With mini-placentas and mini-brains, scientists try to unravel the roots of psychiatric disorders

BALTIMORE — They float in ordinary lab dishes, no more eye-catching than a plain beige lentil, and often so delicate they shrivel up and die if temperatures or food or the very air around them deviates from Goldilocks perfection. No wonder scientists coddle such “organoids” — Lilliputian versions of kidneys and hearts and intestines and even brains, all created from human stem cells. Their fragility can sink months of work.

Biologist Jennifer Erwin of the Lieber Institute for Brain Development, however, has no intention of babying her organoids: the world’s first human placentas in a dish that were made from stem cells. Challenging as the half-millimeter-across organoids were to create, she intends to starve them of oxygen and douse them with stress hormones, among other assaults. It’s all for a good cause: to mimic pregnancy complications that raise the risk of brain development going off the rails, resulting in conditions including schizophrenia, autism, attention deficit hyperactivity disorder, and intellectual disability.

If those experiments work, Erwin has a sequel in mind. She plans to grow the placental organoids with brain organoids. Then, when the mini-placenta suffers physiological distress, she will be able to measure what goes wrong in the mini-brain, gathering clues about how these disorders arise — Are genes overactive? Or underactive? Are neurons forming too few synapses? Too many? — and, ideally, identifying ways to prevent them.

The results of the experiments could fill in a huge blank in science’s understanding of brain development. Years of studies show that if a mother experiences stress during pregnancy, it increases the chances that the child will develop schizophrenia or other neurodevelopmental disorders. “In some cases the risk is two to four times greater,” said neuroscientist Dr. Daniel Weinberger, director of the Lieber Institute. By comparison, genetic variants increase the odds of developing schizophrenia by only a fraction.

But between “complicated pregnancy” and “schizophrenia” is the black box of the brain. By measuring how placental distress affects brain organoids, Erwin hopes she’ll learn what happens in the full-size versions.

Human organoids are being used to understand how diseases develop, to screen drugs, and much more. Mini-me kidneys and intestines hardly draw a second thought from bioethicists. But cerebral organoids, like the brain itself, have a different ethical status, spurring intense discussions (and government-funded studies) of how far they should be allowed to develop, what moral considerations they demand, and whether they might become conscious … probably without anyone knowing it.

Growing brain organoids alongside placental organoids, however, did not strike bioethicists as inherently problematic. To the contrary.

“I would argue that there are strong scientific and ethical reasons to build and study such [placental/brain] organoid models,” said bioethicist Jeantine Lunshof of Boston’s Wyss Institute for Biologically Inspired Engineering, who previously advised George Church’s Harvard Medical School lab on cerebral organoids. “There is no other way, to my knowledge, to test the effects of placental dysfunction on the developing brain. This is such a huge unanswered problem, if you found clues that could lead to prevention or treatment it would be a great medical and moral good.”

That doesn’t erase the boundary-pushing nature of the planned studies.

“This is in many ways science fiction,” said Weinberger. “It’s very far-out.”

The Lieber Institute, which is affiliated with but independent of the Johns Hopkins University School of Medicine, has no shortage of full-size human brains. Through agreements with four medical examiners, it receives some 500 brains per year, adding to a collection that now numbers about 3,000. Its labs in an 11-year-old glass-and-steel building in Hopkins’ Science + Technology Park in north Baltimore look like those of countless biomedical institutes, with genome sequencers and Zeiss microscopes as ubiquitous as Kind bars at a startup.

But the institute differs in a notable way from other labs. Here, the covers of the gray, knee-high carts that staffers wheel through the corridors warn, in Magic Marker scribble, that the dry ice inside is “not for public use.” It’s to keep brains cold. When not in transit, the brains sit in 28 freezers in a bare, concrete-floored room. Digital thermometers set into the doors show readings of 80 below 0 Celsius.

Those brains have gotten Weinberger and his colleagues pretty far in their quest to understand how pregnancy problems can cause schizophrenia. Medical records indicate which brain donors had schizophrenia and, sometimes, whose mothers had problematic pregnancies; those data clearly show that obstetrical complications such as diabetes, oxygen deprivation, poor nutrition, and infection raise the risk of schizophrenia. But they are silent on how pregnancy problems knocked brain development off course so radically as to cause a severe mental disorder.

That’s where Erwin’s placental and brain organoids come in.

Although the placenta seems like it should be associated with the mother, in fact it comes entirely from the same fertilized egg that eventually creates a baby: The second time the egg divides, the daughter cells start creating this little-studied organ. As a result, it is genetically identical to the fetus and the adult that fetus becomes.

Like other organoid makers, Erwin starts with cells from adults — in her studies, both brain-healthy donors and those who died with schizophrenia. After gently removing a tiny piece of the dura, the tough membrane that covers the brain like shrink-wrap, and teasing apart the dura’s cells, she and her colleagues genetically reprogram them so that they are no longer dura cells but instead revert to an embryo-like state, when they were stem cells able to develop into any of the body’s specialized cells.

The scientists then expose these induced pluripotent stem cells to biochemicals that nudge them down a new developmental pathway: to become placental cells. Just like the macro version, the cells spontaneously organize themselves. “There’s a layered, three-dimensional structure to the organoids,” Erwin said. “It’s not one big blob.”

The genome of a placental organoid is identical to the adult whose cells it was created from. It is therefore also identical to the placenta that person was attached to for nine months.

Erwin grows the organoids in a “shaker,” a little dish that moves gently back and forth inside an incubator so they stay well nourished, she describes in a presentation being prepared for a medical meeting. It takes the induced pluripotent stem cells about 10 days to develop into a placental organoid similar enough to the real thing to mimic its response to outside attack such as a flood of stress hormones.

Once they’re at that point, Erwin stresses them. In one test, she simulates a problematic pregnancy by growing the organoids in air containing 5% oxygen rather than the standard 20%. According to preliminary results, the organoids develop tiny villi, protruding fingers that reach out in search of life-sustaining oxygen — just as a real placenta that’s oxygen-deprived grows villi into the uterine wall. Other planned assaults include a flood of stress hormones, flu viruses, and other factors known to raise the risk of schizophrenia.

Soon, Erwin plans to create cerebral organoids and grow them in the same dishes as placental organoids, each type separated from the other by a membrane that allows biochemicals to pass through. If all goes well, experiments should show how assaults on the placental organoids affect the brain organoids.

Because organoids created from the cells of people with schizophrenia have the same schizophrenia-causing genetic variants as the people they come from, the organoids are ideal ways to test how genes and the environment — in this case, the environment as filtered through the placenta — interact to cause schizophrenia.

“There are very few disorders where genes are the only risk factor,” Weinberger said. To the contrary: The right environment can silence what would otherwise be a disease-causing gene, while the wrong environments don’t merely unleash, but actually exacerbate, genes’ disease-causing effects.

In a 2018 study of 2,038 adults with schizophrenia, Weinberger and his colleagues discovered something odd about genes previously identified as increasing the risk of schizophrenia. “When you look at these genes, they’re not necessarily about the brain,” he said. Instead, they are highly expressed in the placenta, of all things. The more complications in the pregnancy, the higher the level of expression of some of the 350 genes implicated in schizophrenia.

Almost none of the highly-expressed genes are related to the classic functions of neurons, as one might expect of schizophrenia-linked genes. Instead, they affect general cellular functions such as running energy-producing mitochondria and exporting proteins.

Weinberger calls it the first mechanism “connecting early-life complications, genetic risk, and mental illness. And that connection is the placenta.”

If Erwin’s placental/brain organoid pairing succeeds, it should explain how elevated activity of hundreds of genes in the placenta affect brain development, sometimes strongly enough to cause schizophrenia 20 years later. That approach, Weinberger said, “opens up a new chapter in placental medicine.”