WO2024063109A1 - Senescent cell remover - Google Patents

Abstract

The purpose of the present invention is to provide a pharmaceutical composition and a senescent cell remover to be used in the treatment or prevention of NASH. The present invention is: a pharmaceutical composition for the treatment or prevention of NASH, having as an active ingredient a substance that inhibits PD-L1 on the cell membrane from binding with PD-1; the pharmaceutical composition, wherein the substance is selected from the group consisting of anti-PD-L1 antibody and anti-PD-1 antibody; the pharmaceutical composition, wherein the substance is a low-molecular compound that binds specifically with PD-L1; and a remover of p16+ senescent cells, having as an active ingredient one or more selected from the group consisting of anti-PD-L1 antibody and anti-PD-1 antibody.

Description

Senescent cell remover

The present invention relates to a senescent cell removing agent and a method for improving various diseases associated with aging using the same. The present invention also relates to a pharmaceutical composition for the treatment or prevention of non-alcoholic steatohepatitis (NASH).
This application claims priority from US Patent No. 63/408105, filed in the United States on September 20, 2022, the contents of which are hereby incorporated by reference.

Accumulation of aging is one of the main causes of age-related inflammation and causes various age-related diseases (Non-Patent Document 1). However, little is known about the molecular basis underlying this accumulation and its potential as a target to ameliorate the aging process.

Several lines of evidence suggest that senescent cells accumulate in various tissues with age, causing excessive inflammation and, consequently, tissue homeostasis imbalance. A simple explanation for this accumulation is an age-dependent increase in senescence-inducing stimuli, such as telomere shortening and DNA damage. Immune system impairment may also be involved, as it has been recently reported that oncogene-induced senescent hepatocytes and injury-induced senescent stellate cells are eliminated by activated T cells and natural killer cells, respectively (Non-Patent Documents 2 and 3). However, exactly how the immune system monitors cellular senescence during natural aging remains largely unknown.

Childs, B. G. et al. Senescent cells: An emerging target for diseases of aging. Nature Reviews Drug Discovery vol. 16 (2017). Kang, T. W. et al. Senescence surveillance of pre-malignant hepatocytes limits liver cancer development. Nature 479, (2011). Krizhanovsky, V. et al. Senescence of Activated Stellate Cells Limits Liver Fibrosis. Cell 134, (2008). Pardoll, D. M. The blockade of immune checkpoints in cancer immunotherapy. Nature Reviews Cancer vol. 12 (2012). Johmura, Y. et al. Necessary and sufficient role for a mitosis skip in senescence induction. Molecular Cell 55, (2014). Pieren, D. K. J., Smits, N. A. M., van de Garde, M. D. B. & Guichelaar, T. Response kinetics reveal novel features of aging in murine T cells. Scientific Reports 9, ( 2019). Lages, C. S., Lewkowich, I., Sproles, A., Wills-Karp, M. & Chougnet, C. Partial restoration of T-cell function in aged mice by in vitro blockade of the PD-1/PD-L1 pathway. Aging Cell 9, (2010). Channappanavar, R., Twardy, B. S., Krishna, P. & Suvas, S. Advancing age leads to predominance of inhibitory receptor expressing CD4 T cells. Mechanisms of Ageing and Development 130, (2009). Decman, V. et al. Defective CD8 T Cell Responses in Aged Mice Are Due to Quantitative and Qualitative Changes in Virus-Specific Precursors. The Journal of Immunology 188, (2012). Omori, S. et al. Generation of a p16 Reporter Mouse and Its Use to Characterize and Target p16high Cells In Vivo. Cell Metabolism 32, (2020). Sadasivam, S. & DeCaprio, J. A. The DREAM complex: Master coordinator of cell cycle-dependent gene expression. Nature Reviews Cancer vol. 13 (2013). Helmbold, H., Koemm, N., Deppert, W. & Bohn, W. Rb2/p130 is the dominating pocket protein in the p53-p21 DNA damage response pathway leading to senescence. Oncogene 28, (2009). Fan, J. & Bertino, J. R. Functional roles of E2F in cell cycle regulation. Oncogene 14, (1997). Dyson, N., Guida, P., Muenger, K. & Harlow, E. Homologous sequences in adenovirus E1A and human papillomavirus E7 proteins mediate interaction with the same set of cellular proteins. Journal of Virology 66, (1992). Janky, R. et al. iRegulon: From a Gene List to a Gene Regulatory Network Using Large Motif and Track Collections. PLoS Computational Biology 10, (2014). Verfaillie, A., Imrichova, H., Janky, R. & Aerts, S. iRegulon and i-cisTarget: Reconstructing Regulatory Networks Using Motif and Track Enrichment. Current Protocols in Bioinformatics 52, (2015). Zhang, J. et al. Cyclin D-CDK4 kinase destabilizes PD-L1 via cullin 3-SPOP to control cancer immune surveillance. Nature 553, (2018). Lim, S. O. et al. Deubiquitination and Stabilization of PD-L1 by CSN5. Cancer Cell 30, (2016). Rechsteiner, M. & Hill, C. P. Mobilizing the proteolytic machine: Cell biological roles of proteasome activators and inhibitors. Trends in Cell Biology vol. 15 (2005). Radhakrishnan, S. K. et al. Transcription Factor Nrf1 Mediates the Proteasome Recovery Pathway after Proteasome Inhibition in Mammalian Cells. Molecular Cell 38, (2010). Zhang, Y. & Hayes, J. D. The membrane-topogenic vectorial behaviour of Nrf1 controls its post-translational modification and transactivation activity. Scientific Reports 3, (2013). Rodier, F. et al. Persistent DNA damage signalling triggers senescence-associated inflammatory cytokine secretion. Nature Cell Biology 11, (2009). Munioz-Espin, D. & Serrano, M. Cellular senescence: From physiology to pathology. Nature Reviews Molecular Cell Biology vol. 15 (2014). Griffin, G. K. et al. Epigenetic silencing by SETDB1 suppresses tumour intrinsic immunogenicity. Nature 595, (2021). Zhang, S. M. et al. KDM5B promotes immune evasion by recruiting SETDB1 to silence retroelements. Nature 598, (2021). Kalbasi, A. et al. Uncoupling interferon signaling and antigen presentation to overcome immunotherapy resistance due to JAK1 loss in melanoma. Science Translational Medicine 12, (2020). Grosse, L. et al. Defined p16High Sensescent Cell Types Are Indispensable for Mouse Healthspan. Cell Metabolism 32, (2020). Maeso-Diaz, R. et al. Effects of aging on liver microcirculatory function and sinusoidal phenotype. Aging Cell 17, (2018). Vaddeepally, R. K., Kharel, P., Pandey, R., Garje, R. & Chandra, A. B. Review of indicators of FDA-approved immune checkpoint inhibitors per NCCN guidelines with the level of evidence. Cancers vol. 12 (2020). Liu, J. Y. et al. Cells exhibiting strong p16 INK4a promoter activation in vivo display features of senescence. Proc Natl Acad Sci U S A 116, (2019). Schulte, H., Muehlfeld, C. & Brandenberger, C. Age-Related Structural and Functional Changes in the Mouse Lung. Frontiers in Physiology 10, (2019). Gregg, S. Q. et al. A mouse model of accelerated liver aging caused by a defect in DNA repair. Hepatology 55, (2012). Sheth, K. A. et al. Muscle strength and size are associated with motor unit connectivity in aged mice. Neurobiology of Aging 67, (2018). Tung, V. W. K., Burton, T. J., Quail, S. L., Mathews, M. A. & Camp, A. J. Motor performance is impaired following vestibular stimulation in aging mice. Frontiers in Aging Neuroscience 8, (2016). Ogrodnik, M. et al. Cellular senescence drives age-dependent hepatic steatosis. Nature Communications 8, (2017). Matsumoto, M. et al. An improved mouse model that rapidly develops fibrosis in non-alcoholic steatohepatitis. International Journal of Experimental Pathology 94, (2013). Chang, J. et al. Clearance of senescent cells by ABT263 rejuvenates aged hematopoietic stem cells in mice. Nature Medicine 22, (2016). Kazankov, K. et al. The role of macrophages in nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. Nature Reviews Gastroenterology and Hepatology vol. 16 (2019). Gate, D. et al. Clonally expanded CD8 T cells patrol the cerebrospinal fluid in Alzheimer’s disease. Nature 577, (2020). Hashimoto, K. et al. Single-cell transcriptomics reveals expansion of cytotoxic CD4 T cells in supercentenarians. Proc Natl Acad Sci U S A 116, (2019). McDonnell, W. J. et al. High CD8 T-cell receptor clonality and altered CDR3 properties are associated with elevated isolevuglandins in adipose tissue during diet-induced obesity. in Diabetes vol. 67 (2018). Koda, Y. et al. CD8+ tissue-resident memory T cells promote liver fibrosis resolution by inducing apoptosis of hepatic stellate cells. Nature Communications 12, (2021). Baruch, K. et al. PD-1 immune checkpoint blockade reduces pathology and improves memory in mouse models of Alzheimer’s disease. Nature Medicine 22, (2016). Rosenzweig, N. et al. PD-1/PD-L1 checkpoint blockade harnesses monocyte-derived macrophages to combat cognitive impairment in a tauopathy mouse model. Nature Communications 10, (2019). Pfister, D. et al. NASH limits anti-tumour surveillance in immunotherapy-treated HCC. Nature 592, (2021). Martins, F. et al. Adverse effects of immune-checkpoint inhibitors: epidemiology, management and surveillance. Nature Reviews Clinical Oncology vol. 16 (2019). Kondo, T. et al. Notch-mediated conversion of activated T cells into stem cell memory-like T cells for adoptive immunotherapy. Nature Communications 8, (2017). Hogquist, K. A. et al. T cell receptor antagonist peptides induce positive selection. Cell 76, (1994). Simma, O. et al. Identification of an indispensable role for tyrosine kinase 2 in CTL-mediated tumor surveillance. Cancer Research 69, (2009). Merry, T. L. et al. The CSF1 receptor inhibitor pexidartinib (PLX3397) reduces tissue macrophage levels without affecting glucose homeostasis in mice. International Journal of Obesity 44, (2020). Nakanishi, M. et al. Exit from G0 and entry into the cell cycle of cells expressing p21Sdi1 antisense RNA. Proc Natl Acad Sci U S A 92, (1995). Parrinello, S. et al. Oxygen sensitivity severely limits the replicative lifespan of murine fibroblasts. Nature Cell Biology 5, (2003). Johmura, Y. et al. SCF Fbxo22-KDM4A targets methylated p53 for degradation and regulates senescence. Nature Communications 7, (2016). Dobin, A. et al. STAR: Ultrafast universal RNA-seq aligner. Bioinformatics 29, (2013). Anders, S., Pyl, P. T. & Huber, W. HTSeq-A Python framework to work with high-throughput sequencing data. Bioinformatics 31, (2015). Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 15, (2014). Liao, Y., Smyth, G. K. & Shi, W. The R package Rsubread is easier, faster, cheaper and better for alignment and quantification of RNA sequencing reads. Nucleic Acids Research 47, (2019). Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: A Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, (2009). Ritchie, M. E. et al. Limma powers differential expression analyzes for RNA-sequencing and microarray studies. Nucleic Acids Research 43, (2015). Yu, G., Wang, L. G., Han, Y. & He, Q. Y. ClusterProfiler: An R package for comparing biological themes among gene clusters. OMICS A Journal of Integrative Biology 16, (2012). Storer, J., Hubley, R., Rosen, J., Wheeler, T. J. & Smit, A. F. The Dfam community resource of transposable element families, sequence models, and genome annotations. Mob DNA 12, (2021 ). Wolf, F. A., Angerer, P. & Theis, F. J. SCANPY: Large-scale single-cell gene expression data analysis. Genome Biology 19, (2018). Lewis, M. D. et al. A reproducible method for the expansion of mouse CD8+ T lymphocytes. Journal of Immunological Methods 417, (2015). Kitamura, T., Doughty-Shenton, D., Pollard, J. W. & Carragher, N. O. Real time detection of in vitro tumor cell apoptosis induced by CD8 + T cells to study immune suppressive functions of tumor-infiltrating myeloid cells. Journal of Visualized Experiments 2019, (2019).

The main object of the present invention is to provide a pharmaceutical composition and a senescent cell remover for use in the treatment or prevention of NASH.

As a result of extensive research, the present inventors have found that by inhibiting PD-L1 on the cell membrane from binding to PD-1, senescent cells are removed in NASH model animals, leading to fat accumulation and fibrosis in the liver. The inventors have completed the present invention by discovering that liver function is significantly reduced and that indicators of liver function are also improved accordingly.

Pharmaceutical compositions and the like according to the present invention are as follows.
[1] A pharmaceutical composition for the treatment or prevention of non-alcoholic steatohepatitis (NASH), which contains as an active ingredient a substance that inhibits the binding of PD-L1 on cell membranes to PD-1.
[2] The pharmaceutical composition of [1] above, wherein the substance is one selected from the group consisting of anti-PD-L1 antibodies and anti-PD-1 antibodies.
[3] The pharmaceutical composition of [1] above, wherein the substance is a low molecular compound that specifically binds to PD-L1.
[4] The pharmaceutical composition according to any one of [1] to [3], wherein the PD-L1 is a cell membrane protein of liver cells, and the PD-1 is a cell membrane protein of CD8-positive T cells.
[5] The pharmaceutical composition according to any one of [1] to [4] above, which is administered only once to an animal that has developed NASH.
[6] The pharmaceutical composition of [5] above, wherein the animal that has developed NASH has not developed cancer other than liver cancer.
[7] An agent for removing p16-positive senescent cells, which contains as an active ingredient one or more selected from the group consisting of anti-PD-L1 antibodies and anti-PD-1 antibodies.

The pharmaceutical composition and senescent cell removal agent according to the present invention can reduce the proportion of p16-positive (p16 ⁺ ) senescent cells in the liver, and can further normalize the functions in the liver. Therefore, the pharmaceutical composition according to the present invention is effective in treating or preventing NASH.

Diagram showing the indicated PD-L1 ⁺ population (Quie: quiescent; n-Sen: nutlin3a-induced senescence; d-Sen: doxorubicin-induced senescence) of human HCA2 cells as determined by FACS in Example 1. It is. A representative scatter plot is shown in the left panel. Confocal immunofluorescence images using the indicated antibodies of organs from 9-month-old p16-Tom mice labeled by TAM administration at 2.5 months of age in Example 1 (scale bars, 25 μm, white arrows, Tom ⁺ /PD-L1 ⁺ cells, pink arrows, Tom ⁺ /PD-L1 ⁻ cells). In Example 1, Tom ⁺ cells (upper (panel) and the proportion of PD-L1 ⁺ cells in the Tom ⁺ cell population (lower panel). In Example 1, the scheme of the experimental schedule (upper panel) and 3 months old (n=6), 6 months old (n=6), or 9 months old (n=6) (lower panel) ) is a diagram showing the percentage of PD-L1+ cells in Tom+ cells of the indicated organs. FIG. 2 is a diagram showing the median fluorescence intensity (MFI) of ProteoStat signals of protein aggregate staining analyzed by FACS in PD-L1 ⁺ and PPD-L1 ⁻ d-Sen HCA2 cells in Example 1. A representative scatter plot is shown in the right panel. FIG. 2 is a diagram showing a proteasome activity assay (n=4) of selected PD-L1 ⁺ and PPD-L1 ⁻ d-Sen HCA2 cells in Example 1. Relative light units (RLU) per cell represent the luminescent signal for assessing chymotrypsin, trypsin, or caspase-like protease activity. Immunoblotting images using the indicated antibodies on d-Sen HCA2 cells treated with mock, epoxomicin (1 μM for 24 hours), and bortezomib (1 μM for 48 hours) in Example 1. In Example 1, PD-L1 ⁺ population of indicated cells similar to FIG. 1g determined by FACS. In Example 1, the indicated PD-L1 ⁺ population of cells was determined by FACS. In Example 2, the indicated PD-L1 ⁺ populations (Quie: quiescent, n-Sen: nutlin3a-induced senescence, d-Sen: doxorubicin-induced senescence) of mouse lung fibroblasts (MPF) were determined by FACS. FIG. A representative histogram is shown in the left panel. In Example 2, the cytotoxicity of the indicated MPFs (target cells) was evaluated by incubation with mouse activated CD8 ⁺ T cells (effector cells) for 18 hours. FIG. 2 is a diagram showing a volcano plot of differentially expressed genes (DEGs) between n-Sen MPF and Quie MPF in Example 2. Red dots (gray dots in the figure) indicate DEGs identified by FDR<0.05 and Log2FC>0.6. Figure 2 shows a dot plot of gene ontology (GO) enriched for biological process terms in DEGs upregulated in d-Sen MPF compared to Quie MPF in Example 2. Enriched terms were identified by FDR<0.05. FIG. 3 shows the cytotoxicity of Quie MPF supplied by conditioned medium from Quie MPF or n-Sen MPF expressing doxycycline-inducible 3xFlag-IκBα (S32/36A) in Example 2. 2 is a volcano plot of transposable elements (TE) differentially expressed between n-Sen MPF and Quie MPF in Example 2. Blue dots (Sig. TEs) indicate all significant TEs identified by FDR<0.05 and Log2FC>1, red dots (Sig. ERVs) indicate significant ERVs annotated by the Dfam database. shows. FIG. 2 is a diagram showing qPCR analysis of ERV expression in Tom ⁻ cells and Tom ⁺ /CD45 ⁻ cells selected from 6-month-old male p16-Tom mice in Example 2. 2B is a diagram showing the cytotoxicity of PD-L1 ⁺ and PD-L1 ⁻ n-Sen MPFs selected in Example 2, similar to FIG. 2b. FIG. In Example 2, cytotoxicity of n-Sen MPFs expressing mock or PD-L1 was inhibited by 10 μg/mL isotype IgG or anti-PD-1 (αPD-1) treatment during incubation with CD8 ⁺ T cells. FIG. FIG. 3 is a diagram showing UMAP visualization of liver sinusoidal endothelial cells (LSEC) of a 7-month-old p16-Tom mouse in Example 3. FIG. 3 shows a violin plot of the expression of Cd274 (PD-L1) in Tom ⁺ and Tom ⁻ LSECs in Example 3. The red dotted line represents the cutoff threshold for determining PD-L1 ⁺ and PD-L1 ⁻ cells. 3 is a diagram showing a volcano plot of DEGs between PD-L1 ⁺ cells and PD-L1 ⁻ cells in Example 3. FIG. DEGs were identified by FDR<0.05 and Log2FC>0.15. Red dots (gray dots in the figure) represent important DEGs. Dot plot of GO enriched in biological process terms in upregulated DEGs of Tom ⁺ /PD-L1 ⁺ cells compared to Tom ⁺ /PD-L1 ⁻ cells in Example 3. It is. Enriched terms were identified by PD-L1 ⁺ . FIG. 3 shows a category-gene net plot of DEGs between Tom ⁺ /PD-L1 ⁺ cells and Tom ⁺ /PD-L1 ⁻ cells containing the GO terms listed in FIG. 3d in Example 3. FIG. 7 is a diagram showing GSEA between Tom ⁺ /PD-L1 ⁺ LSEC and Tom ⁺ /PD-L1 ⁻ LSEC in Example 3. All GSEA terms were identified by FDR<0.05. Figure 4 shows the experimental schedule (left panel) and the percentage of Tom+ cells (right panel) in the indicated organs from p16-Tom mice with the indicated treatments in Example 4 (IgG+IgG: n = 7, αPD-1 + IgG: n = 7, IgG + αCD8: n = 6, αPD-1 + αCD8: n = 5). All p16-Tom male mice in each group were housed in three independent cages. In Example 4, the proportion of PD-L1 ⁺ cells in Tom ⁺ and Tom ⁻ cells in the indicated organs, as in Figure 4a. Figure 4 shows the experimental schedule (upper panel) and quantification of alveolar size (lower panel) for the indicated mice in Example 4. Representative images (H&E staining) are shown in the lower left panel (scale bar: 100 μm; young individual IgG: n = 6; young individual αPD-1: n = 6; old individual IgG: n = 7: old Individual αPD-1: n=8). All male mice in each group were placed in separate cages of 2 (young individuals) or 3 (old individuals). FIG. 4c shows representative images of livers from the indicated mice (left panel: H&E staining) and quantification of lipid droplet area (right panel) similar to FIG. 4c (scale Bar: 100 μm). Diagram showing the experimental schedule (top panel), representative images of the liver (bottom left panel: H&E staining), and quantification of lipid droplet area (bottom right panel) for the indicated mice in Example 4. (Scale bar: 100 μm, each group n=6). All male mice in each group were placed in two separate cages. FIG. 4 shows a representative fluorescence image (upper panel) of the liver of a p16-Tom mouse treated in the same manner as in FIG. 4e and the percentage of Tom ⁺ cells (lower panel) in Example 4. Tom ⁺ cells were labeled by administering TAM to 3-month-old mice (normal IgG: n = 5, CDA-HFD IgG: n = 6, CDA-HFD αPD-1: n = 7, scale Bar: 200 μm). In Example 1, the gating strategy for populations of cells expressing the indicated immune checkpoint proteins in quiescent Quie or nutlin 3a senescence-induced (n-Sen) HCA2 cells is shown in the lower left panel. A representative scatter plot is shown in the right panel. In Example 1, the expression of SASP factors in tdTomato negative (Tom ⁻ ) and Tom ⁺ /CD45 cells selected from 6-month-old male p16-Tom mice was analyzed by qPCR. It is a figure showing this result. 1c is a diagram showing the gating strategy of the experiment shown in FIG. 1c in Example 1. FIG. In Example 1, in young (2 months old, n=7) and old (24-26 months old, n=4) p16-Tom mice, the ratio of PD-L1 ⁺ cells in Tom ⁻ cells was Determined by FACS. It is a figure showing this result. In Example 1, HCA2 cells were isolated from Quie, replicative senescent cells (r-Sen), n-Sen, doxorubicin-induced senescent cells (d-Sen), and proliferative cells (Prolifer) using the indicated antibodies. Immunoblot of lysate. FIG. 2 is a diagram showing qPCR analysis of PD-L1 and CDC20 expression in selected G1, S, and G2/M HCA2 cells in Example 1. Results were expressed as relative fold change. FIG. 2 shows an immunoblot of lysate from HCA2 cells expressing mock, p130 and Flag-E2F97 using the indicated antibodies in Example 1. Figure 6c shows qPCR analysis of PD-L1 and CDC20 expression in HCA2 cells similar to Figure 6c. In Example 1, HCA2 of Prolifer and d-Sen expressing doxycycline (Doxy)-inducible 3xFlag-E7) in the presence or absence of Doxy (1 μg/mL) using the indicated antibodies. FIG. 3 shows an immunoblot of lysate from cells. The duration of senescence induction by doxorubicin is shown (3rd or 12th day). FIG. 2 is a diagram showing a volcano plot of differentially expressed genes between PD-L1 ⁺ and PD-L1 ⁻ d-Sen HCA2 cells in Example 1. Blue dots indicate DEGs identified by FDR<0.05. Red arrow indicates Cd274(PD-L1). FIG. 6f shows transcription factor (TF) candidates enriched in the promoter regions of genes positively correlated with PD-L1 (R ² <0.95) in the RNA-Seq results in FIG. 6f. The normalized enrichment score threshold was 3 or higher. The transcriptional control network is shown in the right panel (pink: target correlated genes; blue: non-target correlated genes; green: TF candidates; red boxes: E2F1 and CD274). FIG. 2 shows an immunoblot of lysates from sorted PD-L1 ⁻ and PD-L ⁺ 1d-Sen HCA2 cells using the indicated antibodies in Example 1. Relative quantification values of PD-L1 are labeled. 2 is an SA-β-Gal stained image of Quien, n-Sen, and d-Sen mouse lung fibroblasts (MPFs) in Example 2. Representative images are displayed in the right panel (scale bar: 500 μm). Figure 7a shows immunoblotting of lysates from MPF similar to Figure 7a using the indicated antibodies. Figure 2b shows a representative histogram of the results shown in Figure 2b (right panel). The gating strategy for the in vitro T cell killing assay is shown in the left panel. FIG. 2D shows a categorical gene net plot of DEGs between n-Sen MPF and Quie MPF that are involved in GO results as in FIG. 2d. FIG. 6 is a diagram showing GSEA of RNA-Seq results between n-Sen MPF and Quie MPF in Example 2. All GSEA terms were identified by FDR<0.05. In Example 2, doxy-inducible 3×Flag-IκBα (S32/36A) expressing n-Sen MPFs using the indicated antibodies with or without 1 μg/mL doxy treatment for 2 days FIG. 3 shows an immunoblot of lysate. FIG. 2 is a diagram showing qPCR analysis of SASP factor expression in n-Sen MPF with or without doxy-inducible 3×Flag-IκBα (S32/36A) in Example 2. A representative histogram of the results shown in Figure 2e. A representative histogram of the results shown in Figure 2h. In Example 2, the cytotoxicity of sorted PD-L1 ⁺ and PD-L1 ^- n-Sen MPFs (target cells) was evaluated for 18 hours with mouse activated CD8 ⁺ T cells (effector cells) derived from OT-1 mice. It was evaluated by incubating. During co-culture, ovalbumin (OVA) peptide 257-264 was added at 1 μg/mL to induce antigen-specific responses. A representative histogram is shown in the right panel. Figure 2 is an immunoblot of lysate from mock PD-L1 overexpressing (OE) MPF using the indicated antibodies in Example 2. A representative histogram of the results shown in Figure 2i. In Example 2, mock or PD-L1 expressing n-Sen MPFs were treated with 10 μg/mL isotype IgG or αPD-1 during incubation with CD8 ⁺ T cells from OT-1 mice. It is a figure showing cytotoxicity. A representative histogram is shown in the right panel. Figure 3 shows UMAP visualization of Tom ⁺ population (top panel) and Cd274 (PD-L1) expression levels (bottom panel) in the indicated scRNA-Seq dataset and cell types in Example 3. . NASH samples were prepared from 7-month-old p16-Tom mice that were fed CDA-HFD for 6 months. Kidney samples were collected from 5.5 month old p16-Tom mice. Violin plot of the expression levels of Cd274 (PD-L1) in Tom ⁺ and Tom ⁻ cells of the indicated cell types in Example 3. The red dotted line represents the cutoff threshold for determining PD-L1 ⁺ and PD-L1 ⁻ cells. Figure 3 is a volcano plot of DEGs between PD-L1 ⁺ and PD-L1 ^- /Tom ⁺ cells in the indicated samples in Example 3. DEGs were identified by FDR<0.05 and Log2FC>0.15. Red dots (gray dots in the figure) represent important DEGs. Figure 3 is a volcano plot of DEGs between PD-L1 ⁺ and PD-L1 ^- /Tom ⁺ cells in the indicated samples in Example 3. DEGs were identified by FDR<0.05 and Log2FC>0.15. Red dots (gray dots in the figure) represent important DEGs. Figure 3 is a volcano plot of DEGs between PD-L1 ⁺ and PD-L1 ^- /Tom ⁺ cells in the indicated samples in Example 3. DEGs were identified by FDR<0.05 and Log2FC>0.15. Red dots (gray dots in the figure) represent important DEGs. In Example 3, GO terms enriched in upregulated DEGs of PD-L1 ⁺ /Tom ⁺ cells compared to PD-L1 ⁻ /Tom ⁺ cells in the indicated samples. In Example 3, GO terms enriched in upregulated DEGs of PD-L1 ⁺ /Tom ⁺ cells compared to PD-L1 ⁻ /Tom ⁺ cells in the indicated samples. In Example 3, GO terms enriched in upregulated DEGs of PD-L1 ⁺ /Tom ⁺ cells compared to PD-L1 ⁻ /Tom ⁺ cells in the indicated samples. 5 is a diagram showing the results of FACS analysis of splenic CD8 ⁺ T cells of mice treated with αPD-1 and/or αCD8 as in FIG. 4. FIG. The gating strategy and representative scatter plots are shown in the right panel. Representative images of the immunofluorescence staining shown in Figure 4a with the indicated treatments for lung, liver, and kidney (blue: DAPI; red: PD-L1; yellow: tdTomato; green: CD3; scale bar: 500μm). Figure 1 shows the percentage of PD-L1 ⁺ /Tom ⁺ cells (upper panel) and PD-L1 ^- /Tom ⁺ cells from immunofluorescent stained sections of lung, liver, and kidney with the indicated treatments in Example 4. Fold change (FC) between the IgG group and the αPD-1 group without αCD8 treatment is shown in the figure. Representative images of immunofluorescence staining of IgG and αPD-1 treated liver shown in Figure 4a (blue: DAPI; red: PD-L1; yellow: tdTomato; green: F40/80; scale bar: 500 μm) . Quantified percentage of F40/80 in Tom ⁺ cells from immunofluorescence-stained sections of liver with the indicated treatments in Example 4 (IgG: n=5, αPD-1: n=6). FIG. 3 is a diagram showing the percentage of PD-L1 ⁺ cells in F40/80 ⁻ cells (left panel) and F40/80 ⁺ cells (right panel) from immunofluorescence-stained liver sections in Example 4. FIG. 4 is a diagram showing the percentage of PD-L1 ⁺ cells in Tom ⁺ /F40/80 ⁻ cells from an immunofluorescence-stained section of the liver in Example 4. FIG. 4 is a diagram showing the results of a motor ability test of young (3 months old) and old (17.5 months old) wild type mice in Example 4. Dwell time was recorded from the start of rotation of the rotarod model until the mouse fell into the switch. FIG. 11e is a diagram showing the results of a mouse grip strength test similar to FIG. 11e. Similar to Figure 4e, wild animals were fed a normal diet (Normal) or an L-amino acid diet (CDA-HFD) containing 60 kcal of fat, 0.1% methionine, and no choline for 10 weeks together with αPD-1 treatment or ABT263 treatment. FIG. 3 shows activation and infiltration of hepatic CD8 ⁺ T cells analyzed by FACS from type mice. The gating strategy is shown in the right panel. FIG. 12 is a diagram showing the measurement results of serum levels of AST, ALT, and LDH in mice similar to FIG. 12a. Figure 12a is Sirius red staining of mouse liver similar to Figure 12a. Representative images are shown in the top panel. The percentage of red area (collagen area) in each field was quantified and described as liver fibrosis (scale bar: 100 μm). Body weight and serum total cholesterol (T-CHO) values of mice similar to FIG. 12a. Figure 4 is an experimental schedule of PLX3397 treatment for young (3 months old, n=5) and old (17.5 months old, n=5) wild type mice in Example 4. Treatment with 40 mg/kg BW of PLX3397 was administered by oral gavage twice a week for a total of 3 weeks. All male mice in each group were placed in two separate cages. FIG. 13 is a diagram showing the results of a motor ability test of mice shown similarly to FIG. 13a. The dwell time from the start of rotation of the rotarod model until the mouse fell into the switch was recorded. FIG. 13A is a diagram showing the results of a grip strength test of a mouse shown similarly to FIG. 13a. 13A is a diagram showing the results of quantifying the alveolar size of the mouse shown in the same manner as FIG. 13a in Example 4. FIG. Representative images (H&E staining) are shown in the right panel. Schedule of αPD-1 and PLX3397 treatment in wild-type mice fed normal diet or CDA-HFD (n=5 per group). Figure 13e shows the measurement results of serum levels of AST, ALT, and LDH in mice shown in the same manner as in Figure 13e. Representative images of the liver from mice shown as in Figure 13e (bottom panel: H&E staining) and quantification of lipid droplet area (top panel) results. Sirius Red staining of mouse liver shown similarly to Figure 13e. Representative images are shown in the bottom panel (scale bar, 100 μm).

We showed that senescent cells heterogeneously express the immune checkpoint PD-L1 and that PD-L1 ⁺ senescent cells accumulate with age in vivo. PD-L1 ⁻ cells are sensitive to T cell surveillance, whereas PD-L1 ⁺ cells are resistant even in the presence of senescence-associated secretory phenotype (SASP). Interestingly, single cell analysis of p16 ⁺ cells in vivo revealed that PD-L1 expression correlated with higher levels of SASP. Consistent with this, administration of anti-PD-1 antibodies to naturally aging or NASH-induced mice not only reduces the total number of p16 ⁺ cells in vivo but also reduces PD in an activated CD8 ⁺ T cell-dependent manner. -L1 ⁺ population also decreases, and various symptoms of age-related diseases are improved. These results suggest that the heterogeneous expression of PD-L1 plays an important role in senescent cell accumulation and aging-associated inflammation, and that the removal of PD-L1 ⁺ senescent cells by immune checkpoint inhibition may be an effective strategy for anti-aging therapy. This suggests that it may be a promising strategy.

As shown in the Examples below, the accumulation of senescent cells (p16 ⁺ cells) is suppressed by inhibiting the binding between PD-L1 and PD-1. This is because PD-L1-positive (PD-L1 ⁺ ) senescent cells bind to PD-1 on CD8-positive T cells (CD8 ⁺ T cells), suppress the activity of CD8 ⁺ T cells, and perform immune surveillance. However, by inhibiting the binding of PD-L1 and PD-1, the activity of CD8 ⁺ T cells cannot be suppressed, and PD-L1 ⁺ senescent cells are eliminated by CD8 ⁺ T cells. It's for a reason.

Cellular aging plays an important role in the pathology of NASH. As shown in the Examples below, by reducing the amount of senescent cells (p16 ⁺ cells) in the liver, the pathological condition of NASH is improved. , ALT, and LDH serum levels can be improved.

The pharmaceutical composition of the present embodiment is a pharmaceutical composition for treating or preventing NASH, which contains as an active ingredient a substance that inhibits PD-L1 on the cell membrane from binding to PD-1. By inhibiting the binding of PD-L1 on the cell membrane to PD-1, the pathological condition of NASH can be improved by removing senescent cells. Due to this mechanism of action, the pharmaceutical composition of this embodiment is effective in treating and preventing NASH.

The active ingredient of the pharmaceutical composition of the present embodiment is not particularly limited as long as it is a substance that can inhibit PD-L1 on the cell membrane from binding to PD-1. It may be a substance that binds to PD-L1 and inhibits the binding to PD-1, or it may be a substance that binds to PD-1 and inhibits the binding to PD-L1. Since the effect of preventing or treating NASH can be obtained more effectively, the active ingredients of the pharmaceutical composition of this embodiment include PD-L1, which is a cell membrane protein of liver cells, and a cell membrane protein of CD8 ⁺ T cells. Preferably, it is a substance that inhibits binding to PD-1.

Examples of the active ingredient of the pharmaceutical composition of the present embodiment include one or more selected from the group consisting of anti-PD-L1 antibodies and anti-PD-1 antibodies. As the anti-PD-L1 antibody, any known anti-PD-L1 antibody may be used, or it may be an antibody obtained by appropriately modifying a known antibody without impairing its affinity for PD-L1. As the anti-PD-1 antibody, any known anti-PD-1 antibody may be used, or it may be an antibody obtained by appropriately modifying a known antibody without impairing its affinity for PD-1. Improving the specificity and safety of antibodies is a technique that has been used for a long time, and can be carried out as appropriate using these techniques. Furthermore, for example, anti-PD-L1 antibodies and anti-PD-1 antibodies that can be used in immune checkpoint inhibition therapy and the like can also be used as active ingredients in the pharmaceutical composition of the present embodiment.

The active ingredient of the pharmaceutical composition of the present embodiment may be, for example, a low-molecular compound that specifically binds to PD-L1 or a low-molecular compound that specifically binds to PD-1. Further, the active ingredient of the pharmaceutical composition of the present embodiment may be a nucleic acid aptamer or a peptide that specifically binds to PD-L1 or PD-1. PD-L1 and low-molecular compounds that specifically bind to PD-1 can be obtained by screening various libraries. Furthermore, it is also possible to synthesize using the structural data of PD-L1 and PD-1.

By inhibiting the binding of PD-L1 on the cell membrane to PD-1, the proportion of p16-positive (p16 ⁺ ) senescent cells in the liver can be reduced. Therefore, the active ingredient of the pharmaceutical composition of this embodiment also functions as an agent for removing senescent cells, in particular, an agent for removing p16-positive senescent cells.

The pharmaceutical composition of this embodiment can be prepared by appropriately mixing a substance capable of inhibiting the binding of PD-L1 on cell membranes with PD-1, which is an active ingredient, and a pharmaceutically acceptable carrier. , can be prepared. The pharmaceutical composition can be manufactured by a method commonly used in the field of pharmaceutical manufacturing by using appropriate additives as necessary.

Pharmaceutically acceptable carriers are diluents, excipients, binders, Solvents, etc. As the carrier, specifically, for example, water, physiological saline, various buffer solutions, etc. are used. In addition, additives that can be used include adjuvants, diluents, excipients, binders, stabilizers, tonicity agents, buffers, solubilizing agents, suspending agents, preservatives, antifreeze agents, and cryoprotectants. agents, lyoprotectants, bacteriostatic agents, etc.

The animal to which the pharmaceutical composition of the present embodiment is administered is not particularly limited as long as it is an animal for which treatment or prevention of NASH is required, and may be a human or a non-human animal. However, mammals are preferred. Non-human mammals include cows, pigs, horses, sheep, goats, monkeys, dogs, cats, rabbits, mice, rats, hamsters, guinea pigs, and the like. Since a more sufficient therapeutic effect on NASH can be expected, the animal to which the pharmaceutical composition of this embodiment is administered is preferably an animal that has not developed cancer other than liver cancer. Furthermore, the administration route for administering the pharmaceutical composition of the present embodiment to animals is not particularly limited, and may include oral administration, intravenous administration, enteral administration, intramuscular administration, subcutaneous administration, transdermal administration, Examples include nasal administration and pulmonary administration.

The dosage per administration of the pharmaceutical composition of the present embodiment is not particularly limited as long as the active ingredient can achieve the effect of removing senescent cells, and the dosage may include the species, sex, and age of the animal to which it is administered. , weight, presence or absence of underlying disease, pathological condition, etc. can be appropriately determined. For example, when using an anti-PD-L1 antibody or anti-PD-1 antibody used in immune checkpoint inhibition therapy as an active ingredient, refer to the dosage used in immune checkpoint inhibition therapy, etc. , the dosage of the pharmaceutical composition of this embodiment can be determined.

The dosage of the pharmaceutical composition of the present embodiment is not particularly limited as long as the active ingredient has the effect of removing senescent cells, but the administration period is short in order to suppress side effects. It is preferable. For example, the pharmaceutical composition of the present embodiment is preferably administered multiple times within 3 weeks, and preferably multiple times within 2 weeks, to animals that have developed NASH and require treatment. More preferably, it is administered multiple times within one week, and particularly preferably only once.

Next, the present invention will be explained in more detail with reference to examples, but the present invention is not limited to these examples.

[Materials and Methods]

<Mouse>
Mice were placed in cages of 2 to 5 mice at an ambient temperature of 23 to 25°C in a humidity-controlled room, fed standard food (CA-1, manufactured by CLEA Japan), and kept on a 12-hour light-dark cycle ( The lighting was maintained from 08:00 to 20:00). Water was provided freely with access to environmental enrichment. All animals were treated in accordance with the Animal Experiment Guidelines and Institutional Laboratory Animal Management of the Institute of Medical Science, The University of Tokyo. All p16-Tom mice are heterozygous and were generated by crossing p16 ^Ink4a -Cre ^ERT2 mice (Non-Patent Document 10) with Rosa26-CAG-LSL-tdTomato mice (obtained from Jackson Laboratory). Age-matched male mice of the same strain were used in all groups in the same experiment. C57BL/6 OT-1-TCR transgenic mice have already been disclosed (Non-Patent Documents 47, 48). In this mouse, most CD8 ⁺ T cells are reactive to OVA peptide/MHC class I complexes. OT-1 mice were housed in a specific pathogen-free facility at Keio University. All mice used were bred in-house. All p16-Tom mice received tamoxifen (TAM, 80 mg/kg BW) intraperitoneally daily at the indicated times for 5 days. For ICB treatment, all mice received 250 μg/dose of isotype control IgG (BioLegend, clone: RTK2758) for 3 days or anti-mouse PD-1 (BioLegend, clone: 29F.1A12) for 3 weeks. 8 intraperitoneal injections were given within 8 days, and mice were sacrificed 1 day after the last injection. To deplete CD8 ⁺ T cells, all mice were treated with 200 μg/dose of isotype control IgG (Bio X Cell, clone: 2A3) or anti-mouse CD8a (Bio X Cell, clone: 2A3) twice weekly. : 53-6.7) was intraperitoneally administered 14 times within 7 weeks (Non-Patent Document 49).

For the induction of non-alcoholic steatohepatitis (NASH), 1.5-month-old p16-Cre ^ERT2 -tdTomato mice were fed a diet containing 60 kcal% fat, 0.1% methionine, and a choline-free L-amino acid diet (CDA). -HFD: A06071302, manufactured by Research Diets) for a total of 10 weeks. For senolytic reagent treatment, all mice received placebo (vehicle control) or ABT263 by oral gavage. ABT263 (manufactured by Selleck) was suspended in ethanol: polyethylene glycol 400: Phosal 50 PG at a ratio of 10:30:60, and then 50 mg/kg BW was administered daily for 7 days per cycle. A total of 2 cycles were performed with a 2 week interval between cycles (Non-Patent Document 37). For CSF1R inhibition, PLX3397 (Selleck) was suspended in 5:25:70 DMSO:PEG300: _ddH2O and then administered orally by gavage at 40 mg/kg BW twice a week ( Non-patent document 50).

<Cell culture>
Cell culture, cell cycle synchronization, and treatment with various drugs were performed as previously described (Non-Patent Document 5). Early passage hHCA2 (Non-Patent Document 51), primary mouse lung fibroblasts (MPF), and HEK293T cells (obtained from RIKEN Cell Bank) were incubated with 10% fetal bovine serum (FBS) (Sigma) and 1× penicillin. The cells were cultured in DMEM (manufactured by Nacalai Tesque) supplemented with /streptomycin/amphotericin B (manufactured by Nacalai Tesque). All cells were cultured at 37° C. under 5% CO ₂ in normoxic conditions, but only MPFs were cultured in hypoxic conditions (3-5% O ₂ ) (Non-Patent Document 52). For quiescence induction, cells were cultured in DMEM supplemented with 0.5% FBS for 3 days until cell cycle arrest was observed. For senescence induction with RO3306 and Nutlin3a, cells were synchronized to G2 phase by treatment with 9 μM RO3306 (Roche) for 16 to 20 hours, followed by treatment with 9 μM RO3306 and 5 μM Nutlin3a (Sigma-Aldrich). The cells were treated for 8 hours, and then 5 μM nutlin3a was administered for 2 days. For DNA damage aging, cells were treated with 100 nM doxorubicin (Sigma) for 24 hours. Cells were then treated with 100 nM BI-2536 until 12 days after induction to remove proliferating cells. Replicative senescent cells were generated by culturing nearly senescent HCA2 cells (41 passages). Doxycycline, epoxomicin, and bortezomib were used at concentrations of 1 μg/mL, 1 μM, and 1 μM, respectively. Cell number was determined by staining cells with 0.4% trypan blue dye and counting viable cells with a hemocytometer.

<Plasmid construction>
Lentivirus-based shRNA constructs and Tet-on inducible lentivirus constructs were generated as previously described (53). To construct lentivirus constructs, PCR-generated EcoRI/NotI fragments containing human PD-L1, P130, Flag-E2F97, and NRF1Δ1-103 cDNAs were inserted into the CSII-CMV-IRES2-Bsd vector (Invitrogen). To construct Tet-on inducible lentivirus constructs, BamHI/NotI PCR fragments containing IκBα (S32/36A) and HPV16E7 cDNAs were inserted into the BamHI/NotI digested pENTR-1A vector (Invitrogen) containing the 3xFlag epitope. The obtained plasmid was mixed with the CSIV-TRE-RfA-UbC-Puro vector and reacted with Gateway ^™ LR Clonase ^™ Enzyme mix (manufactured by Thermo Fisher Scientific) to generate a lentiviral plasmid.

<Lentivirus production and infection>
Lentivirus generation and cell infection were performed as previously described (53). Lentiviruses expressing the respective genes were generated by cotransfection of HEK293T cells with pCMV-VSV-G-RSV-RevB, pCAG-HIVgp, and the respective CSII-CMV-IRES2-Bsd, or CSIV-TRE-RfA-UbC-Puro plasmids using calcium phosphate coprecipitation. HCA2 cells or MPF infected with the indicated viruses were treated with 10 μg/mL blasticidin (Gibco) or 2 μg/mL puromycin (Gibco) for 2 to 4 days. For inducible expression of the CSIV plasmid system, doxycycline (Sigma-Aldrich) was added to the medium at a concentration of 1 μg/mL.

<Bulk RNA sequencing>
For HCA2 samples, total RNA was extracted using ISOGEN (Nippon Gene) according to the manufacturer's instructions, and quality was assessed using an Agilent Bioanalyzer (Agilent Technologies, Santa Clara, Calif.). PolyA-tailed mRNA was bead-selected from 200 μg of total RNA using the NEBNext Poly(A) mRNA Magnetic Isolation Module (New England Biolabs, Ipswich, Mass.). RNA-seq libraries were prepared using the NEBNext Ultra II Directional RNA Library Prep Kit (New England Biolabs, Ipswich, Mass.) according to the manufacturer's instructions. Sequencing of 100 nucleotide long paired-end reads was performed on a HiSeq2500 (Illumina, San Diego, CA).

Genomon (ver. 2.5.2), an in-house pipeline built at the Human Genome Center, University of Tokyo, was used to identify differentially expressed genes (https://github.com/Genomon-Project). Briefly, sequencing data were aligned to the human genome (hg19) using STAR aligner (v2.5.2a) (54). Raw gene counts were obtained from read alignments using HTSeq (v0.11.2) (55). Differential expression analysis was performed using DESeq2 (v1.22.2) (56). Genes with a B-H adjusted P value < 0.05 were considered significantly differentially expressed. To construct the transcriptional regulatory network, 230 genes whose FPKM showed a positive correlation with PD-L1 (R ² >0.95) were entered into iRegulon (v1.3) (Non-Patent Document 15) on the Cytoscape platform (v3.8.2). The 20 kb region centered on the TSS was considered as the promoter region of each gene, and the normalized enrichment score (NES) was set as the threshold with a cutoff of 3.

For static and d-Sen MPF samples, RNA was extracted using the RNeasy Mini Kit (QIAGEN) according to the manufacturer's instructions. rRNA was removed from total RNA (1 μg) using NEBNext rRNA Depletion Kit (manufactured by New England Biolabs). The RNA-seq library was prepared using NEBNext Ultra II Directional RNA Library Prep Kit (New England Biolabs). After converting the linear dsDNA library into a single-stranded circular DNA library (Universal Library Conversion Kit, MGI Tech, Guangdong, China), the pooled library was converted to DNBSEQ-G400RS ( Sequence was determined using MGI Tech).

Sequencing data were aligned to the mouse reference genome (mm10) using Rsubread (v2.4.3) (Non-Patent Document 57). Raw counts were obtained from read alignments via refGene for gene counts and RepeatMasker annotations for TE, and further transferred to CPM by edgeR (v3.32.1) (58). Both the mm10 refGene and RepeatMasker annotation databases are from the University of California, Santa Cruz (UCSC) Genome Browser. After filtering low expressed genes with CPM <0.1, differential expression was analyzed with a linear model using limma (v3.46.0) (Non-Patent Document 59). Genes with log2FC > 0.6 and FDR < 0.05 adjusted by the BH method are considered as significant differentially expressed genes (DEGs), and are further analyzed using clusterProfiler (v3.18.1) (Non-Patent Document 60) was registered for gene ontology analysis using . For gene set enrichment analysis (GSEA), a list of genes ranked in order by log2 fold change is input into clusterProfiler's pre-ranked GSEA function, and mouse hallmark gene sets are obtained from msigdb (v7.4.1) It was done. All figures were plotted with enrichplot (v1.10.2). In the classification of TEs, significantly changed TEs with log2FC > 1 and FDR < 0.05 were further annotated by Dfam (Release 3.5) (Non-Patent Document 61).

<Single cell RNA sequence data analysis>
For analysis of Tom+/PD-L1+ and Tom+/PD-L1- cells of p16-Tom mice, processed droplet-based single-cell RNA-seq dataset 10 (GSE155182) available from GEO was used. Single-cell RNAseq data (Non-Patent Document 62) was processed using the Scanpy package (ver. 1.4.5.post1) on Python (ver. 3.6.8). The threshold for gating PD-L1+ cells was defined as a Cd274 UMI count greater than 0. For differentially expressed gene (DEG) analysis, genes with FDR less than 0.05 and log2 fold change greater than 0.15 were identified using the dffxpy package (ver. 0.7.3). For gene set enrichment analysis (GSEA), a list of genes ranked in order by log2 fold change was input into the pre-ranked GSEA function of clusterProfiler, and the mouse characteristic gene set was extracted from msigdb (v7.4.1). Obtained.

<RNA isolation and real-time PCR>
Total RNA from cultured cells was extracted using ISOGEN (Nippon Gene) according to the manufacturer's instructions. For RNA from the selected primary cells, Single Cell RNA Purification Kit (manufactured by Norgen) was used according to the manufacturer's instructions. For qPCR analysis, cDNA was synthesized using ReverTra Ace qPCR RT Master Mix (manufactured by Toyobo). Real-time PCR amplification was performed in a 96-well optical reaction plate using THUNDERBIRD SYBR qPCR Master Mix (Toyobo) and StepOnePlus (Applied Biosystems). The relative expression level of each gene was determined by normalization to the β-actin expression level of each sample. All primers are listed in Table 1.

Protein extraction and immunoblotting
For whole cell lysates, cells were directly lysed in Laemmli buffer (2% SDS, 10% glycerol, 5% 2-mercaptoethanol, 0.002% bromophenol blue, and 62.5 mM Tris HCl, pH 6.8). Whole cell lysates were separated by SDS-PAGE, transferred to PVDF membranes (Immobileon-P; Millipore), and immunoblotted with the indicated antibodies using the ECL detection system. β-actin antibody (C-4, Santa Cruz) was used as a loading control, and images were captured using an Amersham Imager 680 (Cytiva). All antibodies are listed in Tables 2 and 3.

<Histology>
For immunofluorescence staining, kidneys, lungs, liver, and colon preserved in OCT compound were frozen sectioned and stained with DAPI. After imaging the fluorescent signals using INCell Analyzer 2500HS (manufactured by Cytiva), this instrument automatically counted Tom+ cells by recognizing DAPI signals as nuclei through IN Cell Developer Toolbox v1.9.2. The percentage was calculated. Tom+ signal was 10 times background. To determine the proportion of PD-L1+ cells among Tom+ cells, anti-PD-L1 (Abcam, clone: EPR20529) and anti-CD3 (Abcam, clone: KT3) staining was performed. PD-L1 intensity was quantified for each Tom+ cell, excluding CD3+ cells and cells exhibiting strong autofluorescence (such as oil droplets in NASH samples). PD-L1+ cells were recognized by PD-L1 intensity that was 1.8 times (for kidney) or 2.5 times (for lung and liver) greater than the median. 10-30 fields of each organ section were randomly selected and out-of-focus images were further excluded. The representative image in Figure 1b was observed on a Carl Zeiss LSM 710 NLO (Zeiss). All antibodies are listed in Tables 2 and 3.

For histochemical staining, the liver and lungs were fixed with paraffin, sectioned, and stained with H&E or Sirius Red. The stained tissue sections were observed using an Olympus BX51 microscope (manufactured by Olympus) under bright field illumination. To quantify alveolar size, lipid droplet area, and Sirius Red area via ImageJ software, define the area using the Huang threshold, 1500, 2000, and 400, respectively, in three fields for each section. The average area over the pixels was measured.

<In vitro T cell killing assay>
Spleens were harvested from C57BL/6 or OT-1 mice and triturated with the flat end of a syringe over a 70 μm cell strainer in 3 mL of PBS. All cells that passed were collected in a tube and pelleted by centrifugation at 400 g for 5 minutes. The cell pellet was resuspended in 1.5 mL of PBS containing 2% FBS and 1 mM EDTA, and the cell concentration was then determined by a hemocytometer. CD8+ T cells were isolated using the EasySep ^™ Mouse CD8+ T Cell Isolation Kit (STEMCELL Technologies) according to the manufacturer's instructions. Then, 6 × ¹⁰ CD8+ T cells were plated in a 24-well plate precoated with 5 μg/mL anti-CD3α (BioLegend, clone: 145-2C11) and 5 μg/mL anti-CD28 (BioLegend, clone: 37.51). , and cultured for 72 hours. Primary mouse CD8+ T cells were incubated with 10% fetal bovine serum (FBS) (Sigma), 1x penicillin/streptomycin/amphotericin B (Nacalai Tesque), 2mM L-glutamine (Thermo Fisher Scientific), and 50μM 2- Cultured in RPMI (manufactured by Nacalai Tesque) supplemented with mercaptoethanol, 1× insulin-transferrin-selenium (manufactured by Thermo Fisher Scientific), and 30 U/mL recombinant mouse IL-2 (manufactured by PeproTech) (non-patent literature). 63). Activated CD8+ T cells were collected and co-cultured with target cells at a ratio of 5:1 for 18 hours (Non-Patent Document 64). During co-culture with T cells isolated from OT-1 mice, 1 μg/mL of ovalbumin peptide (OVA 257-264, SynPeptide) was supplied to activate antigen-specific responses. After co-culture, anti-CD45 (manufactured by BioLegend, clone: 30-F11) and propidium iodide (50 μg/mL) staining were performed to exclude effector cells, and cytotoxicity was evaluated. Cells were analyzed by BD FACSCanto ^™ II (BD Biosciences). All antibodies are listed in Tables 2 and 3.

<Flow cytometry and sorting>
To measure lung, liver, and kidney PD-L1 populations from in vivo senescent cells, young (2 months old) and old (24-26 months old) p16-Tom mice were treated with TAM for 2 weeks. It was later slaughtered. After anesthesia with Avertin (Sigma), the mice were perfused with PBS via the vascular system from the heart. The lungs and kidneys were shredded with a sharp blade and placed in Liberase solution (RPMI (manufactured by Nacalai Tesque) at 50 μg/mL Liberase TM (manufactured by Sigma) for lungs and Liberase TH (manufactured by Sigma) for kidneys. , 2 Kunitz units/mL DNase I (manufactured by Sigma), and 10 mM HEPES (manufactured by Nacalai Tesque) at 37°C for 1 hour. During digestion, the tissue suspension was passed sequentially through 18G and 21G needles to further homogenize the remaining tissue. The liver was shredded with a sharp blade and mixed with 1 mg/mL collagenase IV (Thermo Fisher Scientific), 1 mg/mL pronase (Sigma), and 1 Kunitz unit in collagenase solution (HBSS (+) (Wako)). /mL DNase I (manufactured by Sigma)) at 37°C for 20 minutes with shaking. After digestion, all samples were passed through a 70 μm cell strainer and pelleted by centrifugation at 400 g for 5 min. After RBC lysis (Thermo Fisher) and FcR blocking (Miltenyi Biotec), cells were stained with anti-PD-L1 (Proteintech, catalog: 17952-1-AP) for 1 hour at 4°C and then incubated. , AF647 (manufactured by Thermo Fisher Scientific) conjugated anti-rabbit IgG for 30 minutes at 4°C. Immune cells and dead cells were excluded by anti-CD45 staining (manufactured by BioLegend, clone: 30-F11) and DAPI (0.1 μg/mL) staining, respectively. Fluorescent signals were analyzed using a BD FACSCanto ^™ II (BD Biosciences).

To determine the PD-L1 population in cultured cells, cells were detached by trypsinization and resuspended in wash buffer (4% FBS in PBS), after which the cells were incubated with anti-PD-L1 for HCA2 (Abcam, clone: 28-8) or anti-PD-L1 for MPS (Proteintech, catalog: 17952-1-AP) at 4°C for 1 h, followed by incubation with anti-rabbit IgG conjugated to AF647 (Thermo Fisher Scientific) for 30 min at 4°C. For other immune checkpoints, cells were incubated with anti-CD80 (BioLegend, clone: W17149D), anti-CD86 (BioLegend, clone: BU63), anti-CD276 (BioLegend, clone: MIH42), anti-B7-H4 (BioLegend, clone: MIH43), anti-CD270 (BioLegend, clone: 122), and anti-galectin-9 (BioLegend, clone: 9M1-3). DAPI (0.1 μg/mL) staining was performed to exclude dead cells, and the fluorescent signals were analyzed using a BD FACSCanto ^™ II (BD Biosciences).

For liver infiltration or splenic CD8+ T cell analysis, liver or spleen was harvested from mice and triturated with the flat end of a syringe over a 70 μm cell strainer in 3 mL of PBS. All cells that passed were collected in a tube and pelleted by centrifugation at 400 g for 5 minutes. After RBC lysis (manufactured by Thermo Fisher) and FcR blocking (manufactured by Miltenyi Biotec), cells were treated with anti-CD45 (manufactured by BioLegend, clone: 30-F11), anti-CD44 (manufactured by BioLegend, clone: IM7), and anti-CD3. (manufactured by BioLegend, clone: 145-2C11) and anti-CD8 (manufactured by BioLegend, clone: 53-6.7). Dead cells were removed by propidium iodide staining (2 μg/mL). Fluorescent signals were analyzed using a BD FACSCanto ^™ II (BD Biosciences).

For ProteoStat staining, cells were labeled with anti-PD-L1 (Abcam, clone: 28-8) and AF405-conjugated anti-rabbit IgG (Abcam), and then the cell suspension was added to PBS containing 4% formaldehyde. It was added dropwise and incubated at room temperature for 10 minutes. Cells were washed once with wash buffer, and the cell suspension was added dropwise to permeabilization solution (0.2% Triton X-100, 4% FBS in PBS) and incubated for 5 minutes at room temperature. After washing the cells with a washing buffer, they were stained with ProteoStat staining solution (1/5000-fold dilution) at room temperature for 30 minutes. AF405 and ProteoStat (PE-Texas Red channel) signals were analyzed using BD FACSAria SORP (manufactured by BD Biosciences). All antibodies are listed in Tables 2 and 3.

<Proteasome activity assay>
After sorting PD-L1+ cells and PD-L1- cells, 10,000 cells were seeded into a white 96-well plate. Chymotrypsin, trypsin, and caspase-like proteasome activities were tested using the Proteasome-Glo Cell-Based Assay kit (Promega) according to the manufacturer's instructions. The luminescence intensity was measured using FLUOstar OPTIMA (manufactured by BMG LABTECH).

<Measurement of serum AST, ALT, LDH, and total cholesterol (T-CHO) levels>
Blood was collected from mice by cardiac puncture and left at room temperature for 15 min. To obtain serum, samples were centrifuged at 3400 g for 15 min at 4°C. Serum samples with severe hemolysis were excluded from further analysis of AST, ALT, and LDH (one sample from a mouse receiving normal diet and ABT263 treatment). Serum levels of AST, ALT, LDH, and T-CHO were measured by Fujifilm Wako Pure Chemical Corporation (Japan).

<Grip strength measurement and performance test>
Grip strength was measured using a BIO-GS3 grip strength testing device (manufactured by BIOSEB). I grabbed the mouse by its tail and had it grip a metal grid with its two front paws. As the grid is pulled backwards in a horizontal plane until it is released, the strength of the force displayed on the device is recorded.

To measure coordination, a rotarod model MK-630B (Muromachi Co., Ltd.) was used. The rotation speed was set to 30 rpm, and the time the mouse stayed on the rod from the start of rotation until it fell onto the magnetic switch was recorded. To prevent learning effects, both grip strength and rotarod tests were performed three times on the same day. The average value for each mouse was used for statistical inference.

<Statistics and reproducibility>
In animal experiments, all male mice with the same genotype were randomly assigned to each group and performed independently according to the same age-dependent schedule in each experimental design. The sample size was not predetermined by a pilot study. A blind design was not employed in this study, as automated analysis was performed using an image analyzer and with the same criteria. Data are presented as mean ± SEM unless otherwise noted. Comparisons between two groups were made by an unpaired two-tailed Student's t test. Multiple comparisons of univariate data were performed by one-way analysis of variance (ANOVA) followed by post hoc Tukey's or Dunnett's test. Multiple comparisons of multivariate data were performed by two-way analysis of variance followed by Sidak's multiple comparison test. To maintain consistency in comparisons, significance in all figures is indicated as follows: *P<0.05, **P<0.01, ***P<0.001, *** *P<0.0001. For all representative results, three or more independent experiments were performed with similar results.

[Example 1]
The properties of senescent cells in vivo vary depending on their origin and stimulus, and certain types of senescent cells may accumulate with aging. Immune surveillance against incompetent cells, such as cancer cells, is negatively regulated by immune checkpoints. Therefore, it was thought that this might also apply to senescent cells. To address this question, we investigated the expression of several immune checkpoints in the human fibroblast cell line HCA2, a well-established in vitro model of aging fibroblasts. was analyzed. Interestingly, compared to starvation-induced quiescence (Quie), nutlin-3a-induced cells (n-Sen) (Non-Patent Document 5) and DNA damage-induced senescent cells (d-Sen) show that programmed death ligand 1 ( PD-L1) population was the only one whose expression was significantly and specifically elevated (Figure 1a, Figure 5a).

Expression of programmed cell death protein 1 (PD-1), a receptor for PD-L1, increases with age in mouse T cells (Non-Patent Documents 6 to 9). PD-L1 prevents cytotoxicity caused by attack of PD-1-expressing T cells, and therefore, PD-L1 expression in senescent cells may escape immune surveillance and lead to age-related accumulation of senescent cells. I hypothesized that there is. To examine the expression of PD-L1 in senescent cells in vivo, we used the p16-CreERT2-tdTomato mouse model (p16-Tom) and treated these cells (p16 ^high cells) as tdTomato-expressing cells (Tom ⁺ cells). It was visualized by fluorescence imaging (Non-Patent Document 10). In the lung, liver, and kidney, Tom ⁺ non-immune cells generally show increased expression of senescence-associated secretory phenotype (SASP)-related genes compared to Tom- cells, and Tom ⁺ cells are at least a subset of senescent cells ^. This suggests that (Figure 5b). Some Tom ⁺ cells expressing PD-L1 were detected (Fig. 1b), and the population of Tom ⁺ cells and PD-L1-expressing cells within Tom ⁺ cells was smaller than in the younger group (2 months old). It was significantly higher in the older age group (24-26 months). (Fig. 1c, Fig. 5c). An increase in PD-L1 cell population in Tom- cells of aged mice was also observed in liver and kidney, but not in lung (Fig. 5d). To further characterize the in vivo dynamics of senescent cells, we pulse-labeled these cells with tamoxifen (TAM) at 2.5 months of age and then sacrificed the mice at 3, 6, and 9 months of age (Figure 1d). The proportion of PD-L1 ⁺ cells among Tom ⁺ cells in the lungs, liver, and kidneys was higher in the 9-month-old group than in the 3-month-old and 6-month-old groups; A tendency for ⁺ cells to accumulate was suggested.

To clarify the regulatory mechanism of PD-L1 expression in senescent cells, we used an in vitro aging model. PD-L1 levels were high in proliferating cells and low in quiescent and senescent cells (Fig. 6a). PD-L1 transcription was cell cycle dependent and was highest in S phase (Fig. 6b). p130, a dominant pocket protein that forms the DREAM complex and suppresses E2F activity in quiescent cells and senescent cells (Non-Patent Documents 11, 12), or the dominant-negative mutant E2F (E2F97) (Non-Patent Document 13). Both suppressed PD-L1 expression when overexpressed in proliferating cells (Figures 6c and 6d). Consistent with this, overexpression of the human papillomavirus type 16E7 protein, which inhibits the pocket protein family (14), restored PD-L1 expression regardless of the time elapsed since senescence induction (Fig. 6e). Therefore, the very low activity of E2F may suppress PD-L1 expression in quiescent and senescent cells.

In cultured cells, only 6-10% of senescent cells expressed PD-L1, whereas its heterogeneous expression was not observed in quiescent cells (Fig. 1a). PD-L1 ⁻ and PD-L1 ⁺ d-Sen HCA2 cells were sorted for bulk RNA-seq analysis (Fig. 6f). The present inventors constructed a transcriptional regulatory network of 230 PD-L1 positively correlated genes (R ² >0.95) using the transcription factor binding motif enrichment tool iRegulon (Non-Patent Documents 15, 16). (Figure 6g). Among the predicted master regulator candidates, only E2F1-binding motifs are enriched in the promoter region of PD-L1, and weak E2F1 activity remains in PD-L1 ⁺ senescent cells and slightly induces PD-L1 transcription. It was suggested that there is a possibility that This weak E2F1 activity may not be sufficient to trigger the G1/S transition of senescent cells. However, this alone cannot explain the significantly higher level of PD-L1 protein expression in PD-L1 ⁺ cells (Fig. 6h). In this context, large amounts of protein aggregates were also detected in PD-L1 ⁺ senescent cells (Fig. 1e). Considering that proteasomal degradation plays an important role in regulating PD-L1 stability in cancer cells (Non-Patent Documents 17, 18), the heterogeneity of PD-L1 expression in senescent cells may be due to individual This may be due to transcription-independent mechanisms such as proteasome activity in senescent cells. This idea was supported by the fact that levels of protein aggregates differ between senescent cells. In PD-L1 ⁺ senescent cells, chymotrypsin-like protease activity, trypsin-like protease activity, and caspase-like protease activity (Non-Patent Document 19) were lower than in PD-L1 ⁻ cells (Fig. 1f). Treatment of senescent cells with proteasome inhibitors such as epoxomicin and bortezomib increased the expression of PD-L1 and PD-L1 ⁺ senescent cell populations (Figures 1g-1h). To enhance proteasome activity in senescent cells, we investigated the effect of expression of a constitutively active NRF1 mutant (NRFΔ1-103) (Non-Patent Documents 20, 21) on PD-L1 expression. Examined. Overexpression of NRFΔ1-103 significantly reduced the number of PD-L1 ⁺ senescent cells (Fig. 1i). These results suggest that the heterogeneous accumulation of PD-L1 in senescent cells is due, at least in part, to reduced E2F1-mediated transcription and proteasome activity.

[Example 2]
Next, we investigated whether PD-L1 functions as an immune checkpoint for T cell immunity during cellular senescence. Mouse activated primary CD8 ⁺ T cells were co-cultured as effector cells and primary mouse lung fibroblasts (MPF) as target cells, and a syngeneic in vitro T cell killing assay was performed to analyze cytotoxicity. Senescence induction was quantified by SA-β-gal staining and p16 expression in n-Sen and d-Sen MPFs (Figures 7a and 7b). In MPF, heterogeneous expression of PD-L1 was observed in senescent cells but not in quiescent cells (Fig. 2a). After co-culture, senescent cells were more sensitive to T cells than quiescent cells in all induction methods (Fig. 2b, Fig. 7c). SASP is a typical phenotype of cellular aging (Non-Patent Document 22), and is also known to be involved in immune surveillance of senescent cells by attracting immune cells to damaged tissues (Non-Patent Document 23). ). Differences in transcriptional signatures between n-Sen MPF and Quie MPF were investigated by bulk RNA-seq (Fig. 2c).

Gene ontology (GO) and gene set enrichment analysis (GSEA) revealed that terms in n-Sen MPF related to inflammation, cytokine production, T cell activation, and antigen presentation are highly enriched. (Fig. 2d, Fig. 7d, Fig. 7e). Although resting cells did not exhibit T cell cytotoxicity, incubation with conditioned medium from senescent cells resulted in a small but significant increase in cytotoxicity (Fig. 2e). Importantly, incubation with conditioned medium from senescent cells expressing non-degradable IκBα (S32/36A) did not increase cytotoxicity of T cells, and enhancement of cytotoxicity by senescent cell conditioned medium was due to NF -κB dependence was suggested (Fig. 1, Fig. 2e, Fig. 7f to Fig. 7h).

Recently, it has been shown that transposable elements (TEs) and TE-encoded endogenous retroviral elements (ERVs) are presented in MHC class I as neoantigens and play an important role in tumor-specific immunogenicity. (Non-patent Documents 24, 25). RNA-seq analysis revealed that 44 TEs and 20 ERVs were transcriptionally activated in senescent cells (Fig. 2f). Among them, some ERVs were found to be upregulated in Tom+ cells sorted from p16-Tom mice compared to Tom- cells (Fig. 2g). Furthermore, type I interferon signaling, a positive regulator of the MHC class I antigen processing machinery (Non-Patent Document 26), was enhanced in senescent cells (Fig. 2d). Therefore, senescent cells generally have the potential to activate cytotoxic T cell immunity through secretion of proinflammatory cytokines and presentation of endogenous antigens.

Furthermore, activated T cells significantly induced cytotoxicity against PD-L1 ⁻ senescent cells, but not against PD-L1 ⁺ senescent cells (Fig. 2h, Fig. 8a). To determine whether this cytotoxicity is dependent on antigen presentation, we repeated the experiment using T cells from OT-1 mice and found that ovalbumin (OVA) peptide treatment resulted in dramatic cytotoxicity toward senescent cells. (Figure 8b). This suggests that the cytotoxicity of CD8 ⁺ T cells toward senescent cells is dependent on antigen presentation. These increases for PD-L1 ⁺ senescent cells were of a significantly smaller extent than for PD-L1 ⁻ cells (Fig. 8b). Consistent with this, overexpression of PD-L1 significantly suppressed cytotoxicity in senescent cells, which was inhibited by incubation with a neutralizing anti-PD-1 antibody (αPD-1). It was recovered by blocking the body (Fig. 2i, Fig. 8c, Fig. 8d). Similar results were obtained with OVA peptide-activated T cells from OT-1 mice (Fig. 8e). These results suggest that PD-L1 expression in senescent cells is required for escape from T cell immunity.

[Example 3]
To determine the impact of PD-L1 ⁺ senescent cells on in vivo physiological phenotypes in specific cell types, we examined a single-cell RNA-sequencing (scRNA-seq) dataset provided by p16-Tom mice ( Non-Patent Document 10). For each dataset, we included (1) liver sinusoidal endothelial cells (LSEC) from normal liver and nonalcoholic steatohepatitis (NASH) liver, (2) NASH-associated macrophages (NAM) from NASH liver, and ), and (3) a rich set of cell types were enrolled, including vascular cells from normal kidney (Fig. 3a, Fig. 9a). Tom ⁺ and Tom ⁻ cells were differentiated as PD-L1 ⁺ and PD-L1 ⁻ based on Cd274 transcripts (Fig. 3b, Fig. 9b). Differentially expressed genes (DEGs) in PD-L1 ⁺ and PD-L1 ^- LSECs were identified separately in Tom ⁺ and Tom ^- LSECs (Fig. 3c). Compared to the number of DEGs in Tom ⁻ LSEC (1 up DEG, 2 down DEGs), a much larger number of DEGs (71 up DEGs, 9 down DEGs) were detected in Tom ⁺ LSEC. Ta. LSECs have been shown to be moderately inflammatory in aged mice, with upregulation of p16 ^Ink4a (27, 28). GO analysis showed that DEGs (71 genes) upregulated from PD-L1 ⁺ /Tom ⁺ LSECs were enriched in cytokine responses and inflammatory signaling (Fig. 3d, Fig. 3e). Furthermore, GSEA results revealed that interferon, TNF-α, and inflammatory signaling were more activated in PD-L1 ⁺ /Tom ⁺ LSECs than in PD-L1 ⁻ /Tom ⁺ LSECs. (Fig. 3f). Inflammatory signatures were also abundant in PD-L1 ⁺ /Tom ⁺ NASH LSECs, NAMs, and renal vascular cells (Figures 5c-5h). These results suggested that aging-induced phenotypes, such as pro-inflammation, were positively correlated with the presence and selective accumulation of PD-L1 ⁺ senescent cell populations.

[Example 4]
Immune checkpoint blockade, such as treatment with monoclonal antibodies against PD-1 or PD-L1, has contributed to significant advances in cancer immunotherapy over the past decade, redirecting T cells to PD-L1-expressing cancer cells. has been successfully used to treat various malignant tumors (Non-Patent Document 29). To assess the effect of anti-PD-1 (αPD-1) antibodies on senescent cell dynamics in vivo, we inoculated 7-month-old p16-Tom mice that had been labeled with TAM for senescent cells 2 weeks earlier. αPD-1 antibody or isotype control (IgG) was administered for 3 weeks (Figure 4a). Furthermore, to verify whether the effect of αPD-1 antibody was due to CD8 ⁺ T cells, cytotoxic T cells were systemically reduced by anti-CD8 (αCD8) antibody treatment throughout the experiment ( Figure 10a). The total number of Tom ⁺ cells in the lungs, liver and kidneys of mice injected with αPD-1 antibody was significantly lower than in the IgG group (Fig. 4a, Fig. 10b). Furthermore, αPD-1 antibody treatment decreased the PD-L1 ⁺ subpopulation in Tom ⁺ cells, but not in Tom ⁻ cells (Fig. 10b). Furthermore, we noticed that αPD-1 treatment reduced the population of PD-L1 ⁻ /Tom ⁺ cells to a lesser extent than PD-L1 ⁺ /Tom ⁺ cells. This indicates that the αPD-1 antibody preferentially targets PD-L1 ⁺ /Tom ⁺ cells (Figure 10c). However, these reducing effects were mainly reversed by the reduction of CD8 ⁺ T cells, whereas αPD-1 tended to slightly reduce Tom ⁺ cells in the lungs, which was not statistically significant (Fig. 4a) ). These results indicated that immunity by cytotoxic T cells is indeed involved in immune surveillance of cellular aging. We distinguished F4/80-expressing Tom ⁺ cells from other Tom ⁺ cells in the livers of p16-Tom mice as described above, as a subset of macrophages also express p16 and possess some hallmarks of cellular senescence ( Figure 11a). The majority of Tom ⁺ cells in the liver were F4/80 negative (Fig. 11b), and the proportion of Tom ⁺ cells in both F4/80 positive and negative cells was lower under αPD-1 antibody treatment (Fig. 11b, Figure 11c). Furthermore, αPD-1 antibody treatment reduced the PD-L1 ⁺ subpopulation in F4/80 negative Tom ⁺ cells (Fig. 11d).

αPD-1 antibody-mediated senescent cell ablation also ameliorates the senescent phenotype, as targeting senescent cell ablation in vivo has been reported to ameliorate various aging and age-related disorders. I tested whether. The present inventors have discovered that age-related increases in alveolar volume (Non-Patent Document 31), hepatic lipidosis (Non-Patent Document 32), decrease in grip strength (Non-Patent Document 33), and decrease in exercise capacity (Non-Patent Document 33) (34) found that injecting old (17.5 months old) wild-type mice with αPD-1 antibody significantly improved the results compared to the IgG-treated group. (Fig. 4c, Fig. 4d, Fig. 11e, Fig. 11f). No such effects were observed in young (3 month old) mice.

Cellular senescence plays an important role in the pathogenesis of NASH (35). NASH is induced by feeding mice a choline-deficient L-amino acid defined high-fat diet (CDA-HFD) (Non-Patent Document 36). 1.5-month-old mice were fed normal diet or CDA-HFD for 7 weeks and then injected with αPD-1 antibody or IgG for 3 weeks (Figure 4e). The present inventors further found that the elimination of PD-L1 ⁺ senescent cells by immune checkpoint blockade (ICB) therapy is more effective than the senolytic drug ABT263 in the pathogenesis of NASH (37). We investigated whether In the livers of mice fed a normal diet, CD8 ⁺ T cells were moderately activated by αPD-1 antibody administration, whereas in NASH conditions they were strongly activated independent of αPD-1 antibody administration. It was found that T cell infiltration into the liver was enhanced by αPD-1 antibody administration only in the NASH state (Fig. 12a). Mice fed CDA-HFD had higher hepatic hypolipidemia, fibrosis, and serum levels of AST, ALT, and LDH than mice fed a normal diet, demonstrating that CDA-HFD successfully induces NASH pathogenesis. (Fig. 4e, Fig. 12b, Fig. 12c). Both αPD-1 antibody and ABT263 ameliorated the pathogenesis, but the αPD-1 group showed more improvement in hepatic steatosis than the ABT263 group (Fig. 4e). Moreover, body weight and serum total cholesterol (T-CHO) levels showed similar trends (Fig. 12d), suggesting that PD-L1 ⁺ senescent cell clearance increases LDL secretion from the liver and reduces lipid accumulation in the progression of NASH. suggested that it was possible.

Because ICB administration depleted a portion of the Tom ⁺ macrophages mentioned above, we administered PLX3397 (a CSF1R inhibitor) to mice to further clarify whether the amelioration of the disease state by ICB administration was mainly due to macrophage depletion. administered. The results showed that administration of PLX3397 did not significantly affect alveolar size, grip strength, or exercise capacity in aged mice (Figures 13a-13d). It has been reported that macrophage activation is involved in the progression of NASH (Non-Patent Document 38). Indeed, CSF1R inhibition was found to improve several indicators of NASH pathology, including serum AST, ALT, LDH, hepatic lipidosis, and fibrosis (Figures 13e-13h). Importantly, co-treatment of αPD-1 antibody and PLX3397 further improved NASH pathology except liver fibrosis. These results suggest that the effect of PD-1 on NASH progression is not due to removal of the inflammatory macrophage population. To confirm that the effect of αPD-1 antibody was due to the reduction of senescent cells, senescent cells of p16-Tom mice were labeled by TAM administration after NASH induction. NASH-induced Tom ⁺ cells were significantly reduced by αPD-1 treatment (Fig. 4f). These results suggest that aging and NASH-dependent accumulation of senescent cells involve the PD-L1/PD-1 axis to escape from cell-mediated immunity. Therefore, administration of ICB may attenuate inhibitory signals to T cells, reactivate immune surveillance, eliminate senescent cells, and ameliorate various age-related phenotypes.

In conclusion, our results suggest that CD8 ⁺ T cells are involved in the immune surveillance of senescent cells and that the PD-L1/PD-1 immune checkpoint is at least partially involved in the age-dependent accumulation of senescent cells. suggests that it is. Recently, clonal expansion of CD8 ⁺ T cells and cytotoxic CD4 ⁺ T cells with specific TCRs has been discovered in patients with age-related neurodegeneration and centenarians (Non-Patent Documents 39, 40). Furthermore, the clonality of CD8 ⁺ memory T cells is expanded by the induction of age-related diseases such as diabetes and liver fibrosis (Non-Patent Documents 41, 42). These lines of evidence suggest that antigen presentation at MHC-I and activation of the adaptive immune system may commonly occur during the progression of age-related diseases.

Our previous single-cell analysis suggested that senescent cells are highly heterogeneous depending on the cell type and the cause of senescence induction in vivo. Therefore, senescent cells may exhibit heterogeneity even among cells of the same type. In this study, PD-L1 ⁺ senescent cells showed much more inflammatory transcriptome than PD- ^L1− cells, even within the same cell type (LSEC, NAM, or renal vascular cells). Therefore, the accumulation of PD-L1 ⁺ senescent cells may have more deleterious consequences than a simple increase in the number of senescent cells with age. Therefore, selective inhibition of PD-L1 ⁺ senescence accumulation by ICB treatment is a more promising senescence treatment strategy than conventional senescence treatments. In this regard, it has been reported that in Alzheimer's disease models, blocking the PD-L1/PD-1 axis alleviates the disease state and improves cognitive impairment (Non-Patent Documents 43, 44). It has been reported that relatively long-term (8 weeks) administration of αPD-1 antibody to a NASH model with 10 months of CDA-HFD induction increases the burden of hepatocellular carcinoma (HCC) (Non-patent literature). 45). Furthermore, long-term αPD-1 antibody administration often induced systemic autoimmune diseases (Non-Patent Document 46). However, these were not observed with our short-term treatment (3 weeks of αPD-1 antibody treatment followed by 10 weeks of CDA-HFD induction). Therefore, controlling immunotherapy in age-related diseases requires not only reducing the dose and frequency, but also considering the balance between enhanced immune clearance, resistance to acute inflammation, and tissue repair rate. It may be necessary to optimize the use of each treatment with consideration to reducing

Claims (7)

A pharmaceutical composition for the treatment or prevention of non-alcoholic steatohepatitis (NASH), which contains as an active ingredient a substance that inhibits the binding of PD-L1 on cell membranes to PD-1. The pharmaceutical composition according to claim 1, wherein the substance is one selected from the group consisting of anti-PD-L1 antibodies and anti-PD-1 antibodies. The pharmaceutical composition according to claim 1, wherein the substance is a low molecular compound that specifically binds to PD-L1. The pharmaceutical composition according to claim 1, wherein the PD-L1 is a cell membrane protein of liver cells, and the PD-1 is a cell membrane protein of CD8-positive T cells. The pharmaceutical composition according to claim 1, which is administered only once to an animal that has developed NASH. The pharmaceutical composition according to claim 5, wherein the animal that has developed NASH has not developed any cancer other than liver cancer. An agent for removing p16-positive senescent cells, which contains as an active ingredient one or more selected from the group consisting of anti-PD-L1 antibodies and anti-PD-1 antibodies.