Inhibition of BKCa channels protects neonatal hearts against myocardial ischemia and reperfusion injury

BKCa channels are large-conductance calcium and voltage-activated potassium channels that are heterogeneously expressed in a wide array of cells. Activation of BKCa channels present in mitochondria of adult ventricular cardiomyocytes is implicated in cardioprotection against ischemia-reperfusion (IR) injury. However, the BKCa channel’s activity has never been detected in the plasma membrane of adult ventricular cardiomyocytes. In this study, we report the presence of the BKCa channel in the plasma membrane and mitochondria of neonatal murine and rodent cardiomyocytes, which protects the heart on inhibition but not activation. Furthermore, K+ currents measured in neonatal cardiomyocyte (NCM) was sensitive to iberiotoxin (IbTx), suggesting the presence of BKCa channels in the plasma membrane. Neonatal hearts subjected to IR when post-conditioned with NS1619 during reoxygenation increased the myocardial infarction whereas IbTx reduced the infarct size. In agreement, isolated NCM also presented increased apoptosis on treatment with NS1619 during hypoxia and reoxygenation, whereas IbTx reduced TUNEL-positive cells. In NCMs, activation of BKCa channels increased the intracellular reactive oxygen species post HR injury. Electrophysiological characterization of NCMs indicated that NS1619 increased the beat period, field, and action potential duration, and decreased the conduction velocity and spike amplitude. In contrast, IbTx had no impact on the electrophysiological properties of NCMs. Taken together, our data established that inhibition of plasma membrane BKCa channels in the NCM protects neonatal heart/cardiomyocytes from IR injury. Furthermore, the functional disparity observed towards the cardioprotective activity of BKCa channels in adults compared to neonatal heart could be attributed to their differential localization.


INTRODUCTION
Large-conductance calcium and voltage-activated potassium channels (BK Ca channels) encoded by the Kcnma1 gene are the key electrochemical couplers between cellular metabolic state and Ca 2+ homeostasis. The Kcnma1 gene is highly conserved through a wide spectrum of species, including Drosophila Melanogaster [1,2], mus musculus [3], and homo sapiens [4]. BK Ca channels are expressed in a broad range of excitable and non-excitable cells and have been implicated in various fundamental physiological processes, including regulation of gene expression [5], urinary and erectile autonomic functions [6,7], vascular tone [8], neuronal excitability [9], cardiac rhythmicity [10][11][12], and aging [2]. Recently, several Kcnma1associated channelopathies have been identified in humans and BK Ca channels have been projected as key therapeutic targets [4]. They are ubiquitously present in the plasma membrane of a majority of the cell types, are highly selective for K + with a large unitary conductance, and are sensitive to iberiotoxin (IbTx) as well as paxilline.
In addition to the plasma membrane, BK Ca channels are also expressed in intracellular organelles like the endoplasmic reticulum, nuclei, lysosomes, and mitochondria [13,14]. Channel expressed in the cell membrane and intracellular organelles is a product of the Kcnma1 gene, but in cardiomyocyte mitochondria, the c-terminal splice variation of the Kcnma1 gene (DEC) determines its inner mitochondria membrane (IMM) localization [13]. In physiological conditions, Ca 2+ and voltage simultaneously control BK Ca channel currents, but in the absence of Ca 2+ , membrane depolarization alone can generate BK Ca channel currents. On the other hand, Ca 2+ binding decreases the energy required to open the channel, shifting the open probability (P O ) towards the more polarized state [15][16][17]. In smooth muscle cells, BK Ca channels are key determinants in controlling the resting membrane potential and vascular tone, and hence, blocking the BK Ca channel results in membrane depolarization and increased contractile tone [18][19][20]. BK Ca channels are also present in the sinoatrial node (SAN) where their inhibition showed an increase in action potential duration (APD) [11]. Furthermore, when wild-type mice were injected with paxilline (a highly specific BK Ca channel inhibitor), their heart rate decreased by 30%, implicating BK Ca channels in maintaining cardiac rhythmicity [12].
The use of pharmacological agents and genetically modified animal models demonstrated that the expression and opening of BK Ca channels present in adult ventricular cardiomyocytes are directly involved in cardioprotection from IR injury [13,[21][22][23][24]. The cardioprotective mechanism is mediated by BK Ca channels present in the mitochondria of adult cardiomyocytes [13,21,23]. On the other hand, there is growing evidence that BK Ca channels are present in the plasma membrane at the embryonic stage in chick ventricular myocytes [25]. Potassium currents measured in the plasma membrane of cardiomyocytes from 10 to 12 days old chick embryos exhibited all the hallmarks of the BK Ca channel properties such as: large single-channel conductance, voltage, and Ca 2+ dependence and sensitivity to tetramethylammonium (TEA)-and charybdotoxin [25]. Additionally, human-induced pluripotent stem cells-derived cardiomyocytes (hiPSC-CMs), showed sensitivity to IbTx treatment that shifted membrane polarization towards more positive potentials and caused a disturbance in action potential wave, which was not observed in cardiomyocytes isolated from the adult human left ventricle [26]. However, the precise role and impact of BK Ca channels in the plasma membrane of cardiomyocytes during the early stages of development are not yet elucidated.
Although activation of cardiomyocyte mitochondrial BK Ca channels (mitoBK Ca ) has been long recognized in cardioprotection against ischemia-reperfusion injury in adults, the role and function of plasma membrane BK Ca channels in neonates are completely unknown. Here, we present our novel findings showing a stagedependent localization of BK Ca channels in cardiomyocytes during early development, and its impact on cardiomyocyte function and cardioprotection.

RESULTS
BK Ca channels are present in the plasma membrane of neonatal cardiomyocytes BK Ca channels are known to localize to the mitochondria of adult mouse and rat ventricular cardiomyocytes [13,22,23]. However, their localization in NCMs is unknown. We evaluated the localization of BK Ca channels in NCMs isolated from rats and mice using a highly specific anti-BK Ca antibody ( Supplementary Fig. 1) [13,22]. Similar to adult ventricular cardiomyocytes [13,22], in both rat and mouse NCMs, BK Ca channels showed a strong localization to mitochondria (co-labeled with anti-ATP5A) (Fig. 1). Fig. 1 Localization of native BK Ca channel in neonatal and cardiomyocytes isolated from the rodent heart. A, B, C, and D Isolated cardiomyocytes loaded with WGA (magenta), fixed, permeabilized, and labeled with BK Ca (green), and ATP5A (red), antibodies. Nuclei were stained with DAPI (blue). The right panels in A, B, C, and D show merged images at higher magnification. E The purple bar represents percentage colocalization between the BK Ca channel and plasma membrane (WGA), which is significantly higher in neonatal cardiomyocytes compared to adult cardiomyocytes. The red bar represents percentage colocalization between the BK Ca channel and mitochondria, which is lower in neonatal cardiomyocytes as compared to adult cardiomyocytes isolated from rodent hearts. All experiments were repeated independently at least four times and colocalization data are represented as mean ± SEM from 10 cells in each "n" number. P-values were determined by a one-tailed paired student's t-test; **≤ 0.001, ***≤ 0.0001, ### ≤ 1.0*10 -8 .
In rats, localization of BK Ca channels to mitochondria increased from 37 ± 5% (n = 20) in postnatal day 3 (P3) NCMs to 56 ± 4% (n = 20) in adult ventricular cardiomyocytes. Similarly in mice, BK Ca channels localization to ATP5A labeled-mitochondria was 40 ± 3% (n = 20), which increased to 62 ± 4% (n = 25) in P3 NCMs and adult ventricular cardiomyocytes, respectively. These findings were similar to earlier reports on the preferential localization of BK Ca channels to cardiomyocyte mitochondria [27]. In adult ventricular cardiomyocytes isolated from mice or rats, there was a small amount of localization of BK Ca channels to the plasma membrane 12 ± 1% (shown by WGA staining), which could be attributed to mitochondria present near the plasma membrane ( Fig. 1B, D, E). However, in NCMs isolated from rats and mice, BK Ca channels showed 26 ± 3% (n = 20) and 32 ± 3% (n = 15), localization to WGA labeled plasma membrane ( Fig. 1A, C, E), respectively.
BK Ca channels are functional in the plasma membrane of neonatal cardiomyocytes Previous studies [14,28,29] have shown that BK Ca channel-specific currents are not present in the membrane of adult ventricular cardiomyocytes, due to its absence in the plasma membrane [13]. Since we observed localization BK Ca channels in the plasma membrane of NCMs, we evaluated whether these BK Ca channels are functional NCMs using patch-clamp analysis. In the whole-cell configuration, cells were held at -70 mV holding potential and voltage steps of 20 mV (40 ms) from -70 to +150 mV were applied. As shown in Fig. 2A, D, we recorded a large outward current in NCMs isolated from both rats and mice. In rat NCMs, the open probability of potassium currents was decreased by 46 ± 5% (n = 10, Fig. 2A-C) and in mouse NCMs, 32 ± 5% (n = 10, Fig. 2D, E, F) after application of 100 nM IbTx, indicating the presence of functional BK Ca channels in the plasma membrane of NCMs.
Localization of BK Ca channel during cardiomyocyte development Owing to the changes in the localization of BK Ca channels from the plasma membrane to mitochondria in neonatal and adult cardiomyocytes, we investigated changes in their age-specific localization. Hence, we studied the localization of BK Ca channels in cardiomyocytes isolated from pups at P3, P7, P14, P21, P28 days, and adult rats. In NCMs, BK Ca channels were present in the plasma membrane as well as in mitochondria ( Fig. 3A-F). Protein proximity index analysis [30] showed that BK Ca localization to mitochondria increases from P3 to the adult stage, while localization of BK Ca to plasma membrane decreases (Fig. 3G). The major determining factor attributed to the localization of BK Ca to mitochondria is the C-terminus DEC splice variant (BK Ca DEC) [13]. We quantified the DEC splice variant and compared it with the total BK Ca (BK Ca FL) signal present in rat cardiomyocytes isolated at different ages. We observed an increased expression of the BK Ca DEC from P3 to adult, which corresponds to increased localization of BK Ca channels to mitochondria in cardiomyocytes (Fig. 3G, H).
Activation of BK Ca channels in neonatal rat heart increases myocardial infarction Activation of BK Ca channels either pharmacologically [13,23,24,[31][32][33] or genetically [22] in adult animal models results in cardioprotection from IR injury whereas inhibition of BK Ca channels increases myocardial infarction [21]. We tested whether activation of BK Ca channels can protect neonatal hearts from IR injury as observed in adult hearts. Hearts isolated from 6 days old rats (P6) were subjected to ischemia and immediate post-conditioning with a BK Ca channel activator (10 µM NS1619) or inhibitor (100 nM IbTx) (Fig. 4A). Hearts were reperfused after post-conditioning. Infarct size was measured by staining cardiac sections at the end of reperfusion by 2, 3, 5 -triphenyl tetrazolium chloride (TTC) (Fig. 4B-D). Postconditioning with NS1619 significantly increased the infarct size (58 ± 3%) as compared to vehicle control (48 ± 5%) as shown in Fig. 4B, D, E. In contrast, hearts post-conditioned with IbTx showed reduced infarct size (32 ± 5%) (Fig. 4C, E) compared to the vehicle. To corroborate the expression of functional BK Ca channels on the plasma membrane, we recorded potassium currents in NCM isolated from P6 rat pups. The potassium currents were significantly Traces of whole-cell BK Ca channel currents were recorded from A rat neonatal cardiomyocytes and D mouse neonatal cardiomyocytes with or without iberiotoxin (100 nM). Neonatal cardiomyocytes B rat and E mouse, membrane currents were activated by 40 ms voltage steps of 20 mV between −70 to +150 mV from resting −70 mV holding potential. The current versus voltage relationships are plotted with or without iberiotoxin. C and F Percentage block of BK Ca current in the presence or absence of iberiotoxin. Cells were isolated from ten independent litters from mice and rats, and data are represented as mean ± SEM. P-values were determined by a one-tailed paired student's t-test; ****≤ 0.00001, *****≤ 0.000001.
reduced by 33 ± 3% after the addition of 100 nM IbTx (Fig. 4F). Thus, our results indicate that contrary to adults, in neonates, the activation of the BK Ca channel after IR injury is cardio-deleterious whereas BK Ca channel inhibition is cardioprotective.
Activation of the BK Ca channel induces apoptosis after hypoxia-reoxygenation in NCM One of the major roles attributed to the activation of BK Ca channels is protecting adult cardiac cells and tissues from ischemia/hypoxia-reoxygenation injury [13,21,23,34]. These functions are specifically ascribed to the presence of BK Ca channels located in mitochondria of adult ventricular cardiomyocytes [21,34]. Since we discovered BK Ca channels in the plasma membrane of NCMs ( Figs. 1 and 2), we determined the effect of plasma membrane BK Ca channels in isolated NCMs after hypoxiareoxygenation injury. We used cell impermeable (IbTx, 100 nM) to tease out the role of plasma membrane BK Ca channels in HRinduced apoptosis in NCMs (P3). IbTx showed comparable apoptotic positive nuclei 21 ± 8% after HR injury to dimethyl sulfoxide (DMSO) control 30 ± 9% as indicated by TUNEL assay (Fig. 5A, B and quantified in D). Surprisingly, NS1619 treatment showed 63 ± 14% TUNEL-positive cells as compared to DMSO control 30 ± 8% after HR injury (Fig. 5A, C, D).
We have earlier shown that gain of function transgenic adult mice presented cardioprotection after ischemic preconditioning as well as after IR injury [22]. Here we isolated NCMs from wild type and Tg-BK Ca R207Q mice [9] and subjected P3 NCMs to HR injury. After hypoxia and reoxygenation, we discovered that NCMs Mitochondrial targeting BK DEC splicing increases with age in neonatal cardiomyocytes. A-F Cardiomyocytes isolated from aging (P3, P7, P14, P21, P28, and adult) rats loaded with WGA (magenta), fixed, permeabilized, and labeled with BK Ca (green) and ATP5A (red). The nucleus was stained with DAPI (blue). The right panels of A-F show colocalization of merged images between BK Ca to WGA and BK Ca to ATP5A at higher magnification. G Protein proximity index analysis showing percentage colocalization of BK Ca to WGA decreases, whereas BK Ca to mitochondria increases in cardiomyocytes isolated from P3 to the adult stage. H Relative levels of BK Ca FL and BK Ca DEC mRNA expression in cardiomyocytes isolated from P3, P7, P14, P21, P28, and adult hearts were measured by qRT-PCR and were normalized to their relative GAPDH expression. Data represented as fold-change to P3 BK Ca FL and P3 BK Ca DEC; mean ± SEM; at least three independent experiments. P-values were determined by a one-tailed paired student's t-test; *≤ 0.05, **≤ 0.001, ***≤ 0.0001, ****≤ 0.00001.
from Tg-BK Ca R207Q mice showed a significant increase in TUNELpositive cells 66 ± 4% as compared to wild-type mice 17 ± 2% ( Fig. 5E-G). Our data from pharmacological treatment of WT NCMs and genetically activated transgenic mice NCMs indicate that activation of the BK Ca channel increases apoptosis and blocking of the BK Ca channel protects cells.
Plasma membrane BK Ca channel modulates cellular reactive oxygen species upon HR injury Mitochondria isolated from adult guinea pig heart showed that activation of BK Ca channels by NS1619 lowers reactive oxygen species (ROS), which is implicated in cardioprotection from IR injury [35][36][37]. We also investigated the role of plasma membranelocalized BK Ca channels in regulating ROS in NCMs upon HR injury.
As shown in Fig. 6, after hypoxia and reoxygenation, activation of BK Ca channel with NS1619, intracellular ROS significantly increased compared to control and IbTx-treated cells. These results also support our findings that inhibition, and not activation of BK Ca channel in NCMs or neonates is cardioprotective from IR injury.
Expression of the BK Ca channel causes delayed repolarization of the plasma membrane BK Ca channel generates a large conductance for potassium in cellular membranes. Ectopic expression of the BK Ca channel in murine HL-1 cardiomyocytes indicates that the introduction of BK Ca shortens the action potential duration [38]. In human iPSCderived cardiomyocytes, the BK Ca channel induces irregular action potential shapes resembling very early afterdepolarizations [26]. Inhibition of BK Ca channel protects the neonatal heart from ischemia injury. A Schematics of the ischemia-reperfusion protocol. Rat P6 pup hearts were subjected to 20 min ischemia, 5 min post-conditioned with DMSO, Iberiotoxin (100 nM), or NS1619 (10 μM) followed by 60 min of reperfusion. Hearts post-conditioned with C and E iberiotoxin exhibited significantly less infarction (white) as compared to the B and E DMSO control. In hearts post-conditioned with NS1619 displayed significantly higher infarction compared to DMSO control and IbTx treatment (E). The current versus voltage relationships are plotted with or without iberiotoxin (F). Data represented as mean ± SEM for percentage infarction from four independent experiments and electrophysiology data represented from four independent litters P6 rat pups. P-values were determined by a one-tailed paired student's t-test; *≤ 0.05, ***≤ 0.0001.
On these lines, we performed an in silico simulation on cardiac action potential (AP) in presence of the BK Ca channel. We assumed the open probability to 0.5 and the unitary conductance at 300 pS. The addition of the BK Ca channel current creates an early voltage "notch", which is noted during the depolarization phase of the AP ( Supplementary Fig 2). The peak depolarization amplitude is also reduced when compared to the native AP. Perhaps the most notable change is during the repolarization phase, which is much more rapid with the addition of the BK Ca channel current. This rapid repolarization back to the resting state noticeably decreases the duration of the AP. Although these approaches indicate that induction of BK Ca channel-specific current will disrupt the AP, in the absence of experimental action potential data from native cardiomyocyte BK Ca channels, the precise role of the BK Ca channel in cardiac AP could not be established.
Since our data indicate that a functional BK Ca channel is present in NCMs, we tested whether the NCM BK Ca channel plays an active role in modulating cardiac action potential. We isolated NCMs from P3 rats and seeded them on a microelectrode array for recording their activity. As shown in Fig. 7, NCMs are highly active and present rapid action potentials. We used two different concentrations of NS1619 (low: 10 μM and high: 20 μM) and IbTx (low: 100 nM and high: 200 nM). Low-dose NS1619 or IbTx had no significant impact on APD30, APD50, APD90, and rise time of action potential, however, a high dose of NS1619 (Fig. 7A) impacted the NCMs AP. NCMs treated with 20 µM NS1619 showed a significantly altered AP morphology, increased APD30 (Fig. 7C), APD50 (Fig. 7D), and APD90 (Fig. 7E) by 4%, 11%, and 21%, respectively compared to DMSO control.
Moreover, NS1619 increased the rise time ( Fig. 7F) and APDc (Fig. 7H) by 67% and 31%, respectively but decreased the triangulation ratio (Fig. 7G) by 7% when compared to DMSO. These changes in the AP morphology indicate the opening of the BK Ca channel in NCMs might increase arrhythmogenic risk. Similarly, BK Ca channel opening increases in the beat period ( Fig. 7K) and field potential duration (Fig. 7L) by 14%, and 30% as compared to DMSO. The spike amplitude (Fig. 7M) showed a 30% decrease along with conduction velocity (Fig. 7N), which showed a 23% decrease. There were no changes observed with a high dose of IbTx (Fig. 7B, J) in agreement with rat and human ventricular myocyte AP's insensitivity to IbTx [26,39]. Adverse impact on APs and cardiomyocyte function upon opening of plasma membrane BK Ca channels in NCMs implicate them in cardio-deleterious effect.

DISCUSSION
Activation of the BK Ca channel located in the inner mitochondrial membrane has been associated with cardioprotection and neuroprotection [29,40]. Opening of mitoBK Ca in cardiomyocytes has been implicated to increase mitochondrial K + ion flux, which depolarizes the mitochondrial membrane, reduces the Ca 2+ influx and Ca 2+ overload during injury [40]. Similarly, the opening of mitoBK Ca in neurons inhibits hydrogen peroxide production, which reduces ROS production and mediated neuroprotection [41]. To our knowledge, plasma membrane BK Ca channelmediated currents have not been reported in cardiomyocytes [40]. Our study for the first time indicates the presence of BK Ca ) were subjected to 6 h hypoxia at 1% O 2 followed by 12 h of reoxygenation at 21% O 2 . Cells were fixed and stained for TUNEL-positive nuclei (red), and Nucleus (blue). Neonatal cardiomyocytes isolated from Tg-BK Ca R207Q pups showed higher TUNELpositive nuclei as compared to WT-BK Ca (G). Data represented as percentage TUNEL-positive cells (D and G) mean ± SEM from eight out of ten independent experiments. P-values were determined by a one-tailed unpaired student's t-test followed by a non-parametric Mann-Whitney test for comparison between the ranks; ns is not significant, *≤ 0.05, **≤ 0.001. channels in the plasma membrane of the NCMs isolated from mice and rat hearts, as well as in cardiac cells of human infant hearts. Furthermore, we have recorded K + currents sensitive to IbtX and functional consequences of the activation and/or inhibition of plasma membrane BK Ca channels in NCMs. Here we determined the physiological role of neonatal plasma membrane BK Ca channels in cardioprotection and cardiomyocyte function.
Cardiac potassium channels are encoded by over 40 distinct genes [42]. In the heart around 10 distinct potassium channels function in tandem to tightly regulate the cardiac repolarization to ensure stable and consistent AP propagation [43]. Although coded by different genes, some of these channels have overlapping functions, which result in some degree of functional redundancy and is known to contribute to repolarization reserve. Cardiac potassium channels are classified into transient outward, delayed rectifier outward, and inward rectifier currents. Canonical cardiac potassium channels play key roles during phase 1, phase 2, phase 3, and phase 4 of action potentials. More recently Ca 2+ -activated small conductance potassium channels (K Ca 2.x) [44] and several two-pore domain potassium channels (K 2 P) [45] have been characterized in atrial repolarization. However, BK Ca channels have been largely reported in cardiomyocytes' mitochondria but not in the plasma membrane so their role in cardiac AP is not known.
Cardiac potassium channels have been associated with several cardiac diseases such as long and short QT syndromes (LQTS, SQTS), Brugada syndrome (BrS), Andersen-Tawil syndrome, and AF [46]. The occurrence of arrhythmias and ischemic heart diseases increases with cardiac development and age [47,48]. One of the reasons attributed to the lower occurrence of cardiac electrophysiological dysfunction in the younger population could be due to differential expression of ion channels in cardiac tissues during the development. For example, in cardiac development, the expression of I Kir , and I KATP was shown to increase from embryonic day 10 to neonate day 1 in rats, and the current density decreases after day 30 whereas I KAch presented no change [49]. Ion channels play a key role in cell division and differentiation [50]. These channels need to be tightly regulated to maintain the fate and physiological role of cells. Any abnormal expression of the activity of ion channels in developing organs could have a detrimental impact.
BK Ca channels are ubiquitously expressed in the plasma membrane of cardiac fibroblast [51], neurons [9,[52][53][54][55][56], endothelial [57,58], and vascular smooth muscle cells [59][60][61]. However, in adult ventricular cardiomyocytes, a unique C-terminal splice variant of the BK Ca channel targets it to the mitochondrion [13]. In the cardiovascular system, BK Ca channels have been suggested to promote vascular relaxation, regulate heart rate, and protect against IR-mediated injury [13,[21][22][23]. In vascular smooth muscle cells, the plasma membrane-localized BK Ca channel works synchronously with the ryanodine receptor and L-type calcium channel to regulate cellular contraction and relaxation [62]. Interestingly, BK Ca channels have never been detected in the plasma membrane of the adult ventricular cardiomyocytes [28,29]. On the contrary, in chick embryonic cardiac cells and hiPSC-CMs BK Ca channel-specific-potassium currents were recorded [26,63], suggesting that at the early stages of cardiac development, functional BK Ca channels are present in the plasma membrane of myocytes. In adult ventricular cardiomyocytes, there are no BK Ca channel-specific currents [28] in the plasma membrane, which also is further corroborated by the insensitivity of IbTx on ventricular APs [26], and by immunocytochemistry [13]. In this study, for the first time in murine and rodent models, we have detected functional BK Ca channels in the plasma membrane of NCMs. We have also discovered that BK Ca channels localize to the plasma membrane of cardiomyocytes in human infant hearts (Fig. 8). The localization of BK Ca channels in the plasma membrane could be transitional as cells mature and proteins reach their destined organelles. During development cardiac muscle undergoes two major changes; (1) electrical conduction and (2) energy metabolism. The presence of BK Ca channels in the plasma membrane could facilitate potassium fluxes in cardiomyocytes, which are essential for development. At the same time in the later part of the development, the energy demand of heart muscle increases, which brings the focus to mitochondria. The presence of BK Ca channels in mitochondria in adult cardiomyocytes is essential for mitochondrial structure and function [21]. Another possibility is the presence of BK Ca channels in both plasma and mitochondrial membranes are essential for K + homeostasis for mitochondrial function. Alternatively, it is possible that expression of BK Ca channels is required to selectively remove the cells undergoing stress (hypoxia or increase in ROS) by triggering the opening of plasma membrane BK Ca channels and causing apoptosis (Fig. 6).
Though no BK Ca channel-specific currents were reported in the plasma membrane of adult ventricular cardiomyocytes, BK Ca Fig. 6 Inhibition of BK Ca channel protects the neonatal cardiomyocytes from oxidative damage post hypoxia injury. Isolated rat neonatal cardiomyocytes were subjected to 6 h hypoxia at 1% O 2 , 5 min of post-conditioning with DMSO, Iberiotoxin (100 nM), or NS1619 (10 μM) followed by 12 h of reoxygenation. Cells were stained with CM-H 2 DCFDA (green) for 45 min and NucBlue™ for nucleus (blue). A total of 100,000 cells were seeded per assay. B and D Cardiomyocytes post-conditioned with IbTx show less cellular ROS compared to DMSO control (A and D). C Cardiomyocytes postconditioned with NS1619 exhibited significantly higher ROS generation as compared to IbTx and DMSO control and quantified (D). Data represented as mean fluorescence unit ± SEM from eight of ten independent experiments. P-values were determined by a one-tailed unpaired student's t-test followed by a non-parametric Mann-Whitney test for comparison between the ranks; ns is not significant, *≤ 0.05, **≤ 0.001. channels are present in the plasma membrane of smooth muscle cells. In smooth muscle cells, activation of the plasmalemma, the BK Ca channel using NS1608 hyperpolarizes the membrane potential [64]. However, we have shown that blocking with IbTx implies that over~30% of recorded K + currents originate from BK Ca channels in rat and mouse NCM. In a quantitative model of a human ventricular cardiomyocyte action potential with the BK Ca channel integrated into the cell membrane [26], a deep notch was shown during the depolarization phase. Our simulation of the action potential of model cardiomyocyte cells presented a similar phenotype including a notch in the depolarization phase [26]. In an overexpression model where human BK Ca channels were introduced in murine HL-1 cardiac cell line [38], rapid repolarization as well as a shortened action potential were reported, which is consistent with our simulation. These in silico and experimental data indicate an opening of the BK Ca channel allows rapid efflux of K + in response to voltage and Normalized LEAP waveforms for NS1619 (A; green trace) and IbTx (B; red trace) were overlaid with DMSO control (black trace). Bar plots represent C APD30, D APD50, E APD90, F rise time, G triangulation ratio, and H APDc for IbTX and NS1619 normalized to DMSO group and represented as fold-change mean ± SEM. Field potential duration waveform (I, J of neonatal cardiomyocytes post-treatment with NS1619 (I; green trace), and IbTx (J; red trace) overlaid with their respective pre-treatment baseline electrodes. Bar plots represent K beat period, L field potential duration, M spike amplitude, and N conduction velocity for DMSO, IbTx, and NS1619 were normalized to their respective pretreatment electrodes and represented as fold-change mean ± SEM from three independent littles. P-values were determined by a one-tailed paired student's t-test; ns is not significant, *≤ 0.05, **≤ 0.001. [Ca 2+ ] i , which aids in rapid repolarization and thus shortening of the AP. However, little is known about the functional consequences of uniquely localized native plasma membrane BK Ca channel in neonatal cardiomyocytes.
Owing to the distinct localization of the native BK Ca channel in the plasma membrane, we studied the effect of its activation and inhibition on the electrical activation of NCMs. In HL-1 cardiac cells, overexpression of the BK Ca channel affects repolarization K + currents and shortening of action potential duration without depolarizing the membrane or impacting intracellular Ca 2+ influx [38]. In agreement, our data indicated that activation of endogenous BK Ca channel present on NCM plasma membrane shows no change in the membrane depolarization. However, activation of endogenous BK Ca channel delays the repolarization and severely prolongs the duration of the action potential. On the contrary, inhibiting the BK Ca channel activity using IbTx, did not affect the repolarization or duration of the action potential. Since these adverse consequences were observed on the opening of BK Ca channels, our results suggest that under physiological conditions plasma membrane BK Ca channels in NCM are not active.
In adult experimental models, activation of BK Ca channels by NS1619 and NS11021 protected the hearts against IR injury [24,32,[65][66][67][68][69][70][71] and inhibition of BK Ca channels by IbTx or paxilline increased myocardial infarction or reversed cardioprotective effects caused by the opening of BK Ca channels [23,32,33,69,72,73]. Pharmacological data were inadvertently supported by genetic models comprising global [13], cardiomyocyte-specific [23], or gain-of-function transgenic mice for BK Ca channels [22]. The cardioprotective mechanism mediated by the BK Ca channel is associated with an increase in mitochondrial ROS generation and Ca 2+ overload, which in turn triggers the formation and opening of mitochondrial permeability transition pore [21]. These findings are in agreement with the notion that stored potassium in intracellular organelles such as mitochondria protects cells from apoptosis [74,75].
In our study, we discovered that the opening of BK Ca channels at an organ or cellular level causes apoptosis and increases myocardial infarction. The contrasting outcome of activation of BK Ca channels in neonates as opposed to adults could be attributed to the plasma membrane localization and opening of BK Ca channels. Our data indicate that the opening of BK Ca channels at the plasma membrane in neonates will cause membrane depolarization, which in turn will trigger cell death. Under physiological conditions, BK Ca channels are not anticipated to open in neonates, which will not cause any adverse effects on cardiac cells. However, under pathological conditions, such as dilated cardiomyopathy in children, there is evidence of increased sensitivity of myocytes to Ca 2+ , and pediatric cardiomyocytes have decreased cooperativity when compared with adult cardiomyocytes [76]. The increase in cellular Ca 2+ could trigger the opening of the plasma membrane BK Ca channel, which will immediately cause depolarization and heart failure. This could be one of the key reasons why the use of proven adult heart failure pharmacological interventions in pediatric or children with dilated cardiomyopathy do not present the same beneficial outcome as adults. Therefore, it is important to take the presence of the BK Ca channel into account before selecting the appropriate pharmacological therapies in the pediatric population.

MATERIALS AND METHODS
All experiments were conducted in accordance with guidelines and approved by The Ohio State University IACUC committee. C57BL/6NCrL mice and Sprague-Dawley (SD) were used. BK Ca R207Q mice were obtained from Andrea Meredith (University of Maryland School of Medicine, Baltimore) and locally bred.

Neonatal P3 cardiomyocytes isolation
Hearts were surgically excised from postnatal day 3 (P3) pups (from rats and mice) and placed in a dissociation buffer (NaCl 16 mM, HEPES 20 mM, Na 2 HPO 4 0.8 mM, glucose 5.4 mM, KCl 5.4 mM, MgSO 4 0.8 mM, pH 7.35). Ventricles were minced in dissociation buffer followed by centrifugation at 200 x g for 5 min at 4°C. The pellet was then resuspended in a dissociation buffer containing 0.25% Trypsin and the digestion was carried out at 37°C for 20 min in the water bath shaker. After 20 min, the cells were spun down (200 x g, 5 min, 4°C) and the dissociated neonatal cardiomyocytes were passed through a cell strainer (100 µm) and enriched with 20% (v/v) horse serum. The digestion of the remaining tissue pellet was repeated at least twice or until a clear tissue pellet was observed. The dissociated neonatal cardiomyocytes were then seeded on gelatin (0.1%, w/v) coated coverslips and cultured in DMEM containing 20% (v/v) FBS and penicillin (100 I.U.)/streptomycin (100 µg/ ml), cultured in humidified 5% (v/v) CO 2 incubators at 37°C and medium change every 2 days.
Neonatal cardiomyocyte isolation (P7, P14, P21, and P28) Cardiomyocytes were isolated from postnatal day 7 (P7), day 14 (P14), day 21 (P21), and day 28 (P28) pups, respectively using the simplified Langendorff-free method of isolation (1). Briefly, pups were anesthetized, and the chest was cut open to expose the heart. The descending aorta was cut followed by injection of 7 ml EDTA buffer [NaCl 130 mM, KCl 5 mM, NaH 2 PO 4 0.5 mM, HEPES 10 mM, glucose 10 mM, Butanedione Monoxime (BDM) 10 mM, Taurine 10 mM, EDTA 5 mM, pH 7.8] into the right ventricle. The ascending aorta was clamped, and the heart was transferred to a Petri dish containing EDTA buffer. The left ventricle was injected with 10 ml of EDTA buffer followed by 3 ml of perfusion buffer (NaCl 130 mM, KCl 5 mM, NaH 2 PO 4 0.5 mM, HEPES 10 mM, glucose 10 mM, BDM 10 mM, Taurine 10 mM, MgCl 2 1 mM, pH 7.8), and 10-20 ml of collagenase enzyme solution (collagenase type 2 0.5 mg/ml, collagenase type 4 0.5 mg/ml, protease XIV 0.05 mg/ml prepared in perfusion buffer). The heart chambers were separated and gently pulled into 1-mm pieces. The cells were dissociated using gentle trituration and the collagenase activity was inhibited by 5 ml of stop buffer [perfusion buffer containing 5% (v/v) sterile fetal bovine serum (FBS)]. The cells were washed three times with perfusion buffer and allowed to settle with gravity.

Hypoxia/reoxygenation treatment
Neonatal cardiomyocytes were subjected to hypoxia/reoxygenation injury. After 4 days of culturing the neonatal cardiomyocytes with 70% confluency in cell culture dishes or coverslips coated with 0.1% gelatin, cells were subjected to hypoxia in a modular humidified 37°C hypoxia chamber (Biospherix, C127) with 1% O 2 , 5% CO 2 and balanced N 2 for 6 h. After hypoxia incubation, cells were conditioned with DMSO control, BK Ca channel inhibitor (iberiotoxin), and activator (NS1619) for 30 min followed by reperfusion with DMEM supplemented with 10% FBS and 1% penicillinstreptomycin antibiotics.

Total RNA isolation and qPCR
Total RNA from the gravity-settled cardiomyocytes was then purified using TRIzol reagent (Invitrogen) followed by digestion with on column-RNasefree DNAse digestion kit (Qiagen) and clean up with RNeasy mini kit (Qiagen). Cleaned-up RNA (0.5 µg) was reverse transcribed with Omniscript Reverse Transcription kit (Qiagen) using oligo dT primers in a 20 µL reaction volume. Real-time PCR was performed using SYBRTM Green master mix (Applied Biosystems), 1 µL of RT reaction product, and 200 nM primer pairs (supplementary table 1) in a 10 µL reaction volume. The thermal cycling conditions included an initial denaturation at 95°C for 5 min, and 40 cycles of 95°C for 45 s, 60°C for 45 s, and 72°C for 45 s. The controls used in the qPCR are (-)RT (cDNA with no reverse transcriptase), Water control (water instead of the template), and primers used to amplify GAPDH (2). All samples were run in duplicates. The efficiency of primers was calculated to be more than 94% for BK Ca -DEC and BK Ca -Total (3). The fold-change in expression of BK Ca -Total and BK Ca -DEC have been plotted relative to GAPDH.

Immunocytochemistry
Rodent and mouse neonatal cardiomyocytes were incubated with wheat germ agglutinin (WGA) at 37°C on ice for 60 min and/or mitotracker for 10 min at 37°C. Cells were fixed with 4% (w/v) PFA and permeabilized with 0.5% (v/v) Triton-X-100. Neonatal cardiomyocytes were incubated with anti-ATP synthase (mouse, Abcam; ab14748), and anti-BK Ca antibodies, overnight at 4°C. Secondary antibodies conjugated with anti-mouse Alexa-488 (CST; 4408) and anti-rabbit Atto-647N (Sigma-Aldrich; 40839) were added for 60 min at room temperature. To label nuclei, DAPI was added (1:10,000) in the wash solution. Coverslips were mounted with Mowiol®. Cells were imaged with Nikon A1R highresolution confocal microscopy. The colocalization index was calculated using protein proximity index analysis (4). Images were filtered by custom-built software as described earlier.

TUNEL assay
The extent of the apoptotic damage was monitored using a deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay kit (Thermo Scientific; C10618). Briefly, P3 neonatal cardiomyocytes were cultured on 24-well plates coated with 0.1% gelatin at 70% confluency for 96 h, hypoxia injury, and postconditioning with DMSO, NS1619, IbTx, and Pax was induced as described earlier. After 12 h of reoxygenation, the cells were fixed with 4% (v/v) PFA for 10 min, permeabilized with 0.25% (v/v) Triton-X-100 for 20 min, and apoptotic nuclei were stained using a TUNEL assay kit as described by the manufacturer's protocol. Fluorescent images were taken using Nikon A1R confocal microscope.

Patch-clamp
Patch-clamp experiments were performed on mouse and rat neonatal cardiomyocytes 96 h post isolation. Cells were cultures on coverslips coated with poly-D-lysine. Patch electrodes were fabricated from borosilicate glass (Sutter Instrument, Navato, CA) on an M-87 horizontal puller (Sutter Instrument, Navato, CA), with an average resistance of 3-6 MΩ. All the recordings were performed at room temperature and currents were acquired using an EPC10 USB amplifier (HEKA Electronic, Germany) with accompanying PatchMaster (HEKA, Germany) software that was also used for analysis. Patch-clamp experiments were performed in whole-cell and inside-out mode. The bath solution contained (KCH 3 SO 3 140 mM, MgCl 2 2 mM, KCl 2 mM, HEPES 20 mM, pH 7.3). Pipette solution contained (KCH 3 SO 3 140 mM, CaCl 2 10 mM, HEDTA 5 mM, HEPES 20 mM, pH 7.3. During whole-cell experiments cells were held at -70 mV, 40 ms pulses were applied from -70 to +150 mV in 20 mV steps. After establishing stable baseline currents 100 nM IbTx was added to the bath solution, to validate the ion channel responsible for detected K + currents. Open probabilities (P o ) for all the recordings were obtained from a minimum of 20 s single-channel recordings held at +80 mV before and after the addition of IbTx, as described earlier [77].

Langendorff isolated perfused neonatal rat hearts
Six-day-old Sprague-Dawley (SD) rat pups were injected intraperitoneal with heparin (100 IU/Kg) to prevent blood coagulation. After 20 min, animals were anesthetized in an isoflurane chamber. Hearts were rapidly excised and arrested in ice-cold filtered Krebs-Henseleit (KH) buffer containing (Glucose 11.1 mM, NaCl 118 mM, KCl 4.7 mM, MgSO 4 1.2 mM, KH 2 PO 4 1.2 mM, NaHCO 3 25 mM, and CaCl 2 2 mM at pH 7.4). The hearts were cleaned to remove non-cardiac tissue and excess fat before cannulating the aorta onto a 21-gauge cannula. The hearts were retrogradely perfused with KH buffer at a constant rate of 2-3 ml/min and maintained at 37°C in the water-jacketed Langendorff chamber. The buffer was constantly aerated with a mixture of 95% O 2 (v/v) and 5% CO 2 (v/v). After equilibrating the hearts by perfusing with KH buffer for 15 min, the hearts were subjected to global normothermic ischemia for 20 min by stopping the perfusion. Following the ischemia, hearts were then randomly post-conditioned with NS1619 (10 μM), IbTx (100 nM), and DMSO (0.001%) for 5 min and reperfused with KH buffer for 60 min at 37°C.

Myocardium infarct size
Myocardium infarction was measured by staining the heart sections with 2, 3, 5-triphenyl tetrazolium chloride (TTC) stain. After reperfusion for 60 min, the hearts were cut into five transverse sections parallel to the atrioventricular groove. Heart slices were incubated in 1% TTC (w/v) at 37°C for 20 min followed by fixation in 4% PFA (w/v). The stained sections were imaged using Leica S9i. To demarcate the infarcted region (pale) versus viable myocardial tissue (brick red) ImageJ was used to quantify the area. The total infarcted area was calculated against the total heart sections and expressed as the percentage.

Multi-electrode array (MEA) measurements
Rat neonatal cardiomyocytes were isolated from P3 pups using the Pierce primary cardiomyocyte isolation kit (Thermo Scientific). The neonatal cardiomyocytes were seeded on (50 μg/ml) fibronectin-coated CytoView MEA 24-well plates (Axion BioSystems, Inc.) at 100,000 cells per well and data were acquired 72 h post isolation using Maestro Edge MEA platform (Axion BioSyetems, Inc.). The voltage data were sampled for 16 electrodes/ well simultaneously at (15 µV). LEAP inductions were performed using the AxIS Navigator software (Axion BioSystems, Inc.) using the dedicated stimulator in the Maestro Edge on selected planar microelectrodes on 24well CytoView plates. For data analysis of (1) cardiac field potential duration (FPD), (2) beat period, (3) spike amplitude, and (4) conduction velocity were plotted as fold-change after comparing with treatment groups (DMSO, IbTx, or NS1619) to their respective baseline wells. Graphical representation of the data was automatically detected and plotted using CiPA Analysis Tool software (Axion BioSystems, Inc.). Cardiac LEAP signals were detected using AxIS Navigator and representative data shown were plotted against DMSO control. Cardiac LEAP signals (a) rise time and (b) action potential duration (APD), APD30, APD50, and APD90 against DMSO were calculated and detected using CiPA Analysis Tools. Recorded APDs were corrected by the beating frequency (APDc) with Fridericia's formula (APDc = APD/interspike interval1/3) [78].

Immunohistochemistry of human hearts
Human right ventricular myocardial sections were obtained from the Heart Center Biorepository at Nationwide Children's Hospital. This study was approved by the Institutional Review Board (IRB) at Nationwide Children's Hospital (IRB #07-00298). The myocardial sections were from infants under 3 months of age, with the following congenital heart defects (tetralogy of Fallot, truncus arteriosus, and pulmonary stenosis). The myocardial tissues were fixed in PFA and sectioned for histological analysis. For immunohistochemical staining, tissue sections were deparaffinized in xylene and rehydrated in grades of ethanol and PBS, followed by antigen retrieval using citrate-based buffer (Vector Laboratories# H-3300-250). Sections were incubated at room temperature with 3% (v/v) H 2 O 2 for 10 min to quench endogenous peroxidase activity and blocked by 5% (v/v) normal goat serum (Vector Laboratories, #S-1000) in PBS for 1 h. After blocking, sections were incubated with WGA (1:500) for 60 min at room temperature to the label plasma membrane. After 60 min sections were washed with PBS three times. Sections were incubated in 0.5% (v/v) PBS-Triton-X-100 for 10 min at room temperature for permeabilization. After washes with PBS sections were incubated in a primary antibody against BK Ca channels (Alomone Labs, APC021) overnight at 4°C. Following primary antibody incubation, sections were incubated with secondary antibodies. During washing DAPI (1:10,000 dilution) was added and sections were washed with PBS. Cardiac sections were visualized with a high-resolution confocal microscope (Nikon A1R).

Statistical analysis
All experiments were repeated independently for a minimum of four biological repeats. Data were analyzed using Prism (GraphPad) or Excel and reported as mean ± SEM. To assess the significant difference, comparisons were measured between two groups using a paired onetailed Student's t-test. The significance between the groups in apoptotic nuclei and ROS were compared using a one-tailed unpaired student's t-test followed by a non-parametric Mann-Whitney test for comparison between the ranks. P ≥ 0.05 indicated non-significant, P ≤ 0.05 indicated significant.

DATA AVAILABILITY
All data and materials are available on request to the corresponding author.