Gene manipulation applications in autoimmune diseases 2

What we might have to do, then, is to do gene manipulation. To somehow circumvent this shortcoming of if you have total polyclonal Tregs, so few of them see the target of interest that it doesn't work. Can you use gene engineering, gene manipulation of Tregs, to really effectively block immune attack in humans? That is the goal.

As I mentioned, the antigen-specific Tregs are very rare. So rare then, if you inject polyclonal Tregs, who knows what frequency is there. But we know from measurements in blood for other specificities that antigen-specific T cells are one in a few million in blood. This is why, if you take blood, expand total Tregs, very few of them are going to be the ones you care about. How do you circumvent this problem? We go back to how T cells see the world. There is actually a technology taken from the cancer world, that's why it says tumor cell. T cells see other cells via the T cell receptor. There, the T cell receptor sees a peptide antigen presented by the MHC. It's MHC class 1 in this example, there's this B2M accessory protein, which we're going to mention a few slides later in this lecture, but so you see there what we already talked about. T cells, T cell receptor recognize peptide-MHC on the surface of a target cell, that's Signal 1. They need some encouragement as CD28 binding to CD80/86, which might be in the same cell or a different cell. Usually, tumor cells do not have these costimulatory molecules. It's more the job of what we call professional APCs, such as macrophages, dendritic cells, and B cells, sometimes, also endothelial cells. That's how T cells normally work. Now, the field has thought, this is complex, hard to engineer. Can we expand the horizons of what T cells can see and that was called a chimeric antigen receptor, CAR. What the CAR is, is this man-made synthetic molecule that the antigen-binding domain is a single-chain fragment variable, scFv, and it's linked to an intracellular domain, which combines everything a T cell could hope for. It has a costimulatory domain, CD28 or 4-1BB are the two common ones, CD28 being the one that we just talked about for the natural T cells, and it also has Signal 1 domain CD3ζ. I didn't specify, but the T cell receptor itself does not signal. The T cell receptor complexes with CD3, and so CD3 molecules, there's CD3ζ, γ, ε, δ. CD3ζ has the most ITAMs, which are activating motifs that cause single transduction. By really having a minimal design, if you will, the CAR is taking the best costimulatory domain and the best TCR-CD3 domain. This way, every time a T cell or Treg sees a target on the surface of the target cell, it signals Signal 1 and Signal 2, so it's happy to spring into action. That's one important point, it's that the CAR is a one-stop shop. These CAR T cells and CAR Tregs do not need any help from antigen-presenting cells for them to carry their function. Point Number 2 is that, unlike the T cell receptor, which is limited to peptides of defined length presented by MHC complex, CARs because the antibody base, the recognition, they can bind anything on the surface of a cell. So all of a sudden, you're making T cells that can see any cell surface molecule that you can make an antibody against. That can be really empowering, going back to one of our previous statements in this lecture that we might be able to treat autoimmune disease without necessarily knowing what caused it. Because now, with this technology, we can target these curative regulatory cells to tissues using completely different mechanisms from how the immune system normally brings T cells to tissues.