By David N. Leff

Before a trio of Canadian physicians in 1921 saved the lives of dogs with acute diabetes by injecting them with insulin, the disease spelled certain death for people who contracted it.

Those experimental canines were rendered diabetic by removal of their pancreases, the organ that churns out insulin to meet the body's demands for energy.

Today, despite ever-improving forms of the hormone since 1921, (including recombinant insulin since 1983), diabetes remains the sixth leading cause of death by disease in the U.S.

But insulin does a lot more than merely control diabetes, by regulating glucose in the body. It's the master molecule of mammalian metabolism, orchestrating the activities of enzymes critical to the control of metabolism by dephosphorylation.

Cell biologist Alan Saltiel said, "the critical role of protein phosphorylation in the regulation of glucose and lipid metabolism is played by insulin. It stimulates the dephosphorylation of anabolic [protein-building] enzymes, such as glycogen synthase. Protein phosphatase 1 [PP 1] catalyzes this process.

"And insulin," he said, "is the one hormone that regulates glucose metabolism in this way. It can't be substituted for by any other hormone."

Saltiel, who heads cell biology at the Parke-Davis Division of Pharmaceutical Research, in Ann Arbor, Mich., is senior author of a paper in today's Science, (dated March 7, 1997). Its title: "PTG, a protein phosphate 1-binding protein with a role in glycogen metabolism."

"That's pretty straightforward, so far," Saltiel told BioWorld Today. "We knew that insulin activates PP 1, which in turn dephosphorylates glycogen synthase, and causes its activation. But there's a paradox in that process."

He explained: "PP 1 is present in almost all mammalian cells, not just cells that respond to insulin. What's more, PP 1 is found in almost all compartments of the cell, in the nucleus, in the membranes, in the cytoplasm, everywhere. However," Saltiel continued, "insulin dephosphorylates only a very small subset of phosphate-bearing proteins. So there must be a mechanism whereby PP 1 can be activated only by insulin, and only at certain places in certain tissues.

A Molecular Scaffold In Cell's Compartments

"That's the paradox we set out to solve."

In so doing, he and his team at Parke-Davis, and the University of Michigan, came up with a novel concept — a molecular scaffold in the cell.

They began by looking for proteins that might serve to target PP 1 to its substrate, glycogen synthase. "We used a two-hybrid screen to detect proteins in yeast that interact with each other." Saltiel said. "And we identified a new protein, which we call PTG — for protein targeting to glycogen. (Glycogen is a storage form of glucose.)

"To clone PTG," he recounted, "we constructed our own two-hybrid cDNA library from the most insulin-sensitive cell line, that of a mouse fat cell. This is probably the secret of our success.

"PTG turns out to be what we call a molecular scaffold," he continued. "It binds to PP 1, the phosphatase that causes dephosphorylation of these proteins. And it also binds to its own two substrates, glycogen synthase and glycogen, as well as to a protein kinase, which can inactivate glycogen synthase. And PTG can also bind phosphorylase, which cleaves glycogen to form glucose.

"So PTG," he summed up, "is a kind of scaffolding protein that binds four different proteins. It assembles them all into a complex that we call the signal reception module," which allows glycogen to receive a localized hormonal signal from the insulin receptor."

He suggested that "the next frontier in signal transduction will try to determine the role of cellular compartmentalization in signaling to certain domains of the cell, for ensuring specificity and fidelity of the signaling pathway. So I think PTG will be one of the first examples of scaffolding protein involved in this issue."

PTG: Candidate Diabetes Gene

Meanwhile, he and his colleagues have recently cloned the gene for human PTG, and are studying it as a candidate gene for insulin resistance, which marks type II diabetes.

This is non-insulin-dependent diabetes mellitus (NIDDM) of which about 595,000 new cases are diagnosed each year in the U.S.

"One of the earliest defects that's been seen in studies of patients with NIDDM," Saltiel pointed out, "is the inability of their insulin to regulate the non-oxidative metabolism of glucose. This really boils down to the stimulation of glycogen synthesis. So PTG, I think," he continued, "is a candidate gene in diabetes, and we're now looking at diabetic families, and sequencing DNA from patients."

Their hope is to find some patients with a defect in the coding or promoter region of their PTG gene. "That's potentially significant," he said, "and time will tell if it really plays an important role in diabetes."

Beyond that disease, he concluded, "If this theory about scaffolding is correct, there will be other scaffolding proteins — we've already isolated some — involved in the regulation of phosphorylation and dephosphorylation. These may be involved in the regulation, synthesis and degradation of lipids, in response to hormones." *