Journal Logo


Frequency Dependence in the Action of the Class III Antiarrhythmic Drug Dofetilide Is Modulated by Altering L-Type Calcium Current and Digitalis Glucoside

Gjini, Viktor; Schreieck, Jürgen; Korth, Michael*; Weyerbrock, Sonja; Schömig, Albert; Schmitt, Claus

Author Information
Journal of Cardiovascular Pharmacology: January 1998 - Volume 31 - Issue 1 - p 95-100
  • Free


Most class III antiarrhythmics show a reversed rate-dependent effect, rendering their antiarrhythmic profile clinically unfavorable (1,2). This implies that action-potential duration (APD) prolongation is pronounced at slow heart rates, leading to an increased proarrhythmic risk, but strongly attenuated at rapid rates, rendering these drugs less likely successfully to control tachyarrhythmias. This phenomenon is related to a selective IKr block (3-6), although such IKr blockers were shown to block the channel responsible for IKr, not in a reversed rate-dependent manner (7,8). Carmeliet (9) demonstrated that dofetilide preferentially blocks the open IKr channel, and recently Spector et al. (10) demonstrated open channel blockade of HERG with another selective IKr blocker (MK-499). An accumulation of not completely deactivated IKs(11) and a lower binding rate of IKr blockers at increased interstitial potassium concentrations at rapid heart rates (12) have been proposed to explain the reversed rate-dependent effect of IKr blockers on APD. Nevertheless, this phenomenon is not fully elucidated. Reactivation of L-type calcium currents is thought to be involved in the generation of early afterdepolarizations (13), which can be induced by slow heart rates and IKr block (2,14,15). We supposed that the pronounced APD at slow heart rates in the presence of IKr blockers may also depend on L-type calcium currents. Therefore we hypothesized that a block of L-type calcium currents can reduce this pronounced APD prolongation of pure class III antiarrhythmics at slow heart rates. We tested this hypothesis with the selective IKr blocker dofetilide, a new class III antiarrhythmic agent, whose effect has been previously found to be reversed rate dependent (3,9,14-17), and combined it with an inhibitor (diltiazem) or a promotor (Bay K 8644) of the L-type calcium current. Additionally, we used a digitalis glucoside (dihydroouabain) to investigate the effects of another clinically important treatment, which affects the calcium system, on the class III antiarrhythmic effect of dofetilide.


Guinea pigs of either sex weighing 250-350 g were killed by cervical dislocation. Right ventricular papillary muscles (0.5-0.8 mm in diameter) were rapidly excised from the isolated heart and mounted in a two-chambered vessel with internal circulation of the bath solution (50 ml). The bath solution containing (in mM) NaCl, 123; KCl, 5.4; CaCl2, 1.8; MgCl2, 1.0; NaH2PO4, 0.42; NaHCO3, 22; and glucose, 11 was continuously gassed with 95% O2 and 5% CO2, resulting in a pH of 7.4. The temperature was maintained at 36 ± 0.5°C. Furthermore, 30 nM tetrodotoxin was added to the bath solution to suppress the stimulation-evoked release of endogenous catecholamines from sympathetic nerve endings (18). The papillary muscles were fixed at the bottom of the chamber, and the free tendinous end was connected via a small stainless steel hook to an inductive force transducer, allowing resting force to be kept constant at 4 mN. The preparations were stimulated at 0.5 Hz by bipolar platinum electrodes with square-wave pulses of 1 ms and an amplitude slightly above threshold. Transmembrane potentials were recorded with conventional glass microelectrodes filled with 3 M KCl solution (tip resistance of 15-30 MΩ) and with an amplifier providing capacity compensation (model 773; World Precision Instruments, Sarasota, FL, U.S.A.). The data were digitized with 12 kHz and stored on a digital audiotape (DAT-recorder DTR-1202; Biologic, Claix/Grenoble, France) for later analysis. The following parameters of the AP were evaluated: resting potential, AP amplitude, and APD at 90% of repolarization (APD90).

The following experimental protocol was used: AP parameters were first evaluated under control conditions after a 30-min period of equilibration at 0.5 Hz. Muscles were then stimulated at 1, 2, and 3 Hz until steady-state conditions were achieved. Dofetilide was then added to the bath solution, and the same stimulation protocol was repeated after 30 min. Thereafter, 10 μM diltiazem, 0.1 μM Bay K 8644, or 10 μM dihydroouabain was added to the bath solution, and the same stimulation protocol as outlined above for control conditions was repeated 30 min after addition of diltiazem or dihydroouabain and 20 min after addition of Bay K 8644. The effects of diltiazem, Bay K 8644, and dihydroouabain also were tested in the absence of dofetilide. When Bay K 8644 was used, the experimental setup was darkened to reduce light-sensitive decomposition.

Dofetilide (kindly provided by Pfizer, Sandwich, England) was dissolved in distilled water acidified to a pH of 3 by addition of HCl. Appropriate portions of this solution were added to the bath solution to achieve the final concentration of 10 nM.

The concentration of dofetilide (10 nM) was chosen to achieve a potent IKr blockade without affecting any other current. This concentration of dofetilide induced >50% inhibitor of IKr in guinea pig ventricular myocytes (9) and induced a significant AP prolongation of ∼20% in guinea pig papillary muscles (14), which is in the range of clinical importance.

The concentration of diltiazem (10 μM) was chosen to achieve a potent ICa blockade, which significantly reduces APD in control conditions (19,20). For blockade of ICa, we did not use dihydropyridines or verapamil, because they affect potassium currents such as Ito and IK(21,22). We are aware that diltiazem may have some effects on potassium currents as well, but these effects are less than those of the previously mentioned calcium antagonists.

The concentration of Bay K 8644 (0.1 μM) was chosen to achieve a potent ICa increment, which prolongs APD in control conditions (23). This concentration is near the median effective concentration (EC50) value for stimulation of ICa. Blockade of potassium currents is negligible at this concentration (21,22).

We used dihydroouabain, because for this digitalis glucoside, the time for equilibration is short in Tyrode's solution. We used a concentration of 10 μM to achieve a prominent inotropic effect [EC50 for dihydroouabain in guinea pig papillary muscle was 14 μM(24)] without inducing major toxic effects such as arrhythmias.

Statistical analysis

The results are presented as means ± SEM. Comparison within groups was performed by using paired two-tailed Student's t test. A p value <0.05 was considered statistically significant. To describe the rate-dependent effect of dofetilide, the prolongation of the APD90 also was analyzed as a function of the diastolic interval (i.e., the paced cycle length minus APD90) by using linear regression (5). A value of p < 0.05 was considered significant.


Diltiazem and Bay K 8644 had no significant effect on resting membrane potential and AP amplitude (data not shown). On the other hand, dihydroouabain slightly depolarized resting potential from −87 ± 1.1 mV (mean ± SEM) to −83 ± 1.1 mV at 0.5 Hz and from −83 ± 1.2 mV to −80 ± 1.3 mV at 3-Hz stimulation frequency. Additionally ouabain reduced AP amplitude from 120 ± 0.6 to 118 ± 0.7 mV at 1 Hz and from 117 0.8 to 115 ± 0.9 mV at 3-Hz stimulation frequency.

Table 1 shows the effect of all single drugs used in this study on APD90 under control conditions. It can be seen that the APD-prolonging effect of dofetilide is markedly attenuated by increasing the stimulation frequency. APD90 was markedly prolonged by 51 ± 6 ms at 0.5 Hz and only by 21 ± 3 ms at 3 Hz. The linear regression between diastolic interval (DI) and ΔAPD90 was significantly correlated (r = 0.57; p < 0.0001). Diltiazem shortened the APD, and Bay K 8644 prolonged it, independent of stimulation frequency. Dihydroouabain induced a slight shortening in the APD with a tendency to a rate-dependent effect.

Effect of the drugs on action-potential duration

In the presence of 10 μM diltiazem, the APD90-prolonging effect of dofetilide was reduced preferentially at relatively low frequencies (Fig. 1). Interestingly, the effect of dofetilide was not influenced at 3 Hz. The linear regression between DI and ΔAPD90 was not significantly correlated (r = 0.28; p = 0.08). This means that APD prolongation is not significantly diminished by increasing the stimulation frequency. Thus the effect of dofetilide in the presence of the calcium channel blocker should be considered rate independent.

FIG. 1
FIG. 1:
Left: Effect of 10 nM dofetilide and 10 nM dofetilide with 10 μM diltiazem on action-potential duration at 90% repolarization (APD90) at different stimulation frequencies in isolated guinea pig right papillary muscles. Shown are mean values ± SEM. *p < 0.01, dofetilide vs. control; #p < 0.05, dofetilide vs. dofetilide + diltiazem. Right: Superimposed original action-potential recordings at 0.5- and 3-Hz stimulation frequencies.

Quite opposite results were obtained after the addition of 0.1 μM Bay K 8644 (Fig. 2). The calcium channel agonist further prolonged APD90 at low frequencies but had almost no effect at 3 Hz. The regression coefficient between DI and ΔAPD90 was increased to r = 0.65 (p < 0.0001), indicating that in the presence of Bay K 8644, the effect of dofetilide is more reversed rate dependent.

FIG. 2
FIG. 2:
Left: Effect of 10 nM dofetilide and 10 nM dofetilide with 0.1 μM Bay K 8644 on action-potential duration at 90% repolarization (APD90) at different stimulation frequencies in isolated guinea pig right papillary muscles. Shown are mean values ± SEM. *p < 0.01, dofetilide vs. control; #p < 0.05, dofetilide vs. dofetilide + Bay K 8644. Right: Superimposed original action-potential recordings at 0.5- and 3-Hz stimulation frequencies.

The addition of dihydroouabain in the presence of dofetilide leads to strikingly different effects than seen in control conditions (Fig. 3). Dihydroouabain further prolonged the AP in the presence of dofetilide at slow frequencies (0.5 and 1 Hz) but shortened APD at more rapid frequencies (2 and 3 Hz). Therefore the regression coefficient between DI and ΔAPD90 was clearly increased to r = 0.72 (p < 0.0001), resulting in a more pronounced reversed frequency-dependent effect of dofetilide in the combination with dihydroouabain.

FIG. 3
FIG. 3:
Left: Effect of 10 nM dofetilide and 10 nM dofetilide with 10 μM dihydroouabain (DHouabain) on action-potential duration at 90% repolarization (APD90) at different stimulation frequencies in isolated guinea pig right papillary muscles. Shown are mean values ± SEM. *p < 0.01, dofetilide vs. control; #p < 0.05, dofetilide vs. dofetilide + dihydroouabain. Right: Superimposed original action-potential recordings at 0.5- and 3-Hz stimulation frequencies.

The original AP recordings (Figs. 1-3) demonstrate that all effects of dofetilide, of dofetilide and calcium current modulation, or dofetilide and dihydroouabain shown in our study are primarily induced by a modulation of the duration of the plateau phase of the APs, whereas the terminal repolarization phase was not or was little affected. During all experiments, no afterdepolarizations were observed.


The main finding of our study is that the frequency-dependent effect of dofetilide, a new class III antiarrhythmic agent that prolongs APD by selectively blocking (IKr), can be modulated by altering L-type calcium currents. In summary, the reversed frequency-dependent profile in the AP-prolonging effect of dofetilide was more pronounced by increasing L-type calcium currents with Bay K 8644 or by the use of dihydroouabain but was prevented by decreasing the calcium currents with diltiazem.

Dofetilide was previously reported to have a frequency-independent effect on APD (25-27). However, aside from the data of this and our recent studies (16), there is growing evidence for a reversed rate-dependent class III effect of dofetilide, especially when rapid stimulation frequencies were used in studies with guinea pigs and with humans (3,9,14-17). A selective IKr block as a principal mechanism for a class III antiarrhythmic action seems to be related to an reversed rate-dependent effect on APD (4,5,28).

The pronounced AP prolongation at a slow heart rate of class III antiarrhythmics has been claimed as a proarrhythmic mechanism for class III antiarrhythmics, leading to polymorphic ventricular tachycardia (1,2). This can be the result of increased dispersion of APD or conducted early afterdepolarizations (29).

A reduction in L-type calcium currents with diltiazem prevents the pronounced AP prolongation at slow heart rates, whereas the class III effect at more rapid heart rates was preserved, resulting in a frequency-independent effect. This is surprising, regarding the frequency-independent AP shortening induced by diltiazem alone. So the effects of diltiazem and dofetilide are not additive, suggesting different frequency-dependent interactions.

On the other hand, the calcium channel agonist Bay K 8644 markedly increased the class III effect of dofetilide at slow heart rates, in contrast to rapid heart rates. Similar to diltiazem, Bay K 8644 without dofetilide had a frequency-independent effect, suggesting different frequency-dependent interactions with the effect of dofetilide.

Although the effects of Bay K 8644 and diltiazem on potassium currents is small at the concentrations used in our study (21,22), we cannot exclude that effects on other currents aside from ICa are involved in the effects on APs in our study. However, evidence for a primary ICa modulation as a mechanism for the observed effects comes from the clearly diverse effects of Bay K 8644 and diltiazem on APD with and without the presence of dofetilide.

Therefore, aside from other explanations for the reversed rate-dependent effect of the IKr blocker, such as accumulation of nondeactivated IKs(3,11) or of extracellular potassium (12,16), our findings point to the role of pronounced L-type calcium currents at slow stimulation frequencies for the reversed rate dependence of such class III antiarrhythmics. Reversed rate dependence seems to be a multifactorial phenomenon.

A possible explanation for the effect of calcium channel modulation can be that a reactivation of the L-type calcium current becomes possible during slow heart rates (30) and during block of repolarizing potassium currents since small changes in inward currents are able to induce large changes in APD during the end of the plateau phase (30). This phenomenon may be responsible for the more pronounced effect of L-type calcium modulation on the class III effect at slow heart rates. Although this phenomenon was found to induce early afterdepolarizations (13), we did not observe any early afterdepolarizations in the presence of either Bay K 8644 or dihydrooubain. This may be caused by the low concentration of dofetilide or the low concentration of Bay K 8644, which induced early afterdepolarizations in a concentration ≥0.1 μM in sheep Purkinje fibers (13). Another explanation for the absence of early afterdepolarizations may the use of papillary muscles without Purkinje fiber cells, because these muscles are less vulnerable to early afterdepolarizations than are Purkinje fibers (31).

The lesser degree of APD variation at more rapid heart rates by calcium current modulation remains unclear. Our finding of a more pronounced reversed frequency-dependent AP prolongation of dofetilide in the presence of a digitalis glucoside (dihydroouabain) are in line with the possibility that an increased intracellular calcium concentration contributes to the observed effects of Bay K 8644. So it can be speculated that activation of calcium-dependent inward currents at slow stimulation frequencies [for example, the Na+/Ca2+ exchange (30)] or calcium-dependent activation of outward currents at more rapid frequencies [for example, IKs(32)] may contribute to a prolongation at slow stimulation frequencies and a shortening at more rapid frequencies, respectively. However, the effect of dihydroouabain resulting from the block of the Na+K+ pump current is much more complex. To prove this hypothesis, control of intracellular calcium is required, which could not be determined in our model of the isolated papillary muscle.

Nevertheless, our results have particular implications for the clinical use of a combination of class III antiarrhythmic drugs and calcium-modulating drugs. Our findings point out the possibility of improving the class III antiarrhythmic profile by an additional calcium channel block. Indeed calcium channel blockers suppress early afterdepolarization-related tachycardia in relation to QT-interval prolongation (33). This may be an advantageous impact of drugs like amiodarone or azimilide, whose effects combine calcium channel block and class III effects (34,35) and may contribute to the frequency-independent effect of amiodarone (36).

Because of the different currents responsible for repolarization in the guinea pig and humans, a extrapolation of our conclusions to humans should be made with caution. Our conclusion, that L-type calcium block improves the class III antiarrhythmic profile of dofetilide, is based on the theoretic concept of class III antiarrhythmics (1,2). Whether this conclusion is valid in the clinical setting remains to be determined. Nevertheless, our data point to an additional mechanism, the L-type calcium current, for the multifactorial phenomenon of reversed rate dependence of specific IKr blockers as class III antiarrhythmic drugs.


1. Hondeghem L, Snyders D. Class III antiarrhythmic agents have a lot of potential but a long way to go. Circulation 1990;81:686-90.
2. Roden DM. Current status of class III antiarrhythmic therapy. Am J Cardiol 1993;72:44-9B.
3. Jurkiewicz N, Sanguinetti M. Rate-dependent prolongation of action potentials by a methanesulfonanilide class III antiarrhythmic agent: specific block of rapidly activating delayed K+ current by dofetilide. Circ Res 1993;72:75-83.
4. Schmitt C, Brachmann J, Karch M, et al. Reverse use-dependent effects of sotalol demonstrated by recording monophasic action potentials of the right ventricle. Am J Cardiol 1991;68:1183-7.
5. Sager P, Nademanee K, Antimisiaris M, et al. Antiarrhythmic effects of selective prolongation of refractoriness: electrophysiologic actions of sematilide HCl in humans. Circulation 1993;88:1072-82.
6. Gintant G. Two components of delayed rectifier current in canine atrium and ventricle: does Iks play a role in the reverse rate dependence of the class III agents? Circ Res 1996;78:26-37.
7. Carmeliet E. Use-dependent block of the delayed K+ current in rabbit ventricular myocytes. Cardiovasc Drugs Ther 1993;7:599-604.
8. Ohler A, Amos GJ, Wettwer E, Ravens U. Frequency-dependent effects of E-4031, almokalant, dofetilide and tedisamil on action potential duration: no evidence for "reverse use dependent" block. Naunyn Schmiedebergs Arch Pharmacol 1994;349:602-10.
9. Carmeliet E. Voltage- and time-dependent block of the delayed K+ current in cardiac myocytes by dofetilide. J Pharmacol Exp Ther 1992;262:809-17.
10. Spector P, Curran M, Keating M, Sanguinetti M. Class III antiarrhythmic drugs block HERG, a human cardiac rectifier K+ channel: open-channel block by methanesulfonanilides. Circ Res 1996;78:499-503.
11. Sanguinetti M, Jurkiewicz N. Two components of cardiac delayed rectifier K+ current: differential sensitivity to block by class III antiarrhythmic agents. J Gen Physiol 1990;96:195-215.
12. Yang T, Roden DM. Extracellular potassium modulation of drug block of IKr: implication for torsade de pointes and reverse use-dependence. Circulation 1996;93:407-11.
13. January CT, Riddle JM, Salatta JJ. A model for early afterdepolarizations: induction with the Ca2+ channel agonist Bay K 8644. Circ Res 1988;62:563-71.
14. Tande PM, Bjornstad H, Yang T, Retsum H. Rate-dependent class III antiarrhythmic action, negative chronotropy, and positive inotropy of a novel IK blocking drug, UK-68,798: potent in guinea pig but no effect in rat myocardium. J Cardiovasc Pharmacol 1990;16:401-10.
15. Gwilt M, Arrowsmith J, Blackburn K, et al. UK-68,798: a novel, potent and highly selective class III antiarrhythmic agent which blocks potassium channels in cardiac cells. J Pharmacol Exp Ther 1991;256:318-24.
16. Gjini V, Korth M, Schreieck J, Weyerbrock S, Schömig A, Schmitt C. Differential class III antiarrhythmic effects of ambasilide and dofetilide at different extracellular potassium and pacing frequencies. J Cardiovasc Pharmacol 1996;28:314-20.
17. Sager P. Frequency-dependent electrophysiologic effects of dofetilide in humans. Circulation 1995;92(suppl I):774.
18. Honerjäger P. The contribution of Na channel block to the negative inotropic effect of antiarrhythmic drugs. Basic Res Cardiol 1986;81(suppl 1):33-7.
19. Morad M, Tung L, Greenspan AM. Effect of diltiazem on calcium transport and development of tension in heart muscle. Am J Cardiol 1982;49:595-601.
20. Klitzner TS, Chen F, Raven RR, Wetzel GT, Friedman WF. Calcium current and tension generation in immature mammalian myocardium: effects of diltiazem. J Mol Cell Cardiol 1991;23:807-15.
21. Hume JR. Comparative interactions of organic Ca++ channel agonists with myocardial Ca++ and K+ channels. J Pharmacol Exp Ther 1985;234:134-40.
22. Gotoh Y, Imaizumi Y, Watanabe M, Shibata EF, Clark RB, Giles WR. Inhibition of transient outward K+ current by DHP Ca2+ antagonists and agonists in rabbit cardiac myocytes. Am J Physiol 1991;260:H1737-42.
23. Rampe D, Anderson B, Rapien-Pryor V, Li T, Dage RC. Comparison of the in vitro and in vivo cardiovascular effects of two structurally distinct Ca++ channel activators, Bay K 8644 and FPL 64176. J Pharmacol Exp Ther 1993;265:1125-30.
24. Reiter M. The use of the isolated papillary muscle for the evaluation of positive inotropic effects of cardioactive steroids. In: Greeff K, ed. Cardiac glycosides. Part I: Experimental pharmacology. Berlin: Springer-Verlag, 1981:153-9.
25. Knilans TK, Lathrop DA, Nánási PP, Schwartz A, Varró A. Rate and concentration-dependent effects of UK-68,789, a potent new class III antiarrhythmic, on canine Purkinje fibre action potential duration and Vmax. Br J Pharmacol 1991;103:1568-72.
26. Sedgwick ML, Rasmussen HS, Cobbe SM. Effects of the class III antiarrhythmic drug dofetilide on ventricular monophasic action potential duration and QT interval dispersion in stable angina pectoris. Am J Cardiol 1992;70:1432-7.
27. Yuan S, Wohlfart B, Rasmussen HS, Olsson S, Blomström-Lundqvist C. Effect of dofetilide on cardiac repolarization in patients with ventricular tachycardia: a study using simultaneous monophasic action potential recordings from two sites in the right ventricle. Eur Heart J 1994;15:514-22.
28. Baskin E, Lynch J. Comparative effects of increased extracellular potassium and pacing frequency on the class III activities of methansulfonanilide IKr blockers dofetilide, d-sotalol, E-4031, and MK-499. J Cardiovasc Pharmacol 1994;24:199-208.
29. Surawicz B. Electrophysiologic substrate of torsade de pointes: dispersion of repolarization or early afterdepolarizations. J Am Coll Cardiol 1989;14:172-84.
30. Carmeliet E. Mechanisms and control of repolarization. Eur Heart J 1993;14(suppl H):3-13.
31. Li ZY, Maldonado C, Zee-Cheng C, Hiromasa S, Kupersmith J. Purkinje fibre-papillary muscle interaction in the genesis of triggered activity in a guinea pig model. Cardiovasc Res 1992;26:543-8.
32. Hiraoka M, Furukawa T, Sawanobori T, Yamane T. Regulation of delayed rectifier K+ channels and ATP-sensitive K+ channels by intracellular factors. In: Vereecke J, van Bogaert P, Verdonck F, eds. Potassium channels in normal and pathological conditions. Leuven, Belgium: University Press, 1995:103-17.
33. Jackman WN, Szabo B, Friday JK, et al. Ventricular tachyarrhythmias related to early afterdepolarizations and triggered firing: relationship to QT interval prolongation and potential therapeutic role for calcium channel blocking agents. J Cardivasc Electrophysiol 1990;1:170-95.
34. Lubic SP, Nguyen KP, Dave B, Giacomini J. Antiarrhythmic agent amiodarone possesses calcium channel blocker properties. J Caridovasc Pharmacol 1994;24:707-14.
35. Fermini B, Jurkiewicz NK, Jow B, et al. Use-dependent effects of the class III antiarrhythmic agent NE-10064 (azimilide) on cardiac repolarization: block of delayed rectifier potassium and L-type calcium current. J Cardiovasc Pharmacol 1996;26:259-71.
36. Sager P, Uppal P, Follmer C, Antinisiaris M, Pruitt C, Singh B. Frequency-dependent electrophysiologic effects of amiodarone in humans. Circulation 1993;88:1063-71.

Calcium current; Dofetilide; Digitalis glucoside

© Lippincott-Raven Publishers