Mechanisms by which calcium modulates diastolic depolarization in sheep cardiac Purkinje fibers

Mechanisms by Which Calcium Modulates Diastolic Depolarization in Sheep Cardiac Purkinje Fibers

Juan Tamargo, MD,* and Mario Vassalle, MD

Abstract: The mechanisms by which calcium modulates diastolic depolarization (DD) in sheep cardiac Purkinje fibers were studied in vitro. Increasing [Cal, from 2.7 mM to 10.8 mM increased both the slope and amplitude ofDD, induced oscillatory potentials (V,,), and prolonged depolarization (V,,). The steepening of DD occurred even in the absence of an obvious V,,. The increase in DD amplitude was due both to an increase in the maximum diastolic potential and to a less negative steady-state level. At constant [Cal,, increasing the driving rate had effects similar to those induced by increasing [Cal,. The increase in DD slope and amplitude was least at the slowest rates and leveled off at the fastest rates in high [Cal,. Lowering [Cal, decreased DD slope and amplitude, but spontaneous activity could be present during interruption of the drive. In slowly driven fibers, increasing [Cal,, to 10.8 mM initially shifted the maximum diastolic potential and steady state DD to more negative values, and subsequently shifted the latter (but not the former) to less negative values. On recovery, a transient depolarization occurred. Quiescent fibers exposed to high [Cal, also underwent a transient hyperpolarization and a subsequent depolarization, whereas reciprocal effects occurred when [Cal, was lowered. It is concluded that [Cal0 modulates DD through several different mechanisms and that most (but not all) modifications induced are brought about by changes in [Cal,. Key words: calcium, rate of discharge, diastolic depolarization, calcium overload, oscillatory potential, prolonged depolarization.

In cardiac

Purkinje

fibers,

the slope

of diastolic

and vice versa.1,2 The mechanisms underlying this calcium-related change in the slope of diastolic depolarization have not been clarified. Since high [Cal, causes an oscillatory potential (V,,),‘-’ the steepening could be due to the superimposition of V,, on DD. It seems less likely that V,, would steepen DD at normal or low [Cal,, but this possibility cannot be ruled out u priori since the rate of discharge must be taken into consideration. Thus, even in normal [Cal,, repetitive voltage clamp steps can induce the transient inward oscillatory current (I,,) that underlies V,,,4 and an increased rate of dis-

de-

(DD) becomes steeper when the extracellular calcium concentration ([Cal,) is increased polarization

From the Department of Physiology, State University of New York, Health Science Center, Brooklyn, New York. *Dr. Tamargo is currently at the Departamento de Farmacologia. Far&ad de Medicina, Universidad Complutense, Madrid, Spain. Supported by NIH grants HL1745 1 and HL27038. Dr. Tamargo

was supported by CAICYT grant 2074/83. Reprint requests: Dr. Mario Vassalle, Department of Physiology Box 31, SUNY, Health Science Center, 450 Clarkson Avenue, Brooklyn,

NY 11203.

349

350

Journal of Electrocardiology Vol. 24 No. 4 October 1991

charge increases intracellular calcium activity (a&,).* Reciprocally, a slow driving rate could prevent the induction of V,, in high [Cal,. In evaluating the role of V,, in DD, it should be noted that V,, and I,, peak after several hundred milliseconds,2-4 that is, well after the end of the diastolic depolarization visible in between action potentials at the driving rates usually used. For this reason, it is difficult to identify even the presence of V,, (let alone the modifications it induces on the slope of DD) unless the drive is periodically interrupted: only then does V,, become apparent as it peaks and decays. In addition, DD reaches a steadystate after several seconds since during diastole the pacemaker current varies with a time constant of 24 seconds depending on [K]0.6 Furthermore, driveinduced calcium overload causes a prolonged depolarization (V,,) that slowly decays after the drive and may be superimposed on DD.7 Finally, calcium also could modify DD independently of V,,, as voltage clamp experiments suggest.8,9 The aim of the present experiments was to investigate whether calcium modulates DD by different mechanisms and whether these mechanisms are both calcium-dependent and frequency-dependent. The calcium loading of the fibers was carried out by changing [Cal,, the driving rate, or both. After driving the fibers at different rates for fixed periods of time, the electrical drive was interrupted until DD reached a steady-state and this procedure was repeated in the presence of different [Cal,. To further investigate the role of [Cal, versus [Cali, several types of experiments were conducted in driven and quiescent fibers. (A preliminary report has appeared in abstract form. lo )

Materials and Methods Sheep of either sex weighing 12-30 kg were anesthetized with sodium pentobarbital (30 mg/kg, intravenously). The heart was rapidly excised and freerunning strands of Purkinje fibers were cut from the left ventricle. The preparations were perfused in a narrow channel (volume of the perfusing solution about 0.2 ml) in a tissue bath at 37°C with oxygenated (97% O2 and 3% CO*) Tyrode solution of the following composition (mM): NaCl 137, KC1 4, CaCl, 2.7, NaHC03 11.9, NaH2P04 0.45, MgC12 1.05, and glucose 5.5 (pH = 7.3-7.4). The rate of perfusion was 60 drops/min. CaCl, was added as powder to oxygenated, freshly prepared Tyrode solution.

The preparations were driven by means of electrical stimuli delivered by Tektronix 162 Waveform Generator and 161 Pulse Generator through a stimulus isolation unit (Grass SIU 4678). The basal rate was 60 beats/min, but it was changed successively to 7.5, 15, 30,60, 120, and 180 beats/min in different tests. The duration of the stimulus was 1-2 ms and the voltage was approximately 50% above the threshold. The stimulating electrodes were two stainless steel pins insulated except for the tip; one electrode was placed near the preparation and the other was used to immobilize one end of the preparation to the floor of the tissue bath. The other end of the preparation was connected to a force transducer (Grass FT03C) by means of a short silk thread and a stainless steel rod. The force transducer was connected to a Grass Model 7D Polygraph. The transmembrane potentials were recorded by means of microelectrodes filled with 3 M KC1 and connected to a Dagan headstage and to a Dagan model 8500 operational amplifier. The action potential and twitch were displayed on a Tektronix model 5 111 storage oscilloscope and recorded on paper on a three-channel chart recorder (Gould Brush 2400) and in some experiments also on a FM tape recorder (Store 4, Lockheed Electronics, Inc.) at 15 inches/s. The preparations were allowed to equilibrate approximately 1 hour in the tissue bath before beginning the experiment. In order to measure both the slope and the steady-state amplitude of DD, the fibers were driven at a given rate for 5 minutes and then the drive was interrupted for 30 seconds. When a solution was changed, a small air bubble was introduced in the tubing: the arrival of the air bubble in the tissue bath monitored the arrival of the new solution. The amplitude of DD was measured as the difference between E,,, and the steady-state value at the end of a long pause. The amplitude of V,, was measured as the difference between the maximum diastolic potential and the peak of V,, . The amplitude of the decaying V,, was measured as the difference between the least negative point during diastole and the steady-state DD at the end of the pause. All measurements were carried out under magnification on the Gould millimeter paper where the traces were recorded. Strophanthidin was obtained from Sigma Chemical Co. The results are expressed as mean f standard error of the mean. A Student’s t test was used and a p value < 0.05 was considered significant and is indicated by an asterisk. Analysis of variance with repeated measures was also carried out and the results are reported in the text. A least squares method (SAS, Proc Reg Procedure) was used to compute the correlation coefficient (R).

Calcium and Diastolic Depolarization

Results

In a first approach, the preparations were driven at different rates in normal [Ca] 0 (2.7 mM) . In Figure 1, as the driving rate was increased, the action potential duration decreased, whereas the twitch amplitude increased (up to 12O/min). During the pause, the initial DD slope increased progressively as the rate was increased, as emphasized by the dashed lines. At 120/min, an inflection between the lower and upper part of the initial DD appeared (arrow), which became a V,, at 180/min (arrow). Also, DD amplitude increased at the faster rates as shown by

351

Effects of Different [Cal, at Constant Rate If the above findings result from an increased [Cali, then similar effects should be obtained by pro-

Beats/min

Fig. 1. .Effects of progressively

depo-

driving rate in beats/min is indicated to the left of the respective traces (15, 30, 60, 120, 180 beats/ mm). In each strip, the first panel shows the action potential and the twitch curves recorded near the end of the 5-minute drive. The second panel shows at higher gain the lower part of the action potential and DD that follows the interruption of the drive. The dashed lines extrapolate the initial diastolic depolarization. The empty arrowheads point to an inflection on the initial DD or to a V,,. In the inset above the bottom trace, the traces have been aligned by superimposing E,,,. The continuous line under DD in the bottom right panel emphasizes the V,, above it. In this and in the other figures, the voltage, time, and force calibrations are shown next to the respective traces.

Tamargo and Vassalle

the DD curves superimposed for the 15, 30, 60, and 120/min drives (1, 2, 3, and 4, respectively, in the inset above the bottom trace). Finally, following the 180/min drive, DD amplitude was maximal after about 15 seconds of quiescence and then the membrane potential became somewhat more negative (decay of the prolonged depolarization V,,, see below). Thus, the development of a V,, can also occur at normal [Cal,, but only at rates higher than normal.

Effects of Drive at Different Rates on Diastolic Depolarization

faster rates on diastolic larization (DD). The

l

30 -‘

L--

t;

( ‘1, I,

60 J

‘L-

L!

-

‘._

I^

352

1Ca I

o

Journal of Electrocardiology

Vol. 24 No. 4 October 1991

.mM

Bo6Wmln

Fig. 2.

Effects of extracellular calcium concentration on dia-

stolic depolarization. The results are shown for two driving rates: 60 and 120 beats/min. The left-hand panels show the action potential and twitch curves at various [Cal, (rate dO/min); the number next to each panel indicates the extracellular calcium concentration in mhJ. The middle and righthand panels show the termination of 60lmin and 120/min drives, respectively, in different [Cal,, . The insets above the bottom traces show the superimposition of the initial DD at 1.35, 5.4, and 10.8 mM [Cal, ( 1,2, and 4 respectively) at the two rates shown. Other explanations as in Figure 1.

8.1

10.8 -

.

4

10 0

gressively increasing [Cal, while the fiber is driven at a given rate (so that rate-dependent changes do not occur during the exposure to a given [Cal,.) The preparations were driven at the same rates as in Figure 1 and [Cal, was increased during each drive rate. In Figure 2, the progressive increase in [Cal, increased the twitch amplitude and decreased the action potential duration. At 60/min, increasing [Cal, increased the initial DD slope and 10.8 mM calcium caused an obvious inflection (arrow), even at this normal rate. At 120/min, increasing [Cal, increased the initial DD slope even more and 5.4, 8.1, and 10.8 mM [Cal, caused a V,, (arrows). In addition, after the higher driving rates, DD amplitude was greater especially at the lower [Cal,. With each rate, increasing [Cal, also increased DD amplitude as shown in the insets above the bottom trace. These findings indicate that increasing [Cal, at the same rate (Fig. 2) reproduces several of the results obtained by increasing the rate at the same [Cal, (Fig. I), presumably by causing the same effect, that is, an increase in [Cali.

Changes in DD as a Function of [Cal, and Rate

In Figure 3 (n = 6), the changes in slope (y axis in [A]) and in amplitude (y axis in [B]) of DD are plotted as a function of [Cal, (x axis) and of all driving rates tested (z axis). In (A), the slope increased at each rate as a function of [Cal, and as a function of the driving rate at each [ Ca] 0. The increase in slope became much steeper at the two highest [Cal, and two highest rates, reflecting the more frequent development of V,, (the numbers in the graph indicate the number of experiments in which a V,, was present). In (B), DD amplitude (mV) also became larger when [Cal, or the rate was higher. In contrast to the slope, the amplitude increased markedly with the first step increase in [Cal, but additional increases at the higher [Cal, and rates were smaller. Analysis of variance shows that in Figure 3 (A) the difference in the DD slope among groups is signifi-

Calcium and Diastolic Depolarization

Tamargo and Vassalle

l

353

Effects of Increasing Durations of Drive

180

If the effects of increasing driving rates are related to an increasing [Cali, similar results should be obtained by overdriving the preparations at a constant rate and at a constant [Cal, but for progressively longer periods of time, a procedure which also increases calcium loading of the cell. In Figure 4, in normal [Cal, (left-hand traces), increasing the duration of lSO/min drive from 10 seconds (A) to 30 seconds (B) increased DD slope and amplitude. After the 60-second drive (C), a distinct inflection in DD (arrowhead) was present, which then progressed to the threshold and initiated spontaneous activity. The longest drive (D) was followed by a clear V,, (arrowhead) and spontaneous repetitive activity at a faster rate than in (C). In the high [Cal, (right-hand traces), an inflection in DD after the lo-second drive and a clearcut V,, after longer drives (arrowheads) occurred, although V,, failed to attain the threshold potential and initiate spontaneous activity. Also, DD amplitude was greater in high than in normal [Cal,, as expected.

[ca]o.mM

2.7 mM

Fig. 3. Tridimensional

plots of diastolic depolarization as a function of the rate and extracellular calcium concentration. The y axis is the scale for diastolic depolarization (DD) slope in mV/s in (A) and that for DD amplitude in mV in (B). In both panels, [Cal, in mM is shown in the x axis and the rate in beats/min on the z axis. The asterisks indicate that the values are statistically different (p < 0.05) with respect to both the control at the lowest rate ( 15/min) and the control at the lowest [Cal, ( 1.3 5 mM). The squares indicate that the values are statistically different from the control at the lowest [Cal,.

f = 85.6) and so is that among [Cal, groups (p = 0.0001, f = 11.0). The interaction between rate and [Cal, groups is also significant (p = 0.02, f = 2.7). The difference in the amplitude of DD among rate groups is significant (p = 0.0001, f = 85.1), although that among [Cal, groups is not (p = 0.22). The difference in the DD slope between the five rate groups (group 1 vs. 2, 2 vs. 3, 3 vs. 4, and 4 vs. 5) is consistently significant (p = 0.0001 in each case; f = 24.3, 38.7, 39.9, and 51.4, respectively). The difference in the DD amplitude between rate groups is significant (group 1 vs. 2, 2 vs. 3, 3 vs. 4, p = 0.0001 in each case; groups 4 vs. 5, p = 0.02; f = 63.1, 60.8, 65.76, and 5.3, respectively) .

A

B

l-

G

1

cCa O

10.8

mtvl

10 s

30 s

cant (p = 0.0001,

Fig. 4. Effects of increasing

the drive duration at normal and high [Cal,. The left-hand panels were recorded in 2.7 mM [Cal, and the right-hand panels in 10.8 mM [Cal0 at the termination of 1SO/min drives of the duration indicated by the numbers in between the panels. Note that the righthand panels were recorded at a faster time base to visualize the V,, more clearly.

354

Journal of Electrocardiology

Vol. 24 No. 4 October 1991

A

B [Ca],

10.6

rnM

30 -

z AMPLITUDE

*

10.8

,” >

25z

E

AMPLITUOE f 0

2.7 -202 I:

ii SLOPE

-15

2.7*

8 a 0

10.

-10

,“‘,I!

5 15

I

30

45

I

60

a

-20:

0 t 2-

-16

? a z

ii F

-16

: >

z

>

o

lTIME

I

“‘C”’

120 DURATION

5 15 OF

w’ 0

Y w P

20.

0 D

E

-22

3-

DRIVE.

30

0

1

45

60

TO

PEAK 4

120

s

Fig. 5.

Effects of drive duration on slope and amplitude of diastolic depolarization (DD) as well as on amplitude and time to peak of the oscillatory potential. (A) The left ordinate shows that scale for DD slope in mV/s and the right ordinate that for DD amplitude in mV. The numbers next to the end of the curves indicate [Cal, for the respective curve. The empty symbols show the results obtained in normal [Cal, and the filled symbols indicate those in high [Cal, (10.8 n-&%). (B) The results were obtained in high [Cal, ( 10.8 mh4). The left ordinate is the scale for V,, time to peak (filled circles) in s and the right ordinate is that for V,, amplitude (empty circles) in mV. The abscissae show the drive duration in s. An asterisk next to a symbol indicates a p < 0.05 with respect to the control value obtained after 5 s (A) or 120 s (B) drive.

As shown in Figure 5 (n = 6), at normal [Cal,, both DD slope and amplitude increased as a function of drive duration and the increase of these parameters was correlated (R = 0.97). In high [Cal,, DD amplitude was larger in absolute terms than at normal [Cal, (different intercept of the regression lines, p < 0.001 ), but its increase with longer drive durations was similar (R = 0.98) to that in normal [Cal,; instead, DD slope increased steeply with increasing drive durations. In Figure 5(B), V,, amplitude increased up to 30 seconds of drive but then remained constant and eventually declined with longer drives. In contrast, with longer drives, the time to peak of V,, decreased monotonically and markedly to level off after the 60second drive. This behavior is similar to that in other conditions of calcium overload.‘,’ ’ Analysis of variance shows that in Figure 5 (A) the difference in DD amplitude among groups with increasing duration of drive is significant (p = 0.000 1, f = 15.8), although that among the [Cal, groups is not (p = 0.07). A significant difference exists in the DD amplitude between groups 5 and 6 (p = 0.002), 6 and 7 (p = 0.02), 7 and 8 (p = 0.03), and 8 and 9 (p = 0.04). The difference in the DD slope among groups with increasing duration of drive is significant (p = 0.0001, f = 36.2), as it is among [Cal, groups (p = 0.0001, f = 74.8). The interaction between drive duration and [Cal, groups is also significant (p

= 0.0003, f = 15.2). There was a significant difference in DD slope between group 1 and 2 (p = 0.3), 2 and 3 (p = 0.004), 3 and 4 (p = 0.003), 4 and 5 (p = 0.02), 5 and 6 (p = 0.003), 6 and 7 (p = 0.0003), 7 and 8 (p = 0.002), and 8 and 9 (p = 0.0001). In Figure 5 (B), the difference in V,, amplitude among the 10.8 mM Ca groups is not significant, whereas the difference in time to peak is (p = 0.001, F = 3.6). Also, group 1 was statistically different from all other groups as was group 2.

Decay of the Prolonged V,, at Different [Cal,

Depolarization

In some traces, during 30-second pauses DD peaked and then the membrane repolarized to some extent, suggesting that a decaying V,, was superimposed on DD. To better visualize this phenomenon, the preparations were perfused in different [Cal, and the duration of the pause was prolonged up to 120 seconds. As shown in Figure 6, a 5-minute 120/min overdrive was carried out in 1.35 (A), 2.7 (B) and 10.8 (C) mh4 [Cal,. When the 120/min overdrive was interrupted, DD slope became steeper as [Cal, was increased and a V,, and an aftercontraction in the highest [Cal, (arrowheads) occurred. Also, DD

Calcium and Diastolic Depolarization Cm Cl

Fig. 6. Decay of the prolonged depolarization V,, after overdrive in different [Cal,. The action potential and twitch curves are shown in the lefthand panels at the [Cal” indicated. The preparation was driven at 60/min (second panel in each strip) and overdriven at 120/min for 5 min. The cessation of the overdrive is shown in the third panel in each strip where the straight lines under DD emphasize V,, The arrowheads point to V,,, and the aftercontraction.

Tamargo and Vassalle

355

mM 0’

B 2.7

::

li

J II!

-A-

OO/mln peaked after several seconds during the pause, and then the potential slowly returned toward the control value, as emphasized by the straight lines drawn below the DD traces. In four experiments, in 2.7 mM [Cal,, V,, involved a depolarization of 2.5 5 0.4 mV (p < 0.01) (with respect to the steady value) and peaked after 36.9 -t 3.4 seconds after the end of drive. The corresponding values in 10.8 mM [Cal, were 3.7 _’ 0.3 mV (p < 0.01) and 43.9 + 2.4 seconds, respectively.

DD Slope and Amplitude Drive and Calcium Load

l

as a Function

of

To load the fibers at different [Cal, during overdrive and yet separate (at least partially) the effects induced by overdrive from those of different [Cal,, the preparations were driven at 7.5/min and overdriven at 120/min for 3-4 minutes but during (and only during) overdrive [Cal, was 2.7 mM (control), 1.35 or 10.8 mM. In this fashion, [Cal, before and after overdrive was the same in each test (ie, 2.7 mM), and changes seen during recovery should be

120/mln

mainly related to the degree of calcium loading during overdrive. As shown in Figure 7, during overdrive [Cal, was 1.35 mM (A), 2.7 mM (B), and 10.8 mM (C). It is apparent that during overdrive the maximum diastolic potential (E,,,) increased more when [Cal, was higher. After overdrive in (A), DD slope and amplitude were increased enough to cause one spontaneous beat and then progressively declined. In (B), the changes were exaggerated and nine spontaneous beats were present. In (C), DD slope was even steeper due to the appearance of a V,, and a superimposed V,, that affected both E,,, and the value of DD just prior to each excitation. The V,, disappeared after the eighth beat. No spontaneous beats were present in spite of the large DD, as calcium loading apparently shifted the threshold potential.” In Figure 7(D), (n = 5) overdrive increased DD slope above control value whether [Cal, during overdrive was changed or not. However, DD slope increased more and initially decayed more rapidly when [Cal, was higher during overdrive. The opposite finding was present when [Cal, was low. The amplitude of DD behaved somewhat differently (E) . In high [Cal,, it was larger and decreased slowly

356

Journal

of Electrocardiology

Vol. 24 No. 4 October

1991 Fig. 7.

2.7

6

with successive beats, whereas in low [Cal,, it was slightly smaller than that in normal [Cal,. The different behavior of DD slope and amplitude in high [Cal, is likely to reflect the more rapid decay of V,, after overdrive as calcium overload is quickly reduced. Analysis of variance shows that the difference in DD amplitude among the consecutive beat groups is significant in 1.35 mM [Cal, (p = 0.002, f =4.65), in 2.7 mM [Cal, (p = 0.004, f = 4.0), and in 10.8 mM [Cal, (p = 0.0004, f = 5.9). The difference between groups 1, 2, 3, and 4 versus the last group is significant in all Ca solutions. In fact, in 10.8 mM [Cal,, the difference between the first seven groups and the last group is significant. Another way to increase [Cali (at the same [Cal,) is to inhibit the Na-K pump by means of strophanthidin since this in turn increases intracellular sodium activity (ak,)13-” and therefore a&, through the Na-Ca exchange.16,‘7 In one experiment, DD slope and amplitude increased when the driving rate and [Cal, were increased in the absence and more so in the presence of 0.1 PM strophanthidin. In 10.8 mM [Cal,, the values of DD slope and amplitude at 60/min were fairly close to those at 120/min, suggesting that these parameters reach a plateau when the preparation is calcium overloaded. In fact, in

Effects of different calcium loads applied during overdrive. The preparation was driven at 7.5/min and overdriven at 12O/min for 4 min. During overdrive, [Cal, was below normal (1.35 rnM, [A]), normal (2.7 mM, [B]) or above normal (10.8 mM, [Cl). The graphs show on the ordinate the slope (D) 0; lmplitude of DD (E) for the control beats (isolated symbols) or for each of the consecutive beats after overdrive. The abscissae indicate the controls and the number of consecutive beats. The numbers next to the respective curves indicate [Cal, during overdrive. All values after overdrive were recorded in 2.7 mM [Cal, at the same rate as control and the different symbols show the recovery after overdrives carried out in different [Cal,. An asterisk next to a symbol indicates a p < 0.05 with respect to control values.

high [Cal, plus strophanthidin, even a 15/min drive was associated with V,,. In two more experiments in which only the 120/min drive was tested, in the presence of 0.1 p,M strophanthidin the drive increased the amplitude of DD in one and induced spontaneous discharge in the other experiment.

Effect of [Cal, During a Slow Drive and Quiescence Changing [ Ca] ,, should modify [ Ca] extracellularly sooner than intracellularly, thus possibly initially inducing different changes from those in the steadystate. This was studied by changing [Cal, while the preparations were driven at a slow rate (7.51min) so that the changes in E,,, as well as in DD slope and amplitude could be followed on a beat-to-beat basis. In Figure 8, when [Cal, was increased from 2.7 mM to 10.8 mM, E,,, began to be more negative within two beats (B). The slope of initial DD progressively steepened while the steady-state value of DD at first became more negative and then less negative than control. Because of the shift of E,,, in a negative direction and the shift of steady-state DD in a positive direction, the amplitude of DD increased

Calcium

C,a 10.8

m,M

Co

and Diastolic

1.35

Dz

C.,

h.351----_

Ca

10.8

mu

Depolarization

l

Tamargo

and Vassalle

357

2 min

-

p] 1OmV

Co

1.35

mM-

Fig. 8.

Transient and steady-state effects of changing [Cal, on transmembrane potentials in active and quiescent fibers. (A) The action potentials are shown at different time bases in control and in high [Cal,. (B) [Cal<, was increased from 2.7 mM to 10.8 mM as indicated below the traces showing the bottom part of the action potentials and the twitch curves. In the same fiber (but not driven and quiescent), [Cal0 was increased to 10.8 mM (C) or decreased to 1.35 mM (D). (E) Another fiber was driven at 6/min and [Cal, was changed from 1.35 mM to 10.8 mM, as indicated next to the traces.

by + 39%. It should be noted that the initial negative shift of steady-state DD occurred when force was not yet increased. The beginning of the second (B) panel shows that E,,, had grown even more negative and the twitch was also greater. As recovery in 2.7 mM [Cal, began, the steady-state DD shifted initially to less negative values while there was a progressive decrease in both E,,, and twitch amplitude. When the drive was interrupted and the preparation was quiescent (Fig S[C]), 10.8 mM [Cal, induced a transient hyperpolarization followed by a depolarization. At the beginning of recovery, a transient depolarization occurred. In (D), decreasing [Cal, from 2.7 to 1.35 mM caused a transient depolarization followed by a transient hyperpolarization during recovery. When [Cal, was changed from 1.35 mM to 10.8 mu (Fig. 8[El), E,,, increased promptly when force had not yet increased. The increase in E,,, was subsequently associated with a V,,, a less negative steady-state DD, and an increase in force. The record at higher speed (arrowheads pointing downward) shows quickly damped V,, following the action potential. Returning to low [Cal, rapidly induced a transient depolarization at the time when neither V,, nor force were as yet reduced. It is of interest that in low [Cal,, voltage oscillations (arrowheads point-

ing upward) grew during diastole (in contrast to V,, in high [Cal,) and actually led to the onset of spontaneous activity. The amplitude of DD and of twitch was already decreased at that time. In six experiments, 10.8 mM [Cal, initially hyperpolarized E,,, by 3.2 ? 0.6 mV (p < 0.001) during the slow drive and the resting potential by 4.5 2 0.4 mV (p < 0.001) during quiescence. The total amplitude of DD increased from 12.5 +- 0.8 mV to 17.6 ? 1.4 mV in high [Cal, (+40.8%, p < 0.001). Exposure to 1.35 mM [Cal, transiently depolarized Emax by 3.6 -+ 0.5 mV (p < 0.001) during the slow drive and the resting potential by 4.2 +- 0.8 mV (p < 0.001) during quiescence.

Discussion The present results show that increasing calcium modifies diastolic depolarization of Purkinje fibers through several mechanisms, including a steepening of the initial DD, a more negative E,,,, an increase in DD amplitude, and the development of V,, and of V,,. The systematic interruption of the drive under different conditions made it possible to demonstrate that [Cal, can alter the slope of DD directly and

358

Journal of Electrocardiology

Vol. 24 No. 4 October 1991

through a superimposed V,, and that high [Cal, increased the amplitude of DD through both a more negative maximum diastolic potential and a less negative steady-state potential. At any given [Cal,, slope and amplitude of DD are a function of driving rate and duration of overdrive. Also, changing [Cal, has immediate effects on both resting and active membranes that are different from those in the steadystate. Diastolic depolarization is influenced by both the external and internal [Cal. However, the effects of [Cali predominate, as shown by the fact that a change in rate mimics the effects of a change in [Cal, and that similar results are obtained when calcium loading is applied only during overdrive or in the presence of strophanthidin. The calcium-induced repetitive activity is caused by these various mechanisms, but a shift in the threshold potential has a predominant role in low [Cal,. Thus, diastolic depolarization of Purkinje fibers is modulated by calcium through different mechanisms that may operate also at normal [Cal, in conditions that modify [Cali (eg, rate changes).

[Cal, and Diastolic Depolarization The present findings show that, at a given rate, high [Cal, steepens diastolic depolarization in sheep as it does in dog Purkinje fibers driven at 9Wmin.l In spontaneously discharging calf and sheep Purkinje fibers, high [Cal, did not change the slope of DD,12 but this could be due to the fact that high [Cal, decreased the spontaneous rate.’ Also, in the present experiments, V,, was larger in higher [Cal, and peaked sooner at faster rates in sheep as it does in dog Purkinje fibers.2,3 And, in retrospect, it is apparent that the transient “local depolarization” observed on interruption of the drive in 10.8 mM [Cal,’ was also a V,,. The results of this study show that the immediate effects of an increase in [Cal, are a more negative E,, and steady-state DD. The fact that these changes occur within a few beats when force is still low suggests that they result from an action of [Cal,. This conclusion is supported by the finding that high [Cal, also induces a transient hyperpolarization in quiescent fibers. Taken together, the findings might indicate an increase in potassium conductance,‘,” but, if so, this effect is quickly overcome by other changes. Thus, DD of active fibers and the resting potential of quiescent fibers (but not E,,,) soon become less negative. High [Cal, is known to modify the delayed rectifier current Ik, ’ 9 but this cannot be the explanation of the findings since the hyperpo-

larization was also present in fibers quiescent at the resting potential. Also, an increase in IK does not necessarily result in an increase (and may cause a decrease) in E,,,. It should be noted that in resting’5*20,21 and active fibers,22 high [Cal, decreases and low [Cal, increases ak through the Na-Ca exchange. However, this effect could not be immediate since the changes in ah, are relatively slow. This implies that in slowly driven fibers the first effects are due to a change in Ca at the outer aspect of the membrane, whereas the subsequent changes might be influenced by slowly changing intracellular concentrations of calcium, sodium, or both. Furthermore, the events in quiescent and active fibers are not necessarily similar since high [ Ca] 0 increases radioactive potassium movements in active Purkinje fibers but decreases them in quiescent Purkinje fibers.23 It should also be noted that electrical stimuli were applied by means of a point source and not by field stimulation, and therefore release of catecholamines are not expected to play a role in the observed changes. As for a release of catecholamines by high [Cal,, in Purkinje fibers high [Cal, slows spontaneous discharge and shifts the threshold potential to less negative values12; if high [Cal, acted by releasing catecholamines, it would have had opposite effects.

High [Cal,, Driving Rate, and V,, Increasing [Cal, should increase [Cali in the steady-state. Force certainly increases as a result of a larger I,i24 acting on larger calcium stores. Therefore, changes in [Cali should contribute to the changes in the slope of DD for at least two reasons: ( 1) V,, results from intracellular calcium loading3,25 and (2) intracellular injection of calcium affects the pacemaker current.9 The oscillatory potential is believed to be due to a release of calcium from an overloaded SR in the presence25 and in the absence4 of cardiotonic steroids. Therefore, it is not too surprising that V,, should appear in the presence of high [Cal,.‘-’ However, the present results show that a V,, can also appear at normal [Cal, if the rate of discharge is suitably high and may not be present even in high [Cal, if the driving rate is suitably low. This clearly indicates that (for a given [Cal,) the rate of discharge conditions [Cali. These findings provide the explanation for the decrease in the DD slope when the driving rate was decreased from 125 to 95 and 751

Calcium

and Diastolic

min in 10.8 mh5 [Cal,’ and show that such a decrease in slope occurs also at normal [Cal,. The similarities in the effects of high [Cal, and faster driving rates elicit the question as to why the two interventions should have similar effects. The answer appears to be that both interventions increase [Cali. Thus, an increased rate of discharge increases calcium exchangez6 and also [Ca]i.27’28 Furthermore, an increased rate of discharge increases aL 29*30,which in turn increases cellular calcium through the Na-Ca exchange.16,” In fact, the increase in rate should also increase [Cali directly because of the more numerous action potentials (and therefore more numerous activations of Isi) in the unit of time. This increase occurs even if during each action potential I,i is smaller due to the shorter diastolic interval available for reactivation.3’ The shorter diastole results in an overall increase in the time the membrane is depolarized at the plateau with a consequent loading of calcium stores32,33 and in a reduced extrusion of calcium through the Na-Ca exchange. 16,17The net result is that an increase in rate increases a C,. 5 The development of V,, at normal [Cal, with a suitably fast rate is in agreement with the finding that repeated short clamps induce or enhance4 the transient inward oscillatory current (L) underlying V,,. The lack of V,, in high [Cal, when the driving rate is suitably low may be accounted for by the fact that while a low rate diminishes the numbers of I,i activations in the unit of time it also increases the time (diastole) when most of the extrusion of intracellular calcium occurs. In addition, a slower rate leads to a decreased a&a,29,3owhich also decreases [ Ca] i .

Direct Effects of Calcium Depolarization

on Diastolic

As a result of these findings, the question arises as to whether a change in [Cali acts only by inducing a V,, superimposed on DD or if it also alters the process underlying diastolic depolarization. In this regard, the results show that a change occurs in DD even at the lowest [Cal, and the lowest rates. Since apparently there is no V,, at normal [Cal, when the rate is normal, it seems unlikely that decreasing [Cal, at a rate below normal should change DD by eliminating a V,,. One could object that a shallower V,, is present at normal [Cal, or even at low [Cal,; it might be indistinguishable from DD even during the interruption of the drive if V,, only slightly accelerates DD but fails to have a distinct peak. This has been ruled

Depolarization

l

Tamargo

and Vassalle

359

out by exploiting one of the characteristics of V,,. Under voltage clamp conditions, the appearance of V,, requires a previous depolarization to at least - 20 mV; if hyperpolarizing clamps are applied from the resting potential, there is no I,, superimposed on the pacemaker current.4 When hyperpolarizing clamps are applied from the resting to more negative potentials in the presence of low calcium (0.54 mM), the pacemaker current is smaller and the time constant of its decay is greater. These changes could not be due to the disappearance of V,,, for no I,, existed under the same conditions in normal [Cal,,.’ This finding indicates that calcium also influences DD through a direct effect on the pacemaker current, in agreement with the results of Isenberg who found that Ca injection increases the pacemaker current.’ However, this point is not agreed upon since no increase in pacemaker current by calcium has been reported by others. 34

DD Amplitude

and Intracellular

Calcium

V,, increases the slope of the initial diastolic depolarization since I,, is inward.4,25 However, V,, cannot explain the increase in amplitude of steadystate DD since V,, (and the underlying I,,) peaks and decays within a few seconds2-4 whereas DD was less negative than control at the end of 30-second pause. In addition, high [Cal, altered DD even at low rate when no V,, was apparent and decreased the resting potential in quiescent fibers. One possible explanation for these findings is that the neutralization of membrane fixed charges modifies the pacemaker current in such a fashion as to increase the net inward current during diastole. This explanation does not apply to the prolonged depolarization since V,, decays with time. Rather, V,, might be related to the extrusion of calcium through an electrogenic Na-Ca exchange,36 which may create a transient inward current during diastale. Thus, during quiescence (as [Cali decreases), the electrogenic extrusion of intracellular calcium should decrease with time as should the depolarization caused by such an extrusion. The relation of V,, to calcium is suggested by the fact that calcium overload causes not only an I,, but also a tail inward current (Lx) upon which I,, is superimposed.’ 1,36-38 The tail inward current lasts longer than I,, and both are decreased by low [Cal, and increased by high [Cal,.” Therefore, these two phenomena are related to calcium overload but they may have a different origin since high doses of caffeine abolish I,, but increase I,, . 36-38

360 Journal of Electrocardiology

Vol. 24 No. 4 October 1991

It should be noted that an increase in [Cal, shifts the voltage dependence of the fast Na current in a depolarizing direction; greater depolarization is required in spontaneously active fibers to reach the threshold for activation. l2 However, the amplitude of diastolic depolarization at different [Cal, was studied during a period of quiescence when the threshold potential was not attained. Therefore, the diastolic depolarization reached the steady-state value that was determined by the [Cal, tested (independently of shifts in threshold potential).

References 1.

Res 20:32, 1967

2. Musso E, Vassalle M: The role of calcium in overdrive

3.

4. 5.

Conclusion 6.

The present results show that some of the modifications induced by calcium are related to effects occurring predominantly at the outer aspect of the cell membrane whereas most modifications are related to an increase in [Cali. Because the increase in [Cal, (as indicated by changes in contractile force and by the manifestations of calcium overload) plays a major role, the modifications induced by high [Cal, can be reproduced by increasing the rate or the duration of the drive at normal [Cal,. Reciprocally, the effects of high [Cal, on DD can be substantially reduced by decreasing the driving rate. If the changes depended on [Cal,, it would have made no difference that the rate was changed. Both high [Cal, (through V,, and V,,) and low [Cal D can induce spontaneous discharge, but in low [Cal, it is the negative shift in threshold potential (and not a steeper and larger DD) that is responsible for the initiation of the activity. Thus, Ca can modify diastolic depolarization in Purkinje fibers either by modifying the pacemaker potential itself or by inducing abnormal events (V,, and V,,) (or by both mechanisms). Because of the influence of [Cali on diastolic depolarization, the pacemaker potential (and pacemaker activity) of Purkinje fibers is influenced also by those factors that modify [Cali at normal [Cal,, such as a change in the rate of discharge of the sinus node and the onset of tachyarrhythmias.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

Acknowledgment

17. 18.

The authors thank Mr. Ping-Wu Li for his help with statistical elaboration of some of the data.

Temte JV, Davis LD: Effect of calcium concentration on transmembrane potentials of Purkinje fibers. Circ

suppression of canine cardiac Purkinje Fibers. Circ Res 51:167, 1982 Ferrier GR, Moe GK: Effect of calcium on acetylstrophanthidin-induced transient depolarizations in canine Purkinje tissue. Circ Res 33:508, 1973 Vassalle M, Mugelli A: An oscillatory current in sheep cardiac Purkinje fibers. Circ Res 48:618, 1981 Lado MG, Sheu S-S, Fozzard HA: Changes in intracellular Ca* + activity with stimulation in sheep cardiac Purkinje strands. Am J Physiol 243:H133, 1982 Vassalle M: Analysis of cardiac pacemaker potential technique. Am J Physiol using a “voltage-clamp” 210:1335, 1966 Valenzuela F, Vassalle M: Overdrive excitation and cellular calcium load in canine cardiac Purkinje fibers. J Electrocardiol 18:21, 1985 Kotake H, Lin C-I, Vassalle M: On the mechanism of action of low Na solution on the pacemaker current (Abstr). Physiologist 26:A-90, 1983 Isenberg G: Cardiac Purkinje fibres. [Ca*+]i controls the potassium permeability via the conductance components g,, and g,,. Pfluegers Arch 371:77, 1977 Tamargo J, Vassalle M: Effects of calcium on diastolic depolarization of cardiac Purkinje fibers (Abstr). Fed Proc 46:336, 1987 Hasegawa J, Satoh H, Vassalle M: Induction of the oscillatory current by low concentrations of caffeine in sheep cardiac Purkinje fibres. Naunyn Schmiedebergs Arch Pharmacol 335:310, 1987 Weidmann S: Effects of calcium ions and local anaesthetics on electrical properties of Purkinje fibres. J Physiol (Lond) 128:568, 1955 Lee CO, Kang DH, Sokol JH, Lee KS: Relation between intracellular Na ion activity and tension of sheep cardiac Purkinje fibers exposed to dihydroouabain. Biophys J 29:315, 1980 Lee CO, Uhm DY, Dresdner K: Sodium-calcium exchange in rabbit heart muscle cells: direct measurement of sarcoplasmic Ca*+ activity. Science 209:699, 1980 Sheu S-S, Fozzard HA: Transmembrane Na+ and Ca* + electrochemical gradients in cardiac muscle and their relationship to force development. J Gen Physiol 80:325, 1982 Mullins LJ: The generation of electrical currents in cardiac fibers by Na/Ca exchange. Am J Physiol 236:C103, 1979 Chapman RA: Control of cardiac contractility at the cellular level. Am J Physiol 245:H535, 1983 Isenberg G: Cardiac Purkinje ftbres. [Ca2+11 controls steady state potassium conductance. Pfhtegers Arch 371:71, 1977

Calcium and Diastolic Depolarization 19. Kass RS, Tsien RW: Control of action potential duration by calcium ion in cardiac Purkinje fibers. J Gen Physiol 67:599, 1976 20. Ellis D: The effects of external cations and ouabain on the intracellular sodium activity of sheep heart Purkinje fibres. J Physiol (Lond) 273:211, 1977 21. Deitmer JW, Ellis D: Changes in the intracellular sodium activity of sheep heart Purkinje tibres produced by calcium and other divalent cations. J Physiol (Lond) 277:437, 1978 22. Lee CO, Vassalle M: Modulation of intracellular Na+ activity and cardiac force by norepinephrine and Ca’+. Am J Physiol 244:CllO, 1983 cal23. Musso E, Vassalle M: Effects of norepinephrine, cium and rate of discharge on 42K movements in canine cardiac Purkinje fibers. Circ Res 42:276, 1978 24. Reuter H: The dependence of slow inward current in Purkinje fibres on the extracellular calcium concentration. J Physiol (Lond) 192:479, 1967 25. Kass RS, Tsien RW, Weingart R: Ionic basis of transient inward current induced by strophanthidin in cardiac Purkinje fibres. J Physiol (Lond) 281:209, 1978 26. Grossman A, Furchgott RF: The effects of frequency of stimulation and calcium concentration on Ca45 exchange and contractility on the isolated guinea-pig auricle. J Pharmacol Exp Ther 143: 120, 1964 27. Sands DS, Winegrad S: Treppe and total calcium content of the frog ventricle. Am J Physiol218:908, 1970 coupling. 28. Langer GA: Heart: excitation-contraction Annu Rev Pharmacol Toxic01 35:55, 1973 29. Cohen CJ, Fozzard HA, Sheu S-S: Increase in intra-

30

31

32.

33.

34.

35.

36.

37.

38.

l

Tamargo and Vassalle

cellular sodium ion activity during stimulation in mammalian cardiac muscle. Circ Res 50:651, 1982 Abete P, Vassalle M: Relation among Na+-K+ pump, Na+ activity and force in strophanthidin inotropy in sheep cardiac Purkinje fibers. J Physiol (Lond) 404:275, 1988 Gibbons WR, Fozzard HA: Slow inward current and contraction of sheep cardiac Purkinje fibers. J Gen Physiol 65:367, 1975 Fabiato A: Simulated calcium current can both cause calcium loading in and trigger calcium release from the sarcoplasmic reticulum of a skinned canine cardiac Purkinje cell. J Gen Physiol 85:291, 1985 Kotake H, Vassalle M: Rate-force relationship and calcium overload in canine Purkinje fibers. J Mol Cell Cardiol 18:1047, 1986 DiFrancesco D, McNaughton PA: The effects of calcium on outward membrane currents in the cardiac Purkinje fibres. J Physiol (Lond) 289:347, 1979 McAllister RE, Noble D, Tsien RW: Reconstruction of the electrical activity of cardiac Purkinje fibres. J Physiol (Lond) 251:l. 1975 Satoh H, Hasegawa J, Vassalle M: On the characteristics of the inward tail current induced by calcium overload. J Mol Cell Cardiol 2 1:5, 1989 Vassalle M, DiGemraro M: Caffeine actions on currents induced by calcium-overload in Purkinje fibers. Eur J Pharmacol 106: 12 1, 1984 Hasegawa J, Vassalle M: Enhancement and suppression of currents related to calcium-overload by different concentrations of methylxanthines. Arch Int Pharmacodyn Ther 282:68, 1986