Cerebral organoid. The brown bit is retinal tissue. Credit: Madeline A. Lancaster
Cerebral organoid. The brown bit is retinal tissue. Credit: Madeline A. Lancaster

The Cerebral Organoid, a Lab-Grown Model Brain

A team of Austrian and British scientists have created 3-D models of the (embryonic) human brain, and given them the fantastic name of “cerebral organoids”. I’ve written about these in The Scientist so head over there for the details. In the meantime, here’s an explainer to clarify what these things are, and what they are definitely not.

What’s a “cerebral organoid”?

An organoid is a lab-grown, three-dimensional ball of cells that has some of the features of a normal organ. In recent years, scientists have made organoids for intestines, kidneys and eyes. A “cerebral organoid” is the name of the brain version. It would also be a good insult, but a rubbish name for a Jaeger.

Oh, so it’s a brain in a jar?

It is emphatically not a brain in a jar. Jürgen Knoblich, who developed the organoids, told me that’s a very misleading description. The organoids resemble a brain during the first nine weeks of pregnancy and although they show some of the features of a normal brain, “they don’t form anything that resembles a neuronal network,” says Knoblich. In other words, they’re not sensing and thinking, any more than a splinter of your thigh bone would suddenly start walking about.

But they’re a little brain-like?

Yes. The organoids look like white, pea-sized balls but they’re not unstructured blobs of tissue. They have distinctive zones that correspond to parts of the human brain, like the prefrontal cortex or hippocampus. Some of them develop an immature retina. They also produce neurons in the right way. In the cerebral cortex—the outermost slab of tissue in your brain—neurons are made at the bottom and migrate to the top. The same thing happens in the organoids.

Then again, those regions aren’t as organised as they’d be in a normal embryonic brain, and they’re not found in the usual places relative to one another. As I said, it’s a rough model of a brain, but it isn’t one.

How are they made?

With surprising ease, and that’s the cool bit. You might think that to create something with distinctive brain regions, you’d need to carefully orchestrate its growth, adding the right ingredients at the right time, or maybe pulling and shaping it with some tiny forceps.

You’d be wrong. Madeline Lancaster, a member of Knoblich’s group, made them by: bathing clusters of stem cells in nutrients that persuade them to make neurons; embedding them in a gel for structural support; and putting them in a spinning vat so they get enough air and nutrients. And that’s it. They grew and assembled all on their own. Knoblich told me: “This demonstrates the enormous self-organizing power of human cells. Even the most complex organ—the human brain—can start to form without any micro-manipulation.”

What are they for?

Without sophisticated neural networks, you can’t use them to study the connectivity in the brain, or to probe abilities like decision-making or consciousness. Instead, they’re meant to be tools for studying the early development of the brain, and disorders that disrupt this development.

Can’t you just do that by studying animals?

To a point. But many developmental disorders are hard to duplicate in animals like mice, because their brains grow in fundamentally different ways to ours. Consider microcephaly—a condition where people grow up with a small brain. Knoblich’s team worked with a Scottish patient with microcephaly, thanks to mutations in a gene called CDK5RAP2—one of several genes associated with the disorder. They took the patients’ skin cells, converted them into stem cells, and used them to make organoids… which ended up much smaller than usual. They could also work out exactly why this happened – see The Scientist piece for more.

Now, if you introduce the same mutations into a mouse embryo, it would grow up with a slightly smaller brain, but it wouldn’t come close to recapitulating the same extreme stunting. So, in this case, the organoid’s a better model for what’s going on in the human. After all, it’s made from a human—it’s a personalised model brain. Well, a quasi-brain. Brain-ish. You get the idea.

That doesn’t mean that animal experiments, or indeed human ones, are suddenly unnecessary. The organoids are more a tool for testing hypotheses generated through other studies.

Could you ever grow a full brain?

It’s unlikely. For a start, the organoids don’t have any blood vessels so they can only get a few millimetres wide before the cells in the middle start to starve and die. Even if the team solved this problem, human brains don’t grow in isolation. They’re connected to eyes and ears and bodies. From an early age, they start picking up information. Growing a ball of neurons in the absence of any of that doesn’t get you a proper brain, any more than erecting a set of shelves gets you a library.