Synthesis of the cyclohexane core of phomactins and a new route to the bicyclo[9.3.1]pentadecane diterpenoid skeleton

Article abstract of DOI:10.1021/ol1026816

Conjugate reduction of an enone accompanied by in situ intramolecular aldol condensation was used to construct the tetrasubstituted cyclohexane nucleus of phomactins. Subsequent relay ring-closing metathesis completed the nine-membered ansa bridge of the

Full text of DOI:10.1021/ol1026816

ORGANIC
LETTERS
2011
Vol. 13, No. 2
248-251
Synthesis of the Cyclohexane Core of
Phomactins and a New Route to the
Bicyclo[9.3.1]pentadecane Diterpenoid
Skeleton
Keith D. Schwartz and James D. White*
Department of Chemistry, Oregon State UniVersity, CorVallis, Oregon 97331, United States
james.white@oregonstate.edu
Received November 4, 2010
ABSTRACT
Conjugate reduction of an enone accompanied by in situ intramolecular aldol condensation was used to construct the tetrasubstituted cyclohexane
nucleus of phomactins. Subsequent relay ring-closing metathesis completed the nine-membered ansa bridge of the diterpenoid framework.
The phomactins comprise a family of diterpenoids containing
the unusual bicyclo[9.3.1]pentadecane skeleton,1 phomactins
A (1,2 D (2)3 and G (3)4 being representative of the class
(Figure 1). The core unit of these structures is a substituted
cyclohexane or cyclohexene that is bridged by an unsaturated
ansa chain. The intriguing structures of phomactins together
with their activity as platelet activating factor (PAF) antago-
nists5 have drawn intense synthetic activity toward these
Figure 1. Representative members of the phomactin family of
diterpenoids.
molecules with the result that there are now completed
syntheses of 1,6 2,7 and 3.6b In addition, Wulff has published
an elegant synthesis of phomactin B2,8 and there are
numerous reports of partially completed routes to these and
other phomactins.9
Our plan for building the cyclohexane core of phomactins
is shown in Scheme 1 and envisioned construction of 4 from
an acyclic precursor 5 containing all of the carbons needed
(1) Goldring, W. P. D.; Pattenden, G. Acc. Chem. Res. 2006, 39, 354–
361. The bicyclo[9.3.1]pentadecane skeleton is found only in one other class
of diterpenoids, the verticillanes, although the Taxane framework can be
viewed as a tricyclic relative of this family.
(2) Sugano, M.; Sato, A.; Iijima, Y.; Oshima, T.; Furuya, K.; Kuwano,
H.; Hata, T.; Hanzaw, H. J. Am. Chem. Soc. 1991, 113, 5463–5464.
(3) Sugano, M.; Sato, A.; Iijima, Y.; Furuya, K.; Haruyama, H.; Yoda,
K.; Hata, T. J. Org. Chem. 1994, 59, 564–569.
(6) (a) Goldring, W. P. D.; Pattenden, G. Chem. Commun. 2002, 1736–
1737. (b) Goldring, W. P. D.; Pattenden, G. Org. Biomol. Chem. 2004,
466–473. (c) Mohr, P. J.; Halcomb, R. L. J. Am. Chem. Soc. 2003, 125,
1712–1713. (d) Tang, Y.; Cole, K. P.; Buchanan, G. S.; Li, G.; Hsung,
R. P. Org. Lett. 2009, 11, 1591–1594.
(4) Sugano, M.; Sato, A.; Iijima, Y.; Furuya, K.; Kuwano, H.; Hata, T.
J. Antibiot. 1995, 48, 1188–1190.
(5) Braquet, P.; Touqui, L.; Shen, T. Y.; Vargaftig, B. B. Pharmacol.
ReV. 1987, 39, 97–145. Phomactin D (2), with an IC50 of 0.12 µM in PAF
receptors, is the most biologically active member of the group. (Sugano,
M.; Sato, A.; Saito, K.; Takaishi, S.; Matsushita, Y.; Iijima, Y. J. Med.
Chem. 1996, 39, 5281–5284)
(7) Miyaoka, H.; Saka, Y.; Miura, S.; Yamada, Y. Tetrahedron Lett.
1996, 37, 7107–7110.
(8) Huang, J.; Wu, C.; Wulff, W. D. J. Am. Chem. Soc. 2007, 129,
13366–13367.
10.1021/ol1026816  2011 American Chemical Society
Published on Web 12/16/2010

Scheme 1
.
Retrosynthetic Analysis of the Phomactin Core
Scheme 2. Synthesis of Ketophosphonate 7
ozonolysis furnished aldehyde 13 which underwent Wittig
olefination to give heptenoate 14. Treatment of 14 with the
lithio anion of diethyl methylphosphonate gave ꢀ-ketophos-
phonate 15.
for the phomactin framework as well as three key stereo-
centers.10 The configuration of 5 was considered crucial for
the final stage of a phomactin synthesis that would close the
nine-membered bridge across a preformed cyclohexane.
Assembly of 1,8,15-hexadecatriene 5 was projected from
coupling of aldehyde 6 with ketophosphonate 7, the latter
being derived from octenoate 8 which is accessible from
geraniol. Oxidative cleavage of cyclopentene 9 was pro-
grammed as the means for acquiring 6.
Bode’s clever synthesis of ꢀ-hydroxy esters from R,ꢀ-
epoxyaldehydes provided a convenient entry to (S)-11 from
aldehyde 10 (Scheme 2).11 This aldehyde was prepared from
geraniol via Sharpless asymmetric epoxidation12 followed
by Parikh-Doering oxidation.13 Because our projected
closure of the phomactin ansa bridge was to employ ring-
closing metathesis,14 we decided to replace the trisubstituted
alkene of 11 with a terminal vinyl group in the expectation
that this change would facilitate complexation with the RCM
catalyst. After protection of alcohol 11 as its TES ether 12,
Ketoaldehyde 6, projected as the coupling partner for 15,
required placing a pair of methyl substituents at vicinal
stereogenic carbons, one of which is quaternary. Cyclopen-
tene 9, as the precursor to 6, must therefore have a (4R,5S)
configuration in order for oxidative scission of the trisub-
stituted alkene to deliver 6. The starting point for our route
to 9 was (R)-pulegone which was converted via a known
route involving Favorskii ring contraction and reductive
ozonolysis to keto ester 16 (Scheme 3).15 Michael addition
of 16 to methyl vinyl ketone gave major isomer 17
accompanied by 11% of the diastereomeric diketo ester.16
Selective ketalization of the methyl ketone of 17 under
Noyori’s conditions17 afforded 18 which was converted to
enol triflate 19. Palladium-catalyzed methoxycarbonylation
of 19 then afforded diester 20.
Our goal with 20 was selective reduction of the methyl
ester to a primary alcohol, to be followed by exhaustive
reduction of the ethyl ester to a methyl group. This scenario
proved to be impractical, and 20 was therefore reduced
nonselectively to diol 21 with the prospect of differentiating
the two primary alcohols. The latter operation was ac-
complished by converting 21 to mono-p-methoxybenzyl ether
22. Various methods were examined for reducing the primary
alcohol of 22 to a methyl substituent, but the only successful
(9) (a) Seth, P. P.; Totah, N. I. Org. Lett. 2000, 2, 2507–2509. (b) Teng,
D.; Wang, B.; Augatis, A. J.; Totah, N. I. Tetrahedron Lett. 2007, 48, 4605–
4607. (c) Chemler, S. R.; Danishefsky, S. J. Org. Lett. 2000, 2, 2695–
2698. (d) Chemler, S. R.; Iserloh, U.; Danishefsky, S. J. Org. Lett. 2001, 3,
2949–2951. (e) Houghton, T.; Choi, S.; Rawal, V. H. Org. Lett. 2001, 3,
3615–3617. (f) Mi, B.; Maleczka, R. Org. Lett. 2001, 3, 1491–1494. (g)
Balnaves, A. S.; McGowan, G.; Shapland, P. D. P.; Thomas, E. J.
Tetrahedron Lett. 2003, 44, 2713–2716.
(10) Schwartz, K. D.; White, J. D. Book of Abstracts, 239th ACS
National Meeting, San Francisco, CA, March 21-25, 2010; American
Chemical Society: Washington, DC, 2010; ORG-129.
(11) Chow, K. Y. K.; Bode, J. W. J. Am. Chem. Soc. 2004, 126, 8126–
8127.
(14) Grubbs, R. H.; Miller, S. J.; Fu, G. C. Acc. Chem. Res. 1995, 28,
446–452.
(15) Marx, J. N.; Norman, L. R. J. Org. Chem. 1975, 40, 1602–1606.
(16) Ouvrard, N.; Rodriguez, J.; Santelli, M. Angew. Chem., Int. Ed.
1992, 31, 1651–1653.
(12) Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102, 5974–
5976.
(13) Narco, K.; Baltas, M.; Escudier, J.-M.; Gorrichon, L. Tetrahedron
1996, 52, 9047–9056.
(17) Tsunoda, Y.; Suzuki, M.; Noyori, R. Tetrahedron Lett. 1980, 21,
1357–1358.
Org. Lett., Vol. 13, No. 2, 2011
249

Scheme 3
.
Synthesis of Cyclopentene 24
Scheme 4. Assembly of Enedione 26 and Its Cyclization via
Conjugate Reduction and Aldol Condensation
the side chain of cyclopentene 9, i.e. prior to HWE
condensation with 15. Since ozonolytic cleavage of the
cyclopentene would no longer be selective in this case, a
modified sequence was introduced that first took 24 to cis
diol 30 (Scheme 5). The diol was protected as cyclic
carbonate 31 and acid hydrolysis then gave ketone 32.
Although Wittig methylenation of this ketone was straight-
forward and the product was advanced to 5 without incident,
a decision was made at this juncture to invest 32 in an
olefination that afforded diene 33. This choice was motivated
by the worrisome prospect that formation of the phomactin
ansa bridge from a diene such as 4 by conventional ring-
closing metathesis would be difficult. On the other hand,
relay ring-closing metathesis (RRCM)20 appeared to have a
better chance of success.21 To this end, cyclic carbonate 33
was hydrolyzed to diol 34 which underwent clean oxidative
cleavage with lead tetraacetate to keto aldehyde 35.
In order to ensure that the metathesis catalyst would deploy
initially at the vinyl group of the sacrificial pentenyl unit
rather than the distal olefin, the latter was replaced in 15 by
a propenyl terminus. This modification simply required a
different olefination of 13 to give 36. Ester 36 was advanced
to ketophosphonate 37 which upon HWE condensation with
aldehyde 35 afforded diketone 38 (Scheme 6). As before,
exposure of 38 to Stryker’s copper hydride reagent resulted
in conjugate reduction and in situ aldol cyclization to yield
hydroxy ketone 39.
route to 24 proved to be Wolff-Kishner reduction of
aldehyde 23 obtained by oxidation of 22.18
Ozonolytic ring cleavage of 24 cleanly furnished keto
aldehyde 25 which underwent Horner-Wadsworth-Emmons
(HWE) condensation with ketophosphonate 15 to give (E)-
enone 26 as the sole isomer (Scheme 4). As we had
anticipated, conjugate reduction of 26 resulted in spontaneous
aldol cyclization of the intermediate enolate to produce
hydroxy ketone 27. The most effective reagent for this
transformation was Stryker’s copper hydride system.19 An
attempt was made to extend this useful ring construction by
exposing 26 to tetra-n-butylammonium fluoride in the hope
that the alkoxide generated from silyl ether cleavage would
trigger a tandem sequence that would install the carbocyclic
and tetrahydropyran rings of phomactin A concomitantly.
Instead, the enolate generated from R-alkoxy ketone 26 under
these conditions led to intramolecular Michael adduct 28. A
further surprise awaited us when ethylene ketal 27 was
subjected to acidic hydrolysis. Expecting that Wittig meth-
ylenation of the resultant methyl ketone could lead to 4, we
found instead that internal ketal 29 was formed.
The flaw in our synthesis plan exposed by 29 was corrected
by installing the exo methylene function needed for 5 within
(18) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis
(20) Wallace, D. J. Angew. Chem., Int. Ed. 2005, 44, 1912–1915.
(21) For an analogous RRCM, see: McGrath, N. A.; Lee, C. A.; Araki,
H.; Brichacek, M.; Njardarson, J. T. Angew. Chem., Int. Ed. 2008, 47, 9450–
9453.
1994, 639–666.
(19) Chiu, P.; Szeto, C. P.; Geng, Z.; Cheng, K. F. Org. Lett. 2001, 3,
1901–1903.
250
Org. Lett., Vol. 13, No. 2, 2011

Scheme 5
.
Synthesis of Keto Aldehyde 35 for RRCM
Scheme 6. Synthesis of RRCM Precursors 40 and 41
Scheme 7. RRCM of 40
Initial attempts at RRCM of 39 were not encouraging, but
after dehydration of 39 to separable exo and endo alkenes,
40 and 41, a more favorable outcome resulted. In particular,
RRCM of 40 (mixture of E and Z isomers) with Grubbs
second generation catalyst gave 42 in high yield (Scheme
7). Although 42 was formed exclusively with the (Z) olefin
configuration in the nine-membered bridge, protocols for
olefin inversion to the C7-C8 (E) configuration of natural
phomactins are available and are currently under study.22
In summary, we have described a new route to the
tetrasubstituted cyclohexane core of phomactins from an
acyclic precursor that incorporates all of the carbon atoms
present in the diterpenoid skeleton. The key step is conjugate
reduction of an enone accompanied by intramolecualr aldol
condensation. The assembled core was shown to undergo
efficient relay ring-closing metathesis to generate a bicyclo-
[9.3.1]pentadecane system that is potentially a precursor to
members of the phomactin family.
Acknowledgment. We are grateful to Dr. William Martin
of this laboratory for helpful suggestions and to the National
Science Foundation for financial support (0413994-CHE).
Support for the Oregon State University NMR Facility from
the Murdock Charitable Trust (Grant 2005265) and the
National Science Foundation (CHE-0722319) is gratefully
acknowledged.
Supporting Information Available: Experimental details,
1
characterization data, and H and 13C NMR spectra of new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(22) Smith, A. B.; Mesaros, E. F.; Meyer, E. A. J. Am. Chem. Soc. 2006,
128, 5292–5299.
OL1026816
Org. Lett., Vol. 13, No. 2, 2011
251