Mycobacterium tuberculosis Limits Host Glycolysis and IL-1β by Restriction of PFK-M via MicroRNA-21

Abstract

Increased glycolytic metabolism recently emerged as an essential process driving host defense against Mycobacterium tuberculosis (Mtb), but little is known about how this process is regulated during infection. Here, we observe repression of host glycolysis in Mtb-infected macrophages, which is dependent on sustained upregulation of anti-inflammatory microRNA-21 (miR-21) by proliferating mycobacteria. The dampening of glycolysis by miR-21 is mediated through targeting of phosphofructokinase muscle (PFK-M) isoform at the committed step of glycolysis, which facilitates bacterial growth by limiting pro-inflammatory mediators, chiefly interleukin-1β (IL-1β). Unlike other glycolytic genes, PFK-M expression and activity is repressed during Mtb infection through miR-21-mediated regulation, while other less-active isoenzymes dominate. Notably, interferon-γ (IFN-γ), which drives Mtb host defense, inhibits miR-21 expression, forcing an isoenzyme switch in the PFK complex, augmenting PFK-M expression and macrophage glycolysis. These findings place the targeting of PFK-M by miR-21 as a key node controlling macrophage immunometabolic function.

Keywords: glycolysis; interferon gamma; interleukin-1b; macrophage; metabolic reprogramming; miR-21; microRNA; mycobacterium tuberculosis; phosphofructokinase; tuberculosis.

Figure 1.

Mtb infection attenuates macrophage metabolic reprogramming and drives miR-21. A) Extracellular lactate from bone-marrow derived macrophages (BMDM) infected with live Mycobacterium tuberculosis H37Ra (Mtb) or treated with heat-killed Mtb H37Ra (hk-Mtb) at a multiplicity of infection (MOI) of 5 bacteria/cell alongside LPS treatment (100 ng/mL) for indicated times. B) qPCR of indicated genes in BMDM treated as in A) for 24h. C) Extracellular lactate from BMDM infected with viable Mtb H37Ra or H37Rv strains at indicated MOI or treated with hk-Mtb H37Ra or γ-irradiated Mtb H37Rv (MOI 5) for 24h. D) Extracellular lactate from mature human monocyte-derived macrophages (hMDM) infected with Mtb as in A). E) qPCR analysis of the indicated genes in human alveolar macrophages (hAM) treated with viable Mtb H37Rv (Mtb) or γ-irradiated form (γMtb) at MOI of 2 or LPS (100 ng/mL) for 24-48 h. F-H) qPCR of mature miR-21 or primary miR-21 transcript (pri-miR-21) in BMDM (F,H) or hMDM (G) treated with hk-Mtb or infected with viable Mtb H37Ra (Mtb) at the indicated MOI for 24h (F-G) or at MOI of 5 for various times (0-72h – H). I) qPCR of mature miR-21 in BMDM treated as in C) for 72h. J) qPCR of the indicated RNA species in hAM treated as in E). K) qPCR of pri-miR-21 expression in lung tissue from mice infected with Mtb (Erdmann strain) for the indicated time (days). Data is mean concentration ± sem for n=3 independent experiments (A,C-D,) mean fold-induction over uninfected cells ± sem for n=3 (B, G, I) or n=5 (F,H) independent experiments, n=4 donors (E,J), or n=5 mice (K). * P < 0.05, n.s. P> 0.05 for the indicated group comparisons (ANOVA with post-hoc Tukey tests).

Figure 2.

Restraint of glycolysis by miR-21 permits Mtb growth in macrophages. A-C) Extracellular lactate from wild-type (WT) or miR-21-deficient (miR-21^−/−) BMDM infected with Mtb H37Ra (A) or Mtb H37Rv (B) (MOI 5) for indicated times or hMDM transfected with 50 nM miR-21-specific antisense (Anti-miR-21) or a control non-specific antisense (Anti-miR-Ctrl) prior to infection with Mtb H37Ra (MOI 5, indicated times) or treated with γMtb (MOI 5) or LPS (100 ng/mL) for 24 h (C). D) qPCR of indicated genes in BMDM infected with Mtb H37Ra at the indicated MOI or treated with LPS (100 ng/mL) for 24 h. E) Metabolic flux analysis of basal metabolism in untreated or γMtb-treated BMDM (indicated genotype, MOI 5, 24h) by measurement of extracellular acidification rate (ECAR) and oxygen consumption rate (OCR). F) Glycolytic Reserve (GR) of untreated or γMtb-treated BMDM (genotype indicated) treated with the indicated metabolic manipulations (Glu – addition of glucose, OM – oligomycin (1 μM), FCCP (1 μM) and Rotenone + antimycin A (0.5 μM)). G) ECAR:OCR ratio in BMDM treated with γMtb (MOI 5) or LPS (100 ng/mL) 24h. H) GR and Spare Respiratory Capacity (SRC) of resting hMDM transfected as in C) and treated with inhibitors as in G). I) Bacterial colony forming units (CFU) in BMDM infected with Mtb H37Ra (MOI 5) for the indicated times. J) qPCR of bacterial 1S6110 gDNA relative to 16s rRNA in BMDM infected with Mtb H37Rv (MOI 5) for the indicated times. K) Relative expression of indicated mediators in BMDM infected with Mtb H37Ra (Mtb, MOI 5, 24h). L) Bacterial CFU in BMDM treated with 2DG (10 mM) and subsequently infected with Mtb H37Ra (MOI 5) as before. M) Nitrite species in supernatants from BMDM pre-treated with 2DG (10 mM) and subsequently treated with γMtb (MOI 5) or LPS (100 ng/mL) for 24 h. Data is mean concentration ± sem (A-C, M), mean fold-induction over uninfected cells ± sem (D,J), mean fold-change ± sem (F-H, K-L) or mean CFU ± sem (I) for n=3 independent experiments. Energy plot (E) presents mean values and H is mean ± sem from n=5 independent experiments. * P < 0.05, n.s. P> 0.05 for the indicated group comparisons (ANOVA with post-hoc Tukey tests or Student t-test (H).

Figure 3.

Control of IL-1β production by miR-21 regulates anti-Mtb responses. A-B) IL-1β ELISA in supernatants from BMDM (genotype indicated) infected with Mtb H37Ra (A, left) or H37Rv (A, right) or hMDM transfected with 50 nM miR-21-specific antisense (Anti-miR-21) or a control non-specific antisense (Anti-miR-Ctrl) and subsequently infected with Mtb H37Ra (B) (MOI 5, indicated times. C) IL-1β ELISA in supernatants from BMDM transfected with 50 nM of miR-21-specific mimic (miR-21 mimic) or a control non-specific molecule (Ctrl-miR) infected with Mtb H37Ra (Mtb, MOI 5) for 24h. D) Bacterial growth in BMDM treated with anti-IL-1β specific antibody (αIL-1β, 1 μg/mL) or control IgG 3h post-infection with Mtb H37Ra as in A left). E) qPCR of Il1b mRNA in BMDM (genotype indicated) infected with Mtb H37Ra or H37Rv as before (A). F) qPCR of Il1b mRNA in BMDM pre-treated with 2DG (10 mM) infected with Mtb H37Ra (MOI 5) or treated with LPS (100 ng/mL) for 24 h. G) IL-1β ELISA in supernatants from hMDM transfected as in B), 24h prior to infection with Mtb H37Ra (MOI 5, 24h) in glucose-containing media (GLU) or glucose-free galactose-containing media (GAL). H) ELISA of indicated cytokines in lavage fluid taken from mice (genotype indicated) 24 h post-intraperitoneal injection of heat-killed Mtb (hk-Mtb, 750 μg per mouse) or PBS. Data (A-C, G-H) is mean concentration, fold-change in baseline CFU (3h) (D) and fold-induction over uninfected cells (E-F) and represents mean ± sem for n=4 (A, left panel) or n=3 (A right panel, B-G) independent experiments and for n=3 mice per group (H). * P < 0.05, n.s. P> 0.05 for the indicated group comparisons (ANOVA with post-hoc Tukey tests).

Figure 4.

miR-21 targets the expression of the glycolytic enzyme PFK-m. A) Schematic of PFK mRNAs indicating the miR-21 site & mutant constructs. B) 3’UTR luciferase reporter activity of PFK-m 3’UTR (WT 3’UTR) or mutant construct (Mutant 3’UTR) following co-transfection of HEK293T cells (left) or RAW267.4 cells (right) with control plasmid (−) or increasing amounts of miR-21 overexpressing plasmid (pCMV-miR-21, (+) 25ng per point, (++) 50ng per point). C-D) qPCR of PFK genes in BMDM infected with Mtb H37Ra at the indicated MOI or treated with LPS (100 ng/mL) for 24 h (C) or Mtb H37Rv (MOI 5, 24 h) (D). E) Relative expression of the indicated RNA species in RAW267.4 cells following 48 h transfection with increasing amounts of miR-21 overexpressing plasmid (pCMV-miR-21, 0-75ng per point). F) PFK-m immunoblot in RAW267.4 macrophages transfected with increasing amounts pCMV-miR-21 as in E (top panel) or BMDM (genotype indicated) before or after infection with Mtb H37Ra (MOI 5, 24h, bottom panel) alongside β-actin expression (loading control). G) In-vitro enzyme activity for phosphofructokinase (PFK) or hexokinase (HK) in BMDM lysates treated as in F. H-J) In-vitro PFK (H,J) or HK (I) enzyme activity in the indicated tissues from WT & miR-21^−/− mice (20 mg tissue). Data is % inhibition relative to control transfected cells (A), fold-induction relative to uninfected or control transfected cells (C-E) or mean concentration with individual mouse values indicated (G-J). Blots are 1 experiment representative of n=3 independent experiments with normalized PFK-m band intensity provided underneath (F). Data is mean ± sem for n=5 (B-C, E, G) or n=3 (D) independent experiments or n=6 (J-I) or n=3 (J) mice per group (H-J). * P < 0.05, n.s. P> 0.05 for the indicated group comparisons (ANOVA with post-hoc Tukey tests).

Figure 5.

Targeting of PFK-m by miR-21 controls Mtb responses. A) PFK-m mmunoblot in BMDM transfected with increasing amounts of Pfk-m specific siRNA (25 nM (+), 50 nM (++)) or control non-targeting siRNA (−), treated with hk-Mtb (MOI 5, 24h) alongside β-actin expression (loading control). B) qPCR of Pfk-m mRNA in BMDM (genotype indicated) transfected with 50 nM Pfk-m specific (si-Pfk-m) or control non-targeting (si-Ctrl) siRNA and subsequently infected with Mtb H37Ra (MOI 5, 24h). C) Bacterial growth in BMDM treated as B) upto 72h. D) Target protection schematic showing PFK-m 3’UTR and miR-21/PFK-m specific morpholino (m21/PFKm TP – top) and PFK-m immunoblot of BMDM transfected with 10 μM control non-targeting morpholino (Ctrl TP) or m21/PFK-m TP or empty transfected cells (−) and subsequently infected with Mtb H37Ra (MOI 5, 24h) or β-actin expression (loading control). E-F) Lactate production (E) or IL-1β ELISA (F) in supernatants from BMDM transfected with 10 μM non-targeting morpholino (Ctrl TP) or m21/PFKm TP or empty transfected cells (no Txt) and infected with Mtb H37Ra (Mtb, MOI 5, 24h). G) Bacterial CFUs of BMDM treated as in E) upto 72h. Blot is 1 experiment representative of n=2 (A) or n=3 independent experiments (D). mRNA data is fold-induction relative to uninfected si-Ctrl transfected WT BMDM (B). Bacterial growth is presented as fold-change in baseline CFU (3h) (C) or CFU values at 72h. Lactate and ELISA data is mean concentration (E-F). All data is mean ± sem of n=3 independent experiments. * P < 0.05, n.s. P> 0.05 for the indicated group comparisons (ANOVA with post-hoc Tukey tests).

Figure 6.

miR-21 limits alveolar macrophage responses to Mtb infection. A) qPCR of mature miR-21 in resting BMDM or mouse alveolar macrophages (mAM). B) GR and SRC of resting BMDM or mAM (genotype indicated). C) Extracellular lactate from mAM (genotype indicated) treated with γ-irradiated Mtb (γMtb) or infected with Mtb H37Ra (Mtb, MOI 5, 24h). D) qPCR of indicated genes in mAM (genotype indicated) infected with Mtb H37Ra (MOI 5, 24h). E) IL-1β ELISA in supernatants from mAM treated as in D). Data in A) is normalized ΔCt values across cell types, B) is fold change in indicated parameter relative to WT mAM, C) is mean fold-change in Lactate production relative to uninfected WT mAM, D) is fold-induction relative to uninfected WT mAM and E) is mean concentration. All experiments are mean ± sem for n=3 mice per group. * and # P < 0.05 for the indicated group comparisons (ANOVA with post-hoc Tukey tests (C-E) or Student t-test (A-B).

Figure 7.

Activation of macrophages by IFNγ targets glycolysis via miR-21. A) ECAR:OCR ratio in BMDM (genotype indicated) primed with IFNγ (5 ng/mL) before treatment with LPS (100 ng/mL). B) Extracellular lactate from BMDM (genotype indicated) primed with IFNγ (5 ng/mL) before treatment with LPS (100 ng/mL) or infection with Mtb H37Ra (MOI 5, 24h). C) IL-1β ELISA in supernatants from hMDM transfected with 50 nM miR-21-specific antisense (Anti-miR-21) or a control non-specific antisense (Anti-miR-Ctrl) 24h prior to treatment with IFNγ (5 ng/mL) and/or infection with Mtb H37Ra (MOI 5, 24h) in glucose-containing media (GLU) or glucose-free galactose-containing media (GAL). D) qPCR of indicated genes in BMDM (genotype indicated) primed with IFNγ (5 ng/mL) before infection with Mtb H37Ra (MOI 5, 24h). E) qPCR of indicated RNA species in BMDM primed with IFNγ (5 ng/mL) before treatment with LPS (100 ng/mL) or infection with Mtb H37Ra (MOI 5) for 24 h. F) PFK-m immunoblot of BMDM treated as in E). G) In-vitro enzyme activity for phosphofructokinase (PFK) or hexokinase (HK) in lysates from BMDM treated as in D. H) Bacterial growth in BMDM (genotype indicated) primed with IFNγ (5 ng/mL) before infection with Mtb H37Ra (MOI 5, 72h), with treatment with 10 mM 2DG or vehicle 3h post-infection. Data is mean fold-change over uninfected (A, D-E, H) or mean concentration (B-C, G). Data is mean ± sem of n=3 independent experiments (A-E, H), 1 blot representative of n=3 (F) or mean activity ± sem for n=6 (PFK) or n=5 (HK) independent experiments (G). * and # P < 0.05 for the indicated group comparisons (ANOVA with post-hoc Tukey tests).