.
Angewandte
Communications
acyclic nucleophile is not favored, and the use of a bulky
modifier at the a position is necessary to discriminate the
reactive enolate site.[10] In 2010, Shibasaki and co-workers
disclosed the first direct asymmetric vinylogous aldol reaction
of acyclic allyl cyanide to ketones, catalyzed by a soft Lewis
acid/hard Brønsted base, with excellent enantioselectivities
for the quaternary center and (Z) selectivity for the alkene
moiety.[11] We envisioned that allyl ketones, which are similar
to allyl cyanide, but are more readily prepared and modified,
could be used as the precursors of D to achieve the desired
direct vinylogous aldol reaction. It is worth noting that, over
the past few years, allyl ketones have been widely employed
as valuable intermediates in organic synthesis,[12] but that
there is no reported example on their applications in catalytic
asymmetric reactions. Herein, we report a highly enantio- and
E-selective direct vinylogous aldol reaction of allyl ketones
with isatins, and the divergent synthesis of various related 3-
hydroxy-2-oxindole derivatives with important biological
activities.
At the onset of our studies to probe the feasibility of the
proposed strategy, allyl phenyl ketone 8a was reacted with N-
methyl isatin 9a in the presence of Et3N in toluene at 258C.
The reaction worked smoothly, and the proposed vinylogous
aldol product 10a, which bears an E-configured olefin, was
obtained as the sole adduct (Table 1, entry 1). Encouraged by
our preliminary results, we examined the enantioselective
variant in the presence of various bifunctional tertiary amine/
thiourea catalysts (CAT-1–CAT-5), which can be conveniently
prepared from l-amino acids and are effective in a variety of
asymmetric reactions, as reported by us and other
groups.[6c,8d,13] Our screening of reaction conditions revealed
that all reactions proceeded well within 3–5 h in toluene at
258C in the presence of 10 mol% of catalyst, and afforded the
chiral vinylogous aldol E-adduct 10a in high yields with
moderate to good enantioselectivities (Table 1, entries 2–6).
The best selectivity was achieved with catalyst CAT-3, which
was derived from l-valine (Table 1, entry 4, 88% ee). To
further optimize the reaction conditions, the reaction was
performed in different solvents with CAT-3 as the catalyst
(Table 1, entries 7–9),[14] and the use of Et2O resulted in the
highest enantioselectivity with enhanced reaction rates
(Table 1, entry 8). A variation of the reaction temperature
showed that the enantioselectivity improved only slightly at
a lower temperature (Table 1, entries 10–12), and that the
best results were achieved at ꢀ208C (entry 11). Moreover, the
catalyst loading could be reduced to 1.0 mol%, and after 12 h,
10a was isolated with an excellent yield and without
compromising the enantiomeric excess (Table 1, entry 14).
With the optimal conditions established, we investigated
the scope of the reaction (Table 2). First, we attempted the
direct vinylogous aldol reaction of a variety of allyl ketones
with N-methyl isatins in the presence of 10 mol% of CAT-3 in
Et2O at ꢀ208C (Table 2, entries 1–14). The corresponding
products 10a–n were obtained in 57–98% yields with 88–
96% ee and E/Z ratios of 19:1 to more than 99:1. Notably, the
vinylogous aldol product 10g, derived from allyl tert-butyl
ketone, was obtained with 89% ee (entry 8). Subsequently, we
examined the vinylogous aldol reaction of allyl ketones with
N-benzyl isatins, and the desired products 10o–q were
obtained with excellent yields, enantio- and diastereoselec-
tivities (entries 15–17). We also found that simple filtration
was enough to isolate these E-configured adducts with an
ee value of 98% to more than 99%.[15] Furthermore, various
other allyl ketones 8 and commercially available isatins 9
were reacted under the established conditions to give the
vinylogous aldol adducts (10r–z) with yields of 81–92%,
enantioselectivities of 90–96%, and E/Z ratios of 13:1 to more
than 99:1 (entries 18–26). These results demonstrated that the
best reaction outcome was achieved with unprotected isa-
tins.[2,16] Most importantly, we carried out the reactions
toward 10d and 10x on a gram scale, and found that the
reactivity and enantioselectivity remained excellent, and the
corresponding adducts were obtained after a simple filtration.
Since all adducts precipitate from their reaction mixture, this
methodology is suitable for large-scale production.
Table 1: Screening studies.[a]
Entry
Catalyst
Solvent
T [8C]
t [h]
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
7
8
Et3N
toluene
toluene
toluene
toluene
toluene
toluene
hexane
Et2O
THF
Et2O
Et2O
Et2O
25
25
25
25
25
25
25
25
25
12
3
3
3
3
88[d]
89
91
86
82
83
99
99
99
99
99
84
93
96
n.a.
57
85
88
79
85
85
88
87
90
93
93
92
92
CAT-1
CAT-2
CAT-3
CAT-4
CAT-5
CAT-5
CAT-5
CAT-5
CAT-5
CAT-5
CAT-5
CAT-5
CAT-5
Density-functional theory (DFT) was employed to eluci-
date the observed enantioselectivity and preference for g over
a alkylation.[17] We suspected that the pendent pyrrole moiety
of bifunctional catalyst CAT-3 first deprotonates allyl ketone
8a at the g position, and that the protonated catalyst then
brings the resulting enolate and the isatin electrophile
together to form the hydrogen-bonded complex for the
5
1
0.1
0.2
0.5
2
12
4
9
10
11
12
13[e]
14[f]
0
ꢀ20
ꢀ40
ꢀ20
ꢀ20
ꢀ
crucial C C bond formation. We found that binding of
ꢀ
a) the electrophile to the two N H bonds of the thiourea
Et2O
Et2O
moiety and b) the enolate to the pyrrolium arm gives the
transition state (TS) with the lowest Gibbs free energy (Si-R-
TS), which leads to the g-alkylated R product, consistent with
our experimental observation (see the Supporting Informa-
tion for all structures). Upon a close analysis of the Mayer
bond order,[18] this preference of the thiourea to be H-bonded
12
[a] Reactions were performed with 8a (0.075 mmol), 9a (0.05 mmol),
and catalyst (0.005 mmol) in 0.5 mL solvent. [b] Yield of isolated product
1
(E/Z>99:1, determined by H NMR analysis of the crude reaction
mixture). [c] Determined by HPLC. [d] E/Z=49:1. [e] 5 mol% catalyst
was used. [f] 1 mol% catalyst was used. n.a.=not applicable.
2
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!