1. Cone photoreceptors in several vertebrate species generate Ca-dependent regenerative depolarizations (e.g., Ca spikes lasting up to 2 s) in response to current injection or surround illumination and may remain in a state of prolonged depolarization (e.g., a permanent plateau near 0 mV) after these stimuli. This paper, while confirming the role of Ca channels in the regenerative depolarization, demonstrates that Ca-activated Cl channels either enhance or hinder prolonged depolarization, depending on the value of the chloride equilibrium potential (E(Cl)). 2. Current- and voltage-clamp recordings obtained with the whole-cell patch-clamp technique were compared in 158 isolated tiger salamander cones to determine the contribution of specific ion channel types to the two forms of depolarizing response. Cones dialyzed with CsCl or KCl intracellular solution (such that E(Cl) = 0 mV) that had sustained negative slope regions in their current-voltage (I-V) relations recorded under voltage clamp, were, under current clamp, bistable with respect to their resting potential. Injection of approximately 20-pA steps of depolarizing current resulted in transitions from the negative stable membrane potential (near -50 mV) to a long lasting plateau around 0 mV. Injection of 200-300 pA of hyperpolarizing current could then force a return to the negative stable resting potential, although once repolarization occurred, current injection had to be reduced or terminated to prevent damaging hyperpolarization of the cell. 3. The inward currents accounting for the negative slope region of the I-V relation were carried in Ca and Ca-activated Cl channels. Specific block of Ca-activated Cl current (I(Cl(Ca))) by 100 muM niflumic acid (NFA) eliminated the prolonged depolarization, even though the negative slope conductance region in the I-V persisted and the cone could still produce the briefer Ca-dependent regenerative depolarizations. Application of 100 muM Cd2+ blocked both forms of depolarization. 4. Substitution of Ba2+, which among other actions did not activate I(Cl(Ca))), usually supported regenerative depolarizations of shortened duration, demonstrating the role of Ca channels in the initial phase of these responses. 5. A difference was observed in the regenerative depolarization when E(Cl), was shifted away from 0 mV, where it had been in the experiments described above. With E(Cl), set to -40 or -60 mV by reduction of [Cl-] in the pipette, steady-state membrane bistability was eliminated and prolonged depolarization did not occur. Under these conditions, application of the Cl channel blocker NFA showed that I(Cl(Ca)) contributes to membrane hyperpolarization. 6. These results indicate a significant role for Ca-activated Cl channels in the depolarizations evoked in cones. Ca currents and the concomitant negative slope region of the cone's I-V relation account for the initial regenerative phase of the response, which may then be prolonged by I(Cl(Ca)) if E(Cl) is positive. Although the exact value of E(Cl) in unperturbed cones is uncertain, should it lie negative to the Ca current activation range, activation of I(Cl(Ca)) will then oppose membrane depolarization and contribute to repolarization. Clarification of the role of I(Cl(Ca)) in prolonged depolarization should aid our understanding of other Cl--dependent functions of the cone inner segment and terminal.
