Synaptic transmission mediated by single club endings on the goldfish Mauthner cell. II: Plasticity of excitatory postsynaptic potentials

Simultaneous pre- and postsynaptic intracellular recordings were used to analyze the properties of chemically mediated synaptic transmission between single club endings of eighth nerve afferents and the goldfish Mauthner (M-) cell lateral dendrite. The EPSPs exhibited pronounced facilitation when the presynaptic fiber fired high-frequency bursts of 2 or 3 impulses at intervals of 2–4 msec. The amplitudes of the EPSPs evoked by the second and third presynaptic impulses of a burst were, on average, 99 and 108% larger than that evoked by the first impulse. A cross-correlation analysis showed that the amplitudes of the control and facilitated EPSPs fluctuated independently, indicating that the facilitation was mediated by a presynaptic mechanism. This conclusion was supported by a comparison of the coefficient of variation for the control and facilitated EPSPs, on the basis of a binomial release model. In addition, the value of binomial n, the number of presynaptic release units, was not changed during facilitation. The origin of EPSP fluctuations was analyzed by examining the correlation between the amplitudes of EPSPs and those of the electrotonic coupling potentials associated with them. The absence of correlation between the 2 variables suggested that the fluctuations of EPSPs were not due to a variable presynaptic impulse invasion. The EPSP fluctuations were further analyzed by assuming that the facilitation was associated with an increase in the probability (p) of transmitter release and that the release process followed simple binomial statistics. The binomial variables thus calculated were n = 6–11, p = 0.29–0.44, and q = 31–61 microV, values comparable to the estimates for other CNS synapses. More importantly, these parameters provided satisfactory fits to the amplitude histograms of the control and facilitated EPSPs. The number of release units, n, was smaller than, but in a range similar to, the number of active zones identified in the freeze-fracture study of the club endings (Kohno and Noguchi, 1986). This correlation is consistent with the notion that active zones are the structural correlates of quantal release units. In the preceding paper, it was shown that impulses in a majority of club endings electrotonically coupled to the M-cell do not produce a detectable chemically mediated EPSP, although the contacts have the morphological correlates of chemical synapses. In an attempt to activate these “silent” connections, 2 approaches were used. First, the burst-firing paradigm, which could effectively facilitate EPSPs already present, failed to reveal any EPSPs at the silent junctions.