Abstract Chimeric antigen receptors (CAR) expressed by T cells recognize tumor cells via single chains antibodies and activate T cell cytotoxic machinery and costimulation. In clinical studies, costimulation mediated by CD28 and 4-1BB endodomains integrated into the CD19-specific CAR has been shown to be equally effective in causing tumor regression. However, CD28 and 4-1BB costimulation differentially modulates the kinetic, metabolism and persistence of CAR-T cells, and the mechanisms governing these differences are not fully understood. In this study, we have identified that LCK recruited by co-receptors into the synapse of the CAR encoding CD28 leads to antigen-independent CAR-CD3ζ endodomain phosphorylation and imprints T cell activation upon antigen engagement. In contrast, the synapse formed by the CAR encoding 4-1BB recruits the THEMIS-SHP1 phosphatase complex that attenuates CAR-CD3ζ endodomain phosphorylation and T cell activation. We have also proved that the CAR synapse can be engineered to tune down the activity of CD28 costimulation or to tune up the activity of the 4-1BB costimulation. This discovery has been recently published in Cancer Cell. Remarkably, we observed that LCK mediated constitutive phosphorylation of CAR-CD3ζ in 4-1BB- costimulated CAR-Ts does not lead to premature exhaustion of CAR-Ts in xenotransplant models. Therefore, we hypothesize that the LCK-mediated imprinting in 4-1BB costimulated CAR-Ts leads to unique and critical signaling pathways in CAR-Ts. Furthermore, in addition to proximal signaling, CARs profoundly affect downstream T cell signaling. We found that NF-κB activity is influenced by the type of CAR costimulation. Precisely, 4-1BB induces more pronounced NF-κB activity than CD28 in CAR-Ts. NF-κB hyperactivity in 4-1BB is not caused by NF-κB overexpression, but rather by reduced A20 activity. Therefore, we hypothesize that 4-1BB sequesters A20 reducing its inhibitory effects on NF-κB. Furthermore, since NF-κB/A20 interplay is critical in controlling T cell function at multiple levels, we hypothesize that regulating NF-κB/A20 may enhance efficacy, persistence and safety of CAR-Ts. We will develop two specific aims: Aim 1: To mechanistically assess how LCK-mediated imprinting of 4-1BB costimulated CAR-Ts promotes rapid antitumor activity without causing T cell exhaustion. We will assess whether LCK overexpression in the 4-1BB CAR activates unique phosphorylation, transcriptome and metabolic pathways. Aim 2: To mechanistically assess how 4-1BB affect and how the NF-κB/A20 interplay can be manipulated to modulate CAR-T cell functions. Since A20/NF-κB interplay is differentially regulated in 4-1BB vs. CD28 costimulated CAR-Ts, we propose to understand how this interplay functions and to develop pharmacologic and genetic interventions to transiently or permanently modulate NF-κB activity to enhance safety, persistence and efficacy of CAR-Ts.

Project Narrative Adoptive T cell therapy to treat human malignancies has been revolutionized by the use of chimeric antigen receptor (CAR) T cells in the past decade. However, despite the recent FDA approval of CAR T cells for the treatment of leukemia and lymphoma, how these CAR molecules really function remains largely unknown. Therefore, understanding how these novel molecules shape the function of T cells remains critical to logically advancing the field by designing molecules that reflect their true mode of action. In turn, these discoveries will be instrumental to generate more effective and safer therapies for patients with cancer.