A section of brain from a woman who died at 69, obtained from a tissue bank in Belmont, Massachusetts, was placed under fluorescent probes for the X and Y chromosomes. Some of the cells in it carried a Y. She never had one.

That specimen was one of 59 examined for a 2012 paper in PLOS ONE, which reported male DNA in the brain tissue of 37 of them. The oldest woman in whom it was detected was 94.

The study, Male Microchimerism in the Human Female Brain, was led by William F. N. Chan, then at the Fred Hutchinson Cancer Research Center, with J. Lee Nelson as senior author. It was the first published search for fetal cells in the human brain. The women had died between the ages of 32 and 101. None of us are researchers or clinicians, and what follows is a reading of the published work.

This is one study, not settled consensus. Pregnancy history was documented for only eleven of the 59 women, too thin a base for testing how long anything persists, and the follow-up work that solved that problem was done on brain tumours.

What the paper measured, and what it inferred

Chan and colleagues extracted DNA from specimens held at the Department of Pathology at the University of Washington and the Harvard Brain Tissue Resource Center, then ran quantitative PCR targeting DYS14, a sequence found only on the Y chromosome. Male DNA in a woman’s brain has to have come from somewhere. The most likely source, in their reading, is a pregnancy with a male fetus.

The quantities were small. Across all subjects, the median concentration was 0.2 male genome equivalents per 100,000 total genomes, and the ninetieth percentile was 3.7. Of 183 individual specimens tested, 64 came back positive, and those positives were spread across multiple regions, including the temporal lobe, the cingulate gyrus and the pons. The medulla had the highest prevalence of the regions sampled. Neither the frontal lobe nor the putamen returned any positives, though very few specimens from either were available.

One subject stands apart, and she is the woman from Belmont. She appears in the paper as B6388. Her three positive specimens measured 296.1, 481.8, and 512.5 genome equivalents per 100,000, roughly ten times the next highest reading in the dataset, and hers is the only case in which the researchers looked for the cells themselves. The probes were applied to a section of her pons, and rare male nuclei showed up.

DNA, cells, and the difference between them

That distinction matters more than most write-ups of this research allow. PCR detects a sequence. It does not, on its own, establish that an intact living cell is sitting in the tissue. Fragments of DNA that arrived and lodged there would produce the same signal. For 58 of the 59 women in the study, what was measured was male DNA, not confirmed male cells.

Evidence for living cells rests on other work. In 1996, Diana Bianchi and four co-authors published findings in PNAS that male fetal progenitor cells were still circulating in women’s blood as long as 27 years after they had given birth to a son, identified by flow cytometry using markers on the cells themselves. Later work located chimeric cells in bone marrow, liver, heart, thyroid, skin and lymph nodes of women with sons, in several cases with the morphology and antigen expression of the surrounding tissue.

Cell-level evidence from inside the human brain arrived in 2018. Lauren Broestl, Joshua Rubin and Sonika Dahiya at Washington University School of Medicine in St Louis published a study in Brain Pathology that tested tumour samples from women with a documented history of carrying a male pregnancy, then applied X and Y chromosome FISH to check whether intact male cells were there. They were, in roughly 80 per cent of the glioblastoma cases and 50 per cent of the meningioma cases, with no relationship to any of the clinical or molecular features used to classify either tumour.

The tissue was diseased, which limits what that paper can say about the ordinary ageing brain.

The traffic runs in both directions

Cell traffic in this system moves both ways, which is the part that tends to get lost. Cells pass from fetus to mother and from mother to fetus, and both populations can persist. Sean Maloney and colleagues, working in the same Seattle group as Nelson, reported in the Journal of Clinical Investigation in 1999 that maternal microchimerism was detectable in the blood of adult children, in subjects as old as 49, using HLA-specific PCR.

Which produces a genuinely odd arithmetic. A pregnant woman may be carrying material acquired from her own mother while she was in the womb, alongside newly arriving material from the child she is carrying. Hilary Gammill’s group examined exactly this in a 2011 PLOS ONE paper on microchimerism and the maternal grandmother, typing families for HLA loci rather than Y-chromosome markers, and found grandmaternal microchimerism detectable in women during normal pregnancy.

Three generations, briefly, in one bloodstream.

Where else the male DNA could have come from

Chan and Nelson are candid about the limits, and the candour is easy to skip past. Nine of the women were known to have had at least one son, and five of those nine carried male DNA in at least one brain region. Two were known to have had no sons, and one of those two tested positive anyway. For the remaining 48, the record is silent.

The authors list the alternative routes plainly: a miscarriage or termination involving a male fetus, a recognised or vanished male twin, an older male sibling, a non-irradiated blood transfusion. Because the interval between any pregnancy and death was mostly unknown, the paper could not test whether the DNA had persisted for six years or sixty. The 94-year-old is a striking data point, but she is evidence that male DNA can be present in an old brain, not a measured record of how long a specific pregnancy left its trace.

What the cells might be doing

Nobody knows. What the literature offers is a set of competing associations with no mechanism to connect them. Fetal microchimerism has been linked to autoimmune conditions including systemic sclerosis, and separately implicated in tissue repair and immune surveillance. A 2008 review in Obstetric Medicine by Keelin O’Donoghue at University College Cork lays out how often the same cells have been found in healthy and in diseased tissue, and how little that lets anyone conclude about cause.

The most interesting attempt to make sense of the contradiction comes from evolutionary biology. Amy Boddy, Angelo Fortunato, Melissa Wilson Sayres and Athena Aktipis argued in a 2015 BioEssays review that fetal cells in maternal tissue might work as an extension of the placenta, with interests that sometimes align with the mother’s and sometimes compete. Under that framing, organs involved in allocating resources, including the brain, thyroid and breast, would be predicted destinations. The authors offer these as testable predictions, and that is how they should be taken.

One result in the 2012 paper deserves particular care. Women with Alzheimer’s disease showed a lower prevalence of male microchimerism in the brain than women without neurological disease, with an adjusted odds ratio of 0.40. The authors call the result unexpected, note the modest sample, and offer no explanation. It sits awkwardly against earlier reports that Alzheimer’s is more common in women who have given birth. Nobody should read it as evidence that pregnancy protects the brain, and the result needs replication before it means much at all.

Thirteen years on, the follow-up has gone in a different direction. Broestl and her co-authors answered the two questions Chan’s team could not, and answered them in tumours. The missing study is that original design at larger scale, in ordinary brain tissue, in women whose pregnancies and their timing are on record. That is the work that would turn the woman who was 94 into a measurement instead of a line in a table.