Presynaptic excitability.

Based on functional characterizations with electrophysiological techniques, the channels in nerve terminals appear to be as diverse as channels in nerve cell bodies (Table I). While most presynaptic Ca2+ channels superficially resemble either N-type or L-type channels, variations in detail have necessitated the use of subscripts and other notations to indicate a nerve terminal-specific subtype (e.g., Wang et al., 1993). Variations such as these pose a serious obstacle to the identification of presynaptic channels based solely on the effects of channel blockers on synaptic transmission. Pharmacological sensitivity alone is not likely to help in determining functional properties. Crucial details, such as voltage sensitivity and inactivation, require direct examination. It goes without saying that every nerve terminal membrane contains Ca2+ channels as an entry pathway so that Ca2+ can trigger secretion. However, there appears to be no general specification of channel type, other than the exclusion of T-type Ca2+ channels. T-type Ca2+ channels are defined functionally by strong inactivation and low threshold. Some presynaptic Ca2+ channels inactivate (posterior pituitary and Xenopus nerve terminals), and others have a somewhat reduced voltage threshold (retinal bipolar neurons and squid giant synapse). Perhaps it is just a matter of time before a nerve terminal Ca2+ channel is found with both of these properties. The high threshold and strong inactivation of T-type Ca2+ channels are thought to be adaptations for oscillations and the regulation of bursting activity in nerve cell bodies. The nerve terminals thus far examined have no endogenous electrical activity, but rather are driven by the cell body. On functional grounds, it is then reasonable to anticipate finding T-type Ca2+ channels in nerve terminals that can generate electrical activity on their own. The rarity of such behavior in nerve terminals may be associated with the rarity of presynaptic T-type Ca2+ channels. In four of the five preparations reviewed in this chapter--motor nerve, squid giant synapse, ciliary ganglion, and retina bipolar neurons--evidence was presented that supports a location for Ca2+ channels that is very close to active zones of secretion. All of these synapses secrete from clear vesicles, and the speed and specificity of transduction provided by proximity may be a common feature of these rapid synapses. In contrast, the posterior pituitary secretion apparatus may be triggered by higher-affinity Ca2+ receptors and lower concentrations of Ca2+ (Lindau et al., 1992). This would correspond with the slower performance of peptidergic secretion, but because of the large stimuli needed to evoke release from neurosecretosomes, the possibility remains that the threshold for secretion is higher than that reported. While the role of Ca2+ as a trigger of secretion dictates a requirement for voltage-activated Ca2+ channels as universal components of the presynaptic membrane, the presence of other channels is more difficult to predict. Depolarizations caused by voltage-activated Na+ channels activate the presynaptic Ca2+ channels, but whether this depolarization requires Na+ channels in the presynaptic membrane itself may depend on the electrotonic length of the nerve terminal. Variations in density between motor nerve terminals may reflect species differences in geometry. The high Na+ channel density in the posterior pituitary reflects the great electrotonic length of this terminal arbor. Whether Na+ channels are abundant or not in a presynaptic membrane, K+ channels provide the most robust mechanism for limiting depolarization-induced Ca2+ entry. K+ channel blockers enhance transmission at most synapses. In general, K+ channels are abundant in nerve terminals, although their apparent lower priority compared to Ca2+ channels in the eyes of many investigators leaves us with fewer detailed investigations in some preparations. Most nerve terminals have more than