Biologists have long labored at understanding how our body develops. Centuries of embryology and morphology laid the groundwork for the discipline of developmental biology. Then came the discovery of DNA. And by the late 1970s, the heyday of molecular biology and developmental genetics had arrived: fruit fly geneticists discovered Hox and other major genes that set up the fruit fly body plan.
These genes are connected in intricate gene regulatory networks in which they turn one another on and off like switches. Subsequent discoveries showed that novel features in animal bodies are related to changes in where and when these organizing genes are expressed. So it is not so hard to see why the genome has long been considered the master blueprint for building bodies.
It turns out this story is at best incomplete. Even back when Alfonso Martinez Arias was a young developmental biologist at the University of Cambridge in the 1980s, he had suspicions that not everything about fruit fly development could be explained by genes alone. When cloning and stem cell technology arrived in the late 1990s, Martinez Arias immediately saw that these areas had the potential to address questions in developmental biology that had been previously unanswerable. Over the past two decades, he and his colleagues have worked with embryonic stem cells to tease out fundamental principles for how embryos develop. In doing so, they have discovered that even in the absence of external cues, stem cells can be reliably coaxed to initiate gastrulation—to form the beginnings of an entire body plan, the defined structure and shape of an organism, in a lab dish—revealing an unexpected self-organizing ability.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
In recent years the refinement of techniques to induce embryonic stem cells to differentiate and build small structures known as organoids has enabled the field to flourish—and caused a reconsideration of previously overlooked factors that matter in development. Now, elucidating the chemical and mechanical cues that underlie the extraordinary self-organizing property of stem cells is the next frontier for developmental biology, Martinez Arias argues in his recent book, The Master Builder: How the New Science of the Cell Is Rewriting the Story of Life(Basic Books, 2023).
Scientific American recently spoke with Martinez Arias, who is now a Catalan Institution for Research and Advanced Studies (ICREA) research professor at Pompeu Fabra University in Barcelona.
[An edited transcript of the interview follows.]
How did the idea of genes as the blueprint for development emerge? And what were your first inklings that this wasn’t the whole story?
Developmental genetics showed us that you can disrupt development by mutating genes, but that is not the same as understanding what the products of those genes contribute to development. It is one thing to break something, another to understand how each of its parts works and contributes to the whole. If you remove a screw from a car, and the car appears later, smashed against a tree, and now you have to figure out what that screw is normally doing in the car, well, it is going to be a difficult job. I think sometimes mutations in development work like that. It is very difficult to figure out what the gene that has been mutated does.
When you start asking questions about why we have five fingers, or why our eyes are spherical, you realize that the answer does not lie in the genes. Rather it is because cells are able to generate those shapes, and they are controlling the genes under those conditions.
Cells have properties, which we are starting to discover, such as their ability to read the environment—not just in terms of nutrients but also pressure, forces, geometry, numbers of neighbors. Those things are impinging on what cells do and the genes they will use. You could say that cells have proteins that sense these things. This is true, but it is not one protein; it is an ensemble of proteins that now acquire properties that they don’t individually have. This is what is called “emergence.” This is not a mystical property of matter; it is a real property that today we can explore and understand and that lies at the heart of how cells work.
What are gastruloids, and how are they revealing these self-organizing properties of cells?
Gastruloids are structures that begin as an aggregate of embryonic stem cells and mimic aspects of early embryonic development, in particular gastrulation—the process whereby the early embryo folds inward and establishes the cell lineages that will make ectoderm, endoderm and mesoderm [different cell layers]. Gastruloids also go through the establishment of the body axes [head-to-tail axis; front-to-back axis; left-to-right axis]. What is important is that gastruloids recapitulate some features of early development even without the external cues from the placenta or yolk sac that typically direct the organization of an early embryo.
When you put stem cells in a two-dimensional culture, they don’t do very much other than differentiate into various cell types, which is already important. But if you put them together in a little aggregate, now they start doing wonderful things. But those wonderful things depend on the number of initial cells: if you put in too many or too few, nothing happens. They are the same cells—they’re the same genes—but they do totally different things depending on the number of cells in the aggregate. The fact that this only happens when we have a defined number of cells raises a whole lot of questions that we cannot easily map onto genes.
How are gastruloids changing our fundamental understanding of development?
I like gastruloids because they pose questions. For example, the fact that gastruloids do not receive any cues from extraembryonic tissues and yet organize themselves perfectly well gives us an opportunity to figure out how this works.
It is possible to make gastruloids from many different organisms—from fish, frog, pig, mouse and human stem cells. When you take pluripotent stem cells from these species and separate the cells from the maternal organization that they have and put them in the same chemical conditions, they produce indistinguishable structures across species. And they do it very reproducibly. It’s not a fish; it’s not a mouse; it’s a gastruloid.
What is exciting is that no matter the species from which we make it, all gastruloids look the same. I found that very remarkable, because what it tells us is that when you remove the physical constraints of an early embryo, its cells revert to some basic shape, which I would call a morphogenetic ground state. That is telling us that it is not the genes that are creating the shape. It’s the actual mechanical, physical and nutritional constraints that the cells have.
By modifying the physical and chemical environment of these cells, we can start to decode the same mechanical and chemical signals that cells use to sense and respond and communicate among themselves when they are building a body.
What are some possible applications of these gastruloids?
It is early days, but we have used gastruloids to study the process of somitogenesis, the process that generates the vertebral column and the muscles. And we’ve introduced some mutations that affect the development of that process. A lot of pathologies have their origin in very early embryos, and that’s very difficult to study in humans. So one of the ways we are trying to use these systems is to model diseases that happen during gastrulation—in this case, ones associated with abnormalities of the spinal cord. Toxicology of early pregnancy is also a very important field where there are no appropriate nonanimal models. I think that gastruloids can provide a very useful model for these studies.
What are some of the big remaining open questions in the field of developmental biology?
The embryo of a whale or an elephant is not very different from the embryo of a sheep. But then it will grow in a proportionate manner to make a whale or elephant or sheep. And how is that regulated? That’s a really profound question, and I don’t think we have the answer. I think this is a question about how cells sense space, how they measure size. That’s really what I think some of these structures that we can create from stem cells can teach us.
We feel comfortable with genes because that’s what the 20th century has given us. But if you look back at the beginning of the 20th century, we didn’t know very much about genes. And that didn’t stop people from asking questions related to genes [for example, questions about the mechanisms of inheritance and evolution]. I think we are in a similar situation today with regard to cells. The problem is that we have something, genes, that we use to explain everything instead of asking questions about cells. Fortunately, there are people that are asking those questions, and it’s going to be very exciting, in 20 or 30 years, to see what they will have discovered. We have to be bold and go into the unknown.