You are here

Article Text

Article menu
Immune cell therapies and immune cell engineering
Original research
NR4A ablation improves mitochondrial fitness for long persistence in human CAR-T cells against solid tumors
  1. Kensuke Nakagawara1,2,
  2. Makoto Ando2,
  3. Tanakorn Srirat2,
  4. Setsuko Mise-Omata2,3,
  5. Taeko Hayakawa2,
  6. Minako Ito4,
  7. Koichi Fukunaga1 and
  8. Akihiko Yoshimura2,3
  1. 1Division of Pulmonary Medicine, Department of Medicine, Keio University School of Medicine, Tokyo, Japan
  2. 2Department of Microbiology and Immunology, Keio University School of Medicine, Tokyo, Japan
  3. 3Research Institute for Biomedical Sciences, Tokyo University of Science, Noda, Chiba, Japan
  4. 4Division of Allergy and Immunology, Medical Institute of Bioregulation, Kyushu University, Fukuoka, Japan
  1. Correspondence to Professor Akihiko Yoshimura; yoshimura{at}keio.jp; Dr Kensuke Nakagawara; k.nakagawara{at}keio.jp

Abstract

Background Antitumor effect of chimeric antigen receptor (CAR)-T cells against solid tumors is limited due to various factors, such as low infiltration rate, poor expansion capacity, and exhaustion of T cells within the tumor. NR4A transcription factors have been shown to play important roles in T-cell exhaustion in mice. However, the precise contribution of each NR4a factor to human T-cell differentiation remains to be clarified.

Methods In this study, we deleted NR4A family factors, NR4A1, NR4A2, and NR4A3, in human CAR-T cells recognizing human epidermal growth factor receptor type 2 (HER2) by using the CRISPR/Cas9 system. We induced T-cell exhaustion in these cells in vitro through repeated co-culturing of CAR-T cells with Her2+A549 lung adenocarcinoma cells and evaluated cell surface markers such as memory and exhaustion phenotypes, proliferative capacity, cytokine production and metabolic activity. We validated the antitumor toxicity of NR4A1/2/3 triple knockout (TKO) CAR-T cells in vivo by transferring CAR-T cells into A549 tumor-bearing immunodeficient mice.

Results Human NR4A-TKO CAR-T cells were resistant against exhaustion induced by repeated antigen stimulation in vitro, and maintained higher tumor-killing activity both in vitro and in vivo compared with control CAR-T cells. A comparison of the effectiveness of NR4A single, double, and TKOs demonstrated that triple KO was the most effective in avoiding exhaustion. Furthermore, a strong enhancement of antitumor effects by NR4A TKO was also observed in T cells from various donors including aged persons. Mechanistically, NR4A TKO CAR-T cells showed enhanced mitochondrial oxidative phosphorylation, therefore could persist for longer periods within the tumors.

Conclusions NR4A factors regulate CAR-T cell persistence and stemness through mitochondrial gene expression, therefore NR4A is a highly promising target for the generation of superior CAR-T cells against solid tumors.

  • Lung Cancer
  • Chimeric antigen receptor - CAR
  • Solid tumor
  • Stem cell
  • Adoptive cell therapy - ACT

Data availability statement

Data are available in a public, open access repository. Data are available upon reasonable request. The RNA sequencing data supporting the findings of this study have been deposited in the DNA Data Bank of the Japan Sequence Read Archive (https://www.ddbj.nig.ac.jp/index.html) with the accession numbers GSE241456.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/jitc-2023-008665

Statistics from Altmetric.com

    Request Permissions

    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.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • NR4A transcription factors have been shown to play significant roles in T-cell exhaustion in mice. However, the precise contribution of each NR4A factor to human T-cell differentiation remains to be elucidated.

    WHAT THIS STUDY ADDS

    • We clarified the comparison of the effectiveness of NR4A single, double, and triple knockouts demonstrated that triple knockout was the most effective in avoiding exhaustion, maintaining persistence, and enhancing antitumor effect in human chimeric antigen receptor (CAR)-T cells.

    • NR4A factors regulate mitochondrial gene expression, and NR4A triple knockout CAR-T cells sustained energy production even after repeated antigen stimulation.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • We propose that NR4A transcription factors play important roles in determining T-cell fate, and that NR4A deletion in CD8+T cells strongly enhances CAR-T-cell therapeutic effects in solid tumors. Targeting NR4A might provide a novel therapeutic approach of immunotherapies for solid tumor.

    Introduction

    Chimeric antigen receptor (CAR)-T cell therapy has a higher therapeutic efficacy, mainly for B-cell malignancies.1 However, current CAR-T cell therapies have limited efficacy against solid tumors, with little or no tumor regression or rapid inactivation of CAR-T cells after an initial response.2 Several attempts have been made to overcome the limitations of CAR-T cell therapy for solid tumors.3 However, these have not been completely successful owing to various reasons, including the adverse effects of CAR-T cells on normal tissues,4 inhibition of T-cell infiltration into tumors by tumor-associated fibroblasts and abnormal tumor vasculature,5 6 inactivation and exhaustion of T cells due to prolonged exposure to tumor antigens in the tumor microenvironment,7 8 and interaction with immunosuppressive factors, such as regulatory T (Treg) and transforming growth factor-β.9 10 In some cases, the self-oligomerization of CARs in the absence of ligands induces tonic signals that11 result in exhaustion phenotypes.12 Among them, T-cell exhaustion is characterized by a loss of effector and memory phenotypes, inability to produce cytokines, such as interferon gamma (IFN)-γ, tumor necrosis factor (TNF)-α, and interleukin (IL)-2, and increased expression of inhibitory receptors.13 An elevated expression of inhibitory receptors on T cells and their binding to inhibitory ligands within the tumor microenvironment further promotes T-cell dysfunction.

    Understanding the cellular and molecular regulation of exhausted T cells has become a central focus in cancer immunotherapy. T cells under chronic antigenic stimuli, such as solid tumors and chronic infections, have been shown to differentiate into terminally exhausted T (TEX) cells from the precursor of exhausted T (TPEX) cells.14 15 TPEX cells are thought to play an important role in adoptive immune therapy in solid tumors because they can self-renew, express stem cell markers, such as TCF7, and express higher levels of cytokines.16 17 Immune checkpoint inhibitors, such as anti-programmed cell death (PD)-1 antibodies, increase the number of TPEX cells; however, they are not so effective for TEX cells, which express various inhibitory molecules, including PD-1, cytotoxic T-lymphocyte antigen-4 (CTLA4), T-cell immunoglobulin and mucin domain 3 (TIM3), lymphocyte activation gene-3 (LAG3), and T-cell immunoreceptor with Ig and ITIM domains (TIGIT), and are epigenetically fixed.18 Therefore, targeting the transcription factors involved in T-cell exhaustion may improve the efficacy of adoptive T-cell therapy against solid tumors. Various transcription factors have been reported to regulate T-cell exhaustion, including BLIMP1, BATF, IRF4, TOX, and NR4A.19 20 The deletion of these factors in CAR-T cells has been shown to enhance their antitumor effects.17 19–23

    NR4A is a member of the orphan nuclear receptor family consisting of NR4A1 (Nur77/TR3/NGFI-B), NR4A2/Nurr1 (TINUR/NOT), and NR4A3/NOR1 (MINOR/CSMF), which act as ligand-independent transcription factors and regulate the CD8+ T-cell signaling pathway such as T-cell receptor signaling and JAK-STAT signaling pathways.24 25 In chronic immune responses in mice models, NR4A transcription factors are induced by NF-AT and promote dysfunctional/exhausted states of CD8+ T cells by suppressing cytokine expression via competing with AP-1/NF-kB and accelerating the expression of inhibitory receptors through directly promoting transcription.24 25 In previous studies, NR4A1, NR4A2, and NR4A3 triple knockout (TKO) CAR-T cells exhibited strong antitumor effects against solid tumors in in vivo mouse models.24 26 However, the effects of NR4A knockdown in human T cells and the potential mechanisms for avoiding exhaustion remain to be investigated.

    In this study, we generated NR4A TKO CAR-T cells targeting human epidermal growth factor receptor type 2 (HER2) from healthy donor peripheral blood T cells using the CRISPR/Cas9 system and evaluated their phenotypes and antitumor responses. We believe our study would serve as a basis for future studies regarding targeting NR4A as a promising pharmaceutical strategy against solid tumors.

    Methods

    Flow cytometry and antibodies, cells lines, generation of CAR-T cells, and western blot analysis are described in the online supplemental methods.

    Supplemental material

    Continuous antigen exposure assay

    Induction of T-cell exhaustion in vitro was performed using a modified procedure described by Good et al.27 Tumor cells were seeded in a complete Dulbecco's Modified Eagle Medium (DMEM) high glucose medium 1-day prior to co-culturing. The next day, the DMEM high glucose medium was replaced with a T-cell medium, and CAR-T cells were seeded on top of the adherend tumor cells at a 1:1 effector : target (E:T) ratio with IL-2 at 50 IU mL-1. Subsequent co-cultures were set-up every 72–96 hours. The co-cultures were thoroughly suspended by frequent pipetting and the cell suspension was spun down and the supernatant was removed. The cells were resuspended in a fresh R10 medium. T cells were not separated using magnetic beads or flow cytometry. The resulting T-cell suspension was transferred into adherend A549-coated plate cells at a 1:1 E:T ratio for continuous co-culture.

    Proliferative capacity, cytotoxicity, and IL-2 and other cytokine production

    After 14 days of continuous antigen exposure (CAE),27 to accurately assess the proliferation, antitumor effect, and cytokine production capacity of CD8+CAR-T cells, we sorted CD8+CAR-T cells an SH800S Cell Sorter (Sony Biotechnology) before the assay. T-cell immunophenotyping was carried out using the following antibodies: CD8-APC (1 µL/100 µL; BioLegend, #300912), CD4-PerCP/Cyanine5.5 (1 μL/100 µL; BioLegend, #300530), CD3-PE (1 µL/100 µL; BioLegend, #300407), Streptavidin-Bv421 (1 µL/100 µL; BD Biosciences, #563259), and fixable viability Dye eFluor 780-APC/Cy7 (1 µL/4000 µL; eBioscience, #65–0865–14). The sorted CAR-T cells were labeled with 2.5 µmol/L of CellTrace Violet (Thermo Fisher Scientific, #C34571) at 37°C for 8 min in phosphate-buffered saline (PBS) and stimulated with A549 cells (E:T ratio=1:1) for 3 days. The cell division, phenotype, and number of live T cells were assessed using an FACS CytoFLEX Flow Cytometer (Beckman).

    To assess IL-2, TNF-α, and IFN-γ production, the culture supernatants were collected 22–24 hours after the T cells were stimulated with A549 cells (E:T ratio=1:1). The concentration of IL-2, TNF-α, and IFN-γ in the culture supernatant was measured using an ELISA (eBioscience).

    To assess the cytotoxicity, A549 cells were labeled with 2.5 µmol/L of CellTrace Violet at 37°C for 8 min in PBS, and then T cells were co-cultured with the labeled A549 cells (E:T ratio=4:1, 2:1 and 1:1) for 24 hours. Dead A549 cells (CD8α- CellTrace Violet+PI+ cells) were assessed using flow cytometry analysis.

    Quantitative assessment of the mitochondria in CAR-T cells using MitoTracker

    For MitoTracker probe staining, the T cells were incubated in a 6-well plate at 1,000,000 cells per well in 25 nM of MitoTracker Green (Cell Signaling Technology, #9074), or MitoTracker Deep Red (Cell Signaling Technology, #8778), or MitoSOX (Thermo Fisher Scientific, # M36008) in 100 µL of a warm Roswell Park Memorial Institute (RPMI) 1640 medium in an incubator for 30 min. The cells were then quenched with a warm complete medium at a 1:1 volume, spun down, washed twice with a warm medium, resuspended in 5% Fetal Bovine Serum (FBS)/PBS, and then analyzed using flow cytometry.

    Measurement of extracellular acidification rates and oxygen consumption rates

    The oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were measured using an XFe24 Extracellular Flux Analyzer (Agilent Technologies). The ECAR and OCR were measured using the Seahorse XF Cell Mito Stress Test Kit (Agilent Technologies, #103015–100) as described previously.28

    Morphological evaluation of mitochondria using electron microscopy

    For electron microscopy analysis, CD8+ CAR-T cells were sorted using an SH800S Cell Sorter (Sony Biotechnology). Sectioning, mounting, staining, and observations were carried out by the Hanaichi Ultrastructure Research Institute (Okazaki, Aichi, Japan).

    Briefly, live CD8+CAR-T cells were sorted and prefixed with 2% glutaraldehyde in 100 mM phosphate buffer. Following fixation, samples were post-fixed in 2% osmium tetroxide for 2 hours at 4°C. Subsequently, samples were dehydrated in ethanol and embedded in EPON812. Thin sections were stained with uranyl acetate and lead citrate, then imaged using an H-7600 transmission electron microscope at 100 kV (HITACHI). The sizes of mitochondria, number of mitochondria and cristae were quantified using the analytic application BZ-X analyzer (Keyence).

    RNA sequencing analysis

    For RNA sequencing (RNA-seq), CD8+ CAR-T cells were sorted using an SH800S Cell Sorter (Sony Biotechnology). RNA-seq analysis was performed using Rhelixa. Total messenger RNA (mRNA) was isolated using an RNeasy Plus Mini Kit (QIAGEN). RNA-seq by Rhelixa was performed using the Illumina HiSeq 4000 platform. After trimming adapters using Cutadapt and checking for quality using FastQC, single-end reads were aligned to the human reference genome (hg38) using HISAT2. RNA libraries for RNA-seq were prepared using the NEBNext Poly(A) mRNA Magnetic Isolation Module (NEB E7490) and NEBNext Ultra RNA Library Prep Kit for Illumina (E7530) following the manufacturer’s protocol. Gene expression data files were obtained from Rhelixa, and further statistical analyses were performed using the R statistical software environment and an integrated web application (iDEP V.1.1).

    Principal component (PC) analysis was performed using the prcomp function of the statistics package and visualized using RIAS. The statistical significance of differentially expressed genes (DEGs) was assessed using the edgeR package (V.3.22.1) and visualized using base RIAS. Gene expression was visualized using the heatmap.2 function of the gplots package. To generate Gene Set Enrichment Analysis (GSEA)-enrichment plots, processed RNA-seq data sets were downloaded from the GEO database (GSE9650), CD8+ CAR-T cell dysfunction-related gene set (27(data set, accession number GSE160174)), and stem-like memory and exhausted CD8+ T-cell gene set based on tumor-infiltrating lymphocytes (TILs) in non-small cell lung cancer (NSCLC) (29(data set, accession number GSE176022) and 30(data set, accession number GSE99254)).

    A549 xenograft model in immunodeficient mice

    Female NOD.Cg-PrkdcscidIl2rgtm1Wjl/Szj (NSG) mice were purchased from the Charles River Laboratory, Japan. All the mice were housed in specific pathogen-free facilities at Keio University (Tokyo, Japan).

    Female NSG mice that were 5–7 weeks old were engrafted with 1×106 A549 lung adenocarcinoma cells via subcutaneous injection 14 days before CAR-T cell transfer. CAR-T cells (5×106) were injected intravenously into the tail vein. A549 cell progression was assessed using calipers. The mice were weighed weekly. On day 7, after CAR-T cell transfer, tumor masses and spleen were excised and minced. Tumor masses were digested with 2 mg/mL collagenase D and 300 µg/mL DNase I at 37°C for 30 min with gentle shaking. The resulting cell suspension of tumor masses and minced spleen were then passed through a 100 µm filter to remove large debris. Erythrocytes were lysed using an ammonium-chloride-potassium lysis buffer. The numbers of CAR-T cells (mCD45−FVD−hCD8α+CAR+ and mCD45−FVD−hCD4+CAR+) were measured and analyzed using flow cytometry. When the mice reached the humane endpoint in the tumor model, they were sacrificed as approved by the Animal Ethics Committee of Keio University (Tokyo, Japan).

    For the rechallenge experiment, a separate cohort of mice were then engrafted with the same 1×106 A549 lung adenocarcinoma cells via subcutaneous injection, and then 14 days later they were intravenously injected with 5.0×106 NR4A TKO CAR-T cells as before. In this case, after 28 days of the CAR-T cell injection, mice were then engrafted with 1×106 A549 cells rechallenge injections. Tumor masses were followed with calipers and monitored by weight and for any signs of morbidity per our protocol as listed above.

    Statistics

    Statistical analyses were performed using Student’s t-test, one-way or two-way analysis of variance with post hoc Tukey test, or the Kaplan-Meier method using GraphPad Prism V.8 software (GraphPad Software). All data are presented as the mean±SEM. Statistical significance was set at p<0.05.

    Results

    NR4A knockout by CRISPR-mediated genome editing and HER2 CAR transduction

    First, we evaluated the efficiency of using CRISPR/gRNA gene editing to knock out NR4A. HER2 CAR-T cells were generated from the pan-T cell fraction of peripheral blood mononuclear cells (PBMCs) by lentiviral CAR complementary DNA (cDNA) transduction, followed by electroporation with ribonucleoproteins (RNPs) targeting NR4A1, NR4A2, and NR4A3 2 days later, and then expansion with IL-2 for 5 days.

    As shown in figure 1A, CRISPR/gRNA gene editing efficiently diminished the expression of NR4A family proteins. We presented the efficacy of each gRNA and confirmed the loss of NR4a in a different donor by our method and online supplemental figure S1A. We employed three gRNAs targeting each of NR4A1, 2, and 3 for deletion (online supplemental table S1A). We compiled the indel frequencies for NR4A single KO (sKO), double KO (dKO), and TKO CAR T cells for each gRNA using the primer sequences for PCR (online supplemental S1B). Although there were differences in the efficacy of gene deletion among the gRNAs, the indel frequencies did not vary greatly across sKO, dKO, and TKO (online supplemental S1C). Next, we assessed the occurrence of deletions, insertions, and translocations at the on-target sites using Synthego’s Inference of CRISPR Edits Analysis in online supplemental figure S1B. Additionally, unintended on-target effect such as large deletions and off-target effects at the top predicted off-target sites were not detected through structural variation analysis using whole genome sequencing. In the copy number variation analysis comparing NR4A TKO CAR-T cells to control, no significant copy number proliferation or reduction were detected (online supplemental figure S1C).

    Figure 1
    Figure 1

    Generation of NR4A TKO CAR-T cells and characterization before and after CAE. (A) NR4A protein expression in chimeric antigen receptor (CAR)-T cells 5 days after electroporation using western blot analysis in donor 1 (20s woman). The red-arrowed bands indicate the NR4A expression. (B) Overview of the procedure for generating NR4A triple knockout (TKO) CAR-T cells. (C) Overview of the procedure of the continuous antigen exposure (CAE) model of human epidermal growth factor receptor type 2 CAR-T cells co-cultured with A549. (D) Proliferation rate of control CAR-T cells and NR4A single (sKO), double (dKO), and TKO CAR-T cells in the CAE model. The data represents mean±SD from different three donors (one 20s woman, one 30s man, and one 40s man). (E) Expression of memory phenotype markers (CD28 and CD62L) in control CAR-T cells and NR4A sKO, dKO, and NR4A1/2/3 TKO CAR-T cells 17 days after CAE in donor 1 (20s woman). (F) Populations of CD28+/CD62L+ CAR-T cells in the control and NR4A TKO CAR-T groups. The mean fluorescence intensity (MFI) of CD62L, CCR7, and CD28 in the control CAR-T cells and NR4A sKO, dKO, and TKO CAR-T cells 17 days after CAE is shown (data are presented as the means of n=3 replicate wells). (G) Representative flow cytometry plots showing the exhaustion markers (PD-1 and TIM-3) in the control CAR-T cells and NR4A sKO, dKO, and TKO CAR-T cells co-cultured with A549 for 17 days in donor 1 (20s woman). (H) Expression and MFI of the T-cell exhaustion markers PD-1, LAG-3, TIM-3, and CD39 in the control CAR-T cells and NR4A sKO, dKO, and TKO CAR-T cells 17 days after CAE (data are presented as the means of n=3 replicate wells). The data represents mean±SD regarding the CAR-T cells from different three donors (one 20s woman, one 30s man, and one 40s man). Data are representative of at least two independent experiments. *p<0.05; **p<0.01; ns, not significant; LAG-3, lymphocyte activation gene-3; PD-1, Programmed cell Death 1; TIM-3, T-cell immunoglobulin and mucin domain 3.

    We then determined the sequences of CAR-transduction and RNP-electroporation; CAR-transduction then RNP-electroporation or RNP-electroporation then CAR-transduction (online supplemental figure S2A). CAR-transduction followed by the RNP-electroporation procedure was superior to the opposite order for both T-cell proliferation and CAR-transduction efficiency (online supplemental figure S2B, C). Therefore, we developed a procedure to generate NR4A-knockout HER2 CAR-T cells, as shown in figure 1B. This involved the isolation and activation of the pan-T cell fraction from PBMCs on day 0, lentiviral CAR cDNA transduction on day 1, electroporation with the NR4A RNP complex on day 3, and expansion with IL-2 alone for 9 days. The cell expansion rate was comparable between control and NR4A1/2/3-TKO HER2 CAR-T cells (online supplemental figure S2B). At this stage, NR4A protein expression was low in CAR-T cells without T cell receptor (TCR) stimulation (figure 1A). Further, both control and CD8+ NR4A1/2/3-TKO HER2 CAR-T cells retained a high expression of the naïve/early memory markers, CD62L and CD28, and low levels of the exhaustion markers, PD-1 and TIM3 (online supplemental figure S2D).

    We induced T-cell exhaustion in these cells in vitro through repeated co-culturing of CAR-T cells with Her2+ A549 lung adenocarcinoma cells using CAE27 (figure 1C). NR4A TKO CAR-T cells exhibited improved proliferative capacity in a CAE model regardless of E/T ratio (online supplemental figure S2E). The CAR-transduction efficiency was similar between the control and NR4A1/2/3 TKO genome editing conditions before CAE, and the CAR expression was nearly 100% after CAE in both cases (online supplemental figure S2F). The CD4/CD8 ratio was slightly reduced in the control and NR4A TKO CAR-T cells following CAE (online supplemental figure S2G). The Treg fraction within the CD4+ T cells was significantly increased in control and NR4A TKO CAR-T cells after CAE (p<0.001 and p<0.01). However, NR4A TKO CAR-T cells consistently exhibited a significantly lower proportion of CAR-Treg cells (p<0.001, online supplemental figure S2G). These results align with our previous findings that NR4A factors play key roles in FOXP3 expression in mouse CD4+T cells.31 32

    Effects of NR4A knockout by CRISPR on early memory phenotypes and T-cell exhaustion after CAE in vitro

    We evaluated proliferation after CAE in eight groups of CAR-T cells: control CAR-T cells treated with scrambled gRNA-CRISPR complex; NR4A1, NR4A2, and NR4A3 sKOs; NR4A1/2, NR4A1/3, and NR4A2/3 dKOs; and NR4A1/2/3 TKO CAR-T cells. Control CAR-T cells underwent growth arrest after the fourth CAE; however, NR4A-TKO CAR-T cells continued to proliferate even after the sixth CAE round (figure 1D). A single deletion of NR4A1 or NR4A2 had minimal impact on growth arrest, whereas the sKO of NR4A3 slightly improved growth inhibition. This observation aligns with a previous study indicating that the deletion of NR4A3, rather than NR4A1 and NR4A2, in CAR-T cells confers resistance to T-cell exhaustion.20 NR4A1/2 dKO exhibited a proliferation pattern similar to that of NR4A3 sKO under CAE conditions. NR4A1/3 and NR4A2/3 dKOs also exhibited significantly improved proliferative capacities after CAE. However, NR4A1/2/3 TKOs demonstrated the highest resistance to CAE-induced growth arrest (p<0.0001, figure 1D). The final cell count of NR4A TKO CAR-T cells was more than 10 times greater than that of the control CAR-T cells. Furthermore, the NR4A TKO CAR-T cells ceased to grow after the seventh round of CAE, suggesting that NR4A TKO did not lead to indefinite T-cell stemness.

    Next, we examined cell surface markers related to early memory and exhaustion after 17 days of CAE (figure 1E–H). The cell surface markers used in this study were chosen based on their common usage in previous studies.23–28 17 days of CAE drastically reduced the levels of early memory markers, such as CD62L, CD28, and CCR7, whereas exhaustion markers, such as PD-1, LAG3, TIM-3, and CD39 were strongly induced in control CD8+ CAR-T cells. NR4A1/2, NR4A2/3, and NR4A1/3 dKOs and NR4A1, NR4A2, and NR4A3 sKOs exhibited slightly higher levels of early memory marker expression and reduced levels of exhaustion markers than the control (figure 1F,H). However, NR4A1/2/3 TKO CAR-T cells exhibited a higher percentage of CD62L+CD28+CD8+ memory fraction, and the expression levels determined by mean fluorescence intensity (MFI) of CD62L and CD28 were drastically increased compared with control CD8+ CAR-T cells (figure 1E,F). Similarly, the expression levels of CCR7 were also elevated in TKO CD8+ CAR-T cells. In contrast, the PD-1−LAG3−TIM3−CD39−CD8+ fraction was significantly decreased in TKO CAR-T cells, and the expression levels of PD-1, LAG3, TIM-3, and CD39 were also markedly decreased (p<0.05, figure 1H). The levels of early memory and exhaustion markers in TKO CAR-T cells after CAE were comparable to those in T cells before CAE (online supplemental figure S2D), suggesting that NR4A TKO CAR-T cells were highly resistant to T-cell differentiation from early memory to exhaustion. These results also confirmed that NR4A1, NR4A2, and NR4A3 were redundant, at least for T-cell exhaustion following CAE and that the simultaneous deletion of all three isoforms was necessary to obtain full resistance to exhaustion. The consistency of these results was observed across multiple different donors under biological replication conditions.

    NR4A TKO CAR-T cells are superior to BLIMP1/NR4A3 dKO in proliferation, expression of memory markers and resistance to exhaustion during CAE

    Jung et al found that PRDM1 deletion (which encodes BLIMP1) and NR4A3 knockout confers not only exhaustion resistance but also stemness, which leads to a strong antitumor effect20 because BLIMP1 is antagonistic to TCF1. Therefore, we compared BLIMP1/NR4A3 double deletion and NR4A1/2/3 TKO using our Her2 CAR and CAE conditions. As shown in online supplemental figure S3A, CRISPR/gRNA gene editing efficiently diminished the expression of BLIMP1 and NR4A3 proteins. BLIMP1/NR4A3 dKO CAR-T cells demonstrated improved cell proliferative potential compared with control CAR-T cells, however inferior to NR4a TKO CAR-T cells (online supplemental figure S3B). Moreover, NR4A TKO CAR-T cells were much better than BLIMP1/NR4A3 dKOs in maintaining early memory markers and resistance to exhaustion during CAE online supplemental figure S3C–F. In particular, TIM3 and CD39, which are highly expressed in TEX, were not different between control and BLIMP1/NR4A3 CAR-T cells, whereas these expressions were markedly decreased in NR4A TKO CAR-T cells (online supplemental figure S3E, F). These results indicate more robustly that the triple deficiency of NR4A1/2/3 can confer the strongest stemness and resistance to exhaustion.

    NR4A TKO induced consistent naive memory and effector-like effects on CAR-T cells irrespective of the donors

    To ensure the broad applicability of NR4A deletion in CAR-T cell therapy, it is imperative to confirm the consistency of these effects across different donors. This is especially pertinent when considering T cells sourced from older individuals, whose T-cell populations typically exhibit a decrease in naive cells and an increase in terminally differentiated effector memory cells.33 34 Furthermore, the efficacy of CAR-T cell therapy is lower in older patients than in younger ones.35 36 Hence, we conducted experiments to verify whether similar effects of NR4A1/2/3 TKO would be observed in five distinct donors, including a woman in her 20s, a man in his 30s, a man in his 40s, a woman in her 50s, and a man in his 60s.

    First, we assessed the proliferation rate during CAE. All control CAR-T cells exhibited a decline in proliferation rate following CAE, and a more pronounced reduction occurred as the donor’s age increased (figure 2A red). In contrast, NR4A1/2/3 TKO CAR-T cells maintained their proliferative capacity across all donors, although at a slightly slower rate with increasing donor age (figure 2A blue). Control CAR-T cells derived from older donors, in particular, exhibited a significant decline in proliferation capability, with the expansion curve plateauing by day 14 of CAE. Therefore, we evaluated some cell surface markers-related memory and exhaustion on day 15.

    Figure 2
    Figure 2

    Characterization of the control and NR4A TKO CAR-T cells from different donors. (A) Proliferation rate of the control (red) and NR4A TKO CAR-T cells (blue) from five different donors in the continuous antigen exposure (CAE) model. (B) Comparison of CD8+ CAR-T fractions expressing the early memory markers (CD62L+CD28+ and CCR7+CD45RA+) before (Pre) and after (Post) CAE. (C) Comparison of the exhaustion markers (PD-1+TIM3+ and PD-1+LAG3+TIM3+CD39+) expressing CD8+ CAR-T fractions before (Pre) and after (Post) CAE. (D) MFI of the indicated exhaustion markers in the control and NR4A TKO CD8+ CAR-T cells. N=5 different human donors. Data are representative of at least two independent experiments. *p<0.05; **p<0.01; ns, not significant; CAR, chimeric antigen receptor; LAG-3, lymphocyte activation gene-3; MFI, mean fluorescence intensity; PD-1, Programmed cell Death 1; TIM-3, T-cell immunoglobulin and mucin domain 3; TKO, triple knockout.

    Next, we investigated the expression of early memory and exhaustion markers (figure 2B–D, with representative fluorescence activated cell sorter (FACS) data available in online supplemental figure S4A–D. As depicted in figure 2B, the CD62L+CD28+ and CCR7+CD45RA+ early memory fractions significantly decreased in control CD8+ CAR-T cells on following CAE (p<0.0001, red), whereas these markers were relatively preserved, although not entirely, in CD8+ NR4A TKO CAR-T cells (blue). CAR-T cells from donors in their 40s, 50s, and 60s exhibited a more substantial reduction in the CD62L+CD28+ fraction than those from younger donors in their 20s and 30s, even under NR4A TKO conditions (online supplemental figure S4B). Both the PD-1+TIM3+ and PD-1+LAG3+TIM3+CD39+ exhausted fractions increased significantly during CAE in control CD8+CAR T cells from all donors (p<0.0001, red), whereas this increase was effectively mitigated by NR4A TKO (blue in figure 2C,D). Moreover, NR4A TKO substantially suppressed the elevated expression levels of these exhaustion markers as determined by MFI following CAE (figure 2D). The PD-1+TIM3+ fraction post-CAE was also higher in NR4A TKO CAR-T cells as the age of the donors increased (online supplemental figure S4D).

    These findings collectively demonstrated that NR4A TKO consistently maintained the proliferative capacity and preserved the expression of early memory markers while attenuating the induction of exhaustion markers following sustained tumor antigen stimulation, irrespective of the donor source. Nevertheless, these effects exhibited some variability, particularly depending on donor age.

    NR4A TKO CAR-T cells had stronger antitumor potentials than control CAR-T cells in vitro

    Next, we examined the antitumor activity of Her2 CAR-T cells before and after the 14-day CAE. First, we sorted CD8+ CAR-T cells before and after CAE using the SH800S Cell Sorter (Sony Biotechnology, online supplemental figure S5A). The cell proliferative capacity was evaluated by labeling T cells with cell trace violet (figure 3A). Although there was no significant difference in the proliferative capacity between NR4A TKO and control CAR-T cells immediately after CAR transduction (pre-CAE), the proliferative capacity of NR4A TKO CAR-T cells of post-CAE was significantly higher than the control (p<0.01, post-CAE in figure 3A). The expression levels of TNF-α and IL-2 were not different between NR4A TKO and control CAR-T cells before CAE. Moreover, the IFN-γ levels were slightly higher in NR4A TKO CAR-T cells than in control CAR-T cells before CAE. The cytokine levels significantly decreased in both control and NR4A TKO CAR-T cells after CAE (p<0.0001). However, NR4A TKO CAR-T cells exhibited a superior ability to produce these cytokines compared with control cells (figure 3B).

    Figure 3
    Figure 3

    In vitro tumor cell-killing activity of the control and NR4A TKO CAR-T cells. (A) Cell division profile after antigen stimulation measured using the dilution of cell trace violet. Control or NR4A TKO CAR-T cells isolated before (Pre) and after (Post) continuous antigen exposure (CAE) were labeled with cell trace violet and then cultured with A549 cells for 3 days. The proliferation index is calculated based on the ratio of CellTrace Violet’s mean fluorescence intensity before and after stimulation. The data represents mean±SD regarding the CAR-T cells from different four donors (one 20s woman, one 30s man, one 40s man, and one 50s woman). (B) CAR-T cells isolated before (Pre) and after (Post) CAE were co-cultured with A549 cells for 24 hours. The concentration of TNF-α, IFN-γ, and IL-2 in the culture supernatant was measured using ELISA. The data represents mean±SD regarding the CAR-T cells from different four donors (one 20s woman, one 30s man, one 40s man, and one 50s woman). (C) Cytotoxicity of the control and NR4A TKO CAR-T cells before (Pre) and after (Post) CAE. Representative imaging of 549 cells stably expressing EGFP co-cultured with CAR-T cells for indicated periods are shown on the left. Real-time quantitative data of EGFP imaging using the IncuCyte system is shown on the right. Statistical analysis was performed regarding the control and NR4A TKO CAR-T cells before and after CAE with p values indicated on the graph. The data represents mean±SD regarding the CAR-T cells from different three donors (one 20s woman, one 30s man, and one 40s man). (D) Cytokine levels in the culture supernatant of the control and NR4A TKO CAR-T cells from two independent donors (one 30s man and 60s man) after (Post) CAE (data are presented as the means of n=4 replicate wells). (E) Cytotoxicity of the control and NR4A TKO CAR-T cells from two independent donors (one 30s man and 60s man) after (Post) CAE measured using the IncuCyte system. Data are representative of at least two independent experiments. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; ns, not significant; CAR, chimeric antigen receptor; EGFP, enhanced green fluorescent protein; IFN, interferon; IL, interleukin; TKO, triple knockout; TNF, tumor necrosis factor.

    To evaluate the in vitro killing activity, CAR-T cells were co-cultured with A549 cells labeled with trace violet and monitored using flow cytometry. The cytotoxic activity of NR4A TKO CAR-T cells was significantly higher than that of the control CAR-T cells after CAE (p<0.0001). The results remained consistent even when the E/T ratio was varied, demonstrating the robustness of the potent antitumor effects of NR4A TKO CAR-T cells (online supplemental figure S5B). The cytotoxic potential of NR4A TKO CAR-T cells was also analyzed in real-time using IncuCyte Imaging System. As shown in figure 3C, before CAE, co-culture with the control and NR4A TKO CAR-T cells reduced the number of GFP-labeled A549 cells, especially NR4A TKO CAR-T cells (purple) eradicated A549 cells more rapidly than control CAR-T cells (red). In contrast, after CAE, the control CAR-T cells failed to exert antitumor effects, and GFP-positive A549 cells exhibited strong proliferation (vermilion). However, NR4A TKO CAR-T cells strongly reduced the number of tumor cells (light blue). This indicated that NR4A TKO conferred strong antitumor activities on CAR-T cells immediately after gene manipulation and maintained this activity even after CAE. These results remained consistent even when varying the E/T ratio before and after CAE, and real-time analysis using the IncuCyte Imaging System also demonstrated the robust antitumor efficacy of NR4A TKO CAR T cells (online supplemental figure S5C, D).

    We also evaluated the antitumor activity of NR4A sKO, dKO, and TKO CAR-T cells after CAE. In correlation with the proliferative capacity during CAE illustrated in figure 1D, dKO resulted in higher antitumor activity than sKO, with NR4A TKO CAR-T cells showing the strongest cell-killing activity (online supplemental figure S5E).

    Next, we examined the enhancement of antitumor activity by NR4A TKO using two different typical donors. Donor 5 was older, and the generated CAR-T cells were more exhausted than those from donor 2 (figure 2 and online supplemental figure S4). Before CAE, the CAR-T cells derived from the older donor 5 (purple) exhibited tumor-killing activities; however, they were inferior to CAR-T cells derived from the younger donor 2 (vermilion) (online supplemental figure S5 F, G). After CAE, donor 5-derived CAR-T cells were severely exhausted and produced marginal amounts of IFN-γ and TNF-α (figure 3D). However, NR4A TKO CAR-T cells from donor 5 maintained high production levels of these cytokines, although not as high as those from donor 2 (figure 3D). Following the TKO of NR4A, donor 5-derived CAR-T cells exhibited sufficient killing activity, comparable to that of donor 2-derived CAR-T cells, even after CAE (figure 3E and online supplemental figure S5H). These experiments indicated that the deletion of NR4A suppressed exhaustion and improved the proliferative capacity and antitumor activity, regardless of the donor status.

    Energy production capacity was enhanced in NR4A TKO CAR-T cells and retained even in exhausted conditions

    We examined the mechanism underlying exhaustion resistance following NR4A TKO. Less-differentiated early memory T cells have been reported to exist in an oxidative phosphorylation (OXPHOS)-dominant, glycolysis-limited metabolic state, enhancing the survival and persistence of memory T cells.34 37 However, exhausted T cells are quiescent and exhibit low ATP synthesis activity. Moreover, the metabolic states of NR4A TKO CAR-T cells have not yet been investigated. We examined the OCRs and ECARs using a flux analyzer (figure 4A,B and representative flux data are shown in, online supplemental figure S6A, B). Before CAE, NR4A TKO CAR-T cells exhibited improved basal and maximal OCRs and spare respiratory capacity (SRC) compared with control CAR-T cells. SRC has been shown to reflect the mitochondrial metabolic potential linked to longevity and memory formation in T cells.38 Moreover, the basal ECARs were higher in NR4A TKO CAR-T cells than in control CAR-T cells (figure 4A). After CAE (post-CAE), the control CAR-T cells were quiescent, with basal and maximal OCRs, SRC, and glycolysis-dependent ECARs ((ECAR in the presence of oligomycin)—(ECAR at basal level)), all showing very low and weak responses to oligomycin. In contrast, NR4A TKO CAR-T cells exhibited significant OCR responses to oligomycin and FCCP, with higher glycolysis-dependent ECARs (figure 4A and online supplemental figure S6A, B). Based on the ATP production calculated using the OCRs and ECARs, NR4A TKO CAR-T cells had a higher ATP production capacity than control CAR-T cells, even before CAE (figure 4B). Control CAR-T cells significantly lost their ATP-producing ability (p<0.001), whereas NR4A TKO CAR-T cells maintained a level similar to that of the control pre-CAE CAR-T cells (figure 4B and online supplemental figure S6C). Thus, even after repeated antigen stimulation, NR4A depletion resulted in a higher metabolic potential for energy synthesis in CAR-T cells.

    Figure 4
    Figure 4

    NR4A TKO enhanced energy production and mitochondrial fitness in CAR-T cells. (A) Metabolic rates assessed using Seahorse analysis of the basal and maximal oxygen consumption rate (OCR), spare respiratory capacity (SRC), and basal extracellular acidification rate (ECAR) of the control or NR4A TKO CAR-T cells before (Pre) or after (Post) continuous antigen exposure (CAE). Representative actual measurement figures are shown in online supplemental figure S6A, B. Data are presented as the means of n=5 replicate wells from different five donors. (B) Mitochondria-derived and glycolysis-derived ATP production rates were calculated based on changes in the OCR and ECAR, respectively. The total ATP production rate is the sum of mitochondria-derived ATP and glycolytic ATP. Data are presented as the means of n=5 replicate wells. (C) Representative electron microscopy images of T cells and mitochondria of CAR-T cells before (Pre) and after (Post) CAE. Mitochondria were indicated by the red arrows, and cristae were indicated by the blue arrows. (D) Mitochondrial area and number of cristae per mitochondria of CAR-T cells measured from electron microscopy images. (E) The mean fluorescence intensity (MFI) was measured using flow cytometry for the mitochondrial activity (MitoTracker deep red) and reactive oxygen species (MitoSOX) in control CAR-T cells and NR4A TKO CAR-T cells before (Pre) and after (Post) CAE. Data are presented as the mean±SEM of different four donors (one 20s woman, one 30s man, one 40s man, and one 50s woman). Data are representative of at least two independent experiments. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; ns, not significant; one-way analysis of variance. CAR, chimeric antigen receptor; TKO, triple knockout.

    Next, we assessed the effects of NR4A depletion on the mitochondrial morphology of CAR-T cells before and after CAE. An increase in the mitochondrial area was observed in NR4A TKO CAR-T cells both pre-CAE and post-CAE. In addition, the density of the internal cristae (indicated by blue arrows) appeared to be higher in NR4A TKO CAR-T cells than in control CAR-T cells (figure 4C,D). After CAE, the mitochondrial shape of the control CAR-T cells was reduced, and the internal structure collapsed. However, in NR4A TKO CAR-T cells, both the mitochondrial size and structure were preserved even after CAE (figure 4C,D). The number of mitochondria per cell did not differ significantly between groups (online supplemental figure S6D).

    We further assessed the mitochondrial mass, activity, and reactive oxygen species (ROS) production using flow cytometry. Before CAE, NR4A TKO CAR-T cells exhibited higher mitochondrial membrane potentials and ROS production than the control CAR-T cells. After CAE, all these parameters in the control CAR-T cells were significantly decreased (p<0.0001), whereas NR4A TKO CAR-T cells maintained these parameters reasonably well (figure 4E and online supplemental figure S6E). Thus, NR4A TKO CAR-T cells exhibited mitochondrial structural changes and increased metabolic activity before and after CAE. As these changes were observed in pre-CAE CAR-T cells, NR4A depletion may affect mitochondria and improve metabolic function independent of TCR/CAR signaling.

    Gene expression profiles of NR4A TKO CAR-T cells

    To define the genetic characteristics of NR4A TKO CD8+CAR-T cells, total RNA-seq of NR4A TKO and control CD8+ CAR-T cells before and after CAE from three independent donors were performed. PC analysis revealed that the gene profiles of NR4A TKO and control CAR-T cells from three different donors showed changes before CAE (online supplemental figure S7A). However, the gene profiles of NR4A TKO cells in post-CAE conditions were apparently different from those of control CAR-T cells (figure 5A). This may be because NR4A accumulated after continuous TCR stimulation and contributed to the late stages of memory T-cell differentiation.

    Figure 5
    Figure 5

    NR4A TKO enhanced stemness-related and oxidative phosphorylation-related genes and reduced exhaustion-related genes. (A) Principal component (PC) analysis of RNA sequencing data for the control and NR4A TKO CAR-T cells 14 days after continuous antigen exposure (CAE) from three independent donors (one 30s man, one 40s man, and one 50s woman). (B) Volcano plot of differentially expressed genes in NR4A TKO CAR-T cells compared with control CAR-T cells 14 days after CAE. The average of the RNA-seq data of the CAR-T cells 14 days after CAE from three donors were analyzed. Transcripts with FDR<0.05 are highlighted in blue and red. Among them, stemness-related and exhaustion-related genes that were upregulated in NR4A TKO CAR-T cells are indicated. (C, D) Gene Ontology and Gene Set Enrichment Analysis of RNA-seq data and enrichment plot of naïve, memory, exhaustion and T-cell dysfunction associated gene of the CAR-T cells 14 days after CAE. (E) Heat map of the selected genes associated with the mitochondria, OXPHOS, memory, exhaustion, and effector of control and NR4A TKO CAR-T cells 14 days after CAE. CAR, chimeric antigen receptor; FDR, false discovery rate; NES, normalized enrichment score; NSCLC, non-small cell lung cancer; OXPHOS, oxidative phosphorylation; RNA-seq, RNA sequencing; TKO, triple knockout.

    Next, we analyzed the differences between NR4A TKO and control CD8+ CAR-T cells and generated volcano plots and heat maps. After CAE, the expressions of memory, effector-related, mitochondria, and OXPHOS-related genes were higher in NR4A TKO CAR-T cells than in controls, including SELL, IL2RA, IL2, IRF1, IRF8, GZMB, TNFSF11, IFNG, NDUFS6, NDUFB8, MRPL35, and ATP5MC1.28 34 39 Furthermore, NR4A TKO CAR-T cells expressed less exhaustion-associated genes, including PDCD1, LAG 3, TIGIT, CD38, ETNTPD1 (CD39), CTLA4, and TOX24 34 40 (figure 5B). Gene Ontology (GO) analysis of DEGs and GSEA demonstrated that the pathways associated with MYC targets V1 and V2 were related to T-cell activation and proliferation,41 42 GSE9650 (pathway of enhancing CD8 T-cell memory and effector function), MTORC1 signaling, and OXPHOS in NR4A TKO CAR-T cells (figure 5C–E). In addition, NR4A TKO CAR-T cells had distinctly lower expression of the gene set associated with CD8+T cell dysfunction (27(data set, accession number GSE160174)) . In GSEA using gene sets associated with stem-like memory and exhaustion identified in several studies (29(data set, accession number GSE176022) and 30(data set, accession number GSE99254)) aiming to analyze TILs in patients with NSCLC, it was revealed that NR4a TKO CAR-T cells also maintain stem-like memory function and evade exhaustion. These results suggest that NR4A TKO CAR-T cells would be resistant to exhaustion and retain high stemness in clinical applications.

    Although the deletion of NR4A in CAR-T cells before CAE was not altered in PC analysis, GO analysis demonstrated enhanced GSE22886 (expression profiles from various resting and activated human immune cells), MYC targets V1 and V2, and mitochondria-related pathways, such as MTORC1 signaling and mitochondrial inner membrane. GSEA revealed the enhancement of pathways associated with stem-like T cells, OXPHOS, and IFN-γ response-related pathways in NR4A TKO CAR-T cells (online supplemental figure S7B, C). Many genes related to OXPHOS were upregulated in NR4A TKO CAR-T cells before CAE (online supplemental figure S7D). These results may explain the differences in cytotoxicity (figure 3) and energy production status (figure 4) between NR4A TKO and control CAR-T cells before CAE.

    NR4A TKO CAR-T cells exhibited strong antitumor responses in vivo

    To validate the antitumor toxicity of NR4A TKO CAR-T cells in vivo, NR4A TKO and control HER2 CAR-T cells were transferred into NSG mice engrafted with A549 cells (figure 6A). The proportion of CD4+ and CD8+ CAR-T cells before CAR-T cell administration to mice was generally similar (online supplemental figure S8A). Tumors continuously grew in mice treated with non-transduced T cells and control CAR-T cells. However, the tumors of mice treated with NR4A TKO CAR-T cells showed drastic shrinkage 7 days after transfer (figure 6B). Over 40 days after subcutaneous implantation of the tumor cells, most mice injected with control CAR-T cells required patronizing sacrifices as the tumor volume exceeded 2,000 mm3. Nevertheless, the tumor volume remained consistently small in the mice treated with NR4A TKO CAR-T cells (figure 6C). There was no significant difference in the body weights of mice treated with control CAR-T cells and NR4A TKO CAR-T cells or those without T-cell treatment (online supplemental figure S8B). Mice that received NR4A TKO CAR-T cells survived significantly longer than those that received non-transduced T cells or control CAR-T cells (p=0.001, figure 6D). Furthermore, the tumors were completely eradicated in more than 50% of the mice treated with NR4A TKO CAR-T cells.

    Figure 6
    Figure 6

    In vivo antitumor effects of NR4A TKO CAR-T cells. (A) Schema of the tumor cell and CAR-T cell transplantation. NSG mice were inoculated with 1.0×106 A549 cells subcutaneously on the right side of their backs. Tumor growth was monitored using a digital caliper. Then, 5.0×106 CD8+ CAR-T cells were injected through the tail vein 2 weeks after tumor cell transplantation. (B) Average tumor growth (n=7 mice per group). Unpaired t-test on day 40. (C) Pictures of tumors in each group on day 40. (D) Kaplan-Meier survival plots regarding the treated mice (n=10 mice per group). (E) Schema regarding the analysis of TILs. NSG mice were inoculated with 1.0×106 A549 cells subcutaneously on the right side of their backs. Two weeks after tumor cell transplantation, 5.0×106 CD8+ CAR-T cells were injected through the tail vein. TILs were isolated and analyzed 1 week after the administration of CAR-T cells. (F) Pictures of tumors and average tumor size in each group on day 21. Unpaired t-test. (G) Flow cytometry analysis of CD4+ and CD8+ CAR-T cells in the TILs. The numbers of CD4+ and CD8+ T cells in the control and NR4A TKO CAR-T cells are shown. The data represents the mean±SD (n=5). (H) Representative flow cytometry plots of memory markers (CD28 and CD62L) in tumor-infiltrated CD8+CAR-T cells. Quantitative data of CD28+CD62L+ population are shown on the right. (I, J) Representative flow cytometry plots of the exhaustion markers. PD-1+LAG-3+TIM-3+CD39+ population and mean fluorescence intensity (MFI) of the indicated T-cell exhaustion markers in CD8+ CAR-T cells infiltrating the tumors on day 21. (K) Average tumor growth (n=5 mice per group) in the rechallenge model. Arrows on day 14 indicate administration of NR4A TKO CAR-T cells and arrows on day 42 indicate rechallenge of A549 cells. The data represents the mean±SD (n=5). The experiment was performed twice with CAR T cells generated from two different healthy donors (one 20s woman and one 30s man). *p<0.05; **p<0.01; CAR, chimeric antigen receptor; i.v., intravenous; s.c., subcutaneous; TIL, tumor-infiltrating lymphocyte; TKO, triple knockout.

    Next, we collected tumors 7 days after the injection of CAR-T cells into the mice and evaluated the cell surface markers of tumor-infiltrating CAR-T cells (figure 6E). At that time, the tumor volumes were smaller in mice treated with NR4A TKO CAR-T cells than in those treated with control CAR-T cells (figure 6F). The analysis of CAR-T cells infiltrating the tumors indicated that only a few CD4+ and CD8+ T cells were present in tumors treated with control CAR-T cells. However, both CD4+ and CD8+ positive T cells were present in large amounts in tumors treated with NR4A TKO CAR-T cells (figure 6G). In addition, the NR4A TKO CD8+ CAR-T cells that infiltrated into the tumors highly expressed early memory markers CD28 and CD62L (figure 6H), while they reduced the expression of checkpoint molecules, including PD-1, TIM-3, LAG3, and CD39 (figure 6I). Moreover, the MFI of these immune checkpoint molecules was significantly lower in NR4A TKO CD8+ CAR-T cells compared with control CD8+ CAR-T cells (p<0.01, figure 6J). Similarly, the analysis of the CD4+ CAR-T cells in the tumors revealed that the memory phenotype was retained, and the expression of exhaustion-related markers was reduced (online supplemental figure S8C–E). In the spleens of mice 7 days after the injection of CAR-T cells, few CAR-T cells remained in CAR-T-treated mice, while abundant NR4A TKO CAR-T cells were present (online supplemental figure S8F). The NR4A TKO CAR-T cells within the spleen, similar to tumor-infiltrating T cells, maintained a memory phenotype and reduced exhaustion markers compared with control CAR-T cells (online supplemental figure S8G–I).

    Finally, we conducted an experiment to evaluate the persistence of NR4A TKO CAR-T cells by tumor rechallenge. One month after NR4A TKO CAR-T cells administration, A549 cells were subcutaneously re-injected to NSG mice (online supplemental figure S9A). While untreated mice exhibited significant tumor growth, mice administered with NR4A TKO CAR-T cells showed no further tumor growth (figure 6K). There was no significant difference in body weight between the two groups, and many mice administered with NR4A TKO CAR-T cells achieved long-term survival (online supplemental figure S9B, C).

    These results indicated that NR4A TKO CAR-T cells were resistant to exhaustion and retained their memory function, resulting in potent long-term antitumor effects in vivo.

    Discussion

    In this study, we demonstrated that NR4A deletion in human CAR-T cells greatly improved their antitumor activity against solid tumors. The potent antitumor effect of NR4A TKO CAR-T cells is attributed to their lower expression of inhibitory receptors, high levels of effector cytokines, such as IFN-γ and TNF-α, and higher proliferation and energy production even after repeated antigen stimulation.

    NR4a has been implicated in Foxp3 induction and maintenance in CD4+ T cells.32 43 The role of NR4A in T-cell exhaustion has been investigated and elucidated by Rao et al using mouse models.25 Nr4a factors are induced by TCR signaling and bind to the regions of genes involved in T-cell exhaustion. Specifically, Nr4a factors directly bind to the enhancer region of the Pdcd1 gene and the promoter region of Havcr2 (TIM3), leading to the upregulation of these inhibitory receptors in collaboration with NFAT in murine T cells. Furthermore, Nr4a suppresses the expression of effector molecules, including cytokines and granzyme genes, as well as Il2ra, by inhibiting AP-1 and NF-kB activity.25 Notably, Nr4a may collaboratively contribute to T-cell exhaustion along with Tox and Tox2.26 In addtion, loss of Nr4a1 and NR4a2 have been shown to promote accumulation of TCF1+ stem-like precursors of exhausted CD8+ T cells in the tumor microenvironment in mice44 . However, the significance of NR4A factors in human T cells remains unclear.

    In our study, in human, NR4A TKO also conferred resistance to exhaustion of CAR-T cells induced by the repeated antigen stimulation. We found that NR4A TKO in human CAR-T cells maintained the expression of naïve markers even after repeated stimulation, indicating that NR4A deficiency confers stemness. This is consistent with a previous report by Odagiu et al that showed that Nr4a3 deletion in mice resulted in increased memory generation and effector function.45 They demonstrated that NR4A3 deficiency contributes to an early effect on the transcriptional memory program of CD8+ T cells and the chromatin accessibility of the bZIP transcription factor. Importantly, however, our study demonstrated that NR4As are not redundant; the loss of NR4A3 alone is insufficient, and the triple loss of NR4A1/2/3 can confer the strongest stemness and resistance to exhaustion. In Jung et al’s study,20 it was reported that BLIMP1/NR4A3 dKO CAR-T cells maintained stemness and enhanced resistance to exhaustion. In our study, we compared NR4A TKO CAR-T cells with BLIMP1/NR4A3 dKO CAR-T cells in the CAE model, suggesting that NR4A TKO is much better than BLIMP1/NR4A3 dKO in terms of proliferation, expression of early memory markers and resistance to exhaustion during CAE. It is too early to draw conclusions because of the different conditions, including CAR constructs in Jung et al’s study. However, it is worth noting that NR4A TKO can confer stemness without further genetic manipulations. Additionally, from the investigation of our study, it is anticipated that knocking out NR4A1, 2, 3, and PRDM1 altogether could lead to further enhancement of proliferative capacity and antitumor effects. However, further careful study is necessary, since this would increase the complexity of the experiments with the addition of more gRNAs, potentially leading to a decrease in knockout efficiency for each gene.

    NR4A has previously been reported to be involved in the induction of Foxp3 and suppression of proliferation and cytokine production in CD4+ T cells. Further, the deletion of NR4A results in decreased Treg levels and increased effector CD4+ T cells.31 Our study suggests that the deletion of NR4A also affects CD4+ T cells. In our exhaustion-induced and murine xenograft models, control CD4+ CAR-T cells were rapidly depleted, whereas NR4A TKO CAR-T cells exhibited long-term survival of effector CD4+ T cells after CAR activation. Control CAR-T cells also exhibited an increase in Foxp3+ cells after repeated stimulation; however, no such increase was observed in NR4A TKO CAR-T cells. This suggests that NR4A TKO enhances antitumor immunity in CD8+ and CD4+ T cells.

    One of the striking features of NR4A TKO CAR-T cells is their high ATP-producing capacity, even under exhaustion-induced conditions. NR4A receptors are primarily known for their involvement in gene regulation; however, there is emerging evidence suggesting a link to energy metabolism.46 NR4A has been implicated in regulating OXPHOS, fatty acid oxidation, and glucose metabolism. However, there are conflicting reports regarding the effects of NR4A transcription factors on mitochondrial biogenesis, mitochondrial respiration, and oxidative stress responses. For example, the deletion of NR4A1 and NR4A3 in pancreatic beta cells inhibits mitochondrial respiration.47 Conversely, the loss of Nr4a1 in helper T cells promotes OXPHOS and glycolysis.48 Similarly, in macrophages, the loss of Nr4a1 is associated with enhanced glucose metabolism through the regulation of isocitrate dehydrogenase.49 Recent studies have suggested that NR4A1 localizes to the mitochondria and promotes mitophagy.50 Collectively, these studies indicate that NR4A exerts different metabolic effects in different cell types and environmental contexts. Previous studies have reported that sustained antigen stimulation or exposure to hypoxic conditions within the tumor microenvironment leads to increased production of ROS in TILs resulting in decreased mitochondrial capacity and an increase in the number of mitochondria within TILs.51 52 Our study demonstrated that the loss of NR4A promotes an increase and maintenance of mitochondrial functions and renews the glycolytic system in human T cells. We suspect that NR4A directly downregulates mitochondrial and glycolysis genes, therefore loss of NR4A enhances OXPHOS and glycolysis. However, we could rule out the possibility that these are indirect effects by gaining the stemness. Thus, further studies are necessary to understand how the entire mitochondrial and glycolytic pathways are upregulated in the absence of NR4A.

    This study had several limitations. First, although NR4A TKO CAR-T cells demonstrated excellent antitumor efficacy regardless of age, the fact that there was only one donor per age group may not fully represent the target population. To demonstrate the efficacy of NR4A TKO CAR-T cells in old patients, it is necessary to either conduct experiments with older donors or increase the number of donors in the higher age groups. Second, we did not conduct experiments with matched CD4/8 ratios during the assays. In online supplemental figure S2G, we demonstrated that the loss of NR4A reduces FOXP3 expression and decreases the proportion of CAR-Tregs, which is also expected to contribute to the enhancement of the antitumor effect of NR4A TKO CAR-T cells. To more accurately evaluate the improvement of CD8+ CAR-T cells and CAR-Tregs for antitumor effects, the CD4/CD8 ratio should be adjusted.

    Overall, our findings demonstrated that deletions in the NR4A family can enhance the efficacy and durability of CAR-T cell therapy, which is an important clinical issue to overcome when establishing the efficacy of CAR-T cell therapy for solid tumors.

    Data availability statement

    Data are available in a public, open access repository. Data are available upon reasonable request. The RNA sequencing data supporting the findings of this study have been deposited in the DNA Data Bank of the Japan Sequence Read Archive (https://www.ddbj.nig.ac.jp/index.html) with the accession numbers GSE241456.

    Ethics statements

    Patient consent for publication

    Ethics approval

    All human studies were approved by the Institutional Review Board of Keio University School of Medicine (Tokyo, Japan; approval number 20120039), and written informed consent was obtained from all participants. All experiments involving mice were approved by the Institutional Animal Care and Use Committee (IACUC; approval number, 08004) of Keio University (Tokyo, Japan) and performed according to the IACUC guidelines. Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    We would like to thank Yasuko Hirata, Yukiko Tokifuji, Mari Ikeda, Shunsuke Chikuma, Shiro Otake, Shizuko Kagawa, Chika Yonekawa, and Junko Hamamoto (Keio University, Tokyo, Japan) for their technical support.

    References

    1. ↵
      1. Maude SL,
      2. Frey N,
      3. Shaw PA, et al
      . Chimeric antigen receptor T cells for sustained remissions in leukemia. N Engl J Med 2014;371:1507–17.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Ahmed N,
      2. Brawley VS,
      3. Hegde M, et al
      . Human epidermal growth factor receptor 2 (HER2) -specific chimeric antigen receptor-modified T cells for the immunotherapy of HER2-positive sarcoma. J Clin Oncol 2015;33:1688–96. doi:10.1200/JCO.2014.58.0225
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Martinez M,
      2. Moon EK
      . CAR T cells for solid tumors: new strategies for finding, infiltrating, and surviving in the tumor microenvironment. Front Immunol 2019;10:128. doi:10.3389/fimmu.2019.00128
    4. ↵
      1. Drent E,
      2. Themeli M,
      3. Poels R, et al
      . A rational strategy for reducing on-target off-tumor effects of CD38-chimeric antigen receptors by affinity optimization. Mol Ther 2017;25:1946–58. doi:10.1016/j.ymthe.2017.04.024
      OpenUrlCrossRefPubMed
    5. ↵
      1. Spranger S,
      2. Bao R,
      3. Gajewski TF
      . Melanoma-intrinsic β-catenin signalling prevents anti-tumour immunity. Nature 2015;523:231–5. doi:10.1038/nature14404
      OpenUrlCrossRefPubMed
    6. ↵
      1. Spranger S,
      2. Dai D,
      3. Horton B, et al
      . Tumor-residing batf3 dendritic cells are required for effector T cell trafficking and adoptive T cell therapy. Cancer Cell 2017;31:711–23. doi:10.1016/j.ccell.2017.04.003
      OpenUrlCrossRefPubMed
    7. ↵
      1. Wei J,
      2. Long L,
      3. Zheng W, et al
      . Targeting REGNASE-1 programs long-lived effector T cells for cancer therapy. Nature 2019;576:471–6. doi:10.1038/s41586-019-1821-z
      OpenUrlCrossRef
    8. ↵
      1. Vardhana SA,
      2. Hwee MA,
      3. Berisa M, et al
      . Impaired mitochondrial oxidative phosphorylation limits the self-renewal of T cells exposed to persistent antigen. Nat Immunol 2020;21:1022–33. doi:10.1038/s41590-020-0725-2
      OpenUrl
    9. ↵
      1. Poggi A,
      2. Varesano S,
      3. Zocchi MR
      . How to hit mesenchymal stromal cells and make the tumor microenvironment immunostimulant rather than immunosuppressive. Front Immunol 2018;9:262. doi:10.3389/fimmu.2018.00262
    10. ↵
      1. Plitas G,
      2. Konopacki C,
      3. Wu K, et al
      . Regulatory T cells exhibit distinct features in human breast cancer. Immunity 2016;45:1122–34. doi:10.1016/j.immuni.2016.10.032
      OpenUrlCrossRefPubMed
    11. ↵
      1. Kouro T,
      2. Himuro H,
      3. Sasada T
      . Exhaustion of CAR T cells: potential causes and solutions. J Transl Med 2022;20:239. doi:10.1186/s12967-022-03442-3
    12. ↵
      1. Renner K,
      2. Singer K,
      3. Koehl GE, et al
      . Metabolic hallmarks of tumor and immune cells in the tumor microenvironment. Front Immunol 2017;8:248. doi:10.3389/fimmu.2017.00248
    13. ↵
      1. Speiser DE,
      2. Utzschneider DT,
      3. Oberle SG, et al
      . T cell differentiation in chronic infection and cancer: functional adaptation or exhaustion? Nat Rev Immunol 2014;14:768–74. doi:10.1038/nri3740
      OpenUrlCrossRefPubMed
    14. ↵
      1. Kallies A,
      2. Zehn D,
      3. Utzschneider DT
      . Precursor exhausted T cells: key to successful immunotherapy? Nat Rev Immunol 2020;20:128–36. doi:10.1038/s41577-019-0223-7
      OpenUrlPubMed
    15. ↵
      1. Zehn D,
      2. Thimme R,
      3. Lugli E, et al
      . 'Stem-like’ precursors are the fount to sustain persistent CD8(+) T cell responses. Nat Immunol 2022;23:836–47. doi:10.1038/s41590-022-01219-w
      OpenUrlCrossRef
    16. ↵
      1. Gupta PK,
      2. Godec J,
      3. Wolski D, et al
      . CD39 expression identifies terminally exhausted CD8+ T cells. PLoS Pathog 2015;11:e1005177. doi:10.1371/journal.ppat.1005177
    17. ↵
      1. Seo H,
      2. González-Avalos E,
      3. Zhang W, et al
      . BATF and IRF4 cooperate to counter exhaustion in tumor-infiltrating CAR T cells. Nat Immunol 2021;22:983–95. doi:10.1038/s41590-021-00964-8
      OpenUrlCrossRef
    18. ↵
      1. Seo W,
      2. Jerin C,
      3. Nishikawa H
      . Transcriptional regulatory network for the establishment of CD8(+) T cell exhaustion. Exp Mol Med 2021;53:202–9. doi:10.1038/s12276-021-00568-0
      OpenUrlCrossRefPubMed
    19. ↵
      1. Yoshikawa T,
      2. Wu Z,
      3. Inoue S, et al
      . Genetic ablation of PRDM1 in antitumor T cells enhances therapeutic efficacy of adoptive immunotherapy. Blood 2022;139:2156–72. doi:10.1182/blood.2021012714
      OpenUrlCrossRef
    20. ↵
      1. Jung I-Y,
      2. Narayan V,
      3. McDonald S, et al
      . BLIMP1 and NR4A3 transcription factors reciprocally regulate antitumor CAR T cell stemness and exhaustion. Sci Transl Med 2022;14:eabn7336. doi:10.1126/scitranslmed.abn7336
    21. ↵
      1. Battistello E,
      2. Hixon KA,
      3. Comstock DE, et al
      . Stepwise activities of mSWI/SNF family chromatin remodeling complexes direct T cell activation and exhaustion. Mol Cell 2023;83:1216–36. doi:10.1016/j.molcel.2023.02.026
      OpenUrlCrossRefPubMed
    22. ↵
      1. Guo A,
      2. Huang H,
      3. Zhu Z, et al
      . cBAF complex components and MYC cooperate early in CD8(+) T cell fate. Nature 2022;607:135–41. doi:10.1038/s41586-022-04849-0
      OpenUrlCrossRef
    23. ↵
      1. Zhang X,
      2. Zhang C,
      3. Qiao M, et al
      . Depletion of BATF in CAR-T cells enhances antitumor activity by inducing resistance against exhaustion and formation of central memory cells. Cancer Cell 2022;40:1407–22. doi:10.1016/j.ccell.2022.09.013
      OpenUrlCrossRef
    24. ↵
      1. Chen J,
      2. López-Moyado IF,
      3. Seo H, et al
      . NR4A transcription factors limit CAR T cell function in solid tumours. Nature 2019;567:530–4. doi:10.1038/s41586-019-0985-x
      OpenUrlPubMed
    25. ↵
      1. Liu X,
      2. Wang Y,
      3. Lu H, et al
      . Genome-wide analysis identifies NR4A1 as a key mediator of T cell dysfunction. Nature 2019;567:525–9. doi:10.1038/s41586-019-0979-8
      OpenUrlPubMed
    26. ↵
      1. Seo H,
      2. Chen J,
      3. González-Avalos E, et al
      . TOX and TOX2 transcription factors cooperate with NR4A transcription factors to impose CD8(+) T cell exhaustion. Proc Natl Acad Sci U S A 2019;116:12410–5. doi:10.1073/pnas.1905675116
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Good CR,
      2. Aznar MA,
      3. Kuramitsu S, et al
      . An NK-like CAR T cell transition in CAR T cell dysfunction. Cell 2021;184:6081–100. doi:10.1016/j.cell.2021.11.016
      OpenUrlCrossRefPubMed
    28. ↵
      1. Kondo T,
      2. Ando M,
      3. Nagai N, et al
      . The NOTCH-FOXM1 axis plays a key role in mitochondrial biogenesis in the induction of human stem cell memory-like CAR-T cells. Cancer Res 2020;80:471–83. doi:10.1158/0008-5472.CAN-19-1196
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Caushi JX,
      2. Zhang J,
      3. Ji Z, et al
      . Transcriptional programs of neoantigen-specific TIL in anti-PD-1-treated lung cancers. Nature 2021;596:126–32. doi:10.1038/s41586-021-03752-4
      OpenUrlPubMed
    30. ↵
      1. Guo X,
      2. Zhang Y,
      3. Zheng L, et al
      . Global characterization of T cells in non-small-cell lung cancer by single-cell sequencing. Nat Med 2018;24:978–85. doi:10.1038/s41591-018-0045-3
      OpenUrlPubMed
    31. ↵
      1. Sekiya T,
      2. Kagawa S,
      3. Masaki K, et al
      . Regulation of peripheral Th/Treg differentiation and suppression of airway inflammation by Nr4a transcription factors. i Sci 2021;24:102166. doi:10.1016/j.isci.2021.102166
    32. ↵
      1. Sekiya T,
      2. Kashiwagi I,
      3. Yoshida R, et al
      . Nr4a receptors are essential for thymic regulatory T cell development and immune homeostasis. Nat Immunol 2013;14:230–7. doi:10.1038/ni.2520
      OpenUrlCrossRefPubMed
    33. ↵
      1. Nikolich-Žugich J
      . Aging of the T cell compartment in mice and humans: from no naive expectations to foggy memories. J Immunol 2014;193:2622–9. doi:10.4049/jimmunol.1401174
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Ando M,
      2. Kondo T,
      3. Tomisato W, et al
      . Rejuvenating effector/exhausted CAR T cells to stem cell memory-like CAR T cells by resting them in the presence of CXCL12 and the NOTCH ligand. Cancer Res Commun 2021;1:41–55. doi:10.1158/2767-9764.CRC-21-0034
      OpenUrl
    35. ↵
      1. Chihara D,
      2. Liao L,
      3. Tkacz J, et al
      . Real-world experience of CAR T-cell therapy in older patients with relapsed/refractory diffuse large B-cell lymphoma. Blood 2023;142:1047–55. doi:10.1182/blood.2023020197
      OpenUrl
    36. ↵
      1. Noll JH,
      2. Levine BL,
      3. June CH, et al
      . Beyond youth: understanding CAR T cell fitness in the context of immunological aging. Semin Immunol 2023;70:101840. doi:10.1016/j.smim.2023.101840
      OpenUrl
    37. ↵
      1. Pilipow K,
      2. Scamardella E,
      3. Puccio S, et al
      . Antioxidant metabolism regulates CD8+ T memory stem cell formation and antitumor immunity. JCI Insight 2018;3:e122299. doi:10.1172/jci.insight.122299
    38. ↵
      1. van der Windt GJW,
      2. Everts B,
      3. Chang C-H, et al
      . Mitochondrial respiratory capacity is a critical regulator of CD8+ T cell memory development. Immunity 2012;36:68–78. doi:10.1016/j.immuni.2011.12.007
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      1. Simeonov DR,
      2. Gowen BG,
      3. Boontanrart M, et al
      . Discovery of stimulation-responsive immune enhancers with CRISPR activation. Nature 2017;549:111–5. doi:10.1038/nature23875
      OpenUrlCrossRefPubMed
    40. ↵
      1. Galletti G,
      2. De Simone G,
      3. Mazza EMC, et al
      . Two subsets of stem-like CD8(+) memory T cell progenitors with distinct fate commitments in humans. Nat Immunol 2020;21:1552–62. doi:10.1038/s41590-020-0791-5
      OpenUrlPubMed
    41. ↵
      1. Preston GC,
      2. Sinclair LV,
      3. Kaskar A, et al
      . Single cell tuning of Myc expression by antigen receptor signal strength and interleukin-2 in T lymphocytes. EMBO J 2015;34:2008–24. doi:10.15252/embj.201490252
      OpenUrlAbstract/FREE Full Text
    42. ↵
      1. Liberzon A,
      2. Birger C,
      3. Thorvaldsdóttir H, et al
      . The molecular signatures database (MSigDB) hallmark gene set collection. Cell Syst 2015;1:417–25. doi:10.1016/j.cels.2015.12.004
      OpenUrlCrossRef
    43. ↵
      1. Sekiya T,
      2. Kondo T,
      3. Shichita T, et al
      . Suppression of Th2 and Tfh immune reactions by Nr4a receptors in mature T reg cells. J Exp Med 2015;212:1623–40. doi:10.1084/jem.20142088
      OpenUrlAbstract/FREE Full Text
    44. ↵
      1. Srirat T,
      2. Hayakawa T,
      3. Mise-Omata S, et al
      . NR4a1/2 deletion promotes accumulation of TCF1+ stem-like precursors of exhausted CD8+ T cells in the tumor microenvironment. Cell Rep 2024;43:113898. doi:10.1016/j.celrep.2024.113898
    45. ↵
      1. Odagiu L,
      2. Boulet S,
      3. Maurice De Sousa D, et al
      . Early programming of CD8(+) T cell response by the orphan nuclear receptor NR4A3. Proc Natl Acad Sci U S A 2020;117:24392–402. doi:10.1073/pnas.2007224117
      OpenUrlAbstract/FREE Full Text
    46. ↵
      1. Herring JA,
      2. Elison WS,
      3. Tessem JS
      . Function of Nr4a orphan nuclear receptors in proliferation, apoptosis and fuel utilization across tissues. Cells 2019;8:1373. doi:10.3390/cells8111373
    47. ↵
      1. Reynolds MS,
      2. Hancock CR,
      3. Ray JD, et al
      . β-cell deletion of Nr4a1 and Nr4a3 nuclear receptors impedes mitochondrial respiration and insulin secretion. Am J Physiol Endocrinol Metab 2016;311:E186–201. doi:10.1152/ajpendo.00022.2016
      OpenUrlCrossRefPubMed
    48. ↵
      1. Liebmann M,
      2. Hucke S,
      3. Koch K, et al
      . Nur77 serves as a molecular brake of the metabolic switch during T cell activation to restrict autoimmunity. Proc Natl Acad Sci USA 2018;115:E8017–26. doi:10.1073/pnas.1721049115
      OpenUrlAbstract/FREE Full Text
    49. ↵
      1. Koenis DS,
      2. Medzikovic L,
      3. van Loenen PB, et al
      . Nuclear receptor Nur77 limits the macrophage inflammatory response through transcriptional reprogramming of mitochondrial metabolism. Cell Rep 2018;24:2127–40. doi:10.1016/j.celrep.2018.07.065
      OpenUrlCrossRefPubMed
    50. ↵
      1. Peng S-Z,
      2. Chen X-H,
      3. Chen S-J, et al
      . Phase separation of Nur77 mediates celastrol-induced mitophagy by promoting the liquidity of p62/SQSTM1 condensates. Nat Commun 2021;12:5989. doi:10.1038/s41467-021-26295-8
    51. ↵
      1. Scharping NE,
      2. Menk AV,
      3. Moreci RS, et al
      . The tumor microenvironment represses T cell mitochondrial biogenesis to drive intratumoral T cell metabolic insufficiency and dysfunction. Immunity 2016;45:374. doi:10.1016/j.immuni.2016.07.009
      OpenUrlCrossRefPubMed
    52. ↵
      1. Yu Y-R,
      2. Imrichova H,
      3. Wang H, et al
      . Disturbed mitochondrial dynamics in CD8+ TILs reinforce T cell exhaustion. Nat Immunol 2020;21:1540. doi:10.1038/s41590-020-0793-3
      OpenUrlCrossRefPubMed

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors KN performed conceptualization, data collection and analysis, funding acquisition, validation, investigation, and writing of the original draft. MA, TS, and SM-O developed the methods. TH and MI collected RNA sequencing data. KF and AY supervised the study. AY wrote the manuscript. AY as the guarantor for the overall content, approved and supervised the project. All authors read and approved the final version of the manuscript.

    • Funding This work was supported by KAKENHI 21H05044, 22K1944, 23K06723, 23H04785, 21H02719, the Japan Agency for Medical Research and Development (AMED)-CREST JP23gm1110009, AMED-Moonshot JP23zf0127003 AMED-PRIME 22gm6210012 (for MI), Chemo-Sero-Therapeutic Research Institute, Takeda Science Foundation, Uehara Memorial Foundation, the Princess Takamatsu Cancer Research Fund, the Yasuda Medical Foundation, and Keio Gijuku Academic Developmental Funds.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.