Partially synchronous 40-Hz oscillations of cortical neurons have been implicated in cognitive function. Specifically, coherence of these oscillations between different parts of the cortex may provide conjunctive properties to solve the 'binding problem': associating features detected by the cortex into unified perceived objects. Here we report an emergent 40-Hz oscillation in networks of inhibitory neurons connected by synapses using GABAA (gamma-aminobutyric acid) receptors in slices of rat hippocampus and neocortex. These network inhibitory postsynaptic potential oscillations occur in response to the activation of metabotropic glutamate receptors. The oscillations can entrain pyramidal cell discharges. The oscillation frequency is determined both by the net excitation of interneurons and by the kinetics of the inhibitory postsynaptic potentials between them. We propose that interneuron network oscillations, in conjunction with intrinsic membrane resonances and long-loop (such as thalamocortical) interactions, contribute to 40-Hz rhythms in vivo.

An increasingly large body of data exists which demonstrates that oscillations of frequency 12᎐80 Hz are a consequence of, or are inextricably linked to, the behaviour of inhibitory interneurons in the central nervous system. Ž . Ž . Ž . This frequency range covers the EEG bands beta 1 12᎐20 Hz , beta 2 20᎐30 Hz and gamma 30᎐80 Hz . The pharmacological profile of both spontaneous and sensory-evoked EEG potentials reveals a very strong influence on Ž these rhythms by drugs which have direct effects on GABA receptor-mediated synaptic transmission general A .Ž . anaesthetics, sedativerhypnotics or indirect effects on inhibitory neuronal function opiates, ketamine . In addition, a number of experimental models of, in particular, gamma-frequency oscillations, have revealed both common denominators for oscillation generation and function, and subtle differences in network dynamics between the different frequency ranges. Powerful computer and mathematical modelling techniques based around both clinical and experimental observations have recently provided invaluable insight into the behaviour of large networks of interconnected neurons. In particular, the mechanistic profile of oscillations generated as an emergent property of such networks, and the mathematical derivation of this complex phenomenon have much to contribute to our understanding of how and why neurons oscillate. This review will provide the reader with a brief outline of the basic properties of inhibition-based oscillations in the CNS by combining research from laboratory models, large-scale neuronal network simulations, and mathematical analysis. ᮊ 2000 Elsevier Science B.V. All rights reserved.

Experimental and modeling efforts suggest that rhythms in the CA1 region of the hippocampus that are in the beta range (12-29 Hz) have a different dynamical structure than that of gamma (30 -70 Hz). We use a simplified model to show that the different rhythms employ different dynamical mechanisms to synchronize, based on different ionic currents. The beta frequency is able to synchronize over long conduction delays (corresponding to signals traveling a significant distance in the brain) that apparently cannot be tolerated by gamma rhythms. The synchronization properties are consistent with data suggesting that gamma rhythms are used for relatively local computations whereas beta rhythms are used for higher level interactions involving more distant structures. Rhythms in the gamma range (30-80 Hz) and the beta range (12-30 Hz) are found in many parts of the nervous system and are associated with attention, perception, and cognition (1-3). It has been noted in electroencephalogram (EEG) signals that rhythms of different frequencies are found simultaneously (4). Beta oscillations are readily observable immediately after evoked gamma oscillations in sensory evoked potential recordings (5). This beta activity has been correlated with the long-range synchronous activity of neocortical regions during visuomotor reflex activation (6).This paper concerns the correlation between the frequency band of coherent oscillations and conduction delays between the sites participating in the coherent rhythm. It has been noted (7) in human EEG subjects that gamma rhythms are prevalent in local visual response synchronization, but more distant coherence occurring during multimodal integration between parietal and temporal cortices uses rhythms at frequencies of 12-20Hz (the so-called beta 1 range).We shall use data from the CA1 region of the hippocampus (8-10) as a paradigm to address the questions of how long-distance synchrony is achieved and why there is a correlation between oscillation frequency and the temporal distances between participating sites. The data available from the rat hippocampus slice preparation give clues about details of dynamics that are important to the synchronization process.The work builds on earlier work (11-12) describing and analyzing the role of doublet spikes in interneurons in producing synchrony when there are significant conduction delays. Earlier work (13) using rate models showed, via simulations, that longer conduction delays could be tolerated and still produce synchrony if the carrier rhythm had lower frequencies. However, a rate model is not consistent with the situation in which excitatory cells fire at most one spike per cycle, and with high precision in phase. An alternative solution was suggested by data and large-scale models of the gamma rhythm in the hippocampus (8, 9). In both data and models, the ability to synchronize happened in those parameter regimes in which interneurons produced a spike doublet in many of the cycles. This mechanism was analyzed by Ermentrout and Kopell (11), wh...

Coherent oscillations, in which ensembles of neurons fire in a repeated and synchronous manner, are thought to be important in higher brain functions. In the hippocampus, these discharges are categorized according to their frequency as theta (4-10Hz), gamma (20-80 Hz) and high-frequency (approximately 200 Hz) discharges, and they occur in relation to different behavioural states. The synaptic bases of theta and gamma rhythms have been extensively studied but the cellular bases for high-frequency oscillations are not understood. Here we report that high-frequency network oscillations are present in rat brain slices in vitro, occurring as a brief series of repetitive population spikes at 150-200 Hz in all hippocampal principal cell layers. Moreover, this synchronous activity is not mediated through the more commonly studied modes of chemical synaptic transmission, but is in fact a result of direct electrotonic coupling of neurons, most likely through gap-junctional connections. Thus high-frequency oscillations synchronize the activity of electrically coupled subsets of principal neurons within the well-documented synaptic network of the hippocampus.

1. We have shown previously, with experimental and computer models, how a '40 Hz' (gamma) oscillation can arise in networks of hippocampal interneurones, involving mutual GABAA-mediated synaptic inhibition and a source of tonic excitatory input. Here, we explore implications of this model for some hippocampal network phenomena in the rat in vitro and in vivo. 2. A model network was constructed of 1024 CA3 pyramidal cells and 256 interneurones.AMPA (a-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid), NMDA (N-methyl-D-aspartate), GABAA and GABAB receptors were simulated on pyramidal cells and on interneurones. 3. In both model and experiment, the frequency of network oscillations, in the gamma range, depended upon three parameters: GABAA conductance and decay time constant in interneurone-*interneurone connections, and the driving current to the interneurones. 4. The model of gamma rhythm predicts an average zero phase lag between firing of pyramidal cells and interneurones, as observed in the rat hippocampus in vivo. The model also reproduces a gamma rhythm whose frequency changes with time, at theta frequency (about 5 Hz). This occurs when there is 5 Hz modulation of a tonic signal to chandelier and basket cells. 5. Synchronized bursts can be produced in the model by several means, including partial blockade of GABAA receptors or of AMPA receptors on interneurones, or by augmenting AMPA-mediated EPSCs. In all of these cases, the burst can be followed by a 'tail' of transiently occurring gamma waves, a phenomenon observed in the hippocampus in vivo following sharp waves. This tail occurs in the model because of delayed excitation of the interneurones by the synchronized burst. A tail of gamma activity was found after synchronized epileptiform bursts both in the hippocampal slice (CA3 region) and in vivo. 6. Our data suggest that gamma-frequency EEG activity arises in the hippocampus when pools of interneurones receive a tonic or slowly varying excitation. The frequency of the oscillation depends upon the strength of this excitation and on the parameters regulating the inhibitory coupling between the interneurones. The interneurone network output is then imposed upon pyramidal neurones in the form of rhythmic synchronized IPSPs.EEG rhythms at frequencies of 30-100 Hz are prominent possible significance for information processing (for review,

Synchronous neuronal oscillations in the 30-70 Hz range, known as gamma oscillations, occur in the cortex of many species. This synchronization can occur over large distances, and in some cases over multiple cortical areas and in both hemispheres; it has been proposed to underlie the binding of several features into a single perceptual entity. The mechanism by which coherent oscillations are generated remains unclear, because they often show zero or near-zero phase lags over long distances, whereas much greater phase lags would be expected from the slow speed of axonal conduction. We have previously shown that interneuron networks alone can generate gamma oscillations; here we propose a simple model to explain how an interconnected chain of such networks can generate coherent oscillations. The model incorporates known properties of excitatory pyramidal cells and inhibitory interneurons; it predicts that when excitation of interneurons reaches a level sufficient to induce pairs of spikes in rapid succession (spike doublets), the network will generate gamma oscillations that are synchronized on a millisecond time-scale from one end of the chain to the other. We show that in rat hippocampal slices interneurons do indeed fire spike doublets under conditions in which gamma oscillations are synchronized over several millimetres, whereas they fire single spikes under other conditions. Thus, known properties of neurons and local synaptic circuits can account for tightly synchronized oscillations in large neuronal ensembles.

Perisomatic inhibition provided by a subgroup of GABAergic interneurons plays a critical role in timing the output of pyramidal cells. To test their contribution at the network and the behavioral level, we generated genetically modified mice in which the excitatory drive was selectively reduced either by the knockout of the GluR-D or by conditional ablation of the GluR-A subunit in parvalbumin-positive cells. Comparable cell type-specific reductions of AMPA-mediated currents were obtained. Kainate-induced gamma oscillations exhibited reduced power in hippocampal slices from GluR-D-/- and GluR-A(PVCre-/-) mice. Experimental and modeling data indicated that this alteration could be accounted for by imprecise spike timing of fast-spiking cells (FS) caused by smaller interneuronal EPSPs. GluR-D-/- and GluR-A(PVCre-/-) mice exhibited similar impairments in hippocampus-dependent tasks. These findings directly show the effects of insufficient recruitment of fast-spiking cells at the network and behavioral level and demonstrate the role of this subpopulation for working and episodic-like memory.

The questions of how a large population of neurons in the brain functions, how synchronized firing of neurons is achieved, and what factors regulate how many and which neurons fire under different conditions form the central theme of this book. Using a combined experimental-theoretical approach unique in neuroscience, the authors present important techniques for the physiological reconstruction of a large biological neuronal network. They begin by discussing experimental studies of the CA3 hippocampal region in vitro, focusing on single-cell and synaptic electrophysiology, particularly the effects a single neuron exerts on its neighbours. This is followed by a description of a computer model of the system, first for individual cells then for the entire detailed network, and the model is compared with experiments under a variety of conditions. The results shed significant light into the mechanisms of epilepsy, electroencephalograms, and biological oscillations and provide an excellent test case for theories of neural networks. Researchers in neurophysiology and physiological psychology, physicians concerned with epilepsy and related disorders, and researchers in computational neuroscience will find this book an invaluable resource.