DNA damage and apoptosis in hydrogen peroxide-exposed Jurkat cells: bolus addition versus continuous generation of H2O2

Free Radical Biology & Medicine, Vol. 33, No. 5, pp. 691–702, 2002 Copyright © 2002 Elsevier Science Inc. Printed in the USA. All rights reserved 0891-5849/02/$–see front matter

PII S0891-5849(02)00967-X

Original Contribution DNA DAMAGE AND APOPTOSIS IN HYDROGEN PEROXIDE-EXPOSED JURKAT CELLS: BOLUS ADDITION VERSUS CONTINUOUS GENERATION OF H2O2 ALEXANDRA BARBOUTI, PASCHALIS-THOMAS DOULIAS, LAMBROS NOUSIS, MARGARITA TENOPOULOU, DIMITRIOS GALARIS

and

Laboratory of Biological Chemistry, University of Ioannina Medical School, Ioannina, Greece (Received 29 January 2002; Accepted 24 May 2002)

Abstract—Aspects of the molecular mechanism(s) of hydrogen peroxide-induced DNA damage and cell death were studied in the present investigation. Jurkat T-cells in culture were exposed either to low rates of continuously generated H2O2 by the action of glucose oxidase or to a bolus addition of the same agent. In the first case, steady state conditions were prevailing, while in the latter, H2O2 was removed by the cellular defense systems following first order kinetics. By using single-cell gel electrophoresis (also called comet assay), an initial increase in the formation of DNA single-strand breaks was observed in cells exposed to a bolus of 150 ␮M H2O2. As the H2O2 was exhausted, a gradual decrease in DNA damage was apparent, indicating the existence of an effective repair of single-strand breaks. Addition of 10 ng glucose oxidase in 100 ␮l growth medium (containing 1.5 ⫻ 105 cells) generated 2.0 ⫾ 0.2 ␮M H2O2 per min. This treatment induced an increase in the level of single-strand breaks reaching the upper limit of detection by the methodology used and continued to be high for the following 6 h. However, when a variety of markers for apoptotic cell death (DNA cell content, DNA laddering, activation of caspases, PARP cleavage) were examined, only bolus additions of H2O2 were able to induce apoptosis, while the continuous presence of this agent inhibited the execution of the apoptotic process no matter whether the inducer was H2O2 itself or an anti-Fas antibody. These observations stress that, apart from the apparent genotoxic and proapoptotic effects of H2O2, it can also exert antiapoptotic actions when present, even at low concentrations, during the execution of apoptosis. © 2002 Elsevier Science Inc. Keywords—Hydrogen peroxide, Apoptosis, Comet assay, PARP, Single-strand breaks, Glucose oxidase, Jurkat cells, Free radicals

INTRODUCTION

activate several transcriptional factors with consequent expression of a great number of genes [8,9], to provoke cell proliferation and differentiation [3,10,11], and finally to induce cell death either by apoptosis or necrosis [12–14]. Cellular DNA is especially sensitive to the action of H2O2 and this DNA damage is widely believed to be mediated by transition metal ions, mainly iron and/or copper, which are able to catalyze the formation of hydroxyl radicals (HO•) by Fenton-type reactions [15– 17]. The location of redox-active metals is likely to be of utmost importance for the ultimate effect because HO•, due to their extreme reactivity, interact exclusively in the vicinity of the bound metal [18]. Formation of HO• close to DNA (due to bound Fe or Cu ions) results in its damage, including base modifications, single- and double-strand breakage, and sister chromatid exchange [15–

Reactive oxygen species (ROS) are continuously generated in vivo but increases in their steady states are regarded to be responsible for a variety of pathological conditions, including cardiovascular disease, cancer, and aging [1,2]. Among a great variety of ROS, hydrogen peroxide (H2O2) plays a pivotal role because it is generated from nearly all sources of oxidative stress and can diffuse freely in and out of cells and tissues. Moreover, it has been shown that H2O2 has the ability to modulate signal transduction pathways [3,4], to change the homeostasis of ions such as calcium and iron [5–7], to Address correspondence to: Dr. Dimitrios Galaris, University of Ioannina Medical School, Laboratory of Biological Chemistry, University Campus, P.O. Box 1186, 451 10 Ioannina, Greece; Tel: ⫹30 6510-97562; Fax: ⫹30 6510-97868; E-Mail: [email protected]. 691

692

A. BARBOUTI et al.

17]. However, there are indications that location of iron at positions other than DNA may contribute indirectly to DNA damage and ensuing apoptosis. Extensive work from Brunk’s group in Sweden has shown that lysosomal iron may be a key player in peroxide-dependent cell damage and apoptosis [19 –21]. Moreover, results from our laboratory as well as others, indicated that the formation of single-strand breaks in cellular DNA was Ca2⫹-dependent, indicating an obligatory intermediary role for Ca2⫹ and, consequently, a signaling pathway leading to DNA damage [22–25]. It is generally believed that extensive DNA damage leads invariably to cell death either by apoptosis or necrosis. In particular, H2O2 may induce both apoptosis and necrosis depending on the concentration of the oxidant employed and/or the type of the cell being studied [12–14,26,27]. It has to be noted, however, that in the vast majority of studies with H2O2 it was added directly to the cells as a bolus, so that cells were initially exposed to relatively high concentrations followed by a fast decrease as H2O2 is gradually consumed [28]. Consequently, if the mode of action of H2O2 is concentration dependent (as most probably is the case), the results might appear inconsistent. In vivo the rate of H2O2 generation, although different for various kinds of cells, is continuous, with the steady state levels fluctuating at nanomolar concentrations (10⫺8 to 10⫺7 M) [29]. Hence, exposing cells to a continuous flow of H2O2, as opposed to bolus additions, represents a superior method of delivery that mimics physiological conditions. In previous works from our laboratory, we described the short-term effects (up to 60 min exposure) of continuously generated H2O2 (by the action of added glucose oxidase) on DNA damage as assessed by “singlecell gel electrophoresis” or “comet assay” [23–25,30]. In the present study, we extended our previous observations by following the cells for longer times (6 h) and attempted to correlate the formation of single-strand breaks to apoptotic cell death. Contrary to the prevailing idea of the proapoptotic action of H2O2, the results of the present study indicate that H2O2, when continuously generated even at low concentrations, may exert antiapoptotic actions.

DTT (DL-dithiothreitol), pepstatin A, leupeptine, and Hoechst 33342 were from Sigma Chemical Co. (St. Louis, MO, USA). Fetal bovine calf serum, Nunc tissue culture plastics, low melting point agarose, PMSF (phenylmethyl sulfonyl fluoride), penicillin/streptomycin, and proteinase K were obtained from Gibco BRL (Grand Island, NY, USA). Normal melting point agarose was obtained from Serva GmbH (Heidelberg, Germany). Microscope glass super frosted slides were supplied by Menzel-Glaset (Menzel, Germany); 4,6-diamidine-2phenylindole dihydrochloride (DAPI) and RNase A were supplied by Boehringer Mannheim (Mannheim, Germany). Mouse monoclonal antibody raised against amino acids 764-1014 at the carboxy terminus of poly(ADPribose)polymerase (PARP) was from Santa Cruz Biotechnology, Inc. sc 8007 (Santa Cruz, CA, USA). Aprotinin was from Roche Diagnostics (Mannheim, Germany). Mouse anti-Fas monoclonal antibody (clone DX2) and Ac-DEVD-AMC (caspase-3 substrate) were obtained from Calbiochem (Schwalbach, Denmark). Ethidium bromide and H2O2 was from Merck (Darmstadt, Germany). All other chemicals used were of analytical grade. Cell culture and treatment Jurkat cells (ATCC, clone E6-1) were grown in RPMI-1640 containing 10% heat-inactivated fetal calf serum, 2 mM glutamine, 100 U/ml penicillin, and 100 ng/ml streptomycin, at 37°C in 5% CO2 in air. Cells in the log phase were harvested by centrifugation (250 ⫻ g, 10 min, room temperature), resuspended in a density of 1.5 ⫻ 106 cells per ml, and allowed to stay for 1.5 h under standard culturing conditions. Cells were then treated with H2O2 and/or glucose oxidase for the doses and times indicated. Finally, cells were collected and checked for Trypan blue exclusion before any further analysis. In order to assess whether the product of the reaction of glucose oxidase (D-glucono-␦-lactone) or the oxygen during the reaction affected the results, cells were incubated with both glucose oxidase and excess catalase. No significantly different results were observed (not shown).

MATERIALS AND METHODS

Single-cell gel electrophoresis (comet assay) Materials RPMI 1640 growth medium supplemented with Lglutamine, MTT (3-(4,5-dimethyl)-2,5-diphenyl tetrazolium bromide), SDS (sodium dodecyl sulfate), glucose oxidase (from Aspergillus niger, 18,000 units/g), catalase (from bovine liver), triton X-100, CHAPS (3-[(3-cholamidoproyl) dimethylammonio]-1 propanesulfonate),

One hundred microliters RPMI 1640 growth medium containing 1.5 ⫻ 105 Jurkat cells were placed into each of 96 wells of ELISA plastic plates and treated with glucose oxidase or H2O2 for the time periods indicated as described above. The comet assay performed in this work was essentially the same as previously described [23,31,32]. Cells

H2O2-induced DNA damage and apoptosis

were suspended in 1% low melting point agarose in PBS, pH 7.4, and pipetted on to super frosted glass microscope slides precoated with a layer of 1% normal melting point agarose (warmed to 37°C prior to use). The agarose was allowed to set at 4°C for 10 min and then the slides were immersed in lysis solution (2.5 M NaCl, 100 mM EDTA, 10 mM Tris at pH 10, 1% Triton X-100 v/v) at 4°C for 1 h in order to dissolve cellular proteins and lipids. Slides were then placed in single rows in a 30 cm wide horizontal electrophoresis tank containing 0.3 M NaOH and 1 mM EDTA, pH ⬎ 13, at 4°C for 40 min in order to allow for separation of the two DNA strands (alkaline unwinding). Electrophoresis was performed in the unwinding solution at 30 V (1 V/cm), 300 mAmps for 30 min. The slides were then washed three times for 5 min each with 0.4 M Tris, pH 7.5, at 4°C before staining with DAPI (5 mg/ml). Image analysis and scoring DAPI-stained nucleoids were examined under a UV microscope with an excitation filter of 435 nm and a magnification of 400. The damage was not homogeneous and visual scoring of the cellular DNA on each slide was based on characterization of 100 randomly selected nucleoids. The comet-like DNA formations were categorized into 5 classes (0, 1, 2, 3, and 4) representing an increasing extent of DNA damage seen as a “tail.” Each comet was assigned a value according to its class. Accordingly, the overall score for one hundred comets ranged from 0 (100% of comets in class 0) to 400 (100% of comets in class 4). In this way, the overall DNA damage of the cell population can be expressed in arbitrary units [23]. Visual scoring expressed in this way correlated near-linearly with other parameters such as percent of DNA in the tail estimated after computer image analysis using a specific software package (Comet Imager, MetaSystems, Altlussheim, Germany) (results not shown). The same linear correlation between visual scoring and computer image analysis has also been reported by other laboratories [33,34]. Visual observations and analyses of the results were always carried out by the same experienced person, using a specific pattern when moving along the slide.

693

bated for 5 min at room temperature and centrifuged before addition of 0.3 ml of staining solution (0.7 ␮g/ml propidium iodide (PI), 70 ␮g/ml RNAase A) for 30 min in the dark. Finally, 1.5 ⫻ 106 cells were suspended in 0.2 ml PBS and analyzed on a FACScan Becton Dickinson flow cytometer (Becton Dickinson, Mountain View, CA, USA). Extraction of DNA and fragmentation analysis After treatment, cellular DNA was isolated from 3 ⫻ 106 cells for each sample. Cells were collected and washed twice in cold PBS (450 ⫻ g, 4°C), resuspended in 500 ␮l lysis buffer (50 mM Tris-HCl, pH 8.0, 100 mM EDTA, 1% SDS, 0.1 M NaCl), and proteinase K was added to a final concentration of 0.6 mg/ml. The cell suspension was allowed to digest overnight at 37°C. DNA samples were then extracted with an equal volume of phenol:cloroform (v/v, 1:1) for 1 h and centrifuged at 16,000 ⫻ g, for 10 min. This step was repeated two more times before the aqueous phase was mixed with 500 ␮l chloroform and centrifuged again for 5 min at 16,000 ⫻ g. The upper phase containing the DNA was then transferred to new eppendorf tubes and precipitated from the supernatant with one-tenth volume of 3 M sodium acetate, pH 5.2, and 2 volumes of ethanol at ⫺20°C overnight. The samples were then centrifuged at 16,000 ⫻ g for 15 min, the supernatant was removed, and 1 ml of 70% ethanol was added to the DNA pellet. The samples were centrifuged again at 16,000 ⫻ g for 15 min and the supernatant removed before the DNA pellet was allowed to dry at room temperature. After that, DNA was solubilized in Tris-EDTA buffer (10 mM Tris-Cl, 1mM EDTA, pH 8.0) containing boiled RNAase (1.7 mg/ml). The samples were incubated for 2 h at 37°C. The resulted solution of DNA was quantitated spectrophotometrically at 260/280 nm and mixed with loading buffer (0.02% bromophenol blue, 40% glycerol in Tris, boric acid, EDTA (89:89:2, pH 8.0)) before being loaded in 1.8% agarose gel containing 0.5 ␮g/ml ethidium bromide. After electrophoresis, gels were illuminated with ultraviolet light for examination and photography. Measurement of caspase-3-like activity

Flow cytometric analysis of cellular DNA content For flow cytometric analysis, cells cultured and treated by H2O2 or glucose oxidase as described above, were fixed overnight in 70% ice-cold ethanol at 4°C. After one washing, cells were resuspended in 0.5 ml PBS and 1 ml of DNA extraction buffer (192 mM Na2HPO4, 4 mM citric acid, pH 7.8) was added. Cells were incu-

In order to measure caspase-3-like activity, cells (7.5 ⫻ 106 cells) were lysed by incubation in lysis buffer (10 mM HEPES/KOH, pH 7.4, 2 mM EDTA, 0.1% CHAPS, 5 mM dithiothreitol, 1 mM phenylmethyl sulfonyl fluoride (PMSF), 10 ␮g/ml pepstatin A, 20 ␮g/ml leupeptin, and 10 ␮g/ml aprotinin) for 20 min in ice. Lysates were centrifuged at 14,000 ⫻ g for 20 min at 4°C. In order to estimate caspase activity, cell lysates

694

A. BARBOUTI et al.

(150 ␮g protein) were incubated for 1 h in 1 ml reaction buffer at 37°C (100 mM HEPES, pH 7.5, 10% sucrose, 0.1% CHAPS, and 10 mM DTT) containing 100 ␮M of the fluorogenic peptide substrate Ac-DEVD-AMC. The fluorescence intensity of the released 7-amino-4methylcoumarin (AMC) was measured by using a fluorospectrophotometer with excitation at 380 nm and emission at 460 nm. Western blotting Cleavage of the enzyme poly(ADP-ribose)polymerase (PARP) from 116 kDa to an 85 kDa part was observed by using Western blotting analysis. Briefly, 4.5 ⫻ 106 cells were washed twice in cold PBS and resuspended in a lysis buffer (20 mM Tris-Cl, pH 7.5, 1% SDS, 2 mM EDTA, 2 mM EGTA, 5 mM DTT, 1 mM PMSF,10 ␮g/ml pepstatin, 10 ␮g/ml aprotinin, and 20 ␮g/ml leupeptin) and incubated for 10 min on ice. The mixture was sonicated (3 ⫻ 10 s) at 4°C. For immmunoblotting analysis, 30 ␮g of protein were applied on 8% SDSpolyacrylamide gel electrophoresis and transferred to nitrocellulose membranes by electroblotting. After blocking with 5% nonfat milk, membranes were incubated with a mouse monoclonal anti-PARP antibody (sc 8007, Santa Cruz Biotechnology, Inc.) followed by horse radish peroxidase-conjugated secondary antibody. Membranes were developed using the ECL reagent. MTT cell survival assay The MTT assay was used as previously described [35] to estimate cell viability. This method is based on the ability of viable cells to reduce MTT (3-(4,5-dimethyl)2,5-diphenyl tetrazolium bromide) and form a blue formazan product. In brief, cells were treated by H2O2 or glucose oxidase as described above before the addition of MTT solution (stock solution of 5 mg/ml in RPMI without phenol red in the dark and filtered through a 0.2 ␮m filter before use). The final concentration of MTT was 0.2 mg/ml. The cells were left for 4 h at 37°C, followed by the addition of 80 ␮l 10% SDS to each well to dissolve the cells; then the cells were left for another 8 h. Finally, plates were vigorously shaken for 1 min and the optical density of each well measured using an automatic plate reader with a 550 nm test wavelength and a 690 nm reference wavelength. Measurement of hydrogen peroxide generation The amount of hydrogen peroxide generated by the action of glucose oxidase in PBS containing 5.0 mM glucose was estimated either by following the increase in

Fig. 1. Formation of single-strand breaks after exposure of Jurkat cells to H2O2. Cells (1.5 ⫻ 106 per ml) in complete growth medium, containing 10% fetal calf serum, were exposed to either a bolus concentration of 150 ␮M H2O2 (䊐), or continuously generated H2O2 by the action of 100 ng glucose oxidase per ml (‚), while control cells were treated with the same volume of PBS (〫). After initiation of the treatment and at the time points indicated, the cells were centrifuged and the levels of single-strand breaks formed were estimated by comet assay and expressed as arbitrary units as described in Materials and Methods. Every point represents duplicate measurements, the values of which did not differ more than 5%. This experiment was repeated two more times with essentially the same results.

the absorbance at 240 nm (Molar Extinction Coefficient ⫽ 43.6 M⫺1 cm⫺1), or polarographically by using an oxygen electrode (Hansatech Instruments, Norfolk, UK) detecting the liberation of O2 following the addition of excess catalase. Protein determination Protein concentrations were determined by the Bradford method, using bovine serum albumin as a standard. Statistical analysis A Student’s paired t-test was used in order to examine statistically significant differences. RESULTS

We have shown previously that H2O2 rapidly and efficiently induces formation of single-strand breaks in the nuclear DNA of exposed cells [23–25,30]. In the experiment shown in Fig. 1, the formation of singlestrand breaks in nuclear DNA was followed by longer incubation time intervals (up to 6 h) after exposure of the

H2O2-induced DNA damage and apoptosis

695

Fig. 2. Flow cytometric analysis of H2O2-induced apoptosis. Cells (1.5 ⫻ 106 per ml) were treated for 6 h with: PBS (A), 150 ␮M H2O2 (B), 100 ng glucose oxidase per ml (C), and 150 ␮M H2O2 plus 0.1 ␮g glucose oxidase per ml (D). At the end of the treatment, cellular DNA was stained by propidium iodide (0.7 ␮g/ml) and the cellular DNA content was analyzed on a FACScan Becton Dickinson flow cytometer as described in Materials and Methods. Cells containing less DNA than that of G0/G1 phase were regarded as apoptotic. These experiments were repeated several times with essentially the same results.

cells (1.5 ⫻ 105 per 100 ␮l) to either a bolus of 150 ␮M H2O2 or glucose oxidase (100 ng/ml, generating 2.02 ⫾ 0.2 ␮M H2O2 per min). In the first case, the DNA damage, as indicated by tail formation after electrophoresis (comet assay), was initially increased during the first 15 min of incubation but subsequently decreased again as H2O2 was removed from the medium by cellular defense mechanisms. However, the new steady state level of DNA damage did not reach the initial control values indicating possibly the formation of irreversible lesions. In a separate experiment, it was shown that the concentration of H2O2, under these experimental conditions, was continuously decreased by first-order kinetics (half-life about 9 min) and was not apparent about 60 min after its addition (results not shown). In contrast, continuous generation of H2O2 produced a fast increase in the number of single-strand breaks, reaching the upper limit of detection for this particular method (Fig. 1). It has to be noted that 4 to 5 h after the initial exposure of the cells to continuously generated H2O2, the DNA in many cells was completely dispersed in small parts without any organized structure, indicating total disintegration of the cellular DNA by this treatment. It is obvious

from the above results that the kinetics of single-strand break formation differ, depending on the way of exposure (bolus addition vs. continuous generation) to H2O2, with more profound DNA effects in the case of continuous generation. In spite of the damaged DNA observed above, when Jurkat cells were exposed to either a bolus addition (150 ␮M H2O2) or glucose oxidase (100 ng/ml) and the effects on cellular DNA content were examined 6 h later by flow cytometry (propidium iodide staining of DNA), characteristic changes indicating an apoptotic process (lower DNA content than that of G0/G1 phase) was apparent only in the case of bolus addition of H2O2 (32.4% apoptotic cells compared to 10.8% of control) (Figs. 2A and 2B). Surprisingly, cells exposed to continuously generated H2O2 did not undergo such DNA changes (12.0%) (Fig. 2C), although a high degree of DNA damage was observed under the same conditions by comet assay (Fig. 1). Moreover, when cells were exposed simultaneously to bolus addition plus continuous generation, the number of apoptotic cells was decreased significantly (14.9% apoptotic cells compared to 32.4% and 10.8% of bolus-treated and control cells,

696

A. BARBOUTI et al.

Fig. 3. Ladder pattern formation after electrophoresis of DNA from H2O2-exposed cells. Cells (1.5 ⫻ 106 per ml) were treated with increasing concentrations of H2O2 (A) or increasing amounts of glucose oxidase (B) for 6 h. At the end of the treatment, DNA was isolated from 3 ⫻ 106 cells per sample and the isolated DNA was electrophoresed in 1.8% agarose gels as described in Materials and Methods. After electrophoresis, gels were illuminated with ultraviolet light for examination and photography. The numbers on the lanes represent: (A) 1 ⫽ nontreated cells; 2 ⫽ 30; 3 ⫽ 50; 4 ⫽ 70; 5 ⫽ 100; 6 ⫽ 150; 7 ⫽ 250; 8 ⫽ 500; and, 9 ⫽ 1000 ␮M H2O2; (B) 1 ⫽ molecular weight markers; 2 ⫽ control, nontreated cells; 3 ⫽ 1; 4 ⫽ 10; 5 ⫽ 25; 6 ⫽ 50; 7 ⫽ 75; 8 ⫽ 100; 9 ⫽ 200; 10 ⫽ 500; and, 11 ⫽ 750 ng glucose oxidase per ml. These experiments were repeated one more time with the same results.

respectively) (Fig. 2D), indicating an inhibition of apoptosis when H2O2 is continuously present, even at low concentrations, during the apoptotic process. The conclusion drawn above was further substantiated by the experiment shown in Fig. 3. Treatment of the cells with increasing concentrations of H2O2 (30 to 1000 ␮M) led to the appearance of a ladder pattern in agarose gel electrophoresis of extracted DNA, at concentrations higher than 100 ␮M and reaching a maximum at 250 to 500 ␮M H2O2 (Fig. 3A, lanes 6 – 8). No such pattern appeared in the case of continuous generation (1 to 750 ng glucose oxidase per ml, generating H2O2 in a range

from about 0.02 to 15 ␮M per min) (Fig. 3B), further supporting the inability of continuously generated H2O2 to induce apoptosis although the same treatment was much more efficient to induce formation of single-strand breaks (Fig. 1). Interestingly, when the addition of H2O2 was followed by glucose oxidase treatment 1 or 2 h after initiation, the formation of a DNA ladder (observed 6 h later) was inhibited, indicating the ability of H2O2 to inhibit the apoptotic process at some point(s) after it has been initiated (Fig. 4A). However, when glucose oxidase was added 3 or 5 h after the bolus addition, the characteristic

H2O2-induced DNA damage and apoptosis

697

Fig. 4. Effects of H2O2 on H2O2- and anti-Fas-induced apoptosis. Cells (1.5 ⫻ 106 per ml) were treated either with 150 ␮M H2O2 (A) or 0.5 ␮g/ml anti-Fas mouse monoclonal antibody (B) for initiation of apoptosis. At the time points indicated after initiation, 0.1 ␮g glucose oxidase per ml was added in the growth medium. Six hours after initiation, cells (3 ⫻ 106 cells per sample) were harvested and DNA was isolated. Ten micrograms of the isolated DNA was electrophoresed in an 1.8% agarose gel as in Fig. 3. The lanes represent: (A) C ⫽ control, untreated cells; H2O2 ⫽ 150 ␮M H2O2; GO ⫽ 0.1 ␮g glucose oxidase per ml; 1 h, 2 h, 3 h, and 5 h represent DNA extracted from cells in which glucose oxidase (0.1 ␮g per ml) was added at the indicated times after the addition of 150 ␮M H2O2. (B) 123 bp ⫽ markers; C ⫽ control, untreated cells; anti-Fas ⫽ 0.5 ␮M anti-Fas mouse monoclonal antibody alone; 1 h, 2 h, 3 h, and 5 h represent the times of addition of glucose oxidase (0.1 ␮g per ml) after the initiation of apoptosis by the anti-Fas antibody. These experiments were repeated one more time with the same results.

ladder pattern was apparent, although not as intense as in the case of bolus addition alone. Essentially the same results were obtained when a mouse anti-Fas monoclonal antibody (0.5 ␮g/ml) was used as initiator of apoptosis (Fig. 4B). Addition of glucose oxidase (100 ng/ml) 1 or 2 h after initiation by the anti-Fas antibody resulted in

complete abrogation of the formation of ladder pattern. However, when the addition was made 3 or 5 h after initiation, the apoptotic process proceeded normally. These observations indicate that the presence of H2O2, even at low concentrations, is able to inhibit a step (common for H2O2- and anti-Fas-induced apoptosis) that

698

A. BARBOUTI et al.

Fig. 5. Effects of H2O2 on H2O2-induced activation of caspases. Cells (1.5 ⫻ 106 per ml) were treated with PBS alone (control) (⽧), 150 ␮M H2O2 (■), 100 ng/ml glucose oxidase (Œ), or simultaneous addition of 150 ␮M H2O2 and glucose oxidase (x). At the time points indicated, cell extracts from the various samples were prepared as described in Materials and Methods. The ability of these extracts to release 7-amino4methyl-coumarin (AMC) from the substrate (Ac-DEVD-AMC) was estimated fluorometrically with excitation at 380 nm and emission at 460 nm. Each value represents the mean from two different experiments that did not differ more than 10%.

takes place before the third hour after initiation. It has to be noted that inhibition of apoptosis by H2O2 has been reported previously [36 –38], but it remains unclear whether physiological sources of continuously generated H2O2 are able to inhibit apoptosis. Since consecutive activation of a number of caspases is thought to represent the final common pathway for the execution of apoptosis, the enzymatic activities of caspases following exposure of the cells to conditions exactly as above were measured. As shown in Fig. 5, treatment of cells with 150 ␮M H2O2 led to a significant increase in caspase-3-like activity after 3 h. The activity was steadily increased to reach 8-fold levels after 6 h. No increase was observed when cells were exposed to a constant rate of H2O2 generation (about 2.0 ␮M per min). Moreover, simultaneous addition of glucose oxidase with H2O2 resulted in inhibition of H2O2-induced caspase activation. In accordance with Fig. 4, continuous generation of H2O2 by glucose oxidase inhibited activation of caspase3-like proteases even when glucose oxidase was added 1 or 2 h after the initiation of the apoptotic process regardless of whether the activator was H2O2 or an anti-Fas monoclonal antibody (Figs. 6A and 6B). However, only a partial inhibition was observed when glucose oxidase was added 3 or 5 h later (Figs. 6A and 6B). These results indicate that the presence of H2O2 is able to inhibit a step (or steps) at or upstream of caspase-3 activation that takes place mainly before the third hour after initiation. One among several physiological substrates of the activated caspase-3 is the enzyme poly(ADP-ribose)

Fig. 6. Effects of H2O2 on H2O2- and anti-Fas-induced activation of caspases. Cells (1.5 ⫻ 106 per ml) were treated either with 150 ␮M H2O2 (A) or 0.5 ␮g/ml of the mouse anti-Fas monoclonal antibody (B) for initiation of apoptosis. At the time points indicated after initiation, 100 ng/ml glucose oxidase was added in the growth medium. Six hours after initiation, cell extracts (3 ⫻ 106 cells per sample) were prepared and caspase activation was estimated as described in Fig. 5. Each value represents the mean from two different experiments that did not differ more than 10%.

polymerase (PARP), which has been proposed to be involved in the cellular response to genetic damage [39]. This enzyme has a molecular weight of 116 kDa and is cleaved at a specific position by caspase-3, producing two parts with molecular weights of about 85 and 30 kDa [40]. Exposure of Jurkat cells (1.5 ⫻ 106 cells per ml) to increasing bolus concentrations of H2O2 led to cleavage of PARP (6 h after exposure) at H2O2 concentrations higher than 100 ␮M (Fig. 7A). On the contrary, in the case of exposure of the cells to increasing concentrations of continuously generated H2O2, no significant formation of the 85 kDa PARP product was occurred (Fig. 7B), further supporting the notion that when oxidizing conditions are prevailing, the caspase-3 activation is prevented. The various amounts of glucose oxidase added in this experiment generated H2O2 in a range from 0.02 to

H2O2-induced DNA damage and apoptosis

699

Fig. 7. Effects of H2O2 on poly(ADP-ribose)polymerase (PARP) cleavage. Cells (1.5 ⫻ 106 per ml) were exposed to: (A) increasing concentrations of H2O2 for 6 h, (B) increasing amounts of glucose oxidase for 6 h, and (C) 150 ␮M H2O2 for the time points indicated. Total cell extracts were prepared as described in Materials and Methods and detection of proteins that crossreact with a mouse monoclonal antibody against PARP were evaluated by Western blotting as described in Materials and Methods. Arrows at the right indicate the molecular weights for uncleaved 116 kDa and cleaved 85 kDa parts of PARP. Numbers 1–7 in (A) represent: control, 70, 150, 250, 500, 750, and 1000 ␮M H2O2, respectively. Numbers 1– 8 in (B) represent: control, 1, 10, 25, 75 100, 200, and 500 ng glucose oxidase per ml, respectively. Numbers 1–7 in (C) represent: total cell extract preparation at 0, 1, 2, 3, 4, 5, and 6 h after the addition of H2O2. These experiments were repeated one more time with essentially the same results.

15 ␮M per minute. The possibility that uncleaved PARP remains active consuming cellular NAD⫹ and contributing to depletion of ATP, which is needed for execution of the apoptotic process as previously proposed [37], is currently under investigation in our laboratory. In accordance with the time course of caspase activation (Fig. 5), Fig. 7C shows that the cleavage of PARP starts about 3 h after the addition of 150 ␮M H2O2 and the cleavage continued to increase progressively 6 h later. Cell membrane permeability, when evaluated by Trypan blue exclusion, was apparently unaffected during the periods of treatments described above (results not shown). However, effects on cell viability were apparent when estimated by the MTT assay, which is based on the ability of living cells to reduce MTT and form a blue formazan product. It is assumed that mitochondrial dehydrogenases are mainly responsible for the reduction of MTT. As shown in Fig. 8, addition of increasing concentrations of H2O2 (70, 100, and 150 ␮M H2O2 per 1.5 ⫻ 105 Jurkat cells) did not induce any significant impairment in the reducing ability of the cells for the first 2 h, while a small decrease, about 15%, was observed at 100 and 150 ␮M H2O2, 4 h after treatment (Fig. 8A). A significant and concentration-dependent decrease was observed 6 h after initiation of the H2O2 treatment. It has to be noted that at 100 and 150 ␮M H2O2 treatment, cells

showed characteristic signs of apoptosis (Figs. 3 and 7). When cells were exposed to continuously generated H2O2 by the action of glucose oxidase, a decreasing ability to reduce MTT was observed at 100 ng/ml of glucose oxidase after 4 h, while impaired reduction ability was observed in cells exposed to 50 and 100 ng/ml after 6 h exposure (Fig. 8B). The effects observed were insignificant at lower concentrations of glucose oxidase at all time points tested. These results indicate that continuously generated H2O2, although able to induce DNA damage and mitochondrial dysfunction, is unable to lead to apparent apoptotic cell death. DISCUSSION

It has been proposed that H2O2, at relatively low concentrations, is able to play important roles in signaling pathways [41– 43]. Moreover, since H2O2 is able to penetrate freely across cell membranes, it can transfer information to nearby cells or tissues acting in a paracrine fashion, similar to nitric oxide. However, although extensively studied, the exact molecular mechanisms underlying the mode of action of H2O2 remain elusive. A large number of molecules have been proposed or identified as direct or indirect targets of intracellular H2O2. Protein kinases and phosphatases, proteins containing

700

A. BARBOUTI et al.

Fig. 8. Estimation of cell viability by the MTT assay after exposure of cells to H2O2. Cells (1.5 ⫻ 106 per ml) cultured in complete growth medium at ELISA plates were exposed to either bolus concentrations of H2O2 (70, 100, and 150 ␮M) (A) or glucose oxidase (1, 5, and 10 ng per 100 ␮l) (B). At the time points indicated, the ability of the treated cells to reduce MTT to formazan was estimated as described in Materials and Methods and expressed as a percentage of control unexposed cells. Each value represents triplicate measurements ⫾ SD, *p ⬍ .001.

sulfhydryl groups or iron sulfur clusters, lipids, DNA, and others are among the potential targets [44,45]. At higher concentrations, it is generally believed that H2O2 induces cell death mainly by initiating the programmed cell death or apoptosis as it is usually called. This proposal is based mainly on experiments in which H2O2 is added directly as a bolus to the growth medium containing the cells. However, in most cases in vivo, cells are exposed to H2O2 continuously generated for periods of time. In this regard, experimental systems mimicking as closely as possible the situation in vivo are advantageous over bolus addition of H2O2 to cell suspensions [12,19, 23]. Under the experimental conditions used in this work (150 ␮M H2O2 to a cell suspension of 1.5 ⫻ 106 cells per ml), the concentration of H2O2 was continuously decreased by first-order kinetics (half-life about 9 min) and was not apparent about 60 min after its addition (results not shown). Cells continued to have a normal appearance

and only after about 3 to 5 h did they start to show the first signs of apoptosis as indicated by DNA cleavage, caspase activation, PARP cleavage, and so on (Figs. 2–7). There are two main questions arising from the above general observations: (i) what are the primary targets of the action of H2O2 during the initial phase, and (ii) what processes are taking place during the time between H2O2 elimination and the appearance of the earliest signs of apoptosis? Regarding the first question, it is plausible to imagine that DNA represents one of the most sensitive targets among all the basic cellular constituents. We have reported recently the formation of single-strand breaks a few minutes after exposure of Jurkat cells in suspension to H2O2 [23–25,30]. This rapid effect on cellular DNA, however, was mainly Ca2⫹ dependent, at least in low H2O2 concentrations, as indicated by the protection offered by the membrane-permeable Ca2⫹ chelator BAPTA-AM. This observation indicates the need for mediators in the process of H2O2-induced DNA damage and excludes the possibility of interaction of H2O2 (when present at low concentrations) with transition metals bound on the DNA. The possibility exists that calcium may exert its action by activating specific endonuclease(s) able to induce single-strand breaks in DNA. However, preincubation of the cells with aurintricarboxylic acid, a nonspecific inhibitor of endonucleases, did not show any protective effects either on H2O2-induced DNA damage or H2O2-induced apoptosis (Galaris et al., unpublished results). The H2O2-induced DNA damage described above takes place within a few minutes after the exposure of the cells to H2O2 (Fig. 1) and seems to be reversible since addition of catalase brings it to control levels within a few minutes (Galaris et al., unpublished results). However, 3 to 4 h after bolus addition, the first signs of apoptosis become apparent. Interestingly, removal of H2O2 from the growth medium after its initial action on the cells, although leading to a decreased level of singlestrand breaks, did not prevent apoptosis. On the contrary, the presence of H2O2 after initiation inhibits the apoptotic process (Figs. 2–7). Although extensively studied, many aspects of the processes that take place during the intermediate period between initiation and execution of apoptosis remain unclear. According to the schemes proposed, a variety of protein activation cascades are taking place, which ultimately converge to a common caspase pathway that leads to the cleavage of chromatin at pieces of 180 to 200 base pairs. Based on the results of the present work and in accordance with other reports [36 –38], it is concluded that the presence of even low concentrations of H2O2 during the intermediate period between the initiation and

H2O2-induced DNA damage and apoptosis

the execution of apoptosis inhibits this process regardless of the initiation agent being H2O2 itself or other inducers, like an anti-Fas antibody acting through the Fas receptor. Since the amount of glucose oxidase that was able to inhibit apoptosis generated about 2.0 ␮M H2O2 per min, the actual extracellular concentration of H2O2 must be much lower. In addition, if we assume that a gradient of extra- to intracellular H2O2 exists with a ratio 7:1 [28], we can estimate that subnanomolar intracellular concentrations of H2O2 are able to inhibit the apoptotic process. The exact mechanism of this inhibitory action is not known, but a direct oxidation of specific cysteine residues of some caspases as well as inhibition of steps upstream caspase activation due to decreased synthesis of ATP have been proposed [36 –38]. However, a number of other critical points may be potential targets of H2O2 actions. One prominent question arising from the results of this work is regarding the fate of the cells beyond the period of 6 h. Although not mentioned, preliminary results from our laboratory indicate that exposure of the cells for 6 h to 2 ␮M H2O2 per min resulted in growth arrest. However, after the removal of the H2O2-generating system, cells continued to proliferate with a rate similar to control cells (results not shown). Cells exposed to 150 ␮M H2O2 followed by continuous low levels of H2O2 for 6 h die later by apoptosis. An extensive search regarding the long-term fate of Jurkat cells after exposure to various forms of oxidative stress is currently under way in our laboratory. In conclusion, the presence of continuously generated H2O2, although able to induce the formation of singlestrand breaks in DNA, interrupt the cascade of events that leads to apoptotic cell death. The exact mechanism(s) of this inhibitory action of H2O2 is still unknown, but it takes place before the third hour after initiation and acts at or upstream of caspase-3 activation. Acknowledgement — This research was supported by grant PENED 99, No 99ED181 of General Secretariat of Research and Technology, Athens, Greece.

REFERENCES [1] Ames, B.; Shigenaga, M. K.; Hagen, T. M. Oxidants, antioxidants, and the degenerative diseases of aging. Proc. Natl. Acad. Sci. USA 90:7915–7922; 1993. [2] Griendling, K. K.; Harrison, D. G. Dual role of reactive oxygen species in vascular growth. Circ. Res. 85:562–563; 1999. [3] Sundaresan, M.; Zu-Xi, Y.; Ferrans, V. J.; Irani, K.; Finkel, T. Requirement for generation of H2O2 for platelet-derived growth factor signal transduction. Science 270:296 –299; 1995. [4] Forman, H. J.; Torres, M. Redox signaling in macrophages. Mol. Asp. Med. 22:189 –216; 2001. [5] Roveri, A.; Coassin, M.; Maiorino, M.; Zamburlini, A.; Amsterdam, F. T. M.; Ratti, E.; Ursini, F. Effects of hydrogen peroxide on calcium homeostasis in smooth muscle cells. Arch. Biochem. Biophys. 297:265–270; 1992.

701

[6] Golconda, M. S.; Ueda, N.; Shah, S. V. Evidence suggesting that iron and calcium are interrelated in oxidant-induced DNA damage. Kidney Intl. 44:1228 –1234; 1993. [7] Ikebuchi, Y.; Masumoto, N.; Tasaka, K.; Koike, K.; Kasahara, K.; Miyake, A.; Tanizawa, O. Superoxide anion increases intracellular pH, intracellular free calcium, and arachidonate release in human amnion cells. J. Biol. Chem. 266:13233–13237; 1991. [8] Jin, N.; Hatton, N. D.; Harrington, M. A.; Xia, X.; Larsen, S. H.; Rhoades, R. A. H2O2-induced egr-1, fra-1, and c-jun gene expression is mediated by tyrosine kinase in aortic smooth muscle cells. Free Radic. Biol. Med. 29:736 –746; 2000. [9] Nakamura, H.; Nakamura, K.; Yodoi, J. Redox regulation of cellular activation. Ann. Rev. Immunol. 15:351–369; 1997. [10] Burdon, R. Superoxide and hydrogen peroxide in relation to mammalian cell proliferation. Free Radic. Biol. Med. 18:775– 794; 1995. [11] Brown, M. R.; Miller, F. J. Jr.; Li, W. G.; Ellingson, A. N.; Mozena, J. D.; Chatterjee, P.; Engelhardt, J. F.; Zwacha, R. M.; Oberley, L. W.; Fang, X.; Spector, A. A.; Weintraub, N. L. Overexpression of human catalase inhibits proliferation and promotes apoptosis in vascular smooth muscle cells. Circ. Res. 85:524 –533; 1999. [12] Antunes, F.; Cadenas, E. Cellular titration of apoptosis with steady state concentrations of H2O2: submicromolar levels of H2O2 induce apoptosis through Fenton chemistry independent of the cellular thiol state. Free Radic. Biol. Med. 30:1008 –1018; 2001. [13] Yang, Y.; Cheng, J. Z.; Singhal, S. S.; Sanai, M.; Pandya, U.; Awasthi, S.; Awasthi, Y. C. Role of glutathione S-transferases in protection against lipid peroxidation. Overexpression of hGSTA2-2 in K562 cells protects against hydrogen peroxide-induced apoptosis and inhibits JNK and caspase 3 activation. J. Biol. Chem. 276:19220 –19230; 2001. [14] Chandra, J.; Samali, A.; Orrenius, S. Triggering and modulation of apoptosis by oxidative stress. Free Radic. Biol. Med. 29:323– 333; 2000. [15] Meneghini, R. Iron homeostasis, oxidative stress, and DNA damage. Free Radic. Biol. Med. 23:783–792; 1997. [16] Halliwell, B.; Aruoma, O. I. DNA damage by oxygen-derived species: its mechanism and measurement in mammalian systems. FEBS Lett. 281:9 –19; 1991. [17] Imlay, J.; Chin, S.; Linn, S. Toxic DNA damage by hydrogen peroxide through the Fenton reaction in vivo and in vitro. Science 240:640 – 642; 1988. [18] Chevion, M. A site-specific mechanism for free radical induced biological damage: the essential role of redox-active transition metals. Free Radic. Biol. Med. 5:27–37; 1988. [19] Antunes, F.; Cadenas, E.; Brunk, U. T. Apoptosis induced by exposure to a low steady state concentration of H2O2 is a consequence of lysosomal rupture. Biochem. J. 356:549 –555; 2001. [20] Brunk, U. T.; Neuzil, J.; Eaton, J. W. Lysosomal involvement in apoptosis. Redox Rep. 6:91–97; 2001. [21] Li, W.; Yuan, X.; Nordgren, G.; Dalen, H.; Dobowchik, G. M.; Firestone, R. A.; Brunk, U. T. Induction of cell death by the lysosomotropic detergent MSDH. FEBS Lett. 470:35–39; 2000. [22] Cantoni, O.; Sestili, P.; Cattabeni, F.; Bellomo, G.; Pou, S.; Cohen, M.; Cerutti, P. Calcium chelator Quin 2 prevents hydrogen peroxide-induced DNA breakage and cytotoxicity. Eur. J. Biochem. 182:209 –212; 1989. [23] Panagiotidis, M.; Tsolas, O.; Galaris, D. Glucose oxidase-produced H2O2 induces Ca2⫹-dependent DNA damage in human peripheral blood lymphocytes. Free Radic. Biol. Med. 26:548 – 556; 1999. [24] Doulias, P.-T.; Barbouti, A.; Galaris, D.; Ischiropoulos, H. SIN1-induced DNA damage in isolated human peripheral blood lymphocytes as assessed by single-cell gel electrophoresis (comet assay). Free Radic. Biol. Med. 30:679 – 685; 2001. [25] Barbouti, A.; Doulias, P.-T.; Zhu, B.-Z.; Frei, B.; Galaris, D. Intracellular iron, but not copper, plays a critical role in hydrogen peroxide-induced DNA damage. Free Radic. Biol. Med. 31:490 – 498; 2001.

702

A. BARBOUTI et al.

[26] Kim, D. K.; Cho, E. S.; Um, H.-D. Caspase-dependent and -independent events in apoptosis induced by hydrogen peroxide. Exp. Cell Res. 257:82– 88; 2000. [27] Gardner, A. M.; Xu, F.-H.; Fady, C.; Jacoby, F. J.; Duffy, D. C.; Tu, Y.; Lightenstein, A. Apoptotic vs. nonapoptotic cytotoxicity induced by hydrogen peroxide. Free Radic. Biol. Med. 22:73– 83; 1997. [28] Antunes, F.; Cadenas, E. Estimation of H2O2 gradients across biomembranes. FEBS Lett. 475:121–126; 2000. [29] Chance, B.; Sies, H.; Boveris, A. Hydroperoxide metabolism in mammalian organs. Physiol. Rev. 59:527– 605; 1979. [30] Tselepis, A.; Doulias, P.-T.; Lourida, E.; Glantzounis, G.; Tsimoyiannis, E.; Galaris, D. Trimetazidine protects low-density lipoproteins from oxidation and cultured cells exposed to H2O2 from DNA damage. Free Radic. Biol. Med. 30:1357–1364; 2001. [31] Ostling, O.; Johanson, K. J. Microelectrophoretic study of radiation-induced DNA damages in individual mammalian cells. Biochem. Biophys. Res. Commun. 123:291–296; 1984. [32] Singh, N. P.; McCoy, M. T.; Tice, R. R.; Schneider, E. L. A simple technique for quantitation of low levels of DNA damage in individual cells. Exp. Cell Res. 184:461– 470; 1988. [33] Collins, A. R.; Ma, A.; Duthie, S. J. The kinetics of repair of oxidative DNA damage (strand breaks and oxidized pyrimidines) in human cells. Mutat. Res. 336:69 –77; 1995. [34] Duthie, S. J.; Haedon, A. DNA instability (strand breakage, uracil misincorporation, and defective repair) is increased by folic acid depletion in human lymphocytes in vitro. FASEB J. 12:1491– 1497; 1998. [35] Garmichael, J.; DeGraff, W. G.; Gazdar, A. F.; Minna, J. B.; Michell, J. B. Evaluation of a tetrazolium-based semiautomatic

[36]

[37]

[38] [39]

[40]

[41] [42] [43] [44]

[45]

assay: assessment of chemosensitivity testing. Cancer Res. 47: 936 –942; 1987. Hampton, M. B.; Orrenius, S. Dual regulation of caspase activity by hydrogen peroxide: implications for apoptosis. FEBS Lett. 414:552–556; 1997. Lee, Y.-J.; Shacter, E. Hydrogen peroxide inhibits activation, not activity of cellular caspase-3 in vivo. Free Radic. Biol. Med. 29:684 – 692; 2000. Lee, Y.-J.; Shacter, E. Oxidative stress inhibits apoptosis in human lymphoma cells. J. Biol. Chem. 274:19792–19798; 1999. Yung, M. C. T.; Satoh, M. S. Functional competition between poly(ADP-ribose)polymerase and its 24 kDa apoptotic fragment in DNA repair and transcription. J. Biol. Chem. 276:11279 – 11286; 2001. Lazebnik, Y. A.; Kaufmann, S. H.; Desnoyers, S.; Poirier, G. G.; Earnshaw, W. C. Cleavage of poly(ADP-ribose) polymerase by a proteinase with properties like ICE. Nature 371:346 –347; 1994. Suzuki, Y. J.; Forman, H. J.; Sevanian, A. Oxidants as stimulators of signal transduction. Free Radic. Biol. Med. 22:269 –285; 1997. Forman, H. J.; Cadenas, E., eds. Oxidative stress and signal transduction. New York: Chapman and Hall; 1997. Finkel, T. Oxygen radicals and signaling. Curr. Opin. Cell Biol. 10:243–253; 1998. Abe, J.-I.; Kusuhara, M.; Ulevitch, R. J.; Berk, B. C.; Lee, J.-D. Big mitogen-activated protein kinase 1 (BMK1) is a redox-sensitive kinase. J. Biol. Chem. 271:16586 –16590; 1996. Hirpara, J. L.; Clement, M.-V.; Pervaiz, S. Intracellular acidification triggered by mitochondrial-derived hydrogen peroxide is an effector mechanism for drug-induced apoptosis in tumor cells. J. Biol. Chem. 276:514 –521; 2001.