Chemically Homogenous Compounds with Antagonistic Properties at All α1-Adrenoceptor Subtypes but not β1-Adrenoceptor Attenuate Adrenaline-Induced Arrhythmia in Rats

Studies proved that among all α1-adrenoceptors, cardiac myocytes functionally express only α1A- and α1B-subtype. Scientists indicated that α1A-subtype blockade might be beneficial in restoring normal heart rhythm. Therefore, we aimed to determine the role of α1-adrenoceptors subtypes (i.e., α1A and α1B) in antiarrhythmic effect of six structurally similar derivatives of 2-methoxyphenylpiperazine. We compared the activity of studied compounds with carvedilol, which is β1- and α1-adrenoceptors blocker with antioxidant properties. To evaluate the affinity for adrenergic receptors, we used radioligand methods. We investigated selectivity at α1-adrenoceptors subtypes using functional bioassays. We tested antiarrhythmic activity in adrenaline-induced (20 μg/kg i.v.), calcium chloride-induced (140 and 25 mg/kg i.v.) and barium chloride-induced (32 and 10 mg/kg i.v.) arrhythmia models in rats. We also evaluated the influence of studied compounds on blood pressure in rats, as well as lipid peroxidation. All studied compounds showed high affinity toward α1-adrenoceptors but no affinity for β1 receptors. Biofunctional studies revealed that the tested compounds blocked α1A-stronger than α1B-adrenoceptors, but except for HBK-19 they antagonized α1A-adrenoceptor weaker than α1D-subtype. HBK-19 showed the greatest difference in pA2 values—it blocked α1A-adrenoceptors around seven-fold stronger than α1B subtype. All compounds showed prophylactic antiarrhythmic properties in adrenaline-induced arrhythmia, but only the activity of HBK-16, HBK-17, HBK-18, and HBK-19 (ED50 = 0.18–0.21) was comparable to that of carvedilol (ED50 = 0.36). All compounds reduced mortality in adrenaline-induced arrhythmia. HBK-16, HBK-17, HBK-18, and HBK-19 showed therapeutic antiarrhythmic properties in adrenaline-induced arrhythmia. None of the compounds showed activity in calcium chloride- or barium chloride-induced arrhythmias. HBK-16, HBK-17, HBK-18, and HBK-19 decreased heart rhythm at ED84. All compounds significantly lowered blood pressure in normotensive rats. HBK-18 showed the strongest hypotensive properties (the lowest active dose: 0.01 mg/kg). HBK-19 was the only compound in the group, which did not show hypotensive effect at antiarrhythmic doses. HBK-16, HBK-17, HBK-18, HBK-19 showed weak antioxidant properties. Our results indicate that the studied 2-methoxyphenylpiperazine derivatives that possessed stronger α1A-adrenolytic properties (i.e., HBK-16, HBK-17, HBK-18, and HBK-19) were the most active compounds in adrenaline-induced arrhythmia. Thus, we suggest that the potent blockade of α1A-receptor subtype is essential to attenuate adrenaline-induced arrhythmia.

Studies proved that among all α 1 -adrenoceptors, cardiac myocytes functionally express only α 1A -and α 1B -subtype. Scientists indicated that α 1A -subtype blockade might be beneficial in restoring normal heart rhythm. Therefore, we aimed to determine the role of α 1 -adrenoceptors subtypes (i.e., α 1A and α 1B ) in antiarrhythmic effect of six structurally similar derivatives of 2-methoxyphenylpiperazine. We compared the activity of studied compounds with carvedilol, which is β 1 -and α 1 -adrenoceptors blocker with antioxidant properties. To evaluate the affinity for adrenergic receptors, we used radioligand methods. We investigated selectivity at α 1 -adrenoceptors subtypes using functional bioassays. We tested antiarrhythmic activity in adrenaline-induced (20 µg/kg i.v.), calcium chloride-induced (140 and 25 mg/kg i.v.) and barium chloride-induced (32 and 10 mg/kg i.v.) arrhythmia models in rats. We also evaluated the influence of studied compounds on blood pressure in rats, as well as lipid peroxidation. All studied compounds showed high affinity toward α 1 -adrenoceptors but no affinity for β 1 receptors.
Biofunctional studies revealed that the tested compounds blocked α 1A -stronger than α 1B -adrenoceptors, but except for HBK-19 they antagonized α 1A -adrenoceptor weaker than α 1D -subtype. HBK-19 showed the greatest difference in pA 2 values-it blocked α 1A -adrenoceptors around seven-fold stronger than α 1B subtype. All compounds showed prophylactic antiarrhythmic properties in adrenaline-induced arrhythmia, but only the activity of  was comparable to that of carvedilol (ED 50 = 0.36). All compounds reduced mortality in adrenaline-induced arrhythmia.  showed therapeutic antiarrhythmic properties in adrenaline-induced arrhythmia. None

INTRODUCTION
Arrhythmias are the most common causes of sudden cardiac death (Deo and Albert, 2012). Despite numerous antiarrhythmic drugs, pharmacotherapy is still ineffective in majority of patients. Moreover, all antiarrhythmic agents acting via different ion channels possess life-threatening proarrhythmic potential. Thus, scientists are still looking for effective and safe compounds, which will protect against arrhythmia and/or restore normal heart rhythm.
Antiarrhytmic activity of pharmacological agents resulting from their receptor-based mechanisms might be equally efficient and much safer than that observed for classical antiarrhythmic drugs. According to many studies, α 1 -adrenolytics may have potential in the treatment of arrhythmias. Scientists agree that the blockade of α 1 -, and particularly α 1A -adrenoceptor may be beneficial in restoring normal heart rhythm (reviewed in Hieble, 2000 andShannon andChaudhry, 2006). The α 1 -adrenoceptor stimulation results in inositol trisphosphate (IP 3 ) production, and subsequent Ca 2+ release from the sarcoplasmatic reticulum (SR; Escobar et al., 2012). Although the regulation of Ca 2+ level in cardiomyocytes mainly depends on ryanodine receptors and SR Ca 2+ pump, the increased basal level of Ca 2+ induced by IP 3 may also alter the electrical excitability of cardiomyocytes, thus contributing to the development of arrhythmia e.g., atrial (Zima and Blatter, 2004) or ventricular fibrillation (Proven et al., 2006). Thereby, the blockade of α 1 -adrenoceptors may lead to the stabilization of Ca 2+ level producing antiarrhythmic effect in arrhythmias induced by catecholamines e.g., catecholaminergic polymorphic ventricular tachycardia. The above hypothesis was supported by Suita et al. (2015), who demonstrated that prazosin not only shortened norepinephrine-induced elongation of atrial fibrillation in mice, but also attenuated norepinephrine-induced SR Ca 2+ leak and spontaneous SR Ca 2+ release in cultured atrium cardiomyocytes. This proves that α 1 -adrenoceptors may have role in preventing cardiac arrhythmias. Numerous animal studies confirmed this theory, showing antiarrhythmic properties of α 1 -adrenolytics (Sapa et al., 2011;Kubacka et al., 2013a;Rapacz et al., 2014Rapacz et al., , 2015b. Since in our earlier experiments 2-methoxyphenylpiperazine derivatives showed high affinity toward α 1 -adrenoceptors , in this study we aimed to determine the role of α 1 -adrenoceptors subtypes (i.e., α 1A , α 1B ) in antiarrhythmic effect of six structurally similar derivatives of 2methoxyphenylpiperazine. We compared the activity of studied compounds with carvedilol, which is β 1 -and α 1 -adrenoceptors blocker with antioxidant properties.

Animals
The experiments were carried out using male normotensive Wistar rats [Krf: (WI) WU], weighing 200-250 g. Animals were kept in plastic cages (three rats per cage) at constant room temperature of 22 ± 2 • C, with 12:12 h light/dark cycle. Rats had free access to food (standard pellet diet) and water. Each experimental and control groups consisted of four to six animals. The animals were killed by cervical dislocation immediately after the experiment. All injections were given in a volume of 1 ml/kg. All experimental procedures were approved by the Local Ethics Committee for Experiments on Animals of the Jagiellonian University in Krakow, Poland (approval numbers 110/2014 and 246/2015) and cared for in accordance with the Guide to the Care and Use of Experimental Animals.

Radioligand Binding Assay
The α 1 -and β 1 -adrenoceptor radioligand binding assay was performed on rat cerebral cortex. [ 3 H]-prazosin (19.5 Ci/mmol, α 1 -adrenoceptor) and [ 3 H]-CGP-12177 (48 Ci/mmol, β 1adrenergic receptor) were used. The brains were homogenized in 20 volumes of an ice-cold 50 mM Tris-HCl buffer (pH 7.6) and were centrifuged at 20,000 × g for 20 min (0−4 • C). The cell pellet was resuspended in the Tris-HCl buffer and centrifuged again. Radioligand binding assays were performed in plates (MultiScreen/Millipore). The final incubation mixture (final volume 300 µl) consisted of 240 µl of the membrane suspension, 30 µl of [ 3 H]-prazosin (0.2 nM), [ 3 H]-CGP-12177 (0.2 nM) solution and 30 µl of the buffer containing seven to eight concentrations (10 −11 −10 −4 M) of the tested compounds. In order to measure the unspecific binding, 10 µM phentolamine (for [ 3 H]-prazosin) and 1 µM of propranolol (for [ 3 H]-CGP-12177) were applied. The incubation was terminated by rapid filtration through glass fiber filters (Whatman GF/C) using a vacuum manifold (Millipore). The filters were then washed twice with the assay buffer and placed in scintillation vials with a liquid scintillation cocktail. Radioactivity was measured in a WALLAC 1409 DSA liquid scintillation counter (Perkin Elmer, USA). All the assays were made in duplicate and the inhibitory constants (K i ) were estimated.
rubbing in the Krebs-Henseleit solution. The isolated artery, cut into ∼4 mm long rings, was then horizontally put up between two stainless steel hooks (diameter 0.15 mm). One hook was fastened to the bottom of the chamber and the other to an isometric FDT10-A force displacement transducer (DMT Model 750TOBS, Denmark), linked with Powerlab 4/26 analyzer (ADInstruments), and processed by LabChart 7 software. The isolated rings were incubated in the Krebs-Henseleit solution (20 ml) at the temperature 37 • C and pH 7.4 with constant oxygenation (O 2 /CO 2 , 19:1). The initial optimal tissues tension was set at 0.75 g. Chloroethylclonidine (3 µM) as the preferential α 1B -adrenoreceptor alkylating agent was used during incubation and after 30 min it was completely washed off. Through 100 min of equilibration tissues were stimulated with noradrenaline (1 µM) four times with washing out until the contractile response become constant. Two cumulative concentrationresponse curves to noradrenaline, at an interval of 60 min, were established on each arterial ring both in the presence and absence of antagonist. The incubation with antagonists went on for 30 min. The experiments were carried out in the constant presence of yohimbine (0.1 µM) and propranolol (1 µM) to block α 2 -and β-adrenoceptors, respectively in order to minimize the involvement of these adrenoceptors in the response to noradrenaline.

Influence on α 1B -Adrenoceptors: Mouse Spleen
The influence on α 1B -adrenoceptors was evaluated using the isolated mouse spleen. The spleen was removed from male mice right after killing the anesthetized animal by cervical dislocation. The isolated tissue was incubated in 20 ml cup filled with the Krebs-Henseleit solution at the temperature 37 • C and pH 7.4 with constant oxygenation (O 2 /CO 2 , 19:1). The initial optimal tissues tension was set at 1.0 g. Through the 100 min of equilibration tissues were stimulated with noradrenaline (0.1−10.0 µM) three times with washing out until the contractile response become constant. Two cumulative concentrationresponse curves to noradrenaline, at an interval of 60 min, were established on each tissue both in the presence and absence of antagonist. The incubation with antagonists went on for 30 min. The experiments were carried out in the constant presence of propranolol (1 µM) to block β-adrenoceptors, and minimize the involvement of these adrenoceptors in the response to noradrenaline.

Influence on α 1D -Adrenoceptors: Rat Aorta
Finally, the influence on α 1D -adrenoreceptors was investigated using the isolated rat aorta. Rats anesthetized with thiopental and killed by cervical dislocation had aorta removed, denuded of endothelium and incubated in the Krebs-Henseleit solution at the temperature 37 • C and pH 7.4 with constant oxygenation (O 2 /CO 2 , 19:1). The aorta rings were maintained at optimal tension of 2.0 g. During 3 h of equilibration tissues were stimulated with noradrenaline three times (0.3 µM). Two cumulative concentration-response curves to noradrenaline, at an interval of 60 min, were established on each tissue both in the presence and absence of antagonist. The incubation with antagonists went on for 30 min. The experiments were carried out in the constant presence of yohimbine (0.1 µM) and propranolol (1 µM) to block α 2 -and β-adrenoceptors, respectively.
Prophylactic Antiarrhythmic Activity in Adrenaline-, Barium Chloride-, and Calcium Chloride-Induced Arrhythmias The procedures were performed according to the method described by Szekeres and Papp (1975). The heart rate disturbances were evoked by the intravenous administration of adrenaline (20 µg/kg), barium chloride (32 and 10 mg/kg) or calcium chloride (140 and 25 mg/kg) solution into the caudal vein in anesthetized rats (thiopental, 75 mg/kg). The studied compounds were administered i.p. 45 min before the injection of adrenaline, calcium chloride, or barium chloride. The electrocardiogram (ECG) was recorded during the first 2 min as well as in the 5th, 10th, and 15th min after the arrhythmogen injection. The criterion of antiarrhythmic activity was the lack of extrasystoles and inhibition of cardiac arrhythmia in comparison with the control group in adrenaline-induced arrhythmia or the progressive disappearance of the arrhythmia and reinstatement of the sinus rhythm in barium chloride-and calcium chlorideinduced arrhythmias. The ED 50 (a dose producing a 50% inhibition of ventricular contractions) with 95% confidence limits was determined by computer log-probit analysis according to Litchfield and Wilcoxon (1949) and Szekeres and Papp (1975). The compounds were administered at the dose 10 mg/kg. We gradually decreased the dose by half until the disappearance of antiarrhythmic activity.

Therapeutic Antiarrhythmic Activity in Adrenaline-Induced Arrhythmia
The experiment was performed according to the method described by Szekeres and Papp (1968). The arrhythmia was evoked by the intravenous administration of adrenaline (20 µg/kg) into the caudal vein in anesthetized rats (thiopental, 75 mg/kg). The tested compounds were injected i.v. at the peak of arrhythmia directly after adrenaline injection, at the ED 84 (a dose producing a 84% inhibition of premature ventricular contractions established in prophylactic adrenaline-induced arrhythmia). The range of doses was 0.325−0.504 mg/kg. The ECG was recorded constantly for 5 min as well as in the 10th and 15th min of the experiment. The criterion of antiarrhythmic activity was the reduction of premature ventricular contractions in comparison with the control group (Sapa et al., 2011).

The Effect on Normal Electrocardiogram in Rats
The experiment was carried out to exclude the influence of tested compounds on normal ECG. The ECG was recorded (ASPEL ASCARD B5 apparatus, standard lead II and paper speed of 50 mm/s) prior and also 5,10,15,20,30,40,50, and 60 min after the i.p. administration of tested compounds. The influence on QRS complex, PQ interval, heart rate (RR), and QTc interval was determined. The QTc was calculated according to the Bazzett's formula: QTc = QT/ √ RR (De Clerck et al., 2002). The compounds were administered at the ED 84 (a dose producing a 84% inhibition of premature ventricular contractions established in prophylactic adrenaline-induced arrhythmia).

Influence on Blood Pressure in Normotensive Rats
Normotensive rats were anesthetized with thiopental (75 mg/kg ip). The right carotid artery was cannulated with a polyethylene tube filled with heparin solution to allow pressure measurements, using a Datamax apparatus (Columbus Instruments, USA; Kubacka et al., 2013b). The tested compounds were administered i.p. after 15 min of stabilization period. The compounds were administered at the dose 10 mg/kg. We gradually decreased the dose by half until the disappearance of antiarrhythmic activity.

Influence on Blood Vasopressor Response in Rats
To verify if the hypotensive activity was a result of α-adrenolytic properties, we studied the influence of tested compounds on the pressor response to methoxamine (150 µg/kg) The experiment was carried out for all active compounds, which were administered (i.v.) to the caudal vein at the lowest hypotensive dose (Kubacka et al., 2013b). Pressor response to methoxamine injected i.v. was measured before (control) and 5 min after the tested compounds. We administered the tested compounds at the lowest possible doses, not to lose selectivity for α 1 -adrenoceptors.

Antioxidant Effect-Lipid Peroxidation in Rat Brain Homogenate
This experiment was performed according to the method described by Yue et al. (1992). The rat brain homogenate containing 10 mg tissue/ml was prepared in 0.9% saline. The rates of membrane lipid peroxidation were measured by the formation of thiobarbituric acid reactive substances (TBARS). Rat brain homogenates (1 ml) were incubated at 37 • C for 5 min with 10 µl of a tested compound or vehicle. Lipid peroxidation was initiated by the addition of 50 µl of 0.5 mM FeCl 2 and 50 µl of 2.0 mM ascorbic acid. After 30 min of incubation, the reaction was stopped by adding 0.1 ml of 0.2% butylated hydroxytoluene (BHT). Thiobarbituric acid reagent was then added and the mixture was heated for 15 min in a boiling water bath. Carvedilol was used as reference compound. The compounds were tested at a concentration of 10 −3 M. The TBARs were measured at 532 nm.

Data Analysis
In radioligand binding studies, the obtained data were fitted to a one-site curve-fitting equation with Prism 6.0 (GraphPad Software), and inhibition constants (K i ) values were estimated from the Cheng−Prusoff equation (Cheng and Prusoff, 1973): In functional bioassays the concentration-response curves were analyzed using GraphPad Prism 6.0 software (GraphPad Software Inc., San Diego, CA, USA) as previously described by Kubacka et al. (2013b). Data are means ± S.E.M. of at least 4 separate experiments. To establish Hill slopes for the agonist concentration-response curves and calculate EC 50 values, curves were fitted to all the data by non-linear regression. The EC 50 value in the presence and absence of antagonists was used to ascertain the concentration ratio (CR). Schild analysis was performed. If the slope was not significantly different from unity, the relative antagonistic potencies (pA 2 : −log of the concentration of an antagonist that doubles the concentration of the agonist needed to elicit the original submaximal response obtained in the absence of antagonist) were determined by plotting the log (CR-1) against the −log of antagonist concentration (Arunlakshana and Schild, 1959).
In case of in vivo experiments the results are presented as the means ± S.E.M. Statistically significant differences between groups were calculated using one-way analysis of variance (ANOVA) with repeated measurements followed by Dunnett's or Bonferroni's test post-hoc or Student's t-test. The criterion for significance was set at p < 0.05.
Antioxidant activity was expressed as the percentage reduction of the sample absorbance during the reaction at wavelength 532 nm.
The log-probit method described by Litchfield and Wilcoxon (1949) was used to determine median effective doses (ED 50doses producing 50% inhibition of premature ventricular contractions) and doses producing 84% of the maximal effect (ED 84 ) for compounds in arrhythmia models.

Affinity for Adrenoceptors
All studied compounds possessed high affinity for α 1adrenoceptors but none of them bound to β 1 -adrenoceptors ( Table 1).
Functional Affinity for α 1A -, α 1B -, and α 1D -adrenoceptors The tested compounds antagonized noradrenaline evoked contraction in isolated rat aorta and tail artery, as well as mouse spleen, and concentration-dependently, shifted the noradrenaline response to the right, without affecting the maximum response. The obtained pA 2 values with Schild slopes not significantly different from unity indicated a competitive antagonism at α 1A -, α 1B -, and α 1D -adrenoceptors (Tables 2A,B).

The Effect on Normal Electrocardiogram in Rats
HBK-14 and HBK-15 administered at the dose 6.154 and 20.218 mg/kg, respectively did not influence the ECG parameters throughout the experiment. HBK-16 at the dose 0.363 mg/kg did not influence PQ interval, QRS complex or QTc interval but it significantly reduced heart rate by 11−13%, since the 40th min of the observation. Similarly, HBK-17 at the dose 0.504 mg/kg significantly decreased heart rate by 9−12%, 20 min after administration, without affecting PQ interval, QRS complex or QTc interval. HBK-18 at the dose 0.325 mg/kg significantly  Cheng and Prusoff (1973). The compounds were tested in three separate experiments in duplicates. n.a., compound did not bind to the receptor at the concentration 10 −5 M. a Pytka et al. (2015). b Pönicke et al. (2002).
reduced the heart rate by 13−23%, since the 15th min of the observation and did not affect PQ interval, QRS complex or QTc interval. HBK-19 at the dose 0.444 mg/kg between the 15th and 30th min after administration reduced heart rate by 21−22% without the influence on PQ interval, QRS complex, or QTc interval.
Prophylactic Antiarrhythmic Activity in Adrenaline-, Barium Chloride-, and Calcium Chloride-Induced Arrhythmias In anesthetized rats i.v. injection of adrenaline (20 µg/kg) caused atrioventricular disturbances, ventricular and supraventricular extrasystoles in 100% of the animals, which led to the death of ∼70% of the animals. The studied compounds administered 45 min (i.p.) prior to adrenaline, decreased the number of extrasystoles and mortality (Figures 2, 3, Table 4). Table 5 shows the ED 50 values (doses producing 50% inhibition of premature ventricular contractions).
The injection (i.v.) of barium chloride (32 mg/kg) caused rapid ventricular extrasystoles in all animals (100%), which led to the death within 3−5 min. Lower dose of barium chloride (10 mg/kg) caused ventricular extrasystoles in around 60% of rats and did not lead to the death of animals. We did not observe a reproducible negative effect on heart rhythm when barium chloride was used at the dose 10 mg/kg. None of the tested compounds administered (i.p.) 45 min before barium chloride were active in barium chloride-induced model of arrhythmia (data not shown).
In anesthetized rats injection (i.v.) of calcium chloride (140 mg/kg) caused rapid ventricular fibrillation in all animals (100%), which led to death within 3−5 min. The intravenous administration of calcium chloride (25 mg/kg) caused ventricular fibrillation in all animals, but did not lead to the death of rats. The functional affinities were determined in the rat tail artery (α 1A -adrenoceptor subtype), mouse spleen (α 1B -adrenoceptor subtype), rat aorta (α 1D -adrenoceptor subtype). Noradrenaline was used as α 1 -adrenergic receptor agonist. The highest pA 2 values for each receptor subtype are in bold font. Antagonist potency of the tested compounds was expressed as pA 2 ± S.E.M. pA 2 was defined as −log of the concentration of an antagonist that doubles the concentration of the agonist necessary to elicit the original submaximal response in the absence of antagonist. pA 2 values were obtained from the linear regression of Schild plot (Arunlakshana and Schild, 1959). Each value was the mean ± S.E.M. of 4-8 experimental results. a Parés-Hipólito et al. (2006). b Eltze (1996). c Jähnichen et al. (2004). d Eltze et al. (1999).
None of the tested compounds administered (i.p.) 45 min prior to calcium chloride were active in the above model of arrhythmia (data not shown).

Therapeutic Antiarrhythmic Activity in Adrenaline-Induced Arrhythmia
All tested compounds administered i.v. at the peak of adrenaline-induced arrhythmia (20 µg/kg) reduced the number of premature ventricular contractions (Figure 4). Table 6 presents the lowest dose of each tested compound, which significantly lowered systolic and diastolic blood pressure in rats. HBK-14 at the dose 0.625 mg/kg, 10 min after injection, significantly reduced systolic blood pressure by 5−10%, and diastolic blood pressure by 9−14% since the 20th min of the observation. HBK-15 at the dose 5.0 mg/kg significantly reduced systolic blood pressure by 13−19%, and diastolic blood pressure by 13−20%, 5 min after administration. HBK-16 at the dose 0.1 mg/kg, 20 min after administration, significantly reduced systolic blood pressure by 8−11% and diastolic blood pressure by 6−10%. HBK-17 at the dose 0.1 mg/kg significantly reduced systolic blood pressure by 8−11% and diastolic blood pressure by 10−16%, 20 min after administration. HBK-18 at the dose 0.01 mg/kg, from the 20th min of the observation, significantly reduced systolic and diastolic blood pressure by 23−28 and 16−18%, respectively. HBK-19 at the dose 0.625 mg/kg since the 5th min after i.p. administration, significantly reduced systolic blood pressure by 5−17%, whereas diastolic blood pressure by 9−15% since the 10th min of the observation.

Influence on Blood Vasopressor Response in Rats
In the control group the increase of blood pressure after methoxamine (150 µg/kg) was ranging from 62.7 ± 10.4 to 94.2 ± 2.7 mmHg. Figure 5 shows that all tested compounds at the lowest hypotensive doses, significantly antagonized the pressor response to methoxamine.

DISCUSSION
We found that the studied 2-methoxyphenylpiperazine derivatives that possessed stronger α 1A -adrenolytic properties (i.e., HBK-16, HBK-17, HBK-18, and HBK-19) were the most active compounds in adrenaline-induced arrhythmia. Their antiarrhythmic (but not antioxidant) activity was comparable to that of carvedilol. The tested compounds showed hypotensive
Scientists indicated that agents which block α 1A -adrenoceptors may have antiarrhythmic potential (Hieble, 2000). Harrison et al. (1998) showed that hearts from transgenic rats expressing constitutively active α 1B -adrenoceptors, and having 50% reduced α 1A -mRNA levels, were less sensitive to ischemia-induced ventricular tachycardia than normal rats. Lee and Rosen (1993) proved that the blockade of α 1B receptors by chlorethylclonidine increased the amplitude of delayed afterdepolarizations induced by calcium and phenylephrine. Altogether, these findings suggest that agents, which block α 1A -adrenoceptors stronger than α 1B subtype, may have antiarrhythmic potential. Therefore, we determined the selectivity of studied compounds at α 1 -adrenoceptor subtypes. Biofunctional assays revealed that all compounds competitively blocked α 1A , α 1B , and α 1D subtypes. HBK-19 and HBK-16 were the strongest α 1A -adrenoceptor antagonists, while HBK-14 and HBK-15 were the weakest. Although we did not find highly selective compounds, all 2-methoxyphenylpiperazine derivatives antagonized α 1A -adrenoceptors stronger than α 1B subtype. HBK-19 showed the greatest difference in pA 2 values-it blocked α 1A -adrenoceptors around seven-fold stronger than α 1B subtype. Since all studied compounds blocked α 1A -and α 1Badrenoceptors, we decided to examine their antiarrhythmic activity. We also investigated whether antiarrhythmic activity depended on the strength of α 1A -adrenoceptor blockade or the differences between pA 2 values for α 1A -and α 1B -adrenoceptors.
To determine antiarrhythmic effect, we used three models of arrhythmia i.e., adrenaline-, barium chloride-, and calcium chloride-induced.
All compounds showed antiarrhythmic activity in adrenalineinduced model of arrhythmia, and reduced mortality of rats. HBK-18 possessed the strongest prophylactic antiarrhythmic properties, but ED 50 values for HBK-16, HBK-17, and HBK-19 were also very low. Except for HBK-14 and HBK-15, prophylactic antiarrhythmic activities of compounds in adrenaline-induced arrhythmia were comparable to that of carvedilol (reference drug). We think that the weak antiarrhythmic activity of HBK 14 and HBK 15 may be due to their weaker α 1 -adrenolytic properties (see Table 2), which are crucial for antiarrhythmic effect in the applied model of arrhythmia. Since the compounds used in the   Rats were anesthetized with intraperitoneal (i.p.) injection of thiopental (75 mg/kg). The tested compounds were administered (i.p.) 45 min before the experiment. The observation was carried out for 15 min after the intravenous injection of adrenaline (20 µg/kg) i.e., during the first 2 min, in the 5, 10, and 15th min. n = 5−6 animals per group. Rats were anesthetized with intraperitoneal (i.p.) injection of thiopental (75 mg/kg). The tested compounds were administered (i.p.) 45 min before the experiment. The observation was carried out for 15 min after the intravenous injection of adrenaline (20 µg/kg) i.e., during the first 2 min, in the 5, 10, and 15th min. The ED 50 values with confidence limits were calculated according to the methods described by Litchfield and Wilcoxon (1949). Each value was obtained from three experimental groups. n = 5−6 animals per group.
experiments did not present potent selectivity toward different subtypes of α 1− adrenoceptors, we cannot definitely conclude which receptor subtype should be primarily blocked to achieve antiarrhythmic effect. Although, HBK-19 showed the greatest difference in pA 2 values for the α 1A -and α 1B receptor subtypes, it did not possess the strongest antiarrhythmic properties. Similarly, according to the studies performed by Koshimizu et al. (2004), the pA 2 value for α 1A− adrenoceptor subtype for carvedilol was 9.0, whereas for α 1B− adrenoceptor was 10.0. Carvedilol showed comparable properties in adrenaline-induced arrhythmia model. Thus, we claim that the potent blockade of α 1A -receptor subtype is essential to attenuate adrenaline-induced arrhythmia, but the role of α 1B -adrenoceptor blockade needs further studies. Carvedilol is a potent β 1 -and α 1 -adrenoceptors blocker with antioxidant activity. Surprisingly, despite the fact that carvedilol blocked both β 1 -and α 1 -adrenoceptors, and the studied compounds only α 1 -adrenoceptors, their antiarrhythmic effect was comparable. Therefore, we suggest that this may indicate more important role of α 1 -than β 1 -adrenoceptors blockade in adrenaline-induced arrhythmia model. Arrhythmia models induced by calcium or sodium chloride are associated with the changes in intracellular ion concentration. These changes are ion channel dependent, and their dynamics and amplitude are high. In arrhythmias induced by adrenaline, the stimulation of adrenergic receptors also leads to ion level changes (primarily Ca 2+ ), but these changes are not as rapid. Their amplitude and dynamics are much lower than the above. None of the compounds showed activity in barium chloride-or calcium chloride-induced arrhythmias. Therefore, we can assume that the blockade of sodium or calcium channels was not the predominant mechanism of antiarrhythmic effect of the studied compounds.
We decided to test therapeutic antiarrhythmic potential of the most active compounds (i.e., HBK-16, HBK-17, HBK-18, and HBK-19) in adrenaline-induced model of arrhythmia. The studied compounds restored normal heart rhythm administered at the peak of arrhythmia, but the strongest therapeutic antiarrhythmic activity showed HBK-18. The results of this experiments correlate with the results obtained in prophylactic adrenaline-induced arrhythmia model. Antiarrhythmic agents have proarrhythmic potential, thus we evaluated the influence of studied compounds on normal ECG in rats. Only HBK-14 and HBK-15 did not influence ECG at ED 84 obtained in prophylactic adrenaline-induced arrhythmia model. The rest of compounds significantly decreased heart rate. Williamson et al. (1994) proved that stimulation of α 1Aadrenoceptors resulted in positive chronotropic effect. This suggests that the decrease in heart rate observed after treatment with HBK-16, HBK-17, HBK-18, and HBK-19 was a result of α 1A receptor blockade.
The QT interval represents electrical depolarization and repolarization of ventricles. The prolongation of QT interval indicates the potential of a drug to cause ventricular tachyarrhythmias like torsades de pointes. The QTc adjusts the QT interval for heart rate extremes. In this study we used Bazett's equation to calculate QTc. Nevertheless, we need to point out that even though Bazett's formula is very often used for QT correction, it has many limitations (e.g., over-and under-correction of high or low heart rhythms). Rodents' heart rate values can be several times higher than those observed in humans (Kmecova and Klimas, 2010). Since heart rhythm significantly influences QTc, this might be the reason for the observed differences in baseline QTc in our experiments. We showed that none of the compounds affected QTc at ED 84 , therefore we can assume that they did not have proarrhythmic potential at antiarrhythmic doses.
Our findings are in agreement with the results obtained by other researchers, showing that phenylpiperazine derivatives possessed α 1 -adrenolytic properties, as well as prophylactic and/or therapeutic activity in adrenaline-induced arrhythmia (Dylag et al., 2004;Handzlik et al., 2012;Kubacka et al., 2013a,b).
When discussing adrenaline-induced arrhythmias we could neglect the role of α 1D -adrenoceptors, since their blockade should not have a direct influence on cardiac myocytes. Nevertheless, in animal studies on drug candidates, we cannot entirely ignore the role of α 1D -subtype. α 1D -Adrenoceptors blockade in blood vessels might significantly lower blood pressure, which due to the baroreflex might increase heart rate, and contribute to arrhythmia.
Since all studied compounds blocked α 1D -adrenoceptor, and these receptors among others regulate blood pressure, we evaluated their influence on blood pressure in rats. All tested compounds showed hypotensive properties. HBK-18 showed the strongest hypotensive activity, while HBK-15 the weakest. Interestingly, the results of this experiment did not correlate with the functional bioassays, where the strongest α 1D -adrenoceptor blocking properties showed HBK-16. We suspect that this may be due to the differences in receptor binding dynamics, but this issue would require further experiments. Regarding the case of HBK-19, the lowest dose that reduced blood pressure was around three-fold higher than ED 50 value in prophylactic adrenalineinduced arrhythmia. For antiarrhythmic drugs, hypotensive activity is not desirable, since α 1 -adrenoceptor blockers acting in the periphery, may induce reflex tachycardia, and contribute FIGURE 5 | The effect of tested compounds on the blood pressure response to methoxamine. Rats were anesthetized with intraperitoneal injection of thiopental (75 mg/kg). The studied compounds were administered intravenously (i.v.), at the lowest hypotensive doses i.e., 0.625 mg/kg (HBK-14), 5.0 mg/kg (HBK-15), 0.1 mg/kg (HBK-16), 0.1 mg/kg (HBK-17), 0.01 mg/kg (HBK-18), 0.625 mg/kg . Methoxamine (MET) was administered at the dose 150 µg/kg (i.v.). All values represent means ± S.E.M. Statistical analysis: Student's t-test; *p < 0.05, **p < 0.01, compared to the initial maximal response (obtained before the administration of tested compounds). n = 5−6 animals per group. The rates of membrane lipid peroxidation were measured using rat brain homogenate by the formation of thiobarbituric acid reactive substances (TBARs). The studied compounds were tested at a concentration of 10 −3 M. The TBARs were measured at 532 nm.
to cardiac arrhythmias. In our opinion, the lack of hypotensive properties at antiarrhythmic doses makes HBK-19 the most interesting compound in the studied group. Our results suggest that the receptor profile of HBK 19 (the highest affinity for α 1A and the lowest for α 1D ) is the most beneficial in preventing adrenaline-induced arrhythmia. Interestingly, this suggests that in in vivo conditions the selectivity between α 1A and α 1D is much more important in achieving the optimal profile of α 1adrenolytics acting as antiarrhythmic agents, than the selectivity between α 1A and α 1B .
In order to prove that hypotensive properties of tested compounds were a result of their α 1 -adrenolytic properties, we performed the experiments with methoxamine (selective α 1 -adrenoceptor agonist). Drugs that selectively block α 1adrenergic receptors significantly inhibit pressor response to methoxamine. All studied compounds blocked the effect caused by methoxamine, thus we concluded that their hypotensive activity was due to α 1 -adrenolytic properties.
Oxidative stress plays an important role in arrhythmias (Dudek et al., 2014;Sovari, 2016). Reactive oxygen species (ROS) prolong action potential duration, induce early afterdepolarizations, and delayed afterdepolarizations in rats and guinea-pigs (Beresewicz and Horackova, 1991). Scientists indicated that oxidative stress activated Ca2+/CaM-dependent kinase II (CaMKII), and consequently caused arrhythmias (Xie et al., 2009). Rabbits with cardiac hypertrophy pretreated with CaMKII inhibitor were less likely to develop ventricular arrhythmias (Ke et al., 2007). Kirshenbaum et al. (1990) showed that vitamin E (antioxidant) protected rats with chronic heart hypertrophy against adrenaline-induced arrhythmias. This suggests that oxidative stress plays role in adrenaline-induced arrhythmias.
Given the significant antiarrhythmic effect of the studied compounds, we decided to investigate whether 2methoxyphenylpiperazine derivatives possess additional antioxidant activity. Strong antioxidant activity might have contributed to their antiarrhythmic effect. This would explain their significant effect in adrenaline-induced model of arrhythmia. Our experiments showed that among all studied compounds only HBK-16, HBK-17, HBK-18, and HBK-19 weakly inhibited lipid peroxidation. The effect elicited by HBK-19 was the strongest in the group. However, its activity was around eight-fold weaker than the effect caused by carvedilol. Although HBK-16, HBK-17, HBK-18, and HBK-19 possessed weaker antioxidant properties than carvedilol, they showed stronger antiarrhythmic activity. This confirms that in adrenaline-induced arrhythmia model, the blockade of α 1 -adrenoceptors is more important for antiarrhythmic activity than antioxidant properties of the compound. Moreover, our findings suggest that antiarrhythmic properties of studied compounds resulted predominantly from α 1 -adrenolytic properties.
The levels of cardiac α 1 -adrenoceptor are around 10-fold higher in rats than in humans. This may suggest that the role of α 1 -adrenoceptor blockade in arrhythmia is not as significant in humans. Interestingly, despite lower expression of α 1 -adrenoceptors in human heart, scientists proved that they play a significant role in arrhythmias (Furushima et al., 2001). Kurtzwald-Josefson et al. (2014) identified a contribution of α-adrenergic pathway to pathogenesis of catecholamineinduced arrhythmia, and suggested α-blockade as an effective therapy in the murine model of catecholaminergic polymorphic ventricular tachycardia. The Authors suggested α-adrenolytics as an alternative treatment in humans resistant to β-blockers. Thus, it would be reasonable to keep searching for antiarrhythmic agents among α 1 -adrenolytics.
Since structural similarity of studied compounds reduces the likelihood of various mechanisms of action, in future studies we plan to investigate another set of structurally similar 2-methoxyphenylpiperazine derivatives with higher selectivity toward α 1A -adrenoceptor subtype. This might give more insight into the role of α 1A -adrenoceptor subtype in antiarrhythmic effect.