Novel reaction pathways for 2-pyridone [4+4] photoadducts

Article abstract of DOI:10.1021/ol061266z

Transannular ring closure of a pyridone-furan [4+4] photoadduct has been evaluated in a model system. A combination of nitrogen substitution and an isopropyl group gave full control of the four new stereogenic centers. Chlorination transformed the 1,5-cyc

Full text of DOI:10.1021/ol061266z

ORGANIC
LETTERS
2
006
Vol. 8, No. 15
367-3370
Novel Reaction Pathways for 2-Pyridone
4] Photoadducts
3
[4+
§
Peiling Chen, Yanping Chen, Patrick J. Carroll, and Scott McN. Sieburth*
Department of Chemistry, Temple UniVersity, Philadelphia, PennsylVania 19010,
and Department of Chemistry, UniVersity of PennsylVania,
Philadelphia, PennsylVania 19104
scott.sieburth@temple.edu
Received May 24, 2006
ABSTRACT
Transannular ring closure of a pyridone−furan [4+4] photoadduct has been evaluated in a model system. A combination of nitrogen substitution
and an isopropyl group gave full control of the four new stereogenic centers. Chlorination transformed the 1,5-cyclooctadiene to a [4.2.0] ring
system instead of to the expected [3.3.0] ring system. Changing an alkene to an enone suppressed this pathway and led to a new rearrangement,
converting the 5-8-5 ring system to a 6-7-5 system.
Cycloadditions, such as the Diels-Alder reaction, are prized
1
for the controlled leap in complexity that they engender.
Scheme 1. Approach to the Crinipellins
Higher-order cycloadditions such as the [4+4] cycloaddition
of 2-pyridones exhibit similar traits of regio- and stereo-
selectivity, but within the more difficultly accessible cyclo-
2
octadiene framework. As part of our exploration of 2-py-
ridone photocycloadditions, we have investigated a pyri-
done-furan approach to the challenging tetraquinane natural
3
product crinipellin A 1 (Scheme 1). The two published
syntheses of 1 utilized either sequential cyclopentane annu-
4
5,6
lation or an arene-alkene cycloaddition route.
§
University of Pennsylvania.
(
1) Takao, K.; Munakata, R.; Tadano, K. Chem. ReV. 2005, 105, 4779-
4
807. Corey, E. J. Angew. Chem., Int. Ed. 2002, 41, 1650-1667. Nicolaou,
K. C.; Snyder, S. A.; Montagnon, T.; Vassilikogiannakis, G. Angew. Chem.,
Int. Ed. 2002, 41, 1668-1698.
As outlined in Scheme 1, we envisioned that intramolecu-
lar cycloaddition of a furan with a cyclopentane-substituted
2-pyridone 3 would carry all but one carbon of the natural
(2) Sieburth, S. McN. Photochemical Reactivity of Pyridones. In CRC
Handbook of Organic Photochemistry and Photobiology; Horspool, W.,
Lenci, F., Eds.; CRC: Boca Ratan, 2004; pp 103/1-103/18.
(
3) Anke, T.; Heim, J.; Knoch, F.; Mocek, U.; Steffan, B.; Steglich, W.
Angew. Chem., Int. Ed. Engl. 1985, 24, 709-711.
(4) Piers, E.; Renaud, J.; Rettig, S. J. Synthesis 1998, 590-602.
1
0.1021/ol061266z CCC: $33.50
© 2006 American Chemical Society
Published on Web 06/29/2006

product. An isopropyl group, derived from (-)-limonene 5,
would block one face of the pyridone, and cis-selective
cycloaddition would lead to pentacyclic 1,5-cyclooctadiene
A cycloaddition with two tethered 2-pyridones, both
unsubstituted on nitrogen, had been found to heavily favor
1
2
cis cycloaddition in nonpolar solvents, a consequence of
the strong intermolecular hydrogen bonding of 2-pyridones.13
Although not the same system, a cis-selective cycloaddition
motif for 11a could be imagined. In the event, however,
irradiation of nitrogen-unsubstituted 11a in toluene led to
mostly the trans isomer 15a. In contrast, when the N-iso-
propyl 11b was irradiated in the same solvent, only the cis
isomer 14b was formed. In both cases, the cis:trans ratio
was readily determined by warming the photochemistry
product mixture and observing the quantitative Cope rear-
rangement of the cis isomer to give cyclobutane product 16
2. Transannular ring closure would then form a product with
stereochemistry and functionality closely related to the
natural product in only two steps from 3. We report here
our first two generations of this investigation.
The chemistry outlined in Scheme 1 was predicated, in
part, on the two reactions shown in Scheme 2. During a study
Scheme 2. Two Reaction Precedents for the Synthetic Plan
1
by H NMR spectroscopy (Scheme 3).
Scheme 3. Steric-Induced Cis-Selective Photocycloaddition
of 2-pyridone photodimer chlorination, the cis dimer 7
derived from 1,5-dimethyl-2-pyridone 6 was treated with
chlorine. The major product isolated from that reaction was
7
8
, with two quaternary carbons created during the chlorina-
tion step. A similar assembly of quaternary carbons was
anticipated in the reaction of 2 (Scheme 1). The photochem-
istry of 3 would be analogous to that of 9 from which the
[
4+4] adduct 10 was isolated as a mixture of cis and trans
8
isomers.
To evaluate the salient features of the proposed synthesis,
we elected to use ether 11, easily derived from the known
9
furan 4, and the enantiomerically pure isopropyl-substituted
1
-pyrindin-2-one, assembled using chemistry described
10
earlier (see Scheme 5). In the four possible transition states
leading to [4+4] adducts, we expected the isopropyl to inhibit
the approach of the furan from the same face (not shown).
Although the lone stereogenic center of 11 is adjacent to
one of the carbons participating in the cycloaddition, the
degree of this steric control had not been evaluated in earlier
studies. A second stereocontrol element, involving the pro-
cis and pro-trans conformations 12 and 13, was also an
unsolved challenge. The closely related furan-pyridone
photocycloaddition of 9 gave a very low level of stereocon-
trol, favoring the undesired trans isomer of 10.
The cis-selective cycloaddition of 11b was expected to
be a consequence of a destabilization of conformation 13,
where the isopropyl group R is in close proximity to the
methyl group on the furan, thereby favoring conformation
11
1
2.
The anticipated stereocontrol engendered by the isopropyl
group on the cyclopentane of 11 was found to be at useful
levels, with isomeric products resulting from the approach
of the furan to the face of the pyridone syn to the isopropyl
group (not shown) making up no more than 20% of the
product mixture.
(
5) Wender, P. A.; Dore, T. M. Tetrahedron Lett. 1998, 39, 8589-8592.
(6) See also: Curran, D. P.; Sisko, J.; Balog, A.; Sonoda, N. K. N.; Ryu,
I. J. Chem. Soc., Perkin Trans. 1 1998, 10, 1591-1593. Mehta, G.; Rao,
K. S.; Reddy, M. S. J. Chem. Soc., Perkin Trans. 1 1991, 4, 693-700.
Schwartz, C. E.; Curran, D. P. J. Am. Chem. Soc. 1990, 112, 9272-9284.
(7) Kuznetsova, L. SUNY Stony Brook. Research notes, 1999.
(
8) Sieburth, S. McN.; McGee, K. F. J.; Zhang, F.; Chen, Y. J. Org.
(12) Sieburth, S. McN.; McGee, K. F., Jr.; Al-Tel, T. H. J. Am. Chem.
Soc. 1998, 120, 587-588. Sieburth, S. McN.; McGee, K. F. J. Org. Lett.
1999, 1, 1775-1777.
(13) Krackov, M. H.; Lee, C. M.; Mautner, H. G. J. Am. Chem. Soc.
1965, 87, 892-896. Beak, P.; Covington, J. B.; Smith, S. G.; White, J. M.;
Zeigler, J. M. J. Org. Chem. 1980, 45, 1354-1362. Ducharme, Y.; Wuest,
J. D. J. Org. Chem. 1988, 53, 5787-5789.
Chem. 2000, 65, 1972-1977.
(
9) Burness, D. M. Org. Synth. 1963, Coll. Vol. IV, 649-652.
(10) Chen, Y.; Li, T.; Sieburth, S. McN. J. Org. Chem. 2001, 66, 6826-
6
828.
(11) Sieburth, S. McN.; Hiel, G.; Lin, C.; Kuan, D. P. J. Org. Chem.
1
994, 59, 80-87.
3368
Org. Lett., Vol. 8, No. 15, 2006

With 14b in hand, the critical chlorination reaction was
investigated. Addition of a chlorine solution in methylene
chloride to 14b in the same solvent led to two products. Each
a proton to yield tetrasubstituted alkene 18. The third isolated
product appears to be closely related to 18 (NMR), but its
full identity is unresolved.
1
contained a single alkene, and in the H NMR, the chemical
The migration of the amide carbonyl (21, path b) may
appear unusual, but it has been observed in other 2-pyridone
photodimer chlorinations. Indeed, product 8 in Scheme 2
results from carbonyl migration (although the desired
diquinane structure and two quaternary carbons were formed
in this case). A cationic ring closure of a constrained
cyclooctadiene to yield a 4-6 ring system has been observed
shift of the vinyl methyl singlet of 14b (1.44 ppm) had shifted
upfield to 1.28 ppm, characteristic of an aliphatic methyl
group. The spectral data were difficult to rationalize as the
hoped for product, however. When the chlorination solvent
was changed to DMF, two products again resulted, one of
which was the same. Use of ICl in methylene chloride also
gave two products, each closely related (NMR) to the unique
products formed in the two chlorination reactions (17 and
1
4
with 2-methoxy naphthalene photodimers.
Considering the proposed pathways outlined in Scheme
4, a ketone adjacent to the undesired carbocation 21, e.g.,
24 (Scheme 5), was expected to shut down the cyclobutane
18). The structures of two products were determined by X-ray
crystallography. Each of these resulted from a transannular
closure of the cyclooctadiene of 14b to a 4-6 ring system
instead of the anticipated 5-5 product. This can be rational-
ized (Scheme 4) as involving a chloronium ion that opens
Scheme 5. Anticipated Effect of Ketone Substitution and
Preparation of the Photosubstrates
Scheme 4. Chlorination of Cis Adduct 14
formation (23) and favor the desired diquinane product
formation via 25. The preparation of 24, however, would
require a different starting point. Photosubstrate 11 had been
15
prepared from (-)-limonene 5 via the known ester 26 and
the sequence shown in Scheme 5. Deprotonation of ester 26
and condensation with the Weinreb amide 30, followed by
refluxing with ammonia, led to substrate 11a.
To introduce the desired ketone and prepare 24, we began
with (-)-carvone 31 in place of limonene 5. Hydrogenation,
to the tertiary carbocation 19. This cation is then intercepted
by the proximal alkene, but instead of closing to the desired
5-5 ring system and a secondary carbocation 20, the 4-6 ring
system and a tertiary carbocation 21 are formed. This, in
turn, either loses a proton to give alkene 17 (path a) or
undergoes a migration of the amide carbonyl group to
generate another tertiary carbocation (path b) and then loses
(
14) Teitei, T.; Wells, D.; Sasse, W. H. F. Tetrahedron Lett. 1974, 367-
370. Teitei, T.; Wells, D.; Spurling, T. H.; Sasse, W. H. F. Aust. J. Chem.
978, 31, 85-96.
1
(15) Chuman, T.; Noguchi, M. Agric. Biol. Chem. 1975, 39, 567-568.
Chen, Y.; Li, T.; Sieburth, S. McN. J. Org. Chem. 2001, 66, 6826-6828.
Org. Lett., Vol. 8, No. 15, 2006
3369

followed by reduction of the ketone and protection with tert-
1
6
butyldimethylsilyl chloride, gave the cis product 32.
Scheme 6. Rearrangement of Enone 36
Ozonolysis, aldol cyclization, oxidation of the aldehyde, and
amide formation with ethylamine led to 33. Condensation
of this amide with Weinreb amide 30 gave 34. With the silyl
ether and the isopropyl groups blocking the same face of
the pyridone, the relative stereochemistry of the subsequent
cycloaddition was high; however, purification was far easier
with the alcohol. Photocycloaddition was therefore run on
the alcohol. The size of the nitrogen substitution required to
effect the cis selectivity observed for N-isopropyl 11b was
titrated with the smaller N-ethyl 34. In this case, cis adduct
35 again dominated the product but was not the exclusive
outcome, with a cis/trans ratio of 9:1.
Ketone 36 was then prepared by Swern oxidation of the
alcohol (Scheme 6). Treatment of this substrate with chlorine
in methylene chloride led to the isolation of largely one new
product in good yield. Confusingly, however, this new
structure exhibited two ketones in its carbon NMR spectrum,
as well as what appeared to be a secondary amide in the
proton NMR spectrum. The solution to the structural identity
of this product was solved, once again, by X-ray crystal-
lography, which revealed it to be 37.
The rearrangement leading to 37 can be explained by the
mechanism shown in Scheme 6. Chlorinium ion formation
at the least-hindered face of the more electron-rich alkene
opens to generate a tertiary cation. This cation is then
intercepted in the anticipated fashion by the enone alkene
to form the tetraquinane 38. Surprisingly, this cation is not
bond reorganization of [4+4] photoadducts. It is notable that
product 40 has close analogy to several natural product
structures. These studies are continuing.
1
7
intercepted by migration of the adjacent amide nitrogen
1
8
but instead suffers migration of the carbon-carbon bond,
expanding a furan ring to a pyran and contracting the newly
formed cyclopentane to a cyclobutane. The driving force for
this rearrangement is presumably the stabilization resulting
from conversion of the carbocation to the iminium ion. The
strain of the cyclobutane intermediate is then relieved by
removal of the intially added chlorine, possibly by the
chloride ion, and enol(ate) formation. Hydrolysis of the
resulting 40 leads to the observed keto amide product 37.
Although this was not the anticipated outcome, formation
of a 6-7 ring system 37 is the product of an hitherto unseen
Acknowledgment. The efforts of Tindy Li (SUNY Stony
Brook) and Luca Faille (Temple University) in supporting
chemistries are gratefully acknowledged.
Supporting Information Available: Proton NMR spectra
for new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL061266Z
(
18) Engler, M.; Anke, T.; Sterner, O. Phytochemistry 1998, 49, 2591-
(
16) Naves, Y. HelV. Chim. Acta 1964, 47, 1617-21. Gemal, A. L.;
Luche, J. J. Am. Chem. Soc. 1981, 103, 5454-5459.
17) Ader, T. A.; Champey, C. A.; Kuznetsova, L. V.; Li, T.; Lim, Y.;
Rucando, D.; Sieburth, S. McN. Org. Lett. 2001, 3, 2165-2167.
2593. Wellington, K. D.; Cambie, R. C.; Rutledge, P. S.; Bergquist, P. R.
J. Nat. Prod. 2000, 63, 79-85. Kobayashi, J.; Morita, H. The Daphniphyl-
lum Alkaloids. In The Alkaloids; Cordell, G. A., Ed.; Academic: New York,
2003; Vol. 60, pp 165-205.
(
3370
Org. Lett., Vol. 8, No. 15, 2006