A chemoenzymatic synthesis of the cis-decalin core associated with the novel anti-mitotic agent phomopsidin: Some observations concerning a high-pressure-promoted diels-alder cycloaddition reaction of (1S,2R)-3-methyl-cis-1,2-dihydrocatechol and the anionic oxy-cope rearrangement of compounds derived from the adduct

Article abstract of DOI:10.1071/CH04036

The enantiomerically pure and enzymatically derived cis-1,2-dihydrocatechol 2 engages in a diastereofacially selective Diels-Alder cycloaddition reaction with commercially available lactone 3 at 19 kbar to afford adduct 4, which is readily elaborated to the diene-ol 13. Treatment of this last compound with KH/18 [crown]-6 resulted in successive anionic oxy-Cope and 1,2-Wittig rearrangements to afford acyloin 14 embodying the cis-decalin core associated with the natural product phomopsidin (1). Compound 16 also engages in an anionic oxy-Cope rearrangement reaction to give, depending on the molar equivalents of base used, either the cis-decalin 17 or the hexahydroindene 18. The structure of compound 18 has been established by single-crystal X-ray diffraction analysis.

Full text of DOI:10.1071/CH04036

Rapid Communication
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2004, 57, 641–644
A Chemoenzymatic Synthesis of the cis-Decalin Core Associated with
the Novel Anti-Mitotic Agent Phomopsidin: Some Observations
Concerning a High-Pressure-Promoted Diels–Alder Cycloaddition
Reaction of (1S,2R)-3-Methyl-cis-1,2-dihydrocatechol and the Anionic
Oxy–Cope Rearrangement of Compounds Derived from the Adduct∗
Martin G. Banwell,A,B Alison J. Edwards,A Malcolm D. McLeod,A and Scott G. StewartA
A Research School of Chemistry, Institute of Advanced Studies, Australian National University,
Canberra ACT 0200, Australia.
B Author to whom correspondence should be addressed (e-mail: mgb@rsc.anu.edu.au).
The enantiomerically pure and enzymatically derived cis-1,2-dihydrocatechol 2 engages in a diastereofacially selec-
tive Diels–Alder cycloaddition reaction with commercially available lactone 3 at 19 kbar to afford adduct 4, which is
readily elaborated to the diene-ol 13. Treatment of this last compound with KH/18[crown]-6 resulted in successive
anionic oxy-Cope and 1,2-Wittig rearrangements to afford acyloin 14 embodying the cis-decalin core associated
with the natural product phomopsidin (1). Compound 16 also engages in an anionic oxy-Cope rearrangement reac-
tion to give, depending on the molar equivalents of base used, either the cis-decalin 17 or the hexahydroindene 18.
The structure of compound 18 has been established by single-crystal X-ray diffraction analysis.
Manuscript received: 12 February 2004.
Final version: 6 April 2004.
CO2H
In 1997 Kobayashi and coworkers first reported on a
Phomopsis sp. (strain TUF95F47) fungus collected from a
fallen mangrove branch found on the bottom of a coral reef in
Pohnpei, Micronesia.[1] They observed that filtered samples
21
19
H
H
from the cultured broth of this fungus induced deformations
of mycelia in germinated conidia of Pyricularia oryzae P-
2b thus prompting further assay and isolation work.[1,2] As
a result the cis-decalin 1 (Scheme 1), named phomopsidin,
was obtained and identified as the biologically active mate-
rial. Follow-up studies revealed that phomopsidin exhibits
an IC50 value of 5.7 µM in a porcine brain microtubule
assembly assay and compares well, therefore, with the per-
formance of two prominent and potent anti-mitotic agents,
namely rhizoxin (4 µM) and colchicine (10 µM), under the
same conditions. Interestingly, the 16Z-isomer of phomop-
sidin is also a natural product (MK8383),[3] isolated from a
terrestrial Phoma sp., and displays similar anti-microtubule
activity (IC50 8.0 µM).[1,2] The constitution and relative
stereochemistry of phomopsidin (1) follows from extensive
spectroscopic studies while the absolute configuration was
determined by the exciton chirality method.[2,4]
HO
HO
16
HO
1
2
Scheme 1.
development of a total synthesis that would deliver analogues
of value in elucidating the key structural elements respon-
sible for its tubulin-binding properties. On this basis, and
because of our ongoing interest in developing methods for
the enantioselective preparation of cis-decalins via anionic
oxy-Cope rearrangement (AOCR) of appropriately function-
alized and enantiomerically pure 2-vinylbicyclo[2.2.2]octen-
2-ols,[5] we now report on the rapid construction of the
carbobicyclic (cis-decalin) core of phomopsidin (1) from
the cis-1,2-dihydrocatechol 2 (Scheme 1). Starting material
2 can be obtained on a large scale and in an enantiopure
Since phomopsidin has an unusual structure for a com-
pound that exhibits colchicine-type anti-mitotic activity it
clearly warrants further study, especially in terms of the
∗ Dedicated to our colleague Professor Lew Mander on the occasion of his 65th birthday and in recognition of his manifold contributions to the discipline of
organic synthesis.
© CSIRO 2004
10.1071/CH04036
0004-9425/04/070641

642
M. G. Banwell et al.
O
O
O
O
O
19kbar
TBDMS-Cl
MeLi
O
RꢀO
RꢀO
HO
ꢂ
2
OH
OH
OR
OR
Compound 6
O
3
4
5 R ꢁ TBDMS, Rꢀ ꢁ H
6 R ꢁ H, Rꢀ ꢁ TBDMS
7 R ꢁ TBDMS, Rꢀ ꢁ H
8 R ꢁ H, Rꢀ ꢁ TBDMS
NaH
MeOH, Hꢂ
Compound 8
O
O
O
Dess–Martin
periodinane
RꢀO
TBDMSO
TBDMSO
H2CꢁCHMgBr
OMe
OH
OR
OMe
OMe
O
11 R ꢁ Rꢀ ꢁ H
12 R ꢁ H, Rꢀ ꢁ TBDMS
10
9
PMB-Cl Compound 11
O
O
O
PMBO
OH
PMB
O
H
H
H
H
KH,
18[crown]-6
PMBO
O
OH
OMe
OMe
OMe
13
14
15
(not observed)
Scheme 2.
form through the whole-cell biotransformation of toluene
using the genetically engineered micro-organism E. coli
JM109(pDTG601) that over-expresses the responsible
enzyme, namely toluene dioxygenase (TDO).[6] The present
work serves to further highlight the potential of compound
2 as a starting material in natural products synthesis[7]
and its capacity to participate in facially selective Diels–
Alder cycloaddition reactions with various dienophiles at
high pressure.[8] The novel base-promoted rearrangements
of some of the 2-vinylbicyclo[2.2.2]octen-2-ols derived from
the products of such Diels–Alder reactions is another note-
worthy feature of the work presented here.
The reaction sequence leading to a 2-vinylbicyclo[2.2.2]-
octen-2-ol capable of engaging in the required AOCR is
shown in Scheme 2. Thus, in the opening stages, reaction of
a dichloromethane solution of an approximately 1 : 2 mixture
of compounds 2 and 3 at 19 kbar for 24 h afforded the syn-
Diels–Alderadduct 4(mp186–187◦C, 56%basedon 2)asthe
majorproductofreaction. Only9%ofthecorresponding anti-
adduct was isolated. The selective formation of compound
4, the structure of which was confirmed by a single-crystal
X-ray diffraction analysis that will be detailed elsewhere,[8b]
is attributed to a stabilizing interaction between the C O
bonds within diene 2 and the π*-orbital of dienophile 3 at
the transition state for the relevant Diels–Alder process.[9,10]
Selective mono-protection of the diol 4 as the corresponding
tert-butyldimethylsilyl (TBDMS) ether was achieved under
standard conditions to give a chromatographically separa-
ble mixture of the undesired mono-ether 5 (22%) and the
desired isomer 6 (62%). The assigned structures follow from
an X-ray crystallographic study (see below) and the selective
formation of compound 6 is presumably the result of a kinetic
and/or thermodynamic preference for introduction of the
bulkyTBDMS protecting group at that hydroxy group, within
4, remote from the bridgehead methyl group. Consistent with
this notion, treatment of the minor regioisomer 5 with sodium
hydride effected its equilibration to a 3 : 1 mixture favoring
the required isomer 6. Installation of what would become the
C21 methyl group of phomopsidin 1 was achieved by treat-
ing compound 6 with six equivalents of methyllithium and in
this manner a chromatographically separable mixture of the
two isomeric lactols 7 (17%) and 8 (59%) was obtained. The
structures of these products follow from nuclear Overhauser
effect (nOe) difference measurements which establish that in
each isomer the newly introduced methyl group is in the less
hindered β-orientation. Protection of the labile lactol moiety
within compound 8 was necessary at this point and could be
achieved by its reaction with methanol at 0◦C in the presence
of catalytic amounts of camphorsulfonic acid. In this way
the acetal 9 (84%) was obtained as a single diastereoisomer,
the configuration of which follows from X-ray studies and
is presumably determined through operation of the anomeric

Chemoenzymatic Synthesis of the cis-Decalin Core Associated with Phomopsidin
643
11
O1
O3
NaH, MeI
O2
O4
O5
O
MeO
OH
OMe
Fig. 1. A molecule of compound 18 derived from a single-crystal X-ray
diffraction analysis.
KH (2 equiv.),
18[crown]-6
KH (4 equiv.),
18[crown]-6
16
rearrangementprocess, which wouldhaveservedtointroduce
the C19 methyl group, was observed. In an effort to promote
the tandem AOCR/[1,2]-Wittig rearrangement sequence
substrate 16 was treated with four equivalents of each of KH
and 18[crown]-6. However, under such conditions an inter-
converting mixture of the tetrahydroindanone 18 and its C2
epimer (43% combined yield) was observed and the struc-
ture of the former product was established by single-crystal
X-ray diffraction analysis (Fig. 1).[12] The origins of this
material are not entirely clear at this stage but a Favorskii-
type ring-contraction process[13] operating on a species such
as the enolate derived from ketone 17 might be involved.
Furthermore, the anion arising from methoxide ion-promoted
cleavage of the intermediate cyclopropanone could react with
adventitiousdioxygen aspart of aprocessleadingtointroduc-
tion of the ketone carbonyl moiety observed in product 18 and
its C2 epimer.Alternately, the more highly substituted enolate
derived from compound 17 could be trapped with dioxy-
gen providing, after protonation, a peroxyhemi-acetal. This
last species could then react, via initial nucleophilic addition
of methoxide ion to the carbonyl moiety, to give a ring-
cleaved diester which subsequently engages in a Dieckmann
cyclization[14] thus affording compound 18.
O
O
MeO
O
H
H
O
O
OMe
OMe
2
MeO
H
H
17
18
Scheme 3.
effect. In anticipation of introducing the vinyl moiety neces-
sary for theAOCR, alcohol 9 was oxidized to the correspond-
ing ketone 10 (91%) using the Dess–Martin periodinane.
Reaction of the latter compound with vinyl magnesium bro-
mide then afforded a chromatographically separable mixture
of the desired vinyl alcohol 11 (89%) and traces (<5%) of
the mono-ether 12. The facial selectivity of the nucleophilic
addition process leading to compound 11 is dictated by the
bulky TBDMSO moiety adjacent to the ketone carbonyl unit
within substrate 10 and leads to the stereochemistry required
for operation of the pivotal AOCR process. However, before
examining this step, diol 11 was first converted into the cor-
responding p-methoxybenzyl (PMB) ether 13 (97%) under
standard conditions. In keeping with expectations, the lat-
ter compound engaged in a smooth reaction on exposure to
two molar equivalents of KH in the presence of 18[crown]-6
although the product of reaction proved to be the acyloin 14
(50%) and not the anticipated ether 15. The structure of com-
pound 14 follows from, among other things, extensive nOe
difference, HMBC, and HMQC experiments. Presumably,
compound 14 arises through initial operation of the expected
AOCR process but the resulting enolate then engages in
an in situ enolate equilibrium followed by a [1,2]-Wittig
rearrangement to give, after protic work-up, the observed
product. The operation of related sequences has been noted
by Paquette during the course of his work on the synthesis of
taxoids.[11]
The reaction sequences outlined above provide an efficient
means for the construction, in enantiomerically pure form, of
the pivotal cis-decalin moiety associated with phomopsidin
(1). Work aimed at exploiting such results in developing a
total synthesis of this fascinating natural product and certain
analogues will be reported in due course.
Note added in proof (21 May 2004): The first total synthesis
of (+)-phomopsidin has been reported. T. Suzuki, K. Usui,
Y. Miyake, M. Namikoshi, M. Nakada, Org. Lett. 2004, 6,
553. doi:10.1021/OL036338Q
References
[1] M. Namikoshi, H. Kobayashi, T. Yoshimoto, T. Hosoya, J.
Antibiot. 1997, 50, 890.
In seeking to exploit the above-mentioned and unantici-
pated [1,2]-Wittig rearrangement for the purposes of intro-
ducing the C19 methyl group of phomopsidin (1), the diol
11 was converted (Scheme 3) into the corresponding mono-
O-methyl ether 16 (76%) by standard methods. However, in
contrast to the situation detailed above, subjection of com-
pound 16 to the same rearrangement conditions as applied
to substrate 13 afforded only the AOCR product 17 (45%).
Thus, no material arising from a subsequent [1,2]-Wittig
[2] H. Kobayashi, S. Meguro, T. Yoshimoto, M. Namikoshi,
Tetrahedron 2003, 59, 455. doi:10.1016/S0040-4020(02)01566-1
[3] F. Wakui, K. Harimaya, M. Iwata, R. Sashita, N. Chiba,
T. Mikawa, Jpn. Kokai, Tokkyo Koho JP 07 126 211 1995
[Chem. Abstr. 1995, 123, 105272b].
[4] Biosynthetically, phomopsidin is most likely produced via an
intramolecular Diels–Alder cycloaddition reaction of an open-
chain fatty acid precursor.[1,2]
[5] (a) M. G. Banwell, J. R. Dupuche, Chem. Commun. 1996, 869.
doi:10.1039/CC9960000869

644
M. G. Banwell et al.
(b) M. G. Banwell, J. R. Dupuche, R. W. Gable, Aust. J. Chem.
1996, 49, 639.
[9] (a) S. D. Kahn, W. J. Hehre, J. Am. Chem. Soc. 1987, 109, 663.
For recent reviews dealing with such effects see:
(c) M. G. Banwell, D. C. R. Hockless, J. W. Holman,
R. W. Longmore, K. J. McRae, H. T. T. Pham, Synlett
1999, 1491.
(b) J. M. Coxon, R. D. J. Froese, B. Ganguly, A. P. Marchand,
K. Morokuma, Synlett 1999, 1681.
(c) G. Mehta, R. Uma, Acc. Chem. Res. 2000, 33, 278. doi:10.
1021/AR990123S
[6] For an excellent review on methods for generating cis-
1,2-dihydrocatechols by microbial dihydroxylation of the
corresponding aromatics as well the synthetic applications of
these metabolites see: T. Hudlicky, D. Gonzalez, D. T. Gibson,
Aldrichimica Acta 1999, 32, 35.
[10] For examples of thermally induced reactions of cis-1,2-
dihydrocatechols which proceed in a syn-selective fashion see:
J. R. Gillard, D. J. Burnell, J. Chem. Soc., Chem. Commun.
1989, 1439. doi:10.1039/C39890001439
[7] (a) M. G. Banwell, A. J. Edwards, G. J. Harfoot, K. A. Jolliffe,
M. D. McLeod, K. J. McRae, S. G. Stewart, M. Vögtle, Pure
Appl. Chem. 2003, 75, 223.
[11] L. A. Paquette, Q. Zeng, Tetrahedron Lett. 1999, 40, 3823.
doi:10.1016/S0040-4039(99)00631-0
[12] Atomic coordinates, bond lengths and angles, displacement
parameters, and other details associated with this analysis have
been deposited at the Cambridge Crystallographic Data Centre
(CCDC reference number 230238).
[13] V. Moliner, R. Castillo, V. S. Safont, M. Oliva, S. Bohn, I. Tunon,
J. Andres, J. Am. Chem. Soc. 1997, 119, 1941, and references
therein. doi:10.1021/JA962571Q
(b) M. G. Banwell, D. C. R. Hockless, M. D. McLeod, New J.
Chem. 2003, 27, 50. doi:10.1039/B206372G
(c) M. G. Banwell, A. J. Edwards, G. J. Harfoot, K. A. Jolliffe,
Tetrahedron 2004, 60, 535. doi:10.1016/J.TET.2003.10.122
[8] (a) M. G. Banwell, C. Chun, A. J. Edwards, M. M. Vögtle,
Aust. J. Chem. 2003, 56, 861. doi:10.1071/CH03112
(b) D. Appels, M. G. Banwell, A. J. Edwards, G. J. Harfoot,
D. C. R. Hockless, K. J. McRae, H. Schuhbauer, S. G. Stewart,
A. C. Willis, unpublished.
[14] J. P. Schaefer, J. J. Bloomfield, Org. React. 1967, 15, 1.