Metabolic tug-of-war in tumors results in diminished T cell antitumor immunity

ABSTRACT

Glycolytic activity in T cells governs T cell responses by controlling various cellular processes, including proliferation, survival, and effector functions. We recently showed that the tumor microenvironment diminishes T cell antitumor responses by depriving glucose to infiltrating T cells. Moreover, metabolic rewiring tumor-reactive T cells sustain T cell metabolic fitness and antitumor responses.

KEYWORDS:

• Immunotherapy
• immune checkpoint
• metabolic reprogramming
• PEP
• tumor infiltrating T cell

Abbreviations

GLUT1	=	Glucose transporter 1
HIF-1α	=	Hypoxia-inducible factor 1-alpha
NFAT	=	Nuclear factor of activated T-cells
PCK-1	=	phosphoenolpyruvate carboxykinase 1
PD-1	=	Programmed cell death protein 1
PEP	=	phosphoenolpyruvate
PKM2	=	pyruvate kinase M2 isoform
SERCA	=	sarcoplasmic or endoplasmic reticulum Ca2+-ATPase

Malignant transformation is an accumulative process of multiple genomic alterations, including metabolic reprogramming, oncogenic mutations, epigenetic modifications, and cellular signaling alterations.¹ Among these traits of tumor cells, deregulated cellular metabolism is an important feature to support unrestricted growth and survival of tumor cells.² Tumor cells can utilize glucose to generate ATPs in the presence of oxygen, which is well known as Warburg effects (Warburg glycolysis). Although Warburg glycolysis is an inefficient way to consume glucose, this process allows tumor cells to fulfill both bioenergetic and biosynthetic demands. Importantly, oncogenic mutation and signaling cascades have been revealed to initiate metabolic alterations in tumor cells. Therefore, targeting cancer metabolism is considered as a promising strategy for antineoplastic drugs development. However, it remains elusive whether deregulated cancer metabolism promotes tumor progression by other unexplored mechanisms and if targeting cancer metabolism impairs functions of other cells in the tumor microenvironment, such as antitumor immunity.

T cells undergo metabolic reprogramming when they receive activation signal from T cell receptor (TCR) and co-stimulatory receptors.³ Recent studies showed that metabolic reprogramming, including aerobic glycolysis and amino acid metabolism, modulates T cell differentiation and effector functions.⁴ In addition to influencing the balance between mTOR and AMPK activity, studies suggested that nutrient metabolism modulates T cell function and exhaustion by the activation of metabolic enzymes and the generation of metabolic intermediates.^{5, 6} Even through studies suggested that TCR signaling cascades are controlled by nutrient metabolism, the underlying mechanisms remain mostly unclear. Moreover, the similarity between cancer cell and T cell metabolism raises two important issues on cancer immunotherapy. First, common metabolic demands in tumor cells and T lymphocytes may result in metabolic competition in the tumor microenvironment, which could lead to T cell dysfunction/exhaustion. Secondly, targeting cancer metabolism may not only suppress tumor growth but also impairs T cell antitumor immune responses. Therefore, understanding the metabolic regulation involved in T cell activation and antitumor responses is a dire need. Also, it is critical to determine whether metabolic regulation can be exploited to improve cancer immunotherapy.

In our recent study, we first found that the tumor microenvironment was glucose deprived and tumor-infiltrating CD4⁺ T cells displayed gene signature of glucose deprivation in genetically modified murine melanoma model and B16 melanoma model.⁷ Since hexokinase 2 (HK2) controls the rate-limiting step of glycolysis, we postulated that increased aerobic glycolysis in tumor cells by overexpressing HK2 could lead to immune evasion accompanied with reduced antitumor responses in tumor-infiltrating T cells. Using immune-deficient mice and melanoma cell engraftment model, we found that tumor-infiltrating CD4⁺ T cells failed to sustain antitumor effector functions and HK2-overexpressing melanoma cells were resistant to CD4⁺ T-cell-mediated antitumor immunity. In addition to decreased glucose availability in tumors, we also reported that HK2-overexpressing melanoma cells were more potent to suppress glucose uptake ability in CD4⁺ T cells. Taken together, our results uncover that tumor cells with elevated glycolytic rates are capable to promote T cell dysfunction/exhaustion by restricting glucose to tumor-infiltrating T cells.

We further investigated the underlying metabolic regulation by which glycolysis sustains T cell antitumor responses. Surprisingly, we found that sufficient glycolytic activity in T cells was a critical event to sustain cytoplasmic Ca²⁺ accumulation and promote nuclear translocation of NFAT1 and IFNγ and CD40L production. Then, we determined that a glycolytic metabolite, phosphoenolpyruvate (PEP), regulated Ca²⁺-NFAT signaling pathway and the generation of effector molecules by modulating sarcoplasmic or endoplasmic reticulum Ca²⁺-ATPase (SERCA) activity with genetic and pharmacological approaches. Of note, metabolic adaption can support T cell survival by utilizing oxidative phosphorylation metabolism when T cells experienced glucose starvation. We then investigated if manipulating tumor-reactive T cells to generate PEP from other metabolic intermediates is able to restore T cell antitumor responses in glucose-deprived conditions. To do so, we overexpressed phosphoenolpyruvate carboxykinase 1 (PCK1), which can converts TCA cycle intermediate, oxaloacetate (OAA), into PEP in tumor-specific T cells. Indeed, PCK1-overexpressing T cells sustained Ca²⁺ flux, NFAT nuclear localization and the effector functions in the glucose-deprived condition. Moreover, PCK1-overexpressing T cells displayed higher antitumor abilities in tumors, but not in other peripheral tissues and spleens. These findings suggest that PCK1 overexpression only boosts effector functions of T cells in the glucose-deprived circumstance. Altogether, the study provided a proof-of-concept evidence that modulating metabolic checkpoints is a promising strategy to manipulate T cell functions in the specialized microenvironment, like tumors. (Fig. 1)

Figure 1. (A) Cancer cells create a glucose deprivation tumor microenvironment by their own glycolytic activity. Deficiency of glucose in the tumor microenvironment promotes exhaustion/dysfunction in tumor infiltrating T cells. (B) PCK-1 overexpression restores the calcium-NFAT signaling pathway and antitumor responses in tumor-reactive T cells infiltrating into the glucose-deprived tumor microenvironment.

In summary, our findings uncovers that tumor cells cause T cell dysfunction in the tumor microenvironment by depriving glucose in infiltrating T cells. This study further implies that deregulated cancer metabolism could be one of the underlying mechanism which promote immune evasion. Most importantly, it also revealed that metabolic rewiring tumor-reactive T cell is a promising strategy to enhance T cell tumoricidal activity. Of note, PD-1 blockade has also been shown recently to improve metabolic fitness of T cells on fighting against chronic viral infection and tumor cells. ^{8, 9} Taken together, these studies suggest that modulating metabolic activities in immune cells, especially T cells, has therapeutic potential to suppress diseases progression. In addition to aerobic glycolysis, tumor cells also display other metabolic abnormality, such as increased fatty acid synthesis and amino acid metabolism. Therefore, a global understanding of how metabolic state in the tumor microenvironment modulates immune infiltration and the activation status of infiltrating immune cells is an important future direction. Additionally, determining the metabolic regulation mediated by different nutrients in immune cells would be of great interest for future cancer immunotherapy.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

P.-C. H. is supported by ISREC research grant (26075483), SNSF Project Grant (31003A_163204), and Ludwig Center for Cancer Research (2607011910).