Framework for systematically incorporating epigenetic information into traditional genetic studies

Jul 29 2004

Scientists at Johns Hopkins are calling for simultaneous evaluation of both genetic and epigenetic information in the search to understand contributors to such common diseases as cancer, heart disease and diabetes.

Writing in the August issue of Trends in Genetics, available now online, the scientists provide a framework for systematically incorporating epigenetic information into traditional genetic studies, something they say will be necessary to understand the genetic and environmental factors behind common diseases.

"Epigenetics doesn't underlie all human disease, but we definitely need to develop the technology to figure out when and where epigenetic changes do influence health and disease," says Andrew Feinberg, M.D., King Fahd Professor of Medicine.

Much as the genetic sequence is passed from parent to child, epigenetic "marks" that sit on our genes are also inherited. These "marks," usually small methyl groups, are attached to genes' backbones and convey information, such as identifying which parent the gene came from. The marks also normally turn genes on or off. But just as changes in DNA sequences can cause diseases such as cancer, gain or loss of epigenetic marks can, too.

To date, only small, targeted regions of DNA have been analyzed for accompanying epigenetic marks. But the Hopkins researchers say now is the time to begin studying epigenetics on the same mammoth scale used to probe the sequence of creatures' genetic building blocks.

But to establish what's "normal," epigenetically speaking, for the entire genome, Feinberg says scientists will need new technologies to quickly, accurately and inexpensively determine which epigenetic marks are present and where they are, much as "high-throughput" technology revolutionized genetics.

"That kind of power is needed to create comprehensive epigenetic information, but right now the technology doesn't exist," says Feinberg, who pioneered the study of epigenetics in cancer. "Developing that technology, and the necessary statistical approaches to analyze the data, will require a major collaborative effort and should be first on the to-do list."

A first step toward these goals is the new, multi-institutional Center for the Epigenetics of Common Human Disease, which Feinberg is directing. With a $5 million, five-year grant from the National Human Genome Research Institute and the National Institute of Mental Health, Center researchers will develop the tools they need and then begin systematically examining the epigenetics of autism and bipolar disorder. But a broader effort will be necessary, Feinberg says.

"Just as geneticists are probing samples to identify genotypes and haplotypes [lengthy genetic sequences that are inherited in blocks], we need to examine multiple samples from different tissues and different people to establish 'epigenotypes,'" says Feinberg. "Only by superimposing genetic and epigenetic information will we get a complete picture of how genes' functions are affected in healthy people and in those with particular diseases."

Feinberg and co-authors Hans Bjornsson and epidemiologist Daniele Fallin say current attempts to establish genetic contributors to common diseases will fall short without equivalent epigenetic information.

It's widely believed that common, complex diseases like cancer, diabetes and heart disease stem from collections of changes in a number of as-yet-unknown genes because of the wide variation in severity, age of onset, ease of treatment, rate of progression and other factors.

But the Hopkins scientists point out that epigenetic variation adds another layer of complexity that also could contribute broadly to diseases' variability. How much epigenetic marks vary normally is still unknown, but faulty epigenetics are already known to be at work in cancer and relatively rare Beckwith-Wiedemann syndrome, among others conditions.

"The epigenetic marks a person has could influence disease directly, but they also could affect whether an underlying genetic mutation or genetic variation can actually result in biologic or physiologic changes," says Bjornsson, a physician from Iceland who is pursuing his Ph.D. in human genetics at Hopkins. "We think the latter is going to be very important in explaining the variability of the most common diseases."

For example, if a disease-causing mutation is present in a gene turned "off" by its epigenetic marks, then the mutation can't cause disease. More subtly, alteration of epigenetic marks could "tune" gene expression to cause a full spectrum of effects. Epigenetic variation is also likely to help relate environment and age to disease incidence and risk, the researchers say.

"Common complex diseases occur more often as people age, and genetics alone hasn't fully been able to account for that by accumulation of mutations," says Feinberg. "But epigenetic marks might more easily change as cells and people age or be more easily influenced by environmental factors than the actual DNA sequence is. We need to look at those possibilities."

The researchers were supported by grants from the National Human Genome Research Institute, the National Cancer Institute, and by a Fulbright Scholarship to Bjornsson.