New ligands with affinity for the α4β2 subtype of nicotinic acetylcholine receptors. Synthesis, receptor binding, and 3D-QSAR modeling

Article abstract of DOI:10.1021/jm058058h

A new series of piperazines, diazepanes, diazocanes, diazabicyclononanes, and diazabicyclodecanes with affinity for the α₄β ₂ subtype of nicotinic acetylcholine receptors were synthesized on the basis of results from a previous computational study. A predictive 3D-QSAR model was developed using the GRID/GOLPE approach (R² = 0.94, Q ² = 0.83, SDEP = 0.34). The SAR was interpreted in terms of contour maps of the PLS coefficients and in terms of a homology model of the α₄β₂ subtype of the nicotinic acetylcholine receptors. The results reveal that hydrogen bonding from both hydrogens on the protonated amine and from the pyridine nitrogen to a water molecule as well as van der Waals interactions between the substituent bearing the protonated amine and the receptor is of importance for ligand affinity. The combination of 3D-QSAR and homology modeling proved successful for the interpretation of structure-affinity relationships as well as the validation of the individual modeling approaches.

Full text of DOI:10.1021/jm058058h

J. Med. Chem. 2006, 49, 3159-3171
3159
New Ligands with Affinity for the r4â2 Subtype of Nicotinic Acetylcholine Receptors. Synthesis,
Receptor Binding, and 3D-QSAR Modeling
Karine Audouze,† Elsebet Østergaard Nielsen,† Gunnar M. Olsen,† Philip Ahring,† Tino Dyhring Jørgensen,† Dan Peters,†
Tommy Liljefors,‡ and Thomas Balle*,‡
NeuroSearch A/S, 93 PederstrupVej, DK-2750 Ballerup, Denmark, and Danish UniVersity of Pharmaceutical Sciences, 2 UniVersitetsparken,
DK-2100 Copenhagen, Denmark
ReceiVed December 30, 2005
A new series of piperazines, diazepanes, diazocanes, diazabicyclononanes, and diazabicyclodecanes with
affinity for the R4â2 subtype of nicotinic acetylcholine receptors were synthesized on the basis of results
from a previous computational study. A predictive 3D-QSAR model was developed using the GRID/GOLPE
approach (R2 ) 0.94, Q2 ) 0.83, SDEP ) 0.34). The SAR was interpreted in terms of contour maps of the
PLS coefficients and in terms of a homology model of the R4â2 subtype of the nicotinic acetylcholine
receptors. The results reveal that hydrogen bonding from both hydrogens on the protonated amine and from
the pyridine nitrogen to a water molecule as well as van der Waals interactions between the substituent
bearing the protonated amine and the receptor is of importance for ligand affinity. The combination of
3D-QSAR and homology modeling proved successful for the interpretation of structure-affinity relationships
as well as the validation of the individual modeling approaches.
Chart 1
Introduction
Neuronal nicotinic acetylcholine receptors (nAChRs) are
members of a gene superfamily of ligand-gated ion channels
that also comprises 5-HT3, glycine, and GABAA receptors.1 The
nAChRs are composed of five subunits, each spanning the
membrane four times and aligning around the central ion
channel.2 The ion channel of the nAChRs is cation-selective,
and the nicotinic receptors participate in the mediation of fast
excitatory transmission between neurons.3 Furthermore, presyn-
aptic nAChRs modulate the release of other neurotransmitters,4-5
and novel ligands for neuronal nAChRs may have great potential
as pharmaceuticals aiming at several diseases in the CNS.6
Presently, six neuronal R (R2-R7) and three neuronal â (â2-
â4) subunits have been cloned from humans.7-9 Neuronal
nAChRs containing R7 subunits bind R-bungarotoxin with high
affinity.10,11 The current status of the nAChR nomenclatureture
has recently been published.12 Most nAChRs that bind nicotine
with high affinity in the brain are composed of R4 and â2
subunits,13 and the stoichiometry of the receptors has been
14
shown to be (R4)2(â2)3.
In a previous article by Nielsen et al., a 3D-QSAR model
based on 25 ligands, with affinity for the R4â2 subtype of
nAChRs, was described.15 The analysis of the coefficient plots
revealed that steric interactions between the target and the
compounds are of major importance for describing the differ-
ences in affinity. All of the compounds studied include a
pyridine ring and substituents in the pyridine 3-position, in
particular, with bulky ring systems including an sp3 nitrogen
atom, were predicted to increase the affinity of the ligands. In
the present article, we describe the synthesis of a series of
3-substituted pyridines designed on the basis of the 3D-QSAR
model developed by Nielsen et al.15 The novel compounds
comprise piperazines, diazepanes, diazocanes, diazabicy-
clononanes, and diazabicyclodecanes (Chart 1 and Table 1)
because, according to the previous 3D-QSAR model, large
substituents in the pyridine 3-position were predicted to lead to
increased affinity. With the purpose of increasing the robustness
of the previously developed 3D-QSAR model with regard to
the area around the pyridine 3-position and thereby obtaining a
more general model, we have improved the model by including
the new compounds in the training set. The structure-affinity
relationships for the compounds are interpreted using contour
maps of the PLS coefficients obtained from the new 3D-QSAR
analysis. In addition, the results are interpreted in light of a
homology model of the R4â2 receptor reported by Le Novere
et al.16
* To whom correspondence should be addressed. Tel: +45 35 30 64
09. Fax: +45 35 30 60 40. E-mail: tb@dfuni.dk.
† NeuroSearch A/S.
‡ Danish University of Pharmaceutical Sciences.
10.1021/jm058058h CCC: $33.50 © 2006 American Chemical Society
Published on Web 05/05/2006

3160 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
Table 1. Affinities of the Compoundsa
Audouze et al.
lopyridines. The syntheses of seven different 3-pyridylpipera-
zines are described in Scheme 1. Compound 1 was prepared
by a palladium catalyzed reaction of 1-tert-butoxycarbonylpip-
erazine (1b), 3-bromo-5-methoxypyridine (1a), and potassium-
tert-butoxide, followed by deprotection with TFA (Scheme 1).
5-Vinyloxypyridyl analogue 2 was prepared in the absence of
the palladium catalyst from 3-chloro-5-vinyloxypyridine (2d),
potassium tert-butoxide, and unprotected piperazine (2a).
Compounds 3 and 50 (Scheme 1) were prepared by NBS
bromination of the tert-butoxycarbonyl protected piperazines
1c and 50b, respectively, followed by treatment with TFA. The
5-bromo-6-chloropyridine derivative, 4 (Scheme 1), was pre-
pared from the corresponding 5-bromopyridine derivative 4b
by a sequence consisting of tert-butoxycarbonyl protection, NBS
bromination, TFA deprotection, and finally heating in concen-
trated hydrochloric acid to replace the bromine in the 6-position
with a chlorine. The 6-methylpyridyl substituted derivative, 5
(Scheme 1), was obtained from 5a by tert-butoxycarbonyl
protection followed by bromination to give 5c. The methyl group
in the pyridine 6-position was introduced via a palladium
catalyzed methylation reaction, using tetramethyltin as the
methyl source. N-formylation from solvent DMF required
heating in concentrated hydrochloric acid to obtain the final
product. The acetylenic derivative 6 (Scheme 1) was prepared
from 4c by a palladium catalyzed reaction with 2-methyl-3-
butyn-2-ol followed by heating with sodium hydride, which
resulted in a cleavage of the 3-carbon alcoholic fragment
connected to the triple bond.
IC 50
(nM)
compd
R3
R4
obsd.
pred.
training set
1
2
3
4
5
6
7
8
OCH3
OCHdCH2
OCH3
Br
Cl
CtCH
Cl
H
H
Br
Cl
CH3
H
H
H
H
500
190
180
230
520
110
310b
500b
1.3
340
305
209
269
313
269
425
247
2.3
2.0
1.9
5.7
6.4
3.7
3.0
3.5
3.3
16
OCH2CH3
cisOCHdCHCH3
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
54
Cl
Br
Br
Cl
I
H
H
H
H
H
H
H
H
H
H
Br
1.4
1.4
22
5.8
OH
cyclopropyl
3-anilinyl
OCHdCH2
3-pyridyl
SC6H5
CONH2
OH
OCH2CH2OH
C6H5
OCH3
H
H
H
OCH2CH3
OCH3
OCH3
quinoline
cisOCHdCHCH3
OCH2CH3
cisOCHdCHCH3
OCH3
OCH2CH3
Cl
Cl
H
OCH3
O(CH2)3CH3
H
H
4.0b
5.0b
2.8b
4.0b
20b
2.4b
1.8b
3.0b
1.9b
1.0b
170b
170b
0.87b
170b
0.68b
6.6b
1.9
11
1.9
3.4
3.2
2.5
NIb
61
0.59
135
1.1
3.6
1.8
2.8
4.7
6.6
4.2
13
272
265
0.89
1.0
1.0
1.3
0.8
0.43
92
5.7
3.3
1.5
140
SCH2CH3
OCH3
Br
OCH3
Cl
The 1,4-diazepanes 9-13 and 49 were prepared by similar
procedures. Compound 9 (Scheme 1) was obtained from
homopiperazine, (Z)-3-bromo-5-(prop-1-enyloxy)pyridine (9b)
(Scheme 2), and potassium-tert-butoxide. The position of the
double bond and the cis configuration of the final product 9
was confirmed by NMR. This pattern was seen for all of the
cis-propene derivatives (9, 30, and 32). Bromo analogues 10
and 49 (Scheme 1) were both prepared by NBS bromination of
precursors 10b and 49a, respectively, followed by reaction with
TFA to remove the protection group. Chloro and iodo analogues
11 and 12 (Scheme 1) were prepared by halogenation of 11b
and 12d using 1,3-dichlorodimethylhydantoine and N-iodo
succinimide (NIS), respectively. Compound 13 (Scheme 1) was
prepared from its corresponding tert-butoxycarbonyl protected
bromo precursor 11a by a palladium catalyzed reaction with
tributylcyclopropylstannane followed by deprotection with TFA.
H
H
H
H
H
H
H
H
H
H
H
Br
H
Br
H
H
H
H
H
6.5
3.1
10
3.0
8.4
240
190b
1.0
1.8
0.9
0.45
2.0
Cl
OCHdCH2
H
H
H
OCH3
Cl
0.38
100b
6.7b
4.5b
3.0b
220b
test set
1,5-Diazocanes 38-43 and 48 (Scheme 1) were prepared in
a manner analogous to that of the reactions above from two
different precursors, either from tert-butoxycarbonyl derivative
38b, obtained from [1,5]diazocane dihydrobromide17 or from
[1,5]diazocane 39b, also obtained from [1,5]diazocane dihy-
drobromide by treatment with sodium methoxide.
48
49
50
51
52
53
55
56
Br
H
2.8
1.3
0.94
335
3.9
4.1
2.6
8.0
3.3
OCHdCH2
H
OCH2CH2OCH3
OCH2CH(CH3)2
SCH2C5H6
H
Br
Br
H
H
H
0.64
180
2.4b
2.2b
8.0b
1.9
H
A-85380
1.5b
Two different sets of isomers of the 3,9-diazabicyclo[4.2.1]-
nonanes, 30 and 31, were prepared (Scheme 3), which are
connected in their 3-positions with the corresponding prope-
nyloxy and ethoxypyridines and 32-34 (Scheme 4), connected
in their 9-positions with the corresponding propenyloxy and
alkyloxypyridines. The precursor of 30 and 31 was (()-methyl-
3,9-diazabicyclo[4.2.1] nonane (30b), prepared from commercial
tropinone18 (30a). The final step to give 30 and 31 was a
demethylation reaction using diethyl azodicarboxylate. Com-
pounds 32-34 were prepared from benzyltropinone (32a) as
outlined in Scheme 4. 3,10-Diazabicyclo[4.3.1]decane (35)
(Scheme 5), a homoanalogue of the 3,9-diazabicyclo[4.2.1]-
nonanes with the pyridyl connected to its 3-position, was
prepared as outlined in Scheme 5.
a Structures are shown in Chart. b Data from ref 15. NI ) not included
in training set.
Results and Discussion
Chemistry. Compounds 7-8, 14-29, 37, 44-47, and 51-
56 (Chart 1) have been previously reported,15 and the preparation
of compounds 1-6, 9-13, 30-36, 38-43, and 48-50 (Chart
1) is described in the Experimental Section. The key step in
the preparation of all 3-pyridyl substituted diazacycles and
diazabicycles (Chart 1) was the reaction of various diazacy-
cloalkanes (piperazine, 1,4-diazepane and 1,5-diazocane) and
diazabicycloalkanes (3,9-diazabicyclo[4.2.1]nonane and 3,10-
diazabicyclo[4.3.1]-decane) with adequately substituted 3-ha-

R4â2 Subtype of Nicotinic Acetylcholine Receptors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3161
Scheme 1a
a Reagents: (a) t-BuOK; (b) TFA; (c) Boc2O; (d) NBS; (e) Na, 1,2-ethanediol; (f) Et3N; (g) 130°; (h) K2CO3, Cu(I)I, Pd/C, PPh3; (i) HCl; (j) PdCl2(PPh3)2,
tetramethyltin; (k) 1,3-dichloromethylhydantoin; (l) PdCl2(PPh3)2, tributylcyclopropylstannane; (m) NIS; (n) Na; (o) tetrakis(triphenylphosphine)palladium(0);
(p) n-BuONa.
Pyridyl ether 36 (Scheme 6) was prepared by a Mitsunobu
reaction with a Boc-protected alcohol followed by deprotection
using TFA, analogous to previously described procedures.15
Receptor Binding. The predominant nAChR subtype found
in brain tissue is composed of R4â2 subunits, which can be
selectively labeled by the nicotine agonist 3H-cytisine.19,20 The
in vitro affinities of the compounds for the R4â2 subtype of
nAChRs were determined in a 3H-cytisine binding assay using
rat cerebral cortical membranes as described in the Experimental
Section. The IC50 values are given in Table 1.
Comparing the affinities of compounds with a monocyclic
ring system containing the sp3 nitrogen atom and the same
substitution in the pyridyl ring, the data in Table 1 shows that
compound 38 containing an eight-membered ring displays a
somewhat higher affinity (IC50 ) 1.0 nM) than that of the
previously studied seven-membered compound 22 (IC50 ) 1.9

3162 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
Audouze et al.
Scheme 2a
Scheme 4a
a Reagents: (a) Na; (b) 4a.
Scheme 3a
a Reagents: (a) H2SO4, NaN3; (b) LiAlH4; (c) R3ONa; (d) t-BuOK; (e)
diethyl azodicarboxylate.
a Reagents: (a) NaN3, H2SO4; (b) LiAlH4; (c) Et3N, Boc2O; (d) Ac2O;
(e) Pd/C, H2; (f) t-BuOK, 30d (for compound 32f), 33a (for compound
33d), 31a (for compound 34a); (g) HCl; (h) TFA.
nM) and significantly higher affinity than that of the corre-
sponding compound with a six-membered ring system (1, IC50
) 500 nM). Similarly, compound 41 (eight-membered ring, IC50
) 0.45 nM) displays a slightly higher affinity than 23 (seven-
membered ring, IC50 ) 1.0 nM) and an affinity that is 2 orders
of magnitude higher than that of 50 (six-membered ring, IC50
) 180 nM). The eight-membered ring compound 43 displays
the highest affinity (IC50 ) 0.38 nM) in the present series of
compounds. This is in agreement with the prediction from our
previous 3D-QSAR study that a bulky ring system that includes
the ammonium nitrogen atom is favorable for affinity.15 Thus,
larger monocyclic ring systems up to at least eight-membered
rings display increased affinity, although the affinity differences
between eight- and seven-membered rings are significantly
smaller than that between seven- and six-membered rings.
Bicyclononanes 30-34 and bicyclodecane 35 have affinities
(IC50) in the range of 1.9-10 nM, which is similar to those
observed for the previously studied bicyclooctene 45 and
bicyclononenes 46 and 47 (Table 1). The large bicyclic nine-
and ten-membered ring systems are well accommodated by the
receptor but do not increase the affinity compared to that of
the less bulky bicyclooctene and bicyclononenes and display
lower affinities than that of the eight-membered monocyclic ring
compounds.
shows that the new ringsystems are significantly more bulky
than the previously studied ring systems and extend outside the
volume spanning the these ringsystems.
3D-QSAR Analysis. With the purpose of increasing the
robustness of the previous 3D-QSAR model and obtaining a
more general model, we have improved the previous model by
including the new compounds in the training set. The 3D-QSAR
analysis is based on a training set of 47 compounds (1-23,
25-47, 54, and 56) and tested with a validation set of 7
compounds (48-53, 55). In contrast to the study by Nielsen et
al.,15 the thioethyl-substituted homopiperazine 24 was excluded
from the training set. The substituent in this compound is one
of a kind, and the bioactive conformation is ambiguous. The
compounds were aligned to homopiperazine 55 in the proposed
bioactive conformation calculated by Nielsen et al.15 using the
centroid of the pyridine ring, the pyridine nitrogen, the proto-
nated nitrogen, and the two protons attached to this nitrogen as
fitting points. For each compound, a conformational analysis
was performed by using the Macromodel program.21 The
calculations were performed for the N-protonated compound
in aqueous solution. The low energy conformer giving the best
fit to template compound 55 was used for the 3D-QSAR analysis
(for details see the Experimental Section). A superimposition
of compounds 5, 25, 30, 32, 35, 36, 40, 44-46, and 56,
representing the 11 different structural classes included in the
present study, is shown in Figure 1.
Figure 1 shows a least-squares superimposition of compounds
containing the previously studied15 and the new rings systems.
The superimposition is obtained according to the procedure
described by Nielsen et al.15 by superimposing the centroid of
the pyridine ring, the pyridine nitrogen, the protonated nitrogen,
and the two protons attached to this nitrogen using homopip-
erazine 55 as a template. The procedure is described below and
in the Experimental Section in more detail. Figure 1 clearly
The molecular descriptors were generated with the GRID
program22 by calculating the interaction energies between two
probes, water (OH2) and methyl (C3), and the target molecule
as described by Nielsen et al.15 Charged probes O- and N1+
do not improve the quality of the model.15

R4â2 Subtype of Nicotinic Acetylcholine Receptors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3163
Scheme 5a
a Reagents: (a) benzylamine, glutaric dialdehyde, H2O; (b) NaN3, H2SO4; (c) LiAlH4; (d) Et3N, Boc2O; (e) Pd/C, H2; (f) 3,5-dicholoropyridine, t-BuOK;
(g) TFA.
Scheme 6a
a Reagents: (a) Boc2O, Et3N, (b) diethyl azodicarboxylate, PPh3, (c) TFA.
Figure 2. Experimental and predicted affinities (pIC50) for the R4â2
nAChR. Training set (0); validation set (2).
selection was performed using the smart region definition25
(SRD) in combination with fractional factoral design (FFD)
procedures implemented in GOLPE. The SRD procedure groups
the variables close in space and collapses the groups of variables
containing virtually the same information to smart regions. These
regions are then evaluated using the FFD procedure. In the FFD
procedure, a large number of models are calculated with some
regions left out to evaluate the importance of the individual
regions. Only regions contributing in a positive manner to the
predictivity of the model are included in the final model. The
FFD procedure reduced the number of variables to 1413 with
a significant improvement of the predictivity of the model as
determined by Q2 (from Q2 ) 0.68 to Q2 ) 0.83, Table 2). The
model was further tested using an external validation set (i.e.,
a set of compounds not included in the training set). The
affinities of these compounds were predicted with a standard
deviation on error of prediction (SDEP) value of 0.34. The
experimental and calculated affinities for the compounds in the
training set and in the external validation set are listed in Table
1 and graphically shown in Figure 2.
Interpretation of Contour Plots. With the aim of interpreting
the 3D-QSAR model, the weighed PLS coefficients from the
most informative first component were plotted for both probes
together with diazabicyclononane 30 to illustrate the size of the
regions. The negative and positive fields for the water (OH2)
probe are shown in Figure 3a, and the corresponding plot of
the fields for the C3 probe are shown in Figure 3b. The two
sets of plots appear similar to each other with only minor
differences, suggesting that steric factors are of major impor-
tance in explaining the differences in the observed affinities of
the molecules. The contribution of the OH2 probe to the PLS
model is ambiguous because this probe, apart from the hydrogen
Figure 1. Alignment of compounds 5, 25, 30, 32, 35, 36, 40, 44-46,
and 56. Compounds 30, 32, 35, and 40, containing the new ring systems,
are displayed with orange carbon atoms. Nonpolar hydrogens are
removed for clarity.
Table 2. Properties of the 3D-QSAR Models
no. of
variables
LVS
R2
q2
SDEPa
after pretreatment
after FFD selection
4314
1413
3
3
0.92
0.94
0.68
0.83
0.39
0.34
a Standard deviation on error of prediction (SDEP) for external validation
set (log(10) units), SDEP ) (Σ(YExp - YCalc)2/N).
The 3D-QSAR model was calculated using the program
GOLPE.23,24 The properties of the calculated model are sum-
marized in Table 2. A large number of the descriptors (variables)
calculated by GRID do not contribute to the explanation of the
observed differences in affinity and can, therefore, be character-
ized as noise. On the basis of the variable pretreatment protocol
described in the Experimental Section, the initial 11 638
variables were reduced to 4314 without affecting the predictivity
of the 3D-QSAR model (as judged by Q2). Further variable

3164 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
Audouze et al.
Figure 3. (a) Contour maps of the PLS coefficients from the first
component for the OH2 probe. Negative coefficients are shown in blue
(-0.001 level), and positive coefficients are shown in red (+0.0006
level). Diazabicyclononane 30 is shown to illustrate the size of the
region. (b) Contour maps of the PLS coefficients from the first
component for the C3 probe. Negative coefficients are shown in blue
(-0.001 level), and positive coefficients are shown in red (+0.0006
level). Diazabicyclononane 30 is shown to illustrate the size of the
region.
Figure 4. Contour map for the positive PLS coefficients (+0.0006
level) for the C3 probe and diazabicyclononane 30 manually docked
into the binding pocket of the homology model of the R4â2 nicotinic
receptor. Amino acids Val 109, Tyr 195, Tyr 188, Tyr 91, Trp 55, and
Trp 147 and a water molecule are highlighted.
bonding component, also contains a steric component. Conse-
quently, the regions of interest for the OH2 probe are the regions
in space surrounding the molecules where the plot deviates from
the corresponding plot of the C3 probe. The largest deviations,
marked A, B, and C in Figure 3a, are seen in the regions with
negative coefficients (blue regions). A favorable interaction (e.g.,
a hydrogen bonding interaction) between the molecule and a
water probe in these regions is predicted to lead to increased
affinity. The most important of these regions (most negative
coefficients) is region A, which suggests that a hydrogen bond
from the protonated amine to the receptor is of major importance
in explaining the observed differences in affinity. Another major
difference is region C, suggesting that the pyridine nitrogen has
an important electrostatic interaction with the receptor and
finally region B, suggesting that the second hydrogen on the
protonated amine has a hydrogen bond to the receptor, although
of lower importance compared to that of region A.
The most dominant regions in the plot for the C3 probe in
Figure 3b are the positive regions (shown in red). An unfavor-
able interaction (steric clash) betwen the molecule and the probe
in these regions is predicted to lead to increased affinity. The
most important positive region surrounds the substituents
holding the protonated amine and suggests that a large sub-
stituent in this position leads to increased affinity. This is evident
when comparing piperazine 1 (six-membered ring) with corre-
sponding diazocane 38 (eight-membered ring), the latter being
more potent by a factor of 500. Another important positive
region is in the area of the R3-substituents, suggesting that large
substituents in this position are tolerated and could lead to a
slightly increased affinity. This is evident when comparing
homopiperazines 15 (R3: OCHdCH2) and 9 (R3: OCHdCH2-
CH3), the latter having a 4-fold higher affinity.
The present model complements and extends the previous
model by Nielsen et al. by including more compounds with
larger substituents containing the protonated nitrogen, thus
improving the predictivity of the model with regard to such
compounds. One major difference is the size of the region of
negative coefficients for the C3 probe (shown in blue) around
the R4-substituent, which is large in the model by Nielsen et
al. and small in the present model. An unfavorable interaction
between a molecule and the methyl probe in areas with negative
coefficients is predicted to lead to decreased affinity. The
difference between the two models is due to homopiperazine
24, which was removed from the present study because of the
ambiguous orientation of the thioethyl side chain.
bioactive conformations of the compounds, then the positive
and negative regions calculated by GOLPE should represent
interactions in the binding pocket of the R4â2 receptor. To
illustrate this, diazabicyclononane 30 was manually docked into
the homology model of the R4â2 receptor reported by Le
Novere.16 The docking was guided by the binding mode and
interactions displayed by nicotine in the binding pocket of the
acetylcholine binding protein (AChBP, pdb code: 1UW6).26 The
nicotine-AChBP complex displays a water molecule that is
hydrogen bonding to the pyridyl nitrogen atom of nicotine. Thus,
we introduced a water molecule at the corresponding position
in the homology model of R4â2. Compound 30 was docked into
the binding pocket in the conformation employed in the 3D-
QSAR analysis with the constraints of forming a hydrogen bond
between its pyridyl nitrogen and the introduced water molecule
and between the backbone carbonyl of Trp147 and one of the
ammonium group hydrogen atoms of 30. It is then observed
that in this binding mode the ammonium group of 30 could
also form a hydrogen bond to Tyr91. In Figure 4, compound
30 together with a plot of the positive PLS coefficients for the
C3 probe is shown as docked into the binding cavity of R4â2.
The highlighted amino acids, Val109, Tyr195, Tyr 188, Tyr91,
Trp55, Trp 147, and the water molecule form a cavity sur-
rounding the plot of the positive PLS-coefficients displayed at
the +0.0006 contour level. The Figure nicely illustrates that
there are important lipophilic interactions between the five
aromatic residues and the ringsystems containing the protonated
nitrogen atom. The regions with positive coefficients around
the R3 substituent extend into a small channel flanked by
Val109. The size and shape of this channel extends beyond the
borders of the present 3D-QSAR model, suggesting that the
affinity could be increased further by extending the chain length
of the R3 substituent.
The proposed binding mode and interactions of 30 are
strongly supported by a recent crystal structure of a AChBP-
epibatidine complex.27 In particular, the ammonium group of
epibatidine forms the two hydrogen bonds shown in Figure 4,
and the pyridyl group is located in a position very close to that
displayed by 30. Because water molecules are not resolved in
the crystal structure (3.4 Å resolution), the presence of a water
molecule interacting with the pyridyl group cannot be confirmed
by this structure. However, when comparing the position of the
water molecule in Figure 4 with the negative region C in Figure
3a, it becomes obvious that region C represents a water molecule
Receptor Interactions. If the alignment of the compounds
studied is a good representation of the binding modes and

R4â2 Subtype of Nicotinic Acetylcholine Receptors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3165
Furthermore, lipopilic interactions between the ringsystems
containing the protonated nitrogen seem to be of major
importance with regard to the difference in the affinities of the
described compounds. A previous 3D-QSAR study of R4â2
ligands related to epibatidine, cytosine, anatoxin-a, and ferrugi-
nin using the CoMFA and CoMSIA methods resulted in high
quality models.30 However, the small variations in the data set
with respect to the size of the ringsystems containing the sp3
nitrogen atom could not properly map the importance of steric
properties on the affinities of various ring systems.
The strength of the present study is the interpretability of
the 3D-QSAR results. The combination of different methods
has proven very successful for the validation of the individual
modeling approaches. The previously reported homology model
of the R4â2 receptor and the coefficient plots obtained from the
present 3D-QSAR study complement each other. When manu-
ally docked into the receptor model, the important fields of the
coefficient plot overlay the areas of the homology model known
from crystallographic ligand-receptor complexes to be impor-
tant for ligand binding.
Figure 5. Molecular superimposition of low affinity piperazine
compound 5 and high affinity diazabicyclononane 30.
in the receptor. Furthermore, the hydrogen bonds from the
protonated amine to the backbone carbonyl of Trp A147 and
to the oxygen of Tyr A91 are represented in the QSAR model
by regions A and B, respectively. The relative importance of
region A compared to that of B may be explained by the
backbone carbonyl being a much better hydrogen bond acceptor
compared to the phenol of Tyr A91.
Experimental Section
When interpreted in light of the 3D-QSAR model discussed
above and in light of the homology model of the R4â2 receptor,
it is easy to understand why piperazines 1-8 and 50 all have
low affinities. The hydrogens of the protonated amine has a
nonoptimal direction for interaction with the two hydrogen bond
acceptors in the receptor (corresponding to field A and B in
Figure 3a) as displayed in Figure 5. Also, as shown in Figure
5, the piperazine ring lies closer to the areas of negative
coefficients for the C3 probe (blue areas, Figure 3b) compared
to the diazabicyclononane ring in 30. The interaction between
a molecule and the C3 probe in these areas is predicted to lead
to decreased affinity. Similarly, the bulk of the piperidine ring
in piperidin-3-yl-oxy derivatives 36 and 37 and the pyrrolidine
ring of pyrrolidin-3-yl-oxy derivatives 44 and 54 are situated
close to the negative areas, which may explain the low affinities
(>100 nM) of these compounds.
The compounds substituted with the most bulky ring systems,
that is, bicyclic derivatives 30-35 and 45-47, all bind with
high affinity to the R4â2 receptor. These compounds have an
almost perfect hydrogen-bonding geometry as indicated for
diazabicyclo-nonane 30, depicted in Figure 4. Furthermore, the
compounds at least partly occupy the favorable area of positive
coefficients for the C3 probe shown in Figure 3b. The
compounds with the highest affinity are homopiperazine 28 and
diazocanes 40, 41, and 43, all binding to the R4â2 receptor with
subnanomolar affinity. Despite being slightly smaller than the
bicyclic derivatives discussed above, these compounds also form
almost perfect hydrogen bonds with the receptor and occupy
space in the predicted favorable region shown in red in Figures
3b and 4.
Chemistry. 3-Bromo-5-methoxypyridine (1a). Compound 1a
was prepared from a mixture of 4a (25.0 g, 105.0 mmol), sodium
methoxide (2.91 g, 126.0 mmol), and DMSO (100 mL) stirred at
90 °C for 3 h. Aqueous NaOH (400 mL, 1 M) was added, and the
mixture was extracted with diethyl ether (2 × 200 mL). Yield 12.1
g (61%); mp 35.2-36.8 °C.
4-(5-Methoxypyridin-3-yl)-piperazine-1-carboxylic Acid tert-
Butyl Ester (1c). Procedure A. Compound 1c was obtained from
a mixture of tetrakis(triphenylphosphine)palladium(0) (0.11 g, 0.10
mmol), 1-tert-butoxycarbonylpiperazine (1b) (5.20 g, 28.0 mmol),
and 3-bromo-5-methoxypyridine (1a) (5.30 g, 28.0 mmol), prepared
according to Nielsen et al.,15 and potassium tert-butoxide (t-BuOK)
(6.20 g, 56.0 mmol) stirred in anhydrous toluene (40 mL) at 80 °C
for 3 h. The reaction mixture was cooled to room temperature.
Aqueous sodium hydroxide (NaOH) (80 mL, 1 M) was added, and
the mixture was extracted with ethyl acetate (EtOAc) (2 × 80 mL).
CC with dichloromethane (CH2Cl2), methanol (MeOH), and
concentrated aqueous ammonia (89:10:1) gave 1c as an oil. Yield
2.59 g (31%).
1-(5-Methoxypyridin-3-yl)-piperazine Fumaric Acid Salt (1).
Procedure B. Compound 1 was prepared from a solution of 1c
(2.50 g, 8.50 mmol), TFA (19.4 g, 170.0 mmol), and CH2Cl2 (30
mL) stirred at 20 °C for 4 h. The mixture was evaporated to dryness,
and aqueous NaOH (20 mL, 1 M) was added. The mixture was
extracted with CH2Cl2 (4 × 20 mL). CC with CH2Cl2, MeOH, and
concentrated aqueous ammonia (89:10:1) gave 1. Yield 0.43 g
(16%); mp 168.5-170.5 °C. The corresponding salt was obtained
by the addition of a diethyl ether and MeOH mixture (9:1) saturated
with fumaric acid. Anal. (C10H15N3O‚0.5C4H4O4‚H2O) C, H, N.
2-(5-Chloropyridin-3-yloxy)-ethanol (2c). Compound 2c was
prepared using 1,2-ethanediol (135 mL) and sodium (6.20 g, 268.0
mmol) stirred at 80 °C for 3 h. 3,5-Dichloropyridine (2b) (33.0 g,
223.0 mmol) and DMSO (100 mL) were added, and the mixture
was stirred at 90 °C overnight. Aqueous NaOH (200 mL, 1 M)
was added and the compound was extracted with EtOAc (3 × 150
mL). Yield 14.5 g (36%); mp 41.6-43.6 °C.
3-Chloro-5-vinyloxypyridine (2d). Compound 2d was prepared
from a mixture of 2c (14.5 g, 82.1 mmol) and thionyl chloride
(SOCl2) (58.6 g, 0.49 mol) stirred in THF (100 mL) at 50 °C for
1 h. The mixture was evaporated to dryness. Potassium hydroxide
(KOH) (100 mL, 4 M) and tert-butyl alcohol (t-BuOH) (100 mL)
were added, and the mixture was stirred at 100 °C overnight. The
mixture was evaporated. Water (100 mL) was added and the mixture
extracted with diethyl ether (2 × 100 mL). CC with CH2Cl2. The
product was obtained as an oil. Yield 9.44 g (74%).
Conclusions
In the present study, the preparation of a series of new ligands
with affinity for the R4â2 subtype of the nicotinic receptor is
reported. The compounds were designed on basis of a previously
reported 3D-QSAR model and revealed, in particular, a series
of diazocanes, 40, 41, and 43, with subnanomolar affinity for
the R4â2 subtype of the nicotinic receptor. A combination of
3D-QSAR and homology modeling was used to interpret the
structure-affinity relationships of the ligands. These methods
revealed that hydrogen bonding from both protons of the
protonated amine to the receptor as well as from the pyridine
nitrogen to a water molecule is important for ligand binding.
1-(5-Vinyloxypyridin-3-yl)-piperazine Fumaric Acid Salt (2).
Compound 2 was prepared according to procedure A from a mixture

3166 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
Audouze et al.
of 2d (2.00 g, 12.9 mmol), piperazine (2a) (2.20 g, 25.7 mmol),
t-BuOK (3.63 g, 25.7 mmol), and DME (50 mL). Yield 1.26 g
(45%). The corresponding salt was obtained by the addition of a
diethyl ether and MeOH mixture (9:1) saturated with fumaric acid;
mp 136.4-138.2 °C. Anal. (C11H15N3O‚1.2C4H4O4) C, H, N.
4-tert-Butoxycarbonyl-1-(6-bromo-5-methoxypyridin-3-yl)-
piperazine (3a). Compound 3a was prepared from compound 1
(23.1 g, 74.7 mmol), aqueous sodium hydrogen carbonate (224 mL,
1 M), and Boc2O (16.3 g, 74.7 mmol) stirred in CH2Cl2 (200 mL)
stirred at room temperature for 1 h. The phases were separated.
NBS (1.80 g, 10.2 mmol) and acetonitrile (50 mL) were added to
the organic phase at 0 °C. The reaction was allowed to warm to
room temperature and stirred for 2 h. Aqueous NaOH (100 mL, 4
M) was added, and the mixture was extracted with EtOAc (2 ×
100 mL). CC (petroleum ether/ EtOAc (3:1)) gave 3a. Yield 2.59
g (59%); mp 169.9-171.3 °C.
1-(6-Bromo-5-methoxypyridin-3-yl)-piperazine Fumaric Acid
Salt (3). Compound 3a (2.30 g, 6.23 mmol) was deprotected
according to procedure B with TFA (7.30 g, 64.0 mmol) and CH2-
Cl2 (30 mL). Yield 2.28 g (100%). The corresponding salt was
obtained by the addition of a diethyl ether and MeOH mixture (9:
1) saturated with fumaric acid; mp 202-218 °C. Anal. (C10H14N3-
OBr‚1.2C4H4O4‚0.5H2O) C, H, N.
1-(5-Bromopyridin-3-yl)-piperazine (4b). Compound 4b was
prepared from a mixture of 3,5-dibromopyridine (4a) (100 g, 422.0
mmol) and 2a (72.7 g, 844.0 mmol) stirred at 130 °C for 100 h.
H2O (200 mL) was added at 100 °C followed by HCl (4 M) to
obtain pH 7. Aqueous NaOH (200 mL, 1 M) was added followed
by extraction with EtOAc (3 × 600 mL). The product was isolated
as an oil. Yield 44.0 g (43%).
1-(5-Bromopyridin-3-yl)-piperazine-1-carboxylic Acid tert-
Butyl Ester (4c). Compound 4c was prepared according to
procedure D from a mixture of 4b (3.00 g, 12.4 mmol), Et3N (2.51
g, 24.8 mmol), Boc2O (2.70 g, 12.4 mmol), and CH2Cl2 (30 mL).
Yield 3.8 g (89%); mp 112.9-114.1 °C.
mmol), and DMF (2.0 mL) stirred at 160 °C in a sealed stainless
steel vessel overnight. The reaction mixture was stirred in concen-
trated hydrochloric acid (30 mL) at reflux for 15 min. The mixture
was evaporated. Aqueous NaOH (50 mL, 1 M) was added, and the
mixture was extracted with CH2Cl2 (3 × 50 mL). CC (CH2Cl2,
MeOH and concentrated ammonia (89:10:1)) gave compound 5.
Yield 0.70 g (43%). The corresponding salt was obtained by the
addition of a diethyl ether and MeOH mixture (9:1) saturated with
fumaric acid; mp 178.1-180.5 °C. Anal. (C10H14N3Cl‚1.4C4H4O4‚
0.5H2O) C, H, N.
1-(5-Ethynylpyridin-3-yl)-4-tert-butoxycarbonyl-piperazine (6a).
Compound 6a was prepared from a mixture of 4c (53.5 g, 156.0
mmol), K2CO3 (54.0 g, 391.0 mmol), copper(I) iodide (5.90 g, 31.3
mmol), Pd/C 5% (20.0 g, H2O, 50%), triphenylphosphine (PPh3)
(4.10 g, 15.6 mmol), 2-methyl-3-butyn-2-ol (131 g, 1.60 mol), NaH
(100 mg), and dioxane (300 mL) stirred at reflux for 10 days. The
crude mixture was filtered through Celite and isolated by CC
(petroleum ether/EtOAc (1:3) to give the corresponding alcohol.
The alcohol was stirred with sodium hydride (0.50 g, 23.8 mmol)
in toluene (200 mL) at reflux 2 days. Compound 6a was isolated
by CC (petroleum ether/EtOAc (1:3). Yield 16.6 g (35%); mp 129-
130.7 °C.
1-(5-Ethynyl-pyridin-3-yl)-piperazine Fumaric Acid Salt (6).
Compound 6 was prepared according to procedure B from a mixture
of 6a (1.02 g, 3.50 mmol), TFA (4.00 g, 35.3 mmol), and CH2Cl2
(10 mL). Yield 0.60 g (91%). The corresponding salt was obtained
by the addition of a diethyl ether and MeOH mixture (9:1) saturated
with fumaric acid; mp 175.1 °C. Anal. (C11H13N3‚1.2C4H4O4) C,
H, N.
(Z)-3-bromo-5-(prop-1-enyloxy)pyridine (9b). Compound 9b
was prepared from allyl alcohol (9a) (12.2 g, 210.0 mmol) and
sodium (Na) (1.20 g, 50.2 mmol) stirred at 80 °C for 3 h. Compound
4a (10.0 g, 42.0 mmol) and DMSO (50 mL) were added to the
mixture and stirred at 80 °C for 30 min. Aqueous NaOH (100 mL,
1M) was added, and the product was extracted with diethyl ether
(2 × 100 mL). The product was obtained as an oil. Yield 8.0 g
(65%).
1-(5,6-Dibromopyridin-3-yl)-piperazine-1-carboxylic Acid tert-
Butyl Ester (4d). Compound 4d was prepared according to
procedure F from a mixture of 4c (3.75 g, 11.0 mmol), NBS (1.94
g, 11.0 mmol), and acetonitrile (60 mL) stirred at 0 °C for 3 h.
Aqueous NaOH (100 mL, 1 M) was added, and the mixture was
extracted with EtOAc (100 mL). Yield 4.25 g (91%); mp 140.9-
142.7 °C.
1-(5,6-Dibromopyridin-3-yl)-piperazine (4e). Compound 4e
was prepared according to procedure B from a mixture of 4d (4.20
g, 10.0 mmol), TFA (11.3 g, 99.7 mmol), and CH2Cl2 (40 mL).
Yield 1.9 g (43%); mp 199.5 °C.
cis-1-(5-Propen-1-yloxypyridin-3-yl)-[1,4]diazepane Fumaric
Acid Salt (9). Compound 9 was prepared according to procedure
A from a mixture of 9b (10.0 g, 46.7 mmol), homopiperazine (9c)
(23.4 g, 233.0 mmol), t-BuOK (11.4 g, 93.4 mmol), and tetrakis-
(triphenylphosphine)palladium(0) (20.0 mg, 18.0 mol) stirred in
DME (100 mL) at reflux for 1.5 h. Yield 3.30 g (30%); mp 124-
126 °C. The corresponding salt was obtained by the addition of a
diethyl ether and MeOH mixture (9:1) saturated with fumaric acid.
Anal. (C13H19N3O‚C4H4O4) C, H, N.
1-(5-Chloropyridin-3-yl)-[1,4]diazepane (10a). Procedure E.
Compound 10a was prepared from 3,5-dichloropyridine (2b) (60.0
g, 405.0 mmol), homopiperazine (40.0 g, 405.0 mmol), t-BuOK
(68.0 g, 608.0 mmol), and DME (500 mL) stirred at room
temperature for 1.5 h. Aqueous NaOH (100 mL, 1 M) was added,
and the mixture was extracted with EtOAc (2 × 150 mL). The
product was obtained as an oil. Yield 16.4 g (19%).
4-(5-Chloropyridin-3-yl)-[1,4]diazepane-1-carboxylic Acid tert-
Butyl Ester (10b). Compound 10b was prepared according to
procedure D from a mixture of 10a (16.4 g, 77.5 mmol), Boc2O
(16.9 g, 77.6 mmol) and Et3N (7.90 g, 77.6 mmol). Yield 10.2 g
(42%); mp 63.8-65.6 °C.
4-(6-Bromo-5-chloropyridin-3-yl)-[1,4]diazepane-1-carboxyl-
ic Acid tert-Butyl Ester (10c). Procedure F. Compound 10c was
prepared from a mixture of 10b (2.30 g, 7.50 mmol) and NBS (1.34
g, 7.50 mmol) stirred in acetonitrile (75 mL) for 1 h at room
temperature. Compound 10c was obtained by CC (petroleum ether/
EtOAc; 2:1). Yield 2.50 g (85%); mp 113-114 °C.
1-(6-Bromo-5-chloropyridin-3-yl)-[1,4]diazepane Fumaric Acid
Salt (10). Compound 10 was prepared according to procedure B
from a mixture of 10c (2.50 g, 6.40 mmol), TFA (7.30 g, 64.0
mmol), and CH2Cl2 (65 mL) for 6 h. Yield 1.75 g (94%). The
corresponding salt was obtained by addition of a diethyl ether and
MeOH mixture (9:1) saturated with fumaric acid; mp 213-216 °C.
Anal. (C10H13N3BrCl‚C4H4O4) C, H, N.
1-(5-Bromo-6-chloropyridin-3-yl)-piperazine Fumaric Acid
Salt (4). Compound 4 was prepared from a mixture of 4e (1.50 g,
4.70 mmol) and concentrated hydrochloric acid (30 mL) stirred
for 52 h at reflux. The reaction mixture was evaporated, aqueous
NaOH (100 mL, 1 M) was added, and the mixture was extracted
with EtOAc (2 × 50 mL). Yield 1.00 g (54%). The corresponding
salt was obtained by the addition of a diethyl ether and MeOH
mixture (9:1) saturated with fumaric acid; mp 181.2-185.9 °C.
Anal. (C9H11N3BrCl‚1.2C4H4O4) C, H, N.
1-(5-Chloropyridin-3-yl)-4-tert-butoxycarbonylpiperazine (5b).
Compound 5b was prepared according to procedure D from a
mixture of 1-(5-chloropyridin-3-yl)piperazine (5a) (8.10 g, 41.0
mmol), prepared according to Nielsen et al.,15 Boc2O (8.90 g, 41.0
mmol), Et3N (8.30 g, 82.0 mmol), and CH2Cl2 (80 mL). The product
was obtained as an oil. Yield 11.9 g (98%).
1-(6-Bromo-5-chloropyridin-3-yl)-4-tert-butoxycarbonylpip-
erazine (5c). Compound 5c was prepared according to procedure
F from a mixture of 5b (11.9 g, 40.0 mmol), NBS (7.10 g 40.0
mmol), and acetonitrile (150 mL) stirred at 20 °C for 100 h.
Aqueous NaOH (200 mL, 1 M) was added, and the product was
extracted with diethyl ether (200 mL). Yield 10.7 g (71%); mp
107.4-109.2 °C.
1-(5-Chloro-6-methylpyridin-3-yl)-piperazine Fumaric Acid
Salt (5). Compound 5 was prepared using 5c (2.00 g, 5.30 mmol),
tetramethyltin (1.90 g, 10.6 mmol), PdCl2(PPh3)2 (0.18 g, 0.26

R4â2 Subtype of Nicotinic Acetylcholine Receptors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3167
4-(5-Bromopyridin-3-yl)-[1,4]diazepane (11a). Compound 11a
was prepared according to procedure E from a mixture of 4a (60.0
g, 253.0 mmol), 9c (25.0 g, 253.0 mmol), t-BuOK (42.0 g, 380.0
mmol), and DME (600 mL). The product was obtained as an oil.
Yield 19.8 g (30%).
4-(5-Bromopyridin-3-yl)-[1,4]diazepane-1-carboxylic Acid tert-
Butyl Ester (11b). Compound 11b was prepared according to
procedure E from a mixture of 11a (19.0 g, 74.2 mmol), Boc2O
(16.2 g, 74.2 mmol), Et3N (7.50 g, 74.2 mmol), and CH2Cl2 (200
mL). Yield 14.6 g (55%); mp 58.1-60.7 °C.
4-(5-Bromo-6-chloropyridin-3-yl)-[1,4]diazepane-1-carboxyl-
ic Acid tert-Butyl Ester (11c). Compound 11c was prepared
according to procedure A from a mixture of 11b (2.68 g, 7.50
mmol), 1,3-dichlorodimethylhydantoin (0.88 g, 4.50 mmol), and
acetonitrile (75 mL) for 45 min. Compound 11c was obtained by
CC (petroleum ether/EtOAc; 3:1). The product was isolated as an
oil. Yield 0.80 g (51%).
4-(5-Bromo-6-chloropyridin-3-yl)-[1,4]diazepane Fumaric Acid
Salt (11). Compound 11 was prepared according to procedure B
from a mixture of 11c (0.80 g, 2.00 mmol), TFA (2.10 g, 19.0
mmol), and CH2Cl2 (20 mL) stirred for 2 h. Compound 11 was
isolated by CC. Yield 0.23 g (40%). The corresponding salt was
obtained by the addition of a diethyl ether and MeOH mixture (9:
1) saturated with fumaric acid; mp 188.2-189.2 °C. Anal.
(C10H13N3ClBr‚1.2C4H4O4‚0.1H2O) C, H, N.
3-Chloro-5-methoxymethoxypyridine (12b). Compound 12b
was prepared from a mixture of 5-chloro-3-hydroxypyridine (12a)
(10.0 g, 77.2 mmol), bromomethoxymethane (10.6 g, 84.9 mmol),
K2CO3 (10.6 g, 77.2 mmol), and DMF (100 mL) stirred 45 min at
70 °C. Aqueous NaOH (100 mL, 1 M) was added. The product
was extracted with diethyl ether (3 × 100 mL) and isolated as an
oil. Yield 10.0 g (75%). The product was extracted as an oil.
4-(5-Methoxymethoxypyridin-3-yl)-[1,4]diazepane (12c). Com-
pound 12c was prepared from a mixture of 12b (10.0 g, 57.6 mmol),
homopiperazine (28.8 g, 288.0 mmol), t-BuOK (12.9 g, 115.0
mmol), palladium-dichloro-bistriphenylphosphine (PdCl2(PPh3)2)
(0.18 g, 0.26 mmol), and DME (100 mL) stirred at reflux for 3 h.
Aqueous NaOH (100 mL, 1 M) was added, and the product was
extracted with EtOAc (2 × 150 mL). Yield 9.7 g (71%); mp 129-
131 °C.
4-(5-Methoxymethoxypyridin-3-yl)-[1,4]diazepane-1-carboxyl-
ic Acid tert-Butyl Ester (12d). Compound 12d was prepared from
a mixture of 12c (1.00 g, 2.80 mmol), Boc2O (0.60 g, 2.80 mmol),
sodium hydrogen carbonate (10 mL, 1 M), and CH2Cl2 (10 mL).
Compound 12d was obtained by CC (MeOH /EtOAc (1:9)). The
product was obtained as an oil. Yield 0.79 g (83%).
4-(6-Iodo-5-methoxymethoxypyridin-3-yl)-[1,4]diazepane-1-
carboxylic Acid tert-Butyl Ester (12e). Compound 12e was
prepared according to procedure F from a mixture of 12d (0.80 g,
2.30 mmol), N-iodosuccinimide (0.50 g, 2.30 mmol), and aceto-
nitrile (10 mL). The product was isolated as an oil. Yield 0.87 g
(82%).
was obtained by the addition of a diethyl ether and MeOH mixture
(9:1) saturated with fumaric acid; mp 155.7 °C. Anal. (C13H19N3‚
1.5C4H4O4) C, H, N.
(()-9-Methyl-3,9-diazabicyclo[4.2.1]nonan-4-one (30b). Com-
pound 30b was prepared according to Michaels et al.,18 using
tropinone (30a) (40.0 g, 287.0 mmol), sodium azide (NaN3) (37.3
g, 574.0 mmol), and concentrated H2SO4 (100 mL). The product
was obtained as an oil. Yield 18.5 g (42%).
(()-9-Methyl-3,9-diazabicyclo[4.2.1]nonane (30c). Compound
30c was prepared according to Michaels et al.18 using 30b (255 g,
1.66 mol), diethyl ether (7.0 L), and lithium aluminum hydride
(LiAlH4) (158 g, 4.16 mol). Yield 174 g (75%); bp 90 °C (18
mmHg).
3-Chloro-5-propen-cis-yloxypyridine (30d). Compound 30d
was prepared from allyl alcohol (12.2 g, 210.0 mmol) and Na (1.20
g, 50.2 mmol) stirred at 80 °C for 3 h. 3,5-Dichloropyridin (25.0
g, 168.0 mmol) and DMSO (75 mL) were added, and the mixture
was stirred at 60 °C for 1 h. H2O (150 mL) was added, and the
product was extracted with diethyl ether (50 mL) and EtOAc (2 ×
100 mL). The product was obtained as an oil. Yield 29 g (100%).
(()-cis-9-Methyl-3-(5-propen-1-yl-oxypyridin-3-yl)-3,9-diaza-
bicyclo [4.2.1]nonane (30e). Compound 30e was prepared accord-
ing to procedure A from a mixture of 30d (3.63 g, 21.4 mmol),
sodium tert-butoxide (2.70 g, 28.6 mmol), and 30c (2.00 g, 14.3
mmol) stirred in DME (30 mL) at 20 °C for 1 h. t-BuOK (3.20 g,
28.6 mmol) was added, and the mixture was stirred at 20 °C for
72 h. Yield 0.92 g (24%); mp 132-134 °C.
(()-cis-3-(5-Propen-1-yloxypyridin-3-yl)-3,9-diazabicyclo-
[4.2.1]nonane Fumaric Acid Salt (30). Procedure C. Compound
30 was prepared from a mixture of 30e (0.60 g, 2.20 mmol), diethyl
azodicarboxylate (1.96 g, 11.2 mmol) and toluene (10 mL). The
mixture was stirred at reflux for 30 min and cooled to room
temperature. Aqueous hydrochloric acid (20 mL, 4 M) was added,
followed by stirring for 5 min. Aqueous NaOH (25 mL, 4 M) was
added, and the mixture was extracted with EtOAc (2 × 20 mL).
Compound 30 was isolated by CC. Yield 0.15 g (27%). The
corresponding salt was obtained by the addition of a diethyl ether
and MeOH mixture (9:1) saturated with fumaric acid; mp 144-
146. Anal. (C15H21N3O‚1.3C4H4O4‚2H2O) C, H, N.
3-Chloro-5-ethoxypyridine (31a). Compound 31a was prepared
according to Nielsen et al.15 using 2b (295 g, 1.99 mol). Yield 278
g (89%); bp 95-105 °C (16 mmHg).
(()-3-(5-Ethoxypyridin-3-yl)-9-methyl-3,9-diazabicyclo[4.2.1]-
nonane (31b). Compound 31b was prepared according to procedure
A from a mixture of 30c (20.0 g, 142.0 mmol), t-BuOK (32.0 g,
285.0 mmol), 31a (45.0 g, 285.0 mmol), and DME (200 mL). The
product was obtained as an oil. Yield 10.4 g (28%).
(()-3-(5-Ethoxypyridin-3-yl)-3,9-diazabicyclo[4.2.1]nonane Fu-
maric Acid Salt (31). Compound 31 was prepared according to
procedure C from a mixture of 31b (1.00 g, 3.80 mmol) and diethyl
azodicarboxylate (2.00 g, 11.5 mmol). Compound 31 was isolated
by CC. Yield 0.19 g (20%). The corresponding salt was obtained
by the addition of a diethyl ether and MeOH mixture (9:1) saturated
with fumaric acid; mp 143-146 °C. Anal. (C14H21N3O‚1.5C4H4O4‚
0.5H2O) C, H, N.
(()-9-Benzyl-3,9-diazabicyclo[4.2.1]nonane-4-one (32b). Com-
pound 32b was prepared from benzyltropinone (32a) (ABCD
GmbH) (15.0 g, 69.7 mmol) added in chloroform (CHCl3) (150
mL) at -5 °C. Concentrated H2SO4 (35 mL) was added dropwise
to keep the temperature below 5 °C. NaN3 (9.06 g, 139.0 mmol)
was added, and the mixture was stirred at 20 °C overnight, followed
by stirring for 2 h at 50 °C. The mixture was cooled to room
temperature, and a mixture of H2O/ice (120 mL) was slowly added.
The mixture was neutralized with solid NaOH and stirred overnight
at 25 °C. Aqueous NaOH (40 mL, 4 M) was added, and the product
was extracted with CH2Cl2 (3 × 80 mL). Yield 12.9 g (81%); mp
110-111.5 °C.
5-[1,4]Diazepan-1-yl-2-iodo-pyridin-3-ol Hydrochloric Acid
Salt (12). Compound 12 was prepared from a mixture of 12e (0.87
g, 1.80 mmol) and hydrochloric acid (10 mL, 4 M) stirred in MeOH
(6 mL) for 30 min at 20 °C. The reaction mixture was evaporated.
The crystalline product was triturated with diethyl ether and filtered.
Yield 0.65 g (40%); mp 200 °C. Anal. (C10H14N3OI‚2HCl‚0.3H2O)
C, H, N.
4-(5-Cyclopropylpyridin-3-yl)-[1,4]diazepane-1-carboxylic Acid
tert-Butyl Ester (13a). Compound 13a was prepared from a mixture
of 11b (20.0 g, 56.0 mmol), tributylcyclopropylstannane (46.4 g,
140.0 mmol), PdCl2(PPh3)2 (1.90 g, 2.80 mmol), and DMF (20 mL)
stirred at 120 °C for 24 h. Aqueous NaOH (20 mL, 1 M) was added
followed by extraction with diethyl ether (3 × 100 mL). CC
(petroleum ether/ EtOAc(1:4)) gave compound 13a. Yield 4.21 g
(9%); mp 91.3-93.8 °C.
1-(5-Cyclopropylpyridin-3-yl)-[1,4]diazepane Fumaric Acid
Salt (13). Compound 13 was prepared according procedure B from
a mixture of 13a (2.80 g, 8.80 mmol), TFA (10.0 g, 88.2 mmol),
and CH2Cl2 (30 mL). Yield 0.44 g (23%). The corresponding salt
(()-9-Benzyl-3,9-diazabicyclo[4.2.1]nonane (32c). Compound
32c was prepared by using LiAlH4 (4.08 g, 107.0 mmol) in diethyl
ether (250 mL) and 32b (11.8 g, 51.2 mmol) stirred overnight at
20 °C. H2O (15 mL) was slowly added under H2, and the mixture

3168 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
Audouze et al.
was extracted with diethyl ether (2 × 50 mL). The product was
obtained as an oil. Yield 10.1 g (91%).
procedure B from a mixture of 34a (0.23 g, 0.70 mmol), TFA (0.75
g, 6.60 mmol), and CH2Cl2 (10 mL). Yield 0.60 g (21%). The
corresponding salt was obtained by the addition of a diethyl ether
and MeOH mixture (9:1) saturated with fumaric acid; mp 191-
192 °C. Anal. (C14H21N3O‚2C4H4O4) C, H, N.
9-Benzyl-9-azabicyclo[3.3.1]nonan-3-one (35b). Compound
35b was prepared according to Mach29 from a mixture of acetone-
1,3-dicarboxylic acid (35a) (40.5 g, 249.0 mmol), aqueous glutaric
dialdehyde (25%, 140 mL), benzylamine (24 mL), and H2O (400
mL) stirred for 1 h at 60 °C. The product was obtained as an oil.
Yield 34.3 g (60%).
(()-9-Benzyl-3,9-diazabicyclo[4.2.1]nonane 3-carboxylic Acid
tert-Butyl Ester (32d) Procedure D: Compound 32d was prepared
from a mixture of 32c (5.00 g, 23.1 mmol), triethylamine (Et3N)
(4.68 g, 46.2 mmol), and di-tert-butyl-dicarbonate (Boc2O) (5.04
g, 23.1 mmol) stirred overnight in CH2Cl2 at 20 °C. The organic
phase was washed with aqueous NaOH (50 mL, 1 M), and CC
(CH2Cl2, MeOH, and concentrated ammonia (89:10:1)) gave the
protected compound as an oil. Yield 7.79 g (100%).
(()-3,9-Diaza-bicyclo[4.2.1]nonane-3-carboxylic Acid tert-
Butyl Ester (32e). Compound 32e was prepared using 32d (7.60
g, 24.0 mmol) in ethanol (EtOH) (100 mL) stirred with palladium
on carbon 5% (Pd/C) (0.80 g) for 28 h under a hydrogen
atmosphere. CC (CH2Cl2, MeOH, and concentrated ammonia (89:
10:1)) gave the title compound. Yield 4.51 g (83%); mp 66.7-
69.7 °C.
(()-cis-9-(5-Propen-1-yloxy-pyridin-3-yl)-3,9-diaza-bicyclo-
[4.2.1]nonane-3-carboxylic Acid tert-Butyl Ester (32f). Com-
pound 32f was prepared according to procedure A from a mixture
of 32e (1.00 g, 4.40 mmol), 30d (0.90 g, 5.30 mmol), t-BuOK (0.90
g, 8.80 mmol), and DME (20 mL) stirred at 20 °C for 1h. The
product was obtained as an oil. Yield 0.86 g (55%).
(()-cis-9-(5-Propen-1-yloxypyridin-3-yl)-3,9-diazabicyclo-
[4.2.1]nonane Fumaric Acid Salt (32). Compound 32 was prepared
according to procedure B from 32f (0.18 g, 0.50 mmol), TFA (6.46
g, 56.6 mmol), and CH2Cl2 (10 mL). Yield 0.13 g (69%). The
corresponding salt was obtained by the addition of a diethyl ether
and MeOH mixture (9:1) saturated with fumaric acid; mp 161-
163 °C. Anal. (C15H21N3O‚1.1C4H4O4) C, H, N.
3-Chloro-5-methoxypyridine (33a). Compound 33a was pre-
pared from a mixture of sodium (18.6 g, 810.0 mmol) and MeOH
(500 mL) followed by evaporation. Compound 2b (100 g, 675.0
mmol) and DMSO (500 mL) were added, and the mixture was
stirred 2 h at 70 °C. Aqueous NaOH (2.0 L, 1 M) was added, and
the mixture was extrated with diethyl ether (3 × 500 mL). The
product was isolated as an oil. Yield 90.1 g (77%).
(()-1-(9-Benzyl-3,9-diazabicyclo-[4.2.1]non-3-yl)-ethanone (33b).
Compound 33b was prepared from 32c (5.00 g, 23.1 mmol), acetic
anhydride (3.54 g, 34.7 mmol), CH2Cl2 (50 mL) stirred for 3 h at
20 °C. Aqueous NaOH (20 mL, 1 M) was added, and the product
was extracted with EtOAc (4 × 30 mL). The product was obtained
as an oil. Yield 5.90 g (100%).
(+)-10-Benzyl-3,10-diazabicyclo[4.3.1]decan-4-one (35c). Com-
pound 35c was prepared from a mixture of 35b (15.0 g, 65.4 mmol)
and CHCl3 (150 mL) at -5 °C. Concentrated H2SO4 (33 mL) was
added dropwise to keep the temperature below 5 °C. NaN3 (8.50
g, 130.0 mmol) was added, and the mixture was stirred overnight
at 20 °C, followed by stirring for 2 h at 50 °C. The mixture was
cooled to room temperature, and a mixture of H2O/ice (100 mL)
was slowly added. The mixture was neutralized with solid NaOH
and stirred overnight at 25 °C. Aqueous NaOH (40 mL, 10 M)
was added, and the product was extracted with CH2Cl2 (2 × 100
mL). CC (CH2Cl2, MeOH and concentrated ammonia (89:10:1))
gave compound 35c. Yield 9.6 g (60%); mp 149-153 °C.
(+)-10-Benzyl-3,10-diazabicyclo[4.3.1]decane (35d). Com-
pound 35d was prepared from a mixture of 35c (9.50 g, 38.9 mmol),
LiAlH4 (3.09 g, 81.6 mmol), and diethyl ether (200 mL) stirred at
20 °C overnight. H2O (15 mL) was added under N2, and the solid
mixture was extracted with diethyl ether (2 × 50 mL). Aqueous
NaOH (100 mL, 4 M) was added, and the mixture was extracted
with diethyl ether (100 mL). The product was obtained as an oil.
Yield 6.47 g (72%).
(+)-10-Benzyl-3,10-diazabicyclo[4.3.1]decane-3-carboxylic Acid
tert-Butyl Ester (35e). Compound 35e was prepared according to
procedure D from a mixture of 35d (5.45 g, 23.7 mmol), Et3N (2.39
g, 23.7 mmol), Boc2O (5.17 g, 23.7 mmol), and CH2Cl2 (80 mL).
The product was isolated as an oil. Yield 8.06 g (100%).
(()-3,10-Diazabicyclo[4.3.1]decane-3-carboxylic Acid tert-
Butyl Ester (35f). Compound 35f was prepared from a mixture of
35e (8.00 g, 24.2 mmol) and palladium on carbon 10% (1.00 g)
stirred in ethanol (100 mL) under a hydrogen atmosphere. Yield
5.20 g (90%); mp 97.8-101.9 °C.
(()-10-(5-Chloropyridin-3-yl)-3,10-diazabicyclo[4.3.1]decane-
3-carboxylic Acid tert-Butyl Ester (35g). Compound 35g was
prepared according to procedure E from a mixture of 35f (1.50 g,
6.20 mmol), t-BuOK (1.40 g, 12.5 mmol), 3,5-dichloropyridine
(1.85 g, 12.5 mmol), and DME (30 mL). The product was obtained
as an oil. Yield 0.21 g (10%).
(()-1-(3,9-Diazabicyclo-[4.2.1]non-3-yl)-ethanone (33c). Com-
pound 33c was prepared from a mixture of 33b (5.90 g, 22.8 mmol),
Pd/C 5% (600 mg), and EtOH (100 mL) stirred under an atmosphere
of hydrogen for 24 h. The mixture was filtered and evaporated.
The product was isolated as an oil.Yield 2.73 g (71%).
(()-10-(5-Chloro-pyridin-3-yl)-3,10-diaza-bicyclo[4.3.1]decane
Fumaric Acid Salt (35). Compound 35 was prepared according
to procedure B from a mixture of 35f (0.20 g, 0.60 mmol), TFA
(0.70 g, 6.00 mmol) and CH2Cl2 (10 mL). Yield 60 mg (29%).
The corresponding salt was obtained by the addition of a diethyl
ether and MeOH mixture (9:1) saturated with fumaric acid; mp
181.4 °C. Anal. (C13H18ClN3‚1.2C4H4O4) C, H, N.
(()-3-Hydroxypiperidine-1-carboxylic Acid tert-Butyl Ester
(36b). Compound 36b was prepared from a mixture of (()-3-
hydroxypiperidine (36a) (10.0 g, 98.9 mmol), aqueous sodium
hydrogen carbonate (15 mL, 2 M), Et3N (9.90 g, 98.5 mmol), and
Boc2O (21.6 g, 98.9 mmol) stirred overnight in CH2Cl2 (150 mL)
at 20 °C. Compound 36b was isolated by CC (CH2Cl2, MeOH,
and ammonia (89:10:1)). The product was obtained as an oil. Yield
21.5 g (100%).
(()-3-Chloro-5-(piperidin-3-yloxy)-pyridine-1-carboxylic Acid
tert-Butyl Ester (36c). Compound 36c was prepared from a mixture
of PPh3 (19.6 g, 74.8 mmol) and diethyl azodicarboxylate acid (13.0
g, 74.8 mmol) stirred in THF (250 mL) at 20 °C for 30 min.
5-Chloropyridin-3-ol (9.70 g, 74.8 mmol) and 36b (10.0 g 49.8
mmol) were added, and the mixture was stirred overnight at 60
°C. Aqueous NaOH (100 mL, 1 M) was added, and the mixture
extracted with diethyl ether (2 × 100 mL). Compound 36c was
(()-9-(5-Methoxypyridin-3-yl)-3,9-diazabicyclo[4.2.1]non-3-
yl-ethanone (33d). Compound 33d was prepared according to
procedure A from a mixture of 33a (0.80 g, 5.90 mmol), 33c (1.00
g, 5.90 mmol), t-BuOK (1.30 g, 11.9 mmol), and DME (20 mL).
The product was obtained as an oil. Yield 0.45 g (27%).
(()-9-(5-Methoxypyridin-3-yl)-3,9-diazabicyclo[4.2.1]nonane
Fumaric Acid Salt (33). Compound 33 was prepared from a
mixture of 33a (0.40 g, 1.50 mmol) and concentrated hydrochloric
acid (25 mL, 25%) stirred at reflux for 72 h. The mixture was
evaporated. Aqueous NaOH (30 mL, 1 M) was added followed by
extraction with EtOAc (3 × 20 mL). CC (CH2Cl2, MeOH, and
concentrated ammonia (89:10:1)) gave the title compound 33. Yield
60 mg (17%). The corresponding salt was obtained by the addition
of a diethyl ether and MeOH mixture (9:1) saturated with fumaric
acid; mp 183-185 °C. Anal. (C13H19N3O‚1.2C4H4O4) C, H, N.
(()-9-(5-Ethoxypyridin-3-yl)-3,9-diazabicyclo[4.2.1]nonane-
3-carboxylic Acid tert-Butyl Ester (34a). Compound 34a was
prepared according to procedure E from a mixture of 32e (1.10 g,
4.90 mmol), t-BuOK (1.10 g, 8.70 mmol), 31a (0.70 g, 4.90 mmol),
and DME (20 mL). The product was obtained as an oil. Yield 0.24
g (14%).
(()-9-(5-Ethoxypyridin-3-yl)-3,9-diazabicyclo[4.2.1]nonane Fu-
maric Acid Salt (34). Compound 34 was prepared according to

R4â2 Subtype of Nicotinic Acetylcholine Receptors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3169
isolated by CC (CH2Cl2/ ethanol (9:1)). The product was isolated
as an oil. Yield 4.61 g (20%).
g, 0.70 mmol), and acetonitrile (7.0 mL). The product was isolated
as an oil. Yield 0.11 g (5%).
1-(6-Bromopyridin-3-yl)-[1,5]diazocane Fumaric Acid Salt
(41). Compound 41 was prepared according to procedure B from
a mixture of 41b (0.10 g, 0.30 mmol), TFA (0.68 g, 6.00 mmol),
and CH2Cl2 (5.0 mL). Yield 0.66 mg (82%). The corresponding
salt was obtained by the addition of a diethyl ether and MeOH
mixture (9:1) saturated with fumaric acid; mp 176-177 °C. Anal.
(C11H16N3Br‚1.5C4H4O4) C, H, N.
(()-3-Chloro-5-(piperidin-3-yloxy)-pyridine Fumaric Acid
Salt (36). Compound 36 was prepared according to procedure B
from a mixture of 36c (4.50 g, 14.4 mmol), TFA (16.2 g, 143.0
mmol), and CH2Cl2 (100 mL). Yield 2.70 g (50%). The corre-
sponding salt was obtained by the addition of a diethyl ether and
MeOH mixture (9:1) saturated with fumaric acid; mp 170.5-172.7
°C. Anal. (C10H13ClN2O‚C4H4O4) C, H, N.
1-(5-Chloropyridin-3-yl)-[1,5]diazocane Fumaric Acid Salt
(42). Compound 42 was prepared from a mixture of 2b (2.70 g,
18.1 mmol) and 39b (2.00 g, 18.1 mmol) stirred overnight at 150
°C. Aqueous NaOH (50 mL, 1 M) was added, and the mixture
was extracted with EtOAc (2 × 30 mL). Yield 0.10 g (2.5%). The
corresponding salt was obtained by the addition of a diethyl ether
and MeOH mixture (9:1) saturated with fumaric acid; mp 174-
176 °C. Anal. (C11H16ClN3‚1.6C4H4O4) C, H, N.
5-(5-Vinyloxypyridin-3-yl)-[1,5]diazocane-1-carboxylic Acid
tert-Butyl Ester (43a). Compound 43a was prepared according to
procedure E from a mixture of 2d (1.42 g, 9.10 mmol), 38b (1.94
g, 9.10 mmol), t-BuOK (2.00 g, 18.2 mmol), and DME (40 mL).
The product was obtained as an oil. Yield 0.14 g (4.6%).
5-(6-Bromo-5-vinyloxypyridin-3-yl)-[1,5]diazocane-1-carboxy-
lic Acid tert-Butyl Ester (43b). Compound 43b was prepared
according to procedure F from a mixture of 43a (0.14 g, 0.40
mmol), NBS (75.0 mg, 0.40 mmol), and acetonitrile (10 mL). Yield
86.0 mg (52%); mp 135.4-138 °C.
1-(6-Bromo-5-vinyloxypyridin-3-yl)-[1,5]diazocane Fumaric
Acid Salt (43). Compound 43 was prepared according to procedure
B from a mixture of 43b (0.83 g, 0.20 mmol), TFA (0.23 g, 2.00
mmol), and CH2Cl2 (10 mL). Yield 30 mg (56%). The correspond-
ing salt was obtained by the addition of a diethyl ether and MeOH
mixture (9:1) saturated with fumaric acid; mp 155.7-158.9 °C.
Anal. (C13H18N3OBr‚C4H4O4‚H2O) C, H, N.
1-(5-Bromopyridin-3-yl)-[1,5]diazocane Fumaric Acid Salt
(48). Compound 48 was prepared from a mixture of 39b (2.07 g,
18.1 mmol) and 4a (4.28 g, 18.1 mmol) stirred for 48 h at 150 °C
in the absence of solvent. The mixture was allowed to reach room
temperature. Aqueous NaOH (100 mL, 1 M) was added, followed
by extraction with EtOAc (2 × 30 mL). Compound 48 was isolated
by CC (CH2Cl2, MeOH, and concentrated ammonia (89:10:1)).
Yield 0.31 g (63%). The corresponding salt was obtained by the
addition of diethyl ether and MeOH mixture (9:1) saturated with
fumaric acid; mp 151-153 °C. Anal. (C11H16N3Br‚0.7C4H4O4‚
0.8H2O) C, H, N.
[1,5]Diazocane-1-carboxylic Acid tert-Butyl Ester (38b). Com-
pound 38b was prepared from a mixture of NaOMe, prepared from
Na (833 mg, 36.2 mmol), MeOH (30 mL), and [1,5]diazocane
dihydrobromide (38a) prepared according to Hernandez-Mora17
(5.00 g, 18.1 mmol), and stirred for 30 min at room temperature.
Et3N (3.66 g, 36.2 mmol), Boc2O (3.16 g, 14.5 mmol), and CH2-
Cl2 (150 mL) were added. The mixture was stirred overnight at 20
°C, and aqueous NaOH (150 mL, 1 M) was added. Yield 2.02 g
(10%); mp 92-96 °C.
5-(5-Methoxypyridin-3-yl)-[1,5]diazocane-1-carboxylic Acid
tert-Butyl Ester (38c). Compound 38c was prepared according to
procedure A from a mixture of 1a (1.70 g, 9.30 mmol), 38b (2.00
g, 9.30 mmol), t-BuOK (2.10 g, 18.6 mmol), and tetrakis-
(triphenylphosphine)palladium(0) (0.50 g) stirred in DME (50 mL)
at reflux for 3 h. The product was isolated as an oil. Yield 0.3 g
(10%).
1-(5-Methoxypyridin-3-yl)-[1,5]diazocane Fumaric Acid Salt
(38). Compound 38 was prepared according to procedure B from
a mixture of 38c (0.30 g, 0.90 mmol), TFA (0.50 g, 4.70 mmol),
and CH2Cl2 (10 mL). Yield 0.11 g (55%). The corresponding salt
was obtained by the addition of a diethyl ether and MeOH mixture
(9:1) saturated with fumaric acid; mp 158-160 °C. Anal. (C12H19N3O
× 1.5 C4H4O4) C, H, N.
3-n-Butoxy-5-chloropyridine (39a). Compound 39a was pre-
pared from a mixture of sodium (3.89 g, 168.0 mmol) and n-butanol
(200 mL) stirred at 80 °C for 3 h. DMSO (100 mL) and 2b (20.0
g, 135.0 mmol) were added, and the mixture was stirred 10 h at 70
°C. Aqueous NaOH (400 mL, 1 M) was added, and the mixture
was extracted with diethyl ether (200 mL). The product was
obtained as an oil. Yield 28 g (100%).
[1,5]Diazocane (39b). Compound 39b was prepared from sodium
(3.67 g, 159 mmol) dissolved in MeOH (80 mL). Compound 38a
(20.0 g, 72.5 mmol) was added, and the mixture was stirred for 10
min at reflux and evaporated. The product was obtained as an oil.
Yield 9.8 g (42%).
5-[(5-(1-Butoxy)-pyridin-3-yl]-[1,5]diazocane Fumaric Acid
Salt (39). Compound 39 was prepared according to procedure E
from a mixture of 39a (1.50 g, 5.40 mmol), 39b (1.50 g, 5.40
mmol), t-BuOK (1.20 g, 10.9 mmol), and DME (30 mL). Yield 90
mg (63%). The corresponding salt was obtained by addition of a
diethyl ether and MeOH mixture (9:1) saturated with fumaric acid;
mp 146-148 °C. Anal. (C15H25N3O × 1.5C4H4O4) C, H, N.
1-Pyridin-3-yl-[1,5]diazocane Fumaric Acid Salt (40). Com-
pound 40 was prepared from a mixture of 3-fluoropyridine (40a)
(3.00 g, 30.9 mmol), 39b (9.80 g, 30.9 mmol), and 1,4 dioxane
(3.0 mL) heated to 160 °C in a sealed stainless steel vessel for 18
h. Aqueous NaOH (50 mL, 1 M) was added, and the mixture
extracted with EtOAc (3 × 40 mL). Compound 40 was isolated by
CC (CH2Cl2, MeOH, and concentrated ammonia (89:10:1)). Yield
70 mg (1.5%). The corresponding salt was obtained by the addition
of a diethyl ether and MeOH mixture (9:1) saturated with fumaric
acid; mp 137.7-141.1 °C. Anal. (C11H17N3‚2C4H4O4‚0.3H2O) C,
H, N.
4-(Pyridin-3-yl)-[1,5]diazocane Acid tert-Butyl Ester (41a).
Compound 41a was prepared according to procedure D from a
mixture of 40 (0.25 g, 1.30 mmol), Et3N (0.26 g, 2.60 mmol), Boc2O
(0.34 g, 1.60 mmol), and CH2Cl2 (10 mL). The product was
obtained as an oil. Yield 0.60 g (11%).
5-(6-Bromopyridin-3-yl)-[1,5]diazocane-1-carboxylic Acid tert-
Butyl Ester (41b). Compound 41b was prepared according to
procedure F from a mixture of 41a (2.04 g, 0.70 mmol), NBS (1.37
1-(5-Vinylpyridin-3-yl)-[1,4]diazepane-1-carboxylic Acid tert-
Butyl Ester (49a). Compound 49a was prepared from a mixture
of 1-(5-vinyloxy-pyridin-3-yl)-[1,4]diazepane fumarate (15) (1.00
g, 3.00 mmol), prepared according to Nielsen et al.,15 aqueous
sodium hydrogen carbonate (10 mL, 1 M), and Boc2O (0.72 g, 3.30
mmol) stirred in CH2Cl2 (6.0 mL) at room temperature for 1 h.
Compound 49a was isolated by CC (CH2Cl2, MeOH, and concen-
trated ammonia (89:10:1)). The product was isolated as an oil.Yield
1.0 g (100%).
1-(6-Bromo-5-vinylpyridin-3-yl)-[1,4]diazepane-1-carboxyl-
ic Acid tert-Butyl Ester (49b). Compound 49b was prepared
according to procedure F from a mixture of 49a (0.63 g, 2.00
mmol), NBS (0.42 g, 2.40 mmol), and acetonitrile (20 mL). Yield
0.64 g (80%); mp 115.2-116.8 °C.
1-(6-Bromo-5-vinylpyridin-3-yl)-[1,4]diazepane Fumaric Acid
Salt (49). Compound 49 was prepared according to procedure B
from a mixture of 49b (0.64 g, 1.60 mmol), TFA (1.80 g, 16.0
mmol), and CH2Cl2 (25 mL). Yield 0.24 g (50%). The correspond-
ing salt was obtained by the addition of a diethyl ether and MeOH
mixture (9:1) saturated with fumaric acid; mp 128.6 °C. Anal.
(C12H16N3BrO‚1.2C4H4O4) C, H, N.
4-(Pyridin-3-yl)-piperazine-1-carboxylic Acid tert-Butyl Ester
(50b). Compound 50b was prepared from a mixture of 1-(pyridin-
3-yl)-piperazine (50a) (2.50 g, 8.90 mmol), aqueous sodium
hydrogen carbonate (25 mL, 1 M), Boc2O (1.90 g, 8.90 mmol),
and CH2Cl2 (25 mL) stirred overnight at room-temperature. The

3170 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11
Audouze et al.
organic phase was washed with NaOH (2 × 50 mL, 1 M). The
product was isolated as an oil.Yield 1.97 g (84%).
reduced models were then used to predict the affinity of the
molecules that were left out. The final model was used to predict
the affinities of an external validation set of seven compounds.
4-(6-Bromopyridin-3-yl)-piperazine-1-carboxylic Acid tert-
Butyl Ester (50c). Compound 50c was prepared according to
procedure F from a mixture of 50b (1.90 g, 7.20 mmol), NBS (1.30
g, 7.20 mmol), and acetonitrile (20 mL). Yield 2.0 g (81%); mp
122.6-123.5 °C.
1-(6-Bromopyridin-3-yl)-piperazine Fumaric Acid Salt (50).
Compound 50 was prepared according to procedure B from a
mixture of 50c (1.90 g, 5.60 mmol), TFA (0.63 g, 5.60 mmol),
and CH2Cl2 (20 mL). Yield 0.20 g (15%). The corresponding salt
was obtained by the addition of a diethyl ether and MeOH mixture
(9:1) saturated with fumaric acid; mp 170.8-171.6 °C. Anal.
(C9H12N3Br‚0.8C4H4O4‚0.5H2O) C, H, N.
Acknowledgment. We are grateful to Jørgen Bach, Tine
Sparre, and Karina Borup from the Department of Medicinal
Chemistry and the technicians from the Department of Receptor
Biochemistry for excellent technical assistance.
Supporting Information Available: Experimental details,
spectroscopic data, and elemental analysis results. This material is
available free of charge via the Internet at http://pubs.acs.org.
In Vitro Inhibition of [3H]-Cytisine Binding. Tissue Prepara-
tion. Preparations were performed at 0-4 °C unless otherwise
indicated. Cerebral corticies from male Wistar rats20 (150-250 g)
were homogenized for 20 s in 15 mL of Tris HCl (50 mM, pH
7.4) containing 120 mM NaCl, 5 mM KCl, 1 mM MgCl2, and 2.5
mM CaCl2 using an Ultra-Turrax homogenizer. The homogenate
was centrifuged at 27 000g for 10 min. The supernatant was
discarded, and the pellet was resuspended in fresh buffer and
centrifuged a second time. The final pellet was resuspended in fresh
buffer (35 mL/g of original tissue) and used for binding assays.
Assay. Aliquots of 500 µL of homogenate were added to 25 µL of
drug solution and 25 µL of 3H-cytisine (1 nM, final concentration),
mixed, and incubated for 90 min at 2 °C. Nonspecific binding was
determined using (-)-nicotine (100 µM, final concentration). After
incubation, the samples were added to 5 mL of ice-cold buffer and
poured directly onto Whatman GF/C glass fiber filters under suction
and immediately washed with 2 × 5 mL of ice-cold buffer. The
amount of radioactivity on the filters was determined by conven-
tional liquid scintillation counting using a Trib-carb liquid scintil-
lation analyzer with a counting efficiency of 58%. Specific binding
was equal to total binding minus nonspecific binding.
Conformational Analysis. Monte Carlo sampling and the
MMFF force field as implemented in Macromodel v. 9.021 were
used for conformational analysis. In all cases, the calculations were
performed on the N-protonated species in aqueous solution. The
low energy conformer (<1 kcal/mol) that gave the best fit to the
template molecule (55) was used to calculate the 3D-QSAR model.
Molecular Alignment. The fitting points used were the pyridine
nitrogen atom, the centroid of the pyridine ring, the protonated
nitrogen atom, and the two hydrogens attached to this atom.
GRID Calculations. The interaction energies were calculated
using GRID version 22A.22 The grid size (X ) 23 Å, Y ) 23 Å, Z
) 22 Å) was automatically determined to exceed the inclusive grid
box by 5 Å. Energies were calculated with a grid spacing of 1 Å
for water (OH2) and methyl (C3) probes.
References
(1) Karlin, A.; Akabas, M. H. Toward a structural basis for the function
of nicotinic acetylcholine receptors and their cousins. Neuron 1995,
15, 1231-1244.
(2) Changeux, J. P.; Edelstein, S. J. Allosteric mechanisms in normal
and pathological nicotinic acetylcholine receptors. Curr. Opin.
Neurobiol. 2001, 11, 369-377.
(3) McGehee, D. S.; Heath, M. J. S.; Gelber, S. M.; Devay, P.; Role, L.
W. Nicotine enhancement of fast excitatory synaptic transmission
in CNS by presynaptic receptors. Science 1995, 269, 1692-1696.
(4) Role, L. W.; Berg, D. K. Nicotinic receptors in the development and
modulation of CNS synapses. Neuron 1996, 16, 1077-1085.
(5) Wonnacott, S. Presynaptic nicotinic ACh receptors. TINS 1996, 20,
92-98.
(6) Holladay, M. W.; Dart, M. J.; Lynch, J. K. Neuronal nicotinic
acetylcholine receptors as targets for drug discovery. J. Med. Chem.
1997, 40, 4169-4194.
(7) Lindstrom, J. Neuronal nicotinic acetylcholine receptors. In Ion
Channels 4; Narahashin, T., ed; Plenum Press: New York, 1996;
pp 377-437, .
(8) Gotti, C.; Fornasari, D.; Clementi, F. Human neuronal nicotinic
receptors. Progress in Neurobiology 1997, 53, 199-237.
(9) Paterson, D.; Nordberg, A. Neuronal nicotinic receptors in the human
brain. Progress in Neurobiology 2000, 61, 75-111.
(10) Alkondon, M.; Albuquerque, E. X. Initial characterization of the
nicotinic acetylcholine receptors in rat hippocampal neurons. J.
Recept. Res. 1991, 11, 1001-1022.
(11) Alkondon, M.; Albuquerque, E. X. Diversity of nicotinic acetylcholine
receptors in rat hippocampal neurons: I: Pharmacological and
functional evidence for distinct structural subtypes. J. Pharmacol.
Exp. Ther. 1993, 265, 1455-1473.
(12) Lukas, R. J.; Changeux, J. P.; Le Nove`re, N.; Albuquerque, E. X.;
Balfour, D. J. K.; Berg, D. K.; Bertrand, D.; Chiappinelli, V. A.;
Clarke, P. B. S.; Collins, A. C.; Dani, J. A.; Grady, S. R.; Kellar, K.
J.; Lindstrom, J. M.; Marks, M. J.; Quik, M.; Taylor, P. W.;
Wonnacott, S. International union of pharmacology. XX. Current
status of the nomenclature for nicotinic acetylcholine receptors and
their subunits. Pharmacol. ReV. 1999, 51, 397-401.
(13) Whiting, P. J.; Lindstrom, J. M. Characterization of a bovine and
human neuronal nicotinic receptors using monoclonal antibodies. J.
Neurosci. 1988, 8, 2295-3404.
(14) Anand, R.; Conroy, W. G.; Schoepfer, R.; Whiting, P. J.; Lindstrom,
J. Chicken neuronal nicotinic receptors expressed in Xenopus oocytes
have a pentameric quaternary structure J. Biol. Chem. 1991, 266,
11192-11198.
(15) Nielsen, S. F.; Nielsen, E. O.; Olsen, G. M.; Liljefors, T.; Peters, D.
Novel potent ligands for the central nicotinic acetylcholine receptor:
Synthesis, receptor binding, and 3D-QSAR analysis. J. Med. Chem.
2000, 43, 2217-2226.
GOLPE Analysis. The partial least-squares (PLS) models were
calculated using GOLPE v. 4.5.23,24 The dependent variables were
imported as pIC50 values.
Variable Pretreatments. GOLPE automatically rejects X-
variables having a total sum of square less than 10-7. Further
pretreatment was performed as follows. Absolute critical X-values
below 0.05 were deleted. Minimum sd cutoff was set to 0.10 for
all probes. Second, third, and fourth level X-variables were deleted.
The initial PLS-model was build using these data.
Smart Region Definition (SRD).25 The groups of variables were
generated using the default number of seeds. The variables within
a distance of 1 Å to the seeds were grouped. Neighboring groups
within a distance of 2 Å containing the same information were
collapsed.
Variable Selection FFD. The effect of the individual groups of
variables were evaluated using the FFD procedure including 20%
dummy variables. Only groups having a positive influence on the
model larger than the average dummy variable were included in
the final model.
(16) Le Novere, N.; Grutter, T.; Changeux, J.-P. Models of the extracellular
domain of the nicotinic receptors and of agonist and Ca2+ binding
sites. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 3210-3215.
(17) Hernandez-Mora, J. Derivatives of 1,5-Diazacyclooctane. Ph.D.
Dissertation, University of Michigan, Ann Arbor, MI, 1959.
(18) Michaels, R. J.; Zaugg, H. E. Synthesis of 9-methyl-3,9-diazabicyclo-
[4.2.1]-nonane. J. Org. Chem. 1960, 25, 637.
(19) Flores, C. M.; Rogers, S. W.; Pabreza, L. A.; Wolfe, B. B.; Kellar,
K. J. A subtype of nicotinic cholinergic receptor in rat brain is
composed of R4â2 subunits and is up-regulated by chronic nicotine
treatment. Mol. Pharmacol. 1992, 41, 31-37.
Cross-Validation. Twenty reduced models were built and
evaluated. Each time, 20% of the molecules were randomly chosen
and left out of the model (leave five random groups out). The
(20) Pabreza, L. A.; Dhawan, S.; Kellar, K. J. [3H]-Cytisine binding to
nicotinic cholinergic receptors in brain. Mol. Pharmacol. 1991, 39,
9-12.

R4â2 Subtype of Nicotinic Acetylcholine Receptors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 11 3171
(21) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton,
M.; Caulfield, C.; Chang, G.; Hendrickson, T.; Still, W. C. Macro-
Model - An integrated software system for modeling organic
and bioorganic molecules using molecular mechanics. J. Comput.
Chem. 1990, 11, 440-467. The program is available from Schro¨d-
inger, 120 West 45th Street, 32nd Floor, New York, NY 10036-
4041
(27) Hansen, S. B.; Sulzenbacher, G.; Huxford, T.; Marchot, P.; Taylor,
P.; Bourne, Y. Structures of Aplysia AChBP complexes with nicotinic
agonists and antagonists reveal distinctive binding interfaces and
conformations. EMBO J. 2005, 24, 3635-3646.
(28) Abreo, M. A.; Lin, N.-H.; Garvey, D. S.; Gunn, D. E.; Hettinger,
A.-M.; Wasicak, J. T.; Pavlik, P. A.; Martin, Y. C.; Donnelly-Roberts,
D. L.; Anderson, D. J.; Sullivan, J. P.; Williams, M.; Arneric, S. P.;
Holladay, M. W. Novel 3-pyridyl ethers with subnanomolar affinity
for central neuronal nicotinic acetylcholine receptors. J. Med. Chem.
1996, 39, 817-825.
(29) Mach, R. H.; Luedtke, R. R.; Unsworth, C. D.; Boundy, V. A.;
Nowak, P. A.; Scripko, J. G.; Elder, S. T.; Jackson, J. R.; Hoffman,
P. L.; Evora, P. H.; Rao, A. V.; Molinoff, P. B.; Childers, S. R.;
Ehrenkaufer, R. L. 18F-labeled benzamides for studying the dopamine
D2 receptor with positron emission tomography. J. Med. Chem. 1993,
36, 3707-3720.
(30) Gohlke, H.; Schwarz, S.; Gu¨ndisch, D.; Tilotta, M. C.; Weber, A.;
Wegge, T.; Seitz, G. 3D QSAR analyses-guided rational design of
novel ligands for the (R4)2(â2)3 nicotinic acetylcholine receptor. J.
Med. Chem. 2003, 46, 2031-2048.
(22) GRID, version 22A; Molecular Discovery Ltd.: Oxford, England,
2004.
(23) Baroni, M.; Costantino, G.; Cruciani, G.; Rigagnelli, D.; Valigi, R.;
Clementi, S. Generating optimal linear PLS estimations (GOLPE):
An advanced chemometric tool for handling 3D-QSAR problems.
Quant. Struct.-Act. Relat. 1993, 12, 9-20.
(24) Clementi, S. GOLPE 4.5.12; Multivariate Infometric Analysis
(MIA): Viale del Castagni 16 Perugia, Italy, 2002.
(25) Pastor, M.; Cruciani, G.; Clementi, S. Smart region definition: A
new way to improve the predictive ability and interpretability of three-
dimensional quantitative structure-activity relationships. J. Med.
Chem. 1997, 40, 1455-1464.
(26) Celie, P. H. N.; van Rossum-Fikkert, S. E.; van Dijk, W. J.; Brejc,
K.; Smit, A. B.; Sixma, T. K. Nicotine and Carbamylcholine binding
to nicotinic acetylcholine receptors studied in AChBP crystal
ctructures. Neuron 2004, 41, 907-914.
JM058058H