Synthesis of the putative structure of (±)-amarbellisine

Article abstract of DOI:10.1021/ol3034093

The title compound was synthesized mainly by palladium catalytic coupling, cyclopropyl ring-opening rearrangement, epoxidation, Swern oxidation, demethanol reactions, and selective reduction. This synthesis was achieved in 16 steps with 9.7% overall yield

Full text of DOI:10.1021/ol3034093

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis of the Putative Structure
of (()-Amarbellisine
Dandan Liu, Jingbo Chen, Long Ai, Hongbin Zhang, and Jianping Liu*
Key Laboratory of Medicinal Chemistry for Natural Resource (Yunnan University),
Ministry of Education, School of Chemical Science and Technology, Yunnan University,
Kunming, Yunnan 650091, P. R. China
jpliu@ynu.edu.cn
Received December 13, 2012
ABSTRACT
The title compound was synthesized mainly by palladium catalytic coupling, cyclopropyl ring-opening rearrangement, epoxidation, Swern
oxidation, demethanol reactions, and selective reduction. This synthesis was achieved in 16 steps with 9.7% overall yield. Unfortunately, the
published spectroscopic data do not match with those of our synthetic compound.
Due to their potential biological activities as antitumor,1
antiviral, and antimicrobial activities,2 as well as the penta-
cyclic structures, lycorine-type alkaloids have attracted
both medicinal and synthetic chemists’ interests for a long
time.3 More than 100 structurally diverse alkaloids, pos-
sessing a wide spectrum of biological activities, have
been isolated from various Amaryllidaceae species.4 Most
of those isolated alkaloids have a trans-B/C ring system
as lycorine (1),5 and only a few showed cis-B/C ring
configurations6 as γ-lycorane (2). Evidente’s group found
Figure 1. Lycorine-type alkaloids.
a new lycorine-type alkaloid, which was named amarbelli-
sine (3) (Figure 1). They reported this natural compound
had both cis-B/C and C/D rings and showed interesting
bioactivities.7 In this context we report the synthesis of
putative (()-amarbellisine, and to the best of our knowl-
edge, this is the first total synthesis of the compound
reported.
(1) Tang, W.; Hemm, I.; Bertram, B. Planta Med. 2003, 69, 97 and
references cited therein.
Retrosynthetic analysis is showed in Scheme 1. The
target molecule ((3) could be afforded from correspond-
ing dihydroxyl compound 5, which could be obtained
from olefin 6. The previous methodology developed in
our group8 would be employed for the construction of
lycorine-type skeleton 6 from bromodiene 8.
(2) Lewis, J. R. Nat. Prod. Rep. 1998, 15, 107. 1994, 11, 329.
(3) (a) Padaw, A.; Zhang, H. J. Org. Chem. 2007, 72, 2570. (b)
Martin, S. F. The Amaryllidaceae Alkaloids. In The Alkaloids; Brossi,
A., Ed.; Academic Press: New York, 1987; Vol. 30, Chapter 3, p 251. (c)
Dong, L.; Xu, Y.; Cun, L.; Cui, X.; Mi, A.; Jiang, Y.; Gong, L. Org. Lett.
2005, 7, 4285.
(4) Hoshino, O. In The Alkaloids; Cordell, G. A., Ed.; Academic Press:
London, 1998; Vol. 51, pp 323À376.
(5) (a) Takagi, S.; Taylor, W. I.; Uyeo, S.; Yajima, H. J. J. Chem. Soc.
1955, 4003. (b) Mason, L. H.; Puschett, E. R.; Wilman, C. W. J. Am.
Chem. Soc. 1995, 77 (5), 1253. (c) Herrera, M. R.; Machiocho, A. K.;
Nair, J. J.; Campbell, W. E.; Brun, R.; Bastida, J. Fitoterapia 2001, 72,
444. (d) Feng, T.; Wang, Y.; Su, J.; Li, Y.; Cai, X.; Luo, X. Helv. Chim.
Acta 2011, 94 (1), 178.
(7) (a) Evidente, A.; Andolfi, A.; Abou-Donia, A. H.; Touema, S. M.;
Hammoda, H. M.; Shawky, E.; Motta, A. Phytochemistry 2004, 65,
2113–2118. (b) Van Goietsenoven, G.; Andolfi, A.; Lallemand, B.;
Cimmino, A.; Lamoral-Theys, D.; Gras, T.; Abou-Donia, A.; Dubois,
J.; Lefranc, F.; Mathieu, V.; Kornienko, A.; Kiss, R.; Evidente, A.
J. Nat. Prod. 2010, 73, 1223–1227.
(6) (a) Batida, J.; Codina, C.; Peeters, P.; Rubiralta, M.; Orozco, M.
Phytochemistry 1995, 40 (4), 1291. (b) Richomme, P.; Gozler, T.;
Shamma, M. J. Nat. Prod. 1989, 52 (5), 1150.
(8) Shao, Z.; Chen, J.; Tu, Y.; Li, L.; Zhang, H. Chem. Commun.
2003, 1918.
r
10.1021/ol3034093
XXXX American Chemical Society

Scheme 1. Retrosynthetic Analysis of Putative Amarbellisine (3)
Scheme 2. Synthesis of the Intermediate 14
Figure 2. Crystal structures of compounds 11 and 14.
our previous work.9 Unsaturated ketone 7 was obtained
by demethylation of 8 with 2 M HCl. Bromocyclization
of 7 with bromine followed by base treatment at low
temperature afforded bromoketone 9. Cyclopropyl ring
compound 10 was formed when 9 was further exposed
to base. Compound 11, with a full lycorine-type skeleton,
was obtained from cycloketone 10 by a palladium catalytic
coupling reaction.10 The reduction of 11 with sodium
borohydride afforded cyclopropyl alcohol 12, which could
be converted to corresponding cyclopropyl bromide 13 in
the presence of PBr3. Secondary bromide 13 could undergo
a cyclopropyl ring-opening rearrangement reaction to
form homoallylic bromide 14. An interesting phenomena
was observed with treatment of either 13 or 14 with
aqueous hydrogen bromide: the final ratio of about 1:4
for compounds 13 and 14 was always observed. The
cyclopropyl ring-opening rearrangement reaction was first
reported by Julia over 50 years ago and was followed by
considerable development and synthetic applications,11
while the reverse reaction from homoallylic halides
to cyclopropyl ring compounds was reported mainly in
(10) (a) Johansson, C. C. C.; Colacot, T. J. Angew. Chem., Int. Ed.
2010, 49, 676. (b) Bellina, F.; Rossi, R. Chem. Rev. 2010, 110, 1082.
(11) (a) Julia, M.; Julia, S.; Guegan, R. Bull. Soc. Chim. Fr. 1960,
1072. (b) Hoffsommer, R. D.; Taub, D.; Wendler, N. L. J. Org. Chem.
1962, 27, 4134. (c) Julia, M.; Descoinis, C.; Risse, C. Tetrahedron Suppl.
1966, 8, 443. (d) Sarel, S.; Yovell, J.; Sarel-Imber, M. Angew. Chem., Int.
Ed. Engl. 1968, 7, 577. (e) Brady, S. F.; Ilton, M. A.; Johnson, W. S.
J. Am. Chem. Soc. 1968, 90, 2882. (f) McCormick, J. P.; Barton, D. L.
J. Org. Chem. 1980, 45, 2566–2570. (g) C. Ferreri, C.; Ambrosone, M.
Synth. Commun. 1995, 25, 3351–3356. (h) Michelliza, S.; Al-Mourabit,
A.; Gateau-Olesker, A.; Marazano, C. J. Org. Chem. 2002, 67, 6474–
6478.
Our journey started from diene bromide compound 8
(Scheme 2), which could be prepared from 2-(p-
methoxyphenyl)ethylamine and piperonal according to
(9) Shao, Z.; Chen, J.; Huang, R.; Wang, C.; Li, L.; Zhang, H. Synlett
2003, 2228.
B
Org. Lett., Vol. XX, No. XX, XXXX

Scheme 3. Synthesis of Putative Amarbellisine (3)
Figure 3. Crystal structures of 22 and 3.
obtained by acid hydrolysis15 of epoxide 15, and further
investigation showed thatitcouldbeaffordedfromolefin6
directly by treatment with the TFA/Cl3CCN/H2O2 system
for a longer reaction time. Acetate 18 could be afforded
from diol 16byselectivehydroxyl protection/deprotection.
Corresponding ketone 19 was obtained from acetate 18 by
Swern oxidation.16 Dimethyl ketal 20 was formed from
compound 19 by treatment with trimethyl orthoformate
and p-toluenesulfonic acid;17 the acetyl was also removed
in this manipulation. Methoxyl ethylene compound 22
was afforded by careful removal of methanol from di-
methoxyl ketal 21, which could be formed by oxida-
tion from ketal 20. Putative rac-amarbellisine (3) was
obtained as the only isomer by reduction of ketone 22
with DIBAL-H, while the target compound ((3) and its
epi-isomer 23 could be formed in a ratio of about 3:1 if
NaBH4/CeCl3 or Red-Al were employed as a reduction
system (Scheme 3).18
anhydrous media,12,13 but there was no such equilibrium
reported.
The structures as well as the relative configurations of
cyclopropyl ketone 11 and rearrangement product 14 were
confirmed by crystal X-ray crystallography (Figure 2).
The epoxidation of alkene compound 6, which was
obtained by reduction of bromide 14, encountered some
problems. Desired product 15 could not be obtained when
compound 6 was reacted directly with mCPBA, or
Cl3CCN/H2O2 according to von Holleben’s report,14 until
trifluoroacetic acid (TFA) was employed. We thought that
TFA shielded nitrogen by formation of a corresponding
salt in 6 and kept it nonoxidizable, so selective epoxidation
with olefin became possible. The diol 16 could be easily
(12) (a) Hrubiec, R. T.; Smith, M. B. J. Chem. Soc., Perkin Trans. 1
1984, 107–110. (b) Previtera, L.; Monaco, P.; Mangoni, L. Tetrahedron
Lett. 1984, 25, 1293.
(16) Huang, S. L.; Omura, K.; Swern, D. J. Org. Chem. 1976, 41,
3329.
(13) For more related research, see the following review and references
herein: Taylor, R. E.; Engelhardt, F. C.; Schmitt, M. J. Tetrahedron 2003,
59, 5623–5634.
(14) Von Holleben, M. L. A.; Livotto, P. R.; Schuch, C. M. J. Braz.
Chem. Soc. 2001, 12 (1), 42.
(17) Rapoport, H.; Lovell, C. H.; Reist, H. R.; Warren, M. E., Jr.
J. Am. Chem. Soc. 1967, 89 (8), 1942.
(18) (a) Antoine, J.; Emilie, L.; Jean-luc, V. Org. Lett. 2010, 12 (22),
5128. (b) Joseph, E.; Jeremy, W.; David, G. J. Am. Chem. Soc. 2006, 128,
10370. (c) Thomas, E.; Michael, L.; Rajesh, I. J. Org. Chem. 1999, 64,
2391. (d) Luche, J. L. J. Am. Chem. Soc. 1978, 100, 2226.
(15) Takeda, K.; Kotera, K. Chem. Pharm. Bull. 1957, 5, 234.
Org. Lett., Vol. XX, No. XX, XXXX
C

Evidente’s group, and the results suggested they were dif-
ferentisomers.19 Since the obvious differences in resonating
positions for the two compounds appeared at C3a, C3, C1,
and C11c from 13C NMR spectra, and H11c, H3, H1, H5,
and H7 from 1H NMR spectra, it seemed that the B/C or
C/D ring configuration of the two compounds might be
different. The X-ray crystallography data showed that the
C ring of the synthesized product adopted a half-chairlike
conformation (Figure 4), and six carbons on the ring
remained nearly in one plane except for C11b, so the values
of the axialÀequatorial or axialÀaxial coupling in the
system would not be so typical. It suggested that the
configuration of the natural amarbellisine reported should
be reconfirmed.
Figure 4. A half-chairlike C ring conformation of the synthe-
sized compound 3 from another angle of crystal.
Acknowledgment. This work was supported by grants
(Nos. 20562012, 21162033, 21062029) from the National
Natural Science Foundation of China and Natural Science
Foundation of Yunnan Province (2009CD007). The
authors thank Prof. Mingjin Xie, School of Chemical
Science and Technology of Yunnan University, for
X-ray crystallography analysis.
Finally, the synthesis of putative amarbeillisine was
achieved in 16 steps with 9.7% overall yield; the structures
and relative stereochemistry of the target compound ( (3)
and its precursor 22 were confirmed by X-ray crystal-
lography (Figure 3).
The NMR data of our synthesized compound ((3)
were compared with those of natural amarbellisine from
Supporting Information Available. Detailed experimen-
tal procedures and compound characterization data. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(19) 13C NMR signals for C3, C3a of ((3) appeared at 96.9 and
35.3 ppm, while Evidente’s compound signals appeared at 112.9 and
58.6 ppm respectively; 1H NMR signals for H1, H5, H7 and H11c of
( (3) appeared at3.98, 3.24, 3.29, and2.68ppm, while those for Evidente’s
compound were at 3.48, 3.02, 3.79, and 4.08 ppm. Whole comparison
tables were shown in the Supporting Information, pp S11ÀS12.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX