TSC1 regulates the balance between effector and regulatory T cells

Abstract

Mammalian target of rapamycin (mTOR) plays a crucial role in the control of T cell fate determination; however, the precise regulatory mechanism of the mTOR pathway is not fully understood. We found that T cell-specific deletion of the gene encoding tuberous sclerosis 1 (TSC1), an upstream negative regulator of mTOR, resulted in augmented Th1 and Th17 differentiation and led to severe intestinal inflammation in a colitis model. Conditional Tsc1 deletion in Tregs impaired their suppressive activity and expression of the Treg marker Foxp3 and resulted in increased IL-17 production under inflammatory conditions. A fate-mapping study revealed that Tsc1-null Tregs that lost Foxp3 expression gained a stronger effector-like phenotype compared with Tsc1-/- Foxp3+ Tregs. Elevated IL-17 production in Tsc1-/- Treg cells was reversed by in vivo knockdown of the mTOR target S6K1. Moreover, IL-17 production was enhanced by Treg-specific double deletion of Tsc1 and Foxo3a. Collectively, these studies suggest that TSC1 acts as an important checkpoint for maintaining immune homeostasis by regulating cell fate determination.

Figure 1. TSC1 function in T cells preserves intestinal homeostasis.

(A) H&E staining and histological scores of colon and liver tissue sections from 6-month-old mice. Original magnification, ×100. Data are representative of (left) and compiled from (right) 6 mice. Error bars indicate the mean ± SD. *P < 0.05 by two-tailed, unpaired Student’s t test. (B) Overview of DSS-induced chronic colitis model. Mice were administrated 2% DSS for 7 days followed by water and were analyzed up to 4 weeks later. (C) Body weight changes in WT and Cd4^CreTsc1^f/f mice after DSS treatment. Weight loss of individual mice was monitored every 2 days. Data are compiled from three independent experiments with three mice each. Error bars indicate the mean ± SD. *P < 0.05 by two-tailed, unpaired Student’s t test. (D–F) Colon length (D); photograph of representative spleen (SP) and mesenteric lymph nodes (MLNs) (E); H&E staining and histology scores of colon (F). Original magnification, ×100 (F). (G) Flow cytometric analysis of cytokine production (left) and frequencies (right) in colonic lamina propria (cLP) and splenic (SP) CD4⁺ T cells. Cells were obtained from WT and Cd4^CreTsc1^f/f mice 3 weeks after DSS removal (day 28) and restimulated in vitro. Data are compiled from (D) or representative of (E–G) three independent experiments. Each symbol represents an individual mouse (n = 5–6). Error bars indicate the mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 by two-tailed, unpaired Student’s t test.

Figure 2. TSC1 deficiency promotes Th1 and Th17 differentiation.

(A) Naive CD4⁺ T cells from WT and Cd4^CreTsc1^f/f mice were cultured under Th1- and Th17-polarizing condition for 5 days. IFN-γ–, IL-4–, or IL-17A–producing cells were analyzed by intracellular cytokine staining (ICCS) 6 hours after restimulation with anti-CD3/CD28. (B) Th1 and Th17 cell cytokine production was measured by ELISA 24 hours after restimulation with anti-CD3/CD28. (C) RT-PCR analysis of mRNA expression levels of Th subset–specific transcription factors in polarized Th1 or Th17 cells. Data are representative of (A) or compiled from (B and C) three to five independent experiments. Error bars indicate the mean ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 by two-tailed, unpaired Student’s t test . (D) IFN-γ and IL-17 production (upper panel) and frequencies (lower panel) in T cells from recipient mice immunized with OVA/CFA. CD4⁺ T cells from WT or Cd4^CreTsc1^f/f OT-II mice were adoptively transferred into CD45.1 congenic B6 mice, followed by immunization with OVA/CFA. Splenocytes and lymph node cells obtained 6 days later were stimulated with OVA_323–339 peptide for 24 hours and analyzed by ICCS and flow cytometry. (E) Immunization of WT and Cd4^creTsc1^f/f mice with KLH/CFA. CD4⁺T cells obtained 6 days later were stimulated with KLH for 72 hours, and the culture supernatants were analyzed by Bio-Plex multicytokine assay. Data are representative of (D) or compiled from (E) three independent experiments. Each symbol represents an individual mouse (n = 3). Error bars indicate the mean ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 by two-tailed, unpaired Student’s t test.

Figure 3. Treg cells from Cd4^creTsc1^f/f mice have a defect in colitis suppression.

(A) Sorted naive CD4⁺CD62L⁺CD25^– T cells from WT and Cd4^CreTsc1^f/f mice were stimulated with anti-CD3/CD28 in the presence of the indicated concentrations of TGF-β. Induction of Foxp3⁺ Treg cells was examined by flow cytometry on day 4. (B) Cells in A were cocultured with Violet-labeled CD4⁺CD25^– naive T cells at a 1:1 ratio in the presence of irradiated T cell–depleted splenocytes and anti-CD3. Violet dilution was assessed 4 days later by flow cytometry. Data are representative of three to five independent experiments (A and B). (C) Rag1^–/– mice were given sorted WT or TSC1-deficient CD4⁺CD25⁺ (CD45.2⁺) Treg cells, together with CD4⁺CD45RB^hi (CD45.1⁺) T cells or CD4⁺CD45RB^hi (CD45.1⁺) T cells alone (None). Weight loss in individual mice was monitored every week for 12 weeks. (D–F) H&E staining and colon histological scores. Original magnification, ×100 (D); ratios of CD4⁺CD45RB^hi (CD45.1⁺) to CD4⁺CD25⁺ (CD45.2⁺) Treg cells in the SP, LNs, MLNs, and cLP (E); absolute numbers of CD4⁺CD45.2⁺ T cells (F) in Rag1^–/– recipient mice, as in C, 12 weeks after transfer. (G and H) Flow cytometric analysis of Foxp3⁺ expression in sorted CD4⁺CD45RB^loCD25⁺ Treg cells before adoptive transfer (G) or 12 weeks after transfer (H) in Rag1^–/– recipient mice, as in C. Data are compiled from (C, E, and F) or representative of (D and H) three independent experiments with two mice each. Error bars indicate the mean ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 by two-tailed, unpaired Student’s t test.

Figure 4. TSC1 deficiency in Foxp3⁺ Treg cells results in loss of Treg function under inflammatory condition.

(A and B) Flow cytometric analysis (left), frequencies (right) of YFP expression (A), and absolute numbers of CD4⁺YFP⁺ T cells (B) from spleen or cLP of Foxp3^YFPCreTsc1^+/+ and Foxp3^YFPCreTsc1^f/f mice. (C) Flow cytometric analysis of Ki-67 expression in CD4⁺YFP⁺T cells (upper panel) and 7-AAD staining in CD4⁺YFP⁺T cells 48 hours after anti-CD3/CD28 stimulation (lower panel) in Foxp3^YFPCreTsc1^+/+ and Foxp3^YFPCreTsc1^f/f mice. Data are representative of (A and C) or compiled from (B) three independent experiments. Error bars indicate the mean ± SD by two-tailed, unpaired Student’s t test. (D) Rag1^–/– mice were given sorted CD4⁺YFP⁺ (CD45.2⁺) Treg cells from Foxp3^YFPCreTsc1^+/+ or Foxp3^YFPCreTsc1^f/f mice, together with CD4⁺CD45RB^hi (CD45.1⁺) T cells, or CD4⁺CD45RB^hi (CD45.1⁺) T cells alone (None). Weight loss of individual mice was monitored every week for 12 weeks. (E–G) Ratios of CD4⁺CD45RB^high (CD45.1⁺) to CD4⁺YFP⁺ (CD45.2⁺) Treg cells in the SP, LNs, MLNs, and cLP (E); absolute numbers of CD4⁺CD45.2⁺ T (Treg) cells (F); frequencies of the indicated cytokine-producing CD4⁺CD45.1⁺ T cells in spleen and cLP (G) in Rag1^–/– recipient mice as in D 12 weeks after transfer. (H and I) Flow cytometric analysis of Foxp3⁺ expression in sorted CD4⁺CD45RB^loYFP⁺ Treg cells before adoptive transfer (H) or 12 weeks after transfer (I) in Rag1^–/– recipient mice as in D. Data are compiled from (D–G) or representative of (I) three independent experiments with two mice each. Error bars indicate the mean ± SD. *P < 0.05 and **P < 0.01 by two-tailed, unpaired Student’s t test.

Figure 5. TSC1-deficient Foxp3⁺ Treg cells produce IL-17 and convert into effector-like T cells.

(A) Flow cytometric analysis of IL-17A–producing CD45.2⁺CD4⁺ Treg cells from cLP of recipient mice as in Figure 4D, 6 weeks after transfer, followed by restimulation with PMA/ionomycin for 6 hours. (B) Flow cytometric analysis of YFP and RFP expression in CD4⁺ T cells (upper panel) or Foxp3 expression in CD4⁺RFP⁺ T cells (lower panel) in Foxp3^YFPCreR26^RFPTsc1^+/+ and Foxp3^YFPCreR26^RFPTsc1^f/f mice. (C) Immunoblot analysis of TSC1 in sorted CD4⁺RFP^–YFP^–, CD4⁺RFP⁺YFP^–, and CD4⁺RFP⁺YFP⁺ T cells from Foxp3^YFPCreR26^RFPTsc1^+/+ and Foxp3^YFPCreR26^RFPTsc1^f/f mice. (D and E) Sorted CD4⁺RFP⁺YFP⁺ Treg cells (D, right panel) or CD4⁺RFP⁺YFP^– ex-Treg cells (D, left panel), or CD4⁺CD62L⁺RFP^–YFP^– naive T cells (E) from Foxp3^YFPCreR26^RFPTsc1^+/+ and Foxp3^YFPCreR26^RFPTsc1^f/f mice were stimulated with anti-CD3/CD28 for 36 hours. Cytokine production was measured by Bio-Plex multicytokine assay. (F) Sorted CD4⁺RFP⁺YFP⁺ Treg cells were stimulated with anti-CD3/CD28, IL-2, and the indicated cytokines for 48 hours. Cytokine production was measured by Bio-Plex multicytokine assay. (G and H) Rag1^–/– mice were given sorted CD4⁺RFP⁺YFP⁺ (CD45.2⁺) Treg cells from Foxp3^YFPCreR26^RFPTsc1^+/+ or Foxp3^YFPCreR26^RFPTsc1^f/f mice, together with CD4⁺CD45RB^hi (CD45.1⁺) T cells. Flow cytometric analysis of RFP and YFP expression (G) or cytokine production (H) in CD45.2⁺CD4⁺RFP⁺ T cells of the recipient mice 6 weeks after transfer. Data are representative of (A, B, G, and H) or compiled from (D–F) two to five independent experiments. Error bars indicate the mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 by two-tailed, unpaired Student’s t test.

Figure 6. TSC1-deficient Treg cells acquire a Th17-like phenotype in female Foxp3^YFPCre/+Tsc1^f/f mice.

(A) Flow cytometric analysis of YFP, CD25, and Foxp3 expression in CD4⁺ T cells from spleen of female Foxp3^YFPCre/+Tsc1^f/f mice. (B) Sorted CD4⁺CD25⁺NRP1⁺YFP^– or CD4⁺CD25⁺NRP1⁺YFP⁺ Treg cells from spleen of female Foxp3^YFPCre/+Tsc1^f/f mice were stimulated with anti-CD3/CD28 for 36 hours. Cytokine production was measured by Bio-Plex multicytokine assay. (C) Flow cytometric analysis of IL-17 production by CD4⁺Foxp3⁺YFP^– or CD4⁺Foxp3⁺YFP⁺ Treg cells from MLNs, Peyer’s patch (PP), and cLP of female Foxp3^YFPCre/+Tsc1^f/f mice. Cells were restimulated with PMA/ionomycin for 6 hours. Data are representative of (A and C) or compiled from (B) three to four independent experiments. Error bars indicate the mean ± SD. *P < 0.05 and ***P < 0.001 by two-tailed, unpaired Student’s t test.

Figure 7. S6K1 knockdown in TSC1-deficient T cells leads to attenuated IL-17 production.

(A) Analysis of the phosphorylation status of the indicated downstream targets of TSC1. CD4⁺ T cells from WT or CD4^creTsc1^f/f mice were stimulated with anti-CD3/CD28 for the indicated time periods, and the cell lysates were subjected to immunoblotting with the indicated antibodies. (B) Analysis of GFP expression in peripheral blood from bone marrow chimeric mice 2Πmonths after reconstitution of control (shCON) or S6K1 or GRB10 shRNAs expressing WT or Cd4^CreTsc1^f/f bone marrow cells (upper panel). Immunoblot analysis of S6K1 and GRB10 was performed in sorted CD4⁺GFP⁺ T cells (lower panel). (C and D) Flow cytometric analysis (C, left panel), frequencies (C, right panel), or cytokine production (D) of Th17-polarized naive CD4⁺GFP⁺ T cells from bone marrow chimeric mice as in B. (E) Analysis of the phosphorylation status of S6K1 in CD4⁺YFP⁺ Treg cells. (F) Analysis of mAmetrine expression in peripheral blood from bone marrow chimeric mice 2 months after reconstitution of control or S6K1 shRNAs expressing Foxp3^YFPCreTsc1^+/+ or Foxp3^YFPCreTsc1^f/f bone marrow cells. (G) Cytokine production in CD4⁺YFP⁺mAmetrine⁺ T cells from bone marrow chimeric mice as in F was measured by Bio-Plex multicytokine assay 36 hours after stimulation with anti-CD3/CD28. Data are representative of (B, C, and F) or compiled from (D and G) three to five independent experiments. Error bars indicate the mean ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 by two-tailed, unpaired Student’s t test.

Figure 8. Double deletion of TSC1 and Foxo3a heightens Treg conversion to Th17-like cells.

(A) Flow cytometric analysis of YFP and RFP expression in CD4⁺T cells of Foxp3^YFPCreR26^RFPTsc1^+/+, Foxp3^YFPCreR26^RFPTsc1^f/f, and Foxp3^YFPCreR26^RFPTsc1^f/fFoxo3a^f/f mice. (B) Sorted CD4⁺RFP⁺YFP⁺ Treg cells (right) or CD4⁺RFP⁺YFP^– ex-Treg cells (left) from the mice in A were stimulated with anti-CD3/CD28 for 36 hours. Cytokine production was measured by Bio-Plex multicytokine assay. (C) Sorted CD4⁺RFP⁺YFP⁺ Treg cells (right) or CD4⁺RFP⁺YFP^– ex-Treg cells (left) from the mice in A were stimulated with anti-CD3/CD28 for 72 hours in the presence or absence of 250 nM rapamycin. Cytokine production was measured by Bio-Plex multicytokine assay. Data are representative of (A) or compiled from (B and C) three independent experiments. Error bars indicate the mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 by two-tailed, unpaired Student’s t test.