Quantal analysis of inhibitory synaptic transmission in the dentate gyrus of rat hippocampal slices: a patch‐clamp study.

Abstract

1. Synaptically connected neurones were identified in the granule cell layer of slices of 17- to 21-day-old rat hippocampus. Whole-cell current recording using the patch-clamp technique revealed synaptic currents ranging from less than 10 to 200 pA in symmetrical Cl- conditions, at a holding potential of -50 mV. These currents were blocked by 2 .mu.M-bicuculline, indicating that they result from the activation of postsynaptic .gamma.-aminobutyric acid receptor (GABAA-receptor) channels. 2. Addition of tetrodotoxin (TTX, 1 .mu.M) resulted in the loss of most currents of more than 40 pA in amplitude. Currents which disappeared after TTX treatment were assumed to be the result of spontaneous presynaptic action potentials. The currents seen in the absence of TTX are referred to as spontaneously occurring inhibitory postsynaptic currents (IPSCs); those remaining in the presence of TTX were defined as miniature IPSCs. 3. Similar currents were observed when recording in the whole-cell configuration while extracellular stimulation was applied to a nearby neurone. These currents were also completely blocked by 2 .mu.M-bicuculline and by 0.5 .mu.M-TTX. They were thus defined as stimulus-evoked IPSCs. 4. The half rise time of both miniature and stimulus-evoked IPSCs was fast (< 1 ms). The time course of decay of both miniature IPSCs and stimulus-evoked IPSCs could be well fitted with the sum of two exponentials. At a membrane potential of -50 mV, the mean decay time constants of the two components were 2.0 .+-. 0.38 and 54.4 .+-. 18 ms (mean .+-. S.D.) for miniature IPSCs (six cells) and 2.2 .+-. 1.3 and 66 .+-. 20 ms (three cells) for stimulus-evoked IPSCs. 5. Stimulus-evoked IPSCs varied in amplitude from less than ten to hundreds of picoamperes. In eight of eleven cells histograms of IPSC amplitudes showed several clear peaks which, when fitted with the sum of Gaussian curves, were found to be equidistant. This is consistent with the view that stimulus-evoked IPSC amplitudes vary in a quantal fashion. The quantal size varied between 7 and 20 pA, at a membrane potential of -50 mV. 6. Decreasing the Ca2+ and increasing the Mg2+ concentration in the extracellular solution decreased the number of peaks in the IPSC amplitude histogram but did not affect the size of the quantal event. 7. In one cell where the recording was stable for more than an hour, changing the membrane voltage from -50 to -120 mV increased the quantal size by a factor of 2.1, close to that expected if the current-voltage relation of IPSC peak amplitudes were linear. 8. The peak of miniature IPSC amplitude histograms measured in the presence of 1 .mu.M-TTX was comparable with the quantal size of stimulus-evoked IPSC. However, in all cells, a tail of larger amplitude miniature IPSCs was observed. The amplitudes in the tail in six of twelve cases were quantally distributed. 9. Single-channel currents activated by GABA, applied locally to outside-out patches isolated from the soma membrane of granule cells, indicated that GABAA-receptor channels had two conductance levels of 14 and 23 pS. Thus for IPSCs in hippocampal granule cells, one quantal current represents the simultaneous opening of less than thirty GABAA-receptor channels. 10. The small number of postsynaptic channels mediating a quantal IPSC, the small variation in quantal size and the fast rise of the IPSCs are consistent with an ''all-or-none'' hypothesis of synaptic transmission at granule cell synapses where the size of the quantal event is determined by the number of available postsynaptic GABAA-receptor channels opposite a release site.