Vol. 274, Issue 6, H1902-H1913, June 1998
Determinants of action potential initiation in isolated rabbit
atrial and ventricular myocytes
David A.
Golod,
Rajiv
Kumar, and
Ronald W.
Joyner
Todd Franklin Cardiac Research Laboratory, The Children's Heart
Center, Department of Pediatrics, Emory University, Atlanta,
Georgia 30322
 |
ABSTRACT |
Action potential
conduction through the atrium and the ventricle of the heart depends on
the membrane properties of the atrial and ventricular cells,
particularly with respect to the determinants of the initiation of
action potentials in each cell type. We have utilized both current- and
voltage-clamp techniques on isolated cells to examine biophysical
properties of the two cell types at physiological temperature. The
resting membrane potential, action potential amplitude, current
threshold, voltage threshold, and maximum rate of rise measured from
atrial cells (
80 ± 1 mV, 109 ± 3 mV, 0.69 ± 0.05 nA,
59 ± 1 mV, and 206 ± 17 V/s, respectively; means ± SE) differed significantly (P < 0.05) from those values measured from ventricular cells (82.7 ± 0.4 mV, 127 ± 1 mV, 2.45 ± 0.13 nA, 46 ± 2 mV,
and 395 ± 21 V/s, respectively). Input impedance, capacitance, time
constant, and critical depolarization for activation also were
significantly different between atrial (341 ± 41 M
, 70 ± 4 pF, 23.8 ± 2.3 ms, and 19 ± 1 mV, respectively) and ventricular
(16.5 ± 5.4 M, 99 ± 4.3 pF, 1.56 ± 0.32 ms, and 36 ± 1 mV, respectively) cells. The major mechanism of these differences is
the much greater magnitude of the inward rectifying potassium current
in ventricular cells compared with that in atrial cells, with an
additional difference of an apparently lower availability of inward Na
current in atrial cells. These differences in the two cell types may be
important in allowing the atrial cells to be driven successfully by
normal regions of automaticity (e.g., the sinoatrial node), whereas
ventricular cells would suppress action potential initiation from a
region of automaticity (e.g., an ectopic focus).
electrophysiology; arrhythmia; inward rectifier current; cell
coupling
| |
INTRODUCTION |
DIFFERENCES IN resting membrane potential (RMP),
amplitude, duration, and threshold of the action potentials produced by
cardiac myocytes from different regions of the heart are important when trying to understand the underlying mechanisms responsible for conduction through the heart. As current flows ahead of
the advancing wave front through gap junctions, both the conduction
velocity and the safety factor for conduction depend critically on the efficacy of this current to depolarize cells in advance of the wave
front and bring them to their activation threshold. Thus the cellular
properties of input impedance, cell capacitance, and the voltage
threshold for activation of a net inward current are critical
components of the process for both normal and abnormal action potential
conduction. Differences in the activation properties of atrial and
ventricular cells might be expected from the very different
characteristics of action potential conduction in the two regions. The
cells of the atrial wall and septum have significant electrotonic
interactions with cells that are of the slow response, intrinsically
automatic, action potential type at the margins of both the sinoatrial
node and the atrioventricular node, whereas cells within the
ventricular wall do not normally interact electrotonically with cells
of the slow response type or with cells of high intrinsic automaticity.
Although there have been numerous publications investigating action
potential initiation properties and waveforms of cells isolated from
the ventricle, sinoatrial node, and atrioventricular node, there is
much less information available from studies of cells isolated from the
atrial walls and septum, which make up most of the atria. Data from
whole cell voltage-clamp experiments on atrial cells have generally
been performed at room temperature and have utilized various
pharmacological conditions and pulse protocols to isolate a single
ionic conductance or transport system for study. These voltage-clamp
studies have shown several fundamental differences between the ionic
conductances of atrial cells compared with ventricular cells. Hume and
Uehara (7) compared myocytes isolated from guinea pig atria and
ventricles using the whole cell voltage-clamp technique at room
temperature and showed marked differences in background potassium
currents thought to be due to different gating kinetics. Giles and
Imaizumi (4) further investigated the differences in potassium currents
between cells isolated from rabbit atria and ventricles, also using the
whole cell voltage-clamp technique at room temperature. They noted that the transient outward current
(It) is larger
in atrial cells than in ventricular cells, but the inward
rectifying potassium current (IK1) is larger
in ventricular cells than in atrial cells. Whalley et al. (19) showed
that IK1 currents
in freshly isolated rabbit ventricular cells were much larger than
those in cultured rabbit atrial cells.
We have previously studied the activation properties at physiological
temperature of rabbit ventricular cells either as single isolated cells
(9, 14) or as cell pairs consisting of either two real isolated rabbit
ventricular cells (10) or one real isolated rabbit ventricular cell
coupled to a mathematical model of another cell (18). In this work we
showed that the properties of action potential initiation were
significantly altered by changes in the coupling conductance and in the
extracellular potassium concentration that could be
explained by alterations in the strength-duration relationship for the
isolated ventricular cells. Our values of RMP for the isolated
ventricular cells have been generally in the range of 80 to
86 mV, and values for the maximum rate of rise of the action
potential
(Vmax) have
been in the range of 250-400 V/s.
In the present study we have used both whole cell current- and
voltage-clamp techniques at physiological temperature to compare the
membrane characteristics and study the mechanisms responsible for
differences in action potential generation between isolated rabbit
atrial and ventricular myocytes. Data from whole cell current-clamp protocols allowed measurements of cell RMP, amplitude,
Vmax, and current
stimulus threshold for the two cell types. Further comparisons between
atrial and ventricular cells were made from calculations of input
impedance, membrane time constant, critical depolarization for action
potential generation, and membrane capacitance. Voltage-clamp protocols
for ventricular cells were used to investigate the underlying mechanisms responsible for differences in the activation properties of
the two cell types.
| |
METHODS |
Cell isolation and electrodes.
Single atrial and ventricular myocytes were prepared from adult New
Zealand White rabbits weighing 2.5-3.5 kg. The rabbits were
anesthetized intravenously with 50 mg/kg pentobarbital sodium and 500 U
heparin, the heart was rapidly extracted via thoracotomy with
artificial respiration, and the aorta was cannulated for Langendorff
perfusion. Single cells were isolated according to the methods
described previously by Hancox et al. (5). Briefly, the cannulated
heart was perfused sequentially at 37°C with a base solution plus
750 µM CaCl2 for 3 min, base
solution plus 100 µM EGTA for 4 min, and enzyme solution
for 6 min.
The intra-atrial septum was then excised, cut into thin strips, and
further digested in the recirculated enzyme solution used above, with
2% BSA added, for 10 min. Cells were isolated by triturating the
tissue strips and were then placed in a potassium glutamate solution
plus 3% BSA for 1 h at room temperature. To clean the membrane
further, cells were separated from the potassium glutamate solution by
centrifugation at 500 g for 3 min, the
supernatant was replaced with potassium glutamate plus 1 mg/ml
protease, and the centrifugation tube was placed in a shaker bath at
37°C for 5 min. The cells were again centrifuged at 500 g for 3 min, the supernatant was
replaced with the potassium glutamate solution, and the cells were
refrigerated until use.
Endocardial pieces of the right ventricular wall and intraventricular
septum were excised and placed in the potassium glutamate solution. The
tissue was then cut into small chunks, triturated, and filtered through
nylon gauze (200-µm-diameter mesh). The filtered cells were stored in
potassium glutamate and refrigerated until use.
The cells were placed in a chamber that was continuously perfused with
Tyrode solution at 2 ml/min, with the temperature always maintained at
35 ± 0.5°C. Only cells that were quiescent and had a rod-shaped
appearance were used in this study. The pipettes were pulled from
borosilicate glass that, after fire polishing, had resistances of
3-6 M when filled with the internal solution. High-resistance
seals were formed with the cell membrane by applying light suction, and
the membrane was disrupted by applying transient suction. The
junctional potential was corrected by zeroing the potential before the
pipette tip touched the cell membrane.
Solutions.
The base solution contained (in mM) 130 NaCl, 4.5 KCl, 3.5 MgCl2, 0.4 NaH2PO4,
5 HEPES, and 10 dextrose, pH 7.25. The enzyme solution contained 1 mg/ml collagenase (type IIA, Worthington), 0.07 mg/ml protease (type
XIV, Sigma), and base solution plus 240 µM
CaCl2. The potassium glutamate
solution contained (in mM) 100 potassium glutamate, 25 KCl, 10 KH2PO4,
0.5 EGTA, 1 MgSO4, 20 taurine, 5 HEPES, and 10 dextrose, pH 7.2. The normal Tyrode solution contained (in mM) 148.8 NaCl, 4 KCl, 1.8 CaCl2, 0.53 MgCl2, 0.33 NaH2PO4,
5 HEPES, and 5 dextrose, pH 7.4. The internal solution was composed of
(in mM) 135 KCl, 5 Na2CrPh, 5 MgATP, and 10 HEPES, pH 7.2.
Current- and voltage-clamp studies on isolated cells.
Membrane potentials were recorded using the whole cell patch-clamp
technique with an Axoclamp 2A dual amplifier (Axon Instruments, Foster
City, CA) in the current-clamp mode, as previously described (9, 14).
Series resistance was carefully compensated by internal bridge balance
adjustments after recording of the membrane potential was established.
For voltage-clamp studies we used an Axopatch 200 voltage-clamp
amplifier with the same external solution and pipette solution as for
the current-clamp recordings. We used a holding potential of 84
mV to approximate the measured RMP of the ventricular cells. Cell
capacitance and series resistance were measured and compensated. Step
pulses were applied from the holding potential in 2-mV steps with
durations of 50 ms and an interpulse interval of 1 s.
Statistical analysis.
Statistical analysis was performed using SigmaStat for Windows (Jandel
Scientific, San Rafael, CA). Statistical significance between atrial
and ventricular cells was determined by using Student's t-test for unpaired data.
P values <0.05 were regarded as
significant. Data are presented as means ± SE in the
text.
| |
RESULTS |
Differences in atrial and ventricular action potentials.
To examine the characteristics of action potentials generated by
isolated rabbit atrial and ventricular cells, whole cell current-clamp
studies were performed in which each cell was stimulated with current
pulses of 2-ms duration and an amplitude slightly greater than the
stimulus current threshold of that cell (Fig. 1A).
The stimulus frequency was set to a physiological rate of 3 Hz, and all
experiments were performed at 35°C. The data for the ventricular
cell in Fig. 1 are indicated by dotted lines, whereas the data for the
atrial cell are indicated by solid lines. Figure
1A shows that the current required
to initiate an action potential in the two cells is very different. The
threshold current for the ventricular cell was 2.6 nA, whereas that for
the atrial cell was 0.62 nA for this short stimulus duration. Figure
1B shows the action potentials
generated by the rabbit atrial and ventricular cells in response to the
currents plotted in Fig. 1A. As
expected from previous studies (see introduction), there are marked
differences between the atrial and ventricular cells with respect to
the Vmax and
action potential duration. The RMP of the atrial cell is
80 mV, and the
Vmax is 121 V/s,
compared with an RMP of 85 mV and a
Vmax of 279 V/s
for the ventricular cell.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1.
Comparison of current threshold (A)
and action potential waveform (B)
for an isolated ventricular (V) myocyte (dotted lines) and an isolated
atrial (A) myocyte. A: threshold
current pulse for a pulse duration of 2 ms.
B: recorded action potentials from the
two cells with maximum rate of rise of upstroke (V/s) indicated for
each cell.
|
|
To further characterize differences between atrial and ventricular
cells in the initiation of an action potential, we used current pulses
of either 2- or 15-ms duration and a magnitude slightly greater than
the stimulus current threshold of each cell for the given stimulus
duration. Figure 2 shows the recorded
action potentials generated by such stimulus protocols and the critical depolarization from the resting potential level required to initiate an
action potential. The critical depolarization is much
smaller in the atrial cell (19.3 mV) than in the ventricular cell (36.4 mV) and was not affected by increasing the stimulus duration from 2 to
15 ms for either the atrial or ventricular cell. An interesting phenomenon was noticed when measuring the delay from stimulation, defined as the time delay from turning off the current stimulus to the
Vmax of the
action potential upstroke. As we applied current pulses that were very
close to the current threshold, atrial cells were able to generate
action potentials with delays of more than two to three times those of
ventricular cells without affecting Vmax. To begin to
probe for answers as to why there are such differences between these
cells of a common organ, we needed to investigate the membrane
characteristics of each cell type.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 2.
Comparison of voltage threshold and critical depolarization required
for activation for an isolated A cell
(A) and an isolated V cell
(B) in response to current stimuli
of either 2- or 15-ms duration. Solid lines indicate successful
activations, and dotted lines indicate largest subthreshold
responses.
|
|
In excitable cells, a classic experiment for characterizing the
excitability properties of a cell is to create a strength-duration curve by finding the magnitude of the required current for action potential initiation as a function of the duration of the stimulating pulse. Figure
3A
demonstrates the mean strength-duration curves obtained from 10 atrial
and 6 ventricular cells. From these curves it is apparent that,
overall, much less current is needed to generate an action potential in
atrial cells than in ventricular cells at all stimulus durations. This
observation is to be expected because atrial cells have been reported
(see introduction) to have a much higher input impedance than
ventricular cells. In Fig. 3B, the
mean strength-duration curves for the atrial and ventricular cells were
normalized by scaling the values by a factor of the average threshold
stimulus current at the 2-ms duration for each cell type to compare the
shapes of the curves. The dotted line in Fig.
3B represents a theoretical hyperbolic
curve that would have resulted from a condition in which a constant
amount of charge (stimulus current magnitude × stimulus duration)
was required for each value of stimulus duration. The atrial
strength-duration curve closely approximates the theoretical hyperbolic
curve, whereas the ventricular strength-duration curve deviates
significantly toward greater values of required current magnitude than
that predicted by the "constant charge" relationship. This
deviation suggests that, although almost all of the current being
injected into the atrial cell is used to charge membrane capacitance,
most of the current being injected into the ventricular cell, for
stimuli of longer duration, is lost from the cell as a membrane ionic current flow.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 3.
Comparison of strength-duration relationships for V and A cells.
A: averaged data for 10 A cells and 6 V cells. B: data from
A normalized such that, for each cell
type, data have been scaled by the mean current threshold for 2-ms
duration stimulus. Dashed line represents common relationship for both
V and A cells if relationships for each cell type were produced by a
constant charge injection for all stimulus durations (see text).
Inset: ratio of V cell current
threshold (Ith)
to A cell threshold as a function of stimulus duration.
|
|
Another way of expressing the differences in the dependence of the
current threshold on the stimulus duration for ventricular versus
atrial cells is to compute the ratio of the current threshold for the
two cell types as a function of stimulus duration (see inset, Fig.
3B). For short-duration stimuli the
ventricular cells require a current magnitude 3-4 times as great
as do the atrial cells, whereas for stimuli of longer durations the
ventricular cells require ~10 times as much current amplitude as do
the atrial cells. The shape of the current threshold ratio plot that
describes the difference between the two strength-duration curves could be due to differences in the magnitude or voltage dependence of an
inward current, such as inward sodium current
(INa; which is more difficult to turn on in ventricular cells than in atrial cells),
or to an increased magnitude of an outward current, such as
IK1 (which may be present
in ventricular cells and not in atrial cells, or at least more
prevalent in ventricular cells). More current would thus be needed for
a ventricular cell, compared with an atrial cell, to overcome these
obstacles before the membrane could be depolarized enough to generate
an action potential. If the outward current were specific for potassium
ions, this might also explain the more negative RMP in the ventricular
cells but would not account for the higher value of
Vmax for the
ventricular cells compared with the atrial cells.
Membrane characteristics of atrial and ventricular cells.
Further investigation of the charging of the membranes of atrial and
ventricular myocytes can be performed if the ionic conductances of the
membrane do not change significantly as the cell membrane is
depolarized or hyperpolarized over a narrow range of potential from the
value of RMP. We can express the voltage waveform with time in response
to a small step of positive or negative current through the pipette as
an exponential charging function for which the time constant (in ms) is
equal to the product of the membrane resistance (in M) and the
membrane capacitance (in pF). To compute the membrane time
constant of atrial and ventricular cells, a small stimulus current of
50-ms duration was injected into an atrial or a ventricular cell, the
final depolarization was measured and the time for the voltage to rise
(or fall) to 63% of final depolarization (or hyperpolarization) was
measured. Figure 4 shows atrial and
ventricular cell membrane potential changes after injection of such
currents. The membrane time constant for a small depolarization of the
atrial cell (21 ms; Fig. 4A) is much
longer than that for the ventricular cell (2 ms; Fig.
4B), which means that the atrial cell membrane charges more slowly in response to a stimulus current and
also remains charged for a much longer period than the ventricular cell
membrane after the stimulus current is turned off. The long membrane
time constant of atrial cells may explain the long delays from
stimulation that can be produced in atrial cells (Fig.
2A). Membrane capacitance was
obtained by dividing the membrane time constant by the measured input
impedance (the ratio of the final polarization to the current
amplitude). The values of input impedance were 341 ± 41 M for
atrial cells and 16.5 ± 5.4 M for ventricular cells; these
values were then used to calculate the respective values for cell
membrane capacitance of 70 ± 4 pF and 99 ± 4 pF (Table
1). The 41% higher membrane capacitance in
ventricular cells provides one explanation as to why the stimulus
current threshold is higher in ventricular cells than in atrial cells, but this doesn't explain the large difference in current threshold (255%) for short-duration stimuli or the even larger difference in
current threshold for longer stimuli.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 4.
Comparison of response to a 50-ms duration small depolarizing or
hyperpolarizing current step for an A cell
(A; stimulus magnitude 20 pA) and for
a V cell (B; stimulus magnitude 100 pA) with membrane time constants ( ) indicated on traces. RMP,
resting membrane potential.
|
|
Initiation of action potentials in atrial and ventricular cells.
To improve our understanding of the membrane potential changes that
occur within the voltage range between the RMP and the voltage
threshold for action potential generation, we used a current-step protocol in which cells were injected with current stimuli of increasing magnitude (each stimulus was of 50-ms duration) to a level
at which the threshold for action potential initiation was reached.
Figure 5 shows the resulting voltage traces
for a ventricular cell (Fig. 5A) and
an atrial cell (Fig. 5B) recorded from this protocol. For the atrial cell we used current steps incremented by 10 pA, whereas for the ventricular cell we used current
steps incremented by 100 pA. Note that subthreshold depolarizations produced by current pulses 
40 pA for the atrial cell and 500 pA for
the ventricular cell show clear differences in waveform. The atrial
depolarizations show a very gradual increase in depolarization, as was
expected because of the long time constant of the atrial cells. The
ventricular depolarizations show a rapid response for the lowest
current strengths, but for larger current strengths there is a
secondary component of the depolarization that is much slower than can
be accounted for by the membrane time constant. A slight, further
increase in stimulus current produces an action potential in the atrial
cell (current threshold 43 pA) and in the ventricular cell (current
threshold 540 pA).
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 5.
Comparison of activation of a V cell and an A cell.
A: successive application of
increasing positive or negative current steps of 50-ms duration with a
step increment of 100 pA for a V cell, with threshold response to a
current magnitude of 540 pA also shown.
B: successive application of
increasing positive or negative current steps of 50-ms duration with a
step increment of 10 pA for an A cell, with threshold response to a
current magnitude of 43 pA also shown.
C: relationship between magnitude of
current step and membrane potential at end of current step for V cell
( ) and A cell ( ). Also shown are data for A cell scaled by a
factor of 7.4 ( ), which were computed from relative slopes of V and
A cell data to make the slopes of the two relationships equal as they
cross the horizontal axis.
|
|
We have plotted in Fig. 5C a
current-voltage (I-V) relationship
that was obtained by plotting the amplitude of the injected current
versus the value of polarization at the end of the current pulse. This
is not a true "steady-state" relationship because the membrane
potential is still changing with time at the end of the stimulus pulse,
but it does give an indication of the degree of rectification of the
membrane and the relative input impedance of the two cells. The values
for the ventricular cell are plotted as filled squares, whereas the
values for the atrial cell are plotted as open circles. Note that the
slopes of these two relationships are very different, with the data for
the ventricular cell having a significantly higher slope. To better
compare the shapes of the two I-V
relationships, we have also plotted the data for the atrial cell after
scaling all of the current values for the atrial cell by a factor of
7.4, indicated as open triangles. The scaling factor was computed from
the relative slopes of the ventricular and the atrial data to make the
slope of the two relationships the same as they cross the horizontal
axis. The data for both cells have a curvature suggesting inward
rectification, but the actual conductances being changed cannot be
determined from this presentation. The rectification may correspond to
the IK1, which has been shown (see introduction) to be more prevalent in ventricular cells than in atrial cells, but a slowly activating inward current with
depolarization could also be partly responsible.
Table 1 displays the data measured and calculated from our whole cell
current-clamp studies. As described earlier, the RMP, amplitude,
current threshold, voltage threshold, and
Vmax measured from atrial cells (80 ± 1 mV, 109 ± 3 mV, 0.69 ± 0.05 nA, 59 ± 1 mV, and 206 ± 17 V/s,
respectively) differed significantly (P < 0.05) from those values
measured from ventricular cells (82.7 ± 0.4 mV, 127 ± 1.12 mV, 2.45 ± 0.13 nA, 46 ± 2 mV, and 395 ± 21 V/s, respectively). Input impedance, capacitance, time constant, and
critical depolarization, which were calculated to quantitate the
membrane characteristics of the two cell types, also were significantly
different between atrial (341 ± 41 M, 70 ± 4 pF, 23.8 ± 2.3 ms, and 19 ± 1 mV, respectively) and ventricular (16.5 ± 5.4 M, 99 ± 4.3 pF, 1.56 ± 0.32 ms, and 36 ± 1 mV,
respectively) cells.
One very interesting contrast between atrial and ventricular cells is
that the apparent voltage threshold for action potential initiation is
significantly more negative in atrial cells than in ventricular cells
(which might suggest a greater density of sodium channels or a
hyperpolarized shift in the voltage dependence for sodium channels for
the atrial cells), whereas the actual Vmax for atrial
cells is significantly lower than that for ventricular cells (which
might suggest either a lower density of sodium channels or a
significant contribution of outward current during the upstroke of the
action potential for the atrial cells). The data in Fig. 5 show that
the atrial cells have much less outward current over the voltage range
from the RMP to the threshold potential.
Whole cell voltage-clamp studies to examine "threshold" in
ventricular myocytes.
The interplay between the currents over this voltage range and the
voltage threshold for activation is difficult to interpret. One
approach is to voltage clamp the cells with step potentials over this
voltage range and use the resulting ionic membrane currents to estimate
the voltage dependence of activation. By definition, a cell cannot
produce an "action potential" while under voltage control
conditions, but the voltage clamp does provide a way of determining at
what value of step depolarization the net ionic membrane current (the
sum of the inward and outward currents) becomes negative (inward)
during the pulse, which would establish the minimum depolarization for
which a cell might generate an action potential under current-clamp
conditions. Our hypothesis was that the presence of a large outward
current in the ventricular cells, such as
IK1, is
responsible for the differences in the voltage threshold for action
potential generation in the two cell types. We thus proposed to define
the activation threshold for ventricular cells using the whole cell
voltage-clamp technique, block the outward current
IK1 with 100 µM
BaCl2, and compare the new
activation threshold with the control value to test whether the shift
in activation threshold was comparable to the difference in activation
threshold for atrial cells compared with that for ventricular cells. If
the activation threshold for ventricular cells after
IK1 was blocked
was unchanged from control, this would suggest that a shift in the
voltage dependence for
INa might be responsible for the differences in activation threshold for atrial cells compared with ventricular cells.
To correlate these results with those measured using the whole cell
current-clamp technique, the whole cell voltage-clamp studies were
performed at 35°C, using the same external solution and pipette
solution as for the current-clamp experiments. A standard step protocol
with 2-mV increments was used from a holding potential equal to the
ventricular cell RMP (84 mV) to a voltage level at which a net
inward current was produced. Figure 6,
A and
B, shows the recordings from a
ventricular cell. The current tracings were separated to show the
results for pulse potentials to 2-mV increments from 86 to
62 mV (Fig. 6A,
a-m) and from 60 to
44 mV (Fig. 6B,
n-v). This separation better
shows the transition from purely outward net current to an increasing
early phase of inward current that finally becomes a net inward current
at a voltage-step level of 44 mV, which we define as the
activation threshold. Note that this value of activation threshold
corresponds quite well to the value of 46 ± 2 mV obtained
for ventricular cells under current-clamp conditions. We then applied
100 µM BaCl2 [which has
been shown (3) to specifically block
IK1] to the external Tyrode solution and repeated the voltage-clamp protocol for
the same cell, as shown in Fig. 7,
A and
B. Note that the currents are
considerably smaller after 100 µM
BaCl2 were added. Data for pulse
potentials to 2-mV increments from 86 to 70 mV are shown
in Fig. 7A
(a-i), and data from 68
to 54 mV are shown in Fig. 7B
(j-q). In the 100 µM
BaCl2 solution there is no longer the initial surge of net outward current that then decays over ~10
ms, as seen in Fig. 6 (control solution). In Fig.
7A there is a very rapid small surge
in net outward current, which then declines rapidly (see
tracing i) and increases slowly
again. As shown in Fig. 7B, with
stronger depolarizing step potentials, this declining phase in the net
current becomes more and more pronounced until it finally becomes a net
inward membrane current (shown by tracing
q) that defines an activation threshold of 54 mV for this cell in the 100 µM
BaCl2 solution. Note that this new
activation threshold is 10 mV more negative than that obtained in the
control solution for the same cell. This more negative value of
activation threshold, produced by blocking
IK1, corresponds quite well to the value of 59 ± 1 mV obtained from the
atrial cells in the current-clamp conditions.
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 6.
Voltage-clamp responses of a V cell in control external solution, with
a holding potential of 84 mV and successive steps with a 2-mV
increment from 86 to 62 mV
(A, tracings
a-m) and from 60 to 44 mV
(B, tracings
n-v).
|
|
View larger version (36K):
[in this window]
[in a new window]
|
Fig. 7.
Voltage-clamp responses of same V cell in Fig. 6 in an external
solution containing 100 µM
BaCl2, with a holding potential of
84 mV and successive steps with a 2-mV increment from 86
to 70 mV (A,
tracings a-i) and from
68 to 54 mV (B,
tracings j-q).
|
|
To show the effects of blocking
IK1, an
I-V relationship for the voltage-clamp
data was plotted in Fig. 8 by using the
values of net membrane current at the end of the voltage-clamp pulse of
50-ms duration plotted against the value of the membrane potential during the test pulse, using a range from 94 to 60 mV to
exclude the higher depolarizations for which some sodium current was
activated. The control curve (filled squares) is plotted as a nearly
linear slope over the voltage range from 90 to 80 mV and
then shows a significant rectification with depolarizations from
70 to 60 mV producing no increased net outward current,
similar in shape to the relationship we obtained for the ventricular
cell under current-clamp conditions (Fig.
5C). Over this voltage range there is no actual negative slope of the
IK1
I-V relationship, consistent with
previous results from Whalley et al. (19) for rabbit ventricular cells.
The results obtained after the addition of 100 µM
BaCl2 are shown as open circles
and demonstrate significantly less current for each of the test
potentials, similar to those obtained for the atrial cell of Fig. 5. To
better compare the shape of the two curves, the values plotted for the
ventricular cell in 100 µM BaCl2
were scaled by a factor of 12, which matched the slope of the control
data near the zero-crossing point (open triangles). Note that this
residual current not blocked by 100 µM
BaCl2 shows no rectification in
the voltage range from 84 to 60 mV, although there does
appear to be some rectification in the more negative voltage range from
94 to 84 mV.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 8.
Comparison of membrane current at end of voltage-step pulses of Figs. 6
and 7 under conditions of either normal external solution (control) or
100 µM BaCl2, with a holding
potential of 84 mV. Also shown are data for
BaCl2 solution scaled by a factor
of 12, which were computed from relative slopes of data in control and
100 µM BaCl2 solutions to make
the slopes of the two relationships equal as they cross the horizontal
axis.
|
|
Results obtained from five ventricular cells in which we used this
voltage-clamp protocol for the control and 100 µM
BaCl2 solutions demonstrated a
shift in the voltage threshold and the critical depolarization.
Addition of 100 µM BaCl2 caused
a change in threshold from 49.2 ± 9.3 to 60.8 ± 11.5 mV and a change in critical depolarization from 36.0 ± 6.8 to
24.4 ± 4.8 mV. By specifically blocking
IK1 with 100 µM
BaCl2, the voltage threshold and
the critical depolarization for ventricular myocytes, as determined using the voltage-clamp technique, were significantly
(P < 0.05) changed by 11.6 mV, a
value similar to the difference of 13 mV in voltage threshold and the
difference of 17 mV in critical depolarization observed between
ventricular and atrial myocytes (see Table 1).
Figure 9 shows results we obtained from a
rabbit atrial cell by using the same voltage-clamp protocol and
solutions as for the ventricular cell in Fig. 6. Comparison of the
atrial cell data with the ventricular cell data reveals that the atrial
cell data have a much smaller magnitude of currents, even smaller than the magnitude of the ventricular cell currents in the
BaCl2 solution. As shown in Fig.
9B, the current becomes net inward at
a pulse potential of 56 mV, which agrees quite well with the
value of 59 ± 1 mV for the voltage threshold obtained in the
current-clamp conditions for atrial cells. Note that the voltage
threshold determined using the voltage-clamp technique is very similar
when the atrial cell in control solution (Fig.
9) is compared with the ventricular cell in
the BaCl2 solution (Fig. 7).
Similar results were obtained from two additional atrial cells studied
with this technique.
View larger version (36K):
[in this window]
[in a new window]
|
Fig. 9.
Voltage-clamp responses of an A cell in control external solution, with
a holding potential of 84 mV and successive steps with a 2-mV
increment from 86 to 70 mV
(A, tracings
a-i) and from 68 to 54 mV
(B, tracings
j-p).
|
|
The voltage-clamp results with ventricular cells suggested that
current-clamp experiments with ventricular cells in control solutions
versus those in BaCl2 solution
would also show differences in cell activation that might mimic some of
the differences between ventricular and atrial cells. Figure
10 shows the responses of a ventricular
cell in control solution (Fig. 10A)
and in 50 µM BaCl2 solution
(Fig. 10B) to a protocol using steps
of depolarizing current pulses of 50-ms duration. This ventricular cell
was quite large, with a current threshold of 3.7 nA for a 2-ms duration stimulus in the control solution and a threshold of only 1.5 nA in the
50 µM BaCl2 solution (a ratio of
2.5). For the 50-ms duration pulses, we used current step increments of
100 pA for the control solution, with subthreshold responses shown for
current steps from 100 to 900 pA and including the threshold response
for 920 pA. In the 50 µM BaCl2
solution, we used current step increments of 10 pA, with subthreshold
responses shown for depolarizing steps from 10 to 120 pA and including
the threshold response for 130 pA (a ratio of current thresholds of 7.1 for the 50-ms duration compared with the ratio of 2.5 for the current
stimuli of 2-ms duration). Note also that the time course of the
subthreshold responses is clearly changed, with a much slower rise of
potential in the 50 µM BaCl2
solution, similar to that shown for atrial cells. The changes in the
voltage threshold under current-clamp conditions produced by 50 µM
BaCl2 are shown more clearly in
Fig. 11, for which we carefully
determined the voltage threshold in response to current pulses of 2- or
15-ms duration for a ventricular cell in control solution (Fig.
11B) and in 50 µM
BaCl2 solution (Fig.
11A). In the control solution the
voltage threshold is 51.4 mV for the 2-ms duration pulse, with a
critical depolarization from RMP of 33.9 mV. For the 15-ms duration
pulse these values are not significantly changed in the control
solution. In the 50 µM BaCl2
solution (Fig. 11A) for the same
cell, the voltage threshold has become more negative (59.3 or
58.4 mV for the 2- or 15-ms duration pulse, respectively), and
the critical depolarization has been reduced to 26.1 mV. The changes in
the voltage threshold and the critical depolarization produced in the
ventricular cell by the 50 µM
BaCl2 solution are very similar to
those we observed in comparing ventricular cell to atrial cells in
control solutions.
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 10.
Comparison of activation of a V cell in control external solution
(A) with activation of same V cell
in a solution containing 50 µM
BaCl2
(B).
A: successive application of
increasing positive current steps of 50-ms duration with a step
increment of 100 pA, with threshold response to a current magnitude of
920 pA also shown. B: successive
application of increasing positive current steps of 50-ms duration with
a step increment of 10 pA for same V cell as in
A, but after addition of 50 µM
BaCl2, with threshold response for
a current magnitude of 130 pA also shown.
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 11.
Comparison of voltage threshold and critical depolarization required
for activation for an isolated V cell in a solution with 50 µM
BaCl2
(A) and same isolated V cell in
control external solution (B) in
response to current stimuli of either 2- or 15-ms duration. Solid lines
indicate successful activations, and dashed lines indicate largest
subthreshold responses.
|
|
| |
DISCUSSION |
Conduction through the heart can be thought of as a process of current
supply and demand. Conduction fails when either not enough current is
supplied to a region or, equivalently, the amount of current demanded
by a region is greater than that which can be supplied by neighboring
regions that have already undergone activation. An understanding of the
mechanisms of initiation of an action potential and what factors
influence the critical amount of depolarization required for action
potential initiation is necessary to address the latter part of this
supply-demand theory of conduction. Previous studies (4, 7, 19) have
examined the ionic conductances that appear to differ between atrial
and ventricular cells and thus account for some of the differences in
action potential waveforms. In particular, Whalley et al. (19) showed
that the steady-state I-V relationship
for rabbit ventricular cells with 30 µM tetrodotoxin (TTX) and 100 µM CdCl2 showed a prominent
IK1 for
ventricular cells (demonstrated by block with BaCl2) and a much smaller
current for atrial cells. The focus of our work is on the biophysical
properties of the cell within the potential range between the RMP and
the threshold potential and how these differences in biophysical
properties between atrial and ventricular cells determine the
conditions for action potential initiation for that cell type. Because
our studies were done at the isolated cell level, we do not try to
account for other variables present in the whole organ such as the
number or distribution of gap junctions that would alter the cable
properties of atrial or ventricular tissue (1, 15) and, in doing so,
might account for some of the differences in action potential
initiation or propagation.
In previous work, Hume and Uehara (7) used isolated atrial and
ventricular myocytes from guinea pigs, and the RMP,
Vmax, input
impedance, and membrane time constant were 73.4 ± 5.1 mV, 83.9 ± 21.4 V/s, 108.8 ± 58.6 M, and 5.5 ± 2.6 ms,
respectively, for atrial cells and 74.1 ± 3.3 mV, 80.8 ± 17.7 V/s, 32.1 ± 13.4 M, and 2.3 ± 0.9 ms, respectively, for
ventricular cells. Hume and Uehara (7) stated that increasing the
experimental temperature to 35°C only affected action potential
shape by decreasing action potential duration and increasing
Vmax. Giles and
Imaizumi (4) recorded action potentials from isolated rabbit atrial and
ventricular cells and obtained values for RMP, input impedance,
capacitance, and membrane time constant of 66.9 ± 4.8 mV,
617 ± 401 M, 54.3 ± 5.9 pF, and 34 ms, respectively, for
atrial cells and 74.2 ± 2.6 mV, 33.7 ± 22.7 M, 72.5 ± 18.8 pF, and 2.4 ms, respectively, for ventricular cells. Giles
and Imaizumi (4) noted that increasing the stimulus frequency from 0.5 to 1 Hz in rabbit atrial and ventricular cells increased the action
potential duration and plateau height, whereas decreasing the frequency
from 1 to 0.1 Hz resulted in a very rapid early repolarization phase in
atrial cells that was thought to be due to
It. Whalley et
al. (19), using the whole cell current-clamp technique at room
temperature to compare action potentials generated by cultured rabbit
atrial cells that had assumed a spherical shape versus freshly isolated
rabbit ventricular cells, measured RMP,
Vmax,
capacitance, and input impedance as 66.4 ± 1.3 mV, 112.2 ± 4.8 V/s, 15-25 pF, and 958 ± 158 M, respectively, for
atrial myocytes and 70.0 ± 0.9 mV, 161 ± 18 V/s,
100-140 pF, and 29.7 ± 3.8 M, respectively, for ventricular
myocytes. All of these studies determined RMP values for atrial cells
that were somewhat less negative than those for ventricular cells under comparable conditions, but the values obtained were significantly less
negative than those obtained from microelectrode studies from intact
atrial tissue at physiological temperature [e.g., 80 to
84 mV by Spach et al. (16), and 82 mV by Nawrath
(13)].
In the present study, we have compared the initiation properties of
isolated rabbit atrial and ventricular myocytes, with specific
reference to differences in RMP,
Vmax, action
potential amplitude, action potential duration, and the voltage and
current thresholds. From these data, we also computed the differences in the membrane time constant, input impedance, critical
depolarization, and membrane capacitance for atrial and ventricular
cells. Although the membrane capacitance was 41% higher in ventricular
myocytes than in atrial myocytes, the input impedance for small
depolarizations was 20-fold greater in atrial myocytes than in
ventricular myocytes, producing membrane time constants that were
15-fold greater for atrial myocytes than for ventricular myocytes.
Also, the ventricular myocyte RMP was only 2.7 mV more negative,
Vmax was 92%
greater, and the action potential amplitude was 17% greater than the
same values for the atrial myocytes. A comparison of strength-duration curves for the two cell types also revealed that, although the ventricular myocytes required 3.6 times as much current to initiate an
action potential for a short (2 ms) duration, for longer durations this
ratio was much increased, with ventricular myocytes requiring 10 times
as much current as atrial cells for stimulus durations in the range of
25-50 ms. The critical depolarization from the RMP required to
initiate an action potential was 90% greater for ventricular myocytes
than for atrial myocytes, with the voltage threshold as determined from
the current-clamp experiments being 13 mV less negative for ventricular
myocytes than for atrial myocytes.
To test the hypothesis that many of these differences could be
accounted for by the greater magnitude of
IK1 in
ventricular myocytes compared with that in atrial myocytes, we compared
the properties of ventricular myocytes in control solution with the same ventricular myocytes in a solution containing 50-100 µM
BaCl2 to selectively block the
IK1 current.
Using both voltage- and current-clamp techniques, we showed that the
block of IK1 in
the ventricular cells produced 1) a
small depolarization of the RMP, 2)
a large increase in the input resistance and a corresponding large
increase in the membrane time constant,
3) a negative shift in the voltage
threshold for producing a net inward current under voltage-clamp
conditions, 4) a negative shift in
the voltage threshold under current-clamp conditions, and
5) a decrease in the current threshold for stimuli of short duration and an even larger decrease in
the current threshold for stimuli of longer duration. All of these
changes in the ventricular cells produced by 50-100 µM
BaCl2 were comparable to the
differences we observed between ventricular and atrial myocytes,
suggesting that many of the differences could be accounted for by the
relative lack of
IK1 in atrial
cells.
Our use of the voltage-clamp technique to determine the voltage level
for activation of a net inward current is different from the usual use
of this technique to pharmacologically isolate a particular current and
then study the voltage and time dependence of that current. We used a
normal external and internal solution and a physiological temperature
to assess the voltage level at which a net inward current occurs. Thus
our voltage-clamp data are not designed to isolate
IK1 other than to
show the effect of BaCl2 (as a
blocker of IK1)
on the voltage threshold for activating a net inward current during the
potential step for ventricular cells. In particular, we have restricted
the voltage range of Fig. 8, in which we plot the voltage-clamp current
at the end of a 50-ms pulse, to those voltage levels below which the
sodium current is activated. The I-V
relationship thus produced does not show the negative slope of the
IK1
rectification, which is present only at more depolarized voltage levels
in rabbit ventricular cells (19). As the pulse potential is increased
toward this threshold level (see Fig.
6B, tracings
s, t, and
u), there is a phasic component of
inward current, but the presence of a large outward current prevents
the expression of a net inward current, thus raising the voltage level
for activation.
The block of IK1
in ventricular cells did not produce a decrease in
Vmax or a
decrease in the action potential amplitude, suggesting that these
differences between ventricular and atrial myocytes may be produced by
a decreased sodium current in atrial cells. However, we have no direct
data on the magnitude of sodium current in the atrial or ventricular
cells. We previously studied the relationship between the sodium
current and the strength-duration curve for isolated rabbit ventricular
cells (8). This study showed that 3 µM TTX reduced the
Vmax of the
ventricular cell by ~50%, produced only a 13% increase in the
current threshold for a 2-ms duration stimulus, depolarized the voltage
threshold by ~5 mV, and shifted the strength-duration curve in the
positive direction with a nearly constant ratio over a duration range
of 1-10 ms. Thus it seems unlikely that the lower
Vmax of the
atrial myocytes compared with that of the ventricular myocytes plays a
significant role in the mechanism of the differences between the
strength-duration curves and the voltage threshold shifts we observed
in atrial and ventricular myocytes. In fact, the lower amount of
outward current in the atrial myocytes, compared with that in the
ventricular myocytes, during the upstroke of the action potential may
actually serve to partially compensate for a lowered value of sodium
conductance in the atrial myocytes, an effect that has been proposed in
previous studies relating changes in the sodium conductance and
Vmax (6, 19).
With these differences in the activation properties of isolated
myocytes from the atrium and the ventricle established, it is
reasonable to discuss how these differences might be related to the
process of normal and abnormal conduction in the two regions. The
properties we have shown for the atrial cells make them ideally designed to be activated by current flow from a slowly depolarizing region of automaticity comprising cells with high input resistance (e.g., the sinoatrial node). The ventricular cells, on the contrary, are ideally designed to suppress propagation from such a region (which
in the ventricle would be an ectopic focus) because of their lower
input resistance, shorter time constant, larger critical depolarization, and larger current threshold, especially for a prolonged stimulus. In fact, the
IK1 current has
recently been shown to be reduced in ventricular cells under conditions
of hypoxia, decreased intracellular ATP, or by the actions of
lysophosphatidylcholine (2, 11, 12, 17, 20). This partial block of
IK1 in
ventricular myocytes under conditions associated with myocardial
ischemia may make them more susceptible to activation from an
ectopic focus by the mechanism of a greatly reduced current threshold
for prolonged duration stimuli, as we showed with the
BaCl2 solution.
| |
ACKNOWLEDGEMENTS |
This work was partially supported by National Heart, Lung, and
Blood Institute Grant HL-22562 and the Emory Egleston Children's Research Center.
| |
FOOTNOTES |
Address for reprint requests: R. W. Joyner, Dept. of Pediatrics, Emory
Univ., 2040 Ridgewood Dr. NE, Atlanta, GA 30322.
Received 2 September 1997; accepted in final form 10 February
1998.
| |
REFERENCES |
1.
Beyer, E. C.,
L. M. Davis,
J. E. Saffitz,
and
R. D. Veenstra.
Cardiac intercellular communication: consequences of connexin distribution and diversity.
Braz. J. Med. Biol. Res.
28:
415-425,
1995[Medline].
2.
Clarkson, C. W.,
and
R. E. Ten Eick.
On the mechanism of lysophosphatidylcholine-induced depolarizations of cat ventricular myocardium.
Circ. Res.
52:
543-556,
1983[Abstract].
3.
DiFrancesco, D.,
A. Ferroni,
and
S. Visentin.
Barium-induced blockade of the inward rectifier in calf Purkinje fibres.
Pflügers Arch.
402:
446-453,
1984[Medline].
4.
Giles, W. R.,
and
Y. Imaizumi.
Comparison of potassium currents in rabbit atrial and ventricular cells.
J. Physiol. (Lond.)
405:
123-145,
1988[Abstract/Free Full Text].
5.
Hancox, J. C.,
A. J. Levi,
C. O. Lee,
and
P. Heap.
A method for isolating rabbit atrioventricular node myocytes which retain normal morphology and function.
Am. J. Physiol.
265 (Heart Circ. Physiol. 34):
H755-H766,
1993[Abstract/Free Full Text].
6.
Hondeghem, L. M.
Validity of Vmax as a measure of the sodium current in cardiac and nervous tissues.
Biophys. J.
78:
147-152,
1978.
7.
Hume, J. R.,
and
A. Uehara.
Ionic basis of the different action potential configurations of single guinea-pig atrial and ventricular myocytes.
J. Physiol. (Lond.)
368:
524-544,
1985.
8.
Joyner, R. W.,
B. M. Ramza,
T. Osaka,
and
R. C. Tan.
Cellular mechanisms of delayed recovery of excitability in ventricular tissue.
Am. J. Physiol.
260 (Heart Circ. Physiol. 29):
H225-H233,
1991[Abstract/Free Full Text].
9.
Joyner, R. W.,
B. M. Ramza,
R. C. Tan,
J. Matsuda,
and
T. T. Do.
Effects of tissue geometry on initiation of a cardiac action potential.
Am. J. Physiol.
256 (Heart Circ. Physiol. 25):
H391-H403,
1989[Abstract/Free Full Text].
10.
Joyner, R. W.,
H. Sugiura,
and
R. C. Tan.
Unidirectional block between isolated rabbit ventricular cells coupled by a variable resistance.
Biophys. J.
60:
1038-1045,
1991[Abstract/Free Full Text].
11.
Kakei, M.,
A. Noma,
and
T. Shibasaki.
Properties of adenosine-triphosphate-regulated potassium channels in guinea-pig ventricular cells.
J. Physiol. (Lond.)
363:
441-462,
1985[Abstract/Free Full Text].
12.
Kiyosue, T.,
and
M. Arita.
Effects of lysophosphatidylcholine on resting membrane conductance of isolated guinea pig ventricular cells.
Pflügers Arch.
406:
296-302,
1986[Medline].
13.
Nawrath, H.
Action potential, membrane currents and force of contraction in mammalian heart muscle fibers treated with quinidine.
J. Pharmacol. Exp. Ther.
216:
176-182,
1981[Abstract/Free Full Text].
14.
Ramza, B. M.,
R. C. Tan,
T. Osaka,
and
R. W. Joyner.
Cellular mechanism of the functional refractory period in ventricular muscle.
Circ. Res.
66:
147-162,
1990[Abstract/Free Full Text].
15.
Saffitz, J. E.,
L. M. Davis,
B. J. Darrow,
H. L. Kanter,
J. G. Laing,
and
E. C. Beyer.
The molecular basis of anisotropy: role of gap junctions.
J. Cardiovasc. Electrophysiol.
6:
498-510,
1995[Medline].
16.
Spach, M. S.,
P. C. Dolber,
and
J. F. Heidlage.
Interaction of inhomogeneities of repolarization with anisotropic propagation in dog atria.
Circ. Res.
65:
1612-1631,
1989[Abstract/Free Full Text].
17.
Trube, G.,
and
J. Hescheler.
Inward-rectifying channels in isolated patches of the heart cell membrane: ATP-dependence and comparison with cell-attached patches.
Pflügers Arch.
401:
178-184,
1984[Medline].
18.
Wagner, M. B.,
D. Golod,
R. Wilders,
E. E. Verheijck,
R. W. Joyner,
R. Kumar,
H. J. Jongsma,
A. C. van Ginneken,
and
W. N. Goolsby.
Modulation of propagation from an ectopic focus by electrical load and by extracellular potassium.
Am. J. Physiol.
272 (Heart Circ. Physiol. 41):
H1759-H1769,
1997[Abstract/Free Full Text].
19.
Whalley, D. W.,
D. J. Wendt,
C. F. Starmer,
Y. Rudy,
and
A. O. Grant.
Voltage-independent effects of extracellular K+ on the Na+ current and phase 0 of the action potential in isolated cardiac myocytes.
Circ. Res.
75:
491-502,
1994[Abstract/Free Full Text].
20.
Wilde, A. A. M.,
and
G. Aksnes.
Myocardial potassium loss and cell depolarisation in ischaemia and hypoxia.
Cardiovasc. Res.
29:
1-15,
1995[Medline].
Am J Physiol Heart Circ Physiol 274(6):H1902-H1913
0002-9513/98 $5.00
Copyright © 1998 the American Physiological Society
Top
Abstract
Introduction
