UC Berkeley

The nervous system directly controls the muscles of the body, and thus, the behavior of the animal. An understanding of the neural circuits and cell types that mediate and control behavior would give us insight into the mechanisms by which the nervous system operates and would also contribute to the development of therapies and treatments for neurological disorders and diseases that result in locomotor deficits. Some behaviors are organized by constant voluntary drive, while others, which require regular and repeated patterns of activity, are organized by pattern generators. These central patterns generators (CPGs) include neural networks that generate the rhythmic patterns for basic behaviors like walking, swimming and breathing. Extensive research spanning decades in lamprey, Xenopus larvae, chick and mouse have provided great insight into the function and mechanisms of these circuits, but these models are poor for genetic studies or, in the case of the mouse, circuit complexity makes a cellular breakdown of the network challenging. As such, we know little about the specific cell types that are involved in the development and control of locomotor circuits. Using optogenetic tools, which allow for the manipulation and recording of neural activity with light in genetically-defined populations of cells, we investigated the cellular components that control the CPG in the spinal cord and the cells that mediate its proper formation during development. We applied these tools to the genetically-accessible and transparent zebrafish which permitted these studies in an intact and sometimes even behaving animal.

It was previously known that drivers exist in the brain that trigger the activation of the spinal CPG and induce behavior, but it was unknown whether similar such drivers exist in the spinal cord itself. To investigate spinal cell types that control the spinal CPG and mediate distinct behavioral programs, I, in collaboration with others, activated a variety of defined populations of spinal cord neurons and observed any consequent behavior. To activate neurons we targeted the light-gated ionotropic glutamate receptor LiGluR and targeted it to genetically-defined populations of cells in the zebrafish spinal cord using a variety of Gal4 driver lines. Through these experiments we were able to identify a specific cell type, the Kolmer-Agduhr cerebrospinal-contacting interneuron, which drives the CPG during forward swimming and whose function had been previously unknown. To investigate the development of the spinal CPG, I, in collaboration with others, used the genetically-encoded calcium indicator GCaMP to watch how patterns of neural activity evolve during development across the spinal cord and how this relates to the formation of a functional CPG. We found that early spontaneous neural activity progresses from sporadic to CPG-like rapidly in less than three hours, through the coalescence of local microcircuits. I next asked whether the early, uncorrelated activity is required for the acquisition of the later coordinated CPG-like activity. We applied the light-driven chloride pump Halorhodopsin to inhibit this early sporadic activity and found that activity is required for the proper integration of maturing neurons into the coordinated CPG network, which had not been known before. My work also included the application of the photoactivatable fluorophore Dronpa to the zebrafish and the identification of Gal4 driver lines that have targeted cell type specific expression advantageous for studies of the spinal CPG, both of which advanced our ability to perform effective optogenetic studies of the spinal CPG circuitry, function and development.