Antiarrhythmic properties of phenylpiperazine derivatives of phenytoin with α1-adrenoceptor affinities

Article abstract of DOI:10.1016/j.bmc.2012.02.009

An association between α₁-adrenoceptor affinities, hERG K⁺-antagonistic properties and antiarrhythmic activities for a series of phenylpiperazine derivatives of hydantoin (2a-21a) was investigated. New compounds were synthesized and

Full text of DOI:10.1016/j.bmc.2012.02.009

Bioorganic & Medicinal Chemistry 20 (2012) 2290–2303
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Antiarrhythmic properties of phenylpiperazine derivatives of phenytoin
with a1-adrenoceptor affinities
Jadwiga Handzlik a, Marek Bajda b, Małgorzata Zygmunt c, Dorota Macia˛g d,e, Małgorzata Dybała d,
c
c
b
a,
⇑
´
Marek Bednarski , Barbara Filipek , Barbara Malawska , Katarzyna Kiec-Kononowicz
a Department of Technology and Biotechnology of Drugs, Jagiellonian University Medical College, Medyczna 9, PL 30-688 Kraków, Poland
b Department of Physicochemical Drug Analysis, Jagiellonian University Medical College, Medyczna 9, PL 30-688 Kraków, Poland
c Department of Pharmacodynamics, Jagiellonian University Medical College, Medyczna 9, PL 30-688 Kraków, Poland
d Department of Pharmacobiology, Jagiellonian University Medical College, Medyczna 9, PL 30-688 Kraków, Poland
e Department of Psychiatry and Human Behavior, University of Mississippi Medical Center, Jackson, MS, USA
a r t i c l e i n f o
a b s t r a c t
An association between a
1-adrenoceptor affinities, hERG K+-antagonistic properties and antiarrhythmic
activities for a series of phenylpiperazine derivatives of hydantoin (2a–21a) was investigated. New com-
Article history:
Received 4 November 2011
Revised 30 January 2012
Accepted 4 February 2012
Available online 13 February 2012
pounds were synthesized and tested for their affinity for a1-adrenoceptors in radioligand binding assay
using [3H]-prazosin as a selective radioligand. Antiarrhythmic activities in adrenaline- and barium chlo-
ride-induced arrhythmia models, an influence of the phenylpiperazine derivatives on the ECG-compo-
nents and blood pressure were tested in vivo in normotensive rats. The hERG K+-antagonistic
properties of the most potent antiarrhythmic agents were investigated in silico by the use of program
Keywords:
Phenylpiperazine
Hydantoin
Antiarrhythmic activity
hERG
QikProp. The highest a1-adrenoceptor affinity (Ki = 4.7 nM) and the strongest antiarrhythmic activity in
adrenaline induced arrhythmia (ED50 = 0.1 mg/kg) was found for 1-(4-(4-(2-methoxyphenyl)piperazin-
1-yl)butyl)-3-methyl-5,5-diphenylimidazolidine-2,4-dione hydrochloride (19a). The results indicated a
significant correlation between
a1-AR affinities (pKi) and antiarrhythmic activity (ED50) in adrenaline
model (R2 = 0.92, p <0.005). Influence of the examined phenylpiperazine hydantoin derivatives on hERG
K+ channel, predicted by means of in silico methods, suggested their hERG K+-blocking properties.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
effect of selective a1-AR agonists against arrhythmias in isolated
hearts,9,10 whereas the others clearly show that stimulation of
these receptors promotes arrhythmias.11,12 Several lines of evi-
Cardiac arrhythmia is one of the main reasons of sudden death
in cardiovascular diseases. Therefore a search for new antiarrhyth-
mic drug targets is a current therapeutic challenge. The first classi-
fication of antiarrhythmic agents, introduced by Singh and
Vaughan Williams in 19701,2 has been revised and modified by
newer approaches3–5 giving the current version, Vaughan Williams
Classification—2001, which divides antiarrhythmic agents into four
main groups in accordance to their blocking properties for Na+
channels, b-adrenergic receptors, K+ and Ca+ channels, respec-
dence suggest that prazosin and other
a
1-adrenoceptor antagonists
possess antiarrhythmic properties.13,14 The
a1-adrenergic antago-
nistic activity of prazosin distinctly decreases the incidence of
malignant ventricular arrhythmias associated with either myocar-
dial ischemia or subsequent reperfusion.15
The human hERG K+ is another point of interest in search for
antiarrhythmic agents.16–21 Potassium hERG channels regulate
heart function because they are responsible for K+ current (IKr),
which plays a very important role in the repolarization phase.17
Blockade of the K+ current however leads to the prolongation of
the QT interval and it may be associated with a life-threatening
arrhythmia called torsades de pointes.18 Due to their function,
hERG channels can be treated as antitarget or target, respec-
tively.19 Some drugs such as cisapride, terfenadine or sertindole
are able to block hERG channels and they can cause serious heart
rate disturbances. It is very important to design new non-cardio-
logic drugs which are deprived of the ability to block the potassium
channels.20 On the other hand, it is well known that class III of
antiarrhythmic drugs acts by blocking potassium channels. In the
tively.6 Although
Vaughan Williams classification, recent studies indicate their
increasing role in arrhythmia mechanisms, especially, in the case
of ischemic arrhythmia.7,8 Currently available pharmacological
studies on the effect of
mias are not consistent. Some of them describe cardioprotective
a1-adrenergic agents are not considered in the
a1-adrenoceptor stimulation and arrhyth-
⇑
Corresponding author. Tel.: +48 012 620 55 81; fax: +48 012 620 55 96.
E-mail address: mfkonono@cyf-kr.edu.pl (K. Kiec´-Kononowicz).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2012.02.009

J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
2291
literature various channel mutations have been described. Among
2. Results and discussion
2.1. Synthesis
them channelopathy of hERG can be found. It is the base of the
newly described disorder called congenital short QT syndrome
(SQTS).21 In this case hERG blockers could serve as useful therapeu-
tic agents. Thus, the hERG channels can be treated as antitarget or
target, depending on the proposed therapeutic indication. It should
be also pointed out that not only blockers are used. There are also
known hERG activators (e.g., mallatoxin), which are very interest-
ing in the treatment of long QT syndrome.22
Synthesis of the first generation (2–11) of lead AZ-99 was de-
scribed earlier.25,28 Synthesis of compounds 12–21 was performed
according to Scheme 1. Phenylpiperazine hydantoin derivatives
with 2-hydroxypropyl linker 12–17 were obtained within three-
step synthesis, starting from 5,5-diphenylhydantoin (22). Two
steps of N-alkylation of hydantoin, giving compounds 23–25 and
26, 27, were performed using methods described previously.25,28
In the last step derivatives with oxirane ring (26 and 27) were used
as alkylating agents to react with corresponding phenylpiperazines
Our previous studies were focused on the search for new antiar-
rhythmic agents among derivatives of phenytoin23,24 and as a result
several active compounds have been characterized. The compounds
displayed cardioprotective properties against reperfusion induced
arrhythmia in vitro24 or against barium chloride- and adrenaline
induced arrhythmia in anesthetized rats.25 Especially, phenylpiper-
azine derivatives of phenytoin, synthesized as the I-st generation of
chemical modifications of compound 1 (AZ-99, Fig. 1), have been an
interesting chemical group that showed high structural similarity
28–30 under microwave irradiation in solvent free conditions.28
A
structure-yield relationship was observed as all N3-methyl deriva-
tives (12–14) were obtained with satisfying yields (60–84%) while
N3-ester derivatives (15–17) with significantly lower one (35–
49%).
to known
a
1-adrenoceptor antagonists.26,27 The previously tested
As starting materials for synthesis of compounds 18–21, N3-
methyl-(23) or N3-ester derivatives (24, 25) were used.
Compounds 23–25 were alkylated at N1-position of hydantoin
within two-phase alkylation using suitable dibromoalkanes, 1,3-
dibromopropane and 1,4-dibromobutane, respectively, to give
N1-bromoalkylhydantoin derivatives 31–34. The reaction was car-
ried out using two methods, standard (A) or microwave-aided
method (B). The standard method was performed at room temper-
ature using long-term stirring (64–92 h) in acetone with K2CO3 and
TEBA. An excess (33%) of dibromolkanes was used to decrease the
risk of an appearance of bis-substituted side-products. Neverthe-
less, TLC-control indicated a presence of a small amount of some
side products during the long-term stirring without heating. Pri-
mary tests to elaborate the process conditions indicated that the
number and an amount of side products increased if the tempera-
ture of the process was turned up. A main reason was a nucleo-
philic substitution within bromide-alkyl fragments in the alkaline
environment to give some alcohols that was easy to go in the basic
condition. If the temperature of the process was close to 20 °C the
appearance of side alcohols was imperceptible, easy to eliminate
by crystallization with alcohol. Long-term stirring method gave
compounds 31 and 33 with satisfying yields (74–75%). In the case
of bromobutyl derivative 32 the yield was slightly lower (57%) and
the synthesis was repeated under microwave irradiation using
methoxyphenylpiperazine derivatives25 showed antiarrhythmic
activity in the adrenaline-induced model of arrhythmia but they
were totally inactive in the barium chloride-induced arrhythmia.
We suggested that antiarrhythmic activity of these compounds
could be relative to their potential antagonistic properties for a1
-
adrenoceptors. Our further investigations, including binding tests
and functional bioassays,28,29 confirmed that the compounds dis-
played significant affinity for
a1-adrenoceptor with antagonistic
properties. In the same investigation, new phenylpiperazine deriv-
atives of phenytoin (2–9) were synthesized and tested in radioli-
gand binding assays giving
a series of active compounds
(generation II, Fig. 1) with affinities for
submicromolar range (Table 1).
a1-adrenoceptors in the
In the present work, further modifications are described as the
III-rd and IV-th generation of AZ-99 (Fig. 1). The main goal of the
studies was to assess antiarrhythmic activity of the phenylpiper-
azine derivatives (2a–21a). Antiarrhythmic activity of the com-
pounds and an influence on the rat heart rhythm as well as an
influence on the blood pressure were investigated in vivo. The
association between
a1-adrenoceptor affinity and antiarrhythmic
properties for a series of
a
1-adrenoceptor agents (Table 1, Table
2) was evaluated. Theoretical studies to predict hERG-affinities
for the most promising antiarrhythmic agents were carried out.
R
O
OH
O
N
N
N
N
A
CH3
N
N
N
N
B
n
C2H5
O
O
1 (AZ-99)
Generation II[28]
Generation III
Generation I[25]
Generation IV
RA
O
RA
RA
RA
A=
A=
A=
A=
O
N
RA=H, 2-Cl, 2-F,
RA=H, 2-Cl, 3-Cl,
RA=MeO-, EtO-
RA=MeO-, EtO-
4-Cl, 2-MeO
B= -, -CH2COO-,
-CH(CH3)COO-,
-CH(C2H5)COO-
B= -, -CH2COOCH2-,
-CH(CH3)COOCH2-,
-CH(C2H5)COOCH2-
B= -, -CH2COO-,
B= -, -CH2COO-,
-CH(CH3)COO-,
R=OH; n =1
R=H; n = 1,2
R=OH, OCOCH3; n=1
R=OH; n =1
Figure 1. Chemical modifications of AZ-99: previous modifications (generation I-II), present modifications (generation III-IV).

2292
J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
Table 1
Structure and
a
1-adrenoceptor affinity for compounds 2a–11a obtained previously25,28
OH
O
R1
R2
N
N
N
N
O
x HCl
Compd
R1
R2
a1-AR Ki (nM)
a1-AR affinity pKi
OCH3
2a
–CH3
160.7 13.6
121.6 14.9
6.79
6.92
OC2H5
O
3a
4a
–CH3
–CH3
>100 000
—
O
5a
6a
7a
8a
9a
10a
–CH3
691.9 16.5
197.8 25.4
251.6 3.8
103.9 4.2
6.16
6.70
6.60
6.98
6.78
6.87
N
OCH3
OC2H5
OCH3
OC2H5
OCH3
OCH3
–CH2COOCH3
–CH2COOCH3
–CH(CH3)COOCH3
–CH(CH3)COOCH3
–CH(CH3)COOC2H5
167.7
8
135.7 31.3
3100 200
11a
–CH(C2H3)COOCH3
5.51
Table 2
Structure and
a
1-adrenoceptor affinity for new compounds (12a–21a)
R2'
R3
O
R1
N
N
N
n
N
O
Compd
R1
R2’
R3
n
a1-AR Ki (nM)
a
1-AR affinity pKi
12a
13a
14a
15a
16a
17a
18a
19a
20a
21a
–CH3
–CH3
–CH3
–CH2COOCH3
–CH2COOCH3
–CH2COOCH3
–CH3
–CH3
–CH(CH3)COOCH3
–CH2COOCH3
H
Cl
F
H
Cl
F
OCH3
OCH3
OC2H5
OCH3
OH
OH
OH
OH
OH
OH
H
H
H
H
1
1
1
1
1
1
1
2
1
2
542.3 19.4
483.0 29.4
564.6 80.5
413.1 75.8
354.7 21.6
3300 200
412.9 21
4.7 1.5
6.27
6.32
6.25
6.39
6.45
5.48
6.38
8.33
6.40
7.60
401.2 25.9
25.0 2,4
method B. The reactants were irradiated under reflux in chemical
microwave oven ‘Plazmatronika’ in temperature 40–50 °C for
2.5 h. The method B allowed to shorten the time of reaction giving
compound 32 with a bit higher yield (67%).
Final products 18–21 were obtained by two-phase alkylation of
appropriate alkoxyphenylpiperazines (35, 36) using an excess
(10%) of alkylating agents, bromoalkyl derivatives 31–34. The
process was performed in acetone in similar conditions to those
described for compounds 31–34. A progress of reaction was
observed for 0–90 h. The further prolongation of stirring did not
improve a yield. In all cases (18–21), a small amount of unreacted
starting materials was still observed. Especially, a presence of aryl-
piperazines complicated purification of final products. To eliminate
the rest of unreacted arylpiperazines, four-steps extraction with

J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
2293
23: R1= CH3
O
O
R1X a
24: R1= CH2COOCH3
25: R1= CH(CH3)COOCH3
HN
NH
HN
NR1
O
O
i
,
Br
Cl
22
O
n
23-25
Br
O
NR1
O
NR1
N
Br
N
O
n
O
O
R2'
R2'
31-34
26,27
26: R1= CH3
NH
N
NH
, i
25: R1=CH2COOCH3
N
31: R1= CH3, n=1
32: R1= CH3, n=2
, ii
28: R2'=H
29: R2'=Cl
30: R2'=F
28-30
35, 36
33: R1= CH(CH3)COOCH3, n=1
34: R1= CH2COOCH3, n=2
35: R2'=OCH3
36: R2'=OC2H5
R2'
R2'
OH
O
O
N
N
N
NR1
N
N
N
NR1
n
O
O
12-17
18-21
HCl
12a-21a
Scheme 1. Synthesis of phenylpiperazine derivatives 12a–21a; a-synthesis described earlier25,28; i-acetone, TEBA, K2CO3; rt (met. A) or mv-irradiation (met. B); ii-mv-
irradiation.
diluted HCl (2%) was used and a residue from organic phase was
crystallized with methanol. The method gave pure crystals of
compounds 20 and 21 with different yields (32–69%). In the case
of compounds 18 and 19, no crystal was obtained. Methanol solu-
tions of the purified compounds were saturated with gaseous HCl
giving pure crystals of corresponding hydrochlorides 18a and 19a.
Additionally, all compounds obtained in basic form were converted
into corresponding hydrochlorides to give compounds 12a–17a,
20a and 21a, useful in pharmacological assays. Compounds 2a–
17a, possessing chirality centers were synthesized and tested as
racemates.
reduce undesirable death of animals by elimination of preliminary
acute toxicity tests and rational selection of compounds for tests
in vivo based on their activity in vitro. The route was divided into
two levels (Fig. 2) based on the results of previous tests in vivo25
and radioligand binding assays.28 The whole group of compounds
(2a–21a) was tested in the level I, which included four steps
(S1–S4). Based on the testing of level I the most promising antiar-
rhythmic agents were selected for further tests within level II. In
the first step (S1), a1-adrenoceptor affinities for all new phenylpip-
erazine derivatives 12a–21a were tested in radioligand binding
studies. Then, primary tests in vivo (S2) were performed for com-
pounds of the II-nd and III-rd generations (2a–9a, 11a–17a). The
compounds were tested at starting doses of 5–10 mg/kg iv in two
models of arrhythmia (adrenaline and BaCl2-induced) and for their
influence on both, the ECG-components and the blood pressure.
Selection of compounds of IV-th generation for the primary tests
2.2. Pharmacology
2.2.1. Route of pharmacological screening
The main goal of all branches of medicinal science is life protec-
tion. Although the current pharmacology and medicinal chemistry
introduced various bloodless in vitro- or in silico methods helpful
in primary screening, they still cannot fully substitute tests in vivo
which are necessary before clinical trials. In order to minimize the
number of animals used in the study, we designed and applied a
new route of primary pharmacological screening (Fig. 2) which al-
lowed us to obtain desirable information about antiarrhythmic
properties and structure–activity relationship in the group of
phenylpiperazine hydantoin derivatives (2a–21a). The classical
Szekeres’ method was modified.30,31 The main objective was to
in vivo was performed based on their affinity for
a1-AR (S3).
Results for generations I-III indicated that most of highly active
compounds (2a, 3a, 6a–10a), totally protected animals from
arrhythmia syndrome at starting doses and had affinities for a1
-
AR with the pKi values >6.5. Thus, the pKi >6.5 was established
as a criterion for selection of compounds of IV-th generation for
adrenaline induced arrhythmia test in vivo. Only two compounds
from this generation, 19a and 21a were selected (S3) for primary
tests in vivo at starting doses (S4). The most active compounds
from this and previous25,28 studies (2a, 3a, 6a–10a, 14a, 19a and

2294
J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
Level I
α1-AR binding
Assays for
Primary tests
in vivo for
selected
compounds of
generation IV
in starting
dose
generation I-II28
Primary tests in
vivo for
Primary tests in vivo
for generation II-III
in a starting dose:
antiarrhythmic,
influence on ECG
and blood pressure
Selection of
α AR
compounds from
generation IV for
primary in vivo
tests
binding
assays for
generation
III-IV
generation I 25
α AR: pKi> 6.5)
Previous
studies
S1
S2
S3
S4
Selection of most active compounds for Level II (lack of extrasystoles) (2a, 3a, 6a-10a, 14a, 19a, 21a)
Level II
Quantitative
ED50 –α AR
relationship
hERG K+ channel blockage in
silico
for selected compounds
ED50 estimation
(adrenaline arrhythmia
model)
S7
S5
S6
Figure 2. Route of pharmacological screening—levels I and II.
21a), which totally decreased an occurrence of the arrhythmogens-
induced ventricular extrasystoles at their starting doses, were se-
lected for the level 2 testing. The level 2 included an estimation
of effective doses (ED50) in full antiarrhythmic assays (S5), the
quantitative analysis of the ED50–pKi relationship (S6) and studies
in silico to predict the hERG-inhibitory properties (S7).
weakly soluble compounds (14a–16a) and for the most potent
a1-AR antagonist (19a), respectively.
2.2.3.1. Influence on the ECG-components. The effect of a com-
pound on the ECG-components is the key factor for its classification
as an antiarrhythmic agent. The effect on the ECG intervals and heart
rate was determined for the compounds 2a–9a, 11a–17a, 19a, 21a
during the 15-minutes observation period immediately following
compound administration. Results are shown in Tables 3 and 4 .
All of the compounds, excluding 8a, tended to slow down the heart
rhythm. In particular, compounds 4a, 7a, 9a, 11a–17a and 21a
significantly decreased the number of cardiac beats per minute.
The highest impact was observed for halogen- or unsubstituted phe-
nylpiperazine derivatives 12a–15a. In the case of the most active
2.2.2. Radioligand binding assays
Affinities of compounds 2a–11a for
a1-adrenoceptors evaluated
previously28 are presented in Table 1. Compounds 12a–21a were
tested for their in vitro affinity for a1-adrenoceptors in rat cerebral
cortex by the radioligand binding assays using [3H]-prazosin as a
specific radioligand. The affinities, described by Ki values (nM)
and their dimensionless value pKi, are shown in Table 2. The new
synthesized compounds showed different affinity at
a1-AR in
a1-AR agents 19a and 21a, a statistically significant decrease (p
nano- or micromolar range. A structure–activity relationship can
be observed. The most active compounds 19a (Ki = 4.7 nM) and
21a possess both 2-methoxyphenylpiperazine fragment and tetra-
methylene linker (Table 2). Especially, a role of the linker seems to
be crucial as butyl derivatives (19a, 21a) were much more potent
than corresponding hydroxypropyl- (2a, 6a) or propyl derivatives
(18a, 20a). In the case of the most active compounds (19a, 21a),
a profitable influence of methyl substituent at N3-hydantoin is
observed that increased (ꢀ5.5-fold) Ki-value comparing to the
methyl acetate N3-terminal fragment (21a). This relationship for
N3-substituents is also shown in case of following hydroxypropyl
derivatives: 2-methoxy- (2a, 6a), 2-ethoxy- (3a, 7a) and 2-fluoro-
phenylpiperazine derivatives (14a, 17a). Additionally, the results
indicated an influence of substituents at phenylpiperazine phenyl
ring on the receptor affinity. It is shown that replacement of 2-alk-
oxy- substituent ( 2a, 3a, 6a, 7a) with 2-halogen ( 13a, 14a, 16a,
<0.05) was observed only for ester derivative 21a.
Most of the tested compounds prolonged the P–Q intervals (2a,
5a, 12a, 14a, 15a 17a, 21a). The strongest effect was observed for
compounds with N3-ester terminated fragment, 2a, 15a and 17a
(9–18%, p <0.01). In the group of ten compounds with higher affin-
ities for a1-AR (pKi >6.5), six compounds ( 3a, 7a, 8a, 9a, 10a and
19a) did not significantly influence the P–Q intervals. Considering
the Q–T interval, a tendency towards the prolongation was
observed for most of the tested arylpiperazine derivatives. The
highest, statistically significant, Q–T prolongation was observed
for N3-methylhydantoin derivatives (12a, 14a) and N3-ester (15a)
whereas active
a1-adrenergic receptor antagonists 3a, 7a, 8a, 9a
and 19a practically did not affect the Q–T intervals at the tested
doses. Most of the tested compounds did not affect the QRS com-
plex. Slight, statistically significant, widening was observed only
for compounds 15a and 13a. The most active
a1-AR antagonist
17a) or free hydrogen (12a, 15a) decreased the affinity at
a
1-AR.
19a did not significantly influence the ECG-components at its start-
The binding range for halogen substituted- and free phenylpiper-
ing dose.
azine derivatives is similar. It is contrary to the pharmacophore
model of
a
1-adrenoceptor antagonists which underlines the role
2.2.3.2. Antiarrhythmic activities in two models of arrhyth-
mia. Arylpiperazine hydantoin derivatives were tested for their
antiarrhythmic properties in two models of arrhythmia, barium
chloride- and adrenaline-induced. Results of our previous studies
indicated that various hydantoin derivatives have distinct antiar-
rhythmic activity in barium chloride induced arrhythmia, and an
additional introduction of phenylpiperazine moiety is responsible
for prophylactic antiarrhythmic activity in adrenaline induced
of presence of lipophilic substituent (HY-2) at phenylpiperazine.27
2.2.3. Tests in vivo
All tests in vivo were performed in male Wistar rats using intra-
venous (iv) administration of tested compounds 2a–9a, 11a–17a,
19a and 21a as their water solutions at starting dose of 10 mg/kg.
The lower starting doses of 5 m/kg and 2.5 mg/kg were used for

J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
2295
Table 3
induced-arrhythmia assay (Fig. 3). Particularly, all compounds pos-
sessing 2-alkoxyphenylpiperazine fragment ( 2a, 3a, 6a, 8a and
11a) were totally inactive. Only two compounds, halogenphenyl-
piperazine derivatives with N3-methyl substituent (13a and 14a),
displayed antiarrhythmic properties in barium chloride-model,
protecting more than 50% of tested animals against BaCl2-induced
heart disturbances and death.
Effects of an iv injection of the investigated compounds on the heart rate and ECG
intervals in anesthetized male Wistar rats (60 mg thiopental/kg ip)
Compd Parameters
Time of observation (min)
5
0
1
15
2a
3a
4a
5a
6a
7a
8a
9a
11a
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
298 12.5
57 1.5
270 11.1
64 2.6⁄⁄⁄
274 12.8
60 1.1
79 5.9
17 2.2
241 12
79 1.8
114 13.3
16 1.2
285 9⁄⁄
61 0.8
80 2.3
24 3.1
267 19.3
59 0.7
90 1.1⁄
268 12.6
59 0.4
77 2.7
72 3.5
17 1.4
104.4 9.2⁄⁄⁄
18 2.4
Compounds (2a–17a, 19a, 21a) were tested for their prophylac-
tic antiarrhythmic activity in adrenaline-induced arrhythmia mod-
16
2
285 17.7
68 6.1
103
15 1.8
329 10.5
61 1.2
246 12
75 5.3
242 14.2
79 0.7
el in rat.30,31 Intravenous (iv) injections of adrenaline (20
lg/kg)
caused sinus bradycardia (100%), atrioventricular disturbances,
ventricular and supraventricular extrasystoles (96%), which led to
death of approximately 66% of animals within 10 5 min. Com-
pounds injected 15 min before adrenaline administration,
decreased the occurrence of extrasystoles, bigeminy and reduced
mortality (Fig. 4). The most active compounds (2a, 3a, 6a–10a,
14a, 19a and 21a) were given at starting doses of 2.5–10 mg/kg.
They totally protected the animals from the death and all above
disturbances. The ED50 values were assessed for the most promis-
ing compounds (Fig. 4). Most of the compounds in this selected
group (2a, 3a, 6a–10a and 19a) displayed ED50 values lower than
those of tolazoline and propranolol (Fig. 4). The highest activity
in this model of arrhythmia was displayed by compound 19a,
which had an ED50 value of 0.1 mg/kg. This compound (19a), con-
taining 2-methoxyphenylpiperazine moiety attached to N3-methyl
9
109 15
18 0.6
118 14.4
17 0.7
284 12⁄⁄
60 2.5
287 15⁄
64 1.7
86 4.2
26 1.9
86 5.1
26 2.9
255 23.1
62 1.2⁄
87 4.7
85 7.8
26
290 27.1
58 1.2
83 2.9
17 0.7
261 19.6
63
74 4.3
17 1.4
2
320 31.7
57 1.5
77 3.5
17 0.7
17 0.3
17
0
292 17.7
268 6.2
64 2.3
262 29.1
62
72
16
2
4
2
63 0.9⁄
2
98 8.9⁄⁄⁄
20 2.6
76
6
18 2.4
325 7.9
57 0.6
97 10.1
15 1.3
360 15
55 1.3
82 1.1
17 0.7
376 12.6
55 1.5
75 2.9
17 0.8
337 10.6
293 10.3
281 19.1⁄ 274 8⁄⁄
60
101 10
14 0.3
2
62 3.2
98 12
15 0.7
363 7.6
60 0.3
98 11.1
15 0.3
362 12.7
57 1.3
350 12.8
hydantoin with butyl linker, had the highest affinity for a1-adreno-
ceptors (Table 2) among the tested compounds (2a–18a, 21a). It
was in accordance with our previous observation that affinities
57
1
55
1
82 1.4
16 0.3
81 0.7
15 0.3
82
2
16 0.3
325 12.1
59 3.1
78
16
326 16.9⁄⁄ 316 9.7⁄
for
a1-AR are closely associated with antiarrhythmic activities in
adrenaline model. To confirm this thesis, quantitative pKi-ED50
relationship was calculated using linear regression method
(Fig. 5). Results obtained for eight out of ten tested compounds
indicated almost a linear correlation (R2 = 0.92, p <0.005). Two
compounds (19a and 21a) with the highest affinities (the outliers)
were excluded from the correlation analysis.
59 2.9
78 2.6
58 2.1
80 2.0
17 0.7
1
1
17
1
261 15.9⁄⁄⁄ 294 12.1⁄⁄ 291 7.6⁄⁄
56
1
56 2.4
82 8.9⁄⁄
25 2.2
57 3.4
78 4.7
23 3.9
56 2.1
71 1.8
25 2.5
75 4.2
23 3.7
Data represent the mean of 6 experiments SEM. Statistical analysis was performed
using one-way ANOVA test ⁄p <0.05, ⁄⁄p <0.02, ⁄⁄⁄p <0.01.
2.2.3.3. Influence on the blood pressure. The hypotensive activ-
ity of compounds (2a–9a, 12a, 14a–17a, 19a, 21a) was determined
following iv administration at a dose of 2.5–10 mg/kg in normoten-
sive rats. An observation was carried out during the 60 minutes.
The results are presented in Figures 6 and 7. Most of the arylpiper-
azine derivatives of phenytoin (2a, 6a–9a, 12a, 14a–17a, and 21a)
showed similar behavior during the test, causing sudden decrease
of both systolic and diastolic blood pressure in the first minutes
after their administration. Then, the blood pressure returned to
the baseline level and remained unchanged until the end of the as-
say. Compounds 3a (Fig. 6) and 19a (Fig. 7), 2-methoxyphenylpi-
perazine derivatives of N3-methylhydantoin differing in the area
of alkyl linker, showed a decrease of blood pressure until the end
of the assay. Particularly, compound 3a displayed significant hypo-
tensive properties causing reduction of both, systolic and diastolic
pressure during the whole test. Furoylpiperazine derivative 4a
model of heart disturbances.23–25 Therefore these two models of
arrhythmia were chosen for primary screening within the present
studies. Mechanism of barium chloride-arrhythmia is not fully
understood. Barium chloride is considered as the most selective
blocker of inward rectifier potassium current IK1 (Kir2). IKl blockers
prolong atrioventricular (AV) node and ventricular action potential
duration (APD) and are effective against various types of ventricular
reentrant tachycardias. Moreover, IKl blockers produce membrane
depolarization (an effect that slows conduction velocity due to a
voltage-dependent inactivation of Na+ channels) and prolongs the
QT interval, both actions being proarrhythmic. Furthermore, bar-
ium is known to interfere with calcium-mediated inactivation of
L-type calcium channels and thus to prolong the open time of the
channel, which could contribute to APD and QT interval prolonga-
tion. Several lines of evidence25,32 indicated that ions Ba2+ cause
an increase of the sodium inward current in Purkinje fibers leading
to the heart rhythm disturbances whereas adrenergic receptors are
involved in adrenaline-arrhythmia mechanism.25
which did not bind to
a1-adrenoceptors as well as pyridylpipera-
zine derivative 5a did not affect the blood pressure in normoten-
sive rats.
2.2.4. hERG-affinity in silico
Compounds of generation II and III ( 2a, 3a, 6a–8a, 11a–17a)
were tested for their prophylactic activity in barium induced mod-
el of arrhythmia in rat. This drastic arrhythmogen causes progres-
sively increasing disturbances of cardiac rhythm, associated with
premature ventricular beats, ventricular tachycardia and ventricu-
lar fibrillation, leading to death in 100% of investigated rats in
2–5 min at the absence of any protecting compounds. Most of
the tested arylpiperazine derivatives (2a, 3a, 6a–8a, 11a, 12a and
15a–17a) of hydantoin were not active in barium chloride
For the most active compounds (2a, 3a, 6a–10a, 14a, 19a and
21a) inhibiting properties at hERG K+ channel were estimated in
silico. Affinities for hERG K+ channel were expressed as logIC50
values (Table 5). For compounds with chirality centers (2a, 3a,
6a–10a, 14a), all possible absolute configurations were consid-
ered, and mean values of logIC50 were calculated as the com-
pounds were synthesized and tested in their racemic forms.
Results of the prediction show that all tested phenylpiperazine
derivatives (2a, 3a, 6a–10a, 14a, 19a and 21a) are likely to possess

2296
J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
Table 4
Effects of an iv injection of the investigated compounds on the heart rate and ECG intervals in anesthetized male Wistar rats (60 mg thiopental/kg ip)
Compd
Parameters
Time of observation (min)
5
0
1
15
263 9⁄⁄⁄⁄
59 3.3
74 2.8
12a
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
Beats/min
P–Q [ms]
Q–T [ms]
QRS [ms]
322 11.1
57 1.4
63 1.5
18 1.5
343 23.4
57 1.4
68 2.8
15 0.4
322 4.9
54 3.1
62 1.6
15 0.6
15 1.8
332 10.4
270 7.5⁄⁄⁄⁄
260 5.3⁄⁄⁄⁄
59 1.2
66 4.9⁄
109 7.9⁄⁄⁄⁄
23 2.4
79 5.8⁄
20 1.5
18 0.5
13a
14a
279 5.6⁄⁄⁄⁄
60 0.8
278 9⁄⁄⁄⁄
60 1.3
76 5.5
303 12.7⁄⁄
60 1.4
83 5.7
72 5.7
17 0.3⁄
16 0.6
16 0.7
272 8.9⁄⁄⁄
64 3.4
251 13.⁄7⁄⁄⁄
258 8.6⁄⁄⁄⁄
59 2.9
62 2.9
81 5.3⁄⁄⁄⁄
20 2.4⁄
74 3.3⁄⁄
17 0.6
68 1.9
16 1.1
18 0.6
16 1.2
17 0.7
15a
16a
17a
19a
21a
246 20.9⁄⁄⁄⁄
72 3.4⁄⁄⁄
97 10.9⁄⁄⁄
33 2.4⁄⁄
290 17.5
62 4.8
261 8.7⁄⁄⁄⁄
65.2 3⁄
81 1.7
272 8.6⁄⁄⁄⁄
64 2.6
57
1
68 4.8
21 1.7
315 15.3
56 2.3
68 3.9
74 1.6
22 3.5
268 18.5
59 2.6
69 3.7
26 3.7
256 21.1⁄
63 2.9
76 4.8
76 6.2
15
1
16
1
15 0.5
17 0.6
328 1.8
56 2.6
68 3.7
15.4 0.7
353 14.4
50 3.1
84 5.1
22 1.5
396,7 31,7
48,3 5,6
62,3 6,9
19,0 2,8
284 6.4⁄⁄
66 3.1⁄⁄⁄
86 4.7⁄⁄
17 0.9
256 10.2⁄⁄
61 1.3
281 13.9⁄⁄⁄
58 1.5
82 6.3
71 3.6
15 0.5
346 21.2
49 3.6
86 8.6
17
1
338 15.6
50 3.8
83 8.6
333 18.6
51 2.9
81 6.9
22 0.4
22 0.4
22 1.1
355,8 23,8⁄
62,6 7,8⁄
74,6 8,6⁄
19,6 2,6
354,3 21,0⁄
63,3 4,6⁄
73,3 9,3⁄
19,3 1,3
357,8 19,2⁄
62,8 9,3⁄
75,6 4,8⁄
19,5 2,5
Data represent the mean of 6 experiments SEM. Statistical analysis was performed using one-way ANOVA test ⁄p <0.05, ⁄⁄p <0.02, ⁄⁄⁄p <0.01, ⁄⁄⁄⁄p <0.001.
Adrenaline induced arrhythmia model
Barium chloride induced arrhythmia
2a, 3a, 6a-10a, 14a, 19a, 21a
75
50
25
0
100
80
60
40
20
0
4a
5a
extrasystoles
fibrillation
mortality
11a, 12a, 17a
13a, 16a
15a
extrasystoles
bigeminy
mortality
Figure 3. Prophylactic antiarrhythmic activity in the barium-induced arrhythmia.
Cpd
ED50 (mg/kg
i.v.)
Cpd
ED50 (mg/kg
i.v.)
2a
3a
6a
7a
8a
9a
0.51 (0.23-1.1)
0.33 (0.1-1.1)
0.69 (0.43-1.34)
0.88 (0.55-1.4)
0.36 (0.13-1.0)
0.34 (0.1-1.15)
10a
14a
19a
21a
0.41 (0.21-0.7)
1.92 (1.09-3.4)
0.1 (0.01-0.13)
1.8 (1.55-2.3)
3.4 (2.6-3.4)
1.05 (0.64-1.73)
hERG K+- blocking potency with significant logIC50 values of À8.40
to (À7.72). In our reasoning, two values of logIC50 were considered
as criteria for potent hERG K+-blocking properties: À6 and À8.
Compounds with logIC50 <À8 were classified as highly active
blockers, whereas those with logIC50 >À6 were considered as com-
pounds deprived of hERG K+-blocking abilities. Compounds with
logIC50 values in the range between À8 to (À6) were considered
as weak or moderate hERG-inhibitors. Taking these criteria into
account, it seems that all compounds except 14a, 2a and 3a could
strongly block hERG channels and prolong the QT interval. Struc-
ture–activity relationship indicated that the highest hERG-blocking
properties were found for 2-methoxyphenylpiperazine derivatives
of hydantoin with butyl linker (21a > 19a). These compounds
Tolazoline
Propranolol
Figure 4. Prophylactic antiarrhythmic activity in the adrenaline-induced arrhyth-
mia. Each value was obtained from 3 experimental groups. Each group consisted of
6 animals. The ED50 values and their confidence limits were calculated according to
the methods of Litchfield and Wilcoxon.30
were also among the most active as
agents (Table 2, Fig. 4). In the group of compounds with 2-
a1-AR- and antiarrhythmic

J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
2297
2
1
0
170
(a)
ED50 = -2,24 pKi + 15,76
R2 = 0,92
150
130
110
90
*
**
**
***
****
*
*
****
***
12a
15a
17a
21a
14a
16a
19a
****
*
***
***
****
****
****
****
*
6
6,5
7
7,5
70
pKi
0
1
2
3
5 10 15 20 30 40 50 60
time (min)
Figure 5. Association between
a1-adrenoceptor affinity (pKi) and antiarrhythmic
properties (ED50) in the adrenaline model for compounds 2a, 3a, 6a–10a and 14a.
Two compounds with the highest affinities (19a and 21a) were excluded.
Quantitative relationship is expressed by linear equation. The correlation was
calculated using Pearson worksheet function (R2 = 0.92; p <0.005).
150
(b)
130
110
90
*
***
*
**
**
****
160
**
(a)
12a
15a
17a
21a
14a
16a
19a
*
*
***
****
70
***
****
***********
140
****
****
*
****
****
*
*
*
*
***
*
*
50
***
***
**
***
0
1
2
3
5
10 15 20 30 40 50 60
time (min)
120
100
80
*
**
****
***
*
Figure 7. Influence on the blood pressure for compounds 12a–21a. Data represent
the mean of 6 experiments SEM. Statistical analysis was performed using a one-
way ANOVA test: ⁄p <0.05, ⁄⁄p <0.02, ⁄⁄⁄p <0.01, ⁄⁄⁄⁄p <0.001.
2a
4a
6a
8a
3a
5a
7a
9a
****
***
****
****
****
****
****
0
1
2
3
5
10 15 20 30 40 50 60
time (min)
2.3. Estimation of antiarrhythmic properties and SAR-studies
150
In the tested phenylpiperazine derivatives of phenytoin 2a–
21a, a number of compounds with antiarrhythmic properties were
found. Chemical modifications of the lead compound 1 changed
pharmacological profile of phenytoin, a member of class Ib of anti-
arrhythmic drugs. Furthermore, an evolution of pharmacological
profile of lead 1 can be observed, as the most active new com-
pounds (2a, 3a, 6a–10a, 19a or 21a) displayed more selective activ-
ities toward the adrenaline induced arrhythmia, whereas they
were totally inactive in the barium chloride one. This could be ex-
(b)
130
110
90
*
*
**
**
***
***
**
*
****
**
**
**
**
****
****
2a
4a
6a
8a
3a
5a
7a
9a
****
****
****
****
******
70
plained by their stronger
a1-adrenolytic properties (pKi = 6.60–
****
8.33) in comparison with that of lead 1 (pKi = 6.28).
50
According to our previous conclusions,25 phenylpiperazine
derivatives of hydantoin displayed properties similar to those of
class Ia and III of antiarrhythmic drugs in Vaughan Williams clas-
sification.3–7 The most active compounds slightly decreased the
number of cardiac beats per minute. This effect was similar to that
of quinidine (class Ia) and it can also correspond to the effect
caused by procainamide (Ia) and bretylium (III). A lack of influence
or a weak prolongation of the PQ interval caused by phenylpiper-
azine hydantoin derivatives 2a–21a was similar to that of bretyl-
ium (III), amiodarone (III), sotalol (III) or procainamide (Ia). A
tendency of the compounds to prolong time of repolarization (QT
interval) was weaker than that of class Ia members (quinidine, pro-
cainamide) and some members of class III (amiodarone, sotalol).
This property was comparable with that of bretylium (class III).
No influence on the QRS complex was observed by most of the
tested hydantoin derivatives (2a–21a) in contrast to the members
of class Ia. Most of the antiarrhythmic drugs belonging to the class
Ib and II–IV of Vaughan Wiliams classification do not influence the
QRS complex as well.3
0
1
2
3
5
10 15 20 30 40 50 60
time (min)
Figure 6. Influence on the blood pressure for compounds 2a–9a. Data represent the
mean of 6 experiments SEM. Statistical analysis was performed using a one-way
ANOVA test: ⁄p <0.05, ⁄⁄p <0.02, ⁄⁄⁄p <0.01, ⁄⁄⁄⁄p <0.001.
hydroxypropyl linker (2a, 3a, 6a–10a, 14a), compounds with
branched
ester
terminate
fragment
at
N3-hydantoin
(10a > 9a > 8a) had higher affinities than those with N3-methyl
acetate (7a > 6a). Compounds with N3-methyl terminate fragment
(3a > 2a > 14a) were less potent than the rest of evaluated com-
pounds but their predicted hERG-blocking properties were stron-
ger than theoretical properties of known hERG-antagonist,
sertindole (logIC50 = À7.17) and ibutilide (logIC50 = À6.85), calcu-
lated with the same method. Some influence of the phenylpiper-
azine phenyl ring substitution was also observed. Thus, the
presence of alkoxyl (2a and 3a) seems to induce stronger pre-
dicted hERG-blocking effect than that of fluorine (14a).

2298
J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
Table 5
Prediction of hERG K+ channel blockage properties in silico
Compd
Config at CHOH
Config at side chain
hERG K+ Channel Blockage: logIC50
2a
R
S
R
S
R
S
R
S
R
S
R
S
R
S
R
S
R
S
R
S
R
S
—
—
R/S
—
—
—
—
—
—
—
—
—
—
S
S
R
R
S
S
R
R
S
S
R
R
À7.77
À7.66
À7.88
À7.64
À8.19
À7.90
À8.17
À7.94
À8.14
À8.03
À8.04
À8.09
À8.24
À8.13
À8.13
À7.94
À8.21
À8.11
À8.24
À8.28
À7.66
À7.52
À8.33
À8.40
À7.72
3a
6a
7a
8a
À7.76
À8.04
À8.05
À8.08
9a
À8.11
À8.21
À7.59
10a
14a
—
—
—
—
19a
21a
Ibutilide
Sertindole
Tolazoline
À6.85 (À7.55)*
À7.17 (À7.85)*
À4.21
—
—
*
Evaluated experimentally.43,44
Adrenergic receptors (ARs) play many roles in the myocardium
ranging from positive inotropic and chronotropic effects, cardiac
preconditioning, arrhythmogenesis and cardiac hyperthrophy. The
Although our theoretical studies on hERG inhibition properties of
the phenylpiperazine derivatives of phenytoin have not been
compared with corresponding experimental data, some effects of
the compounds in tests in vivo can suggest their influence on hERG
K+ currents. A prolongation oftheQTintervalmaybethemainexper-
imental evidence for hERG K+ blocking action in the group of the
tested compounds. Results of primary tests in vivo in rats indicated
that most of the phenylpiperazine hydantoin derivatives (2a–21a)
tend to prolong the QT-interval (Tables 3 and 4). This observation
is in agreement with results of our studies in silico which ascribed
hERG-inhibition properties to all tested compounds (2a, 3a, 6a–
10a, 14a, 19a and 21a) with significant IC50 values (logIC50 <À7).
The order of ability to prolong the QT interval in Wistar rats was as
follows: 14a P 2a ꢁ 6a > 10a ꢁ 21a > 3a ꢁ 9a > 7a > 19a ꢁ 8a.
Interestingly, compound 14a caused the distinctly strongest prolon-
gation of the QT interval, but its predicted potency to diminish hERG
K+ currents was relatively lower than those calculated for the other
compounds (Table5). Inturn, themostpotenthERG-blockerinsilico,
21a, slightlyprolongedtheQTintervalinvivo(18–21%,p<0.05, Table
4), and compound 19a (logIC50 = À8.33) did not significantly influ-
enced this interval. Observed pharmacological properties of tested
compounds could be partially explained by both, hERG-action and
mechanism of the positive inotropic effect induced by
a1-adreno-
ceptor stimulation is still a matter of debate: the 1-adrenoceptor
a
stimulation-induced activation of phospholipase C (PLC) in the cell
membrane, and increased 1,4,5-triphosphate-formation (IP3),
which might mediate the release of Ca2+ from intracellular stores,
and diacylglycerol, which activates protein kinase C (PKC). In addi-
tion, sustained
a1-adrenoceptors stimulation modulates cardiac
repolarizing hERG/IKr potassium currents via protein kinase A
(PKA) and protein kinase C—dependent and PKC-independent
phosphorylation sites in hERG. Stimulation of
a1A-adrenergic
receptors produced a reduction in current amplitude of the rapidly
activating delayed rectifier K+ current in cardiomyocytes what may
induce ventricular arrhythmias, in particular in patients with ische-
mic heart disease or hereditary long QT syndrome. The possible
proarrhythmic action of hERG current inhibition might be counter-
acted by the prevention of arrhythmia via
a1A-adrenoceptor block-
ade, since 1A-adrenergic stimulation may induce arrhythmias
a
through modification of hERG channel activation. Thus, hERG cur-
rents are a primary target for the pharmacological management
of some arrhythmias. Pharmacological blockade of IKr causes
lengthening of the cardiac action potential, which may produce a
beneficial class III antiarrhythmic effect of amiodarone and verapa-
mil. Class III antiarrhythmic drugs with hERG channel blocking
properties are effective in the treatment of supraventricular and
ventricular tachycardia. From the other side reduction in hERG cur-
rents due to either genetic defects or adverse effects can lead to
hereditary or acquired long QT syndromes characterized by action
potential prolongation, lengthening of the Q–Tc interval on the sur-
face ECG, and an increased risk for ‘torsade the pointes’ arrhythmias
and sudden death. In several cardiovascular drugs such as prazosin,
doxazosin, terazosin or carvedilol hERG current block does not gen-
erally lead to severe cardiac arrhythmias with the risk of sudden
cardiac death which may be due to additional pharmacological ef-
fects on ion currents or adrenergic receptors.33–36
a1-addrenoceptor antagonistic properties. Compound 14a is a sig-
nificantly weaker 1-adrenoceptor antagonist than 21a and much
a
weaker than compound 19a. As the results of hERG-blocking proper-
ties in silico have only theoretical meaning, it is highly probable that
a direct hERG-blocking potency34 of compound 14a in real condi-
tions is higher than those of 21a and 19a. In the context of interac-
tions between adrenergic system and hERG currents, the real
hERG-blocking properties of compound 14a could be an effect of
synergistic action of its direct hERG inhibition (suggested by the
studies in silico) and partial indirect hERG -inhibition by adrenergic
receptors. Our previous studies for compound 14a indicated its only
moderate,
antagonistic
properties
at
a1A-adrenoceptors
(pA2 = 6.05).29 Thus, in the presence of compound 14a, in contrary
tothepotentantagonists21aand19a, itispossiblesomestimulation
of
a1A-adrenoceptors induced by native ligands in the rat heart.

J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
2299
Therefore, an additional hERG-inhibition action might be caused by
following results of 1A-adrenoceptor signal transduction path-
antagonists with the affinities in low-nanomolar range. Most of
the tested 2-alkoxyphenylpiperazine derivatives displayed strong
and selective antiarrhythmic activities in the adrenaline induced
arrhythmia model. Their effective dose ED50, lower than that of
a
ways: (i) activation of PKC and PKA, and (ii) a decrease of concentra-
tion of PIP2, a phospholipid enhancing hERG K+ in myocytes,33,34
which is converted into IP3 during
the case of 21a, and particularly, 19a, the blockade of
notproduceanactivation ofPKCand PKA withinthe 1-AR transduc-
a
1A-adrenoceptors action. In
tolazoline, showed a significant correlation with affinities for a1-
a1-ARs does
AR (pKi) evaluated in radioligand binding assays. Analysis of the
pharmacological profile of the most active compounds was per-
formed based on the radioligand binding assay, tests in vivo in
male Wistar rats and theoretical studies in silico on their hERG
a
tion pathways, and in consequence, it restores higher concentration
of PIP2. The higher concentration of PIP2 may enhance hERG cur-
rents to cause an effect opposite to direct hERG inhibition action
which is hypothetized for compounds 19a and 21a, based on results
of our studies in silico. Thus, their actual hERG-inhibitory effect may
be weaker than that of compound 14a, what can be observed as a
much slighter QT-prolongation in the case of 21a, and particularly,
in the case of 19a. However, the data coming from our studies are
not sufficient to confirm or reject this hypothesis, particularly, as
the investigated compounds can be involved into other interactions
with functional proteins contributing to the cardiovascular regula-
tion of the heart rhythm. Especially, the experimental studies on
hERG-inhibition in this chemical group could explain more and we
will focus on them in the future.
K+ blocking properties. The results indicated that both
a1-AR-
and hERG K+-antagonistic properties could determine the pharma-
cological profile of the compounds, which has some similarities to
class III and Ia of the antiarrhythmic agents. The presented studies
proposed two levels of primary pharmacological screening which
seems to be sufficient for the successful selection of promising
antiarrhythmic agents for further pharmacological assays. Based
on this, 1-(4-(4-(2-methoxyphenyl)piperazin-1-yl)butyl)-3-methyl-
5,5-diphenylimidazolidine-2,4-dione hydrochloride (19a) was
found as the most potent antiarrhythmic agent suitable for more
advanced pharmacological assays.
Interestingly, most of the tested phenylpiperazine derivatives
2a–21a showed higher antiarrhythmic potency than that of known
4. Experimental
4.1. Chemistry
a
1-adrenolytic, tolazoline (Table 5) which does not involve hERG K+
channels in its action. As we know the hERG K+ blockade could
modulate the antiarrhythmic effect of an inhibition of 1-adreno-
1H NMR spectra were recorded on a Varian Mercury VX
300 MHz PFG instrument (Varian Inc., Palo Alto, CA, USA) in
DMSO-d6 at ambient temperature using the solvent signal as an
internal standard. IR spectra were recorded on a Jasco FT/IR-410
apparatus using KBr pellets and are reported in cm–1. Thin-layer
chromatography was performed on pre-coated Merck Silica Gel
60 F254 aluminium sheets, the used solvent systems were: (I) tolu-
ene/acetone 40:3; (II) toluene/acetone/methanol 15:5:1; (III) tolu-
ene/acetone/methanol 5:5:1. Melting points were determined
using Mel-Temp II apparatus and are uncorrected. Elemental
analyses were carried out on a Vario EL III Model Elemental
Analyzer. Microwave aided reactions were performed in micro-
wave reactor for organic synthesis ‘Plazmatronika’ in solvents or
in household microwave oven ‘Samsung M1618’ in case of sol-
vent-free condition. Commercial phenylpiperazines (28–30, 35
and 36) were used. Synthesis of compounds 2–11, 22–27 and 34
was described within other works.25,28
a
ceptors in cardiovascular system. The discovery of adrenergic
mechanisms of hERG channel regulation as well as the develop-
ment of strategies to enhance hERG currents and to modify intracel-
lular hERG protein processing may provide novel antiarrhythmic
options in the repolarization disorders.
The obtained results of tests in vivo, and in silico showed that
antiarrhythmic properties of the tested phenylpiperazine
derivatives 2a–21a were not homogenous. At least two groups of
promising antiarrhythmic agents can be selected including 2-alk-
oxyphenylpiperazine derivatives (2a, 3a, 6a–10a, 19a and 21a)
and 2-halogenphenylpiperazine derivatives (13a, 14a). In the first
group, the antiarrhythmic activity arises mainly from
a1-adreno-
ceptor antagonistic action of the compounds. Although their pre-
dicted strong hERG K+ blocking properties could affected the
antiarrhythmic abilities, it can be confirmed only in the case of
compounds 2a, 6a and 21a which caused the QT-prolongation in
rats in vivo (Table 3 and 4). In this context, the most of compounds
2a, 3a, 6a–10a, 19a and 21a displayed highly selective antiarrhyth-
mic action in the model of adrenaline induced arrhythmia
4.1.1. General procedure for bromoalkyl derivatives (31–33)
Method A: A mixture of the appropriate N3-substituted deriva-
tive of 5,5-diphenylhydantoin 23–25 (15 mmol), TEBA (2 mmol,
0.45 g) and potassium carbonate (44 mmol, 6 g) in acetone
(30 mL) was stirred under reflux for 30 min, then an appropriate
dibromoalkane (20 mmol) in acetone (15 mL) was added. The mix-
ture was stirred at room temperature for 64–92 h, according to a
progress controlled by TLC (I). Then, inorganic precipitate was fil-
tered off, the mother liquor was evaporated. The residue was crys-
tallized with methanol or isopropanol.
Method B: A suspension of the appropriate N3-substituted
derivative of 5,5-diphenylhydantoin 23–25 (15 mmol), TEBA
(2 mmol, 0.45 g) and potassium carbonate (44 mmol, 6 g) in ace-
tone (30 mL) was prepared in the oval flask ‘Plazmatronika’. The
suspension was placed in the microwave reactor, stirred under re-
flux for 15 min using 45% of power of irradiation at 40–45 °C. Then,
an appropriate dibromoalkane (20 mmol) in acetone (15 mL) was
added. The mixture was stirred under microwave irradiation using
following program: 1Â cycle 1 (time: 30 min, temperature: 40–
50 °C, max power: 40%) and 4Â cycles 2 (time: 30 min, tempera-
ture: 40–60 °C, max power: 40%) according to a progress controlled
by TLC (I). Then, inorganic precipitate was filtered off, the filtrate
distinctly corresponding to their individual affinities for
a1-AR.
However they were not active in the BaCl2-induced arrhythmia
and also weakly influenced the time of repolarisation (QT-interval).
The derivatives of 2-halogenophenylpiperazine (13a, 14a), possess-
ing moderate affinities for a1-AR (pKi <6.5), did not display selec-
tive antiarrhythmic activities. Interestingly, in the case of
compounds 13a and 14a, a decrease of activities in the adrenaline
model of arrhythmia was accompanied by the significant prophy-
lactic antiarrhythmic activities in BaCl2-induced arrhythmia.
Compounds 13a and 14a showed also a distinct prolongation of
the QT interval, much higher than that of alkoxyphenylpiperazine
derivatives (2a, 3a, 6a–10a, 19a and 21a). .
3. Conclusion
The performed studies allowed us to identify and gain informa-
tion on the mechanisms of the antiarrhythmic activity of a series of
phenylpiperazine derivatives 2a–21a. Among new synthesized
compounds 12a–21a, 2-alkoxyphenylpiperazine derivatives with
butyl linker (19a and 21a) were the most potent a1-adrenoceptor

2300
J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
was evaporated and the residue was crystallized with methanol or
isopropanol.
tinued for the next 4 min at 600 W(2 Â 2 min). White crystals of
compound 12. Yield 84%, mp 136–138 °C; TLC: Rf (III): 0.71; Anal.
Calcd for C29H32N4O3: C, 71.81; H, 6.66; N, 11.56. Found: C,
71.86; H, 6.64; N, 11.27. 1H NMR d (ppm) of base 12: 1.98–2.01
(dd, J1 = 6.05 Hz, J2 = 3.86 Hz, 2H, Pp-CH2), 2.18–2.21 (t, J = 4.81,
4H, Pp-2,6-H), 2.90–3.01(m, 7H, Pp-3,5-H, CHOH, N1-CH2); 3.31
(s, 3H, N-CH3); 4.43–4.45 (d, J = 4.95 Hz, 1H, OH) 6.70–6.75 (t
def., 1H, PpPh-4-H), 6.84–6.87 (d def, 2H, PpPh-2,4-H), 7.13–7.18
(t def., 2H, PpPh-3,4-H); 7.19–7.26 (m, 4H, 2Â Ph-3,5-H); 7.41–
7.47 (m, 6H, 2Â Ph-2,4,6-H).
4.1.1.1.
1-(3-Bromopropyl)-3-methyl-5,5-diphenylimidazoli-
dine-2,4-dione (31). Method A: The reactants were stirred for
72 h. Pure white crystals of product 31 were crystallized with
methanol. Yield 74%; mp 116–117 °C, TLC: Rf (I):0.67. Anal. Calcd
for C19H19BrN2O2: C, 58.93; H, 4.95; N, 7.23. Found: C, 59.01; H,
4.91; N, 7.15. 1H NMR d (ppm): 1.22 (qu, J = 7.44 Hz, 2H, N1-CH2-
CH2), 2.98 (s, 3H, N-CH3), 3.15 (t, J = 6.70 Hz, 2H, N1-CH2), 3.40–
3.45 (t def., 2H, Br-CH2), 7.20–.44 (m, 10 H, 2Â Ph)
White crystals of 12a. Yield 38%, mp 266–268 °C, TLC: Rf(III):
0.71. Anal. Calcd for C29H33N4O3Cl: C, 66.85; H, 6.38; N, 10.75.
Found: C, 66.90; H, 6.36; N, 10.68. 1H NMR for 12a d (ppm):
2.64–2.95 (m, 7H, Pp-CH2, Pp-2,6-H, CHOH), 2.98 (s, 3H, N3-CH3),
3.14–3.45 (m, 4H, Pp-3,5-H), 3.66–3.77 (m, 2H, N1-CH2), 4.22 (br
s, 1H, OH), 6.81 (t, J = 7.20 Hz, 1H, PpPh-4-H), 6.93 (d, J = 8.30 Hz,
2H, PpPh-2,6-H), 7.03–7.05 (m, 2H, 2ÂPh-4-H), 7.21–7.37 (m, 4H,
2ÂPh-2,6-H), 7.44–7.49 (m, 6H, 2ÂPh-3,5-H, PpPh-3,5-H), 9.80
(br s, 1H, NH+). IR: 3242 (OH), 2944, 2895, 2848 (CH), 2541
(NH+); 1772 (C2@O), 1716 (C4@O), 1599 (Ar).
4.1.1.2. 1-(4-Bromobutyl)-3-methyl-5,5-diphenylimidazolidine-
2,4-dione (32). Method A and method B were used. The reactants
were stirred for 64 h (method A). Pure white crystals of product 32
were crystallized with methanol. Yield 57% (method A), 66% (meth-
od B); mp 120–122 °C, TLC: Rf (I):0.62. Anal. Calcd for
C
20H21BrN2O2: C, 59.86; H, 5.27; N, 6.98. Found: C, 59.80; H, 5.30;
N, 7.12. 1H NMR d (ppm): 0.82 (qu, J = 7.60 Hz, 2H, N1-CH2-CH2),
1.38 (qu, J = 6.54 Hz, 2H, Br-CH2-CH2), 2.97 (s, 3H, N-CH3), 3.19 (t,
J = 6.70 Hz, 2H, N1- CH2), 3.30–.35 (m, 2H, Br-CH2), 7.20–7.25 (m,
4H, 2Â Ph-3,5-H), 7.37–7.45 (m, 6H, 2Â Ph-2,4,6-H)
4.1.2.2. 1-[3-(4-(2-Chlorophenyl)-piperazin-1-yl)-2-hydroxypro-
pyl]-3-methyl-2,4-dioxo-5,5-diphenylimidazolidine hydrochlo-
4.1.1.3. Methyl 2-(1-(3-bromopropyl)-2,4-dioxo-5,5-diphenyl-
imidazolidin-3-yl)propionate (33). Method A: The reactants
were stirred for 92 h. Pure white crystals of product 33 were crys-
tallized with i-PrOH. Yield 75%; mp 89–90 °C, TLC: Rf (I):0.64. Anal.
Calcd for C22H23BrN2O4: C, 57.53; H, 5.05; N, 6.10. Found: C, 57.68;
H, 5.01; N, 6.13. 1H NMR d (ppm): 1.18-1.33 (m, 2H, N1-CH2-CH2),
1.47 (d, J = 7.18 Hz, 3H, CHCH3), 3.15 (t, J = 6.70 Hz, 2H, N1-CH2),
3.41–3.46 (t def., 2H, Br-CH2), 3.65 (s, 3H, OCH3), 4.86 (q,
J = 7.18 Hz, 1H, CHCH3), 7.22–7.28 (m, 4H, 2Â Ph-3,5-H), 7.42–
7.51 (m, 6H, 2Â Ph-2,4,6-H)
ride (13a). Method C: A mixture of 26 (1.55 g) and 2-Cl-
phenylpiperazine (0.9 g) was irradiated (450 W) for 6 min
(3 Â 2 min). The irradiation was continued for the next 6 min at
600 W(3 Â 2 min). White crystals of compound 13. Yield 69%, mp
148–149 °C; TLC: Rf(III): 0.74; Anal. Calcd for C29H31ClN4O3: C,
67.11; H, 6.02; N, 10.79. Found: C, 67.42; H, 6.03; N, 10.82. 1H
NMR for 13 d (ppm): 1.95–2.07 (m, 2H, Pp-CH2), 2.22 (br s, 4H,
Pp-2,6-H), 2.80 (br s, 4H, Pp-3,5-H), 2.90 (s, 1H, CHOH), 2.97 (s,
3H, N3-CH3), 3.28–3.35 (t, J = 9.60 Hz, 2H, N1-CH2), 4.43–4.44 (d,
J = 5.00 Hz, 1H, OH), 6.96–7.02 (m, 1H, PpPh-4-H), 7.07–7.11
(m, 2H, 2Â Ph-4-H), 7.22–7.27 (m, 4H, 2Â Ph-2,6-H), 7.32–7.37
(m, 1H, PpPh-6-H), 7.41–7.47 (m, 6H, 2ÂPh-3,5-H, PpPh-3,5-H).
White crystals of 13a. Yield 89%, mp 280–281 °C, TLC: Rf(III):
0.74. Anal. Calcd for C29H32Cl2 N4O3: C, 62.70; H, 5.81; N, 10.09.
Found: C, 62.70; H, 5.83; N, 9.96. 1H NMR for 13a d (ppm):
2.68–2.71 (m, 1H, CHOH), 2.85–3.07 (m, 6H, PpCH2, Pp-2,6-H),
2.99 (s, 3H, N3-CH3), 3.14–3.22 (d def., 2H, N1CH2), 3.26–3.47
(m, 4H, Pp-3,5-H), 5.63 (br s, 1H, OH), 7.11–7.16 (m, 2H, PpPh-
4,6-H), 7.22–7.34 (m, 6H, 2Â Ph-3,5-H, PpPh-3,5-H), 7.42–7.49
(m, 6H, 2Â Ph-2,4,6-H), 9.60 (br s, 1H, NH+). IR: (KBr) (cmÀ1):
3230 (OH), 2984 (CH), 2534 (NH+), 1775 (C2@O), 1718 (C4@O),
1586 (Ar).
4.1.2. General procedure for N1-phenylpiperazinehydroxy
propyl-5,5-diphenylhydantoin derivatives (12a–17a)
Commercially available monohydrochloride salts of N-2- and N-
3-chlorophenylpiperazine as well as N-4-chlorophenylpiperazine
dihydrochloride were converted into free bases according to the
method described earlier.21
Method C: A mixture of oxiran derivative 26 or 27 (5 mmol) and
suitable phenylpiperazine 28–30 (5 mmol) was dissolved in CH2Cl2
(20 ml) in a flat-bottomed 50 ml flask. The solvent was evaporated.
The homogeneous residue covered with a small inverted funnel.
The flask was placed in a standard household microwave oven
and irradiated for several 1–2 min-cycles using various irradiation
powers, according to TLC-controlled progress of the reaction. If
longer heating did not cause any increase of the product spot
(TLC) intensity, the process was interrupted. The reaction mixture
was crystallized from MeOH to afford pure compounds (12–17).
The compounds (12–17) were converted into hydrochlorides
(12a–17a) by saturation with gaseous HCl in dry methanol solution
(50 mg/mL) until acidic pH. The solutions were cooled at 4 °C over-
night for precipitation. The solids of pure hydrochlorides were sep-
arated by filtration.
4.1.2.3. 1-[3-(4-(2-Fluorophenyl)-piperazin-1-yl)-2-hydroxypro-
pyl]-3-methyl-2,4-dioxo-5,5-diphenylimidazolidine hydrochlo-
ride (14a). Method D:
A mixture of 26 (1.55 g) and 2-
fluorophenylpiperazine (0.93 g) was irradiated (450 W) for 3 min.
The irradiation was continued for the next 4 min at 300 W
(2 Â 2 min). White crystals of compound 14. Yield 61%, mp 149–
150 °C; TLC: Rf(III): 0.82; Anal. Calcd for C29H31FN4O3: C, 69.30;
H, 6.22; N, 11.15. Found: C, 69.29; H, 6.20; N, 11.15. 1H NMR for
14 d (ppm): 1.94–2.06 (m, 2H, Pp-CH2), 2.21 (br s, 4H, Pp-2,6-H),
2.83 (br s, 4H, Pp-3,5-H), 2.97 (m, 4H, CHOH, N3-CH3), 3.21–3.37
(m, 2H, N1-CH2); 4.41 (d, J = 4.95 Hz, 1H, OH) 6.93–7.10 (m, 4H,
PpPh-4,6-H, 2ÂPh-4-H), 7.22–7.27 (m, 4H, 2Â Ph-2,6-H), 7.41–
7.47 (m, 6H, 2ÂPh-3,5-H, PpPh-3,5-H).
Method D: A mixture of oxiran derivative 26 or 27 (5 mmol) and
suitable phenylpiperazine 28–30 (5 mmol) was placed in
a
flat-bottomed 50 ml flask immediately under microwave irradia-
tion to melt at first. Then, the procedure went according to the
method C.
White crystals of 14a. Yield 91%, mp 278–279 °C, TLC: Rf(III):
0.82. Anal. Calcd for C29H32ClF N4O3: C, 64.62; H, 5.98; N, 10.39.
Found: C, 64.15; H, 5.93; N, 10.30. 1H NMR for 14a d (ppm):
2.63–2.80 (m, 1H, CHOH), 2.82–3.18 (m, 6H, Pp-CH2, Pp-2,6-H),
2.98 (s, 3H, N3-CH3), 3.27–3.42 (m, 6H, N1-CH2, Pp-3,5-H), 5.83
(br s, 1H, OH), 6.98–7.90 (m, 8H, PpPh-4,6-H, 2ÂPh-2,4,6-H),
4.1.2.1. 1-[2-Hydroxy-3-(4-phenylpiperazin-1-yl)-propyl]-2,4-dioxo-
3-methyl-5,5-diphenylimidazolidine hydrochloride (12a). Method
C: A mixture of 26 (1.55 g) and N-phenylpiperazine (0.81 g) was
irradiated (450 W) for 9 min (3 Â 3 min). The irradiation was con-

J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
2301
7.42–7.48 (m, 6H, PpPh-3,5-H, 2ÂPh-3,5-H), 10.17 (br s, 1H, NH+).
IR: (KBr) (cmÀ1): 3455, 3163 (OH), 2945, 2831 (CH), 2634 (NH+),
1767 (C2@O), 1713 (C4@O), 1599 (Ar).
3.22–3.36 (m, 2H, N1-CH2,), 3.68 (s, 3H, OCH3), 4.34 (s, 2H,
N3-CH2), 4.44 (br s, 1H, OH), 6.88–7.11 (m, 4H, PpPh-4,6-H, 2Â
Ph-4-H), 7.30–7.38 (m, 4H, 2Â Ph-2,6-H), 7.40–7.49 (m, 6H,
2ÂPh-3,5-H, PpPh-3,5-H).
4.1.2.4. (1-(2-Hydroxy-3-(4-phenylpiperazin-1-yl)-propyl)-2,4-
dioxo-5,5-diphenylimidazolidin-3-yl)-acetic acid methyl ester
hydrochloride (15a). Method D: A mixture of 27 (1.9 g) and
N-phenylpiperazine (0.81 g) was irradiated at 450 W for 2 min
The irradiation was continued for the next 6 min at 300 W
(3 Â 2 min). White crystals of compound 15. Yield 44%, mp 132–
134 °C; TLC: Rf(III): 0.79. Anal. Calcd for C31H34N4O5: C, 68.61; H,
6.32; N, 10.32. Found: C, 68.39; H, 6.29; N, 10.11. 1H NMR for 15
d (ppm): 1.98 (d, J = 6.05 Hz, 2H, Pp-CH2), 2.15–2.30 (m, 4H, Pp-
2,6-H), 2.93–2.96 (m, 5H, Pp-3,5-H, CHOH), 3.32 (s, 2H, N1-CH2,),
3.67 (s, 3H, OCH3), 4.34 (s, 2H, N3-CH2), 4.47 (d, J = 5.00 Hz, 1H,
OH), 6.73 (t, J = 7.30 Hz, 1H, PpPh-4-H), 6.86 (d, J = 8.0 Hz, 2H,
PpPh-2,6-H), 7.16 (t, J = 7.90 Hz, 2H, 2Â Ph-4-H), 7.28–7.40 (m,
4H, 2Â Ph-2,6-H), 7.40–7.55 (m, 6H, 2ÂPh-3,5-H, PpPh-3,5-H).
White crystals of 15a. Yield 94%, mp 210–211 °C, TLC: RF(III):
0.79. Anal. Calcd for C31H35ClN4O5: C, 64.30; H, 6.09; N, 9.68.
Found: C, 64.25; H, 6.09; N, 9.33. NMR for 15a d (ppm): 2.76 (br
s, 3H, Pp-CH2, CHOH), 2.98–3.15 (m, 4H, Pp-2,6-H), 3.35–3.43 (m,
4H, Pp-3,5-H), 3.64–3.75 (m, 2H, N1-CH2), 3.66 (s, 3H, OCH3),
4.20–4.23 (d def., 1H, OH), 4.37 (s, 2H, N3-CH2), 6.82–6.87 (t def.,
1H, PpPh-4-H), 6.94–6.97 (d def., 2H, PpPh-2,6-H), 7.21–7.36 (m,
6H, 2ÂPh-2,4,6-H), 7.47–7.53 (m, 6H, PpPh-3,5-H, 2Â Ph-3,5-H),
10.09 (br s, 1H, NH+). IR: (KBr) (cmÀ1): 3388 (OH), 2980, 2950
(CH), 2533 (NH+), 1775 (C2@O), 1749 (C@O ester), 1719 (C4@O),
1597 (Ar)
White crystals of 17a. Yield 89.4%; mp 209–210 °C; TLC: Rf(III):
0.81. Anal. Calcd for C31H34ClFN4O5 Â H2O: C, 60.53; H, 5.89; N,
9.11. Found: C, 60.39; H, 5.84; N, 9.11. 1H NMR for 17a d
(ppm): 2.76–2.86 (m, 4H, Pp-CH2, Pp-2,6-Ha), 3.04–3.14 (m, 5H,
Pp-2,6-He, CHOH, Pp-3,5-Ha) 3.17–3.37 (m, 4H, Pp-3,5-He, N1-
CH2), 3.66 (s, 3H, OCH3), 4.37 (s, 2H, N3-CH2), 5.70 (br s, 1H,
OH), 6.99–7.19 (m, 4H, PpPh-4,6-H, 2ÂPh-4-H), 7.29–7.36 (m,
4H, 2ÂPh-2,6-H), 7.47–7.53 (m, 6H, 2ÂPh-3,5-H, PpPh-3,5-H),
10.09 (br s, 1H, NH+). IR (KBr) (cmÀ1): 3450, 3244 (OH), 2951
(CH), 2586 (NH+), 1776 (C2@O), 1745 (C@O ester), 1722
(C4@O), 1612 (Ar).
4.1.3. General procedure for phenylpiperazine derivatives 18a–
21a
An appropriate phenylpiperazine (5 mmol), K2CO3 (2 g), TEBA
(15 g) were suspended in acetone (20 ml) and stirred under reflux
for 30 min. N-1-bromoalkyl-5,5-diphenylhydantoin derivatives
31–34 (6 mmol) dissolved in acetone (15 ml) were added. The mix-
ture was stirred at room temperature for 72–110 h. The inorganic
precipitate was separated by filtration and washed with acetone.
Combined filtrates were evaporated. The obtained residue was dis-
solved in methylene chloride (15 ml), washed with 1% solution of
HCl (2 Â 15 ml) and water (2 Â 10 ml), dried with Na2SO4
anhydrous.
Method E: After evaporation of the solvent, the pure compounds
(20 and 21) were obtained from the residue by crystallization with
methanol. Compounds (20 and 21), obtained as crystals in basic
forms were converted into hydrochlorides using the method de-
scribed for 12a–17a.
4.1.2.5.
(1-(3-(4-(2-Chlorophenyl)-piperazin-1-yl)-2-hydroxy-
acid
propyl)-2,4-dioxo-5,5-diphenylimidazolidin-3-yl)-acetic
ethyl ester hydrochloride (16a). Method D: A mixture of 27
(1.9 g) and N-2-chlorophenylpiperazine (0.9 g) was irradiated at
450 W for 3 min The irradiation was continued for the next 6 min-
utes at 300 W (3 Â 2 min). White crystals of compound 16. Yield
49%, mp 120–122 °C; TLC: Rf(III): 0.73. Anal. Calcd for C31H33ClN4O5:
C, 64.52; H, 5.76; N, 9.72. Found: C, 64.58; H, 5.72; N, 9.30. 1H NMR
for 16 d (ppm): 2.02 (d def., 2H, Pp-CH2), 2.23 (br s, 4H, Pp-2,6-H),
2.80 (br s, 4H, Pp-3,5-H), 3.01–3.07 (m, 1H, CHOH), 3.21–3.32 (m,
2H, N1-CH2,), 3.68 (s, 3H, OCH3), 4.37 (s, 2H, N3-CH2), 4.45 (d,
J = 4.70 Hz, 1H, OH), 6.96 (t, J = 7.50 Hz, 1H, PpPh-4-H), 7.08 (d,
J = 8.00 Hz, 1H, PpPh-6-H), 7.23–7.37 (m, 6H, 2Â Ph-2,4,6-H),
7.41–7.59 (m, 6H, 2ÂPh-3,5-H, PpPh-3,5-H).
Method F: For compounds (18 and 19) which did not crystallize
with methanol, the alcohol solution was immediately saturated
with gaseous HCl to give a precipitate of compounds 18a and
19a in hydrochloride form.
4.1.3.1. 1-(3-(4-(2-methoxyphenyl)piperazin-1-yl)propyl)-3-methyl-
5,5-diphenylimidazolidine-2,4-dione hydrochloride (18a). Method
F: N-2-methoxyphenylpiperazine (0.96 g) and compound 31 (2.30 g)
were stirred for 90 h. White crystals of compound 18a. Yield 69%,
mp 266–267 °C, TLC: Rf (II): 0.74. Anal. Calcd for C30H35ClN4O3: C,
67.34; H, 6.59; N, 10.47. Found: C, 67.60; H, 6.34; N, 10.27. 1H NMR
d (ppm): 1.2 (qu def., 2H, CH2CH2CH2), 2.76–2.97 (m, 4H, Pp-CH2, Pp-
2,6-Ha), 2.99 (s, 3H, N3-CH3), 3.10–3.12 (m, 2H, Pp-2,6-He), 3.32–3.44
(m, 6H, Pp-3,5-H, N1-CH2), 3.76 (s, 3H, OCH3), 6.87–6.88 (m, 2H,
PpPh-4,6-H), 6.96–7.01 (m, 2H, 2ÂPh-4-H), 7.23–7.69 (m, 4H, 2ÂPh-
2,6-H), 7.43–7.48 (m, 6H, 2ÂPh-3,5-H, Pp-3,5-H), 10.12 (br s, 1H,
NH+). IR (KBr) (cmÀ1): 2950 (CH), 2398 (NH+), 1768 (C2@O), 1710
(C4@O), 1592 (Ar).
White crystals of 16a. Yield 97%, mp 190-192 °C, TLC: Rf(III):
0.73. Anal. Calcd for C31H34Cl2N4O5: C, 60.69; H, 5.59; N, 9.13.
Found: C, 60.45; H, 5.49; N, 9.20. 1H NMR for 16a d (ppm): 2.79
(br s, 3H, Pp-CH2, CHOH), 2.99–3.16 (m, 4H, Pp-2,6-H), 3.22–3.43
(m, 4H, Pp-3,5-H), 3.36 (s, 2H, N1-CH2), 3.66 (s, 3H, OCH3), 4.26
(s., 1H, OH), 4.37 (s, 2H, N3-CH2), 7.06–7.16 (m, 2H, PpPh-4,6-H),
7.29–7.39 (m, 5H, 2ÂPh-3,5-H, PpPh-3-H), 7.41–7.50 (m, 7H,
3ÂPh-2,4,6-H, PpPh-5-H), 10.08 (br s, 1H, NH+). IR (KBr) (cmÀ1):
3455, 3257 (OH), 2952 (CH), 2580 (NH+), 1775 (C2@O), 1747
(C@O ester), 1721 (C4@O), 1620, 1588 (Ar).
4.1.3.2. 1-(4-(4-(2-Methoxyphenyl)piperazin-1-yl)butyl)-3-methyl-
5,5-diphenylimidazolidine-2,4-dione hydrochloride (19a). Method
F: N-2-Methoxyphenylpiperazine (0.96 g) and compound 32 (2 g)
were stirred for 72 h. White crystals of compound 19a. Yield 42%,
mp 130–131 °C, TLC: Rf (II): 0.74. Anal. Calcd for C31H37ClN4O3 Â 0.8
H2O: C, 66.07; H, 6.90; N, 9.96. Found: C, 66.09; H, 6.89; N, 9.97. 1H
NMR d (ppm): 0.76 (qu, J = 7.40 Hz, 2H, Pp-CH2CH2), 1.37 (qu,
J = 7.40 Hz, 2H, N1-CH2CH2), 2.74 (m, 2H, Pp-CH2), 2.90–3.03 (m, 4H,
Pp-2,6-H), 2.98 (s, 3H, N3-CH3), 3.30–3.43 (m, 6H, Pp-3,5-H, N1-
CH2), 3.76 (s, 3H, OCH3), 6.84–7.03 (m, 4H, PpPh-2,6-H, 2ÂPh-4-H),
7.19–7.27 (m, 4H, 2ÂPh-2,6-H), 7.37–7.48 (m, 6H, 2Â Ph-3,5-H,
PpPh-3,5-H), 10.92 (br s, 1H, NH+). IR (KBr) (cmÀ1): 2942 (CH), 2401
(NH+), 1769 (C2@O), 1712 (C4@O), 1595 (Ar)
4.1.2.6. (1-[3-(4-(2-Fluorphenyl)-piperazin-1-yl)-2-hydroxypro-
pyl)-2,4-dioxo-5,5-diphenylimidazolidin-1-yl)-acetic acid ethyl
ester hydrochloride (17a). Method D: A mixture of 27 (1.9 g) and
N-2-fluorophenylpiperazine (0.93 g) was irradiated at 450 W for
4 min The irradiation was continued for the next 6 minutes at
300 W (2 Â 3 min). White crystals of compound 17. Yield 36%,
mp 87–88 °C; TLC: Rf(III): 0.81. Anal. Calcd for C31H33FN4O5: C,
66.41; H, 5.93; N, 10.00. Found: C, 66.92; H, 6.09; N, 10.26. 1H
NMR for 17 d (ppm): 2.00 (d, J = 6.05 Hz, 2H, Pp-CH2), 2.23 (br s,
4H, Pp-2,6-H), 2.84 (br s, 4H, Pp-3,5-H), 2.96–3.08 (m, 1H, CHOH),

2302
J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
4.1.3.3. Methyl 2-(2,4-dioxo-5,5-diphenyl-1-(3-(4-(2-ethoxy-
phenyl)piperazin-1-yl)propyl)imidazolidin-3-yl)propionate
hydrochloride (20a). Method E: N-2-ethoxyphenylpiperazine
(1.03 g) and compound 34 (2.4 g) were stirred for 96 h. White crys-
tals of compound 20. Yield 32%, mp 138–139 °C, TLC: Rf (II): 0.75.
Anal. Calcd for C34H40N4O5: C, 69.84; H, 6.90; N, 9.58. Found: C,
69.64; H, 6.79; N, 9.61. 1H NMR for 20 d (ppm): 1.19–1.25 (qu def.,
2H, CH2CH2CH2), 1.32 (t, J = 6.90 Hz, 3H, OCH2CH3), 1.49 (d,
J = 7.18 Hz, 3H, N3-CHCH3), 1.98 (t, J = 6.80 Hz, 2H, Pp-CH2), 2.26
(br s, 4H, Pp-2,6-H), 2.83 (br s, 4H, Pp-3,5-H), 3.26–3.35 (m, 2H,
N1-CH2,), 3.64 (s, 3H, OCH3), 3.97 (q, J = 7.00 Hz, 2H, OCH2CH3),
4.86 (q, J = 7.22 Hz, 1H, N3-CHCH3), 6.81–6.94 (m, 4H, PpPh-4,6-H,
2Â Ph-4-H), 7.27–7.36 (m, 4H, 2Â Ph-2,6-H), 7.41–7.52 (m, 6H,
2ÂPh-3,5-H, PpPh-3,5-H). White crystals of compound 20a. Yield
51%, mp 149–150 °C, Rf (II): 0.75. Anal. Calcd for C34H41ClN4O5 Â 1.7
H2O: C, 63.29; H, 6.86; N, 8.60. Found: C, 62.69; H, 6.84; N, 8.50. 1H
NMR for 20a d (ppm): 1.21–1.28 (qu def., 2H, CH2CH2CH2), 1.32 (t,
J = 6.92 Hz, 3H, OCH2CH3), 1.48 (d, J = 7.44 Hz, 3H, N3-CHCH3)
2.77–2.96 (m, 4H, Pp-CH2, Pp-2,6-Ha), 3.01–3.29 (m, 4H, Pp-2,6-He,
N1-CH2), 3.38–3.50 (m, 4H, Pp-3,5-H), 3.65 (s, 3H, OCH3), 3.97 (q,
J = 6.92 Hz, 2H, OCH2CH3), 4.87 (q, J = 7.20 Hz, 1H, N3-CHCH3),
6.82–6.99 (m, 4H, Pp-4,6-H, 2ÂPh-4-H), 7.21–7.39 (m, 4H, 2ÂPh-
2,6-H), 7.42–7.52 (m, 6H, 2ÂPh-3,5-H, PpPh-3,5-H), 10.99 (br s,
1H, NH+). IR (KBr) (cmÀ1): 2950 (CH), 2346 (NH+), 1770 (C2@O),
1741 (C@O ester), 1718 (C4@O), 1592 (Ar).
4.2.3. Statistical analysis
The data are expressed as the means SEM. The statistical sig-
nificance was calculated using a one-way ANOVA test. Differences
were considered significant when p <0.05.
4.2.4. Radioligand binding assay
The compounds were evaluated for their affinity for a1-adrener-
gic receptors by determining their ability to displace [3H]prazosin
from its specific binding sites in rat cerebral cortex. [3H]prazosin
(19.5 Ci/mmol) was used.
The tissue was homogenized in 20 vol of ice-cold 50 mM Tris–
HCl buffer (pH 7.6 at 25 °C) and centrifuged at 20000Âg for
20 min. The cell pellet was resuspended in Tris–HCl buffer and cen-
trifuged again. The final pellet was resuspended in Tris–HCl buffer
(10 mg of wet weight/ml). Two hundred and forty
suspension, 30 l of analyzed compound
l of [3H]prazosin and 30
were incubated at 25 °C for 30 min. To determine unspecific bind-
ing 10 M phentolamine was used. Automated pipetting system
ll of the tissue
l
l
l
epMotion 5070 (Eppendorf, Germany) was used for the pipetting
steps.
After incubation reaction mix was filtered immediately onto
presoaked GF/B glass fiber filter mate using 96-well FilterMate
Harvester (PerkinElmer, USA).
The radioactivity retained on the filter was counted in Micro-
Beta TriLux 1450 scintillation counter (PerkinElmer, USA). Non-lin-
ear regression of the normalized (percent radioligand binding com-
pared to that observed in the absence of test or reference
compound - total binding) raw data representing radioligand bind-
ing was performed in GraphPad Prism 3.0 (GraphPad Software)
using the built-in three parameter logistic model describing ligand
competition binding to radioligand-labeled sites.37,38
4.1.3.4. Methyl 2-(2,4-dioxo-5,5-diphenyl-1-(4-(4-(2-methoxy-
phenyl)piperazin-1-yl)butyl)imidazolidin-3-yl)acetate hydro-
chloride (21a). Method E: N-2-methoxyphenylpiperazine (0.96 g)
and compound 33 (2.4 g) were stirred for 96 h. White crystals of
compound 21. Yield 51%, mp 130–131 °C, TLC: Rf (II): 0.58. Anal.
Calcd for 21 C33H38N4O5: C, 69.45; H, 6.71; N, 9.82. Found: C,
69.31; H, 6.70; N, 9.80. 1H NMR d (ppm): 0.78 (qu, J = 7.30 Hz,
2H, Pp-CH2CH2), 1.05 (t, J = 7.05 Hz, 2H, N1-CH2CH2), 1.94 (t,
J = 6.80 Hz, 2H, Pp-CH2), 2.27 (br s, 4H, Pp-2,6-H), 2.81 (br s, 4H,
Pp-3,5-H), 3.34 (m, 2H, N1-CH2), 3.70 (s, 3H, COOCH3), 3.73 (s,
3H, OCH3), 4.34 (s, 2H,N3-CH2), 6.82–6.92 (m, 4H, PpPh-3,4,5,6-
H), 7.27–7.29 (m, 4H, 2ÂPh-3,5-H), 7.39–7.49 (m, 6H, 2Â Ph-
2,4,6-H). White crystals of compound 21a. Yield 81%, mp 178–
180 °C, TLC: Rf (II): 0.58. Anal. Calcd for C33H39ClN4O5: C, 65.28;
H, 6.47; N, 9.23. Found: C, 65.20; H, 6.46; N, 9.19. 1H NMR d
(ppm): 0.79 (br s, 2H, Pp-CH2CH2), 1.39 (br s, 2H, N1-CH2CH2),
2.79 (m, 2H, Pp-CH2), 2.90–3.14 (m, 4H, Pp-2,6-H), 3.32–3.43 (m,
6H, Pp-3,5-H, N1-CH2), 3.71 (s, 3H, OCH3), 3.76 (s, 3H, OCH3),
4.35 (s, 2H, N3CH2), 6.87–7.20 (m, 4H, PpPh), 7.15–7.48 (m, 10H,
2xPh), 10.36 (br s, 1H, NH+). IR (KBr) (cmÀ1): 2948(CH), 2408
(NH+), 1771 (C2@O), 1742 (C@O ester), 1716 (C4@O), 1590 (Ar).
4.2.5. Antiarrhythmic activity
4.2.5.1. Adrenaline-induced arrhythmia according to the Szek-
eres31
.
The arrhythmia was evoked in rats anesthetised with
thiopental (60 mg/kg, ip) by iv injection of adrenaline (20 g/kg).
l
The tested compounds were administered iv 15 min before adren-
aline. The criterion of antiarrhythmic activity was the lack of pre-
mature beats and inhibition of cardiac arrhythmia in comparison
with the control group.
4.2.5.2. Barium chloride-induced arrhythmia according to the
Szekeres31
. Barium chloride solution was injected into the caudal
vein of rat (32 mg/kg, in a volume of 1 ml/kg). The tested com-
pounds were given iv 15 min before the arrhythmogen. The crite-
rion of antiarrhythmic activity was a gradual disappearance of
the arrhythmia and restoration of the sinus rhythm.
4.2. Pharmacology
4.2.6. The effect on the normal electrocardiogram (ECG)
Electrocardiographic investigations were carried out using
Multicard 30 apparatus, standard lead II and paper speed of
50 mm/s. The tested compounds were administered intravenously
(iv) at a dose of 10 mg/kg. The ECG was recorded just before and 1,
5, 15 min following the administration of compounds.
4.2.1. Materials
Compounds. Barium chloride (POCh, Poland), [3H]clonidine
(Amersham), epinephrine (adrenalinum hydrochloricum, Polfa, Po-
land), [3H]prazosin (Amersham), heparin sodium (Polfa, Poland),
thiopental sodium (Biochemie, GmbH, Vienna, Austria).
4.2.2. Animals
4.2.7. Influence on the blood pressure
The experiments were carried out on male Wistar rats (180–
250 g). Animals were housed in constant temperature facilities ex-
posed to 12:12 h light–dark cycle and maintained on a standard
pellet diet and tap water given ad libitum. Control and experimen-
tal groups consisted of 6 animals each. All procedures were per-
formed according to the Animal Care and Use Committee
Guidelines, and approved by the Ethical Committee of Jagiellonian
University, Kraków.
Male Wistar normotensive rats were anesthetized with thio-
pental (50–75 mg/kg) by ip injection. The right carotid artery was
cannulated with polyethylene tubing filled with heparin solution
(in saline) to enable the blood pressure measurements using the
Datamax apparatus (Columbus Instruments). The studied com-
pounds were injected at a dose of 10 mg/kg iv into the caudal vein,
after a 5 min of stabilization period, in a volume equivalent to
1 ml/kg.

J. Handzlik et al. / Bioorg. Med. Chem. 20 (2012) 2290–2303
2303
4.3. hERG K+ Channel blockage calculation
16. Varró, A.; Biliczki, P.; Iost, N.; Virág, L.; Hála, O.; Kovács, P.; Mátyus, P.; Papp, J.
G. Curr. Med. Chem. 2004, 11, 1.
17. Taglialatela, M.; Castaldo, P.; Pannaccione, A.; Giorgio, G.; Annunziato, L.
Biochem. Pharmacol. 1998, 55, 1741.
18. Recanatini, M.; Poluzzi, E.; Masetti, M.; Cavalli, A.; De Ponti, F. Med. Res. Rev.
2005, 25, 133.
19. Raschi, E.; Vasina, V.; Poluzzi, E.; De Ponti, F. Pharmacol. Res. 2008, 57, 181.
20. De Ponti, F.; Poluzzi, E.; Montanaro, N. Eur. J. Clin. Pharmacol. 2000, 56, 1.
21. Bellocq, C.; van Ginneken, A. C.; Bezzina, C. R.; Alders, M.; Escande, D.;
Mannens, M. M. Circulation 2004, 109, 2394.
The 3D structure of each compound was created by Corina on-
line version39 and then read into Maestro.40 After checking on the
types of atoms, the structures were neutralized and optimized
with OPLS-2005 force field by LigPrep.41 To predict the ability of
hERG K+ channel blockage QikProp program42 was applied. Activity
against hERG channels was expressed as logIC50 and referred to the
selected experimental data.43,44
22. Zeng, H.; Lozinskaya, I. M.; Lin, Z.; Willette, R. N.; Brooks, D. P.; Xu, X. J.
Pharmacol. Exp. Ther. 2006, 319, 957.
23. Ciechanowicz-Rutkowska, M.; Stadnicka, K.; Kiec´-Kononowicz, K.; Byrtus, H.;
Filipek, B.; Zygmunt, M.; Macia˛g, D. Arch. Pharm. Med. Chem. 2000, 333, 357.
24. Kiec´-Kononowicz, K.; Stadnicka, K.; Mitka, A.; Pe˛kala, E.; Filipek, B.; Sapa, J.;
Zygmunt, M. Eur. J. Med. Chem. 2003, 38, 555.
Acknowledgment
The work was partly supported by Polish State Committee for
Scientific Research, Grant No.3P05F 03125 and program K/ZDS/
001915.
25. Dyla˛g, T.; Zygmunt, M.; Macia˛g, D.; Handzlik, J.; Bednarski, M.; Filipek, B.; Kiec´-
Kononowicz, K. Eur. J. Med. Chem. 2004, 39, 1013.
26. Bremner, J. H.; Coban, B.; Griffith, R.; Groenewoud, K. M.; Yates, B. F. Bioorg.
Med. Chem. 2000, 8, 201.
27. Barbaro, R.; Betti, L.; Botta, M.; Corelli, F.; Giannaccini, G.; Maccari, L.; Manetti,
F.; Strappaghetti, G.; Corsano, S. J. Med. Chem. 2001, 44, 2118.
28. Handzlik, J.; Macia˛g, D.; Kubacka, M.; Mogilski, S.; Filipek, B.; Stadnicka, K.;
Supplementary data
´
Kiec-Kononowicz, K. Bioorg. Med. Chem. 2008, 16, 5982.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2012.02.009.
29. Handzlik, J.; Pertz, H. H.; Gornemann, T.; Jahnichen, S.; Kiec´-Kononowicz, K.
Bioorg. Med. Chem. 2010, 20, 6152.
30. Litchfield, J. T.; Wilcoxon, F. J. Pharmacol. Exp. Ther. 1949, 96, 99.
31. Szekeres, L.; Papp, J. G. Handbook of Experimental Pharmacology In Schmier, J.,
Eichler, O., Eds.; Springer: Berlin, Heidelberg, New York, 1975; pp 131–182.
32. Shibata, J. J. Pharmacol. Exp. Ther. 1973, 185, 418.
References and notes
33. Thomas, D.; Kiehn, J.; Katus, H. A.; Karle, C. A. Basic Res. Cardiol. 2004, 99, 279.
34. Thomas, D.; Wimmer, A.; Wu, K.; Hammerling, B. C.; Ficker, E. K.; Kuryshev, Y.
A.; Kiehn, J.; Katus, H. A.; Schoels, W.; Karle, C. A. Naunyn-Schmiedeberg’s Arch.
Pharmacol. 2004, 369, 462.
35. Tamargo, J.; Caballero, R.; Gomez, R.; Valenzuela, C.; Delpon, E. Cardiovasc. Res.
2004, 62, 9.
36. Schram, G.; Pourrier, M.; Wang, Z.; White, M.; Nattel, S. Cardiovasc. Res. 2003,
59, 328.
37. Maj, J.; Klimek, V.; Nowak, G. Eur. J. Pharmacol 1985, 119, 113.
38. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22, 3099.
39. http://www.molecular-networks.com/online_demos/corina_demo.
40. Maestro 8.0, Schrodinger LLC, 1999–2007.
41. LigPrep 2.1, Schrodinger LLC, 1999–2007.
42. QikProp 3.0, Schrodinger LLC, 1999–2007.
43. Perry, M.; de Groot, M. J.; Helliwell, R.; Leishman, D.; Tristani-Firouzi, M.;
Sanguinetti, M. C.; Mitcheson, J. Mol. Pharmacol. 2004, 66, 240.
44. Rampe, D.; Murawsky, M. K.; Grau, J.; Lewis, E. W. J. Pharmacol. Exp. Ther. 1998,
286, 788.
1. Singh, B. N.; Vaughan Williams, E. M. Br. J. Pharmacol. 1970, 39, 657.
2. Singh, B. N.; Vaughan Williams, E. M. Br. J. Pharmacol. 1970, 39, 675.
3. Rosen, M. R.; Janse, M. J.; Schwartz, P. J. Cardiovasc. Res. 1992, 26, 568.
4. Singh, B. N. Am. J. Cardiol. 1998, 81, 3D.
5. Karmazyn, M. Exper. Opin. Investig. Drugs 2000, 9, 1099.
6. Pugsley, M. H. Drug Develop. Res. 2002, 55, 3.
7. Kurz, T.; Schneider, I.; Tolg, R.; Richardt, G. Cardiovas. Res. 1999, 42, 48.
8. Amirahmadi, F.; Turnbull, L.; Du, X. J.; Graham, R. M.; Woodcock, E. A. Clin. Sci.
2008, 114, 157.
9. Ravingerova, T.; Pancza, D.; Ziegelhoffer, A. J. Styk. Pysiol. Res. 2002, 51, 109.
10. Imani, A.; Faghihi, M.; Keshavarz, M.; Karimian, S. M.; Niaraki, S. S. Indian
Pacing Electrophysiol. J. 2009, 9, 35.
11. Schomig, A.; Richardt, G.; Kurz, T. Herz 1995, 20, 169.
12. Tölg, R.; Kurz, T.; Ungerer, M.; Schreieck, J.; Görge, B.; Richardt, G. Naunyn-
Schmiedeberg’s Arch. Pharmacol. 1997, 356, 62.
13. Billman, G. E. J. Pharmacol. Exp. Ther. 1994, 269, 409.
14. Hieble, J. P. Pharm. Acta Helv. 2000, 74, 163.
15. Corr, P. B.; Heathers, G. P.; Yamada, K. A. Am. J. Med. 1989, 87, 19S.