Traces of Trauma in Sperm RNA

A mouse study shows that molecular remnants of early-life stress can be passed on to future generations.

Apr 13, 2014
Jeffrey M. Perkel

WIKIMEDIA, RAMAThe offspring of mice that suffered early-life stress show signs of the disturbance their parent experienced, researchers from the University of Zurich in Switzerland and their colleagues have found, pointing to a potential RNA-based mechanism by which trauma may be epigenetically inherited. The team’s work was published today (April 13) in Nature Neuroscience.

Zurich’s Isabelle Mansuy and her collaborators sought to evaluate the environmental and genetic factors behind complex neurological diseases associated with childhood trauma, such as borderline personality disorder. The team used a mouse model of “unpredictable maternal separation combined with unpredictable maternal stress,” or MSUS. The MSUS mothers were separated from their young at different times once daily for two weeks. While the unpredictability of this event traumatized the pups, the mothers, too, were stressed during these separation periods, for instance, by being confined to a narrow tube.

When the MSUS young became adults, the researchers found that they were generally more likely to take risks than control adult mice. The MSUS mice were less hesitant to enter wide-open spaces or brightly lit areas than animals that had not been separated from their mothers, for instance. They also exhibited altered glucose metabolism, implying that early developmental trauma had caused permanent behavioral and metabolic changes.

And those changes were not confined to the once-traumatized adults. When the researchers bred those MSUS mice mated with control animals, they found that the progeny, too, were less risk-averse and showed more signs of depression than control animals. Even the offspring’s offspring exhibited altered glucose metabolism.

“This paradigm induced behavioral changes across generations,” said Mansuy.

To discern how such stress could be passed on, the researchers isolated RNA from MSUS animals’ sperm, sera, and hippocampi, and compared it with control animals. They detected changes in the abundance of several noncoding microRNAs (miRNAs) and PIWI-interacting miRNAs (piRNAs) in sperm, including some that were previously implicated in epigenetic regulation. These molecular changes typically carried over to the F2 generation, where differences in RNA abundance were seen in the hippocampus and serum, but were largely gone by F3. The behavioral changes remained in both generations.

Then, to solidify the observed link between RNA and behavior, Mansuy and her colleagues isolated RNA from sperm of MSUS mice, injected it into fertilized eggs, and reimplanted the eggs into pseudopregnant females. When the resulting pups became adults, they exhibited the same metabolic and behavioral traits as the MSUS mice.

“That was the best and most causal evidence we could provide,” said Mansuy.

The study represents “quite an exciting piece of work,” said Stephen Krawetz, the associate director of the C.S. Mott Center for Human Growth and Development at Wayne State University School of Medicine in Detroit, Michigan, who studies sperm RNA but was not involved in the research.

“What it’s doing is building on the notion that dad’s contribution is actually more than just his genes when he fertilizes the oocyte,” Krawetz continued. “[It] really adds a new dimension in terms of what impact dad can have.”

The mechanisms that mediate such impact, however, remain unclear, said Minoo Rassoulzadegan of the University of Nice in France, who advised Mansuy’s team on how to isolate RNA from sperm. The microinjection experiments demonstrated that stress-associated RNA can induce phenotype, she said, “but we don’t know how. This is the question for the future, to find out what [the RNA] is doing to the genome.”

Mansuy’s team is now focusing on pinning down how stress alters the germ cells’ RNA profiles, and what, exactly, those altered RNAs target.

K. Gapp et al., “Implication of sperm RNAs in transgenerational inheritance of the effects of early trauma in mice,” Nature Neuroscience, doi:10.1038/nn.3695, 2014.

Correction (April 13): This article has been updated to reflect that when examining RNA from mouse sperm, serum, and hippocampus, Mansuy’s team detected changes in the abundance of miRNAs and piRNAs only in spem, not in “all three sources” as originally stated. The Scientist regrets the error.