Design, synthesis, and SAR analysis of novel selective σ1 ligands

Article abstract of DOI:10.1016/j.bmc.2006.10.048

A new series of arylalkyl- and alkenylamines was designed, synthesized, and evaluated for binding to σ₁ and σ₂ receptors. Many compounds exhibited nanomolar affinity for σ₁ subtype receptor with good selectivity over σ

Full text of DOI:10.1016/j.bmc.2006.10.048

Bioorganic & Medicinal Chemistry 15 (2007) 771–783
Design, synthesis, and SAR analysis of novel selective r ligands
1
a,
a
a
b
*
Simona Collina, Guya Loddo, Mariangela Urbano, Laura Linati,
c
d
d
e
Athos Callegari, Francesco Ortuso, Stefano Alcaro, Christian Laggner,
Thierry Langer, Orazio Prezzavento, Giuseppe Ronsisvalle and Ornella Azzolina
e
f
f
a
a
Dipartimento di Chimica Farmaceutica, Universit a` di Pavia, Viale Taramelli 12, 27100 Pavia, Italy
b
Centro Grandi Strumenti, Universit a` di Pavia, Via Bassi 21, 27100 Pavia, Italy
Dipartimento di Scienze della Terra, Universit a` degli Studi di Pavia, Via Ferrata 1, 27100 Pavia, Italy
c
d
Dipartimento di Scienze Farmacologiche ‘Complesso Nin ı` Barbieri’ Universit a` di Catanzaro ‘Magna Graecia’,
8
8021 Roccelletta di Borgia (CZ), Italy
e
Department of Pharmaceutical Chemistry and Center for Molecular Biosciences (CMBI), University of Innsbruck,
A-6020 Innsbruck, Austria
f
Dipartimento di Scienze Farmaceutiche, Universit a` di Catania, Viale A. Doria 6, 95125 Catania, Italy
Received 8 September 2006; revised 18 October 2006; accepted 23 October 2006
Available online 25 October 2006
Abstract—A new series of arylalkyl- and alkenylamines was designed, synthesized, and evaluated for binding to r and r receptors.
1
2
Many compounds exhibited nanomolar affinity for r
was conducted in order to rationalize the experimental data, and the structure-receptor affinities are discussed.
2006 Elsevier Ltd. All rights reserved.
1 2
subtype receptor with good selectivity over r . A molecular modeling study
ꢀ
1
. Introduction
duction systems, such as the glutamatergic, cholinergic,
and catecholaminergic systems, are sensitive to r -medi-
1
9
Although sigma receptors were discovered about 30 years
ago as a subtype of opioid receptors, the importance of
their biological properties is not yet completely clari-
fied. At present it is accepted that sigma receptors are
a unique family of proteins different from opioid and
N-methyl-D-aspartate (NMDA) phencyclidine recep-
ated neuromodulation. The r receptor has been cloned,
1
and the purified protein is a 25–30 kDa single polypeptide
characterized by a single transmembrane domain an-
chored in mitochondrial membranes and endoplasmic
1
,2
1
0
reticulum. The deduced amino acid sequence of the
purified protein, seemingly correspondent to r receptor,
1
3
tors. The sigma receptors bind numerous xenobiotics
does not resemble that of any known mammalian protein.
Progesterone has been suggested to be a possible endog-
enous ligand, since it was found to inhibit tritiated (+)-
SKF10,047 ((+)-N-allylnormetazocine) binding in rat
of unrelated classes of compounds (Fig. 1) such as many
clinical drugs used in psychiatric disorders including
haloperidol, imipramine, and selective serotonin reup-
take inhibitors. Pharmacological studies have identified
two distinct subtypes of sigma receptor, namely r and r ,
2
11
brain at nanomolar concentrations. However, other
steroids, such as pregnenolone, dehydroepiandrosterone
(DHEA), and testosterone, have moderate affinity. Preg-
1
2
which are expressed in various areas of the brain and in
4
–7
peripheral organs. The r subtype, classified as a typi-
nenolone and DHEA activate the r protein, while pro-
1
1
1
2
cal membrane receptor, can be considered an important
intracellular regulatory protein since it can be translocat-
ed, when activated, from the endoplasmic reticulum,
gesterone is the most potent inactivator. Thus, sigma
receptors could mediate some aspects of steroid-induced
mental disturbances and may have neuroprotective prop-
13
8
where it is present under resting conditions. Many trans-
erties. Since endogenous ligands for the r receptor
1
have not yet been identified and its physiological function
is still to be wholly elucidated, it might be better defined
Keywords: Arylalkylamines; Sigma receptors; Molecular modeling;
SAR studies.
r1 protein. Likewise, it may be better to consider its role
*
Corresponding author. Tel.: +39 382 987376; fax: +39 382 422975;
e-mail: simona.collina@unipv.it
as an activator/inactivator rather than agonist/
antagonist.
0
968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.10.048

772
S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
(
+)-Pentazocine
SKF 10,047
N
N
HO
HO
(
+)-1,2,3,4,5,6-Hexahydro-6,11-dimethyl-3-
(+)-1,2,3,4,5,6-Hexahydro-3-(3-methyl-2-butenyl)
-2,6-methano-3-Benzazocin-8-ol
(
3-methyl-2-butenyl)-2,6-methano-3-Benzazocin-8-ol
Haloperidol
Igmesine
OH
O
N
Cl
N
F
4
-[4-(4-Chlorophenyl)-4-hydroxy-1-piperidinyl-1'-
Cyclopropylmethyl-(1-ethyl-1,4-diphenyl-but-3-
enyl)-methyl-amine
(
4-fluorophenyl)-1-butanone
DTG
Fenpropimorph
NH
N
O
N
N
H
H
1,3-Di-O-tolylguanidine
4-(3-(4-tert-butylphenyl)-2-methylpropyl)
2,6-dimethylmorpholine
-
Figure 1.
The r subtype of receptor, a 18–21 kDa protein, has
sigma ligands represents the only valid approach for
characterizing sigma receptor subtypes better.
2
1
2
not yet been cloned. The binding profile of ligands
for r receptors, unlike those for r , exhibits low affini-
2
1
ties and no stereoselectivity for dextrorotatory benzo-
morphan. r2 receptor activation induces several
biological effects including changes in cell morphology
and apoptosis, and produces both transient and sus-
tained increases in calcium ions. The important roles
played by sigma receptors in certain biological systems
suggest that the development of novel high-selective
sigma ligands may provide potential drugs for CNS-re-
lated disorders and cancer diseases, since both sigma
receptor subtypes are distributed within the central
There are published reports on several highly selective r1
ligands synthesized to gain SAR information for a more
detailed understanding of the pharmacological function
of the r protein. In the last years molecular modeling
1
1
3
studies have been performed to define the binding phar-
macophore model for different classes of sigma ligands.
Selective r pharmacophore models have been proposed
1
1
8,19
by Glennon and Gund. Recently, results of our re-
search on a series of naphthylalkylamines indicated that
(R/S)-4-(dimethylamino)-2-(naphthalen-2-yl)butan-2-ol
1 (Fig. 2) is a ligand for r receptors. Compound 1 had
nervous system (particularly r ) and over-expressed in
1
1
5
,7
several tumor cell lines (particularly r2). r1 receptor
neuromodulation could be the target of therapeutic
strategies in many neurodegenerative diseases. For
example, the relevant antidepressant efficacy of activa-
been synthesized, together with some other compounds,
to determine its possible involvement in opioid analge-
2
0
sia. Pharmacological investigations using the hot plate
test showed that this compound exerts antinociceptive
properties and binding assays evidenced a significant
affinity for r receptors (K = 25 nM).
tors of r protein, especially progesterone, could be an
1
effective approach to alleviate symptoms of depression
1
i
1
4
in Alzheimer’s disease. It has been pointed out that
r1 selective agonists have potent anti-amnesic, antide-
pressant, and anti-stress effects, while selective antago-
1
5
nists have antipsycotic properties. r1 receptors are
also involved in the modulation of opioid analgesia:
(
+)-pentazocine and 1,3-di(2-tolyl)guanidine (DTG)
N
potently antagonize antinociceptivity, whereas r antag-
1
OH
1
6,17
onists enhance opioid analgesia.
For this reason,
there is a growing interest in the potential clinical use
of sigma ligands. Moreover, the discovery of selective
Figure 2.

S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
773
the afore-mentioned pharmacophore models. The sim-
pler model, proposed by Glennon et al., consisted of
an amine-binding site flanked by a primary hydrophobic
aromatic domain A and a secondary hydrophobic group
Ar
N
Ar
N
R
R
Series I
Series II
18
B associated with a region of bulk tolerance. The more
2
-6, 12-13
7-11, 14-15
recent model by Gund complied with the topological
arrangements of A, B, and N, but took into account
an additional site corresponding to an electronegative
Comp.
Ar
R
1
9
atom such as oxygen or sulfur.
2
, 7
CH3
A preliminary analysis of compound 1 was carried out
with the aim of evaluating the possible role of oxygen
in the r binding with respect to Gund’s model. This
1
conformational analysis was performed taking into ac-
count the physiological environment of the drug-recep-
tor interaction (see Section 5.7). Despite the relatively
high flexibility of the side chain of compound 1, the dis-
tances between the alcoholic oxygen and the amino
nitrogen as well as the aromatic centroid were not com-
patible with Gund’s model in the most stable conform-
ers (Fig. 4).
CH3
2
,8
CH3
CH3
CH3
4
,9
HO
On the basis of these findings, we prepared the olefinic
and saturated analogues 2–6 (Series I) and 7–11 (Series
II) (Fig. 3) modifying both the aromatic nucleus and
the 3-carbon dimethylamino side chain of the hit
compound.
5
,10
H CO
3
6
,11
Moreover, the evaluation of structure–activity relation-
ships of literature compounds suggested introducing a
bulk substituent at the amino group. On the basis of
these considerations, N-methyl-N-benzyl analogues of
the most interesting compounds of both series were pre-
pared, bearing the naphthalenic or the biphenylic moiety
1
2,14
H C
2
1
8
H C
2
1
3,15
12–13 (Series I) and 14–15 (Series II), in order to evalu-
Figure 3.
ate the influence of the amino substituent on the binding
mode.
On the basis of these encouraging results, in this paper,
we report on the design, synthesis, and biological evalu-
ation of new arylalkyl- and alkenylamines with different
functionalities both on the aromatic moiety and the
alkylamine chain (Fig. 3). We considered r selective
1
pharmacophore models and investigated synthetic meth-
ods in order to prepare compounds with two indepen-
dent points of diversity. SAR information on our
novel r ligands was drawn on the basis of a new phar-
1
macophore model developed by us.
2. Chemistry
2
.1. Compounds’ design
In order to develop r selective ligands, we selected
1
(
R/S)-4-(dimethylamino)-2-(naphthalen-2-yl)butan-2-ol
1 (Fig. 2) as the hit compound; we designed new ana-
logues, using a computational approach and taking into
account the selective r pharmacophore binding models
1
1
8,19
proposed by Glennon and Gund.
Figure 4. Global minimum conformer of 1 in polytube rendering. The
distances reported in violet dashed lines were measured from the
oxygen, the protonable nitrogen, and a dummy atom centered into the
naphthalene ring. Hydrogen atoms were removed for clarity.
The compounds were designed according to a confor-
mational study that evaluated their degree of match with

774
S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
With regard to the drug-like properties, a preliminary
evaluation of all the new derivatives was performed,
considering the Lipinski’s rule-of-five (molecular weight,
H-bond donors, H-bond acceptors, and calculated logP
derivatives, that is, without hydroxyl or methoxyl
functions.
2.2. Synthesis
2
1
values). The logP values of the examined compounds
were calculated using the Daylight computational meth-
od (web version 4.91) that combines fragment lipophilic-
ity contributions carefully parameterized with
experimental data and calculated ‘from scratch’ values
The general procedure for the preparation of the alkyl
and alkenyl amines is reported in Scheme 1. We started
our synthetic approach with the synthesis of b-aminok-
etones 16 and 17, essentially prepared according to the
convenient methodology described in our previous
2
2
(
analogue compounds by comparison with chromato-
Table 1). We had already validated this method for
2
0
work, with suitable modifications (Scheme 2). The Mi-
chael addition of the dimethylamine to but-3-en-2-one
was performed in absolute ethanol and glacial acetic
acid producing 16, purified by fractional distillation.
For compound 17, benzyl-methyl-amine was employed
and the reaction was performed in anhydrous toluene,
thus producing the desired product in good yields, with-
out any further purification. Thereafter, the compounds
of Series I, 2–6 and 12–13, were prepared by nucleophil-
ic addition of the aromatic anion to b-aminoketones 16
and 17, followed by direct dehydration in order to avoid
isolation of the alcoholic intermediate (Scheme 2). An
acid–base work-up allowed the isolation of the desired
olefinic compounds with a good purity. In the case of
compound 2, quenching with HCl 37% led to the con-
current cleavage of the tetrahydropyranyl protecting
group of the aromatic precursor 2-bromo-6-OTHP-
naphthalene.
2
3
graphic hydrophobicity values. All the designed com-
pounds were within the limits of Lipinski’s rule-of-five
with the exception of the compounds bearing an N-ben-
zyl group. Despite this, compounds 12–15 were consid-
ered in the design strategy with the different purpose
of testing the maximum length of the ligands that are
able to fit into the r cleft. Therefore, we limited our
1
evaluation to biphenyl and simple naphthalene aryl
Table 1.
Ar
N
Ar
N
R
R
(
E)-2-6,12-13
(R/S)- 7-11, 14-15
Series I
Series II
1
H NMR analyses of all crude compounds of Series I re-
Compound
Ar
R
Clog P
vealed the formation of a C2–C3 double bond and the
absence of the signals related to C1–C2 olefinic regioi-
somers (Scheme 2); the H NMR of compound 12 is
1
2
3
4
5
6
7
8
9
Naphth-2-yl
Naphth-2-yl
Naphth-1-yl
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
3.052
3.963
3.963
3.296
3.882
4.677
4.067
4.067
3.400
3.986
4.781
5.921
6.635
6.025
6.739
1
reported in Figure 5 as an example. Regarding the con-
figuration at the C2–C3 double bond, in all cases only
one isomer was obtained (see Section 2.3). Actually,
6-Hydroxy-naphth-2-yl
6-Methoxy-naphth-2-yl
Biphen-4-yl
1
H NMR analyses of compounds 2–6 and 12–13 showed
only one signal corresponding to the olefinic proton; for
the configurational assignment, see Section 2.3.
Naphth-2-yl
Naphth-1-yl
6-Hydroxy-naphth-2-yl
6-Methoxy-naphth-2-yl
Biphen-4-yl
1
1
1
1
1
1
0
1
2
3
4
5
In conclusion, this procedure enabled easy generation of
novel compounds at high purity.
Naphth-2-yl
Biphen-4-yl
C
C
C
C
6
6
6
6
H
5
5
5
5
H
H
H
Naphth-2-yl
Biphen-4-yl
Series II compounds 7–11 were synthesized (Scheme 3)
by catalytic reduction of the corresponding alkenes
HN
+
R
O
CH3
CH3
O
Ar -
+
N
Ar
N
Ar
N
R
R
R
16, 17
Series I
Series II
Comp. 16 R = CH
3
Comp. 17 R = CH
2
6 5
-C H
Scheme 1.

S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
775
CH2
Ar
N
*HCl
R
CH3
Br
a, b, c
Ar
Ar
N
OH
R
CH3
Ar
N
*HCl
Comp.
Ar
R
(E)-2-6
12-13
R
2
3
4
5
6
naphth-2-yl
naphth-1-yl
CH
CH
CH
CH
CH
3
3
3
3
3
6
6
6-hydroxy-naphth-2-yl
6-methoxy-naphth-2-yl
biphen-4-yl
naphth-2-yl
biphen-4-yl
12
CH
CH
2
C
C
H
H
5
5
13
2
Scheme 2. Synthesis of (E)-4–8 and (E)-14–15. Reagents and conditions: (a) t-BuLi; (b) 2 or 3, anhydrous Et
2
O, À78 ꢁC; (c) HCl 37%, rt.
CH3
Ar
N
a, c
R
(
R,S)-7-11
CH3
Ar
N
R
(
E)-2-6
1
2, 13
b, c
CH3
Ar
N
R
(
R,S)-14, 15
Comp
Ar
R
7
8
9
naphth-2-yl
naphth-1-yl
6-hydroxy-naphth-2-yl
6-methoxy-naphth-2-yl
biphen-4-yl
CH
CH
CH
CH
CH
3
3
3
3
3
1
0
1
4
5
1
1
1
naphth-2-yl
biphen-4-yl
CH
CH
2
C
C
6
H
H
5
5
2
6
Scheme 3. Synthesis of (R/S)-9–13 and (R/S)-16–17. Reagents and conditions: (a) H
EtOH, rt; (c) HCOONH
, EtOH, 10% Pd/C, 120ꢁ, 120 s.
2
, 10% Pd/C, CH
3
4
OH, rt; (b) HCOONH , 10% Pd/C, absolute
4
2
–6 with 10% Pd/C in a hydrogen atmosphere. Among
Pd/C in methanol showed that the olefinic bond was
completely reduced and cleavage of the N-benzyl group
was avoided (Scheme 3). Compounds 14 and 15 were
produced in this way, although long reaction times were
required (12–24 h). It has recently been demonstrated
that CTH can be conducted very rapidly and with good
yields by employing microwave-assisted heating with a
domestic microwave oven. We, therefore, assayed the
CTH procedure for all our compounds using a micro-
wave oven specifically designed for organic synthesis
(Discover LabMate single mode microwave, CEM Cor-
the newly designed compounds, the alkylamines 14
and 15 could not be prepared using the above-described
procedure. In fact, when 12 and 13 were reduced using
conventional catalytic hydrogenation, concurrent cleav-
age of the N-benzyl group was observed.
2
4
The reaction was successful employing catalytic transfer
hydrogenation (CTH) with ammonium formiate as the
hydrogen donor. Preliminary test runs with 2.5 equiva-
lents of ammonium formiate in the presence of 10%

776
S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
1
3.0
12.0
11.0
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
(
ppm)
1
Figure 5. H NMR spectrum of compound 12.
showed the presence of a significant NOE effect corre-
sponding to the interaction of h, l and h, m protons
(
Fig. 7). Moreover, both h and i protons interact with
the aromatic proton nucleus. No interactions were ob-
served between l and h. These results are in agreement
with the crystallographic analysis, confirming the (E)
configuration also for compound 12. On the basis of
these findings, and taking into account the reaction
mechanism, it is reasonable to attribute the E configura-
tion to all compounds.
Figure 6. ORTEP drawing of 2 hydrobromide with the atom labeling
scheme. Ellipsoids are drawn at the 30% probability level.
porate). This procedure was investigated varying the
reaction times and the power. The best results were
achieved with one step of irradiation at 140 ꢁC for
3. Results and discussion
3.1. In vitro binding assays
9
0 s. Thus, all alkenes were rapidly reduced and the N-
benzyl group of compounds 12–13 was not
hydrogenolized.
The binding properties of the novel compounds (Table
2) were investigated for sigma receptor subtypes (r1
and r ) expressed in guinea pig brain membranes.
2
3
[ H]-(+)-pentazocine and [ H]DTG [1,3-di-2-tolylguani-
3
2
.3. Configurational study
dine] were used as radioligands.
In order to determine the stereochemistry of the olefinic
bond, crystallographic analysis of 2 hydrobromide was
performed together with 2D-NOESY NMR of all the
compounds.
All new compounds 2–15 displayed interesting binding
affinities for r subtype receptors (K values ranging
from 1.02 to 47 nM, Table 2), with the exception of
compounds 3 and 4, confirming that the presence of
the alcoholic group on the aliphatic chain of (R/S)-1 is
1
i
The structure of compound 2 was established by means
of X-ray diffraction study. Single crystals suitable for the
crystallographic analysis were obtained from slow crys-
not essential for r affinity, according to the preliminary
1
molecular modeling study (see Section 2.1).
tallization (EtOH/H O) of the pure 2 hydrobromide. A
2
stereoscopic view, with the atomic numbering scheme
of the compound, obtained using the ORTEP 3 pro-
gram, is shown in Figure 6. Crystallographic data
undoubtedly showed that the C@C bond is in the (E)
configuration.
First, we considered the new N,N-dimethylamino deriv-
atives evaluating both the effects of the aromatic moiety
and of the 3-carbon side chain on r affinity and r /r
2
5
1
2
1
selectivity.
The influence of the aromatic moiety on affinity (Table 2,
Fig. 8) was particularly relevant for the compounds of
Series I. Worse results were evidenced following replace-
ment of the naphth-2-yl group (hit compound) with 6-hy-
The NOESY analysis of compound 12 was compared
with that of analogous 2, bearing an N,N-dimethyl or
an N-methyl-N-benzyl group, respectively. Both spectra

S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
777
-
Cl
h
l
+
CH3
R
CH3
H C
N
2
b
e
a
f
C
C
H
c
H
i
d
g
ppm
2
3
4
5
6
7
8
8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 ppm
Compound 2
Figure 7. 2D H H NOESY of compounds 2 and 12.
Compound 12
1
1
Table 2. Binding affinities of the synthesized compounds toward r
1
and r
2
receptors
Compound
Ar
R
K
i
(nM) ± SEM
a
1
b
r
r
2
2 1
r /r
1
2
3
4
5
6
7
8
9
Naphth-2-yl
Naphth-2-yl
Naphth-1-yl
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
25 ± 0.4
171 ± 5
148 ± 4
701 ± 21
6.8
7.5
19.8 ± 0.3
102 ± 5
229 ± 3
6.87
6.1
6-Hydroxy-naphth-2-yl
6-Methoxy-naphth-2-yl
Biphen-4-yl
1394 ± 32
985 ± 16
106 ± 6
21.4 ± 0.8
1.40 ± 0.2
1.95 ± 0.2
47.2 ± 1.0
19.0 ± 0.8
2.30 ± 0.2
1.02 ± 0.2
7.88 ± 0.3
6.32 ± 0.3
5.4 ± 0.3
5.8 ± 0.4
46
75.7
22.5
4.3
Naphth-2-yl
Naphth-1-yl
43.8 ± 3
205 ± 5
6-Hydroxy-naphth-2-yl
6-Methoxy-naphth-2-yl
Biphen-4-yl
1328 ± 62
60.4 ± 7
161 ± 5
70
1
1
1
1
1
1
0
1
2
3
4
5
26.3
157.8
110.8
12
Naphth-2-yl
Biphen-4-yl
C
C
C
C
6
6
6
6
H
5
H
5
H
5
H
5
873.1 ± 12
75.7 ± 4
139 ± 4
Naphth-2-yl
Biphen-4-yl
25.7
102
591 ± 25
Values are means ± SEM of three experiments.
a
3
Displacement of 3 nM [ H]-(+)-pentazocine.
3
Displacement of 3 nM [ H]-DTG [1,3-di-(2-tolyl)-guanidine in the presence of (+)-SKF10,047 (400 nM).
b
droxy-naphth-2-yl and naphth-1-yl. This was mainly evi-
dent for compound 4 (Series I) that showed a r affinity of
Next, we investigated the influence of the aminic substit-
uents on the binding mode by introducing a bulk substi-
tuent at the amino group, as suggested by the
literature. We, therefore, prepared the N-methyl-N-
benzyl analogues of the most interesting compounds of
both series, bearing the 2-naphthalenic or the 4-bipheny-
lic moiety. In contrast to our hypothesis, the presence of
1
2
receptor affinities (K = 1.40 and 1.02 nM) and better
29 nM and only a 6-fold selectivity over r . Higher r
2 1
1
8
i
selectivities (K r /K r = 75.7 and 157.8, respectively)
i
2
i
1
were evidenced for compounds 6 (Series I) and 11 (Series
II), leading to the identification of the 4-biphenyl group as
a promising aromatic moiety for both series.
the N-benzyl group increased both r affinity and selec-
1
tivity in the naphth-2-yl derivatives 12 and 14 only. In
fact, the introduction of an N-benzyl group in the 4-bi-
As regards the effect of the 3-carbon dimethylamino side
chain, binding data showed that Series II compounds
phenyl derivatives lowered both r affinity and selectiv-
1
had a better r affinity with respect to Series I (Table
ity. Moreover, all N-benzylamino derivatives showed
remarkable affinity for r (K nM ranging from 5.4 to
1
2
, Fig. 8).
1
i

778
S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
2
2
1
1
50
00
50
00
1
1
1
1
1
80
60
40
20
00
80
Serie1
Serie2
Serie1
Serie2
6
0
50
40
2
0
0
0
a
b
c
d
e
a
b
c
d
e
HO
3
H CO
a
c
b
d
e
1 1 2
Figure 8. Effects of the aromatic substituents on r affinity (A) and on r /r selectivity (B) of N,N-dimethylamino derivatives.
7
.9 nM), but only compound 12 also showed a good
selectivity (110.8).
scribed in detail in a previous paper where it has also
been successfully used to find novel screening candi-
2
6,27
dates.
and one positive ionizable features. Our own model
This model consisted of four hydrophobic
2
7
These results indicate that the best influence on r affin-
1
ity and selectivity is related to the presence of the 4-bi-
phenyl moiety. In contrast, the introduction of a
benzyl group at the amine site is not always a convenient
modification. On the basis of these findings, it can be
postulated that when the sigma ligands possess two aro-
matic moieties, the balance between their steric hin-
drances is particularly relevant for the binding.
was in good agreement with the model proposed by
Glennon, which was not a computational model and
therefore not fit for in silico comparison with compound
models and for database screening. Our model did not
possess the additional electronegative site reported by
Gund either. This additional feature is present in many
known r -active compounds, but does not seem to be
1
important for affinity toward this target. Since most of
our compounds also lacked this feature, our own model
seemed to be better suited for predicting their binding
modes. A short overview of the three models is given
in Table 3.
3
.2. Molecular modeling
The compound design allowed us to obtain novel inter-
esting r ligands, even though neither of the pharmaco-
1
phore models considered was able to explain some of the
structure–activity relationships.
This model was able to fit all our molecules, although
their affinities were always overestimated by factors
ranging from 2 to 4900. Such an overestimation may
be attributed to both interlaboratory differences—the
training set for the Catalyst model and the compounds
presented in this paper were tested in different laborato-
ries—and also to the very high r affinity (K = 0.01 nM)
To gain more insights into binding modes and to per-
form a more rational SAR study, we mapped conform-
ers of the synthesized compounds onto a 5-point
pharmacophore model developed by us using the Hypo-
Gen module in Catalyst 4.9 and which we have de-
1
i
1
Table 3. Comparison of previously reported r models including our own model
18
19
27
Reference
Glennon et al.
Gund et al.
Laggner et al.
Data used for model generation
SAR of 49 compounds
Guided overlay of low-energy
conformations of four
compounds
Automatic overlay of
conformational models from 23
compounds, inclusion of binding-
affinity data (QSAR model)
Catalyst 4.9
Software package used
Distance between basic site and
Is not an in silico model
6–10
Sybyl 6.5
7.14
˚
Two features at 6.3 and 9.8 A,
˚
primary hydrophobic site (A)
Distance between basic site and
˚
secondary hydrophobic site (A)
Additional interaction sites
respectively
4.1
2.5–3.9
None
2.80
Hydrogen-bond accepting
feature in direction of primary
˚
Fourth hydrophobic region in
direction of primary hydrophobic
˚
hydrophobic group, 4.17 A from group, 3.6 A from basic site
basic site
Additional restrictions
Basic site (‘proton donor site’)
tollerates small groups only,
secondary binding site
tolerates bulk
Aromatic rings at the primary
hydrophobic feature lie inside a
For database searching (not used
in this paper): predicted K value
i
plane—two additional points for <100 lM, molecular weight
orientation
<600, and inclusion of molecular
shape of fenpropimorph

S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
779
1
Figure 9. Mapping of compounds 2 and 12 onto our r pharmacophore model: positive ionizable feature, red sphere; hydrophobic features, cyan
spheres.
of fenpropimorph (Fig. 1) in the training set of selected
compounds, which tuned the model toward high affinity
values. It should be pointed out that this compound is
very similar to our own compounds, both consisting of
an aromatic ring connected to a tertiary amine through
a chain of three carbon atoms. Methyl groups of the side
chain at the a and b positions of fenpropimorph and of
our compounds are able to map the same pharmaco-
phore feature. Additionally, the second phenyl group
in our biphenyl derivatives may play a similar role to
that of the tert-butyl group of fenpropimorph.
Fitting studies for both isomers of all synthesized alk-
enes (2–6 and 12, 13) indicated that (E)-alkenes, which
were the sole products retrieved under our reaction con-
ditions, should be much more active than their (Z)-
counterparts. (E)-13 showed a significantly better fit
than its (Z)-isomer (best fit values: 10.17 vs 8.35),
whereas very small differences were found between the
two stereoisomers of 5 and 6 (difference between the best
fit values: 0.54, 0.05).
The role of the stereochemistry in binding requirements
was further investigated extending fitting studies to the
pure enantiomeric (R) and (S) forms of alkanes 7–11
and 14, 15. With the exception of 8 and 10, (S) isomers
were predicted to be slightly more active than the (R)
isomers. Only slight differences were found for each pair
of enantiomers: the highest difference in fit values was
0.7 for the enantiomers of 7.
Because of the observed overestimation, instead of con-
sidering the absolute K values we compared best fit val-
i
ues ranging from 0 to 10.7, the latter value describing
the perfect fit of 14.
According to the SAR study, removal of the alcoholic
group from 1 resulted in a better fitting of 2 and 7
(
best fit values: 8.19 and 8.46 vs 6.44). The replacement
In the future, pure enantiomers of the most interesting
saturated ligands will be prepared in order to assess
our model’s prediction about the relationship between
stereochemistry at the benzylic carbon atom and the
receptor binding affinity.
of the N,N-dimethyl group by an N-methyl-N-benzyl
group was expected to be advantageous since the N-
benzyl derivatives were associated with the highest best
fit values (e.g., 12 showed the highest value of 10.7,
whereas its N,N-dimethyl analogue 2 had a value of
8
.19) (Fig. 9).
4. Conclusion
Interestingly, 1-naphthyl substitution was also predicted
to have negative effects since 3 and 8, not being able to
map the hydrophobic site furthest away from the posi-
tive ionizable center, showed values of 6.51 and 6.54,
respectively, whereas the 2-naphthyl derivatives 2 and
We present the design, synthesis, and biological evalua-
tion of a novel class of arylalkyl- and alkenylamines,
new molecules with novel structures, strong affinities,
and high selectivity.
7 that failed only to map the hydrophobic site closest
to the amino group showed fit values of 8.19 and 8.46.
The application of indirect modeling approaches involv-
ing the analysis of known pharmacophore models
However, the influence of the 4-biphenyl substitution was
not correctly predicted by our Catalyst pharmacophore
model, according to which the effects of the benzyl group
at the amine site were always the most important. For
example, our model was not able to explain the signifi-
cantly enhanced activity for N,N-dimethyl substituted
allowed us to design novel r ligands efficiently. The start-
1
ing point of our study was (R/S)-4-(dimethylamino)-2-
(naphthalene-2-yl)butan-2-ol 1, previously synthesized
in our laboratory and characterized by promising affinity
for r receptors. Racemic N,N-dimethyl-3-(6-hydroxy-
1
naphthalen-2-yl)butan-1-amine 11 showed not only po-
tent affinity for r receptor, but good selectivity over the
4
-biphenyl derivatives 6 and 11, for which it predicted
slightly worse affinities than for their naphthyl analogues
and 7. In good accordance with the binding data, com-
parable best fit values were found for 6 and 11 (8.05 vs
.35), so far identified as the most active compounds of
our series.
1
r2 subtype.
2
Moreover, useful information concerning the structure
and properties of the ligand-binding site in the r receptor
was obtained. Evaluation of the degree of fit between all
8
1

780
S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
the new ligands and a five-point pharmacophore model,
developed using the HypoGen module within the soft-
atmosphere and then HCl 1 N was added until pH 2
was reached. The aqueous phase was washed with dieth-
yl ether, made alkaline with a saturated solution of
NaHCO to pH 8, and extracted with CHCl . The com-
2
7
ware package Catalyst, allowed us to acquire more in-
sights to improve this model and to establish more
coherent structure–activity relationships. Despite some
weaknesses in quantitative predictions, this pharmaco-
phore model never produced false-negative results and
consequently encourages us to use it for the design and
3
3
bined organic layers were dried over anhydrous Na SO
2
4
and concentrated under reduced pressure yielding the
desired product as a yellow oil. The yield was 83%. IR
(cm ): 3025, 2946, 1710, 1600, 1494, 1455, 1355,
À1
1
synthesis of more potent and selective r ligands. With
1163, 1025, 916, 749. H NMR (CDCl ) d 2.16 (s, 3H,
1
3
this aim, the model will be used for virtual screening of li-
braries of potential novel ligands, based on the structures
of the most active and selective compounds so far
identified.
CH –CO), 2.21 (s, 3H, CH –N), 2.70 (dt, 4H, CH –
CH –N, J = 6.4 Hz), 3.52 (s, 2H, CH –C H ), 7.31 (m,
3
3
2
2
2
6
5
5H, aromatic). Elemental analysis calculated for
C H NO: C, 75.35; H, 8.96; N, 7.32. Found: C,
1
2
17
7
5.16; H, 8.88; N, 7.54.
Our aim for the future is to prepare and test the pure
enantiomers of the most active and selective compounds
but also perform functional assays studying the interac-
tion with some neuroactive steroids. The results of such
investigations will allow us to evaluate both their ago-
nistic/antagonistic activity.
5.3. General procedure for the preparation of compounds
(E)-2–8 hydrochlorides
t-BuLi (17.16 mmol, 1.7 M in pentane) was added drop-
wise to a solution of the corresponding aromatic precur-
sor (8.58 mmol) in anhydrous diethyl ether (0.2 M)
cooled to À78 ꢁC, under nitrogen atmosphere, keeping
the temperature for 15 min. The reaction mixture was
then allowed to warm slowly to room temperature. Stir-
ring was continued for 1 h and a solution of the ketone
5. Experimental
5
.1. General methods
(
6.86 mmol) in dry diethyl ether (10 mL) at À78 ꢁC was
Unless otherwise specified, commercially available re-
agents and solvents were used as received from the sup-
plier. Anhydrous solvents were obtained according to
standard procedures.
then added drop-wise. After 2 h, the reaction mixture
was quenched with HCl 10 N until pH 2. The precipitate
obtained was filtered off and, after an acid–base work-
up, was crystallized from acetone providing the expected
product as a white solid.
Melting points were measured on SMP3 Stuart Scientific
apparatus and are uncorrected. Elemental analyses were
performed on a Carlo Erba 1106 C, H, N analyzer.
5.3.1. (E)-N,N-Dimethyl-3-(naphthalen-2-yl)but-2-en-1-
yl-amine hydrochloride [(E)-2ÆHCl]. Yield 76%; white sol-
À1
id, mp 201–203 ꢁC. IR (cm ): 3051, 2892, 2563, 2477,
2366, 2310, 1632, 1475, 1384, 1130, 950, 820, 741. H
1
Analytical TLC were carried out on silica gel pre-coated
glass-backed plates (Fluka Kieselgel 60 F254, Merck)
and visualized by ultra-violet radiation, acidic ammoni-
um molybdate (IV) or potassium permanganate.
NMR (CD OD) d 2.30 (s, 3H, CH C@CH), 2.94 (s,
3
3
6H, N(CH ) ), 4.03 (d, 2H, CH –N, J = 7.8 Hz), 6.02
(t, 1H, CH C@CH, J = 7.8 Hz), 7.48 (m, 2H, aromatic),
3
2
2
3
7
.64 (d, 1H, aromatic, J = 7.8 Hz), 7.80 (m, 3H, aromat-
X-ray crystallographic analysis was performed on Philips
Pw 1100 computer-controlled four circle diffractometer,
graphite-monochromated MoKa radiation, x scan tech-
nique (±h, ±k, and ±l), scan width 1.5ꢁ, scan speed 0.05ꢁ/
min, range 2–20ꢁ theta. IR spectra were recorded on a
Jasco FT/IR-4100 spectrophotometer; only noteworthy
ic), 7.95 (s, 1H, aromatic). Elemental analysis calculated
for C H NCl: C, 73.41; H, 7.70; N, 5.35. Found: C,
73.14; H, 7.57; N, 5.16.
1
6
20
5.3.2. (E)-N,N-Dimethyl-3-(naphthalen-1-yl)but-2-en-1-
yl-amine hydrochloride [(E)-3ÆHCl]. Yield 59%; white sol-
1
À1
absorptions are given. H NMR spectra were measured
id, mp 192–194 ꢁC. IR (cm ): 3011, 2899, 2617, 2361,
1
1656, 1460, 1376, 1161, 1010, 777. H NMR (CD OD)
with an ADVANCE 400 spectrometer Bruker, Germany;
TMS was used as internal standard d = 0 and chemical
shifts are given in ppm.
3
d 2.27 (s, 3H, CH C@CH), 3.32 (s, 6H, N(CH ) ), 4.05
3
3 2
(d, 2H, CH –N, J = 8.3 Hz), 5.65 (dt, 1H, CH C@CH,
2
3
J = 7.8 Hz), 7.34 (dd, 1H, aromatic, J = 6.9 Hz), 7.47
(d, 1H, aromatic, J = 8.3 Hz), 7.52 (m, 2H, aromatic),
7.90 (m, 3H, aromatic). Elemental analysis calculated
for C H NCl: C, 73.41; H, 7.70; N, 5.35. Found: C,
Microwave dielectric heating was performed in a Dis-
cover LabMate instrument, CEM Corporate specifical-
ly designed for organic synthesis and following an
appropriate microwave program. The reaction tempera-
tures were monitored by an IR probe.
ꢂ
1
6
20
73.30; H, 7.68; N, 5.44.
5
.3.3. (E)-N,N-Dimethyl-3-(6-hydroxy-naphthalen-2-
5
.2. 4-(Benzyl-methyl-amino)butan-2-one (17)
yl)but-2-en-1-yl-amine hydrochloride [(E)-4ÆHCl]. Yield
À1
3
2700, 2640, 2510, 1630, 1425, 1260, 1150, 950, 860. H
8%; white solid, mp 219–220 ꢁC. IR (cm ): 3200,
1
Methylvinylketone (12 mL, 146 mmol) was added drop-
wise to a stirred solution of N-benzyl-N-methyl-amine
NMR (CD OD) d 2.38 (s, 3H, CH C@CH), 3.0 (s, 6H,
3
3
(13 mL, 102.2 mmol) in anhydrous toluene (1 M). The
reaction mixture was refluxed for 4 h under a nitrogen
N(CH ) ), 4.11 (d, 2H, CH –N, J = 7.8 Hz), 6.06 (t, 1H,
3 2 2
CH C@CH, J = 7.7 Hz), 7.16 (d, 1H, aromatic,
3

S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
781
J = 2.4 Hz), 7.19 (s, 1H, aromatic), 7.65 (d, 1H, aromatic,
J = 8.7 Hz), 7.73 (d, 1H, aromatic, J = 8.7 Hz), 7.84 (d,
6ÆHCl (4.44 mmol) in methanol (0.1 M, 44.4 mL) in a
hydrogen atmosphere. After 24 h, the reaction mixture
was filtered on a pad of Celite and washed thoroughly
with methanol. After evaporation of the solvent under
vacuum, the crude product was crystallized from ace-
tone giving the desired compound as a white solid.
1H, aromatic, J = 8.5 Hz), 7.95 (s, 1H, aromatic). Ele-
mental analysis calculated for C H NOCl: C, 69.18;
H, 7.26; N, 5.04. Found: C, 68.79; H, 7.54; N, 4.92.
1
6
20
5
.3.4. (E)-N,N-Dimethyl-3-(6-methoxy-naphthalen-2-
yl)but-2-en-1-yl-amine hydrochloride [(E)-5ÆHCl]. Yield
5.4.2. Method B. 10% Pd/C (0.33 g) and ammonium
formiate (8.31 mmol) were added to a stirred solution
of the corresponding alkenes (E)-12–13 (3.32 mmol) in
absolute ethanol (28 mL). After 6 h, the reaction mix-
ture was filtered on a pad of Celite and washed thor-
oughly with absolute ethanol. After an acid–base
work-up, the reaction solvent was evaporated under
vacuum and the crude product was transformed into
the corresponding hydrochloride and crystallized from
acetone providing the product as a white solid.
À1
7
2
0%; white solid, mp 240–241 ꢁC. IR (cm ): 3008,
938, 2589, 2485, 2350, 1598, 1482, 1246, 1028, 851.
1
H NMR (CD OD) d 2.17 (s, 3H, CH C@CH,), 2.76
3
3
(
s, 6H, N(CH ) ), 3.80 (m, 5H, CH –N + OCH ), 5.87
3 2 2 3
(
t, 1H, CH C@CH, J = 7.8 Hz), 7.03 (dd, 1H, aromatic,
3
J = 8.8 Hz), 7.12 (s, 1H, aromatic), 7.50 (dd, 1H, aro-
matic, J = 8.8 Hz), 7.66 (m, 2H, aromatic), 7.76 (s, 1H,
aromatic).
Elemental
analysis
calculated
for
C H NOCl: C, 69.97; H, 7.70; N, 4.80. Found: C,
1
7
22
6
9.87; H, 7.64; N, 4.45.
5.4.3. Method C. Ammonium formiate (0.75 mmol) and
5
.3.5. (E)-N,N-Dimethyl-3-(biphen-4-yl)but-2-en-1-yl-
10% Pd/C (30 mg) were added to alkenes (E)-2–6, 12–13
(0.3 mmol) in absolute ethanol (3 mL) placed in a MW
vial. The mixture was heated for 90 s at 100 W. After
the usual work-up, the desired products were isolated.
amine hydrochloride [(E)-6ÆHCl]. Yield 63%; white sol-
id, mp 251–252 ꢁC. IR (cm ): 3007, 2950, 2569, 2513,
2
À1
1
478, 2376, 1642, 1485, 1300, 952, 841, 758. H NMR
(
N(CH ) ), 3.89 (d, 2H, CH –N, J = 7.8 Hz), 5.83 (dt,
CD OD) d 2.13 (s, 3H, CH C@CH), 2.87 (s, 6H,
3
3
5.4.4. (R/S)-N,N-Dimethyl-3-(naphthalen-2-yl)butan-1-
amine hydrochloride [(R/S)-9ÆHCl]. Yield 59%; white
3
2
2
1
H, CH C@CH, J = 7.3 Hz), 7.22 (dt, 1H, aromatic,
3
À1
J = 7.3 Hz), 7.33 (dt, 2H, aromatic, J = 7.3 Hz), 7.48
dd, 2H, aromatic, J = 8.8 Hz), 7.52 (m, 4H, aromat-
ic). Elemental analysis calculated for C H NCl: C,
solid, mp 177–179 ꢁC. IR (cm ): 3362, 3010, 2776,
1
(
2577, 2467, 1599, 1474, 1190, 1014, 963, 751.
NMR (CD OD)
H
d 1.35 (d, 3H, CH –CH,
3
J = 7.1 Hz), 1.89 (m, 2H, CH–CH ), 2.18 (s, 6H,
1
8
22
3
75.11; H, 7.70; N, 4.87. Found: C, 75.22; H, 7.94;
N, 5.03.
2
N(CH ) ), 2.30 (m, 2H, CH –N), 2.87 (m, 1H, CH –
3
2
2
3
CH), 7.4 (m, 3H, aromatic), 7.63 (s, 1H, aromatic),
7.80 (m, 3H, aromatic). Elemental analysis calculated
for C H NCl: C, 72.85; H, 8.41; N, 5.31. Found:
5
.3.6. (E)-N-Benzyl-N-methyl-3-(naphthalen-2-yl)but-2-
en-1-yl-amine hydrochloride [(E)-7ÆHCl]. Yield 43%;
white solid, mp 179–180 ꢁC. IR (cm ): 3049, 2969,
2
1
6
22
À1
C, 73.01; H, 8.33; N, 5.32.
1
585, 2514, 1647, 1472, 1444, 1389, 906, 818, 745. H
NMR (CD OD) d 2.29 (s, 3H, CH C@CH), 2.88 (s,
5.4.5. (R/S)-N,N-Dimethyl-3-(naphthalen-1-yl)butan-1-
amine hydrochloride [(R/S)-10ÆHCl]. Yield 32%; white
3
3
3
H, N–CH ), 4.09 (d, 2H, CH –N, J = 7.8 Hz), 4.45 (s,
3 2
À1
2
H, CH –C H ), 6.08 (dt, 1H, CH C@CH,
solid, mp 195–199 ꢁC. IR (cm ): 3371, 3015, 2873,
2577, 2477, 1593, 1479, 1396, 964, 803, 780. H NMR
2
6
5
3
1
J = 7.8 Hz), 7.55 (m, 7H, aromatic), 7.67 (dd, 1H, aro-
matic, J = 8.8 Hz), 7.88 (m, 3H, aromatic), 7.97 (s, 1H,
aromatic). Elemental analysis calculated for C H NCl:
(CD OD) d 1.38 (d, 3H, CH –CH, J = 7.3 Hz), 2.16
3
3
(m, 2H, CH–CH ), 2.75 (s, 6H, N(CH ) ), 2.92 (dt,
1H, HCH–N, J = 5.4, 12.2 Hz), 3.09 (dt, 1H, HCH–N,
J = 5.4, 12.2 Hz), 3.72 (m, 1H, CH –CH), 7.43 (m, 3H,
2
2
24
2
3 2
C, 78.20; H, 7.16; N, 4.15. Found: C, 77.81; H, 7.40; N,
4
.40.
3
aromatic), 7.48 (dt, 1H, aromatic, J = 6.9 Hz), 7.69
(dd, 1H, aromatic, J = 6.4 Hz), 7.81 (d, 1H, aromatic,
J = 7.8 Hz), 8.1 (d, 1H, aromatic, J = 8.3 Hz). Elemental
analysis calculated for C H NCl: C, 72.85; H, 8.41; N,
5
.3.7. (E)-N-Benzyl-N-methyl-3-(biphen-4-yl)but-2-en-1-
yl-amine hydrochloride [(E)-8ÆHCl]. Yield 27%; white sol-
À1
id, mp 223–224 ꢁC. IR (cm ): 3030, 2979, 2487, 2381,
1
6
22
1
644, 1468, 1420, 1375, 1100, 1054, 918, 765. H NMR
1
5.31. Found: C, 72.60; H, 8.65; N, 5.34.
(
CH ), 3.93 (d, 2H, CH –N, J = 7.6 Hz), 4.30 (s, 2H,
CD OD) d 2.08 (s, 3H, CH C@CH), 2.73 (s, 3H, N–
3
3
5.4.6. (R/S)-N,N-Dimethyl-3-(6-hydroxy-naphthalen-2-
yl)butan-1-amine hydrochloride [(R/S)-11ÆHCl]. Yield
3
2
CH –C H ), 5.87 (dt, 1H, CH C@CH, J = 7.8 Hz),
2
6
5
3
À1
7
.23 (t, 1H, aromatic, J = 7.3 Hz), 7.33 (dt, 2H, aromat-
69%; white solid, mp 165–167 ꢁC. IR (cm ): 3211,
2706, 2575, 2475, 1602, 1476, 1387, 1363, 1185, 853.
ic, J = 7.3 Hz), 7.42 (m, 6H, aromatic), 7.48 (s, 1H, aro-
matic), 7.52 (m, 4H, aromatic). Elemental analysis
calculated for C H NCl: C, 79.21; H, 7.20; N, 3.85.
1
H
J = 6.6 Hz), 2.03 (m, 2H, CH–CH ), 2.72 (s, 6H,
NMR (CD OD) d 1.30 (d, 3H, CH –CH,
3
3
2
4
26
2
Found: C, 79.36; H, 7.38; N, 3.74.
N(CH ) ), 2.83 (m, 2H, CH –N), 3.01 (m, 1H, CH –
3
3
2
2
CH), 7.0 (sb+d, 2H, aromatic, J = 9.8 Hz), 7.24 (d,
1H, aromatic, J = 8.4 Hz), 7.5 (s, 1H, aromatic), 7.55
(d, 1H, aromatic, J = 8.5 Hz), 7.59 (d, 1H, aromatic,
J = 8.7 Hz). Elemental analysis calculated for
C H NOCl: C, 68.68; H, 7.93; N, 5.01. Found: C,
5.4. General procedure for the preparation of (R/S)-7-11
and (R/S)-14-15 hydrochlorides
5.4.1. Method A. 10% Pd/C (0.044 mmol) was added to a
stirred solution of the corresponding alkenes (E)-2–
1
6
22
69.07; H, 8.09; N, 4.88.

782
S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
5
.4.7. (R/S)-N,N-Dimethyl-3-(6-methoxy-naphthalen-2-
refined parameters are the following: R = 0.0589,
1
yl) butan-1-amine hydrochloride [(R/S)-12ÆHCl]. Yield
wR = 0.1007 (R = 0.1292, wR = 0.1219 for all 1382
2
data). Atomic coordinates and anisotropic displacement
parameters are for deposit.
2
1
À1
5
2
3%; white solid, mp 217–224 ꢁC. IR (cm ): 3398,
957, 2597, 2518, 2485, 1604, 1482, 1217, 1024, 849.
1
H
J = 6.7 Hz), 2.09 (m, 2H, CH–CH ), 2.80 (s, 6H,
NMR (CD OD) d 1.39 (d, 3H, CH –CH,
3
3
1
5.6. H NMR spectroscopy
2
N(CH ) ), 2.92 (m, 2H, CH –N), 3.12 (m, 1H, CH –
3
CH), 3.88 (s, 3H, OCH ), 7.10 (dd, 1H, aromatic,
3
2
2
3
1
1
H, H COSY, and 2D-NOESY NMR spectra of E-2
J = 8.8 Hz), 7.19 (s, 1H, aromatic), 7.36 (d, 1H, aromat-
ic, J = 8.8 Hz), 7.60 (s, 1H, aromatic), 7.69 (d, 1H, aro-
matic, J = 8.4 Hz), 7.75 (d, 1H, aromatic, J = 8.4 Hz).
Elemental analysis calculated for C H NOCl: C,
and E-12 were performed at 9.4 T, in CD OD at room
3
temperature (TMS as internal standard d = 0).
5.7. Molecular modeling
1
7
24
6
4
9.49; H, 8.23; N, 4.77. Found: C, 69.87; H, 8.56; N,
.94.
5.7.1. Conformational search of compound 1. Compound
1
MODEL ver 7.2. Energy was minimized with the
was built by the Maestro user interface of MACRO-
2
8
5
.4.8. (R/S)-N,N-Dimethyl-3-(biphen-4-yl)butan-1-amine
2
8
hydrochloride [(R/S)-13ÆHCl]. Yield 47%, white solid, mp
MMFFs force field and GB/SA implicit model of solva-
29
À1
2
1
03–204 ꢁC. IR (cm ): 3051, 2881, 2550, 2469, 2408,
tion. As previously reported in a similar compound
analysis, the nitrogen atom was considered to be in pro-
1
600, 1485, 1406, 1006, 966, 835. H NMR (CD OD)
3
3
0
d 1.38 (d, 3H, CH –CH, J = 6.9 Hz), 2.07 (m, 2H,
CH–CH ), 2.85 (s, 6H, N(CH ) ), 2.88 (m, 2H, CH –
N), 3.15 (m, 1H, CH –CH), 7.35 (m, 3H, aromatic),
7.44 (t, 2H, aromatic, J = 8.3 Hz), 7.60 (m, 4H, aromat-
ic). Elemental analysis calculated for C H NCl: C,
tonated form. The automatic setup included five tor-
3
3
1
sional bonds. A Monte Carlo (MC) searCH was used
to generate 2000 conformations and energy was mini-
mized with the same force field and solvation conditions.
The deduplication of identical conformers after superim-
position was performed according to the standard MAC-
2
3 2
2
3
1
8
24
7
4
4.59; H, 8.35; N, 4.83. Found: C, 74.41; H, 8.35; N,
.45.
˚
ROMODEL criteria (max rms deviation equal to 0.25 A,
max energy difference equal to 1 kcal/mol).
5
.4.9. (R/S)-N-Benzyl-N-methyl-3-(naphthalen-2-yl)bu-
tan-1-amine hydrochloride [(R/S)-14ÆHCl]. Yield 25%,
The pharmacophore feature measurements were carried
out with the Maestro graphical interface considering for
the aromatic moiety the centroid averaging the position
of the naphthalene carbon atoms, for the protonated
amine the nitrogen atom, and for the hydroxyl moiety
the oxygen atom. All calculations were performed on
an Intel Pentium IV workstation equipped with Linux
Redhatver. 7.3.
À1
white solid, mp 143–144 ꢁC. IR (cm ): 3033, 2526,
1
348, 1867, 1747, 1650, 1396, 1147, 1027, 818, 748. H
2
NMR (CDCl ) d 1.44 (d, 3H, CH –CH, J = 6.8 Hz),
3
3
2
.20 (m, 2H, CH–CH ), 2.78 (s, 3H, N–CH ), 2.98 (m,
2 3
1
CH –CH), 4.26 (s, 2H, CH –C H ), 7.36 (m, 6H, aro-
H, HCH–N), 3.16 (m, 1H, HCH–N), 4.25 (m, 1H,
3
2
6
5
matic), 7.48 (m, 2H, aromatic), 7.67 (s, 1H, aromatic),
.83 (m, 3H, aromatic). Elemental analysis calculated
for C H NCl: C, 77.74; H, 7.71; N, 4.12. Found: C,
7
5.7.2. ClogP for compounds 1–15. The logP values of
the examined compounds were calculated using the
Daylight computational method (version 4.01) that
combines fragment lipophilicity contributions carefully
parameterized with experimental data and calculated
2
2
26
7
7.37; H, 8.04; N, 4.01.
5
.4.10. (R/S)-N-Benzyl-N-methyl-3-(biphen-4-yl)butan-1-
amine hydrochloride [(R/S)-15ÆHCl]. Yield 15%, white
À1
22
solid, mp 144–146 ꢁC. IR (cm ): 3054, 2849, 2566,
‘from scratch’ values.
1
502, 1670, 1484, 1454, 1074, 835, 763. H NMR
2
(
CDCl ) d 1.28 (d, 3H, CH –CH, J = 7.8 Hz), 1.85 (m,
5.7.3. Pharmacophore modeling. All computations were
performed with Catalyst 4.9 on an Indigo O2 Worksta-
tion (255 MHz, MIPS R10000, 320 MB RAM) running
Irix 6.5. Conformational models were generated with the
BEST option, a maximum number of 250 conformers
per compound, and the default energy cutoff value of
20 kcal/mol. When compounds were tested as a mixture
of optical stereoisomers (Series II), the relevant stereoa-
toms were drawn as sterically undefined. In this case,
Catalyst automatically chooses the best fitting stereoiso-
mer during model generation and fitting. The construc-
3
3
2
CH –N), 2.85 (m, 1H, CH –CH), 3.46 (s, 2H, CH –
C H ), 7.29 (m, 8H, aromatic), 7.44 (t, 2H, aromatic,
5
H, CH–CH ), 2.17 (s, 3H, N–CH ), 2.35 (m, 2H,
2 3
2
3
2
6
J = 7.6 Hz), 7.52 (d, 2H, aromatic, J = 8.1 Hz), 7.60 (d,
H, aromatic, J = 7.3 Hz). Elemental analysis calculated
for C H NCl: C, 78.77; H, 7.71; N, 3.83. Found: C,
2
2
4
28
7
8.74; H, 7.79; N, 3.81.
5
.5. Crystal data for C H NBr 2
20
1
6
˚
Monocyclic, space group P2 /c, a = 20.922(5) A,
tion of the pharmacophore hypothesis for r has been
1
1
˚
˚
˚
27
b = 5.5923(11) A, c = 12.634(2) A, b = 93.141 (18)ꢁ,
reported previously. The graphics in Figure 9 were cre-
32
ated with WebLab Viewer Lite 3.5.
3
V = 1476.0(5) A , Z = 4, d = 1.378 g cm , T = 293 K.
À3
x
Data were collected for a colorless needle-shaped crystal
with dimensions of 0.12 mm · 0.15 mm · 0.30 mm on a
Philips Pw 1100 computer-controlled four circle diffrac-
tometer, graphite-monochromated MoKa radiation. Fi-
nal R indices for 770 reflections with I > 2r (I) and 169
5.8. Radioligand binding study
The guinea pig brain membranes were prepared using
the procedure described by Matsumoto et al. Briefly,
33

S. Collina et al. / Bioorg. Med. Chem. 15 (2007) 771–783
783
guinea pig brain minus cerebella was homogenized in 10
volumes of ice-cold 0.32 M sucrose, pH 7.4, with a
glass–Teflon homogenizer and then centrifuged at
7. Vilner, B. J.; John, C. S.; Bowen, W. D. Cancer Res. 1995,
5, 408–413.
5
8
. Itzhak, Y. Sigma Receptors; Academic Press: San Diego,
CA, 1994.
. Hayashi, T.; Su, T. P. CNS Drugs 2004, 18, 269–284.
3
000 rpm for 10 min at 4 ꢁC. The pellet was resuspended
in 2 vol and centrifuged again at 3000 rpm for 10 min at
ꢁC. The supernatants were collected and recentrifuged
9
1
1
0. Hanner, M.; Moebius, F. F.; Flandorfer, A.; Knaus, H.
G.; Striessnig, J.; Kempner, E.; Glossmann, H. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 8072–8077.
1. Teruo, Hayashi; Tsung-Ping, Su CNS Drugs 2004, 18,
269–284.
4
at 19,000 rpm for 20 min at 4 ꢁC. The pellet was resus-
pended in 3 vol of 10 mM Tris–HCl, pH 7.4, incubated
for 15 min at room temperature, and centrifuged again
at 19,000 rpm for 15 min at 4 ꢁC. The final pellet was
resuspended in 1.53 vol and frozen at À80 ꢁC until use.
12. Moebius, F. F.; Striessnig, J.; Glossmann, H. Trends
Pharmacol. Sci. 1997, 18, 67–70.
1
1
3. Bowen, W. D. Pharm. Acta. Helv. 2000, 74, 211–218.
4. Maurice, T. Drug News Perspect. 2002, 15, 617–625.
r1 binding assays were performed as described by
DeHaven and co-workers. Briefly, 1 mL of membrane
3
4
15. Tam, S. W.; Cook, L. Neurobiology 1984, 81, 5618–5621.
16. Chien, C. C.; Pasternak, G. W. J. Pharmacol. Exp. Ther.
3
suspension (500 lg of protein) and 3 nM [ H]-(+)-pen-
1
994, 271, 1583–1590.
7. Mei, J.; Pasternak, G. W. J. Pharmacol. Exp. Ther. 2002,
00, 1070–1074.
8. Glennon, R. A.; Ablordeppey, S. Y.; Ismaiel, A. M.; El-
Ashmawy, M. B.; Fischer, J. B.; Howie, K. B. J. Med.
Chem. 1994, 37, 1214–1219.
tazocine (28 Ci/mmol; K = 1.6 ± 0.3 nM, n = 3) in
d
1
1
5
pounds (from 10 to 10
0 mM Tris–HCl (pH 7.4) were incubated with test com-
M). Nonspecific binding
3
À5
À11
was assessed in the presence of 10 lM haloperidol.
The reaction was performed for 150 min at 37 ꢁC and
terminated by filtering the solution through a Whatman
GF/B glass fiber filter which had been presoaked for 1 h
in a 0.5% poly(ethylenimine) solution. Filters were
washed twice with 4 mL of ice-cold buffer.
19. Gund, T. M.; Floyd, J.; Jung, D. J. Mol. Graphics Modell.
004, 22, 221–230.
2
2
0. Azzolina, O.; Collina, S.; Brusotti, G.; Rossi, D.; Calle-
gari, A.; Linati, L.; Barbieri, A.; Ghislandi, V. Tetrahe-
dron: Asymmetry 2002, 13, 1073–1081.
2
2
1. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Freeney,
P. J. Adv. Drug Delivery Rev. 1997, 23, 3–25.
2. Leo, A. J.; Hoekman, D. Perspect. Drug Discovery Des.
r2 binding assays were carried out as reported by
Mach et al. Briefly, 0.5 mL of membrane suspension
3
5
3
(
guanidine] (31 Ci/mM; K = 15.3 ± 0.6 nM, n = 3 nM)
360 lg protein) with 3 nM [ H]DTG [1,3-di-2-tolyl-
2
000, 18, 19–38.
d
23. Azzolina, O.; Collina, S.; Rossi, D.; Alcaro, S.; Ghislandi,
V. Experimental and computational approaches to esti-
mate the hydrophobicity of antinociceptive agents, First
Magna Grecia Medicinal Chemistry Workshop on New
Perspectives in Drug Research, Copanello, Catanzaro,
Italy, 10–13 June 2001.
in the presence of 400 nM (+)-SKF10,047 to block
r1 sites was incubated with test compounds. Incuba-
tion was carried out in 50 mM Tris–HCl (pH 8.0)
for 120 min at room temperature. Each assay was
terminated by the addition of ice-cold 10 mM Tris–
HCl, pH 8.0, followed by filtration through a What-
man GF/B glass fiber filter which had been presoaked
for 1 h in a 0.5% poly-(ethylenimine) solution. Filters
were washed twice with 4 mL of ice-cold buffer. Non-
specific binding was defined using 5 lM DTG. Inhibi-
2
4. Banik, B. K.; Barakat, K. J.; Wagle, D. R.; Manhas, M.
S.; Bose, A. K. J. Org. Chem. 1999, 64, 5746–5753.
5. Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
6. Catalyst 4.9, Accelrys Inc., San Diego, CA, http://
www.accelrys.com, 2003.
2
2
27. Laggner, C.; Schieferer, C.; Fiechtner, B.; Poles, G.;
Hoffmann, R. D.; Glossmann, H.; Langer, T.; Moebius, F.
F. J. Med. Chem. 2005, 18, 4754–4764.
tion constants (K values) for test compounds were
calculated using the EBDA/LIGAND program, pur-
i
3
6
2
8. Mohamadi, F.; Richards, N. G. J.; Guida, W. C.;
Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.;
Hendrickson, T.; Still, W. C. J. Comput. Chem. 1990, 11,
chased from Elsevier/Biosoft.
4
40–467.
References and notes
29. Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson,
T. J. Am. Chem. Soc. 1990, 112, 6127–6129.
30. Collina, S.; Loddo, G.; Barbieri, A.; Linati, L.; Alcaro, S.;
Chimenti, P.; Azzolina, O. Tetrahedron: Asymmetry 2004,
15, 3601–3608.
1
. Martin, W. R.; Eades, C. G.; Thompson, J. A.; Huppler,
R. E.; Gilbert, P. E. J. Pharmacol. Exp. Ther. 1976, 197,
5
17–532.
2
3
. Skuza, G. Polish J. Pharmacol. 2003, 55, 923–934.
. Quirion, R.; Chicheportiche, R.; Contreras, P. C.; John-
son, K. M.; Lodge, D.; Tam, S. W.; Woods, J. H.; Zukin,
S. R. Trends Pharmacol. Sci. 1987, 10, 444–446.
31. Chang, G.; Guida, W. C.; Still, W. C. J. Am. Chem. Soc.
1989, 111, 4379–4386.
32. WebLab Viewer Lite 3.5; Accelrys (Molecular Simulations
Inc.): San Diego, CA, 1999.
4
5
6
. Walker, J. M.; Bowen, W. D.; Walker, F. O.; Matsumoto,
R. R.; de Costa, B.; Rice, K. C. Pharmacol. Rev. 1990, 42,
33. Matsumoto, R. R.; Bowen, W. D.; Tom, M. A.; Vo, V. N.;
Truong, D. D.; De Costa, B. R. Eur. J. Pharmacol. 1995,
280, 301–310.
34. DeHaven-Hudkins, D. L.; Ford-Rice, Y. Eur. J. Pharma-
col. 1992, 227, 371–378.
35. Macth, R. H.; Smith, C. R.; Childers, S. R. Life Sci. 1995,
57, 57–62.
36. McPherson, G. A. J. Pharmacol. Methods 1985, 14, 213–
228.
3
35–402.
. Quirion, R.; Bowen, W. D.; Itzhak, Y.; Junien, J. L.;
Musacchio, J. M.; Rothman, R. B.; Su, T. P.; Tam, S. W.;
Taylor, D. P. Trends Pharmacol. Sci. 1992, 13, 85–86.
. Hellewell, S. B.; Bruce, A.; Feinstein, G.; Orringer, J.;
Williams, W.; Bowen, W. D. Eur. J. Pharmacol. 1994, 268,
9
–18.