Angewandte
Communications
Chemie
Table 2: Reaction scope.[a]
a large amount of by-product was detected and the starting
materials were completely consumed. Further optimization of
the Yb(OTf)3 system failed to achieve a better result.[6]
We then employed a variety of ligands aimed at regulating
the Lewis acidity of the catalyst. As shown in Table 1, using
the bipyridine ligand (L1) with Cu(SbF6)2 only gave the
desired product in 31% yield with 78:22 d.r. (entry 1), and
was accompanied by a lot of the Friedel–Crafts alkylation by-
product.[5] Bisoxazoline (BOX; L2) also failed to improve the
yield and diastereoselectivity (entry 2). When changing to L3
as the ligand, the desired product was afforded in 42% yield
with 80:20 d.r. (entry 3). It was found that a bulkier R group
led to a better d.r. value. When R was changed to 2,6-
dimethylbenzyl, 88:12 d.r. was obtained, albeit still with
moderate yield (entry 4). Further screening of the reaction
conditions[6] showed that the side reactions could be greatly
suppressed by using 4 ꢀ M.S. as additives. By a combination
of optimal reaction conditions, including temperature, sub-
strates ratio, etc.,[6] finally, 90% yield with 90:10 d.r. was
achieved (entry 5).
Entry
Product
t [h] Yield
d.r.[c]
[%][b]
1
2
3
4
5
6
7
8
X=H (3a)
X=5-Me (3b)
X=5-F (3c)
X=5-Cl (3d)
X=5-Br (3e)
X=6-Me (3 f)
X=7-Me (3g)
X=7-MeO (3h)
11
18
17
18
18
17
17
18
90
97
85
92
90
79
87
57
90:10
88:12
83:17
88:12
90:10
88:12
88:12
83:17
9[e]
10[e]
11[e]
12[e]
13[e]
X=H (3i)
X=5-Me (3j)
X=5-MeO (3k)
X=H (3a)
17
20
20
17
21
71
81
61
80
74
91:9
88:12
91:9
91:9
91:9
X=5-F (3l)
The substrate scope was investigated under the optimized
reaction conditions. As shown in Table 2, a variety of indoles,
including those containing both electron-rich and electron-
poor substituents on the aryl ring, reacted smoothly with 2b to
give the desired products in up to 97% yield with up to 90:10
d.r. (entries 2–8). It is noteworthy that indoles containing
various functional groups at the C3-position, like allyl, benzyl,
hydroxy, and amino groups, were also tolerated (entries 9–
17). Considering the difficulty of removing the 2,6-dimethyl-
benzyl group, the TBSO-substituted cyclobutane 2c was
employed, thus delivering the products that were suitable for
further transformation (entries 18 and 19). Additionally,
a tryptophol-derived substrate reacted smoothly, thus afford-
ing the product 3t in 76% yield with 83:17 d.r. (entry 20).
Notably, with the PMB-protected tryptamine as a substrate,
and Cu(PF6)2 as a Lewis acid, the desired product 3u was
furnished in 50% yield with > 99:1 d.r. (entry 21).
14[e]
15[e]
R1 =Bn (3n)
18
19
82
53
92:8
>99:1[d]
16[e]
17[e]
25
27
72
64
90:10
93:7
18[e]
19[e]
R1 =Me (3r)
18
42
54
46
84:16
89:11
20[e]
21[f]
(3t)
6
77
50
86:14
(3u)
32
>99:1
Further studies showed that the asymmetric version of this
method can also be realized. Although cyclobutanes with
oxygen-donating groups only gave poor enantioselectivity,[6]
cyclobutanes with sulfur-donating groups resulted in excellent
ee values. As shown in Scheme 2, in the presence of a chiral
the side-arm-modified BOX (SaBOX) ligand L4, a variety of
enantiotopic chiral cyclohexa-fused indolines (4a–e) were
furnished with excellent enanatioselectivity (90–94% ee)
through the enantioselective [4+2] annulation of indoles
with the p-MeO-phenylthio-substituted cyclobutane 2d.[7,8]
Many easily transformed functional groups on the indoles
were tolerated by the current catalyst system, including an
allyl group at the 3-position and halo substituents at the 5- and
6-positions.
The [3+2] annulation reactions of D-A cyclopropanes
have been successfully applied in the total synthesis of many
natural products.[9,10] Although the reactions of D-A cyclo-
butanes have been well studied, few of them were employed
in the total synthesis of complex molecules. Inspired by the
synthetic efficiency of forming cyclohexa-fused indolines with
the current method, we tried to employ it to construct the
basic skeletons of some well-known members of the strych-
[a] Reaction conditions: 1/2=2:1, 1 (0.4 mmol), 2 (0.2 mmol), AgSbF6
(0.04 mmol), CuBr2 (0.02 mmol), and L3 (0.024 mmol) in DCM (3 mL).
[b] Yield of isolated product. [c] Determined by 1H NMR analysis of the
crude reaction mixture unless otherwise noted. [d] The d.r. value was
determined by 1H NMR analysis of the isolated products. [e] At À408C.
[f] L3/Cu(PF6)2 was used, at À508C. R3 =2,6-dimethylbenzyl; Bs=ben-
zenesulfonyl, M.S.=molecular sieves, Ns=4-nitrobenzenesulfonyl,
TBS=t-butyldimethylsilyl, TMSE=2-(trimethylsilyl)ethyl, Ts=4-methyl-
benzenesulfonyl.
nos, aspidosperma, and kopsia alkaloids.[4,11] As shown in
Scheme 3, the [4+2] annulation reaction of PMB-protected
tryptamine with D-A cyclobutane was applied as the key step
to construct 3u with excellent diastereoselectivity. The
compound 3u was then deprotected, followed by an intra-
molecular Mitsunobu reaction, thus affording the cyclization
product 6 in excellent yield. The compound 6 was then treated
with LiCl to give the decarboxylated product 7 in 92% yield.
However, trials on converting 7 into the unsaturated ester by
a PhSeBr/H2O2 oxidation failed. Thus, we treated 7 with Na/
naphthalene to give the deprotection product, which was in
turn converted into the Boc-protected product 8, a compound
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
These are not the final page numbers!