Irreversible cellular senescence induced by ... - Semantic Scholar

The International Journal of Biochemistry & Cell Biology 37 (2005) 1407–1420

Irreversible cellular senescence induced by prolonged exposure to H2O2 involves DNA-damage-and-repair genes and telomere shortening Jianming Duan a, b, 1 , Jianping Duan a, c , Zongyu Zhang a, ∗ , Tanjun Tong a, ∗ b

a Department of Biochemistry and Molecular Biology, Peking University Health Science Center, Beijing 100083, China Tumor Biology Section, Head and Neck Surgery Branch, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Bldg. 10 Room 5D51, 10 Central Drive, Bethesda, MD 20892, USA c Department of Internal Medicine, Shiji People’s Hospital of Guangzhou Panyu, Guangzhou 511450, China

Received 23 June 2004; accepted 18 January 2005

Abstract H2 O2 has been the most commonly used inducer for stress-induced premature senescence (SIPS), which shares features of replicative senescence. However, there is still uncertainty whether SIPS and replicative senescence differ or utilize different pathways. ‘Young’ human diploid fibroblasts (HDFs), treated with prolonged low doses of hydrogen peroxide, led to irreversible cellular senescence. Cells exhibited senescent-morphological features, irreversible G1 cell cycle arrest and irreversible senescence-associated ␤-galactosidase positivity. The appearance of these cellular senescence markers was accompanied by significant increases of p21, gadd45 expression and p53 binding activity, as well as a significant decline in DNA repair capability and accelerated telomere shortening. Our results suggest that multiple pathways might be involved in oxidative SIPS, including genes related to DNA-damage-and-repair and telomere shortening, and that SIPS shares the same mechanisms with replicative senescence in vivo. Our findings indicate that several aging theories can be merged together by a common mechanism of oxidative damage, and that the level of oxidative DNA-damage-and-repair capacity may be exploited as reliable markers of cell senescence. © 2005 Elsevier Ltd. All rights reserved. Keywords: Cellular senescence; Oxidative stress; Hydrogen peroxide; DNA damage; DNA repair; Telomere; Human fibroblast

∗ 1

Abbreviations: SIPS, stress-induced premature senescence; HDF, human diploid fibroblast; PD, population doublings; SA-␤-gal, senescenceassociated ␤-galactosidase; SCGE, single cell gel electrophoresis; MV-MDD, mean value of migration distance of DNA; UDS, unscheduled DNA synthesis; EMSA, electrophoretic mobility shift assay; TRF, terminal restriction fragment Corresponding authors. Tel.: +86 10 8280 1454; fax: +86 10 8280 2931. E-mail address: [email protected] (T. Tong). Tel.: +1 301 435 5004; fax: +1 301 402 1140.

1357-2725/$ – see front matter © 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2005.01.010

1408

J. Duan et al. / The International Journal of Biochemistry & Cell Biology 37 (2005) 1407–1420

1. Introduction Human diploid fibroblasts (HDFs) exhibit finite proliferative potential in vitro, the so-called Hayflick limit (Hayflick & Moorhead, 1961). They undergo a limited number of population doublings (PD) before entering a state of permanent growth arrest, referred to as “replicative senescence,” “cellular senescence” or “cellular aging” (Campisi, 1996; Campisi, Kim, Lim, & Rubio, 2001), in which they remain alive and metabolically active but are completely refractory to mitogenic stimuli. HDFs offer the typical model for studying the process of aging in vitro. Various oxidative stresses have been used to study the onset of cellular senescence, such as exposure of cells to ultraviolet (UV) light (Rodemann, Bayreuther, Francz, Dittmann, & Albiez, 1989), hyperoxia (Honda, Hjelmeland, & Handa, 2001), tertbutylhydroperoxide (t-BHP) (Toussaint, Houbion, & Remacle, 1992), and H2 O2 (Frippiat et al., 2001; Frippiat, Dewelle, Remacle, & Toussaint, 2002). The early onset of cellular senescence induced by these stresses is termed stress-induced premature senescence (SIPS) (Toussaint, Medrano, & von Zglinicki, 2000). H2 O2 has been the most commonly used inducer of SIPS, which shares features of replicative senescence: similar morphology, senescence-associated ␤galactosidase activity, cell cycle regulation, etc. (Chen & Ames, 1994; Dimri et al., 1995; Frippiat et al., 2001, 2002). However, what triggers SIPS and whether it shares the same mechanisms and pathways with replicative senescence are still not well understood. DNA damage is by far the most detrimental consequence of oxidative stress. It causes a series of irreversible dysfunctions and eventually cell death (Kawanishi, Hiraku, & Oikawa, 2001). Through a common p53-dependent mechanism, DNA damaged cells may undergo death, usually by apoptosis, or growth arrest (Itahana, Dimri, & Campisi, 2001); but alternatively, cells can also repair damaged DNA by complex enzymatic mechanisms like base excision repair, nucleotide excision repair, mismatch repair, etc. (de Boer & Hoeijmakers, 2000). Cellular senescence may occur if DNA damage is not serious enough to induce cell death but DNA damage cannot be completely repaired (Beckman & Ames, 1998). In fact, it was found that the levels of oxidative DNA damage are significantly increased in senescent cells, especially in post-mitotic tissues (Hamilton et al., 2001), and that

the levels are correlated with aging and age-related diseases (Bayens, 2002; Mattson, 2003). Meanwhile, the DNA repair capacity related to oxidative DNA damage significantly declined in senescent cells (Parrinello et al., 2003). Therefore, the DNA repair capacity and the accumulation of DNA damage play important roles in the initiation and the process of replicative senescence. However, controversy exists as to whether the DNA damage observed in senescent cells reflects prolonged oxidative attack (Hall et al., 2001), a decline in repair mechanisms (Klungland et al., 1999; von Zglinicki, Burkle, & Kirkwood, 2001), or a combination of these factors. Also, arguments exist regarding whether oxidative SIPS shares the same mechanisms and pathways with replicative senescence. Some research has shown that senescent phenotypes induced by sub-lethal doses of H2 O2 are uncoupled from telomere shortening (Chen, Prowse, Tu, Purdom, & Linskens, 2001) which is another theory of aging. To elucidate the underlying mechanism of oxidative SIPS, i.e., what triggers it, how similar it is to replicative senescence, and whether the oxidative damage and DNA repair capacity could be considered biomarkers of aging, we designed experiments to induce SIPS by H2 O2 treatment and observed the pathways involved, and compared these with replicative senescence. We treated ‘young’ human diploid fibroblasts (2BS) with prolonged low doses of hydrogen peroxide instead of using acute treatment by sub-lethal doses of H2 O2 in order to mimic chronic oxidative stress under pathophysiological conditions. We then evaluated oxidative DNA damage and DNA repair capacity after the oxidative damage in 2BS cells using the single cell gel electrophoresis (SCGE) assay and the unscheduled DNA synthesis (UDS) assay, and then correlated the levels of these two markers to cellular senescence monitored by proliferation rate and senescenceassociated-␤-galactosidase (SA-␤-galactosidase) expression. We also examined the role of DNA damage genes and telomere shortening to explore the relationship between DNA-damage-and-repair and telomere shortening during oxidative damage induced SIPS. 2. Materials and methods 2.1. Cell culture and treatment protocols Human diploid fibroblasts 2BS, derived from embryonic human lung primary culture (obtained from


the National Institute of Biological Products, Beijing, China), were grown in Dulbecco’s modified Eagle’s medium (DMEM) (GIBCO Life Technologies Inc.) supplemented with 10% heat-inactivated fetal bovine serum, 100 ␮g/ml penicillin and 100 ␮g/ml streptomycin (GIBCO) at 37 ◦ C in 5% CO2 /air incubator (Duan, Zhang, & Tong, 2001; Li, Zhang, & Tong, 1995). The early passage cells with population doublings (PD) 26 ± 2 were used as ‘young’ cells. Cells at PD 56 ± 4 were virtually growth arrested and used as senescent (‘old’) cells. Cells were seeded at 1 × 104 cells/ml in 24-well plates (Becton–Dickinson, Franklin Lakes, NJ), treated with H2 O2 (Sigma–Aldrich) at the concentrations as indicated, and incubated at 37 ◦ C until the collection time points. For the acute treatment, sub-lethal amounts of H2 O2 (100 ␮M) were added at specific time intervals. At the end of incubations, media was washed out by PBS and the cells were collected. Prolonged H2 O2 treatment was performed by adding 10 ␮M H2 O2 with medium change every 3 days; then 3 days of recovery after treatment was allowed before conducting any further experiments. Cells examined after treatment with H2 O2 were compared to parallel cultured control cells grown in the media without H2 O2 . 2.2. Cell morphology and proliferation assays Cell morphology changes with H2 O2 treatment were examined by microscopy at 10×, 40× and 100× magnification. For cell proliferation assays, 1 × 104 of 2BS cells were inoculated in 6-well plates in 6 replicates for each time point. The cells were harvested after trypsinization and the numbers of cells were counted under the microscope using a hemocytometer at 24 h intervals for 16 days. Data were presented as the mean ± standard deviation from the six replicates, and the statistical significance was calculated by Student’s t-test. 2.3. Cell cycle analysis Cultured cells were harvested at desired time points after trypsinization, fixed and permeablized with 70% ice-cold ethanol for 12 h. The cells were treated with 100 ␮g/ml RNase A (Sigma, St. Louis, MO) at 37 ◦ C for 30 min. DNA content was determined by staining with propidium iodide (Sigma) at l ␮g/ml with 0.1%

1409

Triton for 30 min, and analyzed by FACScan (fluorescence activated cell sorting, Becton–Dickinson, San Jose, CA). Cell cycle distribution was analyzed with CellFit software (Becton–Dickinson, San Jose, CA). 2.4. SA-β-gal staining SA-␤-gal staining has been widely used as a biomarker of cellular senescence in vivo and in vitro (Dimri et al., 1995). The positive blue colored staining of ␤-galactosidase at pH 6.0 has been reported to remarkably increase in senescent cells. To detect SA␤-gal staining, cells were washed twice in PBS, fixed for 3–5 min at room temperature in 3% formaldehyde, and washed with PBS again. Then cells were incubated overnight at 37 ◦ C (without CO2 ) with freshly prepared SA-␤-gal stain solution (1 mg/ml X-gal, 40 mM citric acid/sodium phosphate, pH 6.0, 5 mM potassium ferrocyanide, 5 mM potassium ferricyanide, 150 mM NaCl, 2 mM MgCl2 ). Positive staining appeared after 2–4 h and was evaluated after 12–16 h incubation at 37 ◦ C in a CO2 -free atmosphere. The blue stained cells were counted in 10 fields (5 × 105 cells) under the microscope with 100× magnification and expressed as a percentage of positive cells. To avoid staining due to cell confluence rather than to proliferative senescence (Severino, Allen, Balin, Balin, & Cristofalo, 2000), the assay was performed in sub-confluent cultures displaying comparable cell density. 2.5. Single cell gel electrophoresis assays SCGE assay, also called comet assay, was performed by the method of Singh, McCoy, Tice, and Schneider (1988) to detect DNA strand breaks. Slides were prepared with 85 ␮l of 0.5% normal melting agarose. The cells were then mixed with 75 ␮l of 0.7% low melting agarose and applied to the prepared slides. Alkaline lysis (10% DMSO, 1% Triton-X in alkaline lysis buffer: 2.5 M NaCl, 10 mM Tris, 100 mM Na2 EDTA, pH 10) followed for 1 h. The slides were placed in a horizontal gel electrophoresis chamber containing alkaline buffer solution with NaOH (10 mM) and Na2 EDTA (200 mM) at pH 13.2. After a 20-min DNA “unwinding” period, electrophoresis was performed at 25 V and 300 mA for 20 min. Following neutralization (Tris–HCl, pH 7.5), the cells were stained with ethidium bromide. The slides were examined and the tail length (diameter of

1410


the nucleus plus migrated DNA), i.e., migration distance of DNA (MDD), was measured with a fluorescence microscope (Nikon). The DNA migration of 60 randomly selected cells was examined for each sample. 2.6. Unscheduled DNA synthesis assays 105 Cells were inoculated into each of 35 mm cell culture dish in six replicates (Corning Costar). In the next day, cells were washed once with serum-free medium and incubated in medium containing 0.5% serum for 3 days to reach quiescence. Cells were washed, then incubated with serum-free medium containing 5 mM hydroxyurea (HU, Sigma) for 1 h, and exposed to 200 ␮M H2 O2 and 5 mM HU in serumfree medium for 1 h at 37 ◦ C. H2 O2 -treated cells were washed once with serum-free medium containing 5 mM HU and then incubated in serum-free medium containing 5 mM HU and 1.0 mCi/ml [3 H]thymidine for 10 h. The incorporation of [3 H]thymidine was determined by liquid scintillation spectroscopy (Beckman Instrument Inc., LS ANALYZER) (Pero et al., 1990). 2.7. Telomere length analysis Blots containing genomic DNA were prepared as previously described for Southern blot analyses, and hybridized with 5 end-labeled telomeric oligonucleotide probe [␥-32 P-(TTAGGG)4]. After washing, autoradiographs were obtained from the blots by exposing Kodak XAR-5 X-ray film for 12–24 h at room temperature. To calculate the TRF (terminal restriction fragment), the images were scanned with a densitometer and the data were analyzed as described by Harley, Futcher, and Greider (1990). The mean TRF length was defined as: (ODi )/(Odi /Li ), where ODi is the densitometer output and Li the length of the DNA at position i. The amount of telomeric DNA was calculated by integrating the volume of each smear in ImageQuant software (Molecular Dynamics, Sunnyvale, CA). 2.8. Measurement of mRNA level by Northern blotting For Northern blot, total RNA was extracted with TRIZOL reagent following the manufacturer’s protocol (Invitrogen) at desired time points after H2 O2 treatment. Total RNA (15 ␮g) was electrophoresed

on a formaldehyde-agarose denaturing gel, transferred to the nylon membrane (Hybond-N, Amersham Life Science, Buckinghamshire, England), and hybridized with [␣-32 p] dCTP-labeled human p21WAF1 (2.1 kb cDNA), GADD45 (0.6 kb SacI–KpnI cDNA fragment), p53 (1.7 kb EcoRI–PstI cDNA fragment) and β-actin (0.9 kb PstI cDNA fragment) probes. The membrane was exposed to a Kodak XAR-5 X-ray film at −80 ◦ C. 2.9. Western blot analysis Cultured cells were washed with ice-cold PBS and then were lysed with 2×SDS lysis buffer. Protein concentrations were measured by the Bradford method (Bio-Rad protein Assay; Bio-Rad Laboratories GmbH, Munich, Germany). Protein (50 ␮g) from each sample was loaded on a 12% SDS-polyacrylamide gel followed by blotting on a nitrocellulose membrane (Bio-Rad Laboratories, CA). Membranes were blocked for 1 h with 5% non-fat milk in TBS-T (0.1% Tween20 in TBS), and probed with antibodies against p21 (sc-817), GADD45 (sc-6850) and p53 (sc-263), actin (sc-8432) as the control. All antibodies were purchased from Santa Cruz Biotechnology Inc., Santa Cruz, CA. 2.10. Electrophoretic mobility shift assay (EMSA) Nuclear extracts from H2 O2 treated cells were subjected to EMSA analysis as previously described (Hupp, Meek, Midgley, & Lane, 1992) with the following modifications. Double-stranded oligonucleotides of the core sequence of p53 binding element (sense: 5 GGATCCAAGCTTGGACATGCCCGGCTCGAG 3 ) (GIBCO) were labeled with [32 P] dATP (50 ␮Ci at 3000 Ci/mmol from Dupont-New England Nuclear, Boston, MA) using Klenow fragment of E. coli DNA polymerase I (Boehringer Mannheim, Indianapolis, IN). Equal amounts (6–10 (g) of nuclear extracts were incubated in 20 (l binding buffer containing 10 mM Tris–HCl, pH7.5, 50 mM NaCl, 0.5 mM DTT, 0.5 mM EDTA, 1 mM MgCl2 , 4% glycerol, 2.5 (g of poly (dI/dC) (Pharmacia) and 2 (l of the 32 P-labeled probe for 30 min at room temperature. The DNA-protein binding complexes were analyzed by electrophoretic mobility shift assay (EMSA) on a nondenaturing 5% polyacrylamide gel using a Tris/glycine/EDTA buffer. After being dried, the gel was exposed to film at −70 ◦ C.


1411

2.11. Statistical analysis All experiments were performed at least three times. Data were analyzed by Student’s t-test and one-way analysis of variance (ANOVA) followed by post-hoc multiple comparison tests. Significance was accepted at p < 0.05.

3. Results 3.1. Prolonged treatment with low doses of H2 O2 induces senescent phenotypes As determined by Trypan Blue uptake, H2 O2 above 500 ␮M induces apoptosis and cell death (data not shown). With acute sub-lethal doses of H2 O2 (100–200 ␮M) treatment (1–2 h), cells later exhibit SIPS, with morphology, growth and cell cycle arrest similar to replicative senescence, senescence-associated ␤-galactosidase activity, and gene expression (Dimri et al., 1995; Frippiat et al., 2002). However, to maximally mimic oxidative stress under pathophysiological conditions we treated ‘young’ 2BS cells (PD < 26) with prolonged low doses of H2 O2 (10 ␮M) instead of using acute treatment by sub-lethal doses of H2 O2 . The replication of ‘young’ 2BS cells were found irreversibly arrested after 2 weeks of prolonged low doses of H2 O2 (10 ␮M) treatment. Replication did not recover and the replication curve was consistent with that of senescent cells (Fig. 1). The cells were irreversibly arrested in G1 phase like senescent cells (Fig. 2). There were no significant changes of cell morphology with only 1

Fig. 1. Prolonged low-dose H2 O2 treatment induces irreversible replication arrest. Growth curves were determined by counting cell numbers at indicated time points. 3.0 × 104 cells per well were plated into 6-well plates. At the indicated time points, cells were washed with PBS, harvested by trypsinization, and counted every 24 h for 16 ¯ ± S.E.; days. Each experiment was performed at least twice. Value, X n = 6.

day of treatment (Fig. 3), but cells began to develop the senescent-morphology after 3–5 days of treatment (data not shown). Two weeks later, the morphology of the treated cells resembled that of senescent cells: with gross enlargement, flattening, and accumulation of granular cytoplasmic inclusions (Fig. 3). The specific senescence-associated marker, pH 6.0 optimum ␤-galactosidase (SA-␤-gal), was measured by X-gal staining. Only sporadic SA-␤-gal-positive cells were seen in untreated ‘young’ 2BS (PD26) cells, whereas the percentage of SA-␤-gal positive cells increased along with the time of H2 O2 treatment (1 day–2 weeks). The percentage of SA-␤-gal positive cells of the ‘young’ 2BS cells after 2 weeks of treatment was

Fig. 2. Prolonged low-dose H2 O2 treatment induces irreversible cell cycle arrest. Cell cycle distribution was determined by FACScan flow cytometry. (A) Untreated ‘young’ cells (PD25). (B) ‘Young’ cells treated with 10 ␮M of H2 O2 for 2 weeks and recovered for 3 days. (C) Untreated senescent cells (PD58).

1412


Fig. 3. Morphology and SA-␤-gal staining changes of 2BS cells induced by prolonged low-dose H2 O2 treatment. (A) Untreated ‘young’ cells (PD25). (B) ‘Young’ cells treated with 10 ␮M H2 O2 for 1 day. (C) ‘Young’ cells treated with 10 ␮M for 2 weeks. (D) Untreated senescent cells (PD58). Cells were microphotographed at a magnification of 100×.

more that 95% and similar to that of senescent cells (PD 58) (Fig. 3). As seen above, the prolonged low doses of H2 O2 treatment not only induced the senescent like morphologic changes but also abolished their proliferative potential and induced the expression of senescence-associated activity. These results indicate that prolonged low doses of H2 O2 treatment induced SIPS resembles replicative senescence at both the cellular and molecular level. 3.2. Accumulation of DNA damage and decline of DNA repair capacity induced by prolonged oxidative stress Oxidative DNA damage in 2BS cells caused by H2 O2 treatment and DNA repair capacity of cells were evaluated by SCGE assay. DNA strand breaks of single fibroblasts were measured by ‘comets’ (Fig. 4A and B). The un-pretreated ‘young’ cell, the ‘young’ cell pretreated with 10 ␮M H2 O2 for 2 weeks, and the un-pretreated ‘old’ cell were treated with 100 ␮M H2 O2 for 1 h and allowed to recover in H2 O2 -free media for desired time periods. The MV-MDD of samples was measured from different time points before and after 100 ␮M H2 O2 treatment. Before treatment with 100 ␮M H2 O2 , similar levels of significant DNA damage were observed in senescent cells and 2 weeks pre-

treated ‘young’ cells, but not in untreated ‘young’ cells (p < 0.05) (Fig. 4A and B). This indicates that the former two cell lines have accumulated DNA damage accumulation. Similar levels of DNA strand breaks in all three cell lines could be detected after 1 h H2 O2 treatment (Fig. 4A(b) and B). However, the un-pretreated ‘young’ cells showed a significantly faster rate of repair than the other two (Fig. 4A(c) and B); after 12 h of recovery, the MV-MDD level of un-pretreated ‘young’ cells almost returns to its original level but in the other two cell lines the level of MV-MDD remained higher than their original levels, indicating an accumulation of damage. The 2-week pre-treated ‘young’ cells showed a significant decline in repair rate for DNA strand breaks compared with un-pretreated ‘young’ cells, but similar to that of old cells (Fig. 4A and B). We further tested the total DNA repair capacity by the unscheduled DNA synthesis assay. Similarly, it was shown that the total DNA repair capacity significantly declined in pretreated ‘young’ cells after 2 weeks of H2 O2 treatment, when compared with the control cells, and their DNA repair capacity is similar to that of senescent cells (Fig. 4C). These results indicate that prolonged treatment with low doses of H2 O2 not only causes the accumulation of DNA damage, but also results in decreased repair capacity of the cell. Thus, the level of DNA damage accumulation and repair capacity may serve as markers for cellular senescence.


1413

3.3. DNA-damage-and-repair related gene changes induced by prolonged H2 O2 treatment H2 O2 caused DNA damage usually induces the p53 pathway, which plays an important role in cellular response to DNA damage by controlling DNA repair, cell cycle arrest and apoptosis (Kahlem, Dorken, & Schmitt, 2004). To gain further insight into the correlation between H2 O2 -induced DNA damage and senescence-associated growth arrest, we characterized the effects of H2 O2 treatment on the gene expression and/or DNA binding activity of p53, p21 and GADD45, which are key components of the p53-DNA damage repair pathway. We first tested the effect of acute treatment by a sub-lethal concentration of H2 O2 (100 ␮M) for 1 h, and found that H2 O2 treatment caused significant elevation of p21 and GADD45 gene expression in mRNA but only a slight elevation of p53 (Fig. 5A). The elevation began after 1 h of treatment and reached the highest level after about 6 h, then gradually reduced to a level that was still significantly higher than the original expression of untreated cells. We then further tested the long-term effect of the prolonged low doses of H2 O2 (10 ␮M) treatment on the protein products of these genes by western blotting. In contrast to the acute treatment, there was only moderate elevation of p21 and GADD45 on day 1. However, the levels of p21 and GADD45 continually and significantly increased with the prolonged treatment, which was shown on days 7 and 14, and reached similar levels as in “old” cells on day 14 (Fig. 5B). Similar to the mRNA level changes during acute treatment, the p53 protein level only showed a slight increase on day 1, and no further significant changes during the prolonged treatment.

Fig. 4. Prolonged (2 weeks) low-dose H2 O2 treatment induces the accumulation of DNA damage and the decline of DNA repair capacity. (A) SCEG assay, the comet image of the fluorescence photomicrographs showing DNA strand breaks and repair before and after treatment with 100 ␮M H2 O2 for 1 h. Rows: 1, un-pretreated ‘young’ cells; 2, 2-weeks pretreated ‘young’ cells; 3, un-pretreated senescent cells. Columns: a, untreated control cells; b, cells after 1 h 100 ␮M H2 O2 treatment; c, cells after repairing for 12 h. (B) The time course of SCEG assays before and after 100 ␮M H2 O2 treatment for 1 h. Value, X ± S.E.; n = 20. S: un-pretreated senescent cells; Y1: unpretreated ‘young’ cells; Y2: 2-weeks pretreated ‘young’ cells. (C) A comparison of DNA repair capacity measured with UDS. Value, ¯ ± S.E.; n = 6. Y1: un-pretreated ‘young’ cells; Y2: 2-weeks preX treated ‘young’ cells.

1414



1415

Fig. 6. Prolonged low-dose H2 O2 treatment induces telomere shortening. Terminal restriction fragment (TRF) assay was performed to measure the length of telomere. (A) Electrophoresis of genomic DNA digested with EcoRI. (B) Southern blot analysis using [␥-32 P](TTAGGG)4 as probe to detect TRF and measure the telomere length. Lane 1: Control ‘young’ cells (PD25). Lane 2: PD25 cells treated with 10 ␮M H2 O2 for 1 day. Lane 3: Untreated PD35 cells (control of lane5). Lane 4: PD25 cells treated with 10 ␮M H2 O2 for 7 days. Lane 5: PD25 cells treated with 10 ␮M H2 O2 for 30 days. Lane 6: Control senescent cells (PD 60). Experiments were repeated three times and produced similar results.

Considering p53 binding activity is usually more important than its mRNA and protein expression level in regulating its down-stream pathways, it is not a surprise that there is no significant change in its mRNA level and protein level by H2 O2 treatment. As a result, we investigated p53 binding activity by EMSA assay. Our data showed that the p53 binding activity of the ‘old’ 2BS cell was significantly higher than that of ‘young’ cells. Interestingly, with the prolonged H2 O2 treatment of ‘young’ cells, their p53 binding activity was gradually elevated. The p53 binding activity of 2-week H2 O2 -treated ‘young’ cells is similar with that of ‘old’ cells (Fig. 5C). These results further underline the contribution of the accumulation of DNA damage to cell senescence and suggest that the prolonged H2 O2 treatment-induced cell cycle arrest is related to the p53-dependant pathway, while the DNA repair

capacity did not increase along with the elevation of p21 level and p53 binding activity. 3.4. Prolonged low doses of H2 O2 treatment accelerates telomere shortening Telomeres are thought to serve as a ‘replicometer’ for replicative senescence (Hayflick, 1997; von Zglinicki, 2001). Not only do they trigger replicative senescence under conventional conditions, but also act as cumulative sensors for oxidative stress by accumulation of single strand breaks and accelerated shortening. Induction of premature senescence by hyperoxia is thought to result from accelerated telomere shortening (Toussaint et al., 2000; von Zglinicki, Pilger, Sitte, 2002). To test whether telomere shortening occurs during the induction of premature

Fig. 5. DNA-damage-and-repair related gene changes induced by acute and prolonged H2 O2 treatment. (A) Northern blotting analysis for GADD45, p21 and p53 gene expression after acute sub-lethal H2 O2 treatment (100 ␮M) for 1 h in ‘young’ 2BS cells. The amount of RNA loading was normalized by ␤-actin. (B) Western blotting analysis for GADD45, p21 and p53 expression after prolonged low-dose H2 O2 treatment (10 ␮M) in ‘young’ 2BS cells compared with ‘old’ cells. The amount of protein loading was normalized by actin. (C) p53 binding activity with the prolongation of exposure to low doses of H2 O2 (10 ␮M). Values for H2 O2 treated samples and senescent sample were normalized to the ‘young’ un-pretreated control. Results are expressed as mean ± S.E. of samples.

1416


senescence via prolonged low dose H2 O2 treatment, we determined telomere length by measuring the mean length of TRF in 2BS cells. Cells were harvested at different time points after H2 O2 treatment to determine whether onset of the senescent phenotype correlated with shortening of telomeres. As shown in Fig. 6, with prolonged H2 O2 treatment, significant telomere shortening occurred: 1-week treatment (7.34 kb) and 30-day treatment (6.92 kb) versus untreated 2BS (8.16kb) (p < 0.01). After 30 days of H2 O2 treatment, the TRF length of ‘young’ 2BS was similar to that of ‘old’ 2BS cells (Fig. 6) (p = 0.17). However, the acute treatment by sub-lethal doses of H2 O2 did not show significant telomere shortening (8.04 kb) (p = 0.13). Our results show that significant telomere shortening occurs as a consequence of prolonged H2 O2 treatment but not acute treatment.

4. Discussion Unlike cells that have undergone necrosis or apoptosis caused by cytotoxic doses of H2 O2 , sub-lethal H2 O2 treatment induces senescence-like growth arrest of human fibroblasts (Dimri et al., 1995; Frippiat et al., 2002). However, this type of premature senescence model is mostly produced by acute treatment with sublethal H2 O2 , and it differs from physiological cellular senescence in such ways as transient G1 arrest, increase in p53 protein, and lack of significant telomere shortening. Some researchers think that different pathways may contribute to H2 O2 induced premature senescence and physiologically replicative senescence, and that DNA damage accumulation and telomere shortening are two uncoupled pathways for the onset of cellular senescence (Chen et al., 2001). In this study, instead of using acute treatment with sub-lethal doses of H2 O2 , we treated ‘young’ human diploid fibroblasts with prolonged low doses of hydrogen peroxide to mimic prolonged oxidative stress in vivo, and successfully triggered an accelerated cellular senescence. The prolonged low doses of H2 O2 treatment not only induced irreversible cell arrest and senescentlike morphology, but also caused sustained elevation of the expression and/or activity of DNA-damage-andrepair and senescence-associated genes. Furthermore, we showed that chronic DNA damage accumulation and decline of DNA repair capacity along with the

senescent process were also accompanied by telomere shortening. These results support our hypothesis that premature senescence induced by prolonged oxidative stress shares common mechanisms with pathological aging in vivo and thus can serve as a useful senescence model in vitro. Prolonged low doses of H2 O2 treatment in our study induced senescent-like morphological changes (Fig. 3), irreversible G1 cell cycle arrest (Fig. 2) and a significant increase in senescence-associated marker (pH 6.0 optimum ␤-galactosidase (SA-␤-gal) positivity) (Fig. 3) in ‘young’ 2BS cells, similar to that seen with acute sub-lethal doses of H2 O2 -treatment induced SIPS in other studies (Dimri et al., 1995; Frippiat et al., 2002). However, it differed in that it was a gradual and irreversible process, and only high enough sub-lethal doses and long enough acute H2 O2 treatment (150 ␮M for 2 h or 75 ␮M twice for 2 h) in those studies produce similar irreversible morphologic changes and cell cycle arrest. This indicates that the onset of SIPS is a dose-dependent process and is supported by the free radical theory of aging (Harman, Holliday, & Meydani, 1998), which proposes that normal aging results from accumulation of oxidative DNA damage. Our data also support the concept of “critical threshold of error accumulation” (Toussaint et al., 1992), whereby ‘young’ 2BS cells submitted to successive oxidative stresses of H2 O2 accumulated high levels of unrepaired DNA damage similar to that of senescent cells (Fig. 4). The accumulation of DNA damage is a kinetic process and it must not only be attributed to the extent of damage but also to the efficiency of the systems for elimination and/or repair of damage (Toussaint et al., 2000), as seen in our SCEG study. Besides the accumulation of DNA damage itself, prolonged H2 O2 also apparently wore down the repair capacity of ‘young’ 2BS cells to a level similar to that of senescent cells (Fig. 4). The elevation of the accumulation of DNA damage by prolonged H2 O2 treatment may contribute to a reduction in DNA repair capacity. As a result, the accumulation of DNA damage may reach a threshold and trigger cell senescence. This indicates that irreversible SIPS of ‘young’ cells by prolonged H2 O2 treatment results from the combination of DNA damage accumulation and DNA repair capacity decline, and it may also explain why the low doses or short treatment of H2 O2 only induce temporary cell cycle arrest instead of SIPS. These observations are


consistent with the possibility that decreased repair and, thus, persistence of unrepaired DNA damage, might contribute to senescence and SIPS. However, this hypothesis is still speculative due to the observation that decreased DNA damage repair capacity has also been found in telomere-dependent senescence (Duan, Chen, Liu, Zhang, & Tong, 2004) (Duan et al., 2001; Shin et al., 2004), and might well be a consequence of telomere-dependent senescence rather than a cause. Replicative senescence is thought likely to be a result of a genetic program or changes in gene expression. A group of DNA-damage-and-repair genes, such as p53, p21 and GADD45, can be activated during H2 O2 treatment. Activation of these checkpoint proteins ultimately contributes to halting cell proliferation to allow DNA repair to occur. Interestingly, these genes are also senescence-associated genes (Chen, 2000; Hollander et al., 1999). In our study, the acute treatment by sub-lethal amounts of H2 O2 (100 ␮M, 1 h) caused significant elevation of p21 and GADD45 gene expression in mRNA (Fig. 5A), but these gradually were reduced and returned to the original level after recovery (data not show). This is correlated with the reversible cell cycle arrest and the need for DNA repair. However, in contrast to the acute treatment, the prolonged low dose (10 ␮M) H2 O2 treatment caused a sustained increase of p21 and GADD45 protein levels, and p53 binding activity when we examined them on days 7 and 14. In contrast, only a slight increase was seen on day 1 (Fig. 5B and C). These results further underline the contribution of the accumulation of DNA damage to cell senescence and suggest that prolonged H2 O2 treatment-induced cell cycle arrest is still through the p53-dependant pathway, while the DNA repair capacity did not increase along with the elevation of p21, GADD45 levels and p53 binding activity, indicating telomere shortening may play a role as mentioned above. Therefore, there is no substantial difference between SIPS induced by prolonged low dose H2 O2 treatment in our study and replicative senescence. Besides oxidative DNA damage accumulation, telomere shortening has been widely considered as another important mechanism to trigger replicative senescence in human fibroblasts (Hayflick, 2000). Telomere shortening has been proven to be a cell division counter in proliferating fibroblasts, or ‘replicometer.’ Telomere shortening down to a threshold length seems to be the best-known predictor of senescence. However, what

1417

causes telomere shortening and how this trigger functions is still not clear. Recent studies by von Zglinicki et al. (Saretzki & von Zglinicki, 1999; Toussaint et al., 2000; von Zglinicki, 2000, 2001, 2003) showed that telomere shortening is largely dependent on the interplay of oxidative stress and antioxidant defense rather than the counter clock of numbers of cell divisions. In this study, we found that prolonged H2 O2 treatment significantly accelerated the process of telomere shortening (Fig. 6), indicating that prolonged oxidative stress by H2 O2 can modify the rate of telomere shortening. This result is reinforced by other researchers’ observations. For example, by increasing the ambient oxygen partial pressure from 20% to 40%, Honda et al. (2001) found that the telomere shortening rate was significantly faster in RPE340 and WI38 fibroblasts. Similarly, von Zglinicki et al. detected 4- to 10-fold increases of the shortening rate (von Zglinicki et al., 1995). However, some researchers argue that telomere shortening may not be one of the mechanisms of H2 O2 induced SIPS and that DNA damage is uncoupled with telomere shortening because they did not see significant telomere shortening in their H2 O2 induced SIPS (Chen et al., 2001). However, they agree that cell replication is required for the telomeres to be shortened. Only using prolonged low levels of oxidative stress, as used in our experiments, may allow the detection of the acceleration of the telomere shortening rate because of the presence of cell division. Chen et al. (2001) used acute sub-lethal doses of H2 O2 treatment, which ceases proliferation quickly. Thus the cells likely had no time to exhibit their telomere shortening process. Even though they used mild doses of H2 O2 and repetitively treated and reseeded the cells in new plates, the LTRF decrease was not observed in cells because the majority cells were not dividing after the treatment. Having shown that prolonged low dose H2 O2 stress accelerates telomere shortening, the question of the mechanism of this telomere shortening still remains. A number of groups have provided evidence that the accumulation of single-strand breaks might lead to premature termination of replication and to more telomere shortening because of extremely low repair efficiency. In fact, single-strand breaks that accumulate during G0 arrest are quantitatively transferred into telomere loss within 1–2 PD following release of cells from the arrest (Sitte et al., 1998). Cells that are held for a time in a non-proliferative state accumulate telomeric

1418


single-strand breaks. But when they are re-plated and allowed to proliferate, their telomeres initially shorten more quickly until the frequency of single-strand breaks returns to normal (von Zglinicki, 2002). In our SCGE study (Fig. 4), double DNA strand breaks may have contributed to temporary cell cycle arrest for repairing DNA damage, but single strand breaks may have contributed to telomere shortening. Adda di Fagagna et al. (2003) recently demonstrated that telomere-initiated senescence reflects a DNA damage checkpoint response that is activated with a direct contribution from dysfunctional telomeres. Li et al. (2003) also demonstrated that the induction of an identical senescent phenotype by treatment of fibroblasts with oligonuceotides homologous to the 3 telomere overhang. They proposed a model in which senescence is induced through the same p53-mediated DNA damage response pathway whether caused by exogenous oligonucleotides, serial passage, or acute/chronic DNA damage. These studies are consistent with our gene expression data for p21, GADD45 and the elevation of p53 activity. We demonstrate that telomere shortening is not independent of oxidative stress but a stress-dependent process. On the other hand, telomere shortening induced by prolonged oxidative stress may further act as the trigger of senescence through the ATM and p53-dependant pathway. Herbig et al. (2004) found that the DNA damage foci localized to the telomere, and that the telomeric DNA damage involved ATM and activated the p53/p21 pathway. Therefore, the nature of telomere shortening is not different from DNA damage (Harrington, 2004; Shin et al., 2004); rather lack of a repair mechanism for telomere damage may serve as a long-term DNA damage signal and be responsible for the consistent elevation of p21 expression and p53 activation shown in our study, and long-term cell cycle arrest that is the hallmark of senescence. Taken together, prolonged low doses of oxidative stress-induced SIPS shares not only common features but also common pathways with replicative senescence. From this point of view, the program-controlled cellular senescence in vivo may be easily considered as a stress response to the accumulation of DNA damage. Senescence-associated gene expression and telomere shortening further trigger the onset of cell senescence. Therefore, several aging theories perhaps share as their mechanism of action oxidative damage, the common underlying cause for age-related morbidity and mor-

tality and telomere shortening. Thus levels of oxidative DNA-damage-and-repair capacity might be exploited as reliable markers of cell senescence.

Acknowledgements We are grateful to Dr. Carter van Waes (National Institutes of Health) and Ms. Patricia Soochan (Howard Hughes Medical Institute) for their critical reading and editing of the manuscript. This work was supported by grant G2000517001 from the Major State Basic Research Development Program of China and grant 39930170 from the National Science Foundation of China.

References Bayens, J. W. (2002). The Maillard hypothesis on aging: Time to focus on DNA. Ann. N. Y. Acad. Sci., 959, 360–367. Beckman, K. B., & Ames, B. N. (1998). The free radical theory of aging matures. Physiol. Rev., 78, 547–581. Campisi, J. (1996). Replicative senescence: An old lives’ tale? Cell, 84, 497–500. Campisi, J., Kim, S. H., Lim, C. S., & Rubio, M. (2001). Cellular senescence, cancer and aging: The telomere connection. Exp. Gerontol., 36, 1619–1637. Chen, Q. M. (2000). Replicative senescence and oxidant-induced premature senescence – beyond the control of cell cycle checkpoints. Mol. Cell. Gerontol., 908, 111–125. Chen, Q., & Ames, B. (1994). Senescence-like growth arrest induced by hydrogen peroxide in human diploid fibroblast F65 cells. PNAS, 91, 4130–4134. Chen, Q. M., Prowse, K. R., Tu, V. C., Purdom, S., & Linskens, M. H. K. (2001). Uncoupling the senescent phenotype from telomere shortening in hydrogen peroxide-treated fibroblasts. Exp. Cell Res., 265, 294–303. d’Adda di Fagagna, F., Reaper, P. M., Clay-Farrace, L., Fiegler, H., Carr, P., Von Zglinicki, T., et al. (2003). A DNA damage checkpoint response in telomere-initiated senescence. Nature, 426, 194–198. de Boer, J., & Hoeijmakers, J. H. J. (2000). Nucleotide excision repair and human syndromes. Carcinogenesis, 21, 453– 460. Dimri, G. P., Lee, X., Basile, G., Acosta, M., Scott, G., Roskelley, C., et al. (1995). A biomarker that identifies senescent human cells in culture and in aging skin in vivo. Proc. Natl. Acad. Sci. U.S.A., 92, 9363–9367. Duan, J., Chen, Z., Liu, P., Zhang, Z., & Tong, T. (2004). Wild-type p16INK4a suppresses cell growth, telomerase activity and DNA repair in human breast cancer MCF-7 cells. Int. J. Oncol., 24, 1597–1605.

J. Duan et al. / The International Journal of Biochemistry & Cell Biology 37 (2005) 1407–1420 Duan, J., Zhang, Z., & Tong, T. (2001). Senescence delay of human diploid fibroblast induced by anti-sense p16INK4a expression. J. Biol. Chem., 276, 48325–48331. Frippiat, C., Chen, Q. M., Zdanov, S., Magalhaes, J. P., Remacle, J., & Toussaint, O. (2001). Subcytotoxic H2 O2 stress triggers a release of transforming growth factor-beta 1, which induces biomarkers of cellular senescence of human diploid fibroblasts. J. Biol. Chem., 276, 2531–2537. Frippiat, C., Dewelle, J., Remacle, J., & Toussaint, O. (2002). Signal transduction in H2 O2 -induced senescence-like phenotype in human diploid fibroblasts. Free Radic. Biol. Med., 33, 1334– 1346. Hall, D. M., Sattler, G. L., Sattler, C. A., Zhang, H. J., Oberley, L. W., Pitot, H. C., et al. (2001). Aging lowers steady-state antioxidant enzyme and stress protein expression in primary hepatocytes. J. Gerontol. A Biol. Sci. Med. Sci., 56, B259–B267. Hamilton, M. L., Van Remmen, H., Drake, J. A., Yang, H., Guo, Z. M., Kewitt, K., et al. (2001). Does oxidative damage to DNA increase with age? PNAS, 98, 10469–10474. Harley, C. B., Futcher, A. B., & Greider, C. W. (1990). Telomeres shorten during ageing of human fibroblast. Nature, 345, 458– 460. Harman, D., Holliday, R., & Meydani, M. (1998). Toward prolongation of the healthy life span. Ann. N. Y. Acad. Sci., 854. Harrington, L. (2004). Those dam-aged telomeres!. Curr. Opin. Genet. Dev., 14, 22–28. Hayflick, L. (1997). Mortality and immortality at the cellular level. A review. Biochemistry (Mosc.), 62, 1180–1190. Hayflick, L. (2000). The illusion of cell immortality. Br. J. Cancer, 83, 841–846. Hayflick, L., & Moorhead, P. S. (1961). The serial cultivation of human diploid cell strains. Exp. Cell Res., 25, 585–621. Herbig, U., Jobling, W. A., Chen, B. P., Chen, D. J., & Sedivy, J. M. (2004). Telomere shortening triggers senescence of human cells through a pathway involving ATM, p53, and p21 (CIP1), but not p16 (INK4a). Mol. Cell., 14, 501–513. Hollander, M. C., Sheikh, M. S., Bulavin, D. V., Lundgren, K., Augeri-Henmueller, L., Shehee, R., et al. (1999). Genomic instability in Gadd45a-deficient mice. Nat. Genet., 23, 176– 184. Honda, S., Hjelmeland, L. M., & Handa, J. T. (2001). Oxidative stress–induced single-strand breaks in chromosomal telomeres of human retinal pigment epithelial cells in vitro. Invest. Ophthalmol. Vis. Sci., 42, 2139–2144. Hupp, T. R., Meek, D. W., Midgley, C. A., & Lane, D. P. (1992). Regulation of the specific DNA binding function of p53. Cell, 71, 875–886. Itahana, K., Dimri, G., & Campisi, J. (2001). Regulation of cellular senescence by p53. Eur. J. Biochem., 268, 2784–2791. Kahlem, P., Dorken, B., & Schmitt, C. A. (2004). Cellular senescence in cancer treatment: friend or foe? J. Clin. Invest., 113, 169– 174. Kawanishi, S., Hiraku, Y., & Oikawa, S. (2001). Mechanism of guanine-specific DNA damage by oxidative stress and its role in carcinogenesis and aging. Mutat. Res., 488, 65–76. Klungland, A., Rosewell, I., Hollenbach, S., Larsen, E., Daly, G., Epe, B., et al. (1999). Accumulation of premutagenic DNA le-

1419

sions in mice defective in removal of oxidative base damage. PNAS, 96, 13300–13305. Li, G.-Z., Eller, M. S., Firoozabadi, R., & Gilchrest, B. A. (2003). Evidence that exposure of the telomere 3 overhang sequence induces senescence. PNAS, 100, 527–531. Li, J., Zhang, Z., & Tong, T. (1995). The proliferative response and anti-oncogene expression in old 2BS cells after growth factor stimulation. Mech. Ageing Dev., 80, 25–34. Mattson, M. P. (2003). Gene–diet interactions in brain aging and neurodegenerative disorders. Ann. Intern. Med., 139, 441– 444. Parrinello, S., Samper, E., Krtolica, A., Goldstein, J., Melov, S., & Campisi, J. (2003). Oxygen sensitivity severely limits the replicative lifespan of murine fibroblasts. Nat. Cell. Biol., 5, 741– 747. Pero, R. W., Anderson, M. W., Doyle, G. A., Anna, C. H., Romagna, F., Markowitz, M., et al. (1990). Oxidative stress induces DNA damage and inhibits the repair of DNA lesions induced by N-acetoxy-2-acetylaminofluorene in human peripheral mononuclear leukocytes. Cancer Res., 50, 4619–4625. Rodemann, H. P., Bayreuther, K., Francz, P. I., Dittmann, K., & Albiez, M. (1989). Selective enrichment and biochemical characterization of seven human skin fibroblasts cell types in vitro. Exp. Cell. Res., 180, 84–93. Saretzki, G., & von Zglinicki, T. (1999). Replicative senescence as a model for aging: the role of oxidative stress and telomere shortening. Zeitschrift Fur Gerontologie Und Geriatrie, 32, 69–75. Severino, J., Allen, R. G., Balin, S., Balin, A., & Cristofalo, V. J. (2000). Is [beta]-galactosidase staining a marker of senescence in vitro and in vivo? Exp. Cell Res., 257, 162– 171. Shin, K.-H., Kang, M. K., Dicterow, E., Kameta, A., Baluda, M. A., & Park, N.-H. (2004). Introduction of human telomerase reverse transcriptase to normal human fibroblasts enhances DNA repair capacity. Clin. Cancer Res., 10, 2551–2560. Singh, N. P., McCoy, M. T., Tice, R. R., & Schneider, E. L. (1988). A simple technique for quantitation of low levels of DNA damage in individual cells. Exp. Cell Res., 175, 184–191. Sitte, N., Saretzki, G., & von Zglinicki, T. (1998). Accelerated telomere shortening in fibroblasts after extended periods of confluency. Free Radic. Biol. Med., 24, 885–893. Toussaint, O., Houbion, A., & Remacle, J. (1992). Aging as a multistep process characterized by a lowering of entropy production leading the cell to a sequence of defined stages. II. Testing some predictions on aging human fibroblasts in culture. Mech. Ageing Dev., 65, 65–83. Toussaint, O., Medrano, E. E., & von Zglinicki, T. (2000). Cellular and molecular mechanisms of stress-induced premature senescence (SIPS) of human diploid fibroblasts and melanocytes. Exp. Gerontol., 35, 927–945. von Zglinicki, T. (2001). Telomeres and replicative senescence: is it only length that counts? Cancer Lett., 168, 111–116. von Zglinicki, T. (2002). Oxidative stress shortens telomeres. Trends Biochem. Sci., 27, 339–344. von Zglinicki, T. (2003). Replicative senescence and the art of counting. Exp. Gerontol., 38, 1259–1264.

1420


von Zglinicki, T., Burkle, A., & Kirkwood, T. B. L. (2001). Stress, DNA damage and ageing—an integrative approach. Exp. Gerontol., 36, 1049–1062. von Zglinicki, T., Pilger, R., & Sitte, N. (2000). Accumulation of single-strand breaks is the major cause of telomere short-

ening in human fibroblasts. Free Radic. Biol. Med., 28, 64– 74. von Zglinicki, T., Saretzki, G., Docke, W., & Lotze, C. (1995). Mild hyperoxia shortens telomeres and inhibits proliferation of fibroblasts: A model for senescence? Exp. Cell Res., 220, 186–193.