Revealing neural circuits underlying zebrafish behavior using mesoscopic light field microscopy

One of the holy grails of neuroscience is to record the activity of every neuron in the brain while an animal moves freely and performs complex behavioral tasks. While important steps forward have been taken recently in large-scale neural recording in rodent models, single neuron resolution across the entire mammalian brain (even without behavior) remains elusive. In contrast the larval zebrafish offers great promise in this regard. This is a vertebrate model with substantial homology to the mammalian brain, but at larval stages zebrafish can be engineered to be entirely transparent. This allows whole-brain recordings of genetically-encoded fluorescent indicators at single-neuron resolution using optical microscopy techniques. In addition zebrafish begin to show a complex repertoire of natural behavior from an early age, including hunting small, fast-moving prey using visual cues. Up to now work to address the neural bases of these behaviors has mostly relied on assays where the fish is immobilised under the microscope objective, and stimuli such as prey are presented virtually. However this is a impoverished simulacrum of natural behavior, and the extent to which the results obtained generalise to freely moving and behaving fish is unknown. While there have been recent attempts to develop assays which allow brain imaging in moving fish, these impose severe behavioral constraints on the fish, again limiting the ecological validity of the results obtained. Here we propose a new approach which will allow the neural bases of fully natural behavior to be examined in zebrafish. It will combine for the first time two recent technological developments. Light field microscopy (LFM) allows single-shot imaging of an entire volume, avoiding the time-consuming process of scanning a point or sheet of light as is required in more conventional approaches. The mesolens is a specialist microscope objective which allows for the first time imaging of regions 5 mm wide by several mm thick. In combination, which we term MesoLFM, it will be possible to image zebrafish whole-brain activity during natural behavior under ecologically relevant conditions. In this R34 proposal we will establish this technology by addressing the following aims. Aim 1. We will perform modeling and simulations to determine the optimal optical configuration for the MesoLFM. Aim 2. We will build the first MesoLFM microscope, develop an image deconvolution pipeline, and calibrate it using fluorescent beads. Aim 3. We will use MesoLFM to demonstrate whole brain imaging of behaving zebrafish. By developing this critical enabling technology we will in future proposals be able to address questions about unconstrained natural behaviors such as the following. What are the neural bases of the perception-action cycle of unconstrained zebrafish hunting? How do these neural circuits emerge during development? How are they affected by the environment in which the animal is reared? What are the relations between spontaneous neural activity and spontaneous behavior? How far in advance does neural activity predict behavior? What are the circuit mechanisms underlying the processing and prioritization of competing stimuli? How do neural representations vary between individuals, and how is this connected with behavioral variations? Northwestern University will be responsible for developing the software modeling and simulation pipeline developed in Aim 1, and the image deconvolution pipeline developed in Aim 2. Work completed by NU in Aim 1 and Aim 2 will be performed in close collab