Journal Logo

Articles

Effects of Azimilide on Cardiovascular Tissues from Normo- and Hypertensive Rats

Nand, Vinita; Doggrell, Sheila A.

Author Information
Journal of Cardiovascular Pharmacology: August 2000 - Volume 36 - Issue 2 - p 209-217
  • Free

Abstract

Potassium channel blockers, which inhibit repolarization to prolong the cardiac action potential, are commonly used in the treatment of cardiac arrhythmias (1,2). Prolonging the action potential with nonselective potassium channel blockers such as tetraethylammonium and 4-aminopyridine is associated with a positive inotropic effect on the spontaneously hypertensive rat (SHR) left ventricle strip (3). Potassium channel blockers are not generally considered a potential positive inotropic mechanism for the treatment of heart failure. There are probably a number of reasons for this. One reason for not considering potassium channel blockade in heart failure is that the nonselective blockers have been shown to cause vasoconstriction [tetraethylammonium, 4-aminopyridine (3,4)], and this would be detrimental in failure. The appropriate studies have not been undertaken to ascertain whether potassium channel blockers have positive inotropic effects that are maintained in the presence of cardiac disease.

Another possible reason for not considering potassium channel blockers in heart failure is that they may have proarrhythmic effects in the absence or presence of β-adrenoceptor activation. Cardiac force is augmented by noradrenaline released from the sympathetic nervous system acting at β-adrenoceptors. Sympathetic nervous system activity is high in the failing heart to maintain cardiac output (5). Stimulation of cardiac β-adrenoceptors is associated with an increase in the amplitude of the action potential due to the opening of calcium channels. There is also an increase in the rate of repolarization with the stimulation of cardiac β-adrenoceptors due to the opening of the slow component of the delayed outward rectifying potassium channel (6). It is possible that the combination of a potassium channel blocker and a β-adrenoceptor agonist could be arrhythmogenic.

Many class III antiarrhythmic agents are selective inhibitors of the fast component of the delayed outward rectifying potassium channel. Azimilide is a unique potassium channel blocker, as it shows similar inhibitory potency against the fast and slow components of the delayed outward rectifying potassium channel (7). Azimilide is under clinical investigation as an antiarrhythmic agent (8), but has not been considered a positive inotrope for the treatment of heart failure.

The overall aim of this study was to test whether azimilide has some potential for use in the treatment of heart failure. First, we determined the effects of azimilide on the isolated aorta, portal vein, and intralobar pulmonary and mesenteric arteries from Wistar-Kyoto (WKY) rats. Second, we determined the effects of azimilide on the action potentials of left ventricles isolated from WKY rats, and showed a prolongation of the potentials. Third, we determined whether azimilide was a positive inotrope. Isolated cardiac muscles were used to define positive inotropic responses because loading conditions can be well controlled, in contrast to studies in the whole animal. Thus, we have determined the effects of azimilide on the force responses of WKY left ventricles, and showed positive inotropism. These studies were performed in the absence and presence of isoprenaline to determine whether azimilide was proarrhythmogenic.

Finally, we determined whether the positive inotropic effects of azimilide were maintained in the presence of advanced cardiac hypertrophy and failure by studying the effects of the blockers on the left ventricle of 12- and 22-month-old SHRs. As hypertrophy in humans is usually associated with chronic hypertension, the SHR at 12 months is a realistic model of human hypertrophy (9); 22-month-old SHRs are in cardiac failure (10).

METHODS

Rats

Breeding pairs of WKY and Okamoto SHRs were purchased from the Animal Resources Centre, Perth, Western Australia, and then colonies of these rats were established in the Animal Resource Unit, School of Medicine, The University of Auckland. Adult rats were housed two to a cage with free access to standard rat chow and water.

Beginning of the experiment

Animal protocols were approved by the University of Auckland Animal Ethics Committee for the Helsinki guidelines. Rats were stunned and exsanguinated. The aorta, portal vein, lungs, mesenteric bed, and/or heart were rapidly removed and placed in Krebs solution that was saturated with 5% CO2 in oxygen and at 37°C. The free wall of the left ventricle was immediately excised. Aorta and portal veins were cleared of surrounding tissue. Third-branch 1.6-mm lengths of mesenteric arteries and intralobar pulmonary arteries were dissected. All experiments were performed in the presence of a modified Krebs solution [composition (mM): NaCl, 116; KCl, 5.4; CaCl2, 2.5 (aorta, portal vein and left ventricle) or 1.5 (intralobar pulmonary and mesenteric arteries); MgCl2, 1.2; NaH2PO4, 1.2; NaHCO3, 22.0; D-glucose, 11.2]. In the ventricle experiments, the Krebs contained guanethidine at 10−5M to prevent the release of noradrenaline from nerve endings and atropine at 10−6M to block muscarinic receptors at 37°C. In all experiments, mean values ± SEM were determined. Tests of significance between two groups were made by Student's paired or unpaired t test, as appropriate, or when more than two groups of data from several rats were involved, by analysis of variance followed by Students unpaired t test.

Recording of contractions from the aorta

From 14-week-old WKY rats, endothelium-intact aortic rings were prepared and suspended under 1.5 g tension in 5-ml organ baths and equilibrated for 60 min, during which 250 ml Krebs solution superfused the tissue. Contractility was measured isometrically with force-displacement transducers (Grass model FT03.C, Grass, MA, U.S.A.) and displayed on an eight-channel Grass polygraph (model 79B, Grass Polyview, RI, U.S.A.). When testing the effects of azimilide on the quiescent aorta, the aorta was cumulatively challenged on a 5-min cycle. When testing the effect of azimilide on the KCl-contracted aorta, the aortic rings were contracted by the addition of KCl at 15 mM and challenged with azimilide on a 5-min cycle. The relaxation to azimilide was calculated as a percentage of the KCl contraction.

Recording of contractions from portal vein

The method used by us previously (3) was followed. Portal veins (∼12 mm) derived proximally to the liver of 14-week-old WKY rats were mounted longitudinally under 1 g tension in 5-ml organ baths containing Krebs solution and allowed to equilibrate for 30 min. During this equilibration period, the tissues were washed by the overflow of 300-ml Krebs. Contractions were measured isometrically with Grass force transducers, displayed on a Grass polygraph. The wash was stopped, and the contractions were allowed to stabilize, which took 20-30 min. Then a cumulative exposure to azimilide on a 5-min cycle was made. When investigating the effects of propranolol on the attenuating effect of azimilide, the portal vein was treated with propranolol for 60 min before the azimilide challenge. The amplitudes of the final three contractions before the addition of each concentration of azimilide were measured and averaged. Responses were calculated as a percentage change of the spontaneous contractile activity and were corrected for changes in untreated tissues.

Recording of contractions from intralobar pulmonary and mesenteric arteries

The method used by us previously (3) was followed. The contractions of the intralobar pulmonary and mesenteric arteries of 14-week-old WKY rats were recorded in the Mulvany myograph containing Krebs solution at 37°C. After equilibration for 1 h, the pulmonary and mesenteric arteries was set up under the equivalent of 15 and 60 mm Hg, respectively. The tissues were equilibrated for 15 min before two challenges with K+PSS (NaCl, 37.9; KCl, 85.8; CaCl2, 1.5; MgCl2, 1.2; NaHPO4, 1.2; MgCl2, 1.2; NaHCO3, 22.0; D-glucose, 11.2). The tissues were then washed for 15 min before a cumulative challenge to azimilide on a 3-min cycle. Contractions were calculated as a percentage of K+PSS response.

Recording of the action potentials from the electrically driven rat ventricle

The method used by us previously (3) was followed. A strip of 14-week-old WKY left ventricle that represented about a fifth of the ventricle was pinned endocardium-side uppermost to the silica gel base of a recording chamber between two platinum-stimulating electrodes. The tissues were superfused at ∼2 ml/min with Krebs solution that had been vigorously bubbled with 5% CO2 in oxygen before entering the recording chamber. Tissues were equilibrated for 60 min before the tip of a 3 M KCl-filled microelectrode was inserted into a cell close to the surface of the tissue. The tissues were then stimulated at 2 Hz, 5-ms duration and 30 V, and the resulting action potentials were amplified through a microelectrode preamplifier and displayed on an oscilloscope. After 3 min of stimulation, three action potentials were recorded from a single cell before the stimulator was turned off and azimilide was added to the Krebs solution perfusing the tissue. After a 60-min equilibration, a further period of stimulation and recording of three action potentials was undertaken from the same cell. This procedure was repeated with higher concentrations of azimilide. The diastolic potential, peak amplitude, and action potential duration at 50 and 90% repolarization were measured.

Measurement of blood pressure and heart rate

The 12- and 22-month-old WKY rats and/or SHRs were weighed, and then the tail cuff (systolic) blood pressure was measured by using a tail plethysmograph (IITC Life Science model 29). The rats were placed in Perspex holding cylinder and left in the dark for 30 min, during which time they routinely went to sleep. The occlusion cuff was placed around the tail, which had been warmed to 33°C under a reading light. The tail cuff was inflated to 250 mm Hg so that the arterial pulse displacements were no longer apparent. The pressure was gradually reduced until the pulse was observed on the chart recorder. The pulse point was recorded as the tail-cuff pressure, and the rate of the pulses as the in vivo pulse (heart) rate. Three readings were taken per rat, and these were usually very similar and were averaged.

Recording of the contractions from the electrically driven rat ventricle

The method used by us previously (3) was followed. Five strips were prepared from the left ventricle free wall. Four of the individual strips were mounted longitudinally between two platinum electrodes under 1 g tension in 5-ml organ baths in Krebs solution being vigorously bubbled with 5% CO2 in oxygen and allowed to equilibrate for 75 min. Stimulation at 2 Hz (5-ms duration, 30 V) was commenced, and after 3-6 min, isoprenaline at 10−10M was added, with the contractions being recorded through a Grass polygraph onto a Grass Polyview. The cumulative addition of isoprenaline (10−9, 10−8M, etc.) occurred on a 3-min cycle until an isoprenaline maximal response was obtained. Stimulation was then stopped, and three ventricle strips were treated with differing concentrations of azimilide, while the other strip remained untreated. Strips were superfused with ∼500 ml of Krebs solution over 75 min before a second challenge to cardiac stimulation and isoprenaline. This procedure was repeated, with the azimilide-treated tissues receiving higher concentrations of the drug, and the untreated tissue remaining untreated or being vehicle treated.

The time to peak force (TP) and the times from the peak to 50 and 90% relaxation (TR50 and 90) were measured in milliseconds. The force of contractions to electrical stimulation was measured as milligrams tension. The force responses just before the second and third challenges to isoprenaline were calculated as a percentage of the force responses to stimulation before the first challenge with isoprenaline. If these force responses to electrical stimulation between treated and untreated tissues were significantly different, the percentage difference of the values from the individual treated tissues from the mean of the untreated tissues was calculated. The maximal combined responses to electrical stimulation and isoprenaline were measured as milligrams tension and as a percentage of the maximal response to the first challenge to stimulation and isoprenaline, and compared. The maximal combined responses were not significantly different between azimilide-treated and untreated tissues in any of our experimental groups; thus the combined data were normalized.

Tissue characteristics

The atria and the free wall of the left and right ventricles and septum were separated, blotted, and weighed. At the end of the contractility experiment, the strips of ventricle were removed from the organ baths, and the lengths, circumference, and wall radius were measured before the tissues were blotted and weighed.

Drugs

The drugs used in this study were azimilide (donated by Proctor & Gamble Pharmaceuticals), atropine sulfate, guanethidine sulfate, (−)-isoprenaline bitartrate, and (±)-propranolol hydrochloride (Sigma Chemical Co.).

RESULTS

Effect of azimilide on blood vessels

Azimilide was tested on four aortae, nine portal veins, four mesenteric, and four intralobar pulmonary arteries from 14-week-old WKY rats. Azimilide at 10−8 to 3 × 10−5M had no effect on the quiescent aortae and mesenteric and intralobar pulmonary arteries (data not shown). Azimilide at 10−8-10−5M did not alter, but at higher concentrations, 3 × 10−5 to 3 × 10−4M, significantly (p < 0.05) relaxed the KCl-contracted aortae (Fig. 1). Azimilide at 3 × 10−6 to 10−4M caused much smaller, but significant (p < 0.05) attenuation of the contractions of portal veins (Fig. 1). The attenuating effects of azimilide on the portal vein were not altered in the presence of propranolol at 10−7M (data not shown).

FIG. 1
FIG. 1:
Effects of azimilide on the portal vein (▪) and aorta (▵) the Wistar-Kyoto (WKY) rat. The percentage relaxation is plotted against the negative logarithm of the molar concentration of azimilide. Each value is the mean ± SEM from nine (portal vein) or four (aorta) rats.

Effects of azimilide on left ventricular action potentials

The left ventricular action potentials from 14-week-old WKY rats had a mean diastolic membrane potential of −82 mV, amplitude of 96 mV, and APD50 and 90 values of 19 and 56 ms, respectively. Azimilide at 10−7-10−5M had no effect on the diastolic membrane potential or amplitude of left ventricular action potentials (data not shown), but prolonged the action potentials (Table 1). Azimilide at 10−7 and 10−6M caused a similar prolongation at the APD50 (action potential duration to 50% repolarization) and APD90 levels, but at 10−5M, caused a prolongation that was greater at the APD90 than APD50 level (Table 1).

TABLE 1
TABLE 1:
Effects of azimilide on left ventricle action potentials

SHR model of hypertrophy and heart failure

The WKY rats and SHRs were age-matched at 12 and 22 months, and were weight-matched at 12 months (Table 2). The 22-month-old SHRs were lighter than age-matched WKY rats (Table 2). The 12- and 22-month-old SHRs had higher tail-cuff blood pressures than age-matched WKY rats (Table 2). Heart rates were higher by 32 and 120 beats/min in 12- and 22-month-old SHRs compared with age-matched WKY rats (Table 2).

TABLE 2
TABLE 2:
Characteristics of 12- and 22-month-old WKY rats and SHRs

The SHRs had greater heart weights than the WKY rats (Table 2). To take into account that 22-month-old SHRs had lower body weights than age-matched WKY rats, heart weights also were calculated as percentage body weight. As percentage body weight, 12-month-old SHRs had greater left ventricle and septum, but not right ventricle or atria, weights than WKY rats (Table 2). The 22-month-old SHRs also had greater right ventricle and atria weights than age-matched WKY rats (Table 2).

The 12- and 22-month-old SHR left ventricle strips were longer and had a greater wall radius and circumference than those of age-matched WKY rats (Table 2). The greater wall radius is a marker of hypertrophy, and the greater length and circumference indicate dilation of the left ventricle.

The times to peak (TP values), magnitude of the contractions in response to electrical stimulation, and times to relaxation (TR values) of left ventricle strips were similar on 12- and 22-month-old WKY rats (Table 2). The TP values also were similar on 12-month-old SHRs, but were much prolonged on the left ventricles of 22-month-old SHRs (Table 2). The magnitude of contraction was less and the TR values were prolonged on the 12-month-old SHR left ventricles (Table 2). Further impairment of contraction and relaxation was observed on the left ventricles of 22-month-old SHRs (Table 2).

Effects of azimilide on left ventricular contractility

Azimilide, ≤1 × 10−5M, had no effect on the TP of the contractions of left ventricles of 12- and 22-month-old WKY rats or SHRs (data not shown). Azimilide at 10−7-10−5M significantly (p < 0.05) augmented the peak force of WKY left ventricle contractions to electrical stimulation (Fig. 2). There was also an augmentation of the submaximal, but not maximal, responses in the presence of isoprenaline (Fig. 2). The augmentation of force to electrical stimulation was similar on the 12-and 22-month-old WKY rats and 12-month-old SHRs but reduced on the 22-month-old SHR left ventricle; Fig. 3 shows augmentation with azimilide on 22-month-old ventricles.

FIG. 2
FIG. 2:
The effects of azimilide on the contractions of left ventricles from 22-month-old Wistar-Kyoto (WKY) rats (left) and spontaneously hypertensive rats (SHRs; right). Contractions from untreated ventricles (▪) and from ventricles treated with azimilide at 10−7 (❑), 10−6 (▴), and 10−5M (▵). Contractions were calculated as a percentage of the maximal response to cardiac stimulation and isoprenaline, and plotted against cardiac stimulation alone (CS) and then the negative logarithm of the molar concentration of isoprenaline. Each value is the mean ± SEM from six to eight preparations.
FIG. 3
FIG. 3:
The effects of azimilide on the contractions of left ventricles from 22-month-old Wistar-Kyoto (WKY) rats (❑) and spontaneously hypertensive rats (SHRs) (▪).Top: percentage augmentation of force; middle and bottom: percentage lengthening of TR50 and TR90, respectively, are plotted against the negative logarithm of the molar concentration of azimilide. Each value is the mean ± SEM from six to eight preparations. *p < 0.05, unpaired t test, with the same concentration of azimilide on 22-month-old WKY rats.

Azimilide at 10−7-10−5M significantly (p < 0.05) prolonged the TR90 of WKY left ventricles to a significantly greater extent than TR50 values (Fig. 3). The prolonging effect with azimilide was similar on 12- and 22-month-old WKY rats and 12-month-old SHRs but reduced on the 22-month-old SHR left ventricle; Fig. 3 shows prolongation with azimilide on 22-month-old ventricles.

Drugs that augment force with minimal effects on relaxation probably have the best profile for the treatment of heart failure. Similar augmentation/(relaxation lengthening) ratios for azimilide were obtained on 12- and 22-month-old WKY rats and SHRs. For azimilide at 10−7M on 12-month-old WKY rat left ventricles, the ratios at the TR50 and TR90 levels were significantly different, and were 2.8 ± 0.2 and 1.6 ± 0.2 (8), respectively. Similar ratios were obtained with azimilide at 3 × 10−7M (data not shown). For azimilide at 10−6M on left ventricles from 12-month-old WKY rats, the ratios at the TR50 and TR90 levels were 4.2 ± 0.2 and 3.2 ± 0.2 (8), respectively. The values with azimilide at 10−6M at the TR50 and TR90 levels were different from each other, and also greater than with the lower concentrations of azimilide. Similar ratios were obtained with higher concentrations of azimilide.

DISCUSSION

Nonselective potassium channel blockers cause vasoconstriction, which is an unwanted effect in heart failure. This study shows that azimilide does not cause vasoconstriction, but, in high concentrations, relaxes some vessels. Azimilide attenuates the contractions of the portal vein by a propranolol-insensitive mechanism, and thus β-adrenoceptors are not involved in this attenuation. The effects of azimilide on vascular Ca2+ currents have not been reported, but azimilide at 10−5M inhibits the L-type Ca2+ current of cardiomyocytes from rat, guinea pig, and dog (7). Thus, it is possible that azimilide relaxes the KCl-contracted aorta and attenuates the contractions of the portal vein by blocking the L-type Ca2+ current.

Azimilide prolongs the cardiac action potential at low frequencies [sheep Purkinje fibers (11); calf Purkinje and ventricular fibers, guinea pig papillary muscles (7); guinea pig isolated ventricular myocytes (12)]. On sheep and calf Purkinje fibers, azimilide prolongs the APD90 to a greater extent than it prolongs the APD50(7,11). In contrast, azimilide has similar effects on the APD50 and APD90 of calf ventricular fibers (7) and guinea pig cardiomyocytes (11). In our study, the lower concentrations of azimilide tested had similar effects on the WKY left ventricle APD50 and APD90 values, but at 10−5M caused a larger prolongation of the APD90 than of the APD50. Excessive prolongation of action potentials may be proarrhythmic. The lower concentrations of azimilide tested, 10−7-10−6M, are within the therapeutic range (13) and do not cause excessive prolongation. These concentrations of azimilide are unlikely to be proarrhythmic in normal tissue. The highest concentration of azimilide tested, 10−5M, does cause a pronounced prolongation of the left ventricle action potential at 2 Hz, which could be a predictor of proarrhythmia at low frequencies. It seems unlikely that azimilide will be proarrhythmic at more physiologic frequencies in normal tissue, as the ability to prolong action potentials is reduced or lost at higher frequencies (12,14).

We have studied the effects of azimilide on the action potentials of WKY ventricles. Action potentials are prolonged in hypertrophy [SHR (15), human (16), and SHR heart failure (17)]. It is possible that azimilide has proarrhythmic actions on diseased, but not normal, ventricle action potentials. Thus, further studies of the effects of azimilide on the action potentials from SHR hypertrophied and failing ventricle may be indicated.

Potassium channel blockers are used as antiarrhythmic agents (18). The beneficial effects of potassium channel blockers in arrhythmias are due to their ability to prolong the action potential and prevent reexcitation. However, proarrhythmic effects of potassium channel blockers also have been described, presumable due to excessive prolongation of the action potential or the appearance of afterdepolarization. This is a potential problem with the use of potassium channel blockers as positive inotropes. In our contractility studies, azimilide alone or in the presence of isoprenaline did not induce arrhythmias in left ventricles from normotensive or hypertensive rats. Anti-arrhythmic but not proarrhythmic effects have been reported with azimilide in numerous animal studies (7,19,20). In healthy volunteers, the electrophysiologic effects of azimilide lead to the expected prolongation of the QTc, with no reports of arrhythmias (13). It is likely that the definitive assessment of the anti- and/or proarrhythmic effects of long-term treatment with azimilide will be determined in the ongoing ALIVE clinical trial (21). The ALIVE trial is assessing the potential of azimilide for improving survival in patients after myocardial infarction at high risk of sudden death (21).

Our study was not designed to determine with which ion channels azimilide was interacting. Azimilide is a selective inhibitor of both components of the delayed outward rectifying current, IK, over other potassium channels (7). Thus, azimilide is a potent inhibitor of the IKs and IKr of guinea pig and dog ventricles, with little effect on the inward rectifier, IKl, current of these tissues and on the transient outward potassium current, Ito of dog ventricles (7). To date, only two outward rectifying potassium currents have been characterized in rat ventricles, Ito or Ito-f, and Ik or Ito-s(22), and azimilide has been shown not to block Ito(7). The effects of azimilide on rat ventricular IK have not been reported. Higher concentrations of azimilide also block the L-type calcium channel (7). Cardiac calcium channel blockers have negative inotropic effects. In our study, azimilide, <10−5M, was a positive inotrope, and thus it seems unlikely that azimilide has a marked effect on the ventricular calcium channels. As azimilide has no effect on Ito of rat ventricles, it seems likely that the effects of azimilide in our study are predominantly due to inhibiting IK.

Both positive and negative inotropic effects have been reported with azimilide. On the ferret right ventricular papillary muscle, azimilide (3 × 10−7 to 3 × 10−6M) was a negative inotrope, and the effect was greater at 3 than at 1 Hz (12). In isolated guinea pig papillary muscle, azimilide increased the force by 12% at 10−6M, had no effect at 3 × 10−6 and 10−5M, and decreased force by 26% at 3 × 10−5M(7). In rat papillary muscle, azimilide at ≤10−4M has no effect on force (7). The frequencies at which the guinea pig and rat papillary muscles were being driven are not given, and thus it is possible that the frequency was too low to observe a positive inotropic effect with azimilide. By using the WKY rat left ventricle driven at 2 Hz, we obtained much larger positive inotropic effects than those previously reported with azimilide. These large positive inotropic effects to azimilide observed in this study may reflect the use of a different tissue (ventricle in this study compared with papillary muscle in previous studies), different experimental conditions (e.g., Hz) or different exposure times (60 min in this study). Given that azimilide has been shown to have a positive inotropic effect in this study, it may be worth-while to test the effects on isolated human heart force.

One of the main aims of this study was to determine whether the positive inotropic effects of azimilide were maintained in the presence of advanced cardiac hypertrophy and failure by studying the effects of azimilide on the SHR left ventricles. As hypertrophy and then failure in humans is usually associated with chronic hypertension, the 12- and 22-month-old SHRs are realistic models of human hypertrophy and failure, respectively (9). We showed that at 12- and 22-month-old SHRs have left ventricular hypertrophy, impaired contractility of the left ventricle, and increased in vivo pulse rates. The increased pulse rate is probably due to reflex activation of the sympathetic nervous system to the heart to maintain cardiac output as the heart fails, and is much greater in SHRs at 22 than at 12 months.

The SHR as a model of heart failure has been characterized by Bing et al. (10). Of many markers tested, the most consistent marker of the SHR in failure was right ventricular hypertrophy (10). All of our SHRs at 22 months had right ventricular hypertrophy. Bing et al. (10) have divided their 18- to 24-month-old SHRs into two groups: SHR-F (failing), which have right ventricular hypertrophy, and SHR-NF (nonfailing), which do not have right ventricular hypertrophy. Our 22-month-old SHRs are a homogeneous SHR-F group, as they all had right ventricular hypertrophy.

It seems likely that our 22-month-old SHRs have a mixture of diastolic and systolic dysfunction in their heart failure. The thickened ventricles and prolonged relaxation of the left ventricle are markers of diastolic dysfunction. The dilated ventricles and prolongation of time to peak of the left ventricle contraction are indicative of systolic dysfunction.

Action potentials are prolonged in hypertrophy [SHR (15), human (16), and SHR heart failure (17)]. In our study, the inotropic effect of azimilide was maintained in SHR hypertrophy but reduced in heart failure. Thus, the loss of inotropic effect with azimilide is not closely associated with the prolongation of action potential, as this is observed in hypertrophy and heart failure.

Ito, but not IK, is reduced in SHR hypertrophy (23,24). The positive inotropic effects of blockers of the IKr, bretylium and clofilium, are similar on the left ventricles of 12-month-old SHRs and WKY rats (3,25). This suggests that the function of the IKr is not altered in the hypertrophied left ventricles of 12-month-old SHRs. Our study demonstrates that the positive inotropic effects of azimilide also are not altered in hypertrophied left ventricles of 12-month-old SHRs. Thus, it seems likely that the function of any IKs is also not altered in hypertrophied left ventricles at 12 months.

In cardiomyocytes from 18-month-old SHR heart failure, the Ito remained reduced with no effect on IK(23). In our study we show that the positive inotropic effects of azimilide were reduced on the failing left ventricles of 22-month-old SHRs. It seems unlikely that this reduced effect of azimilide is due to the reduced Ito, as Ito is reduced at 12 months when the positive inotropic effects of azimilide are not. Also azimilide has been shown not to inhibit Ito(7). It is possible that there may be further alterations in potassium currents between 18 and 22 months. One possible explanation for the loss of inotropic effect with azimilide is an alteration in the function of the IK in failing SHR left ventricles at 22 months. Further ion channel studies and studies with selective inhibitors of the fast and slow components are needed to elucidate whether the IK is altered as heart failure progresses.

Prolongation of relaxation is an undesirable property in a drug for heart failure. Ideally drugs used as positive inotropes in heart failure should increase force with little or no effect on relaxation. The relaxation of 12-month-old SHR left ventricles was prolonged, and azimilide further prolonged relaxation. In this study, azimilide at ≥ 10−6M had a percentage augmentation/percentage lengthening ratio of ≥3.2, which is greater than that of bretylium, clofilium, tetraethylammonium, or the lower concentrations of 4-aminopyridine (3,25). This means that it is likely that azimilide would have a more favorable profile than clofilium, bretylium, tetraethylammonium, or 4-aminopyridine on contractility in heart failure.

We tested azimilide on the SHR as our model of hypertrophy and failure. In most species including the human, IK has two components, and azimilide is a selective inhibitor of IKs. IKs has not been demonstrated on rat ventricle. As rat ventricle may lack IKs, there may be major species differences in the effects of azimilide, and our results with azimilide may not be accurately predict the effects in humans. Further animal testing of the cardiovascular effects of azimilide in a model ventricle containing IKs may be warranted.

In summary, azimilide is a vasodilator on isolated rat blood vessels, and this would be beneficial rather than detrimental in heart failure. Azimilide prolongs the rat left ventricle action potential and has not been reported to be proarrhythmic. Azimilide has positive inotropic effects on the rat left ventricles that are fully or partially maintained in hypertrophy and failure. The trough and peak steady-state azimilide concentrations in healthy volunteers are between 2 × 10−7 and 2 × 10−6M(13). Thus, most of the effects observed in our study are with clinically relevant concentrations. In conclusion, the potential of azimilide as a positive inotrope in the treatment of heart failure should be further considered.

Acknowledgment: This study was supported by a Project grant from the Auckland Medical Research Foundation.

REFERENCES

1. Colatsky T. Antiarrhythmic drugs: where are we going? Pharm News 1995;2:17-23.
2. Singh BN, Ahmed R. Class III antiarrhythmic drugs. Curr Opin Cardiol 1994;9:12-22.
3. Nand V, Doggrell SA. Effects of tetraethylammonium, 4-aminopyridine and bretylium on cardiovascular tissue from normo- and hypertensive rats. J Pharmacol Pharm 1999;51:631-40.
4. Halliday FC, Aaranson PL, Evans AM, Gurney AM. The pharmacological properties of K+ currents from rabbit isolated aortic smooth muscle cells. Br J Pharmacol 1995;116:3139-48.
5. Katz A. Heart failure. In: Katz A, ed. Physiology of the heart. Raven Press, 1992:638-68.
6. Sanguinetti MC, Jurkiewicz NK, Siegl PKS. Isoproterenol antagonizes prolongation of refractory period by the class III antiarrhythmic agent E-4031 in guinea pig myocytes. Circ Res 1991;68:77-84.
7. Salata JJ, Brooks RR. Pharmacology of azimilide dihydrochloride (NE-10064): a class III antiarrhythmic agent. Cardiovasc Drug Rev 1997;2:137-56.
8. Singh BN. Current antiarrhythmic drugs: and overview of mechanisms of action and potential clinical utility. J Cardiovasc Electrophysiol 1999;10:283-301.
9. Doggrell SA, Brown L. Rat models of hypertension, cardiac hypertrophy and failure. Cardiovasc Res 1998;39:89-105.
10. Bing OHL, Brooks WW, Robinson KG, et al. The spontaneously hypertensive rat as a model of the transition from compensated left ventricular hypertrophy to failure. J Mol Cell Cardiol 1995;27:383-96.
11. McIntosh MA, Tanira M, Pacini D, Kane KA. Comparison of the electrophysiological effects of NE-10064 with sotalol and E-401 and their modification by simulated ischaemia. J Cardiovasc Pharmacol 1994;23:653-7.
12. Corey A, Al-Khalidi H, Brezovic C, et al. Azimilide pharmacokinetics and pharmacodynamics upon multiple oral dosing. Biopharm Drug Dispos 1999;20:59-68.
13. Fermini B, Jurkiewicz KN, Jow B, et al. Use-dependent effects of class III antiarrhythmic agent NE-10064 (azimilide) on cardiac repolarization: block of delayed rectifier potassium and L-type calcium currents. J Cardiovasc Pharmacol 1995;26:259-71.
14. Groh WJ, Gibson KJ, Maylie JG. Comparison of the rate-dependent properties of the class III antiarrhythmic agents azimilide (NE-10064) and E-4031: considerations on the mechanism of reverse rate-dependent action potential prolongation. J Cardiovasc Electrophysiol 1997;8:529-36.
15. Brooksby B, Levi AJ, Jones JV. The electrophysiological characteristics of hypertrophied ventricular myocytes from the spontaneously hypertensive rat. J Hypertens 1993;11:611-22.
16. Winterton SJ, Turner MA, O'Gorman DJ, Flores NA, Sheridan DJ. Hypertension causes delayed conduction in human and guinea-pig myocardium: accentuation during ischaemic perfusion. Cardiovasc Res 1994;28:47-54.
17. Nand V, Doggrell SA. Effects of BDF 9148 on the action potentials and contractions of left ventricles from normo- and hypertensive rats. Clin Exp Pharmacol Physiol 1999;26:212-20.
18. Roden DM. Antiarrhythmic drugs. In: Hardman JG, Limbird LE, Molinoff PB, Ruddon RW, Gilman AG, eds. Goodman's and Gilman's the pharmacological basis of therapeutics. 9th ed. New York: McGraw-Hill, 1996:839-74.
19. Drexler AP, Micklas JM, Brooks RR. Suppression of inducible ventricular arrhythmias by intravenous azimilide in dogs with previous myocardial infarction. J Cardiovasc Pharmacol 1996;28:848-85.
20. Nattel S, Liu L, St-Georges D. Effects of the novel antiarrhythmic agent azimilide on experimental atrial fibrillation and atrial electrophysiological properties. Cardiovasc Res 1998;37:627-35.
21. Camm AJ, Karam R, Pratt CM. The azimilide post-infarct survival evaluation (ALIVE). Am J Cardiol 1998;81:35D-9D.
22. Gidh-Jain M, Huang B, Jain P, el-Sherif N. Differential expression of voltage-gated K+ channel genes in left ventricular remodeled myocardium after experimental myocardial infarction. Circ Res 1996;79:669-75.
23. Cerbai E, Barbieri M, Li Q, Mugelli A. Ionic basis of action potential prolongation of hypertrophied cardiac myocytes isolated from hypertensive rats of different ages. Cardiovasc Res 1994;28:1180-7.
24. Yokoshiki H, Kohya T, Tomita F, et al. Restoration of action potential duration and transient outward current by regression of left ventricular hypertrophy. J Mol Cell Cardiol 1997;29:1331-9.
25. Doggrell SA, Nand V. The effects of clofilium on ventricles from normo- and hypertensive rats [Abstract]. Proc ASCEPT 1999;6:186.
Keywords:

Azimilide; Positive inotropism; SHR hypertrophy/failure

© 2000 Lippincott Williams & Wilkins, Inc.