A total synthesis of (±)-α-cyclopiazonic acid using a cationic cascade

Article abstract of DOI:10.1039/b417625c

The indolic terpene alkaloid α-cyclopiazonic acid 1 has been prepared in 11 steps from indole-4-methanol 6; the key step is a carbocationic cascade, terminated by a 4-nitrosulfonamide group and initiated by benzylic carbocation formation directly from the

Full text of DOI:10.1039/b417625c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A total synthesis of (¡)-a-cyclopiazonic acid using a cationic cascade{
Charlotte M. Haskins and David W. Knight*
Received (in Cambridge, UK) 19th November 2004, Accepted 27th January 2005
First published as an Advance Article on the web 18th May 2005
DOI: 10.1039/b417625c
cyclisation (Scheme 1).5,6 The key intermediate 3 could, we
The indolic terpene alkaloid a-cyclopiazonic acid 1 has been
reasoned, be generated rapidly by a cascade cyclisation, initiated
by formation of the benzylic carbocation 5. Concerns about its
eventual removal led us to substitute the original toluenesulfonyl
group, which was used throughout our initial studies to attenuate
the reactivity of the cascade terminator,7 with a nitrophenyl-
sulfonyl function, in the anticipation that this would be much more
readily cleaved when necessary.8 The stereochemical outcome of
the cascade cyclisation(s), if observed, was clearly also of concern.
Fortunately, Dreiding models showed both diastereomers of the
intermediates [cf. 5] to be rather crowded and strongly suggested
that access to a transition state conformation which would lead to
the desired cis-ring fusion would be much easier than that leading
to trans-ring fusion.
Our synthesis began with indole-4-methanol 4,9 which was
protected selectively at oxygen using TBDPSCl in THF containing
imidazole, then formylated under standard Vilsmeier conditions10
and further protected by tosylation of the indolic nitrogen. The
resulting aldehyde 7 was then homologated by a Horner–
Wadsworth–Emmons reaction,11 which gave an acceptable yield
of the unsaturated ester 8 (Scheme 2) which was then subjected to
a Michael addition of 2-methylpropenylmagnesium bromide in the
presence of phenylthiocopper.12 This gave the expected product in
an unoptimised isolated yield of 53%. Subsequent enolization,
specifically using KHMDS at low temperature, followed by brief
prepared in 11 steps from indole-4-methanol 6; the key step is a
carbocationic cascade, terminated by a 4-nitrosulfonamide
group and initiated by benzylic carbocation formation directly
from the intermediate 9, which gives the tetracyclic product 10
in 74% yield.
a-Cyclopiazonic acid (a-CPA) 1 is a major toxic principle of the
fungus Penicillin cyclopium Westling, which has a world wide
distribution and which is often found in stored grain and cereal.1
Its structure was elucidated some thirty five years ago;2 since then,
its biosynthesis has been the subject of many studies3 which have
shown that b-cyclopiazonic acid 2 is its immediate biological
precursor. More recently, it has enjoyed a heightened profile, by
reason of its ability to specifically inhibit the Ca2+-ATPase of the
sarcoplasmic reticulum, the very origin of its toxicity.4 This renders
it useful as a biological standard: in 2002, some 193 papers were
published which featured this application.
a-Cyclopiazonic acid 1 is distinguished by a pentacyclic array
containing a 3,4-disubstituted indole and a highly substituted
tetramic acid residue, together with a central cis-ring fusion. To
date, only two total syntheses of a-CPA 1 have been reported. In
the first, by the Kozikowski group,5 the central carbocyclic ring
was established using an intramolecular Michael addition of a
suitable 3,4-disubstituted indole. The pyrrolidine ring was then
elaborated followed by the tetramic acid motif from the
corresponding pyrrolidine-2-carboxylate and diketene. A final
base-catalysed epimerization led to the correct stereochemistry at
the C-9 (tetramic acid) stereogenic centre. A second synthesis by
Muratake and Natsume featured formation of
a suitable
4-substituted indole from N-Moc-pyrrole by the less common
tactic of constructing the benzene ring.6 The central ring system
was again generated using an intramolecular Michael addition,
followed by formation of a 2-methyl-l-pyrroline and the addition
of nucleophilic methyl to produce the gem-dimethyl feature.
The tetramic acid was again elaborated using diketene and the
stereochemistry adjusted by epimerisation. Both routes are
relatively brief (16 and 19 steps respectively), but both suffer from
a few rather inefficient steps.
Our idea was to synthesise the basic ring structure of a-CPA 1
using a cationic cascade cyclisation, terminated by a sulfonamide
function, which we have recently demonstrated to be an efficient
tactic for pyrrolidine synthesis.7 Both previous syntheses have
successfully constructed the tetramic acid residue from
a
pyrrolidine-2-carboxylate 3 and diketene 4, by a Dieckmann
{ Electronic supplementary information (ESI) available: experimental
details. See http://www.rsc.org/suppdata/cc/b4/b417625c/
*knightdw@cf.ac.uk
Scheme 1
3162 | Chem. Commun., 2005, 3162–3164
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Scheme 4
centre also had occurred during exposure to acid. While we have
previously observed such an equilibration to a more thermo-
dynamically stable isomer in simpler models,7 the ease with which
this occurred here was unexpected. We then benefited from a novel
observation: while removing the nosyl group using thioglycolate,8
the indolic tosyl function was also cleaved, leading directly to the
fully deprotected pyrrolidino-indole 11 in an excellent 82% yield.
The ratio of epimers remained essentially unchanged. We have
subsequently shown that this is a relatively general and convenient
method for the N-detosylation of indoles.15
Completion of the synthesis followed the chemistry described
above:5,6 treatment of the deprotected indole 11 with diketene and
potassium t-butoxide in dichloromethane led smoothly to a-CPA 1
(Scheme 4). The sample proved to be identical, except for its lack
of optical rotation, to an authentic sample (Tocris) according to its
m.p. of 238–242 uC [authentic: m.p. 244–245 uC (Tocris sample);
mixed m.p. 240–242 uC], 1H and 13C NMR data, mass spectra and
tlc mobility (EtOAc : petrol 3 : 2). A trace (ca. 5%) of an
epimer was detectable in our synthetic sample by 1H NMR
[dH 4.45, d, J 5 3.9 Hz] which was probably isomeric at C-9 (i.e.
the all-cis-isomer); a-CPA itself shows the C-9 proton at dH 4.00
(d, J 5 11.1 Hz).
Scheme 2
exposure to trisyl azide then engendered introduction of the
necessary nitrogen functionality, as the azide.13 Conversion into
the corresponding free amino-ester was then achieved under
standard conditions.14 Finally, this approach work was completed
by immediate N-nosylation8 to give the precursor 9 in 43% overall
and unoptimised yield for these last two steps, as a ca. 60 : 40
mixture of diastereoisomers.
We reasoned that the desired benzylic carbocation (cf. 5,
Scheme 1) might be generated directly from the O-silyl derivative 9
although, of course, the corresponding alcohol could still be an
intermediate. We were therefore delighted to find that, after some
optimization, compound 9 was converted cleanly into the
advanced tetracyclic precursor 10, in 74% isolated yield, upon
exposure to one equivalent of triflic acid at ambient temperature in
chloroform for 1 h (Scheme 3). Furthermore, our stereochemical
conjectures proved correct: the product 10 possessed entirely the
desired cis-ring fusion (J6a,9a 5 4.1 Hz) and was mostly the epimer
shown (J9,9a 5 9.5 Hz) along with a small amount (ca. 8%) of the
b-ethoxycarbonyl epimer (J9,9a 5 4.2 Hz).5,6 Although we had
anticipated obtaining the correct stereochemistry at C9 during
formation of the tetramic acid ring,5,6 evidently steric crowding
was sufficiently severe that almost complete epimerization at this
Despite some unoptimised and not especially efficient steps in
the approach work, the relatively spectacular yield of 74% from the
key cascade cyclisation step, together with the relative brevity of
this synthesis (14 steps from 2-methyl-3-nitrobenzoate) suggests
that this type of chemistry should find many other useful
applications.
We thank the EPSRC Mass Spectrometry Centre, University
College Swansea for the provision of high resolution mass
spectrometric data and the EPSRC for financial support and the
provision of high resolution NMR facilities.
Charlotte M. Haskins and David W. Knight*
School of Chemistry, Cardiff University, Main Building, Park Place,
Cardiff, UK CF10 3AT. E-mail: knightdw@cf.ac.uk;
Fax: +44(0) 2920 874030; Tel: +44(0) 2920 874210
Notes and references
1 B. J. Wilson, C. H. Wilson and A. W. Hayes, Nature, 1968, 220, 77;
J. Harrison, Top. Sci., 1971, 13, 57.
2 C. W. Holzapel, Tetrahedron, 1968, 24, 2101.
3 A. A. Chalmers, C. P. Gorst-Allman and P. S. Steyn, J. Chem. Soc.,
Chem. Commun., 1982, 1367; J. C. Schabort and D. J. J. Potgieter,
Biochem. Biophys. Acta, 1973, 309, 440; D. C. Neethling and
R. M. McGrath, Can. J. Microbiol., 1977, 23, 856.
4 N. W. Siedler, I. Jona, M. Vegh and A. Martonsi, J. Biol. Chem., 1989,
264, 17816.
5 A. P. Kozikowski, M. N. Grecco and J. P. Springer, J. Am. Chem. Soc.,
1984, 106, 6873.
Scheme 3
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 3162–3164 | 3163

View Article Online
and ester reduction using Red-Al1 [J. G. Cannon and B. J. Dempoulos,
J. Heterocycl. Chem., 1982, 19, 1195].
6 H. Muratake and M. Natsume, Heterocycles, 1985, 23, 1111.
7 C. M. Haskins and D. W. Knight, Chem. Commun., 2002, 249. For a
very similar approach see B. Schlummer and J. F. Hartwig, Org. Lett.,
2002, 4, 1471.
8 T. Fukuyama, C.-K. Jow and M. Cheung, Tetrahedron Lett., 1995, 36,
6373; T. Fukuyama, M. Cheung, C.-K. Jow and Y. Hidai, Tetrahedron
Lett., 1997, 38, 5831; W. Kunosawa, T. Kan and T. Fukuyama, J. Am.
Chem. Soc., 2003, 115, 8112 and references cited therein.
10 D. Peterson, M. Shaw and J. Locker, Tetrahedron Lett., 2000, 41, 6901.
11 Cf. S. Hubino, E. Sugion, T. Yamochi, M. Kuwaki and H. Hashimoto,
Chem. Pharm. Bull., 1987, 35, 2261.
12 Cf. M. Behforouz, T. T. Curran and J. L. Bolan, Tetrahedron Lett.,
1986, 27, 3107.
13 M. C. Evans and R. L. Johnson, Tetrahedron, 2000, 56, 9801.
14 Cf. L. He, H. S. Byun and R. Bittman, J. Org. Chem., 2000, 65,
7618.
9 Formed from methyl 2-methyl-3-nitrobenzoate by the Leimgruber–
Batchko method using Fe–HOAc for nitro group reduction
[G. S. Ponticello and J. J. Baldwin, J. Org. Chem., 1979, 44, 4003]
15 C. M. Haskins and D. W. Knight, Tetrahedron Lett., 2004, 45, 599.
3164 | Chem. Commun., 2005, 3162–3164
This journal is ß The Royal Society of Chemistry 2005