Blocking Interaction of Proteins Prevents Diabetes in Mice

From the JDRF and UC San Francisco:

JDRF-funded researchers have identified a protein on immune T cells that triggers type 1 diabetes in mice when it interacts with another protein found in cells of the pancreas. The scientists report that blocking the interaction prevents the disease without compromising normal immune defenses or causing measurable side effects.

The finding points to a possible strategy for preventing type 1 diabetes in humans, since the mouse T cell protein has a counterpart in the human immune system. The study, led by Jeffrey Bluestone, Ph.D., and Lewis Lanier, Ph.D., of the University of California, San Francisco (UCSF), was published in the June issue of the journal Immunity. The research was funded as part of the JDRF Center for Islet Transplantation at University of California, San Francisco/University of Minnesota

The T cell protein, called NKG2D, is a receptor on the surface of certain T cells called CD8 lymphocytes. Scientists became interested in NKG2D after observing the development of diabetes in nonobese diabetic (NOD) mice, the animal model for type 1 diabetes. In NOD mice, some CD8 T cells invade the pancreas when the animals are only three weeks old. These invading T cells, unlike most others, express the NKG2D receptor. Ten to 20 weeks later, the animals develop the disease.

The other protein of interest, called RAE-1, has been found on cells infected by bacteria or viruses. Normally, the RAE-1 protein binds to NKG2D, alerting CD8 T cells and other immune system molecules to attack and eliminate the pathogen. By this mechanism, the body ensures that foreign, or “non-self” material is flagged and destroyed.

The UCSF researchers were surprised to find that RAE-1 also is present in the pancreas of nonobese diabetic (NOD) mice. This presence of RAE-1 in the pancreatic cells, and the expression of NKG2D by the invading T cells, suggested than an interaction between the two proteins may play a role in type 1 diabetes.

When the scientists used antibodies to prevent RAE-1 from binding its receptor (NKG2D) on the CD8 lymphocytes, the mice were completely protected from the disease. (The effect is like plugging a keyhole so that even the correct key won’t turn the ignition). “You don’t need a calculator to tell the treatment group from the placebo group. It’s 100 percent effective,” Dr. Lanier, said in a press statement.

The researchers expect that a “humanized” antibody to human NKG2D, working in a similar manner, could provide an effective type 1 diabetes treatment in people. The finding is important, because unlike other treatments reported in NOD mice, this therapy was effective even when given relatively late in disease progression.