Each T cell and each B cell has a different antigen receptor. Collectively T cells and B cells are referred to as lymphocytes. How many lymphocytes do we have? The answer is between 100 million and 1 billion. Now, let’s do a bit of math. Antigen receptors are constructed from two proteins (we call them subunits when they combine to make a single structure). Each of the proteins is coded in DNA as a gene. The problem here is that we may have 1 billion different antigen receptors but we sure don’t have 1 billion genes. In point of fact we have somewhere around 20,000 genes. This was made certain when both Francis Collins and Craig Venter succeeded in sequencing the entire human genome. Even a Bear Sterns accountant must accept that we have some explaining to do.
How can one gene become 1 billion different proteins? The answer is through a process called recombination. In thinking about an analogy that lots of people can relate to I hit upon the lottery. Huge numbers of people play the lottery and, of course, very few win. Since I live in Colorado, I’ll use one of the local lotteries as an example. For the Colorado Cash 5 game, you pick five numbers from a field of 32. How many different combinations of these 5 numbers are possible? If you pick a different number each time the number of choices is 32 x 31 x 30 x 29 x 28. This comes to 24,165,120. It’s a pretty big number for 5 simple choices. Biologists refer to this as combinatorial complexity. For both B cells and T cells the receptor comes in 3 parts (called V, D, and J regions) and for some of those parts there are several hundred “choices” which can be made. Additionally, there are two subunits that come together to make the full receptor so now we have 6 times where we can make one of these choices. One can see how quickly the math can generate a billion different combinations. As a matter of fact the number of possible combinations is much greater than a billion. It’s just that we can support somewhere between 100 million and 1 billion cells so that is what limits the immune repertoire.
How good is this process? Proteins seem like pretty complex structures so can we simply “mix and match” and get every thing to work right out of the box? Think about the last computer you bought. Did all of the peripherals work seamlessly right out of the box? Unless you got a Mac…probably not. The same is true for recombination. At least 2/3 of these recombination events are totally non-functional.
What will these antigen receptors recognize? The answer is pretty much anything. The possibility of a T cell receptor recognizing human insulin is as probable an event as another T cell receptor recognizing a coat protein of HIV (the AIDS virus). The most common kind of antigen is a protein fragment of between 7 and 13 amino acids. Continuing with my math fetish, we have around 20 different amino acids. The number of possible 7 amino acid fragments would be 20^7 (20 x20 x 20….7 times) or about 1.3 billion. This is the possible diversity of antigens of that size. The largest antigen has a diversity of 20^13. This has a diversity of 80,000,000,000,000,000 (a very big number). If we refer to this as antigen space you can see that despite the very large number of unique antigen receptors in our immune system, antigen space is vastly larger. So, our immune system samples antigen space.
Each of us is a unique being and each of us undergoes antigen receptor rearrangement differently (it is a random process after all). Since the earth currently supports 6 billion individuals the human species as a whole has about a billion billion different antigen receptors. With that many lottery tickets out there someone must be a winner.
Why does any of this matter? Well, for one thing, it demonstrates that our ability to recognize antigens really is founded on a random set of events. Our immune systems generate receptors for proteins that may or may not exist. In doing so they generate receptors for proteins that are found on the surface of pathogens as well as receptors for our own proteins. There must be a way to distinguish “self” from “non-self”. That will be the topic of tomorrow’s post and will bring us at least a little closer to understanding the pathology underlying type 1 diabetes.