- Email: [email protected]
Oxidative DNA damage as a marker of aging in WI-38 human fibroblasts
Experimental Gerontology 37 (2002) 647±656
www.elsevier.com/locate/expgero
Oxidative DNA damage as a marker of aging in WI-38 human ®broblasts Federica I. Wolf*, Angela Torsello, Valeria Covacci, Silvia Fasanella, Micaela Montanari, Alma Boninsegna, Achille Cittadini School of Medicine, Institute of General Pathology and Giovanni XXIII Cancer Research Center, Catholic University of Sacred Heart, L. go F. Vito 1, 00168 Rome, Italy Received 2 October 2001; received in revised form 10 January 2002; accepted 11 January 2002
Abstract Cause±effect relationships between oxidative stress, DNA damage and aging were investigated in WI-38 human diploid ®broblasts at 21, 41 or 58 population doublings (PDs), corresponding to young, middle age or old ®broblasts, respectively. Oxidative DNA damage was evaluated by immunohistochemical detection of 8-hydroxy-2 0 deoxyguanosine (8-OHdG) adducts or by single cell microgel electrophoresis (COMET assay). Aging was evaluated by growth rate, senescence-associated-bgalactosidase (SA-b galactosidase) activity, cell cycle distribution, and expression of p21. Our results demonstrate that (i) oxidative DNA damage is proportional to the age of cells (ii) DNA damage in old/58 PDs cells re¯ects both an increased susceptibility to oxidative stress, induced by acute exposure to sub-lethal concentrations of hydrogen peroxide (H2O2), and a reduced ef®ciency of repair mechanisms. We also show that mild chronic oxidative stress, induced by prolonged exposure to 5 mM H2O2, accelerates aging in ®broblasts. In fact, this treatment increased 8-OHdG levels, SA-b-galactosidase activity, and G0/G1 cell cycle arrest in middle age/41 PDs, making them similar to H2O2-untreated old/58 PDs cells. Although other mechanisms may concur in mediating the effects of H2O2, these results lend support to the concept that oxidative stress may be a key determinant of aging. Measurements of oxidative DNA damage might therefore be exploited as reliable marker of cellular aging. q 2002 Elsevier Science Inc. All rights reserved. Keywords: 8-OHdG; H2O2; DNA strand breaks; SA-b-galactosidase; p21; Cell cycle
1. Introduction Living organisms are constantly exposed to oxidative stress induced by environmental agents or endogenous metabolic processes. Oxidative DNA damage * Corresponding author. Tel.: 139-63016619; fax: 139-63012753. E-mail address: [email protected] (F.I. Wolf). Abbreviations: DAB, di-amino benzidine; DT, doubling time (h); 8OHdG, 8-hydroxy-2 0 deoxyguanosine; OD, optical density; PDs, population doublings; SA-b-galactosidase, senescence-associated b-galactosidase (pH 6); THF, tetrahydrofurane
is by far the most detrimental consequence of oxidative stress, causing a large array of irreversible dysfunctions and eventually cell death (Kawanishi et al., 2001 for rev.). DNA damage may be an important determinant also in the pathophysiology of aging (Beckman and Ames, 1998 for rev.); in fact, the levels of oxidative DNA damage are signi®cantly increased in senescent cells, especially in post-mitotic tissues (Hamilton et al., 2001; Hosokawa et al., 2000). DNA damaged cells can undergo death, usually by apoptosis, or growth arrest, probably through a common p53-dependent mechanism (Johnson et al.,
0531-5565/02/$ - see front matter q 2002 Elsevier Science Inc. All rights reserved. PII: S 0531-556 5(02)00005-0
648
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
1996; Itahana et al., 2001 for rev.). Alternatively, cells can repair DNA by complex enzymatic mechanisms like base excision repair, nucleotide excision repair, mismatch repair, etc. (Lindahl and Wood, 1999; De Boer and Hoeijmakers, 2000). Controversy exists on whether the DNA damage observed in senescent cells re¯ects prolonged oxidative attacks (Hall et al., 2001), or a decline in repair mechanisms (Klungland et al., 1999; von Zglinicki et al., 2001), or a combination of the two factors. Regardless of such uncertainties, evaluating the levels of oxidative DNA damage in a given cell population might be exploited as a novel marker to predict the biologic age of such cells, and hence their capability to survive, to proliferate, or to undergo pathophysiologic modi®cations. Oxidative DNA damage can be evaluated by means of different techniques. Alkaline elution and single cell electrophoresis are sensitive methods for detecting low levels of double (Kohn, 1991) or single (Singh et al., 1988) strand breaks, respectively. 8-Hydroxy2 0 deoxyguanosine (8-OHdG) or MDA±DNA adducts are selective markers of oxidative damage to speci®c DNA residues (reviewed in Marnett (2000)). In particular, 8-OHdG has gained popularity because it is one of the most abundant oxidative DNA adducts, correlates with mutagenesis, carcinogenesis and aging (Choi et al., 1995; Ottender and Luz, 1999; Xu et al., 1999; Romano et al., 2000), and can be measured by either HPLC±mass spectroscopy, gas chromatography, 32P post-labeling, or monoclonal antibodies (Collins et al., 1997). The latter offers the unique advantage of measuring oxidative DNA damage in whole cells or tissues, thus avoiding artifacts due to DNA damage occurring during extraction procedures. Diploid ®broblasts offer the typical model for studying the process of aging in vitro. Once explanted human diploid ®broblasts proliferate rapidly, undergoing four or more population doublings (PDs) per week, but then reach a phase when growth arrest may occur, regardless of culture conditions. This behaviour is referred to as `®nite proliferative life span in culture', or `cellular aging', or `replicative senescence' (Cristofalo and Pignolo, 1993). Growth arrest, typically at G0/G1, is not the only characteristic of senescent diploid ®broblasts. These cells exhibit also phenotype changes, such as cytoplasm enlargement and pseudopode appearance; modi®ed cell functions, like diminished heat shock response or
decreased production of IL-6 (Bonelli et al., 1999; Goodman and Stein, 1994); and expression of senescence-associated enzymes like pH 6.0-b galactosidase (Dimri et al., 1995). In this study we evaluated oxidative DNA damage in WI-38 diploid ®broblasts, using both single strand breaks and 8-OHdG, and correlated the levels of these two markers to the aging of cells, monitored by proliferation rate and senescence-associated-b-galactosidase (SA-b-galactosidase) expression. The aim was to establish whether DNA oxidative damage could be considered a biomarker of aging. 2. Materials and methods 2.1. Chemicals All chemicals and reagents were of the highest analytical grade and were purchased from Sigma Chemicals (Milan, Italy). a-Tocopherol (Fluka, Milan, Italy) was dissolved in tetrahydrofurane (THF). The amount of THF delivered to the cells never exceeded 0.5% (v/v). 2.2. Cell and treatment protocols WI-38 diploid ®broblasts derived from embryonic human lung primary culture were purchased from Istituto Zoopro®lattico dell'Emilia-Romagna (Brescia, Italy). Cells were received from the provider at the 11 PDs and were grown on Petri dishes in Eagle basal medium (BME) (Sigma Italia, Milano) supplemented with 10% fetal bovine serum (FBS) (Life Technologies Italia, Milano) and 1 mM sodium pyruvate, at 37 8C under 5% CO2/air atmosphere, and maintained in culture untill 58 PDs. After this passage cells were viable but virtually growth arrested. Experimental incubations were carried out at 37 8C in sub-con¯uent cultures that had been seeded at 10 £ 10 4 cells/cm 2 in BME 1 10% FBS 1 1 mM sodium pyruvate, in the presence of sub-lethal amounts of H2O2 from 50 to 300 mM for 30 min (acute treatment). Pretreatment with antioxidant was carried out by incubating cells for 24 h in the presence of 50 mM a-tocopherol. At the end of incubations the medium was washed out and replaced with a fresh one which contained H2O2 at the desired concentration. Chronic oxidative stress was induced by adding
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
5 mM H2O2 at every medium change. This treatment was started when ®broblasts were between 11 and 12 PDs, and was prolonged up to 21, 41 or 58 PDs. Cells examined at these PDs after treatment with H2O2 were compared to age-matched control cells grown in the absence of H2O2. 2.3. Assays for 8-OHdG and DNA strand breaks Cells were cultured directly on glass slides placed into 10 cm dishes. Slides were air dried for 10± 30 min, ®xed in 95% cold ethanol (220 8C) for 10 min, and stored at 220 8C. Detection of 8-OHdG by immunohistochemistry coupled with DAB (Vector, Burlingham, USA) was carried out essentially as described by Yarbourough et al. (1996). 1F7 monoclonal antibody for 8-OHdG was kindly provided by Dr R.M. Santella, Columbia School of Public Health, NY. Semi-quantitative evaluation of the staining was carried out by an optical microscope (Zeiss, Germany, 400 £ ) connected to an Optimas 5 morphometer (Optimas Corporation, USA). Nuclear staining of duplicate samples was evaluated in approximately 50 cells of randomly chosen ®elds by operators who were blind to the status of cell treatment, as recommended by the antibody provider. Positive controls were prepared by treating the cells with 0.5 mM H2O2. Positive and negative controls were included in each assay. Data are reported as units of optical density (OD). Detection of DNA strand breaks by single cell microgel electrophoresis (COMET assay) was performed by the method of Singh et al. (1988). Data are reported as the ratio between tail/nucleus areas, evaluated by Optimas 5 morphometer. 2.4. SA-b -galactosidase assay Detection of pH 6 SA-b-galactosidase was performed essentially as described by Dimri et al. (1995). Brie¯y, cell were washed in PBS, ®xed for 3 min in 2% formaldehyde, washed and incubated at 37 8C with SA-b-Gal stain solution (1 mg of 5-bromo4-chloro-3-indolyl b-d-galactoside/ml (Labtek Italia, Milano), 40 mM citric acid/sodium phosphate, pH 6.0, 5 mM potassium ferrocyanide, 5 mM potassium ferricyanide, 150 mM NaCl, 2 mM MgCl2). Staining was evaluated after 16 h incubation at oven temperature in CO2-free atmosphere. The number of blue
649
stained cells were counted in 20 ®elds (from a total of approximately 1.4 £ 10 6 cells) at 400 £ magni®cation, and expressed as percentage of positive cells. To avoid staining due to cell con¯uence rather than to proliferative senescence (Severino et al., 2000), the assay was performed in sub-con¯uent cultures displaying comparable cell density. 2.5. Cyto¯uorimetric analysis of cell cycle distribution Trypsinized cells were collected and washed twice with PBS. About 1 £ 10 6 cells were suspended in 1 ml PBS, ®xed in 5 ml of 70% ethanol and stored at 4 8C. At the time of analysis cells were collected by centrifugation and the pellets were resuspended in 0.2 mg/ ml of propidium iodide in PBS containing 0.6% Nonidet P-40 and RNAase (1 mg/ml); suspensions were incubated in the dark at room temperature for 30 min. The cell suspensions were then analyzed for propidium iodide-stained DNA content in a Coulter EPICS 753 ¯ow cytometer (excitation at 488 nm/ emission at 580±650 nm). The percentage of cells in the different phases of the cell cycle was determined using the Multicycle software version 2.53. 2.6. p21 expression Cell pellets were suspended in 2±4 volumes of sonication buffer containing protease and phosphatase inhibitors (50 mM Hepes/pH 7.5, 150 mM NaCl, 2.5 mM EGTA, 1 mM EDTA, 10% glycerol, 0.1% Tween 20, 0.1% DTT, 10 mg/ml aprotinin, 10 mg/ ml leupeptin and 1 mM PMSF) and sonicated at 4 8C with a Soni®er Cell Disruptor (Ultrasonic Instruments Intl. Inc., Farmigdale, NY). Homogenates were incubated on ice for 30 min and centrifuged at 14,000 rpm for 15 min at 4 8C. The supernatants were assayed for protein content by the Biorad protein assay method (Biorad laboratories GmbH, Munchen, Germany) and stored at 280 8C. Western blot analysis was performed on 100 mg proteins from each sample, separated by SDS-PAGE (12.5%) and transferred to immobilion-P membranes (Millipore, Bedford, MA) at 100 V for 1 h at 4 8C. Immunodetection was performed using the enhanced chemiluminescence kit for western blotting detection (Amersham Pharmacia Biotech, Freiburg, Germany). The monoclonal antibody to p21 (1:400 dilution) was from Santa
650
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
Cruz Biotechnology (Santa Cruz, CA). The monoclonal antibody to b-actin (1:500 dilution) (Sigma Chemicals) was utilized as internal control. 2.7. Other assays and conditions Cells were counted with a Coulter cell sorting (Coulter Z1). Doubling time (DT) of ®broblasts at the various passages in culture was calculated during the exponential growth phase. Data are given as means ^ SE of at least three separate experiments. In the ®gures, SE are indicated by vertical bars; values without vertical bars have SE within the symbols. Statistical analyses were performed by unpaired Student's t test, and differences were considered signi®cant when P , 0.05. Other details are given in the legends to ®gures and tables. 3. Results 3.1. DNA oxidative damage as a function of aging Oxidative DNA damage in WI-38 ®broblasts was evaluated by immunohistochemical detection of 8OHdG. Fig. 1 shows 8-OHdG nuclear staining in ®broblasts treated with H2O2. The intensity of nuclear staining detected by DAB reaction and calculated by OD, was proportional to the concentration of 8OHdG. DNA damage was evaluated by 8-OHdG in ®broblasts at different passages in culture. As shown in Fig. 2(A), `old ®broblasts' (58 PDs) exhibited steady state levels of 8-OHdG signi®cantly higher than those detected in `middle age' (41 PDs) or `young' (21 PDs) ®broblasts (173 ^ 7 vs 120 ^ 6 and 107 ^ 11 OD, respectively; P , 0.05). This suggested that in WI-38 ®broblasts the aging process was characterized by a persistent increase of oxidative DNA damage. Fibroblasts at different PDs were exposed to an acute sub-lethal oxidative treatment
Fig. 1. Immunohistochemical analysis of 8-OHdG in WI-38 ®broblasts treated with H2O2. Labeling of monoclonal antibody (1F7) with 8-OHdG was evidentiated by DAB reaction. The intensity of nuclear staining evaluated by OD, was proportional to the levels of 8-OHdG. (a) control ®broblasts; (b) and (c) cells treated for 30 min with 100 and 200 mM H2O2, respectively. Technical details are given under Section 2.
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
651
PDs ®broblasts was in between those observed in old or young counterparts. While 200 mM H2O2 increased 8-OHdG to the same level as observed in old/58 PDs ®broblasts, 50 and 100 mM H2O2 had relatively little effects on 8-OHdG levels as compared with old ®broblasts (see also Fig. 2(A)). Collectively, old/58 PDs ®broblasts were much more sensitive than young/21 PDs ®broblasts to the action of both low and high concentrations of H2O2, whereas middle age/41 PDs ®broblasts seemed to be sensitive only to the highest concentration. To establish a correlation between H2O2 treatment and changes in 8-OHdG levels, we pretreated ®broblasts with a-tocopherol (50 mM), an antioxidant able to counteract oxidative events induced by H2O2 (Covacci et al., 2001). Fig. 2(B) shows that pre-treatment with a-tocopherol had no effect on the increase of 8-OHdG induced by H2O2 in young/21 PDs ®broblasts, but clearly protected old/58 PDs ®broblasts. 3.2. Recovery of oxidative DNA damage
Fig. 2. Steady state and H2O2-induced 8-OHdG levels in WI-38 ®broblasts at various passages in culture, in the absence or presence of a-tocopherol pretreatment. (A) 8-OHdG was evaluated immunohistochemically before and after 30 min treatment with the indicated concentrations of H2O2 in young/21, middle age/41 and old/ 58 PDs ®broblasts. (B) Effect of 24 h pre-treatment with 50 mM a-tocopherol on 8-OHdG levels in young/21 and old/58 PDs ®broblasts. Data are expressed as optical density units (OD), and represent the means ^ SE of at least three separate experiments. *P , 0.05 vs corresponding control.
consisting of 30 min exposure to increasing concentrations of H2O2 (50, 100 and 200 mM). As also shown in Fig. 2(A), old/58 PDs ®broblasts responded to such treatment with a rapid elevation of the levels of 8OHdG, which increased by 47 or 78% at 50 or 200 mM H2O2, respectively. Fifty micromolar H2O2 had a marginal effect on the levels of 8-OHdG in young/21 PDs ®broblasts. In these cells, only 200 mM H2O2 proved to induce a signi®cant (P , 0.05) increase of 8-OHdG over pre-treatment steady state levels; nevertheless, the ®nal levels of 8-OHdG remained much below those detected in old ®broblasts. The response observed in middle age/41
To investigate on the capability of different PDs ®broblasts to repair oxidative DNA damage, we performed experiments in which ®broblasts were exposed to H2O2 for 30 min and subsequently allowed to recover for 120 min in H2O2-free medium. As shown in Fig. 3(A), young/21 PDs ®broblasts always removed the H2O2-induced accumulation of 8-OHdG, regardless of whether they had been treated with 100, 200 or 300 mM H2O2. This repair occurred within a 60 min period of post-treatment recovery. In contrast, H2O2-induced levels of 8-OHdG always remained above the pre-treatment levels in old/58 PDs ®broblasts. In particular, measurements of 8-OHdG after 60 min recovery from 100, 200 or 300 mM H2O2 demonstrated that only 85, 70 or 45% of DNA damage, respectively, had been repaired in these cells (Fig. 3(B)). Further evidence that the capability of ®broblasts to withstand oxidative DNA damage was in¯uenced by aging was obtained by measuring DNA strand breaks by the COMET assay, consisting of single cell electrophoresis after DNA solubilization (Singh et al., 1988). Fig. 4 reports DNA damage evaluated with this technique at steady state levels, after treatment with 50±200 mM H2O2, and after 60 min recovery in fresh medium. It can be seen that young/21 PDs
652
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
Fig. 3. Effects of aging on 8-OHdG repair after treatment of ®broblasts with H2O2. Cells were treated for 30 min with the indicated concentration of H2O2, the medium was washed out, and cells were re-incubated in fresh medium for up to 120 min. 8-OHdG was assayed before and after H2O2 treatment at the indicated times. (A) young/21 PDs ®broblasts; (B) old/58 PDs ®broblasts. Data are taken from duplicate experiments with .90% agreement.
®broblasts displayed lower steady state levels of DNA damage than old/58 PDs counterparts. DNA damage increased proportionally to H2O2 concentrations, but less extensively in young/21 PDs than in old/58 PDs ®broblasts (cfr. panels A and B). After 60 min recovery from H2O2 treatment, DNA damage returned to pre-treatment steady state levels in young/21 PDs ®broblasts, regardless of the concentration of H2O2. In contrast, old/58 PDs ®broblasts retained levels of DNA damage, which always exceeded pre-treatment levels, and increased proportionally to the concentration of H2O2. These results con®rmed those obtained by immunohistochemical analysis of 8-OHdG, and suggested that old ®broblasts not only were more susceptible to oxidative
stress but also displayed a decreased capability to repair DNA damage. 3.3. Effect of chronic exposure to low doses of H2O2 Because oxidative stress is an unavoidable consequence of aerobic metabolism, these results also implied that the higher levels of DNA damage observed in old cells could have been caused by a prolonged exposure to oxidative stress. To probe this hypothesis, ®broblasts between 21 and 58 PDs were exposed chronically to very low concentrations of H2O2 (5 mM H2O2), mimicking a prolonged exposure to spontaneous conditions of oxidative stress. As shown in Table 1, this treatment induced a signi®cant
Fig. 4. Single cell electrophoresis evaluation of oxidative DNA damage in young/21 PDs and old/58 PDs ®broblasts. Cultures of young/21 PDs (A) and old/58 PDs ®broblasts (B) were assayed for DNA damage by single cell electrophoresis at steady state (black columns), after 30 min incubation with 50±200 mM H2O2 (white columns), and after additional 60 min recovery in fresh medium (dashed columns). Data are expressed as the tail/nucleus ratios, and are taken from duplicate experiments with .90% agreement.
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
653
Table 1 Effect of chronic H2O2 treatment on 8-OHdG and SA-b-galactosidase in WI-38 ®broblasts at various passages in culture (*P , 0.05 vs control. Mean ^ SE from 3 to 8 experiments) PDs
21 41 58 a b c
8-OHdG a
SA-b-GAL b
Control
1 H2O2 c
Control
1 H2O2
107 ^ 11 120 ^ 6 173 ^ 7
116 ^ 3 153 ^ 7* 185 ^ 6
2.7 ^ 0.5 14.0 ^ 0.7 52.1 ^ 1.1
9.4 ^ 0.8* 36.0 ^ 2.0* 60.4 ^ 2.7
Fig. 5. Effect of chronic treatment with H2O2 on the expression levels of p21 in young to old WI-38 ®broblasts. p21 was detected by Western blot in young/21, middle age/41, and old/58 PDs ®broblasts, grown in the presence (1) or absence (2) of chronically added 5 mM H2O2. b-Actin served as control. Other conditions are given under Section 2.
OD Percentage of positive cells. H2O2 (5 mM) was delivered to cells as described under Section 2.
characterized the effects of low H2O2 on the cell cycle distribution of young, middle age and old ®broblasts. As shown in Table 2, old/58 PDs ®broblasts exhibited the highest percentage of cells in G0/G1. Chronic treatment with low H2O2 had no effect on the cell cycle distribution of young/21 PDs or old/58 PDs ®broblasts, but increased the percentage of middle age/41 PDs in G0/G1 (76.9 vs 54.9 in control cells). The latter increase was accompanied by a reduced percentage of cells in S phase (22.2 vs 32%) or in G2/M phase (0.9 vs 13.1%) and coincided with the greatest increases in the levels of 8-OHdG and SA-bgalactosidase positive cells (cf. Table 1). From these data it clearly appears that H2O2 treatment induced early signs of growth arrest in 41 PDs. Previous studies have shown that growth arrest of senescent cells was accompanied by up-regulation of p21, a negative regulator of cyclin E-CDK2 complexes (Stein and Dulic, 1995). In agreement with this report, we found that old/58 PDs ®broblasts exhibited higher levels of p21 expression than young/21 or middle age/ 41 PDs counterparts (Fig. 5). Moreover, we found that chronic treatment from 11 to 58 PDs ®broblasts with low concentrations of H2O2 increased p21 expression in young/21 PDs and, even more pronouncedly, in
increase of 8-OHdG in middle age/41 PDs ®broblasts when compared to the corresponding untreated cells (P , 0.05). This increase in oxidative DNA damage rendered middle age/41 PDs ®broblasts more similar to old/58 PDs counterparts. Thus, chronic H2O2 treatment seemed to accelerate the aging of ®broblasts. We therefore assayed for SA-b-galactosidase, an enzyme whose expression and activity are induced by H2O2 and associate with in vitro proliferative senescence (Dimri et al., 1995; Severino et al., 2000). As also shown in Table 1, the percentage of SA-b-galactosidase positive cells did increase with the age of ®broblast cultures (2.7, 14.0 and 52.1% in cells at 21, 41 and 58 PDs, respectively). Chronic treatment with `low' H2O2 increased SA-b-galactosidase positivity in all cells examined, but the increase was more pronounced in 21 and 41 PDs ®broblasts as compared to 58 PDs ®broblasts. These data con®rmed that a prolonged exposure to oxidative stress, induced by chronic treatment with low concentrations of H2O2, accelerated the process of aging in ®broblasts. To obtain correlations between H2O2-induced DNA damage and senescence-associated growth arrest we
Table 2 Effect of chronic treatment with H2O2 on cell cycle distribution in WI-38 ®broblasts at different passages in culture (*P , 0.005 vs controls (n 3)) 21 PDs
Control 1 H2O2 a a
41 PDs
58 PDs
% G0/G1
S
G2/M
G0/G1
S
G2/M
G0/G1
S
G2/M
55.0 ^ 1.2 53.9 ^ 1.2
34.6 ^ 1.8 34.5 ^ 3.8
10.4 ^ 2.0 11.6 ^ 3.5
54.9 ^ 1.8 76.9 ^ 0.9*
32.0 ^ 1.3 22.2 ^ 1.0*
13.1 ^ 1.9 0.9 ^ 0.1
78.2 ^ 2.5 79.8 ^ 1.8
17.3 ^ 2.8 16.0 ^ 0.5
4.5 ^ 0.3 4.2 ^ 0.8
H2O2 (5 mM) was delivered to cells as described under Section 2.
654
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
Table 3 Effect of chronic treatment with H2O2 on the DT of WI-38 ®broblasts at various passages in culture (*P , 0.0005; **P , 0.005 vs controls. Means ^ SE of three separate experiments) PDs
21 41 58 a
Doubling time (h) Control
H2O2 a
54.2 ^ 0.1 63.4 ^ 0.2 97.3 ^ 4.0
40.3 ^ 0.2* 55.7 ^ 1.0** 86.9 ^ 2.9
H2O2 (5 mM) was delivered to cells as described under Section 2.
middle age/41 PDs cells (Fig. 5). Concerning the mechanism of H2O2-induced senescence, it has been shown that the effect of H2O2 largely depends on the cell type, and on the dose and duration of treatment, resulting in biological responses which may vary from cell proliferation (at lower concentrations), to growth arrest and apoptosis or necrosis (at higher concentrations) (reviewed by Davies (1999)). To characterize how chronic/low dose H2O2 treatment accelerated the aging of WI-38 ®broblasts, we measured the effect of such treatment on the proliferation rate of these cells. As reported in Table 3, the DT decreased in all cells treated with H2O2, but especially in young/21 PDs ®broblasts (226% vs 212 or ±11% in middle age/ 41 PDs or old/58 PDs ®broblasts, respectively). These results suggested that H2O2 treatment was able to accelerate cell proliferation, particularly at the earlier passages, thereby inducing a more rapid progression to the so-called replicative senescence.
4. Discussion We have shown that the steady state levels of 8OHdG increase in WI-38 human ®broblasts at different passages in culture. These data con®rm and extend previous observations, and lend support to the concept that oxidative DNA damage may be responsible for the appearance of senescent phenotype in vitro as well as in vivo (von Zglinicki et al., 2001; Hamilton et al., 2001). In this study we have shown that aging may in¯uence also the vulnerability of WI-38 human ®broblasts to an acute oxidative stress, mediated by sublethal concentrations of H2O2 (50±200 mM). In fact, old/58 PDs ®broblasts proved to be more sensitive
than young/21 PDs ®broblasts to the action of H2O2, and hence accumulated more DNA damage measured by either 8-OHdG or COMET assays (cf. Figs. 2 and 4). a-Tocopherol, a powerful antioxidant, prevented H2O2-induced DNA damage in old ®broblasts but not in the more H2O2-resistant young ®broblasts. While consistent with the well known ef®cacy of vitamin E in preventing H2O2-induced elevation of free radicals in the cell and formation of DNA single strand breaks (Covacci et al., 2001; Duthie et al., 1996), these results suggest that old cells may have less antioxidant defences than their young counterparts (see also Fig. 2 and cf. Hall et al., 2001). Other antioxidants, like N-tbutyl hydroxylamine, similarly retard spontaneous or H2O2-accelerated senescence and DNA damage in cultured ®broblasts (Atamna et al., 2000a). Conversely, ®broblasts grown in 5% O2 gain more PDs and display lower oxidative damage than ®broblasts grown in 20% O2 (Chen et al., 1995). Thus, several lines of evidences concur in demonstrating that oxidative DNA damage is a clue event in senescence. DNA repair mechanisms similarly may be less ef®cient in old ®broblasts; in fact, a complete return of 8-OHdG to pre-treatment levels was observed in young/21 PDs ®broblasts but not in old/58 PDs ®broblasts (cf. Figs. 3 and 4). In accordance to our observation, an age-dependent decline was shown in the activity of the glycosylase that removes methylated bases in ®broblasts and in human lymphocytes (Atamna et al., 2000b). The increased steady state levels of DNA damage observed in senescent ®broblasts might be the consequence of a chronically induced oxidative stress, such as that mediated by oxyradicals produced during mitochondrial respiration. Consistent with this interpretation, previous studies have shown that the aging of cultured ®broblasts was accelerated by hyperoxia (Saretzki et al., 1998) or energetic stress, induced by growing cells in hypertonic NaCl and resulting in increased oxygen consumption (Reichelt and Schachtschabel, 2001). Here we have shown that aging may be accelerated also by exposing ®broblasts to chronic treatment with low concentrations of H2O2 (5 mM). In fact, this treatment induced elevation of 8-OHdG and expression of SA-b-galactosidase activity, another marker of in vitro senescence, in middle age/41 PDs ®broblasts, rendering them more similar to old/58 PDs ®broblasts (cf. Table 1). More importantly,
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
chronic treatment with low H2O2 was able to anticipate at 41 PDs the G0/G1 growth arrest usually observed at 58 PDs, an effect which was accompanied by consistent expression of a negative regulator of cell cycle progression like p21 (Stein and Dulic, 1995). The latter effects underscore the importance of DNA oxidative damage as a trigger of molecular events leading to growth arrest (Chen et al., 1998; Stein and Dulic, 1995; Itahana et al., 2001). Interestingly, chronic treatment with low H2O2 seems to accelerate the aging of 21 PDs and especially of 41 PDs cells while also enhancing their proliferation, an effect evidenced by a decreased DT (cf. Table 3). On the one hand, these results lend support to earlier evidence for H2O2-induced proliferation, involving activation of MAP kinase-mediated signal transduction pathways and transcription of early response genes like c-fos and AP-1 (Choi et al., 1995; Kim et al., 2001). On the other hand, the observation that increased proliferation coincides with an early senescent phenotype suggests that WI-38 ®broblasts react to chronic treatment with H2O2 by reaching growth arrest in a shorter time. In conclusion, we have shown that DNA oxidative damage is an important determinant of aging in WI-38 ®broblasts. This information has been obtained by exposing ®broblast to very low concentrations of H2O2 and by prolonging treatment from 11 to 21 or 41 or 58 PDs, so as to reproduce conditions in which cells accumulate damage during their life span in response to chronic exposure to oxidative challenges. Measurements of oxidative DNA damage could therefore be exploited as molecular markers of aging, in addition to those identi®ed by others in cells exposed to sub-lethal concentrations of H2O2 (®bronectin, apolipoprotein J, osteonectin, and SM22 mRNA) (Frippiat et al., 2001). We have also shown that DNA damage might be the consequence of a reduced ability of senescent cells to withstand oxidative insults. Whereas other mechanisms might contribute to the effect of H2O2 in accelerating aging, these results identify antioxidant interventions as potential new strategies to prevent DNA damage and retard the progression of aging. Acknowledgements Work co-®nanced by MURST ex 60% and
655
MURSTCOFIN # 9906241379. We are indebted to Dr R.M. Santella (Columbia School of Public Health, New York) for providing us with monoclonal antibody 1F7 against 8-OHdG. References Atamna, H., Paler-Martinez, A., Ames, B.N., 2000a. N-t-Butyl hydroxylamine, a hydrolysis product of a-phenyl-N-t-butyl nitrone, is more potent in delaying senescence in human lung ®broblasts. J. Biol. Chem. 275, 6741±6748. Atamna, H., Cheung, I., Ames, B.N., 2000b. A method for detecting abasic sites in living cells: age-dependent changes in base excision repair. Proc. Natl. Acad. Sci. USA 97, 686±691. Beckman, K.B., Ames, B.N., 1998. The free radical theory of aging matures. Physiol. Rev. 78, 547±581. Bonelli, M.A., Al®eri, R.R., Petronini, P.G., Brigotti, M., Campanini, C., Borghetti, A., 1999. Attenuated expression of 70-kDa heat shock protein in WI-38 ®broblasts during aging in vitro. Exp. Cell Res. 252, 20±32. Chen, Q., Fischer, A., Reagan, J.D., Yan, L.J., Ames, B.N., 1995. Oxidative DNA damage and senescence of human diploid ®broblast cells. Proc. Natl. Acad. Sci. USA 92, 4337±4341. Chen, Q.M., Bartholomew, J.C., Campisi, J., Acosta, M., Reagen, J.D., Ames, B.N., 1998. Molecular analysis of H2O2-induced senescent-like growth arrest in normal human ®broblasts: p53 and Rb control of G1 arrest but not cell replication. Biochem. J. 332, 43±50. Choi, A.M., Pignolo, R.J., apRhys, C.M., Cristofalo, V.J., Holbrook, N.J., 1995. Alteration in the molecular response to DNA damage during cellular aging of cultured ®broblasts: reduced AP-1 activation and collagenase gene expression. J. Cell. Physiol. 164, 65±73. Collins, A., Cadet, J., Epe, B., Gedik, C., 1997. Problems in the measurement of 8-oxoguanine in human DNA. Report of a workshop, DNA oxidation, held in Aberdeen, UK, 19±21 January, 1997. Carcinogenesis 18, 1833±1836. Covacci, V., Torsello, A., Palozza, P., Sgambato, A., Romano, G., Boninsegna, A., Cittadini, A., Wolf, F.I., 2001. DNA oxidative damage during differentiation of HL-60 human promyelocytic leukemia cells. Chem. Res. Toxicol. in press. Cristofalo, V.J., Pignolo, R.J., 1993. Replicative senescence of human ®broblast-like cells in culture. Physiol. Rev. 73, 617± 635. Davies, K.J., 1999. The broad spectrum of responses to oxidants in proliferating cells: a new paradigm for oxidative stress. IUBMB Life 48, 41±47. 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., Medrano, E.E., Linskens, M., Rubelj, I., Pereira-Smith, O., Peacocke, M., Campisi, J., 1995. A biomarker that identi®es senescent human cells in culture and in aging in vivo. Proc. Natl. Acad. Sci. USA 92, 9363±9367. Duthie, S.J., Ma, A., Ross, M.A., Collins, A.R., 1996. Antioxidant
656
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
supplementation decreases oxidative DNA damage in human lymphocytes. Cancer Res. 56, 1291±1295. Frippiat, C., Chen, Q.M., Zdanov, S., Magalhaes, J.P., Remacle, J., Toussaint, O., 2001. Subcytotoxic H2O2 stress triggers a release of transforming growth factor-beta 1, which induces biomarkers of cellular senescence of human diploid ®broblasts. J. Biol. Chem. 276, 2531±2537. Goodman, L., Stein, G.H., 1994. Basal and induced amounts of interleukin-6-mRNA decline progressively with age in human ®broblasts. J. Biol. Chem. 269, 19,250±19,255. Hall, D.M., Sattler, G.L., Sattler, C.A., Zahng, H.J., Oberley, L.W., Pitot, H.C., Kregel, K.C., 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 Richardson, A., 2001. Does oxidative damage to DNA increase with age? Proc. Natl. Acad. Sci. USA 98, 10,469±10,474. Hosokawa, M., Fujisawa, H., Ax, S., Zahn-Daimler, G., Zahn, R.K., 2000. Age-associated DNA damage is accelerated in senescence-mice. Mech. Ageing Dev. 118, 61±70. Itahana, K., Dimri, G., Campisi, J., 2001. Regulation of cellular senescence by p53. Eur. J. Biochem. 268, 2784±2791. Johnson, T.M., Yu, Z.X., Ferrans, V.J., Lowenstein, R.A., Finkel, T., 1996. Reactive oxygen species are downstream mediators of p53-dependent apoptosis. Proc. Natl. Acad. Sci. USA 93, 11,848±11,852. Kawanishi, S., Hiraku, Y., Oikawa, S., 2001. Mechanism of guanine-speci®c DNA damage by oxidative stress, role in carcinogenesis and aging. Mutat. Res. 488, 65±76. Kim, B.J., Ham, M.J., Chung, A.S., 2001. Effects of reactive oxygen species on proliferation of Chinese hamster lung ®broblast (V79) cells. Free Rad. Biol. Med. 30, 686±698. Klungland, A., Rosewell, I., Hollenbach, S., Larsen, E., Daly, G., Epe, B., Seebeg, E., Lindahl, T., Barnes, D., 1999. Accumulation of premutagenic DNA lesions in mice defective in removal of oxidative base damage. Proc. Natl. Acad. Sci.USA 96, 13,300±13,305.
Kohn, K.W., 1991. Principles and practice of DNA ®lter elution. Pharmacol. Ther. 49, 55±77. Lindahl, T., Wood, R.D., 1999. Quality control by DNA repair. Science 286, 1897±1905. Marnett, L.J., 2000. Oxyradicals and DNA damage. Carcinogenesis 21, 361±370. Ottender, M., Lutz, W.K., 1999. Correlation of DNA adduct levels with tumor incidence: carcinogenic potency of DNA adducts. Mutat. Res. 424, 237±247. Reichelt, J., Schachtschabel, D.O., 2001. Energetic stress induces premature aging of diploid human ®broblasts (Wi-38) in vitro. Arch. Gerontol. Geriatr. 32, 219±231. Romano, G., Sgambato, A., Mancini, R., Capelli, G., Giovagnoli, M.R., Flamini, G., Boninsegna, A., Vecchione, A., Cittadini, A., 2000. 8-Hydroxy-2 0 -deoxyguanosine in cervical cells: correlation with grade of displasia and human papillomavirus infection. Carcinogenesis 21, 1143±1147. Saretzki, G., Feng, J., von Zglinicki, T., Villeponteau, B., 1998. Similar gene expression pattern in senescent and hyperoxic-treated ®broblasts. J. Gerontol. A Biol. Sci. Med. Sci. 53, B438±B442. Severino, J., Allen, R.G., Balin, S., Balin, A., Cristofalo, V.J., 2000. Is b-galactosidase staining a marker of senescence in vitro and in vivo?. Exp. Cell Res. 257, 162±171. 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. Stein, G.H., Duliñ, V., 1995. Origins of G1 arrest in senescent human ®broblasts. Bioessay 17, 537±543. Xu, A., Wu, L-J., Santella, R.M., Hei, T.K., 1999. Role of oxyradicals in mutagenicity and DNA damage induced by crocidolite asbestos in mammalian cells. Cancer Res. 59, 5922±5926. Yarborough, A., Zhang, Y.J., Hsu, T.M., Santella, R.M., 1996. Immunoperoxidase detection of 8-hydroxydeoxyguanosine in a¯atoxin B1-treated rat liver and human oral mucosal cells. Cancer Res. 56, 683±688. von Zglinicki, T., Burkle, A., Kirkwood, T.B., 2001. Stress DNA damage and ageingÐan integrative approach. Exp. Gerontol. 36, 1049±1062.
www.elsevier.com/locate/expgero
Oxidative DNA damage as a marker of aging in WI-38 human ®broblasts Federica I. Wolf*, Angela Torsello, Valeria Covacci, Silvia Fasanella, Micaela Montanari, Alma Boninsegna, Achille Cittadini School of Medicine, Institute of General Pathology and Giovanni XXIII Cancer Research Center, Catholic University of Sacred Heart, L. go F. Vito 1, 00168 Rome, Italy Received 2 October 2001; received in revised form 10 January 2002; accepted 11 January 2002
Abstract Cause±effect relationships between oxidative stress, DNA damage and aging were investigated in WI-38 human diploid ®broblasts at 21, 41 or 58 population doublings (PDs), corresponding to young, middle age or old ®broblasts, respectively. Oxidative DNA damage was evaluated by immunohistochemical detection of 8-hydroxy-2 0 deoxyguanosine (8-OHdG) adducts or by single cell microgel electrophoresis (COMET assay). Aging was evaluated by growth rate, senescence-associated-bgalactosidase (SA-b galactosidase) activity, cell cycle distribution, and expression of p21. Our results demonstrate that (i) oxidative DNA damage is proportional to the age of cells (ii) DNA damage in old/58 PDs cells re¯ects both an increased susceptibility to oxidative stress, induced by acute exposure to sub-lethal concentrations of hydrogen peroxide (H2O2), and a reduced ef®ciency of repair mechanisms. We also show that mild chronic oxidative stress, induced by prolonged exposure to 5 mM H2O2, accelerates aging in ®broblasts. In fact, this treatment increased 8-OHdG levels, SA-b-galactosidase activity, and G0/G1 cell cycle arrest in middle age/41 PDs, making them similar to H2O2-untreated old/58 PDs cells. Although other mechanisms may concur in mediating the effects of H2O2, these results lend support to the concept that oxidative stress may be a key determinant of aging. Measurements of oxidative DNA damage might therefore be exploited as reliable marker of cellular aging. q 2002 Elsevier Science Inc. All rights reserved. Keywords: 8-OHdG; H2O2; DNA strand breaks; SA-b-galactosidase; p21; Cell cycle
1. Introduction Living organisms are constantly exposed to oxidative stress induced by environmental agents or endogenous metabolic processes. Oxidative DNA damage * Corresponding author. Tel.: 139-63016619; fax: 139-63012753. E-mail address: [email protected] (F.I. Wolf). Abbreviations: DAB, di-amino benzidine; DT, doubling time (h); 8OHdG, 8-hydroxy-2 0 deoxyguanosine; OD, optical density; PDs, population doublings; SA-b-galactosidase, senescence-associated b-galactosidase (pH 6); THF, tetrahydrofurane
is by far the most detrimental consequence of oxidative stress, causing a large array of irreversible dysfunctions and eventually cell death (Kawanishi et al., 2001 for rev.). DNA damage may be an important determinant also in the pathophysiology of aging (Beckman and Ames, 1998 for rev.); in fact, the levels of oxidative DNA damage are signi®cantly increased in senescent cells, especially in post-mitotic tissues (Hamilton et al., 2001; Hosokawa et al., 2000). DNA damaged cells can undergo death, usually by apoptosis, or growth arrest, probably through a common p53-dependent mechanism (Johnson et al.,
0531-5565/02/$ - see front matter q 2002 Elsevier Science Inc. All rights reserved. PII: S 0531-556 5(02)00005-0
648
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
1996; Itahana et al., 2001 for rev.). Alternatively, cells can repair DNA by complex enzymatic mechanisms like base excision repair, nucleotide excision repair, mismatch repair, etc. (Lindahl and Wood, 1999; De Boer and Hoeijmakers, 2000). Controversy exists on whether the DNA damage observed in senescent cells re¯ects prolonged oxidative attacks (Hall et al., 2001), or a decline in repair mechanisms (Klungland et al., 1999; von Zglinicki et al., 2001), or a combination of the two factors. Regardless of such uncertainties, evaluating the levels of oxidative DNA damage in a given cell population might be exploited as a novel marker to predict the biologic age of such cells, and hence their capability to survive, to proliferate, or to undergo pathophysiologic modi®cations. Oxidative DNA damage can be evaluated by means of different techniques. Alkaline elution and single cell electrophoresis are sensitive methods for detecting low levels of double (Kohn, 1991) or single (Singh et al., 1988) strand breaks, respectively. 8-Hydroxy2 0 deoxyguanosine (8-OHdG) or MDA±DNA adducts are selective markers of oxidative damage to speci®c DNA residues (reviewed in Marnett (2000)). In particular, 8-OHdG has gained popularity because it is one of the most abundant oxidative DNA adducts, correlates with mutagenesis, carcinogenesis and aging (Choi et al., 1995; Ottender and Luz, 1999; Xu et al., 1999; Romano et al., 2000), and can be measured by either HPLC±mass spectroscopy, gas chromatography, 32P post-labeling, or monoclonal antibodies (Collins et al., 1997). The latter offers the unique advantage of measuring oxidative DNA damage in whole cells or tissues, thus avoiding artifacts due to DNA damage occurring during extraction procedures. Diploid ®broblasts offer the typical model for studying the process of aging in vitro. Once explanted human diploid ®broblasts proliferate rapidly, undergoing four or more population doublings (PDs) per week, but then reach a phase when growth arrest may occur, regardless of culture conditions. This behaviour is referred to as `®nite proliferative life span in culture', or `cellular aging', or `replicative senescence' (Cristofalo and Pignolo, 1993). Growth arrest, typically at G0/G1, is not the only characteristic of senescent diploid ®broblasts. These cells exhibit also phenotype changes, such as cytoplasm enlargement and pseudopode appearance; modi®ed cell functions, like diminished heat shock response or
decreased production of IL-6 (Bonelli et al., 1999; Goodman and Stein, 1994); and expression of senescence-associated enzymes like pH 6.0-b galactosidase (Dimri et al., 1995). In this study we evaluated oxidative DNA damage in WI-38 diploid ®broblasts, using both single strand breaks and 8-OHdG, and correlated the levels of these two markers to the aging of cells, monitored by proliferation rate and senescence-associated-b-galactosidase (SA-b-galactosidase) expression. The aim was to establish whether DNA oxidative damage could be considered a biomarker of aging. 2. Materials and methods 2.1. Chemicals All chemicals and reagents were of the highest analytical grade and were purchased from Sigma Chemicals (Milan, Italy). a-Tocopherol (Fluka, Milan, Italy) was dissolved in tetrahydrofurane (THF). The amount of THF delivered to the cells never exceeded 0.5% (v/v). 2.2. Cell and treatment protocols WI-38 diploid ®broblasts derived from embryonic human lung primary culture were purchased from Istituto Zoopro®lattico dell'Emilia-Romagna (Brescia, Italy). Cells were received from the provider at the 11 PDs and were grown on Petri dishes in Eagle basal medium (BME) (Sigma Italia, Milano) supplemented with 10% fetal bovine serum (FBS) (Life Technologies Italia, Milano) and 1 mM sodium pyruvate, at 37 8C under 5% CO2/air atmosphere, and maintained in culture untill 58 PDs. After this passage cells were viable but virtually growth arrested. Experimental incubations were carried out at 37 8C in sub-con¯uent cultures that had been seeded at 10 £ 10 4 cells/cm 2 in BME 1 10% FBS 1 1 mM sodium pyruvate, in the presence of sub-lethal amounts of H2O2 from 50 to 300 mM for 30 min (acute treatment). Pretreatment with antioxidant was carried out by incubating cells for 24 h in the presence of 50 mM a-tocopherol. At the end of incubations the medium was washed out and replaced with a fresh one which contained H2O2 at the desired concentration. Chronic oxidative stress was induced by adding
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
5 mM H2O2 at every medium change. This treatment was started when ®broblasts were between 11 and 12 PDs, and was prolonged up to 21, 41 or 58 PDs. Cells examined at these PDs after treatment with H2O2 were compared to age-matched control cells grown in the absence of H2O2. 2.3. Assays for 8-OHdG and DNA strand breaks Cells were cultured directly on glass slides placed into 10 cm dishes. Slides were air dried for 10± 30 min, ®xed in 95% cold ethanol (220 8C) for 10 min, and stored at 220 8C. Detection of 8-OHdG by immunohistochemistry coupled with DAB (Vector, Burlingham, USA) was carried out essentially as described by Yarbourough et al. (1996). 1F7 monoclonal antibody for 8-OHdG was kindly provided by Dr R.M. Santella, Columbia School of Public Health, NY. Semi-quantitative evaluation of the staining was carried out by an optical microscope (Zeiss, Germany, 400 £ ) connected to an Optimas 5 morphometer (Optimas Corporation, USA). Nuclear staining of duplicate samples was evaluated in approximately 50 cells of randomly chosen ®elds by operators who were blind to the status of cell treatment, as recommended by the antibody provider. Positive controls were prepared by treating the cells with 0.5 mM H2O2. Positive and negative controls were included in each assay. Data are reported as units of optical density (OD). Detection of DNA strand breaks by single cell microgel electrophoresis (COMET assay) was performed by the method of Singh et al. (1988). Data are reported as the ratio between tail/nucleus areas, evaluated by Optimas 5 morphometer. 2.4. SA-b -galactosidase assay Detection of pH 6 SA-b-galactosidase was performed essentially as described by Dimri et al. (1995). Brie¯y, cell were washed in PBS, ®xed for 3 min in 2% formaldehyde, washed and incubated at 37 8C with SA-b-Gal stain solution (1 mg of 5-bromo4-chloro-3-indolyl b-d-galactoside/ml (Labtek Italia, Milano), 40 mM citric acid/sodium phosphate, pH 6.0, 5 mM potassium ferrocyanide, 5 mM potassium ferricyanide, 150 mM NaCl, 2 mM MgCl2). Staining was evaluated after 16 h incubation at oven temperature in CO2-free atmosphere. The number of blue
649
stained cells were counted in 20 ®elds (from a total of approximately 1.4 £ 10 6 cells) at 400 £ magni®cation, and expressed as percentage of positive cells. To avoid staining due to cell con¯uence rather than to proliferative senescence (Severino et al., 2000), the assay was performed in sub-con¯uent cultures displaying comparable cell density. 2.5. Cyto¯uorimetric analysis of cell cycle distribution Trypsinized cells were collected and washed twice with PBS. About 1 £ 10 6 cells were suspended in 1 ml PBS, ®xed in 5 ml of 70% ethanol and stored at 4 8C. At the time of analysis cells were collected by centrifugation and the pellets were resuspended in 0.2 mg/ ml of propidium iodide in PBS containing 0.6% Nonidet P-40 and RNAase (1 mg/ml); suspensions were incubated in the dark at room temperature for 30 min. The cell suspensions were then analyzed for propidium iodide-stained DNA content in a Coulter EPICS 753 ¯ow cytometer (excitation at 488 nm/ emission at 580±650 nm). The percentage of cells in the different phases of the cell cycle was determined using the Multicycle software version 2.53. 2.6. p21 expression Cell pellets were suspended in 2±4 volumes of sonication buffer containing protease and phosphatase inhibitors (50 mM Hepes/pH 7.5, 150 mM NaCl, 2.5 mM EGTA, 1 mM EDTA, 10% glycerol, 0.1% Tween 20, 0.1% DTT, 10 mg/ml aprotinin, 10 mg/ ml leupeptin and 1 mM PMSF) and sonicated at 4 8C with a Soni®er Cell Disruptor (Ultrasonic Instruments Intl. Inc., Farmigdale, NY). Homogenates were incubated on ice for 30 min and centrifuged at 14,000 rpm for 15 min at 4 8C. The supernatants were assayed for protein content by the Biorad protein assay method (Biorad laboratories GmbH, Munchen, Germany) and stored at 280 8C. Western blot analysis was performed on 100 mg proteins from each sample, separated by SDS-PAGE (12.5%) and transferred to immobilion-P membranes (Millipore, Bedford, MA) at 100 V for 1 h at 4 8C. Immunodetection was performed using the enhanced chemiluminescence kit for western blotting detection (Amersham Pharmacia Biotech, Freiburg, Germany). The monoclonal antibody to p21 (1:400 dilution) was from Santa
650
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
Cruz Biotechnology (Santa Cruz, CA). The monoclonal antibody to b-actin (1:500 dilution) (Sigma Chemicals) was utilized as internal control. 2.7. Other assays and conditions Cells were counted with a Coulter cell sorting (Coulter Z1). Doubling time (DT) of ®broblasts at the various passages in culture was calculated during the exponential growth phase. Data are given as means ^ SE of at least three separate experiments. In the ®gures, SE are indicated by vertical bars; values without vertical bars have SE within the symbols. Statistical analyses were performed by unpaired Student's t test, and differences were considered signi®cant when P , 0.05. Other details are given in the legends to ®gures and tables. 3. Results 3.1. DNA oxidative damage as a function of aging Oxidative DNA damage in WI-38 ®broblasts was evaluated by immunohistochemical detection of 8OHdG. Fig. 1 shows 8-OHdG nuclear staining in ®broblasts treated with H2O2. The intensity of nuclear staining detected by DAB reaction and calculated by OD, was proportional to the concentration of 8OHdG. DNA damage was evaluated by 8-OHdG in ®broblasts at different passages in culture. As shown in Fig. 2(A), `old ®broblasts' (58 PDs) exhibited steady state levels of 8-OHdG signi®cantly higher than those detected in `middle age' (41 PDs) or `young' (21 PDs) ®broblasts (173 ^ 7 vs 120 ^ 6 and 107 ^ 11 OD, respectively; P , 0.05). This suggested that in WI-38 ®broblasts the aging process was characterized by a persistent increase of oxidative DNA damage. Fibroblasts at different PDs were exposed to an acute sub-lethal oxidative treatment
Fig. 1. Immunohistochemical analysis of 8-OHdG in WI-38 ®broblasts treated with H2O2. Labeling of monoclonal antibody (1F7) with 8-OHdG was evidentiated by DAB reaction. The intensity of nuclear staining evaluated by OD, was proportional to the levels of 8-OHdG. (a) control ®broblasts; (b) and (c) cells treated for 30 min with 100 and 200 mM H2O2, respectively. Technical details are given under Section 2.
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
651
PDs ®broblasts was in between those observed in old or young counterparts. While 200 mM H2O2 increased 8-OHdG to the same level as observed in old/58 PDs ®broblasts, 50 and 100 mM H2O2 had relatively little effects on 8-OHdG levels as compared with old ®broblasts (see also Fig. 2(A)). Collectively, old/58 PDs ®broblasts were much more sensitive than young/21 PDs ®broblasts to the action of both low and high concentrations of H2O2, whereas middle age/41 PDs ®broblasts seemed to be sensitive only to the highest concentration. To establish a correlation between H2O2 treatment and changes in 8-OHdG levels, we pretreated ®broblasts with a-tocopherol (50 mM), an antioxidant able to counteract oxidative events induced by H2O2 (Covacci et al., 2001). Fig. 2(B) shows that pre-treatment with a-tocopherol had no effect on the increase of 8-OHdG induced by H2O2 in young/21 PDs ®broblasts, but clearly protected old/58 PDs ®broblasts. 3.2. Recovery of oxidative DNA damage
Fig. 2. Steady state and H2O2-induced 8-OHdG levels in WI-38 ®broblasts at various passages in culture, in the absence or presence of a-tocopherol pretreatment. (A) 8-OHdG was evaluated immunohistochemically before and after 30 min treatment with the indicated concentrations of H2O2 in young/21, middle age/41 and old/ 58 PDs ®broblasts. (B) Effect of 24 h pre-treatment with 50 mM a-tocopherol on 8-OHdG levels in young/21 and old/58 PDs ®broblasts. Data are expressed as optical density units (OD), and represent the means ^ SE of at least three separate experiments. *P , 0.05 vs corresponding control.
consisting of 30 min exposure to increasing concentrations of H2O2 (50, 100 and 200 mM). As also shown in Fig. 2(A), old/58 PDs ®broblasts responded to such treatment with a rapid elevation of the levels of 8OHdG, which increased by 47 or 78% at 50 or 200 mM H2O2, respectively. Fifty micromolar H2O2 had a marginal effect on the levels of 8-OHdG in young/21 PDs ®broblasts. In these cells, only 200 mM H2O2 proved to induce a signi®cant (P , 0.05) increase of 8-OHdG over pre-treatment steady state levels; nevertheless, the ®nal levels of 8-OHdG remained much below those detected in old ®broblasts. The response observed in middle age/41
To investigate on the capability of different PDs ®broblasts to repair oxidative DNA damage, we performed experiments in which ®broblasts were exposed to H2O2 for 30 min and subsequently allowed to recover for 120 min in H2O2-free medium. As shown in Fig. 3(A), young/21 PDs ®broblasts always removed the H2O2-induced accumulation of 8-OHdG, regardless of whether they had been treated with 100, 200 or 300 mM H2O2. This repair occurred within a 60 min period of post-treatment recovery. In contrast, H2O2-induced levels of 8-OHdG always remained above the pre-treatment levels in old/58 PDs ®broblasts. In particular, measurements of 8-OHdG after 60 min recovery from 100, 200 or 300 mM H2O2 demonstrated that only 85, 70 or 45% of DNA damage, respectively, had been repaired in these cells (Fig. 3(B)). Further evidence that the capability of ®broblasts to withstand oxidative DNA damage was in¯uenced by aging was obtained by measuring DNA strand breaks by the COMET assay, consisting of single cell electrophoresis after DNA solubilization (Singh et al., 1988). Fig. 4 reports DNA damage evaluated with this technique at steady state levels, after treatment with 50±200 mM H2O2, and after 60 min recovery in fresh medium. It can be seen that young/21 PDs
652
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
Fig. 3. Effects of aging on 8-OHdG repair after treatment of ®broblasts with H2O2. Cells were treated for 30 min with the indicated concentration of H2O2, the medium was washed out, and cells were re-incubated in fresh medium for up to 120 min. 8-OHdG was assayed before and after H2O2 treatment at the indicated times. (A) young/21 PDs ®broblasts; (B) old/58 PDs ®broblasts. Data are taken from duplicate experiments with .90% agreement.
®broblasts displayed lower steady state levels of DNA damage than old/58 PDs counterparts. DNA damage increased proportionally to H2O2 concentrations, but less extensively in young/21 PDs than in old/58 PDs ®broblasts (cfr. panels A and B). After 60 min recovery from H2O2 treatment, DNA damage returned to pre-treatment steady state levels in young/21 PDs ®broblasts, regardless of the concentration of H2O2. In contrast, old/58 PDs ®broblasts retained levels of DNA damage, which always exceeded pre-treatment levels, and increased proportionally to the concentration of H2O2. These results con®rmed those obtained by immunohistochemical analysis of 8-OHdG, and suggested that old ®broblasts not only were more susceptible to oxidative
stress but also displayed a decreased capability to repair DNA damage. 3.3. Effect of chronic exposure to low doses of H2O2 Because oxidative stress is an unavoidable consequence of aerobic metabolism, these results also implied that the higher levels of DNA damage observed in old cells could have been caused by a prolonged exposure to oxidative stress. To probe this hypothesis, ®broblasts between 21 and 58 PDs were exposed chronically to very low concentrations of H2O2 (5 mM H2O2), mimicking a prolonged exposure to spontaneous conditions of oxidative stress. As shown in Table 1, this treatment induced a signi®cant
Fig. 4. Single cell electrophoresis evaluation of oxidative DNA damage in young/21 PDs and old/58 PDs ®broblasts. Cultures of young/21 PDs (A) and old/58 PDs ®broblasts (B) were assayed for DNA damage by single cell electrophoresis at steady state (black columns), after 30 min incubation with 50±200 mM H2O2 (white columns), and after additional 60 min recovery in fresh medium (dashed columns). Data are expressed as the tail/nucleus ratios, and are taken from duplicate experiments with .90% agreement.
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
653
Table 1 Effect of chronic H2O2 treatment on 8-OHdG and SA-b-galactosidase in WI-38 ®broblasts at various passages in culture (*P , 0.05 vs control. Mean ^ SE from 3 to 8 experiments) PDs
21 41 58 a b c
8-OHdG a
SA-b-GAL b
Control
1 H2O2 c
Control
1 H2O2
107 ^ 11 120 ^ 6 173 ^ 7
116 ^ 3 153 ^ 7* 185 ^ 6
2.7 ^ 0.5 14.0 ^ 0.7 52.1 ^ 1.1
9.4 ^ 0.8* 36.0 ^ 2.0* 60.4 ^ 2.7
Fig. 5. Effect of chronic treatment with H2O2 on the expression levels of p21 in young to old WI-38 ®broblasts. p21 was detected by Western blot in young/21, middle age/41, and old/58 PDs ®broblasts, grown in the presence (1) or absence (2) of chronically added 5 mM H2O2. b-Actin served as control. Other conditions are given under Section 2.
OD Percentage of positive cells. H2O2 (5 mM) was delivered to cells as described under Section 2.
characterized the effects of low H2O2 on the cell cycle distribution of young, middle age and old ®broblasts. As shown in Table 2, old/58 PDs ®broblasts exhibited the highest percentage of cells in G0/G1. Chronic treatment with low H2O2 had no effect on the cell cycle distribution of young/21 PDs or old/58 PDs ®broblasts, but increased the percentage of middle age/41 PDs in G0/G1 (76.9 vs 54.9 in control cells). The latter increase was accompanied by a reduced percentage of cells in S phase (22.2 vs 32%) or in G2/M phase (0.9 vs 13.1%) and coincided with the greatest increases in the levels of 8-OHdG and SA-bgalactosidase positive cells (cf. Table 1). From these data it clearly appears that H2O2 treatment induced early signs of growth arrest in 41 PDs. Previous studies have shown that growth arrest of senescent cells was accompanied by up-regulation of p21, a negative regulator of cyclin E-CDK2 complexes (Stein and Dulic, 1995). In agreement with this report, we found that old/58 PDs ®broblasts exhibited higher levels of p21 expression than young/21 or middle age/ 41 PDs counterparts (Fig. 5). Moreover, we found that chronic treatment from 11 to 58 PDs ®broblasts with low concentrations of H2O2 increased p21 expression in young/21 PDs and, even more pronouncedly, in
increase of 8-OHdG in middle age/41 PDs ®broblasts when compared to the corresponding untreated cells (P , 0.05). This increase in oxidative DNA damage rendered middle age/41 PDs ®broblasts more similar to old/58 PDs counterparts. Thus, chronic H2O2 treatment seemed to accelerate the aging of ®broblasts. We therefore assayed for SA-b-galactosidase, an enzyme whose expression and activity are induced by H2O2 and associate with in vitro proliferative senescence (Dimri et al., 1995; Severino et al., 2000). As also shown in Table 1, the percentage of SA-b-galactosidase positive cells did increase with the age of ®broblast cultures (2.7, 14.0 and 52.1% in cells at 21, 41 and 58 PDs, respectively). Chronic treatment with `low' H2O2 increased SA-b-galactosidase positivity in all cells examined, but the increase was more pronounced in 21 and 41 PDs ®broblasts as compared to 58 PDs ®broblasts. These data con®rmed that a prolonged exposure to oxidative stress, induced by chronic treatment with low concentrations of H2O2, accelerated the process of aging in ®broblasts. To obtain correlations between H2O2-induced DNA damage and senescence-associated growth arrest we
Table 2 Effect of chronic treatment with H2O2 on cell cycle distribution in WI-38 ®broblasts at different passages in culture (*P , 0.005 vs controls (n 3)) 21 PDs
Control 1 H2O2 a a
41 PDs
58 PDs
% G0/G1
S
G2/M
G0/G1
S
G2/M
G0/G1
S
G2/M
55.0 ^ 1.2 53.9 ^ 1.2
34.6 ^ 1.8 34.5 ^ 3.8
10.4 ^ 2.0 11.6 ^ 3.5
54.9 ^ 1.8 76.9 ^ 0.9*
32.0 ^ 1.3 22.2 ^ 1.0*
13.1 ^ 1.9 0.9 ^ 0.1
78.2 ^ 2.5 79.8 ^ 1.8
17.3 ^ 2.8 16.0 ^ 0.5
4.5 ^ 0.3 4.2 ^ 0.8
H2O2 (5 mM) was delivered to cells as described under Section 2.
654
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
Table 3 Effect of chronic treatment with H2O2 on the DT of WI-38 ®broblasts at various passages in culture (*P , 0.0005; **P , 0.005 vs controls. Means ^ SE of three separate experiments) PDs
21 41 58 a
Doubling time (h) Control
H2O2 a
54.2 ^ 0.1 63.4 ^ 0.2 97.3 ^ 4.0
40.3 ^ 0.2* 55.7 ^ 1.0** 86.9 ^ 2.9
H2O2 (5 mM) was delivered to cells as described under Section 2.
middle age/41 PDs cells (Fig. 5). Concerning the mechanism of H2O2-induced senescence, it has been shown that the effect of H2O2 largely depends on the cell type, and on the dose and duration of treatment, resulting in biological responses which may vary from cell proliferation (at lower concentrations), to growth arrest and apoptosis or necrosis (at higher concentrations) (reviewed by Davies (1999)). To characterize how chronic/low dose H2O2 treatment accelerated the aging of WI-38 ®broblasts, we measured the effect of such treatment on the proliferation rate of these cells. As reported in Table 3, the DT decreased in all cells treated with H2O2, but especially in young/21 PDs ®broblasts (226% vs 212 or ±11% in middle age/ 41 PDs or old/58 PDs ®broblasts, respectively). These results suggested that H2O2 treatment was able to accelerate cell proliferation, particularly at the earlier passages, thereby inducing a more rapid progression to the so-called replicative senescence.
4. Discussion We have shown that the steady state levels of 8OHdG increase in WI-38 human ®broblasts at different passages in culture. These data con®rm and extend previous observations, and lend support to the concept that oxidative DNA damage may be responsible for the appearance of senescent phenotype in vitro as well as in vivo (von Zglinicki et al., 2001; Hamilton et al., 2001). In this study we have shown that aging may in¯uence also the vulnerability of WI-38 human ®broblasts to an acute oxidative stress, mediated by sublethal concentrations of H2O2 (50±200 mM). In fact, old/58 PDs ®broblasts proved to be more sensitive
than young/21 PDs ®broblasts to the action of H2O2, and hence accumulated more DNA damage measured by either 8-OHdG or COMET assays (cf. Figs. 2 and 4). a-Tocopherol, a powerful antioxidant, prevented H2O2-induced DNA damage in old ®broblasts but not in the more H2O2-resistant young ®broblasts. While consistent with the well known ef®cacy of vitamin E in preventing H2O2-induced elevation of free radicals in the cell and formation of DNA single strand breaks (Covacci et al., 2001; Duthie et al., 1996), these results suggest that old cells may have less antioxidant defences than their young counterparts (see also Fig. 2 and cf. Hall et al., 2001). Other antioxidants, like N-tbutyl hydroxylamine, similarly retard spontaneous or H2O2-accelerated senescence and DNA damage in cultured ®broblasts (Atamna et al., 2000a). Conversely, ®broblasts grown in 5% O2 gain more PDs and display lower oxidative damage than ®broblasts grown in 20% O2 (Chen et al., 1995). Thus, several lines of evidences concur in demonstrating that oxidative DNA damage is a clue event in senescence. DNA repair mechanisms similarly may be less ef®cient in old ®broblasts; in fact, a complete return of 8-OHdG to pre-treatment levels was observed in young/21 PDs ®broblasts but not in old/58 PDs ®broblasts (cf. Figs. 3 and 4). In accordance to our observation, an age-dependent decline was shown in the activity of the glycosylase that removes methylated bases in ®broblasts and in human lymphocytes (Atamna et al., 2000b). The increased steady state levels of DNA damage observed in senescent ®broblasts might be the consequence of a chronically induced oxidative stress, such as that mediated by oxyradicals produced during mitochondrial respiration. Consistent with this interpretation, previous studies have shown that the aging of cultured ®broblasts was accelerated by hyperoxia (Saretzki et al., 1998) or energetic stress, induced by growing cells in hypertonic NaCl and resulting in increased oxygen consumption (Reichelt and Schachtschabel, 2001). Here we have shown that aging may be accelerated also by exposing ®broblasts to chronic treatment with low concentrations of H2O2 (5 mM). In fact, this treatment induced elevation of 8-OHdG and expression of SA-b-galactosidase activity, another marker of in vitro senescence, in middle age/41 PDs ®broblasts, rendering them more similar to old/58 PDs ®broblasts (cf. Table 1). More importantly,
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
chronic treatment with low H2O2 was able to anticipate at 41 PDs the G0/G1 growth arrest usually observed at 58 PDs, an effect which was accompanied by consistent expression of a negative regulator of cell cycle progression like p21 (Stein and Dulic, 1995). The latter effects underscore the importance of DNA oxidative damage as a trigger of molecular events leading to growth arrest (Chen et al., 1998; Stein and Dulic, 1995; Itahana et al., 2001). Interestingly, chronic treatment with low H2O2 seems to accelerate the aging of 21 PDs and especially of 41 PDs cells while also enhancing their proliferation, an effect evidenced by a decreased DT (cf. Table 3). On the one hand, these results lend support to earlier evidence for H2O2-induced proliferation, involving activation of MAP kinase-mediated signal transduction pathways and transcription of early response genes like c-fos and AP-1 (Choi et al., 1995; Kim et al., 2001). On the other hand, the observation that increased proliferation coincides with an early senescent phenotype suggests that WI-38 ®broblasts react to chronic treatment with H2O2 by reaching growth arrest in a shorter time. In conclusion, we have shown that DNA oxidative damage is an important determinant of aging in WI-38 ®broblasts. This information has been obtained by exposing ®broblast to very low concentrations of H2O2 and by prolonging treatment from 11 to 21 or 41 or 58 PDs, so as to reproduce conditions in which cells accumulate damage during their life span in response to chronic exposure to oxidative challenges. Measurements of oxidative DNA damage could therefore be exploited as molecular markers of aging, in addition to those identi®ed by others in cells exposed to sub-lethal concentrations of H2O2 (®bronectin, apolipoprotein J, osteonectin, and SM22 mRNA) (Frippiat et al., 2001). We have also shown that DNA damage might be the consequence of a reduced ability of senescent cells to withstand oxidative insults. Whereas other mechanisms might contribute to the effect of H2O2 in accelerating aging, these results identify antioxidant interventions as potential new strategies to prevent DNA damage and retard the progression of aging. Acknowledgements Work co-®nanced by MURST ex 60% and
655
MURSTCOFIN # 9906241379. We are indebted to Dr R.M. Santella (Columbia School of Public Health, New York) for providing us with monoclonal antibody 1F7 against 8-OHdG. References Atamna, H., Paler-Martinez, A., Ames, B.N., 2000a. N-t-Butyl hydroxylamine, a hydrolysis product of a-phenyl-N-t-butyl nitrone, is more potent in delaying senescence in human lung ®broblasts. J. Biol. Chem. 275, 6741±6748. Atamna, H., Cheung, I., Ames, B.N., 2000b. A method for detecting abasic sites in living cells: age-dependent changes in base excision repair. Proc. Natl. Acad. Sci. USA 97, 686±691. Beckman, K.B., Ames, B.N., 1998. The free radical theory of aging matures. Physiol. Rev. 78, 547±581. Bonelli, M.A., Al®eri, R.R., Petronini, P.G., Brigotti, M., Campanini, C., Borghetti, A., 1999. Attenuated expression of 70-kDa heat shock protein in WI-38 ®broblasts during aging in vitro. Exp. Cell Res. 252, 20±32. Chen, Q., Fischer, A., Reagan, J.D., Yan, L.J., Ames, B.N., 1995. Oxidative DNA damage and senescence of human diploid ®broblast cells. Proc. Natl. Acad. Sci. USA 92, 4337±4341. Chen, Q.M., Bartholomew, J.C., Campisi, J., Acosta, M., Reagen, J.D., Ames, B.N., 1998. Molecular analysis of H2O2-induced senescent-like growth arrest in normal human ®broblasts: p53 and Rb control of G1 arrest but not cell replication. Biochem. J. 332, 43±50. Choi, A.M., Pignolo, R.J., apRhys, C.M., Cristofalo, V.J., Holbrook, N.J., 1995. Alteration in the molecular response to DNA damage during cellular aging of cultured ®broblasts: reduced AP-1 activation and collagenase gene expression. J. Cell. Physiol. 164, 65±73. Collins, A., Cadet, J., Epe, B., Gedik, C., 1997. Problems in the measurement of 8-oxoguanine in human DNA. Report of a workshop, DNA oxidation, held in Aberdeen, UK, 19±21 January, 1997. Carcinogenesis 18, 1833±1836. Covacci, V., Torsello, A., Palozza, P., Sgambato, A., Romano, G., Boninsegna, A., Cittadini, A., Wolf, F.I., 2001. DNA oxidative damage during differentiation of HL-60 human promyelocytic leukemia cells. Chem. Res. Toxicol. in press. Cristofalo, V.J., Pignolo, R.J., 1993. Replicative senescence of human ®broblast-like cells in culture. Physiol. Rev. 73, 617± 635. Davies, K.J., 1999. The broad spectrum of responses to oxidants in proliferating cells: a new paradigm for oxidative stress. IUBMB Life 48, 41±47. 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., Medrano, E.E., Linskens, M., Rubelj, I., Pereira-Smith, O., Peacocke, M., Campisi, J., 1995. A biomarker that identi®es senescent human cells in culture and in aging in vivo. Proc. Natl. Acad. Sci. USA 92, 9363±9367. Duthie, S.J., Ma, A., Ross, M.A., Collins, A.R., 1996. Antioxidant
656
F.I. Wolf et al. / Experimental Gerontology 37 (2002) 647±656
supplementation decreases oxidative DNA damage in human lymphocytes. Cancer Res. 56, 1291±1295. Frippiat, C., Chen, Q.M., Zdanov, S., Magalhaes, J.P., Remacle, J., Toussaint, O., 2001. Subcytotoxic H2O2 stress triggers a release of transforming growth factor-beta 1, which induces biomarkers of cellular senescence of human diploid ®broblasts. J. Biol. Chem. 276, 2531±2537. Goodman, L., Stein, G.H., 1994. Basal and induced amounts of interleukin-6-mRNA decline progressively with age in human ®broblasts. J. Biol. Chem. 269, 19,250±19,255. Hall, D.M., Sattler, G.L., Sattler, C.A., Zahng, H.J., Oberley, L.W., Pitot, H.C., Kregel, K.C., 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 Richardson, A., 2001. Does oxidative damage to DNA increase with age? Proc. Natl. Acad. Sci. USA 98, 10,469±10,474. Hosokawa, M., Fujisawa, H., Ax, S., Zahn-Daimler, G., Zahn, R.K., 2000. Age-associated DNA damage is accelerated in senescence-mice. Mech. Ageing Dev. 118, 61±70. Itahana, K., Dimri, G., Campisi, J., 2001. Regulation of cellular senescence by p53. Eur. J. Biochem. 268, 2784±2791. Johnson, T.M., Yu, Z.X., Ferrans, V.J., Lowenstein, R.A., Finkel, T., 1996. Reactive oxygen species are downstream mediators of p53-dependent apoptosis. Proc. Natl. Acad. Sci. USA 93, 11,848±11,852. Kawanishi, S., Hiraku, Y., Oikawa, S., 2001. Mechanism of guanine-speci®c DNA damage by oxidative stress, role in carcinogenesis and aging. Mutat. Res. 488, 65±76. Kim, B.J., Ham, M.J., Chung, A.S., 2001. Effects of reactive oxygen species on proliferation of Chinese hamster lung ®broblast (V79) cells. Free Rad. Biol. Med. 30, 686±698. Klungland, A., Rosewell, I., Hollenbach, S., Larsen, E., Daly, G., Epe, B., Seebeg, E., Lindahl, T., Barnes, D., 1999. Accumulation of premutagenic DNA lesions in mice defective in removal of oxidative base damage. Proc. Natl. Acad. Sci.USA 96, 13,300±13,305.
Kohn, K.W., 1991. Principles and practice of DNA ®lter elution. Pharmacol. Ther. 49, 55±77. Lindahl, T., Wood, R.D., 1999. Quality control by DNA repair. Science 286, 1897±1905. Marnett, L.J., 2000. Oxyradicals and DNA damage. Carcinogenesis 21, 361±370. Ottender, M., Lutz, W.K., 1999. Correlation of DNA adduct levels with tumor incidence: carcinogenic potency of DNA adducts. Mutat. Res. 424, 237±247. Reichelt, J., Schachtschabel, D.O., 2001. Energetic stress induces premature aging of diploid human ®broblasts (Wi-38) in vitro. Arch. Gerontol. Geriatr. 32, 219±231. Romano, G., Sgambato, A., Mancini, R., Capelli, G., Giovagnoli, M.R., Flamini, G., Boninsegna, A., Vecchione, A., Cittadini, A., 2000. 8-Hydroxy-2 0 -deoxyguanosine in cervical cells: correlation with grade of displasia and human papillomavirus infection. Carcinogenesis 21, 1143±1147. Saretzki, G., Feng, J., von Zglinicki, T., Villeponteau, B., 1998. Similar gene expression pattern in senescent and hyperoxic-treated ®broblasts. J. Gerontol. A Biol. Sci. Med. Sci. 53, B438±B442. Severino, J., Allen, R.G., Balin, S., Balin, A., Cristofalo, V.J., 2000. Is b-galactosidase staining a marker of senescence in vitro and in vivo?. Exp. Cell Res. 257, 162±171. 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. Stein, G.H., Duliñ, V., 1995. Origins of G1 arrest in senescent human ®broblasts. Bioessay 17, 537±543. Xu, A., Wu, L-J., Santella, R.M., Hei, T.K., 1999. Role of oxyradicals in mutagenicity and DNA damage induced by crocidolite asbestos in mammalian cells. Cancer Res. 59, 5922±5926. Yarborough, A., Zhang, Y.J., Hsu, T.M., Santella, R.M., 1996. Immunoperoxidase detection of 8-hydroxydeoxyguanosine in a¯atoxin B1-treated rat liver and human oral mucosal cells. Cancer Res. 56, 683±688. von Zglinicki, T., Burkle, A., Kirkwood, T.B., 2001. Stress DNA damage and ageingÐan integrative approach. Exp. Gerontol. 36, 1049±1062.