Indole amide derivatives: Synthesis, structure-activity relationships and molecular modelling studies of a new series of histamine H1-receptor antagonists

Article abstract of DOI:10.1016/S0223-5234(99)80044-0

A number of indole amide derivatives bearing a basic side chain, in which the indole ring replaces the isoster benzimidazole nucleus typical of some well-known antihistamines, were prepared and tested for their H₁- antihistaminic activity. The 1-benzyl-3-indolecarboxamides 32-42 showed antihistaminic (H₁) activity (pA₂ 6-8); the 3-indolylglyoxylylamides 7-16 and the 2-indolecarboxamides 48-56 showed little or no activity. Insertion of the basic side chain of the active 3-indolecarboxamide derivatives into a piperazine ring (compounds 57-59) led to a dramatic loss of activity. All the active compounds proved to be competitive antagonists, since the values of the regression slope were not statistically different from 1. The most active compounds, 32, 33, 38-41, were also tested both in vitro for their anticholinergic activity and in vivo for their ability to antagonize histamine-induced cutaneous vascular permeability in rats. The biological results and the structure-activity relationships of the novel compounds are discussed in the light of molecular modelling studies, taking the molecule of astemizole as a model, and referring to proposed H₁-receptor pharmacophore models.

Full text of DOI:10.1016/S0223-5234(99)80044-0

Eur. J. Med. Chem. 34 (1999) 93−105
© Elsevier, Paris
93
Original article
Indole amide derivatives: synthesis, structure–activity relationships and
molecular modelling studies of a new series of histamine H1-receptor antagonists
Sandra Battagliaa, Enrico Boldrinib, Federico Da Settimoa*, Giulio Dondioc, Concettina La Mottaa,
Anna Maria Marinia, Giampaolo Primofiorea
aDipartimento di Scienze Farmaceutiche, Università di Pisa, Via Bonanno 6, 56126 Pisa, Italy
bFarmigea SpA, Via Carmignani 2, 56127 Pisa, Italy
cSmithKline Beecham SpA, Via Zambeletti, 20021 Baranzate, Milano, Italy
(Received 7 August 1998; accepted 1 September 1998)
Abstract – A number of indole amide derivatives bearing a basic side chain, in which the indole ring replaces the isoster benzimidazole
nucleus typical of some well-known antihistamines, were prepared and tested for their H1-antihistaminic activity. The 1-benzyl-3-
indolecarboxamides 32–42 showed antihistaminic (H1) activity (pA2 6–8); the 3-indolylglyoxylylamides 7–16 and the 2-indolecarboxamides
48–56 showed little or no activity. Insertion of the basic side chain of the active 3-indolecarboxamide derivatives into a piperazine ring
(compounds 57–59) led to a dramatic loss of activity. All the active compounds proved to be competitive antagonists, since the values of the
regression slope were not statistically different from 1. The most active compounds, 32, 33, 38–41, were also tested both in vitro for their
anticholinergic activity and in vivo for their ability to antagonize histamine-induced cutaneous vascular permeability in rats. The biological
results and the structure–activity relationships of the novel compounds are discussed in the light of molecular modelling studies, taking the
molecule of astemizole as a model, and referring to proposed H1-receptor pharmacophore models. © Elsevier, Paris
indole derivatives / non-classic H1-antagonists / astemizole / structure–activity relationships
1. Introduction
sedative effects. Examples of these compounds are ter-
fenadine, cetirizine, loratadine etc. On the other hand,
stucturally completely different compounds devoid of the
unwanted side-effects of classic H1-antagonists have been
developed. Because of their different chemical structure
these compounds were called non-classic compounds.
Astemizole (figure 1) is one of the few non-classic
compounds known to be used therapeutically [11–14].
H1-antagonists can be classified into two major groups:
classic and non-classic [1]. Classic H1-antagonists all
share the following basic structure I (figure 1), which
consists of a basic nitrogen atom, predominantly proto-
nated at physiological pH, and two aromatic groups,
connected via a linking group, which can be of different
chemical nature.
For these non-classic derivatives of 2-amino-
benzimidazole, a model of the binding site has been
proposed by Iemura et al. [15]. It consists of an anionic
site which binds the positively-charged piperazinyl nitro-
gen atom, a slit-shaped cavity occupied by the
1-substituent group, and a site perpendicular to the
previous one which interacts with the benzimidazole
nucleus.
As side effects (anticholinergic activity and CNS
depression such as sedation, hypnosis, etc.) commonly
occur with all classic antihistamines, extensive studies
have been carried out with the aim of developing anti-
histamine drugs with minimal side effects [2–10]. Two
approaches have been followed to reach this goal. In the
first place hydrophilic substituent groups have been
introduced in classic compounds to reduce their
blood–brain barrier penetration and, consequently, their
Although the non-classic antihistamines seem to pos-
sess new structural features, they present elements of the
general structure I [1, 16].
*Correspondence and reprints

94
dialkylaminoethyl chain was linked at position 3 of the
indole system by a COCONH bridge (compounds 7–16),
as the indolylglyoxylylamide moiety proved to be a
suitable substructure for many compounds synthesized by
us, which showed a good antidepressant [18] and anti
gastric secretory [19] activity, and interacted with a high
affinity at the benzodiazepine receptor [20]. Indeed, a
preliminary study had shown a very good superimposi-
tion of the groups thought to be responsible for antihis-
taminic activity in the structure of these substituted
1-benzylindole derivatives and that of astemizole, taken
as the model compound (see Section 5).
For a better and wider definition of structure–activity
relationships (SAR), we also considered compounds in
which the distance between the aromatic moiety (indole
nucleus) and the cationic centre was varied by substitu-
tion of the oxalylamide group COCONH with a car-
boxyamide residue CONH, bound at position 3 (com-
pounds 29–43) or 2 (compounds 48–56) of the indole
system. It is interesting to note that there are a number of
examples in literature of non-classic H1-antagonists, in
which one of the aromatic groups is separated from the
cationic centre by means of an intermediate chain CON-
H(CH2)2 [21, 22]. Derivatives in which the aminoethyla-
mide chain at position 3 of the indole nucleus was
inserted into a piperazine ring (compounds 57–59) were
also prepared.
Figure 1.
The SARs (antihistaminic) of the indole amides are
discussed in the light of molecular modelling studies by
comparing the structures of selected derivatives with that
of the related astemizole, taken as the reference model.
In a previous paper [17], we reported the synthesis and
the antihistaminic activity of some 1-substituted
2-benzylaminobenzimidazole derivatives, whose general
structure presents few similarities to that of classic
antihistamines I, even if some common elements can be
discerned.
2. Chemistry
In the present paper we report the synthesis and the
evaluation of the antihistaminic activity of a series of
amide derivatives bearing a basic chain, in which the
benzimidazole nucleus has been replaced by the isoster
indole nucleus. In the initially prepared compounds, the
Compounds 7–16 were obtained by reacting under
mild conditions the indolylglyoxylyl chlorides 1–6 [19]
with the appropriate amine in the presence of triethy-
lamine
in
benzene
solution
(figure 2). The
Figure 2.

95
Figure 3.
1-benzylindole-3-ylglyoxylyl chloride 6, never described
This experimental procedure was also used for the
synthesis of the N-(2-indolecarbonyl)amine derivatives
48–56 (figure 4). The 1-benzyl-5-chloro-2-indole-
carboxylic acid 47, never described in literature, was
prepared by Fisher indole synthesis starting from ethyl
pyruvate and 1-benzyl-1-(4-chlorophenyl)hydrazine.
N,N’-carbonyldiimidazole in DMF was also employed
as the condensing agent to obtain the N-(1-benzyl-3-
indolecarbonyl)piperazine derivatives 57 and 58 from the
1-benzylindole-3-carboxylic acid 24 and the appropriate
N-alkylpiperazine. The 1-benzylindole-3-carboxylic acid
24, after conversion into its chloride by treatment with
excess thionyl chloride in ethyl ether, was condensed
with N-(â-hydroxyethyl)piperazine in anhydrous THF, in
the presence of triethylamine, to give the target com-
pound 59.
in
literature,
was
prepared
starting
from
1-benzylindole [23] essentially following the experimen-
tal procedure used for the synthesis of the other indolylg-
lyoxylyl chlorides.
The synthesis of the N-(3-indolecarbonyl)amine de-
rivatives 29–43 involved the preparation of the appropri-
ate 3-indolecarboxylic acids 23–28. These were obtained,
with the exception of the commercial unsubstituted acid,
by alkylation of indole-3-carbaldehyde and 5-chloro-
indole-3-carbaldehyde [24], and successive oxidation
with potassium permanganate in acetone, essentially in
accordance with a literature procedure [25] (figure 3).
The target compounds 29–43 were prepared from the
acids and the appropriate amines, in anhydrous DMF,
under
a
nitrogen
atmosphere,
using
N,N’-
carbonyldiimidazole (CDI) as the condensing agent (fig-
ure 3).
All products were purified by recrystallization from the
appropriate solvent, after a first purification, when nec-
Figure 4.

96
Table I. N-(Indol-3-ylglyoxylyl)amine derivatives 7–16: physical properties and histamine antagonist activity.
a
b
Compound
R1
R2
R3–N–R3
Yield (%) Recryst. solvent
M.p. (°C) Formula
pA2 (slope)
7
8
9
10
11
12
H
Cl
Br
NO2
OCH3
H
H
H
H
H
H
CH3 / CH3
CH3 / CH3
CH3 / CH3
CH3 / CH3
CH3 / CH3
47
55
73
72
48
40
Benzene
Methanol
Methanol
Methanol
Benzene
Pet. ether 60–80°
174–175
220–221
221–223
228–230
194–195
102–104
C14H17N3O2
C14H16ClN3O2 NA
C14H16BrN3O2 NA
C14H16N4O4
C15H19N3O3
C21H23N3O2
NA
NA
NA
6.01 ± 0.25
(0.91 ± 0.18)
NA
NA
NA
NA
CH2C6H5 CH3 / CH3
13
14
15
16
H
H
H
H
CH2C6H5 Pyrrolidine
CH2C6H5 Piperidine
CH2C6H5 N-methylpiperazine 39
CH2C6H5 Morpholine 47
46
47
Pet. ether 60–80°
Pet. ether 100–140° 131–132
Pet. ether 60–80°
Pet. ether 60–80°
120–122
C23H25N3O2
C24H27N3O2
C24H28N4O2
C23H25N3O3
132–133
119–121
Astemizole
8.61 ± 0.14
(1.22 ± 0.28)
9.50 ± 0.41
(0.90 ± 0.19)
Mepyramine
b
a All values for C, H, N were within ±0.4% of calculated values; NA = not active at a concentration lower than 10–6 M.
essary, on a silica gel column. Their structure was
confirmed by IR, 1H-NMR, MS spectral data and elemen-
tal analyses (table I–III).
4. Results
Table I shows the in vitro data of the histamine
antagonist activity of indolylglyoxylylamide derivatives
7–16. Compounds 7–11 did not exhibit any antagonist
activity up to a concentration of 10–6 M. The presence of
a substituent at position 5 of the indole nucleus with
varying electronic and lipophilic characteristics did not
give any significant activity improvement. Compound 12,
bearing a benzyl group at the indole nitrogen atom,
proved to be a weak H1-antagonist, showing a pA2 value
of 6.01, with a regression slope not statistically different
from 1 (slope = 0.91 ± 0.18). Further studies regarding
the basic nitrogen substitution in the 1-benzyl series
showed that cyclic amines were not well tolerated (com-
pounds 13–16 compared with 12).
The 3-indolecarboxamide derivatives 29–43 showed,
in general, a good antihistaminic activity, with pA2 values
ranging from 8.23 to 5.94 and slopes not statistically
different from 1 (table II). Also in this series of com-
pounds, the presence of a benzyl substituent at position 1
of the indole moiety was beneficial for the activity
(compare compounds 29, 30 versus 32, 33). The intro-
duction of a halo substituent, i.e. fluoro and chloro, in the
para position of the N-benzyl moiety did not significantly
3. Pharmacology
All compounds synthesized were assessed in vitro for
their ability to antagonize histamine-induced contraction
of guinea-pig ileum as previously described [17]
(table I–III), using mepyramine and astemizole [26, 27],
selective anti-H1 drugs, as the reference standards. This
test is a reliable in vitro measure of H1-antagonist
activity. Those products which showed a significant
inhibiting effect at the starting concentration of 10–6 M,
were then tested with lower doses to determine their pA2
values.
Selected compounds (32, 33, 38–41) which demon-
strated significant in vitro inhibition in the guinea-pig
ileum assay (pA2 > 7) were then tested in vivo for their
ability to antagonize histamine-induced cutaneous vascu-
lar permeability in rats [28] (table IV). Moreover, with
the aim to evaluate their potential side effects, they were
also assayed in vitro for anticholinergic activity.

97
Table II. N-(3-Indolecarbonyl)-2-dialkylaminoethylamine derivatives 29–43 and N-(2-indolecarbonyl)-2-dialkylaminoethylamine derivatives
48–56: physical properties and histamine antagonist activity.
a
b
Compound R1
R2
R3–N–R3
Yield (%) Recryst. solvent
M.p. (°C) Formula
pA2 (slope)
29
30
31
32
H
H
H
H
H
H
H
CH3 / CH3
Piperidine
Pyrrolidine
CH3 / CH3
45
70
59
68
Benzene
Pet. ether 60–80°
Benzene
116–118
136–137
122–124
138–140
C13H17N3O
C16H21N3O
C15H19N3O
C20H23N3O
NA
NA
NA
7.03 ± 0.22
(0.97 ± 0.20)
8.23 ± 0.34
(0.80 ± 0.28)
6.36 ± 0.21
(1.02 ± 0.16)
NA
CH2C6H5
Benzene
33
34
H
H
CH2C6H5
CH2C6H5
Piperidine
35
Benzene/Pet.
ether 60–80°
Benzene
130–131
126–128
C23H27N3O
C23H28N4O
C29H32N4O
C20H22ClN3O 6.23 ± 0.53
(1.00 ± 0.34)
C23H26ClN3O NA
N-methylpiperazine 36
35
36
H
Cl
CH2C6H5
CH2C6H5
N-benzylpiperazine 32
Toluene
Benzene
120–122
126–128
CH3 / CH3
Piperidine
43
37
38
Cl
H
CH2C6H5
36
32
Toluene
Toluene
138–140
168–170
CH2C6H4-p-F CH3 / CH3
CH2C6H4-p-F Piperidine
CH2C6H4-p-Cl CH3 / CH3
CH2C6H4-p-Cl Piperidine
CH2C6H4-p-Cl CH3 / CH3
CH2C6H4-p-Cl Piperidine
C20H22FN3O
7.05 ± 0.13
(0.89 ± 0.16)
8.00 ± 0.27
(0.91 ± 0.13)
39
40
41
42
43
H
39
69
76
32
43
Toluene
Toluene
151–152
185–186
158–159
132–134
115–117
C23H26FN3O
H
C20H22ClN3O 7.75 ± 0.37
(0.79 ± 0.13)
C23H26ClN3O 7.13 ± 0.20
(0.88 ± 0.11)
C20H11Cl2N3O 5.94 ± 0.21
(1.19 ± 0.33)
C23H25Cl2N3O NA
H
Toluene/Pet.
ether 60–80°
Benzene
Cl
Cl
Benzene
48
49
50
51
H
H
Cl
Cl
H
H
H
H
CH3 / CH3
Piperidine
CH3 / CH3
Piperidine
74
81
91
73
Benzene
Benzene
Benzene
Benzene
137–138
159–161
203–205
219–220
C13H17N3O
C16H21N3O
C13H16ClN3O NA
C16H20ClN3O 6.53 ± 0.22
(0.93 ± 0.23)
NA
NA
52
Cl
H
Pyrrolidine
77
Benzene
204–206
C15H18ClN3O 5.90 ± 0.33
(0.93 ± 0.31)
53
54
H
H
CH2C6H5
CH2C6H5
CH3 / CH3
Piperidine
32
36
Benzene
74–76
123–125
C20H23N3O
C23H27N3O
NA
6.42 ± 0.28
(1.08 ± 0.22)
Benzene/Pet.
ether 60–80°
Benzene
55
Cl
Cl
CH2C6H5
CH2C6H5
CH3 / CH3
Piperidine
38
25
118–120
127–128
C20H22ClN3O 6.50 ± 0.31
(1.10 ± 0.28)
C23H26ClN3O 6.22 ± 0.38
(1.01 ± 0.30)
56
Benzene/Pet.
ether 60–80°
Astemizole
Mepyramine
8.61 ± 0.14
(1.22 ± 0.28)
9.50 ± 0.41
(0.90 ± 0.19)
b
a All values for C, H, N were within ±0.4% of calculated values; NA = not active at a concentration lower than 10–6 M.

98
Table III. 4-Substituted-1-(1-benzyl-3-indolecarbonyl)piperazine derivatives 57–59: physical properties and histamine antagonist activity.
a
b
Compound
R1
R2
Yield (%)
Recryst. solvent
M.p. (°C)
Formula
pA2
57
58
59
CH2C6H5
CH2C6H5
CH2C6H5
CH3
CH2C6H5
CH2CH2OH
27
30
42
Oil
127–129
145–146
C21H23N3O
C27H27N3O
C22H25N3O2
NA
NA
NA
Toluene
Toluene
b
a All values for C, H, N were within ±0.4% of calculated values; NA = not active at a concentration lower than 10–6 M.
alter the overall activity (compounds 32, 33 compared
with 38, 39, 40, 41). The presence of a chlorine atom at
position 5 of the indole moiety resulted in compounds
with a lower activity (see for example compounds 32, 33,
40, 41 compared with 36, 37, 42, 43). Also in this series
of compounds, the substitution pattern of the basic head
was considered. The piperidine derivative 33 was found
to be the most interesting compound, showing a pA2
value of 8.23, one order of magnitude more potent than
that of the dimethylamino compound 32, and comparable
with that of astemizole taken as the reference standard.
Bulky amines, such as N-methyl and N-benzylpiperazine,
gave compounds 34, 35, respectively, with a dramatic
reduction in the H1-antagonist activity.
The 2-indolecarboxamide derivatives 48–56 showed,
in general, little or no H1-antagonist activity, with pA2
values ranging between 6.53 and 5.90 (table II). In this
series of compounds, the combination of the most inter-
esting moieties, i.e. 1-benzyl group and piperidine as
basic head, selected from previous SAR analysis, did not
give positive results (compare compound 54 with respect
to 33).
The incorporation of the aminoethyl side chain of the
3-indole substituted derivatives into a piperazine ring
resulted in compounds 57–59, which did not show any
activity (table III).
The most active compounds (32, 33, 38–41 with pA2 >
7 in the guinea-pig ileum assay), tested for their anticho-
linergic activity, did not show any effect up to a concen-
tration of 10–4 M. These compounds were also assayed in
vivo for their ability to antagonize histamine-induced
cutaneous vascular permeability in rats. The data reported
in table IV show that only compounds 38–41 possessed
any significant activity (i.e. ca. 50% of inhibition at
10 mg/kg s.c.). Moreover, it has to be pointed out that
compounds 40 and 41 presented a pharmacodynamic
profile similar to that observed for the reference antago-
nist mepyramine, as they inhibited dose-dependently the
histamine-induced effect.
Table IV. Effects of selected compounds and reference drugs on cutaneous permeability in rats.
b
c
Compound
Inhib. (%) a (mg/kg s.c.)
ED50 (mg/kg s.c.) Potency ratio
2
3
4.5
6.7
10
32
33
38
39
40
41
14.0 ± 4.1
37.1 ± 6.7
48.0 ± 7.8
45.2 ± 3.7
58.5 ± 2.6
63.0 ± 1.9
77.5 ± 5.5
13.8 ± 0.8
15.1 ± 1.5
32.7 ± 3.9
41.8 ± 2.8
8.9 ± 1.2
7.5 ± 1.1
5.1 ± 1.1
0.57
0.68
Mepyramine 12.4 ± 10
35.6 ± 6.7
a Percent of inhibition vs. control; b each compound was tested with three doses ranging between 2 and 10 mg/kg. Each dose was tested s.c.
c
in triplicate; ED50(mepyramine)/ED50(compound).

99
Table V. Atomic distances (Å) of selected minimum conformations of interest used for superimposition studies. d1 is the distance between
the protonated nitrogen and the centre of gravity of the benzene ring of the indole or benzimidazole nucleus; d2 is the distance between the
centres of gravity of the two unsaturated moieties of the molecule. Root mean square (RMS) values for each superimposition with the two
conformations considered for astemizole vs. the compounds described in this studies are also reported.
Compound
d1
d2
RMS vs. conf. 1
RMS vs. conf. 2
Astemizole (conf. 1)
Astemizole (conf. 2)
12
32
53
57
8.21
6.77
8.26
6.53
7.71
7.51
5.58
5.18
5.27
5.20
5.89
5.19
0.145
1.111
1.553
1.117
0.753
0.374
2.043
0.684
5. Discussion
activity vs. histamine (demonstrated by the value of the
regression slope not statistically different from 1) this
compound was considered to be a good tool to optimize
the overall activity of this new class of compounds. As
the low H1-antagonist activity of compound 12 could be
due to bad steric and/or electrostatic interactions of the
oxalyl bridge with the receptor site, optimization studies
were focused on the modification of the COCONH
residue. A first attempt was made by removing one of the
carbonyl groups of the glyoxyl moiety, leaving the
resulting CONH bridge linked to position 3 of the indole
ring. The compounds obtained, 29–43, proved to be very
interesting. In particular, compound 32 (table II) was 10
times more potent as an antagonist than the parent
compound 12.
The 3-indolylglyoxylylamides 12–16 have a molecular
framework very similar to that of non-classic antihista-
mines, such as astemizole. Moreover, they fulfil all the
requirements reported by Iemura et al. [15] to interact
with the H1-receptor binding site: a positively-charged
nitrogen atom, a planar lipophilic area, and a perpendicu-
lar substituent at the 1-position entering a slit-shaped
cavity. Figure 5 shows the superimposition of compound
12, selected as a model for the indolylglyoxylylamide
class, in its absolute minimum energy conformation, on
the reference compound astemizole. The conformation of
astemizole was obtained after minimization of the origi-
nal X-ray structure, while a molecular model of 12 was
generated by the MacroModel/BatchMin V 5.0 pro-
gramme (computational details are summarized in the
Experimental Section). As can be seen, the groups con-
sidered essential for binding with the H1 receptor, i.e. two
aromatic rings and the basic nitrogen, lie for both
products in the same spatial region. The presence of a
strong electrostatic interaction between the oxygen of the
amide function and the protonated basic nitrogen limited
the conformational freedom of the side chain of com-
pound 12, thus providing a good fit with astemizole, as
demonstrated by the low value of the root mean square
(RMS) of the distances between the selected groups
thought to be responsible for the activity (RMS = 0.145)
(table V). However, in vitro characterization in the
guinea-pig ileum assay showed only a weak H1-
antagonist potency for compound 12 (pA2 = 6.01) (ta-
ble I). Nevertheless, in view of its competitive antagonist
Molecular modelling studies were undertaken on com-
pound 32 to rationalize the activity of this new class of
derivatives. A first comparison of the astemizole mol-
ecule with a model of compound 32 in its minimum
energy conformation (figure 6) showed a different spatial
arrangement with a high value of the RMS of 1.111
(table V) between the selected moieties supposed to be
involved in the receptor interaction. As the conforma-
tional freedom of the side chain of 32 is limited by the
strong electrostatic interaction between the oxygen of the
amide function and the cationic head, a better fit (RMS =
0.374) (figure 7) was obtained considering a different
astemizole conformation (conf. 2), still accessible for
receptor interaction (Econf. 2 – Emin ≤ 5 kcal/mol). It may
be noted that in this conformation (conf. 2), the distance
between the protonated nitrogen and the lipophilic area

100
Figure 6. Superimposition of astemizole (minimum energy
conformation, gray) and 32 (black). Points used were the basic
nitrogens and the benzene rings of the indole and benzyl
moieties (RMS = 1.111).
Figure 5. Superimposition of astemizole (gray) and 12 (black).
Points used were the basic nitrogens and the centroids of the
benzene rings of the indole and benzyl moieties (RMS = 0.145).
represented by the benzene moiety of the benzimidazole
ring (d1) was shorter than in the minimum energy
conformation (conf. 1) (table V), in better agreement with
those originally reported by Naruto [29] and Borea [30]
in their proposed pharmacophore models (i.e. d1 about
6.5 Å and 6.2 Å, respectively). In the light of these
results, it can be speculated that astemizole could not
interact with the H1-receptor site in its minimum energy
conformation (conf. 1), but instead in another accessible
conformation (conf. 2) which fulfils the H1-
pharmacophore requirements, as discussed above. This
hypothesis could account for the inactivity of compounds
7–16, as the superimposition of 12 on astemizole in its
proposed active conformation (conf. 2) did not show a
good fit between the groups responsible for the activity
(RMS of 0.753 vs. conf. 1 RMS of 0.145).
In conclusion, a good H1-antagonist activity was
shown by the derivatives in which the basic chain was
linked at position 3 of the indole system by means of a
carboxamide bridge, with a benzyl group at the indole
nitrogen atom, which can fit into an appropriate cavity on
the receptor site. The molecular modelling results seem to
suggest that the pharmacologically active conformation
of astemizole could be similar to that indicated as conf. 2
Different results were obtained when the same amidic
side chain was linked at position 2 of the indole moiety.
Compounds 48–56 showed little or no activity in the
guinea-pig ileum bioassay. It was found that none of the
accessible conformations of compound 53, taken as a
model for this class of compounds, fits with either of the
conformations of astemizole considered in this study.
This is reflected by the higher d1 distance showed by the
minimum energy conformation of compound 53 com-
pared with that of compound 32 (table V). It is also worth
noting that in this case the presence of a strong electro-
static interaction with the protonated basic nitrogen and
the amidic carbonyl group greatly limits the conforma-
tional freedom of the molecule. A similar situation occurs
in the case of the piperazine derivative 57. Its greater d1
value of 7.51 Å may explain its poor activity as an H1-
antagonist.
Figure 7. Superimposition of astemizole (conformation 2,
gray) and 32 (black). Points used were the same as those in
figure 6 (RMS = 0.374).

101
(table V, figure 7), even though this conformation is not
the minimum energy one, but it is still accessible for
receptor interaction. This conformation possesses the
right distances between the protonated nitrogen and the
centroid of at least one of the aromatic rings for potent
H1-antagonist activity.
0 °C, to a solution of 2.02 g (9.8 mmol) of 1-benzylindole
in 35 mL of the same solvent. The reaction mixture was
left to warm at room temperature, stirred for 4 h and then
cooled again at 0 °C. The precipitate formed was col-
lected, washed with small portions of anhydrous ethyl
ether and dried under vacuum to give pure 6 as a yellow
microcrystalline solid (1.85 g). M.p. 104–105 °C, yield
64%. IR ν cm–1: 1800, 1670, 1220, 1000, 750. 1H-NMR:
δ 5.58 (s, 2H, CH2); 7.12–8.63 (m, 10H, ArH). MS: m/e
297 (M)+, 91, base. Anal. (C17H12ClNO2) C, H, N.
6. Experimental protocols
6.1. Chemistry
6.1.2. General procedure for the synthesis of N-[(1- or
5-substituted indole-3-yl)glyoxylyl]-2-dialkylaminoethyl-
amine derivatives 7–16
Melting points were determined on a Köfler hot-stage
apparatus and are uncorrected. IR spectra were recorded
with a Pye Unicam Infracord Model PU 9516 in Nujol
A solution of the appropriate amine (10 mmol) in
20 mL of anhydrous benzene was added, dropwise, at
0 °C, to a solution of the appropriately substituted in-
dolylglyoxylyl chloride 1–6 (9.4 mmol) and triethy-
lamine (1.5 mL, 11 mmol) in 120 mL of anhydrous ben-
zene. The reaction mixture was left to stir at room
temperature for 48–72 h, monitoring the reaction by TLC
analysis. The precipitate present was filtered off and the
filtrate was concentrated to dryness. The oily residue was
solubilized in chloroform, washed with saturated sodium
hydrogen carbonate aqueous solution and water, and
concentrated again. The crude product obtained was
purified by recrystallization from the appropriate solvent.
Yields, recrystallization solvents and melting points are
listed in table I. The spectrometric data for 12, which is
representative of the title compounds, are listed below.
1
mulls. Routine H NMR spectra were determined on a
Varian CFT 20 spectrometer operating at 80 MHz, using
tetramethylsilane (TMS) as the internal standard.
DMSO-d6 was used as the solvent, unless otherwise
indicated. Mass spectra were obtained on a Hewlett-
Packard 5988 A spectrometer using a direct injection
probe and an electron beam energy of 70 eV. Magnesium
sulfate was always used as the drying agent. Evaporations
were made in vacuo (rotating evaporator). Analytical
TLC was carried out on Merck 0.2 mm precoated silica
gel (60 F-254) aluminium sheets, with visualization by
irradiation with a UV lamp. Silica gel 60 (70–230 mesh)
was used for column chromatography and silica gel 60
(230–400 mesh) was used for flash chromatography.
Elemental analyses were performed by our Analytical
Laboratory and agreed with theoretical values to within
±0.4%.
The following products were prepared according to
literature: 1-methyl-4-(â-aminoethyl)piperazine, b.p.
70–72 °C/4 mmHg (lit. [31] b.p. 61–62 °C/1 mmHg);
1-benzyl-4-(â-aminoethyl)piperazine, b.p. 170–173 °C/5
mmHg (lit. [22] b.p. 180–187 °C/13 mmHg); 1-benzyl-1-
phenylhydrazine, m.p. 167–169 °C (lit. [32] m.p.
169–171 °C); 1-benzylindole, m.p. 44–45 °C (lit. [23]
m.p. 43–45 °C); 1-benzylindole-2-carboxylic acid, m.p.
199–200 °C (lit. [33] m.p. 198–201 °C); 5-chloroindole-
3-carbaldehyde, m.p. 214–215 °C (lit. [24] m.p.
6.1.2.1. N-[(1-Benzylindole-3-yl)glyoxylyl]-2-dimethyl-
aminoethylamine 12:
IR ν cm–1: 3300, 1660, 1625, 1500, 1170, 750.
1H-NMR: δ 2.20 (s, 6H, (CH3)2N); 2.40 (t, 2H,
(CH3)2NCH2); 3.33 (q, 2H, NHCH2CH2); 5.62 (s, 2H,
CH2C6); 7.83–9.16 (m, 10H, ArH); 8.63 (br t, 1H, NH,
exch. with D2O). MS: m/e 349 (M)+, 58, base.
6.1.3. General procedure for the synthesis of 5-
substituted 1-(4-substituted benzyl)indole-3-carbal-
dehydes 17–22
Sodium hydride (0.576 g, 24 mmol, 50% dispersion in
mineral oil) was added, in small portions, to a solution of
the appropriately substituted indole-3-carbaldehyde
(20 mmol) in 33 mL of freshly distilled DMF, then,
dropwise, the appropriate benzylchloride (24 mmol). The
reaction mixture was heated at 90 °C for 2.5–3 h (TLC
analysis), and then, after cooling, was poured into ice.
After acidification to pH 6 with 2 N hydrochloric acid, the
solid formed was collected, washed with water and
purified by recrystallization, after filtration, when neces-
sary, on a silica gel column. Yields, recrystallization
215–216 °C);
1-benzylindole-3-carbaldehyde,
m.p.
110–112 °C (lit. [34] m.p. 113–114 °C); 1-(4-chloro-
benzyl)indole-3-carbaldehyde, m.p. 120–121 °C (lit. [35]
m.p. 121–122 °C); 1-benzylindole-3-carboxylic acid,
m.p. 199–200 °C (lit. [25] m.p. 198–201 °C); 1-(4-chloro-
benzyl)indole-3-carboxylic acid, m.p. 209–211 °C
(lit. [25] m.p. 207–209 °C).
6.1.1. 1-Benzylindole-3-ylglyoxylyl chloride 6
A solution of oxalyl chloride (1.72 mL, 17.9 mmol) in
3 mL of anhydrous ethyl ether was added, dropwise at

102
solvents, melting points and spectrometric data for the
newly synthesized compounds are listed below.
6.1.4.3. 1-(4-Chlorobenzyl)-5-chloroindole-3-carboxy-
lic acid 28:
M.p. 212–214 °C (toluene); yield 62%. IR ν cm–1:
3150–2400, 1640, 1520, 1200, 1080. 1H-NMR: δ 5.48 (s,
2H, CH2); 7.10–8.40 (m, 8H, ArH); 12.00 (br s, 1H,
COOH, exch. with D2O). MS: m/e 319 (M)+, 125, base.
Anal. (C16H11Cl2NO2) C, H, N.
6.1.3.1. 1-Benzyl-5-chloroindole-3-carbaldehyde 19:
M.p. 141–143 °C (ethanol); yield 72%. IR ν cm–1:
1
1640, 1170, 1050, 800. H-NMR: δ 5.51 (s, 2H, CH2);
7.15–8.41 (m, 9H, ArH); 9.89 (s, 1H, CHO). MS: m/e 269
(M)+, 91, base. Anal. (C16H12ClNO) C, H, N.
6.1.5. 5-Chloro-1-benzylindole-2-carboxylic acid 47
The acid 47 was prepared starting from ethyl pyruvate
N-benzyl-N-(4-chlorophenyl)hydrazone. This intermedi-
ate was obtained by refluxing for 1 h a mixture of ethyl
pyruvate (1.2 mL, 11 mmol) and N-benzyl-N-(4-
chlorophenyl)hydrazine (2.3 g, 10 mmol). The crude
product obtained was purified by flash chromatography
(eluting system: petroleum ether 60–80°/dichloromethane
= 95:5), yielding 2.4 g of pure hydrazone as an oil. Yield
72%. IR ν cm–1: 2995, 1710, 1490, 1270, 1140, 730.
1H-NMR (CDCl3): δ 1.38 (t, 3H, CH2CH3); 1.93 (s, 3H,
CH3); 4.33 (q, 2H, CH2CH3); 5.00 (s, 2H, CH2C6);
6.86–7.66 (m, 9H, ArH). MS: m/e 330 (M)+, 91, base.
Anal. (C18H19ClN2O2) C, H, N.
A mixture of 1.65 g (5 mmol) of the hydrazone ob-
tained in 5 mL of ethanol and 15 mL of 12% hydrochloric
acid was refluxed for 2.5 h. After cooling, the precipitate
formed was collected, dried under vacuum and purified
by recrystallization from benzene, giving 0.45 g of pure
acid. M.p. 151–153 °C; yield 32%. IR ν cm–1:
3000–2400, 1650, 1500, 1240, 1180. 1H-NMR: δ 4.42 (s,
2H, CH2); 7.15–7.54 (m, 9H, ArH); 11.55 (br s, 1H,
COOH, exch. with D2O). MS: m/e 285 (M)+, 44, base.
Anal. (C16H12ClNO2) C, H, N.
6.1.3.2. 1-(4-Fluorobenzyl)indole-3-carbaldehyde 20:
M.p. 106–108 °C (ethanol); yield 80%. IR ν cm–1:
1670, 1600, 1290, 1160, 770. H-NMR: δ 5.50 (s, 2H,
1
CH2); 7.13–8.45 (m, 9H, ArH); 9.89 (s, 1H, CHO). MS:
m/e 253 (M)+, 95, base. Anal. (C16H12FNO) C, H, N.
6.1.3.3. 1-(4-Chlorobenzyl)-5-chloroindole-3-carbal-
dehyde 22:
M.p. 135–137 °C (ethanol); yield 52%. IR ν cm–1:
1
1630, 1510, 1150, 790. H-NMR: δ 5.52 (s, 2H, CH2);
7.15–8.45 (m, 8H, ArH); 9.89 (s, 1H, CHO). MS: m/e 303
(M)+, 125, base. Anal. (C16H11Cl2NO) C, H, N.
6.1.4. General procedure for the synthesis of 5-
substituted 1-(4-substituted benzyl)indole-3-carboxylic
acids 23–28
A solution of 2.99 g (19 mmol) of potassium perman-
ganate in 60 mL of water was added, dropwise, to a
solution of the appropriate indole-3-carbaldehyde 17–22
(10 mmol) in 150 mL of acetone. The reaction mixture
was left to stir, in the dark, at room temperature, for 16 h.
After cooling at 0 °C, the mixture was decolorized by a
10% aqueous solution of hydrogen peroxide and then
filtered. The resulting solution was concentrated to half
volume and acidified to pH 3 with 2 N hydrochloric acid.
The precipitate formed was collected, washed with water,
and purified by recrystallization. Yields, recrystallization
solvents, melting points and spectrometric data for the
newly synthesized compounds are listed below.
6.1.6. General procedure for the synthesis of N-(1,5-
substituted 3-indolecarbonyl)-2-dialkylaminoethylamine
derivatives 29–43, N-(1,5-substituted 2-indolecarbonyl)-
2-dialkylaminoethylamine derivatives 48–56 and 4-
substitutedN-(1-benzyl-3-indolecarbonyl)piperazinederi-
vatives 57, 58
N,N’-carbonyldiimidazole (0.405 g, 2.5 mmol) was
added, under stirring in a nitrogen atmosphere, to a
solution of the appropriate indolecarboxylic acid 23–28,
44–47 (2.5 mmol) in 4 mL of anhydrous DMF. After
carbon dioxide evolution had ceased, a solution of the
appropriate amine (2.5 mmol) in 2 mL of anhydrous
DMF was added to the reaction mixture, which was left
to stir at room temperature for 2.5–4 h (TLC analysis).
The resulting solution was concentrated to dryness, and
the oily residue, dissolved in chloroform, was washed
with saturated sodium hydrogen carbonate aqueous solu-
tion, and then with water. After drying, the chloroform
solution was concentrated to dryness, and the crude
6.1.4.1. 1-Benzyl-5-chloroindole-3-carboxylic acid 25:
M.p. 196–197 °C (toluene); yield 50%. IR ν cm–1:
3150–2400, 1650, 1550, 1300, 1230, 1170, 760. 1H-
NMR: δ 5.53 (s, 2H, CH2); 7.30–8.33 (m, 9H, ArH); 8.83
(s, 1H, COOH, exch. with D2O). MS: m/e 285 (M)+, 91,
base. Anal. (C16H12ClNO2) C, H, N.
6.1.4.2. 1-(4-Fluorobenzyl)indole-3-carboxylic acid
26:
M.p. 190–192 °C (toluene); yield 45%. IR ν cm–1:
3100–2400, 1630, 1500, 1250, 1190, 1060, 850, 740.
1H-NMR: δ 5.45 (s, 2H, CH2); 6.90–8.16 (m, 9H, ArH);
12.00 (br s, 1H, COOH, exch. with D2O). MS: m/e 269
(M)+, 109, base. Anal. (C16H12FNO2) C, H, N.

103
1
product was purified by flash chromatography (eluting
system: dichloromethane/methanol/ammonia conc. =
96:5:0.5 – 79:15:1) and recrystallization (table II and III).
The spectrometric data for 32, 55 and 58, which are
representative of the title compounds, are listed below.
IR ν cm–1: 3950, 1600, 1540, 1000, 745. H-NMR; δ
2.37–2.52 (m, 8H, (CH2)2NCH2CH2OH); 3.43–3.68 (m,
4H, (CH2)2NCO); 4.35 (br t, 1H, OH, exch. with D2O);
5.43 (s, 2H, CH2C6); 7.04–7.82 (m, 10H, ArH). MS: m/e
364 (MH)+, 91, base.
6.2. Molecular modelling
6.1.6.1.
N-(1-Benzyl-3-indolecarbonyl)-2-dimethyl-
aminoethylamine 32:
1
Models of compounds 12, 32, 53 and 57 were con-
structed with standard bond lengths and angles from the
fragment database in MacroModel V 5.0 (Columbia
University, New York, NY 10027) using a Silicon Graph-
ics workstation (Indigo II). A starting model of astemi-
zole was built using its X-ray coordinates (Cambridge
Structural Database reference code ZENREP 0001). All
compounds were modelled as protonated bases and mini-
mized by the MacroModel/BatchMin V 5.0 programme
using the MM2 force field. In order to perform an
extensive conformational search, a MonteCarlo/Energy
minimization [36] was carried out for all the compounds
considered in the study (Ei – Emin ≤ 5 kcal/mol). Repre-
sentative minimum energy conformations of each com-
pound were used to perform superimposition studies (see
figure captions).
IR ν cm–1: 3375, 1620, 1540, 1185, 730. H-NMR: δ
2.24 (s, 6H, (CH3)2N); 2.48 (t, 2H, (CH3)2NCH2); 2.48
(q, 2H, NHCH2); 5.43 (s, 2H, CH2C6); 7.10–8.18 (m,
10H, ArH); 7.73 (br t, 1H, NH, exch. with D2O). MS: m/e
277 (M – (CH3)2N)+, 58, base.
6.1.6.2.
N-(1-Benzyl-5-chloro-2-indolecarbonyl)-2-
dimethylaminoethylamine 55:
IR ν cm–1: 3310, 1630, 1545, 1275, 720. H-NMR: δ
2.22 (s, 6H, (CH3)2N); 2.45 (t, 2H, (CH3)2NCH2); 3.36
(q, 2H, NHCH2); 5.86 (s, 2H, CH2C6); 7.00–7.83 (m, 9H,
ArH); 8.60 (br t, 1H, NH, exch. with D2O). MS: m/e 355
(M)+, 58, base.
1
6.1.6.3. N-(1-Benzyl-3-indolecarbonyl)-4-benzylpipe-
razine 58:
IR ν cm–1: 1620, 1535, 1190, 1000, 745. H-NMR: δ
1
6.3. Pharmacology
2.38–2.54 (m, 4H, (CH2)2NCH2); 3.45–3.71 (m, 4H,
(CH2)2NCO); 3.53 (s, 2H, (CH2)2NCH2); 5.45 (s, 2H,
NCH2C6); 7.06–7.85 (m, 15H, ArH). MS: m/e 409 (M)+,
91, base.
6.3.1. Materials and methods
Experiments were carried out using male or female
Durkin–Hartley guinea-pigs, 250–300 g b.w., and male
Sprague Dawley albino rats, 120–140 g b.w., supplied by
Charles River (Calco, Italy). Reference compounds were
obtained from Sigma Chemical Co. (St. Louis, MO,
USA).
6.1.7. N-(1-Benzyl-3-indolecarbonyl)-4-(â-hydroxy-
ethyl)piperazine 59
Compound 59 was obtained essentially using an ex-
perimental procedure similar to that employed for the
above-reported
indolylglyoxylylamine
derivatives.
6.3.2. In vitro test – Histamine H1 and acetylcholine
antagonism: guinea-pig ileum
1-Benzylindole-3-carbonyl chloride was obtained by add-
ing, dropwise, at 0 °C, thionyl chloride (0.53 mL,
7.3 mmol) to a well-stirred suspension of 0.65 g
(2.6 mmol) of the acid 24. The reaction mixture, after
stirring at room temperature for 2.5 h, was concentrated
to dryness giving a solid, which was washed repeatedly
with anhydrous ethyl ether to yield 0.59 g of pure acid
chloride. M.p. 95–97 °C; yield 84%. IR ν cm–1: 1730,
In accordance with the method described by Mag-
nus [37], segments of pre-terminal ileum, about 2 cm
long, were promptly excised from guinea-pigs of either
sex and placed under a resting tension of 1.0 g in a 10 mL
bath with Tyrode’s solution, kept at 37 °C and gassed
with 5% CO2 in O2. After 1 h equilibration time, graded
concentrations of histamine or acetylcholine were added
to the bath to obtain cumulative dose–response curves, in
accordance with the method reported by Van Rossum [38,
39]. Concentrations in response to histamine or acetyl-
choline were measured as changes in isometric tension by
a force transducer.
1
1510, 1390, 1250, 1030, 730. H-NMR: δ 5.46 (s, 2H,
CH2); 7.03–8.13 (m, 10H, ArH). MS: m/e 269 (M)+, 91,
base. Anal. (C16H12ClNO) C, H, N.
The acid chloride obtained was reacted with 4-(2-
hydroxyethyl)piperazine in anhydrous THF. After the
usual work-up, the crude product was purified by flash
chromatography (eluting system: dichloromethane/
methanol/ammonia conc. = 90:7:0.5) and recrystalliza-
tion (table III).
H1-histamine or cholinergic antagonism was assessed
by adding to the bath a solution (0.1 mL) at various
concentrations of the compound under test and, after
3 min incubation, the cumulative dose–response curves

104
[2] Janssens F., Torremans J., Janssen M., Stokbroekx R.A., Luyckx
M., Janssen P.A.J., J. Med. Chem. 28 (1985) 1925–1933.
were constructed. Mepyramine and astemizole [27, 28]
were employed as a reference drugs for H1-histamine
antagonism, and atropine was used as standard for
acetylcholine antagonism.
The pA2 values for the competitive H1-antagonists
were calculated in accordance with the method described
by Schild [40, 41], using the calculation program re-
ported by Tallarida and Murray [42]. The Student’s t-test
was used to estimate statistical significance.
[3] Janssens F., Torremans J., Janssen M., Stokbroekx R.A., Luyckx
M., Janssen P.A.J., J. Med. Chem. 28 (1985) 1943–1947.
[4] Iemura I., Kawashima T., Fukuda T., Ito K., Tsukamoto G., J. Med.
Chem. 29 (1986) 1178–1183.
[5] Ife R.J., Catchpole K.W., Durant G.J., Ganellin C.R., Harvey C.A.,
Meeson M.L., Owen D.A.A., Parsons M.E., Slingsby B.P., Theobald C.J.,
Eur. J. Med. Chem. 24 (1989) 249–258.
[6] Sleevi M.C., Cale A.D., Gero T.W., Jaques L.W., Welstead W.J.,
Johnson A.F., Kilpatrick B.F., Demian I., Nolan J.C., Jenkins H., J. Med.
Chem. 34 (1991) 1314–1328.
6.3.3. In vivo test – Histamine-induced coutaneous
vascular permeability
[7] Zhang M.Q., ter Laak A.M., Timmerman H., Eur. J. Med. Chem.
28 (1993) 165–173.
The ability of compounds under investigation to inhibit
histamine-induced skin weal formation was investigated.
Anaesthetized male Sprague Dawley rats, weighing
approximately 150 g, were used. The products were given
as aqueous solutions, at a constant volume of 5 mL /kg
subcutaneously, in doses from 2 to 10 mg/kg. The same
volume of the vehicle was administered to control ani-
mals. The agents to be assayed were administered 45 min
before an intravenous injection of 0.30 mL 0.9% sodium
chloride containing 3.75 mg Evans’ blue dye. This treat-
ment was immediately followed by two distinct intracu-
taneous injections on the shaved backs of 50 µL saline-
buffered solution containing 10 µg histamine. The
reaction was evaluated after 30 min.
[8] Carceller E., Merlos M., Giral M., Balsa D., Almansa C., Bartrolí
J., García-Rafanell J., Forn J., J. Med. Chem. 37 (1994) 2697–2703.
[9] Abou-Gharbia M., Moyer J.A., Nielsen S.T., Webb M., Patel U., J.
Med. Chem. 38 (1995) 4026–4032.
[10] Iawasaki N., Ohashi T., Musoh K., Nishino H., Kado N., Yasuda
S., Kato H., Ito Y., J. Med. Chem. 38 (1995) 496–507.
[11] Janssens F., Torremans J., Janssen M., Stokbroekx R.A., Luyckx
M., Janssen P.A.J., J. Med. Chem. 28 (1985) 1934–1943.
[12] Wauquier A., Niemegeers C.J.E., Eur. J. Pharmacol. 72 (1981)
245–248.
[13] Van Wauwe J., Awouters F.H.L., Niemegeers C.J.E., Janssens F.,
VanNueten J.M., Janssen P.A.J., Arch. Int. Pharmacodyn. 251 (1981)
39–51.
[14] Awouters F.H.L., Niemegeers C.J.E., Janssen P.A.J., Arzneim.
Forsch. 33 (1983) 381–388.
[15] Iemura R., Ohtaka H., Chem. Pharm. Bull. 37 (1989) 967–972.
Animals were sacrificed, then the skin was removed
and biopsies were taken around the injection sites. The
dye contained in each cutaneous fragment was extracted
and, after centrifugation, the optical density of the super-
natant was measured at 620 nm. Quantization of Evans’
blue dye was calculated by standard calibration curves
obtained by injecting the dye intradermally at various
concentrations in other animals.
[16] ter Laak A.M., Venhorts J., Donné-op den Kelder G.M., Timmer-
man H., J. Med. Chem. 38 (1995) 3351–3360.
[17] Da Settimo A., Primofiore G., Da Settimo F., Calzolari L.,
Cazzulani P., Passoni A., Tofanetti O., Eur. J. Med. Chem. 27 (1992)
395–400.
[18] Primofiore G., Marini A.M., Da Settimo F., Franzone J.S., Mason
U., Reboani M.C., Cirillo R., Eur. J. Med. Chem. 23 (1988) 397–401.
[19] Da Settimo A., Primofiore G., Marini A.M., Ferrarini P.L., Fran-
zone J.S., Cirillo R., Reboani M.C., Eur. J. Med. Chem. 23 (1988) 21–24.
The amount of dye that leaked into the tissue was
considered the parameter of increased vascular perme-
ability.
The effects of test drugs were expressed as percentage
inhibition of the dye amounts extracted from the control
group.
[20] Da Settimo A., Primofiore G., Da Settimo F., Marini A.M.,
Novellino E., Greco G., Martini C., Giannaccini G., Lucacchini A., J. Med.
Chem. 39 (1996) 5083–5091.
[21] Saxena A.K., Ram S., Prog. Drug. Res. 23 (1979) 199–232.
[22] Saxena M., Agarwal S.K., Patnaik G.K., Saxena A.K., J. Med.
Chem. 33 (1990) 2970–2976.
[23] Rubottom G.M., Chabala J.C., Synthesis 566. (1972)
[24] Young E.H.P., J. Chem. Soc. (1958) 3493–3496.
[25] Andreani A., Rambaldi M., Bonazzi D., Andreani F., Bossa R.,
Galatulas I., Arch. Pharm. 317 (1984) 847–851.
Acknowledgements
[26] Kamei C., Mio M., Izushi K., Kitazumi K., Tsujimoto S., Fujisawa
K., Adachi Y., Tasaka K., Arzneim. Forsch. 41 (1991) 932–936.
This work was supported by grants from the Ministry
of University and Scientific and Technological Research
(MURST) (Research fund 60%).
[27] Kii Y., Inui A., Ito T., Arch. Int. Pharmacodyn. 331 (1996) 59–73.
[28] Pedersen L.G., Hoskins R.E., Cable H., J. Pharm. Pharmacol. 23
(1971) 216–218.
[29] Naruto S., Motoc I., Marshall G.R., Eur. J. Med. Chem. 20 (1985)
529–532.
References
[30] Borea P.A., Bertolasi V., Gilli G., Arzneim. Forsch. 36 (1986)
895–899.
[1] ter Laak A.M., van Droodge M.J., Timmerman H., Donnè-op den
Kelder M., Quant. Struct. Act. Relat. 11 (1992) 348–363.
[31] Najer H., Giudicelli R., Sette J., Bull. Soc. Chim. (1962) 556–559.

105
[32] Namis A.J., Cortes E., Collera O., Walls F., Bol. Inst. Quim. Univ.
Nac. Auton. Mex. 18 (1966) 64–77; Chem. Abstr. (1967) 67 73028h.
[37] Magnus R., Pflugers Arch. 102 (1904) 123–151.
[38] Van Rossum J.M., Van den Brink F.G., Arch. Int. Pharmacodyn.
143 (1963) 240–246.
[33] Daou B., Soufiaoui M., Tetrahedron 45 (1989) 3351–3361.
[34] Kalir A., Szara S., J. Med. Chem. 9 (1966) 793–794.
[39] Van Rossum J.M., Arch. Int. Pharmacodyn. 143 (1963) 299–330.
[40] Schild H.O., Brit. J. Pharmacol. 2 (1947) 189–206.
[41] Schild H.O., Brit. J. Pharmacol. 4 (1949) 277–280.
[35] Cavrini V., Gatti R., Roveri P., Cesaroni M.R., Mazzoni A.,
Fiorentini C., Arch. Pharm. 317 (1984) 662–668.
[36] Chang G., Guida W.C., Still W.C., J. Am. Chem. Soc. 111 (1989)
4379–4386.
[42] Tallarida R.J., Murray R.B., Manual of Pharmacologic Calcula-
tions with Computer Programmes, Springer-Verlag, New York, 1987.