Synthesis of Z-α,β-ethylenic N-boc-γ-amino esters from nitrones

Article abstract of DOI:10.1055/s-1999-2672

The reduction of γ-N-hydroxylamino-α,β-acetylenic esters bearing a methoxy-substituted benzyl group on the nitrogen atom by zinc in a MeOH/AcOH mixture leads stereoselectively to Z-γ-amino-α,β-ethylenic esters.

Full text of DOI:10.1055/s-1999-2672

602
LETTER
Synthesis of Z-a,b-Ethylenic N-Boc-g-Amino Esters from Nitrones
Christelle Dagoneau, Jean-Noël Denis, Yannick Vallée*
LEDSS, UMR CNRS-Université Joseph Fourier, B.P. 53X, 38041 Grenoble, France
Fax +33 (4) 76514382, E-mail yannick.vallee@ujf-grenoble.fr
Received 23 February 1999
responding oximes by NaBH3CN in acidic medium and
condensed them with propanal, 2-methylpropanal and 3-
methylbutanal to give the nitrones 2a-f in good yields.5,6
When these nitrones were condensed with t-butyl 3-lithi-
opropiolate, the expected hydroxylamines 3a-f were iso-
lated in yields ranging from 59 to 81% (Table). The
condensations of 2e and 2f with ethyl 3-lithiopropiolate
were also attempted. However, the yields being lower (38
and 58% respectively) than in the previous examples, the
following reactions were all effected with the t-butyl es-
ters.
Abstract: The reduction of g-N-hydroxylamino-a,b-acetylenic es-
ters bearing a methoxy-substituted benzyl group on the nitrogen
atom by zinc in a MeOH/AcOH mixture leads stereoselectively to
Z-g-amino-a,b-ethylenic esters.
Key words: nitrones, 3-lithiopropiolates, unsaturated g-amino es-
ters
We have recently reported a synthesis of g-amino-a,b-un-
saturated esters starting from nitrones and propiolates.1
The first step was the condensation of the lithium anion of
propiolates with aliphatic N-benzylnitrones leading to
acetylenic esters (Scheme 1).2 The following step was ef-
fected using zinc in an acidic medium as the reducing
agent. This method allows the one pot reduction of both
the N-hydroxyl group and the C∫C triple bond. However,
this strategy suffers from a lack of stereoselectivity. In
many cases mixtures of E ethylenic esters and lactams, re-
sulting from the cyclisation of the Z esters, were obtained.
The Z non-cyclized alkene was detected in one case. Me-
rino et al. proposed an interesting two-step alternative to
our strategy : (i) hydrogenation of the C∫C triple bond
with H2/Lindlar catalyst and (ii) reduction of the hydroxyl
group by TiCl3.3 This way, they were able to produce se-
lectively the g-lactams. In this communication, we would
like to report a one-pot strategy towards non-cyclized Z-
g-amino-a,b-ethylenic esters.
Treatmentb of the obtained acetylenic hydroxylamines
3a-f with zinc in a methanol-acetic acid mixture1 led, as
expected, to the cleavage of the N-O bond and to the hy-
drogenation of the triple bond into a double bond. Howev-
er, we were surprised to notice that the isolated products
were the Z-g-amino-a,b-ethylenic esters 4a-f.6 In no case
were the E-esters detected and only in the case of the
monomethoxybenzyl derivative 3d, the lactam resulting
from the possible cyclisation of the Z-amino ester was ob-
served as a by-product (ca. 5%). With all the tested
dimethoxybenzyl derivatives, the 1H NMR spectra of the
crude product, recorded immediately after the work up of
the reactions, showed only the Z-esters. However, these Z-
esters were rather poorly stable. Upon standing for longer
time in the NMR tubes and/or upon purification over sili-
ca gel, they were partly transformed into their E-isomers
(maybe by a reversible [1,4]-addition process) and the
corresponding lactams. Thus we found it more convenient
to mask the nucleophilicity of the nitrogen atom by treat-
ing the crude amines with t-butoxycarbonyl anhydride.
The amines were then isolated as their Boc-protected de-
rivatives 5a-f. The global yields from the acetylenic hy-
droxylamino esters 3 to the N-Boc-protected ethylenic
amino esters 5 ranged from 51 to 66% (Table 1).6
The fact that the Z-compounds 4 did not cyclize under the
reduction conditions was rather surprising. The question
was: why did the benzylamines give the lactams, when the
methoxybenzylamines gave the non-cyclized products?
Indeed, at first glance, we predicted that they should cyc-
lize more readily because of the electron-donating proper-
ties of the methoxy groups which enhance the nucleo-phi-
licity of the amine as compared to the simple benzyl
derivatives. The solution of the problem is probably not to
be found in the more or less nucleophilic properties of the
amine, but in its greater basicity. We propose that the
more basic methoxybenzylamines are more efficiently
protonated than the benzylamines under the reaction con-
ditions, thus preventing their reaction with the ester group.
Scheme 1
Looking for an easily cleavable protective group for the
obtained amine function, we were attracted by various
methoxy-substituted benzyl groups.4 In these cases, the
removal of the protection can be obtained without making
use of hydrogenation, a method which can be inconve-
nient in the presence of a C = C double bond. We prepared
the hydroxylamines 1 (Scheme 2) by reduction of the cor-
Synlett 1999, No. 5, 602–604 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Synthesis of Z-a,b-Ethylenic N-Boc-g-Amino Esters from Nitrones
603
References and Notes
(1) Denis, J.-N.; Tchertchian, S.; Tomassini, A.; Vallée, Y.
Tetrahedron Lett. 1997, 38, 5503.
(2) Concerning the reaction of nitrones with LiC∫CSiMe3, see :
Merino, P.; Anoro, S.; Castillo, E.; Merchan, F.; Tejero, T.
Tetrahedron : Asym. 1996, 7, 1887.
(3) Merino, P.; Castillo, E.; Franco, S.; Merchan, F. L.; Tejero, T.
Tetrahedron : Asym. 1998, 9, 1759. Merchàn, F. L.; Merino,
P.; Tejero, T. Electronic Conference on Heterocyclic
Chemistry (ECHET-96); Internet : http://www.ch.ic.ac.uk/
ectoc/echet96/. Rzepa, H. S.; Snyder, J. P.; Leach, C. Eds.
(4) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis; J. Wiley & Sons Inc.: New York, 1991; Kocienski,
P. J. Protecting Groups; G. Thieme Verlag: Stuttgart, 1994.
(5) Dondoni, A.; Franco, S.; Junquera, F. L.; Merchàn, F.;
Merino, P.; Tejero, T. Synth. Commun. 1994, 24, 2537.
(6) All new compounds gave spectroscopic and analytical data in
agreement with the assigned structures. Typical experimental
procedures and spectroscopic data. Synthesis of nitrone 2c:
Anhydrous MgSO4 and 2-methylpropanal (8.2mmol) were
added to a solution of N-(2,4-dimethoxy)benzyl
hydroxylamine (8mmol) in CH2Cl2 (85mL). The resulting
mixture was stirred at r.t. for 12h and then filtrated over celite.
The solvent was removed under vacuum and the resulting
crude product was purified by washing with pentane. 2c was
obtained as a pale yellow solid. 2c : mp 51-52 °C; 1H NMR
(CDCl3, 200MHz) 1.04 (d, 6H, J = 6.5Hz), 3.17 (m, 1H), 3.81
(s, 6H), 4.48 (s, 2H), 6.37 (d, 1H, J = 7.2Hz), 6.47 (s, 1H), 6.49
(d, 1H, J = 7.9Hz), 7.26 (d, 1H, J = 7.9Hz); 13C NMR (CDCl3,
75MHz) 18.28, 25.21, 54.94 (2 MeO), 63.09, 97.67, 104.11,
113.17, 132.03, 143.32, 158.36, 161.21. Condensation of 2c
with t-butyl propiolate : To a solution of LDA (prepared from
diisopropylamine (1.85mmol), n-butyllithium (1.6M in
hexane, 1.85mmol) and THF (10mL)) was added, in one
portion at -78 °C, t-butylpropiolate (1.98mmol). The reaction
mixture was stirred for 1h before addition of a solution of
nitrone 2c (1.32mmol) in THF (5mL). After another 1.5h, the
reaction was quenched with water and CH2Cl2 was added. The
two liquid layers were separated. The organic layer was
washed with water and brine, dried over anhydrous MgSO4
and concentrated. The resulting residue was purified by
column chromatography over silica gel (eluent AcOEt/
pentane 1/9) giving the expected hydroxylamine as a pale
yellow solid. 3c : mp 134-135 °C; 1H NMR 1.01 (d, 3H,
J = 6.5Hz), 1.06 (d, 3H, J = 6.5Hz), 1.52 (s, 9H), 2.04 (m, 1H),
3.23 (d, 1H, J = 9.2Hz), 3.78 (s, 6H), 3.92 (ABq, 2H,
Scheme 2
We have then tested the possibility to remove the meth-
oxybenzyl groups from the nitrogen atom. The best results
were obtain with the 2,4-dimethoxybenzyl group. In this
case, the use of DDQ led to nearly quantitative yields of
the NHBoc products 6-86 (Scheme 3). The 4-methoxy-
benzyl group of 5d was removed using CAN as the depro-
tecting reagent (yield :86%).
J
AB = 13Hz, dA-dB = 41.5Hz), 4.84 (s, 1H), 6.46 (m, 2H), 7.27
(d, 1H, J = 8.9Hz); 13C NMR 19.68, 28.02, 30.71, 55.07, 55.29
(2 MeO), 66.03, 80.43, 82.56, 83.02, 98.48, 104.11, 117.73,
132.38, 152.63, 159.10, 160.56; IR (neat) 2229 (nC∫C), 1707
(nC = O). Synthesis of 5c : To a solution of compound 3c
(0.29mmol) in a mixture of acetic acid and methanol (1/9,
4mL) heated at 60 °C was added zinc (5.8mmol). The reaction
mixture was vigorously stirred for 0.5h and then filtrated over
celite. The solids were washed with CH2Cl2 and the collected
solution was neutralized by NaHCO3 at O °C. After an usual
extraction procedure, the resulting crude product (4c) was
dissolved in CH2Cl2 (3mL). Triethylamine (0.58mmol) and
BocOBoc (0.435mmol) were added and the reaction mixture
was stirred for 12h at r.t. Water was then added. A classical
work up, followed by column chromatography (silica gel,
AcOEt/pentane:1/9) afforded the expected N-Boc amine 5c as
a colorless oil. 5c :1H NMR 0.84 (t, 6H, J = 6.5Hz), 1.43 (s,
18H), 1.9-2.4 (m, 1H), 3.77 (s, 3H), 3.78 (s, 3H), 4.25-4.55
(m, 2H), 4.85-5.15 (m, 1H), 5.67 (d, 1H, J = 11.6Hz), 6.0-6.3
(m, 1H), 6.38 (s, 1H), 6.41 (d, 1H, J = 8.9Hz), 7.08 (d, 1H,
J = 8.9Hz); 13C NMR 14.01, 19.35, 19.58, 22.30, 28.09, 28.47,
In conclusion, a stereoselective synthesis of Z-a,b-ethyl-
enic-g-amino esters has been reported. We are currently
studying the chemistry of these compounds with the aim
of preparing various g-amino acids. Furthermore, it
should be noticed that the obtained Z-esters are examples
7
of ’conformationaly restricted’ GABA-analogues.
Scheme 3
Synlett 1999, No. 5, 602–604 ISSN 0936-5214 © Thieme Stuttgart · New York

604
C. Dagoneau et al.
LETTER
55.10, 55.26, 79.00, 80.17, 98.09, 103.69, 119.61, 122.58,
129.46, 144.35, 155.51, 157.81, 159.72, 165.12; IR (neat)
1725, 1694 (nC = O). Synthesis of 7: A solution of ester 5c
(0.08mmol) and DDQ (0.28mmol) in CH2Cl2/H2O (19/1,
6mL) was stirred for 1 day at r.t. The mixture was then diluted
with CH2Cl2 and treated by a saturated aqueous solution of
NaHCO3. Classical work-up, followed by column
1.74-1.92 (m, 1H), 4.89 (broad s, 1H), 5.73 (d, 1H, J = 12Hz),
5.94 (m, 1H); 13C NMR 17.92, 19.07, 28.15, 28.36, 32.65,
53.89, 77.65, 80.64, 122.38, 146.88, 155.00, 165.31; IR (neat)
1712, 1683 (nC = O), 1658 (nC = C).
(7) Simonyi, M. Enantiomer 1996, 1, 403.
chromatography (silica gel, AcOEt/pentane:1/9) afforded the
N-Boc amine 7 as a colorless oil. 7:1H NMR 0.93 (d, 3H,
J = 6.8Hz), 0.94 (d, 3H, J = 6.8Hz), 1.43 (s, 9H), 1.49 (s, 9H),
Article Identifier:
1437-2096,E;1999,0,05,0602,0604,ftx,en;L02499ST.pdf
Synlett 1999, No. 5, 602–604 ISSN 0936-5214 © Thieme Stuttgart · New York