Organocatalytic asymmetric neber reaction for the synthesis of 2H-azirine carboxylic esters

Article abstract of DOI:10.1021/ol2026747

The first enantioselective Neber reaction of β-ketoxime sulfonates catalyzed by a bifunctional thiourea has been developed. The reaction proceeds stereoselectively with 5 mol % of the catalyst to give the 2H-azirine carboxylic esters in good yields with u

Full text of DOI:10.1021/ol2026747

ORGANIC
LETTERS
2011
Vol. 13, No. 24
6374–6377
Organocatalytic Asymmetric Neber
Reaction for the Synthesis of 2H-Azirine
Carboxylic Esters
Shota Sakamoto, Tsubasa Inokuma, and Yoshiji Takemoto*
Graduate School of Pharmaceutical Sciences, Kyoto University, Yoshida, Sakyo-ku,
Kyoto 606-8501, Japan
takemoto@pharm.kyoto-u.ac.jp
Received October 3, 2011
ABSTRACT
The first enantioselective Neber reaction of β-ketoxime sulfonates catalyzed by a bifunctional thiourea has been developed. The reaction
proceeds stereoselectively with 5 mol % of the catalyst to give the 2H-azirine carboxylic esters in good yields with up to 93% ee. In addition, the
resulting azirines can be successfully employed in the stereoselective synthesis of di- and trisubstituted aziridines.
2H-Azirines are the smallest heterocycles that contain
a CdN double bond in a three-membered ring. Several
naturally occurring 2H-azirines such as azirinomycin and
dysidazirine have been isolated as promising antibiotics
(Figure 1).1
Due to the unique structural and chemical properties,
these strained molecules have been extensively studied as
synthetic intermediates as well as for theoretical and bio-
logical applications.2 Although the reactivity of 2H-azirines
is highly dependent on the nature of the substituents,
chiral 2H-azirine-2-carboxylic acid derivatives are of par-
ticular interest as nonproteinogenic amino acids. So far,
enantiomerically enriched 2H-azirines have been pre-
pared by either a β-elimination of chiral nonracemic
N-substituted aziridines3 or an asymmetric Neber reaction
of ketoxime sulfonates (Figure 2).4
Figure 1. Biologically active 2H-azirines.
(1) (a) Molinski, T. F.; Ireland, C. M. J. Org. Chem. 1988, 53, 2103.
(b) Skepper, C. K.; Molinski, T. F. J. Org. Chem. 2008, 73, 2592.
(c) Stapley, E. O.; Hendlin, D.; Jackson, M.; Miller, A. K. J. Antibiot.
1971, 24, 42. (d) Miller, T. W.; Tristram, E. W.; Wolf, F. J. J. Antibiot.
1971, 24, 48.
(2) (a) Palacious, F.; Ochoa de Retana, A. M.; Martınez de Marigorta,
E.; Manuel de los Santos, J. Eur. J. Org. Chem. 2001, 2401. (b) Stevens,
K. L.; Jung, D. K.; Alberti, M. J.; Badiang, J. G.; Peckham, G. E.; Veal,
J. M.; Cheung, M.; Harris, P. A.; Chamberlain, S. D.; Peel, M. R. Org.
Lett. 2005, 7, 4753. (c) Alves, M. J.; Lemos, A.; Rodriguez-Borges, J. E.;
Carcıa-Mera, X.; Fortes, A. G. Synthesis 2009, 3263. (d) Alves, M. J.;
Figure 2. Synthetic strategy for 2H-azirines.
(3) (a) Davis, F. A.; Liang, C.-H.; Liu, H. J. Org. Chem. 1997, 62,
3796. (b) Gentilucci, L.; Grijzen, Y.; Thijs, L.; Zwanenburg, B. Tetra-
hedron Lett. 1995, 36, 4665. (c) Davis, F. A.; Reddy, G. V.; Liu, H. J. Am.
Chem. Soc. 1995, 117, 3651. (d) Davis, F. A.; Liu, H.; Liang, C.-H.;
Venkat Reddy, G.; Zhang, Y.; Fang, T.; Titus, D. D. J. Org. Chem. 1999,
64, 8929.
~
Duraes, M. M.; Fortes, A. G. Tetrahedron Lett. 2003, 44, 5079. (e) Banert,
K.; Meier, B. Angew. Chem., Int. Ed. 2006, 45, 4015. (f) Novikov, M. S.;
Amer, A. A.; Khlebnikov, A. F. Tetrahedron Lett. 2006, 47, 639. (g) Pinho
e Melo, T. M. V. D.; Cardoso, A. L.; Rocha Gonsalves, A. M. Tetrahedron
2003, 59, 2345.
r
10.1021/ol2026747
Published on Web 11/22/2011
2011 American Chemical Society

The former method provides the desired compounds
with high ee, but the preparation of the chiral aziridines is
sometimes rather cumbersome. In contrast, the latter route
is more concise and practical. However, except for a highly
diastereoselective synthesis using a chiral amino acid as a chiral
auxiliary, no asymmetric reactions which proceed in a highly
enantioselective manner have been achieved with a stoichio-
metric or catalytic amount of chiral ligands (the reported ee’s
are up to 82%).4 Therefore, development of the catalytic
enantioselective Neber reaction would be highly desireable.
We have developed several organocatalyzed asymmetric
reactions5 with bifunctional thioureas in the past decade.6
In these reactions, the catalysts were designed to work
as a dual activator of both nucleophile and electrophile
in bimolecular reactions, guiding the two reactants to get
close in a restricted manner and enabling them to react
smoothly and stereoselectively [(i) in Figure 3]. After our
work, similar bifunctional thioureas bearing an amino or
phosphine group were applied to intramolecular reactions
of some multifunctional substrates.7 The reaction was
considered to proceed via the same dual activation mech-
anism as seen in the bimolecular reactions.
Wethenenvisaged thatthe sameamino thiourea catalyst
might be used effectively in the Neber reaction, if the
thiourea and amine moieties of the catalyst interact with
the leaving group (LG) and ester of the substrate, respec-
tively, through triple hydrogen-bonding interaction [(ii) in
Figure3]. Inthispaper, we describe the first catalytic asym-
metric Neber reaction of ketoxime sulfonates in the pre-
sence of bifunctional thiourea 2a (5 mol %) to give 2H-
azirines as well as the asymmetric synthesis of related
aziridine derivatives.
We selected O-Ts-oxime tert-butyl ester 1aa, prepared as
an inseparable mixture of E/Z isomers in a ratio of 65/35,
for the optimization of the reaction conditions (Table 1).
The reactions of 1aa were carried out in toluene with
10 mol % of 2a in the presence of more than a stoich-
iometric amount of base to trap the generated TsOH
(entries 1ꢀ6). We found that the desired product 3a was
obtained in comparable yields in all cases, but the enan-
tioselectivity was significantly affected by the base used.
Inorganic bases generally gave better results than organic
base and biphasic conditions (entries 1 and 6). In contrast,
the solvents examined had no significant effect on the ee
(entries 7 and 8). We therefore chose Na2CO3 and toluene
for further examination. Although various types of bifunc-
tional organocatalysts 2bꢀg were investigated to enhance
the stereoselectivity, 2a was revealed to be the best catalyst
in terms of the ee (entries 9ꢀ14). Neither cyclic tertiary
amines 2band 2cnorother types of hydrogen-bond donors
2eꢀ2g8 exhibited higher reactivity and improved stereo-
selectivity. The poor results of the same reaction with amide
2h and sulfonamide 2i strongly suggested the importance
of the thiourea moiety of the catalysts for the high enantio-
selectivity (entries 15 and 16).
We next examined substituent effects of the ketoxime
sulfonates 1abꢀ1df (Table 2). Regarding the leaving
group, electron-withdrawing substituents tend to give im-
proved enantioselectivities (entries 1ꢀ6) and the 3,5-bis
(trifluoromethyl)benzenesulfonate proved to be the best
one, furnishing 3a in 72% ee. It is noteworthy that an
ortho-substitution with a nitro group dramatically in-
creased the reaction rate in the case of O-2-Ns-oxime 1ac,
whereas the reaction of O-2-Ts-oxime 1ad led to a decrease
in ee (entry 2 vs 3). Moreover, the use of mesitylsulfonate
completely suppressed the Neber reaction, resulting in the
recovery of the starting material.9 These results are reason-
able, if the oxygen atom of the sulfonate group is consid-
ered to coordinate to the thiourea of the catalyst in the
transition state. We next investigated other esters and
amides as substrates. Tertiary amide 1ba and esters 1ca
or 1da gave comparable or slightly improved outcomes
(entries6ꢀ8). By lowering the reaction temperature to ꢀ20 °C,
the corresponding product 3d was obtained in 78%
Figure 3. Dual activation of bifunctional thioureas.
(4) (a) Palacios, F.; Aparicio, D.; Ochoa de Retana, A. M.; Manuel
ꢀ
de los Santos, J.; Gil, J. I.; Lopez de Munain, R. Tetrahedron: Asym-
metry 2003, 14, 689. (b) Ooi, T.; Takahashi, M.; Doda, K.; Maruoka, K.
J. Am. Chem. Soc. 2002, 124, 7640. (c) Skepper, C. K.; Dalisay, D. S.;
Molinski, T. F. Org. Lett. 2008, 10, 5269. (d) Verstappen, M. M. H.;
Ariaans, G. J. A.; Zwanenburg, B J. Am. Chem. Soc. 1996, 118, 8491.
(5) (a) Doyle, A. G.; Jacobsen, E. N. Chem. Rev. 2007, 107, 5713.
(b) Takemoto, Y. Org. Biomol. Chem. 2005, 3, 4299. (c) Yu, X.; Wang,
W. Chem.;Asian J. 2008, 3, 516. (d) Zhang, Z.; Schreiner, P. R. Chem.
Soc. Rev. 2009, 38, 1187. (e) Moyano, A.; Rios, R. Chem. Rev. 2011, 111,
4703.
(6) (a) Okino, T.; Hoashi, Y.; Takemoto, Y. J. Am. Chem. Soc. 2003,
125, 12672. (b) Okino, T.; Hoashi, Y.; Furukawa, T.; Xu, X.; Takemoto,
Y. J. Am. Chem. Soc. 2005, 127, 119. (c) Xu, X.; Furukawa, T.; Okino,
T.; Miyabe, H.; Takemoto, Y. Chem.;Eur. J. 2006, 12, 466.
(d) Inokuma, T.; Hoashi, Y.; Takemoto, Y. J. Am. Chem. Soc. 2006,
128, 9413.
ꢀ
ꢀ
(8) (a) Vakulya, B.; Varga, S.; Csampai, A.; Soos, T. Org. Lett. 2005,
7, 1967. (b) Konishi, H.; Lam, T. Y.; Malerich, J. P.; Rawal, V. H. Org.
Lett. 2010, 12, 2028. (c) Wang, C.-J.; Zhang, Z.-H.; Dong, X.-Q.; Wu,
X.-J. Chem. Commun. 2008, 1431.
(7) (a) Nodes, W. J.; Nutt, D. R.; Chippindale, A. M.; Cobb, A. J. A.
J. Am. Chem. Soc. 2009, 131, 16016. (b) Liu, X.; Lu, Y. Org. Lett. 2010,
12, 5592.
(9) The reaction of the similar substrate which bears the mesitylsul-
fonate moiety as a leaving group resulted in completely no reaction. The
spectral data of this substrate are shown in the Supporting Information.
Org. Lett., Vol. 13, No. 24, 2011
6375

Table 1. Optimization of the Reaction Conditionsa
Table 2. Effect of the Substituents (R1 and R2)a
time
(h)
yield
(%)b
ee
(%)c
entry
1
3
1
1ab
1ac
1ad
1ae
1af
24
4
3a
3a
3a
3a
3a
3b
3c
3d
3d
3d
3d
78
78
78
81
80
80
81
81
78
75
72
68
62
49
54
72
64
67
68
81
82
86
2
3
24
24
24
13
24
24
48
48
48
4
5
yield
(%)b
ee
(%)c
6
1ba
1ca
1da
1da
1da
1df
entry
2
base
time (h)
7
8
1
2a
2a
2a
2a
2a
2a
2a
2a
2b
2c
2d
2e
2f
Et3Nd
6
24
24
24
24
24
24
24
12
48
24
24
24
24
24
24
78
80
78
79
63
81
75
78
81
80
72
75
56
72
56
21
10
45
52
64
62
38
64
63
51
56
22
57i
28
56
8i
9d
10d,e
11d,e
2
Cs2CO3
K2CO3
3
4
Na2CO3
NaHCO3
5
a Unless otherwise noted, the reaction was performed with 1, thiour-
ea 2a (10 mol %), and Na2CO3 (10 equiv) in toluene. b Yield of isolated
product. c Determined by chiral HPLC analysis. d The reaction was
carried out at ꢀ20 °C e 5 mol % of 2a was used.
f
6e
7g
8h
9
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
Na2CO3
10
11
12
13
14
15
16
the corresponding (trifluoromethyl)benzenesulfonyl oxime
could not be synthesized; however the corresponding
p-toluenesulfonyl oxime gave 5d and 5e in good enantios-
electivities (entries 4 and 5). The reaction of alkenyl-
substituted ketoximes 4fꢀh generally occurred with high
enantioselectivity, providing the cyclized products 5fꢀh
(entries 6ꢀ8). However, the same treatment of alkynyl-
substituted ketoxime 4i resulted in a somewhat lower ee of
azirine 5i (entry 9). Azirine 5j, which can be an important
synthetic intermediate of pleurocybellaziridine,10 could be
prepared from 4j (entry 10).11
2g
2h
2i
5
a Unless otherwise noted, the reaction was performed with 1aa,
catalyst 2 (10 mol %) and base (10 equiv) in toluene. b Yield of isolated
product. c Determined by chiral HPLC analysis. d 1 equiv of base was
used. e Water was added (toluene/H2O = 10:1). f 2 equiv of base was
used. g TBME was used as a solvent. h CH2Cl2 was used as a solvent.
i The antipode of (S)-3a was obtained.
Finally, we explored the synthetic versatility of the
products obtained in the asymmetric Neber reaction. As
shown in Scheme 1, the asymmetric synthesis of (S)-
dysidazarine, antipode of the natural product, was accom-
plished in five steps from commercially available com-
pound 6. β-Keto ester 7 was prepared by the successive
treatment of 6 with N,N0-carbonyldiimidazole (CDI) and
Mg(OCOCH2CO2Me)2 in one pot. Subsequently, 7 was
transformed into dysidazirine by the standard three-
step procedure. The reaction of 7 with hydroxylamine,
sulfonylation of the resulting oxime with 3,5-bis(CF3)2-
C6H3SO2Cl, and asymmetric Neber reaction provided the
yield and 81% ee (entry 9). In addition, the Neber reaction
proceeded equally well with 5 mol % of 2a without
decreasing the enantioselectivity (entry 10). Finally, when
we used substrate 1df, which was designed by the combina-
tion of the two most positive substituent effects, the first
high enantioselectivity (86% ee) was achieved by the cat-
alytic asymmetric Neber reaction (entry 11).
After optimizing the substrates, our attention was direc-
ted toward the scope of the reactions (Table 3). Azirine 5a
bearing a propyl group was synthesized in 71% yield with
86% ee from O-3,5-bis(trifluoromethyl)benzenesulfonyl
oxime 4a (entry 1). When the reaction was applied to
the substrates with longer alkyl chains, the enantioselecti-
vity slightly increased (entries 2 and 3). In the cases of
ketoximes 4d and 4e possessing secondary alkyl groups,
(10) Wakimoto, T.; Asakawa, T.; Akahoshi, S.; Suzuki, T.; Nagai,
K.; Kawagishi, H.; Kan, T. Angew. Chem., Int. Ed. 2011, 50, 1168.
(11) Obara, K.; Wada, C.; Yoshioka, T.; Enomoto, K.; Yagishita, S.;
Toyoshima, I. Neuropathology 2008, 28, 151.
6376
Org. Lett., Vol. 13, No. 24, 2011

The azirine 5j was successfully converted into the aziridine 9,
which is the synthetic precursor of Pleurocybella porrigens.
During the course of the Neber reaction and Grignard
reaction, the enantiomeric excess did not decrease
and comparison of the specific rotation of 9 with that
of the reported data also supported that the Neber
reaction catalyzed by 2a gave an (S)-configuration
(Scheme 2, eq 2).10
Table 3. Scope of the Substratesa
Scheme 2. Transformation of the Azirines
temp
time
yield
(%)b
ee
(%)c
entry
4
(°C)
(h)
5
1
2
3
4
5
6
7
8
9
10
4a
4b
4c
4d
4e
4f
ꢀ10
ꢀ10
ꢀ10
ꢀ20
ꢀ20
ꢀ20
ꢀ20
ꢀ20
ꢀ20
ꢀ20
48
48
48
72
72
24
24
24
72
72
5a
5b
5c
5d
5e
5f
71
78
82
82
73
96
73
77
81
64
86
93
88
85
88
90
93
90
82
80
4g
4h
4i
5g
5h
5i
4j
5j
a Unless otherwise noted, the reaction was performed with 4
(1.0 equiv), thiourea 2a (5 mol %), and Na2CO3 (10 equiv) in toluene.
b Yield of isolated product. c Determined by chiral HPLC analysis.
In conclusion, we have developed the first highly en-
antioselective catalytic Neber reaction for the synthesis
of chiral 2H-azirines. The reaction requires the bifunc-
tional aminothiourea 2a as a chiral catalyst. In addition we
found that it is effective to use the 3,5-bis(trifluoromethyl)-
benzenesulfonyl group as a leaving group for high
enantioselectivity. In particular, the thiourea moiety of
catalyst 2a is important, and use of amide and sulfonamide
catalysts possessing no thiourea results in a dramatic
decrease in ee. Research to clarify the detailed mechanism
of the asymmetric Neber reaction is currently underway in
our laboratory.
Scheme 1. Asymmetric Synthesis of (þ)-Dysidazirine
Acknowledgment. This work was partially supported
by a Grant-in-Aid for Scientific Research on Innovative
Areas “Advanced Molecular Transformations by Organo-
catalysts”, ‘Targeted Proteins Research Program’ from
the Ministry of Education, Culture, Sports, Science and
Technology of Japan.
desired (S)-azirine in 75% yield with 93% ee. The ee value
of the synthetic product is comparable to the one reported
for the same product obtained by diastereoselective azir-
idination using a chiral sulfoxide auxiliary.3d
Furthermore, various tri- and disubstituted chiral aziridines
8aꢀc could be prepared from 3d by the diastereoselective
nucleophilic addition of methyl, alkynyl, and hydride
groups according to the reported methods (Scheme 2, eq 1).2
Supporting Information Available. Experimental pro-
1
cedures, characterization data, and H and 13C NMR
spectra for new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
Org. Lett., Vol. 13, No. 24, 2011
6377