New aromatic iminoimidazolidine derivatives as α1-adrenoceptor antagonists: A novel synthetic approach and pharmacological activity

Article abstract of DOI:10.1016/S0968-0896(00)00089-4

The design, synthesis and α₁-adrenoceptor antagonism of a series of bis-imidazoline (1a, 2a, 3a and 4a)Figure 2New families of compounds prepared in this work: (a) bis-2-iminoimidazolidinium derivatives, and (b) bis-guanidinium derivatives.Scheme 1 and bis-guanidine (1b, 2b, 3b and 4b) diphenyl derivatives are reported. All of these compounds fulfil the conditions of the most recent pharmacophore proposed for α₁-adrenoceptors and found in the literature. Besides, a novel synthetic approach to the preparation of 2-(arylimino)imidazolidine derivatives is described. All the tested compounds, except the bis-guanidinium derivative 3b, inhibit the contractile responses induced by noradrenaline in aortic rings of rat and rabbit in a dose-dependent manner. Our results indicate that, even though some discrepancies are observed in terms of the α₁ subtype targeted by this new family of compounds, they show an interesting profile as antagonists of α₁-adrenoceptors and a new prototype, compound 1a, has been found deserving further development. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00089-4

Bioorganic & Medicinal Chemistry 8 (2000) 1567±1577
New Aromatic Iminoimidazolidine Derivatives as ꢀ1-Adrenoceptor
Antagonists: A Novel Synthetic Approach and Pharmacological
Activity
Christophe Dardonville,a Pilar Goya,a Isabel Rozas,a,* Angela Alsasua,b
Ma Isabel Martõnb and Ma Jose Borregob
a
Â
Instituto de Quõmica Medica (CSIC), c/ Juan de la Cierva 3, 28006 Madrid, Spain
Â
Departamento de Farmacologõa, Facultad de Medicina, Universidad Complutense de Madrid, 28040 Madrid, Spain
b
Â
Received 7 December 1999; accepted 28 February 2000
AbstractÐThe design, synthesis and a1-adrenoceptor antagonism of a series of bis-imidazoline (1a, 2a, 3a and 4a) and bis-guanidine
(1b, 2b, 3b and 4b) diphenyl derivatives are reported. All of these compounds ful®l the conditions of the most recent pharmaco-
phore proposed for a1-adrenoceptors and found in the literature. Besides, a novel synthetic approach to the preparation of 2-(aryl-
imino)imidazolidine derivatives is described. All the tested compounds, except the bis-guanidinium derivative 3b, inhibit the
contractile responses induced by noradrenaline in aortic rings of rat and rabbit in a dose-dependent manner. Our results indicate
that, even though some discrepancies are observed in terms of the a1 subtype targeted by this new family of compounds, they show
an interesting pro®le as antagonists of a1-adrenoceptors and a new prototype, compound 1a, has been found deserving further
development. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
development of selective drugs for them would be of
great importance.
a-Adrenergic receptors play a primary role in the reg-
ulation of a variety of physiological processes, particu-
larly within the cardiovascular system. In the last
decades, the interest in these receptors, in their physi-
ological relevance and in their classi®cation has been
renewed. Whereas a2-adrenoceptors mediate in hyper-
tension, sedation, antianxiety, and analgesia among
other biological actions,1 several studies have shown
that stimulation of a1-adrenergic receptors can decrease
the propensity for abnormal heart rhythm2 whereas the
antagonists of those receptors decrease the development
of benign prostatic hyperplasia (BPH).3
Imidazoline derivatives have traditionally been con-
sidered as one of the major types of drugs that interact
with a-adrenergic receptors.6 Thus, compounds such
as clonidine or naphazoline (Fig. 1), which contain a
2-iminoimidazolidine or an imidazoline ring respectively,
show a1- and a2-adrenoceptor activities,6 and more
speci®cally, phentolamine5a (Fig. 1), which contains an
imidazoline ring, is a known a1-adrenergic antagonist.
In a previous study, carried out by us, on the similarity
between guanidine and 2-aminoimidazoline,7 the results
obtained showed close semblance between both groups
not only in terms of geometrical parameters but also at
electronic levels (in ref 7 we computed the molecular
electrostatic potential and electron density, and the
corresponding similarity indexes for both groups).
Therefore, taking into account those results we con-
sidered the preparation of two new families of derivatives
containing the guanidine or the 2-aminoimidazoline
groups.
The a1-adrenergic receptors belong to the superfamily
of G protein-coupled receptors (GPCR) which transmit
signals over the cell membrane, starting di€erent bio-
chemical events.4 They have been classi®ed into di€er-
ent subtypes and they are now designated as a1A, a1B
and a1D.5 Thus, since these adrenoceptors are present in
multiple blood vessels and other smooth muscles the
Di€erent pharmacophores have been proposed to
explain the a-adrenergic activity8 (we will discuss them
*Corresponding author. Fax: +34-91-564-4853; e-mail: rozas@iqm.
csic.es
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00089-4

1568
C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
Chemistry Results
As already mentioned, we had found by means of a
similarity study7 that the 2-aminoimidazolinium cation
is very alike structurally (bond distances, atomic char-
ges) and electronically (molecular electrostatic potential,
electron density) to the guanidinium group and, hence,
many of the methods used to introduce the guanidinium
moiety in a molecule have been applied for the intro-
duction of the 2-aminoimidazolinium group.
Figure 1. Imidazoline derivatives containing a 2-iminoimidazolidine
or an imidazoline ring which are known to interact with a-adrenergic
receptors.
To date, several routes have been used to form the 2-
iminoimidazolidine group. For instance, the direct
introduction of the imidazoline heterocycle through the
nucleophilic displacement reaction of a primary amine
was carried out with 2-nitramino-4,5-dihydro-1H-imi-
dazolium,13 2-sulfonate-4,5-dihydro-1H-imidazolium,14
or 2-methylmercapto-4,5-dihydro-1H-imidazolium
iodide.15 Recently, an intramolecular variant of the lat-
ter method was developed.16 This procedure takes place
via the ®ve-membered ring cyclization between an amine
and a thiouronium salt, previously formed in a three
step procedure. This alternative method leads, in some
cases, to higher yields although being more lengthy.17
in another section) and in all of them the presence of an
aromatic ring and a polar centre seems to be essential.
Thus, two phenyl groups connected by a group con-
taining polar atoms (NH, CO, and SO2, and CH2 for
comparison) were considered for the skeleton of these
new families of compounds (see Fig. 2).
The introduction of guanidine moieties by nucleophilic
substitution of deactivated aromatic amines is a prob-
lem which has lately been approached by some
authors.9 Their methodology has been extended to the
preparation of our diphenyl derivatives.
Classically, 2-(arylimino)imidazolidines such as cloni-
dine (see Fig. 1) and its derivatives were prepared in low
yields by reaction of N-aryl-S-methylisothiouronium
iodides or N-aryldichloroimines with ethylenediamine
(the N-aryldichloroimine precursor giving slightly better
yields18). However, puri®cation of the products may be
tedious since one has to work with very polar com-
pounds as salts or the free base of the 2-iminoimidazo-
lidine. Moreover, the poor manageability of such
unprotected compounds in a multistep synthesis made
attractive the development of a new synthetic approach
via the formation of fully Boc-protected 2-iminoimida-
zolidines in one step, in very mild conditions.
To evaluate the functional activity of a-adrenoceptor
antagonists, isolated tissues containing a receptors
showing easy and faithful responses were used. To carry
out these studies the basic pharmacological prepara-
tions are isolated rabbit and rat aorta, where the e€ect
to be measured is a change in tension induced by vaso-
constrictor compounds (a agonists).10 E€ects mediated
through other receptors have been discarded using rab-
bit aorta for bradykinin B1 receptors,11 and guinea pig
ileum for muscarinic, histaminergic and bradykinin B2
receptors.12
In the present paper we report the design, synthesis and
pharmacological activity of a series of bis-imidazoline
(1a, 2a, 3a and 4a) and bis-guanidine (1b, 2b, 3b and 4b)
diphenyl derivatives, all of them ful®lling the conditions
of the latest pharmacophore published8f and the former
of which show a1-adrenoceptor antagonism. Besides, a
novel synthetic approach to the preparation of imida-
zoline-aryl derivatives is described.
To date, protection of 2-iminoimidazolidine derivatives
has been done via N-acylated, N-benzylated19 or 3,4-
dimethoxybenzyl derivative formation.20 Recently, Kim
and Qian9 described an improved method for the prep-
aration of protected guanidines from either sterically
or electronically highly deactivated amines, using N,N0-
bis-(tert-butoxycarbonyl)thiourea (5) in the presence
of mercuric chloride. Lately, it has been found that
Mukaiyama's reagent could replace the mercuric chlor-
ide in the previous method.21
Curiously, the direct synthesis of N,N0-bis-Boc-protected
2-iminoimidazolidines from N,N0-Bis-(tert-butoxycarb-
onyl)imidazoline-2-thione (6), a compound similar to 5,
has not been described. Nevertheless, the preparation of
6 was reported in a patent in 199422 and the same
authors published in 1995 the use of N,N0-dibenzyl-
and N,N0-di-tert-butoxycarbonyl-carbamates of cyclic
thioureas as alkoxycarbonyl transfer agents for the
protection of amines.23
We have extended the methodology of Kim and Qian9
to the synthesis of N,N0-bis-Boc protected 2-iminoimi-
dazolidines of a series of diaminodiphenyl derivatives
Figure 2. New families of compounds prepared in this work: (a) bis-2-
iminoimidazolidinium derivatives, and (b) bis-guanidinium derivatives.

C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
1569
(1, 2, 3 and 4 in Scheme 1), some of them being strongly
deactivated amines such as 4,40-diaminodiphenylsulfone
(3) and 4,40-diaminobenzophenone (2).
N,N0-Bis-(tert-butoxycarbonyl)imidazoline-2-thione (6)
was prepared with di-tert-butyldicarbonate in THF
in the presence of sodium hydride in a 70±80% yield
(Scheme 2).
Two equivalents of 6 (three equivalents in the case of the
deactivated amines 2 and 3) reacted with one equivalent
of diamine in the presence of mercuric chloride and
excess of triethylamine, either in DMF or CH2Cl2, to
give, after ¯ash chromatography on silica, the corre-
sponding fully Boc-protected bis-(2-iminoimidazolidine)
derivatives (1c, 2c, 3c and 4c) in good to excellent yields
(see Table 1). The guanidine derivatives (1d, 2d, 3d, 4d)
were prepared using N,N0-bis(tert-butoxycarbonyl)-
thiourea (5).24
Scheme 2.
Molecular Modelling Results
Despite the fact that the tridimensional structure of the
a1-adrenoceptors is not known, several pharmacophores
have been developed for the antagonism of these recep-
tors.8 The most recent is that proposed by Bremner et
al.8f in 1996 which contains an aromatic ring, a basic N
atom and a polar group included or not in a non-aro-
matic ring. These authors suggest that an a1A-adreno-
ceptor antagonist should have the basic N at a distance
of 5.2±5.8 A from the aromatic ring and at a distance of
6±8 A from the polar group. On the contrary, an
antagonist of an a1B-adrenoceptor would have the basic
N atom at a distance of 6.2±7.8 A from the aromatic
ring and at 5±6 A from the polar group (Fig. 3).
An attempt to remove the Boc groups with SnCl4 led
directly to the hydrochloride salts of the products as
described recently.25 However, as revealed by elemental
analysis, the products that we obtained contained
`heavy metal' impurities, probably stannic complexes
1
that are not detectable in H and 13C NMR, and which
could not be removed by recrystallization. For that
reason and due to the incompatibility of heavy metals
from the pharmacological point of view, this method
was abandoned. Full deprotection of the Boc groups
was easily accomplished by treating the products with
an excess of tri¯uoroacetic acid, leading to the tri-
¯uoroacetate salts of the products. The hydrochloride
salts (1a, 2a, 3a and 4a) were obtained using strongly
basic anion-exchange resin.
We carried out the conformational analyses of the four
2-iminoimidazolidinium (1a, 2a, 3a, 4a) and four guani-
dinium derivatives (1b, 2b, 3b, 4b) by means of random
search techniques which combine the Monte Carlo
Scheme 1.

1570
C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
Table 1. Reaction solvents and time conditions used and melting point and yields obtained for compounds 1c±4c and 1d±4d
Product
R1
Reaction solvent
Time conditions
mp (ꢀC)
165±169
Isolated yield
1c
DMF
CH2Cl2
0 ꢀC/20 min rt/20 h
0 ꢀC/90 min rt/20 h
76%
61%
2c
3c
4c
DMF
5 ꢀC/1 h rt/5 days
0 ꢀC/1 h rt/24 h
106±108
140±141
90±93
43%
89%
a,b
CH2Cl2
CH2Cl2
DMF
0 ꢀC/30 min rt/23 h
0 ꢀC/15 min rt/4 h
68%
94%
1d
2d
3d
4d
DMF
0 ꢀC/20 min rt/20 h
0 ꢀC/30 min rt/5 days
0 ꢀC/1 h rt/24 h
130±135
142±143
121±123
118±122
66%
93%
86%
81%
CH2Cl2/DMFa,c
a,d
CH2Cl2
CH2Cl2
0 ꢀC/1 h rt
a3 equiv of reactants were used.
b0.62 equiv of unreacted N,N0-bis-(tert-butoxycarbonyl)imidazoline-2-thione were recovered.
cDMF was added to the reaction mixture to help solubilization of 4,40-diaminobenzophenone.
d0.3 equiv of unreacted N,N0-bis-(tert-butoxycarbonyl)imidazoline-2-thione were recovered.
methodology for the random generation of structures
and the molecular mechanics approach for the mini-
mization of the generated structures. In this method,
included in the SYBYL package,26 only di€erent con-
formations with di€erent optimized energies are stored
for their subsequent analysis. In all the molecules stud-
ied the distances between the possible elements of the
pharmacophore (aromatic rings, all basic N atoms and
polar centres) were measured and from those distances
the pharmacophore elements were localized.
method of distinguishing between imino and amino
forms in aryl iminoimidazolidines and aryl guanidines.
In the case of derivatives 1a and 1b we found that, for
all the conformers obtained, the -Nˆ atoms which con-
nect the aromatic rings with the extremities of the
molecules play the role of the basic N atom of the
pharmacophore and they are at a distance of about 5.6
A from the farthest aromatic rings (see Fig. 3 and Table
2). Moreover, these basic Ns are at 7.4 A from the cen-
tral NH groups which would correspond to the polar
centre of the pharmacophore.
It should be noticed that all the structures (1a±4a and
1b±4b) have been considered in their imino form (see
Fig. 2), which means that the N connecting the phenyl
rings with the extremities of the molecules is an sp2 N
atom, better described as an imino nitrogen (-Nˆ)
conjugated with the aromatic ring, and the most basic
of all the N atoms. This structural assumption is based
For the rest of the compounds the basic N atoms are the
same as for compounds 1a and 1b, the aromatic rings to
consider will also be the farthest from each of those
basic N atoms and the polar groups would be those
connecting the phenyl rings. Thus, for compounds 2a
and 2b the polar centre was a CO group, for compounds
3a and 3b an SO2 group, and for compounds 4a and 4b
we used the CH2 group. The distances between all the
pharmacophore elements are gathered in Table 2. These
1
on the H and 13C NMR results obtained for all the
compounds by comparison with those published by
Jackman and Jen.27 These authors characterized the
clonidine (Fig. 1) in its imino form and generalized a

C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
1571
Figure 3. (a) Pharmacophore proposed by Bremner et al.8f for a1A- and a1B-adrenoceptor antagonists. (b) General formula for the aromatic bis-
guanidinium (right side) and bis-imidazolinium (left side) derivatives 1a,b to 4a,b, indicating the pharmacophore elements and the ranges of dis-
tances between them.
Table 2. Ranges of distances (A) obtained for the conformations of minimum energy resulting from the random search of bis-2-iminoimidazoli-
dinium derivatives 1a±4a and bis-guanidinium derivatives 1b±4b to ful®l the pharmacophore conditions of Bremner et al.8f
Product
X
Na
Nb
1a
NH
Polar centre [N]
Ph (opposite)
Polar centre [N]
Ph (opposite)
5.579±5.600
7.575±7.470
5.578±5.600
7.488±7.535
5.579±5.600
7.471±7.541
5.578±5.600
7.487±7.535
1b
NH
2a
2b
CO
CO
Polar centre [C(O)]
Ph (opposite)
Polar centre [C(O)]
Ph (opposite)
5.685 (6.330±6.336)
7.801±7.817
5.684 (6.329±6.332)
7.809±7.816
5.685 (6.330±6.336)
7.801±7.817
5.684 (6.329±6.332)
7.808±7.815
3a
3b
SO2
SO2
Polar centre [S(O)(O)]
Ph (opposite)
Polar centre [S(O)(O)]
Ph (opposite)
5.941 (6.547±6.556)
7.280±7.323
5.940 (6.548±6.552)
7.310±7.320
5.941 (6.548±6.556)
7.281±7.323
5.940 (6.548±6.552)
7.310±7.319
4a
4b
CH2
CH2
Polar centre [C]
Ph (opposite)
Polar centre [C]
Ph (opposite)
5.708
5.708
7.313±7.360
5.707±5.714
7.584±7.339
7.314±7.359
5.707±5.714
7.585±7.339
results indicate that the conditions of the pharmaco-
phore are ful®lled by both ends of these molecules.
bis-iminoimidazolidinium derivatives 1a, 2a, 3a and 4a,
and bis-guanidinium derivatives 1b, 2b, 3b and 4b to
reduce the force of the contraction induced by nor-
adrenaline (NA) on isolated aorta was tested.
Pharmacological Results
To discard unspeci®ed e€ects not mediated by a-adre-
nergic receptor blockade, the e€ect of the active deriva-
tives was also evaluated on contractions induced by
KCl10b,29 on both rabbit and rat aorta. Their activity
The action of novel compounds with hypothetical a-
adrenergic antagonist activity was studied in isolated
rat and rabbit aortic rings.28 Thus, the capacity of the

1572
C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
was also tested on contractions elicited by administra-
tion of Des-Arg-bradykinin (B1 agonist) on rabbit
aorta, since this tissue contains B1 receptors.30
molecules, a basic N atom, an aromatic ring opposite to
each basic N atom, and a certain polar group in the
middle of the molecule. The conformational studies
carried out on all these compounds yielded a number of
minimum energy conformations, and the distances
between the di€erent elements of the Bremner pharma-
cophore (see Table 2) obtained for all these conforma-
tions correspond better to an a1A receptor subtype
(basic N at 5.2±5.8 A from the aromatic ring and at 6±8
A from the polar group).
E€ects mediated through activation or blockade of
other receptors were evaluated using myenteric plexus±
longitudinal muscle strips from guinea pig ileum. The
e€ect of new compounds on non-stimulated tissues and
on contractions induced by bradykinin (B2 receptors) or
by electrical stimulation was tested.
None of the tested compounds (1a to 4a and 1b, 2b and
4b) displayed intrinsic activity in the aortic preparations
or on the guinea pig ileum when administered alone.
From this result it could be suggested that these com-
pounds lack agonist activity on the receptors located on
these tissues: adrenergic, serotoninergic, cholinergic, of
bradykinin or histaminergic.
E€ects of the studied compounds
8
On resting tension. At concentrations between 10
and 10 4 M, the bis-imidazolinium derivatives 1a, 2a, 3a
and 4a had no e€ect on basal tension in rat and rabbit
aortic rings or on guinea pig ileum. The same results
were found with bis-guanidinium derivatives 1b, 2b, 3b
and 4b.
With regard to their functional antagonist activity in
inhibiting the contractile responses induced by a sub-
maximal concentration of NA (5Â10 8 M) in aortic
rings, the tested compounds, except the bis-guanidinium
derivative 3b, prevented the e€ect of NA in a dose-
dependent manner. It is well known that NA induces a
dose-dependent contraction through activation of a1-
receptors,10a,31 so that our results indicate that these
new compounds show `in vitro' a1 antagonist activity.
On noradrenaline-induced contractions. The contractions
induced by NA in rat and rabbit aortic rings were
inhibited in a dose-dependent manner by the bis-imida-
zolinium derivatives 1a, 2a, 3a and 4a as well as by the
bis-guanidinium derivatives 1b, 2b and 4b. The bis-gua-
8
nidinium derivative 3b at concentrations from 10 and
up to 10 4 M had no e€ect abolishing the contractions
induced by NA in both arteries. These inhibitions are
completely reversed after washing the preparations with
Krebs solution. The IC50 values obtained, in rat and
rabbit aorta, for the di€erent compounds are gathered
in Table 3. As a reference, the corresponding IC50
values of prazosin (a known a1 antagonist) were deter-
mined following the same procedure, the resulting value
On the other hand, the absence of antagonist activity in
guinea pig ileum or in blood vessel stimulated with KCl
or Des-Arg-bradykinin con®rms the selectivity of the
e€ect observed on contractions mediated through NA
receptors.
9
in rat aorta was 1.7Æ1.0 10 M (n=13), and in rabbit
8
aorta was 1.1Æ1.0 10 M (n=14).
As we mentioned in the Introduction, di€erent subtypes
of a1-receptors have been described (a1A, a1B, a1D, a1H
,
Speci®city assays. The contractions produced by the
addition of: (i) high concentrations of KCl (80 mM) in
rabbit and rat aorta; (ii) Des-Arg-bradykinin on rabbit
aorta; and (iii) by electrical stimulation in guinea pig
ileum, were not modi®ed in the presence of the com-
pounds studied.
a
1N, and a1L5,11a,32,33). The identity of the receptor sub-
types mediating contraction to a1-adrenoceptor agonists
in the vasculature, especially in the rat aorta, has been
controversial.34 36 Whereas Testa et al.34a described a1B
as the most common subtype in that tissue, recent
investigations indicate that the most common receptor
in the rat aorta is the a1D subtype.35,37
The potencies of the antagonists in this tissue parallel
its receptor binding anity of this subtype of the a1-
adrenoceptor population.34a Signi®cant heterogeneity
exists in a1-adrenoceptors in mammalian aorta. In the
rabbit, the coexistence of both a1A, a1D, and a1L-adre-
noceptors has been demonstrated.33b,34b,38 This hetero-
geneity in the rabbit aorta may explain the dissimilar
results obtained between rat and rabbit aortic rings in
response to the compounds studied in this work.
Discussion and Conclusion
All the compounds designed and synthesized (1a,b to
4a,b) exhibit all the elements of the pharmacophore
proposed by Bremner et al. by both extremities of the
Table 3. Concentrations needed for a 50% inhibition (IC50) of the
contraction provoked by noradrenaline in rat and rabbit aortic rings
Product
X
Rat aorta
IC50 (10 5 M)
n
Rabbit aorta
IC50 (10 5 M)
n
The di€erent potency observed in the antagonist activity
of these series of compounds in the rat aorta tests
(Table 3), where a single a1 subtype (B or D) has been
proposed, could be explained as follows. On the one
hand, the bis-2-iminoimidazolidinium derivatives (1a±
4a) are better antagonists than the bis-guanidinium
derivatives (1b±4b) as can be observed in Table 3. This
could probably be due to the additional hydrophobicity
1a
2a
3a
4a
1b
2b
4b
NH
CO
SO2
CH2
NH
CO
0.7Æ0.02
1.9Æ0.1
2.8Æ0.2
4.6Æ0.5
9.4Æ2.7
12Æ0.6
8.7Æ0.1
6
6
6
8
5
5
5
4.3Æ0.6
1.3Æ0.3
11Æ0.4
6
5
6
7
5
6
6
5.8Æ5.1
2.0Æ2.7
0.2Æ0.02
2.2Æ2.3
CH2

C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
1573
added to these systems by the ethylene portion of the
imidazolidine ring.
for compound 1a. Chemical shifts were internally refer-
enced to the residual proton resonance in CDCl3 (d
7.27 ppm) and D2O (d 4.6 ppm). Chemical shifts of the
13C NMR spectra in D2O were referenced with a capil-
lary of DMSO-d6 (d 39.5 ppm). IR spectra were recor-
ded on a Perkin±Elmer 681 spectrophotometer as KBr
pellets. Melting points were determined with a Reichert
Jung Thermovar apparatus and are uncorrected. Mass
spectra were recorded on a Hewlett±Packard Series 1100
MSD spectrometer (ES, APCI) and on a VG Autospec
spectrometer (FAB). FAB spectra were recorded in a NBA
matrix. Elemental analysis were performed on an Heraeus
CHN-O Rapid analyzer. Column chromatography was
performed on silica gel 60 (230±400 mesh ASTM, E.
Merck). All the reactions were monitored on aluminium
sheets precoated with silica gel 60 F254 (E. Merck).
On the other hand, and also in the rat aorta test, the
CH2 derivative, 4a, is the worse antagonist among the
bis-imidazolinium derivatives; this could be a con-
sequence of the lack of polarity of that group and,
therefore, the requisite of a polar group to ful®l the
pharmacophore would not be reached.
Therefore, even though some discrepancies are observed
between the molecular modelling results and the phar-
macological ones, in terms of the a1 subtype targeted by
this new family of compounds, it can be concluded that
these compounds show an interesting pro®le as antago-
nists of a1-adrenoceptors and that a new prototype,
compound 1a, has been found which deserves further
development.
Method A: preparation of Boc-protected
2-iminoimidazolidine derivatives. General procedure
To a 0.25 M solution of the appropriate diamine 1, 2, 3
or 4 (1 equiv), the bis-Boc-protected imidazolidine-
thione 6 (2.2 equiv for 1, 2 and 4, and 3 equiv for 3),
and an excess of Et3N, in DMF or CH2Cl2 at 0 ꢀC, were
added HgCl2 (same molar quantity as 6) at once. A
precipitate formed immediately and the resulting mixture
was stirred at that temperature for the time indicated in
Table 1. Then, the ice-bath was removed and the reaction
was stirred at room temperature for the appropriate
duration. The resulting dark grey reaction mixture was
diluted with EtOAc and ®ltered through a pad of Celite.
The ®lter cake was rinsed with EtOAc. The organic
phase was washed with brine, then dried over MgSO4
and ®nally concentrated under vacuum. The crude prod-
uct was puri®ed by ¯ash chromatography on silica gel.
Experimental
Random search conditions
The following conditions were established in all the con-
formational analyses performed using the random search
method and starting from two di€erent conformations:
i. Bonds to search: only the C(arom)±X bonds [X=
N(guanidinium, 2-imino-imidazolidinium), N(H), C(H2),
C(O), S(O2)] were rotated.
ii. Energy set-up: atomic charges were evaluated by the
Gasteiger±Huckel method.
iii. Minimization details: each generated conformer was
minimized during 300 cycles using the Conjugate Gra-
dient method.
N,N0-Di(tert-butoxycarbonyl)imidazolidine-2-thione (6).
To a cooled solution (ice-water bath) of imidazolidine-
2-thione (4.21 g, 41.3 mmol) in dry THF (600 mL) under
argon, 7.49 g (187 mmol) of NaH (60% in mineral oil)
were added. After 5 min, the ice-bath was removed and
the reaction was stirred 10 min at room temperature.
Then, the reaction mixture was cooled until 0 ꢀC and di-
tert-butyldicarbonate (19.88 g, 91.2 mmol) was added
neat. After 30 min, the ice-bath was removed and the
reaction mixture was stirred another 4 h at room tem-
perature. The reaction was quenched carefully, drop-by-
drop, with a saturated NaHCO3 solution (100 mL). The
mixture was poured into 300 mL of water and the aqu-
eous layer was extracted with EtOAc (3Â150 mL). The
combined organic phases were washed with brine, dried
over MgSO4 and concentrated under vacuum. Recrys-
tallization from hexane:EtOAc (8:2) a€orded 5 as
yellow needles (9 g, 72% yield): mp 117±119 ꢀC; IR (KBr)
2980, 1745, 1720, 1370, 1270, 1150, 770 cm 1; 1H NMR
(CDCl3) d 3.84 (s, 4H), 1.46 (s, 18H); 13C NMR
(CDCl3) d 176, 150.65, 84.24, 44.99, 28.46; anal. calcd
for C13H22 N2O4S: C, 51.63; H, 7.33; N, 9.26; S, 10.60.
Found: C, 51.97; H, 7.38; N, 9.40; S, 10.31.
iv. Random search details: the maximum number of
cycles in the search is set to 6000; the energy cut-o€ is
set to 10 kcal/mol; the RMS-threshold is set to 0.2; the
maximum number of hits is set to 6; and the data con-
vergence is set to 0.005.
For each starting conformation of each derivative, 15±25
di€erent minimum energy conformations were obtained
and considered in the study of the pharmacophore
requirements.
Chemistry
Reagents were used as received except for 4,40-
(diaminodiphenyl)amine, which was generated from its
sulfate precursor by basi®cation with an aqueous solu-
tion of sodium hydroxide (10% weight) followed by
extraction with diethyl ether. THF was distilled from
sodium±benzophenone under argon. Other reaction sol-
vents were purchased anhydrous and used as received.
Other solvents used were reagent grade. 1H and 13C
NMR spectra were recorded at 200 and 50 MHz,
respectively, on a Varian Gemini 200 spectrometer and at
400 and 100 MHz on a Varian 400 Unity spectrometer
4,40-Bis[1,3-di(tert-butoxycarbonyl)-2-imidazolidinylimino]-
diphenylamine (1c). A solution of 99 mg (0.5 mmol) of 1,
334 mg (1.1 mmol) of 6 and 0.5 mL (7 equiv) of Et3N in

1574
C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
2 mL of CH2Cl2 was treated with 308 mg (1.1 mmol) of
HgCl2 and stirred ®rst 1 h 45min at 0 ꢀC, and then 20h at
room temperature. Usual work up and ¯ash-chromato-
graphy (hexane:EtOAc, 1.5:1) gave 1c as a brown solid
(226 mg, 61% yield); mp 165±169 ꢀC; IR (KBr) 2990, 1755,
1700, 1510, 1380, 1310,1155, 980 cm 1; 1H NMR (CDCl3)
d 6.88 (d, 4H, J=8.34 Hz), 6.84 (d, 4H, J=8.93 Hz),
3.79 (s, 8H), 1.32 (s, 36H); 13C NMR (CDCl3) d 150.91,
142.08, 139.67, 138.98, 122.94, 118.61, 83.08, 43.62,
28.43; anal. calcd for C38H53N7O8.1/2H2O : C, 61.27; H,
7.31; N, 13.16. Found: C, 60.92; H, 7.21; N, 12.71.
N,N0-Di(tert-butoxicarbonyl)thiourea (5). That com-
pound was synthesized in 70% yield following the pro-
cedure described in ref 21.
4,40 - Bis[2,3 - di(tert - butoxycarbonyl)guanidino]diphenyl -
amine (1d). Method B: ¯ash-chromatography (cyclo-
hexane:EtOAc, gradient 9:1 to 2.5:1) gave 1d as an o€-
white foam (66%); IR (KBr) 2950, 1708, 1628, 1395,
1220, 1140 cm 1; 1H NMR (CDCl3) d 11.6 (broad, NH),
10.12 (broad, NH), 7.32 (d, 4H, J=8.8 Hz), 6.89 (d, 4H,
J= 8.6 Hz), 1.53 (s, 18H), 1.45 (s, 18H); 13C NMR
(CDCl3) d 164.13, 154.41, 153.80, 141.25, 129.57,
124.54, 118.11, 84.94, 79.90, 28.60, 28.51; anal. calcd for
C34H49N7O8: C, 59.72; H, 7.22; N, 14.34. Found: C,
59.68; H, 6.94; N, 13.98.
4,40-Bis[1,3-di(tert-butoxycarbonyl)-2-imidazolidinylimino]-
diphenylketone (2c). A solution of 159 mg (0.75 mmol)
of 2, 470 mg (1.56 mmol) of 6 and 0.5 mL (5 equiv) of
Et3N in 4 mL of DMF was treated with 450 mg
(1.57 mmol) of HgCl2 and stirred ®rst 1 h at 5 ꢀC, and
then 24 h at room temperature. Usual work up and ¯ash-
chromatography (cyclohexane:EtOAc, 2:1) gave 2c as a
colourless solid (198 mg, 43% yield); mp 106±108 ꢀC; IR
(KBr) 2990, 1760, 1705, 1595, 1310, 1150, 978 cm 1; 1H
NMR (CDCl3) d 7.60 (d, 4H, J=8.4Hz), 6.93 (d, 4H,
J=8.4Hz), 3.78 (s, 8H), 1.27 (s, 36H); 13C NMR (CDCl3)
d 195.39, 152.93, 150.36, 140.63, 132.5, 131.56, 121.2,
83.43, 43.73, 28.33; anal. calcd for C39H52N6O9: C, 62.55;
H, 7.00; N, 11.22. Found: C, 62.61; H: 7,28; N, 11.15.
4,40 - Bis[2,3 - di(tert - butoxycarbonyl)guanidino]diphenyl -
ketone (2d). Method B: ¯ash-chromatography (hexane:
EtOAc, 5:1) gave 2d as a colourless foam (93%); IR
(KBr) 2990, 2940, 1728, 1655, 1415, 1310, 1244,
1160 cm 1; 1H NMR (CDCl3) d 11.63 (broad, NH), 10.6
(broad, NH), 7.79 (d, 4H, J=8.8 Hz), 7.76 (d, 4H, J=
8.8 Hz), 1.55 (s, 18H), 1.52 (s, 18H); 13C NMR (CDCl3)
d 194.52, 163.26, 153.24, 140.70, 133.63, 131.19, 120.99,
84.07, 79.96, 28.08; anal. calcd for C35H48N6O9: C,
60.33; H, 6.94; N, 12.06. Found: C, 60.11; H: 6.81; N,
11.97.
4,40-Bis[1,3-di(tert-butoxycarbonyl)-2-imidazolidinylimino]-
diphenylsulfone (3c). A solution of 131 mg (0.53 mmol)
of 3, 484 mg (1.6 mmol, 3 equv) of 6 and 0.5 mL (7
equiv) of Et3N in 2 mL of CH2Cl2 was treated with
495 mg (3.4 equiv) of HgCl2 and stirred ®rst 45 min at
0 ꢀC, and then 15 h at room temperature. Usual work up
and ¯ash-chromatography (hexane:EtOAc, 1:1) gave 3c
as a colourless solid (371 mg, 89% yield); mp 140±
141 ꢀC; IR (KBr) 2940, 1747, 1740, 1690, 1566, 1350,
4,40-Bis[2,3-di(tert-butoxycarbonyl)guanidino]diphenylsul-
fone (3d). Method B: ¯ash-chromatography (CH2Cl2)
gave 3d as a colourless solid (86%); IR (KBr) 2940,
1
1705, 1628, 1575, 1390, 1304, 1223, 1135 cm
;
1H
NMR (CDCl3) d 11.6 (broad, NH), 10.61 (broad, NH),
7.87 (d, 4H, J=8 Hz), 7.82 (d, 4H, J=8 Hz), 1.54 (s,
18H), 1.51 (s, 18H); 13C NMR (CDCl3) d 163.38, 153.53,
141.61, 137.08, 128.97, 122.07, 84.61, 80.38, 28.42,
28.34; anal. calcd for C34H48N6O10S: C, 55.72; H, 6.60;
N, 11.47; S, 4.37. Found: C, 55.58; H, 6.48; N, 11.25; S,
4.15.
1
1280, 1130, 960 cm 1; H NMR (CDCl3) d 7.56 (d, 4H,
J=8.7 Hz), 6.81 (d, 4H, J=8.7 Hz), 3.68 (s, 8H), 1.13 (s,
36H); 13C NMR (CDCl3) d 153.54, 149.95, 141.20,
135.23, 128.48, 121.62, 83.34, 43.54, 28.05; anal. calcd
for C38H52N6O10S: C, 58.15; H, 6.68; N, 10.70; S, 4.08.
Found: C, 57.97; H, 6.47; N, 10.59; S, 3.95.
4,40 -Bis[2,3-di(tert-butoxycarbonyl)guanidino]diphenyl-
methane (4d). Method B: ¯ash-chromatography (hex-
ane:EtOAc, 9:1) gave 4d as a colourless foam (81%); IR
4,40-Bis[1,3-di(tert-butoxycarbonyl)-2-imidazolidinylimino]-
diphenylmethane (4c). A solution of 102 mg (0.5 mmol)
of 4, 400 mg (1.3 mmol) of 6 and 0.5 mL (7 equiv) of
Et3N in 2 mL of CH2Cl2 was treated with 380 mg
(1.35 mmol) of HgCl2 and stirred ®rst 1 h at 0 ꢀC, and
then 23 h at room temperature. Usual work up and
¯ash-chromatography (hexane:EtOAc, 2:1) gave 4c as a
colourless solid (302 mg, 81% yield); mp 90±93 ꢀC; IR
(KBr) 2950, 1705, 1627, 1395, 1135 cm
;
1H NMR
1
(CDCl3) d 11.3 (broad, NH), 10.26 (broad, NH), 7.49
(d, 4H, J=8.5 Hz), 7.11 (d, 4H, J=8.6 Hz), 3.9 (s, 2H),
1.52 (s, 18H), 1.49 (s, 18H); 13C NMR (CDCl3) d
164.22, 154.11, 153.91, 138.15, 135.43, 129.97, 122.91,
84.21, 80.12, 41.35, 28.79, 28.72; anal. calcd for C35H50
N6O8 : C, 61.58; H, 7.33; N, 12.31. Found: C, 61.30; H,
7.27; N, 12.18.
1
(KBr) 2990, 1760, 1705, 1370, 1310, 1155, 978 cm 1; H
NMR (CDCl3) d 7.00 (d, 4H, J=8.4 Hz), 6.88 (d, 4H,
J=8.4 Hz), 3.81 (bs, 10H), 1.31 (s, 36H); 13C NMR
(CDCl3) d 150.79, 146.61, 139.46, 136.25, 129.61, 121.83,
83.11, 43.57, 41.27, 28.35; anal. calcd for C39H54N6O8: C,
63.74; H, 7.40; N, 11.43. Found: C, 63.51; H, 7.15; N, 11.30.
General procedure for the generation of the
hydrochloride salts
The Boc-protected products were treated either with an
excess of CF3COOH neat or a 50% solution in CH2Cl2
and the resulting solution stirred at room temperature
(in general, the reaction was complete after only
30 min). Excess of CF3COOH was removed under
vacuum and the tri¯uoroacetate salt lyophilized over-
night. The hydrochloride salt was generated by stirring
an aqueous solution of the tri¯uoroacetate salt with
Method B
Preparation of Boc-protected guanidines (1d, 2d, 3d, 4d)
using compound 5 and following the same procedure as
described above.

C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
1575
.
IRA400 Amberlyte resin in its Cl form during 24 h at
room temperature. The resin was removed by ®ltration
and the aqueous solution washed twice with EtOAc.
Lyophilization a€orded the pure hydrochloride salt.
Absence of tri¯uoroacetate salt was checked by 13C
NMR.
100%); anal. calcd for C14H18Cl2N6O2S 2H2O: C, 38.10;
H, 5.02; N, 19.04; S, 7.26. Found: C, 37.83; H, 4.76; N,
19.19; S, 6.82.
Dihydrochloride salt of 4,40-diguanidinodꢀiphenylmethane
(4b). White solid (81%); mp 199±202 C (lit.40 199±
200 ꢀC); 1H NMR (D2O) d 7.21 (d, 4H, J=8.4 Hz), 7.07
(d, 4H, J=8.4 Hz), 3.88 (s, 2H); 13C NMR (D2O) d
155.45, 140.39, 131.21, 129.22, 125.31, 39.48; MS
(FAB+): 283 (MH+/100%); anal. calcd for C15H20
Hydrochloride salt of 4,40-di(2-imidazolidinylimino)di-
phenylamine (1a). Grey solid (87%); mp 165±170 ꢀC;
1H NMR (100 MHz/D2O) d 7.11 (d, 4H, J=8.94 Hz),
7.07 (d, 4H, J=9.09 Hz), 3.62 (s, 8H); 13C RMN (D2O)
d 159.43, 142.69, 128.23, 126.47, 118.80, 43.02; MS
(FAB+): 336 (MH+/100%); anal. calcd for C18H23
.
Cl2N6 2.5H2O: C, 45.01; H, 6.29; N, 20.99. Found: C,
45.53; H, 6.12; N, 20.92.
.
Cl2N7 2H2O: C, 48.65; H, 6.12; N, 22.06. Found: C,
48.88; H, 5.81; N, 21.60.
Pharmacology: materials and methods
Rabbit and rat aorta. Male Wistar rats weighing 250±
300 g and rabbits (New Zealand) weighing 2.5±3 kg were
killed by decapitation. Descending thoracic aortas
were removed and placed in a Krebs physiological
solution.
Hydrochloride salt of 4,40-di(2-imidazolidinylimino)di-
phenylketone (2a). White solid (90%); mp>159 ꢀC; H
1
NMR (D2O) d 7.63 (d, 4H, J=8.6 Hz), 7.20 (d, 4H,
J=8.6 Hz), 3.62 (s, 8H); 13C RMN (D2O) d 196.96,
157.0, 139.05, 133.28, 131.12, 121.37, 41.83; MS
(FAB+): 349 (MH+/100%); anal. calcd for C19H22
Cl2N6O.3H2O: C, 48.00; H, 5.94; N, 17.68. Found: C,
48.34; H, 5.73; N, 17.73.
The excess of surrounding tissue was removed and the
arteries were cut into rings and mounted in organ baths
(under 1 g of tension for rats tissues and 2 g for rabbits
tissues) containing 5 mL of Krebs, maintained at 37 ꢀC,
and bubbled with 95% O2 and 5% of CO2 gas mixture.
The aortas were allowed to equilibrate during 1 h.28a
Hydrochloride salt of 4,40-di(2-imidazolidinylimino)di-
phenylsulfone (3a). White solid (85%); mp 180±185 ꢀC;
1H NMR (D2O) d 7.83 (d, 4H, J=8.71 Hz), 7.27 (d, 4H,
J=8.87 Hz), 3.59 (s, 8H); 13C RMN (D2O) d 158.95,
142.1, 137.78, 130.50, 124.25, 43.94; MS (ES+): 385
(MH+/100%); anal. calcd for C18H22Cl2N6O2S.3H2O:
C, 42.27; H, 5.52; N, 16.43; S, 6.27. Found: C, 41.98; H,
5.53; N, 16.43; S, 6.15.
Isometric tension was recorded in a polygraph (Grass
Mod. 7D) through force±displacement transducers (Grass
FT07).
After the equilibration period, the arteries were pre-
contracted with NA (5Â10 8 M) until two successive
responses were almost identical in height. This was fol-
lowed by exposure to the new compounds for 5 min
prior to the addition of NA.10b After 20 min the prep-
arations were washed and left to recover the basal
response before testing the next doses. Concentration±
response curves of the derivatives were constructed
(10 8±10 4 M). Only one antagonist was used in each
experiment.
Hydrochloride salt of 4,40-di(2-imidazolidinylimino)di-
phenylmethane (4a). Brown solid (81%); mp 140±
145 ꢀC; 1H NMR (D2O) d 7.07 (d, 4H, J=8 Hz), 6.92 (d,
4H, J=8 Hz), 3.71 (s, 2H), 3.46 (s, 8H); 13C RMN
(D2O) d 158.96, 140.97, 133.45, 130.43, 124.74, 43.04,
40.5; MS (FAB+): 335 (MH+/100%); anal. calcd for
.
C19H24Cl2N6 3.2H2O: C, 49.08; H, 6.59; N, 18.07.
Found: C, 48.94; H, 6.59; N, 18.24.
Dihydrochloride salt of 4,40-diguanidinodiphenylamine
The e€ect on the basal tension was also studied. Vaso-
constriction produced by KCl (80 mM) was tested after
every experiment.29 Inhibition of contraction produced
by NA was measured and was expressed as percentage.
The IC50 of each compound was calculated.
(1b). Grey solid (90%); mp>170 ꢀC (dec.); H NMR
1
(D2O) d 6.96 (m); 13C NMR (D2O) d 157.04, 143.27,
128.05, 127.06, 119.05; MS (FAB+): 284 (MH+/100%);
.
anal. calcd for C14H19Cl2N7 1.5H2O: C, 43.87; H, 5.78;
N, 25.58. Found: C, 43.88; H, 5.39; N, 25.19.
To discard e€ects mediated through activation of B1
receptors the e€ects of the new compounds on rabbit
aorta after an equilibrium period of 5 h to induce B1
receptors11a were tested. Control contractions were
elicited by repeated administration of Des-Arg9-brady-
kinin, then the derivatives (10 8±10 4 M) were added to
the organ bath 10 min before the following administra-
tion of Des-Arg9-bradykinin (3Â10 7 M).
Dihydrochloride salt of 4,40-diguanidinodiphenylketone
(2b). White solid (90%); mp 149±152 ꢀC; 1H NMR
(D2O) d 7.64 (d, 4H, J=8.4 Hz), 7.25 (d, 4H, J=
8.2 Hz); 13C NMR (D2O) d 195.6, 153.28, 136.74,
132.22, 129.44, 121.57; MS (FAB+): 297 (MH+/100%);
.
anal. calcd for C15H18Cl2N6O 1.3H2O: C, 45.88; H,
5.24; N, 21.40. Found: C, 45.92; H, 4.98; N, 21.40.
Dihydrochloride salt of 4,40-diguanidinodiphenylsulfone
Guinea pig isolated ileum. Guinea pigs of either sex
(300±400 g) were used. Myenteric±plexus±longitudinal
muscle (MP±LM) strips were isolated from distal ileum
as previously described41 and suspended in a 5 mL
organ bath containing Krebs solution at 35 ꢀC aerated
(3b). White solid (85%); mp 247±250 ꢀC (lit.39 263±
265 ꢀC); H NMR (D2O) d 7.85 (d, 4H, J=8.71 Hz),
1
7.32 (d, 4H, J=8.56 Hz); 13C NMR (D2O) d 156.81,
141.63, 138.19, 130.52, 125.88; MS (FAB+): 333 (MH+/

1576
C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
with 95% O2 and 5% CO2. Tissues were maintained
under a resting tension of 1 g, and the contractile activ-
ity of the strips was recorded isometrically using an
Omniscribe polygraph D-5000. After an equilibration
period of 20 min following sets of experiments were
carried out.
8. (a) De Marinis, R. M.; Wise, M.; Hieble, J. P.; Ru€olo, R.
R. In The Alpha-1 Adrenergic Receptor; Ru€olo, R. R., Ed.;
Humana Press: Clifton, 1987. (b) Ru€olo, R. R.; Bondinel,
W.; Hieble, J. P. J. Med. Chem. 1995, 38, 3416. (c) Rastelli, G.;
De Benedetti, P. G. J. Mol. Struct. 1991, 251, 307. (d) Ven-
turelli, M; Menziani, M. C.; Cocchi, M.; Fanelli, F.; De Ben-
edetti, P. G. J. Mol. Struct. 1992, 276, 327. (e) Cocchi, M.;
Menziani, M. C.; Fanelli, F.; De Benedetti, P. G. J. Mol.
Struct. 1995, 331, 79. (f) Bremner, J. B.; Coban, B.; Grith, R.
J. Comput.-Aided Mol. Des. 1996, 10, 545. (g) Grith, R.;
Bremner, J. B. J. Comput.-Aided Mol. Des. 1999, 13, 69.
9. Kim, K. S.; Qian, L. Tetrahedron Lett. 1993, 34, 7677.
10. (a) Furchgott, R. F.; Bradrakom, S. J. Pharmacol. Exp.
Ther. 1953, 108, 129. (b) Duarte, J.; Perez-Vizcaõno, F.; Zar-
zuelo, A.; Jimenez, J.; Tamargo, J. Eur. J. Pharmacol. 1993,
239, 1.
8
Agonist activity. All the synthesized compounds (10
to 5Â10 6 M) were added to the organ bath in order to
discard agonist activity; each concentration was left in
the organ bath a 10 min period.
Antagonist activity. The e€ect of all the derivatives on
the contractions induced by electrical stimulation (single
square waves: 0.3 Hz, 2 ms and supramaximal voltage)
or by bradykinin (5Â10 7±5Â10 6 M) on MP±LM
11. (a) Bouthillier, J.; Deblois, D.; Marceau, F. Br. J. Phar-
macol. 1987, 92, 257. (b) Regoli, D.; Gobeil, F.; Nguyen, Q.
T.; Jukic, D.; Seoane, P. R.; Salvino, J. M.; Sawutz, D. J. Life
Sci. 1994, 55735.
strips was tested. In electrically stimulated tissues the
8
new compounds were added to the organ bath at 10
±
5Â10 6 M concentrations. The bradykinin was applied
repeatedly at 10 min intervals and 1 min after each
addition the bath solution was replaced. When a stable
contractile response was reached the studied com-
pounds (10 8±5Â10 7 M) were added to the organ
bath 10 min before the following dose of bradykinin.
Only strips producing reproducible responses (>0.5 g)
were used, the last control response was taken as 100
and subsequent results obtained were expressed as a
percentage of this.
12. (a) Paton, W. D. M. Br. J. Pharmacol. Chemother. 1957,
12, 119. (b) Cousins, H. M.; Edwards, F. R.; Hirst, G. D. S.;
Wendt, I. R. J. Physiol. 1993, 471, 61. (c) Munro, A. F. J.
Physiol. 1953, 120, 41. (d) Wirth, K.; Hock, F. J.; Albus, U.;
Linz, W.; Alpermann, H. G.; Anagnostopoulus, H.; Heenke,
S.; Breiphol, G. Br. J. Pharmacol. 1991, 102, 774.
13. McKay, A. F.; Buchanan, M. N.; Grant, G. A. J. Am.
Chem. Soc. 1949, 71, 766.
14. (a) Muche, M.-S.; Gobel, M. W. Angew. Chem. Int. Ed.
Engl. 1996, 35, 2126. (b) Gluchowski, C. US Patent 5 130 441
A, 1992.
15. Short, J. H.; Biermacher, U.; Dunnigan, D. A.; Leth, T. D.
J. Med. Chem. 1963, 6, 275.
16. (a) Kneeland, D. M.; Ariga, K.; Lynch, V. M.; Huang, C.-
Y.; Anslyn, E. V. J. Am. Chem. Soc. 1993, 115, 10042. (b)
Perreault, D. M.; Chen, X.; Anslyn, E. V. Tetrahedron 1995,
51, 353.
Drugs. The following drugs were used: noradrenaline
bitartrate, bradykinin, Des-Arg-bradykinin (Sigma
Chemical Co.). All drugs were dissolved in distilled
deionized water at 10 2 M stock solution.
17. Ariga, K.; Anslyn, E. V. J. Org. Chem. 1992, 57, 417.
18. Timmermans, P. B. M. W. M.; Van Zwieten, P. A.;
Speckamp, W. N. Rec. Trav. Chim. Pays-Bas 1978, 97, 51.
19. Rouot, B.; Leclerc, G. Bull. Soc. Chim. Fr. 1979, 9±10, 520.
20. Huber, D.; Leclerc, G.; Ehrhardt, J. D.; Andermann, G.
Tetrahedron Lett. 1987, 28, 6453.
21. Yong, Y. F.; Kowalski, J. A.; Lipton, M. A. J. Org. Chem.
1997, 62, 1540.
22. Inoe, H.; Matsumara, N. Japan Patent: Tokuyama Kk, JP
94-42320, 14 March 1994.
Acknowledgements
The authors thank the Spanish DGICYT (Project No.
SAF 97-0044) for ®nancial support. One of us (CD) is a
holder of a Marie Curie Research Training Grant for
which he thanks the European Commission, Directorate
General XII (Science, Research and Development).
23. Matsumara, N.; Noguchi, A.; Kitayoshi, A.; Inoue, H. J.
Chem. Soc. Perkin Trans. 1 1995, 2953.
24. Iwanowicz, E. J.; Poss, M. A.; Lin, J. Synthetic Comm.
1993, 23, 1443.
References
1. (a) Fornai, F.; Blandizzi, C.; del Tacca, M. Pharmacol. Res.
1990, 22, 541. (b) Piascik, M. T.; Soltis, E. E.; Piascik, M. M.;
Macmillan, L. B. Pharmacol. Ther. 1996, 72, 215.
25. Miel, H.; Rault, S. Tetrahedron. Lett. 1997, 38, 7865.
26. SYBYL 6.4. Copyright 1991±1997 Tripos, Inc. Molecular
Silverware copyright (c) 1990 Rohm and Haas Company.
27. Jackman, L. M.; Jen, T. J. Am. Chem. Soc. 1975, 97, 2811.
28. (a) Furchgott, R. F. Fed. Proc. 1952, 11, 217. (b) Perez-
Vizcaõno, F.; Fernandez, C.; Fernandez del Pozo, B.; Cogol-
ludo, A.; Zaragoza, F.; Tamargo, J. J. Cardiovasc. Pharmacol.
1998, 32, 745. (c) Cogolludo, A.; Perez-Vizcaõno, F.; Fajardo,
S.; Ibarra, M.; Tamargo, J. Br. J. Pharmacol. 1999, 126, 1025.
29. Godfraind, T.; Miller, R.; Wibbo, M. Pharmacol. Rev.
1986, 38, 324.
30. Marceau, F. Inmunopharmacology 1995, 30, 1.
31. Ruiz, E.; Tejerina, T. Br. J. Pharmacology 1998, 125, 186.
32. (a) Ru€olo, R. R.; Nichols, A. J.; Stadel, J. M.; Hieble, J.
P. Pharmacol. Rev. 1991, 43, 475. (b) Bylund, D. B. FASEB J.
1992, 6, 832. (c) Graham, R. M.; Perez, D. M.; Hwa, J.; Pias-
cik, M. K. Circ. Res. 1996, 78, 737. (d) Graham, R. M.; Perez,
D. M.; Hwa, J.; Piascik, M. K. TIPS 1999 (Suppl.), 11.
2. Anyuovsky, E. P.; Rosen, M. R. Circulation 1991, 2076.
3. Matyus, P.; Horvath, K. Med. Res. Rev. 1997, 17, 523.
4. Barlocco, D.; Cignarella, G.; Montesane, F.; Leonardi, A.;
Mella, M.; Toma, L. J. Med. Chem. 1999, 42, 173.
5. (a) Ford, A. P. D. W.; Williams, T. J.; Blue, D. R.; Clarke,
D. E. Trends Pharmacol. Sci. 1994, 15, 167. (b) Hieble, J. P.;
Bylund, D. B.; Clarke, A. E.; Eikenberg, D. C.; Langer, S. Z.;
Lefkowitz, R. J.; Minneman, K. P.; Ru€olo, R. R. Pharmacol.
Rev. 1995, 47, 267. (c) Hieble, J. P.; Ru€olo, R. R. Prog. Drug
Res. 1996, 47, 81.
6. Amemiya, Y.; Hong, S. S.; Venkataraman, B. V.; Patil, P.
N.; Shams, G.; Romstedt, K.; Feller, D. R.; Hsu, F.-L.;
Miller, D. D. J. Med. Chem. 1992, 35, 750.
7. Dardonville, C.; Rozas, I.; Alkorta, I. J. Mol. Graphics
Mod. 1999, 16, 150.

C. Dardonville et al. / Bioorg. Med. Chem. 8 (2000) 1567±1577
1577
33. (a) Faure, C.; Pimoule, C.; Arbilla, S.; Langer, S. Z.; Gra-
ham, D. Eur. J. Pharmacol. 1994, 268, 141. (b) Docherty, J. R.
Eur. J. Pharmacol. 1998, 36, 1.
34. (a) Testa, R.; Guarneri, L.; Poggesi, E.; Simonazi, I.; Tad-
dei, C.; Leonardi, A. Br. J. Pharmacol. 1995, 114, 745. (b)
Muramatsu, I.; Omura, T.; Kicoshi, S.; Hashimoto, S.; Oshita,
M. Br. J. Pharmacol. 1990, 99, 197.
1990, 190, 97. (b) Kong, J. Q.; Taylor, D. A.; Fleming, W. W.
J. Pharmacol. Exp. Ther. 1994, 268, 1153. (c) Piascik, M. T.;
Sparks, M. S.; Pruit, T. A.; Soltis, E. E. Eur. J. Pharmacol.
1991, 199, 279. (d) Wenham, D.; Marshall, I. Br. J. Pharma-
col. 1994 (Proc. Suppl.), 112.
37. Hussain, M. B.; Marshall, I. Br. J. Pharmacol. 1997, 122,
849.
35. (a) Kenny, B. A.; Chalmers, D. H.; Philpott, P. C.; Naylor,
A. M. Br. J. Pharmacol. 1995, 115, 981. (b) Quaglia, W.;
Pigini, M.; Piergentili, A.; Giannella, M.; Marucci, G.; Pog-
gesi, E.; Leonardi, A.; Melchiorre, C. J. Med. Chem. 1999, 42,
2961.
38. Oriowo, M. A.; Ru€olo, Jr., R. R. J. Vasc. Res. 1992, 29,
33.
39. Kazumi, F. J. Pharm. Soc. Japan 1945, 65, 4.
40. Braun, C. E.; Erit, J. D.; Crooks, G. C. J. Org. Chem.
1938, 3, 146.
36. (a) Han, C.; Li, J.; Minneman, K. P. Eur. J. Pharmacol.
41. Ambache, N. J. Physiol. 1954, 125, 53.