PROPERTIES OF UNITARY IPSPS EVOKED BY ANATOMICALLY IDENTIFIED BASKET CELLS IN THE RAT HIPPOCAMPUS

Hippocampal pyramidal cells receive GABA-mediated synaptic input from several distinct interneurons. In order to define the effect of periso matic synapses, intracellular recordings were made with biocytin-conta ining microelectrodes from synaptically connected inhibitory and pyram idal cell pairs in subfields CA1 and CA3 of the rat hippocampus. Subse quent physiological analysis was restricted to the category of cells, here referred to as basket cells (n = 14), which had an efferent synap tic target profile (n = 282 synaptic contacts) of predominantly somati c (48.2%) and proximal dendritic synapses (45.0%). Electron microscopi c analysis revealed that in two instances identified postsynaptic pyra midal cells received a total of 10 and 12 labelled basket cell synapse s respectively. At an average membrane potential of -57.8 +/- 4.6 mV, unitary inhibitory postsynaptic potentials (IPSPs; n = 24) had a mean amplitude of 450 +/- 238 mu V, a 10-90% rise time of 4.6 +/- 3.2 ms an d, measured at half-amplitude, a mean duration of 31.6 +/- 18.2 ms. In most instances (n = 19) the IPSP decay could be fitted with a single exponential with a mean time constant of 32.4 +/- 18.0 ms. Unitary bas ket cell-evoked IPSPs fluctuated widely in amplitude, ranging from the level of detectability to > 2 mV. The response reversal of IPSPs (n = 5) was extrapolated to be at -74.9 +/- 6.0 mV. Averages of unitary IP SPs had a mean calculated conductance of 0.95 +/- 0.29 nS, ranging fro m 0.52 to 1.16 nS. Unitary basket cell IPSPs (n = 3) increased in ampl itude by 26.3 +/- 19.9% following bath application of the GABA(B) rece ptor antagonist CGP 35845A (1-4 mu M), whereas subsequent addition of the GABA(A) receptor antagonist bicuculline (10-13 mu M) reduced the I PSP amplitude to 13.5 +/- 3.1% of the control response. Rapid presynap tic trains of basket cell action potentials resulted in the summation of up to four postsynaptic responses (n = 5). However, any increase in the rate of tonic firing (2- to 10-fold) led to a > 50% reduction of the postsynaptic response amplitude. At depolarized membrane potential s, averaged IPSPs could be followed by a distinct depolarizing oversho ot or postinhibitory facilitation (n = 4). At firing threshold, pyrami dal cells fired postinhibitory rebound-like action potentials, the lat ter in close temporal overlap with the depolarizing overshoot. In conc lusion, hippocampal basket cells have been identified as one source of fast, GABA(A) receptor-evoked perisomatic inhibition. Unitary events are mediated by multiple synaptic release sites, thus providing an eff ective mechanism to avoid total transmission failures.