relatively few reports of the asymmetric alkylation of
â-alanine or its derivatives. Those syntheses described in the
literature using Oppolzer’s sultam,4 Evans’ chiral auxiliary,5
or Seebach and Juaristi’s chiral pyrimidinone methodology
appeared to give highly scalemic R-substituted â-amino
acids.6 However, the enolate derived from Oppolzer’s sultam
was reportedly unstable at temperatures greater than -45
°C. Moreover, Evans’ chiral auxiliary was prohibitively
expensive for the large-scale synthesis of these derivatives,
while the hydrogenation conditions required for the prepara-
tion of the chiral pyrimidinone derivatives did not scale well
in our hands. Hence, these methodologies appeared to be
unacceptable for the preparation of bulk quantities (>100
g) of these â-amino acid analogues.
Myers has shown pseudoephedrine to be an efficient and
inexpensive chiral auxiliary in the stereoselective synthesis
of unnatural R-amino acids.7 Moreover, this methodology
has been extended to the synthesis of chiral â-hydroxy acids8
and chiral R-substituted acids.9 These reports suggested that
the use of pseudophedrine as a chiral auxiliary should be
applicable to a wide variety of substrates, including â-amino
acids.
of the pseudoephedrine amide 2 was accomplished with
LiHMDS in the presence of excess LiCl.8 The alkylation
step generally gave the desired product in good yield and
with a high degree of stereoselectivity (Table 1).
Table 1. Overall Yields from 2 and Optical Rotations and
Enantiomeric Excesses of 4a-d
product
R-X
yield (%)a [R]n (deg), n ee (%)
D
4a
4b
4c
4d
CH3I
CH3CH2I
CH3(CH2)2I
CH2dCHCH2Br
74
74
63
52
-12.6, 26b
-2.9, 26
3.5, 27
84
94
>99
75
-4.6, 26
a Yield from pseudoephedrine amide 2. b Lit. [R]29 ) -11.8 (c ) 1,
D
1.1 M HCl).2
Analysis of the diastereomeric excesses of 3a-d proved
to be nontrivial. Determination of the de values of the
pseudoephedrine amides by their 1H NMR spectra was
complicated by the presence of rotational isomers.7a These
1
rotational isomers sufficiently complicated the H NMR
We have found that the extension of this method to the
alkylation of â-alanine provides an inexpensive, efficient,
and enantioselective route to R-alkyl â-amino acids (Scheme
2). Boc-â-alanine (1) was coupled to (R,R)-pseudoephedrine
spectra of the amides such that key signals could not be
resolved. Moreover, the diastereomeric protons of MTPA
ester 5 (from the reduction/esterification of Fmoc-4a) were
1
also unresolvable by H NMR. The diastereomeric separa-
tions of Marfey’s derivative 6,10 pseudoephedrine amide 3a,
and 7 were not achieved by HPLC (Figure 1). The diaster-
Scheme 2a
a (i) (1R,2R)-(+)-Pseudoephedrine, PvCl, TEA, THF, 0 °C; HCl,
1:1 H2O/CH3OH; (ii) R-X (X ) Br, I), LHMDS, LiCl, THF, -5-0
°C; (iii) H2O, 4.
(2) using a mixed anhydride method.7b Deprotection of the
amine was accomplished with HCl, and the final product
was obtained after recrystallization from toluene. Lithiation
(4) Ponsinet, R.; Chassaing, G.; Vaissermann, J.; Lavielle, S. Eur. J.
Org. Chem. 2000, 83-90.
Figure 1. â-Amino acid derivatives used to determine ee values.
(5) (a) Evans, D. A.; Urp´ı, F.; Sommers, T. C.; Clark, J. S.; Milodeau,
M. T. J. Am. Chem. Soc. 1990, 112, 8215. (b) Hinternamm, T.; Seebach,
D. HelV. Chim. Acta 1998, 81, 2093-2126.
eomeric separations of 5-7 by chiral HPLC on a Leucine
Pirkle column were also unsuccessful. The enantiomeric
excess was finally determined by GC/MS of the trifluoro-
acetamide-isopropyl ester derivatives 8a-d of amino acids
5a-d on a Chirasil-Val capillary GC column (Table 1).
(6) Juaristi, E.; Quintana, D. Tetrahedron: Asymmetry 1992, 3, 723-
726.
(7) (a) Myers, A.; Schnider, P.; Kwon, S.; Kung, D. J. Org. Chem. 1999,
64, 3322-3327. (b) Myers, A.; Gleason, J.; Yoon, T.; Kung, D. J. Am.
Chem. Soc. 1997, 119, 656-673. (c) Myers, A.; Gleason, J.; Yoon, T. J.
Am. Chem. Soc. 1995, 117, 8488-8489.
3528
Org. Lett., Vol. 2, No. 22, 2000
Nagula
