Cell DistensionInduced Increase of the Delayed Rectifier K+ Current in Guinea Pig Ventricular Myocytes
Abstract Single ventricular myocytes of guinea pig heart were distended by applying a positive pressure of 5 to 20 mm Hg in the pipette during the whole-cell voltage clamp. The amplitude of delayed rectifier K+ current (IK) was increased by ≈1.5 times, whereas the inward rectifier K+ current was scarcely affected. The increase of IK was reversible by applying a negative pressure of −10 to −30 mm Hg accompanied by shrinkage of the inflated cell. This response of IK was largely attributed to the E-4031–insensitive component of IK. The fully activated current amplitude, measured using long-lasting depolarizing pulses (>30 seconds) to +60 mV, was increased by the cell distension. The activation time course of IK during the long pulse consisted of more than three exponential components, and the slowest time constant was decreased by the distension from control 20.2±7.7 seconds (n=4) to 7.6±1.6 seconds (n=5). We failed to detect an involvement of microtubules or microfilaments, protein kinase C, and Ca2+ in the inflation-mediated increase of IK.
Depolarization of the ventricular cell membrane immediately shuts off the inward rectifier K+ channels, providing the low background membrane conductance for generation of the long-lasting plateau of the action potential. Continuous depolarization then gradually increases the K+ conductance (IK). It is generally considered that the activation of IK is mainly responsible for gradual repolarization during the plateau, and finally, the outward current through the inward rectifier K+ channel terminates the ventricular action potential. The amplitude of IK is modulated by various factors, such as stimulation of the β-adrenergic receptors,1 2 phorbol ester stimulation,3 4 and intracellular Ca2+.5
Recently, whole-cell voltage-clamp experiments revealed that the magnitude of IK is dramatically changed by altering the osmolarity of the bathing solution.6 7 Exposure to a 70% hypotonic solution almost doubles the magnitude of IK, whereas a 130% hypertonic solution halves it. This response of IK was consistently observed and appeared earlier than the delayed increase of the swelling-activated Cl− conductance.7 Superfusion of myocytes with hypotonic or hypertonic solution may increase or decrease cell volume, respectively, and the changes in the cell volume may alter the tension applied to the sarcolemma. However, our previous study6 failed to observe changes in IK during mechanical stretch applied along the longitudinal axis of the ventricular myocytes. The present study demonstrates that the increase of IK similar to that induced by the hypotonic superfusion is seen when a pipette solution is injected directly into the cell. It is suggested that the number of available IK channels is increased by the membrane stretch in a reversible manner. However, the mechanism of this effect remains unresolved and does not seem to involve PKC, Ca2+, or the cytoskeleton.
Materials and Methods
Single ventricular cells were isolated from guinea pig (300 to 400 g body weight) hearts using essentially the same enzymatic dissociation technique as described previously.8 9 In brief, each guinea pig was deeply anesthetized with pentobarbital sodium (30 mg/kg IP), and the heart was dissected out and perfused for 20 minutes with a Ca2+-free Tyrode’s solution containing 0.04% collagenase (type I, Sigma Chemical Co) using the Langendorff perfusion method. After the enzymatic treatment, single myocytes were dispersed from the left ventricle in a high-K+ low-Cl− solution and stored in the cell culture medium (Eagle’s medium, Flow Laboratories) at room temperature (24°C to 26°C) for later use.
The composition of the control Tyrode’s solution was (mmol/L) NaCl 140.0, KCl 5.4, CaCl2 1.8, MgCl2 0.5, NaH2PO4 0.3, glucose 5.5, and HEPES 5.0, and the pH was adjusted to 7.4 with NaOH. CaCl2 was omitted if necessary. In some experiments, a hypotonic (Figs 5B⇓ and 7⇓) or a hypertonic (Fig 5C⇓) solution was prepared as described previously.7 In brief, the hypotonic solution (70% osmolarity) was prepared by decreasing the concentration of NaCl to 100 mmol/L in the control Tyrode’s solution, and the hypertonic solution (130% osmolarity) was prepared by adding 90 mmol/L mannitol to the control Tyrode’s solution. An isotonic solution was also prepared by adding 80 mmol/L mannitol to the hypotonic solution. As noted previously,7 no significant difference was noticed in IK between the two isotonic solutions (Tyrode’s solution and hypotonic solution+mannitol). The L-type Ca2+ current was blocked by adding 2 mmol/L NiCl2 to all bath solutions. The bath solution was exchanged by switching the perfusates at the inlet of the chamber. In all experiments, the temperature of the bath solution was kept at 35±0.5°C.
Colchicine, cytochalasin B and D, and vinblastine sulfate (Sigma) were used to disrupt the cytoskeleton. Different doses of those agents were tested in the present study compared with previous studies.10 11 12 Staurosporine (Wako) was used to inhibit PKC. These chemicals were prepared freshly before the experiment. Cytochalasin B and D and staurosporine were dissolved in dimethyl sulfoxide and vinblastine sulfate was dissolved in acetone before addition to the external solutions. The final concentration of dimethyl sulfoxide or acetone in the salines was <0.1%. Experiments with cytochalasin B and D and vinblastine sulfate were performed in relative darkness to avoid photodestruction. E-4031 was kindly provided by Eisai Pharmaceutical Co and was dissolved in distilled water to make 10 mmol/L stock solution.
The patch pipettes were filled with an internal solution containing (mmol/L) potassium aspartate 140, MgCl2 5.0, K2-ATP 5.0, Na2-phosphocreatine 5.0, EGTA 10, and HEPES 5.0 (pH 7.2 with KOH).
Whole-Cell Current Recording
Single ventricular cells were voltage-clamped using the whole-cell configuration of the patch-clamp technique13 (patch-clamp amplifier, Axopatch-lD, Foster). The resistance of the pipette was 2.5 to 4.0 MΩ when filled with the internal solution. The membrane current and potential were recorded on a digital magnetic tape (RD-101, TEAC) for later computer analysis (PC98, NEC). The cell membrane capacitance was calculated by integrating current recorded by 10-mV hyperpolarizing steps from a holding potential of −40 mV. The liquid junction potential of −10 mV between the pipette solution and the external solution was corrected for all membrane potential recordings.
The changes of the ventricular cell volume during the osmotic cell swelling or shrinkage have been approximated by geometric measurements14 15 and were compared with the activation of a volume-sensitive Cl− channel.16 According to these previous studies, we measured the cell area under the microscope as an approximation of the cell volume. The image of the single ventricular cell under the whole-cell voltage clamp was recorded with a video recorder, and the single cell area was measured on the video screen. The cell was inflated or deflated by applying a positive or negative pressure into the pipette, which was monitored by a pressure transducer (Spectramed).
Data are given as mean±SEM. Statistical significance was determined using Student’s paired t test.
Increase of IK by Mechanical Inflation of the Ventricular Cell
The ventricular myocytes were slowly inflated by applying a positive pressure of 5 to 20 mm Hg into the glass pipette during the whole-cell voltage clamp. In the experiment shown in Fig 1A⇓ and 1B⇓, when a positive pressure of 5 mm Hg was applied into the pipette, the amplitude of IK slowly but continuously increased, accompanied by an enlargement of the cell area observed under the microscope. We failed to observe the activation of a volume-sensitive Cl− current, which was typically represented by an increase of instantaneous current jump at the pulse onset in the previous hypotonic cell swelling.7 The amplitude of IK usually remained constant or slightly decreased after the release of the positive pressure and was decreased by applying a negative pressure. The recovery of IK was accelerated by increasing the negative pressure as shown in Fig 1A⇓.
The current-voltage relations were measured when the current change reached a steady state in each experimental condition. The original current recordings in Fig 1B⇑ indicate that both IK during depolarization and the tail current on repolarization were increased by the mechanical cell distension. No obvious change was noticed in the double exponential time course of activation or deactivation of IK as examined by the semilogarithmic plot of the time-dependent current.7 Namely, the fast and slow time constants of activation were 106±5 and 708±24 milliseconds in the control condition and 96±8 and 739±19 milliseconds after the inflation; those for deactivation were 95±7 and 310±25 milliseconds in the control condition and 94±9 and 290±27 milliseconds after the inflation (n=7).
In Fig 1C⇑, the isochronal current-voltage relations measured at the onset of the pulse (left) showed no significant difference among the control (circles), inflated (squares), and recovery (triangles) values, suggesting that the inward rectifier K+ current was not affected by the inflation. The slope conductance between −70 and −90 mV as a measure of the inward rectifier K+ conductance was 104±12% (n=10) of the control value in the inflated cells. However, in the current-voltage relations measured near the end of the pulse (Fig 1C⇑, right), the outward current was markedly increased at potentials more positive than −10 mV by the cell distension. It seems that the voltage range of activation of IK was not changed. Essentially the same findings were obtained in seven experiments. The amplitudes of the outward current measured near the end of the pulse and the tail current were 2.26±0.29 and 0.73±0.09 nA (n=7) in the control cells, respectively; those in the inflated cells were increased to 152±9% and 148±8% (n=7) of the control values, respectively. The input capacitances of the cell, measured before and after the inflation, were not significantly different (P>.1). The capacitance was 212±23 pF in the control condition and 196±16 pF (n=7) after the cell distension.
A small outward shift of the holding current was observed in 6 inflated cells, and the amplitude of shift ranged between 0.10 and 0.25 nA (0.16±0.03 nA). The mechanisms of this shift were not examined in the present study. In 3 of 10 experiments, we failed to observe any change in IK or in the cell shape, and increasing pressure above 30 mm Hg simply destroyed the tight seal at the pipette tip. The positive pressure used for the inflation of the cell was between 5 and 20 mm Hg, and the negative pressure for the deflation was −10 to −30 mm Hg.
The E-4031–Insensitive Component Is Responsible for the IK Response
IK has been separated into two components on the basis of their sensitivity to some K+ channel blocking compounds such as E-4031. There is a rapidly activated drug-sensitive component (IKr) and a slowly activated drug-insensitive component (IKs) in guinea pig ventricular myocytes.17 In Fig 2⇓, the whole-cell current in response to a depolarizing pulse to +10 mV was recorded with and without E-4031. In the control condition, the amplitude of IK was slightly decreased by 5 μmol/L E-4031 (Fig 2A⇓), and the drug-sensitive component (difference current, IKr) showed a relatively rapid rise and decay on depolarization and repolarization, respectively. The relative amplitude of E-4031–sensitive current was measured from IK,tail on repolarization and was 31±10% of the control value (n=4). Under the effect of the drug, however, IK was still enhanced by the mechanical cell distension (201±17% of the control value, n=3).
In a separate series of experiments, E-4031 was applied only after IK was enhanced by the cell inflation. The drug decreased the enhanced tail amplitude by 10.5% in the experiment shown in Fig 2B⇑. In eight cells, IK,tail was increased to 177±10% of the control value, and the drug-sensitive component amounted to 23±3% (n=8).
Effects of E-4031 on IK were also examined in the hypotonic cell swelling (70% osmolarity; data not shown). As in the case of the mechanical cell distension, IK was enhanced by the hypotonic cell swelling even under the effect of E-4031. This finding is in agreement with the recent report by Rees et al,18 who reported that osmotic cell swelling significantly increased IKs. On the basis of the above experiments, we conclude that the increase of IK in inflated ventricular cells is largely due to the drug-insensitive component.
IK During a Continuous Depolarization
When IK was enhanced by the hypotonic solution,7 no obvious change was observed in the double-exponential time course of activation and deactivation or in the voltage dependence of activation if the currents induced by depolarizing pulses of several seconds were analyzed by assuming a steady current level by a best-fit routine.19 However, it is essential to determine the steady state current level to determine the activation kinetics in detail. So we made an attempt to record the fully activated amplitude of IK by prolonging the depolarizing pulse.
The current recordings induced by depolarizing pulses of different durations are shown in Fig 3A⇓. The outward current did not reach steady state even using a 32-second depolarization. The extremely slow late current change induced by the continuous depolarization might be due to unknown current systems, which might overlap on the relatively fast activation of IK. For example, [Ca2+]i may increase during depolarization accompanied with a gradual activation of the Ca2+-dependent Cl− current.20 21 22 This might cause an increase of the outward current but not the tail current at −40 mV, which is near the Cl− equilibrium potential. To test the possibility of overlapping currents, we examined the relationship between the amplitude of the outward current (IK) during the depolarization and that of IK,tail. The ratio of IK,tail to IK was plotted against the duration of the depolarizing pulse on a logarithmic time scale (Fig 3⇓). Although the ratio of IK,tail to IK was larger for short depolarizing pulses (0.60±0.04, n=6 for 125-millisecond pulse) as reported previously,17 it became quite constant for the pulses >1 second. The value of 0.36 was almost equal to the ratio of the driving forces for K+ at +50 mV and −40 mV. This finding may support the view that the extremely slow activation of the outward current is due to IK.
The membrane potential, which saturates the steady state activation of IK, was determined in Fig 4⇓. A pulse duration of 30 seconds was used, and the amplitude of IK,tail at −40 mV was measured. The relationship between the amplitude of the tail current and the test potential was fitted with a Boltzmann curve. The half saturation was at 26.8±3.5 mV, and the slope factor was 16.1±1.2 mV (n=4). These values are different from those determined previously using relatively short pulses.7 17
Fully Activated Amplitude of IK During the Cell Distension
IK was recorded by depolarization to +60 mV, where activation is almost complete (Fig 4⇑), before and during the mechanical inflation of ventricular cells. In the experiment shown in Fig 5A⇓, the gradual enhancement of IK by applying a 10 mm Hg pressure was monitored by repeating the usual 2-second depolarizing pulses to +60 mV, and the long-lasting depolarizations were applied as required. It was evident that the saturating amplitude of IK was increased by the cell distension. In four experiments using 5 to 20 mm Hg positive pressure, the saturating amplitude of IK during depolarization was increased to 170±30% of the control value. This finding suggests that the inflation of the ventricular myocyte increases the number of available IK channels.
The increase of fully activated IK induced by the cell distension was also confirmed during a continuous depolarization. Because of the variable time lag for the development of IK after applying the positive pressure into the pipette, we used the osmotic method. In Fig 5B⇑, hypotonic solution was applied under a continuous depolarization to +60 mV after saturating IK activation in the control solution. It is evident that IK was further increased by the hypotonic inflation. After terminating the continuous depolarization, further increase of IK in the hypotonic solution was confirmed by repeating the 2-second depolarization. Accordingly, the saturating level of IK in the hypotonic solution was further increased as revealed by the second continuous depolarization. During this depolarization, switching back to the isotonic solution almost completely reversed the increase of IK. Thus, in three cells, the fully activated amplitude of IK was 139±7% of the control value in the hypotonic solution.
The examination of the fully activated amplitude of IK was further extended to the hypertonic solution (Fig 5C⇑). The superfusion of hypertonic (130% osmolarity) solution decreased the fully activated current level in a reproducible manner. On the average, the saturating current level of IK decreased to 65±7% (n=5) of the control value in the hypertonic solution.
Kinetic Properties of IK Examined by the Long-Lasting Depolarization
Changes in the time courses of IK activation and deactivation by the cell inflation were examined using the long-lasting depolarizing pulse. Fig 6A⇓ shows the semilogarithmic plot of the time-dependent component of IK during the pulse to +60 mV before (left) and after (right) the mechanical cell inflation by applying 10 mm Hg. It is evident that the initial rapid increase of IK within 2 seconds is followed by a very slow exponential activation phase. By fitting a line to the late current change during depolarization, the exponential time constant of the slowest current component was determined to be 20.2±7.7 seconds (n=4) in control cells and 7.6±1.6 seconds (n=5) in inflated cells. This kinetic change may be supplemented by measuring the steady state activation. However, such measurements using long-lasting pulses before and after the cell inflation were not feasible and accurate because of spontaneous current fluctuations during the extremely long experimental time. In three experiments, a trend of negative shift of the half-activation voltage was suggested; namely, it was shifted from 23.9, 35.4, and 28.9 mV to 16.3, 11.9, and 13.3 mV after the cell inflation, respectively.
The IK,tail recorded on repolarization to a holding potential of −40 mV was fitted by a sum of two exponentials (Fig 6B⇑), as in the previous study, using a short (<3-second) conditioning depolarization.7 The deactivation time course was not significantly changed; the fast and slow time constants of the tail were 138±19 and 457±93 milliseconds, respectively, in the control state and 134±16 and 415±77 milliseconds, respectively, in the inflated state (n=3).
Taken together, we conclude that only the slowest (τ>5-second) phase of activation of IK during the extremely long pulse is markedly accelerated by the cell distension, whereas the fast double-exponential time course of both activation and deactivation remains unchanged.
Effects of Disrupting Cytoskeleton
It has been postulated that a viscoelastic cytoskeletal network is essential for the regulation of the stretch-activated channels.23 24 In the present study, the involvement of microfilaments or microtubules in the distension-induced modulation of IK was tested by disrupting the cytoskeletal elements by directly introducing cytoskeletal modifiers into the cytoplasm through the patch pipette. Myocytes were dialyzed with an internal solution including cytochalasin B (2 μmol/L). In order to ensure the equilibration of the pipette solution with intracellular medium and the steady effect of cytochalasin B, the effect of the cell distension was examined at least 7 minutes after the whole-cell current recording was started. In a series of four experiments, IK recorded with a 2-second depolarizing pulse was increased by inflating cells (5 to 10 mm Hg positive pressure), which was reversed by applying a negative pressure. The amplitudes of the outward current and IK,tail during the cell inflation were 160±9% and 150±11% (n=4) of the control value, respectively. No obvious change was noticed in the time course of IK enhancement during the application of positive pressure. The response of IK to cell distension was also observed when 50 nmol/L (n=3) and 2 μmol/L vinblastine sulfate (n=2) or 2 μmol/L colchicine (n=4) was added in the pipette solution to disrupt microtubules.
To further confirm that neither the microfilaments nor the microtubules are involved in the IK response to the cell distension, ventricular cells were incubated for 3 to 5 hours in cell culture medium containing cytoskeletal modifiers. Then the whole-cell clamp was carried out in the continuous presence of the agent under investigation. Agents used were cytochalasin B (10 μmol/L, n=3; 100 μmol/L, n=2), cytochalasin D (10 μmol/L, n=3; 50 μmol/L, n=2), vinblastine sulfate (10 μmol/L, n=3), and colchicine (0.1 mmol/L, n=3; 0.2 mmol/L, n=4). None of these agents significantly affected the reversible increase of IK by the cell distension. Similarly, the hypotonic solution increased IK in a reversible manner under these conditions.
Other Putative Mechanisms for the Distension-Induced Modulation of IK
An increase of [Ca2+]i or protein kinases was ruled out as a mechanism of the IK response to hypotonic cell inflation in a previous study.7 Since inflation of the ventricular cells was achieved by injecting an internal solution containing 10 mmol/L EGTA in the present study, an increase of [Ca2+]i is unlikely to mediate the observed response. Confirming this view, the IK response was also observed when an elevation of [Ca2+]i was prevented by the use of a nominally Ca2+-free external solution. The amplitude of IK at the end of a 1-second depolarizing pulse to +60 mV and the amplitude of IK,tail were increased to 175±13% and 183±28%, respectively, of the control amplitude (n=3). The increase of IK was also observed in the inflated cells when 2 μmol/L staurosporine, a specific inhibitor of PKC, was included in the external solution. The amplitudes of the outward current and tail current of IK were increased to 143±8% and 130±7%, respectively, of the control amplitude (n=7). Thus, we conclude that the IK response to the mechanical cell inflation is not mediated by an elevation of [Ca2+]i or PKC.
Increase of IK Parallel to the Osmotic Swelling of the Ventricular Cell
The inflation of the cell was always observed when IK was increased by applying positive pressure into the pipette. However, the increase of cell area was not always uniform over the entire cell, probably because of the point source of the fluid injection. Therefore, the relation between the cell swelling and the amplitude of IK was compared using the hypotonic cell swelling method, which caused homogeneous cell inflation. In Fig 7A⇓, the increase of IK as examined by 2-second depolarizing pulses (upper panel) is compared with the normalized cell area (lower panel) plotted on the same time scale. Both IK and the cell area were increased with a similar time course to reach a new plateau level after the hypotonic superfusion, and they were reversed by the superfusion of the control solution. The half-times for the enhancement and recovery of IK were 48.6±6.3 seconds (n=7) and 41.0±2.5 seconds (n=5); those for the cell area were 61.4±6.0 seconds (n=7) and 45.0±9.6 seconds (n=4), respectively.
The relative amplitude of IK was plotted against the relative cell area in Fig 7B⇑ (n=6), which indicates that the increase of IK was nearly proportional to the increase of the cell area. However, in one experiment the increase of the current was much delayed compared with that of the cell area.
Shortening of the Action Potential by Mechanical Inflation
In our previous study, we concluded that IK takes the role of osmoelectrical signal transduction.7 The role of IK in the mechanotransduction is clearly shown in Fig 8A⇓, where the action potential was recorded before (a) and after (b) the mechanical cell inflation. Inflating the cell shortened APD50 from 340 to 238 milliseconds and APD90 from 365 to 250 milliseconds. The time derivative of the membrane potential is shown in Fig 8B⇓, which suggested that the reduction of action potential is due to the increased density of the outward current during phase 2 and 3 repolarization. In five experiments, mean APD50 was decreased to 68±4% and APD90 was decreased to 69±4% of the control value. These observations are consistent with the enhancement of IK during cell inflation.
Mechanisms of the IK Response
The increase of IK induced by the mechanical cell inflation in the present study is quite comparable to that induced by the hypotonic cell swelling in the previous studies.6 7 18 In both cases, the increase of IK was reversible, and the gating kinetics remained almost constant after the cell swelling, except for the very slow activation that was revealed using a pulse longer than 30 seconds in the present study. The increase of IK during mechanical cell inflation was also observed when an elevation of [Ca2+]i was prevented by the use of Ca2+-free internal and external solutions or by inhibiting PKC with staurosporine, excluding Ca2+-mediated mechanisms from the signal transduction.6 Furthermore, the present study demonstrated an almost linear relationship between the cell surface area and the magnitude of IK during the course of the osmotic cell swelling (Fig 7⇑). Thus, these findings strongly suggest a direct coupling between the activity of the IK channels and the membrane stretch irrespective of osmotic or mechanical cell inflation. It should be noted that the present findings using the cell distension technique are free from complications of dilution or loss of the intracellular components, which remained to be examined in the previous studies of the osmotic cell swelling.26 27
The mechanical cell distension of the myocyte induced by injecting pipette solution consistently caused an increase of IK. However, IK was not significantly changed when the cardiac myocyte was stretched along its longitudinal axis.6 This difference might be explained by different deformations of the cell membrane. In the latter case, the diameter of the elongated cell might be decreased by the cell extension to maintain a constant cell volume, whereas the cell distension might cause a two-dimensional stretch on the cell membrane, which might be essential for the IK response, as suggested in the stretch-activated ion channels.23 28
The present study failed to observe any modulation of the IK response to the mechanical inflation after treating myocytes with vinblastine, colchicine, or cytochalasin, excluding the involvement of the microtubules or microfilaments. The involvement of the cytoskeletal elements for the channel activity has been examined in a variety of cells and tissues, and the contribution of microtubules was consistently excluded.11 24 On the other hand, a wide variety of effects have been reported when microfilaments were destructed: an inhibitory effect on the IKs channel expressed in oocytes,10 no effect on the swelling induced Cl− current,29 and a stimulatory effect on the stretch-activated channel.11 23 The inhibition of the channel response may suggest an essential role of the microfilaments in the transduction of the membrane mechanical stress to the channel activity, whereas the enhanced channel activity may suggest that stretching the membrane was more effective when the cytoplasmic viscoelasticity was eliminated by the disruption of the microfilaments. It should be noted, however, that the present findings do not completely exclude the role of the cytoskeletal elements other than the microfilaments and microtubules. Sokabe et al30 suspected that spectrin/dystrophin submembrane cytoskeleton might be the tension-bearing element in stretch-activated channels because the lipid membrane was stress free and the destruction of microfilaments does not abolish the increases of the channel activity.
Busch et al10 expressed K+ channels by injecting cRNA from rat kidney in Xenopus oocytes and recorded IKs, which showed voltage-dependent activation similar to IK in the cardiac muscle. Interestingly, the amplitude of IKs increased in response to the hypotonic superfusion of the oocyte, and this response was present even in the presence of the protein kinase inhibitors. However, different from the cardiac IK, the hypotonic increase of IKs was suppressed by removal of Ca2+ in the external solution and by treatment of oocytes with cytochalasin D. These authors suggested that increase in [Ca2+]i changes the cytoskeletal organization of oocytes, thereby increasing IKs.
Stretch-Induced State Transitions of the IK Channel
In guinea pig ventricular cells, the amplitude of the E-4031–sensitive component of IK was relatively small (Fig 2⇑), and the increase of IK by cell inflation was largely due to the drug-insensitive component. The amplitude of the drug-insensitive component of IK (I) is described as follows:
where N is the total number of available channels, i the unit amplitude of the single-channel current, and Po the open probability of the channel. Our noise analysis7 suggested that i was not modified during the IK increase in the hypotonic cell swelling. Thus, the increase in I is attributable to an increase in NPo. However, a simple increase of N due to an increase of membrane area (eg, the fusion of intracellular membrane vesicles to the surface membrane) is difficult to assume, since the membrane capacitance remained constant after the cell distension, as in the case of the hypotonic cell inflation. A model that we use to explain our experimental findings includes an unavailable (resting) mode and an available (active) mode (see also the stretch-activated channel model proposed by Sachs24 ). The transition between these two modes may be determined by both the membrane potential and the membrane stretch:
The active mode includes several closed states and an open state. This model assumes that more than half of the IK channels are in the active mode in the isotonic solution and that the rest of the channel population is in the unavailable resting mode. The gating kinetics in the active mode (Po) is assumed not to be affected by the cell distension, according to the experimental finding that the voltage-dependent gating kinetics remained the same as far as IK during short depolarizing pulses and IK,tail are concerned. The slowest activation process during the continuous depolarization (Fig 6A⇑) is attributed to the recruitment of the unavailable channels through a voltage-dependent transition from the resting mode to the active mode. The cell inflation may shift equilibrium to the right between the resting and active modes, probably by increasing the forward rate constant. The increase in the number of functional IK channels due to an aggregation of subunits into a channel was proposed in the IKs channels expressed in Xenopus oocytes.31 The hypothesized transition from the resting mode to the active mode in the present study should also include this aggregation mechanism if applicable to the IK response. Unfortunately, the quantitative comparison of the present model with experimental recordings was not practical because of the difficulty in analyzing the unusually slow activation kinetics of IK.
Action Potential Configuration
The duration of the action potential was shortened by the mechanical cell inflation in the present study (Fig 8⇑). This finding is in good agreement with the shortening of the action potential in the whole-heart preparation, which was overloaded by the intraventricular balloon method.32 The increase in ventricular volume might cause a two-dimensional stretch of the membrane. The amplitude of the net current calculated from the time derivative of the action potential was not obviously changed by the cell inflation during the plateau, whereas IK under the voltage clamp increased by 1.5-fold. The increase of the net outward current by cell inflation was evident only during the rapid repolarization (phase 3). To explain this apparent discrepancy, the activation of IK was reconstructed by using the IK model described by Kiyosue et al25 along the time course of the recorded action potential as shown in Fig 8C⇑. When the limited conductance of IK was increased by 1.5-fold in the model calculation, the amplitude of IK was increased only by 1.05 times at 150 milliseconds after the onset of the action potential during cell inflation. This is because the negative shift of the plateau potential level decreased Po of the IK channel; eg, the Po at 150 milliseconds was 0.27 for the control condition and 0.22 for the depressed action potential in the model calculation. However, the model calculation clearly indicated that the increase of IK is primarily responsible for the shortening of the action potential. The outward current during the rapid repolarizing phase (Fig 8B⇑) was increased when the action potential was shortened. At present, we cannot specify the underlying membrane current change for this increase of the net outward current.
Selected Abbreviations and Acronyms
|APD50, APD90||=||action potential duration at 50% and 90% repolarization|
|IK||=||delayed rectifier K+ current|
|IK,tail||=||K+ tail current|
|IKr||=||rapidly activating component of IK|
|IKs||=||slowly activating component of IK|
|PKC||=||protein kinase C|
This study was supported by a scientific research grant from the Ministry of Education, Science, and Culture of Japan. Dr Wang was also supported by the Kato Asao International Scholarship Foundation. The authors thank Dr K. Manabe for his discussion, Dr Sian A. Rees for reading the manuscript and discussion, M. Fukao for his technical assistance, and K. Fujida for her secretarial work.
- Received May 30, 1995.
- Accepted December 1, 1995.
- © 1996 American Heart Association, Inc.
Tohse N, Kameyama M, Irisawa H. Intracellular Ca2+ and protein kinase C modulate K+ current in guinea pig heart cells. Am J Physiol. 1987;253:H1321-H1324.
Walsh KB, Kass RS. Regulation of a heart potassium channel by protein kinase A and C. Science. 1988;242:67-69.
Tohse N. Calcium-sensitive delayed rectifier potassium current in guinea pig ventricular cells. Am J Physiol. 1990;258:H1200-H1207.
Sasaki N, Mitsuiye T, Wang Z, Noma A. Increase of the delayed rectifier K+ and Na+-K+ pump currents by hypotonic solutions in guinea pig cardiac myocytes. Circ Res. 1994;75:887-895.
Roos KP. Length, width, and volume changes in osmotically stressed myocytes. Am J Physiol. 1986;251:H1373-H1378.
Drewnowska K, Baumgarten CM. Regulation of cellular volume in rabbit ventricular myocytes: bumetanide, chlorothiazide, and ouabain. Am J Physiol. 1991;260:C122-C131.
Tseng GN. Cell swelling increases membrane conductance of canine cardiac cells: evidence for a volume-sensitive Cl channel. Am J Physiol. 1992;262:C1056-C1068.
Sanguinetti MC, Jurkiewicz NK. Two components of cardiac delayed rectifier K+ current: differential sensitivity to block by class III antiarrhythmic agents. J Gen Physiol. 1990;96:195-215.
Rees SA, Vandenberg JI, Wright AR, Yoshida A, Powell T. Cell swelling has differential effects on the rapid and slow components of delayed rectifier potassium current in guinea pig cardiac myocytes. J Gen Physiol. In press.
Zygmunt A, Gibbons WR. Calcium-activated chloride current in rabbit ventricular myocytes. Circ Res. 1991;68:424-437.
Zygmunt A, Gibbons WR. Properties of calcium-activated chloride current in heart. J Gen Physiol. 1992;99:391-414.
Rasmusson RL, Davis DG, Lieberman M. Amino acid loss during volume regulatory decrease in cultured chick heart cells. Am J Physiol. 1993;264:C136-C145.
Lerman BB, Burkoff D, Yue DT, Sagawa K. Mechanoelectrical feedback: independent role of preload and contractility in modulation of canine ventricular excitability. J Clin Invest. 1985;76:1843-1850.