Abstract
1. The kinetics of the slow inactivation process of Na+ channels were examined by recording single-channel currents from cultured neuroblastoma cells. 2. In order to directly examine slow inactivation, fast inactivation was first removed irreversibly by briefly exposing the internal surface of excised membranes to papain. Following treatment, the time constant for the inactivation of averaged membrane Na+ current increased by over two orders of magnitude, while the open time of individual channels increased by a factor of three. The two effects are consistent with the idea that papain can selectively remove fast inactivation of Na+ channels. 3. In the absence of fast inactivation, Na+ channels continued to open during maintained depolarization of the membrane to potentials less negative than -60 mV. Under these conditions, the opening occurred in bursts 50 ms to hundreds of milliseconds long, followed by silent periods lasting many seconds. The average burst length was found to be equal to the time constant of the decline in average evoked current measured at the same potential, indicating that a burst was terminated by entry of the channel into the slow inactivated state. 4. Histograms of open times revealed two populations of open states at any potential. Bursts could also be classified as either short or long bursts. Bursts appeared to be due to the gating of a single channel, and long bursts contained both types of open states, suggesting that a Na+ channel could have more than one open state. 5. The kinetics of bursts of Na+ channels were voltage dependent. As the membrane was depolarized, the burst length, interval between bursts, and open time all increased. Although the probability of an open channel during a burst increased to almost 1.0 with depolarization, any channel was open less than 0.5% of the time when measured throughout the depolarization. The increase in burst duration with depolarization would occur if the rate of slow inactivation is faster from closed states of the channel than from open states. 6. Records of membrane current evoked by a series of step depolarizations were clustered into those with openings of Na+ channels and those without openings. Records in which a channel did not inactivate during the depolarization were less likely to lead to hibernation, suggesting that this phenomenon is caused by the slow inactivation process.