Dendritic integration in mouse retinal ganglion cells

Abstract

Understanding how a neuron integrates the large number of synaptic inputs across its dendritic arbour is critical to understand neural computations. The central nervous system comprises a large variety of neuron types that differ in their morphology, physiology and functional role within the circuit. However, little is known about how cell-type-specific differences in dendritic integration arise from general features such as neuronal morphology and intrinsic membrane properties. Here, retinal ganglion cells, which relay the visual system’s first computations to the brain, represent an exquisite model. They are functionally and morphologically diverse yet defined, and they allow studying dendritic integration in a functionally relevant context. In this thesis, I systematically investigate the dendritic integration of visual information in four types of mouse retinal ganglion cells (transient Off alpha, transient Off mini, sustained Off, and F-miniOff), which receive similar excitatory inputs, but display different visual responses and dendritic morphologies. Using two-photon imaging of dendritic calcium signals from individual cells, and biophysical modelling, I demonstrate that these retinal ganglion cells exhibit diverse type-specific spatio-temporal dendritic integration profiles: In transient Off alpha cells, dendritic receptive fields displayed little spatial overlap, indicating a dendritic arbour that is partitioned in largely isolated regions. In contrast, dendritic receptive fields in the other three cell types overlapped greatly and were offset to the soma in transient Off mini and sustained Off cells, suggesting strong synchronization of dendritic signals likely due to back-propagation of somatic signals. Also, the temporal correlation of dendritic signals varied extensively among these types, with transient Off mini cells displaying the highest correlation across their dendritic arbour. Modelling suggests that morphology alone cannot explain these differences in dendritic integration, but instead specific combinations of dendritic morphology and ion channel densities are required. Together, these results reveal how neurons exhibit distinct dendritic integration profiles, tuned towards their type-specific computations in their circuits and highlight the interplay between morphology and channel complement as a key contributor in shaping dendritic integration.