Generation and reactions of α-trifluoromethyl stabilized aziridinyl anion, a general synthetic precursor for stereospecific construction of α-amino-α-trifluoromethylated quaternary carbon

Article abstract of DOI:10.1016/S0040-4039(03)01520-X

Optically pure α-trifluoromethylated aziridinyl anions react with various electrophiles to give the corresponding optically pure 2-trifluoromethyl-2-substituted aziridines, which are general synthetic precursors for optically pure α-amino-α-trifluoromethylated compounds, such as trifluoromethylated α/β-amino acids, in good yields.

Full text of DOI:10.1016/S0040-4039(03)01520-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 6319–6322
Generation and reactions of a-trifluoromethyl stabilized
aziridinyl anion, a general synthetic precursor for stereospecific
construction of a-amino-a-trifluoromethylated quaternary carbon
Yoshihiro Yamauchi, Tomomi Kawate, Hiromi Itahashi, Toshimasa Katagiri and Kenji Uneyama*
Department of Applied Chemistry, Faculty of Engineering, Okayama University, Tsushimanaka 3-1-1, Okayama 700-8530, Japan
Received 30 April 2003; revised 13 June 2003; accepted 13 June 2003
Abstract—Optically pure a-trifluoromethylated aziridinyl anions react with various electrophiles to give the corresponding
optically pure 2-trifluoromethyl-2-substituted aziridines, which are general synthetic precursors for optically pure a-amino-a-tri-
fluoromethylated compounds, such as trifluoromethylated a/b-amino acids, in good yields.
© 2003 Published by Elsevier Ltd.
Development of new synthetic methodologies for intro-
duction of fluorine atoms into various bioactive com-
pounds,1 such as amino acids,2 has been eagerly
demanded. Among such demands, construction of a
chiral trifluoromethylated quaternary carbon center is a
tough problem which remains to be solved. Introduc-
tion of a trifluoromethyl group toward the a-carbon of
amino acid to construct the trifluoromethylated quater-
nary carbon sometimes resulted in providing higher
chemical stability, strict conformational fixation, and
remarkable change in reactivity compared to the parent
amino acids. Such quaternary carbons have mainly
been prepared from starting materials such as
CF3TMS,3 CF3I,4 1,1,2-perfluoroalkyl(aryl)ethylenes,5
trifluoroacetoimidoyl halides,6 and trifluoropyruvates;7
thus further resolution of the racemate or individual
stereoselective methodology was needed for obtaining a
chiral one. Zanda et al. succeeded in construction of
optically active quaternary trifluoromethylated a-amino
acids via diastereoselective alkylation of chiral sulfi-
mines derived from trifluoropyruvate with Grignard
reagents7b,c or the reaction of non-chiral imine from the
pyruvate with chiral nucleophiles,7d–f although individ-
ual effort was needed for every amino acid to tune the
reaction conditions to attain high diastereoselectivity.
Meanwhile, no deprotonation followed by stereospecific
alkylation on the a-trifluoromethylated methine carbon
(CF3-CH-RR%) has been realized.
We recently reported that the trifluoromethylated three-
membered ring system is useful for a stereospecific
construction of quaternary trifluoromethylated com-
pounds; generation of a trifluoromethyl-stabilized oxi-
ranyl anion from highly available optically pure
2,3-epoxy-1,1,1-trifluoropropane,8,9 and stereospecific
alkylation.10 This result clearly demonstrates that the
oxiranyl anion does not racemize in the course of the
reaction, and is a very efficient precursor for stereospe-
cific construction of tertiary trifluoromethylated alco-
hols with optically pure quaternary carbon. On this
basis, optically pure 2-trifluoromethyl-2-substituted
aziridines would be general precursors for b-
trifluoroamines11 with a quaternary stereogenic carbon
center and could be prepared by the stereospecific
alkylation of the trifluoromethylaziridinyl anion.8,12
This protocol would be useful for the preparation of
optically pure trifluoromethylated quaternary amino
acids and b-amino-a-hydroxy-b-trifluoromethyl car-
boxylic acids; the latter would be a key structure of
protease inhibitors (Scheme 1).
The preparation of N-substituted-2-trifluoromethyl
aziridines is summarized in Scheme 2. The common
starting material, optically pure 2,3-epoxy-1,1,1-trifl-
uoropropane 1, was prepared via hydrolytic kinetic
resolution.10 Preparative methods for N-benzyl
aziridine 5, N-(p-anisyl) aziridine 6 and N-(o-anisyl)
Scheme 1.
* Corresponding author. E-mail: uneyamak@cc.okayama-u.ac.jp
0040-4039/$ - see front matter © 2003 Published by Elsevier Ltd.
doi:10.1016/S0040-4039(03)01520-X

6320
Y. Yamauchi et al. / Tetrahedron Letters 44 (2003) 6319–6322
aziridine 7 were already reported.13 N-Tosyl aziridine
10 was prepared via intramolecular SN2 substitution of
tosylate 9 of a-trifluoromethyl amino alcohol 8.
rized in Table 2. The aziridinyl anion from N-Ts
aziridine 10 reacted with aldehydes and ketones
smoothly to give the corresponding alcohols in moder-
ate to good yields (entries 1–5). The reaction with
glyoxylate gave 12e,14 a general a-hydroxy-b-amino
acid precursor, predominantly (entry 5). The reactions
with acid chloride and halo formate gave the expected
products in good yields (entries 6 and 7). The product
12g is the general precursor for a-trifluoromethyl-a-
amino acids.15 The reaction with benzyl bromide gave
alkylated product 12h in 13% yield (entry 8).
Generation of aziridinyl anions was confirmed by the
reaction with PhCHO. The generation of the anion was
markedly affected by the structure of the N-substituent,
as summarized in Table 1. Detailed experimental condi-
tions are described in the footnote to Table 1. The
combination of N-(anisyl)-aziridines 6 and 7 with s-
BuLi and N-Ts-aziridine 10 with n-BuLi gave the
expected aziridinyl alcohols 11a in 68%, 11b in 45% and
12a in 83% yields, respectively (Table 1, entries 3, 5,
and 6). The diastereoselectivity of the product 12a was
75/25. The present results suggested that the electron-
withdrawing property of N-substituents and the depro-
tonating ability of the alkyllithium are essential to
promote the reaction. Thus, the Ts group would act as
the best N-substituent among those tested.
The stereochemistry of the product was confirmed by
an X-ray crystallographic analysis of a ring-opened
derivative of compound 12g. Compound 12g was
allowed to react with optically pure phenethyl amine to
give compound 13, 1,1,1-trifluoromethyl-2,3-diamino
acid (Scheme 3). The ORTEP view of 13 is shown in
Figure 1.16 Retention of the configuration at the tri-
fluoromethylated quaternary carbon center in the
course of the reaction was confirmed.
The scope of the electrophiles of the reaction is summa-
Scheme 2.
Table 1. Preparations and reactions of aziridinyl anion with PhCHO
Entry
Aziridine
Base
Time (min)
Isolated yield (%)
1
2
3
4
5
6
5
6
6
7
7
10
n-BuLi
n-BuLi
s-BuLi
n-BuLi
s-BuLi
n-BuLi
30
30
30
30
30
10
No reaction
11a (11)a
11a, 68
(Recovery of 5, 97%)
(Recovery of 6, 84%)
11b (8)a
11b, 45
(Recovery of 7, 83%)
12a, 83
General reaction procedure: To a solution of aziridine (1 mmol) in dry THF (5 ml), cooled to −102°C, base (1.1 equiv.) was added. After 10–30
min, benzaldehyde (1.5 mmol) was added and then stirred for a further 10 min.
a Yield was determined by 19F NMR.

Y. Yamauchi et al. / Tetrahedron Letters 44 (2003) 6319–6322
6321
In conclusion, we have succeeded in the generation and
the stereospecific alkylation of a-trifluoromethylated
aziridinyl anions. The N-substituents were found essen-
tial for generation of the anion. The aziridinyl anion
reacted well with various electrophiles in moderate to
good yields. Moreover, the reaction proceeded with
retention of the absolute configuration at the trifl-
uoromethylated quaternary carbon center throughout
the reactions. The present products, especially 12e and
12g can be general precursors for optically pure trifl-
uoromethylated a/b-amino acids. Further optimization
of the reaction conditions and synthetic applications
are now in progress.
Scheme 3.
Table 2. Scope of the electrophiles

6322
Y. Yamauchi et al. / Tetrahedron Letters 44 (2003) 6319–6322
1998, 39, 7771; (f) Bravo, P.; Fustero, S.; Guidetti, M.;
Volonterio, A.; Zanda, M. J. Org. Chem. 1999, 64, 8731.
8. Yamauchi, Y.; Katagiri, T.; Uneyama, K. Org. Lett. 2002,
4, 173.
9. (a) Katagiri, T. In Enantiocontorolled Synthesis of Fluoro-
Organic Compounds: Stereochemical Challenges and
Biomedical Targets; Soloshonok, V. A., Ed.; J. John Wiley
& Sons: New York, 1999; p. 161; (b) Katagiri, T.;
Uneyama, K. J. Fluorine Chem. 2000, 105, 285.
10. Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga,
M.; Hansen, K. B.; Gould, A. E.; Furrow, M. E.; Jacob-
sen, E. N. J. Am. Chem. Soc. 2002, 124, 1307.
11. Review on optically active b-fluorinated amines, see: (a)
Bravo, P.; Crucianelli, M.; Ono, T.; Zanda, M. J. Fluorine
Chem. 1999, 97, 27. Some preparations and reactions of
racemic aziridines have been reported, see: (b) Osipov, S.
N.; Kolomiets, A. F.; Fokin, A. V. Izv. Akad. Nauk.
SSSR, Ser. Khim. 1988, 132; (c) Rozhkov, V. V.;
Makarov, K. N.; Osipov, S. N.; Chervin, I. I.; Ignatenko,
A. V.; Kostyanovsky, R. G. Mendeleev Commun. 1997,
229.
Figure 1. The ORTEP view of 13.
12. For recent reviews on aziridinyl anions, see: (a) Satoh, T.
Chem. Rev. 1996, 96, 3303; (b) McCoull, W.; Davis, F. A.
Synthesis 2000, 1347.
13. Katagiri, T.; Ihara, H.; Takahashi, M.; Kashino, S.;
Furuhashi, K.; Uneyama, K. Tetrahedron: Asymmetry
1997, 8, 2933.
Acknowledgements
We thank SC NMR Laboratory of Okayama Univer-
1
sity for 19F and H NMR analyses and the Venture
Business Laboratory of Graduate School of Okayama
University for X-ray crystallographic analysis. This
work was financially support by Monbusho Grant-in-
Aid for Scientific Research (No. 12650854 and No.
13555254).
14. Spectroscopic data for 14e (a mixture of diastereomers):
Overall yield was 27%, ds=67/33, viscous colorless liquid.
1
IR (neat) 3520, 1750 cm−1; H NMR (500 MHz, CDCl3)
l 7.90 (d, J=9 Hz, 2H, minor), 7.87 (d, J=9 Hz, 2H,
major), 7.35–7.38 (m, 2H, major and minor), 5.04 (d, J=7
Hz, 1H, major), 5.02 (br, 1H, minor), 4.22–4.38 (m, 2H,
major and minor), 3.61 (d, J=4 Hz, 1H, minor), 3.55 (d,
J=7 Hz, 1H, major), 3.16 (s, 1H, minor), 3.02 (d, J=2
Hz, 1H, major), 2.85 (s, 1H, minor), 2.74 (s, 1H, major),
2.46 (s overlap?, 3H, major and minor), 1.33 (t, J=7 Hz,
3H, minor), 1.28 (t, J=7 Hz, 3H, major) ppm; 19F NMR
(282 MHz, CDCl3) l 91.4 (s, CF3, minor), 91.1 (s, CF3,
major) ppm; DI/MS m/z (%) 367 (tr, M+), 294 (7), 212 (5),
195 (4), 155 (47), 138 (27), 91 (100), 65 (30); Anal calcd for
C14H16F3NO5S: C, 45.77; H, 4.39; N, 3.81. Found: C,
45.78; H, 4.58; N, 4.15; [h]2D5=−0.4 (diastereo mixture, c
1.06, MeOH).
References
1. (a) Welch, J. T.; Eswarakrishnan, S. Fluorine in Bioorganic
Chemistry; John Wiley & Sons: New York, 1991; (b)
Soloshonok, V. A. Enantiocontrolled Synthesis of Fluoro-
Organic Compounds: Stereochemical Challenges and
Biomedicinal Targets; Wiley: Chichester, 1999.
2. Kukhar, V. P.; Soloshonok, V. A. Fluorine-containing
Amino Acids Synthesis and Properties; John Wiley & Sons:
New York, 1995.
3. Prakash, G. K. S.; Mandal, M. J. Fluorine Chem. 2001,
123, 112.
4. (a) Aiet-Moham, S.; Takeuchi, N.; Medebielle, M.; Dol-
bier, W. R. Org. Lett. 2001, 3, 4271; (b) Petrov, V. A.
Tetrahedron Lett. 2001, 42, 3267.
5. Ramachandran, P. V.; Jennings, M. P. Org. Lett. 2001, 3,
3789.
6. (a) Uneyama, K.; Yan, F.; Hirama, S.; Katagiri, T.
Tetrahedron Lett. 2000, 37, 2045; (b) Bravo, P.; Capelli, S.;
Meille, S. V.; Seresini, P.; Volonterio, A.; Zanda, M.
Tetrahedron: Asymmetry 1996, 7, 2321.
15. Spectroscopic data for 12g: Overall yield was 85%, white
1
solid. IR (KBr) 1760 cm−1; mp=81–82°C; H NMR (300
MHz, CDCl3) l 7.82–7.86 (m, 2H), 7.26–7.39 (m, 2H),
3.93 (s, 3H), 3.54 (q, J=2 Hz, 1H), 2.79 (s, 1H), 2.47 (s,
3H) ppm; 19F NMR (282 MHz, CDCl3) l 89.0 (s, CF3)
ppm; GC/MS m/z (%) 323 (2: M+), 292 (2), 168 (20), 155
(56), 91 (100), 65 (23). Anal. calcd for C12H12F3NO4S: C,
44.58; H, 3.74; N, 4.33. Found: C, 44.63; H, 4.03; N, 4.31;
[h]2D5=−60.7 (c 1.16, MeOH).
16. Crystal data for 13 at 150 K: C20H23F3N2O4S; Mr=
7. (a) Gathergood, N.; Zhuang, W.; Jorgensen, K. A. J. Am.
Chem. Soc. 2000, 122, 12517; (b) Assensio, A.; Bravo, P.;
Crucianelli, M.; Farina, A.; Fustero, S.; Soler, G. J.;
Meille, V. S.; Panzeri, W.; Viani, F.; Volonterio, A.;
Zanda, M. Eur. J. Org. Chem. 2001, 1449; (c) Volonterio,
A.; Bravo, P.; Panzeri, W.; Pesenti, C.; Zanda, M. Eur. J.
Org. Chem. 2002, 3336; (d) Kukhar, V. P.; Soloshonok, V.
A.; Bravo, P.; Capelli, S.; Meille, S. V.; Viani, F.; Zanda,
M. Tetrahedron: Asymmetry 1994, 5, 2009; (e) Bravo, P.;
Crucianelli, M.; Vergani, B.; Zanda, M. Tetrahedron Lett.
444.47; orthorhombic; P212121 (c19); a=7.2660(2), b=
3
,
,
13.3922(4), c=21.2120(9) A, V=2064.1(1) A , Z=4,
Dx=1.430 g/cm3; v=2.13 cm−1 for Mo Ka radiation
,
(u=0.7107 A). The structure was solved by a direct
method (SIR92), expanded using Fourier techniques
(DIRDIF94), and refined by a full-matrix least-square
method. Final R was 0.030 and Rw was 0.033 for 2290
reflections with I0>3.00| (I0). Reflection/parameter ratio
was 6.31, Goodness of fit indicator was 1.73. Max shift/
error in final cycle was 0.06.