Parallel synthesis of a series of potentially brain penetrant aminoalkyl benzoimidazoles

Article abstract of DOI:10.1016/j.bmc.2007.11.068

Alpha7 agonists were identified via GOLD (CCDC) docking in the putative agonist binding site of an alpha7 homology model and a series of aminoalkyl benzoimidazoles was synthesised to obtain potentially brain penetrant drugs. The array was prepared startin

Full text of DOI:10.1016/j.bmc.2007.11.068

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 2313–2328
Parallel synthesis of a series of potentially brain
penetrant aminoalkyl benzoimidazoles
Iolanda Micco,*,ꢀ Arianna Nencini,*,ꢀ Joanna Quinn,*,ꢀ Hendrick Bothmann,
Chiara Ghiron, Alessandro Padova and Silvia Papini
Siena Biotech S.p.A., Therapeutic Research, Via Fiorentina 1, Siena 53100, Italy
Received 7 September 2007; revised 16 November 2007; accepted 23 November 2007
Available online 19 December 2007
Abstract—Alpha7 agonists were identified via GOLD (CCDC) docking in the putative agonist binding site of an alpha7 homology
model and a series of aminoalkyl benzoimidazoles was synthesised to obtain potentially brain penetrant drugs. The array was pre-
pared starting from the reaction of ortho-fluoronitrobenzenes with a selection of diamines, followed by reduction of the nitro group
to obtain a series of monoalkylated phenylene diamines. N,N0-Carbonyldiimidazole (CDI) mediated acylation, followed by a par-
allel automated work-up procedure, afforded the monoacylated phenylenediamines which were cyclised under acidic conditions.
Parallel work-up and purification afforded the array products in good yields and purities with a robust parallel methodology which
will be useful for other libraries. Screening for alpha7 activity revealed compounds with agonist activity for the receptor.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
tion of novel compounds. One of the key strategies to
address this demand has led to the development and
use of parallel and automated procedures to not only
speed up and simplify the synthesis of potential drugs,
but also to reduce company costs. Virtual screening
technologies, such as docking and pharmacophore
screening, also provide an expeditious route to the iden-
tification of molecules ready to enter hit to lead optimi-
sation. Herein we describe how docking experiments of
commercially available compounds were carried out
utilising GOLD software5 on an alpha7 homology mod-
el. After the initial virtual screening, visual inspection of
the results allowed the manual selection of compounds
for purchasing. These compounds were assayed in an al-
pha7 FLIPR assay. The assay displayed three com-
pounds as the most promising (Table 1). Even if
compounds 1 and 2 were more active (0.6 and 0.2 lM,
respectively), compound 3 was selected as the starting
point for Medicinal Chemistry exploration. The choice
was dictated by analysis of the literature around the
structures of the three compounds. The quinuclidine
core of compound 1 and tropane of compound 2 are al-
ready widely described for alpha7 nAChR agonists.6–8
In addition, the same scaffolds often play a key role in
the binding to other receptors such as 5-HT3, muscarinic
subtypes and Serotonin Reuptake transporters.8–10 An
initial library was therefore enumerated around com-
pound 3. The virtual set of compounds so obtained were
analysed for drug-like properties using the Accord SDK
An extensive number of studies have demonstrated the
important role of nicotinic acetylcholine receptors within
the CNS and their role in processes such as memory, cog-
nition, sensory gating and anxiety.1 Evidence of the asso-
ciation between cholinergic deficit and cognitive
impairment in patients with Alzheimer’s disease (AD)
has led to much research into possible therapies designed
to replace loss of function of these receptors.2 Neuronal
nicotinic acetylcholine receptors (nAChRs) are penta-
meric ligand-gated ion channels consisting of alpha (2–
10) and beta (2–4) subunits in various combinations. Spe-
cifically, the alpha7 and alpha4beta2 subtypes are found
in abundance in the CNS, suggesting their importance
in a variety of biological processes.3 In the present report,
we focus on the challenge to find novel alpha7 nicotinic
receptor agonists showing affinity for these brain
nAChRs.4
A reoccurring issue for medicinal chemists within the
pharmaceutical industry is the need for fast identifica-
Keywords: Alpha7; Alzheimer’s disease; Benzoimidazole.
*
Corresponding authors. Tel.: +39 0577 381441; fax: +39 0577
381410; e-mail addresses: imicco@sienabiotech.it; anencini@siena-
biotech.it; jquinn@sienabiotech.it
ꢀ
These authors contributed equally to the work.
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.11.068

2314
I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
Table 1. Virtual hits tested in concentration-response analysis in Ca-flux assay
Compound
Structure
MW
264
AlogP
FPSA
42
Alpha7 nAChR EC50 (lM)
O
H
F
N
N
1
2
0.6
O
Br
N
2
338
2.1
46
0.2
NH
O
N
H
O
N
3
309
3.8
29
2.6
N
N
from Accelrys.11 The criteria for substituent selection
were based on modified Lipinski, Veber and Norinder
rules (Table 2)12 and evaluated using Spotfire visualisa-
tion.13 An array of molecules, which fulfilled the selec-
tion criteria for about 80% of its components and
suitably represented the chemical space covered by the
virtual array, was chosen for synthesis. Synthetic valida-
tion was undertaken with consideration to library
expansion employing automation tools. It was funda-
mental that the synthetic procedures were optimised
with regard to the capabilities of the automation plat-
form to be used, for example, tube size, reaction vol-
umes, block format, etc., to prevent any problems
being encountered downstream.
Schapiro17 have recently developed homology models
of alpha7, based on the structure of the highly homol-
ogous acetyl choline binding protein (AChBP).18 For
our docking simulations, we used a previously pub-
lished homology structure of chicken alpha7, contain-
ing a docked nicotine molecule in the agonist binding
site located at the interface of two subunits.19 Nicotine
was then eliminated from the pdb file. The active site
was defined by residues Tyr91, Trp147, Thr148,
Tyr186, Cys188, Cys189 and Tyr193 from the alpha
subunit and Met36, Trp53, Asn105, Gln115 and
Leu117 from the beta subunit. Sdf files for commer-
cially available compounds were directly obtained
from Maybridge and Specs.20 Filters were applied in
order to identify hits with a physicochemical property
profile compatible with CNS penetration. In particu-
lar, all compounds complied with our internal criteria
(Table 2). The molecules were then minimised using
Corina21 and protonated at pH 7 using LigPrep.22
Commercial compounds’ databases were docked
utilising Gold software.5 A prerequisite for virtual
hit selection was the presence of a protonated nitrogen
contributing to binding via a hydrogen bond with the car-
bonyl side chain of Trp147 or via a cation-pi interaction
with the aromatic cage composed of Trp53, Trp147,
Tyr186 and Tyr193.23 Virtual hits were then selected man-
ually based on highest GoldScore fitness values, after vi-
sual inspection to ensure optimal geometries for
putative binding conformations.
Benzoimidazole derivatives are often exploited in drug
discovery programmes14 and the intention of this study
was also to find a robust process for the synthesis of this
chemotype, not only for the project in hand but also for
future compound library generation.
2. Methods
2.1. Docking calculations
The alpha7 nicotinic acetylcholine receptor (nAChR)
is composed of five homologous subunits positioned
symmetrically around an axis perpendicular to the cell
membrane. Each subunit consists of an extracellular
domain and a transmembrane domain.15 The agonist
binding site is found at the interface between two
adjacent subunits in the extracellular domain. The al-
pha7 X-ray crystal structure has not been determined
to date, but the research groups of Le Novere16 and
Selected compounds were then purchased from commer-
cially available vendors and screened in an alpha7
FLIPR assay. Given the nature of the FLIPR assay,
we identified compound showing both agonist and
antagonist properties. The overall hit-rate was

I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
Table 2. Physicochemical properties-criteria for compound selection
2315
Parameter
Selection range
O
N
N
A
B
C
MW
FPSA
ClogP
<400
<70
<4.4
0–2
<7
Hydrogen bond donors
Hydrogen bond acceptors
Carboxylic acids
N
D
0
Figure 2. Possible variation points of the hit compound.
around 3%. Data for selected agonists are reported in
Table 1.
tonated nitrogen in the binding to nicotinic receptors,
several amines with different basicity and steric hin-
drance were evaluated. An amidic moiety was also intro-
duced in order to confirm the necessity of a basic
nitrogen also in this class of molecules.
Compounds 1 and 2 are analogues of compounds re-
ported to be alpha7 agonists and therefore were deprior-
itised. To the best our knowledge, benzoimidazole 3 was
a novel agonist of alpha7 and was therefore selected for
further investigation.
2.2. Chemistry
Two synthetic routes to the benzoimidazole series were
explored. The first, depicted in Scheme 1, involved the
preparation of intermediates 6 following a reported
nucleophilic substitution procedure of ortho-fluoronitro-
benzenes 4a–c with the designated diamines 5a–h.24
In the putative binding conformation of benzoimidazole
3 (Fig. 1), the protonated nitrogen of the dimethylami-
nopropyl side chain makes a hydrogen bond with the
carbonyl of Trp147. The benzoimidazole ring makes
hydrophobic interactions with Trp53 and Tyr193. The
oxygen atom of the phenoxy methyl moiety is within
H-bond distance from the OH of Tyr193. The phenyl
ring experiences a stacking interaction with the same res-
idue. In light of these considerations four main variation
points of the hit were identified for scaffold exploration
(Fig. 2). The strategy essentially consisted in the modifi-
cation of four main portions of the hit molecule: differ-
ent substituents on the benzoimidazole ring (A), various
aryl moieties (B), two kinds of linker (C) and a few ami-
no groups (D). Considering the importance of the pro-
Optimisation of the work-up allowed the synthesis to be
carried out using a parallel approach without chromato-
graphic purification and afforded the desired intermedi-
ates in medium to high yield (Table 3).
Monoacylation of compounds 6a–j with in situ gener-
ated acid chlorides, followed by cyclisation-reduction
of the nitro-derivatives 7, was attempted in a one-pot
approach (Scheme 1). Unfortunately, complete amide
hydrolysis was observed, due to the strongly acidic con-
ditions resulting from the use of tin (II) chloride as a
reducing agent.
The second route (Scheme 2) employed the tin(II) chlo-
ride reduction of the nitro group on intermediates (6a–j)
to afford the corresponding anilines,25 resulting in high
purity products and quantitative yields (8a–j). At this
point, small trials could be performed to investigate
the cyclisation step further. In the first trial, direct acyl-
ation-cyclisation in 37% HCl was tested.26 A small set of
carboxylic acids was used to verify the ring closure con-
ditions’ reliability. 4-Bromophenylacetic acid gave com-
plete conversion to the corresponding benzoimidazole,
but the conditions did not appear to be suitable for
the other substrates (Table 4).
In a further attempt (Scheme 1) a 2-step approach was
employed, consisting of monoacylation of phenylenedi-
amine (affording 9 and/or 10) followed by cyclisation
under acidic conditions.27 Activation of the carboxylic
acids using oxalyl chloride was first tested, resulting in
an undesired mixture of mono- and diacylated products.
Further validation using milder activating agents was
therefore required to broaden the scope of the benzoim-
idazole synthesis to a variety of substitutions on the aro-
matic ring. Mild coupling agents were explored and it
was found that when HATU, PyBOP and CDI were
used, complete conversion to a mixture of the two pos-
Figure 1. Putative binding conformation of compound 3 with homol-
ogy structure of chicken alpha7. The protonated nitrogen of the
dimethylaminopropyl side chain makes a hydrogen bond with the
carbonyl of Trp147. The benzoimidazole ring makes hydrophobic
interactions with Trp53 and Tyr193. The oxygen atom of the phenoxy
methyl moiety is within H-bond distance from the OH of Tyr193. The
phenyl ring experiences a stacking interaction with the same residue.

2316
I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
NO2
NO2
R1
R1
(i)
+
H2N
R2
F
R2
N
5a-h
H
6a-j
4a R2=H
4b R2=CH3
4c R2=F
(ii)
Ar
X
NO2
R1
NH2
R1
(iii)
N
N
R2
N
R2
N
R1
7a-j
O
X
X
O
Ar
Ar
R2
Scheme 1. Reagents and conditions: (i) K2CO3, DMF, 70 ꢁC; (ii) (COCl)2, ArXCOOH, DCM, rt; (iii) SnCl2, EtOH, reflux.
Table 3. Structure and yields of the intermediates 6a–j
Table 3 (continued)
Compound Fluoronitro
benzenes
R1NH2
Yield
Compound Fluoronitro
benzenes
R1NH2
Yield
77%
NO2
N
NH2
NO2
N
NH2
6a
78%
6i
F
F
5a
5a
4a
4b
N
NH2
N
NH2
F
NO2
NO2
O
6j
65%
6b
92%
84%
F
F
5a
5b
5c
4c
4a
N
NH2
NO2
6c
X
Ar
F
NO2
NH2
R1
N
(ii)
(i)
4a
N
R1
R1
R2
N
H
R2
N
H
6a-j
8a-j
11-116
NO2
R2
N
NH2
6d
89%
(iii)
F
5d
4a
X
O
NH2
R1
Ar
(iv)
NH
N
NO2
NH2
R2
N
N
6e
74%
93%
R1
O
R2
N
H
X
F
9
10
5e
Ar
4a
Scheme 2. Reagents and conditions: (i) SnCl2, EtOH, reflux; (ii)
ArXCOOH, HCl 37%, reflux; (iii) ArXCOOH, coupling agents; (iv)
glacial acetic acid, 80 ꢁC.
NO2
N
NH2
6f
F
4a
5f
sible monoacylated derivatives was achieved. However,
this was not the case where EDC was used; in this in-
stance, LC–MS analysis of the reaction suggested the
formation of a stable intermediate with the acid, rea-
soned to be the rearrangement N-acylurea product.
Overall, it was observed that the reactions using CDI
activation procedure gave the cleanest profile by LC–
MS analysis. For this reason, for the lower cost of this
coupling agent and for the ease of removal of its
byproducts, CDI was finally selected for array produc-
tion. The crude intermediates obtained in the monoacy-
N
NO2
6g
57%
88%
F
NH2
4a
5g
O
NO2
N
NH2
6h
F
4a
5h

I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
2317
Table 4. Cyclisation conversions using HCl 37%
probably due to steric hindrance. Products 83–94 were
then purified by Si-column or HPLC-prep.
Ar
N
NH2
ArCH2COOH
N
In the case where ring-substituted phenylenediamines
were utilised, it was noted that after successful monoa-
cylation, the cyclisation step proved difficult. The major
reaction impurity was found to be the acetylated starting
material (Scheme 4), suggesting that reaction with sol-
vent was favoured over ring closure. Due to these reac-
tivity problems only a smaller set of compounds was
synthesised (Table 6). Where enough product was pres-
ent by LC–MS crude analysis, Si-column purification
was attempted and compounds passing internal purity
requirements were submitted for screening.
HCl
N
N
H
N
8a
11-15
Compound
ArCH2COOH
Conversion (%)
10
N
11
COOH
F
12
25
COOH
O
3. Results and conclusions
13
14
15
10
Ninety-three out of 106 potentially brain penetrant
benzoimidazole compounds were successfully synthes-
ised to a yield and purity sufficient for internal com-
pound registration (over 90% purity by LC–MS at
215 nm with no single impurity above 5%). Overall,
isolated yields were in the range of 15–60% allowing
for sufficient compound to be delivered to both screen-
ing and archive. The 93 so-prepared compounds were
initially tested at single concentration (5 lM) in a cellu-
lar assay based on GH4C1 cells stably expressing the
rat alpha7-nAChR using the FLuorometric Imaging
Plate Reader (FLIPR).29 The FLIPR methodology al-
lows the measurements of real time Ca2+-concentration
changes in living cells using a Ca2+ sensitive fluores-
cence dye (Fluo-4). Compounds which at 5 lM showed
activity greater than 20% of the response given by
10 lM nicotine were further validated in concentra-
tion-response curve analysis (Table 7).
COOH
Br
100
30
COOH
COOH
N
lation trials were then subjected to cyclisation in glacial
acetic acid to validate the final step. This proceeded suc-
cessfully with clean conversion to the desired benzoimi-
dazoles. With the optimised conditions a library of 96
compounds was prepared, exploiting automation tech-
nology to assist in the combinatorial amide formation
step. In Table 5 structures and reaction yields of the li-
brary are reported. A robust work-up procedure using
automation was also performed in order to have the
most efficient parallel procedure. All the reported auto-
mation steps described in this paper were carried out
using a Zinsser ‘Speedy’ liquid handling platform.28 In
the optimised procedure, when amide coupling comple-
tion was observed, the reaction blocks were moved to
the automation platform for work-up, and a stock solu-
tion of sodium hydroxide was dispensed across the vials
containing the reaction mixture in dichloromethane.
The vials were moved to a Zinsser shaker station for fas-
ter mixing and the samples recovered via probe aspira-
tion/dispensing action. Therefore, the samples were
aspirated from the bottom of the vials and transferred
to phase separators equipped with collection blocks,
also located on the platform bed.
Compounds 14 and 50 maintained the same activity of
the original hit (3), while compound 111 was slightly
more active. As the level of activity was only moderate
for the majority of compounds synthesised, at this point
we could not elaborate a full SAR around the structure.
Nevertheless some features appeared to be common in
the structure of the active compounds: the presence of
a para bromine substituent in the aryl moiety (B in
Fig. 2) and the dimethylamine as ionisable centre (D
in Fig. 2). These results support the putative binding
mode of benzoimidazole 3, which describes the space
around the protonated nitrogen to be limited, with more
hindered amines making steric clashes within the aro-
matic cage. The pyrrolidinone series was inactive con-
firming the importance of the protonated nitrogen.
The insertion of a methyl group in position 5 of the ben-
zoimidazole (compound 111) improved the interaction
with the hydrophobic pocket of the receptor.
During the preproduction phase it became evident that
highly hindered phenylenediamines (such as 8g) were
not reactive under the previously validated acylation
and cyclisation conditions. The preparation of com-
pounds 83–94 required further validation. Acylation dif-
ficulties could be related to steric hindrance of the
secondary amino group (Scheme 3). This problem was
overcome by performing the reaction in DMF at 80 ꢁC
with 2 equiv of the designated carboxylic acids as a li-
brary ‘subset’. The cyclisation step was achievable only
under microwave irradiation conditions at 140 ꢁC, again
The four best compounds were additionally tested for
their action against the alpha7-nAChR subtype
(Fig. 3). The fluorescence response was evaluated at
10 lM in the absence and in the presence of 1 lM meth-
yllycaconitine (MLA), a selective alpha7 receptor antag-
onist. The fluorescence induced by the compounds was
dramatically reduced when MLA was added to the cells,

2318
I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
Table 5. Structures and yields of compounds 11–106
Table 5 (continued)
X
Ar
Compound R1
Ar-X
Yield (%)
—
N
N
N
N
*
R1
27
O
*
*
R2
N
N
*
*
28
44
60
O
Compound R1
Ar-X
Yield (%)
—
Cl
N
N
*
11
12
*
29
*
O
O
F
N
N
*
*
Cl
F
44
35
N
*
*
30
50
*
*
O
O
O
13
*
N
N
N
*
*
*
31
54
45
44
O
O
*
N
N
Br
N
*
*
14
15
29
—
32
O
*
O
*
*
*
33
O
O
N
N
*
*
16
17
21
30
O
N
*
*
34
23
Cl
O
O
*
O
N
*
*
N
35
36
11
31
Cl
F
*
N
N
N
N
N
*
*
*
*
*
*
18
19
3
28
30
26
27
44
F
O
N
N
*
*
*
O
O
*
37
19
*
O
N
N
*
*
Br
N
38
39
15
26
*
*
21
22
O
O
*
*
*
O
N
N
*
*
40
41
30
23
N
N
*
N
23
24
35
44
O
O
*
Cl
F
*
*
*
*
O
*
*
Cl
F
O
N
N
*
*
42
43
21
14
N
N
25
26
54
23
O
O
O
*
*
O
Br
*

I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
2319
Table 5 (continued)
Table 5 (continued)
Compound R1
Ar-X
Yield (%)
27
Compound R1
Ar-X
Yield (%)
O
*
N
N
N
*
*
*
44
45
46
N
O
*
61
40
N
*
*
24
28
O
N
N
*
*
Br
N
62
63
40
33
N
N
*
O
*
O
*
*
N
47
—
40
N
N
N
*
*
*
*
*
64
65
18
41
N
N
F
Cl
48
N
*
*
*
O
O
Cl
F
49
39
N
N
N
*
*
*
*
66
67
26
35
N
N
O
Br
N
50
43
41
29
32
*
N
*
*
*
*
*
O
51
N
*
N
N
*
*
*
*
68
69
18
41
O
N
N
52
N
*
O
Cl
*
53
N
O
O
N
*
*
70
71
72
73
74
75
76
40
36
36
32
43
19
23
N
O
Cl
F
*
54
37
49
27
26
31
N
*
*
*
*
*
O
N
N
N
N
N
N
N
*
*
*
*
*
*
*
*
55
O
F
N
*
*
*
56
O
O
N
*
57
O
N
Br
N
*
O
*
58
N
O
*
*
N
N
*
*
N
59
—
25
N
*
F
60
(continued on next page)
N
*

2320
I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
Table 5 (continued)
Table 5 (continued)
Compound R1
Ar-X
Yield (%)
41
Compound R1
Ar-X
Yield (%)
20
Cl
N
*
*
77
78
79
80
81
82
N
N
N
N
N
N
*
*
*
*
*
*
93
O
O
*
*
Cl
F
O
N
*
41
35
39
44
36
*
94
—
57
55
64
64
20
57
40
56
67
46
40
59
O
O
O
*
N
O
95
N
N
N
N
N
N
N
N
N
N
N
N
*
*
*
*
*
*
*
*
*
*
*
*
*
F
O
O
O
O
O
O
O
O
O
O
O
*
O
96
*
*
O
*
O
97
O
*
Br
N
98
O
*
N
N
83
—
—
14
17
—
—
15
17
13
—
99
*
*
*
*
*
*
*
*
*
*
*
*
*
F
N
84
100
101
102
103
104
105
106
*
*
O
Cl
N
*
85
O
Cl
F
N
Br
N
*
86
O
*
N
*
87
O
*
*
N
*
88
O
Cl
N
*
*
89
O
O
Cl
F
O
N
*
*
90
O
O
N
*
91
O
proving that activity was indeed related to the alpha7
subtype.
N
*
92
O
Although we were disappointed to find that only a small
fraction of the compounds synthesised showed activity,

I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
2321
Ar
N
N
N
X
O
H
N
(i)
N
N
H
O
N
H
NH2
NH2
X Ar
8g
10
9
(ii)
Ar
X
N
N
N
83-94
Scheme 3. Reagents and conditions: (i) ArXCOOH, CDI, DMF, 80 ꢁC; (ii) glacial acetic acid, 140 ꢁC, microwave irradiation.
Ar
X
O
R2
N
H
N
R2
N
NH
N
Ar
NH2
X
O
9
10
(i)
Ar
Ar
O
X
X
O
N
R2
N
R2
N
N
N
N
N
NH
NH
Ar
R2
X
O
O
107-116
Scheme 4. Reagent and condition: (i) glacial acetic acid, 80 ꢁC.
the results obtained suggest that additional future mod-
ifications around the structure could improve activity
further. This includes varying the C and D portions,
chain lengths and further investigations into substitution
on the benzoimidazole ring.
using a Thermo LTQ Orbitrap Mass Spectromer. Ana-
lytical thin-layer chromatography (TLC) was carried
out using pre-coated plates of silica gel (5 · 20 cm). Sil-
ica gel chromatography was conducted using Isolute
prepacked 2 g Flash Si-columns. Microwave reactions
were conducted using a CEM Discover Synthesis Unit
(CEM Corp., Matthews, NC). Specs and Maybridge
compounds were purchased and screened without preli-
minary analysis.
4. Experimental
4.1. General information
4.2. Docking methods
All solvents and reagents were used as supplied, unless
otherwise stated. Reactions using air/moisture sensitive
reagents were run in a nitrogen atmosphere. High-per-
formance liquid chromatography (HPLC) analysis was
performed using a Waters 2795 module system with
Waters Micromass ZQ and Waters PDA 2996 detectors
connected to Masslynx 4.0 software. The column used
was an XTerra MS C18 (3.5 lm, 2.1 · 50 mm). 1H
NMR spectra were recorded on a NMR Varian Mer-
cury Plus 400 MHz, 5 mm PFG ATB Broadband probe.
Chemical shifts are reported in parts per million (ppm)
units (s: singlet, d: doublet, t: triplet, dd: doublet dou-
blet, m: multiplet). HRMS analysis was carried out
4.2.1. Generation of database for virtual screening. An sdf
file of commercially available molecules was obtained
directly from SPECS and Maybridge. The sdf file was
loaded onto an internal Oracle 10 g database using the
Accelrys Chemistry Cartridge. Physicochemical proper-
ties calculation was performed utilising the Accelrys Ac-
cord SDK. The collections were filtered to exclude
undesirable reactive groups by substructure searches.
A script was prepared in order to obtain suitable mini-
mised structures and protonation states. The procedure
entails addition of hydrogens and structure minimisa-
tion by Corina (Corina version 3.2 2005. Molecular

2322
I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
Table 6. Structures and yields of compounds 107–116
16000
14000
12000
10000
8000
6000
4000
2000
0
no MLA
MLA 1µM
Compound
R2
Ar-X
Yield (%)
20
O
*
107
Me
F
108
109
110
111
112
113
114
115
116
Me
Me
Me
Me
F
18
—
13
20
22
—
22
21
22
*
*
O
O
*
Nicotine
3
14
50
111
Compound
F
Figure 3. FLIPR/Ca2+ influx assay with MLA.
*
*
Br
Networks GmbH) followed by protonation at pH 7 and
re-minimisation using Ligprep (LigPrep version 1.6
2005. Schro¨dinger, LLC).
F
F
*
*
4.2.2. Preparation of homology model for docking. For
our docking simulations, we used the homology struc-
ture of alpha7 containing a docked nicotine molecule
in the agonist binding site located at the interface of
two subunits (a7gg.pdb).19 Nicotine was eliminated
from the pdb file. Active sites were selected manually
and contained residues Y91, W147, T148, Y186, C188,
C189 and Y193 from the alpha subunit and M36,
W53, N105, Q115 and L117 from the beta subunit.
F
O
*
*
F
F
F
Br
Table 7. Results from concentration-response analysis in functional Ca-flux assay for selected compounds
N
X
N
Ar
R2
R1
Compound
Nicotine
R2
R1
Ar-X
Curve top plateaua (%)
97
Alpha7 EC50 (lM)
1.5
O
*
*
N
*
*
3
H
52
2.6
2.6
N
14
50
66
111
H
135
Br
*
H
128
2.7
N
*
Br
Cl
N
*
b
*
H
10.0
1.7
N
N
O
*
*
Me
111
Br
a Curve top plateau of concentration-reponse analysis. Normalised to nicotine 10 lM (positive control).
b Did not reach the top plateau.

I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
2323
4.2.3. Docking calculation. Docking calculations were
performed using GOLD (GOLD version 2.2 2005.
CCDC). Default settings were generally used. Thus,
ten poses for each ligand were calculated. The command
allowing early termination was switched on and used as
one of the criteria to select virtual hits, when at least 3
docked conformations for the same ligand were within
6.86 (m, 1H), dH 7.41–7.45 (m, 1H), dH 8.15–8.17 (m,
1H), dH 8.40 (s, broad, 1H).
4.3.1.5. [3-(4-Methyl-piperazin-1-yl)-propyl]-(2-nitro-
phenyl)-amine (6e). Following general procedure A,
1
4.11 g was obtained (74% yield) H NMR (400 MHz,
CDCl3): dH 1.86–1.91 (m, 2H), dH 2.30 (s, 3H), dH
2.46–2.50 (m, 10H), dH 3.36–3.41 (m, 2H), dH 6.61–
6.65 (m, 1H), dH 6.87–6.89 (m, 1H), dH 7.40–7.45 (m,
1H), dH 8.16–8.18 (m, 1H), dH 8.24 (s, broad, 1H).
˚
1.5 A rms deviation. Both GoldScore and ChemScore
fitness functions were used.
4.3. Chemistry methods
4.3.1.6. [3-(2-Methyl-piperidin-1-yl)-propyl]-(2-nitro-
phenyl)-amine (6f). Following general procedure A,
4.3.1. General procedure A (6a–j). To a suspension of
potassium carbonate (40 mmol) in DMF (15 mL),
22 mmol of the required ortho-fluoronitrobenzenes
(1a–c) were added, along with 20 mmol of the appropri-
ate aminopropylamine (2a–j). The reaction was left mix-
ing at 90 ꢁC for 24 h. On reaction completion, the crude
was filtered and the collected solution concentrated.
This was then diluted with dichloromethane (25 mL)
and extracted twice with water (20 mL per extraction).
Products were then extracted from dichloromethane
phase using 20 mL of 6 N HCl solution. This aqueous
phase was then treated with dichloromethane (20 mL)
to remove the excess of ortho-fluoronitrobenzenes pres-
ent. The aqueous phase containing the desired com-
pound was basified with NaOH pellets and the
product extracted with dichloromethane (20 mL) which
was evaporated to leave the desired products in good
yield (60–90%).
1
5.15 g was obtained (93% yield) H NMR (400 MHz,
CDCl3): dH 1.05–1.07 (m, 1H), dH 1.25–1.37 (m, 2H),
dH 1.56–1.69 (m, 4H), dH 1.83–1.92 (m, 2H), dH 2.10–
2.17 (m, 1H), dH 2.28–2.36 (m, 1H), dH 2.37–2.43 (m,
1H), dH 2.81–2.90 (m, 3H), dH 3.32–3.37 (m, 2H), dH
6.62–6.64 (m, 1H), dH 6.86–6.88 (m, 1H), dH 7.40–7.45
(m, 1H), dH 8.16–8.18 (m, 1H), dH 8.24 (s, broad, 1H).
4.3.1.7. (8-Methyl-8-aza-bicyclo[3.2.1]oct-3-yl)-(2-nitro-
phenyl)-amine (6g). Following general procedure A,
1
3.39 g was obtained (65% yield) H NMR (400 MHz,
CDCl3): dH 1.81–1.85 (m, 2H), dH 1.96–1.99 (m, 2H),
dH 2.13–2.17 (m, 2H), dH 2.33–2.34 (m, 5H), dH 3.22–
3.24 (m, 2H), dH 3.71–3.95 (m, 1H), dH 6.60–6.65 (m,
1H), dH 6.73–6.75 (m, 2H), dH 7.40–7.44 (m, 1H), dH
8.17–8.20 (m, 1H), dH 8.69 (s, broad, 1H).
4.3.1.8. 1-[3-(2-Nitro-phenylamino)-propyl]-pyrrolidin-
2-one (6h). Following general procedure A, 4.63 g was
obtained (88% yield) H NMR (400 MHz, CDCl3): dH
1.92–1.99 (m, 2H), dH 2.04–2.07 (m, 2H), dH 2.39–2.43
(m, 2H), dH 3.33–3.38 (m, 2H), dH 3.41–3.46 (m, 4H),
dH 6.64–6.67 (m, 1H), dH 6.84–6.86 (m, 1H), dH 7.42–
7.46 (m, 1H), dH 8.10 (s, broad, 1H), dH 8.16–8.18 (m
1H).
4.3.1.1. N,N-Dimethyl-N0-(2-nitro-phenyl)-propane-1,3-
diamine (6a). Following general procedure A, 3.48 g was
obtained (78% yield) H NMR (400 MHz, CDCl3): dH
1.84–1.90 (m, 2H), dH 2.26 (s, 6H), dH 2.41–2.44 (m,
2H), dH 3.35–3.41 (m, 2H), dH 6.59–6.63 (m, 1H), dH
6.86–6.88 (m, 1H), dH 7.39–7.45 (m, 1H), dH 8.14–8.16
(m, 1H), dH 8.45 (s, broad, 1H).
1
1
4.3.1.2. (3-Morpholin-4-yl-propyl)-(2-nitro-phenyl)-
amine (6b). Following general procedure A, 4.87 g was
obtained (92% yield) H NMR (400 MHz, CDCl3): dH
1.86–1.91 (m, 2H), dH 2.42–2.50 (m, 6H), dH 3.38 (dd,
J = 12.03 Hz, 2H), dH 3.70–3.76 (m, 4H), dH 6.59–6.63
(m, 1H), dH 6.87 (d, J = 8.69 Hz, 1H), dH 7.39–7.43
(m, 1H), dH 8.14–8.16 (m, 1H), dH 8.26 (s, broad, 1H).
4.3.1.9. N,N-Dimethyl-N0-(4-methyl-2-nitro-phenyl)-
propane-1,3-diamine (6i). Following general procedure
A, 3.65 g was obtained (77% yield) 1H NMR
(400 MHz, MeOD): dH 1.87–1.94 (m, 2H), dH 2.31 (s,
6H), dH 2.35 (s, 3H), dH 2.50–2.54 (m, 2H), dH 3.39–
3.43 (m, 2H), dH 6.49–6.51 (m, 1H), dH 6.82 (s, 1H),
dH 7.98–8.00 (m, 2H).
1
4.3.1.3. N,N-Diethyl-N0-(2-nitro-phenyl)-propane-1,3-
diamine (6c). Following general procedure A, 4.22 g
was obtained (84% yield) H NMR (400 MHz, CDCl3):
dH 1.01–1.05 (t of d, J = 1.6, 5.2, 6H), dH 1.82–1.90 (m,
2H), dH 2.52–2.58 (m, 6H) dH 3.35–3.39 (m, 2H), dH
6.60–6.64 (m, 1H), dH 6.86–6.98 (d, 1H), dH 7.40–
7.44 (m, 1H), dH 8.16–8.18 (m, 1H), dH 8.41 (s, broad
1H).
4.3.1.10. N0-(4-Fluoro-2-nitro-phenyl)-N,N-dimethyl-
propane-1,3-diamine (6j). Following general procedure
A, 3.13 g was obtained (65% yield) 1H NMR (400 MHz,
MeOD): dH 1.84–1.92 (m, 2H), dH 2.29 (s, 6H), dH 2.44–
2.47 (m, 2H), dH 3.35–3.39 (m, 2 H), dH 6.39–6.44 (m,
1H), dH 6.71–6.75 (m, 1H), dH 8.18–8.22 (m, 1H).
1
4.3.2. General procedure B (8a–j). The nitrodiamine ob-
tained from procedure A (13 mmol) was dissolved in
ethanol (30 mL) and SnCl2ÆH2O was added (13 mmol,
1 equiv). The reaction mixture was stirred at reflux for
16 h. The reaction progress was controlled by LC–MS
analysis and if required, a further equivalent of SnCl2Æ-
H2O was added. Upon reaction completion, the crude
was cooled to room temperature and the solution trea-
ted with 10% NaOH to pH 8. The reaction mixture
4.3.1.4. [2-(1-Methyl-pyrrolidin-2-yl)-ethyl]-(2-nitro-
phenyl)-amine (6d). Following general procedure A,
1
4.43 g was obtained (89% yield) H NMR (400 MHz,
CDCl3): dH 1.55–1.63 (m, 1H), dH 1.72–1.80 (m, 3H),
dH 1.94–2.05 (m, 2H), dH 2.14–2.19 (m, 1H), dH 2.21–
2.31 (m, 1H), dH 2.36 (s, 3H), dH 3.08–3.13 (m, 1H),
dH 3.28–3.43 (m, 2H), dH 6.60–6.64 (m, 1H), dH 6.84–

2324
I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
was then treated with solid sodium tartrate until a clear
solution was obtained and the mixture stirred for
around 1 h, after which the reaction mixture was filtered
and the retained solution treated with NaOH to pH 12.
The desired product was extracted using dichlorometh-
ane (40 mL), drying the collected organic phase with so-
dium sulfate before evaporation of the solvent to leave
the pure desired product in good yield (ꢀ70%).
(s, broad, 1H), dH 6.56–6.58 (m, 1H), dH 6.66–6.76 (m,
2H), dH 6.79–6.84 (m, 1H).
4.3.2.8. 1-[3-(2-Amino-phenylamino)-propyl]-pyrrolidin-
2-one (8h). Following general procedure B, 2.21 g was
1
obtained (73% yield). H NMR (400 MHz, CDCl3): dH
1.82–1.88 (m, 2H), dH 2.00–2.06 (m, 2H), dH 2.39–2.43
(m, 2H), dH 3.14–3.18 (m, 2H), dH 3.67–3.43 (m, 4H),
dH 3.60 (s, broad, 2H), dH 6.63–6.71 (m, 3H), dH
6.76–6.80 (m, 1H).
4.3.2.1. N-(3-Dimethylamino-propyl)-benzene-1,2-dia-
mine (8a). Following general procedure B, 3.8 g was ob-
tained (76% yield). 1H NMR (400 MHz, CDCl3): dH
1.82–1.88 (m, 2 H), dH 2.27 (s, 6H), dH 2.42–2.45 (m,
2H), dH 3.17–3.20 (m, 2H), dH 3.34–3.35 (s, 2H), dH
6.64–6.68 (m, 2H), dH 6.70–6.72 (m, 1H), dH 6.80–6.84
(m, 1H).
4.3.2.9. N-(3-Diethylamino-propyl)-4-methylbenzene-
1,2-diamine (8i). Following general procedure B, 2.20 g
was obtained (82% yield). 1H NMR (400 MHz, CDCl3):
dH 1.84–1.88 (m, 2H), dH 2.25 (s, 3H), dH 2.30 (s, 6H),
dH 2.47–2.50 (m, 2H), dH 3.16–3.19 (m, 2H), 3.35
(s, broad, 2H), dH 6.44–6.46 (m 2H), dH 6.59–6.61 (m,
1H)
4.3.2.2. N-(3-Morpholin-4-yl-propyl)-benzene-1,2-dia-
mine (8b). Following general procedure B, 2.5 g was ob-
tained (83% yield). 1H NMR (400 MHz, CDCl3): dH
1.81–1.87 (m, 2H), dH 2.46–2.50 (m, 6H), dH 3.15–3.19
(m, 2H), dH 3.35 (s, broad, 2H), dH 3.72–3.74 (m, 4H),
dH 6.62–6.97 (m, 3H), dH 6.78–6.82 (m, 1H).
4.3.2.10. N-(3-Diethylamino-propyl)-4-fluorobenzene-
1,2-diamine (8j). Following general procedure B, 2.14 g
was obtained (78% yield). 1H NMR (400 MHz, CDCl3):
dH 1.82–1.85 (m, 2H), dH 2.26 (s, 6H), dH 2.42–2.45 (m,
2H), dH 3.13–3.16 (m, 2H), 3.35 (s, broad, 2H), dH 6.26–
6.35 (m, 2H), dH 6.58–6.61 (m, 1H).
4.3.2.3. N-(3-Diethylamino-propyl)-benzene-1,2-diamine
(8c). Following general procedure B, 2.61 g was ob-
tained (91% yield). 1H NMR (400 MHz, CDCl3): dH
1.04 (t, J = 7.14 Hz, 6H), dH 1.80–1.86 (m, 2H), dH
2.52–2.62 (m, 6H), dH 3.15–3.18 (m, 2H), dH 3.35
(s, broad, 2H), dH 6.62–6.70 (m, 3H), dH 6.78–6.82 (m,
1H).
4.3.3. General synthesis of 11–116. The required acid was
weighed into clean vials (0.39 mmol per reaction) and
dissolved in dichloromethane (1 mL per reaction). A
solution of N,N0-carbonyldiimidazole (CDI) in dichlo-
romethane (0.39 mmol per reaction, 1 mL per reaction)
was then added as a solution. The vials were then left
to shake for 2 h before adding a solution of the desired
amine in dichloromethane (0.39 mmol per reaction in
1 mL dichloromethane per reaction). The reactions were
then left to mix for 16 h. When LC–MS analysis showed
reaction completion, the blocks were moved to the auto-
mation platform for work-up. Here, the samples were
treated with 10% NaOH solution (2 mL per reaction)
and the organic phase collected and dried. (NMR data
were not taken at this stage). In the case of highly hin-
dered phenyleneamines, 2 equiv of the required acid
was used (0.78 mmol) as well as DMF in place of dichlo-
romethane (same amount as indicated above). The reac-
tion mixtures were heated to 80 ꢁC for 16 h before
solvent removal and work-up as with the remainder of
the library. The crude samples were dried and treated
with glacial acetic acid (3 mL per reaction) and the reac-
tion mixtures left to heat to 80 ꢁC for 4 h. After LC–MS
analysis to check for complete cyclisation, the reaction
mixtures were concentrated via solvent evaporation. In
the case of highly sterically hindered intermediates, the
reaction was carried out under microwave irradiation
exposure at 140 ꢁC for 15 min and solvent was removed
before work-up using the general procedure. The blocks
were then moved to the Zinsser platform for redissolu-
tion in dichloromethane (2 mL per reaction) followed
by an aqueous wash with 15% NaOH (2 mL per reac-
tion) using the procedure described above. The collected
organic phases were dried under nitrogen before parallel
purification using flash Si-columns. The desired prod-
ucts were found to elute using a dichloromethane/
MeOH mix (96:4).
4.3.2.4. N-[2-(1-Methyl-pyrrolidin-2-yl)-ethyl]-benzene-
1,2-diamine (8d). Following general procedure B, 1.76 g
was obtained (62% yield). 1H NMR (400 MHz, CDCl3):
dH 1.54–1.63 (m, 1H), dH 1.68–1.81 (m, 3H), dH 1.93–
2.02 (m, 2), dH 2.12–2.27 (m, 2H), dH 2.36 (s, 3H), dH
3.06–3.22 (m, 3H), dH 3.32 (s, broad, 2H), dH 6.64–
6.72 (m, 3H), dH 6.80–6.84 (m, 1H).
4.3.2.5. N-[3-(4-Methyl-piperazin-1-yl)-propyl]-benzene-
1,2-diamine (8e). Following general procedure B, 2.77 g
was obtained (86% yield). 1H NMR (400 MHz, CDCl3):
dH 1.83–1.90 (m, 2H), dH 2.30 (s, 3H), dH 2.41–2.54 (m,
10H), dH 3.17–3.20 (m, 2H), dH 3.35 (s, broad, 2H), dH
6.63–6.71 (m, 3H), dH 6.79–6.83 (m, 1H).
4.3.2.6. N-[3-(2-Methyl-piperidin-1-yl)-propyl]-benzene-
1,2-diamine (8f). Following general procedure B, 2.53 g
was obtained (79% yield). H NMR (400 MHz, CDCl3):
1
dH 1.06 (d, J = 6.24 Hz, 3H), dH 1.25–1.35 (m, 2H),
dH 1.51–1.68 (m, 4H), dH 1.77–1.92 (m, 2H), dH 2.07–
2.14 (m, 1H), dH 2.24–2.30 (m, 1H), dH 2.35–2.42 (m,
1H), dH 2.84–2.96 (m, 2H), dH 3.09–3.20 (m, 2H), dH
3.34 (s, broad, 2H), dH 6.62–6.71 (m, 3H), dH 6.78–6.82
(m, 1H).
4.3.2.7. N-(8-Methyl-8-aza-bicyclo[3.2.1]oct-3-yl)-ben-
zene-1,2-diamine (8g). Following general procedure B,
1.95 g was obtained (65% yield). H NMR (400 MHz,
1
CDCl3): dH 1.77 (d, J = 14.02 Hz, 2H), dH 2.02–2.08
(m, 4H), dH 2.26–2.34 (m, 5H), dH 3.16–3.22 (m, 2H),
dH 3.35 (s, broad, 2H), dH 3.42–3.46 (m, 1H), dH 3.64

I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
2325
For LC–MS analysis, an automated analytical plate
preparation procedure was also utilised to eliminate
the need for manual manipulation and hence possible er-
rors in sampling and dilution.
4.3.4.6. {3-[2-(4-Bromo-benzyl)-benzoimidazol-1-yl]-
propyl}-diethyl-amine (38). 1H NMR (400 MHz,
CDCl3): dH 0.97–1.00 (t, J = 7.2, 6H), dH 1.71–1.78
(m, 2H), dH 2.33–2.37 (t, J = 6.8, 2H), dH 2.45–2.50
(dd, J = 6.8, 7.2, 4H), dH 4.02–4.07 (t, J = 7.6, 2H), dH
4.32 (s, 2H). C21H26BrN3 Mass (calculated) [400.37];
(found) [M+H]+ = 401; LC tR = 1.32 min, 96% (10 min
method). HRMS m/z 400.13744 (M+1)+. Calcd mass
C21H27BrN3 400.13829.
4.3.4. Analytical data of 20 representative samples
4.3.4.1. Dimethyl-[3-(2-phenoxymethyl-benzoimidazol-
1
1-yl)-propyl]-amine(3). H NMR (400 MHz, CDCl3): dH
1.62 (s, 2H), dH 2.02–2.04 (m, 2H), dH 2.20 (s, 6H), dH
2.28 (t, J = 6.8, 2H), dH 5.42 (s, 2H), dH 6.97–7.00 (m,
1H), dH 7.07–7.09 (m, 1H), dH 7.28–7.32 (m, 5H), dH
7.42–7.44 (m, 1H), dH 7.78–7.80 (m, 1H). C19H23N3O,
Mass (calculated) [309.41]; (found) [M+H]+ = 310; LC
tR = 1.16 min, 100% (10 min method). HRMS m/z
310.19141 (M+1)+. Calcd mass for C19H24N3O
310.19139.
4.3.4.7. Diethyl-[3-(2-p-tolyloxymethyl-benzoimidazol-
1-yl)-propyl]-amine (45). H NMR (400 MHz, CDCl3):
1
dH 0.95–0.99 (t, J = 7, 6H), dH 1.97–2.04 (m, 2H), dH
2.28 (s, 3H), dH 2.44–2.52 (m, 6H), dH 4.33 (t, J = 7.6,
2H), dH 5.37 (s, 2H), dH 6.96 (d, J = 8, 2H), dH 7.08
(d, J = 8, 2H), dH 7.24–7.32 (m, 2H), dH 7.40–7.42
(m, 1H), dH 7.77–7.79 (m, 1H). C22H29N3O, Mass (cal-
culated) [351.50]; (found) [M+H]+ = 352; LC
tR = 1.62 min, 100% (10 min method). HRMS m/z
352.23768 (M+1)+. Calcd mass C22H30N3O 352.23834.
4.3.4.2. {3-[2-(2-Chloro-phenoxymethyl)-benzoimidazol-
1-yl]-ropyl}-dimethyl-amine (17). 1H NMR (400 MHz,
CDCl3): dH 2.01–2.08 (m, 2H), dH 2.19 (s, 6H), dH
2.28–2.31 (t, J = 6.6, 2H), dH 4.41–4.45 (t J = 7.4, 2H),
dH 5.51 (s, 2H), dH 6.90–6.94 (m, 1H), dH 7.19–7.23
(m, 1H), dH 7.27–7.37 (m, 4H), dH 7.44–7.46 (m, 1H),
dH 7.77–7.79 (m, 1H). C19H22ClN3O Mass (calculated)
[343.86]; (found) [M+H]+ = 344; LC tR = 1.73 min,
100% (10 min method). HRMS m/z 344.15252 (M+1)+.
Calcd mass for C19H23ClN3O 344.15242.
4.3.4.8. Diethyl-{3-[2-(2-methoxy-phenoxymethyl)-ben-
zoimidazol-1-yl]-propyl}-amine (46). 1H NMR (400 MHz,
CDCl3): dH 0.95–0.98 (t, J = 7.0, 6H), dH 1.98–2.05 (m,
2H), dH 2.46–2.51 (m, 6H), dH 3.85 (s, 3H), dH 4.39–
4.43 (t, J = 7.4, 2H), dH 5.44 (s, 2H), dH 6.86–6.90 (m,
2H), dH 6.95–6.99 (m, 1H), dH 7.17–7.19 (m, 1H), dH
7.25–7.31 (m, 2H), dH 7.40–7.43 (m, 1H), dH 7.76–7.78
(m, 1H). C22H29N3O2 Mass (calculated) [367.50]; (found)
[M+H]+ = 368; LC tR = 1.36 min, 100% (10 min meth-
od). HRMS m/z 368.23258 (M+1)+. Calcd mass for
C22H30N3O2 368.23325.
4.3.4.3. Dimethyl-{4-[1-(3-morpholin-4-yl-propyl)-1H-
benzoimidazol-2-ylmethyl]-phenyl}-amine (23). H NMR
1
(400 MHz, CDCl3): dH 2.08 (s, 2H), dH 2.22 (t, J = 6.8,
2H), dH 2.33 (s, 4H), dH 2.90 (s, 6H), dH 3.71 (t,
J = 4.4, 4H), dH 4.08 (t, J = 7.2, 2H), dH 4.29 (s, 2H),
dH 6.65 (d, J = 8.4, 2H), dH 7.11 (d, J = 8.4, 2H), dH
7.23–7.30 (m, 3H), dH 7.78–7.80 (m, 1H). C23H30N4O,
Mass (calculated) [378.41]; (found) [M+H]+ = 379; LC
tR = 0.35 min, 100% (10 min method). HRMS m/z
379.24939 (M+1)+. Calcd mass for C23H31N4O
379.24924.
4.3.4.9. 2-(4-Fluoro-phenoxymethyl)-1-[2-(1-methyl-pyrr-
olidin-2-yl)-ethyl]-1H-benzoimidazole (55). 1H NMR
(400 MHz, CDCl3): dH 1.61 (s, 2H), dH 1.68–2.01 (m,
4H), dH 2.12–2.2 (m, 2H), dH 2.28 (s, 3H), dH 3.04–
3.09 (m, 1H), dH 4.24–4.39 (m, 2H), dH 5.36 (s, 2H),
dH 6.95–7.05 (m, 4H), dH 7.28–7.33 (m, 2H), dH 7.38–
7.40 (m, 1H), dH 7.78–7.80 (m, 1H). C21H24FN3O Mass
(calculated) [353.44]; (found) [M+H]+ = 354; LC
tR = 1.50 min, 100% (10 min method). HRMS m/z
354.19715 (M+1)+. Calcd mass for C21H25FN3O
354.19647.
4.3.4.4. 2-(2-Chloro-phenoxymethyl)-1-(3-morpholin-
4-yl-propyl)-1H-benzoimidazole
(29).
1H
NMR
(400 MHz, CDCl3): dH 2.07–2.12 (m, 3H), dH 2.37–
2.40 (m, 6H), dH 3.64–3.67 (t, J = 4.4, 3H), dH 4.44–
4.47 (t, J = 7.2, 2H), dH 5.52 (s, 2H), dH 6.92–6.96 (m,
1H), dH 7.21–7.25 (m, 1H), dH 7.28–7.37 (m, 4H), dH
7.43–7.46 (m, 1H), dH 7.78–7.81 (m, 1H). C21H24ClN3O2
Mass (calculated) [385.90]; (found) [M+H]+ = 387; LC
tR = 1.56 min, 100% (10 min method). HRMS m/z
386.16301 (M+1)+. Calcd mass for C21H25ClN3O2
386.16298.
4.3.4.10. 2-(2-Fluoro-benzyl)-1-[3-(4-methyl-piperazin-
1-yl)-propyl]-1H-benzoimidazole (60). 1H NMR
(400 MHz, CDCl3): dH 1.69 (s, 2H), dH 1.80–1.87 (m,
2H), dH 2.22–2.25 (t, J = 6.8, 2H), dH 2.30 (s, 3H), dH
2.36–2.46 (m, 6H), dH 4.12–4.15 (t, J = 7.2, 2H), dH
4.42 (s, 2H), dH 7.02–7.10 (m, 2H), dH 7.15–7.27 (m,
4H), dH 7.35–7.37 (m, 1H), dH 7.75–7.77 (m, 1H).
C22H27FN4 Mass (calculated) [366.49]; (found)
[M+H]+ = 367; LC tR = 0.53 min, 100% (10 min meth-
od). HRMS m/z 367.22893 (M+1)+. Calcd mass for
C22H28FN4 367.22925.
4.3.4.5. 2-(2-Methoxy-phenoxymethyl)-1-(3-morphol-
in-4-yl-propyl)-1H-benzoimidazole (34). 1H NMR
(400 MHz, CDCl3): dH 2.05–2.09 (m, 2H), dH 2.33–
2.36 (m, 6H), dH 3.65 (t, J = 4.4, 4H), dH 3.85 (s, 3H),
dH 4.45 (t, J = 7.2, 2H), dH 5.45 (s, 2H), dH 6.88–6.91
(m, 2H), dH 6.96–6.99 (m, 1H), dH 7.17–7.19 (m, 1H),
dH 7.25–7.31 (m, 2H), dH 7.42–7.44 (m, 1H), dH 7.76–
7.79 (m, 1H). C22H27N3O3, Mass (calculated) [381.48];
(found) [M+H]+ = 382; LC tR = 1.24 min, 100%
(10 min method). HRMS m/z 382.21190 (M+1)+. Calcd
mass for C22H28N3O3 382.21253.
4.3.4.11. 2-(3-Chloro-phenoxymethyl)-1-[3-(4-methyl-
piperazin-1-yl)-propyl]-1H-benzoimidazole (66). H NMR
1
(400 MHz, CDCl3): dH 2.03–2.04 (m, 4H), dH 2.31 (s,
3H), dH 2.33–2.48 (m, 6H + H2O), dH 4.34–4.37 (t,
J = 7.0, 4H), dH 5.41 (s, 2H), dH 6.97–7.00 (m, 2H), dH
7.07–7.08 (m, 1H), dH 7.21–7.25 (m, 1H), dH 7.29–7.32

2326
I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
(m, 2H), dH 7.42–7.44 (m, 1H), dH 7.79–7.81 (m, 1H).
C22H27ClN4O Mass (calculated) [398.94]; (found)
[M+H]+ = 400; LC tR = 1.79 min, 100% (10 min meth-
od). HRMS m/z 399.19436 (M+1)+. Calcd mass for
C22H28ClN4O 399.19462.
tR = 1.79 min, 98% (10 min method). HRMS m/z
364.20190 (M+1)+. Calcd mass C22H26N3O2 364.20195.
4.3.4.17. 1-[3-(2-Pyridin-3-ylmethyl-benzoimidazol-1-
yl)-propyl]-pyrrolidin-2-one (99). 1H NMR (400 MHz,
CDCl3): dH 1.75–1.79 (m, 2H), dH 1.96–2.03 (m, 2H),
dH 2.37–2.41 (m, 2H), dH 3.17–3.21 (m, 2H), dH 3.28–
3.32 (m, 2H), dH 4.06–4.10 (m, 2H), dH 4.34 (s, 2H),
dH 7.23–7.29 (m, 4H), dH 7.55–7.62 (m, 1H), dH 7.75–
7.78 (m, 1H), dH 8.51–8.52 (m, 1H), 8.58–8.59 (m,
1H). C20H22N4O Mass (calculated) [334.42]; (found)
[M+H]+ = 335; LC tR = 0.67 min, 100% (10 min meth-
od). HRMS m/z 335.18658 (M+1)+. Calcd mass
C20H23N4O 335.18664.
4.3.4.12. 2-(2-Methoxy-phenoxymethyl)-1-[3-(4-methyl-
piperazin-1-yl)-propyl]-1H-benzoimidazole (70). H NMR
1
(400 MHz, CDCl3): dH 1.66 (m, 2H), dH 2.02–2.09 (m,
2H), dH 2.27 (s, 3H), dH 2.33–2.43 (m, 8H), dH 3.85 (s,
3H), dH 4.42–4.45 (t, J = 7.2, 2H), dH 5.45 (s, 2H),
6.87–6.91 (m, 2H), dH 6.96–6.99 (m, 1H), dH 7.18–7.20
(m, 1H), dH 7.25–7.31 (m, 2H), dH 7.41–7.45 (m, 1H),
dH 7.76–7.78 (m, 1H). C23H30N4O Mass (calculated)
[394.52]; (found) [M+H]+ = 395; LC tR = 1.18 min,
100% (10 min method). HRMS m/z 379.24900 (M+1)+.
Calcd mass C23H31N4O 379.24924.
4.3.4.18. 1-{3-[2-(3-Chloro-phenoxymethyl)-benzoimi-
dazol-1-yl]-propyl}-pyrrolidin-2-one (102). 1H NMR
(400 MHz, CDCl3): dH 1.88–1.96 (m, 2H), dH 2.05–2.12
(m, 2H), dH 2.52–2.36 (m, 2H), dH 3.27–3.30 (t, J = 7.0,
2H), dH 3.41–3.44 (t, J = 7.0, 2H), dH 4.28–4.31 (m, 2H),
dH 5.36 (s, 2H), dH 6.98–7.02 (m, 2H), dH 7.06–7.07 (m,
1H), dH 7.21–7.25 (m, 1H), dH 7.30–7.39 (m, 3H), dH
7.79–7.81 (m, 1H). C21H22ClN3O2 Mass (calculated)
[383.88]; (found) [M+H]+ = 384; LC tR = 2.69 min,
100% (10 min method). HRMS m/z 384.14752 (M+1)+.
Calcd mass C21H23ClN3O2 384.14733.
4.3.4.13. 2-(2-Methoxy-benzyl)-1-[3-(2-methyl-piperi-
din-1-yl)-propyl]-1H-benzoimidazole (73). 1H NMR
(400 MHz, CDCl3): dH 0.95–0.96 (d, J = 6, 3H), dH
1.24–1.30 (m, 2H), dH 1.66–1.82 (m, 3H), dH 1.96–2.02
(m, 1H), dH 2.18–2.27 (m, 2H), dH 2.57–2.69 (m, 2H),
dH 3.88 (s, 3H), dH 3.97–4.11 (m, 2H), dH 4.30–4.39
(d,d, J = 16,6.4, 2H), dH 6.82–6.90 (m, 2H), dH 7.04–
7.06 (m, 1H), dH 7.19–7.26 (m, 3H), dH 7.30–7.33 (m,
1H), dH 7.74–7.77 (m, 1H). C24H31N3O Mass (calculated)
[377.53]; (found) [M+H]+ = 378; LC tR = 0.77 min, 98%
(10 min method). HRMS m/z 378.25393 (M+1)+. Calcd
mass C23H32N3O 378.25399.
4.3.4.19.
{3-[2-(4-Fluoro-phenoxymethyl)-5-methyl-
benzoimidazol-1-yl]-propyl}-dimethyl-amine (110). 1H
NMR (400 MHz, DMSO-d6): dH 1.84–1.91 (m, 2H),
dH 2.08 (s, 6H), dH 2.15–2.18 (t, J = 6.4, 2H), dH 2.43
(s, 3H), dH 4.23–4.27 (t, J = 7.2, 2H), dH 5.34 (s, 2H),
dH 7.01–7.04 (m, 1H), dH 7.08–7.16 (m, 4H), dH 7.38
(s, 1H), dH 7.49–7.51 (d, J = 8.4, 1H). C20H24FN3O
Mass (calculated) [341.43]; (found) [M+H]+ = 342;
LC tR = 1.86 min, 100% (10 min method). HRMS
m/z 342.19745 (M+1)+. Calcd mass C20H25FN3O
342.19762.
4.3.4.14. 2-Benzyl-1-[3-(2-methyl-piperidin-1-yl)-propyl]-
1H-benzoimidazole (76). H NMR (400 MHz, CDCl3):
1
dH 0.94–0.96 (d, J = 6.4, 3H), dH 1.29–1.33 (m, 2H),
dH 1.51–1.80 (m, 5H + H2O), dH 1.94–2.23 (m, 4H), dH
2.55–2.68 (m, 2H), dH 3.94–4.10 (m, 2H), dH 4.32–4.43
(dd J = 15.6, 12.8 2H), dH 7.22–7.31 (m, 8H), dH 7.76–
7.78 (m, 1H). C23H29N3 Mass (calculated) [347.51];
(found) [M+H]+ = 348; LC tR = 0.71 min, 100%
(10 min method). HRMS m/z 348.24328 (M+1)+. Calcd
mass C23H30N3 348.24342.
4.3.4.20.{3-[2-(4-Bromo-benzyl)-5-fluoro-benzoimidazol-
1-yl]-propyl}-dimethyl-amine (116). H NMR (400 MHz,
1
DMSO-d6): dH 1.59–1.66 (m, 2H), dH 2.04–2.08 (m, 8H),
dH 4.10–4.13 (t, J = 7.2, 2H), dH 4.28 (s, 2H), dH 6.96–
7.03 (m, 1H), dH 7.21–7.24 (m, 2H), dH 7.39–7.42 (d,d
J = 2.6,6.8, 1H) dH 7.49–7.51 (m, 2H), dH 7.52–7.56 (m,
1H). C19H21BrFN3 Mass (calculated) [390.30]; (found)
[M+H]+ = 392; LC tR = 1.93 min, 100% (10 min method).
HRMS m/z 390.09759 (M+1)+. Calcd mass C19H22BrFN3
390.09757.
4.3.4.15. 2-(2-Chloro-phenoxymethyl)-1-[3-(2-methyl-
piperidin-1-yl)-propyl]-1H-benzoimidazole (77). H NMR
1
(400 MHz, CDCl3): dH 1.04–1.05 (d, J = 6.4, 3H), dH
1.29–1.69 (m, 6H), dH 2.10–2.16 (m, 3H), dH 2.49–2.56
(m, 2H), dH 2.86–2.97 (m, 2H), dH 4.33–4.47 (m, 2H), dH
5.46–5.54 (d,d, J = 12.8, 8, 2H), dH 6.92–6.95 (m, 1H), dH
7.20–7.25 (m, 1H), dH 7.29–7.37 (m, 4H), dH 7.46–7.47
(m, 1H). C23H28ClN3O Mass (calculated) [397.95]; (found)
[M+H]+ = 399; LC tR = 2.11 min, 100% (10 min method).
HRMS m/z 398.19932 (M+1)+. Calcd mass C23H29ClN3O
398.19937.
LC–MS analysis of all compounds (12–116) is reported
in Supporting information.
4.4. Biological methods
4.3.4.16. 1-{3-[2-(2-Methoxy-benzyl)-benzoimidazol-1-
yl]-propyl}-pyrrolidin-2-one (97). 1H NMR (400 MHz,
CDCl3): dH 1.69–1.75 (m, 2H), dH 1.92–1.99 (m, 2H),
dH 2.33–2.37 (m, 2H), dH 3.13–3.17 (m, 2H), dH 3.23–
3.27 (t, J = 7.2, 2H), dH 3.91 (s, 3H), dH 4.04–4.08 (t
J = 8, 2H), dH 4.31 (s, 2H), dH 6.83–6.87 (m, 1H), dH
6.90–6.92 (d, J = 8, 1H), dH 7.07–7.09 (m, 1H), dH
7.20–7.28 (m, 4H), dH 7.75–7.70 (m, 1H). C22H25N3O2
Mass (calculated) [363.46]; (found) [M+H]+ = 364; LC
4.4.1. Cell culture. GH4C1-F7 cells stably expressing the
rat- alpha7-nAChR were cultivated in HAM’s F10 med-
ium supplemented with 2.5% FBS, 15% horse serum, 1%
penicillin/streptomycin, 0.5% GLUTAMAX and
0.2 mg/ml Hygromycin at 37 ꢁC and 5% CO2 and prop-
agated in a Monday–Friday routine. For cellular assays
the cells were seeded on poly-D-lysine treated 96-well
plates; black/clear bottom; 100,000 cells/well.

I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
2327
4.4.2. FLIPR/Ca2+ influx assay. A plate of GH4C1-F7
Acknowledgments
cells was prepared and the medium was removed. It
was then incubated with 100 lL of labelling solution
(HBSS-20 mM Hepes supplemented with 4 mM Fluo-
4, 0.02% pluronic acid and 5 mM probenecid) at
37 ꢁC for 30 min. After removing the labelling solution
200 lL of HBSS-20 mM Hepes supplemented with
2.5 mM probenecid was added and the plate trans-
ferred into the FLIPR. On an additional plate1 the test
compounds and controls were prepared as five times
concentrated solutions in HBSS-20 mM Hepes. In a
second plate2 each of the wells was filled with six times
concentrated nicotine solution; except for the assay
buffer controls. Both plates were transferred into the
FLIPR. During a FLIPR run a baseline was recorded
first. After addition from the compound plate (first
addition, 5 lM final concentration) the data were col-
lected every second for 1 min, than each 30 s for
10 min. For the second addition nicotine was added
(10 lM final concentration) to plate2 followed by an
additional data collection every second for 1 min, then
each 3 s for 3 min. The height of the peaks for the first
and second addition was exported using the MAX–
MIN function of the FLIPR software and normalised
against the positive control (10 lM nicotine signal).
Compounds showing only a signal for the first addition
were classified as agonists. Compounds showing no sig-
nal at the first addition and a signal for the second
addition were classified as inactive, compounds show-
ing no signal for first and second addition were classi-
fied as antagonists. Identified hits were further
validated in concentration-response curves.
We thank Dr. Eva Genesio for her assistance in NMR
analysis and Dr. Stefano Gotta for the exact mass deter-
mination. Dr. Graeme M. Robertson’s careful reading
of the manuscript is also acknowledged.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmc.2007.
11.068.
References and notes
1. Picciotto, M. R.; Caldarone, B. J.; King, S. L.; Zachariou,
V. Neuropsychopharmacology 2000, 22, 451–465.
2. Davis, K. L.; Yamamura, H. I. Life Sci. 1978, 23, 1729–
1734.
3. Holladay, M. W.; Dart, M. J.; Lynch, J. K. J. Med. Chem.
1997, 40, 4169–4194.
4. (a) 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. J. Med.
Chem. 1996, 39, 817–825; (b) Tatsumi, R.; Seio, K.; Fujio,
M.; Katayama, J.; Horikawa, T.; Hashimoto, K.; Tanaka,
H. Bioorg. Med. Chem. Lett. 2004, 16, 3781–3784; (c)
Bodnar, A. L.; Cortes-Burgos, L. A.; Cook, K. K.; Dinh,
D. M.; Groppi, V. E.; Hajos, M.; Higdon, N. R.;
Hoffmann, W. E.; Hurst, R. S.; Myers, J. K.; Rogers, B.
N.; Wall, T. M.; Wolfe, M. L.; Wong, E. J. Med. Chem.
2005, 48, 905–908; (d) Bunnelle, W. H.; Dart, M. J.;
Schrimpf, M. R. Curr. Med. Chem. 2004, 4, 299–334; (e)
Tatsumi, R.; Fujio, M.; Takanashi, S.; Numata, A.;
Katayama, J.; Satoh, H.; Shiigi, Y.; Maeda, J.; Kuriyama,
M.; Horikawa, T.; Murozono, T.; Hashimoto, K.;
Tanaka, H. J. Med. Chem. 2006, 49, 4374–4383.
5. CCDC, GOLD version 2.2 2005.
6. Leonik, F. M.; Papke, R. L.; Horenstein, N. A. Bioorg.
Med. Chem. Lett. 2007, 17, 1520–1522.
7. Mazurov, A.; Klucik, J.; Miao, L.; Phillips, T. Y.;
Seamans, A.; Schmitt, J. D.; Hauser, T. A.; Johnson, R.
T., Jr.; Miller, C. Bioorg. Med. Chem. Lett. 2005, 15,
2073–2077.
8. Macor, J. E.; Gurley, D.; Lanthorn, T.; Loch, J.; Mack, R.
A.; Mullen, G.; Tran, O.; Wright, N.; Gordon, J. C.
Bioorg. Med. Chem. Lett. 2001, 11, 319–321.
9. Lanzafame, A. A.; Sexton, P. M.; Christopoulos, A. Mol.
Pharmacol. 2006, 70, 736–746.
4.4.3. Data analysis. EC50 and IC50 values were calcu-
lated using the IDBS XLfit4.1 software package employ-
ing the sigmoidal concentration-response (variable
slope) equation 205:
HillSlope
Y ¼ Bottom þ ððTop–BottomÞ=ð1 þ ððEC50=XÞ
ÞÞ
X is the logarithm of the concentration, Y is the re-
sponse. Bottom is the bottom plateau of the curve,
Top the top plateau.
4.4.4. FLIPR/Ca2+ influx assay with MLA. A plate of
GH4C1-F7 cells was prepared and the medium was re-
moved. It was then incubated with 100 lL of labelling
solution (HBSS-20 mM Hepes supplemented with
4 mM Fluo-4, 0.02% pluronic acid and 5 mM probenecid)
at 37 ꢁC for 30 min. After removing the labelling solution,
200 lL of washing solution (HBSS-20 mM Hepes supple-
mented with 2.5 mM probenecid) with or without MLA
1 lM was added. The plate was then transferred into the
FLIPR. On an additional plate, test compounds and con-
trols were prepared as five times concentrated solutions in
HBSS-20 mM Hepes. Both plates were transferred into
the FLIPR. At the beginning of the FLIPR assay the base-
line was recorded. After addition of compounds, data
were collected every second for 1 min, then each 30 s for
10 min. The peak height was exported using the MAX–
MIN function of the FLIPR software. Compounds show-
ing a nicotinic signal that disappeared in the presence of
MLA are considered selective alpha7 agonists.
10. Schmitz, W. D.; Denhart, D. J.; Brenner, A. B.; Ditta, J.
L.; Mattson, R. J.; Mattson, G. K.; Molski, T. F.; Macor,
J. E. Bioorg. Med. Chem. Lett. 2005, 15, 1619–1621.
11. http://www.accelrys.com.
12. (a) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.;
Feeney, P. J. Adv. Drug Del. Rev. 1997, 23, 3–25; (b)
Lipinski, C. A. J. Pharm. Tox. Meth. 2000, 44, 235; (c)
Norinder, U.; Haeberlein, M. Adv. Drug Del. Rev. 2002,
54, 291–313; (d) Veber, D. F.; Johnson, S. R.; Cheng, H-
Y.; Smith, B. R.; Ward, K. W.; Kopple, K. D. J. Med.
Chem. 2002, 45, 2615; (e) Mahar Doan, K. M.; Humph-
reys, J. E.; Webster, L. O.; Wring, S. A.; Shampine, L. J.;
Serabjit-Singh, J. C.; Adkison, K. K.; Polli, J. W. J.
Pharmacol. Exp. Ther. 2002, 303, 1029–1037.
13. http://www.spotfire.com.

2328
I. Micco et al. / Bioorg. Med. Chem. 16 (2008) 2313–2328
14. (a) Nakane, M.; Cowart, M. D.; Hsieh, G. C.; Miller, L.;
Uchic, M. E.; Chang, R.; Terranova, M. A.; Donnelly-
Roberts, D. L.; Namovic, M. T.; Miller, T. R.; Wetter, J.
M.; Marsh, K.; Stewart, A. O.; Brioni, J. D.; Moreland, R.
B. Neuropharmacology 2005, 49, 112–121; (b) Huang, S.;
Lin, R.; Yu, Y.; Lu, Y.; Connolly, P. J.; Chiu, G.; Li, S.;
Emanuel, S. L.; Middleton, S. A. Bioorg. Med. Chem.
Lett. 2007, 17, 1243–1245; (c) Ng, R. A.; Lanter, J. C.;
Alford, V. C.; Allan, G. F.; Sbriscia, T.; Lundeen, S. G.;
Sui, Z. Bioorg. Med. Chem. Lett. 2007, 17, 1784–1787.
15. Karlin, A.; Akabas, M. H. Neuron 1995, 15, 1231–1244.
16. Le Novere, N.; Grutter, T.; Changeux, J. P. Proc. Natl.
Acad. Sci. U.S.A. 2002, 99, 3210–3215.
20. http://www.maybridge.com; http://www.specs.net.
21. Corina version 3.2 2005. Molecular Networks GmbH.
22. LigPrep version 1.6 2005. Schro¨dinger, LLC.
23. Lester, H. A.; Dibas, M. I.; Dahan, D. S.; Leite, J. F.;
Dougherty, D. A. Trends in Neurosciences 2004, 27, 329–
336.
24. Poulain, R.; Horvath, D.; Bonnet, B.; Eckhoff, C.;
´
Chapelain, B.; Bodinier, M.-C.; Deprez, B. J. Med. Chem.
2001, 21, 3378–3390.
25. Bellamy, F. D.; Ou, K. Tetrahedron Lett. 1984, 25, 839–
842.
26. Elderfield, R. C.; Meyer, V. B. J. Am. Chem. Soc. 1954, 76,
1883–1886.
17. Schapiro, M.; Abagyan, R.; Totrov, M. BMC Struct. Biol.
2002, 2, 1–9.
18. Brejc, K.; van Dijk, W. J.; Klaassen, R. V.; Schuurmans,
M.; van Der Oost, J.; Smit, A. B.; Sixma, T. K. Nature
2001, 411, 269–276.
27. White, W.; Almassy, R.; Calvert, A. H.; Curtin, N. J.;
Golding, B. T.; Griffin, R. J.; Maegley, K.; Newell, D. R.;
Srinivasan, S. J. Med. Chem. 2000, 43, 4084–4097.
28. http://www.zinsser.com.
29. Dunlop, J.; Roncarati, R.; Jow, B.; Bothmann, H.; Lock,
T.; Kowal, D.; Bowlby, M.; Terstappen, G. C. Biochem-
ical Pharmacology 2007, 74, 1172–1181.
19. a7gg_nic.pdb available from http://www.ebi.ac.uk/comp-
neur-srv/LGICdb/HTML/ACHa7gaga.php.