Enantioselective synthesis of the [6,6] spiroketal core of reveromycin A.

Article abstract of DOI:10.1021/ol991290v

[structure: see text] Reveromycin A (1) belongs to a family of microbial polyketides with unusual structural features and biological activities. The structure of 1 is composed of a [6,6] spiroketal core decorated with highly unsaturated side chains. As a

Full text of DOI:10.1021/ol991290v

ORGANIC
LETTERS
2000
Vol. 2, No. 2
207-210
Enantioselective Synthesis of the [6,6]
Spiroketal Core of Reveromycin A
Keith E. Drouet, Taotao Ling, Hung V. Tran, and Emmanuel A. Theodorakis*
Department of Chemistry and Biochemistry, UniVersity of California, San Diego,
9500 Gilman DriVe, La Jolla, California 92093-0358
etheodor@ucsd.edu
Received November 30, 1999
ABSTRACT
Reveromycin A (1) belongs to a family of microbial polyketides with unusual structural features and biological activities. The structure of 1
is composed of a [6,6] spiroketal core decorated with highly unsaturated side chains. As a prelude to the synthesis of 1, we present herein
a short, efficient, and enantioselective synthesis of the C9−C21 fragment 5 (spiroketal core) of reveromycin A.
Reveromycins A-D (1-4, Figure 1) constitute a novel class
of polyketide-type natural products that have recently been
isolated from Streptomyces sp. and display intriguing bio-
logical1 and structural features.2 From a biological standpoint,
these compounds exhibit strong antiproliferative activities,
which presumably derive from inhibition of the mitogenic
activity of the epidermal growth factor (EGF).3,4 Moreover,
reveromycins A, C, and D inhibit protein synthesis selectively
in eukaryotic cells and induce morphological reversion of
scrts-NRK cells.1
From a structural standpoint, the reveromycins are triacids
composed of a [6,6] or [5,6] spiroketal core bearing a
hemisuccinate ester, two highly unsaturated side chains, and
two alkyl groups. The combination of exquisite structure and
(1) (a) Takahashi, H.; Osada, H.; Koshino, H.; Sasaki, M.; Onose, R.;
Nakakoshi, M.; Yoshihama, M.; Isono, K. J. Antibiot. 1992, 45, 1414-
1419. (b) Takahashi, H.; Yamashita, Y.; Takaoka, H.; Nakamura, J.;
Yoshihama, M.; Osada, H. Oncology Res. 1997, 9, 7-11. (c) Osada, H.;
Koshino, H.; Isono, K.; Takahashi, H.; Kawanishi, G. J. Antibiot. 1991,
44, 259-261.
(2) (a) Takahashi, H.; Osada, H.; Koshino, H.; Kudo, T.; Amano, S.;
Shimizu, S.; Yoshihama, M.; Isono, K. J. Antibiot. 1992, 45, 1409-1413.
(b) Koshino, H.; Takahashi, H.; Osada, H.; Isono, K. J. Antibiot. 1992, 45,
1420-1427. (c) Ubukata, M.; Koshino, H.; Osada, H.; Isono, K. J. Chem.
Soc., Chem. Commun. 1994, 1877-1878.
(3) For recent reviews on EGF receptor, see: Growth Factors and
Receptors: a Practical Approach; McCay, I. A., Brown, K. D., Eds.;
University Press: New York, 1998. Growth Factors and Signal Transduc-
tion in DeVelopment; Nilsen-Hamilton, M., Ed.; Wiley-Liss: New York,
1994.
(4) For selected references on the importance of EGF receptor in
anticancer drug design, see: Stearns, M. E.; Stearns, M. Cancer Metastasis
ReV. 1993, 12, 39-52. Klijn, J. G. M.; Berns, P. M. J. J.; Schmitz, P. I.
M.; Foekens, J. A. Endocrine ReV. 1992, 13, 3-17. Modjtahedi, H.; Dean,
C. Intern. J. Oncol. 1994, 4, 277-296.
Figure 1. The reveromycin family of natural products.
10.1021/ol991290v CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/17/1999

biological activity displayed by these compounds has at-
tracted the interest of the synthetic community and has led
recently to the development of two independent total
syntheses of reveromycin B (4).5,6 Despite some reported
efforts, however, no synthesis of reveromycin A (1) has been
accomplished to date.7
Scheme 1. Synthesis of Aldehyde 8a
Structural inspection of reveromycin A (1) and B (4)
reveals that these molecules are constructed by a different
folding of an identical C1-C24 backbone. In the case of
reveromycin A, this folding creates a unique mosaic com-
posed of a [6,6] spiroketal core,8 in which the C18 tertiary
hydroxyl group and C20-C24 side chain are axially
oriented.2c The strain associated with such an arrangement
is at least partially alleviated in reveromycin B, in which
the C18 hydroxyl group is engaged in the spiroketal
formation. Respectful of the inherent instability of the [6,6]
core of 1,7 we set out to approach its synthesis keeping the
C18 hydroxyl group protected as a robust silyl ether.
Moreover, we projected that functionalization of the C20
carbon center with an acetylene unit could provide the
flexibility needed for further construction of the C20-C24
side chain. These considerations led us to define compound
5 as our synthetic target (Figure 2). Disassembly of the
Figure 2. Retrosynthetic analysis of the [C9-C21] spiroketal
fragment of reveromycin A (1).
spiroketal unit of 5 unveiled ketone 6 as a putative precursor.
In the synthetic direction, we envisioned that concurrent
release of the C11 and C19 hydroxyl groups would lead to
the desired spiroketal 5 without any interference from the
C18 silyl ether. Further disconnection across the C14-C15
bond revealed iodide 79 and aldehyde 8 as coupling partners,
the latter being synthetically accessible from compound 9.10
Our venture to bring this strategy to fruition is described
below.
The synthesis of fragment 8 proceeded as shown in
Scheme 1. Treatment of the readily available aldehyde 910
with n-BuLi at -78 °C afforded a 3:1 mixture of alcohols
at the newly installed C18 center in favor of compound 11
(85% combined yield). The observed stereochemical outcome
208
Org. Lett., Vol. 2, No. 2, 2000

of this reaction was consistent with addition of the nucleo-
phile according to the modified Felkin-Ahn model 10.11
Formation of the above diastereomeric mixture was incon-
sequential for our synthetic plan, since both compounds were
eventually converted to a single ketone, 13. Nonetheless, to
obtain analytical and spectroscopic data after each step, we
separated a small amount of these alcohols and carried out
the transformation to ketone 13 with the predominant
diastereomer.12
this stereochemistry had to be postponed until construction
of spiroketal 5, for which the structure was confirmed using
the transformations described in Scheme 3. Compound 20
was then treated with TBAF, and the resulting C20 alcohol
was oxidized to the aldehyde and transformed to dibromide
22 upon exposure to HMPT and CBr4 (three steps, 82%
overall).14 After silylation of the tertiary C18 hydroxyl group
(TBSOTf, lutidine), the geminal dibromoalkene functionality
was converted to alkyne 24 using the modified Corey-Fuchs
conditions (two steps, 88% overall).15 Finally, ozonolysis of
the terminal olefin of 24 gave rise to the desired aldehyde 8
in 85% yield.
In our initial trial we attempted to produce ketone 13 from
alcohol 12, the latter being easily obtained from 11 via a
sequence of three steps involving: oxidation of the C18
hydroxyl group to the corresponding ketone, acid-catalyzed
removal of the acetonide unit, and selective monoprotection
of the resulting diol at the primary C20 hydroxyl center (73%
overall yield). However, our efforts to protect the C19
hydroxyl group of 12 as a p-methoxybenzyl ether met with
failure, due to a concomitant scrambling and removal of the
primary silyl group that occurred under both acidic (PM-
BONHCCl3, CSA) and basic (PMBCl, NaH) treatment. To
overcome this problem, we pursued a synthetic maneuver,
which began with exposure of compound 11 to Dowex
50WX4-400 resin to afford triol 14 (81% yield). Treatment
of 14 with p-methoxybenzaldehyde dimethyl acetal under
acid catalysis produced the six-membered acetal 15 in 95%
yield, thereby rendering the C19 hydroxyl group available
for further functionalization. Treatment of 15 with p-
methoxybenzyl chloride, followed by removal of the acetal
unit under carefully controlled acidic conditions furnished
diol 17, through the intermediacy of compound 16 (two steps,
77% overall yield). Selective monoprotection of 17 at the
primary C20 carbon center, followed by oxidation of the C18
secondary alcohol, then gave rise to the desired ketone 13
(two steps, 76% combined yield). This maneuver allowed
for the smooth conversion of aldehyde 9 to ketone 13 with
a combined yield of about 35%.
Assembly of the spiroketal core 5 of reveromycin A
proceeded as described in Scheme 2. Lithiation of iodide 7
Scheme 2. Synthesis of Spiroketal Core Fragment 5a
The stage was now set for the installation of the C18
tertiary center. This was accomplished by reaction of 13 with
4-butenyllithium, affording a 8:1 mixture of separable
alcohols (88% combined yield). Formation of the major
diastereomer 20 was predicted to occur via a nonchelated
controlled nucleophilic attack, as shown in intermediate 19.13
Nevertheless, additional and unambiguous confirmation of
(5) Drouet, K. E.; Theodorakis, E. A. J. Am. Chem. Soc. 1999, 121, 456-
457.
(6) Masuda, T.; Osako, K.; Shimizu, T.; Nakata, T. Org. Lett. 1999, 1,
941-944.
(7) (a) Shimizu, T.; Kobayashi, R.; Osako, K.; Osada, H.; Nakata, T.
Tetrahedron Lett. 1996, 37, 6755-6758. (b) McRae, K. J.; Rizzacasa, M.
A. J. Org. Chem. 1997, 62, 1196-1197.
(8) For a recent review on spiroketals, see: Perron, F.; Albizati, K. F.
Chem. ReV. 1989, 89, 1617-1661.
(t-BuLi, -78 °C), followed by addition of aldehyde 8
(9) The synthesis of iodide 7 is presented in the Supporting Information.
(10) Andrews, G. C.; Crawford, T. C.; Bacon, B. E. J. Org. Chem. 1981,
46, 2976. Schmid, C. R,; Bradley, D. A. Synthesis 1992, 587-590.
(11) Heathcock, C. H.; Young, S. D.; Hagen, J. P.; Pirrung, M. C.; White,
C. T.; vanDerveer, D. J. Org. Chem. 1980, 45, 3846-3856. Nagano, H.;
Ohno, M.; Miyamae, Y.; Kuno, Y. Bull. Chem. Soc. Jpn. 1992, 65, 5, 2814-
2820. Nagano, H.; Ohno, M.; Miyamae, Y. Chem. Lett. 1990, 463-466.
Still, W. C.; McDonald, J. H. Tetrahedron Lett. 1980, 21, 1031-1034. Still,
W. C.; Schneider J. A. Tetrahedron Lett. 1980, 21, 1035-1038.
(12) All new compounds exhibited satisfactory spectral and analytical
data (see Supporting Information).
afforded a mixture of secondary alcohols at C15, which upon
(13) For selected reviews on this topic, see: Jurczak, J.; Pikul, S.; Bauer,
T. Tetrahedron 1986, 42, 447-488. Reetz, M. T. Acc. Chem. Res. 1993,
26, 462-468. Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1984, 23, 556-
569.
(14) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769-3772.
(15) Humphrey, J. M.; Eggen, J. B.; Chamberlin, A. R. J. Am. Chem.
Soc. 1996, 118, 11759-11770. Cliff, M. D.; Pyne, S. G. J. Org. Chem.
1997, 62, 1023-1032.
Org. Lett., Vol. 2, No. 2, 2000
209

oxidation with Dess-Martin periodinane16 furnished ketone
6 (two steps, 79% yield). Selective removal of the C11 and
C19 p-methoxybenzyl ethers was readily accomplished with
DDQ and produced spiroketal 25 as a mixture of diastere-
omers at the C15 center in 81% combined yield. Our attempts
to equilibrate the above mixture toward one of the diaster-
eomers using acid catalysis proved unsuccessful. Nonethe-
less, upon exposure to acid (CSA, MeOH) we observed
selective removal of the C9 silyl group resulting in a 1:1.5
mixture of spiroketals 26 and 5 (91% combined yield) that
was readily separated by column chromatography. The
stereochemistry at the C15 spiroketal center was assigned
on the basis of spectroscopic data. Notably, in spiroketal 26
we observed an NOE cross-peak between the H19 and H11
protons, while in spiroketal 5 such a cross-peak was not
detected. Moreover, in compound 26 the C19 hydrogen
appears at 5.06 ppm in accordance with the deshielding effect
of the axial oxygen, while in spiroketal 5 this hydrogen
appears at 4.53 ppm.17 For the same reason, in spiroketal 5
the C11 hydrogen is deshielded and resonates at 4.11 ppm,
while in compound 26 this hydrogen appears at 3.08 ppm
(Scheme 2). The remarkable difference in these chemical
shifts may also be attributed to the orientation of the C19
acetylene group.18 In addition, confirmation that compounds
26 and 5 were diastereomers at the C15 spiroketal center
was obtained by submitting spiroketal 26 to an acid-catalyzed
equilibration, which, as expected, gave rise to a new mixture
of spiroketals 26 and 5 in a 1.5:1 ratio and quantitative
yield.19 These results are in good agreement with computa-
tional studies that predict a difference of 0.25 kcal/mol
between the two spiroketals, in favor of 5.20
Although the above data confirmed the relative stereo-
chemistry at the C15 and C19 centers, the stereochemistry
at the C18 center merited further verification. To address
this issue, we sought to convert compound 25 to triacetate
29 and compare its data with those of the known material.5,7a
This conversion was executed as shown in Scheme 3.
Fluoride-induced removal of all silicon-based protecting
groups led to a concomitant trans-spiroketalization, giving
rise to spiroketal 27. The observed trans-spiroketalization
was evidenced by the downfield shifting of the C15 carbon
center (107.4 ppm in the 5,6 spiroketal system versus 95.9
and 97.2 ppm in the 6,6 spiroketal ring). LiAlH4-mediated
reduction of the terminal acetylene (assisted by the presence
of the C19 propargyl alcohol) produced olefin 28, which
Scheme 3. Conversion of 25 to Triacetate 29a
upon exposure to ozonolysis and reduction (NaBH4) fur-
nished the corresponding triol. Finally, acetylation using
acetic anhydride/pyridine gave rise to triacetate 29. As
expected, compound 29 exhibited spectroscopic and analyti-
cal data identical with those of the one previously described.5,7a
This observation confirmed unambiguously the previous
predictions for the formation of the C18 stereocenter and
further secured the stereochemical assignment for five out
of the seven stereocenters of the reveromycin A framework.
In conclusion, we have presented a concise, enantioselec-
tive approach to the [6,6] spiroketal core fragment of
reveromycin A (1). Fundamental to our strategy was the
coupling and subsequent spiroketalization of two suitably
functionalized components. Of particular interest is the
observation that the desired spiroketal 5 is formed as an
equilibrium mixture with its epimer 26. This result attests
to the strain of the core fragment of reveromycin A, in which
the C18 hydroxyl group and the C19 side chain reside in
axial orientations. Nonetheless, the desired spiroketal 5 can
be separated by chromatographic techniques and is amenable
to further manipulation. Extension of the above strategy to
the synthesis of reveromycin A (1) and related compounds
is currently underway in our laboratories.
Acknowledgment. Financial support from the Cancer
Research Coordinating Committee, the American Cancer
Society (RPG CDD-9922901), and the Hellman Foundation
is gratefully acknowledged.
(16) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277-
7287. Ireland, R. E.; Liu, L. J. Org. Chem. 1993, 58, 2899.
(17) Ireland, R. E.; Daub, J. P. J. Org. Chem. 1983, 48, 1303-1312.
Baker, R.; Herbert, R. H.; Patron, A. H. J. Chem. Soc., Chem. Commun.
1982, 601-604.
Supporting Information Available: Experimental pro-
cedures and spectral data (including 1H and 13C NMR
spectra) for compounds 5-8, 13, 25-27, and 29. This material
is available free of charge via the Internet at http://pubs.acs.org.
(18) We thank the referees for this comment.
(19) Deslongchamps, P.; Rowan, D. D.; Pothier, N.; Sauve, T.; Saun-
ders: J. K. Can. J. Chem. 1981, 59, 1105-1121. Evans, D. A.; Sacks, C.
E.; Kleschick, W. A.; Taber, T. R. J. Am. Chem. Soc. 1979, 101, 6789-
6791.
(20) Computational studies were performed using MM2 calculations with
CS Chem3D Pro. See Supporting Information for more details.
OL991290V
210
Org. Lett., Vol. 2, No. 2, 2000