By David N. Leff

Science Editor

It's no secret that overweight people are at risk of developing diabetes mellitus. Paradoxically, babies born underweight also have a heightened risk of diabetes, as well as high blood pressure and heart disease.

Diabetologist and molecular geneticist Andrew Hattersley, at the University of Exeter, in the U.K., is researching this apparent association between low weight at birth and Type II diabetes later in life. In a family study to dissect the presumed genetic and environmental causes of the disease, he told BioWorld Today, "we're seeing a one-pound loss in utero as a result of the genetics."

He and his colleagues have tracked this small-fetus syndrome to mutations in a gene that senses when glucose (sugar) levels in the blood produce more insulin, or less.

"Glucose quite freely crosses the pancreatic beta-cell membrane," explained molecular geneticist and endocrinologist Mark McCarthy, at Imperial College, London. "So the glucose inside the cell is closely parallel to the glucose outside it. [Beta cells, inside the pancreatic Islets of Langerhans, secrete the body's insulin.]

"What happens," McCarthy continued, "is that when glucose is acted upon by the enzyme glucokinase [GCK], it's phosphorylated to glucose-6 phosphate. And the rate at which that happens affects the energy levels of the cell.

"That conversion of glucose to glucose-6 phosphate," he pointed out, "will proceed faster if there's more glucose, and slower if there's less. This is the way the beta cell tells how much glucose there is out there and therefore ultimately how much insulin to secrete in order to close that homeostatic loop, which balances the body's needs."

McCarthy compares this feedback loop to a home thermostat, which is constantly switched on by cold and off by heat.

But trouble arises in this insulin/glucose equilibrium, when as a result of mutations in the GCK gene, beta cells produce less insulin for a given amount of glucose.

Hattersley is principal author of a paper in the July issue of Nature Genetics, published today, July 1, 1998. Its title is "Mutations in the glucokinase gene of the fetus result in reduced birth weight."

McCarthy discusses Hattersley's article in a "News & Views" editorial, "Weighing in on diabetes risk."

Hattersley explained the hypothesis his paper aimed to demonstrate. "Our hypothesis was," he told BioWorld Today, "that this particular genetic change meant that those low birth-weight people didn't produce the correct amount of insulin. And we knew in life that meant they had high blood glucose, and a mild form of diabetes.

"Because insulin is a growth factor in the baby," Hattersley pointed out, "we felt that maybe this hyperglycemia was also occurring before they were born. In which case we'd expect them to be small."

Genes + Environment = Nature + Nurture

Hattersley was "struck by the fact others had shown that low birth-weight babies are more likely to get diabetes and high blood pressure. They were suggesting this has to do with the environment. We were saying that maybe it could be genetic.

"Obviously, the environment has a role," he went on. "Maternal smoking, for example, has an effect of about 400 grams less fetal weight. Yes, all sorts of environmental factors that we know about are important. But I think this Nature Genetics paper reports one of the first clear genetic factors."

He and his co-authors looked in families that were known to be segregating a GCK mutation, and compared infants harboring that mutation with siblings who didn't.

"They'd obviously been born to the same mother and father in the same womb," McCarthy pointed out. The only difference between the mutation-bearing and mutation-free siblings was their weight at birth."

McCarthy described the combinations and permutations of this fetal insulin hypothesis, which he likened to a four-handed poker game with one wild card — the GCK mutation.

Maternal, Paternal, Fetal Genes Play In Mutation Game

"Let's imagine," he began, "that we've got a pregnant mother who has the mutation, but whose fetus does not. So the mother is modestly hyperglycemic during pregnancy because she's got GCK-deficient diabetes.

"Her glucose crosses the placenta, giving the fetus a high sugar level, and since it hasn't got a GCK mutation, its own beta cells are normal. Fetus notices that there's a lot of glucose around, so quite normally secretes excess insulin.

"That system would work quite nicely," McCarthy observed, "if the fetus weren't attached to its mother. But because the fetus has got nowhere for that glucose to go, it can't really normalize its sugar level. It's got its own feedback loop, but because the mother is so much bigger than the fetus, it can never win. It can never get its glucose levels down, because there's always loads more washing across the placenta.

"So it just remains a high-insulin-secreting fetus and that's why it grows up big.

"Now take the situation where both have got the mutation," McCarthy continued. "Mother's high glucose crosses the placenta. But fetus too has the mutation, so it really doesn't sense a high glucose level. It just puts out a normal amount of insulin, and everything cancels out. It ends up a more-or-less normal-size baby."

"A situation where mother doesn't have the mutation and fetus does, means it got it from its father," McCarthy explained. "Both mother and fetus have a normal glucose, but the fetus's beta cells aren't sensing it normally, so it puts out an inadequate amount of insulin.

"Now if that fetus were out there on its own," he pointed out, "the glucose would rise. But the glucose can't rise because fetus is attached to mother and [it] equilibrates across the placenta. Fetus ends up making feeble amounts of insulin and is born a small baby because it's in a low-insulin state."

Glucokinase deficiency "accounts for maybe half of one percent of all adult-onset types of diabetes," McCarthy said.

"How important is this rare diabetes in general?" he asked. "That's one of the things Hattersley intends going on to do: to look at more common defects related to diabetes, and see if they also relate to the same defect in birth weight.

"If that's the case," he concluded, "then finding adults who were low birth-weight babies might be a way of identifying individuals with a particular risk of diabetes in genetic terms and modulating preventive therapy according to the defect they have." *