Responses of tetrodotoxin-sensitive (TTX-s) and insensitive (TTX-i) Na(+) channels, in frog dorsal root ganglion (DRG) cells and frog heart Na(+) channels, to two grayanotoxin (GTX) analogs, GTX-I and alpha-dihydro-GTX-II, were examined using the patch clamp method. GTX-evoked modification occurred only when repetitive depolarizing pulses preceded a single test depolarization; modification, during the test pulse, was manifested by a decrease in peak Na(+) current accompanied by a sustained Na(+) current. GTX-evoked modification of whole-cell Na(+) currents was quantified by normalizing the conductance for sustained currents through GTX-modified Na(+) channels to that for the peak current through unmodified Na(+) channels. The dose-response relation for GTX-modified Na(+) channels was constructed by plotting the normalized slope conductance against GTX concentration. With respect to DRG TTX-i Na(+) channels, the EC(50) and maximal normalized slope conductance were estimated to be 31 microM and 0.23, respectively, for GTX-I, and 54 microM and 0.37, respectively, for alpha-dihydro-GTX-II. By contrast, TTX-s Na(+) channels in DRG cells and Na(+) channels in ventricular myocytes were found to have a much lower sensitivity to both GTX analogs. In single-channel recording on DRG cells and ventricular myocytes, Na(+) channels modified by the two GTX analogs (both at 100 microM), had similar relative conductances (range, 0.25-0.42) and open channel probabilities (range, 0.5-0.71). From these observations, we conclude that the differences in responsiveness of DRG TTX-i, and ventricular whole cell Na(+) currents to the GTX analogs studied are related to the number of Na(+) channels modified.

Inactivation of the fast Na+ current of heart muscle occurs in two kinetically distinct phases: a fast process operating on a millisecond time scale and a considerably slower process, the kinetic properties of which have not been explored fully. In this study, we analysed the slow inactivation process in isolated frog ventricular myocytes using the whole-cell variation of the patch-clamp method. Slow inactivation of the Na+ current followed a double-exponential time course, corresponding to slow and ultraslow components of Na+ channel inactivation. The individual time constants were 2-7 s (slow component) and 40-560 s (ultraslow component). Recovery from these slow inactivation processes also followed a double-exponential time course, but was characterized by significantly briefer time constants than those for the inactivation process. The relationship between transmembrane potential and steady-state slow or ultraslow inactivation was well described by the Boltzmann equation. The membrane potential at which half the Na+ channels are inactivated (V1/2) and the slope factor were estimated to be -48.1 and 13.6 mV, respectively, for the slow component alone. Under conditions in which the slow and ultraslow inactivation components were both present, these parameters were -53.1 and 8.7 mV respectively. When the fast and the two slow inactivation processes occurred concomitantly, the resultant steady-state inactivation curves were shifted to more negative potentials and the slope factor was decreased. Treatment with 1 mM Cd2+ externally did not affect the time course of slow inactivation, but produced a 3-7 mV depolarizing shift in its steady-state voltage dependency by virtue of cadmium's known effect on the cell surface potential. This study has thus identified two components of slow Na+ inactivation in heart muscle, operating on a time scale of seconds (slow inactivation) and minutes (ultraslow inactivation).