Synthesis of the C(1)-C(18) segment of lophotoxin and pukalide. Control of 2-alkenylfuran (E/Z)-configuration.

Article abstract of DOI:10.1021/ol025861m

[reaction: see text] The convergent synthesis of the fully functionalized C(1)-C(18) segment 24 of the furanocembranes lophotoxin and pukalide was accomplished in 11 steps and 10% overall yield. The key step was a stereoselective conversion of alkynoate 2

Full text of DOI:10.1021/ol025861m

ORGANIC
LETTERS
2002
Vol. 4, No. 10
1787-1790
Synthesis of the C(1)−C(18) Segment of
Lophotoxin and Pukalide. Control of
2-Alkenylfuran (E/Z)-Configuration
Peter Wipf* and Michael J. Soth
Department of Chemistry, UniVersity of Pittsburgh, Pittsburgh, PennsylVania 15260
pwipf+@pitt.edu
Received March 12, 2002
ABSTRACT
The convergent synthesis of the fully functionalized C(1)−C(18) segment 24 of the furanocembranes lophotoxin and pukalide was accomplished
in 11 steps and 10% overall yield. The key step was a stereoselective conversion of alkynoate 21 to trimethylsilyl 2-alkenylfuran 22.
The major biomedical interest in lophotoxin, pukalide, and
the closely related bipinnatins results from their selective
irreversible binding to nicotinic acetylcholine receptors.1
Lophotoxin is a member of the family of furanocembrane
natural products (Figure 1), and no total synthesis has been
reported to date.2 In related work, Paquette and co-workers
reported the synthesis of gorgiacerone and acerosolide, using
as the key cyclization step in both syntheses an allylchro-
mium attack on an aldehyde.3 Marshall and co-workers have
synthesized (-)-kallolide B by a diastereoselective [2,3]
Wittig rearrangement and, most recently, (-)-deoxypukalide
using as a key step an intraannular SiO2-mediated cyclization
of a 4-oxopropargylic â-keto ester.4
2-alkenylfurans.5 In our approach, R-propargyl â-keto esters
are cyclized to the desired 2-alkenylfurans, under either
palladium or mild base catalysis. A major advantage of this
approach is that the entire 2-alkenylfuran segment is as-
sembled in one step from readily available precursors; in
We recently reported a new method for synthesizing
(1) Abramson, S. N.; Fenical, W.; Taylor, P. Drug DeV. Res. 1991, 24,
297.
(2) For relevant synthetic approaches, see: (a) Cases, M.; Gonzalez-
Lopez de Turiso, F.; Pattenden, G. Synlett 2001, 1869. (b) Marshall, J. A.;
McNulty, L. M.; Zou, D. J. Org. Chem. 1999, 64, 5193. (c) Paterson, I.;
Brown, R. E.; Urch, C. J. Tetrahedron Lett. 1999, 40, 5807. (d) Astley, M.
P.; Pattenden, G. Synthesis 1992, 101. (e) Tius, M. A.; Trehan, S. J. Org.
Chem. 1986, 51, 765. (f) Paterson, I.; Gardner, M.; Banks, B. J. Tetrahedron
1989, 45, 5283.
(3) Rayner, C. M.; Astles, P. C.; Paquette, L. A. J. Am. Chem. Soc. 1992,
114, 3926.
(4) (a) Marshall, J. A.; Bartley, G. S.; Wallace, E. M. J. Org. Chem.
1996, 61, 5729. (b) Marshall, J. A.; Liao, J. J. Org. Chem. 1998, 63, 5962.
(c) Marshall, J. A.; Van Devender, E. A. J. Org. Chem. 2001, 66, 8037.
Figure 1. Structures of common furanocembranes.
10.1021/ol025861m CCC: $22.00 © 2002 American Chemical Society
Published on Web 04/24/2002

addition, the cyclization conditions should be compatible with
a variety of functional groups. However, this approach
suffered in that the cyclization reaction showed poor (E/Z)-
selectivity with regard to the alkene portion of the 2-al-
kenylfuran. In a relevant example in our original report, we
observed a 3:1 isomeric ratio for a 1,2-disubstituted alkene
(Scheme 1). It seems likely that the ratio would be worse
directing effect; the two relevant substituents are similar in
size. We reasoned that this issue could be resolved by
retrosynthetically substituting a trimethylsilyl group for one
of the substituents (Scheme 3). The bulky TMS group was
Scheme 3
Scheme 15
for the trisubstituted alkene products required for furano-
cembrane syntheses.
A possible solution to this problem arose from consider-
ation of the probable mechanism for 2-alkenylfuran formation
(Scheme 2). Both the palladium- and base-catalyzed reactions
expected to provide the desired selectivity in the 2-alkenyl-
furan-forming reaction and could subsequently be trans-
formed to the desired (methyl) group. Realization of this
strategy required an extension of our furan synthesis protocol5
to acylsilanes.
â-Keto ester 2 was easily obtained in one step from known
propargylic ester 17 by reaction with the anion of syn-
benzaldehyde oxime in a procedure slightly modified from
a report by Go´mez et al. (Scheme 4).8-10 Commercially
available acetyltrimethylsilane was reacted with the anion
of propargyl chloride, affording propargyl alcohol 3.11
Propargylic alcohol 3 was benzoylated using Vedej’s pro-
tocol,12 and the resulting benzoate 4 was used for alkylation
of the anion of â-keto ester 2, affording coupling product 5.
In this alkylation, it was important to first transform the
propargylic chloride to the corresponding iodide; attempts
to form the iodide in situ resulted in low yields.
Scheme 2
After much experimentation, we were able to efficiently
cyclize 2-alkenylfuran precursor 5 under palladium catalysis
in a heated acetonitrile/water solvent mixture.13 The silyl
2-alkenylfuran product 6 was formed as one predominant
alkene isomer (ca. 14:1). Furthermore, this isomer could be
almost completely isomerized to the alternative isomer (7,
>30:1) by reaction with a catalytic amount of diphenyl
likely afford an initial allene product, which then isomerizes
to a 2-alkenylfuran via a protonation-deprotonation se-
quence. The (E/Z)-ratio would accordingly depend on the
facial selectivity of the allene protonation step. It follows
that if one face of the allene was blocked by a bulky group
(R1 in Scheme 2), then one alkene isomer should dominate.6
This model predicts that the large group (R1) should end up
cis to the furan.
(7) Baxter, J.; Mata, E. G.; Thomas, E. J. Tetrahedron 1998, 54, 14359.
(8) Go´mez, V.; Pe´rez-Medrano, A.; Muchowski, J. M. J. Org. Chem.
1994, 59, 1219.
(9) â-Keto ester 2 has been reported,10 but the details on its preparation
are lacking. Reference 10b reports in a footnote that 2 can be prepared by
reaction of the dianion of methyl acetoacetate with [(p-methoxybenzyl)-
oxy]methyl chloride.
Unfortunately, in the furanocembrane targets of interest,
there is no sterically large group that could have the desired
(10) (a) Jones, A. B.; Yamaguchi, M.; Patten, A.; Danishefsky, S. J.;
Ragan, J. A.; Smith, D. B.; Schreiber, S. L. J. Org. Chem. 1989, 54, 17.
(b) Nakatsuka, M.; Ragan, J. A.; Sammakia, T.; Smith, D. B.; Uehling, D.
E.; Schreiber, S. L. J. Am. Chem. Soc. 1990, 112, 5583.
(11) We did not encounter significant problems with rearrangements in
our preparation of 3. However, both a Brook rearrangement and a [1,2]
TMS group shift have been reported for this same coupling under different
conditions. See: (a) Cunico, R. F.; Nair, S. K. Synth. Commun. 1996, 26,
803. For related studies, see: (b) Kuwajima, I.; Kato, M. Tetrahedron Lett.
1980, 21, 623. (c) Reich, H. J.; Eisenhart, E. K.; Olson, R. E.; Kelly, M. J.
J. Am. Chem. Soc. 1986, 108, 7791. (d) Bienz, S.; Enev, V.; Huber, P.
Tetrahedron Lett. 1994, 35, 1161.
(5) Wipf, P.; Rahman, L. T.; Rector, S. R. J. Org. Chem. 1998, 63, 7132.
(6) For related examples that exploit the axial asymmetry of allenes as
a stereocontrolling element, see: (a) Aso, M.; Ikeda, I.; Kawabe, T.; Shiro,
M.; Kanematsu, K. Tetrahedron Lett. 1992, 33, 5787. (b) Ikeda, I.; Gondo,
A.; Shiro, M.; Kanematsu, K. Heterocycles 1993, 36, 2669. (c) Ikeda, I.;
Honda, K.; Osawa, E.; Shiro, M.; Aso, M.; Kanematsu, K. J. Org. Chem.
1996, 61, 2031. (d) De Schrijver, J.; De Clercq, P. J. Tetrahedron Lett.
1993, 34, 4369. (e) Yang, F.-Y.; Cheng, C.-H. J. Am. Chem. Soc. 2001,
123, 761. (f) Trost, B. M.; Oi, S. J. Am. Chem. Soc. 2001, 123, 1230. (g)
Vanderwal, C. D.; Vosburg, D. A.; Sorensen, E. J. Org. Lett. 2001, 3, 4307.
For reviews on allenes, see: (h) Smadja, W. Chem. ReV. 1983, 83, 263. (i)
Pasto, D. J. Tetrahedron 1984, 40, 2805.
(12) Vedejs, E.; Daugulis, O. J. Org. Chem. 1996, 61, 5702.
1788
Org. Lett., Vol. 4, No. 10, 2002

Scheme 4a
Scheme 5a
a (a) syn-Benzaldehyde oxime, NaH, THF/HMPA; then 1 (75%);
(b) propargyl chloride, n-BuLi, Et2O, -78 °C, then TMSCOCH3
(94%); (c) Bz2O, TEA, MgBr2, CH2Cl2 (90%); (d) (i) NaI, acetone,
∆, (ii) 2/NaH, THF (79%, two steps); (e) K2CO3, Pd(OAc)2, dppf,
10:1 CH3CN/H2O, 80 °C (73%); (f) PhSeSePh, THF, ∆ (96%).
a (a) (i) TMS-dithiane, n-BuLi, THF, -20 °C, then 8, -20 °C;
(b) Fe(NO3)3‚9H2O, basic alumina, hexane, 43 °C (70%, two steps);
(c) propargyl chloride, n-BuLi, Et2O, -78 °C, then 10 (71%); (d)
Bz2O, TEA, MgBr2, CH2Cl2 (80%); (e) (i) NaI, acetone, ∆, (ii)
2/NaH, THF (87%, two steps); (f) K2CO3, Pd(OAc)2, dppf, 10:1
CH3CN/H2O, 84 °C (72%); (g) I2, AgClO4, pyridine, THF (88%);
(h) Me2Zn, Pd(PPh3)4, THF (98%).
diselenide in refluxing THF.14 The two isomers could be
1
unequivocally assigned as (Z)-6 and (E)-7 by H and 13C
1
NMR chemical shift values15 and H NMR NOE spectro-
scopic studies. The highly stereoselective formations of 6
and 7 provide a solid proof of concept for our mechanistic
hypothesis outlined in Scheme 2.
that has previously been used for dithiane but not for silyl
dithiane hydrolysis.19 Initial attempts using silica gel as the
support resulted in multiple products; however, a switch to
basic alumina as the support provided 10 in 70% overall
yield from 8. Acylsilane 10 was then converted to silyl
2-alkenylfuran 14 according to the reaction sequence used
for the preparation of furan 6. Reaction with lithiated
propargyl chloride afforded alcohol 11, which was benzo-
ylated to afford 12. This benzoate was used to alkylate the
anion of â-keto ester 2, and the resulting product 13 was
cyclized to give (Z)-2-alkenylfuran 14 in excellent selectivity
(ca. 15:1).20
We next turned to a more complex acylsilane precursor
containing a side chain relevant to our planned furanocem-
brane syntheses (Scheme 5). Trimethylsilyl dithiane was
alkylated with iodide 8.16 The hydrolysis of the resulting silyl
dithiane 9 to afford acylsilane 10 proved to be problematic.
Of the variety of known methods attempted for this
transformation,17,18 only the use of mercuric salt (HgClO4,
CaCO3, THF/H2O; 72% from iodide 8)17b worked well, but
we were reluctant to use this method on a large scale, as it
is both costly and leads to toxic side products. Finally, we
investigated the use of supported iron(III) nitrate, a reagent
For the conversion of the TMS group to the desired methyl
substituent, 14 was subjected to silane-iodine exchange.21
The resulting vinyl iodide 15 was reacted with dimethylzinc
under palladium catalysis to afford 2-alkenylfuran 16.22 NOE
(13) These conditions were first developed with ethyl acetoacetate as
the â-keto ester component. The corresponding 2-alkenylfuran product was
difficult to purify, however. Importantly, our originally reported conditions
employing THF as a solvent were completely unsuccessful, resulting in no
reaction. Use of ethanol at reflux led to product, but in low yields; reaction
in dimethylformamide (starting at room temperature and warming to ca.
85 °C) was successful (55%) but not always reproducible. Reaction in
acetonitrile at reflux was slow; but use of a hot 10:1 acetonitrile/water
mixture resulted in reproducible yields of ca. 55%.
(14) Ali, M. A.; Tsuda, Y. Chem. Pharm. Bull. 1992, 40, 2842.
(15) (a) Dorman, D. E.; Jautelat, M.; Roberts, J. D. J. Org. Chem. 1971,
36, 2757. (b) Chan, T. H.; Mychajlowskij, W.; Amouroux, R. Tetrahedron
Lett. 1977, 18, 1605.
(18) For alternative protocols for silyl dithiane hydrolysis, see: (a) Corey,
E. J.; Seebach, D.; Freedman, R. J. Am. Chem. Soc. 1967, 89, 434. (b)
Suda, K.; Watanabe, J.; Takanami, T. Tetrahedron Lett. 1992, 33, 1355.
(c) Chuang, T.-H.; Fang, J.-M.; Jiaang, W.-T.; Tsai, Y.-M. J. Org. Chem.
1996, 61, 1794. (d) Patroc´ınio, A. F.; Moran, P. J. S. J. Organomet. Chem.
2000, 603, 220.
(19) (a) Cornelis, A.; Laszlo, P. Synthesis 1985, 909. (b) Hirano, M.;
Ukawa, K.; Yakabe, S.; Morimoto, T. Synth. Commun. 1997, 1527. (c)
Hirano, M.; Ukawa, K.; Yakabe, S.; Clark, J. H.; Morimoto, T. Synthesis
1997, 858.
(20) (Z)-14 can be isomerized to the corresponding (E)-isomer, but this
process is much slower than the isomerization of (Z)-6. Treatment of (Z)-
14 with 10 equiv of diphenyl diselenide in tetrahydrofuran at reflux affords,
after 8 d, a 4.4:1 (E):(Z) ratio of alkenes.
(16) For a preparation of iodide 8, see: Paquette, L. A.; Doherty, A.
M.; Rayner, C. M. J. Am. Chem. Soc. 1992, 114, 3910.
(17) (a) Plantier-Royon, R.; Portella, C. Tetrahedron Lett. 1996, 37, 6113.
(b) Bouillon, J.; Portella, C. Eur. J. Org. Chem. 1999, 1571. (c) Saleur, D.;
Bouillon, J.; Portella, C. Tetrahedron Lett. 2000, 41, 321.
Org. Lett., Vol. 4, No. 10, 2002
1789

studies confirmed that 16 was the expected (E)-isomer;
however, a minor alkene isomerization during the silane-
iodine exchange step reduced the (E):(Z)-ratio to ca. 12:1.23
The preparation of the advanced C(1)-C(18) segment of
lophotoxin and pukalide was accomplished in 10% overall
yield by this strategy (Scheme 6). â-Keto ester 20 was
prepared from 1,4-butanediol, which was monoprotected and
then subjected to a combination Swern-Wittig reaction24 to
afford R,â-unsaturated ester 17 (Scheme 6). Reduction with
diisobutylaluminum hydride afforded alcohol 18, which was
subjected to a Johnson ortho ester-Claisen rearrangement25
to yield ester 19. After hydrolysis of 19, the resulting acid
was converted to â-keto ester 20 according to the method
of Li and Franck.26 Alkylation of the sodium enolate of 20
with the iodide of benzoate 12, and cyclization of â-keto
ester 21 to silyl 2-alkenylfuran 22, was followed by silane-
iodine exchange to afford vinyl iodide 23. Finally, reaction
of 23 with dimethylzinc under palladium catalysis afforded
(E)-24 in excellent yield and stereoselectivity (ca. 15:1).
In conclusion, we have accomplished an extension of our
2-alkenylfuran synthesis that employs allene stereochemistry
for control of alkene configuration. The face-selective
protonation of a silyl allene intermediate provides trisubsti-
tuted 2-alkenylfurans in good overall yield. Other noteworthy
features of our approach include the use of Al2O3-supported
iron(III) nitrate for hydrolysis of silyl dithioketals and the
stereoselective isomerization of vinylsilane (Z)-6 to (E)-7 in
the presence of catalytic diphenyl diselenide. The C(1)-
C(18) segment of lophotoxin and pukalide was thus prepared
in 11 steps and 10% yield.
Scheme 6a
a (a) (i) MEMCl, DIEA, CH2Cl2, (ii) (COCl)2, DMSO, CH2Cl2,
-78 °C; TEA, 0 °C; (carbethoxyethylidene)triphenylphosphorane,
rt (67%, two steps); (b) DIBALH, CH2Cl2, -78 °C (83%); (c)
triethylorthoacetate, EtCO2H, ∆ (81%); (d) LiOH‚H2O, 3:1 EtOH:
H2O; (e) (i) Meldrum’s acid, DCC, DMAP, CH2Cl2, (ii) MeOH, ∆
(58%, three steps); (f) NaH, THF, 0 °C, then add the iodide of 12;
(g) K2CO3, Pd(OAc)2, dppf, 10:1 CH3CN:H2O, 84 °C; (h) I2,
AgClO4, pyridine, THF (39%, three steps); (i) Me2Zn, Pd(PPh3)4,
THF (98%).
(21) For methods for iododesilylation of vinylsilanes, see: (a) Chan, T.
H.; Lau, P. W. K.; Mychajlowskij, W. Tetrahedron Lett. 1977, 18, 3317.
(b) Chan, T. H.; Koumaglo, K. Tetrahedron Lett. 1986, 27, 883. (c) Stamos,
D. P.; Taylor, A. G.; Kishi, Y. Tetrahedron Lett. 1996, 37, 8647. (d)
Alimardanov, A.; Negishi, E. Tetrahedron Lett. 1999, 40, 3839.
(22) Marshall, J. A.; Zou, D. Tetrahedron Lett. 2000, 41, 1347.
(23) Silyl 2-alkenylfurans 6 and 7 were also subjected to the same silane-
iodine exchange conditions used to prepare 15; these reactions proceeded
with retention of alkene configuration as determined by 1H and 13C chemical
shift and 1H NMR NOE spectroscopic studies on the vinyl iodide products.
In the case of (Z)-6, a significant amount of alkene isomerization was noted,
and the vinyl iodide product was isolated as a ca. 1:6 (E):(Z) mixture. In
the case of (E)-7, little to no isomerization occurred.
(24) Ireland, R. E.; Norbeck, D. W. J. Org. Chem. 1985, 50, 2198.
(25) Wipf, P. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Paquette, L. A., Eds.; Pergamon: Oxford, 1991; Vol. 5; pp
827-874.
(26) (a) Li, B. Q.; Franck, R. W. Bioorg. Med. Chem. Lett. 1999, 9,
2629. For a similar approach using an acid chloride, see: (b) Oikawa, Y.;
Sugano, K.; Yonemitsu, O. J. Org. Chem. 1978, 43, 2087.
Acknowledgment. This work has been supported by a
grant from the National Science Foundation (CHE-0078944).
M.J.S. gratefully acknowledges a National Institutes of
Health Postdoctoral Fellowship and thanks Todd Bosanac
for helpful discussions.
Supporting Information Available: Experimental pro-
cedures and spectral data for all new compounds, including
copies of H and 13C NMR for 5, 6, 7, 13, 14, 15, 16, 23,
and 24. This material is available free of charge via the
Internet at http://pubs.acs.org.
1
OL025861M
1790
Org. Lett., Vol. 4, No. 10, 2002