Comparison between spontaneous and kainate-induced gamma oscillations in the mouse hippocampus in vitro

Neuronal synchronization at gamma frequency, implicated in cognition, can be evoked in hippocampal slices by pharmacological activation. We characterized spontaneous small-amplitude gamma oscillations (S gamma O) recorded in area CA3 of mouse hippocampal slices and compared it with kainate-induced gamma oscillations (K gamma O). S gamma O had a lower peak frequency, a more sinusoidal waveform and was spatially less coherent than K gamma O, irrespective of oscillation amplitude. CA3a had the smallest oscillation power, phase-led CA3c by similar to 4 ms and had the highest S gamma O frequency in isolated subslices. During S gamma O CA3c neurons fired at the rebound of inhibitory postsynaptic potentials (IPSPs) that were associated with a current source in stratum lucidum, whereas CA3a neurons often fired from spikelets, 3-4 ms earlier in the cycle, and had smaller IPSPs. Kainate induced faster/larger IPSPs that were associated with an earlier current source in stratum pyramidale. S gamma O and K gamma O power were dependent on alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors, gap junctions and gamma-aminobutyric acid (GABA)(A) receptors. S gamma O was suppressed by elevating extracellular KCl, blocking N-methyl-d-aspartate (NMDA) receptors or muscarinic receptors, or activating GluR5-containing kainate receptors. S gamma O was not affected by blocking metabotropic glutamate receptors or hyperpolarization-activated currents. The adenosine A(1) receptor antagonist 8-cyclopentyl-1,3-dimethoxyxanthine (8-CPT) and the CB1 cannabinoid receptor antagonist N-(piperidin-1-yl)-5-(4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide (AM251) increased S gamma O power, indicating that endogenous adenosine and/or endocannabinoids suppress or prevent S gamma O in vitro. S gamma O emerges from a similar basic network as K gamma O, but differs in involvement of somatically projecting interneurons and pharmacological modulation profile. These observations advocate the use of S gamma O as a natural model for hippocampal gamma oscillations, particularly during less activated behavioural states.