2009 Banting Award

George Eisenbarth and autoimmune type 1 diabetes

One of the pleasures of the annual meeting of the American Diabetes Association in June is the plenary address of the winner of the highest award given for scientific achievement, the Banting Award (named for one of the discoverers of insulin).  This year’s winner was George Eisenbarth of The Barbara Davis Center for Childhood Diabetes at the University of Colorado Denver and his achievement was nothing less than the demonstration that type I diabetes is an autoimmune disease.

I remember when the involvement of the immune system in diabetes was controversial.  A pathologist named Willy Gepts had studied the pancreases of many people who had died of diabetes, and published in 1965 that most had inflamed islets infiltrated with lymphocytes (white blood cells). It was clear that in diabetes the beta cells died and no longer made insulin; why they died was less clear.  The inflammation shows that the immune system was involved, either as cause or as bystander.

The concept of ‘autoimmune disease’ can be pretty slippery. Common autoimmune diseases include lupus and rheumatoid arthritis.  In every case, you can imagine that there is an unknown agent causing the immune reaction, say a virus.  How can scientists be sure that the disease is truly autoimmune, that is, caused solely by immune system dysfunction?

Eisenbarth worked many years to assemble information from many sciences related to diabetes to answer this question.  It has long been known, for instance, that diabetes runs in families. In fact a person with a diabetic identical twin has about a 50% chance of developing diabetes. Is this because they live in the same environment?  Is it because they have been infected by the same virus?  Or perhaps the genes governing their immune response are the cause because, of course, their genes are identical.

Massive genomic screening has found the genes that correlate with diabetes.  All of the top twenty genes have known functions, and they are all in two categories: insulin secretion (not surprising) and immune response.  This is strong evidence for autoimmune disease, all of which has a strong genetic component.

Before molecular screening was practical Eisenbarth and colleagues identified certain tissue markers – immune system proteins encoded by immune genes – that correlated with disease.  These proteins are involved closely in the presentation of antigens.  Think of the immune system as a vast memory of all the components that should be present in the body, and the immune response as the attack on novel components, those associated with pathogens.  The diabetes-related proteins present a fragment to the cells that ‘remember’ what is self.  They are like the card swipe on the subway turnstile. Did you pay for your ticket or are you faking it?

The diabetes-related proteins come in many models, and some of them are much more likely to trigger a response to the islets than others.  Some are in fact protective – they never respond inappropriately.  If you measure the presence of these various types of proteins in children, you can predict their likelihood of developing diabetes, from essentially zero to over 50%.

One of the most interesting things scientists have found lately is that the T cells that actually kill islets seem to function in swarms.  The evidence is shown in the picture of the pancreas of a diabetic who has died and donated his pancreas for study.  In most diabetics there are no beta cells at all, but in about 10% a small number remain (not enough to prevent diabetes).  Look in the micrograph, and you will see a small number of beta cells (stained brown). But the surprise is that they are not found randomly. In one lobe of the pancreas the islets have a fair number of living beta cells (on the left, brown cells contain insulin), but in the other, not one (right).  We use to think that the death of beta cells was random and relentless.  But this pancreas shows that the T cells seem to attack in a swarm, moving from islet to islet and killing all the beta cells they find.

All of this science serves to design methods to prevent the autoimmune disease and therefore prevent diabetes.  In relatives of diabetes these therapies have been shown to be effective at slowing the disease (but not yet very effective at preventing diabetes).  There is a long way to go but in time Eisenbarth’s work should prove essential at preventing new diabetes.

This entry was posted on Friday, June 12th, 2009 at 9:15 am and is filed under Current Research. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.