Child abuse permanently modifies stress genes in brains of suicide victims

By Ed Yong | February 22, 2009 1:00 pm

Blogging on Peer-Reviewed ResearchThe trauma of child abuse can last a lifetime, leading to a higher risk of anxiety, depression and suicide further down the line. This link seems obvious, but a group of Canadian scientists have found that it has a genetic basis.

By studying the brains of suicide victims, Patrick McGowan from the Douglas Mental Health University Institute, found that child abuse modifies a gene called NR3C1 that affects a person’s ability to deal with stress. The changes it wrought were “epigenetic”, meaning that the gene’s DNA sequence wasn’t altered but it’s structure was modified to make it less active. These types of changes are very long-lasting, which strongly suggests that the trauma of child abuse could be permanently inscribed onto a person’s genes.

Child abuse, from neglect to physical abuse, affects the workings of an important group of organs called the “hypothalamic-pituitary-adrenal axis” or HPA. This trinity consists of the hypothalamus, a funnel-shaped part of the brain; the pituitary gland, which sits beneath it; and the adrenal glands, which sit above the kidneys.  All three organs secrete hormones. Through these chemicals, the HPA axis controls our reactions to stressful situations, triggering a number of physiological changes that prime our bodies for action.

The NR3C1 gene is part of this system. It produces a protein called the glucocorticoid receptor, which sticks to cortisol, the so-called “stress hormone”. Cortisol is produced by the adrenal glands in response to stress, and when it latches on to its receptor,  it triggers a chain reaction that deactivates the HPA axis. In this way, our body automatically limits its own response to stressful situations.

Without enough glucocorticoid receptors, this self-control goes awry, which means that the HPA is active in normal situations, as well as stressful ones. No surprise then, that some scientists have found a link between low levels of this receptor and schizophrenia, mood disorders and suicide. So, childhood trauma alters the way the body reacts to stress, which affects a person’s risk of suicide or mental disorders later in life. Now, McGowan’s group have revealed part of the genetic (well, epigenetic) basis behind this link.

Silencing a promoter

The group looked at 24 samples of brain tissue taken from autopsies of male suicide victims, half of whom had been abused as children and half of whom had not. They compared these people to a dozen others, who had never been abused and had died suddenly through accidents. All the samples came from the hippocampus, a part of the brain where NR3C1 is particularly active.

McGowan found that the activity of the NR3C1 gene was much lower in abuse victims who took their own lives, than in either of the other groups. In their brains, he found lower levels of NR3C1’s mRNA – a molecule that is transcribed from the gene’s DNA sequence and provides the recipe for building the glucocorticoid receptor. With a shortfall of this mRNA, the abused suicide victims would also have had lower levels of this critical receptor.

It was clear that this deficit was because of the abuse rather than the suicide, since suicide victims who hadn’t been abused had the same levels of NR3C1 mRNA as people who died of  other causes. McGowan even accounted for whether the victims suffered from mood disorders, alcoholism or drug addictions – none of these factors swayed the results.

Why was the NR3C1 gene less active in the abused victims? To find out, McGowan focused on the gene’s “promoter region“, a stretch of DNA that controls its activity. NR3C1 is activated by a protein called NGFI-A, which needs to dock at special landing sites in the promoter to switch on the gene.

In the abused suicide victims, these landing sites had been heavily altered. McGowan found that small molecules called methyl groups had been attached to the promoter region’s DNA, which prevented the NGFI-A protein from docking. The underlying DNA hadn’t changed – in fact, the sequence of NR3C1’s promoter region was exactly the same in all 36 samples.

This is an example of an “epigenetic” change, where structural modifications change the way a gene behaves even though its DNA sequence is still the same. The methyl groups are the chemical equivalent of Post-It notes that obscure the underlying text without actually editing it. McGowan found over twice as many of these chemical add-ons in the abused suicide victims than in the others. It was their presence that slashed the activity of the NR3C1 gene.

Putting it together

Together, McGowan’s results paint a very suggestive picture that links child abuse to later illness and death. Abuse leads to epigenetic changes that modify genes including NR3C1. These changes, especially those involving methyl groups, are very stable and they could slash the gene’s activity for a very long time, potentially even for life. With the gene not working properly in many cells, the body can’t produce enough glucocorticoid receptor.

As a result, the HPA trinity can’t turn itself down properly and is constantly on high-alert. The body behaves as if it were stressed, even when nothing stressful is happening. The result is a higher risk of anxiety, depression and suicide. McGowan admits that this the whole picture is still speculative, but the individual steps make sense in the light of his results.  

Other groups have found incredibly similar results in the brains of rats. The quality of care that a baby rat receives from its mother is reflected in the degree of methylation in the promoter region of its NR3C1 gene. If the bond between a mother rat and her baby is disrupted at an early age, the number of methyl groups in this critical area goes up, and the baby’s HPA system becomes hyper-responsive to stress. And mice that lack the glucocorticoid receptor altogether behave in ways that are similar to depressed humans.

Mothers could affect the fates of their children through this chain of events, even before they are born. One fascinating study found that a newborn baby’s NR3C1 gene is more heavily methylated if its mother was depressed or anxious during her third trimester. The epigenetic changes to this gene could allow children to inherit a vulnerability for depression from their mothers. 

As a final note (and I hope he doesn’t mind me pointing this out) but the lead author of this study is one Dr. Meaney. Which is a rather ironic name to crop up on a paper about abuse! A bit of unfortunate nominative determinism, surely?

Reference: Patrick O McGowan, Aya Sasaki, Ana C D’Alessio, Sergiy Dymov, Benoit Labonté, Moshe Szyf, Gustavo Turecki, Michael J Meaney (2009). Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse Nature Neuroscience, 12 (3), 342-348 DOI: 10.1038/nn.2270

More on epigenetics:

Subscribe to the feed

Bookbanner.jpg

Comments (29)

  1. Lily

    Well, this definitely strikes a chord for me. I don’t experience depression on a day-to-day basis; I’m not schizophrenic; and I’m not suicidal. But I do really overreact to ordinary stressful events, and I have a really hard time recovering from them. It’s fairly disruptive to my life. I wonder who’s doing research on how to reset NR3C1, if that’s even possible.

  2. Hank Roberts

    Has anyone sorted out ‘child abuse’ from ‘traumatic brain injury’? I’m figuring a kid could get a concussion even in a normal happy family, back in the era of metal dashboards, no seat belts, and no helmets on tricycle and bicycle riders — there should be enough data on that.

  3. Heatherly

    I would assume they were able to account for differences between abuse and physical injury–child abuse is a hell of a lot more than physical–neglect, sexual, mental, emotional, medical, etc. And if this is similar to other studies I read during grad school, I assume their focus is on the holistic affect of abuse: the tension, build-up, actual event and aftermath, not just physical injuries.

  4. Fishwit

    This article would explain my almost life long trouble with high level anxiety.(By the way I was not sexually abused or brain damaged). Conventional professional treatments have reduced it somewhat but not eliminated the problem. Until knee problems stopped me, I controlled cortisol levels by pretty ferocious jogging. So can someone tell me how to remove the metyhl molecules from my NR3C2 DNA? If my cortisol levels stay high it does not bode well for my brain functioning as I grow older.

  5. Ellen

    The work that is being done here is “good work”. There is a difference between public and private thoughts.
    Since sexual abuse is in my past, I object to you adding the comment about Dr. Meany.
    You owe an apology to him.
    May no child ever suffer abuse, Ed.

  6. I think Hank Roberts asks a good question. Perhaps a mild injury, a concussion, or one that does not result in long-lasting unconsciousness might not have such an effect.
    Along with that question, I would ask: Can and does this epigenetic change occur in adults? And, are there other things that can cause the change?
    I also wonder if research that does not lead to further questions is valid?? Offhand, I can’t think of any except pseudo-stuff that has “all” the answers.

  7. Fishwit

    I doubt that the test of “validity” is whether research leads to other questions. That may be a test of “utility” but as I understand the word “valid”, research is valid if it is methodologically sound. In any case this research does lead to at least one other question, namely how does one detach excess methyl molecules from the NR3C2 DNA, regardless of how they got there?

  8. Garrick

    Thank God they are finally figuring this out. I had a very abusive childhood. I’ve had therapy afterwards which didn’t amount to much except going over memories of abuse. I’ve had “cognitive therapy” which tried to change how I remember things, how I think and who I am. I don’t need that as I’d rather have the truth than be a relaxed idiot.
    I’m tense and alert all the time, but otherwise feel fine. I’m tired of people I meet occasionally trying to probe my past and help me “let it all out” which I soon figure out is what they’re trying to do. Very insulting. This is who I am, because when I was growing up, only physical abuse was classified as child abuse. Emotional abuse, even severe chronic emotional abuse, was not really considered abuse for “ethical reasons”. Children are “resilient” I was told as a child.

  9. Interesting stuff. Are these epigenetic changes reversible? And (aside from the ethical concerns) could children be ‘primed’ against these changes?

  10. speedwell

    Does this only happen after child abuse? At what age does it stop happening? Might an abused young wife, 19 or 20, whose brain is in the last stages of development, suffer this effect? How about a frightened young soldier of the same age? What about older people?

  11. Gil

    What an interesting post – and piece of research. I know from my own practical experience working in a CAMHS service that we often need to “stablise” adolescents presenting to us with a past history of abuse by sorting out lifestyle, home life and social networks before considering any therapeutic work (whatever that may be). Without that a high level of anxiety can prevent any cognitive work that may take place.
    Also, I found the “Dr Meany” comment to be a entertaining side-note which it didn’t detract from the bulk of the piece.

  12. Garrick, the problem is that we weren’t designed to be stressed out all the time. Hypertension is not a normal condition for human beings to have. As for being a “happy idiot” — I’ve seen how easy it is to roll reflexively and unthinkingly cynical people. Their unhappiness doesn’t automatically make them smarter. (It does, however, often make them easy marks for things like antivaccination and “9/11 truth” movements.)

  13. Ed Yong

    Great comments folks – keep it coming. To address two quick point:
    – Some people have asked loads of follow-up questions and obviously, the study ended with what I described. So those are great questions for follow-up but I don’t have any answers now.
    – Many people (Lily, Fishwit, Frank and others) have asked about resetting epigenetic marks. I’m pretty sure this is theoretically possible and being heavily investigated, especially because epigenetic changes are implicated in a wide range of diseases and disorders, including cancer and obesity. As an example, check out the post I link to at the end, about a mouse experiment which shows that epigenetic changes are responsible for obesity “cascading down the generations”. In that study, this effect was reversible through folate-rich diets. Also just try Googling “epigenetic therapy” or “epigenetic drugs” but as always, please bear in mind that this is preliminary research.

  14. Lily

    Hey, Ed, thanks for the response–I think your reply cut off the example that you were about to recommend. :) Repost? Thank you!
    (Also, LOL @ Phoenix Woman.)

  15. Art

    Phoenix Woman – “As for being a “happy idiot” — I’ve seen how easy it is to roll reflexively and unthinkingly cynical people. Their unhappiness doesn’t automatically make them smarter. (It does, however, often make them easy marks for things like antivaccination and “9/11 truth” movements.)”
    As a victim of abuse and a studier of the various conspiracy and denialist movements I have to note it isn’t the true ‘cynics’ who fall for them. It is the people who are romantics and are feeling a loss of power and control that get sucked in.
    The hidden romantic side to conspiracies is that someone is in control. This is like the religious sorts who claim that God controls every little thing. The vast unseen forces assumed to be engaged in the conspiracy are, in effect, a substitute for an absent God. It is a comfort knowing that everything is under control. Even if the controller is an arbitrary, thundering git.
    Denialist claims comfort not by assuming that there is a controller but rather by conferring the illusion of control onto the denialists by way of their having ‘inside information’ and ‘seeing through’ the ruse and ‘disinformation matrix’.
    In both cases it is romantic notions that reinforce the delusion. In the first case it is the romantic notion that there is a ‘Big Daddy’ figure out there controlling things. The later is reinforced by the romantic notion that he denier is ‘heroic’ and an ‘outlaw’ fighting against ‘the machine’.
    Such people are not cynics. They are romantics who sneer a lot and play at cynicism. That they are unhappy provides some motivation by doesn’t explain much. Conspiracy theorists can be either smart or dumb as bricks, while still assuming they are smart (Dunning-Kruger effect), but it has little to do with their susceptibility to self delusion.
    People who have lost hope to the point of suicide with an actual intention of dying, versus staging suicide to get attention, are, IMHO, seldom romantics. The romantic side has been burned out. They tend toward being comfortable, as comfortable as anyone who’s nervous system is always screaming it is under attack can be, in their awareness that nobody is in control. As such they are highly resistant to denialism, conspiracy theories and the go-along-to-get-along culture that surrounds them.
    Granted conspiracy theorists are damaged. They just aren’t damaged enough to give up the romantic illusions.
    This is a remarkable study. That explains a lot about what is going on and why abuse victims are so resistant to therapy. Why the effects can last a lifetime.
    Now, if someone could point out, as Fishwit asks, “how does one detach excess methyl molecules from the NR3C2 DNA, regardless of how they got there?” Understanding how your borked is great. Having a way of repairing or compensating for the damage, divine.

  16. Ed Yong

    Ahem – truncated comment now untruncated.
    Also Art, lovely post and interesting theory regarding lack of control and belief in conspiracies. I think this post, which provides direct experimental evidence for it, will make you happy.
    And finally, to everyone who has personally suffered abuse and had chimed in, you have my compassion for your experiences and my gratitude for your comments.

  17. Anna K.

    Hmmmmm. Somewhere I seem to remember coming across studies that claimed long-term meditation practices changed brain structures, specifically, helping people become less anxious. I wonder if the changes said to occur with a habit of meditation affect glucocorticoid receptors?

  18. Fishwit

    Very good point Anna K. As Art points out, abuse victims are resistant to therapy and this research may explain why. But some are completely cured or, as in my case, considerably improved by therapies of one kind or another. It may be that some therapies do, among other things, alter the receptors. I suspect that altering the receptors is only part of what has to be done, although an important part. Integrating aberrant emotions into the whole personality is just as important, and very hard for an adult because it is not achieved by purely intellectual means (peace, all fanatical cognitive therapists.)
    In what is now getting to be a longish life, by far the best help I have found is in a book by two doctors, Conrad Baars and Anna Terruwe, called “Psychic Wholeness and Healing.” It is a bit hard to get in bookshops, but if you Google “Conrad Baars” you will get to a web site run by his son (Dr Baars is dead) which sells this and other of their books plus some very helpful tapes/CDs. But I think I recall that Baars & Terruwe say somewhere to the effect that their methods are not helpful, or at least not the whole answer, for victims of sexual abuse. My childhood was a lot less than positive, to put it euphemistically, but thank God it did not contain any sexual abuse. And Baars/Terruwe and the tapes have been enormously helpful, although I am not yet “there.” Maybe still some especially sticky methyl molecules??

  19. Nathan Myers

    As I understand it, those methyls are also what makes a cell “know” to behave as a brain cell, or a blood cell, or a muscle cell. Stripping the methyls off, willy-nilly, might change the working cells into stem cells. Stem cells are nice to have ready to hand but don’t actually do anything useful. Of course it’s bound to be easier to get all the methyls stripped off than to replace the ones that belong where they belong, or to select just some to strip.
    We have seen cases of stem cells that change to match the type of cells they find around them, but I don’t know that anyone has any idea how that works. Do you?

  20. notoverthehill

    I have no intent to change the subject of this discussion, however I decided to respond to the subject of 9/11. In my experience with some 9/11 truth types, I find they are in two sets. Those who want to see in 9/11 the evidence of a hidden super-government planning and controlling our reality are one set. The second set are those who think the video of the destruction of the WTC fails the reality test. Based on such disbelief, they took the effort to collect evidence and learn the physics of how such an event could possibly occur. There is a difference between a religious zealotry on one hand, and on the other hand a sense of certainty based on science and physical evidence.
    In keeping the discussion theme about early abuse, consider a defensive strategy by some who in the past have been severely disappointed by authority. We all need to maintain a some trust in authority to live in society. For some abuse survivors, a firm decision was made to detach reality testing from the need to have trust in authority. If dependency for reality is shifted from authority trust to the scientific method, a valid and objective judgment can result.
    One can perform an arm-chair psychological analysis of 9/11 “conspiracy de-bunkers” as well. For example, de-bunkers who claim that anyone who disagrees with the government version of events are people who “disrespect the families of the victims” are demonstrating a classic denial defense of shaming the messenger. Such a denial pattern results from having forbidden knowledge. This suggests that these people also believe in what the “truthers” believe, subconsciously.
    For example, consider the spouse who has evidence based subconscious knowledge that her husband is cheating on her. The defense activates when an acquaintance quietly suggests evidence of infidelity, and for reasons which focus on preserving the family and by extension the authority structure inherent in the family, the in-denial spouse will violently reject and may even counter attack the person who suggested her husband is cheating.
    By pathologizing everyone with a distrust in the government’s version of events on 9/11, a disservice is done to many competent engineers, scientists and architects who are motivated by a moral obligation to preserve and protect our country.

  21. Ed Yong

    Nathan – not sure about the stem cells question, but you’re right about the methyls.
    As a general note, let’s be careful about oversimplifying here – epigenetic changes are not A Bad Thing. For the most part, they’re a normal and necessary part of the workings of our cells.

  22. cam

    This was a very helpful post, so I truly thank you. We recently completed a case study in my psychology class about a young boy who was experiencing severe stress when it came to test taking and we were told to evaluate the causes, symptoms, and effects of his condition. This post provided information that I was able to learn from.
    Thanks!

  23. MTran

    What an interesting study. Does anyone know how this information might correlate with the presence or absence of so-called “resilience” genes (5HTTP?) which might mitigate the long term effects of early abuse? I hope that some good therapies and remedies grow out of this research.

  24. Grace

    I actually read an article that fully discussed the original rat study that initiated this human study. According to the rat study, the methylation WAS reversible. The study noted that rat pups who were returned to a nurturing environment began to have lower cortisol levels. When the gene was analyzed, there was a significant decrease in the levels of methylation at the receptor sites.
    So, assuming the human outcome would parallel the rat study, there is a lot of hope out there for us “abused”.
    I actually believe that people who have been abused have a great deal of “healing” when they surround themselves in an environment where it is safe and nurturing. This healing isn’t just metaphorical or spiritual–I actually believe there is a biological component to it.

  25. Fishwit

    This is very encouraging, Grace. This might explain why some people are able to be helped by therapy but others are not at all or only very slowly. The external environment may be crucial to the effectiveness of meditation or Baars/Terruwe or whatever other therapy is attempted, and, as you say, where that environment is favourable may facilitate the decrease in levels of methylation without the need for some super pill or other. Your phrase “safe and nurturing” is almost a synonym for “loving.”

  26. Liz

    Terrific post, many thanks. I’d be interested to read more on the link between abuse and auto-immune disease. I have rheumatoid arthritis that started at age 20 and I firmly believe was a result of the very genetic changes you refer to here. Too much stress in childhood, continual flight or fight response, and eventual immune breakdown. When i first hit on this theory my medical specialists thought I was nuts. 25 years on, this abuse theory seems to be gaining more credence.

  27. pariah

    I would draw no conclusion from this tiny retrospective study. On the scale of evidence it is terrible. We’re talking about 24 subjects. If you want to obsess over this data and accept it as valuable and meaningful, you are wasting your time in my opinion. These conclusions are terribly premature and the publication has been popularized by the media for its sensational value.

  28. James Bannon

    I echo pariah’s comments. In addition, there is something not quite right about the description. Are we into Lamarkism perchance? Supposing that abuse did, indeed, alter the structure of NRC31, what does this mean in terms of heritability? Not much, unless there is plain evidence that the altered structure is indeed heritable. I don’t see anything here to suppose that this is the case. I wish people would use the label “genetic” much more carefully!

  29. Sunset Mission

    Hello to you,
    It’s not easy to read and understand everything for me, but your article is very very interesting for me. I write to you but it’s like a bottle in the sea for me.
    I was abused and I need to do a test for my NR3C1 for 2 things : first because I need to know, second it’s for justice.
    It’s hard to find in France a place where I can ask to do the test and I don’t know where, or who exactly could do it here.
    I asked to my doctor. No idea. I asked in Paris hospitals in genetic department : nobody knows…
    If you can tell me where, how, who, exact informations, or just an idea, you will be very very nice.
    You know, I don’t want to modify my genes, because I learned to live with my traumas with a therapy. But I need to know if I lost 40% of the capacity of my NR3C1 and if I can know it, I will give the test to the judge in a few month (the monster will be judged).
    Cheers and thanks to be interested by “our” problem.
    We are survivors…

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Not Exactly Rocket Science

Dive into the awe-inspiring, beautiful and quirky world of science news with award-winning writer Ed Yong. No previous experience required.
ADVERTISEMENT

See More

ADVERTISEMENT
Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »