- Email: [email protected]
Effects of 60 Hz magnetic field exposure on polymorphonuclear leukocyte activation
Biochimica et Biophysica Acta 1472 (1999) 359^367 www.elsevier.com/locate/bba
E¡ects of 60 Hz magnetic ¢eld exposure on polymorphonuclear leukocyte activation Raddassi Khadir, James L. Morgan, John J. Murray * Departments of Medicine and Pharmacology, Division of Allergy/Immunology, Vanderbilt University Medical Center, 843 Light Hall, Nashville, TN 37232-0111, USA Received 9 March 1999; received in revised form 27 July 1999; accepted 29 July 1999
Abstract We have investigated the effects of a sinusoidal 60 Hz magnetic field on free radical (superoxide anion) production, degranulation (L-glucuronidase and lysozyme release) and viability in human neutrophils (PMNs). Experiments were performed blindly in very controlled conditions to examine the effects of a magnetic field in resting PMNs and in PMNs stimulated with a tumor promoter: phorbol 12-myristate 13-acetate (PMA). Exposure of unstimulated human PMNs to a 60 Hz magnetic field did not affect the functions examined. In contrast, exposure of PMNs to a 22 milliTesla (mT), 60 Hz magnetic field induced significant increases in superoxide anion (O3 2 ) production (26.5%) and in L-glucuronidase release (53%) when the cells were incubated with a suboptimal stimulating dose of PMA. Release of lysozyme and lactate dehydrogenase was unchanged by the magnetic field, whether the cells were stimulated or not. A 60 Hz magnetic field did not have any effect on O3 2 generation by a cell-free system xanthine/xanthine oxidase, suggesting that a magnetic field could upregulate common cellular events (signal transduction) leading to O3 2 generation and L-glucuronidase release. In conclusion, exposure of PMNs to a 22 mT, 60 Hz magnetic field potentiates the effect of PMA on O3 2 generation and Lglucuronidase release. This effect could be the result of an alteration in the intracellular signaling. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: Magnetic ¢eld; Superoxide; Polymorphonuclear leukocyte; Degranulation; Neutrophil activation
1. Introduction During the last two decades, there has been an increasing concern about the possible health hazards subsequent to electric and magnetic ¢elds (EMF) exposure. The use of electricity is so widespread that it is impossible to avoid exposure to EMF produced by the transmission and distribution of electric power (power lines) or those devices used in homes and
* Corresponding author. Fax: +1 (615) 936-5828; E-mail: [email protected]
work places (electric appliances, cellular telephones, medical imaging, computers, etc.). Questions about the possible adverse health e¡ects due to EMF exposure were ¢rst raised by Wertheimer and Leeper [1]. Further epidemiological studies investigating the possible link of EMF exposure to certain cancers like childhood leukemia have reported either a correlation [2,3] or none [4]. Besides the health issue link, EMF a¡ects the economy, since elimination of EMF exposure is impossible and alternative solutions to minimize the exposure turned out to be very costly and could exceed one billion dollars/year [5]. In order to explain the epidemiological observations associ-
0304-4165 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 1 6 5 ( 9 9 ) 0 0 1 4 2 - 7
BBAGEN 24891 1-10-99
360
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
ated with EMF exposure, experiments have been conducted in multiple laboratories to examine alterations of biological functions by EMF at the cellular and molecular levels. Cellular studies have described a variety of EMF e¡ects on biological and biochemical responses such as cell proliferation [6,7], cell surface properties [8], gene expression [9^11], DNA damage [12], apoptosis induction [13], secretion [14], ion transport [15,16] and free radical generation [17,18]. Most of these studies used 50 or 60 Hz sine wave magnetic ¢elds. However, in some others, a static [14] or pulsed [9] magnetic ¢eld induced the biological e¡ects. EMF e¡ects have been observed in a variety of cell types including nerve and immune cells. The modi¢cation of immune cell activity by EMF has a particular biological signi¢cance. The immune system is essential in protection against invasion by pathogens and against tumor development. It is also involved in in£ammatory processes. Previous reports have shown e¡ects of EMF on thymocytes [16], Blymphocytes [19], T-cell lines [20] and polymorphonuclear leukocytes (PMNs) [17]. The mechanisms by which EMF modi¢es cell functions are yet to be clearly determined. However, several reports suggested that the cell membrane is one of the most likely target sites of EMF-cell interaction [21]. Calcium transport across the membrane has been reported in many studies to be in£uenced by EMF exposure [22]. Calcium alteration is a strong candidate for mediating e¡ects of EMF, due to its important role in signal transduction and enzyme regulation. Another pathway consists of a direct interaction of EMF with free radicals [23] and with iron containing enzymes and molecules such as respiratory cytochromes [24,25]. The purpose of the present work was to establish the e¡ects of a single £ux density (22 milliTesla (mT)) 60 Hz sine wave magnetic ¢eld on human PMNs. We investigated the possible alteration of two important functions in these cells, free radical generation [17] and secretion (degranulation). Human PMNs are involved in numerous immunologic functions. In addition, PMNs are excellent cells to study these parameters as their regulation and activation are dependent upon many of the biochemical processes that could be a¡ected by EMF. Moreover, the oxidative burst can be triggered in PMNs leading
to free radical formation. This is of importance, since free radicals can react with proteins, lipids and DNA resulting in cell damage and transformation. In the current study, we used phorbol 12-myristate 13-acetate (PMA) to stimulate the cells, which acts intracellularly by inducing protein kinase C translocation to the membrane and its activation. PMN activation by this agent triggers a cascade of biochemical events, leading to the production of reactive oxygen species and degranulation. In this report, we show that while a 60 Hz magnetic ¢eld had no e¡ect per se, it enhanced the PMA-induced superoxide anion (O3 2 ) generation and L-glucuronidase release by PMNs. The magnetic ¢eld had no signi¢cant e¡ect on lysozyme release, PMN viability (lactate dehydrogenase (LDH) release) or on O3 2 production in a cellfree system. 2. Materials and methods PMA, Ficoll-Hypaque and all other routine reagents were obtained from Sigma Chemical Company (St. Louis, MO, USA). Hank's balanced salt solution (HBSS) was obtained from Gibco (Grand Island, NY, USA). PMA was dissolved in DMSO and aliquots were diluted in HBSS to the indicated concentrations prior to use in cell stimulation. The ¢nal DMSO concentration was always 90.01% and had no e¡ect on cell viability or control responses. 2.1. Neutrophil suspensions Human neutrophils were isolated from heparinized venous blood of healthy adult volunteers by FicollHypaque density gradient centrifugation, as previously described [26], and suspended in HBSS, pH 7.4. The cell suspension was adjusted to 5U106 cells/ml and warmed at 37³C for 10 min before use. 2.2. Conditions of exposure to magnetic ¢eld The exposure system consisted of a water-jacketed solenoidal coil (n = 392 turns of wire in the solenoid, 16 cm diameter, 16 cm length, 2.78 6, 21.2 mH) [10,27] and allowed for exposure to EMF magnitudes from 0.1 mT to 44 mT, with a 0.3 mV/cm induced electric ¢eld. The generator consisted of a variable
BBAGEN 24891 1-10-99
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
autotransformer Powerstat type F1126. Inside this exposure system, we placed a cylindrical incubation chamber. This incubation chamber was water-jacketed so the incubation temperature could be maintained with 37³C circulating water. The advantage of this incubation system was that it allowed for performance of as many as 29 samples at one time. Samples were arranged in three concentric circles, so that each group of 12, 11 and six samples was subjected to the same EMF intensity and characteristics. By measuring the magnetic ¢eld in every tube location, it was determined that every group was exposed to an uniform ¢eld. A second incubation chamber identical to the one described above was used as a simultaneous control for non-EMF-exposed cells (sham-exposed). For each experimental condition, three tubes were placed in the ¢rst incubation chamber (magnetic ¢eld-exposed), while simultaneously, the same number of tubes was placed in the second incubation chamber (sham-exposed). The ambient DC geomagnetic ¢eld and the ambient 60 Hz magnetic ¢eld were 50 and 1 WT, respectively. 2.3. Oxidative metabolism determination by chemiluminescence The oxidative metabolism of PMNs was measured by the luminol-enhanced chemiluminescence method as previously described [28]. Brie£y, PMNs (105 ) were sham-exposed or magnetic ¢eld-exposed at 37³C. After 10 min, the cells were removed from the incubation chambers and mixed in glass vials with 1 WM luminol and bu¡er or PMA. Chemiluminescence emitted was monitored in the dark at room temperature (for 15 min) with a Beckman LS6000SE scintillation spectrometer in the 14 C-3 H window. 2.4. Superoxide anion production O3 2 was measured spectrophotometrically by the superoxide dismutase (SOD)-inhibitable reduction of ferricytochrome c [29]. A suspension of 2.5U106 neutrophils containing ferricytochrome c was incubated at 37³C with PMA in the presence or absence of magnetic ¢eld. The reaction was halted at speci¢ed times by addition of 20 Wg/ml SOD. The extent of ferricytochrome c reduction in each supernatant was determined by the change in absorbance at 550 nm in
361
reference to controls in which SOD was added before any stimulus or solvent. O3 2 production was quanti¢ed using the extinction coe¤cient of 21.1 mM31 cm31 of ferricytochrome c and expressed as the increase in O3 2 over resting (bu¡er-treated) cells. The ¢nal concentration of ferricytochrome c was 80 WM. Generation of O3 2 by a cell-free system has been assayed under the same conditions as described above except that PMNs were replaced by 0.05 U/ ml xanthine oxidase and 0.5 mM xanthine. 2.5. Degranulation assay Degranulation was measured as the release of Lglucuronidase and lysozyme. PMNs (5U106 ) were incubated at 37³C with PMA in the presence or absence of a 60 Hz magnetic ¢eld. After 10 or 60 min, the cells were rapidly centrifuged for 20 s at 12 000 rpm and supernatant or cell lysate were assayed for the L-glucuronidase and lysozyme content. Lysozyme hydrolyzes the mucopeptide cell wall structure of Micrococcus lysodeikticus. Lysozyme release was measured as described previously [14], except that we miniaturized the assay which was performed in 96 well plates to get faster and more sensitive results. Aliquots (50 Wl) of supernatant were added to 200 Wl of M. lysodeikticus (0.25 mg/ ml). The absorbance decrease, which re£ects the lysis of M. lysodeikticus at 25³C, was monitored at 450 nm. Lysozyme release by PMNs was quanti¢ed by comparison to a standard solution of egg white lysozyme. The results were expressed as the percentage of total lysozyme released after PMN sonication. For the L-glucuronidase release assay, aliquots (15 Wl) of supernatant or cell lysates were incubated for 12 h with phenolphtalein glucuronic acid at 37³C. The product released (phenolphtalein) by L-glucuronidase present in the sample was measured by its absorbance at 540 nm and compared to a standard solution of phenolphthalein. 2.6. Apoptosis assessment DNA was extracted from 5U106 cells using DNAzol reagent (MRC, Cincinnati, OH, USA). After digestion with RNAse A (0.5 mg/ml) for 1 h at 50³C, DNA was washed twice and samples were applied to a 2% agarose gel containing 1 Wg/ml ethidium bro-
BBAGEN 24891 1-10-99
362
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
mide for electrophoresis. A DNA ladder l DNA Eco digest was used as a DNA fragmentation marker. LDH was measured using a microassay kit (Sigma). 2.7. Statistical analysis Before starting the experiments, tubes containing the samples were coded by assigning numbers to each one. At the end of each experiment, data were collected and means for each triplicate calculated. The codes were then broken to identify sham- and magnetic ¢eld-exposed samples. Data are expressed as the mean þ S.D. of the average of the values obtained in di¡erent experiments. A two-tailed paired Student's t-test was utilized to compare results of sham-exposed and magnetic ¢eld-exposed samples obtained in n independent experiments. P 6 0.05 was considered to be signi¢cant. 3. Results 3.1. E¡ect of a 60 Hz magnetic ¢eld on PMN oxidative burst Data published previously suggested an e¡ect of EMF on free radical generation [17,30]. One of the most important functions of PMNs is the generation of reactive oxygen species (H2 O2 , O3 2 ) through the oxidative burst. The e¡ect of a 60 Hz magnetic ¢eld on H2 O2 formation was ¢rst assessed by measuring chemiluminescence, since this assay requires relatively few cells and gives continuous information about the initial steps and the kinetics of PMN oxidative metabolism. PMNs were sham-exposed or magnetic ¢eld-exposed, then, they were removed from the incubation chambers and chemiluminescence was immediately monitored in the presence or absence of PMA. Unstimulated PMNs produced a very small amount of chemiluminescence (14 000 cpm). Upon stimulation with 15 nM PMA, a sustained increase was observed re£ecting massive H2 O2 production (Fig. 1). Fig. 1 represents the e¡ect of 10 min exposure to the magnetic ¢eld on chemiluminescence generated by PMNs. There was no signi¢cant di¡erence in chemiluminescence between sham- and magnetic ¢eld-exposed groups either in unstimulated or stimulated cells. Di¡erent times of
Fig. 1. E¡ect of a 60 Hz magnetic ¢eld on chemiluminescence generation by human PMNs. PMNs (105 /ml) were incubated for 10 min at 37³C in the presence (closed symbols, solid lines) or absence (open symbols, broken lines) of a 22 mT magnetic ¢eld. The suspension (1 ml) was mixed with luminol, followed by addition of bu¡er (control) or 15 nM PMA. Chemiluminescence was recorded for 15 min and expressed as 106 cpm/105 PMNs. The ¢gure is a typical recording of one experiment representative of three similar others.
exposure (1, 5, 10, 15 and 30 min) were tested, but the only di¡erences we observed were in the range of experimental £uctuations. The previous method of chemiluminescence presented a technical limitation. Stimulation and exposure could be performed only sequentially and demonstrated that a 60 Hz magnetic ¢eld did not alter H2 O2 generation by PMNs. We used another experimental procedure consisting of measuring directly the production of O3 2 during magnetic ¢eld exposure. 3.2. E¡ect of a 60 Hz magnetic ¢eld on superoxide production Superoxide (O3 2 ) is the initial one electron reduction product of oxygen in the respiratory burst of PMNs. Its generation in stimulated PMNs occurs after NADPH oxidase translocates to the membrane and catalyzes the electron transfer from the donor NADPH to the acceptor O2 . The possible alteration of this electron transfer by EMF could be re£ected by a change in the amount of O3 2 generated. We examined the e¡ect of a sinusoidal 60 Hz, 22 mT magnetic ¢eld on O3 2 production. In resting PMNs,
BBAGEN 24891 1-10-99
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
Fig. 2. E¡ect of a 60 Hz magnetic ¢eld on superoxide production by PMA-stimulated PMNs. PMNs (2.5U106 /ml) were stimulated with 15 nM PMA and exposed or not to a 22 mT magnetic ¢eld. O3 2 production was measured after di¡erent periods of exposure. Data represent the mean þ S.D. of seven experiments and are expressed as the % of the corresponding non-exposed sample values (as a ratio magnetic ¢eld/shamU 100%). Italicized numbers are P values for magnetic ¢eld versus control (sham). Control values were 2 þ 0.4, 2.8 þ 0.3, 5.2 þ 0.8, 11.2 þ 1.2, 24.4 þ 1.7, 31 þ 4.2 nmol O3 2 /0.5, 1, 3, 5, 8, 10 min, respectively.
the production of superoxide was less than 2 nmol/ 2.5U106 cells (Fig. 2) and remained unchanged even after 60 min exposure to the magnetic ¢eld (1.8 þ 0.7 versus 1.7 þ 0.8, for sham versus magnetic ¢eld, respectively, n = 10, P = 0.56). Stimulation of PMNs with PMA (15 nM) induced the production of large amounts of O3 2 . 15 nM PMA is a suboptimal dose that produces 60^80% of the maximal oxidative response in PMNs. To examine the possible alteration of this electron transfer by a magnetic ¢eld, cells were simultaneously sham- or magnetic ¢eld-exposed with or without PMA. At di¡erent times of exposure/ stimulation, the reaction was stopped and superoxide generation was measured. PMA-stimulated PMNs showed a signi¢cant increase in O3 generation 2 when exposed to a magnetic ¢eld (Fig. 2). The e¡ect of a 60 Hz magnetic ¢eld was more important during the ¢rst minutes of O3 2 production, becoming statistically insigni¢cant after 10 min of exposure. As an additional control experiment, cells were incubated in the presence of PMA and placed randomly in either of the two incubation chambers without any
363
applied magnetic ¢eld (sham versus sham). PMAstimulated PMNs incubated in one or the other incubation chambers did not show any di¡erences (Fig. 3), indicating that the incubation chambers were identical. Since studies have reported e¡ects of EMF on radical generation [25] and on iron containing molecules [24], e.g. cytochrome. We examined the hypothesis that a magnetic ¢eld could interfere with the method we employed to measure O3 2 production. As described in Section 2, this assay is based on the reduction of ferricytochrome c to the ferrous form produced by O3 2 . We used the same conditions of incubation except that PMNs were replaced by a cell-free system consisting of the oxidation of xanthine by xanthine oxidase to form O3 2 and urate. Exposure of this cell-free system generating superoxide to a 22 mT, 60 Hz sine wave magnetic ¢eld was not a¡ected (Fig. 4) and there was no di¡erence between sham- and magnetic ¢eld-exposed groups. These results suggested that the target of the observed 60 Hz magnetic ¢eld e¡ect could be located
Fig. 3. Comparison of superoxide anion generation in the absence of a magnetic ¢eld in the two exposure chambers. PMNs (2.5U106 /ml) were split into two identical groups. Both groups were stimulated with 15 nM PMA and placed in one of the two incubation chambers. The reaction was stopped at the indicated times and superoxide production measured. Results are the means þ S.D. of three independent experiments. No di¡erences were noted in cells incubated in either chamber, indicating that the two chambers were identical.
BBAGEN 24891 1-10-99
364
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
suboptimal dose of PMA (Fig. 5). We examined the e¡ect of a 60 Hz magnetic ¢eld on lysozyme release under the same conditions as for L-glucuronidase. In our study, exposure of PMNs up to a 22 mT magnetic ¢eld had no e¡ect on either the basal or the stimulated release of lysozyme (Fig. 6). Furthermore, exposure of unstimulated cells to a 40 mT magnetic ¢eld for various periods of time up to 60 min did not a¡ect lysozyme release (data not shown). 3.4. E¡ect of a 60 Hz magnetic ¢eld on cell viability
Fig. 4. E¡ect of a 60 Hz magnetic ¢eld on superoxide anion generation in a cell-free system. A mixture of xanthine (0.5 mM) and ferricytochrome c (80 WM) was split into tubes and placed in either chamber. The reaction was started by adding xanthine oxidase (0.05 U/ml) to each tube and exposing the chambers to a 22 mT magnetic ¢eld (solid line) or none (broken line). The reaction was stopped at di¡erent times and superoxide production measured. Results are the means þ S.D. of four independent experiments.
We examined the e¡ect of 22 mT magnetic ¢eld exposure on viability of PMNs assessed by LDH release. Fig. 7 shows that as for lysozyme, LDH release was not a¡ected by exposure to a 22 mT, 60 Hz magnetic ¢eld, either in resting or PMA-stimulated PMNs. Furthermore, analysis of DNA fragmentation on an agarose gel con¢rmed that the cells were not undergoing apoptosis upon exposure to the
intracellularly upstream from the enzymatic reaction generating O3 2 . If this were true, we hypothesized that on other PMN functions such as degranulation could be similarly a¡ected by a magnetic ¢eld. 3.3. E¡ect of a 22 mT, 60 Hz magnetic ¢eld on degranulation Degranulation is a microbicidal and in£ammatory function of PMNs that has been shown to be associated with ion £uxes [31]. EMF exposure could affect this PMN function by interfering with ion movements. We tested this hypothesis by measuring Lglucuronidase and lysozyme. Lysozyme is a constituent of both primary (azurophile) and secondary (speci¢c) granules, while L-glucuronidase is localized in azurophile granules and smaller storage organelles. Our results showed that the baseline level of L-glucuronidase release was increased by 19% upon exposure to a 22 mT magnetic ¢eld, a more signi¢cant increase (53%) was observed when PMNs were simultaneously exposed to the magnetic ¢eld and a
Fig. 5. Alteration of L-glucuronidase release from PMNs during exposure to a 60 Hz magnetic ¢eld. PMNs (5U106 /ml) were incubated for 60 min with bu¡er or 15 nM PMA, in the presence or absence of a 22 mT magnetic ¢eld. 15 nM PMA is a suboptimal concentration providing 65% of the optimal response. LGlucuronidase release was quanti¢ed in supernatants by incubating with phenolphthalein glucuronic acid for 12 h and measuring the optical density of the product phenolphthalein. Quantities of L-glucuronidase released are expressed as the percentage of total L-glucuronidase in PMNs. Data are the mean þ S.D. of three separate experiments. *P 6 0.05 compared to sham. 2P 6 0.05 compared to control (without PMA).
BBAGEN 24891 1-10-99
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
Fig. 6. Lysozyme release from PMNs during exposure to a 60 Hz magnetic ¢eld. PMNs (5U106 /ml) were incubated for 10 or 60 min with bu¡er or 15 nM PMA, in the presence (hatched bars) or absence of a 22 mT magnetic ¢eld. Lysozyme release was quanti¢ed by measuring the optical density of a suspension of M. lysodeikticus. Quantities of lysozyme released are expressed as the percentage of total lysozyme in PMNs. Data are the mean þ S.D. of six separate experiments. *P 6 0.05 compared to control (without PMA).
magnetic ¢eld for 60 min (data not shown). Incubation of PMNs with TNFK (200 U/ml) alone used as a positive control showed that a signi¢cant apoptosis induction can be observed after 60 min of treatment. Moreover, it has been shown in a recent report that exposure of PMNs to a 50 Hz magnetic ¢eld for up to 3.5 h did not induce apoptosis [13]. 4. Discussion Several studies have shown e¡ects of EMF on calcium £ux [32] and free radical generation [17,34]. The e¡ect of EMF has been observed in many immune cell types [33]: thymocytes [16], B-lymphocytes [19], T-lymphocytes [20] and peripheral blood monocytes [34]. Only one study has been done previously in human PMNs, using a strong static magnetic ¢eld [14], and another in rat peritoneal neutrophils, using a 60 Hz magnetic ¢eld [17]. Stimulation of PMNs induces a membrane potential change and transloca-
365
Fig. 7. LDH release from PMNs during exposure to a 60 Hz magnetic ¢eld. PMNs (5U106 /ml) were incubated for 10 or 60 min with bu¡er or 15 nM PMA, in the presence (hatched bars) or absence of a 22 mT magnetic ¢eld. LDH release was quanti¢ed using a LDH detection kit (Sigma). Data are the mean þ S.D. of ¢ve separate experiments.
tion of enzymes to the cell membrane. These events could be a¡ected by EMF exposure. Furthermore, calcium plays an important role in signal transduction and in the control of PMN functions such as enzyme secretion and production of free radicals. The production of free radicals, which is the primary function of PMNs, consists of electron transfer from NADPH to molecular oxygen, leading to the formation of reactive oxygen species. In this study, we assessed more speci¢cally the impact of a high £ux density (22 mT), 60 Hz sinusoidal magnetic ¢eld exposure on O3 2 production, degranulation (L-glucuronidase and lysozyme release) and cell viability (LDH release, apoptosis induction). The experimental conditions we used were similar to those described previously [10]. The exposure system described in Section 2 is exactly the same as used in these previous experiments. However, the present experiments were performed to provide maximum controls and to eliminate potential bias. The modi¢cations we introduced included the following: (1) performing experiments in two identical incubation chambers, one used as control (sham), while the other one is used to expose the cells to a 60 Hz magnetic
BBAGEN 24891 1-10-99
366
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
¢eld (this procedure allows for simultaneous treatment of the cells and eliminates any possible e¡ect due to di¡ering pre-incubation conditions), (2) sham exposure and magnetic ¢eld exposure were performed simultaneously, (3) every treatment was performed in triplicate to eliminate variations due to di¡erences between samples both for the sham-exposed group and for the magnetic ¢eld-exposed one, (4) exposures and data analyses were done blindly. We ¢rst examined the possible alteration of oxidative burst in human PMNs by exposure to a 22 mT, 60 Hz sinusoidal magnetic ¢eld. Many publications suggested an e¡ect of EMF on free radical formation [23,35]. Moreover, the oxidative burst process is dependent on calcium, it occurs after translocation of components of the NADPH-oxidase to the plasma membrane and involves iron containing enzymes. All these events could be altered by electromagnetic ¢elds. We utilized the SOD-inhibitable reduction of ferricytochrome c to detect speci¢cally the major initial product of the oxidative burst (O3 2 ). Under the described experimental conditions, we found a significant e¡ect of the 60 Hz magnetic ¢eld on O3 2 production. The maximal e¡ect has been observed at 3 and 5 min after PMA stimulation and EMF exposure (18^35% increase at 5 min). These results are consistent with the previous reported e¡ect of a magnetic ¢eld on rat neutrophils [17], despite using a di¡erent experimental system. In fact, in those experiments, the authors used rat peritoneal neutrophils elicited by casein, which are di¡erent from the human PMNs freshly isolated from the circulating blood that we examined. Also, a di¡erent method to measure free radical production was utilized [17]. We next examined the possibility that a magnetic ¢eld could alter the method of superoxide measurement or the chemical formation of O3 2 by utilizing a . We did not ¢nd any cell-free system generating O3 2 e¡ect of the 60 Hz magnetic ¢eld on superoxide generation by a system xanthine/xanthine oxidase, suggesting that the observed e¡ect may occur upstream of the O3 2 generation step. If this was true, we should be able to detect the same e¡ect on other PMN functions such as degranulation. We examined the e¡ect of a 60 Hz sine wave magnetic ¢eld on degranulation (lysozyme and L-glucuronidase release) and on viability (LDH release). Ex-
posure of PMNs to intensities up to a 40 mT magnetic ¢eld with or without PMA stimulation did not induce any di¡erences between sham- and magnetic ¢eld-exposed cells regarding lysozyme and LDH release. In contrast, our results showed that exposure to a 22 mT, 60 Hz sinusoidal magnetic ¢eld induced a signi¢cant increase of L-glucuronidase release, which was bigger in PMA-stimulated PMNs. The explanation for this could be attributed to the subcellular distribution of L-glucuronidase (azurophile granules and smaller storage organelles) and lysozyme (azurophile and speci¢c granules). This would indicate that the e¡ect of EMF on degranulation is speci¢c and could a¡ect the products localized in smaller storage organelles. Such a mechanism is not impossible and has been reported in human neutrophils stimulated with phosphatidic acid which induced the release of L-glucuronidase but not lactoferrin. PMA induced the same response. However, at a high intracellular calcium concentration, the release of both enzymes was induced [36]. Together, these results demonstrate that a 60 Hz magnetic ¢eld affects PMN functions. The mechanism of these changes may occur through e¡ects on intracellular signaling such as calcium, phospholipase D and protein kinase C and remains to be elucidated. Further experiments are also required to examine the range of £ux densities producing these e¡ects. Epidemiological studies indicate that £ux densities around 0.2 WT may be threshold values for possible health e¡ects [1,2,37]. In conclusion, in the conditions we described here, we showed a signi¢cant e¡ect of a speci¢c magnetic ¢eld (22 mT, 60 Hz sine wave magnetic ¢eld) on superoxide anion production and L-glucuronidase release induced by PMA in human PMNs, while no e¡ect was detected on O3 2 generation by a cell-free system. These results suggest an interaction of the 60 Hz magnetic ¢eld with the early signal transduction mechanism of the PMN. Elucidation of these results of the magnetic ¢eld on signal coupling mechanisms may clarify the possible basis for the described alterations of biological responses by EMF. Acknowledgements This work was supported in part by a National
BBAGEN 24891 1-10-99
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
Institutes of Health Grant, GM15431. The authors would like to thank Dr Robert Liburdy and Dr John Wikswo for their expert assistance and instructive discussions. References [1] N. Wertheimer, E. Leeper, Am. J. Epidemiol. 109 (1979) 273^284. [2] D.A. Savitz, H. Wachtel, F.A. Barnes, E.M. John, J.G. Tvrdik, Am. J. Epidemiol. 128 (1988) 21^38. [3] S.J. London, D.C. Thomas, J.D. Bowman, E. Sobel, T.C. Cheng, J.M. Peters, Am. J. Epidemiol. 134 (1991) 923^937. [4] M.S. Linet, E.E. Hatch, R.A. Kleinerman, L.L. Robinson, W.T. Kaune, D.R. Friedman, R.K. Severson, C.M. Haines, C.T. Hartsock, S. Niwa, S. Wacholder, R.E. Tarone, New Engl. J. Med. 337 (1997) 1^7. [5] H.K. Florig, Science 257 (1992) 468^469. [6] A.M. Khalil, W. Qassem, Mutat. Res. 247 (1991) 141^146. [7] C.F. Blackman, J.P. Blanchard, S.G. Benane, D.E. House, J.A. Elder, Bioelectromagnetics 19 (1998) 204^209. [8] S. Paradisi, G. Donelli, M.T. Santini, E. Straface, W. Malorni, Bioelectromagnetics 14 (1993) 247^255. [9] R. Goodman, L.-X. Wei, J.-C. Xu, A.S. Henderson, Biochim. Biophys. Acta 1009 (1989) 216^220. [10] R.P. Liburdy, D.E. Callahan, J. Harland, E. Dunham, T.R. Sloma, P. Yaswen, FEBS Lett. 334 (1993) 301^308. [11] R. Karabakhtsian, N. Broude, N. Shalts, S. Kochlatyi, R. Goodman, A.S. Henderson, FEBS Lett. 349 (1994) 1^6. [12] G. D`Ambrosio, A. Scaglione, D. DiBerardino, M.B. Lioi, L. Iannuzzi, E. Mostcciuolo, M.R. Scar¢, J. Bioelectr. 4 (1985) 279^284. [13] T. Hisamitsu, K. Narita, T. Kasahara, A. Seto, Y. Yu, K. Asano, Jpn. J. Physiol. 47 (1997) 307^310. [14] F.J. Papatheofanis, Radiat. Res. 122 (1990) 24^28. [15] W.R. Adey and A. Sheppard, in: M. Blank and E. Findl (Eds.), Mechanistic Approaches to Interactions of Electromagnetic Fields with Living Systems, Plenum Press, New York, 1987, pp. 365^387.
367
[16] J. Walleczek, R. Liburdy, FEBS Lett. 271 (1990) 157^160. [17] S. Roy, Y. Noda, V. Eckert, M.G. Traber, A. Mori, R. Liburdy, L. Packer, FEBS Lett. 376 (1995) 164^166. [18] J.C. Scaiano, F.L. Cozens, N. Mohtat, Photochem. Photobiol. 62 (1995) 818^829. [19] F.M. Uckun, T. Kurosaki, J. Jin, X. Jun, A. Morgan, M. Takara, J. Bolen, R. Luben, J. Biol. Chem. 270 (1995) 27666^27670. [20] E. Lindstro«m, P. Lindstro«m, A. Berglund, K.H. Mild, E. Lungren, J. Cell. Physiol. 156 (1993) 395^398. [21] R.A. Luben, C.D. Cain, in: W.R. Adey and A.F. Lawrence (Eds.), Non-linear Electrodynamics in Biological Systems, Plenum Press, New York, 1984, pp. 23^34. [22] M.J. Azanza, A. DelMoral, Prog. Neurobiol. 44 (1994) 517^ 601. [23] F.L. Cozens, J.C. Scaiano, J. Am. Chem. Soc. 115 (1993) 5204^5211. [24] J.L. Philips, Immunol. Lett. 13 (1986) 295^299. [25] R. DelCarratore, E. Morichetti, C. DellaCroce, G. Bronzetti, Bioelectromagnetics 16 (1995) 324^329. [26] J.C. Gay, K. Raddassi, A.P. Truett III, J.J. Murray, Biochim. Biophys. Acta 1336 (1997) 243^253. [27] R.P. Liburdy, FEBS Lett. 301 (1992) 53^59. [28] B.D. Cheson, R.L. Christensen, R. Sperling, B.E. Kohler, B.M. Babior, J. Clin. Invest. 58 (1976) 789^796. [29] J.M. McCord, I. Fridovich, J. Biol. Chem. 244 (1969) 6049^ 6063. [30] B. Brocklehurst, K.A. McLauchlan, Int. J. Radiat. Biol. 69 (1996) 3^24. [31] P.H. Naccache, H.J. Showell, E.L. Becker, R.I. Sha'a¢, J. Cell Biol. 75 (1977) 635^649. [32] J. Walleczek, FASEB J. 6 (1992) 3177^3185. [33] A. Lacy-Hulbert, J.C. Metcalf, R. Hesketh, Radiat. Res. 144 (1998) 9^17. [34] C. Petrini, M.L. Dupuis, A. Polighetti, C. Ramoni, P. Vecchia, Bioenergetics 44 (1997) 121^125. [35] T.T. Harkins, C.B. Grissom, Science 263 (1994) 958^960. [36] W. Zaman, T. Mitsuyama, M. Hatakenaka, D. Kang, S. Minakami, K. Takeshige, J. Biochem. 115 (1994) 238^244. [37] S. Preston, W. Navidi, D. Thomas, P.J. Lee, J. Bowman, J. Pagoda, Am. J. Epidemiol. 143 (1996) 105^119.
BBAGEN 24891 1-10-99
E¡ects of 60 Hz magnetic ¢eld exposure on polymorphonuclear leukocyte activation Raddassi Khadir, James L. Morgan, John J. Murray * Departments of Medicine and Pharmacology, Division of Allergy/Immunology, Vanderbilt University Medical Center, 843 Light Hall, Nashville, TN 37232-0111, USA Received 9 March 1999; received in revised form 27 July 1999; accepted 29 July 1999
Abstract We have investigated the effects of a sinusoidal 60 Hz magnetic field on free radical (superoxide anion) production, degranulation (L-glucuronidase and lysozyme release) and viability in human neutrophils (PMNs). Experiments were performed blindly in very controlled conditions to examine the effects of a magnetic field in resting PMNs and in PMNs stimulated with a tumor promoter: phorbol 12-myristate 13-acetate (PMA). Exposure of unstimulated human PMNs to a 60 Hz magnetic field did not affect the functions examined. In contrast, exposure of PMNs to a 22 milliTesla (mT), 60 Hz magnetic field induced significant increases in superoxide anion (O3 2 ) production (26.5%) and in L-glucuronidase release (53%) when the cells were incubated with a suboptimal stimulating dose of PMA. Release of lysozyme and lactate dehydrogenase was unchanged by the magnetic field, whether the cells were stimulated or not. A 60 Hz magnetic field did not have any effect on O3 2 generation by a cell-free system xanthine/xanthine oxidase, suggesting that a magnetic field could upregulate common cellular events (signal transduction) leading to O3 2 generation and L-glucuronidase release. In conclusion, exposure of PMNs to a 22 mT, 60 Hz magnetic field potentiates the effect of PMA on O3 2 generation and Lglucuronidase release. This effect could be the result of an alteration in the intracellular signaling. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: Magnetic ¢eld; Superoxide; Polymorphonuclear leukocyte; Degranulation; Neutrophil activation
1. Introduction During the last two decades, there has been an increasing concern about the possible health hazards subsequent to electric and magnetic ¢elds (EMF) exposure. The use of electricity is so widespread that it is impossible to avoid exposure to EMF produced by the transmission and distribution of electric power (power lines) or those devices used in homes and
* Corresponding author. Fax: +1 (615) 936-5828; E-mail: [email protected]
work places (electric appliances, cellular telephones, medical imaging, computers, etc.). Questions about the possible adverse health e¡ects due to EMF exposure were ¢rst raised by Wertheimer and Leeper [1]. Further epidemiological studies investigating the possible link of EMF exposure to certain cancers like childhood leukemia have reported either a correlation [2,3] or none [4]. Besides the health issue link, EMF a¡ects the economy, since elimination of EMF exposure is impossible and alternative solutions to minimize the exposure turned out to be very costly and could exceed one billion dollars/year [5]. In order to explain the epidemiological observations associ-
0304-4165 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 1 6 5 ( 9 9 ) 0 0 1 4 2 - 7
BBAGEN 24891 1-10-99
360
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
ated with EMF exposure, experiments have been conducted in multiple laboratories to examine alterations of biological functions by EMF at the cellular and molecular levels. Cellular studies have described a variety of EMF e¡ects on biological and biochemical responses such as cell proliferation [6,7], cell surface properties [8], gene expression [9^11], DNA damage [12], apoptosis induction [13], secretion [14], ion transport [15,16] and free radical generation [17,18]. Most of these studies used 50 or 60 Hz sine wave magnetic ¢elds. However, in some others, a static [14] or pulsed [9] magnetic ¢eld induced the biological e¡ects. EMF e¡ects have been observed in a variety of cell types including nerve and immune cells. The modi¢cation of immune cell activity by EMF has a particular biological signi¢cance. The immune system is essential in protection against invasion by pathogens and against tumor development. It is also involved in in£ammatory processes. Previous reports have shown e¡ects of EMF on thymocytes [16], Blymphocytes [19], T-cell lines [20] and polymorphonuclear leukocytes (PMNs) [17]. The mechanisms by which EMF modi¢es cell functions are yet to be clearly determined. However, several reports suggested that the cell membrane is one of the most likely target sites of EMF-cell interaction [21]. Calcium transport across the membrane has been reported in many studies to be in£uenced by EMF exposure [22]. Calcium alteration is a strong candidate for mediating e¡ects of EMF, due to its important role in signal transduction and enzyme regulation. Another pathway consists of a direct interaction of EMF with free radicals [23] and with iron containing enzymes and molecules such as respiratory cytochromes [24,25]. The purpose of the present work was to establish the e¡ects of a single £ux density (22 milliTesla (mT)) 60 Hz sine wave magnetic ¢eld on human PMNs. We investigated the possible alteration of two important functions in these cells, free radical generation [17] and secretion (degranulation). Human PMNs are involved in numerous immunologic functions. In addition, PMNs are excellent cells to study these parameters as their regulation and activation are dependent upon many of the biochemical processes that could be a¡ected by EMF. Moreover, the oxidative burst can be triggered in PMNs leading
to free radical formation. This is of importance, since free radicals can react with proteins, lipids and DNA resulting in cell damage and transformation. In the current study, we used phorbol 12-myristate 13-acetate (PMA) to stimulate the cells, which acts intracellularly by inducing protein kinase C translocation to the membrane and its activation. PMN activation by this agent triggers a cascade of biochemical events, leading to the production of reactive oxygen species and degranulation. In this report, we show that while a 60 Hz magnetic ¢eld had no e¡ect per se, it enhanced the PMA-induced superoxide anion (O3 2 ) generation and L-glucuronidase release by PMNs. The magnetic ¢eld had no signi¢cant e¡ect on lysozyme release, PMN viability (lactate dehydrogenase (LDH) release) or on O3 2 production in a cellfree system. 2. Materials and methods PMA, Ficoll-Hypaque and all other routine reagents were obtained from Sigma Chemical Company (St. Louis, MO, USA). Hank's balanced salt solution (HBSS) was obtained from Gibco (Grand Island, NY, USA). PMA was dissolved in DMSO and aliquots were diluted in HBSS to the indicated concentrations prior to use in cell stimulation. The ¢nal DMSO concentration was always 90.01% and had no e¡ect on cell viability or control responses. 2.1. Neutrophil suspensions Human neutrophils were isolated from heparinized venous blood of healthy adult volunteers by FicollHypaque density gradient centrifugation, as previously described [26], and suspended in HBSS, pH 7.4. The cell suspension was adjusted to 5U106 cells/ml and warmed at 37³C for 10 min before use. 2.2. Conditions of exposure to magnetic ¢eld The exposure system consisted of a water-jacketed solenoidal coil (n = 392 turns of wire in the solenoid, 16 cm diameter, 16 cm length, 2.78 6, 21.2 mH) [10,27] and allowed for exposure to EMF magnitudes from 0.1 mT to 44 mT, with a 0.3 mV/cm induced electric ¢eld. The generator consisted of a variable
BBAGEN 24891 1-10-99
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
autotransformer Powerstat type F1126. Inside this exposure system, we placed a cylindrical incubation chamber. This incubation chamber was water-jacketed so the incubation temperature could be maintained with 37³C circulating water. The advantage of this incubation system was that it allowed for performance of as many as 29 samples at one time. Samples were arranged in three concentric circles, so that each group of 12, 11 and six samples was subjected to the same EMF intensity and characteristics. By measuring the magnetic ¢eld in every tube location, it was determined that every group was exposed to an uniform ¢eld. A second incubation chamber identical to the one described above was used as a simultaneous control for non-EMF-exposed cells (sham-exposed). For each experimental condition, three tubes were placed in the ¢rst incubation chamber (magnetic ¢eld-exposed), while simultaneously, the same number of tubes was placed in the second incubation chamber (sham-exposed). The ambient DC geomagnetic ¢eld and the ambient 60 Hz magnetic ¢eld were 50 and 1 WT, respectively. 2.3. Oxidative metabolism determination by chemiluminescence The oxidative metabolism of PMNs was measured by the luminol-enhanced chemiluminescence method as previously described [28]. Brie£y, PMNs (105 ) were sham-exposed or magnetic ¢eld-exposed at 37³C. After 10 min, the cells were removed from the incubation chambers and mixed in glass vials with 1 WM luminol and bu¡er or PMA. Chemiluminescence emitted was monitored in the dark at room temperature (for 15 min) with a Beckman LS6000SE scintillation spectrometer in the 14 C-3 H window. 2.4. Superoxide anion production O3 2 was measured spectrophotometrically by the superoxide dismutase (SOD)-inhibitable reduction of ferricytochrome c [29]. A suspension of 2.5U106 neutrophils containing ferricytochrome c was incubated at 37³C with PMA in the presence or absence of magnetic ¢eld. The reaction was halted at speci¢ed times by addition of 20 Wg/ml SOD. The extent of ferricytochrome c reduction in each supernatant was determined by the change in absorbance at 550 nm in
361
reference to controls in which SOD was added before any stimulus or solvent. O3 2 production was quanti¢ed using the extinction coe¤cient of 21.1 mM31 cm31 of ferricytochrome c and expressed as the increase in O3 2 over resting (bu¡er-treated) cells. The ¢nal concentration of ferricytochrome c was 80 WM. Generation of O3 2 by a cell-free system has been assayed under the same conditions as described above except that PMNs were replaced by 0.05 U/ ml xanthine oxidase and 0.5 mM xanthine. 2.5. Degranulation assay Degranulation was measured as the release of Lglucuronidase and lysozyme. PMNs (5U106 ) were incubated at 37³C with PMA in the presence or absence of a 60 Hz magnetic ¢eld. After 10 or 60 min, the cells were rapidly centrifuged for 20 s at 12 000 rpm and supernatant or cell lysate were assayed for the L-glucuronidase and lysozyme content. Lysozyme hydrolyzes the mucopeptide cell wall structure of Micrococcus lysodeikticus. Lysozyme release was measured as described previously [14], except that we miniaturized the assay which was performed in 96 well plates to get faster and more sensitive results. Aliquots (50 Wl) of supernatant were added to 200 Wl of M. lysodeikticus (0.25 mg/ ml). The absorbance decrease, which re£ects the lysis of M. lysodeikticus at 25³C, was monitored at 450 nm. Lysozyme release by PMNs was quanti¢ed by comparison to a standard solution of egg white lysozyme. The results were expressed as the percentage of total lysozyme released after PMN sonication. For the L-glucuronidase release assay, aliquots (15 Wl) of supernatant or cell lysates were incubated for 12 h with phenolphtalein glucuronic acid at 37³C. The product released (phenolphtalein) by L-glucuronidase present in the sample was measured by its absorbance at 540 nm and compared to a standard solution of phenolphthalein. 2.6. Apoptosis assessment DNA was extracted from 5U106 cells using DNAzol reagent (MRC, Cincinnati, OH, USA). After digestion with RNAse A (0.5 mg/ml) for 1 h at 50³C, DNA was washed twice and samples were applied to a 2% agarose gel containing 1 Wg/ml ethidium bro-
BBAGEN 24891 1-10-99
362
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
mide for electrophoresis. A DNA ladder l DNA Eco digest was used as a DNA fragmentation marker. LDH was measured using a microassay kit (Sigma). 2.7. Statistical analysis Before starting the experiments, tubes containing the samples were coded by assigning numbers to each one. At the end of each experiment, data were collected and means for each triplicate calculated. The codes were then broken to identify sham- and magnetic ¢eld-exposed samples. Data are expressed as the mean þ S.D. of the average of the values obtained in di¡erent experiments. A two-tailed paired Student's t-test was utilized to compare results of sham-exposed and magnetic ¢eld-exposed samples obtained in n independent experiments. P 6 0.05 was considered to be signi¢cant. 3. Results 3.1. E¡ect of a 60 Hz magnetic ¢eld on PMN oxidative burst Data published previously suggested an e¡ect of EMF on free radical generation [17,30]. One of the most important functions of PMNs is the generation of reactive oxygen species (H2 O2 , O3 2 ) through the oxidative burst. The e¡ect of a 60 Hz magnetic ¢eld on H2 O2 formation was ¢rst assessed by measuring chemiluminescence, since this assay requires relatively few cells and gives continuous information about the initial steps and the kinetics of PMN oxidative metabolism. PMNs were sham-exposed or magnetic ¢eld-exposed, then, they were removed from the incubation chambers and chemiluminescence was immediately monitored in the presence or absence of PMA. Unstimulated PMNs produced a very small amount of chemiluminescence (14 000 cpm). Upon stimulation with 15 nM PMA, a sustained increase was observed re£ecting massive H2 O2 production (Fig. 1). Fig. 1 represents the e¡ect of 10 min exposure to the magnetic ¢eld on chemiluminescence generated by PMNs. There was no signi¢cant di¡erence in chemiluminescence between sham- and magnetic ¢eld-exposed groups either in unstimulated or stimulated cells. Di¡erent times of
Fig. 1. E¡ect of a 60 Hz magnetic ¢eld on chemiluminescence generation by human PMNs. PMNs (105 /ml) were incubated for 10 min at 37³C in the presence (closed symbols, solid lines) or absence (open symbols, broken lines) of a 22 mT magnetic ¢eld. The suspension (1 ml) was mixed with luminol, followed by addition of bu¡er (control) or 15 nM PMA. Chemiluminescence was recorded for 15 min and expressed as 106 cpm/105 PMNs. The ¢gure is a typical recording of one experiment representative of three similar others.
exposure (1, 5, 10, 15 and 30 min) were tested, but the only di¡erences we observed were in the range of experimental £uctuations. The previous method of chemiluminescence presented a technical limitation. Stimulation and exposure could be performed only sequentially and demonstrated that a 60 Hz magnetic ¢eld did not alter H2 O2 generation by PMNs. We used another experimental procedure consisting of measuring directly the production of O3 2 during magnetic ¢eld exposure. 3.2. E¡ect of a 60 Hz magnetic ¢eld on superoxide production Superoxide (O3 2 ) is the initial one electron reduction product of oxygen in the respiratory burst of PMNs. Its generation in stimulated PMNs occurs after NADPH oxidase translocates to the membrane and catalyzes the electron transfer from the donor NADPH to the acceptor O2 . The possible alteration of this electron transfer by EMF could be re£ected by a change in the amount of O3 2 generated. We examined the e¡ect of a sinusoidal 60 Hz, 22 mT magnetic ¢eld on O3 2 production. In resting PMNs,
BBAGEN 24891 1-10-99
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
Fig. 2. E¡ect of a 60 Hz magnetic ¢eld on superoxide production by PMA-stimulated PMNs. PMNs (2.5U106 /ml) were stimulated with 15 nM PMA and exposed or not to a 22 mT magnetic ¢eld. O3 2 production was measured after di¡erent periods of exposure. Data represent the mean þ S.D. of seven experiments and are expressed as the % of the corresponding non-exposed sample values (as a ratio magnetic ¢eld/shamU 100%). Italicized numbers are P values for magnetic ¢eld versus control (sham). Control values were 2 þ 0.4, 2.8 þ 0.3, 5.2 þ 0.8, 11.2 þ 1.2, 24.4 þ 1.7, 31 þ 4.2 nmol O3 2 /0.5, 1, 3, 5, 8, 10 min, respectively.
the production of superoxide was less than 2 nmol/ 2.5U106 cells (Fig. 2) and remained unchanged even after 60 min exposure to the magnetic ¢eld (1.8 þ 0.7 versus 1.7 þ 0.8, for sham versus magnetic ¢eld, respectively, n = 10, P = 0.56). Stimulation of PMNs with PMA (15 nM) induced the production of large amounts of O3 2 . 15 nM PMA is a suboptimal dose that produces 60^80% of the maximal oxidative response in PMNs. To examine the possible alteration of this electron transfer by a magnetic ¢eld, cells were simultaneously sham- or magnetic ¢eld-exposed with or without PMA. At di¡erent times of exposure/ stimulation, the reaction was stopped and superoxide generation was measured. PMA-stimulated PMNs showed a signi¢cant increase in O3 generation 2 when exposed to a magnetic ¢eld (Fig. 2). The e¡ect of a 60 Hz magnetic ¢eld was more important during the ¢rst minutes of O3 2 production, becoming statistically insigni¢cant after 10 min of exposure. As an additional control experiment, cells were incubated in the presence of PMA and placed randomly in either of the two incubation chambers without any
363
applied magnetic ¢eld (sham versus sham). PMAstimulated PMNs incubated in one or the other incubation chambers did not show any di¡erences (Fig. 3), indicating that the incubation chambers were identical. Since studies have reported e¡ects of EMF on radical generation [25] and on iron containing molecules [24], e.g. cytochrome. We examined the hypothesis that a magnetic ¢eld could interfere with the method we employed to measure O3 2 production. As described in Section 2, this assay is based on the reduction of ferricytochrome c to the ferrous form produced by O3 2 . We used the same conditions of incubation except that PMNs were replaced by a cell-free system consisting of the oxidation of xanthine by xanthine oxidase to form O3 2 and urate. Exposure of this cell-free system generating superoxide to a 22 mT, 60 Hz sine wave magnetic ¢eld was not a¡ected (Fig. 4) and there was no di¡erence between sham- and magnetic ¢eld-exposed groups. These results suggested that the target of the observed 60 Hz magnetic ¢eld e¡ect could be located
Fig. 3. Comparison of superoxide anion generation in the absence of a magnetic ¢eld in the two exposure chambers. PMNs (2.5U106 /ml) were split into two identical groups. Both groups were stimulated with 15 nM PMA and placed in one of the two incubation chambers. The reaction was stopped at the indicated times and superoxide production measured. Results are the means þ S.D. of three independent experiments. No di¡erences were noted in cells incubated in either chamber, indicating that the two chambers were identical.
BBAGEN 24891 1-10-99
364
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
suboptimal dose of PMA (Fig. 5). We examined the e¡ect of a 60 Hz magnetic ¢eld on lysozyme release under the same conditions as for L-glucuronidase. In our study, exposure of PMNs up to a 22 mT magnetic ¢eld had no e¡ect on either the basal or the stimulated release of lysozyme (Fig. 6). Furthermore, exposure of unstimulated cells to a 40 mT magnetic ¢eld for various periods of time up to 60 min did not a¡ect lysozyme release (data not shown). 3.4. E¡ect of a 60 Hz magnetic ¢eld on cell viability
Fig. 4. E¡ect of a 60 Hz magnetic ¢eld on superoxide anion generation in a cell-free system. A mixture of xanthine (0.5 mM) and ferricytochrome c (80 WM) was split into tubes and placed in either chamber. The reaction was started by adding xanthine oxidase (0.05 U/ml) to each tube and exposing the chambers to a 22 mT magnetic ¢eld (solid line) or none (broken line). The reaction was stopped at di¡erent times and superoxide production measured. Results are the means þ S.D. of four independent experiments.
We examined the e¡ect of 22 mT magnetic ¢eld exposure on viability of PMNs assessed by LDH release. Fig. 7 shows that as for lysozyme, LDH release was not a¡ected by exposure to a 22 mT, 60 Hz magnetic ¢eld, either in resting or PMA-stimulated PMNs. Furthermore, analysis of DNA fragmentation on an agarose gel con¢rmed that the cells were not undergoing apoptosis upon exposure to the
intracellularly upstream from the enzymatic reaction generating O3 2 . If this were true, we hypothesized that on other PMN functions such as degranulation could be similarly a¡ected by a magnetic ¢eld. 3.3. E¡ect of a 22 mT, 60 Hz magnetic ¢eld on degranulation Degranulation is a microbicidal and in£ammatory function of PMNs that has been shown to be associated with ion £uxes [31]. EMF exposure could affect this PMN function by interfering with ion movements. We tested this hypothesis by measuring Lglucuronidase and lysozyme. Lysozyme is a constituent of both primary (azurophile) and secondary (speci¢c) granules, while L-glucuronidase is localized in azurophile granules and smaller storage organelles. Our results showed that the baseline level of L-glucuronidase release was increased by 19% upon exposure to a 22 mT magnetic ¢eld, a more signi¢cant increase (53%) was observed when PMNs were simultaneously exposed to the magnetic ¢eld and a
Fig. 5. Alteration of L-glucuronidase release from PMNs during exposure to a 60 Hz magnetic ¢eld. PMNs (5U106 /ml) were incubated for 60 min with bu¡er or 15 nM PMA, in the presence or absence of a 22 mT magnetic ¢eld. 15 nM PMA is a suboptimal concentration providing 65% of the optimal response. LGlucuronidase release was quanti¢ed in supernatants by incubating with phenolphthalein glucuronic acid for 12 h and measuring the optical density of the product phenolphthalein. Quantities of L-glucuronidase released are expressed as the percentage of total L-glucuronidase in PMNs. Data are the mean þ S.D. of three separate experiments. *P 6 0.05 compared to sham. 2P 6 0.05 compared to control (without PMA).
BBAGEN 24891 1-10-99
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
Fig. 6. Lysozyme release from PMNs during exposure to a 60 Hz magnetic ¢eld. PMNs (5U106 /ml) were incubated for 10 or 60 min with bu¡er or 15 nM PMA, in the presence (hatched bars) or absence of a 22 mT magnetic ¢eld. Lysozyme release was quanti¢ed by measuring the optical density of a suspension of M. lysodeikticus. Quantities of lysozyme released are expressed as the percentage of total lysozyme in PMNs. Data are the mean þ S.D. of six separate experiments. *P 6 0.05 compared to control (without PMA).
magnetic ¢eld for 60 min (data not shown). Incubation of PMNs with TNFK (200 U/ml) alone used as a positive control showed that a signi¢cant apoptosis induction can be observed after 60 min of treatment. Moreover, it has been shown in a recent report that exposure of PMNs to a 50 Hz magnetic ¢eld for up to 3.5 h did not induce apoptosis [13]. 4. Discussion Several studies have shown e¡ects of EMF on calcium £ux [32] and free radical generation [17,34]. The e¡ect of EMF has been observed in many immune cell types [33]: thymocytes [16], B-lymphocytes [19], T-lymphocytes [20] and peripheral blood monocytes [34]. Only one study has been done previously in human PMNs, using a strong static magnetic ¢eld [14], and another in rat peritoneal neutrophils, using a 60 Hz magnetic ¢eld [17]. Stimulation of PMNs induces a membrane potential change and transloca-
365
Fig. 7. LDH release from PMNs during exposure to a 60 Hz magnetic ¢eld. PMNs (5U106 /ml) were incubated for 10 or 60 min with bu¡er or 15 nM PMA, in the presence (hatched bars) or absence of a 22 mT magnetic ¢eld. LDH release was quanti¢ed using a LDH detection kit (Sigma). Data are the mean þ S.D. of ¢ve separate experiments.
tion of enzymes to the cell membrane. These events could be a¡ected by EMF exposure. Furthermore, calcium plays an important role in signal transduction and in the control of PMN functions such as enzyme secretion and production of free radicals. The production of free radicals, which is the primary function of PMNs, consists of electron transfer from NADPH to molecular oxygen, leading to the formation of reactive oxygen species. In this study, we assessed more speci¢cally the impact of a high £ux density (22 mT), 60 Hz sinusoidal magnetic ¢eld exposure on O3 2 production, degranulation (L-glucuronidase and lysozyme release) and cell viability (LDH release, apoptosis induction). The experimental conditions we used were similar to those described previously [10]. The exposure system described in Section 2 is exactly the same as used in these previous experiments. However, the present experiments were performed to provide maximum controls and to eliminate potential bias. The modi¢cations we introduced included the following: (1) performing experiments in two identical incubation chambers, one used as control (sham), while the other one is used to expose the cells to a 60 Hz magnetic
BBAGEN 24891 1-10-99
366
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
¢eld (this procedure allows for simultaneous treatment of the cells and eliminates any possible e¡ect due to di¡ering pre-incubation conditions), (2) sham exposure and magnetic ¢eld exposure were performed simultaneously, (3) every treatment was performed in triplicate to eliminate variations due to di¡erences between samples both for the sham-exposed group and for the magnetic ¢eld-exposed one, (4) exposures and data analyses were done blindly. We ¢rst examined the possible alteration of oxidative burst in human PMNs by exposure to a 22 mT, 60 Hz sinusoidal magnetic ¢eld. Many publications suggested an e¡ect of EMF on free radical formation [23,35]. Moreover, the oxidative burst process is dependent on calcium, it occurs after translocation of components of the NADPH-oxidase to the plasma membrane and involves iron containing enzymes. All these events could be altered by electromagnetic ¢elds. We utilized the SOD-inhibitable reduction of ferricytochrome c to detect speci¢cally the major initial product of the oxidative burst (O3 2 ). Under the described experimental conditions, we found a significant e¡ect of the 60 Hz magnetic ¢eld on O3 2 production. The maximal e¡ect has been observed at 3 and 5 min after PMA stimulation and EMF exposure (18^35% increase at 5 min). These results are consistent with the previous reported e¡ect of a magnetic ¢eld on rat neutrophils [17], despite using a di¡erent experimental system. In fact, in those experiments, the authors used rat peritoneal neutrophils elicited by casein, which are di¡erent from the human PMNs freshly isolated from the circulating blood that we examined. Also, a di¡erent method to measure free radical production was utilized [17]. We next examined the possibility that a magnetic ¢eld could alter the method of superoxide measurement or the chemical formation of O3 2 by utilizing a . We did not ¢nd any cell-free system generating O3 2 e¡ect of the 60 Hz magnetic ¢eld on superoxide generation by a system xanthine/xanthine oxidase, suggesting that the observed e¡ect may occur upstream of the O3 2 generation step. If this was true, we should be able to detect the same e¡ect on other PMN functions such as degranulation. We examined the e¡ect of a 60 Hz sine wave magnetic ¢eld on degranulation (lysozyme and L-glucuronidase release) and on viability (LDH release). Ex-
posure of PMNs to intensities up to a 40 mT magnetic ¢eld with or without PMA stimulation did not induce any di¡erences between sham- and magnetic ¢eld-exposed cells regarding lysozyme and LDH release. In contrast, our results showed that exposure to a 22 mT, 60 Hz sinusoidal magnetic ¢eld induced a signi¢cant increase of L-glucuronidase release, which was bigger in PMA-stimulated PMNs. The explanation for this could be attributed to the subcellular distribution of L-glucuronidase (azurophile granules and smaller storage organelles) and lysozyme (azurophile and speci¢c granules). This would indicate that the e¡ect of EMF on degranulation is speci¢c and could a¡ect the products localized in smaller storage organelles. Such a mechanism is not impossible and has been reported in human neutrophils stimulated with phosphatidic acid which induced the release of L-glucuronidase but not lactoferrin. PMA induced the same response. However, at a high intracellular calcium concentration, the release of both enzymes was induced [36]. Together, these results demonstrate that a 60 Hz magnetic ¢eld affects PMN functions. The mechanism of these changes may occur through e¡ects on intracellular signaling such as calcium, phospholipase D and protein kinase C and remains to be elucidated. Further experiments are also required to examine the range of £ux densities producing these e¡ects. Epidemiological studies indicate that £ux densities around 0.2 WT may be threshold values for possible health e¡ects [1,2,37]. In conclusion, in the conditions we described here, we showed a signi¢cant e¡ect of a speci¢c magnetic ¢eld (22 mT, 60 Hz sine wave magnetic ¢eld) on superoxide anion production and L-glucuronidase release induced by PMA in human PMNs, while no e¡ect was detected on O3 2 generation by a cell-free system. These results suggest an interaction of the 60 Hz magnetic ¢eld with the early signal transduction mechanism of the PMN. Elucidation of these results of the magnetic ¢eld on signal coupling mechanisms may clarify the possible basis for the described alterations of biological responses by EMF. Acknowledgements This work was supported in part by a National
BBAGEN 24891 1-10-99
R. Khadir et al. / Biochimica et Biophysica Acta 1472 (1999) 359^367
Institutes of Health Grant, GM15431. The authors would like to thank Dr Robert Liburdy and Dr John Wikswo for their expert assistance and instructive discussions. References [1] N. Wertheimer, E. Leeper, Am. J. Epidemiol. 109 (1979) 273^284. [2] D.A. Savitz, H. Wachtel, F.A. Barnes, E.M. John, J.G. Tvrdik, Am. J. Epidemiol. 128 (1988) 21^38. [3] S.J. London, D.C. Thomas, J.D. Bowman, E. Sobel, T.C. Cheng, J.M. Peters, Am. J. Epidemiol. 134 (1991) 923^937. [4] M.S. Linet, E.E. Hatch, R.A. Kleinerman, L.L. Robinson, W.T. Kaune, D.R. Friedman, R.K. Severson, C.M. Haines, C.T. Hartsock, S. Niwa, S. Wacholder, R.E. Tarone, New Engl. J. Med. 337 (1997) 1^7. [5] H.K. Florig, Science 257 (1992) 468^469. [6] A.M. Khalil, W. Qassem, Mutat. Res. 247 (1991) 141^146. [7] C.F. Blackman, J.P. Blanchard, S.G. Benane, D.E. House, J.A. Elder, Bioelectromagnetics 19 (1998) 204^209. [8] S. Paradisi, G. Donelli, M.T. Santini, E. Straface, W. Malorni, Bioelectromagnetics 14 (1993) 247^255. [9] R. Goodman, L.-X. Wei, J.-C. Xu, A.S. Henderson, Biochim. Biophys. Acta 1009 (1989) 216^220. [10] R.P. Liburdy, D.E. Callahan, J. Harland, E. Dunham, T.R. Sloma, P. Yaswen, FEBS Lett. 334 (1993) 301^308. [11] R. Karabakhtsian, N. Broude, N. Shalts, S. Kochlatyi, R. Goodman, A.S. Henderson, FEBS Lett. 349 (1994) 1^6. [12] G. D`Ambrosio, A. Scaglione, D. DiBerardino, M.B. Lioi, L. Iannuzzi, E. Mostcciuolo, M.R. Scar¢, J. Bioelectr. 4 (1985) 279^284. [13] T. Hisamitsu, K. Narita, T. Kasahara, A. Seto, Y. Yu, K. Asano, Jpn. J. Physiol. 47 (1997) 307^310. [14] F.J. Papatheofanis, Radiat. Res. 122 (1990) 24^28. [15] W.R. Adey and A. Sheppard, in: M. Blank and E. Findl (Eds.), Mechanistic Approaches to Interactions of Electromagnetic Fields with Living Systems, Plenum Press, New York, 1987, pp. 365^387.
367
[16] J. Walleczek, R. Liburdy, FEBS Lett. 271 (1990) 157^160. [17] S. Roy, Y. Noda, V. Eckert, M.G. Traber, A. Mori, R. Liburdy, L. Packer, FEBS Lett. 376 (1995) 164^166. [18] J.C. Scaiano, F.L. Cozens, N. Mohtat, Photochem. Photobiol. 62 (1995) 818^829. [19] F.M. Uckun, T. Kurosaki, J. Jin, X. Jun, A. Morgan, M. Takara, J. Bolen, R. Luben, J. Biol. Chem. 270 (1995) 27666^27670. [20] E. Lindstro«m, P. Lindstro«m, A. Berglund, K.H. Mild, E. Lungren, J. Cell. Physiol. 156 (1993) 395^398. [21] R.A. Luben, C.D. Cain, in: W.R. Adey and A.F. Lawrence (Eds.), Non-linear Electrodynamics in Biological Systems, Plenum Press, New York, 1984, pp. 23^34. [22] M.J. Azanza, A. DelMoral, Prog. Neurobiol. 44 (1994) 517^ 601. [23] F.L. Cozens, J.C. Scaiano, J. Am. Chem. Soc. 115 (1993) 5204^5211. [24] J.L. Philips, Immunol. Lett. 13 (1986) 295^299. [25] R. DelCarratore, E. Morichetti, C. DellaCroce, G. Bronzetti, Bioelectromagnetics 16 (1995) 324^329. [26] J.C. Gay, K. Raddassi, A.P. Truett III, J.J. Murray, Biochim. Biophys. Acta 1336 (1997) 243^253. [27] R.P. Liburdy, FEBS Lett. 301 (1992) 53^59. [28] B.D. Cheson, R.L. Christensen, R. Sperling, B.E. Kohler, B.M. Babior, J. Clin. Invest. 58 (1976) 789^796. [29] J.M. McCord, I. Fridovich, J. Biol. Chem. 244 (1969) 6049^ 6063. [30] B. Brocklehurst, K.A. McLauchlan, Int. J. Radiat. Biol. 69 (1996) 3^24. [31] P.H. Naccache, H.J. Showell, E.L. Becker, R.I. Sha'a¢, J. Cell Biol. 75 (1977) 635^649. [32] J. Walleczek, FASEB J. 6 (1992) 3177^3185. [33] A. Lacy-Hulbert, J.C. Metcalf, R. Hesketh, Radiat. Res. 144 (1998) 9^17. [34] C. Petrini, M.L. Dupuis, A. Polighetti, C. Ramoni, P. Vecchia, Bioenergetics 44 (1997) 121^125. [35] T.T. Harkins, C.B. Grissom, Science 263 (1994) 958^960. [36] W. Zaman, T. Mitsuyama, M. Hatakenaka, D. Kang, S. Minakami, K. Takeshige, J. Biochem. 115 (1994) 238^244. [37] S. Preston, W. Navidi, D. Thomas, P.J. Lee, J. Bowman, J. Pagoda, Am. J. Epidemiol. 143 (1996) 105^119.
BBAGEN 24891 1-10-99