Original article

A. RADWANSKA1, J. DLUGOKECKA1, R. WASILEWSKI2, R. KALISZAN1


TESTING CONCEPTION OF ENGAGEMENT OF IMIDAZOLINE RECEPTORS
IN IMIDAZOLINE DRUGS EFFECTS ON ISOLATED RAT HEART ATRIA



1Department of Biopharmaceutics and Phamacodynamics, Medical University of Gdansk, Gdansk, Poland;
2Department of Pharmacology, Medical University of Gdansk, Gdansk, Poland


  Recently, attention has been payed to the role of imidazolines in physiology of the heart. However, no systematic comparative studies were reported regarding the activity of a representative set of specific ligands towards imidazoline receptors in the heart preparations. The aim of this project was to test effects of a set of ligands on the pharmacological function of putative imidazoline receptors in isolated rat heart atria. Known imidazoline drugs with a postulated high affinity to imidazoline I1 receptor: AGN192403, rilmenidine, moxonidine and clonidine were used. The specific ligands of imidazoline I2 receptor: 2-BFI, BU239 and putative natural ligand for imidazoline I1, I2 and I3 receptors, agmatine, were tested also. The spontaneously beating right and left atria, driven electrically, were studied. Dose-response curves for amplitude and rate of the contractions of the atria were produced by administration of increasing doses of the agents. Phentolamine as 1/2 adrenergic receptors blocker and idazoxan as I2/I1/2 receptors blocker were added in order to inhibit ino- and chronotropic effects of the compounds studied. The -log EC50 parameters were calculated. The positive inotropic effect on left atria were evoked with the rank order of potency: agmatine >> clonidine > BU239 > rilmenidine moxonidine and these effects were generally diminished by idazoxan. Moxonidine produced a weak positive inotropic effect potentiated by idazoxan. Rilmenidine and moxonidine were assumed to act as partial agonists of imidazoline I1 receptor. AGN192403 did not change the amplitude of beating of left atria. The positive chronotropic effects on spontaneously beating right heart atria were with in the following order of potency: BU239 agmatine >>> clonidine > AGN192403. Idazoxan markedly antagonized chronotropic effect of both BU239 and agmatine. 2-BFI weakly diminished the rate of beating of atria; moxonidine and rilmenidine had no effect. In conclusion, imidazoline receptors of the I1 subtype may be involved in inotropic reaction of the agents studied, but this effect depends mainly on the 2/1 adrenergic receptors. Engagement of I2 imidazoline receptors, along with the 2 adrenergic ones, in chronotropic activity of isolated right atria of rat has been demonstrated.

Key words: imidazoline receptors, rat heart atria, chronotropic and inotropic effect



INTRODUCTION

Various imidazolines have cholinomimetic, sympathomimetic, histaminelike, antihistamine and adrenergic properties. Second generation central antihypertensives, such as moxonidine and rilmenidine, have attenuated sedative (2 adrenoceptor-mediated) side effects at equihypotensive doses compared with the drugs of first generation (1). While recognized as potent agonists at peripheral 2 adrenergic receptors, several studies suggested that the drugs also had another mode of action. Clonidine-like drugs, such as moxonidine, rilmenidine and dexmedetomidine, reduce blood pressure by acting centrally at both 2 adrenergic and imidazoline receptors. Imidazoline-binding sites (IBS) were discovered in central nervous system (2) as well as in peripheral tissues from various species (3).

Imidazoline receptors engaged in circulatory system are classified in two groups: the I1 type, sensitive to clonidine and idazoxan, an antagonist with an imidazoline structure, and the I2 type, displaying a high affinity for idazoxan (4), guanabenz (5, 6) cirazoline (7), and a medium-to-low affinity for clonidine (4).

Imidazoline I1 receptors are reported to play a role in the central regulation of blood pressure (8). The selective activation of central I1imidazoline receptors results in an inhibition of peripheral sympathetic activity and produces arterial vasodilatation (9, 10). Imidazoline agents evoke diverse pharmacological responses in both peripheral tissues and the central nervous system, so that are difficult to attribute to known receptor signaling system (11).

The presence of presynaptic imidazoline receptors has been suggested in the human and rat heart but their functional role is unknown (12). Imidazoline I1 receptors over 2 adrenoceptors in the heart atria and ventricles have been identified with the affinity to imidazolines at nanomolar range (13). An 85 kD protein may correspond to the functional I1 receptor in atria. It has been shown that atrial I1 receptors are up-regulated in spontaneously hypertensive rats (SHR) (14, 15). El-Ayoubi et al. (14) observed that I1 receptors are increased in hypertensive rats or humans. Therefore, there are remisses to suppose the engagement of imidazoline receptors in ino- and chronotropism of isolated heart atria (16).

Functional cardiac imidazoline I1 receptors are tissue-specific, being differentially regulated in atria and ventricles in hypertension by chronic exposure to agonists (17).

In contrast to the I1 imidazoline receptors, physiological role for I2 sites has not yet been determined but it has been proposed that they play an active role in various physiological processes (18). For example, antiproliferative action of the imidazolines correlated with their affinity to the I2 imidazoline binding sites in blood vessels (2, 19). The imidazoline I2 binding sites (IBS) were described as imidazoline-guanidinium receptive site (IGRS) with idazoxan binding selectivity (20, 21). IBS appear to be heterogeneous in nature (22). Up today, their molecular structure, functional significance and their second-messenger system are unknown. I2 receptor ligands interact with a domain on MAO, a catecholamine metabolizing enzyme, but this mechanism is not equally accessible in all tissues (11, 23, 24).

Clonidine is twice as potent as moxonidine at the I1 receptor but has similar affinity for 2 and 1 adreceptors (25). In binding assay on cow brain the Ki values for clonidine at 2A and I1 receptors were 3.8 and 1.0 nM, respectively (26). Hypotensive effect of clonidine is mainly through the 2 adrenergic over I1 imidazoline receptors (27). Significant bradycardia of isolated rat heart was observed with clonidine, and less with moxonidine, at 10-6 M concentration. It suggests that postsynaptic cardiac imidazoline I1 receptors may be involved in these effects (28).

Moxonidine is pharmacologically similar to clonidine, but its affinity to imidazoline I1 receptors over 2 adrenoceptors is 100-fold higher (29). The hypotensive mechanism of moxonidine originally suggested was through activation of central 2 adrenoceptors, but it appeared that primary action in hypertension is due to binding of moxonidine at imidazoline I1 receptors in the rostroventrolateral part of the brainstem (RVLM) (30). Moxonidine is three times more selective for the I1 receptor in RVLM than rilmenidine and has 40-70 times greater affinity for I1 receptors than for 2 adrenoceptors (25). Moxonidine and rilmenidine injected intravenously lowered blood pressure, decreased plasma norepinephrine concentrations and inhibited stimulation-evoked cardioacceleration in pithed rabbits via 2 adrenoceptor mechanism (31). Inhibition of norepinephrine release by moxonidine in pithed SHR was demonstrated by Raasch et al. (32): the authors explain this effect by interaction with imidazoline I1 receptor. Intravenous moxonidine may activate imidazoline I1 receptors and 2 adrenoceptors present in the rat heart. Compared with clonidine selectivity, moxonidine and rilmenidine has approximately 3 to 10 times greater affinity for imidazoline receptors (14, 17).

Rilmenidine exhibited antiarrhythmic effects in different animal models of arrhythmia (33). This effect is likely to originate from effects on the central nervous system as well as from an action at peripheral sites (34).

Rilmenidine is neutral regarding metabolic parameters, but it influences left ventricular hypertrophy, microalbuminuria and insulin resistance positively in hypertensives at risk (35). Widimsky and Sirotiakova (36) observed decrease of arterial pressure and reduction in heart rate in hypertensive patients treated with 1mg rilmenidine daily. That may be a clinically relevant benefit in patients with an increased cardiovascular risk and metabolic disorders.

Agmatine (decarboxylated arginine) is widely distributed through the body, attaining high levels in the rat aorta (57.41 ng/g) and in the rat heart (6.03 ng/g), with the concentration in brain being relatively small (2.4 ng/g) (37). Molecular mass of agmatine is 130 Da (24). The agent was found to be regionally distributed in the rat brain (38). Agmatine is relased from the neurons and interacts with various pre- and postsynaptic receptors including the I1 imidazoline receptor, 2 adrenoceptor, NMDA receptor, nicotinic cholinergic receptor and 5-HT3. receptor which might be of physiological importance (39, 40). All that suggests that agmatine may be a neurotransmitter (41). Their affinity (Ki) in human brain for 2A and I1 imidazoline receptors is 46 980 and 33.4 nM, respectively (26), although the interaction of agmatine with 2 adrenoceptors is unclear (42, 43). It has been postulated that agmatine may change uptake of norepinephrine, like clonidine does, thus reducing sympathetic tone through imidazoline receptors (42).

Agmatine concentration-dependently releases adrenaline and noradrenaline from chromaffin cells. This effect can be blocked by antagonists of the I1 receptor. In chromaffin cells agmatine has a high affinity to 2 adrenoceptors and I1 and I2 binding sites, namely 4.0, 0.7 and 1.0 µM, respectively. It has a low affinity to the 1 and ß adrenoceptors as well as to the 5-HT3 serotonin and D2 dopamine binding sites. Agmatine-uptake into synaptosomes is blocked by the imidazoline derivative idazoxan and by phentolamine, but not by clonidine, moxonidine and rilmenidine (41).

Agmatine has no effect on vascular contraction and blood pressure, contrary to moxonidine, which increases the vascular contraction and decreases blood pressure (39). However, Herman (44) suggested that it can regulate cardiovascular functions and modulate some processes in the peripheral and central nervous system, whereas it has only a weak blood pressure effect when applied within the RVLM area. The affinity of agmatine for IBS is rather low, while it also binds 2 adrenoceptors (41). In the experiments on rat hearts, agmatine increased norepinephrine level indicating a synergistic inhibitory action at I1 imidazoline receptors under conditions of stimulated 2 adrenergic autoinhibition (15).

Research on newly synthesized imidazoline compounds with a high affinity to imidazoline receptors is still carried on. It is believed that the binding sites specifically recognizing the imidazoline structure or similar chemical structures, both in the brain and in certain peripheral tissues, including the heart, participate in the control of blood pressure (14).

Imidazoline receptors bind some agents that are not imidazolines, such as guanabenz (45), guanidinium, rilmenidine (46) and oxazole. Newly synthesized analogues of imidazolines and reference oxazolines elicit an appreciably increased selectivity for I1 and I2 receptors. A few of these compounds, namely "AGN" and "BU families", show prevailing affinities for I1 and I2 receptors. These compounds are the first imidazoline analogues described that are devoid of any significant affinity to 2 adrenergic receptors.

The isofuran derivative of imidazoline, AGN 192403, is about 10,000 time more selective for imidazoline receptors than for 2 adrenoceptors (47). It had no effect on blood pressure when injected intravenously in monkeys and rabbits. AGN was found to be an antagonists of imidazoline receptor in a number of studies (15, 48).

The benzofuran derivative, 2-BFI, appeared 1,000 to 10,000 times more selective for [3H]idazoxan imidazoline specific binding sites than for 2 adrenoceptors. Selective for the I2 imidazoline binding sites agent 2-BFI has Ki of about 1 nM and a low activity for 2 adrenoceptors. It can elevate extracellular levels of norepinephrine in the frontal cortex and hippocampus in rats (49).

BU239 was described as a ligand with a high selectivity to I2 receptors in rabbit brain. In competitive binding assays on rat kidney membranes Ki for BU239 was 4.3 nM (16, 50).

Distinction between the imidazoline receptor and the 2 adrenoceptor-mediated mechanism for imidazoline compounds is difficult. A combination of agonists and antagonists differing in affinities for each receptor is required for that (26). The most frequently used imidazoline receptor antagonist, idazoxan, is also a potent 2 adrenoceptor antagonist. It seems that important for affinity of agents towards I1 receptors is their hydrophobicity, as expressed by log P ranging from 1 to 2 (51). Therefore, structural alterations of idazoxan can result in molecules with a marked selectivity for either 2 adrenoceptors or imidazoline receptors.

Beside phentolamine, potent nonselective 1 /2 adrenoceptor antagonists with a low affinity at histaminergic receptors and less potent at imidazoline receptors are used. For now, there are no endogenous agonists selective for imidazoline receptors. All drugs binding to imidazoline receptors bind to 2 adrenergic receptors as well. It has not been possible to determine unambiguously whether drugs binding to I receptors act as agonists or antagonists and have actions comparable to their actions at the 2 adrenergic receptors.

Imidazoline derivatives have been demonstrated to interact with the sympathetic neurotransmission via nonadrenergic presynaptic receptors in different experimental models. However, no systematic comparative studies were reported regarding the ino- and chronotropic activity of a representative set of imidazoline drugs towards imidazoline receptors in the heart preparations. On the other hand, single individual imidazolines have been reported to elicit pronounced pharmacological effects mediated through cardiac receptors. The direct effect of an imidazoline receptor ligand on cardiac receptors has not been established. Still, new imidazoline derivatives are potential drugs for antihypertensive therapy.

The aim of this study was to determine in vitro the inotropic and chronotropic effect of newly synthesized imidazoline receptor ligands: 2-BFI, BU239 and AGN192403 as well as the known imidazoline drugs, like clonidine, rilmenidine, moxonidine and agmatine on isolated rat atria. The agents studied were selected from the point of view of their hypothetical interaction with the adrenergic/imidazoline receptors in cardiac cells. The main task of this project was to help to direct rationally the further search for original circulatory and antihypertensive imidazoline agents based on their cardiotropic properties.


MATERIALS AND METHODS

The procedure applied was designed in accordance with the respective Polish and European regulations and the guidelines established by the Ethics Committee for Animal Experiments of the Medical University of Gdansk, Poland.

Materials

Male Wistar rats (200-350 g) were used for in vitro studies. The animals were housed and fed in a laboratory kept at constant temperature of 22°C under the standard conditions (12:12 h L:D cycle, standard pellet diet, tap water).

Methods

The animals were anesthetized with urethane (1.5 g/kg i.p.) and the heart was rapidly excised. After cervical dislocation, thorax was quickly opened, the still beating heart removed and placed in the preparation dish with a modified Krebs-Henseleit solution. The ventricular tissue was cut away as far as possible. The atria, left and right separately, were placed in the solution, gassed with 95 % O2 and 5 % CO2, giving pH of 7.3-7.4, and kept at 35.5-37°C. The incubation medium contained NaCl 118 mmol, KCl 4.7 mmol, CaCl2 6 mmol, NaH2PO4 1 mmol, MgCl2 1.2 mmol, NaHCO3 25 mmol, glucose 11.1 mmol, EDTA 0.04 mmol, and ascorbic acid 0.1 mmol.

The muscle was tied at either end to stainless hooks under a tension of 0.5-1.0 g in an organ bath and was allowed to stabilize for 45-60 min. The left atrium was electrically stimulated with two platinum electrodes by square-wave electrical pulses (2.5 Hz, 4 ms) and voltage 10 V. Amplitude of contractile tension (mm) of the left atrium and the rate of contractile action (min-1) of the spontaneously beating right atrium were recorded by means of an isometric force transducer (Bio-Sys-Tech, Bialystok, Poland).

Experimental protocol

After equilibration period, the cumulative concentration-response curves were constructed for increasing concentrations of the imidazolines studied ranging from 10-11 to 10-3 M. Structure of compounds studied shows Fig. 1.

Fig. 1. Chemical structure of compounds studied.

In the next stage of the experiment, the inotropic and chronotropic responses to imidazolines studied were measured at the presence of fixed concentrations (from 10-9 to 10-3 M) of the imidazoline blockers (idazoxan or/and phentolamine).

Data presentation and statistical evaluation

Cumulative concentration-response curves with variable slope were constructed and analysed by means of GraphPad Prism4 software (GraphPad Software Inc., San Diego, Ca). Each point of the curves was a mean of at least 6 experiments. The changes of responses to each concentration was expressed as percent of the control value (100 %) of atrial rate or amplitude preceding the administration of cumulative concentrations of an agent with or without idazoxan or phentolamine. Data are reported as mean SEM (standard error of mean). Based on the profile of concentration-response behavior of isolated organs for the compounds studied, the -log EC50 parameters were calculated (the concentration of the ligand producing the half of the maximal effect observed). A significance level was taken as p 0.05 or p 0.01 in comparison of the compound studied alone and the same compound pretreated with idazoxan or phentolamine. Nonparametric analysis was done by U'Mann-Whitney (unpaired) test using Statistica 7.1 software (StatSoft Inc., Tulsa, OK, USA)

Data obtained are shown in Figs. 2-6 and in Tables 1-3.

Drugs

Pure chemical substances were used for the preparation of bath solutions of the drugs studied.

Phentolamine and idazoxan were obtained from Sigma (Steinheim, Germany); clonidine from Boehringer (Ingelheim, Germany); agmatine [(4-aminobutyl)guanidine], AGN192403 (2-endo-amino-3-exo-isopropylbicyclo[2.2.1]heptane) and BU239 (2-(4,5-dihydroimidazol-2-yl)quinoxaline) were from Tocris (Bristol, UK); 2-BFI (2-(2-benzofuranyl)-2-imidazoline) was from Tocris (London, UK); rilmenidine was a gift from Servier (Paris, France) and moxonidine was a gift from Dr. B. I. Armah, (BDF Research Laboratories, Hamburg, Germany). Stock solutions of each agent for in vitro studies were 10-3 M. The concentrations of idazoxan were 10-3, 10-5, 10-6, 10-7 and 10-9 M and phentolamine 10-9 M. Stock solutions were diluted with water ex tempore before individual experiments.


RESULTS

Inotropic activity

It has been demonstrated that clonidine, moxonidine and rilmenidine elicit the positive inotropic activity on electrically stimulated left atria with maximum effects of 132.1, 116.2 and 118.3 per cent, respectively (Fig. 2A, Table 1). The -log EC50 values observed for clonidine, moxonidine and rilmenidine were 5.2, 6.2 and 5.1, respectively (Table 2).

Fig. 2. Effect of cumulative concentrations of the imidazoline compounds studied on contractility of the left rat heart atria (A) and on beating rate of the right rat heart atria (B).
The data are shown as means of at least 6 experiments SEM

Table 1. Inotropic and chronotropic effects of the cumulative concentrations of the compounds studied. Maximum effect observed are expressed as % of control (at indicated molar concentration).

The presence of idazoxan 10-5 M and 10-3 M diminished positive inotropic effect of clonidine and rilmenidine (Figs. 3A, 3C). The -log EC50 values for clonidine and rilmenidine increased after pretreatment with idazoxan 10-3 and 10-5 M (Table 2). Surprisingly, moxonidine produced positive inotropic effect at the presence of idazoxan 10-5 M (Fig. 3B). The antagonism at the presence of idazoxan 10-3 M manifested itself at the high concentrations (10-5 - 10-3 M) of moxonidine. The -log EC50 for moxonidine alone was 6.2 and it increased to 7.2 after pretreatment with the 10-5 M idazoxan (Table 2).

Fig. 3. Effect of cumulative concentrations of clonidine (Clo) (A), moxonidine (Mox) (B) and rilmenidine (Ril) (C) alone and in the presence of fixed concentrations of idazoxan (Ida) on contractility of the left rat heart atria.
The data are shown as means of 6 experiments SEM; p 0.05 denotes significance level of differences between results obtained for the compounds alone and after pretreatment with idazoxan (Ida) as compared with the U'Mann-Whitney, unpaired test.

Table 2. The -log EC50 value for compounds studied alone and at the presence of idazoxan (Ida) or phentolamine (Phen).

AGN192403 in cumulative concentrations from 10-11 to 10-3 M does not act on amplitude of beating of left atria. A pretreatment with idazoxan 10-9 and 10-6 M or phentolamine 10-9 M increases inotropic activity of AGN 192403 but these effects are of no statistical significance.

Positive inotropic effect of compound 2-BFI appeared only at very low agent's concentrations (10-11-10-8 M). In higher concentrations, 2-BFI decreased the amplitude of contraction of left atria up to 76.1 %, in comparison to the 100% of control. The presence of idazoxan 10-3 M partially diminished inotropic effect of 2-BFI but concentration 10-5 M of idazoxan remained without effect on 2-BFI (Fig. 4A). The -log EC50 values of 2-BFI alone and pretreated with idazoxan 10-5 M were equally 7.0, and for 2-BFI pretreated with idazoxan 10-3 M both values were 9.5 (Table 3).

Fig. 4. Effect of cumulative concentrations of 2-BFI (A), BU239 (B) and Agmatine (Agm) (C) alone and in the presence of fixed concentrations of idazoxan (Ida) on contractility of the left rat heart atria.
The data are shown as means of 6 experiments SEM; p 0.05 denotes significance level of differences between results obtained for the compounds alone and after pretreatment with idazoxan (Ida) as compared with the U'Mann-Whitney, unpaired test.

Table 3. The -log EC50 value for compounds studied alone and at the presence of idazoxan (Ida) or phentolamine (Phen).

Compound BU239, which is structurally related to 2-BFI and has been reported to label I2 receptors, produced positive inotropic activity with the maximum effect observed of 117.4 % (Fig. 4B). Idazoxan at concentration 10-7 M slightly diminished the left atria amplitude of contraction evoked by BU239 whereas idazoxan 10-9 M increased the activity of BU239. The presence of phentolamine 10-9 M had no effect on inotropy of BU239. The -log EC50 values were also diminished from 8.8 for BU239 alone to 8.3 for BU239 with idazoxan 10-5 M, to 6.6 for BU239 with idazoxan 10-7 M and to 7.2 for BU239 with phentolamine 10-5 M (Table 3).

The most marked positive inotropy was observed for agmatine with maximal effect observed of 142.0 % and the -log EC50 of 8.2 (Fig. 4C, Table 3). Pretreatement with idazoxan 10-7 M or 10-3 M agmatine decreased the amplitude of left atria. Phentolamine 10-9 M antagonized the inotropism of agmatine also. The -log EC50 values of agmatine with antagonists were diminished as has been presented in Table 3.

Idazoxan in cumulative concentrations did not affect the amplitude of beating of left atria: the maximum effect was 105.7 %. Also phentolamine remained without effect on the inotropic effect of left atria.

Chronotropic activity

Clonidine produced a weak positive chronotropic effect on the right atria up to maximum of 110.2 % (Figs. 2B and 5A). Idazoxan at concentrations 10-5 and 10-3 M markedly antagonized positive chronotropic activity of clonidine and decreased the -log EC50 value about 100-fold (Table 2).

Rilmenidine and moxonidine had no effect, either alone or pretreated with idazoxan, on right atria (Figs. 5B, 5C).

Fig. 5. Effect of cumulative concentrations of clonidine (Clo) (A), moxonidine (Mox) (B), rilmenidine (Ril) (C) and AGN192403 (D) alone and in the presence of fixed concentrations of idazoxan (Ida) on beating rate of the right rat heart atria.
The data are shown as means of 6 experiments SEM; p 0.05, p 0.01 denotes significance level of differences between results obtained for the compounds alone and after pretreatment with idazoxan (Ida) as compared with the U'Mann-Whitney, unpaired test.

AGN192403 increased weakly the rate of beating of the right atria (maximum effect was 110.4 %). In experiments with preexposure to idazoxan 10-9 M or 10-6 M, the antagonism with regards to chronotropic action occured with the statistical significance of p< 0.05. In the case of preexposure with phentolamine 10-9 M, there was no change of this effects (Fig. 5D). The -log EC50 values were changed for AGN192403 when pretreated with idazoxan 10-6 M and with phentolamine 10-9 M, from 7.0 to 3.9, and 4.0, respectively.

2-BFI decreased weakly the rate of beating of the right atria (from maximum effect of 106.5 to minimum of 93.0 %) (Fig. 6A). In the presence of various concentrations of idazoxan, the chronotropic effect of 2-BFI was not changed (Fig. 6A). The -log EC50 values from experiments with 2-BFI untreated and treated with either idazoxan or phentolamine were ever in the range from 6.9 to 8.1 (Table 3).

Fig. 6. Effect of cumulative concentrations of 2-BFI (A), BU239 (B) and Agmatine (Agm) (C) alone and in the presence of fixed concentrations of idazoxan (Ida) on beating rate of the right rat heart atria.
The data are shown as means of 6 experiments SEM; p 0.05 denotes significance level of differences between results obtained for the compounds alone and after pretreatment with idazoxan (Ida) as compared with the U'Mann-Whitney, unpaired test.

BU239 in cumulated concentrations significantly increased the rate of beating of the right atria and the maximum effect observed was 131.0 %. However, in the presence of idazoxan 10-7 and 10-9 M the chronotropic effect of BU239 was reduced (Fig. 6B). Phentolamine 10-9 M, added to BU239 preincubated with idazoxan 10-9, 10-7 or 10-5 M, had no effect. All the values of -log EC50 from experiments involving BU239 were similar, ranging only from 7.3 to 8.3 (Table 3).

Agmatine in cumulative concentrations accelerated the rate of the spontaneously beating right heart atria to a maximum of 125.7 % (Fig. 6C). Idazoxan in concentrations 10-7, 10-5 or 10-3 M attenuated agmatine's positive chronotropic effect. Phentolamine 10-9 M, added to agmatine pretreated with idazoxan, significantly decreased the rate of beating of the atria. The -log EC50 value decreased about 100-fold (Table 3).

Neither idazoxan nor phentolamine in cumulative concentrations had any marked effect on chronotropic activity of right atria.


DISCUSSION

Inotropy

In this work manifestations of positive inotropic activity were shown in case of agmatine, clonidine, rilmenidine and moxonidine. The narrow range of -log EC50 values for clonidine, rilmenidine and moxonidine from 5.1 to 6.2 indicates a similar receptor-mediated effect of those drugs. Idazoxan diminished significantly statistically the positive inotropic effect of clonidine and rilmenidine, but not that of moxonidine. The positive inotropic effect of rilmenidine was diminished by idazoxan, unless idazoxan was applied at a high concentration 10-3M, suggesting that rilmenidine exhibits some selectivity for imidazoline receptors. Rilmenidine is usually considered to be an I1 receptor ligand. However, recent evidences show that it may also label an I2-like site (52). Moreover, it was reported that the hypotensive effect of rilmenidine in humans was potently antagonized by idazoxan, whereas it was weakly or not at all antagonized by yohimbine. At the oral dose of 2 mg, rilmenidine has no effect on beating rate of the heart (53, 54).

Molderings et al. (52) observed that rilmenidine and oxymetazoline are potent full agonists to 2 adrenoceptors in rabbit hearts, whereas in the human atrial appendages both agents are antagonists at the 2 autoreceptors, like rauwolscine and idazoxan are. Prazosin is ineffective in that preparation. The antagonistic activity of rilmenidine towards human 2A adrenoceptors indicates that, in contrast to the suggestion based on rabbit data, the hypotensive effect on humans is not due to activation of 2A adrenoceptors but other, presumably I1 imidazoline, receptors are involved (52). In our work, the increasing -log EC50 values for clonidine and rilmenidine at the presence of idazoxan 10-3 M, suggest dual interaction of the imidazolines with the 2 adenergic and the imidazoline I1 receptors. In the case of clonidine it would confirm the hypothesis that its positive inotropic effect at low doses is mainly due to a stimulation of postsynaptic a adrenoceptors, whereas additional stimulation of the imidazoline receptors occurs at higher drug doses. Certainly our results do not exclude the possibility that rilmenidine and clonidine elicit their effects through putative presynaptic imidazoline receptors, at least as regards the inotropic activity.

In our study moxonidine elicited a weak positive inotropic effect on left atrium. This finding is in accordance with the results obtained by Raasch et al. (55). Evidently the presence of idazoxan potentiates the inotropic effect of moxonidine. It suggests that moxonidine acts as an agonist-antagonist on both adrenoceptors and imidazoline receptors. The -log EC50 value increases in the presence of idazoxan at concentration 10-5 M, but it decreases when 10-3 M idazoxan is added to the incubation medium. On the other hand, it has been reported, that moxonidine reduced norepinephrine release independently of I1 receptor, thus suggesting the prominent effect of 2 adrenoceptors in cardiac tissue (52). Moxonidine binds with different affinities to cardiac imidazoline I1 receptors, 2 adrenoceptors (56), and, at some conditions, to 1 adrenoceptors. Raasch et al. (55) explain the increase of contractility of left rat atria by stimulation of postsynaptic 1 adrenoceptors rather than the imidazoline I1 receptors. Hovewer, in experiments consisting in chronic moxonidine treatment of the spontaneously hypertensive rats Mukaddam-Daher and Gutkowska (16) and El-Ayoubi et al. (17) observed the specific binding with moxonidine at the atrial I1 subtype receptors.

A structurally related to moxonidine compound AGN192403 did not change the amplitude of beating of the left atria. However, in the presence of various concentrations of idazoxan, AGN192403 showed a weak nonsignificant positive inotropic effect, potentiated with 10-9 M of either phentolamine or idazoxan, similarly as in experiments with moxonidine. Most authors suggest that AGN192403 has no effect on circulatory system. However, its general behaviour may suggest it to be a selective ligand of I1 receptor (15, 57, 58) in the experiments on rat hearts, demonstrated that AGN192403 had no influence on norepinephrine level. According to these authors AGN192403 seems to be an antagonist to the imidazoline I1 receptor and the potentiation of inotropic activity by idazoxan seems to result from synergistic interactions.

In the case of agmatine most important are increases of the amplitude of the left atria contraction. Idazoxan at concentration of 10-7, 10-5 and 10-3 M, and phentolamine at concentration of 10-9 M, diminished inotropic activity in a dose dependent manner. Except of imidazoline receptors, agmatine has affinity to both the /ß adrenergic and dopaminergic receptors. Its effect is mediated probably by all those receptor sites in cardiac tissue.

Compound 2-BFI and a more potent ligand at I2 imidazoline receptor, BU239, evoked a very weak positive inotropic activity. The presence of idazoxan diminished inotropic effect of 2-BFI and BU239 without statistical significance. The -log EC50 values for BU239, 2-BFI and idazoxan were 8.8, 7.0 and 7.0, respectively. The reported binding affinities, Ki for I2 receptors in rat brain membrane are 4.2 and 7 nM for BU239 and 2-BFI, respectively (19). Pharmacometric analysis of the data obtained involving 2-BFI and BU239 lead to conclusion that these ligands cannot be clearly identified as either agonists or antagonists of the I2 receptor. This hypothesis is in accordance with the conclusion from the study on the relaxation of rat jejnum evoked by 2-BFI and idazoxan (59).

In our conclusion, the positive inotropic action on isolated rat heart left atria is with the following rank order for the agents studied: agmatine >> clonidine > BU239 rilmenidine moxonidine. Rilmenidine and moxonidine act as partial agonists of the imidazoline I1 receptors. In inotropic effects of these imidazolines both the I1 and 2 receptors are engaged. Inotropic effect of clonidine and agmatine is mostly due to the adrenoceptors activation. The role of I2 imidazoline receptors is not to convince.

Chronotropy

Agmatine and clonidine were found to elicit positive chronotropic effect on the right rat heart atria. Idazoxan markedly antagonized activity of clonidine, but independly of the dose used. It is well known that clonidine is an agonist not only of the 2 but also of the 1 adrenoceptors, present in the right atria.

Phentolamine 10-9 M, added to agmatine, significantly decreased positive chronotropic effect of agmatine. In opposite, in the experiments of some authors (41, 60) agmatine did not influence contractions of isolated rat heart atria. Agmatine has affinity to both the adrenergic and the imidazoline receptors. Some investigators demonstrated that agmatine recognizes 2 adrenoceptors but is without effect on these receptors (61). To date, there are no proofs of action of agmatine attributable to agonism or antagonism at the site in vitro.

In the present study, rilmenidine and moxonidine had almost no effect on the spontaneously beating right atria. Pretreatment with idazoxan attenuated the chronotropic effect of the drugs, but this antagonism against the chronotropic effect of rilmenidine and moxonidine never reached statistical significance, presumably because the induced effects were very small. Moxonidine and rilmenidine are the most selective agonists for I1 receptors among imidazoline agents. Nevertheless, some authors classify moxonidine and rilmenidine among selective 2 adrenoceptor agonists clainning that this receptor may be predominant in the chronotropic activity of the assumed I1 imidazoline receptor agonists.

Compound AGN192403 possesses a small positive chronotropic activity on right atrium and this effect is blocked by idazoxan 10-6 M and phentolamine 10-9 M as proved by the decreased -log EC50 values.

An I2 imidazoline receptor ligand, 2-BFI, had a very weak negative chronotropic activity, probably due to its antagonistic activity towards the I2 receptor subtype (19). After pretreatement with idazoxan the effect of 2-BFI remains unchanged.

Another I2 receptor ligand, BU239, is the most potent of all the agents studied in increasing the beating rate of the right atria. An antagonist, idazoxan, applied at various concentrations diminished this activity in an irregular manner. The -log EC50 values determined after the preincubation with idazoxan in different concentrations, were closely similar. It may suggest that the chronotropic mechanism of BU239 involves the imidazoline I2 receptors.

In conclusion, regarding to the maximal efffect observed, the positive chronotropic action on isolated rat heart right atria were with the rank order for the agents studied: BU239 ³ agmatine >> clonidine > AGN192403. In view of our research, the engagement of imidazoline receptors in the chronotropic response of rat heart atria to imidazoline drugs still remains disputable. Certainly, regarding to chronotropic effect of agmatine and clonidine we feel obliged to acknowledge an involvement of the 2/1 adrenergic receptors. However, as concerns BU239, the results obtained by us demonstrate that its activity on the rat right heart atrial beating rate is exerted via the imidazoline I2 receptors.


REFERENCES
  1. Molderings GJ. Imidazoline receptors: basic knowledge, recent advances and future prospects for therapy and diagnosis. Drugs Future 1997; 22: 757-772.
  2. De Vos H, Bricca G, De Keyser J, De Backer J-P, Bousquet P, Vauquelin G. Imidazoline receptors, non-adrenergic idazoxan binding sites and 2 adrenoceptors in the human central nervous system. Neuroscience 1994; 59: 589-598.
  3. Bousquet P, Greney H, Bennai F et al. Imidazoline receptors and cardiovascular regulations. A statement. Ann NY Acad Sci 1995; 763: 526-530.
  4. Ernsberger P. Heterogeneity of imidazoline binding sites: proposed I1 and I2 subtypes. Fund Clin Pharmacol 1992; 6(Suppl.1): 55.
  5. Miralles A, Olmos G, Sastre M, Barturen F, Martin J, Garcia-Sevilla JA. Discrimination and pharmacological characterization of I2-imidazoline sites with [3H]idazoxan and 2-adrenoceptor with [3H]RX821002 (2-methoxyidazoxan) in the human and rat brains. J Pharmacol Exp Ther 1993; 264: 1187-1197.
  6. Wikberg JES, Uhlen S, Chjalari V. Evidence that drug binding to non-adrenergic [3H]idazoxan binding sites (I-receptors) occurs to interacting or interconvertible affinity forms of the receptor. Pharmacol Toxicol 1992; 70: 208-219.
  7. Greney H, Molines D, Bousquet P, Dontenwill M. Heterogeneity of imidazoline binding sites revealed by a cirazoline derivative. Eur J Pharmacol 1994; 271: 533-536.
  8. Morgan NG, Chan SLF. Imidazoline receptors and their ligands as potentiators of nutrient-induced insulin secretion. Drug Design Reviews 2004; 1: 185-193.
  9. Scheen AJ. Pharma Clinics. Medication of the month. Moxonidine (Moxon). Rev Med Liege 2000; 55: 61-63.
  10. Camilleri G, Portal B, Quiniou G, Clerson P. Comparison of the efficacy and the safety of two imidazoline receptors agonists: rilmenidine and moxonidine. Ann Cardiol Angiol 2001; 50: 169-174.
  11. Raddatz R, Savic SL, Laniers SM. Imidazoline binding domains on MAO-B. Localization and accessibility. Ann NY Acad Sci 1999; 881: 26-31
  12. Molderings GJ, Göthert M. Imidazoline binding sites and receptors in cardiovascular tissue. Gen Pharmacol 1999; 32: 17-22.
  13. Trendelenburg A-U, Sutej I, Wahl CA, Molderings GJ, Rump LC, Starke K. A re-investigation of questionable subclassifications of presynaptic 2-autoreceptors: rat vena cava, rat atria, human kindey and guinea-pig urethra. Naunyn-Schmiedeberg's Arch Pharmacol 1997; 356: 721-737.
  14. El-Ayoubi R, Gutkowska J, Regunathan S, Mukaddam-Daher S. Imidazoline receptors in the heart: characterization, distribution and regulation. J Cardiovasc Pharmacol 2002; 39: 875-883.
  15. Schafer U, Burgdorf C, Engelhardt A, Raasch W, Kurz T, Richardt G. Moxonidine displays a presynaptic alpha-2-adrenoceptor-dependent synergistic sympathoinhibitory action at imidazoline-I1 receptors. Ann NY Acad Sci 2003; 1009: 265-269.
  16. Mukaddam-Daher S, Gutkowska J. Imidazoline receptors in the heart: a novel target and a novel mechanism of action that involves atrial natriuretic peptides. Braz J Med Biol Res 2004; 37: 1239-1245.
  17. El-Ayoubi R, Menaouar A, Gutkowska J, Mukaddam-Daher S. Imidazoline receptors, but not alpha2-adrenoceptors are regulated in SHR heart by chronic moxonidine treatment. J Pharmacol Exp Ther 2004; 310: 446-451.
  18. Bousquet P, Bruban V, Schann S, Feldman J. Imidazoline receptors: a challenge. Pharm Acta Helv 2000; 74: 205-209.
  19. Regunathan S, Youngson C, Raasch W, Wang H, Reis DJ. Imidazoline receptors and agmatine in blood vessels: a novel system inhibiting vascular smooth muscle proliferation. J Pharmacol Exp Ther 1996; 276: 1272-1282.
  20. Coupry I, Limon I, Tesson F, Lachaud V, Gargalidis-Moudanos C, Parini A. Imidazoline-guanidine site: a subtype of imidazoline receptors. Therapie 1992: 47: 519-524.
  21. Holt A, Wieland B, Baker GB. Allosteric modulation of semicarbazide-sensitive amine oxidase activities in vitro by imidazoline receptor ligands. Br J Pharmacol 2004; 143: 495-507.
  22. MacInnes N, Handley S. Autoradiographic localization of [3H]2-BFI imidazoline I2 binding sites in mouse brain. Eur J Pharmacol 2005; 516: 139-44.
  23. Romer L, Wurster S, Savola J-M, Raasmaja A. Iidentification and characterization of the imidazoline I2-binding sites in the hamster brown adipose tissue as a study model for imidazoline receptors. Arch Physiol Biochem 2003; 111: 159-166.
  24. Raasch W, List B, Hauser W, Schafer U, Dominiak P. Presynaptic release of noradrenaline is mediated not only through 2-adrenoceptors but also through imidazoline binding sites. Dtsch Med Wschr 1999; 124: S114.
  25. Khan ZP, Ferguson C, Jones RM. Alpha-2 and imidazoline receptor agonists. Their pharmacology and therapeutic role. Anaesthesia 1999; 54: 146-165.
  26. Head GA, Mayorov DN. Imidazoline receptors, novel agents and therapeutic potential. Cardiovasc Hematol Agents Med Chem 2006; 4: 17-32.
  27. Chan CKS, Sannajust F, Head GA. Role of imidazoline receptors in the cardiovascular actions of moxonidine, rilmenidine and clonidine in conscious rabbits. J Pharmacol Exp Ther 1996; 276: 411-420.
  28. Mukaddam-Daher S, Menaouar A, Gutkowska J. Receptors in moxonidine stimulated atrial natriuretic peptide release from isolated normotensive rat hearts. Eur J Pharmacol 2006; 541: 73-79.
  29. Mukaddam-Daher S, Gutkowska J. Atrial natriuretic peptide is involved in renal actions of moxonidine. Hypertension 2000; 35: 1215-1220.
  30. Urban R, Szabo B, Starke K. Involvement of peripheral presynaptic inhibition in the reduction of symphathetic tone by moxonidine, rilmenidine and UK14.304. Eur J Pharmacol 1995; 282: 29-37.
  31. Szabo B, Bock C, Nordheim U, Niedewrhoffer N. Mechanism of the sympatho-inhibition produced by the clonidine-like drugs rilmenidine and moxonidine. Ann NY Acad Sci 1999; 881: 264.
  32. Raasch W, Jungbluth B., Schäfer U, Hauser W, Dominiak P. Modification of noradrenaline release in pithed spontaneously hypertensiverats by I1-binding sites in addition to alph2-adrenoceptors. J Pharmacol Exp Ther 2003; 304: 1063-1071.
  33. Bousquet P. I1 receptors, cardiovascular function and metabolism. Am J Hypertens 2001; 14: 317-321.
  34. Alemany R, Olmos G, Escriba PV, Menargues A, Obach R, Garcia-Sevilla JA. LNS, a selective ligand for imidazoline I2 receptors, on glial fibrillary acidic protein concentration. Eur J Pharmacol 1995; 280: 205-210.
  35. Kourilsky O. ITERIUM - Clinical benefits from an innovative antihypertensive treatment. J Cardiol 2003; 10(Suppl. D): 8-13.
  36. Widimsky J, Sirotiakova J. Efficacy and tolerability of rilmenidine compared with israpidine in hypertensive patients with features of metabolic syndrom. Curr Med Res Opin 2006; 22: 1287-1294.
  37. Raasch W, Regunathan S, Li G, Reis DJ. Agmatine, the bacterial amine, is widely distributed in mammalian tissues. Life Sci 1995; 56: 2319-2330.
  38. Otake K, Ruggiero DA, Regunathan S, Wang H, Milner TA, Reis DJ. Regional localization of agmatine in the rat brain: an immunocytochemical study. Brain Res 1998; 787: 1-14.
  39. Raasch W, Schafer U, Chun J, Dominiak P. Biological significance of agmatine, an endogenous ligand at imidazoline binding sites. Brit J Pharmacol 2001; 133: 755-780.
  40. Reis DJ, Regunathan S. Agmatine: an endogenous ligand at imidazoline receptors may be a novel neurotransmitter in brain. J Auton Nerv Syst 1998; 72: 80-85.
  41. Li G, Regunathan S, Barrow CJ, Eshraghi J, Cooper R, Reis DJ. Agmatine: an endogenous clonidine displacing substance in brain. Science 1994; 263: 966-969.
  42. Pinthong D, Wright IK, Hannar C et al. Agmatine recognizes alpha-2 adrenoceptor binding sites but neither activates nor inhibits alpha-2 adrenoceptors. Naunyn Schmiedebergs Arch Pharmacol 1995; 351: 10-16.
  43. Pineda J, Ruiz-Ortega JA, Martin-Ruiz, Ugedo L. Agmatine does not have activity at alpha 2-adrenoceptors which modulate the firning rate of locus coreuleus neurones: an electrophysiological study on rats. Neurosci Lett 1996; 219: 103-106.
  44. Herman ZS. Agmatine - a novel endogenous ligand of imidazoline receptors. Pol J Pharmacol 1997; 49: 85-88.
  45. Bricca G, Greney H, Dontenwill-Kieffer M, Zhang J, Belcourt A, Bousquet P. Heterogenity of the specific imidazoline binding of [3H]-idazoxan in the human cerebral cortex. Neurochem Int 1993; 22: 153-163.
  46. Reis DJ, Regunathan S, Meeley MP. Imidazole receptors and clonidine-displacing substances in relationship to control blood pressure, neuroprotection and adrenomedullary secretion. Am J Hypertens 1992; 5: S51-S57.
  47. Munk SA, Lai RK, Burke JE et al. Synthesis and pharmacologic evaluation of 2-endo-amino-3-exo-isopropylbicyclo[2,2,1]heptane: a potent imidazoline1 receptor specific agent. J Med Chem 1996; 39: 193-195.
  48. Velliquette RA, Ernsberger PJ. The role of I1-imidazoline and 2-adrenergic receptors in the modulation of glucose metabolism in the spontaneously hypertensive obese rat model of metabolic syndrome X. J Pharmacol Exp Ther 2003; 306: 646-657.
  49. Angel I, Le Rouzic M, Pimoule C, Graham D, Arbilla S. [3H]-Cirazoline as a tool for the characterization of imidazoline sites. Ann NY Acad Sci 1995; 763: 112-124.
  50. Hudson AL, Gough R, Tyacke RJ et al. Novel selective compounds for the investigation of imidazoline receptors. Ann NY Acad Sci 1999; 881: 81-91.
  51. Garaj V, Remko M. Conformational study of drugs with effect on I1-imidazoline receptors. Ceska Slov Farm 2002; 51: 196-199.
  52. Molderings GJ, Menzel S, Kathmann M, Schlicker E, Gothert M. Dual interaction of agmatine with the rat alpha (2D)-adrenoceptor: competitive antagonism and allosteric activation. Br J Pharmacol 2000; 130: 1706-1712.
  53. Head GA, Godwin SJ, Sannajust F. Differential receptors involved in the cardiovascular effects of clonidine and rilmenidine in conscious rabbits. J Hypertens Suppl 1993; 11(5): S322-S325.
  54. Feldman J, Tibirica E, Bricca G, Dontenwill M, Belcourt A, Bousquet P. Evidence for the involvement of imidazoline receptors in the central hypotensive effect of rilmenidine in the rabbit. Br J Pharmacol 1990; 100: 600-604.
  55. Raasch W, Chun KRJ, Dendorfer A, Dominiak P. Positive inotropic effects of imidazoline derivatives are not via imidazoline binding sites, but alpha1-adrenergic receptors. Jpn J Pharmacol 2000; 84: 1-6.
  56. Haxiu MA, Dreshaj I, Schafer SG, Ernsberger P. Selective antihyprtensive action of moxonidine is mediated by I1-imidazoline receptors in the rostral ventrolateral medulla. J Cardiovasc Pharmacol 1994; 24(Suppl. 1): S1-S8.
  57. Brede M, Philipp M, Kaus A, Muthing V, Hein L. 2-Adrenergic receptor subtypes - novel functions uncovered in gene-targeted mouse models. Biol Cell 2004; 96: 343-348.
  58. Dardonville C, Rozas I. Imidazoline binding sites and their ligands: an overview of the different chemical structures. Med Res Rev 2004; 24: 639-661.
  59. Kaliszan W, Petrusewicz J, Kaliszan R. Imidazoline rceptors in relaxation of acetylocholine-constricted isolated rat jejnum. Pharmacol Reports 2006; 58: 700-710.
  60. Raasch W, Schafer U, Qadri F, Dominiak P. Agmatine, an endogenous ligand of imidazoline binding sites, does not antagonize the clonidine mediated blood pressure reaction. Br J Pharmacol 2002; 135: 663-672.
  61. Pintong D, Wright LK, Hammer C et al. Agmatine recognizes alpha-2 adrenoceptor binding sites but neither activates nor inhibits alpha-2-adrenoceptors. Naunyn-Schmiedeberg's Arch Pharmacol 1995; 351: 10-16.

R e c e i v e d : July 27, 2008
A c c e p t e d : February 20, 2009

Author’s address: Dr Aleksandra Radwanska, Department of Biopharmaceutics and Phamacodynamics, Medical University of Gdansk, Gen. J. Hallera 107, 80-416 Gdansk, Poland. Phone: +48 58 349 32 60; Fax: +48 58 349 32 62; e-mail: alex@amg.gda.pl