A 4D Atlas Maps Every Gene and Cell in the Developing Zebrafish Embryo

How does a ball of identical-looking cells become an animal with a head, a spine, and a tail? The question has driven developmental biology for over a century. Now, for the first time, researchers can watch the answer unfold across an entire vertebrate embryo, gene by gene, cell by cell, in three dimensions over time.

A team led by Professor Alex Schier at the Biozentrum, University of Basel, has created what amounts to a 4D atlas of early development in zebrafish. Using a new imaging technology called weMERFISH, they can measure the activity of nearly 500 genes simultaneously throughout an entire embryo at subcellular resolution. The results, combined with existing single-cell datasets, produce spatial maps of thousands of genes and around 300,000 potential regulatory regions. The study was published in Science.

The problem with slices

Previous methods for measuring gene activity in embryos relied on thin 2D tissue sections. These slices captured gene expression in one plane but made whole-embryo visualization impossible. They missed spatial relationships between distant tissues and could not resolve patterns at the subcellular level. Trying to understand embryonic development from 2D slices is a bit like trying to understand a city from a handful of cross-section photographs.

weMERFISH solves this by enabling direct measurement of gene activity across intact, whole embryos. First author Dr. Yinan Wan explains that by combining their measurements with previous single-cell data, the team could compute spatial patterns for thousands of genes - far more than the 500 they measured directly.

The resulting atlas is freely accessible through the web platform MERFISHEYES (http://schier.merfisheyes.com), intended as a resource for developmental biologists worldwide.

Seeing time written in space

One of the most striking observations came from studying tail formation. Along the body axis, cells are arranged in a spatial sequence that mirrors their developmental age. At the tip of the growing tail sit immature stem cells. Moving forward along the body, cells become progressively more mature - differentiating into muscle, neural tissue, and other specialized types.

The spatial arrangement acts as a developmental timeline frozen in tissue. As Wan puts it: you can see time in space. The team also found that changes in gene activity closely track how cells move through the embryo, linking gene expression dynamics directly to the physical rearrangements - called morphogenetic movements - that shape the developing body.

Boundaries that form without sorting

The atlas also resolved a long-standing question about how sharp boundaries form between different tissue types - for example, between muscle and backbone tissue. One hypothesis held that cells from different tissue fates initially mix together and then sort themselves out, migrating to join their correct tissue. The data tell a different story.

The researchers identified a transition zone where the activity of many genes changes dramatically from one side to the other. Comparing early and later developmental stages revealed that genes initially active on both sides of the boundary become restricted to one side over time. And critically, very few cells cross this boundary. The sharp borders arise not because cells sort themselves but because cells on either side change their genetic programs independently.

What the atlas cannot yet do

The current atlas covers early zebrafish embryogenesis - a specific window of development in a specific organism. While zebrafish share many developmental genes and pathways with mammals, they are not mammals. How directly these spatial gene-expression patterns translate to mouse or human embryonic development is unknown.

The technology measures gene activity at the RNA level, which does not always correspond to protein levels or protein function. Genes can be transcribed without producing functional protein, and post-transcriptional regulation adds layers of complexity that weMERFISH does not capture.

The study also focused on wild-type (normal) embryos. Understanding what goes wrong in development - the mutations and environmental disruptions that cause birth defects - will require applying the same technology to perturbed embryos, work that has not yet been done.

Schier's team plans to extend the atlas to additional developmental stages to build a more complete picture. Their long-term goal is to understand which combinations of gene activity and cellular behavior are required to form specific organs and tissues. As Schier puts it, one day they may discover how many ways there are to build a heart or a spinal cord. That day is not here yet, but the atlas provides the foundation.