Synthesis and antagonistic activity at muscarinic receptor subtypes of some 2-carbonyl derivatives of diphenidol

Article abstract of DOI:10.1016/S0968-0896(99)00124-8

A series of 2-carbonyl analogues of the muscarinic antagonist diphenidol bearing 1-substituents of different lipophilic, electronic, and steric properties was synthesized and their affinity for the M₂ and M₃ muscarinic receptor subty

Full text of DOI:10.1016/S0968-0896(99)00124-8

Bioorganic & Medicinal Chemistry 7 (1999) 1837±1844
Synthesis and Antagonistic Activity at Muscarinic Receptor
Subtypes of Some 2-Carbonyl Derivatives of Diphenidol
Lucilla Varoli, a,* Piero Angeli, b Silvia Burnelli, a Gabriella Marucci b
and Maurizio Recanatini a
a
Á
Dipartimento di Scienze Farmaceutiche, Universita degli Studi di Bologna, Via Belmeloro 6, 40126 Bologna, Italy
Á
Dipartimento di Scienze Chimiche, Universita degli Studi di Camerino, Via S. Agostino 1, 62032 Camerino (MC), Italy
b
Received 30 November 1998; accepted 1 March 1999
AbstractÐA series of 2-carbonyl analogues of the muscarinic antagonist diphenidol bearing 1-substituents of di€erent lipophilic,
electronic, and steric properties was synthesized and their anity for the M2 and M3 muscarinic receptor subtypes was evaluated by
functional tests. Two derivatives (2g and 2d) showed an M2-selective pro®le which was con®rmed by functional tests on the M1 and
M4 receptors. A possible relationship between M2 selectivity and lipophilicity of the 1-substituent was suggested by structure±
activity analysis. This work showed that appropriate structural modi®cation of diphenidol can lead to M2-selective muscarinic
antagonists of possible interest in the ®eld of Alzheimer's disease. # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
of a carbonyl group in position 2 of the butyl chain of
diphenidol led to compound 2 with enhanced anity for
both the M2 and the M3 receptor subtypes and a selec-
tivity ratio M3/M2 better than that of diphenidol (1,
Table 1).
The heterogeneity among muscarinic receptor has been
widely demonstrated, and ®ve unique gene sequences
coding for muscarinic receptors (m1±m5) have been
cloned;1 four of them (M1±M4) have been pharmacolo-
gically de®ned.2 The search for potent and selective
ligands is still in progress, mostly because of the need
for reliable pharmacological standards which would
make for a more complete classi®cation and represent
potential therapeutic agents. In fact, muscarinic recep-
tor subtypes are variously involved in secretory and
cardiovascular functions, smooth muscle control and in
central nervous system transmission.
Considering the important role of the benzilic OH in
position 1 of diphenidol (1), we designed and synthe-
sized a number of 1-substituted-2-carbonyl derivatives
(2a±i, Table 1) with the aim to evaluate the contribution
of 1-substituents to the anity for the muscarinic
receptor subtypes. Substituents were chosen in such a
way as to provide orthogonality and as much variability
as possible with respect to the classical lipophilic (p),
electronic (s) and steric (MR) parameters.
As a part of our studies on diphenidol (1,1-diphenyl-4-
piperidin-1-yl-butan-1-ol, 1), we initially turned our
attention to the part of the molecule interacting with the
hydrophobic receptor pocket introducing a substituent
in either position para or meta or ortho to one phenyl
ring of the lead compound.3 To extend our research, we
then modi®ed the intermediate chain connecting the
lipophilic head of the molecule to the cationic center
rendering it more polar or less ¯exible.4 The introduction
In this paper, we report the synthesis and the evaluation
by functional studies of the anity and selectivity for
muscarinic receptor subtypes of the title compounds,
and discuss their structure±activity and structure±selec-
tivity relationships.
Key words: Receptors; cholinergic activity; antagonists; substituent
e€ects.
* Corresponding author. Fax: +39-(0)51-209-734; e-mail: quark
@alma.unibo.it
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00124-8

1838
L. Varoli et al. / Bioorg. Med. Chem. 7 (1999) 1837±1844
Table 1. Anity values at the muscarinic M2 and M3 receptor subtypes, selectivity ratios, and physicochemical parameters of the studied compounds
pKb‹SEM
Selc
p
s*
MR
a
a
b
No.
R
M2
M3
M2/M3
2d
2a
2b
2c
2d
2e
2f
2g
OH
H
CH3
C2H5
C6H5
OCH3
OC2H5
SC2H5
SO2C2H5
Cl
7.48‹0.05
6.64‹0.06
7.13‹0.21
6.69‹0.01
7.06‹0.14e
7.60‹0.21
6.97‹0.07
7.19‹0.10e
5.20‹0.10
7.89‹0.02
[6.72‹0.02]
8.12‹0.03
6.54‹0.13
6.75‹0.06
6.10‹0.03
5.85‹0.14f
6.94‹0.01
6.21‹0.14
5.40‹0.02f
6.02‹0.07f
7.40‹0.16
7.02‹0.04
0.2
1.3
2.4
3.9
16
4.6
5.7
62
0.64
0.10
0.38
0.59
1.21
0.66
0.76
1.79
0.82
0.49
[Ð]
0.67
0.00
0.56
1.02
1.96
0.02
0.38
1.07
1.07
0.71
[Ð]
1.37
0.49
0.00
0.10
0.60
1.77
1.68
1.44
3.74
2.94
[Ð]
0.285
0.103
0.565
1.030
2.536
0.787
1.247
1.842
1.810
0.603
[Ð]
2h
2i
0.2
3.1
[0.5]
Diphenidol (1)d
a
Anity constants calculated from the equation log(DR-1)=log[ant] logKb for a single concentration of the antagonist, according to van
Rossum,23 are reported.
b
Antilog of the di€erence between the pKb values for M2 and M3 muscarinic receptor subtypes.
Log (M2/M3).
c
d
Ref 4.
pA2 values.
e
f
5
A decrease of the maximum e€ect of the reference agonist was present at 3Â10 M.
Methods
Chemistry
Series design
The compounds reported in this study were synthesized
by means of the Mannich reaction on the appropriate 1-
substituted-1,1-diphenylpropan-2-one (3a±h) and piperi-
dine hydrochloride in 2-methoxyethanol as high-boiling
solvent (2a±h, Scheme 1). Compound 2i was prepared
The substituents to be introduced in the benzilic posi-
tion of 2 (R in Table 1) were selected following, as far
as possible, the criteria of minimum redundancy and
maximum variance in physicochemical properties. Syn-
thetic restraints limited the choice to a small number of
groups. The selection was based on the consideration of
the lipophilic, electronic and steric properties which
were parameterized by means of the substituent con-
stants5 p, s* and MR, respectively. The R groups and
the physicochemical descriptors are shown in Table 1.
The squared correlation matrix6 of the parameters is
reported in Table 2, and it shows that a reasonable
orthogonality among the properties was obtained with
the selected series of substituents. This means that the
variation of each parameter within the series is inde-
pendent from the variation of the others as indicated by
the low values of the squared correlation coecients. It
is remarkable that almost no collinearity exists between
the lipophilic (p) and steric (MR) parameters.
Table 2. Squared correlation matrix for the physicochemical para-
meters of the substituents of Table 1
p
s*
MR
p
s*
MR
Ð
0.269
Ð
0.207
0.030
Ð
Scheme 1. (a) CH2O, piperidine hydrochloride, HCl, CH3OCH2±
CH2OH, re¯ux, (b) Cl2, THF, rt.

L. Varoli et al. / Bioorg. Med. Chem. 7 (1999) 1837±1844
1839
by reaction of 2a with chlorine in THF. Compounds
2a±i were obtained as hydrochloride salts, because of
the better characterizability and stability. Starting
ketones were commercially available (3a) or synthesized
according to literature methods (3b±f). Compound 3g
was obtained from 1-chloro-1,1-diphenylpropan-2-one
and ethylmercaptan, in presence of anhydrous CaCO3.
Oxidation of the ethylthio-derivative 3g with m-chloro-
perbenzoic acid (MCPBA) in CH2Cl2 led to the sulfone
3h (Scheme 2).
Pharmacology
All the compounds were tested on M2 (guinea pig heart)
and M3 (guinea pig ileum) muscarinic receptor subtypes
for their antimuscarinic activity using arecaidine pro-
pargyl ester (APE) as agonist. Compounds 2d and 2g
were also tested on M1 (rabbit vas deferens) and M4
(guinea pig lung) subtypes using p-Cl-McN-A-343 and
APE as agonists, respectively. The antagonist potency
was expressed as dissociation constant (pKb), and in
some cases pA2 values were also determined.
Figure 1. Schild plots obtained for the antagonists 2d and 2g on gui-
nea-pig atria force (M2). The points are the means of ®ve to seven
experiments. S.E.M.s are not reported for clarity and are less than
10%. Ordinate: log(DR-1) where DR=EC50 ratio measured from the
displacement of the agonist concentration-response curves by the
antagonist. Abscissa: log[antagonist]. Schild correlations, based on
three points, do not deviate from linearity, suggesting a competitive
behavior of the two antagonists.
Results and Discussion
The antagonistic activities at the M2 and M3 muscarinic
receptors and the subtype selectivity of the new com-
pounds studied are reported in Table 1. The corre-
sponding data of the previously published4 diphenidol
analogue 2 are also shown. Compounds 2a±i behave as
competitive (causing a parallel shift to the right of the
agonist dose±response curves) and noncompetitive (a
decrease of the maximum e€ect of the reference agonist
for the M3 receptor subtype in a range reaching 100-fold
di€erences (5±525, for compounds 2i and 2g, respec-
tively). The e€ects of the substitution are di€erent in the
case of the M2 receptor subtype. In fact, while the Cl
and OCH3 substituents slightly enhance the anity of
the corresponding compounds (2i and 2e, respectively)
for this subtype, all the others cause a moderate decrease
(up to sevenfold with compound 2a) in this parameter,
the only exception being the SO2C2H5 substituent which
causes a remarkable fall (190 times) in the anity of
5
was present at 1Â10 M) antagonists at M2 and M3
subtypes, respectively (Figs 1 and 2).
Examination of the pKb values of Table 1 reveals that
the introduction in position 1 of substituents di€erent
from the hydroxylic group, always decreases the anity
Figure 2. Experimental dose±response curves of APE on guinea-pig
ileum (M3) in the absence (&) and in the presence of 10 mM of 2d (^)
and 2g (&). The decrease of the maximum e€ect of APE suggests a
non-competitive behavior of the two antagonists.
Scheme 2. (a) C2H5SH, CaCO3, 80ꢀC, (b) MCPBA, CH2Cl2, rt.

1840
L. Varoli et al. / Bioorg. Med. Chem. 7 (1999) 1837±1844
compound 2h. As a consequence, the M2/M3 selectivity
pro®le of the antagonists 2a±g and 2i results reversed
compared with that of the reference compound 2, which
shows higher anity for the M3 subtype.
Since M1 receptors are apparently untouched in these
areas of the brain of Alzheimer's disease patients, a
possible mean of restoring central cholinergic tone is the
substitution of the reduced amount of acetylcholine
with exogenous M1 selective agonists.
Another interesting point is the di€erence in the selec-
tive pro®le observed between compounds 2g and 2h
carrying sul®de and sulfone substituents, respectively.
The 310-fold di€erence in selectivity (Table 1) between
these two antagonists suggests that the two oxygens
in the substituent of compound 2h could play a
role in binding with the muscarinic receptor sub-
sites. In particular, it seems possible to suppose that the
presence of two oxygen atoms improves the binding of
compound 2h at M3 subtype, while it hinders the bind-
ing of this compound at the M2 one. Moreover, as
already pointed out,7 it seems that oxygen and sulfur
can di€erentiate the behaviour of the muscarinic
antagonists: in fact, compounds 2f and 2g, which pre-
sent substituents di€ering for oxygen (2f) and sulfur
(2g), display a tenfold di€erence in selectivity (Table 1).
On the other hand, since stimulation of M2 presynaptic
autoreceptors will reduce, and blockade will enhance the
acetylcholine release, the use of M2 selective antagonists
could improve memory and learning, amplifying the
physiological cholinergic transmission. The increase of
acetylcholine release will be so possible by blocking the
presynaptic M2 receptors with selective M2 antagonists
possessing the ability to penetrate the blood±brain barrier
but leaving una€ected the postsynaptic M1 receptors.10
Attempts to quantitatively correlate the observed an-
ity data at both the M2 and M3 muscarinic receptor
subtypes with the substituents' physico-chemical prop-
erties were carried out following the traditional Hansch
method, but no signi®cant QSAR equation could be
obtained. As an alternative approach, a factor analysis6
was performed on the matrix formed by the biological
and physico-chemical data of Table 1. This method
allows one to extract the relevant information from a
data matrix and to calculate linearly independent
factors from the original variables. Each factor corres-
ponds to one of the eigenvalues of the matrix and, gen-
erally, only a few of them are needed to account for
most of the variance of the data. In the present case, as
shown in Table 4, the ®rst three factors are able to
explain 94% of the variance, indicating that some
intercorrelation exists within the six (biological and
physicochemical) variables. Looking at the loadings of
each original variable on to the relevant factors, it is
possible to see which variables are correlated. Con-
sidering Factor1, it appears that the selectivity ratio
(Sel) and p are correlated to some extent, although
other variables contribute to that factor. Factor2 is
mainly composed of pKb(M2) correlated with pKb(M3)
and MR; Factor3 is mostly loaded by s*. Factor1 also
shows a negative correlation between pKb(M3) and p,
which might explain the favorable e€ect of the latter
parameter on the M2 versus M3 selectivity. As regards
pKb(M2), it is not easy to say which physicochemical
variable (if any) has a quantitative e€ect on it, even if
MR (negative association in Factor2) might play a role.
Since the M2/M3 selectivity ratios for compounds 2g
and 2d showed interesting values (62 and 16, respec-
tively), we decided to complete the study of these two
antagonists testing them at M1 (rabbit vas deferens) and
M4 (guinea pig lung) muscarinic receptor subtypes as
well. In Table 3, the whole results for these two ligands
are reported, from which their M2-selective pro®le
clearly emerges. Actually, these compounds show M2
selectivity ratios ranging from 62 to >155 (2g) and
from 16 to >115 (2d), and therefore they can be con-
sidered as candidates for further development in an area
of great therapeutic interest as that of the treatment of
Alzheimer's disease.
It is well known that the cholinergic system is involved
in memory and learning.8 Although acetylcholine is not
the only neurotransmitter involved in Alzheimer's disease,
an illness presenting degenerative alterations which occur
in certain brain regions involved in the cognitive pro-
cesses, nevertheless the `cholinergic hypothesis' suggests
the restoration of central cholinergic tone in order to
alleviate and slow down the symptoms and the progress
of this disease.9 The presence of M1 and M2 muscarinic
receptors has been investigated in cortex and hippo-
campus with selective radioligands. M1 receptors con-
trol excitatory processes and are postsynaptically
located, whereas M2 receptors are mostly located pre-
synaptically and modulate the release of acetylcholine.10
The important outcome of this analysis is the relation-
ship between selectivity and lipophilicity, which can also
Table 3. Anity values, pKba or (pA2)‹SEM, and selectivity ratiosb for the muscarinic antagonists 2d and 2g on rabbit vas deferens (M1), guinea-
pig atria force (M2), ileum (M3) and lung (M4)
pKb or (pA2)‹SEM
No.
M1
M2
M3
M4
M2/M1
M2/M3
M2/M4
M3/M1
M3/M4
M4/M1
2d
2g
<5
<5
(7.06)‹0.14
(7.19)‹0.10
5.85c‹0.14
5.40c‹0.02
5.32c‹0.15
5.35c‹0.10
>115
>155
16
62
55
69
>7.1
>2.5
3.4
1.1
>2.1
>2.1
a
b
c
See note a of Table 1.
See note b of Table 1.
A decrease of the maximum e€ect of the reference agonist was present at 1Â10 M.
5

L. Varoli et al. / Bioorg. Med. Chem. 7 (1999) 1837±1844
1841
Table 4. Factor analysis
Factor1
Factor2
Factor3
Factor4
Factor5
Factor6
Eigenvalues:
Variance explained:
2.87
0.48
1.96
0.81
0.82
0.94
0.26
0.98
0.09
1.00
0.00
1.00
Variables
pKb(M2)
pKb(M3)
Sel
p
s*
MR
Loadings
0.15
0.75
0.91
0.91
0.43
0.67
0.90
0.59
0.24
0.30
0.53
0.62
0.41
0.18
0.20
0.03
0.72
0.25
0.00
0.24
0.25
0.19
0.10
0.30
0.08
0.01
0.06
0.22
0.11
0.14
0.00
0.00
0.00
0.00
0.00
0.00
be shown by the plot of Figure 3. The correlation
between the two variables is not sharp (r2=0.718), but
the trend is evident: increasing the lipophilicity of the
benzilic substituent (R) favors the anity for the M2
muscarinic receptor subtype rather than that for the M3
subtype. From inspection of the plot, it appears that the
e€ect of the SC2H5 substituent on selectivity is not only
strong, but also much stronger than its lipophilicity
would predict. The same seems to hold, even if at a les-
ser extent, also for the OC2H5 and OCH3 substituents.
to the binding energy, repulsion forces favor oxygen
(due to its smaller size), but dispersion and electrostatic
forces favor sulfur (due to its higher polarizability and
lower electron density).
The similarity of the molecular skeletons on which the
substituent groups are introduced (the carbonyl ana-
logue of diphenidol, 2, and adiphenine, 4, RˆH) leads
us to believe that the results reported in the present
paper, as regards the e€ect of the SC2H5 group on the
M2/M3 selectivity, can be ascribed to the same reasons
outlined above. In such a context, it is reasonable that
lipophilicity exerts a generalized in¯uence on selectivity
(perhaps limiting the M3 anity), while more speci®c
interactions determine the binding anity of each single
compound to the di€erent receptor subtypes.
The e€ect of the SC2H5 substituent on the M2 (and M1)
versus M3 selectivity was reported for a series of 2,2-
diphenyl-2-ethylthioacetic acid esters (4, RˆSC2H5)
developed as muscarinic antagonists structurally corre-
lated to adiphenine (4, RˆH).11 In an attempt to
understand the chemical reasons at the basis of such an
e€ect, Romanelli et al.12 performed an accurate theore-
tical study comparing the electronic and steric proper-
ties of the ethoxy- and ethylthio-groups. Based on some
ligand±receptor models, and postulating that these sub-
stituents bind to a receptor pocket made by several
aromatic residues, the authors reached the conclusion
that sulfur derivatives are favored over oxygen ana-
logues, because in the overall balance of the contributions
Finally, in a previous work,4 we hypothesized that the
bioactive conformation of these carbonyl-derivatives of
diphenidol might be one that exposes the pharmaco-
phoric groups to the receptor in a favorable arrange-
ment (5). At the light of the present results, namely the
relevant e€ect of the R substituents on both anity and
selectivity towards the M2 and M3 receptor, we can
con®rm that hypothesis and point out the critical role of
R in recognizing and binding the di€erent muscarinic
receptor subtypes.
Conclusion
A series of analogues of 2 was synthesized with the aim
of probing the e€ects of the benzilic substituent on the
anity and selectivity towards the M2 and M3 musca-
rinic receptor subtypes. In almost all the compounds
studied, replacing the OH group of 2 reduced the a-
nity for both M2 and M3 subtypes, but the introduction
of some substituent caused di€erent e€ects at the two
receptors. All substituents except SO2C2H5 reversed the
M3-selective pro®le of the parent compound (2) up to
the point that we obtained some interesting M2-selective
Figure 3. Plot of the M2/M3 selectivity (Sel) versus lipophilicity (p).

1842
L. Varoli et al. / Bioorg. Med. Chem. 7 (1999) 1837±1844
muscarinic antagonists. The increase in M2/M3 selectiv-
ity throughout the series roughly parallels the increase
in lipophilicity of the substituents, with the remarkable
exception of the SC2H5 group (compound 2g), which
leads to a much higher M2-selectivity than expected on
the basis of the lipophilicity alone. The most selective
compounds 2g and 2d were further characterized with
respect to the other muscarinic receptor subtypes (M1
and M4), and their M2-selective antimuscarinic pro®le
was con®rmed. In conclusion, our study led us to obtain
two M2-selective muscarinic antagonists, which might
be of potential interest in the ®eld of the Alzheimer's
disease treatment, when the present data will be con-
®rmed with human cloned receptor binding studies.
Further work will be devoted to the increase of the
antimuscarinic potency and to the improvement of the
physicochemical properties (lipophilicity) critical for
the pharmacokinetics of the central nervous system
drugs.
1,1-Diphenyl-1-methyl-4-piperidin-1-ylbutan-2-one hydro-
chloride (2b). From 1,1-diphenyl-1-methyl-propan-2-
one (3b)14 (1.97 g): 1.20 g (yield 38%), mp 190±191ꢀC
from abs. EtOH:Et2O (lit.15 mp 188±189ꢀC). H NMR
1
(DMSO-d6) d 1.33±1.70 (m, 6H), 1.93 (s, 3H), 2.70±2.80
(m, 2H), 3.01±3.28 (m, 6H), 7.11±7.14 (m, 4H), 7.29±
7.39 (m, 6H), 10.15 (bs, 1H exch. D2O).
1,1-Diphenyl-1-ethyl-4-piperidin-1-ylbutan-2-one hydro-
chloride (2c). From 1,1-diphenyl-1-ethyl-propan-2-one
(3c)16 (2.10 g): 1.18 g (yield 36%), mp 143±145ꢀC from
1
abs. EtOH:Et2O. H NMR (DMSO-d6) d 0.60 (t, 3H,
J=7.3 Hz), 1.27±1.64 (m, 6H), 2.39 (q, 2H, J=7.3 Hz),
2.60±2.75 (m, 2H), 2.90±3.23 (m, 6H), 7.20±7.38 (m,
10H), 10.09 (bs, 1H exch. D2O).
1,1,1-Triphenyl-4-piperidin-1-ylbutan-2-one hydrochloride
(2d). From 1,1,1-triphenyl-propan-2-one (3d)17 (2.52 g):
1.59 g (yield 43%), mp 180±181ꢀC from abs. EtOH:
1
Et2O. H NMR (DMSO-d6) d 1.26±1.63 (m, 6H), 2.28±
2.72 (m, 2H), 2.87±2.96 (m, 2H), 3.00±3.14 (m, 4H),
7.26±7.39 (m, 15H), 10.28 (bs, 1H exch. D2O).
Experimental
Chemistry
1,1-Diphenyl-1-methoxy-4-piperidin-1-ylbutan-2-one
hydrochloride (2e). From 1,1-diphenyl-1-methoxy-pro-
pan-2-one (3e)18 (2.11 g): 0.46 g (yield 14%), mp 164±
Melting points were taken on Electrothermal open
capillary apparatus and are uncorrected. Elemental
analysis was performed for compounds 2a±i and the
results (not shown) were within ‹0.4% of the theor-
etical values. Infrared spectra (IR) were recorded on a
Perkin±Elmer 683 instrument for all compounds and
were consistent with the assigned structures; because of
165ꢀC from abs. EtOH:Et2O. H NMR (DMSO-d6) d
1
1.24±1.68 (m, 6H), 2.65±2.80 (m, 2H), 3.00 (s, 3H),
3.08±3.28 (m, 6H), 7.30±7.40 (m, 10H), 10.41 (bs, 1H
exch. D2O).
1,1-Diphenyl-1-ethoxy-4-piperidin-1-ylbutan-2-one hydro-
chloride (2f). From 1,1-diphenyl-1-ethoxy-propan-2-one
(3f)19 (2.24 g): 0.51 g (yield 15%), mp 159±161ꢀC from
1
the lack of unusual features, they are not included. H
NMR spectra were registered on a Varian VXR 300
spectrometer, peak positions are given in parts per mil-
lion (d) relative to the standard chemical shift of the
solvent. Merck silica gel 60 (230±400 mesh) was used for
column chromatography. Thin-layer chromatography
(Merck silica gel 60 F254 analytical plates) was used to
monitor reactions. The term `dried' refers to the use of
anhydrous sodium sulfate.
abs. EtOH:Et2O (lit.20 mp 160ꢀC). H NMR (DMSO-
1
d6) d 1.16 (t, 3H, J=7.0 Hz), 1.30±1.63 (m, 6H), 2.62±
2.80 (m, 2H), 3.05 (q, 2H, J=6.9 Hz), 3.12±3.23 (m,
6H), 7.28±7.41 (m, 10H), 10.10 (bs, 1H exch. D2O).
1,1-Diphenyl-1-ethylthio-4-piperidin-1-ylbutan-2-one
hydrochloride (2g). From 1,1-diphenyl-1-ethylthio-pro-
pan-2-one (3g) (2.38 g): 0.89 g (yield 25%), mp 149±
150ꢀC from CHCl3:Et2O. H NMR (DMSO-d6) d 1.09
1
General procedure for the preparation of 1,1-diphenyl-4-
piperidin-1-ylbutan-2-one hydrochloride (2a) and of 1-
substituted derivatives 2b±h. A solution of 1,1-diphenyl-
propan-2-one (3a) (1.85 g, 8.8 mmol), paraformaldehyde
(0.64 g) and piperidine hydrochloride (1.48 g, 12.2 mmol)
in 2-methoxyethanol (10 mL) was re¯uxed with stirring
at 140ꢀC for 10 min. A suspension of paraformaldehyde
(0.64 g) in 2-methoxyethanol (3 mL) was added during
20 min. Concentrated hydrochloric acid (0.5 mL) was
added and re¯uxing continued for a further 10 min to
produce a clear solution. The cooled solution was
poured into brine (10 mL) and extracted with CHCl3
(2Â30 mL). The combined organic layers were washed
with brine and dried. The CHCl3 was evaporated and
Et2O was added to induce separation of 2a as white
solid, which was collected by ®ltration and recrystallized
from abs. EtOH to give 2.06 g (yield 68%); mp 206±
(t, 3H, J=7.4 Hz), 1.30±1.71 (m, 6H), 2.08 (q, 2H,
J=7.4 Hz), 2.70±2.86 (m, 2H), 2.90±3.04 (m, 2H), 3.19±
3.27 (m, 4H), 7.28±7.51 (m, 10H), 10.13 (bs, 1H exch.
D2O).
1,1-Diphenyl-1-ethylsulfonyl-4-piperidin-1-ylbutan-2-one
hydrochloride (2h). From 1,1-diphenyl-1-ethylsulfonyl-
propan-2-one (3h) (2.66 g): 0.38 g (yield 10%), mp 160±
ꢀ
1
162 C from CHCl3:Et2O. H NMR (CDCl3) d 1.19 (t,
3H, J=7.5 Hz), 1.40±2.16 (m, 8H), 2.40±2.64 (m, 2H),
2.86 (q, 2H, J=7.5 Hz), 3.04±3.25 (m, 4H), 7.44±7.58
(m, 10H), 9.45 (bs, 1H exch. D2O).
1-Chloro-1,1-diphenyl-4-piperidin-1-ylbutan-2-one hydro-
chloride (2i). Chlorine was passed into a stirred suspen-
sion of 2a (0.50 g) in THF (4 mL). Re¯uxing began after
few min and the solution cleared. After a further 5 min
the passage of chlorine was stopped and the solvent
removed. The resulting residue was recrystallized
from CHCl3:Et2O to give 0.25 g of 2i (yield 45%), mp
208ꢀC (lit.13 mp 204±205ꢀC). H NMR (DMSO-d6) d
1
1.30±1.70 (m, 6H), 2.70±2.85 (m, 2H), 3.10±3.30 (m,
6H), 5.44 (s, 1H), 7.2±7.34 (m, 10H), 9.98 (bs, 1H exch.
D2O).

L. Varoli et al. / Bioorg. Med. Chem. 7 (1999) 1837±1844
1843
127±129ꢀC. H NMR (DMSO-d6) d 1.34±1.73 (m, 6H),
one being discarded and the second one being taken as
the control.
1
2.70±2.86 (m, 2H), 3.20±3.35 (m, 6H), 7.26±7.49 (m,
10H), 10.45 (bs, 1H exch. D2O).
Guinea pig stimulated left atria. The heart was rapidly
removed, and the right and left atria were separately
excised. Left atria were mounted in PSS (the same used
for ileum) at 30ꢀC and stimulated through platinum
electrodes by square-wave pulses (1 ms, 1 Hz, 5±10 V)
(Tetra Stimulus, N. Zagnoni). Inotropic activity was
recorded isometrically. Tissues were equilibrated for 2 h
and a cumulative dose±response curves to APE was
constructed.
1,1-Diphenyl-1-ethylthio-propan-2-one (3g). 1-Chloro-1,1-
diphenylpropan-2-one21 (7.34 g, 30 mmol) and anhy-
drous CaCO3 (4.5 g, 45 mmol) were added to ethyl-
mercaptan (25 mL) in a steel bomb and kept at 80ꢀC for
4 days. After cooling, the excess of ethylmercaptan was
carefully removed and the residue treated with a 10%
solution of K2CO3 and extracted with Et2O (2Â50 mL).
The organic phase was washed, dried and evaporated
and a€orded an oily residue, which was puri®ed on a
silica gel column (toluene), to give 4 g (yield 49%) of
Guinea pig lung strips. The lungs were rapidly removed
and strips of peripheral lung tissue were cut either from
the body of a lower lobe with the longitudinal axis of
the strip parallel to the bronchus or from the peripheral
margin of the lobe. The preparations were mounted,
with a preload of 0.3 g, in PSS with the following com-
compound 3g, mp 40±42ꢀC from petroleum ether. H
1
NMR (DMSO-d6) d 1.06 (t, 3H, J=7.5 Hz), 2.09 (q,
2H, J=7.5 Hz), 2.15 (s, 3H), 7.25±7.47 (m, 10H).
1,1-Diphenyl-1-ethylsulfonyl-propan-2-one (3h). To
a
.
solution of 3g (2.7 g, 10 mmol) in CH2Cl2 (40 mL)
MCPBA (50%) (6.90 g, 20 mmol) in CH2Cl2 (30 mL)
was added dropwise, at room temperature, and left
under stirring for 2 h at room temperature. The reaction
mixture was extracted with a 10% solution of Na2CO3
(2Â20 mL) and dried. The solvent was evaporated and
the residue was puri®ed on a silica gel column (toluene:
acetone, 98:2), to give 2.0 g (yield 66%) of 3h, mp 142±
position (mM): NaCl (118.78), KCl (4.32), CaCl2 2H2O
.
(2.52), MgSO4 7H2O (1.18), KH2PO4 (1.28), NaHCO3
(25), glucose (5.55). Contractions were recorded iso-
tonically at 37ꢀC after tissues were equilibrated for 1 h,
then two cumulative dose±response curves to APE
(0.01, 0.1, 1, 10, 100 mM) were obtained at 45 min inter-
vals, the ®rst one being discarded and the second one
being taken as the control.
144ꢀC from MeOH. H NMR (CDCl3) d 1.21 (t, 3H,
1
J=7.5 Hz), 2.09 (s, 3H), 2.89 (q, 2H, J=7.5 Hz), 7.44±
7.47 (m, 6H), 7.61±7.65 (m, 4H).
Rabbit stimulated vas deferens. This preparation was set
up according to Eltze.22 Vasa deferentia were carefully
dissected free of surrounding tissue and were divided
into four segments, two prostatic portions of 1 cm and
two epididymal portions of approximately 1.5 cm
length. The four segments were mounted in PSS with
the following composition (mM): NaCl (118.4), KCl
(4.7), CaCl2 (2.52), MgCl2 (0.6), KH2PO4 (1.18),
Pharmacology
General considerations. Male guinea pigs (200±300 g)
and male New Zealand white rabbits (3.0±3.5 kg) were
killed by cervical dislocation. The organs required were
set up rapidly under 1 g of tension in 20 mL organ baths
containing physiological salt solution (PSS) maintained
at an appropriate temperature (see below) and aerated
with 5% CO2±95% O2. Dose±response curves were
constructed by cumulative addition of the reference
agonist. The concentration of agonist in the organ bath
was increased approximately threefold at each step,
with each addition being made only after the response
to the previous addition had attained a maximal level
and remained steady. Following 30 min of washing, tis-
sues were incubated with the antagonist for 30 min, and
a new dose±response curve to the agonist was obtained.
Contractions were recorded by means of a force trans-
ducer connected to a two-channel Gemini polygraph
(U. Basile). In all cases, parallel experiments in which
tissues received only the reference agonist were run in
order to check any variation in sensitivity.
6
NaHCO3 (25), glucose (11.1); 10 M yohimbine was
included to block a2-adrenoceptors. The solution was
maintained at 30ꢀC and tissues were stimulated through
platinum electrodes by square-wave pulses (0.1 ms,
2 Hz, 10±15 V). Contractions were measured iso-
metrically after tissues were equilibrated for 1 h, then a
cumulative dose±response curve to p-Cl-McN-A-343
was constructed.
Determination of antagonist potency. To quantify
antagonist potency, pKb values were calculated from the
equation pKb=log(DR-1) log[B], where DR is the
ratio of ED50 values of agonist after and before treat-
ment with one or two antagonist concentration [B].23 In
some cases, antagonist potency is expressed in terms of
pA2, estimated by Schild plots constrained to slope
1.0, as required by the theory.24,25
Guinea pig ileum. Two-centimeter-long portions of
terminal ileum were taken at about 5 cm from the
ileum±cecum junction and mounted in PSS, at 37ꢀC, of
the following composition (mM): NaCl (118), NaHCO3
Statistical analysis. Values are given as mean ‹ stan-
dard error of four or ®ve independent observations.
Student's t-test was used to assess the statistical sig-
ni®cance of the di€erence between two means.
.
(23.8), KCl (4.7), MgSO4 7H2O (1.18), KH2PO4 (1.18),
CaCl2 (2.52), and glucose (11.7). Tension changes were
recorded isotonically. Tissues were equilibrated for
30min, and dose±response curves to arecaidine propargyl
ester (APE) were obtained at 30 min intervals, the ®rst
Acknowledgements
This work was supported by a grant from MURST.

1844
L. Varoli et al. / Bioorg. Med. Chem. 7 (1999) 1837±1844
11. Scapecchi, S.; Angeli, P.; Dei, S.; Gualtieri, F.; Marucci,
References
G.; Romanelli, M. N.; Teodori, E. Pharm. Pharmacol. Lett.
1993, 2, 220.
12. Romanelli, M. N.; Holtje, H.-D.; Scapecchi, S. Quant.
Struct. -Act. Relat. 1995, 14, 126.
1. Buckley, N. J.; Bonner, T. I.; Buckley, C. M.; Brann, M. R.
Mol. Pharmacol. 1989, 35, 469.
2. Caul®eld, M. P. Pharmac. Ther. 1993, 58, 319.
13. Katz, L.; Karger, L. S. J. Am. Chem. Soc. 1952, 74, 4085.
14. Sisido, K.; Nozaki, H. J. Am. Chem. Soc. 1948, 70, 776.
15. Zaugg, H. E.; Freifelder, M.; Horrom, B. W. J. Org.
Chem. 1950, 15, 1191.
3. Varoli, L.; Burnelli, S.; Budriesi, R.; Guarnieri, A.; Chiarini,
A.; Recanatini, M. Med. Chem. Res. 1994, 4, 588.
4. Varoli, L.; Burnelli, S.; Budriesi, R.; Guarnieri, A.; Chiarini,
A.; Recanatini, M. Med. Chem. Res. 1996, 6, 571.
5. Parameters were obtained from the recent compilation by
Hansch, C.; Leo, A.; Hoekman, D. Exploring QSAR. Hydro-
phobic, Electronic, and Steric Constants, ACS Professional
Reference Book: American Chemical Society: Washington
(DC), 1995.
6. Calculations were performed by means of the C-QSAR
program Ver. 1. 87, BioByte Corp., Claremont (CA), 1994.
7. Angeli, P. Il Farmaco 1998, 53, 1.
8. Drachman, D. A.; Leavitt, J. L. Arch. Neurol. 1974, 30, 113.
9. Gualtieri, F.; Dei, S.; Manetti, D.; Romanelli, M. N.; Sca-
pecchi, S.; Teodori, E. Il Farmaco 1995, 50, 489.
16. Meerwein, H. Ann. 1919, 419, 157.
17. McKenzie, A.; Myles, J. R. Chem. Ber. 1932, 65, 209.
18. Stevens, C. L.; Lenk, C. T. J. Org. Chem. 1954, 19, 538.
19. Temnikova, T. I. J. Gen. Chem. 1945, 15, 514.
20. Schlesinger, A.; Gordon, S. M. U.S. 3,073,842. Chem.
Abstr. 1963, 59, 1652.
21. Richard, G. Compt. Rend. 1935, 200, 1944.
22. Eltze, M. Eur. J. Pharmacol. 1988, 151, 205.
23. Van Rossum, J. M. Arch. Int. Pharmacodyn. Ther. 1963,
143, 299.
24. Arunlakshana, O.; Schild, H. O. Br. J. Pharmacol. 1959,
14, 48.
25. Tallarida, R. J.; Cowan, A.; Adler, M. W. Life Sci. 1979,
25, 637.
10. Eberlein, W. G.; Engel, W.; Hasselbach, K. M.; Mayer, N.;
Mihm, G.; Rudolf, K.; Doods, H. Trends in Receptor
Research, Angeli, P.; Gulini, U.; Quaglia, W., Eds.; Elsevier
Science B.V.: Amsterdam, 1992, Vol 18, pp 231±249.