Samarium(II) iodide mediated radical/polar crossover reactions of cyclobutenes. An efficient approach to the BCD ring system of the penitrems.

Article abstract of DOI:10.1021/ol0272491

[reaction: see text] Radical/polar crossover reactions of derivatives of 1-(2-cyclobutenyl)-2-(2-iodoaryl)ethanones with acetone promoted by samarium diiodide and HMPA provide 1-(1-hydroxy-1-methylethyl)-2,2a,4,8b-tetrahydro-1H-cyclobuta[a]naphthalen-3-on

Full text of DOI:10.1021/ol0272491

ORGANIC
LETTERS
2003
Vol. 5, No. 4
419-422
Samarium(II) Iodide Mediated Radical/
Polar Crossover Reactions of
Cyclobutenes. An Efficient Approach to
the BCD Ring System of the Penitrems
Alexey Rivkin, Tadamichi Nagashima, and Dennis P. Curran*
Department of Chemistry, UniVersity of Pittsburgh, Pittsburgh, PennsylVania 15260
curran@pitt.edu
Received November 7, 2002
ABSTRACT
Radical/polar crossover reactions of derivatives of 1-(2-cyclobutenyl)-2-(2-iodoaryl)ethanones with acetone promoted by samarium diiodide
and HMPA provide 1-(1-hydroxy-1-methylethyl)-2,2a,4,8b-tetrahydro-1H-cyclobuta[a]naphthalen-3-one derivatives in about 50% isolated yield.
This reaction shows promise for construction of the BCD ring fragment of the penitrems.
The penitrems are a small but important family of structurally
complex and biologically active indole alkaloids. Typified
by penitrem D (Figure 1), family members contain at least
penitrems that features a cyclization of an aryl radical to a
cyclobutene followed by reduction and trapping with acetone.
Although cyclobutenes are expected to be good radical
acceptors6 and have been used occasionally in bimolecular
reactions,7 we could not locate any examples of intra-
molecular additions (cyclizations) of radicals to cyclobutenes.
(1) (a) Wilson, B. J.; Wilson, C. H.; Hayes, A. W. Nature 1968, 220,
77. (b) de Jesus, A. E.; Steyn, P. S.; van Heerrden, F. R.; Vleggar, R.;
Wessels, P. L.; Hull, W. E. J. Chem. Soc., Perkin Trans. 1 1983, 1857-
1861.
(2) Smith, A. B., III; Kanoh, N.; Ishiyama, H.; Hartz, R. A. J. Am. Chem.
Soc. 2000, 122, 11254-11255.
(3) (a) Nagashima, T., Ph.D. Thesis, University of Pittsburgh, 1999. (b)
Rivkin, A., Ph.D. Thesis, University of Pittsburgh, 2001.
(4) We use here the “radical/polar crossover” terminology of Murphy,
but such reactions are also called by other names such as “cascade radical/
ionic” reactions. Bashir, N.; Patro, B.; Murphy, J. A. In AdVances in Free
Radical Chemistry; Zard, S. Z., Ed.; Jai Press: Stamford, CT, 1999; Vol.
2, pp 123-150.
(5) (a) Krief, A.; Laval, A. M. Chem. ReV. 1999, 99, 745-778. (b)
Molander, G. A.; Harris, C. R. Chem. ReV. 1996, 96, 307-338. (c)
Molander, G. A.; Harris, C. R. Tetrahedron 1998, 54, 3321-3354. (d)
Curran, D. P. In ComprehensiVe Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon: Oxford, UK, 1991; Vol. 4, pp 779-831. (e) Curran,
D. P.; Fevig, T. L.; Jasperse, C. P.; Totleben, M. J. Synlett 1992, 943-961.
(6) For example, the ketone group of cyclobutanones has often been used
in radical cyclization reactions: Dowd, P.; Zhang, W. Chem. ReV. 1993,
93, 2091-2115. Cyclizations to methylene cyclobutanes are also known:
Zhang, W.; Dowd, P. Tetrahedron Lett. 1995, 36, 8539-8542.
Figure 1. Penitrem D.
nine interlocking rings in sizes ranging from four to seven
members.1 The unusual juxtaposition of rings B-F on the
periphery of ring A provides a challenge to existing
methodology that has already been met by Smith and co-
workers.2 We are interested in developing novel approaches
to the A-F ring core of the penitrems by using radical/polar
crossover reactions3,4 mediated by samarium(II) iodide,5 and
we deploy here a new strategy to make the BCD unit of the
10.1021/ol0272491 CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/23/2003

The general plan for the model study is outlined in Figure
2. Cyclization of 1 with samarium iodide and acetone by
The synthesis of a representative substrate 13 for the
radical/polar cascade introduces a new approach to make acyl
cyclobutenes that is shown in Scheme 1. Generation of
Scheme 1
Figure 2. SmI2 radical/polar crossover reaction.
benzyl cuprate 89 from the corresponding benzyl bromide
followed by quenching with the 3-chlorocyclobutane car-
boxylic acid chloride 910 provided a 1/1 mixture of the cis
and trans diastereomers of ketone 10 in a combined yield of
78%. These were easily separated by flash chromatography,
and pure cis and trans diastereomers of ketone 10 were
independently carried through the synthesis to cycliza-
tion precursor 13 to simplify characterization of the inter-
mediates.
Protection of ketones 10-trans and 10-cis as the 1,3-
dioxolane ketals via Dean-Stark azeotropic distillation with
ethylene glycol and pTSA in benzene provided 11-trans and
11-cis in 93% and 91% yield, respectively. Exposure of
chlorides 11 to the benzeneselenolate anion in DME at reflux
for 12 h provided 12-trans (67%, from 11-cis) and 12-cis
(63%, from 11-trans). Separate oxidation of 12-trans and 12-
either Grignard or Barbier procedures8 is expected to yield
2 through the sequence of one-electron reduction to give aryl
radical 3, cyclization to give cyclobutyl radical 4, further
reduction to cyclobutyl samarium species 5 (or its equiva-
lent), and addition to acetone to give samarium alkoxide 6.
In addition to determining the viability of the proposed
sequence and its stereochemical outcome, we selected
substrates with different R1 and R2 groups to probe generality,
keeping in mind that these should be precursors of the exo
methylene group present in the penitrems.
The viability of the polar stage of the proposed sequence
was readily demonstrated by the experiments shown in eq
1. Addition of cyclobutyl bromide to 2-octanone was effected
(7) Examples of radical additions to cyclobutenes: (a) Ferjancic, Z.;
Cekovic, Z.; Saicic, R. N. Tetrahedron Lett. 2000, 41, 2979-2982. (b)
Campbell, E. F.; Park, A. K.; Kinney, W. A.; Fengel, R. W.; Liebskind, L.
S. J. Org. Chem. 1995, 60, 1470-1472. (c) Legrand, N.; Quiclet-Sire, B.;
Zard, S. Z. Tetrahedron Lett. 2000, 41, 9815-9818. (d) Chen, X.-P.; Sufi,
B. A.; Padias, A. B.; Hall, H. K., Jr. Macromolecules 2002, 35, 4277-
4281.
(8) Curran, D. P.; Totleben, M. J. J. Am. Chem. Soc. 1992, 114, 6050-
6058. Grignard conditions: addition of halide to SmI2‚4HMPA followed
by addition of ketone. Barbier conditions: addition of halide and ketone
together in THF to SmI2‚4HMPA.
by treatment with SmI2 (2 equiv) and HMPA (8 equiv) under
Grignard and Barbier conditions.5d,8 Both reactions succeeded
but the Barbier procedure gave the superior yield (69% for
Barbier compared to 33% for Grignard). Accordingly, the
Barbier procedure was used for the subsequent cascade
reactions.
(9) (a) Berk, S. C.; Knochel, P.; Yeh, M. C. P. J. Org. Chem. 1988, 53,
5789-5791. (b) Knochel, P.; Singer R. D. Chem. ReV. 1993, 93, 2117-
2188.
(10) Hall, H. K.; Smith, C. D.; Blanchard, E. P.; Cherofsky, S. C.; Sieja,
J. B. J. Am. Chem. Soc. 1971, 93, 121-130.
420
Org. Lett., Vol. 5, No. 4, 2003

cis with m-CPBA in dichloromethane followed by selenoxide
elimination in 1,2-dichloroethane at reflux for 12 h pro-
vided the same cyclobutene 13 in 60% and 56% yields after
silica gel chromatography. The other substrates shown in
Table 1 were made from ketone 10-cis/trans by similar
sequences.3 Conveniently, these cyclobutenes were not prone
to thermal electrocyclic ring opening either under the
conditions of elimination or on storage at or below room
temperature.
Table 1. Examples of Radical/Polar Crossover Reactions
The results of the samarium(II) iodide mediated tandem
cyclization of 13 are shown in eq 2. A solution of the iodide
and acetone (4 equiv) in THF was added to a THF solution
of SmI2 (4 equiv) and HMPA (16 equiv). Standard aqueous
workup and flash chromatography gave the target radical/
polar crossover product 14 in 59% yield as a single
stereoisomer alongside the directly reduced product 16 (19%)
and a small amount of the product 15 (2%) derived from
successful radical cyclization but failed polar addition. The
cis ring fusion can readily be anticipated from much
precedent in formation of bicycles of other sizes by radical
cyclization,11 and the exo addition of acetone is supported
by the observation of a strong cross-peak between one of
the methyl groups and the adjacent ring fusion hydrogen in
a 2D noe NMR experiment. The directly reduced product
most likely results from hydrogen abstraction by the inter-
mediate aryl radical from the medium,5e although this has
not been demonstrated experimentally.
The generality of the process is shown by the examples
in Table 1. Reductive cyclization of the dioxane acetal gave
results comparable to its lower homologue (compare entry
1 to the results in eq 2). Cyclization of a chiral acetal (entry
2) gave an inseparable mixture of diastereomers in a 1/1 ratio
in 55% isolated yields. The diastereomeric alcohols in entries
3 and 4 cyclized to give the corresponding products in 40%
and 45% isolated yield, respectively. Reductive cyclization
of exo-methylene substrate in entry 5 provided the hydroxy-
alkylated tricycle in 41% isolated yield. Finally, the o-methyl-
substituted analogue of 13 in entry 6 provided the tricycle
a The product is a 1/1 mixture of diastereomers with the ring fusion
hydrogens and the hydroxypropyl group on the cyclobutyl ring either all R
or all â.
in 53% isolated yield.12 This success is important since the
substituent pattern of the precursor better models that needed
for the penitrems.
These successful model studies introduce radical cycliza-
tions to cyclobutenes, expand the capabilities of the already
powerful radical/polar crossover reactions of SmI2, and
encourage further development of the planned strategy
toward the penitrem class of natural products. The reaction
(11) Curran, D. P.; Porter, N. A.; Giese, B. Stereochemistry of Radical
Reactions: Concepts, Guidelines, and Synthetic Applications; VCH: Wein-
heim, Germany, 1996.
(12) For more detailed studies on the effects of ortho substituents on
aryl radical reactions, see: Fairweather, N., M.S. Thesis, University of
Pittsburgh, 2002.
Org. Lett., Vol. 5, No. 4, 2003
421

tolerates several substituents at the key carbon atom that bears
a methylene group in the penitrems, but trends in yields
suggest that acetal (and perhaps other quaternary) substituents
give better yields than protected hydroxy groups or meth-
ylene groups. In a subsequent full paper, we will report on
rate constant studies that confirm and quantify this trend and
we will also describe other reductive and organometallic
methods to cyclize these types of substrates.
Fairweather for preparing several intermediates and Drs. Fu-
Tyan Lin (NMR) and Kasi Somayajula (mass spectrometry)
for their assistance. Alexey Rivkin thanks the Bayer Cor-
poration for a Bayer graduate fellowship.
Supporting Information Available: Detailed experi-
mental procedures and compound characterization for syn-
thesis and cyclization of 13. This material is available free
of charge via the Internet at http://pubs.acs.org.
Acknowledgment. We thank the National Institute of
Health for funding this work. We also thank Mr. Neil
OL0272491
422
Org. Lett., Vol. 5, No. 4, 2003