The mushroom clouds produced by more than 500 nuclear bomb tests during the Cold War may have had a silver lining, after all. More than 50 years later, scientists have found a way to use radioactive carbon isotopes released into the atmosphere by nuclear testing to settle a long-standing debate in neuroscience: Does the adult human brain produce new neurons? After working to hone their technique for more than a decade, the researchers report that a small region of the human brain involved in memory makes new neurons throughout our lives—a continuous process of self-renewal that may aid learning.
For a long time, scientific dogma held that our brains did not produce new neurons during adulthood, says Pasko Rakic, a neuroscientist at Yale University who was not involved in the study. In 1998, however, a group of Swedish researchers reported the first evidence that neurons are continually born throughout the human lifespan. The researchers injected a compound normally used to label tumor cell division into patients who had agreed to have their brains examined after death. When the scientists examined the postmortem brain tissue, they found that new neurons had indeed sprung forth during adulthood. The cells were located in a part of the hippocampus—a pair of seahorse-shaped structures located deep within the brain and involved in memory and learning. The compound was later found to be toxic, however, and the experiment was never repeated.
Since 1998, a number of studies have demonstrated that new neurons are generated in the same small region of the hippocampus in mice and appear to play an important role in memory and learning, says Kirsty Spalding, a molecular biologist at the Karolinska Institute in Stockholm and lead author of the new study. Because the 1998 work was never confirmed by independent research, however, scientists have fiercely argued over whether the neuron birth seen in mice also occurs in people.
More than 10 years ago, Spalding's adviser, Jonas Frisén, a stem cell researcher at the Karolinska Institute and study co-author, urged her to take on a project aimed at settling this debate  by using an unconventional approach. The method, which has taken Spalding more than a decade to develop, hinges on a massive pulse of radioactive carbon-14 isotopes released by nuclear explosions in the 1950s and '60s, which doubled the amount of carbon-14 in the atmosphere. This pulse stopped with the Limited Test Ban Treaty of 1963, which banned aboveground tests of nuclear weapons, and the unstable carbon-14 isotopes have steadily decayed. Because cells incorporate carbon from the atmosphere into their DNA as they divide, the proportion of carbon-14 to the more stable carbon isotope carbon-12 acts as a time stamp for when a cell was born.
Spalding has been using this ratio to determine the age of teeth in forensic investigations and the turnover rate of  fat cells . But she had to improve the sensitivity of the technique so that it could detect the isotopic ratio in DNA from the roughly 6-gram sliver of neural tissue in the hippocampus thought to produce new neurons, the dentate gyrus. At best, the isotope is present in only one out of every 15 neurons, she says, making it difficult to detect in small amounts of tissue.
For the first 5 years, Spalding worked on finding an effective way of separating the roughly 20 million neurons in the dentate gyrus from other types of hippocampal cells and then extracting their DNA. Discovering that she could use a fluorescence-activated cell sorting machine to distinguish non-neuronal cells from neurons by making them glow in different colors was "a high point," she says. The next 5 years were largely spent on finding ways to purify the DNA samples and extract and analyze the carbon atoms using high-powered particle accelerators. "We had many years without any results," Frisén says. "It was fun, but frustrating."
After finally getting the technique down pat, Spalding decided that it was time to try it on some real human brain tissue. She and her colleagues extracted hippocampi from 55 deceased people who had given informed consent to have their brains studied. They then ground up the tissue samples, sorted the cells, and extracted the DNA. Next, she sent the purified genetic material to the Lawrence Livermore National Laboratory in California, where it was reduced to pure carbon pellets and split into different carbon isotopes by weight in a particle accelerator, allowing the researchers to calculate the ratio between carbon-12 and carbon-14.
Spalding, Frisén, and colleagues then created a mathematical model estimating, based on those ratios, the rate of cellular turnover within the hippocampal neurons. More than a third of hippocampal neurons were regularly replaced, with roughly 1400 new neurons added each day during adulthood , they report online today in Cell. "Some cells are dying, some are being replaced," Spaulding says. "There is a constant flux of life and death."
"This is a spectacular independent confirmation" of the 1998 study suggesting that new neurons are born during adulthood in the dentate gyrus, writes Gerd Kempermann, a neuroscientist at the German Center for Neurodegenerative Diseases in Dresden, in an e-mail. "It will likely settle the case."
Kempermann says that his own and other's studies in mice indicate that fresh adult neurons have a specific function in the hippocampus—for example, in helping the brain distinguish between things that belong to the same category, or comparing new information to what it has already learned from experience. The ability to distinguish between the Beatles and Rolling Stones, yet still identify both as "rock bands," is one example of this type of task in humans, Frisén says.
There is another possibility, however: Our ability to replace hippocampal neurons could be an evolutionary vestige that is not all that important today, Rakic says. He argues that human survival may have depended not so much on our ability to produce new neurons, but on our ability to keep old ones in order to accumulate memories over the entire lifespan. Compared with fishes, frogs, reptiles, and birds, some of which can regrow entire brain structures, he says, "it is interesting that neuronal turnover in humans is limited to a single population of neurons in only one relatively small structure, and it is worthwhile to examine why it persists."