Formal total synthesis of okadaic acid via regiocontrolled gold(I)-catalyzed spiroketalizations

Article abstract of DOI:10.1021/ol101833h

Both C19 and C34 spiroketal domains of okadaic acid were assembled using gold(I) chloride catalyzed spiroketalizations, and the two resulting fragments were coupled to give the C15 - C38 fragment of okadaic acid, a known intermediate for the total synthesis of this important natural product.

Full text of DOI:10.1021/ol101833h

ORGANIC
LETTERS
2010
Vol. 12, No. 20
4528-4531
Formal Total Synthesis of Okadaic Acid
via Regiocontrolled Gold(I)-Catalyzed
Spiroketalizations
Chao Fang, Yucheng Pang, and Craig J. Forsyth*
Department of Chemistry, The Ohio State UniVersity, Columbus, Ohio 43210-1185
forsyth@chemistry.ohio-state.edu
Received August 5, 2010
ABSTRACT
Both C19 and C34 spiroketal domains of okadaic acid were assembled using gold(I) chloride catalyzed spiroketalizations, and the two resulting
fragments were coupled to give the C15-C38 fragment of okadaic acid, a known intermediate for the total synthesis of this important natural product.
Okadaic acid (OA, 1, Figure 1) is a polyether natural product
originally isolated from the marine sponges Halichondria
okadai and H. melanodocia.1 The combination of OA’s
attractive structural features and broad range of biological
activities2-5 has stimulated considerable effort within the
synthetic community, culminating in three reported total
syntheses of OA6-8 and one reported total synthesis of the
natural product 7-deoxy-okadaic acid.9
Figure 1. Structure of Okadaic acid.
A common retrosynthetic strategy of the published total
syntheses is disconnection of the natural product into three
components of comparable complexity. We have continued
to refine the fragment syntheses, as well as develop more
reliable coupling methods10 to enhance access to the natural
products and analogs. Herein we report some of our recent
(1) (a) Tachibana, K.; Scheuer, P. J.; Tsukitani, Y.; Kikuchi, H.; Van
Engen, D.; Clardy, J.; Gopichand, Y.; Schimtz, F. J. J. Am. Chem. Soc.
1981, 103, 2469. (b) Dounay, A. B.; Forsyth, C. J. Curr. Med. Chem. 2002,
9, 1851.
(2) Scheuer, P. J. Nat. Prod. 1995, 58, 335
.
(3) Yasumoto, T.; Murata, M.; Oshima, Y.; Sano, M.; Matsumoto, G. K.;
Clardy, J. Tetrahedron 1985, 41, 1019
.
(4) Suganuma, M.; Fujiki, H.; Suguri, H.; Yoshizawa, S.; Hirota, M.;
Nakayasu, M.; Ojika, M.; Wakamatsu, K.; Sugimura, T. Proc. Natl. Acad.
(8) Ley, S. V.; Humphries, A. C.; Eick, H.; Downham, R.; Ross, A. R.;
Boyce, R. J.; Pavey, J. B. J.; Pietruszka, J. J. Chem. Soc., Perkin Trans. 1
1998, 3907.
Sci. U.S.A. 1988, 85, 1768
.
(5) Bialojan, C.; Takai, A. Biochem. J. 1988, 256, 283
.
(6) (a) Isobe, M.; Ichikawa, Y.; Goto, T. Tetrahedron Lett. 1986, 27,
963. (b) Isobe, M.; Ichikawa, Y.; Bai, D.-L.; Masaki, H.; Goto, T.
Tetrahedron 1987, 43, 4767, and references therein.
(9) Dounay, A. B.; Urbanek, R. A.; Sabes, S. F.; Forsyth, C. J. Angew.
Chem., Int. Ed. 1999, 38, 2258.
(10) (a) Dounay, A. B.; Forsyth, C. J. Org. Lett. 1999, 1, 451. (b)
Dounay, A. B.; Urbanek, R. A.; Frydrychowski, V. A.; Forsyth, C. J. J.
Org. Chem. 2001, 66, 925. (c) Trygstad, T. M.; Pang, Y.; Forsyth, C. J. J.
Org. Chem. 2009, 74, 910. (d) Barfknecht, S. A. M.S. Thesis, University
of Minnesota, 2007.
(7) (a) Forsyth, C. J.; Sabes, S. F.; Urbanek, R. A. J. Am. Chem. Soc.
1997, 119, 8381. (b) Urbanek, R. A.; Sabes, S. F.; Forsyth, C. J. J. Am.
Chem. Soc. 1998, 120, 2523. (c) Sabes, S. F.; Urbanek, R. A.; Forsyth,
C. J. J. Am. Chem. Soc. 1998, 120, 2534.
10.1021/ol101833h  2010 American Chemical Society
Published on Web 09/17/2010

efforts in the synthesis of the C15-C38 fragment of OA,
which highlights the dual use of Au(I)-catalyzed spiroket-
alizations.
Scheme 1. Retrosynthesis of the C15-C38 Domain of OA
Transition metal catalyzed spiroketalizations11-13 have
recently emerged as a versatile method for the syntheses of
spiroketal moieties of natural products, as exemplified by
(+)-broussonetine G,14 azaspiracid,15 (+)-spirolaxine methyl
ether,16 and ushikulide A17 (Figure 2). Compared with the
Scheme 2. Synthesis of C15-C27 Fragment of OA
Figure 2
synthesis.
. Metal catalyzed spiroketalization used in natural product
conventional approach toward spiroketals using acid-
catalyzed dehydrative cyclization of dihydroxyl ketones, the
use of an alkyne as a dehydrated surrogate for a ketone
avoids some potential sensitivity issues associated with highly
functionalized ketones.15,16 With this in mind, we embarked
upon the syntheses of the C19 and C34 spiroketals of OA
using gold catalysis.
Retrosynthetically, the C15-C38 domain of OA is sepa-
rated into two fragments: the C27 aldehyde 2 and the C28
iodide 3, which can be coupled via a derived C28 organo-
metallic species (Scheme 1).7 The C19 spiroketal in 2 is
derived from alkyne 4, while the synthesis of the C34
spiroketal 3 starts with triol 5. With the latter 1,3-anti triol,
the cyclization regioselectivity can be controlled to give the
desired 1,7-dioxaspiro[5.5]-undecane system, as recently
reported by Aponick et al.13
mercially available methyl-R-D-glucopyranoside 6 in 10
steps.7b Treatment of aldehyde 7 with the Bestmann-Ohira
reagent (8)18 and K2CO3 in methanol afforded terminal
alkyne 9 in 82% yield. BF3-mediated nucleophilic opening19
of oxirane 10 with the lithium acetylide generated from 9
and nBuLi produced alcohol 11 smoothly. Desilylation of
11 with TBAF gave dihydroxyl alkyne 4 as the spiroketal-
ization precursor.
On the basis of previous experience,10d,15 AuCl was
selected as the catalyst for the spiroketalization reaction.
Gratifyingly, the desired spiroketal was formed after mixing
4 with a catalytic amount of AuCl in CH2Cl2. Any acid
cocatalyst was initially avoided here to prevent the removal
of the anisylidene protecting group. However, the Lewis
acidity of AuCl itself triggered partial removal of the
protecting group, but the in situ liberated hydroxyl groups
did not seem to affect the spiroketalization efficiency.
The synthesis of the C19 spiroketal started with known
aldehyde 7 (Scheme 2), which was prepared from com-
(11) Utimoto, L. Pure Appl. Chem. 1983, 55, 1845
(12) Liu, B.; De Brabander, J. L. Org. Lett. 2006, 8, 4907
(13) Aponick, A.; Li, C.-Y.; Palmes, J. A. Org. Lett. 2009, 11, 121
.
.
.
(14) Trost, B. M.; Horne, D. B.; Woltering, M. J. Angew. Chem., Int.
Ed. 2003, 42, 5987.
(15) Li, Y.; Zhou, F.; Forsyth, C. J. Angew. Chem., Int. Ed. 2007, 46,
279.
(16) Trost, B. M.; Weiss, A. H. Angew. Chem., Int. Ed. 2007, 46, 7664.
(17) (a) Trost, B. M.; O’Boyle, B. M. J. Am. Chem. Soc. 2008, 130,
16190. (b) Trost, B. M.; O’Boyle, B. M.; Hund, D. J. Am. Chem. Soc.
2009, 131, 15061.
(18) (a) Ohira, S. Synth. Commun. 1989, 19, 561. (b) Roth, G. J.; Liepold,
B.; Muller, S. G.; Bestmann, H. J. Synlett 1996, 521. (c) Roth, G. J.; Bernd,
L.; Muller, S. G.; Bestmann, H. J. Synthesis 2004, 59.
(19) Yamaguchi, M.; Hirao, I. Tetrahedron Lett. 1983, 24, 391.
Org. Lett., Vol. 12, No. 20, 2010
4529

Subsequent addition of TsOH in methanol to the reaction
mixture helped to completely remove the anisylidene group
to give diol 12 in 81% yield.
1,6-dioxaspiro[4.6]undecane ring system (33S)-21b was
formed as the major product, followed by the 1,7-
dioxaspiro[5.5]undecane ring system (34S)-21a, and (33R)-
21c in a 7.6:1.6:1.0 ratio, respectively (82% combined yield,
Scheme 3). In contrast, the 1,3-anti triol 5 gave the
unsaturated OA spiroketal 21a as the exclusive product, albeit
in lower yield. The introduction of the alkene into 21a would
be inconsequential en route to 3. The divergent regioselec-
tivity observed upon cyclization of 20 and 5 is consistent
with that reported by Aponick.13
The synthesis of the C34 spiroketal began with known
diol 137c (Scheme 3). Selective protection of the primary
Scheme 3. Synthesis of C28-C38 Fragment of OA
These results suggest a kinetic preference for the C30
hydroxyl of 1,3-syn triol 20 versus the C30 hydroxyl of 1,3-
anti triol 5 to participate in initial oxy-auration of the alkyne
via 5-exo addition at C33 (Scheme 4). The 5-exo cyclization
Scheme 4. Proposed Spiroketalization Mechanisms
hydroxyl group using BnBr and KOH in benzene at reflux20
gave benzyl ether 14 in 55% yield. The remaining hydroxyl
group was protected as triethylsilyl ether 15. Ozonolysis of
15 yielded aldehyde 16 in 94% yield. Addition of the lithium
acetylide derived from alkyne 17 to aldehyde 16 afforded a
mixture of epimers at C32 (anti:syn ) 1.0: 1.5). These were
separated then treated with TsOH in methanol to generate
1,3-syn triol 20 and 1,3-anti triol 5 as spiroketalization
precursors.
Aponick reported that the relative 1,3-stereochemistry of
the propargylic and nucleophilc hydroxyls within triols
similar to 20 and 5 influenced the gold catalyzed spiroket-
alization regioselectivity.13 Thus, it was of interest to probe
the fate of 1,3-syn triol 20 versus 1,3-anti triol 5 upon gold(I)
catalysis. When 1,3-syn triol 20 was treated with AuCl, the
of 20 would afford an enol ether gold intermediate 23,
protiodeauration of which followed by capture of the primary
hydroxyl would lead to spiroketals 21b/c. However, an
analogous 5-exo cyclization of 5 would proceed via a more
sterically hindered all syn C30-C32 trisubstituted five
membered ring transition state (e.g., 26, Scheme 4). Alter-
natively, the primary hydroxyl of 5 likely initiates its
spiroketalization via a less encumbered 6-exo oxy-auration.
The original C32 propargylic hydroxyl group of 20 is
retained in 21b/c, whereas it is eliminated in 21a. This may
reflect a concerted loss of gold hydroxide from an R-hydroxy
vinyl gold species (e.g., 28, Scheme 4), whereas an allylic
hydroxy vinyl gold intermediate (e.g., 23) undergoes simple
protiodeauration. Isomerization of the resulting exocyclic
allenyl ether intermediate 29 into a vinyl substituted oxac-
(20) Roulland, E. Angew. Chem., Int. Ed. 2008, 47, 3762.
4530
Org. Lett., Vol. 12, No. 20, 2010

arbenium ion (30) followed by addition of the C30 hydroxyl
would afford 21a.
favoring the natural product’s configuration (27S/27R ) 2:1).
The undesired diastereomer 34b was converted into 34a
using the traditional oxidation-reduction sequence.6,7 Thus,
34a was obtained in 48% combined yield, including stere-
ochemical correction, which is in line with the 52 and 47%
yields for coupling and (27S) alcohol installation in the other
OA total syntheses.6,8 Protection of the free hydroxyl group
at C27 as a benzyl ether followed by removal of the PMB
protecting group afforded alcohol 35, a known intermediate
for the total syntheses of OA.6,8
Once the dependence of spiroketalization regioselectivity
on polyol stereochemistry was confirmed, an initial measure
to enhance mass throughput was to convert syn-18b into anti-
18a (Scheme 3). This was accomplished in a two-step
sequence of oxidation/asymmetric reduction.21 Thus both
18a/b diastereomers converged to spiroketal 21a. Hydroge-
nation of 21a followed by iodination gave 3.
The next stage was fragment coupling (Scheme 5). Prior
to this, the C25 exocyclic alkene of OA was installed. This
In our original approach to OA, the use of an unmodified
C28 organolithium species in THF gave the C27-C28
coupled product in poor yield (30-35%), while the major
byproduct was the adduct derived from the addition of tBuLi
to the aldehyde.7c As reported in 1997, transmetalation of
the organolithium species to an organocerium derivative
improved the yields, although the diastereomeric ratio was
unfavorable (27R/27S ) 2.5:1.0).7 In the present case, it was
beneficial to use diethyl ether as the bulk solvent and to warm
the mixture to room temperature after the lithium-iodine
exchange of 3 to decompose any residual tBuLi.22 After the
alkyllithium solution was recooled to -78 °C, the addition
of aldehyde 2 gave coupled products in yields comparable
to those obtained previously with the difficult to handle
organocerium reagent (53 vs 50%). However, the diastereo-
selectivity observed here in forming the C27 carbinol using
the organolithium in diethyl ether was favorably reversed
(2.5:1 vs 1:2). This may reflect better chelation controlled
addition of the C28 organolithium to the C27 aldehyde in
diethyl ether than with the organocerium reagent in THF.7
In conclusion, the C19 and C34 spiroketals of OA were
synthesized using regiocontrolled AuCl catalyzed spiroket-
alizations of dihydroxy alkynes. The fragments were coupled
via C27-C28 bond formation using an improved procedure
to afford the C15-C38 domain of OA. Compound 35 can
serve as an intermediate in Isobe’s and Ley’s total syntheses
of OA,6,8 and therefore its preparation here constitutes a
formal total synthesis.
Scheme 5. Coupling of C15-C27 and C28-C38 Fragments
Acknowledgment. We thank S. A. Barfknecht (U. of
Minnesota) and Dr. C. Wang (Enanta) for preliminary
experimental contributions, the Ohio BioProducts Innovation
Center for support of the OSU Chemistry Department Mass
Spectrometry Facility, and the NIEHS (grant R01ES010615)
and OSU for financial support.
was initiated by selective silyl protection of the primary
hydroxyl of diol 12. The resulting secondary alcohol 31 was
oxidized to the corresponding ketone then converted into
alkene 32 under Wittig conditions. TBAF induced removal
of the C27 silyl group gave primary alcohol 33. This was
oxidized into the sensitive ꢀ,γ-unsaturated aldehyde 2 using
the Dess-Martin periodinane reagent immediately prior to
use. Treatment of iodide 3 with tBuLi in diethyl ether
followed by the addition of aldehyde 2 afforded a mixture
of C27 carbinols 34a and 34b in 50% combined yield
Supporting Information Available: Experimental pro-
1
cedure, characterization data, H NMR and 13NMR spectra
for previously unreported compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL101833H
(21) Parker, K. A.; Ledeboer, M. W. J. Org. Chem. 1996, 61, 3214.
(22) Bailey, W. F.; Punzalan, E. R. J. Org. Chem. 1990, 55, 5404.
Org. Lett., Vol. 12, No. 20, 2010
4531