Axon degeneration plays a critical but ill-defined role in Parkinson disease (PD), and is actively regulated by pathways that are not well understood. Mitochondria orchestrate many of the processes that underlie axonal homeostasis, and mitochondrial dysfunction is implicated in PD pathogenesis. A better understanding of axonal mitochondria may therefore identify the cell biological processes that mediate degeneration and protection of the axon. Recent studies have suggested that mitochondrial transport is critical to axon protection after injury. I studied mitochondrial morphology, transport, and function in living zebrafish embryos. Using time-lapse confocal imaging of peripheral sensory neurons, I investigated the effect of injury on mitochondrial transport. I used two-photon laser axotomy to induce Wallerian degeneration (WD) of the distal axon. I found that acute transport arrest occurs in the distal axon, and motility does not recover before fragmentation. Although transport arrest was preserved in the proximal axon, which does not degenerate, increased mitochondrial trafficking after injury did not always correlate with axon protection. To determine whether mitochondrial ROS production is relevant to WD, I expressed the redox-sensitive biosensor roGFP2 in the mitochondrial matrix. After injury, roGFP2 was rapidly and persistently oxidized in the distal, but not the proximal, axon. The axon-protective protein WldS, which had only a mild and temporary effect on transport arrest after axotomy, robustly inhibited roGFP2 oxidation and degeneration. To further investigate the importance of mitochondrial ROS production after injury, I expressed the transcriptional co-activator PGC-1alpha, which has roles in mitochondrial biogenesis and ROS detoxification. I saw that overexpression of this protein delays roGFP2 oxidation after injury, and delays WD. Mitochondrial ROS production is therefore a better predictor of axonal vulnerability than mitochondrial transport, and ROS detoxification may be a relevant therapeutic target to prevent axon degeneration.

I then studied degeneration induced by alpha-synuclein (aSyn), a protein that is associated with PD. I expressed human wild-type aSyn in peripheral sensory neurons and saw that axon pathology precedes cell death in these cells. Early changes in mitochondrial morphology were consistent with increased fragmentation, and axonal varicosities were filled with swollen mitochondria. I also saw reduced mitochondrial trafficking in aSyn-expressing cells. Motile mitochondria favored retrograde transport towards the cell body. I hypothesized that protection of mitochondria might prevent degeneration in this model, as it had after axotomy. I therefore expressed PGC-1alpha in cells expressing aSyn, and saw robust protection against axon degeneration and cell death. These results suggest that axon degeneration pathways converge on mitochondrial dysfunction. Protection of mitochondria may therefore be a promising therapeutic target in the prevention and treatment of neurodegenerative disease.

Synapses link neuronal populations into circuits, which in turn regulate behavior. Axon damage is a hallmark of traumatic injuries and neurodegenerative diseases such as Alzheimer’s. Its ravages upon individuals and societies alike are well known, yet effective treatment of the consequences remains a challenge to modern medicine. Axon regeneration in humans is quite poor due to the limited regenerative capacities of the peripheral nervous system (PNS) and the inhibitory conditions of the central nervous system (CNS). While axon regeneration is regulated by a fine-tuned balance between intrinsic and extrinsic factors, the contribution of intrinsic factors to this process is not fully understood. The MAP3Ks DLK and LZK are two such intrinsic factors. These axon damage sensors are activated in response to injury and direct diverse, context-specific outcomes to injury. Accounting for the environment around the cell, magnitude of the injury, cell type specificity, and subcellular localization of key actors offers insight into these disparate resolutions. This work designs an experimental approach to investigate the context-specific capacities for axon regeneration and finds that DLK and LZK have cell type- and injury type-specific responses to axon damage. In chapter 1, I describe a new protocol for precise axon injury in single cells using zebrafish. Unlike in mammals, the zebrafish PNS and CNS are permissive to axon regeneration, making them a perfect environment to study the contribution of intrinsic factors. I take advantage of zebrafish genetic tractability and optic clarity to mosaically label neuronal cell types. I then use a precise laser to cut individual axons with minimal damage to the surrounding tissue. Using live imaging, I capture the post-axotomy processes of Wallerian degeneration and the axonal dynamics of regenerating axons. In chapter 2, I describe a new zebrafish model to address the variability in DLK and LZK outcomes following axon injury. DLK and LZK CRISPR mutants, as well as DLK LZK double mutants, are adult viable and fertile. Using the protocol described in chapter 1, I found that DLK and LZK are redundantly required for axon regeneration in motor neurons, but not in touch-sensing neurons. However, DLK and LZK regulate collateral sprouting in non-injured axons of touch-sensing neurons, at both the individual neuron and population levels. In chapter 3, I describe an effort to create lines labeling DLK- and LZK- expressing neurons. In different organisms, RNA studies indicate that DLK and LZK translation is mostly neuronal; however, these studies lack cell-type specificity. Using CRISPR/Cas9 technology, I attempted to insert the signal amplifying GAL4 gene upstream of DLK and LZK start sites. These enhancer traps were characterized using confocal microscopy, and several techniques were used to validate the lines. Although validation revealed incorrect insertion of GAL4, identifying the cell types in which these proteins are expressed under basal and pathological conditions would be a powerful tool in elucidating their multifaceted activation.

The brain is composed of complex neuronal circuits that provide the physiological basis for our cognition, perception, and behavior. Unraveling how neurons establish these circuits and the myriad of molecular signals that guide their development is one of the most daunting tasks in neurobiology. Down Syndrome Cell Adhesion Molecules (DSCAMs) play an evolutionarily conserved role in regulating key aspects of neuronal wiring, including programmed cell death, neuronal migration, axon guidance, neurite branching, branch spacing, and synaptic targeting. However, despite being expressed broadly throughout the vertebrate nervous system, there remains a paucity of in vivo investigations of the functions of DSCAM family members across different regions of the brain. Additional studies could shed light on how well DSCAM functions are conserved across the diversity of neuronal cells types in different neuronal systems.

In the present study, I leveraged the genetic tractability and optical accessibility of the larval zebrafish model to investigate the expression and function of a DSCAM family member, dscamb, throughout the developing brain. Using targeted mutagenesis, I created the first dscamb loss-of-function mutant lines, and through a combination of transgenic approaches, I provided the first characterization of dscamb expression. Using these genetic tools, I uncovered that, similar to DSCAM family members in other species, Dscamb is expressed broadly throughout the brain, spinal cord, and peripheral nervous system. However, I also identified several regions of expression, particularly in the peripheral nervous system and muscle cells, that have not been described for other DSCAM family members. Unlike other vertebrate DSCAMs, I found no evidence that Dscam is required for retinal development. Moreover, dscamb loss-of-function did not affect the overall structural organization of the brain and spinal cord. Despite an absence of apparent anatomical defects a series of behavioral analyses revealed that dscamb mutants are severely deficient in their ability to find or capture food, suggesting that this protein has a critical, and perhaps subtle, function in the wiring of neuronal systems that underlie feeding behavior.

The concept that form adapts to function has long been at the core of biological science. However, there are many cellular forms that we do not fully understand. Microridges are actin-based cell membrane protrusions that wind into labyrinthine patterns on the apical surfaces of mucosal epithelial cells, conferring a unique cellular morphology. Though the form of microridges has been studied in fixed tissues for many years, their morphogenesis is not understood. To investigate the formation of these enigmatic structures, I observed the process directly on the epidermis of developing zebrafish larvae using live fluorescence microscopy. In Chapter 2 of this dissertation I describe the mechanism of microridge formation and the role of contractile force in this process. Prior to microridge formation, epithelial cells are covered in “pegs”, short finger-like protrusions that resemble microvilli. Pegs are dynamic, appearing and disappearing within a matter of minutes, as well as fusing with one another to form short microridges. As epithelial cells constrict, short microridges and pegs gradually assemble into elongated microridges in a pattern that begins near cell borders and progresses toward the center of the cell surface. A biophysical model of this process demonstrated that contraction of the apical actomyosin cortex in a similar concentric pattern leads to cellular constriction and microridge formation by reducing membrane tension. Observation of myosin activity in live epithelial cells demonstrated that contraction progresses in a concentric manner and small molecule inhibition of contraction blocked both cellular constriction and microridge formation. Reduction of membrane tension through hyperosmolar shock and reorganization of the apical cortex through rapid cell elongation demonstrated that microridge formation is directly regulated by membrane tension and the organization of the cortical cytoskeleton. Chapter 3 describes the mechanics of microridge formation and pattern maturation with high spatial resolution. After formation, microridges elongate further, align with their neighbors, and become more regularly spaced. Microridge fusion and fission continue during this period of pattern maturation and these events correlate spatially and temporally with cortical myosin contraction. However, microridge rearrangements diminish after microridge formation and negatively correlate with the degree of microridge alignment. Inhibiting myosin contraction prevented microridge rearrangements and disrupted the microridge pattern. High-resolution Airyscan microscopy revealed individual myosin minifilaments in the cortex that connect microridges to one another and mediate microridge rearrangements through contraction. Overall, this work elucidates the mechanism of microridge formation. Pegs and microridges fuse, fission, and gradually rearrange, a process driven by the web-like network of myosin minifilaments in the apical cortex which contracts to reduce membrane tension. As myosin contraction diminishes, the input needed to overcome the energy barrier associated with the changes in membrane curvature during microridge rearrangement is decreased, and microridges slowly settle into their highly organized patterns.

Larval zebrafish are emerging as a model for describing the development and function of simple neural circuits. To analyze somatosensory neuron diversity in larval zebrafish, we identified several enhancers from the zebrafish and pufferfish genomes and used them to create five new reporter transgenes. Sequential deletions of three of these enhancers identified small sequence elements sufficient to drive expression in zebrafish trigeminal and Rohon-Beard (RB) neurons. One of these reporters highlighted a somatosensory neuron subtype that expressed both the p2rx3a and pkca; genes, as well as a previously described trpA1b reporter. To determine whether neurons of this subtype possess characteristic peripheral branching morphologies or central axon projection patterns, we analyzed the morphology of single neurons. Surprisingly, although these analyses revealed diversity in peripheral axon branching and central axon projection, PKCa/p2rx3a/trpA1b-expressing RB cells did not possess characteristic morphological features, suggesting that even within this molecularly defined subtype, individual neurons may possess distinct properties.

Due to their external fertilization, rapid development, and optical clarity, zebrafish larvae are particularly well suited for optogenetic approaches to investigate neural circuit function. Here we demonstrate a procedure for expressing a light-sensitive ion channel in larval zebrafish somatosensory neurons, photo-activating single cells, and recording the resulting behaviors. Specifically, we created a transgene using a somatosensory neuron enhancer to drive the expression of the tagged channelrhodopsin variant, ChEF-tdTomato. Illuminating identified cells in these animals with light from a 473 nm DPSS laser, guided through a fiber optic cable, elicited behaviors that can be recorded with a high-speed video camera and analyzed quantitatively. By introducing a few modifications to previously established methods, this approach was used to elicit behavioral responses from single neurons activated up to at least 4 days post-fertilization (dpf). The new transgenes created in this study will be powerful tools for further characterizing the molecular, morphological, and development diversity of larval somatosensory neurons and in combination with genetic or pharmacological perturbations will be a powerful way to investigate circuit formation and function.

Significance statement