Synthesis, biological activity, QSAR and QSPR study of 2- aminobenzimidazole derivatives as potent H3-antagonists

Article abstract of DOI:10.1016/j.bmc.2003.11.030

We report the design, synthesis, QSPR and QSAR of a new class of H ₃-antagonists, having a 2-aminobenzimidazole moiety connected to the 4(5) position of an imidazole ring through di- or tri-methylene chains. Eleven substituents, selected by experimental design to obtain broad and non-correlated variation in their lipophilic, electronic and steric properties, were introduced at the 5(6) position of the benzimidazole nucleus. The compounds were tested for their H₃-receptor affinity, by displacement of [³H]-(R)-α-methylhistamine ([ ³H]-RAMHA) binding to rat brain membranes (pK_i), for intrinsic activity, evaluating their effect on [³⁵S]GTPγS binding to rat brain membranes, and for H₃-antagonist potency, on electrically stimulated guinea-pig ileum (pK_B). The pK_i values of the derivatives with longer chain (5a-k) ranged over 2 orders of magnitude, with the 5(6)-methoxy derivative 5d endowed with sub-nanomolar affinity (pK_i=9.37). The series having two methylene groups in the chain spacer (4a-k), showing a small variation in affinity, revealed to be somewhat insensitive to ring substitution. Lipophilicity (log P) and basicity (pK_a) of the newly synthesized compounds were measured and related to receptor affinity in a QSAR study. Multiple regression analysis (MRA) showed an approximate parabolic dependence of pK_i on log P, while an additional electronic effect of the substituents on benzimidazole tautomerism is suspected.

Full text of DOI:10.1016/j.bmc.2003.11.030

Bioorganic & Medicinal Chemistry 12 (2004) 663–674
Synthesis, biological activity, QSAR and QSPR study of
2-aminobenzimidazole derivatives as potent H3-antagonists
Marco Mor,a,* Fabrizio Bordi,a Claudia Silva,a Silvia Rivara,a Valentina Zuliani,a
Federica Vacondio,a Mirko Rivara,a Elisabetta Barocelli,b Simona Bertoni,b
Vigilio Ballabeni,b Francesca Magnanini,b Mariannina Impicciatoreb
and Pier Vincenzo Plazzia
a
`
Dipartimento Farmaceutico, Universita degli Studi di Parma, Parco Area delle Scienze 27/A, I-43100 Parma, Italy
b
`
Dipartimento di Scienze Farmacologiche, Biologiche e Chimiche Applicate, Universita degli Studi di Parma,
Parco Area delle Scienze 27/A, I-43100 Parma, Italy
Received 22 September 2003; accepted 25 November 2003
Abstract—We report the design, synthesis, QSPR and QSAR of a new class of H3-antagonists, having a 2-aminobenzimidazole
moiety connected to the 4(5) position of an imidazole ring through di- or tri-methylene chains. Eleven substituents, selected by
experimental design to obtain broad and non-correlated variation in their lipophilic, electronic and steric properties, were intro-
duced at the 5(6) position of the benzimidazole nucleus. The compounds were tested for their H3-receptor affinity, by displacement
of [3H]-(R)-a-methylhistamine ([3H]-RAMHA) binding to rat brain membranes (pKi), for intrinsic activity, evaluating their effect
on [35S]GTPgS binding to rat brain membranes, and for H3-antagonist potency, on electrically stimulated guinea-pig ileum (pKB).
The pKi values of the derivatives with longer chain (5a–k) ranged over 2 orders of magnitude, with the 5(6)-methoxy derivative 5d
endowed with sub-nanomolar affinity (pKi=9.37). The series having two methylene groups in the chain spacer (4a–k), showing a
small variation in affinity, revealed to be somewhat insensitive to ring substitution. Lipophilicity (log P) and basicity (pKa) of the
newly synthesized compounds were measured and related to receptor affinity in a QSAR study. Multiple regression analysis (MRA)
showed an approximate parabolic dependence of pKi on log P, while an additional electronic effect of the substituents on benzi-
midazole tautomerism is suspected.
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
increase the synthesis and the release of histamine,
blocking a negative feed-back mechanism,3,5,6 and the
release of several other neurotransmitters, such as acet-
ylcholine, noradrenaline, dopamine and serotonin.7À10
The high density of H3-receptors in the CNS suggests
that H3-antagonists could be useful for the treatment of
central disorders in which an enhancement of neuro-
transmitter release is required, such as diseases invol-
ving impaired cognitive functions.11,12 In the last few
years a large number of potent H3-receptor antagonists
have been reported in the literature. Most of them con-
tain the classical structure consisting of an imidazole
ring connected by a spacer to a polar group, which is
attached to a lipophilic ending group.13 This general
formula can be recognized in many derivatives (see
Chart 1), having different polar groups, such as thiour-
eas (e.g., thioperamide14), isothioureas (e.g., clobenpro-
pit15 and iodophenpropit16), guanidines (1),17
carbamates (2),18 sulfonamides (3,4).19 In some series of
H3-antagonists, the polar group is represented by a
Histamine acts as a neurotransmitter in the peripheral
and central nervous system (CNS), exerting its pharma-
cological actions through the activation of distinct
receptor subtypes, named H1, H2, H3 and H4,1,2
belonging to the family of G-protein coupled receptors.
The recent cloning of the gene encoding the human
H3-receptor, discovered in 1983 by Arrang et al.,3 has
shown a 93% identity with rat H3-receptor.4 The
H3-receptor can be located both presynaptically and
postsynaptically, acting both as an autoreceptor and as
a heteroreceptor. Histamine H3-receptor antagonists
Keywords: Histamine; H3-receptor antagonists; 2-aminobenzimida-
zole; QSAR.
Abbreviations: Et2O; diethyl ether; EtOH; ethanol; Acet; acetone;
DMSO; dimethyl sulfoxide.
* Corresponding author. Tel.: +39-0521-905062; fax: +39-0521-
905006; e-mail: marco.mor@unipr.it
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.11.030

664
M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
imidazole25, Chart 2), we observed that a moderate
increase in basicity led to an improvement in affinity for
compounds with longer chain spacers, thus revealing a
dependence of the optimal length of the alkyl chain on
the polar group basicity. These observations also indi-
cated that an intermediate basicity of the polar fragment
could yield H3-antagonists with high affinity values,
while maintaining a significant unionized fraction at
physiological pH, useful for drug distribution.
On the basis of these considerations, we designed and
synthesized the series of H3-antagonists reported in this
work, choosing the 2-aminobenzimidazole fragment as a
moderately basic polar group, also including in its struc-
ture a lipophilic ending domain. Two series of 5(6)-sub-
stituted 2-[o-[4(5)-imidazolyl]alkyl]amino]benzimidazole
derivatives (Tables 1 and 2), characterized by a di- or a
tri-methylene alkyl chain, respectively, were thus
designed, synthesized and tested. Physico-chemical prop-
erties, which could be relevant for receptor affinity, that
is, log P and pKa, were also measured and these properties
were employed for QSAR and quantitative structure–
property relationship (QSPR) analysis.
Chart 1. Classical histamine H3-receptor antagonists.
heterocyclic nucleus (Chart 2), such as oxadiazole (5),20
pyridine (6),21 thiazole or benzothiazole (7),22,23 or
imidazole (8).24,25 More recently, several classes of non
classical H3-antagonists, lacking the imidazole ring,
have also been described in the literature.26À28
2. Design
We applied an experimental design to modulate the
basicity of the new compounds by introducing, on the
5(6) position of the benzimidazole nucleus, substituents
with different electronic features. The pKa values of a
series of 5(6)-substituted 2-aminobenzimidazoles had
been shown to be well correlated to the average of sm
and sp (referred to as smÀp henceforth), accounting for
benzimidazole tautomerism,33 which was then employed
as a descriptor of substituent electronic effect.
In spite of the large number of known potent classical
H3-antagonists, the role of the polar group still has to
be fully understood. In fact, H3-antagonists such as
ciproxifan29 and imoproxifan30 (Chart 1), having a
phenolic ether fragment, endowed with high lipophili-
city and low propensity to accept hydrogen bonds, have
displayed high affinity for the H3-receptor and high in
vivo potency. Moreover, the alkyne derivative ciprali-
sant31 and some H3-antagonists recently described are
devoid of a polar group.32
In order to build QSAR models by multiple regression
analysis (MRA), the variation in the electronic proper-
ties had to be as non-correlated as possible to that in the
lipophilic and steric ones. The substituent selection was
therefore made starting from a factorial design on a
variable space defined by the physico-chemical para-
meters smÀp, p and MR, taken from the list of Skagerberg
et al.34 An arbitrary pre-selection was based on the
supposed synthetic accessibility of the final compounds.
Referring to the classical antagonists, we started from
the hypothesis that they could be grouped into two
classes, depending on polar group basicity: compounds
having a neutral polar group (e.g., thioperamide), and
those having a strongly basic one, mainly positively
charged at physiological pH (e.g., clobenpropit). In
principle, a strong basicity should be detrimental for in
vivo potency, because of reduced access to the CNS.
Nevertheless, during our past investigations on
H3-receptor antagonists having different heterocycles as
polar groups (e.g., 2-alkylthioimidazole22,24 and 2-alkyl-
The first set of 8 substituents (a–h in Table 1 and Table
2) allowed for a good exploration of the property space,
although some correlation between p and MR (r=0.44)
resulted from the fact that the experimental domain was
limited.
Variable ranges were as follows: smÀp, À0.41 (NH2) to
0.75 (NO2); p, À1.27 (CONHCH3) to 2.13 (n-C4H9);
MR, 1.03 (H) to 19.69 (n-C4H9). Correlation coefficients
were as follows: p-smÀp,-0.02; p-MR, 0.44; smÀp-MR,
0.15.
To further explore the effect of the electronic properties
on receptor affinity, new compounds were synthesized,
having small electron-donor (CH3) or withdrawing (CF3)
groups, or a methylenedioxy fragment (O–CH2–O).
Chart 2. Histamine H3-receptor antagonists with heterocyclic polar
groups.

M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
665
3. Chemistry
(o-aminoalkyl)imidazole43 to give the desired products
4a–g, 4i–k, 5a–g and 5i–k. The final compounds 4h and
5h were prepared by reduction of the nitro derivatives
4c and 5c, respectively. The carboxyethyl substituent on
the intermediate 2f was hydrolized to COOH during the
oxidation, and it was treated with abs EtOH saturated
with gaseous HCl to obtain 3f.
The 2-aminobenzimidazole derivatives described in this
paper were prepared following a general synthetic route
represented in Scheme 1.
The 4-substituted o-phenylendiamines (1), commercially
available or prepared according to different literature
methods,35À41 were cyclized to 5(6)-substituted 2-mer-
captobenzimidazoles (2) (intermediates 2a–d and 2i were
purchased). 5(6)-Substituted 2-benzimidazolesulphonic
acids (3) were synthesized, according to Sorba,42 by
oxidation from the 2-mercaptobenzimidazole deriva-
tives 2, and then condensed with the appropriate 4(5)-
4. Pharmacology
H3-Receptor affinity of the newly synthesized com-
pounds was measured by displacement of [3H]-(R)-a-
methylhistamine ([3H]RAMHA) bound to rat cerebral
Table 1. Histamine H3-receptor affinity (pKi), antagonist potency (pKB), partition and distribution coefficients (log P and log D7.4) and dissociation
constants (pKa) of compounds 4a–k
b
d
d
Compd
R
pKia
pKB
log Pc
log D7.4
pKa,1
pKa,2
4a
4b
4c
4d
4e
4f
4g
4h
4i
H
Cl
NO2
OCH3
n-C4H9
COOC2H5
CONHCH3
7.25Æ0.05e
7.82Æ0.07
7.73Æ0.06
7.78Æ0.06
7.24Æ0.16
7.24Æ0.12
7.10Æ0.16
7.30Æ0.14
7.16Æ0.10
7.83Æ0.05
6.91Æ0.06
7.33Æ0.20f
7.98Æ0.14
8.42Æ0.24
8.05Æ0.17
7.84Æ0.25
7.41Æ0.05
7.68Æ0.20
7.71Æ0.22
7.42Æ0.15
7.32Æ0.09
7.47Æ0.12
1.49Æ0.02
2.57Æ0.04
1.83Æ0.01
1.43Æ0.01
3.64Æ0.03
2.39Æ0.07
0.48Æ0.01
0.01Æ0.03
1.93Æ0.01
3.01Æ0.01
1.44Æ0.04
1.09Æ0.01
2.31Æ0.04
1.81Æ0.01
1.02Æ0.01
3.30Æ0.07
2.28Æ0.03
0.34Æ0.01
À0.56Æ0.01
1.49Æ0.01
2.82Æ0.01
1.12Æ0.01
6.20
5.76
4.76
6.20
6.27
5.50
5.67
6.54
6.27
5.49
6.19
7.40
7.16
6.96
7.42
7.57
7.09
7.20
7.75
7.54
7.07
7.40
g
NH2
CH3
CF3
4j
4k
O–CH2–O
a Inhibition of [3H]RAMHA binding to rat brain membranes.
b Inhibition of RAMHA-induced effects on guinea-pig isolated ileum. pKB Values obtained according to Furchgott’s method.55
c For the neutral species.
d The values of pKa,1 were mainly attributed to the protonation of the 2-aminobenzimidazole nucleus (see text); s.d. was around 0.01 (n=3).
e pKi versus [3H]NAMHA=7.14Æ0.05.22
f pA2 Value.23
g For the amino group pKa,3=4.16.
Table 2. Histamine H3-receptor affinity (pKi), antagonist potency (pKB), partition and distribution coefficients (log P and log D7.4) and dissociation
constants (pKa) of compounds 5a–k
log Pc
log D7.4
pKa,1
pKa,2
b
d
d
Compd
R
pKia
pKB
5a
5b
5c
5d
5e
5f
5g
5h
5i
H
Cl
NO2
OCH3
n-C4H9
COOC2H5
CONHCH3
8.90Æ0.05
8.95Æ0.03
8.68Æ0.12
9.37Æ0.28
8.06Æ0.18
8.31Æ0.18
7.57Æ0.19
7.33Æ0.09
8.52Æ0.06
8.56Æ0.13
8.46Æ0.17
9.46Æ0.57
8.53Æ0.11
8.32Æ0.09
8.34Æ0.16
7.38Æ0.12
7.48Æ0.12
8.53Æ0.49
7.91Æ0.15
8.18Æ0.08
8.39Æ0.14
8.19Æ0.13
1.78Æ0.01
2.84Æ0.08
2.22Æ0.01
1.77Æ0.01
3.93e
2.64Æ0.03
0.80Æ0.01
0.17Æ0.01
2.26Æ0.01
3.31Æ0.01
1.89Æ0.10
1.11Æ0.01
2.62Æ0.08
2.09Æ0.02
1.14Æ0.04
N.A.f
2.43Æ0.02
0.47Æ0.01
À0.64Æ0.04
1.56Æ0.01
3.03Æ0.03
1.21Æ0.01
6.70
6.20
5.10
6.65
6.70
5.94
6.04
6.82
6.76
5.92
6.68
7.65
7.44
7.34
7.65
7.78
7.40
7.41
7.85
7.75
7.41
7.67
g
NH2
CH3
CF3
5j
5k
O–CH2–O
a Inhibition of [3H]RAMHA binding to rat brain membranes.
b Inhibition of RAMHA-induced effects on guinea-pig isolated ileum. pKB Values obtained according to Furchgott’s method.55
c For the neutral species.
d The values of pKa,1 were mainly attributed to the protonation of the 2-aminobenzimidazole nucleus (see text); s.d. was around 0.01 (n=3).
e Estimated by adding the contribution of n-C4H9 in series 4 to the log P of 5a.
f N.A., Not Available.
g For the amino group pKa,3=4.27.

666
M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
Figure 1. Effect of (R)-a-methylhistamine (RAMHA), clobenpropit
and 5d on specific [35S]GTPgS binding to rat cerebral membranes.
Error bars represent standard errors of the means.
to smÀp. The compounds described in the present work
have two basic groups, the imidazole ring and the 2-
aminobenzimidazole nucleus. The experimentally deter-
mined (macroscopic) constants, pKa,i, are a composite
function of the microscopic ones, but tend to approx-
imate them when the two ionizable centers have sizeable
differences in basicity.45
Scheme 1. Reagents: (a) CS2, EtOH, KOH or C2H5OCS2K, EtOH;
(b) 10% NaOH, 30% H2O2 or 10% NaOH, KMnO4. Note: Com-
pounds 4h and 5h were prepared by reduction of the nitro derivatives
4c and 5c, respectively.
cortex synaptosomes. Histamine H3-receptor antagonist
potency was evaluated on electrically stimulated guinea-
pig ileum, by inhibition of RAMHA-induced responses.44
Titration curves for the compounds tested revealed two
sites of protonation (three for the NH2 derivatives 4h
and 5h, see Table 1 and 2). Although complex equilibria
exist in solution, the pKa,1 was mainly attributed to the
protonation of the 2-aminobenzimidazole fragment, as
it is strongly influenced by the electronic effect of the
substituent in the 5(6) position, and because its values
(4.7<pKa,1<6.9) approach those observed by Serafino
et al.33 The best Hammett equation was obtained
employing sm-p for compounds 4a–k (eq 1), while for
the longer chain derivatives 5a–k statistically equivalent
models could be obtained from smÀp (eq 2) and sm (not
reported). The average of sm and sp seems therefore to
encode a sufficiently accurate balance of inductive and
resonance effects to describe the influence on protona-
tion constants of the 2-aminobenzimidazole nucleus.
The ability of compounds 5a–k to influence [35S]GTPgS
binding to rat cerebral cortex synaptosomes was also
tested, to evaluate their intrinsic activity on this tissue.
In the experimental conditions employed, the agonist
RAMHA effectively stimulated [35S]GTPgS binding,
which, conversely, was unaffected by thioperamide and
clobenpropit.
5. Results and discussion
Biological data relating to the affinity (pKi) for rat brain
H3-receptor and to H3-antagonist potency (pKB) on
guinea-pig ileum, for the 2-aminobenzimidazole deriva-
tives having two (4a–k) and three (5a–k) methylene
groups in the chain spacer, are reported in Tables 1 and
2, respectively. In these tables the results of ionization
(pKa) and lipophilicity (log P and log D7.4) measure-
ments are also reported. The compounds belonging to
the more active class, 5a–k, did not significantly stimu-
late, nor inhibited, [35S]GTPgS binding to rat cerebral
cortex synaptosomes at concentrations able to displace
[3H]RAMHA from rat brain H3-receptors (0.01nM–
1mM), thus confirming their antagonist behavior on this
tissue (Fig. 1).
For compounds 4a–k:
pKa;1 ¼ À1:43ðÆ0:11ÞꢀmÀp þ 6:08ðÆ0:04Þ
ð1Þ
ð2Þ
n ¼ 11; r2 ¼ 0:950; s ¼ 0:13; F ¼ 158:4;
q2 ¼ 0:887; SDEP ¼ 0:17
For compounds 5a–k:
pKa;1 ¼ À1:46ðÆ0:16ÞꢀmÀp þ 6:51ðÆ0:06Þ
n ¼ 11; r2 ¼ 0:910; s ¼ 0:17; F ¼ 87:2;
5.1. Ionization, lipophilicity and QSPR
q2 ¼ 0:798; SDEP ¼ 0:23
Investigating the ionization of a series of 5(6)-sub-
stituted 2-aminobenzimidazole derivatives, Serafino et
al.33 reported pKa values in the range 4.5–6.8 for the 2-
aminobenzimidazole moiety, slightly influenced by the
presence of a second basic group, but strictly correlated
The substituent constants employed for the above-men-
tioned eq 1 and 2, and for the regression models repor-
ted below, are shown in Table 3. For common

M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
667
Table 3. Substituent constantsa employed for experimental design, or
in QSAR and QSPR models
by the observation that experimental log D7.4 values
were in good agreement with those calculated from
log P, pKa,1 and pKa,2, according to the formula: log
D=log P- log [1+10 (pKa,2ÀpH)+10 (pKa,1+pKa,2À2pH)]:48
for the available values of the 19 diprotic bases reported
in Tables 1 and 2 (i.e., 4a–g, 4i–k, 5a–d, 5f, 5g and 5i–k),
we observed a mean difference between measured and
calculated log D7.4 of 0.03, and a mean absolute differ-
ence of 0.15.
b
Compd Substituent
p
MR
sm
sp
smÀp
a
b
c
d
e
f
g
h
i
H
Cl
NO2
0.00
0.71
À0.28
À0.02
2.13
1.03
6.03
7.36
7.87
19.69
17.47
14.57
5.42
5.65
5.02
0.00
0.37
0.71
0.12
0.00
0.23
0.78
À0.27
À0.16
0.45
0.000
0.300
0.745
À0.075
À0.120
0.410
OCH3
n-C4H9
COOC2H5
À0.08
0.51
0.37
0.35
CONHCH3 À1.27
0.36
0.355
NH2
CH3
CF3
À1.23
0.56
0.88
À0.05
À0.16
À0.07
0.43
À0.66
À0.17
0.54
À0.410
À0.120
0.485
Moreover, the formation of ion pairs was prevented by
the use of a zwitterionic buffer (see Experimental).
Therefore, the dependence of log P on smÀp should
reflect an interesting, but as yet unexplained, effect of
charge distribution through the 2-aminobenzimidazole
nucleus on its interaction with one or both of the two
solvents employed in partition experiments. In fact,
compounds with electron-withdrawing substituents
appear more lipophilic than expected from the relative p
values.
j
k
c
O-CH2-O
9.00c À0.16c À0.16c À0.160
a Taken or calculated from Ref. 34, unless differently indicated.
b smÀp=(sm+sp)/2.
c From Ref. 47.
substituents, identical values of the substituent con-
stants were reported in the compilation of Skagerberg34
and in that of van de Waterbeemd and Testa;46 the
constants values for the methylenedioxy group (k) were
obtained from the Hansch data bank.47
QSAR. The comparison of the pKi values with the pKB
ones shows that, while all the compounds tested are
potent H3-antagonists, there is poor correlation between
the two series of biological data, at least within the two
classes of compounds (r=0.51 for 4a–k; r=0.41 for 5a–
k). No statistically sound QSAR model was observed,
for the two subsets 4a–k and 5a–k, relating pKB on
guinea-pig ileum to the experimentally derived proper-
ties reported in Tables 1 and 2, or to the substituent
constants reported in Table 3, by linear or multi-linear
regression analysis. For the first subset, this could be
due to limited variation of potency (the standard
deviation of the pKB column being only 0.43), compared
to the uncertainty of the individual pKB values. More-
over, most of the variation in the antagonist potency of
compounds 5a–k is due to the high pKB value of the
parent compound, 5a.
log P Experimental values of the neutral species in the two-
phase n-octanol/water system were correlated to the p
values of the substituents, showing unexpectedly high
residuals (r2=0.921, s=0.31 for 4a–k; r2=0.890,
s=0.33 for 5a–k). Nevertheless, significant improve-
ments in the fitting were obtained by introducing the
electronic constant sm-p, as shown by eqs 3 and 4 (for
4a–k and 5a–k, respectively).
logP ¼ 1:04ðÆ0:03Þꢁ þ 0:82ðÆ0:08ÞꢀmÀp
þ 1:55ðÆ0:03Þ
ð3Þ
n ¼ 11; r2 ¼ 0:995; s ¼ 0:08; F ¼ 812:4;
q2 ¼ 0:990; SDEP ¼ 0:10; rðꢁ=ꢀmÀpÞ ¼ 0:039
On the other hand, more significant structure–activity
relationships were observed for the competitive binding
(pKi) data on rat brain. First of all, we observed that
compounds having three methylene groups in the chain
spacer (5a–k) showed very high affinity for rat brain H3-
receptor and pKi variation spanning two orders of
magnitude. In contrast, the corresponding di-methylene
chain derivatives (4a–k) showed lower pKi values and a
limited influence of substitution on receptor affinity.
logP ¼ 1:06ðÆ0:05Þꢁ þ 0:85ðÆ0:11ÞꢀmÀp
þ 1:86ðÆ0:04Þ
ð4Þ
n ¼ 10; r2 ¼ 0:989; s ¼ 0:11; F ¼ 306:2;
q2 ¼ 0:981; SDEP ¼ 0:10; rðꢁ=ꢀmÀpÞ ¼ 0:278
A QSAR study was undertaken for the series 5a–k,
investigating the correlation between the affinity values
and a multivariate description of physico-chemical
properties and substituent constants. Traditional
descriptors of substituent electronic (sm, sp, smÀp, F, R)
and steric (MR, L, B1, B5, Sb) effects46 were used toge-
ther with the experimental parameters; the square of the
lipophilicity (log P2) was included to consider a second-
degree effect. Given the small number of compounds,
MRA was limited to models with no more than two
variables.
In these equations, the coefficients of the two variables
are practically the same for the two subsets, indicating
an influence of the electronic features of the 5(6) sub-
stituent on the partition of the neutral species which is
independent of chain length. All the residuals were
lower than 0.2, with the exception of 5k, being 0.22. The
effect of ionization can be ruled out, the log P being
measured at high pH values, where a plateau was
reached in the log D/pH profiles, and where the fraction
of ionized species, calculated from the pKa values, was
negligible. The fact that the measured log P values cor-
rectly refer to the unionized species is further confirmed
Surprisingly, no MRA model resulted definitely accep-
table from a statistic point of view. The best one

668
M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
obtained described a parabolic dependence of pKi on
experimental log P (eq 5 and Fig. 2):
However, the hypothesis that the benzimidazole nucleus
can be strongly involved in receptor binding is sup-
ported by the fact that longer chain derivatives (5a–k)
showed both higher affinity than the shorter ones (4a–
k), and higher sensitivity to the effect of 5(6) substitu-
tion, revealing more specific interactions. This could be
due to strict geometrical requirements for simultaneous
interaction at the imidazole and 2-aminobenzimidazole
binding sites at the H3-receptor, which should be ful-
filled by the three-methylene derivatives only.
pKi ¼ 1:51ðÆ0:37ÞlogP À 0:31ðÆ0:09ÞlogP2
þ 6:96ðÆ0:38Þ
ð5Þ
n ¼ 11; r2 ¼ 0:682; s ¼ 0:38; F ¼ 8:6;
q2 ¼ 0:478; SDEP ¼ 0:41
This model seems to indicate that a suitable range of
lipophilicity (around log P 2.40) could be required for
optimal pKi values. No additional steric or electronic
parameter could improve the fitting of the model
obtained with experimental lipophilicity.
6. Conclusions
2-Aminobenzimidazole derivatives represent a new class
of H3-receptor antagonists, having a polar group (i.e.,
the 2-aminobenzimidazole itself) of intermediate basi-
city, compared with those of the classical antagonists
thioperamide and clobenpropit, yet able to strongly
interact with its binding site at rat H3-receptor. The
introduction of substituents at the benzimidazole 5(6)
position exerts the expected influence on compound
basicity, and an interesting effect on lipophilicity, also
dependent on their electronic properties.
The same data were also submitted to PLS analysis:49
starting from a matrix including the variables log P
(expanded to the squared term), pKa,1, sm, sp and MR,
the best q2 (0.45) was observed for a two-latent variable
model. The first latent variable, explaining 60% of pKi
variation, was mainly related to lipophilicity (with a
positive coefficient for log P, and a negative one for its
squared term), and the second one, explaining a further
18% (total r2 was therefore 0.78) was dominated by the
electronic variables. The pseudo coefficients for the ori-
ginal variables, calculated from their weights on the two
latent ones, were:
Compounds with an additional imidazole ring and a
three-methylene chain (5a–k) showed very high affinity
for rat brain H3-receptor, with pKi variations depending
on the properties of the substituents at the 5(6) position
of the benzimidazole moiety. An exploration of the
substituent space led to the 5(6)-methoxy derivative
(5d), having sub-nanomolar affinity for rat brain recep-
tor; a dependence of pKi from experimental lipophilicity
indicated an optimal log P value around 2.40. Although
compounds 5a–k also showed high potency as H3-
antagonists on guinea-pig ileum, this pharmacological
test revealed a structure–activity profile slightly differ-
ent from that observed for the displacement of
[3H]RAMHA from rat brain membranes; this difference
is as yet unexplained. The observed inter-species differ-
ences, recently reported in the literature, in receptor
binding of H3-antagonists could explain these differ-
ences, and indicate that caution is necessary in the
extrapolation of SAR profiles from one species to
another.
pKi ¼ 1:40logP À 0:28logP2 þ 0:14pKa;1 þ 0:07ꢀm
À 0:03ꢀp À 0:04MR þ 4:48
ð6Þ
This model confirmed the marginal influence of the
electronic variables on affinity variation. Even if QSAR
analysis did not lead to conclusive deductions, as a ten-
tative explanation of the SAR observed, we concluded
that the 2-aminobenzimidazole moiety seems to interact
with the receptor in its neutral state, as suggested by the
lack of correlation between pKi and basicity, and that
the influence of the 5(6) substituents on receptor affinity
could be related to their lipophilicity (see Fig. 2), even if
their precise role remains unexplained.
Compounds with shorter chain (4a–k) had lower H3-
receptor affinity and antagonist potency, and showed a
different SAR profile, suggesting non-optimal interac-
tion with the two putative binding regions, one for the
imidazole nucleus and the other for the 2-aminobenzi-
midazole one.
Compound 5a represents a lead structure, endowed with
high affinity for the H3-receptor, which could be con-
sidered as a pharmacological tool, or further optimized
for in vivo administration.
7. Experimental
Melting points were not corrected and were determined
with a Buchi instrument (Tottoli) and with Gallenkamp
Figure 2. pKi versus log P for compounds 5a–k.

M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
669
melting point apparatus. The final compounds were
analyzed for C, H and N. The percentages we found
were within Æ0.4% of the theoretical values. The H
ponding 4-substituted o-phenylendiamine with carbon
disulphide and aqueous potassium hydroxide in 95%
ethanol, following the general method described by Rao.52
5(6)-Trifluoromethyl-2-mercaptobenzimidazole (2j) was
synthesized, according to Ref. 50, using potassium ethyl
xanthogenate.
1
NMR spectra were recorded on a Bruker 300 spectro-
meter (300 MHz); chemical shifts (d scale) are reported
in parts per million (ppm) relative to the central peak of
the solvent. 1H NMR spectra are reported in order:
multiplicity, approximate coupling constant (J value) in
hertz (Hz) and number of protons. Abbreviations are
the following: Im, imidazolyl; Bz, benzimidazolyl. Mass
spectra were recorded using a Finnigan MAT SSQ 710
instrument. Reactions were monitored by TLC, on
Kieselgel 60 F 254 (DC-Alufolien, Merck). Final com-
pounds and intermediates were purified by chromato-
graphy on preparative Gilson MPLC, using a SiO2
column (LiChroprep, Si 60, 25–40 mm, Merck); the elu-
ents were mixtures of CH2Cl2/CH3OH at various
volume ratios. When indicated, the methanolic phase
was saturated with gaseous NH3.
7.2. 5(6)-n-Butyl-2-mercaptobenzimidazole (2e)
ꢀ
1
80%, mp 230–233 C, H NMR (DMSO-d6) d 0.88 (t,
3H, CH3), 1.28 (m, 2H, CH2), 1.53 (m, 2H, CH2), 2.60
(t, 2H, CH2), 6.93 (m, 2H, Bz-H), 7.03 (d, 1H, Bz-H).
7.3. 5(6)-Carboxyethyl-2-mercaptobenzimidazole (2f)
ꢀ
1
80%, mp 281–283 C, H NMR (DMSO-d6) d 1.31 (t,
3H, CH3), 4.40 (m, 2H, CH2), 7.22 (d, J=8.3 Hz, 1H,
Bz-H), 7.66 (s, 1H, Bz-H), 7.76 (d, J=8.3 Hz, 1H, Bz-
H).
Yields, reaction times and characteristic data of the final
compounds are described in Table 4.
7.4. 5(6)-N-Methylcarbamoyl-2-mercaptobenzimidazole
(2g)
80%, 336–340 ꢀC, H NMR (DMSO-d6) d 2.77 (d, 3H,
CH3), 7.16 (d, J=8.2 Hz, 1H, Bz-H), 7.62 (s, 1H, Bz-H),
7.64 (d, J=8.3 Hz, 1H, Bz-H), 8.40 (m, 1H, NH).
1
7.1. General method of preparation of 5(6)-substituted 2-
mercaptobenzimidazole derivatives (2)
5(6)-Substituted 2-mercaptobenzimidazole derivatives
(2e–g, 2j–k50,51) were prepared starting from the corres-
7.5. General method of preparation of 5(6)-substituted
2-benzimidazolesulphonic acid derivatives (3)
5(6)-Substituted 2-benzimidazolesulphonic acid deriva-
tives (3a–b,42 3d,42 3e–g, 3i,42 3j–k, sodium salt for 3c42)
were prepared starting from the corresponding 5(6)-
substituted 2-mercaptobenzimidazole with 10% sodium
hydroxide and 30% hydrogen peroxide (KMnO4 for 3c)
following the methods reported by Sorba.42 The car-
boxyethyl derivative 3f was prepared by treating 5(6)-
carboxy-2-benzimidazolesulphonic acid (12 mmol, 3g)
with abs ethanol (20 mL) saturated with gaseous HCl.
The reaction mixture was stirred at 100 ꢀC for 3 h,
under dry nitrogen. The solvent was then evaporated
under reduced pressure and the crude product crystal-
lized from abs EtOH/Et2O.
Table 4. Yields, reaction times and characteristic data of the final
compounds 4a–k and 5a–k
Compd
Reaction
Time
(min)
Yield
(%)
Crystallization
solvent
mp
(ꢀC)
4aa
4b
4c
4d
4e
4f
4g
4h
4i
60
240
45
240
240
240
240
—
60
60
240
60
240
45
85
240
240
60
65
25
25
26
30
22
22
90d
20
21
18
22
25
25
21
22
22
19
90d
19
17
17
95% EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH/Et2O
abs EtOH
abs EtOH/Acet
abs EtOH
abs EtOH
abs EtOH/Acet
95% EtOH
223–225b
261–264c
267–269c
196–198b (dec.)
204–206b (dec.)
253–255c
260–262b (dec.)
299–302e (dec.)
257–260c
4j
243–246c
4k
5a
5b
5c
5d
5e
5f
5g
5h
5i
247–249c
7.6. 5(6)-n-Butyl-2-benzimidazolesulphonic acid (3e)
196–198f
241–244c
68%, 335–338 ꢀC, H NMR (DMSO-d6) d 0.89 (t, 3H,
1
246–247c
CH3), 1.31 (m, 2H, CH2), 1.59 (m, 2H, CH2), 2.76 (t,
2H, CH2), 7.43 (d, 1H, Bz-H), 7.48 (s, 1H, Bz-H), 7.61
(d, 1H, Bz-H).
196–200g
183–185f
237–240h
abs EtOH/Et2O
abs EtOH/Acet
abs EtOH/Et2O
abs EtOH/Acet
abs EtOH/Acet
194–197i (dec.)
296–298j
—
60
60
260–262c
7.7. 5(6)-Carboxyethyl-2-benzimidazolesulphonic acid (3f)
5j
5k
174–177f
89%, 324–327 ꢀC, H NMR (DMSO-d6) d 1.34 (t, 3H,
1
60
188–191f
CH3), 4.37 (m, 2H, CH2), 7.79 (d, J=8.8 Hz, 1H, Bz-
H), 8.10 (dd, J=8.7 Hz and J=1.6 Hz, 1H, Bz-H), 8.21
(d, 1H, Bz-H).
a From Ref. 24.
Dihydrochloride 1/2H2O.
c Dihydrochloride.
d Reduction from nitroderivatives 4c and 5c.
b
.
e
.
Trihydrochloride 1/2C2H5OH.
f Dioxalate.
7.8. 5(6)-N-Methylcarbamoyl-2-benzimidazolesulphonic
acid (3g)
g Difumarate.
h Oxalate.
ꢀ
1
i
71%, >340 C, H NMR (DMSO-d6) d 2.81 (d, 3H,
CH3), 7.75 (d, J=8.7 Hz, 1H, Bz-H), 8.04 (dd, J=8.7
.
Dihydrochloride H2O.
Trihydrochloride.
j

670
M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
Hz and J=0.9 Hz, 1H, Bz-H), 8.17 (d, 1H, Bz-H), 8.70
(m, 1H, NH).
Hz, 1H, Bz-H), 7.58 (s, 1H, Im-5-H), 9.04 (s, 1H, Im-2-
+
.
H). MS (CI) 262 [M+1] . Anal. (C12H12N5Cl 2HCl) C,
H, N.
7.9. 5(6)-Trifluoromethyl-2-benzimidazolesulphonic acid
(3j)
7.16. 5(6)-Nitro-2-[2-[4(5)-imidazolyl]ethyl]amino]benzi-
.
midazole dihydrochloride (4c 2HCl)
62%, 301–303 ꢀC, 1H NMR (DMSO-d6) d 7.78 (d,
J=8.5 Hz, 1H, Bz-H), 7.87 (d, J=8.5 Hz, 1H, Bz-H),
7.98 (s, 1H, Bz-H).
1H NMR (DMSO-d6) d 3.08 (t, 2H, CH2-Im), 3.80 (t,
2H, CH2-N), 7.58 (d, 1H, Bz-H), 7.60 (s, 1H, Im-5-H),
8.13 (dd, 1H, Bz-H), 8.17 (d, 1H, Bz-H), 9.05 (s, 1H, Im-
+
.
2-H). MS (CI) 273 [M+1] . Anal. (C12H12N6O2 2HCl)
C, H, N.
7.10. 5,6-Methylenedioxy-2-benzimidazolesulphonic acid
(3k)
70%, >340 ꢀC, 1H NMR (DMSO-d6) d 6.17 (s, 2H,
CH2), 7.13 (s, 2H, Bz-H).
7.17. 5(6)-Methoxy-2-[2-[4(5)-imidazolyl]ethyl]amino]-
.
benzimidazole dihydrochloride (4d 2HCl)
1H NMR (DMSO-d6) d 3.04 (t, 2H, CH2-Im), 3.73 (t,
2H, CH2-N), 3.77 (s, 3H, CH3), 6.82 (dd, J=8.7 Hz and
J=2.4 Hz, 1H, Bz-H), 6.94 (d, J=2.4 Hz, 1H, Bz-H),
7.29 (d, J=8.7 Hz, 1H, Bz-H), 7.58 (s, 1H, Im-5-H),
9.05 (s, 1H, Im-2-H). MS (EI) 257 [M+]. Anal.
7.11. 5(6)-Carboxy-2-benzimidazolesulphonic acid
68%, 343–345 ꢀC, 1H NMR (DMSO-d6) d 7.78 (d,
J=8.6 Hz, 1H, Bz-H), 8.10 (dd, J=8.7 Hz and J=1.1
Hz, 1H, Bz-H), 8.21 (d, 1H, Bz-H).
.
.
(C13H15N5O 2HCl 1/2H2O) C, H, N.
7.12. General method of condensation of 5(6)-substituted
2-benzimidazolesulphonic acids with 4(5)-(!-aminoalkyl)-
imidazole (4a,22 4b–g, 4i–k, 5a–g, 5i–k)
7.18. 5(6)-n-Butyl-2-[2-[4(5)-imidazolyl]ethyl]amino]benzi-
.
midazole dihydrochloride (4e 2HCl)
A mixture of the 5(6)-substituted 2-benzimidazolesul-
phonic acid (10 mmol) (3a–g, 3i–k) and the appropriate
4(5)-(o-alkylamino)imidazole (15 mmol) was heated at
130 ꢀC for the period of time reported in Table 4. The
residue was then treated with a saturated aqueous
sodium bicarbonate solution and extracted with ethyl
acetate. The organic layer was dried over Na2SO4 and
evaporated under reduced pressure. The crude products
were purified by column chromatography (SiO2, 4a,5a
CH2Cl2:CH3OH(NH3)=10:1; 4g,5g CH2Cl2:CH3OH-
(NH3)=5:1; 4c,5c CH2Cl2:CH3OH(NH3)=20:1; 4b,4d,
4e,4f,4i,4j,4k,5b,5d,5e,5f,5i,5j,5k CH2Cl2:CH3OH(NH3)
=15:1).
1H NMR (DMSO-d6) d 0.91 (t, 3H, CH3), 1.23–1.36 (m,
2H, CH2), 1.50–1.60 (m, 2H, CH2), 2.50 (t, 2H, CH2),
2.61 (t, 2H, CH2-Im), 3.43 (t, 2H, CH2-N), 6.92 (d,
J=7.9 Hz, 1H, Bz-H), 7.12 (s, 1H, Bz-H), 7.20 (d,
J=7.9 Hz, 1H, Bz-H), 7.22 (s, 1H, Im-5-H), 8.57 (s, 1H,
+
.
Im-2-H). MS (EI) 283 [M ]. Anal. (C16H21N5 2HCl
1/2H2O) C, H, N.
.
7.19. 5(6)-Carboxyethyl-2-[2-[4(5)-imidazolyl]ethyl]amino]-
.
benzimidazole dihydrochloride (4f 2HCl)
1H NMR (DMSO-d6) d 1.33 (t, 3H, CH3), 3.07 (t, 2H,
CH2-Im), 3.82 (t, 2H, CH2-N), 4.32 (m, 2H, CH2),
7.49 (d, J=8.4 Hz, 1H, Bz-H), 7.60 (s, 1H, Im-5-H),
7.86 (dd, J=7.9 Hz and J=1.3 Hz, 1H, Bz-H), 7.94
7.13. Method of preparation of the 5(6)-aminoderivatives
(4h,5h)
(d, J=1.3 Hz, 1H, Bz-H), 9.05 (s, 1H, Im-2-H). MS
+
.
(CI) 300 [M+1] . Anal. (C15H17N5O2 2HCl) C, H,
N.
The 5(6)-aminoderivatives 4h and 5h were obtained by
hydrogenation (50 psi) of the nitro compounds 4c and
5c in Parr apparatus with Pd/C (10%), in abs EtOH, at
room temperature, for 3h.
7.20. 5(6)-N-Methylcarbamoyl-2-[2-[4(5)-imidazolyl]ethyl]-
.
amino]benzimidazole dihydrochloride (4g 2HCl)
1H NMR (DMSO-d6) d 2.78 (s, 3H, CH3), 3.06 (t, 2H,
CH2-Im), 3.80 (t, 2H, CH2-N), 7.43 (d, J=8.3 Hz, 1H,
Bz-H), 7.60 (s, 1H, Im-5-H), 7.75 (dd, J=8.4 Hz and
J=1.5 Hz, 1H, Bz-H), 7.87 (d, J=1.5 Hz, 1H, Bz-H),
9.05 (s, 1H, Im-2-H). MS (CI) 285 [M+1]+. Anal.
7.14. 2-[2-[4(5)-Imidazolyl]ethyl]amino]benzimidazole di-
.
hydrochloride (4a 2HCl)
1H NMR (DMSO-d6) d 3.06 (t, 2H, CH2-Im), 3.78 (m,
2H, CH2-N), 7.19–7.24 (m, 2H, Bz-H), 7.37–7.43 (m,
2H, Bz-H), 7.59 (s, 1H, Im-5-H), 9.06 (s, 1H, Im-2-H),
9.38 (t, 1H, NH). MS (EI) 227 [M+]. Anal.
.
.
(C14H16N6O 2HCl 1/2H2O) C, H, N.
.
.
(C12H13N5 2HCl 1/2H2O) C, H, N.
7.21. 5(6)-Amino-2-[2-[4(5)-imidazolyl]ethyl]amino]benzi-
.
midazole trihydrochloride (4h 3HCl)
7.15. 5(6)-Chloro-2-[2-[4(5)-imidazolyl]ethyl]amino]benzi-
.
midazole dihydrochloride (4b 2HCl)
1H NMR (DMSO-d6) d 3.06 (t, 2H, CH2-Im), 3.77 (m,
2H, CH2-N), 7.17 (dd, 1H, Bz-H), 7.42 (s, 1H, Bz-H),
7.45 (d, 1H, Bz-H), 7.59 (s, 1H, Im-5-H), 9.05 (s, 1H,
Im-2-H), 9.53 (t, 1H, NH). MS (CI) 287 [M+1]+. Anal.
1H NMR (DMSO-d6) d 3.05 (t, 2H, CH2-Im), 3.74 (t,
2H, CH2-N), 7.26 (dd, J=8.6 Hz and J=1.8 Hz, 1H,
Bz-H), 7.39 (d, J=8.6 Hz, 1H, Bz-H), 7.43 (d, J=1.8
.
.
(C13H14N6O2 3HCl 1/2C2H5OH) C, H, N.

M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
671
7.22. 5(6)-Methyl-2-[2-[4(5)-imidazolyl]ethyl]amino]benzi-
.
midazole dihydrochloride (4i 2HCl)
7.29. 5(6)-n-Butyl-2-[3-[4(5)-imidazolyl]propyl]amino]benzi-
.
midazole dioxalate (5e 2C2H2O4)
1H NMR (DMSO-d6) d 2.36 (s, 3H, CH3), 3.05 (t, 2H,
CH2-Im), 3.76 (m, 2H, CH2-N), 7.03 (d, J=8.2 Hz, 1H,
Bz-H), 7.20 (s, 1H, Bz-H), 7.27 (d, J=8.1 Hz, 1H, Bz-
1H NMR (DMSO-d6) d 0.89 (t, 3H, CH3), 1.23–1.35 (m,
2H, CH2), 1.50–1.60 (m, 2H, CH2), 1.89–1.98 (m, 2H,
CH2), 2.63 (t, 2H, CH2), 2.72 (t, 2H, CH2-Im), 3.41 (t, 2H,
CH2-N), 7.00 (dd, 1H, Bz-H), 7.16 (d, 1H, Bz-H), 7.24–
H), 7.59 (s, 1H, Im-5-H), 9.05 (s, 1H, Im-2-H), 9.27 (t, 1H,
+
.
NH). MS (CI) 242 [M+1] . Anal. (C13H15N5 2HCl) C,
H, N.
7.26 (m, 2H, Bz-H, Im-5-H), 8.53 (s, 1H, Im-2-H). MS
+
.
(CI) 298 [M+1] . Anal. (C17H23N5 2C2H2O4) C, H, N.
7.23. 5(6)-Trifluoromethyl-2-[2-[4(5)-imidazolyl]ethyl]-
.
amino]benzimidazole dihydrochloride (4j 2HCl)
7.30. 5(6)-Carboxyethyl-2-[3-[4(5)-imidazolyl]propyl]-
.
amino]benzimidazole oxalate (5f C2H2O4)
1H NMR (DMSO-d6) d 3.07 (t, 2H, CH2-Im), 3.80 (m,
2H, CH2-N), 7.58–7.61 (m, 3H, Bz-H), 7.67 (s, 1H, Im-
1H NMR (DMSO-d6) d 1.31 (t, 3H, CH3), 1.92–1.95 (m,
2H, CH2), 2.72 (t, 2H, CH2-Im), 3.38 (t, 2H, CH2-N), 4.28
(m, 2H, CH2), 7.25 (d, J=8.3 Hz, 1H, Bz-H), 7.36 (s, 1H,
Im-5-H), 7.65 (dd, J=8.3 Hz and J=1.6 Hz, 1H, Bz-H),
5-H), 9.07 (s, 1H, Im-2-H), 9.76 (t, 1H, NH). MS (CI)
+
.
296 [M+1] . Anal. (C13H12N5F3 2HCl) C, H, N.
7.77 (d, J=1.4 Hz, 1H, Bz-H), 8.75 (s, 1H, Im-2-H). MS
+
.
(CI) 314 [M+1] . Anal. (C16H19N5O2 C2H2O4) C, H, N.
7.24. 5,6-Methylenedioxy-2-[2-[4(5)-imidazolyl]ethyl]-
.
amino]benzimidazole dihydrochloride (4k 2HCl)
7.31. 5(6)-N-Methylcarbamoyl-2-[3-[4(5)-imidazolyl]pro-
.
pyl]amino]benzimidazole dihydrochloride (5g 2HCl)
1H NMR (DMSO-d6) d 3.01 (t, 2H, CH2-Im), 3.67 (m,
2H, CH2-N), 6.04 (s, 2H, CH2), 7.01 (s, 2H, Bz-H), 7.55
(s, 1H, Im-5-H), 8.93 (t, 1H, NH), 9.02 (s, 1H, Im-2-H).
+
1H NMR (DMSO-d6) d 1.99–2.03 (m, 2H, CH2), 2.78 (s,
3H, CH3), 2.80 (t, 2H, CH2-Im), 3.30 (t, 2H, CH2-N),
7.41 (d, 1H, Bz-H), 7.46 (s, 1H, Im-5-H), 7.73 (dd, 1H,
.
MS (CI) 272 [M+1] . Anal. (C13H13N5O2 2HCl) C, H,
N.
Bz-H), 7.83 (s, 1H, Bz-H), 8.98 (s, 1H, Im-2-H). MS (CI)
+
.
299 [M+1] . Anal. (C15H18N6O 2HCl H2O) C, H, N.
.
7.25. 2-[3-[4(5)-Imidazolyl]propyl]amino]benzimidazole (5a)
1H NMR (DMSO-d6) d 1.86–1.89 (m, 2H, CH2), 2.57 (t,
2H, CH2-Im), 3.32 (t, 2H, CH2-N), 6.77 (s, 1H, Im-5-
H), 6.85–6.90 (m, 2H, Bz-H), 7.12–7.16 (m, 2H, Bz-H),
7.54 (s, 1H, Im-2-H). MS (EI) 241 [M+]. Anal.
7.32. 5(6)-Amino-2-[3-[4(5)-imidazolyl]propyl]amino]benzi-
.
midazole trihydrochloride (5h 3HCl)
1H NMR (DMSO-d6) d 2.00–2.03 (m, 2H, CH2), 2.80 (t,
2H, CH2-Im), 3.48 (t, 2H, CH2-N), 7.18 (dd, J=8.5 Hz,
1H, Bz-H), 7.43 (d, J=8.5 Hz, 1H, Bz-H), 7.47–7.49 (m,
.
(C13H15N5 2C2H2O4) C, H, N.
2H, Bz-H, Im-5-H), 9.01 (s, 1H, Im-2-H). MS (CI) 257
+
7.26. 5(6)-Chloro-2-[3-[4(5)-imidazolyl]propyl]amino]benzi-
.
midazole dihydrochloride (5b 2HCl)
.
[M+1] . Anal. (C13H16N6 3HCl) C, H, N.
1H NMR (DMSO-d6) d 1.99–2.02 (m, 2H, CH2), 2.80 (t,
2H, CH2-Im), 3.48 (t, 2H, CH2-N), 7.23 (dd, J=8.5 Hz
and J=1.8 Hz, 1H, Bz-H), 7.38 (d, J=8.5 Hz, 1H, Bz-
H), 7.42 (d, J=1.8 Hz, 1H, Bz-H), 7.47 (s, 1H, Im-5-H),
9.01 (s, 1H, Im-2-H). MS (CI) 276 [M+1]+. Anal.
7.33.
benzimidazole dihydrochloride (5i 2HCl)
5(6)-Methyl-2-[3-[4(5)-imidazolyl]propyl]amino]-
.
1H NMR (DMSO-d6) d 1.99–2.03 (m, 2H, CH2), 2.36 (s,
3H, CH3), 2.79 (t, 2H, CH2-Im), 3.47 (m, 2H, CH2-N),
7.01 (dd, 1H, Bz-H), 7.19 (s, 1H, Im-5-H), 7.26 (d, 1H,
Bz-H), 7.47 (d, 1H, Bz-H), 9.01 (s, 1H, Im-2-H), 9.27 (t,
1H, NH). MS (CI) 256 [M+1]+. Anal.
.
(C13H14N5Cl 2HCl) C, H, N.
7.27. 5(6)-Nitro-2-[3-[4(5)-imidazolyl]propyl]amino]benzi-
.
midazole dihydrochloride (5c 2HCl)
.
(C14H17N5 2HCl) C, H, N.
1H NMR (DMSO-d6) d 1.99–2.02 (m, 2H, CH2), 2.79 (t,
2H, CH2-Im), 3.49 (t, 2H, CH2-N), 7.49 (d, 1H, Bz-H),
7.52 (s, 1H, Im-5-H), 8.10 (d, 1H, Bz-H), 8.13 (s, 1H,
Bz-H), 9.00 (s, 1H, Im-2-H). MS (EI) 286 [M+]. Anal.
7.34. 5(6)-Trifluoromethyl-2-[3-[4(5)-imidazolyl]propyl]-
.
amino]benzimidazole dioxalate (5j 2C2H2O4)
1H NMR (DMSO-d6) d 1.90–1.94 (m, 2H, CH2), 2.72 (t,
2H, CH2-Im), 3.36 (t, 2H, CH2-N), 7.25 (dd, 1H, Bz-H),
7.32 (d, 1H, Bz-H), 7.37 (d, 1H, Bz-H), 7.46 (s, 1H, Im-
5-H), 8.80 (s, 1H, Im-2-H). MS (CI) 310 [M+1]+. Anal.
.
(C13H14N6O2 2HCl) C, H, N.
7.28. 5(6)-Methoxy-2-[3-[4(5)-imidazolyl]propyl]amino]-
benzimidazole (5d)
.
(C14H14N5F3 2C2H2O4) C, H, N.
1H NMR (DMSO-d6) d 1.86–1.89 (m, 2H, CH2), 2.58 (t,
2H, CH2-Im), 3.30 (t, 2H, CH2-N), 3.70 (s, 3H, CH3),
6.52 (dd, J=8.1 Hz and J=2.2 Hz, 1H, Bz-H), 6.76 (d,
J=2.2 Hz, 1H, Bz-H), 6.81 (s, 1H, Im-5-H), 7.04 (d,
7.35. 5,6-Methylenedioxy-2-[3-[4(5)-imidazolyl]propyl]-
.
amino]benzimidazole dioxalate (5k 2C2H2O4)
1H NMR (DMSO-d6) d 1.90–1.94 (m, 2H, CH2), 2.72 (t,
2H, CH2-Im), 3.32 (m, 2H, CH2-N), 5.97 (s, 2H, CH2),
6.96 (s, 2H, Bz-H), 7.30 (s, 1H, Im-5-H), 8.72 (s, 1H,
J=8.5 Hz, 1H, Bz-H), 7.65 (s, 1H, Im-2-H). MS (EI)
+
.
271 [M ]. Anal. (C14H17N5O 2C4H4O4) C, H, N.

672
M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
+
.
Im-2-H). MS (EI) 285 [M ]. Anal. (C14H15N5O2 2-
C2H2O4) C, H, N.
electrical pulses were delivered to the tissue (0.1 Hz, 1
ms, 1.5–3.0 V) from a stimulator (LACE Elettronica
model ES-3, Ospedaletto PI, Italy). Cumulative con-
centration-response curves for the inhibition of elec-
trically stimulated contractions were determined for the
H3 selective agonist RAMHA (1 nM–1mM). The tissues
were allowed to equilibrate with the compounds under
study (1 nM–10mM) for 30 min before the generation of
concentration-response curves to the agonist. pKB
Values (‘apparent pA2’) were determined according to
Furchgott’s equation.55
7.36. Lipophilicity
log P And log D7.4 values were determined by the shake-
flask technique. The two-phase system consisted of
mutually saturated n-octanol and aqueous buffers
0.05M CAPS (3-cyclohexylamino-1-propanesulfonic
acid) pH 11 for log P and 0.05M MOPS [3-(N-mor-
pholino)propanesulfonic acid] pH 7.4 for log D. Ionic
strength was adjusted at 0.15M with KCl.
7.40. [35S]GTPꢀS binding assay
After partitioning, samples were centrifuged at 3000
rpm for 10 min, each of the two phases was manually
separated and diluted with analytical grade CH3OH
before injection in HPLC apparatus equipped with UV
detector (Gilson); the volume injected was 20 mL per
run. The column was a Spherisorb CN-S10, 250Â4.6
mm (Waters). Mobile phases were mixtures of analytical
CH3CN and KH2PO4 0.01 M (pH between 5.0 and 7.4).
log P And log D7.4 values were calculated from the ratio
of the mean peak areas in the two phases, correcting for
instrumental attenuation and dilution; they are the
means of at least four different measurements.
[35S]GTPgS binding assays were performed according to
Nickel et al.56 with modifications. Cerebral cortex and
striatum from female Wistar rats (150–200g) were
homogenized (Potter–Elvehjem) in 6 volumes of ice-
cold membrane buffer (Tris–HCl 50 mM, MgCl2 3 mM,
EGTA 1 mM, pH 7.4) and centrifuged twice at 14,500
rpm for 15 min. The final pellet, resuspended and
homogenized in 6 volumes of assay buffer (Tris–HCl 50
mM, MgCl2 3 mM, EGTA 0.2 mM, NaCl 100 mM, pH
7.7), was then preincubated for 10 min at 30 ꢀC with
adenosine deaminase (0.6 U/mL) to remove endogenous
adenosine.57,58 Membranes in aliquots of 0.5 mL (50 mg
protein) were incubated for 60 min at 30 ꢀC with
0.05nM [35S]GTPgS and, when required, with the dif-
ferent drugs tested, in a final volume of 1 mL of assay
buffer containing 100mM GDP. The non specific bind-
ing was determined using GTPgS (20mM). Incubations
were terminated by rapid filtration through Whatman
GF/B glass filters. Filters were washed three times with
3 mL of ice-cold Tris–HCl 50 mM buffer, pH 7.7, and
bound radioactivity was measured by liquid scintillation
spectrometry. Protein concentration was assayed apply-
ing Bradford’s method.59 Following incubation with
0.05 nM [35S]GTPgS, specific binding to rat brain
membranes gave 1500–3000 d.p.m.; non specific binding
was 15–20% of total binding. Specific binding was sig-
nificantly increased (about 27%) by RAMHA
(EC50=34.67 nM, maximal effect at 300 nM), which
had been reported to behave as a full agonist in H3-
receptor mediated stimulation of [35S]GTPgS binding in
cerebral cortex rat membranes.60 The drugs under study
were used at concentrations ranging from 0.01 nM to 1
mM. The H3-antagonists thioperamide and clobenpropit
did not significantly affect [35S]GTPgS binding in the
range 0.01–100 nM, while they abolished the effect of
100 nM RAMHA at the concentrations of 100 and 10
nM, respectively.
7.37. Dissociation constants
pKa Values were measured using an automatic poten-
tiometric apparatus with automatic burette addition of
KOH of known normality (0.1 N). 0.04–0.05 mmol Of
the compound (salt) were dissolved in CO2-free water
inside a jacketed vessel, keeping the temperature at
25Æ0.1 ꢀC. Air was displaced with a slow stream of N2.
Ionic strength was kept fixed at 0.15 M by the addition
of KCl. The standard potential E0 and the pKw values
were determined from titrations of fixed volumes of HCl
of known concentration with KOH, determining the
neutralization point with the Gran method. The dis-
sociation constant values are the means of at least three
different titrations with a mean s.d.=Æ0.01 and were
refined by employing the SUPERQUAD program.53
7.38. Pharmacology. Binding assays
Rat (Wistar) brain membranes were incubated for 30
min with [3H]RAMHA 0.5 nM and the compounds
under study (1 nM–10mM), in Tris–HCl 50 mM, pH 7.4,
NaCl 50 mM, EDTA 0.5 mM, then rapidly filtered
(AAWP0.8 mM) under vacuum and rinsed with ice-cold
buffer. Specific binding was defined as the binding
inhibited by thioperamide 10mM, and the IC50 values
were estimated from the displacement curves of the
compounds tested versus [3H]RAMHA bound to cere-
bral membranes.22 pKi Values were calculated according
to Cheng and Prusoff’s equation.54
7.41. QSAR
Multiple regression analysis calculations were per-
formed with an Excel (Microsoft Co., version 97)
spreadsheet, employing the built-in statistical functions
and automated macro procedures. For each subset of
compounds, and for the dependent variables employed,
MRA models were calculated with one or two variables
simultaneously included; all the possible combinations
were tested. Partial least squares (PLS) analysis was
performed, after scaling of the variables to unit
7.39. Functional assays
Portions of guinea-pig ileum were mounted on a coaxial
platinum electrode assembly in a 10 mL water-jacketed
organ-bath containing Krebs–Henseleit solution aerated
with 95%O2:5%CO2 and maintained at 37 ꢀC. Single

M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
673
Elemental analysis data (found data within Æ0.4% of theoretical values)
Compd.
C calc.
C found
H calc.
H found
N calc.
N found
4a
4b
4c
4d
4e
4f
4g
4h
4i
C=46.61
C=43.07
C=41.75
C=46.02
C=52.60
C=48.39
C=45.91
C=41.67
C=49.69
C=42.40
C=45.36
C=48.46
C=44.78
C=43.46
C=52.48
C=52.82
C=53.59
C=46.28
C=42.69
C=51.22
C=44.17
C=46.45
C=46.57
C=43.27
C=41.80
C=46.28
C=52.83
C=48.36
C=45.71
C=41.34
C=49.73
C=42.45
C=45.10
C=48.50
C=44.56
C=43.10
C=52.48
C=52.72
C=53.66
C=46.66
C=43.00
C=51.12
C=44.12
C=46.40
H=5.21
H=4.22
H=4.09
H=5.35
H=6.62
H=5.14
H=5.23
H=5.38
H=5.45
H=3.83
H=4.39
H=4.55
H=4.63
H=4.49
H=5.00
H=5.70
H=5.24
H=5.70
H=5.24
H=5.83
H=3.71
H=4.12
H=4.88
H=4.18
H=4.14
H=5.54
H=6.79
H=5.22
H=5.24
H=5.02
H=5.52
H=3.82
H=4.39
H=4.93
H=4.62
H=4.60
H=5.13
H=5.83
H=5.32
H=5.96
H=5.51
H=6.01
H=3.85
H=4.20
N=22.65
N=20.93
N=24.35
N=20.65
N=19.17
N=18.81
N=22.95
N=22.43
N=22.29
N=19.02
N=20.35
N=16.62
N=20.08
N=23.40
N=13.91
N=14.67
N=17.36
N=21.59
N=22.98
N=21.34
N=14.31
N=15.05
N=22.80
N=20.84
N=24.23
N=20.28
N=19.08
N=18.59
N=22.84
N=22.76
N=22.12
N=18.98
N=20.10
N=16.85
N=19.84
N=23.07
N=13.60
N=14.31
N=17.09
N=21.34
N=22.59
N=21.20
N=14.34
N=14.79
4j
4k
5a
5b
5c
5d
5e
5f
5g
5h
5i
5j
5k
variance, with the program Simca 6.0;61 the lipophilicity
variable (log P) was squared, in order to include para-
bolic effects. Standard deviation of the errors in predic-
tion (SDEP), and the relative predictivity parameter, q2,
were calculated by cross-validation, omitting one com-
pound at a time from the set, according to the leave-
one-out technique (LOO).62
10. Schlicker, E.; Betz, R.; Gothert, M. Arch. Pharmacol.
1988, 337, 588.
11. Philippu, A.; Prast, H. Behav. Brain Res. 2001, 124, 151.
12. Baciottini, L.; Passani, M. B.; Mannaioni, P. F.; Blan-
dina, P. Behav. Brain Res. 2001, 124, 183.
13. Lipp, O. R.; Stark, H.; Schunack, W. Recept. Biochem.
Methodol. 1992, 16, 57.
14. Arrang, J.-M.; Garbarg, M.; Lancelot, J. C.; Lecomte,
J. M.; Pollard, H.; Robba, M.; Schunack, W.; Schwartz,
J.-C. Nature 1987, 327, 117.
Acknowledgements
15. Van der Goot, H.; Schepers, M. J. P.; Sterk, G. J.; Tim-
merman, H. Eur. J. Med. Chem. 1992, 27, 511.
16. Jansen, F. P.; Wu, T. S.; Voss, H.-P.; Steinbusch,
H. W. M.; Vollinga, R. C.; Rademaker, B.; Bast, A.;
Timmerman, H. Br. J. Pharmacol. 1994, 113, 355.
17. Stark, H.; Krause, M.; Arrang, J.-M.; Ligneau, X.;
Schwartz, J.-C.; Schunack, W. Bioorg. Med. Chem. Lett.
1994, 4, 2907.
Financial support from Italian MIUR is gratefully
acknowledged. We are grateful to the Centro Inter-
facolta Misure of the University of Parma for providing
the NMR and MS instrumentation.
18. Reidemeister, S.; Stark, S.; Ligneau, X.; Ganellin, C. R.;
Schwartz, J.-C.; Schunack, W. Pharmazie 2000, 55, 83.
19. Tozer, M. J.; Buck, I. M.; Cooke, T.; Kalindjian, S. B.;
Pether, M. J.; Steel, K. I. Bioorg. Med. Chem. Lett. 2002,
10, 425.
20. Clitherow, J. W.; Beswick, P.; Irving, W. J.; Scopes,
D. I. C.; Barnes, J. C.; Clapham, J.; Brown, J. D.; Evans,
D. J.; Hayes, A. G. Bioorg. Med. Chem. Lett. 1996, 6,
833.
21. Ganellin, C. R.; Hosseini, S. K.; Khalaf, Y. S.; Tertiuk,
W.; Arrang, J.-M.; Garbarg, M.; Ligneau, X.; Schwartz,
J.-C. J. Med. Chem. 1995, 38, 3342.
22. Plazzi, P. V.; Bordi, F.; Mor, M.; Silva, C.; Morini, G.;
Caretta, A.; Barocelli, E.; Vitali, T. Eur. J. Med. Chem.
1995, 30, 881.
23. Barocelli, E.; Ballabeni, V.; Caretta, A.; Chiavarini, M.;
Bordi, F.; Mor, M.; Impicciatore, M. Pharm. Sci. 1995, 1,
383.
24. Mor, M.; Bordi, F.; Silva, C.; Rivara, S.; Crivori, P.;
Plazzi, P. V.; Ballabeni, V.; Caretta, A.; Barocelli, E.;
Impicciatore, M.; Carrupt, P.-A.; Testa, B. J. Med. Chem.
1997, 40, 2571.
References and notes
1. Hill, S. J.; Ganellin, C. R.; Timmerman, H.; Schwartz, J.-
C.; Shankley, N. P.; Young, J. M.; Schunack, W.; Levi,
R.; Haas, H. L. Pharmacol. Rev. 1997, 49, 253.
2. Nakamura, T.; Itadani, H.; Hidaka, Y.; Ohta, M.; Tanaka,
K. Biochem. Biophys. Res. Commun. 2000, 279, 615.
3. Arrang, J.-M.; Garbarg, M.; Schwartz, J.-C. Nature
(London) 1983, 302, 832.
4. Lovenberg, T. W.; Pyaty, J.; Chang, H.; Wilson, S. J.;
Erlander, M. G. J. Pharm. Exp. Ther. 2000, 293, 771.
5. Arrang, J.-M.; Garbarg, M.; Schwartz, J.-C. Neuroscience
1985, 15, 553.
6. Arrang, J.-M.; Deveaux, B.; Chodkiewicz, J.-P.;
Schwartz, J.-C. J. Neurochem. 1988, 51, 105.
7. Molderings, G. J.; Weibenborn, G.; Schlicker, E.;
Likungu, J.; Gothert, M. Naunyn-Schmiedeberg’s Arch.
Pharmacol. 1992, 346, 46.
8. Clapham, J.; Kilpatrick, G. J. Br. J. Pharmacol. 1992,
107, 919.
9. Schlicker, E.; Fink, K.; Detzner, J.; Gothert, M. J.
Neural. Transm. 1993, 93, 1.
25. Mor, M.; Bordi, F.; Silva, C.; Rivara, S.; Zuliani, V.;

674
M. Mor et al. / Bioorg. Med. Chem. 12 (2004) 663–674
Vacondio, F.; Morini, G.; Barocelli, E.; Ballabeni, V.;
Impicciatore, M.; Plazzi, P. V. Farmaco 2000, 55, 27.
43. Vollinga, R. C.; Menge, W. M. P. B.; Timmerman, H.
Recl. Trav. Chim. Pays-Bas 1993, 112, 123.
26. Miko, T.; Ligneau, X.; Pertz, H. H.; Ganellin, C. R.;
Arrang, J.-M.; Schwartz, J.-C.; Schunack, W.; Stark, H.
J. Med. Chem. 2003, 46, 1523.
27. Turner, S. C.; Esbenshade, T. A.; Bennani, Y. L.;
Hancock, A. A. Bioorg. Med. Chem. Lett. 2003, 13,
2131.
28. Apodaca, R.; Dvorak, C. A.; Xiao, W.; Barbier, A. J.;
Boggs, J. D.; Wilson, S. J.; Lovenberg, T. W.; Carruthers,
N. I. J. Med. Chem. 2003, 46, 3938.
29. Stark, H.; Sadek, B.; Krause, M.; Huls, A.; Ligneau, X.;
Ganellin, C. R.; Arrang, J.-M.; Schwartz, J.-C.; Schu-
nack, W. J. Med. Chem. 2000, 43, 3987.
30. Sasse, A.; Sadek, B.; Ligneau, X.; Elz, S.; Pertz, H. H.;
Luger, P.; Ganellin, C. R.; Arrang, J.-M.; Schwartz,
J.-C.; Schunack, W.; Stark, H. J. Med. Chem. 2000, 43,
3335.
31. Ali, S. A.; Tedford, C. E.; Gregory, R.; Handley, M. H.;
Yates, S. L.; Hirth, W. W.; Phillips, J. G. J. Med. Chem.
1999, 42, 903.
32. Stark, H.; Ligneau, X.; Arrang, J.-M. .; Schwartz, J.-C.;
Schunack, W. Bioorg. Med. Chem. Lett. 1998, 8, 2011.
33. Serafino, A.; Sorba, G.; Daniele, P. G. Annali di Chimica
1989, 79, 81.
34. Skagerberg, B.; Bonelli, D.; Clementi, S.; Cruciani, G.;
Ebert, C. Quant. Struct.-Act. Relat. 1989, 8, 32.
35. Tomioka, H.; Ichikawa, N.; Komatsu, K. J. Am. Chem.
Soc. 1992, 114, 8045.
44. Barocelli, E.; Ballabeni, V.; Caretta, A.; Bordi, F.; Silva, C.;
Morini, G.; Impicciatore, M. Agents Actions 1993, 38, 158.
45. Perrin, D.D.; Dempsey, B.; Serjeant, E.P. In: pKa Predic-
tion for Organic Acids and Bases; Chapman and Hall:
London and New York, 1981; pp 17–18.
46. van de Waterbeemd, H; Testa, B. In: Advances in Drug
Research; Testa, B. (Ed.), Academic Press: London, 1987;
Vol. 16, pp 85–225.
47. Hansch, C.; Leo, A.; Hoekman, D. In: Exploring QSAR -
Hydrophobic, Electronic and Steric Constants; American
Chemical Society: Washington, DC, 1995.
48. Avdeef, A. In: Lipophilicity in Drug Action and Toxicol-
ogy; Pliska, V., Testa, B., van de Waterbeemd, H. (Eds.),
VCH: Weinheim, 1996; pp 109–139.
49. Wold S. In: Chemometric Methods in Molecular Design;
van de Waaterbeemd, H. (Ed.), VCH Publishers Inc.:
New York, 1995; pp 195–218.
50. Uchida, M.; Morita, S.; Chihiro, M.; Kanbe, T.; Yama-
saki, K.; Yabuuchi, Y.; Nakagawa, K. Chem. Pharm.
Bull. 1989, 37, 1517.
51. Lavagnino, E. R.; Thompson, D. C. J. Heterocycl. Chem.
1972, 9, 149.
52. Rao, K. V. B.; Charles, E. S.; Sharma, S.; Iyer, R. N.;
Gupta, S.; Katiyar, C. Eur. J. Med. Chem.-Chim. Ther.
1981, 16, 35.
53. Gans, P.; Sabatini, A.; Vacca, A. J. Chem. Soc. Dalton
Trans. 1985, 6, 1195.
36. Clark, R. L.; Pessolano, A. A. J. Am. Chem. Soc. 1958,
80, 1657.
54. Cheng, Y. C.; Prusoff, W. H. Biochem. Pharmacol. 1973,
22, 3099.
37. Kumar, S.; Seth, M.; Bhaduri, A. P.; Visen, P. K. S.;
Misra, A.; Gupta, S.; Fatima, N.; Katiyar, J. C.; Chat-
terjce, R. K.; Sen, A. B. J. Med. Chem. 1984, 27, 1083.
38. Silk, J. A. J. Chem. Soc., part II 1956, 2058.
39. Keana, J. F. W.; Kher, S. M.; Cai, S. X.; Dinsmore, C. M.;
Glenn, A. G.; Guastella, J.; Huang, J.-C.; Ilyin, V.; Lu,
Y.; Mouser, P. L.; Woodward, R. M.; Weber, E. J. Med.
Chem. 1995, 38, 4367.
40. Cai, S. X.; Kher, S. M.; Zhou, Z.-L.; Ilyin, V.; Espitias,
S. A.; Tran, M.; Hawkinson, J. E.; Woodward, R. M.;
Weber, E.; Keana, F. W. J. Med. Chem. 1997, 40, 730.
41. Goker, H.; Ertan, R. Arch. Pharm. (Weinheim) 1991, 324,
283.
55. Furchgott, R.F. In: Handbook of Experimental Pharma-
cology; Blanschko, E., Muscholl, E. (Eds), Springer: Ber-
lin, 1972; Vol. 33, pp 283–335.
56. Nickel, T.; Bauer, U.; Schlicker, E.; Kathmann, M.;
Gothert, M.; Sasse, A.; Stark, H.; Schunack, W. Br. J.
Pharmacol. 2001, 132, 1665.
57. Laitinen, J. T.; Jokinen, M. J. Neurochem. 1998, 71, 808.
58. Moore, R. J.; Xiao, R.; Sim-Selley, L. J.; Childers, S. R.
Neuropharmacology 2000, 39, 282.
59. Bradford, M. M. Anal. Biochem. 1976, 72, 248.
60. Clark, E. A.; Hill, S. J. Eur. J. Pharmacol. 1996, 296, 223.
61. Simca-S 6.01, Umetri AB & Ericsson AB, 1997.
62. Wold, S.; Eriksson, L. In: Chemometric Methods in
Molecular Design; van de Waaterbeemd, H. (Ed.), VCH
Publishers Inc.: New York, 1995; pp 309–318.
42. Sorba, G.; Garrone, A.; Serafino, A.; Gasco, A.; Orsetti,
M. Eur. J. Med. Chem.-Chim. Ther. 1986, 21, 391.