ORIGINAL ARTICLE E6AP is required for replicative and oncogene-induced senescence in mouse embryo fibroblasts Y Levav-Cohen 1 , K Wolyniec 2 , O Alsheich-Bartok 1 , A-L Chan 2 , SJ Woods 2 , Y-H Jiang 3 , S Haupt 2 and Y Haupt 2 1 Lautenberg Center for General and Tumor Immunology, The Hebrew University Hadassah Medical School, Jerusalem, Israel; 2 Research Division, The Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia and 3 Division of Medical Genetics, Department of Pediatrics and Neurobiology, Duke University School of Medicine, Durham, NC, USA Cellular senescence is important for the maintenance of tissue homeostasis, and has recently been shown to pose a natural barrier to tumorigenesis. The E3 ubiquitin ligase, E6AP, has been linked to a number of protein regulators of the cell cycle as well as the cellular stress response. We therefore explored the role of E6AP in the cellular response to stress. We found that mouse embryo fibro- blasts (MEFs) lacking E6AP escape replicative senescence, as well as Ras-induced senescence associated with impaired markers. E6AP-deficient MEFs exhibit a range of trans- formed phenotypes: these include the ability to grow under stress conditions (such as low serum and DNA damage), enhanced proliferation, anchorage independent growth and enhanced growth of xenografts in mice. The transformed phenotype of E6AP-deficient MEFs is asso- ciated with lower basal and stress-induced accumulation of p53. Overall, our study implicates E6AP as an important regulator of the cellular response to stress, in particular through the regulation of replicative and oncogene- induced senescence. Oncogene (2012) 31, 2199–2209; doi:10.1038/onc.2011.402; published online 19 September 2011 Keywords: E6AP; cellular senescence; stress; p53; Ras Introduction Cellular senescence, the state of a stable long-term loss of replicative capacity, restricts cells from propagation (Kuilman et al., 2010). Senescence is ultimately driven by DNA disruption. One form of senescence, replicative senescence, occurs in cultured human cells that reach an intolerable telomeric shortening or uncapping (Dimri, 2005). The second, accelerated or premature senescence, is independent of telomeres and occurs in mouse embry- onic fibroblasts (MEFs) that endure culture-stress, and also arises in human and mouse cells following geno- toxic stress or oncogene activation (Campisi and d’Adda di Fagagna, 2007). Senescence involves a program of signal transduction that culminates in terminal, irrever- sible growth cessation, accompanied by a distinct set of alterations in the cellular phenotype, such as an enlarged and flattened cell shape (Campisi and d’Adda di Fagagna, 2007). Regardless of the signals that induce premature senescence, the two major tumor suppressor pathways p53-p21 WAF1 and pRb-p16 INK4a lie at the heart of the machinery controlling the execution and maintenance of this response (Campisi and d’Adda di Fagagna, 2007; Collado and Serrano, 2010; and Kuilman et al., 2010). In murine cells the ARF-p53 pathway has a dominant role, whereas in human cells pRb-p16 INK4a seems to be equally important (Campisi and d’Adda di Fagagna, 2007). Recent studies support the role for cellular senescence as a natural barrier to tumor development (Collado and Serrano, 2010). This concept has been demonstrated by exploiting p53-driven senescence to limit cancer progression in an existing malignancy using a mouse sarcoma model (Ventura et al., 2007). P53 has been demonstrated to be essential for cellular senescence in cultured cells and in vivo mouse models. Signals that induce DNA damage response, such as ionizing irradiation (IR) and chemotherapeutic drugs, or telomere dysfunction, drive senescence through the p53-p21 pathway (Zuckerman et al., 2009). Impairment of DNA repair genes, such as BRCAI or DNA ligase IV, allows DNA to accumulate damage, which induces premature senescence in MEFs; this can be evaded by p53 inactivation (Frank et al., 2000). Irreparable DNA lesions sustain the ATM/ATR-p53 response, maintain- ing the senescence phenotype (Campisi and d’Adda di Fagagna, 2007; Di Micco et al., 2008). Likewise, oncogenic activation triggers senescence, which may involve a DNA damage response, as in the case of Ras or Mos, or be independent of DNA damage as with Runx1 (Wolyniec et al., 2009). In both cases the response involves p53. Oncogenic activation of p53 is mediated by ARF, encoded by the INK4a locus (Gil and Peters, 2006). The mechanism by which p53 promotes senescence is only partially understood. P53 target genes have been implicated in the induction of senescence, these include the CDK inhibitor p21, the plasminogen activator inhibitor-1 (PAI-1) (Mu and Received 6 February 2011; revised 5 August 2011; accepted 6 August 2011; published online 19 September 2011 Correspondence: Professor Y Haupt, Research Division, The Peter MacCallum Cancer Centre, St Andrew’s Place, East Melbourne, Melbourne, Victoria 3002, Australia. E-mail: ygal.haupt@petermac.org Oncogene (2012) 31, 2199–2209 & 2012 Macmillan Publishers Limited All rights reserved 0950-9232/12 www.nature.com/onc