Connectomic analyses of songbird Area X

Connectomics is a field dedicated to the high-resolution analysis of neural circuits. Making use of nanometer scale resolution provided by volume electron microscopy (EM), analysis allows information on the level of individual synapses and other subcellular structures. This thesis presents a comprehensive analysis of a connectomic dataset from the songbird brain, specifically Area X—a nucleus involved in song learning and part of the avian basal ganglia. The work includes improvements to the processing pipeline, morphological characterization of neurons and glia, and detailed analyses of synaptic connectivity and subcellular organelles. To facilitate this analysis, the existing EM data processing pipeline was extended to include classification of glial cells, migratory neurons, and three newly described GABAergic interneurons. Additionally, to enable analysis of further subcellular structures, the endoplasmic reticulum (ER), the Golgi apparatus (GA), and individual synaptic vesicles were predicted. This culminated in the first dataset that, alongside dense reconstructions of over 8,500 neurons, includes densely segmented synapses, mitochondria, ER, GA, and individual vesicles. Mitochondria, vesicles and axonal ER densities were found to correlate with known firing rates and help to bridge information between EM and functional methods. The dataset provides the first connectomic overview of basal ganglia circuitry in any species. Analyses revealed analogues of the direct, indirect, and hyperdirect pathways. In terms of synaptic connectivity, the direct pathway is the strongest pathway through Area X, while the hyperdirect pathway appears to be its strongest antagonist. The indirect pathway is poorly segregated from the direct pathway and also weak in terms of synaptic connectivity. Since one cell type of the indirect pathway, the GPe, is more involved in a feedback loop with novel GABAergic interneurons, the results challenge the idea of basal ganglia connectivity being mainly in the form of feedforward pathways. As not all cell types transmit information via synaptic transmission, individual vesicles were used to further analyze the possibility of volume transmission. The presence of vesicles outside synaptic loci was common in cholinergic and dopaminergic axons, consistent with their reported use of volume transmission and their surroundings were in line with a global signal. Finally, dense reconstructions of glial cells and migratory neurons allowed for organelle-level comparisons, providing insight into their metabolic demands. To highlight their role in neuronal plasticity, contact sites with neurons were analyzed as a first proxy for neuron-glia interactions. In summary, this thesis presents the first dense connectomic analysis of an avian basal ganglia nucleus, integrating synaptic, morphological, and subcellular data. These findings not only provide new insights into the structure and function of Area X and into connectivity within the vertebrate basal ganglia, but also demonstrate how extending connectomic pipelines to include organelles and non-neuronal cells can unlock richer biological interpretations from EM datasets.