PROJECT SUMMARY The courtship song of male zebra finches is a classical model for learning complex motor behaviors and shows important parallels to human speech and communication. Male zebra finches learn a song from an adult tutor and then reproduce this song throughout adulthood. The zebra finch model offers outstanding behavioral control that permits the investigation of general principles of the circuit basis of vocal learning and motor control. Many highly interconnected brain regions across the zebra finch brain, ranging from forebrain to brainstem motor nuclei, form a song control system required for song learning and song production. A major bottleneck in understanding this system stems from unknown connectivity properties in the songbird brain and little knowledge of how these connectivity properties interface with transcriptomic and functional diversity of individual neurons to produce robust behavioral output. BRAIN Initiative supported tools have revolutionized research in the laboratory mouse by addressing this same bottleneck. However, these technologies are currently limited to a small set of model systems and not yet adapted to the zebra finch. We recently developed a novel class of flexible, high-throughput connectivity mapping tools for investigating neural circuit function. Our strategy relies on nucleic acid barcodes to translate neuronal connectivity into a format that can be read out by DNA sequencing and can bridge connectomics, transcriptomics, and functional data with single-cell resolution. Importantly, our tools are virus-based and can therefore be applied across species. To enable a multi-modal investigation of the circuit function of integrated cell types in the zebra finch song control system, and in particular the song nucleus HVC, in this capacity-building proposal we will adapt our next-generation connectomics and spatial transcriptomics tools to the zebra finch. In Aim 1, we bring the single-cell tracing method MAPseq and its combination with single-cell RNA sequencing to the zebra finch to discover, in an unbiased way, the multi-omic cell type identities in HVC. In Aim 2, we first optimize in situ sequencing method BARseq for zebra finches to map endogenous gene expression and barcodes in space. We then establish a pipeline to routinely register functional imaging data of HVC with BARseq images to interrogate the interaction of neuronal activity, gene expression, and long-range connectivity at the resolution of single cells for hundreds of cells per experiment. Successful completion of these aims paves the way for integrative TargetedBCP R01 projects conducted in collaboration between the Kebschull and Long labs. These projects will update existing models of song learning in the zebra finch by incorporating long-range connectivity and transcriptomic cell identity information that is critical for achieving a mechanistic understanding of circuit function. Our first target will test two competing hypotheses about how the connectomic and transcriptomic features of HVC neurons determine network properties that give rise to acoustic structure in the songbird brain.

NARRATIVE The zebra finch courtship song shows striking similarities to human speech, and the zebra finch system provides a well-controlled animal model with exceptional behavior to understand the neural circuit mechanisms of vocal learning and motor control. We will develop next generation connectomics and spatial transcriptomics methods to simultaneously measure connectivity, gene expression, and activity of single neurons in the zebra finch song control system during singing. The proposed research paves the way for future R01 applications investigating the role of diverse HVC neurons in producing acoustic structure in the brain and is relevant to NIH's mission by furthering our understanding of the neuronal mechanisms underlying skilled motor behaviors.