Rotigaptide (ZP123) Reverts Established Atrial Conduction Velocity Slowing

Abstract

Rotigaptide (ZP123) increases gap junction intercellular communication (GJIC) and prevents stress-induced cardiac conduction velocity (CV) slowing. However, the effect of rotigaptide on established cardiac conduction slowing and the duration of effect on rotigaptide during washout is unknown. Metabolic stress (induced by superfusion with nonoxygenated glucose-free Tyrodes buffer) was associated with a 30% decrease in atrial CV in vehicle-treated rat atria. Rotigaptide treatment initiated after a period of 30 minutes of metabolic stress produced a rapid and significant increase in CV compared to vehicle-treated time controls. During washout of rotigaptide for 30 min (while subjected to metabolic stress), there was a minor decrease in atrial CV; however, this was not significantly different from atrial CV in a rotigaptide-treated time control group. Rotigaptide treatment rapidly normalizes established conduction slowing in atria subjected to metabolic stress. However, the cessation of effect was considerably slower than the onset of action.

Keywords:

• conduction velocity
• atrial
• antiarrhythmic agents
• gap junctions

INTRODUCTION

Rotigaptide (formerly known as ZP123) is a synthetic analogue of the antiarrhythmic peptide (AAP) (1). Rotigaptide increases gap junction intercellular communication (GJIC) in isolated pairs of ventricular cardiomyocytes (2). This effect on cardiac GJIC is considered to be the mechanism whereby rotigaptide prevents acidosis-induced ventricular conduction slowing (3), metabolic stress-induced atrial conduction slowing (4), and pacing-induced reentry ventricular tachycardia in dogs with acute myocardial infarction (2).

The studies examining the effects of rotigaptide on cardiac conduction have focused on the preventive effect of rotigaptide on cardiac conduction slowing when administered prior to and during various types of stress (acidosis or glucose-deprivation combined with hypoxia). However, it is unknown whether rotigaptide can normalize cardiac conduction in cardiac tissue with established conduction slowing. The first aim of the present study was to investigate whether rotigaptide could rescue established cardiac conduction slowing, and if so, to characterize the onset and the time course of such an effect. Based on the observed antiarrhythmic effect of rotigaptide in studies where rotigaptide treatment was initiated after a period of ischemia (dogs with myocardial infarction (2) or myocardial infarction and reperfusion (5)), we hypothesised that rotigaptide would be able to revert established conduction slowing. The second aim of this study was to examine the duration of the effect of rotigaptide after washout of rotigaptide to characterize the cessation of the effect, as this has not been studied previously.

MATERIALS AND METHODS

Isolated Atrial Strip Preparation

Male Sprague Dawley rats (Taconic, Lille Skensved, Denmark) weighing 350–450 g were sacrificed by a quick blow to the neck. The left atrium was carefully dissected and a tissue sample of approximately 2× 6 mm with clearly visible longitudinal fiber direction was taken from the atrial appendage and placed with the endocardial surface upwards in a tissue chamber (volume 5 ml), (Steiert Organ Bath, Hugo Sach Electronic, March-Hugstetten, Germany). The chamber was perfused throughout the study with 37°C oxygenated Tyrodes buffer at a rate of 10 ml/min using a roller-pump. The buffer contained (in mM): NaCl 136, KCl 4.0, MgCl₂ 0.80, CaCl₂ 1.8, HEPES 5.0, MES 5.0, glucose 10, pH 7.3. The atrial tissue was fixed to two hooks in the chamber. One hook was connected to a force-transducer mounted on a micrometer screw built into the tissue chamber and the other was fixed in the tissue chamber. In all experiments the atrial tissue was allowed a 30 min equilibration period before stimulation was started.

Measurement of Atrial Conduction Velocity

Atrial CV was recorded continuously using a simple two-electrode technique (see Figure 1). A bipolar stimulation electrode (Teflon coated stainless steel, diameter 75 μ m) was placed at one end of the tissue. Stimulation was performed at 1 Hz using rectangular pulses at double threshold (stimulus width: 0.2 ms). The stimulus was delivered by a stimulator (Hugo Sachs type 215) through an isolation unit (Hugo Sachs UISO unit type 263). Two separate microelectrodes of pure iridium (World Precision Instruments, Sarasota, Florida, US, tip-impedance 3.5–4.0 MΩ) were placed on a line along the long-axis of the preparation for recording of atrial CV. Each microelectrode was connected to a Hugo Sachs amplifier (BPA module) via a head-stage preamplifier (10× amplification of the signals). The BPA module was connected to the data acquisition system through a Hugo Sachs PLUGSYS. Signals were filtered at 1 kHz and sampled at 10 kHz. The distance between the two electrodes was measured using a stereomicroscope with a calibrated ocular grid.

Figure 1. Schematic presentation of the experimental setup used for the measurement of atrial conduction (CV). Two microelectrodes were placed on a line along the long-axis of the preparation and the distance between them was measured using a stereomicroscope with a calibrated ocular grid. The extracellular action potentials were used for determination of conduction time at the two electrodes. Atrial CV was calculated as the interelectrode distance (mm) divided by the time (ms) between the peaks of the differentiated extracellular signals from the proximal and distal microelectrodes. For detailed description see the methods.

Measurement of Atrial Contractility

Tissue contractility was recorded continuously using the isometric force-transducer built into the tissue-chamber. Resting tension (preload) applied to the preparation was adjusted to a level where the developed tension in response to stimulation under control conditions was 50% of the maximally developed tension (200–500 mg). Preload adjustment was performed at the end of the equilibration-period.

Study Design and Experimental Groups

During the first part of the experiment, atrial CV and contractility were measured during perfusion with oxygenated Tyrodes buffer in two periods of 20 min each (baseline and pretreatment periods). Then, the perfusion buffer was switched to a nonoxygenated glucose-free Tyrodes buffer (metabolic stress periods) and measurement of atrial CV and contractility were continued during metabolic stress in two periods of 30 min each.

Four groups (n = 7–8/group) were studied (see Figure 2). The first group received vehicle treatment throughout the entire study and served as an untreated time control group. A second group received vehicle treatment during the first metabolic stress period and rotigaptide treatment during the second metabolic stress period. This was the rotigaptide rescue group. A third group received rotigaptide treatment during the 20 min pretreatment period and during the entire 60 min of metabolic stress. This group served as a rotigaptide–treated time control group. The fourth group also received rotigaptide treatment during the pretreatment period and during the first metabolic stress period. However, during the second metabolic stress period rotigaptide treatment was stopped to study the duration of the effect during washout. This was the rotigaptide washout group.

Figure 2. Study design and treatment in the four experimental groups.

Drugs

Rotigaptide (proposed INN name for the drug formerly known as ZP123) was synthesized at Bachem AG (Bubendorf, Switzerland) for Zealand Pharma A/S. Concentration of rotigaptide was 10 nM in all groups and in all of the experimental periods.

Data Analysis

The differentiated extracellular action potentials used for analysis of conduction time were calculated beat-by-beat using a personal computer with the Notocord v3.5 software (Notocord Systems, Croissy sur Seine, France). Atrial CV was calculated as the interelectrode distance (mm) divided by the time (ms) between the peaks of the differentiated extracellular signals from the proximal and distal microelectrodes. Atrial CV was averaged for each of the one hundred consecutive 60 seconds period. Atrial maximal contractile force (dF/dt_max) was calculated using the analysis modules in the Notocord software.

Statistics

Overall differences among groups were analyzed for the four experimental periods using a 2-way analysis of variance (ANOVA) for repeated measurements. If a significant difference among groups were detected, a posthoc analysis (Fisher's LSD test) was performed. Moreover, for the two 30 min periods of metabolic stress, differences in the changes in CV over time (time-group interaction) were analyzed between the vehicle time control and the rotigaptide rescue group, as well as between the rotigaptide time control and rotigaptide washout. For all comparisons, p<0.05 was considered statistically significant. Data are expressed as mean ± SEM.

RESULTS

Rotigaptide Rapidly Increased CV in Atria With Established CV Slowing

Table 1 shows the absolute CV at the end of the four periods in the four different experimental groups. There were no statistically significant differences in the absolute atrial CV among groups at baseline at the end of the pretreatment period. However, in order to compare the magnitude of change in CV in response to metabolic stress among the four groups, the absolute CV values were normalized to CV at t = 0 min (just prior to metabolic stress was started). Figure 3 shows the relative CV over time in the four experimental groups. During the baseline and pretreatment periods atrial CV was stable and there were no differences between the groups. Thus, in accordance with previous observations (4), rotigaptide had no effect on CV during physiological conditions.

Figure 3. Atrial conduction velocity (CV) during the baseline, the pretreatment, and the two metabolic stress periods in the four experimental groups. Values are relative to t = 0 min. Data are mean ± SEM, n = 7–8/group.

TABLE 1 Absolute conduction velocity (m/sec) at the end of the four experimental periods in the four treatment groups

Group	Baseline	Pretreatment	Metabolic stress period 1	Metabolic stress period 2
Vehicle time control	0.70 ± 0.03	0.72 ± 0.03	0.59 ± 0.04	0.53 ± 0.05
Rotigaptide time control	0.68 ± 0.02	0.69 ± 0.02	0.69 ± 0.05	0.68 ± 0.02
Rotigaptide rescue	0.65 ± 0.01	0.66 ± 0.01	0.54 ± 0.02	0.64 ± 0.03
Rotigaptide washout	0.69 ± 0.02	0.66 ± 0.02	0.67 ± 0.04	0.60 ± 0.05

*p < 0.05 vs pretreatment level within the same treatment group (2-way ANOVA). Data are mean ± SEM, n = 7–8/group.

Whereas atrial CV was stable during physiological conditions, changing the buffer to the nonoxygenated glucose-free buffer had significant effects on conduction. Thus, during the 60 min of metabolic stress, CV decreased significantly in the vehicle time control group reaching 78 ± 4.9% and 70 ± 6.8% of baseline CV after 30 and 60 min, respectively. In the rotigaptide rescue group, CV also decreased significantly during the first 30 min of metabolic stress in the absence of rotigaptide (reaching 82 ± 2.1% of baseline CV after 30 min). Importantly, this was not different from the vehicle time control group during the same period (p = 0.30). During the following 30 min of metabolic stress (in the presence of rotigaptide), atrial CV increased significantly in the rotigaptide rescue group reaching 96 ± 5.4% of baseline CV. Overall, CV during the second 30 min period of metabolic stress was significantly different from the vehicle time control group (p = 0.016). Moreover, the change in CV over time (time-group interaction) was significantly different between the two groups (p<0.0001). In the rotigaptide rescue group, the increase in CV occurred within the first minutes after rotigaptide was added to the perfusion buffer, suggesting a rapid onset of action of rotigaptide.

The Effect of Rotigaptide on Atrial CV Was Not Reversible During 30 Min Washout

In atria treated with rotigaptide throughout the metabolic stress periods (rotigaptide time control group) CV did not change during the 60 min of metabolic stress (Figure 3). This is in accordance with previous reports and confirms that rotigaptide effectively prevents stress-induced cardiac CV slowing (4). Similarly, in the rotigaptide washout group, atrial CV was stable during the first 30 min of metabolic stress in the presence of rotigaptide and there were no differences between the two groups (p=0.84). During the second 30-min period of metabolic stress (in the absence of rotigaptide) there was a minor decrease in atrial CV. Thus, after 30 min of drug washout, atrial CV had reached 89 ± 6.9% of the baseline level. However, atrial CV during washout was not significantly different from atrial CV in the rotigaptide time control group during the same period (p=0.24). Moreover, the change in CV over time was not different between the two groups (time-group interaction, p=0.99).

Rotigaptide Had no Effect on Atrial Contractility

Figure 4 shows the changes in maximal contractility (dF/dt_max) during the 100 min experimental protocol in the four groups. There was a spontaneous rundown in dF/dt_max (∼ 0.5–1%/min) during the first 40 min. Metabolic stress led to a significant reduction in atrial contractility within the first 10–15 min. Thus, after 15 min of metabolic stress atrial contractility had decreased by ∼ 65%. There were no differences in contractility among the four groups, suggesting that rotigaptide treatment had no effect on contractility. Both the marked decrease in contractility during metabolic stress and the lack of effect of rotigaptide on contractility is in accordance with what we have reported previously (4).

Figure 4. Atrial contractility (dF/dt_max) during the entire experiment in the four experimental groups. Values are relative to dF/dt_max max at t = 0 min. Data are mean, n = 7–8/group (for the sake of clarity the SEM values are omitted in the figure).

DISCUSSION

Rotigaptide is a synthetic analogue of the endogenous AAP that was originally isolated from bovine atria (1). Similar to the AAP analogue AAP10, rotigaptide increases GJIC between isolated pairs of ventricular cardiomyocytes without affecting basal membrane currents (2, 6). The selective effect of rotigaptide on GJIC is believed to underlie the effect on cardiac conduction slowing reported by two different groups (3, 4). However, whereas the preventive effect of rotigaptide on stress-induced cardiac conduction slowing is well described, the effect of rotigaptide when administered during established conduction slowing has not been examined previously.

In the present study, we demonstrate that rotigaptide is able to rapidly revert established conduction slowing. The effect of rotigaptide occurred within a few minutes after treatment was started, and after 30 min treatment conduction velocity was normalized. To our knowledge this is the first direct demonstration that rotigaptide can revert established conduction slowing. In previous in vivo studies, it was reported that rotigaptide exerts significant antiarrhythmic effects when rotigaptide treatment was initiated after a period of myocardial ischemia. Thus, rotigaptide prevented inducible VT in dogs with acute MI in a study where treatment was initiated 1–3 hours after the onset of ischemia (2). Moreover, in dogs with ischemia-reperfusion injury, rotigaptide treatment initiated after 1 h of ischemia significantly reduced the incidence of spontaneously occurring ventricular tachyarrhythmias during reperfusion (5). The electrophysiological mechanism whereby rotigaptide exerted the antiarrhythmic effect in the two studies is unknown but the present study supports the notion that the antiarrhythmic effect of rotigaptide may have been the result of an increase in cardiac conduction velocity.

Whereas the onset of action of rotigaptide was rapid, the cessation of action appeared to be slower. During 30 min washout of the drug there was a minor decrease in CV compared to the rotigaptide-treated time control group, however the difference was not statistically significant. Little is known about the duration of the effect of the AAP-analogues during drug washout. The only study examining the effect of an AAP-analogue during washout comes from the lab of Stefan Dhein. It was reported that the effect of the AAP-analogue AAP10 on GJIC (studying pairs of isolated cardiomyocytes) was absent after 10 min washout, whereas the effect on conduction velocity in isolated papillary muscle was absent after 30 min washout (6). Thus, the lack of significant changes in CV during 30 min washout of rotigaptide is in contrast to what has been reported for AAP10 (6). Structurally, rotigaptide is closely related to AAP10; however, all L-amino acids have been substituted with D-isomers, which protect against enzymatic degradation and thereby increases the stability of the peptide (7). Therefore, increased stability of rotigaptide is likely to contribute to the longer duration of action of rotigaptide during washout relative to AAP10.

Is the effect of rotigaptide on established conduction slowing the result of an increase in GJIC? Theoretically, the metabolic stress induced cardiac conduction slowing in the current study could have been caused by a decrease in the degree of intercellular coupling, by a reduced excitability, or by a combination of both (8). Thus, the effect of rotigaptide on conduction slowing could have been exerted through either mechanism. Rotigaptide has been reported to have no effect on the shape and duration of cardiac action potential in guinea pigs, rabbits, and dogs (3, 7, 9). In double cell patch clamp experiments on pairs of cardiomyocytes, rotigaptide had no effect on the basal membrane currents. Moreover, patch clamp studies on single cardiomyocytes demonstrated that rotigaptide does not affect the cardiac sodium current (3), an important determinant of cardiac conduction. The lack of effect of rotigaptide on the cardiac sodium current is in line with the recently reported lack of specific binding of rotigaptide to the cardiac sodium channel (4). Taken together, these data suggest that the effect of rotigaptide on cardiac conduction could not have been caused by an effect on the cardiac excitability.

Gating of cardiac gap junction channels is regulated by posttranslational phosphorylation of the C-terminal tail of connexins as recently reviewed (10) and it has been reported that both AAP10 and rotigaptide increases the phosphorylation of the major cardiac gap junction protein, connexin 43 (11, 12). Recently, it was demonstrated that rotigaptide site-specifically suppresses dephosphorylation of various serine residues in the C-terminal tail of connexin43 in rat heart subjected to low-flow ischemia (13). Taken together, the effect of rotigaptide on connexin phosphorylation as well as the original observation that rotigaptide increases GJIC in cardiomyocytes (2) suggests that the effect on established conduction slowing in the present study was the result of the effect of rotigaptide on GJIC.

LIMITATIONS OF THE STUDY

The method used to record cardiac conduction velocity in the present study is rather simple as it only uses two electrodes. Compared to high-resolution methods (e.g., optical mapping and multielectrode array technique), the method does not allow studies of more complex conduction velocity parameters (e.g., conduction velocity heterogeneity). However, compared to data obtained using the optical mapping technique the data obtained in the present study is not confounded by the presence of voltage sensitive dyes and excitation-contraction uncouplers–drugs that are known to alter the cardiac conduction pattern (14, 15, 16). Indeed, the absence of excitation-contraction uncouplers allowed us to examine the effect of rotigaptide on contractility, which would have been difficult to do in an unconfounded way using optical mapping.

We used a relatively low, and for the rat, nonphysiological pacing frequency of 1 Hz in the present study. This pacing frequency was used based on our experience that it is difficult to keep the atrial preparations viable for 60 min of metabolic stress if the tissue is paced at a physiological pacing frequency.

CONCLUSION

Rotigaptide treatment reverts established conduction slowing in atria subjected to metabolic stress. Whereas the onset of action of rotigaptide is rapid, the cessation of the effect was slow.

ACKNOWLEDGEMENT

The authors would like to thank Henrik Holm-Kjar for his excellent technical assistance.