PDK1 regulation of mTOR and hypoxia-inducible factor 1 integrate metabolism and migration of CD8+ T cells

Abstract

mTORC1 (mammalian target of rapamycin complex 1) controls transcriptional programs that determine CD8+ cytolytic T cell (CTL) fate. In some cell systems, mTORC1 couples phosphatidylinositol-3 kinase (PI3K) and Akt to the control of glucose uptake and glycolysis. However, PI3K-Akt-independent mechanisms control glucose metabolism in CD8+ T cells, and the role of mTORC1 has not been explored. The present study now demonstrates that mTORC1 activity in CD8+ T cells is not dependent on PI3K or Akt but is critical to sustain glucose uptake and glycolysis in CD8+ T cells. We also show that PI3K- and Akt-independent pathways mediated by mTORC1 regulate the expression of HIF1 (hypoxia-inducible factor 1) transcription factor complex. This mTORC1-HIF1 pathway is required to sustain glucose metabolism and glycolysis in effector CTLs and strikingly functions to couple mTORC1 to a diverse transcriptional program that controls expression of glucose transporters, multiple rate-limiting glycolytic enzymes, cytolytic effector molecules, and essential chemokine and adhesion receptors that regulate T cell trafficking. These data reveal a fundamental mechanism linking nutrient and oxygen sensing to transcriptional control of CD8+ T cell differentiation.

Figure 1.

mTORC1 regulates glucose uptake and glycolysis in TCR-stimulated CD8⁺ T cells. (A) Immunoblot analysis of Glut1 expression in naive OTI CD8⁺ T cells ± TCR (SIINFEKL) stimulation for 20 h. (B and C) Naive P14-LCMV CD8⁺ T cells ± TCR (gp33-41/anti-CD28) stimulation were assayed for glucose uptake (B) and lactate production (C). (D) Immunoblot analysis of naive OTI CD8⁺ T cells ± TCR (SIINFEKL) stimulation for 20 h. mTORC1 activity was determined by analyzing the phosphorylation of target sequences on S6K1 (T389 and S241/242) and phosphorylation of the S6K1 substrate S6 ribosomal protein. PTEN was used as a loading control. (E–G) Immunoblot analysis (E) and analysis of glucose uptake (F) and lactate production (G) for naive P14-LCMV CD8⁺ T cells ± TCR (gp33-41/anti-CD28) stimulation with or without rapamycin for 20 h. For all panels, data are mean ± SEM or representative of at least three experiments. All metabolic assays were preformed in triplicate (**, P < 0.01; ***, P < 0.001). Molecular mass is indicated in kilodaltons.

Figure 2.

mTORC1 regulates glucose uptake and glycolysis in IL-2–maintained CTLs. (A) Immunoblot analysis for Glut1 expression in naive OTI CD8⁺ T cells ± TCR (SIINFEKL) stimulation for 20 h and also mature OTI CTLs. The black line indicates that intervening lanes have been spliced out. (B) Immunoblot analysis for Glut1 expression in CTLs treated with or without IL-2 for 20 h. 4EBP1 was used as a loading control. (C and D) Analysis of glucose uptake (C) and lactate production (D) in P14-LCMV CTLs treated with or without IL-2 for 20 h. (E and F) Analysis of lactate production (E) and glucose uptake (F) in P14-LCMV CTLs treated with or without rapamycin for 20 h. (G) Immunoblot analysis for Glut1 expression in P14-LCMV CTLs treated with or without rapamycin for 20 h. 4EBP1 was used as a loading control. (A–G) Data are mean ± SEM or representative of at least three experiments. All metabolic assays were preformed in triplicate (**, P < 0.01; ***, P < 0.001). Molecular mass is indicated in kilodaltons. (H) SILAC-based proteomic analysis of P14-LCMV CTLs treated with and without rapamycin for 48 h. Shown is the relative expression of rate-limiting glycolytic enzymes (rapamycin/untreated). Data are mean ± SEM for three experiments and were analyzed by ANOVA (**, P < 0.01; ***, P < 0.001). CD8α was used as a control protein with unchanged expression.

Figure 3.

mTORC1 controls glucose uptake and glycolysis via HIF1. (A–D) Immunoblot analysis of HIF1α and HIF1β expression in naive P14-LCMV CD8⁺ T cells ± TCR (gp33-41/anti-CD28) stimulation for 20 h (A and C) and P14-LCMV CTLs (A, B, and D) treated with and without IL-2 (B) or rapamycin (C and D) for 20 h. Phospho-S6K1 and phospho-S6 were used as a measure of mTORC1 activity. (E) Immunoblot analysis of c-myc expression in CTLs treated with or without rapamycin for 20 h. Phospho-S6K1 and phospho-S6 were used as a measure of mTORC1 activity. (F) Flow cytometric analysis of HIF1β^WT/WT CD4Cre (WT) and HIF1β^flox/flox CD4Cre (HIF1^−/−) CD8⁺ T cells after TCR (2c11) stimulation for 20 h. (G) Immunoblot analysis of WT and HIF1^−/− CTLs. (H and I) Analysis of glucose uptake (H) and lactate production (I) in WT and HIF1^−/− naive CD8⁺ T cells after TCR (2c11/anti-CD28) stimulation for 20 h. Glucose uptake in unstimulated WT naive T cells is also shown (uptake in unstimulated HIF1^−/− naive T cells is equivalent to WT; not depicted). (J–L) Analysis of glucose uptake (J), Glut1 expression (K), and lactate production (L) in WT versus HIF1^−/− CTLs. (M) A comparison of the transcriptional profile of HIF1 WT versus HIF1^−/− CTLs was performed by microarray. Shown here are KEGG pathway analysis of genes down-regulated in HIF1^−/− CTLs (top) and a heat map of the relative normalized expression of selected genes that are significantly different in expression in WT versus HIF1^−/− CTLs, as determined by microarray. For all panels, data are mean ± SEM or representative of at least three experiments. All metabolic assays were preformed in triplicate (**, P < 0.01; ***, P < 0.001). Molecular mass is indicated in kilodaltons. Dotted lines indicate that intervening lanes have been spliced out.

Figure 4.

The HIF1 complex regulates the CD8⁺ T cell transcriptional program but is not essential for T cell proliferation. (A and B) Proliferation analysis of WT and HIF1^−/− CTLs (A) and P14-LCMV CTLs treated with and without rapamycin (B). Cells were seeded at 0.3 × 10⁶/ml, and CTL numbers were counted after 24 h. Data are mean ± SEM of five experiments. (C) Heat map showing the relative normalized expression of selected genes that are significantly different in expression in WT versus HIF1^−/− CTLs, as determined by microarray. (D) Real-time PCR analysis of IFN-γ expression in WT and HIF1^−/− CTLs. Data are mean ± SEM of three experiments in triplicate. (E) Immunoblot analysis of T-bet and Blimp1 expression in WT and HIF1^−/− CTLs. Data are representative of two experiments. (F) Real-time PCR analysis of Perforin mRNA expression in WT and HIF1^−/− CTLs. Data are mean ± SEM of three experiments in triplicate (**, P < 0.01). (G) Immunoblot analysis of Perforin protein expression in WT and HIF1^−/− CTLs (top) and P14-LCMV CTLs treated ± rapamycin for 20 h (bottom). Data are representative of at least three experiments. (H) IL-2–maintained CTLs were placed in either hypoxic (1%) or normoxic (20%) oxygen for 24 h before being subjected to immunoblot analysis for HIF1α, Glut1, and perforin expression. Data are representative of three experiments. (I) ChIP was performed with anti–Pol II, and the changes in Pol II binding to the Perforin transcription start site and the second exon were quantified by real-time PCR. Data were normalized to input DNA amounts and plotted as fold over the values for Pol II binding to the HPRT proximal promoter. Data are mean ± SEM of three experiments performed in duplicate. (J) P14-LCMV T cells were activated for 2 d with gp33-41 and then cultured for a further 4 d with IL-2 in different glucose concentrations. Cells were then subjected to immunoblot analysis for perforin expression. Data are representative of two experiments. Molecular mass is indicated in kilodaltons. Black lines (solid or dotted) indicate that intervening lanes have been spliced out.

Figure 5.

HIF1 regulation of chemokines and chemokine receptors. (A) A comparison of the transcriptional profile of WT versus HIF1^−/− CTLs was performed by microarray. Shown here are KEGG pathway analysis of genes up-regulated in HIF1^−/− CTLs (top) and a heat map showing the relative normalized expression of selected genes that are significantly different in expression in WT versus HIF1^−/− CTLs, as determined by microarray. (B) Real-time PCR analysis of CD62L expression in WT and HIF1^−/− CTLs. (C) Analysis of CD62L surface expression on WT and HIF1^−/− CTLs by flow cytometry. (D) Real-time PCR analysis of CCR7 expression in WT and HIF1^−/− CTLs. (E) WT and HIF1^−/− CTLs were labeled with CFSE or CellTracker orange (CMTMR) and mixed at a ratio of 1:1 before being injected into C57BL/6 host mice. Values indicate recovery of WT or HIF1^−/− cells as a percentage of the total recovered transferred cells from the blood and lymph nodes 4 h after transfer. Each dot indicates a mouse; horizontal bars indicate mean. (F) P14 T cells were activated for 2 d with cognate peptide and then cultured for a further 4 d with IL-2 in different glucose concentrations. Cells were then analyzed for the surface expression of CD62L by flow cytometry. In B and D, mean ± SEM of three experiments performed in triplicate is shown; in C and F, data are representative of at least three experiments (*, P < 0.05; **, P < 0.01).

Figure 6.

mTORC1 and HIF1 do not regulate Akt activity or Foxo phosphorylation. (A) Immunoblot analysis of phosphorylated AKT and Foxos in P14-LCMV CTLs treated with and without Akti1/2 or rapamycin for 24 h. (B) P14-LCMV CTLs were treated with and without rapamycin or Akti1/2 for 24 h and subjected to nuclear/cytoplasmic fractionation before immunoblot analysis for Foxo1 and Foxo3a expression. Purity of cytoplasmic and nuclear fractions was confirmed by IκBα and Smc1 expression. (C) Immunoblot analysis of WT or HIF1^−/− CTLs for Akt–Foxo and mTORC1 signaling. HIF1^−/− CTLs were treated with rapamycin as a negative control for mTORC1 activity. For all panels, data are representative of at least three experiments. Molecular mass is indicated in kilodaltons. Dotted lines indicate that intervening lanes have been spliced out.

Figure 7.

PI3K and Akt do not regulate mTORC1 activity. (A and B) CTLs were cultured in the presence or absence of Akti1/2, IC87114, rapamycin, or LY294002 for 60 min (A and B) or 24 h (A) and subjected to immunoblot analysis with the indicated antibodies. (C) CTLs generated from WT or p110δ^D910A mice were subjected to immunoblot analysis with or without rapamycin treatment (30 min). Data are representative of two experiments. (D) CTLs were cultured in the presence or absence of Akti1/2, IC87114, or rapamycin for 24 h and subjected to immunoblot analysis with the indicated antibodies. (E–G) CTLs generated from PDK1^flox/flox TamoxCre (PDK1^Flox) and PDK1^WT/WT TamoxCre (WT) mice were treated ± 4′OHT for 3 d to delete PDK1 and subjected to immunoblot analysis. For, A, B, and D–G, data are representative of at least three experiments. Molecular mass is indicated in kilodaltons.