Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Tumor cells and tumor infiltrating lymphocytes (TIL) competes for glucose and other metabolites within the tumor microenvironment for their survival. Glucose consumption by tumors metabolically restricts T cell's ability to produce effector cytokines and therefore approaches to improve the overall metabolic fitness of TIL may improve tumor regression in cancer patients. Long-term survival and anti-tumor immunity critically depends on their metabolic fitness but approaches to select metabolically robust T cells for adoptive immunotherapy remains less clear. Here we introduce a simple and clinically translatable method using a lipophilic cationic dye (tetramethylrhodamine methyl ester-TMRM) to identify and isolate metabolically-robust T cells based on mitochondrial membrane-potential (∆Ψm). Cells with lower membrane-potential (low-∆Ψm) had a molecular profile characteristic of memory precursors and displayed an enhanced ability to enter the memory pool as compared to cells displaying higher mitochondrial potential (high-∆Ψm) characteristic of short-lived effectors. Interestingly, we also found that multiple distinct negative inhibitory receptors such as programmed death-1 (PD-1), cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), B- and T-lymphocyte attenuator (BTLA), Lymphocyte-activation gene 3 (LAG-3) and T cell immunoglobulin mucin receptor 3 (TIM-3) were enriched in the high-∆Ψm subset compared to the low-∆Ψm subset. Comprehensive metabolomic and gene expression profiling demonstrated global features of metabolic fitness in low ∆Ψm sorted CD8+ T cells—including reduced glycolysis, enhanced fatty-acid oxidation and robust spare respiratory capacity. Transfer of low-∆Ψm T cells was associated with superior long-term in vivo persistence as evidenced by 100 fold increase in the frequency of T cells 300 days after adoptive transfer, augmented autoimmunity and an enhanced capacity to eradicate established cancer compared with high-∆Ψm cells. High-∆Ψm T cells exhibited elevated ROS levels, increased effector cytokines and had up-regulation of genes involved in DNA replication, DNA repair and cell-cycle arrest genes compared to low-∆Ψm T cells. Surprisingly, use of Ψm to enrich for cells with superior metabolic features was observed within central-memory (TCM) and effector (Tc17, Th1, Th17) T cells as well as long-term hematopoietic stem cells (LT-HSC). Finally, we also demonstrate that mitochondrial membrane potential based sorting can identify CD45RO- CCR7+ human CD8+ T cells. These findings demonstrate that metabolic-sorting serves as a complementary strategy to the use of conventional cell surface markers for identifying cells with the capacity for long-term survival and ongoing effector function after adoptive-transfer. This novel metabolism-based approach may be broadly applicable to therapies involving transfer of hematopoietic stem cells or lymphocytes for treatment of viral-associated illnesses and advanced cancer.