Heterogeneous populations of cells mediate spontaneous synchronous bursting in the developing hippocampus through a frequency-dependent mechanism

Abstract

Under normal conditions, hippocampal slices from newborn rats and rabbits (postnatal days 0–8) show spontaneous synchronous bursts known as giant depolarizing potentials. These bursts are recorded from CA3, CA1 and the fascia dentata in both intact slices and isolated hipocampal regions. Giant depolarizing potentials are network-driven events resulting from the synergistic activation of N-methyl- d-aspartate, α-amino-3-hydroxy-5-methyl-4-isoxadepropionate and GABA A receptors, the latter playing an excitatory role. Recently, we showed that they spontaneously emerge in an all-or-none manner after the increase of synaptic and cellular activity beyond a threshold frequency [Menendez de la Prida L. and Sanchez-Andres J. V. (1999) J. Neurophysiol. 82, 202–208]. Under this framework, background levels of spontaneous activity at individual neurons build up network synchronization 100–300 ms prior to the onset of giant depolarizing potentials. However, the role of distinct cellular populations and connectivity in determining the threshold frequency has not been examined. By performing simultaneous intracellular recordings from pyramidal cells, non-pyramidal cells and interneurons, we investigated their participation in the generation of giant depolarizing potentials. Electrodes containing Neurobiotin were used to examine the cellular morphology. We found that giant depolarizing potentials were not initiated from a single pacemaker cellular group; instead, they involved recurrent cooperation among these groups, which contributed differently according to their intrinsic firing capability. In all the neurons examined, the onset of these bursts took place in an all-or-none frequency-dependent manner, both spontaneously (depending on the frequency of the excitatory postsynaptic potentials) or when triggered by extracellular stimulation. The CA3 threshold of frequency was at 12 Hz in both pyramidal cells and interneurons, while in the fascia dentata it was 17 Hz. The application of 6-cyano-7-nitroquinoxaline-2,3-dione increased CA3 threshold of frequency up to 50 Hz, suggesting that it is determined by combined synaptic components. We examined the role of postsynaptic summation on the threshold of frequency. Heterogeneity is present among the cellular groups, pyramidal neurons from CA1 and CA3 showing less evidence of postsynaptic summation prior to giant depolarizing potentials. Cells showing stronger evidence of postsynaptic summation were more typically recorded at the hilus, the granule layer of the fascia dentata and the CA3/CA4 area. Nevertheless, for a given cell, not all the giant depolarizing potentials were preceded by summation of postsynaptic potentials. These outcomes, together with the long and variable time delays recorded between different areas, strongly suggest that giant depolarizing potentials are locally generated from different initiation sites and not from a single region. We discuss these results in view of the principles underlying hyperexcitability in hippocampal slices, i.e. the intrinsic firing properties of individual cells and the connectivity patterns.