A Modular Approach to the Asymmetric Synthesis of Cytisine

Article abstract of DOI:10.1002/ejoc.201501435

The asymmetric synthesis of (+)-and (-)-cytisine starts with Matteson homologations for the construction of a chiral C3-building block. Conversion of the C3-building block into a dihydropyridone is achieved by straightforward functional group interconversions and ring closing metathesis. After bromination, this central building block was diastereospecifically converted into cytisine in five steps.

Full text of DOI:10.1002/ejoc.201501435

DOI: 10.1002/ejoc.201501435
Full Paper
Asymmetric Synthesis
A Modular Approach to the Asymmetric Synthesis of Cytisine
Felix R. Struth[a] and Christoph Hirschhäuser*[a]
Dedicated to the memory of Dr. Uwe Kayser (1940–2013)
Abstract: The asymmetric synthesis of (+)- and (–)-cytisine tional group interconversions and ring closing metathesis. After
starts with Matteson homologations for the construction of a
chiral C3-building block. Conversion of the C3-building block
into a dihydropyridone is achieved by straightforward func-
bromination, this central building block was diastereospecifi-
cally converted into cytisine in five steps.
Introduction
Stead and O'Brien concluded their 2007 review[6a] on synthe-
ses of cytisine by highlighting two remaining challenges: (i) the
need for a synthesis “that delivers a late stage intermediate that
is equipped with appropriate functionality for analogue prepara-
tion,” and (ii) “an efficient asymmetric synthesis of cytisine”.
Against this background, a modular synthesis has been pub-
lished by Gallagher in 2011;[6b] this work specifically addresses
the need for derivatives with a core-modified pyridone motif.
The Gallagher approach (Scheme 1) relies on a common di-
hydropyridone precursor (rac-3) and its conversion to bromide
rac-4. Subsequent bromide coupling with appropriate hetero-
aryl-stannanes provides access to compounds of type rac-5. De-
pending on the heteroaryl units used, these intermediates were
easily converted into rac-cytisine or related derivatives in high
yields.
Cytisine (1) is a member of the lupin alkaloid family, which can
be isolated from parts (preferably the seeds) of the common
laburnum tree.[1] Although it was first isolated in 1862,[2] the
structure of this bispidine derivative eluded chemists for 70
years.[3] After the discovery of its remarkable pharmacological
profile as a selective, but partial agonist at the α4ꢀ2 subtype
nicotinic acetylcholine receptor,[4] interest in direct chemical
modification of the natural product[5] as well as its (diversity
orientated) synthesis[6] has strongly increased. Furthermore,
cytisine (1) has served as a lead structure in the development
of Pfizer's smoking cessation agent varenicline (2) (Figure 1).[7]
Figure 1. Structure of lupin alkaloid cytisine 1, and smoking cessation agent
Varenicline 2.
Recent investigations into the mechanistic details of the ac-
tivity profile for 1 have employed (i) the synthesis and subse-
quent testing of artificial ligands,[4a,5a–5j,6c–6f] (ii) in silico experi-
ments,[8] as well as (iii) co-crystallization and X-ray analysis us-
ing suitable models for nicotinic acetylcholine receptors like the
Aplysia californica acetylcholine-binding protein (AChBP).[9]
These results indicate that the C-ring piperidine moiety of the
partial agonist is a highly conserved region, whereas the pyr-
idone moiety is amenable to some degree of structural varia-
tion.[10]
Scheme 1. Key steps of the 2011 Gallagher cytisine synthesis.[6b]
Although the Gallagher synthesis provides access to core-
modified derivatives in the variable pyridone region, their work
does not address the 2nd challenge outlined by Stead and O′
Brien.[6a] We now report a synthetic strategy to cytisine that
succeeds in combining (and enhancing) the modular nature of
Gallagher's approach and also provides asymmetric access to
both enantiomers.[11]
[a] Institut für Organische Chemie, Universität Duisburg-Essen,
Universitätsstraße 7, 45141 Essen, Germany
Results and Discussion
E-mail: Christoph.Hirschhaeuser@uni-due.de,
Although the endgame of Gallagher's synthesis is high yielding
in its final steps, it relies on a moderately effective selenoxide
elimination (50 % yield) to produce rac-3 from the correspond-
ing saturated lactam and subsequent bromine addition/elimi-
https://www.uni-due.de/akschmuck/ch_home.php
Supporting information and ORCID(s) from the author(s) for this article
are available on the WWW under http://dx.doi.org/10.1002/
ejoc.201501435.
Eur. J. Org. Chem. 2016, 958–964
958
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
nation to produce rac-4 (63 % yield). In addition, cleavage of room temperature this ate complex undergoes a 1,2-rearrange-
the OTBS-ether was observed regularly in the latter reaction. ment to deliver α-bromoboronate 9 in quantitative yield. Inter-
Thus, channeling a precious, enantiomerically pure substrate estingly, the manner of ZnCl2 addition was found to be crucial.
through this bottleneck is unattractive.[12] We reasoned that When a commercially available solution of ZnCl2 (1
M in Et2O)
avoiding the SeO-elimination altogether and using a more sta- was added 0.5–1 h after the addition of LDA, as described re-
ble protecting group might improve yields. Consequently, our cently for similar systems,[17] the de did not exceed 80 %. Only
retrosynthetic analysis starts with benzyl ether 3a as outlined by strictly following the procedure originally published by Mat-
in Scheme 2. First, clear cut disconnections are used to open teson for this substrate,[15] was the reported diastereoselectivity
the ring (RCM from 6) and to then reduce the size of the mol- (de > 94 %) achieved. This approach required removal of the
ecule (SN2 next to N), leading to precursor 7. Although this volatile constituents at <0 °C following formation of the ate
asymmetrically protected diol is still relatively complex,[13] it is complex and before addition of ZnCl2 solution at –78 °C.[18]
easily attained from C1-fragments using Matteson homologa- This consideration needs to be especially highlighted since the
tions.[14]
diastereomers are not distinguishable in the 1H NMR (300 MHz)
of 9, but can be identified by 13C NMR when an appropriate
signal to noise ratio is achievable.[19] The subsequent substitu-
tion with vinylmagnesium bromide yielded C2-building block
10 in 82 % yield over both steps. Initially, yields for this reaction
varied between 67 % and 88 %, and, in those cases where lower
yields were obtained, pinanediol vinylboronate was identified
as the major byproduct (up to 25 %). Two relevant factors for
this variation became apparent: (i) the age/quality of the vinyl-
magnesium bromide solution, and (ii) the ability to maintain
strict control over the reaction temperature. Temperature con-
trol was achieved by transferring the reaction mixture to a
freezer (–20 °C) following addition of the Grignard reagent at
–78 °C instead of simply allowing the cooling bath to warm up
to room temperature overnight. In employing this method
yields > 80 % can be achieved reproducibly. The de for this
reaction was checked at a later stage (vide infra). For the prepa-
ration of 11, reaction of allylboronate 10 with a suitable carben-
oid (LiCH2X) and subsequent oxidation was necessary. Initially,
this was considered a rather critical step since allylboronates
are known to rearrange at elevated temperatures,[20] and the
only report of an attempt to react a moderately complex allyl-
boronate derivative under similar conditions was unsuccess-
ful.[21] Indeed, this reaction was initially quite capricious, often
yielding < 50 % of 11 after chromatography, although it did
appear to be relatively pure on the basis of TLC and 1H NMR of
the crude. A possible explanation for this observation is shown
in Scheme 4. We considered that, upon hydrolysis, oxidation
product 12 may very well form 11 as well as some stable borate
esters[22] such as 13, which would be stabilized by intramolecu-
lar interactions. To break down these side products we stirred
the crude product in Et2O/CyHex with pinanediol (same
Scheme 2. Retrosynthetic analysis of dihydropyridone 3a.
Thus, the synthesis of 3a, as shown in Scheme 3, com-
menced with C1-building block 8,[15] which was easily prepared
from the corresponding pinacol-boronate and (–)-pinanediol
(see Supporting Information) with no need for chromatography.
Pinanediol was made from α-pinene, both enantiomers of
which are cheaply available in enantiomerically enriched form.
The ee of 8 was determined by HPLC as 77 %, corresponding
well to the optical purity of commercially available (–)-α-pinene
(determined to be 79 %). If higher ee values are desired, enan-
tiomerically pure or nearly pure pinanediol is commercially
available or can be obtained by recrystallization from hept-
ane.[16]
enantiomer) and aq. KOH (5 M) in order to form the more stable
Scheme 3. Synthesis of pivotal precursor 3a.
Addition of LDA to a solution of 8 and CH2Br2 at –78 °C
generates the carbenoid LiCHBr2 which forms an ate complex
with 8. Following the addition of ZnCl2 and warming up to
Scheme 4. Proposed mechanism for the hydrolysis of oxidation product 12.
Eur. J. Org. Chem. 2016, 958–964
www.eurjoc.org
959
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
potassium bispinanediol borate 14. As this white precipitate Information).[25] After derivatization with (S)-(+)-O-acetyl-
can be simply removed by filtration and used again for the mandelic acid, an ee of > 91 % was determined (see Supporting
preparation of pinanediol boronates[23] this method potentially Information) for ent-11. Notably, even better results are
allows for the recycling of the chiral director, but more impor- probably possible using C2-symmetric chiral directors such as
tantly, the isolation of alcohol 11 in 81 % yield.
DIPED or DICHED, which usually facilitate degradation of un-
wanted diastereomers.[14]
To obtain metathesis precursor 6, the C3-building block 11
was first tosylated and then reacted with an excess of benzyl-
amine. Isolation of the resulting amine required minimal
workup, although the removal of excess benzylamine by distil-
lation prior to the reaction with acryloyl chloride was necessary
to suppress formation of undesired polymerization products.
Building block 6 was obtained in 73 % yield using this ap-
proach. Ring closing metathesis furnished desired dihydropyr-
idone 3a in 74 % yield and with an overall yield of 36 % from
C1 building block 8, requiring column chromatography only
four times.
Conclusions
In summary, we have developed an enantioselective synthesis
for cytisine that proceeds in 13 linear steps, starting from
pinanediol both enantiomers of which are commercially avail-
able. The overall yield is 9.7 %. Key to our approach is the syn-
thesis of bromo dihydropyridone 4a from chiral C3-building
block 11, which was prepared by iterative boronate homologa-
tions. Although 4a contains only the atoms ultimately compos-
ing the C-ring piperidine moiety of cytisine (a biologically
highly conserved region) it can be converted into target com-
pound 1 in 5 steps with 34 % overall yield. This chiral precursor
should therefore be a useful tool for the synthesis of highly
enantiomerically enriched derivatives bearing modifications in
the biologically important pyridine moiety.
With the enantiomerically enriched precursor 3a in hand, the
synthesis of cytisine was completed as shown in Scheme 5.
Experimental Section
General: All reactions were carried out in dried glassware under
argon. Solvents for chromatography, unless purchased as pro anal-
ysi (p. a.) grade, were distilled from a rotary evaporator before use.
THF was always freshly distilled from sodium/benzophenone.
MeCN, DMF and DCM were distilled from CaH2 and stored under
argon over appropriate molecular sieves. Diisopropylamine was dis-
tilled from CaH2 and stored in a Schlenk tube over 4 Å molecular
sieves. BuLi, LiHMDS, vinylmagnesium bromide and ZnCl2 were pur-
chased as solutions and stored under argon at +7 °C. The precipi-
tate formed by vinylmagnesium bromide under these conditions
was carefully dissolved by gentle heating and shaking before use.
CuCl was purchased in > 99 % purity and stored at +7 °C in a
Schlenk tube under argon and shielded from light. LiCl was dried
by vigorously heating (heatgun) under vacuum (<5 mbar) and was
subsequently stored under argon. (+)- and (–)-(α)-Pinene were pur-
chased from Acros: (+)-(α)-Pinene: [α]D22 = +42.71 (neat) ־81 % ee //
(–)-(α)-Pinene: [α]D22 = –41.26 (neat) ־79 % ee.[26] All other reagents
were used as supplied from commercial sources and stored appro-
priately. Analytical HPLC was carried out using a Jasco HPLC appara-
tus: PU-980 pump, UV-975 UV detector, LG-980-02 gradient unit and
GT-103. Commercially available solvents were used as pro analysi
(p. a.) grade eluents and degassed with a Labotec Gastorr GT-103.
1H and 13C NMR spectra were recorded in CDCl3 with Bruker
DMX 300 and Bruker DRX 500 spectrometers. IR spectra were meas-
ured with a Jasco FT/IR-430 spectrometer bearing an ATR attach-
ment. Low and high resolution ESI mass spectra were recorded
with Bruker amaZon SL and Bruker maXis 4G spectrometers, respec-
tively.
Scheme 5. Completion of the enantioselective synthesis of cytisine (1).
Introduction of the C3-bromide by addition of Br2 in DCM
followed by Et3N mediated elimination delivered the benzyl
congener of pivotal precursor 4a in 79 % yield, which was then
subjected to a Stille-coupling, delivering alkene 5a. Hydrogen-
ation of this electron deficient alkene over palladium on char-
coal in MeOH required an H2 pressure of 6 bar. O-benzyl cleav-
age required the addition of 10 equiv. concentrated aqueous
HCl.[24] In this way the cyclization-precursor 15, was obtained
as a 3:1 mixture of cis/trans epimers. As reported earlier,[6b] the
diastereoselectivity of this hydrogenation is of little concern
since the next step employs an in situ epimerization/cyclization
that converts both epimers into desired bispidine 16. Subse-
quent reduction and debenzylation yields cytisine.[11a]
To facilitate HPLC analysis of 16 for ee determinations, the
reaction sequence was also applied to the enantiomeric series,
starting from (+)-α-pinene to deliver the corresponding bis-
pidine ent-16. The ee values obtained from both bispidine 16
(ee = 78 %) and its enantiomer ent-16 (ee = 81 %) corresponded
to those of their respective starting materials, (–)-α-pinene
(79 % optical purity) and (+)-α-pinene (81 % optical purity), as
well as to that of the C1-building block 8 (vide supra).
(3aS,4S,6S)-2-[(S)-1-(Benzyloxy)but-3-en-2-yl]-3a,5,5-trimethyl-
hexahydro-4,6-methanobenzo[d][1,3,2]dioxaborole (ent-10):
LDA was freshly prepared by adding BuLi (1.9
M solution in hexane,
0.68 mL, 1.3 mmol) to diisopropylamine (1.6 mmol, 0.22 mL) in dry
THF (1.5 mL) at –78 °C and stirring for 20 min at room temp. The
resulting LDA solution was added dropwise to a mixture of benzyl-
As the de of C2-building block 10 was not directly evident
from NMR (vide supra), ent-11 was prepared from recrystallized
(+)-pinanediol (> 95 % ee, by HPLC and 1H NMR, see Supporting oxymethylboronate ent-8 (300 mg, 1.0 mmol) and CH2Br2 (0.70 mL,
Eur. J. Org. Chem. 2016, 958–964
www.eurjoc.org
960
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
10 mmol) in THF (2 mL) at –78 °C. After stirring for 1 h at the same
and cooled to 0 °C, before aq. NaOH (2
M, 3 mL) and H2O2 (30 %,
temperature the solvent was removed under reduced pressure 0.06 mL) were added simultaneously. A white precipitate formed,
(3 mbar) and cooling with an ice/NaCl bath (–10 to –5 °C) to yield
the borate salt as a brown solid. The flask was returned to a –78 °C
bath, before a solution of ZnCl2 in Et2O (1 M, 2.8 mL, 2.5 mmol) was
which was still visible after stirring for 1 h at room temp. The reac-
tion mixture was heated to 45 °C for 1 h after which the precipitate
had disappeared. The reaction was quenched by the addition of aq
Na2S2O3 (20 % by wt., 1 mL), poured onto saturated aq NH4Cl and
extracted into Et2O (2 ×). The combined organic layers were dried
with Na2SO4 and concentrated in vacuo. The residue was dissolved
in Et2O/CyHex, 1:1 (0.6 mL) and cooled to 0 °C. Recrystallized (+)-
pinanediol (62 mg, 0.33 mmol) was dissolved in the mixture before
added and the reaction mixture was allowed to slowly warm up to
room temp. overnight, after which time the solid had dissolved. The
reaction mixture was diluted with an equal volume of CyHex and
washed with saturated aq. NH4Cl (10 mL). The aqueous phase was
re-extracted twice with CyHex/Et2O = 4:1. The combined organic
phases were filtered through a column of MgSO4 into a dried
Schlenk tube and concentrated in vacuo to yield pinanediol (2-
benzyloxy-1(R)-bromoethyl)boronate ent-9 (396 mg) as a pale
brown oil. An NMR sample (18 mg, 5 %) was removed and the rest
of the crude product was used immediately for the next step. 1H
NMR (300 MHz, CDCl3): δ = 7.41–7.28 (m, 5 H), 4.61 (s, 2 H), 4.38
aq. KOH* (5 M, 0.07 mL, 0.33 mmol) was added at the same temper-
ature. To suspend the thick precipitate further Et2O/CyHex (1:1,
2 mL) was added and the suspension was stirred overnight. The
precipitate (14) was filtered and thoroughly washed with Et2O/Cy-
Hex (1:1) until no more product was detectable by TLC in the wash-
ings. The combined filtrates were concentrated in vacuo and the
(dd, J = 1.9, 8.8 Hz, 1 H), 3.93–3.74 (m, 2 H), 3.54 (dd, J = 6.4, 8.0 Hz, residue was purified by flash chromatography (silica gel; CyHex/
1 H), 2.42–2.30 (m, 1 H), 2.23 (s, 1 H), 2.13–2.06 (m, 1 H), 1.96–1.85
(m, 2 H), 1.42 (s, 3 H), 1.30 (s, 3 H), 1.28 (d, J = 6.3 Hz, 1 H), 0.85 (s,
3 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 138.0, 128.3, 127.7, 127.6,
86.8, 78.6, 73.1, 71.5, 51.2, 39.3, 38.3, 35.2, 28.3, 27.0, 26.2, 23.9 ppm.
A solution of this crude product (< 0.95 mmol) in dry THF (10 mL)
was cooled to –78 °C before a solution of vinylmagnesium bromide
EtOAc, 4:1), yielding alcohol ent-11 (46 mg, 0.24 mmol, 81 %) as a
colorless oil. Data for ent-11: Rf = 0.16 (silica gel; CyHex/EtOAc, 4:1).
= +26.0 (CHCl3, c = 1.00). 1H
24
24
24
[α]
= +32.0. [α]
= +28.0. [α]
546
578
589
NMR (300 MHz, CDCl3): δ = 7.40–7.07 (m, 5 H), 5.65 (ddd, J = 7.7,
10.2, 17.8 Hz, 1 H), 5.17–4.98 (m, 2 H), 4.46 (s, 2 H), 3.83–3.33 (m, 4
H), 2.70–2.45 (m, 1 H), 2.19 (t, J = 5.9 Hz, 1 H) ppm. 13C NMR
(75 MHz, CDCl3): δ = 137.9, 135.8, 128.6, 128.5, 127.8, 127.6, 117.5,
in THF (1
M, 1.4 mL, 1.4 mmol) was added dropwise. The reaction
mixture was stored in the freezer at –20 °C overnight and heated
to room temp. for 1 h, before it was diluted with an equal amount
of CyHex, poured on saturated aq NH4Cl and extracted into Et2O
(2 ×). The combined organic phases were dried with Na2SO4, fil-
tered, and concentrated in vacuo. The residue was purified by flash
chromatography (silica gel; CyHex/Et2O, 19:1) yielding allylborane
ent-10 (266 mg, 0.78 mmol, 82 %) as a colourless oil. Rf = 0.11 (silica
73.4, 72.5, 65.1, 45.7 ppm. IR (film): νmax = 3432 (m, br), 2862 (m),
˜
2360 (s), 2341 (s), 1496 (w), 1455 (m), 1361 (m), 1206 (w), 1099 (s),
1029 (s), 918 (m), 736 (m), 698 (s) cm–1. MS (ESI-ion trap): m/z = [M
+ Na]+ calcd. for C12H16O2Na 215.10, found 215.11; [M + H]+ calcd.
for C12H16O2H 193.12, found 193.08. HRMS (ESI-Q-TOF) m/z: [M +
Na]+ calcd. for C12H16O2Na 215.1043, found 215.1026. Analysis of
14: The precipitate was dried for several days under air after which
24
24
24
gel; CyHex/EtOAc, 19:1). [α] = +14.5. [α] = +13.0. [α] = +12.0 94 mg (≈ 0.24 mmol, ≈ 80 %) of a white solid remained. ESI-MS
546
578
589
1
(CHCl3, c = 1.00). H NMR (300 MHz, CDCl3): δ = 7.46–7.12 (m, 5 H),
5.90 (ddd, J = 8.4, 10.3, 17.2 Hz, 1 H), 5.19–4.97 (m, 2 H), 4.54 (s, 2
H), 4.29 (dd, J = 1.9, 8.8 Hz, 1 H), 3.69–3.64 (m, 2 H), 2.43–2.26 (m,
2 H), 2.22–2.11 (m, 1 H), 2.06 (t, J = 5.3 Hz, 1 H), 1.93–1.81 (m, 2 H),
1.39 (s, 3 H), 1.29 (s, 3 H), 1.17 (d, J = 11.3 Hz, 1 H), 0.84 (s, 3 H) ppm.
confirmed the formation of 14 and other pinanediol borate
esters after redissolution in MeOH/EtOH: MS (ESI-ion trap)
m/z: [Pinanediol2B]– 347 (45), [PinanediolB(OEt)2]– 283 (45),
[PinanediolB(OEt)(OMe)]– 255 (100), [PinanediolBO]– 195 (35). *The
rest of the crude homologation product was oxidized accordingly, but
13C NMR (75 MHz, CDCl3): δ = 138.6, 136.6, 128.2, 127.5, 127.3, 114.9, the order of addition of (+)-pinanediol and KOH was reversed upon
85.8, 77.8, 72.8, 71.0, 51.2, 39.4, 38.1, 35.4, 28.5, 27.0, 26.4, 26.2,
which the yield dropped to 71 %.
23.9 ppm. IR (film): νmax = 2919 (m), 2869 (w), 2360 (s), 2341 (m),
˜
(S)-N-Benzyl-N-{2-[(benzyloxy)methyl]but-3-en-1-yl}acrylamide
(6): A solution of alcohol 10 (125 mg, 0.65 mmol), Et3N (0.34 mL,
2.45 mmol) and TsCl (320 mg, 1.68 mmol) in MeCN (6.5 mL) was
heated at 60 °C for 4 h. RfTosylate 7 = 0.30 (silica gel; CyHex/EtOAc,
4:1). The reaction mixture was cooled to room temp., BnNH2
(2.5 mL) was added and heating at 60 °C was continued for 14 h.
Solvent and excess BnNH2 were removed by kugelrohr distillation
at 80 °C and 5 mbar pressure. After cooling to room temp. the
1454 (w), 1376 (s), 1339 (s), 1281 (m), 1239 (m), 1209 (w),
1098 (m), 1077 (s), 1030 (m), 991 (w), 938 (w), 904 (w), 735 (m), 697
(m) cm–1. MS (ESI-ion trap): m/z = [M + Na]+ calcd. for C21H29BO3Na
363.21, found 363.17. HRMS (ESI-Q-TOF) m/z: [M + Na]+ calcd. for
C21H29BO3Na 363.2106, found 363.2109.
(R)-2-[(Benzyloxy)methyl]but-3-en-1-ol (ent-11): BuLi (1.79
M so-
lution in hexanes, 0.67 mL, 1.2 mmol) was added dropwise over
5 min to a solution of allylborane ent-10 (340 mg, 1.00 mmol) and residue was partitioned between aq NaOH (1
M) and EtOAc. The
iodochloromethane (0.15 mL, 2.0 mmol) in absolute THF (10 mL) at
–78 °C. The reaction mixture was warmed up to room temp. over-
night, after which it was diluted with an equal volume of CyHex
and extracted with saturated aq NH4Cl. The aqueous layer was re-
extracted with Et2O. The combined organic phases were dried with
aqueous phase was re-extracted with EtOAc (2 ×), dried with
Na2SO4, filtered, and concentrated in vacuo. Further BnNH2 was re-
moved by kugelrohr distillation at 70 °C and 5 mbar pressure. The
residue was cooled to room temp. and dissolved in DCM (8 mL).
The resulting solution was cooled to –5 °C before Et3N (1.7 mL,
Na2SO4 and concentrated in vacuo yielding the homologation prod- 12.3 mmol) was added. Acryloylchloride (0.64 mL, 7.85 mmol) was
uct (354 mg, 1.00 mmol, quant.), which was used without further
then added dropwise over 5 min. The reaction mixture was stirred
for 10 min at –5 °C and then for 3 h at room temp. before saturated
purification. 1H NMR of the crude material (300 MHz, CDCl3): δ =
7.43–7.22 (m, 5 H), 5.81 (ddd, J = 7.7, 10.2, 17.5 Hz, 1 H), 5.18–4.94 aq NaHCO3 (10 mL) was added and the mixture extracted with
(m, 2 H), 4.53 (d, J = 2.2 Hz, 2 H), 4.23 (dd, J = 2.2, 8.8 Hz, 1 H),
3.48–3.25 (m, 2 H), 2.70 (qd, J = 7.2, 14.1 Hz, 1 H), 2.39–2.25 (m, 1
H), 2.23–2.11 (m, 1 H), 2.03 (t, J = 5.6 Hz, 1 H), 1.94–1.74 (m, 2 H),
EtOAc (3 ×). The combined organic phases were dried with Na2SO4,
filtered, and concentrated in vacuo. The residue was purified by
flash chromatography (silica gel; CyHex/EtOAc, 4:1) yielding metath-
1.33 (s, 3 H), 1.28 (s, 3 H), 1.14 (d, J = 10.6 Hz, 1 H), 1.02 (dd, J = esis precursor 6 (158 mg, 0.47 mmol, 73 %) as a pale yellow oil. Rf =
6.6, 15.3 Hz, 1 H), 0.90 (dd, J = 7.8, 15.3 Hz, 1 H), 0.83 (s, 3 H). Part
of the material (106 mg, <0.30 mmol) was dissolved in THF (3 mL)
0.16 (silica gel; CyHex/EtOAc, 4:1). The NMR spectrum recorded at
room temp. showed the presence of two discrete rotamers. 1H NMR
Eur. J. Org. Chem. 2016, 958–964
www.eurjoc.org
961
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(500 MHz, CDCl3): δ = 7.38–7.21 (m, 9 H), 7.14 (d, J = 6.9 Hz, 1 H),
6.73 (dd, J = 10.4, 16.7 Hz, 0.5 H), 6.51 (dd, J = 10.1, 16.7 Hz, 0.5 H),
6.41 (ddd, J = 2.2, 17.0, 34.7 Hz, 1 H), 5.87–5.74 (m, 1 H), 5.68 (ddd,
J = 2.2, 12.6, 28.1 Hz, 1 H), 5.19–5.09 (m, 2 H), 4.77 (d, J = 14.8 Hz,
0.5 H), 4.67–4.57 (m, 1.5 H), 4.54–4.44 (m, 2 H), 3.63 (dd, J = 6.6,
15.1 Hz, 0.5 H), 3.58–3.45 (m, 2 H), 3.40 (dd, J = 6.1, 9.3 Hz, 0.5 H),
3.26 (dd, J = 7.9, 15.1 Hz, 0.5 H), 2.95–2.83 (m, 0.5 H), 2.73–2.61 (m,
0.5 H) ppm. 13C NMR (126 MHz, CDCl3): δ = 167.2, 166.8, 138.2,
137.9, 137.6, 137.5, 137.0, 136.4, 128.8, 128.5, 128.5, 128.4, 128.3,
128.1, 127.8, 127.7, 127.6, 127.5, 127.3, 126.3, 118.0, 117.1, 73.2, 73.1,
1993 (w), 1657 (s), 1613 (m), 1495 (w), 1475 (m), 1453 (m), 1426 (w),
1360 (m), 1297 (w), 1260 (w), 1205 (m), 1173 (w), 1092 (m), 1073
(m), 1028 (w), 1002 (w), 957 (w), 909 (w), 882 (w), 863 (w), 838 (w),
735 (s), 697 (s), 613 (m), 585 (w), 553 (w) cm–1. MS (ESI-ion trap):
m/z = [M + Na]+ calcd. for C20H20BrNO2Na 408.06, found 408.06.
HRMS (ESI-Q-TOF) m/z: [M + Na]+ calcd. for C20H20BrNO2Na
408.0570, found 408.0576.
(S)-1′-Benzyl-5′-[(benzyloxy)methyl]-6-methoxy-5′,6′-dihydro-
[2,3′-bipyridin]-2′(1′H)-one (5a): 2-Methoxy-6-(tributylstannyl)pyr-
idine (129 mg, 0.32 mmol) was added to a solution of bromide 4a
(90 mg, 0.23 mmol), CuCl (30 mg, 0.30 mmol), LiCl (12 mg,
0.30 mmol) and Pd(PPh3)4 (24 mg, 21 μmol) in absolute THF (3 mL).
The reaction mixture was heated to 60 °C overnight, cooled to room
temp. and concentrated in vacuo. The residue was purified by flash
chromatography (silica gel; CyHex/EtOAc, 4:1) yielding coupling
product 5a (89 mg, 0.22 mmol, 92 %) as a pale yellow oil. Rf = 0.26
71.6, 70.5, 51.9, 49.3, 48.3, 48.1, 44.4, 43.0, 26.9 ppm. IR (film): ν
=
˜
max
2856 (w), 1717 (w), 1649 (s), 1612 (s), 1495 (m), 1445 (s), 1359 (m),
1269 (w), 1214 (s), 1175 (w), 1148 (w), 1098 (m), 1078 (m), 1028 (m),
977 (m), 919 (m), 793 (m), 735 (s), 697 (s), 614 (w), 596 (w), 580 (w),
567 (w) cm–1. MS (ESI-ion trap): m/z = [M + Na]+ calcd. for
C
22H25NO2Na 358.18, found 358.17; [M + H]+ calcd. for C22H25NO2H
336.20, found 336.17. HRMS (ESI-Q-TOF): m/z = [M + Na]+ calcd. for
22
22
22
(silica gel; CyHex/EtOAc, 4:1). [α]
= +22.0. [α]
= +18.0. [α]
=
546
578
589
C
22H25NO2Na 358.1778, found 358.1798; [M + H]+ calcd. for
1
+18.0 (c = 0.50, CHCl3). H NMR (300 MHz, CDCl3): δ = 7.60 (d, J =
7.5 Hz, 1 H), 7.54–7.45 (m, 1 H), 7.32–7.11 (m, 11 H), 6.59 (d, J =
8.1 Hz, 1 H), 4.74 (d, J = 14.7 Hz, 1 H), 4.49 (d, J = 14.4 Hz, 1 H), 4.25
(s, 2 H), 3.53–3.30 (m, 3 H), 3.23 (t, J = 8.9 Hz, 1 H), 2.93–2.65 (m, 1
H) ppm. 13C NMR (75 MHz, CDCl3): δ = 163.5, 163.0, 150.1, 139.4,
138.8, 137.8, 137.5, 134.1, 128.6, 128.4, 127.8, 127.7, 127.5, 117.1,
C22H25NO2H 336.1958, found 336.1960.
(S)-1-Benzyl-5-[(benzyloxy)methyl]-5,6-dihydropyridin-2(1H)-
one (3a): A solution of metathesis precursor 6 (100 mg, 0.30 mmol)
and Grubbs 2nd generation catalyst (11 mg, 14 μmol, 0.05 equiv.)
in DCM (15 mL) was refluxed for 5 h. The reaction mixture was
concentrated in vacuo and the residue was purified by flash chro-
matography (silica gel; CyHex/EtOAc, 4:1 → 7:3), yielding dihydro-
pyridone 3a (68 mg, 0.22 mmol, 74 %) as a pale yellow oil. Rf = 0.10
109.8, 73.1, 69.2, 53.1, 50.3, 46.8, 35.3 ppm. IR (film): νmax = 3901
˜
(w), 3069 (w), 3853 (w), 3839 (w), 3819 (w), 3800 (w), 3750 (w), 3734
(w), 3710 (w), 3689 (w), 3675 (w), 3649 (w), 3628 (w), 3566 (w), 2952
24
24
(silica gel; CyHex/EtOAc, 7:3). [α]
= +144.0. [α]
= +126.0. (m), 2924 (m), 2855 (m), 2360 (s), 2341 (s), 2158 (w), 2024 (w), 1967
546
578
24
[α]
= +118.0 (c = 0.50, CHCl3). 1H NMR (300 MHz, CDCl3): δ = (w), 1868 (w), 1845 (w), 1827 (w), 1792 (w), 1771 (w), 1733 (w), 1716
589
7.33–7.19 (m, 8 H), 7.19–7.12 (m, 2 H), 6.38 (dd, J = 4.2, 9.8 Hz, 1 H),
5.97 (dd, J = 1.7, 9.8 Hz, 1 H), 4.67 (d, J = 14.7 Hz, 1 H), 4.41 (d, J =
14.7 Hz, 1 H), 4.26 (s, 2 H), 3.39–3.12 (m, 4 H), 2.75–2.57 (m, 1
H) ppm. 13C NMR (75 MHz, CDCl3): δ = 163.8, 139.8, 137.6, 137.2,
128.5, 128.3, 128.2, 127.7, 127.6, 127.4, 126.0, 73.1, 69.3, 49.7, 46.6,
(w), 1698 (w), 1684 (w), 1654 (m), 1577 (m), 1558 (w), 1541 (w), 1521
(w), 1507 (w), 1457 (m), 1437 (w), 1417 (w), 1362 (w), 1318 (w), 1259
(w), 1230 (w), 1173 (w), 1119 (w), 1092 (w), 1076 (w), 1029 (w),
873 (w), 808 (w), 736 (w), 696 (w), 668 (m), 650 (w), 614 (w), 566
(w) cm–1. MS (ESI-ion trap): m/z = [M + Na]+ calcd. for C26H26N2O3Na
437.18, found 437.17; [M + H]+ calcd. for C26H26N2O3H 415.20, found
415.21. HRMS (ESI-Q-TOF) m/z: [M + Na]+ calcd. for C26H26N2O3Na
437.1836, found 437.1840; [M + H]+ calcd. for C26H26N2O3H
415.2016, found 415.2016.
35.1 ppm. IR (film): νmax = 3734 (w), 3060 (w), 3030 (w), 2922 (m),
˜
2856 (m), 2360 (s), 2341 (m), 2159 (w), 1967 (w), 1733 (w), 1662 (s),
1609 (s), 1558 (w), 1541 (w), 1521 (w), 1482 (m), 1454 (m), 1435 (m),
1380 (w), 1361 (m), 1261 (m), 1204 (w), 1170 (w), 1090 (s), 1028 (w),
819 (s), 728 (s), 697 (s), 669 (w), 609 (m), 566 (w) cm–1. MS (ESI-ion
trap): m/z = [M + Na]+ calcd. for C20H21NO2Na 330.15, found 330.09.
HRMS (ESI-Q-TOF) m/z: [M + Na]+ calcd. for C20H21NO2Na 330.1465,
found 330.1464.
(3S,5S)-1-Benzyl-5-(hydroxymethyl)-3-(6-methoxypyridin-2-yl)-
piperidin-2-one (cis-12) and (3R,5S)-1-Benzyl-5-(hydroxy-
methyl)-3-(6-methoxypyridin-2-yl)piperidin-2-one (trans-12): A
mixture of coupling product 5a (28 mg, 67 μmol) and 10 wt.-% Pd/
C (14 mg) in MeOH (0.7 mL) was stirred under an atmosphere of
H2 (6 bar) for 4 h. In order to achieve debenzylation as well, 2 drops
of conc. aq HCl were added and stirring under an atmosphere of H2
(6 bar) was continued overnight. The reaction mixture was filtered
through celite and concentrated in vacuo. The residue was taken
up in a mixture of saturated aq NaHCO3 and EtOAc. The phases
(S)-1-Benzyl-5-[(benzyloxy)methyl]-3-bromo-5,6-dihydropyr-
idin-2(1H)-one (4a): Bromine (46 μL, 0.91 mmol) was added to a
solution of dihydropyridone 3a (140 mg, 0.46 mmol) in DCM (5 mL),
which was then refluxed for 6 h. A solution of Na2S2O3 (20 % by
weight, 5 mL) was added and the mixture was extracted into DCM
(3 ×). The combined organic phases were dried with Na2SO4, fil-
tered, and concentrated in vacuo. The residue was dissolved in DCM were separated and the aqueous layer was re-extracted into EtOAc
(5 mL) and Et3N (0.64 mL, 4.6 mmol) was added. The reaction mix- (2 ×). The combined organic layers were dried with Na2SO4, filtered,
ture was refluxed overnight, cooled to room temp. and stirred for
a further 10 h at room temp., before it was poured into saturated
aq NH4Cl and extracted into DCM (3 ×). The combined organic
phases were dried with Na2SO4, filtered, and concentrated in vacuo.
The residue was purified by flash chromatography (silica gel; CyHex/
and concentrated in vacuo. 1H NMR confirmed reduction of the
alkene but not debenzylation. The material thus obtained was dis-
solved in MeOH (0.7 mL), to which 10 wt.-% Pd/C (14 mg) and
conc. aq HCl (0.05 mL, 10 equiv.) were added under Ar. The reaction
mixture was stirred under an atmosphere of H2 (6 bar) for 6 h. The
EtOAc, 4:1) yielding bromide 4a (138 mg, 0.37 mmol, 79 %) as a reaction mixture was filtered through celite and concentrated in
pale yellow oil. Rf = 0.17 (silica gel; CyHex/EtOAc, 4:1). [α]52446 = +93.0. vacuo. The residue was taken up in a mixture of saturated aq NaH-
1
24
24
[α]
= +80.0. [α]
= +76.0 (c = 1.00, CHCl3). H NMR (300 MHz, CO3 and EtOAc. The phases were separated and the aqueous layer
589
578
CDCl3): δ = 7.33–7.18 (m, 8 H), 7.18–7.07 (m, 2 H), 6.79 (d, J = 4.7 Hz, was re-extracted into EtOAc (2 ×). The combined organic layers
1 H), 4.69 (d, J = 14.7 Hz, 1 H), 4.42 (d, J = 14.4 Hz, 1 H), 4.24 (s, 2
were dried with Na2SO4, filtered, and concentrated in vacuo. The
residue was purified by flash chromatography (silica gel; CyHex/
H), 3.46–3.24 (m, 3 H), 3.23–3.09 (m, 1 H), 2.77–2.55 (m, 1 H) ppm.
13C NMR (75 MHz, CDCl3): δ = 159.5, 140.4, 137.4, 136.7, 128.6, 128.4, EtOAc, 1:1) yielding a 3:1 mixture of isomers cis-12 and trans-12
128.4, 127.9, 127.7, 120.2, 73.2, 68.4, 51.2, 46.7, 37.2 ppm. IR (film):
(15 mg, 46 μmol, 69 %) as a pale yellow oil. Rf = 0.29 (EtOAc). 1H
νmax = 3030 (w), 2923 (m), 2854 (m), 2359 (w), 2341 (w), 2089 (w), NMR (300 MHz, CDCl3): δ = 7.56–7.46 (m, 2 H, C4′-H-cis, C4′-H-trans),
˜
Eur. J. Org. Chem. 2016, 958–964
www.eurjoc.org
962
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[7]
a) J. W. Coe, P. R. Brooks, M. G. Vetelino, M. C. Wirtz, E. P. Arnold, J. Huang,
S. B. Sands, T. I. Davis, L. A. Lebel, C. B. Fox, A. Shrikhande, J. H. Heym, E.
Schaeffer, H. Rollema, Y. Lu, R. S. Mansbach, L. K. Chambers, C. C. Rovetti,
D. W. Schulz, F. D. Tingley, B. T. O'Neill, J. Med. Chem. 2005, 48, 3474–
3477; b) J. W. Coe, H. Rollema, B. T. O'Neill, Annu. Rep. Med. Chem. 2009,
44, 71–101. Cytisine itself has been used for smoking cessation as well,
see: c) J. F. Etter, Arch. Intern. Med. 2006, 166, 1553–1559.
a) J. A. Abin-Carriquiry, M. P. Zunini, B. K. Cassels, S. Wonnacott, F. Dajas,
Bioorg. Med. Chem. Lett. 2010, 20, 3683–3687; b) B. Ivanova, M. Spiteller,
J. Mol. Struct. 2013, 1034, 173–182.
For AChBP as a model, see: a) C. D. Dellisanti, Y. Yao, J. C. Stroud, Z. Z.
Wang, L. Chen, Nat. Neurosci. 2007, 10, 953–962. For cytisine in this re-
ceptor, see: b) P. Rucktooa, C. A. Haseler, R. van Elk, A. B. Smit, T. Gal-
lagher, T. K. Sixma, J. Biol. Chem. 2012, jbc.M112.360347. For other nico-
tinic pharmacophores, see: c) A. B. Smit, N. I. Syed, D. Schaap, J. van Min-
nen, J. Klumperman, K. S. Kits, H. Lodder, R. C. van der Schors, R. van Elk,
B. Sorgedrager, K. Brejc, T. K. Sixma, W. P. M. Geraerts, Nature 2001, 411,
261–268; d) K. Brejc, W. J. van Dijk, R. V. Klaassen, M. Schuurmans, J.
van der Oost, A. B. Smit, T. K. Sixma, Nature 2001, 411, 269–276; e) P. H. N.
Celie, S. E. van Rossum-Fikkert, W. J. van Dijk, K. Brejc, A. B. Smit, T. K.
Sixma, Neuron 2004, 41, 907–914; f) Y. Bourne, T. T. Talley, S. B. Hansen,
P. Taylor, P. Marchot, EMBO J. 2005, 24, 1512–1522; g) S. B. Hansen, G.
Sulzenbacher, T. Huxford, P. Marchot, P. Taylor, Y. Bourne, EMBO J. 2005,
24, 3635–3646.
Carriquiry et al.[8a] refer to an “aromatic-box” that encloses the piperidine
unit. For a detailed discussion see ref.[6b]. More recently examples of
cytisine-like bispidine systems with variations in the relevant area have
been reported, see: a) C. Eibl, I. Tomassoli, L. Munoz, C. Stokes, R. L.
Papke, D. Gündisch, Bioorg. Med. Chem. 2013, 21, 7283–7308; b) C. Eibl,
L. Munoz, I. Tomassoli, C. Stokes, R. L. Papke, D. Gündisch, Bioorg. Med.
Chem. 2013, 21, 7309–7329.
7.35–7.29 (m, 10 H, 5 × ArC-H-cis, 5 × ArC-H-trans), 6.88 (m, 2 H, C5′-
H-cis, C5′-H-trans), 6.62 (d, J = 8.0 Hz, 2 H, C3′-H-cis and trans), 4.94
(d, J = 14.5 Hz, 1 H, CH2Ph-trans), 4.93 (d, J = 14.5 Hz, 1 H, CH2Ph-
cis), 4.43 (d, J = 14.5 Hz, 1 H, CH2Ph-trans), 4.40 (d, J = 14.5 Hz, 1 H,
CH2Ph-cis), 3.87–3.91 (m, 4 H, OCH3-cis, C3-H-trans), 3.86 (s, 3 H,
OCH3-trans), 3.78 (dd, J = 10.5, 7.0 Hz, 1 H, C3-H-cis), 3.66–3.37 (m,
6 H, C5a-H2-cis, C5a-H2-trans, C6-H2-trans, C5a-H2-cis, C5a-H2-trans,
C6-H2-cis), 3.25–3.14 (m, 2 H, C6-H2-cis, C6-H2-trans), 2.42 (m, 1 H,
C5-H-trans), 2.27–2.05 (m, 5 H, C4-H2-trans, C5-H-cis, C4-H2-cis, C4-
H2-trans) ppm. 13C NMR only signals for cis-12 are reported:
(75 MHz, CDCl3): δ = 170.1, 163.8, 158.1, 138.9, 137.2, 128.5, 128.1,
127.4, 116.6, 108.7, 65.0, 53.2, 50.9, 50.4, 50.2, 36.3, 31.4 ppm. Spec-
troscopic data were consistent with those reported in the litera-
ture.[6b]
[8]
[9]
Acknowledgments
The authors thank Prof. Dr. C. Schmuck for excellent laboratory
conditions and fruitful discussions. Prof. Dr. G. Haberhauer and
Mrs. H. Wöll are thanked for help with the HPLC experiments.
Funding from the University of Duisburg, Germany and the Sci-
ence Support Center, Essen, Germany
[10]
Keywords: Asymmetric synthesis · Natural products ·
Alkaloids · Metathesis · Rearrangement · Carbenoids ·
Boronate homologation
[11]
[12]
Three asymmetric syntheses of cytisine have been described previously,
all of which were reviewed by O'Brien in 2007, see ref.[6a]: a) D. Gray, T.
Gallagher, Angew. Chem. Int. Ed. 2006, 45, 2419–2423; Angew. Chem.
2006, 118, 2479; b) T. Honda, R. Takahashi, H. Namiki, J. Org. Chem. 2005,
70, 499–504; c) B. Danieli, G. Lesma, D. Passarella, A. Sacchetti, A. Silvani,
A. Virdis, Org. Lett. 2004, 6, 493–496.
Gallagher described the asymmetric synthesis of a suitable lactam pre-
cursor to 3 by enzymatic resolution, see ref.[11a]. Although this would
add another step with a maximum yield of 50 % to the sequence, it
would – in theory – grant access to both enantiomers as well.
A recent racemic synthesis of the alcohol precursor 11 required 4 steps
including a selenium–oxide elimination: S.-H. Yang, G. R. Clark, V. Caprio,
Org. Biomol. Chem. 2009, 7, 2981–2990.
For recent reviews, see: a) D. S. Matteson, J. Org. Chem. 2013, 78, 10009–
10023; b) S. P. Thomas, R. M. French, V. Jheengut, V. K. Aggarwal, The
Chemical Record 2009, 9, 24–39; c) D. S. Matteson, D. Majumdar, J. Am.
Chem. Soc. 1980, 102, 7588–7590; d) D. S. Matteson, R. Ray, J. Am. Chem.
Soc. 1980, 102, 7590–7591.
A synthesis of 8 requiring some chromatography has been described,
see: D. S. Matteson, M. L. Peterson, J. Org. Chem. 1987, 52, 5116–5121.
For a chromatography free synthesis see the Supporting Information.
J. Wityak, R. A. Earl, M. M. Abelman, Y. B. Bethel, B. N. Fisher, G. S. Kauff-
man, C. A. Kettner, P. Ma, J. L. McMillan, J. Org. Chem. 1995, 60, 3717–
3722. The authors controlled the enantiomeric purity of pinanediol by
1H-NMR after converting the material into the Mosher ester. Convenient
alternatives are described in the Supporting Information.
[1] For a recent review, see: J. Rouden, M.-C. Lasne, J. Blanchet, J. Baudoux,
Chem. Rev. 2014, 114, 712–778. For an isolation procedure, see: A. J.
Dixon, M. J. McGrath, P. O'Brien, Org. Synth. 2006, 83, 141.
[2] T. S. Gray, Edinburgh Med. J. 1862, 7, 908–919.
[3] H. R. Ing, J. Chem. Soc. 1932, 2778–2780.
[4] a) E. G. Pérez, C. Méndez-Gálvez, B. K. Cassels, Nat. Prod. Rep. 2012, 29,
555–567; A. A. Jensen, B. Frølund, T. Lijefors, P. Krogsgaard-Larsen, J. Med.
Chem. 2005, 48, 4705–4745. See also: b) R. A. Glennon, M. Dukat, Bioorg.
Med. Chem. Lett. 2004, 14, 1841–1844; c) J. E. Tønder, P. H. Olesen, Curr.
Med. Chem. 2001, 8, 651–674.
[13]
[14]
[5] For asymmetric syntheses C3 and C5 modifications, see: a) P. Imming, P.
Klaperski, M. T. Stubbs, G. Seitz, D. Gündisch, Eur. J. Med. Chem. 2001,
36, 375–388; b) E. Marriére, J. Rouden, V. Tadino, M.-C. Lasne, Org. Lett.
2000, 2, 1121–1124; c) S. K. Chellappan, Y. Xiao, W. Tueckmantel, K. J.
Kellar, A. P. Kozikowski, J. Med. Chem. 2006, 49, 2673–2676; d) A. P. Kozik-
owski, S. K. Chellappan, Y. Xiao, K. M. Bajjuri, H. Yuan, K. J. Kellar, P. A.
Petukhov, ChemMedChem 2007, 2, 1157–1161; e) G. Roger, B. Lagnel, J.
Rouden, L. Besret, H. Valette, S. Demphel, J. Gopisetti, C. Coulon, M. Otta-
viani, L. A. Wrenn, S. R. Letchworth, G. A. Bohme, J. Benavides, M.-C.
Lasne, M. Bottlaender, F. Dollé, Bioorg. Med. Chem. 2003, 11, 5333–5343.
For C10 modifications, see: f) N. Houllier, S. Gouault, M.-C. Lasne, J.
Rouden, Tetrahedron 2006, 62, 11679–11686; g) J. Rouden, A. Ragot, S.
Gouault, D. Cahard, J.-C. Plaquevent, M.-C. Lasne, Tetrahedron: Asymmetry
2002, 13, 1299–1305; h) F. Frigerio, C. A. Haseler, T. Gallagher, Synlett
2010, 729–730.
[15]
[16]
[17]
[18]
a) B. J. Kim, J. Zhang, S. Tan, D. S. Matteson, W. H. Prusoff, Y.-C. Cheng,
Org. Biomol. Chem. 2012, 10, 9349–9358; b) D. Fan, R. L. Jarvest, L. Laza-
rides, Q. Li, X. Li, Y. Liu, L. Liao, J. E. Mordaunt, Z. Ni, J. Plattner, WO
2009046098, 2009.
In ref.[15] Matteson mentions addition of excess ZnCl2 (as applied in
ref.[17a]) or the removal of diisopropylamine (a ligand that potentially
disables ZnCl2) as possible ways to avoid loss of de when LiCHBr2 is
generated from LDA and CH2Br2. In our case only the latter worked relia-
bly.
[6] For a review on previous syntheses of cytisine see: a) D. Stead, P. O′
Brien, Tetrahedron 2007, 63, 1885–1896. For a later synthesis, see: b) C.
Hirschhäuser, C. A. Haseler, T. Gallagher, Angew. Chem. Int. Ed. 2011, 50,
5162–5165; Angew. Chem. 2011, 123, 5268. For total syntheses of deriva-
tives (C4-substituted), see: c) B. T. O'Neill, D. Yohannes, M. W. Bundes-
mann, E. P. Arnold, Org. Lett. 2000, 2, 4201–4204; d) P. Durkin, P. Magrone,
S. Matthews, C. Dallanoce, T. Gallagher, Synlett 2010, 18, 2789–2791. For
syntheses of norcytisines, see: e) D. Yohannes, K. Procko, L. A. Lebel, C. B.
Fox, B. T. O'Neill, Bioorg. Med. Chem. Lett. 2008, 18, 2316; f) D. Yohannes,
C. P. Hansen, S. R. Akireddy, T. A. Hauser, M. N. Kiser, N. J. Gurnon, C. S.
Day, B. Bhatti, W. S. Caldwell, Org. Lett. 2008, 10, 5353–5356. For highest
yielding racemic syntheses, see: g) D. Stead, P. O'Brien, A. J. Sanderson,
Org. Lett. 2005, 7, 4459–4462; h) J. W. Coe, Org. Lett. 2000, 2, 4205–4208.
The de was estimated by integrating the 13C-NMR signals at
86.71 ppm (for 9) and 86.79 ppm (for epi-9) as described by Matteson,[15]
although we did not employ chromium(III) acetylacetonate.
[19]
[20]
H. C. Brown, M. V. Rangaishenvi, Tetrahedron Lett. 1990, 31, 7115–7118.
Eur. J. Org. Chem. 2016, 958–964
www.eurjoc.org
963
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[21]
[22]
age of the OBn group often required a quick filtration and renewal of
the catalyst.
[25] Following functionalization with (S)-(+)-O-acetylmandelic acid, no sign of
the other diastereomer was detectable; see the Supporting Information.
[26] For reference values for (+)-(α)- and (–)-(α)-pinene, see: J. A. Faraldos,
B. M. Kariuki, R. M. Coates, Org. Lett. 2011, 13, 836–839.
D. S. Matteson, J. J. Yang, Tetrahedron: Asymmetry 1997, 8, 3855–3861.
After one attempt, an impure product was isolated after rinsing the col-
umn with MeOH following the isolation of 11 in only 20 % yield. 1H NMR
analysis of this highly polar material was consistent with a 2:1 borate
adduct of 11 and pinanediol.
[23] D. S. Matteson, K. M. Sadhu, M. L. Peterson, J. Am. Chem. Soc. 1986, 108,
810–819.
[24] To avoid any chance of racemization, aqueous HCl was only added after
the reduction of the alkene was complete as determined by TLC. Cleav-
Received: November 13, 2015
Published Online: January 15, 2016
Eur. J. Org. Chem. 2016, 958–964
www.eurjoc.org
964
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim