Inactivation of calcium channel current in the calf cardiac Purkinje fiber. Evidence for voltage- and calcium-mediated mechanisms

We have studied the influence of divalent cations on Ca channel current in the calf cardiac Purkinje fiber to determine whether this current inactivates by voltage- or Ca-mediated mechanisms, or by a combination of the two. We measured the reversal (or zero current) potential of the current when Ba, Sr, or Ca were the permeant divalent cations and determined that depletion of charge carrier does not account for time-dependent relaxation of Ca channel current in these preparations. Inactivation of Ca channel current persists when Ba or Sr replaces Ca as the permeant divalent cation, but the voltage dependence of the rate of inactivation is markedly changed. This effect cannot be explained by changes in external surface charge. Instead, we interpret the results as evidence that inactivation is both voltage and Ca dependent. Inactivation of Sr or Ba currents reflects a voltage-dependent process. When Ca is the divalent charge carrier, an additional effect is observed: the rate of inactivation is increased as Ca enters during depolarizing pulses, perhaps because of an additional Ca-dependent mechanism.     

Delayed rectification in the cardiac Purkinje fiber is not activated by intracellular calcium

This study was designed to test the hypothesis that an outward current (Ix) responsible for action potential repolarization in the cardiac Purkinje fiber is activated by intracellular calcium (Cai). Pharmacological probes were combined with the measurement of membrane current and contractile activity under voltage clamp conditions. Experiments were designed to examine properties of Ix that have previously linked activation of this current to changes in Cai. The independence of Ix from Cai was demonstrated for each case tested. Thus, the results of these experiments support the view that Ix is not a calcium-activated current.     

Phase resetting in a model of cardiac Purkinje fiber.

The phase-resetting response of a model of spontaneously active cardiac Purkinje fiber is investigated. The effect on the interbeat interval of injecting a 20-ms duration depolarizing current pulse is studied as a function of the phase in the cycle at which the pulse is delivered. At low current amplitudes, a triphasic response is recorded as the pulse is advanced through the cycle. At intermediate current amplitudes, the response becomes quinquephasic, due to the presence of supernormal excitability. At high current amplitudes, a triphasic response is seen once more. At low stimulus amplitudes, type 1 phase resetting occurs; at medium amplitudes, a type could not be ascribed to the phase resetting because of the presence of effectively all-or-none depolarization; at high amplitudes, type 0 phase resetting occurs. The modeling results closely correspond with published experimental data; in particular type 1 and type 0 phase resetting are seen. Implications for the induction of ventricular arrhythmias are considered.     

Cadmium-induced blockade of the cardiac fast Na channels in calf Purkinje fibres

The fast transient inward current elicited by depolarizations above about -60 mV in calf Purkinje fibres was found to be depressed by Cd in concentrations less than 1 mM. The Cd-sensitive current, which strongly depended on external Na, was recorded in the presence of 2 mM MnCl2 and was blocked by TTX, indicating that a contamination from slow Ca-dependent currents could be discounted. The current reduction caused by Cd was also observed in nominally Ca-free solutions. The Cd-induced depression of the fast Na current was not accompanied by changes in the current kinetic parameters, as revealed by comparing inactivation curves and peak current voltage relations at different Cd concentrations, and could be attributed to a voltage-independent channel blocking action. Half-blockade occurred at 0.182 +/- 0.06 mM (n = 4). Plots of peak current amplitude as a function of the Cd concentration showed that the cooperation of two Cd ions was required to block a single channel.     

Open sodium channel properties of single canine cardiac Purkinje cells.

Open channel properties of canine cardiac Purkinje cell Na+ channels were studied with single channel cell-attached recording and with whole cell macroscopic current recording in internally perfused cells. Single channel currents and membrane currents increased with an increase in Na+ concentration, but showed evidence of saturation. Assuming first-order binding, the Km for Na+ was 370 mM. PCs/PNa was 0.020 and PK/PNa was 0.094. The current-voltage relationship for single channels showed prominent flattening in the hyperpolarizing direction. This flattening was accentuated by 10 mM Ca2+ and was greatly reduced in O mM Ca2+, indicating that the rectification was a consequence of Ca2+ block of the Na+ channels. A similar instantaneous current-voltage relationship was seen for the whole cell membrane currents. These results demonstrate that the cardiac channel shows substantial Ca2+ block, although it is relatively insensitive to tetrodotoxin. The Na+ and Ca2+ binding properties could be modeled by the four-barrier Eyring rate theory model, with similar values to those reported for the neuroblastoma Na+ channel (Yamamoto, D.,J.Z. Yeh, and T. Narahashi, 1984, Biophys J., 45:337-344).     

Inactivation properties of T-type calcium current in canine cardiac Purkinje cells.

The kinetic behavior of T-type Ca2+ current (ICa-T) was studied in canine cardiac Purkinje cells using a single suction-pipette whole-cell voltage clamp method. ICa-T was studied without contamination of conventional L-type Ca2+ current (ICa-L). Ca2+, Sr2+, or Ba2+ were used as the charge carrier. During maintained depolarization ICa-T decayed rapidly, and under most conditions the decay showed a voltage-dependent single exponential time course that did not depend on the species of charge carrier. The development of inactivation did not depend on Ca2+, but the time course required more than a single exponential process. Just negative to the threshold voltage for activating ICa-T, inactivation slowly developed and there was a delay in its onset. The time course of recovery from inactivation was dependent on the protocol used to measure it. As the duration of an inactivating voltage step was increased, recovery slowed markedly and there was a delay in its onset. The time course of recovery could be fit as a biexponential. The fast and slow time constants of recovery were relatively constant, however, the relative amplitudes were dependent on the duration of the inactivating voltage step. Recovery was not dependent on Ca2+, and it was slower at a less negative voltage. These results suggest that the T-type Ca2+ channel in cardiac Purkinje cells follows a complex kinetic scheme dependent only on voltage. This behavior can be accounted for by incorporating into a Markovian model several inactivated and closed states.     

The effects of calcium channel blockade on proliferation and differentiation of cardiac progenitor cells

Cardiogenesis depends on a tightly regulated balance between proliferation and differentiation of cardiac progenitor cells (CPCs) and their cardiomyocyte descendants. While exposure of early mouse embryos to Ca²⁺ channel antagonists has been associated with abnormal cardiac morphogenesis, less is known about the consequences of Ca²⁺ channel blockade on proliferation and differentiation of CPCs at the cellular level. Here we showed that at embryonic day (E) 11.5, the murine ventricles express several L-type and T-type Ca²⁺ channel isoforms, and that the dihydropyridine Ca²⁺ channel antagonist, nifedipine, blunts isoproterenol induced increases in intracellular Ca²⁺. Nifedipine mediated Ca²⁺ channel blockade was associated with a reduction in cell cycle activity of E11.5 CPCs and impaired assembly of the cardiomyocyte contractile apparatus. Furthermore, in cell transplantation experiments, systemic administration of nifedipine to adult mice receiving transplanted E11.5 ventricular cells (containing CPCs and cardiomyocytes) was associated with smaller graft sizes compared to vehicle treated control animals. These data suggest that intracellular Ca²⁺ is a critical regulator of the balance between CPC proliferation and differentiation and demonstrate that interactions between pharmacological drugs and transplanted cells could have a significant impact on the effectiveness of cell based therapies for myocardial repair.     

Adrenergic modulation of the delayed rectifier potassium channel in calf cardiac Purkinje fibers.

We have investigated the modulation of the delayed rectifier potassium channel in calf cardiac Purkinje fibers by the neurohormone norepinephrine. We find that 0.5 microM norepinephrine increases this K channel current by a factor of 2.7. A maximal increase of about four was found for concentrations of 1 microM and above. Norepinephrine produced a small (less than 5 mV) and variable shift of the K channel reversal potential toward more negative values. The kinetics of the potassium channel are well described by a two-exponential process, both in the absence and presence of norepinephrine. However, norepinephrine substantially decreases the slower time constant with no significant effect on the fast time constant. Potassium channel activation curves in the presence of norepinephrine are very similar to control curves except at large positive potentials. A simple sequential three-state model for this channel can reproduce these data both with and without norepinephrine. The logarithms of the rate constants derived from this model are quadratic functions of voltage, suggesting the involvement of electric field-induced dipoles in the gating of this channel. Most of the kinetic effects of norepinephrine appear to be on a single rate constant.     

Voltage-dependent block by tetrodotoxin of the sodium channel in rabbit cardiac Purkinje fibers.

The two-microelectrode, voltage-clamp technique was applied to rabbit cardiac Purkinje fibers to study the interaction of tetrodotoxin (TTX) with the slowly inactivating Na current. Binding of TTX to rested, inactivated, and activated channels was estimated by measuring the relative decrease of current at the beginning (rested and inactivated channels) and the end (activated channels) of a 1 s depolarizing clamp to -45 mV. The accelerated decline of the Na current in the presence of a submaximal dose of TTX was interpreted as an increase in blocking efficiency upon depolarization. The experiments show that activated as well as inactivated channels are more sensitive to TTX than are rested channels. The dissociation equilibrium constants for the three states are 3.5 X 10(-6) M for the rested, 0.94 X 10(-6) M for the activated, and 0.75 X 10(-6) M for the inactivated channels. The time course of activation block was dependent on TTX concentration. Rate constants for association and dissociation of the activated state are 1.3 X 10(6) M-1 X s-1 and 1.5 s-1, respectively.     

Delayed rectification in the calf cardiac Purkinje fiber. Evidence for multiple state kinetics.

We have investigated the delayed rectifier current (Ix) in the calf cardiac Purkinje fiber using a conventional two-microelectrode voltage clamp arrangement. The deactivation of Ix was monitored by studying decaying current tails after the application of depolarizing voltage prepulses. The reversal potential (Vrev) of these Ix tails was measured as a function of prepulse magnitude and duration to test for possible permeant ion accumulation- or depletion-induced changes in Vrev. We found that prepulse-induced changes in Vrev were less than 5 mV, provided that prepulse durations were less than or equal to 3.5 s and magnitudes were less than or equal to +35 mV. We kept voltage pulse structures within these limits for the remainder of the experiments in this study. We studied the sensitivity of Vrev to variation in extracellular K+. The reversal potential for Ix is well described by a Goldman-Hodgkin-Katz relation for a channel permeable to Na+ and K+ with PNa/PK = 0.02. The deactivation of Ix was always found to be biexponential and the two components shared a common reversal potential. These results suggest that it is not necessary to postulate the existence of two populations of channels to account for the time course of the Ix tails. Rather, our results can quantitatively be reproduced by a model in which the Ix channel can exist in three (two closed, one open) conformational states connected by voltage dependent rate constants.     

Diffusion around a cardiac calcium channel and the role of surface bound calcium.

The diffusion of Ca as it converges to the external mouth of a Ca channel is examined. Diffusional limitation on Ca ions entering Ca channels during current flow, cause local extracellular Ca depletions. Such extracellular Ca depletions have been reported in cardiac muscle. The cardiac sarcolemma has a large number of low-affinity Ca binding sites that can buffer these local Ca depletions. For a hemisphere of extracellular space (of radius less than 0.33 microns) centered on the external mouth of a Ca channel the amount of Ca bound at the membrane surface exceeds that which is free within the associated hemisphere. The ratio of bound Ca/free Ca increases as r decreases, such that the [Ca] nearest the Ca channel is the most strongly buffered by sarcolemmal bound Ca. It is demonstrated that Ca ions coming from these sarcolemmal Ca binding sites contribute quantitatively to the integrated Ca current. The electric field generated by the local depletion of Ca near the channel mouth has little impact on the extent of Ca depletion, but if an additional electric field exists at the mouth of the channel, Ca depletion can be significantly altered. Other low-affinity Ca binding sites in the interstitium may also contribute to the buffering of extracellular Ca. The complex geometry of the extracellular space in cardiac muscle (e.g., transverse tubules and restrictions of extracellular space between cells) increases both the predicted Ca depletions (in the absence of binding) and the bound/free ratio. Thus, the impact of this surface Ca binding is greatly increased. By considering arrays of Ca channels in transverse tubules or in parallel planes (e.g., membranes of neighboring cells), extracellular Ca depletions are predicted which agree with those measured experimentally. Membrane Ca binding may also be expected to buffer increases in [Ca] around the inner mouth of Ca channels. It is demonstrated that in the absence of other intracellular systems most of the Ca entering the cell via Ca channels might be expected to be bound to the inner sarcolemmal surface. It is concluded that surface Ca binding may have a substantial impact on the processes of extracellular Ca depletion (and intracellular Ca accumulation).     

Numerical simulation of local calcium movements during L-type calcium channel gating in the cardiac diad.

Computer simulation was used to investigate the calcium levels after sarcolemmal calcium influx through L-type calcium channels (DHPRs) into the narrow diadic space of cardiac muscle. The effect of various cytosolic and membranebound buffers, diad geometry, DHPR properties (open time and current), and surface charge were examined. The simulations showed that phospholipid binding sites on the sarcolemmal membrane are the major buffer affecting free calcium ([Ca2+]) levels in the diad. The inclusion of surface charge effects calculated from Gouy-Chapman theory resulted in a marked decrease in [Ca2+] levels at all times and a faster decay of [Ca2+] after termination of DHPR influx. For a DHPR current of 200 fA, [Ca2+] at the center of the diad reached peak levels of approximately 73 microM. In larger diads (> or = 400 nm diameter), [Ca2+] decayed more slowly than in smaller diads (100-200 nm diameter), although peak [Ca2+] levels reached during typical DHPR open times were similar. For a wide range of DHPR single-channel current magnitudes (Ica = 25-200 fA), [Ca2+] levels in the diad were approximately proportional to ICa. The decrease in calculated [Ca2+] levels due to the effects of surface charge can be interpreted as resulting from an effective "volume expansion" of the diad space. Furthermore, the layer of increased [Ca2+] close to the sarcolemmal membrane can act as a fast buffer.     

Effect of ryanodine on cardiac calcium current and calcium channel gating current.

The effects of 100 microM ryanodine on the L-type calcium channel were studied using the pacth-clamp technique in isolated guinea pig ventricular myocytes. The inactivation kinetics of the calcium current were slowed down in the presence of ryanodine in agreement with the blockade of the release of calcium from the sarcoplasmic reticulum by the drug. The I-V and steady-state inactivation curves of the calcium current were shifted to negative values by ryanodine. A similar shift was observed in the activation and inactivation curves of the intramembrane charge movement associated with the calcium channel. Due to this shift, ryanodine slightly reduced the maximal amount of displaced charge although it did not modify the transition from the inactivated to the activated state (i.e., charge movement repriming). This result is in notable contrast with that obtained in skeletal muscle, where it has been found that ryanodine interferes with charge movement repriming. These results provide additional evidence of the postulated differences between the architecture of the excitation-contraction coupling system in cardiac and skeletal muscle.     

A minimal gating model for the cardiac calcium release channel.

A Markovian model of the cardiac Ca release channel, based on experimental single-channel gating data, was constructed to understand the transient nature of Ca release. The rate constants for a minimal gating scheme with one Ca-free resting state, and with two open and three closed states with one bound Ca2+, were optimized to simulate the following experimental findings. In steady state the channel displays three modes of activity: inactivated 1 mode without openings, low-activity L mode with single openings, and high-activity H mode with bursts of openings. At the onset of a Ca2+ step, the channel first activates in H mode and then slowly relaxes to a mixture of all three modes, the distribution of which depends on the new Ca2+. The corresponding ensemble current shows rapid activation, which is followed by a slow partial inactivation. The transient reactivation of the channel (increment detection) in response to successive additions of Ca2+ is then explained by the model as a gradual recruitment of channels from the extant pool of channels in the resting state. For channels in a living cell, the model predicts a high level of peak activation, a high extent of inactivation, and rapid deactivation, which could underlie the observed characteristics of the elementary release events (calcium sparks).     

Control of action potential duration by calcium ions in cardiac Purkinje fibers

It is well known that cardiac action potentials are shortened by increasing the external calcium concentration (Cao). The shortening is puzzling since Ca ions are thought to carry inward current during the plateau. We therefore studied the effects of Cao on action potentials and membrane currents in short Purkinje fiber preparations. Two factors favor the earlier repolarization. First, calcium-rich solutions generally raise the plateau voltage; in turn, the higher plateau level accelerates time- and voltage-dependent current changes which trigger repolarization. Increases in plateau height imposed by depolarizing current consistently produced shortening of the action potential. The second factor in the action of Ca ions involves iK1, the background K current (inward rectifier). Raising Cao enhances iK1 and thus favors faster repolarization. The Ca-sensitive current change was identified as an increase in iK1 by virtue of its dependence on membrane potential and Ko. A possible third factor was considered and ruled out: unlike epinephrine, calcium-rich solutions do not enhance slow outward plateau current, ikappa. These results are surprising in showing that calcium ions and epinephrine act quite differently on repolarizing currents, even though they share similar effects on the height and duration of the action potential.     

Fluctuations in membrane current driven by intracellular calcium in cardiac Purkinje fibers

Spontaneous oscillatory fluctuations in membrane potential are often observed in heart cells, but their basis remains controversial. Such activity is enhanced in cardiac Purkinje fibers by exposure to digitalis or K-free solutions. Under these conditions, we find that voltage noise is generated by current fluctuations that persist when membrane potential is voltage clamped. Power spectra of current signals are not made up of single time-constant components, as expected from gating of independent channels, but are dominated by resonant characteristics between 0.5 and 2 HZ. Our evidence suggests that the periodicity arises from oscillatory variations in intracellular free Ca that control ion movements across the surface membrane. The current fluctuations are strongly cross-correlated with oscillatory fluctuations in contractile force, and are inhibited by removing extracellular Ca or exposure to D600. Chelating intracellular Ca with injected EGTA also abolishes the current fluctuations. The oscillatory mechanism may involve cycles of Ca (or Sr) movement between sarcoplasmic reticulum and myoplasm, as previously suggested for skinned cardiac preparations. Our experiments in intact cells indicate that changes in surface membrane potential can modulate cytoplasmic Ca oscillations in frequency and perhaps amplitude as well. A two-way interaction between surface membrane potential and intracellular Ca stores may be a common feature of heart, neuron, and other cell types.     

Multiple effects of calcium antagonists on plateau currents in cardiac Purkinje fibers

We studied the influence of Mn, La, and D600 on action potentials and plateau currents in cardiac Purkinje fibers. The Ca antagonists each abolished the second inward current, but they failed to act selectively. Voltage clamp experiments revealed two additional effects: decrease of slow outward current (iotachi) activation, and increase of net outward time-independent plateau current. These effects occurred at inhibitor concentrations used in earlier studies, and were essential to the reconstruction of observed Ca antagonist effects on electrical activity. The inhibitory influence of Mn, La, and D600 on iotachi suggested that iotachi activation might depend upon prior Ca entry. This hypothesis was not supported, however, when [Ca]omicron was varied: elevating [Ca]omicron enhanced Ca entry, but iotachi was nevertheless depressed. Thus, the results suggested instead that Ca antagonists and Ca ions have rather similar effects on iotachi, possibly mediated by changes in membrane surface charge.     

Diffusion of dihydropyridine calcium channel antagonists in cardiac sarcolemmal lipid multibilayers.

A membrane bilayer pathway model has been proposed for the interaction of dihydropyridine (DHP) calcium channel antagonists with receptors in cardiac sarcolemma (Rhodes, D.G., J.G. Sarmiento, and L.G. Herbette. 1985. Mol. Pharmacol. 27:612-623) involving drug partition into the bilayer with subsequent receptor binding mediated (though probably not rate-limited) by diffusion within the bilayer. Recently, we have characterized the partition step, demonstrating that DHPs reside, on a time-average basis, near the bilayer hydrocarbon core/water interface. Drug distribution about this interface may define a plane of local concentration for lateral diffusion within the membrane. The studies presented herein examine the diffusional dynamics of an active rhodamine-labeled DHP and a fluorescent phospholipid analogue (DiIC16) in pure cardiac sarcolemmal lipid multibilayer preparations as a function of bilayer hydration. At maximal bilayer hydration, the drug diffuses over macroscopic distances within the bilayer at a rate identical to that of DiI (D = 3.8 X 10(-8) cm2/s), demonstrating the overall feasibility of the membrane diffusion model. The diffusion coefficients for both drug and lipid decreased substantially as the bilayers were dehydrated. While identical at maximal hydration, drug diffusion was significantly slower than that of DiIC16 in partially dehydrated bilayers, probably reflecting differences in mass distribution of these probes in the bilayer.