Cell Biology
Broken Hearts May
Mend After All
Neuroscience
Breathing Restored After Severe Spinal-cord Injury
Immunology
Insulin Prods Development of Type 1 Diabetes
Publishing
Publishing Partnership Issues First Six Consumer Health Books
Overweight Undermines Health
Gene Network Discovered Supporting Cell Migration
Voracious Kudzu Drains Thirst for Alcohol
Proceedings of the HMS
Faculty Council
HMS and Children’s Announce New Endowed Chairs
Three HMS Professors Elected to NAS
New Appointments to Full Professor
New Members Join Academy of Arts and Sciences
Gawande to Speak on Class Day
In Memoriam
Unexpected Tragedy in a Little
Girl’s Expected Death
Insulin Prods Development of Type 1 Diabetes
Autoreactive T Cells Recognize Insulin Epitope, Mount Attack on Islet Cells in Pancreas
In an unexpected convergence of mouse and human studies, researchers have shown that insulin appears to be a crucial trigger for the onset of type 1 diabetes. The findings will likely stimulate development and testing of therapies aimed at blocking or suppressing the autoimmune response to insulin antigens to prevent or treat the disease.
Photo by Steve Gilbert
Insulin is a critical autoantigen in diabetes, according to a new study of human tissue by (from left) Yahua “Jenny” Chen, David Hafler, Sally Kent, Lisa Bregoli, and their collaborators.
“If we’re right, at one level we may have figured out what causes type 1 diabetes,” said David Hafler, HMS professor of neurology at Brigham and Women’s Hospital. “It’s a wonderful story of how mouse immunology, clinical, and translational science need to work together.”
The
Damage Done
The disease strikes about one in every 300 children and adults in the United
States. In type 1 diabetes, the immune system destroys islet cells in the
pancreas, which normally produce the insulin needed to move sugar from the
blood into
the body’s cells. Without insulin, glucose can build up to fatal levels
in the blood while the body’s cells are starved of life-sustaining
fuel.
The new findings “provide convincing evidence that insulin is an essential autoimmune target in the initiation of diabetes,” wrote Matthias von Herrath of the La Jolla Institute for Allergy and Immunology in San Diego in a commentary that accompanies two of the latest studies in the May 12 Nature. The findings “also suggest that insulin could be the target of the autoimmune response in humans, providing a potential candidate for treatment.”
Of the many islet cell proteins attacked by immune cells in type 1 diabetes, insulin has long been suspected of being especially important. Insulin makes up about one third of all the protein in islet cells, and checking for elevated insulin antibodies in the blood is one of several islet cell antibody tests that now predict high risk of the disease. But recently, proteins such as glutamic acid decarboxylase 65 (GAD) have taken center stage in diabetes research.
“It was known that mouse and man can develop certain antibodies to certain proteins before coming down with type 1 diabetes,” said Harald von Boehmer, HMS professor of pathology at the Dana–Farber Cancer Institute. “The question for decades has been, which and how many different antigens can initiate the disease? If there are only a few, and you know what they are, you can do something about it.”
T Cell Target
About 15 years ago, working with non-obese diabetic (NOD) mice, Joslin
Diabetes Center researchers Diane Mathis and Christophe Benoist discovered
that the lymph node draining the pancreas was ground zero for the autoimmune
response. There, miscued T cells multiplied for their attack on the islet
cells.
This finding made Hafler wonder if something similar was happening in people. Now, he and his colleagues report in the May 12 Nature that about one quarter of the CD4 T cells sampled and cloned from the pancreatic lymph nodes of two women with long-term diabetes were reactive to a particular piece of the insulin excreted by islets.
T cells from two healthy people and one person with type 2 diabetes showed no such expansion of insulin-reactive T cells. In another control experiment, T cells from the spleen of one of the two women showed no reactivity to insulin, suggesting that the insulin reaction was limited to the and its lymph node and was not a systemic response to daily insulin injections.
Illustration by Rachel Meyer Insulin trigger. In genetically susceptible people, insulin helps trigger the islet-cell destruction that leads to type 1 diabetes. Some immune cells carry pieces of insulin into lymph nodes draining the pancreas. In response, miscued T cells react by increasing their ranks and going on the attack. Other T cells target other proteins released by the damaged islet cells. |
Such human tissue is difficult to obtain. Collaborators in Miami and Minnesota provided the samples of fresh tissue from people who had just died. One woman, 39, had been diagnosed with type 1 diabetes at age 10, and the other, 24, had lived with the disease since age 9.
“Hopefully, we will make good use of their legacy,” said first author Sally Kent, HMS instructor in neurology at BWH. “We need to look at a lot more samples. Other antigens are probably involved as well. This gives us a good clue about which antigens to focus on and try to change the immune response.”
Minuses and Pluses of Insulin
In a companion paper, Denver researchers prevented diabetes in NOD mice
by knocking out their two insulin genes and replacing them with an insulin
gene missing the autoreactive epitope. Female mice with the altered insulin
did not develop insulin autoantibodies, the T cell invasion of islet
cells known as insulitis, or autoimmune diabetes. Mice with one or
two copies
of the native insulin gene were not so lucky.
“It looks like this epitope is crucial for the initiation of disease,” said senior author George Eisenbarth, executive director of the Barbara Davis Center for Childhood Diabetes at the University of Colorado Health Sciences Center. “It might turn out that there are other crucial molecules.” After islet cell destruction begins, lymphocytes react to pieces of other proteins released by the damaged tissue, a process known as “antigenic spreading.”
A mouse study from von Boehmer’s group in the October 2004 Nature Immunology initially highlighted the importance of insulin. Using a different tactic to squelch T cell reactivity, they overexpressed insulin in the NOD mice to rev up insulin in the thymus, where more of the protein instructs those T cells to calm down.
As a result, the incidence of diabetes dropped dramatically, and the few mice that came down with it did so with delayed and less severe disease. “It makes a lot of sense to use this knowledge and design protocols to convert disease-causing T cells into disease-preventing suppressor cells,” said von Boehmer, who nearly 20 years ago designed another animal model in which he and others proved how all this happens in the thymus.
Finally, tantalizing clinical-trial results in the May Diabetes Care suggest that oral insulin may delay onset of diabetes for 4.6 years in about half of the highest-risk relatives of people with type 1 diabetes. The double-blind, placebo-controlled study was testing the idea from animal studies by Howard Weiner, HMS professor of neurology at BWH, that oral insulin also would teach tolerance.
The rules of statistics frown on drawing conclusions from such post hoc subset analysis in clinical trials, but the researchers received approval from the National Institutes of Health to conduct a more definitive study with that high-risk group to test the hypothesis. Joslin will be one of the study sites. “If we can confirm that benefit, we would advocate using insulin to delay diabetes,” said Jay Skyler of the University of Miami, chair of the Type 1 Diabetes TrialNet. For more information, see www.diabetestrialnet.org.
The story is likely more complex than it appears. Methods of eliminating, redirecting, or suppressing the insulin-reactive T cells need to be refined, or even invented, and tested in mice before dependable results can be expected in people. As Mathis says, “This will stimulate a lot of research.”