Situational Stress Makes Short-Term Epigenetic Changes

Or so it appears, based on research by Gunther Meinlschmidt and colleagues. When they exposed 76 people to a stressful simulated social situation, they found changes in the methylation of two genes within an hour. What’s more, those two genes—oxytocin receptor (OXTR) and brain-derived neurotrophic factor (BDNF)—are important to human behavior. The oxytocin receptor conveys the hormone oxytocin’s effect at several sites in the body, including the brain. BDNF supports existing neurons, encourages their growth, and functions in memory and learning.

The case isn’t perfectly conclusive yet, of course—with 76 subjects between 61 and 67 years old, the study could be larger. And the team measured gene methylation in blood samples—and not brain samples, of course—so it’s not clear that the same changes happen in our big thinking organs when we’re stressed.

Still, it’s a cool finding. The study’s big enough to identify real things, and it’s reasonable to suspect that blood and brain OXTR and BDNF genes get methylated under similar circumstances. Also, this study seems to come the closest to what the more, uh, optimistic and fanciful epigenetics fans have been claiming. Now, it definitely doesn’t show that gene regulation responds to intentional happy thought, but the mental state of psychosocial stress does seem to trigger methylation—and quick.

Gunther Meinlschmidt is director of the Research Department of Psychobiology, Psychosomatics, and Psychotherapy at Ruhr-University Bochum’s LWL University Hospital, and the rest of the team works from the University of Basel, the University of Trier, King’s College London, and Ruhr-University Bochum. I spoke with him the other day to ask him a few things about his research. Here’s what we talked about.

Is this the first study you know of that looks at the epigenetic effects of stress on specific genes?

As far as I know, yes.

It depends a little bit upon what you mean by stress. You’re probably aware of this early life-stress—looking at stress in terms of early adversity, how it has long-term effects on the epigenome. Most of these studies have looked at specific genes, most of the studies have looked at the GR [glucocorticoid receptor gene]. But some studies—there was a recent one in the Archives of General Psychiatry by scientists from Montreal—in which they look genome wide, so using chip technology.

But again this was looking at early adversity. So as far as we know, we are really the first to look at acute effects of psychosocial stress on specific genes.

Can you describe the stress test that the subjects go through?

We applied a stress test called the TSST, the Trier Social Stress Test. It’s perhaps the most applied test to induce psychosocial stress—so it’s been applied in hundreds of studies.

It consists of two phases. The participants are required to do mental arithmetic and to undergo a kind of mock job interview, so they have to present themselves [favorably]. What they do is relevant to them. They do this in a social situation, so there are two collaborators sitting in front of them with a camera and a microphone—it’s not real or working, but the two collaborators usually wear white clothes, and they’re instructed to be very neutral and strict. For example they ask the participants to start calculating again, if they make any mistakes.

The participants know, of course, that this is just for research. They know that this is not a real job interview. But the mere situation induces stress because of this individual personal relevance together with the social evaluative threat. It induces subjective changes in stress, but also a stress reaction in terms of activity of the HPA [hypothalamic-pituitary-adrenal] axis.

What was the most surprising result in your research?

One thing is that we didn’t know at all if you could see changes occurring so quickly after a psychosocial event. Of course, histones change very quickly [when triggered]. But in terms of DNA methylation, [there is a widespread impression] that these changes are very long-lasting changes, and that they respond rather slowly.

But we found changes occurring within an hour after a psychosocial stress [session] that itself lasted ten minutes. I think it really adds to the picture of DNA methylation as something that can respond more quickly—not as quickly as some of the histone modifications probably, but it’s rather quick. In the literature, there’s data in a Nature paper on cyclic changes in DNA methylation that regularly turn genes on and off. That showed that DNA methylation changed within—I think—one or two hours.

So it seems like we have to look at changes in DNA methylation following external triggers, like social events.

The genes that became methylated were the oxytocin receptor (OXTR) and brain-derived neurotrophic factor (BDNF). About the oxytocin receptor, what tissues is it most associated with?

I’d have to look at the literature to be perfectly accurate, but it depends. The oxytocin receptor is sensitive to estrogen, so it’s not just be a question of tissue. It also depends on [a person's] endocrine status.

Peripherally the oxytocin receptor is relevant when it comes to contracting smooth muscle, such as in the uterus, or to elicit milk ejection. For example in the uterus, the hormonal milieu can upregulate the receptor. And of course, there is the brain, where the oxytocin receptor is involved in a lot of behavioral effects. You’ll find it in different parts of the hypothalamus, such as the medial preoptic nucleus, for example.

You discovered methylation changes in oxytocin receptor in blood samples—can’t take brain samples, I guess.

Of course, brain is not possible. You could take biopsies [of other tissues], but it’s invasive and ethically problematic. It probably wouldn’t be possible to do this before and after stress. So very quickly it gets very complicated.

Scientists from Montreal did a couple of nice studies in post-mortem brains [Here's another. —Ed.], where you have really good target tissues. But of course, you can’t use this to address quick changes in methylation. They studied only long-term consequences of some early adversities by reconstructing them using databases of information from people who donated the brains.

How much does the state of oxytocin receptor in blood correspond to the situation in other tissues—brain and so forth?

That’s an important open question. There is data on the comparison of blood cells with mucosa from the gut, and there was quite a bit of correspondence in the epigenome across tissues.

But really, to say that what we picked up in the blood is related to the epigenetic status in specific brain areas—we really don’t know. That’s one question that we’d like to answer, for example, with animal models now.

What’s the next step in your research?

We picked up some of the genes that were interesting to us. We’re really moving in one of three directions. One is to assess epigenetic changes more globally, so not only specific genes, but gene networks. Or with chips, even more.

The other thing is to have a better understanding of the time frame of the changes, to have a more narrow description of the epigenetic changes.

And then a third question is to identify the mechanism. So: What drives these changes in DNA methylation? Are there some enzymes involved in the methylation machinery that are up- or down-regulated following psychosocial stress? And what are the factors that are changing them? Are they factors that we’re already aware of, such as the usual candidates, cortisol and other factors? Or are there further factors involved that really drive this methylation change?

_

[The picture above wasn't intended to be about stress, but I thought it fit pretty well anyway. It's by Flickr user sashafatcat, and it's used here under a Creative Commons license.]

_
Unternaehrer E, Luers P, Mill J, Dempster E, Meyer AH, Staehli S, Lieb R, Hellhammer DH, & Meinlschmidt G (2012). Dynamic changes in DNA methylation of stress-associated genes (OXTR, BDNF ) after acute psychosocial stress. Translational psychiatry, 2 PMID: 22892716