Reactivity of 2-acylaminoacrylates with ketene diethyl acetal; [2 + 2] cycloadditions vs. tandem condensations

Article abstract of DOI:10.1039/b302000b

The reactivity of 2-acylaminoacrylates with ketene diethyl acetal can be modulated by means of thermal conditions to yield cyclobutanes for the preparation of protected β-hydroxycyclobutane-α-amino acids, or catalytic conditions that yield cyclohexanes by

Full text of DOI:10.1039/b302000b

Reactivity of 2-acylaminoacrylates with ketene diethyl acetal; [2 + 2]
cycloadditions vs. tandem condensations†
Alberto Avenoza,* Jesús H. Busto, Noelia Canal and Jesús M. Peregrina*
Departamento de Química, Universidad de La Rioja, UA-CSIC, 26006 Logroño, Spain.
E-mail: alberto.avenoza@dq.unirioja.es; Fax: +34 941 299655; Tel: +34 941 299655
Received (in Cambridge, UK) 20th February 2003, Accepted 22nd April 2003
First published as an Advance Article on the web 13th May 2003
The reactivity of 2-acylaminoacrylates with ketene diethyl
acetal can be modulated by means of thermal conditions to
yield cyclobutanes for the preparation of protected b-
hydroxycyclobutane-a-amino acids, or catalytic conditions
that yield cyclohexanes by tandem condensations to obtain
interesting building blocks that are alternatives to Da-
nishefsky’s diene.
Unexpectedly, the cyclobutane core was not obtained and the
reaction led to a cyclohexane ring by a sequential Michael-
aldol-type process. To the best of our knowledge, this reactivity
of ketene diethyl acetal has been explored only in quinone
chemistry and has recently been used in the synthesis of
optically active lactones.13 The yield was increased by the use
of methylaluminium bis(2,6-di-tert-butyl-4-bromophenoxide)
(MABR) as catalyst (Table 1, Scheme 2).
Due to the main role that 2-acylaminoacrylates have played in
the field of novel amino acids,1 these compounds have been the
subject of several synthetic studies2 involving reactions such as
cyclopropanations,3 Diels–Alder reactions,4 hydrogenations,5
and nucleophilic additions.6 However, these compounds have
never been investigated in terms of [2 + 2] reactions, while in
this field methyl acrylate, an olefin acceptor, has been widely
used as a starting material in [2 + 2] cycloadditions.7 In the
context of our research programme on the synthesis of
conformationally restricted hydroxy amino acids8 and in order
to obtain 1-aminocyclobutane-1-carboxylic acids with oxygen
groups at C-2, which are significant targets in bioorganic
chemistry,9 we studied the cycloaddition of 2-acylaminoacry-
lates 1–4 with ketene diethyl acetal 5. The reaction gave
cyclobutane-a-amino acid derivatives 6–7 and these could be
easily transformed into b-hydroxycyclobutane-a-amino acids
(c4Ser), which are analogues of serine. Since c5Ser and c6Ser
have been described and c6Ser even incorporated into pep-
tides,10 the kind of restricted amino acid described here is a very
important target in peptide chemistry due to the lack of available
c4Ser (Scheme 1).
Table 1 Reactivity of 2-acylaminoacrylates with ketene diethyl acetal
Sub-
strate R1
R2
Methoda
Catalyst
Product Yieldb
1
2
3
4
4
4
4
OH
Ac
Ac
Bz
Ac
Ac
Ac
Ac
A
A
A
A
B
B
B
–
–
–
–
–
–
6
7
7
–
–
NHiPr
OMe
OMe
OMe
OMe
OMe
51%
64%c
traces
51%
80%
AlMe3
MAD
MABR
10
10
a Method A corresponds to thermal conditions in tert-butanol at 83 °C and
method B corresponds to catalytic conditions in CH2Cl2 at 20 °C.
b Obtained after purification of cycloadduct by column chromatography.
c Conversion of 80% measured by HPLC.
Taking into account the mechanism proposed for the thermal
[2 + 2] cycloaddition between acrylates and ketene diethyl
acetal, we attempted the reaction in two polar solvents
(acetonitrile and tert-butanol).11 In the case of substrates 1 or 2
the reaction did not proceed at all. Fortunately, the reactions
with methyl 2-benzamido- and 2-acetamidoacrylates (3 and 4)
gave the desired cyclobutane core in yields similar to those
obtained with acrylates7 (Table 1). The structure of compound
7 was unambiguously determined by X-ray diffraction‡ (Figure
1).
Fig. 1 X-ray structure of compound 7.
In order to assess catalytic conditions, we carried out the
reaction at low temperature in the presence of aluminium
catalysts. When AlMe3 was employed only traces of 7 were
detected because of the propensity of this compound to undergo
the ring-opening reaction. To prevent this, we used the soft and
bulky Lewis acid methylaluminium bis(2,6-di-tert-butyl-4-me-
thylphenoxide) (MAD)12 (Table 1).
Scheme 1 Reaction of 2-acylaminoacrylates and ketene diethyl acetal.
† Electronic supplementary information (ESI) available: general proce-
dures. See http://www.rsc.org/suppdata/cc/b3/b302000b/
Scheme 2 Mechanism proposed for different reaction conditions.
1376
CHEM. COMMUN., 2003, 1376–1377
This journal is © The Royal Society of Chemistry 2003

The first addition of one molecule of ketene diethyl acetal 5
onto amidoacrylate 4 by Michael reaction leads to the zwitterion
8. In the case of thermal conditions this intermediate gives
directly the cyclobutane core 7. Nevertheless, when MAD or
MABR are used as the Lewis acid, zwitterion 8 undergoes
another ketene diethyl acetal addition to give the zwitterion 9,
which in turn gives the cyclohexane nucleus 10 exclusively,
without traces of compound 7. Moreover, to confirm this
mechanism we carried out the reaction of compound 7 with 5 in
the presence of MABR and using CH2Cl2 as a solvent and after
10 min. at rt, compound 10 was obtained in good yield (Scheme
2).
framework by a Michael-aldol tandem condensation, adding a
new use of Yamamoto catalyst.15 The use of thermal conditions
therefore gives the cyclobutane derivatives 6 and 7, while
catalytic conditions give rise to the unexpected cyclohexane
derivative. Both pathways open the door to important com-
pounds, exemplified by the novel Ac–c4Ser(OBn)–OH. An
asymmetric approach to exploit this new reactivity of 2-acyla-
minoacrylates will be explored in the near future.
We thank the Ministerio de Ciencia y Tecnología (project
PPQ2001-1305), the Gobierno de La Rioja (project ANGI2001/
30 and grant of N. C.) and the Universidad de La Rioja (project
API-02/03).
In order to explore the possible synthetic use of cyclobutane
and cyclohexane rings, we developed two synthetic routes.
Compound 7 was reduced using LiBH4 to give the correspond-
ing cyclobutanol, which was hydrolysed with HCl to give keto
alcohol 11 (Scheme 3).
Notes and references
‡ Crystal data: (a) C12 H21 N O5, Mw = 259.30, colourless prism of 0.50 3
0.20 3 0.10 mm, T = 293(2) K, orthorhombic, space group P 21 21 21, Z =
8, a = 9.3229(3), b = 13.8767(7), c = 22.3925(9) Å, V = 2896.9(2) Å3,
dcalc = 1.189 g cm23, F(000) = 1120, l = 0.71073 Å (Mo-Ka), µ = 0.092
mm21, Nonius kappa CCD diffractometer, q range 1.91–27.89°, 3068
collected reflections, 3068 unique (Rint = 0.000), full-matrix least-squares
(SHELXL97, see ref. 16), R1 = 0.0522, wR2 = 0.1145, (R1 = 0.0697, wR2
= 0.1252 all data), goodness of fit = 1.063, residual electron density
between 0.123 and 20.118 e Å23. Hydrogen atoms were located from
mixed methods (electron-density maps and theoretical positions). CCDC
204647. See http://www.rsc.org/suppdata/cc/b3/b302000b/ for crystallo-
graphic data in .cif or other electronic format
§ Crystal data: C23 H31 N O3 Si, Mw = 397.58, colourless prism of 0.40 3
0.35 3 0.25 mm, T = 173(2) K, monoclinic, space group P 21/c, Z = 4, a
= 15.8810(3), b = 8.8310(2), c = 18.6380(4) Å, V = 2238.44(8) Å3, dcalc
= 1.180 g cm23, F(000) = 856, l = 0.71073 Å (Mo-Ka), µ = 0.127
mm21, Nonius kappa CCD diffractometer, q range 1.91–27.89°, 16905
collected reflections, 5311 unique (Rint = 0.0392), full-matrix least-squares
(SHELXL97),16 R1 = 0.0501, wR2 = 0.1314, (R1 = 0.0682, wR2 = 0.1440
all data), goodness of fit = 1.046, residual electron density between 0.642
and 20.334 e Å23. Hydrogen atoms were located from mixed methods
(electron-density maps and theoretical positions) CCDC 204646.
Protection of the alcohol group with tert-butyldiphenylsilyl
chloride, followed by hydride addition to the Si face of the
carbonyl group, gave alcohol 12 as a single isomer. This
compound was assigned unambiguously by X-ray diffraction§.
Protection of the secondary alcohol with benzyl 2,2,2-tri-
chloroacetimidate (BTCA), cleavage of the silyl group with
TBAF and oxidation in the presence of Jones reagent gave the
desired Ac–c4Ser(OBn)–OH. Purification of this compound
was achieved through esterification with CH2N2 to obtain, after
column chromatography, the pure Ac–c4Ser(OBn)–OMe (15).
As an alternative, and in order to increase the yield of the last
steps, we protected alcohol 12 as an acetyl ester to give 14.
Compound 16 was obtained from 14 following the same
procedure as described above. This scheme represents, to the
best of our knowledge, the first synthesis of protected c4Ser
(Scheme 3).
On the other hand, compound 10 was transformed into the
interesting building block 17 by simple hydrolysis and further
treatment with DBU in EtOH. The position of the OEt group
was assigned by NOE experiments. These polyfunctional
cyclohexanes can be used as alternatives to Danishefsky’s diene
in Diels–Alder reactions with 2-acetamido- or 2-benzamidoa-
crylates14 to obtain enones with an additional OEt group
(Scheme 3).
1 C. Cativiela and M. D. Díaz-de-Villegas, Tetrahedron: Asymmetry,
1998, 9, 3517; C. Cativiela and M. D. Díaz-de-Villegas, Tetrahedron:
Asymmetry, 2000, 11, 645.
2 U. Schmidt, A. Lieberknecht and J. Wild, Synthesis, 1988, 159.
3 C. Cativiela, M. D. Díaz-de-Villegas, J. A. Mayoral and E. Melendez, J.
Org. Chem., 1985, 50, 3167.
4 A. Avenoza, C. Cativiela, M. A. Fernández-Recio and J. M. Peregrina,
Tetrahedron: Asymmetry, 1999, 10, 3999.
5 A. Wolfson, S. Janssens, I. Vankelecom, S. Geresh, M. Gottlieb and M.
Herskowitz, Chem. Commun., 2002, 388.
In conclusion, the absence or presence of bulky aluminium
derivatives in the reaction between ketene diethyl acetal and
2-amidoacrylates allows the synthesis of either the cyclobutane
skeleton by a [2 + 2] cycloaddition or the cyclohexane
6 R. Labia and C. Morin, J. Org. Chem., 1986, 51, 249.
7 K. C. Brannock, R. D. Burpitt and J. G. Thweatt, J. Org. Chem., 1964,
29, 940; Ph. Amice and J. M. Coina, Bull. Soc. Chim. Fr., 1974, 1015;
C. S. Kniep, A. B. Padias and H. K. Hall Jr., Tetrahedron, 2000, 56,
4279.
8 A. Avenoza, C. Cativiela, M. A. Fernández-Recio and J. M. Peregrina,
J. Chem. Soc., Perkin Trans. 1, 1999, 3375.
9 R. D. Allan, J. R. Hanrahan, T. W. Hambley, G. A. R. Johnston, K. N.
Mewett and A. D. Mitrovic, J. Med. Chem., 1990, 33, 2905; V. N.
Balaji, K. Ramnarayan, M. F. Chan and S. N. Rao, Pept. Res., 1995, 8,
178; E. Gershonov, R. Granoth, E. Tzehovol, Y. Gaoni and M. Fridkin,
J. Med. Chem., 1996, 39, 4833.
10 A. Avenoza, J. I. Barriobero, C. Cativiela, M. A. Fernández-Recio, J. M.
Peregrina and F. Rodríguez, Tetrahedron, 2001, 57, 2745.
11 R. Huisgen, Acc. Chem. Res., 1977, 117.
12 K. Maruoka, H. Imoto, S. Saito and H. Yamamoto, Synlett, 1993,
197.
13 J. Banville, J.-L. Grandmaison, G. Lang and P. Brassard, Can. J. Chem.,
1974, 52, 80; D. W. Cameron, M. J. Crossley and G. I. Feutrill, J. Chem.
Soc., Chem. Commun., 1976, 275; D. W. Cameron, M. J. Crossley, G.
I. Feutrill and P. G. Griffiths, J. Chem. Soc., Chem. Commun., 1977,
297; H. Audrain and K. A. Jørgensen, J. Am. Chem. Soc., 2000, 122,
11543.
14 A. Avenoza, J. I. Barriobero, J. H. Busto, C. Cativiela and J. M.
Peregrina, Tetrahedron: Asymmetry, 2002, 13, 625.
15 H. Yamamoto and S. Saito, Chem. Commun., 1997, 1585.
16 G. M. Sheldrick, SHELXL-97, Program for refinement of crystal
structures, University of Göttingen, Germany, 1997.
Scheme 3 (a) i) LiBH4, Et2O, rt; ii) 1N HCl, THF/H2O, rt, 60%; (b) i)
TBDPSCl, imidazole, DMF, rt; ii) NaBH4, THF/EtOH, rt, 52%; (c) BTCA,
TfOH (cat), Et2O, rt to obtain 13 or Ac2O, DMAP, pyridine, rt to obtain 14;
(d) i) TBAF, THF, rt; ii) Jones reagent, acetone, 0 °C; iii) CH2N2, Et2O, rt
20% from 12 in Bn route and 33% from 12 in Ac route; (e) i) THF/1N HCl,
rt, ii) DBU, EtOH, rt, 78%.
CHEM. COMMUN., 2003, 1376–1377
1377