The role of TASK potassium channels in theta oscillations and behaviour

One of the great challenges of neuroscience is determining how the working of our brains leads to thoughts, memories and actions. Some aspects of cognition, such as our emotional states and consciousness itself, are profoundly daunting problems. Understanding how these states arise is many years away. More defined behavioural questions, such as how we lay down memories or how we generate motor activity in response to a specific sensory input, while still very difficult problems, seem more tractable. One powerful concept that has emerged during the last few decades is that co-ordinated electrical oscillations in the brain reflect internal computations which may underlie specific behaviours. The best studied of these 'brain waves' occurs at frequencies of about 4-12 cycles per second and have been termed 'theta oscillations'. Theta oscillations are generated by a complex neuronal network which involves several brain regions, the best studied of which is the hippocampus. Theta oscillations in the hippocampus are hypothesized to be causally linked to several different behaviours. One important behaviour - planning to initiate movement in response to a meaningful sensory stimulus - for example, when an animal is confronted by a predator, is associated by a particular type of theta oscillation (so-called Type II or 'arousal' theta) which can be distinguished from other theta oscillations by its sensitivity to the drug atropine. The neuronal networks, cellular mechanisms and behavioural correlates of Type II theta are poorly understood, but a chance discovery that we have made promises new insights into these questions. A recently discovered type of potassium channel - the two-pore-domain potassium channel - critically influences the baseline excitability of neurons; several channel types are found in the brain, each with its own specific distribution. We have been investigating the TASK-1 and TASK-3 members of this family and have been studying behaviour and brain oscillations in genetically modified animals which lack either, or both, of these channels. We have made the unexpected discovery that removing the TASK-3 channel, but not the TASK-1 channel, completely ablates Type II theta while having no effect on the ability of the animal to generate other theta oscillations. Not only will this discovery allow us to investigate how the specific ablation of Type II theta affects behaviour, it will also give us a powerful new way of determining which neuronal networks generate these oscillations. TASK-3 channels are particularly highly expressed in certain types of hippocampal neuron. One of these neuronal types is postulated to be especially important for synchronising theta oscillations and a subset of these cells, so-called parv-positive GABAergic interneurons, contain high levels of TASK-3 channels. In this proposal we plan to explore the hypothesis that those parv-positive interneurons which contain TASK-3 channels are necessary for sustaining Type II theta oscillations. We will investigate behaviours thought to be linked to Type II theta in both wild-type mice and mice lacking TASK-3 channels, and we will compare the membrane properties of hippocampal neurons from wild-type and TASK-3 knock-out animals. Using an electrophysiological technique often referred to as dynamic current-clamp, we will 'restore' TASK-3 channels into neurons recorded from knock-out animals to see if we can re-establish wild-type electrical properties. Using in vivo genetics we will selectively remove TASK-3 channels from parv-positive interneurons to see if this is sufficient to ablate Type II theta. Finally, we will selectively restore TASK-3 channels to their parv-positive interneurons to see if Type II theta and the associated animal behaviours can be restored.

As animals explore their environments, as they lay down memories, as they process sensory input, and as they sleep, characteristic electrical oscillations appear in the hippocampus. A prominent and well-studied class of oscillations occurs at 'theta' frequencies (4-12 Hz): Type I theta predominates during exploratory behaviour, and Type II theta (also known as atropine-sensitive theta) is associated with immobility during the processing of sensory stimuli relevant to initiating, or intending to initiate, motor activity. We have made the surprising discovery that animals lacking the TASK-3 potassium channel display normal Type I theta oscillations, whereas Type II oscillations are completely absent. Animals lacking the TASK-1 potassium channel display normal theta oscillations. Within the CA1 hippocampal region, TASK-3 expression is particularly strong in subsets of parvalbumin-positive GABAergic interneurons, cells which are critical in the production of theta oscillations. Our hypothesis is that TASK-3 channel expression in parv-positive interneurons is essential for maintaining Type II theta oscillations and the associated whole animal behaviours. We will test this hypothesis by using novel genetically-engineered mice in which TASK-3 is selectively ablated from parv-interneurons, and mice in which TASK-3 is present only in these neurons but absent elsewhere. We will study the behaviours of these site-specific knock-out and knock-in animals, as well as the total knock-out animals using several assays, including a paradigm which characterises an animal's response when confronted by a predator (where type II theta oscillations are thought to be instrumental in the behavioural response). We will also investigate the properties of hippocampal neurons in these animals using slice electrophysiology and dynamic current-clamp to account for the differences in animal behaviour in terms of changes in neuronal excitability produced by removing a single class of ion channel.