Search

Elimination of senescent cells (SnC) is anti-atherogenic, but the specific contribution of senescent vascular endothelial cells (EC) is unknown. We inactivated angiopoietin like-2 (angptl2), a marker of SnEC and a pro-atherogenic cytokine in LDLr -/- , hApoB 100 +/+ atherosclerotic (ATX) mice. Three months after a single vascular delivery of a small hairpin (sh)Angptl2 in 3-month old ATX mice using an adeno-associated virus serotype 1 (AAV1), aortic atheroma plaque progression was slowed by 58% (p<0.0001). In the native aortic endothelium, angptl2 expression was decreased by 80%, in association with a reduced expression of p21 , a cyclin-dependent kinase inhibitor overexpressed in growth-arrested SnC. Endothelial activation was reduced (lower Icam-1, Il-1β and Mcp-1 expression), decreasing monocyte Cd68 expression in the endothelium. One week post-injection, the ratio Bax/Bcl2 increased in the endothelium only, suggesting that angptl2 + /p21 + SnEC were eliminated by apoptosis. Four weeks post-injection, the endothelial progenitor marker Cd34 increased, suggesting endothelial repair. In arteries of atherosclerotic patients, we observed a strong correlation between p21 and ANGPTL2 (r=0.727, p=0.0002) confirming the clinical significance of angptl2 -associated senescence. Our data suggest that therapeutic down-regulation of vascular angptl2 leads to the clearance of SnEC by apoptosis, stimulates endothelial repair and reduces atherosclerosis.

The mechanism by which p53 suppresses tumorigenesis remains poorly understood. In the context of aberrant activation of the JAK/STAT5 pathway, SOCS1 is required for p53 activation and the regulation of cellular senescence. In order to identify p53 target genes acting during the senescence response to oncogenic STAT5A, we characterized the transcriptome of STAT5A-expressing cells after SOCS1 inhibition. We identified a set of SOCS1-dependent p53 target genes that include several secreted proteins and genes regulating oxidative metabolism and ferroptosis. Exogenous SOCS1 was sufficient to regulate the expression of p53 target genes and sensitized cells to ferroptosis. This effect correlated with the ability of SOCS1 to reduce the expression of the cystine transporter SLC7A11 and the levels of glutathione. SOCS1 and SOCS1-dependent p53 target genes were induced during the senescence response to oncogenic STAT5A, RasV12 or the tumor suppressor PML. However, while SOCS1 sensitized cells to ferroptosis neither RasV12 nor STAT5A mimicked the effect. Intriguingly, PML turned cells highly resistant to ferroptosis. The results indicate different susceptibilities to ferroptosis in senescent cells depending on the trigger and suggest the possibility of killing senescent cells by inhibiting pathways that mediate ferroptosis resistance.

Mutants of lamin A cause diseases including the Hutchinson-Gilford progeria syndrome (HGPS) characterized by premature aging. Lamin A undergoes a series of processing reactions, including farnesylation and proteolytic cleavage of the farnesylated C-terminal domain. The role of cleavage is unknown but mutations that affect this reaction lead to progeria. Here we show that interphase serine 22 phosphorylation of endogenous mutant lamin A (progerin) is defective in cells from HGPS patients. This defect can be mimicked by expressing progerin in human cells and prevented by inhibition of farnesylation. Furthermore, serine 22 phosphorylation of non-farnesylated progerin was enhanced by a mutation that disrupts lamin A head to tail interactions. The phosphorylation of lamin A or non-farnesylated progerin was associated to the formation of spherical intranuclear lamin A droplets that accumulate protein kinases of the CDK family capable of phosphorylating lamin A at serine 22. CDK inhibitors compromised the turnover of progerin, accelerated senescence of HGPS cells and reversed the effects of FTI on progerin levels. We discuss a model of progeria where faulty serine 22 phosphorylation compromises phase separation of lamin A polymers, leading to accumulation of functionally impaired lamin A structures.

Eukaryotic cells respond to various forms of stress by blocking mRNA translation initiation via the phosphorylation of the alpha (α) subunit of eIF2 at serine 51 (S51) (eIFαP). An important role of eIF2αP is the regulation of redox homeostasis and adaptation of cells to oxidative stress. Herein, we demonstrate that eIF2αP guards cells from intracellular reactive oxygen species (ROS) via the inhibition of senescence. Specifically, genetic inactivation of either eIF2αP or eIF2α kinase PERK in primary mouse or human fibroblasts leads to proliferative defects associated with increased DNA damage, G2/M accumulation and induction of premature senescence. Impaired proliferation of either PERK or eIF2αP-deficient primary cells is caused by increased ROS and restored by anti-oxidant treatment. Contrary to primary cells, impaired eIF2αP in immortalized mouse fibroblasts or human tumor cells provides tolerance to elevated intracellular ROS levels. However, eIF2αP-deficient human tumor cells are highly susceptible to extrinsic ROS generated by the pro-oxidant drug doxorubicin by undergoing premature senescence. Our work demonstrates that eIF2αP determines cell destiny through its capacity to control senescence in response to oxidative stress. Also, inhibition of eIF2αP may be a suitable means to increase the anti-tumor effects of pro-oxidant drugs through the induction of senescence.

Members of the signal transducers and activators of transcription (STATs) family of proteins, which connect cytokine signaling to activation of transcription, are frequently activated in human cancers. Suppressors of cytokine signaling (SOCS) are transcriptional targets of activated STAT proteins that negatively control STAT signaling. SOCS1 expression is silenced in multiple human cancers suggesting a tumor suppressor role for this protein. However, SOCS1 not only regulates STAT signaling but can also localize to the nucleus and directly interact with the p53 tumor suppressor through its central SH2 domain. Furthermore, SOCS1 contributes to p53 activation and phosphorylation on serine 15 by forming a ternary complex with ATM or ATR. Through this mechanism SOCS1 regulates the process of oncogene-induced senescence, which is a very important tumor suppressor response. A mutant SOCS1 lacking the SOCS box cannot interact with ATM/ATR, stimulate p53 or induce the senescence phenotype, suggesting that the SOCS box recruits DNA damage activated kinases to its interaction partners bound to its SH2 domain. Proteomic analysis of SOCS1 interaction partners revealed other potential targets of SOCS1 in the DNA damage response. These newly discovered functions of SOCS1 help to explain the increased susceptibility of Socs1 null mice to develop cancer as well as their propensity to develop autoimmune diseases. Consistently, we found that mice lacking SOCS1 displayed defects in the regulation of p53 target genes including Mdm2, Pmp22, PUMA and Gadd45a. The involvement of SOCS1 in p53 activation and the DNA damage response defines a novel tumor suppressor pathway and intervention point for future cancer therapeutics.