Above and beyond DNA

Epigenetics, literally meaning 'above the DNA', is the study of how proteins and other molecules that bind our DNA and chromosomes can change gene expression without changing the DNA sequence. Cells use epigenetic changes during embryonic development to regulate gene expression as needed, and to prevent genetic problems through life.

Faulty epigenetics have been linked to many diseases, including cancer and psychiatric disorders, and mounting evidence indicates that diet and other environmental influences could directly affect epigenetics. The challenge for researchers in this field is to study these phenomena meaningfully, on both the single-gene and genome-wide scale, and to ascertain how epigenetic changes could impact on health status in both children and adults.

In 2006, Jeff Craig set up an epigenetics laboratory at the Murdoch Children's Research Institute in Melbourne, in partnership with Dr Richard Saffery. Their aim was to investigate epigenetic components of childhood and adolescent disease and to identify epigenetic factors such as DNA methylation and histone modification that contribute to human disease.

After a set-up phase and some pilot studies, work at the new lab is just starting in earnest, inspired by two bodies of evidence. "Firstly, we became really interested in the Developmental Origins of Health and Disease (DOHaD) theory proposed by researchers in the UK several years ago," Craig says. "This emerging field is based on the hypothesis that chronic illnesses such as heart disease, diabetes, osteoporosis, mental health, cancer and obesity originate during foetal development or shortly after birth."

To test the hypothesis, infants with low birth weight for gestational age were followed through life and found to have a much higher rate of these complex diseases, particularly cardiovascular disease. Craig was fascinated by these results from the viewpoint of epigenetics, and this interest was stimulated further by recent animal experiments showing that alterations in the foetal environment, such as in diet or stress levels, could influence the health and gene expression of the offspring.

The second major influence on Craig and his team was a low-resolution study by a group in Spain that compared an epigenetic 'barcode' within pairs of twins. The Spanish scientists compared old and young twins and found that DNA methylation states were more different between older twins than between the younger ones. Craig and his colleagues were intrigued to see if the same changes could be measured at birth and thereafter, which nobody had looked at so far.

So, they set about investigating the epigenetic differences between newborn twins, looking at the correlation of maternal nutrition with epigenetics at birth and into early life. "Twins are the best model to study this link as they share similar, but not identical, environments, and some share identical genetics," he says.

---PB--- You are what you eat

Embryonic development comes with a program that is basically written in epigenetics - genes that are no longer needed are turned off for good once they have done their job, such as those needed for growing an arm, and others keep working, Craig says.

"The genes are turned off mainly by DNA methylation and histone modification. We and others now strongly suspect that environmental conditions can affect these genetic changes and push the programming off course and that this happens via epigenetics - sometimes called re-programming or epigenetic de-regulation."

Nutrition is an important aspect of environment, and folate is an example of a dietary component that has been shown to influence DNA methylation and gene expression. For example, studies of famine events reveal that if a foetus receives less food it re-programs its metabolism to deal more efficiently with less food - to expect less so make the most out of everything you get. There is one hypothesis that when the baby is born and gets an excess of food, conditions such as obesity can result. However, the timing and the overall contribution of environmental influences to epigenetic de-regulation remain largely unknown.

There is also some evidence from the famine studies of a transgenerational effect on health - meaning that an individual's health problems are seen in their grandchildren despite improvements in diet and other aspects of environment. The most critical times for the programming seem to be in utero in males and females and just before puberty in men. If the program is disturbed at these crucial points, not only can it re-program your life but the epigenetic changes and any associated health predisposition can pass on to the next generation.

To address their suspicions, Craig and his team are studying a group of women pregnant with twins. Participating mothers are asked to complete a number of questionnaires throughout the pregnancy and submit samples for testing of factors such as nutritional status. It is known that folate and other vitamins as well as omega 3 and 6 affect gene activity, although the precise involvement of epigenetics in these effects is not known.

Further samples are taken at birth, including placenta, cord blood and whole cords, which are very important for analysis of complex diseases. "We are interested in a few things, the simplest being what are the epigenetic differences between identical twins (monozygotic) at birth compared to non-identical twins (dizygotic)," Craig says. "We think there will be some differences, especially in twins with significantly different birth weights. It is hoped that genes exhibiting the largest epigenetic differences within these pairs will provide some clue to the underlying pathways linking low birth weight and risk of chronic disease in later life."

Once all mothers and twins are recruited and tested, differences in several target genes will be examined. A whole-genome microarray approach will be used to assess changes in gene expression (expression array studies are currently underway on same birth-weight twins to get baselines) and DNA methylation state (using a technique called methylation-dependent immunoprecipitation to attain essentially a methylation signature).

"We really do not know quite what to expect until we do it, but we hypothesise that the discordant-weight twins will show some differences, and the epigenetics differences we find may reflect the twins' health in later life," Craig says. "The current plan is to track them to 18 months of age. No other studies have monitored epigenetics changes over time and with different environments."

Ideally, the group would like to follow the twin pairs further into life but this will depend on the first set of results and of course on funding. The group is looking at expression and methylation data for two twin pairs by microarray and RT-PCR analysis to back up any of the interesting microarray data. Already, subtle expression differences are starting to emerge, although the significance of these differences will become apparent only when more twin pairs are born and analysed.

---PB--- Full circle

The final aim of the twins study is to relate differences in epigenetics between all twin pairs to pair-specific differences in maternal nutrition. For instance, the team will examine the general relationships between global levels of DNA methylation and levels of metabolites such as folate.

"Analysing genetics entails a lot of 'noise' that can mask epigenetic changes," Craig says. "Studying monozygotic twins avoids this by controlling for genotype and therefore any differences detected are either random or due to small changes in environment such as share of the placenta. On the other hand, the dizygotic or fraternal twins are interesting because they are essentially controlled for environment but have different genetics."

Craig is coming full circle at this year's Lorne conference after being 'converted' to microarray-based approaches to epigenetics at Lorne about four years ago. "It blew me away when I saw a talk on epigenetics by someone who had put the whole yeast genome on a microarray and hybridised their particular fractions of chromatin," he says. "They could then make conclusions about the kind of epigenetic modifications that were going on in the yeast genome.

"I thought there and then that is the field for me, and I saw the future - a diagnostic test. Any test that can help predict the onset or severity of disease, even in combination with genetic tests, will be useful, especially for those complex conditions such as psychiatric, metabolic and cardiovascular diseases for which few tests are currently available.

"The most exciting area of epigenetics will be finding associations between disease, looking at epigenetic changes when someone has a disease, but also analysing the changes over time, and more importantly the health predisposition of an individual. Can we look at people at birth and predict the kinds of diseases they might get? And because epigenetic marks are potentially reversible, can we then treat with specific drugs and prevent or even reverse the process?"