Early Stress Loosens DNA Packaging in Mice, and One Enzyme Holds It Open

The material arrived as frozen tissue punches from the ventral tegmental area, a dopamine hub in the mouse midbrain, pooled three animals to a sample. Half of the mice had been taken from their mothers for three or four hours a day between postnatal days 10 and 17, with the nesting material in their home cage cut roughly in half on separation days. The researchers pulled the histone proteins out of the chromatin, cut their tails into 27 short fragments, and ran them through a mass spectrometer that can read more than 200 chemical states at once. Three quarters of the changes it turned up in the stressed animals pointed the same way, toward chromatin held open.

That survey was run with Ian Maze’s group at the Icahn School of Medicine at Mount Sinai, one of three labs behind the paper. By then the mice were about two months old and had spent their entire adult lives in ordinary housing. The separations were long finished, yet the way these cells packaged their DNA still carried them, which is the sort of persistence that has made chromatin marks a leading candidate for how early environments reach forward into adult behavior, as a 2022 review of cell type specific epigenetic effects sets out.

DNA does not sit loose inside a nucleus. It winds around histone proteins in tight turns, and the tightness is adjustable: small chemical tags stuck onto the protruding histone tails tell the cell’s machinery which stretches to keep wound shut and which to leave slack enough to read. Peña likens the arrangement to a coiled slinky, its genes shut away while the coil sits compressed and within reach once it stretches open. One of those tags is H3K4me1, a single methyl group parked on the fourth amino acid of histone H3, and it collects mostly on enhancers, the regulatory stretches that set how hard a gene fires once a signal arrives. The mark is not just a label on those stretches: work in differentiating stem cells indicates that H3K4me1 helps enhancers make physical contact with the genes they control. A gene sitting under a tagged enhancer is not switched on. It is easier to switch on.

Only One Enzyme Had Moved

Tags do not appear on their own, so the team went looking for the machinery. They measured sixteen enzymes in the same tissue: four that add methyl groups at H3K4, two that strip them off, and ten more that write or erase marks at neighboring positions.

One had moved. Setd7 was up in the stressed mice and stayed up in a second, independent group of males and females, while a rival writer called Kmt2a went the other way and fell. Setd7 is unusual company: where the MLL family enzymes can add a second and third methyl group in sequence, Setd7 places one and stops, which makes it about as close to a dedicated switch for this particular mark as biology offers. Mice given stress only in adulthood showed no change in it at all. And by postnatal day 21, four days after the separations ended, SETD7 protein was already piling up inside the nuclei of dopamine neurons, picked out by staining for the enzyme that manufactures dopamine.

Nothing Changed Until the Second Stress

To test whether the enzyme was driving the priming or riding along beside it, the researchers packaged a Setd7 gene into a virus and injected it into the midbrain of ordinary pups on day 14. It raised H3K4me1 in the region by about a third while leaving two related marks, H3K4me3 and H3K27ac, untouched. That specificity matters: it means the experiment tests one tag rather than a general loosening of chromatin.

Then, for weeks, nothing happened. Recorded in adulthood in Meaghan Creed‘s lab at Washington University School of Medicine in St. Louis, treated dopamine neurons from animals that had never been stressed behaved like any others, matching controls on how many spikes they produced per unit of injected current, on the voltage at which they fired, and on membrane resistance. Only after three days of unpredictable stress did the difference surface. Cells carrying the extra enzyme now fired far more, and they carried a larger hyperpolarization activated current, the slow inward drift that pulls a neuron back toward its firing threshold after something has pushed it down.

Behavior tracked the cells. After a mild bout of adult social defeat, half the Setd7 mice avoided a novel aggressive mouse strongly enough to be scored susceptible, against 8 percent of controls, and they spent less time in the exposed center of a test arena. None of these animals had experienced early-life stress. The enzyme by itself had been enough to build the vulnerability.

That delay between cause and consequence is the part the researchers keep returning to. “This work is exciting because it reveals a clear mechanism, and also helps explain why the impact of stress is both latent and broad,” said Catherine Jensen Peña of the Princeton Neuroscience Institute, the study’s senior author.

The Stress Response Ran Backward

Sequencing the tissue showed how odd the primed state is. In control mice, adult defeat mostly turned genes down: 72 percent of the genes that shifted went quieter. In mice carrying extra Setd7, the same stressor produced close to the mirror image, with 94 percent of the shifted genes rising instead. The primed brain was not responding more; it was responding in the opposite direction.

It is also not quite the response early-life stress itself produces, which this lab had mapped previously, and the authors say directly that Setd7 priming is probably only one component of what early adversity does. Two limits sit alongside that. The overexpression virus used a promoter that switches on in every neuron it reaches, so the effects cannot be pinned specifically on dopamine neurons, and the enzyme drops its mark wherever chromatin is already accessible rather than at genes of the researchers’ choosing.

Turning the Enzyme Down

The reverse experiment used a second virus carrying a hairpin that suppresses Setd7, delivered after the separations had begun. It cut H3K4me1 by roughly a third. In standard-reared mice it did essentially nothing, which is what a targeted intervention should look like. Among mice that had been through early-life stress, though, it held the line: 85 percent of the stressed animals given a scrambled control sequence scored as susceptible after adult defeat, compared with 33 percent of those with the enzyme knocked down. Their dopamine neurons did not develop the extra excitability, and the inward current stayed at normal levels. The mark, not the memory of the separations, was carrying the vulnerability forward.

Creed, an associate professor of anesthesiology and one of the paper’s two corresponding authors, described the result in human terms. “This finding reveals a physical scar left by trauma experienced during development inside brain cells, providing scientists with a concrete biological target to develop new treatments and interventions,” she said. The scar, for now, is a characterization of what the mark does to a mouse, and the human picture is not a simple copy of the rodent one. When institutionally reared Romanian children were randomly assigned to high quality foster care, those left behind in institutions turned out to have blunted rather than heightened responses to stress, a direction opposite to the hyperreactivity these mouse models produce.

Blocking the enzyme is one route to protection. Changing the surroundings is another, and it has already been tried: rats given an enriched cage during the same weeks they were being separated from their mothers did not grow into the anxious adults that separation alone produced, an effect that ran through glucocorticoid receptor traffic in the amygdala rather than through anything measured here.

The same dopamine neurons that grew louder about defeat also report unexpected rewards. If early stress leaves them primed to react harder to whatever arrives, rather than to threat in particular, a primed mouse should also respond more strongly to a good week. Nobody has run that experiment. The authors raise it as the obvious next one, and the answer would decide whether this mark is best understood as a scar or as a wider aperture. Further reading: What Is Neuroplasticity? The Science of How the Brain Rewires Itself

Frequently Asked Questions