Doi:10.1016/j.bbrc.2005.08.243

Biochemical and Biophysical Research Communications 336 (2005) 1144–1149 Induction of tamoxifen resistance in breast cancer cells Rainer Girgert a,*, Hartmut Schimming b, Wolfgang Ko¨rner c, Carsten Gru¨ndker a, a Department of Obstetrics and Gynecology, University of Go¨ttingen, D-37099 Go¨ttingen, Germany b Facility for Electronic Equipment, University of Ulm, D-89075 Ulm, Germany c Bavarian Environmental Protection Agency, D-86179 Augsburg, Germany The incidence of breast cancer in western societies has been rising ever since the Second World War. Besides the exposure to a mul- titude of new chemical compounds, electromagnetic field exposure has been linked to breast cancer through a radiation-mediated anti-melatonin pathway. We investigated, whether low-frequency electromagnetic field exposure interferes with the anti-estrogenic activity oftamoxifen. Two different clones of the breast cancer cell line MCF-7 were exposed to highly homogeneous 50 Hz electromagnetic fieldsand IC50 values were calculated from dose–response curves of tamoxifen at various field intensities. An intensity-dependent shift oftamoxifen dose–response curves to higher concentrations with a maximal response at 1.2 lT was observed. Hypothetically, electromag-netic field exposure could contribute to tamoxifen resistance observed in breast cancer after long-term treatment.
Ó 2005 Elsevier Inc. All rights reserved.
Keywords: Breast cancer; Estrogen receptor; Electromagnetic fields; Tamoxifen resistance; Dose–response The effect of extremely low-frequency electromagnetic suffer of a high rate of mammary tumors if treated with the field (ELF/EMF) exposure on human health has been widely debated. A number of epidemiological studies have (DMBA). Exposure of these rats to a 100 lT electromag- pointed to a slight increase in malignant diseases in popu- netic field for 27 weeks increased the number of tumor lations exposed to electromagnetic fields through the vicin- bearing rats to 65% compared to 50% in sham exposed rats ity of power lines. A significant positive association was . Although radiation energy of an extremely low-fre- observed between childhood leukemia and exposure of quency magnetic field (50 Hz) is considered to be by far children to magnetic fields during the night In two stud- too low to induce DNA strand breaks, Lai and Singh ies, premenopausal women exposed to environmental fields observed an increase in DNA single- and double-strand stronger than 0.2 lT had an increased risk of breast cancer breaks in brain cells of rats exposed to electromagnetic (BC) Conversely, studies from Finland and Taiwan fields as low as 10 lT. This effect was attributed to the gen- did not find any increased BC risk in populations living eration of oxygen radicals in the presence of iron ions .
in the proximity (100–500 m) of power lines In addition, EMF was reported to suppress the nocturnal These epidemiological observations prompted the exam- synthesis of melatonin in the pineal gland in animals and ination of the impact of electromagnetic fields on breast human As melatonin may physiologically inhibit cancer incidence in an animal model. Sprague–Dawley rats estrogen production by the ovary, the EMF-suppressedmelatonin secretion would favor the growth of estrogen-de- pendent BC A direct oncostatic effect of melatonin on Corresponding author. Fax: +49 5513912784.
E-mail address: (R. Girgert).
breast cancer cells was first demonstrated by Blask and Hill 0006-291X/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.bbrc.2005.08.243 R. Girgert et al. / Biochemical and Biophysical Research Communications 336 (2005) 1144–1149 coiled around energized with an anti-parallel current, so that the net applied Influence of low-frequency magnetic fields on tumor cell gene expression static magnetic field by the heating coil is annulled. On top of the heating coila second layer of copper wire is coiled around and connected to a signal generator delivering a 50 Hz sinusoidal alternating current. The current inducing ELF/EMF is regulated by electronic feedback stabilizing the chosen field intensity. Feedback signals are generated by a Hall sensor measuring the field intensity in the incubatorÕs center. Due to dimensions of the field inducing coil, homogeneity of induced magnetic fields in a centralspace harboring the culture plates varies by less than ±5%. The temperatureinside the incubator is measured by a thermistor probe regulating the current and many other investigators thereafter. Melatonin re- to the bifilar heating coil. CO2 chamber concentration is kept at 5.0 ± 0.1% duced the growth of the estrogen receptor positive breast by an infrared sensor (Vaisala, Vanha, Finland) that regulates CO2 influx cancer cell line MCF-7 in vitro by 18–27%. When the cells through a magnetic valve placed at a distance of more than 1 m outside the were exposed to a 60 Hz electromagnetic field of 1.2 lT flux Proliferation assay. Five hundred cells per well were plated into 96-well density, this inhibitory effect of melatonin was completely plates (Falcon, Heidelberg) in 100 ll DMEM/5% fetal calf serum (FCS, blocked This surprising observation has been indepen- Biochrom, Berlin) without phenol red, 2 mM glutamine, 50 U/ml peni- dently replicated by several other authors Since the cillin/streptomycin, 2.5 lg/ml amphotericin B, and 1:100 non-essential oncostatic effect of melatonin was estrogen-dependent, amino acids. After cell attachment, 100 ll medium or 100 ll tamoxifen Harland et al. tested whether the growth-inhibitory effect solution at increasing final concentrations of 10À8–5 · 10À6 M was addedto the wells in six replicates. Cells were exposed to magnetic field inten- of the estrogen receptor (ER) antagonist tamoxifen was sities of 0, 0.2, 1.2, 10 or 100 lT, respectively, for seven days at 37 °C, 5% modulated by ELF/EMF exposure. Using the same exper- CO2. Cell number was determined by a colorimetric assay using Alamar imental set-up a reduced growth inhibition by tamoxifen Blue (Biosource, Solingen, Germany). The optical density (OD) of the on MCF-7 cells was observed at 1.2 lT These results reduced dye is assessed at 570 nm vs 630 nm after 4 h at 37 °C.
were also reproducible by other laboratories Calculation of dose–response curves. Means and standard deviations of the OD of six replicates were calculated. The proliferative effect (PE) at The reduced tamoxifen activity in the presence of elec- each tamoxifen concentration was determined tromagnetic fields appears similar to a phenomenon, called‘‘tamoxifen resistance.’’ Tamoxifen has been used for treat- ment of ER positive BC for nearly thirty years. While most Dose–response curves for tamoxifen were obtained for each field exposure patients with advanced estrogen-responsive BC initially condition by plotting the mean PE of all experiments versus the concentra- profit from tamoxifen treatment, most of their tumors re- tion of tamoxifen on a half-logarithmic scale.
cur and respond no longer to tamoxifen treatment For calculating EC50 values of growth stimulation and the IC50 values Numerous investigations on EMF-regulated gene expres- for growth inhibition by tamoxifen, dose–response curves were split intotwo ranges, one, at lower concentrations (10À8–10À7 M) where tamoxifen sion in tumor cells yielded controversial results agonistically stimulated the growth of the MCF-7 cells, and the other The authors employed different cellular systems and ranging from 10À7 to 5 · 10À6 M where tamoxifen inhibited the cell exposure conditions making comparisons between reported growth in an anti-estrogenic manner. Calculations of EC50 and IC50 were results difficult. However, there is agreement in the necessi- performed using a VBA program for EXCEL 5 written by Josef Greve at ty of further investigations. In order to minimize uncon- the Fraunhofer Institute for Molecular Biology and Applied Ecology,Schmallenberg, Germany trolled external interferences with exposure conditions,particular caution must be paid to the generation of a sta-ble and reproducible magnetic field.
For the analysis of EMF-induced modulation of tamox- ifen activity we developed and constructed a novel incuba- Influence of EMF on the anti-proliferative effect of tor for the reproducible exposure of cells to defined ELF/ EMF. Maximum effort was employed to achieve highlyhomogeneous sinusoidal fields and control of exposure Dose–response curves of tamoxifen were calculated for two different subclones (MCF-7 p40 and MCF-7 p181)and compared at various field intensities.
The tamoxifen dose–response curves in either a shielded configuration excluding surrounding environmental fields Cell culture. The human BC cell line MCF-7 was obtained from ATCC (0 lT), at the ambient (%0.2 lT) field, and at sinusoidal (Manassas, USA). A second MCF-7 clone (MCF-7 p181) was provided by artificial fields of 1.2 and 100 lT intensity are shown in Dr. W. Ko¨rner, Augsburg. Cells were maintained in DulbeccoÕs modified . The results of the measurements at 10 lT are not MEM supplemented with 5% fetal calf serum (Biochrom, Berlin), 2 mMglutamine, 50 U/ml penicillin/streptomycin, 2.5 lg/ml amphotericin B, included in for a better clarity but the calculations and 1:100 non-essential amino acids (Biochrom, Berlin, Germany).
for IC50 of tamoxifen are listed in .
Exposure of cells to electromagnetic fields. We exposed MCF-7 cells to The dose–response curves of tamoxifen differ clearly in various field intensities (0, 0.2, 1.2, 10, and 100 lT) of a synthetic sinusoidal the two MCF-7 subclones examined (). The dose–re- 50 Hz alternating electromagnetic field. Exposure- incubators with sinu- sponse curves of clone MCF-7 p40 (A) show an soidal current generator/regulators and separated CO2 blenders consistedeach of a copper tube, 30 cm in diameter and 75 cm in length, closed at either inhibitory effect of tamoxifen on the growth of the BC cells end by heat accumulating copper plates. For heating, a bifilar copper wire is at concentrations >10À7 M. In the absence of any alternat- R. Girgert et al. / Biochemical and Biophysical Research Communications 336 (2005) 1144–1149 would be exposed to an average ambient magnetic flux den-sity of %0.2 lT present in the laboratory. The dose–re-sponse curve of tamoxifen at 0.2 lT resembled the one recorded under shielded conditions. At 1.2 lT the dose–re-sponse curve is slightly shifted to the right, resulting in an IC50 value of 2.3 · 10À6 M. At a substantially higher field shielded
intensity of 100 lT this shift of the dose–response curve is no longer observed and the IC50 of tamoxifen is reducedto about 0.9 · 10À6 M ( In the cell clone MCF-7 p181, the described effects of magnetic fields on the dose–response curves of tamoxifen In the shielded situation (0 lT), the dose–response curve of tamoxifen in MCF-7 p181 cells showed a similar sigmoi- dal pattern as the one seen with the p40 clone. Even weak ambient flux densities (0.2 lT) resulted already in a markedproliferative activity of tamoxifen at concentrations The maximal proliferative gain in MCF-7-p181 cells at 0.2 lT and a tamoxifen concentration of 10À7 M was 26% compared to the absence of tamoxifen.
Already at 0.2 lT the dose–response curve of tamoxifen was clearly shifted to higher concentrations. This shift waseven more pronounced at a field intensity of 1.2 lT. The maximal proliferative effect of tamoxifen at 1.2 lT was ob- shielded
served at a concentration close to 1 lM. At higher field intensities (10 and 100 lT) the shift of the dose–response curve was lower as compared to 1.2 lT, but did not return to the values measured in the absence of the EMF These measurements clearly show a ‘‘window effect’’ of the applied EMF with a maximum between 1.2 and10 lT as has also been observed in other biological systems Fig. 1. Dose–response curves of tamoxifen at various intensities of 50 Hz electromagnetic fields. (A) Clone MCF-7 p40. (B) Clone MCF-7 p181.
Cells were grown at increasing concentrations of tamoxifen either in ashielded configuration (0 lT) (closed circle) or at 0.2 lT (open square) or Calculation of IC50- and EC50 values of tamoxifen at at 1.2 lT (upright triangle) or 100 lT (diamond). Cell number was estimated after 7 days of culture by a colorimetric assay. Control: cellnumber achieved in the absence of tamoxifen = 100%. Means of at least The dose–response curves of tamoxifen in MCF-7 three independent experiments with six replicates at each concentration.
(p181) cells (were separated into a proliferativebranch (10À8–10À7 M) and an anti-proliferative branch ing EMF (ambient field shielded by a container of mu-met- (10À7–5 · 10À6 M) and EC50 values of the proliferative ef- al), the IC50 of tamoxifen was calculated at 1.4 · 10À6 M. If fect at low tamoxifen concentrations (clone p181 only) this mu-metal shielding were omitted, cells in culture plates and the IC50 values of the anti-proliferative effect of tamox- Table 2IC50- and EC50 values of tamoxifen at various field intensities IC50: tamoxifen concentration for half-maximal growth inhibition.
EC50: tamoxifen concentration for half-maximal growth stimulation.
R. Girgert et al. / Biochemical and Biophysical Research Communications 336 (2005) 1144–1149 ifen at high concentrations were calculated from the sepa- selection process favoring cells in the tumor that are al- rate dose response curves for all applied field intensities ready sensitized to growth stimulation by tamoxifen or are at least insensitive to the growth inhibition or to cellu- In the shielded configuration, clone p181 was double lar alterations induced by the drug or other environmental as sensitive to the inhibitory effect of tamoxifen as clone factors. Wiseman et al. observed a sensitization of p40 If p181 cells were exposed to the ambient tumor cells to the proliferative activity of IGF-I after treat- EMF of about 0.2 lT, a threefold higher tamoxifen con- ment with tamoxifen. Tamoxifen treatment would select centration was needed to achieve 50% growth inhibition for these IGF-1-dependent cells ultimately producing a as compared to the shielded situation. In cells of clone p40, sensitivity to tamoxifen was only slightly reduced The modulated tamoxifen effects that we observed in at 0.2 lT. A strong shift in the IC50 occurred in both cell clones at 1.2 lT and similarly high concentrations of incompatible with a selection process because the time tamoxifen were needed for a half-maximal inhibition at a of exposure was too short to allow a hypothetically magnetic field of 10 lT. Surprisingly, at 100 lT the effect tamoxifen-stimulated or at least tam-insensitive subpopu- on tamoxifen inhibition was clearly lower than at 10 lT.
lation to overgrow the majority of tamoxifen-sensitive From the data in it can be seen that ELF/EMF clearly reduce the growth-inhibitory effect of tamoxifen One further hypothesis for the development of tamoxi- with a maximum efficacy between 1.2 and 10 lT, and that fen resistance in breast tumors suggests that this resistance this effect is waning at higher field intensities.
is associated with an inappropriate expression of receptor A marked estrogen-like proliferative effect at low tamox- interacting proteins (RIPs) A multitude of receptor ifen concentrations was only observed in clone p181 in the interacting proteins (RIPs) regulate gene transcription by presence of EMF. The proliferative EC50 is reduced with nuclear hormone receptors, e.g., ER, for review, see increasing field intensities, reaching its strongest effect at In a preliminary clinical study, high levels of SRC-1 were detected in breast tumors showing good response to tamox-ifen treatment In a comparison of the expression of various RIPs in wild type MCF-7 breast cancer cells and MCF-7/TAMR- Here we show that the anti-estrogenic activity of tamox- 1 cells that acquired a tamoxifen resistant phenotype ifen is reduced in two subclones of MCF-7 cells under the after permanent treatment with tamoxifen revealed no influence of ELF/EMF to different extent. Dose–response differences in the expression of TIF-1, SUG-1, and curves of the growth-inhibitory effect of tamoxifen are SMRT but RIP140 expression was lower in non-stimu- shifted towards higher concentrations leading to a reduced lated cells of the resistant strain. Stimulation of the growth inhibition at a given concentration. Our observa- resistant cells by E2 or tamoxifen increased the level tion confirms results from a previous report describing a re- of RIP140 mRNA but not in the parental MCF-7 cells duced inhibitory effect of tamoxifen at 10À7 M from 40% to only 17% by exposure to an EMF of 1.2 lT . More rel- When expression levels of the corepressor N-CoR are evant from a therapeutic point of view, in our experiments low, patients receiving tamoxifen therapy experience poor tamoxifen even enhanced growth of the MCF-7 cells at outcomes. This observation suggests that tamoxifen antag- concentrations below 10À6 M if cells were exposed to onism requires high levels of N-CoR function .
EMF. The behavior of breast cancer cells exposed to Tamoxifen can act as an agonist through ERa/ERb het- EMF appears similar to the frequently observed tamoxifen erodimers, thus, in breast cancer cells where sufficient con- resistance in tamoxifen-treated patients.
centrations of ERa and ERb are present, tamoxifen could About 40% of ER-positive breast tumors fail to respond induce cell proliferation An imbalance of ERa- and to anti-estrogen therapy by tamoxifen from the beginning ERb-expression may determine a breast tumor to become (intrinsic resistance), while most of the residual tumors that initially respond to tamoxifen develop resistant relapse in Exposure to ELF/EMF is omnipresent in our electrified the course of treatment (acquired resistance—AR) environment but the strength of the EMF generated by the Tamoxifen is known as a partial estrogen antagonist be- electric wiring in usual households varies between 0.01 and cause it can either stimulate or inhibit ER-dependent 1 lT, in occupational situations exposure values of 1 lT tumor growth in a tissue-, cell-, and promoter-specific man- and more are occasionally achieved At 1.2 lT the ner. Like other selective estrogenic response modifiers enhancing/augmenting influence of ELF/EMF on the pro- (SERMs) tamoxifen acts estrogen antagonistic in certain liferative effect of tamoxifen is strongest and is surprisingly tissues, e.g., breast tissue, and agonistic in other tissues like waning at higher field intensities. Such kind of ‘‘window ef- bone and uterus . Resistant tumors behave like tissues fect’’ of EMF activity has also been observed in other where tamoxifen acts as an estrogen agonist.
Several mechanisms have been hypothesized as to how In the clinical situation where BC is frequently treated AR to tamoxifen could arise. AR may be due either to a with tamoxifen, it could be speculated that EMF exposure R. Girgert et al. / Biochemical and Biophysical Research Communications 336 (2005) 1144–1149 may also contribute to the induction of a tamoxifen-resis- [11] R.P. Liburdy, T.R. Sloma, R. Sokolic, P. Yaswen, ELF magnetic tance-like behavior in some breast tumors.
fields, breast cancer, and melatonin: 60 Hz fields block melatoninÕsoncostatic action on ER+ breast cancer cell proliferation, J. Pineal From a medical point of view it is disturbing that max- imal induction of cell proliferation by tamoxifen at a field [12] C.F. Blackman, S.G. Benane, D.E. House, The influence of strength of 1.2 lT is observed at a concentration of 1.2 microT, 60 Hz magnetic fields on melatonin- and tamoxifen- 10À6 M. This is exactly the serum concentration achieved induced inhibition of MCF-7 cell growth, Bioelectromagnetics 22 in BC patients under standard oral therapy . Given [13] M. Ishido, H. Nitta, M. Kabuto, Magnetic fields (MF) of 50 Hz the great number of BC patients under long-term oral at 1.2 microT as well as 100 microT cause uncoupling of inhib- tamoxifen treatment and more so in the light that in Octo- itory pathways of adenylyl cyclase mediated by melatonin 1a ber 1998 the US Food and Drug Administration (FDA) receptor in MF-sensitive MCF-7 cells, Carcinogenesis 22 (2001) approved the use of tamoxifen to reduce the incidence of breast cancer in healthy women at increased risk of the dis- [14] J.D. Harland, R.P. Liburdy, Environmental magnetic fields inhibit the antiproliferative action of tamoxifen and melatonin in a human ease, clearly more research efforts are warranted to exclude breast cancer cell line, Bioelectromagnetics 18 (1997) 555–562.
the fact that EMF exposure could induce breast epithelial [15] I.A. Jaiyesimi, A.U. Buzdar, D.A. Decker, G.N. Hortobagyi, Use of proliferation in tamoxifen users. Such research is continu- tamoxifen for breast cancer: 28 years later, J. Clin. Oncol. 13 (1995) ingly being supported by the German Radiation Protection [16] R. Girgert, C. Bartsch, S.M. Hill, R. Kreienberg, V. Hanf, Tracking the elusive antineoplastic effect of melatonin: a new methodological Our results confirming earlier reports on the modulation approach, Neuroendocrinol. Lett. 24 (2003) 433–437.
of tamoxifen activity through exposure of BC cells to ELF/ [17] T.A. Litovitz, C.J. Montrose, W. Wang, Dose–response implications EMF suggest that clones of MCF-7 cells are suitable mod- of the transient nature of electromagnetic field induced bioeffects, els to study cellular changes associated with the induction Bioelectromagnetics Suppl. 1 (1992) 237–246.
[18] R.E. Curtis, J. Boice, D.A. Shriner, B.F. Hankey, J.F. Fraumeni, Second cancers after adjuvant tamoxifen therapy for breast cancer, J.
Natl. Cancer Inst. 88 (1996) 832–834.
[19] L.R. Wiseman, M.D. Johnson, A.E. Wakeling, A.E. Lykkesfeldt, F.E. May, B.R. Westley, Type I IGF receptor and acquired tamoxifen This work was supported by Grant StSch4219 of the resistance in oestrogen-responsive human breast cancer cells, Eur. J.
Federal Ministry for the Environment, Nature Conserva- [20] C.M.W. Chan, A.E. Lykkesfeldt, M.G. Parker, M. Dowsett, Expression of nuclear receptor interacting proteins TIF-1, SUG-1,receptor tamoxifen-resistant breast cancer, Clin. Cancer Res. 5 (1999)3460–3467.
[1] J. Schu¨z, J.P. Grigat, K. Brinkmann, J. Michaelis, Residential [21] N.J. McKenna, R.B. Lanz, B.W. OÕMalley, Nuclear receptor coreg- magnetic fields as a risk factor for childhood acute leukaemia: results ulators: cellular and molecular biology, Endocr. Rev. 20 (1999) 321– from a German population-based case–control study, Int. J. Cancer [22] E.M.J.J. Berns, I.L. van Staveren, J.G.M. Klijn, J.A. Foekens, [2] N. Wertheimer, E. Leeper, Adult cancer related to electrical wires Predictive value of SRC-1 for tamoxifen response of recurrent breast near the home, Int. J. Epidemiol. 11 (1982) 345–355.
cancer, Breast Cancer Res. Treat. 48 (1998) 87–92.
[3] M. Feychting, U. Forssen, L.E. Rutqvist, A. Ahlbom, Magnetic fields [23] I. Girault, F. Lerebours, S. Amarir, Expression analysis of estrogen and breast cancer in Swedish adults residing near high-voltage power receptor alpha coregulators in breast carcinoma: evidence that N- lines, Epidemiology 9 (1998) 392–397.
CoR 1 expression is predictive of the response to tamoxifen, Clin.
[4] P.K. Verkasalo, E. Pukkala, J. Kaprio, K.V. Heikkila, M. Kos- kenvuo, Magnetic fields of high voltage power lines and risk of cancer [24] V. Speirs, A.T. Parkes, M.J. Kerin, D.S. Walton, P.J. Carelton, J.N.
in Finnish adults: nationwide cohort study, BMJ 313 (1996) 1047– Fox, S.L. Atkin, Coexpression of estrogen receptor a and b: poor prognostic factors in human breast cancer, Cancer Res. 59 (1999) [5] C.Y. Li, G. Theriault, R.S. Lin, Residential exposure to 60-Hertz magnetic fields and adult cancers in Taiwan, Epidemiology 8 (1997) [25] S. Davis, W.T. Kaune, D.K. Mirick, C. Chen, R.G. Stevens, Residential magnetic fields, light-at-night, and nocturnal urinary 6- [6] S. Thun-Battersby, M. Mevissen, W. Loscher, Exposure of Sprague– sulfatoxymelatonin concentration in women, Am. J. Epidemiol. 154 Dawley rats to a 50-Hertz, 100-microTesla magnetic field for 27 weeks facilitates mammary tumorigenesis in the 7,12-dimethylbenz[a]-an- [26] A. De Cupis, R.E. Favoni, Oestrogen/growth factor cross-talk in thracene model of breast cancer, Cancer Res. 59 (1999) 3627–3633.
breast carcinoma: a specific target for novel antioestrogens, TIPS 18 [7] H. Lai, N.P. Singh, Magnetic field induced DNA strand breaks in brain cells of the rat, Environ. Health Perspect. 112 (2004) 687–694.
[27] R. Goodman, L.X. Wei, J. Bumann, A. Henderson, Exposure to [8] R.G. Stevens, S. Davis, The melatonin hypothesis: electric power and electric and magnetic (EM) fields increases transcripts in HL-60 cells: breast cancer, Environ. Health Perspect. 104 (Suppl. 1) (1996) 135– does adaptation to EM fields occur? Bioelectrochem. Bioenerg. 29 [9] W.S. Baldwin, J.C. Barrett, Melatonin: receptor-mediated events that [28] R.D. Owen, MYC mRNA abundance is unchanged in subcultures of may affect breast and other steroid hormone-dependent cancers, Mol.
HL60 cells exposed to power-line frequency magnetic fields, Radiat.
[10] D.E. Blask, S.M. Hill, Effects of melatonin on cancer: studies on [29] L.I. Loberg, J.R. Gauger, J.L. Buthod, W.R. Engdahl, D.L.
MCF-7 human breast cancer cells in culture, J. Neural Transm.
McCormick, Gene expression in human breast epithelial cells exposed to 60 Hz magnetic fields, Carcinogenesis 20 (1999) 1633–1636.
R. Girgert et al. / Biochemical and Biophysical Research Communications 336 (2005) 1144–1149 [30] M. Mevissen, M. Kietzmann, W. Lo¨scher, In vivo exposure of rats to [33] A. DiCarlo, N. White, F. Guo, P. Garrett, T. Litovitz, Chronic a weak alternating magnetic field increases ornithine decarboxylase electromagnetic field exposure decreases HSP70 levels and lowers activity in the mammary gland by a similar extent as the carcinogen cytoprotection, J. Cell. Biochem. 84 (2002) 447–454.
DMBA, Cancer Lett. 90 (1995) 207–214.
[34] B. Shi, B. Farboud, R. Nuccitelli, R.R. Isseroff, Power-line frequency [31] J.M. Mullins, L.M. Penafiel, J. Juutilainen, T.A. Litovitz, Dose– electromagnetic fields do not induce changes in phosphorylation, response of electromagnetic field-enhanced ornithine decarboxylase localization, or expression of the 27-kilodalton heat shock protein in activity, Bioelectrochem. Bioenerg. 48 (1999) 193–199.
human keratinocytes, Environ. Health Perspect. 111 (2003) 181–188.
[32] G.A. Boorman, R.D. Owen, W.G. Lotz, M.J. Galvin Jr., Evaluation [35] H. Lin, M. Blank, K. Rossol-Haseroth, R. Goodman, Regulating of in vitro effects of 50 and 60 Hz magnetic fields in regional EMF genes with electromagnetic response elements, J. Cell. Biochem. 81 exposure facilities, Radiat. Res. 153 (2000) 648–657.

Source: http://www.subestacionpatraixfuera.com/documentos/tamoxifen.pdf