Efficient synthesis of okadaic acid. 1. Convergent assembly of the C15- C38 domain

Article abstract of DOI:10.1021/ja973287h

A convergent synthesis of the C15-C38 domain of the marine natural product okadaic acid is reported. This involved the preparation of intermediates representing the C16-C27 and C28-C38 portions of okadaic acid, their direct coupling, and elaboration to th

Full text of DOI:10.1021/ja973287h

J. Am. Chem. Soc. 1998, 120, 2523-2533
2523
Efficient Synthesis of Okadaic Acid. 1. Convergent Assembly of the
C15-C38 Domain
Rebecca A. Urbanek, Steven F. Sabes, and Craig J. Forsyth*
Contribution from the Department of Chemistry, UniVersity of Minnesota, Minneapolis, Minnesota 55455
ReceiVed September 18, 1997
Abstract: A convergent synthesis of the C15-C38 domain of the marine natural product okadaic acid is
reported. This involved the preparation of intermediates representing the C16-C27 and C28-C38 portions
of okadaic acid, their direct coupling, and elaboration to the complete C15-C38 intermediate. A C16-C27
intermediate bearing an aldehyde at C27 was constructed in 14-steps from methyl 3-O-benzyl-R-D-
altropyranoside. A C28-C38 intermediate with a primary alkyl bromide at C28 was prepared in 10 steps
from methyl (S)-3-hydroxy-2-methylpropionate. These fragments were then joined in ∼55% yield by conversion
of the bromide into an alkylcerium reagent then addition to a sensitive â,γ-unsaturated C27 aldehyde to give
a mixture of C27 carbinols (27R:27S ) 2.5:1). The configuration at C27 of the major coupling product was
inverted by a simple oxidation-reduction sequence to establish the 27S-configuration of okadaic acid.
Elaboration into a C15 â-keto phosphonate completed the synthesis of the fully functionalized C15-C38
portion of okadaic acid in 19 linear steps and ∼3% overall yield from methyl 3-O-benzyl-R-D-altropyranoside.
Introduction
natural products have since been identified as okadaic acid type
inhibitors of PP1 and PP2A. These include calyculin A,14
dephosphonocalyculin A,15 tautomycin,16 microcystin-LR,17
motuporin,18 cantharidin,19 and thyrsiferyl 23-acetate,20 all of
which have drawn considerable attention from the synthetic
community. Okadaic acid itself, however, has remained the
most widely used tool for probing the roles of these ubiquitous
enzymes in a wide variety of cellular processes.21-23 Recent
X-ray crystal structures of the PP1 catalytic subunit covalently
linked with microcystin24 and complexed with tungstate25 have
provided detailed structural information of the PP1 active site
and plausible binding domains for members of the okadaic acid
class of inhibitors. These results have been augmented by
molecular modeling26,27 and molecular biology experiments28,29
Okadaic acid (1) is a marine natural product originally isolated
from the Pacific and Caribbean sponges Halichondria okadai
and H. melanodocia, respectively.1 Subsequently, it was found
to have its biogenetic origins in the marine dinoflagellates
Prorocentrum lima, Dinophysis fortii, and D. acuminata.2,3 Since
the complex structures of okadaic acid1 and the related C9-
C10 episulfide acanthifolicin4 determined by X-ray crystal-
lography were published in 1981, the biomedical and commer-
cial importance of compounds of this class has been well-
established and widely recognized. In addition to being one of
the chief causative agents of diarrhetic shellfish poisoning,5,6
okadaic acid has been characterized as a potent nonphorbol ester-
type tumor promoter on mouse skin.7 In contrast to the phorbol
ester class of natural products that may perturb intracellular
signal transduction pathways by stimulating protein kinase C
serine/threonine kinase activity,8 okadaic acid was found to be
a potent inhibitor of protein serine/threonine phosphatases 1 and
2A (PP1 and PP2A, respectively).9 Okadaic acid’s inhibition
of PP1- and PP2A-like enzymes may account for the natural
product’s induction of a diverse array of acute cellular responses,
which, depending upon the cell type, may range from mitogenic
stimulation10 to apoptosis.11-13 A number of structurally diverse
(10) Ghosh, S.; Schroeter, D.; Paweletz, N. Exp. Cell Res. 1996, 227,
165.
(11) Inomata, M.; Saijo, N.; Kawashima, K.; Kaneko, A.; Fujiwara, Y.;
Kunikane, H.; Tanaka, Y. J. Cancer Res. Clin. Oncol. 1995, 121, 729.
(12) Morimoto, Y.; Ohba, T.; Kobayashi, S.; Haneji, T. Exp. Cell Res.
1997, 230, 181.
(13) Kawamura, K.-I.; Grabowski, D.; Weizer, K.; Bukowski, R.;
Ganapathi, R. Br. J. Cancer 1996, 73, 183.
(14) Ishihara, H.; Martin, B. L.; Brautigan, D. L.; Karaki, H.; Ozaki,
H.; Kato, Y.; Fusetani, N.; Watabe, S.; Hashimoto, K. Biochem. Biophys.
Res. Commun. 1989, 159, 871.
(15) Matsunaga, S.; Wakimoto, T.; Fusetani, N. Tetrahedron Lett. 1997,
38, 3763.
(16) MacKintosh, C.; Klumpp, S. FEBS Lett. 1990, 227, 137.
(17) MacKintosh, C.; Beattie, K. A.; Klumpp, S.; Cohen, P.; Codd, G.
A. FEBS Lett. 1990, 264, 187.
(18) Dilip de Silva, E.; Williams, D. E.; Anderson, R. J.; Klix, H.;
Holmes, C. F. B.; Allen, T. M. Tetrahedron Lett. 1992, 33, 1561.
(19) Li, Y.-M.; Casida, J. E. Proc. Natl. Acad. Sci. U.S.A. 1992, 89,
11867.
(20) Matsuzawa, S.; Suzuki, T.; Suzuki, M.; Matsuda, A. FEBS Lett.
1994, 356, 272.
(1) Tachibana, K.; Scheuer, P. J.; Tsukitani, Y.; Kikuchi, H.; Engen, D.
V.; Clardy, J.; Gopichand, Y.; Schmitz, F. J. J. Am. Chem. Soc. 1981, 103,
2469.
(2) Murakami, Y.; Oshima, Y.; Yasumoto, T. Bull. Jpn. Soc. Sci. Fish.
1982, 48, 69.
(3) Yasumoto, T.; Seino, N.; Murakami, Y.; Murata, M. Biol. Bull. 1987,
172, 128.
(4) Schmitz, F. J.; Prasad, R. S.; Gopichand, Y.; Hossain, M. B.; van
der Helm, D.; Schmidt, P. J. Am. Chem. Soc. 1981, 103, 2467.
(5) Murata, M.; Shimatani, M.; Sugitani, H.; Oshima, Y.; Yasumoto, T.
Bull. Jpn. Soc. Sci. Fish. 1982, 48, 549.
(21) Fujiki, H. Mol. Carcinog. 1992, 5, 91.
(6) Yasumoto, T.; Murata, M.; Oshima, Y.; Sano, M.; Matsumoto, G.
K.; Clardy, J. Tetrahedron 1985, 41, 1019.
(7) Suganuma, M.; Fujiki, H.; Suguri, H.; Yoshizawa, S.; Hirota, M.;
Nakayasu, M.; Ojika, M.; Wakamatsu, K.; Yamada, K.; Sugimura, T. Proc.
Natl Acad. Sci. U.S.A. 1988, 85, 1768.
(22) Cohen, P.; Holmes, C. F. B.; Tsukitani, Y. Trends Biochem. Sci.
1990, 15, 98.
(23) Schonthal, A. New Biol. 1992, 4, 16.
(24) Goldberg, J.; Huang, H.; Kwon, Y.; Greengard, P.; Nairn, A. C.;
Kuriyan, J. Nature 1995, 376, 745.
(8) Nishizuka, Y. Nature 1984, 308, 693.
(9) Bialojan, C.; Takai, A. Biochem. J. 1988, 256, 283.
(25) Egloff, M.-P.; Cohen, P. T. W.; Reinemer, P.; Barford, D. J. Mol.
Biol. 1995, 254, 942.
S0002-7863(97)03287-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/06/1998

2524 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
Urbanek et al.
Scheme 1
Figure 1. Cyclic conformation of okadaic acid.
Results and Discussion
Synthetic Plan. Okadaic acid is a functionally and topologi-
cally complex molecule. The 38 contiguous carbons of its
backbone, as well as single carbon branches are believed to
arise biosynthetically via a unique polyketide pathway.38-40
Among okadaic acid’s skeleton are 23 functionalized carbons
that include 17 stereogenic carbons and 3 spiroketal moieties.
Much of this functionality contributes to a cyclic conformation
of 1 in solution41,42 that is similar to the solid-state conformation1
of the o-bromobenzyl ester derivative (Figure 1). This
pseudomacrolide allows the hydroxyl group at C24 to form an
intramolecular hydrogen bond with the natural product’s C1
carboxylate moiety. Additional structural features that may
contribute to a rigidified cyclic conformation are the opportunity
for another hydrogen bond between the C2 hydroxyl and the
C4 oxygen, thermodynamically enforced anomeric configura-
tions at C8 and C19, the C14-C15 trans-alkene, and the C19-
C26 trans-dioxadecalin system.43 In particular, the C8 and C19
spiroketal carbons define two reinforcing 90° turns in the
pseudomacrolide domain. Extending from okadaic acid’s cyclic
core is a lipophilic domain that terminates in the C30-C38 1,7-
dioxaspiro[5.5]undecane system. This degree of structural
complexity contributes to the challenge presented for total
synthesis and to the unique potential of molecules based upon
the conformationally well-defined okadaic acid architecture to
delineate the specific structural requirements for phosphatase
binding and inhibition.
aimed at illuminating the interactions of okadaic acid type
inhibitors with their phosphatase receptors. In addition, earlier
work has implicated the importance of several of okadaic acid’s
functional groups for phosphatase inhibition.30-33 However,
despite substantial synthetic efforts34-37that have culminated in
a previous total synthesis34,35 of okadaic acid, comprehensive
and definitive studies to determine the structural basis of PP1
and PP2A inhibition by okadaic acid have been limited by a
lack of designed structural variants.
As part of a program aimed at understanding the molecular
interactions between 1 and its phosphatase receptors, we have
recently developed an efficient and flexible total synthesis of
okadaic acid from intermediates representing the C1-C14 (2)
and C15-C38 (3) portions of the natural product (Scheme 1).36
The full details of the synthesis of the advanced okadaic acid
intermediate 3 are described here. In an accompanying paper
the synthesis of the complementary fragment 2 and its utilization
in conjunction with 3 for the total synthesis of 1 are detailed.37
In the original total synthesis of okadaic acid,34,35 Isobe and
co-workers disconnected 1 into three fragments, representing
C1-C14 (A), C15-C27 (B), and C28-C38 (C) of the natural
product. Each fragment was assembled independently, then
joined sequentially. This resulted in a substantial total synthesis
effort that spanned 54 steps in the longest linear sequence.34,35
This included a 47 linear step preparation from triacetyl D-glucal
of a C15-C38 fragment (B and C) that was coupled to the
C1-C14 intermediate (A). Although laborious, this landmark
synthesis illustrated the utility of several synthetic methods for
reliable carbon-carbon bond formation and for cyclic and
acyclic stereocontrol. In particular, phenyl sulfone stabilized
carbanions were used to form the C14-C15, C18-C19, C27-
C28, and C34-C35 bonds;34,44,45 the C13 and C29 methyl
groups were installed stereoselectively using heteroconjugate
(26) Bagu, J. R.; Sykes, B. D.; Craig, M. M.; Holmes, C. F. B. J. Biol.
Chem. 1997, 272, 5087.
(27) Gauss, C. M.; Sheppeck, J.; Nairn, A. C.; Chamberlin, A. R. Bioorg.
Med. Chem. 1997, 5, 1751.
(28) Zhang, L.; Zhang, Z.; Long, F.; Lee, E. Y. C. Biochemistry 1996,
35, 1606.
(29) Huang, H.-B.; Horiuchi, A.; Goldberg, J.; Greengard, P.; Nairn, A.
C. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 3530.
(30) Nishiwaki, S.; Fujiki, H.; Suganuma, M.; Furuya-Suguri, H.;
Matsushima, R.; Iida, Y.; Ojika, M.; Yamada, K.; Uemura, D.; Yasumoto,
T.; Schmitz, F. J.; Sugimura, T. Carcinogenesis 1990, 11, 1837.
(31) Yanagi, T.; Murata, M.; Torigoe, K.; Yasumoto, T. Agric. Biol.
Chem. 1989, 53, 525.
(38) Yasumoto, T.; Torigoe, K. J. Nat. Prod. 1991, 54, 1487.
(39) Wright, J. L. C.; Hu, T.; McLachlan, J. L.; Needham, J.; Walter, J.
A. J. Am. Chem. Soc. 1996, 118, 8757.
(32) Takai, A.; Murata, M.; Torigoe, K.; Isobe, M.; Mieskes, G.;
Yasumoto, T. Biochem. J. 1992, 284, 539.
(33) Holmes, C. F. B.; Luu, H. A.; Carrier, F.; Schmitz, F. J. FEBS Lett.
1990, 270, 216.
(34) Isobe, M.; Ichikawa, Y.; Goto, T. Tetrahedron Lett. 1986, 27, 963.
(35) Isobe, M.; Ichikawa, Y.; Bai, D.-L.; Masaki, H.; Goto, T. Tetra-
hedron 1987, 43, 4767.
(36) Forsyth, C. J.; Sabes, S. F.; Urbanek, R. A. J. Am. Chem. Soc. 1997,
119, 8381.
(37) Sabes, S. F.; Urbanek, R. A.; Forsyth, C. J. J. Am. Chem. Soc. 1998,
120, 2534-2542 (following article in this issue).
(40) Norte, M.; Padilla, A.; Fernandez, J. J. Tetrahedron Lett. 1994, 35,
1441.
(41) Norte, M.; Gonzalez, R.; Fernandez, J. J.; Rico, M. Tetrahedron
1991, 47, 7437.
(42) Matsumori, N.; Murata, M.; Tachibana, K. Tetrahedron 1995, 51,
12229.
(43) Uemura, D.; Hirata, Y. In Studies in Natural Products Chemistry;
Rahman, A. U., Ed.; Elsevier: Amsterdam, 1988; Vol. 5; p 377.
(44) Ichikawa, Y.; Isobe, M.; Goto, T. Tetrahedron Lett. 1984, 25, 5049.

Synthesis of Okadaic Acid. 1
J. Am. Chem. Soc., Vol. 120, No. 11, 1998 2525
addition chemistry;45,46 and the C2 stereogenic center was set
by a highly stereoselective oxymercuration.47
Scheme 2
A more flexible and streamlined synthetic entry to the okadaic
acid system than that represented by the original synthetic effort
was desired for the purpose of rapidly generating a variety of
analogues to probe the structural requirements for phosphatase
binding and inhibition. Therefore, we designed a new and
efficient total synthesis which employed major skeletal discon-
nections similar to those used in the original synthesis.
Key features of our synthetic plan were to minimize the
number of transformations required to prepare each fragment,
to install a maximal degree of functionality into each advanced
synthetic intermediate so as to minimize the extent of post-
coupling transformations, to utilize direct and chemoselective
reactions to couple the fragments, and to rely upon thermody-
namic equilibration45,48 to set the configurations of the three
spiroketal moieties of 1.36 In particular, it was anticipated that
the natural configuration at C19 could be established readily at
a late stage by intramolecular ketalization of a C16 hydroxyl
upon a masked C19 ketone.49 In contrast to the Isobe synthesis
where the C16-C23 spiroketal was constructed prior to C14-
C15 bond formation, we chose to defer formation of this central
spiroketal moiety until after joining the C1-C14 and C15-
C38 fragments. Hence, our initial retrosynthetic dissection
involved cleavage of the central C16-C23 spiroketal to afford
an allylic alcohol at C16 and a mixed methyl acetal at C19.
The C16 carbinol could be derived from the diastereoselective
reduction of a C16 ketone, and the trans-C14-C15 alkene could
be obtained via a Horner-Wadsworth-Emmons coupling of
an aldehyde (2) and a â-keto phosphonate (3, Scheme 1).
The choice of a keto phosphonate-aldehyde pair would allow
for mild coupling50 in the presence of the intact C1 carboxylate
resident in 2, and provide the C14-C15 (E)-alkene stereose-
lectively. Further, the two final coupling partners en route to
1, C14 aldehyde (2) and C15-C38 â-keto phosphonate (3),
contain all of the functionality present in okadaic acid in
appropriately masked form. With C16 at the ketone, rather than
the alcohol oxidation state, and just benzyl ether and acetal
protecting groups present, only a few transformations would
be required to convert 2 and 3 into 1.
Dissection of intermediate 3 at the C27-C28 carbon-carbon
bond and retrosynthetic simplification of the â-keto phosphonate
moiety yields the C27 aldehyde 4 and C28 bromide 5. Ideally,
a direct coupling of aldehyde 4 with an unstablized primary
carbanion derived from 5 would generate the C27 carbinol with
a useful degree of stereoselectivity and without the previous
necessity of post coupling removal of a carbanion stabilizing
group.34,35 Of course, â,γ-alkenyl aldehyde 4 is expected to
be susceptible to base-induced enolization-conjugation to give
the corresponding R,â-unsaturated aldehyde.44,51 Nevertheless,
we anticipated that, if successful, the convergency of adding a
primary organometallic species derived from 5 directly to
aldehyde 4 to generate the C27 carbinol would provide a
substantial gain in synthetic efficiency. Ideally, only protection
of the C27 hydroxyl and elaboration of the C16 terminus to the
keto phosphonate would then be required to complete the
assembly of 3. The synthesis of okadaic acid thus began with
the preparation of aldehyde 4 and bromide 5, and an examination
of their direct and convergent coupling.
Synthesis of the C16-C27 Fragment (4). Examination of
the central C22-C26 ring of 1 immediately suggested that a
D-altropyranoside could provide much of the stereochemistry
(C23, C24, and C26) and functionality present in this portion
of the natural product. Conversion of D-altropyranose into
intermediate 4 (Scheme 1) would entail methylenation at C25,
R-selective C-glycosidation to form the C21-C22 bond, an-
nulation to construct the functionalized C19-C23 oxane ring,
and final functional group transformations to provide the C27
aldehyde (Scheme 2). While commercially available, the high
costs of D-altrose and its derivatives precluded the selection of
any of these sugars as an actual starting point for the synthesis
of 4. However, it was recognized that inversion of both the
C2 and C3 configurations of inexpensive D-glucose would
provide D-altrohexose with the correct stereochemistry for C23
and C24, respectively. Therefore, a two-step protocol based
upon conversion of methyl 4,6-di-O-benzylidene-R-D-glucopy-
ranoside (6)52 into the corresponding 2,3-anhydromannopyran-
oside (7),53 followed by regioselective opening of the oxirane
at C3 with sodium benzoxide was adopted to provide the
requisite altropyranoside54 (Scheme 2). In the event, treatment
of 7 with sodium benzoxide in benzyl alcohol gave an
approximate 1:4 ratio of gluco (8) and altro (9) isomers in ∼70%
combined yield.54 Altropyranoside 9 could be separated
conveniently by fractional crystallization from the minor glu-
coisomer (8), which arose from nucleophilic attack of sodium
benzoxide at C2 of 7. Although the oxirane moiety of 7 distorts
the conformation of the parent pyranoside, trans-diaxial-like
opening still predominates to generate 9. Hence, ready access
was established to large quantities of a chiral building block
for the C22-C27 portion of 1 with the correct absolute
configurations at C23, C24, and C26.
The next task was to append stereoselectively carbons C16-
C21 to the anomeric position of 9. An obvious approach to
accomplish this was via an R-selective C-glycosidation. Ini-
tially, we examined convergent C-glycosidations under Hosami-
Sakurai-type conditions55 using highly functionalized allylic
silanes that contained the entire six carbons of the C16-C21
side chain, with C16 and C19 functionalized as masked alcohol
and ketone moieties, respectively. However, construction of
the C16-C21 side chain three carbons at a time, although less
convergent, was more successful. The one-pot procedure of
Gray56 for C-glycosidation of methyl glycosides with allyltri-
(45) Isobe, M.; Ichikawa, Y.; Masaki, H.; Goto, T. Tetrahedron Lett.
1984, 25, 3607.
(46) Isobe, M.; Ichikawa, Y.; Bai, D.-L.; Goto, T. Tetrahedron Lett. 1985,
26, 5203.
(47) Isobe, M.; Ichikawa, Y.; Goto, T. Tetrahedron Lett. 1985, 26, 5199.
(48) Deslongchamps, P. Stereoelectronic Effects in Organic Chemistry;
Pergamon: Oxford, 1983.
(49) Only the natural C19 spiroketal configuration was reportedly
observed upon C16-C23 spiroketalization at a much earlier stage of the
original synthesis of 1.44
(50) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183.
(51) Ichikawa, Y.; Isobe, M.; Goto, T. Tetrahedron 1987, 43, 4749.
(52) Evans, M. E. Carbohydr. Res. 1972, 21, 473.
(53) Hicks, D. R.; Fraser-Reid, B. Synthesis 1974, 203.
(54) Kunz, H.; Weissmueller, J. Liebigs Ann. Chem. 1984, 66.
(55) Hosami, A. Acc. Chem. Res. 1988, 21, 200.
(56) Bennek, J. A.; Gray, G. R. J. Org. Chem. 1987, 52, 892.

2526 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
Urbanek et al.
Scheme 3
provided 20 with C19 at the ketone oxidation state required for
1. The C19 carbon of 20 was to remain at this oxidation state
throughout the duration of the synthesis. However, this
necessitated masking the carbonyl from nucleophilic attack or
premature spiroketalization. A mixed-methyl ketal seemed an
ideal choice for this. In addition to providing requisite
protection of the ketone, intramolecular mixed ketal formation
was expected to provide a rigid and stable trans-dioxadecalin
system with a well-defined single anomeric configuration.
Further, a mixed-methyl acetal would be expected to participate
readily in an acid-triggered C16-C23 spiroketalization at the
appropriate juncture late in the total synthesis. The targeted
mixed-methyl acetal (22) could be obtained directly from silyl
ether 20 by treatment with TsOH or CSA in methanol.
However, silyl ether cleavage was sluggish under these condi-
tions, requiring prolonged reaction times or forcing conditions,
both of which led to competitive cleavage of the p-methoxy-
benzyl ether and subsequent C16-C23 spiroketalization. In-
stead, a two-step process involving cleavage of the silyl ether
of 20 by treatment with TBAF, followed by ketalization of the
resultant keto alcohol 21 in acidic methanol was favored. The
C25-C27 anisylidene group was conveniently dispensed with
in the process. Acetal 22 produced under these conditions
methylsilane provided a convenient and efficient method for
installation of carbons 19-21. Acidic methanolysis removed
the benzylidene group from 9. The resultant triol 10 was treated
with N,O-bis(trimethylsilyl)trifluoroacetamide to generate the
tris-TMS ether derivative. Without isolation, the tris-TMS ether
was treated with trimethylsilyl trifluoromethanesulfonate and
allyltrimethylsilane to produce, after aqueous work up, prope-
nylated product 11 as the major component of a ∼10:1 mixture
of R: â epimers (Scheme 3). Conversion of crude 11 into the
corresponding anisylidene derivative facilitated product purifica-
tion and yielded the C19-C27 intermediate 12. Before install-
ing carbons C16-C18, the free C23 hydroxyl group of 12 was
strategically capped as a silyl ether to give 13. A routine
hydroboration-oxidation sequence then converted alkene 13
into aldehyde 15.
1
appeared as only one anomeric isomer by 500-MHz H NMR
spectroscopy, in analogy to a model of the C16-C23 spiroketal
moiety of 1 prepared previously.44
Treatment of 21 with acid under dehydrating conditions but
without sufficient methanol present led to formation of the glycal
corresponding to 22. In addition, the mixed methyl acetal was
susceptible to hydrolysis to give the corresponding hemiacetal
in the presence of acid.58 However, careful handling allowed
the mixed acetal moiety present in 22 to be carried forward
throughout the remainder of the synthetic sequence until called
upon for end-game spiroketalization.
In the course of converting keto alcohol 21 into methyl acetal
22, the central pyran ring is allowed to undergo a chair-chair
ring flip concomitant with loss of the anisylidene group.
Whereas the chair-chair conformation of 22 is buttressed by
anomeric stabilization at C19 and equatorial deployment of the
C19 alkyl side chain and C24 benzyloxy group, it places the
The remaining three carbons of the side chain were derived
from 1,3-propane diol via bromide 18 (Scheme 4). Conversion
of 18 into the corresponding organolithium followed by addition
to aldehyde 15 gave a 1:1 diastereomeric mixture of C19
carbinols 19. Treatment of 19 with Dess-Martin periodinane57
Scheme 4

Synthesis of Okadaic Acid. 1
J. Am. Chem. Soc., Vol. 120, No. 11, 1998 2527
C27 hydroxymethyl substituent axial with respect to the trans-
dioxadecalin system. Thus, enolization of a derived C27
aldehyde may either lead to epimerization to place the C27 side
chain in an equatorial position, or in the case of â,γ-unsaturated
carboxaldehyde (4), lead to endocyclic migration of the alkene
into conjugation to give 27.
well as other products of base-induced side reactions were
observed. Hence, there seemed no advantage to attempt to
optimize conditions for C27-C28 bond formation before
installing the C25 exocyclic methylene group.
C25 was set up for methylenation by selective silylation of
the primary alcohol of 22 to give silyl ether 23 (Scheme 4).
Treatment of 23 with Dess-Martin periodinane gave ketone
24, which could be methylenated using either Lombardo-
Takai59 or Wittig conditions to give 25. Whereas the Lom-
bardo-Takai olefination proceeded cleanly and without detect-
able epimerization, it required a large excess of reagent which
made workup and isolation of 25 difficult. Wittig olefination
according to Miljkovic and Glisin60 was trouble-free. TBAF-
induced removal of the silyl group from 25 provided primary
alcohol 26, which was converted into the corresponding
aldehyde 4 with Dess-Martin periodinane. As was the case
with axial carboxaldehyde 30, â-methylenated aldehyde 4 was
quite prone to isomerization, but in this case, isomerization to
the R,â-conjugated carboxaldehyde 27 occurred. The equatorial
carboxaldehyde 26-epi-4 was never observed. Thus, 4 was best
prepared from 26 immediately prior to use.
Aldehyde 4 was thus prepared in 14 steps from methyl 3-O-
benzyl-R-D-altropyranoside 10. Although preliminary studies
aimed at defining useful conditions for C27-C28 bond forma-
tion were performed with simple organometallic species, the
unique characteristics of carbanions generated from the C28-
C38 intermediate 5 dictated that optimization be carried out with
the actual synthetic intermediates. Hence, a facile synthesis of
the C28-C38 portion of 1 was developed.
This latter concern is related to the base-induced conversion
of 28 into 29 observed by Isobe and co-workers in model studies
associated with the original synthesis of 1.51 The reported
conjugation of 28 effected with Et3N at 55 °C raised concerns
that the â,γ-alkene might pose problems in a low-temperature
coupling of a stabilized carbanion with a â,γ-unsaturated
carboxaldehyde such as 28. Therefore, installation of the C25
alkene was deferred until after C27-C28 bond formation in
the Isobe synthesis of 1. To minimize the extent of postcoupling
transformations, however, we were interested in pursuing an
alternative sequence.
Synthesis of the C28-C38 Fragment (5). The C28-C38
domain of okadaic acid contains four stereogenic centers,
including a contiguous triad spanning C29-C31, as well as the
C34 spiroketal carbon. Because the two chair perhydropyran
rings of the C30-C38 spiroketal benefit from mutual anomeric
stabilization, while the bulkier C30 substituent adopts an
equatorial position in preference to the axially disposed C31
methyl branch, the natural product’s configuration at C34 was
expected to result preferentially from simple spiroketalization
and thermodynamic equilibration of the corresponding δ,δ′-
dihydroxy ketone.61,62 In contrast to the original synthesis of 1
where a phenylsulfonyl group was installed at C28 to stabilize
a carbanion for C27-C28 bond formation,34,35 an unstablized
primary carbanion at C28 was planned here to simplify the
overall synthetic sequence. Thus, the C28 bromide 5 was
selected as a precursor to a range of potential organometallic
species that could be used to explore formation of the C27-
C28 bond. Because the bromide functionality could be elabo-
rated from the corresponding primary alcohol, acid-induced
spiroketalization-equilibration of an acyclic trihydroxy ketone
(32) was planned (Scheme 5). Ketotriol 32 would derive from
the corresponding trisbenzyloxy R,â-unsaturated ketone (33),
which, in turn, could be obtained by Horner-Emmons coupling
of a C33-C38 â-keto phosphonate (35) and a C28-C32
aldehyde (34). All of the stereochemistry of 5 would derive
from the latter. This general approach could also provide
synthetic access to the related dinophysistoxin natural prod-
ucts6,63,64 by simply varying the methyl substitution at C31 and
C35 in the aldehyde and keto phosphonate.
Although the C26 R-proton of the targeted â,γ-unsaturated
carboxaldehyde 4 is both allylic and adjacent to the C27
carbonyl group, the C25 alkene is not expected to contribute
much to the kinetic acidity of the C26 proton. This is because
the local conformation of the rigid trans-dioxadecalin system
maintains the C26 carbon-proton σ-bond nearly perpendicular
to the π-system of the C25 alkene in 4. However, once enolized,
γ-protonation may readily occur to migrate the alkene into
conjugation (27). To determine the extent to which the presence
of the C25 alkene might complicate a direct coupling between
an unstabilized organometallic species and a â,γ-unsaturated
carboxaldehyde, we first performed baseline studies with
aldehyde derivatives of 22 that lacked the C25 alkene. Thus,
routine protecting and functional group manipulations converted
22 into â-silyloxy aldehyde 30 (Scheme 4). Compound 30 did
not undergo carbon-carbon bond formation with unstabilized
primary organolithium and magnesium species cleanly, even
at -78 °C. Instead, the equatorial C27 carboxaldehyde 31, as
(59) Lombardo, L. Tetrahedron Lett. 1982, 23, 4293.
(60) Miljkovic, M.; Glisin, D. J. Org. Chem. 1975, 40, 3357.
(61) Ichikawa, Y.; Isobe, M.; Masaki, H.; Kawai, T.; Goto, T. Tetrahe-
dron 1987, 43, 4759.
(57) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277.
Ireland, R. E.; Liu, L. J. Org. Chem. 1993, 58, 2899.
(58) This was first realized upon dissolving 22 in untreated CDCl3
containing trace amounts of HCl.
(62) Tsuboi, K.; Ichikawa, Y.; Isobe, M. Synlett 1997, 713.
(63) Hu, T.; Doyle, J.; Jackson, D.; Marr, J.; Nixon, E.; Pleasance, S.;
Quilliam, M. A.; Walter, J. A.; Wright, J. L. C. J. Chem. Soc., Chem.
Commun. 1992, 39.

2528 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
Urbanek et al.
Scheme 5
Scheme 7
42 was planned. This was to involve reductive cleavage of the
benzyl ethers and concomitant saturation of the alkene under
Pd-catalyzed hydrogenation conditions, followed by a discrete
acid-induced dehydration step. However, vigorous stirring of
enone 33 with 20% palladium hydroxide on carbon in absolute
ethanol under 1 atm of H2 not only reduced the alkene and
cleaved the three benzyl ethers, but also led directly to spiro-
ketalization to give 42. This in situ reduction-spiroketalization
sequence was not only reproducible, but also stereoselective,
giving the expected product of thermodynamic spiroketalization
(42) in greater than 90% yield. Rather than being promoted
by trace acid arising from the solvent,67 it is likely that the
commercially obtained Pd(OH)2 on carbon used in this sequence
provides a source of acid that promotes in situ thermodynamic
spiroketalization.68 In the original synthesis of 1, prolonged
exposure of benzyloxy spiroketal 44 to Pd-catalyzed hydrogena-
tion conditions designed to simply cleave the benzyl ether and
give 45, led to competitive formation of the bridged spiroketal
ketal isomer 46.61 No similar complication was observed with
42, however, presumably because unfavorable steric interactions
would be encountered between a C30 axial isopropyl-like
substituent and an adjacent C31 equatorial methyl group on the
pyran chair of hypothetical bridged spiroketal 47. In a recently
reported synthesis of the C28-C38 domain of 1,62 an alternative
ordering of events precludes the possibility of undesired bridged
spiroketal formation (cf. 46).
Scheme 6
Carbons 34-38 were derived conveniently from δ-valero-
lactone via methyl ester 3665 (Scheme 5). Addition of the
lithium anion of dimethyl methyl phosphonate to 36 provided
â-keto phosphonate 35. The stereochemical triad spanning
C29-C31 was established in a direct fashion using Keck’s
crotylstannane methodology. Silylation of commercially avail-
able methyl (S)-3-hydroxy-2-methylpropionate (37), followed
by reduction of the resulting ester 38 with DIBAL gave aldehyde
39 (Scheme 6). Keck had shown that the use of a silyl
protecting group in aldehydes similar to 39 was required for
high syn-syn diastereoselectivity in BF3‚OEt2-mediated crotyl
stannane additions.66 However, uniform protection of the
hydroxyl groups as benzyl ethers was desired here to facilitate
a subsequent reduction-spiroketalization sequence. Thus, the
crude product of crotyl addition to 39 was desilylated with
TBAF to afford diol 40 in 81% overall yield (∼18:1 syn-syn:
syn-anti) from 39. Silyl group removal also facilitated
chromatographic isolation of the crotyl addition product. After
40 was converted into dibenzyl ether 41, ozonolytic cleavage
of the alkene and reductive workup gave aldehyde 34.
Coupling of 34 and 35 under Masamune-Roush conditions50
provided trans-enone 33 without detectable epimerization of
the R-aldehyde stereogenic center of 34 (Scheme 7). A two-
step conversion of the trisbenzyloxy enone 33 into the spiroketal
(64) Sakai, R.; Rinehart, K. J. Nat. Prod. 1995, 58, 773.
(65) Hoye, T.; Kurth, M.; Lo, V. Tetrahedron Lett. 1981, 22, 815.
(66) Keck, G. E.; Abbott, D. E. Tetrahedron Lett. 1984, 25, 1883.
(67) Hayes, C. J.; Heathcock, C. H. J. Org. Chem. 1997, 62, 2678.
(68) Luss, E.; Forsyth, C. J. Unpublished results.

Synthesis of Okadaic Acid. 1
J. Am. Chem. Soc., Vol. 120, No. 11, 1998 2529
Having successfully constructed the fully functionalized
C28-C38 domain of 1 in eight-steps from methyl (S)-3-
hydroxy-2-methylpropionate, two additional steps were used to
convert the primary alcohol 43 into the corresponding bromide
5 (Scheme 7). With multigram quantities of 5 thus available,
direct C27-C28 bond formation could be thoroughly examined.
Scheme 8
Convergent C27-C28 Bond Formation. The direct addi-
tion of an organometallic species derived from bromide 5 to
aldehyde 4 would provide the fully functionalized C16-C38
portion of 1. However, stereochemical control over C27
carbinol formation and the previously established base sensitivity
of 4 were two important issues to be addressed. Intramolecular
metal chelation involving the C27 aldehyde and the R-pyranyl
oxygen would be expected to preferentially reveal the pro-S
face of the aldehyde to nucleophilic species. Hence, the natural
product’s 27S-configuration might result from a chelation-
controlled addition of a C28 nucleophile to the C27 aldehyde.
ucts. However, the organocerium species69 derived from
lithium-halogen exchange of 5 followed by addition of a THF
70
suspension of CeCl3 to the organolithium at -78 °C added
successfully to aldehyde 4 to give the C27 carbinols 48 and 49
(2.5:1 diastereomeric ratio, respectively) in useful yield (Scheme
8).
Several experiments were performed to assign the configura-
tions at C27 of the chromatographically separable adducts 48
and 49. First, an oxidation-reduction sequence analogous to
that used in the original synthesis of 1 and in the preparation
of [27-3H]okadaic acid71 was performed. Treatment of the major
coupling diastereomer 48 with Dess-Martin periodinane gave
ketone 50. With a minimum of handling, the ketone was
reduced with NaBH4 at 0 °C to generate a 9:1 mixture of 49
and 48, respectively. It had been established previously that
the major, or exclusive, diastereomers resulting from NaBH4
reduction of ketones similar to 50 have the same 27S config-
uration as 1.35,44 Next, carbinol 49 was subjected to the
modified Mosher ester analysis.72 Separate treatment of 49 with
either (R)- or (S)-MTPA chloride and DMAP gave the corre-
sponding (S)- and (R)-MTPA esters 51 and 52, respectively.
This led to an empirical assignment based upon the measured
∆δ values of the C26 methine proton that was in accord with
the above 27S configurational assignment. Final proof of
configuration was, of course, later provided by conversion of
49 into 1.36,37
Inversion of the C27 carbinol configuration via the conversion
of 48 into 49 via ketone 50 not only supported the stereochem-
ical assignment, but also provided a facile preparative route to
overcome the unfavorable coupling diastereoselectivity. Al-
though modest in both yield and diastereoselectivity, this
coupling-inversion protocol represented a reliable and useful
method for the direct joining of okadaic acid intermediates 4
and 5. Unfortunately, the use of an organocerium species for
the formation of the C27-C28 bond did not provide the desired
chelation control outcome to set the C27 stereogenic center.
While this methodology is useful for the present purpose, the
realization of a direct and diastereoselectively favorable coupling
between an unstablized C28 carbanion and a base-sensitive
aldehyde will require further study.
However, for any nucleophilic species to be useful for C27-
C28 bond formation, enolization of the C27 aldehyde would
have to be avoided. Preliminary studies revealed that simple
Grignard reagents (EtMgBr, Ph(CH2)3MgBr) could be added
cleanly to 4 at low temperature, but with essentially no
stereoselectivity (∼1:1 27R:27S). The simple cuprate (n-
Bu)2CuCNLi also added cleanly to 4 to give only a slight excess
(<2:1) of the desired 27S-configuration. However, attempts
to convert 5 into such species were unsuccessful and frustrating.
The organolithium derived from 5 by metal-halogen exchange
with tert-butyllithium was itself short-lived and difficult to
handle. THF solutions of the generated organolithium could
be maintained at low temperature and added to 4, in poor yield
(30-35%) and nonstereoselectively (1:1 ds). Use of the
unmodified organolithium species gave predominantly R,â-
unsaturated aldehyde 27, as well as numerous other byproducts.
Attempts to trans-metalate the organolithium derived from 5 at
or below -78 °C with MgBr2, CuBr‚DMS, CuCN, or CrCl3‚
THF prior to addition of 4 led to combinations of poor yields,
unfavorable diastereoselectivities, and byproduct formation.
Similar attempts to trans-metalate at temperatures much above
-60 °C were also ineffective, yielding species that were either
too basic or insufficiently nucleophilic. In the most stereose-
lectively favorable case, the use of stoichiometric 2-thienyl-
CuCNLi in conjunction with the organolithium derived from 5
by metal-halogen exchange with tert-butyllithium appeared to
give only the desired (27S)-alcohol diastereomer (49), but in
∼10% yield and accompanied by 27, as well as polar byprod-
(69) Corey, E. J.; Ha, D.-C. Tetrahedron Lett. 1988, 29, 3171.
(70) Obtained by treatment of commercially available CeCl3‚7H2O under
conditions (see the Experimental Section) expected to generate the mono-
hydrate CeCl3‚H2O: Evans, W. J.; Feldman, J. D.; Ziller, J. W. J. Am.
Chem. Soc. 1996, 118, 4581.
(71) Levine, L.; Fujiki, H.; Yamada, K.; Ojika, M.; Gjika, H. B.; van
Vumakis, H. Toxicon 1989, 26, 1123.
(72) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 3, 4092.

2530 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
Urbanek et al.
This premature spiroketalization demonstrated the facility
with which the C16-C23 1,6-dioxaspiro[4.5]decane ring system
is formed from the mixed C19 methyl acetal, but it also diverted
useful material. Thus, considerable attention was paid to
preventing this side reaction from occurring, for example, by
adding Et3N to the crude product and the chromatography
solvents. Oxidation of the primary alcohol of 54 with Dess-
Martin periodinane buffered with NaHCO3 gave the aldehyde
55 uneventfully. Thereafter, addition of the lithium anion of
dimethyl methylphosphonate followed by a second oxidation
with Dess-Martin periodinane delivered the â-keto phosphonate
3.
The phenyl sulfone carbanion mediated coupling of a C27
aldehyde with a C28-C38 fragment that was used in the original
synthesis of 1 reportedly provided the (27S)-carbinol in 52%
overall yield, after removal of the carbanion stabilizing group
and correction of the C27 stereogenic center. Although the
efficiency of this sequence is comparable to that of the direct
coupling between 4 and 5 before the stereochemical adjustment
via oxidation-reduction, having the C25 alkene and an optimal
array of protecting groups already installed in 4 expedites the
advancement of 49 toward 1. More careful handling of the
ketone 50 and/or a lower temperature reduction35 than was
performed here may further enhance the efficiency of generating
49.
Preparation of a C15-C38 â-Keto Phosphonate (3). With
the central and C38 terminal fragments of okadaic acid joined,
preparation for the attachment of the C1-C14 domain required
only elaboration of the C16 terminus into a â-keto phosphonate.
This was preceded by conversion of the C27 hydroxyl of 49
into the corresponding benzyl ether 53 (Scheme 9). Oxidative
Conclusions
The present synthesis provides the C15-C38 okadaic acid
intermediate 3 in ∼19 linear steps from methyl 3-O-benzyl-R-
D-altropyranoside 9. The utilization of a readily available
D-altrose derivative54 for C22-C27, as well as the expeditious
assembly of the C28-C38 spiroketal-containing fragment
contribute substantially to the overall efficiency of this approach.
The direct, convergent coupling of the central and lipophilic
domains of 1 illustrated here may allow for the facile preparation
of analogues varying in substitution and functionalization in both
of these regions. Hence, the dinophysistoxin natural prod-
ucts,6,63,64 as well as nonnatural derivatives,62 which may be
designed to probe specific interactions with the phosphatase
receptors, may be accessed by variations of this synthetic theme.
Opportunities to improve upon this route include the correction
and enhancement of the stereoselectivity of C27-C28 bond
formation, and the development of a more convergent annulation
of C15-C21 upon the C22-C27 core. Nonetheless, the current
synthesis provides relatively succinct access to the fully
functionalized C15-C38 portion of okadaic acid. As described
in an accompanying paper, this advanced intermediate (3) was
elaborated into 1 in only five additional steps.37
Scheme 9
Experimental Section
General Methods. Unless otherwise noted, all reactions were
carried out under an argon or nitrogen atmosphere in oven-dried
glassware using standard syringe, cannula, and septa techniques.
Benzene, diethyl ether, toluene, and tetrahydrofuran were distilled from
sodium or potassium/benzophenone ketyl under N2 immediately prior
to use; dichloromethane, triethylamine, acetonitrile, allyltrimethylsilane,
oxalyl chloride, dimethyl sulfoxide, pyridine, and diisopropylethylamine
were distilled from CaH2. DMF was distilled from BaO under vacuum.
All other solvents were used as received. Pd(OH)2 on carbon [20%
by wt Pd(OH)2] was obtained from Aldrich Chemical Co. LiBr and
LiCl were dried by heating at 140 °C for 14 h at ∼0.4 Torr. Silica gel
chromatography of 22 and all subsequent intermediates bearing the C19
mixed-methyl acetal was performed with triethylamine (0.5 vol %)
present in the chromatography solvents, even if this was not specified
in experimental procedures given below. Analytical TLC was per-
formed with 0.25 mm EM silica gel 60 F254 plates. NMR spectra are
referenced to residual CHCl3 at 7.25 (1H) and 77.0 ppm (13C). The
mass spectrometers used show deviations of less than 5 ppm. Melting
points are uncorrected. Combustion analyses were performed by
M-H-W Laboratories (Phoenix, AZ).
removal of the p-methoxybenzyl group using DDQ73 liberated
the C16 hydroxyl to give 54. However, unless the DDQ reaction
mixture and the alcohol product were protected against low pH,
spontaneous formation of the C16-C23 spiroketal occurred to
1
give 56, which appeared as a single isomer by 500 MHz H
NMR spectroscopy, by analogy to 22.
(73) Horita, K.; Yoshioka, T.; Tanaka, T.; Oikawa, Y.; Yonemitsu, O.
Tetrahedron 1986, 42, 3021.

Synthesis of Okadaic Acid. 1
J. Am. Chem. Soc., Vol. 120, No. 11, 1998 2531
1. Synthesis of the C16-C27 Fragment. 1-Deoxy-1-C-(2-
propenyl)-r-D-3-O-benzyl-4,6-di-O-p-anisylidenealtropyranoside (12).
To a stirred room-temperature solution of 10 (10.8 g, 37.8 mmol) in
acetonitrile (150 mL) was added N,O-bis(trimethylsilyl)trifluoroacet-
amide (40.2 mL, 151 mmol) under Ar. The resultant solution was
heated to 70-80 °C for 2.5 h, and then cooled to room temperature
before allyltrimethylsilane (30.0 mL, 189 mmol) and trimethylsilyl
trifluoromethanesulfonate (36.4 mL, 189 mmol) were added. After the
solution was stirred at room temperature for 12 h, it was cooled to 0
°C and ethyl acetate (800 mL) and saturated aqueous NaHCO3 (200
mL) were added. The separated organic phase was washed with
saturated aqueous NaHCO3 (2 × 100 mL), H2O (2 × 100 mL), and
saturated aqueous NaCl (2 × 100 mL). The aqueous phases were
combined and extracted with ethyl acetate. The combined organic
phases were dried (Na2SO4), filtered, and concentrated. The residue
was suspended in dichloromethane and gravity-filtered through filter
paper, and the filtrate was concentrated. The resultant crude triol (37.8
mmol theor) was dissolved in CH2Cl2 (350 mL) and cooled to 0 °C
under Ar. p-Anisaldehyde dimethyl acetal (6.58 mL, 38.3 mmol), (()-
camphorsulfonic acid (404 mg, 1.74 mmol), and 4 Å molecular sieves
(∼50 g) were added, and the resultant mixture was allowed to warm
to room temperature and stir for 3.5 h, at which time the TLC showed
no remaining triol. Ethyl acetate (120 mL) and saturated aqueous
NaHCO3 (25 mL) were added, the mixture was filtered through a fritted
glass funnel and the separated organic phase was washed with saturated
aqueous NaHCO3 (30 mL), H2O (30 mL) and saturated aqueous NaCl
(30 mL). The aqueous phases were combined and extracted with ethyl
acetate (2 × 50 mL). The combined organic phases were dried (Na2-
SO4), filtered, and concentrated. Silica gel column chromatography
(hexanes-ethyl acetate, 5:1-2:1, v/v) of the residue gave 12 (10.1 g,
24.5 mmol, 65% from 10) as an off-white solid: mp 84-87 °C
(hexanes-ethyl acetate); Rf 0.25 (hexanes-ethyl acetate, 2:1, v/v);
75 MHz, 1:1 diastereomeric mixture) δ 168.0, 160.1, 159.2, 138.8,
130.5, 130.3, 130.2, 129.4, 129.3, 128.6, 128.3, 127.6, 127.56, 114.0,
113.9, 113.6, 102.2, 80.4, 80.1, 76.7, 73.5, 73.4, 72.7, 72.1, 71.6, 71.1,
70.3, 69.9, 59.9, 59.8, 55.3, 55.28, 35.1, 34.8, 34.7, 34.5, 26.4, 26.3,
25.9, 25.8, 25.5, 18.06, -5.00, -5.03. Anal. Calcd for C41H58O9Si:
C, 68.11; H, 8.09. Found: C, 68.28; H, 7.96.
Acetal 22. A solution of 21 (170 mg, 280 µmol), benzene (12 mL),
anhydrous methanol (5 mL), and p-toluenesulfonic acid monohydrate
(5.3 mg, 28 µmol) was heated at reflux under a Dean-Stark trap and
N2 for 3.5 h, at which point TLC showed no remaining 21. The solution
was cooled to room temperature, diluted with diethyl ether (50 mL),
and washed with 10% aqueous NaOH (5 mL), H2O (20 mL), and
saturated aqueous NaCl (20 mL). The combined aqueous phases were
extracted with ethyl acetate (3 × 25 mL), and the combined organic
phases were dried over Na2SO4, filtered, and concentrated to give crude
22 (143 mg) as a yellow oil. This was used without further purification.
Preparative TLC (hexanes-ethyl acetate, 1:5, v/v) provided an analyti-
cal sample of 22 as a white solid: mp 56-58 °C (hexanes-ethyl
acetate); Rf 0.22 (hexanes-ethyl acetate, 1:5, v/v); [R]23 ) -7.3 (c
D
0.55, CHCl3); IR (neat) 3292, 2956, 1613, 1514, 1454, 1248, 1101,
1030 cm-1; 1H NMR (CDCl3, 300 MHz) δ 7.34 (m, 5H), 7.25 (d, J )
8.5 Hz, 2H), 6.87 (d, J ) 8.5 Hz, 2H), 4.86 (d, J ) 11.8 Hz, 1H), 4.70
(d, J ) 11.8 Hz, 1H), 4.44 (s, 2H), 4.07 (dd, J ) 5, 9 Hz, 1H), 3.95
(m, 1H), 3.79-3.87 (m, 2H), 3.78 (s, 3H), 3.55 (m, 2H), 3.46 (t, J )
6 Hz, 2H), 3.32 (ddd, J ) 4.5, 10, 10 Hz, 1H), 3.20 (s, 3H), 2.88 (s,
1H), 2.19 (br s, 1H), 1.78-1.92 (m, 4H), 1.52-1.65 (m, 4H); 13C NMR
(CDCl3, 75 MHz) δ 159.2, 138.3, 130.6, 129.2, 128.5, 127.8, 127.6,
113.8, 99.4, 78.3, 76.7, 73.0, 72.6, 71.5, 70.0, 69.6, 68.3, 59.7, 55.3,
47.5, 32.4, 32.3, 25.2, 24.0. Anal. Calcd for C28H38O8: C, 66.91; H,
7.62. Found: C, 66.84; H, 7.82.
C25 Ketone 24. To a stirred room-temperature solution of 23 (225
mg, 365 µmol) in CH2Cl2 (5 mL) was added NaHCO3 (2.45 g, 29.1
[R]23 ) +57.9 (c 1.14, CHCl3); IR (neat) 3420, 2900, 1639, 1610,
57
D
mmol), followed by the Dess-Martin periodinane reagent (1.24 g,
1515, 1450, 1246, 1090 cm-1; 1H NMR (CDCl3, 500 MHz) δ 7.43 (d,
J ) 9 Hz, 2H), 7.26-7.38 (m, 5H), 6.91 (d, J ) 9 Hz, 2H), 5.79 (m,
1H), 5.55 (s, 1H), 5.12 (m, 1H), 5.10 (m, 1H), 4.93 (d, J ) 12.5 Hz,
1H), 4.65 (d, J ) 12.5 Hz, 1H), 4.27 (dd, J ) 5, 10 Hz, 1H), 4.20
(ddd, J ) 5, 9.5, 9.5 Hz, 1H), 4.03 (dd, J ) 2.5, 9.5 Hz, 1H), 3.95 (m,
2H), 3.85 (dd, J ) 7.5, 7.5 Hz, 1H), 3.82 (s, 3H), 3.73 (dd, J ) 10, 10
Hz, 1H), 2.88 (dddd, J ) 1.5, 7.8, 7.8, 15.5 Hz, 1H), 2.63 (ddd, J )
7, 7, 14 Hz, 1H), 1.96 (d, J ) 5.5 Hz, 1H); 13C NMR (CDCl3, 125
MHz) δ 160.1, 138.7, 134.8, 130.3, 128.3, 127.5, 127.4, 117.5, 113.7,
102.4, 79.2, 77.8, 76.3, 73.8, 70.5, 69.7, 60.4, 55.3, 34.1. Anal. Calcd
for C24H28O6: C, 69.89; H, 6.84. Found: C, 69.77; H, 6.69.
2.91 mmol). The resultant mixture was allowed to stir for 2 h. Diethyl
ether (50 mL), saturated aqueous NaHCO3 (10 mL), and 10% aqueous
Na2S2O3 (10 mL) were added, and the mixture was stirred until the
organic layer became clear. The separated organic phase was washed
with H2O (2 × 10 mL) and saturated aqueous NaCl (2 × 10 mL). The
aqueous phases were combined and extracted with diethyl ether (2 ×
25 mL). The organic layers were combined, dried over Na2SO4, filtered,
and concentrated. The residue was filtered through a pad of silica gel
with hexanes-ethyl acetate (2:1, v/v) and the eluant concentrated to
yield crude 24 (223 mg) as an oil. This was used without further
purification. Preparative TLC (hexanes-ethyl acetate, 2:1, v/v)
provided an analytical sample of 24 as a colorless oil: Rf 0.50
(hexanes-ethyl acetate, 2:1, v/v); [R]22D ) -17.3 (c 1.95, CHCl3); IR
(neat) 2954, 2857, 1731, 1612, 1513, 1462, 1249, 1098 cm-1; 1H NMR
(CDCl3, 500 MHz) δ 7.43 (d, J ) 8.5 Hz, 2H), 7.26-7.30 (m, 5H),
6.88 (d, J ) 8.5 Hz, 2H), 5.00 (d, J ) 12 Hz, 1H), 4.80 (d, J ) 12 Hz,
1H), 4.46 (s, 2H), 4.16 (ddd, J ) 6, 10, 10 Hz, 1H), 4.12 (dd, J ) 2.5,
2.5 Hz, 1H), 4.05 (d, J ) 10.5 Hz, 1H), 3.95 (m, 2H), 3.79 (s, 3H),
3.74 (dd, J ) 10, 10 Hz, 1H), 3.47 (t, J ) 6 Hz, 2H), 3.21 (s, 3H),
1.80-1.92 (m, 4H), 1.55-1.60 (m, 4H), 0.81 (s, 9H), 0.02 (s, 3H),
-0.05 (s, 3H); 13C NMR (CDCl3, 75 MHz) δ 206.8, 159.2, 138.2, 130.6,
129.2, 128.2, 127.7, 127.6, 113.8, 99.0, 83.6, 83.3, 74.5, 74.1, 72.6,
72.1, 70.0, 66.6, 55.3, 47.5, 32.2, 31.9, 25.9, 25.8, 25.5, 24.1, 18.0,
-5.7, -5.8. Anal. Calcd for C34H50O8Si: C, 66.42; H, 8.20. Found:
C, 66.36; H, 8.12.
C25 Alkene 25. To a room temperature solution of methyltri-
phenylphosphonium bromide (325 mg, 910 µmol) in toluene (8.5 mL)
under N2 was added potassium hexamethyldisilazide (1.45 mL of a
0.5 M solution in toluene, ∼728 µmol). The resulting deep yellow
mixture was heated to 90 °C for 30 min and then cooled to room
temperature before a solution of ketone 24 (365 µmol theor) in toluene
(2 mL) was added via cannula. The resultant solution was heated to
90 °C for 30 min and then cooled to room temperature before saturated
NH4Cl (4 mL) was added. The toluene was removed by rotary
evaporation, and the aqueous phase was extracted with ethyl acetate
(35 mL). The organic phase was washed with H2O (2 × 10 mL) and
saturated aqueous NaCl (2 × 10 mL). The combined aqueous phases
were extracted with ethyl acetate (2 × 25 mL), and the combined
1-Deoxy-1-C-[(R,S)-3-hydroxy-6-[(4-methoxybenzyl)oxy]hexanyl]-
r-D-3-O-benzyl-2-O-tert-butyldimethylsilyl-4,6-di-O-p-anisylidene-
altropyranosides (19). To a stirred -78 °C solution of 18 (404 mg,
1.55 mmol) in diethyl ether (7.8 mL) under Ar was added tert-
butyllithium (1.74 mL of a 1.7 M solution in pentane, 3.0 mmol). The
resulting solution was stirred for 20 min before a solution of 15 (423
mg, 780 µmol) in diethyl ether (7.8 mL) was added slowly via cannula.
The resultant solution was stirred for an additional 15 min before
saturated aqueous NH4Cl (5 mL) was added. After the mixture was
warmed to 0 °C, it was extracted with diethyl ether (25 mL), and the
separated organic phase was washed with saturated aqueous NH4Cl
(15 mL), H2O (2 × 15 mL) and saturated aqueous NaCl solution (2 ×
15 mL). The combined aqueous layers were extracted with diethyl
ether (2 × 25 mL). The organic layers were combined, dried over
MgSO4, filtered, and concentrated. Silica gel column chromatography
(hexanes-ethyl acetate, 2:1, v/v) of the residue gave 19 (402 mg, 555
µmol, 71%) as a 1:1 diastereomeric mixture and colorless oil: Rf 0.34
(hexanes-ethyl acetate, 2:1, v/v); IR (neat) 3457, 2930, 1614, 1515,
1464, 1249, 1098, 836 cm-1 1H NMR (CDCl3, 300 MHz, 1:1
;
diastereomeric mixture) δ 7.43 (d, J ) 8.5 Hz, 4H), 7.23-7.34 (m,
10H), 6.89 (d, J ) 7.2 Hz, 4H), 6.87 (d, J ) 8.2 Hz, 4H), 5.54 (s, 2H),
4.89 (d, J ) 12.2 Hz, 1 H), 4.88 (d, J ) 12.3 Hz, 1H), 4.61 (d, J )
12.3 Hz, 1H), 4.60 (d, J ) 12.2 Hz, 1H), 4.44 (s, 4H), 4.23 (m, 2H),
4.14 (m, 2H), 4.01 (m, 2H), 3.80 (s, 6H), 3.78 (s, 6H), 3.59-3.76 (m,
10H), 3.47 (t, J ) 5.6 Hz, 4H), 2.73 (d, J ) 3.8 Hz, 1H), 2.62 (d, J )
4.2 Hz, 1H), 2.29 (m, 2H), 1.50-1.70 (m, 14H), 0.86 (s, 18H), -0.01
(s, 3H), -0.02 (s, 3H), -0.04 (s, 3H), -0.05 (s, 3H); 13C NMR (CDCl3,

2532 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
Urbanek et al.
organic extracts were dried over Na2SO4, filtered, and concentrated.
Silica gel column chromatography (hexanes-ethyl acetate, 8:1-5:1,
v/v) of the residue gave 25 (189 mg, 308 µmol, 84% from 23) as a
colorless oil: Rf 0.33 (hexanes-ethyl acetate, 5:1, v/v); [R]25D ) -11
(c 0.89, CHCl3); IR (neat) 2940, 1737, 1610, 1585, 1510, 1460, 1245,
mmol) in acetonitrile (10 mL) was added via cannula. After the reaction
mixture was stirred for 6 h, it was diluted with ethyl acetate (250 mL),
washed with aqueous 5% HCl, H2O, and saturated aqueous NaCl (100
mL ea). The organic phase was dried over MgSO4, filtered, and
concentrated. The residue was purified by silica gel column chroma-
tography (hexanes-ethyl acetate, 10:1-5:1, v/v) to give 33 (2.318 g,
4.51 mmol, 87%) as a clear, colorless oil: Rf 0.39 (hexanes-ethyl
acetate, 5:1, v/v); [R]25D ) +3.2 (c 0.75, CHCl3); IR (neat) 3300, 3030,
2930, 1693, 1454, 1098, 1028, 986 cm-1; 1H NMR (CDCl3, 500 MHz)
δ 7.36-7.27 (m, 15H), 6.78 (d, J ) 8.5, 16.0 Hz, 1H), 6.11 (d, J )
16.0 Hz, 1H), 4.58 (d, J ) 11.5 Hz, 1H), 4.52 (d, J ) 11.5 Hz, 1H),
4.50 (s, 2H), 4.46 (d, J ) 3.5 Hz, 2H), 3.56 (dd, J ) 3.0, 7.5 Hz, 1H),
3.49 (t, J ) 6.5 Hz, 2H), 3.43 (dd, J ) 8.5, 8.5 Hz, 1H), 3.32 (dd, J
) 5.5, 9.5 Hz, 1H), 2.65 (m, 1H), 2.55 (t, J ) 7.5 Hz, 2H), 1.97 (m,
1H), 1.73-1 64 (m, 4H), 1.15 (d, J ) 6.5 Hz, 3H), 0.91 (d, J ) 7.0
Hz, 3H); 13C NMR (CDCl3, 125 MHz) δ 200.4, 149.2, 138.6, 138.3,
129.5, 128.4, 128.3, 128.2, 127.7, 127.6, 127.5, 127.3, 82.2, 73.1, 73.0,
72.9, 70.0, 40.4, 39.8, 36.6, 29.3, 20.9, 16.4, 11.0; ΗRMS calcd for
C34H43O4 [M + H]+ 515.3161, found 515.3181, calcd for C34H46NO4
[M + NH4]+ 532.3427, found 532.3425.
1
1090 cm-1; H NMR (CDCl3, 500 MHz) δ 7.40 (d, J ) 7.5 Hz, 2H),
7.26-7.34 (m, 5H), 6.88 (d, J ) 7.5 Hz, 2H), 5.42 (dd, J ) 2, 2 Hz,
1H), 5.06 (s, 1H), 4.91 (d, J ) 12 Hz, 1H), 4.79 (d, J ) 12 Hz, 1H),
4.45 (s, 2H), 4.30 (dd, J ) 5, 5 Hz, 1H), 4.19 (d, J ) 9.5 Hz, 1H),
3.80 (s, 3H), 3.78-3.86 (m, 2H), 3.65 (dd, J ) 9, 16 Hz, 1H), 3.47 (t,
J ) 6 Hz, 2H), 3.40 (dd, J ) 10, 10 Hz, 1H), 3.21 (s, 3H), 1.77-1.87
(m, 4H), 1.56-1.61 (m, 4H), 0.89 (s, 9H), 0.03 (s, 3H), 0.02 (s, 3H);
13C NMR (CDCl3, 75 MHz) δ 159.2, 143.3, 139.1, 130.6, 129.2, 128.3,
127.5, 127.4, 113.8, 111.8, 99.2, 80.7, 78.2, 73.9, 72.6, 71.0, 70.1, 65.2,
55.3, 47.4, 32.4, 32.3, 25.9, 25.7, 24.1, 18.3, -5.4. Anal. Calcd for
C35H52O7Si: C, 68.59; H, 8.55. Found: C, 68.49; H, 8.37.
2. Synthesis of the C28-C38 Fragment. (2S,3S,4R)-2,4-dimeth-
ylhex-5-ene-1,3-diol (40).66,74 To a stirred -98 °C solution of (S)-3-
[(tert-butyldimethylsilyl)oxy]-2-methylpropanal (39, 7.546 g, 37.4
mmol) in CH2Cl2 (300 mL) under Ar was added borontrifluoride
etherate (9.65 mL, 78.4 mmol) via syringe over 5 min. Tri-n-
butylcrotylstannane75,76 (15 mL, ∼37 mmol) was added via cannula
over 10 min. After 1 h, the solution was allowed to warm to room
temperature and stir for an additional 2 h before tetra-n-butylammonium
fluoride in THF (50 mL of a 1 M solution in THF, 50 mmol) was
added. After 1 h, the solution was cooled to 0 °C and diluted with
diethyl ether, and saturated aqueous NaHCO3 was added until the
aqueous phase was neutral to pH paper. The mixture was filtered, and
the separated organic phase was washed with H2O and saturated aqueous
NaCl (150 mL ea) and then concentrated by rotary evaporation. The
biphasic residue was extracted with ethyl acetate, and the combined
organic extracts were concentrated to a residue that was purified by
silica gel column chromatography (hexanes-ethyl acetate, 5:1-1:2,
v/v) to give 40 (4.335 g, 30.1 mmol, 81%), as a clear, colorless liquid:
Rf 0.21 (hexanes-ethyl acetate, 1:1, v/v); [R]23D ) +37 (c 0.7, CHCl3);
[2R(1′S),3R]-2-[2-Hydroxy-1-methylethyl]-3-methyl-1,7-dioxaspiro-
[5.5]undecane (42). A mixture of 33 (2.04 g, 3.97 mmol) and 20%
Pd(OH)2 on carbon (0.04 g, 0.3 mmol) in absolute ethanol (40 mL)
was stirred vigorously under 1 atm of H2 for 15 h. The mixture was
filtered through Celite with ethyl acetate. The filtrate was concentrated
and the residue was purified by silica gel column chromatography
(hexanes-ethyl acetate, 4:1, v/v) to give 42 (901 mg, 3.65 mmol, 92%)
as a clear, colorless oil: Rf 0.59 (hexanes-ethyl acetate, 1:1, v/v); [R]25
D
) +69 (c 1.55, CHCl3); IR (neat) 3402, 2936, 2875, 1452, 1383, 1181,
1
1071, 1044 cm-1; H NMR (CDCl3, 500 MHz) δ 3.68 (m, 1H), 3.65
(dd, J ) 5.0, 11.0 Hz, 1H), 3.56 (m, 1H), 3.54 (dd, J ) 2.5, 9.5 Hz,
1H), 3.48 (dd, J ) 6.5, 11.0 Hz, 1H), 2.03 (dddd, J ) 4.5, 4.5, 9.0,
9.0, Hz, 1H), 1.19-1.72 (m, 3H), 1.70-1.34 (m, 8H), 1.13 (d, J ) 6.5
Hz, 3H), 0.94 (d, J ) 6.5 Hz, 3H); 13C NMR (CDCl3, 125 MHz) δ
95.7, 72.6, 64.9, 60.4, 37.6, 35.9, 30.3, 28.1, 26.4, 25.4, 18.7, 14.4,
11.4; ΗRMS calcd for C13H25O3 [M + H]+ 229.1804, found 229.1794,
calcd for C13H28NO3 [M+NH4]+ 246.2069, found 246.2070.
1
IR (neat) 3312, 2967, 1467, 1325, 1128, 1094, 1037, 992 cm-1; H
NMR (CDCl3, 500 MHz) δ 5.63 (ddd, J ) 8.5, 10.0, 17.0 Hz, 1H),
5.08 (dd, J ) 1.5, 17.0 Hz, 1H), 5.00 (dd, J ) 1.5, 10.0 Hz, 1H), 3.74
(dd, J ) 4.0, 10.5, Hz, 1H), 3.69 (dd, J ) 5.0, 10.5, Hz, 1H), 3.58 (dd,
J ) 2.5, 9.0 Hz, 1H), 2.31 (m, 1H), 1.84 (m, 1H), 1.10 (d, J ) 6.5 Hz,
3H), 0.96 (d, J ) 7.0 Hz, 3H); 13C NMR (CDCl3, 75 MHz) δ 140.7,
114.7, 77.4, 67.9, 42.1, 36.5, 16.9, 8.9; ΗRMS calcd for C8H17O2 [M
+ H]+ 145.1229, found 145.1224.
3. Synthesis of the C15-C38 Fragment. C27-C28 Bond
Formation [(27R)-48 and (27S)-49]. CeCl3‚7H2O (714 mg, 1.92
mmol) was magnetically stirred in a 10-mL conical two-necked flask
at 150 °C and 0.4 Torr for 18 h. After the CeCl3 was cooled to 0 °C
under Ar (1 atm), THF (3.3 mL) was added. The resultant slurry was
allowed to warm to room temperature and stir for ∼5 h. Bromide 5
(176 mg, 605 µmol) was azeotropically dried from benzene, dissolved
in THF (3.4 mL), and the resulting solution was cooled to -78 °C
under Ar. tert-Butyllithium (616 µL of a 1.77 M solution in pentane,
1.09 mmol) was added dropwise under Ar. The resultant deep yellow
solution was stirred for 45 min, before the CeCl3/THF slurry prepared
above was added dropwise via syringe (16 Ga needle), resulting in a
brown suspension. After the suspension was stirred for 15 min, a
solution of aldehyde 4 (135 µmol theor, azeotropically dried from
benzene) in THF (2 mL) was added via cannula. After 30 min,
saturated aqueous NH4Cl (2 mL) was added, and the mixture was
allowed to warm to room temperature. THF (5 mL) was added and
the mixture was filtered through Celite, followed by ethyl acetate
washes. The organic solvents were removed by rotary evaporation,
and the aqueous phase was extracted with ethyl acetate (35 mL). The
separated organic phase was washed with H2O (1 × 10 mL) and
saturated aqueous NaCl (2 × 10 mL). The aqueous phases were
extracted with ethyl acetate, and the organic phases were combined,
dried over Na2SO4, filtered, and concentrated. Silica gel column
chromatography (hexanes-ethyl acetate, 5:1-2:1, v/v) of the residue
gave 48 and 49 (50.4 mg, 71.1 µmol, 53%) as a 2.5:1 mixture on the
(2S,3R,4S)-1,3-Bis(benzyloxy)-2,4-dimethylpentanal (34). A stream
of O2/O3 was bubbled through a solution of 41 (1.280 g, 3.95 mmol)
in CH2Cl2 (50 mL) at -78 °C until a faint blue color persisted. A
stream of N2 was then bubbled through the solution until it became
colorless. Triphenylphosphine (1.553 g, 5.93 mmol) was added, and
the mixture was allowed to warm to room temperature and stir for 1.5
h. The solution was concentrated and the residue purified by silica
gel column chromatography (hexanes-ethyl acetate, 10:1, v/v) to give
34 (1.041 g, 3.20 mmol, 81%) as a clear, colorless oil: Rf 0.72
(hexanes-ethyl acetate, 5:1, v/v); [R]25 ) +30 (c 1.95, CHCl3); IR
D
(neat) 3064, 3030, 2971, 1721, 1496, 1454, 1361, 1207, cm-1; 1H NMR
(CDCl3, 500Mz): δ 9.79 (s, 1H), 7.37-7.26 (m, 10H), 4.54 (d, J )
11.0 Hz), 4.49 (d, J ) 11.0 Hz, 1H), 4.47 (s, 2H), 4.00 (dd, J ) 5.0,
7.0 Hz, 1H), 3.45 (dd, J ) 7.0, 9.5 Hz, 1H), 3.36 (dd, J ) 5.0, 9.0 Hz,
1H), 2.69 (dt, J ) 5.0, 7.0 Hz, 1H), 2.04 (m, 1H), 1.16 (d, J ) 7.0 Hz,
3H), 1.01 (d, J ) 6.5 Hz, 3H); 13C NMR (CDCl3, 75 MHz) δ 204.4,
138.3, 138.2, 128.4, 128.3, 127.7, 127.6, 127.5, 73.9, 73.1, 72.7, 50.0,
36.8, 12.7, 9.4; ΗRMS calcd for C21H26O3 [M + H]+ 327.1960, found
327.1960.
(2S,3S,4R)-1,3,11-Tris(benzyloxy)-2,4-dimethylundec-5-en-7-
one (33). To a mixture of LiCl (381 mg, 8.27 mmol) and 35 (2.360 g,
7.52 mmol) in acetonitrile (90 mL) at room temperature under Ar was
added diisopropylethylamine (1.298 mL, 7.52 mmol). The resultant
mixture was stirred for 10 min before a solution of 34 (1.686 g, 5.17
1
basis of integration of the 300-MHz H NMR resonances at δ 5.14
and 5.06, respectively. An analytical sample of each diastereomer was
obtained as an oil upon further silica gel column chromatography. Data
1
for 48: Rf 0.35 (hexanes-ethyl acetate, 2:1, v/v); H NMR (CDCl3,
300 MHz) δ 7.26-7.41 (m, 7H), 6.88 (d, J ) 8.4 Hz, 2H), 5.58 (dd,
J ) 1.8, 1.8 Hz, 1H), 5.14 (s, 1H), 4.96 (d, J ) 12 Hz, 1H), 4.78 (d,
J ) 12 Hz, 1H), 4.45 (s, 2H), 4.13 (m, 1H), 3.96 (m, 1H), 3.89 (d, J
) 8.7 Hz, 1H), 3.80 (s, 3H), 3.69 (ddd, J ) 3, 11, 11 Hz, 1H), 3.57
(74) Mulzer, J.; Autenrieth-Ansorge, L.; Kirstein, H.; Matsuoka, T.;
Munch, W. J. Org. Chem. 1987, 52, 3784.
(75) Hull, C.; Mortlock, S. V.; Thomas, E. J. Tetrahedron 1989, 45, 1007.
(76) Still, W. C. J. Am. Chem. Soc. 1978, 100, 1481.

Synthesis of Okadaic Acid. 1
J. Am. Chem. Soc., Vol. 120, No. 11, 1998 2533
(m, 1H), 3.44 (m, 5H), 3.20 (s, 3H), 1.76-1.99 (m, 8H), 1.48-1.66
(m, 11H), 1.37-1.45 (m, 4H), 1.17 (d, J ) 6.6 Hz, 3H), 0.91 (d, J )
7.2 Hz, 3H); HRFABMS calcd for C41H57O8 [M - OCH3]+ 677.4053,
found 677.4080.
J ) 6.3 Hz, 3H), 0.88 (d, J ) 6.9 Hz, 3H); 13C NMR (CDCl3, 75
MHz) δ 201.3, 143.5, 138.7, 138.6, 128.3, 128.2, 127.6, 127.5, 127.4,
112.7, 98.7, 95.7, 85.0, 77.2, 75.1, 74.7, 73.6, 73.0, 70.5, 60.4, 47.5,
38.7, 35.9, 34.6, 32.2, 31.2, 30.4, 27.6, 27.4, 26.4, 25.5, 25.4, 18.8,
16.7, 10.8; HRFABMS calcd for C40H53O7 [M - OCH3]+ 645.3791,
found 645.3813.
Data for 49: Rf 0.26 (hexanes-ethyl acetate, 2:1, v/v); [R]23
)
D
+1.2 (c 1.4, CHCl3); IR (neat) 3480, 2930, 1610, 1510, 1450, 1240,
1090, 990 cm-1; H NMR (CDCl3, 300 MHz) δ 7.25-7.39 (m, 7H),
1
â-Keto Phosphonate 3. To a stirred -78 °C solution of dimethyl
methylphosphonate (38 µL, 0.35 mmol) in THF (1.5 mL) under Ar
was added tert-butyllithium (160 µL of a 1.77 M solution in pentane,
280 µmol) dropwise. The solution was stirred for 45 min before a
solution of 55 (22 mg, 33 µmol) in THF (1 mL) was added slowly via
cannula. The resultant pale yellow solution was stirred for an additional
45 min, at which time TLC showed no remaining 55. Saturated aqueous
NaCl (1 mL) was added, and the mixture was allowed to warm to room
temperature. THF was removed by rotary evaporation and the aqueous
residue was extracted with ethyl acetate (15 mL). The separated organic
phase was washed with H2O (2 × 3 mL) and saturated aqueous NaCl
(2 × 3 mL). The combined aqueous phases were extracted with ethyl
acetate, and the combined organic phases were dried over Na2SO4,
filtered, and concentrated. The residue was filtered through silica gel
with hexanes-ethyl acetate-triethylamine (1:5:0.3, v/v) and the filtrate
concentrated to yield crude â-hydroxy phosphonate 55a (29 mg) as an
oil, which was used without further purification.
6.88 (d, J ) 8.7 Hz, 2H), 5.47 (s, 1H), 5.06 (s, 1H), 4.90 (d, J ) 12
Hz, 1H), 4.77 (d, J ) 12 Hz, 1H), 4.46 (s, 2H), 4.00 (dd, J ) 10, 10
Hz, 1H), 3.90 (m, 2H), 3.80 (s, 3H), 3.69 (ddd J ) 3, 11, 11 Hz, 1H),
3.39-3.61 (m, 5H), 3.26 (dd, J ) 2, 10 Hz, 1H), 3.21 (s, 3H), 2.47 (s,
1H), 1.77-2.06 (m, 8H), 1.48-1.68 (m, 11H), 1.38-1.45 (m, 3H),
0.97 (d, J ) 6.3 Hz, 3H), 0.92 (d, J ) 6.9 Hz, 3H); 13C NMR (CDCl3,
75 MHz) δ 159.1, 147.0, 141.9, 138.6, 130.6, 129.2, 128.3, 127.6, 113.8,
99.3, 95.6, 85.1, 76.9, 75.1, 73.8, 72.6, 70.1, 69.9, 64.4, 60.4, 55.3,
47.5, 35.9, 35.2, 32.3, 31.0, 30.4, 27.4, 26.4, 25.5, 24.0, 18.8, 16.2,
10.7; HRFABMS calcd for C41H57O8 [M - OCH3]+ 677.4053, found
677.4003.
Alcohol 54. To a stirred room temperature solution of 53 (27 mg,
34 µmol) in CH2Cl2 (2.2 mL) in a 25-mL round-bottom flask was added
aqueous phosphate buffer (pH 7, 2.2 mL), tert-butyl alcohol (420 µL),
and DDQ (23 mg, 0.10 mmol). The flask was immersed and sonicated
in a H2O bath for 15 min then removed from the bath. Triethylamine
(100 µL) and diethyl ether (15 mL) were added, and the resulting
mixture was washed with saturated aqueous NaHCO3 (2 × 3 mL), H2O
(2 × 3 mL), and saturated aqueous NaCl (2 × 3 mL). The combined
aqueous phases were extracted with diethyl ether (2 × 5 mL), and
triethylamine (100 µL) was added to the combined organic extracts.
Drying over Na2SO4, filtration and concentration gave a residue that
was purified by silica gel column chromatography (hexanes-ethyl
acetate-triethylamine, 5:1:0.3, v/v/v) to give 54 (17 mg, 25 µmol, 75%)
and 53 (7.1 mg, 8.9 µmol) as colorless oils. Data for 54: Rf 0.14
To a stirred room temperature solution of 55a (33 µmol theor) in
CH2Cl2 (1 mL) was added NaHCO3 (120 mg, 1.4 mmol) and Dess-
Martin periodinane reagent57 (60 mg, 0.14 mmol). The resultant
mixture was stirred for 45 min, at which time TLC showed no remaining
55a. Diethyl ether (15 mL), saturated aqueous NaHCO3 (1 mL), and
10% aqueous Na2S2O3 (1 mL) were added, and the mixture was stirred
vigorously until the organic layer became clear. The separated organic
phase was washed with H2O (2 × 3 mL) and saturated aqueous NaCl
(2 × 3 mL). The aqueous phases were extracted with diethyl ether,
and the combined organic phases were dried over Na2SO4, filtered,
and concentrated. The residue was purified by silica gel column
chromatography (hexanes-ethyl acetate, 1:1-1:5, v/v) to give 3 (17
mg, 21 µmol, 64% from 55) as a pale oil: Rf 0.23 (hexanes-ethyl
acetate, 1:5, v/v); [R]26D ) +15 (c 1.7, CHCl3); IR (neat) 2950, 1715,
(hexanes-ethyl acetate, 2:1, v/v); [R]25 ) +19 (c 0.60, CHCl3); IR
D
1
(neat) 3430 (broad), 2930, 1495, 1450, 1350, 1230, 1090 cm-1; H
NMR (CDCl3, 300 MHz) δ 7.25-7.38 (m, 10H), 5.44 (dd, J ) 1.8,
1.8 Hz, 1H), 5.06 (s, 1H), 4.86 (d, J ) 12.3 Hz, 1H), 4.74 (d, J ) 12.3
Hz, 1H), 4.73 (d, J ) 11.1 Hz, 1H), 4.58 (d, J ) 11.1 Hz, 1H), 4.27
(d, J ) 7.8 Hz, 1H), 3.91 (m, 2H), 3.69 (m, 3H), 3.64 (m, 2H), 3.43
(dd, J ) 9.6, 9.6 Hz, 1H), 3.24 (s, 3H), 3.22 (m, 1H), 1.79-2.06 (m,
8H), 1.49-1.67 (m, 11H), 1.38-1.45 (m, 3H), 1.26 (s, 1H), 0.94 (d, J
) 6.3 Hz, 3H), 0.88 (d, J ) 6.9 Hz, 3H); 13C NMR (CDCl3, 75 MHz)
δ 143.5, 138.7, 128.3, 128.2, 127.63, 127.6, 127.5, 127.4, 112.7, 99.3,
95.7, 85.2, 77.3, 77.2, 75.1, 74.4, 73.6, 73.0, 70.6, 63.0, 60.4, 47.4,
35.9, 34.6, 32.3, 32.2, 31.2, 30.4, 27.4, 27.0, 26.3, 25.5, 25.4, 18.8,
16.6, 10.8 HRFABMS calcd for C40H55O7 [M - OCH3]+ 647.3948,
found 647.3906.
1
1455, 1315, 1095, 1030 cm-1; H NMR (CDCl3, 300 MHz) δ 7.25-
7.37 (m, 10H), 5.44 (dd, J ) 1.8, 1.8 Hz, 1H), 5.06 (s, 1H), 4.84 (d,
J ) 12.3 Hz, 1H), 4.73 (d, J ) 12.3 Hz, 1H), 4.72 (d, J ) 11 Hz, 1H),
4.58 (d, J ) 11 Hz, 1H), 4.26 (d, J ) 7.8 Hz, 1H), 3.88-3.94 (m,
2H), 3.79 (d, JP-H ) 11.4 Hz, 6H), 3.61-3.69 (m, 3H), 3.40 (dd, J )
9.8, 9.8 Hz, 1H), 3.23 (dd, J ) 2.1, 10.2 Hz, 1H), 3.20 (s, 3H), 3.12
(d, JP-H ) 22.8 Hz, 2H), 2.66 (dd, J ) 7.8, 7.8 Hz, 2H), 1.97-2.13
(m, 2H), 1.77-1.90 (m, 7H), 1.47-1.68 (m, 8H), 1.35-1.45 (m, 3H),
0.95 (d, J ) 6.3 Hz, 3H), 0.88 (d, J ) 6.9 Hz, 3H); 13C NMR (CDCl3,
75 MHz) δ 201.0, 143.5, 138.7, 138.6, 128.3, 128.2, 127.6, 127.5, 127.4,
112.7, 98.7, 95.6, 85.1, 77.4, 77.2, 75.1, 74.5, 73.6, 73.0, 70.4, 60.4,
53.0, 47.5, 41.5 (d, JC-P ) 127 Hz), 38.6, 35.9, 34.7, 32.2, 31.2, 30.4,
29.0, 27.4, 26.3, 25.5, 25.4, 18.8, 16.6, 10.8; HRFABMS calcd for
C43H60O10P [M - OCH3]+ 767.3924, found 767.3958.
Aldehyde 55. To a stirred room-temperature solution of 54 (24 mg,
35 µmol) in CH2Cl2 (1 mL) was added NaHCO3 (120 mg, 1.40 mmol)
followed by the Dess-Martin periodinane reagent57 (60 mg, 0.14
mmol). This mixture was stirred for 45 min, at which time TLC showed
no remaining 54. Diethyl ether (15 mL), saturated aqueous NaHCO3
(1 mL) and 10% aqueous Na2S2O3 (1 mL) were added. This mixture
was stirred vigorously until the organic layer became clear (∼20 min).
The separated organic phase was washed with H2O (2 × 3 mL) and
saturated aqueous NaCl solution (2 × 3 mL). The aqueous phases
were extracted with diethyl ether, and the combined organic phases
were dried over Na2SO4, filtered, and concentrated. The residue was
filtered through silica gel with hexanes-ethyl acetate (2:1, v/v) and
concentrated to yield aldehyde 55 (22 mg, 33 µmol, 94%) as an oil,
which was used without further purification: Rf 0.31 (hexanes-ethyl
acetate, 2:1, v/v); [R]26D ) +15 (c 1.1, CHCl3); IR (neat) 2940, 2870,
Acknowledgment. Financial support from the NIH
(CA62195) and a University of Minnesota McKnight Land-
Grant Professorship (C.J.F.) is gratefully acknowledged. We also
thank Mr. D. Demeke for technical assistance.
Supporting Information Available: Experimental proce-
dures and characterization data for compounds 4, 5, 9, 10, 13-
18, 20, 21, 23, 26, 35, 41, 43, 48, and 49 via NaBH4 reduction
of 50, and 50-53; photocopies of 1H NMR spectra for
1
2720, 1725, 1495, 1455, 1095 cm-1; H NMR (CDCl3, 300 MHz) δ
compounds 3, 4, 33-35, 40, 41, 48, 49, and 53-55 and of 13
C
9.81 (s, 1H), 7.26-7.36 (m, 10H), 5.44 (s, 1H), 5.06 (s, 1H), 4.82 (d,
J ) 12.3, 1H), 4.72 (d, J ) 12.3 Hz, 1H), 4.71 (d, J ) 11 Hz, 1H),
4.58 (d, J ) 11 Hz, 1H), 4.26 (d, J ) 7.5 Hz, 1H), 3.91 (m, 2H),
3.63-3.69 (m, 3H), 3.40 (dd, J ) 9.6, 9.6 Hz, 1H), 3.26 (m, 1H), 3.22
(s, 3H), 2.50 (dd, J ) 7.5, 7.5 Hz, 2H), 1.19-2.10 (m, 20H), 0.96 (d,
NMR spectra for compounds 3, 53-55 (31 pages). See any
current masthead page for ordering and Web access instructions.
JA973287H