Synthesis of the complete carbocyclic skeleton of vinigrol

Article abstract of DOI:10.1021/ol034217k

(Matrix presented) An efficient entry to the fully elaborated skeleton of vinigrol is described. The installation of the desired stereochemistry at C(12) and the construction of the eight-membered ring were achieved in one operation by a remarkably facile anionic oxy-Cope rearrangement of Z-isopropenyl isomer 24.

Full text of DOI:10.1021/ol034217k

ORGANIC
LETTERS
2003
Vol. 5, No. 7
1139-1142
Synthesis of the Complete Carbocyclic
Skeleton of Vinigrol
,†
Lionel Gentric,† Issam Hanna,* and Louis Ricard‡
Laboratoire de Synthe`se Organique associe´ au CNRS and Laboratoire
“He´te´roe´le´ments et Coordination” associe´ au CNRS, Ecole Polytechnique,
F-91128 Palaiseau Cedex, France
hanna@poly.polytechnique.fr
Received February 6, 2003
ABSTRACT
An efficient entry to the fully elaborated skeleton of vinigrol is described. The installation of the desired stereochemistry at C(12) and the
construction of the eight-membered ring were achieved in one operation by a remarkably facile anionic oxy-Cope rearrangement of Z-isopropenyl
isomer 24.
Vinigrol (1) is a diterpenoid isolated in 1987 from a culture
of the fungal strain identified as Virgaria nigra.1 Very early
on, it was demonstrated that this natural product decreased
arterial blood pressure in rats in a dose-dependent manner
and inhibited platelet activating factor and epinephrine
induced platelet aggregation.2 In addition, it was found that
1 is a powerful tumor necrosis factor (TNF) antagonist.
Therefore, vinigrol may be used for the treatment of
endotoxic shock inflammation, infections, and cachexia and
to arrest progression from AIDS-related complex to AIDS.3
Structurally, vinigrol possesses a unique tricyclic skeleton
2, involving a bridged eight-membered ring. The unusual
structure of this natural product combined with its interesting
biological activities provide a challenging synthetic target.
However, despite the efforts of many groups,4 a complete
total synthesis of 1 has yet to be reported. As for cyclo-
octanoid substances, the most critical issue in the synthesis
of vinigrol is the construction of the eight-membered ring.
Our strategy is based on the recognition that the oxygenated
tricyclic skeleton 2 of vinigrol can be quickly elaborated via
an anionic oxy-Cope rearrangement5 of a tricyclic vinyl
carbinol such as 3, which could arise from stereoselective
alkylation of enone 4 (Scheme 1). Previous results from our
laboratory established the feasibility of this approach and
described the first successful entry into the functionalized
decahydro-1,5-butanonaphthalene ring system of this natural
product in a very concise manner.6 However, the previously
described model was not adequately functionalized for the
expeditious completion of the synthesis. In particular, the
stereoselctive introduction of the isopropyl group at C(12)7
seemed to be a major problem. We now disclose an efficient
(4) (a) Mehta, G.; Reddy, K. S. Synlett 1996, 625-627. (b) Matsuda,
F.; Kito, M.; Sakai, T.; Okada, N.; Miyashita, M.; Shirahama, H.
Tetrahedron 1999, 55, 14369-14380. (c) Efremov, I. V. Ph.D. Dissertation,
The Ohio State University, 2001. (d) Goodman, S. N. Ph.D. Dissertation,
Harvard University, 2000.
(5) Evans, D. A.; Golob, A. M. J. Am. Chem. Soc. 1975, 97, 4765-
4766. For reviews on stereocontrolled construction of complex cyclic
ketones via Oxy-Cope rearrangements see: Paquette, L. A. Angew. Chem.,
Int. Ed. Engl. 1990, 29, 609-626. Paquette, L. A. Tetrahedron 1997, 53,
13971-14020.
(6) Devaux, J.-F.; Hanna, I.; Lallemand, J.-Y.; Prange´, T. J. Org. Chem.
1993, 58, 2349-2350. Devaux, J.-F.; Hanna, I.; Lallemand, J.-Y. J. Org.
Chem. 1997, 62, 5062-5068.
(7) The numbering used in this paper refers to the corresponding centers
of vinigrol.
† Laboratoire de Synthe`se Organique associe´ au CNRS.
‡ Laboratoire “He´te´roe´le´ments et Coordination” associe´ au CNRS.
(1) Uchida, I.; Ando, T.; Fukami, N.; Yoshida, K.; Hashimoto, M.; Tada,
T.; Koda, S.; Morimoto, Y. J. Org. Chem. 1987, 52, 5292-5293.
(2) (a) Ando, T.; Tsurumi, Y.; Ohata, N.; Ushida, I.; Hoshida, K.;
Okuhara, M. J. Antibiot. 1988, 41, 25-30. (b) Ando, T.; Yoshida, K.;
Okuhara, M. J. Antibiot. 1988, 41, 31-35.
(3) (a) Norris, D. B.; Depledge, P.; Jakson, A. P. PCT Int. Appl. WO 91
07,953; Chem. Abstr. 1991, 115, 64776. (b) Nakajima, H.; Yamamoto, N.;
Kaisi, T. Jpn. Kokai Tokyo Koho JP 07206668; Chem. Abstr. 1995, 123,
246812.
10.1021/ol034217k CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/14/2003

Scheme 1
Scheme 2
entry to the fully elaborated vinigrol carbon skeleton. In this
study, methyl groups at C(8) and C(9) were introduced at
the first stage and installation of the isopropyl at C(12) was
planned before the oxy-Cope rearrangement as outlined in
Scheme 1. To this end the tricyclic ketone 4, bearing the
methyl groups and conveniently positioned oxygen func-
tionalities, was first elaborated.
The synthesis began with the readily available trimethyl-
silyl enol ether 58 (Scheme 2). Heating this diene in the
presence of 1,4-benzoquinone in THF gave 7 in quantitative
yield. Treatment of 7 with a catalytic amount of BF3-Et2O
in THF at -70 °C cleanly led to the mixed trimethylsilyl
ketal 8. This alcohol was submitted to the Mitsunobu
inversion,13 using 4-nitrobenzoic acid,14 followed by cleavage
of the resulting ester under mild conditions (K2CO3, MeOH-
Et2O) to give 9. It is worthy of note that the first chromato-
graphic purification only occurred at this stage. The overall
yield of this five-step sequence, conveniently executed on a
20-g scale, was 65-70%.
Next, the hydroxymethyl group at C(3) was stereoselec-
tively introduced, using the Stork procedure.15 Thus, bromo-
methyldimethylsilyl ether 10, prepared by silylation of 9
(BrCH2SiMe2Cl, DMAP, Et3N), was subjected to standard
high-dilution, radical-generating conditions followed by
Tamao oxidation16 (H2O2, Na2CO3,THF-MeOH) to furnish
diol 11.
Treatment of 11 with 2,2-dimethoxypropane in the pres-
ence of p-toluenesulfonic acid resulted in the formation of
the acetonide protecting group concomitant with hydrolysis
of the mixed TMS ketal affording 12 in 67% overall yield
for the four-step sequence. Dehydration of 12 with phos-
phorus oxychloride in pyridine led to the key tricycyclic
ketone 4 in 93% yield.
Exposure of 4 to 3-methylbutynylmagnesium bromide17
in THF gave rise to a 4.2:1 mixture of exo- and endo-
propargyl alcohols 13 and 14 (94%) which were readily
separated by flash chromatography. In contrast to previous
observations in the model study,6 the presence of the exo-
methyl group at C(9) in 420 promoted the addition of the
Grignard reagent from the sterically less hindered endo face.
Since the oxy-Cope rearrangement proceeds via a chairlike
transition state, introduction of an isopropyl at C(12) with
the desired configuration requires a Z-geometry of the double
bond in the allylic alcohol precursor 15. This compound was
(8) Diene 5 was prepared from the commercialy available 2,6-dimethyl-
cyclohexanone by oxidation with NBS in refluxing CCl4 (4 h) in the
presence of a catalytic amount of AIBN (90%)9 followed by silylation of
the resulting 2,6-dimethylcyclohex-2-enone.10
(9) Kende, A. S.; Fludzinski, P. H.; Hill, J. H.; Swenson, W.; Clardy, J.
J. Am. Chem. Soc. 1984, 106, 3551-3562.
(10) Rubottom, G. M.; Gruber, J. M. J. Org. Chem. 1977, 42, 1051-
1056.
(16) Tamao, K.; Kakui, T.; Akita, M.; Iwahara, T.; Kanatani, R.; Yoshida,
J.; Kumada, M. Tertrahedron 1983, 39, 983-990. Tamao, K.; Ishida, N.;
Ito, Y.; Kumada, M. Org. Synth. 1990, 69, 96-105.
(11) For similar transformations see: (a) Hung, S. C.; Liao, C. C.
Tetrahedron Lett. 1991, 32, 4011-4014. Hung, S. C.; Liao, C. C. J. Chem.
Soc., Chem. Commun. 1993, 1457-1458. (b) Chu-Moyer, M. Y.; Dan-
ishefsky, S. J.; Schulte, G. K. J. Am. Chem. Soc. 1994, 116, 11213-11228.
(12) Gemal, A. L.; Luche, J. L. J. Am. Chem. Soc. 1981, 103, 5454-
5459.
(17) This Grignard reagent was prepared by the action of ethylmagnesium
bromide on isopropylacetylene, which was obtained from 2-methylbut-3-
yn-1-ol in two steps: bromination (47% aq HBr, CuBr, NH4Br, Cu)18
followed by reduction of the resulting bromoallene with LiAlH4 in
DME.19
(18) Landor, S. R.; Patel, A. N.; Whiter, P. F.; Greaves, P. M. J. Chem.
Soc. C 1966, 1223-1226.
(13) Mitsunobu, O. Synthesis 1981, 1-28. Hughes, D. L. Org. React.
1992, 42, 335-656.
(19) Fleming, I.; Lawrence, N. J. J. Chem. Soc., Perkin Trans. 1 1992,
3309-3326.
(14) Dodge, J. A.; Trujillo, J. I.; Presnell, M. J. Org. Chem. 1994, 59,
234-236.
(20) The exo configuration of the methyl group at C(9) was confirmed
by X-ray crystallographic anallysis of an analogue compound.21
(15) Stork, G.; Kahn, M. J. Am. Chem. Soc. 1985, 107, 500-501.
1140
Org. Lett., Vol. 5, No. 7, 2003

Scheme 3
Scheme 4
easily obtained by semihydrogenation of 14 (H2, BaSO4,
quinoline) (Scheme 3). However, all attempts of oxy-Cope
rearrangement on 15 under a variety of anionic (KH or
KHMDS in refluxing THF or toluene, with 18-Cr-6) or
thermolytic (refluxing Decalin) conditions failed. The starting
alcohol was either recovered unchanged or decomposed,
depending upon the experimental conditions. Obviously,
there is considerable steric crowding between the olefinic
moiety and the isopropyl group when the side chain of 15 is
forced to reach the transition state, which impedes any
concerted rearrangement. In contrast, when the E-isomer 17
was subjected to potassium hydride in refluxing THF,
compound 18 was isolated in 78% yield. This structure was
assigned on the basis of spectroscopic data and the config-
uration of alkyl goups was confirmed later by X-ray
crystallographic analysis of epoxyalcohol 20. Both alkyl
groups at C(9) and C(12) have the opposite stereochemistry
to that of vinigrol (Scheme 3). While epimerization at C(9)
could be achieved under basic conditions, the inversion of
the configuration of isopropyl at C(12) seemed to be a major
hurdle again.
Many approaches were surveyed to solve this problem
without success.21 After considerable experimentation, the
solution came from an unexpected observation. We were
pleased to discover that when the isopropyl group in 15 was
replaced by an isopropenyl (compound 24, Scheme 4), the
oxy-Cope rearrangement took place affording the skeleton
of vinigrol with the correct relative configuration at C(12).
Alcohol 24 was prepared from ketone 4 as indicated in
Scheme 4. Treatment of 4 with an excess of Grignard reagent
2122 gave a separable mixture of exo and endo alcohols 22
and 23 in 80% and 14% yields, respectively. Reduction of
22 with Rieke zinc23 cleanly provided diene 24 in good yield
(61%, 94% based on recovered enyne).24 Exposure of 24 to
KH in refluxing THF for 1 h led to the rearrangement product
25 in 50% along with the starting material (20%). The
structure of 25 was deduced from the spectroscopic data,
and the relative configuration of isopropenyl at C(12) was
confirmed by X-ray crystallographic analysis. Catalytic
hydrogenation of 25 in the presence of rhodium on alumina
gave 16 in quantitative yield. After a brief survey of reaction
conditions, we found that simply heating 24 and NaH (instead
of KH) in THF for 18 h cleanly produced 25 in 72% yield
(92% based on recovered starting material) with high
reproducibility.25 This observed rate acceleration is probably
the result of stabilization of the transition state of the anionic
oxy-Cope rearrangement by an additional unsaturation on
the terminal position.26 Under the same conditions, E-isomer
26 underwent the same rearrangement affording 27 in 50%
(22) Grignard reagent 22 was prepared from 2-methyl-1-en-3-yne
(Brandsma, L. In PreparatiVe Acetylenic Chemistry, 2nd ed.; Elsevier:
Amsterdam, The Netherlands, 1988; p 203) and ethylmagnesium bromide
in THF-ether at room temperature.
(23) Rieke, R. D. Aldrichim. Acta 2000, 33, 52-60. Chou, W. N.; Clark,
D. L.; White, J. B. Tetrahedron Lett. 1991, 32, 299-302.
(24) Semihydrogenation of the triple bond with Lindlar’s catalyst led to
a mixture of products from which the desired diene 23 was isolated in 30-
40% yield.
(21) Details will be reported later in the full account of this work.
Org. Lett., Vol. 5, No. 7, 2003
1141

synthesis: removal the carbonyl group at C(10) and inversion
of C(9) methyl stereochemistry. To this end ketone 16 was
reduced with LiAlH4 and the resulting alcohol 28 was
dehydrated with POCl3 to afford diene 29 (Scheme 5).
Treatment of 29 with 1 equiv of m-CPBA at 0 °C selectively
affected the tetrasubstituted double bond leading to epoxide
30 in 95% yield.27 The remaining double bond was then
hydrogenated in the presence of rhodium on alumina to give
a 2:3 mixture of 31 and 32 in 94% yield. The relative
configuration of the methyl group at C(9) in 32, which was
established by NOESY and confirmed by X-ray crystal-
lographic analysis, was found to be the same compared with
that of vinigrol.
Scheme 5
In summary, this work describes an efficient synthesis of
the complete carbocyclic skeleton of vinigrol. Our finding
of the remarkable rate acceleration of the oxy-Cope re-
arrangement of Z-isopropenyl isomer provided a neat solution
to the major hurdle in this synthesis. As a result, construction
of the eight-membered ring and introduction of isopropenyl
at C(12) with the correct stereochemistry were achieved in
one operation. With compound 31, we have in hand an
advanced precursor that may hold the key to a successful
total synthesis of vinigrol.
Supporting Information Available: Experimental pro-
cedure and characterization data for compounds 4-32. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(75%) yield. Subsequent catalytic hyrogenation of 27 pro-
vided 18, previously obtained from 17.
OL034217K
With the required relative configuration of the isopropyl
at C(12) in hand, we turned to the next phase of the
(26) To our knowledge, only one example of such an effect has hitherto
been reported. See: Paquette, L. A.; DeRussy, D. T.; Rogers, R. D.
Tetrahedron 1988, 44, 3139-3148.
(27) Initial attempts to carry out the epoxidation of 29 failed: the resulting
epoxide rapidly underwent an intramolecular ring opening by the hydroxyl
group producing the corresponding cyclic ether.
(25) In this reaction NaH (80% dispersion in mineral oil) was used
without washing, and the solvent was not degassed. In contrast, exposure
of 23 to oil-free KH in carefully degassed THF at reflux for more than 3
h led to a mixture of products from which 24 was isolated in modest yield.21
1142
Org. Lett., Vol. 5, No. 7, 2003