Cellular senescence is a program of arrested proliferation and altered gene expression triggered by many stresses. Although it is a potent tumor-suppressive mechanism, senescence has been implicated in several pathological processes including aging, age-associated diseases, and (counterintuitively) tumorigenesis. One potential mechanism through which senescent cells exert such pleiotropic effects is the secretion of proinflammatory cytokines, chemokines, growth factors, and proteases, termed the senescence-associated secretory phenotype (SASP), which affects senescent cells and their microenvironment. The mechanism by which the SASP is initiated and maintained is not well characterized beyond the classical regulators of inflammation, including the transcription factors NF-κB and C/EBPβ.

In senescence growth arrest, two core senescence-regulating pathways, p53 and p16^INK4a/Rb, play a critical role. By contrast, the SASP does not depend on either p53 or p16^INK4a, which suggests the existence of an independent senescence regulatory network that controls the SASP. Having observed high levels of induction of microRNA miR-146a during induced senescence in human fibroblasts, we developed a green fluorescent protein–tagged senescence reporter based on a miR-146a promoter fragment. This reporter responded to senescence-inducing stimuli, including replicative exhaustion, DNA damage, and oncogenic RAS activation—all of which activate the SASP. This system allowed us to identify additional regulators of senescence and the SASP.

Through miR-146a promoter analysis, we mapped the critical region for senescence-induced activity and identified the transcriptional regulator responsible for this regulation, GATA4, previously known as a regulator of embryonic development. Ectopic expression of GATA4 induced senescence, whereas disruption of GATA4 suppressed it, thus establishing GATA4 as a senescence regulator. GATA4 protein abundance, but not mRNA, increased during sene1scence, primarily as a result of increased protein stability.

Under normal conditions, GATA4 binds the p62 autophagy adaptor and is degraded by selective autophagy. Upon senescence induction, however, this selective autophagy was suppressed through decreased interaction between GATA4 and p62, thereby stabilizing GATA4. GATA4 in turn induced TRAF3IP2 (tumor necrosis factor receptor–associated factor interacting protein 2) and IL1A (interleukin 1A), which activate NF-κB to initiate and maintain the SASP, thus facilitating senescence. GATA4 pathway activation depends on the key DNA damage response (DDR) kinases ATM (ataxia telangiectasia mutated) and ATR (ataxia telangiectasia and Rad3–related), as does senescence-associated activation of p53 and p16^INK4a. However, the GATA4 pathway is independent of p53 and p16^INK4a. Finally, GATA4 protein accumulated in multiple tissues in mice treated with senescence-inducing stimuli and during normal mouse and human aging, including many cell types in the brain; these findings raise the possibility that the GATA4 pathway drives age-dependent inflammation.

Our results indicate that GATA4 connects autophagy and the DDR to senescence and inflammation through TRAF3IP2 and IL1A activation of NF-κB. These findings establish GATA4 as a key switch activated by the DDR to regulate senescence, independently of p53 and p16^INK4a.

Our in vivo data indicate a potential role of GATA4 during aging and its associated inflammation. Because accumulation of senescent cells is thought to promote aging and aging …