Total synthesis of (-)-exiguolide

Article abstract of DOI:10.1021/ol902829e

The first total synthesis of the naturally occurring enantiomer of exiguolide ((-)-1) has been completed. This very convergent synthesis features the following as main steps: (I) a Trost's ruthenium-catalyzed ene-yne cross-coupling reaction (this complex transformation allows the challenging control of the C5-C28 double bond geometry along with the stereoselective construction of the tetrahydropyran ring A) and (II) a very efficient one-pot, two-step stereoselective conjugated allyllc alcohol substitution that allowed the control of the C15 stereogenic center.

Full text of DOI:10.1021/ol902829e

ORGANIC
LETTERS
2010
Vol. 12, No. 4
744-747
Total Synthesis of (-)-Exiguolide
Cyril Cook, Xavier Guinchard, Fre´de´ric Liron, and Emmanuel Roulland*
Centre de Recherche CNRS de Gif-sur-YVette, Institut de Chimie des Substances
Naturelles, AVenue de la Terrasse, 91198 Gif-sur-YVette, France
emmanuel.roulland@icsn.cnrs-gif.fr
Received December 8, 2009
ABSTRACT
The first total synthesis of the naturally occurring enantiomer of exiguolide ((-)-1) has been completed. This very convergent synthesis
features the following as main steps: (i) a Trost’s ruthenium-catalyzed ene-yne cross-coupling reaction (this complex transformation allows
the challenging control of the C5-C28 double bond geometry along with the stereoselective construction of the tetrahydropyran ring A) and
(ii) a very efficient one-pot, two-step stereoselective conjugated allylic alcohol substitution that allowed the control of the C15 stereogenic
center.
In 2006, the Ikegami group reported the isolation of (-)-
exiguolide ((-)-1) from the marine sponge Geodia exigua.1
From a structural point of view, 1 is a macrolide that displays
a number of salient motifs rendering this challenging target
quite attractive for synthetic chemists. (-)-Exiguolide ((-)-
1) is a 16-membered macrolactone bearing five C-C double
bonds, and seven stereogenic centers. The macrocycle is
fused with two tetrahydropyran rings A and B, ring A bearing
a methoxycarbonylmethylidene function at C5 with a Z
configuration. This target does not only represent a synthetic
challenge, since biological tests revealed interesting proper-
ties.1 Thus, (-)-1 inhibits the fertilization of sea urchin
gametes, which indicates that this compound could inhibit
the fusion of viruses with cell membranes.2
product.3 They designed an approach featuring the macro-
cyclization by ring closing metathesis creating then the
C16-C17 double bond. The installation of the methoxycar-
bonylmethylidene function at C5 was made by a Horner-
Wadsworth-Emmons reaction but required a stoichiometric
amount of an asymmetric phosphonoacetate to finally only
deliver a Z/E mixture (5.8/1).
This original structure attracted our attention and teased
our imagination leading to the retrosynthetic plan presented
in the abstract. We identified the control of the configuration
of the methoxycarbonylmethylidene function at C5 as one
of the main challenges of this synthesis. Focusing on this, it
appeared from the literature that one of the most expeditious
methodologies allowing full control of this double bond
would be the ene-yne cross-coupling reaction described by
Trost et al.,4 which is catalyzed by the cationic RuII complex
[CpRu(MeCN)3]PF6 (4).5 Furthermore, in the course of this
reaction, the tetrahydropyran ring A would ideally be closed
In 2008, Lee et al. reported the synthesis of ent-exiguolide
((+)-1), the enantiomer of the naturally occurring compound,
thus establishing the right absolute configuration for this
(1) Ohta, S.; Uy, M. M.; Yanai, M.; Ohta, E.; Hirata, T.; Ikegami, S.
Tetrahedron Lett. 2006, 47, 1957–1960.
(3) Kwon, M. S.; Woo, S. K.; Na, S. W.; Lee, E. Angew. Chem., Int.
Ed. 2008, 47, 1733–1735.
(2) Ikegami, S.; Kobayashi, H.; Myotoishi, Y.; Ohto, S.; Kato, K. H.
J. Biol. Chem. 1994, 269, 23262–23267.
(4) (a) Gill, T. P.; Mann, K. R. Organometallics 1982, 1, 485–488. (b)
Trost, B. M.; Older, C. M. Organometallics 2002, 21, 2544–2546.
10.1021/ol902829e  2010 American Chemical Society
Published on Web 01/15/2010

Scheme 1. Control of C15 Stereogenic Center by SN2′ and Synthesis of Fragment 3
stereoselectively in the course of an intramolecular oxa-
Michael addition. One must notice that a similar framework
occurs in some other important natural product families of
high biological interest such as in the bryostatins.6 In 2008,
as we were developing this study, Trost described the total
synthesis of bryostatin 16 for which he made use of catalyst
4.7 For the construction of ring B we would use the classical
reduction of the corresponding hemiketal. Securing the C15
stereochemistry was identified as another challenge, and here
we thought it possible to perform a stereoselective SN2′
reaction on an asymmetric Z-allylic alcohol activated as a
phosphate 5. The macrolactonic ring would be closed by a
classical Yamaguchi macrolactonization and we imagined
bringing the C21-C27 side chain in the late steps of the
synthesis by using a cross-metathesis reaction.
We commenced with the synthesis of the C6-C21
fragment 3 (Scheme 1). First, we accessed the enantio-
enriched epoxide (+)-7 using the Jacobsen hydrolytic kinetic
resolution (HKR)8 on the corresponding racemic epoxide
rac-7 obtained by m-CPBA epoxidation of alkene 6. Alkyne
8 was furnished by reaction of epoxide (+)-7 with lithium
trimethylsilylacetylide, followed by removal of the TMS
alkyne protective group and protection of the alcohol
function. The lithium acetylide of 8 was condensed with the
known Weinreb amide 9.9 This efficient cross-coupling step
afforded propargylic ketone 10 in 95% yield. The ketone
function of 10 was reduced into alcohol 11 with a good
diastereoselectivity by using Luche conditions.10 The semire-
duction of the alkyne function of 11, using the Lindlar
catalyst, led to Z allylic alcohol 12. Our SN2′ strategy for
the control of the allylic C15 stereogenic center is based on
the assumption that the constrained geometry of Z allylic
alcohols blocks the rotation of the allylic C-C bonds, which
is not the case with the corresponding E isomers. As a
consequence, SN2′ on activated Z allylic alcohols would
furnish only one product by attack of the nucleophile anti
to the leaving group allowing then a good transfer of chirality
(E isomers usually give mixtures). A similar strategy has
been reported in the literature that made use of perfluo-
robenzoate esters as leaving groups.11 Unfortunately we met
troubles in synthesizing the required perfluorobenzoate ester
from alcohol 12 as this ester appeared unstable and was
obtained in poor yields. After many experiments, screening
various leaving groups and cuprates, we finally found simple
and efficient one-pot, two-step conditions allowing a direct
transformation of the Z allylic alcohol 12 into the desired
alkene 13 in high yield (96%) and a very good transfer of
chirality (dr around 95:5 by H NMR spectroscopy). In
this protocol, alcohol 12 was first activated as phosphate 5
(KHMDS in Et2O, and addition of ClPO(OEt)2) and subse-
quently substituted by methyl cuprate (Me2CuLi) formed in
Et2O. This protocol has been applied to various kinds of Z
allylic alcohols with success.
Next, the TBDPS protective group was selectively re-
moved under basic conditions13 affording alcohol 14, which
was oxidized into aldehyde 15 under Swern conditions.14
The latter was transformed with use of the Luche15 procedure
into homoallylic alcohol 16 (1/1 mixture of diastereoisomers)
12
1
(5) Trost, B. M.; Frederiksen, M. U.; Rudd, M. T. Angew. Chem., Int.
Ed. 2005, 44, 6630–6666.
(11) Harrington-Frost, N.; Leuser, H.; Calaza, M. I.; Kneisel, F. F.;
Knochel, P. Org. Lett. 2003, 5, 2111–2114.
(6) Cossy, J. C. R. Chimie 2008, 11, 1477–1482.
(12) We have synthesized the unwanted diastereosomer of 13 in order
1
(7) Trost, B. M.; Dong, G. Nature 2008, 456, 485–488.
(8) Tokunaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science
1997, 936–938.
to locate it in H NMR spectra of 13 (see the Supporting Information for
the full reference).
(13) Kabat, M. M.; Lange, M.; Wovkulich, P. M.; Uskokovic´, M. R.
Tetrahedron Lett. 1992, 33, 7701–7704.
(9) Funel, J.-A.; Prunet, J. J. Org. Chem. 2004, 69, 4555–4558.
(10) (a) Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226–2227. (b)
Gemal, A. L.; Luche, J.-L. J. Am. Chem. Soc. 1981, 103, 5454–5459.
(14) Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651–1660.
(15) Pe´trier, C.; Luche, J.-L. J. Org. Chem. 1985, 50, 910–912.
Org. Lett., Vol. 12, No. 4, 2010
745

Scheme 2. Ruthenium-Catalyzed Coupling of Fragments 2 and 3, Macrolactonization, and Final Steps
and readily oxidized by Dess-Martin periodinane16 into ꢀ,γ-
unsaturated ketone 3, the key “ene” coupling partner of the
RuII catalyzed ene-yne coupling (53.4% over 11 steps from
(+)-7).
a 8:1 mixture of diastereomers easily separated by HPLC.
Anyway, this result compares favorably with the 34% yield
(dr not given) obtained by Trost for his bryostatin 16
synthesis, on comparable substrates. Furthermore one should
notice that in our case, the reaction was regioselective as
the C20-C21 terminal double bond was not engaged in the
cross-coupling. The TMS group of 20 being particularly acid
sensitive, we performed its iodolysis leading to 21, prior to
the acid-promoted removal of the two TBS protective groups
and subsequent cyclization into the hemiketal precursor of
cycle B. The crude mixture of hemiketals was subsequently
reduced by Et3SiH in the presence of BF3·OEt2 at -40 °C
affording compound 22 (dr >20:1) now featuring the two
tetrahydropyran rings A and B. After saponification we
obtained acid 23. Then, the methoxycarbonyl function was
cleanly installed at C28 by a Pd0-catalyzed carbonylation21
of iodinated derivative 23 in MeOH furnishing the desired
(Z)-R,ꢀ-unsaturated ester 24 (92:8 Z/E mixture). We pro-
ceeded then to the macrolactonization step using the Yamagu-
chi conditions22 and obtained macrolactone 25 in a good 74%
yield (9.2% over 17 steps). The order of these steps
(saponification-carbonylation-macrolactonization) is slightly
On the other hand, the known “yne” partner 217 was
accessed from trans-crotonoyl chloride 17, which was
transformed into ꢀ,γ-unsaturated ester 18 (Scheme 2).18 An
epoxidation by m-CPBA furnished rac-19, which was
enantio-enriched by using the Jacobsen HKR method fur-
nishing (+)-19.8 Finally, 2 was cleanly obtained after reaction
with an aluminum acetylide nucleophile,19 the more classical
reaction of the corresponding lithium acetylide in the
presence of BF3 led to decomposition.
We have already used RuII catalyst 4 to build the
tetrahydropyran ring of (+)-neopeltolide,20 and discovered
that a catalytic amount of acetic acid accelerated the reaction
also allowing a remarkable enhancement of the diastereo-
selectivity. The same conditions of cross-coupling of alkyne
2 with alkene 3 led to tetrahydropyran 20 in a 47% yield as
(16) (a) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4156–4158.
(b) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277–7287.
(17) Shen, R.; Lin, C. T.; Bowman, E. J.; Bowman, B. J.; Porco, J. A.,
Jr. J. Am. Chem. Soc. 2003, 125, 7889–7901.
(18) (a) Ozeki, T.; Kusaka, M. Bull. Chem. Soc. Jpn. 1966, 39, 1995–
1998. (b) Roulland, E.; Ermolenko, M. S. Org. Lett. 2005, 7, 2225–2228.
(19) Larcheveˆque, M.; Petit, Y. Bull. Chim. Soc. Fr. 1989, 130–139.
(20) Guinchard, X.; Roulland, E. Org. Lett. 2009, 11, 4700–4703.
(21) Schoenberg, A.; Bartoletti, I.; Heck, R. F. J. Org. Chem. 1974, 39,
3318–3326.
(22) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull.
Chem. Soc. Jpn. 1979, 52, 1989–1993.
746
Org. Lett., Vol. 12, No. 4, 2010

unusual but it was the way we found to make the carbonyl-
ation reaction work in a reproducible manner.
For the last step of the synthesis it was first envisioned to
introduce the C21-C27 side chain directly by a cross-
metathesis reaction23 on alkene 25. Unfortunately, none of
the catalysts (Grubbs II and Hoveyda-Grubbs II (Figure 1)),
mixture of diols with 20% yield over two steps and partial
recovery of the starting material 25 (31%). The Takai-
Utimoto28 olefination reaction on aldehyde 26 furnished
vinylic iodide derivative (-)-27 in high yield, the enantiomer
of which was described by Lee.3 In the two final steps of
this synthesis we followed the Lee strategy. Thus we
introduced the C22-C27 side chain by Sonogashira cross-
coupling of 27 with alkyne 28,3,29 which furnished dienyne
29. The final step of semihydrogenation led to the targeted
(-)-exiguolide (1), the characterization data of which are
identical with those reported for the naturally occurring
compound ([R]20D -95.0 (c 0.28, CHCl3), lit.1 [R]20D -92.5
(c 0.069, CHCl3)).
In summary the first total synthesis of the naturally
occurring enantiomer of exiguolide ((-)-1) has been com-
pleted. This very convergent synthesis furnished the target
in 21 steps as the longest linear path (from (+)-7) in a 0.52%
yield. This total synthesis features as main steps a Trost’s
ruthenium-catalyzed ene-yne cross-coupling reaction, al-
lowing for control of the challenging C5-C28 double bond
geometry, and a very efficient one-pot, two-step stereose-
lective conjugated allylic alcohol substitution used to control
the C15 stereogenic center.
Figure 1. Catalysts and alkenes used for the cross-metathesis
reaction at C20.
cross-coupling partners (i, ii,24 and iii25 and allylic alcohol
(Figure 1)), and solvents (CH2Cl2, toluene, rt or reflux) we
tried gave any product, in all cases the starting material 25
was totally recovered and not even dimers of 25 were formed.
We then reconsidered our retrosynthetic analysis, and we
thought it reasonable to expect selectivity from the OsO4
promoted dihydroxylation reaction26 toward the less hindered
and not electron-deficient C20-C21 double bond of 25.
Surprisingly we observed a poor selectivity and identified
various diols and tetraols, some resulting from the unexpected
dihydroxylation of the R,ꢀ-unsaturated ester at C5-C28, the
C16-C17 double bond remaining untouched. A mild reac-
tion with Pb(OAc)427 finally delivered aldehyde 26 from this
Acknowledgment. This work has been financially sup-
ported by the Institut de Chimie des Substances Naturelles
(ICSN) and the Centre National pour la Recherche Scien-
tifique (CNRS).
Note Added after ASAP Publication. Reference 15 was
updated on January 22, 2010.
Supporting Information Available: Experimental pro-
cedures and characterization data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL902829E
(23) Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. J. Am.
Chem. Soc. 2003, 125, 11360–11370.
(27) Bunton, C. A. In Oxidation in Organic Chemistry Part A; Wiberg
K., Ed.; Academic: New York. 1965; pp 398-405.
(28) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108,
7408–7410.
(24) Uno, B. E.; Gillis, E. P.; Burke, M. D. Tetrahedron 2009, 6, 3130–
3138.
(25) Kujawa-Welten, M.; Pietraszuk, C.; Marciniec, B. Organometallics
2002, 21, 840–845.
(29) For the synthesis of the precursor of 28 see: Kigoshi, H.; Kita, M.;
(26) Schroeder, M. Chem. ReV. 1980, 80, 187–213.
Ogawa, S.; Itoh, M.; Uemura, D. Org. Lett. 2003, 5, 957–960.
Org. Lett., Vol. 12, No. 4, 2010
747