Molecular Vision 2004; 10:65-73 <http://www.molvis.org/molvis/v10/a10/> Received 25 February 2003 | Accepted 9 January 2004 | Published 27 January 2004

Download Reprint

L-DNase II associated with active process during ethanol induced cell death in ARPE-19

Jean-Yves Brossas,¹ Ronan Tanguy,¹ Francoise Brignole-Baudouin,² Yves Courtois,¹ Alicia Torriglia,¹ Jacques Tréton¹
 
 

¹INSERM Unité 450, Association Claude Bernard, Paris, France; ²Laboratoire d'Immuno-Hématologie, Hópital Ambrois-Paré, AP-HP, Université Paris V, Paris, France

Correspondence to: Jacques Tréton, INSERM Unité 450 - ACB, Institut des Cordeliers, 15, Rue de l'Ecole de Médecine, F-75006 Paris, France; Phone: 033 01 40 46 78 68; FAX: 01 40 46 78 65; email: treton@infobiogen.fr

Abstract

Purpose: To analyse the mechanism of ethanol-induced cell death, and particularly, the activation of the leucocyte elastase inhibitor (LEI) pathway.

Methods: Cultured ARPE-19 cells were exposed to 0-13% ethanol for 24 h. Cytotoxicity was estimated by morphologic changes within the nucleus and breakdown of DNA, assessed by agarose gel electrophoresis or flow cytometry cell sorter. Poly(ADP-ribose)polymerase cleavage (PARP) was determined by western blot analysis.Changes in transcription and translation of LEI were assessed by analysis of mRNA levels and expression of protein product (immunohistochemistry), respectively.

Results: We established the ability of ethanol to induce cell death in ARPE-19 cells. After a 24 h incubation with 4% ethanol, 50% of the cells died; all the cells died in the presence of 10% ethanol. After ethanol incubation, we observed nuclear condensation and DNA fragmentation; the amount of fragmentation was proportional to the ethanol level. By flow cytometry analysis and agarose gel electrophoresis, the pattern of DNA cleavage exhibited a sub-G1 peak, suggesting necrotic cell death. However, other observations, i.e. nuclei shrinkage, PARP cleavage and inhibition of cell death by cycloheximide, and activation of a caspase independant LEI/DNase II pathway were observed and are features associated with apoptotic cell death. During ethanol stress, an LEI/L-DNase II intermediate was lost, leading to complete activation L-DNase II (24 kDa). RT-PCR analysis showed an early and specific increase of the LEI mRNA. Cycloheximide inhibited LEI synthesis and protected cells against apoptosis.

Conclusions: Our data indicate that ethanol stress on ARPE-19 cells can induce a pathway which is a form of programmed cell death with characteristics of both apoptosis and necrosis, possibly by triggering conversion of LEI to L-DNase II.

Introduction

Pigmented epithelial cells (RPE) play a major role in retina homeostasis by phagocytosis of photoreceptors outer segments and by its involvement in the hematoretinal barrier. The reaction of these cells to stress has been studied under different conditions. Senescence of the RPE, as a result of replicative exhaustion, has been studied as an in vitro phenomenon [1]. Variable results have been reported [2]. Hyperoxia induces mild oxidative stress in RPE cells in vitro [3]. The distribution of topographical and age-related changes in IGFBP-2 expression in cultured RPE cells isolated from human donor eyes is likely not to be representative of the situation in vivo [4]. RPE cryopreservation could also induce telomere-shortening [5]. Oxidative stress induces single strand breaks in chromosomal telomeres of RPE cells in culture [6] and down-regulates differentiation-specific gene expression [7]. DNA array studies have shown that matrix containing advanced glycation end products could induce an RPE cell phenotype characterized by age-related degenerative changes in the RNA [8]. A recent study in ARPE-19 cells compared the effect of different oxidative stressors on gene expression and found down-regualtion in genes associated with apoptosis, cell cycle regulation, cell-cell communication, signal transduction, and transcriptional regulation [9].

Of all biochemical pathways controlling cell death, the caspase-mitochondrial pathway is one of the most important in apoptosis biochemistry. However, even this pathway has its limits. Inhibition of the caspase group of enzymes does not prevent cell death in many mammalian models [10-14], suggesting the existence of several apoptotic pathways. A common feature of the activation of the different apoptotic pathways appears to be the activation of other proteases and endonucleases that degrade molecules critical to cell survival. Thus, the activation of different proteases or endonucleases may reflect the activation of different apoptotic pathways. One of them is the LEI-L Dnase II pathway. This is a caspase independent pathway. The key molecule is leukocyte elastase inhibitor (LEI), a member of the serpin (Serine Protease Inhibitor) superfamily. When apoptosis is induced under specific conditions, LEI is transformed into L-DNase II, a molecule with endonuclease activity.

Recently, stress-induced premature senescence and replicative senescence in human diploid fibroblasts (HDF) was obtained by a repeated subtoxic exposure to ethanol [15]. There are indications that small doses of alcohol act directly on the RPE independently of light [16-20]. Ethanol induces apoptosis in the developing visual system during synaptogenesis [21]. Finally, differential effects of alcohol on the electro-oculographic responses of patients with age-related macular disease have been shown [20].

In the study we report here, we evaluated the effect of ethanol exposure on survival of ARPE-19 cells, and investigated the LEI biochemical pathway activated during ethanol-induced cell death.

Methods

Reagents and antibodies

Cycloheximide (CHM), propidium Iodide (PI), and fluoro-Di-Acetate (FDA) were obtained from Sigma Chemical Co., France. N-tosyl-L-phenylalanyl-chloromethyl-ketone (TPCK) was obtained from Calbiochem (La Jolla, CA). Ethanol (100%) was obtained from (Prolabo, Paris). Polyclonal pig anti-DNase II was prepared and fully characterized as described previously [22]. This antibody recognizes human L-DNase II in its fully processed or intermediate form (24 and 38 kDa respectively) as well as its precursor, Leukocyte Elastase Inhibitor (LEI; 42 kDa) and the complex between LEI and elastase (60 kDa) [23]. This antibody completely inhibits 4 U of L-DNase II at 1:100 dilution [22]. Polyclonal antibody directed against one synthetic peptide from LEI was prepared in the same way. Anti-peptide 181-196 has higher affinity for the native form of LEI (42 kDa) [22].

Cell culture and ethanol treatment

The ARPE-19 cell line, graciously provided by Dr. Hjelmeland (University of California, Davis, CA), was derived from one globe of a 19 year-old patient and propagated as previously described [24]. Cultures were maintained in Dulbecco's modified Eagle medium/Nutrient mixture F12 with 15 mM HEPES buffer (DMEM/F12, Life Technology GIBCO BRL, Cergy Pontoise, France), 10% fetal bovine serum (FBS; Life Technology), 0.348% additional sodium bicarbonate, and 2 mM L-glutamine (Life Technology), under conditions of 5% CO₂ and 100% relative humidity at 37 °C. For experiments, cells were grown in 150 cm² flasks (Falcon, VWR International Fontenay-sous-Bois, France) at an initial seeding density of 10,000 cells/cm². Cells were plated near confluence to avoid cell detachment. After three days of culture, cells are treated with ethanol at various concentrations (v/v) for 24 h. by replacing the medium with HAM F12 containing 10% FBS and different concentrations of ethanol.

Cell viability and quantification of cell death by flow cytometry

FDA diffuses freely into cells. In viable cells, FDA is converted by intracellular esterase to highly fluorescent fluorescein. As a result, viable cells have bright green fluorescence when analyzed by flow cytometry, whereas non-viable cells are non-fluorescent. PI is a highly fluorescent compound that under isotonic conditions can only enter cells that lack membrane integrity. Thus, after incubation with PI, non-viable cells have bright red fluorescence and viable cells are non-fluorescent. Florescence was measured with a flow cytometer EPICS-XL (Beckman Coulter, Hialeah, Fl, USA) using argon laser excitation at 488 nm (blue light). FDA fluorescence emission was collected at 530 nm (green) and PI fluorescence emission at 620 nm (red). Fluorescence intensities were expressed on a logarithmic mode. Three independent experiments were conducted.

Cells were plated at 10,000/cm² in DMEM/F12 plus 10% FBS, grown for 3 days, then either treated with ethanol or not treated. Ethanol-treated and control cells were collected after detachment by a 3 min incubation in the presence of trypsin/EDTA. Cells were rinsed twice with ice-cold Hank's balanced salt solution (HBSS) and centrifuged for 5 min at 300 g. Cells were then incubated in 1 ml of Phosphate Buffered Saline (PBS) containing 50 μg/ml of PI and 5 ng/ml of FDA for 15 min at 37 °C and kept on ice until assay.

DNA fragmentation assay

DNA integrity was analyzed by agarose gel electrophoresis, as detailed elsewhere [25]. Briefly, 1.5x10⁶ cells were pelleted by centrifugation (1600 rpm) and were incubated in 30 μl of lysing buffer (10 mM Tris-HCl pH 7.4, 100 mM NaCl, 25 mM EDTA, and 1% N-lauryl Sarkosyl) by gentle vortexing, and digested with 4 μl of proteinase K (10 μg/μl) for 1-2 h at 45 °C. The sample was then incubated in 2 μl of RNase (10 μg/μl) for 1 h at room temperature. The cell lysate was added directly to the wells of a 1.8% agarose gel. Electrophoresis was carried out in 90 mM Tris-Borate buffer containing 1 mM EDTA, and DNA was visualized by ethidium bromide staining.

Immunocytochemistry

ARPE-19 cells were seeded in 6 well plates. After incubation and treatment cells were washed twice with HBSS and then fixed in 4% paraformaldehyde (10 min) and washed twice with PBS. The cells were then permeabilized with Triton 0.03% (15 mn). Non-specific protein binding sites were blocked by incubation in a blocking buffer containing 5% skim milk in PBS (45 min). Cells were then incubated with L-DNase II polyclonal antibody at a 1:100 dilution in 1% skim milk in PBS. This was followed by five washes with 1% skim milk in PBS and incubation for 1 h with an 1/200 dilution of fluorescein iso-thio-cyanate (FITC)-conjugated goat anti-rabbit IgG and rhodamine-conjugated goat anti-mouse IgG (Pierce, Perbio, Brévières, France). Cells were finally washed with 1% skim milk in PBS, incubated for 5 min with 4'-6-di-amindino-2-phenyl indoledichloride (DAPI) and washed twice with PBS. Immunoreactivity was visualized using fluorescence microscopy with an Aristoplan microscope (Leitz, France). Cells that did not stain with FITC-conjugated secondary antibodies alone and cells stained with isotypically matched control immunoglobulin were run in parallel as negative controls.

Preparation of nuclei, cytoplasm, and total protein extract

About 3x10⁷ cells were used for preparation of the nuclear extract. Cells were grown in 150 cm² culture flasks and then either treated with ethanol or not treated. Control cells were collected after treatment by trypsin/EDTA. Cells treated with ethanol were collected from supernatant. After collection, cells were washed twice with ice-cold HBSS and pelleted by centrifugation (1600 rpm). Pellets were resuspended in 800 microliter of 1.5 mM MgCl₂, and homogenized using a Dounce homogenizer using a type B pestle (20 strokes) in an ice bath. The extract was centrifuged for 10 min at 300 g. The pellet containing the nuclei was considered the nuclear fraction, and the supernatant as the cytoplasmic fraction. The nuclear fraction was solubilized in 100 μl of radio immuno-precipitation assay (RIPA) buffer (50 mM Tris-HCl pH 7.4, 50 mM EDTA pH 8.0, 150 mM NaCl, 1% Triton, 2% SDS, 10 mM DOC, and 2 mM PMSF), and centrifuged at 10,000 g for 10 min. The supernatant, designated the nuclear extract, was stored at -20 °C. The cytoplasm fraction was concentrated with a Centricon device (Millipore, Saint-Quentin-en-Yvelines, France; cut off 5 kDa). The resulting volume was mixed with 2X RIPA buffer (v/v) and then stored at -20 °C. In the case of the total protein extract, cells were solubilized as described for the nuclear fraction, after washing with ice-cold PBS and centrifugation.

SDS-PAGE analysis

Protein concentration was determined using the BCA assay (Pierce Chem. Co., Rockford, Il) according to the manufacturer' specifications. Forty μg of protein from each sample were diluted with Laemli sample buffer and boiled for 5 min. The proteins were separated on 14% polyacrylamide gels at 25 mA/gel constant current for 2 h. A ten μl sample of prestained protein ladder (BenchMark, Gibco BRL) was included in the gels for calibration. Gels were either stained with Coomassie brilliant blue (Sigma, Saint-Quentin Falavier, France) for visualization of the electrophoretic pattern or transferred onto PVDF membranes (Immobilon, Millipore Corp., Bedford, MA) for western blot analysis.

Western blot analysis of PARP and LEI

The PVDF membranes were pre-treated with 5% skim milk in PBS (pH 7.4), containing 0.1% Tween, followed by incubation for 1 h with polyclonal anti-B antibody with the polyclonal antibody prepared against B peptide 181-191 from LEI, diluted at 1:1000 in PBS-Tween 20 (0.1%), or anti-PARP antibody (Zymed, Cliniscience, Montrouge, France). After this treatment, the membranes were washed 15 min with PBS-Tween 20 (0.1%) and subjected to further incubation for 1 h with horseradish peroxidase conjugated human anti-rabbit IgG. ECL+ reagent (Amersham Pharmacia biotech, UK) was used for the detection of peroxidase activity according to the manufacturer' specifications.

Reverse transcriptase-PCR analysis of LEI

ARPE-19 cells were seeded in 6 well culture plates. After three days of culture, cells were treated with ethanol. Every ten minutes total RNA was extracted from one well using Tri-Instapure (Eurogentec, Serain, Belgium) according to the manufacturer's specifications. RNA samples (0.4 μg) were subjected to reverse transcription for 60 min at 42 °C with 200 U of Moloney murine leukemia virus reverse transcriptase (Life Technologies SARL, Eragny, France), and random hexamers (70 μM) in retrotranscript buffer (50 mM Tris-HCl pH 8.3, 3 mM MgCl₂, 65 mM KCl, 0.5 mM dNTPs, 5 mM DDT, and 24 U RNase inhibitor). Heating to 95 °C for 5 min terminated the reaction. Six μl of cDNA was then used for RT-PCR. To verify a uniform amplification process, β-actin mRNA was also transcribed and amplified for each sample. PCR mixtures contained PCR buffer (75 mM Tris-HCl pH 9.0, 20 mM (NH₂)2SO₄, 0.01% Tween 20, 200 μM dNTP, 1.5 mM MgCl₂) and 1.25 U of Taq polymerase (Eurogentec, Belgium) in a total volume of 50 μl. Amplification was performed as follows: 94 °C for 1 min, 94 °C for 30 s, 55 °C for 30 s, 72 °C for 45 s, and then 72 °C for 10 min. β-Actin primers were included to verify that equal amounts of RNA were added in each PCR reaction within an experiment. Reactions were routinely run 38 cycles for human LEI and 24 cycles for β-actin. The amplified fragments were separated in a 1.4% agarose gel and visualized after ethidium bromide staining. The intensity of the bands was quantified by using a high-resolution camera coupled to an image processor (Ultra-Lum Inc., Carson, CA) driven by One Descan densitometric software (Scanalytics, Billerica, MA).

The nucleotide sequences of the oligonucleotide primers used for RT-PCR were as follows: human β-actin anti-sense (AGG GGC CGG ACT CGT CAT AC), human β-actin sense (AGG AGA AGC TGT GCT ACG TC), human LEI sense (CAT TCA AGA TTC CAG AGT CTG), and human LEI anti-sense (AAA GAG ATC CTG CAC ACC TAG).

Statistics

Each figure shows the results of experiments repeated at least three times. All data are expressed as the mean plus or minus the standard error of the mean. Statistical comparisons were performed using two-tailed Student's test (Gaussian populations with equal standard deviation) and the Wilcoxon or Mann and Whitney test (non-parametric).

Results

The Effect of ethanol on cell viability

To determine the cytotoxicity of ethanol on cultured ARPE-19 cells, cultured cells were exposed to 0-13% ethanol for 24 h. Cytotoxicity was estimated by flow cytometry, using FDA and isotonic PI methods.

Figure 1A shows a three-dimensional distribution map of the cells. In Figure 1A control cells are shown in the left graph and cells treated with 10% ethanol are shown in the right graph. For control cells, we observed that 87.2% of the population emitted an intense green fluorescence, but no red fluorescence. This percentage represented viable ARPE-19 cells. For ethanol-treated cells, we observed a large peak representing 98.0% of the cell population that emitted an intense red fluorescence and little green fluorescence. This peak represented non-viable ARPE-19 cells. Under these conditions less than 1% of cells were alive.

Using this method, the percentage of viable cells was calculated at different ethanol concentrations and reported in the histogram shown in Figure 1B. The data indicate that cellular viability decreases with ethanol concentration. All cells were dead after 24 h exposure to 10% ethanol.

Effects of ethanol on DNA Fragmentation

DNA cleavage is a characteristic event that occurs during cell death. Three approaches were undertaken to identify DNA fragmentation in ethanol treated ARPE-19 cells. In the first approach, flow cytometry analysis was used to quantify DNA fragmentation. Results shown in Figure 2A represent DNA patterns after 24 h of treatment with 10% and 13% ethanol. A similar pattern was observed for cells treated with 10% ethanol and for untreated cells (results not shown). The majority of cells are in G1 and S phases. In these experiments, 7.8% of cells incubated with 10% ethanol and 38.8% of cells incubated with 13% ethanol showed a sub-G1 peak reflecting a DNA degradation. The percentage of cells in sub-G1 for every ethanol concentration was calculated and reported in Figure 2B. In the presence of 10% ethanol, most of the cells were in G1 phase, whereas in the presence of 13% ethanol, the numbers of cells in sub-G1 phase increased 5 fold. Moreover, the record sub-G1 peak generated above 10% ethanol was not detached from the G1 peak, suggesting a partial degradation of the DNA.

To ensure that all cells were killed and to avoid a heterogeneous cell population, we used 10% ethanol to induce cell death in the following experiments. In order to better characterize DNA degradation induced by 10% ethanol exposure, agarose gel electrophoresis was used to detect small DNA fragments. The results are shown in Figure 3. Intact DNA was seen for control cells and a smear of damaged DNA was observed for treated cells. This is in agreement with the pattern previously seen by flow cytometry analysis.

Finally, nuclear morphology was examined by fluorescence microscopy after DAPI staining. Compared with untreated cells (Figure 4A), distinct morphological changes characterized by chromatin condensation and nuclear shrinkage were noted in ethanol-treated cells (Figure 4C).

Taken together, these results indicate that ethanol-treated ARPE-19 cells degrade their DNA in a dose-dependent fashion. The pattern of DNA cleavage is a smear. However, the nuclei present exhibit condensed chromatin, a feature related to an apoptotic-like cell death.

Effects of ethanol on PARP Cleavage

A central event in some cell death mechanisms is the activation of cysteine proteases called caspases [10]. Among the different caspases, caspase-3 plays a major function in the execution of the cell death process [26]. Activation of caspase-3 results in cleavage of proteins with the amino acid sequence aspartate-glutamate-valine-aspartate (DEVD), such as poly(ADP-ribose)polymerase (PARP) [27]. The products of PARP cleavage are distinctively different in necrosis and apoptosis. During apoptosis, PARP generates an 89 kDa fragment, but during necrosis PARP cleavage generates several fragments with a predominant product of 50 kDa [28,29]. In order to better characterize ethanol-induced cell death, we performed western blot analyses using an antibody against PARP that recognizes an epitope in the DNA binding domain of PARP [30]. Figure 5 shows intact PARP at 116 kDa in control cells (lane 1). The ethanol treated cells (lane 2) were characterized by intact PARP and a major band of 89 kDa. Minor additional degradation products of PARP at 50 kDa were also observed. Intact and cleaved PARP bands correspond to those found in a HL-60 cell lysate and an etoposide-treated HL-60 cell lysate from Zymed [31], which were used as controls (lanes 3 and 4). This observation implies that, in spite of the DNA degradation pattern observed, the cell death induced by ethanol in ARPE-19 leads to cleavage of PARP in an apoptotic-like fashion.

L-DNase II is involved in ARPE-19 cell death

We have reported that L-DNase II is derived from LEI and that DNase-II is implicated in DNA degradation occurring in several models of cell death [22,32]. We investigated whether this DNase was also implicated in cell death induced by ethanol. Two approaches were undertaken.

First, immunocytochemistry was performed. At 24 h after inducing cell death, the cultured cells were fixed and treated for immunofluorescence with anti-L-DNase-II, which does not recognize LEI. In the case of untreated cells (Figure 4B), we observed that L-DNase II labeling was very faint and was localized to the cytoplasm. After induction of cell death for 24 h with 10% ethanol (Figure 4D), condensed cells showed a strong increase in anti L-DNase II labeling. The labeling was not confined to the cytoplasm but was also extended to the nucleus of the cells.

In a second approach, western blot analysis was used with an antibody directed against the 181-191 peptide of LEI that recognizes both LEI and L-DNase II. For untreated cells we observed, in the total protein extract (Figure 6, lane 1), a faint band, which represented a protein of 42 kDa. The molecular weight of this protein corresponds to the molecular weight of LEI, precursor of L-DNase II. It is interesting to note the presence of an immunoreactive protein of 38 kDa, which may represent an intermediate between LEI and L-DNase II.

In the total protein extract for cells treated with ethanol, there was an increase of expression of the immunoreactive 42 kDa protein and the presence of a major band of 24 kDa (Figure 6, lane 2). The size of this protein corresponds to the molecular weight of human L-DNase II [32]. We also noted the disappearance of the immunoreactive protein of 38 kDa.

Our results suggest that treatment of ARPE-19 cells for 24 h with 10% ethanol induces an increased expression of LEI, precursor of L-DNase II, and the presence of L-DNase II. L-DNase II is not present in normal cells. This suggests that L-DNase II is implicated in ethanol-induced ARPE-19 cell death.

Protein synthesis in ethanol-induced apoptosis

The results described above suggested that L-DNase II was activated and that its synthesis was increased. To verify this, we first performed immunocytochemistry analysis in the presence of protein synthesis inhibitor. One hour before inducing cell death with 10% ethanol, the cultured cells were treated with cycloheximide (CHM). After ethanol treatment, the cultured cells were fixed and stained for immunofluorescence with anti-L-DNase II. Under these conditions, cells showed less condensed nuclei (Figure 4E) than cells treated only with ethanol alone (Figure 4A) as well as a diffuse cytoplasmic L-DNase II labeling (Figure 4F). This may suggest that CHM is inhibiting LEI synthesis and also the apoptotic process.

In a second approach, we investigated the expression of LEI mRNA. Total RNA was extracted from cells cultured in the presence of 10% ethanol for different periods of time. LEI mRNA and β-actin mRNA expression were examined by RT-PCR analysis. PCR products (amplified by using LEI-specific primers or β-actin-specific primers) showed a specific band at the predicted size of 420 bp for LEI and 243 bp for β-actin (Figure 7A). The relative ratios of LEI and β-actin PCR product quantities were determined by densitometric analysis. Three experiments were carried out at each time point. We observed that LEI mRNA expression increased very rapidly after initial exposure to ethanol, with a maximal 2.5 fold increase at 30 min, and remained high at 120 min (Figure 7B).

We conclude that treatment of ARPE-19 cells for 24 h with 10% ethanol induced an increase in LEI mRNA expression. This expression is increased soon after inducing cell death by exposure to ethanol, and persists for at least 24 h.

Discussion

In this study we have investigated the effects of ethanol in ARPE-19 cells, a cell line derived from human RPE. The results obtained by labelling the cells with FDA indicate that 10% ethanol was capable of killing the entire population of cultured cells, whereas 50% of the cells died after incubation with 4% ethanol. In order to characterize this cell death, we have studied DNA degradation. Flow cytometry indicates a sub-G1 DNA population while gel electrophoresis shows the absence of DNA ladders. Interestingly, cell death induced by another oxidizing agent (tert-butyl hydroperoxide) shows a similar form of DNA degradation [33]. Both techniques indicate a "necrotic-like" DNA degradation [34]. However, other indicators, i.e., the degradation of PARP and the inhibition of cell death by cycloheximide, suggest an apoptotic-like cell death. During apoptosis, PARP is cleaved into an 89 kDA fragment, while during necrosis it is cleaved into fragments of differing molecular weight [28]. In addition, cycloheximide inhibits apoptotic cell death, whereas necrosis is usually independent of protein synthesis [35]. The morphology of the cells also suggests an apoptotic process since shrinkage of both cytoplasm and nucleus is seen. These phenomena are usually related to apoptosis [36]. It seems then that ARPE-19 cells induced to die by ethanol undergo cell death with characteristics of both apoptosis and necrosis. Similar features have been reported in other cells [24,25,37]. According to accepted criteria, it seems that ethanol induces an apoptosis with a necrotic-like DNA degradation in ARPE-19 cells [38]. Notably, this degradation occurs quite late in the death of this cell line, as illustrated by the gap seen between the rate of death (Figure 1B) and the amount of DNA degradation (Figure 2B), that is, DNA degradation seems to occur long after the initial insult, as opposed to other models of apoptosis in which death begins within minutes of initial insult.

Thus, we investigated whether the LEI-L-DNase II pathway is activated in ARPE-19 cells. Western blot experiments indicate a transformation of human LEI into L-DNase II, as identified by molecular weight (42 kDa LEI, 24 kDa for L-DNase II). Moreover, in ethanol induced apoptosis, the entire 38 kDa band, representing the intermediate form between LEI and L-DNase II, disappears, leading to complete activation of L-DNase II. In these experiments we have also noticed an increase in the 42 kDa band compared to the controls, suggesting an increased synthesis of LEI. This is also in agreement with immunocytological experiments that show an increase in cell labeling with anti-L-DNase II antibody when apoptosis is induced. These results suggest a neosynthesis of LEI. This protein neosynthesis is also the result of the specific increase in transcription of LEI mRNA, as shown by RT-PCR. It is interesting to note that this increase in LEI mRNA is a very early event: a 50% increase in mRNA is detected only 10 min after adding ethanol to the cultured layer, and the maximum amount of mRNA is identified 20 min later. This is shorter than the time needed to activate caspases in Fas-induced apoptosis [39]. Also, the rate of increase in LEI mRNA is higher than for other molecules implicated in apoptosis [40].

In this study we observed cleavage of PARP, which is converted from the 116 kDa form to a 89 kDa fragment, a feature which is characteristic of apoptosis [31]. This cleavage indicates that caspases 3 and/or 7 are also activated and may be implicated in the fate of the cell. This issue merits investigation. However, in HeLa cells induced to die by long term culture [32], we found the same cleavage of PARP could be inhibited by caspase inhibitors. This inhibition did not have any influence on the apoptotic fate of cells, such as ARPE-19 cells, that activate the LEI/ L-DNase II pathway [41]. Moreover, cycloheximide inhibits LEI synthesis (Figure 4F) and protects cells against apoptosis, indicating that synthesis of LEI is important to induce apoptosis in these cells.

In conclusion, we have determined that ethanol-induced death in ARPE cells appears to be, at least in part, apoptotic, and that activation of the LEI/L-DNase II pathway may underlie this apoptosis. Whether the nature of stress induced by ethanol in this condition results from oxidative stress, membrane modification, or both, must be resolved in future experiments.

Acknowledgements

ARPE-19 cells were a kind gift of Dr. L. M. Hjelmeland (University of California, Davis, CA). This work was supported in part by a grant from Retina France (AT).

References

1. Hjelmeland LM, Cristofolo VJ, Funk W, Rakoczy E, Katz ML. Senescence of the retinal pigment epithelium. Mol Vis 1999; 5:33 <http://www.molvis.org/molvis/v5/a33/>.

2. Matsunaga H, Handa JT, Gelfman CM, Hjelmeland LM. The mRNA phenotype of a human RPE cell line at replicative senescence. Mol Vis 1999; 5:39 <http://www.molvis.org/molvis/v5/a39/>.

3. Honda S, Hjelmeland LM, Handa JT. The use of hyperoxia to induce chronic mild oxidative stress in RPE cells in vitro. Mol Vis 2001; 7:63-70 <http://www.molvis.org/molvis/v7/a10/>.

4. Miyamura N, Mishima K, Honda S, Aotaki-Keen AE, Morse LS, Handa JT, Hjelmeland LM. Age and topographic variation of insulin-like growth factor-binding protein 2 in the human rpe. Invest Ophthalmol Vis Sci 2001; 42:1626-30.

5. Honda S, Weigel A, Hjelmeland LM, Handa JT. Induction of telomere shortening and replicative senescence by cryopreservation. Biochem Biophys Res Commun 2001; 282:493-8.

6. Honda S, Hjelmeland LM, Handa JT. Oxidative stress--induced single-strand breaks in chromosomal telomeres of human retinal pigment epithelial cells in vitro. Invest Ophthalmol Vis Sci 2001; 42:2139-44.

7. Alizadeh M, Wada M, Gelfman CM, Handa JT, Hjelmeland LM. Downregulation of differentiation specific gene expression by oxidative stress in ARPE-19 Invest Ophthalmol Vis Sci 2001; 42:2706-13.

8. Honda S, Farboud B, Hjelmeland LM, Handa JT. Induction of an aging mRNA retinal pigment epithelial cell phenotype by matrix-containing advanced glycation end products in vitro. Invest Ophthalmol Vis Sci 2001; 42:2419-25.

9. Weigel AL, Handa JT, Hjelmeland LM. Microarray analysis of H2O2-, HNE-, or tBH-treated ARPE-19 cells. Free Radic Biol Med 2002; 33:1419-32.

10. Carmody RJ, Cotter TG. Oxidative stress induces caspase-independent retinal apoptosis in vitro. Cell Death Differ 2000; 7:282-91.

11. Assefa Z, Vantieghem A, Garmyn M, Declercq W, Vandenabeele P, Vandenheede JR, Bouillon R, Merlevede W, Agostinis P. p38 mitogen-activated protein kinase regulates a novel, caspase-independent pathway for the mitochondrial cytochrome c release in ultraviolet B radiation-induced apoptosis. J Biol Chem 2000; 275:21416-21.

12. Mathiasen IS, Lademann U, Jaattela M. Apoptosis induced by vitamin D compounds in breast cancer cells is inhibited by Bcl-2 but does not involve known caspases or p53. Cancer Res 1999; 59:4848-56.

13. Deshmukh M, Johnson EM Jr. Staurosporine-induced neuronal death: multiple mechanisms and methodological implications. Cell Death Differ 2000; 7:250-61.

14. Mateo V, Lagneaux L, Bron D, Biron G, Armant M, Delespesse G, Sarfati M. CD47 ligation induces caspase-independent cell death in chronic lymphocytic leukemia. Nat Med 1999; 5:1277-84.

15. Dierick JF, Eliaers F, Remacle J, Raes M, Fey SJ, Larsen PM, Toussaint O. Stress-induced premature senescence and replicative senescence are different phenotypes, proteomic evidence. Biochem Pharmacol 2002; 64:1011-7.

16. Skoog KO, Textorius O, Nilsson SE. Effects of ethyl alcohol on the directly recorded standing potential of the human eye. Acta Ophthalmol (Copenh) 1975; 53:710-20.

17. Arden GB, Wolf JE. The human electro-oculogram: interaction of light and alcohol. Invest Ophthalmol Vis Sci 2000; 41:2722-9.

18. Arden GB, Wolf JE. The electro-oculographic responses to alcohol and light in a series of patients with retinitis pigmentosa. Invest Ophthalmol Vis Sci 2000; 41:2730-4.

19. Arden GB, Wolf JE, Singbartl F, Berninger TE, Rudolph G, Kampik A. Effect of alcohol and light on the retinal pigment epithelium of normal subjects and patients with retinal dystrophies. Br J Ophthalmol 2000; 84:881-3.

20. Arden GB, Wolf JE. Differential effects of light and alcohol on the electro-oculographic responses of patients with age-related macular disease. Invest Ophthalmol Vis Sci 2003; 44:3226-32.

21. Tenkova T, Young C, Dikranian K, Labruyere J, Olney JW. Ethanol-induced apoptosis in the developing visual system during synaptogenesis. Invest Ophthalmol Vis Sci 2003; 44:2809-17.

22. Torriglia A, Chaudun E, Chany-Fournier F, Jeanny JC, Courtois Y, Counis MF. Involvement of DNase II in nuclear degeneration during lens cell differentiation. J Biol Chem 1995; 270:28579-85.

23. Torriglia A, Perani P, Brossas JY, Chaudun E, Treton J, Courtois Y, Counis MF. L-DNase II, a molecule that links proteases and endonucleases in apoptosis, derives from the ubiquitous serpin leukocyte elastase inhibitor [published erratum in Mol Cell Biol 1998; 18:4947]. Mol Cell Biol 1998; 18:3612-9.

24. Hamatake M, Iguchi K, Hirano K, Ishida R. Zinc induces mixed types of cell death, necrosis, and apoptosis, in molt-4 cells. J Biochem (Tokyo) 2000; 128:933-9.

25. Liu J, Akahoshi T, Jiang S, Namai R, Kitasato H, Endo H, Kameya T, Kondo H. Induction of neutrophil death resembling neither apoptosis nor necrosis by ONO-AE-248, a selective agonist for PGE2 receptor subtype 3. J Leukoc Biol 2000; 68:187-93.

26. Thornberry NA, Lazebnik Y. Caspases: enemies within. Science 1998; 281:1312-6.

27. Keane RW, Srinivasan A, Foster LM, Testa MP, Ord T, Nonner D, Wang HG, Reed JC, Bredesen DE, Kayalar C. Activation of CPP32 during apoptosis of neurons and astrocytes. J Neurosci Res 1997; 48:168-80.

28. Shah GM, Shah RG, Poirier GG. Different cleavage pattern for poly(ADP-ribose) polymerase during necrosis and apoptosis in HL-60 cells. Biochem Biophys Res Commun 1996; 229:838-44.

29. Casiano CA, Ochs RL, Tan EM. Distinct cleavage products of nuclear proteins in apoptosis and necrosis revealed by autoantibody probes. Cell Death Differ 1998; 5:183-90.

30. Shah GM, Poirier D, Duchaine C, Brochu G, Desnoyers S, Lagueux J, Verreault A, Hoflack JC, Kirkland JB, Poirier GG. Methods for biochemical study of poly(ADP-ribose) metabolism in vitro and in vivo. Anal Biochem 1995; 227:1-13.

31. Duriez PJ, Desnoyers S, Hoflack JC, Shah GM, Morelle B, Bourassa S, Poirier GG, Talbot B. Characterization of anti-peptide antibodies directed towards the automodification domain and apoptotic fragment of poly (ADP-ribose) polymerase. Biochim Biophys Acta 1997; 1334:65-72.

32. Torriglia A, Negri C, Chaudun E, Prosperi E, Courtois Y, Counis MF, Scovassi AI. Differential involvement of DNases in HeLa cell apoptosis induced by etoposide and long term-culture. Cell Death Differ 1999; 6:234-44.

33. Cai J, Wu M, Nelson KC, Sternberg P Jr, Jones DP. Oxidant-induced apoptosis in cultured human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci 1999; 40:959-66.

34. Walker NI, Harmon BV, Gobe GC, Kerr JF. Patterns of cell death. Methods Achiev Exp Pathol 1988; 13:18-54.

35. Fiers W, Beyaert R, Declercq W, Vandenabeele P. More than one way to die: apoptosis, necrosis and reactive oxygen damage. Oncogene 1999; 18:7719-30.

36. Hengartner MO. The biochemistry of apoptosis. Nature 2000; 407:770-6.

37. Tan S, Wood M, Maher P. Oxidative stress induces form of programmed cell death with characteristics of both apoptosis and necrosis in neuronal cells. J Neurochem 1998; 71:95-105.

38. Nicotera P, Leist M, Ferrando-May E. Apoptosis and necrosis: different execution of the same death. Biochem Soc Symp 1999; 66:69-73.

39. Bajt ML, Lawson JA, Vonderfecht SL, Gujral JS, Jaeschke H. Protection against Fas receptor-mediated apoptosis in hepatocytes and nonparenchymal cells by a caspase-8 inhibitor in vivo: evidence for a postmitochondrial processing of caspase-8. Toxicol Sci 2000; 58:109-17.

40. Ray SK, Matzelle DC, Wilford GG, Hogan EL, Banik NL. E-64-d prevents both calpain upregulation and apoptosis in the lesion and penumbra following spinal cord injury in rats. Brain Res 2000; 867:80-9.

41. Torriglia A, Perani P, Brossas JY, Altairac S, Zeggai S, Martin E, Treton J, Courtois Y, Counis MF. A caspase-independent cell clearance program. The LEI/L-DNase II pathway. Ann N Y Acad Sci 2000; 926:192-203.