Stereoselective synthesis of the conformationally constrained glutamate analogue, (-)-(2R,3S)-cis-2-carboxyazetidine-3-acetic acid, from (S)-N-tosyl-2-phenylglycine

Article abstract of DOI:10.1055/s-2005-869866

The stereoselective synthesis of a novel cis conformationally constrained glutamate analogue containing an azetidine framework was accomplished from (S)-N-tosyl-2-phenylglycine in moderate overall yield. The key steps in the synthesis involved a N-H carbe

Full text of DOI:10.1055/s-2005-869866

LETTER
1559
Stereoselective Synthesis of the Conformationally Constrained Glutamate
Analogue, (–)-(2R,3S)-cis-2-Carboxyazetidine-3-acetic Acid, from
(S)-N-Tosyl-2-phenylglycine
Synthesis of the
C
o
n
nformationa
t
lly Con
o
strained Glu
n
tamate Anal
i
ogue o Carlos B. Burtoloso, Carlos Roque D. Correia*
Instituto de Química, Universidade Estadual de Campinas, UNICAMP, C.P. 6154, CEP. 13083-970, Campinas, São Paulo, Brazil
Fax +55(19)37883023; E-mail: roque@iqm.unicamp.br
Received 17 February 2005
no acids, the naturally occurring (S)-azetidine-2-carboxy-
lic acid (Figure 1) represents the prototype of an a-amino
acid containing an azetidine motif.
Abstract: The stereoselective synthesis of a novel cis conforma-
tionally constrained glutamate analogue containing an azetidine
framework was accomplished from (S)-N-tosyl-2-phenylglycine in
moderate overall yield. The key steps in the synthesis involved a N–
H carbenoid insertion promoted by Cu(acac)2, a very efficient Wit-
tig olefination of an azetidin-3-one, followed by a highly stereose-
lective rhodium-catalyzed hydrogenation. Epimerization of the cis
to the trans analogue was performed using DBU as base in toluene
at reflux.
O
OH
N
H
(S)-azetidine-2-carboxylic acid
Key words: azetidin-3-ones, glutamate analogues, N–H insertion,
conformationally constrained amino acids, Wittig reaction
Figure 1
Two illustrative examples of azetidine containing amino
acids displaying important biological activities are the
chiral glutamate analogues 1 and 2 (Figure 2).6a–6c
Glutamate 1 has been shown to act as an activator of the
metabotropic receptors, whereas analogue 2 appears to be
a potent agonist of the kainate receptor, as well as a potent
inhibitor of sodium-dependent glutamate uptake.
Glutamate and aspartate are the predominant excitatory
amino acid (EAAs) neurotransmitters in the mammalian
brain.1a,b These excitatory amino acids activate a family of
ligand-gated ion channels, called ionotropic receptors
(AMPA, KA and NMDA), and a family of receptors cou-
pled through GTP-binding proteins, called metabotropic
receptors, implicated in a variety of intracellular signaling
molecules.2a–2d EAA receptors participate in fast excitato-
ry transmission as well as in more complex signaling pro-
cesses, such as those required for synaptic plasticity and
higher cognitive functions.3a–3c In contrast to these normal
signaling pathways, excessive activation of the ionotropic
EAA receptors can trigger a cascade of events that even-
tually leads to neuronal death. This process, referred to as
excitotoxicity, is thought to be an underlying pathological
mechanism in a wide variety of neurological insults and
degenerative disorders, such as ischemia, trauma, hy-
poglycemia, epilepsy, Huntington’s and Parkinson’s dis-
eases.4a–4d
O
O
OH
HO
OH
N
HO
O
H
N
H
O
1
2
Figure 2 Chiral glutamate analogues displaying biological activity
In spite of several examples of the synthesis of azetidinic
amino acids available in the literature,6 the synthesis of
chiral glutamate analogues has been accomplished with
rather low stereoselectivity or has been restricted to the
synthesis of the trans-glutamate analogues.6a–6c,6i
In the last decades many research groups have been in-
volved with the synthesis of conformationally restricted
glutamate and aspartate analogues and the majority of the
glutamate analogues synthesized so far display a pyrroli-
dine ring as their rigid element.5 Four-membered rings as
conformational constraining elements are less common
and have been attracting considerable attention lately, es-
pecially for their increased rigidity and interesting physi-
ological activities exhibited by four-membered ring
containing amino acids.6 Regarding four-membered ami-
Our research group has been involved in the application of
azetidin-3-ones as key building blocks for the synthesis of
azetidines and azetidine alkaloids. As illustrated in
Scheme 1, chiral azetidin-3-ones can be readily prepared
from commercially available amino acids, which make
them potential intermediates for the synthesis of the cis-
and trans-glutamate analogues. Herein, we report a short
stereoselective synthesis of the (2R,3S) cis isomer of
glutamate 2 from (S)-N-tosyl-2-phenylazetidin-3-one (4,
Scheme 2), and a successful epimerization of the cis-
glutamate analogue to the trans analogue using DBU.
This approach extends the applicability of the metal-cata-
SYNLETT 2005, No. 10, pp 1559–1562
1
7
.0
6
.2
0
0
5
Advanced online publication: 07.06.2005
DOI: 10.1055/s-2005-869866; Art ID: S02205ST
© Georg Thieme Verlag Stuttgart · New York

1560
A. C. B. Burtoloso, C. R. D. Correia
LETTER
Table 1 Conditions Employed for the Reduction of Enoate 5
O
O
R
HO
O
OH
Conditiona
Yield of diester 6 (%) cis:trans
N
N
P
H2, Pd/C
19b
0
95:05
H
azetidin-3-one
cis-glutamate analogue
H2, Pt/C
R = CH2OP or Ph
isomerization
O
Et3SiH/ Wilkinson cat.
H2, Rh/C
80b
90c
88:12
92:08
O
HO
OH
OH
O
N
a Reductions were carried out at atmospheric pressure using a balloon
filled with hydrogen.
H
HNTs
trans-glutamate analogue
(S)-N-tosyl-2-phenylglycine
b Average of two experiments.
c Average of three experiments.
Scheme 1 Strategy to construct the cis- and trans-glutamate ana-
logues
Hydrogenation using platinum over carbon failed to pro-
vide the diester 6. Better results were obtained using rhod-
ium on carbon or Et3SiH in the presence of the Wilkinson
catalyst.13 Under these conditions the cis-ester 614 was ob-
tained in good yields (90% and 80%, Table 1) and in good
diastereoselectivities (92:8 and 88:12, cis:trans ratio, re-
spectively).
lyzed N–H insertion methodology commonly used for the
preparation of azetidin-3-ones.7
We started our synthesis with the preparation of chiral
azetidin-3-one 48 (Scheme 2). Based on a previous
protocol9 compound 4 was readily prepared in two steps
from (S)-N-tosyl-2-phenyl glycine in good overall yields.
The protocol involved the conversion of N-protected phe-
nyl glycine to the diazoketone 3 in 64% (Scheme 2), fol-
lowed by the reaction of 3 with Cu(acac)2 in reflux
benzene for just one minute to promote the N–H insertion
reaction in 50–55% yield.10
As planned, the phenyl ring of 6 was then converted to the
carboxylic acid by oxidation with RuCl3 hydrate and
NaIO415 followed by addition of diazomethane to furnish
the cis-diester 716 in moderate yields ranging from 40% to
54%. A small 3% epimerization was observed at this step,
as the cis:trans ratio dropped from 92:08 to 89:11 as ob-
served by GC (average of three experiments). Next, hy-
drolysis of the diester 7 with LiOH17 gave the N-protected
glutamate acid derivative 8 as a white solid (91% yield,
cis:trans = 89:11). The major cis-compounds 6, 7 and
even 8 could not be separated from their respective trans
isomers by column chromatography along the synthetic
pathway. Fortunately, after a careful recrystallization
(ethanol–hexane) of diacid 8 (cis:trans = 89:11) we were
The azetidin-3-one 4 was then converted to the enoate 5 in
quantitative yield by a Wittig olefination reaction11 using
the stabilized ylide carboethoxyethylidene-triphenylphos-
phorane. Catalytic hydrogenation of enoate 5 was carried
out under a variety of conditions aiming at optimization of
the stereoselectivity and yield of this key reduction step
(Table 1). Hydrogenation using palladium on carbon as
catalyst gave only 19% of the desired ester 6 in a dia-
stereomeric ratio of 95:05 for the cis:trans stereoisomers,
together with 81% of some unidentified material.12
Cu(acac)2, 10 mol%
O
1. ClCOCOCl,
CH2Cl2, DMF, 2 h
O
O
C6H6 reflux, 1 min
H
OH
N
2. THF, CH2N2
50–55%
Ts
HNTs
4
N2
3
HNTs
64%
(S)-N-tosyl-2-phenylglycine
[α]D +272.0 (c 1.05, CHCl3)
[α]D +112.4 (c 4.30, MeOH)
[α]D +219.5 (c 2.58, CHCl3)
PPh3CHCOOEt
C6H6, reflux, 1 h
H2, MeOH, C6H6
Rh/C 10 mol%, 24 h
EtO
EtO
1. NaIO4, RuCl3,
O
O
N
N
EtOAc, H2O,
MeCN, 3 h
6
5
Ts
Ts
quantitative (E+Z mixture)
90%
ds = 92:08 (GC)
2. CH2N2, Et2O
NOE = 3%
O
O
H
O
LiOH
THF:H2O
Na/C10H8
HO
H
H
N
HO
EtO
THF
OH
OH
OMe
dowex 50 H+
O
O
O
N
N
8
7
9
Ts
Ts
91%
40–54%
quantitative
ds = >98:02 (GC)
after recrystallization (EtOH/hexane)
ds = > 95:05 (1HNMR)
[α]D – 4.8 + 0.5 (c 0.3, H2O)
(ORD)
ds = 89:11 (GC)
[α]D +1.6 (c 2.44, THF)
Scheme 2 Synthesis of glutamate analogue 9
Synlett 2005, No. 10, 1559–1562 © Thieme Stuttgart · New York

LETTER
Synthesis of the Conformationally Constrained Glutamate Analogue
1561
Neurotransmission, In Psychopharmacology: The Fourth
Generation of Progress; Bloom, F. E.; Kupfer, D. J., Eds.;
Raven Press: New York, 1995, 75–85.
able to obtain this key compound in almost pure form (ds
> 98:02 after a single recrystallization).18
Finally, completion of the synthesis of the novel confor-
mational restricted cis-azetidine glutamate 9 was carried
out by N-deprotection of diacid 8 with Na/naphthalene in
quantitative yield, after purification on ion exchange resin
(Dowex 50 H+).19
(4) (a) Choi, D. W. Ann. Rev. Neurosci. 1990, 13, 171.
(b) Meldrum, B. S. Brain Pathol. 1993, 3, 405. (c) Choi, D.
W. Prog. Brain Res. 1994, 100, 47. (d) Rothman, S. M.;
Olney, J. W. Trends Neurosci. 1995, 18, 57.
(5) Chamberlin, A. R.; Bridges, R. J. Conformationally
Constrained Amino Acids as Probes of Glutamate Receptors
and Transporters, In Drug Design for Neuroscience;
Kozikowski, A. P., Ed.; Raven Press: New York, 1993, 231–
259.
(6) For the synthesis and studies of azetidinic a-amino acids,
see: (a) Kozikowski, A. P.; Tückmantel, W.; Reynolds, I. J.;
Wroblewski, T. J. J. Med. Chem. 1990, 33, 1561.
(b) Kozikowski, A. P.; Tückmantel, W.; Liao, Y.;
Wroblewski, T. J.; Wang, S.; Pshenichkin, S.; Surin, A.;
Thomsen, C. Bioorg. Med. Chem. Lett. 1996, 6, 2559.
(c) Kozikowski, A. P.; Tückmantel, W.; Liao, Y.;
Wroblewski, T. J.; Manev, H.; Ikonomovic, S. J. Med.
Chem. 1993, 36, 2706. (d) Bridges, R. J.; Lovering, F. E.;
Humphrey, J. M.; Stanley, M. S.; Blakely, T. N.; Cristofaro,
M. F.; Chamberlin, R. Bioorg. Med. Chem. Lett. 1993, 3,
115. (e) Arakawa, Y.; Murakami, T.; Arakawa, Y.;
Yoshifugi, S. Chem. Pharm. Bull. 2003, 51, 96. (f) De
Kimpe, N.; Boeykens, M. Tetrahedron 1998, 54, 2619.
(g) Hanessian, S.; Bernstein, N.; Yang, R.; Maguire, R.
Bioorg. Med. Chem. Lett. 1999, 9, 1437. (h) Couty, F.;
Evano, G.; Rabasso, N. Tetrahedron: Asymmetry 2003, 14,
2407. (i) Couty, F.; Carlin-Sinclair, A.; Rabasso, N. Synlett
2003, 726.
After completion of the synthesis of the cis-glutamate an-
alogue 9, we have also examined the conversion of the
cis-diester 10 (prepared from diacid 8 with diazomethane)
to its trans stereoisomer. The use of bases such as LH-
MDS, KHMDS, t-BuOK, proton sponge, Me2NH20 and
pyridine led to no epimerization21 at C2 or led to decom-
position of the diester 10. However, reaction of diester 10
with DBU (10 equiv) in toluene at reflux for seven hours
provided a diastereomeric mixture of diester 10 and 11 in
a 20:80 (GC) ratio as described in Scheme 3. Attempts to
carry out this epimerization step beyond the cis:trans ratio
of 20:80 were fruitless.
NOE = 0.8%
O
O
H
H
MeO
MeO
OMe
OMe
O
DBU, 10 equiv
PhMe, reflux, 7 h
N
11
O
N
10
Ts
Ts
cis:trans = 20:80
cis:trans >98:2
(7) For a review on azetidin-3-ones, see: Dejaegher, Y.;
Kuz’nenok, N. M.; Zvonok, A. M.; De Kimpe, N. Chem.
Rev. 2002, 102, 29.
(8) Compound 4 was also prepared by Pussino: Pusino, A.;
Saba, A.; Desole, G. Gazz. Chim. Ital. 1985, 115, 33.
(9) Burtoloso, A. C. B.; Correia, C. R. D. Tetrahedron Lett.
2004, 45, 3355.
(10) For the synthesis of N-tosyl-azetidin-3-ones employing
Cu(acac)2, see: Wang, J.; Hou, Y.; Wu, P. J. Chem. Soc.,
Perkin Trans. 1 1999, 2277.
(11) For previous applications of the Wittig olefination of
azetidin-3-ones, see: (a) Hanessian, S.; Fu, J.; Chiara, J. L.;
Di Fabio, R. Tetrahedron Lett. 1993, 34, 4157. (b) Podlech,
J.; Seebach, D. Helv. Chim. Acta 1995, 78, 1238.
(c) Emmer, G. Tetrahedron 1992, 48, 7165.
Scheme 3 Epimerization of 10 with DBU
In summary, we have accomplished for the first time the
stereoselective synthesis of the novel cis-glutamate ana-
logue 9 containing an azetidine nucleus in seven steps in
15% yield from the chiral (S)-N-tosyl-phenylglycine.
Epimerization of the cis-glutamate analogue 10 with DBU
allowed the synthesis of the trans-glutamate analogue 11
with good diastereoselectivity. The synthesis of other con-
strained azetidine glutamates and aspartates will be re-
ported in due course.
Acknowledgment
(12) Inseparable mixture of compounds by column
chromatography.
We thank FAPESP (Research Supporting Foundation of the State of
São Paulo) for financial support and a student fellowship and Ange-
lo H. L. Machado (Unicamp) for useful suggestions during the de-
velopment of this work.
(13) Liu, H.; Ramani, B. Synth. Commun. 1985, 15, 965.
(14) The cis compound 6 could not be separated from its trans
isomer by column chromatography.
(15) Matsuura, F.; Hamada, Y.; Shioiri, T. Tetrahedron Lett.
1992, 33, 7921.
(16) Compound 7 was obtained as an inseparable mixture of the
cis and trans stereoisomers. The ratio of these compounds
was determined by 1H NMR and GC.
References
(1) (a) Kanai, Y.; Smith, C. P.; Hediger, M. A. Trends Neurosci.
1993, 16, 365. (b) Szatkowski, M.; Attwell, D. Trends
Neurosci. 1994, 17, 359.
(17) Clayden, J.; Menet, C. J.; Tchabanenko, K. Tetrahedron
2002, 58, 4727.
(2) (a) Conn, P. J.; Patel, J. In The Metabotropic Glutamate
Receptors; Humana Press: Totowa / New Jersey, 1994, 1–
277. (b) Hollmann, M.; Heinemann, S. Annu. Rev. Neurosci.
1994, 17, 31. (c) Nakanishi, S. Neuron 1994, 13, 1031.
(d) Nicolletti, F.; Bruno, V.; Copani, A.; Casabona, G.;
Knöpfel, T. Trends Neurosci. 1996, 19, 267.
(3) (a) Daw, N. W.; Stein, P. S.; Fox, K. A. Rev. Neurosci. 1993,
16, 207. (b) Collingridge, G. L.; Bliss, T. V. P. Trends
Neurosci. 1995, 18, 54. (c) Cotman, C. W.; Kahle, J. S.;
Miller, S. E.; Ulas, J.; Bridges, R. J. Excitatory Amino Acid
(18) Mp 201–202 °C (dec). IR: 3300–2500, 1698, 1437, 1346,
1229, 1161, 943 cm–1. 1H NMR (300 MHz, acetone-d6):
d = 7.79 (d, J = 8.8 Hz, 2 H), 7.49 (d, J = 8.8 Hz, 2 H), 4.62
(d, J = 9.5 Hz, 1 H), 3.90 (t, J = 8.1 Hz, 1 H), 3.50 (dd,
J = 8.1, 4.4 Hz, 1 H), 2.97 (m, 1 H), 2.67 (d, J = 8.1 Hz, 2 H),
2.46 (s, 3 H). 13C NMR (75 MHz, acetone-d6): d = 172.5,
169.6, 145.1, 134.1, 130.8, 129.0, 63.8, 54.2, 34.5, 28.4,
21.6.
Synlett 2005, No. 10, 1559–1562 © Thieme Stuttgart · New York

1562
A. C. B. Burtoloso, C. R. D. Correia
LETTER
[M + 1], 114, 97, 96, 84, 78, 68. HRMS: m/z calcd for
C6H9NO4: 159.05316; found: 159.06178.
(20) Alcaide, B.; Aly, M. F.; Rodríguez-Vicente, A. Tetrahedron
Lett. 1998, 39, 5865.
(19) IR: 3077, 1679, 1625, 1574, 1421, 1184, 1129, 974, 801, 723
cm–1. 1H NMR (300 MHz, D2O): d = 4.90 (d, J = 9.5 Hz, 1
H), 4.30 (dd, J = 10.9, 8.8 Hz, 1 H), 3.77 (dd, J = 10.9, 6.6
Hz, 1 H), 3.40 (m, 1 H), 2.60 (dd, J = 16.1, 5.1 Hz, 1 H), 2.44
(dd, J = 16.1, 11.7 Hz, 1 H). 13C NMR (75 MHz, D2O):
d = 176.0, 170.6, 61.8, 48.5, 34.5, 30.5. ESI-MS: m/z = 160
(21) In some occasions, a small amount of epimerization at C2
was observed.
Synlett 2005, No. 10, 1559–1562 © Thieme Stuttgart · New York