Researchers have converted astrocytes (red) into neurons (green) in a living mouse brain.

ZHENG WU/GONG CHEN'S LAB

Reprogrammed cells could tackle brain damage

SAN DIEGO, CALIFORNIA—If a diseased or injured brain has lost neurons, why not ask other cells to change jobs and pick up the slack? Several research teams have taken a first step by "reprogramming" abundant nonneuronal cells called astrocytes into neurons in the brains of living mice.

"Everybody is astonished, at the moment, that it works," says Nicola Mattugini, a neurobiologist at Ludwig Maximilian University in Munich, Germany, who presented the results of one such experiment here at the annual meeting of the Society for Neuroscience last week.

Now, labs are turning to the next questions: Do these neurons function like the lost ones, and does creating neurons at the expense of astrocytes do brain-damaged animals any good? Many researchers remain skeptical on both counts. But Mattugini's team, led by neuroscientist Magdalena Götz, and two other groups presented evidence at the meeting that reprogrammed astrocytes do, at least in some respects, impersonate the neurons they're meant to replace. The two other groups also shared evidence that reprogrammed astrocytes help mice recover movement lost after a stroke.

Some see the approach as a potential alternative to transplanting stem cells (or stem cell–derived neurons) into the damaged brain or spinal cord. Clinical trials of that strategy are already underway for conditions including Parkinson's disease and spinal cord injury. But Gong Chen, a neuroscientist at Pennsylvania State University in State College, says he got disillusioned with the idea after finding in his rodent experiments that transplanted cells produced relatively few neurons, and those few weren't fully functional. The recent discovery that mature cells can be nudged toward new fates pointed to a better approach, he says. His group and others took aim at the brain's most abundant cell, the star-shaped astrocyte.

Astrocytes are glial cells, named for the misconception that they're merely the brain's structure-giving "glue." In fact, they nourish and communicate with neurons and help control blood flow. After an injury, subsets of astrocytes proliferate, promote inflammation, and contribute to the formation of a scar. Many scientists think astrocytes' effects on recovery are contradictory—some helpful and some harmful.

"I cannot imagine another technology to be more efficient than using the neighboring glial cell" to repair the brain, Chen says. His group enlisted a harmless virus that, injected into the brain, infects astrocytes and introduces DNA that codes for NeuroD1, a transcription factor that activates genes typically expressed in neurons. The reprogramming apparently prompts other astrocytes to multiply, which he thinks might prevent the treatment from depleting the brain of astrocytes.

The approach, under development in several labs working with various transcription factors, is "super provocative," says Timothy Murphy, a neuroscientist at The University of British Columbia in Vancouver, Canada, who studies how brain circuits change after stroke. But, he adds: "These cells need to survive, and they need to reconnect."

No group has yet shown that the reprogrammed cells do wire up into circuits to carry out the functions of lost neurons. But several have evidence that the cells take on key neural features. In the weeks after inducing a stroke in a mouse's brain, Chen's team saw reprogrammed astrocytes retract some of their starlike tendrils and begin to produce hallmark neural proteins. Reprogrammed astrocytes also appear to fire electrical signals and extend new fibers across the brain and into the spinal cord.

Götz's team, meanwhile, documented that newly reprogrammed neurons around the site of a stab wound resemble pyramidal neurons, which send excitatory signals. (Her group, like others, is now teasing out how different combinations of transcription factors prompt astrocytes to become different types of neurons.) The researchers also found that newly reprogrammed neurons express different markers and send out different projections depending on which layer of the cortex they are in, just as native neurons do.

That's "very surprising," says Chun-Li Zhang, a neuroscientist at the University of Texas Southwestern Medical Center in Dallas. He is exploring a different reprogramming process, which turns astrocytes into primitive neural progenitor cells that then become neurons more gradually. Both approaches will have to overcome skepticism, he says. Many researchers don't expect neural newcomers, introduced abruptly into the adult brain, to mature and function normally.

"To really convince people," he says, "we need to be really careful" to document the steps in the transformation of these cells—and to prove that they begin as astrocytes and finish as mature neurons.

Researchers have also begun to look for indications that the approach helps animals heal. In a study posted in April on the preprint server bioRxiv, Chen's group reported that reprogrammed cells improved a mouse's ability to walk and use its front limbs after a stroke. At the meeting, he hinted that the same approach had restored neural tissue in the brains of stroke-injured monkeys; experiments to gauge their recoveries are ongoing at a collaborator's facility in China, he says.

Chen has founded a company to develop therapies with astrocyte reprogramming, including a cocktail of small molecules that could reprogram cells without brain surgery or the use of a virus. "I believe this is the future," he told the audience at his conference presentation. "It's the next frontier in regenerative medicine."

Stem cell biologist Cindi Morshead of the University of Toronto (U of T) in Canada is more circumspect. Scientists don't fully understand the role of astrocytes in the brain after an injury, she says, but "they're there for a purpose." As her group prepared to test the strategy, she expected it to make injured animals worse.

She's more optimistic now. At the conference, her U of T collaborator Maryam Faiz revealed that mice injected with NeuroD1 a week after a stroke recovered motor function more quickly than untreated mice, some of which were permanently disabled. By 2 months after the treatment, mice performed about as well as healthy controls on walking tests. Fully 20% of their neurons were reprogrammed cells.

The results in stroke are among the first glimmers of benefit. Last year, Swedish researchers also reported that they had restored some motor function in a mouse model of Parkinson's disease by reprogramming astrocytes into dopamine-producing neurons.

Morshead's results have encouraged her to continue experimenting. She now wants to wait longer after a stroke to inject her mice. Once stroke disability becomes chronic in humans, "we have nothing for them," she says. If long-disabled mice benefit from their new neurons, she says, "now, that would be the coolest thing."