Male mice trained to fear a particular harmless smell before they sire offspring can pass the memory of that specific fear on to their children and grandchildren.
Might this finding mean anything for humans?
As John Timmer points out at Ars Technica, inheriting memories is controversial. If an organism is exposed to a particular environmental stimulus before any of its descendants were conceived, the experience can influence the nervous systems of those descendants even though the organism had no contact with them. The proposition is that this is a biological transmission of a specific fear, or maybe the memory of that fear, not only to the organism’s offspring, but to their offspring.
And it was a clever and persuasive experiment. For example, the researchers showed that the effect was biological by ruling out parental social behavior as an influence. They did this by training male mice to fear the cherry-smelling chemical acetophenone by pairing it with an electric shock, and then using the trainees’ sperm to inseminate females in a different lab across the campus. The offspring shuddered in the presence of acetophenone even though they had never smelled it before and were not being shocked. And so did their offspring.
J. David Sweatt, who directs the McKnight Brain Institute at the University of Alabama and was neither involved in the research nor a reviewer on the paper, calls the study “a landmark set of findings.” He told me in an email that he expects the work to face skepticism because it so challenges accepted ideas. He is emphatically not among the skeptics. “This is a beautiful study, executed with great rigor, of high conceptual importance to the emerging field of behavioral neuroepigenetics.”
The most extensive general-audience account of this work has come from Virginia Hughes, at her blog Only Human. She wrote about it first in mid-November following a talk by the researchers at the recent Society for Neuroscience meeting. Hughes said, “I’ve been to a lot of scientific talks. The excitement around this one was notable, with many scientists whispering about it in the room and more loudly buzzing in the hallways outside.” When the paper was published December 1, Hughes took another look, this time explicitly connecting the smell of fear study to transgenerational epigenetics.
What is transgenerational epigenetics?
Epigenetics is not a concept with a consensus definition, which is to say that just what it includes is a matter of scientific dispute. For our purposes, it’s enough to say that epigenetics encompasses a heterogeneous set of biochemical mechanisms that change the behavior of genetic material without changing any DNA sequences. Epigenetics seeks to explain how the environment turns genes off and on in particular cells at particular times. It’s about how nurture shapes nature.
Transgenerational epigenetics is, literally, epigenetics at a whole other level. The idea is that epigenetic alterations acquired in one generation can be passed on to future generations. Transgenerational epigenetics is a pretty well-established phenomenon in plants. There is evidence that transgenerational epigenetics happens in animals, but it’s often controversial, especially when it is about our own species.
How does transgenerational epigenetics happen?
A major stumbling block to scientific acceptance of work on transgenerational epigenetics is that it’s been hard to explain how such a counterintuitive thing—some say a crazy thing—might happen. What biological mechanism can account for it?
The smell-of-fear researchers did spend a bit of their paper’s space pondering how handing down sensitivity to a particular odor could happen. For example, they speculate that blood-borne odorants might activate odorant receptors on sperm.
It’s not in the paper, but an intriguing candidate for a way to achieve transgenerational epigenetics is gene regulation by small noncoding RNAs, in particular the ncRNA known as piRNA. One reason that’s an attractive idea is that piRNAs do nearly all their work in the germline—eggs and sperm. So they are perfectly placed to influence what gets passed on to the next generation.
What piRNAs do in the germline is silence the genomic parasites known as transposable elements, transposons. Transposon DNA is not “genes” and doesn’t code for proteins, but it’s a huge presence in the human genome, accounting for more than half of our DNA. When you hear the term “junk DNA,” you’re hearing mostly about transposons.
Transposons can occasionally be useful, for example by providing raw material for evolution to work on. But they are also destructive, hopping around in the genome and disrupting genes. I wrote about transposons and noncoding DNA and their role in evolution here a few weeks ago.
piRNAs do their transposon silencing in the germline across the animal kingdom, so they have been around for a very long time. One recent review says that the “piRNA pathway is at the forefront of defence against transposons in germ cells.”
But wait, there’s more. The piRNA pathway also affects the best-known epigenetic mechanism, DNA methylation. Adding methyl groups to a gene usually silences it. In addition, there is some experimental evidence that piRNAs also target and regulate expression of protein-coding genes.
So far as I know, there’s as yet no step-by-step account of how piRNA activity could help pass something so undefined as the memory of a fear on to the next generation. But there piRNAs are in eggs and sperm, shutting things down. Scientists are looking at it hard.
Here comes Jean-Baptiste Lamarck?
An editorial in New Scientist wonders whether the smell of fear research means new respectability for a dirty word. The dirty word is “Lamarck.” That would be Jean-Baptiste Lamarck, the biologist whose work was mocked and rejected even in his own 18th century times.
Poor Lamarck, still mocked and rejected today even though some of his work was praised by such illustrious figures as Stephen Jay Gould and Charles Darwin himself. Lamarck is most often associated with the supposedly discredited notion of inheritance of acquired characteristics. That’s the idea that traits taken on during life can be passed on to descendants. The oft-given example, also trotted out by New Scientist: “if giraffes stretch their necks to reach high leaves, their offspring have longer necks.” The New Scientist editorialist says the fear memory work “seems consistent with Lamarckian inheritance . . . It fits with natural selection—and may yet give Lamarck’s name a sheen of respectability.”
About this, New Scientist is mistaken. The smell of fear research really doesn’t fit with natural selection, at least not yet. Lamarckian inheritance of acquired characteristics has been controversial because the characteristics are supposed to be permanent. They shape evolution. All giraffes have long necks, and so do their descendants, even when they are snacking on greenery of moderate height.
The new paper describes the fear memory persisting in the grandchildren of the research subjects. There’s no claim that this process is affecting evolution. In fact, as far as I can tell, the length of this study is typical of the work on transgenerational epigenetics. When biological responses in one generation have been passed on to descendants, the changes characteristically don’t persist beyond three or four generations. They peter out. They do not become an example of evolution. The idea that a fear can be passed on to grandchildren through some so-far-unidentified biological mechanism is startling. But it is not really Lamarckian.
Not so far, anyway. Perhaps there will eventually be papers telling us what happens in the grandchildren of the affected grandchildren in this study. Will the sixth generation be fearful of acetophenone too?
There will almost certainly be papers attempting to replicate this work. Will they confirm it? Or contradict it? Mice have a relatively short generation time, so we might find out quite quickly whether the smell-of-fear results hold up.
Tabitha M. Powledge is a long-time science and medical writer whose work has appeared all over the place, in print and online.