GLUT5 armouring enhances adoptive T-cell therapy anti-tumour activity under glucose-limiting conditions

Abstract

Background: Cancer immunotherapy with engineered T cells has become a standard treatment for certain haematological cancers. However, clinical trial outcomes for solid tumours are significantly lagging. A primary challenge in solid tumours is the lack of essential metabolites in the tumour microenvironment, such as glucose, due to poor vascularization and competition with tumour cells.

Methods: To address this, we modified T cells to use fructose as an alternative energy source by introducing ectopic GLUT5 expression.

Results: We show that "GLUT5-armored" T cells, engineered with either chimeric antigen receptors (CARs) or an ectopic T-cell receptor (TCR), achieve enhanced anti-tumour activity in low-glucose environments in both in vitro and in vivo models.

Conclusion: This straightforward modification is compatible with current clinical approaches and may improve the efficacy of T-cell therapies for solid tumours.

Keywords: CAR T cells; immunometabolism; immunotherapy.

Graphical Abstract

Figure 1.

Glucose deprivation attenuates P28ζ CAR T-cell function. (a) PC3-LN3-PSMA viability after 72 h co-culture with P28ζ or PTr CAR T cells in media supplemented with increasing concentrations of glucose. b-d (b) Fold proliferation; (c)CD69 and PD-1 expression; and (d)Annexin V binding of P28ζ or PTr CAR T cells co-cultured with PC3-LN3-PSMA cells for 72 h in media supplemented with increasing concentrations of glucose. e. PC3-LN3-PSMA viability after each restimulation round when co-cultured with serially restimulated P28ζ cells in the presence of increasing concentrations of glucose. Statistical significance relative to 25mM glucose conditions was calculated by two-way ANOVA, with P < 0.05 considered significant. p-values are displayed where statistical significance was found. Data represents mean ± SEM of three independent healthy donors.

Figure 2.

Ectopic expression of GLUT5 and media supplementation with fructose rescues 4P28ζ cytotoxicity and proliferation in the absence of glucose. a. 4P28ζ, 4PTr and GLUT5_4P28ζ construct design. b. 4P28ζ, 4PTr, and 4P28ζ_GLUT5 transduction efficiency and CAR mean fluoresce intensity (MFI) as defined by staining for the CAR Myc tag. Statistical significance was calculated by two-way ANOVA. c. Representative GLUT5 and β-actin expression by 4P28ζ, 4PTr, and GLUT5_4P28ζ as determined by western blotting. d. PC3-LN3-PSMA viability after co-culture with 4P28ζ, 4PTr, or 4P28ζ_GLUT5 cells for 72 h in media containing no glucose/fructose, 10 mM glucose, or 10 mM fructose. e. Relative proliferation of 4P28ζ, 4PTr, or 4P28ζ_GLUT5 cells after co-culture with PC3-LN3-PSMA cells for 72 h in media containing no glucose, 10 mM glucose, or 10 mM fructose. f. PC3-LN3-PSMA viability after each restimulation round when co-cultured with serially-restimulated 4P28ζ or 4P28ζ_GLUT5 cells in media no glucose, 10 mM glucose, or 10 mM fructose. Statistical significance relative to no glucose/fructose conditions was calculated by two-way ANOVA with P < 0.05 considered significant. P-values are displayed where statistical significance was found. Data represent mean ± SEM of three to seven independent healthy donors.

Figure 3.

Ectopic GLUT5 expression enables fructose consumption and lactate production by activated T cells. a. Relative proliferation of 4P28ζ, 4PTr or GLUT5_4P28ζ cells after stimulation with PMA and Ionomycin for 72 h in media containing no glucose, 10 mM glucose or 10 mM fructose. (b-d) Cell culture supernatant from the experiment described in (a) was collected at 0, 24, 48, and 72 h and analysed for (b) glucose, (c) fructose and (d) lactate concentrations. Statistical significance relative to no glucose/fructose conditions was calculated by two-way ANOVA, with P < 0.05 considered significant. P-values are displayed where statistical significance was found. Data represent mean ± SEM of three independent healthy donors.

Figure 4.

Fructose acts as an ACS for GLUT5-expressing CAR T cells in vivo. a. in vivo. tumour challenge experimental design. b-d (b) tumour growth for grouped and individual animals; (c) survival; and (d) weight change for mice inoculated with PSMA-LN3-PSMA cells and treated with either PBS (n = 2), 4P28ζ (n = 4), 4PTr (n = 4) or GLUT5_ 4P28ζ (n = 4) while receiving daily intraperitoneal (IP) fructose injections. Dashed line indicates time-point of PBS or CAR T-cell infusion. Statistical significance of tumour volume was calculated by two-way ANOVA, statistical significance of mice survival was calculated by logrank test. P < 0.05 was considered significant. P-values are displayed where statistical significance was found.

Figure 5.

Fructose can be utilized as ACS by CAR T cells targeting HER-2 and by TCR-T cells targeting gp100. a. 4H28ζ, 4HTr, and 4H28ζ_GLUT5 construct design. b. MCF7 viability after co-culture with 4H28ζ, 4HTr, or GLUT5_4H28ζ cells for 72 h in media containing no glucose/fructose, 10 mM glucose, or 10 mM fructose. c. Relative proliferation of 4H28ζ, 4HTr, or GLUT5_4H28ζ cells after co-culture with PC3-LN3-PSMA cells for 72 h in media containing no glucose/fructose, 10 mM glucose, or 10 mM fructose. d. 4gp100, 4 and 4gp100_GLUT5 construct design. e. Hela-gp100 viability after co-culture with 4gp100, 4, or 4gp100_GLUT5 cells for 72 h in media containing no glucose, 10 mM glucose, or 10 mM fructose. f. Relative proliferation of 4gp100, 4, or 4gp100_GLUT5 cells after co-culture with Hela-gp100 cells for 72 h in media containing no glucose, 10 mM glucose, or 10 mM fructose. Statistical significance relative to no glucose/fructose conditions was calculated by two-way ANOVA, with P < 0.05 considered significant. P-values are displayed where statistical significance was found. Data represent the mean ± SEM of three independent healthy donors.