Gen. Physiol. Biophys. (1983), 2,163—167
163
A Mathematical Model of Afterdepolarizations
I. T I P Ä N S and
E.
LAVENDELS
Riga Polytechnical Institute, Lenin str. 1, 226355 Riga, Latvian SSR, USSR
Abstract. Conditions for the occurrence of delayed afterdepolarizations were
analyzed in a mathematical model of the action potential in atrial and ventricular
conduction system cells. First, the influence of the rapid repolarization component
of the potassium current, the slow inward (calcium) and the sodium backward
currents respectively upon the main parameters of afterdepolarizations was examined. In the second series of computation experiments also the other two
potassium current components were included. The estimation of ionic currents was
carried out in order to determine the oscillation zones of transmembrane potential.
Key words: Mathematical models — Action potential — Afterdepolarizations
— Heart muscle — Oscillations
Introduction
When developing mathematical models of action potentials in atrial and ventricular
conduction system cells we noted (Lavendels et al. 1981) that at certain values of
ionic current conductances after the ordinary action potential, the depolarization
and repolarization processes are followed by another depolarization. In some cases
this depolarization reached the threshold and another action potential occurred. It
was also noted that, at some values of ionic current conductances, not one but
a train of oscillations of transmembrane potential occurred in cells which normally
do not show automaticky. According to Cranefield (1977) these delayed afterdepolarizations may cause tachycardias, characterized by accelerated rhythm and
transition of sites of pacemaker activity. The aim of this work was to determine the
influence of ionic current conduction upon the main parameters of afterdepolarization, the maximum level and duration, diastolic potential as well as to
estimate the range of conduction in which these afterdepolarizations occur.
Methods
We started with the model of atrial and ventricular cells, developed by Lavendels et al. (1981) and
derived from the model (McAllister et al. 1975) of Purkinje fibres membrane action potential in which
only the values of ionic currents were varied; terminology and all other values of kinetic variables,
Tipans and Lavendels
164
Fig. 1. Transmembrane potential oscillations at different values of coefficients k,„ kxi, £N.. b.
reversal potentials etc. remained the same as in the model of McAllister et al. (1975). To denote the
variations of currents, /), the coefficient kj, was used,
i'i = k,. i j ,
where i\ is current which was varied.
The values of ionic currents and transmembrane potential were calculated by solving the 10-th order
system of nonlinear differential equations described by McAllister et al. (1975) using the Runge-Kutta's
procedure with variable integration step. For the calculation an EC-1050 computer was used, and the
total computation time of one action potential was about 15 minutes.
Results and Discussion
First of all, only some currents — slow inward /si, sodium backward iN«, b and the fast
component of potassium repolarization current i*i — were varied with k* = 0.5; the
values of kxi and kN*,b are shown in Table 1, and the corresponding changes of
transmembrane potential as the function of time are presented in Fig. 1. It should
be noted that in all cases the shape of the action potential remained the same
during the first 200 ms. With k^ = 0.5 and kxi = 3 the afterdepolarizations occurred
when ArNa.be(0.55; 0.8). It is to be noticed that an increase in kN,,b changes the
value of the resting potential from —80 mV to —45 mV. It was experimentally
shown (Morton and Arnsdorf 1977) that there are two stable values of resting
potential and that it is possible to induce transition from one value to another by
the action of procainamide on Purkinje fibres. Within a narrow range of
A:N»,bG (0.55; 0.7) several transmembrane potential oscillations are generated
instead of one afterdepolarization. The highest amplitude was observed at kN., b =
0.6. With the increase of &N«,b the frequency of oscillations increased, but their
amplitude decreased.
v.
Afterdepolarization Modelling
165
Table 1. Coefficients of variations of ionic currents
No
ÄNa. b
1
2
3
4
5
6
7
8
3
0.6
3
0.62
3
0.55
2.7
0.7
3
0.8
3.3
0.9
3.3
0.7
3.1
07
When £Na,b was further increased, the oscillations ceased and only one
afterdepolarization was observed, again, followed by a low resting potential of
- 5 0 mV. Finally, at &Na,b = 0.8, no afterdepolarization could be induced. If in the
existence area of afterdepolarizations the potassium current component ixi was
increased (kxi = 3.1 and kxi = 3.3), the amplitude of the action potential following
afterdepolarization was increased. When ixi and iNa, b were increased simultaneously (£xi = 3.3; &Na,b = 0.9) the afterdepolarizations were absent and a low level of
resting potential was reached. It should be noted that an action potential of the
same shape as in the previous case exists when kxi=2.7 and /tNa,b = 0.7.
Therefore a more general question arises concerning the possibility of the
existence of afterdepolarizations at other sets of values of ionic currents.
To answer this question, two series of computation experiments were performed. First, 20 experiments with ksi, kxU ArNa,b varied as follows
ksi e (0.4; 0.6);
fc:e(2.8; 3.2);
W e (0.5; 0.7),
i.e. in the zone certainly including the oscillations and in another, which was wider,
where
**e(0;l)
kxl e (2; 4)
*x2e(l;2)
*Kie(0;2)
ArNa, b e ( 0 ; 2)
consisting of 40 experiments where two other potassium current components — kKi
and kx2 were also changed.
In the first series of experiments afterdepolarizations were always encountered. A common feature of all of them was that each cycle of depolarization was
characterized by an increased frequency and a decreased amplitude. The correspondance of repolarization and depolarization levels is also remarkable, with
Tipäns and Lavendels
166
Table 2. The relation between the main parameters of the action potential and coefficients of the ionic
currents Ic, kmi, k&, kkU kN„ b
NO
fc,
fc,l
kx2
fckl
fcta,
b
Eäp
1.05
1.95
1.8
1.0
1.6
1.2
0.8
1.3
1.65
0.5
0.45
1.25
0.65
0.95
0.55
0.05
1.4
1.5
0.15
1.7
0.3
1.75
0.75
0.35
1.15
1.1
0.7
0.2
1.35
0.9
0.4
1.9
0.6
2.00
1.45
0.25
0.1
1.55
1.85
0.85
36
34
29
27
38
44
27
36
38
33
34
41
24
32
34
39
36
31
31
24
45
37
34
32
50
39
39
29
31
30
35
28
34
31
44
40
32
31
47
44
fid.
to
mv
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
0.125
0.1
0.55
0.175
0.6
0.45
0.05
0.5
0.525
1.0
0.775
0.9
0.025
0.3
0.275
0.475
0.975
0.225
0.425
0.075
0.7
0.15
0.575
0.75
0.95
0.85
0.925
0.325
0.2
0.825
0.675
0.65
0.35
0.625
0.8
0.4
0.875
0.25
0.725
0.375
2.3
3.25
3.05
4.0
3.3
2.75
2.15
2.05
3.9
2.8
3.1
3.2
3.35
2.85
3.65
3.7
3.15
3.5
3.4
2.6
3.75
2.7
3.6
2.55
3.45
2.45
3.85
2.5
2.65
2.1
2.25
2.2
2.95
2.9
2.35
2.4
3.0
3.95
3.8
3.55
1.8
1.625
1.575
1.375
1.975
1.125
1.425
1.15
1.275
1.45
2.0
1.875
1.325
1.075
1.95
1.5
1.475
1.025
1.2
1.35
1.675
1.525
1.05
1.1
1.225
1.725
1.55
1.85
1.925
1.9
1.4
1.75
1.775
1.175
1.25
1.3
1.6
1.7
1.65
1.825
0.35
0.85
2.0
1.350
0.55
0.05
1.25
0.75
0.5
1.7
1.15
0.6
1.9
1.05
0.65
0.3
1.3
1.45
1.5
1.75
0.2
0.4
0.95
1.6
0.15
0.7
0.8
1.4
1.2
1.8
1.0
1.95
0.9
1.55
0.45
0.25
1.65
1.1
0.1
0
-40
-29
-34
-49
-47
-28
-51
-31
-33
-92
-87
-63
-87
-44
-66
-93
-35
-43
-100
-41
-82
-28
-65
-97
-29
-32
-73
-99
-40
-73
-86
-34
-64
-32
-26
-77
-102
-40
-22
-36
tä.
-ms
105
70
63
40
120
140
30
115
120
85
85
125
25
65
75
100
105
50
50
30
113
115
85
80
150
120
105
45
55
80
95
75
74
85
150
120
75
60
140
115
250
275
350
225
475
250
275
450
340
350
350
850
275
305
"oscillations"
325
365
335
225
385
"oscillations"
525
"oscillations"
365
340
405
"oscillations"
280
360
"oscillations"
355
335
"oscillations"
370
465
"oscillations"
285
330
380
280
fe — time of reaching the zero level of transmembrane potential, Et, — resting potential, Eäp — maximal
depolarization level, h, — approximate duration of the action potential.
Afterdepolarization Modelling
167
higher depolarization spike occurring when the repolarization reached more
negative values. In order to exceed the zero level of transmembrane potential the
preceding maximum repolarization level should be less negative than — 70 mV. If
this condition was not satisfied the following depolarization levels were still
negative and the steady state value of transmembrane potential was of about
—45 mV. The results of the second series of experiments are presented in Table 2.
Oscillations of transmembrane potential were observed in seven experiments. It is
significant that they occurred at the set of values of ksi, kxi and km, b which differ
considerably from those in the previous series of experiments, where oscillations
always existed. It is therefore supposed that the zone of oscillations is not as narrow
as expected before or that there exist more than one such zone.
The existence of similarly shaped action potentials at different values of ionic
currents means that in some cases the same electric properties of cell membranes
may be achieved by distinct action of different pharmacologic agents.
References
Lavendels E. E., Matschabeli L. I., Tipäns I. O. (1981): Construction of a model on the spread of
excitation in the conducting system of the heart. In: Theory and Applications of Automatization of
the Electrocardiological Clinical Research (Ed. Z. Yanushkevichus), pp. 252—254, Kaunas (in
Russian)
Cranefield P. J. (1977): Action potentials, afterpotentials and arhythmias. Circ. Res. 41, 415—423
Mc Allister R. E., Noble D., Tsien R. W. (1975): Reconstruction of the electrical activity of cardiac
Purkinje fibres. J. Physiol. (London) 251, 1—59
Morton J., Arnsdorf (1977): The effect of antiarhythmic drugs on triggered sustained rhythmic activity
in cardiac Purkinje fibres. J. Pharmacol. Exp. Ther. 201, 689—700
Received August 23, 1982 / Accepted December 6, 1982