Editors' ChoiceCancer

Metabolic vulnerabilities of intratumoral T cells and tumor cells

See allHide authors and affiliations

Science Translational Medicine  16 Dec 2020:
Vol. 12, Issue 574, eabf7739
DOI: 10.1126/scitranslmed.abf7739

Abstract

Tumor-infiltrating CD8+ T cells exhibit more metabolic vulnerabilities than tumor cells.

Recent pioneering studies have revealed that cellular metabolism fundamentally controls T cell effector function, which underpins therapeutic efficacy of immunotherapies such as immune checkpoint inhibitors. Thus, substantial research has been focused on modulating metabolism of T cells to boost anti-tumor responses, but largely to no avail due to essential functions of many same metabolic pathways in both effector T cells and tumor cells. To solve this metabolic conundrum, defining differential metabolic sensitivities between T cells and tumor cells is an immediate next step.

As an initial effort, Drijvers et al. developed a high-throughput in vitro co-culture system consisting of activated OT-I CD8+ T cells recognizing ovalbumin (OVA) and B16 melanoma or MC38 colon adenocarcinoma cells expressing OVA. By screening 240 compounds that target a wide array of metabolic pathways, they found three of the top 10 compounds with the highest specific killing of CD8+ T cells inhibited glutathione peroxidase 4 (GPX4), the only glutathione peroxidase enzyme capable of reducing hydroperoxides in the cell membrane, thereby disrupting redox homeostasis and leading to lipid peroxidation. This in turn induced the iron-dependent cell death pathway, ferroptosis, and caused significant death of CD8+ T cells, which was associated with reduced killing of co-cultured tumor cells. Both pharmacological inhibition of ferroptosis and genetic overexpression of GPX4 and FSP1 (ferroptosis suppressor protein 1) reversed these effects, supporting an essential role of ferroptosis in this process. Of note, unlike in vitro results, transferred CD8+ T cells overexpressing GPX4 and FSP1 into B16-OVA–bearing mice did not show survival advantages over wild-type cells. Further, pharmacological inhibition or genetic deletion of ACSL4 (acyl-CoA synthetase long-chain family member 4), a promoter of ferroptosis, also suppressed ferroptosis sensitivity of CD8+ T cells to GPX4 inhibitors. However, in the absence of GPX4 inhibition, ACSL4 deletion in CD8+ T cells dampened their ability to kill co-cultured B16-OVA cells and their survival both in vitro and in vivo, highlighting an important CD8+ T cell-intrinsic role of ACSL4 for optimal anti-tumor immunity.

Taken together, these studies demonstrate that CD8+ T cells are more metabolically vulnerable to redox perturbation by GPX4 inhibition compared to tumor cells in vitro. Yet, it remains unclear how to target this pathway to boost anti-tumor immunity. It would also be important to assess tumor growth upon GPX4 inhibition as a direct readout of anti-tumor responses. From the standpoint of therapeutics development, more efforts should be put toward evaluating compounds that selectively kill tumor cells and spare T cells.

Highlighted Article

Stay Connected to Science Translational Medicine

Navigate This Article