Total synthesis of the proposed structure of mycosporulone: Structural revision and an unexpected retro-aldol/aldol reaction

Article abstract of DOI:10.1021/ol302234a

The proposed structure of the fungal metabolite, mycosporulone 1, was prepared starting from the cyclohexenone ester 11 and the D-(R)-glyceraldehyde acetonide 12. The spectroscopic data for both 1 and its C2 epimer 1a did not match those reported for the natural product. A revised structure 29 for mycosporulone is proposed.

Full text of DOI:10.1021/ol302234a

ORGANIC
LETTERS
2012
Vol. 14, No. 18
4898–4901
Total Synthesis of the Proposed Structure
of Mycosporulone: Structural Revision
and an Unexpected Retro-Aldol/Aldol
Reaction
Michael E. Jung* and Jonah J. Chang
Department of Chemistry and Biochemistry, University of California, Los Angeles,
405 Hilgard Avenue, Los Angeles, California 90095, United States
jung@chem.ucla.edu
Received August 11, 2012
ABSTRACT
The proposed structure of the fungal metabolite, mycosporulone 1, was prepared starting from the cyclohexenone ester 11 and the
D-(R)-glyceraldehyde acetonide 12. The spectroscopic data for both 1 and its C2 epimer 1a did not match those reported for the natural
product. A revised structure 29 for mycosporulone is proposed.
Mycosporulone 1 (Figure 1), a polyketide-derived fun-
gal metabolite first reported in 1993,1a has shown activity
against several bacterial, fungal, and cancer strains while
exhibiting no toxicity toward human lung fibroblasts
(MRC5).1b It belongs to the 3-methylidene-2-oxaspiro-
[4.5]decan-1-one class of compounds which all possess
similar carbon connectivity and varying degrees of oxida-
tion (Figure 2).2 With the exception of an analogue synthe-
sis by Kraus and co-workers,3 no member of this class has
Figure 1. Structures of mycosporulone and its 2-epimer.
been synthesized; moreover, very little is known about
their biological activity. The structure of mycosporulone 1
1
was determined by H and 13C NMR and mass spectro-
analysis. Investigation of the reported data revealed an
NOE between the protons on C2 and C6, suggesting a cis
and not a trans relationship as depicted. According to
calculations, the cis diastereomer 1a would be the lowest
metric fragmentation patterns, with the relative stereo-
chemistry being assigned by NOE and Dreiding model
energy conformation by approximately 2.5 kcal/mol.4 Due
(1) (a) Kaouadji, M.; de Gusmao, N. B.; Steiman, R.; Seigle-
to its varied biological activity and questionable structural
Murandi, F. J. Nat. Prod. 1993, 56, 2189–2192. (b) Guiraud, P.; Steiman,
assignment, we therefore sought to develop a synthesis of
mycosporulone 1 that would prove its structure and that
also could be applied to its structural congeners.
The retrosynthetic analysis (Scheme 1) suggested that
the methylene tetrahydrofurandione moiety of 1 could be
prepared by dehydration of the lactone 8, which would
R.; Seigle-Murandi, F.; Buarque de Gusmao, N. J. Nat. Prod. 1999, 62,
1222–1224.
(2) (a) Ayer, W. A.; Craw, P. A.; Neary, J. Can. J. Chem. 1992, 70,
1338–1347. (b) Hirota, A.; Nakagawa, M.; Hirota, H. Agric. Biol. Chem.
1991, 55, 1187–1188. (c) Abdel-Wahab, M. A.; Asolkar, R. N.;
Inderbitzin, P.; Fenical, W. Phytochemistry 2007, 68, 1212–1218. (d) Sun,
Z.-L.; Zhang, M.; Zhang, J.-F.; Feng, J. Phytomedicine 2011, 18, 859–862.
(e) Albinati, A.; Bruckner, S.; Camarda, L.; Nasini, G. Tetrahedron 1980,
36, 117–21.
(3) For a synthesis of desmethyl mycosporulone, see: Kraus, A. G.;
Cui, W. Synlett 2003, 1, 95–96.
(4) Liu, P.; Houk, K. H. UCLA, private communication.
r
10.1021/ol302234a
Published on Web 09/06/2012
2012 American Chemical Society

This observed regioselectivity is predicated on the increased
basicity of the enone over the ester carbonyl.
Scheme 2. Preparation of Silyl Enol Ether 15
The silyloxy cyclohexadiene 15 was deprotonated with
LDA at ꢀ78 °C, trapped with D-(R)-glyceraldehyde aceto-
nide 12, and quenched at the same temperature to produce
a 1:1 mixture of the aldol diastereomers 16 and 17 (Scheme 3).
These diastereomers differ in the absolute stereochemistry
about the methyl and ethyl ester groups but have the
same alcohol configuration consistent with a FelkinꢀAhn
approach of the aldehyde trans to the methyl group.10
When the same mixture was allowed to warm to ꢀ40 °C
and then quenched, a different 1:1 mixture of products
was obtained, namely a yellow oil, 18, along with 17. An
X-ray crystal structure of 17 shows the stereochemistry11
(Figure 3). The aldol product 17 possesses the requisite
stereodiad for the synthesis of mycosporulone.
Figure 2. 3-Methylidene-2-oxaspiro[4.5]decan-1-one class.
arise from an acid-mediated acetonide hydrolysis/lactoni-
zation of the acyloin 9. We noted that these acidic condi-
tions might epimerize the acyloin functionality by the
Lobry de BruynꢀAlberda van Ekenstein reaction to give
two diastereomers.5 The acyloin group would be intro-
duced by a Rubottom oxidation of the silyl enol ether 10,6
which would result from an aldol reaction of the kinetic
silyl enol ether of the cyclohexadienone 11 and the en-
antiopure D-(R)-glyceraldehyde acetonide 12.
Scheme 3. Aldol Reaction of 15 and 12
Scheme 1. Mycosporulone Retrosynthetic Analysis
Condensation of ethyl 4-oxo-2-pentenoate 137 and pro-
pionaldehyde 14 employing pyrrolidine as an organo-
catalyst followed by standard DeanꢀStark dehydration
conditions8 afforded the enone 11 in 65% yield as a
mixture of diastereomers (Scheme 2). Soft enolization9 of
the enone 11 in the presence of TBSOTf gave the silyloxy
cyclohexadiene 15 in 83% yield as a mixture of diastereomers.
The configuration of the aldol product 18 was assigned
as follows. First, the lowest energy conformations of the
four possible lithium alkoxides were calculated (Scheme 4).12
The alkoxide 19 arises from FelkinꢀAhn addition of
the aldehyde trans to the methyl group of the (S)-enolate
eventually leading to the aldol product 17. The alkoxide 20
corresponds to the same aldehyde addition on the opposite
(R)-enolate leading to 16. The alkoxides 21 and 22 would
be generated from either FelkinꢀAhn or Cram-chelation
addition of the aldehyde cis to the methyl group on the (R)-
enolate, respectively. Second, the observed protonꢀproton
(5) For examples of the Lobry de BruynꢀAlberda van Ekenstein
reaction in synthesis, see: (a) Grieco, P. A.; Nargund, R. P.; Parker,
D. T. J. Am. Chem. Soc. 1989, 111, 6287–6294. (b) White, J. D.; Cutshall,
N. S.; Kim, T.-S.; Shin, H. J. Am. Chem. Soc. 1995, 117, 9780–9781.
(6) Rubottom, G. M.; Vazquez, M. A.; Pelegrina, D. R. Tetrahedron
Lett. 1975, 15, 4319–4322.
(7) McMurry, J. E.; Blaszczak, L. C. J. Org. Chem. 1974, 39, 2217–
2222.
(8) Wang, J.; Ma, A.; Ma, D. Org. Lett. 2008, 10, 5425–5428.
(9) Jung, M. E.; Ho, D. Org. Lett. 2007, 9, 375–378.
(10) For selectivity in enolate additions to D-(R)-glyceraldehyde
acetonide, see: Heathcock, C. H.; Young, S. D.; Hagan, J. P.; Pirrung,
M. C.; White, C. T.; VanDerveer, D. J. Org. Chem. 1980, 45, 3846–3856.
(11) Since the absolute stereochemistry of the acetonide carbon was
established from 12, one can assign the absolute stereochemistry of 17.
(12) Pham, H. V.; Houk, K. H. UCLA, private communication.
Org. Lett., Vol. 14, No. 18, 2012
4899

Scheme 5. NMR Coupling Constant Analysis of 18
stereochemical configuration of the aldol product 18 was
assigned based on this analysis.
Figure 3. X-ray crystallographic structure of 17.
When the racemic silyloxycyclohexadiene 15 was depro-
tonated, the two enantiomeric enolates 24 and 25 were
formed in solution (Scheme 6). Each one added to the
glyceraldehyde acetonide 12 at ꢀ78 °C to give the lithium
alkoxides 26 and 27, leading to the aldol products 17 and
16 upon quenching. The alkoxide 27 is 4.1 kcal/mol higher
in energy than the alkoxide 26 which can be attributed to a
syn-pentane-like interaction between the methyland aceto-
nide substituent in 27. Upon warming to ꢀ40 °C, this
interaction in the lithium alkoxide 27 is alleviated by
undergoing a retro-aldol/aldol sequence to give the alk-
oxide 28, which when quenched forms the aldol 18.14
Scheme 4. Calculated Lowest Energy Conformations
Scheme 6. Retro-Aldol/Aldol Sequence
coupling constants of the aldol product 18 (Scheme 5)
also are consistent with the assigned structure, i.e., a large
coupling (J = 12.0 Hz) for both HA and HB, implying an
anticoplanar orientation of the two protons. This is sup-
ported by the structure minimization, which indicates a
HAꢀOꢀCꢀHB dihedral angle of 159°. Furthermore, HC
exists as a doublet of doublets (dd) due to the coupling
between it and the protons on the adjacent methylene
group with no coupling observed with HB, implying that
HC and HB are nearly orthogonal. This is again borne out
by the structure minimization which shows that the
HBꢀCꢀCꢀHC dihedral is 62°.13 In contrast, proton HC
of the ester 23 now exists as a ddd, J = 6.8, 6.8, 2.1 Hz.
Several conditions were screened for the oxidation of the
aldol 18, but all resulted in unreacted starting material.
After the initial alcohol activation, the inability to remove
the proton HB, which is sterically hindered due to a pseudo
1,3-diaxial interaction with the methyl group, presumably
prevents the second E2 elimination step. Accordingly, the
(14) For examples of retro-aldol/aldol sequences, see: (a) Silverman,
R. B. J. Org. Chem. 1981, 46, 4789–4791. (b) Brocksom, T. J.; Coelho, F.;
Depres, J.-P.; Greene, A. E.; Freire de Lima, M. E.; Hamelin, O.;
ꢀ
Hartmann, B.; Kanazawa, A. M.; Wang, Y. J. Am. Chem. Soc. 2002,
124, 15313–15325. (c) Wang, J.; Cole, K. P.; Wei, L.-L.; Zehnder, L. R.;
Hsung, R. P. Tetrahedron Lett. 2002, 43, 3337–3340. (d) Xu, K.; Lalic,
G.; Sheehan, S. M.; Shair, M. D. Angew. Chem., Int. Ed. 2005, 44, 2259–
2261. (e) Akai, S.; Tsujino, T.; Fukuda, N.; Iio, K.; Takeda, Y.;
Kawaguchi, K.-I.; Naka, T.; Higuchi, K.; Akiyama, E.; Fujioka, H.;
Kita, Y. Chem.;Eur. J. 2005, 11, 6286–6297. (f) Flock, A. M.; Reucher,
C. M. M.; Bolm, C. Chem.;Eur. J. 2010, 16, 3918–3921.
(13) Although this dihedral angle is not 85°, the J for the two protons
is close to zero which may be due to the fact that the two heteroatoms on
the adjacent carbons change the Karplus equation values.
4900
Org. Lett., Vol. 14, No. 18, 2012

The required aldol product 17 was next oxidized with
DessꢀMartin periodinane to give the ketone 10 in 68%
yield (Scheme 7). The ketone 10 was then oxidized with
DMDO15 to give the acyloin 9 as a single diastereomer, the
structure of which was assigned by NMR, especially the
NOE between the methyl and the acyloin proton. The
acyloin 9 was treated with concentrated HCl in THF at
25°C followed bydehydration withthe Burgessreagent3 to
give the proposed structure of mycosporulone as a mixture
of the alcohol epimers 1 and 1a in 21% yield. The spectral
data of the mixture of 1 and 1a were not consistent with the
data reported for mycosporulone, strongly suggesting a
structural misassignment. We were able to access both
epimers 1 (OH trans to Me) and 1a (OH cis to Me) and can
conclude that neither has the required spectral data of
mycosporulone. It should be pointed out that the steps in
thesynthesisfromthe aldolproduct 17to the epimers 1 and
1a cannot epimerize the quaternary center; had that oc-
curred by some unknown route, one would have expected
to see four final products instead of two. Based on these
observations, we postulate that the quaternary center (C1)
is of the wrong configuration.
Scheme 8. Synthesis of 1 from 3 and Structural Reassignment
In summary, a total synthesis of the proposed structure
of mycosporulone 1 and its C2-epimer 1a has been com-
pleted. The synthetic material 1 does not correspond to the
reported data for mycosporulone, strongly suggesting that
there has been a structural misassignment. Based on the
current synthesis and the previously reported spectro-
scopic data, we propose the structural revision of mycos-
porulone to compound 29. While attempting to access 29
via oxidation of the known 6-epi-50-hydroxymycosporu-
lone 3, the previously synthesized mixture of 1 and 1a was
produced instead. Thus we have reason to believe that
6-epi-50-hydroxymycosporulone 3 has also been misas-
signed and now propose its structure to be that of com-
pound 30. Lastly, a retro-aldol/aldol reaction sequence
was observed, generating the aldol products 17and 18. The
retro-aldol reaction takes place to alleviate the syn-pentane-
like interactions found in the lithium alkoxide 27. Future
efforts will be focused on elaborating the aldol product 18
by an analogous synthesis to the revised structure of
mycosporulone 29.
Scheme 7. Completion of Synthesis of 1 and 1a
In an attempt to prove that the revised structure of
mycosporulone was indeed 29, we requested and received a
gift of the known natural product, 6-epi-50-hydroxymycos-
porulone 3. After screening several conditions (Scheme 8),
we found that oxidation with SeO2 gave a mixture of
epimers in 37% yield (59% brsm) that was spectroscopi-
cally consistent with the previously synthesized proposed
structure for mycosporulone 1. Consequently, we con-
cluded that 6-epi-50-hydroxymycosporulone 3 also has
the configuration in which the methyl group and the ester
of the lactone are cis to one another, leading us to believe
that it has likewise been misassigned. The revised structure
30 as drawn is consistent with the NOE data reported for
6-epi-50-hydroxymycosporulone 3.2c,16
Acknowledgment. We thank Professor William Fenical
(Scripps Oceanography) and the Compound Transfer
Program of Pfizer, Inc., Groton, CT for providing authen-
tic samples of 3. Dr. P. Liu and H. V. Pham of the Houk
Laboratories are gratefully acknowledged for assistance
with calculations. Dr. S. Khan is acknowledged for X-ray
crystallographic assistance. J.J.C. gratefully acknowledges
a UCLA Excellence in First Year Academics and Research
Award and an American Chemical Society Graduate
Research Symposium Award.
Supporting Information Available. Experimental pro-
cedures and proton and carbon NMR for all new com-
pounds. This material is free of charge via the Internet at
http://pubs.acs.org.
(15) Murray, R. W.; Singh, M. Org. Synth. 1997, 74, 91.
(16) (a) Fukami, A.; Taniguchi, Y.; Nakamura, T.; Rho, M.-C.;
Kawaguchi, K.; Hayashi, M.; Komiyama, K.; Omura, S. J. Antibiot.
1999, 52, 501–504. (b) Roll, D. M.; Tischler, M.; Williamson, R. T.;
Carter, G. T. J. Antibiot. 2002, 55, 520–523.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 18, 2012
4901