Multiple Ionic Mechanisms of Early Afierdepolarizations in Isolated Ventricular Myocytes from Guinea-pig Hearts MASAYASU HIRAOKA,u AKIHIKO SUNAMI, ZHENG FAN, AND TOHRU SAWANOBORl Depa-t of cardiovacuhr Lliseases Medical hseanh Institute Tokyo Medual and Dental Uniaenity 1-5-45, Tiushima, Bunkyo-Ku Tokyo-113, Japan
INTRODUCTION Afterpotentials can be an important mechanism for the generation of arrhythmias since they induce triggered activity under certain settings. These afterpotentials are divided into two categories depending on the timing of repolarization. One is an early afterdepolarization which is a depolarizing afterpotential developing before the end of repolarization. The other is a delayed afterdepolarization starting after the end of repolarization.2 Both types of afterpotentials have been shown to develop under various conditions, and different mechanisms among them are thought to be involved. The early afterdepolarization (EAD) is seen at various potential levels from the plateau to the final repolarization (phase 3). Because of wide ranges of voltages for its development under various conditions, an ionic mechanism of EAD may not be explained by a single ionic current. As for the ionic mechanism of EAD, two inward currents were suggested as major contributing factors. They are the L-type Ca2+ current and the window or non-inactivating Na+ curIn addition, decreased K+ conductance was also indicated to be a contributing factor for development of EAD, since Cs+ as a blocker of K+ channels were frequently used as a model of this type of arrhythmia^.^,^ Among various conditions, low external K+ concentration is known to decrease the K+ conductance and to induce EAD,2 but its actual mechanism a Addresscorrespondence to Masayasu HIRAOKA, M.D., Ph.D. Tel. 81-3-3813-6111,Ext. 6160; FAX 81-3-5684-6295.
33
34
ANNALS NEW YORK ACADEMY OF SCIENCES
has not been clarified. The alkaloid neurotoxin, veratridine, is known to modify the Na+ channel properties in many excitable membranes9 and cause EAD in heart muscles,2 while the mechanism of EAD is not known. Therefore, the ionic mechanism for low-K+-induced and veratridine-induced EADs was examined in isolated ventricular myocytes from guinea-pig hearts.
MATERIALS AND METHODS Preparations All the experiments were done with isolated ventricular myocytes from guinea-pig hearts. Our method for an enzymatic isolation procedure was described elsewhere.’O Solutions The composition of Tyrode’s solution was (mM) NaCl 144, NaH2P04 0.33, KCI 4.0, CaC12 1.8, MgClz 0.53, glucose 5.5, and HEPES 5.0, with p H adjusted to 7.3-7.4 by addition of NaOH. The K+-free solution was made up by omitting KCI from the Tyrode’s solution. The Ni2+- or Cd2+containing solutions were prepared by adding 50 pM NiC12 and/or 500 pM CdC12 to the Tyrode’s or to the K+-free solutions, respectively. Tetrodotoxin (TTX)(Sankyo Co., Tokyo) was directly dissolved in the Tyrode’s or the K+free solutions at concentrations of 30-60 pM as indicated in the text. Veratridine (Sigma Chem. Co., St. Louis, MO) was prepared as 50 mM stock solution containing 0.1% ethanol and diluted into the Tyrode’s solution as the final concentration indcated in the text.
Recding Methods and Data Analysis The patch clamp technique of whole-cell and cell-attached configurations” was used to record membrane potentials, currents and single Na channel currents. In the whole-cell experiments, a patch-clamp amplifier (Model 8900, Dagan Corp., Minneapolis, MN) was used. Details of the recording technique and the data-acquisition systems were described in our previous reports.’O Action potentials were elicited by a square-wave pulse of 2 msec duration applied through the patch-pipette in a current-clamp mode at frequency of 0.1 H z or less. When the ramp-voltage clamp method was used, an intelligent arbitrary function synthesizer (Model 1731, NF Instrument, Yokohama) was used to supply command pulses. The pipette solution con-
HIRAOKA et al.: MULTIPLE IONIC MECHANISMS OF EAD
35
tained (mM) KCI 130, K2ATP (Sigma Chem. Co., St. Louis) 5.0, creatine phosphate (Sigma Chem. Co.) 5.0, EGTA 0.1 and HEPES-KOH buffer 5.0. Recordings were made at 32-34OC. Single-channel current recorchngs were made at room temperature in the cell-attached configuration with a patch-clamp amplifier (Axopatch, Axon Inst. Inc., Burlingame, CA). Current and voltage signals were recorded with a videocassette recorder (HR-S 5500, Victor Co., Tokyo) through a PCM converting system (RP-882, NF Instruments, Yokohama). Recorded signals were filtered off-line through an eight-pole Bessel low-pass filter (48 dB/octave, FV665, NF Instruments) at 1-2 kHz and sampled a t 4-10 kHz on the dlsc of a personal computer (IBM-PSI2) using an analog-to-digital converter (TL-1 DMA interface, Axon Inst., Burlingame). Data were collected and analyzed using a software, pClamp 5 . 5 . The external solution contained (mM) K aspartate 140, NaCl 4.5, MgCh 0.5, EGTA 1.0, glucose 5.5 and HEPESKOH buffer 5.0 adjusted to p H 7.35. The composition ofthe pipette solution was (mM) NaCl 140, KCI 4.0,MgC12 1.0, CaC12 0.18, glucose 5.5, and HEPES-NaOH buffer 5.0 adjusted to p H 7.35. Veratridine (50 pM) was applied to the pipette solution.
RESULTS Early A@a?epo&rization
Induced by the K+-Eree Solution
When the myocytes were exposed to the K+-free solution under a constant stimulation rate at 0.1 Hz, resting membrane potential gradually hyperpolarized from around -85 mV in the control to - 100 mV or more negative voltage with little change in action potential repolarization initially, and then action potential duration became prolonged due to slowing of the final repolarization phase (phase 3) forming a hump. The hump eventually turned into depolarizing afterpotential (EAD) and induced triggered activity (FIG. 1,A). The appearance of humps and EAD were always seen in six examined preparations exposed to the K+-free solution. In any case, EAD started at high negative membrane potential and the maximum diastolic potential before the start of the first EAD was -68 3.4 mV (mean SD, n = 6). The depolarizing phase of EAD shifted into a rapid upstroke a t voltages between -50 and -40 mV in most of the cells. Once triggered activity was initiated in these preparations, the maximum diastolic potentials stayed around -50 mV and spontaneous firings continued as long as the K+-free solution was p e h s e d . These changes of membrane potentials were reversible upon wash-out of the K+-free solution. Changes in the background current-voltage (I-v) relation exposed to the K+-free solution were examined with a ramp voltage clamp of slow depolari-
*
*
36
ANNALS NEW YORK ACADEMY OF SCIENCES
FIGURE 1. Changes in action potentials and membrane currents after exposure to the Kt-free solution. A shows successive changes in action potentials after exposure to the K+-free solution. Initially, hyperpolarization of resting membrane potential developed with little changes in the repolarization phase. Later, prolongation of action potential started with a formation o f a hump o n the phase 3. Then, the hump eventually turned into a depolarizing afterpotential and induced triggered activity. B illustrates the background I-V relations after exposure to the Kt-free solution. Open circles indicate the I-V curve recorded in the control solution containing 4 mM K+. Numbers 1 to 4 represent the I-V curves recorded after the start of the K+-free solution in that order. The voltage protocol is indicated in the inset shown at the top left. Note that the I-V curves show a "crossover" phenomenon in the K+-fme solution and net-inward current at potentials negative to -50 rnV with inward hump between -40 and - 10 rnV. A and B were recorded from different cells.
HIRAOKA et al.: MULTIPLE IONIC MECHANISMS OF EAD
37
zation (150 mV per 30 sec) from -100 mV to +50 mV. FIGUREl,B shows successive changes in the I-V relations after the pefisate was switched from the Tyrode’s solution to the K+-free solution. Currents at negative voltages were gradually decreased and, therefore, the I-V curves showed “cross-over” phenomenon12 in successive recordings. Finally, current at potentials negative to - 50 mV became inward and a negative hump was observed at potential between -40 and - 10 mV. On the other hand, outward currents at positive voltages were increased in the K+-free solution. In six examined preparations, nearly complete suppression of the inward rectifier K+ current with a negative shift of the I-V relation, an increase in a steady inward current at the plateau level and increased outward current at positive voltages were general findings in the K+-free solution. To examine a possible involvement of inward currents in the formation of EAD, the effects of inhibitors of the Ca2+and Na+ currents were tested (FIG. 2). When the myocyte was exposed to the K+-free solution in the presence of 50 pM Ni2+ to block the T-type Caz+-current,13 the appearance of hump and EAD was not affected (A). Additional application of 500 pM CdZ+to inhibit the L-type Ca2+ current as we1113 abolished the appearance of EAD and membrane potential stayed at the plateau level during action potential repolarization, although prolongation of action potentials at the beginning of the K+-free perfusion was similarly observed as in the absence of these cations. 4 mV (n = 5) in the Cd2+The maximal diastolic potential was -32 containing K+-free solution and n o spontaneous activity was initiated. When myocytes were exposed to the K+-free solution in the presence of 30-60 pM tetrodotoxin (TTX) to block the window or non-inactivating N a + - c ~ r r e n t , ’ ~the -~~ hump and EAD were similarly induced as in the control, but triggered activity from EAD was not elicited (C). The latter finding was possibly due to a partial inhibition of the Na+ current. These results of the TTX effect were confirmed in four preparations. FIGURE3 shows the effects of the inward current inhibitors on the background I-V relations. Application of 50 pM Ni2+ did not affect the I-V relation in the K+-free condition. Further application of 500 pM Cd2+ abolished increased steady inward current between -40 and - 10 mV, and partially suppressed the outward current at positive voltages (A), while little change in the I-V relation was noted at potentials negative to -50 mV. These results were confirmed in 5 myocytes. B shows the effect of 30 pM T T X on the background I-V relation. TTX slightly depressed inward current at around -60 to -30 mV but left the steady inward current at the plateau level to remain (n = 3). The contribution of the outward current to formation of EAD was SUSpected, since EAD started around -68 mV which was too high to activate the L-type CaZ+ current, and Ni2+ or TTX could not abolish EAD. One POS-
*
38
ANNALS NEW YORIC ACADEMY OF SCIENCES
FIGURE 2. Effects of the inward current inhibitors on the low-K+-induced EAD. A shows effects of 50 pM Ni2+ on action potential after exposure to the K+-free solution. Ni2+ was contained in both the Tyrode's and K+-free solutions. B shows effects of hrther application of 500 pM Cd2+.While prolongation of action potentials were seen, no EAD was induced and the maximal diastolic potential stayed at around - 30 mV due to incomplete repolarization. C represents effect of 30 pM T T X on the appearance of EAD after exposure to the K+-free solution. The humps and EAD were similarly observed as in the absence of the inhibitors, but triggered activity was not initiated from EAD probably due to a partial block of the Na+ current. A and B were recorded from the same cell but C was recorded from a different cell.
HIRAOKA et al.: MULTIPLE IONIC MECHANISMS OF EAD
39
0 ; K-free A ; " + Ni
m;
"+Ni+Cd
0 ; K-free
Id;
"+
TTX
I/ FIGURE 3. Effects of the inward current inhibitors on the background current in the K+-free condition. A shows effects of 50 pM Ni2+ (A)and plus 500 pM Cd2+ (W) on the background I-V relation. Ni2+ had n o effects on the I-V relation compared to the control (O), while Cd2+ completely depressed negative hump of the I-V relation between -40 and - 10 mV, and partially suppressed the outward currents at positive voltages. B illustrates effects of 30 pM TTX (8)o n the background I-V relation. T T X slightly depressed the steady inward current at potentials between -60 and -20 mV, but left the negative hump in the I-V relation at the plateau voltage to remain. A and B were recorded from different cells.
ANNALS NEW YORK ACADEMY OF SCIENCES
40
sible candidate is the time-dependent decay of the delayed outward K+ current (IK).FIGURE4 presents the effects of the K+-free solution on IK and its decay upon repolarization. In the K+-free solution, not only the outward current during depolarizing steps but also the tail current amplitude were increased compared to the control, while the time course of decay was not much affected since its half-time was not changed. Increases in the tail current amplitude were observed at all the test voltages examined and these changes were reversible on the wash-out of the K+-free solution (n = 3).
Early A~poluririwttimtInduced by Veratricline When 25-100 pM veratridme was applied to the myocytes under the constant stimulation rate of 0.1 Hz, action potentials were gradually prolonged due to the slowing of the plateau phase and regenerative EADs appeared on the plateau after 8-10 minutes of application (FIG. 5). At this stage, prolongation of action potentials was so marked to be its duration of several seconds. EADs appeared regeneratively on the plateau phase increasing their amplitudes as the starting voltages became negative. Application of 500 pM Cd2+ shortened the action potential duration and decreased the amplitudes of EADs with slowed frequency of their appearance, but complete abolishment
V l p . Q ' 5 n A A J
I=O
~
A
A~
.*'
O
.0
+20
+40
O +60
+BonN
FIGURE 4. Effects of the K+-free solution on the delayed outward K+ current. A represents current records in responses to 500 msec voltage steps from a holding potential of -50 mV to +70 (q) and +30 rnV (bonmn), and subsequent repolarization to -50 mV. Open circles (0)indicate currents in the control Tyrode's solution and triangles (A)are those in the K+-free solution. B shows the tail current amplitude (ordinate) Venus test voltage (VT).
HIRAOKA cz af.: MULTIPLE IONIC MECHANISMS OF EAD
41
FIGURE 5. Veratridine-induced action potential prolongation and appearance of EAD. A shows action potential changes after application of 25 pM veratridine. Action potentials were recorded 5-10 minutes in the presence of veratridine. Progressive lengthening of action potentials and development of regenerative EADs o n the plateau phase were noted in the later records. B illustrates effects of CdZ+(A)and plus T T X (M) on action potentials and EADs. Cd2+ shortened action potential considerably, and decreased the amplitude and frequency of EADs. Further application ofTTX in the presence of veratridine shortened the action potential and completely abolished EADs. A and B were recorded from the same cell.
of EADs was not achieved (n = 3). Further application of 30 pM T T X completely eliminated residual EADs and action potential duration was further shortened. These effects of T T X developed rapidly within 2-3 minutes of application. Application of TTX alone (30-60 p M ) caused a rapid abolishment of EADs and shortened action potential duration in other two preparations. The effects of veratridine on the single Na+ channel currents were examined by the cell-attached patch recordings (FIG. 6 ) . In the control, openings of the single Na+ channel currents were seen at the beginning of pulses and were seldom after 10 msec. An ensemble average current showed a decay time
42
ANNALS NEW Y O U ACADEMY OF SCIENCES
constant of 3.6 msec. The conductance of this channel current was 16pS (n = 5). When 50 pM veratridine was applied to the pipette solution, the amplitude of the single channel currents decreased to the half or one-third of the control (5-8pS). In addition, markedly prolonged openings, sometimes lasting throughour 500 msec pulses, or delayed reopenings after 100 msec were frequently observed. Because of these prolonged openings or delayed reopenings, channel activities were seen upon repolarization to the resting potential level of - 120 mV after 500 msec step pulses. An ensemble average current decayed with very slow time constant of 609 msec, which was about 200 times slower than the control. Prolonged channel openings or delayed reopenings were seen at various voltages, once the channel was activated (FIG. 7) (n = 4).These results suggest that the m curve of the Na+ channel was shifted to the left along the voltage axis by veratridine.
DISCUSSION The present study demonstrated that EAD was induced easily and repeat-
FIGURE 6. Effects of veratridine on the single Na' channel currents. A shows records in the control and B after application of veratridine. The top indicates the voltage protocol, mirlltle tnues are single channel current sweeps and the botarm is an ensemble average current of 500 (A) and 100 (B) sweeps.
HIRAOKA et al.: MULTIPLE IONIC MECHANISMS OF EAD
43
FIGURE 7. Effects of veratridine on single Na' channel currents at different voltages. Three different test voltages are indicated at the left. Records were taken from a different cell from that shown in FIGURE6, B.
edly in ventricular myocytes by petfusing the K+-free solution or exposing to a low concentration of veratridine. The ionic mechanism of the low-K+induced EAD seems to be different from that of the veratridine-induced EAD because of differences in modes of appearance, the level of their developing voltages, and pharmacological responses. The low-K+-induced EAD was seen at high negative voltage of -68 mV with a single depolarizing afterpoten-
ANNALS NEW YORK ACADEMY OF SCIENCES
44
tial, whereas the veratridine-induced EAD initiated as repeated appearance of afterpotentials at relatively low membrane potentials during the plateau. The former type was easily blocked by Cd2+ which inhibited the L-type Ca2+ channel and partially suppressed the delayed outward K+ current, while the latter was abolished easily by T T X ; in addition Cd2+partly eliminated its appearance. Therefore, the low-K+-induced EAD was mainly produced by the L-type Ca2+ current and time-dependent decay of IK, while the veratridineinduced EAD was caused by the Na+ current with partial contribution of the Ca2+ current. The prolongation of action potential duration seems to be a common factor to predspose the appearance of EAD in both conditions, which was similar to other reported case^.^-^ In the K+-free solution, the prolongation was mainly caused by lengthening or hump formation at the phase 3 which could be explained by inhibition of the inward rectifier K+ current. This condition was similar to the Cs+-induced or quinidine-induced EAD,3.416 while the veratridine-induced EAD was produced by the altered properties of the Na+ channel as like the cases by aconitine16 or anthopleurin-A.6 The single channel recording showed that veratridine produced long-lasting openings or delayed reopenings, causing markedly delayed decay in ensemble average current. These changes in opening behaviors were seen at high negative potentials around -90 mV once the channels were opened (not shown), which indicated the m curve of the Na+ channel was shifted to negative direction along the voltage axis. In addition, veratridine decreased the conductance of the channels. These findings were similar to the veratridine effects on the Na+ channels of skeletal muscle and neuroblastoma cell^.^^^^^ The actions also resembled those of aconitine in cardiac cells as modifying the open Na+ channels. Although veratridine caused increased inward current with slow and steady decay, EADs appeared as regenerative afterpotentials repeatedly during the plateau. The reason for recycling appearance of EADs was not known, but it may be partly explained by contribution of the L-type Ca2+ current, since application of Cd2+slowed their frequencies and decreased amplitudes. It was also noted that at the plateau voltages the total membrane current was small and, therefore, small changes in outward or inward currents would cause potentials to depolarizing or hyperpolarizing directions. In this context, time- and voltage-dependent changes in the background current mainly produced by delayed outward K+ current and inward rectifier K+ current might be involved in the repeated appearance of EADs, in association with recovery from inactivation of the L-type Ca2+ c ~ r r e n t This . ~ point needs hrther clarification by additional experiments. On the contrary, the low-K+induced EAD was composed of a single depolarizing afterpotential following action potential repolarization. The initial depolarizing portion of EAD was 19320
'
HIRAOKA
et
al.: MULTIPLE IONIC MECHANISMS OF EAD
45
probably brought about by a time-dependent decay of IK, since the T- and L-type Ca2+ current were not involved, and T T X did not block it. The conditions of the present experiment are too far to be seen in clinical situations, and may be questioned as to their practical significance. Although the K+-free condition is not relevant to clinical situations, decreased K+ conductance seen in hypokalemia, acidosis or combinations of both may provide an easy appearance of EAD with acceleration of sympathetic activity to increase the Ca2+ current. There have been no reports of modification of the Na+ channel behaviors like veratridine action in any of clinical settings, and, therefore, the clinical significance of the case of the veratridine-induced EAD is not known. The study indicates that EAD is a potential change which can be produced by multiple ionic mechanisms depending on the basal conditions. This fact must always be considered for the mechanism of arrhythmias when triggered activity from EAD is suspected.
SUMMARY Ionic mechanisms of early afterdepolarization (EAD) induced by the K+fi-ee solution or veratridine were studied with guinea-pig ventricular myocytes using the patch-clamp technique of whole-cell and cell-attached patch configurations. In the K+-free solution, myocytes exhibited prolonged action potential duration with humps on the final repolarization phase, which eventually turned into EAD starting around -70 mV and induced triggered activity. Application of 0.5 mM Cd2+ inhibited the development of EAD and caused depolarization of maximum diastolic potentials around - 30 mV, although Cd2+ did not prevent prolongation of the action potential. Application of 50-100 pM Ni2+ or 30 pM tetrodotoxin had little effects on EAD and diastolic potentials. The background current-voltage relation examined by a ramp voltage clamp showed inhibition of the inward rectifier K+ current, induction of steady inward current between -40 and - 10 mV, and increase in the outward tail current upon repolarization in the K+-free solution. Cd2+completely blocked the steady inward current at the plateau level and partially depressed the delayed outward K+ current, while Ni2+ had no effects on the background I-V relation. Tetrodotoxin showed a mild inhibitory effect on the inward component of the background current negative to -50 mV, but left the steady inward current at the plateau level. Therefore, EAD in the K+-fiee condition is mainly formed by decreased inward rectifier K+ current, activation of the L-type Ca2+ current, and time-dependent decay of the delayed outward K+ current upon repolarization. Application of 25-100 pM veratridine caused marked prolongation of action potential with appearance of regenerative EADs. Action potential prolon-
46
ANNALS NEW YORK ACADEMY OF SCIENCES
gation and EADs were partially abolished by CdZ+and completely eliminated by tetrodotoxin. T h e single channel current recordings showed a decreased current amplitude, and prolonged and delayed openings of t h e N a + channel currents by veratridine. Thus, a n ensemble average current showed markedly prolonged decay time constant of 609 msec in veratridine from 3.6 msec in the control. These results indicate that veratridine-induced EAD is mainly formed by altered properties of the N a + channel current and partly by the L-type Ca2+ current d u e to slowed repolarization. Thus, EAD can be induced by different ionic mechanisms depending o n the basal conditions.
REFERENCES 1. ZIPES, D. P. 1989. Cardiac electrophysiology: Promises and contributions. J. Am. Coll. Cardiol. 13: 1329-1352. 2. CRANEFIELD, P. F. 1975. The Conduction of the Cardiac Impulses. Futura Publishing Co. Mt. Kisco, NY. L. V. ROSENSHTRAUKH & R LAZARA.1983. 3. BRACHMANN, J., B. J . SCHERLAG, Bradycardia-dependent triggered activity: Relevance to drug-induced multiform ventricular tachycardia. Circulation 68: 8%-856. 4. DANIANO,B. P. & M. R ROSEN. 1984. Effects of pacing on triggered activity induced by early afterdepolarizations. Circulation 69: 1013-1025. 5. JANUARY, C. T. & J. M. LIDDLE.1989. Early afterdepolarization: Mechanism of induction and block. A role for L-type CaZ+ current. Circ. Res. 64: 977-990. 6. EL-SHERIF,N., R H. ZIELER,W. CRAELIUS,W. B. GOUGH& R HENKEN. 1988. QKJprolongation and polymorphic ventricular tachycardia due to bradycardia-dependent early afterdepolarizations. Afterdepolarizations and ventricular arrhythmias. Circ. Res. 63: 286-305. 7. NATTEL,S. & M. A. QUANTZ.1988. Pharmacologic responses of quinidine induced early afterdepolarization in canine cardiac Purkinje fibers- Insights into underlying ionic mechanisms. Cardiovasc. Res. 22: 808-825. 8. LEVINE,J. H., J. F. SPEAR, T. GUARNIERI,M. L. WEISFELD,C. D. J. DELANGEN, L. C. BECKER & E. N. MOORE.1985. Cesium chloride-induced long Qr syndrome: Demonstration of afterdepolarizations and triggered activity in vivo. Circulation 72: 1092-1103. W. A. 1980. Neurotoxins that act on voltage-sensitive sodium chan9. CATTERALL, nels in excitable membranes. Ann. Rev. Pharmacol. Toxicol. 20: 15-43. 10. HIRANO,Y. & M. HIRAOKA. 1988. Barium-induced automatic activity in isolated ventricular myocytes from guinea-pig hearts. J. Physiol. 395: 455-472. & F. J. SIGWORTH. 1981. 11. HAMILL,0. P., A. MARTY,E. NEHER,B. SAKMANN Improved patch-clamp technique for high resolution current recording from cells and cell-free membrane patches. Pfliigers Arch. 391: 85-100. 1968. The kinetics and rectifier properties ofthe slow 12. NOBLE,D. & R W. TSIEN. potassium current in cardiac Purkinje fibres. J. Physiol. 195: 185-214. R W., P. HESS,E. W. MCCLESKEY & R L. ROSENBERG. 1987. Calcium 13. TSIEN, channels: Mechanism of selectivity, permeation, and block. Ann. Rev. Biophys. Chem. 16: 265-290. 14. ATTWELL, D. J., J. COHEN,D. EISNER,M. OHBA & C. OJEDA.1979. The steady-
HIRAOKA et ul.: MULTIPLE IONIC MECHANISMS OF EAD
15. 16. 17. 18. 19.
20.
47
state ‘ITX-sensitive (“Window”) sodium current in cardiac Purkinje fibres. Pflugers Arch. 379: 137-142. CORABOEUF, E., E. DEROUBAINX & A. COWLOMBE. 1979. Effects of tetrodotoxin on action potentials of the conducting system in dog heart. Am. J. Physiol. 236: H561-H567. PEPER,K. & W. TRAUTWEIN. 1967. The effect ofaconitine o n the membrane current in cardiac muscle. Pfliigers Arch. 296: 328-336. SUTRO, J. B. 1986. Kinetics of veratridine action o n Na channels of skeletal muscle. J. Gen. Physiol. 87: 1-24. BARNES,S. & B. HILLE.1988. Veratridine modifies open sodium channels. J. Gen. Physiol. 91: 421-443. NILIUS,B., W . BOLDT& K. BENNDORF.1986. Properties of aconitine-modified sodium channels in single cells of mouse ventricular myocardium. Gen. Physiol. Biophys. 5: 473-484. NILIUS, B., K. BENNDORF & F. MAFUCWARDT. 1986. Modified gating behaviour of aconitine treated single sodium channels from adult cardiac myocytes. Pflugers Arch. 407: 691-693.
INTRODUCTION Afterpotentials can be an important mechanism for the generation of arrhythmias since they induce triggered activity under certain settings. These afterpotentials are divided into two categories depending on the timing of repolarization. One is an early afterdepolarization which is a depolarizing afterpotential developing before the end of repolarization. The other is a delayed afterdepolarization starting after the end of repolarization.2 Both types of afterpotentials have been shown to develop under various conditions, and different mechanisms among them are thought to be involved. The early afterdepolarization (EAD) is seen at various potential levels from the plateau to the final repolarization (phase 3). Because of wide ranges of voltages for its development under various conditions, an ionic mechanism of EAD may not be explained by a single ionic current. As for the ionic mechanism of EAD, two inward currents were suggested as major contributing factors. They are the L-type Ca2+ current and the window or non-inactivating Na+ curIn addition, decreased K+ conductance was also indicated to be a contributing factor for development of EAD, since Cs+ as a blocker of K+ channels were frequently used as a model of this type of arrhythmia^.^,^ Among various conditions, low external K+ concentration is known to decrease the K+ conductance and to induce EAD,2 but its actual mechanism a Addresscorrespondence to Masayasu HIRAOKA, M.D., Ph.D. Tel. 81-3-3813-6111,Ext. 6160; FAX 81-3-5684-6295.
33
34
ANNALS NEW YORK ACADEMY OF SCIENCES
has not been clarified. The alkaloid neurotoxin, veratridine, is known to modify the Na+ channel properties in many excitable membranes9 and cause EAD in heart muscles,2 while the mechanism of EAD is not known. Therefore, the ionic mechanism for low-K+-induced and veratridine-induced EADs was examined in isolated ventricular myocytes from guinea-pig hearts.
MATERIALS AND METHODS Preparations All the experiments were done with isolated ventricular myocytes from guinea-pig hearts. Our method for an enzymatic isolation procedure was described elsewhere.’O Solutions The composition of Tyrode’s solution was (mM) NaCl 144, NaH2P04 0.33, KCI 4.0, CaC12 1.8, MgClz 0.53, glucose 5.5, and HEPES 5.0, with p H adjusted to 7.3-7.4 by addition of NaOH. The K+-free solution was made up by omitting KCI from the Tyrode’s solution. The Ni2+- or Cd2+containing solutions were prepared by adding 50 pM NiC12 and/or 500 pM CdC12 to the Tyrode’s or to the K+-free solutions, respectively. Tetrodotoxin (TTX)(Sankyo Co., Tokyo) was directly dissolved in the Tyrode’s or the K+free solutions at concentrations of 30-60 pM as indicated in the text. Veratridine (Sigma Chem. Co., St. Louis, MO) was prepared as 50 mM stock solution containing 0.1% ethanol and diluted into the Tyrode’s solution as the final concentration indcated in the text.
Recding Methods and Data Analysis The patch clamp technique of whole-cell and cell-attached configurations” was used to record membrane potentials, currents and single Na channel currents. In the whole-cell experiments, a patch-clamp amplifier (Model 8900, Dagan Corp., Minneapolis, MN) was used. Details of the recording technique and the data-acquisition systems were described in our previous reports.’O Action potentials were elicited by a square-wave pulse of 2 msec duration applied through the patch-pipette in a current-clamp mode at frequency of 0.1 H z or less. When the ramp-voltage clamp method was used, an intelligent arbitrary function synthesizer (Model 1731, NF Instrument, Yokohama) was used to supply command pulses. The pipette solution con-
HIRAOKA et al.: MULTIPLE IONIC MECHANISMS OF EAD
35
tained (mM) KCI 130, K2ATP (Sigma Chem. Co., St. Louis) 5.0, creatine phosphate (Sigma Chem. Co.) 5.0, EGTA 0.1 and HEPES-KOH buffer 5.0. Recordings were made at 32-34OC. Single-channel current recorchngs were made at room temperature in the cell-attached configuration with a patch-clamp amplifier (Axopatch, Axon Inst. Inc., Burlingame, CA). Current and voltage signals were recorded with a videocassette recorder (HR-S 5500, Victor Co., Tokyo) through a PCM converting system (RP-882, NF Instruments, Yokohama). Recorded signals were filtered off-line through an eight-pole Bessel low-pass filter (48 dB/octave, FV665, NF Instruments) at 1-2 kHz and sampled a t 4-10 kHz on the dlsc of a personal computer (IBM-PSI2) using an analog-to-digital converter (TL-1 DMA interface, Axon Inst., Burlingame). Data were collected and analyzed using a software, pClamp 5 . 5 . The external solution contained (mM) K aspartate 140, NaCl 4.5, MgCh 0.5, EGTA 1.0, glucose 5.5 and HEPESKOH buffer 5.0 adjusted to p H 7.35. The composition ofthe pipette solution was (mM) NaCl 140, KCI 4.0,MgC12 1.0, CaC12 0.18, glucose 5.5, and HEPES-NaOH buffer 5.0 adjusted to p H 7.35. Veratridine (50 pM) was applied to the pipette solution.
RESULTS Early A@a?epo&rization
Induced by the K+-Eree Solution
When the myocytes were exposed to the K+-free solution under a constant stimulation rate at 0.1 Hz, resting membrane potential gradually hyperpolarized from around -85 mV in the control to - 100 mV or more negative voltage with little change in action potential repolarization initially, and then action potential duration became prolonged due to slowing of the final repolarization phase (phase 3) forming a hump. The hump eventually turned into depolarizing afterpotential (EAD) and induced triggered activity (FIG. 1,A). The appearance of humps and EAD were always seen in six examined preparations exposed to the K+-free solution. In any case, EAD started at high negative membrane potential and the maximum diastolic potential before the start of the first EAD was -68 3.4 mV (mean SD, n = 6). The depolarizing phase of EAD shifted into a rapid upstroke a t voltages between -50 and -40 mV in most of the cells. Once triggered activity was initiated in these preparations, the maximum diastolic potentials stayed around -50 mV and spontaneous firings continued as long as the K+-free solution was p e h s e d . These changes of membrane potentials were reversible upon wash-out of the K+-free solution. Changes in the background current-voltage (I-v) relation exposed to the K+-free solution were examined with a ramp voltage clamp of slow depolari-
*
*
36
ANNALS NEW YORK ACADEMY OF SCIENCES
FIGURE 1. Changes in action potentials and membrane currents after exposure to the Kt-free solution. A shows successive changes in action potentials after exposure to the K+-free solution. Initially, hyperpolarization of resting membrane potential developed with little changes in the repolarization phase. Later, prolongation of action potential started with a formation o f a hump o n the phase 3. Then, the hump eventually turned into a depolarizing afterpotential and induced triggered activity. B illustrates the background I-V relations after exposure to the Kt-free solution. Open circles indicate the I-V curve recorded in the control solution containing 4 mM K+. Numbers 1 to 4 represent the I-V curves recorded after the start of the K+-free solution in that order. The voltage protocol is indicated in the inset shown at the top left. Note that the I-V curves show a "crossover" phenomenon in the K+-fme solution and net-inward current at potentials negative to -50 rnV with inward hump between -40 and - 10 rnV. A and B were recorded from different cells.
HIRAOKA et al.: MULTIPLE IONIC MECHANISMS OF EAD
37
zation (150 mV per 30 sec) from -100 mV to +50 mV. FIGUREl,B shows successive changes in the I-V relations after the pefisate was switched from the Tyrode’s solution to the K+-free solution. Currents at negative voltages were gradually decreased and, therefore, the I-V curves showed “cross-over” phenomenon12 in successive recordings. Finally, current at potentials negative to - 50 mV became inward and a negative hump was observed at potential between -40 and - 10 mV. On the other hand, outward currents at positive voltages were increased in the K+-free solution. In six examined preparations, nearly complete suppression of the inward rectifier K+ current with a negative shift of the I-V relation, an increase in a steady inward current at the plateau level and increased outward current at positive voltages were general findings in the K+-free solution. To examine a possible involvement of inward currents in the formation of EAD, the effects of inhibitors of the Ca2+and Na+ currents were tested (FIG. 2). When the myocyte was exposed to the K+-free solution in the presence of 50 pM Ni2+ to block the T-type Caz+-current,13 the appearance of hump and EAD was not affected (A). Additional application of 500 pM CdZ+to inhibit the L-type Ca2+ current as we1113 abolished the appearance of EAD and membrane potential stayed at the plateau level during action potential repolarization, although prolongation of action potentials at the beginning of the K+-free perfusion was similarly observed as in the absence of these cations. 4 mV (n = 5) in the Cd2+The maximal diastolic potential was -32 containing K+-free solution and n o spontaneous activity was initiated. When myocytes were exposed to the K+-free solution in the presence of 30-60 pM tetrodotoxin (TTX) to block the window or non-inactivating N a + - c ~ r r e n t , ’ ~the -~~ hump and EAD were similarly induced as in the control, but triggered activity from EAD was not elicited (C). The latter finding was possibly due to a partial inhibition of the Na+ current. These results of the TTX effect were confirmed in four preparations. FIGURE3 shows the effects of the inward current inhibitors on the background I-V relations. Application of 50 pM Ni2+ did not affect the I-V relation in the K+-free condition. Further application of 500 pM Cd2+ abolished increased steady inward current between -40 and - 10 mV, and partially suppressed the outward current at positive voltages (A), while little change in the I-V relation was noted at potentials negative to -50 mV. These results were confirmed in 5 myocytes. B shows the effect of 30 pM T T X on the background I-V relation. TTX slightly depressed inward current at around -60 to -30 mV but left the steady inward current at the plateau level to remain (n = 3). The contribution of the outward current to formation of EAD was SUSpected, since EAD started around -68 mV which was too high to activate the L-type CaZ+ current, and Ni2+ or TTX could not abolish EAD. One POS-
*
38
ANNALS NEW YORIC ACADEMY OF SCIENCES
FIGURE 2. Effects of the inward current inhibitors on the low-K+-induced EAD. A shows effects of 50 pM Ni2+ on action potential after exposure to the K+-free solution. Ni2+ was contained in both the Tyrode's and K+-free solutions. B shows effects of hrther application of 500 pM Cd2+.While prolongation of action potentials were seen, no EAD was induced and the maximal diastolic potential stayed at around - 30 mV due to incomplete repolarization. C represents effect of 30 pM T T X on the appearance of EAD after exposure to the K+-free solution. The humps and EAD were similarly observed as in the absence of the inhibitors, but triggered activity was not initiated from EAD probably due to a partial block of the Na+ current. A and B were recorded from the same cell but C was recorded from a different cell.
HIRAOKA et al.: MULTIPLE IONIC MECHANISMS OF EAD
39
0 ; K-free A ; " + Ni
m;
"+Ni+Cd
0 ; K-free
Id;
"+
TTX
I/ FIGURE 3. Effects of the inward current inhibitors on the background current in the K+-free condition. A shows effects of 50 pM Ni2+ (A)and plus 500 pM Cd2+ (W) on the background I-V relation. Ni2+ had n o effects on the I-V relation compared to the control (O), while Cd2+ completely depressed negative hump of the I-V relation between -40 and - 10 mV, and partially suppressed the outward currents at positive voltages. B illustrates effects of 30 pM TTX (8)o n the background I-V relation. T T X slightly depressed the steady inward current at potentials between -60 and -20 mV, but left the negative hump in the I-V relation at the plateau voltage to remain. A and B were recorded from different cells.
ANNALS NEW YORK ACADEMY OF SCIENCES
40
sible candidate is the time-dependent decay of the delayed outward K+ current (IK).FIGURE4 presents the effects of the K+-free solution on IK and its decay upon repolarization. In the K+-free solution, not only the outward current during depolarizing steps but also the tail current amplitude were increased compared to the control, while the time course of decay was not much affected since its half-time was not changed. Increases in the tail current amplitude were observed at all the test voltages examined and these changes were reversible on the wash-out of the K+-free solution (n = 3).
Early A~poluririwttimtInduced by Veratricline When 25-100 pM veratridme was applied to the myocytes under the constant stimulation rate of 0.1 Hz, action potentials were gradually prolonged due to the slowing of the plateau phase and regenerative EADs appeared on the plateau after 8-10 minutes of application (FIG. 5). At this stage, prolongation of action potentials was so marked to be its duration of several seconds. EADs appeared regeneratively on the plateau phase increasing their amplitudes as the starting voltages became negative. Application of 500 pM Cd2+ shortened the action potential duration and decreased the amplitudes of EADs with slowed frequency of their appearance, but complete abolishment
V l p . Q ' 5 n A A J
I=O
~
A
A~
.*'
O
.0
+20
+40
O +60
+BonN
FIGURE 4. Effects of the K+-free solution on the delayed outward K+ current. A represents current records in responses to 500 msec voltage steps from a holding potential of -50 mV to +70 (q) and +30 rnV (bonmn), and subsequent repolarization to -50 mV. Open circles (0)indicate currents in the control Tyrode's solution and triangles (A)are those in the K+-free solution. B shows the tail current amplitude (ordinate) Venus test voltage (VT).
HIRAOKA cz af.: MULTIPLE IONIC MECHANISMS OF EAD
41
FIGURE 5. Veratridine-induced action potential prolongation and appearance of EAD. A shows action potential changes after application of 25 pM veratridine. Action potentials were recorded 5-10 minutes in the presence of veratridine. Progressive lengthening of action potentials and development of regenerative EADs o n the plateau phase were noted in the later records. B illustrates effects of CdZ+(A)and plus T T X (M) on action potentials and EADs. Cd2+ shortened action potential considerably, and decreased the amplitude and frequency of EADs. Further application ofTTX in the presence of veratridine shortened the action potential and completely abolished EADs. A and B were recorded from the same cell.
of EADs was not achieved (n = 3). Further application of 30 pM T T X completely eliminated residual EADs and action potential duration was further shortened. These effects of T T X developed rapidly within 2-3 minutes of application. Application of TTX alone (30-60 p M ) caused a rapid abolishment of EADs and shortened action potential duration in other two preparations. The effects of veratridine on the single Na+ channel currents were examined by the cell-attached patch recordings (FIG. 6 ) . In the control, openings of the single Na+ channel currents were seen at the beginning of pulses and were seldom after 10 msec. An ensemble average current showed a decay time
42
ANNALS NEW Y O U ACADEMY OF SCIENCES
constant of 3.6 msec. The conductance of this channel current was 16pS (n = 5). When 50 pM veratridine was applied to the pipette solution, the amplitude of the single channel currents decreased to the half or one-third of the control (5-8pS). In addition, markedly prolonged openings, sometimes lasting throughour 500 msec pulses, or delayed reopenings after 100 msec were frequently observed. Because of these prolonged openings or delayed reopenings, channel activities were seen upon repolarization to the resting potential level of - 120 mV after 500 msec step pulses. An ensemble average current decayed with very slow time constant of 609 msec, which was about 200 times slower than the control. Prolonged channel openings or delayed reopenings were seen at various voltages, once the channel was activated (FIG. 7) (n = 4).These results suggest that the m curve of the Na+ channel was shifted to the left along the voltage axis by veratridine.
DISCUSSION The present study demonstrated that EAD was induced easily and repeat-
FIGURE 6. Effects of veratridine on the single Na' channel currents. A shows records in the control and B after application of veratridine. The top indicates the voltage protocol, mirlltle tnues are single channel current sweeps and the botarm is an ensemble average current of 500 (A) and 100 (B) sweeps.
HIRAOKA et al.: MULTIPLE IONIC MECHANISMS OF EAD
43
FIGURE 7. Effects of veratridine on single Na' channel currents at different voltages. Three different test voltages are indicated at the left. Records were taken from a different cell from that shown in FIGURE6, B.
edly in ventricular myocytes by petfusing the K+-free solution or exposing to a low concentration of veratridine. The ionic mechanism of the low-K+induced EAD seems to be different from that of the veratridine-induced EAD because of differences in modes of appearance, the level of their developing voltages, and pharmacological responses. The low-K+-induced EAD was seen at high negative voltage of -68 mV with a single depolarizing afterpoten-
ANNALS NEW YORK ACADEMY OF SCIENCES
44
tial, whereas the veratridine-induced EAD initiated as repeated appearance of afterpotentials at relatively low membrane potentials during the plateau. The former type was easily blocked by Cd2+ which inhibited the L-type Ca2+ channel and partially suppressed the delayed outward K+ current, while the latter was abolished easily by T T X ; in addition Cd2+partly eliminated its appearance. Therefore, the low-K+-induced EAD was mainly produced by the L-type Ca2+ current and time-dependent decay of IK, while the veratridineinduced EAD was caused by the Na+ current with partial contribution of the Ca2+ current. The prolongation of action potential duration seems to be a common factor to predspose the appearance of EAD in both conditions, which was similar to other reported case^.^-^ In the K+-free solution, the prolongation was mainly caused by lengthening or hump formation at the phase 3 which could be explained by inhibition of the inward rectifier K+ current. This condition was similar to the Cs+-induced or quinidine-induced EAD,3.416 while the veratridine-induced EAD was produced by the altered properties of the Na+ channel as like the cases by aconitine16 or anthopleurin-A.6 The single channel recording showed that veratridine produced long-lasting openings or delayed reopenings, causing markedly delayed decay in ensemble average current. These changes in opening behaviors were seen at high negative potentials around -90 mV once the channels were opened (not shown), which indicated the m curve of the Na+ channel was shifted to negative direction along the voltage axis. In addition, veratridine decreased the conductance of the channels. These findings were similar to the veratridine effects on the Na+ channels of skeletal muscle and neuroblastoma cell^.^^^^^ The actions also resembled those of aconitine in cardiac cells as modifying the open Na+ channels. Although veratridine caused increased inward current with slow and steady decay, EADs appeared as regenerative afterpotentials repeatedly during the plateau. The reason for recycling appearance of EADs was not known, but it may be partly explained by contribution of the L-type Ca2+ current, since application of Cd2+slowed their frequencies and decreased amplitudes. It was also noted that at the plateau voltages the total membrane current was small and, therefore, small changes in outward or inward currents would cause potentials to depolarizing or hyperpolarizing directions. In this context, time- and voltage-dependent changes in the background current mainly produced by delayed outward K+ current and inward rectifier K+ current might be involved in the repeated appearance of EADs, in association with recovery from inactivation of the L-type Ca2+ c ~ r r e n t This . ~ point needs hrther clarification by additional experiments. On the contrary, the low-K+induced EAD was composed of a single depolarizing afterpotential following action potential repolarization. The initial depolarizing portion of EAD was 19320
'
HIRAOKA
et
al.: MULTIPLE IONIC MECHANISMS OF EAD
45
probably brought about by a time-dependent decay of IK, since the T- and L-type Ca2+ current were not involved, and T T X did not block it. The conditions of the present experiment are too far to be seen in clinical situations, and may be questioned as to their practical significance. Although the K+-free condition is not relevant to clinical situations, decreased K+ conductance seen in hypokalemia, acidosis or combinations of both may provide an easy appearance of EAD with acceleration of sympathetic activity to increase the Ca2+ current. There have been no reports of modification of the Na+ channel behaviors like veratridine action in any of clinical settings, and, therefore, the clinical significance of the case of the veratridine-induced EAD is not known. The study indicates that EAD is a potential change which can be produced by multiple ionic mechanisms depending on the basal conditions. This fact must always be considered for the mechanism of arrhythmias when triggered activity from EAD is suspected.
SUMMARY Ionic mechanisms of early afterdepolarization (EAD) induced by the K+fi-ee solution or veratridine were studied with guinea-pig ventricular myocytes using the patch-clamp technique of whole-cell and cell-attached patch configurations. In the K+-free solution, myocytes exhibited prolonged action potential duration with humps on the final repolarization phase, which eventually turned into EAD starting around -70 mV and induced triggered activity. Application of 0.5 mM Cd2+ inhibited the development of EAD and caused depolarization of maximum diastolic potentials around - 30 mV, although Cd2+ did not prevent prolongation of the action potential. Application of 50-100 pM Ni2+ or 30 pM tetrodotoxin had little effects on EAD and diastolic potentials. The background current-voltage relation examined by a ramp voltage clamp showed inhibition of the inward rectifier K+ current, induction of steady inward current between -40 and - 10 mV, and increase in the outward tail current upon repolarization in the K+-free solution. Cd2+completely blocked the steady inward current at the plateau level and partially depressed the delayed outward K+ current, while Ni2+ had no effects on the background I-V relation. Tetrodotoxin showed a mild inhibitory effect on the inward component of the background current negative to -50 mV, but left the steady inward current at the plateau level. Therefore, EAD in the K+-fiee condition is mainly formed by decreased inward rectifier K+ current, activation of the L-type Ca2+ current, and time-dependent decay of the delayed outward K+ current upon repolarization. Application of 25-100 pM veratridine caused marked prolongation of action potential with appearance of regenerative EADs. Action potential prolon-
46
ANNALS NEW YORK ACADEMY OF SCIENCES
gation and EADs were partially abolished by CdZ+and completely eliminated by tetrodotoxin. T h e single channel current recordings showed a decreased current amplitude, and prolonged and delayed openings of t h e N a + channel currents by veratridine. Thus, a n ensemble average current showed markedly prolonged decay time constant of 609 msec in veratridine from 3.6 msec in the control. These results indicate that veratridine-induced EAD is mainly formed by altered properties of the N a + channel current and partly by the L-type Ca2+ current d u e to slowed repolarization. Thus, EAD can be induced by different ionic mechanisms depending o n the basal conditions.
REFERENCES 1. ZIPES, D. P. 1989. Cardiac electrophysiology: Promises and contributions. J. Am. Coll. Cardiol. 13: 1329-1352. 2. CRANEFIELD, P. F. 1975. The Conduction of the Cardiac Impulses. Futura Publishing Co. Mt. Kisco, NY. L. V. ROSENSHTRAUKH & R LAZARA.1983. 3. BRACHMANN, J., B. J . SCHERLAG, Bradycardia-dependent triggered activity: Relevance to drug-induced multiform ventricular tachycardia. Circulation 68: 8%-856. 4. DANIANO,B. P. & M. R ROSEN. 1984. Effects of pacing on triggered activity induced by early afterdepolarizations. Circulation 69: 1013-1025. 5. JANUARY, C. T. & J. M. LIDDLE.1989. Early afterdepolarization: Mechanism of induction and block. A role for L-type CaZ+ current. Circ. Res. 64: 977-990. 6. EL-SHERIF,N., R H. ZIELER,W. CRAELIUS,W. B. GOUGH& R HENKEN. 1988. QKJprolongation and polymorphic ventricular tachycardia due to bradycardia-dependent early afterdepolarizations. Afterdepolarizations and ventricular arrhythmias. Circ. Res. 63: 286-305. 7. NATTEL,S. & M. A. QUANTZ.1988. Pharmacologic responses of quinidine induced early afterdepolarization in canine cardiac Purkinje fibers- Insights into underlying ionic mechanisms. Cardiovasc. Res. 22: 808-825. 8. LEVINE,J. H., J. F. SPEAR, T. GUARNIERI,M. L. WEISFELD,C. D. J. DELANGEN, L. C. BECKER & E. N. MOORE.1985. Cesium chloride-induced long Qr syndrome: Demonstration of afterdepolarizations and triggered activity in vivo. Circulation 72: 1092-1103. W. A. 1980. Neurotoxins that act on voltage-sensitive sodium chan9. CATTERALL, nels in excitable membranes. Ann. Rev. Pharmacol. Toxicol. 20: 15-43. 10. HIRANO,Y. & M. HIRAOKA. 1988. Barium-induced automatic activity in isolated ventricular myocytes from guinea-pig hearts. J. Physiol. 395: 455-472. & F. J. SIGWORTH. 1981. 11. HAMILL,0. P., A. MARTY,E. NEHER,B. SAKMANN Improved patch-clamp technique for high resolution current recording from cells and cell-free membrane patches. Pfliigers Arch. 391: 85-100. 1968. The kinetics and rectifier properties ofthe slow 12. NOBLE,D. & R W. TSIEN. potassium current in cardiac Purkinje fibres. J. Physiol. 195: 185-214. R W., P. HESS,E. W. MCCLESKEY & R L. ROSENBERG. 1987. Calcium 13. TSIEN, channels: Mechanism of selectivity, permeation, and block. Ann. Rev. Biophys. Chem. 16: 265-290. 14. ATTWELL, D. J., J. COHEN,D. EISNER,M. OHBA & C. OJEDA.1979. The steady-
HIRAOKA et ul.: MULTIPLE IONIC MECHANISMS OF EAD
15. 16. 17. 18. 19.
20.
47
state ‘ITX-sensitive (“Window”) sodium current in cardiac Purkinje fibres. Pflugers Arch. 379: 137-142. CORABOEUF, E., E. DEROUBAINX & A. COWLOMBE. 1979. Effects of tetrodotoxin on action potentials of the conducting system in dog heart. Am. J. Physiol. 236: H561-H567. PEPER,K. & W. TRAUTWEIN. 1967. The effect ofaconitine o n the membrane current in cardiac muscle. Pfliigers Arch. 296: 328-336. SUTRO, J. B. 1986. Kinetics of veratridine action o n Na channels of skeletal muscle. J. Gen. Physiol. 87: 1-24. BARNES,S. & B. HILLE.1988. Veratridine modifies open sodium channels. J. Gen. Physiol. 91: 421-443. NILIUS,B., W . BOLDT& K. BENNDORF.1986. Properties of aconitine-modified sodium channels in single cells of mouse ventricular myocardium. Gen. Physiol. Biophys. 5: 473-484. NILIUS, B., K. BENNDORF & F. MAFUCWARDT. 1986. Modified gating behaviour of aconitine treated single sodium channels from adult cardiac myocytes. Pflugers Arch. 407: 691-693.
