Cortical parvalbumin-expressing interneurons sample network oscillations in their synaptic activity

Abstract

Synaptic activity is thought to be the primary input of the frequency bands conveyed in the electroencephalogram (EEG) and local field potentials (LFPs) recorded on the cortex. Here we ask whether synaptic activity observed in parvalbumin expressing (PV + ) neurons recorded in isolated cortical tissue bear these frequency bands. The muscarinic agonist carbachol (CCh) was used to increase cortical excitability. PV + neurons play a significant role in perisomatic inhibition and the synchronization of cortical ensembles to generate gamma (γ) oscillations during cholinergic modulation. γ-oscillations associate with cognitive activities co-existing with slower rhythms. While CCh induces depolarization and firing in pyramidal neurons, it only causes barrages of synaptic potentials without firing in most PV + neurons. We show that the frequency spectra of CCh-induced synaptic events recorded onto layer 5 PV + neurons display the various frequency bands generated by cortical networks: from δ to γ. Isolation of inhibitory events shows potency increases in the δ band and decreases in other bands. Isolated excitatory events exhibit a decrease in the β-band. Excitatory potentials appear to drive the circuitry while inhibitory ones appear to regulate events frequency. Muscarinic M₁-class receptors are mainly responsible for the synaptic activity from which oscillatory bands emerge. These results demonstrate that PV + interneurons “sample” network activity through the ligand-gated synaptic events that receive from it. We conclude that random synaptic events recorded in single neurons contain the wide range of brain oscillations as revealed by frequency spectra and power density analyses.