Cycloaddition protocol for the assembly of the hexacyclic framework associated with the kopsifoline alkaloids

Article abstract of DOI:10.1021/ol062029z

(Chemical Equation Presented) An approach to the hexacyclic framework of the kopsifoline alkaloids has been developed and is based on a Rh(II)-catalyzed cyclization-cycloaddition cascade. The resulting [3+2]-cycloadduct was readily converted into the TBS

Full text of DOI:10.1021/ol062029z

ORGANIC
LETTERS
2006
Vol. 8, No. 22
5141-5144
Cycloaddition Protocol for the Assembly
of the Hexacyclic Framework Associated
with the Kopsifoline Alkaloids
Xuechuan Hong, Stefan France, Jose´ M. Mej´ıa-Oneto, and Albert Padwa*
Department of Chemistry, Emory UniVersity, Atlanta, Georgia 30322
chemap@emory.edu
Received August 17, 2006
ABSTRACT
An approach to the hexacyclic framework of the kopsifoline alkaloids has been developed and is based on a Rh(II)-catalyzed cyclization
−
cycloaddition cascade. The resulting [3 2]-cycloadduct was readily converted into the TBS enol ether 23. Oxidation of the primary alcohol
+
present in 23 followed by reaction with CsF afforded compound 24 that contains the complete hexacyclic skeleton of the kopsifolines.
The Malaysian members of the genus Kopsia have yielded
a prodigious harvest of new natural products possessing novel
Scheme 1
carbon skeletons as well as useful bioactivities.1 In 2003,
Kam and Choo reported the structures of several new
hexacyclic monoterpenoid indoline alkaloids isolated from
a Malaysian Kopsia species known as K. fruticosa.2 The
major components of the alkaloidal extracts were identified
as the kopsifolines.3 The structures were secured by spectral
analysis and were characterized by an unprecedented carbon
skeleton, in which the C(18) carbon is linked to C(16). The
kopsifolines are structurally intriguing compounds, related
The interesting biological activities of these compounds
combined with their fascinating and synthetically challenging
structure make them attractive targets for synthesis.
Our own work on the synthesis of the kopsifolines
represents one component of a broad program focused on
the synthesis of various aspidosperma alkaloids using [3+2]-
cycloaddition chemistry.4 Our retrosynthetic analysis of 3 is
shown in Scheme 2 and envisions the core skeleton to arise
to and possibly derived from an aspidosperma-type alkaloid
precursor 1. A possible biogenetic pathway to the kopsifo-
lines from 1 could involve an intramolecular epoxide-ring
opening followed by loss of H2O as shown in Scheme 1.
(1) (a) Kam, T. S. In Alkaloids; Chemical and Biological PerspectiVes;
Pelletier, S. W., Ed.; Elsevier: Oxford, 1999; Vol. 14, pp 285-435. (b)
Kam, T. S.; Sim, K. M.; Koyano, T.; Komiyama, K. Phytochemistry 1998,
50, 75. (c) Kam, T. S.; Yoganathan, K.; Li, H. Y.; Harada, N. Tetrahedron
1997, 53, 12661. (d) Kam, T. S.; Yoganathan, K.; Koyano, T.; Komiyama,
K. Tetrahedron Lett. 1996, 37, 5765. (e) Uzir, S.; Mustapha, A. M.; Hadi,
A. H. A.; Awang, K.; Wiart, C.; Gallard, J. F.; Pais, M. Tetrahedron Lett.
1997, 38, 1571. (f) Awang, K.; Sevenet, T.; Hadi, A. H. A.; David, B.;
Pais, M. Tetrahedron Lett. 1992, 33, 2493.
(4) For some leading references, see: (a) Padwa, A.; Hornbuckle, S. F.
Chem. ReV. 1991, 91, 263. (b) Padwa, A.; Weingarten, M. D. Chem. ReV.
1996, 96, 223. (c) Padwa, A. Top. Curr. Chem. 1997, 189, 121. (d) Padwa,
A. Pure Appl. Chem. 2004, 76, 1933. (e) Padwa, A.; Brodney, M. A.; Lynch,
S. M.; Rashatasakhon, P.; Wang, Q.; Zhang, H. J. Org. Chem. 2004, 69,
3735.
(2) Kam, T. S.; Choo, Y. M. Tetrahedron Lett. 2003, 44, 1317.
(3) (a) Kam, T. S.; Choo, Y. M. Phytochemistry 2004, 65, 2119. (b)
Kam, T. S.; Choo, Y. M. HelV. Chim. Acta 2004, 87, 991.
10.1021/ol062029z CCC: $33.50
© 2006 American Chemical Society
Published on Web 09/27/2006

in benzene at 80 °C afforded cycloadduct 11 in 90% yield
as a single diastereomer (Scheme 3). The isolation of 11 is
Scheme 2
Scheme 3
from a metal carbene-cyclization-cycloaddition cascade
that our group has been studying for some time.5,6 Using
this metal-catalyzed domino reaction as a key step, the
heterocyclic skeleton of the kopsifolines could eventually
be built by a 1,3-dipolar cycloaddition of a carbonyl ylide
dipole derived from diazo ketoester 4 across the indole
π-bond.7,8 Ring opening of the resulting cycloadduct 5
followed by a reductive dehydroxylation step9 would lead
to the critical silyl enol ether 6 necessary for the final F-ring
closure. Although the very last step (i.e., 6 f 7) appears to
be reasonable, no example of such a reaction had been
previously reported in the literature. Accordingly, we decided
to study the facility and stereoselectivity of this process with
model substrates prior to commencing the total synthesis of
the kopsifolines. In this communication, we detail the
successful implementation of this strategy.
To test what appeared to be a logical but not directly
precedented F-ring closure, we first carried out a model study
using cycloadduct 11 to test the viability of our design as
well as the specific reactions to be used in a total synthesis
effort. Diazoimide 10 was easily prepared by reacting acid
chloride 8 with amide 9 in the presence of 4 Å molecular
sieves at room temperature followed by a diazo transfer with
MsN3.10 Heating a sample of 10 with catalytic Rh2(OAc)4
the consequence of endo cycloaddition with regard to the
dipole,7 and this is in full accord with the lowest-energy
transition state. The cycloaddition can also be considered
doubly diastereoselective in that the indole moiety ap-
proaches the dipole exclusively from the side of the 2-(ben-
zyloxy)ethyl group and away from the more sterically
encumbered piperidone ring.
On the basis of our previous studies using related aza-
oxabicyclic systems,11 we expected that treating cycloadduct
11 with Et3SiH and BF3‚OEt2 would result in a reductive
ring-opening reaction. We found, however, that the rear-
ranged nine-membered lactone 14 was obtained as the only
identifiable product in 60% yield. This unusual reorganization
can be rationalized by the pathway proposed in Scheme 4.
Scheme 4
(5) For some leading references, see: Padwa, A. HelV. Chim. Acta 2005,
88, 1357.
(6) For some related work using this metal-catalyzed domino reaction,
see: (a) Hodgson, D. M.; LeStrat, F.; Avery, T. D.; Donohue, A. C.; Bru¨ckl,
T. J. Org. Chem. 2004, 69, 8796. (b) Hodgson, D. M.; LeStrat, F. Chem.
Commun. 2004, 822. (c) Chiu, P. Pure Appl. Chem. 2005, 77, 1183. (d)
Chen, B.; Ko, R. Y. Y.; Yuen, M. S. M.; Cheng, K. F.; Chiu, P. J. Org.
Chem. 2003, 68, 4195. (e) Chiu, P.; Chen, B.; Cheng, K. F. Org. Lett. 2001,
3, 1721. (f) Graening, T.; Bette, V.; Neudo¨rfl, J.; Lex, J.; Schmalz, H. G.
Org. Lett. 2005, 7, 4317.
(7) (a) Padwa, A.; Price, A. T. J. Org. Chem. 1995, 60, 6258. (b) Padwa,
A.; Price, A. T. J. Org. Chem. 1998, 63, 556. (c) Mej´ıa-Oneto, J. M.; Padwa,
A. Org. Lett. 2004, 6, 3241. (d) Padwa, A.; Lynch, S. M.; Mej´ıa-Oneto, J.
M.; Zhang, H. J. Org. Chem. 2005, 70, 2206.
(8) Boger and co-workers have recently described [3+2]-cycloaddition
chemistry across the indole π-bond and have elegantly exploited this
chemistry for the construction of a variety of aspidosperma alkaloid targets.
See: Choi, Y.; Ishikawa, H.; Velcicky, J.; Elliott, G. I.; Miller, M. M.;
Boger, D. L. Org. Lett. 2005, 7, 4539 and references therein.
(9) Mej´ıa-Oneto, J. M.; Padwa, A. Org. Lett. 2006, 8, 3275.
(10) (a) Regitz, M. Chem. Ber. 1966, 99, 3128. (b) Regitz, M.; Hocker,
J.; Liedhegener, A. Org. Synth. 1973, 5, 179.
We assume that the first step involves a Lewis acid promoted
debenzylation with Et3SiH, and the transient alcohol 12 so
5142
Org. Lett., Vol. 8, No. 22, 2006

produced undergoes ready cyclization onto the adjacent
carbonyl group to generate a hemiketal intermediate. A
subsequent ring opening of the oxabicyclic ring affords
N-acyliminium ion 13 which then undergoes cleavage of the
C-C bond to dissipate the positive charge on the nitrogen
atom thereby giving rise to the observed lactone 14.
groups were sequentially removed using Cs2CO3 fol-
lowed9,12,13 by a samarium iodide reduction.14 The remaining
keto group was converted into the corresponding TBS enol
ether, and this was followed by a two-step reduction/
oxidation of the ester functionality to give 20 in 59% yield
for the five-step sequence (Scheme 6). Gratifyingly, when
We next investigated the key F-ring closure step that is
the foundation of our kopsifoline strategy by making use of
hemiketal 15. For these feasibility studies (Scheme 5), we
Scheme 6
Scheme 5
20 was treated with CsF in CH3CN at reflux, it was cleanly
converted to the hexacyclic skeleton of the kopsifoline
alkaloids (i.e., 21) in 78% yield whose structure was
confirmed by a single-crystal X-ray analysis.15
Because all members of the kopsifoline family of alkaloids
contain a carbomethoxy group at the C(16)-position of the
core skeleton, we decided to selectively remove the hydroxyl
group from hydroxy ester 19. To this end, the reduction of
19 was carried out with SmI2 in HMPA and the expected
ketoester 22 was isolated in 93% yield (Scheme 7). This
compound was easily converted into the corresponding TBS
converted 15 into keto-ester 17 which possesses a side chain
mesylate group that could undergo intramolecular displace-
ment with the carbanion derived from ketoester 17. Hemiket-
al 15 is easily available from cycloadduct 11 via lactam
reduction followed by catalytic hydrogenation which causes
reductive ring opening of the oxabicyclic ring, debenzylation,
and cyclization to give 15 in 43% yield for the three-step
sequence. Treatment of 15 with triethylamine/mesyl chloride
furnished 16 (85%) which was then reduced with SmI2 to
give 17 in 60% yield. Unfortunately, all of our efforts to
induce F-ring closure of 17 with various bases and under
different experimental conditions were unsuccessful.
This outcome suggested that the mesylate group was not
sufficiently reactive to undergo reaction with the anion
resulting from deprotonation of the â-ketoester. Conse-
quently, we turned our attention to replacing the mesylate
group with a more reactive aldehyde functionality on the
side chain. To access a suitable and useful substrate for the
key aldol reaction, we first synthesized the related cycload-
duct 18 which bears a 2-methyl acetate group at the ring
juncture. This compound was readily available from the
corresponding diazoimide in 98% yield as a single diaste-
reomer by treating it with Rh2(OAc)4 in refluxing benzene.
We then converted 18 into the ring-opened hydroxy ester
19 (68%) using the same three-step protocol that was used
in Scheme 5. At this point, the carbomethoxy and hydroxyl
Scheme 7
(11) Padwa, A.; Brodney, M. A.; Marino, J. P., Jr.; Sheehan, S. M. J.
Org. Chem. 1997, 62, 78.
Org. Lett., Vol. 8, No. 22, 2006
5143

silyl enol ether in 96% yield. Reaction of the enol ether with
LAH at 0 °C resulted in the selective reduction of the12
nonconjugated carbomethoxy group affording 23 in 91%
yield. The primary alcohol so formed was then subjected to
oxidation using TPAP/NMO at 0 °C, and the crude reaction
mixture was subsequently treated with CsF in refluxing CH3-
CN to give the desired F-ring-cyclized ester 24 in 34% yield
for the two-step sequence. Despite the additional manipula-
tions associated with the processing of intermediate 19 and
the somewhat lower yield, only five steps are required to
prepare a complex hexacyclic structure that contains the
complete skeleton of the kopsifoline alkaloids.
In summary, the new cyclization chemistry described
herein gives rapid access to the hexacyclic core skeleton of
the kopsifoline alkaloids, and the products synthesized appear
to be suitable for further manipulation toward the natural
products and their close analogues. Further application of
the metal carbene cyclization-cycloaddition cascade toward
several other monoterpenoid indoline alkaloids is currently
being investigated.
Acknowledgment. We appreciate the financial support
provided by the National Institutes of Health (GM 059384)
and the National Science Foundation (CHE-0450779).
(12) (a) Rubin, M. B.; Inbar, S. J. Org. Chem. 1988, 53, 3355. (b) Rubin,
M. B.; Inbar, S. Tetrahedron Lett. 1979, 20, 5021. (c) Davis, F. A.; Clark,
C.; Kumar, A.; Chen, B.-C. J. Org. Chem. 1994, 59, 1184.
Supporting Information Available: Spectroscopic data
and experimental details for the preparation of all new
compounds together with an ORTEP drawing for compound
21. This material is available free of charge via the Internet
at http://pubs.acs.org.
(13) For a review of dealkoxycarbonylations, see: (a) Krapcho, A. P.
Synthesis 1982, 805. (b) Krapcho, A. P. Synthesis 1982, 893.
(14) Molander, G. A. In ComprehensiVe Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 4, pp 251-282.
(15) The authors have deposited coordinates for structure 21 with the
Cambridge Crystallographic Data Centre. The coordinates can be obtained
on request, from the Director, Cambridge Crystallographic Data Centre,
12 Union Road, Cambridge, B2 1EZ, U.K.
OL062029Z
5144
Org. Lett., Vol. 8, No. 22, 2006