker synthesis; sterically hindered amine; protection; ether-
ification.
* Corresponding author. Fax: 81-6-6605-3153; e-mail: ohfune@
sci.osaka-cu.ac.jp
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02810-1

1236
M. Kawasaki et al. / Tetrahedron Letters 44 (2003) 1235–1238
(1)
i.e. the relative configuration of the nitrile group is
usually anti to that of the amino acid side chain.
ine equilibrium (incorporation of 70% D atom at C6),
and (3) the configuration of the product at C5 was
1
assigned to be S by the H and 13C NMR data which
Accordingly, the use of
L-amino acid as the acyloxy
group would give rise to the desired (2R)-enantiomer 1.
were consistent with those of the related compounds, in
3
particular, JC–H values were 2.1 and 8.9 Hz indicating
axial orientation of the nitrile group.10
The Strecker precursor 5, possessing
L-valine as the
acyloxy group, was prepared from methyl 3-(chlorocar-
bonyl)propanoate in two steps (Scheme 1): (1) addition
of diazomethane and (2) insertion of the resulting dia-
Removal of the chirality transferring group leading to 1
requires oxidation of a-amino nitrile 7 to a-imino nitrile
8 followed by acidic hydrolysis. The classical method
was an initial chlorination with t-BuOCl and subse-
quent dechlorination with triethylamine. This method
gave 8 in 78% yield. On the other hand, the use of
ozone as the oxidant gave a mixture of 8 and a-amide
nitrile 9 in 94% yield as previously reported.14 The
mixture was subjected to acidic hydrolysis to give 1 in
98% yield. The sign of the optical rotation and 1H
NMR data of synthetic 1 were identical with those of
the reported data.15
zoketone 4b to Boc-L-valine using a catalytic amount of
Cu(acac)2.11 After removal of the Boc group with tri-
fluoroacetic acid, the resulting amine–TFA salt was
treated with powdered NaCN in 2-PrOH to give crude
a-amino nitrile 7 which was composed of a mixture of
(5S)- and (5R)-isomers (18:1) by 1H NMR spectral
analysis.12 Upon purification using short-pass column
chromatography on SiO2, the minor isomer was com-
pletely removed from the mixture to give in 64% yield
the pure (5S)-7.13 The stereochemical outcome of the
reaction was examined by the use of 2-PrOD as the
solvent or Na13CN. These experiments in combination
1
with the H and 13C NMR data of 7 indicated that (1)
We turned our attention to introduce an alkyl group
onto the a-hydroxymethyl group of 1 where trouble-
some g-lactam or d-lactone formation would compete.
Protection of the amino group with a Boc group was
no racemization was encountered during the reaction
since no D atom was incorporated into the C3 position
of 7, (2) the reaction accompanied certain imine–enam-
Scheme 1. Reagents and conditions: (a) CH2N2, 0°C, 5 min, 90%; (b) Boc-L-valine, 0.1 equiv. Cu(acac)2, toluene, rt, 1.5 h, 40%;
(c) (i) TFA, CH2Cl2, 0°C to rt, 2 h; (ii) 2 equiv. NaCN, 2-PrOH, rt, 18 h, 64%; (d) O3, AcOEt, −78°C, 30 min, 94%; (e) conc.
HCl, 100°C, 15 h, 98%.

M. Kawasaki et al. / Tetrahedron Letters 44 (2003) 1235–1238
1237
Scheme 2. Reagents and conditions: (a) Boc2O, 5.0 equiv. n-Bu4NOH, CH3CN–H2O (10:1), rt, 72 h, then, Dowex 50W×4; (b)
CH2N2, 0°C, 5 min, 57% (two steps); (c) 4.0 equiv. MeOTf, 4.4 equiv. DBMP, CH2Cl2, rt, 20 h, 65%; (d) 1.5 equiv. BTCA, 0.2
equiv. TMSOTf, CH2Cl2, 0°C, 3.5 h, 38%; (e) (i) 1N NaOH, THF, rt; (ii) TFA, CH2Cl2, rt, 89% for 2 and 93% for 3.
group. These results would enable the preparation of
useful tools to investigate glutamate receptors. Neu-
ropharmacological characterization of 2 and 3, and
preparation of the resin-connected glutamate through a
hydroxymethyl linker are in progress in our
laboratories.
examined first. Under the usual reaction conditions
using Boc2O in the presence of a tertiary amine or
NaOH as the base, only 5–27% of the desired N-Boc
derivative 10 was obtained due probably to steric rea-
sons (Scheme 2). The use of a phase transfer catalyst
(tetramethylammonium hydroxide in CH3CN) com-
monly used for the protection of a sterically hindered
amine16 formed a white suspension and gave in only 5%
yield the protected product due probably to poor solu-
Acknowledgements
bility of
1
in the solvent. We chose tetra-
butylammonium hydroxide as a hydrophobic base to
increase solubility of 1, but the solution, again, formed
a white suspension. Finally, it was found that addition
of a small amount of H2O (CH3CN/H2O=10:1) formed
a homogeneous solution. Using this reaction system,
the desired protected compound 11 was obtained after
diazomethane treatment in 57% yield. It is noted that
hazardous contamination of the ammonium salt in the
reaction mixture was easily removed by means of
Dowex 50W×4 (H+ form) prior to CH2N2 treatment.17
This work was supported by a Grant-in-Aid for Scien-
tific Research from the Ministry of Education, Culture,
Sports, Science and Technology, Japan, and the
Research for the Future Program (JSPS 99L01204).
References
1. (a) Collingridge, G. L.; Lester, R. A. Pharmacol. Rev.
1989, 40, 143–210; (b) Parthasarathy, H. S., Ed. Neuro-
logical disorders. Nature 1999, 399 (Suppl.), A1–A47.
2. (a) Nakanishi, S. Science 1992, 258, 597–603; (b) Conn,
P. J.; Pin, J.-P. Annu. Rev. Pharmacol. Toxicol. 1997, 37,
205–237 and references cited therein.
3. Schoepp, D. D.; Jane, D. E.; Monn, J. A. Neuropharma-
cology 1999, 38, 1431–1476.
4. Shimamoto, K.; Ishida, M.; Shinozaki, H.; Ohfune, Y. J.
Org. Chem. 1991, 56, 4167–4176.
5. Shimamoto, K.; Ohfune, Y. J. Med. Chem. 1996, 39,
407–423.
6. Monn, J. A.; Valli, M. J.; Massey, S. M.; Wright, R. A.;
Salhoff, C. R.; Johnson, B. G.; Howe, T.; Alt, C. A.;
Rhodes, G. A.; Robey, R. L.; Griffey, K. R.; Tizzano, J.
P.; Kallman, M. J.; Helton, D. R.; Schoepp, D. D. J.
Med. Chem. 1997, 40, 528–537.
Next, we examined the introduction of a methyl or a
benzyl group onto the a-hydroxymethyl group. A com-
bination of MeI or benzyl bromide with NaH, t-BuOK,
LDA, or LHMDS was not successful and gave d-lac-
tone as the major product. Since a weak base would
prevent undesired d-lactone formation, a combination
of 2,6-di-tert-butyl-4-methylpyridine (DBMP)18 and
MeOTf was examined next. The reaction proceeded
smoothly to give in 65% yield the desired methyl ether
12. In the case of the benzyl ether, a combination of
benzyltrichloroacetoimidate (BTCA) and TMSOTf19
was found to be an effective method to give benzyl
ether 13. Hydrolysis of the protected 12 and 13 with 1N
NaOH followed by TFA afforded 2 (89%) and 3 (93%),
respectively.20
7. Bertrand, H.-O.; Bessis, A.-S.; Pin, J.-P.; Acher, F. C. J.
Med. Chem. 2002, 45, 3171–3183.
8. Zhang, J.; Flippen-Anderson, J. L.; Kozikowski, A. P. J.
Org. Chem. 2001, 66, 7555–7559.
In summary, (2R)-a-(hydroxymethyl)glutamate (1), an
agonist of group II mGluRs, was synthesized in short
steps in a highly enantioselective manner. Its conversion
into two types of a-alkoxymethyl-substituted glutamate
analogs 2 and 3 accompanied a useful method for
protection of the sterically hindered amine and the
introduction of an alkyl group onto the hydroxymethyl
9. Choudhury, P. K.; Le Nguyen, B. K.; Langlois, N.
Tetrahedron Lett. 2002, 43, 463–464.
10. (a) Moon, S.-H.; Ohfune, Y. J. Am. Chem. Soc. 1994,
116, 7405–7406; (b) Horikawa, M.; Nakajima, T.;
Ohfune, Y. Synlett 1997, 253–254; (c) Ohfune, Y.;

1238
M. Kawasaki et al. / Tetrahedron Letters 44 (2003) 1235–1238
Nanba, K.; Takada, I.; Kan, T.; Horikawa, M.; Naka-
16. Khalil, E. M.; Subasinghe, N. L.; Johnson, R. L. Tetra-
hedron Lett. 1996, 37, 3441–3444.
17. General applications of the use of n-Bu4NOH for the Boc
protection of sterically hindered amines will be reported,
separately.
18. (a) Marshall, J. M.; Xie, S. J. Org. Chem. 1995, 60,
7230–7237; (b) Berry, J. M.; Hall, L. D. Carbohydr. Res.
1976, 47, 307–310.
jima, T. Chirality 1997, 9, 459–462; (d) Ohfune, Y.;
Horikawa, M. J. Synth. Org. Chem. Jpn. 1997, 55, 982–
993; (e) Namba, K.; Shinada, T.; Teramoto, T.; Ohfune,
Y. J. Am. Chem. Soc. 2000, 122, 10708–10709; (f) Iwama,
S.; Gao, W.-G.; Shinada, T.; Ohfune, Y. Synlett 2000,
1631–1633.
11. Shinada, T.; Kawakami, T.; Sakai, H.; Takada, I.;
19. Eckenberg, P.; Groth, U.; Huhn, T.; Richter, N.;
Schmeck, C. Tetrahedron 1993, 49, 1619–1624.
20. Compound 2: amorphous powder; [h]2D3 +27.0 (c 1.68,
Ohfune, Y. Tetrahedron Lett. 1998, 39, 3757–3760.
12.
L
-Valine was a superior acyloxy group to other amino
acids in the present substrate. The use of -phenylalanine
L
1
MeOH); H NMR (400 MHz, D2O): l 3.76 (d, J=10.5
gave a 10:1 mixture of (5S)- and (5R)-a-amino nitrile in
60% yield.
Hz, 1 H), 3.56 (d, J=10.5 Hz, 1 H), 3.36 (s, 3 H),
2.28–2.24 (m, 2 H), 1.93–2.05 (m, 2 H); 13C NMR (100
MHz, D2O): l 179.9, 174.4, 75.4, 65.3, 59.9, 31.4, 29.2.
Compound 3: amorphous powder; [h]2D4 −55.0 (c 0.70,
MeOH); 1H NMR (300 MHz, D2O): l 7.20–7.40 (m, 5
H), 4.47 (s, 2 H), 3.72 (d, J=10.5 Hz, 1 H), 3.52 (d,
J=10.5 Hz, 1 H), 2.05–2.25 (m, 2 H), 1.84 (t, J=7.3 Hz,
2 H); 13C NMR (75 MHz, D2O): l 181.4, 174.6, 162.1,
138.2, 129.2, 74.1, 72.9, 65.5, 32.4, 29.6.
13. An intermediary cyclic imine was isolated in a small
amount, presumably, because of elimination of HCN
from the products.
14. Namba, K.; Kawasaki, M.; Takada, I.; Iwama, S.; Izu-
mida, M.; Shinada, T.; Ohfune, Y. Tetrahedron Lett.
2001, 42, 3733–3736.
15. Hygroscopic amorphous powder: [h]1D9 −11.2 (c 0.51,
H2O). HCl salt of 1: [h]1D9 −1.29 (c 1.03, H2O).8