The axon terminal of goldfish retinal horizontal cells: a low membrane conductance measured in solitary preparations and its implication to the signal conduction from the soma.

1. Mechanical dissociation of the enzyme-treated goldfish retina yielded somata and axon terminals of horizontal cells. The membrane properties of these solitary axon terminals were investigated using the whole-cell patch-clamp technique. 2. Axon terminals had a large input resistance, comparable to the seal resistance (approximately 30 G omega). Most axon terminals (greater than 80%) showed a nearly linear current-voltage relation between -60 and +10 mV, where the slope conductance was as small as 5 muS/cm2. Some axon terminals showed a shallow negative slope conductance in the same potential range. 3. The membrane current consisted of two components: transient and sustained. The transient component was carried by sodium ions, and the sustained component was a mixture of calcium and potassium currents. The sodium current (INa) was activated by depolarization of beyond -45 mV and was maximal (approximately 60 pA) at -10 mV. It was blocked by 5 microM tetrodotoxin and disappeared in Na+-free medium. The maximum amplitude of INa was less than 10% of INa of the soma. 4. A small calcium current (less than 6 pA) was isolated in a small proportion of cells, with an amplitude approximately 5% of the calcium current evoked in the soma under the identical recording conditions. 5. A small amount of potassium current through the anomalous rectifier was induced in the axon terminal when the membrane potential was below -60 mV. Its conductance was 15-20 muS/cm2, only 1/20 of the estimate in the soma. Other types of potassium currents were not detected. 6. It is concluded that the soma and the axon terminal have a similar set of membrane currents, but the specific membrane conductance of the axon terminal is extremely low. The signal conductivity from soma to axon terminal was assessed using a passive cable model together with numerical values obtained from the present experiments. Although the membrane conductance of the connecting axon was not measurable directly, the calculation strongly suggests that low conductance of the axon terminal membrane minimizes the leakage of signals arriving electrotonically through the thin connecting axon, even if the membrane conductance of the axon was overestimated as being identical to the soma membrane. 7. These results can explain why light-evoked responses recorded from the axon terminal are similar in amplitude as well as in waveform to those recorded from the soma, despite the lack of direct inputs from photoreceptors.