Today I want to talk about something full of strangeness and beauty which also is fundamental to the onset of type 1 diabetes – namely a process by which baby T cells mature into adult T cells. The process is called thymic selection. In the last couple of posts I talked about how T cells have receptors that recognize antigen. They arise from a fundamentally random process called recombination and are as likely to recognize insulin as to recognize a viral coat protein. How is it that the insulin reactive T cell is removed while the viral reactive T cell is retained?
The thymus, where all this happens, sits just above the heart. Precursor cells differentiate from bone marrow stem cells, enter the blood stream, and find their way to the thymus. When they enter the thymus they begin to interact with the resident cells. Because of the way they are programmed, a new gene expression program begins and they become a new cell type called a thymocyte; a baby T cell.
Thymocytes undergo the gene rearrangements we talked about last time to produce a T cell receptor. Now the fun begins. One by one, resident cells in the thymus present antigens to the thymocytes. Thousands. Millions. Eventually one of three things will happen. Either the T cell receptor will react vigorously in response to an antigen or it will react just a little bit, or it won’t react at all. This is beginning to sound like Goldilocks and the 3 bears.
Before we get into the fate of the thymocytes depending on how it responds to antigen, let’s think for a moment about where that antigen came from. The immune system forms very early in the life of an organism. The neonate has yet to be exposed to pathogens so there are no bacterial proteins nor are there viruses to serve as antigens. All of the millions of antigens presented to thymocytes are actually bits of self; mirrors if you like – bits of our own genes, translated into proteins, chopped up, and presented buffet style to the waiting line of hungry thymocytes. If, after all of this varied buffet, the T cell has not responded; it dies. It actually commits cellular suicide, releasing a host of enzymes within itself to kill itself from the inside. Now, importantly, if the T cell responds exuberantly in response to a bit of self, it is likely to cause autoimmune disease. Again, those suicide enzymes are activated and the cell dies. This is called negative selection. All that we have left then is the Goldilocksian middle choice; T cells that react to self antigen but do so halfheartedly. Think about this for a minute. The system which is exquisitely designed to protect us from the dangers outside never actually sees the outside during its formative period. All surviving thymocytes must be (and indeed are) self reactive. Each and every T cell that survives this deadly course of thymic education has the capacity to turn upon us and devour us. A sobering thought.
In the case of insulin, this happens in roughly 1 out of 5,000 people. Why? First of all, we don’t have all the answers. But, of course, we have some pretty good theories. One idea is that insulin is highly visible to the immune system. It is a major hormone that goes everywhere. The likelihood that many damaged insulin molecules (with slightly altered shapes) will come into contact with T cells is high. Under the right conditions this can lead to an immune response against insulin. A second important idea is referred to as molecular mimicry. We know for sure that this happens. Certain viruses such as rubella have an epitope that looks very much like a common insulin epitope. Once a person is exposed to this virus the chances are much higher that insulin can now cross react with the T cells generated via that immune response. Another important idea is that of a breakdown in tolerance mechanisms. It turns out that while some T cells function as soldiers, others (for want of a better image) function as UN peace keepers. Their job is to block inappropriate conflicts. Genetic factors that weaken the function of this type of T cell increases the possibility of developing type 1 diabetes. Finally, we know that there is a significant genetic component contributing to the risk of developing type 1 diabetes. Interestingly, while there are many genes that contribute to this risk just one gene contributes 50% of the risk. It happens to be the protein that is responsible for presenting antigens to T cells.
Putting all of this together, we see that the immune system has an extraordinary method of establishing the boundary between self and non-self. T cells must be self reactive to an extent because this is the only way to test for their function. This is a dangerous balancing act and sometimes the person falls off of that wire. For some time now we have been focused on building the net that will catch them safely. Soon, however, we will understand enough to keep them from falling in the first place. That will be the greatest show on earth.