New neurons (green) were visible in a newborn's brain (top), but not in brain tissue from a 35-year-old man (bottom).

KEN PROBST

Study questions whether adults can really make new neurons

Over the past 20 years, evidence that adult humans can produce hundreds of new neurons per day has fueled hope that ramping up cell birth could be therapeutic. Boosting neurogenesis, researchers speculate, might prevent or treat depression, Alzheimer's disease, and other brain disorders. But a controversial study in Nature this week threatens to dash such hopes by suggesting that the production of neurons declines sharply after early development and grinds to a halt by adulthood.

The results of the "exhaustive search" for new neurons in adult human and monkey brains "will disappoint many," says neuroscientist Paul Frankland of the Hospital for Sick Children in Toronto, Canada. "It raises concern that levels of neurogenesis are too low to be functionally important" in humans, adds another observer, René Hen, a neuroscientist at Columbia University. But he and others suggest that the study left much room for error. The way the tissue was handled, the deceased patients' psychiatric history, or whether they had brain inflammation could all explain why the researchers failed to confirm earlier, encouraging studies, Hen says.

The first evidence of neurogenesis in adult humans came in 1998 from the brains of deceased cancer patients who had received injections of a chemical called bromodeoxyuridine while they were still alive. The chemical labels newly divided cells, and in their brain tissue, it showed up in a sprinkling of neurons in the hippocampus—a seahorse-shaped structure involved in memory and learning. In 2013, Jonas Frisén's lab at the Karolinska Institute in Stockholm buttressed the case by carbon dating individual neurons in brain tissue from 55 deceased people. From the cells' ages, the group calculated that every day, humans replace 700 of their neurons in the dentate gyrus, a sliver of hippocampus thought to encode memories.

Arturo Alvarez-Buylla of the University of California, San Francisco, who has been studying the brain's capacity to produce new cells since the 1980s, was skeptical. He is known for showing how niches of neural stem cells constantly regenerate parts of the rodent brain. But the carbon-dating evidence did not persuade him that people maintain similar stem cell reserves: The method involves "a lot of assumptions and steps in which there can be contamination or false positives," he says.

For the new analysis, his team spent 5 years collecting brain tissue from 59 people who had died or had such tissue removed during surgery for epilepsy at different ages, ranging from before birth to 77 years of age. They used fluorescent antibodies to label proteins specific to cells at different states of maturity. With an electron microscope, they also looked for the characteristic long, slender, simple shapes of young neurons.

The team found that people have large numbers of neural stem cells and progenitors early in life—an average of 1618 young neurons per square millimeter of brain tissue at birth. But these cells did not go on to form a proliferating layer of neural stem cells, and production of new neurons dropped 23-fold between 1 and 7 years of age, the team reports. By adulthood the supply of young neurons had petered out entirely. "We just don't see what other people are claiming" in adults, Alvarez-Buylla says.

Frisén counters that the antibody markers used are far from reliable, because background fluorescence can muddy the results. He adds that other teams using the same techniques have seen adult neurogenesis. "I think this debate will rage on," Frankland says. "A lot rests on this."