Article In Brief
Researchers created 3D models of miniature brains—organoids—grown from human induced pluripotent stem cells, and they appear to produce oscillations similar to those of preterm babies.
Cortical “organoids” grown from human induced pluripotent stem (iPS) cells develop neurons, form synapses, and fire in an oscillatory pattern reminiscent of the preterm infant brain, according to an August 23 study in the journal Cell Stem Cell. The finding supports the use of these model systems for studying human diseases, such as epilepsy and autism, in which early developmental events may lay the foundation for aberrant neuronal firing patterns much later in life.
“The primary way brain organoids have been studied so far is to assess development at the cellular and genetic level,” said Marla Feller, PhD, professor of neurobiology at the University of California, Berkeley, who was not part of the new study. “There have been very few studies recording from neurons in organoids, to see if they are maturing normally, so this was a nice study in that regard. I think this was an important achievement.”
But John Huguenard, PhD, professor of neurology and neurologic science at Stanford University, was more cautious. Cortical organoids can display electrical activity, and even fire in a temporal pattern, but whether that firing provides insight into actual developing circuits is another question, he said. “You can get something that looks like an oscillation but doesn't in any way resemble what is happening in the developing brain. That's my huge caution on this.”
Organoids have emerged in the past several years as a new type of model system for studying the development and self-organization of actual organs. Like other organoids, cortical organoids are derived from iPS cells and are guided to develop—in this case into brain cortex—by exposure to a specific suite of factors. Unlike in standard cell culture, organoids are grown in suspension, allowing more naturalistic three-dimensional interactions among developing cells, leading to higher-level structures not seen in two dimensions.
“In the earliest stages, most of the cells are neural progenitor cells,” explained Alysson Muotri, PhD, professor in pediatrics at the University of California, San Diego School of Medicine, who led the new study. “But after several months, they self-assemble and form a small ventricle, with cells that migrate out and start forming the cortical plate.” The organoid reaches a maximum diameter of about half a centimeter, constrained by the lack of a circulatory system to grow any thicker, he explained, added that even at that size, distinct cortical layers form.
“The biggest advantage of three-dimensional culture is to gain these structures. We don't see stratification of layers in the cortex in two-dimensional culture. What we think is happening is that when you have a 2D culture, the neurons make synapses, but they are probably random. In 3D, they take the time to really form connections where they are supposed to, between neighboring neurons and between different layers.”
But while brain organoids have mimicked important structural features of the developing brain, they have as yet not shown electrical oscillations—patterns of firings among connected neurons. “The emergence of oscillations is the basis for all complex tasks that the brain will eventually perform,” Dr. Muotri said. “Without these oscillations, you will never have any complex networks that can dictate behavior.
Study Design, Findings
In the new study, Dr. Muotri and colleagues grew cortical organoids for over nine months—they can live for several years. By three to six months, glutamatergic neurons predominated, along with glial cells, which were joined by GABAergic cells from six to ten months.
Using multi-electrode arrays, the team recorded from the organoids, finding an increasingly complex firing pattern over time. As they described in their paper, the neurons switched between “long periods of quiescence and short bursts of spontaneous network-synchronized spiking,” about every 20 seconds early on. By four months, a second, faster oscillatory pattern emerged, superimposed on the first. These regular oscillations became stronger and more variable over time.
“Once we had the oscillations, the next question we asked was whether they were similar to the human brain,” Dr. Muotri said. There are no recordings of the developing brain in utero, but there are EEG recordings from pre-term infants, and so Dr. Muotri turned to those recordings.
“The EEG has much more information that we can extract from the organoid, but there are several features in common, so we isolated what was comparable,” he said, including timing and frequency of bursts, and the distribution of specific frequencies of oscillatory waves. “When we do that, we see they are very similar, and we see a nice progression. The organoid develops in similar ways as the human brain would do in terms of these EEG features. That was a big surprise.”
The results, Dr. Muotri said, suggest that the development of electrical activity in the organoid may follow a stable genetic program, and so may be a valid model for examining the earliest wiring up of the cortex and how it can go wrong. “I think we now have an experimental model to study that.”
The team hopes to use their cortical organoids to ask questions about human developmental diseases. “The low-hanging fruit is to make brain organoids from people carrying specific genetic variants for autism or seizure disorders, where the EEG is really different from normal,” he said. They are beginning with early-onset epilepsy, to see whether the organoids display an aberrant firing pattern. “If they do, then I think we have a powerful human model for those diseases.”
“Organoids are likely to be a great model for studying genetic diseases of development,” commented Dr. Feller of UC Berkeley, who studies retinal development in rodents. “Our understanding of diseases tends to be postnatal, but many of the events that will lead to the phenotypes are happening much earlier in development, which then propagate into a quite complicated phenotype much later.” Studying development in brain-mimicking organoids “could give us the opportunity to find at what point in development things begin to go awry.”
Oscillatory activity has been shown to be critical for development of complex circuits in the visual system, the auditory system, and the spinal cord, and within the brain, between cortex and thalamus. “One of the features of an immature brain is the existence of these correlated network events, these big depolarizations that happen in many regions of the brain, so it was exciting to see the oscillatory activity occurring in this study.”
The promise of the brain-mimicking organoids “is that they will ultimately allow us to study patient-specific issues in neural development and function,” said Dr. Huguenard of Stanford, who studies the origins of thalamocortical epilepsy. But it is “a huge challenge” to capture a large number of meaningful features in vitro, and especially so when it comes to modeling the intricate electrophysiologic behavior of the real developing brain.
“Migration and connections can be captured pretty well by these organoids. I think right now that is the sweet spot for this overall approach. What is not the sweet spot is starting to study complex, higher-order neural oscillatory activity.”
The firing patterns seen in this study “are more representative of what happens when you hook up only excitatory cells, creating an auto-excitation circuit that would drive itself for as long as it could, and then stop. That is reminiscent of what we are seeing here.”
More realistic networks are definitely within sight, Dr. Huguenard said. One approach is to merge a thalamic organoid with a cortical one, to provide more sophisticated inhibitory regulation. That work is just starting, he said.
“It is exciting that neurons are being produced, and showing action potentials, and beginning to connect with one another, and beginning to communicate. But it is such an immature form that we are very far away from creating realistic patterns. These organoids represent a highly, highly immature part of the nervous system.”
Alysson Muotri is a co-founder and has equity interest in TISMOO, a company dedicated to genetic analysis and brain organoid modeling focusing on therapeutic applications customized for autism spectrum disorder and other neurological disorders with genetic origins. The terms of these arrangements have been reviewed and approved by the University of California, San Diego in accordance with its conflict of interest policies. Drs. Feller and Huguenard had no disclosures.