Dual intracellular recordings and computational models of slow inhibitory postsynaptic potentials in rat neocortical and hippocampal slices

Dual intracellular recordings in slices of adult rat neocortex and hippocam pus investigated slow, putative GABA(B) receptor-mediated inhibitory postsy naptic potentials. In most pairs tested in which the interneuron elicited a fast inhibitory postsynaptic potential in the pyramid, this GABA(A) recept or mediated inhibitory postsynaptic potential was entirely blocked by bicuc ulline or picrotoxin (3:3 in neocortex, 6:8 in CA1, all CA1 basket cells), even when high-frequency presynaptic spike trains were elicited. However, i n three of 85 neocortical paired recordings involving an interneuron, altho ugh no discernible response was elicited by single presynaptic interneurona l spikes, a long latency (greater than or equal to 20 ms) inhibitory postsy naptic potential was elicited by a train of greater than or equal to 3 spik es at frequencies greater than or equal to 50-100 Hz. This slow inhibitory postsynaptic potential was insensitive to bicuculline (one pair tested). In neocortex, slow inhibitory postsynaptic potential duration reached a maxim um of 200 ms even with prolonged presynaptic spike trains. In contrast, sum ming fast, GABA(A) inhibitory postsynaptic potentials, elicited by spike tr ains, lasted as long as the train. Between four and 10 presynaptic spikes, mean peak slow inhibitory postsynaptic potential amplitude increased sharpl y to 0.38, 2.6 and 2.9 mV, respectively, in the three neocortical pairs (me mbrane potential -60 to -65 mV). Thereafter increases in spike number had l ittle additional effect on amplitude. In two of eight pairs in CA1, one inv olving a presynaptic basket cell and the other a putative bistratified inte rneuron, the fast inhibitory postsynaptic potential was blocked by bicucull ine revealing a slow inhibitory postsynaptic potential that was greatly red uced by 100 mu M CGP 35348 (basket cell pair). The sensitivity of this slow inhibitory postsynaptic potential to spike number was similar to that of n eocortical 'pure' slow inhibitory postsynaptic potentials, but was of longe r duration, its plateau phase outlasting 200 ms spike trains and its maximu m duration exceeding 400 ms. Computational models of GABA release, diffusio n and uptake suggested that extracellular accumulation of GABA cannot alone account for the non-linear relationship between spike number and inhibitor y postsynaptic potential amplitude. However, cooperativity in the kinetics of GABA(B) transduction mechanisms provided non-linear relations similar to experimental data. Different kinetic models were considered for how G-prot eins activate K+ channels, including allosteric models. For all models, the best fit to experimental data was obtained with four G-protein binding sit es on the K+ channels, consistent with a tetrameric structure for the K+ ch annels associated with GABA(B) receptors. Thus some inhibitory connections in neocortex and hippocampus appear mediat ed solely by fast GABA(A) receptors, while others appear mediated solely by slow, non-ionotropic, possibly GABA(B) receptors. In addition, some inhibi tory postsynaptic potentials arising in proximal portions of CA1 pyramidal cells are mediated by both GABA(A) and GABA(B) receptors. Our data indicate that the GABA released by a single interneuron can saturate the GABA(B) re ceptor mechanism(s) accessible to it and that 'spillover' to extrasynaptic sites need not necessarily be proposed to explain these slow inhibitory pos tsynaptic potential properties. (C) 1999 IBRO. Published by Elsevier Scienc e Ltd.