mTORC1 Inhibition Protects Human Regulatory T Cells From Granzyme-B-Induced Apoptosis

Abstract

Regulatory T cells (T_regs) have shown great promise as a means of cellular therapy in a multitude of allo- and auto-immune diseases-due in part to their immunosuppressive potency. Nevertheless, the clinical efficacy of human T_regs in patients has been limited by their poor in vivo homeostasis. To avert apoptosis, T_regs require stable antigenic (CD3ζ/T-cell-receptor-mediated), co-stimulatory (CD28-driven), and cytokine (IL-2-dependent) signaling. Notably, this sequence of signals supports an activated T_reg phenotype that includes a high expression of granzymes, particularly granzyme B (GrB). Previously, we have shown that aside from the functional effects of GrB in lysing target cells to modulate allo-immunity, GrB can leak out of the intracellular lysosomal granules of host T_regs, initiating pro-apoptotic pathways. Here, we assessed the role of inhibiting mechanistic target of rapamycin complex 1 (mTORC1), a recently favored drug target in the transplant field, in regulating human T_reg apoptosis via GrB. Using ex vivo models of human T_reg culture and a humanized mouse model of human skin allotransplantation, we found that by inhibiting mTORC1 using rapamycin, intracytoplasmic expression and functionality of GrB diminished in host T_regs; lowering human T_reg apoptosis by in part decreasing the phosphorylation of S6K and c-Jun. These findings support the already clinically validated effects of mTORC1 inhibition in patients, most notably their stabilization of T_reg bioactivity and in vivo homeostasis.

Keywords: granzyme B; grapoptosis; human tregs; mTORC1; rapamycin; treg homeostasis.

Graphical Abstract

Figure 1

mTORC1 inhibition prevents GrB upregulation and rescues T_regs from apoptosis. Human T_regs were isolated and expanded for three days with α-CD3, α-CD28, and IL-2, without additional inclusions (CT), or with CT stimulants plus 100 nM rapamycin (Rapa). (A–D) Flow cytometric analyses of CD25^hiCD127^lo subset among CD4⁺ T cells (A), FoxP3⁺ subset among CD25^hiCD127^lo cells (B), and GrB⁺ (C) and Annexin V⁺ subsets (D) among FoxP3⁺ T_regs, including the respective box plots (n = 3 technical replicates/condition; 2 representative experiments of 5). Data represent boxplots with median, interquartile range, minimum, maximum, and all individual data points of the denoted experimental groups. P values were calculated with independent samples two-tailed Student’s t-tests, and non-parametric Mann-Whitney U-tests were performed when the assumption of homoscedasticity could not be met. CT, control; GrB, granzyme B; Rapa, rapamycin.

Figure 2

mTORC1 inhibition attenuates intracytoplasmically active GrB. Human T_regs were isolated and expanded for three days with α-CD3, α-CD28, and IL-2 (CT), or with CT stimulants plus 100 nM rapamycin (Rapa). (A) The GranToxiLux assay was used to determine functionally active GrB in the cytoplasmic T_reg compartment, where active GrB cleaves the GrB-sensitive GranToxiLux substrate to yield a fluorescent GranToxiLux reagent detectable in the FITC channel. (B) Flow cytometric plots of active GrB in the T_reg cytoplasm. (C, D) Box plots of GranToxiLux⁺ percentages (C) and mean fluorescence intensities (D) among the T_reg population (n = 3 technical replicates/condition; 2 representative experiments of 4). Data represent boxplots with median, interquartile range, minimum, maximum, and all individual data points of the denoted experimental groups. P values were calculated with independent samples two-tailed Student’s t-tests, and non-parametric Mann-Whitney U-tests were performed when the assumption of homoscedasticity could not be met. CT, control; GrB, granzyme B; MFI, mean fluorescence intensity; Rapa, rapamycin; T_regs, regulatory T cells.

Figure 3

mTORC1 inhibition attenuates S6K and c-Jun phosphorylation. Human T_regs were isolated and expanded in a 24-hour cell culture window without inclusions (NC), with α-CD3, α-CD28, and IL-2 (CT), or with CT stimulants and 100 nM rapamycin (Rapa). (A) Flow histograms of p-S6K expression among CD25^hiCD127^lo T_regs, including inset mean fluorescence intensities (MFIs) and the 24-hour period XY plot (n = 3 technical replicates/condition/timepoint; 1 representative experiment of 3). (B) Flow histograms of p-c-Jun expression among CD25^hiCD127^lo T_regs, including inset mean fluorescence intensities (MFIs) and the 24-hour period XY plot (n = 3 technical replicates/condition/timepoint; 1 representative experiment of 3). All individual datapoints are displayed. P values were calculated using mixed-effects model analyses with the Geisser-Greenhouse correction followed by Holm-Šídák multiple comparison tests. AF647, Alexa Fluor 647; CT, control; MFI, mean fluorescence intensity; NC, negative control; Rapa, rapamycin.

Figure 4

Observed interplay between mTORC1 and GrB pathways. (A) In a physiological setting, antigenic T_reg stimulation via the T-cell receptor, co-stimulatory CD28-based activation, and IL-2Rα (CD25)-dependent cytokine signaling, sets in motion a phosphorylation cascade involving phosphoinositide 3-kinase (PI3K), Akt, mTORC1, S6 kinase 1 (S6K), and c-Jun. Phosphorylated c-Jun is known to permit its nuclear translocation where it can undergo dimerization with p-c-Fos and generate transcription factor Activator Protein-1 (AP-1). (B) Our hypothesis is that AP-1 can unwind heterochromatic DNA in the promotor region of GrB in T_regs to boost the transcription of GrB mRNA. Conventionally, upon mRNA translation, the mature GrB proteins are stored in designated cytotoxic granules to save their potent cytolytic activity for target effector T cells. (C) In the context of protracted antigenic stimulation, however, GrB can inadvertently leak from its harboring lysosomes to precipitate caspase-dependent and independent cell death. (D) Treatment of T_regs with mTORC1 inhibitors such as rapamycin provide a multipronged effect, notably, decreasing phosphorylation of p-S6K and p-c-Jun, attenuating GrB expression and activity, and finally diminishing T_reg apoptosis. The exact biochemical constituents that drive this phenotypic and functional shift in the T_reg population upon mTORC1 inhibition remain to be elucidated and warrant further research efforts.