Reduction in hypoxia‐reoxygenation‐induced myocardial mitochondrial damage with exogenous methane

Abstract Albeit previous experiments suggest potential anti‐inflammatory effect of exogenous methane (CH4) in various organs, the mechanism of its bioactivity is not entirely understood. We aimed to investigate the potential mitochondrial effects and the underlying mechanisms of CH4 in rat cardiomyocytes and mitochondria under simulated ischaemia/reperfusion (sI/R) conditions. Three‐day‐old cultured cardiomyocytes were treated with 2.2% CH4‐artificial air mixture during 2‐hour‐long reoxygenation following 4‐hour‐long anoxia (sI/R and sI/R + CH4, n = 6‐6), with normoxic groups serving as controls (SH and SH + CH4; n = 6‐6). Mitochondrial functions were investigated with high‐resolution respirometry, and mitochondrial membrane injury was detected by cytochrome c release and apoptotic characteristics by using TUNEL staining. CH4 admixture had no effect on complex II (CII)‐linked respiration under normoxia but significantly decreased the complex I (CI)‐linked oxygen consumption. Nevertheless, addition of CH4 in the sI/R + CH4 group significantly reduced the respiratory activity of CII in contrast to CI and the CH4 treatment diminished mitochondrial H2O2 production. Substrate‐induced changes to membrane potential were partially preserved by CH4, and additionally, cytochrome c release and apoptosis of cardiomyocytes were reduced in the CH4‐treated group. In conclusion, the addition of CH4 decreases mitochondrial ROS generation via blockade of electron transport at CI and reduces anoxia‐reoxygenation‐induced mitochondrial dysfunction and cardiomyocyte injury in vitro.


| INTRODUC TI ON
Ischaemic heart disease is a leading cause of death worldwide. The underlying pathophysiology is multifactorial, but mitochondrial dysfunction, is thought to be the common denominator in ischaemia or ischaemia/reperfusion (I/R)-mediated cardiomyocyte-damaging events. 1,2 Methane (CH 4 ) forms part of the gaseous environment, which maintains the metabolism within living aerobic cells. Though it is considered to be biologically inert, several studies have demonstrated bioactivity for exogenous CH 4 in animal models of ischaemia and inflammation, [3][4][5] and accumulating experimental data suggest that exogenous CH 4 can influence mammalian energy homeostasis as well. [3][4][5][6] More importantly, the administration of CH 4 has improved cardiac function, reduced the level of necroenzymes and prevented myocardial fibrosis and remodelling in acute and chronic rodent models of myocardial infarction. 4 Whereas these results suggest a causal link between increased CH 4 input and the hypoxia-induced oxido-reductive stress response, the subcellular mechanisms of action are still unclear.
The major goal of this study was to outline a mitochondrial pathway, which explains the bioactivity of CH 4 . In designing our experiments, we took into account that CH 4 can easily traverse cell membranes and that the molecules move down their concentration gradient into subcellular compartments. 7 Further, previous findings have demonstrated that CH 4 treatment can preserve adenosine-triphosphate (ATP) production after I/R injuries to the liver and eyes. 5,6 Therefore, these results strengthened the view that the mitochondrion is among the expected intracellular targets of CH 4 and led us to hypothesize that increased exogenous CH 4 input can influence the respiratory activity of cardiac mitochondria. 8 Against this background, we carried out a sequential exploration of the mitochondrial effects of exogenous CH 4 in normoxic and simulated I/R environments using a high-resolution respirometry (HRR) system to quantify the electron transport chain (ETC) responses. Ischaemia can impair the mitochondrial respiration and other oxygen-dependent cellular functions, leading to reversible or irreversible structural damage; therefore, we also detected cell viability and apoptosis of cardiomyocytes as final outcomes.

| Photoacoustic spectroscopy (PAS) measurement of CH 4 concentration
The dynamics of the CH 4 concentration changes were detected by photoacoustic spectroscopy (PAS), as described previously. 9 Briefly, PAS is a special mode of spectroscopy, which measures optical absorption indirectly via the conversion of absorbed light energy into acoustic waves. The set-up allows for online measurements of CH 4 concentrations with a minimum detectable concentration of 0.25 ppm. The CH 4 concentration in the medium was measured over a period of 120 min, and samples were taken every 2 min.

| Cardiomyocyte cell culture
Neonatal rat cardiac myocytes (NRMCs) were isolated, as described previously. 10 Briefly, 1-3-day-old Wistar rats were sacrificed by cervical dislocation, and the hearts were excised and collected in ice-cold phosphate-buffered saline. After the atria were removed, the ventricles were minced with scissors and digested with 0.25% trypsin for 25 min. The cell suspension was centrifuged at 2000 rpm for 15 min at 4°C. Pelleted cells were pre-plated for 90 min at 37°C to separate the cardiac myocyte-enriched fraction. Cardiac myocytes were collected and counted in a Burker chamber and plated into 24-well plates (7 × 10 4 cells/well) and into 75 cm 2 flasks (4 × 10 6 cells/flask).
Cells were harvested in Dulbecco's Modified Eagle's growth medium (DMEM) supplemented with a 10% foetal bovine serum (FBS), 1% glutamine and 1% antibiotic/antimycotic solution for 24 hours, and then, the medium was changed to 1% FBS-containing growth medium to promote the differentiation of the cardiomyocytes. At the end of the three-day isolation protocol, the phenotype of NRMCs corresponds to that of cardiomyocytes isolated from adult rats. The cardiac myocytes were kept in a normoxic incubator to maintain physiological conditions (37°C, 5% CO 2 and 95% air).

| Isolation of cardiac mitochondria
Adult Sprague Dawley rats were anaesthetized with sodium pentobarbital (45 mg/kg ip) to harvest the heart. The hearts were homogenized with a glass Potter homogenizer, and the mitochondria were isolated with Gnaiger's method. 11 The isolated mitochondria were suspended in a 2.5 mL mitochondrial respiration medium (MiRo5) for respirometric analysis and were treated as follows: 2 hours normoxia (95% air and 5%CO 2 ) or anoxia (100%N 2 ) was followed by reoxygenation (with or without CH 4 ) for 30 minutes. At the end of the experiments, mitochondrial function was tested.
In the second series, isolated cardiac mitochondria were treated as follows: anoxia was induced using 100% N 2 persufflation for 2 hours into a 2 mL volume cuvette containing 1 mL respiratory medium and 1 mL airspace. Anoxia was followed by a reoxygenation period (95% air and 5% CO 2 ) with or without 2.2% CH 4 supplementation for 30 minutes (the A/R and A/R + CH 4 groups, respectively) (n = 12-16). In the control groups, the mitochondria were kept in normoxic cuvettes (95% air and 5% CO 2 ) with or without 2.2% CH 4 supplementation (the normoxia and normoxia + CH 4 groups, respectively). Then, the mitochondria were subjected to HRR ( Figure S2).

| Examination of mitochondrial functions
High-resolution respirometry with an Oxygraph-2k (Oroboros Instruments, Innsbruck, Austria) was used to examine the oxygen consumption of the NRMCs and isolated cardiac mitochondria in various mitochondrial metabolic states, mitochondrial hydrogen peroxide (H 2 O 2 ) production and changes to mitochondrial membrane potential. The mitochondrial protein content of the samples was determined by Lowry's method.

| Coupling control protocol
Before the mitochondrial metabolic states were examined, a cell permeabilization protocol of the NRMCs was applied in the respirometer chamber ( Figure S3).
Next, we applied a coupling control protocol to the permeabilized NRMCs. First, endogenous routine respiration was defined without substrates. Then, the cells were permeabilized with digitonin, and the oxidative phosphorylation capacity (OxPhos, State 3) of the NRMCs was measured by adding 10 mmol/L succinate (Succ) and 5 mmol/L ADP substrates. Subsequently, ATP-independent respiration was measured using 0.5 µmol/L oligomycin (Omy). Maximal mitochondrial respiratory capacity was then measured by titration of 1 µmol/L carbonyl cyanide p-trifluoromethoxy-phenyl-hydrazine (FCCP). Finally, residual oxygen consumption (ROX) was determined by adding 1 μmol/L rotenone (Rot) and 1 μmol/L antimycin-A (Ama).

| Mitochondrial hydrogen peroxide (H 2 O 2 ) production
In this series, mitochondrial H 2 O 2 release as a marker of reactive oxygen species (ROS) (ie superoxide anion) production was moni-

| Mitochondrial membrane potential
Mitochondrial membrane potential was measured fluorimetrically using the fluorophore safranin. First, we added 1 µmol/L Rot, 10 mmol/L succinate and 1 µmol/L FCCP. Finally, ROX was determined by adding 1 μmol/L antimycin-A (Ama). Extramitochondrial Ca 2+ movement (as it indicates strong correlation with membrane potential 12 ) was examined with the use of blue fluorescence sensor of HRR (excitation 465 nm; gain for sensor: 1000 and polarization voltage: 500 mV) ( Figure S4).

| Detection of cytochrome c oxidase activity
Cytochrome c oxidase activity was calculated via the time-dependent oxidation of cytochrome c at 550 nm, as described previously.
Briefly, a cytochrome c stock solution was freshly prepared by dissolving 10.6 mg cytochrome c (Sigma-Aldrich, Budapest, Hungary) in 20 mL distilled water. The cytochrome c was then reduced by adding 50 µL 0.1 mol/L sodium dithionite, with the absorbance of the solution determined at 550 nm; the photometer was calibrated to this level. Heart samples were homogenized with a Potter grinder in 10× ice-cold Miro5 medium and then centrifuged at 800 g for 5 minutes at 4°C. 50 μL supernatant was added to 2.5 mL cytochrome c stock solution, and the decrease in optical density at 550 nm was measured spectrophotometrically during 1 minute intervals at 0, 30 and 60 minutes.

| TUNEL and DAPI staining
Apoptosis of the NRMCs was detected with the TUNEL method. First, the cell number was detected, and then, a cytocentrifuge (6 minutes, 600 rpm, 50 000 cells/slide) was used to create cytospin samples.

| CH 4 concentrations
The background CH 4 concentration in the airspace of the incubation chambers was 1.46.10 4 ± 94.95 ppm, and a rapid two orders of increase (to 1.5.10 6 ± 58.12 ppm) was detected after the start of persufflation with a 2.2% CH 4 -artificial air mixture. This concentration was steadily maintained during the 2 hours reoxygenation period ( Figure 1A

| Effect of CH 4 on the mitochondrial functions of the neonatal rat cardiomyocytes (NRMCs)
The coupling control protocol provides an opportunity to analyse the leak respiration of the mitochondria. As a result, significantly lower OxPhos was measured in the sI/R group in comparison with the normoxia group (19.17 ± 9.37 pmol/s*mL vs. 50.51 ± 12.87 pmol/s*mL; P < .001) (Figure 2A). CH 4 treatment in the sI/R + CH 4 group significantly enhanced oxygen consumption (to 40.88 ± 15.08 pmol/ s*mL; P = .004) (Figure 2A). The leak respiration decreased during sI/R (13.54 ± 2.66 pmol/s*mL; P < .001); however, it was ameliorated as a result of CH 4 administration in the sI/R + CH 4 group (19.94 ± 3.15 pmol/s*mL; P < .001) (Figure 2A). The sI/R significantly lowered the maximum respiratory capacity in comparison with the normoxia group (17.35 ± 4.46 pmol/s*mL vs. 35.72 ± 6.55 pmol/ s*mL; P < .001) (Figure 2A). CH 4 treatment had no effect on the maximum respiratory capacity during sI/R (18.41 ± 2.99 pmol/s*mL; P = .986) (Figure 2A). Flux values in different states were corrected for ROX (data not shown). There was a significant rise in mitochondrial H 2 O 2 levels in the I/R group as compared to the normoxia group.
Incubation with CH 4 during reoxygenation lowered the amount of H 2 O 2 production ( Figure 2B). The free radical leak was increased in the sI/R group; however, this rise was ameliorated as a result of CH 4 administration in the sI/R + CH 4 group (8.74 ± 3.85 vs. 4.09 ± 1.57; P < .05) ( Figure 2B). The CH 4 incubation had no effect on the free radical leak in the normoxia + CH 4 group compared with the normoxia group (3.97 ± 2.80 vs. 2.91 ± 0.64; P > .05) ( Figure 2B).

| Apoptosis, cytochrome c oxidase activity and viability of NRMCs
Neonatal rat cardiac myocytes were marked with TUNEL/DAPI staining to examine the presence of apoptosis. As expected, few TUNELpositive cells were observed in the normoxia and normoxia + CH 4 groups (26 ± 9% and 26.3 ± 12% of cells, respectively; P = 1.00) ( Figure 3A-B,E). sI/R was accompanied by an increased TUNEL positivity (sI/R: 52.4 ± 12% of cells) ( Figure 3C In the case of the LDH activity assay, there was no difference in this parameter between the two normoxic groups (0.22 ± 0.12

| Effects of CH 4 on isolated cardiac mitochondria
Changes to mitochondrial membrane potential have been characterized by means of the potential-sensitive fluorophore safranin.
Substrates of respiratory complexes induced a significant hyperpolarization in the mitochondrial membrane under normoxic conditions ( Figure 5A). In contrast, hyperpolarization was eliminated in the AR group. Substrate-induced changes in membrane potential were partially preserved by CH 4 supplementation ( Figure 5A). CH 4 applied during the anoxic period lowered the amount of H 2 O 2 production in leak states ( Figure 5B). In terms of oxygen consumption, we investigated complex I and succinate-semialdehyde dehydrogenase (complex II)-linked respiration separately. CH 4 significantly decreased the oxygen consumption of complex I, whereas it had no effect on complex II-linked respiration under normoxic conditions.
In contrast, CH 4 treatment in the sI/R + CH 4 group significantly improved the oxygen consumption of complex II compared with complex I ( Figure 5B).

| D ISCUSS I ON
Our primary aim was to outline a possible mechanism linked to the in vivo biological efficacy of CH 4 . The expected mitochondrial effects of CH 4 have been characterized by HRR, and we have shown that the administration of CH 4 reduces the sI/R-related mitochondrial ETC disturbance and mitigates subsequent apoptotic consequences.
Of importance, CH 4 preserved mitochondrial membrane potential have been demonstrated in other I/R settings as well. [3][4][5] These data all suggest that the underlying mechanism of action is intimately connected with the mitochondrial functions.
Excessive oxidative stress is a major component of sI/R, and the mitochondrial ETC is a dominant source of ROS generation.
Likewise, the majority of superoxide production is linked to complex I early in reperfusion. [13][14][15][16] This notion has been supported by studies showing that ischaemic preconditioning or pre-treatment with

F I G U R E 5
The effect of CH 4 on isolated cardiac mitochondria. The upper left-hand chart demonstrates representative records of mitochondrial membrane potential measured fluorimetrically by HRR. The continuous black line indicates changes in membrane potential; in parallel, the red line signifies the substrate-fuelled respiration. The upper right-hand chart presents changes in membrane potential in the experimental groups. The A/R group is labelled with a red line, the A/R + CH 4 group with a black line, and the normoxia and normoxia + CH 4 groups with pale and dark grey lines, respectively. The lower left-hand chart shows complex I and II-driven mitochondrial oxygen consumption. The lower right-hand chart demonstrates H 2 O 2 production in the case of reverse electron transfer (RET) reversible complex I inhibitors can limit ROS generation and cardiac IR injury. 14,16 In this primary mitochondrial model of cardiac IR injury, we investigated mitochondrial coupling states with HRR. We tracked mitochondrial ROS production with HRR by using the fluorescent dye Amplex Red, and, in parallel, the mitochondrial membrane potential was measured using the potential-sensitive fluorescence dye safranin. The respiratory activity of complex I remained stunned in the reoxygenation phase in line with the overwhelming ROS production. In the presence of CH 4 , the complex I-linked respiration decreased in both control and simulated ischaemia-damaged mitochondria, but there were no changes in the presence of rotenone, an irreversible complex I inhibitor. This suggests that CH 4 treatment reduced ROS generation via a partial blockade of electron transport in complex I. It should be added that two sites for superoxide production have recently been explored on respiratory complex I, the ubiquinone (Q)-binding and flavin sites. 15 The superoxide production at the flavin site is linked to the forward electron transport, and its rate depends on the reduction state of the matrix nicotinamide adenine dinucleotide (NAD) pool. More importantly, the Q-binding site produces superoxide at much higher rates than the flavin site, driven by the reverse electron transport (RET) from complex II into complex I during reoxygenation. 17 In isolated cardiac mitochondria, the underlying mechanism of RET is the accumulation of succinate during hypoxia and its subsequent rapid oxidation at reoxygenation in the presence of a high membrane potential. 18 Rotenone, an irreversible inhibitor of electron transport at the Q-binding site has been demonstrated to exert cardioprotection by decreasing RET in the early phase of reperfusion. 14 Based on our results, the active site of CH 4 is, as in the case of rotenone, distal to the flavin site, because it enhances mitochondrial ROS generation when the electrons enter complex I from NADH, but it inhibits ROS generation by RET from complex II. Large membrane potential is a prerequisite to drive the electrons against the gradient of redox potentials from complex II into complex I. It has been demonstrated in isolated mitochondria that only a 5% reduction in mitochondrial membrane potential will reduce peroxide production by 95%. 19 Any manipulation of the RET pathway could potentially influence the end outcome of ROS production, even by lowering the driving hyperpolarization of the mitochondrial membrane potential. In our system, the addition of normoxic CH 4 slightly decreased the substrate-induced hyperpolarization in control mitochondria, in contrast to the preservative effect seen in the case of the anoxia-damaged membrane.
Next, the role of complex I and complex II in the post-anoxic cardiac mitochondrial respiration was addressed in more detail.
As a result of CH 4 treatment, respiration was inhibited when glutamate + malate was used as a complex I substrate but not with succinate as a complex II substrate. This finding suggested that the addition of CH 4 resulted in a decreased electron flux through complex I but did not alter the succinate oxidation through complex II.
Based on these findings, the drop in net ROS production from mito-

| CON CLUS ION
We have presented our first data on the effects of CH 4 on transient anoxia-induced respiratory changes in cardiomyocyte cultures.
Evidence suggests that CH 4 treatment decreases myocyte injury through reduced ROS generation via blockade of electron transport in complex I and improved inner mitochondrial membrane integrity.
Further in vivo studies are needed to investigate whether the administration of CH 4 would provide a way to attenuate the potentially harmful mitochondrial consequences of hypoxia-reoxygenation insults.

ACK N OWLED G EM ENT
The authors are grateful to Nikolett Beretka and Csilla Mester for their skilful assistance.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests. The effects of CH 4 on complex I. The mechanism of protection likely included a blockade of the electron transport in complex I and decreased ROS generation. Reversible deactivation of mitochondrial complex I is an intrinsic mechanism, which provides a fast response of the mitochondrial respiratory chain to oxygen deprivation. However, subsequent reoxygenation leads to ROS generation due to the rapid burst of respiration. Under normoxic conditions, a high level of nicotinamide adenine dinucleotide hydride (NADH) can drive forward electron flow with superoxide generation at the flavin mononucleotide moiety located near the NADH binding subunit. During reoxygenation, reverse electron flow driven by a reduced ubiquinone (ubiquinol) pool and high proton motive force can generate ROS when electrons flow back from ubiquinol to Complex I. CH 4 treatment restricts the forward electron transfer within complex I in control mitochondria while effectively inhibiting RET in post-ischaemic mitochondria editing (equal). Petra Hartmann: Conceptualization (supporting); data curation (supporting); methodology (supporting); project administration (supporting); resources (lead); supervision (lead); validation (lead); writing-original draft (supporting); writing-review and editing (equal).

E TH I C A L A PPROVA L
The experimental protocol was in accordance with EU Directive 2010/63 for the protection of animals used for scientific purposes,

CO N S E NT FO R PU B LI C ATI O N
Not applicable.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data are all presented in the manuscript.