Research Article: Cooperation between p21 and Akt is required for p53‐dependent cellular senescence

Date Published: July 09, 2017

Publisher: John Wiley and Sons Inc.

Author(s): Young Yeon Kim, Hye Jin Jee, Jee‐Hyun Um, Young Mi Kim, Sun Sik Bae, Jeanho Yun.


Cellular senescence has been implicated in normal aging, tissue homeostasis, and tumor suppression. Although p53 has been shown to be a central mediator of cellular senescence, the signaling pathway by which it induces senescence remains incompletely understood. In this study, we have shown that both Akt and p21 are required to induce cellular senescence in response to p53 expression. In a p53‐induced senescence model, we found that Akt activation was essential for inducing a cellular senescence phenotype. Surprisingly, Akt inhibition did not abolish p53‐induced cell cycle arrest, but it suppressed the increase in intracellular reactive oxygen species (ROS) levels. The results of the cell cycle and morphological analysis suggest that p53 induced quiescence, not senescence, following Akt inhibition. Conversely, the inhibition of p21 induction abolished cell cycle arrest but did not affect the p53‐induced increase in ROS levels. Additionally, p21 and Akt separately controlled cell cycle arrest and ROS levels, respectively, during H‐Ras‐induced senescence in human normal fibroblasts. The mechanistic analysis revealed that Akt increased ROS levels through NOX4 induction, and increased Akt‐dependent NF‐κB binding to the NOX4 promoter is responsible for NOX4 induction upon p53 expression. We further showed that Akt activation upon p53 expression is mediated by mammalian target of rapamycin complex 2. In addition, p53‐mediated IL6 and IL8 induction was abrogated by Akt inhibition, suggesting that Akt activation is also required for the senescence‐associated secretory phenotype. Collectively, these results suggest that p53 simultaneously controls multiple pathways to induce cellular senescence through p21 and Akt.

Partial Text

Cellular senescence was first described by Hayflick et al. as the limited proliferative capacity of normal human fibroblasts. While this phenomenon, termed ‘replicative senescence’, is known to contribute to organismal aging processes, another type of cellular senescence, termed ‘premature senescence’ or ‘stress‐induced senescence’, is considered a barrier to tumorigenesis because it acts to remove precancerous cells (Collado et al., 2005; Collado & Serrano, 2010). Cellular senescence has also been recently implicated in various pathophysiological conditions, such as wound healing and tissue fibrosis (Munoz‐Espin & Serrano, 2014; Tominaga, 2015). Given the importance of cellular senescence in tissue homeostasis and various diseases, it is important to understand the underlying regulatory pathways.

In this study, we have provided evidence that cell cycle arrest and increased ROS levels are regulated separately by p21 and Akt, respectively, during p53‐dependent premature senescence. Although p21 is known to be essential for p53‐induced cell cycle arrest, this is the first report that Akt is required for the p53‐induced increase in intracellular ROS levels. These separate roles in cell cycle arrest and ROS induction were also demonstrated during H‐Ras‐induced senescence in normal human WI‐38 fibroblasts. These data suggest that Akt and p21 cooperation is involved in a common mechanism underlying the induction of p53‐dependent premature senescence (Fig. 6I). The necessity of cooperation among more than two regulators for the induction of cellular senescence has been previously reported in a variety of cellular contexts. For example, the coexpression of Ras or Mek1 together with p53 is required to induce senescence in p53‐null MEFs (Ferbeyre et al., 2002). The Raf and Akt pathways have been shown to cooperatively control the onset of senescence in response to anticancer drug and hormonal treatments (Taylor et al., 2011). Therefore, our and other researchers’ results suggest that cellular senescence is induced by the contributions of multiple signaling pathways and that this induction is dependent on the cellular context rather than on a specific linear pathway.

Detailed descriptions of the experimental procedures, reagents, and associated references can be found in Appendix S1 (Supporting information).

This work was supported by a grant from National Research Foundation of Korea (2016R1A2B2008887) and Medical Research Center grant (2016R1A5A2007009).

H.J.J., Y.Y.K., and J.Y. initiated and designed the study. Y.Y.K., H.J.J., and J.U. performed the experiments and analyzed and interpreted the data. J.U., Y.M.K., and S.S.B. conceived the specific experiments and participated in writing the manuscript draft. H.J.J., Y.Y.K., and J.Y. wrote the manuscript. All authors reviewed the manuscript and provided editorial input.

None declared.




Leave a Reply

Your email address will not be published.