Voltage-dependent properties of macroscopic and elementary calcium channel currents in guinea pig ventricular myocytes.

Whole-cell Ca channel currents were recorded from guinea pig ventricular myocytes that were internally perfused with Cs solution and bathed in solutions containing 3.6 mM Ca, 3.6 mM Ba or 90 mM Ba (34 degrees C). Single Ca channel currents were recorded from cell-attached membrane patches of similar myocytes; the patch pipettes contained a 90 mM Ba solution. 1. Although the shape of the whole-cell I-V relation was independent of the bathing solution, this was not the case with the location of the inward current maximum (Vpeak); Vpeak in 90 mM Ba was about 30 mV positive to Vpeak in 3.6 mM Ba. 2. The activation and inactivation of whole-cell currents were voltage dependent. Compared to the voltage dependencies in 3.6 mM Ba, those in 90 mM Ba were shifted by about 30 mV to the right, suggesting a neutralization of surface charges. 3. Observations compatible with the ion permeation model proposed by Hess and Tsien (1984) included (a) a depression of current during Ca/Ba solution exchange, (b) a high divalent to monovalent ion permeability, and (c) rectification of the outward limb of the I-V relation. 4. Estimated current densities at Vpeak were similar for myocytes in 3.6 mM Ca and 3.6 mM Ba, and about 10 times larger in 90 mM Ba. 5. Average currents (I) calculated from ensembles of records of single Ca channel current had voltage-dependent time courses resembling those of whole-cell IBa (90 mM). 6. Single-channel I-V relations were superimposable on whole-cell I-V curves suggesting that voltage-dependent single-channel parameters (probability of opening, elementary current amplitude) can be related to the voltage-dependent macroscopic current parameters (activation, instantaneous I-V relation) when scaled by channel number. 7. The density of Ca channels in myocytes was calculated from whole-cell IBa (90 mM) and average current through single channels. The outcome, 3-5 channels/micron 2, agrees with two other recent estimates (Tsien et al. 1983; Lux and Brown 1984). However, it is difficult to reconcile with the much lower density that one would forecast from the frequency of functional channel observation in myocyte membrane patches (Pelzer et al. 1985c).