Skip to main content
NCBI home page
The FASEB Journal logoLink to The FASEB Journal
. 2009 Mar;23(3):783–794. doi: 10.1096/fj.08-114256

Hyperoxia-induced premature senescence requires p53 and pRb, but not mitochondrial matrix ROS

Tatyana A Klimova *, Eric L Bell *, Emelyn H Shroff *, Frank D Weinberg *, Colleen M Snyder *, Goberdan P Dimri †,‡, Paul T Schumacker §, G R Scott Budinger *, Navdeep S Chandel *,‡,1
PMCID: PMC2653981  PMID: 18948382

Abstract

Senescence is a potential tumor-suppressing mechanism and a commonly used model of cellular aging. One current hypothesis to explain senescence, based in part on the correlation of oxygen with senescence, postulates that it is caused by oxidative damage from reactive oxygen species (ROS). Here, we further test this theory by determining the mechanisms of hyperoxia-induced senescence. Exposure to 70% O2 led to stress-induced, telomere-independent senescence. Although hyperoxia elevated mitochondrial ROS production, overexpression of antioxidant proteins was not sufficient to prevent hyperoxia-induced senescence. Hyperoxia activated AMPK; however, overexpression of a kinase-dead mutant of LKB1, which prevented AMPK activation, did not prevent hyperoxia-induced senescence. Knocking down p21 via shRNA, or suppression of the p16/pRb pathway by either BMI1 or HPV16-E7 overexpression, was also insufficient to prevent hyperoxia-induced senescence. However, suppressing p53 function resulted in partial rescue from senescence, suggesting that hyperoxia-induced senescence involves p53. Suppressing both the p53 and pRb pathways resulted in almost complete protection, indicating that both pathways cooperate in hyperoxia-induced senescence. Collectively, these results indicate a ROS-independent but p53/pRb-dependent senescence mechanism during hyperoxia.—Klimova, T. A., Bell, E. L., Shroff, E. H., Weinberg, F. D., Snyder, C. M., Dimri, G. P., Schumacker, P. T., Budinger, G. R. S., Chandel, N. S. Hyperoxia-induced premature senescence requires p53 and pRb, but not mitochondrial matrix ROS.

Keywords: AMPK, cell cycle, telomerase

Cellular senescence, or permanent, irreversible replication arrest, was first characterized in 1961, and occurs in all primary cells studied to date (1). Senescence can occur after a certain number of cell divisions (replicative senescence), or anytime the cell encounters sublethal levels of certain stresses, such as elevation of reactive oxygen species (ROS), oncogene overexpression, or irradiation. Because of the timing mechanism inherent in replicative senescence, it is thought to be a model for cellular aging. Also, since senescence would limit proliferation of potentially transformed cells, it may be a possible tumor-suppressing mechanism.

However, despite the interest in senescence, our understanding of its underlying mechanisms is still incomplete. It is widely accepted that human cells undergo replicative senescence when telomeres, structures that protect the ends of linear chromosomes, shorten during cell division, finally acting like a double-stranded DNA break to induce sustained proliferative arrest (reviewed in ref. 2). DNA damage response protein p53 is often involved, usually acting through its direct downstream target p21 (3, 4). Likewise, p16, a cell cycle checkpoint protein, is also often implicated through signaling to pRb (5,6,7). These factors are also involved in stress-induced senescence, although the exact upstream events that trigger their activation remain unclear, and likely vary in different cell types and the type of senescence-inducing stressor used.

It has been shown that treatment with exogenous reactive oxygen species, such as H2O2, induces premature senescence (8, 9). In fact, nearly all stresses that induce senescence are thought to increase intracellular ROS (10). It was broadly believed that high oxygen levels (hyperoxia) should elevate intracellular ROS production. Indeed, many researchers have found elevated ROS production in cells exposed to hyperoxia, and human fibroblasts cultured under 40% O2 senesce more quickly than their normoxic counterparts (11). Conversely, reducing oxygen levels (hypoxia) extends replicative life span of human fibroblasts compared to their normoxic counterparts (12, 13). The accepted interpretation was that under low oxygen levels, cells generate lower levels of ROS, resulting in less damage to cellular components and delayed senescence.

We recently demonstrated that ROS levels increase, rather than decrease, in human lung fibroblasts cultured under hypoxia (14). Furthermore, we showed that the increase in ROS activated the transcription factor hypoxia inducible factor (HIF), which induces transcription of the hTERT gene, a catalytic component of the enzyme telomerase. The increase in telomerase activity during hypoxia extends replicative life span of fibroblasts during hypoxia, thus resolving the apparent discrepancy that increased levels of ROS can increase life span. This finding indicates that the connection between ROS levels and senescence is not as direct as it had previously been assumed and that a closer exploration of the free radical theory is needed. To further examine the theory, in this study we focused on the mechanisms of senescence under hyperoxia (defined here as 70% O2).

MATERIALS AND METHODS

Cell culture

Normal primary diploid fibroblasts from adult human lung (PHLFs) were purchased from Cambrex (East Rutherford, NJ, USA) and cultured in FGM-2 medium at 37°C in 21% O2/5% CO2 humidified incubators. They were passaged on reaching confluence. For senescence studies, low-passage PHLFs were seeded into 35-mm dishes at 3 × 104 cells/plate, or ∼25% confluency. Hyperoxia exposure (70% O2/5% CO2) for 72 h occurred in a humidified chamber, with gas delivery controlled by Oxycycler model C42 chamber and software (Biospherix, Lacona, NY, USA). Alternatively, hyperoxic conditions were achieved in a humidified Plexiglas chamber supplied with a gas mixture of 70% O2/5% CO2/balance N2. As a positive control, PHLFs were treated with 100 μM H2O2 for 2 h; then medium was changed, and cells were allowed to develop the senescent phenotype for 72 h at 21% O2. Untreated negative controls were left at 21% O2 for 72 h. After 72 h, PHLFs were placed in fresh FGM-2 and allowed to rest at 21% O2 overnight before proceeding to stain for senescence-associated β-galactosidase activity.

Senescence-associated β-galactosidase staining

To determine senescence-associated (SA) β-galactosidase activity in PHLFs, cells were exposed to conditions as above, fixed, and stained at pH 6.0 using the SA-β-galactosidase staining kit (Cell Signaling, Danvers, MA, USA). Twenty-four hours after staining, phase microscopy images were obtained with a Nikon Eclipse TE200 inverted microscope (Nikon, Tokyo, Japan) at ×100. Five random fields were captured per plate, and the numbers of total and β-galactosidase-positive cells were manually counted.

Lactate dehydrogenase assay

To determine whether reduced cell number after 70% O2 exposure is the result of cell death rather than senescence, PHLFs were plated at 50% confluency in 60-mm dishes and exposed to 70% O2 or 100 μM or 1 mM H2O2 as above. Normoxic controls were plated to 25% confluency to prevent overconfluence at 72 h. After 72 h, cell death was determined using the lactate dehydrogenase assay kit (Roche, Basel, Switzerland) following manufacturer’s protocol.

Plasmids and constructs

pBabe-Puro retroviral expression vector (Clontech, Mountain View, CA, USA) was used to overexpress human hTert (Addgene, Cambridge, MA, USA). pBabe-Puro-BMI1 retroviral expression construct was a kind gift from Dr. G. Dimri (Northwestern University, Evanston, IL, USA). pBabe-puro-FLAG-LKB1 wild type (WT) and kinase dead (KD) were generated by Dr. L. Cantley and purchased from Addgene. pSiren-RetroQ retroviral short hairpin (shRNA) expression vector (Clontech) was used to express shRNA sequences for p21 [5′-TGTCAGAACCGGCTGGGGATT-3′; this sequence was previously published (15)] and a scrambled control shRNA [5′-AGCGCGATTTGTAGGATTCGT-3′; this sequence was previously published (16)].

Stable cell lines

Stable cell lines were generated in early-passage PHLFs by retroviral infection, using PT67 packaging cell line (Clontech). PT67 cells were transfected with 10-μg plasmid using Transit-LT1 (Mirus, Madison, WI, USA) reagent according to manufacturer’s protocol and selected with antibiotics. For infection of PHLFs, PT67 cells stably expressing the plasmid were washed once with PBS and placed in 4 ml FGM-2 medium overnight. The following day, virus-containing medium was supplemented with 8 μg/ml polybrene (Sigma-Aldrich, St. Louis, MO, USA) and applied to PHLFs. PHLFs stably expressing plasmids of interest were selected with an appropriate antibiotic. To generate PHLFs expressing p53-DN and mutant or WT E7, PHLFs were first infected with retrovirus expressing pBabe-hygro-p53-DN. After selection with hygromycin, the stable PHLFs were infected with LXSN-HPV-E7 mutant or WT retrovirus and selected with G418. PHLFs overexpressing LXSN-HPV-E7 WT, LXSN-HPV-E6 and E7 together, and empty LXSN vector were generated by the same general protocol, but using stable PA317 packaging cells. These were donated by Dr. L. Laimins (Northwestern University, Chicago, IL, USA). LXSN-HPV-E7 mutant in PHLF packaging cells were also donated by Dr. L. Laimins.

Population doubling determination

PHLFs expressing hTERT or pBabe controls were seeded at 4 × 105 cells/flask and allowed to reach ∼80% confluence. They were then trypsinized, counted, and reseeded at 4 × 105 cells. Population doublings were calculated with the formula log2(D/Do), where Do = number cells seeded and D = number of cells counted at confluence. The plateau point when cells ceased to undergo further population doublings was deemed replicative senescence.

Adenoviral infection

PHLFs were treated with a low volume of serum-free fibroblast basal medium with no serum or other supplementation. Seventy-five plaque-forming units (PFU) per cell of each virus was used, and PHLFs were incubated at 37°C for 4–6 h. Fully supplemented FGM-2 medium was added, and cells were allowed to rest overnight before being placed in conditions or harvested for immunoblotting. MnSOD and mitochondrial catalase adenoviruses were obtained from Iowa Vector Core (Iowa City, IA, USA).

Immunoblotting

Whole-cell extracts from PHLFs plated in 100-mm dishes and exposed to conditions were harvested using cell lysis buffer (Cell Signaling) supplemented with PMSF. In addition, buffer used to harvest PHLFs for analyzing phosphorylated proteins was supplemented with phosphatase inhibitors NaF (50 mM) and Na3VO4 (2 mM). To activate p16 or p21 expression under normoxia, cells were allowed to become overconfluent before harvesting. To determine p16 levels under hyperoxia, control and 70% O2-treated PHLFs were kept below 70% confluency to prevent p16 activation through overcrowding. Samples of 50 to 70 μg were resolved on sodium dodecyl sulfate-acrylamide gels (5% for ACC; 12% for all other proteins) and transferred to nitrocellulose membrane. Antibodies used were p16, p21, and MnSOD (BD, Franklin Lakes, NJ, USA), catalase (Abcam, Cambridge, MA, USA), phosphorylated and total ACC (Cell Signaling), phosphorylated and total AMPK (Cell Signaling), and LKB1 (D-19; Santa Cruz, Santa Cruz, CA, USA). α-tubulin (Sigma-Aldrich) served as a loading control for all samples.

Fluorescent microscopy

To verify correct localization of redox-sensitive green fluorescent protein (roGFP) probes, PHLFs were plated on glass-bottom microwell dishes (MakTek Corp., Ashland, MA, USA) and infected with 75 PFU/cell of adenovirus encoding cytosolic or mitochondrial roGFP probe. After 6 h infection in serum-free medium, FGM-2 was added, and cells were left at 21% O2 overnight. The following day, cells were treated with 50 nM tetramethylrhodamine (TMRE) 10 min prior to imaging to show mitochondrial localization and imaged with a Yokogawa spinning disc confocal microscope from Perkin Elmer (Waltham, MA, USA), fitted on a Nikon TE2000-U fluorescent inverted microscope using the ×100 objective. Excitation at 488 nm was used to visualize roGFP and 568 nm to visualize TMRE.

ROS measurement

To determine whether 70% O2 elevated mitochondrial ROS production, we utilized the roGFP construct targeted to the mitochondria or cytosol. PHLFs were infected with 75 PFU/cell of adenovirus encoding roGFP, along with 75 PFU/cell each of adenovirus encoding MnSOD and mitochondria-targeted catalase, or with 150 PFU/cell of sham virus. After a 6 h of infection in serum-free medium, FGM-2 was added, and cells were placed in 70% O2 or left in normoxia for 18 h. After exposure, PHLFs were harvested in trypsin and analyzed with a CyanADP flow cytometry analyzer (Dako, Glostrup, Denmark). As internal controls, samples were fully reduced with 10 mM dithiothreitol (DTT) and fully oxidized with 1 mM H2O2. The mean fluorescent channel for the ratio of violet excitable to blue excitable signal was determined with Summit software 4.2 (Dako). Percentage oxidized probe was determined with the equation (R − RDTT)/(R − RH2O2), where R is sample without DTT or H2O2 added, RDTT is the fully reduced sample, and RH2O2 is the fully oxidized sample.

AICAr treatment

PHLFs were plated at 3 × 104 cells in 35-mm plates and exposed to 0.5 mM AICAr for 16 h. Medium was then replaced, and cells were allowed to recover for 72 h before staining for SA-β-galactosidase activity as above. For determining AMPK activation, PHLFs grown in 100-mm plates were treated with fresh medium for 1 h before being treated with 0.5 mM AICAr for 30 min and harvested. AMPK activation was then tested by immunoblotting for phosphorylated ACC.

Statistical analysis

Data are presented as means ± se. One-way analysis of variance was performed in Origin 7 to determine the presence of significant differences in the data. When analysis of variance indicated that a significant difference was present, a two-sample Student’s t tests were performed to compare experimental data with appropriate controls (as indicated in each figure legend). Statistical significance was determined at a value of P < 0.05.

RESULTS

Hyperoxia induces premature senescence in primary human lung fibroblasts

To determine whether hyperoxia induces premature senescence, we cultured primary adult human lung fibroblasts (PHLFs) in either 21% O2 (approximately physiological oxygen concentration for these cells) or 70% O2 for 72 h. Besides being a useful tool for correlating senescence with O2 levels, this concentration of O2 is frequently administered to patients in intensive care for pulmonary or cardiac disorders. Unlike PHLFs grown in 21% O2, those cultured in 70% O2 displayed a marked growth arrest, as indicated by reduced number of cells seen per field, and enlarged, flattened phenotype associated with senescence (Fig. 1A). Senescence was further confirmed by staining with 5-bromo-4-chloro-3-indolyl-b″-d-galactopyranoside (X-gal) to test for β-galactosidase activity at pH 6, a well-established and most commonly used marker of senescence (17). Approximately 10% of untreated cells stained blue, while 80% stained blue in the PHLF population exposed to 70% O2 (Fig. 1B). Moreover, the phenotype of PHLFs exposed to hyperoxia was similar to that of PHLFs treated with a sublethal concentration of H2O2 (100 μM for 2 h), a stimulus widely used to induce senescence (Fig. 1A). Together, these data indicate that 70% O2 induces premature senescence in PHLFs.

Figure 1.

Figure 1.

Open in a new tab

Hyperoxia induces premature senescence, but not death, in primary human lung fibroblasts. A) Representative phase microscopy images of PHLFs exposed to 21 or 70% O2 for 72 h, or 100 μM H2O2 for 2 h, and stained with X-gal to show senescence-associated β-galactosidase activity. B) Percentage of β-galactosidase-positive cells after exposure to 21 or 70% O2 or 100 μM H2O2. *P < 0.05 vs. 21% O2 control (P=0.00010, 100 μM H2O2; P=0.00000, 70% O2); n = 4. C) Cell death measured by percentage lactate dehydrogenase released from PHLFs exposed to 21 or 70% O2 for 72 h, or to 100 μM or 1 mM H2O2 for 2 h (LDH release measured 72 h after initial exposure). *P < 0.05 vs. 21% O2 control (P=0.00072, 100 μM H2O2; P=0.00047, 1 mM H2O2); n = 4.

Hyperoxia-exposed PHLFs displayed a marked decrease in cell number compared to normoxic counterparts. A growth arrest by senescence would certainly produce such a decrease. However, high O2 levels could also be lethal (18). As senescent cells are more resistant to apoptotic stimuli (19), the high percentage of cells with senescent phenotype could therefore actually be due to the dying off of normal cells rather than an increase in senescence. To ensure that the comparative reduction in cell number after 70% O2 exposure was not due to increased cell death, we measured cell death through lactate dehydrogenase (LDH) release. LDH is an intracellular enzyme released from dying cells and serves as a global indicator of apoptotic and necrotic cell death. Although 70% O2 exposure doubled the percentage of LDH release compared to controls, cell death levels were still at ∼10% of total cells, indicating that cell death was likely not the main factor in the reduction of cell number (Fig. 1C). Neither was 100 μM H2O2 sufficient to increase cell death to more than 15% of total cells. By contrast, a 2-h treatment with 1 mM H2O2, an established lethal dose, elevated LDH release to more than 60% of total cells (Fig. 1C). We therefore conclude that 70% O2 induces senescence and not cell death.

Telomerase overexpression does not prevent hyperoxia-induced senescence

Previous studies with hyperoxia (40% O2) noted that senescence under this condition was telomere dependent (11, 20, 21). To determine whether senescence under 70% O2 is also telomere dependent, we overexpressed the catalytic component of telomerase, hTERT, in PHLFs. hTERT overexpression is sufficient to extend replicative life span indefinitely in many cell types, including fibroblasts (22, 23). Indeed, hTERT overexpression allowed PHLFs to continue proliferating indefinitely under 21% O2 (Fig. 2A), indicating that hTERT was functional. On the other hand, PHLFs expressing the empty retroviral vector ceased to proliferate after ∼30 days postinfection. However, on exposure to 70% O2, hTERT-overexpressing PHLFs senesced in similar numbers compared to vector controls (Fig. 2B). This indicates that hyperoxia-induced senescence is telomere independent and likely proceeds through a stress-induced senescence mechanism.

Figure 2.

Figure 2.

Open in a new tab

Hyperoxia-induced senescence is telomere independent. A) Population doublings in culture of PHLFs stably expressing empty vector (pBabe; squares) or pBabe-hTERT (hTERT; circles) after retroviral infection. B) Percentage of early-passage β-galactosidase positive PHLFs expressing empty vector (pBabe) or hTERT (gray) after 72-h exposure to 21 or 70% O2; n = 3.

Hyperoxia induces senescence independently of an increase in mitochondrial and cytosolic ROS generation

To date, the vast majority of senescence-inducing stressors have been shown to act through increased ROS generation. To determine whether hyperoxia-induced ROS production (as had previously been demonstrated in other systems, e.g., ref. 20), we quantified intracellular ROS levels with roGFP. This probe consists of GFP containing two additional cysteine thiols (S147C and Q204C) located on the outer surface (24, 25). The position of these residues allows for dithiol formation or breakage, depending on whether the environment of the cell is oxidizing or reducing. roGFP emission at 525 nm is assessed via flow cytometry at excitation wavelengths at 400 and 490 nm. When the probe is oxidized, the emission ratio (400 nm/490 nm) increases, and decreases when the probe is reduced (24, 25). This ratiometric behavior allows the probe to be calibrated to the redox state and renders the redox signal independent of the protein expression level. In our study, we utilized untargeted roGFP (cytosolic), and roGFP targeted to the mitochondrial matrix (Fig. 3A). In addition, roGFP was internally calibrated by treatment with 1 mM H2O2 to fully oxidize the probe and 10 mM DTT to fully reduce it.

Figure 3.

Figure 3.

Open in a new tab

Hyperoxia induces senescence independently of ROS generation. A) Fluorescence microscopy demonstrating cytosolic and mitochondrial localization of roGFP virus (green channel). TMRE used to stain mitochondria is red; colocalization indicated by yellow. B, C) Mitochondrial (B) and cytosolic (C) ROS levels detected after 16 h in 21 or 70% O2 using mitochondrial matrix or cytosolic roGFP, respectively, expressed via adenovirus in PHLFs. Intracellular ROS levels are shown as a percentage of roGFP maximally oxidized by treating PHLFs with 1 mM H2O2 and maximally reduced by addition of 10 mM DTT for 5 min. PHLFs were coinfected with null virus (white) or MnSOD and mitochondrially targeted catalase (gray). *P < 0.05 vs. null virus at 21% O2 (P=0.00039); n=4. D) Percentage β-galactosidase-positive PHLFs expressing null virus (white) or MnSOD and mitochondrial catalase together (gray) after 72 h in 21 or 70% O2; n = 4. E) MnSOD and catalase protein levels in whole-cell lysates of PHLFs infected with sham adenovirus or that expressing MnSOD and mitochondrial catalase.

Exposure to 70% O2 for 18 h increased mitochondrial matrix, but not cytosolic, roGFP oxidation over normoxic controls (Fig. 3B, C, respectively). This indicates that hyperoxia increases mitochondrial matrix ROS generation but that these ROS are confined to that compartment, as no corresponding increase in ROS is seen in the cytosol. The overexpression of mitochondrial matrix antioxidant proteins, MnSOD, and catalase tagged with a mitochondrial localization sequence, lowered mitochondrial roGFP oxidation to normoxic levels (Fig. 3B), indicating that hyperoxia selectively increases mitochondrial matrix ROS generation. (Since cytosolic roGFP was not significantly oxidized by hyperoxia, MnSOD and mitochondrial catalase did not significantly affect cytosolic roGFP oxidation.) However, after hyperoxia exposure, PHLFs overexpressing MnSOD and mitochondria-targeted catalase displayed numbers of β-galactosidase-positive cells comparable to PHLFs infected with sham virus (Fig. 3D) despite efficient overexpression (Fig. 3E) and function (Fig. 3B) of these proteins. Collectively, these results indicate that hyperoxia-induced senescence occurs independently of mitochondrial matrix ROS generation.

Hyperoxia-induced senescence does not involve AMP activated protein kinase

Hyperoxia leads to a metabolic crisis by disrupting oxidative phosphorylation (26, 27). As cells continue to utilize energy without being able to produce more ATP, the AMP to ATP ratio will increase. The rise in the AMP/ATP ratio activates AMP-activated protein kinase, which signals to reduce cellular energy consumption and to increase energy production. AMPK can also signal to both p53 and pRb and induce senescence (28, 29). Furthermore, LKB1, a well-known regulator of AMPK, can induce senescence in certain types of human tumor cell lines (30). Indeed, treatment with 5-aminoimidazole-4-carboxamide-1-ß-d-ribofuranoside (AICAr), an AMP mimetic and known activator of AMPK, is sufficient to activate AMPK (Fig. 4A) and induce senescence in PHLFs (Fig. 4B, C) under normal oxygen conditions. Furthermore, AMPK can be activated by various stressors, including 100 μM H2O2. 70% O2 exposure also rapidly activates phosphorylation of acetyl CoA-carboxylase (ACC), a direct phosphorylation target of activated AMPK (Fig. 5A). Because AMPK can induce senescence and is activated by hyperoxia exposure, it represented a likely candidate to regulate hyperoxia-induced senescence. To test this, we overexpressed WT- or KD-LKB1 in PHLFs. LKB1 is a well-known activator of AMPK; after AMP binds to AMPK, LKB1 recognizes it for phosphorylation and activation. Both WT- and KD-LKB1 were efficiently overexpressed in PHLFs (Fig. 5B). On treatment with 100 μM H2O2 or 70% O2, levels of phosphorylated ACC increased significantly in PHLFs expressing WT-LKB1 (Fig. 5C). However, KD-LKB1 overexpression prevented the increase in phosphorylated ACC, indicating that KD-LKB1 prevents AMPK activation under H2O2 and hyperoxia (Fig. 5C). Surprisingly, both WT-LKB1 and KD-LKB1-expressing PHLFs underwent senescence at similar levels to control cells after hyperoxia exposure (Fig. 5D). Thus, we conclude that AMPK is not required for hyperoxia-induced senescence.

Figure 4.

Figure 4.

Open in a new tab

AMPK activation is sufficient to induce senescence under normoxia. A) Phosphorylated ACC and α-tubulin levels in PHLFs treated with AMPK activator AICAr (0.5 mM) for 30 min. B) Representative phase microscopy image of PHLFs stained with X-gal to show β-galactosidase activity; n = 3. PHLFs were untreated for 72 h or treated with 0.5 mM AICAr for 16 h at 21% O2 and allowed to develop senescent phenotype for 72 h before staining. C) Percentage β-galactosidase-positive PHLFs after treatment with 0.5 mM AICAr as above. *P < 0.05 vs. untreated control; n = 3.

Figure 5.

Figure 5.

Open in a new tab

Suppressing AMPK activation is not sufficient to prevent hyperoxia-induced senescence. A) Levels of phosphorylated and total ACC after treatment with 70% O2 for indicated time. Treatment with 100 μM H2O2 for 5 min serves as a positive control. Blots are representative of 3 independent experiments. B) Levels of LKB1 protein in PHLFs infected with retrovirus expressing empty pBabe vector or WT- or KD-LKB1. C) Levels of phosphorylated and total ACC and AMPK in PHLFs overexpressing empty vector, WT-LKB1, or KD-LKB1. Cells were placed in fresh medium for 1 h before treatment with 70% O2-pre-equilibrated medium for 30 min, or with 100 μM H2O2 for 5 min, and harvested. D) Percentage β-galactosidase-positive PHLFs overexpressing empty vector, WT-LKB1, or KD-LKB1 after exposure to 21 or 70% O2 for 72 h; n = 4.

BMI1 overexpression is insufficient to prevent hyperoxia-induced senescence

p16, which signals to retinoblastoma protein (pRb) to maintain cell cycle arrest, is a widely known regulator of senescence (31, 32). In many cases of senescence, the p16/pRb pathway acts independently of the p53/p21 pathway, though cooperation can also occur (4). p16 is often thought to be activated by stress; indeed, 72-h exposure to 70% O2 elevated p16 levels in PHLFs (Fig. 6A), suggesting an involvement of p16 in hyperoxia-induced senescence. Recently, members of the Polycomb group proteins, particularly BMI1 and CBX7, have been shown to suppress p16 expression (33). While CBX7 affects both the p16 and p21 pathways, BMI1 appears to act only on p16 (34, 35). Overexpressing BMI1 can be sufficient to prevent senescence (33, 36). In addition, aging cells reduce BMI1 levels (37); some other senescence-inducing stimuli also reduce BMI1 levels, leading to senescence (38, 39). Thus, BMI1 overexpression serves not only as a useful tool for suppressing p16 function, but also as a potential regulator of hyperoxia-induced senescence in its own right. However, 70% O2 exposure does not reduce BMI1 levels (Fig. 6B). Since BMI1 is a transcriptional repressor that physically blocks p16 transcription, its activity may be modulated by its DNA binding, rather than affecting its protein levels; therefore, we proceeded with testing BMI1 involvement. We overexpressed BMI1 in PHLFs (Fig. 6C). BMI1 overexpression suppressed p16 protein levels in overconfluent cells (overconfluence activates p16 levels), as expected, but not p21 (Fig. 6C), indicating fully functional BMI1 protein. Indeed, p21 levels were sometimes elevated in BMI1 overexpressing cells (data not shown). This is consistent with previous findings that p21 and p16 pathways can compensate for each other. The majority of BMI1 overexpressing PHLFs still senesced after 70% O2 exposure despite p16 suppression (Fig. 6D). This suggests that BMI1 overexpression and subsequent suppression of the p16/pRb pathway alone is insufficient to prevent hyperoxia-induced senescence. To confirm this finding, we overexpressed the human papillomavirus 16 protein E7 (HPV-E7), which down-regulates pRb function (40). E7-overexpressing PHLFs underwent senescence in similar numbers to PHLFs expressing vector control (Fig. 6E). This further supported our finding that hyperoxia-induced senescence does not involve the p16/pRb pathway alone.

Figure 6.

Figure 6.

Open in a new tab

Inhibiting the p16/pRb pathway is insufficient to prevent hyperoxia-induced senescence. A, B) p16 (A) or BMI1 (B) protein levels after exposure to 21 or 70% O2. Cells were kept below 70% confluence to prevent p16 activation by overcrowding. Blot is representative of 4 independent experiments. C) BMI1, p16, and p21 protein levels in PHLFs overexpressing BMI1 or expressing empty vector. Cells were allowed to reach overconfluence to activate p16. D) Percentage β-galactosidase-positive PHLFs expressing empty vector or BMI1 after exposure to 21 or 70% O2 for 72 h; *P < 0.05 vs. control; n = 4. E) Percentage β-galactosidase-positive PHLFs expressing empty vector or HPV16 E7 after 72 h exposure to 21 or 70% O2; n = 4.

Inactivating the p53 pathway partially protects against hyperoxia-induced senescence

p53 is a known regulator of senescence, usually acting through p21 activation (3, 4). To begin determining whether the p53/p21 pathway is responsible for hyperoxia-induced senescence, we suppressed p53 function by overexpressing a dominant negative p53 (DN-p53) in a retroviral vector, which resulted in effective suppression of p21 expression (Fig. 7A), indicating suppression of p53 function. PHLFs overexpressing DN-p53 showed a significant decrease in the number of β-galactosidase-positive cells after 70% O2 exposure compared to PHLFs expressing empty vector (Fig. 7B). However, since incomplete senescence suppression was observed, it is possible that p53 acts in cooperation with another protein.

Figure 7.

Figure 7.

Open in a new tab

Inhibiting the p53 pathway, but not p21, partially protects against hyperoxia-induced senescence. A) Immunoblot demonstrating suppression of p21 protein expression by overexpression of DN-p53 in PHLFs. B) Percentage β-galactosidase-positive PHLFs expressing vector control or DN-p53 after 72 h exposure to 21 or 70% O2. *P < 0.05 (P=0.03876); n = 4. C) p21 protein levels in PHLFs expressing scrambled shRNA or shRNA targeted against p21. D) Percentage β-galactosidase-positive PHLFs expressing scrambled shRNA or p21 shRNA after 72 h exposure to 21 or 70% O2; *P < 0.05 vs. control; n = 3.

p21 suppression is insufficient to protect against hyperoxia-induced senescence

While p53 has numerous downstream targets, the one most well known to be involved in senescence is p21. We, therefore, used short hairpin RNA (shRNA) to knock down p21 expression. p21 expression was sufficiently suppressed using this shRNA despite growing cells to overconfluence to activate p21 (Fig. 7C). However, under hyperoxia, p21-shRNA-expressing PHLFs senesced in similar numbers to controls expressing a scrambled shRNA, indicating that p21 suppression is insufficient to prevent hyperoxia-induced senescence (Fig. 7D). p53 has many downstream targets besides p21, some of which can induce senescence (41), though p21 is the most well-known p53 target in this role. Our data indicate that the p53 pathway is involved in hyperoxia-induced senescence, but possibly not through p21 activation.

Hyperoxia-induced senescence involves cooperation between the p53 and pRb pathways

Cells often senesce via either the p53/p21 or the p16/pRb pathway, but the pathways can act on each other and cooperate to induce senescence. For example, p21 can also maintain pRB in its hypophosphorylated (active) form. Furthermore, suppressing the p16/pRb pathway can increase p21 expression (as previously reported and seen in our own system), indicating that one pathway can compensate for the lack of the other. To test for the requirement of both the p53/p21 and p16/pRb pathways under hyperoxia, we overexpressed the HPV proteins E6 (which suppresses p53 function; ref. 42)and E7 together in the same cells. While ∼70% of the PHLFs expressing vector controls senesced under 70% O2, only 20% of E6/E7 expressing PHLFs underwent senescence (Fig. 8A), indicating that both the p53 and pRb pathways participate in hyperoxia-induced senescence. This demonstrates that hyperoxia-induced senescence proceeds through genetically coded and previously well-described senescence pathways.

Figure 8.

Figure 8.

Open in a new tab

Hyperoxia-induced senescence requires both the p53 and p16/pRb pathways. A) Percentage β-galactosidase-positive PHLFs expressing empty vector or HPV16/18 E6 and E7 after 72 h exposure to 21 or 70% O2. *P < 0.05 (P=0.00006); n = 4. B) Percentage β-galactosidase-positive PHLFs expressing E7 and either scrambled shRNA (Scr) or shRNA against p21 after 72 h exposure to 21 or 70% O2; n = 2 experiments of 3 separate wells each. C) Western blot showing reduced pRb levels in PHLFs expressing p53DN in combination with WT E7, but not with pLXSN empty vector or E7 mutant deficient in pRb binding (MutE7). D) Percentage β-galactosidase-positive PHLFs expressing DN-p53 mutant and either WT E7- or E7-deficient in pRb binding (MutE7), after 72-h exposure to 21 or 70% O2. *P < 0.05 (P=0.03679); n = 3.

To test whether p21 was necessary for cooperation with pRb and regulation of hyperoxia-induced senescence, we expressed scrambled shRNA or shRNA against p21 in PHLFs expressing E7. Both sets of cells senesced in similar numbers when exposed to hyperoxia (Fig. 8B). This further suggests that p21 alone is not sufficient in regulating hyperoxia-induced senescence or that other targets of p53 are also involved.

E7 interacts with several other Rb-family proteins, including p107 and p130. Likewise, E6 interacts with and affects the function of proteins besides p53. To ensure that hyperoxia-induced senescence depended specifically on p53 and pRb, we overexpressed WT E7 or E7 deficient in binding to pRb in PHLFs expressing DN-p53 (Fig. 8C, D). While WT E7 efficiently reduced pRb protein levels, neither pLXSN vector nor E7 mutant deficient for pRb binding was able to reduce pRb levels (Fig. 8C). As expected, PHLFs expressing mutant E7 underwent senescence at the same rate as PHLFs expressing only DN-p53 when exposed to hyperoxia (compare Figs. 8D and 7B). However, PHLFs expressing WT E7 were able to escape hyperoxia-induced senescence almost entirely (Fig. 8D). Because the mutant E7 is capable of negating the function of p107 and p130, and because it is unable to further reduce hyperoxia-induced senescence in PHLFs expressing DN-p53, we conclude that WT E7 suppresses senescence through its inhibition of pRb.

DISCUSSION

ROS independence of hyperoxia-induced senescence

Despite the biological importance of senescence, its mechanisms remain incompletely understood. A prevalent theory is that senescence occurs through oxidative damage by elevated mitochondrial or cytosolic ROS levels. In the present study, we used a redox-sensitive GFP (roGFP) protein probe to measure intracellular oxidant stress (24, 25). Using roGFP is advantageous over the use of other measures of ROS because it can be calibrated to permit comparisons between experiments. Furthermore, it is a protein and, therefore, a relevant biological target of ROS. Finally, roGFP can be targeted to subcellular compartments, like the mitochondrial matrix, by the addition of a localization tag. Our data demonstrate that hyperoxia up-regulates mitochondrial matrix ROS generation. Notably, neutralizing mitochondrial matrix ROS is insufficient to prevent hyperoxia-induced senescence. Interestingly, hyperoxia did not produce a significant increase in cytosolic ROS generation. This suggests that the mitochondrial matrix ROS generated during hyperoxia do not escape into the cytosol, where ROS-dependent damage or signaling pathways could induce senescence.

Our finding that neutralizing ROS does not protect against hyperoxia-induced senescence is contrary to previous models of stress-induced senescence, which implicate mitochondrial ROS generation in senescence. However, one recent study also reported the onset of premature senescence occurring independently of ROS generation (43); thus, ROS-independent premature senescence may not be as unusual as previously thought. Furthermore, our current data complement our previous finding that hypoxia extends replicative life span despite an up-regulation in mitochondrial ROS production (14). In that system, the mitochondrial ROS were generated not in the matrix, but into the intermembrane space by complex III (14). Thus, unlike hyperoxia-induced ROS, hypoxia-induced mitochondrial ROS can escape easily into the cytosol. However, we showed that the levels of ROS that escape into the cytosol in hypoxia are below the threshold to induce damage and activate senescence pathways. In fact, under hypoxia, the ROS work as signaling molecules to activate HIF-dependent hTERT expression to extend replicative life span. In our current study on hyperoxia, ROS levels increase in the mitochondrial matrix, but not in the cytosol. Thus the levels and localization of ROS are likely to determine whether the ROS serve as adaptive signaling molecules or damaging, senescence-inducing molecules.

A possibility remains that ROS are, indeed, involved in hyperoxia-induced senescence. Our study measured ROS levels only after 18 h, at which point mitochondrial ROS were effectively neutralized by ROS scavengers expressed via adenovirus. However, it is possible that initial exposure to hyperoxia produces a burst of ROS less than 18 h after exposure and that a burst is too great to be completely neutralized by scavengers. These initial high levels of ROS may be sufficient to damage cellular components, particularly mitochondrial DNA, and the cell then responds by inducing senescence in order to avoid propagating a mutated nuclear or mitochondrial genome. For example, pulmonary artery endothelial cells respond to ROS generated by xanthine oxidase with apoptosis resulting from damage to the mitochondrial DNA (44). Alternatively, the roGFP probe may not be as sensitive to ROS levels as the physiological target of ROS, and mitochondrial DNA, lacking protective histones or repair mechanisms, may be such a sensitive target. An intriguing question is the role of mitochondrial DNA in hyperoxia-induced senescence. Mitochondrial DNA is proximal to ROS generated in the mitochondrial matrix. It also has reduced protective and proofreading mechanisms compared to nuclear DNA, and, in fact, sustains greater oxidative damage (45). Cells may engage senescence as an attempt to avoid propagating damaged mitochondrial DNA, or as a response to the metabolic crisis that results when components of the electron transport chain encoded by the mitochondrial DNA sustain mutations from oxidative lesions. Cells are known to engage apoptotic machinery in response to mitochondrial DNA damage (44, 46), but the link to senescence remains less certain. Death from mitochondrial DNA injury can be prevented by targeting 8-oxoguanine DNA glycosylase, an enzyme that repairs oxidative damage, to the mitochondria (46). It would be interesting to determine the effects of enhanced mitochondrial DNA repair on hyperoxia-induced senescence.

Hyperoxia-induced senescence and AMPK

AMPK would appear to be an attractive potential regulator of hyperoxia-induced senescence, as it is activated by various stresses and can signal to p53 and pRb to induce senescence. Indeed, AMPK activation by AICAR led to senescence in our PHLFs, and hyperoxia led to rapid and sustained AMPK activity. However, suppressing AMPK by expressing a KD-LKB1 mutant, while effective at suppressing AMPK activation, did not prevent senescence after hyperoxia. It is possible that AMPK acts coordinately with another pathway to induce senescence under hyperoxia exposure; however, that pathway remains unidentified.

Involvement of telomerase

We demonstrate that hyperoxia-induced senescence is independent of telomere involvement, and thus likely constitutes a premature senescence-type pathway rather than accelerated replicative senescence. It is possible that repeated but brief exposure to 70% O2 could induce senescence via telomere shortening, as previous studies have shown using lower levels of hyperoxia that resulted in telomere-dependent senescence (11, 20, 21). However, the vast majority of stress-induced senescence cases occur independently of telomere involvement. It is even possible that differential exposure to the same stimulus may result in both telomere-dependent replicative senescence and telomere-independent premature senescence, as is seen with exposure to angiotensin II (47). Thus, the finding that 70% O2 can induce senescence without telomere involvement is very possible despite prior data suggesting otherwise.

Involvement of p53 and pRb pathways

Cellular senescence is ultimately regulated by the p53 and/or p16/pRb pathways. We found that the p16/pRb pathway is not by itself required for hyperoxia-induced senescence, while p53 inactivation suppresses hyperoxia-induced senescence incompletely. However, the inactivation of both the p53 and pRb pathways leads to almost complete prevention of hyperoxia-induced senescence. Thus, hyperoxia-induced senescence proceeds through the combined action of two well-established senescence pathways. These data are consistent with other studies showing that p53 and p16 pathways can interact to induce senescence.

Overexpressing HPV type 16 proteins E6 and E7 leads to an almost complete prevention of senescence after hyperoxia exposure. However, E6 and E7 interact with and affect the function of proteins other than just p53 and pRb, respectively, which may also have an affect on senescence. To verify that p53 and pRb were specifically involved, we expressed WT E7, or E7 deficient only in pRb binding, in a p53-DN background. Using a dominant negative p53 allowed us to specifically target p53 rather than all the targets affected by E6. Furthermore, using E7 deficient only in pRb binding allowed us to examine the effect of E7 on pRb in hyperoxia-induced senescence. Only the WT E7 was able to suppress pRb levels and senescence further in p53-DN PHLFs, indicating that hyperoxia-induced senescence is driven specifically by p53 and pRB.

It is generally thought that p53 activation signals to senescence, while p16 activation maintains cells in the senesced state. This may be the case in our system, where p16 activation is apparent, yet p16 inactivation alone is insufficient to prevent senescence. Interestingly, however, p53 may be acting independently of p21 to induce senescence, as simply knocking down p21 expression does not prevent hyperoxia-induced senescence even in combination with E7, while using dominant negative p53 does. It may simply be that the knockdown of p21 was insufficient and that only a small amount of p21 is enough to drive senescence, while DN-p53 resulted in more efficient p21 suppression. However, p53-induced senescence can indeed occur independently of p21, through, for instance, DEC1 (41). It remains to be determined how p53 and/or pRb pathways are being initiated during hyperoxia.

In summary, our data indicate that hyperoxia utilizes p53 and pRB pathways to execute senescence independently of an increase in ROS or an increase in AMPK. Our data therefore suggest that oxidative stress is not always causal in the induction of senescence. While not involved in senescence, hyperoxia-induced ROS may still activate genes to counteract the stress of high oxygen. However, during hyperoxia exposure, other mechanisms are also invoked, which ultimately lead to p53/pRB-induced senescence. Nevertheless, our current data on hyperoxia, along with our previously published data on hypoxia, indicate that the relationship between increases in ROS and senescence are not correlative.

Acknowledgments

We thank Dr. Laimonis A. Laimins (Northwestern University) for his kind gift of packaging cells expressing HPV-E7 WT and mutant, and HPV-E6 and E7 in combination. This work was supported in part by U.S. National Institutes of Health grants to N.S.C. (GM60472-09, P01HL071643-03004, R21AG027093). T.A.K. is supported by an American Heart Association predoctoral grant (0715708Z).

References

  1. Hayflick L. The limited in vitro lifetime of human diploid cell strains. Exp Cell Res. 1965;37:614–636. doi: 10.1016/0014-4827(65)90211-9. [DOI] [PubMed] [Google Scholar]
  2. De Lange T. Protection of mammalian telomeres. Oncogene. 2002;21:532–540. doi: 10.1038/sj.onc.1205080. [DOI] [PubMed] [Google Scholar]
  3. Di Leonardo A, Linke S P, Clarkin K, Wahl G M. DNA damage triggers a prolonged p53-dependent G1 arrest and long-term induction of Cip1 in normal human fibroblasts. Genes Dev. 1994;8:2540–2551. doi: 10.1101/gad.8.21.2540. [DOI] [PubMed] [Google Scholar]
  4. Beausejour C M, Krtolica A, Galimi F, Narita M, Lowe S W, Yaswen P, Campisi J. Reversal of human cellular senescence: roles of the p53 and p16 pathways. EMBO J. 2003;22:4212–4222. doi: 10.1093/emboj/cdg417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Alcorta D A, Xiong Y, Phelps D, Hannon G, Beach D, Barrett J C. Involvement of the cyclin-dependent kinase inhibitor p16 (INK4a) in replicative senescence of normal human fibroblasts. Proc Natl Acad Sci U S A. 1996;93:13742–13747. doi: 10.1073/pnas.93.24.13742. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Hara E, Smith R, Parry D, Tahara H, Stone S, Peters G. Regulation of p16CDKN2 expression and its implications for cell immortalization and senescence. Mol Cell Biol. 1996;16:859–867. doi: 10.1128/mcb.16.3.859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Stein G H, Drullinger L F, Soulard A, Dulic V. Differential roles for cyclin-dependent kinase inhibitors p21 and p16 in the mechanisms of senescence and differentiation in human fibroblasts. Mol Cell Biol. 1999;19:2109–2117. doi: 10.1128/mcb.19.3.2109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chen Q, Ames B N. Senescence-like growth arrest induced by hydrogen peroxide in human diploid fibroblast F65 cells. Proc Natl Acad Sci U S A. 1994;91:4130–4134. doi: 10.1073/pnas.91.10.4130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen Q, Fischer A, Reagan J D, Yan L J, Ames B N. Oxidative DNA damage and senescence of human diploid fibroblast cells. Proc Natl Acad Sci U S A. 1995;92:4337–4341. doi: 10.1073/pnas.92.10.4337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fridovich I. Oxygen toxicity: a radical explanation. J Exp Biol. 1998;201:1203–1209. doi: 10.1242/jeb.201.8.1203. [DOI] [PubMed] [Google Scholar]
  11. Saretzki G, Feng J, von Zglinicki T, Villeponteau B. Similar gene expression pattern in senescent and hyperoxic-treated fibroblasts. J Gerontol A Biol Sci Med Sci. 1998;53:B438–B442. doi: 10.1093/gerona/53a.6.b438. [DOI] [PubMed] [Google Scholar]
  12. Packer L, Fuehr K. Low oxygen concentration extends the lifespan of cultured human diploid cells. Nature. 1977;267:423–425. doi: 10.1038/267423a0. [DOI] [PubMed] [Google Scholar]
  13. Saito H, Hammond A T, Moses R E. The effect of low oxygen tension on the in vitro-replicative life span of human diploid fibroblast cells and their transformed derivatives. Exp Cell Res. 1995;217:272–279. doi: 10.1006/excr.1995.1087. [DOI] [PubMed] [Google Scholar]
  14. Bell E L, Klimova T A, Eisenbart J, Schumacker P T, Chandel N S. Mitochondrial reactive oxygen species trigger hypoxia-inducible factor-dependent extension of the replicative life span during hypoxia. Mol Cell Biol. 2007;27:5737–5745. doi: 10.1128/MCB.02265-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Li C H, Tzeng S L, Cheng Y W, Kang J J. Chloramphenicol-induced mitochondrial stress increases p21 expression and prevents cell apoptosis through a p21-dependent pathway. J Biol Chem. 2005;280:26193–26199. doi: 10.1074/jbc.M501371200. [DOI] [PubMed] [Google Scholar]
  16. Bond J, Jones C, Haughton M, DeMicco C, Kipling D, Wynford-Thomas D. Direct evidence from siRNA-directed “knock down” that p16(INK4a) is required for human fibroblast senescence and for limiting ras-induced epithelial cell proliferation. Exp Cell Res. 2004;292:151–156. doi: 10.1016/j.yexcr.2003.09.021. [DOI] [PubMed] [Google Scholar]
  17. Dimri G P, Lee X, Basile G, Acosta M, Scott G, Roskelley C, Medrano E E, Linskens M, Rubelj I, Pereira-Smith O, Peacocke M, Campisi J. A biomarker that identifies senescent human cells in culture and in aging skin in vivo. Proc Natl Acad Sci U S A. 1995;92:9363–9367. doi: 10.1073/pnas.92.20.9363. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Budinger G R, Tso M, McClintock D S, Dean D A, Sznajder J I, Chandel N S. Hyperoxia-induced apoptosis does not require mitochondrial reactive oxygen species and is regulated by Bcl-2 proteins. J Biol Chem. 2002;277:15654–15660. doi: 10.1074/jbc.M109317200. [DOI] [PubMed] [Google Scholar]
  19. Sasaki M, Kumazaki T, Takano H, Nishiyama M, Mitsui Y. Senescent cells are resistant to death despite low Bcl-2 level. Mech Ageing Dev. 2001;122:1695–1706. doi: 10.1016/s0047-6374(01)00281-0. [DOI] [PubMed] [Google Scholar]
  20. Saretzki G, Murphy M P, von Zglinicki T. MitoQ counteracts telomere shortening and elongates lifespan of fibroblasts under mild oxidative stress. Aging Cell. 2003;2:141–143. doi: 10.1046/j.1474-9728.2003.00040.x. [DOI] [PubMed] [Google Scholar]
  21. Von Zglinicki T, Saretzki G, Docke W, Lotze C. Mild hyperoxia shortens telomeres and inhibits proliferation of fibroblasts: a model for senescence? Exp Cell Res. 1995;220:186–193. doi: 10.1006/excr.1995.1305. [DOI] [PubMed] [Google Scholar]
  22. Bodnar A G, Ouellette M, Frolkis M, Holt S E, Chiu C P, Morin G B, Harley C B, Shay J W, Lichtsteiner S, Wright W E. Extension of life-span by introduction of telomerase into normal human cells. Science. 1998;279:349–352. doi: 10.1126/science.279.5349.349. [DOI] [PubMed] [Google Scholar]
  23. Vaziri H, Benchimol S. Reconstitution of telomerase activity in normal human cells leads to elongation of telomeres and extended replicative life span. Curr Biol. 1998;8:279–282. doi: 10.1016/s0960-9822(98)70109-5. [DOI] [PubMed] [Google Scholar]
  24. Hanson G T, Aggeler R, Oglesbee D, Cannon M, Capaldi R A, Tsien R Y, Remington S J. Investigating mitochondrial redox potential with redox-sensitive green fluorescent protein indicators. J Biol Chem. 2004;279:13044–13053. doi: 10.1074/jbc.M312846200. [DOI] [PubMed] [Google Scholar]
  25. Dooley C T, Dore T M, Hanson G T, Jackson W C, Remington S J, Tsien R Y. Imaging dynamic redox changes in mammalian cells with green fluorescent protein indicators. J Biol Chem. 2004;279:22284–22293. doi: 10.1074/jbc.M312847200. [DOI] [PubMed] [Google Scholar]
  26. Kimura R E, Thulin G E, Wender D, Warshaw J B. Decreased oxidative metabolism in neonatal rat lung exposed to hyperoxia. J Appl Physiol. 1983;55:1501–1505. doi: 10.1152/jappl.1983.55.5.1501. [DOI] [PubMed] [Google Scholar]
  27. Schoonen W G, Wanamarta A H, van der Klei-van Moorsel J M, Jakobs C, Joenje H. Hyperoxia-induced clonogenic killing of HeLa cells associated with respiratory failure and selective inactivation of Krebs cycle enzymes. Mutat Res. 1990;237:173–181. doi: 10.1016/0921-8734(90)90023-k. [DOI] [PubMed] [Google Scholar]
  28. Jones R G, Plas D R, Kubek S, Buzzai M, Mu J, Xu Y, Birnbaum M J, Thompson C B. AMP-activated protein kinase induces a p53-dependent metabolic checkpoint. Mol Cell. 2005;18:283–293. doi: 10.1016/j.molcel.2005.03.027. [DOI] [PubMed] [Google Scholar]
  29. Wang W, Yang X, Lopez de Silanes I, Carling D, Gorospe M. Increased AMP:ATP ratio and AMP-activated protein kinase activity during cellular senescence linked to reduced HuR function. J Biol Chem. 2003;278:27016–27023. doi: 10.1074/jbc.M300318200. [DOI] [PubMed] [Google Scholar]
  30. Gurumurthy S, Hezel A F, Sahin E, Berger J H, Bosenberg M W, Bardeesy N. LKB1 deficiency sensitizes mice to carcinogen-induced tumorigenesis. Cancer Res. 2008;68:55–63. doi: 10.1158/0008-5472.CAN-07-3225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Serrano M, Lin A W, McCurrach M E, Beach D, Lowe S W. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell. 1997;88:593–602. doi: 10.1016/s0092-8674(00)81902-9. [DOI] [PubMed] [Google Scholar]
  32. Xu H J, Zhou Y, Ji W, Perng G S, Kruzelock R, Kong C T, Bast R C, Mills G B, Li J, Hu S X. Reexpression of the retinoblastoma protein in tumor cells induces senescence and telomerase inhibition. Oncogene. 1997;15:2589–2596. doi: 10.1038/sj.onc.1201446. [DOI] [PubMed] [Google Scholar]
  33. Jacobs J J, Kieboom K, Marino S, DePinho R A, van Lohuizen M. The oncogene and Polycomb-group gene bmi-1 regulates cell proliferation and senescence through the ink4a locus. Nature. 1999;397:164–168. doi: 10.1038/16476. [DOI] [PubMed] [Google Scholar]
  34. Gil J, Bernard D, Martinez D, Beach D. Polycomb CBX7 has a unifying role in cellular lifespan. Nat Cell Biol. 2004;6:67–72. doi: 10.1038/ncb1077. [DOI] [PubMed] [Google Scholar]
  35. Jacobs J J, Scheijen B, Voncken J W, Kieboom K, Berns A, van Lohuizen M. Bmi-1 collaborates with c-Myc in tumorigenesis by inhibiting c-Myc-induced apoptosis via INK4a/ARF. Genes Dev. 1999;13:2678–2690. doi: 10.1101/gad.13.20.2678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Liu L, Andrews L G, Tollefsbol T O. Loss of the human polycomb group protein BMI1 promotes cancer-specific cell death. Oncogene. 2006;25:4370–4375. doi: 10.1038/sj.onc.1209454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sasaki M, Ikeda H, Sato Y, Nakanuma Y. Decreased expression of Bmi1 is closely associated with cellular senescence in small bile ducts in primary biliary cirrhosis. Am J Pathol. 2006;169:831–845. doi: 10.2353/ajpath.2006.051237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Itahana K, Zou Y, Itahana Y, Martinez J L, Beausejour C, Jacobs J J, Van Lohuizen M, Band V, Campisi J, Dimri G P. Control of the replicative life span of human fibroblasts by p16 and the polycomb protein Bmi-1. Mol Cell Biol. 2003;23:389–401. doi: 10.1128/MCB.23.1.389-401.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Guney I, Wu S, Sedivy J M. Reduced c-Myc signaling triggers telomere-independent senescence by regulating Bmi-1 and p16(INK4a) Proc Natl Acad Sci U S A. 2006;103:3645–3650. doi: 10.1073/pnas.0600069103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Boyer S N, Wazer D E, Band V. E7 protein of human papilloma virus-16 induces degradation of retinoblastoma protein through the ubiquitin-proteasome pathway. Cancer Res. 1996;56:4620–4624. [PubMed] [Google Scholar]
  41. Qian Y, Zhang J, Yan B, Chen X. DEC1, a basic helix-loop-helix transcription factor and a novel target gene of the p53 family, mediates p53-dependent premature senescence. J Biol Chem. 2008;283:2896–2905. doi: 10.1074/jbc.M708624200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Scheffner M, Huibregtse J M, Vierstra R D, Howley P M. The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of p53. Cell. 1993;75:495–505. doi: 10.1016/0092-8674(93)90384-3. [DOI] [PubMed] [Google Scholar]
  43. Stockl P, Zankl C, Hutter E, Unterluggauer H, Laun P, Heeren G, Bogengruber E, Herndler-Brandstetter D, Breitenbach M, Jansen-Durr P. Partial uncoupling of oxidative phosphorylation induces premature senescence in human fibroblasts and yeast mother cells. Free Radic Biol Med. 2007;43:947–958. doi: 10.1016/j.freeradbiomed.2007.06.005. [DOI] [PubMed] [Google Scholar]
  44. Ruchko M, Gorodnya O, LeDoux S P, Alexeyev M F, Al-Mehdi A B, Gillespie M N. Mitochondrial DNA damage triggers mitochondrial dysfunction and apoptosis in oxidant-challenged lung endothelial cells. Am J Physiol Lung Cell Mol Physiol. 2005;288:L530–L535. doi: 10.1152/ajplung.00255.2004. [DOI] [PubMed] [Google Scholar]
  45. Balaban R S, Nemoto S, Finkel T. Mitochondria, oxidants, and aging. Cell. 2005;120:483–495. doi: 10.1016/j.cell.2005.02.001. [DOI] [PubMed] [Google Scholar]
  46. Dobson A W, Grishko V, LeDoux S P, Kelley M R, Wilson G L, Gillespie M N. Enhanced mtDNA repair capacity protects pulmonary artery endothelial cells from oxidant-mediated death. Am J Physiol Lung Cell Mol Physiol. 2002;283:L205–L210. doi: 10.1152/ajplung.00443.2001. [DOI] [PubMed] [Google Scholar]
  47. Herbert K E, Mistry Y, Hastings R, Poolman T, Niklason L, Williams B. Angiotensin II-mediated oxidative DNA damage accelerates cellular senescence in cultured human vascular smooth muscle cells via telomere-dependent and independent pathways. Circ Res. 2008;102:201–208. doi: 10.1161/CIRCRESAHA.107.158626. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The FASEB Journal are provided here courtesy of The Federation of American Societies for Experimental Biology

RESOURCES