Health Articles

Microsoft word - elf mf alter cancer cells.docx

ISSN: 1536·8378 (print), 1536-8386 (electronic) Electromagn Bioi Med, Early Online: 1-12 AND MEDICINE
2013 lnforma Healthcare USA, Inc. DOl: 10.3109/15368378.2013.817334· ORIGINAL ARTICLE Extra-low-frequency magnetic fields alter cancer cells through
metabolic restriction
Ying Li1 2 and PauI Heroux 2 'lnVitroPJus Laboratol'f, Department of Surgel'f, Royal Victoria Hospital,Montreal, QC,Canada and 2Department of Epidemiology, Biostatistics and Occupational Health, McGiJJ University, Montreal, QC, Canada Background: Biological effects of extra-low-frequency {ELF) magnetic fields {MFs) have lacked a AMP-activated protein kinase, credible mechanism of interaction between MFs and living material. Obiectives: To examine the ATP synthase, chromosome instability, effect of ELF·MFs on cancer cells. Methods: Five cancer cell lines were exposed to ELF·MFs extra-low-frequency, magnetic field within the range of0.025-Sf.1T, and the cells were examined for karyotype changes after 6d.   Results: All cancer cells lines lost chromosomes from MF exposure, with a mostly flat dose· History response. Constant MF exposures for three weeks allow a rising return to the baseline, Received 8 January 2013 unperturbed karyotypes. From this point, small MF increases or decreases are again capable Revised 3 May 201 3 of inducing karyotype contractions {KCs). Our data suggest that the KCs are caused by Accepted 26 May 2013 MF interference with mitochondria's adenosine triphosphate synthase {ATPS), compensated Published online 31 July 2013 by the action of adenosine monophosphate-activated protein kinase {AMPK). The effects of MFs are similar to those of the ATPS inhibitor, oligomycin. They are amplified by metformin, an AMPK stimulator, and attenuated by resistin, an AMPK inhibitor. Over environmental MFs, KCs of various cancer cell lines show exceptionally wide and flat dose-responses, except for those of erythroleukemia cells, which display a progressive rise from 0.025 to 0.4 f.1T. Conclusions: The biological effects of MFs are connected to an alteration in the structure of water that impedes the flux of protons in ATPS channels. These results may be environmentally important, in view of the central roles played in human physiology by ATPS and AMPK, particularly in their links to diabetes, cancer and longevity. (Phillips et al., 1986), inhibition of differentiation with increased cell proliferation (Chen et al., 2000) as well Since the Wertheimer and Leeper (1979) article l:inking wire as DNA breaks with apoptosis and necrosis (Lai and Singh, codes to childhood cancer, the relation between cancer and power-frequency magnetic fields (MFs) has been under In the early days of extra-low-frequency (ELF)-MF investigation (Heroux, 1991). Population, in vivo and research, Senllkhina et al. (Sem.ild:rina et al., 1988; in vitro studies have falled to provide a clear link. The Semikhina and Kiselev, 1981) documented by electrical exception is childhood leukemia (Ahlbom et al., 2000), dissipation factor (roRC, also known in electrical engineering leading the International Agency for Research on Cancer to as tg o) and optical measurements (the dimerization of dilute attach the class 2B carcinogen designation to MFs in June rhodamine 60 solutions) that alternating lviFs in the range 2001 (IARC, 2002). 25nT-879T disrupt the anangement of water molecules, It has been argued that envirorunental 60-Hz lviFs, as particularly under high concentrations of hydrogen bonds and non-ionizing radiation and incapable of raising tissue tem- protons. The effects were absent above 40-50°C, as water peratures, could not have significant impacts on cells. But structure changes. The maximum effect was detected at effects on breast cancer cells, MCF-7, were confirmed by a 156.2Hz and 15.45T for 7oc pure water. NatTow reson- number oflaboratories near 1.2T (Ishido et al, 2001). Many ances were observed, easily broadened by the presence of have also reported a diversity of effects above 2.5 T, higher even small levels of impurities. The MF effects on water than common environmental exposures. These include progressed over 5h and dissipated over 2h after the field was (Goodman et al., 1979), increased soft agar colony formation Interestingly, when alternating lviFs were kept below 25nT, an influence of static MFs on water could also be detected. Removing the static MF acted on water Address correspondence to Dr Paul Heroux, Faculty of Medicine, McGill University, Montreal, QC H3A 1A3, C:mada. Tel: 1-514-398- variables (dissipation factor and optical measurements) in a 6988. E-mail: [email protected] direction opposite to the application of ELF-lviFs larger  

Y Li & P. Heroux Electromagn Bioi Med, Early Online: 1-12 than 25 nT. Thus, it seemed that elimination of both ELF and insulin 1mgll (Sigma15500), iron saturated bovine transferrin static MFs allowed water to "opti.rnlze" its molecular 25mgll (Sigma T1408), sodium bicarbonate 2g/1 (SigmaS- 6014) and bovine serum albumin 4g/1 (Sigma A3311, These observations created ground to attempt an inter- Oakville, Canada). Vented T-25s (Sarstedt 83.1810.502, pretation of ELF-MF health effects based on water structure Ntirnbrecht, Germany) and T-12s (Falcon 353018, BD, alterations brought about by the MF itself, as opposed Franklin Lakes, NJ) were used for experiments, and cells to magnetically induced currents. We investigated this possi- are seeded at 5000/cm2 and kept in the same medium for 6 d. bility by setting up baseline cancer cell lines maintained under In longer tests (3 weeks), new medium is added weekly. power-frequency MFs lower than 4 nT, and also under anoxia. Oxygen was eliminated by enclosing T-25s and T-12s in large As 82% of oxygen readings in solid tumors are less than polycarbonate containers (1.61, Starfrit Lock & Lock, 0.33% (Kizaka-Kondoh et al., 2003), and stem cells are hosted Longueuil, Canada) flushed with medical-grade nitrogen in niches that are very low in oxygen (Hill et al., 2009), anoxia (95%) and C02 (5%). pH readings were conducted under is a better simulation of the tumor environment than routinely isothennal conditions (water bath) for samples as well as used 21% oxygen. Our cells are also hyperploid, displaying calibration buffers, using Corning 445 meters (Corning, NY). a range of chromosomes numbers larger than 46, as a result of the enhanced metabolism typical of cancer cells. The absence of oxygen reduces chromosome numbers to some Unexposed cells for experiments are kept in T-12 or T-25 extent, but not back to normal, and also narrows their range culture flasks under anoxia and MFs below 4 nT.Three 6.3 mm (Li et al., 2012). thick layers of structural steel reduce ELF-MFs from incuba- Metabolic restrictors, chemicals that impair oxygen tors and the environment. Culture vessels are centered in a metabolism, adenosine triphosphate (ATP) synthesis or ATP rectangular st.ructw:al steel pipe 5.1 x 7.6 em, itself contained use, can bring back chromosome numbers in cancer cell lines in a7.6 x 10.2mm pipe, both 20 em long.These two shields are even closer to their original46 than anoxia alone, an effect we placed in a 15.2 x 24.5 x 36 em long pipe. This reduces 60- labeled karyotype contraction (KC). KC is a rapid and HzMFs by a factor of 144, providing unexposed cells with a reversible loss of clu·omosomes resulting from metabolic MF environment of 3nT, slightly below the measurement floor restriction (Li et al., 2012). (5 nT at 60Hz) of our Narda EPA-300 instrument (Hauppauge, A critical enzyme in ATP production is ATP synthase NY).The incubator is a Forma 3310 (ThermoFisher, Waltham, (ATPS).The structure of ATPS is documented in detail (Boyer; MA), with low average MF (0.4 ).11). 2002; Sasada and Marcey, 2010) as a rotating motor-generator MFs are applied by rectangular coils (19 x 25.6 em) with structure activated by the trickle of high-density protons from 20-50 turns of #25 AWG varnished copper wire wound on the inter-membrane space into the matrix of mitochondria. 13mm polycarbonate, providing8 ft The coil is under the Proton diffusion along the 15nm thick inter-membrane space two irmer shields and over an ac1ylic spacer at the bottom of does not limit their transit time of 1-2 JlS (Procopio and Fornes, the outer shield. 60-Hz fields above 0.4 JlT are from sector- 1997). Protons enter the Fo of ATPS along an entry half- cormected variable transformers fitted with a passive low-pass channel made of four hydrophilic ex-helices, to reach a rotating capacitive filters, with all harmonics at less than 20 dB. helix. After rotation, protons flow out through a similar exit Smaller 60-Hz fields and other frequencies were generated half-channel.The rotation is used by the F segment of ATPS to with computer-based synthesizers with a background noise at produce ATP (Procopio and Fornes, 1997). less than 40dB. MFs are within 10% of nominal in the These hydrophilic channels (Fillingame etal.,2003) provide whole cell culture area. a high density of hydrogen bonds, while the mitochondrial The ''NIM'' shield cancelling both alternating and static inter-membrane space feedsATPS a high-density of protons, as MFs is an acrylic cylinder 5.7 em in internal diameter 'vith a required for maximum MF effect on water by Semikhina et al. 0.38 em wall and 38 em in length, covered by six layers of (Semikhina et al., 1988; Semikhina and Kiselev, 1981). The 0.4mm nickel-iron-molybdenum foil (ASTM A753 Type 4) high-density protons (pH 1; Procopio and Fornes, 1997) are wound in a spiral, together with a 1.6 mm neoprene driven through the half-channels by a 180kV/cm electric field membrane spacer (Futurplast, St-Laurent, Canada). (Zorov et al., 2009) across the inner membrane (Mitchell, 60-Hz 5 JlT exposures produce no measurable temperature 1966). In this study, we assess the ability of 1-fFs at common rises. K562 is a good the1mal sentinel, hype1ihermia being environmental levels to induce KCs. detectable from larger cell sizes at +0.5 K, while +1K seriously impairs proliferation and +2K over a few days is Cells and culture conditions Chromosome, cell and adenosine monophosphate- The cell lines, K562 and HEL 92.1.7 (erythroleukemias), activated protein kinase assays MCF7 (breast cancer), NCI-H460 (lung cancer) and COL0320DM (colon cancer) were obtained from ATCC Metaphase preparation and cytogenetic analysis were per- (Manassas, VA). Cells are maintained under 5% carbon formed according to the standard trypsin-Gien1sa banding dioxide and 90% humidity, and grown in synthetic culture technique. Karyotypes are obtained using x100 oil immer- medilll11. because changes in serum can alter chromosome sion, a Laborlux D (Leitz, Leica, Wetzlar, Germany) micro- counts. The medium is RPMI-1640 with L-glutamine (Sigma scope, and an Infinity X (21 Mpixels) CMOS camera 61-030-RM), sodium selenite, 20nM (Sigma S-5261), bovine (Lumenera, Ottawa, Canada).

  DOl: 10.31091153683782()13.817334 Extra-low-frequency magnetic fields alter cancer cells Cell proliferation and cell size histograms of are fi:om a the horizontal coil induces cments six times larger Scepter automated cell counter (Millipore, Billerica, MA). because the exposed culture dish area is 34 x 34 mm for Metformin was obtained from Sigma (D150959) and resistin the horizontal coil, compared to 5.8 x 34mm for the from Prospec Protein Specialists, East Bnmswick, NJ. vertical coil. As KCs after 6 d at 1J.LT come out similarly for both orientations (Figure 1), we conclude that the effect on chromosome numbers are dependent on the MF itself. Ve assume direct MF, rather than induced cment action on Because of the controlled MFs and of anoxia, our reference the basis that variations of cunent density by a factor of 6 do K562 cultures are karyotypically and otherwise exceptionally not affect KC. But this would also occur if induced cments stable. Seventy-five percent of the cells have just had a flat dose-response, already saturated at the lower t vo chromosome numbers, 62 and 61, compared to a wider current. Furthemwre, direct MF action on KC does not range under 21% oxygen (Li et al., 2012). The stability of preclude that other effects of MFs may depend on induced chromosome numbers in baseline anoxic K562 has been pe1iodically confirmed in our laboratory over 5 years. These cultures provide an extremely precise reference point, as shown in the nanow baselines of Figures 1, 2 (top) and 4. This is of great advantage in obtaining statistical Figure 2 (top) shows the chromosome number losses significance in our data. Figure 1 sho -vs little overlap between experienced by previously shielded anoxic K562 cells after baseline and exposed data, yielding small numbers in 6 d in various MFs. Under any exposure, the nanow baseline Student's t-tests. In Figure 2 (top), the p value between expands, and there are substantial KCs. baseline and 0.025 J.LT is 0.00012. In Figure 3, even when Three features are of importance. First, a no-effect-level using 21% oxygen, the large number of karyotypes performed, lower than 25nT. Second, a progression of KCs up to 0.4 J.LT. and the strong shifts in average chromosome numbers Third, the relatively flat dose-response bet veen 0.1 and produced by MFs result in extremely small p values (0.000006 for MCF7). The graph spans time-averaged MFs representing domestic (0-0.2J.LT), commercial (0.07-0.5 J. LT) and occupational (0.1- 1J.lT) environments (Heroux, 1987). Induced currents Later sections of this article will argue that MFs act most 'hether biological effects of power-frequency MFs relate to directly on the structure of water, leading to an alteration in the MF itself, or to the CUlTents induced in tissues by the proton mobility. As proton mobility is tightly related to pH, a fields, has been a perennial question. Many think that effects measurement of hydrogen ion activity, some perturbation of occur through potentials produced by magnetically induced pH values might be expected. A lasting effect of MFs on currents blocked by the thin membranes, within or bordering aqueous fluids is actually observable from pH measurements living cells. Such cments and membrane potentials are in cell culture media, which turn slightly more acidic under familiar to conventional electrophysiology. short MF exposures. After 20h, there is a difference In the results of Figure 1, one aliquot of an anoxic of 0.09 pH units with a 95% confidence interval of K562 cell culture is placed in a vertical and the second in 0.045 between unexposed versus 5 J.LT 60-Hz exposed a horizontal MF exposure system. At the same MF, media (Figure 2; bottom) for the widely used RPMI-1640 Figure 1. Baseline anoxic K562 cells at less 70-- ------------ ------ ------------ ------ -, tl1an 4nT (60Hz) with an average of 61.5 chromosomes (horizontal line) and a very narrow distribution (at left) are simultan- eously transferred for 6 d to 11-1T MFs applied IIOKIZOI<TAI. eitl1er horizontally or vertically. Three inde- pendent 6-d assays show the resulting chromosome numbers. Box plots show median (solid), average (dotted), 25 and 75% (box), 10 and 90% limits (whiskers) and outside values (dots). Fifty-si'< (assay 1), 50 (assay 2) and 51(assay 3) metaphases were karyotyped in each orientation. Inside the box plots are average chromosome losses. TI1e Student's t-test results quantify the probabil- ity that the horizontal and vertical results are

Y Li & P. Heroux Electromagn Bioi Med, Early Online: 1- 12 Figure 2. Top: K562 chromosome numbers as a function of 60Hz magnetic flux density. Six-day assays with, in sequence, 65, 28, 50, 77, 46, 33, 65, 102, 56 and 50 metaphases. 1 vo to 6 experiments at each MF. Approximate ranges for domestic, commer- cial and occupational exposures are shown. Bottom: pH differences between two cell medium aliquots, one exposed for 20h to <4nT at 60Hz and the second to the MF density in the figure. Medium is RPMl- 1640 with 10% FBS. Isotherm measurements using auto read were made with the same probe, alternating between the two aliquots. TI1ree measurements for each aliquot and three repeats at each MF density. ComtnetSlll
IA.SCUHI:
Magnetic Flux Density (1-JT) with 10% serum. The pH shift was confirmed for a variety of numbers is even less than what is observed in the long-term cell culture media baseline culture, as shown in the last measurement of Figure 4 (top) and in the central measurements of Figure 4 (middle and Across cell lines bottom). Chromosome numbers restore earlier than chromo- some number dispersions. Beyond K562, we investigated four more hyperploid cancer After 3 weeks, if the MF is altered by a small percentage of cell lines to determine the generality of KC by MFs.Over two the original value, either positively or negatively, KCs are orders ofMF magnitude, erythroleukemia (HEL 92.1.7), breast again observed, as shown in Figure 4 (middle and bottom). (l 1CF7) and lung (NCI-H460) cancer cells lose between 8 and Starting from low (middle, 0.1 T) or high (bottom. 1 T) 13 chromosomes (Figure 3).BEL, our second erythroleukemia baselines, symmetrical KCs are observed. This bilateral cell line, shows fewer losses at lower fields, similar to K562. sensitivity to changes is unforeseen by conventional toxico- Three of the four results reported in Figure 3 were obtained logical principles. KC is also observed when fields are reduced under standard (21% oxygen) culture conditions. from 50 to 4nT (not shown). Classical toxicology and epidemiology, where smoothly The KCs will be interpreted below as caused by magnet- climbing dose-responses are justified by binding chemistry ically induced perturbations in intra-cellular AlP levels. and the central tendency theorem. do not expect the flat dose- These results cast doubt on the stability of cancer cell models responses observed in Figures 2 (top) and 3.The effects found housed in incubators with MFs that are highly variable over for different cell types are strikingly similar, with similar low- space and time (l 1ild et al., 2009; Su and Heroux, 2012). field deviations in the two erythroleukemia lines, suggesting common, basic mechanisms. Differential action We measured in anoxic K562 6-d tests at 1 T, the average KCs over frequency as follows: K562 cells with magnetic KCs, such as in Figure 2 (top), progressively recover their original chromosome numbers 3.6 ± 0.79 at 50-Hz, 9.36 ± 1.06 at 60-Hz, after 3 weeks, even as the :MF is maintained at a constant level 12.71 ± 1.82 at 120-Hz and 9.8 ± 1.31 at 155-Hz. (Figure 4; top). Surprisingly, in cells recovering over 3 weeks from a MF disturbance, the deviation of chromosome A polynomial fit predicts maximum KC at 113Hz for 1T . DOl: 10.31091153683782()13.817334 Extra-low-frequency magnetic fields alter cancer cells Figure 3. Average chromosome losses in erythroleukemia, breast, lung and colon cancer cells as a function of 60Hz magnetic flux density. The cllromosome number baseline(' '0") averages for <4nT cells at 60Hz, 80% range and metapllase number are: HEL: 66, 62-67 and 32; MCF7: 74, 61-75 and 30;NO-H460: 57,53-65and 30; and COLO 320DM: 54,49-61and 30. Six-day assayswitll, in sequence, 32, 22, 29, 32; 19, 22, 19, 21; 29, 22, 24; 22, 34 and 46 metapllases. Two to five experiments at each MF.HEL, NO-H460 and COL 320DM assays used 21% oxygen, ratller tl1an anoxic conditions, as some anoxic karyotype modes are too close to 46 to allow easy statistical separation from MF-exposed samples. (less than 5 n'I) and the static field to 3 T. Karyotyping Static MF removed revealed a very slow drift downward, but a strong effect on The :influence of the static MF was investigated by observing proliferation rate was observed. After 4 d, cell numbers in the K562 cells transferred from a steel shield that eliminated ELF NIM shield were increased by a factor of 2.05 ± 0.13 (SD) MFs (to less than 5 n'I), but had a static field of 74 T, to over cells kept in the steel shield, indicating enhanced a second shield (NIM) that attenuated both the ELF-MF metabolism. The effect is persistent over time. 6 Y Li & P. Heroux Electromagn Bioi Med, Early Online: 1-12 Figure 4. Top: K562 chromosome numbers return to baseline after 3 weeks of continuous 1f.lT lVIF exposure. Sixty -five, 102, 50 and 37 metaphases. Two experiments at each lVIF. Middle: K562 chromosome numbers obtained after 6 d by altering baseline lVIFs of 0.1f.lT. Twenty, 31, 37 (baseline), 31 and !l 10
35 metaphases. Three to six e.-xperiments at each lVIF. Bottom: For 1f.lT, 28, 28, 37 8
(baseline), 28 and 28 metaphases. Three experiments at each lv!F. Although the symmetry of the chromosome numbers is strong, there is more cell decay with increased than with reduced fields. Weeks of 1 T Exposure
Il l ""0""' 009 010 O lt
";2"": < Field Oeorease : Faeld Increase >
lb'!fllnt'
Magnetic F1eld Density (IJT)
  DOl: 10.31091153683782()13.817334 Extra-low-frequency magnetic fields alter cancer cells MF and oligomycin MF and adenosine monophosphate-activated protein Previous experiments (Li et al., 2012) on the five cancer cell lines used in this article show a link between metabolic The similarity between 0.4 )lT and oligomycin suggests that restriction and KC. Anoxia alone induces partial KCs of 6-8 the MF may be an inhibitor of ATPS, as oligomycin is a chromosomes. Deeper contractions, almost to normalization highly specific inhibitor of ATPS. If this were the case, of the karyotypes to 46, are produced by IC50 doses (allov.ring inhibition of mitochondrial ATPS by lYIFs would activate 50% of the nonnal cell division rate) of the metabolic adenosine monophosphate-activated protein kinase (AMPK), restrictors oligomycin and imatirrib. Similar KCs are because healthy cells must maintain a high level of phos- produced by physiological levels of melatonin and vitamin phorylation capacity (ATP:ADP10) to function well (Hardie and Hawley, 2001). AMPK is a sensitive ATP We believed that comparison of metabolically restricted regulator that switches on catabolic pathways and off many cultures with MF-exposed cultures could provide clues on ATP-consmning processes, both acutely and clu·orrically, action mechanisms, as the different metabolic restrictors through gene expression. mentioned above have different sites of action.Figure 5 (top), The MF>ATPS>AMPK pathway was investigated using displays the similarity in cell size distribution after 6-d metfonrrin and resistin. Metfonnin is a diabetes drug that between two of seven anoxic K562 assays, one exposed to a activates AMPK, leading to reduced glucose production in the very effective MF, 0.4 )lT at 60-Hz, and the second to liver and reduced insuJin resistance in muscle. It is an oligomycin at IC50 (2.5 ng/mJ). TI1e two distributions attractive anti-aging drug that usually causes weight and stand apart, with smaller cell diameters and higher ratios of cells-to-objects below 11J.llll, the decay particles and Resistin, a product of the RSTN gene, is a 9.9 kDa protein apobodies. This suggests that lYIFs and oligomycin share a containing 93 alllino acid residues which, at 20ng/rn1 or more, common mode of action. Despite the closeness between inhibits AMPK. It interferes vlth phosphorylation of Akt oligomycin and MF assays in Figure 5 (top), oligomycin is (serine/threonine protein kinase), active in multiple cellular faster-acting than 0.4 )lT: changes in cell size, revealing of processes such as glucose metabolism, cell proliferation, KC, are visible at 1d, em·lier than for the IVIF. But more apoptosis, transcription and cell migration. efficant MFs, such as 5 )lT at 60-Hz or 1)lT at 120-Hz, show Metfonrrin (0.01mg/1) and resistin (40ngll) alone for 3d effects earlier (not shown). induce average KCs of 9 and 10, respectively, in K562. Figure 5. Top: Object diameter histograms for 6-d anoxic exposures of K562 cultures to 0.4 f!T MF at 60-Hz and oligomycin at ICso (2.5 ngfml). The lower four ICso curves are: imatinib (0.04 J!gfml), resistin (40ngtml), metfonnin (0.01 mg/ml) and melatonin-vita- min C (0.3 J.lg!ml and 26 J.lg!ml). All cultures are adjusted to a common small particle count maximum. Bottom: Object diameter histograms for 7-h 21% oxygen exposures of three K562 cultures under typical incubator MF. Aliquots of RPMl-1640 witl1 10% FBS medium exposed for 15h to very small MF (<4nT at 60Hz, 3 f!T static), incubator MF (2-2.7 f!T at 60Hz) or Inhibitory MF (0.62 f!T at 120Hz) were seeded with cells at Object Diamete•·(l.tm) time 0 and measured with a lvfillipore Scepter at 7h. Average of three repeats for each condition. TI1e p value between average levels (l2-16J!m) for very small MF and inhibitory MF is 0.001 (n-4). Object Diameter (Jtm)   8 Y Li & P. Heroux Electromagn Bioi Med, Early Online: 1-12 'hen, in the 6-d trials routinely used for MF tests, 1T is There is an increase (Figure 5; bottom) in the number of combined with metfornrin, even larger KCs are observed (9 living cells observed nnder the very small MF condition, becomes 11± 0.34). 'hen 1 Tis combined with resistin, the compared to the inhibitory MF condition, with the incubator KC of resistin reduces from 10 to 4 ± 0.46, aJso less than the MF condition rating in between. KC of 1T alone, at 7.5. The conclusion from summary of When stressed cells from a culture with depleted medium Table 1 is that l .1Fs enhance the action of metformin, but (lower pH) were used, the inhibitory MF had the effect of neutralize the effect of resistin, again suggesting a connection increasing the level of decay products (object diameters less between l .1Fs and ATPS. than 11 )in the culture (not shown). Experiments on medium alone NCI-H460 and MCF7 proliferation Cells grown for 7 h nnder identical incubator MF (2-2.7 Tat Beyond strong effects on cancer cells karyotypes, MFs also 60-Hz) conditions fared differently according to whether the impact proliferation rate, adhesion and cell shape, but in culture medium added at Oh originated from closed flasks specific cases, such as for lnng (NCI-H460) and breast cancer exposed for 15 h to: very small MF (<4 nT at 60-Hz, 3 T (MCF7), effects are particularly striking. Figure 6(C) illus- static), incubator MF (2-2.7T at 60-Hz) or inhibitory MF trates that the cell connts of lnng cancer cells after 4 d in our (0.62T at 120Hz). synthetic medium at 50nT, 400nT and 5 Tare 8, 9.2 and After the sealed flasks v.>ith media (only) are exposed 14.8, respectively, times larger than those of nnexposed (4 n'I) to their respective MFs, cell culture aliquots are introduced cells. NCI-H460 and MCF7 cells grown at less than 4nT in into each flask, and incubated for 7 h nnder incubator our synthetic medium do not attach to the growth surface, but iv!F conditions. Measurements of cells numbers of each do so nnder any MF exposure. Figure 6 illustrates that the S1Ze are acquired at 0h, as well as at 7 h for each flask. appearance of MCF7 cells after 4 d.is completely different at <4nT (A) than at 5T (B). These MF-related differences in adhesion, appearance and proliferation last over two passages, Table 1. K562 karyotype contractions under the action of AMPK modulators and MFs. but finally return to the (A) characteristic, the "non- adherent'' style, which is coherent with the restoration of Agent Concentration/intensity chromosome numbers nnder constant l .1F exposures displayed Metformin (activator) in Figure 4 (top). The proliferation results for NCI-H460 are Resistin (inhibitor) also compatible with our metfornrin and resistin experiments, which suggest that MFs stimulate Akt, which in turn activates protein synthesis and apoptosis inhibition in NCI-H460 (Hovelmann et al., 2004).    '" I l l "".tnooc Flo4d Oon•IIJ IIITI ------r------- I-----1 Figure 6. EffectofMFs on MCF7 adhesion and shape and NCI-H460 proliferation. MCF7 (breast cancer) a:nd NO-H460 (lwlg cancer) cells originally under shielded levels (4nT). (A) MCF7 exposed to 4nTfor 4d. (B) MCF7 exposed to 5 J.lT for4 d. (C) Proliferation ofNCI-H460 over4 d as a function of MF density. (D) Evolution of NO-H460 over 8d for 4nT and 5 J.lT exposures. Three experiments per determination. limited number of binding sites, but still manage to elicit a strong 001: 10.'3109/153683782()13.817334 response (Vandenberg et aL, 2012). In our case, the "binding Discussion
2008; Zorov et al., 2009). The transitions between para and ortho forms are forbidden by quantum mechanics. They rarely Possible site of action of Mf s occur during random thermodynantic fluctuations as a proton There are documented examples of hydrogen bonding drifts away from its OH radical, releases Pauli exclusion impacting biological processes. The hydrogen bond is 4% principle constraints and allows spin flipping according to an longer for heavy water than for light water, which results in external MF. The ortho state is sensitive to the l .1F. This lethal rnamrnalian toxicity at about 50% heavy water (Kushner dependence of the transitions on thermodynantic instability is et al., 1999). Small changes in hydrogen bond lengths and convenient to explain the slow rate of action ofMFs on water. angles are critical in determining how binding pockets react, The detailed physics within the water twmel is complex as demonstrated in the selective uptake of phosphorous over because proton tunneling is probably coordinated with arsenic in bacteria (Elias et al, 2012). electron tunneling (Gray and Winkler, 1996; Moser et al., The involvement of water structure disruptions through 1992). Proton-coupled electron transfer underpins many bio- alterations in hydrogen bonding was also anticipated by recent logical reactions and may occur as unidirectional or bidirec- views on EMF bioeffects (De Ninno and Castellano, 2011; tional, and synchronous or asynchronous, transfer of protons Novikov et al., 2010). and electrons (Reece and Nocera. 2009). Most enzymatic If the effect on water described by Semildrina and reactions would involve a single such transfer, wbile ATPS is Klselev is involved, it would be more prominent "the exquisitely sensitive to MFs because its structure is dependent higher the concentrations of hydrogen bonds and proton on a serial layout of such transfers. containing groups in aqueous systems'' (Semikhina et al., Finally, the metabolic ''restriction'' that we attribute to the 1988). The only known location in the biota where these action of MFs on ATPS could be labeled as a metabolic two conditions are met with such intensity are the entry "disruption" and bas at least three characteristics in common and exit water channels of ATPS. That protons can with endocrine disruption. proceed through water channels by tunneling bas been First, dependence on very specific receptors, namely ATPS confirmed in water-filled nanotube models where neutron channels of mitochondria for the MF, and ligand-receptor Compton scattering has detected as double wells the systems in the cell membrane, cytosol or nucleus for quantum delocalization of protons (Reiter et al, 2011). endocrine disrupters. Second, the quickly saturating dose- Proton movement is sensitive to the global configuration response curves observed in Figures 2 (top) and 3 are similar, of hydrogen bonds. for example, to the effects of atrazine on the size of the larynx. The intense electric field (Zorov et al., 2009) applied to Atrazine does not shrink the larynx, but it inhibits the ATPS Fo in mitochondria lowers the potential barriers androgen stimulus (Hayes et al., 2002). Third, thresholds of inhibiting proton movement. In low-barrier hydrogen bonds, action at very low levels: 25nT for l .1Fs and pico-molar to hydrogen is free to move between two oxygen atoms (Cleland, nano-molar levels endocrine-disrupting 2000). According to this mechanism, MFs would impede the (Vandenberg et al, 2012). quantum mechanical proton flow through the hydrophilic The increased metabolism observed when alternating and channels, and MF removal would improve proton flow, static MFs are removed. and the ability of MF-conditioned directly impacting ATPS efficiency. The dose-responses culture media to influence cellular development are aD of Figures 2 (top) and 3 would be determined by rising compatible with Russia11 water data. The presence of a KC proton impedance (decreased soliton tunneling) through resonance wider tha11that observed for pure water by Russian ATPS half-channels. physicist adds support. Finally, the ratio between frequency The involvement of protons relieves the area of EMF and field intensity (fiB) for maximum biological effects is bioeffects of the "kT problem", because EMF would act not suggestive of a coupling with the gyromagnetic ratio of the on molecules, but on particles (protons and electrons), wbich, as ferrnions, do not follow Maxwell-Boltzmann statistics, but In tbis context, similar fingerprints between the 0.4}-LT and Fermi-Dirac statistics, and are governed by quantum oligomycin (Figure 5; top), known to inhibit ATPS by binding to the o subunit of the Po segment of ATPS (also named electrodynamics. Numerous elements documented by Russian physicists in oligomycin sensitivity conferral protein) comes as no surprise. Another intriguing link between MFs and ATPS is provided their studies of :MF's on water (Semikhina et al, 1988; by the fact that rhodamine 6 G, used by Semikbina to detect Senrikbina and Kiselev, 1981) are compatible with our own MF effects on water, also happens to inhibit the F biological observations. It is particularly notable that the KC threshold of 25nT in Figure 2 (top) falls in line with the water effect threshold detected by Russian physicists (Semikhina et al., 1988). The extended, flat response is also compatible with their obser- vations, and there is an interesting biological aspect to this flat dose-response. Quickly saturating dose-responses are observed in endocrine disrupters that attach to a Y Li & P. Heroux Electromagn Bioi Med, Early Online: 1-12 KC,AMPK and diabetes Mfs and cancer epidemiology Perturbations of A1P concentrations trigger AMPK, which For many cancer cell types, the dose-response of KC versus activates p53 and reduces both ATP consumption and DNA MFs is remarkably flat (Figure 3). The deviation from synthesis (Jones et al., 2005; Motoshima et al., 2006). flatness in erythroleukemia cells (Figures 2 (top) and 3; HEL) The suppression ofDNA synthesis, part of AMPK's catabolic is due, we suspect, to extra-mitochondrial A1P secretion in control, leads to KCs through suppression of chromosome the cell membrane (Arakaki et al., 2003) where pH is at a endoreduplication, the mechanism probably responsible physiological 7.3 rather than 1, a probable feature of this cell for rapid chromosome number increases :in cancer cells type (Das et al., 1994). (Li et al., 2012). If KC is indeed a marker of increased malignancy, there is Two unusual aspects of MF action, adaptation to a stable a possibility of carcinogenicity from MF exposures. fu such a field over three weeks (Figure 4; top) and the unusual case, the phenomenon would not be easy to document through shorter-term sensitivity to small MF increases and decreases epidemiology.First, the threshold for the effect (25n1)is very (Figure 4; middle and bottom) are compatible with AMPK low, which means that all the population is "exposed''. physiology. As far as we know, this is the first example of an Second, the dose-response is unusually flat (Figure 3), such agent presenting this kind of symmetry, making it possible to that useful low and high exposure groups vlth otherwise sustain KCs indefinitely by judicious selection of MF similar characteristics would be difficult to assemble. Third, sequences. AlYIPK is easily triggered by small changes in the differential action of MFs may confuse conventional A1P levels (Hardie and Hawley, 2001), and also controls exposure analysis. long-term dynamic adaptation in muscle (Winder et al., Occupational studies are often at the forefront of 2000). The connection between metabolic restrictors, includ- epidemiological discovery because of their higher and better ing MFs, and KC can be explained by AMPK physiology. docwnented exposures. According to Figure 2 (top), occupa- The MF >A1PS > AlYIPK pathway is easily detectable in tional populations of low (0.1JlT) and high exposures (1)l1) cancer cells because of KC, but there is no reason to think that have a KC difference of "1 chromosome" between them. the A1PS of normal cells is spared under MF exposure. A Domestic MF epidemiology on leukemia may have been major regulator of metabolism (Liu et al., 2006), AMPK successful (Ahlborn et al., 2000; Svendsen et al., 2007) modulates insulin secretion by pancreatic beta-cells (Winder because it benefited from a KC of "10 chromosomes" and Hardie, 1999) and is investigated for the treatment of between 0 and 0.4 JlT (Figure 2; top). diabetes (Viollet et al., 2009).AMPK is tied with body weight The increased proliferation rates reported for lung cancer (Kim et al., 2004) as well as with immWle cell behavior cultures may also be important.Lung cancer was pointed in at (Kanellis et al., 2006). Type 2 diabetes has been linked to least three studies related to EMFs (Armstrong et al., 1994; prolonged and persistent exposures to endocrine disruptors l'vfiller et al., 1996; Vagero and Olin, 1983). (Lee et al., 2010). Conclusion
The following evidence swpmt the inhibition of A1PS Cancer cells depend on glycolysis and significantly upregu- late it when respiration is inhibited. The Warburg effect  manifests as increased glycolysis and reduced mitochondrial Mfs alter metabolism respiration (Jezek et al., 2010; Wu et al., 2006). These capabilities of cancer cells allow growth under meta- (1) MFs induce KCs in five cancer cell lines, as do other bolic restriction by concentration of their resources on metabolic restrictors (Li et al., 2012). bio-synthesis through the elimination of detoxification mech- (2) MFs interact with metformin and resistin as would an anisms associated with oxygen exposure, such as glutathione-S-transferase and CYP3A4 expression (Nagai (3) Elimination of alternating and static MFs produces a et al., 2004). The smaller karyotypes maintained under durable increase in cell proliferation. metabolic restriction contribute to tumor core expansion, as fewer chromosomes can be more rapidly duplicated. The survival of twnors could thus be enhanced by certain levels of chronic metabolic restrictions from hypoxia, (4) The KC threshold (25 n1), as well as its extent over two oligomycin or MFs. orders of magnitude, is predicted by the work of Russian It has been repeatedly confirmed that cancer cells physicists on water (Semikhina and Kiselev, 1981). become more malignant under metabolic restriction (Hill Lack of sensitivity to MF intensity or to cell type et al., 2009; Rockel and Vaupel, 2001; Jogi et al., 2003) suggest tl1e knockout of a biological enzyme by physics. in vitro (Anderson et al., 1989) and in the clinic (Brizel et al., (5) MF-exposed culture meditun. without cells, is a vector 1996; Nordsmark et al., 1996), to the point where it has of MF action (proliferation and cell decay). become a central issue in twnor physiology and treatment (6) Measured changes in the pH of cell culture media from (Rockel and Vaupel, 2001). From our data, it is logical to conclude that KC observed under metabolic restriction is a (7) A wide KC resonance (113Hz at 1)l1) is compatible possible indicator of meta-genetic promotion in cancer cells with the work of Russian physicists on water (Semikhina (Li et al., 2012). and Kiselev, 1981). Y Li & P. Heroux Electromagn Bioi Med, Early Online: 1-12 laboratory support and Michel Bourdages, DOl: 10.31091153683782()13.817334 Institut de Recherche d'Hydro-Quebec, for (8) KC is maximized at specific frequency-amplitude (fiB) equipment contributions. We thank Louis Slesin for combinations (Semikhina and Kiselev, 1981), suggest- reviewing the document We also thank an anonymous ing an interaction with water's protons. reviewer for contributions to the article. Mfs alter ATPS Fo Declaration of interest
(9) ATPS Fo is the only site in the biota where conditions We declare no competing financial interests. for maximum sensitivity to MF action (Semik:hina et al., The work was suppmted by Royal Victoria Hospital 1988) happen together: high concentrations of protons Research Institute Fund 65891. and hydrophilic bonds in a narrow channel. (10) Strongly acting MFs induce cell cultme characteristics (Figure 5; top), closely matching those of a specific References
ATPS Fo inhibitor, oligomycin. (11) 1-IF activation of AlVIPK implies a perturbation to ATP Ahlborn, A., Day, N,, Feychting, M., et al. (2000), A pooled analysis of levels, thus a change in ATPS performance. magnetic fields and childhood leukaemia. Br. J. Cane. 83:692-698. Anderson, G. R., Stoler, D. L., Scarcello, L A. (1989). Normal (12) Rhodamine 6 G, used by Russian physicists (Semik:hina :fibroblasts responding to anoxia exllibit features of tlle malignant and Kiselev, 1981) to detect MF effects, is also an phenotype. J. Biol. Chem 264:14885-14892 (accessed 5 Jul2013). inhibitor of ATPS Fo. Andreev, S. N., Makarov, V. P., Tikhonov, V.I., Volkov, A. A. (2008). Environmental :rv!Fs act on the core of human metabolism. Ortho and para molecules of water in electric field. Available from: llttp://arxiv.orgfabstphysicsf0703038v1 (accessed 5 Jul 2013). Past evaluations of MF bio-effects were at a serious Arakaki, N., Nagao, T., Niki, R., et al. (2003). Possible role of cell disadvantage because of traditional toxicological and surface H+-ATP synthase in the extracellular ATP syntl1esis and epidemiological assumptions, that larger exposures induce proliferation of human umbilical vein endotl1elial cells. MoL Cane. larger responses. The controls of in vitro scientists were Res. 1:931-939. Armstrong, B., Theriault, G., Guenel, P., et al. (1994). Association ah·eady randomly exposed by the :rv!Fs of their incubators. between exposure to pulsed electromagnetic fields and cancer in The flatness ofMFs' dose-response impaired epidemiological electric utility workers in Quebec, Canada, and France. Am. J. work, as most studies, except for domestic leukemia, used Epidermiol. 140:805-820. tainted controls (Milham, 2010). The interaction between Boyer, P. D. (2002). A research journey witl1 ATP synt11ase. J. BioL Chern. 277:39045-39061. power-frequency :rv!Fs and living cells may have been Brizel, D. M., Scully, S. P., Harrelson, J. M., et al. (1996). Thmour underestimated for a long time, because of these unexpected oxygenation predicts for tl1e likelihood of distant metastases in human characteristics. soft tissue sarcoma Cancer Res. 56:941-943. Some diseases appear to have strengthened, with no Bulterijs, S. (2011). Metfonnin as ageroprotector. Rejuvenation Res. 14: clear causation, as more advanced technology, in great part Chen, G., Upham, B. L., Sun, W., et al. (2000). Effect ofElVfF exposure based on electricity, has expanded. Chronic diseases that on chemically induced differentiation of friend erythroleukemia cells. increased or decreased in the past century, and that are Ercv. Health Persp. 108:967-972. connected to ATP metabolism, should be examined for a link Cleland, W. W. (2000). Low-barrier hydrogen bonds and enzymatic catalysis. Arch. Biochem Biophys. 382:1-5. with MFs. But our understanding of AMPK and metabolism Das, B., Mondragon, M. 0. H., Sadeghian, M., et al. (1994). A novel is incomplete (Jones and Thompson, 2009), making a link ligand in lymphocyte-mediated cytotoxicity: E:<pression of the/3 between MFs and any specific disease, such as diabetes, subunit of H+ transporting ATP synthase 011 the surface of tumor cell lines. J. Exp. Med. 180:273-281. De Ninno, A., Castellano, A. C. (2011). On tl1e effect of weak magnetic According to our data, MFs are physiological agonists of :field on solutions of glutamic acid: The function of water. J. Phys. metformin, perhaps through inhibition of ATPS. But metfor- Con(. 329:012025. min is a knovm geroprotector (Bulterijs, 2011). Beyond the Elias, M., Wellner, A., Goldin-Azulay, K., et al. (2012). The molecular anti-oxidants we investigated and :rv!Fs, there are other agents basis of phosphate discriminatiotl in arsenate-rich environments. known to impair the use of oxygen. H Nature 491:134-137. 2S, for example, inhibits Fillingame, R. H., Angevine, C. M., Dmitriev, 0. Y. (2003). Mechanics the cytochrome enzymes of mitochondria, reversibly slowing of coupling proton movements to c-ring rotation in ATP syntl1ase. the metabolism of mice, lowering their body temperatme, FEBS Lett. 555:29-34. breathing rate and use of oxygen. Ftuthermore, Goodm Ul, E. M., Greenebaum, B., Marron, M. T. (1979). Bioeffects of extremely low-frequency electromagnetic fields: variation with inten- Caenorhabditis elegans, when grown in 50 ppm H2S, lives sity, waveform, and individual or combined electric and magnetic 70% longer than in normal air (Miller and Roth, 2007). Such :fields. Radiat. Res. 78:485-501. models suggest that the ability of MFs to suppress ATPS Gray, H. B., Winkler, J. R. (1996). Electron transfer in proteins. Annu. (Hootkooper et al., 2013) may have played a role in the Rev. Biochem 65:537-561. Hardie, D. G., Hawley, S. A. (2001). AMP-activated protein kinase: The increased lifespan observed in developed countries in the past energy charge hypothesis revisited. BioEssays 23:1112-1119. Hayes, T. B., Collins, A., Lee, M., et al. (2002). Hermaphroditic, demasculinized frogs after exposure to t11e herbicide atrazine at low ecologically relevant doses. ?roc. Natl. Acad. Sci. USA 99: 5476-5480. Authors acknowledge the late Nancy Wertheimer and Edward Herou:'<, P. (1987). 60Hz electric and magnetic fields generated by a Leeper, who saw it first; Semikhina and Kise]ev, whose work distribution network. Bioelectromagnetics 8:135-148. Heroux, P. (1991). A dosimeter for assessment of exposures to ELF made our analysis possible. We are grateful to Janet Moir :fields. Bioelectronw.gnetics 12:241-257. and Lome Beckman of the Royal Victoria Hospital for Hill, R. P., Marie-Egyptienne, D. T., Hedley, D. W. (2009). Cancer stem cells, hypoxia and metastasis. Semi.rr. Radiat. OncoL 19:106-111. Moser, C. C., Keske, J. K., Warncke, K., et al. (1992). Nature of biological electron transfer. Nacure 355:796-802. , M., Vanpel, P. (2001). Tumour hypoxia: Definitions and current Motoshima, H., Goldstein, B. J., lgata, M., Araki, E. (2006). AMPK and clinical, biologic, and molecular aspects. J. Natl. Cancer lnstic. 93: cell proliferation - AMPK as a therapeutic target for atherosclerosis and cancer. J. Physiol. 574:63-71. Houtkooper, R. H., Moucl1iroud, L., Ryu, D., et al. (2013). Mitonuclear Nagai, F., Kato, E., Tamura, H. (2004). Oxidative stress induces GSTP1 protein imbalance as a conserved longevity mechanism. Nature 497: and CYP3A4 expression in the hwnan erythroleukemia cell line, K562. BioL Pharm. Bull. 27:492-495. Hovelmann, S., Beckers, T. L., Schmidt, M. (2004). Molecular Nordsmark, M., Hoyer, M., Keller, J., et a!. (1996). The relationship alterations in apoptotic pathways after PKB/Akt mediated chemore- between twnour oxygenation and cell proliferation in hwna11 soft sistance in NCI H460 cells. Br. J. Cane. 90:2370-2377. tissue sarcomas. Inc. J. Radiat Oncol. Biol. Phys. 35:701-708. International Agency for Research on Cancer. (2002). IARC Monographs Novikov V. V., Ponomarev, V. 0., Novikov, G. V., et al. (2010). Effects on che Evaluacion of Carcinogenic Risks co Humans. Non-ion ing and molecular mechanisms of tl1e biological action of weak and Radiacion, Part 1: Scatic and Excremely Low-Frequency (ELF) extremely weak magnetic fields. Biophysics 55:565-572. Eleccric and Magnecic fields. Volume 80. Lyon, France: IARC Press. Phillips, J. L., Winters, W. D., Rutledge, L. (1986). In vitro exposure to lshido, M., Nitta, H., Kabuto, M. (2001). Magnetic fields of 50-Hz at electromagnetic fields: Changes in tumour cell properties. Int J. 1.2 f!T as well as 100 f!T cause uncoupling of inhibitory pathways of Radiac. Biol. 49:463-469. adenylyl cyclase mediated by melatonin 1a receptor in MF-sensitive Procopio, J., Fornes, J. A. (1997). Fluctuations of the proton- MCF-7 cells. Carcinogenesis 22:1043-1048. electromotive force across the inner mitochondrial membrane. Phys. Jezek, P., Plecita-Hiavata, L., Smolkova, K., Rossignol, R. (2010). Rev. 55:6285-6288. Distinctions and similarities of cell bioenergetics and t11e role of Reece, S. Y., Nocera, D. G. (2009). Proton-coupled electron transfer in mitochondria in hypoxia, cancer, and embryonic development. Tnt J. biology: Results from synergistic studies in natural and model Biochem. Cell Biol. 42:604-622. Jogi, A., 0ra, 1, Nilsson, H., et al. (2003). Hypoxia-induced dediffer- systems. Ann. Rev. Biochem 78:673-699. http://aiXiv.org/abs/ entiation in neuroblastoma cells. Cancer Lect 197:145-150. 1101.4994v1 (accessed 5 Jul 2013). Jones, R. G., TI10mpson, C. B. (2009). Tumor suppressors and cell Reiter, G. F., Kolesnikov, A. 1,Paddison, S. J., eta!. (2011).Evidence of metabolism: A recipe for cancer growt11. Genes Dev. 23:537-548. a new qua11tum state of nano-confined water. Available from: http:// Jones, R. G., Plas, D. R., Kubek, S., et al. (2005). AMP-activated protein arxiv.org/abs/ll01.4994v1(accessed 5 Jul 2013). kinase induces a p53-dependent metabolic checkpoint MoL CelL 18: Sasada, R., Marcey, D. (2010). ATP Syntl1ase. Available from: http:// Kapralov, P. 0., Artemov, V. G., Leskin, A. A., et al. (2008). On the html#fig1 (accessed 5 Jul2013). possibility of sorting ortho and para water molecules during diffusion Semikhina, L. P., Kiselev, V. F., Levshin, L. V., Salets.kii, A.M. (1988). in nanopores. BulL Lebedev Phys. Insc. 35:221-223. Effect of weak magnetic fields on tl1e luminescence-spectral proper- Kanellis, J., Kandane, R. K., Etemadmoghadam, D., et a!. (2006). ties of a dye in an aqueous solution. J. Appl. Speccros. 48:556-559. Activators of t11e energy sensing kinase AJvlPK inhibit random cell Semiklrina, L. P., Kiselev, V. F. (1981). Effect of weak magnetic fields on movement and chemotaxis in U937 cells. Immunol. Cell Biol. 84: the properties of water and ice. Russ. Plrys. J. 31:5351-5354. Su, D., Heroux, P. (2012). Survey of extra-low frequency and very-low Kim, E. K., },ifiller, 1, Aja, S., et al. (2004). C75, a fatty acid synthase frequency magnetic fields in cell culture incubators. Available from: inhibitor, reduces food intake via hypot11alamic AMP-activated llttp://arxiv.org/abs/1211.2458 (accessed 5 Jul 2013). protein .kinase. J. Biol. Chem 279:1997 19976. Svendsen, A.L., Weihkopf, T., Kaatsch, P., Schtiz, J. (2007). Exposure to Kizaka-Kondoh, S., Inoue, M., Harada, H., Hlraoka, M. (2003). Tumor magnetic fields a11d survival after diagnosis of clri!dllOod leukemia: A hypoxia: A target for selective cancer t11erapy. Cancer Sci 94: german cohort study. Cancer Epidemiol. Biomarkers Prev. 6: Kus!Uler, D. J., Baker, A., Dunstall, T. G. (1999). Pharmacological uses Tikhonov, V. I., Volkov, A. A. (2002). Separation of water into its ortl1o and perspectives of heavy water and deuterated compow1ds. Can. J. and para isomers. Science 296:2363. PhysioL Plw.rmacol. 77:79-88. Vagero, D., Olin, R. (1983). Incidence of ca11cer in tl1e electronics Lai, H., Singh, N. P. (2004). Magnetic-field-induced DNA strand breaks industry: Using the new Swedish Cancer Environment Registry as a in brain cells of the rat. Environ. Healch Perspecc. 112:687-694. screening instrument Br. J. Inti Med 40:188--192. Lee, D., Steffes, M. W., Sjodin, A., et al. (2010). Low dose of some Vandenberg, L. N., Colborn, T., Hayes, T. B., et a!. (2012). Hormones persistent organic pollutants predicts type 2 diabetes: A nested case- and endocrine-disrupting chemicals: Low-dose effects and nonmono- control study. Environ. Healch Perspect 118:1235-1242. tonic dose responses. Endocr. Rev. 33:378--455. Li, Y., Heroux, P., Kyrychenko,1(2012).Metabolic restriction of cancer Viollet, B., Lantier, L., Devin-Leclerc, J., et al. (2009). Targeting the cells in vitro canses karyotype contraction - An mdicator of cancer AMPK pathway for t11e treatment of Type 2 diabetes. Fronc. Biosci. promotion? Tumor BioL 33:195-205. Liu, L., Cash, T. P., Jones, R. G., et al. (2006). Hypoxia-induced energy Wert11eimer, N., Leeper, E. (1979). Electrical wiring configurations and stress regulates mRNA translation and cell growth. Mol. Cell 21: childhood cancer. Am J. EpidemioL 109:273-284. Winder, W. W., Holmes, B. F., Rubink, D. S., eta!. (2000). Activation of },l{ild, K. H., Wilen, J., Mattsson, M. 0., Simko, M. (2009). Background AMP-activated protein kinase increases mitochondrial enzymes in ELF magnetic fields in incubators: A factor of importance in cell skeletal muscle. J. Appl. Physiol. 88:2219-2226. culture work. Cell Biol. Inc. 33:755-757. Winder, W. W., Hardie, D. G. (1999). AMP-activated protein kinase, a },l{i]Jer, A. B., Teresa, T., Agnew, D. A., et a!. (1996). Leukemia metabolic master switch: possible roles in Type 2 diabetes. Am. J. following occupational exposure to 60-Hz electric and magnetic fields PhysioL Endocrinol. Mecab. 277:E1-E10. among Ontario electric utility workers. Am. J. EpidemioL 144: Wu, M., Neilson, A., Swift, A. L., et a!. (2006). Multiparameter metabolic analysis reveals a close link betlveen attenuated mitochon- }, {iller, D. L., Rotl1, M. B. (2007). Hydrogen sulfide increases drial bioenergetic function and enhanced glycolysis dependency in thermotolerulCe and lifespan in Caenorhabdicis elegans. Proc. Nacl. human tumor cells. Am J. PlzysioL Cell. Physiol. 292:C125-C136. Acad. Sci. USA. 104:20618-20622. Zorov, D. B., Juhaszova, M., Yaniv, Y., et al. (2009). Regulation and },l{ilham, S. (2010). Hlstorical evidence that electrification cansed the pharmacology of tl1e mitochondrial permeability transition pore. 20th century epidemic of "diseases of civilization". Med. Hypocheses Cardiovas. Res. 83:213-225. Mitchell, P. (1966). Chemiosmotic coupling in oxidative and photosyn- thetic phosphorylation. Biol. Rev. 41:445-501.

Source: http://www.lifeharmonizer.name/userfiles/file/ELF%20MF%20cancer%20cells.pdf