Biochbrdca et Bmphysica Actao ! 136 (1992) 231-238 ~ 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4889/92/$05/2,0

231

BBAMCR 13222

Growth of human cultured cells exposed to a non-homogeneous static magnetic field generated by Sm-Co magnets

Kei Sato ~, Hisao Yamaguch i a Hiroshi Miyamoto a and Yohsuke Kinouchi b

Depanmem of Physiolo~; School of Medicine, and ~ Depamnenl of Electric and Eleclronic Engineering, Faculo' of Engineering, Tiw Unh'ersity of Tokushima. Tok~lshima (lapan~

(Received 2 MaTch 1992)

Key words: Cell 8ro~lh; Cell morphology:, DNA synthesis; HeLa cell: Normal human giagival fibroblast: Static magnetic field

A slatic magnetic field, with a strong spatial gradient, was established on the surface of cell culture dishes by use of a gilded iron needle .set vertically above an Sm-Co magnet. The calculated magnetic flux density was more than 1.5 T at the center of the needle tip, and the products of the flux densi.~ and its gradient were about 200 and 60 T2/m at distances of 0.1 and 0.3 mm, respectively, from the cenler. The DNA content, DNA synthesis and labeling index of cultured cells located within 0.1 mm from the center of the needle, and the growth rate of cells located within 0.3 mm from the center, were measured. HeLa cells grew at a normal rate for % h in the magnetic field and showed no significant change in shape, deteclable by scanning electron microscopy. The gro'~h of HeLa cells was not influenced by exposure to the magnetic field. Similarly, exposure for 48 h to the magnetic field had no effect on growth of normal human gingival fibroblasls (Gin-l). The DNA content, assayed by microfluorometry Of the nuclei of both types of cells stained by the Feulgen reaction, was not significantly different from that of controls. Moreover, exposure to the magnetic field had no effect on DNA synthesis or the labeling index of HeLa ceils assayed by aumradiography of incoq~oratcd [3H]thymidinc, It is concluded that a non-homogeneous magnetic field of the intensity and the gradient used in Lhis study does not significantly influence the growth of HeLa cells or Gin-I cells.

]ntroductien

About a dozen papers describing the influences of strong magnetic fields on file functions and mor0hol- o ~ of various ceils have been published. Barly studies reveal marked changes in parameters of cell activity following exposure to a magnetic field. For example, there are reports showing inhibitions of growth of mouse leukocytes [1] and DNA synthesis of ascites sarcoma cells [2], whereas another study reveals an increase, rather than decrease, in the number of guinea-pig macrophages [3]. A recent paper [4] also shows inh~ition of cell growth and an increase in chromosomal aberrations of human lymphocytes. How- ever, a similar treatment does not induce a change in the growth rate or ceil morphology of WI 38 cells and human skin fibtoblasts [5]. Therefore, reported results are contradictory, and rftany recent studies indicate insignificant effects Of exposure, mostly to strong ho-

CoValence to: EL Miyamoto, Departmenl of Physiology, School of Me-~icine, The University of Tokushima. Kuramoto-che 3, Tokushima 770. Japan.

mogeneous magnetic fields, on cell growth, cell viabil- it*y, DNA synthesis and genetic properties [6-9].

Though the. interaction of individual diamagnetic biomolecules with :~, strong magnetic field is weak, several types of molecular assemblies, or living cr.lis in aqueous medium, exhibit significant orientation, based on diamagnetic anisotropy during exposure to static magnetic fields [10,11]. Erythrocytes, containing para- magnetic hemoglobin and flowing in a model vessel, are shown to be attracted towards the stronger mag- netic field, when exposed to a non-homogeneous mag- netic field with a steep gradient [12]. These observa- tions suggest that strong static magnetic fields, espe- cially those with large, strong gradients, might influ- ence cellular functions. With the recent development of medical instruments in which strong magnetic fields are applied, including NMR imaging apparatus and dental devices i13], human body or tissues will be exposed for relatively long times to not only strong homogeneous magnetic fields (0.2-2 T) but also to fields with steep spatial gradients. But the biological effects of such non-homogeneous fields are not yet fury elucidated, and studies are required to provide basic information for drawing safety guidelines for human tissues.

232

Previously, we designed a system in which cultured cells can be exposed to a magnetic field with cyclic changes in flux density and gradient [14]. With this system, ceils located in d i f fe rent places are exposed to

different magnetic densities and gradients of the field. In the presert study, we used a simpler system applica- ble to cultured cells, in this system, the magnetic flux is concentrated by an iron needle placed above a Sm-Co magnet, producing a non-homogeneous magnetic field of higher flux density and steeper spatial gradient in a narrow area near the needle than in the previous system. The effects of exposure of cultured HeLa cells and normal human gingival fibroblasts (Gin-l) to the field in the narrow area were examined by microscopy, micro-fluorometry and micro-autoradiography. No marked influences of the field on cell morphology, the ~owth rate, the DNA content or DNA synthesis were detected.

1 fJ1rl

CM ................. "t ....................... N ................. CS

Sin-Co Magnet

Fig. I. Diagram of tile ~3~-lem used for producing a static magnetic field with a large gradient of magnetic flux density on the surface of a cell culture. N, gc~ld-cxrated iron needle (i.04 mm diameter, approx. 20/zm tip diameter);, CS, culture surface; CM, culture medium. The

paris s h o ~ in the ~]tlStVal~Ofl are not exactly proportional in size.

Materials and Methods

Ce//cu/tur~ HeLa $3 cells, purchased from Flow Laboratories

VA, USA were cultured serially in culture flasks in modified Eagle's minimum essential medium (mMEM) [15] supplemented with 10% (v/v) calf serum. Normal human ginglval fibroblasts (Gin-l), purchased from the American Type Culture Collection (Rockville, MD, USA), were cultured serially in flasks in Dulbecco's modified medium supplemented with 10% (v/v) fetal bovine serum. These cultures were grown in a CO 2 incubator (LNA-121, Tabai, Tokyo, Japan) in a humidi- fied atmosphere of 5% CO 2 in a~. ~ a: 37~C; the pH of the media was adjusted to 7.2. The growing cultures were dispersed by treatment with 0.5% trypsin for 3 rain. Then the cells were suspended in the ~me cui- ture media at a density of 5-104 cells per ml, and 5 ml aliquots of the suspensions were inoculated into plastic culture dishes (6 cm diameter, No. 25010, Coming Glass Works, Coming, NY, USA).

Exposure to magnetic field After a few hours, the cultures were placed on

blocks (5 x 4.5 × i.5 cm) of a Sin-Co magnet, with a surface field strength of 0.2 T, as measured by a gauss meter (501, Nihon Denji Soldd, Tokyo, Japan) and grown in the CO 2 incubator for the periods indicated in the Results section. Iron needles (1.04 mm diame- ter) were coated with Au (30 ~tm thick at the middle and less than 5 ~tm thick at the tip of the needle) by an electrolytic method in a factory (Uemura Indust. Ltd., Takarnatsu, Japan). One of these needles was set verti- cal to the surface of the culture plate with its tip (aboat 20/zm diameter) touching the surface (Fig. 1). The position of th,. tip was clearly recognized as a round empty area on the culture surface (see Fig. 3). Gold-

coating of the needle was essential, because when an uncoated needle was used, the needle rusted in the media, cell growth was suppressed and ceils became necrotic around the needle. Therefore, we used only the gold-coated needles, previously confirmed to be rustless by placing in the culture medium for a week. For sham-exposure (control) cultures, the same sized Sin-Co blocks were not magnetized.

After exposure to the magnetic field for the indi- cated period (see Results), the culture dishes, with the attached needles, were removed from the magnets for microscopy, micro-photography, micro-fluorometry and micro-autoradiography.

Micro-photo~'aphy An inverted microscope (IMT, Olympus Optical In-

dust. Co., Tokyo, Japan), equipped with a reflex illumi- nation system in addition to the central illumination system, was used for morphological observation and micro-photography, which were carried out at magnifi- cations of 10 X 10 and 10 X 5, respectively. Reflex illu- mination was necessary for obse~ng cells located near the needle tip; these could not be examined by central illumination because of hindrance by the needle. The micro-photographs were finally enlarged to a size of 16.5 x 12 cm with photographic papers (FM3, Fuji, Tokyo, Japan) for counting cell numbers. After ebser- ration and photography of the cells, the dishes were returned to the CO 2 incubator for further exposure to the magnetic field. As time required for the micro- ~,¢opic observation and photography was less than 1 min, the removal from the incubator did not affect the cells. The total number of the ceils located within 0.3 nun of the needle tip was counted in photographs and expressed relative to the number at the start of expo- sure. Valu~ were given as means and S.D. values for five replicate cultures.

Micro-fluorometry The DNA content was determined by micro-fluo-

rometry (with a fluorescence microscope, modified FXA-RFL Nikon, Tokyo, Japan) of the nuclei stained by the Feu!gen reaction, using excitation and emission wavelengths of 546 nm and 580 nm, respectively. The mean content was expressed relative to the mean con- tent at the beginning of exposure. Measurements were made on the cells located within 0.1 mm of the needle tip and sampled without intentional selection.

Autoradiography Incorporation of tritiated thymidine into the nuclei

of HeLa cells was measured. For this purpose, the cukures were incubated with [6JH]thymidine (specific activity 185 GBq/mmol, A;ncrsham International, Bucks, UK), diluted to 74 kBq/ml with mMEM, for 30 rain and fixed overnight in a mixture of acetic acid and ethanol (1:3). The preparations were processed using an emulsion (Sakura NR-M2, Konishiroku Photo In- dust., Tolq¢o, Japan) and exposed for a week at 4°C. The rate of DNA synthesis was measured by counting the number of grains per nucleus. The fraction of the cell population in the S phase (labeling index) was e~ressed as the percentage of nuclei having more than five grains among the total number of nuclei examined. The ceils used for the autoradiographie assay were chosen from among those located within 0.1 mm of the needle tip and sampled ~thout conscious selection.

Electron microscopy The cultures were first fixed for 20 rain in a solution

of 15% glutalaldehyde in 0.1 M phosphate buffer (pH 7.3), and then were washed with 0.2 M phosphate buffer for 20 rain and post-fixed with 1% osmium in 0.1 M phosphate buffer for 20 min. The samples were dehydrated in an ethanol series of the concentrations of 70, 80, 90, 95 and 99%, with three changes of absolute ethanol. The ethanol was removed in a critical point dryer (HCP-2, Hitachi Ltd., Tokyo, Japan) and the cells were coated with a 100 mm thickness of Au-Pd alloy in a sputtering apparatus (E 5150, Polaron Equipment Ltd., Wafford, UK). The distribution and morphology of the cells were then examined in a scanning electron microscope (S-800, Hitachi Ltd., Tokyo, Japan) at magnifications of 100 x and 1000 x . Photographs were haken with a highly sensitive film (Trix Pan, Eastman Kodak Co, Rochester, NY, USA).

Finite element method The curve for the magnetic flux density B in Fig. 2

was drawn by the finite element method. In this calcu- latiort, the diameter and shape of the needle were taken into consideratien, We used Poison's equation based on the qrlindri~l coordinates for magnetic vec- tor potential, assuming a cylindrical domain (11 mm

233

radius, l I mm height). The tip of the iron needle was regarded to bc placed at the center of the circular base of the domain. The domain was divided into 472 trian- gle elements with 266 nodes. Concerning magnetic saturation characteristics of the needle, we determined the magnetic saturation density of 1.6 T and the rein. tive permeability of 300. The magnetic flux density measured on the culture surface without the needle, was introduced to set up the boundary condition: the flux density at 11 mm from the center of the needle tip was estimated to be 0.2 T. We also computed B- grad B using B.

Statistical analysis Significances of differences of values between con-

trol and exposed cells at the time points indicated in Tables I-IV were tested by a one-way layout of ANOVA. On the other hand, significances of differ- ences of the slopes of regression lines and those of variances of the cell numbers between control and exposed cells in Figs. 4A, B and C, were analyzed by t-test and F-test, respectively.

Results

Magnetic field strength The magnetic flux density (B), as a function o| the

distance from the center of the tip of the iron needle, is shown in Fig. 2 (see Materials and Methods). Exact measurement of the magnetic flux density distributed in the tiny area around the needle tip, where the cells

. 105

"" I I 104

° t / ; 9tad B 10 3 '~

o 0.5 I- \ " m

_ _ I . . . . - - i 0 0J}5 0.1 0.2 0.3 0.4 0.5 0.6

Distance (ram)

Fig. 2. Magnetic flux density ( 3 ) and the product of the density and its gradient (B-grad B) as functions of distance from the center of the needle tip. The curves were simulated by the finite element

method: solid curve, B; broken curve, B.grad B.

2.34

C D

235

A B C

"J21 5

.. 10 ~ ~ ~2 i 4 - g 8

:?, / ~ ' z z = 2 ¢ _ =~ ¢J ;;, o 1 o o

1 ~ t = 1 = 0 24 48 ; 2 96 0 6 1~) 1 ; 2J4 0 12 24 36

Tene (~ } Time (hr) Time(h0

Fig. 4. Effects of the non-homogeneous magnetic field on growlh of HeLa cells and human normal gingival fibroblasts (Gin-I) in relation to time. (A) Effect of exposure of HeLa cells to the field for 96 h; (B) effect of exposure of HeLa cells in the first 24 h: (C) effect of exposure of Gin-I cells in the first 48 h. Closed circles are data points for exposed cultures; open circles for co.trol cultures. Points in (A) and (C) are means of the number of attached cells in 5 control and 5 exposed cultures at indicated times, and bars are S.D.s. The number is expressed relative to the inoculated number of celLs. Regression lines in (A) and (C) were drawn by a least-squares program, in such a way that they pass the point of value I (Le, the inoculated number of ceils) in ordinate. Points in (B) are means of the cell numbers for f, control and 6 exposed cultures, and bars are S.D. values. The cell number is expressed relative to that of attached cells at the beginning of exposure (time 0). Regression lines in (B)

were dw, va by a least squares program using points except that at time 0. See also Materials and Methods.

examined were located, was technically impossible. The product of B and its gradient (B-grad B) was also calculated a~: a function of the distance.

The cells used to examine the effect of the field on the growth ra~e were located in the region between 10 mm and 0.3 ram from the center of the needle tip, implying that ~hese cells are exposed to a magnetic field of B from 1.1 to 0.3 T and of B-grad B from 4.104 to (30 T2/m. Sim,;!arly, the cells used for deter- mining the rate of DNA synthesis, the labeling index and the DNA content were in the field of B from 1.1 to about 0.35 T and of B. grad B from 4-104 to 200 T2/m. The HeLa cells and Gin-1 cells observed and photographed by scanning electron microscopy were located between 30 mm and 0.3 mm from the circum- ference of the needle tip.

Cell morphology Cell morphology was examined by scanning electron

microscopy (Fig. 3). After exposure to a magnetic field, neither type of cell showed any distinct change in shape or surface structure compared with controls. The cells were shown to grow without marked change in cell shape even in the immediate vicinity of the tip of

the needle, where the magnetic field density was high and its gradient was steep. One of the examples is shown in Fig. 3C.

Cell growth The effect of the magnetic field on cell growth is

shown in Fig. 4. In this experiment, HeLa and Gin-I ceils were exposed from 2 h after their inoculation. The number of HeLa cells increased exponentially for at least 96 h, and there was no significant difference in the cell numbers or slopes of the regression lines of exposed and control cells at any time (Fig. 4A). As sparse cultures, obtained ~,~o,-'tly after inoculation, are thought to be more sensitive to various extrinsic agents than dense cultures [16], we examined the effect of a magnetic field on young ":.t:ltures of HeLa cells more precisely (Fig. 4B). However, no significant effect of the field on cell growth was observed during the early period of the first 24 h after inoculation. The appar- ently rapid increase in the number of HeLa cells in the first 6 h was due to delayed attachment to the culture dishes of cells that were still fir: .ling in the medium at the start of the exposure. Sh~ilar experiments with Gin-I cells also indicated no siE,,titicant difference be-

Fig. 3. Morpholo~, of HeLa cells observed by a scanning electron microscope at magnifica'~ons of 100× and 1000×. (A) Unexposed (control) HeLa cells, 100×; (B) unexposed HeLa cells, 10O0x; (C) HeLa ceils exposed to a static magnetic field for 96 h, 100×; (D) the same cells as those in (C), 1000x. Arrows in (A) ~nd (C) indicate where the tips of the iron needles were located. See Fig. 2 for the magnetic flux density and

its gradient.

236

TABLE !

Effect of expasure to a r,~n.homogeneoa~ magnetic field on the DNA content of HeLa cells at various erpo~re times

The DNA content was determined on the Feulgen-staincd cells by micro-fluoromelry and is expressed relative In the c~onlent at the beginning of exposure. Values ate means_+S.D.s of the DNA con- tent for the cells chosen, without conscious selection, from 6 control and 6 exposed cultures, at the indicated times. There was no signifi- cant diffeter~e between the values for exposed and control cells ( P > 0.05) at any time.

Time (h) Rehtive to content at beginning of ~.4~osute

Cenlrol n exposed n

0 l 240 6 0.86 +__ 021 240 1.07 __+ 0.28 2,h3

12 0.87+-0.14 240 1.12+_0A6 240 18 1.01 _+ 0-~6 240 !.08 + 0.36 240 24 1.31 +-0.37 240 1.27+-0.30 240 48 0.89+_0.31 240 1.10±0.27 240

n, the number of celts measured.

tween exposed and contral cells after 48 h (Fig. 4 0 . Further extension of exposure time or increase in the number o|" sampling times was not possible, because of the limited number of the available Sin-Co magnets.

DNA content

Mean values for the nuclear DNA contents of HeLa and Gin-I cells at various times of exposure to a magnetic field are shown in Tables i and !l. The mean DNA contents of both types of cells did not show any marked variation with time during culture for up to 48 h, suggesting that these cells grew exponentially and were not synchronized. Moreover, the differences in the mean DNA contents of exposed and control cells at different exposure times were not significant. Fre-

TABLE !I

Effect of exposure to a nan-htmwgene~ magneti, field on the DNA content of normal human gingirul Ob]9[Jb~a~l~ I Gin~l) at radma expo-

The DNA content was determined on the Feulgen-staiued cells by micro-flumometty and is expressed relative to the conlent al Ihe beginning of expt~ute. In thi~ experiment, 5 control and 5 exposed cultures were used. S¢¢ Table I for d¢'ails. There was no significant difference bclween the values for expo~,:d and conlrol cells (P > 0.05) a! any time.

Time (h) Relative to content at beginning of exposure

conlrol u exposed n

0 ! 200 6 1.09_+0.17 200 126+03~

12 124_+ 0-28 200 1.21 +-0.31 !8 1,05__+0.21 200 0.87+-0.16 24 0.87__+ 022 20{} 0.96 +__ 0.24 48 1.03+026 200 !.18+_0J9

200 200 200 200 200

TABLE Ill

Effect of erpasure to a nwMwmogeneaus magnetic field on the number of grains per nudens of Hr.La cells due to lndse labeling i~ith I ~Hltl~din¢

Values are nlcans±S.D~s of Ihe number of grains per nucleus determined in the cells chosen, without inteqlional selection, from labeled cells wilh more than 5 grains per nucleus. The values were obtained [torn 2 control and 2 exposed cultures at the indicated limes. The~e was ~m significar, t difference hetween Ihe values for exposed and control cells (P > 0.05) al any lime.

Time (h) Number of grains per nucleus

comr~ r exposed n

0 16,8+_ 12 20 ,~ 17.4+0,9 20 17.5+_ 1,1 20

12 16.4_+ i . I 15 17.6+ 1,0 20 18 18.1 ± 1.8 ~ 17.6__+ 1.0 25 24 19.2+__ IA 25 !9.1 + I.! 48 16.5+ l.I ! ~; 17.6_+ 0,9

n, the numb~:t o| ~lls measured,

qucaL~ distlibutiorLs of the DNA contents also vmre examined, but no dctccta'olc differences were ~.oparcnt between control and exposed cells at any time of exposure (da ta no t shown).

D~A synd:esis and labelbtg btdex The effect of the magnetic field on the rate of DNA

synthesis in HeLa cells was examined by measuring the number of grains per nucleus after exposure of the cells to [3H]thymidine (Table ill). "[*he mean numbers of grains did not show marked time-dependent changes during 48 h, in either exlxr~d or control cells. More* over, the mean number was not significantly different from that of controls at any time during exposure to the field. A magnetic field has been reported to have a marked effect on electron emission from tritium in the nuclei of HeLa cells [17], but we did not observe an

"fABLE IV

Effect of eJ~,mw w a magnetic field on the percentage of labeled mtcl¢i amvng tile total nuclei of HeLa cells exon~wd (rite labelblg index) after pulse labeling widi [ 3Hbfovnidine

Values are means+_S.D.s of the lahelirg index of the cells chosen, wilhoul intentional seleclion, from 5 control and 5 exposed cultures at the indicaled times. There was no signifu:anl difference between Ibe values for exIxr~ed a'~d control cells (P > 0.05) al any lime.

Time (h) Labeling index (,c~)

control n exposed

0 57.0+_i1.1 70 6 66,4_+ 9,2 86 67.2_+ 10.1 81

12 695+_ 9.7 90 592_+ 8.9 89 18 64~8_+ 10.1 101 592+_ 8.9 102 24 71.2+_ 8.8 120 65.2_+ 92 131 48 62.7_+ IZI 167 02.9+_ 7.3 193

n, Ibe number of cells measured, n, the number of cells measured.

237

effect of the field on the grain count. This was proba- bly because samp!es were dipped into emulsion after their removal from the magnetic field, and so the field would not have affected emission from tritium during the autoradiographic process. The labeling indices of HeLa cells were determined from the sam,..~utoradio- graphs as those used in the above expcri~e,at (Table IV). No marked time-dependent chan~;e i~. the index, representing a change in the percentage of cells in the S-phase among the total cells, was observed in 48 h, and the magnetic field did not significantly affect the index.

Discussion

Theoretical considerations suggest that a stroJ~g magnetic field, of above 10 T, and a large gradient of magnetic flux density, might affect the cell membrane and distort the ordered process of cell division [18]. Thus, a strong non-homogeneous magnetic field might cause separation of biological macromolecules and so modify cell growth [19]. This would be caused mainly by Maxwell stress, which is proportional to the product of the field density and its gradient. In the present study, we devised a system in which the cultured cells were exposed to a non-homogeneous field with a stee" gradient (Figs. 1 and 2). Considering the cell volume, the cells should be affected by Maxwell stress, in addition to this direct stress on the cells, the field might affect charged particles in the cytoplasm and the culture medium. The diffusion ot" these particles could be affected by the Lorenz force and Maxwell stress. The Lorenz force might inhibit diffusion, but it would not really change diffusion of ,~ons and prn_tein molecules, because the threshold for such an effect is more than 10 4 T [20]. The threshold of the product of the magnetic flux density and its gradient to suppress the diffusion of these particles by Maxwell stress is calculated to be above l0 s T2/m [20]. Therefore, the Maxwell stress produced by the present device would not significantly after' the diffusion. However, these thresholds might be much smaller if the force or stress acts on living cells or ¢u'ganells because of their large volumes.

The effects of magnetic fields on enzyme activities and various parameters reflecting cell activities have been studied. Catalase and xanthine oxidase activities are not influenced by a magnetic fields of the intensity of I T [21]. In contrast, glutamate dehydrogenase activ- ity is reported to be inhibited, while catalase activity is increased; the effect is greater in non-homogeneous strong magnetic fields of 6-7.8 T than in homogeneous fields of similar intensities [22]. Brief exposure of mouse mammary adenoearck, oma cells to a strong field (3.8 T) with a gradient (1.2 T/m) is found to cause inhibi- tion of cell grov,~h and cell viability. [23]. Conversely,

increases in the growth rates of rabbit myocardium and mouse lung fibroblasts in serum-free medium are ob- served in a ~tatic magnetic field of 1.46 T with a 50 T/m gradient [24i. Non-homogeneous magnetic field~ are reported to affect :he de~'elopment of frog eggs [25], whereas recent studies point out that a homoge- neous field of 6.3 T [261 and non-homogeneous fields with gradients from 10 T /m to 10 ~ T /m do nox influence early development of Xeno~s iaeris [27]. Other investigators have detected no significant effects of non-homogeneous fields on the viability or growth of mouse mammary tumor cells [28] or the viability of Chinese hamster lung cells [29].

We also did not observe any significant influence of a non-homogeneous field with a steep gradient on the growth or DNA synthesis and/or the DNA content of cultured HeLa cells and human gingival fibroblasts. The percentage of HeLa cells in the S-phase among the total cells was also not affected by the magnetic field. These results ind[eale that the field has no effect on either I3NA synthesis or on the duration of the S-phase or other phases of the cell cycle of these cells.

HeLa ce!ls have been reported Io show no change in morphology or colony formation following exposure to a steady field of 0.5 T for 10 days [30], or in cell growth following exposure to 0.12 T for 7 days [31]. There is no previous report on the effect of a magnet!c field on human gingival fibroblasts.

We conclude from this work that the gro¢,¢h of cultured human HeLa and gingival cells are not af- fected by a non-homogeneous magnetic field with a variation in the flux density from 1.1 to 0.3 T and a gradient ranging from 4-104 to 60 TZ/m.

Acknowledgements

We thank Dr. T.S. Tenforde of Life Sciences Cen.. ter, Battelle Pacific Northwest Laboratories for helpfid discussion. We also thank Mr. T. Masuya for cultufiug the ceils, along with Mr. M. Fujimoto and Mr. M. Shono for technical assistance in electron microscopy and micro-fluorometry, respectively. This study was supported in part by a Grant-in-Aid on a Prioriiy Area given from the Japanese Ministry of Education, Sci- ence and Culture and by a grant from Nissan Founda- tion for Promoting Scientific Research. Sm-CO magnets were donated by Hitachi Metal lndust. Co. Ltd.

References

1 Bamothy, J.M., Barnozhy, M.F. and Boszormenyi-Nagy, I. (1956) Nature 177, 577-578.

2 D'Souza, L., Reno, V.R., Nutini, L.G. and Cook, E.S. 0969) in Biological Effects of Magnetic Fields (Bamothy, M.E, Ed.}, Vol. 2, pp. 53-59, Plenum Press, New York,

3 Valen'inuzzi. M., Ferrarefi, R,W. and Vazquez, I". 0966) F.xl~ri- entia 22, 312.

238

4 A~d~to, G., Lamberti, L, Bigatti, P. and Prono, G. (1~84) Boll Soc. IL Biol. Sper. 9, 1341-1346.

5 Haigis, E., Gitzelmann, It. and OtL H. (1981) Experientia 37, 255.

6 Wohi, S. Crooks. LE.. Brown, P.. Howard, R~ and Painter, R.B. (1980) Radiology 136, 707-710.

7 Cooke, P. and MmTix, PC;. (1981) Br. J, Radiol. 54, 622-6Z5. 8 Schwartz, J.L and Crooks, L E (1982) Am. J. Roent. 139, 583-

585. 9 Ngo, F.O.I-L Blue, J.W. and Roberts, W.K. (1987) Mag- Reson.

Med. 5, 307-3!7. 10 Tenforde. T.S. and L~urdy, R.P. (1986) Proc. 8th Annu. Conf.

IKFF Eng. MecL Biol. Soc. 1381-1384. 11 Y a m a ~ A. (1990).I. Mag. Mag. Mat. 90 and 91, 43-46. 12 Okazaki, M., Kon, K., Macda, r~. and Shiga, T. (1988} Physiol.

Clu:m. Phys. Me'.'- NMR ~, 3-14. 13 Tsulsui, I~ Kiam~hi, Y, Sasaki, H, Shiota, M and Ushila, T.

(1979) J. i)cnL Rcs. 58,1597-1606. 14 ~ , Y~ Ushi~a, T., Sato, K., Miyamoto, I-L Yamaguchi, H.

and Yo~la, Y. (1984) Bioelectromagnetics 5, 399-410. 15 Mi3~mo~o, H, Rasmussen, L. and Zeuthen, E. (1976) in Methods

in Cell Biology (Pr:~x~tl, D.M~ Ed.), Vol. 13, pp. 15-27, Aca- ~mic Press, New York.

16 Yor~zu, T, Yamaguc~i, H, l-iosokawa, IC, Ikehara. T., Taka- has~ A. (1992) Cell MoL Biol. 38, 217-224.

17 Harford, C.G. and Hamlio, A. (1961) Lab. Invest. 10, 627-635. 18 Winterbe~g, F. (1967) Arch. Biochem. Biophys. 122, 594-598. 19 Marel, Go and l~ransfeld, I~ (1985) Top. Appl. Phys. 57~ 143-204. 20 ~ L Y-, Tanimolo. S, Ushita, 1", Sato, K., Yamaguchi, H.

and Miyam~o, FL (1988) B~ctc~trom~gnelics 9, 159-166. 21 Ueno, S. and Harnda, K. (1986) IEEE Trans. Mag. 2C, 868-873. 22 ttaberditzl, W. (1967) Nature ~13, 72-73. 23 Weber, T. and Caxilli, GJ. (1971) Cancer ~ , 340-343. 24 Pumper, R.Wo and Bamolhy. J.M. (1969) in Bamothy, M.F,

(Ed.), Bioto~cal Effects of Magnetic FieJds Vol. 2, pp. 61-65, Plenum Press, New YorL~

25 lxvcngood, W.C. (1969) J. Emb~ol. Exp. MorphoL 21, 23-31. 26 Ueno, S, Sh~okawa. K. and Iwamoto, M. (1990)£ Appl. Phys. 67,

5841-5843. 27 Ueno, S~ Ha~da, K. and Shiokawa, K. (1984) IEEE Trans. Mag.

20, 1663-1665. 28 Rockwell, S. (1977) inL J. Rad. Biol. 3, 153-160. 29 Nath, R-, Schulz, RJ. and I~n,~omi, P. (1980) Int. J. Rad, Biol.

38, 285-292. 30 HaIL EA, Bedford. J.S. and Lea~k, MJ.H. (1964) Nature 203,

1086-1087. 31 Halpem, M !L and G,-cr~te, A.E (1964) Nature 202, 717.