Akt and STAT5 mediate naïve human CD4+ T-cell early metabolic response to TCR stimulation

Abstract

Metabolic pathways that regulate T-cell function show promise as therapeutic targets in diverse diseases. Here, we show that at rest cultured human effector memory and central memory CD4+ T-cells have elevated levels of glycolysis and oxidative phosphorylation (OXPHOS), in comparison to naïve T-cells. Despite having low resting metabolic rates, naive T-cells respond to TCR stimulation with robust and rapid increases in glycolysis and OXPHOS. This early metabolic switch requires Akt activity to support increased rates of glycolysis and STAT5 activity for amino acid biosynthesis and TCA cycle anaplerosis. Importantly, both STAT5 inhibition and disruption of TCA cycle anaplerosis are associated with reduced IL-2 production, demonstrating the functional importance of this early metabolic program. Our results define STAT5 as a key node in modulating the early metabolic program following activation in naive CD4+ T-cells and in turn provide greater understanding of how cellular metabolism shapes T-cell responses.

Fig. 1

Quiescent EM and CM T-cells are metabolically active. a Glycolytic stress profile of NV, EM and CM T-cells by measuring ECAR before and following injections of oligomycin (0.75 μM), FCCP (1 μM) and antimycin A and rotenone (1 μM) at the time points indicated. Basal (b) and maximal ECAR (c) in NV, EM and CM T-cells. d Representative immunoblots from two different donors per cell type for GLUT1, HKI HKII, GAPDH, PFKP, PKM2 and LDHA and β-actin. Respective densitometry normalised to β-actin is shown. Data are either representative of five independent experiments (a–c) or 3−4 experiments (d). Statistical analysis was performed using a non-matching one-way ANOVA with Tukey’s multiple comparison test (b–d). For non-parametric data, a Kruskal−Wallis test with Dunn’s multiple comparisons test was used. Data expressed as mean ± SEM; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Fig. 2

Activated CD4+ T-cell subsets incorporate glucose into the TCA cycle. a Oxidative phosphorylation profile of NV, EM and CM T- cells by measuring OCR before and following injections of oligomycin (0.75 μM), FCCP (1 μM) and antimycin A and rotenone (1 μM). Basal respiration (b), ATP-linked respiration (c), maximal respiration (d) and OCR/ECAR ratio (e) in NV, EM and CM T- cells. f Mitochondrial content of NV, EM and CM CD4+ T-cells measured using MitoTracker. g Representative images of NV, EM and CM T-cells stained with DRAQ5 (nucleus), cell mask orange (plasma membrane) and MitoTracker green (mitochondria) scale bar = 10 μm and h corresponding MitoTracker signal to area ratios. i Uniformly labelled ¹³C-glucose incorporation into T-cell metabolites via the TCA cycle in NV, EM and CM T-cells activated for 0.5 and 4 h. Relative abundance of ¹²C and ¹³C including citrate, α-ketoglutarate, succinate, fumarate and malate. j Relative abundance of ¹²C and ¹³C in non-essential amino acids glutamate and aspartate . Data are representative of either five independent experiments (a–e), four independent experiments (f), three experiments with <100 cells analysed as technical replicates (h) or three independent experiments (i, j). Statistical analysis was performed using a nonmatching one-way ANOVA with Tukey’s multiple comparison test (b–h). For non-parametric data, a Kruskal−Wallis test with Dunn’s multiple comparisons test was used. Data expressed as mean ± SEM; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Fig. 3

NV T-cells have a greater metabolic response upon activation. a ECAR and b OCR in NV, EM and CM T- cells upon stimulation with anti-CD3/CD28. c Fold ECAR change in NV, EM and CM T-cells upon activation. d NV, e EM and f CM ECAR ‘Pre’ and ‘Post’ activation taken from the measurements in the indicated boxes in Supplementary Fig. 3a. g Fold change in OCR in NV, EM and CM T-cells upon activation. h NV, i EM and j CM OCR ‘Pre’ and ‘Post’ activation taken from the measurements in the indicated boxes in Supplementary Fig. 3a. k ECAR/OCR ratio of ‘Pre’, ‘Peak’ and ‘Post’ activation kinetics of NV, EM and CM T-cells. Statistical analysis was performed using a non-matching one-way ANOVA with Tukey’s multiple comparison test (c, g, k). For non-parametric data, a Kruskal−Wallis test with Dunn’s multiple comparisons test was used. A one-sample t-test was also used comparing values to a theoretical value of 1 (c, g). A paired t-test was also used (d–f, h–j) whereby a Wilcoxon matched-pairs signed rank test was applied to non-parametric data. Data are representative of four experiments (a–k) and are expressed as mean ± SEM; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Fig. 4

Early T-cell activation requires Akt phosphorylation. a Immunoblot for pAkt Thr308 and Ser473 and β-actin in NV, EM and CM T-cells following 0, 15, 30 and 180 min of activation with anti-CD3/CD28. Densitometry of pAkt Thr308 (b) and pAkt Ser473 (c) for NV, EM and CM T-cells normalised to β-actin. ECAR (d) and OCR (e) in NV CD4+ T-cell upon treatment with either 10 μM Akt1/2 kinase inhibitor or vehicle control (DMSO) and anti-CD3/CD28 at the times indicated. Data are representative (a–c) of 3–5 experiments with one representative immunoblot sample of 3–5 is shown, (d, e) three experiments. Statistical analysis was performed using a nonmatching two-way ANOVA with Sidak’s multiple comparison test (b, c). Data are expressed as mean ± SEM; **p ≤ 0.01

Fig. 5

STAT5 orchestrates the metabolic switch in activated NV T-cells. a Immunoblot for pSTAT5 Tyr694 and β-actin in NV, EM and CM T-cells following 0, 15, 30 and 180 min of activation with anti-CD3/CD28. Densitometry of pSTAT5 Tyr694. b Immunoblot for pSTAT5 Tyr694 and β-actin in NV T-cells activated with anti-CD3/CD28 in the presence (1, 10 and 20 μM) or absence (Veh) of a lck inhibitor. OCR (c–e) and ECAR (f–h) was measured in NV (c, f), EM (d, g) and CM (e, h) CD4+ T-cells in the absence (DMSO) or presence of a STAT5 inhibitor (100 μM) before cells were activated with anti-CD3/CD28 at the indicated time points. i Fold change in ECAR upon activation calculated using the measures in the FC box. j Extracellular glucose and lactate production in activated NV T-cells (anti-CD3/CD28) in the absence or presence of STAT5i (100 μM) for 4 h. OCR (k) and ECAR (l) of NV T-cells activated with either IL-2 or IL-7 (10 ng/mL). Immunoblot of pSTAT5 Tyr694 and β-actin in NV T-cells activated with anti-CD3 (2 μg/mL) and anti-CD28 (20 μg/mL) with common γ chain antibody or isotype control (1 μg/mL) for m 0.5 or n 3 h. Statistical analysis was performed using a non-matching two-way ANOVA with Sidak’s multiple comparison test (a, i), a paired t- test (j) or a matched Friedman test with Dunn’s multiple comparisons test (m, n). Data are representative of a 3–5 experiments with one representative immunoblot sample of 3–5 is shown, five (b, c, e, f, h), three (d, g, n), four (j, m) or two independent experiments (k, l) and expressed as mean ± SEM; *p ≤ 0.05, **p ≤ 0.01

Fig. 6

STAT5 regulates glutaminolysis in an mTORC1-dependent manner. a Schematic summarising incorporation of uniformly labelled ¹³C-glutamine into the TCA cycle. NV T-cells were activated (anti-CD3/CD28) in the presence or absence of STAT5i (100 μM) for 4 h. b Relative abundance of glutamine-derived ¹²C and ¹³C TCA cycle metabolites, citrate, α-ketoglutarate, succinate, fumarate and malate. c Relative abundance of glutamine-derived ¹²C and ¹³C glutamine and glutamate. d Mass isotopologue distributions (MID) of TCA cycle intermediates, citrate, α-ketoglutarate, succinate, fumarate and malate. e Relative abundance of glutamine-derived ¹²C and ¹³C amino acids aspartate and proline. f Immunoblot for p70S6K and ribosomal S6 phosphorylation (pS6) in NV T-cells following 4 h activation with (anti-CD3/CD28) in the absence and presence of STAT5i (100 μM). β-actin was used as a loading control. Statistical analysis was performed using a paired t-test (f). Data are representative of four experiments (b–e) or three experiments (c) and are expressed as mean + SEM. **p ≤ 0.01, ***p ≤ 0.001

Fig. 7

Disrupted glutaminolysis causes compensatory TCA cycling. a NV T-cells were activated (anti-CD3/CD28) in the presence or absence of STAT5i (100 μM) for 4 h. Relative abundance of glucose-derived ¹²C and ¹³C TCA cycle metabolites citrate, α-ketoglutarate, succinate, fumarate and malate. b Mass isotopologue distributions (MID) of TCA cycle intermediates citrate, α-ketoglutarate, succinate, fumarate and malate. c Relative abundance of glucose-derived ¹²C and ¹³C amino acids aspartate and proline. Data are representative of four experiments and are expressed as mean + SEM

Fig. 8

Disruption glutaminolysis impacts on NV T-cell IL-2 production. IL-2 production of NV T-cells cultured with anti-CD3 (2 μg/mL) and anti-CD28 (20 μg/mL) in the presence or absence of a STAT5 inhibitor (STAT5i; 100 μM), b glutamine withdrawal (- Q). c Schematic of the different mechanisms used to investigate glutamine metabolism in human NV T-cells. IL-2 production of NV T-cells cultured with d DON (50 μM) or e aminooxyacetic acid (AOA; 0.25–1 mM). f NV T-cells activated and treated as in b with the addition of dimethyl 2-oxoglutarate (DMK; 0.3 mM) with downstream analysis of IL-2 production. NV T-cell cultured with g oligomycin (100 nM) and h BMS303141 (1 μM) with IL-2 production measured. i Schematic of NV T-cell activation via the T-cell receptor leading to downstream STAT5 phosphorylation. Glutaminolysis is regulated by STAT5 in an mTORC1-dependent manner. Statistical analysis was performed using an unpaired t-test (a, b, d, f, g), a Kruskal−Wallis test (e) or a one-way ANOVA (g). Data are representative of four experiments (a, d–e, h), five experiments (b) and three experiments (f) expressed as mean ± SEM; *p ≤ 0.05, **p ≤ 0.01