Methylmercury and H2O2 provoke lysosomal damage in human astrocytoma D384 cells followed by apoptosis

Free Radical Biology & Medicine, Vol. 30, No. 12, pp. 1347–1356, 2001 Copyright © 2001 Elsevier Science Inc. Printed in the USA. All rights reserved 0891-5849/01/$–see front matter

PII S0891-5849(01)00526-3

Original Contribution METHYLMERCURY AND H2O2 PROVOKE LYSOSOMAL DAMAGE IN HUMAN ASTROCYTOMA D384 CELLS FOLLOWED BY APOPTOSIS ELISABETTA DARE´ ,* WEI LI,† BORIS ZHIVOTOVSKY,* XIMING YUAN,†

and

SANDRA CECCATELLI*

*The National Institute of Environmental Medicine, Division of Toxicology and Neurotoxicology, Karolinska Institutet, Stockholm, Sweden; and †Department of Pathology II, Faculty of Health Sciences, Linko¨ping University, Linko¨ping, Sweden (Received 11 December 2000; Accepted 1 March 2001)

Abstract—Methylmercury (MeHg) is a neurotoxic agent acting via diverse mechanisms, including oxidative stress. MeHg also induces astrocytic dysfunction, which can contribute to neuronal damage. The cellular effects of MeHg were investigated in human astrocytoma D384 cells, with special reference to the induction of oxidative-stress-related events. Lysosomal rupture was detected after short MeHg-exposure (1 ␮M, 1 h) in cells maintaining plasma membrane integrity. Disruption of lysosomes was also observed after hydrogen peroxide (H2O2) exposure (100 ␮M, 1 h), supporting the hypothesis that lysosomal membranes represent a possible target of agents causing oxidative stress. The lysosomal alterations induced by MeHg and H2O2 preceded a decrease of the mitochondrial potential. At later time points, both toxic agents caused the appearance of cells with apoptotic morphology, chromatin condensation, and regular DNA fragmentation. However, MeHg and H2O2 stimulated divergent pathways, with caspases being activated only by H2O2. The caspase inhibitor z-VAD-fmk did not prevent DNA fragmentation induced by H2O2, suggesting that the formation of high-molecular-weight DNA fragments was caspase independent with both MeHg and H2O2. The data point to the possibility that lysosomal hydrolytic enzymes act as executor factors in D384 cell death induced by oxidative stress. © 2001 Elsevier Science Inc. Keywords—Methylmercury, Hydrogen peroxide, Lysosomes, Mitochondrial potential, Caspases, DNA fragmentation, Free radicals

INTRODUCTION

goes bioaccumulation and biomagnification in the aquatic food chain [3]. There is concern about possible neurological deficits in infants due to MeHg-prenatal exposure related to the mother’s diet, since it is presently unclear what level of exposure, if any, is without effect on the developing nervous system [4,5]. MeHg accumulates in various areas of the brain, and it is particularly abundant in glial cells. While neurons of the CNS are an established target of MeHg toxicity, less is known about the effects on glial cells. In primates, exposure to relatively high doses of MeHg increased the number of reactive and hypertrophic astrocytes [6]. Interestingly, studies performed in Macaca fascicularis have shown that long-term (6 months) subclinical exposure to MeHg induced accumulation of both MeHg and inorganic mercury and a significant decline in the number of astrocytes [7]. Primary cultures of astrocytes, isolated from rat brain and exposed to MeHg in vitro, have been reported to exhibit alteration of the K⫹ and amino acid transport,

Methylmercury (MeHg) is a highly toxic environmental pollutant that causes irreparable damage to the central nervous system (CNS). Episodes of MeHg intoxication due to accidental release of MeHg induced a condition with specific pathological features denominated “Minamata disease” [1]. Adult brains denoted disappearance of the granule cells of the cerebellum, loss of neurons in the visual area of the calcarine cortex, and damage of the sensory branch of the peripheral nerves. Also, these accidents revealed that the developing brain is particularly sensitive to the toxic effects of MeHg, resulting in widespread damage [2]. Presently, contaminated fish from polluted areas is the most important route of MeHg exposure to the general population, since MeHg underAddress correspondence to: Elisabetta Dare´, Institute of Environmental Medicine, Karolinska Institutet, BOX 210, S-171 77 Stockholm, Sweden; Tel: ⫹46-8-728 7431; Fax: ⫹46-8-329041; E-Mail: [email protected]. 1347

1348

E. DARE´ et al.

rapid cell swelling, mitochondrial dysfunction, and accumulation of reactive oxygen species (ROS) [8,9]. The increased generation of ROS appears to be a hallmark of MeHg toxicity [10 –12]. It is likely to occur in the mitochondria, where MeHg can uncouple oxidative phosphorylation and electron transport [13], and to be related to the homolytic breakdown of MeHg, which produces alkyl radicals and free radicals [14]. The capability of antioxidants to prevent MeHg-induced neuronal death have confirmed the key role of oxidative stress in MeHg toxicity, suggesting that elevation of ROS is upstream in the cell death cascade [15–17]. Increased generation of ROS leads to lipoperoxidation and consequent disruption of membranes. Recent studies have indicated that oxidative stress can elicit, as an early event, lysosomal rupture [18]. The lysosomes represent a potential source of irreversible cell injury, since they contain powerful hydrolytic enzymes that, once released into the cytosol, can digest the whole cell. Lysosomal damage has been detected in J-774 lymphoma cells and in NIT-1 insulinoma cells exposed to nonlethal concentrations of H2O2 [19,20]. Similarly, photo-oxidation has been shown to induce lysosomal rupture in cultured human fibroblasts and neuroblastoma cells [21], leading to sublethal damage reparable by autophagocytosis or cell death, necrotic or apoptotic, depending on the intensity of the stress. Apoptosis is a modality of cell death characterized by cell shrinkage, nuclear pyknosis, chromatin condensation, DNA cleavage into fragments of regular sizes, and activation of a peculiar class of cysteine-proteases, the caspases [22–24]. During this process the cells preserve the plasma membrane integrity, whereas necrosis is typically associated with destruction of the cell membrane, swelling of the cell and of the organelles, and random cleavage of the chromatin. The aim of the present study was to investigate the cellular and molecular mechanisms of MeHg toxicity in astrocytic cells of primates, using as experimental model the clonal cell line D384 established from a human astrocytoma [25]. The cellular effects induced by MeHg were studied by evaluating the loss of plasma membrane integrity, chromatin condensation, DNA fragmentation, and the activation of caspases. Attention was particularly focused on the possible involvement of lysosomes in the early phases of injury induced by MeHg. Lysosomal damage was measured as loss of functional integrity of lysosomes and related to the alteration of the mitochondrial membrane potential (⌬⌿). Since H2O2 has been previously shown to determine cell damage through lysosomal rupture in several model systems [19,20], D384 cells exposed to H2O2 were analyzed for comparison.

MATERIALS AND METHODS

Chemicals All chemicals used were of analytical grade. Methylmercury (II) hydroxyde (Alfa, Johnson Matthey, Karlsruhe, Germany) was purchased from Novakemi ab (Stockholm, Sweden) as a 1 M solution in H2O. H2O2, Trypan blue, propidium iodide (P.I.), cyclosporin A (CsA) and staurosporine (STS) were purchased from Sigma (St. Louis, MO, USA). All other chemicals for cell culture were supplied by Life Technologies (Ta¨by, Sweden). The caspase 3-like substrate DEVD-MCA [AcAsp-Glu-Val-Asp-␣-(4-methyl-coumaryl-7-amide)] and the pan-caspase inhibitor z-VAD-fmk [Z-Val-AlaAsp(Ome)-FMK] were purchased from Peptide Institute (Osaka, Japan). Acridine orange (AO, Euchrysyn 3R) was obtained from Gurr (Poole, UK) and tetramethylrhodamine ethyl ester (TMRE) from Molecular probes (Eugene, OR, USA). Cell culture and exposure to toxic agents The D384 clonal cell line was established from a human astrocytoma [25]. Cells were routinely seeded at a density of 10,000 cells/cm2 in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 ␮g/ml streptomycin, at 37°C in humidified air with 5% CO2. Cells were cultured in complete medium for 24 h before adding the treatments. The monolayers were exposed to MeHg or H2O2 by adding the compounds directly to the conditioned medium. Acridine orange staining and lysosomal integrity measurements Damage of the lysosomes was measured using the lysosomotropic weak base AO, according to a protocol previously described [26]. The vital staining AO becomes charged (AOH⫹), and consequently membrane impermeable, at the low pH of the lysosomes, thus it is retained in these organelles. The AO concentrated in the lysosomes has a monochromatic red fluorescence when activated with green light. The amount of red fluorescence per cell is indicative of lysosomes with high AO concentration and intact proton gradients. D384 cells were grown on coverslips and exposed to the toxic agents, then incubated with AO-medium (5 ␮g/ml AO in complete medium pre-equilibrated at 37°C for 15 min) and washed with fresh medium before the measurements. The coverslips were mounted on a well slide and the red fluorescence was measured in single cells by static cytofluorometer using green light excitation (Nikon B-2A

Lysosomal damage and apoptosis due to oxidative stress

filter cube and an extra 630 nm barrier filter in a Nikon p102 photometer linked to a computer system). At least 50 cells from one dish were measured in each experiment and the mean value was used for further statistical analysis. The experiments were repeated 4 times. Statistical analysis was performed with the Student t-test comparing control and treated samples analyzed at the same time point. Pictures of AO-stained cells were taken with the LSM 410 inverted confocal laser scanning microscope (Carl Zeiss, Oberkochen, Germany) using the 488 nm argon laser as light source. Measurements of the mitochondrial membrane potential TMRE is a cationic, lipophilic dye that partitions to the negatively charged mitochondrial matrix according to the Nernst equation. Decreases in the mitochondrial membrane potentials are paralleled by a reduction of the fluorescence emitted by TMRE, which acts as a voltagesensitive probe [27,28]. Cells grown on coverslips were exposed to the toxic agents, then incubated with 20 nM TMRE for 30 min at room temperature and washed with PBS before further analysis. The fluorescence emitted by single cells was measured by static cytofluorometer using green light excitation. At least 50 cells were measured in each experiment and the mean was calculated and used for statistical analysis. The experiments were repeated 3 or 4 times. Statistical analysis was performed with the Student t-test comparing control and treated samples analyzed at the same time point. Trypan blue exclusion test Cells were detached from the flasks by trypsinization, then an aliquot of the cell suspension was mixed with an equal volume of 0.2% Trypan blue in PBS. Cells were scored at the phase contrast microscope using a Neubauer improved counting chamber. Cells with damaged cell membrane stained blue (dead cells), while cells with plasma membrane integrity prevented the dye entry, remaining unstained (healthy cells and apoptotic cells). The experiments were performed in triplicate and repeated 2 or 3 times. Statistical analysis was performed using either one-way analysis of variance (ANOVA) or the Student t-test comparing treated samples to control samples at the same time point. Nuclear staining with propidium iodide and TUNEL staining Cells were grown on coverslips, treated with MeHg or H2O2 and fixed in ice-cold methanol/water (8/2 ⫽ v/v),

1349

at ⫺20°C for 30 min. Each coverslip was then washed with PBS and covered with reaction mixture (20 ␮M dUTP-bioltinylated, 0.25 U/␮l terminal-transferase, 1 mM CoCl2, 25 mg/ml BSA, 200 mM potassium cacodylate, 25 mM Tris-HCl pH 6.6) (Roche, Bromma, Sweden) and incubated at 37°C for 1 h. Coverslips were then kept in 300 mM NaCl, 30 mM sodium citrate at room temperature for 15 min, blocked with 2% BSA in PBS for 30 min, then incubated with Extravidine-FITC (Sigma; 1/100 in PBS) at 37°C for 30 min. After washing with PBS, the coverslips were stained with P.I. (2.5 ␮g/ml in PBS) for 5 min, and mounted onto glass slides with PBS/glycerol (1/9 ⫽ v/v) containing 0.1% (w/v) phenylenediamine. Stained cells were analyzed at the fluorescence microscope (Olympus BX60). The smaller size, irregular shape, and higher intensity of chromatin stained with P.I. identified apoptotic nuclei. Images of the nuclei were obtained with the LSM 510 laser scanning confocal microscope (Carl Zeiss, Jena, Germany). The light sources were two lasers that provided the two excitation wavelengths, 488 nm (Ar) and 543 nm (HeNe). The following emission filters were used: BP 505-550 and LP 560. Detection of high molecular weight (H.M.W.)-DNA fragments by pulse field gel electrophoresis (FIGE) To monitor the formation of large DNA fragments, field inversion gel electrophoresis (FIGE) was performed as described previously [29]. Briefly, cells were trypsinized, centrifuged at 150 ⫻ g for 5 min and washed with PBS. The cells were immobilized into agarose plugs and subjected to proteinase K digestion prior to loading into agarose gels [30]. Two sets of pulse markers DNA were used for determination of molecular weights: (i) chromosomes from Saccharomyces cerevisiae (225– 2200 kbp) and (ii) a mixture of ␭-DNA, ␭-Hind III fragments, and ␭-DNA concatemers (0.1–200 kbp) purchased from Sigma. The gels were stained with ethidium bromide (50 ␮g/l) to visualize the DNA and photographed on a 305 nm UV-transilluminator with Polaroid 665 positive/negative films. Measurements of caspase activity The group II-caspase activity (caspase 2, 3, and 7) was measured in cell extracts according to a method previously described [31], with some modification [32]. Cleavage of the substrate DEVD-MCA leading to the release of free 4-methyl-coumaryl-7-amide (excitation 355 nm, emission 460 nm) was monitored at 37°C using a Fluoroskan II (Labsystem AB, Stockholm, Sweden). Fluorescent units were converted to pmoles of 4-methyl-

E. DARE´ et al.

1350

coumaryl-7-amide released using a standard curve generated with 4-methyl-coumaryl-7-amide and subsequently related to protein content. The measurements were performed in triplicate and the experiments were repeated 2 times. Statistical analysis was performed using ANOVA. RESULTS

Disruption of lysosomes and decrease of mitochondrial membrane potential in cells exposed to MeHg and H2O2 D384 cells stained with the lysosomotrophic vital dye AO showed the presence of a well-defined granular lysosomal compartment by confocal microscopy (Fig. 1A). Exposure to either MeHg or H2O2 decreased AO uptake in the cells, visualized as a reduction in the number of lysosomes and an increased green, cytosolic and nuclear fluorescence (Fig. 1B and C). For AO alteration, light emitted from lysosomal AO was measured in single cells. A significant (p ⱕ .05) reduction of the total cellular fluorescence was observed in experiments where cells were exposed to 1 ␮M MeHg for 1 h (Fig. 2A), indicating rapid disruption of lysosomal membranes. The lysosomal alterations were persistent, as shown by measurements of AO-fluorescence in cells exposed to MeHg for 5 h (Fig. 2A). The uptake of the vital dye tetramethylrhodamine ethyl ester (TMRE) was used for semiquantitative estimations of the mitochondrial membrane potential [27]. In spite of the decrease in lysosomal AO fluorescence, cells treated with MeHg for 3–5 h exhibited mitochondrial membrane potential similar to control cells (Fig. 2B). A significant (p ⱕ .05) decrease of the TMRE-fluorescence was observed after exposing the cells to MeHg for 8 h, suggesting that substantial loss of the mitochondrial membrane potential occurred at this time of exposure (Fig. 2B). Rapid AO relocalization in the cytoplasm was measured also in cells exposed to 100 ␮M H2O2 (Fig. 2A). Like MeHg, H2O2 induced persistent lysosomal damage (Fig. 2A), whereas at the same time point the mitochondrial membrane potential did not differ from control cells (Fig. 2B). Significant (p ⱕ .05) loss of the mitochondrial membrane potential was observed after 8 h (Fig. 2B). Additional effects induced by MeHg D384 cells exposed to 1 ␮M MeHg for 24 h underwent morphological changes, becoming rounded, shrunken, and more loosely attached to the plastic surface (Fig. 3B). The Trypan blue exclusion test indicated cell membrane integrity during the first 12 h of exposure to MeHg (Fig. 4A), that is at time points when lysosomes were already altered and the mitochondria functionality was impaired. A small

Fig. 1. Confocal images of D384 cells stained with AO. (A) Control. (B) Cells exposed to 100 ␮M H2O2 for 5 h. (C) Cells exposed to 1 ␮M MeHg for 5 h. The lysosomes appear as brightly fluorescent dots. After exposure to H2O2 (B) or MeHg (C) the lysosomes are destabilized and the AO fluorescence decreases.

increase of Trypan-blue permeable cells was found after 24 h (Fig. 4A). At this time point chromatin condensation appeared in cells fixed and stained with propidium iodide (Fig. 5A), suggesting the occurrence of apoptosis.

Lysosomal damage and apoptosis due to oxidative stress

1351

Fig. 2. (A) Exposure to either 1 ␮M MeHg or 100 ␮M H2O2 for 1 h caused a decrease of lysosomal AO fluorescence, indicating alterations of the lysosomes. The lysosomal damage was persistent, as shown by the lower AO fluorescence after exposure for 5 h. (B) The mitochondrial membrane potential, measured with the voltage-sensitive fluorescent dye TMRE in cells exposed to either MeHg or H2O2 for 5 h, was found similar to control cells. Exposure for 8 h, instead, decreased the TMRE fluorescence measured in single cells as compared to control samples, suggesting loss of mitochondrial membrane potential. The values displayed in the figure are means ⫾ SEM of 3 or 4 independent experiments. *Significantly different (p ⱕ .05) from control samples at the same time point (Student t-test).

Apoptotic chromatin is characterized by the presence of DNA fragments with 3⬘-hydroxyl ends, which can be visualized with the TUNEL assay. TUNEL positive cells, which were virtually absent in control samples, were observed sporadically after exposure to MeHg for 24 h, and not all the nuclei with condensed chromatin appeared labeled (Fig. 5B). The cells were further tested for the presence of regular DNA fragmentation, which is considered a marker of apoptosis. The DNA of cells exposed to MeHg for 24 h showed an increase of distinct high-molecular-weight (HMW) fragments (700, 300, and 50 kbp), indicative of cleavage at specific sites, but also a smear, resulting from random DNA fragmentation (Fig. 6A). The ability of MeHg to stimulate the activation of

Fig. 3. Phase contrast micrographs showing cell shrinkage induced by MeHg and H2O2 in D384 cells. (A) Control. (B) 1 ␮M MeHg for 24 h. C. 100 ␮M H2O2 for 24 h.

caspases, proteases involved in apoptosis in a variety of cell models, was also examined. The main effector caspases (caspase 3 and caspase 7) all have preference for cleaving after the sequence DEVD. MeHg did not induce any significant increase in DEVDase activity in D384 cells analyzed at different time points of exposure (Fig. 7). Moreover, pre-incubation with the caspase inhibitor z-VAD-fmk did not ameliorate the loss of membrane integrity caused by MeHg (data not shown).

1352

E. DARE´ et al.

not prevent the occurrence of DNA fragmentation (Fig. 6C). Cyclosporin A (CsA) acts as an inhibitor of the mitochondrial permeability transition (MPT)-pore, a megachannel located in the inner mitochondrial membrane that has been implicated in the release of apoptogenic factors from these organelles [33,34]. In our experimental model, CsA (0.5–10 ␮M) did not prevent the activation of class II caspases induced by H2O2 (Table 2). However, CsA reduced significantly (p ⱕ .05) the DEVDase activity in cells exposed to staurosporine (Table 2), a toxic agent that induces apoptosis with the release of cytochrome c from the mithocondria and activation of caspase 3-like proteases in D384 cells [32]. These results indicate that other mechanisms than the MPT-pore opening are activated in H2O2-induced cell death. DISCUSSION

Fig. 4. (A) Alterations of cell membrane impermeability due to exposure to either 1 ␮M MeHg or 100 ␮M H2O2. D384 cells were stained with Trypan blue and scored with a counting chamber at the microscope. Values are percentage of cells impermeable to Trypan blue. (B) Number of cells after exposure to either 1 ␮M MeHg or 100 ␮M H2O2, relative to the number of control cells at the same time point. Values are means ⫾ SEM of 3 determinations. Statistical analysis was performed with the two-tailed Student t-test, comparing exposed cells with control cells at the same time point (*p ⱕ .05; #p ⱕ .01; ¤p ⱕ .001).

Cellular toxicity induced by H2O2 D384 cells exposed to 100 ␮M H2O2 for 24 h presented morphological alterations similar to the one observed in cells exposed to MeHg (Fig. 3C). A small but significant increase in the number of cells permeable to Trypan blue was observed at all the time points analyzed (Fig. 4A). The cell number was lower than in the control after 6 h (Fig. 4B). Nuclei with condensed chromatin were detected in D384 cells exposed to H2O2 for 24 h (Fig. 5C), and they often appeared TUNEL-positive (Fig. 5D). DNA extracted from cells treated with H2O2 showed a regular pattern of fragmentation with the characteristic appearance of a 50 kbp band indicative of DNA cleavage at specific sites (Fig. 6B). H2O2 also caused a significant activation of class II caspases at 15 h and 24 h (Fig. 7). Although the caspase inhibitor z-VAD-fmk could block the DEVDase activity (data not shown), it had no protective effect on the loss of membrane integrity observed after exposure to H2O2 (Table 1) and did

Here we report the evidence that significant lysosome disruption occurrs rapidly in human astrocytoma cells exposed to 1 ␮M MeHg. Lysosomal damage was found to precede any significant reduction of mitochondrial potential and loss of membrane integrity. Rupture of lysosomes, causing the release of proteolytic enzymes into the cytosol, could represent a critical step of MeHg toxicity in astrocytoma cells. In agreement with our results, experiments with aspartic protease inhibitors have recently supplied indirect evidence that the endosomal/lysosomal compartment might also be partially involved in the cell death pathway stimulated by MeHg in primary culture of rat cerebral microglia [35]. The mitochondrial ⌬⌿ was found significantly decreased after exposure to MeHg for 8 h, that is, several hours after the lysosomes had began to undergo leakage. This confirms that MeHg causes damage of the mitochondria, as observed previously in other neural cell models exposed to this toxic agent, such as primary culture of rat astrocytes and rat cerebellar granule neurons [8,36]. However, the fact that lysosomal destabilization anticipated the alterations of ⌬⌿ questions the previous observation that the mitochondria are the earliest target of MeHg neurotoxicity [8]. Disruption of the ⌬⌿, causing depletion of cellular ATP and possibly resulting in release of apoptogenic factors and Ca2⫹ from the mitochondria, is likely to amplify the damage concomitantly induced by lysosomal rupture. Roberg and collaborators [37] have demonstrated that oxidative stress induced by naphthazarin causes lysosomal destabilization followed by cathepsin D release in the cytoplasm in human foreskin fibroblasts. It is worth mentioning that the leakage of proteases from damaged lysosomes preceded the loss of the mitochondrial ⌬⌿. Li

Lysosomal damage and apoptosis due to oxidative stress

1353

Fig. 5. Morphological alterations of the nuclei in D384 cells exposed to either 1 ␮M MeHg (A, B) or 100 ␮M H2O2 (C, D) for 24 h. Nuclei were simultaneously stained with propidium iodide to visualize chromatin condensation (top panels) and labeled with the TUNEL technique to detect DNA breaks with 3⬘-hydroxyl ends (bottom panels). Arrows indicate double stained nuclei. Scale bar ⫽ 25 ␮m.

and coworkers [38] have recently shown that the synthetic lysosomotropic detergent O-methyl-serine dodecylamide hydrochloride (MSDH) results in partial lysosomal rupture and activation of caspase 3-like proteases, followed by apoptosis after some hours. The mitochondrial transmembrane potential was found to decline early during apoptosis, but also in this case, secondary to lysosomal destabilization. Subsequently to MeHg exposure for 24 h the D384 cells displayed morphological alterations resembling apoptosis, such as cell shrinkage, chromatin condensation, and DNA fragmentation. However, MeHg did not cause any significant increase of the caspase 3-like activity in D384 cells. In agreement, our recent analysis of primary cultures of rat cerebellar granule cells exposed to MeHg has pointed out that activation of caspases is not involved in the apoptotic process occurring in these neurons [16]. DNA damage is a common feature of MeHg toxicity reported in a variety of studies, and evidence has been supplied that MeHg activates apoptosis-related endonucleases that cause regular chromatin cleavage in neuronal models [16,39,40]. The DNA extracted from D384 cells exposed to MeHg was found fragmented both regularly and irregularly, resulting in a smear. Caspase-

activated DNase (CAD) [41] is unlikely to participate in the process of DNA degradation induced by MeHg in D384 cells, since no caspase activity was detected in MeHg-exposed D384 cells. Direct DNA oxidation by MeHg could partly contribute to the formation of DNA breaks. Moreover, our data on the AO-relocalization in D384 cells exposed to MeHg are indicative of lysosomerupture, with consequent leakage of their content into the cytosol. DNase II, an endonuclease that generates double strands breaks with a single hit, is known to be abundant in the lysosomes, and thus might be involved in the DNA cleavage observed upon MeHg exposure in D384 cells [42– 44]. Alternatively, the mitochondrial damage caused by MeHg could determine the translocation of the apoptogenic factor AIF from the space between the inner and outer mitochondrial membrane to the nucleus, where AIF is able to induce chromatin condensation and DNA cleavage into HMW fragments [45]. Besides investigating MeHg toxicity, we have also analyzed the consequences of oxidative stress induced by H2O2 in D384 cells. Differently from MeHg, H2O2 rapidly induced loss of membrane integrity in a small but significant fraction of the cells. These Trypan blue-permeable cells are likely to result from a caspase-indepen-

1354

E. DARE´ et al. Table 1. Effect of the Caspase Inhibitor z-VAD-fmk on the Membrane Integrity of D384 Cells Exposed to H2O2

Treatment

Membrane integrity (%)b

Control 20 ␮M z-VAD-fmk 100 ␮M H2O2 20 ␮M z-VAD-fmk ⫹ 100 ␮M H2O2

99 ⫾ 0.2 97.3 ⫾ 0.5 93 ⫾ 0.8* 88.8 ⫾ 0.2*

a

a Cells were pretreated with z-VAD-fmk 30 min prior to addition of H2O2 for 6 h. b Evaluated by scoring cells after Trypan blue staining. Values are means ⫾ SEM (n ⫽ 3). Statistical analysis was performed using ANOVA [F(3, 8) ⫽ 4.664; p ⫽ 0.0363]. * Significantly different from control ( p ⱕ .01) with the Dunnet multiple comparison test.

Fig. 6. DNA fragmentation induced by MeHg and H2O2. D384 cells were exposed to either MeHg (A) or to H2O2 (B and C) for 24 h, then DNA was separated by FIGE in agarose gels. The figure shows the DNA stained with ethidium bromide. Lane 1, control; lane 2, 1 ␮M MeHg; lane 3, mixture of ␭-DNA, ␭-Hind III fragments, and ␭-DNA concatemers; lane 4, chromosomes from Saccharomyces cerevisiae; lane 5, control; lane 6, 100 ␮M H2O2; lane 7, 100 ␮M H2O2; lane 8, control; lane 9, DNA of cells pre-incubated with 20 ␮M z-VAD-fmk before exposure to 100 ␮M H2O2 for 24 h; lane 10, chromosomes from Saccharomyces cerevisiae.

Fig. 7. Measurement of the caspase 3-like activity in D384 cells exposed to MeHg or H2O2. Cells were harvested at different time points of exposure to either 1 ␮M MeHg or 100 ␮M H2O2 and caspase activity was measured with the fluorimetric DEVDase assay. Values (means ⫾ SEM, n ⫽ 3) are expressed as percentage of control cells at the same time points. Statistical analysis was performed with ANOVA. Exposure to MeHg did not induce any significant increase in caspase activity [F(5, 12) ⫽ 2.002]. Opposite, the DEVDase activity was significantly higher in cells exposed to H2O2 as compared to control [F(8, 18) ⫽ 155.12; p ⬍ .0001]. *Significantly different (p ⬍ .001) from control at time zero (Tukey-Kramer multiple comparison test).

dent cell death process, since there was no increase in DEVDase activity at early time points. Moreover, the caspase inhibitor z-VAD-fmk was not capable of inhibiting the loss of membrane integrity provoked by H2O2 after 6 h. At a later time point, cells displayed morphological alterations similar to the ones caused by MeHg, the DNA was cleaved regularly, the chromatin appeared condensed, and TUNEL-positive nuclei were observed. Opposite to MeHg, H2O2 induced increased caspase 3-like activity. The fact that caspases are activated by H2O2 but not by MeHg, in spite of similar lysosomal and mitochondrial damages, might be related to their inhibition by direct binding of MeHg to the SH-groups present in the active site of caspases. Also, MeHg has higher stability as compared to H2O2, that is rapidly degraded, and excessive oxidative stress is likely to inhibit caspases [46]. Cells exposed to H2O2 were analyzed with the vital dyes AO and TMRE, demonstrating that significant damage of the lysosomes was detectable some hours before these cells displayed any significant loss of the mitochondrial membrane potential. Thus, early lysosomal rupture appears a feature common to the agents inducing oxidative stress MeHg and H2O2. These data confirm previous observations in other cell lines exposed to H2O2 [19,20] and in human foreskin fibroblasts exposed to the redox-cycling quinone naphthazarin [37]. A significant increase in caspase 3-like activity was measured in D384 cells exposed to H2O2 for 15–24 h. There is increasing evidence that mitochondria participate in the apoptotic process. Cytochrome c, let loose from the mitochondria, takes part in the formation of the apoptosome complex and in the activation of class II caspases [47]. The release of cytochrome c has been related to the occurrence of mitochondrial permeability transition (MPT), a process that seems to involve the opening of a pore in the inner mitochondrial membrane

Lysosomal damage and apoptosis due to oxidative stress Table 2. Effects of Pretreatment with Cyclosporin A (CsA) on the Caspase Activity Induced by H2O2 and Staurosporine (STS)

Treatment

DEVDase activity (pmol/ min/mg)

Control 10 ␮M CsA 100 ␮M H2O2 1 ␮M STS 10 ␮M CsA ⫹ 100 ␮M H2O2 10 ␮M CsA ⫹ 1 ␮M STS

1.41 ⫾ 0.30 0.98 ⫾ 0.22 6.26 ⫾ 0.49⽤ 29.81 ⫾ 3.09* 5.29 ⫾ 0.16 9.50 ⫾ 2.02⽤†

a

a Cells were pretreated with CsA 30 min before exposure to the toxic agents for 15 h. Values (means ⫾ SEM) were statistically analyzed with ANOVA [F(5, 17) ⫽ 71.209; p ⬍ .0001], followed by TukeyKramer multiple comparison test. ⽤, * Significantly different from control (⽤ p ⱕ .05; *p ⱕ .001). † Significantly different from 1 ␮M STS († p ⱕ .001).

allowing nonspecific passage of Ca2⫹ and low-molecular-weight solutes [33]. Cyclosporin A acts as an inhibitor of the MPT-pore formation by binding to cyclophilin, a molecule that regulates the conformational changes responsible for the MPT pore opening. Notably, CsA did not prevent the activation of class II caspases induced by H2O2 in D384 cells, suggesting that caspase activation was not determined by the MPT pore opening. Release of cytochrome c from the mitochondria independently from MPT has been shown to occur in certain model systems [48,49] and it could be the mechanism operating in D384 cells exposed to H2O2. Caspases are zimogens that require proteolytic cleavage for activation. It has been shown that lysosomal enzymes, such as cathepsins L, can activate cytosolic procaspase-3 by proteolysis [50,51]. Therefore the DEVDase activity observed in H2O2-exposed D384 cells may be due to the disruption of their lysosomal membranes observed at early time points. It should be pointed out that, although pre-incubation with the caspase inhibitor z-VAD-fmk prevented caspase activation by H2O2 in D384 cells, it did not abolish DNA fragmentation, suggesting that DNA was cleaved to a large extent via caspase-independent processes. As discussed above for MeHg, it remains possible that AIF released from the mitochondria and/or DNase II released from the lysosomes participate to the DNA degradation process. In conclusion, this study contributes to order sequentially the cascade of events induced by MeHg and H2O2 in D384 cells. We have presented the novel finding that lysosome rupture is an early cytotoxic event induced by MeHg, as well as by H2O2, in human astrocytoma D384 cells. These alterations of the lysosomes were persistent and preceded the decrease of the mitochondrial potential and DNA fragmentation observed at later time points. In spite of the similar damage caused to lysosome and mitochondria, MeHg and H2O2 were found to activate

1355

divergent pathways at later time points, with caspases being activated only after H2O2. Our data are compatible with a model where hydrolytic enzymes released from the lysosomes would constitute by themselves “execution factors” in the death process occurring under oxidative stress [44]. However, this hypothesis awaits further experimental evidences to be confirmed. Acknowledgements — The work was supported by grants from the European Commission, the Swedish Environmental Protection Agency, the Foundation for Strategic Environmental Research (MISTRA), the ¨ sterSwedish Medical Research Council (3490), and Landstinget in O go¨tland. The authors are grateful to Prof. U. T. Brunk for critical reading of the manuscript.

REFERENCES [1] Takeuchi, T., eds. Pathology of Minamata disease. Japan: Kumamoto University; 1968. [2] Lapham, L. W.; Cernichiari, E.; Cox, C.; Myers, G. J.; Baggs, R. B.; Brewer, R.; Shamlaye, C. F.; Davidson, P. W.; Clarkson, T. W. An analysis of autopsy brain tissue from infants prenatally exposed to methylmercury. Neurotoxicology 16:689 –704; 1995. [3] Morel, F. M. M.; Kraepiel, A. M. L.; Amyot, M. The chemical cycle and bioaccumulation of mercury. Annu. Rev. Ecol. Syst. 29:543–566; 1998. [4] Clarkson, T. W. Human toxicology of mercury. J. Trace Elem. Exp. Med. 11:303–317; 1998. [5] Grandjean, P.; Weihe, P.; White, R. F.; Debes, F. Cognitive performance of children prenatally exposed to “safe” levels of methylmercury. Environ. Res. 77:165–172; 1998. [6] Shaw, C. M.; Mottet, N. K.; Body, R. L.; Luschei, E. S. Variability of neuropathologic lesions in experimental methylmercurial encephalopathy in primates. Am. J. Pathol. 80:451– 469; 1975. [7] Charleston, J. S.; Body, R. L.; Bolender, R. P.; Mottet, N. K.; Vahter, M. E.; Burbacher, T. M. Changes in the number of astrocytes and microglia in the thalamus of the monkey Macaca fascicularis following long-term subclinical methylmercury exposure. Neurotoxicology 17:127–138; 1996. [8] Yee, S.; Choi, B. H. Oxidative stress in neurotoxic effects of methylmercury poisoning. Neurotoxicology 17:17–26; 1996. [9] Aschner, M.; Rising, L.; Mullaney, K. J. Differential sensitivity of neonatal rat astrocyte cultures to mercuric chloride (MC) and methylmercury (MeHg): studies on K⫹ and amino acid transport and metallothionein (MT) induction. Neurotoxicology 17:107– 116; 1996. [10] InSug, O.; Datar, S.; Koch, C. J.; Shapiro, I. M.; Shenker, B. J. Mercuric compounds inhibit human monocyte function by inducing apoptosis: evidence for formation of reactive oxygen species, development of mitochondrial membrane permeability transition and loss of reductive reserve. Toxicology 124:211–224; 1997. [11] LeBel, C. P.; Ali, S. F.; McKee, M.; Bondy, S. C. Organometalinduced increases in oxygen reactive species: the potential of 2⬘,7⬘-dichlorofluorescin diacetate as an index of neurotoxic damage. Toxicol. Appl. Pharmacol. 104:17–24; 1990. [12] Shenker, B. J.; Guo, T. L.; Shapiro, I. M. Low-level methylmercury exposure causes human T-cells to undergo apoptosis: evidence of mitochondrial dysfunction. Environ. Res. 77:149 –159; 1998. [13] Verity, M. A.; Sarafian, T.; Pacifici, E. H.; Sevanian, A. Phospholipase A2 stimulation by methyl mercury in neuron culture. J. Neurochem. 62:705–714; 1994. [14] Ganther, H. E. Interactions of vitamin E and selenium with mercury and silver. Ann. N. Y. Acad. Sci. 355:212–226; 1980. [15] Kasuya, M. Effect of selenium on the toxicity of methylmercury on nervous tissue in culture. Toxicol. Appl. Pharmacol. 35:11–20; 1976.

1356

E. DARE´ et al.

[16] Dare´, E.; Go¨tz, M. E.; Zhivotovsky, B.; Manzo, L.; Ceccatelli, S. Antioxidants J811 and 17␤-estradiol protect cerebellar granule cells from methylmercury-induced apoptotic cell death. J. Neurosci. Res. 62:557–565; 2000. [17] Park, S. T.; Lim, K. T.; Chung, Y. T.; Kim, S. U. Methylmercuryinduced neurotoxicity in cerebral neuron culture is blocked by antioxidants and NMDA receptor antagonists. Neurotoxicology 17:37– 45; 1996. [18] Ollinger, K.; Brunk, U. T. Cellular injury induced by oxidative stress is mediated through lysosomal damage. Free Radic. Biol. Med. 19:565–574; 1995. [19] Brunk, U. T.; Zhang, H.; Dalen, H.; Ollinger, K. Exposure of cells to nonlethal concentrations of hydrogen peroxide induces degeneration-repair mechanisms involving lysosomal destabilization. Free Radic. Biol. Med. 19:813– 822; 1995. [20] Olejnicka, B. T.; Dalen, H.; Brunk, U. T. Minute oxidative stress is sufficient to induce apoptotic death of NIT-1 insulinoma cells. APMIS 107:747–761; 1999. [21] Brunk, U. T.; Dalen, H.; Roberg, K.; Hellquist, H. B. Photooxidative disruption of lysosomal membranes causes apoptosis of cultured human fibroblasts. Free Radic. Biol. Med. 23:616 – 626; 1997. [22] Kerr, J. F.; Wyllie, A. H.; Currie, A. R. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer 26:239 –257; 1972. [23] Wyllie, A. H.; Kerr, J. F.; Currie, A. R. Cell death: the significance of apoptosis. Int. Rev. Cytol. 68:251–306; 1980. [24] Thornberry, N. A.; Lazebnik, Y. Caspases: enemies within. Science 281:1312–1316; 1998. [25] Balmforth, A. J.; Ball, S. G.; Freshney, R. I.; Graham, D. I.; McNamee, H. B.; Vaughan, P. F. D-1 dopaminergic and betaadrenergic stimulation of adenylate cyclase in a clone derived from the human astrocytoma cell line G-CCM. J. Neurochem. 47:715–719; 1986. [26] Yuan, X. M.; Li, W.; Olsson, A. G.; Brunk, U. T. The toxicity to macrophages of oxidized low-density lipoprotein is mediated through lysosomal damage. Atherosclerosis 133:153–161; 1997. [27] Ehrenberg, B.; Montana, V.; Wei, M. D.; Wuskell, J. P.; Loew, L. M. Membrane potential can be determined in individual cells from the nernstian distribution of cationic dyes. Biophys. J. 53: 785–794; 1988. [28] Krohn, A. J.; Preis, E.; Prehn, J. H. Staurosporine-induced apoptosis of cultured rat hippocampal neurons involves caspase-1like proteases as upstream initiators and increased production of superoxide as a main downstream effector. J. Neurosci. 18:8186 – 8197; 1998. [29] Zhivotovsky, B.; Wade, D.; Gahm, A.; Orrenius, S.; Nicotera, P. Formation of 50 kbp chromatin fragments in isolated liver nuclei is mediated by protease and endonuclease activation. FEBS Lett. 351:150 –154; 1994. [30] Go¨tz, M. E.; Ahlbom, E.; Zhivotovsky, B.; Blum-Degen, D.; Oettel, M.; Ro¨mer, W.; Riederer, P.; Orrenius, S.; Ceccatelli, S. Radical scavenging compound J 811 inhibits hydrogen peroxideinduced death of cerebellar granule cells. J. Neurosci. Res. 56: 420 – 426; 1999. [31] Nicholson, D. W.; Ali, A.; Thornberry, N. A.; Vaillancourt, J. P.; Ding, C. K.; Gallant, M.; Gareau, Y.; Griffin, P. R.; Labelle, M.; Lazebnik, Y. A. Identification and inhibition of the ICE/CED-3 protease necessary for mammalian apoptosis. Nature 376:37– 43; 1995. [32] Gorman, A. M.; Hirt, U. A.; Orrenius, S.; Ceccatelli, S. Dexamethasone pre-treatment interferes with apoptotic death in glioma cells. Neuroscience 96:417– 425; 2000. [33] Zoratti, M.; Szabo, I. The mitochondrial permeability transition. Biochim. Biophys. Acta 1241:139 –176; 1995.

[34] Halestrap, A. P.; Davidson, A. M. Inhibition of Ca2⫹-induced large-amplitude swelling of liver and heart mitochondria by cyclosporin is probably caused by the inhibitor binding to mitochondrial-matrix peptidyl-prolyl cis-trans isomerase and preventing it interacting with the adenine nucleotide translocase. Biochem. J. 268:153–160; 1990. [35] Nishioku, T.; Takai, N.; Miyamoto, K.; Murao, K.; Hara, C.; Yamamoto, K.; Nakanishi, H. Involvement of caspase 3-like protease in methylmercury-induced apoptosis of primary cultured rat cerebral microglia. Brain Res. 871:160 –164; 2000. [36] Castoldi, A. F.; Barni, S.; Turin, I.; Gandini, C.; Manzo, L. Early acute necrosis, delayed apoptosis and cytoskeletal breakdown in cultured cerebellar granule neurons exposed to methylmercury. J. Neurosci. Res. 59:775–787; 2000. [37] Roberg, K.; Johansson, U.; Ollinger, K. Lysosomal release of cathepsin D precedes relocation of cytochrome c and loss of mitochondrial transmembrane potential during apoptosis induced by oxidative stress. Free Radic. Biol. Med. 27:1228 –1237; 1999. [38] Li, W.; Yuan, X.; Nordgren, G.; Dalen, H.; Dubowchik, G. M.; Firestone, R. A.; Brunk, U. T. Induction of cell death by the lysosomotropic detergent MSDH. FEBS Lett. 470:35–39; 2000. [39] Kunimoto, M. Methylmercury induces apoptosis of rat cerebellar neurons in primary culture. Biochem. Biophys. Res. Commun. 204:310 –317; 1994. [40] Miura, K.; Koide, N.; Himeno, S.; Nakagawa, I.; Imura, N. The involvement of microtubular disruption in methylmercury-induced apoptosis in neuronal and nonneuronal cell lines. Toxicol. Appl. Pharmacol. 160:279 –288; 1999. [41] Enari, M.; Sakahira, H.; Yokoyama, H.; Okawa, K.; Iwamatsu, A.; Nagata, S. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 391:43–50; 1998. [42] Bernardi, G. In: Boyer, P. D., ed. The enzymes. New York: Academic Press; 1971:271–287. [43] Bradley, M. O.; Taylor, V. I.; Armstrong, M. J.; Galloway, S. M. Relationships among cytotoxicity, lysosomal breakdown, chromosome aberrations, and DNA double-strand breaks. Mutat. Res. 189:69 –79; 1987. [44] Zamzami, N.; Kroemer, G. Condensed matter in cell death. Nature 401:127–128; 1999. [45] Susin, S. A.; Lorenzo, H. K.; Zamzami, N.; Marzo, I.; Snow, B. E.; Brothers, G. M.; Mangion, J.; Jacotot, E.; Costantini, P.; Loeffler, M.; Larochette, N.; Goodlett, D. R.; Aebersold, R.; Siderovski, D. P.; Penninger, J. M.; Kroemer, G. Molecular characterization of mitochondrial apoptosis-inducing factor. Nature 397:441– 446; 1999. [46] Hampton, M. B.; Fadeel, B.; Orrenius, S. Redox regulation of the caspases during apoptosis. Ann. N. Y. Acad. Sci. 854:328 –335; 1998. [47] Green, D. R.; Reed, J. C. Mitochondria and apoptosis. Science 281:1309 –1312; 1998. [48] Kluck, R. M.; Bossy-Wetzel, E.; Green, D. R.; Newmeyer, D. D. The release of cytochrome c from mitochondria: a primary site for Bcl-2 regulation of apoptosis. Science 275:1132–1136; 1997. [49] Andreyev, A.; Fiskum, G. Calcium induced release of mitochondrial cytochrome c by different mechanisms selective for brain versus liver. Cell Death Differ. 6:825– 832; 1999. [50] Ishisaka, R.; Utsumi, T.; Yabuki, M.; Kanno, T.; Furuno, T.; Inoue, M.; Utsumi, K. Activation of caspase-3-like protease by digitonin-treated lysosomes. FEBS Lett. 435:233–236; 1998. [51] Ishisaka, R.; Utsumi, T.; Kanno, T.; Arita, K.; Katunuma, N.; Akiyama, J.; Utsumi, K. Participation of a cathepsin L-type protease in the activation of caspase-3. Cell. Struct. Funct. 24: 465– 470; 1999.