Skip Navigation

Child abuse leaves lasting 'scars' on DNA

February 20, 2009 By Heidi H Ledford This article courtesy of Nature News.

Lingering marks on DNA could amplify stress responses.

Suicide victims with a history of abuse during childhood are more likely to carry chemical changes to their DNA that could affect how they respond to stress as adults, a study has found.

Those with no history of childhood abuse did not show the same pattern of DNA modification, and had normal expression of NR3C1, a gene linked to stress responses.

But the findings do not mean that the effect of childhood abuse is indelible, cautions Joan Kaufman, a psychologist at Yale School of Medicine in New Haven, Connecticut, who was not involved in the new study. "The long-term effects of early abuse are not inevitable," she says, "and the more you understand about the mechanisms of risk, the more you can devise treatments."

The results, reported today in Nature Neuroscience1, follow on from work showing that rat pups that are stressed because they were raised by negligent mothers have extra methyl groups in their DNA in a region of the genome that controls expression of Nr3c1, the equivalent gene in rats. Such 'methylation' can reduce gene expression. NR3C1 encodes a protein expressed in neurons that responds to hormones called glucocorticoids. Lower expression of NR3C1 could be harmful because reduced responses to these glucocorticoids have been linked to increased stress.

Lasting impressions

To find out whether the results in rodents translated into humans, neurobiologist Michael Meaney of McGill University in Montreal, Canada, and his colleagues collected brain samples from the Quebec Suicide Brain Bank. The researchers looked at samples from 12 suicide victims with a history of childhood abuse, 12 suicide victims with no history of childhood abuse, and 12 controls who had died suddenly from other causes.

People with a history of childhood abuse had lower levels of glucocorticoid receptors than either people who had not been abused or those who had not committed suicide. And childhood-abuse victims had a similar methylation pattern to that seen in rats that had been stressed as pups.

These changes are unlikely to be passed on to the next generation, notes Meaney. Although researchers have not yet looked for effects on egg or sperm DNA, it is doubtful that the changes affect the germline, he says.

Researchers do not yet know whether trauma as an adult produces the same pattern of changes. There may be times during childhood when the developing brain is particularly responsive to abuse, explains Martin Teicher, director of the Developmental Biopsychiatry Research Program at Harvard Medical School in Belmont, Massachusetts. Teicher and his collaborators imaged the brains of women who had been victims of child abuse and found that those who were abused between the ages of three and five, or eleven and thirteen, had a smaller hippocampus — a region in the brain that is important for memory and learning — than those who had not been abused2.

Searching for a way out

Studies in rats have also suggested that the methylation changes can be reversed: if pups reared by negligent mothers are later treated with a chemical that removes DNA methylation, their stress responses return to normal3.

Such drugs are not ready for use in humans, and could carry unwanted side effects. But medication may not be the only way to treat victims of child abuse, and DNA methylation was restored to normal in neglected rat pups if those pups were transferred to more attentive mothers. "Just because there's a biological effect doesn't mean the only way you can intervene with drugs," says Kaufman.

Psychotherapy, for instance, has been shown to produce chemical changes in the brain, and might be able to reset the methylation pattern, says Meaney. "A social event got you into it. Could a social event get you back out?", he asks. "That's a very viable hypothesis."


  1. McGowan, P. O. et al. Nature Neurosci. Advanced online publication doi:10.1038/nn.2270 (2009).
  2. Andersen, S. L. et al. J. Neuropsychiatry Clin. Neurosci. 20, 292–301 (2008).
  3. Weaver, I. C. G. et al. Nature Neurosci. 7, 847–854 (2004).


Need Assistance?

If you need help or have a question please use the links below to help resolve your problem.