Epigenetics

['lizbeth]

A chapter from 100 Most important Science Ideas by Mark Henderson Joanne Baker and Tony Crilly

Epigenetics

It was previously thought that a person’s genetic makeup is arrived at independently of any environmental influences, but it now appears that sometimes acquired characteristics may, in fact, be passed on in the genes to succeeding generations.

In the autumn of 1944, railway workers in the Netherlands, then under German occupation, went on strike to assist the advancing Allies. When the initial British and American assault failed, the Nazis retaliated by imposing a devastating food embargo. At least 20,000 Dutch citizens starved or died of malnutrition in the ensuing famine.
The effects of the Hongerwinter or “Hunger Winter” were to last long beyond the country’s liberation in 1945. Mothers who were pregnant during the famine had clidren with an elevated risk of a wide range of health problems, including diabetes, obesity and cardiovascular disease. In some cases, even their grandchildren were more likely to be born underweight. While damage to the first generation’s health might be explained by malnutrition during pregnancy, the Netherlands was a rich nation by the time the second generation was born. Yet, an inherited effect remained.

The story of the Dutch famine is not unique. The village of Överkalix, in northern Sweden, boats meticulous historical records of harvests, births and deaths, which have allowed Marcus Pembrey, of the Institute of Child Health in London, to conduct and exhaustive study of food availability and life expectancy. He found that when boys grew up during periods of plenty, their grandsons were more likely to die at a young age. Further analysis revealed that this reflected a predisposition only through the male line.

Both examples suggest that people’s health can be affected by the diets followed by their grandparents. Yet according to orthodox evolutionary theory, such an effect should be impossible. Acquired characteristics are not supposed to be inherited – that was the Lamarckian heresy, which has not been fashionable since Darwin.

“We are changing the view of what inheritance is. You can’t, in life, in ordinary development and living, separate out the gene from the environmental effect. They’re so intertwined.” Marcus Pembrey


Genetic memory

The Dutch and Swedish experiences can be explained by a phenomenon know as epigenetics, by which the genome appears to “remember” certain environmental influences to which it has been exposed. Normally, these epigenetic effects act only on the somatic cells of the adult body, switching genes off or otherwise adjusting their activity. Some, however, can also alter sperm and eggs, to be inherited by future generations. Acquired characteristics, it turns out, can sometime be passed on after all.

Epigenetics owes its prefix to the ancient Greek for “over” or “on,” and it generally relies on two broad mechanisms. One is methylation and the process we met in Chapter 20, which silences genes by adding part of a molecule called a methyl group to the DNA base cyosine, or C. The other is modification of chromatin, the combination of DNA and histones (kinds of protein), of which chromosomes are made up. Changes to chromatin structure can affect which genes are made available for transcription into messenger RNA and protein, and which are hidden away out of reach. In neither case is the actual sequence of DNA altered at all, but changes in its organization can still be passed on from one cell to its offspring.

These epigenetic process are central to normal growth, development and metabolism. Every cell contains the full set of instructions that are needed by every type of tissue, and epigenetics determines which of these are actually issued and executed. It ensures that genes act in the body in response to environmental cues. Experiments with mice have demonstrated the lucidly: changes in diet while females are pregnant affect the coat color of their offspring, by modifying the way genes are methylated. This effect, indeed, could account for the surprising observation that many cloned animals differ from their parents in color and markings. While their genomes are identical, their “epigenomes” are not.

Normally, such epigenetic changes are stripped away from the genome during embryonic development, so that they are not passed on to offspring. But they can sometimes be retained, causing environmental effects on health and behavior that cascade down the generations. That could account for what happened in Holland and Sweden/ Prenatal diets seem to have changed the epigenetic programming of their children and grandchildren, to modify metabolism to cope with the prevailing nutritional environment. This, in turn, has influenced health risks such as diabetes.

The importance of the epigenome

As with copy number variation and junk DNA, science is starting to understand that epigenetic effects are just as significant to biology as convention genetic mutations. Epigenetics, for example, plays an important role in cancer. Several chemicals are known to be carcinogenic, even though they are not mutagens that directly damage DNA. They induce epigenetic effects, silencing important tumor suppressor genes, or altering chromatin structure so that oncogenes become more active.

Epigenetic markings also ensure that when cancer cells divide, their daughter cells are cancerous too. An understanding of how these processes work could open a new approach to medicine. The first cancer drug that works by removing methylation, Vidaza, was approved by the U.S. Food and Drug Administration in 2004.

The Human Epigenome Project, recently started by a European consortium, should help to bring more epigenetic therapies into medicine. This ambitious initiative aims to map all the possible methylation patterns of every gene in every type of human tissue. A pilot project has already achieved this for the major histo-compatibility complex – a cluster of genes on chromosome 6 that affect immune response.

Once these methylation sites have been identified, it should be possible to link variation to diseases, in much the same way as can be done for SNPs. Doctors, indeed, may well find their patients’ epigenomes more useful than their genomes. The genetic code, as the early history of gene therapy has proved, is painfully difficult to correct in living organisms; methylation should be comparatively simple to undo. Many of the medicines of the future could be designed to exploit this natural method of genetic control to treat and prevent disease.