Induced regulatory T cells suppress Tc1 cells through TGF-β signaling to ameliorate STZ-induced type 1 diabetes mellitus

Abstract

Type 1 diabetes mellitus (T1D) is a chronic autoimmune condition in which the immune system destroys insulin-producing pancreatic β cells. In addition to well-established pathogenic effector T cells, regulatory T cells (Tregs) have also been shown to be defective in T1D. Thus, an increasing number of therapeutic approaches are being developed to target Tregs. However, the role and mechanisms of TGF-β-induced Tregs (iTregs) in T1D remain poorly understood. Here, using a streptozotocin (STZ)-induced preclinical T1D mouse model, we found that iTregs could ameliorate the development of T1D and preserve β cell function. The preventive effect was associated with the inhibition of type 1 cytotoxic T (Tc1) cell function and rebalancing the Treg/Tc1 cell ratio in recipients. Furthermore, we showed that the underlying mechanisms were due to the TGF-β-mediated combinatorial actions of mTOR and TCF1. In addition to the preventive role, the therapeutic effects of iTregs on the established STZ-T1D and nonobese diabetic (NOD) mouse models were tested, which revealed improved β cell function. Our findings therefore provide key new insights into the basic mechanisms involved in the therapeutic role of iTregs in T1D.

Keywords: TGF-β; induced regulatory T cells; mTOR and TCF1; type 1 cytotoxic T cells; type 1 diabetes mellitus.

Fig. 1

iTregs ameliorated the progression of T1D and improved β cell function. iTregs (CD4⁺ Foxp3⁺) or Med cells (CD4⁺ Foxp3^-) from Thy1.1⁺ FoxP3^GFP mice were injected into Thy1.2⁺ T1D mice via the tail vein. A Schematic diagram for the in vivo experiment. B Nonfasting blood glucose levels of mice in the three groups (n = 6–8 per group). Two-way ANOVA; data are shown as the mean ± s.d. C Incidence of diabetes. The chi-square test was used to compare groups. D IPGTT results for day 15 after iTregs administration (n = 6–10 per group). The Mann–Whitney test was used to compare the Model and iTregs groups; data are shown as the mean ± s.d. E AUC for the IPGTT. Two-tailed unpaired t test; data are shown as the mean ± s.d. F and G Serum insulin and C-peptide levels on day 15 (n = 4 per group). Two-tailed unpaired t-test; data are shown as the mean ± s.d. H Representative images of insulin and glucagon staining. Insulin or glucagon expression is indicated by brown staining in cells. I and J All islets seen in one section per pancreas were used for analysis. Statistical analysis of islet size and β cell area in relation to islet area among the three groups (n = 5 per group). One-way ANOVA; data are presented as the mean ± s.d. Experiments were repeated three times. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001; ns indicates no significance

Fig. 2

Adoptively transferred iTregs migrate to the pancreas and immune organs and maintain their phenotypic features in vivo. iTregs (CD4⁺ Foxp3⁺) from Thy1.1⁺ FoxP3^GFP mice were injected into Thy1.2⁺ T1D mice, and the iTregs organ distribution, engraftment and stability were then examined on day 15 and day 30. A Representative flow cytometry data showing the distributions of iTregs (CD4⁺ GFP⁺ cells) on day 15 (upper) and day 30 (lower) in the spleen (SP), mesenteric and inguinal lymph nodes (LNs) and pancreatic lymph nodes (PLNs). Gated on total live cells. Analysis of the iTregs cell percentage in the total live cells (n = 5) (B) and in relation to the CD4⁺ T cells (n = 5) (C) in the SP, LNs and PLNs. Two-tailed unpaired t-test; data are shown as the mean ± s.d. D Analysis of iTregs cell numbers (n = 7) in the SP, LNs and PLNs. Two-tailed unpaired t-test; data are shown as the mean ± s.d. Representative flow cytometry data (E) and statistical data (F) showing the iTregs distribution in PBMCs gated on total live cells (n = 5). Two-tailed unpaired t-test; data are shown as the mean ± s.d. G Analysis of the iTregs cell percentage in relation to CD4⁺ T cells (n = 5) in PBMCs. Two-tailed unpaired t-test; data are shown as the mean ± s.d. Representative flow cytometry data (H) and statistical data (I) showing the iTregs distribution in the pancreas gated on total live cells (n = 5). Two-tailed unpaired t-test; data are shown as the mean ± s.d. J Analysis of the iTregs cell percentage in the CD4⁺ T cell population (n = 5) in the pancreas. Two-tailed unpaired t-test; data are shown as the mean ± s.d. Representative flow data (K) and statistical data (L) showing Foxp3 expression on day 15 and day 30 in cells in different tissues gated on CD4⁺ Thy1.1⁺ cells (n = 3). Two-way ANOVA; data are shown as the mean ± s.d. M Representative flow cytometry data showing the conversion of iTregs or Med cells into Th1 and Th17 cells on day 30; gated on CD4⁺ Thy1.1⁺ cells. Experiments were repeated three times. *P < 0.05, **P < 0.01, and ***P < 0.001; ns indicates no significance

Fig. 3

iTregs rebalance endogenous IFN-γ-producing CD8 T cells and regulatory T cells (Tregs) in recipient STZ-induced T1D mice. Lymphocytes were isolated from the SP, LNs and PLNs and then stimulated in vitro with PMA (50 ng/ml) and ionomycin (500 ng/ml) for 5 h, with brefeldin A (10 μg/ml) added for the last 4 h, and the intracellular expression of IFN-γ on CD8⁺ T cells was analyzed by flow cytometry. A Representative flow cytometry data showing IFN-γ expression gated on CD8⁺ Thy1.2⁺ T cells. Statistical analysis of the IFN-γ⁺ population in CD8⁺ Thy1.2⁺ T cells on day 15 (B) and day 30 (C) in different tissues (n = 4). Two-way ANOVA; data are shown as the mean ± s.d. D Representative flow cytometry data showing the percentage of Foxp3⁺ cells gated on CD4⁺ Thy1.2⁺ T cells on day 15. Statistical analysis of the Foxp3⁺ cell population in CD4⁺ Thy1.2⁺ cells on day 15 (E) and day 30 (F) in different tissues (n = 4). Two-way ANOVA; data are shown as the mean ± s.d. Ratios of the recipient Treg percentage to the Tc1 percentage on day 15 (G) and day 30 (H) (n = 4 per group). Two-way ANOVA; data are shown as the mean ± s.d. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001; ns indicates no significance

Fig. 4

iTregs suppress Tc1 cells through the TGF-β-dependent combinational action of mTOR and TCF1 signaling in vitro. Representative flow cytometry data and statistical analysis showed that iTregs did not affect CD8⁺ T cell proliferation (A, B) or apoptosis (C, D) in vivo (n = 5). One-way ANOVA; data are shown as the mean ± s.d. Naïve CD8⁺ T cells were polarized into Tc1 cells in vitro, and TGF-β (2 ng/ml), iTregs (iTregs: CD8⁺ T cells = 1:1) or an ALK5 inhibitor (10 µM) was added to the culture system. E, F Representative flow cytometry data and statistical analysis showed that iTregs suppressed IFN-γ induction in CD8⁺ T cells, similar to TGF-β, and this function relied on TGF-β signaling. Two-way ANOVA; data are presented as the mean ± s.d. of five independent experiments. G, H Representative flow cytometry data and statistical analysis showing the phosphorylation of the S6 protein at 24 hours among different groups. Two-way ANOVA; data are presented as the mean ± s.d. of four independent experiments. I, J Representative flow cytometry data and statistical analysis showed that iTregs maintained high expression of TCF1 at 72 h during Tc1 cell induction, and this effect was partially dependent on TGF-β signaling. Two-way ANOVA; data are presented as the mean ± s.d. of four independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001; ns indicates no significance

Fig. 5

The therapeutic role of iTregs in STZ-induced T1D mice is dependent on TGF-β signaling. An ALK5 inhibitor was injected into STZ-T1D mice to verify the role of TGF-β in iTregs-mediated protection. A Schematic diagram for the in vivo experiment. B Nonfasting blood glucose levels and diabetes incidence are shown (n = 5 for the Model group and n = 6 for the iTregs group). C Incidence of diabetes. Chi-square test comparing groups. D HE, insulin and glucagon staining in four groups. Insulin or glucagon expression is indicated by brown staining in cells. E, F Statistical analysis of insulitis scores among the four groups (n = 5 per group). One-way ANOVA; data are presented as the mean ± s.d. G, H All islets seen in one section per pancreas were used for analysis. Statistical analysis of islet size and β cell area in relation to islet area among the four groups (n = 5 per group). One-way ANOVA; data are presented as the mean ± s.d. I, J Relative mTOR and TCF1 mRNA levels in lymphocytes from the LNs on day 15 (n = 5 per group). One-way ANOVA; data are presented as the mean ± s.d. K, L Representative western blot and statistical analysis showing phosphorylated S6 protein expression in lymphocytes from the LNs on day 15 (n = 5 per group). One-way ANOVA; data are presented as the mean ± s.d. M, N Representative western blot and statistical analysis showing TCF1 protein expression in lymphocytes from the LNs on day 15 (n = 5 per group). One-way ANOVA; data are presented as the mean ± s.d. Experiments were repeated twice, with n = 15 for the Model, Model+ALK5 inhibitor, and iTreg+ALK5 inhibitor groups and n = 16 for the iTregs group in total. *P < 0.05, **P < 0.01, and ****P < 0.0001; ns indicates no significance

Fig. 6

iTregs improved β cell function in mice with established T1D. Diabetes was induced in C57BL/6 mice with STZ, and iTregs were injected after disease onset. A Schematic diagram for the therapeutic experiment. iTregs were injected into diabetic mice (n = 10) 7 days after STZ injection, and three mice with improved β cell function were sacrificed eight days after the cell intervention for analysis of the pancreas. A second injection of iTregs was performed 26 days after the STZ injection (n = 7), and blood glucose levels were monitored thereafter. B Nonfasting blood glucose levels are shown (n = 7–10). C Diabetes incidence is shown. D Results for an IPGTT performed one week after iTregs injection are shown. Two-tailed unpaired t-test; data are shown as the mean ± s.d. Three mice per group were sacrificed eight days after the first iTregs injection. Representative images of HE, insulin and glucagon staining (E). Statistical analysis of insulitis scores between the groups (n = 3). Two-tailed unpaired t test; data are shown as the mean ± s.d (F, G). All islets seen in three sections per pancreas were included in the analysis. Statistical analysis of islet area (H) and β cell area in relation to islet area (I). Two-tailed unpaired t test; data are shown as the mean ± s.d. iTregs were injected into 6-week-old (n = 5 for each group) and 10-week-old (n = 5 for each group) NOD mice. Week 0 was the day of iTregs injection, and blood glucose levels were monitored for 81 days. J Diabetes incidence in the two groups. K Representative image of insulin staining. All islets seen in three sections per pancreas were included in the analysis. Statistical analysis of islet area (L) and β cell area in relation to islet area (M). Two-tailed unpaired t-test; data are shown as the mean ± s.d. *P < 0.05, ***P < 0.001, and ****P < 0.0001

Fig. 7

Model of iTregs migration, stability and function in STZ-T1D mice. iTregs migrated to PBMCs, the pancreas, the spleen and the lymph nodes. iTregs survived for at least 30 days in STZ-T1D mice and were stable in vivo, as indicated by the maintained expression of Foxp3 and resistance to secretion of IL-17 and IFN-γ. iTregs could protect β cell function, possibly by upregulating Treg levels in recipient mice while inhibiting type 1 cytotoxic T cell (Tc1) function, thus rebalancing the Treg/Tc1 cell ratio in recipients. The effect of iTregs on Tc1 cells was dependent on TGF-β-mediated combinational action on mTOR and TCF1 signaling