Inhibitory synaptic plasticity: spike timing-dependence and putative network function.

While the plasticity of excitatory synaptic connections in the brain has been widely studied, the plasticity of inhibitory connections is much less understood. Here, we present recent experimental and theoretical findings concerning the rules of spike timing-dependent inhibitory plasticity and their putative network function. This is a summary of a workshop at the COSYNE conference 2012.

Functional relation between interneuron input and population activity in the rat hippocampal cornu ammonis 1 area.

Inhibitory interneurons are important components of the cornu ammonis 1 (CA1) network, as they are strategically positioned to control network information transfer. We investigated in detail synaptic input to individual CA1 interneurons (mainly basket and bistratified cells) after the local circuit was activated through the Schaffer-Commissural pathway and related this input to the population activity of the pyramidal cells. Synaptic responses were measured under whole-cell voltage clamp and population activity was determined from local field potentials. The synaptic input that was evoked in CA1 interneurons fell into two distinct groups. Disynaptic input with a long latency always started after the population spike with a mean latency of 3.0+/-0.3 ms (n=22) in respect to the peak of the population spike. This type of synaptic input to the interneurons was causally linked to the occurrence and amplitude of the population spike and most likely driven by CA1 pyramidal cells. Short-latency monosynaptic input occurred 0.8+/-0.2 ms (n=18) before the peak of the population spike. Its timing was strictly linked to the stimulus and showed significantly less jitter than long-latency input. In the absence of a population spike only short-latency input could be observed. Whether an interneuron receives direct monosynaptic Schaffer input or disynaptic input from the pyramidal cell population determines when that interneuron will be recruited in the network after Schaffer collateral stimulation. In addition, we found that the relation between the strength of the synaptic input and the population activity was different for the two types of input. Short-latency monosynaptic input showed large sensitivity to input changes at stimulus intensities that evoked little activity in the pyramidal cell population. In contrast, the amplitude of the long-latency disynaptic input to the interneurons closely reflected the population activity and increased gradually with stimulus intensity. Interneurons receiving the first type of input may expand the input sensitivity of the network, while interneurons receiving the second type could be involved in overall normalization of the output of the CA1 network. Our results underscore the importance of knowledge of the input to an interneuron for the understanding of its inhibitory role in the network.

Miniature inhibitory postsynaptic currents in CA1 pyramidal neurons after kindling epileptogenesis.

Miniature inhibitory postsynaptic currents (mIPSCs) were measured in CA1 pyramidal neurons from long-term kindled rats (>6 weeks after they reached the stage of generalized seizures) and compared with controls. A large reduction in the number of mIPSCs was observed in a special group of large mIPSCs (amplitude >75 pA). The frequency of mIPSCs in this group was reduced from 0.042 Hz in controls to 0.027 Hz in the kindled animals. The reduction in this group resulted in a highly significant difference in the amplitude distributions. A distinction was made between fast mIPSCs (rise time <2.8 ms) and slow mIPSCs. Fast mIPSCs, which could originate from synapses onto the soma and proximal dendrites, had significantly larger amplitudes than slow mIPSCs, which could originate from more distal synapses (35.4 +/- 1.1 vs. 26.2 +/- 0.4 pA in the kindled group; means +/- SE). The difference in the value of the mean of all amplitudes and frequency of fast and slow mIPSCs did not reach significance when the kindled group was compared with controls. The mIPSC kinetics were not different after kindling, from which we conclude that the receptor properties had not changed. Nonstationary noise analysis of the largest mIPSCs suggested that the single-channel conductance and the number of postsynaptic receptors was similar in the kindled and control groups. Our results suggest a 40-50% reduction in a small fraction of (peri-) somatic synapses with large or complex postsynaptic structure after kindling. This functionally relevant reduction may be related to previously observed loss of a specific class of interneurons. Our findings are consistent with a reduction in inhibitory drive in the CA1 area. Such a reduction could underlie the enhanced seizure susceptibility after kindling epileptogenesis.