UCLA and Florida researchers have produced strong new evidence implicating a common class of viruses in causing Type 1, or insulin-dependent, diabetes--a discovery that could eventually lead to new ways of preventing the disease that affects as many as 1 million Americans.
Type 1 diabetes occurs when the body's immune system attacks and destroys insulin-producing cells in the pancreas. Researchers know that only people with a specific genetic profile are susceptible to this autoimmune attack, but there has been great debate about the identity of the agent that triggers the immune reaction.
Now, two studies in this month's issues of the Journal of Experimental Medicine and the Journal of Clinical Investigation strongly implicate Coxsackie viruses, polio-related viruses that cause upper respiratory infections.
One report shows that Coxsackie viruses trigger diabetes in genetically susceptible mice but not in those that have a different genetic profile. The second study demonstrates that the autoimmune reaction is triggered by a similarity between a Coxsackie virus protein and a protein in the pancreas.
The two studies provide "a strong case of guilt by association," said UCLA molecular biologist Daniel L. Kaufman.
"It's only inferential data so far, but it's pretty strong inferential data," added Dr. Noel Maclaren of the University of Florida. "I think the case is really seriously beginning to build" that Coxsackie viruses are the key triggering agent.
If the viruses are shown to be a primary cause of the disease, vaccines against them could probably prevent diabetes, the researchers said. But "before we can start working on such a vaccine, we need to have a fairly clear smoking gun" strongly confirming that the viruses cause diabetes, said Dr. Joan Harmon, director of the Diabetes Research Program at the National Institutes of Health.
Researchers have a 30-year history of interest in Coxsackie viruses as a diabetes cause. Epidemiological studies in the 1960s, for example, showed that Coxsackie outbreaks in various regions were followed by an increased incidence of diabetes. Studies also showed that perhaps 60% of newly diagnosed diabetics had antibodies to the virus.
More recently, researchers such as Maclaren and his Florida colleague, Mark Atkinson, have demonstrated that people in the very earliest stages of diabetes have antibodies against a specific protein, subsequently shown by Kaufman and others to be an enzyme called glutamic acid decarboxylase, or GAD.
Three years ago, Kaufman and Allen J. Tobin of UCLA demonstrated that a small segment of GAD is structurally similar to a segment of a Coxsackie protein. Their interpretation was that when the immune system attacks this viral protein in fighting off a Coxsackie infection, it also inadvertently attacks GAD, which is found on the surface of insulin-secreting pancreas cells. Ultimately the cells are destroyed and diabetes results.
The new papers buttress this hypothesis.
Maclaren and Atkinson wanted to show that the structural similarity between the two proteins was not simply coincidence. To achieve that, they broke the GAD molecule down into 36 shorter fragments, one of them the segment similar to the Coxsackie protein.
When they exposed blood from several patients in the early stages of diabetes to each of these fragments, they observed that the anti-GAD antibodies strongly attacked the segment similar to the viral protein and reacted only weakly with the other fragments. This reinforced the idea that the attack on GAD was triggered by a Coxsackie infection.
To further explore this relationship, Kaufman and his colleagues exposed two types of mice--one group that was genetically susceptible to Type 1 diabetes and one that was not--to a Coxsackie virus. They found that, shortly after the infection, the susceptible group developed antibodies not only against the Coxsackie protein but also against GAD. The mice then developed diabetes. The non-susceptible mice, in contrast, developed antibodies only against the viral protein and did not develop diabetes.
Because the genes that control diabetes susceptibility in mice are similar to those in humans, Kaufman said, the observation suggests that Coxsackie viruses can trigger the disease in humans as well.
The researchers agreed, however, that further studies will be necessary to actually prove that Coxsackie viruses cause the disease.
Maclaren cautioned that Coxsackie viruses may not be the cause of all Type 1 diabetes. About one in every five diabetics, he noted, does not produce an immune response to GAD, and the disease in those cases may have been triggered by some other agent. Possible triggers that have received a lot of attention include the nitrosamines used as a food preservative and a protein in cow's milk.
"We're really homing in on the cause of this disease," said Dr. Richard Kahn, vice president for research at the American Cancer Society. "This could very well be the major, perhaps the only, environmental trigger for diabetes. "