The New Neuron in the Brain: A Date Left by the Bomb

Bölüm metni

If a neuron was born in your brain today, how could we possibly know how old it is? The answer came not from a century of anatomy but from the Cold War's atomic tests: carbon scattered into the atmosphere wrote an indelible date into the DNA of every new cell. And once those dates were read, the seventy-year-old rule that the adult brain makes no new neurons began to crack.

In the eighteen nineties, in Madrid, a physician sat at his microscope looking at slices of brain stained with silver salts, and saw something no one before him had seen clearly. Santiago Ramón y Cajal showed that the nervous system is not one continuous web but is built from individual cells that touch one another and yet remain separate from one another. The discovery brought him a Nobel Prize in nineteen oh six and became the cornerstone of modern neuroscience. The same man, with the same rigour, said one more thing; and for a hundred years that one thing was read as a law.

In his book on the degeneration and regeneration of the nervous system, published in nineteen twenty-eight, Cajal wrote that the pathways in the adult nerve centres were now fixed, finished and unchangeable. Everything could die, but nothing could be born again. The sentence entered the textbooks in this short form: you die with the neurons you were born with. Unlike every tissue of the body that renews itself, the brain was taken to run for an entire lifetime on the same roster of cells.

What made this assumption persuasive was that it looked less like a shortcoming than like a design choice. You can replace a liver cell, because the identity of a liver cell is largely generic; whatever takes its place does the same job. The value of a neuron lies in its address. A neuron is the sum of the particular connections it has made with thousands of other neurons, and that pattern of connections is what it means for you to recognise a face or to remember a smell from your childhood. Swapping such a cell for a fresh one would be erasing the memory. And observation supported the idea as well: in adult brain tissue, no dividing cells could be seen.

In nineteen sixty-two, Joseph Altman broke that silence. He had a new tool in his hands: tritiated thymidine. Thymidine is the building block of one of the four letters of DNA; tritium is the radioactive form of hydrogen. When you give this substance to an animal, every cell that is copying its DNA at that moment takes it into its own genome. You then coat a slice of the tissue with photographic emulsion and wait; the radiation leaves dark grains in the emulsion. If you can see those grains over the nucleus of a cell, that cell divided after the injection. Altman found labelled cells in the hippocampus and the olfactory bulb of adult rats, and framed the title of his paper as a direct question: are new neurons being formed in the brains of adult mammals?

The answer was a long silence. There were two reasons for it. The first was technical: under the microscopes of that time, telling a small neuron from a support cell by its shape alone was open to argument, and his critics said that these labelled cells were in fact not neurons but glia. The second was authority; the weight of the field lay on the other side. Altman's line of work found no backing. In nineteen seventy-seven, Michael Kaplan used an electron microscope to show synapses, the connection structures specific to neurons, on labelled cells. Again it was not enough.

In nineteen eighty-five, Pasko Rakic applied the same labelling method to adult rhesus monkeys and reported that the dividing cells were not neurons but glia and vascular cells. The conclusion he reached was a reasonable one: whatever the rat was doing, primates were not doing it. The dogma held, and now it held with data from the animal closest to us.

The crack came from an unexpected direction. Fernando Nottebohm showed that male canaries learn new songs every season, that the brain nuclei governing song grow and shrink over the course of a year, and that new neurons are added to those nuclei in adulthood. It was a bird's brain; it was easy to call it exotic and set it aside. But it reopened the question.

The first direct evidence in humans arrived in nineteen ninety-eight, and the door was opened by a side entrance of medicine. A substance called bromodeoxyuridine is an analogue of thymidine that slips into DNA in its place; in those years it was being given intravenously to some patients with advanced cancer in order to measure how fast a tumour was multiplying. Peter Eriksson, working in Gothenburg, and Fred Gage of the Salk Institute in San Diego asked the families of these patients for permission to examine brain tissue after death. Five patients. In the part of the hippocampus called the dentate gyrus, cells carrying this substance were also carrying the markers of neurons. There were new neurons in the adult human brain.

The dogma cracked. But the crack did not turn into a map. You had five people, all of them gravely ill, and all of them labelled for one single moment in time. How many neurons does a healthy person produce in a day, at what age does that production slow and by how much, and where in the brain does it happen? None of these questions had an answer, because answering them would mean giving healthy people a substance that labels DNA. No one could do that. It would have had to have been done already.

And it had been done. Without anyone's consent, to the entire planet.

In the years when nuclear weapons were being tested above ground, the neutrons released by the explosions struck nitrogen nuclei in the atmosphere and turned them into carbon fourteen, the radioactive form of carbon. Carbon fourteen exists in nature as well; cosmic rays produce it continuously, and its level in the air remained very nearly constant for thousands of years. The tests broke that balance. During the intense testing period between nineteen fifty-five and nineteen sixty-three in particular, the proportion of carbon fourteen in the air over the northern hemisphere rose to roughly twice its natural level, and reached its highest point in nineteen sixty-three.

Then, with the treaty signed in Moscow on the fifth of August, nineteen sixty-three, which came into force on the tenth of October, the major powers stopped testing in the atmosphere, under water and in space. The curve turned downward from its peak. The excess carbon fourteen in the air began to fall by roughly four percent a year. Note that this fall is not radioactive decay; the half-life of carbon fourteen is five thousand seven hundred and thirty years, which means that on the scale of a human lifetime it barely diminishes at all. The reason for the decline was that the oceans and the vegetation were drawing that carbon out of the air and taking it into themselves. What was left behind was a sharp peak followed by a long, smoothly descending tail. In other words, into the air of every year a carbon fourteen value belonging to that year alone was written.

Plants take in the carbon dioxide of the air exactly as they find it. We eat plants, and animals that feed on plants. So the carbon that enters our bodies carries the signature of that year's atmosphere. The idea standing at the end of this chain belonged to Jonas Frisén at the Karolinska Institute in Stockholm, and it was very simple: almost every molecule in a cell is continuously renewed, but the DNA in the genome is not. The carbon of DNA settles into place at the moment the cell copies its genome, and it stays there. In that case, the proportion of carbon fourteen in a cell's DNA is that cell's date of birth. Brain samples supplied by the forensic specialist Henrik Druid went to the accelerator mass spectrometer operated by Bruce Buchholz at the Lawrence Livermore National Laboratory in California, the Livermore laboratory for short. The precision of the measurement was between one and a half and two years. The hard part of the work was separating the neurons from the support cells; for that, the nuclei were sorted out one by one with flow cytometry, on the basis of a protein found only in neurons.

The first answer came in two thousand five, and it looked like a disappointment. The carbon in the DNA of the neurons of the cerebral cortex corresponded to the air of the person's own year of birth. That is to say, the cortex with which you think, with which you decide, with which you speak, was exactly as old as you were. For the brain's largest region, Cajal was right.

In twenty thirteen the same method was turned on the hippocampus; tissue from fifty-five people between the ages of nineteen and ninety-two was measured. This time the result was different: the DNA of hippocampal neurons carried air from later than their owners' year of birth. These cells were younger than the people they belonged to. The model that emerged from the data said this: around seven hundred new neurons are added each day to each of the two hippocampi of an adult human being. That amounts to one point seven five percent of the renewing pool of cells turning over each year, and roughly a third of the neurons in the dentate gyrus belong to that pool. There is a decline with age, but a surprisingly measured one; the hippocampus of an eighty-year-old was still adding.

The scale has to be stated honestly: seven hundred cells a day, set beside tens of millions of neurons, is not a reconstruction. It is a leak. But spread across decades, it is a leak large enough to replace the greater part of one particular population of cells.

This looked like the close of the matter. It was not.

In the spring of twenty eighteen, a study came out of Arturo Alvarez-Buylla's laboratory at the San Francisco campus of the University of California, carried out by Shawn Sorrells. Thirty-seven human hippocampi, ranging from the foetal period up to the age of seventy-seven, had been examined. The result was clear: the number of young, maturing neurons fell rapidly during childhood, and was not present at detectable levels in any sample older than thirteen. What was striking was whom it came from; in the early years of his career, Alvarez-Buylla had been part of the work that showed new neurons in the bird brain.

A few weeks later, Maura Boldrini's team at Columbia University in New York reported exactly the opposite. In healthy people between the ages of fourteen and seventy-nine, they found maturing neurons and their progenitor cells in every age group, and what is more, they saw no clear age-related decline in their number.

The same brain region, the same marker proteins, the same year, opposite results. This was not a difference of opinion ripening slowly; it was a collision. And when two careful teams measure the same thing and find the reverse of each other, the disagreement is usually not about the brain. It is about what happens to the tissue between death and the microscope.

Because the protein on which these counts rest is fragile. Brain tissue is fixed with formaldehyde so that it will not decay. The longer the fixation lasts, the more the face of the protein that the antibody recognises is masked; beyond a certain point the protein becomes invisible even when it is there. To this are added dozens of further variables: the time that passes between death and the processing of the tissue, whether the tissue was flushed through the blood vessels or immersed in fluid, how long it waited in the jar, how it was frozen. Not one of the brain banks in the world had designed any of these conditions for the purpose of counting newborn neurons.

In twenty nineteen in Madrid, María Llorens-Martín's team placed that variable at the centre of the equation. They worked with tissue under their own control, briefly and uniformly fixed. In all thirteen neurologically healthy donors between the ages of forty-three and eighty-seven, they found maturing neurons. Then they ran the decisive control: they deliberately left matched samples in formaldehyde for longer, and watched the signal die away. The absence of a signal could turn out to be the work of the jar rather than of the brain. The same study also showed that the number falls markedly in Alzheimer's disease, and that it falls at an early stage of it.

In those same years in Chicago, Orly Lazarov's team looked at the donors of a long-running study of ageing whose participants had been followed with cognitive tests throughout their lives; eighteen people with an average age above ninety. All of them had neural stem cells and young neurons, but the numbers varied enormously from person to person. More importantly, that variation went together with cognition: in those whose minds had stayed clear the count of young neurons was high, and in those with mild cognitive impairment it was low.

Even so, the weakness common to all of these counts was the same: you are inferring the youth of a cell by looking at one or two proteins. In twenty twenty-five, Frisén's team changed the instrument. Single-nucleus RNA sequencing, a method that lists directly which genes a cell is reading at a given moment. First they read more than a hundred thousand nuclei from children between birth and the age of five; the aim was to learn what the gene activity signature of a genuine human hippocampal progenitor cell looks like. They taught that signature to a machine learning model. Then they applied the model to more than two hundred thousand nuclei taken from people between the ages of thirteen and seventy-eight. In adult tissue the model found both progenitor cells and maturing neurons, all of them gathered in the dentate gyrus, and some of them carrying the signature of actively dividing cells. They were no longer looking at a single protein but at an entire profile of what the cell was reading. And here the same thing was seen: a large difference between individuals. Abundant in some adults, almost absent in others.

The real question of the field shifted here as well. The matter is no longer whether it exists or not; it is what determines the amount, and whether that amount means anything.

In February of twenty twenty-six, Lazarov's team gave the most interesting answer to that question. This time the group examined were people who had been followed for years and who, in spite of being past eighty, performed on memory tests at the level of people twenty to thirty years younger than themselves; super-agers. The team did not look only at which genes were being read; they also looked at chromatin accessibility, that is, at which parts of the DNA stand open to being read. In the hippocampus of the super-agers there were markedly more young neurons and progenitor cells than in their normally ageing peers, and what drove this was a regulatory pattern of its own. In Alzheimer's patients the picture was reversed: the stem cells were in place, indeed increased in number, but the steps that follow were blocked. Not an emptied factory, then, but a production line that had come to a halt. What is more, the epigenetic traces of that blockage appeared before the changes at the level of gene reading, which makes it a candidate for an early warning marker.

A limit has to be drawn clearly here. These are associations. Post-mortem tissue is a single frame taken out of a lifetime; from that frame we cannot say whether more new neurons kept the mind clear, or whether a mind that stayed clear kept the production of new neurons going. In the same way, the widespread claims that running, or learning, or certain antidepressants increase the production of new neurons in human beings are largely inferred from rodent experiments; it is a reasonable expectation, not a result proven in people. To say it as a possibility is correct; to say it as knowledge would be wrong.

One last detail, concerning the instrument of the story. The calendar the bomb left behind is being erased. The proportion of carbon fourteen in the air has gone on falling since the peak of nineteen sixty-three and has come close to its natural level; and on top of that, the carbon released into the atmosphere by the burning of fossil fuels, which contains no carbon fourteen at all, is diluting the proportion further still. The curve is flattening out. In the coming decades it will not be possible to read a cell's year of birth by this method. The weapons tests of the nineteen fifties left behind an archive, without intending to, in every body that was alive in those years, and the window in which that archive can be read is closing. The neurons that will be born tomorrow in the hippocampus of a person born today will be cells with no date written upon them.

Cajal's sentence did not turn out to be exactly wrong; it narrowed. For the wide, thinking cortex of the brain it still appears to hold. But in one single structure, in the form of a leak of a few hundred cells a day, there is a place that falls outside that sentence. And the only reason we are able to see it is that humanity, for eight years, entirely without meaning to, wrote a date upon everything that was alive at the time.

Nöron Science'de bugünkü bölümün sonuna geldik. Yeni bir hikâyede yeniden buluşmak üzere.