Angewandte
Chemie
ently pioneered the catalytic asymmetric 6p electrocycliza-
tion.[12] Whereas List et al. developed a highly enantioselec-
tive cycloisomerization of a,b-unsaturated hydrazones to
pyrazolines catalyzed by a chiral phosphoric acid,[12a] Smith
and co-workers used functionalized aldimines as precursors to
2-aza-pentadienyl anions while employing phase transfer
catalysis (PTC) to generate a tight ion pair featuring
a chiral ammonium cation to achieve excellent enantioselec-
tivity.[12b] Smith et al. further used their method to develop
cascade syntheses of polycyclic compounds.[12c] Despite sig-
nificant advances, the use of functionalized ketimines for such
reactions is still unknown, and the identification of new
catalysts for this interesting reaction is still very much in
demand.
a ratio of approximately 10:1. Nevertheless, it was directly
used for the optimization of the reaction conditions. First, we
attempted the electrocyclization of 4a under Smithꢀs con-
ditions using chiral ammonium catalyst C1 along with K2CO3
as the base and toluene as the solvent at À158C; however, the
desired product 5a was obtained in only 6% ee, albeit in 89%
yield (Table 1, entry 1). This result further suggested the
Table 1: Optimization of the reaction conditions.[a]
Considering the versatility of bifunctional tertiary amine/
hydrogen bond donor catalysts,[13] we speculated that such
a dual activation mode would be suitable for the asymmetric
[1,5] electrocyclic reaction of aniline-derived ketimine 4,
which might be activated to produce 2-aza-pentadienyl anion
intermediates through activation of the ketimine moiety
through hydrogen bonding and simultaneous activation of the
methine moiety through deprotonation. The hydrogen bond-
ing interactions might organize a favorable transition state to
control the sense of the disrotatory electrocyclization process
(Scheme 1).
Entry
Catalyst
Solvent
t [d]
Yield[b] [%]
ee[c] [%]
1[d]
2
3
4
5
6
7
8
9
C1
toluene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
EtOAc
EtOH
toluene
THF
1
6
6
4
3
3
3
3
3
3
2
89
trace
trace
78
63
88
94
93
46
90
6
49
16
98
95
97
65
98
97
99
98
(DHQD)2PYR
C2
C3
C4
C4
C4
C4
C4
C4
C4
10
Et2O
Et2O
11[e]
88
Importantly, with bifunctional tertiary amines as the
catalyst, catalyst quenching, a major obstacle in the develop-
ment of the desired tandem process, may be avoided.[1] We
were also concerned that the Brønsted acid required for the
synthesis of the aniline-derived ketimines might be detrimen-
tal to the electrocyclization, as it might protonate the tertiary
amine moiety of the chiral catalyst. However, recent results
from Wang and co-workers and our group revealed that
bifunctional chiral tertiary amines could tolerate the coex-
istence of some acidic additives.[14] Furthermore, bifunctional
tertiary amines are known to be able to tolerate some metal
species,[6k–n] and might be compatible with the palladium
species we wished to use for the hydrogenation of nitroarenes
1 to obtain malonate-anilines 3. It is important to conduct the
ketimine formation and the cyclization process in a one-pot
fashion as the purification of such sensitive imines often
incurs yield losses.[15] Based on the above analyses, it is
worthwhile to combine the three distinct catalytic reactions
into the one-pot process shown in Scheme 1.
[a] Reactions run on a 0.1 mmol scale in 1.0 mL of solvent. [b] Yield of
isolated product. [c] Determined by HPLC analysis on a chiral stationary
phase. [d] With 0.2 mL of aqueous K2CO3 (33%). [e] At 408C.
importance of identifying a new catalyst motif for this
valuable cyclization. In the following, various tertiary amine
catalysts were evaluated at 258C using CH2Cl2 as the solvent.
Not surprisingly, simple chiral tertiary amines proved to be
incapable of catalyzing this transformation. For example,
(DHQD)2PYR (hydroquinidine-2,5-diphenyl-4,6-pyrimidine-
diyl diether) gave only trace amounts of 5a in 49% ee even
after six days (entry 2). On the other hand, it was also
important for the desired reactivity and enantioselectivity
that the amine catalyst featured a suitable hydrogen bond
donor moiety. Cinchona alkaloid derivatives with a single
hydrogen bond donor were inferior as well, and derivative C2,
for instance, afforded product 5a in trace amounts and 16%
ee (entry 3). Fortunately, cinchona alkaloid derivatives with
a dual hydrogen bond donor moiety, squaramide C3[19] or
thiourea C4,[20] were promising, giving 5a in reasonable yield
and ꢀ 95% ee (entries 4 and 5; for the full catalyst evaluation,
see the Supporting Information, Table S1). Encouraged by
these results, we chose the more easily available thiourea
catalyst C4 for further optimization. By replacing CH2Cl2
With our interests in oxindole chemistry,[10,16] we first
designed isatin-derived malonate-ketimine 4a as a starting
point for the study of the asymmetric 6p electrocyclization, to
pave the way for the subsequent development of the tandem
reaction. It should be noted that whereas the catalytic
asymmetric synthesis of spirocyclic oxindoles, a prominent
structural motif in natural products, drugs, and bioactive
products,[17] has been extensively studied, the asymmetric
synthesis of spirocyclic oxindole-indoline derivatives is
unprecedented.
The synthesis of ketimine 4a from malonate-aniline
derivative 1a and N-methylisatin (2a) required the use of
a stronger acid catalyst,[18] namely para-toluenesulfonic acid
(p-TsOH; see Supporting information). Ketimine 4a was
prepared as an inseparable mixture of the Z and E isomers in
Angew. Chem. Int. Ed. 2014, 53, 13740 –13745
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim