Total synthesis of (+)-cyanthiwigin AC

Article abstract of DOI:10.1021/ol062304h

A 13-step synthesis of (+)-cyanthiwigin-AC (2) from (+)-Hajos-Parrish ketone derivative 8b and dimesylate 9c employing deconjugative spirobis-alkylation strategy is described.

Full text of DOI:10.1021/ol062304h

ORGANIC
LETTERS
2006
Vol. 8, No. 24
5585-5588
Total Synthesis of (+)-Cyanthiwigin AC
T. Jagadeeswar Reddy,* Guillaume Bordeau, and Laird Trimble
Department of Medicinal Chemistry, Merck-Frosst Canada & Co.,
16711 Trans Canada Highway, Kirkland, Quebec, Canada H9H 3L1
tj.reddy@enobia.com
Received September 19, 2006
ABSTRACT
A 13-step synthesis of (+)-cyanthiwigin-AC (2) from (+)-Hajos−Parrish ketone derivative 8b and dimesylate 9c employing deconjugative spiro-
bis-alkylation strategy is described.
In 1992,1 Green and co-workers isolated four novel diter-
penes containing a 5,6,7-tricarbocyclic skeleton 1 (cyanthi-
wigins A-D) from the marine sponge Epipolasis reiswigi.
A decade later, Peng and co-workers isolated2a several
cyanthiwigins from the Jamaican sponge Myrmekioderma
styx. Since their first isolation, cyanthiwigins showed a
diverse range of biological activities2 ranging from cytotox-
icity to inhibition of Mycobacterium tuberculosis and nerve-
growth factor stimulation. Recently, Peng et al. have isolated
the completely novel diterpene, cyanthiwigin AC (2), along
with cyanthiwigin AD from the deep-reef collection of the
Jamaican sponge M. styx.2b Minute quantities of material
hampered further biological screening of these natural
products. Cyanthiwigin AC (2) is a member of a unique class
of natural products containing a [6,6]-spiro skeleton. It is
believed to be biogenetically derived from cyanthiwigin U
(3). Despite their biological properties there has been little
synthetic effort on these terpenes.3a Cyanthiwigin AC is a
synthetically complex molecule consisting of five stereogenic
centers, of which four are contiguous. In this array are two
quaternary carbons including a spiro ring fusion. Herein, we
report the first total synthesis of (+)-cyanthiwigin AC (2)
and the second reported synthesis of a cyanthiwigin.
Retrosynthetic analysis of cyanthiwigin AC (2) can be
envisioned from the commercially available (+)-Hajos-
Parrish ketone (Scheme 1).4 The target compound (2) would
be obtained by the regio- and stereoselective addition of
methyllithium or methyl Grignard reagent to cyclohexenone
4, which would arise from intermediate 5 via dehydrogena-
tion of rings A and C (Scheme 1). Compound 5 would be
prepared from precursor 6 via a two-step sequence, 2-pro-
* To whom correspondence should be addressed. Tel: 514-596-2901
ext 216. Fax: 514-596-2749.
(1) Green, D.; Goldberg, I.; Stein, Z.; Ilan, M.; Kashman, Y. Nat. Prod.
Lett. 1992, 1, 193-199.
(2) (a) Peng, J.; Walsh, K.; Weedman Braude, I. A.; Kelly, M.; Hamann,
M. T. Tetrahedron 2002, 58, 7809-7819. (b) Peng, J.; Avery, M. A.;
Hamann, M. T. Org. Lett. 2003, 5, 4575-4578. The optical rotation for 2
was not measured due to minute quantities of the material (confirmed with
Professor Mark Hamann).
(3) Recent syntheses of cyanthiwigin U and other similar families of
natural products: (a) Pfeiffer, M.; Phillips, A. J. J. Am. Chem. Soc. 2005,
127, 5334-5335. (b) Waters, S. P.; Tian, Y.; Li, Y.-M.; Danishefsky, S. J.
J. Am. Chem. Soc. 2005, 127 (39), 13514-13515 and references cited
therein.
(4) Hajos, Z. G.; Parrish, D. R. Organic Syntheses; Wiley: New York,
1990; Collect. Vol. VII, p 363.
10.1021/ol062304h CCC: $33.50
© 2006 American Chemical Society
Published on Web 10/24/2006

the starting material and partially alkylated products. How-
ever, reaction of TBS ether derivative 8b with 95% NaH in
refluxing DME followed by diiodide 9b addition afforded a
mixture of starting material and product (16%). Formation
of spiro ketone 7b was significantly improved when alkylated
with mesylate 9c6c to provide a ∼1:2 mixture of starting
material and product (48%). However, the mixture of starting
material and product could be separated by neither column
chromatography nor HPLC. Therefore, it was essential to
drive the reaction to completion. After a few manipulations,
the best results were obtained by refluxing 8b in THF (0.15
M) with 95% NaH (3.5 equiv) for 3 h, followed by addition
of mesylate (1.5 equiv) at 60 °C for 2 h. Further addition of
0.5 equiv of the mesylate 9c (0.5 equiv) over 2 h provided
a respectable yield of the spiro-product 7b (60%, with an
average yield of 78% for each alkylation). Reaction can be
easily carried out on 5 g scale without deterioration of
reaction yields. Furthermore, excess dimesylate 9c (0.5 equiv)
was easily recovered by trituration of the crude reaction
mixture with hexanes. After successful spiro annulation, we
focused on the olefination reaction of highly hindered spiro
ketone 7b. Wittig olefination of spiro ketone 7b using either
Corey’s reagent (Ph3PdCHLi)10a or the Tebbe10b reaction
provided only low conversion. We attempted the mild and
efficient direct methylenation reported by Yan.11 In our first
attempt, alkene 10 was isolated in 40% yield (80% conver-
sion). This highly functionalized substrate 7b seems to be
sensitive to the reaction conditions. After several attempts,
reaction of spiro-ketone 7b with TiCl4 (4 equiv, 1 M THF),
Mg powder (10 equiv) in DCM-THF (1:1.5, 0.2 M) at 0
°C to rt delivered spiro-alkene 10 (47% yield).
Scheme 1. Retrosynthesis
penyl cuprate addition and hydrogenation of both double
bonds. The tricyclic core of cyanthiwigin AC (6) would be
constructed in enantiopure form from (+)-Hajos-Parrish
ketone via a sequence featuring deconjugative spiro-bis-
alkylation 75,6 followed by isomerization of the double bond
in ring A.
Deprotection of the TBS ether 10 with TBAF followed
by oxidation of the resultant secondary alcohol with Dess-
Martin (D-M) periodinane afforded cyclopentanone deriva-
tive 11 (68% yield from TBS ether). Now the stage was set
for the isomerization of the trisubstituted double bond.
Attempts to isomerize the double bond using either KOH in
methanol12 or concd HCl failed. However, stirring of
cyclopentanone derivative 11 with NaOMe in methanol at
rt cleanly delivered cis-hydrindenone derivative 6 (95%
yield) as the sole product. Tentative assignment of the cis-
ring junction for compound 6 is based on literature prece-
dence,12 which secures the natural product stereochemistry.
After obtaining enone 6 in sufficient quantities, introduction
of the isopropyl group via a 1,4 addition to the unsaturated
ketone was examined. In order to avoid complications,
2-propenyl cuprate was used instead of isopropyl cuprate,
as the former can be easily hydrogenated to the latter. The
alkenyl cuprate was preformed in situ from 2-bromopropene,
n-butyllithium (2-propenyl lithium at -68 °C), and copper(I)
cyanide. Cuprate addition13 to enone 6 at -68 °C for 2 h
Even though (+)-Hajos-Parrish ketone is commercially
available, it can be readily prepared in multigram quantities
with high optical purity.4 Both starting materials 8a (R1,R2
) -OCH2CH2O-)7 and 8b (R1 ) OTBS, R2 ) H)8 were
conveniently prepared from (+)-Hajos-Parrish ketone using
literature procedures. To our surprise, only a single report
of spiro annulation of a ketone similar to 8b was known in
the literature.9 Our initial attempt at reaction of compound
8a and diiodide 9b (Scheme 2) using 60% NaH in DME
provided a complex mixture, without any noticeable amount
of desired spiro ketone 7. We undertook systematic optimi-
zation of several reaction parameters on compound 8a
including variation of the base (KOtBu, KH, 95% NaH),
leaving group of 9a-c, solvent (THF, THF-DME), reaction
time, and temperature. All attempts resulted in a mixture of
(5) Sviridov, S. V.; Vasilevskii, D. A.; Kulinkovich, O. G. Zh. Org. Khim.
1991, 27, 1431-1433.
(6) (a) Gleiter, R.; Ramming, M.; Weigl, H.; Wolfart, V.; Irngartinger,
H.; Oeser, T. Liebigs Ann./Recueil 1997, 1545-1550. (b) Davenport, R.
J.; Regan, A. C. Tetrahedron Lett. 2000, 41, 7619-7622. (c) Dimesylate
is unstable at rt but can be stored in the refrigerator for a few days:
Wasylishen, R. E.; Rice, K. C.; Weiss, U. Can. J. Chem. 1975, 53, 414-
417.
(10) (a) Corey, E. J.; Kang, J.; Kyler, K. Tetrahedron Lett. 1985, 26,
555-558. (b) Pine, S. H.; Shen, G. S.; Hoang, H. Synthesis 1991, 165-
167.
(7) Paquette, L. A.; Wang, T.-Z.; Sivik, M. R. J. Am. Chem. Soc. 1994,
116, 11323-11334.
(8) The compound 8b was prepared from the commercially available
(+)-Hajos-Parrish ketone in two steps (70% yield); see: Isaacs, R.; Di
Grandi, M. J.; Danishefsky, S. J. J. Org. Chem. 1993, 58, 3938-3941.
(9) There is only one report on a similar system; see: Niwa, H.; Nisiwaki,
M.; Tsukada, I.; Ishigaki, T.; Ito, S.; Wakamatsu, K.; Mori, T.; Ikagawa,
M.; Yamada, K. J. Am. Chem. Soc. 1990, 112, 9001-9003.
(11) Yan, T.-H.; Tsai, C.-C.; Chien, C.-T.; Cho, C.-C.; Huang, P.-C. Org.
Lett. 2004, 6, 4961-4963.
(12) Brown, R. F. C.; Burge, G. L.; Collins, D. J. Aust. J. Chem. 1983,
36, 117-134.
(13) Molander, G. A.; Quirmbach, M. S.; Silva, L. S.; Spencer, K. C.;
Balsells, J. Org. Lett. 2001, 3, 2257-2260.
5586
Org. Lett., Vol. 8, No. 24, 2006

Scheme 2
provided product 12 (85% yield) as a single isomer. The
3). No attempt was made to establish the stereochemistry of
phenyl sulfide. Acidic hydrolysis of compound 13 with aq
HCl-THF afforded ketosulfide 14 in quantitative yield. IBX
oxidation of ring C in 14, followed by sulfoxide formation
with magnesium monoperoxyphthalate (MMP) in ethanol and
dehydrosulfenylation with K2CO3 in refluxing toluene,
furnished an ∼1.2:1 mixture of spiro dienones 4a/4b (64%
yield from 14).19 Stereochemistry of the spiro centers was
established on the basis of 2D NMR experiments and the
major isomer contained desired stereochemistry at C-5.
Having established the stereochemistry of the dienones 4a
configuration of the 2-propenyl group was assigned the syn-
relation to the quaternary methyl group as the reagent
approaches from the convex face. Reduction14 of both double
bonds in compound 12 with Adam’s catalyst (PtO2) in
methanol cleanly provided a ∼2:1 mixture of 5a/5b in
quantitative yield which was easily separated by silica gel
column chromatography. The relative stereochemistry of the
minor isomer 5b was unambiguously established using
NOESY, COSY, HSQC, and HMBC.15 Accordingly, the
relative stereochemistry of the secondary methyl group of
major isomer 5a is syn to the quaternary methyl group, ring
junction hydrogen (4-H), and isopropyl side chain.
After securing the stereochemistry of all chiral centers,
the major isomer 5a was carried forward. Introduction of
the two double bonds on 5a proved to be more challenging
than expected. Acidic hydrolysis of compound 5a and one-
pot IBX oxidation16 of resultant diketone did not provide
dienone 4. Direct IBX oxidation of cyclopentanone of 5a
prior to acidic hydrolysis of the ketal failed to provide
cyclopentenone. Palladium(II) acetate catalyzed oxidation17
of TMS ether derived from 5a provided only small amounts
of the desired cyclopentenone. We then focused our attention
on installing the cyclopentenone double bond prior to
cyclohexenone via sulfenylation-dehydrosulfenylation meth-
odology.18
Scheme 3
Reaction of compound 5a with LDA (3.0 equiv) at -70
°C followed by quenching with PhSSO2Ph provided phenyl
sulfide derivative 13 (86% yield) as the sole product (Scheme
(14) Reaction with 10% Pd-C gave an ∼5:1 mixture of 5a/5b in small
scale reaction. Large-scale reaction gave partial isomerized products which
were difficult to reduce.
(15) The major isomer 5a had confounding conformers which precluded
its full assignment.
(16) Nicolaou, K. C.; Zhong, Y.-L.; Baran, P. S. J. Am. Chem. Soc. 2000,
122, 7596-7597.
(17) Ito, Y.; Hirao, T.; Saegusa, T. J. Org. Chem. 1978, 43, 1011-1013.
(18) Kosugi, H.; Ku, J.; Kato, M. J. Org. Chem. 1998, 63, 6939-6946.
Org. Lett., Vol. 8, No. 24, 2006
5587

and 4b, there was one stereocenter that needed to be
introduced on cyclohexenone via methyl addition. Finally,
regioselective addition of methyl magnesium chloride on
cyclohexenone of 4a at -100 to -80 °C afforded a 2:1
separable mixture of cyanthiwigin AC (2) and 12-epi-
cyanthiwigin AC (15) (91% yield). 1H and 13C NMR spectra
of synthetic cyanthiwigin AC (2) were identical to spectra
obtained from the authentic natural product.2b
The first total synthesis of (+)-cyanthiwigin AC (2)20 was
accomplished in 13 steps with 2.0% overall yield starting
from the known TBS ether derivative 8b. Our synthetic
strategy, relying on the efficient spiroannulation step, is
expected to provide rapid access to other unnatural isomers.
Biological screening and syntheses of other isomers of
cyanthiwigin AC are in progress. Optimization of stereo- and
regioselectivities will be addressed after identifying the most
promising isomer by biological screening.
Acknowledgment. We thank Professor Mark Hamann
and Dr. Jiangnan Peng for providing copies of spectra of
authentic cyanthiwigin AC and Dr. Daniel Guay for useful
discussions. G.B. thanks the Merck-Frosst Co-op program
for financial support.
Supporting Information Available: Full experimental
details and 1H and 13C NMR spectra for compounds 2, 4-7,
10-12, and 15. This material is available free of charge via
the Internet at http://pubs.acs.org.
(19) Oxidation of ring C was critical prior to dehydrosulfenylation of
ring A (14). If the sequence is inverted, the olefinic proton of ring A
disappears during IBX oxidation for a compound obtained from 5b.
Presumably, the absence of olefinic proton in 1H and 13C NMR spectra
indicated the formation of tetracylic compound arising from the Michael
addition of cyclohexanone enol onto cyclopentenone.
(20) Spectral data of all compounds are consistent with their structures.
Yields refer to isolated yields.
OL062304H
5588
Org. Lett., Vol. 8, No. 24, 2006