Organic Letters
Letter
Scheme 4. Scale-up Reaction and Derivatization of the
Product 2a
Scheme 6. Plausible Mechanism
Scheme 5. Construction of the Core Skeleton 9 of PARP-1
a
Inhibitor C
compound 1s was successfully transformed to the product 2s
in 76% yield. However, the reaction failed when the C3
position of the indole ring was substituted with a phenyl ring.
Notably, the substrate 1t containing a 6-azaindole unit could
be converted to the expected product 2t in 58% yield as well.
Then, a scale-up reaction and a series of derivatization of the
product 2a were performed to demonstrate the utility of this
method (Scheme 4). When the N-(2-(1H-indol-1-yl)phenyl)-
picolinamide 1a (1.0 mmol) was subjected to the standard
reaction conditions, a 72% yield of the product 2a was
achieved smoothly. The treatment of 2a with NaH at 0 °C
followed by the addition of alkyl halides could provide various
N-alkyl-substituted indolo[1,2-a]quinoxalin-6(5H)-ones 3.
The reaction of 2a with methyl iodide gave a good yield
(77%) of compound 3a. When 2a was reacted with allyl
bromide and propargyl bromide, compounds 3b and 3c were
obtained in 62 and 65% yield, respectively.
a
Reagents and conditions: (a) Cs2CO3, DMF, 60 °C, 4 h. (b) Fe,
NH4Cl, H2O, 4 h. (c) DMAP, picolinic acid, EDCI, DCM, 12 h. (d)
CoCl2, Ag2CO3, PivOH, TFBen, Et3N, dioxane, 130 °C, 20 h.
results are shown in Scheme 3. For compounds bearing either
electron-donating or -withdrawing substituents at the C4
position of the benzene ring, the reaction afforded the desired
products 2b−d in 57−65% yields. It was found that the
reaction of compounds with substituents at the C5 position
proceeded well to give the products 2e−f in high yields.
Surprisingly, only trace product 2g was observed when the
substrate 1g with a methyl group at the C6 position was tested.
Here one possible reason for this phenomenon is the steric
effect from the ortho-methyl group, which makes the indole
ring and the aniline not in the same plane and leads to
difficulty in forming a new C−N bond. In addition, when
compounds had substituents such as Me, OMe, Cl, and
COOMe at the C6 position of the indole ring, the reaction
gave the corresponding products 2h−k in moderate to good
yields (48−88%). It was shown that good yields (65−88%) of
the products 2l−o were obtained when substrates with
functional groups at the C5 position were subjected to the
reaction system. Also, C4-substituted compounds could
undergo the reaction smoothly to give the products 2p−r in
high yields (56−82%). Furthermore, the C3-Me-substituted
Furthermore, the core skeleton of PARP-1 inhibitor C can
be easily established by this protocol as well (Scheme 5). The
coupling reaction of pyrrole 4 and the fluoride 5 led to the
formation of the nitro compound 6 in 65% yield. The
subsequent reduction of 6 with iron gave an excellent yield
(90%) of the amine 7, which was then reacted with picolinic
acid to access the picolinamide 8 in 75% yield. Gratifyingly, the
direct C−H carbonylative reaction of 8 under our standard
conditions could successfully construct the crucial pyrrolo[1,2-
a]quinoxalin-4(5H)-one skeleton 9 in 63% yield. Finally,
according to the known procedures,3 PARP-1 inhibitor C can
be synthesized via the reduction of 9 followed by amination.
On the basis of our results and previous reports,14,15
a
plausible mechanism for this cobalt-catalyzed C−H carbon-
ylation of N-(2-(1H-indol-1-yl)phenyl)picolinamides is pro-
posed as shown in Scheme 6. Initially, the coordination of the
Co(II) catalyst with the N-(2-(1H-indol-1-yl)phenyl)-
picolinamide 1a followed by the oxidation of the Ag(I) salt
forms the Co(III) species A′. Then, C−H bond activation at
the C2 position of A′ generates the Co(III) complex B′.
Subsequently, the insertion of CO that is released from TFBen
C
Org. Lett. XXXX, XXX, XXX−XXX