Oxa-di-π-methane Photochemical Rearrangement of Quinuclidinones. Synthesis of Pyrrolizidinones

Article abstract of DOI:10.1021/ol035202p

(Matrix presented) The oxa-di-π-methane (ODPM) photochemical rearrangement, a triplet-sensitized sigmatropic 1,2-acyl shift of β,γ-enones, was successful utilizing methyl and heptyl 1-aza-3-carboalkoxybicyclo[2.2.2]oct-2-en-5-ones (quinuclidinones) as the

Full text of DOI:10.1021/ol035202p

ORGANIC
LETTERS
2003
Vol. 5, No. 21
3811-3813
Oxa-di-π-methane Photochemical
Rearrangement of Quinuclidinones.
Synthesis of Pyrrolizidinones
Cynthia K. McClure,* Anthony J. Kiessling,† and Jeffrey S. Link‡
Department of Chemistry and Biochemistry, Montana State UniVersity,
Bozeman, Montana 59717
cmcclure@chemistry.montana.edu
Received June 27, 2003
ABSTRACT
The oxa-di-π-methane (ODPM) photochemical rearrangement, a triplet-sensitized sigmatropic 1,2-acyl shift of â,γ-enones, was successful
utilizing methyl and heptyl 1-aza-3-carboalkoxybicyclo[2.2.2]oct-2-en-5-ones (quinuclidinones) as the photoprecursors. The cyclopropane of
the heptyl ester tricyclic photoproduct could be opened with lithium dimethylcuprate or via hydrogenolysis to produce the corresponding
pyrrolizidinone skeletons.
The ongoing discoveries of new biologically active nitrogen
heterocycles such as pyrrolizidine, indolizidine, and tropane
alkaloids continue to attract interest in the development of
new methods for their syntheses.1 We are currently exploring
the use of the photochemical oxa-di-π-methane (ODPM)
rearrangement,2,3 a triplet-sensitized sigmatropic 1,2-acyl
shift, to ultimately produce these various alkaloids. In this
communication, we present the first successful ODPM
rearrangement of the quinuclidinones, 1-azabicyclo[2.2.2]-
octenones 1a,b (R ) CO2R′) to the tricyclic compounds 2a,b
(Scheme 1), and the subsequent ring-opening to the pyr-
rolizidine skeleton.
The synthetic potential of the photochemical ODPM
rearrangement of various all-carbon bicyclo[2.2.2]octenones
to produce bi- and tricyclopentanoid compounds has been
illustrated by several groups.2,3 When we first designed our
original photoprecursor, 1, we wondered if the lone pair of
electrons on the nitrogen would interfere with the photo-
chemistry, as it is known that amines can undergo photo-
activated single-electron transfer (SET) reactions.4e To our
knowledge, the majority of nitrogen-containing photopre-
cursors have been studied in either the di-π-methane rear-
rangement where the nitrogen did not participate in the
† Current address: Department of Chemistry, Wilkes University, Wilkes-
Barre, PA 18766.
‡ Current address: ArQule, Inc., 19 Presidential Way, Woburn, MA
01801.
(1) For recent reviews, see: (a) Liddell, J. R. Nat. Prod. Rep. 2001, 18,
448. Liddell, J. R. Nat. Prod. Rep. 2002, 19, 773. (b) Michael, J. P. Nat.
Prod. Rep. 2001, 18, 543. Michael, J. P. Nat. Prod. Rep. 2002 19, 719. (c)
O’Hagan, D. Nat. Prod. Rep. 2000, 17, 435. Humphrey, A. J.; O’Hagan,
D. Nat. Prod. Rep. 2001, 18, 494.
(2) (a) Zimmerman, H. E.; Armesto, D. Chem. ReV. 1996, 96, 3065 and
references therein. (b) Zimmerman, H. E. Org. Photochem. 1991, 11, 1-36.
(c) Carless, H. A. J. In Photochemistry in Organic Synthesis; Coyle, J. D.,
Ed.; Royal Society of Chemistry: London, 1986; pp 118-140. (d) Schuster,
D. I. In Rearrangements in Ground and Excited States; de Mayo, P., Ed.;
Academic Press: New York, 1980; Vol. 5, p 167. (e) Houk, K. N. Chem.
ReV. 1976, 76, 1.
(3) (a) Demuth, M. Org. Photochem. 1991, 11, 37-109. (b) Demuth,
M.; Schaffner, K. Angew. Chem., Int. Ed. Engl. 1982, 21, 820. (c) Schaffner,
K.; Demuth, M. In Modern Synthetic Methods; Scheffold, R., Ed.; Springer-
Verlag: Berlin, 1986; Vol. 4, p 89. (d) Demuth, M. In ComprehensiVe
Organic Synthesis; Trost, B. M., Fleming, I., Editors-in-chief; Paquette, L.
A., Ed.; Pergamon Press: New York, 1991; Vol. 5, p 215. (e) Yates, P.;
Stevens, K. E. Tetrahedron 1981, 37, 4401. (f) Demuth, M.; Hinsken, W.
HelV. Chim. Acta 1988, 71, 569. (g) Singh, V.; Porinchu, M. Tetrahedron
1996, 52, 7087. (h) Uyehara, T.; Kabasawa, Y.; Furuta, T.; Kato, T. Bull.
Chem. Soc. Jpn. 1986, 59, 539. (i) Corbett, R. M.; Lee, C.-S.; Sulikowski,
M. M.; Reibenspies, J.; Sulikowski, G. A. Tetrahedron 1997, 53, 11099.
(j) Singh, V. Thomas, B. J. Org. Chem. 1997, 62, 5310. (k) Singh, V.;
Prathap, S.; Porchinu, M. J. Org. Chem. 1998, 63, 4011. (l) Singh, V.;
Alam, S. Q. J. Chem. Soc., Chem. Commun. 1999, 2519. (m) Lee, T.-H.;
Rao, P. D.; Liao, C.-C. J. Chem. Soc., Chem. Commun. 1999, 801.
10.1021/ol035202p CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/20/2003

Scheme 1
Scheme 3
rearrangement (as shown in Scheme 2, eq 1)2a,e,4a,b or the
aza-di-π-methane rearrangement (Scheme 2, eqs 2 and
Scheme 3.6 The methyl ester derivative 3a required two steps
to effect the installation of the double bond, followed by
deprotection of the ketal with perchloric acid.5 However, in
the preparation of the less volatile heptyl ester derivative
1b, the perchloric acid served not only to deprotect the
dimethyl ketal but also to oxidize the phenylselenyl group,
which subsequently eliminated under the reaction conditions.
The photochemical irradiation of bicyclic compound 1a
was run under triplet-sensitized conditions (acetophenone,
quartz, 450 W Hanovia medium-pressure lamp).7 Two
possible products, the 1,2-acyl shift (ODPM) product 2a and
the 1,3-acyl shift product 4, are theoretically possible
(Scheme 4).2 The 1,3-acyl shift product normally arises from
Scheme 2
Scheme 4
3).2a,4d,e The oxa-di-π-methane rearrangement of the bicyclo-
hydrazide (Scheme 2, eq 4) is a rare example of a nitrogen-
containing bicyclic system being rearranged via the ODPM
reaction.4f
To test the viability of the ODPM rearrangement on the
quinuclidinone skeleton, 1a and 1b were prepared via a
classical route, as the synthesis of 3a was a literature
preparation from commercially available 3,4-pyridinedicar-
boxylic acid.5 Installations of the double bonds in 3a and
3b were successful via the standard procedures outlined in
photolysis under nonsensitized conditions but can be formed
from certain compounds under sensitized photolysis condi-
tions. The 1,2-acyl shift product 2a was the expected product
under our reaction conditions.
From one-dimensional proton and carbon NMR spectra,
the only product isolated from the photochemical rearrange-
ment did not exhibit an olefin, as would be seen in the R,â-
unsaturated ester 4a. The carbonyl carbon peaks were at
207.6 and 165.9 ppm, consistent with the ketone and ester
groups, respectively, in 2a. A standard COSY spectrum
(4) (a)Kuzuya, M.; Adacji, M.; Noguchi, A.; Okuda, T. Tetrahedron Lett.
1983, 24, 2271. (b) Kuzuya, M.; Mano, E.; Ishikawa, M.; Okuda, T.; Hart,
H. Tetrahedron Lett. 1981, 22, 1613. (c) Armesto, D.; Caballero, O.;
Amador, U. J. Am. Chem. Soc. 1997, 119, 12656. (d) Armesto, D.; Horspool,
W. M.; Mancheno, M. J.; Ortiz, M. J. J. Chem. Soc., Perkin Trans. 1 1990,
2348. (e) Armesto, D.; Ortiz, M. J. J. Am. Chem. Soc. 2001, 123, 9920. (f)
Hart, H.; Murray, R. K., Jr. Tetrahedron Lett. 1969, 4785.
(5) Snow, R. J.; Baker, R.; Herbert, R. H.; Hunt, I. J.; Merchant, K. J.;
Saunders, J. J. Chem. Soc., Perkin Trans. 1 1991, 409. (b) Snow, R. J.;
Street, L. J. Tetrahedron Lett. 1989, 30, 5795.
(6) Robins, D. J.; Sakdarat, S. J. Chem. Soc., Perkin Trans. 1 1979, 1734.
(7) (a) Kiessling, A. J. Ph.D. Dissertation, University of Delaware,
Newark, DE, 1996. (b) Link, J. S. Ph.D. Dissertation, Montana State
University, Bozeman, MT, 2000.
3812
Org. Lett., Vol. 5, No. 21, 2003

established all of the proton-proton coupling networks.7,8
To assist in the assignments, the tricyclic structure 2a was
submitted to molecular mechanics calculations to estimate
the dihedral angles between the protons and thus approximate
the coupling constants using the Karplus equation.7a,8 The
13C and HETCOR spectra, as well as NOE experiments,
further verified the proposed structure 2a. Thus, the rear-
rangement occurred via an oxa-di-π-methane rearrangement
and not by a 1,3-acyl shift mechanism.
Scheme 5
In our initial photolyses, we utilized acetophenone as the
solvent and the sensitizer. However, problems existed in the
isolation of the photoproduct due to the volatility of 2a and
the amount of acetophenone sensitizer used. Therefore, we
transesterified the methyl ester in 3a to the heptyl ester 3b
in order to decrease the volatility of the photoproduct.
Installation of the alkene in 3b and hydrolysis of the ketal
as described above yielded the photoprecursor 1b. Subse-
quent photolysis under sensitized conditions using either
100% acetophenone or 5% acetophenone/95% acetone
yielded the corresponding ODPM product 2b as the only
product isolated in 70% yield (material balance was starting
material 1b). Note that when 100% acetone was used as the
solvent and the sensitizer, very poor conversion to the ODPM
product was seen (2%). Optimization of the conversion was
achieved by increasing the amount of acetophenone in small
increments.
Thus, we have the first experimental evidence that the
ODPM photochemical rearrangements of 1-aza-3-carbo-
alkoxybicyclo[2.2.2]oct-2-en-5-ones did indeed produce the
expected tricyclic photoproducts 2a and 2b.
To form the pyrrolizidine substructure, ring-opening
reactions of the cyclopropane have been performed on the
photoproduct 2b.7b Utilization of “soft” nucleophiles that
were successful in the all-carbon systems, such as Nafion-
TMS and CF3CO2TMS,3a,9 led to no reaction with the
nitrogen-containing photoproduct 2b. However, the harder
nucleophile, lithium dimethylcuprate, worked well to open
the three-membered ring to afford 5 in 77% yield as a single
stereoisomer at the methyl group, but as a mixture of keto-
enol forms (Scheme 5). NOE experiments indicated that the
relative stereochemistry of the methyl group was as shown.
Reductive cleavage of the cyclopropane in 2b was also
readily accomplished using H2/Pd-C (after protonation of
the nitrogen) to produce 6 in 82% yield as a mixture of the
keto-enol forms. We have also successfully prepared the
corresponding ethyl ester (7) in 80% yield via transesteri-
fication of the heptyl ester in 6.7b Benn and Ru¨eger have
reported the hydrochloride salt of 7.10 Further transformations
of 2b to other pyrrolizidine alkaloids are currently underway.
Acknowledgment. We thank the National Science Foun-
dation for partial support of this research under Grants CHE-
9753270 (POWRE Program) and CHE-0079288, as well as
the Montana NSF EPSCoR program for an individual grant.
Acknowledgment is also made to the donors of the Petroleum
Research Fund, administered by the American Chemical
Society, for partial support of this research.
Supporting Information Available: Experimental pro-
cedures and characterization data for 1a,b, 2a,b, 3a,b, and
5-7; two-dimensional NMR spectra of 2a,b; selected one-
dimensional NMR spectra; and a table of calculated proton
dihedral angles with estimated and observed J couplings for
2a. This material is available free of charge via the Internet
at http://pubs.acs.org.
(8) See Supporting Information.
(9) Demuth, M.; Mikhail, G.; George, M. V. HelV. Chim. Acta 1981,
64, 2759.
(10) (a) Ru¨eger, H.; Benn, M. Heterocycles 1982, 19, 1677. (b) Ru¨eger,
H. E. Ph.D. Dissertation, The University of Calgary, Calgary, Alberta, 1983.
OL035202P
Org. Lett., Vol. 5, No. 21, 2003
3813