A domino amidation route to indolines and indoles: Rapid syntheses of anhydrolycorinone, hippadine, oxoassoanine, and pratosine

Article abstract of DOI:10.1021/ol052086c

(Chemical Equation Presented) When subjected to palladium-catalyzed amidation conditions, 2-triflyloxy phenethyl carbonates undergo, in addition to the expected aryl cross-coupling, an additional amidation with net displacement of the carbonate. The result is a one-step synthesis of indolines which may be oxidized to indoles. The utility of the procedure is illustrated by the two- or three-step syntheses of anhydrolycorinone, hippadine, oxoassoanine, and pratosine.

Full text of DOI:10.1021/ol052086c

ORGANIC
LETTERS
2005
Vol. 7, No. 21
4777-4779
A Domino Amidation Route to Indolines
and Indoles: Rapid Syntheses of
Anhydrolycorinone, Hippadine,
Oxoassoanine, and Pratosine
Michael D. Ganton and Michael A. Kerr*
Department of Chemistry, The UniVersity of Western Ontario,
London, Ontario, Canada N6A 5B7
makerr@uwo.ca
Received August 30, 2005
ABSTRACT
When subjected to palladium-catalyzed amidation conditions, 2-triflyloxy phenethyl carbonates undergo, in addition to the expected aryl cross-
coupling, an additional amidation with net displacement of the carbonate. The result is a one-step synthesis of indolines which may be
oxidized to indoles. The utility of the procedure is illustrated by the two- or three-step syntheses of anhydrolycorinone, hippadine, oxoassoanine,
and pratosine.
The recent development of palladium-catalyzed aryl C-N
cross-coupling reactions has revolutionized organic synthe-
sis.1 The more recent use of palladium-catalyzed C-N bond-
forming reactions in domino processes has allowed for the
efficient construction of nitrogen-containing heterocycles,
such as indolines and indoles.2
catalyzed “stitching” of a nitrogen atom to prepare indolines
(and possibly indoles) would be a useful method for the
synthesis of alkaloids and other indole containing targets.
During studies directed toward the synthesis of CC-1065
and related compounds, we explored the installation of the
dihydropyrrole subunits via the amidation of a 5-triflyloxy-
indole such as 1 with benzamide 2. The intention was to
convert the acetate in 3 to a good leaving group and effect
the ring formation to 4 in a nucleophilic manner. We were
surprised to isolate the pyrroloindole 4 as the major product
from the cross-coupling reaction and 3, especially since the
same reaction involving halides in place of the acetoxy
moiety (a poorer leaving group) led to decomposition. We
decided to explore this unusual reaction since the palladium-
Scheme 1 shows an optimization study in which the
aliphatic “leaving group” was varied and the catalyst loading
was probed for its lower limit. While the chloride 5a and
bromide 5b produced appreciable amounts of the desired
compound, the yield was low, and extensive decomposition
was observed. The iodide 5c failed to produce identifiable
products. The mesylate 5d produced, along with the desired
(1) For reviews, see: (a) Schlummer, B.; Scholz, U. AdV. Synth. Catal.
2004, 346, 1599. (b) Muci, A. R.; Buchwald, S. L. Top. Curr. Chem. 2002,
219, 131. (c) Hartwig, J. F. Modern Arene Chemistry 2003, 107.
(2) For recent examples, see: (a) Poondra, R. R.; Turner, N. J. Org.
Lett. 2005, 7, 863. (b) Lira, R.; Wolfe, J. P. J. Am. Chem. Soc. 2004, 126,
13906. (c) Larock, R. C.; Yum, E. K.; Refvik, M. D. J. Org. Chem. 1998,
63, 7652. (d) Aoki, K.; Peat, A. J.; Buchwald, S. L. J. Am. Chem. Soc.
1998, 120, 3068.
10.1021/ol052086c CCC: $30.25
© 2005 American Chemical Society
Published on Web 09/22/2005

Table 1. Substrate Scope4
Scheme 1. Variation of Leaving Group and Catalyst Loading
indoline 6, the heterocycle 7 in which participation by the
amide oxygen occurred. The acetate 5e and benzoate 5f gave
both the indoline 6 and the uncyclized amidation products 8
and 9. The carbonate 5g proved to be the ideal substrate
giving a near-quantitative yield of 6 with catalyst loadings
of as low as 2.5 mol %. Only when the loading was dropped
to 0.5 mol % did the yield drop to an unacceptable level.
To investigate substrate scope using optimized conditions,
a series of compounds bearing an aryl triflyloxy group and
a pendant alkyl carbonate was subjected to the domino
amidation sequence (Table 1). Several examples are worthy
of note. In all cases, the aryl amidation was seen to occur
first, and if the reaction was stopped an early stage, the
uncyclized anilide could be isolated implying that the
N-arylation is a distinct event and is followed by an
intramolecular N-alkylation. The amido component seems
to be general (entries 1-4); benzamide acetamide, N-meth-
ylindole-2-carboxamide, and methyl carbamate gave good
results. Substitution on the aryl moiety of the triflate
component was tolerated, although both examples which had
substitution ortho to the triflyloxy group gave diminished
yields (entries 5 and 9). Surprisingly, extension of the alkyl
chain by one carbon in the hopes of preparing tetrahydro
quinolines met with little success (entry 10). Although a trace
amount of the cyclized material was isolated, the major
product was the uncyclized material 18, in which only aryl
amidation was observed.
styrene. This would explain the failure of the formation of
the six-membered ring (Table 1, entry 10). However,
treatment of 19 (Scheme 2) under the optimized reaction
conditions for indoline formation resulted in almost complete
recovery of starting material, indicating that a styrene was
not a likely intermediate in the indoline formation.
While the amidation of the aryl ring was certainly
expected, the subsequent cyclization puzzled us, especially
since the best substrates by far were the carbonates and not
alkyl halides.3 It occurred to us that the cyclization may also
be a palladium-mediated event and furthermore that the
cyclization was proceeding through a hydroamination of a
Scheme 2. Probing the Cyclization Step
(3) The alkylation of anilides with carbonates is very rare but known.
See, for example: Selva, M.; Tundo, P.; Perrosa, A. J. Org. Chem. 2001,
66, 667 and references therein.
To determine whether the cyclization of the benzanilide
to the N-benzoylindoline was a palladium-catalyzed reaction,
4778
Org. Lett., Vol. 7, No. 21, 2005

the amidocarbonate 20 (Scheme 2) was prepared. While
treatment of 20 under the palladium-catalyzed conditions
shown in Scheme 1 resulted in near-quantitatiVe conVersion
to 6, the same results were obtained when the palladium
catalyst and phosphine were remoVed from the reaction.
Clearly, the second process (the intramolecular amidation)
is not a palladium-catalyzed process and is most likely a
nucleophilic displacement (albeit an unusual one with a
carbonate leaving group) promoted by the Cs2CO3.5
Scheme 3. Syntheses of Pyrrolophenanthridone Natural
Products
With an optimal procedure for the formation of indolines
from the triflyloxy carbonates, we undertook to showcase
this methodology with the synthesis of several simple
Amaryllidaceae pyrrolophenanthridone natural products
(Scheme 3).6 Treatment of 5g with piperonylamide 21 or
veratramide 22 under the standard conditions gave indolines
23 and 24 in 85 and 97% yields, respectively. Oxidative
cyclization of 23 under the influence of phenyliodo bistri-
fluoroacetate (PIFA)7 gave anhydrolycorinone 25 in 83%
yield.8 Similarly, oxidation of 24 with iodosobenzene gave
oxoassoanine in 47% yield. Treatment of 25 and 26 with
freshly recrystallized DDQ in dry benzene gave hippadine
27 and pratosine 28 in 80 and 63% yields.
a unique displacement of the aliphatic carbonate. The
protocol was employed in the rapid synthesis of the natural
products anhdrolycorinone, hippadine, pratosine, and oxoas-
soanine in two or three steps from the triflyloxy carbonate.
In conclusion, we have developed a convenient, one-pot,
domino sp2-sp3 amidation for the formation of indolines
and indoles from o-triflyloxyphenethyl carbonates. This
sequence involves, as the individual components, a pal-
ladium-catalyzed amidation of the aryl triflate followed by
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada, and GlaxoSmith-
Kline for funding. Acknowledgment is made to the donors
of the American Chemical Society Petroleum Research Fund
for partial support of this research. We are grateful to Mr.
Doug Hairsine for performing MS analyses. M.D.G is an
NSERC PGS D scholar.
(4) For the reaction of 5g with 2a, 2b, and 2d, 1.5 equiv of amide was
used. For the reaction of 5g with 2a, 2b, and 2c, 2.5 mol % of Pd2(dba)3
and 5 mol % of XANTPHOS were used.
(5) Treatment of 20 with DBU, NaH, and t-BuOK did not lead to
cyclization; however, K2CO3 and Na2CO3 were effective in the conversion
of 20 to 6.
(6) (a) Anhydrolycorinone and hippadine: Ghosal, S.; Rao, P. H.; Jaiswal,
D. K.; Kumar, Y.; Frahm, A. W. Phytochemistry 1981, 20, 2003. (b)
Oxoassoanine: Llabres, J. M.; Viladomat, F.; Bastida, J.; Codina, C.;
Rubiralta, M. Phytochemistry 1986, 25, 2637. (c) Pratosine: Ghosal, S.;
Saini, K. S.; Frahm, A. W. Phytochemistry 1983, 22, 2305.
(7) Lead references for the PIFA-mediated oxidative biphenyl forma-
tion: (a) Moreno, I.; Tellitu, I.; Etayo, J.; SanMart´ın, R.; Dom´ınguez, E.
Tetrahedron 2001, 57, 5403. (b) Kita, Y.; Egi, M.; Okijama, A.; Ohtsubo,
M.; Takada, T.; Tohma, H. Chem. Commun. 1996, 1481.
Supporting Information Available: Experimental pro-
cedures and compound characterization data. This material
is available free of charge via the Internet at http://pubs.acs.org.
(8) If oxidation was performed with PIFA, a dimeric compound stemming
from biaryl coupling was isolated in significant amounts.
OL052086C
Org. Lett., Vol. 7, No. 21, 2005
4779