The proliferation and contraction of activated effector T cells must be carefully coordinated to maintain immune homeostasis. This is achieved in part through restimulation-induced cell death (RICD), a pre-programmed apoptosis program triggered by antigen restimulation through the T cell receptor (TCR). Forkhead box P3 (FOXP3)+ regulatory T cells (Tregs) also constrain conventional T cell (Tcon) responses, but can resist RICD themselves despite frequent TCR stimulation. We previously showed that FOXP3 protects Tregs from RICD by suppressing SLAM-associated protein (SAP), a key adaptor protein that amplifies TCR signal strength. Mysteriously, FOXP3 expression is also transiently induced in human Tcons after activation, with a kinetic expression profile that correlates inversely with acquired RICD sensitivity. Hence we asked whether FOXP3 protects expanding human effector T cells from premature RICD by modulating SAP expression. Our results show that although siRNA-mediated FOXP3 knockdown sensitizes early effector CD4 and CD8 T cells to RICD, SAP expression remains unaffected. Unlike late stage effector T cells, a low level of RICD in expanding Tcons was entirely dependent on de novo transcription, and knockdown of SAP failed to reduce death. Subsequent RNA-Seq analyses revealed that CD48, a SLAM family receptor, was markedly reduced upon FOXP3 knockdown. Blockade or knockdown of CD48 also increased RICD in CD4 and CD8 T cells. We now show that CD48 protects early effector T cells both by promoting autophagy and upregulation of pro-survival genes such as BATF. These findings implicate FOXP3 as the central governor of a distinct transcriptional program that promotes RICD resistance early in the effector T cell response.

The adaptive immune response relies on specific apoptosis programs to regulate effector T cell expansion and maintain homeostasis. Restimulation-induced cell death (RICD) is an apoptotic pathway triggered by repeated stimulation through the T cell receptor (TCR), which ensures that effector T cell proliferation remains in check. Constitutive fork head box P3 (FOXP3) expression renders regulatory T cells (Tregs) resistant to RICD. Although FOXP3 is also expressed transiently in human conventional T cells (Tcons) during initial rounds of activation-induced proliferation, its function in this context remains unclear. Here we describe a novel role for FOXP3 in protecting both CD4+ and CD8+ human Tcons from premature RICD during the expansion phase. SiRNA-mediated silencing of FOXP3 sensitized early effector Tcons to RICD, and was dependent on de novo transcription. RNA-Seq analysis revealed FOXP3-dependent effects on genes involved in metabolic pathways, autophagy, and cell surface receptors. FOXP3 silencing increased glycolysis and reduced autophagy in Tcons, which has been linked to changes in apoptosis sensitivity. Indeed, blocking autophagy increased RICD in control Tcons without augmenting death of FOXP3-silenced cells, suggesting FOXP3 reduces RICD sensitivity through an autophagy-dependent mechanism. Additionally, FOXP3 knockdown markedly reduced expression of CD48, a SLAM-family receptor. We now show that CD48 protects expanding Tcons from RICD. Surprisingly, CD48 also promoted protective autophagy, illuminating a novel connection between SLAM receptor signaling, autophagy regulation, and RICD sensitivity that is governed by FOXP3 expression in early Tcons.

An efficient adaptive immune response relies on the rapid clonal expansion of effector T cells, followed by timely disposal of those cells in order to avoid damage to host tissues. Effector T cell proliferation is constrained by the action of forkhead box P3 (FOXP3)+ regulatory T cells (Tregs) and restimulation-induced cell death (RICD), an apoptotic pathway triggered by repeated stimulation through the T cell receptor (TCR) in the presence of interleukin-2 (IL-2). Constitutive FOXP3 expression renders Tregs resistant to RICD, in part through repression of the signaling adaptor molecule SLAM-associated protein (SAP) and FAS ligand (FASL). Interestingly, FOXP3 is also expressed transiently in human conventional T cells (Tcons) during initial rounds of activation-induced proliferation, but its function in this context remains unclear. Here we uncover a new role for FOXP3 induction in protecting both CD4+ and CD8+ human Tcons from premature RICD during the expansion phase. SiRNA-mediated silencing of FOXP3 sensitized early effector Tcons to RICD. This increase in RICD was completely dependent on de novo transcription, but FOXP3 knockdown did not alter SAP or FASL expression in these cells. Instead, FOXP3 silencing increased glycolysis and reduced autophagy in Tcons, which we and others have linked to changes in apoptosis sensitivity. Surprisingly, FOXP3-dependent autophagy specifically protected CD4+ but not CD8+ Tcons from RICD. Our results suggest a fundamentally different mechanism connecting transient FOXP3 expression to metabolic activity and RICD resistance in CD4+ versus CD8+ T cells.

In healthy individuals, the balance between immune-mediated pathogen clearance and tissue damage is tightly regulated through multiple processes. One such mechanism is restimulation-induced cell death (RICD), a propriocidal apoptotic mechanism that constrains T cell responses. Activated T cells must surpass a high threshold of TCR signal strength to induce pro-apoptotic genes responsible for executing RICD. Because co-receptors such as TIM-3 are thought to modulate TCR signal strength, we hypothesized that TIM-3 can shape the T cell response by tuning RICD sensitivity. We utilized siRNA knockdowns and transfection of wild-type and mutant TIM-3 expression constructs to assay for changes in RICD sensitivity in both immortalized and primary human T cells in vitro. Strikingly, we observed that the effect of altering TIM-3 expression varied temporally during the effector T cell response. Whereas TIM-3 promoted RICD resistance in newly activated T cells, TIM-3 enhanced RICD sensitivity in late-stage effector T cells by boosting TCR-induced FASL and BIM expression in a ligand-independent fashion. Preliminary evidence suggests that this dichotomy may relate to substantial temporal changes in the expression, cellular localization, and glycosylation of TIM-3 in early versus late-stage effector T cells. Overall, these results indicate that TIM-3 influences apoptosis sensitivity differently at distinct phases of the human T cell response, with important implications for the use of checkpoint blockade immunotherapies targeting TIM-3. Our findings may also help to clarify discrepancies in the literature regarding TIM-3 signaling and function in T cells.