Image courtesy NASA.

Colorimeter used to measure
blood glucose levels in
experiments which led to
the discovery of insulin.
Today, diabetics use a much
more sophisticated device to
measure blood glucose levels.

Diabetes: the pituitary proposition
In Type 1, or insulin-dependent, diabetes, the body's immune system during childhood destroys cells in the pancreas that make the hormone insulin. Lacking these so-called beta cells, the blood fills with sugar, damaging small blood vessels in the heart, eye and kidney.

Type 1 diabetes is a major cause of blindness. It's also behind more than one-third of the 50,000 annual cases of end-stage kidney disease in the United States. Type II diabetes usually occurs in adults, and is less severe. Together, the American Diabetes Association estimates the direct medical costs in the United States at $44 billion per year.

The traditional treatment for Type 1 diabetes, daily insulin injections, is helpful but not perfect as erratic blood-sugar still causes damage. If Type 1 diabetes is an autoimmune disease, what triggers the immune attack? There's considerable speculation that insulin, the very protein that helps store sugars in the cells, attracts the wrath of immune cells.

To test that idea, Myra Lipes and colleagues at Harvard Medical School and the Joslin Diabetes Center taught cells in the pituitary gland of mice to make insulin. She figured that since there's little immune activity inside the pituitary, any attack triggered by insulin would stand out. That would implicate insulin in triggering the immune assault on beta cells -- the normal source of insulin. (The pituitary's job is to produce hormones that help direct the endocrine system, which controls body development.)

Unexpected benefit
That didn't happen. What did happen, however, could open an intriguing tack in the war on diabetes -- "teaching" non-pancreatic cells to make insulin.

Lipes's research started with non-obese diabetic mice -- a special strain of mice that develop autoimmune diabetes similar to human type I diabetes. Using genetic manipulation, Lipes created a strain of these mice that made enough insulin in the pituitary gland to control their blood sugar.

Surprisingly, there was no immune attack, even in mice that were transfused with 20 million immune cells programmed to attack insulin. In contrast, transplanted beta cells from the pancreas died within five days of this transfusion, Lipes reports.

Happy as micro-clams, the altered pituitary cells pumped out enough insulin to reverse the effects of diabetes.

Further research has found that transplanting the adulterated cells helps reverse diabetes in mice that already have symptoms of the disease. The mice showed healthy gains in body weight, and their average blood sugar levels returned to normal.

A step along the way
It didn't take a genius to figure out that the surprising finding that cells that could make insulin could also evade immune attack might be useful in treating diabetes. Although the research is promising, two major hurdles must be jumped before it can be tested with human diabetics.

First, ongoing experiments to "teach" the altered cells to meter blood sugar levels must succeed. Normal beta cells in the pancreas respond to sugar levels, so that after a holiday pig-out, the blood is flooded with insulin, driving sugar from the blood into cells, where it belongs. The Lipes lab is currently trying to introduce glucose-sensing capability into the modified pituitary cells, to create hybrid cells that can serve as replacement beta cells.

And second, the technique would have to be modified to treat people, not rodents. That's so far down the line that Lipes could not speculate on timing. Eventually, however, we may learn to help our bodies overcome deadly Type 1 diabetes, courtesy of another spare part made for the new millennium.