stimulation of ca2+ influx in rat pituitary cells under exposure to a 50 hz magnetic field

Bioelectromagnetics 17:303-311 (1996)

Stimulation of Ca2+ Influx in Rat Pituitary Cells Under Exposure to a 50 Hz

Magnetic Field

Elisabeth Barbier, Bernard Dufy, and Bernard Veyret

Laboratoire de Physique des Interactions Ondes-Matiere, C.N. R.S., E.N.S.C. P.B., Universite Bordeaux I, Talence, France (E. B., B. V.): Laboratoire de

Neurophysiologie, Universite de Bordeaux 11, C. N. R.S., Bordeaux, France (E. B., B. D.)

The effect of exposure of single rat pituitary cells to 50 Hz sine wave magnetic fields of various strengths on the intracellular free Ca2+ concentration ([Ca"],) was studied by using dual-emission microfluorimetry, using indo- 1 as probe. A 30 min exposure of the cells to vertical 50 pT peak magnetic field triggered a long-lasting increase in [Ca*+], from a basal value of about 185 k 4 nM to 326 k 41 nM (S.E.; n = 150). The vertical and horizontal components of the static magnetic field were 57 and 15 pT, respectively. The 50 Hz ambient magnetic field was always below 0.1 pT rms. The effect was observed both at 25 ? 2 "C and at 37? 2 "C. Responsive cells, for which [Ca*+], rose to values above 309 nM, were identified as lactotrophs and represented 29% of the total pituitaries. [Ca2+], increase, for the most part, was due to Ca2+ influx through voltage-dependent dihydropiridine-sensitive calcium channels inhibited by PN 200-1 10. However, neither Ca2+ channel blockers nor removal of Ca2+ from the external medium during exposure completely prevented the field-induced [Ca2+], increase. Additional experiments using an MTT colorimetric assay showed that alteration of Ca*+ homeostasis of lactotrophs was associated with impairment of some mitochondria1 processes. 01996 Wiiey-l.iss, Inc.

Keywords: ELF, lactotrophs, metabolic activity, voltage-dependent calcium channels

INTRODUCTION

Although the matter still remains controversial for lack of conclusive experimental support, sinusoidal, low- intensity (0.1-1 0.0 mT) 50-60 Hz magnetic fields have been shown to produce biological effects, such as stimu- lating nerve regeneration [Rusovan and Kanje, 19911 or altering gene transcription [Greene et al., 1993; Phillips, 19931. They may also play a synergic role in cellular processes that are already activated, such as cell proliferation [Walleczek, 19921. The role of Ca2+ in the trans- duction of these effects has been suggested, and indirect evidence of its involvement has been shown [Walleczek, 1992; Karabakhtsian et al., 19941. One report of a direct action of low-strength ( I 00 pT) 50 Hz magnetic fields on cytosolic free Ca2+ concentration ( [Ca2+Ii) in Jurkat- cells described oscillations resembling those triggered by mitogen activation [Lindstrom et al., 19931. The physiological significance and the underlying mechanisms of the magnetic field effects remain unknown.

For this work, we studied the effects of low-strength magnetic field exposure on the intracellular calcium concentration of endocrine cells from the rat anterior

pituitary. These cells have been extensively studied, and the regulation of [Ca2+Ii, which is involved in the secre- tory process, is well known [Schlegel et al., 1987; Mollard et al., 1989; Stojilkovic and Catt, 19921. In particular, we followed variations of [CaZ+ll in single cells during exposure to magnetic fields of various strengths, assessing some of the biological and physical mechanisms. In addition, we looked for a biological effect of exposure on metabolic activity.

MATERIALS AND METHODS

Cell Culture

Cells were obtained from anterior pituitary lobes of 3-month-old female Wistar rats. The tissue was di- gested by sequential incubations in a solution of 0.1 %

Received for review July 6, 1995; revision received November 20, 1995.

Address reprint requests to E. Barbier, Laboratoire de Physique des Interactions Ondes-Matibre, C.N.R.S. U.R.A. 1506, E.N.S.C.P.B., UniversitC Bordeaux I. BP 108, 33402 Talence cedex, France.

0 1996 Wiley-Liss, Inc.

304 Barbier et al.

trypsin (Sigma, LaVerpillihe, France) at 37 "C for 20 min and in a solution of 0.02% DNase (Boehringer, France) for 1-2 min. It was then successively washed in a Ca2+/Mg2+- free Hank's balanced salt solution (HBSS; see below) containing 2 mM and 1 mM EGTA and, finally, was mechanically dissociated i n Ca2+/Mg2+-free HBSS. The dispersed cells were cultured in 35 mm petri dishes in Ham's F10 nutrient medium (Gibco BRL, France) supple- mented with 10% (vh) fetal calf serum (FCS; Seromed, Strasbourg, France) and maintained at 37 "C in a 5% CO, humidified atmosphere for at least 3 days prior to experiments. One pituitary finally yielded to 2-3 x lo6 viable cells. Viability was assessed by using the trypan blue exclusion test. For [Ca2+], measurements, 2 x lo5 cells per ml were seeded on glass coverslips pretreated with polyornithine ( 5 giliter). Grids were etched on some of the coverslips for postimmuno-identification.

Cytosolic-Free Calcium Measurements

Indo- 1 dual-emission microfluorimetry was used as described by Mollard [Mollard et al., 19891. Briefly, cells were loaded for 30 min at 37 "C with 5 pM indo- 1 penta-acetoxymethylester (indo- l/AM; Sigma, France) and 0.02% (w/v) pluronic F-127 (Molecular Probes; Eu- gene, OR) in a final volume of 1.5 ml of a modified HBSS containing (in mM) NaCl(136.9), KCl(5.9, CaCl, (2.0), MgCl, (2.0), NaHCO, (4.2), Na,HPO, (0.3), KH,PO, (0.4), HEPES ( lo) , and glucose ( S . S ) , pH 7.3-7.4, 300- 3 I0 mosmol/kg. They were then rinsed and kept in HBSS. Indo-1 fluorescence was monitored either at room temperature (25 f 2 "C) or at 37 f 2 "C by using an in- verted microscope equipped for microfluorimetry (Nikon, Paris, France). [Ca2+], was estimated by the ratio method using single wavelength excitation (355 nm) and dual emission (405 nm Ca2+ bound480 nm Ca2+ free) ana- lyzed with Axotape software on an IBM PC (version 1.2.01; Axon Instruments, Foster City, CA).

Temperature Regulation Most experiments were carried out in a

thermostated room (25 k 2 "C). Some [Ca'+], measurements were performed at 37 & 2 "C; the coverslip in the petri dish was placed at the bottom of an open per- fusion system, where both inlet and outlet were con- nected to a peristaltic pump (Gilson Minipuls 2; France). The recording medium superfused the cells at a flow rate of 0.8 ml/min and was kept at 37 If: 2 "C by warming the inlet. In our lab, [Ca2+Ii measurements of pituitaries are routinely made at room temperature, around 25 "C, and those conducted at 37 "C do not lead to any variation of calcium behaviour.

Magnetic Exposure

Magnetic fields were induced through a coil fit- ted around the petri dish and supplied with a 50 Hz sig- nal from a generator (Wavetek 178). The coil was double-wound, allowing either real exposure to the magnetic field or sham exposure when the magnetic field of the windings cancel each other. Controls were performed under this last condition, and cells were only exposed to the static local field (geomagnetic and ambient building fields). The magnetic field strength at the cell location, which was measured with a fluxgate meter composed of three separate probes (10 x 4 x 4 mm each; Bartington, United Kingdom), was expressed as peak value. It was measured while the objective of the microscope was in place. The average value of the vertical component of the static field was 57 pT, and the ambient 50 Hz magnetic field, which was monitored with a gaussmeter (Mag Check 50; MSI, USA), was always below 0.1 pT rms. Recorded cells were all located at the center of the petri dish, where the induced electric field and current density, both proportional to the distance to the center of the coil, were negligible.

Four types of experiments were performed. A total of 60 cells (seven different experiments with 8 I n I 12) were sham exposed for 30 min, and 33 cells (four different experiments with 8 5 n I 10) were sham exposed for 3 h. A total of 150 cells were exposed to the magnetic field for 30 min ( 1 6 different experiments with 8 I n I 12), and 103 cells were field exposed for 3 h (ten different experiments with 8 5 n I 12).

lmmunoidentification of recorded cells

After [Ca2'li measurement, the cells, which were located by their coordinates on the etched coverslips, were identified by double-label immunochemistry monkey-anti-rGH-IC-1 (rGH; 1/4,000) and rabbit-anti- rPRL-IC-5 (rPRL; 1/2,000) of the contents of their secre- tory granules. Cells were fixed in 4% paraformaldehyde, 1 5% saturated picric acid phosphate-buffered saline (PBS) 0.1 M, pH 7.2, rinsed in PBS buffer containing 1.2% Triton X-100, 0.1% BSA, and 6% normal goat serum, and labelled by using the double-label immun- ofluorescence method described by Pryzwanski [Pryzwanski et al., 19781. The secondary antibodies, goat anti-monkey-FITC ( 1 / I 00) and goat anti-rabbit-TRITC (1 /loo), were purchased from TEBU (Le Perray, France) and Tmmunotech (Marseille, France), respectively. An- tisera containing primary antibodies were obtained from NIDDK (Bethesda, MD). The specificity of the label- ing was checked by omitting the primary antibody.

Magnetic Field on [Caz+]i in Pituitaries 305

Metabolic Activity

MTT colorimetric assay [Mosmann, 19831, modified by Tada [Tada et al., 19861, was used to measure the cell activity, as explained by Gerlier [Gerlier and Thomasset, I 9861. Briefly, 3-(4,5-Dimethylthiazo1-2-y1)- 2,5-diphenyltetrazolium bromide (MTT; Sigma, France) was dissolved at a concentration of 5 mg/ml in PBS, ster- ilized by filtration, and kept in the dark at 4 “C. After cells had been exposed for 1 h in HBSS (2 ml) and [Ca2+Ii had been measured, a 200 p1 MTT solution was added and left for 2 h, allowing MTT metabolization. Formazan crystals, which were produced by MTT cleavage, were then observed in single cells under the microscope.

Data Analysis

Results are expressed as mean value k standard deviation (Mv f S.D.). Statistical analysis of the data was performed by using one-sample Kolmogorov- Smirnov tests (K-S 1) for normality of distribution and two-sample Kolmogorov-Smirnov tests (K-S 2) for equality of distribution. Comparisons between mean values were performed by using two-tail Wilcoxon signed rank and two-tail Mann-Whitney tests [Siege] and Castellan, 19881.

RESULTS

[ C a2+] M ea s u re m e n t s In anterior pituitary cells, [Ca’+], often oscillates

spontaneously [Schlegel et al., 1987; Mollard et al., 1989; Stojilkovic and Catt, 19921. Thus, basal [Ca2+], level was defined as the minimal [Ca2+], value reached between two oscillations. In this study, it was approximately 180 nM. Figure 1A presents the protocol of field exposure.

After 30 rnin of sham exposure, the cells (n = 60) displayed a slight decrease in [Ca’+], with respect to the baseline (149 k 60 nM vs. 163 k 48 nM ( P = .0005; Wilcoxon signed rank). No variation of [Ca2+II was observed after a 3 h sham exposure ( P = >.5; Wilcoxon signed rank; Fig. 1B).

Cells (n = 150) that were exposed to the magnetic field for 30 min displayed an increase in their [Ca2*], on average from 185 k 46 to 326 f 504 nM ( P < .0001; Wilcoxon signed rank). In another group of cells that were (n = 103) exposed for 3 h, [Ca2+], increased from 169 k 43 nM to 379 k 407 nM ( P < .0001; Wilcoxon signed rank).

N o difference was observed between the 30 rnin and the 3 h exposed groups ( P > = .5; Mann-Whitney). These results exclude the possibility of errors in [Ca2+], measurements that could occur from weak fluorescent

sham-exposure A indo-1

e

B [Ca2*ll (nM) Sham-exposure

400

300 ns I

200 -I I

loo 0 I baseline 30 min baseline 3 h

sham n=60

sham n=33

IC~*+I , ( n ~ ) Field exposure

C *

* 500

400 6oo 1 100

30 min I baseline 3 h I exposure ex P o s u r e

n= 150 n=103

Fig. 1. A: Experimental protocol of single cell [Caz+]i measurements during magnetic field exposure. After indo-l load- ing, cells were sham exposed for 15 rnin in Hank’s balanced salt solution (HBSS), then [Ca*+]i was measured to determine the basal level or “baseline”. Thus, cells were their own control. These preexposure conditions, i.e., coil on or off sham, were performed in order to eliminate any heating effect due to the current in the coils. 6: Sham exposure led to a decrease in [Caz+]i after 30 min exposure (***f = .0005 vs. baseline; Wilcoxon signed rank). A 3 h exposure did not lead to any variation of [Caz+]i. Magnetic field induced a twofold increase of [Caz+]i both in 30 rnin and 3 h exposure experiments (***P< .0001 vs. baseline; Wilcoxon signed rank). No difference was observed between 30 rnin and 3 h exposed groups (f = .5; Mann-Whitney).

306 Barbier et al.

signals caused by dye leakage or bleaching over a long period of time, e.g., 3 h. However, the large standard deviation in the magnetic field-exposed groups called for a careful statistical analysis of data distribution. Baseline and exposed data are defined as [Ca"], values before and after exposure, respectively. Analysis of the baseline groups. Baselines of both sham (n = 60) and field-exposed groups (n = 150) had a normal distribution ( P c .01; K-S 1). The two normal distributions are the same ( P < .01; K-S 2); thus, the two groups can be considered as issued from the same population. Analysis of sham exposure. The sham exposure data are summarized in Table I . [Ca2+], has a tendency to decrease over a period of 30 min. This was not found i n the 3 h experiment, probably because of the small size of the group, i.e., n = 33. Analysis of field exposure. The field exposure data are summarized in Table 2. The high S.D. of field-exposed groups suggested that they consisted in more than one population. Because the baseline group had a normal distribution (Mv = I 86 f 45 S.D.), this meant that less than 0.003 of the values should have been larger than the ar- bitrary value of Mv k 3 S.D. or, in our case, 309 nM. Nineteen percent of the 30 min exposed cells (28 of 150) had a [Ca'+Ii larger than 309 nM and could reach values above 1 pM. Therefore, these cells were responsive to the magnetic field. Responsive cells represented 30% of the 3 h exposed cells (31 of 103). No difference was found between 30 min and 3 h exposed groups ( P > 2 3 ;

Mann-Whitney for mean values; K-S 2 for distributions). Figure 2 shows two examples of responses to field exposure. The same experiment performed at 37 "C led to the same [Ca"], increase as that observed at 25 "C, in- dicating that the effect was not temperature dependent (n = 18; not shown).

1200 -

1000 -

800 -

600 -

400 -

TABLE 1. Analysis of Sham Exposure*

[Caz+l. (nM)

30 Minutes ( n = 60) 3 Hours ( n = 33)

Measure Baseline Sham Baseline Sham

Mean value I63 I49 222 222 S.D. 47 60 49 54

Median 152 I30 200 205 Mi ni m uni 73 1 3 I65 I42 Maximum 286 355 362 356

K-S 1 P < . 0 l P < . 0 l P<.OI P < . O l "In the 30 min exposure experiment, both baseline and sham-exposed groups are normal IP < .01 ; one-sample Kolmogorov-Smirnov test (K-S I ) ] . In the 3 h exposure experiment, both baseline and sham- exposed groups are normal ( P < .01; K-S 1).

TABLE 2. Analysis of Field Exposure*

ICa2+l (nM)

30 Minutes (n = 150) 3 Hours (n - 103)

Measure Baseline Exposed Baseline Exposed

Mean value 186 326 I69 379 S.D. 45 502 43 405

Median 178 205 I65 223 Minimum 80 I09 90 I03 Maximum 314 4,261 286 2,354

K-S 1 P < .O1 P < P < .01 P < 10-6 *In the 30 min exposure, the field-exposed group distribution was not normal [P < lob; one-sample Kolmogorov-Srnirnov test (K-S I ) ] and was different from its baseline ( P < .0001; Wilcoxon signed rank). In the 3 h exposure, the field-exposed group was not normal ( P < K-S 1 ) and was different from its baseline ( P < .0001; Wilcoxon signed rank). The 30 min and 3 h exposed groups had the same distribution and were not different ( P = .5; Mann-Whitney).

Exposure 800 4 600 1 4 o o L l l 200 3 h

2 0 c

Exposure

11

30min 200

20 sec - Fig. 2. Two significant examples of high calcium responses to exposure to 50 Hz, 50 FT magnetic field. A: In a silent cell, [CaP+]i increased to three times the baseline value after a 3 h exposure. 8: In this oscillating cell, the same response pat- tern was observed; [Caz+]i rose with a small oscillating activity after a 30 min exposure.


the “ion parametric resonance model” [Blanchard and Blackman, 19941. These models are based on the con- cept that specific ions may be the target of the interaction and that the functions in which they are involved may be affected across a range of intensity values of the applied sine wave magnetic field (BA,). Conditions for ion resonance are defined as follows: The frequency of B,, is proportional to B,, q/m, where B,, is the strength of the static field (e.g., geomagnetic), and where q and m are the charge and the mass of the ion involved in the process, respectively. B,, and B,, must be parallel and must be of the same amplitude. In our case, the ion involved is Ca2+. B,, is the vertical component of the static field, B,, is induced perpendicular to the plane of the coil, and, because both of them are vertical, they are parallel. According to the model for calcium resonance, for a frequency of 50 Hz, the theoretical value of B,, is 65 pT (with B, = 65 pT), which is similar to our conditions (BDc = 57 pT). Because the effect was found for B,, = 50 pT, and no variation of [Ca”], was observed at 250, 10, and 2 pT, some type of resonance for Ca2+ may exist near our set of conditions (B,, = 57 pT, frequency = 50 Hz, and B,, = 50 pT).

The ion cyclotron resonance has been the most tested hypothesis involving Ca2+ [Liboff et a]., 1987; Garcia-Sancho et al., 1994; Hojevik et al., 19951. To explore this hypothesis, we suppressed one of the resonance condition parameters: The vertical component of the static magnetic field B,, was set to zero from its original value of 57 pT. This was performed by apply- ing an inversed, vertical, static magnetic field of the same amplitude (57 pT). In this experiment, six of nine monitored cells were responsive, showing a large increase in [Ca”], (not shown). Because cells responded to the applied B,, (50 pT) under both B,, conditions (57 and 0 pT), the cyclotron or ion parametric resonance hypothesis cannot be retained in our case.

Identification of Responsive Cells

Responsive cells were identified after [Ca2+], measurements were performed under magnetic field exposure. Of the 18 cells tested, seven responded to field exposure by a large increase in [Ca2+Il, from a basal level of 207 f 23 nM (Mv k S.D.) to 1,179 k 959 nM ( P = .015; Wilcoxon signed rank; n = 7). Immunochemical labelling revealed that 50% of the pituitary cells (9 of 18) were immunoreactive to rat Prolactin (rPRL) and thus could be identified as lactotrophs, 28% (5 of 18) were immunoreactive to rat growth hormone (rGH), and 22% (4 of 18) were neither labeled with rPRL nor rGH; therefore, they belonged to the other families secreting ACTH, L H, FSH, and TSH [Hopkins and Farquhar, 19731. All

3000 -

2500 - 2000 - 1500 - 1000 -

500 -

Time-course of [Ca2+], Increase

A kinetic study of [Ca2+], enhancement was performed on 24 cells (three different experiments; n = 8) that were exposed to a 50 Hz, 50 pT sinusoidal magnetic field. Basal [Ca2+Ii remained constant during the first 10 rnin of exposure. Then, for 25% of the cells (n = 6; see Fig. 3 ) , [Ca2+II increased twofold after 30 min. In five cells, [Ca2+], reached 2 4 pM at the end of the experiment, i.e., 80 min, remaining high during the 20 min after the field was shut off. According to the Grynkiewicz equation [Grynkiewicz et al., 1985; Tsien et al., 19851, indo-1 can be used to measure [Ca2+], within the range of from 100 nM to 30 pM, but accuracy is poorer for concentrations ranging from 2 to 30 pM, where the calibration curve is asymptotic. Within the range from 100 nM to 2 pm, the calibration curve is linear [Mollard et al., 19891. Therefore, in this experiment, the maximum values of [Ca2+Ii corresponded to the maximum binding of Ca” and indo-1 .

Dose Response

Similar time-course experiments at field strengths of 2, 10, and 250 pT (two experimentslfield strength; n = 1 1 experiments) did not reveal any change in [Ca’+II when cells were exposed (not shown). Our finding of an effect at a single field strength (50 pT) suggested that this effect might fit some of the theoretical models of ELF field interaction with biological systems, including “ion cyclotron resonance” [Liboff et al., 19871 and

(nM) Exposure

0 ! . 1 . 1 . r . I - 1 . i

-

-20 0 20 40 60 a0 100

Time (min)

Fig. 3. Time course of 50 Hz, 50 mT sinusoidal magnetic field exposure of highly responsive cells. Magnetic field was on during the first 60 rnin and then was turned off for the last 20 min. [Gaz+], remained high for 20 rnin after removal of the magnetic field.

308 Barbier et al.

of the highly responsive cells (group B) were immunoreactive to rPRL and constituted 78% (7 of 9) of the total lactotrophs.

Origin of [Ca*+], Increase

To determine the origin of the increase in [Ca2+I1, calcium ions were removed from the recording solution during 50 Hz, 50 pT magnetic field exposure (n = 30). The latency necessary to observe any field effect was at least 20 min, and 0 mM Ca2+ experiments were not conducted, because they would lead to Ca2+ pool deple- tion and a poor viability in a protocol lasting more than 45 min. Removal of Ca2+ during exposure resulted in a dramatic decrease of [Ca'+], (Fig. 4). Cells were exposed to the 50 Hz, 50 pT magnetic field for 30 min in HBSS solution in order to identify responsive cells, then they were rinsed and placed in a CaCI,-free HBSS buffer (2 mM EGTA). [Ca2+], was measured 15 min after the medium had been changed to give time for chelation of remaining Ca2+, whereas cells were still exposed to the magnetic field. Therefore, the sustained [Ca2+II rise, which was triggered by magnetic field exposure, was due largely to Ca2+ influx across the plasma membrane. However, [Ca2+], remained significantly higher than the basal level, suggesting that Ca2+ homeostasis was altered by field exposure.

To further analyze the nature of this influx, PN 200- 110 (Bioblock Scientific, Tllkrich, France), which is a specific blocker of L-type Ca2+ voltage-dependent channels [Cognard et al., 19861 was used (n = 16). Cells were exposed to the magnetic field for 1 h to identify responsive cells. PN 200-1 10 (100 nM) was then added to the recording solution. Figure 5 shows that, following the addition of PN 200- 1 10 (100 nM), the elevated [Ca2+II in responsive cells decreased towards the basal values, but one should note that, in some cells, it still remained high in some cells. The same applied when Ca" was removed from the extracellular medium. Hence, the inflow of calcium ions observed during the sustained rise in [Ca2+II involved a dynamic process through voltage- operated calcium channels.

MTT Test for Cell Metabolic Activity

The metabolic activity of the individual cells was assessed after microfluorimetric [Ca2+Il measurement by watching the formation of formazan crystals in recorded cells under a microscope. Immediately after the last [Ca2+II measurement (i.e., after a 3 h maximum exposure), 200 p1 MTT solution was added to the 2 ml of recording medium, and, 2 h later, crystal formation was observed. Under these conditions, formazan production is proportional to the succinate dehydrogenase activity,

700

6 o o ]

500

400

300

200

100

n I

baseline ' exposed ' o-ca'+

ExDosure

1 6 0 0 i 1400 I I

1200

600

400

200 30 min

11- 15 min

20 sec - Fig. 4. A: Dependence of the response on Ca2+ entry (*P < .05; Wilcoxon signed rank vs. baseline; n = 30). B: Example of the dependence on calcium entry on field response.

which is mostly a mitochondrial enzyme. Hence, by using this test, some mitochondrial processes may be assessed [Gerlier and Thomasset, 19861. In nonresponsive cells, crystals became visible. Conversely, no crystal formation was observed in responsive cells.

Because the extremely high [Ca2+], values that were reached during magnetic field exposure may indicate that cells were damaged, viability was systematically assessed by the exclusion of trypan blue. One hour after the increase in [Ca2+], triggered by field exposure, none of the responsive cells was stained with trypan blue, thus, excluding the possibility of damage to the cell membrane.

DISCUSSION

This report describes the effect of 50 Hz, 50 pT sinusoidal magnetic field on cytosolic free Ca2+ concentration in pituitary cells. Exposure to the field was found


Grissom, 19941, and, moreover, there is no rationale within this hypothesis for the peculiar field strength dependence of the effect that we observed i n this work.

This study provides some clues about the biological mechanisms of the magnetic field effect. Elevated [Ca2+Il, for the most part, is due to Ca2+ entry through voltage- dependent calcium channels. It should be noted that these channels are responsible for Ca2+ oscillations that regulate Ca2+ homeostasis in pituitaries [Schlegel et al., 1987; Stojilkovic and Catt, 19921. This Ca2+ inflow is inhibited by the dihydropiridine PN 200-1 10, which is known to block L-type voltage-dependent calcium channels [Stanfield, 1986; Miller, 19921 and steady-state calcium currents in pituitaries [Mollard et al., 19941. The latter, which is responsible for sustained [ [Ca2+], increase [Mollard et al., 19941, might be activated during field exposure and may be associated with the observed, high sustained [Ca2+],. The use of other Ca2+ voltage-dependent channel blockers and single-channel studies may help to determine their role more precisely.

Variations of [Ca2+], under 50 Hz sine-wave magnetic field exposure were reported in another study conducted by Lindstrom et al. [1993]. In that work, exposure to the field triggered [Ca2+], oscillations starting within the first minutes of exposure and ending within 10 rnin after field was shut off. It was similar to physiological oscillations either occurring spontaneously or triggered by the mitogen anti-CD3. [Ca2+Il oscillations were not found when Ca2+ was omitted from the bath- ing medium, suggesting that these Ca2+ increases have an external origin. Despite the presence of voltage dependent Ca2+ channels in Jurkat cells [Dupuis et al., 19891, Lindstrom et al. did not consider any direct effect of the magnetic field on these channels.

Our data are consistent with the observations reported by Lindstrom et al [ 19931. In our case, the use of voltage-dependent calcium channel blockers showed that most of the [Ca2+II increase was mediated through these channels. However, in pituitary cells, the effect was delayed, suggesting that it was not direct but was mediated through a biochemical process that has not yet been identified. Moreover, because neither the presence of a Ca2+ channel blocker nor the removal of Ca2+ from the medium completely prevented [Ca2+Il increase, Ca2+ entry would not seem to be the only process affected by magnetic fields. Thus, the two sets of results show that, although increases in [CaZ+lI are observed during exposure to 50 Hz magnetic fields, differences in experimental biological material and in the nature of the effects indicate that distinct phenomena may occur.

The MTT test showed that exposure to the magnetic field led to a drastic decrease in metabolic activity.

** T

baseline exposed PN 200-110

Exposure

2 mM Ca2+ 6 I c a 7 , (nM)

1 0 0 0 1 (+) PN 200-110

11

11- 15 rnin

60 m 200

20 sec - Fig. 5. A: Blocking field exposure triggered [Ca*+], increase by PN 200-110 (100 nM; * * P = .0049; Wilcoxon signed rank vs. baseline; n = 16). B: Example of partial blockade of the response by PN 200-1 10. PN 200-1 10 abolished the oscillatory response in this responsive cell.

to increase cytosolic free calcium concentration. In most of the cells tested, the [Ca”], rise was very weak, because only an overall 10% increase was noted. However, a significant proportion (20%) of the tested cells showed a dramatic increase of up to 400%.

Calcium response to field exposure follows a sur- prising dose-response behaviour, because only the 50 pT field elicit the effect. However, its mechanism is unknown, and, as shown, the “ion cyclotron resonance” hypothesis does not seem to hold up under our experimental conditions. Another hypothesis that should be considered involves the effect of magnetic fields on the reac- tion kinetics of free radicals [Steiner and Ulrich, 19891. These species are present within the cell as products of many biochemical reactions, including those triggered by absorption of UV light by indo-1. However, most effects of magnetic fields on free radicals have been observed at field strengths above 1 mT [Harkins and

310 Barbier et al.

Because MTT is reduced mainly by mitochondrial dehydrogenase, defects in its metabolization strongly suggest that exposure to the field was altering mitochondrial processes. Among these, the involvement of mitochondria in ca?+ homeostasis, which is expected to occur when [Ca2+], reaches lethal levels [Meldolesi et al., 19901, should be considered, because, in our case, [Ca2+], did reach over 1-2 pM, which is in the lethal range. This is also consistent with the observation that blockade of Ca2+ entry by either removing extracellular Ca2+ or blocking voltage-operated calcium channels did not completely restore normal [Ca2+],. Obviously, Ca2+ homeostasis was altered. Regulation of mitochondrial enzymes that are Ca” dependent may also be affected [McCormack et al., 19901. Whether the alteration of Ca” homeostasis is a cause or a consequence of the decrease in the metabolic activity of the cell remains to be determined.

Cell identification showed that only lactotrophs were responsive to the magnetic field exposure. A num- ber of studies have emphasized differences in the regulation of Ca2+ homeostasis in the different pituitary cell types [Stojilkovic and Catt, 19923. Indeed, pituitary cells are provided with membrane ionic channels and Ca2+ intracellular pools that are specifically involved in Ca2+ homeostasis but that differ from one cell type to another [Israel and Vincent, 19901. A low-threshold transient Ca” and a high-threshold sustained Ca2+ current have been identified in most pituitary cells. However, the proportion of each current type differs for each pituitary cell type [Lewis et al., 1988; Tse and Hille, 19931. Lactotrophs display large Ca” transients, which are indicative of Ca2+ action potentials that are absent from gonadotrophs. Moreover, regulation of intracellular Ca2+ pools specific to the cell type studied account for the cellular Ca2+ homeostasis [Stojilkovic and Catt, 19921. Therefore, alterations in the buffer activity of intracellular calcium stores may affect the cytosolic free Ca2+ level in lactotroph5 differently from gonadotrophs, for example. Observation of the specific sensitivity of lactotrophs to magnetic fields is important; however, additional experiments are required to understand the processes and the conditions of this magnetic field sensitivity.

Among the whole population of lactotrophs, 78% were highly responsive, suggesting a magnetic field- sensitive cell subpopulation. Several studies have dem- onstrated the heterogeneity of lactotrophs on the basis of their physiological properties, such as their basal prolactin secretion [Cota et al., 1990; Lledo et a]., 19911. Lactotroph separation on a gradient density revealed light and heavy subpopulations that corresponded to two

different functional states of the cells. The light fraction displays a higher basal prolactin secretion, which can be strongly inhibited by dopamine and is poorly sensitive to TRH. The [Ca2+], of this subpopulation oscillates and displays a higher basal level than that of the heavy fraction. The resting membrane potential is also lower [Lledo et al., 19911. The heterogeneity of lactotrophs is related to the proportion of two types of Ca2+ channels that have different electrical properties. Slowly deactivating and fast deactivating Ca2+ currents have been observed [Lledo et al., 19911, the surface density of the latter being a determining fac- tor for prolactin release [Cota et al., 19901. These characteristics may account for the different sensitivity to magnetic fields that we observed among the lactotroph population.

In conclusion, we have shown that a 50 Hz, 50 pT magnetic field can markedly affect endocrine cell physiological processes. This effect is specific to a proportion of the lactotroph cell subtype. Its characteristics include an alteration in cytosolic Ca2+ homeostasis involving voltage-dependent channels and impairment of metabolic activity, such as mitochondrial activity. Our results suggest two types of putative cell targets of the effect of the field: those located at the membrane level (i.e., ion channels) and/or those located at the subcel- lular level (e.g., the mitochondria). Experiments are in progress to assess the importance of the UV used for indo-1 excitation in the field effect. The ultimate consequence of field exposure and its physical mechanisms remain to be determined in order to improve understand- ing of the interaction between ELF magnetic fields and living systems.

ACKNOWLEDGMENTS

We are indebted to Dr. J.-M. Moreau for designing the coils, to Dr. W. Ellison for his precious help in statistical analysis, and to J. Audin for designing the temperature regulation system. We thank M. Bouchkhacheckh, J.-M. Calvinhac, and D. Varoqueaux for technical assis- tance. We are grateful to the National Hormone and Pi- tuitary Program, NIDDK, who provided the antibodies for immunocytochemistry. This research was supported in part by a grant from Electricit6 de France.

REFERENCES

Blanchard JP, Blackman CF (1994): Clarification and application of an ion parametric resonance model for magnetic field interactions with biological systems. Bioelectromagnetics 15:2 I 7-238.

Magnetic Field on [Ca2+]i in Pituitaries 31 1

Cognard C, Lazdunski M, Romey G (1986): Different types of Ca2+ channels in mammalian skeletal muscle cells in culture. Proc Natl Acad Sci USA 83:517-521.

Cota G, Hiriart M, Horta J, Torres-Escalante JL (1990): Calcium channels and basal prolactin secretion in single male rat lactotrophs. Am J Physiol 259:C949-C959.

Dupuis G, Heroux J, Payet MD ( I 989): Characterization of Ca2+ and K+ currents in the human jurkat T cell line: Effect of phytohaemagglutinin. J Physiol 412: 135-154.

Garcia-Sancho J, Montero M , Alvarez J, Fonteriz RI, Sanchez A ( 1 994): Effects of extremely-low-frequency electromagnetic fields on ion transport in several mammalian cells. Bioelectromagnetics 15:579-588.

Gerlier D, Thomasset N (1986): Use of MTT colorimetric assay to measure cell activation. J Immunol Methods 94:57-63.

Greene JJ , Pearson SL, Skowronski WJ, Nardone RM, Mullins JM, Krause D ( I 993): Gene-specific modulation of RNA synthe- sis and degradation by extremely low frequency elecfromag- netic fields. Cell Mol Biol 39:261-268.

Grynkiewicz R, Poenie M, Tsien RY (1985): A new generation of Ca2+ indicators with greatly improved tluorescence properties. J Biol Chem 260:3440-3450.

Harkins TT, Grissom CB (1994): Magnetic field effects on B 12 ethanolamine ammonia lyase: Evidence for a radical mechanism. Science 263:958-960.

Hojevik P, Sandblom J, Galt S, Hamnerius Y (1995): Ca2+ ion transport through patch-clamped cells exposed to magnetic fields. Bioelectromagnetics 16:33-40.

Hopkins CR, Farquhar MG (1973): Hormone secretion by cells dissocated from rat anterior pituitaries. J Cell Biol59:276-303.

Israel J-M, Vincent J-D (1990): The pituitary gland as excitable tissue. Frontiers Neuroendocrinol I I :339-362.

Karabakhtsian R, Broude N, S h a h N, Kochlatyi S, Goodman R, Henderson AS (1994): Calcium is necessary in the cell response to EM fields. FEBS Lett 349: 1-6.

Lewis DL, Goodman MB, St John PA, Barker JL (1988): Calcium currents and fura-2 signals in fluorescence-activated cell sorted lactotrophs and somatotrophs of rat anterior pituitary. Endo- crinology 123:611-621.

Liboff AR, Rozek RJ, Sherman ML, McLeod BR, Smith SD (1987): Ca2+-45 cyclotron resonance in human lymphocytes. J Bioelec- tricity 6: 13-22.

Lindstrom E, Lindstrom P, Berglund A, Mild KH, Lundgren E (1993): lntracellular calcium oscillations induced in a T-cell line by a weak 50 Hz magnetic field. J Cell Physiol 156:395-398.

Lledo P-M, Gutrineau N, Mollard P, Vincent J-D, Israel J-M (199 I ) : Physiological characterization of two functional states in subpopulations of prolactin cells from lactating rats. J Physiol 437:477494.

McCormack JG, Halestrap AP, Denton RM (1990): Role of calcium ions in regulation of mammalian intramitochondrial metabo- lism. Physiol Rev 70:391-425.

Meldolesi J, Madeddu L, Pozzan T (1990): Intracellular Ca” stor- age organelles in non-muscle cells: Heterogeneity and functional assignment. Biochim Biophys Acta 1055: 130-140.

Miller RJ (1992): Voltage-sensitive Ca2+ channels. J Biol Chem

Mollard P, GuCrineau N, Audin J, Dufy B (1 989): Measurement of Ca” transients using simultaneous dual-emission microspectrofluorimetry and electrophysiology in individual pituitary cells. Biochem Biophys Res Commun 164: 1045-1 052.

Mollard P, Theler J-M, GuCrineau N, Vacher P, Chiavaroli C, Schlegel W (1994): Cytosolic CaZ+ of excitable pituitary cells at resting potentials is controlled by steady state Ca2+ currents sensitive to dihydropyridines. J Biol Chem 269:25158-25 164.

Mosmann T ( 1 983): Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays. J Immunol Methods 65:55-63.

Phillips JL (1993): Effect of electromagnetic field exposure on gene transcription. J Cell Biochem 5 1 :38 1-388.

Pryzwanski KB, Martin LE, Spitznagel JK (1978): Immunocytochemi- cal localization of myeloperoxydase, lactoferrin, lysosyme and neutral proteases in human monocytes and neutrophilic granu- locytes. J Reticuloendothel SOC 24:295-309.

Rusovan A, Kanje M (1991): Stimulation of regeneration of the rat sciatic nerve by 50 Hz sinusoidal magnetic fields. Exp Neurol

Schlegel W, Winiger BP, Mollard P, Vacher P, Wuarin F. Zahnd ER, Wolleheim CB, Dufy B (1987): Oscillations in cytosolic Ca” of excitable pituitary cells due to action potentials. Nature

Siege1 S, Castellan NJ Jr (1988): “Non Parametric Statistics for the Behavioral Sciences, Statistics Series, 2nd Ed.” Singapore: McGraw-Hill International Editions, pp 51-55, 128-137, 144- 151.

Stanfield PR (1986): Voltage-dependent calcium channels of excitable membranes Br Med Bull 42:359-367.

Steiner UE, Ulrich T (1989): Magnetic field effects in chemical kinetics and related phenomena. Chem Rev 89:5 1-147.

Stojilkovic SS, Catt KJ ( I 992): Calcium oscillations in anterior pituitary cells. Endocrinol Rev 13:256-280.

Tada H, Shiho 0, Kuroshima KI, Koyama M, Tsukamoto K (1986): An improved colorimetric assay for 11-2. J Immunol Methods

Tse A, Hille B (1993): Role of voltage-gated Na+ and CaZ+ channels in gonadotropin-releasing hormone-induced membrane potential changes in identified rat gonadotropes. Endocrinology

Tsien RY, Rink TJ, Poenie M (1985): Measurement of cytosolic free Ca?’ in individual small cells using fluorescence microscopy with dual excitation wavelengths. Cell Calcium 6: 145-157.

Walleczek J (1992): Electromagnetic field effects on cells of the immune system, the role of calcium signaling. FASEB J

267: 1403-1406.

112:312-316.

329~7 19-72 1.

93: 157-1 65.

132: 1475-148 1.

6:3177-3185.

stimulation of ca2+ influx in rat pituitary cells under exposure to a 50 hz magnetic field

Documents