Light microscope micrograph of the brain white matter stained with hematoxylin-eosin showing, in the center, a row of rounded nuclei belonging to oligodendrocytes. (Credit: © JosLuis - stock.adobe.com)
More Brain Cells Are Not Always Better, Aging Study Suggests
In A Nutshell
- Brain cells that coat nerve fibers were more plentiful in older adults whose thinking declined faster than average.
- Those cells showed signs of low activity in NRF2, a protein that helps cells clear waste and defend against damage.
- Mice with NRF2 switched off in these cells gained about 28% on a memory test, versus about 58% for normal mice.
- Human brain exams covered only 13 to 21 people and no drug was tested, so NRF2 is a lead, not a treatment.
Some people stay mentally sharp into their 80s, while others lose ground far faster, and the reasons have long been unclear. Cognitive decline affects quality of life, and the paper notes that treatments are lacking. A study in Nature Medicine now points to an unlikely suspect: helper cells that coat the brain’s nerve fibers, its wiring. Those cells were more plentiful, not scarcer, in people whose thinking skills declined faster than average.
Called oligodendrocytes (oh-LIG-oh-DEN-droh-sites), they wrap nerve fibers in a fatty layer named myelin, much like the plastic around an electrical wire, so signals travel quickly. Earlier research in mice tied learning to the creation of new cells of this kind, which made the surplus a surprise.
In their paper, the authors say these cells are not always supportive of brain function and cognition, and that they “can become actively detrimental in aging.” That complicates the idea that more of them is always better.
Nearly All 866 Older Scots Showed Cognitive Decline, but at Different Speeds
Researchers drew on the Lothian Birth Cohort 1936, which the paper calls “the longest-running study of human cognition.” Everyone in it was born in 1936 and lived in Scotland’s Lothian region when recruited. Participants were mostly of white European ancestry, men and women alike. Most had taken an IQ test at age 11, then were retested at 70 and roughly every three years until 82 on memory, mental speed and spatial skills.
Of 866 people retested after 70, all but one slipped. The team split them by whether their decline ran faster or slower than the group average. Childhood IQ predicted how people scored at 70, yet it did not predict which side of the average they landed on, suggesting a sharp mind at 11 says little about how quickly thinking fades afterward.
Some participants donated their brains after death. The team focused on a thick band of nerve fibers linking the brain’s two halves, but the detailed exams covered only 13 to 21 people, depending on the test, a small slice of the full group. That limited size is why the human findings need confirmation in larger studies.
Faster Cognitive Decline Came With Thicker Myelin on Large Nerve Fibers
Under the microscope, the brains of faster decliners held a surprise. Thinking problems are usually linked to thinning myelin, yet here the large nerve fibers wore unusually thick coats, and fibers overall ran smaller. Fiber size and myelin thickness are thought to largely set how fast signals travel, so the mismatch could plausibly slow the brain’s messaging. Thicker wrapping sounds protective, but the authors cite evidence that excess myelin precedes nerve damage in other conditions, including multiple sclerosis and aged monkeys.
Faster decliners also had more empty myelin sleeves with no nerve fiber left inside, which can signal early breakdown. None of this traced back to age at death, sex, childhood IQ or damage to tiny blood vessels. Smaller fibers did line up with Alzheimer’s-related changes, though.
Extra Oligodendrocytes Showed Lower Activity of the Protein NRF2
To find out what was happening inside the cells, the team analyzed genetic activity in 14 brains. Oligodendrocytes were the only major cell type that increased with faster decline, a result confirmed by staining tissue from 18 people.
Two subtypes of these cells showed signs of low activity of NRF2, a protein that helps cells clear waste and defend against damage. Earlier work showed that mice lacking NRF2 throughout the body have worse age-related memory problems, but the responsible cell type was unclear. This study points to oligodendrocytes. Staining backed this up: a smaller share of oligodendrocytes carried NRF2 in faster decliners than in slower ones. A reasonable guess, supported by earlier mouse data, is that overcrowding comes first and NRF2 drops as a result.
Mice Missing NRF2 in These Cells Improved Less on a Memory Test
Association is not proof, so researchers switched off NRF2 only in the oligodendrocytes of mice. At 12 months old, when mice start showing learning and memory slips, 38 males faced a water maze that rewards remembering where a hidden platform sits, guided by visual cues on the walls. Female mice were left out of this analysis after showing differences in anxiety-like behavior. All improved over four days, but mice missing NRF2 started ahead and gained less: normal mice spent about 58% more time in the platform’s area by the end, versus about 28% more for the NRF2-deficient group.
Mice that performed worse had narrower fibers and thicker myelin, echoing the human pattern, and mice missing NRF2 showed the same pattern by 18 months, even without extra oligodendrocytes, suggesting that losing the protein alone can reshape the wiring.
NRF2 can be switched on by drugs. One, dimethyl fumarate, has improved cognition in rodents and in multiple sclerosis research, so repurposing such drugs is an idea worth testing.
Why some brains fade faster than others has long lacked an answer, and these helper cells now offer a credible lead. Cause-and-effect evidence comes only from mice, and no drug was tested here, so NRF2 is a lead for researchers, not yet a treatment for anyone.
Disclaimer: This article summarizes a peer-reviewed study for general readers and is not medical advice. Anyone with concerns about memory, aging or treatment options should speak with a qualified healthcare professional.
Paper Notes
Limitations
Human brain analyses relied on postmortem tissue from small subsets of the cohort, with 13 to 21 people per structural measurement and 14 for gene activity sequencing. The cohort is Scottish and mostly of white European ancestry, so results may not apply to other populations. The researchers acknowledge that clinical factors and tissue processing can confound human tissue analyses. Their human findings are associations, which cannot prove cause, and the experiments pointing to a causal role were done in mice, which do not perfectly replicate human aging or cognition. Female mice were excluded from the cognitive analysis because of genotype-specific differences in anxiety-like behavior, so those results cover males only. Smaller nerve fiber size was linked to Alzheimer’s-related pathology, so overlap between the two cannot be ruled out. The link between extra oligodendrocytes and lower NRF2 relied partly on a previously published mouse dataset. Only the corpus callosum was examined, and the authors note that other brain regions may reveal additional contributors. The exact way NRF2 regulates myelin thickness is unknown, and so is whether the increase in oligodendrocytes reflects new cells made late in life or accumulation from earlier periods.
Funding and Disclosures
This study received support from the John David Eaton Chair in Multiple Sclerosis Research and the Hall-Sloan Multiple Sclerosis Basic Science Research Fund (both from St. Michael’s Hospital Foundation), the Canadian Institutes of Health Research (grant 203844), and the UK Medical Research Council Senior Non-Clinical Fellowship (MR/V031260/1). Salary support for one author came from a Deighton Postdoctoral Fellowship in Multiple Sclerosis from Unity Health Toronto. Animal work was supported by the National Institutes of Health (grants 4R01AG072080 and 5R35NS137478) and the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation. Cognitive assessments and human brain tissue processing were funded by a Sir Henry Dale Fellowship from the Wellcome Trust and the Royal Society (221890/Z/20/Z) and the UK Dementia Research Institute (grant UK DRI-4204). The Lothian Birth Cohort 1936 studies received support from the Biotechnology and Biological Sciences Research Council, the Economic and Social Research Council, Age UK, the Milton Damerel Trust, the Medical Research Council, and the University of Edinburgh. The funders had no role in study design, data collection and analysis, the decision to publish, or preparation of the manuscript. Over the past 10 years, corresponding author Véronique E. Miron has received payments for consulting, scientific talks, or collaborative research from Biogen, Novartis, Roche, Astex Pharmaceuticals, Clene Pharmaceuticals, Cognito Therapeutics, and Myrobalan Therapeutics. Co-author Tara L. Spires-Jones has received such payments over the same period from AbbVie, Sanofi, Merck, Boehringer Ingelheim, Scottish Brain Sciences, Jay Therapeutics, Cognition Therapeutics, Ono Pharmaceutical, and Eisai, is a charity trustee for the British Neuroscience Association, and serves as scientific advisor to several charities and nonprofit institutions. All other authors declared no competing interests.
Publication Details
Paper title: Oligodendrocyte dysfunction in human age-related cognitive decline | Authors: Georgina A. Craig, Emeric Merour, Luise A. Seeker, Annette J. Haughian, Keon Arbabi, Jamie Rose, Alana Hoffmann, Jessica Thapar, Andrea Corsinotti, Stephen Mitchell, Janie Corley, Susan D. Shenkin, Colin Smith, Tara L. Spires-Jones, Lu O. Sun, Simon R. Cox, Anna Williams, Shreejoy J. Tripathy, Brian Popko, and Véronique E. Miron. | Journal: Nature Medicine | Published: August 25, 2026 | DOI: https://doi.org/10.1038/s41591-026-04608-y
