MADELINE A. LANCASTERIn an Austrian laboratory, a team of scientists has grown three-dimensional models of embryonic human brains. These “cerebral organoids” are made from stem cells, which are simply bathed in the right cocktail of nutrients and grown in a spinning chamber. Over a few weeks, they arrange themselves into pea-sized balls of white tissue, which recapitulate some of the complex features of a growing brain, including distinct layers and regions.
“This demonstrates the enormous self-organizing power of human cells,” said Jürgen Knoblich from the Institute of Molecular Biotechnology of the Austrian Academy of Science, who led the study published in Nature today (August 28). “Even the most complex organ—the human brain—can start to form without any micro-manipulation.”
Knoblich cautioned that the organoids are not “brains-in-a-jar.” “We’re talking about the very first steps of embryonic brain development, like in the first nine weeks of pregnancy,” he said. “They’re nowhere near an adult human brain and they don’t form anything that resembles a neuronal network.”
These models will not help to unpick the brain’s connectivity or higher mental functions but they are excellent tools for studying both its early development and disorders that perturb those first steps. For example, Knoblich’s team produced unusually small organoids using stem cells taken from a patient with microcephaly—a neurodevelopmental disorder characterized by a small brain. Knocking out microcephaly-associated genes in mice does very little because murine brains develop differently than humans’. The organoids could help to bypass the limitations of these animal models, providing a more accurate representation of human brains.
Madeline Lancaster, a member of Knoblich’s group, created the 3-D models from small clusters of stem cells. She bathed the cells in nutrients that nudge them toward a neural state, embedded them inside a gel for structural support, and grew them in a spinning bioreactor to help them absorb more nutrients. It took a huge amount of work to fine-tune the conditions, but once the team did, the organoids grew successfully within just 20 to 30 days.
Using molecular markers tuned to specific parts of the brain, Lancaster showed that the organoids develop a variety of distinctive zones that correspond to human brain regions like the prefrontal cortex, occipital lobe, hippocampus, and retina. They also included working neurons, which were produced in the right way—they were made by radial glial cells at the innermost layers of the cortex, before migrating to the outer layers.
Other scientists have developed organoids that mimic several human organs, including eyes, kidneys, intestines, and even brains. For example, in 2008, Yoshiki Sasai’s team at the RIKEN Center for Developmental Biology showed that stem cells can be coaxed into balls of neural cells that self-organize into distinctive layers. But compared to this earlier attempt, the new organoids are “the most complete to date in terms of features that directly resemble those in the developing human brain,” according to Arnold Kriegstein, a stem cell biologist from the University of California, San Francisco, who was not involved in the study.
“They really highlight the ability just nudge these human embryonic cells and allow them to self-assemble,” Kriegstein added. “So much of the signalling that goes on and the actual specification of different parts of the brain occur intrinsically in these cells.”
Having refined their technique, the team created a “personal organoid” from a Scottish patient with severe microcephaly, who had several mutations in a gene called CDK5RAP2. They took skin cells from the patient, reprogrammed them into a stem-like state, and used them to grow organoids that ended up much smaller than usual. By dissecting the organoids, the team discovered the reason for this stunted size.
When healthy brains develop, radial glial cells first divide symmetrically to increase their numbers before dividing asymmetrically to produce neurons. In the microcephalic organoids, this switch happens prematurely, and neurons start forming when the pool of radial glial cells is too low. As a result, the brains do not develop enough neurons and end up small. CDK5RAP2 is responsible for this premature switch; when the team added the protein back into the mutant microcephalic organoids, they grew to a normal size.
Wieland Huttner, a neurobiologist from the Max Planck Institute of Molecular Cell Biology and Genetics, said that these results merely confirm what others had already suspected about CDK5RAP2. However, the organoids could be more useful for understanding other microcephaly genes whose roles are still unclear.
For example, mutations in the ASPM gene can shrink a human brain by a third of its normal size, but barely make a dent in the size of a mouse brain. “The mouse brain isn’t good enough for studying microcephaly,” said Huttner. “You need to put those genes into an adequate model like this one. It is, after all, human. It definitely enriches the field. There’s no doubt about that.”
Knoblich cautioned that organoids are unlikely to replace animal experiments entirely. “We can’t duplicate the elegance with which one can do genetics in animal models,” he said, “but we might be able to reduce the number of animal experiments, especially when it comes to toxicology or drug testing.”
In the future, he hopes to develop larger organoids. For the moment, the models cannot get any bigger without a blood supply, and their interiors are dead zones comprised of starving, choking cells. If the team can solve this problem and coax the organoids to continue growing, they might be able to capture later events in brain development, which may be relevant to other disorders, like autism. “That would be a gigantic step forwards,” said Knoblich.
M.A. Lancaster et al., “Cerebral organoids model human brain development and microcephaly,” Nature, doi: 10.1038/nature12517, 2013.