Extremely low-frequency pulses of faint magnetic eld, weaker than the geomagnetic eld, induce mitophagy to rejuvenate mitochondria

Humans are frequently exposed to time-varying and static weak magnetic elds (WMF). However, the effects of faint magnetic elds, weaker than the geomagnetic eld, have not been reported. We found that extremely low-frequency (ELF)-WMF, comprised of serial pulses of 10 µT intensity at 1–8 Hz, which was three or more times weaker than the geomagnetic eld, reduced mitochondrial mass to 70% and the mitochondrial electron transport chain (ETC) complex II activity to 88%. Chemical inhibition of electron ux through the mitochondrial ETC complex II nullied the effect of ELF-WMF. Suppression of ETC complex II subsequently induced mitophagy by translocating parkin and PINK1 to the mitochondria and by recruiting LC3-II. Thereafter, mitophagy induced PGC-1α-mediated mitochondrial biogenesis to rejuvenate mitochondria. The lack of PINK1 negated the effect of ELF-WMF. Thus, ELF-WMF may be applicable for the treatment of human diseases that exhibit compromised mitochondrial homeostasis, such as Parkinson’s disease. of 0 to 300 µT at time-varying frequencies of 1–16 Hz every second. We used a 10 µT magnetic eld of 4 m pulse width with increasing frequencies of 1, 2, 3, 4, 5, 6, 7, and 8 Hz every second (Opti-ELF-WMF), unless indicated otherwise. This condition maximizes the hysteresis of the electronic resistance of pure water 11, 12 . Before and after each experiment, we conrmed the intensity of the magnetic ux using a pulse magnetic eld meter (Aichi Micro Intelligent). To reduce the effects of an electromagnetic eld generated by an incubator and the geomagnetic eld, the ELF-WMF device and a culture dish were sandwiched by two 5-mm thick copper plates, and were placed in a humidied incubator with 5% CO 2 at 37°C. The basal electromagnetic eld was below 20 µT (200 mG) on the culture dish. Control samples were incubated in parallel under same in another incubator without ELF-WMF. activities


Introduction
In the present-day industrialized societies, humans are exposed daily to time-varying and static weak magnetic elds (WMF). The effects of WMF on animals, including humans, have been documented in a few reports. WMF increases intracellular calcium concentrations and induces the development of satellite cells 1 . The viability of breast cancer cells is speci cally decreased by WMF 2 . Extremely low-frequency WMF (ELF-WMF), which is de ned as ELF with a frequency of 300 Hz or less, may or may not reduce the levels of reactive oxygen species (ROS) in cells 3 . ROS are mostly produced during electron transfer through the mitochondrial electron transport chain (ETC). The in cellulo or in vivo effects of faint magnetic elds, weaker than the geomagnetic eld, have not been reported. In addition, the optimal conditions for manifestation of the cellular effects of ELF-WMF remain undetermined. Furthermore, the molecular mechanisms underlying the effects of ELF-WMF have not been elucidated.
Mitophagy and mitochondrial biogenesis cooperate in the maintenance of mitochondrial homeostasis. Mitophagy is a quality-assurance system that selectively eliminates damaged mitochondria using the macroautophagy machinery. Mitophagy-associated proteins include PTEN-induced kinase 1 (PINK1), parkin, BCL2/adenovirus E1B 19 kDa protein-interacting protein 3 (BNIP3), BCL2/adenovirus E1B 19 kDa protein-interacting protein 3-like (NIX/BNIP3L), and FUN14 domain containing 1 (FUNDC1). Mitophagy is induced by mitochondrial damage, excessive levels of mitochondrial ROS, endoplasmic reticulum (ER) stress 4 , circadian rhythm 5 , and hypoxia-inducible factor 6 . In addition, mitophagy induces mitochondrial biogenesis to compensate for the removal of mitochondria 7,8 . The parkin/PINK1 pathway is a key regulator of mitophagy. Parkin and PINK1 translocate from the cytosol to the mitochondria, and ubiquitinate the mitochondria, which are subsequently recognized by the phagosome-lining LC3 to eliminate the mitochondria. Accumulating knowledge points to the notion that the compromised parkin/PINK1 pathway is associated with the development and progression of neurodegenerative diseases, including Parkinson's disease 9 and Alzheimer's disease 10 .
In this study, we investigated the effects of faint magnetic elds, weaker than the geomagnetic eld, on cells. We report that ELF-WMF e ciently suppresses the mitochondrial mass to 70% by inhibiting the mitochondrial ETC complex II, which subsequently induces mitophagy and rejuvenates mitochondria. We expect that ELF-WMF may be applicable to a plethora of human diseases that exhibit compromised mitophagy like neurodegenerative diseases.

Materials And Methods
The ELF-WMF apparatus The ELF-WMF device was manufactured by Mr. Kota Okada at the Technical Center of Nagoya University, Japan. The device had a round coil (1 cm height, 10 cm inner diameter, and 10.7 cm outer diameter, 50 turns of copper wire, and 0.29 mm diameter). It generated 1 -16 ms pulsed magnetic elds of 0 to 300 µT at time-varying frequencies of 1-16 Hz every second. We used a 10 µT magnetic eld of 4 m pulse width with increasing frequencies of 1, 2, 3, 4, 5, 6, 7, and 8 Hz every second (Opti-ELF-WMF), unless indicated otherwise. This condition maximizes the hysteresis of the electronic resistance of pure water 11,12 . Before and after each experiment, we con rmed the intensity of the magnetic ux using a pulse magnetic eld meter (Aichi Micro Intelligent). To reduce the effects of an electromagnetic eld generated by an incubator and the geomagnetic eld, the ELF-WMF device and a culture dish were sandwiched by two 5-mm thick copper plates, and were placed in a humidi ed incubator with 5% CO 2 at 37°C. The basal electromagnetic eld was below 20 µT (200 mG) on the culture dish. Control samples were incubated in parallel under the same conditions in another incubator without ELF-WMF.
Exposure of wild-type mice to ELF-WMF All the studies on mice were approved by the Animal Care and Use Committee of Nagoya University, and were conducted in accordance with the relevant guidelines. Seven-week-old C57BL6/N male mice were purchased from Japan SLC. Two ELF-WMF devices were placed in tandem beneath the mouse cage, and mice were housed in the cage with switch on (ELF-WMF group, n = 4) or off (control group, n = 4) for 4 weeks. The Opti-ELF-WMF condition stated above was applied to the ELF-WMF group.
Test for open-eld locomotor activity in mice Open-eld locomotor activity was evaluated using a photometric actimeter (45 cm × 45 cm, IR Actimeter, Panlab). Fast and slow movements were monitored with a grid of infrared beams every 30 min for 24 h and were used as indices for locomotor activity. To examine the effects of ELF-WMF on the locomotor activity in mice, fast and slow movements were measured before (0 week) and after (4 weeks) exposure. All data were collected using the SEDACOM software (Panlab). Each mouse was tested individually and had no contact with other mice.
Isolation of mitochondria from the mouse liver Mitochondria were isolated from the mouse liver, as described previously 13  2) using a homogenizer. The homogenate was centrifuged at 800 × g for 10 min at 4°C. The supernatant was then centrifuged at 8,000 × g for 10 min at 4℃. The pellet was suspended in the mitochondrial isolation buffer to obtain the mitochondrial fraction.
TMRM assay of the isolated mitochondria The mitochondrial membrane potential of isolated mitochondria was analyzed using TMRM (T668, Thermo Scienti c), following the procedure described previously 14 . Brie y, the isolated mitochondria were incubated with 100 nM TMRM in the mitochondrial isolation buffer for 30 min at 37˚C in a humidi ed incubator. The signal intensities of TMRM were quanti ed using BD FACS Calibur (BD Biosciences).

Measurement of basal oxygen consumption rate of the isolated mitochondria
The basal oxygen consumption rate (OCR) of the isolated mitochondria (20 μg of mitochondrial proteins per well) isolated from the mouse liver was determined using the Seahorse XFp Extracellular Flux Analyzer (Agilent Technologies). The assay was conducted as described previously 15 . Western blot analysis of cell lysates Cells were lysed in PLC buffer containing 50 mM HEPES (pH 7.0), 150 mM NaCl, 10% glycerol, 1% TritonX-100, 1.5 mM MgCl 2 , 1 mM EGTA, 100 mM NaF, 10 mM sodium pyrophosphate, 1 μg/μl aprotinin, 1 μg/μl leupeptin, 1 μg/μl pepstatin A, and 1 mM PMSF. The cell lysates were rotated at 4°C for 20 min and centrifuged at 17,900 × g at 4°C for 15 min. The supernatant was incubated at 37°C for 1 h to analyze the mitochondrial ETC complex proteins or at 95°C for 5 min to analyze other proteins in the sample buffer (62.5 mM Tris-HCl pH 6.8, 2% SDS, 10% glycerol, 0.005% bromophenol blue, and 2% 2-mercaptoethanol). For LC3-II, the lysates were separated by Tricine-SDS-PAGE on a 16% polyacrylamide gel 17 . For the other proteins, the lysates were separated by Tris-SDS-PAGE on a 10%, 12%, or 14% SDS-polyacrylamide gel. The samples were then transferred to a polyvinylidene uoride membrane (Immobilon-P, Millipore). Membranes were washed in Tris-buffered saline containing 0.05% Tween 20 (TBS-T) and blocked for 1 h at 24°C in TBS-T with 5% skimmed milk. The membranes were then incubated overnight at 4°C with speci c antibodies, as indicated below. The Preparation of mitochondrial and cytosolic fractions of cultured cells A mitochondria isolation kit (ab110170, Abcam) was used for the extraction of mitochondrial and cytosolic fractions according to the manufacturer's protocols. After obtaining the mitochondrial fraction by centrifugation at 12,000 × g for 10 min at 4°C, the supernatant was used as the cytosolic fraction.

Detection of mitophagy in cultured cells
To detect mitophagy in AML12 cells, the Mitophagy Detection Kit (Dojindo Molecular Technologies) was used according to the manufacturer's protocol. Brie y, Mtphagy Dye (Dojindo) dissolved in HBSS at 100 nM was added to the cells and incubated for 30 min at 37°C in a humidi ed incubator. After incorporation of Mtphagy Dye into AML12 cells, the cells were exposed to ELF-WMF for 120, 150, and 180 min, washed with PBS, trypsinized, resuspended in PBS, and harvested. Signal intensities of Mtphagy Dye were quanti ed using BD FACS Calibur (BD Biosciences). To visualize both mitophagy and lysosomes after exposure to ELF-WMF, Lyso Dye (Dojindo) dissolved in HBSS at 1 μM was also added to the cells and incubated for 30 min at 37°C in a humidi ed incubator. Images were obtained using a confocal microscope TiE-A1R (Nikon).
Enzyme assay for mitochondrial ETC complex (I, II, III, IV) activities of the mouse liver homogenates Mitochondrial ETC complex activities were measured using homogenates of the liver excised from C57BL/6N mice. The ETC complex activity assay was performed as previously described 18 . Brie y, the protein concentration of each sample was measured using a Pierce 660 nm protein assay reagent. The ETC complex activities of complexes I, II, III, and IV were estimated by determining the decrease in absorbance of NADH at 340 nm, the decrease in absorbance of 2, 6-dichlorophenolindophenol (DCPIP) at 600 nm, the increase in absorbance of reduced cytochrome c at 550 nm, and the decrease in absorbance of reduced cytochrome c at 550 nm, respectively, with NanoDrop ONE C (Thermo Scienti c). The ETC complex activities were measured in the liver homogenate of mice that were exposed to either ELF-WMF or control in vivo. Similarly, the ETC complex activities were measured in the mouse liver homogenates before and after exposure to ELF-WMF for 10 min or less in vitro.
Enzyme assay for mitochondrial ETC complex II activity of the mitochondria isolated from the mouse liver Mitochondria (10 μg protein) isolated from the liver of C57BL/6N mice were used to measure the mitochondrial ETC complex II activities (Fig. 6E).
Assays for measuring the fractional and extended mitochondrial ETC complex II activities of SQR, SCR, and SDH were performed as previously described 18, 19 . The mitochondrial ETC complex II activities were quanti ed before and after exposure to ELF-WMF for 8 min in vitro. The SQR, SCR, and SDH activities were measured by determining the decrease in absorbance of DCPIP at 600 nm, the increase in absorbance of reduced cytochrome c at 550 nm, and the decrease in absorbance of DCPIP at 600 nm, respectively, using NanoDrop ONE C (Thermo Scienti c). The SDHA activity was quanti ed by modifying a method used to measure the SDH (SDHA and SDHB) activity. For measuring the SDHA activity, 10 μg of sonicated mitochondrial fraction was resuspended in 35 mM phosphate buffer (pH 7.3) supplemented with 0.3 mM KCN (Wako), 10 μg/ml antimycin A (Sigma), 4 mM succinate (Wako), 1.6 mM PMS (Sigma), and 40 μM DCPIP (Sigma). The SDHA activity was quanti ed before and after exposure to ELF-WMF for 8 min by determining the decrease in absorbance of DCPIP at 600 nm with NanoDrop ONE c (Thermo Scienti c).
Measurement of mitochondrial mass in AML12 cells exposed to an inhibitor of mitochondrial ETC complex I or II AML12 cells were cultured either with variable concentrations of rotenone (Tokyo Chemical Industry Co.), an inhibitor of mitochondrial ETC complex I, or 3-nitropropionic acid (Cayman Chemical), an inhibitor of mitochondrial ETC complex II, for 12 h. The cells were then exposed to Opti-ELF-WMF for 3 h. Mitochondrial mass was measured by MitoTracker Green, as described above.

RNA-sequencing and GSEA of AML12 cells
Total RNA was extracted from AML12 cells exposed to OPTI-ELF-WMF for 1 h using QuickGene-Mini80 (Kurabo) according to the manufacturer's instructions. The extracted RNA was subjected to RNA-seq at Macrogen, Japan. Brie y, a sequencing library was prepared using the TruSeq Stranded mRNA kit (Illumina), and the library was read on an Illumina NovaSeq 6000 (150 bp paired-end reads). GSEA was conducted with the GSEA v4.1.0 software for Windows (https://www.gsea-msigdb.org/gsea/downloads.jsp) using the RNA-seq dataset. RNA-seq data were deposited in the gene expression omnibus (GEO) with an accession number GSE166811.

Statistical analysis
All values are presented as the mean ± SEM. For in cellulo studies, values were normalized to those of control cells, unless indicated otherwise. Statistical signi cance was estimated either by Student's t-test, one-way ANOVA followed by Dunnett's posthoc test, or false discovery rate of multiple Student's t-tests. P-values less than 0.05 were considered statistically signi cant.

Results
Exposure to Opti-ELF-WMF for 4 weeks increases the mitochondrial activity in the mouse liver Based on our previous observation that 1-8 Hz stimulation of 4 ms pulses of 10 µT magnetic eld (Opti-ELF-WMF) increases the hysteresis of electric resistance of pure water 11, 12 , we examined the effect of 4-week exposure to Opti-ELF-WMF on the liver mitochondria in wild-type C57BL6/N mice.
Using an open-eld locomotor test, we rst con rmed that exposure to Opti-ELF-WMF had no effect on the locomotor activity in mice (Fig. S1A, B). We examined the ETC activity of mitochondria isolated from the mouse liver by measuring the oxygen consumption rate (OCR) and mitochondrial membrane potential using a ux analyzer and tetramethylrhodamine (TMRM), respectively. We found that Opti-ELF-WMF increased both the OCR and mitochondrial membrane potential by approximately 40% (Fig. 1A, B).
Next, we evaluated the enzymatic activity of each ETC complex and the amount of OXPHOS proteins in the mouse liver homogenates. Opti-ELF-WMF increased the activities of mitochondrial ETC complexes I to IV, although statistical signi cance was observed only in complex IV (Fig. 1C). The levels of four nucleus-encoded proteins (NDUFB8 [complex I], SDHB [complex II], UQCRC2 [complex III], and ATP5F1A [complex V]) were also tended to be increased (Fig. 1D). In contrast, the levels of mitochondria-encoded MTCO1 [complex IV] remained unchanged.
These data demonstrate that Opti-ELF-WMF had no effect on the locomotor activity in wild-type mice, but tended to increase the mitochondrial ETC complexes activity and the levels of nucleus-encoded ETC proteins in the mouse liver.
Opti-ELF-WMF temporarily decreases the mitochondrial ROS levels, mitochondrial mass, and mitochondrial membrane potential in cultured cells To further dissect the effect of Opti-ELF-WMF on the mitochondria, AML12 cells were cultured under Opti-ELF-WMF for 1 to 24 h, and were stained with MitoSOX, MitoTracker Green, and TMRM to quantify the levels of mitochondrial superoxide, mitochondrial mass, and mitochondrial membrane potential, respectively. Opti-ELF-WMF decreased the level of mitochondrial superoxide, mitochondrial mass, and mitochondrial membrane potential at 1, 3, and 6 h after treatment, respectively, and increased them at 12 h ( Fig. 2A, B, C). At 24 h, the values reverted to normal levels. Thus, Opti-ELF-WMF suppressed the mitochondrial ETC activity at 1 h, which was likely followed by elimination and/or inactivation of a subset of mitochondria at 3 to 6 h. The mass and function of mitochondria were then increased at 12 h and returned to normal levels at 24 h.
Optimal conditions of ELF-WMF for the reduction of the mitochondrial mass in cultured cells Next, we analyzed the optimal conditions of ELF-WMF that would reduce the mitochondrial mass by changing the intensity, pulse width, and frequency of ELF-WMF. ELF-WMF, less than 10 µT, showed MF strength-dependent reduction in the mitochondrial mass, but the effects were not enhanced when the MF strength ranged from 10 to 200 µT (Fig. S2A). However, compared to 10 µT, 300 µT ELF-WMF had a marginally reduced effect. ELF-WMF with pulse widths of 2, 4, and 8 ms reduced the mitochondrial mass, with a peak at 4 ms (Fig. S2B). In contrast, ELF-WMF with pulse widths of 1 and 16 ms had no effect. The mitochondrial mass was reduced the most upon treatment with increasing frequencies of ELF-WMF (1, 2, 3, 4, 5, 6, 7, and 8 Hz for 1 s each) (Fig. S2C). Static frequencies at 6 and 8 Hz had no effect. Similarly, changing the frequency pro les to 1-4 Hz for 1 s each or to 1-16 Hz for 1 s each had no effect.
As observed for AML12 cells (Fig. 2B, C), Opti-ELF-WMF rst reduced the mitochondrial mass, and thereafter increased the mitochondrial membrane potential in Neuro2A, C2C12, human iPS, HEK293, and HeLa cells (Table 1). Thus, the effects of Opti-ELF-WMF on mitochondria are unlikely to be cell line-speci c.

Decrease in the mitochondrial mass by Opti-ELF-WMF is accounted for by temporary decreases in the levels of mitochondrial ETC proteins and of outer membrane proteins
To identify the mitochondrial proteins that were decreased by the Opti-ELF-WMF exposure, we quanti ed the amounts of mitochondrial ETC proteins and VDAC1, which is an outer membrane protein, by western blot analysis. Opti-ELF-WMF had no effect on the levels of the eight examined mitochondrial proteins at 1 h (Fig. S3), but decreased them at 3 h (Fig. 3). At 12 h, the amounts of ve proteins (NDUFB8, SDHB, UQCRC2, MTCO1, and VDAC1) were restored to their basal levels, and those of three proteins (ATP5A, NDUFS1, and UQCRFS1) were increased compared to their basal levels. Taken together, Opti-ELF-WMF decreased the levels of all the examined mitochondrial proteins at 3 h, which was consistent with the decreased mitochondrial mass at 3 h after exposure (Fig. 2B). The levels of mitochondrial proteins were restored to normal or higher than normal levels at 12 h.

Opti-ELF-WMF induces mitophagy
Mitophagy is one of the autophagic mechanisms in the mitochondrial quality assurance system that eliminates damaged mitochondria. We investigated whether mitophagy is activated by Opti-ELF-WMF. We rst examined the expression levels of mitophagy-related proteins, PINK1 and LC3-II, in whole cell lysates of AML12 cells. PINK1 triggers mitophagy, whereas LC3-II is an effector that eliminates the mitochondria. The amount of PINK1 gradually increased until 90 min and gradually decreased thereafter upon exposure to Opti-ELF-WMF (Fig. 4A). Similarly, the amount of LC3-II gradually increased until 120 min and gradually decreased thereafter (Fig. 4A). These results indicated that the decrease in mitochondrial mass by Opti-ELF-WMF was likely due to the activation of mitophagy.
Next, we evaluated the expression levels of PINK1 and parkin, and mitochondrial ubiquitination at 120 min in the mitochondrial and cytosolic fractions. PINK1 and parkin translocate from the cytosol to the mitochondria, and ubiquitinate them. Opti-ELF-WMF increased the level of parkin in the mitochondria, but had no effect in the whole cells (Fig. 4B). In contrast, Opti-ELF-WMF increased the levels of PINK1 in both mitochondria and whole cells (Fig. 4B). We also found that ELW-WMF induced the ubiquitination of mitochondrial proteins (Fig. 4B). These results indicated that Opti-ELF-WMF accumulated PINK1 and parkin in the mitochondria, and induced mitochondrial ubiquitination. To detect the mitochondria in the lysosomes, we used the Mtphagy Dye that uoresces with decreasing pH around mitochondria. The uorescence intensity of the Mtphagy Dye peaked at 150 min (Fig. 4C).
We also con rmed the colocalization of mitochondria and lysosomes at 150 min by confocal microscopy (Fig. 4D).
To further con rm the effect of PINK1 on Opti-ELF-WMF-induced mitophagy, we examined the mitochondrial mass and the amounts of ATP5A and VDAC1 in PINK1-knocked out (KO) HeLa cells 16 . As expected, exposure of PINK1-KO HeLa cells to Opti-ELF-WMF failed to reduce the mitochondrial mass (Fig. S4A), or the levels of mitochondrial proteins (Fig. S4B).
PGC-1α expression is upregulated for mitochondrial biogenesis after mitophagy PGC-1α is a key player in mitochondrial biogenesis. PPARα and TFAM are regulated by PGC-1α, and are effectors of mitochondrial biogenesis and metabolism. Thus, we examined whether the recovery of mitochondrial mass was mediated by PGC-1α, TFAM, and PPARα. We observed that Opti-ELF-WMF increased the expression of these proteins at 12 h (Fig. 5), indicating that PGC-1α-mediated mitochondrial biogenesis was activated after mitophagy to rejuvenate mitochondria.

Opti-ELF-WMF suppresses the enzymatic activity of ETC complex II in vitro
We conducted RNA-sequence (RNA-seq) analysis along with gene set enrichment analysis (GSEA) using AML12 cells exposed to Opti-ELF-WMF for 1 h. We found that Opti-ELF-WMF reduced the expression of mitochondrial ETC genes (Table. 2). Thus, suppression of mitochondrial ETC genes is likely to be a key factor in the triggering of mitophagy by Opti-ELF-WMF.
To capture the initial event activated by Opti-ELF-WMF, we examined the direct effects of Opti-ELF-WMF on the enzymatic activities of mitochondrial ETC complexes I, II, III, and IV in mouse liver homogenates that were exposed to Opti-ELF-WMF for 10 min or less in vitro. The enzymatic activity of ETC complex II was reduced to 88% by Opti-ELF-WMF, whereas the activities of the other ETC complexes (I, III, and IV) remained unchanged (Fig. 6A).
Mitochondrial ETC complex II is comprised of four succinate dehydrogenase (SDH) subunits: SDHA, SDHB, SDHC, and SDHD (Fig. 6E). To further dissect the effect of Opti-ELF-WMF on ETC complex II, we quanti ed the enzymatic activities of succinate:quinone reductase (SQR), succinate cytochrome c reductase (SCR), SDH, and SDHA in the mitochondria isolated from mouse liver homogenates exposed to Opti-ELF-WMF for 8 min. The activities of SQR, SCR, SDH, and SDHA decreased to 85%, 85%, 90%, and 95%, respectively (Fig. 6B). Thus, ELF was likely to suppress all the four subunits of mitochondrial ETC complex II. We also examined the expression of SDHA, SDHB, SDHC, and SDHD under Opti-ELF-WMF and did not nd any change in the expression levels of these subunits at 1 h after exposure (Fig. S5).
To examine whether mitophagy by Opti-ELF-WMF was indeed due to the suppression of mitochondrial ETC complex II, AML12 cells were incubated with either 3-nitropropionic acid (3-NP), an ETC complex II inhibitor, or rotenone, an ETC complex I inhibitor, for 12 h. The cells were then exposed to Opti-ELF-WMF for 3 h. Inhibition of ETC complex II by 3-NP negated the reduction in the mitochondrial mass induced by Opti-ELF-WMF (Fig. 6C). In contrast, inhibition of ETC complex I had no effect on the reduction in the mitochondrial mass induced by Opti-ELF-WMF (Fig. 6D). Taken together, Opti-ELF-WMF-mediated mitophagy requires electron ow through mitochondrial ETC complex II.

Discussion
We found that Opti-ELF-WMF reduced the amount of mitochondria by ~30% by inhibiting mitochondrial ETC complex II by ~15%, which subsequently induced mitophagy and rejuvenated mitochondria. Mitochondrial ETC complex II is comprised of four subunits, and Opti-ELF-WMF suppressed the activities of all the four subunits. The optimal conditions for ELF-WMF exposure to suppress the mitochondrial ETC activities were 1-8 Hz serial pulses for every 1 s, 10 µT magnetic eld, and 4 ms pulse width, which are referred to as Opti-ELF-WMF in this communication. We previously reported that Opti-ELF-WMF most e ciently increased the hysteresis of the electronic resistance of pure water as a function of temperature 11,12 . We have shown the effects of Opti-ELF-WMF on cultured cells and wild-type mice for the rst time. We found that the conditions of Opti-ELF-WMF exhibited the maximum effect on the reduction in the mitochondrial mass (Fig. S2). The modi cations of the pulse widths (Fig. S2B) and the frequency pro les (Fig. S2C) markedly attenuated the effects of ELF-WMF on the mitochondrial mass. As the conditions of Opti-ELF-WMF exerted the maximum effects on the hysteresis of electronic resistance of pure water in vitro and on the reduction of mitochondrial mass in vivo, the identity of the molecular target of Opti-ELF-WMF on ETC complex II subunits may share a feature similar to that of pure water.
The effects of magnetic elds on cultured cells, animal models, and humans have been reported mostly using static magnetic elds (SMFs) and radio frequency magnetic elds (RF-MF) 3,20,21 . Similarly, the effects of ELF-MF with static frequencies have also been reported 3,22 , but the effect of ELF-MF with time-varying frequencies has not been reported. In addition, the intensities of SMF, RF-MF, and ELF-MF were mostly greater than 1 mT and were rarely ~100 μT. In contrast, Opti-ELF-WMF had an MF intensity of 10 µT. According to the guidelines for limiting exposure to time-varying electric and magnetic elds by the International Commission on Non-Ionizing Radiation Protection (ICNIRP) 23 , the intensities of time-varying MF acceptable for occupational exposure increase with decreasing frequencies. For example, MF intensities of 1 mT and lower are safe at less than 300 Hz. Adverse effects with higher MF intensities include induction of magnetic phosphenes by 5 mT ELF-MFs at 20 Hz 24 ; gross external, visceral, or skeletal malformations by 20 mT LF-MF 25,26 ; and genotoxicity to cells by 50 mT LF-MF 27 . Because bene cial biological effects are sometimes inevitably accompanied by adverse effects, it is reasonable that biological effects of all the modalities of MF have been studied mostly with 1 mT or higher intensities.
Previous studies have shown that SMF, RF-MF, and ELF-MF increase, decrease, or have no effect on the levels of ROS in cultured cells and animal models 3 . In an SMF study, 200 μT decreased and 500 μT increased a surrogate marker of ROS, which suggested MF intensity-dependent changes in the levels of ROS 28 . We show that ELF-WMF decreased the mitochondrial ROS levels to 81% at 1 h and increased them to 114% at 12 h (Fig. 2C). The inconsistent effects of MF on the levels of ROS in previous reports may be at least partly accounted for by the temporal pro les of ROS levels.
We demonstrate that the target of Opti-ELF-WMF is the mitochondrial ETC complex II (Fig 6A, B). Cryptochrome (Cry) forms a complex with a magnetoreceptor protein (MagR) to conduct a nanoscale magnetoreception in many organisms, including mammals 29 . The Cry/MagR complex serves as a biocompass in these animals. The mitochondrial ETC complex II and Cry/MagR complex share the same components: avin adenine dinucleotide (FAD) and iron-sulfur clusters. A moiety in the mitochondrial ETC complex II that is targeted by Opti-ELF-WMF may reside in a structure shared with the Cry/MagR complex.
Compromised mitophagy and accumulation of damaged mitochondria are causally associated with neurodegenerative diseases 30 . Especially, germline mutations in genes encoding PINK1 and parkin, which are essential molecules in mitophagy, cause Parkinson's disease. In addition, heteroplasmic mutations in mitochondrial DNA cause mitochondrial diseases, in which the elimination of defective mitochondria carrying a high number of mutant mitochondrial DNA would enhance the biogenesis of normal mitochondria and ameliorate disease phenotypes 31 . We show that Opti-ELF-WMF induced mitophagy, followed by upregulation of the mitochondrial ETC activity. Similar to our observation, chemical inhibition of mitochondrial ETC complex II potentially provides neuroprotection by inducing autophagy in cultured neuronal cells 32 . Opti-ELF-WMF may be applicable in the treatment of neurodegenerative, mitochondrial, and other diseases, in which amelioration of compromised mitophagy and enhancement of normal mitophagy would be bene cial.
Repetitive transcranial magnetic stimulation (rTMS) is approved by FDA for treating depression, migraine, and compulsive disorder. Similarly, according to the guidelines of evidence-based medicine for rTMS, level A evidence indicating de nite e cacy is reached for depression and stroke-associated motor de cits 33,34 . In addition, level B evidence indicating probable e cacy is reached for Parkinson's disease, multiple sclerosis, bromyalgia, aphasia, and post-traumatic stress disorder. Although the therapeutic mechanisms of rTMS remain mostly elusive, rTMS preserves mitochondrial membrane integrity in a rat model of ischemic stroke 35 and decreases oxidative stress in a rat model of autoimmune encephalomyelitis 36 . If rTMS and Opti-ELF-WMF share similar mechanisms, Opti-ELF-WMF may serve a safe alternative to rTMS, in which the magnetic intensities up to 3.0 T potentially induce epilepsy and distressing sensation by stimulating the nerve and muscle 37,38 .

Limitations of the Study
We demonstrated that Opti-ELF-WMF induced mitophagy by inhibiting the mitochondrial ETC complex II activity, which was followed by hormetic facilitation of the mitochondrial ETC activity. We evaluated the effects under continuous exposure to Opti-ELF-WMF. However, discontinuation of the exposure or intermittent exposure might have exerted more effects. The temporal pro le of Opti-ELF-WMF might be able to be optimized in the future.

Declarations
Resource availability

Lead contact
Requests for resources and reagents should be directed to and will be ful lled by the lead contact, Kinji Ohno (ohnok@med.nagoya-u.ac.jp).

Material availability
This study did not generate new unique reagents.

Data and materials availability
Additional data related to this paper are available upon request to the authors. RNA-seq data were deposited in the Gene Expression Omnibus with an accession number GSE166811.  Figure 1 Exposure to Opti-ELF-WMF for 4 weeks increased mitochondrial electron transport chain (ETC) activities in the mouse liver. A Basal oxygen consumption rate was measured by a ux analyzer using mitochondria isolated from the mouse liver. No statistically signi cant difference was observed by Student's t-test (mean ± SD, n = 4 mice each). b Membrane potential of mitochondria isolated from the mouse liver was measured by ow cytometry with tetramethylrhodamine (TMRM; mean ± SD, n = 4 mice each; **p < 0.01 by Student's t-test). c Relative enzymatic activities of mitochondrial electron transport chain (ETC) complexes I, II, III, and IV of the mouse liver (mean ± SD, n = 4 mice each; *q [false discovery rate] < 0.05 by multiple Student's t-tests). d Western blotting of the mitochondrial oxidative phosphorylation proteins in the mouse liver (mean ± SD n = 4 mice each; no statistical difference was observed by false discovery rate with multiple Student's t-tests). Representative duplicates of western blot analysis are shown. See also Figure S1. ELF-WMF temporarily decreased the levels of mitochondrial reactive oxygen species (ROS), mitochondrial mass, and mitochondrial membrane potential. a Levels of mitochondrial ROS in AML12 cells exposed to Opti-ELF-WMF for 1 to 24 h were evaluated by MitoSOX. b Mitochondrial mass of AML12 cells exposed to Opti-ELF-WMF for 1 to 24 h was evaluated by MitoTracker Green. c Mitochondrial membrane potential of AML12 cells exposed to Opti-ELF-WMF for 1 to 24 h was evaluated by tetramethylrhodamine (TMRM). a-c show mean ± SD, n = 3 culture dishes each; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 by one-way ANOVA followed by Dunnett's posthoc test compared with the value at time 0.

Figure 3
Temporary downregulation of the levels of mitochondrial proteins at 3 h (a) and recovery or slight increase at 12 h (b) in cells exposed to Opti-ELF-WMF. Representative duplicates of western blot analysis of mitochondrial oxidative phosphorylation proteins, as well as of a mitochondrial outer membrane protein, VDAC1, in AML12 cells are shown (mean ± SD, n = 4 culture dishes each; *q < 0.05, and **q < 0.01 by multiple Student's t-tests). See also Figure  S3. ELF-WMF induced mitophagy. a AML12 cells were exposed to Opti-ELF-MF for up to 150 min. The levels of PINK1 and LC3-II were evaluated by western blot analysis (mean ± SD, n = 4 culture dishes each; *p < 0.05, **p < 0.01, and ***p < 0.001 by one-way ANOVA followed by Dunnett's posthoc test compared with the value at time 0). b Representative duplicated western blot analysis for parkin and PINK1 in whole cell lysates, a cytosolic fraction, and a mitochondrial fraction, as well as for ubiquitination in a mitochondrial fraction of AML12 cells exposed to Opti-ELF-WMF for 120 min (mean ± SD, n = 4 culture dishes each; *q < 0.05 and **q < 0.01 by multiple Student's t-tests). c Mitophagy was evaluated using Mtphagy Dye in AML12 cells exposed to Opti-ELF-WMF for up to 180 min (mean ± SD, n = 4 culture dishes each; ****p < 0.0001 by one-way ANOVA followed by Dunnett's posthoc test compared with the value at time 0). d Representative confocal images showing colocalization of mitochondria, detected by Mtphagy Dye, and lysosomes, detected by Lyso Dye, in AML12 cells exposed to Opti-ELF-WMF for 150 min. Scale bar, 50µm. See also Figure S4. Opti-ELF-WMF induced mitochondrial biogenesis. a Representative duplicates of western blot analysis are shown. b Densitometric analysis of western blots (mean ± SD, n = 4 culture dishes each; *q < 0.05 and **q < 0.01 by multiple Student's t-tests).

Figure 6
Opti-ELF-WMF suppressed the activities of the mitochondrial electron transport chain (ETC) complex II. a Relative enzymatic activities of mitochondrial ETC complex I, II, III, and IV of mouse liver homogenates exposed to Opti-ELF-WMF in vitro for 8 min (complexes I, III, and IV) or 10min (complex II) (mean ± SD, n = 3 mice each; **q < 0.01 by multiple Student's t-tests ) of isolated mitochondria exposed to Opti-ELF-WMF in vitro for 8 min (mean ± SD, n = 4 mice each; ****q < 0.0001 by multiple Student's t-tests). See (e) for a schematic of the fractional enzymatic activities. c Mitochondrial mass (MitoTracker Green) of AML12 cells treated with variable concentrations of a mitochondrial ETC complex II inhibitor, 3-nitropropionic acid, exposed to Opti-ELF-WMF in cellulo (mean ± SD, n = 3 culture dishes each; **q < 0.01 and ***q < 0.001 by multiple Student's t-tests). d Mitochondrial mass (MitoTracker Green) of AML12 cells treated with variable concentrations of a mitochondrial ETC complex I inhibitor, rotenone, exposed to Opti-ELF-WMF in cellulo (mean ± SD, n = 3 culture dishes each; *q < 0.05, and **q < 0.01 by multiple Student's ttests). e Schematic to indicate the fractional and extended enzymatic activities of mitochondrial ETC complex II measured in (b). Complex II is composed of SDHA, SDHB, SDHC, and SDHD. See also Figure S5.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.