Synthesis of the western half of the lolicines and lolitrems

Article abstract of DOI:10.1021/ol060338j

The synthesis of the highly substituted indole portion of the complex tremorgenic natural products lolicine A and B is presented. The Diels-Alder reaction of a quinone monoimine enables the synthesis of an appropriately substituted indole. The key step in

Full text of DOI:10.1021/ol060338j

ORGANIC
LETTERS
2006
Vol. 8, No. 11
2209-2212
Synthesis of the Western Half of the
Lolicines and Lolitrems
Dylan B. England, Jakob Magolan, and Michael A. Kerr*
Department of Chemistry, The UniVersity of Western Ontario,
London, Ontario, Canada N6A 5B7
makerr@uwo.ca
Received February 8, 2006
ABSTRACT
The synthesis of the highly substituted indole portion of the complex tremorgenic natural products lolicine A and B is presented. The Diels
−
Alder reaction of a quinone monoimine enables the synthesis of an appropriately substituted indole. The key step in the synthesis is a tandem
isopropenyl cuprate addition/aldol cyclocondensation which provides the necessary functionality for elaboration to the 2,2,5,5-tetramethyltet-
rahydrofuran.
Neotyphodium-type fungi contaminate many of the world’s
agriculturally important ryegrasses and are implicated in the
toxicity of such grasses.1 These toxic ryegrasses are indi-
genous to every continent with the exception of Antarctica
and contribute to a disorder known as perennial ryegrass
staggers, a neurotoxic affliction causing the grazing livestock
to exhibit tremors.2 The lolicines, lolitrems, and lolitriols
(isolated from Neotyphodium lolii)3 and the related penitrems
(from Penicillium cyclopium and Penicillium crustosum)4 are
known as tremorgenic indole terpenoids (Figure 1). Although
other tremorgenic indoles with similar “eastern halves” exist,
the lolicines, lolitrems, and penitrems are privileged in their
possession of a cyclohexane fused to the 4,5 position of the
indole moiety. A notable and unique feature of the lolicines
and lolitrems is the presence of a 2,2,5,5-tetramethyltetrahy-
(1) White, J. F., Jr. Plant Dis. 1987, 71, 340-342.
(2) Neill, J. C. N. Z. J. Sci. Technol. 1941, 23, 185A-193A.
(3) (a) Munday-Finch, S. C.; Wilkins, A. L.; Miles, C. O. J. Agric. Food
Chem. 1998, 46, 590-598. (b) Munday-Finch, S. C.; Wilkins, A. L.; Miles,
C. O. J. Agric. Food Chem. 1996, 44, 2782-2788.
Figure 1. Selected tremorgenic indole natural products.
(4) (a) Wilson, B. J.; Wilson, C. H.; Hayes, A. W. Nature 1968, 220,
77-78. (b) de Jesus, A. E.; Steyn, P. S.; van Heerden, F. R.; Vleggaar, R.;
Wessels, P. L.; Hull, W. E. J. Chem. Soc., Chem. Commun. 1981, 289-
291. (c) de Jesus, A. E.; Steyn, P. S.; van Heerden, F. R.; Vleggaar, R.;
Wessels, P. L.; Hull, W. E. J. Chem. Soc., Perkin Trans. 1 1983, 1847-
1856.
drofuran ring cis or trans fused (C31-C35) to the R,â-
carbons of a tetralone.
These tremorgens, aside from their biological interest,
represent new, complex architectures for assembly by total
10.1021/ol060338j CCC: $33.50
© 2006 American Chemical Society
Published on Web 04/25/2006

synthesis. Indeed, Smith and co-workers have been successful
in preparing the formidable penitrem D.5 Curran and co-
workers have also put forth an inventive approach to a
tricyclic substructure of the penitrems containing the cy-
clobutane moiety.6 There has only been one brief study
nominally directed toward the lolicine-type structure.7 Herein,
we report a successful construction of the western half of
the lolicines/lolitrems.
Scheme 2. Retrosynthesis for a Lolicine Western Half Model
Recently, we reported that highly substituted indoles (e.g.,
8-10) could be rapidly assembled by the oxidative cleavage
of amino dihydronaphthalenes8 4-6 which are prepared via
the Diels-Alder reaction of quinoid imines 1-3 (Scheme
1).9 This indole preparation strategy addresses the longstand-
Michael acceptor in 14 would be available via Horner-
Emmons elaboration of the aldehyde group present in 15.
The synthesis of indole 15 has been reported by us7a and
arises via the strategy in Scheme 1 from dienophile 17 and
trans-piperylene 16.
Scheme 1. General Synthesis of Complex Indoles
Our synthetic efforts commenced (Scheme 3) with the
Scheme 3. Elaboration of the Indole
ing challenge facing indole chemists, namely, a flexible
preparation of indoles bearing complex substitution on the
benzenoid portion of the molecule.
Scheme 2 shows a brief retrosynthesis of a western half
model of the lolicines. The 2,2,5,5-tetramethyltetrahydrofuran
of the target 11 would be prepared from 12 by acid-catalyzed
ring closure of a tertiary alcohol onto an isopropenyl moiety.
The tetralone system in 12 would arise from an aldol-type
ring closure via 13 which in turn would be prepared via
conjugate addition of an isopropenyl nucleophile to a
substrate such as 14. The required 5-formyl moiety in 14
would be installed via the aryl triflate in 15, whereas the
cycloaddition of quinone monoimine 17 with trans-pip-
erylene to yield the expected cycloadduct which was treated
with DBU to effect aromatization (vide supra, Scheme 1).
The phenolic moiety was triflated to yield 18 in 72% yield
over three steps from 17. Conversion of the dihydronaph-
thalene 18 to indole 15 was done in the usual manner in
74% overall yield by oxidative cleavage of the double bond
and treatment with acid. It was necessary to protect the
resulting aldehyde in 15 because it interfered (in an as yet
unknown way) with upcoming cross-coupling chemistry.
Although this could be done via ketalization, reduction and
acetate formation to give 19 were both more convenient and
higher yielding (92%). After exhaustive attempts at direct
formylation of 19, we decided to use a vinyl group as a latent
aldehyde. To this end, Stille coupling proceeded in excellent
yield (90%) to produce styrene 20. Upon reflection, this was
(5) (a) Kanoh, N.; Smith, A. B., III.; Ishiyama, H.; Minakawa, N.; Rainier,
J. D.; Hartz, R. A.; Cho, Y. S.; Cui, H.; Moser, W. H. J. Am. Chem. Soc.
2003, 125, 8228-8237. (b) Smith, A. B., III.; Kanoh, N.; Ishiyama, H.;
Hartz, R. A. J. Am. Chem. Soc. 2000, 122, 11254-11255. See also
references therein.
(6) (a) Rivkin, A.; Gonzalez-Lopez de Turiso, F.; Nagashima, T.; Curran,
D. P. J. Org. Chem. 2004, 69, 3719-3725. (b) Rivkin, A.; Nagashima, T.;
Curran, D. P. Org. Lett. 2003, 5, 419-422.
(7) Harrison, C. A.; Jackson, P. M.; Moody, C. J.; Williams, J. M. J. J.
Chem Soc., Perkin Trans. 1 1995, 1131-1136.
(8) The cleavage of dihydronaphthalenes (usually formed from a Birch
reduction) to indoles is an old but rarely used indole synthesis called a
Plieninger indolization. See: (a) Plieninger, H.; Voekl, A. Chem. Ber. 1976,
109, 2121-2125. (b) Plieninger, H.; Suhr, K.; Werst, G.; Kiefer, B. Chem.
Ber. 1956, 89, 270-278.
(9) (a) England, D. B.; Kerr, M. A. J. Org. Chem. 2005, 70, 6519-
6522. (b) Jackson, S. K.; Banfield, S. C.; Kerr, M. A. Org. Lett. 2005, 7,
1215-1218. (c) Zawada, P. V.; Banfield, S. C.; Kerr, M. A. Synlett 2003,
971-974. (d) Banfield, S. C.; England, D. B.; Kerr, M. A. Org. Lett. 2001,
3, 3325-3327.
2210
Org. Lett., Vol. 8, No. 11, 2006

a shorter route anyway because the aldehyde from formyl-
ation would have to be protected orthogonally from the
acetaldehyde moiety at the indole 4 position. Methanolysis
of the acetate in 20 and subsequent oxidation restored the
necessary aldehyde, giving 21 in 75% yield. Horner-
Emmons olefination of 21 smoothly produced enoate 22 in
87% yield. The requisite aldehyde at the indole 5 position
was revealed by oxidative cleavage of the double bond in a
two-step dihydroxylation/periodate cleavage protocol (71%
overall). With 23 in hand, the stage was set for an isopropenyl
cuprate addition and a subsequent aldol reaction.
Scheme 5. Successful Synthesis of the Western Half of
Lolicine A
Our initial experiments to install an isopropenyl moiety
via cuprate addition were unsuccessful and resulted in a
complex mixture, yielding 24 in small amounts (via nucleo-
philic addition to the aldehyde, Scheme 4). We then turned
Scheme 4. Attempted Ring Closures
benzylic hydroxyl group of the diol (en route to 28) adds in
a conjugate fashion to the enone. Careful control of the
reaction conditions avoided this as a major difficulty.11
Gratifyingly, cuprate addition to enone 28 not only installed
the requisite isopropenyl moiety but also resulted in the
addition of the intermediate enolate to the aldehyde in a
tandem fashion.12 The net result was the formation of 29
(58% overall as a 1:1 mixture, epimeric at the hydroxyl
group) with a benzofused ring and suitable functionality for
elaboration of the tetrahydrofuran ring. The 1:1 epimeric
mixture of benzylic alcohols was inconsequential because
that stereocenter would ultimately be a carbonyl group.
Protection of the benzylic hydroxyl group in 29 as a
triethylsilyl ether yielded 30 in 70% yield which in turn was
subjected to treatment with methyllithium, desilylation, and
oxidation with IBX to give hydroxyl ketone 31 in 65%
overall yield. All attempts to effect tetrahydrofuran formation
with protic or Lewis acids led to dehydration via elimination
of the hydroxyl group, sometimes with subsequent aroma-
to a strategy which would see the benzofused ring in 11
formed prior to the installation of the isopropenyl group,
ideally furnishing a compound such as 26 which would act
as a Michael acceptor for an isopropenyl cuprate. To this
end, 23 was treated with trimethylphosphine in acetonitrile
to effect a Morita-Baylis-Hillman-type ring closure.10 This
proceeded in excellent yield; however, subsequent dehydra-
tion could not be avoided and the aromatized compound 25
was formed as the sole identifiable product. Other Morita-
Baylis-Hillman conditions led to no appreciable reaction.
Next, we turned to a strategy which would see cyclization
via an extended enolate. Deprotonation of 23 with NaHMDS
led to the relatively clean formation of the same product 25.
At this stage, we reexamined the cuprate addition to a
substrate such as 27. It was felt that a better Michael acceptor
may avoid the domination of 1,2-addition. To this end,
aldehyde 21 was treated with a ketophosphonate to give
enone 27 in 88% yield (Scheme 5). Selective dihydroxylation
of the styrenyl moiety followed by oxidative cleavage gave
a 50% yield of the aldehyde 28. The modest yield of 28
arises from the formation of a side product where the
(11) The 50% yield obtained, although moderate, is an optimized and
reproducible yield and is a preferable alternative to a more circuitous
protection/deprotection route.
(12) This type of Michael-initiated ring-closure (MIRC) strategy for the
formation of carbocycles has been known for some time. For a seminal
contribution, see: Little, R. D.; Dawson, J. R. Tetrahedron Lett. 1980,
2609-2612.
(10) (a) Frank, S. A.; Mergott, D. J.; Roush, W. R. J. Am. Chem. Soc.
2002, 124, 2404-2405. (b) Keck, G. E.; Welch, D. S. Org. Lett. 2002, 4,
3687-3690.
Org. Lett., Vol. 8, No. 11, 2006
2211

tization of the tetralone. The required ring closure, however,
was effected with a selenoetherification protocol13 to give
selenide 32 in 80% yield which was reductively deselenated
with Raney nickel to produce the target molecule 11 in 80%
yield.
The fact that the trans stereochemistry in 11 was obtained
was confirmed by comparison to the NMR data for the
natural products. Both cis and trans relative stereochemistries
at C31-C35 are found in nature, and the vicinal coupling
constants between these two methines are, as expected, quite
different. Figure 2 shows representations of lolicine A 11-
coupling to the C35 methine proton (2.68 ppm). The C31-
C35 trans compound lolitrem N on the other hand has a
proton chemical shift at C31 of 3.35 ppm and has a much
smaller 7.3 Hz vicinal coupling to the C35 methine proton
(2.68 ppm). Our compound 11 has a chemical shift for the
C31 methine (lolicine numbering) of 2.72 ppm and has a
14.6 Hz vicinal coupling to the C35 methine proton (2.61-
2.56 ppm). These values are very close to those for lolicine
A 11-O-propionate, and therefore we are reasonably well
assured that we have a trans relative stereochemistry at what
would be the C31-C35 ring junction.
We were subsequently able to prepare crystals of 11 of a
sufficient quality for X-ray analysis to confirm the relative
trans stereochemistry of the C31-C35 ring fusion.14 Far from
being redundant, the X-ray analysis in concert with the NMR
analysis acts as fairly conclusive evidence that lolicine A
and related compounds have, in fact, the trans stereochem-
istry at C31-C35. To date, X-ray analysis of the natural
products has not been possible.
In conclusion, we have developed an efficient synthesis
of the western half of lolicine A and related compounds. A
Diels-Alder/Plieninger indolization allows for the prepara-
tion of a suitably substituted indole which in turn sets the
stage for the key tandem conjugate addition/aldol cyclization.
This step sets the relative stereochemistry at the tetrahydro-
furan in the trans relationship found in the natural product.
We intend to use this synthetic sequence in our efforts to
prepare lolicine A.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada, Medmira Labo-
ratories Inc., the Ontario Ministry of Science and Technol-
ogy, and GlaxoSmithKline for funding. Acknowledgment is
made to the Donors of the American Chemical Society
Petroleum Research Fund for partial support of this research.
We are grateful to Mr. Doug Hairsine for performing MS
analyses and Dr. Michael Jennings for X-ray analysis. D.B.E.
is an NSERC PGS B scholar.
Supporting Information Available: Full experimental
procedures and spectroscopic data for all new compounds
are available as well as the X-ray cif file. This material is
available free of charge via the Internet at http://pubs.acs.org.
Figure 2. X-ray structure of 11 and comparison of NMR data to
the natural products.
OL060338J
(13) Nicolaou, K. C.; Lysenko, Z. Tetrahedron Lett. 1977, 1257-1260.
(14) The representation of 11 shown in Figure 2 was generated from
the cif file directly using Mercury (version 1.4.1) software. The methyl
substituent on the p-toluene sulfonyl group shows disorder.
O-propionate and lolitrem N 10-O-propionate as well as of
our synthetic model 11. The proton chemical shift at the C31
methine for lolicine A is 2.77 ppm and has a 14.1 Hz
2212
Org. Lett., Vol. 8, No. 11, 2006