.
Angewandte
Communications
led to a major increase in yield to 91% (entry 13). A silyl
ether could also be used,[10b] but led to a lower yield (80%,
entry 14). A bromide counterion led to a slightly higher yield
(entry 15) and iodide completely suppressed the reaction
(entry 16). The results obtained with PtCl2 and PtBr2 were
difficult to reproduce, probably because of the poor solubility
of 4a in CH3CN. Using a cosolvent led to more reproducible
results (entries 17–19), and THF was the best, thus giving the
desired product 8a in 96% yield (entry 18).
a benzyl group were well tolerated (entries 2 and 3), as well as
different substituents on an alkyl chain such as methoxy and
trifluoromethyl (entries 4 and 5). Substitution of the alkyne
was examined next. The desired C5-alkynylated product 8 f
was obtained in 70% yield with a methyl group (entry 6).
With aromatic substituents, formation of the C6-alkynylated
product 10 was observed for the first time. The C5-alkyny-
lated product 8 was still the major product with either
a
phenyl, a tolyl, or a para-bromophenyl substituent
With optimized reaction conditions in hand, we then
examined the scope of the cyclization/ethynylation domino
reaction (Table 2). The indole 8a was obtained in 84% yield
upon isolation from a 0.3 mmol scale reaction (entry 1). Only
C5-ethynylation was observed. Modification of the substitu-
ent on the nitrogen atom was investigated first. A tolyl and
(entries 7–9). The latter case also demonstrated that halogen
substituents were tolerated under the reaction conditions,
thus allowing further functionalization of the products by
classical cross-coupling chemistry. With a naphthyl group,
a 1:1 mixture of regioisomers was obtained (entry 10), and
with a methyl substituent in the propargylic position, the C5-
alkynylated product 8k was obtained in 75% yield (entry 11).
In contrast, when the tertiary ether 6l was used, the C6-
alkynylated product 10e was formed (entry 12). The alkyny-
lated carbazole 8m was obtained in 76% yield (entry 13), and
the domino process could also be applied to the synthesis of
the thiophene-fused heterocycle 8n in 78% yield (entry 14).
Having achieved the synthesis of C5-alkynylated indoles
starting from C2-substituted pyrroles, we then turned to C3-
substituted derivatives to achieve C6 functionalization.
(Table 3). The pyrrole 9a, bearing a terminal alkyne, gave
the C6-alkynylated indole 10a exclusively in 77% yield
(entry 1). Substituted alkynes could also be used to access 6,7-
disubstituted indoles in 47–53% yield (entries 2–4). Starting
from tertiary ethers, 4,6-disubstituted products could be
obtained (entries 5 and 6). In this case, the same products
were obtained as when starting from C2-substituted pyrroles.
We then wondered if more-complex trisubstituted indoles
could be accessed. In fact, the 4,6,7- (10g) and 4,5,6-
substituted (10h) indoles were obtained in 44 and 34%
yield, respectively (entries 7 and 8). Although the yields are
moderate, the synthesis of these polysubstituted compounds
using another method would have been extremely difficult.
Finally, the synthesis of the C6-alkynylated carbazoles 10i,j
was also possible (entries 9 and 10).
Table 2: Scope of 5-alkynylated indoles.
Entry Substrate
Product
Yield [%][a]
1
2
3
4
5
R=Me (6a)
R=tolyl (6b)
R=benzyl (6c)
R=3-methoxy propyl (6d) 8d
R=1,1,1-trifluorobutyl (6e) 8e
8a
8b
8c
84
81
83
80
82
6
7
8
9
10
R=Me (6 f)
R=phenyl (6g)
R=tolyl (6h)
R=p-Br-phenyl (6i)
R=naphthyl (6j)
8 f
70
8g/10c (4:1)
8h/10k (3:1)
8i/10l (2.5:1)
8j/10d (1:1)
70 (56)[b]
69 (52)[b]
62 (44)[b]
31/31
The developed domino process is highly complex. Con-
sequently, any proposal about the reaction mechanism is
highly speculative at this stage of development. Based on the
well-established p-acidity of platinum,[11a] the first step of the
catalytic cycle is most probably activation of the alkyne to
give complex A (Scheme 2). At this point, cyclization could
occur either at the C2- or C3-position of the pyrrole ring to
give intermediates B or D, respectively. Attack on the more
nucleophilic C2 position, thus leading to B is more probable.
From B, two different pathways are possible: a 1,2 shift of the
platinum-bearing vinyl group via the platinum carbene cyclo-
propane intermediate C[11a,15] to give the carbocation D, or
1,2-shift of the alkoxy substituent to give intermediate G.[10]
C5- and C6-metalated intermediates E and H, respectively,
will then be generated from the corresponding compounds D
and G after deprotonation and aromatization by methanol
elimination. The alkynylated products 8 and 10 can be formed
by reaction of 4a with organoplatinum intermediates. At this
point, it is not clear if the alkynylation step proceeds through
an oxidative addition/reductive elimination mechanism on
11
12
75
54
13
14
76
78
[a] Reaction conditions: Substrate 6 (0.30 mmol), PtBr2 (10 mol%), EBX
reagent 4a (0.60 mmol), NaHCO3 (0.60 mmol), THF/CH3CN (3 mL, v/
v=5:1), 72–120 h. Yield is that of the product isolated after column
chromatography. [b] Yield determined by 1H NMR spectroscopy. The
yield of the isolated isomer 8 is given within parenthesis.
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 5438 –5442