A new stereocontrolled synthetic route to (-)-echinosporin from d-glucose via Padwa allenylsulfone [3 + 2]-anionic cycloadditive elimination

Article abstract of DOI:10.1021/ol301090v

A new formal total synthesis of (-)-echinosporin has been developed based upon the Padwa [3 + 2]-cycloadditive elimination reaction of allenylsulfone 4 with the d-glucose-derived enone 14 which provides cycloadduct 12.

Full text of DOI:10.1021/ol301090v

ORGANIC
LETTERS
2012
Vol. 14, No. 12
3024–3027
A New Stereocontrolled Synthetic Route
to (ꢀ)-Echinosporin from D-Glucose via
Padwa Allenylsulfone [3 þ 2]-Anionic
Cycloadditive Elimination
Jakub T. Flasz and Karl J. Hale*
The School of Chemistry and Chemical Engineering, and the CCRCB, Queen’s
University Belfast, Stranmillis Road, Belfast BT9 5AG, Northern Ireland, U.K.
k.j.hale@qub.ac.uk
Received April 24, 2012
ABSTRACT
A new formal total synthesis of (ꢀ)-echinosporin has been developed based upon the Padwa [3 þ 2]-cycloadditive elimination reaction of
allenylsulfone 4 with the D-glucose-derived enone 14 which provides cycloadduct 12.
For some time now our group has been interested in the
application of novel, under-exploited cycloaddition reac-
tions in organic synthesis and, in this context, one process
that recently came to our attention is the Padwa, sodium
benzenesulfinate-induced, anionic [3 þ 2]-cycloadditive
elimination reaction of allenylsulfones with R,β-unsatu-
rated carbonyl compounds,1 a process that provides direct
synthetic access to nonsymmetric cyclopentenylsulfones
with complete regiocontrol. Despite the obvious synthetic
advantages that can potentially accrue from use of this
reaction, it has never been deployed successfully in any
complex natural product total synthesis, notwithstanding
the ease with which the resulting cyclopentenylsulfones can
be synthetically manipulated.
Partly out of a desire to apply the Padwa [3 þ 2]-
cycloadditive elimination upon chiral dipolarophile sys-
tems, and partly out of our own natural curiosity to learn
more about the scope and generality of this process, we set
out to examine whether it could be performed stereoselec-
tively on readily available, orthogonally protected
monosaccharide-enones and ene-lactones. If it could, our
aim was to use this chemistry in a new enantioselective
total synthesis of the naturally occurring antitumor agent,
(ꢀ)-echinosporin;2 a challenging, highly strained, acetal-
lactone of shikimate origin2d that has only succumbed to
total synthesis once in Amos B. Smith’s laboratory at Penn.3,4
Our initial efforts toward achieving this goal involved us
investigating the Padwa [3 þ 2]-cycloadditive elimination
of ene-lactone 3 with (phenylsulfonyl)-1,2-propadiene (4)5
to obtain the bicyclic adduct 6, which would then be taken
forward to (ꢀ)-echinosporin by a multistep synthetic
sequence (Scheme 1). However, to our great dismay, the
(2) Isolation: (a) Sato, T.; Kawamoto, I.; Oka, T.; Okachi, R.
J. Antibiot. 1982, 35, 266. (b) Hirayama, N.; Iida, T.; Shirahata, K.;
Ohashi, Y.; Sasada, Y. Bull. Chem. Soc. Jpn. 1983, 56, 287. Bioactivity:
(c) Morimoto, M.; Imai, R. J. Antibiot. 1985, 38, 490. Biosynthetic
studies on (ꢀ)-echinosporin: (d) Dubeler, A.; Krastel, P.; Floss, H. G.;
Zeeck, A. Eur. J. Org. Chem. 2002, 983.
(3) Total synthesis of (ꢀ)-echinosporin: (a) Smith, A. B., III;
Sulikowski, G. A.; Fujimoto, K. J. Am. Chem. Soc. 1989, 111, 8039.
(b) Smith, A. B., III; Sulikowski, G. A.; Sulikowski, M. M.; Fujimoto,
K. J. Am. Chem. Soc. 1992, 114, 2567.
(4) For various other synthetic approaches to echinosporin, see:
(a) Kinsella, M. A.; Kalish, V. J.; Weinreb, S. M. J. Org. Chem. 1990,
55, 105. (b) Forst, J. M. M.S. Thesis, University of Minnesota, 1987.
(c) Wincott, F. E. Ph.D Thesis, Yale University, 1989.
(1) (a) Padwa, A.; Yeske, P. E. J. Am. Chem. Soc. 1988, 110, 1617.
(b) Padwa, A.; Yeske, P. E. J. Org. Chem. 1991, 56, 6386. (c) Padwa, A.;
Watterson, S. H.; Ni, Z. J. Org. Chem. 1994, 59, 3256. (d) Nunez, A., Jr.;
Martin, M. R.; Fraile, A.; Garcia Romano, J. L. Chem.;Eur. J. 2010,
16, 5443.
(5) Stirling, C. J. M. J. Chem. Soc. 1964, 5856.
r
10.1021/ol301090v
Published on Web 05/30/2012
2012 American Chemical Society

Scheme 1. Our Initial Failed Attempt at Applying the Padwa
Allenylsulfone [3 þ 2]-Cycloadditive Elimination in the
Synthesis of (ꢀ)-Echinosporin
Scheme 2. Revised Retrosynthetic Plan for the Formal Total
Synthesis of (ꢀ)-Echinosporin
attempted cycloadditive elimination between 3 and 4 did
not produce 6. Instead, the O-desilylated ene-lactone 5 was
the only significant product formed in a meager 11% yield.
In light of this outcome, we modified our synthetic
approach to that shown in Scheme 2. In this, (ꢀ)-echinos-
porin would derive from Smith’s advanced synthetic inter-
mediate 73 which itself would be prepared from the
primary alcohol 8 and its progenitor 9, whose C(7)-
tertiary-hydroxyl we envisioned installing through the
dihydroxylation of 10 on its less hindered underside. The
centerpiece of our plan would be the alternate, regiochemi-
cally reversed Padwa [3 þ 2]-cycloadditive elimination1
between the more electrophilic pyranoside enone 146 and
allenylsulfone 45 which we believed would yield 12 which
itself would potentially be convertible into 10 via 11.
Enone 14 was secured (Scheme 3) by a modification of a
routeoriginallydevelopedby Card6 whichinitiallyentailed
us performing a Ferrier glycosidation on tri-O-acetyl-
D-glucal with ethanol and BF3 Et2O, O-deacetylating
3
the product with NaOMe/MeOH, and then selectively
O-silylating the 4,6-diol with TBSCl and imidazole in
DMF. A Swern oxidation on 15 thereafter secured 14 in
60% overall yield in four steps.
The key anionic Padwa [3 þ 2]-cycloadditiveelimination
was best conducted with a 4.5-fold excess of the enone 14,
relative to the allenylsulfone 4,5 and it was accomplished in
56% yield on a 60 g scale. The process was fully stereo-
controlled. It also typically allowed 81% of the theoreti-
cally isolable starting enone 14 to be recovered after SiO2
flash chromatography of the crude reaction mixture. A
chemo- and stereoselective reduction of the ketone in 12
ensued with Li-trisec-butylborohydride (L-Selectride) in
CH2Cl2. Here, the upward-pointing cyclopentenylsulfone
and the ꢀCH2OTBS group both guaranteed an exclusive
underside delivery of the hydride ion to the keto group to
provide 16following O-silylation. Potassium osmate there-
after induced underside syn-dihydroxylation to give the
R-hydroxy ketone as the sole product, and this was
O-triethylsilylated with TESOTf and 2,6-lutidine at
ꢀ30 °C for 25 min. We next examined a variety of methods
for the kinetic enolization of ketone 17 at C(7) including
(6) (a) For its previous use in stereoselective [4 þ 2]-cycloaddition,
see: Card, P. J. J. Org. Chem. 1982, 47, 2169. (b) For Trost’s seminal use
of 14 in a stereoselective [3 þ 2]-cycloaddition, see: Trost, B. M.; King,
S. A.; Schmidt, T. J. Am. Chem. Soc. 1989, 111, 5902.
Org. Lett., Vol. 14, No. 12, 2012
3025

Scheme 3. Synthesis of the β-Keto Ester 10
Scheme 4. Completion of Our Formal Synthesis of (ꢀ)-Echi-
nosporin
with LiHMDS and Mander’s reagent7 in THF at ꢀ78 °C,
and LDA/methoxycarbonyl imidazolide8 in THF at ꢀ78 °C
and then at rt, but we did not obtain a satisfactory outcome.
Eventually, after much experimentation, we discovered that
methyl pentafluorophenyl carbonate in the presence of
MgBr2 Et2O, Hunig’s base, and DMAP9 in CH2Cl2 could
3
effect the desired C-acylation efficiently. Significantly, this
new method for β-keto ester synthesis produced 10 in
92ꢀ95% yield.
The time had now come for C(7)-tertiary alcohol
installation, which was best accomplished through the
dihydroxylation of 10 with potassium osmate and NMO
(Scheme 4);it furnished18asa singleproduct in80% yield.
(7) Mander, L.; Sethu, S. P. Tetrahedron Lett. 1983, 24, 5425.
(8) Li, Y.; Hale, K. J. Org. Lett. 2007, 9, 1267.
(9) (a) Brooks, D. W.; Lu, L. D.; Masamune, S. Angew. Chem., Int.
Ed. Engl. 1979, 18, 72. (b) Lim, D.; Zhou, G.; Livanos, A. E.; Fang, F.;
Coltart, D. M. Synthesis 2008, 2148.
(10) Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin Trans.
1 1975, 1574.
3026
Org. Lett., Vol. 14, No. 12, 2012

The TES-group was next detached from 18 with PPTS/
MeOH, and a BartonꢀMcCombie10 deoxygenation was
performed. Ketone 9 was now subjected to Barton’s
excellent iodoolefination procedure,11 which entailed us
converting the ketone into a hydrazone and treating this
with iodine and triethylamine. Stille Pd(0)-induced reduc-
tive cross-coupling with Bu3SnH12 thereafter deiodinated
21 to give the desired cyclopentene in 72% overall yield. Its
primary TBS group was detached with PPTS/EtOH over
5 days to obtain 8 in 65ꢀ80% yield. Alcohol 8 was immedi-
ately oxidized to the acid 22 in a one-pot operation13 prior
to O-allylation. A carefully timed (only 5 min) E1cb
elimination of this ester followed, using KHMDS (only
1 equiv) at low temperature.
(ꢀ)-echinosporin formal total synthesis involved ammo-
nolysis and intersection with Smith’s penultimate inter-
mediate 7.3 The latter chemoselective deprotection was
best accomplished by reacting 25 with 50% aqueous
HBF4,15 a reaction which, for ease of final product pur-
ification, was best stopped at ∼50% conversion. By utiliz-
ing this method, hemiacetal 7 could be reproducibly
obtained pure in 37% yield, or 53% yield based upon a
30% recovery of the starting glycoside 25. Importantly,
our version of 7 matched Smith’s previously reported
NMR data perfectly (see Supporting Information), as
did its di-O-acetate derivative in CDCl3.3
In conclusion, with the new enantioselective route that
we have developed to (ꢀ)-echinosporin, we have demon-
strated the great worth of the Padwa allenylsulfone [3 þ 2]-
cycloadditive elimination reaction in complex natural
product total synthesis for the first time. We have also
uncovered a useful new protocol for the R-C-acylation
of bromomagnesium ketone enolates. Further details of
the wide scope of this new enolate C-acylation method
will be reported in due course, including with other
alkyl pentafluorophenyl carbonates, thiocarbonates, and
thionocarbonates.
O-Deallylation of 23 with phenylsilane and Pd(0)13
subsequently unmasked the acid, to allow formation of
the pentafluorophenyl ester 24 with EDC HCl (N-(3-
3
dimethylaminopropyl)-N0-ethylcarbodiimide hydrochlo-
ride) and pentafluorophenol. The final steps of our new
(11) (a) Barton, D. H. R.; O’Brien, R. E.; Sternhell, S. J. Chem. Soc.
1962, 470. (b) Barton, D. H. R.; Chen, M.; Jaszberenyi, J. C.; Taylor,
D. C.; Hartz, R. A.; Smith, A. B., III. Org. Synth. 1998, 74, 101.
(12) (a) Taniguchi, M.; Takeyama, Y.; Fugami, K.; Oshima, K.;
Utimoto, K. Bull. Chem. Soc. Jpn. 1991, 64, 2593. (b) Uenishi, J.i.;
Kawahama, R.; Yonemitsu, O.; Tsuji, J. J. Org. Chem. 1998, 63, 8965.
(13) For the one-pot Piancatelli/Pinnick oxidation, see: Huang, L.;
Teumelson, N.; Huang, X. Chem.;Eur. J. 2006, 12, 5246.
(14) Dessolin, M.; Guillerez, M.-G.; Thieriet, N.; Guibe, F.; Loffet,
A. Tetrahedron Lett. 1995, 36, 5741.
(15) When we attempted to convert the ethyl glycoside 25 into the
hemiacetal 7 using 6 M aqueous HCl in THF at rt for 28 days, the
reaction was unsuccessful. Instead, the corresponding R-hydroxy car-
boxylic acid 26 was isolated in 47% yield (see the SI for its NMR
spectra):
Acknowledgment. We thank Queen’s University Belfast
for a Ph.D studentship and financial support.
Supporting Information Available. Detailed experimen-
tal procedures and spectral data, as well as copies of the
1
400 MHz H and 100 MHz 13C NMR spectra (1D and
2D), IR, and HR mass spectra of all compounds in the
synthetic route, are provided. This material is available
free of charge via the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 12, 2012
3027