Although each of the fundamental processes involved in excitation-contraction coupling in mammalian heart has been identified, many quantitative details remain unclear. The initial goal of our experiments was to measure both the transmembrane Ca2+ current, which triggers contraction, and the Ca2+ extrusion due to Na(+)-Ca2+ exchange in a single ventricular myocyte. An action potential waveform was used as the command for the voltage-clamp circuit, and the membrane potential, membrane current, [Ca2+]i, and contraction (unloaded cell shortening) were monitored simultaneously. Ca(2+)-dependent membrane current during an action potential consists of two components: (1) Ca2+ influx through L-type Ca2+ channels (ICa-L) during the plateau of the action potential and (2) a slow inward tail current that develops during repolarization negative to approximately -25 mV and continues during diastole. This slow inward tail current can be abolished completely by replacement of extracellular Na+ with Li+, suggesting that it is due to electrogenic Na(+)-Ca2+ exchange. In agreement with this, the net charge movement corresponding to the inward component of the Ca(2+)-dependent current (ICa-L) was approximately twice that during the slow inward tail current, a finding that is predicted by a scheme in which the Ca2+ that enters during ICa is extruded during diastole by a 3 Na(+)-1 Ca2+ electrogenic exchanger. Action potential duration is known to be a significant inotropic variable, but the quantitative relation between changes in Ca2+ current, action potential duration, and developed tension has not been described in a single myocyte. We used the action potential voltage-clamp technique on ventricular myocytes loaded with indo 1 or rhod 2, both Ca2+ indicators, to study the relation between action potential duration, ICa-L, and cell shortening (inotropic effect). A rapid change from a "short" to a "long" action potential command waveform resulted in an immediate decrease in peak ICa-L and a marked slowing of its decline (inactivation). Prolongation of the action potential also resulted in slowly developing increases in the magnitude of Ca2+ transients (145 +/- 2%) and unloaded cell shortening (4.0 +/- 0.4 to 7.6 +/- 0.4 microns). The time-dependent nature of these effects suggests that a change in Ca2+ content (loading) of the sarcoplasmic reticulum is responsible. Measurement of [Ca2+]i by use of rhod 2 showed that changes in the rate of rise of the [Ca2+]i transient (which in rat ventricle is due to the rate of Ca2+ release from the sarcoplasmic reticulum) were closely correlated with changes in the magnitude and the time course of ICa-L.(ABSTRACT TRUNCATED AT 400 WORDS)

2+ ] i (calcium sparks) have been identified in cardiac, smooth and skeletal muscle (reviewed by Cheng et al. 1996b). A number of authors have suggested that Ca 2+ transients are due the spatio-temporal summation of many Ca 2+ sparks Cheng et al. 1995;Tanaka et al. 1998). These Ca 2+ sparks are the result of release of Ca 2+ from the SR and are believed to be the elementary events underlying E-C coupling in cardiac muscle (Cheng et al. 1993;Bridge et al. 1999). Shacklock et al. (1995) have shown that location of sparks coincides with the location of t-tubules in rat ventricular myocytes. Since L-type Ca 2+ channels and SR Ca 2+ release are co-localized at the t-tubule dyad junction, it has been suggested that a local accumulation of [Ca 2+ ] i near one or a cluster of SR Ca 2+ release channels gates their opening Cheng et al. 1996a). However, it is unclear whether Ca 2+ channels or junctions with the sarcolemma are absolutely necessary for the spontaneous occurrence of sparks.In contrast to the information available on sparks and Ca 2+ transients in ventricular muscle, relatively little is known about the spatio-temporal characteristics of Location of the initiation site of calcium transients and sparks in rabbit heart Purkinje cells ] i of ventricular cells showed a more uniform pattern of activation across the cell. Staining with di-8-ANEPPS revealed that Purkinje cells lack t-tubules, whereas ventricular cells have an extensive t-tubular system.3. When we superfused both cell types with a buffer containing 5 mM Ca 2+ -1 µM isoproterenol (isoprenaline) they produced Ca 2+ sparks spontaneously. Ca 2+ sparks occurred only at the periphery of Purkinje cells but occurred throughout ventricular cells. Sparks in both cell types could be completely abolished by addition of the SR inhibitor thapsigargin (500 nM). Brief exposure to nifedipine (10 µM) did not reduce the number of spontaneous sparks.4. Immunofluorescence staining of Purkinje cells with anti-ryanodine antibody revealed that ryanodine receptors (RyRs) are present at both peripheral and central locations.5. Computer simulations of experiments in which the calcium transient was evoked by voltage clamp depolarizations suggested that the increase in calcium observed in the centre of the cell could be explained by simple buffered diffusion of calcium. These computations suggested that the RyRs deep within the cell do not contribute significantly to the calcium transient.6. These results provide the first detailed, spatially resolved data describing Ca 2+ transients and Ca 2+ sparks in rabbit cardiac Purkinje cells. Both types of events are initiated only at subsarcolemmal SR Ca 2+ release sites suggesting that in Purkinje cells, Ca 2+ sparks only originate where the sarcolemma and sarcoplasmic reticulum form junctions. The role of the centrally located RyRs remains unclear. It is possible that because of the lack of t-tubules these RyRs do not experience a sufficiently large Ca 2+ trigger during excitation-contraction (E-C) coupling to become active.

Differences in action potential duration in different regions of the mammalian ventricle are not systematically present when quiescent tissue is first stimulated, but develop rapidly during repetitive activity. The effects of ouabain and temperature suggest the involvement of the Na+-K+ exchange pump.

SUMMARY1. A whole-cell gigaseal suction microelectrode voltage-clamp technique has been used to study slow inward tail currents in single myocytes obtained by enzymatic dispersion of rabbit ventricle and atrium. A variety of stimulation protocols, Tyrode solutions and pharmacological agents have been used to test three hypotheses: (a) that the slow inward tail current is generated by an electrogenic Na+-Ca2+ exchanger; (b) that a rise in [Ca2±]1, due to release from the sarcoplasmic reticulum can modulate the activity of this exchanger; and (c) that the uptake of calcium by the sarcoplasmic reticulum is a major determinant of the time course of the tail current.2. As shown previously in amphibian atrium and guinea-pig ventricle, slow inward tail currents can be observed consistently under conditions in which action potentials and ionic currents are recorded using microelectrode constituents which only minimally disturb the intracellular milieu.3. In ventricular cells, the envelope of these tail currents obtained by varying the duration of the preceding depolarizations shows that (a) the tail currents are activated by pulses as short as 10 ms, and reach a maximum for pulse durations of 100-200 ms, (b) the rate of decay of the tail current gradually increases as the activating depolarizations are prolonged, and (c) the tails cannot be due to deactivation of calcium currents, in agreement with other studies in frog heart. 7. The slow tail currents were changed significantly by increasing the temperature of the superfusing Tyrode solution. The major effect was a speeding up of the decay time. This was most apparent for tails following short depolarizations, which normally decay quite slowly at room temperature.8. Caffeine (2-5 mM) produced a prolongation of the decay time of the slow tails, and a small reduction in slow tail amplitudes. The peak of the 'envelope' of the tails was shifted so that it occurred at longer depolarizations.9. In atrial cells, somewhat similar tail currents could be recorded consistently. However, in atrium they decayed much faster and there was very little difference in decay time between short and long depolarizations. Caffeine also prolonged the decay time of these tail currents.10. The slow tail currents in atrial cells are very sensitive to stimulus rate. They were large after short rest periods (30-60 s), and declined to a steady-state level within one to two pulses (at 1-3 Hz) after stimulation was resumed.11. These results from rabbit ventricular and atrial cells support the hypothesis that the slow tail current reflects the electrogenic activity of the Na'-Ca2" exchanger. However, the observed changes in the slow tail currents caused by indirect manipulations of Ca2+ sequestration into the sarcoplasmic reticulum suggest that intracellular calcium homeostasis involves a complex interaction between Ca2+ sequestration into the sarcoplasmic reticulum and Ca2+ extrusion via the Na+-Ca2+exchanger.

Electrotonic effects of electrically coupling atrioventricular (AV) nodal cells to each other and to real and passive models of atrial and ventricular cells were studied using a technique that does not require functional gap junctions. Membrane potential was measured in each cell using suction pipettes. Mutual entrainment of two spontaneously firing AV nodal cells was achieved with a junctional resistance (Rj) of 500 M omega, which corresponds to only 39 junctional channels, assuming a single-channel conductance of 50 pS. Coupling of AV nodal and atrial cells at Rj of 50 M omega caused hyperpolarization of the nodal cell, decreasing its action potential duration and either slowing or blocking diastolic depolarization in the AV node myocyte. Opposite changes occurred in the atrial action potential. When AV nodal and ventricular cells were coupled at Rj of 50 M omega, nodal diastolic potential was markedly hyperpolarized and diastolic depolarization was completely blocked with little change in ventricular diastolic potential. However, coupling did elicit marked changes in the action potential duration of both cells, with prolongation in the nodal cell and shortening in the ventricular cell. Nodal maximum upstroke velocity was increased by both atrial and ventricular coupling, as expected from the hyperpolarization that occurred. With an Rj of 50 M omega, spontaneous firing was blocked in all single AV nodal pacemaker cells during coupling to a real or passive model of an atrial or ventricular cell. These results demonstrate that action potential formation and waveform in a single AV nodal cell is significantly affected by electrical coupling to other myocytes.