Enantioselective Formal Total Synthesis of (-)-Dysidiolide

Article abstract of DOI:10.1021/ol016073k

(matrix presentd) An enantioselective formal total synthesis of the sesterterpene (-)-dysidiolide 1 beginning with an intermolecular Diels-Alder reaction of the allene ester 3 and the silyloxydiene 10 is reported.

Full text of DOI:10.1021/ol016073k

ORGANIC
LETTERS
2001
Vol. 3, No. 13
2113-2115
Enantioselective Formal Total Synthesis
of (−)-Dysidiolide
Michael E. Jung* and Nobuko Nishimura
Department of Chemistry and Biochemistry, UniVersity of California,
Los Angeles, California 90095-1569
jung@chem.ucla.edu
Received May 5, 2001
ABSTRACT
An enantioselective formal total synthesis of the sesterterpene (−)-dysidiolide 1 beginning with an intermolecular Diels−Alder reaction of the
allene ester 3 and the silyloxydiene 10 is reported.
Dysidiolide 1 was isolated from the Caribbean sponge
Dysidea etheria de Laubenfels and claimed to be an inhibitor
of protein phosphatase cdc25A with an IC50 of 9.4 µM. It
also inhibited the growth of the A-549 human lung carcinoma
and P388 leukemia cell lines, with IC50 values of 4.7 and
1.5 µM, respectively.1 Because of its interesting sesterter-
pinoid structure with six stereocenters, it has attracted a large
amount of synthetic interest which has culminated in several
total or formal total syntheses.2 A few analogues have also
been prepared and tested for their inhibition of both cdc25A
and tumor growth with contradictory results.3 One group3a
cited no activity for dysidiolide and its enantiomeric or
racemic forms (IC50 > 2000 µM) while the second3b reported
good inhibition for the natural product and its close analogues
(IC50 13-35 µM). We now report the enantioselective formal
total synthesis of (-)-dysidiolide 1, which begins with the
intermolecular cycloaddition of an allene and a 2-silyloxy-
diene coupled with a novel stereoselective rearrangement of
the [2 + 2] adduct to afford the normally less favorable exo
Diels-Alder adduct, a process that we had described earlier.4
Recently we reported the cycloaddition of several 2-silyl-
oxydienes, e.g., 2, with the allenecarboxylate 3 which gave
a mixture of exo and endo Diels-Alder adducts and the [2
+ 2] cycloadduct (Scheme 1).4 This latter compound could
be rearranged thermally by a highly diastereoselective process
(1) Gunasekera, S. P.; McCarthy, P. J.; Kelly-Borges, M.; Lobkovsky,
E.; Clardy, J. J. Am. Chem. Soc. 1996, 118, 8759.
Scheme 1
(2) (a) Corey, E. J.; Roberts, B. E. J. Am. Chem. Soc. 1997, 119, 12425.
(b) Boukouvalas, J.; Cheng, Y. X.; Robichaud, J. J. Org. Chem. 1998, 63,
228. (c) Magnuson, S. R.; Sepp-Lorenzino, L.; Rosen, N.; Danishefsky, S.
J. J. Am. Chem. Soc. 1998, 120, 1615. (d) Brohm, D.; Waldmann, H.
Tetrahedron Lett. 1998, 39, 3998. (e) Piers, E.; Caille´, S.; Chen, G. Org.
Lett. 2000, 2, 2483. (f) Demeke, D.; Forsyth, C. J. Org. Lett. 2000, 2, 3177.
(g) Miyaoka, H.; Kajiwara, Y.; Yamada, Y. Tetrahedron Lett. 2000, 41,
911. (h) Takahashi, M.; Dodo, K.; Hashimoto, Y.; Shirai, R. Tetrahedron
Lett. 2000, 41, 2111. (i) Paczkowski, R.; Maichle-Mossmer, C.; Maier, M.
E. Org. Lett. 2000, 2, 3967. (j) Miyaoka, H.; Kajiwara, Y.; Hara, Y.;
Yamada, Y. J. Org. Chem. 2001, 66, 1429.
(3) (a) Blanchard, J. L.; Epstein, D. M.; Boisclair, M. D.; Rudolph, J.;
Pal, K. Bioorg. Med. Chem. Lett. 1999, 9, 2537. (b) Takahashi, M.; Dodo,
K.; Sugimoto, Y.; Aoyagi, Y.; Yamada, Y.; Hashimoto, Y.; Shirai, R.
Bioorg. Med. Chem. Lett. 2000, 10, 2571.
10.1021/ol016073k CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/31/2001

to give only the desired exo adduct so that one could isolate
mainly the exo adduct, e.g., 4, by heating for an extended
period (e.g., 4 days). A four-step sequence (diastereoselective
hydrogenation, desilylation, reduction, and highly regio-
selective dehydration) afforded the bicyclic ring system
analogue of dysidiolide 5. We now report the application of
this route to the synthesis of the natural product itself.
which the dienophile had approached the diene from both
directions, namely the desired face syn to the methyl group
as well as the face anti to the methyl group. The mixture
was separated by careful HPLC on a reverse phase column
using acetonitrile as eluent to produce six distinct compounds
(Scheme 3). These compounds were identified by extensive
Thus, racemic 2-methylcyclohexanone 6 was converted
into the optically active keto ester 7 in 71% yield by the
method of d’Angelo,5 which was first used in the dysidiolide
area by Boukouvalas2b (Scheme 2). The original ee of 7 of
Scheme 3
Scheme 2
1H NMR analysis including the use of HMBC and HMQC
spectra and NOE experiments (shown with arrows in Scheme
3). As shown in Scheme 3, the desired exo isomer A
(compound 11) was the major isomer formed and could be
isolated from HPLC in 30% yield from 10 and 3. Interest-
ingly, the second major isomer isolated is the opposite exo
adduct B with the two endo adducts C and D being formed
in 14% and 6% yield, respectively. It should be noted that it
may be possible to convert the less favorable endo isomer
D into the 4,6-bis-epimer of dysidiolide which has been
shown to be more biologically active than dysidiolide 1
itself.3b Finally small amounts of the [2 + 2] adduct F and
the conjugated diene E were also isolated.
Conversion of 11 into the desired natural product dysid-
iolide followed our earlier route (Scheme 4). Thus, reduction
of the exocyclic methylene was carried out as previously
described4 but with the addition of a pH 7.0 buffer which
was crucial to obtaining consistently high yields and good
diastereoselectivity. Under these conditions one isolates a
4:1 mixture of the compound with the desired R-methyl
86% was raised to >99% by recrystallization of the semi-
carbazone and hydrolysis. Reduction of both functional
groups followed by selective protection of the primary
alcohol and Swern oxidation afforded the keto benzyl ether
8 in 75% yield for the three steps (again by analogy to the
work of Boukouvalas). Formation of the enol triflate (LDA,
PhNTf2, 84%) and Sonogashira coupling with (trimethyl-
silyl)acetylene furnished the enyne 9 in 96% yield. The
alkyne produced by simple desilylation was hydrated under
standard conditions to give in 75% yield the methyl ketone
which was converted into the tert-butyldimethylsilyl (TBS)
enol ether 10 with TBSOTf and triethylamine in 99% yield.
Heating of 10 with the substituted allenecarboxylate 3 at 140
°C for 7 d gave, as expected, a mixture of several products,
the major ones being the two exo Diels-Alder adducts in
(4) Jung, M. E.; Nishimura, N. J. Am. Chem. Soc. 1999, 121, 3529.
(5) d’Angelo, J.; Desma¨ele, D.; Dumas, F.; Guingant, A. Tetrahedron:
Asymmetry 1992, 3, 459.
2114
Org. Lett., Vol. 3, No. 13, 2001

and the ester gave the diol 13 ([R]20 ) +0.7°). In this
D
process, the normally favored axial attack on the cyclohex-
anone is disfavored due to 1,3-diaxial interaction of the
methyl group with the incoming hydride and thus the hydride
attacked the ketone from the normally disfavored equatorial
direction to give the axial alcohol, as we had shown
previously. Selective silylation of the primary neopentyl-
like alcohol in the presence of the hindered axial secondary
alcohol was accomplished using the method of Forsyth2f to
Scheme 4
give the monoalcohol ([R]20 ) -48.7°), which was elimi-
D
nated using thionyl chloride in the presence of pyridine to
give the desired trisubstituted alkene as expected on the basis
of our earlier work.4 This furnished the diene silyl ether 14
which was identical by 1H and 13C NMR6 with an authentic
sample kindly provided by Demeke and Forsyth.2f De-
silylation of the silyl ether 14 with TBAF and oxidation of
the alcohol afforded the aldehyde 15 which has been
converted into dysidiolide 1 by the four-step sequence of
homologation of the aldehyde (reaction with the methoxy-
methylene Wittig reagent and acidic hydrolysis followed by
the well-known process (used often in the synthesis of
dysidiolide) of addition of 3-lithiofuran and photooxidation).
Thus, the preparation of 14 constitutes an enantioselective
formal total synthesis of (-)-dysidiolide 1 from 2-methyl-
cyclohexanone 6 via the key Diels-Alder addition of the
silyloxydiene 10 and the allenecarboxylate 3. Further work
on the preparation of analogues of 1, e.g., the 4,6-bis-epimer,
is currently underway.
Acknowledgment. We thank the National Institutes of
Health for financial support and Drs. Craig Forsyth and
Damtew Demeke for a sample of an intermediate in their
synthesis2f which was identical to the keto ester (before the
reduction to give 13) of our synthesis, except for optical
rotation.
group to the compound with the â-methyl group in 93%
yield. This reduction step also removes the benzyl ether to
give the primary alcohol which was converted into the
tosylate 12 in 86% yield. Coupling of this tosylate with
isopropenylmagnesium bromide in the presence of copper
iodide gave the desired terpene side chain ([R]20D ) -51.2°).
Desilylation afforded the ketone with the trans ring junction
Supporting Information Available: Spectral data and
experimental procedures for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL016073K
(trans-decalin) ([R]20 ) -34.5°) as we had shown in our
(6) The 13C NMR spectra of both 14 and the corresponding alcohol were
D
earlier model study. Hydride reduction of both the ketone
essentially identical to those reported by Forsyth and Demeke.2f
Org. Lett., Vol. 3, No. 13, 2001
2115