A calcium-activated potassium channel causes frequency-dependent action-potential failures in a mammalian nerve...

1. The contribution of a calcium-activated potassium channel to action-potential failure was studied in nerve terminals of the rat posterior pituitary. 2. Depolarizing current injections under current clamp were faithfully followed by action potentials for stimulation frequencies of < or = 12 Hz. Further increases in frequency resulted in action-potential failure within a few hundred milliseconds. The fraction of failures increased with stimulation frequency. This decrease in excitability was concomitant with a hyperpolarization from -57.3 +/- 1.4 to -61.3 +/- 1.4 (SE) mV. 3. The decrease in excitability was dependent on calcium influx through voltage-dependent calcium channels, because action-potential failures did not occur at frequencies < or = 30 Hz in the presence of cadmium. The dihydropyridine agonist BayK 8644 increased the fraction of failed action potentials. 4. Depolarizations from -80 to 10 mV for 3 s evoked macroscopic potassium currents with a rapidly activated, transient component and a slowly developing, noninactivating component. The late outward current was dependent on calcium influx, because it was reduced by cadmium and enhanced by BayK 8644. 5. Tetraethylammonium and 4-aminopyridine effectively blocked potassium outward currents but failed to distinguish this calcium-dependent potassium channel from the other two potassium channels in this preparation. Charybdotoxin and apamin did not affect potassium currents in this preparation. 6. In excised inside-out patches, the calcium-dependent potassium channel had a slope conductance of 193 pS. The open probability changed e-fold per 14.8 mV change in membrane potential with a calcium concentration at the cytoplasmic membrane face ([Ca]i) of 100 nM. 7. The channel was highly sensitive to [Ca]i. Depolarizations to 100 mV at 10 nM [Ca]i activated the channel half-maximally. When [Ca]i was raised to 250 nM, the voltage for half-maximal activation shifted to -16 mV. Calcium also decreased the steepness of the voltage activation curve. 8. At a constant membrane potential, pressure ejection of calcium to the cytosolic face of an excised patch activated the channel with a delay of 82 ms. This slow activation in excised patches was consistent with the slow activation of the delayed component of the macroscopic current. 9. At constant calcium concentration, the time course of activation exhibited a strong voltage dependence. Most of the channels did not inactivate during depolarizations lasting < or = 300 ms. 10. The channel exhibited complex gating, with at least two distinct open and closed states.(ABSTRACT TRUNCATED AT 400 WORDS)