Natural selection in the womb can explain health problems in adulthood

The conditions encountered in the womb can have life-long impact on health. Scientists have previously assumed that this is because embryos respond to adverse conditions by programming their gene expression. Now researchers propose a radically different alternative. Rather than being programmed by the environment, random differences in gene expression may provide some embryos with a survival advantage, in particular when conditions are harsh. By studying DNA methylation, an important mechanism to control gene activity, the researchers found that a specific part of the DNA methylation pattern was missing among famine-exposed individuals.

Growth with limited resources

The new research was motivated by the observation that people conceived during the Dutch Hunger Winter of 1944-1945 suffer from reduced metabolic health in their sixties. This can be attributed to persistent changes in how genes are expressed, through so-called epigenetic modification of the DNA. “We know that a lack of nutrition decreases the likelihood of an embryo to survive. Our new study indicates that surviving famine in the uterus hinged on having a DNA methylation pattern allowing continued growth of the embryo in spite of limited resources. But those same methylation patterns may have adverse health effects much later in life”, says Bas Heijmans, epigeneticist at the Leiden University Medical Center.

Computer model

To understand the interplay between epigenetics and survival of the embryo, the researchers took inspiration from evolutionary biology. In evolution, random genetic variation is filtered by natural selection, resulting in accumulation of variants that best ‘fit’ the environment. A computer model showed that random epigenetic variation between embryos is inevitable, just like genetic mutation. Some of the random DNA methylation variants may enhance an embryo’s chance to survive on low nutrition. As a consequence, those epigenetic variants will become more common in cohorts that were exposed to a famine as embryos. “We have always struggled to explain how early embryos would be able to modify specific epigenetic marks in response to nutrition. It is fascinating that selective survival based on random epigenetic variation fits the data best”, says Tobias Uller, evolutionary biologist at Lund University.

Health does not equal fitness

Some health effects of the Dutch Hunger Winter only show later in life and those exposed during early gestation seem to be most affected. “These findings have often been interpreted as conclusive proof of fetal adaptation in the womb that will lead to adult disease if the environment changes for the better. But our findings point to a different mechanism”, says Professor Lumey, epidemiologist at Columbia University and principal investigator of the Dutch Hunger Winter Families Study. “Indeed, medical fitness does not necessarily equal evolutionary fitness and in human cohorts the data to infer whether adaptation has occurred are often impossible to obtain. Caution is warranted as the hypothesized predictive adaptive plastic response to developmental adversity has evolved in many species, but is completely absent in others”, explains first author Joost van den Heuvel, evolutionary biologist at Wageningen University, “with Professor Bas Zwaan we employ various modelling techniques to test if, and how, such responses can evolve. The key novelty of this paper is that we could test our theoretical model with data that are available, rather than the data we would normally use. The DNA methylation data allowed us to distinguish the underlying mechanism, adaptive plasticity versus “epigenetic” selection.”

New insight

The team came together in European networking and research projects and via repeated discussions on the various possible hypotheses and how to test them, the manuscript slowly came to fruition over many years. “It is a truly cross-disciplinary effort,” says first author Elmar Tobi at Wageningen University & Research, “and shows that evolutionary biology and the techniques employed to study it are relevant to make sense of human biology.” The distinction between adaptive plasticity and selection has implications. When there is adaptation, later life health effects may be preventable by avoiding harmful exposures of the fetus. With selection on embryonic characteristics, one may only hope to reduce, but could never entirely prevent, all late-life health effects linked to harmful exposures during early development.

Professor Bas Zwaan: ”Getting this paper published has also been a bumpy ride. Most journals we contacted were interested, but had trouble fitting our manuscript in. Surprisingly, presenting a novel idea, a hypothesis, with some additional work to showcase how to test it, is uncommon in science these days. Now that our hypothesis has finally found a home at Cell Reports, we hope that others will take up the challenge and look at their data on prenatal exposures and adult health with fresh eyes, and proof it correct or wrong.”