Doubts Arising About Claimed Epigenetics of Holocaust Trauma

Holocaust and other trauma had been reported to be inheritable, but it's impossible to tease out the influence of genetic modification from that of horrific stories.

Surfer standing at attention on the Mediterranean Sea beachfront as a siren sounds in memory of victims of the Holocaust, in Haifa, April 24, 2017.
Rami Shllush

Is it even possible that the effects of trauma, such as experienced during the Holocaust or other horrifying events, can be inherited? That a father or mother can pass post-traumatic stress to their childen, through egg or sperm? Recent studies have insisted they can – but now doubts are emerging about the conclusions reached from the data.

Over decades, studies have linked post-traumatic stress disorder in Holocaust survivors with an increased risk of anxiety, depression, and PTSD in their children. More recently, research has suggested that such trauma may be transmitted via epigenetic means, which means: we can pass on not only our genes, but how our genes are expressed.

As I described in the journal Nature in 2014, I myself participated in one such study, conducted by Rachel Yehuda, a neuroscientist at Mount Sinai School of Medicine in New York. Her research, which has been widely reported in The Guardian and elsewhere, is claimed to be “a clear example in humans of the theory of epigenetic inheritance.” Deepak Chopra, the popular New Age guru, reproduced chunks of my Nature article in his book Super Genes: Unlock the Astonishing Power of Your DNA for Optimum Health and Well-Being.
            Even as Nature was publishing my article, though, I was troubled by a question: How does one separate the impact of horrific stories heard in childhood from the influence of epigenetics?

When I was a child, my father, Gershon Glausiusz, told us of his suffering during the Second World War. As a child, he had been deported with his mother and four brothers from his home-town of Szarvas in Hungary to the concentration camp of Bergen-Belsen. He arrived in the concentration camp on December 6, 1944, and  “celebrated” his tenth birthday in Belsen on the day after their arrival.

After almost five months of incarceration, my father, grandmother and three surviving brothers were liberated on the 23rd of April, 1945. Even so, the trauma of my father’s childhood has remained with him throughout his life. “It’s always there; it’s not on the surface, it’s just below,” he has told me. “When I was a young boy, I was more like an old man.”

Children of Holocaust survivors may suffer depression and anxiety: but its source cannot be ascertained. In theory, hearing horrific stories in childhood could influence the development of anxiety and depression in adulthood. How do we know if horrific stories or epigenetics caused the anxiety or depression? There is no way to tease out the influence of either.

Indeed, some leading researchers in the field of epigenetics have grave reservations about Yehuda’s interpretation of her data. Foremost among them is John Greally, professor of genetics and pediatrics at the Center for Epigenomics at Albert Einstein College of Medicine in the Bronx, New York. In a blog post published a few days after the Guardian story, Greally termed Yehuda’s research the “over-interpreted epigenetics study of the week.”

Over-interpreted epigenetics

Part of the problem, Greally says, is that there are multiple ways in which to define the term “epigenetics”, but in general, it is used to describe the study of potentially heritable changes in gene expression (the process by which DNA is converted into proteins) that do not involve mutations to the underlying DNA itself.

The most commonly studied epigenetic modification of a gene is DNA methylation, in which a chemical modification called a "methyl group" tags and alters the expression of the gene, without changing the gene itself.

Yehuda’s 2014 study in The American Journal of Pyschiatry showed that children of fathers who had survived the Holocaust and had suffered PTSD had higher methylation of a promoter of a gene involved in the body’s stress response.

But if both parents had survived the Holocaust and had experienced PTSD, their children actually turned out to have lower methylation of the gene.

As Greally and his colleagues described in a 2016 paper in PLOS Genetics, studies like these “suffer from multiple problems in design and execution that severely limit their interpretability.”

One problem is that there are many reasons why DNA methylation of cell types can change.

One “huge confounding variable,” Greally explains, is genetic differences in DNA sequences between individuals, which can influence DNA methylation.

In fact, between 22 to 80 percent of the differences in DNA methylation between individuals are due to underlying genetic differences. It is therefore difficult to know whether the methylation of a gene has some external cause, or is due to the underlying genetic sequence.

In other words, theoretically, a parent and child could share methylated sequences because of an external event, or because the child has inherited the DNA, which influences the methylation of the gene.

A second problem is that DNA methylation patterns differ across cell types. The DNA of a white blood cell may have a different pattern of methylation than an inaccessible brain cell, for instance. Mixtures of cells with varying proportions of cell types would therefore result in different patterns of methylation in the lab. (Science can correct for this through isolating on a specific purified cell type, but the process is much more expensive, for one thing.)

Third is the problem of “reverse causation." Trauma or stress could have caused methylation of the gene. Or methylation of a gene could increase the risk of trauma or stress. But that doesn’t prove that the methylation itself was inherited.

Even if you conduct a rigorous study of two groups of people, one stressed, one not, and you found increased methylation in the stressed group, you still would not know if the methylation was the cause or effect of their stress, Greally explains.

Greally is cautious about criticizing Yehuda’s research, though.

“I have to put my hand on my heart and say I have made all the same mistakes that everybody else has made and I’ve published them, so I’m not being some holier-than-thou character in all of this. I’m just as guilty as everybody else,” he says. Even so, he adds, the values in Yehuda’s study “were so marginally different, that that is concerning.”

His doubts are corroborated by Rafael Irizarry, Professor of Applied Statistics at Harvard and the Dana Farber Cancer Institute. Asked to comment on Yehuda’s original paper, he wrote, “I am not convinced that the data supports the conclusions,” and that the differences in average methylation values are “very small,” oscillating between 0.3 and 4.5 percent. “I don't see how these could have an effect on anything,” Irizarry says.

Yehuda does not completely disagree with Greally.  In response to a request for her reaction to Greally’s blog post, she replied, “It is important to neither overinterpret or ignore results from imperfect studies as neither advances the field. We are currently trying to replicate and better understand the findings because an understanding of how effects of parental experiences manifest in the next generation is of interest and would help many who may be affected by parental or historical trauma. Ultimately science is an iterative and self-correcting process which is greatly advanced by peer review and critique.”

Greally concurs that “it remains possible” that stress, anxiety or depression can be inherited trans-generationally, even though a review by Edith Heard, Professor of Epigenetics and Cellular Memory at the Collège de France in Paris, shows “no solid evidence” for trans-generational inheritance of epigenetic associations occurring in anything but plants and nematode worms.

There is plenty of evidence that epigenetic changes experienced during pregnancy or early development, for instance as a result of diet, or exposure to toxins, can affect you as an adult, but this is not the same as inheriting those changes from a parent.

Yet studies of inheritance of trauma that will likely continue to reap media attention. Why?

“It’s a great story! It’s so compelling, it’s so easy for people to believe, it’s intriguing,” Greally says. “That’s the problem with epigenetics, the stories get out of hand, because they’re such wonderful concepts. In a sense you don’t want to destroy that because that’s why this is a very interesting field. But at the same time if we allow the Deepaks of this world to take it over, we’re going to destroy the brand of epigenetics and no-one is going to take even the good science seriously.”