Gene Regions Beyond Protein Instructions Important In Disease

"It's a funny mutation in a funny place," says study leader Aravinda Chakravarti, Ph.D., director of the McKusick-Nathans Institute of Genetic Medicine. "But I think the majority of mutations found in major diseases are going to be funny mutations in funny places."

Far from being a problem, the finding is good news, he suggests. "Mutations in the protein-coding sequence can't really be fixed, but those outside the protein-coding regions -- perhaps we can fiddle with them, perhaps they are 'tunable.' The protein should be fine if we can just get the cells to make the right amount," he says.

"Our finding really underscores the fact that health and disease can be affected by all regions of a gene," he continues. "For diseases like diabetes and heart disease, just as for Hirschsprung disease, multiple inherited factors contribute to the disease, and these factors are not just going to be in protein-coding regions."

The researchers' discovery, described in the April 14 issue of Nature, adds to growing evidence that problems with the amount of protein made from a gene's instructions are likely to be just as important as - and perhaps more important than -- changes in the proteins themselves, they say.

"But finding important mutations outside of protein-coding sequences is a challenge because of the amount of genetic material to sort through," notes postdoctoral fellow Eileen Emison, Ph.D., the study's first author. "Only 1.5 percent of the roughly 3 billion building blocks in our genetic material carry instructions for proteins."

Fortunately, about twice that much has stood the tests of time and evolution and remains the same, or very similar, among various species, indicating the regions' biologic importance. By comparing the genetic sequences of humans and other species to find these regions, and then combining those results with traditional genetic studies of disease in families, the hunt for disease-related mutations in so-called non-coding sequences can be successful, the researchers show.

In fact, the researchers used this combined approach to discover the risk-increasing mutation in the RET gene in individuals with Hirschsprung disease. In this birth defect, the effects of multiple genetic mutations -- many still unknown -- combine to prevent proper development of the nerves that control intestinal function. Only 30 percent of Hirschsprung cases have been tied to a specific protein-changing mutation, even though protein-encoding regions of eight genes already are known to be involved in the disease.

The new risk-confirming mutation confirms Chakravarti's long-held suspicion that some of Hirschsprung's unknowns might be due to mutations in non-coding regions, which usually are not included in the hunt for disease-related mutations. A gene's non-coding regions -- which don't have to be adjacent to or even near a gene's protein-coding sequences -- contain the gene's on-switch (the promoter), areas that tweak whether, when and how the gene is used to make proteins (enhancers or suppressors) and other expanses that still just seem to be filler. The new mutation is in a gene called RET, whose protein-coding sequence had already been tied to the disease.

To hunt for Hirschsprung-related mutations in the largely uncharted non-coding regions, Chakravarti and his team first determined the identities of 28 specific genetic building blocks, or markers, in a large region surrounding the RET gene in samples from 126 people with Hirschsprung disease and their parents. (Earlier work had tied the disease in these families to a large region that includes RET, but no protein-changing mutations had been found in affected individuals.)

The genetic markers' identities act as a sort of signature the researchers can track. Computer analysis identified three large regions of DNA, one including the RET gene, that were passed from parents to affected children (but not unaffected children) more often than one would expect by chance alone. One particular eight-marker signature around RET was most tightly associated with the disease, the researchers found.

Rather than sequencing the entire region in all the families, the researchers turned to comparative genomics to focus the search. Colleague Eric Green, Ph.D., and others at the National Institutes of Health determined the genetic sequences of a large region surrounding the equivalent of RET in 12 nonhuman vertebrates, including the chimpanzee, cow, mouse, dog, chicken and blowfish, for comparison to the human sequence (determined by the Human Genome Project).

"We found 84 areas within the region that were highly conserved, almost half of which were protein-coding areas of the RET gene and two other genes," says Emison. "That left us with 47 areas that didn't carry instructions for proteins, but that were likely to be both biologically important and involved in the disease."

By overlapping the disease-linked regions and the smaller, highly conserved genetic snippets, the researchers uncovered five short areas within the overall RET gene on which to focus. Sequencing these five areas in patients revealed the culprit -- a genetic sequence that was identical in all mammals studied and in all unaffected individuals. In those with Hirschsprung, however, the sequence contained a single change.

In laboratory studies, Andrew McCallion, Ph.D., an assistant professor in Hopkins' Institute of Genetic Medicine, and graduate student Elizabeth Grice determined that this region of RET normally enhances the gene's activity. The mutation diminished that effect.

"Not everyone who has the mutation has the disease, but our analysis shows that the mutation clearly contributes to the risk of disease," says Emison. "Interestingly, the frequency of the mutation in different world populations mirrors that of the disease."

The frequency of the mutation, ranging from almost absent in Africa to 50 percent in Asia, is much higher than the incidence of the disease, which affects roughly 1 in 5,000 births, on average. The mutation is almost twice as common in Asians as in Europeans, and a study in the 1980s showed that Asian Americans in California were twice as likely to have a child with Hirschsprung disease than mothers of European descent. The mutation's distribution also mirrors the greater incidence of the disease in boys, the researchers report.

The Johns Hopkins researchers were funded by the National Institute of Child Health and Development. Authors on the paper are Emison, McCallion, Grice, Chakravarti, Carl Kashuk, Richard Bush, Shin Lin and David Cutler of the McKusick-Nathans Institute of Genetic Medicine at Johns Hopkins; and Matthew Portnoy and Green of the Genome Technology Branch and the Intramural Sequencing Center at the National Institutes of Health.

On the Web: http://www.nature.com/nature