Modulation of Ca(2+)-activated K+ currents and Ca(2+)-dependent action potentials by exocytosis in goldfish bipolar cell terminals

Abstract

Retinal bipolar cells convey light-evoked potentials from photoreceptors to ganglion cells and mediate the initial stages of visual signal processing. They do not fire Na(+)-dependent action potentials (APs) but the Mb1 class of goldfish bipolar cell exhibits Ca(2+)-dependent APs and regenerative potentials that originate in the axon terminal. I have examined the properties of Ca(2+)-dependent APs in isolated bipolar-cell terminals in goldfish retinal slices. All recorded terminals fired spontaneous or evoked APs at frequencies of up to 15 Hz. When an AP waveform was used as a voltage stimulus, exocytosis was evoked by single APs, maintained throughout AP trains and modulated by AP frequency. Furthermore, feedback inhibition of the Ca2+ current (I(Ca)) by released vesicular protons reduced depression of exocytosis during AP trains. In the absence of K+ current inhibition, step depolarizations and AP waveforms evoked a rapidly activated outward current that was dependent on Ca2+ influx I(K(Ca). I therefore investigated whether proton-mediated feedback inhibition of I(Ca) affected the activation of I(K(Ca)). A transient inhibition of I(K(Ca)) was observed that was dependent on exocytosis, blocked by high-pH extracellular buffer, of similar magnitude to inhibition of I(Ca) but occurred with a delay of 2.7 ms. In addition, the amplitude of APs evoked under current clamp was inhibited by the action of vesicular protons released by the APs. Protons released via exocytosis may therefore be a significant modulator of Ca(2+)-dependent currents and regenerative potentials in bipolar-cell terminals.

Figure 1. Isolated bipolar-cell terminals in retinal slices fire spontaneous and evoked APs

A, current-clamp recording from a bipolar-cell terminal with K⁺-based intracellular solution. This terminal did not fire APs spontaneously, but APs were evoked by injection of +4 pA of current. In this and subsequent Figures, membrane voltages have been corrected for liquid-junction potential. B, membrane voltage response from a terminal that fired spontaneous APs. Negative current injection (ramp from 0 to −10 pA) caused hyperpolarization and inhibited AP firing. C, modulation of AP frequency by current injection in control recordings (n = 13) and in the presence of picrotoxin (50 μm; n = 8). Data points show mean ± s.e.m.D, voltage responses from a terminal in which AP amplitude and duration were unaffected by changing AP frequency within the low range. This recording was made in the presence of picrotoxin (50 μm). E, voltage responses from a terminal firing high-frequency APs in which further current injection evoked small, variable voltage oscillations rather than regular APs. F, voltage response from a terminal to a current ramp from 0 to +25 pA. AP frequency increased and amplitude decreased with increasing current. G, voltage response to a 0- to +25-pA ramp from a terminal in the presence of the L-type Ca²⁺ channel antagonist nifedipine (100 μm) plus picrotoxin (50 μm). No APs were observed in the presence of nifedipine.

Figure 2. Ca²⁺ influx and exocytosis in response to AP waveforms

A, voltage-clamp recording from a bipolar-cell terminal with Cs⁺-based intracellular solution in the presence of picrotoxin (50 μm). A typical AP waveform previously recorded under current clamp was used as the stimulus, and membrane capacitance was measured before and after the depolarization using a voltage sinewave. The AP evoked an inward Ca²⁺ current (I_Ca) and an increase in membrane capacitance (ΔC_m). For comparison, I_Ca and ΔC_m evoked in the same terminal by a 25-ms step depolarization to −10 mV are shown on the right. B, I_Ca and ΔC_m responses in the same terminal as A evoked by a 5.6-Hz train of 20 APs (stimulus shown at top). I_Ca was smaller in response to the first AP than subsequent APs in the train. ΔC_m was greatest in response to the first AP, and exocytosis was evoked by APs throughout the train. C, mean ΔC_m evoked by the first, second, and third to 10th APs in a 5.6-Hz train (n = 29). Data from recordings using K⁺-based intracellular solution were included. D, mean Ca²⁺ influx evoked by the first, second, and third to 10th APs in a 5.6-Hz train (n = 14).

Figure 3. Released vesicular protons affect Ca²⁺ influx and exocytosis during AP trains

A, I_Ca evoked by a pair of 25-ms step depolarizations to −10 mV (interval, 100 ms) in the presence of picrotoxin (50 μm), (first, black; second, grey). Under control extracellular pH buffering conditions (24 mm HCO₃⁻), prominent inhibition of I_Ca by released vesicular protons occurred during the first depolarization, but not during the second when exocytosis was depressed. In a different terminal, the addition of 24 mm Hepes to the extracellular solution greatly reduced inhibition of I_Ca. B, I_Ca evoked by a 5.6-Hz train of 10 APs in the same terminals as A, (first AP, black; second to 10th APs, grey). The addition of 24 mm Hepes greatly reduced the inhibition of I_Ca during the first AP. C, in the presence of 24 mm Hepes, the first AP evoked a significantly larger ΔC_m than in control conditions (left), but there was no significant difference in the total ΔC_m evoked by 10 APs (middle: control, n = 29; Hepes, n = 17). There was significantly more Ca²⁺ influx during the 10-AP train in the presence of Hepes than in control conditions (right: control, n = 14; Hepes n = 6).

Figure 4. Bipolar-cell terminal exocytosis is sensitive to AP frequency

A, ΔC_m evoked by two trains of APs (three APs at 2.33 Hz and 10 APs at 7.78 Hz) in one terminal (control conditions). Approximately twice as much exocytosis was evoked by the 7.78-Hz train as by the 2.33-Hz train. B, ΔC_m evoked by the 2.33-Hz and 7.78-Hz AP trains in eight terminals (•) and mean ± s.e.m.ΔC_m values (□).

Figure 5. Inhibition of Ca²⁺-activated K⁺ current (I_K(Ca)) by released vesicular protons

A, membrane currents evoked by step depolarizations to −10 mV with K⁺-based intracellular solution in control conditions (black) and in the presence of nifedipine (100 μm; grey). To the right, the initial 20-ms period of depolarization is shown on an expanded time scale. Nifedipine inhibited I_Ca and both a rapidly activated component and a slowly developing component of I_K(Ca). Recordings were made in the presence of picrotoxin (50 μm) to block GABA-mediated currents. B, membrane currents and ΔC_m evoked by paired 25-ms step depolarizations to −10 mV (interval, 100 ms) in the presence of picrotoxin (50 μm). Black traces show responses early in the recording when the first depolarization evoked a large ΔC_m; superimposed (grey traces) are responses later in the recording when exocytosis had run down. The first depolarization, associated with the largest ΔC_m, showed a prominent transient inhibition of I_K(Ca). C, membrane currents evoked by a pair of 25-ms depolarizations to −10 mV in the presence of picrotoxin (50 μm), (first, black; second, grey), in a terminal that showed little outward current inactivation during the depolarization. Inhibition of I_K(Ca) was observed during the first depolarization. D, as for C, but with the addition of 24 mm Hepes to the extracellular solution. Inhibition of I_K(Ca) during the first depolarization was greatly reduced.

Figure 6. Inhibition of I_K(Ca) during AP trains

A, the current response to an AP waveform with K⁺-based intracellular solution in the presence of picrotoxin (50 μm). A brief inward I_Ca was followed by more sustained outward current. B, the current response to an AP waveform in a different terminal in the presence of nifedipine (100 μm). Both I_Ca and the outward current (I_K(Ca)) were inhibited. C, the currents evoked by APs in a 5.6-Hz train (first AP, black; second to 10th APs, grey) in the presence of picrotoxin (50 μm). I_K(Ca) was smaller during the first AP than during subsequent APs. D, mean outward charge of first AP as a percentage of the average outward charge during the second to tenth APs, in control conditions (n = 12) and in the presence of 24 mm extracellular Hepes (n = 10). The inhibition of I_K(Ca) of the first AP is likely to result from inhibition of Ca²⁺ influx by released vesicular protons.

Figure 7. Released protons modulate AP firing in bipolar-cell terminals

A, example current-clamp recordings from a terminal in control conditions and a terminal in the presence of 24 mm extracellular Hepes. Long-duration, larger-amplitude spikes (*) were occasionally observed in control conditions, and were more prominent in the presence of Hepes. B, recordings from a control terminal and a terminal in the presence of 24 mm Hepes that mainly exhibited regular AP firing. The terminals were held below AP threshold for at least 10 s, followed by injection of a small amount of positive current (2–5 pA) to evoke AP firing. In control conditions, the first AP was noticeably smaller than subsequent APs in the train, whereas this difference was reduced in the presence of Hepes. Both recordings were made in the presence of picrotoxin (50 μm). C, a graph of peak inhibition of I_K(Ca) during paired step depolarizations to −10 mV against the inhibition of the first evoked AP (difference between the first AP maximum amplitude and the average of the third to 10th AP maximum amplitudes) in the same terminals; the data were fitted with a straight line. D, mean peak AP amplitudes for the first evoked AP and average of the third to 10th APs in control conditions and in the presence of 24 mm extracellular Hepes. Hepes significantly reduced the inhibition of the first AP.