Organic Letters
Letter
moderate to good yields. Various substituents on the nitrogen
atom, such as 4-MeBn, 4-FBn, Me, Et, Ph, and allyl groups, were
well-tolerated, and the desired products 3ba−ga were afforded
in 30−75% yields. Importantly, imide 1h with a phenyl group on
the alkene was able to undergo this cascade cyclization to furnish
product 3ha in 48% yield. Regarding the 2-iodobenzamide
moiety, a range of functional groups (Me, OMe, F, and Cl) in an
ortho or meta position relative to the amide group all could
survive, furnishing products 3ia−na in 46−85% yields.
However, imide 1o derived from 2-iodo-4-methylbenzamide
was not a viable substrate. The results indicated that the steric
effect had a vital influence on the construction of a seven-
membered ring. Substrates 1p and 1q containing only one amide
group were also tested, and 1q showed poor reactivity.
Remarkably, substrate derived from 2-bromobenzamide was
perfectly accommodated to provide product 3aa in 70% yield.
The scope of 2-(2-bromoaryl)-2-oxoacetic acids was then
explored (Scheme 2). A range of 2-oxocarboxylic acids 2b−k
were subjected to the optimal reaction conditions to furnish
isoquinolinedione-fused benzocycloheptanones 3ab−ak in 61−
87% yields. Various substituents on the benzene ring, including
both electron-donating (Me, OMe, and OPh) and electron-
withdrawing groups (F and Cl), were well-tolerated (3ab−ah).
Their electronic property and position had no obvious effect on
the reaction efficiency. Note that morpholinyl-, dimethyloxy-, or
dioxolyl-substituted 2-(2-bromoaryl)-2-oxoacetic acids could be
smoothly converted into products 3ai−ak. Gratifyingly, 2-(3-
bromothiophen-2-yl)-2-oxoacetic acid also exhibited good
reactivity, providing product 3al in 45% yield. Furthermore, 2-
(2-iodophenyl)-2-oxoacetic acid was also a suitable substrate.
However, only 35% of 3aa was isolated due to the formation of
the decarboxylation/decarbonylation side product.9d Unfortu-
nately, this protocol was not compatible with 2-(2-chlorophen-
yl)-2-oxoacetic acid.
indole could also furnish product 5aa, albeit with a lower yield.
Subsequently, a series of 2-(2-iodophenyl)-1H-indoles 4b−i
were examined. Substrates 4b−f bearing diverse substituents
(Me, F, Cl, and Br) on the indole undergo this cascade reaction
with 2a to deliver the target products 5ba−fa in 54−80% yields.
2-Phenylacryl- or 2-methylallyl-substituted indoles were also
competent substrates, affording products 5ga and 5ha in
moderate yields. Notably, the reaction of 2-(3-bromothio-
phen-2-yl)-2-oxoacetic acid with 4a proceeded efficiently,
delivering 51% of the target product 5al. Unfortunately, 2-(2-
iodophenyl)-1H-benzo[d]imidazole 1i was unreactive.
To gain insights into the reaction mechanism, the isotope
experiment was carried out by adding 5 equiv of D2O under the
optimal reaction conditions. The result indicated that pallada-
cycle C was formed in the reaction process. On the basis of our
experimental results and previous work,13 a plausible catalytic
cycle for this transformation was illustrated as Scheme 4. First,
Scheme 4. Control Experiment and Possible Reaction
Mechanism
To highlight the applicability of this cyclization cascade, we
next attempted the feasibility of the reaction of alkene-tethered
2-(2-iodophenyl)-1H-indoles 4a with 2-oxocarboxylic acids 2a
(Scheme 3). To our delight, indolo[2,1-a]isoquinolinone-fused
benzocycloheptanone 5aa was formed in 78% yield under the
above reaction conditions. Notably, the replacement of an
iodine atom with a bromine atom on the 2-(2-iodophenyl)-1H-
oxidation addition of the C−I bond to Pd(0) followed by
intramolecular carbopalladation forms intermediate B, which
undergoes C−H activation to generate fused palladacycle C.
Then, palladacycle C undergoes an oxidative addition with the
C−Br bond of 2-oxocarboxylic acid 2a to afford Pd(IV) species
D with the assistance of an ortho-chelating carboxyl group, which
has been demonstrated by the related reports.13 Immediately
after, sequential reductive elimination and decarboxylation of
intermediate D produces eight-membered palladacycle E.
Eventually, this catalytic cycle is accomplished by reductive
elimination of palladacycle E, thus resulting in the formation of
products 3 or 5 as well as active Pd(0) species.
Scheme 3. Synthesis of Indolo[2,1-a]isoquinolinone-Fused
a
Benzocycloheptanones
In conclusion, we have disclosed an efficient method for
synthesizing diverse fused heteropolycycles via a palladium-
catalyzed decarboxylative cascade cyclization of aryl iodides,
including alkene-tethered 2-iodobenzamides and 2-(2-iodo-
phenyl)-1H-indoles. In this novel conversion, 2-oxocarboxylic
acids are employed as coupling reagents to achieve three-carbon
unit insertion via an intramolecular Heck/C−H activation/
decarboxylation sequence, thus forging isoquinolinedione- and
indolo[2,1-a]isoquinolinone-fused benzocycloheptanones in
moderate to good yields. Remarkably, a seven-membered ring
is constructed in this reaction. Further applications of 2-
oxocarboxylic acids are underway in our laboratory.
a
Reaction conditions: 4 (0.2 mmol), 2a (1.5 equiv), PdCl2 (10 mol
%), PCy3 (20 mol %), K3PO4 (3 equiv), and DMF (2 mL) at 120 °C
under N2 for 12 h. Alkene-tethered 2-(2-bromophenyl)-1H-indole
was used.
b
2880
Org. Lett. 2021, 23, 2878−2883