Regio- and diastereoselective functionalization of (-)-cytisine: An unusual N-C acyl migration

Article abstract of DOI:10.1016/S0957-4166(02)00271-9

In order to test (-)-cytisine as a potential chiral inductor, N-propionyl cytisine was treated with LDA and then with benzyl bromide. Instead of the expected α-benzyl substituted propionamide, (-)-N-benzyl 6α-propionyl cytisine was formed, arising from an

Full text of DOI:10.1016/S0957-4166(02)00271-9

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 13 (2002) 1299–1305
Regio- and diastereoselective functionalization of (−)-cytisine: an
unusual N–C acyl migration
Jacques Rouden,a,* Alexis Ragot,a Sonia Gouault,a Dominique Cahard,b
Jean-Christophe Plaqueventb and Marie-Claire Lasnea
aLaboratoire de Chimie Mole´culaire et Thioorganique UMR CNRS 6507, ISMRA, Universite´ de Caen-Basse Normandie,
6 Boulevard du Mare´chal Juin, 14050 Caen Cedex, France
bIRCOF, UMR CNRS 6014, Universite´ de Rouen, Rue Tesnie`re, 76821 Mont Saint Aignan Cedex, France
Received 22 May 2002; accepted 6 June 2002
Abstract—In order to test (−)-cytisine as a potential chiral inductor, N-propionyl cytisine was treated with LDA and then with
benzyl bromide. Instead of the expected a-benzyl substituted propionamide, (−)-N-benzyl 6a-propionyl cytisine was formed,
arising from an unusual diastereoselective nitrogen to carbon acyl migration. Using LDA in the presence of an excess of LiCl, the
6-substituted cytisine was isolated in yields of up to 79%. The efficiency of the N–C acyl transfer was shown to be dependent on
the nature of the N-acyl group. Complete epimerization of the newly created stereocenter was observed under basic conditions.
This methodology allows the stereoelective functionalization of the C-6 position of cytisine, an important agonist of nicotinic
receptors. © 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
Parkinson’s diseases, cytisine 2 became an important
probe for studying nAChRs. In continuation of our
interest in studying neurotransmission using positron
emission tomography,12 we recently reported a formal
synthesis of cytisine,13 the preparation of several
cytisine derivatives and the radiolabelling of one of
them (Fig. 1).14
The lupin alkaloids are found in a wide variety of
plants, especially within the leguminosae family.1 Most
of the compounds belonging to this rather large group
of alkaloids contain an indolizidine nucleus and the
tetracyclic (−)-sparteine 1 is probably the most well
known member of this group. Because of its commer-
cial availability and low cost in enantiomerically pure
form, the diamine 1 has received considerable attention.
It has been used as a chiral bidentate ligand in asym-
metric reactions such as deprotonation,2 enantioselec-
tive addition of organolithium3 and Grignard reagents,4
carbolithiation of alkenes,5 polymerization of
methacrylate esters6 and palladium-catalyzed allylic
alkylation7 and oxidation protocols.8
Due to the large amounts of cytisine needed for this
synthetic work, we developed an easy method for
extracting cytisine from readily available natural
sources. This chiral alkaloid, which, with its half cage-
like structure, shows obvious similarities to sparteine 1,
has the advantage of bearing a secondary amine func-
tionality and a pyridone ring, both of which can be
involved in metal complexation. Having in hand multi-
gram quantities of this alkaloid, we decided to test
cytisine as a chiral auxiliary.
(−)-Cytisine 2, another lupin alkaloid, was isolated
more than 140 years ago9 and its structure was deter-
mined with certainty in 1932.10 It shows high affinity
and selectivity for the a4b2 nicotinic acetylcholine recep-
tors (nAChRs) of the central nervous system.11 Because
there is accumulating evidence that neuronal nAChRs
are involved in various physiological effects of nicotine
and in several pathologies such as Alzheimer’s and
* Corresponding author. Tel.: +33 2 3145 2893; fax: +33 2 3145 2877;
e-mail: rouden@ismra.fr
Figure 1.
0957-4166/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(02)00271-9

1300
J. Rouden et al. / Tetrahedron: Asymmetry 13 (2002) 1299–1305
2. Results and discussion
on the C6 carbon of cytisine and the benzyl group on
the nitrogen N3. The two conformers arise from the two
possible orientations of the benzyl group. When com-
pound 5a was heated in refluxing aqueous sodium
hydroxide, the 1H NMR spectrum of the resulting
product showed that 5a underwent isomerization to the
b-isomer 5b. Indeed, only one conformer was observed
We chose the alkylation reaction of an amide enolate to
evaluate the chiral induction properties of cytisine. The
secondary amine function of cytisine was readily
acylated with propionyl chloride in the presence of
triethylamine to afford the amide 3. Subsequent depro-
tonation of 3 with 1.1 equiv. of lithium diisopropyl-
amide (LDA) in THF at −78°C followed by the addi-
tion of 2 equiv. of benzyl bromide led mainly to the
recovery of starting material (78%) along with 20% of a
new compound (Scheme 1).
1
in the H NMR spectrum, with the signal for the H6
proton occurring as a singlet at 4.93 ppm. Mass spec-
troscopy and 13C NMR data are in good agreement
with the diastereoisomeric structure 5b.
The absolute configuration of the newly formed C-6
stereogenic center was assigned by comparing the
observed vicinal coupling constants between protons 5
and 6 and those calculated from molecular modelling
simulations (AM1, MacSpartan®). After minimization
of the structures, a dihedral angle H5–C–C–H6 of 44°
was measured for 5a and 74° for 5b (Fig. 2). From
these dihedral angles, the expected coupling constants
JH5–H6 would be 4.12 and 0.36 Hz, respectively.16 In the
1H NMR spectrum of the compound directly generated
Whereas the molecular mass (MW=336) matched the
deprotonated-benzylated product of this new com-
1
pound, the H NMR spectrum presented three interest-
ing features: the methylene signal of the benzyl group
appeared as two superimposed AB quartets (lA: 3.24
and lB: 3.57 ppm, J: 13.2 Hz) with no small coupling.
The resonance of the methyl group in the propionyl
chain is seen as two triplets at 0.92 and 1.14 ppm. This
suggests that no substitution occurred at the a-position
to the carbonyl group and that the compound exists as
a mixture of two conformers (see15 for conformers in
cytisine). This latter hypothesis was confirmed by
3
by the LDA reaction, the coupling constants are JH6–
3
H5=6.0 and 6.6 Hz for the two conformers and JH6–
1
H5=0 Hz on the H NMR spectrum of the compound
obtained from the NaOH epimerization reaction.
1
recording the H NMR spectrum in CDCl3 at different
temperatures. At 50°C, partial coalescence of the
methyl triplets was observed, whereas the methylene
signal of the benzyl group is seen as a single AB
quartet. Finally only one proton appears as two dou-
blets (4.76 and 5.03 ppm, two conformers) in the region
(4.5–5 ppm), which is characteristic of the resonance of
the 6-protons in cytisine and cytisine derivatives. Even
though the 13C NMR spectrum is complicated by the
presence of both conformers, the chemical shift
observed at 209.3 ppm is typical of a ketone function
and not of an amide (in the starting material the
chemical shift of the amide carbonyl group is observed
at 172.8 ppm). From these spectral data we suggest the
structure 5a with the propionyl group now positioned
This discrepancy in one case (the measured coupling is
somewhat larger than predicted) is also observed on
cytisine itself between H5 and H6b.17 Factors other than
simple geometry can modulate vicinal proton couplings
especially in cyclic bridged molecules like cytisine, and
one has to take this into account when using the
equation, as originally discussed by Karplus.18 There-
1
fore by comparing the H NMR spectra of compounds
5a and 5b we can deduce the absolute configuration for
both compounds at C6: H6b for 5a with a pseudo-axial
propionyl chain and H6a for 5b with the propionyl
chain being in a pseudo-equatorial disposition. This is
in good agreement with the experimental data: the
more stable epimer, which should have the propionyl
chain in the more stable pseudo-equatorial position, is
isolated when compound 5a is treated with NaOH
under reflux.
A mechanistic proposal for the formation of 5a is
shown in Scheme 2. An initially formed pyramidal
carbanion at the position a to the pyridone ring might
Scheme 1.
Figure 2.

J. Rouden et al. / Tetrahedron: Asymmetry 13 (2002) 1299–1305
1301
attack the carbon atom of the amide function to give a
strained five-membered ring. This intermediate would
then be stabilized by transferring the propionyl group
to the 6-position. The lithiated amine would then be
alkylated by benzyl bromide to afford the N-benzyl-6-
propionylcytisine 5a. Such lithiation a to the nitrogen
of an N-alkyl pyridone is already documented19 and is
directed by the pyridone carbonyl. We have found only
a few reports in the literature of this unusual kind of
nitrogen to carbon acyl migration.20
fer reaction and subsequent quench with three types of
electrophile, water, methyl iodide or benzyl bromide,
afforded, respectively, the free secondary amine, methyl
and benzyl amine 6-substituted cytisine. The results are
summarized in Table 2.
Alkylcarbonyl or branched alkylcarbonyl groups were
transferred with average to good yields (entries 1–3, 6
and 7). Surprisingly, the N-acetyl cytisine 7 and N-ben-
zoyl cytisine 8 gave no ‘rearranged’ product (data not
shown in Table 2). In those two cases the reaction gave
rise to an inseparable mixture of compounds. The
migration of the methoxycarbonyl group (entries 4 and
5) was an interesting result, mainly because an acetyl
group has better migratory aptitude than a carbalkoxy
group under thermal conditions.23 After quenching with
benzyl bromide or water, cytisine derivatives 11a and
11b bearing an ester group at the C-6 position were
isolated in 65 and 79% yields (Scheme 3). As for the
N-benzyl-6-propionyl cytisine 5a, the N-benzyl-6a-
methoxycarbonyl cytisine 11a was completely epimer-
ized to its thermodynamically more stable 6b-derivative
11c under basic non-hydrolytic conditions (isolated
yield: 57%, Scheme 4). Compounds 11a, 11b and 11c
are potential g-amino-acids with a new type of
geometry.
Table 1 shows the main results for the optimization of
the acyl transfer alkylation reaction. Increasing the
amount of LDA or adding a co-solvent20b (DMPU:
N,N%-dimethylpropyleneurea) did not improve the yield
of 5a significantly. However, the addition of 5–6 molar
equiv. of LiCl to the reaction mixture before deproto-
nation led to a complete conversion and a 75% isolated
yield of the ‘rearranged’ product 5a. The influence of
lithium salts on lithium aggregates21 and on the rates of
enolate alkylation reactions22 is well documented.
We next examined the acyl migration reaction of vari-
ous cytisine amide derivatives. For this purpose other
acyl groups were appended to cytisine according to the
conditions described in Scheme 1 for compound 3
(R1COCl, Et3N). N-Acyl cytisine derivatives 6–10 were
isolated with non-optimized yields ranging from 49 to
91% (see Section 4). Then the deprotonation–acyl trans-
Table 2. LDA/LiCl deprotonation of acyl cytisine, acyl
migration then electrophile quenchinga
Entry
Product
R1
R2
Yieldb (%)
1
2
3
4
5
6
7
5a
5c
5d
11a
11b
12
Et
Et
Et
MeO
MeO
tert-Bu
iso-Pr
Bn
Me
H
Bn
H
75
51
70
65
79
57
65
H
H
13
a The electrophiles used were: R2=Bn, benzyl bromide; R2=Me,
methyl iodide, R2=H, water.
b Isolated yield.
Scheme 2.
Table 1. Optimization of the conditions for the reaction
3“5a
Scheme 3.
Entry
LDAa
(equiv.)
Additives
S.M. 3
Yield 5a
(%)b
(%)c
1
2
3
4
1.1
2.1
2.1
2.1
None
none
DMPU (10%)
LiCl (5–6 equiv.)
78
60
60
0
22
37
35
75
a Conditions: LDA, −78°C, 30 min, then BnBr (3 equiv.), −60°C, 3 h.
b S.M.: starting material; yield of recovered 3.
c Yield of isolated product 5a.
Scheme 4.

1302
J. Rouden et al. / Tetrahedron: Asymmetry 13 (2002) 1299–1305
3. Conclusion
4.2.1. Data for N-propionyl cytisine 3. Isolated as a
white solid after flash chromatography (1.185 g, 96%).
Mp 141–142°C (acetone–Et2O). [h]2D0=−241 (c 1,
CHCl3). 1H NMR l (CDCl3, two conformers): 0.89 and
1.02 (t, 3H, J=7.37 Hz), 1.85–1.98 (m, 1H), 2.01
(broad s, 2H), 2.23–2.31 (m, 1H), 2.53 (broad s, 1H),
2.77–2.87 (m, 1H), 3.08 (broad s, 1H), 3.30–3.38 (m,
1H), 3.81–4.00 (m, 1H), 4.00–4.17 (m, 1H), 4.69 and
4.80 (d, 1H, J=13 Hz), 6.06 (dd, 1H, J=3.1 Hz, J=6.6
Hz), 6.44 (dd, 1H, J=3.1 Hz, J=9.0 Hz), 7.27 (dd, 1H,
J=6.6 Hz, J=9.0 Hz). 13C NMR l (CDCl3, two con-
formers): 8.9, 25.5, 25.8, 27.1 and 27.2, 34.1 and 34.7,
47.4 and 48.4, 48.6, 51.2 and 52.4, 104.7 and 105.6,
116.8 and 117.3, 138.3 and 138.9, 148.4, 162.9, 172.8.
IR (KBr, cm−1): 2978, 2942, 2876, 1644, 1610, 1544,
1472, 1446, 1362, 1228, 1156, 1058, 810. MS (EI) m/z:
246 (M+, 97), 146 (100). HRMS (EI) calcd for
C14H18N2O2 246.1368, found 246.1361.
To summarize, this study demonstrates the possibility
of functionalizing (−)-cytisine 2 on the 6-position via an
unusual type of nitrogen to carbon acyl transfer. These
acyl migrations are completely diastereoselective, even
though an excess of LDA is used and the resulting
carbonyl group is appended in the less stable pseudo-
axial position. The total epimerization of the 6-a-isomer
to the more stable 6-b-compound was demonstrated in
two cases.
The newly synthesized cytisine derivatives are potential
novel ligands for nACh receptors.
4. Experimental
4.1. General methods and starting materials
4.2.2. Data for N-methoxycarbonyl cytisine 6. Isolated
as a pale yellow solid after flash chromatography (1.187
g, 91% yield). Mp 108°C. [h]2D0=−209 (c 0.77, CHCl3).
1H NMR l (CDCl3): 1.91–2.05 (m, 2H), 2.48 (broad s,
1H), 3.06 (broad s, 2H), 3.11 (broad s, 1H), 3.55 (s,
3H), 3.85 (dd, H-6b, J=15.7 Hz, J=6.6 Hz), 4.14 (d,
H-6a, J=15.7 Hz), 4.22 (broad s, 2H), 6.07 (d, 1H,
J=6.7 Hz), 6.42 (dd, 1H, J=9.1 Hz, J=1.4 Hz), 7.29
(dd, 1H, J=6.7 Hz, J=9.1 Hz). 13C NMR l (CDCl3):
25.1, 26.6, 33.8, 48.3, 49.6, 50.5, 52.1, 105.0, 116.5,
138.3, 148.4, 155.5, 162.8. IR (NaCl, cm−1): 2994, 2946,
2864, 1696, 1650, 1546, 1448, 1410, 1346, 1264, 1238,
1188, 1158, 1124, 802, 734. MS (EI) m/z: 248 (M+, 13),
146 (100). Anal. calcd for C13H16N2O3: C, 62.89; H,
6.89; N, 11.28; O, 18.94. Found: C, 62.65; H, 6.74; N,
11.15; O, 19.46%.
The reactions were carried out under argon in flasks
dried in an oven at 110°C. THF was distilled from
sodium/benzophenone directly prior to use. Lithium
chloride was dried with a heat gun under high vacuum.
Diisopropylamine and triethylamine were distilled from
calcium hydride. Methyl iodide and benzyl bromide
were distilled before use. Other chemicals were used as
received. Thin layer chromatography (TLC) was per-
formed using aluminum sheets precoated with silica gel
60 F254 (Merck). Flash chromatography was carried out
on silica gel SI 60 (0.040–0.063 mm, Merck). Melting
points were obtained using a Ko¨fler block apparatus
(uncorrected). Optical rotations were measured using a
Perkin–Elmer 241 polarimeter. Infrared spectra (IR)
were recorded with a FT-IR Perkin–Elmer 684 spec-
trometer. Mass spectra were taken on a Nermag R10
apparatus and are reported as fragmentation in m/z
with relative intensities (%) in parentheses. NMR spec-
tra were recorded on a Bruker Avance DPX-250 (1H at
250.1 MHz; 13C at 62.9 MHz; TMS as internal stan-
dard). Elementary analyses were performed on a Ther-
moquest apparatus by the microanalytical service of the
Laboratory of Molecular and Thioorganic Chemistry
of ISMRA, Caen.
4.2.3. Data for N-acetyl cytisine 7. Isolated as a white
solid after flash chromatography (1.05 g, 86% yield).
Mp 212°C. [h]2D0=−200 (c 1, CHCl3). 1H NMR l
(CDCl3, two conformers): 1.67 (s, 2H), 1.98 (broad s,
3H), 2.46 (broad s, 1H), 2.76 (m, 1H), 3.03 (s, 1H), 3.34
(m, 1H, one conformer), 3.72–3.89 (m, 2H, 1 con-
former, H-6a, two conformers), 4.01 and 4.07 (d, 1H,
J=10.1 Hz, H-6b), 4.60 and 4.72 (d, 1H, J=13.1 Hz),
6.00 (d, 1H, J=6.6 Hz), 6.36 (m, 1H), 7.24 (m, 1H). 13
C
Cytisine
[(−)-(1R,5S)-1,2,3,4,5,6-hexahydro-1,5-
NMR l (CDCl3, two conformers): 19.4 and 20.0, 24.7
and 24.8, 26.0 and 26.3, 33.0 and 33.7, 46.2 and 47.1,
47.5, 51.2 and 52.4, 103.5 and 104.6, 116.0 and 116.5,
137.1 and 137.8, 147.0 and 147.2, 161.9 and 162.1,
168.3 and 168.4. IR (KBr, cm−1): 1634, 1616, 1546,
1452, 1424, 1360, 1238. MS (IE) m/z: 232 (M+, 100),
190 (25), 147 (87). HRMS (EI) calcd for C13H16N2O2
232.1212, found 232.1216.
methanopyrido[1,2-a][1,5]diazocin-8-one] was extracted
from commercially available seeds of Cytisus laburnum
anagyro¨ıdes (Vilmorin Company, France). The proce-
dure was reported in the supporting information (free
of charge via the Internet at http://pubs.acs.org) of a
previous publication.14
4.2. General procedure for the acylation of cytisine 2
To a mixture of cytisine 2 (1 g, 5.26 mmol) and
triethylamine (1.467 mL, 10.5 mmol) in CH2Cl2 (20
mL) was added dropwise acid chloride (7.9 mmol) at
0°C. After stirring overnight at rt, the solvent was
evaporated and the crude product was taken into ethyl
acetate. Triethylamine hydrochloride precipitated and
was removed by filtration. After evaporation of the
solvent, the residue was purified by flash chromatogra-
phy (CH2Cl2/MeOH, 95:5) to give pure N-acylcytisine.
4.2.4. Data for N-benzoyl cytisine 8. Isolated as a white
solid after flash chromatography (1.33 g, 86% yield).
1
Mp 193°C. [h]2D0=−277 (c 1.02, CHCl3). H NMR l
(CDCl3): 1.99 (s, 2H), 2.5 (broad s, 1H), 3.12 (broad s,
3H), 3.85 (broad s, 1H), 4.22 (broad s, 1H), 4.8 (broad
s, 1H), 5.95 (broad s, 1H), 6.55 (d, 1H, J=9 Hz), 6.9
(broad s, 1H), 7.30 (s, 1H). All signals are poorly
defined due to the presence of rotamers. 13C NMR l
(CDCl3): 26.6, 28.0, 35.2, 49.2, 106.2, 118.0, 126.9,

J. Rouden et al. / Tetrahedron: Asymmetry 13 (2002) 1299–1305
1303
128.7, 129.9, 135.6, 139.3, 148.8, 163.7, 171.6. IR (KBr,
cm−1): 2936, 1640, 1614, 1574, 1536, 1440. MS (EI)
m/z: 294 (M+, 19), 189 (17), 146 (31), 105 (100), 77 (49).
Anal. calcd for C18H18N2O2: C, 73.45; H, 6.16; N, 9.52.
Found: C, 73.24; H, 6.29; N, 9.55%.
4.3.1. Data for N-benzyl-6a-propionyl cytisine 5a. The
electrophile added was benzyl bromide. A white solid
was obtained after flash chromatography (CH2Cl2/
MeOH, 95:5) (0.408 g, 75% yield). [h]2D0=−284 (c 1,
1
CHCl3). H NMR (CDCl3, two conformers): 0.92 and
1.14 (2t, 3H, J=7.1 Hz and 6.4 Hz), 1.85 (broad s, 2H),
2.17–3.02 (m, 8H), AB system lA: 3.24 and lB: 3.57
ppm (two superimposed AB quartets due to the two
conformers, J: 13.2 Hz), 4.76 and 5.03 (2d, H-6b,
J=6.6 Hz and 6.0 Hz), 5.87 and 6.00 (2d, 1H, J=6.6
Hz and 6.1 Hz), 6.42–6.50 (m, 1H), 7.05–7.38 (m, 6H).
13C NMR l (CDCl3, two conformers): 7.3, 27.0, 29.7
and 31.8, 30.1 and 33.4, 34.9 and 35.3, 55.4 and 55.9,
57.4 and 58.5, 61.5 and 62.1, 65.2 and 67.4, 104.6 and
105.1, 116.5, 126.3, 126.9, 127.5, 128.0, 128.5, 136.1 and
137.5, 138.5 and 139.3, 150.9 and 151.1, 162.8 and
163.7, 209.3. MS (EI) m/z: 336 (M+, 9), 91 (100). IR
(KBr, cm−1): 2936, 2802, 1706, 1654, 1568, 1546, 1454,
1346. HRMS (EI) calcd for C21H24N2O2 336.1838,
found 336.1835.
4.2.5. Data for N-pivaloyl cytisine 9. Isolated as a white
solid after flash chromatography (1.167 g, 81% yield).
1
Mp 163°C. [h]2D0=−261 (c 0.95, CHCl3). H NMR l
(CDCl3): 1.09 (s, 9H), 2.01 (m, 2H), 2.5 (broad s, 1H),
3.04 (d, 1H, J=3.3 Hz), 3.13 (m, 2H), 3.82 (dd, H-6b,
J=6.4, J=15.7 Hz), 4.19 (d, H-6a, J=15.7 Hz), 4.44
(d, 1H, J=12.7 Hz), 4.61 (d, 1H, J=13.1 Hz), 6.05 (dd,
1H, J=1.1 Hz, J=6.8 Hz), 6.43 (dd, 1H, J=1.1 Hz,
J=9.0 Hz), 7.26 (dd, 1H, J=6.8 Hz, J=9.0 Hz). 13C
NMR l (CDCl3): 26.9, 28.0, 28.5 (3CH3), 35.3, 39.2,
49.1, 50.8, 52.3, 53.8, 105.9, 117.9, 139.0, 148.9, 163.6,
118. IR (KBr, cm−1): 2966, 1928, 2858, 1660, 1616,
1578, 1546, 1476, 1420, 1364, 1342, 1144. MS (EI) m/z:
274 (M+, 18), 189 (14), 160 (21), 147 (100), 109 (28), 83
(45), 55 (45), 42 (65). Anal. calcd for C16H22N2O2: C,
70.04; H, 8.08; N, 10.21. Found: C, 69.93; H, 8.18; N,
10.39%.
4.3.2. Data for N-methyl-6a-propionyl cytisine 5c. The
electrophile added was methyl iodide. A white solid was
obtained after flash chromatography (CH2Cl2/MeOH,
95:5) (0.215 g, 51% yield). Mp 164°C (Et2O:MeOH).
[h]2D0=−333 (c 1, CHCl3). 1H NMR l (CDCl3, two
conformers): 1.01 and 1.19 (t and broad s, 3H, J=6.7
Hz), 1.82 (m, 2H), 2.05 (m, 1H), 2.10 (s, 3H), 2.33–2.67
(m, 5H), 2.98–3.02 (m, 2H), 4.76 and 4.97 (d and broad
s, H-6b, J=6.7 Hz), 6.09 (d, 1H, J=6.5), 6.45 (d, 1H,
J=8.6 Hz), 7.33–7.39 (m, 1H). 13C NMR l (CDCl3,
two conformers): 7.5, 26.7, 30.2, 31.8, 33.7, 35.7, 45.5,
57.1, 61.6, 65.4, 67.3, 105.2, 116.9, 139.4, 151.3, 163.9,
209.9. IR (KBr, cm−1): 2970, 2928, 2798, 1708, 1654,
1572, 1548, 1454, 1142, 802. MS (EI) m/z: 260 (M+,
15), 203 (100). Anal. calcd for C15H20N2O2: C, 69.20;
H, 7.74; N, 12.29; O, 10.76. Found: C, 69.29; H, 7.81;
N, 12.32; O, 10.58%.
4.2.6. Data for N-(2-methylpropionyl)cytisine 10. Iso-
lated as a white solid after flash chromatography (0.672
g, 49% yield). Mp 143–144°C. [h]2D0=−168 (c 0.28,
1
CHCl3). H NMR l (CDCl3): 0.67 (s, 1H), 1.00 (d, 6H,
J=6.5 Hz), 2.03 (s, 2H), 2.53 (broad s, 1H), 2.81 (m,
1H), 3.09 (s, 1H), 3.37 (d, 1H, J=10.9 Hz), 3.83 (dd,
H-6b, J=6.4 Hz, J=15.7 Hz), 4.1 (m, 2H), 4.77 (m,
1H), 6.06 (d, 1H, J=6.8 Hz), 6.42 (d, 1H, J=9.0 Hz),
7.27 (dd, 1H, J=6.8 Hz, J=9.0 Hz). 13C NMR l
(CDCl3): 19.04, 26.28, 27.48, 29.50, 35.04, 48.83, 52.51,
104.98, 117.17, 117.26, 138.41, 148.51, 163.17, 176.38.
IR (KBr, cm−1): 2964, 2926, 2868, 1646, 1606, 1574,
1544, 1476, 1446. MS (EI) m/z: 260 (M+, 21), 190 (20),
147 (40), 109 (20), 82 (30), 43 (100). HRMS (EI) calcd
for C15H20N2O2 260.1524, found 260.1508.
4.3.3. Data for 6a-propionyl cytisine 5d. The electrophile
added was water as an aqueous ammonium chloride
solution. A white solid was obtained after flash chro-
matography (CH2Cl2/MeOH, 93:7) (0.279 g, 70%
yield). Mp 133°C (Et2O:MeOH). [h]2D0=−96 (c 1,
4.3. General procedure for the acyl transfer
To a mixture of lithium chloride (0.5 g, 11.79 mmol)
and diisopropylamine (0.487 mL, 3.41 mmol) was
added dropwise butyllithium (1.6 M solution in hex-
anes, 2.12 mL, 3.41 mmol) at −20°C under argon. After
stirring 10 min 5 mL of THF was added and the
temperature was cooled to −78°C. Then a solution of
the acyl cytisine derivative (compounds 3 and 6–10)
(1.62 mmol) in THF (10 mL) was transferred via can-
nula to the LDA solution at −78°C. After stirring for
30 min at −78°C, the electrophile (4.87 mmol, 1.43
equiv.) was added and the reaction was stirred for a
further 3 h at −78°C. The reaction was quenched with
a saturated aqueous ammonium chloride solution and
then ammonium hydroxide (28% aqueous solution) was
added to basify the medium. Extraction of the aqueous
phase was carried out with CH2Cl2 several times and
the combined organic layers were dried over sodium
sulfate, filtered and concentrated under vacuum. The
residue was purified by flash chromatography to give
the pure rearranged product.
1
CHCl3). H NMR l (CDCl3): 1.20 (t, 3H, J=7.2 Hz),
1.85 (broad s, 1H), 1.95–2.15 (m, 2H), 2.43 (broad s,
1H), 2.48–2.64 (m, 1H), 2.80 (broad s, 2H), 2.90 (broad
s, 1H), 2.96–3.06 (m, 1H), 3.09 (broad s, 2H), 5.00 (d,
H-6b, J=6.7 Hz), 6.07 (dd, 1H, J=6.8 Hz, J=1 Hz),
6.44 (dd, 1H, J=1 Hz, J=9.1 Hz), 7.35 (dd, 1H, J=6.8
Hz, J=9.1 Hz). 13C NMR l (CDCl3): 7.1, 26.9, 28.9,
33.9, 35.1, 48.8, 52.2, 65.0, 105.3, 116.6, 139.3, 150.7,
162.8, 206.4. IR (KBr, cm−1): 3422, 2926, 2856, 2362,
1718, 1652, 1544, 1456, 1210, 1148, 802. MS (EI) m/z:
246 (M+, 40), 146 (100). HRMS (EI) calcd for
C14H18N2O2 246.1368, found 246.1371.
4.3.4. Data for N-benzyl-6a-methoxycarbonyl cytisine
11a. The electrophile added was benzyl bromide. A
viscous pale yellow oil was obtained after flash chro-
matography (CH2Cl2/MeOH, 95:5) (0.547 g, 65%
1
yield). [h]2D0=−318 (c 1, CHCl3). H NMR l (CDCl3):
1.86–1.87 (m, 2H), 2.24 (d, 1H, J=9.3 Hz), 2.40 (d, 1H,

1304
J. Rouden et al. / Tetrahedron: Asymmetry 13 (2002) 1299–1305
J=10.3 Hz), 2.74 (broad s, 2H), 2.85–2.90 (m, 2H),
3.27 (d, 1H, J=13.3 Hz), 3.49 (d, 1H, J=13.3 Hz), 3.57
(broad s, 1H), 3.72 (broad s, 2H), 4.90 (d, H-6b, J=5.3
Hz), 5.87 (d, 1H, J=6.7 Hz), 6.46 (dd, 1H, J=1.4 Hz,
J=9.1 Hz), 7.15–7.23 (m, 5H), 7.28 (dd, 1H, J=1.4 Hz,
J=6.7 Hz). 13C NMR l (CDCl3, two conformers):
26.4, 30.2, 30.9, 34.9, 51.4, 56.1, 57.7, 60.2, 61.5, 104.5,
116.5, 126.4, 127.5, 128.0, 128.5, 134.0, 137.2, 138.6,
150.6, 163.0, 167.7, 170.5. IR (KBr, cm−1): 2944, 2800,
2764, 1760, 1730, 1656, 1572, 1546, 1494, 1452, 1436,
1346, 1280, 1192, 1168, 1144, 800. MS (EI) m/z: 338
(M+, 23), 84 (100). HRMS (EI) calcd for C20H22N2O3
338.1630, found 338.1626.
18.0, 19.4, 27.3, 29.6, 35.4, 39.9, 49.3, 52.5, 64.7, 105.7,
117.0, 139.5, 151.1, 163.1, 209.8. IR (NaCl, cm−1): 2968,
2934, 2868, 1712, 1654, 1546. MS (EI) m/z: 260 (M+,
100), 230 (10), 160 (15), 123 (25), 82 (15). HRMS (EI)
calcd for C15H20N2O2 260.1524, found 260.1503.
4.4. Isomerization of 6a to 6b cytisine acyl derivatives
4.4.1. Data for N-benzyl-6b-propionyl cytisine 5b. N-
Benzyl-6a-propionyl cytisine 5a (0.2 g, 0.59 mmol) was
added to a 15% aqueous NaOH solution (10 mL). After
heating the mixture under reflux overnight, the mixture
was extracted three times with CH2Cl2. The combined
organic layers were dried over sodium sulfate, filtered
and concentrated under vacuum. The residue was
purified by flash chromatography (CH2Cl2/MeOH,
95:5) to give the C-6 epimerized product as a colorless
viscous oil (0.170 g, 85% yield). [h]2D0=−228 (c 1,
4.3.5. Data for 6a-methoxycarbonyl cytisine 11b. The
electrophile added was water as an aqueous ammonium
chloride solution. A pale yellow solid was obtained
after flash chromatography (CH2Cl2/MeOH, 93:7)
1
1
(0.317 g, 79% yield). [h]2D0=−151 (c 1.12, CHCl3). H
CHCl3). H NMR l (CDCl3): 1.12 (t, 3H, J=7.3 Hz),
1.60 (dt, 1H, J=3.0 Hz, J=12.9 Hz), 1.99–2.05 (dt, 1H,
J=3.0 Hz, J=12.90 Hz), 2.22 (broad s, 1H), 2.31 (dd,
1H, J=1.9 Hz, J=10.7 Hz), 2.45 (dd, 1H, J=2.2 Hz,
J=11.1 Hz), 2.63 (dq, 1H, J=7.3 Hz, J=18.0 Hz), 2.89
(dq, 1H, J=7.3 Hz, J=18.0 Hz), 2.80–2.85 (m, 1H),
2.95–3.04 (m, 2H), 3.39 (d, 1H, J=13.6 Hz), 3.47 (d,
1H, J=13.6 Hz), 4.93 (s, H-6a), 5.97 (dd, 1H, J=1.2
Hz, J=6.9 Hz), 6.46 (dd, 1H, J=1.2 Hz, J=9.0 Hz),
6.95–6.99 (m, 2H), 7.16–7.22 (m, 3H), 7.31 (dd, 1H,
J=6.9 Hz, J=9.0 Hz). 13C NMR l (CDCl3): 7.4, 22.8,
29.8, 32.6, 34.9, 59.6, 59.8, 61.6, 65.9, 104.5, 116.3,
126.8, 127.9, 128.1, 137.7, 139.1, 151.2, 162.8, 206.7. IR
(KBr, cm−1): 2936, 2796, 1722, 1654, 1568, 1544, 1450,
1352, 1314, 1250, 1144, 1114, 800. MS (EI) m/z: 336
(M+, 95), 279 (100), 146 (20), 91 (94). HRMS (EI) calcd
for C21H24N2O2 336.1838, found 336.1835.
NMR l (CDCl3): 1.94–2.00 (m, 2H), 2.10 (dt, 1H,
J=2.9 Hz, J=13.0 Hz), 2.49 (broad s, 1H), 2.89 (s,
3H), 3.10 (broad s, 2H), 3.73–3.85 (m, 3H), 4.86 (d,
H-6b, J=6.9 Hz), 6.07 (dd, 1H, J=1.2 Hz, J=6.9 Hz),
6.47 (dd, 1H, J=1.2 Hz, J=9.1 Hz), 7.35 (dd, 1H,
J=9.1 Hz, J=6.9 Hz). 13C NMR l (CDCl3): 26.4, 29.2,
35.0, 49.0, 51.9, 52.1, 60.3, 105.2, 116.6, 139.2, 150.2,
163.0, 170.1. IR (NaCl, cm−1): 3458, 3330, 3040, 2944,
2862, 1732, 1652, 1558, 1538, 1456, 1358, 1270, 1158,
1102, 1066, 800. MS (EI) m/z: 248 (M+, 74), 216 (45),
189 (22), 174 (14), 160 (18), 146 (100). HRMS (EI)
calcd for C13H16N2O3 248.1161, found 248.1167.
4.3.6. Data for 6a-(2,2-dimethylpropionyl)cytisine 12.
The electrophile added was water as an aqueous ammo-
nium chloride solution. A pale yellow solid was
obtained after flash chromatography (CH2Cl2/MeOH,
93:7) (0.253 g, 57% yield). Mp 193°C. [h]2D0=−53 (c 0.7,
4.4.2. Data for N-benzyl-6b-methoxycarbonyl cytisine
11c. N-Benzyl-6a-methoxycarbonyl cytisine 11a (0.2 g,
0.59 mmol) was added to a solution of NaH (0.042 g,
1.77 mmol) in THF (10 mL) in the presence of a
catalytic amount of 18-crown-6 (10 mol% wrt the
cytisine derivative). After heating under reflux for 72 h,
the mixture was quenched with water and extracted
three times with CH2Cl2. The combined organic layers
were dried over sodium sulfate, filtered and concen-
trated under vacuum. The residue was purified by flash
chromatography (CH2Cl2/MeOH, 95:5) to give the C-6
epimerized product as a colorless viscous oil (0.115 g,
1
CHCl3). H NMR l (CDCl3): 1.37 (s, 9H), 1.94 (broad
s, 1H), 2.01 (d, 1H, J=2.7 Hz), 2.12 (dt, 1H, J=3 Hz,
J=12.9 Hz), 2.35 (broad s, 1H), 2.78 (dd, 1H, J=2.2
Hz, J=14.4 Hz), 2.88 (d, 2H, J=2 Hz), 3.11 (d, 2H,
J=2.2 Hz), 5.34 (d, H-6b, J=6.3 Hz), 6.06 (dd, 1H,
J=1.1 Hz, J=7 Hz), 6.43 (dd, 1H, J=1.1 Hz, J=9
Hz), 7.33 (dd, 1H, J=7 Hz, J=9 Hz). 13C NMR l
(CDCl3): 27.5, 27.9, 29.6, 35.4, 44.1, 49.4, 52.8, 61.9,
105.8, 117.10, 139.5, 151.4, 162.8, 211.8. IR (KBr,
cm−1): 2960, 2950, 2928, 1696, 1647, 1561, 1542. MS
(EI) m/z: 274 (M+, 23), 217 (95), 189 (57), 146 (30), 57
(51), 44 (88), 43 (57), 42 (100). HRMS (EI) calcd for
C16H22N2O2 274.1681, found 274.1677.
1
57% yield). [h]2D0=−141 (c 0.45, CHCl3). H NMR l
(CDCl3): 1.67 (dt, 1H, J=12.9 Hz, J=2.9 Hz), 2.09
(broad d, 1H, J=12.9 Hz), 2.31 (dd, 1H, J=10.7 Hz,
J=1.9 Hz), 2.42 (dd, 1H, J=11.1 Hz, J=2.4 Hz), 2.45
(broad s, 1H), 2.82 (m, 1H), 2.97–3.07 (m, 2H), 3.38 (d,
1H, J=13.6 Hz), 3.45 (d, 1H, J=13.6 Hz), 3.75 (s, 3H),
4.88 (s, 1H), 5.97 (dd, 1H, J=6.9 Hz, J=1.2 Hz), 6.51
(dd, 1H, J=9.1 Hz, J=1.2 Hz), 6.93–6.97 (m, 2H),
7.16–7.21 (m, 3H), 7.33 (dd, 1H, J=9.1 Hz, J=6.9 Hz).
13C NMR l (CDCl3): 23.8, 31.9, 35.3, 52.4, 59.5, 60.0,
61.2, 61.8, 104.6, 116.9, 127.0, 128.1, 128.2, 137.7,
139.2, 150.7, 163.2, 171.6. IR (KBr, cm−1): 3018, 2952,
2804, 1748, 1734, 1654, 1570, 1548, 1354, 1282, 1216,
1174, 770. MS (EI) m/z: 338 (M+, 23), 234 (89), 220
(22), 105 (73), 84 (100). HRMS (EI) calcd for
C20H22N2O3 338.1630, found 338.1638.
4.3.7. Data for 6a-(2-methylpropionyl)cytisine 13. The
electrophile added was water as an aqueous ammonium
chloride solution. A yellow viscous oil was obtained
after flash chromatography (CH2Cl2/MeOH, 93:7)
1
(0.274 g, of 65% yield). [h]2D0=−52 (c 0.75, CHCl3). H
NMR l (CDCl3): 1.19 (d, 3H, J=6.9 Hz), 1.38 (d, 3H,
J=6.9 Hz), 2.01 (broad s, 2H), 2.11 (dt, 1H, J=2.9 Hz,
J=12.9 Hz), 2.46 (broad s, 1H), 2.081 (broad s, 2H),
2.89 (broad s, 1H), 3.02 (m, 1H), 3.09 (broad s, 2H),
5.12 (d, H-6b, J=6.6 Hz), 6.06 (dd, 1H, J=1.2 Hz,
J=6.6 Hz), 6.42 (dd, 1H, J=1.2 Hz, J=9 Hz), 7.27
(dd, 1H, J=6.6 Hz, J=9 Hz). 13C NMR l (CDCl3):

J. Rouden et al. / Tetrahedron: Asymmetry 13 (2002) 1299–1305
1305
Acknowledgements
Labelled Compd. Radiopharm. 2001, 44, S128–S131; (c)
Gue´ret, C.; Le Bas, M.-D.; Perrio, C.; Dauphin, F.;
Dhilly, M.; Chazaviel, L.; Emond, P.; Chalon, S.; Tym-
ciu, S.; Guilloteau, D.; Lasne, M.-C.; Barre´, L. J.
Labelled Compd. Radiopharm. 2001, 44, S283–285.
13. Nshimyumukiza, P.; Cahard, D.; Rouden, J.; Lasne,
M.-C.; Plaquevent, J.-C. Tetrahedron Lett. 2001, 42,
7787–7790.
This work was supported by the Re´seau Interre´gional
de Chimie Organique Fine, Contrat de Plan Etat-
Grand-Bassin Parisien-Re´gions Haute Normandie,
Basse Normandie.
14. Marrie`re, E.; Rouden, J.; Tadino, V.; Lasne, M.-C. Org.
Lett. 2000, 2, 1121–1124.
15. Gornicka, E.; Raczynska, E. D. Talanta 2002, 57, 609–
616.
16. Spectroscopic Methods in Organic Chemistry; Hesse, M.;
Meier, H.; Zeeh B., Eds., Engl. translation by Linden, A.;
Murray, M.; Thieme Verlag: Stuttgart, 1997; pp. 106–
112.
17. Mascagni, P.; Christodoulou, M.; Gibbons, W. A.; Asres,
K.; Phillipson, J. D.; Niccolai, N.; Mangani, S. J. Chem.
Soc., Perkin Trans. 2 1987, 1159–1165.
References
1. Leonard, N. J. In The Alkaloids; Manske, R. H. F.;
Holmes, H. L., Eds.; Academic Press: New York, 1953;
Vol. 3, pp. 119–199.
2. (a) Hoppe, D.; Hense, T. Angew. Chem., Int. Ed. Engl.
1997, 36, 2282–2316; (b) Nakamura, S.; Nakagawa, R.;
Watanabe, Y.; Toru, T. J. Am. Chem. Soc. 2000, 122,
11340–11347.
3. Chrzanowska, M.; Sokolowska, J. Tetrahedron: Asymme-
try 2001, 12, 1435–1440 and references cited therein.
4. (a) Trost, B. M.; Fraisse, P. L.; Zachary, T. Angew.
Chem., Int. Ed. Engl. 2002, 41, 1057–1059; (b) Nozaki,
H.; Aratani, T.; Toraya, T.; Noyori, R. Tetrahedron 1971,
27, 905–913.
5. (a) Gil, G. S.; Groth, U. M. J. Am. Chem. Soc. 2000, 122,
6789–6790; (b) Bailey, W. F.; Mealy, M. J. J. Am. Chem.
Soc. 2000, 122, 6787–6788; (c) Marek, I. J. Chem. Soc.,
Perkin Trans. 1 1999, 535–544.
18. Karplus, M. J. Am. Chem. Soc. 1963, 85, 2870–2871.
19. Katritzky, A. R.; Arrowsmith, J.; Bahari, Z. B.; Jayaram,
C.; Siddiqui, T.; Vassilatos, S. J. Chem. Soc., Perkin
Trans. 1 1980, 2851–2855.
20. (a) Chayer, S.; Schmitt, M.; Collot, V.; Bourguignon,
B.-G. Tetrahedron Lett. 1998, 39, 9685–9688; (b) Hara,
O.; Ito, M.; Hamada, Y. Tetrahedron Lett. 1998, 39,
5537–5540; (c) Trapani, G.; Reho, A.; Latrofa, A.; Mor-
lacchi, F.; Liso, G. J. Chem. Res. Miniprint 1986, 3,
954–976; (d) Kurebayashi, Y.; Ishikawa, Y.; Terao, Y.;
Sekiya, M. Tetrahedron Lett. 1981, 22, 123–126; (e) Yat-
sunami, T.; Iwasaki, S. Helv. Chim. Acta 1978, 61, 2823–
2830; (f) Hermecz, I.; Meszaros, Z.; Vasvari-Debreczy,
L.; Horvath, A.; Horvath, G.; Pongor-Csakvari, M. J.
Chem. Soc., Perkin Trans. 1 1977, 789–795; (g) Franck-
Neumann, M.; Dietrich-Buchecker, C. Tetrahedron Lett.
1976, 2069–2072; (h) Akasaki, Y.; Ohno, A. J. Am.
Chem. Soc. 1974, 96, 1957–1959.
6. Okamoto, Y.; Suzuki, K.; Ohta, K.; Hatada, K.; Yuki,
H. J. Am. Chem. Soc. 1979, 101, 4763–4765.
7. Togni, A. Tetrahedron: Asymmetry 1991, 2, 683–690.
8. (a) Jensen, D. R.; Pugsley, J. S.; Sigman, M. S. J. Am.
Chem. Soc. 2001, 123, 7475–7476; (b) Ferreira, E. M.;
Stoltz, B. M. J. Am. Chem. Soc. 2001, 123, 7725–7726.
9. Husemann, A.; Marme´, W. Z. Chem. 1865, 1, 161.
10. (a) Spa¨th, E.; Galinovsky, F. Ber. 1932, 65, 1526–1535;
(b) Ing, H. R. J. Chem. Soc. 1932, 2778–2780.
21. (a) Romesberg, F. E.; Collum, D. B. J. Am. Chem. Soc.
1994, 116, 9187–9197; (b) Hall, P. L.; Gilchrist, J. H.;
Collum, D. B. J. Am. Chem. Soc. 1991, 113, 9571–9574.
22. (a) Myers, A. G.; Yang, B. H.; Chen, H.; Gleason, J. L.
J. Am. Chem. Soc. 1994, 116, 9361–9362; (b) Seebach, D.
Angew. Chem., Int. Ed. Engl. 1988, 27, 1624–1654.
23. Schiess, P.; Fu¨nfschilling, P. Tetrahedron Lett. 1972,
5195–5198.
11. (a) Anderson, D. J.; Arneric, S. P. Eur. J. Pharmacol.
1994, 253, 261–267; (b) Hall, M.; Zerbe, L.; Leonard, S.;
Freedman, R. Brain Res. 1993, 600, 127–133; (c) Glen-
non, R. A.; Dukat, M. Med. Chem. Res. 1996, 465–486.
12. (a) Marrie`re, E.; Chazalviel, L.; Dhilly, M.; Toutain, J.;
Perrio, C.; Dauphin, F.; Lasne, M. C. J. Labelled Compd.
Radiopharm. 1999, 42, S69–S71; (b) Lemoucheux, L.;
Rouden, J.; Ibazizene, M.; Sobrio, F.; Lasne, M.-C. J.