Endo-oxacyclizations of polyepoxides: Biomimetic synthesis of fused polycyclic ethers

Article abstract of DOI:10.1021/jo0110092

Boron trifluoride-etherate promotes the endo-selective oxacyclization of polyepoxides derived from various acyclic terpenoid polyalkenes, including geraniol, farnesol, and geranylgeraniol, providing an efficient and stereoselective synthesis of substituted oxepanes and fused polyoxepanes. The mechanism of the oxacyclization reaction probably involves intramolecular nucleophilic addition of epoxide oxygen to open another epoxide that is activated as an electrophile by the Lewis acid. These oxacyclizations proceed stereospecifically with inversion of configuration upon opening of each epoxide to provide trans-fused polycyclic products. The oxacyclization cascade is terminated by a tethered nucleophile, which may be the carbonyl oxygen of a ketone, ester, or carbonate, or a trisubstituted alkene. The best oxacyclization yields are generally observed with tert-butyl carbonate as the terminating nucleophile, although in some cases the oxacyclization products include formation of tert-butyl ethers as a minor product. The oxacyclization transformations described herein may mimic ring-forming steps in the biosynthesis of trans-syn-trans-fused polycyclic ether marine natural products.

Full text of DOI:10.1021/jo0110092

J . Org. Chem. 2002, 67, 2515-2523
2515
En d o-Oxa cycliza tion s of P olyep oxid es: Biom im etic Syn th esis of
F u sed P olycyclic Eth er s
Frank E. McDonald,* Fernando Bravo, Xia Wang, Xudong Wei, Motoki Toganoh,1a
J . Ramo´n Rodr´ıguez,1b Bao Do,1c Wade A. Neiwert,1c and Kenneth I. Hardcastle1c
Department of Chemistry, Emory University, 1515 Pierce Drive, Atlanta, Georgia 30322
fmcdona@emory.edu
Received October 17, 2001
Boron trifluoride-etherate promotes the endo-selective oxacyclization of polyepoxides derived from
various acyclic terpenoid polyalkenes, including geraniol, farnesol, and geranylgeraniol, providing
an efficient and stereoselective synthesis of substituted oxepanes and fused polyoxepanes. The
mechanism of the oxacyclization reaction probably involves intramolecular nucleophilic addition
of epoxide oxygen to open another epoxide that is activated as an electrophile by the Lewis acid.
These oxacyclizations proceed stereospecifically with inversion of configuration upon opening of
each epoxide to provide trans-fused polycyclic products. The oxacyclization cascade is terminated
by a tethered nucleophile, which may be the carbonyl oxygen of a ketone, ester, or carbonate, or a
trisubstituted alkene. The best oxacyclization yields are generally observed with tert-butyl carbonate
as the terminating nucleophile, although in some cases the oxacyclization products include formation
of tert-butyl ethers as a minor product. The oxacyclization transformations described herein may
mimic ring-forming steps in the biosynthesis of trans-syn-trans-fused polycyclic ether marine natural
products.
In tr od u ction
A variety of marine natural product structures consist
of multiple cyclic ether structures.2 Trans-syn-trans-fused
polycyclic ether structures are present in a large family
of dinoflagellate-derived natural products, with the sim-
plest member of these structurally complex compounds
represented by hemibrevetoxin B (3, Figure 1).3,4 The
production of these compounds in nature is associated
with dinoflagellate blooms (“red tide” phenomenon),
which can result in massive kills of fish and other marine
animals in temperate waters in both the Atlantic and
Pacific Oceans.5 Human exposure to these compounds can
cause symptoms ranging from diarrhea to extreme
cardiovascular and neurotoxic effects at exposures as low
as a few parts per billion in contaminated seawater or
(1) (a) Visiting Graduate Fellow, The J apan Society for the Promo-
tion of Science, 2001. (b) Visiting Graduate Fellow, Ministerio de
Educacio´n y Ciencia (Spain), 1999. (c) Emory University X-ray Crystal-
lography Laboratory.
(2) (a) Marine Toxins: Origin, Structure, and Molecular Pharmacol-
ogy; Hall, S., Strichartz, G., Eds.; ACS Symposium Series 418;
American Chemical Society: Washington, DC, 1990. (b) Shimizu, Y.
Chem. Rev. 1993, 93, 1685. (c) Yasumoto, T.; Murata, M. Chem. Rev.
1993, 93, 1897. (d) Yasumoto, T.; Satake, M. Chimia 1998, 52, 63.
(3) Krishna Prasad, A. V.; Shimizu, Y. J . Am. Chem. Soc. 1989, 111,
6476.
(4) Other fused polycyclic ether natural products include: (a)
Brevetoxin B: Lin, Y.-Y.; Risk, M.; Ray, S. M.; Van Engen, D.; Clardy,
J .; Golik, J .; J ames, J . C.; Nakanishi, K. J . Am. Chem. Soc. 1981, 103,
6773. (b) Brevetoxin A: Shimizu, Y.; Chou, H.-N.; Bando, H.; Van
Duyne, G.; Clardy, J . C. J . Am. Chem. Soc. 1986, 108, 514. (c)
Ciguatoxin: Murata, M.; Legrand, A. M.; Ishibashi, Y.; Fukui, M.;
Yasumoto, T. J . Am. Chem. Soc. 1990, 112, 4380. (d) Satake, M.;
Morohashi, A.; Oguri, H.; Oishi, T.; Hirama, M.; Harada, N.; Yasumoto,
T. J . Am. Chem. Soc. 1997, 119, 11325. (e) Gambierol: Satake, M.;
Murata, M.; Yasumoto, T. J . Am. Chem. Soc. 1993, 115, 361. (f)
Maitotoxin: Murata, M.; Naoki, H.; Matsunaga, S.; Satake, M.;
Yasumoto, T. J . Am. Chem. Soc. 1994, 116, 7098. (g) Zheng, W.;
DeMattei, J . A.; Wu, J .-P.; Duan, J . J .-W.; Cook, L. R.; Oinuma, H.;
Kishi, Y. J . Am. Chem. Soc. 1996, 118, 7946. (h) Yessotoxin: Taka-
hashi, H.; Kusumi, T.; Kan, Y.; Satake, M.; Yasumoto, T. Tetrahedron
Lett. 1996, 37, 7087.
F igu r e 1. Proposed biosynthesis of hemibrevetoxin from an
acyclic polyene.
seafood. However, some fused polycyclic ether compounds
also exhibit potent antifungal activity,6 and the demon-
strated activation of sodium ion channels by brevetoxins
and the resulting increase in sodium ion transport7 may
10.1021/jo0110092 CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/19/2002

2516 J . Org. Chem., Vol. 67, No. 8, 2002
McDonald et al.
also have beneficial and highly potent antibiotic effects
similar to those of ionophoric polyethers.
Hemibrevetoxin B (3) is a relatively simple member
of this class of polycyclic ether natural products and has
been synthesized by several laboratories, albeit with long
step counts in each case (38-60 steps).8 However, nature
probably synthesizes hemibrevetoxin and structurally
related natural products in a much more efficient fashion,
constructing several rings and multiple stereocenters via
a tandem oxacyclization transformation. In the mid-
1980s, Shimizu and Nakanishi9 independently proposed
a general scheme for brevetoxin biosynthesis (Figure 1):
an all trans-polyene precursor 1 is biosynthesized, fol-
lowed by multiple enzyme-catalyzed epoxidations of the
alkenes to polyepoxide 2. While the absolute sense of
epoxidation is identical for each alkene, perhaps the most
interesting step is the endo-regioselective tandem oxa-
cyclization to form the fused polycyclic ether structure
3, a reaction that would normally proceed with exo-
regioselectivity due to both kinetic and stereoelectronic
factors.10
F igu r e 2. Tandem exo-cyclizations of polyepoxides.
We plan to mimic the efficient biosynthesis of polycyclic
ether natural products by a nonenzymatic strategy
featuring tandem oxacyclizations from acyclic polyene
precursors. The principal challenge to this strategy is
achieving endo-regioselectivity in oxacyclization trans-
formations. Cascades of exo-selective oxacyclizations of
polyepoxides to afford chain polycyclic ethers are well
precedented, particularly when promoted by protic ac-
ids.11 A good nucleophile (usually an alcohol or carboxylic
acid) intramolecularly adds to the nearest protonated
epoxide at the proximal position to generate a new
hydroxyl group upon oxacyclization (4 to 5, Figure 2).
However, various substituent effects12 have been em-
ployed to elicit endo-selectivity as well as the use of
catalytic antibodies13 or Co-salen catalysis.14
F igu r e 3. Tandem endo-oxacyclizations of polyepoxides.
epoxides and a 1-hydroxy-4,7,10-triepoxide substrate, but
this synthesis required introduction of chelating substit-
uents in order to provide endo-regioselectivity. Although
single endo-oxacyclizations had been reported with larger
ring sizes,12j,k the tandem oxacyclizations reported were
limited to formation of six-membered rings, and the
triepoxide substrate provided the tricyclic polypyran
product in less than 10% yield.
Our mechanistically unique strategy for tandem, endo-
selective oxacyclizations of polyepoxide substrates is to
diminish the reactivity of the oxygen nucleophile, with
epoxides serving as nucleophiles (at oxygen) as well as
electrophiles (at carbon).17 Thus, activation of one epoxide
as an electrophile in 8 (Figure 3) is followed by nucleo-
philic addition of a second epoxide to form an epoxonium
Prior to our initial communication15 there was only one
report of cascade endo-selective oxacyclization.16 Murai
demonstrated the formation of polypyrans via lanthanum
triflate-promoted oxacyclizations of 1-hydroxy-4,7-di-
(5) (a) Shimizu, Y. Marine Natural Products; Scheuer, P. J ., Ed.;
Academic Press: New York, 1978; Vol. 1, Chapter 1. (b) Tayler, D. L.;
Seliger, H. H. Toxic Dinoflagellate Blooms; Elsevier/North-Holland:
New York, 1979. (c) Baden, D. G.; Bikkazi, G.; Decker, S. J .; Foldes,
F. F.; Leung, I. Toxicon 1984, 22, 75.
(6) Gambieric acids: Nagai, H.; Torigoe, K.; Satake, M.; Murata,
M.; Yasumoto, T.; Hirota, H. J . Am. Chem. Soc. 1992, 114, 1102.
(7) (a) Trainer, V. L.; Baden, D. G.; Catterall, W. A. J . Biol. Chem.
1994, 269, 19904. (b) Matile, S.; Nakanishi, K. Angew. Chem., Int. Ed.
Engl. 1996, 35, 757.
(8) (a) Nicolaou, K. C.; Raja Reddy, K.; Skokotas, G.; Sato, F.; Xiao,
X.-Y.; Hwang, C.-K. J . Am. Chem. Soc. 1993, 115, 3558. (b) Kadota, I.;
J ungYoul, P.; Koumura, N.; Pollaud, G.; Matsukawa, Y.; Yamamoto,
Y. Tetrahedron Lett. 1995, 36, 5777. (c) Morimoto, M.; Matsukura, H.;
Nakata, T. Tetrahedron Lett. 1996, 37, 6365. (d) Mori, Y.; Yaegashi,
K.; Furukawa, H. J . Am. Chem. Soc. 1997, 119, 4557. (e) Kadota, I.;
Yamamoto, Y. J . Org. Chem. 1998, 63, 6597. (f) Rainier, J . D.; Allwein,
S. P.; Cox, J . M. J . Org. Chem. 2001, 66, 1380.
(12) (a) Chen, R.; Rowand, D. A. J . Am. Chem. Soc. 1980, 102, 6609.
(b) Cookson, R. C.; Liverton, N. J . J . Chem. Soc., Perkin Trans. 1 1985,
1589. (c) Nicolaou, K. C.; Prasad, C. V.; Somers, P. K.; Hwang, C.-K.
J . Am. Chem. Soc. 1989, 111, 5330. (d) Nicolaou, K. C.; Prasad, C. V.
C.; Somers, P. K.; Hwang, C.-K. J . Am. Chem. Soc. 1989, 111, 5335.
(e) Adiwidjaja, G.; Flo¨rke, H.; Kirschning, A.; Schaumann, E. Tetra-
hedron Lett. 1995, 36, 8771. (f) Mori, Y.; Yaegashi, K.; Furukawa, H.
J . Am. Chem. Soc. 1996, 118, 8158. (g) Oishi, T.; Maeda, K.; Hirama,
M. J . Chem. Soc., Chem. Commun. 1997, 1289. (h) Neogi, P.; Doun-
doulakis, T.; Yazbak, A.; Sinha, S. C.; Sinha, S. C.; Keinan, E. J . Am.
Chem. Soc. 1998, 120, 11279. (i) Mukai, C.; Sugimoto, Y.; Ikeda, Y.;
Hanaoka, M. Tetrahedron 1998, 54, 823. (j) Fujiwara, K.; Hayashi,
N.; Tokiwano, T.; Murai, A. Heterocycles 1999, 50, 561. (k) Fujiwara,
K.; Morishita, H.; Tokiwano, T.; Murai, A. Heterocycles 2001, 54, 109.
(13) J anda, K. D.; Shevlin, C. G.; Lerner, R. A. J . Am. Chem. Soc.
1995, 117, 2659.
(14) Wu, M. H.; Hansen, K. B.; J acobsen, E. N. Angew. Chem., Int.
Ed. 1999, 38, 2012.
(15) McDonald, F. E.; Wang, X.; Do, B.; Hardcastle, K. I. Org. Lett.
2000, 2, 2917.
(16) Tokiwano, T.; Fujiwara, K.; Murai, A. Synlett 2000, 355.
(17) For additional examples in which epoxides have been utilized
as nucleophiles in oxacyclization transformations, see: (a) David, F.
J . Org. Chem. 1981, 46, 3512. (b) Alvarez, E.; D´ıaz, M. T.; Pe´rez, R.;
Ravelo, J . L.; Regueiro, A.; Vera, J . A.; Zurita, D.; Mart´ın, J . D. J .
Org. Chem. 1994, 59, 9, 2848. (c) Tokumasu, M.; Sasaoka, A.; Takagi,
R.; Hiraga, Y.; Ohtaka, K. Chem. Commun. 1997, 875.
(9) (a) Nakanishi, K. Toxicon 1985, 23, 473. (b) Chou, H.-N.;
Shimizu, Y. J . Am. Chem. Soc. 1987, 109, 2184.
(10) (a) Coxon, J . M.; Hartshorn, M. P.; Swallow, W. H. Aust. J .
Chem. 1973, 26, 2521. For computational studies, see: (b) Na, J .; Houk,
K. N.; Shevlin, C. G.; J anda, K. D.; Lerner, R. A. J . Am. Chem. Soc.
1993, 115, 8453. (c) Na, J .; Houk, K. N. J . Am. Chem. Soc. 1996, 118,
9204. (d) Coxon, J . M.; Morokuma, K.; Thorpe, A. J .; Whalen, D. J .
Org. Chem. 1998, 63, 3875. (e) Coxon, J . M.; Thorpe, A. J . J . Org. Chem.
1999, 64, 5530. (f) Coxon, J . M.; Thorpe, A. J . J . Am. Chem. Soc. 1999,
121, 10955.
(11) For a review, see (a) Koert, U. Synthesis 1995, 115. For more
recent examples, see: (b) Evans, D. A.; Ratz, A. M.; Huff, B. E.;
Sheppard, G. S. J . Am. Chem. Soc. 1995, 117, 3448. (c) Xiong, Z.; Corey,
E. J . J . Am. Chem. Soc. 2000, 122, 4831. (d) Morimoto, Y.; Iwai, T.;
Kinoshita, T. J . Am. Chem. Soc. 2000, 122, 7124.

Biomimetic Synthesis of Fused Polycyclic Ethers
J . Org. Chem., Vol. 67, No. 8, 2002 2517
Sch em e 1. Ster eoselective Ep oxid a tion s of
1,5-Dien esa
ion intermediate 9. The reverse reaction may be facile
and, thus, controlled by thermodynamic rather than
exclusively kinetic factors. The attack of an external
nucleophile (another epoxide or carbonyl oxygen) finally
terminates the ring-forming reaction (9 to 10). Endo-
regioselectivity might be due in part to the difference in
stability between the two bicyclic oxonium ions 11 vs 9,
with the smaller ring size in 11 (from exo-cyclization)
more strained than the intermediate 9 arising from initial
endo-cyclization. We also anticipated that endo-selectivity
might be favored in substrates bearing alkyl substitution
at each site of nucleophilic attack.
Resu lts a n d Discu ssion
Ster eoselective Syn th esis a n d Oxa cycliza tion Re-
a ction s of 1,5-Diep oxid es: In itia l Stu d ies. Our stud-
ies began with diepoxides derived from terpene-derived
1,5-dienes, including geranylacetone and geraniol. Al-
though peroxycarboxylic acid epoxidation of geranyl-
acetone (12) provided an inseparable 1:1 mixture of
diastereomeric diepoxides 13 and 14, the Shi enantio-
selective double epoxidation18 of both alkenes of geranyl-
acetone provided an inseparable mixture of diastereomers
favoring 13 in a 3.7:1 ratio in quantitative yield (Scheme
1). Enantioselective epoxidation of both alkenes of ge-
raniol 16 was best obtained by Sharpless enantioselective
epoxidation19 of the alkene nearest the hydroxyl group
(88% ee), followed by Shi epoxidation of the remaining
alkene, providing a 4.6:1 mixture of diastereomers with
the major diastereomer 17 produced in 98% ee as
determined by HPLC analysis (Chiralpak OD-H).20 Fur-
thermore, a diastereomeric diepoxide 18 could be pro-
duced using the opposite catalytic tartrate enantiomer
for the Sharpless epoxidation and the Shi ketone epoxi-
dation catalyst 15. The tert-butyl ester of geranylacetic
acid (19)21 was insoluble in the polar solvent medium
required for Shi epoxidation, but the corresponding
methyl ester 2022 could be stereoselectively epoxidized
to give diepoxide 23 with high diastereo- and enantio-
selectivity. Diepoxide substrate 25 was prepared from
sequential Sharpless and Shi epoxidations of 2-methyl-
ene-6-methyl-5-hepten-1-ol (24).23
Oxacyclization studies were first conducted with gera-
nylacetone-derived diepoxide mixture 13/14. A variety of
Lewis acids were initially explored, including SnCl4 and
Yb(OTf)3, which both gave complex mixtures, before we
observed that reaction of 13/14 with BF3‚OEt2 in CH2-
Cl2 promoted rapid formation of a mixture of compounds
determined to be oxepanediol 26 and a diastereomer in
a 1:1 inseparable mixture after aqueous workup. The
corresponding reaction with the 3.7:1 mixture of di-
epoxides favoring 13 afforded oxepanediol 26, which was
purified after acylation of secondary alcohol to provide
monoacetate 27 as a 3.7:1 inseparable mixture of dia-
a
Reagents and Conditions: (a) m-CPBA, CH2Cl2 (100%). (b) 0.6
equiv 15, Oxone, MeCN/DMM/H2O, pH 10.5, 0 °C (from 12, 100%
yield, 3.7:1 dr; from 19, 61%). (c) 5 mol % Ti(O-i-Pr)4, 6 mol %
(-)-DET, TBHP, CH2Cl2, -25 °C (from 16, 100% yield, 88% ee;
from 24, 77% yield). (d) 0.3 equiv 15, Oxone, MeCN/DMM/H2O,
pH 10.5, 0 °C (93% yield, 4.6:1 dr 17:18, 98% ee for 17; 85% yield,
5.7:1 dr 18:17, 99% ee for 18; 85% yield of 25). (e) 5 mol % Ti(O-
i-Pr)4, 6 mol % (+)-DET, TBHP, CH2Cl2, -25 °C (100% yield, 88%
ee).
stereomers (Scheme 2). We also explored B(C6F5)3 ca-
talysis, which converted diepoxyketone 13 to oxepanediol
26 in a slightly better yield. Structure assignments for
26 were based on the observation that only one of the
alcohols underwent acylation, as indicated by the pres-
1
ence of one new acetate methyl H NMR resonance, as
well as IR stretches in the carbonyl region. Alternative
structures 28 and 29 would have been present predomi-
nantly in the cyclic hemiketal form; thus, the ketone
CdO IR stretch would have been absent or relatively
weak. Structure 30 with two tertiary alcohols was also
excluded by the presence of the 1H NMR doublet of
doublets at 4.83 ppm, which was assigned to H-9 of
structure 27.
The primary hydroxyl group of the geraniol-derived
diepoxide 17 (5.7:1 dr) was protected as acetate ester 31
and then subjected to BF3‚OEt2-promoted oxacyclization
to provide the corresponding oxepane 33 in 60% isolated
(18) (a) Wang, Z.; Tu, Y.; Frohn, M.; Zhang, J .; Shi, Y. J . Am. Chem.
Soc. 1997, 119, 11224. (b) Cao, G.-A.; Wang, Z.-X.; Tu, Y.; Shi, Y.
Tetrahedron Lett. 1998, 39, 4425. (c) Zhu, Y.; Tu, Y.; Yu, H.; Shi, Y.
Tetrahedron Lett. 1998, 39, 7819. (d) Shi, Y.; Wang, Z.-X. J . Org. Chem.
1998, 63, 3099.
(19) Gao, Y.; Hanson, R. M.; Klunder, J . M.; Ko, S. Y.; Masamune,
H.; Sharpless, K. B. J . Am. Chem. Soc. 1987, 109, 5765.
(20) The minor diastereomer was produced in only 32% ee.
(21) Kuwajima, I.; Doi, Y. Tetrahedron Lett. 1972, 13, 1163.
(22) Dyer, U. C.; Robinson J . A. J . Chem. Soc., Perkin Trans. 1 1988,
53.
(23) Takano, S.; Morimoto, M.; Satoh, S.; Ogasawara, K. Chem. Lett.
1984, 1261.

2518 J . Org. Chem., Vol. 67, No. 8, 2002
McDonald et al.
Sch em e 2. En d o,En d o-Oxa cycliza tion of
Diep oxyk eton e 13
We sought to determine if the carbonyl-containing side
chains of substrates 13, 31, and 32 were serving as
terminating nucleophiles, as we had originally hypoth-
esized,25 to provide the tertiary hydroxyls in products 26,
33, and 34. Alternatively, these tertiary hydroxyls might
instead have arisen from stereoselective addition of water
upon aqueous quenching. However, diepoxide substrates
tethered to a carbonate functional group would be
expected to provide a cyclic carbonate fused to the
oxepane ring if the carbonyl was terminating the tandem
oxacyclization reaction. The first experiment with methyl
carbonate diepoxide 35 afforded a mixture of two major
products, one of which was characterized after acylation
as the expected bicyclic carbonate 36 (Scheme 4). The
atom connectivity and relative stereochemistry of 36 were
unambiguously determined by X-ray crystallography.26
The other product, 37, clearly retained the acyclic methyl
carbonate side chain but appeared to be a cyclic oxepane
product. Exchange of the acyclic carbonate of 37 for
acetate showed that the resulting compound 38 was
similar to but slightly different from oxepane 33, and the
presence of covalently bound fluorine was then confirmed
by 19F NMR spectroscopy of 37. However, the tert-butyl
carbonate 39 cleanly underwent endo,endo-oxacyclization
to provide bicyclic product 40 in good yield, which was
converted by acylation into the identical compound 36
produced by oxacyclization of the methyl carbonate
diepoxide substrate 35. Close inspection of the byproducts
arising from oxacyclization of 39 revealed the formation
of the tert-butyl ether 41 as a minor side product, which
apparently results from tert-butyl group transfer from the
acyclic carbonate. On one occasion, a trace amount (<5%)
of the fluoride corresponding to 37 was obtained from
oxacyclization of tert-butyl carbonate diepoxide substrate
39.
Sch em e 3. En d o,En d o-Oxa cycliza tion of
Ger a n iol-Der ived Diep oxid es 31 a n d 32
Similar oxacyclization results were observed with the
geranylacetic acid-derived diepoxides. The methyl ester
diepoxide 23 afforded two products after acylation, the
oxepane 43 and a single isomer of the bicyclic lactone
44. As the tert-butyl ester diepoxide was produced as
racemic mixture of diastereomers 21/22, oxacyclization
provided the lactone 44 combined with a stereoisomeric
product 45; however, the open-chain tert-butyl ester-
oxepane corresponding to 43 was not found. These results
indicate that intramolecular oxygen participation occurs
in the course of oxacyclization, although the hydrolysis
product (lactone vs acyclic ester) is partially dependent
on the nature of the ester substituent (tert-butyl vs
methyl). The oxygen of the carbonyl side chain could also
be positioned five atoms away from the tertiary center
of the proximal epoxide, demonstrated by oxacyclization
of tert-butyl carbonate diepoxide 46 (from 25) to provide
spirocyclic carbonate-oxepane 47 in 75% yield.
yield after acylation of the crude reaction mixture, along
with diastereomer 34 in 10% yield (Scheme 3). For this
substrate, B(C6F5)3-catalyzed cyclization provided only
20% yield of 33. Unambiguous confirmation of the atom
connectivity and relative stereochemistry was obtained
by X-ray diffraction of a single crystal of 33.24 As a
demonstration of the stereospecificity of the oxacycliza-
tion process, the diastereomeric diepoxide 18 (4.6:1 dr)
was also acylated to 32 and cyclized with BF3‚OEt2, which
after acylation of the crude reaction mixture resulted in
predominant formation of oxepane 34 in 46% isolated
yield and diastereomer 33 isolated in 9% yield.
(25) The participation of carbonyl oxygens in stereospecific oxacy-
clizations of monoepoxides tethered to ketones and carboxylic esters
has been demonstrated. (a) Scheidegger, U.; Schaffner, K.; J eger, O.
Helv. Chim. Acta 1962, 45, 400. (b) Demole, E.; Wuest, H. Helv. Chim.
Acta 1967, 50, 1314. (c) Wasserman, H. H.; Barber, E. H. J . Am. Chem.
Soc. 1969, 91, 3674. (d) Costa, M. C.; Tavares, R.; Motherwell, W. B.;
Marcelo Curto, M. J . Tetrahedron Lett. 1994, 35, 8839. (e) Boeckman,
R. K.; Rico Ferreira, M.; Mitchell, L. H.; Shao, P. J . Am. Chem. Soc.
2002, 124, 190.
(24) Colorless crystals of 33 (C14H24O6) were grown from slow
diffusion of hexanes into a solution of 33 in CH2Cl2. Data collection
was conducted at 298 K on a monoclinic crystal, P21; a ) 6.7703(3) Å,
b ) 15.9227(12) Å, c ) 7.3866(5) Å, â ) 99.446(4)°; V ) 785.49(9) Å3;
Z ) 2; R1 ) 0.0358, wR2 ) 0.1129, GOF ) 1.168.
(26) Colorless crystals of 36 (C13H20O6) were grown from slow
evaporation of a solution of 36 in a mixture of hexanes and CH2Cl2.
Data collection was conducted at 100 K on an orthorhombic crystal,
P212121; a ) 8.0614(8) Å, b ) 10.6645(11) Å, c ) 16.1512(16) Å; V )
1388.5(2) Å3; Z ) 4; R1 ) 0.0419, wR2 ) 0.0929, GOF ) 1.123.

Biomimetic Synthesis of Fused Polycyclic Ethers
J . Org. Chem., Vol. 67, No. 8, 2002 2519
Sch em e 4. Dem on str a ted In tr a m olecu la r
P a r ticip a tion of Ca r bon yl Oxygen in
En d o,En d o-Oxa cycliza tion s of Ca r bon a te a n d
Ester -Su bstitu ted Diep oxid es
Sch em e 5. Oxa cycliza tion s of Diep oxid es Der ived
fr om Hom oger a n iol
characterizable compound found to be the tetrahydro-
furan product 49 (Scheme 5). In this case, it appears that
the placement of the carbonyl oxygen of the acetate is
now too distant from the tertiary center of the proximal
epoxide to favor endo-regioselectivity. However, the tet-
rahydropyranyloxy diepoxide 50 (prepared by double
m-CPBA epoxidation of the THP derivative of homoge-
raniol) afforded the bicyclic product 51, consistent with
initial endo-selective tetrahydrofuran formation followed
by exo-cyclization to close the six-membered ring.
The effect of alkyl substituents on the epoxide moieties
was also explored with 1,5-diepoxide substrates 52,28 53,
56,29 and 58.30 Although we had anticipated that oxa-
cyclization of substrate 52 would have occurred with
sequential exo- and endo-selectivity (with each nucleo-
philic addition occurring at the more substituted carbon
of each epoxide), we instead observed formation of meso-
tetrahydrofuran diol 54 as the major product (Scheme
6). Similar results were obtained with bis-tert-butyl
carbonate substrate 53. The reaction rates for these
reactions (based on TLC analysis) were much slower than
those for substrates previously described. Thus, it is
possible that formation of products 54 and 55 resulted
only upon aqueous quenching of the reaction mixture,
upon either TLC analysis of an aliquot or workup of the
entire reaction mixture. Indeed, compound 54 is formed
in 65% isolated yield from the reaction of diepoxide 52
with aqueous perchloric acid in THF. Substrate 56
bearing one trisubstituted epoxide and one disubstituted
epoxide also afforded tetrahydrofuran 57 as the major
product upon BF3-promoted reaction and acylation of free
hydroxyl groups (which assisted in product characteriza-
tion). Substrate 58 (two disubstituted epoxides) gave an
unusual product (assigned as structure 60), which ex-
hibited NMR characteristics of six-membered ring forma-
tion but a mass spectrum consistent with a dimeric
structure.
Tr i- a n d Tetr a cycliza tion Rea ction s. We subse-
quently demonstrated this strategy in formation of trans-
fused bisoxepane compounds from multiple oxacycliza-
(28) The diepoxide diol precursor to substrates 52 and 53 is
known: (a) Hoye, T. R.; J enkins, S. A. J . Am. Chem. Soc. 1987, 109,
6196-6198. (b) Lindel, T.; Franck, B. Tetrahedron Lett. 1995, 36,
9465-9468.
(29) Prepared from trans-5-hepten-2-one (Cane, D. E.; Thomas, P.
J . J . Am. Chem. Soc. 1984, 106, 5295. J ung, M. E.; Angelica, S.;
D’Amico, D. C. J . Org. Chem. 1997, 62, 9182) in four steps: (1)
trimethyl phosphonoacetate, NaOMe, THF, reflux; (2) LiAlH4, THF;
(3) Shi epoxidation; (4) Ac2O, Et3N, catalytic DMAP, CH2Cl2.
(30) Prepared from trans-4-hexenal (Oppolzer, W.; Siles, S.; Snowden,
R. L.; Bakker, B. H.; Petrzilka, M. Tetrahedron 1985, 41, 3497) in four
steps: (1) (triphenylphosphoranylidene)acetaldehyde, C6H6, reflux; (2)
NaBH4, MeOH; (3) Shi epoxidation; (4) Ac2O, Et3N, catalytic DMAP,
CH2Cl2.
Oxa cycliza tion s of 1,5-Diep oxid e Su bstr a tes th a t
Affor d F ive- a n d /or Six-Mem ber ed Rin g P r od u cts.
Several other substrates were briefly explored that
demonstrated the limitations of this methodology. For
instance, BF3-etherate-promoted reaction of substrate
48 (from double Shi epoxidation of homogeraniol27 and
alcohol acylation) gave a complex mixture, with the only
(27) Leopold, E. J . Org. Synth. 1985, 64, 164.

2520 J . Org. Chem., Vol. 67, No. 8, 2002
McDonald et al.
Sch em e 6. Oxa cycliza tion s of 1,5-Diep oxid e
Su bstr a tes 52, 53, 56, a n d 58
F igu r e 4. Mechanism 1: tandem endo,endo,endo-oxacycliza-
tion via nucleophilic addition at tertiary centers.
Sch em e 8. Ta n d em En d o-Oxa cycliza tion s of Tr i-
a n d Tetr a ep oxid es
Sch em e 7. En a n tioselective Ep oxid a tion s of
Hyd r oxytr ien es a n d Hyd r oxytetr a en esa
on the basis of our experience with oxacyclizations of
substrates 39 and 46.
The reaction of triepoxide substrate 62 with BF3‚OEt2
was optimized to provide the tricyclic product 67 in 52%
yield,32 which was derivatized at the secondary alcohol
as a crystalline p-bromobenzoate (p-BrBz) ester 68
(Scheme 8). The atom connectivity and relative and
absolute configuration of compound 68 were unambigu-
ously determined by X-ray diffraction analysis.33 Like-
wise, the triepoxide substrate 64 provided the spiro-
carbonate bisoxepane 69 as the major characterized
product, and the tetraepoxide 66 could be effectively
converted into the tetracyclic product 70. This compound
contains trans-ring fusions between the each of the three
oxepane rings as well as the cyclic carbonate and adjacent
oxepane, reminiscent of the trans-syn-trans stereostruc-
ture of brevetoxins and structurally related fused poly-
cyclic ether marine natural products.
a
Reagents and Conditions: (a) 5 mol % Ti(O-i-Pr)4, 6 mol %
(-)-DET, TBHP, CH2Cl2, -25 °C. (b) 0.6 equiv 15, Oxone, MeCN/
DMM/H2O, pH 10.5, 0 °C. (c) (t-BuOCO)2O, Et3N, catalytic DMAP,
CH2Cl2 (62, 60%, 3 steps; 64, 42%, 3 steps; 66, 30%, 3 steps).
tions of tri- and tetraepoxide substrates. Given our
experiences described above, we began with two 1,5,9-
trienes that were appropriately substituted with alkyl
substituents. Sharpless enantioselective epoxidation of
the 2,3-alkene of farnesol (61) followed by Shi epoxida-
tions on the remaining alkenes and derivatization of the
primary alcohol provided the triepoxide 62 (Scheme 7).
Similar conversions of hydroxytriene 6331 and geranyl-
geraniol 65 afforded the respective triepoxide 64 and
tetraepoxide 66. The tert-butyl carbonate (Boc) terminat-
ing substituent was introduced in all three substrates
Crystals of suitable quality for X-ray diffraction analy-
sis were obtained from compound 68. The presence of the
(32) In the course of optimizing the oxacyclization of 62, we searched
for but have not yet found the tert-butyl ether byproduct corresponding
to 67 with R ) tert-butyl. In general, the material balance appears to
be a mixture of several very minor products that are less polar than
the polycyclic ether product. No epoxide-containing byproducts seem
to be present on the basis of an absence of resonances at ca. 3.0 ppm
in the crude 1H NMR spectra.
(33) Colorless crystals of 68 (C23H29BrO7) were grown from slow
diffusion of hexanes into a solution of 68 in CH2Cl2. Data collection
was conducted at 100 K on a monoclinic crystal, P21; a ) 10.3128(7)
Å, b ) 10.7363(7) Å, c ) 11.5382(8) Å, â ) 94.1510(10)°; V ) 1274.17-
(15) Å3; Z ) 2, R1 ) 0.0448, wR2 ) 0.0575, GOF ) 1.082.
(31) (a) Ogiso, A.; Kitasawa, E.; Kurabayashi, M.; Sato, A.; Taka-
hashi, S.; Noguchi, H.; Kuwano, H.; Kobayashi, S.; Mishima, H. Chem.
Pharm. Bull. 1978, 26, 7. (b) Koohang, A.; Coates, R. M.; Owen, D.;
Poulter, C. D. J . Org. Chem. 1999, 64, 6.

Biomimetic Synthesis of Fused Polycyclic Ethers
J . Org. Chem., Vol. 67, No. 8, 2002 2521
F igu r e 5. Mechanism 2: tandem endo,endo,endo-oxacycliza-
tion via nucleophilic addition at secondary centers.
p-bromobenzoate permitted not only determination of the
atom connectivity and relative stereochemistry but also
confirmation of the absolute stereochemistry for product
68. The unambiguous assignment of the absolute stereo-
chemistry for 68 has important mechanistic implications
for the tandem oxacyclization reaction. For instance, the
proposed mechanism (Figure 4) featuring BF3 activation
of the terminal epoxide, a cascade of intramolecular
nucleophilic additions by the other epoxide oxygens, and
termination by the carbonate oxygen nucleophile is
consistent with the stereostructure determined for 68 (p-
bromobenzoate ester of 67).
In contrast, an alternative mechanism (Figure 5) could
be considered, particularly in the case of inadvertent
addition of water. This mechanism features nucleophilic
addition of water or hydroxide to the less substituted
carbon of the terminal epoxide, generating a tertiary
alcohol that might react with the same regiochemistry
at the neighboring epoxide, until the reaction is termi-
nated by intramolecular alkoxide addition to the carbon-
ate carbon. However, this alternative mechanism is
conclusively disproven by the crystallographic determi-
nation of the absolute stereochemistry of 68, rather than
the enantiomer ent-68 (from acylation of ent-67, mech-
anism 2, Figure 5).
1
F igu r e 6. Key H NMR assignments for compounds 36, 42,
68, and 71.
Sch em e 9. Syn th esis a n d Ta n d em
En d o-P olycycliza tion of Alk en e-Diep oxid e 72
Although the tetracyclic bromobenzoate 71 did not
yield crystals suitable for X-ray diffraction analysis,34 the
structural and stereochemical assignment of 71 is based
1
on careful comparison of the H and 13C NMR spectra of
71 with the bicyclic acetate 36, the bromobenzoate
analogue 42, and tricyclic product 68, with the assign-
ments of 36 and 68 secured by crystallography. Coupling
constants and chemical shifts are similar for hydrogen
atoms located at the same relative positions in each
compound, indicating similar ring conformations for
compounds 36, 42, 68, and 71. Specifically, the hydrogen
atoms located at the ring fusions between oxepanes in
68 and 71 exhibit similar chemical shifts and coupling
constants (Figure 6). In addition, the crystal structures
of bicyclic acetate 36 and tricyclic bromobenzoate 68 show
that the terminal ester substituent is in a pseudoaxial
position, and the hydrogen atom attached to the carbon
at this terminus of each polycyclic product 36, 42, 68,
and 71 exhibits similar NMR chemical shifts for each
compound.
10,11-alkenes, thus providing easy access to the di-
epoxide-alkene substrate 72. To test a hypothesis com-
bining biomimetic alkene and epoxide cyclizations, we
subjected the tert-butyl carbonate derivative 72 to our
standard cyclization conditions to obtain a major tricyclic
product. X-ray crystallographic structure determination
unambiguously confirmed the atom connectivity and
stereochemistry of compound 73 (Scheme 9).35,36
We observed that Shi enantioselective epoxidation of
farnesol (61) proceeded more rapidly with the 6,7- and
(34) After this manuscript was submitted, we obtained colorless
crystals of 71 (C28H37BrO8) from slow diffusion of heptane into a
solution of 71 in CH2Cl2. Data collection was conducted at 100 K on
an orthorhombic crystal, P212121; a ) 7.08380(10) Å, b ) 12.1709(3)
Con clu sion s
A biomimetic strategy has been developed to synthesize
trans-fused polyoxepanes via tandem endo-regioselective
Å, c ) 32.0281(7) Å; V ) 2761.34(10) Å3; Z ) 4; R1 ) 0.0439, wR2
0.0560, GOF ) 1.011.
)

2522 J . Org. Chem., Vol. 67, No. 8, 2002
McDonald et al.
with an ice bath. A solution of Oxone (2.55 g, 4.15 mmol) in
aqueous Na2 (EDTA) (4 × 10-4 M, 20.0 mL) and a solution of
K2CO3 (2.49 g, 18.0 mmol) in water (20.0 mL) were added
dropwise through two separate addition funnels over a period
of 1.5 h. The pH value of the reaction mixture was kept
between 10.3 and 10.5. At this point, the reaction was
immediately quenched by addition of EtOAc and water. The
mixture was extracted with EtOAc, washed with brine, and
dried over MgSO4. After removal of volatiles by rotary evapo-
ration, the crude product 17 was dissolved in anhydrous CH2-
Cl2 (20 mL) and sequentially treated with DMAP (10 mg),
triethylamine (2.10 mL, 15 mmol), and acetic anhydride (0.55
mL, 6.0 mmol). The reaction mixture was stirred for 1 h at
room temperature and then washed with water (2 × 7 mL)
and brine (7 mL); the organic layer was dried over MgSO4 and
then concentrated to give the crude acetylated product.
Purification by silica gel flash chromatography (6:1 hexanes/
EtOAc) provided compound 31 (630 mg, 93% yield). The
enantiomeric excess was determined on the basis of the
corresponding benzoate ester derived from 17 to be 98% ee by
HPLC analysis. Conditions for HPLC analysis: column,
Chiralpak OD(H) 0.46 cm φ × 25 cm; eluent, 9:1 hexanes/2-
propanol; flow rate, 1.0 mL/min; 254 nm; temp ) 25 °C; R,R,R-
isomer, tR ) 6.97 min; S,S,S-isomer, tR ) 16.0 min). IR (neat)
2964, 2930, 1744, 1378, 1235, 1037 cm-1; 1H NMR (300 MHz,
CDCl3) δ 4.31 (AB dd, J ) 12.0, 4.5 Hz, 1H), 4.04 (AB dd, J )
12.0, 6.9 Hz, 1H), 3.01 (dd, J ) 6.9, 4.5 Hz, 1H), 2.70 (m, 1H),
2.09 (s, 3H), 1.90-1.76 (m, 1H), 1.72-1.49 (m, 3H), 1.33 (s,
3H), 1.30 (s, 3H), 1.26 (s, 3H); 13C NMR (75 MHz, CDCl3) δ
170.06, 63.83, 63.45, 60.33, 59.49, 58.68, 34.97, 25.00, 24.54,
20.99, 18.86, 17.20. Anal. Calcd for C12H20O4: C, 63.14; H, 8.83.
Found: C, 63.16; H, 8.86.
and trans-stereoselective oxacyclization of 1,5-, 1,5,9-, and
1,5,9,13-polyepoxides by reaction with Lewis acids. De-
spite some limitations with regard to substituent pat-
terns and the nature of the nucleophilic terminating
group,37 such tandem oxacyclizations have great potential
in providing efficient syntheses of structurally complex
polycyclic ether natural products, particularly when one
considers the ease of substrate synthesis.
Exp er im en ta l Section
1
Gen er a l. H NMR and 13C NMR spectra were recorded at
either 300 MHz on a Varian Mercury-300 or 400 MHz on an
Inova-400 spectrometer. NMR spectra were recorded on solu-
tions in deuterated chloroform (CDCl3), with residual chloro-
form (δ 7.26 ppm for 1H NMR and δ 77.23 ppm for 13C NMR)
taken as the internal standard, and were reported in parts
per million (ppm). Abbreviations for signal coupling are as
follows: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet.
IR spectra were collected on a Mattson Genesis II FT-IR
spectrometer as neat films. Mass spectra (low- and high-
resolution FAB) were recorded on a VG 70-S Nier J ohason
Mass Spectrometer. Elemental analyses were performed by
Atlantic Microlab, Inc, Norcross, GA. Analytical Thin Layer
Chromatography (TLC) was performed on precoated glass-
backed plates purchased from Whatman (silica gel 60 F254; 0.25
mm thickness). Flash column chromatography was carried out
with silica gel 60 (230-400 mesh ASTM) from EM Science.
All reactions were carried out with anhydrous solvents in
oven- or flame-dried and nitrogen- or argon-charged glassware.
All anhydrous solvents except as mentioned were freshly
distilled. All the solvents used in workup extraction procedures
and chromatography were used as received from commercial
suppliers without prior purification. During reaction workup,
the reaction mixture was usually diluted to three times the
original volume and washed with an equal volume of water
and/or aqueous solutions as needed. Purified and redistilled
boron trifluoride diethyl etherate (BF3‚OEt2) was purchased
from Aldrich. All other reagents were purchased from Aldrich.
R ep r esen t a t ive P r oced u r es for Syn t h esis of P oly-
ep oxid e Su bstr a tes: (R,R,R)-Diep oxid e-a ceta te (31). To
a 250 mL three-neck round-bottom flask were added buffer
(0.05 M Na2B4O7‚10H2O in 4 × 10-4 M aqueous Na2(EDTA),
30 mL), acetonitrile (15 mL), dimethoxymethane (30 mL),
(2R,3R)-3-methyl-3-(4-methyl-3-pentenyl)oxiranemethanol ((+)-
2,3-epoxygeraniol,19 509 mg, 2.92 mmol), tetrabutylammonium
hydrogen sulfate (73 mg, 0.215 mmol), and 1,2:4,5-di-O-
isopropylidene-D-erythro-2,3-hexodiuro-2,6-pyranose (Shi’s cata-
lyst, 15,18 280 mg, 1.08 mmol). The reaction mixture was cooled
(R,R,R)-Diep oxid e-ter t-bu tyl Ca r bon a te (39). Diepoxy-
alcohol 17 (135 mg, 0.725 mmol, prepared as described above)
was dissolved in anhydrous CH2Cl2 (20 mL) and sequentially
treated with DMAP (5 mg), triethylamine (0.53 mL, 3.75
mmol), and di-tert-butyl dicarbonate (288 mg, 1.32 mmol). The
reaction mixture was stirred for 1 h at room temperature and
then washed with water (2 × 2 mL) and brine (2 mL); the
organic layer was dried over MgSO4 and then concentrated to
give the crude Boc-protected product. Purification by silica gel
flash chromatography provided compound 39 (191 mg, 92%)
as a colorless oil: IR (neat) 2977, 2932, 1744, 1459, 1373, 1279,
1
1257, 1164, 1095 cm-1; H NMR (400 MHz, CDCl3) δ 4.06-
4.17 (m, 2H), 2.99 (t, J ) 5.4 Hz, 1H), 2.66 (t, J ) 6.0 Hz, 1H),
1.80-1.72 (m, 1H), 1.69-1.50 (m, 3H), 1.43 (s, 9H), 1.27 (s,
3H), 1.24 (s, 3H), 1.20 (s, 3H); 13C NMR (100 MHz, CDCl3) δ
153.1, 82.4, 65.3, 63.5, 60.0, 58.9, 58.3, 34.6, 27.6, 24.7, 24.2,
18.5, 16.8. Anal. Calcd for C15H26O5: C, 62.91; H, 9.15.
Found: C, 63.00; H, 9.19.
(R,R,R,R,R)-Tr iep oxid e-ter t-bu tyl Ca r bon a te (62): IR
(neat) 2971, 2932, 1743, 1460, 1279, 1164, 861 cm-1; 1H NMR
(400 MHz, CDCl3) δ 4.18 (AB dd, J ) 11.6, 5.2 Hz, 1H), 4.11
(AB dd, J ) 12.0, 6.0 Hz, 1H), 3.01 (t, J ) 5.6 Hz, 1H), 2.74-
2.66 (m, 2H), 1.82-1.70 (m, 2H), 1.70-1.54 (m, 6H), 1.47 (s,
9H), 1.31 (s, 3H), 1.28 (s, 3H), 1.25 (s, 3H), 1.24 (s, 3H); 13C
NMR (100 MHz, CDCl3) δ 153.43, 82.81, 65.56, 63.97, 62.64,
60.58, 60.26, 59.21, 58.61, 35.32, 34.82, 27.88, 25.01, 24.74,
24.32, 18.82, 17.12, 16.77.
(R,R,R,R,R,R,R)-Tetr aepoxide-ter t-bu tyl Car bon ate (66):
1H NMR (400 MHz, CDCl3) δ 4.18 (dd, J ) 12.0, 5.2 Hz, 1H),
4.11 (dd, J ) 12.0, 6.0 Hz, 1H), 3.03 (t, J ) 5.6 Hz, 1H), 2.74-
2.64 (m, 3H), 1.80-1.50 (m, 12H), 1.47 (s, 9H), 1.30 (s, 3H),
1.28 (s, 3H), 1.24 (s, 3H), 1.24 (s, 3H), 1.23 (s, 3H); 13C NMR
(100 MHz) δ 153.1, 82.6, 65.3, 63.9, 63.7, 62.5, 62.4, 60.3, 60.0,
59.0, 58.3, 35.1, 35.0, 34.6, 27.6, 24.7, 24.5, 24.3, 24.1, 18.6,
16.9, 16.5 (2C).
Rep r esen ta tive P r oced u r es for BF 3-Ca ta lyzed En d o-
Oxa cycliza tion s: Oxep a n e-F u sed Cyclic Ca r bon a te (40),
t h e Byp r od u ct ter t-Bu t yl E t h er (41), a n d t h e Der ived
Aceta te (36) a n d p-Br om oben zoa te Ester s (42). Diepoxide-
tert-butyl carbonate 39 (248 mg, 0.87 mmol) was added to a
50 mL, oven-dried, round-bottom flask containing a magnetic
stir bar, and anhydrous CH2Cl2 (5 mL) was added under an
inert atmosphere. The solution was stirred and cooled to -40
(35) Colorless crystals of 73 (C16H26O5) were grown from slow
evaporation of a solution of 73 in CD2Cl2. Data collection was conducted
at 100 K on an orthorhombic crystal, P212121; a ) 7.19550(10) Å, b )
11.12840(10) Å, c ) 18.5031(2) Å; V ) 1481.62(3) Å3; Z ) 4; R1
0.0271, wR2 ) 0.0602, GOF ) 1.006.
)
(36) Several natural products exhibit similar cyclohexane structures
trans-fused to oxepanes, which might be derived from compound 73:
(a) Sipholenols: Carmely, S.; Kashman, Y. J . Org. Chem. 1983, 48,
3517. (b) Sodwanones: Rudi, A.; Goldberg, I.; Stein, Z.; Benayahu, Y.;
Schleyer, M.; Kashman, Y. Tetrahedron Lett. 1993, 34, 3943. (c)
Raspacionin: Cimino, G.; Epifanio, R.; Madaio, A.; Puliti, R.; Trivellone,
E. J . Nat. Prod. 1993, 56, 1622. (d) Rudi, A.; Yosief, T.; Schleyer, M.;
Kashman, Y. Tetrahedron 1999, 55, 5555. (e) Review: Ferna´ndez, J .
J .; Souto, M. L.; Norte, M. Nat. Prod. Rep. 2000, 17, 235.
(37) Endo-selective polyene carbocyclizations to fused polycyclohex-
anoid products exhibit similar substrate requirements. For recent
examples, see: (a) Corey, E. J .; Luo, G. L.; Lin, L. S. Angew. Chem.,
Int. Ed. 1998, 37, 1126. (b) Sen, S. E.; Zhang, Y. Z.; Smith, S. M. J .
Org. Chem. 1998, 63, 4459. (c) Zoretic, P. A.; Fang, H.; Ribeiro, A. A.
J . Org. Chem. 1998, 63, 7213. (d) Aggarwal, V. K.; Bethel, P. A.; Giles,
R. J . Chem. Soc., Perkin Trans. 1 1999, 3315. (e) Yang, D.; Ye, X.-Y.;
Gu, S.; Xu, M. J . Am. Chem. Soc. 1999, 121, 5579. (f) Bender, J . A.;
Arif, A. M.; West, F. G. J . Am. Chem. Soc. 1999, 121, 7443. (g)
Bogensta¨tter, M.; Limberg, A.; Overman, L. E.; Tomasi, A. L. J . Am.
Chem. Soc. 1999, 121, 12206. (h) Ishihara, K.; Ishibashi, H.; Yamamoto,
H. J . Am. Chem. Soc. 2001, 123, 1505. (i) Goeller, F.; Heinemann, C.;
Demuth, M. Synthesis 2001, 1114.

Biomimetic Synthesis of Fused Polycyclic Ethers
J . Org. Chem., Vol. 67, No. 8, 2002 2523
°C, and precooled BF3‚OEt2 in CH2Cl2 (4.5 mL of a 0.2 M
solution freshly prepared from redistilled BF3‚OEt2 and an-
hydrous CH2Cl2) was added dropwise. The reaction mixture
was stirred for 30 min while maintaining the temperature
between -40 and -50 °C, and the reaction was then quenched
with saturated aqueous NaHCO3 (2 mL). The reaction mixture
was allowed to warm to room temperature and washed with
CH2Cl2 (3 × 15 mL), and the combined organic layers were
dried over MgSO4. The mixture was purified by silica gel flash
chromatography (gradient elution from 10:1 to 1:3 hexanes/
ethyl acetate), to obtain 107 mg (54%) of the alcohol 40 and a
less polar fraction, which was repurified by flash chromatog-
raphy (2:1 pentane/ether) to furnish the tert-butyl ether 41
(10 mg, 4% yield). Data for alcohol 40: mp 151 °C dec; IR (neat)
3431 (br), 2975, 2935, 1720, 1461, 1404, 1382, 1344, 1257,
1232, 1153, 1131, 1099, 1081, 1070 cm-1; 1H NMR (400 MHz,
CDCl3) δ 4.29 (dd, J ) 10.8, 6.4 Hz, 1H), 4.11 (dd, J ) 10.8,
6.8 Hz, 1H), 3.93 (t, J ) 10.8 Hz, 1H), 3.88 (d, J ) 6.8 Hz,
1H), 2.80 (br s, 1H), 2.26 (td, J ) 13.3, 2.8 Hz, 1H), 1.94-1.84
(m, 1H), 1.76-1.62 (m, 1H), 1.70 (t, J ) 13.2 Hz, 1H), 1.39 (s,
3H), 1.23 (s, 3H), 1.07 (s, 3H); 13C NMR (100 MHz) δ 148.7,
84.2, 79.8, 74.5, 66.8, 64.1, 32.9, 27.8, 24.4, 22.0, 18.9. Partial
data for tert-butyl ether byproduct 41: IR (neat) 2977, 2937,
2874, 1759, 1463, 1404, 1381, 1366, 1349, 1289, 1253, 1225,
1209, 1190, 1152, 1130, 1102, 1072, 1020 cm-1; 1H NMR (400
MHz, CDCl3) δ 4.22 (dd, J ) 11.0, 6.8 Hz, 1H), 4.12 (t, J )
11.0, 6.8 Hz, 1H), 3.95 (t, J ) 10.8 Hz, 1H), 3.56 (d, J ) 7.2
Hz, 1H), 2.3 (td, J ) 13.4, 3.2 Hz, 1H), 1.82-1.62 (m, 3H),
1.41 (s, 3H), 1.17 (s, 12H), 1.12 (s, 3H); 13C NMR (100 MHz) δ
148.4, 83.7, 80.6, 74.7, 74.2, 66.9, 64.2, 33.3, 28.3, 28.0, 24.6,
23.4, 19.0.
(-40 °C) of triepoxide 62 (346 mg, 0.93 mmol) in anhydrous
CH2Cl2 (14 mL). The reaction was monitored by TLC (1:1
hexanes/ethyl acetate) and quenched after 30 min by addition
of 3 mL of saturated sodium bicarbonate. After the reaction
mixture was warmed to room temperature, the contents of the
flask were poured into an aqueous saturated NaHCO3 solution
and extracted with dichloromethane. The organic layers were
combined, dried over MgSO4 and filtered, and the solvent was
removed under reduced pressure. After column chromatogra-
phy (gradient elution from 10:1 to 1:3 hexanes/ethyl acetate),
the tricyclic alcohol product 67 was isolated (152 mg, 52%
yield): IR (neat) 3449 (br), 2977, 2936, 2877, 1752, 1453, 1406,
1382, 1359, 1272, 1254, 1203, 1153, 1104, 1069, 1035 cm-1
;
1H NMR (400 MHz, CDCl3) δ 4.16-3.95 (m, 3H), 3.73 (br s,
1H), 3.63 (dd, J ) 11.2, 2.4 Hz, 1H), 2.14 (br s, 1H), 2.14-1.72
(m, 6H), 1.50-1.34 (m, 2H), 1.39 (s, 3H), 1.23 (s, 3H), 1.19 (s,
3H), 1.04 (s, 3H); 13C NMR (100 MHz) δ 149.1, 83.2, 80.8, 78.1,
77.5, 75.7, 66.8, 63.4, 37.1, 34.7, 28.5, 27.9, 25.3, 22.0, 21.6,
16.1. Data for p-bromobenzoate ester (68, derived from 67):
mp 203-205 °C; [R]23 -10.2 (c 0.98, CHCl3); IR (KBr neat)
D
3130, 2985, 2359, 1750, 1724, 1271, 1100 cm-1; 1H NMR (400
MHz, CDCl3) δ 7.88 (d, J ) 8.4 Hz, 2H), 7.63 (d, J ) 9.2 Hz,
2H), 5.17 (d, J ) 6.8 Hz, 1H), 4.22-4.12 (m, 2H), 4.08-4.02
(m, 1H), 3.63 (dd, J ) 11.2, 2.4 Hz, 1H), 2.25 (dq, J ) 16.0, 2.8
Hz, 1H), 2.12-1.96 (m, 3H), 1.94-1.70 (m, 2H), 1.62-1.50 (m,
2H), 1.46 (s, 3H), 1.31 (s, 3H), 1.20 (s, 3H), 1.17 (s, 3H); 13C
NMR (100 MHz, CDCl3) δ 164.72, 148.82, 132.18, 130.95,
129.06, 128.64, 83.10, 80.60, 79.20, 78.66, 78.13, 66.90, 63.74,
38.14, 36.12, 28.53, 28.16, 23.18, 22.62, 21.91, 16.43.
F u sed Tr isoxep a n e-Cyclic Ca r bon a te (70): IR (neat)
3475 (br), 2975, 2935, 2873, 1750, 1462, 1404, 1381, 1302,
1252, 1212, 1134, 1104, 1069, 992, 886, 769, 737, 703 cm-1
;
In a smaller scale experiment, diepoxide-tert-butyl carbonate
39 (85 mg, 0.3 mmol) was cyclized to 40 by reaction with
BF3‚OEt2 in CH2Cl2 (1.0 mL of a 0.30 M solution) in CH2Cl2
(5 mL) for 20 min while maintaining the temperature between
-40 and -50 °C. After aqueous extractive workup as described
above and removal of volatiles by rotary evaporation, the crude
product 40 was dissolved in anhydrous CH2Cl2 (10 mL) and
sequentially treated with DMAP (1 mg), triethylamine (0.42
mL, 3 mmol), and acetic anhydride (0.11 mL, 1.2 mmol). The
reaction mixture was stirred for 1 h at room temperature and
then washed with water (2 × 1.5 mL) and brine (1.5 mL); the
organic layer was dried over MgSO4 and then concentrated to
give the crude acetylated product. Purification by silica gel
flash chromatography (gradient elution from 9:1 to 4:1 hex-
anes/EtOAc) provided compound 36 (48 mg, 60%). Data for
1H NMR (400 MHz, CDCl3) δ 4.20 (dd, J ) 10.4, 6.4 Hz, 1H),
4.05 (t, J ) 10.4 Hz, 1H), 3.92 (dd, J ) 10.4, 6.4 Hz, 1H), 3.78
(bs, 1H), 3.74 (dt, J ) 11.2, 5.2 Hz, 1H), 3.61 (dd, J ) 11.6, 3.2
Hz, 1H), 2.05-1.61 (m, 12H), 1.46 (s, 3H), 1.27 (s, 3H), 1.23
(s, 3H), 1.20 (s, 3H), 1.07 (s, 3H); 13C NMR (100 MHz) δ 148.3,
83.6, 81.6, 80.0, 78.4, 77.9, 76.4, 72.9, 67.0, 66.5, 40.3, 36.2,
35.4, 28.6, 27.5, 26.8, 25.6, 22.0, 20.4, 18.8, 15.8; FABMS m/z
405.4 (M + Li+); HRMS (FAB) calcd for C21H34O7Li (M + Li+)
405.2465, found 405.2485.
Ack n ow led gm en t. This research was supported by
National Science Foundation Grant CHE-9982400. We
also acknowledge the use of shared instrumentation
(NMR spectroscopy, X-ray diffractometry, mass spec-
trometry) provided by grants from the National Insti-
tutes of Health and the National Science Foundation.
We thank Prof. David J . Goldsmith for several interest-
ing and enlightening conversations during the course
of this research and Prof. Robert K. Boeckman for
bringing our attention to some of the papers cited in
ref 25.
36: mp 119-121 °C; [R]23 -13.28 (c 1.02, CHCl3); IR (KBr)
D
3133, 2361, 2339, 1756, 1629, 1401, 1245, 1106 cm-1; 1H NMR
(400 MHz, CDCl3) δ 5.02 (d, J ) 6.8 Hz, 1H), 4.19 (dd, J )
9.6, 6.0 Hz, 1H), 4.12-4.00 (m, 2H), 2.16 (s, 3H), 2.08-1.92
(m, 2H), 1.84-1.74 (m, 2H), 1.46 (d, J ) 0.8 Hz, 3H), 1.21 (s,
3H), 1.18 (m, 3H); 13C (100 MHz) δ 169.90, 148.08, 83.43, 79.34,
77.03, 66.87, 65.07, 34.08, 28.16, 22.34, 22.06, 21.15, 19.05.
Anal. Calcd for C13H20O6: C, 57.34; H, 7.40. Found: C, 57.26;
H, 7.40. The p-bromobenzoate ester 42 was prepared similarly,
except that p-bromobenzoyl chloride was used instead of acetic
anhydride. Partial data on 42: 1H NMR (400 MHz, CDCl3) δ
7.86 (dt, J ) 8.8, 2.0 Hz, 1H), 7.62 (dt, J ) 8.8, 2.0 Hz, 1H),
5.26 (d, J ) 7.2 Hz, 1H), 4.25 (dd, J ) 10.2, 6.0 Hz, 1H), 4.15
(dd, J ) 10.4, 6.0 Hz, 1H), 4.05 (t, J ) 10.2 Hz, 1H), 2.20-
1.78 (m, 4H), 1.49 (s, 3H), 1.26 (s, 3H), 1.22 (s, 3H).
Su p p or tin g In for m a tion Ava ila ble: Characterization
data for compounds 13, 21-23, 27, 32-34, 43-47, 49, 51, 54,
57, 59-60, 64, 69, 71-73; thermal ellipsoid figures, X-ray
crystallographic tables, and X-ray CIF files for compounds 33,
36, 68, 71, and 73; and 1H NMR spectra for compounds 40-
42, 54, 62, 64, 66-73. This material is available free of charge
via the Internet at http://pubs.acs.org.
F u sed Bisoxep a n e-Cyclic Ca r b on a t e (67). Boron tri-
fluoride dietherate (4.7 mL of an approximately 0.2 M solution
in CH2Cl2, 0.94 mmol) was added dropwise to a cooled solution
J O0110092