Correlative Light and Electron Microscopy in an Intact Larval Zebrafish

Abstract

How does the structure of a living system relate to its function? In neuroscience, an outstanding issue is what can be learned about the brain’s function by knowing how the individual neurons are wired together. We use high resolution serial-section electron microscopy (EM) to map the structure of an entire vertebrate’s nervous system at the synaptic scale. The dataset is collected from an intact larval zebrafish at 4nm x 4nm x 30nm resolution and the imaged volume size is 76 million cubic microns. It contains 170,000 neurons and glia, and offers comprehensive synaptic connectivity information from both the central and peripheral nervous systems. Additionally, we combine complementary information from light and electron microscopy in the same animal. Specifically, we pair electron microscopy with confocal microscopy of the same brain, where the major excitatory and inhibitory neurons are fluorescently labeled. To achieve single-cell precision in the matching of the two datasets, we solve a multi modal image registration challenge, which handles non-uniform deformations introduced during tissue processing between the two imaging modalities. This correlative electron microscopy data is essential to test computational models for circuit function in larval zebrafish. Most generally, these experiments advance larval zebrafish as a vertebrate model system in which vastly disparate scales can be bridged – brain-wide synaptic connectivity, brain-wide neuronal activity, and animal behavior.