TLR-driven early glycolytic reprogramming via the kinases TBK1-IKKɛ supports the anabolic demands of dendritic cell activation

Abstract

The ligation of Toll-like receptors (TLRs) leads to rapid activation of dendritic cells (DCs). However, the metabolic requirements that support this process remain poorly defined. We found that DC glycolytic flux increased within minutes of exposure to TLR agonists and that this served an essential role in supporting the de novo synthesis of fatty acids for the expansion of the endoplasmic reticulum and Golgi required for the production and secretion of proteins that are integral to DC activation. Signaling via the kinases TBK1, IKKɛ and Akt was essential for the TLR-induced increase in glycolysis by promoting the association of the glycolytic enzyme HK-II with mitochondria. In summary, we identified the rapid induction of glycolysis as an integral component of TLR signaling that is essential for the anabolic demands of the activation and function of DCs.

Figure 1

TLR ligation induces a rapid increase in glycolytic metabolism in DCs. (a) Real-time changes in the ECAR and OCR of GM-DCs left untreated (−) or treated with LPS (vertical dotted line indicates initiation of treatment throughout). (b,c) Lactate production (b) and glucose consumption (c) in GM-DCs left unstimulated (−) or stimulated for 6 h with LPS. (d) ¹³C-labeled metabolites in GM-DCs incubated for 1 h with medium (−) or LPS in the presence of 1,2-¹³C-glucose, detected by NMR spectrometry and presented relative to the total pool of that metabolite (m+1, m+2, m+3 (horizontal axes), mass of metabolite (m) + mass units (1, 2, 3) derived from incorporated glucose-derived ¹³C). (e) Total metabolite profiling of GM-DCs stimulated for 1 and 3 h with LPS, assessed by pathway-enrichment analysis, based on changes in the relative abundance of metabolites; key lists the five pathways with the most significant change (in ranked order, i–v). (f,g) Real-time changes in the ECAR and OCR of GM-DCs (f) and mouse DCs generated by stimulation with Flt3L and cultured in vitro (g), treated with various TLR agonists (TLR-L (key); f) or with poly(I:C) or LPS (g). *P < 0.05, **P < 0.01 and ***P < 0.001 (Student’s t-test). Data are from one experiment representative of over ten experiments (a) or three experiments (f,g; mean ± s.e.m. of triplicates) or represent three independent experiments (b–d; mean and s.e.m.) or are representative of five independent experiments (e).

Figure 2

The glycolytic burst is required for the activation and function of DCs. (a) Real-time changes in the ECAR and OCR (as in Fig. 1a) of GM-DCs left untreated (−) or treated with LPS with or without 2-DG (key). (b,c) Expression of markers CD86 and CD40 or of MHC class I and II (b) or cytokines IL-6, IL-12 and TNF (c) in GM-DCs either left untreated (−) or given no pretreatment (LPS) or pretreated with 2-DG (2-DG + LPS) and then stimulated for 6 h with LPS, analyzed by flow cytometry with surface (b) or intracellular (c) staining. Numbers in top right corners (b) indicate percent CD86⁺CD40⁺ cells (top row) or MHCI⁺MHCII⁺ cells (bottom row). Numbers above bracketed lines (c) indicate percent cells with cytokine expression; gray filled curves and black lines (c) indicate unstimulated cells and stimulated cells, respectively. (d) Expression of mRNA (vertical axes) in GM-DCs left unstimulated or stimulated for 3 h with LPS with or without 2-DG (key), normalized to that of mRNA encoding β-actin and presented relative to that of unstimulated cells, set as 1 (10⁰). (e) CCR7 expression by Nos2^−/− GM-DCs left unstimulated or stimulated for 16 h with LPS with or without 2-DG (key); numbers above lines indicate percent cells with CCR7 expression (right of vertical line). Isotype, isotype-matched control antibody. (f) Migration of Nos2^−/− GM-DCs, stimulated for 16 h with LPS with or without 2-DG (horizontal axes), toward CCL19 or no cytokine (−) in a Transwell chamber, assessed after 2 h of migration and presented relative to total input, set as 100%. *P < 0.05 (Student’s t-test). (g) Dilution of the cytosolic dye CFSE in CFSE-labeled OT-I or OT-II T cells (which have transgenic expression of an OVA-specific T cell antigen receptor) cultured for 3 d together with Nos2^−/− GM-DCs (that had been stimulated for 6 h with LPS and OVA with or without 2-DG; key), at a DC/T cell ratio of 1:50 (left) or at various DC/T cell ratios (horizontal axes; right). (h) Expression of IFN-γ, IL-4, IL-17A and Foxp3 in OT-II T cells cultured for 6 d together with GM-DCs treated as in g (DC/T cell, 1:5). Numbers adjacent to outlined areas indicate percent cells with cytokine expression. Data are from one experiment representative of three (a,e–h) or five (b,c) experiments (mean and s.e.m. of triplicates (a,f) or duplicates (g)) or are from three independent experiments (d; mean and s.e.m.).

Figure 3

TLR-induced activation of DCs depends on the flux of glucose-derived carbon into the TCA cycle. (a) Change in mitochondrial membrane potential (ΔΨ_m) of GM-DCs left unstimulated or stimulated for various times (horizontal axis) with LPS, then stained with the mitochondria-selective dye DiOC₆; results are presented relative to the potential at time 0, set as 1. (b) Real-time changes in the OCR of GM-DCs stimulated for 1 h with control medium (−) or LPS, assessed during subsequent sequential treatment with oligomycin (Oligo; inhibitor of ATP synthase), FCCP (ionophore) and rotenone plus antimycin A (Ant-Rot; inhibitors of the electron-transport chain). SRC, spare respiratory capacity (double-headed arrow). (c) 1,2-¹³C-glucose tracing (left and right; as in Fig. 1d) in GM-DCs incubated for 1 h with medium (−) or LPS; middle, TCA cycle. (d) Spare respiratory capacity (as in b) of GM-DCs either left untreated (−) or given no pretreatment (LPS) or pretreated with 2-DG (2-DG + LPS) and then stimulated with LPS. *P < 0.001 (Student’s t-test). (e) Expression of Mpc1 mRNA in GM-DCs transduced with retroviral vector expressing control shRNA (Ctrl-hp) or Mpc1-specific shRNA (Mpc1-hp), normalized to that of mRNA encoding β-actin and presented relative to that of cells transduced with control shRNA, set as 1. (f,g) Expression of CD86 and CD40 (f) or IL-6, IL-12 and TNF (g) in GM-DCs transduced as in e and then left untreated (−) or treated with LPS. Numbers in top right corners (f) indicate percent CD86⁺CD40⁺ cells; numbers above bracketed lines (g) indicate percent cells with cytokine expression (as in Fig. 2b,c). Data are from one experiment representative of three (a,b,f,g) or two (d) experiments (mean and s.e.m. of duplicates (a) or triplicates (b,d)) or are from three (c) or two (e) independent experiments (mean and s.e.m.).

Figure 4

TLR-induced activation of DCs requires de novo synthesis of fatty acids supported by glycolysis. (a) Abundance of metabolites in DCs stimulated for 1 h with LPS, relative to that in untreated control cells, set as 1 (dotted line). (b) Confocal microscopy (left) of Nos2^−/− GM-DCs left unstimulated (−) or stimulated for 24 h with LPS, then stained with the DNA-intercalating dye DAPI and the fluorescent dye BODIPY. Scale bars, 10 μm. Right, quantification of BODIPY staining, assessed by flow cytometry and presented as the geometric mean fluorescence intensity (geoMFI). (c) Uptake of BODIPY by Nos2^−/− GM-DCs left unstimulated or stimulated for 24 h with LPS with or without 2-DG or C75 (key). (d) Incorporation of ¹⁴C into the lipid fraction of GM-DCs left unstimulated or stimulated for 6 h with LPS with or without C75, in the presence of U-¹⁴C-glucose (glc). (e,f) Expression of Slc25a1 mRNA (e) and of the cytokines IL-6, IL-12 and TNF (f) in GM-DCs transduced with control shRNA (Ctrl-hp) or Slc25a1-specific shRNA (Slc25a1-hp); mRNA results (e) are as above (Fig. 3e), and numbers above bracketed lines (f) indicate percent cells with cytokine expression. (g,h) Expression of CD86 and CD40 (g) or of IL-6, IL-12 and TNF (h) in GM-DCs left untreated (far left, g) or given no pretreatment (LPS) or pretreated with TOFA (TOFA + LPS) or C75 (C75 + LPS) and then stimulated with LPS (numbers in plots, as in Fig. 3f,g). (i) Expression of mRNA (vertical axes) in GM-DCs left unstimulated (−) or stimulated for 3 h with LPS with or without C75 (presented as in Fig. 2d). (j,k) Quantification of ER and Golgi (j) and electron microscopy (k) of GM-DCs treated as in i. Scale bars (k), 500 nm. *P < 0.05, **P < 0.01 and ***P < 0.001 (Student’s t-test). Data are from three experiments (a), five (b) or three (e,i) independent experiments (mean and s.e.m.), or one experiment representative of three (c,f), two (d,j,k) or four (g,h) experiments (mean and s.e.m. of triplicates (c,d) or 20 cells (j)).

Figure 5

The PPP supports the accumulation of lipids in DCs stimulated via TLRs. (a) NADPH and NADP⁺ in GM-DCs left unstimulated (−) or stimulated for 1 h with LPS. (b) Quantification of ribose-5-phosphate (R5P) in GM-DCs stimulated for 0–3 h with LPS (horizontal axis), based on total metabolite profiling. (c–e) Expression of G6pdx (which encodes glucose-6-phosphate dehydrogenase) (c) and of the cytokines IL-6, IL-12 and TNF (d), as well as BODIPY uptake (as in Fig. 4b) (e), in GM-DCs transduced with control shRNA (Ctrl-hp) or G6pdx-specific shRNA (G6pdx-hp); mRNA results (e) are as above (Fig. 3e), and numbers above bracketed lines (e) indicate percent cells with cytokine expression. *P < 0.05 (paired Student’s t-test). Data are from one experiment representative of three (a,d; mean and s.e.m. of duplicates in a) or are from four (b), three (c) or two (e) independent experiments (mean and s.e.m.).

Figure 6

The TLR-driven glycolytic burst is dependent on TBK1-IKKε and Akt. (a) Immunoblot analysis of total Akt and of Akt phosphorylated at Thr308 (p-Akt(T308)) or Ser473 (p-Akt(S473)), and of total and phosphorylated (p-) PRAS40, in GM-DCs stimulated for 0, 15 or 30 min (above lanes) with LPS. (b) Real-time changes in the ECAR of GM-DCs left untreated (−) or given no pretreatment (LPS) or pretreated with triciribine (Tric + LPS) and then stimulated with LPS (analyzed as in Fig. 1f). (c,d) Expression of CD86 and CD40 (c) or of IL-6, IL-12 and TNF (d) in GM-DCs treated as in b (numbers in plots, as in Fig. 3f,g). (e) Immunoblot analysis of total and phosphorylated Akt (as in a), PRAS40 and S6 in GM-DCs either left untreated (far left) or given no pretreatment (second from left) or preincubated with BX795 (BX), KU0063794 (KU) or LY294002 (LY) and stimulated for 15 min with LPS. (f) Real-time changes in the ECAR of GM-DCs treated as in e (analyzed as in Fig. 1a); Inh, inhibitor (key). (g) Immunoblot analysis of total and phosphorylated TBK1 in GM-DCs stimulated for 0, 15 or 30 min (above lanes) with LPS. (h) Immunoblot analysis of TBK1 and IKKε in wild-type (WT) and Ikbke^−/− GM-DCs transduced with control shRNA (Ctrl-hp) or Tbk1-specific shRNA (Tbk1-hp) and then sorted on the basis of human CD8 expression (as a reporter for transduction); β-actin serves as a loading control. (i) Real-time changes in the ECAR of GM-DCs transduced and sorted as in h and then left untreated (−) or treated with LPS (analyzed as in Fig. 1a). Data are from one experiment representative of two (a,g,i), four (b,f) or three (c,d,e) experiments (mean ± s.e.m. of triplicates in b,f,i) or are from one experiment (h).

Figure 7

The TLR-induced glycolytic burst and activation of DCs are dependent on the Akt-driven association of HK-II with mitochondria. (a) Immunoblot analysis of HK-II, β-actin (loading control), prohibitin-1 (PHB1; specific marker for mitochondria) and phosphorylated hexokinase (p-HK) in mitochondria-enriched (Mito) and cytosolic (Cyto) fractions of GM-DCs left unstimulated (0) or stimulated for 1 h with LPS (1) in the presence (+) or absence (−) of triciribine; hexokinase phosphorylation was assessed with an antibody to Akt-specific phosphorylation sites. (b) Hexokinase (HK) activity in GM-DCs left unstimulated (−) or stimulated for 1 h with LPS in the presence or absence of triciribine or BX795 (key). (c) Immunoblot analysis of HK-II and prohibitin-1 in mitochondria-enriched fractions of GM-DCs treated with control peptide (Ctrl pep) or HK-II peptide (HK-II pep; for targeted dissociation of HK-II from the mitochondria). (d) Hexokinase activity in GM-DCs left unstimulated or stimulated for 1 h with LPS in the presence of control or HK-II peptide (key). (e) Real-time changes in the ECAR of GM-DCs pretreated with control or HK-II peptide and then left unstimulated or stimulated with LPs (analyzed as in Fig. 1a). (f,g) Expression of CD86 and CD40 (f) or of IL-6, IL-12 and TNF (g) in GM-DCs pretreated with control or HK-II peptide and then stimulated with LPS (numbers in plots, as in Fig. 3f,g). Data are from one experiment representative of three (a,e,f,g) or two (c) experiments (mean ± s.e.m. of triplicates in e) or are from three independent experiments (b,d; mean and s.e.m.).

Figure 8

DCs rely on a rapid increase in glycolytic metabolism and fatty-acid synthesis for their activation and function in vivo. (a) Uptake of 2-NBDG in splenic DCs obtained from mice given no injection (−) or 1 h after injection of LPS (LPS + 2-NBDG) or vehicle control (2-NBDG), analyzed by flow cytometry. (b) Real-time changes in the ECAR of naive splenic DCs left unstimulated or stimulated with LPS (key) and analyzed ex vivo (as in Fig. 1a). (c–e) Real-time changes in the ECAR of naive CD11b⁺ DCs either left untreated (−) or given no pretreatment or pretreated with triciribine (c), BX795, KU0063794 or LY294002 (d) or control or HK-II peptide (e) and then stimulated with LPS (analyzed as in Fig. 1a). (f) Expression of CD40 and CD86 in splenic DCs obtained from mice 6 h after no injection (−) or intravenous injection of LPS with or without 2-DG, analyzed by flow cytometry; results are presented relative to those of untreated mice, set as 1 (each symbol represents an experiment). (g) IL-12 expression in splenic DCs left unstimulated or stimulated ex vivo for 12 h with LPS or without 2-DG. Numbers adjacent to outlined areas indicate percent IL-12⁺CD11c⁺ cells. (h) Dilution of the proliferation-tracking dye CellTrace Violet (CTV) in T cells in spleen of host mice given congenically marked, CellTrace Violet–labeled OT-I or OT-II T cells and then, 1 d later, given splenic DCs activated in vivo with LPS with or without 2-DG (key) and loaded with OT-I and OT-II OVA peptide ex vivo, assessed by flow cytometry 60 h after transfer of DCs (left), and quantification of the results at left, presented as division index (right). (i,j) Flow cytometry (i) and total number (j) of OT-I or OT-II T cells in the draining lymph nodes of host mice 7 d after subcutaneous injection of GM-DCs stimulated with LPS and pulsed with OVA in vitro with or without C75, assessing the endogenous OVA-specific CD8⁺ T cell response. Numbers adjacent to outlined areas (i) indicate percent CD44⁺ cells positive for staining with the H-2K^b–OVA tetramer. Each symbol (j) represents an individual mouse; small horizontal lines indicate the mean. *P < 0.05 and **P < 0.01 (Student’s t-test (a,h,j) or paired Student’s t-test (f)). Data are from one experiment representative of three (a–e,g) or two (h–j) experiments (mean and s.e.m. of three to four mice per group (a,h–j) or triplicates (b–e)) or four experiments with three mice per group (f).