Cellular stress response to extremely low‐frequency electromagnetic fields (ELF‐EMF): An explanation for controversial effects of ELF‐EMF on apoptosis

Abstract Impaired apoptosis is one of the hallmarks of cancer, and almost all of the non‐surgical approaches of eradicating tumour cells somehow promote induction of apoptosis. Indeed, numerous studies have stated that non‐ionizing non‐thermal extremely low‐frequency magnetic fields (ELF‐MF) can modulate the induction of apoptosis in exposed cells; however, much controversy exists in observations. When cells are exposed to ELF‐EMF alone, very low or no statistically significant changes in apoptosis are observed. Contrarily, exposure to ELF‐EMF in the presence of a co‐stressor, including a chemotherapeutic agent or ionizing radiation, can either potentiate or inhibit apoptotic effects of the co‐stressor. In our idea, the main point neglected in interpreting these discrepancies is “the cellular stress responses” of cells following ELF‐EMF exposure and its interplay with apoptosis. The main purpose of the current review was to outline the triangle of ELF‐EMF, the cellular stress response of cells and apoptosis and to interpret and unify discrepancies in results based on it. Therefore, initially, we will describe studies performed on identifying the effect of ELF‐EMF on induction/inhibition of apoptosis and enumerate proposed pathways through which ELF‐EMF exposure may affect apoptosis; then, we will explain cellular stress response and cues for its induction in response to ELF‐EMF exposure; and finally, we will explain why such controversies have been observed by different investigators.


| INTRODUC TI ON
In modern world, electromagnetic fields (EMFs) have become an inseparable part of routine life. Numerous electric power-generating human-made devices are now producing EMFs which are overlaid on those of earth's magnetic field. EMFs are usually identified with a 50 or 60 Hz frequency and therefore are classified under the extremely low-frequency, non-ionizing span of electromagnetic spectrum. 1 Due to these physical characteristics, ELF-EMFs are not capable of breaking molecular bond or inducing thermal effects on tissue.
However, it is now proven that they can interact with human tissues and induce some weak electrical currents. 2 In addition, it is not completely understood whether biological effects induced by EMFs are hazardous for human or environment. During last few decades, a number of studies have reported beneficial effects of ELF-EMFs in treatment of cancer both in vitro and in vivo. [3][4][5][6][7] Despite this, the exact mechanism of these anti-neoplastic effects has not been confirmed yet.
So far, the most probable mechanism proposed for explaining anticancer effects of ELF-EMF is induction of apoptosis through upregulation of intracellular reactive oxygen species (ROS) which has also been confirmed by different experimental studies. In the study performed by Ding et al., 8  induced caspase-dependent apoptosis following 24-h exposure to 50 Hz, 1 mT ELF-MF in SH-SY5Y neuroblastoma cell lines. 11 One of the main mechanisms proposed for defining anticancer effects of ELF-EMF is induction of apoptosis through upregulation of reactive oxygen species (ROS) which has also been confirmed by different experimental studies.
Contrary to above-mentioned studies, several reports propose an anti-apoptotic activity for ELF-EMF. Pirozzoli et al. 12 reported that 24-h exposure to 50 Hz, 1 mT ELF field significantly attenuated apoptosis induced by camptothecin in LAN-5 neuroblastoma cell lines. De Nicola et al reported that puromycin-induced apoptosis in human lymphoblasts was significantly weakened in response to 2-h exposure to a 0.1 mT ELF field. 13 They reported that reduced glutathione (GSH) was the key mediator of the observed effect. In addition, based on the study performed by Palumbo et al., 14 pretreatment of Jurkat leukaemic cell lines with 50 Hz, 1 mT EMF resulted in 22% reduction in caspase 3-dependent apoptosis induced by anti-Fas therapy. Moreover, based on Cid et al., 15 the antiapoptotic activity of melatonin on HepG2 cell lines was completely abrogated in response to 42-h intermittent exposure with a 50 Hz, 10 µT EMF. Similarly, bleomycin-induced apoptotic activity in K562 erythroleukaemia cell line was significantly reduced in response to a short-term (~10 min) exposure period to a 217 Hz, 120 µT ELF-MF. 16 Based on Brisdelli et al., 17  Still, some reports have stated no statistically significant cytotoxic or cytostatic activity for ELF-EMF. Laqué-Rupérez et al. 18 reported no statistically significant changes in methotrexate-induced cytotoxicity in MCF-7 breast cancer cell lines after exposing them to 25 Hz, 1.5 mT pulsed EMF. Similarly, in the study performed by Mizuno et al., 19 no statistically significant changes in survival rates of SV40 cells were observed between cells which were subjected to UV radiation alone and group subjected to concurrent administration of 24-h 60 Hz, 5 mT EMF and UV radiation. Finally, Höytö et al. 20 reported no statistically significant enhancement in antiproliferative and cytotoxic activities of menadione on SH-SY5Y neuroblastoma cells when combined with 24 h exposure to ELF-MF of 100 µT intensity. This discrepancy in observations has made it difficult to come into a unit conclusion, and therefore, application of ELF-EMF in clinic for treatment of cancer still remains a big dilemma. In our idea, the main point neglected in interpreting the discrepancies observed in results is consideration of cellular stress responses induced by ELF-EMF exposure and its interplay with the molecular mechanisms underlying apoptosis. The main purpose of current review was to outline the triangle of ELF-EMF, cellular stress response of cells and apoptosis, and interpret and unify the discrepancies in results based on this theory. Therefore, initially we will explain studies performed on identifying the effect of ELF-EMF on induction/inhibition of apoptosis, enumerate proposed pathways through which ELF-EMF exposure may affect apoptosis; then, we will explain cellular stress response, cues for activation of this phenomenon in response to ELF-EMF exposure and finally under a separate "discussion" section we will try to explain why such controversy has been obtained by different investigators.

| AP OP TOS IS AND ELF-EMF E XP OSURE
Considering hallmarks of cancer, aberrant cellular survival is an important characteristic of malignant cells which is usually attributed to a mis-regulated apoptotic state in cells. Apoptosis is a type of programmed cell death which is abundantly observed under both physiological and pathological conditions, upon interaction of cells with specific stimulators, capable of activating either of intrinsic and extrinsic pathways. Moreover, failure in induction of apoptosis, as a consequence of aberrant expression of antigens, secreted angiogenic growth factors, or their receptors has already been linked to an elevated risk of metastasis, promotion of angiogenesis and an accelerated risk of resistance development to antiangiogenic cancer therapies. [21][22][23][24][25][26][27] Either mediated by the extrinsic (mediated by FASL, TNFα and so on) or intrinsic pathway (most importantly, accumulation of ROS and development of oxidative stress), the rest of the process will be followed by modulation of specific sets of procaspase molecules cleavage (caspase 8 and caspase 9 for extrinsic and intrinsic pathways respectively), ending in degradation of numerous intracellular target proteins, blebbing of cellular membrane, cleavage and degradation of chromosomal DNA, and finally, getting phagocytosed and scavenged by polymorphonuclear cells. 28 Apoptosis can be triggered upon activation of two main pathways which are broadly referred as "intrinsic" and "extrinsic" pathways.
The most prevalent mechanism through which several chemotherapeutic agents trigger apoptosis is induction of mitochondrial membrane permeabilization, the intrinsic apoptosis pathway, which is mainly controlled by Bcl-2 proteins family. This process results in leakage of several pro-apoptotic molecules such as cytochrome c, Smac/DIABLO, apoptosis-inducing factor (AIF) and endonuclease G (Endo G) into the cytoplasm. 29 Released Endo G and AIF initiate nuclear modifications while the others activate caspases. Cytochrome c promotes formation of apoptosis protease activating factor-1 (Apaf-1) oligomers using ATP or dATP. 30,31 This complex in next place recruits procaspase 9 and forms "apoptosome" which in turn induces autoactivation of procaspase 9. 32,33 Matured caspase 9 further activates caspase 3 and 7 which in turn results in initiation of downstream caspase cascades 34 and induction of apoptotic cell death. In parallel, Smac/DIABLO antagonize suppressing effects of inhibitors of apoptosis proteins (IAPs) on activated caspases. 35,36 In some cell types however, chemotherapeutic-induced apoptotic cell death may be initiated through the death receptor Fas (APO-1/ CD95), the extrinsic apoptosis pathway. Ligation of Fas with its natural ligand, FasL, promotes Fas clustering, which in next place attracts FADD 37 and procaspase 8, 38 totally forming a complex referred as death-inducing signalling complex (DISC). The mature caspase 8 would be exhausted from the DISC after oligomerization and autoactivation of procaspase 8. 39 Based on the cell type, mature caspase 8 initiates apoptosis by two distinct pathways. 40 In first pathway, high quantities of mature caspase 8 induce direct cleavage and activation of procaspase 3 without enrolment of mitochondrial pathway. In second pathway however, low quantities of mature caspase 8 are formed which is not capable of directly inducing activation of procaspase 3. Alternatively, herein, caspase 8 promotes cleavage of the "BH3-only protein" Bid and formation of truncated Bid which, in turn, triggers mitochondrial apoptosis pathway. 41,42 Different groups of anticancer agents are capable of activating death receptor pathway through enhancement of Fas or FasL expression. 43 This process is transcription-dependent and involves p53 activity. 44 The activated signalling pathway following Fas/FasL complexation outlines an autocrine/paracrine pathway like that happening during activation-induced cell death in T lymphocytes.
Nevertheless, FasL plays minimal role in chemotherapy-induced apoptosis, as administration of antagonist antibodies or any small molecule preventing from FasL/Fas interaction does not suppress apoptosis. 45 Likewise, the pro-apoptotic effects of chemotherapeutic agents on embryonic fibroblasts from FADD and caspase 8 knockout mice remained unaltered. 46,47 Although apoptosis is usually induced upon overproduction of ROS and development of oxidative stress, a mild-to-moderate level of ROS is required for maintenance and regulation of physiological function of cells including growth, proliferation, differentiation and migration 52 ; regulation of immune system's function and maintaining redox balance 48 ; and promotion of autophagy through activation of different signalling pathways including phosphoinositide 3-kinase (PI3K)/Akt, mitogen-activated protein kinases (MAPK), nuclear factor (erythroid-derived 2)-like 2 (Nrf2)/Kelch-like ECH-associated protein 1 (Keap1), nuclear factor-κB (NF-κB) and the tumour suppressor p53. [48][49][50][51] Hence, manipulation of ROS level in cells is a good strategy for cancer therapy.
If ROS generation and accumulation can be considered the first cellular event of ELF-EMFs exposure, the modification of intracellular Ca 2+ levels could be one of the most important mechanisms by which ROS have their multiple actions in cells. 52 Over the past few years, lots of data have shown that ELF-EMF exposure regulates intracellular Ca 2+ level which can, in turn, activate multiple physiological mechanisms such as differentiation of chromaffin cells into In most of these studies, the strengths of applied ELF-EMFs were more than 100 µT, and none has investigated the relation be- Although effects of EMF exposure on TGFβ/BMP signalling pathway have been studied during the process of bone repair, same pathway is a key player in pathophysiology of cancer and its modulators demonstrate statistically significant anti-metastatic activities. 75 Different studies have shown that exposure to pulsed EMF results in a statistically significant increase in TGFβ, in osteoblastic cells and both atrophic and non-hypertrophic cells. 76,77 In addition, based on a recent study, exposing differentiating osteoblasts to pulsed EMF, promotes activation of TGFβ signalling pathway through Smad2 and increases expression of osteoblastic differentiation markers such as ALP and type I collagen. 78 BMP expression during osteogenesis was also increased after exposure to pulsed EMFs. [79][80][81] Moreover, it has been shown that exposure to pulsed EMFs, stimulates osteogenic differentiation and maturation through the activation of BMP-Smad1/5/8 signalling. In this case, BMP receptor II, BMPRII, regulates differentiation in a cilium-dependent manner. 82 Considering the separate effects of BMP and pulsed EMFs on differentiation and maturity of osteoblasts, many studies have shown that concurrent treatment with BMP and pulsed EMF enhances bone formation to a much greater degree compared to each treatment alone. [83][84][85][86] In addition to the mentioned signalling pathways, electromagnetic fields can also affect pathways underlying VEGF and FGF signalling molecules. 87,88 Based on a recent report, exposure to pulsed EMF significantly increases expression of IGF-1 at mRNA level and promotes bone formation. 89 In addition, pulsed EMF (1.5 mT, 75 Hz) can also increase synthesis of proteoglycans and protect human articular cartilage from further damage. 90 Finally, it has been shown that exposure to pulsed EMF reverses osteoporotic effect of dexamethasone. 91 Notch signalling is a highly conserved pathway that regulates cellular fate and skeletal development. Recent reports have shown that exposure to pulsed EMF can regulate expression levels of Notch4 receptor, as well as DLL4 ligands and target genes (Hey1, Hes1 and Hes5) during the osteogenic differentiation of human mesenchymal stem cells. Interestingly, expression of osteogenic markers, including Runx2, Dlx5, Osterix, Hes1 and Hes5, after pulsed EMF treatment was reversed following treatment of cells with notch pathway inhibitors. 92 Furthermore, exposure to pulsed EMF significantly increases the level of cAMP, protein kinase A activity and accelerates osteogenic differentiation of MSCs. 93,94 Anti-inflammatory effects of pulsed EMFs have also been reported both in vitro 95,96 and in vivo, [97][98][99] as well as in clinical settings. 100

| ELF-EMF AND INDUC TI ON OF CELLUL AR S TRE SS RE S P ON S E
Numerous studies have shown that cells are physiologically well buffered against negative effects of ELF-EMF alone. However, in the presence of stressful condition, including exposure to toxins, viruses, DNA damage and proteotoxic, hypoxic, metabolic and oxidative stress, an additional weak stressor like ELF-EMF might produce large effects. 101 Based on Mattsson and Simko who extensively investigated the oxidative response of cells following ELF-EMF exposure, ROS levels can be consistently altered in different cell types or experimental conditions following exposure to magnetic fields. These effects were prominent for fields with intensities more than 1 mT, but were also documented at or below 100 mT. Despite this, all observed effects where moderate and majority of changes were below 50%. 102 Consequently, the pro- As discussed earlier, antioxidant defence capacity of cells can be changed following exposure to ELF-EMF. For example, it has been shown that exposure to ELF-EMF can significantly increase SOD levels in cells. 107 Furthermore, ELF-EMF can enhance activity of both glutathione-S-transferase and -reductase enzymes in malignant cells. 108 Also, based on Cichon et al., 109  Generally, cellular stress response is characterized by modulation of expression of various genes. The main outcome of this alteration in pattern of gene expression is protection of cells from cytotoxic doses of a harmful agent. This response represents that following exposure to a toxin, cells expect or at least prepare themselves for a lethal concentration of the agent. In addition to mild exposure to toxic agents or stressful conditions, physiological conditions may also promote development of an cellular stress response. 117 For instance, exercise training reduces the extension of lipid peroxidation during acute exercise which has been attributed to induction of oxidative stress. 118,119 Likewise, an enhanced repairing capacity was observed in lymphocytes of workers which were occupationally become exposed with low levels of ionizing radiation. 120 It is now clear that sub-lethal doses of oxidants are capable of inducing cellular stress responses in cells. This phenomenon was initially discovered in bacteria, but now it has also been documented in eukaryotic cells. 121 (Table 1). Here, we will comprehensively review the ways through which cells respond to elevated ROS following exposure to ELF-EMF and orchestrate cellular stress response ( Figures   1 and 2).

| Heat-shock response
In most eukaryotes, heat-shock factors (HSF) [ie transcription factors that regulate expression of heat-shock proteins (HSPs)] are located in cytoplasm in bond with HSP70, HSP90 or other proteins which renders them to be inactive during normal condition. 125,126 During stressful condition however, cells are exposed to a much higher extent of denatured proteins. In this condition, as HSPs prefer to act more like a molecular chaperone instead of a regulatory protein, they become detached from HSF and undergo oligomerization. In next step, oligomerized HSFs translocate to the nucleus where they promote expression of HSP and related heat-responsive genes. 127,128 Different studies have shown that treatment of cells with H 2 O 2 and induced ROS can increase expression of heat-responsive genes. [129][130][131] In the study performed by Volkov et al. 128  can also easily pass through membrane and therefore take role of a signalling molecule. 135

| Unfold protein response
In order to function properly, proteins require a specific threedimensional folding. 148,149 This unique structural folding is mainly stabilized through intramolecular disulphide bonds particularly for

| Autophagic response
Observed during nutrition starvation or related stressful metabolic condition, autophagy is defined as a catabolic event through which cellular components are degraded and recycled following transportation in to lysosomes via specific bilayer structures named autophagosomes. 163 It has been shown that other stressful conditions such as hypoxia or oxidative stress are also capable of inducing autophagy. Likewise, several chemotherapy agents including arsenic trioxide and oxaliplatin are capable of inducing autophagy. 164,165 PI3K type III-Atg6/Beclin 1 complex is responsible for initiation of autophagosomes nucleation while Atg12-Atg5 and Atg8/LC3phosphatidylethanolamine conjugates monitor the process of autophagosomes elongation. These two processes are considered as the main characteristics of autophagy. 163,166,167 It is noteworthy to mention that while accumulation of a large body of autophagic vacuoles may result in initiation of autophagic cell death, a controlled autophagic response guarantees physiological recycling of damaged organelles and biomacromolecules to cope with energy demands following exposure to cytotoxic drugs or a stressful condition. [168][169][170] As mentioned above, ROS accumulation can trigger initiation of

| NF-kB inflammatory response
Based on different studies, during oxidative stress, activation of NF-κB transcription factor can protect cells from injury through inhibition of ROS accumulation. Suppressing activation of NF-κB has shown to be together with an enhancement in TNFα-induced ROS generation, oxidation of proteins and peroxidation of lipids. 174 In addition, NF-κB can also modulate activation of autophagy which is another efficient protective mechanism against oxidative stress.
Exposing retinal pigment epithelial cells to different concentrations of H 2 O 2 , a potent inducer of oxidative stress, resulted in phosphorylation of p65 subunit of NF-κB which, in turn, promoted upregulation of p62 which is a potent promoter of autophagy. 175 ROS can also delay inactivation of JNK pathway by inhibiting phosphatases responsible for inactivating JNKs. This is mainly mediated through conversion of the catalytic cysteine of these enzymes to sulfenic acid. 176 Studies have shown an increase in TNF-alpha-mediated apoptosis in response to a decline in NF-κB-mediated inhibitory effect on JNK activation. 177 Therefore, it has been proposed that the anti-apoptotic effect of NF-κB may be partly mediated through suppression of JNK pathway activation, in which ROS maybe the bridging molecule. 178 Reported by Wu et al., 179

| ANTI -/PRO -AP OP TOTI C EFFEC TS OF ELF-EMF: S TATEMENT OF THE CONTROVER SY
As discussed above, a vast variety of densities ranging from a few micro Tesla up to tens of milli tesla have been applied in studies examining effects of ELF-EMF on apoptosis. [184][185][186] In addition, in some cases, promotion of apoptosis by ELF-EMF exposure was examined in the presence of a co-stressor (eg chemotherapeutic agents). One obvious result of these studies is that contrary to chemotherapeutic agents, ELF-EMF exposure does not demonstrate a clear doseresponse pattern. In another words, increase in intensity of magnetic fields does not necessarily result in enhancement of apoptosis or other biological effects. Also, no threshold can be considered for induction of ELF-EMF biological effects. Despite this, another conclusion from reported data is that very low magnetic flux densities and exposure time are also enough for induction of biological responses.
Based on these facts, the differential responses to magnetic fields cannot be attributed to the exposure conditions. Instead, biological state of the experiment including the studied cell type or animal tis- Ca 2+ can be only consistently detected when biological characteristics of system including pH of the environment, cell cycle phase and response to a Ca 2+ agonist were specifically determined. Likewise, several other investigators have also listed some specific criteria necessary to be fulfilled in order to record a consistent response to ELF-EMF exposure. [188][189][190] In situations where a co-stressor such as treatment with a chemotherapeutic agent is also exist, prediction of results becomes more complex. One of the main points, which must be considered additionally in this context, is that how the study has been scheduled. In another words, how is the order of treating with ELF-EMF exposure and the co-stressor. In order to make discussion easier, in this section, we only focus on the controversies in apoptosis induced by ELF-EMF exposure at the presence of another stressor and will not consider the discrepancies related to the biological nature of the materials in the study. In next few paragraphs, we will classify studies based on their experimental design, propose our theory and enumerate the cues in support of the study.  and melatonin 15 protected cancer cells from pro-apoptotic effects of these agents. Other important finding of these studies was that exposure to ELF-EMF per se is not enough for induction of apoptosis.

| Theory: Cellular stress response to ELF-EMF protects cells from chemotherapy-induced apoptosis
As Numerous data exist in support of this theory. For instance, when SH-SY5Y neuroblastoma cells were concurrently exposed to H 2 O 2 and ELF-EMF with intensity of 1 mT and frequency of 50 Hz for 24 h, the increase in catalase was significantly restricted. 192 Contrarily, reports have shown that exposure to magnetic fields with above-mentioned characteristics alone can induce expression of cytochrome P450 (CYP450) and glutathione S-transferase (GST), both of which play key roles in cellular detoxification process. 192,193 Reported by Patruno et al., 194  activity and improves reduced glutathione's availability. 195,196 This is important, as GSH is a vital co-factor for GPX and several other enzymes involved in phase II drug metabolization. [197][198][199] Recently, it has been shown that long-term exposure to ELF-EMF before exogenous treatment with methylglyoxal (MG) significantly reduces susceptibility of cancer cells to cytotoxic effects of this agent. 195 This has been mainly attributed to the enhanced accessibility of GSH following ELF-EMF exposure which is an important co-factor for directing MG into the glyoxalase-mediated detoxifying system. 200 Recently, it has also been shown that ELF-EMF is capable of inducing sirtuin 3 (SIRT3) expression. 196 The signalling cascade mediated by SIRT3 is capable of improving mitochondrial integrity and fitness following exposure to oxidative proteotoxic stress. 201 206 Contrarily, when cells are exposed to ELF-EMF, a new source of ROS production is introduced in cells F I G U R E 3 A schematic illustration of the hypothesis for explanation of controversial effects of ELF-EMF on apoptosis. Upper side: ELF-EMF exposure prior to treatment with the apoptosis-inducing agent will result in activation of cellular defence system and alteration in expression of a number of genes which, in next place, will end in promotion of DNA repair system, ROS detoxification system and Ca 2+ homeostasis through production of new protective proteins and antioxidative enzymes or restoration of antioxidative stress molecule reservoirs such as glutathione and so on. In next place, upon introduction of the apoptosis-inducing agent, cells will defend themselves with robust protective system and consequently, lower rate of apoptosis will occur. Lower side: Contrarily, ELF-EMF co-treatment with or immediately after chemotherapeutic agent will enhance the rate of injury by ROS overproduction or unbalancing Ca 2+ homeostasis which will end in promotion of apoptosis which can at least partially reverse anticancer effects observed with cell's treatment with melatonin.
Camptothecin is a unique chemotherapeutic agent which induces apoptosis at S phase of cell cycle through inhibition of topoisomerase I. 207 Thus, depending on the doubling time of the cells, a specific time period is required for initiation of anticancer effects of camptothecin which is usually about 24 h. Thus, when cells are simultaneously exposed to ELF-EMF and camptothecin, during the first 24 h, cells have enough time to undergo cellular stress response prior to initiation of action of camptothecin, and therefore, it is not surprising to observe protective effects against apoptosis. However, when the exposure time extends to 48 h, camptothecin is completely active and have produced excess ROS levels through which induced cellular stress response is not capable of coping with it, and therefore, extra ROS produced by ELF-EMF can help in promotion of apoptosis.
Puromycin is a specific chemotherapeutic agent which is capable of inducing cell death through inhibition protein synthesis, accelerating accumulation of misfolded proteins and induction of apoptosis. 208 Grassi et al. 209 have shown that pre-exposing cells to ELF-EMF can significantly reduce apoptosis induced by puromycin which is consistent with the cellular stress response theory. However, the study by Kaszuba-zwoinska et al. 10 also showed that simultaneous treatment with ELF-EMF and puromycin can also protect from apoptotic effect. As protein synthesis process and accumulation of unfolded proteins in cells also requires a time period which is usually about 10-12 h, it can also be concluded that cells during concurrent treatment also have enough time for adaption. It is also noteworthy to mention that observed protective effect in this study was very low and about 5%-10% in its highest point. 10

| CON CLUS I ON AND FUTURE PER S PEC TIVE
As discussed herein, cellular stress response is a unique behaviour of cells following exposure to ELF-EMF which helps them to cope with next more stressful encountered conditions. This response is mainly due to the mild increase in cellular ROS levels mediated by ELF-EMF exposure which is not capable of inducing apoptosis alone. The statement discussed in present article, if proved to be true, will become very important in designation of new therapeutic schedules for treatment of cancer as concurrent ELF-EMF exposure/chemotherapy or ELF-EMF exposure immediately following chemotherapy can significantly improve pro-apoptotic effects of chemotherapeutic agents. Furthermore, following this hypothesis, one can apply ELF-EMF in treatment of resistant cancers as one of the main mechanisms of resistance to chemotherapeutic agents is high capacity of these cells in scavenging free radicals.
Additive effects of ELF-EMF to chemotherapeutic agents in induction of an oxidative stress condition may be helpful in this context. More importantly, ELF-EMF exposure to normal cells in most cases has shown to be safe and un-harmful. Therefore, ELF-EMF therapy may not pose any other adverse effects except for those observed with potentiation of chemotherapeutic agents' cytotoxic effects. As discussed herein, determination of ELF-EMF's intensity window is also very important as no response occurs outside this range, making it a highly personalized therapy. Finally, although this hypothesis apparently sounds rational, future studies comparing results of ELF-EMF exposure before, during and immediately after chemotherapy is highly recommended for further confirmation of this theory.

ACK N OWLED G EM ENT
The authors highly acknowledge Prof. Mats-Olof Mattsson and Prof.
Myrtill Simko for their expert review and opinion about this manuscript and their valuable commentary which has been used to improve the consistency of this paper.

CO N FLI C T O F I NTE R E S T
Authors declare that they have no conflict of interest.

MB, MAJ and AM drafted the main body of the manuscript; BD
and MRE modified the manuscript and extended some sections and also designed illustrations; SPS and AMA conceived the original idea of the manuscript; and AMA furthermore supervised the team.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.