Copyright 2005 by Humana Press Inc. All rights of any nature whatsoever reserved. 1085-9195/05/42:6174/$30.

00

ORIGINAL ARTICLE

Exposure of Human Leukemic Cells to Direct Electric Current

Generation of Toxic Compounds Inducing Cell Death by Different Mechanisms

Venicio F. Veiga,1 Leonardo Nimrichter,1 Cesar A. Teixeira,3 Marcelo M. Morales,2 Celuta S. Alviano,1 Marcio L. Rodrigues,1 and Carla Holandino3,*
de Microbiologia Professor Paulo de Ges, 2Instituto de Biofsica Carlos Chagas Filho, and de Medicamentos-Faculdade de Farmcia, Centro de Cincias da Sade, Universidade Federal do Rio de Janeiro, CCS, Bloco K, Segundo Andar, Sala 50, Ilha do Fund~ ao, 21941590, Rio de Janeiro, Brazil
3Departamento 1Instituto

Abstract
Treatment with direct electric current (DC) influences the growth of several cancer cells. In this work, we evaluated the effects of DC treatment on the human leukemic cell line HL60. Human cells were separately treated in the presence of the cathode or the anode or without contact with the electrodes. In all systems, DCtreated cells presented an impaired ability to proliferate. Growth inhibition was dependent on the generation of soluble products of electrolysis. Cathodic treatment of HL60 cells predominantly induced lysis, whereas treatment without contact with electrodes did not induce alterations in cell viability. In contrast, cell stimulation by the anode resulted in irreversible membrane damage, as demonstrated by trypan blue and 7aminoactinomycin staining. Analysis of these cells by transmission electron microscopy indicated that necrosis is a major mechanism inducing cell death. In addition, apoptotic-like cells were observed under light microscopy after anodic treatment. Accordingly, DNA from anodic-treated cells presented a typical pattern of apoptosis. Apoptotic cell death was only generated after the treatment of HL60 cells in conditions in which the generation of chloride-derived compounds was favored. These results indicate that the nature of the products from cathodic or anodic reactions differently influences the mechanisms of cell death induced by DCderived toxic compounds. Index Entries: Direct electric current; human leukemic cells; tumor growth; apoptosis; necrosis.

INTRODUCTION
Direct electric current (DC) can influence the growth and cell biology of several systems (15). Because of such properties, DC has been widely used for the treatment of tumors (611). However, the use of DC as an alternative therapy for cancer is limited by the poor knowledge of the cellular effects generated in different conditions in which animal cells are exposed to electric fields (2). In this context, different polarities
*Author to whom all correspondence and reprint requests should be addressed. E-mail: cholandino@terra.com.br
Cell Biochemistry and Biophysics

of electric current have been shown to induce diverse effects on animal cells, which is reflected by the varying efficacies of treatments of cancers with DC using positive or negative electrodes (8,1214). In such experimental systems, cells are exposed to DC as they are to the soluble products generated after oxidation or reduction of physiologic compounds (1517). Therefore, the products of electrolysis can certainly influence the cells that are exposed to DC in aqueous systems. Parameters involved in DC treatment such as electric field strength, electrode polarity, and period of stimulation also influence the type of cell response to an electric stimulus. For instance, the suppression or
61
Volume 42, 2005

62

Veiga et al.

enhancement of proliferation of mouse lymphoma cells varied significantly when the cells were electrically stimulated within a narrow range of low-level DC (5). Mechanisms of DC-induced cell growth or inhibition have been described in different experimental systems. For example, DC treatment of multicellular tumor spheroids results in adenosine triphosphate release, which concomitantly activates purinergic receptors, elicits a Ca2+ wave spreading through the tumor spheroid tissue, and stimulates tumor growth (18). Calcium release seems in fact to be involved in DC-induced tumor growth. In a previous study by Wartenberg and coworkers (19), a single electrical eld pulse induces production of reactive oxygen species, which mediated release of intracellular Ca2+ and activated cell-cycle activity in multicellular spheroids tumors. On the other hand, DC-treated tumor cells can have their growth inhibited, depending on cell type and conditions of electric stimulation. The mechanisms involved in such inhibition include changes in membrane conformation, alterations in cell organization, and appearance of mitochondrial lesions (15,16,20). Apoptosis is another biologic process apparently associated with DC-induced tumor reduction. Kurokawa and coworkers (21) described that treatment of human leukemic cell lines with DC intensities varying from 0.2 to 5 mA-induced cell shrinkage and internucleosomal DNA fragmentation. In their model, cells were simultaneously submitted to anodic and cathodic reactions, which impairs the establishment of a direct relation between the products of electrolysis and the cellular damage generated after stimulation with DC. We have previously shown that the nature of cytotoxicity generated by DC is dependent on the polarity of the electrode used for treatment of tumor cells (16,17). In an experimental model in which cells are separately exposed to cathodic or anodic polarities, we observed that distinct alterations are induced by each independent electrolytic reaction (1517). In the present work, we evaluate the occurrence of cell death after DC stimulation using electrodes of different polarities. Our results show that, besides the cellular effects previously described in other models, toxic compounds produced during anodic stimulation induce apoptosis and necrosis in the human leukemic cell line HL60. The induction of apoptosis in anodic-treated cells was dependent on the generation of chlorine-derived electrolysis products. In contrast, cathodic treatment caused cell lysis and necrosis, but not apoptosis. These results demonstrate the relevance of electrode polarity in the generation of cytotoxic compounds killing cancer cells and help to elucidate the mechanisms involved in the tumor inhibition observed after electric stimulation.

MATERIAL AND METHODS Cell Culture

Human promyelocytic leukemia HL60 cells were obtained from the American Type Culture Collection. Cells were grown at 37C, in 25 cm2 culture asks containing Dulbeccos modied Eagles medium (DMEM) supplemented with 10% fetal bovine serum (FBS). pH control was warranted by the addition of 3 g/L N-(2hydroxyethyl)-piperazine-N-(2-ethanesulfonic acid) and 0.2 g/L NaHCO3 to the medium composition, as previously described (22). The initial inoculum was 5 104 cells/mL, which were subcultured every 2 d and maintained in a log-phase growth, as described elsewhere (17,23).

DC Stimulation
Human cells were collected by centrifugation, washed, and suspended (1.0 106 cells/mL) in phosphate-buffered saline (PBS, g/L: 0.20 KH2PO4, 2.17 Na2HPO4-7H20, 8.0 NaCl, 0.20 KCl, pH 7.4) at 308 mOsm. The cellular suspensions were then distributed in a system of three acrylic chambers with internal volumes of 3 cm3, connected in series by filter-paper bridges, and tted with platinum electrodes in their extremities. Chamber conformation and distribution of electrodes are diagramed in Fig. 1. In this system, cell suspensions can be separately exposed to cathodic or anodic reactions, and to electric current without contact with the electrodes, in the intermediary chamber (IC). Cell suspensions (2 mL) were treated with a DC value of 2 mA, which was similar to those used in previous studies (7,16,17,20,24,25). Human cells were stimulated for 0 (control), 2, 4, 6, 8, and 10 min using a DC source (KLD Biossistemas Equipamentos Eletrnicos Ltda, Brazil; model ETM 901). The cross-sectional area of the chamber was 0.6 cm2, as determined by multiplying the depth of the uid (0.6 cm) by the width of the chamber (1 cm). Because the total current passing through the chamber was 2 mA, a current density of 3.3 mA/cm2 was generated, associated with a voltage gradient of 378,38 V/m. In all further experiments, cells were treated for 6 min, an intermediate period that was representative of the alterations in morphology and viability. Treatments were performed at 25C, and temperature variation was monitored with a thermometer. The higher temperature achieved was 25.4C, after 10 min of stimulation, indicating that DC treatment did not signicantly increase temperature levels. Control systems consisted of exposure of human cells to the same chambers exactly as described previously, except for the presence of electric current (Fig. 1B). Alternatively, an additional experiment was performed after reduction of the internal volume of the exposure chamber, allowing treatments using 1.0 and 1.5

Cell Biochemistry and Biophysics

Volume 42, 2005

Generation of Toxic Compounds Inducing Cell Death

63

Fig. 1. Graphic representation of the in vitro system used for direct electric current (DC) treatment of HL60 cells (A). The cell suspensions are distributed over each individual chamber (AC for anodic chamber, IC for intermediary chamber, and CC for cathodic chamber). Platinum electrodes are inserted into the lateral chambers (AC and CC), allowing the system to be connected to a DC source. This distribution permits the separate exposition of cells to cathodic or anodic reactions, and to electric current without contact with the electrodes (IC). Control cells are exposed to the same conditions, except for the use of DC. In this case, they are placed into a control cuvette (B), in which they have direct contact with the platinum electrode and the lter paper bridge. Cell populations treated in this chamber were called nonstimulated cells (NS). Internal volume: 3 cm3.

mL cell suspensions. Variations in electric conductivity during DC stimulation were monitored with a conductivity meter (Microsiemens).

Inuence of the Electrolytic Medium in Cellular Damage

To establish a relationship between the composition of the electrolytic medium and cellular damage, HL60 cells (106/mL) were treated with DC in 2 mL DMEM (without FBS) for 6 min. After stimulation, HL-60 cells were cultivated for 24 h in a 24-well plate at 37C in two different situations: (1) at the same medium of treatment or (2) in fresh DMEM after removal of the medium of stimulation by centrifugation. Alternatively, the medium of treatment of stimulated HL60 cells was used to cultivate untreated cells after the removal of DCtreated populations by centrifugation. To study the toxic effects of chlorine-derived compounds, which are generated in the reaction of electrolysis products with amino acids, HL-60 cells were also treated with DC in PBS supplemented with glutamine (0.5, 1.0, and 1.5 mM) for further cultivation in DMEM for 0, 1, and 3 h. The number of viable cells in the different systems was determined as previously described (16,17), and in all experimental situations the suspensions containing similar cell numbers were cultivated in a 24-well plate. Aliquots of 25 L of the cell culture were then taken after 0, 1, and 3 h, and the total number of cells determined in a Neubauer chamber. Human cells stimulated in these conditions were also screened for the occurrence of apoptosis, as described in a following section.

Cell Growth
To evaluate the inuence of DC treatment on the proliferation of leukemic cells, they were suspended in 2 mL PBS (106 cells/mL) and treated for 6 min with DC. Aliquots of 400 L of the cell suspensions were taken, washed in PBS, counted in a Neubauer chamber, and suspended in 2 mL DMEM supplemented with FBS. The number of viable cells in the different systems was determined as previously described (1517) and the suspensions containing similar cell numbers cultivated in a 24-well plate. Aliquots of 25 L of the cell culture were taken after 0, 4, 18, and 24 h, and the total number of cells determined in a Neubauer chamber.

Cell Viability
Cell viability was evaluated by the trypan blue method, through the determination of the number of remaining cells in a Neubauer chamber after DC treatment, as previously described (16,17,20,27,28). Nonviable cells were divided in two classes, which were (1) the trypan blue-stained cells that retained their shape and (2) lysed leukemic cells. The number of nonviable, lysed cells was considered as the difference between the initial number of cells added to the chamber and the number of remaining cells.

Cell Biochemistry and Biophysics

Volume 42, 2005

64

Veiga et al.

Flow Cytometry With 7-Aminoactinomycin

The progressive increase in membrane permeability induced by DC treatment was evaluated by 7aminoactinomycin (7-AAD) incorporation (29,30). Before 7-AAD staining, a cell suspension (2 mL) containing 106 cells/mL was treated with DC and washed twice in PBS. DC-treated and control cells were immediately incubated with 7-AAD or, alternatively, reintroduced into the culture medium and incubated for 0, 4, 18, or 24 h, as described previously. Control cells were also introduced into the treatment chambers and incubated as performed for stimulated cells, but without any current ow. After these periods, cells were again washed and incubated in a solution of 7-AAD at 20 g/mL (PBS) for 20 min at 4C and protected from light. Cells in this staining solution were then analyzed (10,000 events) in a FACS CALIBUR ow cytometer (Becton Dickinson, Franklin Lakes, NJ). The red uorescence from 7-AAD was ltered through a 650-long pass lter. Multiparameter data analysis was performed with Winmdi software (Salk Flow Cytometry).

PBS and pelleted by centrifugation. DNA fragmentation was assayed by a modication of the method of Duke and Sellins groups (32). Cultured cells were washed twice with ice-cold PBS and suspended in 100 L lysis buffer (10 mM Tris HCl/10 mM EDTA/0.5% Triton X100, pH 8.0), vortex-mixed, sonicated, and incubated on ice for 20 min. After centrifugation for 20 min at 4C (14,000g), the supernatant containing fragmented (soluble) DNA was transferred to another tube. Lysis buffer (100 L) was added to the pellet containing insoluble DNA. Both samples were treated with Rnase A (0.5 mg/mL) for 1 h at 37C and then with proteinase K (Sigma, 0.4 mg/mL) for 1 h at 37C. After adding 20 L 5 M NaCl and 120 L isopropanol, the samples were incubated overnight at 20C. The soluble fraction of DNA was determined by electrophoresis on 1.5% agarose gel and has a ladder-like appearance.

RESULTS Treatment of HL60 Cells With DC Inhibits Proliferation

Treatment with DC has been shown to significantly inhibit the growth of several tumor cells (2,711,14,17,20,24,3343), which led us to evaluate the influence of DC treatment on the proliferation rates of HL60 human leukemic cells. They were treated with a DC intensity of 2 mA for 6 min, according with previously described models of electric stimulation of animal cell lines (16,17,20). Control cells were exposed to the same conditions of stimulation, except for the presence of electric current. DC-treated cells were washed to remove products of electrolysis and reinoculated (105) into fresh media; the number of cells in culture was then determined after 0, 4, 18, and 24 h. Cells that were treated with DC in anodic or cathodic chambers, as well as without contact with electrodes, were unable to proliferate in normal rates (Fig. 2).

Electron Microscopy
Ultrastructural alterations in leukemic cells after DC treatment were evaluated by transmission electron microscopy. Cells were first centrifuged and resuspended in a 2.5% glutaraldehyde solution. After 2 h of incubation in this solution, cells were washed in 0.1 M cacodylate buffer (pH 7.2) containing 0.2 M sucrose and postxed in 1% osmium tetroxide for 45 min. After xation, they were preincluded in 1.5% agar, dehydrated with ethanol and acetone, and embedded in Epon (31). Ultrathin sections were prepared with a diamond knife in an KLB ultramicrotome, collected in 300 mesh cooper grids, counterstained with uranyl acetate and lead citrate, and examined under a Zeiss 900 transmission electron microscope operating at 80 kV.

Assay for Morphological Changes

The morphological features of DC-treated cells were assessed by staining cytocentrifuged preparations by the May-Grunwald-Giemsa method. At least 200 cells in each preparation were examined with a light microscope. For analysis of cellular alterations, three separate experiments were performed.

Kinetics of Trypan Blue Incorporation by DC-Treated Cells

The mechanisms by which DC treatment inhibits the growth of tumor cells are poorly known. Because the cell membrane is commonly affected by the action of DC (16,20,44,45), we evaluated the immediate effects of electric treatment on cell lysis and membrane integrity of HL60 cells. Cells were treated for periods varying from 0 to 10 min and their viability assayed by the trypan blue method. Treatment of HL60 cells in the cathodic chamber (CC) predominantly induced lysis, which was greater in higher periods of treatment (Fig. 3). In contrast, cell stimulation in the anodic chamber (AC) resulted in a progressive augment of permeability to the
Volume 42, 2005

Analysis of DNA Fragmentation

HL60 cells were treated with DC as described previously (DC Stimulation); control systems consisted of exposure of human cells to the chambers under the same conditions, except for the presence of electric current (Fig. 1B). A total of 107 leukemia cells was obtained in combined experiments. The cells were washed with
Cell Biochemistry and Biophysics

Generation of Toxic Compounds Inducing Cell Death

65

Fig. 2. Treatment of HL60 cells with direct electric current (DC) inhibits proliferation. Cells were treated with DC and inoculated into fresh media. The number of cells was determined after 4, 18, and 24 h. Cells that were treated in an anodic chamber (), intermediary chamber (), or cathodic chamber () had their growth inhibited. Proliferation rates of cells that were not treated with DC () are also shown (for details of control preparation, see Material and Methods). Results of ve independent experiments expressed as mean standard deviation are shown.

Fig. 3. Kinetics of trypan blue incorporation by direct electric current (DC)-treated cell. Cells were treated for periods varying from 0 to 10 min and their viability assayed by the trypan blue method. Black bars represent the number of trypan blue-stained cells, whereas white bars are representative of the number of lysed cells. Indices of stained or lysed cells are shown after their treatment in the cathodic (CC), anodic (AC), intermediary (IC), or control (NS) chambers. Results of ve independent experiments expressed as mean standard deviation are shown.
Cell Biochemistry and Biophysics Volume 42, 2005

66

Veiga et al.

Table 1 Effect of Chamber Volume on the Viability of HL60 Cells After Different Periods (min) of DC Treatment Nonviable Cellsa (%) Chamber volume 1.0 mL 1.5 mL 2.0 mL
aNonviable bCathodic-

2 min 60b 33b 33b 37c 33c 37c

4 min 85b 58b 53b 65c 69c 65c

6 min 95b 60b 57b 95c 85c 65c

8 min 99b 72b 58b 100c 99c 83c

10 min 100b 100c 90b 100c 85b 94c

cells were considered as the addition of the number of lysed and trypan-blue stained cells. and canodic-stimulated cells.

trypan blue dye, indicating that membrane integrity was affected by these conditions of electric stimulation. Cell lysis was also observed, although less intensively than observed after cathodic treatment. These results are indicative that the cytoplasmic membrane is indeed a key target for electric stimulation in the presence of electrodes. Expressive levels of lysis or trypan blue staining were not observed when cells were treated in the IC. Exposure of HL60 cells to the chamber of treatment in the absence of DC resulted in high levels of viability. To evaluate the concentration of the products of electrolysis in viability, HL60 cells were stimulated in the same exposure conditions, except for the volume of the cell suspension. In this system, the internal volume of the exposure chamber was reduced by the addition of glass slides at the bottom of the chamber, which made possible the treatment of cell suspensions in the same conditions but in reduced volumes. This experiment demonstrated that the lowest viabilities were generated in smaller volumes (Table 1), indicating that the concentration of the products of electrolysis directly inuences the occurrence of cell death. Cells that were treated in IC or in the CC presented very similar levels of viability, which varied from 80% to 95% of viable cells in the different periods and volumes of treatment (data not shown).

sented an irreversible and progressive augment in their permeability to 7-AAD (Fig. 4, R2), which was accompanied by the generation of cell bodies with a signicantly reduced size (Fig. 4, R3).

Anodic Treatment Induces Necrosis and Apoptosis in HL-60 Cells

The occurrence of 7-AADstained cells presenting reduced cell sizes could be indicative of different processes leading to cell death (46). In fact, necrotic cells are generated after 24 h of cultivation of anodictreated cells, as shown in Fig. 5. The cells presented membrane discontinuity and matrix rarefaction, as observed for DC-treated P815 and multidrug-resistant K562 cells (16,17). Previous studies described that apoptosis is triggered in human leukemic cell lines after treatment with electric current (21). Untreated (control), CC-, and ICstimulated cells presented similar morphological proles (not shown). However, Giemsa staining revealed the occurrence of morphological changes induced by anodic treatment mainly characterized by cell shrinkage and nuclear chromatin clumping (Fig. 6A). Cellular presentations suggestive of apoptotic bodies were also observed, mainly when anodic-treated cells were cultivated for prolonged periods. In fact, biochemical features of apoptosis were detected after DNA analysis by gel electrophoresis (Fig. 6B). DNA extracted from AC-, but not CC- or IC-stimulated cells, was fragmented into segments of 180200 base pairs and presented the typical ladder pattern observed in preparations from apoptotic cells (47). In anodic reactions performed in the presence of chloride, hypochlorous acid (HOCl) is generated as a product of electrolysis (15). The latter compound can react with several amino acids to form chloramines, which are potent inducers of apoptosis (48). To evaluate if such products of electrolysis were in fact the inducers
Volume 42, 2005

Membrane Damage Is Not Reverted by Cultivation of DC-Treated Cells

After the 6-min stimulation, a period of treatment in which the most signicant alterations in cell number and viability are observed in this and others systems (16,17,20), cells were reinoculated into fresh media and cultivated for periods of 4, 18, and 24 h. The cells were then analyzed by ow cytometry, after staining with 7AAD (29,30). Control cells (Fig. 4), as well as IC- and cathodic-treated populations (not shown), remained impermeable to 7-AAD. AC-treated cells, however, preCell Biochemistry and Biophysics

67

Fig. 4. Flow cytometry of direct electric current (DC)-treated HL60 cells after staining with 7-aminoactinomycin (7-AAD). Cells were treated with DC and inoculated into fresh media. After cultivation for 4, 18, and 24 h, cells were harvested and then stained with 7-AAD, followed by ow cytometry analysis. Control cells remained impermeable to 7-AAD, as inferred by the higher concentration of cells in region 1 (R1). Anodictreated cells presented an irreversible and progressive augment in their permeability to 7-AAD, as concluded from the increased amount of cells in region 2 (R2). These events were accompanied by the generation of cell bodies with a signicantly reduced size, mainly distributed over region 3 (R3). Data from these experiments are representative of one analysis in ve identical assays.

68

Veiga et al.

Fig. 5. Transmission electron micrographs showing the morphology of control (A) and anodic-treated (B) HL60 cells. Anodic treatment induced membrane discontinuity and matrix rarefaction. Scale bars: 1 m.

Fig. 6. Anodic stimulation induces apoptosis in human cells. (A) Giemsa staining of cytocentrifuged control or anodic-treated cells. HL60-treated cells were inoculated into fresh media and Giemsa-stained after 4, 18, or 24 h of cultivation. Morphological changes were mainly characterized by cell shrinkage and nuclear chromatin clumping (inset 24 h; scale bar: 20 m). Cellular presentations suggestive of apoptotic bodies were also observed (arrows). Scale bars: 50 m. Data from these experiments are representative of three identical assays. (B) Electrophoretic profile of DNA extracted from control (a), cathodic (CC)-treated (b), intermediary (IC)-treated (c) and anodic (AC)-treated (d) cells. After treatment with DC, cells were cultivated for 4, 18, and 24 h, followed by DNA extraction and analysis in agarose gels. DNA from anodic-stimulated cells (d) presented a typical apoptosis pattern. These data are representative of three identical assays producing similar profiles.

Cell Biochemistry and Biophysics

Volume 42, 2005

Generation of Toxic Compounds Inducing Cell Death

69

gressive augment in trypan blue permeability, indicating loss of viability (Fig. 8A). Exposure of HL60 cells to the same conditions, except for the presence of DC, did not affect viability. HL60 cells were also evaluated microscopically for the presence of signs of apoptosis, which were in fact observed. Cell shrinkage and nuclear chromatin clumping, as well as cellular presentations suggestive of apoptotic bodies, were detected (Fig. 8B).

DISCUSSION
DC can differently inuence the growth of many tumoral cell types (4953). In the present work, we describe an inhibitory effect of DC-derived compounds on the proliferation of human leukemic cells. Necrosis and apoptosis were the biologic events associated with cell death in our experimental conditions. In addition, we demonstrate that the mechanism of cell death is dependent on the products of electrolysis and the polarity of the electrode, showing for the rst time that programmed cell death is induced by toxic compounds produced during anodic stimulation. Electrode polarity and the products generated during anodic or cathodic stimulations can be hallmarked as an important issue influencing tumor inhibition (7,8,12,37,51). Morris and coworkers (52) demonstrated that treatment of tumor in mice yields different results depending on whether AC or CC stimulation is used. In their model, cell death in the region of contact with the anodic electrode is not accompanied by cell lysis, whereas cathodic-stimulated cells are disrupted in a great extent. In addition, anodic stimulation has been shown to be more effective than the cathodic treatment in the clinical use of DC for the treatment of cancer (7,8,24,53). Our studies on the in vitro stimulation of tumor cells with DC confirm that different mechanisms are involved in the growth inhibition caused by DC (1517). This and previous studies demonstrate that tumor cells can have their growth ability impaired even when their morphologies are unaltered, which is a typical feature of IC-treated cells. The mechanisms causing tumor inhibition in the absence of electrodes are still undefined, but free-floating electrons could inactivate ribonucleotide reductase (54), which is responsible by the conversion of building blocks of RNA into DNA during cell division. However, evident cellular alterations are observed in cathodic- and anodic-stimulated cells (1317,37), which confirms that different mechanisms should be involved in DC-induced tumor inhibition and, additionally, that they are influenced by the products generated by electrodes with different polarity. In the present work, an extensive index of cell lysis was observed when the cells were treated in the cathodic
Volume 42, 2005

Fig. 7. Death of HL60 cells after anodic stimulation is induced by the products of electrolysis. The percentage of nonviable cells was determined after cultivation of direct electric current-stimulated cells for 24 h in the same medium of electric treatment (A) or in fresh medium (B). Alternatively, the cells were removed by centrifugation and the medium of treatment was used to cultivate nonstimulated cells (C). The viability of nonstimulated cells cultivated in fresh medium for 24 h is shown in (D).

of cell death after anodic stimulation, the percentage of nonviable cells was determined by the trypan blue method after cultivation of DC-treated cells in different conditions (Fig. 7). When the cells were cultivated for 24 h in the same medium in which they were electrically stimulated, a very low index of viability was observed. In contrast, a lesser number of nonviable cells was observed when they were treated with DC and further cultivated in a fresh medium. In addition, cultivation of nonstimulated cells in a medium in which HL60 cells have been previously stimulated resulted in a large population of nonviable cells, suggesting that toxic molecules generated during electric treatment were the active agents generating cell death. The inuence of chlorine-derived compounds in the generation of apoptotic cell death in our system was evaluated by the anodic treatment of HL-60 cells in PBS supplemented with glutamine, followed by incubation in culture medium for 1 and 3 h. During DC treatments, no signicant variations in electric conductivity were observed, as determined in a conductivity meter (data not shown). In control systems, which were treated in phosphate buffer (identical to PBS except for the presence of sodium chloride), no cellular alterations were observed (not shown). In this system, electric conductivity also remained constant during DC treatment in the different conditions. However, leukemic cells that were treated in PBS plus glutamine presented a proCell Biochemistry and Biophysics

70

Veiga et al.

Fig. 8. Stimulation of HL60 cells in the presence of glutamine. (A) Human cells were suspended in phosphate-buffered saline supplemented with 0.5 mM (squares), 1.0 (circles), or 1.5 (triangles) glutamine and treated with direct electric current (DC) (open symbols). Cell viability was measured immediately after electric stimulation (0 h) or after incubation in fresh medium supplemented with fetal bovine serum for 1 or 3 h. The viability of HL60 cells incubated in the presence of glutamine is also shown (closed symbols). (B) Nonstimulated or DC-treated cells that were incubated in the presence of glutamine (0.5 mM) for 3 h were also analyzed microscopically after Giemsa staining. Control cells presented their typical morphology, whereas DC-stimulated cells resembled apoptotic bodies. Scale bars: 10 m.

chamber, which was accompanied by a progressive increase of pH in the medium used for electric treatment. Lysis of cathodic-treated cells in PBS may be explained by a direct attack of the oxidants generated during DC treatment on the cell membrane, as previously described (13,55). Superoxide, hydroxyl groups, molecular hydrogen, and oxygen are major products of cathodic electrochemical reactions occurring in aqueous solution (14,55). The production of hydroxyl groups in these reactions is probably a major contributing factor in the pH increase, which, in turn, causes cell lysis (37). Superoxide radicals can also inuence the occurrence of cell lysis, because they are generated by cathodic reactions (55) and are efcient agents in the destruction of target cells (56,57).

Electrolysis products actually seem to be involved in cell death in our experimental conditions, because their increased concentrations resulted in higher levels of cell death (Table 1). Microscopic analysis of Giemsa-stained cells revealed that, besides the usual morphologic features of DC-inhibited cells, an alternative mechanism of death was detected. Although leukemic cells treated in the presence of the cathode or in the absence of electrodes presented morphologies that were similar to control systems, several cells stimulated in the presence of the anode resembled apoptotic bodies. This observation was conrmed by DNA electrophoresis and indicated that apoptosis is in fact an alternative pathway of cell death in leukemic cells exposed to DC. The induction of

Cell Biochemistry and Biophysics

Volume 42, 2005

Generation of Toxic Compounds Inducing Cell Death

71

apoptosis by electric and magnetic fields has been described in other systems (21,5860), although this is the rst report demonstrating that toxic compounds whose generation is dependent on electrode polarity regulate the induction of this biologic process. Using DC as the source of electric stimulation, Kurokawa and coworkers (21) described that treatment of human leukemic cell lines had undergone apoptotic cell death, but, in their model, cells were simultaneously treated with anodic and cathodic polarities, which makes an analysis of the products of electrolysis and their inuence on the generation of cellular damage difcult. The occurrence of apoptosis in cells exposed to DC was apparently derived from electrolysis products from anodic reactions. Cisplatinum is a well-known DNAdamaging agent and its presence in a DC-stimulated medium is expected around the anodic region (61), but not in the cathode compartment. Therefore, it is possible that the generation of cisplatinum complexes contributes to the induction of apoptosis in the anodic chamber. However, experiments using steel electrodes instead of platinum probes demonstrated that apoptosis still occurs, which indicates that cisplatinum complexes are not the major inducers of cell death in our system (Holandino et al., unpublished data). We therefore investigated the inuence of chlorine species produced during anodic stimulations possibly involved in the induction of apoptosis. HOCl, which is generated during electric stimulation of aqueous solutions (1315), is known to modify membrane lipids (6264) as well as protein and DNA. Its reaction with extracellular amino acids results in the generation of chloramines, which are less potent oxidants (6567). The ability of these reactive oxygen species in the generation of apoptosis has been recently described (48,68). Most studies of HOCl-induced cytotoxicity have been carried out on cells incubated in buffer solutions instead of cell culture media (6769). Although the mechanism of cell death was not specifically determined in these studies, results were generally reective of necrosis. However, additional studies revealed that treatment of endothelial cells with HOCl for 15 min in buffered saline followed by transfer of cells to complete media resulted in the occurrence of apoptotic cell death in approx 20% of the population (70). Necrosis was the major mechanism involved in cell death in the presence of higher levels of HOCl, which is consistent with our results. In our model, leukemic cells were initially treated in a buffered system, which means a low availability of amino acids and a consequent low concentration of chloramines. If this hypothesis is true, necrosis and, in lower levels, apoptosis should occur after treatment of cells with DC, which is indeed observed in our

experiments. However, treatment of HL-60 cells in PBS supplemented with glutamine resulted in a predominant occurrence of apoptosis, in accordance with the generation of chloramines derived from the reaction between HOCl and glutamine. When chloride ions were removed from the medium of treatment, no cellular damage occurred (data not shown). This result could be explained by the facilitated generation of chloramines from the reaction between the amino acid and HOCl, creating an environment more favorable to apoptotic than necrotic cell death. In this context, the critical task will be to identify the specic molecular targets that initiate the apoptotic cascade. The influence of DC and its soluble products in the biology of animal cells and microorganisms is a largely recognized phenomenon (1,3,7,12,24,25,45). The observation of different cellular DC-induced effects contributes with the comprehension of several physiologic events, and also raises the possibility of the use of electric stimulation in the treatment of cancer (3335,37,4143,52,53,71), wound healing (4,72,73), and fracture repair (74). However, the biologic mechanisms whereby DC and their derived compounds inhibit or stimulate cell growth still require clarification. In this context, the elucidation of the cellular events triggered after electric stimulation under controlled conditions (e.g., electrical field strength, period of treatment, electrode polarity, and toxic compounds generated during stimulation) will support the use of in vivo or in vitro models applying DC as a cell growthinterfering agent.

ACKNOWLEDGMENTS
We thank Dr. Geraldo A. G. Cidade for helping us in the physical determinations after DC treatment in the acrylic chamber and Dr. Damijan Miklavcic for helpful discussions. This work was supported by Financiadora de Estudos e Projetos (FINEP), Conselho Nacional de Desenvolvimento Cientfico e Tecnolgico (CNPq), ~ de Amparo a Pesquisa no Estado do Rio de Funda ao ~ Janeiro (FAPERJ), Funda ao Jos Bonifcio (FUJB), and Programa de Apoio a Ncleos de Excelncia (PRONEX).

REFERENCES
1. Bolton, L., Foleno, B., Means, B., and Petrucelli, S. (1980) Direct-current bactericidal effect on intact skin. Antimicrob. Agents Chemother. 18, 137141. 2. Chou, C. K., Mc Dougall, J. A, Ahn, C., and Vora, N. (1997) Electrochemical treatment of mouse and rat brosarcomas with direct current. Bioelectromagnetics 18, 1424. 3. Chu, C. S., McManus, A. T., Pruitt, B. A. Jr., and Mason, A. D. (1988) Therapeutic effects of silver nylon dressings with

Cell Biochemistry and Biophysics

Volume 42, 2005

72 weak direct current on Pseudomonas aeruginosa-infected burn wounds. J. Trauma 28, 14881492. Lee, C. R., Canaday, D. J., and Doong, H. (1993) A review of the biophysical basis for the clinical application of electric elds in soft-tissue repair. J. Burn Care Rehabil. 14, 319335. Lyte, M., Gannon, J. E., and OClock, G. D. Jr. (1991) Effects of in vivo electrical stimulation on enhancement and suppression of malignant lymphoma cell proliferation. J. Natl. Cancer Inst. 83, 116119. Miklavcic, D., Jarm, T., Cemazar, M., Sersa, G., An, D. J., Belehradek, J. Jr., et al. (1997) Tumor treatment by direct electric current. Tumor perfusion changes. Bioelectrochem. Bioenergetics 43, 253256. David, S. L., Absolom, D. R., Smith, C. R., Gams, J., and Herbert, M. A. (1985) Effect of low level direct current on in vivo tumor growth in hamsters. Cancer Res. 45, 56255631. Grifn, D. T., Dodd, N. J., Moore, J. V., Pullan, B. R., and Taylor, T. V. (1994) The effects of low-level direct-current therapy on a preclinical mammary carcinoma: tumor regression and systemic biochemical sequel. Br. J. Cancer 69, 875878. Xin, Y., Xue, F., and Zhao, F. (1997). Effectiveness of electrochemical therapy in the treatment of lung cancers of middle and late stage. Chin. Med. J. 110, 379383. Xin, Y. L. (1993) Traditional and Western medical treatment of 211 cases of late stage lung cancer. Chung Ku Chung Hsi I Chieh Ho Tsa Chih 13, 135113. Xin, Y., Xue, F-Z., Ge, B-S., Zhao, F-R., Shi, B., and Zhang, W. (1997) Electrochemical treatment of lung cancer. Bioelectromagnetics 18, 813. Li, K., Xin, Y., Gu, Y., Xu, B., Fan, D., and Ni, B. (1997) Effects of direct current on dog liver: possible mechanisms for tumor electrochemical treatment. Bioelectromagnetics 18, 27. Samuelsson, L. and Jnsson, L. (1980) Electrolyte destruction of lung tissue. Electrochemical aspects. Acta Radiol. Diagnosis 21, 711714. Beredson, J. and Simonsson, D. (1994) Electrochemical aspects of treatment of tissue with direct current. Eur. J. Surg. Suppl. 574, 111115. Veiga, V. F. (1996) Masters thesis, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil. Veiga, V. F., Holandino, C., Rodrigues, M. L., Capella, M. A. M., Menezes, S., and Alviano, C. S. (2000) Cellular damage and altered carbohydrate expression in P815 tumor cells induced by direct electric current: an in vitro analysis. Bioelectromagnetics 21, 597607. Holandino C., Veiga, V. F., Rodrigues, M. L., Morales, M. M., Capella, M. A. M., and Alviano, C. S. (2001) Direct current decreases cell viability but not P-glycoprotein expression and function in human multidrug resistant leukaemic cell. Biolectromagnetics 22, 470478. Sauer, H. Stanelle, R., Hescheler, J., and Wartenberg, M. (2002) The DC electrical-eld-induced Ca+2 response and growth stimulation of multicellular tumor spheroids are mediated by ATP release and purinergic receptor stimulation. J. Cell Sci. 115, 32653273. Wartenberg, M., Hescheler, J., and Sauer, H. (1997) Electrical elds enhance growth of cancer spheroids by

Veiga et al. reactive oxygen species and intracellular Ca+2. Am. J. Physiol. 272, R1677R1683. Holandino, C. Q., Veiga, V. F., Capella, M. A. M., Menezes, S., and Alviano, C. S. (2000) Damage induction by direct electric current in tumoral target cells (P815). Ind. J. Exp. Biol. 38, 554558. Kurokawa, M., Sakagami, H., Kokubu, F., Noda, H., Takeda, M., and Adachi, M. (1997) Induction of apoptotic cell death by direct-current treatment in human leukemic cell lines. J. Cancer Res. Clin. Oncol. 123, 370376. Freshney, R. I. (1994) Culture of Animal Cells: A Manual of Basic Technique, Wiley-Liss, Inc., New York. Trindade, G. S., Farias, S. L. A., Rumjanek, V. M., and Capella, M. A. M. (2000) Methylene blue reverts multidrug resistance: sensitivity of multidrug resistant cells to this dye and its photodynamic action. Cancer Lett. 151, 161167. Schauble, M. K., Habal, M. B., and Gullick, H. D. (1977) Inhibition of experimental tumour growth in hamsters by small direct currents. Arch. Pathol. Lab. Med. 101, 249297. Marino, A. A., Morris, D., and Arnold, T. (1986) Electrical treatment of Lewis lung carcinoma in mice. J. Surg. Res. 41, 198201. Patterson, M. K. Jr. (1979) Measurement of growth and viability of cells in culture, in Methods in Enzymology (Jakoby, W. B., Pastan, I. H., ed.), Academic Press, New York, pp. 141152. Trindade, G. S., Capella, M. A. M., Capella, L. S., AffonsoMitidieri, O. R., and Rumjanek, V. M. (1999) Differences in sensitivity to UVC, UVB and UVA radiation of a multidrug-resistant cell line overexpressing P-glycoprotein. Photochem. Photobiol. 69, 694699. Capella, L. S., Alcantara, J. S., Moura-Neto, V., Lopes, A. G., and Capella, M. A. M. (2000) Vanadate is toxic to adherentgrowing multidrug resistant cells. Tumor Biol. 21, 5462. Schmid, I., Uittenbogaart, C. H., Keld, B., and Giorgi, J. V. (1994) A rapid method for measuring apoptosis and dualcolor immunouorescence by single laser ow cytometry. J. Immunol. Methods 170, 145157. Lecouer, H., Ledru, E., Prvos, M. C., and Gougeon, M. L. (1997) Strategies for phenotyping apoptotic peripheral human lymphocytes comparing ISNT, annexin-V and 7AAD cytofluorimetric staining methods. J. Immunol. Methods 209, 111123. Lewis, P. R. and Knight, D. P. (1991) Cytological staining methods in electron microscopy. In Staining Methods for Sectioned Material. (Glauert A. M., ed.), North-holland Publishing Company, Amsterdam, pp. 247. Duke, R. C. and Sellins, C. B. (1989) Cellular Basis of Immune Modulation. (Kaplan, J. G. ed.) New York. Yi-Hong, L., Ting-Giu, G., Xiang-Liang, Z., Jian-Zhe, Z., Ya-Wei, H., Shu-Ming, M., et al. (1994) Electrochemical therapy for intermediate and advance liver cancer: a report of 50 cases. Eur. J. Suppl. 574, 5153. Berry, D. P., Dennison A. R., Ward, R., and Maddern, G. J. (2000) Electrolytic ablation of colorectal liver metastases: 1-year histological patient follow-up. Dig. Surg. 17, 518519.

4.

20.

5.

21. 21

6.

22. 23. 23

7.

8.

24.

9.

25. 25 26.

10.

11.

27. 27

12.

13.

28. 28 29. 29

14. 14

15. 15 16.

30. 30

31.

17. 17

32. 33. 33

18. 18

34. 34

19. 19

Cell Biochemistry and Biophysics

Volume 42, 2005

Generation of Toxic Compounds Inducing Cell Death 35. 35 Wemyss-Holden, S. A., Hall, P. M., Robertson, G. S. M., Dennison, A. R., Vanderzon, P. S., and Maddern, G. J. (2000) A new treatment for unresectable liver tumours: long-term studies of electrolytic lesions in the pig liver. Clin. Sci. 98, 561567. 36. 36 Humphrey, C. E. and Seal, E. H. (1959) Biophysical approach toward tumor regression in mice. Science 130, 388389. 37. Turler, A., Schaefer, H., Schaefer, N., Wagner, M., Maintz, 37 D., Qiao, J. C., et al. (2000) Experimental low-level direct current therapy in liver metastases: inuence of polarity and current dose. Bioelectromagnetics 21, 395401. 38. Nordestrm, B. W. E. (1994) Electrostatic eld interference with cellular and tissue function, leading to dissolution of metastases that enhances the effect of chemotherapy. Eur. J. Surg. Suppl. 574, 93109. 39. 39 Jaroszeski, M. J., Gilbert, R. A., and Heller, R. (1997) In vivo antitumor effects of electrochemotherapy in a hepatoma model. Biochim. Biophys. Acta 1334, 1518. 40. 40 Lie-Cai, S., Cun-Yn, L., Bao-Ping, Z., Tao, W., Yun-Qing, S., and Wa-Wei, L. (1994) Electrochemical therapy (ECT) for thyroid adenoma during acupuncture anaesthesia: analysis of 46 patients. Eur. J. Surg. Suppl. 574, 7981. 41. Wang, H. L (1994) Electrochemical therapy of 74 cases of 41 liver cancer. Eur. J. Surg. Suppl. 574, 5558. 42. 42 Fosh, B. G., Finch, J. G., Anthony, A. A., Texler, M., and Maddern, G. J. (2001) Electrolytic ablation of the rat pancreas: a feasibility trial. BMC Gastroenterol. 1, 111. 43. Teague, B. D., Wemyss-Holden, S. A., Fosh, B. G., 43 Dennison A. R., and Maddern, G. J. (2002) Electrolysis and other local ablative treatments for non-resectable colorectal liver metastases. ANZ J. Surg. 72, 137141. 44. 44 Capella, M. A. M., Fonseca, M. F., and Menezes, S. (1991) Synergism between electricity and ionizing radiation. J. Photochem. Photobiol. 8, 371383. 45. 45 Holandino, C. Q., Capella, M. A. M., Angluster, J., SilvaFilho, F. C., Menezes, S., and Alviano, C. S. (1998) Cell surface alterations induced by methylene blue and direct electric current in Escherichia coli. Ind. J. Biochem. Biophys. 35, 284290. 46. 46 Schmid, I., Hausner, M. A., Cole, S. W., Uittenbogaart, C. H., Giorgi, J. V., and Jamieson, B. D. (2001) Simultaneous ow cytometric measurement of viability and lymphocyte subset proliferation. J. Immunol. Methods 247, 175186. 47. 47 McDonald, E. R. and El-Deiry, W. S. (2000) Cell cycle control as a basis for cancer drug development (review). Int. J. Oncol. 16, 871886. 48. 48 Englert, R. P. and Shacter, E. (2002) Distinct modes of cell death induced by different reactive oxygen species. J. Biol. Chem. 277, 2051820526. 49. 49 Chudomel, V., Soucek, J., Hruba, A., Jerabek, J., Schwarz, J., and Smetana, K. (1989) Positive effect of direct current on cytotoxicity of human lymphocytes. Neoplasma 36, 573582. 50. 50 Batista, U., Miklavcic, D., and Sersa, G. (1994) Low level direct current cell culture broblast model. Bioelectroch. Bioener. 35, 99101.

73 51. 51 Yen, Y., Li, J. R., Zou, B. S., Rojas, F., Yu, J., and Chou, C. K. (1999) Electrochemical treatment of human KB cells in vitro. Bioelectromagnetics 20, 3441. 52. 52 Morris, D. M., Marino, A. A., and Gonzales, E. (1992) Electrochemical modication of tumor growth in mice. J. Surg. Res. 5, 306309. 53. 53 Taylor, T. V., Engler, P., Pullan, B. R., and Holt, S. (1994) Ablation of neoplasia by direct current. Br. J. Cancer 70, 342345. 54. 54 Kulsh, J. (1997) Targeting a key enzyme in cell growth: a novel therapy for cancer. Med. Hypoth. 49, 297300. 55. Forman, H.J. and Fridovich, I. (1972) Electrolytic univalent reduction of oxygen in aqueous solution demonstrated with superoxide dismutase. Science 175, 339. 56. 56 Saran, M., Miche, C., and Bors, W. (1998) Radical functions in vivo: a critical review of current concepts and hypotheses. Z. Naturforsch. 53, 210227. 57. 57 Zhou, M. J. and Petty, H. R. (1993) Superoxide-mediated lysis of erythrocytes: the role of colloid-osmotic forces. J. Cell. Physiol. 157, 555561. 58. 58 Hisamitsu, T., Narita, K., Kasahara, T., Seto, A., Yu, Y., and Asano, K. J. (1997) Induction of apoptosis in human leukemic cells by magnetic elds. Japan J. Physol. 47, 307310. 59. 59 Tofani, S., Barone, D., Cintorino, M., Santi, M. M., Ferrar, A., Orlassino, R., et al. (2001) Static and ELF magnetic fields induce tumor growth inhibition and apoptosis. Bioelectromagnetics 22, 419428. 60. 60 Narita, K., Hankawa, K., Kasahara, T., Hisamitsu, T., and Asano, K. (1997) Induction of apoptotic cell death in human leukemic cell line, HL-60, by extremely low frequency electric magnetic elds: analysis of the possible mechanisms in vitro. In Vivo 11, 329336. 61. 61 Miklavcic, D., Sersa, G., Kryzanovski, M., Novakovi, S., Bobanovi, F., Golouh, R., et al. (1993) Tumor treatment by direct electric currenttumor temperature and pH. Bioelectrochem. Bioenerg. 30, 209220. 62. 62 van den Berg, J. J. Winterbourn, C. C., and Kuypers, F. A. (1993) Hypochlorous acid-mediated modication of cholesterol and phospholipid: analysis of reaction products by gas chromatography-mass spectrometry. J. Lip. Res. 34, 20052012. 63. 63 Heinecke, J. W., Li, W., Mueller, D. M., Bohrer, A., and Turk, J. (1994) Cholesterol chlorohydrin synthesis by the myeloperoxidase-hydrogen peroxide-chloride system: potential markers for lipoproteins oxidatively damaged by phagocytes. Biochemistry 33, 1012710136. 64. 64 Carr, A. C., van den Berg, J. J. M., and Winterbounr, C. C. (1996) Chlorination of cholesterol in cell membranes by hypochlorous acid. Arch. Biochem. Biophys. 332, 6369. 65. 65 Zgliczynski, J. M., Stelmaszynska, T., Ostrowski, W., Naskalski, J., and Sznajd, J. (1968) Myeloperoxidase of human leukaemic leucocytes. Oxidation of amino acids in the presence of hydrogen peroxide. Eur. J. Biochem. 4, 540547. 66. 66 Test, S. T., Lampert, M. B., Ossana, P. J., Thoene, J. G., and Weiss, S. J. (1989) Generation of nitrogen-chlorine oxidants by human phagocytes. J. Clin. Invest. 74, 1341134.

Cell Biochemistry and Biophysics

Volume 42, 2005

74 67. 67 Thomas, E. L., Grisham, M. B., Melton, D. F., and Jefferson, M. M. (1985) Evidence for a role of taurine in the in vitro oxidative toxicity of neutrophils toward erythrocytes. J. Biol. Chem. 260, 33213329. 68. Wagner, B. A., Britigan, B. E., Reszka, K. J., McCormick, M. 68 L., and Burns, C. P. (2002) Hydrogen peroxide-induced apoptosis of HL-60 human leukemia cells is mediated by the oxidants hypochlorous acid and chloramines. Arch. Biochem. Biophys. 401, 223234. 69. 69 Vile, G. F., Rothwell, L. A., and Kettle, A. J. (2000) Initiation of rapid, P53-dependent growth arrest in cultured human skin broblasts by reactive chlorine species. Arch. Biochem. Biophys. 377, 122128. 70. Vissers, M. C., Pullar, J. M., and Hampton, M. B. (1999) 70 Hypochlorous acid causes caspase activation and apoptosis or growth arrest in human endothelial cells. Biochem. J. 344, 443449.

Veiga et al. 71. Xin, Y. L. (1994) Organisation and spread of electrochemical therapy (ECT) in China. Eur. J. Surg. Suppl. 577, 2530. 72. Karba, R. Semrov, D. Vodovnik, L., Benko, H., and Savrin, R. (1997) DC electrical stimulation for chronic wound healing enhancement. Part 1. Clinical study and determination of electrical field distribution in the numerical wound model. Bioelectrochem. Bioenergetics 43, 265270. 73. 73 Baker, L. L., DeMuth, S. K., Chambers, R., and Villar, F. (1997) Effects of electrical stimulation on wound healing in patients with diabetic ulcers. Diabetes Care 20, 405412. 74. Brighton, C. T., Friedenberg, Z. B., Mitchell, E. I., and Booth, R. E. (1977). Treatment of non-union with constant direct current. Clin. Orthop. Relat. Res. 124, 106123.

Cell Biochemistry and Biophysics

Volume 42, 2005