Two long-sought diabetes genes found: Surprise finding opens new avenues of diabetes research, clinical intervention

Two long-sought diabetes genes found

Surprise finding opens new avenues of diabetes research, clinical intervention

December 5, 1996

Two interacting genes, both unlikely suspects as a cause of adult-onset diabetes, have now been convicted. Mutations of either gene trigger an early-onset form of this extremely prevalent disease, report research teams based at the University of Chicago's Howard Hughes Medical Institute, the Rockefeller University (New York), the University of Michigan, the Wellcome Trust Centre for Human Genetics in Oxford, England, and the Institute Pasteur de Lille, France, together with other investigators from Denmark, Japan and the United Kingdom, in back-to-back papers in the December 5, 1996 issue of Nature.

Both genes evaded scrutiny for years because neither was thought to be involved in the control of blood glucose levels.

But two overlapping research teams, each led by Graeme Bell, PhD, Louis Block Professor of Biochemistry & Molecular Biology and of Medicine in the University of Chicago's Howard Hughes Medical Institute, traced the disorder to mutations of two similar genes one on chromosome 12 and one on chromosome 20 that regulate the activity of other genes.

"Both of these are known genes but neither one was considered likely to play a role in diabetes," said Bell.

Both genes turned out to be "transcription factors," proteins that control how and when other genes are turned on or off. Both were known to be active in regulating gene expression in the liver, kidney, and intestine.

"This surprising finding means we have to quit thinking of diabetes purely as a defect in glucose metabolism and starting thinking of it as possibly a defect in gene expression," said Bell. "That makes diabetes a more complicated disorder, but it also opens up new ways of treating it.

"The discovery of these genes means that we have to place more emphasis on treating diabetes in families rather than in individual patients," added Bell. "We need to screen the brothers and sisters and other relatives of those with diabetes and test methods to prevent the disease in those with a genetic propensity to it."

The two genes each cause a form of non-insulin dependent diabetes (NIDDM) known as maturity-onset diabetes of the young (MODY). MODY affects an estimated one percent to 18 percent of those with NIDDM, the prevalence varying widely between different ethnic groups.

Molecular biologists are particularly interested in MODY because it provides a way to unravel the complex genetics of adult-onset diabetes. Most people with NIDDM develop symptoms only after age 40, which has made it difficult to gather the large multi-generational families needed for genetic studies.

But careful analysis of several extended families with MODY has provided the thorough pedigrees needed to study the genetics of this disease.

Since the 1980s, Bell has been working with endocrinologist Stefan Fajans, MD, professor emeritus at the University of Michigan, who spent decades studying one large family--more than 360 members spanning six generations--in which 74 people have diabetes. Most of those 74 have MODY, which can often be detected in early adolescence.

In 1990, Bell mapped MODY1, the gene responsible for the disease in this family, to a small region on chromosome 20, but did not pinpoint the precise gene. This landmark paper was the first time genetic techniques had been used to determine the chromosomal location of a gene that could cause diabetes. It allowed the researchers to successfully predict which children from the family would eventually develop diabetes.

Since then, Bell's team, and that of Phillippe Froguel, Institute Pasteur de Lille, and others have identified other diabetes-causing genes. They identified a specific gene (for the enzyme glucokinase) as MODY2 in 1992. In 1995 they mapped MODY3 to a specific region on chromosome 12. Last spring, Bell and colleagues mapped NIDDM1, the gene responsible for a significant proportion of diabetes in Mexican-Americans, to one end of chromosome 2.

In September, a Scandinavian study indicated that MODY3 may be the same gene as NIDDM2; different forms of the gene, the researchers suggested, may contribute to both early- and late-onset versions of NIDDM. Now both MODY1 and MODY3 have been identified.

Despite a five-year head start in the search for MODY1, MODY3 was found first. It was discovered by brute force, painstaking scrutiny of every gene within the targeted region, an effort led by Kazuya Yamagata, M.D., at the University of Chicago, and Roger D. Cox, PhD, at the Wellcome Trust Centre for Human Genetics. The researchers found that patients from MODY3 families had one of several different mutations in the gene for hepatocyte nuclear factor 1 alpha (HNF-1-alpha) but healthy subjects had normal copies of the gene.

A few young normal subjects from these families had also inherited the mutant gene. They are expected to develop diabetes in the future.

Finding MODY3 led to the rapid discovery of MODY1, a functionally related gene known as HNF-4-alpha, which had been mapped to the MODY1-region of chromosome 20 by Markus Stoffel, MD, Rockefeller University, a collaborator in the search for MODY1 since 1991. One of the functions of HNF-4-alpha (MODY1) is to regulate the expression of HNF-1-alpha (MODY3).

HNF-1-alpha and HNF-4-alpha are members of a regulatory pathway first identified in cells of the liver but present in other cells including those of the kidney and intestine. This pathway was not considered to be important in the insulin-secreting cells of the pancreas and thus was overlooked in the search for diabetes genes.

There are other genes in this pathway in addition to HNF-1-alpha and HNF-4-alpha. Genes from this family have become leading candidates for other forms of MODY as well as common late-onset NIDDM.

The findings have also stimulated the search for a substance that could increase the expression of either of these genes.

"Many transcription factors respond to other proteins that trigger increased activity," said Bell. "The identification of this biochemical signal could lead to new approaches for treating diabetes."

NIDDM affects an estimated 100 million people worldwide, including 15 million in the United States, where it is the seventh leading cause of death. According to the National Institute of Diabetes and Digestive and Kidney Diseases, the direct costs of medical care for people with diabetes exceed $45 billion, with an additional $47 billion lost to disability and premature mortality. Nearly 2,000 new cases are diagnosed every day in the U.S., for a total of 625,000 each year.

Support for these projects was provided by the U.S. Public Health Service; Wellcome Trust; Bristol-Meyers Squibb; the Juvenile Diabetes Foundation International; the Blum-Kovler Foundation; Pew Charitable Trust; Irma T. Hirschl Trust; the Ministries of Education, Science and Culture (Japan); Ministry of Research (France); the European Union, and the British Diabetic Association.

Two interacting genes, both unlikely suspects as a cause of adult-onset diabetes, have now been tied to an early-onset form of diabetes, report research teams based at the University of Chicago. The finding makes diabetes a more complicated disorder, but it also opens up new ways of treating it.