Incredibly detailed embryo maps chart each cell’s developmental fate

The path to a mature organism is much more dynamic than previously thought.

Search for this author in:

False-coloured scanning electron micrograph of a zebrafish embryo.

Researchers have used embryos from zebrafish (pictured) and frogs to track development in the greatest detail yet.Annie Cavanagh/Wellcome Collection

Painstaking genomic analyses of thousands of individual cells in frogs and fish have yielded the most detailed roadmaps yet charting an embryo’s journey from a single cell to a fully formed organism.

The results paint a picture of a developmental pattern more fluid than the one described in many textbooks, says systems biologist Sean Megason of Harvard Medical School in Boston, Massachusetts. “Things are happening at different times, sometimes faster than we appreciated,” he says. “There is often more of a continuum of cell states.”

The work, described by Megason and others in three papers published on 26 April in Science1,2,3, will help researchers to track key developmental stages, and to uncover the processes that contribute to certain conditions, such as autism or cancer. A third of all neurological diseases first appear during development, says Sten Linnarsson, a neuroscientist at the Karolinska Institute in Stockholm. “And they probably happen quite early.”

Researchers have tried to create such gene expression maps, but for many years they could only focus on a single gene or a handful of cells. Technological advances in DNA sequencing and computer algorithms now allow scientists to analyse the expression of thousands of genes in a single cell. This has sparked a rush to apply the techniques in developmental biology. Last July, for example, the Paul G. Allen Frontiers Group, a research initiative based in Seattle, Washington, announced that it would provide US$10 million over four years to projects that harness such methods to create maps of cellular development.

Allon Klein, a systems biologist at Harvard Medical School, wanted to make those kinds of detailed maps when he first started working out methods for sequencing RNA from single cells in 2012. But they were considered impossible at the time, he says. “The landscape now is quite different from where it was when we started.”

Flexible fate

For the latest studies, a group led by Klein and systems biologist Marc Kirschner, also at Harvard Medical School, analysed gene expression in each of nearly 140,000 cells taken from Western clawed frog embryos (Xenopus tropicalis) over the course of 17 hours of development1. Klein also teamed up with Megason and others to map gene expression in 92,000 cells collected from zebrafish (Danio rerio) embryos over the course of a day2 — long enough for them to develop from a single cell to an embryo with a beating heart.

A third team, led by computational biologist Aviv Regev of the Broad Institute of MIT and Harvard and developmental biologist Alexander Schier of Harvard University, both in Cambridge, Massachusetts, analysed almost 39,000 cells from zebrafish embryos during the first 12 hours of development3.

Following the embryos over time allowed the researchers to watch as cells assumed specialized roles, becoming nervous tissue, for example, or skin. Sometimes cells that seemed to be destined for one fate were directed towards another, probably because of their position in the embryo, says Schier.

“The cells are more plastic than maybe we thought,” he says. “This might actually be quite common for cells to go down one path, and then switch and go down another path.”

Branching out

The researchers deposited their maps into databases that others can search to find out where any gene is expressed, and during what stage of development. Developmental biologist Thomas Schilling of the University of California, Irvine, hopes to mine the data for information about the development of the neural crest, a structure found in vertebrates that gives rise to many cell types, including some brain cells. “We’re interested in when cells become specified for different fates, and looking for transitional states,” he says, adding that his lab might consider a similar experiment focused more specifically on neural-crest development.

Now that the techniques and algorithms have been developed, Klein hopes to apply the methods to a wide range of creatures, to learn more about how evolution shapes development. Initially, his team focused on vertebrates, he says, because of their evolutionary proximity to humans. Now, he hopes researchers can study more distant relatives, including invertebrates like lancelets and acorn worms, to learn about the evolution of the spinal cord. “Understanding how evolution has tweaked development gene expression dynamics to create each new cell type and organ is going to be fascinating," he says.

doi: 10.1038/d41586-018-04986-5
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up