The research in this thesis focuses on mechanisms that underlie brain waves (also called oscillations). Brain activity is often rhythmical, and depending on what a person is doing, waves of different frequency occur. In this thesis we describe processes which underlie brain waves typically observed when a person is active. These waves, which are called fast network oscillations (13-80 waves per second), are thought to be involved in many higher brain functions such as attention consciousness and working memory. Although brain activity correlates with behavior there are large differences between individuals in the shape of brain waves. Twin studies showed that the variation between individuals is highly heritable. Yet it is unknown how individual differences relate to cognitive performance or intelligence. There are many different theories on how brain waves could be involved in higher brain functions, however direct evidence for these theories is still lacking. Therefore more insight in the cellular mechanism as well as how variations in genes influence brain waves might help understanding when and why these waves occur.

Brain waves only occur when large groups of brain cells (neurons) are active at the same time. To do this neurons make thousands of contacts with other brain cells at a synapse. The synapse is a connection between two different neurons that is responsible for transmission of activity. When the sending neuron is active it will release a substance called a neurotransmitter at the synaptic connection between two neurons, which in turn open ionic channels at the receiving neuron. Neurons that release the neurotransmitter glutamate activate channels that permeate positive ions that can stimulate activity in the receiving neuron, while neurons that release the neurotransmitter GABA activate channels that permeate negative ions that suppress activity in the receiving neuron. When many neurons are active at the same time, their summed can be recorded as voltage changes that lead to brain waves from the space between neurons. This can be done by either placing an electrode (to record local field potentials) in the tissue or by placing electrodes on the skull (EEG). The research in this thesis uses techniques to record synaptic activities that one neuron receives in combination with recordings of the field potential to monitor brain waves. This way it is possible to see whether the synaptic activity of that a particular neuron receives is contributing to the waves in the field.

To study the neuronal mechanism of fast network waves we use mouse brain slices as a model system. Normally when an animal is awake neurons are activated by a neurotransmitter called acetylcholine. In brain slices we use a drug that binds to the same receptor as acetylcholine and thereby it will activate cells that release glutamate. In this way other neurons are activated among which are neurons that release GABA. This release of GABA will activate currents that suppress the activity of many neurons at the same time. Only when the suppression of activity from GABA releasing neurons is decreased it is possible for neurons to become active again and to start a new cycle. This loop between activating and suppressing neurons is thought to generate fast network oscillations. The duration of the suppressing currents is hypothesized to determine which cells become active and when they become active.

This slice model for fast network oscillations we used in chapter 2 to test whether genetic differences between inbred mouse lines affect the local neuronal network responsible for the generation of these brain waves. The advantage of using inbred mouse lines is that all mice within a line are genetically identical (like monozygotic twins) making it possible to test whether properties of fast network oscillations are heritable. We indeed found that many properties of fast network oscillations vary between inbred mouse lines. In chapter 3 we tried to find the cellular mechanism that is responsible for differences in frequency between inbred mouse lines that we found in chapter 2. By recording suppressing currents activated by GABA we found that variation in the duration of these GABA activated currents correlated with the frequency of fast network oscillations indicating that genes that influence the GABA activated currents.

In the brain there are at least 12 different genes that encode units that form the receptor being responsible for the GABA activated currents. This GABA receptor consists of five building blocks called subunits, which are all encoded by different genes. With 12 different building blocks there are many different ways to assemble a five particle GABA receptor and some of these building blocks determine the function and location of the GABA receptor. As these GABA receptors are thought to be responsible for the shape of fast network oscillations we tried to find out in chapter 4 which subunits of the receptor are important for these oscillations. By using mice that lack the gene encoding a specific subunit of the GABA receptor and mice that have a mutation in a subunit that alters its modulation by a sleeping drug, we found that GABA receptors with a2 subunits are responsible for the frequency of fast network oscillations.

Chapter 5 describes research on how the GABA activated currents control the spread of rhythmic activity in the brain. We found that when the suppressing currents are enhanced by using sleeping drug the there is more build-up of activity during high frequency input. Taken all these findings together our research indicates that GABA mediated currents control the rhythm and spread of fast network oscillations.