PAIRED INDIVIDUAL AND MEAN POSTSYNAPTIC CURRENTS RECORDED IN 4-CELL NETWORKS OF APLYSIA

1. Presynaptic neurons B4 and B5 of Aplysia buccal ganglia produce similar inhibitory postsynaptic currents (PSCs) in several postsynaptic follower cells. Two previous papers have characterized the variability of synaptic current amplitude and decay time both for individual PSCs and also for mean values characterizing synapses and have compared PSC amplitude and time course at different synapses sharing a common presynaptic or postsynaptic neuron. 2. To distinguish similarity in synaptic current amplitude or decay introduced by a common pre- or postsynaptic neuron from similarity because of factors common to the particular ganglion or animal, paired synapses were analyzed in four-cell networks in which each of two identified presynaptic neurons producing similar PSCs in each of two postsynaptic cells. Pairing the same synaptic data by common presynaptic or postsynaptic neuron tests if the presynaptic or postsynaptic element partially specifies a parameter; cross-pairing controls for more global factors. Paired values of peak conductance g(peak) and decay time constant τ were compared for both individual sequential PSCs and for averages characterizing synapses. Analyses of individual PSCs examine processes affecting synaptic plasticity on a time scale of seconds to minutes, while average values compare more slowly varying factors. 3. Peak amplitudes were compared between individual PSCs in each of 24 paired sets. Correlations of g(peak) fluctuations were significantly larger for PSCs produced by the same presynaptic neuron than for postsynaptic or cross pairings (P < 0.05), consistent with partially correlated fluctuations in transmitter release at different presynaptic terminals. 4. Firing rates of individual presynaptic neurons were modulated to induce variability of test PSCs. These manipulations altered synaptic peak amplitudes in paired postsynaptic neurons, although not to the same degree. Manipulation of a single presynaptic neuron modulated input from that neuron alone to common postsynaptic cells without any effect on input from the paired presynaptic neuron. When fluctuations in the amplitude of g(peak) were examined in runs incorporating presynaptic modulation, correlations were strong for sets of PSCs sharing a common presynaptic neuron (R = 0.87), significantly greater (P < 0.001) than for other pairings. 5. In contrast to the partial presynaptic specification of fluctuations of individual PSCs, values of synaptic amplitude and time course averaged over 21-132 PSCs at a given synapse reflect postsynaptic determinants. Mean values of g(peak) characterizing synapses paired by common postsynaptic cell are highly similar (P = 0.0001), in contrast to the lack of similarity seen when the same data are presynaptically (P = 0.11) or cross (P = 0.36) paired. 6. Mean values of τ are again highly similar when paired by common postsynaptic cell (P = 0.0001); however, pairing either by common presynaptic cell or cross pairing also yields significant similarity (P < 0.005). Cell-to-cell variations in decay time constant are not a consequence of temperature differences. 7. Unlike the mean values of amplitude and time course, the coefficients of variation (CV) of these parameters show no significant similarity when paired by common postsynaptic neuron. However, CV of g(peak) is similar at presynaptically paired synapses (P = 0.01). 8. The data confirm earlier findings that presynaptic and postsynaptic neurons specify aspects of plasticity with differing durations. Slowly varying factors determining both g(peak) and τ are specified by the postsynaptic neuron, globally affecting convergent inputs on that neuron. Individual fluctuations are partially introduced presynaptically. Consistent with neural network models, synaptic strengths between similar neurons differ widely, as do individual synapses between the same identified cells in different animals.