enylphosphine oxide8e,f was used as an additive. When the
reaction was carried out with 30 mol % of Ph3PO and 20
mol % of 1a (Table 1, entry 2), the enantioselectivity of the
desired product increased to 55%, albeit with a slow reaction
rate. To improve the enantioselectivity, careful optimization
of the reaction parameters, additives, and catalysts was
performed. During our investigation, it emerged that catalyst
1a in the presence of 50 mol % of Ph3PO in toluene provided
better ee and chemical yields than did catalysts 2 or 1b (Table
1, entries 3-5) and solvents CH2Cl2 or propionitrile. On the
other hand, varying the silyl ketene acetal did not improve
the yield or ee (Table 1, entries 6-8).10 The enantioselec-
tivity of product 3 was excellent (93% ee) in the presence
of mexyl-substituted catalyst 1c (Table 1, entry 9).
cyclohexylcarboxaldehyde resulted in a moderate yield of
product with 85% ee (Table 2, entry 15).
For aromatic aldehydes, substitution with electron-donating
groups provided excellent yields and enantioselectivities
(>90% ee) (Table 2, entries 2 and 4) due to the effective
coordination between the aldehyde oxygen atom and the
boron of the oxazaborolidinium ion (Figure 2). However,
ortho-tolualdehyde gave the desired aldol product with
moderate enantioselectivity (60% ee, Table 2, entry 3). We
suspect that the bulkier 2-methyl substituent of ortho-
tolualdehyde significantly reduces the degree of complexation
with the catalyst in the pretransition-state assembly leading
to the lower enantioselectivity. Alternatively, strong electron-
withdrawing substituents such as a p-nitro group caused a
small reduction in enantioselectivity (85% ee, Table 2, entry
7), which was to be expected due to the reduced carbonyl
basicity and thus diminished degree of aldehyde complex-
ation with the catalyst leading to lower ee. Biphenyl and
naphthyl carboxaldehydes were also reacted successfully
under similar conditions to provide the corresponding ꢀ-hy-
droxy R,R-dimethyl esters in excellent yields and enanti-
oselectivities (Table 2, entries 8-10).
After optimization of the reaction parameters, the scope
and limitations of this methodology were studied using a
variety of aldehydes (Table 2). Very good yields and high
Table 2. Results of the Catalytic Enantioselective Mukaiyama
Aldol Reactiona
The absolute configuration of the major enantiomeric
isomer has been assigned as (R) by measurement of optical
rotation and comparison with known substances.11 The
resulting configuration of the aldol reaction products presented
in Table 2 can be explained by a cyclic complex between
catalyst 1c and the aldehyde as depicted in Figure 2.
The mode of aldehyde complexation is the same as has
previously been postulated in the enantioselective formation
of (R)-cyanohydrins from aldehydes and trimethylsilyl
cyanide.8e The observed configuration can be rationalized
from the transition state model in which the rear face (back)
of the aldehyde is shielded from attack by the silyl ketene
acetal by bulky aryl groups from the catalyst. Thus, nucleo-
philic attack of the silyl ketene acetal from the si (front) face
of the formyl carbon is facilitated leading to the observed
(R)-enantioselectivity. Due to the greater shielding ability
of a mexyl group, catalyst 1c provided higher ee (2-6%)
than catalyst 1a.
To further evaluate the broad feasibility of the present
catalytic system, R,ꢀ-disubstituted acroleins were used
as substrates (Table 3). The results are summarized in
Table 3.
The synthetic utility of this methodology was further
demonstrated by the total synthesis of inthomycin C, which
was isolated from Streptomyces sp. in 1991.12 The intho-
mycins have been shown to be highly specific inhibitors of
cellular biosynthesis, displaying selective in vitro antimi-
crobial activity,13 and to reduce prostate cancer cell growth.14
To date, there has been only one reported synthesis of
inthomycin C by Taylor and co-workers, which produced
a Reactions were run with 1.0 mmol of aldehyde, 1.2 mmol of silyl
ketene acetal, and 0.2 mmol of catalyst, and deprotection of silylated adduct
was carried out with TBAF. b Isolated yield after column chromatography.
c Determined by chiral HPLC analysis. d Determined by chiral GC analysis.
enantioselectivities were obtained for aromatic as well as
aliphatic or unsaturated aldehydes which are typically more
challenging substrates in asymmetric carbon-carbon bond
forming reactions (Table 2, entries 11-15). The reaction of
(10) The Mukaiyama aldol reaction of benzaldehyde and 1-phenyl-1-
(trimethylsilyloxy)ethylene (R1 ) H, R2 ) Ph) provided ꢀ-hydroxy ketone
in lower enantiomeric excess (48%) and 40% yield.
(9) (a) Wang, X.; Adachi, S.; Iwai, H.; Takatsuki, H.; Fujita, K.; Kubo,
M.; Oku, A.; Harada, T. J. Org. Chem. 2003, 68, 10046. (b) Carreira, E. M.;
Singer, R. A. Tetrahedron Lett. 1994, 35, 4323. (c) Hassfeld, J.; Christmann,
M.; Kalesse, M. Org. Lett. 2001, 3, 3561.
(11) (a) Yamashita, Y.; Ishitani, H.; Shimizu, H.; Kobayashi, S. J. Am.
Chem. Soc. 2002, 124, 3292. (b) Fu, F.; Teo, Y.-C.; Loh, T.-P. Tetrahedron
Lett. 2006, 47, 4267.
(12) Henkel, T.; Zeeck, A. Liebigs Ann. Chem. 1991, 367.
5090
Org. Lett., Vol. 12, No. 22, 2010