https://www.nature.com/articles/s41591-019-0375-9

Reports of old brains’ decrepitude have been greatly exaggerated, scientists reported on Monday, unveiling results that contradict a much-discussed 2018 study and instead support the idea that human gray matter is capable of generating new neurons up to the ninth decade of life.

The research, published in Nature Medicine, also found that old brains from people without dementia have much higher rates of such neurogenesis than do the brains of people with Alzheimer’s disease, offering a new clue to a field that is desperate for new ideas.

When scientists at the University of California, San Francisco, reported last year that they “couldn’t find a single new neuron” in human brains older than about 7, as one of the lead researchers told STAT, and that the rate of neurogenesis falls off a cliff during the first year of life, they were greeted skeptically. Their conclusion contradicted 20 years of research in rodents as well as humans showing that the adult brain can give birth to new neurons in the memory-forming structure called the hippocampus. (Claims of neurogenesis in other brain regions, in contrast, have not stood up.)

Critics offered numerous explanations for why the UCSF team saw no neurogenesis, including the difficulty of using the molecular markers of neuron birth. Some of those markers degrade very quickly, for instance, meaning they might have vanished from the brains used in the study (donated to brain banks) before the scientists could detect them.

Because the UCSF study was published in a prominent journal (Nature) by a respected lab, it was not rejected out of hand. But the new research suggests that the critics’ hunch was right: The molecular markers of neurogenesis are as tricky to work with as soufflés in an earthquake.

In particular, said molecular neuropathologist María Llorens-Martín of Spain’s Universidad Autónoma de Madrid, who led the new study, the marker called DCX does very poorly during tissue fixation, the step when minuscule slices from a donated brain are prepared for analysis. “If you go over about 12 hours of fixation time, the markers of new neurons disappear,” she said. “The new cells are there but we cannot detect them.”

She knew new neurons were present because after not finding them in samples that had undergone a long fixation time, she and her colleagues looked again in samples from the same brain that were fixed quickly. Those, she said, showed clear molecular signs of having undergone neurogenesis until right before death — regardless of the age when that occurred — including the presence of a protein called doublecortin, which is associated with neuronal migration.

“We clearly demonstrated that if you go longer than 12 hours” of fixation time, “you lose the signal of neurogenesis,” Llorens-Martín said. She and her colleagues analyzed samples from 58 banked brains, with ages at death from 43 to 87.

In response, Shawn Sorrells, a lead author of the 2018 study failing to find adult neurogenesis, told STAT that although the new study “contains valuable data,” he and his co-authors did not find its evidence for ongoing neurogenesis “convincing.” Accurately identifying new neurons “is a complicated endeavor requiring multiple lines of evidence to rule out alternative explanations, none of which are presented in this study,” said Sorrells, now at the University of Pittsburgh.

He believes that what the Madrid scientists took to be new neurons “are actually a distinct set of mature hippocampal neurons that have been there since childhood.” The supposed new ones have a size and shape more like that of mature neurons, Sorrells added.

The Madrid group disagrees. And their analysis meshes with the more widely held view that healthy, adult brains produce new neurons, but at a rate that declines with age.

In a 43-year-old brain, for instance, the scientists measured roughly 42,000 new neurons per cubic millimeter of hippocampus (approximately the volume of nine grains of table salt). An 87-year-old had 20,000 new neurons per cubic millimeter.

Alzheimer’s disease was even worse for neurogenesis than age. One 78-year-old brain from a person who didn’t have Alzheimer’s had about 23,000 new neurons per cubic millimeter of hippocampus, while a brain of the same age but from someone who died with Alzheimer’s disease had 10,000.

That might solve a longstanding Alzheimer’s mystery: that some elderly people who died with brains full of amyloid showed no signs of dementia. That suggests some sort of resilience or compensatory mechanism: Inflammation, which more and more scientists believe causes Alzheimer’s, kills neurons, but high rates of neurogenesis supply replacements.

Although Llorens-Martín and her colleagues did not have enough brains of the same age to make definitive comparisons of individual neurogenesis rates, they saw hints of person-to-person variation. Among people in their 60s without Alzheimer’s, she said, rates of neurogenesis ranged from about 30,000 to 40,000 new neurons per cubic millimeter of hippocampus; in 80-somethings, it was 20,000 to 30,000.

“If you can increase the rate of neurogenesis, it might be protective” against Alzheimer’s, she said.

Research in both lab animals and people has found that aerobic exercise can increase the rate of neurogenesis in some brain regions; so can some antidepressants. Scientists are trying to figure out whether new neurons do any functional good in people, as research shows they do in rats. Llorens-Martín is planning experiments on that, she said, “but from their morphology and other signs, I would bet they’re doing something.”