Diabetes marker pegged as brain enzyme.

Diabetes marker pegged as brain enzyme

For nearly a decade, diabetes researchers have studied a protein known as 64K, produced in tiny amounts by the insulin-secreting beta cells of the pancreas. Antibodies to 64K circulate in the blood of about 80 percent of people who later develop insulin-dependent (Type I) diabetes or who already have the disease, which appears to involve a misguided immune-system attack on beta cells (SN: 6/18/88, p.389).

But attempts to use the presence of 64K antibodies as a widespread screening test for Type I diabetes have failed because of difficulties in identifying and purifying the relatively rare pancreatic protein they target.

Now, scientists report that 64K appears identical to an abundant and easily isolated enzyme of the central nervous system, known as glutamic acid decarboxylase, or GAD. Since other researchers already have cloned one form of the enzyme, the finding may speed efforts to develop an inexpensive screening test for Type I diabetes. Michele Solimena of Yale University adds that the work may also provide new information about the causes of the disease, which affects up to 1 million people in the United States.

Solimena, with colleagues from Yale and the University of Milan in Italy, proposed an association between GAD and 64K in the May 31 NEW ENGLAND JOURNAL OF MEDICINE. They noted that GAD, which helps synthesize the neurotransmitter GABA, comes under antibody attack in a rare autoimmune disease known as stiff-man syndrome. A significant number of the estimated 300 people with the muscle-stiffening disorder also suffer Type I diabetes. Moreover, 64K antibodies in diabetics without stiff-man syndrome show similarities to those targeting GAD, they found.

Now, in the Sept. 13 NATURE, Solimena and co-workers from Yale and the University of California, San Francisco, report that the two compounds exhibit identical chemical behavior, indicating they are one and the same.

Solimena notes that antibodies against the full GAD molecule cannot directly cause diabetes because GAD lies inside cells, away from immune-system components that might be activated to kill beta cells. He speculates that GAD fragments may migrate to the surface of beta cells to trigger such wholesale destruction. He adds that 64K's identification as a neuronal compound does not indicate that Type I diabetics are prone to neurological disorders -- in part because pancreatic antibodies would have to cross the blood-brain barrier to cause such damage.

But the discovery does emphasize the similarity between nerve and endocrine cells, researchers note. At the European Congress of Cell Biology, held last week in Florence, Pietro De Camilli of Yale and his colleagues reviewed their findings that beta cells and other hormone-secreting cells possess storage compartments similar to those used by some neurons to hold neurotransmitters. They say this suggests beta cells may secrete GABA, possibly to help regulate blood sugar.