Angewandte
Chemie
DOI: 10.1002/anie.201304687
Alkaloid Synthesis
Protic-Solvent-Mediated Cycloisomerization of Quinoline and
Isoquinoline Propargylic Alcohols: Syntheses of
(
ꢀ)-3-Demethoxyerythratidinone and (ꢀ)-Cocculidine**
Stephen T. Heller, Toshihiro Kiho, Alison R. H. Narayan, and Richmond Sarpong*
Novel nitrogen-containing rings (azacycles) are particularly
valuable in the arenas of alkaloid synthesis and medicinal
chemistry. Our research group has been engaged in the
development of strategies for the cycloisomerization of
pyridine propargylic alcohols (e.g., 1, Scheme 1) for the
tertiary alcohols, respectively (see 2 and 3, Scheme 1), are not
formed efficiently under the metal-promoted cycloisomeriza-
tion conditions that were effective for the synthesis of
indolizines and indolizinones. Herein, we demonstrate that
benzindolizinones can be produced in high yields by the
cycloisomerization of quinoline or isoquinoline alcohols in
protic solvents (e.g., ethanol or n-propanol), which presum-
ably form a hydrogen-bonding network that activates the
substrate.
II
generation of indolizinones and indolizines with both Pt and
Surprisingly, the benzindolizinone framework has gone
virtually unexplored: only scattered reports of benz[e]indo-
[
3]
lizinones and only a single report of the isomeric benz[g]in-
[
4]
dolizinone had appeared prior to the study that is described
herein. To demonstrate the utility of the unique azacyclic
products that can now be effectively accessed by the protic-
solvent-mediated cycloisomerization, we applied the benz-
[
g]indolizinone scaffold to short syntheses of the Erythrina
[
5]
[6]
alkaloids (ꢀ)-3-demethoxyerythratidinone (4) and (ꢀ)-
[
7]
cocculidine (5).
The basis of our current studies rests on the concurrent
discovery by Kim et al. and our group that simple protic
solvents can mediate cycloisomerizations of pyridine prop-
[
8,9]
argylic alcohols to yield indolizines and indolizinones.
This
observation suggested a potential route to benzindolizinones,
given that the metal-mediated cycloisomerization (e.g., 6a!
7
a, Scheme 2) proceeds only in low yield (36%) owing to
Scheme 1. Cycloisomerization approaches to indolizines, indolizi-
nones, and benzindolizinones.
a competing ejection of the acetylide unit to give a ketone
[
10]
(e.g., 8a) and an alkyne (in this case hexyne). Alternatively,
[
11]
as demonstrated by Kim and co-workers, the subjection of
quinoline propargylic alcohol 6b to iodocycloisomerization
conditions results in the formation of iodopyrrolo[1,2-a]-
quinoline 9 and not the expected benzindolizinone 7b or its
iodinated derivative.
III
[1]
In catalysts. Similarly, others have shown that a wide
variety of p acids can mediate the cycloisomerization of
pyridine propargylic secondary alcohols to indolizines.
[2]
Unfortunately, benzannulated variants of these underutilized
scaffolds (especially benz[e]indolizinones and benz[g]indoli-
zinones) that would arise from quinoline or isoquinoline
We hypothesized that protic solvents would be less likely
than p Lewis acids to effect the ejection of acetylides from
substrates such as 6. Thus, protic-solvent-mediated cyclo-
isomerization could provide a possible solution to the
unsolved problem of access to benzindolizinones from
propargylic alcohols. Support for this assertion is evident in
the effectiveness of the protic-solvent-mediated cycloisome-
rization conditions (nPrOH, 1208C, 40 h), which led to the
formation of benz[g]indolizinone 7a in 91% yield (for the
isolated product) from the quinoline tertiary alcohol 6a.
Aryl, alkyl, and alkenyl groups functioned as migrating
groups when quinoline propargylic tertiary alcohol substrates
3 containing a terminal alkyne were employed, and the
desired products were obtained in good to excellent yields
(Scheme 3). Generally, these reactions were conducted at
[
*] S. T. Heller, T. Kiho, A. R. H. Narayan, Prof. R. Sarpong
Department of Chemistry, University of California, Berkeley
Berkeley, CA 94720 (USA)
E-mail: rsarpong@berkeley.edu
[**] The research was supported by a grant from the NIH-NIGMS (RO1
084906), the NSF-USA (graduate fellowship to STH), and Daiichi
Sankyo. We thank Daniel Sanner and Christina Kraml of Lotus
Separations (Princeton, NJ (USA)) for the preparative-scale sepa-
ration of the enantiomers of 28.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!