If you put glucose on pancreatic cells in a lab dish, they secrete insulin. But in the body, insulin production actually begins as food reaches the stomach, well before glucose reaches the pancreas. So there must be a signal or series of signals that spur the pancreas into action before food is digested.

Recent experiments in mice have raised other questions. If you remove a key enzyme from the glucose-sensing pathway in the pancreas, the animal is not diabetic. Feed the mouse and it produces 100 percent of the insulin necessary to metabolize incoming sugars. Inject the mouse with glucose, however, and it cannot metabolize the sugar.

David Piston, Ph.D., professor of molecular physiology and biophysics and professor of physics at Vanderbilt University, is investigating these issues on several fronts.

In one line of experiments, he used two types of high-resolution fluorescent imaging along with chemical analyses to uncover a previously unrecognized feedback loop by which insulin can have a direct impact on glucokinase, a key enzyme that regulates blood sugar levels.

By tagging glucokinase with light-emitting proteins, it was possible to follow some of the steps leading to activation of the enzyme inside live beta cells, the cells of the pancreas that make insulin.

Piston demonstrated that high blood glucose causes glucokinase to be released from its bound state in living cells and to begin stimulating insulin release from the pancreas.

He took away the glucose and found that insulin alone was capable of triggering the release and activation of glucokinase and the start of glucose metabolism. But if he restored the glucose and then took away the insulin, the result was little or no glucokinase stimulation. These results suggest that insulin provides a signal necessary to initiate glucose metabolism.

“Taken together, it is reasonable to conclude that the effects of glucose on GK [glucokinase] are best explained by a ...feedback loop,” Piston concluded in The Journal of Biological Chemistry, Sept. 13, 2002.

“Our results indicate the existence of a novel mechanism for linking insulin signaling and glucose metabolism in beta cells,” Piston wrote. “Knowledge of this functional association between insulin signaling and GK is important in light of the growing realization that insulin signaling in the beta cell, and the defects brought about by insulin resistance, may play a causative role in the pathogenesis of type II diabetes mellitus.”

This by no means offers a full explanation of the signaling involved in the body’s response to glucose, but it adds another level of detail to the picture. It is also possible that insulin from neighboring beta cells could affect the response to glucose. Moreover, glucose may have additional effects on glucokinase regulation that have nothing to do with insulin secretion.

“We’d like to have the ability to say, ‘Here are all the things between glucose and the end products of glucose metabolism,’” Piston says. “We’ve looked at a total of about six or seven things, maybe a few more. We have about 40 to look at. So we’re maybe a sixth of the way there after 11 years.”

Early in his research, Piston built most of his imaging devices. He is still creating new ways of imaging biological systems, but these days he often relies on modern commercial instruments, which have improved dramatically in recent years. Moreover, “I don’t have to write the software anymore.”

Piston recently launched a new line of studies using microfluidic devices in which he stimulates different parts of the islet and looks for evidence of intracellular communication. He has also begun using a customized confocal microscope to examine beta cell activity inside the pancreas of live mice.

Piston received a 1994 Whitaker Foundation Biomedical Engineering Research Grant for investigating cell lineages using fluorescent microscopy.