Cascade annulation reactions to access the structural cores of stereochemically unusual strychnos alkaloids

Article abstract of DOI:10.1002/ejoc.200900033

A new cascade annulation reaction has been developed to access the core structures of a novel family of strychnos alkaloids with a unique stereochemical arrangement. The new annulation cascade is facilitated by the development of a robust reaction sequenc

Full text of DOI:10.1002/ejoc.200900033

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200900033
Cascade Annulation Reactions To Access the Structural Cores of
Stereochemically Unusual Strychnos Alkaloids
Ricardo Delgado[a] and Simon B. Blakey*[a]
Keywords: Malagashanine / Strychnos / Alkaloids / Iminium / Cascade / Annulation / Reaction mechanisms / Polycycles /
Natural products
A new cascade annulation reaction has been developed to
access the core structures of a novel family of strychnos alka-
loids with a unique stereochemical arrangement. The new
annulation cascade is facilitated by the development of a ro-
bust reaction sequence to access extremely sensitive N-acyl-
iminium ions.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
Introduction
In 1946, following decades of careful research, Sir Robert
Robinson reported the structure of strychnine (1).[1] In the
60 years that have followed, numerous strychnos and
aspidosperma alkaloids have been isolated, all sharing a
common complex architecture and yet offering diverse bio-
logical activities. The dense functionalization and complex-
ity of these molecules has captured the imagination of gen-
erations of chemists and inspired the development of a daz-
zling array of synthetic methodologies.[2]
In considering these venerable families of natural prod-
ucts, we noted that the vast majority of the molecules
shared a common stereochemical pattern around the D
ring. In particular, the stereochemistry at C-2, C-3, C-7, C-
15 and C-16 is highly conserved,[3] and this sterochemical
pattern has been the focus of all the synthetic attention to
date (Figure 1). In 1995, Caira and Rasoanaivo reported
the structure of malagashanine (2), a unique strychnos al-
kaloid with inverted C-3 stereochemistry, isolated from the
Madagascan shrub Strychnos myrtoides.[4] Malagashanine
was isolated as part of a study to identify the bioactive com-
ponents of traditional herbal remedies for malaria treat-
ment in areas with high incidence of chloroquine (CQ) re-
sistant Plasmodium falciparum. Despite sharing many struc-
tural features with strychnine, malagashanine does not exhi-
bit significant toxicity. Furthermore, it has been shown to
restore the activity of CQ against drug-resistant popula-
tions of P. falciparun.[5] It is postulated that malagashanine
acts by inhibiting a P-glycoprotein efflux pump, similar to
those found in multi-drug-resistant cancer cells.[5d]
Figure 1. Malagashanine – a stereochemically unusual strychnos
alkaloid.
Since the discovery of malagashanine, Galeffi and co-
workers have identified several other strychnos alkaloids
bearing the unusual inverted C-3 stereochemistry [e.g. myr-
toidine (3)].[6] The intriguing biological activity exhibited by
malagashanine combined with its unprecedented stereo-
chemical motif has prompted us to develop a methodology
for the direct entry into this new class of strychnos alka-
loids. Specifically, we felt it was important to develop a sin-
gle methodology that would allow access to the novel mala-
gashanine core, but that would be flexible enough to facili-
tate the synthesis of all members of this new family, as well
as a diverse range of analogues to further explore the bio-
logical activity in the context of both CQ potentiation and
p-glycoprotein inhibition.
Cognizant of recent advances in Brønsted acid and hy-
drogen-bonding catalysis for the asymmetric addition of
nucleophiles to imines,[7] we envisioned a process, in which
imine 4 could be activated by either a chiral phosphoric
acid or by acylation in the presence of a chiral thiourea
[a] Department of Chemistry, Emory University,
1515 Dickey Drive, Atlanta, GA 30306, USA
Fax: +1-404-727-7586
E-mail: sblakey@emory.edu
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
1506
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 1506–1510

Access to Structural Core of Strychnos Alkaloids
catalyst, leading to cascade cyclization and formation of the
desired heterocyclic core of the S. myrtoides alkaloids (5)
(Scheme 1). Conceptually, this process fulfilled many of the
requirements we had outlined. However, in practice, the ex-
isting protic acid imine activation protocols were not com-
patible with this sensitive β,γ-unsaturated imine, and the
β,γ-unsaturated aldehyde 6 was extraordinarily sensitive to
isomerization under both acidic and basic conditions, ren-
dering imine 4 virtually impossible to access.[8]
Scheme 2. Synthesis of N-acyl-O-TMS-aminol 11 as an N-acylimi-
nium ion precursor.
ing groups on both the indole nitrogen atom and the imin-
ium ion nitrogen atom. Removal of the benzyl protecting
group from the indole resulted in almost exclusive forma-
tion of the undesired Pictet–Spengler product (Entry 2),
and use of electron-withdrawing groups on this nitrogen
atom led to substantially longer reaction times, competitive
iminium ion hydrolysis and a complex mixture of undesired
products (not shown).
Scheme 1. Retrosynthetic analysis of the malagashanine core. P =
protecting group.
Herein, we describe the evolution of a general strategy
for the construction of polycyclic indole alkaloids via β,γ-
unsaturated iminium ion intermediates that circumvents the
isomerization issues associated with the generation of these
species.
Table 1. Optimization of the cascade annulation process.
Results and Discussion
Having ruled out the condensation of aldehyde 6 with an
amine as a viable entry to the malagashanine alkaloids, we
adopted a strategy, in which an activated N-acyliminium
ion could be accessed from a stable amide precursor
(Scheme 2).[9] Thus, carboxylic acid 7 was synthesized by
nickel-catalyzed Kumada coupling of [(trimethylsilyl)-
methyl]magnesium chloride with diketene 8.[10] The requi-
site amide 9 was constructed by using standard conditions
and could be acylated in excellent yield with 0.95 equiv. of
LiHMDS and 1.4 equiv. of CbzCl. Under these conditions,
no olefin isomerization was observed. Reduction of amide
10 with DIBAL-H and in-situ trapping of the resulting alu-
minate complex with TMSOTf gave N-acyl-O-TMS-aminol
11 in good yield (82%). These N-acyl-O-TMS-aminol spe-
cies are generally stable to chromatography on silica gel and
can be stored for extended periods, making them conve-
nient precursors to reactive acyclic N-acyliminium ions.
With N-acyl-O-TMS-aminol 11 in hand, our attention
turned to the crucial cascade cyclization reaction. When
aminol 11 was treated with BF3·OEt2 in CH2Cl2 at –78 °C,
With respect to the iminium ion protecting group, steric
the desired tetracyclic product 12 was obtained in only factors did not significantly impact the product distribu-
modest yield (36%), with the majority of the iminium ion tion. Benzyl, tert-butyl and methyl carbamates all gave sim-
intermediate undergoing competitive Pictet–Spengler cycli- ilar results (Entries 1, 3 and 4). However, when the carb-
zation to give tetrahydro-β-carboline 13 (Table 1, En- amate was replaced with more electron-withdrawing amide
try 1).[11] In order to tune the reaction for the desired cas- or sulfamate auxiliaries, an improvement in the ratio of cas-
cade process, we studied the effects exerted by the protect- cade product/Pictet–Spengler rearrangement was observed
Eur. J. Org. Chem. 2009, 1506–1510
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1507

R. Delgado, S. B. Blakey
SHORT COMMUNICATION
(Entries 5 and 6). Under the optimal conditions, the N-to- resulted in only 23% yield of the desired product, with com-
syl-O-TMS-aminol is treated with BF3·OEt2 at 0 °C to give petitive Pictet–Spengler rearrangement dominating the re-
alkaloid 14 corresponding to the A,B,C,D-ring core of the action profile.[12]
malagashanine alkaloids in good yield (79%) with excellent
chemo- and stereoselectivity (Entry 7). The cascade product
is obtained as a single diastereomer, and only trace quanti-
Table 2. Cascade cyclizations of substituted indoles.[a]
ties of the product arising from Pictet–Spengler rearrange-
ment are observed.
In considering the mechanism of this reaction, we hy-
pothesize that both products arise from an initial C-3 attack
of the indole on (E)-iminium ion 15, establishing the rela-
tive stereochemistry between the spirocyclic quaternary cen-
ter and the adjacent amine. The resulting indolium ion 16
can subsequently be trapped by the pendant allylsilane
(Path A) leading to the desired cascade product 14, or a
competitive C-3–C-2 migration (Path B) will result in the
formation of the undesired Pictet–Spengler product 17. We
believe that strongly electron-withdrawing groups on the
iminium ion nitrogen atom destabilize the three-membered
transition state 18, leading to the Pictet–Spengler reaction,
and as such promote selective formation of the cascade
product 14 (Scheme 3).
[a] The relative stereochemistry of 22 was established by X-ray crys-
tallography (see Supporting Information for details). Other com-
pounds in this series were assigned by analogy and extensive NOE
studies.
This observation can be accounted for in our mechanistic
hypothesis, with the electron-donating 6-methoxy group
stabilizing the three-membered transition state leading to
the undesired rearrangement product. In order to overcome
this side reaction, we attempted to tune the electronics of
the substrate by placing an electron-withdrawing tosyl
group on the indole nitrogen atom [24, Equation (1)]. This
simple change resulted in a dramatic shift in the reaction
profile, with the initial electrophilic aromatic attack now
taking place at the 2-position of the indole and subsequent
intramolecular allylation of the C-3 cation resulting in the
formation of the regioisomeric cascade product 25 in 62%
yield [Equation (1)].
Scheme 3. Mechanistic rationale for the observed selectivity.
Having established conditions for the cascade cyclization
to give parent compound 14, we investigated the generality
of this reaction with respect to substitution of the indole
(1)
ring. Electron-withdrawing substituents in the 4-, 5- and 6-
positions were well tolerated (19–21, Table 2), as were elec-
In an extension of this methodology, we have demon-
tron-donating methoxy groups (22, 23). In most cases, the strated that the cascade annulation process can be adapted
use of a tosyl auxiliary on the iminium ion nitrogen atom for the cyclization of analogous oxocarbenium intermedi-
and a benzyl group on the indole nitrogen atom provided ates.[13–15] Tryptophol (26) was condensed with carboxylic
exquisite selectivity for the desired cascade product. How- acid 7 to give ester 27 (Scheme 4). Conversion to the mixed
ever, for the 6-methoxyindole substrate (particularly perti- acetal followed by BF3·OEt2-promoted cyclization provided
nent for the synthesis of myrtoidine), this auxiliary pattern tetrahydrofuran analogue 28 in good yield (64%).
1508
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 1506–1510

Access to Structural Core of Strychnos Alkaloids
group and the tosyl group can be removed by treatment of
14 with sodium in ammonia providing diamine 33 in 68%
yield.
Conclusions
We have developed an efficient cascade annulation strat-
egy to access the unusual structural core of the malagashan-
ine alkaloids. The annulation strategy has proved effective
in both intra- and intermolecular reactions. The use of N-
acyl- and N-tosylamides as iminium ion precursors circum-
vents problems associated with the condensation of primary
amines with sensitive aldehydes and exhibits significant
promise as an alternative entry into iminium ion chemistry.
The total synthesis of malagashanine and SAR studies to
further understand its mode of action are being actively
pursued in our laboratory.
Scheme 4. Oxonium ion cascade annulation reaction.
Furthermore, we have shown that intermolecular cascade
annulations are possible. N-Tosyl-O-TMS-aminol 29 was
synthesized by using our standard protocol.[12] When this
aminol was treated with BF3·OEt2 in the presence of
1.05 equiv. of N-benzylindole (30), annulation product 31
containing three contiguous stereocenters was obtained in
excellent yield (85%) as a single diastereomer (relative
stereochemistry determined by X-ray crystallography)
[Equation (2)].
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures, NMR spectroscopic data and X-ray
structures of key compounds.
Acknowledgments
We thank Syed Hussaini for preliminary experimental investi-
gations. We thank Kenneth Hardcastle, Rui Cao and Sheri Lense
for X-ray crystallography.
[1] a) R. Robinson, Experientia 1946, 2, 28; b) H. L. Holmes, H. T.
Openshaw, R. Robinson, Experientia 1946, 2, 908; c) H. T. Op-
enshaw, R. Robinson, Nature 1946, 157, 438.
(2)
[2] For reviews, see: a) J. Hájícek, Collect. Czech. Chem. Commun.
2004, 69, 1681; b) J. Bonjoch, D. Solé, Chem. Rev. 2000, 100,
3455; c) J. Bosch, J. Bonjoch, M. Amat, Alkaloids 1996, 48, 75.
[3] The biogenetic numbering system is used for the tetracyclic
core structure: J. Le Men, W. I. Taylor, Experientia 1965, 21,
508.
Despite the fact that N-acyliminium ions have been
shown to be excellent substrates for hydrogen-bond-cata-
lyzed Pictet–Spengler reactions,[16] our preliminary investi-
gations with known chiral thiourea and phosphoric acid
catalysts did not furnish the desired cascade annulation
product. The development of an appropriate chiral catalyst
for this new cascade process remains a research priority in
our laboratory.
Finally, we have demonstrated that both the tosyl and
benzyl groups used to tune the chemoselectivity of the reac-
tion are readily removed under standard conditions. Treat-
ment of tetracycle 14 with sodium naphthalide in DME at
–60 °C leads to clean removal of the tosyl group providing
monoamine 32 in 65% yield (Scheme 5). Both the N-benzyl
[4] M. R. Caira, P. Rasoanaivo, J. Chem. Crystallogr. 1995, 25,
725.
[5] a) P. Rasoanaivo, S. Ratsimamanga-Urvrg, R. Milijaona, H.
Rafatro, A. Rakoto-Ratsimamanga, C. Galeffi, M. Nicoletti,
Planta Med. 1994, 60, 13; b) D. Ramanitrahasimbola, S. Ratsi-
mamanga-Urverg, P. Rasoanaivo, A. Rakoto-Ratsimamanga,
Phytomedicine 1999, 6, 331; c) H. Rafatro, D. Ramanitrahasim-
bola, P. Rasoanaivo, S. Ratsimamanga-Urverg, A. Rakoto-Rat-
simamanga, F. Frappier, Biochem. Pharmacol. 2000, 59, 1053;
d) D. Ramanitrahasimbola, P. Rasoanaivo, S. Ratsimamanga,
H. Vial, Mol. Biochem. Parasitol. 2006, 146, 58.
[6] M.-T. Martin, P. Rasoanaivo, G. Palazzino, C. Galeffi, M. Nic-
oletti, F. Trigalo, F. Frappier, Phytochemistry 1999, 51, 479.
[7] For reviews, see: a) . G. Doyle, E. N. Jacobsen, Chem. Rev.
2007, 107, 5713; b) M. S. Taylor, E. N. Jacobsen, Angew. Chem.
Int. Ed. 2006, 45, 1520.
[8] For a related system, in which an α,α-disubstituted (i.e. non-
enolizable) aldehyde participates in a similar cascade process,
see F. He, Y. Bo, J. D. Altom, E. J. Corey, J. Am. Chem. Soc.
1999, 121, 6771.
[9] a) Y.-G. Suh, D.-Y. Shin, J.-K. Jung, S.-H. Kim, Chem. Com-
mun. 2002, 1064; b) J.-W. Jung, D.-Y. Shin, S.-Y. Seo, S.-H.
Kim, S.-M. Paek, J.-K. Jung, Y.-G. Suh, Tetrahedron Lett.
2005, 46, 573; c) for a review of N-acyliminium ion chemistry,
see: W. N. Speckamp, M. J. Moolenaar, Tetrahedron 2000, 56,
3817.
[10] K. Itoh, T. Yogo, Y. Ishii, Chem. Lett. 1977, 103. We note that
Scheme 5. Protecting group removal following cascade annulation.
it is important to purify the commercial diketene by sublima-
Eur. J. Org. Chem. 2009, 1506–1510
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1509

R. Delgado, S. B. Blakey
SHORT COMMUNICATION
tion to ensure reproducibly high yields in this cross-coupling
reaction.
[15]
S. E. Reisman, A. G. Doyle, E. N. Jacobsen, J. Am. Chem. Soc.
2008, 130, 7198.
[11] E. D. Cox, J. M. Cook, Chem. Rev. 1995, 95, 1797.
[12] See Supporting Information for details.
[13] a) D. J. Kopecky, S. D. Rychnovsky, J. Org. Chem. 2000, 65,
191; b) V. H. Dahanukar, S. D. Rychnovsky, J. Org. Chem.
1996, 61, 8317; c) D. J. Kopecky, S. D. Rychnovsky, Org. Synth.
2003, 80, 177.
[16] a) M. S. Taylor, E. N. Jacobsen, J. Am. Chem. Soc. 2004, 126,
10558; b) I. T. Raheem, P. S. Thiara, E. A. Peterson, E. N. Ja-
cobsen, J. Am. Chem. Soc. 2007, 129, 13404; c) J. Seayad, A. M.
Seayad, B. List, J. Am. Chem. Soc. 2006, 128, 1086.
Received: January 13, 2009
Published Online: March 4, 2009
[14] T. V. Lee, R. J. Boucher, C. J. M. Rockell, Tetrahedron Lett.
1988, 29, 689.
1510
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 1506–1510