Universal template approach to drug design: Polyamines as selective muscarinic receptor antagonists

Article abstract of DOI:10.1021/jm981038d

The concept that polyamines may represent a universal template in the receptor recognition process is embodied in the design of new selective muscarinic ligands. Tetraamines 4-7 and 16-20 and diamine diamides 8-15 were synthesized, and their pharmacological profiles at muscarinic receptor subtypes were assessed by functional experiments in isolated guinea pig left atrium (M₂) and ileum (M₃) and by binding assays in rat cortex (M₁), heart (M₂), submaxillary gland (M₃), and NG 108-15 cells (M₄). It has been confirmed that appropriate substituents on the terminal nitrogens of a tetraamine template can tune both affinity and selectivity for muscarinic receptors. The novel tetraamine C-tripitramine (17) was able to discriminate significantly M₁ and M₂ receptors versus the other subtypes, and in addition it was 100-fold more lipophilic than the lead compound tripitramine. Compound 14 (tripinamide), in which the tetraamine backbone was transformed into a diamine diamide one, retained high affinity for muscarinic subtypes; displaying a binding affinity profile (M₂ > M₁ > M₄ > M₃) qualitatively similar to that of tripitramine. Both these ligands, owing to their improved lipophilicity relative to tripitramine and methoctramine, could serve as tools in investigating cholinergic functions in the central nervous system. Furthermore, notwithstanding the fact that the highest affinity was always associated with muscarinic M₂ receptors, for the first time polyamines were shown to display high pA₂ values also toward muscarinic M₃ receptors.

Full text of DOI:10.1021/jm981038d

4150
J . Med. Chem. 1998, 41, 4150-4160
Un iver sa l Tem p la te Ap p r oa ch to Dr u g Design : P olya m in es a s Selective
Mu sca r in ic Recep tor An ta gon ists
Maria L. Bolognesi,† Anna Minarini,† Roberta Budriesi,† Silvia Cacciaguerra,‡ Alberto Chiarini,†
Santi Spampinato,‡ Vincenzo Tumiatti,† and Carlo Melchiorre*,†
Department of Pharmaceutical Sciences, University of Bologna, Via Belmeloro 6, 40126 Bologna, Italy, and Department of
Pharmacology, University of Bologna, Via Irnerio 48, 40126 Bologna, Italy
Received May 29, 1998
The concept that polyamines may represent a universal template in the receptor recognition
process is embodied in the design of new selective muscarinic ligands. Tetraamines 4-7 and
16-20 and diamine diamides 8-15 were synthesized, and their pharmacological profiles at
muscarinic receptor subtypes were assessed by functional experiments in isolated guinea pig
left atrium (M2) and ileum (M3) and by binding assays in rat cortex (M1), heart (M2),
submaxillary gland (M3), and NG 108-15 cells (M4). It has been confirmed that appropriate
substituents on the terminal nitrogens of a tetraamine template can tune both affinity and
selectivity for muscarinic receptors. The novel tetraamine C-tripitramine (17) was able to
discriminate significantly M1 and M2 receptors versus the other subtypes, and in addition it
was 100-fold more lipophilic than the lead compound tripitramine. Compound 14 (tripinamide),
in which the tetraamine backbone was transformed into a diamine diamide one, retained high
affinity for muscarinic subtypes, displaying a binding affinity profile (M2 > M1 > M4 > M3)
qualitatively similar to that of tripitramine. Both these ligands, owing to their improved
lipophilicity relative to tripitramine and methoctramine, could serve as tools in investigating
cholinergic functions in the central nervous system. Furthermore, notwithstanding the fact
that the highest affinity was always associated with muscarinic M2 receptors, for the first time
polyamines were shown to display high pA2 values also toward muscarinic M3 receptors.
Ch a r t 1
In tr od u ction
The identification of multiple muscarinic receptor
subtypes has stimulated the search for ligands with
selectivity for a given receptor subtype.1 Five different
subtypes (m1-m5) have been identified so far by mo-
lecular cloning. Muscarinic receptors that have been
characterized pharmacologically and classified as M1-
M4 appear to correspond to cloned m1-m4 receptors. At
present, little information is available about the nature
and the cellular location of the m5 subtype.1
Nonselective muscarinic receptor antagonists have
long been used for treatment of a variety of human
diseases. Notwithstanding the enormous therapeutic
potential, these compounds have modest clinical benefit
owing to a variety of side effects which limit their
practical use in favor of newer generations of therapeu-
tic agents.2 It is possible that many of the unwanted
side effects of antimuscarinics are due to their interac-
tion with multiple muscarinic receptor subtypes. Clearly,
achievement of selectivity is of paramount importance
for receptor subtype characterization and also for the
development of therapeutically useful muscarinic
drugs.1-3 For instance, muscarinic M1 receptor agonists
and muscarinic M2 receptor antagonists may be useful
to enhance cognitive function, whereas selective mus-
carinic M1 antagonists could have therapeutic potential
for gastrointestinal indication and may be useful as
cognition-impairing tools for research. Furthermore,
muscarinic M2 and M3 antagonists have potential for
the treatment of bradycardic disorders and of airway
obstructions, respectively.
Un iver sa l Tem p la te Con cep t
These considerations prompted us to design new
antimuscarinics which bear no kinship with previously
known classes of muscarinic receptor antagonists. The
starting point was the observation that benextramine
(1) (Chart 1), an irreversible antagonist of both R1- and
R2-adrenoreceptors, also displayed muscarinic antago-
nistic activity with a significant selectivity for cardiac
† Department of Pharmaceutical Sciences.
‡ Department of Pharmacology.
S0022-2623(98)01038-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/12/1998

Universal Template Approach to Drug Design
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21 4151
muscarinic receptors.4 Using benextramine as the
focus, polymethylene tetraamines lacking a disulfide
moiety were designed to achieve specific recognition of
muscarinic receptors.5,6 The finding that appropriate
chain lengths separating the nitrogen atoms and ap-
propriate substituents can modulate both affinity and
selectivity for different receptor systems allowed us to
postulate that a polyamine backbone may represent a
master key (“passe-partout”) in the drug-receptor
recognition process.7 In other words, a polyamine
backbone can be considered a universal template on
which suitable groups (pharmacophores) can be mounted
to achieve selectivity for any given receptor.
there is more chance to distinguish receptor systems.
Thus, an appropriate modification of the chain length
separating the nitrogens of a polyamine might give rise
to an increase of affinity, whereas the insertion of
N-substituents might improve the selectivity as well as
the affinity by increasing the overall number of contacts
between a drug and a receptor. Many naturally occur-
ring or synthetic polyamines with a linear tetraamine
backbone are known. These ligands recognize different
receptors which may be either activated or inhib-
ited.9-11,14
The development of polymethylene tetraamines as
muscarinic receptor antagonists, the prototype of which
is methoctramine (2) (Chart 1), has been the subject of
review articles.5 The replacement of the terminal
2-methoxybenzyl groups of methoctramine with an 11-
acetyl-5,11-dihydropyrido[2,3-b][1,4]benzodiazepin-6-
one (PBD) moiety led to the discovery of tripitramine
(3), resulting in the most potent and the most selective
muscarinic M2 receptor antagonist so far available, able
to discriminate also between muscarinic M2 and M4
The assumption that a polyamine is a universal
template stands on the following: It is known that a
high percentage of homology exists not only within
receptor subtypes but also among different receptor
families, which may account for the difficulty to achieve
receptor selectivity that remains one of the most fasci-
nating challenges to medicinal chemists. Although
receptor homology makes it inherently difficult to
achieve receptor selectivity, it may be used indeed to
design molecular entities which are able, in principle,
to recognize different receptor systems. This working
hypothesis derives from the consideration that neu-
rotransmitter receptors are folded polypeptide chains
which always contain the same amino acids, albeit in a
different proportion and sequence. The sequence of the
amino acids constitutes the primary structure which is
derived by peptide bonds linking the carboxylate groups
to the amino groups. The resulting repeat units in the
chain, i.e., the peptide bond and the R carbon, form the
backbone. Since the backbone of a protein has only a
structural role and cannot be considered a target for
selectivity, it is evident that only lateral chains play a
major role in drug-receptor binding. Among these
functionalities aspartate, glutamate, and aromatic resi-
dues may acquire paramount importance for the binding
with cationic ligands by way of a cation-anion or a
cation-π interaction. Considering that proteins may
bear several carboxylate and/or aromatic residues some-
where in their structure, in principle, it is possible to
design a lead compound having a polyamine backbone
which is able to recognize multiple anionic sites of a
given receptor. Thus, such a ligand may interact with
all receptor proteins, provided that the distance sepa-
rating the amine functions of the ligand fits the distance
between the carboxylate or aromatic residues of the
receptor. In other words, a polyamine could be consid-
ered a master key in the drug-receptor recognition
process owing to its flexibility that permits it to assume
a suitable conformation to allow the interaction between
protonated amine functions and receptor anionic sites.
It was reported that polyamines, such as spermine and
homospermine, are highly protonated at physiological
pH. It turned out that spermine and homospermine are
85% and 97% tetracation, respectively, the remainder
being the trication.8 Thus, the ability of a polyamine
to interact with biological counterions, that is, a set of
carboxylate anions fixed to the backbone of a receptor,
could well be related to its cationic properties. Conse-
quently, the distance between the cationic nitrogens of
a polyamine becomes critical in the recognition step.
It is known that by increasing the number of interac-
tions that take place between a receptor and a ligand
receptors15 but not between M1 and M4 subtypes.16,17
A
further modification of methoctramine structure af-
forded spirotramine that in binding assays displayed an
inverse selectivity profile in comparison with both
methoctramine and tripitramine, further supporting the
hypothesis that a polyamine backbone may represent
a universal template for drug design.18
The observation that the insertion of one PBD moiety
onto only one of the terminal nitrogens of the tetraamine
backbone of methoctramine afforded 4, which displayed
an affinity profile similar to that of methoctramine,17
gave us the opportunity to verify whether the position
of the tricyclic moiety relative to the protonated terminal
nitrogen may be responsible for receptor subtype selec-
tivity as previously suggested by others to rationalize
receptor subtype selectivity displayed by pirenzepine
and its analogues.19 To this end, we synthesized some
homologues (5-7) of the monosubstituted tetraamine
4, bearing from two to four methylenes between the
terminal nitrogen and the tricyclic group. Since the four
amine functions of these tetraamines are likely to be
mostly protonated at physiological pH, making their use
as a tool in improving cognitive function difficult
because they will hardly cross the blood brain-barrier,
we synthesized the corresponding diamine diamides
8-11 of tetraamines 4-7 by transforming the two inner
amine functions into two amide groups to improve
lipophilicity and allow the new compounds to penetrate,
albeit to a modest extent, the central nervous system.
Furthermore, we synthesized all possible symmetrical
and unsymmetrical diamine diamides (12-15), bearing
from one to four PBD groups on the terminal nitrogens,
to improve affinity and hopefully selectivity for musca-
rinic receptor subtypes.
Since the replacement of the pyridine ring of the
tricyclic moiety of AQ-RA 741, a potent and selective
M2 antagonist, with a benzene ring afforded the carbon
analogue DIBA,20 which is 1 order of magnitude more
potent than the parent compound, we thought it of
interest to perform the same structural modification on
the tricyclic system of tripitramine. Thus, we designed
tetraamines 16 and 17 (C-tripitramine), which bear on
their terminal nitrogens the more lipophilic 5-acetyl-
10,11-dihydrodibenzo[b,e][1,4]diazepin-11-one (DBD) tri-

4152 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21
Bolognesi et al.
Sch em e 1a
cyclic ring system of DIBA rather than PBD moieties.
In addition, to further support the view that the affinity
profile of methoctramine-related tetraamines can be
modulated by appropriate substituents on the terminal
nitrogens of a tetraamine backbone, we replaced tripi-
tramine PBD moieties with the isomeric 11-acetyl-6,-
11-dihydropyrido[2,3-b][1,5]benzodiazepin-5-one (inv-
PBD) groups, affording tetraamines 18-20.
carbonyl]-6-aminocaproic acid to afford diamine diamide
24, having the two terminal amine functions protected
with different moieties. Removal of the N-tert-butoxy-
carbonyl group was achieved by hydrolysis with 6 N
HCl. Thus, 24 was transformed into 25 which was
treated with 27 or 2822 to afford 31 and 32 or 33,
respectively. Diamine diamides 8, 9, and 12 were
obtained from 31, 33, and 32, respectively, by catalytic
hydrogenolysis. Finally, monosubstituted diamine dia-
mides 10 and 11 were prepared by alkylating unsub-
stituted diamine diamide 2621 with 29 and 30,22 respec-
tively (Scheme 1).
We describe here the synthesis and the pharmacologi-
cal profile of tetraamines 4-7 and 16-20 and diamine
diamides 8-15 in functional and binding experiments.
Ch em istr y
Diamine diamides 13-15 were synthesized following
the synthetic pathway shown in Scheme 2. Thus,
hydrogenolysis of 24 gave 34 which was transformed
into 35 by reaction with benzaldehyde and subsequent
reduction of the intermediate Schiff base with NaBH4
followed by removal of the tert-butoxycarbonyl group.
Alkylation of 35 and 3616 with 2722 afforded 38 and 37,
respectively, which, in turn, were transformed, upon
catalytic hydrogenolysis, into the corresponding 13 and
14. A further alkylation of 13 with 2722 gave 15
(Scheme 2).
All the compounds were synthesized by standard
procedures (Schemes 1-3) and were characterized by
IR, 1H NMR, fast atom bombardment (FAB) mass
spectra, and elemental analysis.
Tetraamine 2321 was alkylated with the appropriate
chloride (28-30)22 to give the corresponding monosub-
stituted tetraamines 5-7 (Scheme 1). N-(tert-Butoxy-
carbonyl)-6-aminocaproic acid was amidated with
N′-[(benzyloxy)carbonyl]-N,N′-dimethyl-1,8-octanedi-
amine23 to give 21. Removal of the N-(benzyloxy)-
carbonyl group was achieved by catalytic hydrogenation
over 10% palladium on charcoal. Thus, hydrogenolysis
of 21 gave 22 which was reacted with N-[(benzyloxy)-
Tetraamines 16-20 were obtained as shown in Scheme
3. Alkylation of 3916 with 4020 and 4122 afforded 42 and

Universal Template Approach to Drug Design
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21 4153
Sch em e 2a
a
For definition of PBD, see Scheme 1.
Sch em e 3
43, respectively, which were transformed, upon hydro-
genolysis, into 16 and 18. A further alkylation of 16
and 18 with 40 and 41, respectively, gave the corre-
sponding tetraamines 17 and 20, and 19.
verified that the experimental data generated a line
whose derived slope was not significantly different from
unity.
Bin d in g Exp er im en ts. The muscarinic receptor
subtype selectivity was assessed by employing receptor
binding assays as reported previously.16,17 [3H]-N-
Methylscopolamine was used to label M2, M3, and M4
muscarinic receptor binding sites of rat heart, submax-
illary gland, and NG 108-15 cell homogenates, respec-
tively. [3H]Pirenzepine was the tracer to label M1
muscarinic receptor binding sites of the rat cerebral
cortex. Binding affinities are expressed as pKi values
derived using the Cheng-Prusoff equation.26 Methoc-
tramine (2) and tripitramine (3) were used as the
standard compounds.
Biology
F u n ction a l Stu d ies. Functional activity at musca-
rinic receptor subtypes was determined by the use of
the muscarinic M2 receptor-mediated negative inotro-
pism in driven guinea pig left atria (1 Hz) and musca-
rinic M3 receptor-mediated contraction of guinea pig
ileum longitudinal muscle. These methods have been
described in detail earlier.17 The agonist was arecaidine
propargyl ester (APE). To allow comparison of the
results, methoctramine (2) and tripitramine (3) were
used as the standard compounds. The biological results
are expressed as pA2 values determined from Schild
plots24 constrained to slope -1.0,25 as required by
theory. When this method was applied, it was always
Resu lts a n d Discu ssion
The functional activity, expressed as pA2 values, at
peripheral muscarinic M2 and M3 receptors of polyamines
used in the present study is shown in Table 1 and

4154 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21
Bolognesi et al.
Ta ble 1. Antagonist Affinities, Expressed as pA2 Values, in the Isolated Guinea Pig Left Atrium (M2) and Longitudinal Ileum (M3)
Muscarinic Receptors
b
pA2
no.a
n
X
R1
PBD
PBD
PBD
PBD
PBD
PBD
PBD
PBD
PBD
PBD
PBD
PBD
PBD
R2
R3
PBD
H
H
H
H
H
H
H
H
H
H
PBD
PBD
H
DBD
H
R4
PBD
H
H
H
H
H
H
M2
M3
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
7.91 ( 0.03
9.65 ( 0.08
7.95 ( 0.04
8.57 ( 0.06
9.15 ( 0.08
9.01 ( 0.02
8.01 ( 0.05
8.04 ( 0.06
9.34 ( 0.01
9.53 ( 0.05
5.25 ( 0.03
9.30 ( 0.01
8.36 ( 0.04
7.75 ( 0.03
8.82 ( 0.01
9.70 ( 0.02
8.28 ( 0.02
8.60 ( 0.04
7.87 ( 0.01
6.44 ( 0.06
6.55 ( 0.04
6.56 ( 0.01
7.77 ( 0.02
8.32 ( 0.04
7.71 ( 0.03
6.36 ( 0.02
7.74 ( 0.01
8.73 ( 0.03
7.86 ( 0.07
5.14 ( 0.04
7.14 ( 0.01
6.26 ( 0.08
5.67 ( 0.01
6.95 ( 0.02
7.35 ( 0.01
6.27 ( 0.01
6.04 ( 0.05
5.79 ( 0.01
0
0
1
2
3
0
1
2
3
0
0
0
0
0
0
0
0
0
CH2
CH2
CH2
CH2
CH2
CO
CO
CO
CO
CO
CO
CO
CO
CH2
CH2
CH2
CH2
CH2
H
H
H
H
H
H
H
H
H
H
H
H
PBD
H
H
PBD
PBD
PBD
DBD
DBD
inv-PBD
inv-PBD
inv-PBD
PBD
DBD
DBD
inv-PBD
inv-PBD
inv-PBD
H
H
H
H
inv-PBD
inv-PBD
inv-PBD
a
b
2, tetrahydrochloride; 3-7, 16-20, tetraoxalates; 8-15, dihydrochlorides. pA2 values ( SE were calculated from Schild plots,24
constrained to slope -1.0.25 pA2 is the positive value of the intercept of the line derived by plotting log (DR - 1) vs log [antagonist]. The
log (DR - 1) was calculated from at least three different antagonist concentrations, and each concentration was tested from four to six
times. Dose-ratio (DR) values represent the ratio of the potency of the agonist arecaidine propargyl ester (EC50) in the presence of the
antagonist and in its absence. Parallelism of concentration-response curves was checked by linear regression, and the slopes were tested
for significance (p < 0.05).
F igu r e 2. Effect of the carbon chain length separating the
PBD moiety and the terminal nitrogen (n ) 0-3) of diamine
diamides 8-11 on blocking activity of guinea pig left atria (M2)
and guinea pig longitudinal ileum (M3) muscarinic receptor
subtypes. Data from Table 1.
F igu r e 1. Effect of the carbon chain length separating the
PBD moiety and the terminal nitrogen (n ) 0-3) of tet-
raamines 4-7 on blocking activity of guinea pig left atria (M2)
and guinea pig longitudinal ileum (M3) muscarinic receptor
subtypes. Data from Table 1.
nitrogen atom of 4, affording 5-7, on affinity is quali-
tatively similar for both muscarinic M2 and M3 recep-
tors; the affinity for these receptors increased with
increasing the chain length, reaching a maximum for n
) 2 (Table 1), although tetraamine 7 carrying an
additional methylene also retained high activity (Figure
1). Next, changing the two inner amine functions of
tetraamines 4-7 into amides, affording the correspond-
ing diamine diamides 8-11, did not alter dramatically
the affinity profile at muscarinic M2 and M3 receptors,
as only a slight variation in the affinity for the M2
subtype was observed as shown graphically in Figure
Figures 1 and 2. To make relevant considerations on
structure-activity relationships, we included parent
compounds 2 (methoctramine) and 3 (tripitramine) for
comparison. All compounds behaved as competitive
antagonists as revealed by the slopes of their Schild
plots, which were not significantly different from unity
(p > 0.05) (Table 1).
It can be observed that the effect of lengthening the
distance between the tricyclic moiety and the terminal

Universal Template Approach to Drug Design
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21 4155
2. Diamine diamides 8-11 displayed a very high
affinity toward muscarinic M2 receptors, approaching
that of tripitramine as revealed by a comparison of pA2
values of 3 and 11 (9.65 ( 0.08 vs 9.53 ( 0.05).
Interestingly enough, notwithstanding the fact that the
highest affinity was always associated with muscarinic
M2 receptors, for the first time polyamines were shown
to display high pA2 values also toward muscarinic M3
receptors. In particular, diamine diamide 10 was more
than 2 orders of magnitude more potent than both
prototypes, methoctramine (2) and tripitramine (3), at
muscarinic M3 receptors as revealed by their pA2 values
(10, 8.73 ( 0.03; 2, 6.44 ( 0.06; 3, 6.55 ( 0.04). In
previous related studies polyamines were shown to
display invariably pA2 values of about 6.0-6.5, allowing
the conclusion that muscarinic M3 receptors are less
sensitive than muscarinic M2 receptors to structural
modifications performed on tetraamines and that a
tetraamine backbone may not be a suitable carrier for
optimal interaction with this receptor subtype. Now we
have shown that lengthening the distance between the
tricyclic ring of the PBD moiety and the terminal amine
function of a tetraamine template might represent a
first step for the design of tetraamines displaying high
affinity and hopefully selectivity for the M3 subtype. The
high affinity displayed by tetraamine 6 for muscarinic
M3 receptors emphasizes once again that a polyamine
backbone may represent a universal template for recep-
tor recognition. Affinity and selectivity can be tuned
simply by inserting appropriate substituents in the
template structure.
that of tripitramine (9.30 ( 0.01 vs 9.65 ( 0.08). The
insertion of more than one PBD group on the same
terminal nitrogen of a diamine diamide backbone caused,
in comparison with 8, a decrease in affinity, as in 12,
for both muscarinic M2 and M3 receptors or, alterna-
tively, did not markedly modify the affinity profile, as
in 14 and 15.
Concerning the mode of interaction of the above
polyamines, at a first glance, comparing the results
obtained with tetraamines 4-7 and diamine diamides
8-11, one could reach the conclusion that both musca-
rinic M2 and M3 receptors may have the same structural
requirements and, as a consequence, that tetraamines
and diamine diamides may share a common mechanism
of receptor interaction. However, taking into account
also the diverging effect observed in a previous work17,27
and in the present investigation by replacing the
hydrogens with PBD groups on the terminal nitrogens
of 4 and 8, it becomes reasonable to assume that
tetraamines and diamine diamides may have a slightly
different mode of interaction with the receptor.
Previously, we advanced that the high affinity of
tripitramine for muscarinic M2 receptors may be the
result of an interaction with two different sites thanks
to its two differently substituted terminal nitrogens.17
Thus, the N-substituted terminal nitrogen would inter-
act with the active binding site, whereas the other N,N-
disubstituted terminal nitrogen would recognize a sec-
ond accessory site. The finding that nonsymmetrical
polyamines 4-11 displayed high affinity for muscarinic
M2 receptors supports the view that the sites where the
two terminal nitrogens of these ligands are likely to bind
have different structural requirements.
The finding that tetraamine 4 and diamine diamide
8 showed a qualitatively and quantitatively similar
affinity profile, while being 25- and 45-fold, respectively,
more potent at muscarinic M2 receptors than at mus-
carinic M3 receptors, may be relevant (a) to the design
of new selective M2 antagonists based on a diamine
diamide rather than a tetraamine backbone and (b) to
the understanding of the mode of interaction of meth-
octramine (2)-related polyamines at muscarinic M2
receptors.
Since the insertion of additional PBD groups on the
terminal nitrogens of 8, affording 12-15, did not
improve either affinity or selectivity of tripitramine (3)
for muscarinic M2 receptors versus muscarinic M3
receptors, our attention was focused again on modifying
the substituents on the terminal nitrogens of the tet-
raamine backbone, affording 16-20. An analysis of the
results reveals that the replacement of the tricyclic ring
system of tripitramine with a DBD moiety, affording the
carbon analogues 16 and 17 (C-tripitramine), did not
improve the affinity for both muscarinic M2 and M3
receptors as observed following the same modification
performed on AQ-RA 74120 to afford the carbon analogue
DIBA. C-Tripitramine (17) was as active as tripitra-
mine (3) at muscarinic M2 receptors, whereas it was
significantly more potent (6-fold) at the M3 subtype.
Interestingly, the inversion of the endo-amide function
of the tricyclic moiety of tripitramine, as in 18-20,
caused a decrease in affinity for muscarinic M2 receptors
while not affecting that for M3 receptors, in comparison
with tripitramine or its carbon analogue 17.
The use of a diamine diamide backbone as template,
on which appropriate substituents should be mounted
to tune both affinity and selectivity, could afford new
M2 antimuscarinics with improved lipophilic properties
relative to those bearing a tetraamine backbone, which
might be useful in investigating cognitive function.
Since the inclusion of additional PBD moieties on the
terminal nitrogens of 4 led to the discovery of tripitra-
mine (3), which displayed outstanding properties as an
M2 antagonist, we made the same modification on the
terminal nitrogens of 8, affording diamine diamides 12-
15. Surprisingly enough, the affinity of 12-15 for
muscarinic M2 receptors did not follow the same trend
observed for the corresponding tetraamines obtained
through a similar modification performed on tetraamine
4. An analysis of the results reveals that the unsym-
metrical diamine diamide 14 (tripinamide) is signifi-
cantly less potent (about 20-fold) than the corresponding
tetraamine tripitramine (3) at muscarinic M2 receptors
while retaining the same affinity for the M3 subtype.
Furthermore, optimum activity was associated with the
symmetrically monosubstituted diamine diamide 13
whose pA2 value at muscarinic M2 receptors approached
The binding affinities, expressed as pKi values, in rat
cortex (M1), heart (M2), and submaxillary gland (M3) and
in NG 108-15 cells (M4) of muscarinic receptor subtypes
of tetraamines 4, 6, and 16-20 and diamine diamides
6, 8, 11, and 14 are shown in Table 2 and, for selected
compounds, in Figure 3 in comparison with those of
methoctramine (2) and tripitramine (3). Concerning M2
and M3 muscarinic receptor subtypes, it is evident that
binding affinities are qualitatively and quantitatively
similar to pA2 values derived from functional experi-

4156 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21
Bolognesi et al.
Ta ble 2. Affinity Estimates, Expressed as pKi Values, in Rat
Cortex (M1), Heart (M2), and Submaxillary Gland (M3) and NG
108-15 Cell (M4) Muscarinic Receptor Subtypes of Polyamines
14, 17, and 19 in Comparison with the Prototypes 2 and 3
raamine 16, did not affect the affinity for muscarinic
M2 and M3 receptors while causing a significant increase
in the affinity for the M4 type and a significant decrease
for the M1 type.
pKia
It is evident that an appropriate substitution of the
terminal nitrogens of the tetraamine backbone of me-
thoctramine affords potent antimuscarinics that display
different selectivity profiles. The insertion of PBD
groups on a tetraamine backbone afforded tripitramine
which possesses outstanding properties toward musca-
rinic M2 receptors in comparison with the parent
compound methoctramine.8-10 However, tripitramine
fails to discriminate between muscarinic M1 and M4
receptors. Now we have demonstrated that the replace-
ment of the tricyclic ring system of tripitramine with a
DBD moiety gives the carbon analogue C-tripitramine
(17) whose selectivity profile makes it possible to
overcome the limitation of that displayed by tripitra-
mine owing to a higher (about 100-fold) affinity for
muscarinic M1 and M2 versus M3 and M4 receptors.
Furthermore, C-tripitramine (17) was 100-fold more
lipophilic than tripitramine (3) as revealed by the log P
values which were calculated for the free bases by using
CLOGP Program.28 Moreover, changing the inner
amine functions of tripitramine into amides afforded
tripinamide (14) which was significantly less potent
than the prototype but retained the same affinity profile
as shown in Figure 3. Thus, diamine diamide 14 can
serve as a lead for further structural modifications to
design new antimuscarinics and also as tool in inves-
tigating cholinergic function in the central nervous
system, owing to its presumably improved lipophilicity
relative to tripitramine or methoctramine.
no.
M1
M2
M3
M4
2 (metho-
ctramine)
3 (tripi-
tramine)
7.38 ( 0.10 7.70 ( 0.08 6.02 ( 0.15 7.49 ( 0.14
8.45 ( 0.13 9.52 ( 0.07 6.83 ( 0.10 7.94 ( 0.12
4
6
8
11
8.36 ( 0.15 8.11 ( 0.09 6.48 ( 0.11 7.80 ( 0.06
9.30 ( 0.15 9.43 ( 0.14 8.52 ( 0.22 8.84 ( 0.09
6.50 ( 0.07 7.35 ( 0.09 6.80 ( 0.07 7.70 ( 0.08
7.20 ( 0.08 8.25 ( 0.08 7.15 ( 0.20 7.20 ( 0.07
7.30 ( 0.10 8.60 ( 0.07 5.24 ( 0.09 6.80 ( 0.12
14 (tripin-
amide)
16
17 (C-tripi-
tramine)
18
7.83 ( 0.15 9.25 ( 0.10 6.94 ( 0.06 8.55 ( 0.16
9.02 ( 0.20 9.36 ( 0.08 6.96 ( 0.04 7.38 ( 0.08
7.84 ( 0.32 7.68 ( 0.22 5.59 ( 0.10 7.44 ( 0.08
7.90 ( 0.20 9.04 ( 0.08 6.40 ( 0.07 7.40 ( 0.04
7.60 ( 0.07 8.03 ( 0.12 5.60 ( 0.06 6.80 ( 0.08
19
20
a
Values are the mean ( SE of at least three separate experi-
ments performed in triplicate. All Hill numbers (nH) were not
significantly different from unity (p > 0.05). Equilibrium dissocia-
tion constants (Ki) were derived using the Cheng-Prusoff equa-
tion.26 [3H]NMS was used to label muscarinic receptors in rat heart
and submaxillary gland and NG 108-15 binding assays, whereas
[3H]pirenzepine was the tracer in rat cortex homogenates. Scat-
chard plots were linear or almost linear in all preparations tested.
Con clu sion s
We have confirmed that appropriate substituents on
the terminal nitrogens of a tetraamine template can
tune both affinity and selectivity for muscarinic recep-
tors. The insertion of a DBD moiety into a tetraamine
backbone afforded the novel tetraamine C-tripitramine
(17), which was able to discriminate significantly mus-
carinic M1 and M2 receptors versus the other subtypes.
It is evident that the use of C-tripitramine combined
with that of tripitramine may represent a valuable tool
for the pharmacological identification of muscarinic
receptor subtypes. Another additional interesting find-
ing was the observation that the tetraamine backbone
which served as a basis for the discovery of both
methoctramine and tripitramine can be transformed
into a diamine diamide while retaining high affinity for
different muscarinic receptor subtypes. Thus, as an
example, the diamine diamide tripinamide (14) dis-
played a binding affinity profile (M2 > M1 > M4 > M3)
qualitatively similar to, albeit quantitatively different
from, that of tripitramine while being significantly more
lipophilic which might allow it to cross the blood-brain
barrier and to investigate cognition function.
F igu r e 3. Affinity constants (pKi) in rat cortex (M1), heart
(M2), and submaxillary gland (M3) and NG 108-15 cell (M4)
muscarinic receptor subtypes for 14, 17, and 19 in comparison
with the prototypes methoctramine (2) and tripitramine (3).
ments, the only exception being tetraamine 11 and the
diamine diamide tripinamide (14) at muscarinic M2 and
M3 receptors, respectively. However, we have no ex-
planation for this discrepancy.
Interestingly, the inversion of the endo-amide function
of tripitramine, affording 19, caused a decrease in
affinity for all muscarinic receptor subtypes in compari-
son with tripitramine (3). Removal or insertion of one
inv-PBD group from a terminal nitrogen atom of 19,
affording 18 and 20, did not improve the selectivity
profile. However, the substitution of the terminal
nitrogens of the tetraamine backbone with DBD groups
gave different effects on the affinity for muscarinic M1
and M4 receptors. C-Tripitramine (17) resulted as
active as tripitramine at muscarinic M2 and M3 recep-
tors, while being less potent at the M4 subtype and more
potent at the M1 subtype. Removal of one DBD group
from 17, affording the symmetrically substituted tet-
Exp er im en ta l Section
Ch em istr y. Melting points were taken in glass capillary
tubes on a Bu¨chi SMP-20 apparatus and are uncorrected. IR
and NMR spectra were recorded on Perkin-Elmer 297 and
Varian VXR 300 instruments, respectively. Although the IR
and NMR spectra data are not included (because of the lack
of unusual features), they were obtained for all compounds

Universal Template Approach to Drug Design
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21 4157
reported and were consistent with the assigned structures. The
mass spectra were obtained on a VG707EH-F spectrometer.
The elemental compositions of the compounds agreed to within
(0.4% of the calculated value. When the elemental analysis
is not included, crude compounds were used in the next step
without further purification. Chromatographic separations
were performed on silica gel columns by flash (Kieselgel 40,
0.040-0.063 mm; Merck) or gravity column (Kieselgel 60,
0.063-0.200 mm; Merck) chromatography. The composition
and volumetric ratio of eluting mixtures were as follows: A,
CHCl3-EtOH (9:1); B, MeOH-aqueous 28% ammonia (9.7:
0.3); C, CH2Cl2-EtOAc-EtOH (8.7:1:0.3); D, CHCl3-MeOH-
aqueous 28% ammonia (8.5:1.5:0.1); E, CHCl3-MeOH-
aqueous 28% ammonia (8.5:1.5:0.15); F, CHCl3-MeOH-
aqueous 28% ammonia (9:1:0.1); G, CHCl3-MeOH-aqueous
28% ammonia (5:4.5:0.5); CHCl3-MeOH-aqueous 28% am-
monia (9:1:0.07); H, CH2Cl2-EtOH-aqueous 28% ammonia
(9:1.5:0.15); I, CH2Cl2-EtOH-aqueous 28% ammonia (9.1:0.9:
0.09); J , CH2Cl2-EtOH-aqueous 28% ammonia (9.2:0.8:0.08);
K, CHCl3-EtOH-aqueous 28% ammonia (9:1:0.1); L, CHCl3-
EtOH-aqueous 28% ammonia (9:1:0.04); M, CH2Cl2-MeOH-
aqueous 28% ammonia (9:1:0.1). Reactions were followed by
thin-layer chromatography (TLC) on Merck (0.25 mm) glass-
packed precoated silica gel plates (60 F254) that were visualized
in an iodine chamber. The term “dried” refers to the use of
anhydrous sodium sulfate.
N1,N1,N22,N22-Tet r a k is[[(5,11-d ih yd r o-6-oxo-6H -p y-
r id o[2,3-b][1,4]ben zod ia zep in -11-yl)ca r bon yl]m eth yl]-7,-
16-d im e t h y l-6,17-d io x o -7,16-d ia za -1,22-d o c o s a n e d i-
a m in e Dih yd r och lor id e (15). It was obtained from 13 (as
free base) and 27.22 Eluting with solvent J gave an oil that
was transformed into the dihydrochloride salt: 12% yield; mp
210 °C dec (from EtOH/ether). Anal. (C78H84Cl2N16O10) C, H,
N. The second fraction was 14 (27% yield) that was identical
to the compound obtained by another route.
N1,N1,N22-Tr is[[(5,10-dih ydr o-11-oxo-11H-diben zo[b,e]-
[1,4]d ia zep in -5-yl)ca r bon yl]m eth yl]-7,16-d im eth yl-7,16-
d ia za -1,22-d ocosa n ed ia m in e Tetr a oxa la te (17). It was
obtained from 16 (as free base) and 40.20 It was purified by
flash chromatography, eluting with solvent H, followed by
gravity column chromatography. Eluting with toluene-
chloroform-methanol-aqueous 28% ammonia (1:9.5:1:0.1)
gave 17 as the free base (first fraction) that was transformed
into the tetraoxalate salt: 25% yield; mp 150-151 °C (from
EtOH/ether); MS (FAB) calcd for C67H81N10O6 1121 [M + H]+,
found 1121. Anal. (C75H88N10O22) C, H, N. The second
fraction was an unidentified compound.
N1,N1,N22-Tr is[[(6,11-d ih yd r o-5-oxo-5H-p yr id o[2,3-b]-
[1,4]ben zodiazepin -11-yl)car bon yl]m eth yl]-7,16-dim eth yl-
7,16-d ia za -1,22-d ocosa n ed ia m in e Tetr a oxa la te (19) a n d
N1,N1,N22,N22-Tetr a k is[[(6,11-d ih yd r o-5-oxo-5H-p yr id o-
[2,3-b][1,4]b en zod ia zep in -11-yl)ca r b on yl]m et h yl]-7,16-
d im eth yl-7,16-d ia za -1,22-d ocosa n ed ia m in e Tetr a oxa la te
(20). They were obtained from 18 (as free base) and 4122 and
purified by gravity column chromatography eluting with
solvent E. The first fraction was 20 as the free base that was
transformed into the tetraoxalate salt: 20% yield; mp 190-
192 °C (from EtOH/ether); MS (FAB) calcd for C78H87N16O8
1376 [M + H]+, found 1376. Anal. (C86H94N16O24) C, H, N.
The second fraction was 19 as the free base that was
transformed into the tetraoxalate salt: 22% yield; mp 185-
187 °C (from EtOH/ether); MS (FAB) calcd for C64H78N13O6
1125 [M + H]+, found 1125. Anal. (C72H85N13O22) C, H, N.
Gen er a l P r oced u r e for th e Syn th esis of 5-7, 10, 11,
15, 17, 19, 20, 31-33, 37, 38, 42, a n d 43. A mixture of an
amine and a chloride in a 1:1 (17, 19, 20, 31-33), 1:2.16 (37),
1:2.5 (42, 43), 1:3.7 (38), 1.66:1 (15), 3.7:1 (10, 11), or 5:1 (5-
7) ratio in dry DMF (5-7, 10, 11, 15, 17, 19, 20, 37, 38, 42,
and 43) or methylene chloride (31-33) containing triethy-
lamine (chloride’s equivalents) and KI (few crystals) was
stirred at room remperature for several days, following the
reaction by TLC. Removal of the solvent gave a residue that,
unless otherwise specified, was purified by flash chromatog-
raphy.
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]eth yl]-7,16-d im eth yl-7,16-
d ia za -1,22-d ocosa n ed ia m in e Tetr a oxa la te (5). It was
obtained from 2321 and 28.22 Eluting with solvent G gave the
free base that was transformed into the tetraoxalate salt: 45%
yield; mp 118-120 °C (MeOH/i-PrOH); MS (FAB) calcd for
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]m eth yl]-N22-[(ben zyloxy)-
c a r b o n y l]-7,16-d im e t h y l-6,17-d io x o -7,16-d ia za -1,22-
d ocosa n ed ia m in e (31) a n d N1,N1-Bis[[(5,11-d ih yd r o-6-
o x o -6H -p y r i d o [2,3-b ][1,4]b e n z o d i a z e p i n -11-y l)c a r -
bon yl]m eth yl]-N22-[(ben zyloxy)ca r bon yl]-7,16-d im eth yl-
6,17-d ioxo-7,16-d ia za -1,22-d ocosa n ed ia m in e (32). They
were obtained from 25 and 27.22 Eluting with solvent A gave
the desired compounds. The first fraction was 32 as an oil
(23% yield). The second fraction was 31 (27% yield).
C
37H62N7O2 637 [M + H]+, found 637. Anal. (C45H69N7O18
‚
2H2O) C, H, N.
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]p r op yl]-7,16-d im eth yl-7,-
16-d ia za -1,22-d ocosa n ed ia m in e Tetr a oxa la te (6). It was
obtained from 2321 and 29.22 Eluting with solvent G gave the
free base that was transformed into the tetraoxalate salt: 20%
yield; mp 99-101 °C; MS (FAB) calcd for C38H64N7O2 651 [M
+ H]+, found 651. Anal. (C46H71N7O18) C, H, N.
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]bu tyl]-7,16-d im eth yl-7,16-
d ia za -1,22-d ocosa n ed ia m in e Tetr a oxa la te (7). It was
obtained from 2321 and 30.22 Eluting with solvent G gave the
free base that was transformed into the tetraoxalate salt: 30%
yield; mp 80-85 °C; MS (FAB) calcd for C39H66N7O2 664 [M +
H]+, found 664. Anal. (C47H73N7O18) C, H, N.
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]et h yl]-N22-[(ben zyloxy)-
c a r b o n y l]-7,16-d im e t h y l-6,17-d io x o -7,16-d ia za -1,22-
d ocosa n ed ia m in e (33). It was obtained from 25 and 28.22
Eluting with solvent L gave the desired compound as an oil
(43% yield).
N1,N22-Bis[[(5,11-d ih yd r o-6-oxo-6H-p yr id o[2,3-b][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]m eth yl]-N1,N22-d iben zyl-
7,16-d im e t h yl-6,17-d ioxo-7,16-d ia za -1,22-d ocosa n e d i-
a m in e (37). It was obtained from 3616 and 27.22 The residue
was taken up with methylene chloride, washed with water,
and extracted with 3 N HCl. The aqueous layer was made
basic with NaOH pellets and extracted with methylene
chloride. The extracts were washed with brine, dried, and then
evaporated to give 37 as the free base in 97% yield.
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]p r op yl]-7,16-d im eth yl-6,-
17-d ioxo-7,16-d ia za -1,22-d ocosa n ed ia m in e Dih yd r och lo-
r id e (10). It was obtained from 2621 and 29.22 Eluting with
solvent B gave the free base that was transformed into the
dihydrochloride salt: 25% yield; hygroscopic salt. Anal.
(C38H61Cl2N7O4‚H2O) C, H, N.
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]bu tyl]-7,16-d im eth yl-6,17-
d ioxo-7,16-d ia za -1,22-d ocosa n ed ia m in e Dih yd r och lor id e
(11). It was obtained from 2621 and 30.22 Eluting with solvent
B gave the free base that was transformed into the dihydro-
N1,N1,N22-Tr is[[(5,11-d ih yd r o-6-oxo-6H-p yr id o[2,3-b]-
[1,4]ben zod ia zep in -11-yl)ca r bon yl]m eth yl]-N22-ben zyl-
7,16-d im e t h yl-6,17-d ioxo-7,16-d ia za -1,22-d ocosa n e d i-
a m in e Dih yd r och lor id e (38). It was obtained from 35 and
27.22 The residue was taken up with methylene chloride,
washed with water, and extracted with 3 N HCl. The aqueous
layer was made basic with NaOH pellets and extracted with
methylene chloride. The extracts were washed with brine,
dried, and then evaporated to give 38 as the free base which
was transformed into the dihydrochloride salt: 25% yield; mp
175-180 °C (from EtOH/ether).
chloride salt: 25% yield; hygroscopic salt. Anal. (C39H63
-
Cl2N7O4‚H2O) C, H, N.

4158 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21
Bolognesi et al.
N1,N22-Bis[[(5,10-dih ydr o-11-oxo-11H-diben zo[b,e][1,4]-
d ia zep in -5-yl)ca r b on yl]m et h yl]-N1,N22-d ib en zyl-7,16-
d im eth yl-7,16-d ia za -1,22-d ocosa n ed ia m in e Tetr a oxa la te
(42). It was obtained from 3916 and 40.22 Eluting with solvent
F gave the free base that was transformed into the tetraoxalate
salt: 68% yield; mp 180-182 °C (from MeOH/ether).
N1,N22-Bis[[(6,11-d ih yd r o-5-oxo-5H-p yr id o[2,3-b][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]m eth yl]-N1,N22-d iben zyl-
7,16-d im eth yl-7,16-d ia za -1,22-d ocosa n ed ia m in e Tetr a ox-
a la te (43). It was obtained from 3916 and 41.20 Eluting with
solvent D gave the free base that was transformed into the
tetraoxalate salt: 46% yield; mp 185-187 °C (MeOH/ether).
Gen er a l P r oced u r e for th e Syn th esis of 8, 9, 12, 22,
a n d 34. A solution of the N-[(benzyloxy)carbonyl]amine (0.1
mmol) in MeOH (10 mL) was hydrogenated over 10% Pd on
charcoal (10% w/w) for 45 min at room temperature and a
pressure of 15 psi. Following catalyst removal, the solvent
was evaporated, yielding the desired compound as the free
base.
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]eth yl]-7,16-d im eth yl-6,17-
d ioxo-7,16-d ia za -1,22-d ocosa n ed ia m in e Dih yd r och lor id e
(8). It was obtained from 31. The free base was transformed
into the dihydrochloride salt to give a hygroscopic solid: 80%
yield. Anal. (C36H57Cl2N7O4) C, H, N.
N 1-[[(5,11-D i h y d r o -6-o x o -6H -p y r i d o [2,3-b ][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]p r op yl]-7,16-d im eth yl-6,-
17-d ioxo-7,16-d ia za -1,22-d ocosa n ed ia m in e Dih yd r och lo-
r id e (9). It was obtained from 33 and purified by flash
chromatography. Eluting with solvent N gave an oil that was
transformed into the dihydrochloride salt to give a hygroscopic
solid: 70% yield. Anal. (C37H59Cl2N7O4‚1.5H2O) C, H, N.
N1,N1-Bis[[(5,11-d ih yd r o-6-oxo-6H-p yr id o[2,3-b][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]m eth yl]-7,16-d im eth yl-6,-
17-d ioxo-7,16-d ia za -1,22-d ocosa n ed ia m in e Dih yd r och lo-
r id e (12). It was obtained from 32. The free base was
transformed into the dihydrochloride salt to give a hygroscopic
solid. Anal. (C50H66Cl2N10O6‚H2O) C, H, N.
N1-(ter t-Bu toxyca r bon yl)-N15,7-d im eth yl-6-oxo-7-a za -
1,15-p en ta d eca n ed ia m in e (22). It was obtained from 21 as
an oil (75% yield).
N1-(ter t-Bu toxyca r bon yl)-7,16-d im eth yl-6,17-d ioxo-7,-
16-d ia za -1,22-d ocosa n ed ia m in e (34). It was obtained from
24 as an oil in quantitative yield.
Gen er a l P r oced u r e for th e Syn th esis of 13, 14, 16, a n d
18. A solution of N-benzylamine hydrochloride (1.0 mmol) in
MeOH (30 mL) was hydrogenated over 10% Pd on charcoal
(wet, Degussa type E101 NE/W) (0.5 g) overnight at room
temperature and a pressure of 75 psi. Following catalyst
removal, the solvent was evaporated, yielding a residue that
was purified by flash chromatography, unless otherwise speci-
fied.
N1,N22-Bis[[(5,11-d ih yd r o-6-oxo-6H-p yr id o[2,3-b][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]m eth yl]-7,16-d im eth yl-6,-
17-d ioxo-7,16-d ia za -1,22-d ocosa n ed ia m in e Dih yd r och lo-
r id e (13). It was obtained from 37 dihydrochloride. Eluting
with solvent K gave the free base that was transformed into
the dihydrochloride salt: 60% yield; mp 192-194 °C (from
EtOH/ether). Anal. (C50H66Cl2N10O6) C, H, N.
the tetraoxalate salt: 80% yield; mp 182-184 °C (MeOH/
ether); MS (FAB) calcd for C52H71N8O4 871 [M + H]+, found
871. Anal. (C60H78N8O20) C, H, N.
N1,N22-Bis[[(6,11-d ih yd r o-5-oxo-5H-p yr id o[2,3-b][1,4]-
ben zod ia zep in -11-yl)ca r bon yl]m eth yl]-7,16-d im eth yl-7,-
16-d ia za -1,22-d ocosa n ed ia m in e Tetr a oxa la te (18). It was
obtained from 43 (a few drops of 3 N ethanolic HCl added).
The residue was mixed with water, made basic with aqueous
28% ammonia, and extracted with chloroform. Removal of
dried extracts gave 18 as the free base that was transformed
into the tetraoxalate salt: 80% yield; mp 191-194 °C (from
MeOH/ether); MS (FAB) calcd for C50H69N10O4 873 [M + H]+,
found 873. Anal. (C58H76N10O20) C, H, N.
N1-(ter t-Bu toxyca r bon yl)-N15-[(ben zyloxy)ca r bon yl]-
N15,7-dim eth yl-6-oxo-7-aza-1,15-pen tadecan ediam in e (21)
an d N1-(ter t-Bu toxycar bon yl)-N22-[(ben zyloxy)car bon yl]-
7,16-d im e t h yl-6,17-d ioxo-7,16-d ia za -1,22-d ocosa n e d i-
a m in e (24). The procedure used for the synthesis of 21 is
described. Ethyl chlorocarbonate (1.2 mL, 13 mmol) in dry
dioxane (3 mL) was added dropwise to a stirred and cooled (5
°C) solution of N-(tert-butoxycarbonyl)-6-aminocaproic acid
(3.10 g, 13 mmol) and triethylamine (1.3 g, 13 mmol) in dioxane
(70 mL), followed after standing for 30 min by the addition of
N′-[(benzyloxy)carbonyl]-N,N′-dimethyl-1,8-octanediamine23 (4.0
g, 13 mmol) in dioxane (20 mL). After stirring at room
temperature overnight, the mixture was poured into water
(150 mL) and then extracted with methylene chloride (3 × 60
mL). The organic layer was washed with 2 N aqueous KHSO4
(2 × 70 mL), saturated NaHCO3 aqueous solution (2 × 70 mL),
and brine. Removal of dried solvents gave a residue which
was triturated with cyclohexane to give 21 as an oil: 4.2 g
(65% yield).
Similarly, 24 was obtained from N-[(benzyloxy)carbonyl]-
6-aminocaproic acid and 22. The residue was purified by flash
chromatography. Eluting with solvent C gave 24 as an oil:
65% yield.
N1-[(Ben zyloxy)ca r bon yl]-7,16-d im eth yl-6,17-d ioxo-7,-
16-d ia za -1,22-d ocosa n ed ia m in e (25). A solution of 24 (2.30
g, 3.63 mmol) in THF (100 mL) was treated with 6 N HCl (20
mL) and stirred at room temperature for 2 h. THF was
distilled off under reduced pressure, and the aqueous phase
was extracted with ether (3 × 20 mL), made basic with 2 N
NaOH, and extracted with methylene chloride (3 × 50 mL).
Removal of the dried extracts gave 25 as an oil: 1.9 g (96%
yield).
Biology. F u n ction a l An ta gon ism . Guinea pigs of either
sex (200-400 g) were sacrificed by cervical dislocation under
ketamine anesthesia, and the organs required were set up
rapidly under a suitable resting tension in 15-mL organ baths
containing physiological salt solution kept at appropriate
temperature (see below) and aerated with 5% CO2-95% O2 at
pH 7.4. Concentration-response curves were constructed by
cumulative addition of the agonist.29 The concentration of
agonist in the organ bath was increased approximately 5-fold
at each step, with each addition being made only after the
response to the previous addition had attained a maximal level
and remained steady. Contractions were recorded by means
of a force displacement transducer (FT.03 Grass and 7003
Basile) connected to a four-channel pen recorder (Battaglia-
Rangoni KV 380). In all cases, parallel experiments in which
tissues did not receive any antagonist were run in order to
check any variation in sensitivity.
Gu in ea P ig Left Atr ia . The heart of guinea pigs was
rapidly removed and washed by perfusion through the aorta
with oxygenated physiological salt solution, and right and left
atria were separated out. The left atria were mounted under
0.2-0.3-g tension at 35 °C in Tyrode solution of the following
composition (mM): NaCl, 136.9; KCl, 5.4; MgSO4‚7H2O, 1.0;
CaCl2, 2.52; NaH2PO4, 0.4; NaHCO3, 11.9; glucose, 5.5. Tis-
sues were stimulated through platinum electrodes by square-
wave pulses (0.6-0.8 ms, 1 Hz, 1-5 V). Inotropic activity was
recorded isometrically. Tissues were equilibrated for 1 h, and
cumulative concentration-response curves to arecaidine pro-
pargyl ester (APE) (0.01-1 µM) were constructed. Following
N1,N1,N22-Tr is[[(5,11-d ih yd r o-6-oxo-6H-p yr id o[2,3-b]-
[1,4]ben zodiazepin -11-yl)car bon yl]m eth yl]-7,16-dim eth yl-
6,17-d ioxo-7,16-d ia za -1,22-d ocosa n ed ia m in e Dih yd r o-
ch lor id e (14). It was obtained from 38 dihydrochloride.
Eluting with solvent I gave the free base that was transformed
into the dihydrochloride salt: 40% yield mp 186-188 °C (from
EtOH/ether); MS (FAB) calcd for C64H74N13O8 1152 [M + H]+,
found 1152. Anal. (C64H75Cl2N13O8) C, H, N.
N1,N22-Bis[[(5,10-dih ydr o-11-oxo-11H-diben zo[b,e][1,4]-
dia zepin -5-yl)ca r bon yl]m eth yl]-7,16-dim eth yl-7,16-d ia za -
1,22-d ocosa n ed ia m in e Tetr a oxa la te (16). It was obtained
from 42 (a few drops of 3 N ethanolic HCl added). The residue
was mixed with water, made basic with aqueous 28% am-
monia, and extracted with chloroform. Removal of dried
extracts gave 16 as the free base that was transformed into

Universal Template Approach to Drug Design
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21 4159
incubation with the antagonist for 60 min, a new concentra-
tion-response curve to APE was obtained.
research topics), European Community (BMH4-CT97-
2395), CNR (97.02822.CT03), and MURST.
Gu in ea P ig Ileu m Lon gitu d in a l Mu scle. The terminal
portion of the ileum was excised after discarding the 8-10 cm
nearest to the ileo-caecal junction. The tissue was cleaned,
and segments 2-3 cm long of ileum longitudinal muscle were
set up under 1-g tension at 37 °C in organ baths containing
Tyrode solution of the following composition (mM): NaCl, 118;
KCl, 4.75; CaCl2, 2.54; MgSO4, 1.2; KH2PO4‚2H2O, 1.19;
NaHCO3, 25; glucose, 11. Tension changes were recorded
isotonically. Tissues were allowed to equilibrate for at least
30 min during which time the bathing solution was changed
every 10 min. Concentration-response curves to APE (0.01-
0.5 µM) were obtained at 30-min intervals, the first one being
discarded and the second one taken as control. Following
incubation with the antagonist for 60 min, a new concentra-
tion-response curve to the agonist was obtained.
Cell Cu ltu r e a n d Bin d in g Assa ys. The detailed methods
have been published previously.30-32 [3H]-N-Methylscopola-
mine ([3H]NMS; specific activity 79.5 Ci/mmol; NEN Du Pont)
was used to evaluate binding sites in rat heart homogenates
(expressing M2 muscarinic receptors; Kd 0.32 ( 0.04 nM; Bmax
77.8 ( 15.3 fmol/mg of protein), in rat submaxillary gland
homogenates (expressing M3 muscarinic receptors; Kd 0.48 (
0.03 nM; Bmax 1102 ( 85 fmol/mg of protein), and in homoge-
nates obtained from NG 108-15 cells (expressing M4 muscarinic
receptors; Kd 0.54 ( 0.03 nM; Bmax 190 ( 4 fmol/mg of protein).
[3H]Pirenzepine (specific activity 86.2 Ci/mmol; NEN Du Pont)
was the tracer to label M1 muscarinic receptor binding sites
of the rat cerebral cortex (Kd 2.1 ( 0.2 nM; Bmax 1.9 ( 0.13
pmol/mg of protein). In competition studies, fixed concentra-
tions of 0.7-0.8 nM [3H]NMS were used in rat heart, rat
submaxillary gland, and NG 108-15 cells binding assays,
whereas 5 nM was the concentration of [3H]pirenzepine in rat
cortex homogenates. Nonspecific binding was assessed in the
presence of 10 µM atropine.
Competition binding studies were performed using homo-
genates of the indicated cells31,32 or rat tissues30,32 in incubation
buffer (50 mM sodium phosphate, pH 7.4, enriched with 2 mM
MgCl2, 1% bovine serum albumin). Homogenates (200 µg of
protein for cortex, 500 µg of protein for heart and submaxillary
gland, 600 µg of protein for NG 108-15 cells) were incubated
for 2 h at 25 °C in 1 mL of incubation buffer and the indicated
concentrations of tracer and atropine. Binding assays were
terminated by filtration on Whatman GF/C glass-fiber filters
previously soaked in 0.1% poly(ethylenimine) and then rinsed
four times with 5 mL of ice-cold 50 mM phosphate buffer (pH
7.4). Saturation binding studies were performed as indicated
above, in the presence of [3H]NMS (25-4000 pM) or [3H]-
pirenzepine (0.08-20 nM) and in the absence or presence of
atropine. The results were analyzed according to the method
of Scatchard.
Refer en ces
(1) For a review, see: (a) Hulme, E. C.; Birdsall, N. J . M.; Buckley,
N. J . Muscarinic Receptor Subtypes. Annu. Rev. Pharmacol.
Toxicol. 1990, 30, 633-673. (b) Caulfield, M. P. Muscarinic
Receptors-Characterization, Coupling and Function. Pharmacol.
Ther. 1993, 58, 319-379. (c) Eglen, R. M.; Hegde, S. S.; Watson,
N. Muscarinic Receptor Subtypes and Smooth Muscle Function.
Pharmacol. Rev. 1996, 48, 531-565.
(2) Widzowski, D.; Helander, H. F.; Wu, E. S. C. Selective Musca-
rinic M1 Antagonists: Drug Design and Discovery. Drug Dis-
covery Today 1997, 2, 341-350.
(3) McKinney, M.; Coyle, J . T. The Potential for Muscarinic Receptor
Subtype-Specific Pharmacotherapy for Alzheimer’s Disease.
Mayo Clin. Proc. 1991, 66, 1225-1237.
(4) Melchiorre, C.; Yong, M. S.; Benfey, B. G.; Belleau, B. Molecular
Properties of the Adrenergic R Receptor. 2. Optimum Covalent
Inhibition by Two Different Prototypes of Polyamine Disulfides.
J . Med. Chem. 1978, 21, 1126-1132.
(5) (a) Melchiorre, C. Polymethylene Tetraamines: a New Genera-
tion of Selective Muscarinic Antagonists. Trends Pharmacol. Sci.
1988, 9, 216-220. (b) Melchiorre, C.; Minarini, A.; Angeli, P.;
Giardina`, D.; Gulini, U.; Quaglia, W. Polymethylene Tet-
raamines as Muscarinic Receptor Probes. Trends Pharmacol. Sci.
1989, Suppl. IV, 55-59. (c) Melchiorre, C. Polymethylene
Tetraamines: A Novel Class of Cardioselective M2 Antagonists.
Med. Res. Rev. 1990, 3, 327-349.
(6) Melchiorre, C.; Cassinelli, A.; Quaglia, W. Differential Blockade
of Muscarinic Receptor Subtypes by Polymethylene Tetraamines.
Novel Class of Selective Antagonists of Cardiac M-2 Muscarinic
Receptors. J . Med. Chem. 1987, 30, 201-204.
(7) Melchiorre, C.; Angeli, P.; Brasili, L.; Giardina`, D.; Pigini, M.;
Quaglia, W. Polyamines: A Possible “Passe-Partout” for Receptor
Characterization. Actual. Chim. The´r. - 15e se´rie 1988, 149-
168.
(8) Aikens, D.; Bunce, S.; Onasch, F.; Parker, R., III; Hurwitz, C.;
Clemans, S. The Interactions Between Nucleic Acids and
Polyamines. II. Protonation Constants and 13C NMR Chemical
Shift Assignments of Spermidine, Spermine, and Homologs.
Biophys. Chem. 1983, 17, 67-74.
(9) Melchiorre, C. Tetramine Disulfides: A New Tool in R-Adren-
ergic Pharmacology. Trends Pharmacol. Sci. 1981, 2, 209-211.
(10) Melchiorre, C.; Recanatini, M.; Bolognesi, M. L.; Filippi, P.;
Minarini, A. Polyamines: A Case of Multiple Yet Specific
Receptor Recognition. Curr. Top. Med. Chem. 1993, 1, 43-65.
(11) Doughty, M. B.; Chu, S. S.; Miller, D. W.; Li, K.; Tessel, R. E.
Benextramine: A Long-Lasting Neuropeptide Y Receptor An-
tagonist. Eur. J . Pharmacol. 1990, 185, 113-114.
(12) Bergeron, R. J .; Weimar, W. R.; Wu, Q.; Austin, J . K., J r.;
McManis, J . S. Impact of Polyamine Analogues on the NMDA
Receptor. J . Med. Chem. 1995, 38, 425-428.
(13) Bergeron, R. J .; McManis, J . S.; Liu, C. Z.; Feng, Y.; Weimar,
W. R.; Luchetta, G. R.; Wu, Q.; Ortiz-Ocasio, J .; Vinson, J . R.
T.; Kramer, D.; Porter, C. Antiproliferative Properties of
Polyamines Analogues: A Structure-Study. J . Med. Chem. 1994,
37, 3464-3476.
(14) Seiler, N.; Atanassov, C. L. The Natural Polyamines and the
Immune System. In Progress in Drug Research; J ucker, E., Ed.;
Birkha¨user Verlag: Basel, 1994; Vol. 43, pp 87-141.
(15) Dorje, F.; Wess, J .; Lambrecht, G.; Tacke, R.; Mutschler, E.;
Brann, M. R. Antagonist Binding Profiles of Five Cloned Human
Muscarinic Receptor Subtypes. J . Pharmacol. Exp. Ther. 1991,
256, 727-733.
(16) Melchiorre, C.; Bolognesi, M. L.; Chiarini, A.; Minarini, A.;
Spampinato, S. Synthesis and Biological Activity of Some
Methoctramine-Related Tetraamines Bearing a 11-Acetyl-5,11-
dihydro-6H-pyrido[2,3-b][1,4]-benzodiazepin-6-one Moiety as An-
timuscarinics: A Second Generation of Highly Selective M2
Muscarinic Receptor Antagonists. J . Med. Chem. 1993, 36,
3734-3737.
(17) Minarini, A.; Bolognesi, M. L.; Budriesi, R.; Canossa, M.;
Chiarini, A.; Spampinato, S.; Melchiorre, C. Design, Synthesis,
and Biological Activity of Methoctramine-Related Tetraamines
Bearing an 11-Acetyl-5,11-dihydro-6H-pyrido[2,3-b][1,4]benzo-
diazepin-6-one Moiety: Structural Requirements for Optimum
Occupancy of Muscarinic Receptor Subtypes As Revealed by
Symmetrical and Unsymmetrical Polyamines. J . Med. Chem.
1994, 37, 3363-3372.
(18) Melchiorre, C.; Minarini, A.; Spampinato, S.; Tumiatti, V.
Design, Synthesis and Biological Activity of Some Tetraamines
Related to Methoctramine and 4-DAMP. Bioorg. Med. Chem.
Lett. 1995, 5, 785-790.
Deter m in a tion of Dissocia tion Con sta n ts. In functional
experiments, dose ratios at the EC50 values of the agonists
were calculated at three to six antagonist concentrations, and
each concentration was tested from four to six times. The
results are expressed as pA2 values.24,25 Data are presented
as means ( SE of n experiments. Differences between mean
values were tested for significance by Student’s t-test.
Binding data were analyzed using LIGAND.33 Differences
in the slope of the curves were determined by the test of
parallelism as described by Tallarida and Murray34 and were
not different (p > 0.05). Scatchard plots were linear or almost
linear in all preparations. All Hill numbers (nH) were not
significantly different from unity (p > 0.05). Equilibrium
dissociation constants (Ki) were derived from the Cheng-
Prusoff equation,26 Ki ) IC50/(1 + L/Kd), where L and Kd are
the concentration and the equilibrium dissociation constant
of [3H]NMS or [3H]pirenzepine, respectively. pKi values are
the mean ( SE of three separate experiments performed in
triplicate.
Ack n ow led gm en t. This research was supported by
grants from University of Bologna (funds for selected

4160 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21
Bolognesi et al.
(19) Eberlein, W. G.; Engel, W.; Hasselbach, K. M.; Mayer, N.; Mihm,
G.; Rudolf, K.; Doods, H. Tricyclic Compounds as Selective
Muscarinic Antagonists: Structure Activity Relationships and
Therapeutic Implications. In Trends in Receptor Research;
Angeli, P., Gulini, U., Quaglia, W., Eds.; Elsevier Science
Publishers B.V.: Amsterdam, 1992; pp 231-249.
(20) Cohen, V. I.; Baumgold, J .; J in, B.; De La Cruz, R.; Rzeszotarski,
W. J .; Reba, R. C. Synthesis and Structure-Activity Relationship
of Some 5-[[[(Dialkylamino)alkyl]-1-piperidinyl]acetyl]-10,11-
dihydro-5H-benzo[b,e][1,4]diazepin-11-ones as M2-Selective An-
timuscarinics. J . Med. Chem. 1993, 36, 162-165.
(21) Melchiorre, C.; Quaglia, W.; Picchio, M. T.; Giardina`, D.; Brasili,
L.; Angeli, P. Structure-Activity Relationships among Methoc-
tramine-Related Polymethylene Tetraamines. Chain-Length and
Substitution Effects on M-2 Muscarinic Receptor Blocking
Activity. J . Med. Chem. 1989, 32, 79-84.
(22) Eberlein, W. G.; Engel, W. W.; Trummlitz, G.; Schmidt, G.;
Hammer, R. Tricyclic Compounds as Selective Antimuscarinics.
2. Structural-Activity Relationships of M1-Selective Antimusca-
rinics Related to Pirenzepine. J . Med. Chem. 1988, 31, 1169-
1174.
(23) This compound was synthesized from N,N′-dimethyl-1,8-oc-
tanediamine following the procedure reported by Atwell, G. J .;
Denny, W. A. Monoprotection of R,ω-Alkanediamines with the
N-Benzyloxycarbonyl Group. Synthesis 1984, 27, 1032-1033.
(24) Arunlakshana, D.; Schild, H. O. Some Quantitative Uses of Drug
Antagonists. Br. J . Pharmacol. 1959, 14, 48-58.
(25) Tallarida, R. J .; Cowan, A.; Adler, M. W. pA2 and Receptor
Differentiation: A Statistical Analysis of Competitive Antago-
nism. Life Sci. 1979, 25, 637-654.
(26) Cheng, Y. C.; Prusoff, W. H. Relationship Between the Inhibition
Constant (Ki) and the Concentration of Inhibitor Which Causes
50% Inhibition (I50) of an Enzymatic Reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(27) Quaglia, W.; Giardina`, D.; Marucci, G.; Melchiorre, C.; Minarini,
A.; Tumiatti, V. Structure-Activity Relationships Among Me-
thoctramine-Related Polymethylene Tetraamines. 3. Effect of the
Four Nitrogens on M2 Muscarinic Blocking Activity as Revealed
by Symmetrical and Unsymmetrical Polyamines. Farmaco 1991,
46, 417-434.
(28) Partition coefficients (log P) of tetraamines as free bases were
calculated with CLOGP Program implemented in the C-QSAR
program version 1.87, BioByte, Claremont, CA, 1994.
(29) Van Rossum, J . M. Cumulative Dose-Response Curves. II.
Technique for the Making of Dose-Response Curves in Isolated
Organs and the Evaluation of Drug Parameters. Arch. Int.
Pharmacodyn. 1963, 143, 299-330.
(30) Waelbroeck, M.; Gillard, M.; Robberecht, P.; Christophe, J .
Kinetic Studies of [3H]-N-Methylscopolamine Binding to Mus-
carinic Receptors in the Rat Central Nervous System: Evidence
for the Existence of Three Classes of Binding Sites. Mol.
Pharmacol. 1986, 30, 305-314.
(31) Michel, A. D.; Delmendo, R.; Stefanich, E.; Whiting, R. L. Binding
Characteristics of the Muscarinic Receptor Subtype of the NG
108-15 Cell Line. Arch. Pharmacol. 1989, 340, 62-67.
(32) Lazareno, S.; Buckley, N. J .; Roberts, F. F. Characterization of
Muscarinic M4 Binding Sites in Rabbit Lung, Chicken Heart,
and NG 108-15 Cells. Mol. Pharmacol. 1990, 38, 805-815.
(33) Munson, P. J .; Rodbard, D. LIGAND: A Versatile Computerised
Approach for the Characterization of Ligand Binding Systems.
Anal. Biochem. 1980, 107, 220-239.
(34) Tallarida, R. J .; Murray, R. B. Manual of Pharmacological
Calculations with Computer Programs, Version 4.2; Springer-
Verlag: New York, 1991.
J M981038D