The interaction of "K+-like" cations with the apical K+ channel in frog skin.

The apparent permeability of the apical K+ channel in the abdominal skin of the frog (Rana temporaria) for different monovalent cations was tested by comparing the short-circuit current (SCC) obtained after imposition of serosally directed ionic concentration gradients. Furthermore, the SCC was subjected to noise analysis. Of various cations tested, only the "K+-like" ions NH+4, Rb+ and Tl+, besides K+, were found to permeate the apical K+ channel, as reflected by SCC- and fluctuation analysis: (i) The SCC could be depressed by addition of the K+-channel blocker Ba2+ to the mucosal solution. (ii) With the K+-like ions (Ringer's concentration), a spontaneous Lorentzian noise was observed. Plateau values were similar for K+ and Tl+, and smaller for NH+4 and Rb+. The corner frequencies clearly increased in the order K+ less than NH+4 less than Tl+ much less than Rb+. The SCC dose-response relationships revealed a Michaelis-Menten-type current saturation only for pure K+- or Tl+-Ringer's solutions as mucosal medium, whereas a more complicated SCC behavior was seen with Rb+ and especially, NH+4. For K+-Tl+ mixtures an anomalous mole-fraction relationship was observed: At low [Tl+]/[K+] ratios, Tl+ ions appeared to inhibit competitively the K+ current while, at high [Tl+]/[K+] ratios, Tl+ seemed to be a permeant cation. This feature was also detected in the noise analysis of K+-Tl+ mixtures. Long-term exposure to mucosal Tl+ resulted in an irreversible deterioration of the tissue. The SCC depression by Ba2+ was of a simple saturation-type characteristic with, however, different half-maximal doses (NH+4 less than K+ less than Rb+). Ba2+ induced a "blocker noise" in presence of all permeant cations with corner frequencies that depended on the Ba2+ concentration. A linear increase of the corner frequencies of the Ba2+-induced noise with increasing Ba2+ concentration was seen for NH+4, Rb+ and K+. With the assumption of a pseudo two-state model for the Ba2+ blockade the on- and off-rate constants for the Ba2+ interaction with the NH+4/Rb+/K+ channel were calculated and showed marked differences, dependent on the nature of the permeant ion. The specific problems with Tl+ prevented such an analysis but SCC- and noise data indicated a comparably poor efficiency of Ba2+ as Tl+-current inhibitor. We attempted a qualitative analysis of our results in terms of a "two-sites, three-barriers" model of the apical K+ channel in frog skin.