FUNCTIONAL AND STRUCTURAL IMAGING OF MULTIPLE INDIVIDUAL PYRAMIDAL NEURONS IN FREELY BEHAVING MICE OVER SEVERAL WEEKS

According to the classical view of neuronal information processing, synaptic inputs are received by the dendritic tree and propagate towards the soma from where the axon originates. Signals are then integrated at the axon initial segment where they may give rise to an action potential. Within this framework, participation of cortical neurons in network oscillations is determined by convergent excitation and rhythmic perisomatic inhibition, which provides gain control and a temporal scaffold for coordinated activity.

However, we previously showed that in a large fraction of hippocampal principal neurons (~50%), the axon originates from a basal dendrite rather than the soma. This non-canonical morphology profoundly alters intracellular signal processing and firing during network oscillations. Whether neurons with different axon origins contribute differently to behaviourally relevant processes such as learning remains unknown.

To address this question, we combined chronic functional and structural imaging of individual hippocampal pyramidal neurons during a prolonged learning task. We developed a custom holding plate enabling imaging of the same animal with two complementary systems: a UCLA miniscope for calcium imaging during free behaviour, and a two-photon microscope for in vivo morphological reconstruction with sub-micrometer precision. Using sparse expression of tdTomato and GCaMP7f, we simultaneously resolved morphology and activity in ~10% of pyramidal neurons in dorsal CA1 over multiple days.

During a five-day learning task in freely moving mice, we observed neurons with both stable and unstable spatial representations. This approach enables correlation of functional dynamics within local networks to detailed neuronal morphology, including axon initial segment location.