Cellular senescence is characterized by stable cell cycle arrest and a secretory program that modulates the tissue microenvironment 1 , 2 . Physiologically, senescence serves as a tumor suppressive mechanism that prevents the expansion of premalignant cells 3 , 4 and plays a beneficial role in wound healing responses 5 , 6 . Pathologically, the aberrant accumulation of senescent cells generates an inflammatory milieu that leads to chronic tissue damage and contributes to diseases such as liver and lung fibrosis, atherosclerosis, diabetes, and osteoarthritis 1 , 7 . Accordingly, elimination of senescent cells from damaged tissues in mice ameliorates symptoms of these pathologies and even promotes longevity 1 , 2 , 8 – 10 . Here we test the therapeutic concept that chimeric antigen receptor (CAR) T cells targeting senescent cells can be effective senolytics. We identify the urokinase plasminogen activator receptor (uPAR) 11 as a cell surface protein broadly induced during senescence and demonstrate that uPAR-specific CAR T cells efficiently ablate senescent cells in vitro and in vivo . uPAR-directed CAR T cells extend the survival of mice harboring lung adenocarcinoma treated with a senescence-inducing drug combination, and restore tissue homeostasis in chemical- or diet-induced liver fibrosis. These results establish the therapeutic potential of senolytic CAR T cells for senescence-associated diseases.

We used bioluminescence imaging to reveal patterns of metastasis formation by human breast cancer cells in immunodeficient mice. Individual cells from a population established in culture from the pleural effusion of a breast cancer patient showed distinct patterns of organ-specific metastasis. Single-cell progenies derived from this population exhibited markedly different abilities to metastasize to the bone, lung, or adrenal medulla, which suggests that metastases to different organs have different requirements. Transcriptomic profiling revealed that these different single-cell progenies similarly express a previously described "poor-prognosis" gene expression signature. Unsupervised classification using the transcriptomic data set supported the hypothesis that organ-specific metastasis by breast cancer cells is controlled by metastasis-specific genes that are separate from a general poor-prognosis gene expression signature. Furthermore, by using a gene expression signature associated with the ability of these cells to metastasize to bone, we were able to distinguish primary breast carcinomas that preferentially metastasized to bone from those that preferentially metastasized elsewhere. These results suggest that the bone-specific metastatic phenotypes and gene expression signature identified in a mouse model may be clinically relevant.

We used bioluminescence imaging to reveal patterns of metastasis formation by human breast cancer cells in immunodeficient mice. Individual cells from a population established in culture from the pleural effusion of a breast cancer patient showed distinct patterns of organ-specific metastasis. Single-cell progenies derived from this population exhibited markedly different abilities to metastasize to the bone, lung, or adrenal medulla, which suggests that metastases to different organs have different requirements. Transcriptomic profiling revealed that these different single-cell progenies similarly express a previously described "poor-prognosis" gene expression signature. Unsupervised classification using the transcriptomic data set supported the hypothesis that organ-specific metastasis by breast cancer cells is controlled by metastasis-specific genes that are separate from a general poor-prognosis gene expression signature. Furthermore, by using a gene expression signature associated with the ability of these cells to metastasize to bone, we were able to distinguish primary breast carcinomas that preferentially metastasized to bone from those that preferentially metastasized elsewhere. These results suggest that the bone-specific metastatic phenotypes and gene expression signature identified in a mouse model may be clinically relevant.

To reject tumors, T cells must overcome poor tumor immunogenicity and an adverse tumor microenvironment. Providing agonistic costimulatory signals to tumor-infiltrating T cells to augment T cell function remains a challenge for the implementation of safe and effective immunotherapy. We hypothesized that T cells overexpressing selected costimulatory ligands could serve as cellular vehicles mediating powerful, yet constrained, anatomically targeted costimulation. Here, we show that primary human T cells expressing CD80 and 4-1BB ligand (4-1BBL) vigorously respond to tumor cells lacking costimulatory ligands and provoke potent rejection of large, systemic tumors in immunodeficient mice. In addition to showing costimulation of bystander T cells (transcostimulation), we show the effect of CD80 and 4-1BBL binding to their respective receptors in the immunological synapse of isolated single cells (autocostimulation). This new strategy of endowing T cells with constitutively expressed costimulatory ligands could be extended to other ligand-receptor pairs and used to enhance any targeted adoptive transfer therapy.

Two genetic reporter systems were developed for multimodality reporter gene imaging of different molecular-genetic processes using fluorescence, bioluminescence (BLI), and nuclear imaging techniques. The eGFP cDNA was fused at the N-terminus with HSV1-tk cDNA bearing a nuclear export signal from MAPKK (NES-HSV1-tk) or with truncation at the N-terminus of the first 45 amino acids (Delta45HSV1-tk) and with firefly luciferase at the C-terminus. A single fusion protein with three functional subunits is formed following transcription and translation from a single open reading frame. The NES-TGL (NES-TGL) or Delta45HSV1-tk/GFP/luciferase (Delta45-TGL) triple-fusion gene cDNAs were cloned into a MoMLV-based retrovirus, which was used for transduction of U87 human glioma cells. The integrity, fluorescence, bioluminescence, and enzymatic activity of the TGL reporter proteins were assessed in vitro. The predicted molecular weight of the fusion proteins (~130 kDa) was confirmed by western blot. The U87-NES-TGL and U87-Delta45-TGL cells had cytoplasmic green fluorescence. The in vitro BLI was 7- and 13-fold higher in U87-NES-TGL and U87-Delta45-TGL cells compared to nontransduced control cells. The Ki of (14)C-FIAU was 0.49+/-0.02, 0.51+/-0.03, and 0.003+/-0.001 ml/min/g in U87-NES-TGL, U87-Delta45-TGL, and wild-type U87 cells, respectively. Multimodality in vivo imaging studies were performed in nu/ nu mice bearing multiple s.c. xenografts established from U87-NES-TGL, U87-Delta45-TGL, and wild-type U87 cells. BLI was performed after administration of d-luciferin (150 mg/kg i.v.). Gamma camera or PET imaging was conducted at 2 h after i.v. administration of [(131)I]FIAU (7.4 MBq/animal) or [(124)I]FIAU (7.4 MBq/animal), respectively. Whole-body fluorescence imaging was performed in parallel with the BLI and radiotracer imaging studies. In vivo BLI and gamma camera imaging showed specific localization of luminescence and radioactivity to the TGL transduced xenografts with background levels of activity in the wild-type xenografts. Tissue sampling yielded values of 0.47%+/-0.08%, 0.86%+/-0.06%, and 0.03%+/-0.01%dose/g [(131)I]FIAU in U87-NES-TGL, U87-Delta45-TGL, and U87 xenografts, respectively. The TGL triple-fusion reporter gene preserves the functional activity of its subunits and is very effective for multimodality imaging. It provides for the seamless transition from fluorescence microscopy and FACS to whole-body bioluminescence imaging, to nuclear (PET, SPET, gamma camera) imaging, and back to in situ fluorescence image analysis.