A short synthesis of (±)-cytisine

Article abstract of DOI:10.1039/b405275g

The synthesis of racemic cytisine 1 has been completed using (i) N-selective alkylation of 6-bromopyridone with bromide 6 and (ii) Pd(o) mediated intramolecular a-arylation of lactam 8 as key steps to achieve rapid assembly of the tricyclic core skeleton of the lupin alkaloids.

Full text of DOI:10.1039/b405275g

View Article Online / Journal Homepage / Table of Contents for this issue
A short synthesis of ( )-cytisine†
Candice Botuha, Carl M. S. Galley and Timothy Gallagher*
School of Chemistry, University of Bristol, Bristol, UK BS8 1TS.
E-mail: t.Gallagher@bristol.ac.uk; Fax: ϩ44 117 929 8611;
Tel: ϩ44 117 928 8260
Received 7th April 2004, Accepted 7th May 2004
First published as an Advance Article on the web 4th June 2004
The synthesis of racemic cytisine 1 has been completed
using (i) N-selective alkylation of 6-bromopyridone with
bromide 6 and (ii) Pd(0) mediated intramolecular ꢀ-aryl-
ation of lactam 8 as key steps to achieve rapid assembly of
the tricyclic core skeleton of the lupin alkaloids.
(Ϫ)-Cytisine 11a,b is an important representative member of the
lupin class of alkaloids,2 which also includes (ϩ)-sparteine 2, a
molecule that has found broader application as a ligand within
asymmetric synthesis.3 Cytisine, which is a potent nicotinic
agonist4 with specificity for the α4β2 subtype and is currently
marketed as Tabex for use as a smoking cessation aid, was
first synthesised by van Tamelen,5a,b followed closely by both
Bohlmann5c and Govindachari,5d but there was little further
activity in this area for almost fifty years. In 2000, the Pfizer
groups of O’Neill6a and Coe6b each described new routes to
( )-cytisine, as part of a broader drug discovery project aimed
at exploiting the therapeutic potential of nicotinic agonists that
has led to analogues such as 3.7 This resurgence of interest in
this area has continued at pace with the first asymmetric
synthesis of (Ϫ)-cytisine 1 being reported very recently.8
Scheme 1 Reagents and conditions: i, BnNH , Et O, rt, then CH ᎐
2
2
2
᎐
CH2COCl, THF, 70 ЊC (56%) ref. 9; ii, H2 (1 atm), Pd/C, Na2CO3,
EtOH, 20 ЊC (95%); iii, LiAlH4, THF, Ϫ10 ЊC; iv, PBr3, PhMe, 110 ЊC
(57% over two steps); v, 7 (1 equiv.), NaH (2 equiv.), LiBr (4 equiv.),
DME/DMF (24 : 1), concentration of both 6 and 7 = 0.2 M, 10 days (8:
61%; 9: 25%; 10: 14%) vi, Pd(OAc)2 (5 mol%), ( )-BINAP (7.5 mol%),
KHMDS (2 equiv.), THF, 70 ЊC (44%); vii, BH3ؒTHF, THF, 0 ЊC to rt
(55%); viii, HCl in MeOH (1 equiv.), H2 (1 atm), Pd(OH)2/C, MeOH
(88%).
LiBr (4 equiv.), DME/DMF (10 : 1), NaH (1 equiv.), rt to 65 ЊC]
gave no conversion after 6 days. We then examined conditions
similar to those previously used by Curran for more reactive
alkylating agents [(6-bromopyridone (1 equiv.), alkyl halide
(1 equiv.), LiBr (2 equiv.), DME/DMF (4 : 1), NaH (1 equiv.),
rt to 65 ЊC]. Under these conditions a poor yield (20%) of
the desired N-alkylated product 8 was obtained. The major
adduct isolated was the O-alkylated isomer 9 (57%), and the
elimination product 10 (23%) was also observed.
Two strategies were adopted to address this problem. We
screened a wide range of other reaction conditions to improve
access to 8, but with little success. This included examination
of different substrates, and full details of these attempts will
be described at a later date. Our second approach was to
refocus on the direct reaction between 6 and 7 and explore this
process using factorial experimental design methods.12 Follow-
ing our initial studies, a number of parameters were recognised
as potentially important in terms of determining N- vs.
O-selectivity, as well as pyridone alkylation vs. elimination.
These included reaction temperature, substrate concentration,
DMF/DME ratio, and the number of equivalents of both LiBr
and NaH.‡ Using this statistical approach, we have identified
modified conditions (details are presented in Scheme 1)
that serve to suppress elimination (<15% of this pathway is
observed) and provides a 2.4 : 1 ratio of N- to O-alkylated
adducts 8 and 9 respectively, with 8 being obtained in 61% yield.
This represents a comparatively minor adjustment to Curran’s
original conditions, but using 6 as the alkylating agent
these changes are significant in practical terms. This exercise
In this paper we describe a novel strategy for the synthesis
of cytisine 1 that has, we believe, potential for more general
application to the assembly of a range of other, more complex
lupin alkaloids, as well as unnatural, but potentially important
variants such as 3.
Our approach to racemic cytisine has relied on two key steps:
(i) the N-selective alkylation of a 6-halopyridone (to establish
C(10)–N(1) bond) and (ii) the intramolecular α-arylation of a
lactam to establish the C(6)–(7) bond and complete the tricyclic
core of the target. This provides a direct and highly convergent
entry to cytisine, and this paper reports the implementation of
this strategy, together with a discussion of the key issues
involved.
The synthetic route is outlined in Scheme 1. The 5-substituted
piperidinone 5 was assembled in 55% overall yield via 4, an
intermediate that was available from commercially available
materials using the three component coupling procedure
described by Stille.9 LiAlH4 reduction of 5 followed by halo-
genation gave bromide 6 in 57% overall yield. The first key
transformation requires an N-selective alkylation of 6-bromo-
pyridone 710 using bromide 6. While this chemistry has good
precedent, use of Curran’s11 optimised conditions for alkyl
halides [6-bromopyridone (1 equiv.), alkyl halide (2 equiv.),
† Electronic supplementary information (ESI) available: Characteriz-
ation data for key intermediates. See http://www.rsc.org/suppdata/ob/
b4/b405275g/
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 8 2 5 – 1 8 2 6
1825

View Article Online
illustrates further the highly sensitive nature of this type of
alkylation and work continues to improve this process, since
one drawback is the requirement for extended reaction times
(7–10 days) for the conversion of 6 to 8.
Notes and references
‡ We explored the alkylation of pyridone 7 with bromide 6 using
Design Expert program (www.statease.com). This generated an eight
factorial set of 20 experiments including 4 centre points which were
carried out using a carousel reactor assembly, and analysis to determine
the extent of conversion and product distribution was carried out by 1H
NMR.
The second key step, the intramolecular lactam α-arylation
of 8, was achieved using Hartwig’s original conditions
(KHMDS, ( )-BINAP)13 and the arylated adduct 11 was
isolated in 44% yield. This transformation has yet to be
fully optimised but we have examined briefly other general
conditions for α-arylation,14 such as those reported by
Buchwald,14a but these were less effective in our case. One
possible issue here relates to the conformational require-
ments associated with this cyclisation step. Intermediate 8
must adopt the axial conformation shown in Scheme 1 in order
for cyclisation to take place. The fate of the equatorial
conformer of 8 under these conditions is unknown, but
may involve intermolecular reactions leading to oligomer
formation. To date, the only characterisable product from this
reaction is the tricycle 8.
The final steps to complete the synthesis of racemic cytisine
are shown in Scheme 1 and involved selective lactam reduction
of 11, which was achieved after some experimentation with
BH3ؒTHF, followed by N-debenzylation to provide ( )-1 in
21% overall yield from 8.§ This final deprotection step was
employed by O’Neill,6a although we have used somewhat
different conditions to those described earlier.
One aspect of this chemistry merits further comment. The
asymmetric synthesis of cytisine has only very recently been
achieved.8 Earlier, Coe6b used the asymmetric variant of the
Heck reaction that underpins his very direct and elegant
approach to cytisine, but this only proceeded in 22% ee. We have
investigated various methods for the catalytic asymmetric
reduction of the unsaturated ester 4 (as well as the corre-
sponding carboxylic acid), but to date we have also only
achieved modest enantiomeric excesses of the order of ≤24%
ee.¶ An efficient asymmetric entry to cytisine via an inter-
mediate such as 5 remains a goal of this programme.
In summary, we have described a novel strategy for
the synthesis of cytisine 1, which has been achieved in a total
of 8 steps from commercially available starting materials.
The route is also highly convergent but it should be stressed
that this sequence has not yet been optimised in terms
of the key transformations involved. Nevertheless, we
anticipate that the strategy outlined above, and exemplified
for cytisine, will be applicable to a range of other related
targets.
§ Synthetic material was compared directly (TLC, IR, 1H and 13C
NMR) to a commercially available sample of (Ϫ)-cytsine.
¶ Reduction of 4 was carried out in 24% ee (based on chiral HPLC)
using H2 (4 atm), Ru[(R,R)-Me-Duphos]CodؒBF4, in MeOH at 45 ЊC.
1 (a) A. Partheil, Arch. Pharm. (Weinheim, Ger.), 1894, 232, 161;
(b) H. Ing, J. Chem. Soc., 1932, 2778.
2 N. J. Leonard, in The Alkaloids, ed. R. H. F. Manske and
H. L. Holmes, Academic Press, New York, 1953, vol. 3, Chapter 19,
p. 119; K. A. Aslanov, Y. K. Kushmuradov and A. S. Sadykov,
The Alkaloids, ed. A. Brossi, Academic Press, New York, 1997,
vol. 31, Chapter 5, p. 117; H. C. S. Wood and R. Wrigglesworth,
Rodd’s Chemistry of Carbon Compounds, 2nd Edn., ed. S. Coffey,
Elsevier, Amsterdam, 1978; Vol. IV, Part H, Chapter 38, p. 285.
3 T. Schütz, Synlett, 2003, 901.
4 (a) L. A. Pabreza, S. Dhawan and K. J. Kellar, Mol. Pharmacol.,
1991, 39, 9; (b) S. F. Heinemann and R. L. Papke, Mol. Pharmacol.,
1994, 45, 142; (c) D. J. Anderson and S. P. Arneric, Eur. J.
Pharmacol., 1994, 253, 261.
5 (a) E. E. van Tamelen and J. S. Baran, J. Am. Chem. Soc., 1955, 77,
4944; (b) E. E. van Tamelen and J. S. Baran, J. Am. Chem. Soc.,
1958, 80, 4659; (c) F. Bohlmann, A. Englisch, N. Ottawa, H. Sander
and W. Weise, Chem. Ber., 1956, 89, 792; (d ) T. R. Govindachari,
S. Rajadurai, M. Subramanian and B. S. Thyagarajan, J. Chem. Soc.,
1957, 3839.
6 (a) B. T. O’Neill, D. Yohannes, M. W. Bundesmann
and E. P. Arnold, Org. Lett., 2000, 2, 4201; (b) J. W. Coe, Org. Lett.,
2000, 2, 4205. A formal total synthesis based on ref. 6a has also
been reported: P. Nshimyumukiza, D. Cahard, J. Rouden,
M.-C. Lasne and J.-C. Plaquevent, Tetrahedron Lett., 2001, 42, 7787.
7 J. W. Coe and P. R. P. Brooks, U.S. Patent 6,605,610, 2003.
8 B. Danieli, G. Lesma, D. Passarella, A. Sacchetti, A. Silvani and
A. Virdis, Org. Lett., 2004, 6, 493.
9 G. R. Cook, L. G. Beholz and J. R. Stille, J. Org. Chem., 1994, 59,
3575.
10 G. R. Newkome, J. Broussard, S. K. Staires and J. D. Sauer,
Synthesis, 1974, 707.
11 H. Liu, S. B. Ko, H. Josien and D. P. Curran, Tetrahedron Lett.,
1995, 36, 8917. A variety of different conditions are presented in this
paper, and a number of these were examined in our study. Alkyl
halides tend to show lower N-selectivities under conditions suited to
more reactive allyl and propargyl halides.
12 M. R. Owen, C. Luscombe, L.-W. Lai, S. Godbert, D. L. Crookes
and D. Emiabata-Smith, Org. Process Res. Dev., 2001, 5, 308.
13 K. H. Shaughnessy, B. C. Hamann and J. F. Hartwig, J. Org.
Chem., 1998, 63, 6546. See also T. Hama, X. Liu, D. A. Culkin and
J. F. Hartwig, J. Am. Chem. Soc., 2003, 125, 11176; D. A. Culkin
and J. F. Hartwig, Acc. Chem. Res., 2003, 36, 234.
14 (a) Buchwald has described the arylation of a vinylogous amide:
T. Hamada, A. Chieffi, J. Ahman and S. L. Buchwald, J. Am. Chem.
Soc., 2002, 124, 1261; (b) We also evaluated an alternative
zinc-based variant that was recently reported, but without success.
See, J. Cossy, A. de Filippis and D. G. Pardo, Org. Lett., 2003, 5,
3037; (c) δ-Lactams have been produced by intramolecular
α-arylation of amide enolates, see T. Honda, H. Namiki and
F. Satoh, Org. Lett., 2001, 3, 631.
Acknowledgements
We thank EPSRC (CMSG) and the EU (Marie Curie
Fellowship to CB), and we acknowledge the use of the
Chemical Database Service and the EPSRC Mass Spectrometry
Swansea Service Centre. We also thank GSK, and Drs Martin
Owen, Gillian Turner and Daniel Tray (all GSK) for their
support with the experimental design study.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 8 2 5 – 1 8 2 6
1826