We would like to be able to know who will develop type 1 diabetes long before it happens and have a treatment ready that will stop it from happening. This is a tall order and the path to this goal is not a short one. However, an important step along this path was recently accomplished and the results published in this month’s issue of Immunity.
The question addressed by this study is whether T cells which infiltrate the pancreas recognize pancreatic antigens or whether they don’t. Why does this matter? Well, it boils down to understanding the forces that push a person towards disease. The drug discovery process works something like this: first we figure out the forces that drive a pathology, then we identify the molecular chain of events that makes these forces manifest. Next we choose one of the molecules as a target, and finally we examine what happens when we mess with that target; preferably in an animal model.
When we examine pancreatic tissue from diabetic mice (or diabetic patients for that matter) we see that it is full of T cells and these T cells seem to be activated. As you may recall from previous posts, T cells require antigen presented by MHC to become activated. Now, it is not an easy thing to establish what the T cell specificity is for all of the T cells present. Previously, through many ingenious but ultimately indirect experiments, immunologists came to the conclusion that the T cells present in these pancreatic infiltrates included both cells that react to pancreatic antigens as well as something we call bystander T cells, i.e. T cells that do not react to pancreatic antigens but simply dropped by to see what was happening. If this bystander business is correct it would argue that the pancreas is quite leaky and that the primal force driving disease is a non-specific inflammation (one particular molecular chain of events). Conversely, if this bystander business is wrong it would argue that only certain T cells can gain entry into the pancreas and thus the primal force driving disease is an antigen specific T cell attack (a different molecular chain of events).
Lennon and co-workers solved the problem in a beautifully elegant way. They created a simplified immune system. Instead of 1,000,000,000 different types of TCR there were 2 different types of TCR. The way in which this was done was complex and itself was a major paper published in 2006 in Nature Protocols. Simply, one T cell population was diabetogenic (in other words, all of the T cells in that population had one identical T cell receptor that happened to be specific for a pancreatic antigen and caused diabetes). The other population was bystander (in other words the T cell receptor had nothing to do with the pancreas). Consider the work that went into the making of this model system. Countless immunologists cloned rearranged T cell receptors and figured out what antigens bound to those receptors. The work took decades. Others developed mice that conveniently lacked an immune system and could serve as the empty plate upon which a new immune system could be arranged. Still others have generated the tools (primarily from viruses) that allow us to slip a new gene into a tissue as easily as popping a new engine into a Mercedes. All of this and more sits on the shelf for us to purchase and put together like some sort of huge biological Lego® set.
When the two populations were expressed together, the diabetogenic T cells went to the pancreas and began the job of destruction. Meanwhile, the bystander T cells did….nothing. They tried increasing the relative population of bystander T cells as compared to diabetogenic T cells. Again…nothing. The bystanders were simply not interested. However, when the authors created a mouse in which the pancreas now expressed the gene corresponding to the bystander T cell antigen…the bystander T cells took interest. They infiltrated the pancreas and got into the game. They were no longer bystanders.
I mentioned in a previous post that T cells hang out in military bases called lymph nodes. Each organ has one or more dedicated lymph nodes so as might be expected, T cells that attack the pancreas come from the pancreatic lymph node. In a technical tour-de-force, the authors of this study showed that T cells taken from a pancreatic lymph node, when placed in a mouse missing its own immune system, attacked the pancreas. When the T cells were taken from some other lymphoid organ like the spleen, they acted like bystanders and did nothing. The authors then went on to actually clone several of the T cell receptors and showed that those receptors from T cells derived from the pancreatic lymph node were actually focused on a pancreatic antigen while those taken from the spleen were not.
Putting this all together, the authors showed that T cells from the pancreatic lymph node are activated by some antigen that is brought to the lymph node by an antigen presenting cell. Only these antigen specific T cells can gain entry into the pancreas and begin the dance of destruction. However, over time, as destruction gets underway, more antigens get revealed and brought to the lymph node. Different T cells get activated and move into the pancreas. Eventually, the pancreas does get leaky but this is quite late in the disease. In the beginning though, the driving force is not some general inflammatory process, it is antigen driven. The hunt is on for that first antigen. If we find it, we have a target.
Many of you may be disappointed by this. You already have diabetes. Your pancreas is just a memory now. “What about me?” rings in my ear. Don’t worry. A huge amount of work is going into the construction of another Lego® set – one that builds a pancreas. But, gentle readers, that is another story.