Computational 3D histological phenotyping of whole zebrafish by X-ray histotomography

Abstract

Organismal phenotypes frequently involve multiple organ systems. Histology is a powerful way to detect cellular and tissue phenotypes, but is largely descriptive and subjective. To determine how synchrotron-based X-ray micro-tomography (micro-CT) can yield 3-dimensional whole-organism images suitable for quantitative histological phenotyping, we scanned whole zebrafish, a small vertebrate model with diverse tissues, at ~1 micron voxel resolutions. Micro-CT optimized for cellular characterization (histotomography) allows brain nuclei to be computationally segmented and assigned to brain regions, and cell shapes and volumes to be computed for motor neurons and red blood cells. Striking individual phenotypic variation was apparent from color maps of computed densities of brain nuclei. Unlike histology, the histotomography also allows the study of 3-dimensional structures of millimeter scale that cross multiple tissue planes. We expect the computational and visual insights into 3D cell and tissue architecture provided by histotomography to be useful for reference atlases, hypothesis generation, comprehensive organismal screens, and diagnostics.

eLife digest

Diagnosing diseases, such as cancer, requires scientists and doctors to understand how cells respond to different medical conditions. A common way of studying these microscopic cell changes is by an approach called histology: thin slices of centimeter-sized samples of tissues are taken from patients, stained to distinguish cellular components, and examined for abnormal features. This powerful technique has revolutionized biology and medicine. But despite its frequent use, histology comes with limitations. To allow individual cells to be distinguished, tissues are cut into slices less than 1/20th of a millimeter thick. Histology’s dependence upon such thin slices makes it impossible to see the entirety of cells and structures that are thicker than the slice, or to accurately measure three-dimensional features such as shape or volume.

Larger internal structures within the human body are routinely visualized using a technique known as computerized tomography, CT for short – whereby dozens of x-ray images are compiled together to generate a three-dimensional image. This technique has also been applied to image smaller structures. However, the resolution (the ability to distinguish between objects) and tissue contrast of these images has been insufficient for histology-based diagnosis across all cell types. Now, Ding et al. have developed a new method, by optimizing multiple components of CT scanning, that begins to provide the higher resolution and contrast needed to make diagnoses that require histological detail.

To test their modified CT system, Ding et al. created three-dimensional images of whole zebrafish, measuring three millimeters to about a centimeter in length. Adjusting imaging parameters and views of these images made it possible to study features of larger-scale structures, such as the gills and the gut, that are normally inaccessible to histology. As a result of this unprecedented combination of high resolution and scale, computer analysis of these images allowed Ding et al. to measure cellular features such as size and shape, and to determine which cells belong to different brain regions, all from single reconstructions. Surprisingly, visualization of how tightly the brain cells are packed revealed striking differences between the brains of sibling zebrafish that were born the same day.

This new method could be used to study changes across hundreds of cell types in any millimeter to centimetre-sized organism or tissue sample. In the future, the accurate measurements of microscopic features made possible by this new tool may help us to make drugs safer, improve tissue diagnostics, and care for our environment.