Hippocampale Netzwerke für assoziatives Lernen und Gedächtnis

The hippocampus is an important brain area for associative learning and memory. In this project we aim to determine how distinct types of GABAergic interneuron and pyramidal cells of the hippocampus operate in time to generate and retrieve such associations. For this we will train head-fixed mice to navigate in a virtual environment by running on an air-lifted jetball. During task performance we will record neuronal activity in the distal CA1 hippocampus, which receives olfactory information from the lateral entorhinal cortex. The mice will run along a virtual Y maze and in the central arm they may receive an odour stimulus. Depending on the presence or absence of this stimulus the animals have to turn to the right or left arm of the Y maze, respectively, to receive a reward. During this task the animals learn to associate the odour with a reward location. While the mice perform the association task, we will record the activity of identified neurons with glass electrodes and neuronal ensembles and network oscillations with an additional silicon probe. The neuron recorded with the glass electrode will be filled with neurobiotin using the juxtacellular labelling technique and post-hoc histological analysis will determine the cell type of the recorded neuron. With these experiments we will test the hypothesis that distinct types of GABAergic interneuron contribute differentially to the network operations generating associations in hippocampal circuits. Furthermore we will test whether distinct types of pyramidal cells with axons projecting to different target brain areas contribute differentially to network activity during task performance. In addition, we will detect different gamma oscillators within the CA1 hippocampus and we will determine how these gamma oscillators change dynamically during task performance. In particular, we will test the hypothesis that gamma oscillations mediated by input from the entorhinal cortex become synchronised with perisomatic gamma oscillations mediated by parvalbumin-expressing basket cells in the CA1 hippocampus during successful task performance. We will determine how distinct types of GABAergic interneuron contribute to the different gamma oscillations and to synchronisation between different oscillators and between different brain areas. We will also test how the observed activity patterns change and evolve during different stages of learning and how these activity patterns are influenced by errors in the task performance. In international collaborations, we will provide our data for building a computational model of the learning process and for comparisons with a mouse model of Alzheimers disease. Overall, these experiments will determine how distinct types of neuron in the hippocampus interact in time in order to generate associations between independent sets of information.

The hippocampus is an important brain area for associative learning and memory. In this project we aimed to determine how distinct types of GABAergic interneuron and pyramidal cells of the hippocampus operate in time to generate and retrieve such associations. For this, we trained head-fixed mice to navigate in a virtual environment by running on an air-lifted jetball. During task performance we recorded neuronal activity in the distal CA1 hippocampus, a brain area important for learning, which also receives olfactory information from the lateral entorhinal cortex. In this task, the mice ran along a virtual maze and at some point they could receive an odour stimulus. Depending on the presence or absence of this stimulus the animals had to stop running earlier or later, respectively, to receive a reward. During this task the animals learned to associate the odour with a reward location. While the mice performed the association task, we recorded the activity of identified neurons and neuronal ensembles and network oscillations. In some experiments, the neuron recorded could be filled with neurobiotin using the juxtacellular labelling technique and post-hoc histological analysis determined the cell type of the recorded neuron. With these experiments we tested the hypothesis that distinct types of GABAergic interneuron contribute differentially to the network operations generating associations in hippocampal circuits. Furthermore we also tested how the observed activity patterns change and evolve during different stages of learning and how these activity patterns are influenced by errors in the task performance. We observed that some neurons in the hippocampus changed their activity in relation to the presence of the odour stimulus to inform the animal on consecutive behavioural actions. Other neurons adjusted their activity in order to maintain this information in their working memory until the appropriate behavioural action has to be taken. A third subset of neurons indicated the successful completion of the trial and the reception of reward, which may support the positive reinforcement of the learned association. Interestingly, we observed that these firing patterns developed as the animals were learning the association and indicated the level of competence in this task. In contrast, a change of rule during the task, not only induced a change in strategy of the animal, but also caused a major readjustment of the firing patterns of the hippocampal neurons. Overall, the experiments performed during this project showed how distinct types of neuron in the hippocampus interact in time in order to generate associations between independent sets of information.