Cellular senescence in cancer: from mechanisms to detection

Abstract

Senescence refers to a cellular state featuring a stable cell-cycle arrest triggered in response to stress. This response also involves other distinct morphological and intracellular changes including alterations in gene expression and epigenetic modifications, elevated macromolecular damage, metabolism deregulation and a complex pro-inflammatory secretory phenotype. The initial demonstration of oncogene-induced senescence in vitro established senescence as an important tumour-suppressive mechanism, in addition to apoptosis. Senescence not only halts the proliferation of premalignant cells but also facilitates the clearance of affected cells through immunosurveillance. Failure to clear senescent cells owing to deficient immunosurveillance may, however, lead to a state of chronic inflammation that nurtures a pro-tumorigenic microenvironment favouring cancer initiation, migration and metastasis. In addition, senescence is a response to post-therapy genotoxic stress. Therefore, tracking the emergence of senescent cells becomes pivotal to detect potential pro-tumorigenic events. Current protocols for the in vivo detection of senescence require the analysis of fixed or deep-frozen tissues, despite a significant clinical need for real-time bioimaging methods. Accuracy and efficiency of senescence detection are further hampered by a lack of universal and more specific senescence biomarkers. Recently, in an attempt to overcome these hurdles, an assortment of detection tools has been developed. These strategies all have significant potential for clinical utilisation and include flow cytometry combined with histo- or cytochemical approaches, nanoparticle-based targeted delivery of imaging contrast agents, OFF-ON fluorescent senoprobes, positron emission tomography senoprobes and analysis of circulating SASP factors, extracellular vesicles and cell-free nucleic acids isolated from plasma. Here, we highlight the occurrence of senescence in neoplasia and advanced tumours, assess the impact of senescence on tumorigenesis and discuss how the ongoing development of senescence detection tools might improve early detection of multiple cancers and response to therapy in the near future.

Keywords: cancer; cellular senescence; detection; senoprobes; tumour microenvironment.

Fig. 1

Hallmarks of cellular senescence. Senescence is triggered in response to a variety of stimuli, with senescent cells acquiring phenotypes derived from changes in morphology, the nucleus and the cytoplasm. B2M, β₂ microglobulin; BCL‐2, B‐cell lymphoma 2; CCF, cytoplasmic chromatin fragment; DPP4, dipeptidyl‐peptidase 4; MMPs, matrix metalloproteinases; ROS, reactive oxygen species; SAβG, senescence‐associated β‐galactosidase; SAHF, senescence‐associated heterochromatin foci; SASP, senescence‐associated secretory phenotype.

Fig. 2

Signalling pathways of senescence induction in cancer. DNA damage and telomere shortening activate a DNA damage response that imposes cell‐cycle arrest through the p53‐p21 axis while ARF and p16 upregulation due to ageing and CDKN2A de‐repression block cell‐cycle progression via both the p53‐p21 and p16 axis. ROS and metabolic alterations implement senescence through MAPK/p38 signalling whereas SASP reinforces senescence by means of TGFβ signalling. Inactivation of tumour suppressors not only induces the Ras/Raf/MEK signalling pathway as oncogenic signals, but also modulates the p53‐p21 axis via the PI3K/AKT/mTOR pathway. In addition to the conventional CKI‐dependent pathway, oncogenic signals trigger cell‐cycle withdrawal by downregulating ribosome biogenesis, thereby increasing RPS14 for direct inhibition of CDK/cyclin‐mediated RB phosphorylation. ATM, ataxia‐telangiectasia mutated; ATR, ATM‐ and Rad3‐related; CDK, cyclin‐dependent kinase; CHK, checkpoint kinase; MAPK, mitogen‐activated protein kinase; MEK, MAPK/ERK kinase; mTOR, mammalian target of rapamycin; PI3K, phosphatidylinositol 3‐kinase; RB, retinoblastoma protein; RPS14, ribosomal protein S14; TGFβ, transforming growth factor β.

Fig. 3

Dual role of senescence in tumorigenesis. Senescence triggered by OIS or TIS initially halts proliferation of premalignant cells and elicits immunosurveillance of senescent cells via SASP secretion, which in turn mediates clearance of premalignant cells, conferring tumour suppression. In contrast, failure to clear senescent cells leads to chronic inflammation by SASP, which cultivates a pro‐tumorigenic microenvironment that promotes proliferation, EMT and stemness of premalignant/malignant cells. Senescence reversion or escape may result in the re‐emergence of malignant cells that may have higher aggressiveness. SASP also contributes to paracrine senescence and induces chemotaxis of malignant cells, resulting in tumour migration, immune evasion and metastasis in distant organs. EMT, epithelial–mesenchymal transition; OIS, oncogene‐induced senescence; SASP, senescence‐associated secretory phenotype; TIS, therapy‐induced senescence.

Fig. 4

Novel approaches for in vivo senescence detection. In addition to conventional detection methods relying on IHC detection of multiple senescence biomarkers in deep‐frozen or fixed tissues, recent development of approaches combining histochemical, cytochemical and flow cytometry offer higher efficiency for senescence detection in fresh tissues. The fine tuning of nanoparticles for recognising senescent cells strengthens further the targeted delivery of cargoes, that is, image contrasting agents, into senescent tumour cells. Avoiding potential cytotoxicity, OFF‐ON Senoprobes facilitate the real‐time detection and tracking of living senescent cells with elevated SAβG activity. In the human setting, the senescent‐specific PET probe FPyGal may be used to assess senescence burden within tumours pre‐ and post‐treatment, which would provide valuable information in the design of therapeutic strategies and inpatient response. The emerging field of cell‐free DNA (cfDNA) analysis in liquid biopsy provides the least invasive senescence detection tool that is also usable in large‐scale and longitudinal patient screening and monitoring. B2M, β₂ microglobulin; nanoMIP, molecular imprinted nanoparticle; NP, nanoparticle; SAβG, senescence‐associated β‐galactosidase.