One-pot synthesis of N-Cbz-l-BMAA and derivatives from N-Cbz-l-serine

Article abstract of DOI:10.1016/j.tetlet.2007.01.149

We report herein an asymmetric synthesis of the modified amino acid N-Cbz-l-BMAA and seven of its alkyl derivatives (2a-h) from N-Cbz-l-serine via ring-opening of the β-lactone (formed under modified Mitsunobu conditions) by different amines. This procedu

Full text of DOI:10.1016/j.tetlet.2007.01.149

Tetrahedron Letters 48 (2007) 2325–2327
One-pot synthesis of N-Cbz-L-BMAA and derivatives
from N-Cbz-L-serine
Sidnei Moura and Ernani Pinto*
ˆ
ˆ
Departamento de Ana´lises Clı´nicas e Toxicolo´gicas, Faculdade de Ciencias Farmaceuticas, Universidade de Sa˜o Paulo,
Av. Prof. Lineu Prestes, 580, Sa˜o Paulo 05505900, Brazil
Received 20 November 2006; revised 25 January 2007; accepted 26 January 2007
Available online 1 February 2007
Abstract—We report herein an asymmetric synthesis of the modified amino acid N-Cbz-L-BMAA and seven of its alkyl derivatives
(2a–h) from N-Cbz-L-serine via ring-opening of the b-lactone (formed under modified Mitsunobu conditions) by different amines.
This procedure is simple, one-pot and can generate various derivatives that can be investigated for their toxicological effects. In addi-
tion, it can be employed to produce analytical standards for water monitoring as well as labeled compounds for biotransformation
studies. This toxin has been the focus of serious ecological and public concern since its implication in degenerative disease such as
Alzheimer and Parkinsonism dementia.
Ó 2007 Elsevier Ltd. All rights reserved.
The modified amino acid b-N-methylamino-L-alanine
(L-BMAA) plays an important role in degenerative dis-
eases.1 The neurotoxicity of L-BMAA is due to an exci-
totoxic mechanism, involving elevated intracellular
calcium and bicarbonate levels.2 This amino acid is also
a potent glutamate receptor agonist.3 Its mechanism of
action seems to be mediated, in part, by the activation
of N-methyl-D-aspartate (NMDA) receptors.4 This non-
protein amino acid is neurotoxic and is produced as a
secondary metabolite by seeds of the indigenous Guam
cycad (Cycas micronesica).5 This effect was evidenced
by one study with Chamorro natives of Guam that com-
pared the increase of incidence rates of ALS/PDC
(amyotrophic lateral sclerosis/Parkinsonism dementia
complex) with the consumption of seeds.6,7 More re-
cently, it was reported that cyanobacteria of the genus
Nostoc, which are a root symbiont of cycads, also have
the ability to biosynthesize this important metabolite.7
Human poisonings from cyanobacterial blooms and
their toxins (i.e., microcystins, saxitoxins, cylindrosper-
mopsin and others) are well known.8,9
amines for the cleavage of lactones in order to obtain
the N-Cbz-b-amino-L-alanine. In this way, they guaran-
teed the production of the desired optically pure form.
The aim of this work is the asymmetric synthesis of N-
Cbz-L-a-amino-b-methylaminopropionic acid and seven
of its other N-alkyl derivatives from N-Cbz-L-serine,
with low synthetic cost and significant yields. The retro-
synthesis of N-Cbz-L-BMAA is proposed in Scheme 1.
In order to do this, we use direct addition of amines
to open lactones formed under Mitsonubu conditions
(Scheme 2). The synthesis of this molecule is very impor-
tant for further toxicological investigation and for the
production of analytical standards.
Some derivatives and labeled analogues can also be used
for the elucidation of its mechanism of action, which is
not yet completely understood.
General methods for the synthesis of b-substituted ami-
no acids are relatively difficult because of stereochemical
The first synthesis reported for this molecule produced a
racemic mixture via the methylation of a-acetoamino-b-
methylaminopropionic acid, followed by hydrolysis.10
Later, Ratemi and Vederas11 employed trimethylsilyl-
O
O
HO
N
H
HO
OH
NHCbz
-serine
NHCbz
N
-Cbz-
L
-BMAA
N-Cbz-L
*
Corresponding author. Tel.: +55 11 30911505; fax: +55 11
30319055; e-mail: ernani@usp.br
Scheme 1. Retrosynthetic analyses of N-Cbz-L-BMAA.
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.01.149

2326
S. Moura, E. Pinto / Tetrahedron Letters 48 (2007) 2325–2327
O
O
NHCbz NHR1R2
68-81%
O
Ph3P, DEAD
THF, -78 oC to r.t.
R1
HO
CbzHN
N
HO
OH
NHCbz
O
R2
R1 R2
R1
R2
Cbz = COOCH2Ph
2 e
2 f
iso-Prop
ter t-But
Me
H
H
2 a
2 b
2 c
2 d
H
H
H
H
Me
Et
2 g Me
n
-Prop
2 h
Et
Et
n
-But
Scheme 2. Synthesis of N-Cbz-L-BMAA and derivatives.
control, poor yields or the necessity of numerous steps.
For instance, to produce the mono-methylated amine N-
Boc-L-BMAA as an intermediate for the total synthesis
of target drugs, Gopin et al.12 utilized DAP (Boc-L-2,4-
diaminopropionic acid) as starting material in a three
steps synthetic procedure with low overall yields.
In our hands, bubbling gaseous alkylamine derivatives
through the reaction mixture leads only to the amine
derivatives 2a–h under one-pot conditions.17 The experi-
mental conditions utilized were similar to those of
Ratemi and Vederas11 (THF as a solvent and low
temperature for the addition of amine, 0 °C).
Several alternative procedures for the synthesis of mod-
ified amino acids have been proposed in the last years.
By using L-serine, Chhabra et al.13 reported one efficient
method via oxazolidine formation, a complex synthetic
process leading to N-protected amino acids, but without
stereoselective control, after oxidation by Jones reagent.
In another method, the transformation of L-asparagine
using iodobenzene diacetate produced via Hoffmann
rearrangement14 the product N-Cbz-3-dimethylamino-
L-alanine. However, the use of costly reagents and the
complex alkylation of primary amines, which may lead
to isomer formation due to the nonselective alkylation,
make this method complicated.
The great advantage of the procedure developed here is
the retention of the serine stereochemistry, which was
solvent dependent in the work of Ratemi and Vederas.11
Moreover, our strategy joined the functionality of a one-
pot procedure for product formation with mild condi-
tions, leading to goods yields, from 68% to 81%. In
the case of 2a, our method showed a 78% of yield
whereas in a previous report, which uses two steps, it
was 70%11 for purified molecules. Our approach reduces
common problems present in other methodologies (i.e.,
low stability) that use lactones from serine as intermedi-
ates.11,15–17 This efficient method is appropriate for
obtaining N-Cbz-L-BMAA, 2a, and generates the
cyanotoxin L-BMAA. The production of this compound
is very important for toxicological assays as well as an
analytical standard in water monitoring. The analogues
of N-Cbz-L-BMAA, 2b–h, produced by using different
alkylating agents in gaseous form, can be prepared for
assay in toxicological experiments. In conclusion, our
method is more efficient than previous published routes
that use Mitsunobu lactone intermediate.11,15–17 In addi-
tion, these compounds can be easily used as intermedi-
ates in other procedures of total synthesis. All
products were identified by 1H and 13C NMR and
ESI-MSn.18
The use of b-lactones derived from serine in organic syn-
thesis is functional and widely employed, mainly in the
production of modified amino acids used as intermedi-
ate in synthesis. Venderas et al. confirmed the applica-
tion of lactones to produce modified amino acids.11,15
More recently, Schneider et al.16 also applied this proce-
dure, through a lactone intermediate, to obtain selenium
and tellurium substituted amino acids. Nevertheless, the
formation of this lactone depends on controlled condi-
tions, which can complicate the execution of this proce-
dure. Another difficulty with the traditional Mitsunobo
conditions is the purification, since the product polarity
and that of diethyl hydrazodicarboxylate (formed from
DEAD) are almost the same. These problems can be
solved by a simple procedure; make this reaction a
one-pot process by adding to the same flask the nucleo-
philic agent after the formation of the lactone intermedi-
ate. With this one-pot procedure, it is possible to directly
produce the desired compound that is easier to isolate
and purify afterwards. In addition, Arnol et al. had
already used ammonia to open these lactones15 and
described the use of powerful nucleophiles such as
methoxide and ammonia to react with carbonyl groups.
However, by simply changing solvent composition, it is
possible to invert the proportion of the formation of
amide/amine (in the case of ammonia). Other authors
had also discussed and applied different strategies to
obtain only amino acids,11 where trimethylsilylamines
were employed as nucleophilic agents, well known to
be weaker nucleophiles.
Acknowledgements
The authors thank FAPESP, CAPES and CNPq for
financial support.
References and notes
1. (a) Montine, T. J.; Li, K.; Perl, D. P.; Galasko, D.
Neurology 2005, 65, 768; (b) Banack, S. A.; Cox, P. A.
Neurology 2003, 61, 387.
2. Brownson, D. M.; Mabry, T. J.; Leslie, S. W. J.
Ethnopharmacol. 2002, 82, 159.
3. Andersson, H.; Lindqvist, E.; Olson, L. Neuroreport 1997,
8, 1813.
4. (a) Spencer, P. S.; Nunn, P. B.; Hugon, J.; Ludolph, A. C.;
Ross, S. M.; Roy, D. N.; Robertson, C. Science 1987; (b)

S. Moura, E. Pinto / Tetrahedron Letters 48 (2007) 2325–2327
2327
Copani, A.; Canonico, P. L.; Catania, M. V.; Aronica, E.;
Bruno, V.; Ratti, E.; Van Amsterdam, F. T. M.;
Gaviraghi, G.; Nicoletti, F. Brain Res. 1991, 558, 79–86,
237, 517–522.
13.6, 22.5, 54.2, 61.8, 66.4, 127.1 (2C), 127.4, 127.6 (2C),
134.9, 155.8, 169.8; ESI-MS 289 [M+Na]+, 266.9
[M+H]+, 222.9 [MÀCOOH+H]+, 91.3 [PhCH2]+.
N-(Benzyloxycarbonyl)-b-n-propylamine-L-alanine
(2c):
5. (a) Vega, A.; Bell, E. A. Phytochemistry 1967, 6, 759–762;
(b) Oh, C. H.; Brownson, D. M.; Mabry, T. J. Planta
medica 1995, 61, 66.
6. (a) Spencer, P. S.; Nunn, P. B.; Hugon, J.; Ludolph, A.;
Roy, D. N. Lancet 1986, 1, 965; (b) Spencer, P. S.; Nunn,
P. B.; Hugon, J.; Ludoph, A. C.; Ross, S. M.; Roy, D. N.;
Robertson, R. C. Science 1987, 237, 517–522.
7. Cox, P. A.; Banack, S. A.; Murch, S. J. Proc. Natl. Acad.
Sci. 2003, 100, 13380.
Compound 2c was isolated with 81% of yield; mp
114 °C; H NMR (300 MHz, CDCl3, ppm) d 0.82 (t, 3H,
1
J = 7.38, NCH2CH2CH3), 1.23 (sex, 2H, J = 7.68,
NCH2CH2CH3), 3.14 (q, 2H, J = 7.23, NCH2CH2CH3),
3.60 (dd, 1H, J1 = 6.31, J2 = 4.56), 4.05 (dd, 1H,
J1 = 11.43, J2 = 3.87), 4.11 (dd, 1H, J = 3.18), 5.07 (s,
2H, CH2Ph), 7.29 (m, 5H, Ph); 13C NMR (75 MHz,
CDCl3, ppm) d 11.2, 22.6, 41.2, 55.1, 62.8, 67.4, 128.1
(2C), 128.4, 128.6 (2C), 135.9, 156.8, 170.9; ESI-MS 303
[M+Na]+, 280.9 [M+H]+, 263 [MÀOH+H]+, 236.9
[MÀCOOH+H]+, 91.3 [PhCH2]+.
8. Kuiper-Goodman, T.; Falconer, I.; Fitzgerald, J. In Toxic
Cyanobacteira in Water; Chorus, I., Bartram, J., Eds.;
E&F.N. Spon: London, 1999; pp 113–153.
N-(Benzyloxycarbonyl)-b-n-butylamine-L-alanine
(2d):
9. (a) Pouria, S.; de Andrade, A.; Barbosa, J.; Cavalcanti, R.
L.; Barreto, V. T. S.; Ward, C. J.; Preiser, W.; Poon, G.
K.; Neild, G. H.; Codd, G. A. Lancet 1998, 352, 21–26; (b)
Carmichael, W. W. Sci. Am. 1994, 270, 78–86; (c) Codd,
G.; Bell, S. G.; Kaya, K.; Ward, C. J.; Beattie, K. A.;
Metcalf, J. S. Eur. J. Phycol. 1999, 34, 405–415.
10. Vega, A.; Bell, E. A.; Nunn, P. B. Phytochemistry 1986, 7,
1885–1887.
11. Ratemi, E. S.; Vederas, J. C. Tetrahedron Lett. 1994, 35,
7605–7608.
12. Gopin, A.; Raderb, C.; Shabat, Doron. Bioorg. Med.
Chem. 2004, 12, 1853–1858.
13. Chhabra, S. R.; Mahajan, A.; Chan, W. C. J. Org. Chem.
2002, 67, 4017–4029.
14. Andruszkiewicz, R.; Walkowiak, A. Org. Preparations
Procedures Int. 2001, 33, 379–381.
15. Arnol, L. D.; Kalantar, T. H.; Vederas, J. C. J. Am. Chem.
Soc. 1985, 107, 7105–7109.
16. Schneider, A.; Rodrigues, O. E. D.; Paixao, M. W.;
˜
Compound 2d was isolated with 81% of yield; mp 86 °C;
1H NMR (300 MHz, CDCl3, ppm) d 0.82 (t, 3H, J = 7.36,
NCH2CH2CH2CH3),
1.31
(sex,
2H,
J = 7.32,
NCH2CH2CH2CH3), 1.38 (qui, 2H, J = 7.23, NCH2-
CH2CH2CH3), 3.04 (sex, 2H, J = 7.05, NCH2-
CH2CH2CH3), 3.56 (dd, 1H, J1 = 6.33, J2 = 4.50), 3.98
(dd, 1H, J1 = 11.20, J2 = 3.92), 4.10 (dd, 1H, J = 3.18),
5.05 (s, 2H, CH2Ph), 7.28 (m, 5H, Ph); 13C NMR
(75 MHz, CDCl3, ppm) d 11.2, 20.1, 28.6, 41.4, 54.1,
62.6, 67.2, 128.4 (2C), 128.5, 128.6 (2C), 136.2, 156.9,
171.2; ESI-MS 317 [M+Na]+, 294.9 [M+H]+, 250.9
[MÀCOOH+H]+, 91.3 [PhCH2]+.
N-(Benzyloxycarbonyl)-b-iso-propylamine-L-alanine (2e):
Compound 2e was isolated with 71% of yield; mp 96 °C;
1H NMR (300 MHz, CDCl3, ppm) d 1.02 (d, 6H, J = 6.63,
NCH(CH3)2), 4.00 (sep, 1H, J = 6.84, NCH2(CH3)2), 3.57
(dd, 1H, J1 = 5.97, J2 = 5.04), 4.04 (dd, 1H, J1 = 11.43,
J2 = 3.87), 4.06 (dd, 1H, J = 3.51), 5.21 (s, 2H, CH2Ph),
7.28 (m, 5H, Ph); 13C NMR (75 MHz, CDCl3, ppm) d 22.5
(2C), 41.6, 55.4, 62.8, 67.2, 128.0 (2C), 128.2, 128.4 (2C),
136.1, 157.1, 170.0; ESI-Ms 303 [M+Na]+, 280.9 [M+H]+,
263 [MÀOH+H]+, 236.9 [MÀCOOH+H]+, 91.3
[PhCH2]+.
Appelt, H. R.; Braga, A. L.; Wessjohann, L. A. Tetra-
hedron Lett. 2006, 47, 1019–1021.
17. N-(Benzyloxycarbonyl)-b-alkylamine-L-alanine
(2a–h)
general procedure: To a stirred solution of dried Ph3P
(6.43 g, 24.5 mmol) in 100 mL of anhydrous THF at
À78 °C was added distilled diethyl azodicarboxylate
(2.70 mL, 3.58 g, 24.5 mmol) dropwise over 10 min. After
10 min, a solution of dried L-serine (5.84 g, 24.4 mmol) in
100 mL of THF was added dropwise over 15 min to the
stirred slurry at À78 °C. The mixture was stirred for
20 min at À78 °C and then for 2.5 h at 20 °C.14 Subse-
quently, dry amine NHR1R2 was bubbled (ꢀ0.3 mL/min)
through this solution stirred at 0 °C for 30 min. After-
ward, the solvent was removed by vacuum, and the residue
was chromatographed on silica gel. The product was
eluted after other compounds with MeOH. The product
could be recrystallized with MeOH/Et2O.
N-(Benzyloxycarbonyl)-b-tert-butylamine-L-alanine (2f):
Compound 2f was isolated with 69% of yield; mp
1
104 °C; H NMR (300 MHz, CDCl3, ppm) d 1.20 (s, 9H,
NC(CH3)3), 3.58 (dd, 1H, J1 = 6.01, J2 = 5.04), 4.11 (dd,
1H, J1 = 11.43, J2 = 3.87), 4.22 (dd, 1H, J = 3.51), 5.04 (s,
2H, CH2Ph), 7.22 (m, 5H, Ph); 13C NMR (75 MHz,
CDCl3, ppm) d 27.6 (3C), 51.9, 56.9, 66.7, 66.9, 128.0 (2C),
128.1, 128.5 (2C), 136.3, 156.4, 170.1; ESI-MS 317
[M+Na]+, 295.1 [M+H]+, 238.9 [MÀC(CH3)3+H]+, 91.3
[PhCH2]+.
N-(Benzyloxycarbonyl)-b-dimethylamine-L-alanine (2g):
Compound 2g was isolated with 68% of yield; mp
1
103 °C; H NMR (300 MHz, CDCl3, ppm) d 2.97 (s, 3H,
18. N-(Benzyloxycarbonyl)-b-methylamine-L-alanine
(2a):
N(CH3)2), 3.03 (s, 3H, N(CH3)2), 3.69 (m, 2H), 4.66 (dd,
1H, J = 3.66), 5.02 (s, 2H, CH2Ph), 7.25 (m, 5H, Ph); 13C
NMR (75 MHz, CDCl3, ppm) d 35.9, 37.3, 52.1, 63.9,
67.2, 128.0 (2C), 128.2, 128.5 (2C), 136.1, 156.4, 170.5;
ESI-MS 289 [M+Na]+, 266.9 [M+H]+, 222.9
[MÀCOOH+H]+, 91.3 [PhCH2]+.
Compound 2a was isolated with 78% of yield; mp 54 °C;
1H NMR (300 MHz, CDCl3, ppm) d 2.70 (s, 3H, NCH3),
3.58 (dd, 1H, J1 = 6.25, J2 = 4.65), 4.00 (dd, 1H,
J1 = 11.47, J2 = 3.87), 4.11 (dd, J = 3.20, 1H), 5.03 (s,
2H, CH2Ph), 7.38–7.21 (m, 5H, Ph); 13C NMR (75 MHz,
CDCl3, ppm)
d
25.3, 54.3, 61.7, 66.4, 127.2 (2C),
N-(Benzyloxycarbonyl)-b-diethylamine-L-alanine
(2h):
127.4, 127.6 (2C), 134.9, 155.8, 170.6; ESI-MS 275
[M+Na]+, 252.9 [M+H]+, 208.9 [MÀCOOH+H]+, 91.3
[PhCH2]+.
Compound 2h was isolated with 70% of yield; mp 92 °C;
1H NMR (300 MHz, CDCl3, ppm) d 1.02 (t, 3H, J = 7.02,
N(CH2)2(CH3)2), 1.08 (t, 3H, J = 7.12, N(CH2)2(CH3)2),
2.41 (q, 2H, J = 7.06, N(CH2)2(CH3)2), 2.52 (q, 2H,
J = 7.04, N(CH2)2(CH3)2), 3.72 (m, 2H), 4.68 (dd, 1H,
J = 3.68), 5.00 (s, 2H, CH2Ph), 7.27 (m, 5H, Ph); 13C
NMR (75 MHz, CDCl3, ppm) d 13.3, 13.7, 40.9, 41.6,
53.1, 64.6, 67.1, 128.3 (2C), 128.4, 128.6 (2C), 136.4, 156.3,
171.5; ESI-Ms 317 [M+Na]+, 294.9 [M+H]+, 280.9
[MÀCOOH+H]+, 91.3 [PhCH2]+.
N-(Benzyloxycarbonyl)-b-ethylamine-L-alanine (2b): Com-
1
pound 2b was isolated with 76% of yield; mp 128 °C; H
NMR (300 MHz, CDCl3, ppm) d 1.05 (t, 3H, J = 7.08,
NCH2CH3), 3.21 (q, t, 2H, J = 7.08, NCH2CH3), 3.58 (dd,
1H, J1 = 6.31, J2 = 4.56), 4.05 (dd, 1H, J1 = 11.53,
J2 = 3.85), 4.11 (dd, J = 3.25, 1H), 5.06 (s, 2H, CH2Ph),
7.28 (m, 5H, Ph); 13C NMR (75 MHz, CDCl3, ppm) d