Investigating the Effects of TGF-β on T Cell Activation-Induced Cell Death

Achieving immune homeostasis requires the functions and numbers of suppressive regulatory T cells (Tregs) and effector conventional T cells (conv T) to be balanced and precisely controlled. A decrease in Treg numbers is associated with autoimmune diseases such as type I diabetes and multiple sclerosis, while Treg numbers are increased in the tumor microenvironment. Conversely, an increase in conv inflammatory T cells in the tumor microenvironment is associated with better outcomes for cancer patients. Thus, maintaining the appropriate balance between Treg and conventional T cell populations is essential for controlling immune responses against non-self and self-antigens, as well as tumor antigens. Signaling mechanisms that promote the balance between regulatory and conv T cell populations remain largely unknown. One mechanism that could help control the balance between Treg and conv T cell populations is activation induced cell death (AICD). This type of apoptosis occurs when either Tregs or conv T cells undergo repeated stimulation in the presence of the T cell growth factor IL-2. If conv T cells undergo extensive AICD, then Tregs would dominate, and vice versa: if Treg cells undergo extensive AICD, then conv T cells would dominate. Our goal is to determine if conv T cells and Treg cells have different signaling mechanisms that lead to AICD.We previously discovered a novel form of AICD that occurs in vitro with plate bound stimulation of T cells with anti-CD3 and anti-CD28 coated antibodies. Under these conditions, mouse Foxp3+ Tregs expand while Foxp3- conv T cells undergo massive apoptosis, indicating that these cells respond differently to anti-CD3/anti-CD28 stimulation. Unlike classical AICD, this form of apoptosis depends on p53, and thus we termed the process PICA (p53-induced CD28-dependent apoptosis). We wanted to understand why Tregs could expand and survive, while conv T cells died. We found that the expansion and survival of Tregs requires autocrine TGF-β signaling. Conversely, addition of TGF-β could promote conv T cell survival. Our work has identified that Tregs and conventional T cells have differences in the MAPK/ERK signaling pathways, which controls survival after stimulation. Tregs maintain low RasGRP1 expression by autocrine TGF-β signaling, and RasGRP1 expression in conventional T cells is suppressed by TGF-β. Furthermore, we found that RasGRP1 signaling is required for conv T cell apoptosis during PICA. For TGF-β to promote survival of conv CD4 T cells, we found that the transcription factor FoxO3 is critical. Finally, we studied AICD in a more physiological manner in CD19-specific chimeric antigen receptor (CAR) T cells. We found that CAR T cells undergo AICD in vitro when stimulated with large numbers of tumor antigens and the surviving CAR T cells are CD25+Foxp3+. Like PICA, addition of TGF-β can promote CAR T cell survival, but markers of CAR functionality are reduced. Taken together, we have identified distinct signaling pathways that promote AICD in T cells and determined mechanisms by which TGF-β controls survival. We propose that the crosstalk between the RasGRP1 and FoxO3 pathways with TGF-β signaling are critical for T cell survival. Results from this study could inform how conventional and Treg populations survive and undergo apoptosis during AICD in memory formation and the tumor microenvironment, among other disease states