GATING PROPERTIES OF HAIR CELLS CALCIUM CHANNELS IN THE FROG SEMICIRCULAR CANAL

Three distinct components of Ca current were recently demonstrated in hair cells mechanically isolated from the semicircular canals of the frog, using the whole-cell configuration of the patch-clamp recording technique. The L and R components were isolated on the basis of specific voltage dependencies, the differential sensitivities to drugs and toxins, and the fast run-down observed for one current component but not for the others. The L current increases in size during repetitive stimulation (run-up) and is carried by a non-inactivating, nifedipine-sensitive channel type; the second channel carries an R2, mibefradil-sensitive current; the third channel, which is present in 40% of the cells studied, carries an inactivating R1 current, also sensitive to mibefradil, which runs down first. The inactivation of the R1 component is Ca-dependent (mean time constanti about 6.7 ms in 4 mM external Ca); it is slowed down on decreasing extracellular Ca and disappears upon substituting an equiosmolar amount of Ba for external Ca. To further investigate whether modifications in intracellular Ca concentration ([Ca]i) at the cytoplasmic side of the channel affect the inactivation of the R1 component and in general the gating of all channel types, Ca currents were recorded in the presence of increasing amounts of BAPTA in the patch pipette. BAPTA is indeed expected to clamp [Ca]i more efficiently and rapidly than EGTA. Since K currents heavily contaminate recordings obtained using high concentrations of BAPTA in its commercially available K salt form, in the present experiments custom-synthesized tetracesium salt of BAPTA was employed. It resulted that even the highest concentrations of BAPTA used (50 mM) did not preverit Ca channel inactivation (and therefore [Ca]i rise), although, as expected, the kinetics of R1-channel inactivation was progressively slowed down upon increasing BAPTA concentration (up to tenfold in 25 mM BAPTA). Surprisingly, current activation and deactivation were also slowed down (up to two- and threefold, respectively, in 25 mM BAPTA), suggesting a role of [Ca]i in the kinetics of channel opening and closing