Novel mannich bases, 5-arylimidazolidine-2,4-dione derivatives with dual 5-HT1A receptor and serotonin transporter affinity

Article abstract of DOI:10.1002/ardp.201200378

A computer aided ligand design study of imidazolidine-2,4-dione derivatives was conducted in order to obtain compounds with dual 5-HT_1A receptor and serotonin transporter (SERT) affinity. According to molecular modeling results, series of Mannich bases were chosen and synthesized. Investigated compounds were tested for 5-HT_1A, 5-HT_2A, α₁ and SERT affinity. Two selected compounds (5, 9) were characterized in functional experiments and possessed a pharmacological profile which may enhance SERT blocking efficacy - 5-HT_1A partial agonism and 5-HT_2A antagonism in one molecule. Furthermore these compounds displayed satisfactory selectivity over adrenergic α₁ receptors. The most promising compounds, 5-arylimidazolidine-2,4-dione derivatives with 4-(3-chlorophenyl)piperazinylmethyl moiety were tested for antidepressant and anxiolytic activity. In particular, compound 5 (5-(2-methoxyphenyl)-3-{1-[4-(3-chlorophenyl)piperazin-1-yl]methyl} -imidazolidine-2,4-dione), tested in the forced swim test in mice, exhibited a favorable antidepressant-like profile without affecting spontaneous locomotor activity. Copyright

Full text of DOI:10.1002/ardp.201200378

98
Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Full Paper
Novel Mannich Bases, 5-Arylimidazolidine-2,4-dione
Derivatives with Dual 5-HT1A Receptor and Serotonin
Transporter Affinity
Anna Czopek1, Marcin Kołaczkowski1, Adam Bucki1, Hanna Byrtus1, Maciej Pawłowski1,
2
3
4
5
Agata Siwek , Andrzej J. Bojarski , Marek Bednarski , Dagmara Wrobel , and Anna Wesołowska
5
´
1
´
Department of Pharmaceutical Chemistry, Jagiellonian University Medical College, Krakow, Poland
Department of Pharmacobiology, Jagiellonian University Medical College, Krakow, Poland
2
3
´
´
Department of Medicinal Chemistry, Institute of Pharmacology, Polish Academy of Sciences, Krakow,
Poland
4
5
´
Department of Pharmacodynamics, Jagiellonian University Medical College, Krakow, Poland
Department of Clinical Pharmacy, Jagiellonian University Medical College, Krakow, Poland
´
A computer aided ligand design study of imidazolidine-2,4-dione derivatives was conducted in order
to obtain compounds with dual 5-HT1A receptor and serotonin transporter (SERT) affinity. According
to molecular modeling results, series of Mannich bases were chosen and synthesized. Investigated
compounds were tested for 5-HT1A, 5-HT2A, a1 and SERT affinity. Two selected compounds (5, 9) were
characterized in functional experiments and possessed a pharmacological profile which may enhance
SERT blocking efficacy – 5-HT1A partial agonism and 5-HT2A antagonism in one molecule. Furthermore
these compounds displayed satisfactory selectivity over adrenergic a1 receptors. The most promising
compounds, 5-arylimidazolidine-2,4-dione derivatives with 4-(3-chlorophenyl)piperazinylmethyl
moiety were tested for antidepressant and anxiolytic activity. In particular, compound 5 (5-(2-
methoxyphenyl)-3-{1-[4-(3-chlorophenyl)piperazin-1-yl]methyl}-imidazolidine-2,4-dione), tested in
the forced swim test in mice, exhibited a favorable antidepressant-like profile without affecting
spontaneous locomotor activity.
Keywords: 5-HT1A receptor ligands / Antidepressants / Imidazolidine-2,4-dione / Mannich bases / Serotonin transporter
inhibitors
Received: October 3, 2012; Revised: November 3, 2012; Accepted: November 9, 2012
DOI 10.1002/ardp.201200378
Introduction
in mental activity, but they also have social and economic
dimension [1, 2].
Depression has been described by the World Health
Organization as a mental disorder with symptoms of depres-
sed mood, loss of interest or pleasure, feelings of guilt, low
self-esteem, disturbed sleep, loss of appetite, low energy,
and poor concentration. These symptoms, in the chronic
or recurrent form, lead to impaired ability to cope with
everyday duties, or, in the worst case, death. The effects of
long-term depression not only lead to a significant reduction
In past decades, monoamine reuptake inhibitors were the
most significant drugs applied as a standard in the treatment
of depression – in particular, the serotonin and norepi-
nephrine reuptake inhibitors [3, 4]. Besides their side effects,
the major drawback of these is the delayed onset of their
antidepressant activity [5]. The therapeutic lag is believed to
be caused by activation of 5-HT1A autoreceptors, and the
delayed onset is due to the time required for desensitization
of these autoreceptors [6]. Therefore, one of the ways to
accelerate the antidepressant activity may be introduction
of additional interactions with serotonin 5-HT1A receptors
(5-HT1AR). The mechanism of action, combining serotonin
reuptake inhibition and 5-HT1AR antagonism or partial ago-
nism, should prevent the feedback that inhibits release of
Correspondence: Anna Czopek, Department of Pharmaceutical
Chemistry, Jagiellonian University Medical College, 9 Medyczna Str,
´
30-688 Krakow, Poland.
E-mail: aczopek@cm-uj.krakow.pl
Fax: þ48 12 657 02 62
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Novel Mannich Bases with Dual 5-HT1A/SERT Affinity
99
serotonin from neurons, thus leading to increased efficiency
of the antidepressant drug. This assumption has been based on
clinical observations – it was found that patients treated with
serotonin reuptake inhibitors (SSRIs) and pindolol (b-blocker/
5-HT1A antagonist/partial agonist) felt the positive effects of
therapy more quickly than the group who received SSRIs only
[7]. The most recently introduced antidepressant drug vilazo-
done acts as SSRI and partial agonist of 5-HT1A receptor. An
important advantage of vilazodone is that its antidepressant
effect is observed after only 1 week of treatment [8, 9].
In previous studies [10], we reported the synthesis of 5,5-
disubstituted- and 5-spiro-imidazolidine-2,4-diones with aryl-
piperazinylpropyl moiety which exhibited diversified affinity
and intrinsic activity for serotonin 5-HT1A and 5-HT2A recep-
tors (5-HT2AR). The most interesting compounds behaved like
agonists or partial agonists of 5-HT1ARs and they showed
antidepressant activity. Bearing in mind the clinical signifi-
cance of compounds with dual mechanism of action, being
serotonin transporter (SERT) blockers and 5-HT1A ligands, we
decided to incorporate the serotonin reuptake inhibition
into the profile of our compounds, interacting with 5-HT1A
receptors. Therefore we performed a computer-aided design
of the dual ligands, resulting in development of Mannich
bases of the arylpiperazinemethylimidazolidine-2,4-dione
structure. The new compounds were tested for their affinity
for SERT, 5-HT1A, and 5-HT2A receptors as well as for their
selectivity over a1-adrenergic receptors (a1R). The most prom-
ising ones were characterized in functional experiments and
tested for antidepressant and anxiolytic activity.
affinity ligands. Therefore, we obtained a homology model of
SERT, based on LeuT crystal structure (see Experimental part
for details), and we examined the sertraline binding mode
within the binding site. Sertraline was chosen as a model
ligand due to its high affinity and selectivity for SERT as well
as its fairly rigid structure, which made it suitable for ligand-
steered binding site modeling. We also compared the docked
pose of sertraline in the SERT model with its crystal structure
in LeuT, which proved to be highly similar, being also an
argument supporting the accuracy of the obtained homology
model. Moreover, we projected structures of some other SERT
blockers, crystalized in complex with LeuT in the binding
site of modeled SERT. All of the structures presented quite
similar binding mode, with the most notable feature of an
aromatic ring penetrating the hydrophobic cavity, formed by
Leu99, Trp103, Tyr176, Ile179, Phe335, and aliphatic chain of
Arg104. Despite some differences in binding mode, resulting
from diversified structures of the examined ligands, this
interaction was present in all cases (Fig. 2A). As a starting
point for dual ligands design, we used a compound from
previously described series, an arylpiperazinepropyl deriva-
tive of 20,30-dihydrospiro[imidazolidine-4,10-indene]-2,4-dione
[10] (Fig. 1). This compound proved to be a partial agonist of
5-HT1AR, but displayed no affinity for SERT. In order to
explore this fact at a molecular level, we docked this com-
pound to the homology model of SERT and examined its
binding mode, paying particular attention to the possibility
of emergence of the above-mentioned aromatic ring inter-
action. As a result, we found that the compound was docked
to the SERT binding site with arylpiperazine moiety and
therefore was unable to penetrate the hydrophobic binding
cavity with the aromatic ring in the manner observed for
the reference SSRIs (Fig. 2B). Based on this observation, we
decided to shorten the alkyl linker from 3 to 1 unit, to enable
the SERT binding site penetration with the aryl ring of the
Results
Computer aided ligand design
To understand the most significant features of ligand binding
to SERT we examined the binding mode of reference high
N
NH2
O
O
F
F
N
F
N
O
N
H
N
H
(a)
(c)
(d)
Cl
Cl
H
N
R
O
N
Figure 1. Structures of known antidepressants:
fluoxetine (a), sertraline (b), vilazodone (c),
and an arylpiperazinepropyl derivative of 20,30-
dihydrospiro[imidazolidine-4,10-indene]-2,4-dione
(d) [10].
N
N
N
H
O
(b)
R = H, 2-OCH3, 3-Cl
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

100
A. Czopek et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Figure 2. (A) Projection of SSRIs crystals with LeuT in the binding site of the SERT homology model showing the well superimposed
aromatic rings interacting with the hydrophobic cavity between TMHs 1, 3, and 10. (B) Top-scored pose of a compound used as a starting
point for modifications, showing its poor fit with reference sertraline. (C) Top-scored pose of a mid-stage modification compound with
shortened alkyl chain, presenting its improved fit with the reference ligand. (D) Top-scored pose of a final structure in the SERT binding
site. Residues forming important interactions with the ligands are presented as thick sticks. Dotted lines represent H-bonds with polar
residues.
spirohydantoin moiety. The resulting Mannich base was
found to interact with the SERT model in the desired mode,
having the aromatic moiety well superimposable on those
of the reference ligands (Fig. 2C). Bearing in mind the
putative importance of good fit of the aryl portion to
the binding cavity, we made it more flexible by replacing
the spiro ring system with one substituted phenyl ring
at position 5 of the imidazolidine-2,4-dione. The newly
obtained structure presented good fit with the binding site
of the SERT model, especially within the hydrophobic pocket,
with better score of ligand–protein complex quality
(Fig. 2D). In order to verify whether the modified structure
is likely to maintain the originally occurring affinity for
5-HT1AR it was docked to its homology model. The binding
site of the 5-HT1A serotonin receptor is placed on the extra-
cellular side of the protein, along the third transmembrane
helix (TMH 3), where ligands are anchored by a charge-
reinforced hydrogen bonding to Asp3.32 (Ballesteros–
Weinstein nomenclature [11]). It is divided into two pockets:
(i) situated between TMHs 3, 5, and 6, where arylpiperazine
moiety interacts with aromatic amino acids of the latter
helix, particularly Phe6.51 and Phe6.52, and (ii) constituted
by amino acids localized on TMHs 2, 3, and 7. 5-Arylhydantoin
moiety was found to interact with this binding cavity,
establishing van der Waals and polar contacts with
Tyr2.64, Phe3.28, Trp7.40, and Tyr7.43. Moreover, the
o-methoxyphenyl group is likely to form an H-bond with
Asn7.39. The obtained ligand binding mode was in line
with the one previously established for 5-HT1AR ligands,
thus promising its good in vitro activity (Fig. 3). Based
on those findings we proposed a group of 5-phenylimidazo-
lidine-2,4-dione derivatives with an arylpiperazinemethyl
fragment, which were synthesized and subjected to pharma-
cological studies.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Novel Mannich Bases with Dual 5-HT1A/SERT Affinity
101
Figure 3. Top-scored ligand–receptor complex
of compound 5 showing its binding mode in
5-HT1A
R (view from the extracellular side).
Binding site of the receptor is shown as a trans-
parent surface. Residues forming important
interactions with the ligand are presented as thick
sticks. Dotted lines represent H-bonds with polar
residues.
Chemistry
fications described by Goodson et al. [12]. In the next step,
aminoalkylation of the acidic proton of 5-arylhydantoins
carried out in the presence of formaldehyde and substituted
1-phenylpiperazine enabled to obtain final compounds 3–10.
The structure of the final compounds 3–10 was established on
the basis of the results of elemental (C, H, N) and spectral
The designed series of 5-arylhydantoin derivatives was pre-
pared according to the synthetic routes shown in Scheme 1.
The appropriate imidazolidine-2,4-diones were obtained from
the 2-methoxybenzaldehyde and 5-fluoro-2-methoxybenzal-
dehyde by means of the Bucherer–Berg reaction with modi-
X
X
O
a
CHO
X
NH
HN
O
O
O
b
cmpd
1
2
H
F
cmpd
X
R
X
3
4
5
6
7
8
9
H
H
H
H
F
F
F
H
O
2-OCH3
3-Cl
3-CF3
H
2-OCH3
N
O
O
HN
N
N
R
3-Cl
10
F
3-CF3
Reagents, reaction conditions:
(a) KCN, (NH4)2CO3, 50% ethanol.
Scheme 1. Synthesis pathways of investi-
gated compounds.
(b) 4-Substituted piperazine derivatives, formaldehyde, 96% ethanol, reflux.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

102
A. Czopek et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Table 1. Binding data for final compounds on 5-HT1A, 5-HT2A, and
a1 receptors and SERT.
transfected with the human 5-HT1AR (Fig. 4). The assay is
based on the quantification of phosphorylated ERK (pERK)
obtained due to agonist stimulation of 5-HT1AR which
induces phosphorylation of mitogen-activated protein
(MAP) kinases such as extracellular signal-regulated kinase
(ERK 1/2) [16]. In order to determine the central 5-HT2AR
properties of the tested compounds, their ability to evoke
or inhibit the head twitches induced by (ꢀ)-DOI, a 5-HT2AR
agonist, was studied in mice [17, 18]. The results are pre-
sented in Table 2.
Cmpd
Ki (nM)
Ki (mM)
5-HT1A
5-HT2A
a1
SERT
3
4
5
6
7
8
9
10
80 ꢀ 7
25 ꢀ 3
38 ꢀ 5
86 ꢀ 11
693 ꢀ 54
782 ꢀ 38
89 ꢀ 5
nd
4.1 ꢀ 0.5
6.7 ꢀ 0.2
1100 ꢀ 500
555 ꢀ 21
403 ꢀ 21
nd
750 ꢀ 46
807 ꢀ 27
685 ꢀ 20
0.1665 ꢀ 0.0086
0.9644 ꢀ 0.0286
5.0 ꢀ 0.7
95 ꢀ 12
269 ꢀ 35 1084 ꢀ 120
53 ꢀ 4
76 ꢀ 8
73 ꢀ 5
788 ꢀ 54
58 ꢀ 8
5.2 ꢀ 0.2
0.2779 ꢀ 0.0164
0.9072 ꢀ 0.0728
68 ꢀ 9
Antidepressant and anxiolytic activity
The potential antidepressant activity of compounds 5 and 9
was evaluated in the forced swim test in mice [19], whereas
the potential anxiolytic properties were determined in the
four-plate test in mice [20]. To control for the possibility of
occurrence of drug-induced changes in locomotor activity of
mice, which could contribute to behavior in both the
forced swim and the four-plate tests, the influence of the
investigated compounds on the spontaneous locomotor
activity was also carried out. The obtained results are shown
in Tables 3–5.
nd, not determined.
(1H NMR, 19F NMR) analysis. The detailed spectral data of the
synthesized compounds are presented in the Experimental
section.
Pharmacology
Affinity determination
The pharmacological in vitro studies included determination
of affinity for SERT, 5-HT1A, 5-HT2A receptors as well as for Discussion
adrenergic a1 receptors. The compounds (3–10) were tested
in radioligand binding experiments where their ability to
displace [3H]-citalopram, [3H]-8-OH-DPAT, [3H]-ketanserin,
and [3H]-prazosin, respectively, were evaluated [13–15]. The
affinity data are shown in Table 1.
Affinity of the tested compounds for SERT, 5-HT1AR, 5-HT2AR,
and a1R was diversified depending on the biological target, as
well as substitution mode at the main scaffold, and ranged
from nanomolar to low micromolar Ki values. As expected,
based on similarity to the previously studied compounds, as
well as on molecular modeling studies, affinity for the 5-HT1AR
was high in the case of all compounds ranging from 25 to
86 nM, excluding compound 7, displaying moderate affinity
(269 nM) for these receptors. Affinity of the newly synthes-
ized compounds for 5-HT2AR ranged from high to moderate
Intrinsic activity studies
The intrinsic activity of the selected compounds (5, 9) at 5-
HT1AR was assessed by fluorescence detection of phosphoryl-
ated extracellular signal-regulated kinase (pERK) accumu-
lation in Chinese hamster ovary (CHO)-K1 cells stably
cmpd 5
cmpd 9
8-OH-DPAT
8-OH-DPAT
WAY100635 + 10-7 M 8-OH-DPAT
WAY 100635 + 10-7 M 8-OH-DPAT
120
90
60
30
0
120
90
60
30
0
-10
-8
-6
-4
-10
-8
-6
-4
log [cmpd] (M)
log [cmpd] (M)
Figure 4. Concentration-dependence curves of compounds 5 and 9 to active extracellular signal-regulated kinase (ERK) phosphorylation
in CHO-h5-HT1A cells. Obtained values (%) are expressed as percent of the action of 0.1 mM of full agonist, 8-OH-DPAT (100%).
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Novel Mannich Bases with Dual 5-HT1A/SERT Affinity
103
Table 2. The effect of 5 and 9 on the (ꢀ)-DOI (2.5 mg/kg)-induced
implication of a1R antagonism in adverse effects like seda-
tion or hypotension. Unlike in the case of 5-HT1AR, the pre-
dicted good fit between the designed ligands and SERT model
was not reflected in high affinity of the whole group. On the
other hand, the affinity for SERT was very diversified (166–
6700 nM), depending strongly on the substitution pattern in
the aromatic ring of phenylpiperazine moiety, which was not
considered by us as crucial in the initial design approach. Post
hoc analysis of the binding mode suggested that the arylpi-
perazine moiety might be involved in the interaction with
the second hydrophobic pocket between TMHs 10 and 11,
which seems to be as important in terms of affinity as the
previously discussed interaction with the pocket between
TMHs 1, 3 and 10 (Fig. 5). 3-Chloro substitution proved to
be the most advantageous, and, in fact, the only one provid-
ing significant affinity for SERT. Despite a substantial body of
work, both experimental and modeling, devoted to explora-
tion of 3D structure of SERT, either the precise binding site
(vestibular, buried) or the ligand binding mode remain, in
fact, uncertain. The level of ambiguity is even higher than in
the case of monoamine GPCRs, where considerable progress
has been made recently. This may be a source of discrepancies
between the structure–activity relationships, perceived based
on modeling studies, and the actual experimental results.
Nevertheless, the applied ligand design approach resulted
in compounds possessing significant affinity for SERT, higher
than that ever observed in our studies with the previously
described compounds.
head twitch response in mice.
Treatment
Dose
(mg/kg)
Number of head
twitches/20 min
Mean ꢀ SEM
Vehicle þ (ꢀ)-DOI
5 þ (ꢀ)-DOI
–
20.2 ꢀ 2.8
10
20
40
11.0 ꢀ 1.3a)
9.2 ꢀ 1.3a)
6.8 ꢀ 1.4a)
F(3,20) ¼ 10.212 p < 0.001
ID50 ¼ 20.10 (18.60–21.61) mg/kg
19.5 ꢀ 4.1
Vehicle þ (ꢀ)-DOI
9 þ (ꢀ)-DOI
–
10
20
40
11.7 ꢀ 2.3
7.0 ꢀ 1.0a)
5.3 ꢀ 1.0a)
F(3,20) ¼ 6.744 p < 0.01
ID50 ¼ 19.55 (18.05–21.05) mg/kg
Compounds 5 and 9 were administered 30 min before the test.
a)
p < 0.01 versus vehicle þ (ꢀ)-DOI group (Dunnett’s test).
ID50 values were calculated by the method of Litchfield and
Wilcoxon.
(58–1084 nM) indicating the superiority of 4-(3-chloro-
phenyl)piperazine or 4-(3-trifluoromethylphenyl)-piperazine
moiety containing derivatives, which was in line with pre-
viously described SAR for arylpiperazine derivatives [13].
All the tested compounds possessed low affinity for
a1R (> 400 nM), which was also in compliance with the
previously described impact of spacer shortening on a1R
affinity [21]. This property should be considered as an
advantage of the studied compounds, given the putative
Based on the results of affinity studies, compounds 5 and 9,
possessing the desired profile, i.e., displaying significant
affinity for both SERT and 5-HT1AR, as well as 5-HT2AR and
Figure 5. Top-scored pose of compound 5 in
SERT model, showing important interaction
between 3-chlorophenylpiperazine moiety and
the hydrophobic cavity between TMHs 10 and
11 (view from the extracellular side). Binding
site of the transporter is shown as a transparent
surface. Residues forming important interac-
tions with the ligand are presented as thick
sticks. Dotted lines represent H-bonds with
polar residues.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

104
A. Czopek et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Table 3. The effect of 5, 9, and imipramine in the forced swim test in
mice.
low affinity for a1R, were selected for determination of
intrinsic activity. Studies of intrinsic activity at human
5-HT1AR showed that Emax values for stimulation of ERK
phosphorylation by compounds 5 and 9 were 45 and
57.8%, respectively, indicating their partial agonist activity
(0.1 mM 8-OH-DPAT ¼ 100%). The EC50 values for the tested
compounds in the agonist assay were 157 nM and 4.1 mM,
respectively. In the 5-HT2AR intrinsic activity test, both com-
pounds administered alone had no ability to produce head
twitches in mice as did (ꢀ)-DOI. However, both 5 and 9 dose-
dependently and significantly abolished the in vivo effect
produced by (ꢀ)-DOI with similar values of ID50 ꢁ20 mg/kg
(Table 2).
Treatment
Dose
(mg/kg)
Immobility time (s)
Mean ꢀ SEM
Vehicle
5
–
168.8 ꢀ 4.9
149.7 ꢀ 6.3
10
20
30
125.7 ꢀ 8.1b)
88.4 ꢀ 9.9b)
F(3,36) ¼ 21.088
p < 0.0001
Vehicle
9
–
168.8 ꢀ 4.9
10
20
30
151.2 ꢀ 10.3
139.1 ꢀ 5.8a)
101.2 ꢀ 8.3b)
F(3,36) ¼ 14.162
p < 0.0001
The functional profile of the investigated compounds
determined in vitro and in vivo, i.e., partial agonists of 5-HT1A
receptors and 5-HT2A antagonists with affinity for SERT,
suggests that they may show antidepressant- and/or
anxiolytic-like activity. To date, there is a lot of evidence
indicating that such compounds produce behavioral activity
relevant to depression and anxiety. These effects may be
obtained by administration of agonists/antagonists of
serotonin receptors in combination with other therapeutic
compounds, for instance SSRIs, to augment their clinical
effects [22]. Considering the premises described above,
we selected compounds 5 and 9, which combine all
desirable functional activity in one molecule, for further
in vivo preclinical studies employing the forced swim and
the four-plate tests in mice.
Vehicle
Imipramine
–
10
20
167.1 ꢀ 6.7
149.1 ꢀ 10.7
107.8 ꢀ 12.4b)
F(2,27) ¼ 8.760
p < 0.01
Compounds 5, 9, and imipramine were administered 30 min
before the test. n ¼ 9–10 mice per group.
a)
p < 0.05 versus vehicle.
p < 0.01 versus vehicle (Dunnett’s test).
b)
Table 4. The effect of 5, 9, and diazepam in the four-plate test in
mice.
Compounds 5 and 9 (20 and 30 mg/kg) dose-dependently
and significantly reduced the immobility time of mice in the
forced swim test (Table 3). Analog 9 was slightly less active,
which can be connected with its sedative effect observed in
the locomotor activity test (Table 5). However, it is notewor-
thy that the effective antidepressant doses of both 5 and 9
did not stimulate the spontaneous locomotor activity in
mice during the 4–6-min, as well as the 30-min experimental
sessions (Table 5), indicating that both compounds have
specific antidepressant-like activity, and this effect cannot
be explained by competing behaviors, such as locomotor
activity. The results of our successive experiments show
that both analogs produced no anti-anxiety-like effects in
the four-plate test in mice (Table 4).
Treatment
Dose
(mg/kg)
Number of
punished crossings
Mean ꢀ SEM
Vehicle
5
–
3.9 ꢀ 0.3
3.3 ꢀ 0.5
4.6 ꢀ 0.5
4.9 ꢀ 0.6
F(3,36) ¼ 2.152
ns
3.9 ꢀ 0.3
3.8 ꢀ 0.4
3.9 ꢀ 0.4
3.2 ꢀ 0.4
F(3,36) ¼ 0.770
ns
3.5 ꢀ 0.4
5.5 ꢀ 0.5a)
6.8 ꢀ 0.6b)
6.7 ꢀ 0.6b)
F(3,36) ¼ 9.514
p < 0.001
10
20
30
Vehicle
9
–
10
20
30
Vehicleꢂ
Diazepam
–
1.25
2.5
5
Conclusions
A computer-aided design of new imidazolidine-2,4-dione
derivatives as dual 5-HT1A/SERT ligands was performed.
The structure optimization of the starting 5-HT1AR ligands
resulted in Mannich bases with an aryl substituent at
position 5 of imidazolidine-2,4-dione, which showed the best
fit with the SERT binding site. However, the affinity determi-
nation studies of the newly synthesized series revealed that,
Compounds 5 and 9 were administered 30 min before the test.
n ¼ 10 mice per group.
a)
p < 0.05.
p < 0.01 versus vehicle (Dunnett’s test).
b)
ns, non-significant.
ꢂ
Data taken from ref. [32].
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Novel Mannich Bases with Dual 5-HT1A/SERT Affinity
105
Table 5. The effect of 5 and 9 on the spontaneous locomotor
activity of mice.
2009. The models were initially optimized using Protein
Preparation Wizard and afterwards the extracellular loops were
refined using Prime Refinement. In order to optimize the model
taking into account crucial structural features of active ligands,
induced fit docking (IFD) was involved for ligand-based optimiz-
ation [26]. For this purpose a group of 12 possibly rigid com-
pounds with high affinity for a given receptor was chosen. The
top-scored complexes were inspected visually to check the com-
pliance with common binding mode for monoaminergic recep-
tor ligands. The above procedure resulted in 12 structures that
served as molecular targets in further docking studies.
Treatment
Dose
(mg/kg)
Locomotor activity:
number of crossings during:
4–6 min
30 min
Vehicle
5
–
20
30
197.1 ꢀ 20.4
187.4 ꢀ 21.7
201.8 ꢀ 17.1
F(2,27) ¼ 0.135
ns
1344.9 ꢀ 125.0
560.1 ꢀ 94.3a)
630.1 ꢀ 85.5a)
F(2,27) ¼ 7.782
p < 0.01
Homology model of serotonin transporter
Vehicle
9
–
20
30
197.1 ꢀ 20.4
118.9 ꢀ 8.8a)
41.8 ꢀ 5.1a)
F(2,27) ¼ 3.732
p < 0.05
1344.9 ꢀ 126.0
411.4 ꢀ 59.7a)
154.1 ꢀ 16.3a)
F(2,27) ¼ 6.580
p < 0.01
The method of homology modeling developed by our group and
introduced for 5-HT1A serotonin receptor was applied to model
the structure of SERT. The X-ray structure of leucin transporter
(LeuT) was used as the template for homology modeling (PDB id
2A65) since this protein is homologous to SERT. Amino acid
sequence of human SERT, used in this study, figures in
UniProt database under P31645 accession number. Building of
the crude model and the following modifications were analo-
gous to those applied for 5-HT1A modeling. At the IFD stage,
spatial structure of sertraline as a possibly most conformation-
ally rigid ligand of those which were crystalized together with
LeuT [27] was used for binding site optimization. Using the
ligand of known binding mode in LeuT was beneficial since
no unambiguous binding mode information for SERT binding
site is available. Best scored complex was verified in the context
of proper binding mode, which resulted in final SERT model
qualified to perform docking studies.
It is assumed that inhibitors of the SERT may act at two
different active sites: (i) the substrate binding site, so called high
affinity binding site, and (ii) low affinity vestibular binding site.
On the basis of docking studies, we examined both of them to
identify the one that binds our group of compounds. Since the
first LeuT X-ray structures were published (e.g., PDB id 2A65 [28]),
binding mode deliberations focused on the latter site. On the
other hand, there were studies showing that SERT inhibitors act
in a competitive manner in the area of the substrate binding site
[29]. The previously mentioned structure was crystalized in
occluded conformation, therefore there was no possibility to
fit ligands in the tight pocket of the substrate binding site.
Molecular docking to that site was possible since the outward-
facing crystal structure was published [30]. Building homology
model on the template of 3F3A enabled comparison of the bind-
ing modes of our compounds within both binding sites. It turned
out that they were unable to dock correctly to the substrate
binding site. What is noteworthy, in the low affinity binding
site, they take analogous pose as SERT inhibitors (e.g., sertraline)
co-crystallized with LeuT. For that reasons, molecular modeling
studies were performed in the vestibular binding site, using
homology model based on 2A65, which is identical in its spatial
structure to crystals obtained together with SERT inhibitors (e.g.,
PDB id 3GWU).
Compound 5 and 9 were administered 30 min before the test.
n ¼ 10 mice per group.
ns, non-significant.
p < 0.01 versus vehicle (Dunnett’s test).
a)
unlike it had been initially expected, only 3-chlorophenylpi-
perazine derivatives 5 and 9 were proven to possess signifi-
cant affinity for both 5-HT1AR and SERT, indicating high
significance of the arylpiperazine moiety for proper inter-
actions with the transporter binding site. The selected com-
pounds 5 and 9 displayed partial agonist and antagonist
properties at 5-HT1AR and 5-HT2AR sites, respectively, and a
low affinity for a1R. Such profile, combining SERT affinity
with 5-HT1AR partial agonism and 5-HT2AR antagonism,
prompted us to undertake the studies to determine the
antidepressant and anxiolytic properties of these com-
pounds. As a result, compounds 5 and 9 showed the potential
and specific antidepressant activity in the forced swim test
in mice. It is noteworthy that the antidepressant effect of
compound 5 was comparable to that of the reference drug
imipramine in doses not affecting the locomotor activity.
Experimental protocol
Molecular modeling
Homology model of the 5-HT1A serotonin receptor
Homology model of human 5-HT1A serotonin receptor was pre-
pared in accordance with the method published previously [23].
The crystal structure of the b2-a1R (PDB code 2rh1) was used as a
template [24]. Our experience in homology modeling of mono-
aminergic GPCRs as well as the latest studies show the usefulness
of this structure for modeling 5-HT1A receptor [25]. Amino acid
sequence of the receptor was taken from the Uniprot database
(accession number P08908), while sequence alignment was deter-
mined using the GeneSilico Metaserver’s hhsearch method
(https://genesilico.pl/meta2). The preliminary homology model
was generated on SwissModel server (accessible from the pro-
gram DeepView/SwissPdb-Viewer). Further model modifications
were performed utilizing components of Schro¨dinger Suite
Compound docking to the homology models
Docking of all compounds was carried out using Schro¨dinger’s
Glide application with XP (Extra Precision) mode set. In case of
5-HT1A model docking, an interaction constraint to form an
H-bond between Asp3.32 and the protonated nitrogen of
the ligand was applied, since such interaction was proved to
be crucial for aminergic GPCR ligands binding [31]. Docking to
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

106
A. Czopek et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
the SERT model was performed using the same conditions, con-
straining an H-bond interaction between the ligand and Glu493,
since this residue was postulated to be engaged in the ligand
binding in SERT [27]. The ligands were prepared with
Schro¨dinger’s LigPrep application employing OPLS_2001 force
field. Beside of spatial optimization, it produced stereoisomers in
possible protonation states.
(300 MHz, DMSO): d 3.74 (s, 3H), 5.17 (s, 1H), 6.91–7.35 (m, 4H),
8.01 (s, 1H), 10.59 (s, 1H). Anal. calcd. for C10H10N2O3: C, 58.25; H,
4.89; N, 13.59. Found: C, 58.22; H, 5.08; N, 13.14.
5-(5-Fluoro-2-methoxyphenyl)imidazolidine-2,4-dione (2)
Following the general procedure 2 was obtained from 5-fluoro-2-
methoxybenzaldehyde (608 EtOH). Yield: 64%; m.p. 227–2298C;
1H NMR (300 MHz, DMSO): d 3.72 (s, 3H), 5.19 (s, 1H), 7.02–7.20
(m, 3H), 8.07 (s, 1H), 10.73 (s, 1H); 19F NMR (300 MHz, CDCl3): d
ꢄ123.70–123.78 (m, 1F).
Chemistry
Melting points (m.p.) were determined in open capillaries on an
Electrothermal 9300 apparatus and were uncorrected. The
purity of the compounds was confirmed by the thin-layer
chromatography (TLC) performed on Merck silica gel 60 F254
aluminum sheets (Merck; Darmstadt, Germany), using the devel-
oping systems: benzene/ethyl acetate/acetone in the ratio of
10:5:1 by volume and acetone/isopropanol/chloroform in the
ratio of 20:10:1 by volume. Spots were detected by their absorp-
tion under UV light (l ¼ 254 nm). The structures were con-
firmed by spectral (1H NMR, 19F NMR) and elemental (C, H, N)
analysis. Nuclear magnetic resonance spectra were taken with
Varian Mercury 300 MHz spectrometer (Varian Inc., Palo Alto,
CA, USA) in CDCl3 (3–10) or d6-DMSO (1, 2) solution with TMS as
internal standard. Chemical shifts were expressed in d (ppm) and
the coupling constants J in Hertz. Signal multiplicities are
represented by the following abbreviations: s (singlet), br s (broad
singlet), d (doublet), t (triplet), m (multiplet). For selected com-
pounds (1, 3–5) elemental analyses for C, H, N were carried out
with an Elementar Vario EL III apparatus (Hanau, Germany).
Additionally, the liquid chromatography/mass spectrometry
(LC/MS) spectra for chosen compounds (3, 6–10) were obtained
on the Agilent 1100 HPLC coupled to a API 2000 (Applied
Biosystems/MDS Sciex, Concord, Ontario, Canada) triple quadru-
pole mass spectrometer. Electrospray ionization (ESI) was used
for ion production. All the analyses were performed on a C18
XBridgeTM analytical column (2.1 mm ꢃ 30 mm, 3.5 mm,
Waters, Ireland) with gradient elution using a mobile phase
containing acetonitrile and water with an addition of 0.1% of
formic acid.
General procedure for synthesis of compounds 3–10
A mixture of 5-arylimidazolidine-2,4-dione (5 mmol), the substi-
tuted 1-phenylpiperazine (5 mmol) and 36% formalin solution
(15 mmol) in 96% ethanol (30 mL) was refluxed for 25–30 h.
After cooling, the precipitate was filtered off, then purified by
column chromatography (ethyl acetate/methanol, 8:2) and
recrystallized from anhydrous ethanol.
[R,S]5-(2-Methoxyphenyl)-3-[1-(4-phenylpiperazin-1-
yl)methyl]imidazolidine-2,4-dione (3)
Following the general procedure 3 was obtained from 1. Yield:
67%; m.p. 155–1578C; ESI–MS (MþHþ) 381.50; 1H NMR (300 MHz,
CDCl3): d 2.88–2.91 (t, J ¼ 4.80 Hz, 4H), 3.20–3.23 (t, J ¼ 4.80 Hz,
4H), 3.77 (s, 3H), 4.64–4.66 (d, J ¼ 5.40 Hz, 2H), 5.18 (s, 1H),
6.83–6.99 (m, 6H), 7.02–7.29 (m, 4H). Anal. calcd. for
C21H24N4O3: C, 66.30; H, 6.36; N, 14.73. Found: C, 66.60; H,
6.69; N, 14.62.
[R,S]5-(2-Methoxyphenyl)-3-{1-[4-(2-methoxyphenyl)-
piperazin-1-yl]methyl}imidazolidine-2,4-dione (4)
Following the general procedure 4 was obtained from 1. Yield:
55%; m.p. 147–1508C; 1H NMR (300 MHz, CDCl3): d 2.92 (br s, 4H),
3.08 (br s, 4H), 3.82 (s, 3H), 3.85 (s, 3H), 4.63–4.65 (d, J ¼ 5.60 Hz,
2H), 5.20 (s, 1H), 6.83–7.29 (m, 9H). Anal. calcd. for C22H26N4O4: C,
64.37; H, 6.38; N, 13.65. Found: C, 63.98; H, 6.23; N, 13.45.
Appropriate substituted 1-phenylpiperazine derivatives, 2-
methoxybenzaldehyde, 5-fluoro-2-methoxybenzaldehyde, and ot-
her chemicals were commercially available (Aldrich or Fluka)
and were used without purification.
[R,S]5-(2-Methoxyphenyl)-3-{1-[4-(3-chlorophenyl)-
piperazin-1-yl]methyl}imidazolidine-2,4-dione (5)
Following the general procedure 5 was obtained from 1. Yield:
1
71%; m.p. 163–1658C; H NMR (300 MHz, CDCl3): d 2.84–2.88 (t,
J ¼ 5.00 Hz, 4H), 3.19–3.22 (t, J ¼ 5.00 Hz, 4H), 3.75 (s, 3H), 4.62–
4.64 (d, J ¼ 5.60 Hz, 2H), 5.18 (s, 1H), 6.74–7.02 (m, 6H), 7.13–7.26
(m, 3H). Anal. calcd. for C21H23N4O3Cl: C, 60.79; H, 5.59; N, 13.50.
Found: C, 60.45; H, 5.58; N, 13.24.
General procedure for preparation of imidazolidine-
2,4-diones (1, 2)
The imidazolidine-2,4-diones (1, 2) were prepared from the alde-
hyde by Bucherer–Bergs reaction with modifications described
by Goodson et al. [12]. A solution of aldehyde (330 mmol) and
ammonium carbonate (1 mol) in ethanol (330 mL) and water
(220 mL) was warmed to 508C, at which time potassium cyanide
(350 mmol) dissolved in 50 mL of water was dropped in over a
period of 15 min. The mixture was then heated at temperatures
between 56 and 608C for more than 20 h. The reflux condenser
was then replaced by an air condenser and the temperature
raised to 808C for 1 h to remove the excess ammonium carbon-
ate. Then the reaction solution was cooled and acidified. The
precipitated solid was filtered off, washed with water, and recrys-
tallized from a mixture of ethanol and acetone.
[R,S]5-(2-Methoxyphenyl)-3-{1-[4-(3-(trifluoromethyl)-
phenyl)piperazin-1-yl]methyl}-imidazolidine-2,4-dione (6)
Following the general procedure 6 was obtained from 1. Yield:
59%; m.p. 173–1768C; ESI–MS (MþHþ) 449.40; 1H NMR (300 MHz,
CDCl3): d 2.87–2.90 (t, J ¼ 5.00 Hz, 4H), 3.23–3.25 (t, J ¼ 5.00 Hz,
4H), 3.76 (s, 3H), 4.63–4.65 (d, J ¼ 5.10 Hz, 2H), 5.18 (s, 1H),
6.90–7.08 (m, 6H), 7.21–7.26 (m, 3H); 19F NMR (300 MHz,
CDCl3): d ꢄ62.76 (m, F). Anal. calcd. for C23H25F3N4O3: C,
59.73; H, 5.45; N, 12.11. Found: C, 59.79; H, 5.74; N, 12.18.
[R,S]5-(5-Fluoro-2-methoxyphenyl)-3-[1-(4-
5-(2-Methoxyphenyl)imidazolidine-2,4-dione (1)
Following the general procedure 1 was obtained from 2-methox-
ybenzaldehyde (608 EtOH). Yield: 59%; m.p. 194–1958C; H NMR
phenylpiperazin-1-yl)methyl]imidazolidine-2,4-dione (7)
Following the general procedure 7 was obtained from 2. Yield:
61%; m.p. 194–1968C; ESI–MS (MþHþ) 399.40; 1H NMR (300 MHz,
1
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Novel Mannich Bases with Dual 5-HT1A/SERT Affinity
107
CDCl3): d 2.84–2.92 (t, J ¼ 4.80 Hz, 4H), 3.12–3.15 (t, J ¼ 4.80 Hz,
4H), 3.90 (s, 3H), 4.60 (s, 2H), 6.53 (s, 1H), 6.76–6.90 (m, 5H),
6.98–7.18 (m, 4H); 19F NMR (300 MHz, CDCl3): d ꢄ123.71–
123.98 (m, 1F). Anal. calcd. for C21H23FN4O3: C, 63.30; H, 5.82;
N, 14.06. Found: C, 63.44; H, 6.05; N, 14.16.
from seven to eight concentrations (10ꢄ11–10ꢄ4 M) of investi-
gated compounds. For measuring unspecific binding, phentol-
amine – 10 mM was applied. Radioactivity was measured in a
WALLAC 1409 DSA – liquid scintillation counter. All assays were
done in duplicates. Radioligand binding data were analyzed
using iterative curve fitting routines (GraphPAD/Prism,
Version 3.0 – San Diego, CA, USA).
[R,S]5-(5-Fluoro-2-methoxyphenyl)-3-{1-[4-(2-
methoxyphenyl)piperazin-1-yl]methyl}-imidazolidine-2,4-
dione (8)
Serotonin transporter binding assay
The experiment was performed according to the method
described by Owens et al. [15]. Rats’ cerebral cortices were
homogenized in 30 volumes of ice-cold 50 mM Tris–HCl buffer
(pH 7.7 at 258C) containing 150 mM NaCl and 5 mM KCl. Then
the homogenate was centrifuged at 20 000 ꢃ g for 20 min. The
resulting supernatant was decanted and the pellet was resus-
pended in the same quantity of the buffer, and centrifuged as
above. The pellet was resuspended and centrifuged for further
two times. [3H]-Citalopram (50 Ci/mM, NEN Chemicals) was used
for labeling SERT. The final examined mixture consisted of
240 mL tissue suspension, 30 mL of a 1 mM imipramine (dis-
placer), 30 mL of 1 nM [3H]-citalopram and 100 mL buffer con-
taining seven concentrations (10ꢄ10–10ꢄ4 M) of tested
compounds. The incubation was performed in plates
(MAFCNOB 10, Millipore) at 238C. After 60 min the incubation
was terminated by rapid vacuum filtration over glass filters
(Watman GF/B). Next the filters were washed two times with
100 mL of ice-cold buffer (0–48C) and placed in scintillation vials
with scintillation cocktail. Radioactivity was measured in a
WALLAC 1409 DSA – liquid scintillation counter. All assays were
done in duplicates. Radioligand binding data were analyzed
using iterative curve fitting routines (GraphPAD/Prism, version
3.0; San Diego, CA, USA).
Following the general procedure 8 was obtained from 2. Yield:
50%; m.p. 172–1748C; ESI–MS (MþHþ) 429.30; 1H NMR (300 MHz,
CDCl3): d 2.44 (br s, 4H), 2.80 (br s, 4H), 3.67 (s, 3H), 3.73 (s, 3H),
4.81–4.85 (d, J ¼ 5.00 Hz, 2H), 5.27 (s, 1H), 6.35 (s, 1H), 6.74–6.92
(m, 5H), 7.19–7.24 (m, 2H); 19F NMR (300 MHz, CDCl3): d ꢄ123.31
(s, 1F). Anal. calcd. for C22H25FN4O4: C, 61.67; H, 5.88; N, 13.08.
Found: C, 61.97; H, 6.22; N, 13.18.
[R,S]5-(5-Fluoro-2-methoxyphenyl)-3-{1-[4-(3-
chlorophenyl)piperazin-1-yl]methyl}-imidazolidine-
2,4-dione (9)
Following the general procedure 9 was obtained from 2. Yield:
65%; m.p. 190–1928C; ESI–MS (MþHþ) 433.30; 1H NMR (300 MHz,
CDCl3): d 2.83–2.86 (t, J ¼ 4.80 Hz, 4H), 3.19–3.22 (t, J ¼ 4.80 Hz,
4H), 3.81 (s, 3H), 4.61–4.63 (d, J ¼ 5.00 Hz, 2H), 5.14 (br s, 1H),
6.74–6.91 (m, 5H), 6.93–7.25 (m, 3H); 19F NMR (300 MHz, CDCl3): d
ꢄ123.65–123.83 (s, 1F). Anal. calcd. for C21H22ClFN4O3: C, 58.27;
H, 5.12; N, 12.94. Found: C, 58.53; H, 5.43; N, 12.75.
[R,S]5-(5-Fluoro-2-methoxyphenyl)-3-{1-[4-(3-
(trifluoromethyl)phenyl)piperazin-1-
yl]methyl}imidazolidine-2,4-dione (10)
Serotonin 5-HT1A receptor intrinsic activity assay
Following the general procedure 10 was obtained from 2. Yield:
51%; m.p. 170–1738C; ESI–MS (MþHþ) 467.70; 1H NMR (300 MHz,
CDCl3): d 2.86–2.89 (t, J ¼ 4.80 Hz, 4H), 3.23–3.26 (t, J ¼ 4.80 Hz,
4H), 3.74 (s, 3H), 4.63–4.64 (d, J ¼ 5.00 Hz, 2H), 5.14 (s, 1H), 6.83–
6.88 (m, 1H), 6.96–7.09 (m, 7H); 19F NMR (300 MHz, CDCl3): d
ꢄ62.77 (s, 3F), ꢄ122.85 (s, 1F). Anal. calcd. for C22H22F4N4O3: C,
56.65; H, 4.75; N, 12.01. Found: C, 56.82; H, 5.02; N, 11.80.
Chinese hamster ovary (CHO)-K1 cells stably transfected with the
human serotonin 5-HT1A receptor were obtained commercially
from PerkinElmer SA (Brussels, Belgium) and grown in RPMI
supplemented with 10% fetal calf serum, 1.25 mg/mL geneticin
and antibiotics. Cultures were maintained at 378C in an air/CO2
(95:5) water-saturated atmosphere and subcultured weekly. Cells
were plated in 96-well culture plates at about 40 000 cells/well,
grown until 90% confluent. For agonist studies, after removal of
the medium, cells were stimulated for 15 min with compound
diluted in 200 mL serum-free medium, based on preliminary
studies showing the detection of maximum levels of phosphory-
lated extracellular signal-regulated kinase (pERK) after 15 min of
stimulation (data not shown). In antagonist studies, cells were
incubated for 15 min with compound diluted in 200 mL of the
medium and then after removal of the medium, cells were
stimulated with 0.1 mM (þ)-8-hydroxy-2-(di-n-propylamino)tetra-
lin (8-OH-DPAT) in the presence of tested compound dissolved in
200 mL of the medium. At the end of incubation the medium and
compound were removed, and cells were lysed for 15 min at
room temperature with cell lysis buffer (SureFire assay kit,
Perkin Elmer). Lysates were allowed to equilibrate for 15 min
at room temperature. Cell lysates were analyzed for content of
pERK1/2 using an immunometric assay, therefore to 4 mL lysate
was added 7 mL of reaction buffer plus activation buffer mix
containing AlphaScreen1 beads (PerkinElmer). Fluorescence
detection was performed on a plate reader (EnVision, Perkin
Elmer) set, using the method of time-resolved fluorescence
energy transfer (TR-FRET). Efficacies of compounds tested in
In vitro experiments
Serotonin 5-HT1A and 5-HT2A receptor binding assays
Radioligand studies with native 5-HT1A and 5-HT2A receptors
were conducted according to the methods previously described
[12]. Briefly, the following were used: for 5-HT1A assays, rat hippo-
campal membranes, [3H]-8-OH-DPAT (170 Ci/mmol, NEN
Chemicals), and 5-HT (10 mM) for non-specific binding; for 5-
HT2A assays, [3H]-ketanserin (88.0 Ci/mmol, NEN Chemicals)
and methysergide (1 mM) for non-specific binding.
Adrenergic a1 receptor binding assay
The assay was performed according to the method described by
Cheng and Prusoff [14]. Rats’ cerebral cortices were homogenized
in 20 volumes of ice-cold 50 mM Tris–HCl buffer (pH 7.6), and
centrifuged at 20.000 ꢃ g for 20 min (0–48C). The cell pellet was
resuspended in Tris–HCl buffer and centrifuged again.
[3H]Prazosin (19.5 Ci/mM, NEN Chemicals) was used for labeling
a1 receptors. The final incubation mixture (final volume 300 mL)
consisted of 240 mL membrane suspension, 30 mL of
a
[3H]Prazosin (0.2 nM) solution and 30 mL of buffer containing
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

108
A. Czopek et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
CHO-h5-HT1A cells are expressed relative to 0.1 mM 8-OH-DPAT
(100%). Isotherms were analyzed by nonlinear regression, Rusing
GraphPad Prism (GraphPad Software Inc., San Diego, CA) to yield
EC50 and Emax values [16].
Locomotor activity in mice
The locomotor activity of mice was recorded in photoresistor
actometers (24 cm in diameter) illuminated by two light beams,
which were connected to a counter for the recording of light-
beam interruptions. The mice were placed individually in the
actometers, and the number of crossings of the light beams was
counted twice: between 2 and 6 min (i.e., the time equal to the
observation period in the forced swim test), and during 30-min
experimental sessions.
In vivo experiments
The experiments were performed on male Albino Swiss mice
(24–28 g), purchased from a licensed breeder Staniszewska
(Ilkowice, Poland) in an environmentally controlled, experimen-
tal room (ambient temperature 21 ꢀ 28C; relative humidity
50–60%; 12:12 light/dark cycle, lights on at 8:00). Standard
laboratory food and filtered water were freely available.
Animals were assigned randomly to treatment groups. All the
experiments were performed by two observers unaware of the
treatment applied between 9 a.m. and 2 p.m. on separate groups
of animals. All the experimental procedures were approved by
the I Local Ethics Commission for Animal Experiments at
Statistics
The obtained data were analyzed by one-way analysis of variance
followed by Dunnett’s test. ID50 values were calculated by the
method of Litchfield and Wilcoxon.
The authors wish to thank the Polish Ministry of Science and Higher
Education for financial support (grant no. 2P05F04329). Part of this
work was presented at XII BDSHC in Tihany, Hungary. Schro¨dinger
Suite 2009 was licensed for Jagiellonian University Collegium Medicum.
´
Jagiellonian University in Krakow. (ꢀ)-2,5-Dimethoxy-4-iodoam-
phetamine (hydrochloride, (ꢀ)-DOI, Sigma–Aldrich, Inc., USA)
was dissolved in saline. The investigated compounds were sus-
pended in a 1% aqueous solution of Tween 80. (ꢀ)-DOI and the
tested compounds were given intraperitoneally (ip), in a volume
of 10 mL/kg. The tested compounds (5, 9) were administered
30 min before the experiments. Each experimental group con-
sisted of six to ten animals per dose, and all the animals were
used only once.
The authors have declared no conflict of interest.
References
[1] L. E. Schechter, R. H. Ring, C. E. Beyer, Z. A. Hughes, X.
Khawaja, J. E. Malberg, S. Rosenzweig-Lipson, NeuroRx 2005,
2, 590–611.
Head twitch response in mice
To habituate mice to the experimental environment, each
animal was randomly transferred to a 12 cm (diameter) ꢃ 20 cm
20 cm (height) glass cage lined with sawdust, 30 min
before treatment. Head twitches in mice were induced by
(ꢀ)-DOI (2.5 mg/kg). Immediately after the treatment, the
number of head twitches was counted throughout 20 min
[17]. The tested compounds (5, 9) were administered 60 min
before (ꢀ)-DOI.
[2] WHO, (http://www.who.int/mental_health/management/
depression/definition/en/05.12. 2012).
[3] P. Pacher, V. Kecskemeti, Curr. Med. Chem. 2004, 11, 925–943.
[4] D. N. Middlemiss, G. W. Price, J. M. Watson, Curr. Opin.
Pharmacol. 2002, 2, 18–22.
[5] M. M. Katz, J. L. Tekell, C. L. Bowden, S. Brannan, J. P.
Houston, N. Berman, A. Frazer, Neuropsychopharmacology
2004, 29, 566–579.
Four-plate test in mice
[6] D. S. Kreiss, I. Lucki, J. Pharmacol. Exp. Ther. 1995, 274, 866–876.
The box was made of an opaque plastic and was rectangular
(25 cm ꢃ 18 cm ꢃ 16 cm) in shape. The floor was covered with
four rectangular metal plates (11 cm ꢃ 8 cm), separated by a
4 mm gap. The plates were connected to a source of continuous
current which enabled a 120 V difference of potential between
two adjacent plates for 0.5 s when the experimenter pressed the
switch. Individual mice were placed gently onto the plate, and
were allowed to explore for 15 s. Afterwards, each time a mouse
passed from one plate to another, the experimenter electrified
the whole floor which evoked a visible flight reaction of the
animal. If the animal continued running, it received no new
shocks for the following 3 s. The episodes of punished crossing
were counted for 60 s [20].
[7] K. John, Curr. Med. Chem. Cent. Nerv. Syst. Agents 2003, 3, 311–
322.
[8] K. Rickels, M. Athanasiou, D. S. Robinson, M. Gibertini, H.
Whalen, C. R. Reed, J. Clin. Psychiatry 2009, 70, 326–333.
[9] L. A. Dawson, J. M. Watson, CNS Neurosci. Ther. 2009, 15, 107–
117.
[10] A. Czopek, H. Byrtus, M. Kołaczkowski, M. Pawłowski, M.
´
Dybała, G. Nowak, E. Tatarczynska, A. Wesołowska, E.
Chojnacka-Wojcik, Eur. J. Med. Chem. 2010, 45, 1295–1303.
´
[11] J. A. Ballesteros, H. Weinstein, Methods Neurosci. 1996, 25,
366–428.
[12] L. H. Goodson, I. L. Honigberg, J. J. Lehman, W. H. Burton, J.
Org. Chem. 1960, 25, 1920–1924.
Forced swim test in mice
[13] H. Byrtus, M. Pawłowski, A. Czopek, A. J. Bojarski, B.
The experiment was carried out according to the method of
Porsolt et al. [19]. Briefly, mice were individually placed in a
glass cylinder (25 cm high; 10 cm in diameter) containing 6 cm
of water maintained at 23–258C, and were left there for 6 min. A
mouse was regarded as immobile when it remained floating on
the water, making only small movements to keep its head above
it. The total duration of immobility was recorded during the last
4 min of a 6-min test session.
´
Duszynska, G. Nowak, A. Kłodzinska, E. Tatarczynska, A.
Wesołowska, E. Chojnacka-Wojcik, Eur. J. Med. Chem. 2005, 40,
820–829.
´
´
´
[14] Y. Cheng, W. H. Prusoff, Biochem. Pharmacol. 1973, 22, 3099–
3108.
[15] M. J. Owens, W. N. Morgan, S. J. Plott, C. B. Nemeroff, J.
Pharmacol. Exp. Ther. 1997, 283, 1305–1322.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2013, 346, 98–109
Novel Mannich Bases with Dual 5-HT1A/SERT Affinity
109
[16] L. A. Bruins Slot, L. De Vries, A. Newman-Tancredi, D. Cussac,
Eur. J. Pharmacol. 2006, 534, 63–70.
W. I. Weis, B. K. Kobilka, R. C. Stevens, Science 2007, 318,
1258–1265.
[17] N. A. Darmani, B. R. Martin, U. Pandey, R. A. Glennon,
Pharmacol. Biochem. Behav. 1990, 36, 901–906.
[25] V. Katritch, V. Cherezov, R. C. Stevens, Trends Pharmacol. Sci.
2012, 33, 17–27.
[18] D. L. Willins, H. Y. Meltzer, J. Pharmacol. Exp. Ther. 1997, 282,
699–706.
[26] W. Sherman, T. Day, M. P. Jacobson, R. A. Friesner, R. Farid, J.
Med. Chem. 2006, 49, 534–553.
[19] R. D. Porsolt, A. Bertin, M. Jalfre, Arch. Int. Pharmacodyn. Ther.
1977, 229, 327–336.
[27] Z. Zhou, J. Zhen, N. K. Karpowich, C. J. Law, M. E. A. Reith, D.
N. Wang, Nat. Struct. Mol. Biol. 2009, 16, 652–657.
[20] C. Aron, P. Simon, C. Larousse, J. R. Boissier,
Neuropharmacology 1971, 10, 459–469.
[28] A. Yamashita, S. K. Singh, T. Kawate, Y. Jin, E. Gouaux, Nature
2005, 437, 215–223.
´
´
´
[21] M. L. Lopez-Rodrıguez, M. L. Rosado, B. Benhamu, M. J.
Morcillo, A. M. Sanz, L. Orensanz, M. E. Beneitez, J. A.
Fuentes, J. Manzanares, J. Med. Chem. 1996, 39, 4439–4450.
[29] L. K. Henry, J. R. Field, E. M. Adkins, M. L. Parnas, R. A.
Vaughan, M. F. Zou, A. H. Newman, R. D. Blakely, J. Biol. Chem.
2006, 281, 2012–2023.
[22] G. V. Carr, I. Lucki, Psychopharmacology (Berl.) 2011, 213, 265–
287.
[30] S. K. Singh, C. L. Piscitelli, A. Yamashita, E. Gouaux, Science
2008, 322, 1655–1661.
´
´
[23] P. Kowalski, J. Jaskowska, A. J. Bojarski, B. Duszynska, A.
Bucki, M. Kołaczkowski, J. Heterocycl. Chem. 2011, 48, 192–198.
[31] C. D. Strader, I. S. Sigal, M. R. Candelore, E. Rands, W. S. Hill,
R. A. Dixon, J. Biol. Chem. 1988, 263, 10267–10271.
[24] V. Cherezov, D. M. Rosenbaum, M. A. Hanson, S. G. F.
Rasmussen, F. S. Thian, T. S. Kobilka, H. J. Choi, P. Kuhn,
[32] B. Y. Ho, A. Karschin, T. Branchek, N. Davidson, H. A. Lester,
FEBS Lett. 1992, 312, 259–262.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com