Nonsynaptic network rhythms as catalysts for epileptic seizures

Our brains are made up of many thousands of billions of cells called neurons. In order for our brains to function properly clusters of these neurons must communicate with each other to understand information coming from our senses, and to control out movements. Neurons communicate with each other by generating tiny electrical charges but, for the most part, these charges do not pass directly from one neuron to another. Instead, electrical activity passes down long, wire-like extensions of the neuron until it reaches a specialised part of the ?wire? ? the synapse - that sits right next to other neurons. Electrical activity reaching this synapse causes it to release small amounts of a chemical which changes the electrical properties of the target neuron. In using this chemical, non-direct, form of communication the brain sacrifices speed and fidelity of communication in favour of a very broad and exquisitely adaptable range of effects of one neuron on another.

However, another type of communication between neurons exists which DOES involve direct passage of electrical activity. This ?non-synaptic? communication involves neurons directly touching each other and sharing electric charge rapidly. This is a far more crude scheme than the elegant, chemical synaptic method of communication. It does not allow for a broad range of effects of one neuron on another and it is far less adaptable, but it does have the advantage of speed. It is present in humans as the nervous system is forming but was thought not to be important in our adult brains, but this idea is changing. Some forms of electrical activity in adult brains occur with neurons communicating at rates far too fast to be explained by chemical synapses. This type of activity appears critical for the normal function of our brains, but is usually ?drowned out? by the huge amount of chemical synaptic communication. This makes it very hard to study. However, for reasons we don?t yet understand, much larger amounts of fast communication between neurons is seen in the brains of patients suffering from epilepsy. In fact there appears to be a relationship between the amount of this fast communication and the underlying problems which generate the seizures themselves. This proposal intends to examine this relationship and take advantage of the large amount of fast communication present in epilepsy to understand how it occurs and to use it to aid better treatments for this debilitating neurological disorder.

The neuronal doctrine states that nerve cells are functionally separate entities, with communication between them afforded primarily by the chemical synapse. The flow of electrical information through networks of chemically synaptically connected neurons is considered predominantly orthodromic ? synaptic excitation and inhibition being integrated in neuronal dendrites, leading to generation of action potentials in perisomatic compartments which then propagate along axons to the next synaptically coupled target neuron. Consequences of such a pattern of activity include the dynamic assembly hypothesis which states that information is coded in the brain by the synchronous firing of specific subsets (assemblies) of neurons. However, there is increasing evidence to suggest that neurons are not as functionally separate as previously thought. Direct electrical and chemical connectivity via gap junctions allows many neurons, coupled in this way, to respond to the pattern of activation of just one. Similarly, the output of a single neuron can percolate through electrically connected axonal compartments in such a manner that the ?identity? of the original signal, in terms of its discrete neuronal origin, becomes more probabilistic than absolute. Electrophysiological markers for participation of a neuron in a non-synaptic network take the form of active and passive junctional potentials, and many such neurons appear to produce specific types of population dynamic activity which cannot be modelled using orthodromic, chemical synapses alone.

The most promising candidate for a purely non-synaptically connected network dynamic is the very fast oscillation (VFO) ? a field potential rhythm seen at frequencies from 80Hz to greater than 200Hz. In animal models this VFO survives blockade of synaptic activity, but is abolished by drugs which reduce gap junction conductance. It is seen transiently on receipt of sensory information in cortex, is nested within more conventional activity patterns such as alpha and gamma rhythms, and is a prominent component of physiological sharp waves. In each of these cases the magnitude of the VFO is very small. However, much larger, prolonged epochs of VFO are associated with seizure onset in focal epilepsies. This study intends to use this association with epilepsy both as a substrate to allow characterisation of underlying mechanisms of VFO generation in human cortex, and as a diagnostic marker to predict the precise location of seizure foci. We aim to show that nonsynaptic networks are functional entities in adult human cortex, and that quantifying their excessive activation can be used to guide surgical resection in epilepsy.