Short synthesis of the 6,6-spiroketal cores of spirofungins A and B

Article abstract of DOI:10.1021/ol0491078

(Matrix Presented) Initial efforts toward the total synthesis of the antifungal antibiotics spirofungins A and B are reported. A short and efficient synthesis of the C9-C20 6,6-spiroketal fragments of both compounds is described. This asymmetric approach

Full text of DOI:10.1021/ol0491078

ORGANIC
LETTERS
2004
Vol. 6, No. 15
2587-2590
Short Synthesis of the 6,6-Spiroketal
Cores of Spirofungins A and B†
Luiz C. Dias* and Luciana G. de Oliveira
Instituto de Qu´ımica, UniVersidade Estadual de Campinas, UNICAMP,
C.P. 6154 13084-971, Campinas, SP, Brazil
ldias@iqm.unicamp.br
Received May 14, 2004
ABSTRACT
Initial efforts toward the total synthesis of the antifungal antibiotics spirofungins A and B are reported. A short and efficient synthesis of the
C9−C20 6,6-spiroketal fragments of both compounds is described. This asymmetric approach uses a very efficient alkylation of a lithiated
N,N-dimethylhydrazone followed by spiroketal formation under acidic conditions.
Spirofungins A (1) and B (2) were isolated (as a ∼4:1
mixture, respectively) from the culture filtrate and extracts
of Streptomyces Violaceusniger Tu¨ 4113 as new polyketide-
spiroketal-type antibiotics related to reveromycins, antibiotics
produced by another Streptomyces strain (Figure 1).1,2
Spirofungins A and B showed high inhibition activity
against yeasts such as the human pathogen Candida albicans
and a moderate antifungal activity against filamentous fungi
such as Botrytis cinerea and Mucos miehei. Their mode of
antifungal action may be the same as for reveromycin, in
which activity is related to the inhibition of protein synthesis
in eukaryotic cells.
spirofungins A and B resemble those of reveromycin A (3),
except for the C18 position, since spirofungins lack the
Spirofungins A (1) and B (2) are polyketide-spiroketals
with seven stereogenic centers, two diene systems, and
carboxylic acid residues at both side chain termini. The
structure of spirofungins A and B (determined using the ∼4:1
mixture) was determined by NMR spectroscopic methods
in combination with HR-FAB-MS data. The structures of
† This paper is dedicated to Prof. Edmundo Alfredo Ruveda on the
occasion of his 70th birthday.
* Fax: +55-019-3788-3023.
(1) Ho¨ltzel, A.; Kempter, C.; Metzger, J. W.; Jung, G.; Groyh, I.; Fritz,
T.; Fiedler, H. P. J. Antibiot. 1998, 51, 699.
(2) (a) Takahashi, H.; Osada, H.; Koshino, H.; Kudo, T.; Amano, S.;
Shimizu, S.; Yoshihama, M.; Isono, K. J. Antibiot. 1992, 45, 1409. (b)
Takahashi, H.; Osada, H.; Koshino, H.; Sasaki, M.; Onose, R.; Nakakoshi,
M.; Yoshihama, M.; Isono, K. J. Antibiot. 1992, 45, 1414.
Figure 1. Spirofungins A and B and reveromycin A.
10.1021/ol0491078 CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/29/2004

Scheme 1. Retrosynthetic Analysis
Scheme 2. Synthesis of the C14-C20 Fragment
having the functional groups for the synthesis of both 1 and
2a, was based on the treatment of 5 under acidic conditions.
Ketone 5 may be further dissected in a straightforward
manner to give N,N-dimethylhydrazone 6 and primary alkyl
iodide 7. Of the available options, we speculated that the
desired syn stereocenters at C18 and C19 in 6 might be
established through a boron enolate-mediated aldol reaction
and that the anti stereocenters at C11 and C12 in 7 might
arise from a trans-epoxide opening reaction.6
succinate residue and have a methyl group instead of an
n-butyl group (Figure 1).1,2 By analogy to reveromycins, it
was assumed initially that the absolute configuration of the
spiroketal fragments of spirofungins A and B possessed
11R,12S,15S,18S,19R and 11R,12S,15S,18R,19S configura-
tions, respectively. The stereogenic centers at C4 and C5 in
both 1 and 2 were not determined initially but can be
proposed as having 4S,5S configurations in analogy to the
reveromycins.
Our approach began with an asymmetric aldol addition
of the boron enolate derived from oxazolidinone (R)-8 with
aldehyde 9 to give aldol adduct 10 in 94% yield and >95:5
diastereoselectivity (Scheme 2).7,8 Exchange of the oxazo-
lidinone auxiliary in the syn-aldol 10 with N,O-dimethylhy-
droxylamine9 followed by protection of the alcohol func-
tionality as its TBS ether cleanly provided the Weinreb amide
11, in 86% yield (over two steps). This amide was smoothly
reduced to the aldehyde on treatment with diisobutylalumi-
num hydride at 0 °C (Scheme 2).
The unpurified aldehyde was directly subjected to a
Horner-Wadsworth-Emmons homologation with the re-
quired phosphonate reagent to give the R,â-unsaturated
ketone 12 in 90% isolated yield (E:Z > 95:05) over the two-
step sequence.10 Selective hydrogenation11 of the double bond
proceeded smoothly, leaving the PMB group intact to give
the corresponding methyl ketone, which, after treatment with
N,N-dimethyl hydrazine in the presence of TMSCl as a
dehydrating agent, gave the corresponding hydrazone 13
(18S,19R) in 87% yield for the two-step sequence.12 Starting
from oxazolidinone (S)-8, we were able to prepare hydrazone
In very elegant work, Rizzacasa et al.3 recently proposed
a reassignment of the stereochemistry for spirofungin B with
the proposed corrected structure corresponding to 15-epi-
spirofungin A (structure 2a, 11R,12S,15R,18S,19R), having
a spiroketal with one less anomeric stabilization and not
epimeric at C18 and C19, as suggested earlier (Figure 1).1-3
We also were very intrigued by the fact that the initially
proposed structure for spirofungin B possesses different
absolute configurations at C18 and C19, when compared to
spirofungin A and the reveromycins. As the natural supply
is extremely restricted, and attracted by their promising
anticancer activity, we initiated a project directed toward the
total synthesis of spirofungins A and B.3,4
An efficient and flexible synthesis of their spiroketal parts
is essential to confirm the absolute configurations at C4, C5,
C18, and C19 as well as to provide further material for more
extensive biological studies, along with access to novel
analogues.5
Our disconnection strategy summarized in Scheme 1 (for
spirofungin A) involved cleavage of the C8-C9 as well as
the C20-C21 double bonds to give spiroketal 4.6 The latter
contains five of the seven stereogenic centers of the spiro-
fungins. Our synthetic strategy for the 6,6-spiroketal system,
(6) Numbering of 1, 2, and 2a as well as of each intermediate follows
that suggested in ref 1.
(7) Gage, J. R.; Evans, D. A. Org. Synth. 1990, 68, 83.
(8) (a) Evans, D. A.; Bartroli, J.; Shih, T. L. J. Am. Chem. Soc. 1981,
103, 2127. (b) Evans, D. A.; Taber, T. R. Tetrahedron Lett. 1980, 21, 4675.
(9) Levin, J. I.; Turos, E.; Weinreb, S. M. Synth. Commun. 1982, 12, 989.
(10) Maryanoff, B. E.; Reitz, A. B. Chem. ReV. 1989, 89, 863.
(11) Dias, L. C.; Campano, P. L. J. Braz. Chem. Soc. 1998, 9, 97.
(12) (a) Evans, D. A.; Bender, S. L.; Morrisy, J. J. Am. Chem. Soc. 1988,
110, 2506. (b) Evans, D. A.; Kaldor, S. W.; Jones, T. K.; Clardy, J.; Stout,
T. J. J. Org. Chem. 2001, 66, 7001. (c) Panek, J. S.; Jain, N. F. J. Am.
Chem. Soc. 1988, 110, 2747.
(3) Zanatta, S. D.; White, J. M.; Rizzacasa, M. A. Org. Lett. 2004, 6, 1041.
(4) Approaches to the spiroketal parts of spirofungins A and B: (a)
Shimizu, Y.; Kiyota, H.; Oritani, T. Tetrahedron Lett. 2000, 41, 3141. (b)
Shimizu, T.; Kusaka, J.; Ishiyama, H.; Nakata, T. Tetrahedron Lett. 2003,
44, 4965.
(5) For a review on spiroketals, see: Vaillancourt, V.; Pratt, N. E.; Perron,
F; Albizati, K. F. In The Total Synthesis of Natural Products; ApSimon, J.,
ed.; John Wiley & Sons: New York, 1992; Vol. 8, pp 533-691.
2588
Org. Lett., Vol. 6, No. 15, 2004

Scheme 3. Synthesis of Spiroketal 17
Scheme 5. Preparation of Primary Iodide 7
(Scheme 3). The 13C chemical shift of the spiro carbon C15
(95.7 ppm) was typical for a bis-axial orientation with two
anomeric stabilizations. In addition, the 13C chemical shift
of the methyl group at C18 (11.3 ppm) was appropriate for
an axial position, although the 13C chemical shift of the
methyl group at C12 (16.5 ppm) was more deshielded than
expected for a methyl group in an axial position. The axial
orientation for the methyl group was confirmed by NOESY
interactions and the observed coupling constants for Ha (3.93,
dd, J 11.1, 2.6 Hz) and Hf (3.28, brd, J 11.0 Hz).
14 (18R,19S), with the intention of preparing the initially
proposed spiroketal of spirofungin B (see Scheme 6).
To check the best conditions for carrying out the required
hydrazone coupling as well as the spiroketalization reactions
leading to analogues of spirofungins lacking the stereocenter
at C11, we first prepared the two enantiomeric primary alkyl
iodides (R)-15 and (S)-15 from commercially available
methyl (R)- and (S)-3-hydroxy-2-methylpropionate esters.13
The lithiated N,N-dimethylhydrazone derived from hydrazone
13 was used for a very efficient subunit coupling with alkyl
iodides (R)-15 and (S)-15 (Schemes 3 and 4).11 After exten-
sive experimental work, we found that the best conditions
for this coupling involved treatment of hydrazone 13 with
n-BuLi (1.1 equiv) in THF at -78 °C, to give a lithiated
hydrazone that participated in a smooth alkylation reaction
with primary iodide (R)-15. This reaction led to a coupled
product that was directly subjected to hydrazone hydrolysis
using silica gel in CH2Cl2 for 48 h to provide ketone 16 in
87% overall yield.
We next investigated the coupling of hydrazone 13 with
alkyl iodide (S)-15 (Scheme 4). Treatment of hydrazone 13
with n-BuLi (1.1 equiv) in THF at -78 °C, followed by
addition of the primary iodide (S)-15 and hydrazone hy-
drolysis provided ketone 18 in 87% overall yield (Scheme
4).11 Removal of the silicon protecting groups with HF-
pyridine followed by spiroketalization gave the desired
spiroketal 19 as the only isolated product, in 82% yield. The
relative stereochemistry for spiroketal 19 was confirmed by
the illustrated NOESY interactions as well as by NMR
analysis (Scheme 4). The 13C chemical shift of the spiro
carbon C15 (95.1 ppm) was typical for a bis-axial orientation
with two anomeric stabilizations. The 13C chemical shifts
of the methyl groups at C12 (17.3 ppm) and C18 (11.2 ppm)
were appropriate for equatorial and axial positions, respec-
tively. The equatorial orientation of the methyl group at C12
was confirmed by the illustrated NOESY interactions and
the observed coupling constants for Ha (t, J ) 11.0 Hz).
We have also observed that hydrogens Ha (3.93 ppm) and
Hb (3.80 ppm) in spiroketal 17 are more deshielded than
the same hydrogens, Ha (3.80 ppm) and Hb (3.35 ppm), in
spiroketal 19, which further supports the assignment of the
methyl group in an axial position at C12 in spiroketal 17.
Having established the best conditions for hydrazone
coupling and spiroketal formation, we next moved to the
preparation of alkyl iodide 7 (Scheme 5).
Scheme 4. Synthesis of Spiroketal 19
This was accomplished by epoxidation of allylic alcohol
20 using the Sharpless asymmetric procedure15 with D-(-)-
DET, Ti(OiPr)4 and tBuOOH to produce the epoxy alcohol.16
Epoxide opening proceeded smoothly with high regioselec-
(13) Protection of the OH function, DIBALH reduction to the alcohol,
followed by treatment with PPh3 and I2 to give primary iodides 15 in 80%
yield for the three-step sequence.
(14) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis
1994, 639. (b) Bloch, R.; Brillet, C. Synlett 1991, 829.
(15) Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune, H.;
Sharpless, K. B. J. Am. Chem. Soc. 1987, 109, 5765.
Removal of the TBS protecting groups with HF-pyridine
followed by spirocyclization gave the desired spiroketal 17
as the only isolated product, in 78% yield.14 The relative
stereochemistry for spiroketal 17 was confirmed by the
illustrated NOESY interactions as well as by NMR analysis
Org. Lett., Vol. 6, No. 15, 2004
2589

Scheme 6. Synthesis of Spiroketal 24
Scheme 7. Equilibrium between Spiroketals 4 and 4a
tivity after treatment of the epoxy alcohol with Me2CuCNLi2
to give diol 21 with good yield and selectivity.17 Selective
protection of the primary OH function as its TBS ether
followed by protection of the secondary alcohol functionality
as the TIPS ether gave 22 in 86% overall yield (two steps).
Selective removal of the primary TBS group with HF-pyr
in pyridine followed by treatment with PPh3 and I2 gave pri-
mary alkyl iodide 7 in 80% yield for the two-step sequence.
With fragments C11-C13 (7) and C14-C20 (13 and 14)
in hand, their coupling was undertaken (Schemes 6 and 7).
To prepare the initially proposed structure of spirofungin B,
treatment of hydrazone 14 with n-BuLi (1.1 equiv) in THF
at -78 °C, followed by addition of primary alkyl iodide 7
and hydrolysis gave ketone 23 in 87% overall yield (Scheme
6).11 Compound 23 was subjected to acid deprotection and
spiroketal formation by treatment with HF-pyr in THF to
give 24 as the only spiroisomer in 78% yield.
We next moved to the preparation of the desired spiroket-
als of spirofungins A and B (Scheme 7). Treatment of
hydrazone 13 with n-BuLi (1.1 equiv) in THF at -78 °C,
followed by addition of primary alkyl iodide 7, gave an
intermediate hydrazone that was directly subjected to hy-
drolysis without further purification, providing the ketone
25 in 87% overall yield (Scheme 7).11 At this point, all that
remained was to carry out the necessary spiroketalization.
Compound 25 was subjected to acid deprotection and
spiroketal formation by treatment with HF-pyr in THF to
give a 30:70 mixture of spiroketals 4 and 4a, respectively,
in 84% combined yield.
the 13C chemical shift of the spiro carbon C15 (96.6 ppm)
was typical for a bis-axial orientation with two anomeric
stabilizations. In addition, the 13C chemical shifts of the
methyl groups at C12 (13.4 ppm) and C18 (17.9 ppm) were
typical for equatorial positions.
For the major spiroketal 4a (needed for the synthesis of
the corrected structure of spirofungin B), the 13C chemical
shift of the spiro carbon C15 (97.5 ppm) was typical for an
orientation with only one anomeric stabilization. The 13C
chemical shift of the methyl group at C12 (17.4 ppm) was
that expected for an equatorial position, and the 13C chemical
shift of the methyl group at C18 (11.0 ppm) was typical for
an axial position. This equilibrium between 4 and 4a is in
perfect agreement with the fact that spirofungins A and B
were not separated and the structure elucidation was con-
ducted on the 4:1 mixture. On the basis of these results, we
support the assignment of the Rizzacasa group3 that the
spiroketal cores of spirofungins A and B have the same
absolute configurations at C18 and C19 and that spirofungin
B is epimeric at the C15 spiro carbon, being a spiroketal
with one less anomeric stabilization.
Our synthesis requires nine steps from (R)-8 (longest linear
sequence) and produced the 6,6-spiroketal 4 and 4a in good
yields. This approach is, in principle, readily applicable for
the preparation of spirofungins A and B as well as to
additional analogues. Extension of this work to the synthesis
of spirofungins A and B is underway, and the results will
be described in due course.
It is noteworthy that under these conditions, spiroketal 4a,
with only one anomeric stabilization, was isolated as the
major isomer. Although these spiroketals were readily
separated by flash column chromatography, we observed that
each isolated pure spiroketal led to the same 30:70 equilib-
rium mixture of 4:4a during attempts to obtain the NMR
spectra in CDCl3. The relative stereochemistries for spiroket-
als 4 and 4a were confirmed by NMR analysis (Scheme 7).
For spiroketal 4 (needed for the synthesis of spirofungin A),
Acknowledgment. We are grateful to FAEP-UNICAMP,
FAPESP (Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o
Paulo), and CNPq (Conselho Nacional de Desenvolvimento
Cient´ıfico e Tecnolo´gico) for financial support. We thank
also Prof. Carol H. Collins for helpful suggestions about
English grammar and style.
Supporting Information Available: Spectral data for
compounds 4, 4a, 7, 10-13, and 16-25. This material is
available free of charge via the Internet at http://pubs.acs.org.
(16) D´ıez-Martin, D.; Kotecha, N. R.; Ley, S. V.; Mnategani, S.;
Mene´ndez, H. M.; White, A. D.; Banks, B. J. Tetrahedron 1992, 48, 7899.
(17) Relative stereochemistry for diol 21 was determined by conversion
to the PMB acetal followed by analysis of coupling constants and NOESY
1
interactions in the corresponding H NMR spectra.
OL0491078
2590
Org. Lett., Vol. 6, No. 15, 2004