Professor John Jefferys

Department of Neurophysiology, Division of Neuroscience

Brain Waves ("200Hz") and gap junctions

Introduction

Brain waves, or the "EEG", are electrical signals that can be recorded from the brain, either directly or through the scalp. The kind of brain wave recorded depends on the behaviour of the animal, and is the visible evidence of the kind of neuronal (brain cell) processing necessary for that behaviour.

Gyorgy Buzsaki showed that fast ripples at up to 200 Hz occur, superimposed on "sharp waves" in the hippocampus in rats in vivo.

We have recently found that 100-200 Hz oscillations also occur in the hippocampal slice in vitro:-

Draguhn, A., Traub, R.D., Schmitz, D. and Jefferys, J.G.R., Electrical coupling underlies high-frequency oscillations in the hippocampus in vitro, Nature, 394 (1998) 189-192.

1. High frequency oscillations recorded from CA1 pyramidal layer in a hippocampal slice.

Note 200Hz peak in power spectrum (C) and 5 ms period in autocorrelation

High frequency oscillations are synchronised over 120 microns, but not over 220 microns.

(Synchronization is much longer range when we simulate sharp waves using acute convulsant treaments.)

HF oscillations do not depend on synaptic transmission. bathing the slice in zero calcium solution blocked the dentate evoked response (A), confirming that this rtreatment blocked synaptic transmission, but HF oscillations occurred more frequently than before (B), presumably due to the increased excitability causes by loss of charge screening of the membrane by calcium ions. The HF ripples also survived blockade of synaptic transmission by glutamate and GABA antagonists (not shown).

Gap junctions are necessary for HF oscillations. Several blockers of gap junctions, halothane (shown above), carbenoxolone, octanol, amongst others all blocked the HF oscillations. intracellular alkalinization, which opens gap junctions, enhanced HF oscillations. While all these agents have effects in addition to those on gap junctions, the only action they have in common is that on gap junctions, so their consistent effects on HF oscillations can be attributed to gap junctions. The data illustrate are from slices in the absence other drugs; similar results were found when synaptic transmission was blocked.

Intracellular recordings from a CA! pyramidal cell during HF oscillations. Note that this pyramidal cell fires in tight synchrony with the field potential ripple, but that it does not fire on every cycle of the oscillations and occasionally fires independently of the oscillations, all of which suggests that the ripples are indeed produced by the interaction of groups of neurons. Given that no interneurons recorded were synchronised with the HF oscillations we conclude that they are an emergent property of the pyramidal cells.

Computer simulations of pairs of neurons coupled by gap junctions between their axons (left) and their proxomal dendrites (right). Coupling through the axons is required to produce partial spikes as those seen in the experimental records. The physical idea is that the axons have a low threshold and high input resistance so that a relatively small number of gap junctions (hence high coupling resistance) can initiate a action potential in the postjunctional membranem and it is this action potential that is recorded at the soma rather than the "coupling potential" directly produced by the gap junction. Coupling at the dendrites or even the soma leads to inrealistic, slow potentials.

Group members and their research

John Jefferys

Simon Gladwell

Collaborators

Roger Traub - IBM Research Division, Yorktown Heights, NY

Andreas Draguhn & Deitmar Schmitz - Humbolt University, Berlin

Recent Publications

Draguhn, A., Traub, R.D., Schmitz, D. and Jefferys, J.G.R., Electrical coupling underlies high-frequency oscillations in the hippocampus in vitro, Nature, 394 (1998) 189-192.

We would like to hear from you if you have any comments on the work we do.