Read our COVID-19 research and news.

Researchers prompted cells in mouse retinas to make new photoreceptors (illustrated above, right), which turn light into electrical signals.

Nathan Devery/Science Source

Eye regeneration technique lets blind mice see the light

Nestled in the backs of our eyes, there are cells that might be able to repair damage from some vision-impairing diseases. But so far, scientists haven’t managed to kick them into gear. Now, a team of researchers claims to have prompted these cells, called Müller glia, to regenerate one type of light receptor cell in the eyes of mice. According to their study, published today in Nature, these new cells could detect incoming light and network with other cells in the eye to relay signals to the brain, a potential step toward reversing certain genetic eye conditions and injuries. But others are skeptical of that claim and argue the signals could have come from existing light-sensing cells in the eye—not new ones.

“Nobody more than me wants this to be true,” says Seth Blackshaw, a neuroscientist at Johns Hopkins University’s School of Medicine in Baltimore, Maryland, “but I have serious concerns about this study.”

The new work is part of a long effort to regenerate photoreceptors, neurons in the retina that transform incoming light into electrical signals. Cone receptors are responsible for our daytime vision and perception of colors, and the more sensitive rod receptors enable vision in low light. The destruction of these cells—or of the retinal ganglion cells that transmit their signals to the brain—can diminish vision and even cause blindness.

If people were like zebrafish, that loss wouldn’t be so distressing. The Müller glia in fish and amphibian eyes can divide and specialize into cells that replace damaged or lost neurons. But mammalian eyes don’t spontaneously repair themselves like that. Müller glia support and nourish surrounding cells, but they don’t seem to regenerate neurons except after an injury—and even then, they seem to make a relatively small number of new cells.

Bo Chen, a neuroscientist at the Icahn School of Medicine at Mount Sinai in New York City, and his colleagues hoped to restore photoreceptors without damaging the eye. “We’re trying to wake up the self-repair mechanisms that we know are happening in zebrafish,” he says. That approach could be less invasive and damaging than another treatment under development eye diseases: inserting stem cells into the retina to regenerate neurons.

In previous work, Chen’s team got Müller glia in mouse eyes to divide by injecting a harmless virus containing the gene for a protein that helps regulate how cells proliferate. The glia made daughter cells that resembled stem cells, Chen says, but then they got stuck; they didn’t develop any further.

In the new study, he and colleagues tried a second round of gene transfer 2 weeks after the first, injecting the eyes of healthy mice with three more genes that normally direct cells in the developing eye to become rods. They found that the Müller glia targeted by these gene-carrying viruses generated cells that were rodlike in their structure and in signaling ability.

The researchers then tried the procedure in blind mice, which still had rods and cones, but lacked two key genes that allow those photoreceptors to transmit signals. Alongside the three genes that encourage rod development, researchers also introduced the gene that corrects the signaling defect in rods—so that any newly created ones would be functional. When exposed to light, the treated mice showed activity in the part of the brain that receives visual signals. New rods had apparently wired up to retinal ganglion cells and transmitted their messages successfully, the team concluded.

“No one has made a photoreceptor that looks and functions as much like a photoreceptor as they have,” says Deborah Otteson, a cell and developmental neurobiologist at the University of Houston’s College of Optometry in Texas. But even in the mice that regenerated the most new rods, she notes, their density was just 0.2% of what you’d expect in a healthy mouse retina. As a result, the treated mice probably perceived light, but they couldn’t make out shapes or objects.

“They’ve cracked the first part of the problem, and now it’s a question of amping it up,” Otteson says. If the researchers can get Müller glia to make vastly more photoreceptors, she says, the approach could someday restore some vision to people who have lost rods because of a detached retina or the genetic disorder retinitis pigmentosa. To treat other conditions—including age-related macular degeneration—researchers will have to prompt the Müller glia to regenerate cone cells. Plus, they’ll have to identify and correct whatever genetic mutation underlies a given eye disease.

Maureen McCall, a neurobiologist at the University of Louisville in Kentucky, calls the work “a major step forward” in the effort to restore rods, but notes the team still needs to show that the rods develop and function properly in a diseased eye, where retinal cells might not be connecting and interacting normally.

Blackshaw, however, sees an alternate explanation for the new results: that the existing rod and cones in the blind mice were repaired in the procedure either because they took up the virus carrying the corrective gene, or because Müller glia shared that gene’s products with them. In either case, the visual signals to their brains didn’t come from newly created rods, but from existing photoreceptors with restored function. The study is missing the chemical labeling technique that would prove any functioning rods really came from the Müller glia at all, he says.

Chen maintains that he and his team did such a labeling experiment, though it’s not described in the paper, and that they’ve thoroughly demonstrated the origin of the new rods by several other methods. He also cites control experiments where the group transferred the rod-correcting gene to Müller glia without reprogramming them. In that case, there was no visual signal in the brain, meaning existing rods weren’t restored.

Chen and his team are now exploring other genes that might drive Müller glia to make bigger batches of rods. They’re also gearing up for experiments to test whether their method will also work in human retinal cells, for now, in a lab dish.