Ablation of Atp5if1 impairs metabolic reprogramming and proliferation of T lymphocytes and compromises mouse survival

Abstract

T cells experience metabolic reprogramming to an enhanced glycolysis upon activation. Herein, we have investigated whether ATPase Inhibitory Factor 1 (IF1), the physiological inhibitor of mitochondrial ATP synthase, participates in rewiring T cells to a particular metabolic phenotype. We show that the activation of naive CD4⁺ T lymphocytes both in vitro and in vivo is accompanied by a sharp upregulation of IF1, which is expressed only in Th1 effector cells. T lymphocytes of conditional CD4⁺-IF1-knockout mice display impaired glucose uptake and flux through glycolysis, reducing the biogenesis of mitochondria and cellular proliferation after activation. Consequently, mice devoid of IF1 in T lymphocytes cannot mount an effective Th1 response against bacterial infection compromising their survival. Overall, we show that the inhibition of a fraction of ATP synthase by IF1 regulates metabolic reprogramming and functionality of T cells, highlighting the essential role of IF1 in adaptive immune responses.

Keywords: Biological sciences; Immunology; Molecular biology; Physiology.

Graphical abstract

Figure 1

Expression of IF1 in T cells and development of IF1-KO mice in CD4⁺ lymphocytes (A) Representative blot of IF1 and β-F1-ATPase expression in mouse tissue extracts of heart, kidney, liver, lymph nodes and spleen, and in spleen of mice treated with tamoxifen (Tam) or dextran sodium sulfate (DSS). α-tubulin is shown as loading control. (B) Representative blot of IF1 and β-F1-ATPase expression in naive and in vitro activated CD4⁺ mouse lymphocytes. CD4 is shown as loading control. (C) Quantification of IF1 mRNA levels in naive and activated CD4⁺ lymphocytes (n = 4). (D) Extracellular acidification rates (ECAR) of naive and activated CD4⁺ lymphocytes (n = 3). B–D, In vitro activation of CD4⁺ lymphocytes was performed by incubating the cells with anti-CD3 and anti-CD28 during 48 h. (E) Quantification of mRNA levels of IFNγ, IL17, TGFβ and IF1 in CD4⁺ lymphocytes polarized to the Th1, Th17 or Treg subsets (n = 3–4). (F) Representative blot of IF1 and β-F1-ATPase expression in CD4⁺ lymphocytes polarized to the Th1, Th17 or Treg subsets. CD4 is shown as loading control. (G) Spleen weight of mice 10 days after immunization with PBS (−) or BSA/CFA (n = 4). (H) Representative plots of the percentage of IFNγ⁺ immune cells (CD45⁺) in the spleen of mice 10 days after PBS (−) or BSA/CFA (+) injection. (I) Representative blot of IF1 and β-F1-ATPase expression in CD4⁺ lymphocytes 10 days after PBS (−) or BSA/CFA (+) injection. CD4 is shown as loading control. (J) Schematic of the generation of the CD4⁺-IF1-KO mouse model. IF1-KO mice in CD4⁺ lymphocytes were obtained by breeding CD4-Cre with IF1-floxed mouse lines. (K) Histograms show the quantification of IF1 mRNA levels in CRL and IF1-KO CD4⁺ lymphocytes (n = 4). (L) Representative blot of IF1 and β-F1-ATPase expression in extracts of CD4⁺ cells, immune CD4⁻ cells, brain and kidney of CRL and CD4⁺-IF1-KO mice. The histograms show the mean and the error bars ±SEM. ∗∗p ≤ 0.01; ∗∗∗p ≤ 0.001 when compared to PBS (−) or CRL by Student’s t test. #p ≤ 0.05; ##p ≤ 0.01; ###p ≤ 0.001 when compared by one-way ANOVA test and the Tukey multiple comparison test. See also Figure S1.

Figure 2

The biogenesis of mitochondria is restrained in activated CD4⁺ IF1-KO lymphocytes (A and B) Profiles of oxygen consumption rates (OCR) of naive (a) (n = 3) and activated (b) (n = 3) murine CRL and IF1-KO CD4⁺ lymphocytes using glucose as respiratory substrate. The addition of oligomycin (OL), 2,4-dinitrophenol (DNP) and rotenone (Rot) plus antimycin A (Ant) is indicated. (B) Histogram shows the quantification of basal, oligomycin-sensitive (OSR) and maximal respiration rates of CRL and IF1-KO naive and activated CD4⁺ lymphocytes. (C) Mitochondrial membrane potential in CRL and IF1-KO activated CD4⁺ lymphocytes (n = 4). MFI, mean fluorescent intensity. (D) Mitochondrial reactive oxygen species (mtROS) production in CRL and IF1-KO activated CD4⁺ lymphocytes (n = 4). (E) Histogram shows the rate of ATP production per mg of cellular protein in digitonin-permeabilized CRL and IF1-KO activated CD4⁺ lymphocytes (n = 3). (F) Representative ATP hydrolytic activity of complex V in CN-PAGE gels of CRL and IF1-KO activated CD4⁺ lymphocytes. Where indicated, 2 μM oligomycin (OL) was added in the sample to inhibit ATP hydrolysis. (G) Representative images of Proximity Ligation Assays (PLA) using γ-F1-ATPase as target (green dots) in CRL and IF1-KO activated CD4⁺ lymphocytes. DAPI (blue) stained nuclei. Histograms show the number of PLA signals per cell (n = 3). (H) Heatmap shows normalized Z-scores of mRNA levels of genes encoding different mitochondrial proteins in naive and activated CRL and IF1-KO CD4⁺ lymphocytes (n = 3–4). #p ≤ 0.05; ##p ≤ 0.01; ###p ≤ 0.001 when compared naive CRL vs. activated CRL; $p ≤ 0.05; $$p ≤ 0.01; $$$p ≤ 0.001 when compared naive CRL vs. naive IF1-KO; λλp ≤0.01; when compared activated CRL vs. activated IF1-KO; Φp ≤ 0.05; ΦΦp ≤0.01 when compared naive IF1-KO vs. activated IF1-KO by one-way ANOVA and Tukey multiple comparison test. (I) Representative electron micrographs of naive (upper panels) and activated (lower panels) CRL and IF1-KO CD4⁺ lymphocytes. Violin plots show the number of mitochondria per cell (n = 17–36), the mitochondrial area per cytoplasmatic surface (n = 17–28), the mean mitochondrial area (n = 59–256) and the mean mitochondrial circularity (n = 59–256) assessed in micrographs. (J) mtDNA copy number analysis in naive and activated CRL or IF1-KO CD4⁺ lymphocytes (n = 3–4). In vitro activation of CD4⁺ lymphocytes was performed by incubating the cells with anti-CD3 and anti-CD28 during 48 h. The profiles and histograms show the mean and the error bars ±SEM. n.s., no significant. ∗p ≤ 0.05; ∗∗∗p ≤ 0.001 when compared to CRL or naive by Student’s t test.

Figure 3

IF1-KO lymphocytes have restrained rates of glycolysis and proliferation upon activation (A) Rates of glucose uptake in CRL and IF1-KO activated CD4⁺ lymphocytes (n = 4). (B) Rates of lactate production in CRL (n = 4) and IF1-KO (n = 3) activated CD4⁺ lymphocytes. (C) Proliferation rates of CRL and IF1-KO CD4⁺ lymphocytes after 24 h of activation (n = 6). (D) Percentage of dead (AnnexinV⁺/7AAD⁺) CRL and IF1-KO CD4⁺ lymphocytes after 48 h of activation (n = 4). (E) Determination of adenosine released into the culture medium of CRL (n = 9) and IF1-KO (n = 7) CD4⁺ lymphocytes after 48 h of activation. (F and G) Activities of phosphoribosyl pyrophosphate synthetase (PRPP synthetase) (f) (n = 5) and hypoxanthine-guanine phosphoribosyl transferase (HPRT) (n = 3) (k) in CRL and IF1-KO activated CD4⁺ lymphocytes. (H) Representative blot of the expression of adenine phosphoribosyl transferase (APRT) in CRL and IF1-KO activated CD4⁺ lymphocytes. α-tubulin is shown as loading control. Histogram shows the expression level relative to α-tubulin (n = 4). (I–K) Rates of glucose uptake (i), lactate production (j) and proliferation (k) in CRL CD4⁺ lymphocytes treated with the indicated doses of adenosine during activation (n = 3). In vitro activation of CD4⁺ lymphocytes was performed by incubating the cells with anti-CD3 and anti-CD28 during 48 h. The histograms show the mean and the error bars ±SEM. ∗p ≤ 0.05; ∗∗p ≤ 0.01; ∗∗∗p ≤ 0.001 when compared to CRL by Student’s t test. #p ≤ 0.05; ##p ≤ 0.01; ###p ≤ 0.001 when compared by one-way ANOVA and Tukey multiple comparison test.

Figure 4

Survival of CD4⁺-IF1-KO mice is compromised in response to bacterial infection due to a restrained adaptive immune response (A) Progressive weight change of CRL (n = 5) and IF1-KO (n = 4) mice treated with 2.5% dextran sodium sulfate (DSS) in the drinking water for nine days. (B) Kaplan–Meier survival analysis of CRL (n = 9) and IF1-KO mice (n = 9) after treatment with 2.5% DSS in the drinking water. The p-value of the log rank test is shown. (C) Number of CD3⁺ (CD3⁺;CD45⁺;DAPI⁻), CD4⁺ (CD4⁺;CD45⁺;DAPI⁻) and CD8⁺ (CD8⁺;CD45⁺;DAPI⁻) cells in the blood of CRL and IF1-KO mice after 6-day treatment with 2.5% DSS (n = 3–4). (D) Percentage of Th1 (INFγ⁺;CD4⁺;CD45⁺), Th17 (IL17⁺;CD4⁺;CD45⁺) and Treg (FoxP3⁺;CD4⁺;CD45⁺) cells in the blood of CRL and IF1-KO mice after 6-day treatment with 2.5% DSS (n = 3–4). (E) Multiplexed quantitative analysis of cytokines and chemokines. Histograms show the mean and the error bars ±SEM of cytokine levels in the spleen and colon of CRL and IF1-KO mice after 6-day treatment with 2.5% DSS (n = 3–4). ∗p ≤ 0.05; ∗∗p ≤ 0.01 when compared to CRL by Student’s t test. See also Table S1.