Mitochondrial fusion protein 2 (MFN2) and its variants in immunotherapy

Abstract

The present invention belongs to the field of immunotherapy, and more specifically, the present invention relates to mitochondrial fusion protein 2 (MFN2), MFN2 variants capable of interacting with SERCA2, or MFN2 expression promoters in maintaining and / or promoting CD8 + Use of CD8 T cells overexpressing MFN2 or its variants in tumor killing and / or survival + T cells exhibit better adaptive metabolism, higher IFN-γ production levels, stronger tumor killing ability and better survival in the process of cancer treatment.

Description

Technical Field

[0001] The present invention belongs to the field of immunotherapy, and more specifically, the present invention relates to the role of mitochondrial fusion protein 2 (MFN2), MFN2 variants, or MFN2 expression promoters in maintaining and / or promoting CD8 + Use of T cells in tumor killing and / or survival. Background Art

[0002] Although cancer immunotherapy using adoptive transfer of tumor-reactive tumor-infiltrating lymphocytes (TILs) and immune checkpoint blockade (ICB) has been widely used, sustained and complete responses have been observed in only a small number of clinical cases, particularly in solid malignancies (Rosenberg and Restifo, 2015; Wolchok, 2021). Loss of T cell effector function within the tumor microenvironment (TME) is one of the main reasons for the failure of TIL and ICB immunotherapy (Hegde and Chen, 2020). Emerging evidence emphasizes the importance of metabolic adaptability in programming T cell function and fate (Bantug et al., 2018b; DePeaux and Delgoffe, 2021). Competition with tumor cells for nutrients limits the metabolic capacity of TILs, leading to immune evasion (Chang et al., 2015). For TILs under hypoglycemic conditions within the tumor microenvironment, fatty acid oxidation (FAO)-driven oxidative phosphorylation (OXPHOS) is crucial for their effector function and survival (Hamanaka and Chandel, 2012; Zhang et al., 2017). Therefore, metabolic remodeling of TILs is a promising strategy to enhance the clinical efficacy of T cell-based immunotherapy.

[0003] Mitochondria fusion proteins (MFNs) are dynamin-like GTPases responsible for mitochondrial fusion, a fundamental event that enhances OXPHOS capacity (Labbe et al., 2014; Youle and van der Bliek, 2012). Mammalian MFNs have two MFNs, MFN1 and MFN2, which share 80% sequence similarity and have a degree of complementarity in catalyzing mitochondrial outer membrane fusion (Eura et al., 2003; Gao and Hu, 2021). Mitochondria (mito)-endoplasmic reticulum (ER) contacts allow Ca to flow through the mitochondria. 2+MFN2 flows from the ER to the mitochondria, activating key enzymes of the Krebs cycle (also known as the tricarboxylic acid cycle or citric acid cycle) to regulate mitochondrial ATP production (Jouaville et al., 1999). Overall, MFN2 functions as a metabolic hub by controlling mitochondrial behavior (Schrepfer and Scorano, 2016). Mutations or abnormal expression of MFN2 have been linked to the onset of various human diseases, including neuromuscular disorders, diabetes, and cancer (Filadi et al., 2018).

[0004] In ex vivo models, enhancing mitochondrial fusion in effector T cells imposes memory T cell characteristics and promotes anti-tumor capacity (Buck et al., 2016). The importance of balanced mitochondrial dynamics for tumor-infiltrating NK cells has also been demonstrated (Zheng et al., 2019). + Mitochondrial depolarization and mitochondrial autophagy in TILs can lead to an exhaustion phenotype (Yu et al., 2020). However, existing studies and literature have limited understanding of the CD8 + The nature of mitochondrial fusion in T cells remains poorly understood, and the function and regulatory mechanisms of mitochondria-endoplasmic reticulum contacts in CD8w T cells remain poorly understood.

[0005] Therefore, this field urgently needs to + Further study of the function and regulatory mechanism of mitochondria-endoplasmic reticulum contact in T cells will provide methods to improve the effector function and metabolic adaptability of CD8wT cells to enhance CD8 + The effects of T cells in cancer immunotherapy. Summary of the Invention

[0006] As mentioned above, there is an urgent need in the art to provide an improved CD8 + Approaches to T cell effector function and metabolic adaptation.

[0007] The present inventors unexpectedly discovered that mitochondrial fusion protein 2 (MFN2) plays an important role in the expression of functionally active CD8 + TILs were upregulated, and CD8 + High MFN2 levels in TILs are positively correlated with the prognosis of various solid malignancies (such as melanoma and renal clear cell carcinoma). The present inventors also found that MFN2 can regulate the prognosis of TILs by interacting with SERCA2 (a Ca2+ receptor) located in the endoplasmic reticulum (ER). 2+ ATPase) to mediate mitochondrial-endoplasmic reticulum contact to protect mitochondrial Ca 2+ homeostasis, ultimately promoting CD8 + Metabolic adaptability and effector function of TIL; through CD8+ Targeting MFN2 in T cells to enhance mitochondria-endoplasmic reticulum contacts can improve CD8 + The inventors completed the present invention based on the above findings.

[0008] Therefore, in the first aspect, the present invention provides mitochondrial fusion protein 2 (MFN2), a MFN2 variant capable of interacting with SERCA2, or a MFN2 expression promoter in maintaining and / or promoting CD8 + Use of T cells in tumor killing and / or survival.

[0009] In a second aspect, the present invention provides a CD8 + Use of T cells in the preparation of cell therapeutic agents for adoptive cellular immunotherapy.

[0010] In a third aspect, the present invention provides a MFN2 variant capable of interacting with SERCA2, comprising mutations in one or more of R259, V69, L76, R280 and W740.

[0011] In a fourth aspect, the present invention provides a method for treating cancer, comprising: administering to a cancer patient a CD8 T cell that overexpresses MFN2 or overexpresses a MFN2 variant that can interact with SERCA2; + T cells, or administering a MFN2 expression promoter to the cancer patient.

[0012] The beneficial effects of the present invention are: by studying the effect of mitochondrial fusion protein 2 (MFN2) on CD8 + The effects of MFN2, MFN2 variants, or MFN2 expression promoters on the metabolic adaptability and effector function of T cells are provided. + Use of CD8 T cells overexpressing MFN2 or its variants in tumor killing and / or survival + T cells exhibit higher mitochondrial metabolism, higher IFN-γ production levels, stronger tumor killing ability and better survival during cancer treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other implementation plans can be obtained based on these drawings without paying any creative work.

[0014] Figure 1 CD8 in the .TME + MFN2 is upregulated in T cells upon activation and correlates with better effector function, OXPHOS, and patient survival. (A) Representative images of immunohistochemical staining of CD8 and MFN2 in serial sections from human ccRCC cancer samples (n = 116 patients). + TILs were divided into high (strong staining) or low (weak staining) MFN2 expression groups (MFN hi and MFN lo ), scale bar is 100 μm; (B) with MFN hi (n=45) or MFN lo (n=61) Intratumoral CD8 + Kaplan-Meier survival curves of overall survival (left) and disease-free survival (right) in patients with T-cell ccRCC (n=116); (C) MFN hi and MFN lo Intratumoral CD8 in human ccRCC samples (n=116) between groups + Comparison of T cell numbers; (D) Kaplan-Meier survival curves of melanoma patients (n=32) for overall survival, according to MFN2 in CD8 + Expression classification in TIL (n=10, MFN2 hi ; n = 22, MFN2 lo ); (E) CD8 + Relative expression of MFN2 in TILs (n=32); (F) Immunofluorescence analysis of CD8 T cells isolated from human ccRCC samples. + Correlation between MFN2 expression and IFN-γ levels in TILs. Each point represents the fluorescence intensity of MFN2 per cell correlated with the fluorescence intensity of IFN-γ. Pearson correlation coefficient (R) and p value are shown (n = 3 ccRCC patients); (G) CD8 + Correlation between MFN2 mRNA levels and the abundance of ATP5A (left) or CPT1A (right) in TILs. Pearson correlation coefficient (R) and p-value are shown (n = 15 ccRCC patients); (H) CD8 + Representative images (left) and quantification (right) of MFN2 expression and cleaved caspase 3 in TILs (n=7 ccRCC patients).

[0015] Figure 2 MFN2 on CD8 +T cells are crucial for anti-tumor function and mitochondrial metabolism in vivo, including: (A) subcutaneous injection of 4×10 5 WT and Mfn2 in B16 melanoma cells flox / fllox CD4 Cre (Mfn2 CKO , referred to as CKO) mice (n=5 mice / group); (B) Tumor growth at day 14 from WT and Mfn2 CKO IFN-γ isolated from B16 tumor-bearing mice (n=5 mice / group) + CD8 + Representative flow cytometry plots of T cells (left) and IFN-γ + Percentage of CD8w T cells (right); (C) Percentage of CD8w T cells from WT and Mfn2 cells on day 14 CKO CD8 isolated from B16 tumor-bearing mice (n=3 mice / group) + Ki67 in TILs + The percentage of proliferating cells; (D) The percentage of proliferating cells from WT and Mfn2 cells on day 14 CKO Apoptotic (Annexin V) isolated from B16 tumor-bearing mice (n=3 mice / group) + ) Spleen and tumor-infiltrating CD8 + Percentage of T cells; (E and F) WT and Mfn2 cells were treated with anti-CD8 antibody 1 day before tumor injection and every 3 days thereafter (4 injections in total) CKO Tumor growth (E) and survival curves (F) of B16 tumor-bearing mice (n = 5 mice / group); (G and H) WT and Mfn2 treated with anti-PD-1 antibody on days 4, 7, 10, and 13. CKO Tumor growth (G) and survival curves (H) of B16 tumor-bearing mice (n=5 mice / group); (I) Mfn2 isolated from corresponding B16 tumor-bearing mice (n=3 mice / group) on day 14 - / - CD8+ T cells and WT CD8 + Volcano plot of differentially expressed genes in T cells, where P.adj represents the adjusted p-value, No sig represents no significant difference, up represents up-regulation, and down represents down-regulation; (J) Tumor-infiltrating Mfn2 - / - CD8 + T cells and WT CD8 + Gene ontology (GO) enrichment analysis of DEGs between T cells, including -log 10 (Adjusted p-value) >2 was used as cutoff value; (K) tumor-infiltrating Mfn2 - / - CD8 + T cells and WT CD8 +Gene set variation analysis (GSVA) ​​of downregulated (green) and upregulated (blue-gray) pathways between T cells, using an adjusted p-value < 0.05 as a cutoff; (L) Activated Mfn2 was measured using an extracellular flux analyzer. - / - CD8 + T and WT CD8 + Oxygen consumption rate (OCR) of T cells injected with oligomycin (Oligo), FCCP, and rotenone + antimycin A (R / A) at the indicated time points; (M) OCR of T cells from WT and Mfn2 cells measured by flow cytometry targeting BODIPY 500 CKO CD8 isolated from B16 tumor-bearing mice (n=3 mice / group) + Representative histograms (left) and MFI (right) of fatty acid metabolism in TILs, where MFI represents mean fluorescence intensity; the data in the above graphs are expressed as mean ± SD, and the data were analyzed by unpaired two-tailed Student's t test (A, B, C, D, E, G, M) or log-rank test (F, H). ** p<0.01, *** p<0.005.

[0016] Figure 3 Shows Mfn 2 C KO Generation and characterization of mice with: (A) Schematic representation of the CAS9-targeted region in Mfn2; (B) T cell-specific MFN2 knockout (Mfn2 flox / flox CD4 Cre or Mfn2 CKO ) Schematic diagram of C57BL / 6 mice (left), and representative confocal images on the right show the expression of WT or Mfn2 CKO CD8 isolated from mouse spleen + MFN2 expression in T cells, where the scale bar represents 20 μm; (C) 6×10 5 WT and Mfn2 in MC38 colon cancer cells CKO Tumor growth in mice (n=5 mice / group); (D) Tumor growth in WT and Mfn2 mice at day 21 CKO IFN-γ isolated from MC38 tumor-bearing mice (n=5 mice / group) + The percentage of CD8+ T cells; (E) The expression levels of immune cell marker genes (Cd3d, Cd8a, Cd4, Cd14, Cd19, and Cd79a) showed that CD8+ T cells were successfully isolated for RNA sequencing. + T cells; (F) Mfn2- / -CD8 T cells isolated from corresponding B16 tumor-bearing mice (n=3 mice / group) on day 14 + T and WTCD8+ Heat map of selected differentially expressed genes in T cells; the data in the above graphs are expressed as mean ± SD, and the data were analyzed by unpaired two-tailed Student's t-test (C, D), and **p < 0.01; ***p < 0.005.

[0017] Figure 4 MFN2-mediated mitochondria-endoplasmic reticulum contacts contribute to CD8 + Mitochondrial metabolism of TILs: (A) from WT and Mfn2 CKO Isolated spleen and intratumoral CD8 stained for COX IV (mitochondria; red) and calnexin (endoplasmic reticulum; green) from B16 tumor-bearing mice. + Representative 3D renderings of T cells, where magnifications of the boxed areas are shown on the right side of each image, and the scale bar represents 3 μm; (B) CKO Spleen and intratumoral CD8 isolated from B16 tumor-bearing mice + Mitochondrial elongation status of T cells (10 cells in 3 fields of view per sample, n = 3 mice), where fragmented represents <4 μm, medium represents 4-6 μm, and long represents >6 μm; (C) CD8 + Statistical quantification of the colocalization of COX IV and calnexin in T cells, including 10 cells in 3 fields of view per sample, n = 3 mice; (D) from WT and Mfn2 CKO CD8 activated by anti-CD3 / CD28 antibody (αCD3 / CD28) in mice (n=3 independent experiments) + Representative Western blot images (left) and statistical quantification (right) of the indicated proteins in whole-cell lysates (WCL) of T cells and crude mitochondrial fractions containing mitochondria-endoplasmic reticulum junctions (MEJs); (E) Expression of proteins from WT and Mfn2 cells as determined by flow cytometry targeting Rhod-2. CKO The designated CD8 + Mitochondrial Ca in T cells 2+ Representative histograms (left) and MFI (right); (F) mitochondrial Ca 2+ Intratumoral CD8 T cells of B16 tumor-bearing mice (n=3 mice / group) treated with Ru360, an uptake inhibitor + MFI of BODIPY 500 in T cells; (G) IFN-γ isolated from B16 tumor-bearing mice treated with Ru360 + Intratumoral CD8 +Percentage of T cells (n = 3 mice / group); Data in the above graphs are expressed as mean ± SD, and the data were analyzed by chi-square test (B), unpaired two-tailed Student's t test (D, F, G), or two-way one-way ANOVA and Tukey test (C, E), *p < 0.05, **p < 0.01, ***p < 0.005.

[0018] Figure 5 MFN2 in CD8 + Functions in mediating mitochondrial fusion and mitochondria-endoplasmic reticulum contacts in T cells, including: (A) Figure 4 (A) Shown are the WT and Mfn2 CKO Spleen and intratumoral CD8 isolated from B16 tumor-bearing mice + Representative 3D images of mitochondria (COX IV, red) and endoplasmic reticulum (calnexin, green) in T cells, where the scale bar represents 5 μm; (B) from WT and Mfn2 CKO Intratumoral CD8 isolated from MC38 tumor-bearing mice (10 cells from 3 fields per sample, n = 3 mice) + Mitochondrial elongation in T cells, where fragmented represents <4 μm, medium represents 4-6 μm, and long represents >6 μm; (C) CKO Intratumoral CD8 isolated from MC38 tumor-bearing mice (10 cells from 3 fields per sample, n = 3 mice) + Statistical quantification of colocalization of COX IV and calnexin in T cells; (D) Representative Western blot images (left) and statistical quantification (right) of the indicated proteins in whole cell lysates (WCL) and crude mitochondrial fractions containing mitochondria-endoplasmic reticulum junctions (MEJs) of αCD3 / CD28-activated human CD8+ T cells transduced with shRNA control vectors (shCtrl) and MFN2 targeting vectors (shMFN2) (n=3 independent experiments); the data in the above bar graphs are expressed as mean ± SD, and the data were analyzed by chi-square test (B), unpaired two-tailed Student's t test (C, D), *p < 0.05, *** p<0.005.

[0019] Figure 6 MFN2 interacts with SERCA2 on the endoplasmic reticulum to mediate CD8 +Mitochondrial-endoplasmic reticulum contacts in T cells: (A) Mass spectrometry analysis identifies SERCA2 as an MFN2-interacting protein in human T cells and HEK293T cells; (B) Western blot shows co-immunoprecipitation of overexpressed MFN2-Flag and SERCA2-HA in HEK293T cells; (C) Western blot shows co-immunoprecipitation of endogenous MFN2 and SERCA2 in T cells; (D) shows co-immunoprecipitation of human CD8 activated from αCD3 / CD28 + Representative confocal images of colocalization of MFN2 (green) and SERCA2 (red) in crude mitochondrial fractions isolated from T cells, where the scale bar represents 2 μm; (E) Human CD8 T cells transduced with αCD3 / CD28-activated cells carrying either shRNA control vector (shCtrl) or SERCA2-targeting shRNA (shSERCA2) + Western blot of the indicated proteins in whole cell lysates (WCL) of T cells and crude mitochondrial fractions containing mitochondria-endoplasmic reticulum junctions (MEJs); (F) Pull-down assay using MFN2-Flag and SERCA2-His purified from SF9 insect cells confirmed the direct interaction between MFN2 and SERCA2; (G) In tumor-infiltrating CD8 T cells isolated from ccRCC patients, the direct interaction between MFN2 and SERCA2 was confirmed. + The interaction between endogenous SERCA2 and MFN2 is enhanced in T cells; (H) CD8 + Kaplan-Meier survival curves of overall survival of melanoma patients classified by MFN2 and SERCA2 expression levels in TILs (n = 32, log-rank test); (I) Surface representation of the truncated MFN2 structure (Protein Data Bank code 6JFK) showing the locations of point mutations; (J) and (K) Interactions between overexpressed SERCA2-HA and various MFN2 variants in HEK293T cells; For B, C, E, F, G, J, and K, three independent experiments were performed with similar results.

[0020] Figure 7.Interaction between MFN2 and SERCA2, including: (A) Selected MFN2-associated proteins in HEK293T and T cells identified by mass spectrometry analysis; (B) Western blot showing co-immunoprecipitation of endogenous SERCA2 and overexpressed MFN2-Flag in HEK293T cells; (C) Western blot showing co-immunoprecipitation of endogenous MFN2 and overexpressed SERCA2-Flag in HEK293T cells; (D) Western blot showing co-immunoprecipitation of MFN2, but not MFN1, with SERCA2 in HEK293T cells; (E) Representative confocal images showing colocalization of MFN2 and SERCA2 in HeLa cells, where the right panel is an enlarged view of the boxed area in the left panel, showing colocalization of MFN2 and SERCA2; scale bar, 10 μm; For A to E, three independent experiments were performed with similar results.

[0021] Figure 8 MFN2-SERCA2 interaction on CD8 + The anti-tumor function of TIL is crucial, among which: (A) Schematic diagram showing Mfn2 CKO Generation of OT-I TCR transgenic mice and Mfn2 expressing MFN2 variants CKO Adoptive transfer of OT-I CD8+ T cells into B16-OVA melanoma-bearing mice; (B and C) Mfn2 CKO OVA-activated splenic CD8 + Mito-ER contact in T cells was rescued by MFN2 variants. Representative Western blots (B) and statistical quantification results (C) of three independent experiments (n=3) are shown; (D and E) MFN2 variants have an effect on CD8 + The effect of MFN2 variants on the antitumor function of T cells was investigated 22 days after tumor injection. - / - OT-I CD8 + Tumor growth curves (D) and tumor weights (E) of B16-OVA mice adoptively transferred with T cells (n = 4 mice / group); (F and G) Effects of MFN2 variants on CD8 + Effects of intratumoral Mfn2 expression on the effector function of TILs - / - OT-I CD8 + IFN-γ production by T cells isolated from B16-OVA tumor-bearing mice is shown as a representative flow cytometry plot (F) and IFN-γ + The percentage of CD8+ T cells (G) (n=4 mice / group); (H) Effect of MFN2 variants on CD8 +Effects of mitochondrial-endoplasmic reticulum contact status in TILs with Mfn2 variants expressed in tumors - / - OT-I CD8 + Statistical quantification of the colocalization area between COX IV and calnexin in T cells (5 cells from 3 fields per mouse, n = 3 mice / group) isolated from B16-OVA tumor-bearing mice 22 days after tumor injection; (I) Mfn2 expressing MFN2 variants 6 days after OVA activation - / - OT-I CD8 + Relative ATPase activity of SERCA2 isolated from crude mitochondrial fractions of T cells (n = 3 independent experiments); (J and K) Mfn2 expressing MFN2 variants in tumors - / - OT-I CD8 + Representative histograms (J) and MFI (K) of Rhod-2 in T cells (n = 4 mice / group); (L and M) Mfn2 expression in tumors expressing MFN2 variants - / - OT-I CD8 + Representative histograms (L) and MFI (M) of BODIPY in T cells (n = 4 mice / group); (N) apoptosis of T cells expressing MFN2 variants (Annexin V + ) Intratumoral Mfn2 - / - - Percentage of OT-I CD8+ T cells (n=4 mice / group); (O and P) Effect of MFN2 variants on CD8 + The effect of MFN2 variants on the antitumor function of T cells was investigated 22 days after tumor injection. - / - OT-I CD8 + Tumor growth curves (O) and tumor weights (P) of B16-OVA mice adoptively transferred with T cells (n=4 mice / group); (Q) Effects of MFN2 variants on CD8 + Effects of intratumoral Mfn2 expression on the effector function of TILs - / - OT-I CD8 + IFN-γ production by T cells, denoted as IFN-γ + The percentage of CD8+ T cells; The data in the above graphs are expressed as mean ± SD, and the data were analyzed by two-way one-way ANOVA and Tukey test (C, D, E, G, H, I, K, M, N, O, P and Q), *p < 0.05, ** *p<0.005.

[0022] Figure 9 .CD8 +Functional characterization of MFN2 variants in T cells, where: (A) PCR genotyping using tail DNA from Mfn2-loxp- / -;Cd4-Cre(-);OT-I(-) mice (1), Mfn2-loxp+ / +;Cd4-Cre(+);OT-I(+) mice (2, 3), and Mfn2-loxp+ / +;Cd4-Cre(-);OT-I(+) mice (4, 5) targeting Mfn2-loxp (wild type at 257 bp, Mfn2-loxP at 360 bp), Cd4-Cre (252 bp), and OT-I TCR transgenic versions (wild type at 200 bp, OT-I transgenic at 350 bp); (B) flow cytometry sorting of WT or Mfn2-loxp- / -;Cd4-Cre(-);OT-I(+) mice. CKO OT-I CD8 in the spleen of OT-I mice + T cell gating strategy; (C) OVA-activated WT or Mfn2C expressing MFN2 variants KO OT-ICD8 + Statistical quantification of MFN2 expression in T cells (n=3 independent experiments); (D) Effect of MFN2 variants on CD8 + Effect of mitochondrial elongation status on TILs expressing intratumoral MFN2 variants isolated from B16-OVA tumor-bearing mice (n=3 mice / group) CKO OT-ICD8 + Statistical quantification of mitochondrial elongation status in T cells; (E) Statistical quantification of mitochondrial elongation status in T cells expressing Mfn2 variants 6 days after OVA activation. CKO OT-I CD8 + Relative ATPase activity of SERCA2 isolated from whole-cell lysates of T cells (n=3 independent experiments); data in the above graphs are expressed as mean ± SD, and the data were analyzed by two-way one-way ANOVA and Tukey test (C, E) or chi-square test (D), *p < 0.05.

[0023] Figure 10 Targeting MFN2 can improve adoptive CD8 + Efficacy of T cell therapy and ICB-based cancer immunotherapy, wherein: (A) human CD8 T cells with or without MFN2 overexpression under the indicated treatment +T Relative expression of calnexin in crude mitochondrial fractions containing MEJ extracted from cells (n=3 independent experiments); (B) Relative expression of calnexin in human CD8 T cells with or without MFN2 overexpression under the indicated treatments. + Relative oxygen consumption rate (OCR) of T cells (n=3 independent experiments): (C) IFN-γ with or without MFN2 overexpression under the indicated treatments +human CD8 + Relative percentages of T cells (n=3 independent experiments); (D) Viable human CD8 T cells with or without MFN2 overexpression under the indicated treatments. + Relative numbers of T cells (n=3 independent experiments); (E) CD8 T cells activated by primary tumor antigens with or without MFN2 overexpression + Apoptosis rate of primary renal tumor cells from ccRCC patients co-cultured with T cells; (F) Preparation of antigen-specific CD8 + Scheme of adoptive transfer of CD8 T cells into NCG mice transplanted with autologous ccRCC PDX; (G) CD8 T cells 4 hours and 5 weeks after adoptive transfer into ccRCC PDX-bearing mice + Biodistribution of T cells, where the color scale represents light intensity and Luc represents luciferase (n=5 PDX / group); (H) Transferred CD8 T cells in different organs of NCG mice 5 weeks after transfer. + Distribution of T cells (n = 5 PDX / group), color scale indicates light intensity; (I) CD8 T cells in ccRCC PDX samples harvested 5 weeks after transfer + Representative IHC staining of CD8 T cells (top) and CD8 + T cell quantification (lower panel), where FOV represents field of view, scale bar is 50 μm (n=5 PDX / group); (J) adoptive transfer of human CD8 + Fold change of tumor volume in ccRCC PDX after adoptive transfer of human CD8 T cells (n = 5 PDX / group); (K) + Relative IFN-γ levels (normalized to CD8 + T cell count); (L) Representative Western blot showing human CD8 T cells treated with the indicated concentrations of leflunomide. + MFN2 levels in T cells. Three independent experiments were performed with similar results; (M) MFN2 levels in T cells from human CD8 T cells treated with the indicated concentrations of leflunomide. + Representative Western blot of calnexin in the crude mitochondrial fraction containing MEJ of T cells. Three independent experiments were performed with similar results; (N and O) Tumor growth (N) and survival curves (O) of B16 tumor-bearing mice treated with intraperitoneal injection of anti-PD-1 antibody and leflunomide (n = 5 mice / group); Data in the above graphs are expressed as mean ± SD, and the data were analyzed by unpaired two-tailed Student's t test (A, B, C, D, E, I, J, K), two-sided one-way ANOVA and Tukey test (N), or log-rank test (O). * p<0.05, **p<0.01, ** *p<0.005.

[0024] Figure 11 Targeting MFN2 in cancer immunotherapy, including: (A) Schematic diagram of the process for generating conditioned medium (CM) from primary renal tumor cells; (B) human CD8 + Representative Western blot of MFN2 overexpression in T cells; three independent experiments were performed with similar results; (C) HLA-A2 expression in primary renal tumor cells; (D) CD8 + In vitro bioluminescence of T cells; (E) CD8 T cells activated with primary renal tumor antigens with or without MFN2 overexpression + Representative flow cytometry images of primary renal tumor cells labeled with cell proliferation dye (CPD) in co-culture with T cells (Figure 2 Figure 10 (E) (F) Adoptive transfer of human CD8 with or without MFN2 overexpression + Fold change of tumor volume in CRC PDXs after T cell transfer (n = 5 PDXs / group); (G) CD8 + Representative IHC staining of CD8 T cells (left) and CD8 + Quantification of T cells (right) (n = 5 PDX / group), where the scale bar is 50 μm; (H) human CD8 + Relative IFN-γ levels (normalized to CD8 + T cell count); (I) Tumor growth curves of nude mice injected with vehicle- or leflunomide-treated B16 melanoma cells (n = 4 mice / group); Data in the above graphs are expressed as mean ± SD, and the data were analyzed by unpaired two-tailed Student's t test (F, H, I). ** p < 0.01; *** p<0.005. DETAILED DESCRIPTION

[0025] The present invention will be described clearly and completely below in conjunction with the embodiments and accompanying drawings of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments that can be obtained by a person of ordinary skill in the art based on the embodiments of the present invention fall within the scope of protection of the present invention.

[0026] As described in the background art, it has been shown that CD8 + Mitochondrial dynamics in T cells is crucial for maintaining CD8 + The effector function of T cells is very important, but existing research has not yet fully explored the role of CD8+ The function of mitochondria-endoplasmic reticulum contacts in T cells and their regulation are poorly understood.

[0027] As described above, the present inventors unexpectedly discovered that the tumor-infiltrating CD8 + T cells have higher expression of MFN2 protein, and it was found that this protein plays a role by mediating mitochondria-endoplasmic reticulum contact, more specifically, by interacting with SERCA2 (Ca2) on the endoplasmic reticulum. 2+ ATPase) to mediate mitochondrial-endoplasmic reticulum contact. In addition, the inventors have also found that MFN2 variants that can interact with SERCA2, such as R259A and V69F, can also function as wild-type MFN2 protein, although with lower efficacy. Based on the above findings, the inventors have come up with the idea of ​​increasing CD8 + Enhancement of MFN2 or its variant levels in T cells enhances mitochondria-endoplasmic reticulum contact to improve CD8 + The present invention was completed by demonstrating the cancer immunotherapy and / or tumor retention effects of T cells through multiple experiments.

[0028] Therefore, in the first aspect, the present invention provides mitochondrial fusion protein 2 (MFN2), a MFN2 variant capable of interacting with SERCA2, or a MFN2 expression promoter in maintaining and / or promoting CD8 + Use of T cells in tumor killing and / or survival.

[0029] As described in the background art, mitochondrial fusion protein 2 (MFN2) is a transmembrane GTPase responsible for mitochondrial outer membrane fusion. Its key role in mitochondrial fusion is undeniable, and it is involved in mitochondrial-endoplasmic reticulum contact. The amino acid sequence of MFN2 is shown in SEQ ID NO. 1, as follows: + Mitochondrial metabolism and effector function of T cells are key factors, and high levels of MFN2 are associated with CD8 +Therefore, MFN2 or any agent that can increase MFN2 expression can be used to maintain and / or promote CD8 + Tumor killing or survival of T cells.

[0030] The inventors further discovered that MFN2 mediates mitochondrial-endoplasmic reticulum interactions by interacting with sarcoplasmic / endoplasmic reticulum calcium ATPase 1 / 2 / 3 (SERCA1 / 2 / 3, or ATP2A1 / 2 / 3) on the endoplasmic reticulum, especially SERCA2. SERCA is an internal endoplasmic reticulum channel that transfers Ca to the endoplasmic reticulum in an ATP hydrolysis-dependent manner. 2+ Pumped from the cytosol to the ER lumen (Dyla et al., 2020; Zhao et al., 2017).

[0031] To further investigate the effect of MFN2 structure on its function, the inventors initially introduced four single point mutations (T105M, T130A, R94Q, and R259A) into MFN2. The results showed that only the MFN2 variant R259A, which can interact with SERCA2, can mediate mitochondria-endoplasmic reticulum contact and maintain CD8 + On this basis, the inventors further tried to introduce more single point mutations (V69F, L76P, P251A, R280H or W740S) into MFN2. The results showed that the MFN2 variants (V69F, L76P, R280H or W740S) that can interact normally with SERCA2 all maintain the CD8 T cell tumor killing ability to a certain extent. + The tumor killing ability of T cells is not maintained by the MFN2 mutant (P251A) that cannot normally bind to SERCA2. + This further confirms the above conclusion and also indicates that at least the amino acids at the above positions (R259, V69, L76, R280 and W740) are actually not involved in the interaction between MFN2 and SERCA2.

[0032] The term "MFN2 variant" as used herein refers to a mutant protein having one or more (such as two, three, four, five, or more) amino acid mutations relative to the wild-type MFN2 protein. In this article, the symbol "AXXXB" is used to represent a point mutation or a point mutant protein, wherein the amino acid A at position XXX is mutated to B or a protein variant containing such a mutation. For example, R259A indicates that the R (arginine (Arg)) at position 259 of the MFN2 protein is mutated to A (alanine (Ala)) or a protein variant containing such a mutation. Therefore, in the present invention, maintaining and / or promoting CD8 +MFN2 variants that enhance tumor killing and / or survival of T cells may be variants that interact with SERCA2. More specifically, the MFN2 variant may be one that has at least 85% (e.g., at least 87%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.6%, 99.7%, 99.8%, 99.9%, 99.99%, or 99.999%) sequence homology to the amino acid sequence of SEQ ID NO: 1 and that interacts with SERCA2. Alternatively, the MFN2 variant may be one that has one to ten (e.g., one, two, three, four, five, six, seven, eight, nine, or ten) amino acid mutations relative to the amino acid sequence of SEQ ID NO: 1 and that interacts with SERCA2. Therefore, in one embodiment, the MFN2 variant is an MFN2 variant comprising a mutation at one or more of R259, V69, L76, R280, and W740. In a preferred embodiment, the MFN2 variant is an MFN2 variant comprising one or more of R259A, V69F, L76P, R280H, and W740S. In a more preferred embodiment, the MFN2 variant is an MFN2 variant comprising R259A, V69F, L76P, R280H, or W740S.

[0033] It will be understood by those skilled in the art that the MFN2 or MFN2 variant capable of interacting with SERCA2 may be in any suitable form. For example, the MFN2 or its variant may be in the form of a protein itself; in this case, the MFN2 or its variant may be directly administered to CD8 + In another example, the MFN2 or its variant can be in the form of any vector expressing the MFN2 or its variant; in this case, the vector expressing the MFN2 or its variant can be used to transfect CD8 + T cells, thereby enabling them to +It is expressed in T cells, thereby mediating mitochondrial-endoplasmic reticulum contact therein. In some embodiments, the vector can be a viral vector, such as a lentiviral vector, a retroviral vector, and an adenoviral vector. Compared with lentiviral vectors, adenoviral vectors do not integrate into the host cell genome after entering the cell and are only expressed transiently. Adenovirus has a clear advantage in infecting mouse cells, but the advantage is not obvious when infecting human cells. For the transduction of human lymphocytes, since the transduction is in vitro and then re-infused into the body, there is no need to consider the long-term risk of lentivirus integration into the host genome causing carcinogenesis. On the contrary, after lentivirus integrates into the lymphocyte genome, it can stably overexpress and exert its effect for a long time, while adenoviral vectors do not have this advantage. Although retroviral vectors can also achieve long-term and stable expression of exogenous proteins by integrating into the host genome, their integration sites have certain tendencies, thereby increasing the long-term risk of carcinogenesis. In a preferred embodiment, the vector can be a lentiviral vector.

[0034] Furthermore, as used herein, the term "MFN2 expression-promoting agent" refers to any agent known in the art capable of promoting MFN2 expression, such as leflunomide. The MFN2 expression-promoting agent can be administered in any amount effective to promote MFN2 expression. If the MFN2 expression-promoting agent itself can also treat tumors or cancers at higher doses, it can be administered in an amount effective only to promote MFN2 expression or in an amount effective to treat tumors or cancers. It is understood that in the latter case, the MFN2 expression-promoting agent not only promotes MFN2 expression but also has a tumor or cancer treatment effect.

[0035] CD8 + The tumor killing ability of T cells is mainly reflected in whether their effector function can be normally exerted, and the exertion of effector function is mainly affected by the production of interferon-γ. + The amount of MFN2 protein in T cells can interact with SERCA2 on the endoplasmic reticulum, thereby mediating mitochondrial-endoplasmic reticulum contact to protect mitochondrial Ca 2+ Homeostasis, promoting CD8 + T cells produce interferon-γ, ensuring CD8 + T cell metabolic adaptability and effector function, thereby ensuring CD8 + In one embodiment, the MFN2 or MFN2 variant can increase the tumor killing ability of CD8 + T cells produce interferon-γ (IFN-γ).

[0036] In addition, the field has found that CD8 +The tumor killing ability and survival of T cells in the tumor microenvironment are both reduced. The tumor microenvironment (TME) refers to the surrounding microenvironment where tumor cells exist, including surrounding blood vessels, immune cells, fibroblasts, bone marrow-derived inflammatory cells, various signaling molecules and extracellular matrix (ECM). It is a complex integrated system. In solid tumors, due to the rapid growth of tumor tissue, the high expansion of volume and the incomplete vascular system inside the tumor tissue, these will lead to insufficient oxygen supply in the tumor tissue, and the tumor microenvironment presents the characteristics of overall hypoxia. Such a tumor microenvironment will have different effects on T cell effector function or its survival, and the loss of T cell effector function in the tumor microenvironment is one of the main reasons for the failure of TIL and ICB immunotherapy. In the present invention, the inventors evaluated the effect of MFN2 on CD8 + The effects of ccRCC conditioned medium on the tumor killing and survival of T cells were investigated by using primary ccRCC cancer cell lysates to generate ccRCC conditioned medium to simulate the tumor microenvironment. + Compared with CD8 T cells cultured in ccRCC conditioned medium, MFN2 had a significant effect on + The promotion of tumor killing and survival of T cells is more significant, indicating that the tumor microenvironment-induced CD8 + T cell dysfunction can be corrected by enhancing MFN2 expression.

[0037] In a second aspect, the present invention provides a CD8 + Use of T cells in the preparation of cell therapeutic agents for adoptive cellular immunotherapy.

[0038] Adoptive immune cell therapy (ACT) is an immunotherapy used to treat tumors or cancers. It specifically refers to the collection of autoimmune cells, amplification and processing through in vitro culture, and then re-infusion into the patient to increase the immunogenicity of tumor or cancer cells and sensitivity to effector cell killing. In adoptive T cell immunotherapy, maintaining the continued survival and effector function of T cells is a key factor in good clinical effects. Existing clinical results show that the continued existence of T cells or their survival is highly correlated with tumor regression, and the loss of effector function of T cells may be related to intrinsic factors (T cell metabolic adaptability) and extrinsic factors (tumor microenvironment). Therefore, CD8 can be improved by overexpressing MFN2 or its variants that can interact with SERCA2. + T cells' metabolic adaptability in the tumor microenvironment, improving CD8 + T cell effector function and increase CD8 + The survival of T cells in the tumor microenvironment enables CD8 +T cells play a better therapeutic effect in adoptive cellular immunotherapy, so CD8 T cells overexpressing MFN2 or its variants that can interact with SERCA2 + T cells can be used to prepare cell therapeutic agents for adoptive cellular immunotherapy.

[0039] In some embodiments, the CD8 + CD8 T cells were transfected with vectors overexpressing MFN2 or overexpressing MFN2 variants that can interact with SERCA2. + T cells are obtained.

[0040] In some embodiments, the MFN2 variant may include a mutation at one or more of R259, V69, L76, R280, and W740. In a preferred embodiment, the MFN2 variant may include a mutation at one or more of R259A, V69F, L76P, R280H, and W740S. In some more preferred embodiments, the MFN2 variant may include R259A, V69F, L76P, R280H, or W740S.

[0041] In some embodiments, the vector can be a viral vector, such as a lentiviral vector, a retroviral vector, and an adenoviral vector. In a preferred embodiment, the vector can be a lentiviral vector.

[0042] Naive CD8 + T cells require antigen stimulation to become activated CD8 cells with cytotoxic function + In one embodiment, the CD8 + T cells can be further activated by antigen presenting cells, thus having cytotoxicity to kill tumor cells. Dendritic cells (DC) are the most powerful antigen presenting cells. Dendritic cells can phagocytize tumor neoantigens, process them into antigen peptides and present them to CD8 + T cells also express co-stimulatory molecules such as CD80 and CD86, and secrete cytokines such as IL-2, helping to activate T cells and enable them to exert their anti-tumor functions. Compared with other existing methods for activating T cells in vivo, the method using dendritic cells has the advantages of recognizing a wider spectrum of mutant antigens and fewer side effects. Therefore, in a preferred embodiment, the antigen-presenting cells can be dendritic cells.

[0043] In one embodiment, the cell therapy agent can be further used in combination with an immune checkpoint blocker to achieve a better therapeutic effect. The so-called "immune checkpoint" refers to a series of molecules expressed on immune cells that can regulate the degree of immune activation. They play an important role in preventing the occurrence of autoimmune effects (abnormal immune function, attacking normal cells). Tumor cells can use this mechanism of immune cells to inhibit the function of immune cells, thereby escaping and surviving from the human immune system. The so-called "immune checkpoint inhibitor" refers to a class of agents that can relieve the immunosuppressive effect of tumor cells on immune cells, allowing immune cells to be reactivated and thus eliminate cancer cells. Current immune checkpoint inhibitors are mainly CTLA-4 inhibitors and PD-1 inhibitors (PD-1 / PD-L1 inhibitors), among which PD-1 inhibitors (PD-1 / PD-L1 inhibitors) include anti-PD-1 antibodies (PD-1 inhibitors) and anti-PD-L1 antibodies (PD-L1 inhibitors). Therefore, in one embodiment, the immune checkpoint blocker can be an anti-PD-1 antibody. In addition, the cell therapy agent and immune checkpoint blocker can be administered simultaneously or sequentially, depending on the specific circumstances.

[0044] In the experimental process of the present invention, the inventors used three different tumor models (melanoma (B16), renal clear cell carcinoma (ccRCC) and colorectal cancer (CRC) models) to conduct multiple experiments. The results showed that CD8 + T cells achieved better tumor killing and survival in all three tumor models. Therefore, in one embodiment, the cell therapy agent can be used to treat cancer, such as renal cancer, colorectal cancer or melanoma, but is not limited thereto. It is expected that when the cell therapy agent is used to treat other types of cancer, the CD8 + After T cells are activated with corresponding tumor antigens, the desired therapeutic effect of the corresponding cancer can also be achieved.

[0045] In a third aspect, the present invention provides a MFN2 variant capable of interacting with SERCA2, comprising mutations in one or more of R259, V69, L76, R280 and W740.

[0046] The inventors have found that among the many types of MFN2 variants constructed, only the MFN2 variants that retain the activity of interacting with SERCA2 still retain the function of the wild-type MFN2 variant. Therefore, it can be speculated that the MFN2 variants obtained by mutating one or more sites of the MFN2 protein that are not involved in the interaction with SERCA2 (for example, one or more sites of R259, V69, L76, R280 and W740) can still interact normally with SERCA2, thereby maintaining the CD8 + The tumor killing and / or survival of T cells can be used to treat cancer, among other things. It is understood that the class or specific amino acid of the mutated amino acid is not important for the present invention; what is important is that the mutated MFN2 protein still interacts with SERCA2. In a preferred embodiment, the MFN2 variant includes one or more mutations: R259A, V69F, L76P, R280H, and W740S. In a more preferred embodiment, the MFN2 variant includes mutations: R259A, V69F, L76P, R280H, or W740S.

[0047] In a fourth aspect, the present invention provides a method for treating cancer, comprising: administering to a cancer patient a CD8 T cell that overexpresses MFN2 or overexpresses a MFN2 variant that can interact with SERCA2; + T cells, or administering a MFN2 expression promoter to the cancer patient.

[0048] In addition, in addition to the administration of CD8 + In addition to using T cells to kill tumor cells, it is also possible to consider administering MFN2 expression promoters to promote the expression of MFN2, thereby maintaining a relatively high MFN2 level in the patient. When using MFN2 expression promoters, it is possible to consider administering CD8 T cells that overexpress or do not express MFN2 or MFN2 variants that can interact with SERCA2. + Therefore, in an optional embodiment, the method further comprises administering a CD8 T cell that overexpresses or does not overexpress MFN2 or a MFN2 variant that can interact with SERCA2 to the cancer patient at the same time as administering the MFN2 expression promoter. +T cells. In a preferred embodiment, the MFN2 variant capable of interacting with SERCA2 may include a mutation in one or more of R259, V69, L76, R280, and W740. In a more preferred embodiment, the variant may include a mutation in one or more of R259A, V69F, L76P, R280H, and W740S. In a further preferred embodiment, the variant may include a mutation in R259A, V69F, L76P, R280H, or W740S.

[0049] It is understandable that by + When administering T cells to cancer patients for cancer treatment, it is necessary to pre-regulate the CD8 + T cells are activated to enable them to have corresponding tumor killing ability. Therefore, in one embodiment, the CD8 + T cells are activated by antigen presenting cells such as dendritic cells. + The method for activating CD8 T cells in the second aspect of the present invention + T cells in a similar way, both of which work by making CD8 + T cells are mixed with antigen presenting cells such as dendritic cells and then the mixture of the two cells is administered to the cancer patient. + T cells and antigen-presenting cells such as dendritic cells to activate the CD8 + T cells.

[0050] Similar to the second aspect of the present invention, to achieve a better cancer treatment effect, the method for treating cancer of the third aspect of the present invention may also include administering an immune checkpoint blocker to the patient. Similarly, in one embodiment, the immune checkpoint blocker may be an anti-PD-1 antibody. Furthermore, similarly, the cell therapy agent and the immune checkpoint blocker may be administered simultaneously or sequentially, depending on the specific circumstances.

[0051] In addition, as can be understood, after activation with appropriate tumor antigens, the CD8 + When T cells are administered to a patient with the tumor / cancer, they can kill the tumor / cancer cells, thereby achieving treatment for the cancer patient. In one embodiment, the tumor can be a variety of cancer types including renal cancer, colorectal cancer, and melanoma.

[0052] Example

[0053] The present invention will be described in more detail below in conjunction with the examples. The test methods in the following examples are conventional methods unless otherwise specified. The test materials used in the following examples are purchased from conventional reagent stores unless otherwise specified. It should be noted that the above summary of the invention and the detailed description below are only for the purpose of specifically illustrating the present invention and are not intended to limit the present invention in any way.

[0054] Materials and methods

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

[0061] Patients and tissue samples

[0062] Blood samples and clear cell renal cell carcinoma (ccRCC) tissues were obtained from patients at the Sun Yat-sen University Cancer Center and the First Affiliated Hospital of Sun Yat-sen University (Guangzhou, China). Blood samples and colorectal cancer (CRC) tissues were obtained from patients at the Sixth Affiliated Hospital of Sun Yat-sen University (Guangzhou, China). Paraffin-embedded tumor samples were obtained from 116 ccRCC patients admitted to the Sun Yat-sen University Cancer Center (Guangzhou, China) between 2013 and 2015 for Kaplan-Meier survival analysis. All samples were obtained from patients who provided informed consent, and all related procedures were approved by the Internal Review and Ethics Committee of Sun Yat-sen University. This study complied with all relevant ethical regulations for research involving human participants.

[0063] Isolation of CD8 from patient samples + T cells

[0064] The tumor was freshly isolated and washed with PBS to prevent contamination of peripheral blood cells. After tumor tissue resection, the tumor samples were cut into small pieces (1-2 mm 3) and digested with RPMI-1640 containing 1 mg / ml Liberase TM (Roche Diagnostics, 5401119001) and 30 IU / ml DNAse (Takara, 2270A) at 37 ° C with constant shaking for 40 minutes. The digested cell suspension was filtered through a 40-μm cell strainer and washed twice with PBS to remove debris. Infiltrating T cells were enriched by Ficoll-Paque PLUS (GE Healthcare) density gradient separation and collected from the mononuclear cell layer. After washing with PBS, the precipitated cells were resuspended and incubated with Alexa Fluor700 anti-human CD3 antibody (eBioscience, 56-0037-42) and FITC anti-human CD8 antibody (eBioscience, 11-0086-42) at 4 ° C for 20 minutes. 1 μM calcein violet AM (Invitrogen, C34858) was added immediately before sorting to exclude dead cells. Fluorescence activated cell sorting (FACS) was performed on a FACS Astrios (Beckman Coulter) using 488 nm (FITC, 513 / 26 filter), 640 nm (Alexa Fluor 700, 722 / 44 filter), and 405 nm (Calcein Violet AM, 450 / 50 filter) lasers. Standard forward angle width and height criteria were used to remove doublets and capture singlets. Sorted live CD8 + T cells (Calcein 高 CD3 + CD8 + ) viability and number, and the sorted cells were immediately used for further experiments.

[0065] Single-cell RNA sequencing

[0066] CD3 + CD8 + T cells were sorted into PBS containing 0.04% bovine serum albumin (BSA) and kept on ice. The sorted cells were then counted using a Countess II automated counter (ThermoFisher Scientific) and their viability was assessed using trypan blue. The cells were then plated at 2–4 × 10 5The cells / ml were resuspended, and the final survival rate was >90%. Single cell RNA sequencing was performed according to the manufacturer's protocol using a single cell 5 'library and a gel bead kit V2 (10 × Genomics). In brief, living single cells were loaded onto a chromium single cell controller (10 × Genomics) to generate single cell gel beads in an emulsion (GEM). The captured cells were lysed, and the released RNA was barcoded in each GEM by reverse transcription. The amplified cDNA was purified using SPRIselect beads (Beckman Coulter) and sheared to 250-400bp. The quality of the cDNA was assessed using a Qubit 3.0 fluorometer. The library was sequenced (performed by Beijing Novogene) using the Illumina NovaSeq 6000 system.

[0067] Single-cell RNA-Seq data preprocessing

[0068] A raw gene expression matrix was generated for each sample using the Cell Ranger (version 3.0.2) pipeline in conjunction with the human reference genome version GRCh38. The output filtered gene expression matrix was analyzed using the Seurat package (version 3.0.0) (Butler et al., 2018). Briefly, genes expressed at >0.1% of the data and cells with >200 genes detected were selected for further analysis. Low-quality cells were removed if they met the following criteria: 1) <800 UMIs, 2) <500 genes, or 3) >10% of the UMIs were from the mitochondrial genome. After removing low-quality cells, the gene expression matrix was normalized using the NormalizeData function, and 2000 features with high inter-cell variability were calculated using the FindVariableFeatures function. To reduce the dimensionality of the dataset, the RunPCA function was used with default parameters on the data scaled using the linear transformation generated by the ScaleData function. Next, the ElbowPlot, DimHeatmap, and JackStrawPlot functions were used to identify the true dimensionality of each dataset, as recommended by the Seurat developers. Finally, the inventors clustered the cells using the FindNeighbors and FindClusters functions and performed nonlinear dimensionality reduction with the RunUMAP function using default settings. All details about the Seurat analysis performed in this work can be found in the website tutorial ( https: / / satijalab.org / seurat / v3.0 / pbmc3k tutorial.html ).

[0069] Primary human PBMC and CD8 + T cell isolation

[0070] Peripheral blood mononuclear cells (PBMCs) were isolated using Ficoll-Paque (GE Healthcare, 17-5442-02) according to the manufacturer's instructions. Briefly, fresh peripheral blood was collected in EDTA anticoagulant tubes and then layered onto Ficoll-Paque. After density gradient centrifugation, primary PBMCs were collected from the mononuclear cell layer. PBMCs were washed twice with PBS and resuspended in X-VIVO (Lonza, 04-418Q) to produce a single cell suspension. CD8 + T cell isolation kit (Miltenyi, 130-096-495) was used to purify primary human CD8 + T cells. In some experiments, CD8 + T cells were incubated with Alexa Fluor 700 anti-human CD3 (eBioscience, 56-0037-42) and FITC anti-human CD8 (eBioscience, 11-0086-42) antibodies at 4°C for 20 minutes and sorted by flow cytometry (Beckman Coulter). Flow cytometric analysis confirmed that the purity of the cell population was >90%.

[0071] Tumor tissue cytokine assay

[0072] Mice were sacrificed, and tumor tissues were collected and homogenized in PBS. The protein concentration of each homogenized tissue sample was determined, and the IFN-γ level of each sample was determined using a human IFN-γ ELISA kit (Abbkine, KET6011) according to the manufacturer's instructions.

[0073] Immunofluorescence detection

[0074] For immunostaining of T cells, coverslips were pre-coated with PDL (poly-D-lysine, Sigma, P6407) at 37°C for 1 hour and washed three times with PBS. Cells were seeded on slides coated with PDL and fixed with 4% paraformaldehyde for 15 minutes at room temperature, then permeabilized with 0.1% Triton X-100 and blocked with 2% BSA in PBS for 1 hour at room temperature. Next, the cells were incubated with the appropriate primary antibody at 4°C overnight and incubated with Alexa Fluor (Invitrogen)-conjugated secondary antibody for 1 hour at room temperature. The nuclei were counterstained with DAPI (Invitrogen, D3571) and images were acquired using a confocal laser scanning microscope (Olympus FV1000 or Nikon N-SIM). For confocal z-axis stacks, 20 images separated by 0.2 μm along the z-axis were acquired by super-resolution confocal microscopy (Nikon N-SIM). 3D reconstruction and colocalization analysis of mitochondria and endoplasmic reticulum were performed using IMARIS 9.0. Image J was used to determine the Manders colocalization coefficient for the mitochondrial-endoplasmic reticulum overlap region and to analyze mitochondrial elongation. For the latter, cells were defined as fragmented if the majority of mitochondria were less than 4 μm in length, as medium if the majority were 4 to 6 μm in length, and as long if the majority were >6 μm in length.

[0075] Immunohistochemical staining

[0076] Paraffin-embedded tumor samples were serially sectioned at a thickness of 4 μm. Antigen retrieval was performed in 0.01 M citrate buffer (pH 6.0) using a pressure cooker for 3 minutes, followed by treatment with 3% hydrogen peroxide for 5 minutes. Slides were incubated with specific antibodies against CD8 (1:100; MXB Biotech, MAB-0021) and MFN2 (1:100; Abcam, ab218162) at 4°C overnight and stained with an anti-mouse / rabbit IHC secondary antibody kit (ZSGB-BIO, PV-6000) according to the manufacturer's instructions. After staining with hematoxylin, images were captured under a microscope (NIKON ECLIPSE 80i). CD8 + The number of TILs was determined by counting CD8-positive cells at 20× magnification in at least 5 fields per section. + MFN2 expression levels in TILs were measured at high resolution using serial sections from the same patient in at least five fields per section at ×40 magnification. The accuracy of the measurements was visually verified by independent assessment by two pathologists.

[0077] mice

[0078] Tg(Cd4-cre)1Cwi / BfluJ (Cd4-Cre transgenic mice) and C57BL / 6-Tg(TcraTcrb)1100Mjb / J (OT-I transgenic mice) mice were obtained from Jackson Laboratory. em26 IL2rg em26 / Gpt (NCG), C57BL / 6 and nude mice were purchased from GemPharmatech (Nanjing, China). All mice were maintained under specific pathogen-free conditions at the Laboratory Animal Resource Center of Sun Yat-sen University.

[0079] Mfn2 flox / flox The mice were generated by GemPharmatech (Nanjing, China) using CRISPR / Cas9-mediated genome engineering. To generate these mice, Cas9, sgRNA, and a construct consisting of Mfn2-loxP (exon 5)-loxP were microinjected into fertilized eggs of C57BL / 6J mice. The fertilized eggs were transplanted to obtain correctly targeted mice, and the results were confirmed by PCR and sequencing. flox / flox Mice and Cd4 Cre Crossing with OT-I transgenic mice to generate mice with MFN2 conditional knockout in T cells or OT-I T cells (Mfn2 flox / flox Cd4 Cre and Mfn2 flox / f lox Cd4 Cre OT-I). Genotyping was performed by PCR on tail DNA using specific primers. Conditional deletion of Mfn2 was confirmed by immunoblotting and immunofluorescence using T cells isolated from the spleen. CKO For the animal experiments in mice, 6-week-old littermates with normal MFN2 expression (WT) were used. All animal experiments used age- and sex-matched mice that were randomly assigned to experimental groups. The animal experiments were approved by the Institutional Review Board and the Animal Care and Use Committee of Sun Yat-sen University. + and CD8 + T cell differentiation is completed in infancy and there is no specific CD4 cre or CD8 cre Mouse models, existing CD4 cre The mouse model is actually in CD4 + and CD8 + A tool for conditional knockout in T cells, commonly known in the field as "CD4 cre Therefore, in this study, CD4 cre Mouse model.

[0080] cell

[0081] SF9 cells were obtained from the laboratory of Professor Ping Yin (Huazhong Agricultural University). Human PBMCs were donated by healthy donors. Primary ccRCC tumor cells were obtained from fresh tumor samples. HEK293T, HeLa, Jurkat, and B16F10 cell lines were originally from the American Type Culture Collection (ATCC), and the MC-38 cell line was originally purchased from Kerast Inc. and maintained in the laboratory. The B16F10-OVA cell line was generated by lentiviral transduction of OVA antigen. All cell lines tested negative for mycoplasma contamination. HEK293T, HeLa, B16F10, B16F10-OVA, and MC38 cells were cultured in complete DMEM medium containing 10% FBS and 1% penicillin / streptomycin. Human CD8 + T cells and Jurkat cells were cultured in X-VIVO medium, and mouse CD8 + T cells were cultured in complete RPMI1640 medium supplemented with 10% FBS, 1% PS and IL-2 (100 IU / ml). + T cells were activated with 2 μg / ml plate-bound anti-CD3 / CD8 antibodies (BioLegend) for the indicated periods of time. -1- OT-I T cells were incubated with 10 nM OVA in the presence of 100 IU / ml IL-2. 257-264 Peptide (Sigma, S7951) was used to stimulate the cells for the indicated time periods. All cells were grown according to standard protocols.

[0082] Immunoblotting

[0083] The cells were lysed on ice with RIPA buffer (Beyotime, P0013B) for 30 minutes. The cell lysate was centrifuged at 18,000 g for 10 minutes, and the supernatant was resolved by SDS-PAGE, transferred to a PVDF membrane, and blocked with 5% w / v BSA. The membrane was incubated with the primary antibody at 4°C overnight and then incubated with an HRP-conjugated secondary antibody (Cell Signaling Technology) at room temperature for 1 hour. Antigen-antibody reactions were visualized by ECL Western Blotting Substrate (Tanon, 180-5001).

[0084] Metabolic assays

[0085] For fatty acid uptake assays, CD8 +T cells were sorted by FACS and incubated with 1 μM BODIPY 500 (ThermoFisher, B3824) at 37°C for 20 minutes. Cells were analyzed on a Beckman CytoFLEX flow cytometer and metabolic parameters were quantified as mean fluorescence intensity (MFI). To measure cellular oxygen consumption rate (OCR), isolated CD8 + T cells (5×10 5 Cells were plated in XF medium (2 mM glucose, 2 mM glutamine, and 1 mM pyruvate) on PDL-treated Seahorse plates and analyzed using an XF-24 extracellular flux analyzer (Agilent Technologies). Basal OCR was measured for 30 minutes, and then cells were treated sequentially with 1.5 mM oligomycin, 1.0 mM FCCP, and 0.5 mM rotenone / antimycin A (all from Agilent Technologies) at the indicated time points to measure maximal respiration and hyperrespiration capacity.

[0086] Animal experiments

[0087] For tumor implantation, age- and sex-matched WT and MFN2 CKO Mice (age 6-8 weeks) were anesthetized with 150 μl of 4% chloral hydrate and 4 × 10 5 B16F10 cells or 5 × 10 5 MC38 cells were injected subcutaneously into the back of each mouse. Starting from the 6th day, the tumor size and mouse survival rate were recorded every 3 days. The tumor volume was calculated as follows: (length 2 × width) / 2. When the tumor diameter reached approximately 15 mm, the animals were euthanized. For phenotypic analysis and RNA-seq of tumor-infiltrating T cells, mice were euthanized on day 14 (B16F10) or day 21 (MC38). In some experiments, to deplete CD8 + T cells were injected intraperitoneally with CD8 depletion antibody (150 μg per mouse, BioXcell, BP0117) 1 day before tumor injection, and then injected three times every 3 days. For anti-PD-1 treatment, anti-mouse PD-1 antibody (100 μg per mouse, BioXcell, BP0273) or isotype control antibody (IgG) (100 μg per mouse, BioXcell, BP0089) was injected intraperitoneally on day 4, and then every 3 days. For in vivo treatment, mice were administered DMSO or leflunomide (4 mg / kg, MCE, HY-B0083) by intraperitoneal injection every 3 days.

[0088] T cells were isolated from tumors or spleens according to a previously reported method (Hamaidi et al., 2020). Tumor tissue samples were washed with PBS and cut into small pieces, and then digested with RPMI-1640 containing 2 mg / ml collagenase IV (Sigma, C4-BIOC) and 30 IU / ml DNAse at 37 ° C for 1 hour under constant shaking. To isolate tumor-infiltrating CD8 + T cells, the tumor tissue was mechanically separated. The spleen was cut into small pieces and placed on a filter connected to a 50 ml conical tube. The fragments were pressed through the filter using the plunger end of a syringe and the filter was washed with excess PBS to obtain a cell suspension. The cell suspension was filtered through a 40-μm cell strainer and washed twice with PBS, and then the T cells were isolated by Ficoll-Paque PLUS (GE Healthcare) density gradient separation. The isolated cells were stained with antibodies against APC anti-mouse CD3 (BioLegend, 100235) and PE anti-mouse CD8a (BioLegend, 100707) at 4°C for 30 minutes. The cells were stained with Calcein AM (Beyotime, C2012) to exclude dead cells. CD8 in single cell suspensions was analyzed by FACS using a flow cytometer (Beckman Coulter). + T cells were sorted for further experiments, and the purity of the sorted population was verified to be >90% by flow cytometry analysis.

[0089] For adoptive cell transfer therapy against B16F10-OVA melanoma, WT C57BL / 6 mice (male, 6-8 weeks) were anesthetized and 4×105 B16F10-OVA cells were injected subcutaneously into the back of the mice. On day 4, tumor-bearing mice were intravenously injected with 1.5×10 6 WT OT-1CD8 + T cells or MFN2 cells transduced with the indicated MFN2 variants - / - OT-1CD8 + T cells, and tumor size was recorded every 4 days. + T cell phenotype: Tumor-bearing mice were euthanized on day 22, and OT-1CD8 + T cells were used for further experiments.

[0090] Flow cytometry

[0091] Cells were labeled with designated fluorescein-conjugated antibodies for 30 minutes at 4°C to analyze surface markers. To detect cytokine production, cells were stimulated with 50 ng / ml phorbol 12-myristate 13-acetate (PMA), 1 μM ionomycin, and 5 μg / ml BFA at 37°C for 4 hours. For intracellular staining, cells were treated with a fixation and permeabilization solution kit (BD, 554714) according to the manufacturer's instructions and stained with designated primary antibodies. For apoptosis analysis, cells were collected by centrifugation, incubated with 5 μl Annexin V (MultiScience) in 100 μl binding buffer at room temperature for 10 minutes, stained with PI, and immediately analyzed by flow cytometry. Samples were analyzed using a Beckman CytoFLEX flow cytometer, and data were analyzed using FlowJo10 software.

[0092] RNA-Seq analysis

[0093] The WT and Mfn2 cells were isolated from two age- and sex-matched groups of 8-week-old WT mice (n=3) by FACS (at least 95% purity). CKO A total of 600-800 viable tumor-infiltrating CD8 +T cells were directly sorted into 5 μl of lysis buffer containing 10 μM dNTP mix, 10 μM Oligo dT primer, 1% Triton X-100, and 40 IU / ml RNase inhibitor. The tubes were sealed, snap-frozen, and stored at -80°C before further processing according to previously described protocols (Picelli et al., 2014). Paired-end read sequences were aligned to the mouse reference genome version mm10 using default settings in STAR (version 2.6.1b) (Dobin et al., 2013) and quantified using HTSeq (version 0.11.0) (Anders et al., 2015) in "cross stringency" mode. The raw count matrix was normalized using DESeq2 (version 1.32.0) (Love et al., 2014) to estimate gene expression levels and identify differentially expressed genes (DEGs). The Benjamini-Hochberg method was used to estimate the false discovery rate (FDR). DEGs were filtered using a minimum log2-transformed fold change of 1 and a maximum FDR value of 0.05. Enrichment analysis was performed using the Metascape web tool (www.metascape.org) to determine the functions of the DEGs. Gene sets were derived from the Gene Ontology (GO) biological process ontology (http: / / geneontology.org). To assign pathway activity estimates to individual samples, the inventors applied gene set variation analysis (GSVA, version 1.40.1) (Hanzelmann et al., 2013) to 50 landmark pathways using standard settings, as previously described (Xing et al., 2021). Limma (version 3.48.1) (Ritchie et al., 2015) was used to calculate the differential activity levels of pathways between conditions. Each pathway with a Benjamini-Hochberg-corrected p-value <0.05 was considered significantly perturbed.

[0094] Protein expression and purification

[0095] Full-length human MFN2 and SERCA2 cDNAs were cloned into the pFastBac1 vector (Invitrogen) with a C-terminal 3× Flag tag (MFN2) or a His6 tag (SERCA2). Recombinant MFN2 and SERCA2 proteins were expressed in SF9 insect cells using the Bac-to-Bac baculovirus system (Invitrogen). Briefly, bacmid DNA was generated in DH10Bac cells, and the resulting baculovirus was amplified in SF9 insect cells. Following baculovirus infection, cells were cultured at 27°C for 48 hours before harvesting.

[0096] For MFN2 purification, a mitochondrial fraction was prepared. Cells were harvested by centrifugation at 800 × g for 20 minutes, washed with PBS, and resuspended in a buffer containing 20 mM HEPES (pH 7.5), 70 mM sucrose, 210 mM mannitol, 0.5 mM EDTA, 1 mg / ml BSA, and 1 mM PMSF. Cells were homogenized 80 times on ice using a Dounce homogenizer (Sigma), and the homogenate was centrifuged twice at 1,000 × g for 10 minutes at 4°C. The supernatant was further centrifuged at 10,000 × g for 20 minutes at 4°C to obtain a crude mitochondrial fraction. MFN2 (TargetMol, C0001) was extracted from the crude mitochondrial fraction by treatment with 1.2% n-dodecyl-b-D-maltoside (DDM, Anatrace) in lysis buffer containing 20 mM HEPES (pH 7.5), 500 mM NaCl, 1 mM EDTA, and a 1 / 100 protease inhibitor cocktail at 4°C for 2 hours. The extract was centrifuged at 40,000 × g for 1 hour to remove insoluble components. The supernatant was incubated with anti-Flag G1 affinity resin (Genscript, L00432) at 4°C for 2 hours and then washed three times with 10 column volumes of lysis buffer supplemented with 0.1% DDM (Anatrace). Proteins were eluted with lysis buffer supplemented with 1 mM dithiothreitol (DTT), 0.1% DDM, and 400 μg / ml Flag peptide (Genscript, RP10586).

[0097] For SERCA2 purification, cells were lysed using a Dounce homogenizer in lysis buffer containing 50 mM HEPES (pH 7.0), 100 mM NaCl, 5% glycerol, 1 mM CaCl2, 1 mM MgCl2, 1 mM PMSF, and 1 / 100 protease inhibitor cocktail. Next, SERCA2 was extracted from the membrane fraction using 1% DDM at 4°C for 2 hours, and the sample was centrifuged at 40,000 × g for 1 hour to remove insoluble components. The supernatant was collected and incubated overnight with Ni-NTA resin (GE Health, 17-3712-02). The sample was washed with 10 column volumes of 50 mM HEPES (pH 7.0), 100 mM KCl, 5% glycerol, 1 mM CaCl2, 1 mM MgCl2, 30 mM imidazole, and 0.25 mg / ml C 12The eluted samples were washed three times with E8 (Anatrace) buffer and eluted with the same buffer supplemented with 300 mM imidazole. The eluted samples were subjected to size exclusion chromatography using a 50 mM HEPES (pH 7.0) eluted sample containing 100 mM KCl, 5% glycerol, 1 mM CaCl2, 1 mM MgCl2, 0.25 mg / ml C 12 Superdex200 10 / 300 column (GE Healthcare) in E8 and 1 mM DTT buffer. The target protein in the peak fraction was collected and concentrated to 3-5 mg / ml for further experiments.

[0098] MFN2 pull-down and LC-MS / MS analysis

[0099] Human T cells or 293T cells activated with anti-CD3 / CD28 antibodies were lysed with RIPA buffer containing a 1 / 100 protease inhibitor cocktail. 3× Flag-tagged MFN2 was purified from SF9 insect cells as described above, up to the step where the samples were incubated with Flag affinity resin. Control resin and MFN2-binding resin were incubated with T cell or 293T cell lysates overnight at 4°C. The resin was washed five times with RIPA buffer and eluted with 400 μg / ml Flag peptide. Protein samples were separated by SDS-PAGE and analyzed by mass spectrometry to identify interacting proteins.

[0100] Immunoprecipitation

[0101] 293T cells were transfected with the indicated plasmids for 48 hours and then lysed in ice-cold lysis buffer (1% Triton X-100, 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA) with a protease inhibitor cocktail (TargetMol, C0001). For immunoprecipitation of exogenously expressed MFN2-Flag, SERCA2-Flag, or SERCA2-HA, anti-Flag G1 affinity resin or anti-HA magnetic beads (ThermoFisher Scientific, 88836) were used, and the immunoprecipitated proteins were eluted with Flag peptide or HA peptide (TargetMol, TP1276). For co-immunoprecipitation of endogenous MFN2 and SERCA2, cell lysates were incubated with the indicated antibodies (1-2 μg) at 4°C overnight. Protein A / G magnetic beads (ThermoFisher Scientific, 26162) were added and incubated for an additional hour, after which the beads were boiled with SDS loading buffer for 10 minutes. In both cases, the beads were washed thoroughly at least five times with lysis buffer to remove associated proteins. Immunoprecipitated proteins were resolved by SDS-PAGE and immunoblotted with the indicated antibodies.

[0102] SERCA2 activity assay

[0103] The Ca of purified SERCA2 was measured using a phosphate assay kit (Invitrogen, E6646). 2+ Dependent ATPase activity. Briefly, the assay was performed by mixing 5× reaction buffer, 200 μM 2-amino-6-mercapto-7-methylpurine nucleoside (MESG), 0.1 IU purine nucleoside phosphorylase (PNP), and purified SERCA2 in a 100 μl volume in a 96-well plate in the presence or absence of the indicated MFN1 or MFN2 variants. The 96-well plate was incubated at 37°C for 20 minutes. The SERCA2 was activated by adding 1 mM Ca 2+ The reaction was started with 1 mM ATP (Jena Bioscience, NU-1010), and absorbance was measured at 360 nm every 30 seconds using a Tecan Spark™ 10M reader 30 minutes after the reaction was started at 37° C. The ATP turnover rate was calculated based on the standard curve.

[0104] To determine SERCA2 activity in MFN2-deficient T cells, SERCA2 was immunoprecipitated from whole-cell lysates or crude mitochondrial fractions using appropriate antibodies and cultured in a 5% flask containing 50 mM HEPES (pH 7.0), 100 mM KCl, 5% glycerol, 1 mM CaCl2, 1 mM MgCl2, and 0.25 mg / ml C 12 The cells were washed four times in E8 buffer and resuspended in the same buffer. SERCA2 activity was measured by ATPase activity colorimetric assay kit (NJJCBio, A070-4). Briefly, the immune complex was incubated with reaction buffer at 37°C for 10 minutes and centrifuged at 2,200 × g for 10 minutes at room temperature. The supernatant was then transferred to a 24-well fluorescent plate and the absorbance was measured at 636 nm using a Tecan Spark TM10M reader. SERCA2 activity was measured as C 标准孔 ×(A 测试孔 -A 对照孔 )÷(A 标准孔 -A 空白孔 )×V 总 ÷(C Pr ×V 样本 )÷(T÷60), where C 标准孔 is the concentration of phosphorus standard solution, A 测试孔 is the absorbance of the test well, A 对照孔 is the absorbance of the control well, A 标准孔 is the absorbance of the well containing the phosphorus standard solution, A 空白孔 is the absorbance of the blank well containing deionized water, V 总 is the total volume of the enzymatic reaction, C Pr is the protein concentration of the sample, V 样本 is the volume of sample added to the reaction system, T is the reaction time (min). The relative concentration of SERCA2 in the immune complex was determined by immunoblotting.

[0105] Cell fraction separation To isolate the crude mitochondrial fraction enriched in mitochondria-ER junctions (MEJs), cells were washed and resuspended in separation buffer (20 mM HEPES [pH 7.5], 70 mM sucrose, 210 mM mannitol, 0.5 mM EDTA, 1 mg / ml BSA, and 1 mM PMSF) and homogenized with 50-100 strokes using a glass homogenizer. The cell homogenate was centrifuged twice at 1,000 × g for 10 minutes at 4 ° C, discarding the pellet after each rotation and then further centrifuged at 10,000 × g for 10 minutes. The resulting supernatant (containing ER, Golgi apparatus, and cytoplasm) was collected, and the pellet was lysed with RIPA buffer (containing mitochondria enriched in MEJs) for immunoblotting, or the pellet was resuspended in separation buffer and then plated on PDL-treated coverslips for immunofluorescence experiments using designated antibodies. To prepare a pure mitochondrial fraction, the crude mitochondrial fraction was purified by centrifugation over a 30% Percoll gradient in separation buffer at 100,000 × g for 30 minutes. The resulting mitochondrial layer was washed to remove the Percoll and lysed with RIPA buffer for immunoblotting. The cytosolic fraction was further centrifuged at 20,000 × g for 30 minutes at 4°C and at 100,000 × g for 60 minutes to remove the endoplasmic reticulum.

[0106] Ca 2+ Measurement

[0107] Low affinity Ca 2+ Fluo-5N AM (Invitrogen, F14204) was used to detect Ca in the endoplasmic reticulum. 2 + levels. Briefly, T cells were loaded with 2 μM Fluo-5N AM in RPMI-1640 medium at 37°C for 20 minutes, washed twice with HBSS buffer containing 20 mM HEPES (pH 7.4), 150 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, and 10 mM D-glucose, and kept in the same buffer. Cells were analyzed by flow cytometry (Beckman, CytoFlex) with an excitation wavelength of 488 nm and an emission wavelength of 525 nm. 2+ The concentration was quantified as MFI. 2 + indicator Rhod-2 AM (Invitrogen, R1245MP) or mitochondrial-targeted Ca 2+ Fluorescence resonance energy transfer (FRET) reporter gene 4mtD3cpv was used to measure mitochondrial Ca 2+For Rhod-2 AM measurements, T cells were loaded with 2 μM Rhod-2 AM in RPMI-1640 medium at 37°C for 20 minutes, washed twice, and maintained in HBSS buffer. The presence of Rhod-2 in mitochondria was confirmed by confocal fluorescence microscopy (Olympus). Cells were analyzed by flow cytometry (Beckman, CytoFlex) with an excitation wavelength of 561 nm and an emission wavelength of 585 nm. 2+ Concentrations were quantified as MFI. For FRET-based measurements, αCD3 / CD28-activated T cells were electrotransfected with 4mtD3cpv together with the mito-ER linker or a control plasmid. The presence of the mito-ER linker was confirmed by immunoblotting. Cells were washed 48 hours after transfection and maintained in HBSS buffer. The presence of 4mtD3CPV in mitochondria was confirmed by confocal fluorescence microscopy and cells were analyzed at 37°C using a Tecan Spark TM10M reader with excitation at 488 nm and two emission filters (490 nm for CFP and 535 nm for YFP). Mitochondrial Ca 2+ Levels were calculated as the YFP / CFP emission ratio.

[0108] electron microscope

[0109] The T cells separated were fixed at room temperature with 2.5% glutaraldehyde diluted in 0.1M phosphate buffer, and then treated with 1% osmium tetroxide. Next, cells were dehydrated in an ethanol gradient series (50%, 70%, 90%, 99% and 100%), embedded and sliced ​​(70nm) for electron microscopy analysis. Slices were observed on a FEI Tecnai transmission electron microscope (FEI) operated at 80kV, and images were acquired using a 1K × 1K CCD camera (Gatan). Cells were randomly selected on the slices, and images were taken with 5,800 × and 18,500 × magnifications.

[0110] Generation of dendritic cells (DCs) and tumor-specific T cells

[0111] Tumor-specific T cells were generated according to an earlier protocol (Kryczek et al., 2011). +Monocytes were obtained from the peripheral blood of healthy donors and cultured in VIVO medium containing 100 ng / ml GM-CSF (GenScript, Z02983) and 30 ng / ml IL-4 (GenScript, Z02925) for 5 days, and half of the medium (by volume) was replaced with fresh medium and cytokines every 3 days. On day 6, DCs were matured by incubation with 10 ng / ml TNF-α (GenScript, Z02682) for 24 hours and then incubated with HLA-A2 + Tumor cell lysates from primary tumor cells and PDX tumor xenografts were pulsed by freeze-thawing with liquid nitrogen for 24 h (200 μg protein / 1×10 6 To generate tumor-specific CTLs, CD8 + CD8 T cell isolation kit (Miltenyi, 130-096-495) was used to isolate CD8 T cells from the peripheral blood of the same healthy donor. + T cells. CD8 + T cells and DCs were co-cultured at a ratio of 5:1 in VIVO medium supplemented with 25 IU / ml IL-2 (GenScript, Z03074) for 6 days.

[0112] Glucose concentration and fatty acid measurements

[0113] Interstitial fluid from human ccRCC samples was collected by centrifugation and snap-frozen in liquid nitrogen as previously described (Zhang et al., 2017). Glucose concentration and fatty acid levels were measured using a Glucose Colorimetric / Fluorometric Kit (Sigma, MAK263) and a Free Fatty Acid Quantitation Kit (Sigma, MAK044) according to the manufacturer's instructions.

[0114] Preparation of ccRCC conditioned medium

[0115] Conditioned medium was obtained by incubating primary renal tumor cells (clear cell renal cell carcinoma (ccRCC), 80%-90% density) with fresh conventional culture medium for 48 hours before five freeze-thaw cycles. Cancer cell culture supernatant was obtained by centrifugation (15,000 × g, 1 hour, 4°C) and stored at -80°C for subsequent experiments.

[0116] Cytotoxicity assay

[0117] Primary ccRCC tumor cells were labeled with the cell proliferation dye eFluor 670 (Invitrogen, 65-0840) at 37°C for 10 minutes. Tumor-specific CD8 + T cells were co-cultured with relevant target tumor cells at an effector / target (E / T) ratio of 10:1 in round-bottom 96-well plates at 37° C. for 10 hours. Next, all cells were harvested, stained with PI (100 μg / ml, Beyotime, ST511), and immediately analyzed by flow cytometry.

[0118] Transduction of primary T cells

[0119] Primary human or mouse CD8 + T cells. Lentiviral vectors were used to transduce shRNA against MFN2 and SERCA2 (human) or recombinant plasmids encoding MFN2 variants (mouse) into T cells. To produce lentivirus, 293T cells were transfected with lentiviral vectors and packaging vectors by PEI (Polysciences, 24765-1). The supernatant containing the virus was collected 48 hours and 72 hours after transfection and concentrated by centrifugation at 1,600 × g in ultrafiltration tubes (Millipore). Transduction of primary T cells was performed as described previously with some modifications (Liu et al., 2020). Briefly, 2 μg / ml anti-CD3 / CD28 antibodies or 10 nM OVA bound to the plate were used in the presence of 100 IU / ml IL-2. 257-264 Peptide-stimulated CD8 + T cells were incubated for 24 hours. Activated T cells were incubated with concentrated lentivirus (MOI = 25) supplemented with 8 μg / ml polybrene (Sigma, TR-1003), centrifuged at 800 × g for 90 minutes at 32°C, and cultured for 8-10 hours. The infection process was repeated the next day, and cells were cultured in fresh VIVO medium supplemented with 100 IU / ml IL-2.

[0120] In vivo bioluminescence imaging

[0121] To examine the distribution of T cells in vivo, human CD8 + T cells. For in vivo bioluminescence imaging, D-luciferin (PerkinElmer, 122799) was injected intraperitoneally and imaged for 1 min using the In-VivoFX PRO system (Bruker). Bioluminescence flux (photons / s / cm 2 / steradian) was used to determine T cell distribution.

[0122] Adoptive cell transfer therapy

[0123] From healthy HLA2 + CD8 + T cells were stimulated with anti-CD3 / CD28 antibodies in the presence of 25 IU / ml IL-2 for 48 hours. + T cells were infected with lentivirus as described above and incubated with DCs for another 3 days to obtain tumor antigen-specific CD8 + T cells. For patient-derived xenograft (PDX) transfer models, CD8 + T cells were transduced with a lentiviral vector expressing luciferase and with a control vector or a recombinant MFN2 overexpression plasmid. DCs were generated and pulsed as described above and then incubated with transduced CD8 + T cells were co-cultured at a ratio of 1:5 for 4 days. Next, after tumor formation, 2.5×10 6 CD8 + T cells and 0.5×10 6 DCs were intravenously infused into each tumor-bearing mouse. For anti-PD-1 treatment, anti-human PD-1 antibody (100 μg per mouse, BioXcell, BE0188) was injected intraperitoneally every 5 days thereafter. Tumor growth was monitored and recorded weekly, and tumor volume was estimated as follows: V = (length × width 2 ) / 2.

[0124] result:

[0125] Tumor-infiltrating CD8 + MFN2 expression on T cells is associated with better survival in cancer patients

[0126] To understand the role of MFN2 in CD8 + To investigate the clinical significance of CD8 TIL expression, the inventors collected 116 tumor samples from patients with renal clear cell carcinoma (ccRCC) and analyzed these tumor samples by immunohistochemistry (IHC). + Patients with higher MFN2 expression in TILs had longer overall survival and disease-free survival ( Figure 1 A and 1B). High MFN2 expression is associated with CD8 + More frequent tumor infiltration by T lymphocytes ( Figure 1 C) Single-cell transcriptome analysis data obtained from immune cells of melanoma patients observed a consistent trend (Sade-Feldman et al., 2018), as the inventors found that CD8 + TILs had significantly higher levels of MFN2 ( Figure 1 D and 1E). These findings suggest that CD8+ The expression of MFN2 in TILs was positively correlated with the patient's treatment outcome.

[0127] By isolating CD8 + Further analysis of TILs also confirmed a positive correlation between MFN2 expression and key genes involved in effector function and mitochondrial metabolism, such as IFNG, ATP5A, and CPT1A ( Figure 1 F and 1G). The inventors further found that CD8 + TILs have higher expression of MFN2 than CD8 + TILs are more susceptible to apoptosis, as indicated by higher levels of cleaved caspase-3 ( Figure 1 H).

[0128] Ablation of MFN2 impairs CD8 by disrupting mitochondrial metabolism + T cell effector function

[0129] To explore how MFN2 regulates T cell function, the inventors flox / flox Mice and CD4 Cre Mice were crossed to generate mice with T cell-specific deletion of Mfn2 (termed Mfn2 CKO mice, Figure 3 A and 3B), and then used B16 melanoma and MC38 colorectal cancer models to test the importance of MFN2 in anti-tumor immunity. CKO The tumor progression of mice was faster than that of wild-type (WT) mice ( Figure 2 A, Figure 3 C) from Mfn2 CKO Mouse CD8 + TILs exhibited impaired IFN-γ production and proliferation, and showed elevated apoptosis rates ( Figure 2 B-2D, Figure 3 D) When CD8 + When T cells were depleted by anti-CD8 antibody, WT and Mfn2 CKO Mice were also sensitive to B16 tumor inoculation, as shown by tumor growth and mouse survival ( Figure 2 E and 2F). Unlike WT mice, Mfn2 CKO B16 tumor-bearing mice did not respond to anti-PD-1 antibody treatment ( Figure 2 G and 2H). These results show that the elimination of MFN2 disrupts CD8 + Effector function of T cells.

[0130] From WT and Mfn2 CKOCD8 isolated from B16 tumor-bearing mice + Transcriptome analysis of TILs revealed significant transcriptional changes ( Figure 2 I, Figure 3 E and 3F). Gene ontology (GO) enrichment analysis of differentially expressed genes (DEGs) showed that compared with WT CD8 + Compared with MFN2-deficient CD8 T cells + Pathways related to T cell activation, metabolism, mitochondrial membrane organization, and endoplasmic reticulum homeostasis are downregulated in T cells ( Figure 2 J). Gene set variation analysis (GSVA) ​​showed that MFN2-deficient CD8 + Several metabolic pathways are impaired in T cells, including fatty acid metabolism, oxidative phosphorylation, and lipogenesis ( Figure 2 K) Subsequent Seahorse experiments confirmed that MFN2-deficient CD8 + Maximum mitochondrial respiration in T cells was greatly reduced ( Figure 2 L). Compared with WT CD8 + Compared with TIL, MFN2-deficient CD8 + TILs internalized lower amounts of BODIPY-labeled fatty acid analogs, indicating impaired lipid metabolism ( Figure 2 M). Therefore, MFN2-deficient CD8 + Impaired effector function of TILs may be the result of disturbed mitochondrial metabolism.

[0131] MFN2-mediated mitochondria-endoplasmic reticulum contact affects CD8 + Mitochondrial metabolism is crucial in T cells

[0132] MFN2 is known to regulate mitochondrial metabolism by mediating mitochondrial fusion and / or mitochondrial-endoplasmic reticulum contact (Schrepfer and Scorrano, 2016). First, the inventors conducted experiments to determine whether the mitochondrial fusion activity of MFN2 plays a role in regulating CD8 + TIL mitochondrial metabolism plays a major role. CKO Splenic CD8 isolated from B16 tumor-bearing mice + Mitochondria in T cells were compared with splenic CD8 T cells isolated from WT mice. + Mitochondria are more fragmented in T cells, but not in WT CD8 T cells in the B16 and MC38 models. + TIL and Mfn2 CKO CD8 + No significant morphological differences were observed between mitochondria of TILs, implying that MFN2-mediated mitochondrial fusion plays a role in regulating CD8+ TILs play only a minor role in mitochondrial metabolism ( Figure 4 A and 4B, Figure 5 A and 5B).

[0133] On the other hand, compared with WT CD8 + Compared with splenic T cells and TILs, MFN2-deficient CD8 + In splenic T cells and TILs, mitochondria-endoplasmic reticulum contacts were greatly attenuated as indicated by colocalization of COX IV and calnexin staining ( Figure 4 A and 4C, Figure 5 A and 5C). Compared with the + T cell mitochondria compared to those from MFN2-deficient CD8 + Mitochondria extracted from T cells contain fewer endoplasmic reticulum membrane-bound Figure 4 D) and CD8 + Knockdown of MFN2 in T cells leads to decreased mitochondria-endoplasmic reticulum contacts ( Figure 5 D) Mitochondrial-ER contacts promote ER Ca 2+ MFN2 is transported into mitochondria, which is necessary for promoting mitochondrial metabolism (Jouaville et al., 1999). Therefore, the inventors used the specific fluorescent probe Rhod-2 to examine the effect of WT and MFN2-deficient CD8 + Mitochondrial Ca in TILs 2+ Levels. Compared with WT CD8 + Compared with TIL, MFN2-deficient CD8 + Mitochondrial Ca2+ in TILs 2+ Reduced levels ( Figure 4 E), and blocks mitochondrial Ca by Ru360 2+ Influx reduces lipid metabolism and CD in B16 tumor-bearing mice 8+ TIL produces IFN-γ ( Figure 4 F and 4G). These results indicate that MFN2-mediated mitochondria-endoplasmic reticulum contact is essential for CD8 + These data suggest that MFN2-mediated mitochondria-endoplasmic reticulum contact is essential for promoting CD8 + Clear and critical factors for mitochondrial metabolism and effector functions of TILs.

[0134] MFN2 interacts with SERCA2 on the endoplasmic reticulum to mediate mitochondria-endoplasmic reticulum contacts

[0135] To determine how MFN2 mediates mitochondria-endoplasmic reticulum contacts, we performed mass spectrometry analysis of the MFN2 interactome in HEK293T cells and T cells and identified sarcoplasmic / endoplasmic reticulum calcium ATPase 1 / 2 / 3 (SERCA1 / 2 / 3, or ATP2A1 / 2 / 3) as potential MFN2 interactors on the endoplasmic reticulum (ER). Figure 6 A, Figure 7 A).

[0136] SERCA is an endoplasmic reticulum channel that transfers Ca to 2+ Pumped from the cytosol to the endoplasmic reticulum (Dyla et al., 2020; Zhao et al., 2017). Since SERCA2 is widely expressed in human tissues, the inventors selected SERCA2 for subsequent validation experiments. The sequence of SERCA2 is as follows:

[0137]

[0138] When co-expressed in HEK293T cells, HA-tagged SERCA2 (SERCA2-HA) was co-immunoprecipitated with Flag-tagged MFN2 (MFN2-Flag), and vice versa ( Figure 6 B). Overexpressed Flag-tagged MFN2 or Flag-tagged SERCA2 were co-immunoprecipitated with endogenous SERCA2 or MFN2, respectively ( Figure 7 B and 7C), while the SERCA2 binding ability of MFN1 was negligible in the same experiment ( Figure 7 D) Binding between endogenous MFN2 and SERCA2 in CD8 + T cells by co-immunoprecipitation experiments using specific monoclonal antibodies ( Figure 6 C) and in HeLa cells by using immunofluorescence imaging ( Figure 7 E) was verified. + In mitochondria extracted from T cells, SERCA2 and MFN2 colocalized ( Figure 6 D) When CD8 + When SERCA2 was knocked down in T cells, these mitochondria contained much less calnexin, indicating reduced mitochondrial-endoplasmic reticulum contact. Figure 6 E). To examine whether MFN2 and SERCA2 are in direct physical contact, the inventors purified both proteins from insect cells and applied them to in vitro pulldown assays, in which a direct interaction between Flag-tagged MFN2 and His-tagged SERCA2 was confirmed ( Figure 6F). In summary, these results demonstrate that CD8 + Interaction between MFN2 and SERCA2 at mitochondria-endoplasmic reticulum contact sites in T cells. + Melanoma patients with high expression of both MFN2 and SERCA2 in TILs had the best overall survival ( Figure 6 H).

[0139] The function of MFN2 depends on GTP hydrolysis-coupled conformational changes and oligomerization. To explore whether these features are crucial for its interaction with SERCA2, the inventors introduced four single point mutations (T105M, T130A, R94Q and R259A) into MFN2 ( Figure 6 I). These MFN2 mutants are all incapable of mediating mitochondrial fusion, but through different mechanisms: T105M and T130A affect GTP loading and hydrolysis, R94Q prevents the normal conformational change of MFN2 by rendering the hinge between the two domains inoperable, and R259A does not affect intrinsic GTP hydrolysis but prevents MFN2 from homodimerizing through the GTPase domain (Detmer and Chan, 2007; Li et al., 2019). According to co-immunoprecipitation experiments, except for MFN2(R259A), all other mutants failed to strongly interact with SERCA2 ( Figure 6 J). Building on this, the inventors further introduced additional single-point mutations (V69F, L76P, R280H, W740S, and P251A) into MFN2. They found that, with the exception of P251A, all other mutations strongly interacted with SERCA2. These results suggest that the intact GTPase machinery and conformational flexibility of MFN2, rather than its homodimerization ability, are essential for its interaction with SERCA2.

[0140] CD8 + MFN2-SERCA2 interaction in T cells is essential for optimal anti-tumor immunity

[0141] To understand the role of MFN2-SERCA2 interaction in CD8 + The importance of Mfn2 in T cells flox / flox CD4 Cre The OT-I T cell receptor (TCR) transgenic mouse model was established. Figure 8 A, Figure 9 A and 9B). The resulting Mfn2 CKO OT-I mice develop CD8 T cells that specifically recognize ovalbumin (OVA) peptide antigens +T cells and can be easily tracked by flow cytometry after adoptive transfer and tumor residence. CKO mice, compared with spleen CD8 isolated from WT OT-I mice + T cell mitochondria-endoplasmic reticulum contacts compared to those from Mfn2 CKO Splenic CD8 isolated from OT-I mice + The MFN2(R259A) mutant, which interacts with SERCA2, has reduced mitochondria-endoplasmic reticulum contacts in T cells. - / - )OT-I CD8 + Overexpression (OE) in T cells completely or largely restored mitochondrial-endoplasmic reticulum binding to WT levels, whereas overexpression of mutant MFN2(R94Q), a mutant unable to bind SERCA2, had no effect ( Figure 8 B and 8C, Figure 9 C).

[0142] Next, the inventors transformed Mfn2 cells with MFN2 or its mutants. - / - OT-I CD8 + T cells were adoptively transferred into B16-OVA tumor-bearing mice to detect the Mfn2 - / - OT-I CD8 + Compared with the WT group, Mfn2 - / - OT-I CD8 + The anti-tumor ability of T cells is reduced. - / - OT-I CD8 + MFN2-OE (overexpression) in T cells resulted in a significant slowing of tumor growth to that of WT OT-I CD8 + T cell efficacy. Mfn2 with MFN2 mutants (R259A, V69F, L76P, R280H, or W740S)-OE that can bind to SERCA2 - / - OT-I CD8 + T cells maintain partial activity, that is, they can effectively inhibit tumor growth, but cells with MFN2 mutants (R94Q or P251A)-OE that cannot effectively bind to SERCA2 (i.e., weak binding or no binding) have no effect on inhibiting tumor growth ( Figure 8 D, 8E, 8O and 8P).

[0143] The inventors isolated these adoptively transferred OT-I CD8 + T cells and examined their effector functions. + Compared with TIL, Mfn2- / - OT-I CD8 + TILs have reduced IFN-γ production; MFN2-OE can rescue this phenotype, and MFN2(R259A, V69F, L76P, R280H, or W740S)-OE can rescue this phenotype to some extent, that is, effectively increase the secretion level of IFN-γ, but MFN2(R94Q or P251A)-OE cannot rescue this phenotype ( Figure 8 F, 8G and 8Q). The inventors then evaluated these OT-I CD8 + Mitochondrial-endoplasmic reticulum binding status and mitochondrial morphology of TILs. Although similar mitochondrial fragmentation phenotypes were observed in all groups ( Figure 9 D), but their mitochondria-endoplasmic reticulum binding levels were different and closely correlated with the anti-tumor activity of each group. + Compared with TIL, Mfn2 with MFN2-OE or MFN2(R259A)-OE - / - OT-I CD8 + TILs have similar or only slightly fewer mitochondria-endoplasmic reticulum contacts, whereas Mfn2 with MFN2(R94Q)-OE - / - OT-I CD8 + TILs have significantly reduced mitochondria-endoplasmic reticulum contacts ( Figure 8 H) in the corresponding OT-I CD8 + A similar trend was observed in the ATPase activity of SERCA2 bound to mitochondria in T cells ( Figure 8 I, Figure 9 E) In addition, in mitochondrial Ca 2+ level( Figure 8 J and 8K), lipid metabolism ( Figure 8 L and 8M), and survival in the TME ( Figure 8 N) aspect, Mfn2 with MFN2-OE or MFN2(R259A)-OE - / - OT-I CD8 + TIL outperforms Mfn2 with MFN2(R94Q)-OE - / - OT-I CD8 + These data demonstrate that MFN2-SERCA2 interaction plays a crucial role in maintaining CD8 TIL. + The key role of MFN2-SERCA2 interaction in the mitochondria-endoplasmic reticulum contact state of T cells was demonstrated. + Critical role of mitochondrial metabolism in T cells and their anti-tumor activity.

[0144] Promoting MFN2 expression can improve CD8-based+ T cells for cancer therapy

[0145] Since MFN2 is effective for CD8 + T cell metabolism, function, and survival in the TME are crucial, and targeting MFN2 can promote CD8 + To better evaluate the antitumor activity of T cells by manipulating CD8 + To investigate the effect of MFN2 expression in T cells, the inventors generated conditioned medium ( Figure 11 A) CD8 derived from human PBL cultured in normal medium or ccRCC conditioned medium + Overexpression of MFN2 in T cells leads to increased mitochondrial-endoplasmic reticulum contact, mitochondrial metabolism, and IFN-γ production ( Figure 10 A-10D, Figure 11 B) In addition, compared with CD8 + Compared with the T cells, MFN2-OE had an inhibitory effect on CD8 T cells cultured in ccRCC conditioned medium. + This function-promoting effect of T cells is more significant ( Figure 10 A-10D), which shows that TME-induced CD8 + T cell dysfunction can be corrected by promoting MFN2 expression.

[0146] In view of the above results, the inventors tested the possibility of enhancing MFN2 expression as a potential therapeutic strategy. + T cell co-culture with HLA-A2 + Compared with primary renal tumor cells, antigen-specific MFN2-OE CD8 + T cells co-cultured with HLA-A2 + The apoptosis rate of primary renal tumor cells doubled ( Figure 10 E. Figure 11 C and 11E). Next, the inventors used luciferase encoding plasmids to express primary ccRCC antigen-specific CD8 + We overexpressed MFN2 in T cells and then injected these cells into ccRCC PDX mice, followed by anti-PD-1 treatment every 5 days ( Figure 10 F, Figure 11 D) Within 4 hours after injection, adoptively transferred CD8 + T cells accumulated in the lungs of mice ( Figure 10 G). MFN2-OE CD8 + T cells were present in tumors 5 weeks after injection, but control cells were absent ( Figure 10G and 10H). These MFN2-OE CD8 + T cell tumor retention ( Figure 10 I) Compared with the control group, adoptively transferred antigen-specific MFN2-OE CD8 + Tumor growth in PDX mice expressing T cells was strongly inhibited ( Figure 10 J), in which the level of IFN-γ in the tumor was significantly increased ( Figure 10 K). Similar results were observed in experiments evaluating T cell transfer using different human PDX models ( Figure 11 F-11H). Finally, the inventors treated B16 tumor-bearing mice with leflunomide, a commonly used anti-rheumatoid arthritis drug that promotes the expression of MFN2 (Miret-Casals et al., 2018). In the inventors' experiments, leflunomide enhanced the expression of CD8 + MFN2 expression in T cells ( Figure 10 L) and mitochondria-endoplasmic reticulum contacts ( Figure 10 For B16 tumor-bearing mice receiving PD-1 blockade therapy, supplementation with a relatively low dose of leflunomide (4 mg / kg every 3 days, a dose that did not inhibit B16 tumor growth in C57BL / 6 or nude mice) further limited tumor progression and prolonged survival ( Figure 10 N and 10O, Figure 11 I). Collectively, these results suggest that increasing CD8 + MFN2 expression in T cells may be an effective adjuvant strategy to improve the efficacy of cancer immunotherapy.

[0147] References:

[0148] Anders, S., Py1, PT, and Huber, W (2015). HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166-169.

[0149] Bantug, GR, Galluzzi, L., Kroemer, G., and Hess, C. (2018b). The spectrum of T cell metabolism in health and disease. Nature reviews Immunology 18, 19-34.

[0150] Buck,M.D.,OvSullivan,D.,Klein Geltink,R.I.,Curtis,J.D.,Chang,C.H.,Sanin,D.E.,Qiu,J.,Kretz,O.,Braas,D.,van der Windt,G.J.,et al.(2016).Mitochondrial Dynamics Controls T Cell Fate through MetabolicProgramming.Cell 166,63-76.

[0151] Butler,A.,Hoffman,P.,Smibert,P.,Papalexi,E.,and Satija,R.(2018).Integrating single-cell transcriptomic data across different conditions,technologies,and species.Namre biotechnology 36,411-420.

[0152] Chang,C.H.,Qiu,J.,O′Sullivan,D.,Buck,M.D.,Noguchi,T.,Curtis,J.D.,Chen,Q.,Gindin,M.,Gubin,M.M.,van der Windt,G.J.,et al.(2015).MetabolicCompetition in the Tumor Microenvironment Is a Driver of CancerProgression.Cell 162,1229-1241.

[0153] DePeaux,K.,and Delgoffe,G.M.(2021).Metabolic barriers to cancerimmunotherapy.Nature reviews Immunology 21,785-797.

[0154] Detmer,S.A.,and Chan,D.C.(2007).Complememation between mouse Mfh1 andMfn2 protects mitochondrial fusion defects caused by CMT2A diseasemutations.The Journal of cell biology 176,405-414.Dobin,A.,Davis,C.A.,Schlesinger,F.,Drenkow,J.,Zaleski,C.,Jha,S.,Batut,P.,Chaisson,M.,andGingeras,T.R.(2013).STAR:ultrafast universal RNA-seq aligner.Bioinformatics29,15-21.

[0155] Dyla,M.,Kjaergaard,M.,Poulsen,H.,and Nissen,P.(2020).Structure andMechanism of P-Type ATPase Ion Pumps.Annual review of biochemistry 89,583-603.

[0156] Eura,Y.,Ishihara,N.,Yokota,S.,and Mihara,K.(2003).Two mitofusinproteins,mammalian homologues of FZO,with distinct functions are bothrequired for mitochondrial fusion.Joumal of biochemistry134,333-344.

[0157] Filadi,R.,Pendin,D.,and Pizzo,P.(2018).Mitofusin 2:from functions todisease.Cell death&disease 9,330.

[0158] Gao,S.,and Hu,J.(2021).Mitochondrial Fusion: The Machineries In andOut.Trends in cell biology 31,62-74.

[0159] Hamaidi,I.,Zhang,L.,Kim,N.,Wang,M.H.,Iclozan,C.,Fang,B.,Liu,M.,Koomen,J.M.,Berglund,A.E.,Yoder,S.J.,etal.(2020).Sirt2 Inhibition EnhancesMetabolic Fitness and Effector Functions of Tumor-Reaetive T Cells.Cellmetabolism 32,420-436e412.

[0160] Hamanaka,R.B.,and Chandel,N.S.(2012).Targeting glucose metabolism forcancer therapy.The Journal of experimental medicine 209,211-215.

[0161] Hanzelmann,S.,Castelo,R.,and Guinney,J.(2013).GSVA:gene set variationanalysis for microarray and RNA-seq data.BMC bioinformatics14,7.

[0162] Hegde,P.S.,and Chen,D.S.(2020).Top 10Challenges in CanCerImmunotherapy.Immunity 52,17-35.

[0163] Jouaville,L.S.,Pinton,P.,Bastianutto,C.,Rutter,G.A.,and Rizzuto,R.(1999).Regulation of mitochondrial ATP synthesis by calcium: evidence for along-term metabolic priming.Proceedings of the National Academy of Sciencesof the United States ofAmerica 96,13807-13812.

[0164] Kryczek,I.,Zhao,E.,Liu,Y.,Wang,Y.,Vatan,L.,Szeliga,W.,Moyer,J.,Klimczak,A.,Lange,A.,and Zou,W.(2011).Human TH17cells are long-lived effectormemory cells.Science translational medicine 3,104ra100.

[0165] Labbe,K.,Murley,A.,and Nunnari,J.(2014).Determinants and functions ofmitochondrial behavior.Annual review of cell and developmental biology 30,357-391.

[0166] Li,Y.J.,Cao,Y.L.,Feng,J.X.,Qi,Y.,Meng,S.,Yang,J.F.,Zhong,Y.T.,Kang,S.,Chen,X.,Lan,L.,et al.(2019).Structural insights of human mitofusin-2intomitochondrial fusion and CMT2A onset.Nature communications 10,4914.

[0167] Love,M.I.,Huber,W.,and Anders,S.(2014).Moderated estimation of foldchange and dispersion for RNA-seq data with DESeq2.Genome biology 15,550.

[0168] Miret-Casals,L.,Sebastian,D.,Brea,J.,Rico-Leo,E.M.,Palacin,M.,Fernandez-Salguero,P.M.,Loza,M.I.,Albericio,F.,and Zorzano,A.(2018).Identification of New Activators of Mitochondrial Fusion Reveals a Linkbetween Mitochondrial Morphology and Pyrimidine Metabolism.Cell chemicalbiology 25,268-278e264.

[0169] Picelli,S.,Faridani,O.R.,Bjorklund,A.K.,Winberg,G.,Sagasser,S.,andSandberg,R.(2014).Full-length RNA-seq from single cells using Smart-seq2.Nature protocols 9,171-181.

[0170] Ritchie,M.E.,Phipson,B.,Wu,D.,Hu,Y.,Law,C.W.,Shi,W.,and Smyth,G.K.(2015).limma powers differential expression analyses for RNA-sequencing andmicroarray studies.Nucleic acids research 43,e47.Rosenberg,S.A.,and Restifo,N.P.(2015).Adoptive cell transfer as personalized immunotherapy for humancancer.Science 348,62-68.

[0171] Sade-Feldman,M.,Yizhak,K.,Bjorgaard,SL,Ray,JP,de Boer,CG,Jenkins,RW,Lieb,DJ,Chen,JH,Frederick,DT,Barzily-Rokni,M,et al.(2018).Defining T Cell States Associated with Response to CheckpointImmunotherapy in Melanoma.Cell175,998–1013e1020.

[0172] Schrepfer,E.,and Scorrano,L.(2016).Mitofusins,from Mitochondria toMetabolism.Molecular cell 61,683-694.

[0173] Szabadkai,G.,Bianchi,K.,Varnai,P.,De Stefani,D.,Wieckowski,MR,Cavagna,D.,Nagy,AI,Balla,T.,and Rizzuto,R.(2006).Chaperone-mediatedcoupling of endoplasmic reticulum and mitochondrial Ca2+channels.The Journalof Cell Biology 175,901-911.

[0174] Wolchok,J.D.(2021).Checkpoint blockade:the end of thebeginning.Nature reviews Immunology 21,621.Xing,X.,Yang,F.,Huang,Q.,Guo,H.,Li,J.,Qiu,M.,Bai,F.,and Wang,J.(2021).Decoding the multicellular ecosystem oflung adenocarcinoma manifested as pulmonary subsolid nodules by single-cellRNA sequencing.Science advances 7.

[0175] Youle,R.J.,and van der Bliek,A.M.(2012).Mitochondrial fission,fusion,and stress.Science 337,1062-1065.

[0176] Yu,Y.R.,Imrichova,H.,Wang,H.,Chao,T.,Xiao,Z.,Gao,M.,Rincon-Restrepo,M.,Franco,F.,Geno)et,R.,Cheng,W.C.,et al.(2020).Disturbed mitochondrialdynamics in CD8(+)TILs reinforce T cell exhaustion.Nature immunology 21,1540-1551.

[0177] Zhang,Y.,Kurupati,R.,Liu,L.,Zhou,X.Y.,Zhang,G.,Hudaihed,A.,Filisio,F.,Giles-Davis,W.,Xu,X.,Karakousis,G.C.,etal.(2017).Enhancing CD8(+)T CellFatty Acid Catabolism within a Metabolically Challenging TumorMicroenvironment Increases the Efficacy of Melanoma Immunotherapy.Cancer cell32,377-391 e379.

[0178] Zhao,Y.G.,Chen,Y.,Miao,G.,Zhao,H.,Qu,W.,Li,D.,Wang,Z.,Liu,N.,Li,L.,Chen,S.,et al.(2017).The ER-Localized Transmembrane Protein EPG-3 / VMP1Regulates SERCA Activity to Control ER-Isolation Membrane Contacts forAutophagosome Formation.Molecular cell 67,974-989e976.

[0179] Zheng,X.,Qian,Y.,Fu,B.,Jiao,D.,Jiang,Y.,Chen,P.,Shen,Y.,Zhang,H.,Sun,R.,Tian,Z.,et al.(20l9).Mitochondrial fragmentation limits NK cell-basedtumor immunosurveillance.Nature immunology 20,1656-1667.

Claims

1. Mitochondrial fusion protein 2 (MFN2) or MFN2 variants that can interact with SERCA2 play a role in maintaining and / or promoting CD8 + Use of T cells for tumor killing and / or survival, wherein The use is for non-therapeutic purposes; wherein the MFN2 variant is an MFN2 variant with one of the mutations R259A, V69F, L76P, R280H or W740S.

2. The use according to claim 1, wherein The MFN2 or MFN2 variant is in the form of the protein itself or a vector expressing the protein.

3. The use according to claim 2, wherein The vector is a viral vector.

4. The use according to claim 3, wherein The viral vector is a lentiviral vector, a retroviral vector, or an adenoviral vector.

5. The use according to claim 4, wherein The viral vector is a lentiviral vector.

6. The use according to claim 1 or 2, wherein The MFN2 or MFN2 variant increases CD8 + T cells produce interferon-gamma (IFN-gamma).

7. CD8 overexpression of MFN2 or overexpression of MFN2 variants that can interact with SERCA2 + Use of T cells in the preparation of a cell therapy agent for adoptive cell immunotherapy; wherein, The cell therapy agent is used to treat renal cancer, colorectal cancer or melanoma; wherein the MFN2 variant is an MFN2 variant with one of the mutations R259A, V69F, L76P, R280H or W740S.

8. The use according to claim 7, wherein The CD8 + T cells were transfected with a vector that overexpressed MFN2 or a variant of MFN2 that could interact with SERCA2.

9. The use according to claim 8, wherein The vector is a viral vector.

10. The use according to claim 9, wherein The viral vector is a lentiviral vector, a retroviral vector, or an adenoviral vector.

11. The use according to claim 10, wherein The viral vector is a lentiviral vector.

12. The use according to claim 7 or 8, wherein The CD8 + T cells are further activated with antigen-presenting cells.

13. The use according to claim 12, wherein The antigen presenting cells are dendritic cells.

14. The use according to claim 7 or 8, wherein The cell therapy is further used in combination with an immune checkpoint blocker.

15. The use according to claim 14, wherein The immune checkpoint blocker is an anti-PD-1 antibody.

Images