Senescence in premalignant tumours

Abstract

Cellular senescence, a growth-arrest program that limits the lifespan of mammalian cells and prevents unlimited cell proliferation, is attracting considerable interest because of its links to tumour suppression. Using a mouse model in which the oncogene Ras is activated in the haematopoietic compartment of bone marrow, Braig et al. show that cellular senescence can block lymphoma development. Genetic inactivation of the histone methyltransferase Suv39h1 that controls senescence by ‘epigenetic’ modification of DNA-associated proteins, or a pharmacological approach that mimics loss of this enzyme, allow the formation of malignant lymphomas in response to oncogenic Ras. This work has important implications for both tumour development and tumour therapy. Michaloglou et al. report that oncogene-induced senescence may be a physiologically important process in humans, keeping moles in a benign state for many years: unchecked they develop into malignant melanomas. Chen et al. also find that cellular senescence blocks tumorigenesis in vivo: they show that acting together, the p53 tumour suppressor and the cellular senescence system can prevent prostate cancer induction in mice by the PTEN mutation. Collado et al. show that cellular senescence is a defining feature of Ras-initiated premalignant tumours; this could prove valuable in the diagnosis and prognosis of cancer. See the web focus . Oncogene-induced senescence is a cellular response that may be crucial for protection against cancer development1,2, but its investigation has so far been restricted to cultured cells that have been manipulated to overexpress an oncogene. Here we analyse tumours initiated by an endogenous oncogene, ras, and show that senescent cells exist in premalignant tumours but not in malignant ones. Senescence is therefore a defining feature of premalignant tumours that could prove valuable in the diagnosis and prognosis of cancer.