Phosphorylation of PFKL regulates metabolic reprogramming in macrophages following pattern recognition receptor activation

Abstract

Innate immune responses are linked to key metabolic pathways, yet the proximal signaling events that connect these systems remain poorly understood. Here we show that phosphofructokinase 1, liver type (PFKL), a rate-limiting enzyme of glycolysis, is phosphorylated at Ser775 in macrophages following several innate stimuli. This phosphorylation increases the catalytic activity of PFKL, as shown by biochemical assays and glycolysis monitoring in cells expressing phosphorylation-defective PFKL variants. Using a genetic mouse model in which PFKL Ser775 phosphorylation cannot take place, we observe that upon activation, glycolysis in macrophages is lower than in the same cell population of wild-type animals. Consistent with their higher glycolytic activity, wild-type cells have higher levels of HIF1α and IL-1β than Pfkl^S775A/S775A after LPS treatment. In an in vivo inflammation model, Pfkl^S775A/S775A mice show reduced levels of MCP-1 and IL-1β. Our study thus identifies a molecular link between innate immune activation and early induction of glycolysis.

Fig. 1. TLR activation induces PFKL Ser775 phosphorylation in primary macrophages.

a mBMDM were stimulated with the indicated stimuli for 1 h and ECAR was then measured over time in response to the indicated compounds. Glucose (10 mM); Oligomycin (1.5 µM); 2DG (50 mM). Measurement of one representative experiment is shown. Data are presented as mean ± SEM (∅, n = 6; other stimuli, n = 8 technical replicates). b PFKL was identified as one of the top 10 proteins with increased phosphorylation after 30 min of LPS treatment in immortalized mBMDM. c PFKL catalyzes the conversion of F6P to F1,6BP in the glycolytic pathway. d Normalized transcript counts of PFK1 isoforms in immune cells. mBMDM (e) and hMDM (f) were stimulated with TLR agonists for the indicated time periods. PFKL Ser775 phosphorylation and PFKL were analyzed by western blot. g mBMDM were primed overnight with 20 ng/ml mouse IFNγ and then stimulated with zymosan (100 µg/ml) for the indicated time periods (left panel). mBMDM were stimulated with mouse TNF (50 ng/ml) for the indicated time periods (right panel). R848 (TLR7, 1 µg/ml), LPS (TLR4, 200 ng/ml), Pam₃CSK₄ (TLR1/2, 500 ng/ml), poly(I:C) (TLR3, 20 µg/ml). β-actin was used as a loading control. a and f are representative of two independent experiments. e and g are representative of three independent experiments. Source data are provided as a Source Data file.

Fig. 2. The IKK complex and PKC function in PFKL Ser775 phosphorylation.

Immunoblot analysis of indicated proteins. mBMDM were pretreated with DMSO, Takinib (+, 50 µM; ++, 100 µM) (a), TPCA-1 ( +, 5 µM; ++, 10 µM) (b), Wortmannin (+, 10 µM; ++, 20 µM) (c), or MK2206 2HCl (+, 10 µM; ++, 20 µM) (d) for 1 h and then stimulated with R848 (1 µg/ml) for the indicated time periods. hMDM were pretreated with DMSO, Takinib (1, 5, 10 and 20 µM) (e), TPCA-1 (1, 5, 10 and 20 µM) (f), or Wortmannin (1, 5, 10 and 20 µM) (g) for 1 h and then stimulated with R848 (1 µg/ml) for 1 h. h, i mBMDM, primed with 20 ng/ml mouse IFNγ overnight, were pretreated with DMSO, TPCA-1 ( + , 1 µM; ++, 5 µM), or Takinib (+, 1 µM; ++, 5 µM) (h), or Go 6983 (0.1, 1, 5 and 10 µM) (i) for 1 h and then stimulated with zymosan (100 µg/ml) for 1 h. j PFKL Ser775 phosphorylation and total PFKL levels were analyzed using western blot after the in vitro kinase assay. a–j are representative of three independent experiments. Source data are provided as a Source Data file.

Fig. 3. PFKL Ser775 phosphorylation increases its catalytic activity.

a PFKL knockout in HEK293T cells was analyzed by western blot. β-actin was used as a loading control. b ECAR in indicated HEK293T cells was measured over time in response to indicated compounds. Glucose (10 mM); Oligomycin (1.5 µM); 2DG, 2-Deoxy-D-glucose (50 mM). Measurement of one representative experiment is shown. Data are presented as mean ± SEM (WT, n = 5; PFKL^−/−, n = 6 technical replicates). c PFKL knockout reconstituted with either PFKL^WT or PFKL^S775A was analyzed by western blot. β-actin was used as a loading control. d ECAR in indicated HEK293T cells was measured over time in response to indicated compounds. Measurement of one representative experiment is shown. Data are presented as mean ± SEM (WT, n = 6; other genotypes, n = 8 technical replicates). e ECAR in different stages of the measurement from d are presented as mean ± SEM (n = 3), statistics indicate two-way ANOVA with Dunnett’s correction for multiple testing. f The enzymatic activity of purified PFKL^WT and PFKL^S775A was measured by performing in vitro enzymatic assays and represented by the initial reaction rate. Data are presented as mean ± SEM (n = 3), statistics indicate unpaired two-tailed student’s t-test. g Workflow for analyzing the Ser775 phosphorylation ratio of PFKL^WT was illustrated. IP immunoprecipitated samples, FT flow through samples. h PFKL from the indicated fractions was analyzed by immunoblot (top panel), and the phosphorylation ratio at Ser775 of PFKL^WT was analyzed by measuring PFKL band intensities in FT using ImageJ software (bottom panel). Data are presented as mean ± SEM (n = 3). Immunoblot results in a, c, h are representative of three independent experiments. Source data are provided as a Source Data file.

Fig. 4. Characterization of Pfkl^S775A/S775A mice.

a The S775A mutation in Pfkl^S775A/S775A mice was confirmed by Sanger sequencing. Body weight (b), spleen weight (c), and the ratio of spleen weight to body weight (d) of 10-week-old WT and Pfkl^S775A/S775A mice. Data are presented as mean ± SEM (WT mice, n = 8; Pfkl^S775A/S775A mice, n = 9), statistics indicate unpaired two-tailed student’s t-test. e CD11b⁺Ly6G⁺ neutrophils from the bone marrow of WT and Pfkl^S775A/S775A mice were analyzed by flow cytometry. Numbers next to the outlined areas indicate the percentage among living cells. f Percentages of CD11b⁺Ly6G⁺ neutrophils from e are summarized and presented as mean ± SEM (n = 6), statistics indicate unpaired two-tailed student’s t-test. g CD11b⁺Ly6G⁺ neutrophils from the spleen of WT and Pfkl^S775A/S775A mice were analyzed by flow cytometry. Numbers next to the outlined areas indicate the percentage among living cells. h Percentages of CD11b⁺Ly6G⁺ neutrophils from g are summarized and presented as mean ± SEM (WT mice, n = 8; Pfkl^S775A/S775A mice, n = 9), statistics indicate unpaired two-tailed student’s t-test. WT and Pfkl^S775A/S775A mice were injected with NaCl or murine TNF intraperitoneally. The levels of serum MCP-1 (i) and IL-1β (j) in TNF-injected WT and Pfkl^S775A/S775A mice. Data are presented as mean ± SEM (i n = 12; j n = 11 for WT + TNF, n = 12 for Pfkl^S775A/S775A + TNF), statistics indicate unpaired two-tailed student’s t-test. WBC (k) and the percentage of indicated cell types among WBC (l–o) are shown. Data are presented as mean ± SEM (k–n n = 6, n = 20, n = 22 for WT mock, WT + TNF, Pfkl^S775A/S775A + TNF; o n = 6, n = 16, n = 18 for WT mock, WT + TNF, Pfkl^S775A/S775A + TNF), statistics indicate one-way ANOVA with Šidák’s correction for multiple comparisons test. Source data are provided as a Source Data file.

Fig. 5. PFKL Ser775 phosphorylation contributes to TLR-induced glycolysis in primary macrophages.

a ECAR of WT and Pfkl^S775A/S775A mBMDM stimulated with LPS (200 ng/ml) for 1 h. Data are presented as mean ± SEM (∅, n = 5; LPS, n = 6 technical replicates). b ECAR of Pfkl^S775A/S775A relative to WT at different stages of the measurement from a are presented as mean ± SEM (n = 4), statistics indicate one sample two-tailed t test. c ECAR of WT and Pfkl^S775A/S775A mBMDM stimulated with LPS for 24 h. Data are presented as mean ± SEM (∅, n = 5; LPS, n = 6 technical replicates). d ECAR of Pfkl^S775A/S775A relative to WT at different stages of the measurement from c are presented as mean ± SEM (n = 5), statistics indicate one sample two-tailed t test. e ECAR of WT and Pfkl^S775A/S775A mBMDM stimulated with TNF (50 ng/ml) for 1 h. Data are presented as mean ± SEM (∅, n = 5; TNF, n = 6 technical replicates). f ECAR of Pfkl^S775A/S775A relative to WT stimulated with TNF for 1 h at different stages of the measurement from e are presented as mean ± SEM (n = 4), statistics indicate one sample two-tailed t test. g ECAR of WT and Pfkl^S775A/S775A mBMDM stimulated with TNF for 24 h. The data for unstimulated controls in e and g were from the same experiment. Data are presented as mean ± SEM (∅, n = 5; TNF, n = 6 technical replicates). h ECAR of Pfkl^S775A/S775A relative to WT stimulated with TNF for 24 h at different stages of the measurement from g are presented as mean ± SEM (n = 5), statistics indicate one sample two-tailed t test. Relative abundances of F6P m + 6 (i), F1,6BP m + 6 (j), DHAP m + 3 (k), and GA3P m + 3 (l). Data are presented as mean ± SEM (n = 3), statistics indicate unpaired two-tailed student’s t-test. For a, c, e, g, measurement of one representative experiment is shown. Source data are provided as a Source Data file.

Fig. 6. PFKL Ser775 phosphorylation is required for LPS-induced HIF1α and IL-1β production.

a mBMDM from WT and Pfkl^S775A/S775A mice were stimulated with LPS (200 ng/ml) for the indicated time periods. Indicated proteins were analyzed by western blot. β-actin was used as a loading control. b LPS-induced Il1b mRNA levels in WT and Pfkl^S775A/S775A mBMDM were measured by qPCR and normalized to Actb mRNA levels. Data are presented as mean ± SEM (n = 3), statistics indicate unpaired two-tailed student’s t-test. c mBMDM from WT and Pfkl^S775A/S775A mice were stimulated with LPS for the indicated time periods. Indicated proteins were analyzed by western blot (left panel). LPS-induced pro-IL-1β were quantified by normalization to β-actin at indicated time points (right panel). Data are presented as mean ± SEM (n = 3), statistics indicate unpaired two-tailed student’s t-test. d mBMDM from WT and Pfkl^S775A/S775A mice were stimulated with LPS for 24 h and Nos2 mRNA levels were measured by qPCR and normalized to Actb mRNA levels. Data are presented as mean ± SEM (n = 3), statistics indicate two-way ANOVA with Šidák’s correction for multiple comparisons test. e mBMDM from WT and Pfkl^S775A/S775A mice were pretreated with octyl-α-KG ( + , 0.5 mM; ++, 1 mM) or DMOG ( + , 0.25 mM; ++, 0.5 mM) for 2 h and then stimulated with LPS for 24 h. Indicated proteins were analyzed by western blot. β-actin was used as a loading control. Octyl-α-KG octyl-α-ketoglutarate, DMOG dimethyloxalylglycine, s.e. short exposure, l.e. long exposure. a, e are representative of three independent experiments. Source data are provided as a Source Data file.