A gram-scale route to phlegmarine alkaloids: Rapid total synthesis of (-)-cermizine B

Article abstract of DOI:10.1039/c4cc01708k

The synthesis of the Lycopodium alkaloid (-)-cermizine B (1), which establishes its absolute configuration, is achieved by combining asymmetric organocatalysis and an uninterrupted eight-step reaction sequence, followed by a final reduction step. This pot-economy strategy provides access to the cis-phlegmarine stereoparent embedded in 1 for the first time, rapidly and on a gram-scale.

Full text of DOI:10.1039/c4cc01708k

Volume 50 Number 54 11 July 2014 Pages 7077–7206
ChemComm
Chemical Communications
www.rsc.org/chemcomm
ISSN 1359-7345
COMMUNICATION
Ben Bradshaw, Josep Bonjoch et al.
A gram-scale route to phlegmarine alkaloids: rapid total synthesis
of (−)-cermizine B

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A gram-scale route to phlegmarine alkaloids:
rapid total synthesis of (À)-cermizine B†
Cite this: Chem. Commun., 2014,
50, 7099
Ben Bradshaw,* Carlos Luque-Corredera and Josep Bonjoch*
Received 6th March 2014,
Accepted 3rd April 2014
DOI: 10.1039/c4cc01708k
www.rsc.org/chemcomm
The synthesis of the Lycopodium alkaloid (À)-cermizine B (1), which
establishes its absolute configuration, is achieved by combining
asymmetric organocatalysis and an uninterrupted eight-step reaction
sequence, followed by a final reduction step. This ‘‘pot-economy’’
strategy provides access to the cis-phlegmarine stereoparent embedded
in 1 for the first time, rapidly and on a gram-scale.
One of the main impediments to accessing viable quantities of
complex natural products in a timely and cost-effective manner
is the ‘stop-and-go’ approach,1 which requires a purification
step after each synthetic operation. This undoubtedly involves
the greatest investment of time and materials in any total
synthesis endeavour and is a major source of waste generation.
An alternative strategy involving a series of tandem reactions2 in
combination with ‘pot-economy’3 offers a potent solution to this
problem. By eliminating the need for work-up and product isolation
between successive synthetic steps, it becomes possible to complete
an entire multi-step sequence in a single pot,4 a process that
approaches Wender’s definition of an ‘ideal synthesis’.5
With this aim in mind, and based on our recent work on
asymmetric synthesis of decahydroquinolines using organo-
catalysis,6 we decided to apply the aforementioned ‘‘one-pot’’
operational approach in the field of Lycopodium alkaloids7
(Fig. 1). This group of structurally diverse compounds have
elicited major interest in recent years for their potential use in the
treatment of severe neurodegenerative diseases.7d,8 A great number
of approaches to the main classes of Lycopodium alkaloids,9 namely
lycopodine, lycodine, and fawcettimine, have been developed, but
there are limited synthetic entries to their biogenetic precursors, the
Fig. 1 (a) Structures of phlegmarine alkaloids (the three stereoparents),
including the biogenetic numbering. (b) Strategy for synthesis of cermizine B.
phlegmarine alkaloids. To date, the synthesis of cis-phlegmarine studies of MacLean,11 the attainment of lycoposerramine X by
alkaloids has been limited to lycoposerramine Z,6a,10 whereas Takayama10,12 and the comprehensive phlegmarine synthesis
synthetic efforts towards trans-phlegmarines include the pioneering reported by Comins.13 Considering that all the previous syntheses,
apart from our recent work,6a have required more than twenty steps,
a more efficient entry to the phlegmarine alkaloids would not only
´
`
`
Laboratori de Quımica Organica, Facultat de Farmacia, IBUB,
Universitat de Barcelona, Av. Joan XXIII s/n, 08028-Barcelona, Spain.
E-mail: benbradshaw@ub.edu, josep.bonjoch@ub.edu
be a desirable goal in itself, but could also be used as a platform to
access members of the other classes of Lycopodium alkaloids.
We herein report a highly efficient synthesis and the absolute
† Electronic supplementary information (ESI) available: Experimental procedures,
characterization data and copies of NMR spectral data. See DOI: 10.1039/c4cc01708k configuration of the cis phlegmarane-type alkaloid cermizine B (1),14
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 7099--7102 | 7099

View Article Online
Communication
ChemComm
Scheme 1 Optimized synthesis of cis-decahydroquinoline 5.
the first of a member of the cis-phlegmarine subset with an anti one-pot sequence, alkene 10 was hydrogenated in the presence of
stereochemical relationship between the ring-fusion hydrogens and Pd/C in MeOH, which delivered the hydrogen selectively from the
the prototypical methyl substituent (C-16) in the decahydroquino- bottom face of the molecule in a 5 : 1 ratio to give 11. We believe that
line ring.
the facial selectivity was primarily influenced by the tosyl group,
The overall process was initially developed by dividing the which effectively blocked the top face of 10. Removal of the tosyl
synthesis into three sets of tandem reactions (Scheme 1), which group prior to the hydrogenation led to the selective reduction from
were subsequently fused into a single sequence. The first one-pot the top face of the molecule to give the opposite stereochemistry.
operation began with b-keto ester 2, which underwent an organo- Notably, this provides the stereochemical arrangement of another
catalyzed Michael reaction in the presence of just 5% loading of the phlegmarine alkaloid (huperzine M).17 Hydrogenation of the
modified Hayashi catalyst 3,15 improving our preliminary published pyridine ring was then accomplished by direct addition of PtO2
results.6a Removal of the solvent and treatment of 4 with LiOH in the and AcOH to the reaction mixture containing 11.18 Filtration,
presence of iPrOH and water led to the tandem aldol condensation– evaporation and carbamation with methyl chloroformate gave 12
intramolecular aza-Michael reaction, which delivered cis-decahydro- as a stable, readily purified epimeric mixture. Finally, treatment
quinoline 5. With the aim of minimizing waste and preventing with LiAlH4 converted the carbamate into the required methyl
downstream accumulation in the eventual uninterrupted sequence, group and smoothly removed the tosyl group to give cermizine B
the reaction mixture was treated with a sulfonic acid resin, eliminating (1), which showed NMR spectroscopic data identical to those
the need for any work-up procedure by removing the basic residues reported for the natural product (see ESI†). The overall yield of
and also allowing the capture and recuperation of the organocatalyst the complete sequence was 20%. The data obtained for 1 [a]D
=
in excellent yield. We believe that this solution is a simple alternative À3.1 (c 0.7, MeOH) also confirmed that the absolute configu-
to recycling the catalyst via immobilization on a solid support. It ration depicted in Fig. 1 corresponds to that of the natural
should be noted that not only did the modified catalyst (triphenylsilyl product (À)-cermizine B {lit14 [a]D À2.0 (c 0.6, MeOH)}.19
instead of trimethylsilyl diphenylprolinol ether) improve the enantio-
After the successful completion of the synthesis via a sequence
meric excess but its significantly more robust nature also proved involving 3 one-pot reactions, we sought to eliminate the remaining
essential for its recovery with the acidic resin at the end of the purification steps until carbamate 12, which would render the whole
reaction. Decahydroquinoline 5 was isolated in 75% overall yield16 process even more efficient (Scheme 3). Thus, after treatment with
from keto ester 2 with 90% ee.
the resin at the end of the first one-pot sequence, the solvent was
The second one-pot sequence (Scheme 2) commenced with removed by evaporation and the resulting material was fed directly
treatment of 5 with TFA, followed by evaporation and heating to into the second set of tandem reactions, and then, after subsequent
effect the complete decarboxylation of the resulting keto acid moiety. removal of the LiOH by filtration, into the third set. Reduction of this
Direct addition of LiOH and THF to the reaction pot, followed by enantioenriched material, i.e. 12 (after a single purification step to
refluxing overnight resulted in the retro aza-Michael ring opening of remove epi-12), with LiAlH4 then took place in 88% yield to give
6 to 7, and subsequent partial closure to the more stable decahydro- 0.96 g of the target cermizine B (1) from just 5 g of the starting b-keto
quinoline 8,6b with the required stereochemistry present in ester 2. By eliminating the aforementioned purification procedures,
cermizine B. Although the closure did not go to completion, giving the overall yield for the new integrated sequence was increased to
an equilibrium mixture of 6:7:8 (1:4.5:4.5, respectively), it was 26%, and could be completed in only 8 h of operational time, spaced
found that direct addition of phosphonate 9 led to a chemoselective over 10 days. To evaluate the overall efficiency of the process we
reaction with 8, and by means of a shift in the equilibrium, applied the notation introduced by Jørgensen for one-pot reactions4c
vinylpyridine 10 was formed exclusively as an inconsequential (see Scheme 3). The most striking feature of the uninterrupted
mixture of Z/E isomers (1 : 5 ratio). Hydrogenation (see below) of sequence from the starting material to the precursor of cermizine
each isomer gave the same result, obviating the need to separate B is that it consists of eight consecutive reactions for eight manual
them. Furthermore, we noted that this coupling reaction performed operations (nmo) and only one final purification. For the total
better if carried out under solvent-free conditions, giving 10 in an synthesis of cermizine B, which required an additional step and
excellent yield of 89% over the entire sequence from 5. For the final purification, the values of YPBF (76% yield per bond formed) and
7100 | Chem. Commun., 2014, 50, 7099--7102
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
Scheme 2 Total synthesis of (À)-cermizine B (1).
phlegmarane-type alkaloids underlines the efficiency and simplicity
offered by asymmetric organocatalytic one-pot cascades.
In summary, a highly efficient, enantioselective total synthesis of
cermizine B (1) was completed using a carefully orchestrated series
of tandem reactions and the principles of pot economy. Our route
to cermizine B constitutes the first synthesis of this type of
cis-phlegmarine alkaloids. The convergent decahydroquinoline
synthetic entry was key to the successful development of this concise,
sustainable eight-step procedure that has allowed a gram scale
synthesis of the natural product.20 Considering the flexible, highly
convergent and modular nature of this methodology, it could also be
used to access the other phlegmarine alkaloids by changing stereo-
motifs in the decahydroquinoline core. Furthermore, it should
enable the rapid synthesis of other Lycopodium alkaloids and analogs
for the development of a comprehensive structure–activity relation-
ship of this important class of compounds.
Financial support for this research was provided through
Project CTQ2010-14846/BQU by the Ministry of Economy and
Competitiveness of Spain.
Notes and references
1 (a) A. M. Walji and D. W. C. MacMillan, Synlett, 2007, 1477;
(b) N. J. Green and M. S. Sherburn, Aust. J. Chem., 2013, 66, 267.
2 (a) C. Grondal, M. Jeanty and D. Enders, Nat. Chem., 2010, 2, 167;
(b) L. M. Ambrosini and T. H. Lambert, ChemCatChem, 2010,
2, 1373; (c) C. M. R. Volla, I. Atodiresei and M. Rueping, Chem.
Rev., 2014, 114, 2390.
3 (a) H. Ishikawa, T. Suzuki and Y. Hayashi, Angew. Chem., Int. Ed.,
2009, 48, 1304; (b) H. Ishikawa, T. Suzuki, H. Orita, T. Uchimaru and
Y. Hayashi, Chem. – Eur. J., 2010, 16, 12616; (c) P. A. Clarke, S. Santos
and W. H. C. Martin, Green Chem., 2007, 9, 438; (d) Y. Hayashi and
S. Umemiya, Angew. Chem., Int. Ed., 2013, 52, 3450.
Scheme 3 Pot-economy synthesis of cermizine B on a gram-scale and
summary of the process according to Jørgensen’s notation (ref. 4c).
YPMO (86% yield per manual operation) indicate that each step in
this synthesis proceeded in high yield. The very high value of the
purification factor (Pf = 6), which denotes the number of purifica-
tions avoided, is also noteworthy. The considerable difference in the
number of required steps compared with previous approaches to
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 7099--7102 | 7101

View Article Online
Communication
ChemComm
4 (a) H. Ishikawa, M. Honna and Y. Yahashi, Angew. Chem., Int. Ed., 2011, 13 (a) D. L. Comins, A. H. Libby, R. Al-awar and C. Foti, J. Org. Chem.,
50, 2824; (b) C. Vaxelaire, P. Winter and M. Christmann, Angew. Chem.,
Int. Ed., 2011, 50, 3605; (c) L. Albrecht, H. Jiang and K. A. Jørgensen,
1999, 64, 2184; (b) B. H. Wolfe, A. H. Libby, R. Al-awar, C. J. Foti and
D. L. Comins, J. Org. Chem., 2010, 75, 8564.
Angew. Chem., Int. Ed., 2011, 50, 8492; (d) T. Mukaiyama, H. Ishikawa, 14 For isolation of cermizines A–D, see: H. Morita, Y. Hirasawa,
H. Koshino and Y. Hayashi, Chem. – Eur. J., 2013, 19, 17789. T. Shinzato and J. Kobayashi, Tetrahedron, 2004, 60, 7015.
5 (a) P. A. Wender and B. L. Miller, Nature, 2009, 460, 197; (b) P. A. Wender, 15 (a) Y. Wang, P. Li, X. Liang and J. Ye, Adv. Synth. Catal., 2008,
´
Tetrahedron, 2013, 69, 7529; (c) P. A. Wender, Nat. Prod. Rep., 2014, 31, 433.
6 (a) B. Bradshaw, C. Luque-Corredera and J. Bonjoch, Org. Lett., 2013,
350, 1383; (b) E. Gomez-Bengoa, A. Landa, A. Lizarraga, A. Mielgo,
M. Oiarbide and C. Palomo, Chem. Sci., 2011, 2, 353.
15, 326; (b) B. Bradshaw, C. Luque-Corredera, G. Saborit, C. Cati 16 Notably, along with decahydroquinoline 5, small amounts of
viela, R. Dorel, C. Bo and J. Bonjoch, Chem. – Eur. J., 2013, 19, 13881.
7 (a) X. Ma and D. R. Gang, Nat. Prod. Rep., 2004, 21, 752; (b) Y. Hirasawa,
J. Kobayashi and H. Morita, Heterocycles, 2009, 77, 679; (c) P. Siengalewicz,
J. Mulzer and U. Rinner, The Alkaloids, 2013, vol. 72, p. 1; (d) R. A. Murphy
and R. Sarpong, Chem. – Eur. J., 2014, 20, 42.
non-cyclized product and material with the b-keto ester in the keto
rather than enol form were also isolated. While the presence of
these products reduced the yield of 5, not separating them in the
uninterrupted sequence allowed their conversion to the final product,
thereby effectively increasing the yield of the first sequence.
8 J. Xu, M. H. Lacoske and E. A. Theodorakis, Angew. Chem., Int. Ed., 17 W.-Y. Gao, Y.-M. Li, S.-H. Jiamg and D.-Y. Zhu, Helv. Chim. Acta,
2014, 53, 956. 2008, 91, 1031.
9 (a) C. H. Heathcock, E. Kleinman and E. S. Binkley, J. Am. Chem. 18 It should be noted there is no effective asymmetric version for
Soc., 1982, 104, 1054; (b) D. F. Fischer and R. Sarpong, J. Am. Chem.
Soc., 2010, 132, 5926; (c) B. B. Liau and D. M. Shair, J. Am. Chem. Soc.,
2010, 132, 9594; (d) J. Ramharter, H. Weinstabl and J. Mulzer, J. Am.
Chem. Soc., 2010, 132, 1433.
reducing alkylpyridines. A preliminary examination of the methodology
of Glorius using a pyridine analogous to 9 with a chiral auxiliary at C-6
was not sufficiently satisfactory. For the procedure, see: F. Glorius,
N. Spielkamp, S. Holle, R. Goddard and C. W. Lehman, Angew. Chem.,
Int. Ed., 2004, 43, 2850.
10 T. Tanaka, N. Kogure, M. Kitajima and H. Takayama, J. Org. Chem.,
2009, 74, 8675.
11 A. Leniewski, J. Szychowski and D. B. MacLean, Can. J. Chem., 1981,
59, 2479.
19 To further confirm the stereochemistry at the stereogenic center at
C-5, we also prepared the non-natural epi-cermizine B from epi-12 to
compare their NMR data (see ESI,† Tables S1 and S2) and optical
rotation [a]D = +50.0 (c 0.95, MeOH).
12 For the synthesis of lycoposerramines V and W, embodying a
tetrahydroquinoline framework, see: T. Shigeyama, K. Katakawa, 20 C. A. Kuttruff, M. D. Eastgate and P. S. Baran, Nat. Prod. Rep., 2014,
N. Kogure, M. Kitajima and H. Takayama, Org. Lett., 2007, 9, 4069. 31, 419.
7102 | Chem. Commun., 2014, 50, 7099--7102
This journal is ©The Royal Society of Chemistry 2014