Autophagy mediates the mitotic senescence transition

Abstract

As a stress response, senescence is a dynamic process involving multiple effector mechanisms whose combination determines the phenotypic quality. Here we identify autophagy as a new effector mechanism of senescence. Autophagy is activated during senescence and its activation is correlated with negative feedback in the PI3K-mammalian target of rapamycin (mTOR) pathway. A subset of autophagy-related genes are up-regulated during senescence: Overexpression of one of those genes, ULK3, induces autophagy and senescence. Furthermore, inhibition of autophagy delays the senescence phenotype, including senescence-associated secretion. Our data suggest that autophagy, and its consequent protein turnover, mediate the acquisition of the senescence phenotype.

Figure 1.

Autophagic activity is up-regulated in OIS. (A) Immunoblot analysis of LC3 in IMR90 cells expressing H-rasV12 (R) or vector (V) at 7 d post-antibiotic selection (PS7), uninfected growing cells (G) and quiescent cells (Q) generated by 5 d confluency. (HMGA2) Marker of senescence; (β-actin) loading control; (Bafilomycin A1 [Baf] treatment) 100 mM for 3 h before extraction. (B) Long-lived protein degradation analysis in cells expressing Vector, H-rasV12 (Ras), MEKQ56P (MEK), or H-rasV12 and E1A (R/E) at PS9. Quiescent cells were generated either by 3 d 0.1% serum (0.1%) or 5 d confluency (Conf). Cells cultured in DMEM without amino acids or serum for 4 h (Starve) serve as a positive control. Values are mean ± SEM. (*) P < 0.05 relative to vector. (C) Confocal images of cells expressing EGFP-LC3B and either vector or H-rasV12. (D) Confocal images for EGFP-LC3B and LysoTracker in Ras senescent cells. Cells were labeled with 200 mM LysoTracker for 2 h. (E) Electron microscopy images of vector and Ras senescent cells. Arrowheads show representative autophagic vacuoles.

Figure 2.

Correlation between negative feedback on mTOR activity and autophagy induction during stress-induced senescence. (A) ER:Ras-expressing cells were given 100 nM 4-OHT and assessed for SA-β-gal, BrdU incorporation, senescence-associated heterochromatic foci (SAHF) formation, and expression of the indicated proteins at the various time points. Each value of SAHF and BrdU represents the mean ± SEM. For SA-β-gal a representative value is shown. p16 and HMGA2 serve as markers of senescence. (B) Cells were synchronized in G0 using low serum conditions (0.1%), then released by adding 4-OHT together with serum (10%). Cyclin A; cell cycle marker. (C) DNA damage was induced by a 48-h Etoposide (100 μM) treatment. γH2AX, cyclin A, and HMGA2 serve as markers of DNA damage, cell cycle, and senescence, respectively.

Figure 3.

Regulation of autophagy and senescence genes after Ras induction. (A) Transcriptional profiles of autophagy, lysosomal and senescent genes during OIS. Total RNA was extracted from ER:Ras cells at the indicated time points after 4-OHT addition and assayed on Illumina microarrays. The data are the weighted mean of three independent experiments. Log fold changes compared with time 0 are shown. (B) Expression of the indicated genes was assessed by qPCR in cells expressing H-rasV12 (Ras), MEKQ56P (MEK) or Vector at PS8. (Quiescent) Five days confluent. Amino acid-starved cells (Starv) were generated as in Figure 1B. Cyclin A confirms cell cycle arrest. (p16) Senescence marker. Values are mean ± SEM. (C) Growth curves of IMR90 cells expressing vector or HA-ULK3. Data are representative of three independent experiments. Inset shows immunoblot analysis at day 18.

Figure 4.

Autophagy contributes to the transition to the OIS phenotype. (A) SA-β-gal activity and (B) BrdU incorporation in ER:Ras IMR90 expressing the indicated shRNAs at day 4 after 4-OHT treatment. *P < 0.05 relative to vector. (C) The indicated ER:Ras populations of BJ cells were assessed for colony formation in the presence of 4-OHT. (D,E) Autophagy deficiency delays cytokine production as assessed by immunoblotting (D) and qPCR (E) of ER:Ras IMR90 cells. Values are mean ± SEM of at least three experiments (A,B,E).

Figure 5.

Immunohistochemical analysis of Ki67, Dec1, and LC3 in the same DMBA/TPA-induced mouse papillomas. (A) A representative example is shown from a total of seven papillomas from seven different animals. (Red dotted lines) Basement membrane. (B) Model of the relationship between autophagy and OIS.