Hydrogen sulphide increases pulmonary veins and atrial arrhythmogenesis with activation of protein kinase C

Abstract Hydrogen sulphide (H2S), one of the most common toxic air pollutants, is an important aetiology of atrial fibrillation (AF). Pulmonary veins (PVs) and left atrium (LA) are the most important AF trigger and substrate. We investigated whether H2S may modulate the arrhythmogenesis of PVs and atria. Conventional microelectrodes and whole‐cell patch clamp were performed in rabbit PV, sinoatrial node (SAN) or atrial cardiomyocytes before and after the perfusion of NaHS with or without chelerythrine (a selective PKC inhibitor), rottlerin (a specific PKC δ inhibitor) or KB‐R7943 (a NCX inhibitor). NaHS reduced spontaneous beating rates, but increased the occurrences of delayed afterdepolarizations and burst firing in PVs and SANs. NaHS (100 μmol/L) increased IKATP and INCX in PV and LA cardiomyocytes, which were attenuated by chelerythrine (3 μmol/L). Chelerythrine, rottlerin (10 μmol/L) or KB‐R7943 (10 μmol/L) attenuated the arrhythmogenic effects of NaHS on PVs or SANs. NaHS shortened the action potential duration in LA, but not in right atrium or in the presence of chelerythrine. NaHS increased PKC activity, but did not translocate PKC isoforms α, ε to membrane in LA. In conclusion, through protein kinase C signalling, H2S increases PV and atrial arrhythmogenesis, which may contribute to air pollution‐induced AF.

bacterial breakdown of sulphur-containing matter and can be found in various natural environments and industrial settings, such as spas, sewers, landfills, waste water plants and oil refineries. 11,12 Increases in the H 2 S concentration are associated with daily all-natural-cause mortality 13 and cardiovascular hospitalization. 14 However, it is not elucidated whether H 2 S may play a role in the pathophysiology of air pollution-induced AF.
Pulmonary veins (PVs) and left atrium (LA) are the most important AF triggers and substrates. [15][16][17] Calcium dysregulation plays a critical role in the occurrences of AF and PV arrhythmogenesis. 18,19 The activation of Na + /Ca 2+ exchangers (NCXs) induces delayed afterdepolarizations (DADs) and increases PV arrhythmogenic activity. 18 Protein kinase C (PKC)-mediated signalling plays a vital role in NCX activation. 20 H 2 S was known to activate PKC, 21 thus H 2 S may increase I NCX and increase PV arrhythmogenesis leading to AF genesis. Moreover, sinoatrial node (SAN) dysfunction plays an important role in AF pathophysiology 22 and increases PV arrhythmogenesis. 23 H 2 S reduces the electrical activity of SANs, which may modulate the arrhythmogenesis of PVs and increase the risk of AF genesis. 24 H 2 S plays a critical role in cell signalling 25 and attenuates ischaemia-reperfusion injury by activating the ATP-sensitive potassium channel (K ATP ). 26 The activation of the K ATP channel shortens the action potential duration (APD), which may increase the risk of AF by facilitating the genesis of re-entry circuits. Previous study has revealed that I KATP differentially regulates the electrical activity of right atrium (RA) and LA. 27 The different effects of I KATP on APD shortening in the LA and RA can increase the risk of cardiac arrhythmias due to increasing interatrial dispersion. Accordingly, H 2 S may modulate the electrical activity of PVs, atria and SANs and increase the risk of air pollution-induced AF. This study explored the effects of H 2 S on PVs, atria and SANs, and investigated the potential underlying mechanisms.
2 | ME TH ODS 2.1 | Animal and tissue preparation All experimental procedures were approved by the Institutional Animal Care and Use Committee of Taipei Medical University and conformed to the Institutional Guidelines for the Care and Use of Laboratory Animals and the Guide for the Care and Use of Laboratory Animals published by the United States National Institute of Health. As described previously, 28 male rabbits (1.5-2 kg) were anaesthetized with an intravenous injection of sodium pentobarbital (100 mg/kg). The adequacy of the anaesthesia was confirmed by the lack of corneal reflexes and motor responses to pain stimuli induced using a scalpel tip. The heart and lungs were rapidly excised following midline thoracotomy. For SAN tissue preparation, SANs were isolated from the RA and superior vena cava. PVs were separated from the atria at the LA-PV junction and from the lungs at the end of PV myocardial sleeves. One end of the preparation was pinned to the bottom of a tissue bath using needles, and the other end was connected to a Grass Instruments FT03C force transducer (MA, USA) using silk thread. Tissue preparations were superfused with normal Tyrode's solution composed of NaCl (137 mmol/L), KCl (4 mmol/L), NaHCO 3 (15 mmol/L), NaH 2 PO 4 (0.5 mmol/L), MgCl 2 (0.5 mmol/L), CaCl 2 (2.7 mmol/L) and dextrose (11 mmol/L) at a constant rate of 3 mL/min at 37°C as described previously. 18,19,23,27 NaHS (Sigma, MO, USA) was used as a donor of H 2 S. PVs, atria and SANs were exposed to different concentrations of NaHS (1, 10, and 100 lmol/L) in Tyrode's solution for 40 minutes to investigate the electrophysiological effects of H 2 S.

| Electrophysiological and pharmacological studies
The transmembrane APs of PVs, SANs and atria were recorded using machine-pulled glass capillary microelectrodes filled with 3 mol/L KCl; the microelectrodes were connected to a World Precision

| Electropharmacological studies in isolated single PV and atrial cardiomyocytes
Pulmonary vein and atria cardiomyocytes from rabbits were enzymatically dissociated, as previously described. 29,30 The whole-cell patch clamp technique was performed in the PV and atrial cardiomyocytes with pacemaker activity before and after the administration of NaHS with or without chelerythrine; the APs were recorded using CsOH). The I NCX was elicited through depolarization in 10-mV steps from a holding potential of À40 mV to test potentials from À100 to +100 mV for 300 mseconds at a frequency of 0.1 Hz. The I NCX amplitudes were measured as 10-mmol/L nickel-sensitive currents.  were normalized to GAPDH to confirm equal protein loading.

| Protein kinase C activity assay
Protein kinase C activity was assayed using PKC Kinase Activity Assay Kit (Abcam) as manufacturer's instructions. Briefly, total proteins from LA tissues with or without NaHS (100 lmol/L) incubation for 40 minutes were added to PKC substrate coated wells of a 96-well microtitre plate, and reactions were initiated by adding ATP. The phosphorylated substrates were recognized by a phosphor-PKC substrate-specific antibody and a secondary antibody conjugated with HRP after being reacted at 30°C for 90 minutes.
Bound antibodies were detected with TMB substrate, and the absorbance was measured at OD450 nm. Relative kinase activity was calculated from standard curve and normalized to individual control.

| Statistical analyses
All continuous variables are expressed as mean AE standard error of mean. One-way repeated-measures analysis of variance followed by Bonferroni's analysis was used to compare the differences in PVs, SANs and LA before and after drug administration. The chi-square analysis with Fisher's exact test was used to compare the incidences of DADs and burst firing in PVs and SANs before and after drug administration. P < .05 was considered statistically significant. Statistical analysis was performed using SigmaPlot 12 (Systat software).

| RESULTS
3.1 | Effects of H 2 S on the electrical activity of

PVs, and SANs
As shown in Figure 1A, NaHS (1, 10, and 100 lmol/L) significantly reduced the PV beating rates in a concentration-dependent manner.
As shown in Figure 3

| Effects of H 2 S on atrial electrical activity
NaHS at 100 lmol/L, but not at 1 and 10 lmol/L, significantly shortened APD 90 and reduced the contractility of the LA ( Figure 4A).

| Effects of H 2 S on I KATP and I NCX
We investigated the effects of NaHS on I KATP and I NCX in isolated single PV and atrial cardiomyocyte. As shown in Figure 5, NaHS (100 lmol/L) significantly increased the I KATP and the forward mode of the I NCX . However, in the presence of chelerythrine (3 lmol/L), NaHS did not change the I KATP and I NCX in PV cardiomyocytes.
We compared the effects of NaHS on I KATP and I NCX in LA and RA cardiomyocytes. As shown in Figure 6, NaHS significantly increased the I NCX and I KATP in LA cardiomyocytes, but not in RA cardiomyocytes.

| Effect of NaHS on translocation of PKC isoforms, PKC activity and ROS
As shown in Figure 7A,B, NaHS did not change the membrane to cytosol ratios of PKC a and e in LA. However, NaHS-treated LA had larger PKC activity than those without treatment. As shown in 7C, NaHS-treated PV cardiomyocytes had lower ROS in cytosol than did control PV cardiomyocytes.

| DISCUSSION
Air pollution is caused by multiple air pollutants, including H 2 S. This study is the first to report that H 2 S induces the occurrences of DADs and burst firing in PVs and SANs. H 2 S-induced PKC activation is reported to play a role in regulating intracellular calcium handling by facilitating cytosolic calcium clearing through NCX channel in the development of calcium overloading and cardiomyocyte hypercontraction induced by ischaemic-reperfusion insults. 26 In single-cell experiments, we revealed that H 2 S increased the forward mode of the I NCX in the PV cardiomyocytes, which was attenuated by PKC inhibition. In addition, we observed that KB-R7943 and chelerythrine suppressed NaHS-induced PV and SAN arrhythmogenesis. These results suggest that H 2 S-induced PKC signalling increases PV and SAN arrhythmogenesis with the activation of NCX.
Hydrogen sulphide has been reported to exert a negative chronotropic effect in SANs; this effect is inhibited by the K ATP channel F I G U R E 4 Effects of NaHS interacted with PKC inhibitors on action potential morphology and contractility of the left atrial (LA) and right atrial (RA) preparations. A, Superimposed tracings, contractility and average data of the action potential parameters in the LA preparations (n = 7) before and after superfusion with different concentrations of NaHS (1, 10 and 100 lmol/L). B, Superimposed tracings, contractility and average data of the action potential parameters in the RA preparations (n = 8) before and after superfusion with different concentrations of NaHS (1, 10, and 100 lmol/L). C, Superimposed tracings, contractility and average data of the action potential parameters in LA preparations (n = 6) pretreated with chelerythrine (3 lmol/L) and after superfusion with NaHS (100 lmol/L) blocker, glibenclamide. 33,34 Similarly, this study found that H 2 S significantly reduced SAN beating rates, and this effect was attenuated by PKC inhibition. In single-cell experiments, we revealed that NaHS increased the I KATP in PV cardiomyocytes, which was attenuated by PKC inhibition. Because PKC activation is required for K ATP channel opening, these results indicate that H 2 S modulates SAN function by activating PKC and K ATP channels. SAN dysfunction plays an important role in AF pathophysiology 22,35 and increases PV arrhythmogenesis. 23,36 Accordingly, H 2 S may modulate SAN function and result in PV arrhythmogenesis and AF occurrence.
In the present study, NaHS increased I KATP and I NCX in PV cardiomyocytes, which were attenuated by chelerythrine (a selective PKC inhibitor). Additionally, chelerythrine and rottlerin (a specific PKC d inhibitor) attenuated the arrhythmogenic effects of NaHS, thereby suggesting that PKC may mediate the effects on membrane ion currents caused by H 2 S. H 2 S diffuses through the cell membrane directly because the H 2 S molecule is very small and non-polar. H 2 S has been shown to activate different PKC isoforms directly. 21 Protein kinase-catalysed phosphorylation can regulate the activity of ion channels, including the K ATP and NCX channel. 37 The activation of PKC increases the open probability of K ATP channel and acts via phosphorylation of a specific, conserved threonine residue in the K ATP channel. 38 In addition, PKC directly phosphorylates NCX channel, which significantly enhances I NCX . 39 Protein kinase C exists as several different isoforms and six isoforms (a, b, d, e, g, and f) were detected in hearts, among which PKC isoforms a, d and e are the prominent isoforms expressed in the heart. Moreover, chelerythrine is well-known to inhibit PKC a, b1, c and d. Therefore, PKC a and PKC d are more likely to be essential to NaHS-mediated arrhythmogenesis. We found that NaHS did not induce translocation of PKC isoforms a, e from cytosol to membrane but did increase PKC kinase activity. In the presence of F I G U R E 5 Effects of NaHS on I NCX and I KATP in PV cardiomyocytes. A, The tracings and current-voltage relationship of I NCX in PV cardiomyocytes before and after NaHS (100 lmol/L) with (n = 8) and without (n = 9) chelerythrine (3 lmol/L). B, The tracings and current-voltage relationship of I KATP in PV cardiomyocytes before and after NaHS (100 lmol/L) with (n = 8) and without (n = 7) chelerythrine (3 lmol/L). The insets in the current traces show the various clamp protocols. *P < .05, **P < .01, ***P < .005 vs baseline CHAN ET AL.
These findings suggested that H 2 S activates PKC d and results in its arrhythmogenesis.
The present study revealed that H 2 S differentially changed the cardiac electrophysiology of the LA and RA, whereas H 2 S significantly shortened the APD and reduced the contractility of the LA but not of the RA. These effects were attenuated by chelerythrine, suggesting that PKC signalling plays a vital role in the effects of H 2 S.
The different effects of H 2 S on APD shortening in the LA and RA increase the dispersion of the APD, facilitating the maintenance of cardiac arrhythmias. Nevertheless, the mechanisms underlying the different effects of H 2 S in the LA and RA are unclear. A previous study reported that the higher expression of heat stress protein 70 in the RA may attenuate the response of the RA to the activation of the K ATP channel by hypoxia and reperfusion. 40,41 Previous studies have shown that LA plays a critical role in AF genesis compared to RA. Therefore, H 2 S may have different electrophysiological effects on RA and LA cardiomyocytes. We found that NaHS significantly increased the I NCX and I KATP in LA cardiomyocytes but not in RA cardiomyocytes, which may result in the shortening of APD in NaHStreated LA.
Air pollutant is known to increase oxidative stress. We evaluated the effects of H 2 S on oxidative stress in PV cardiomyocytes by measurement of intracellular ROS using a laser scanning confocal microscope and found that NaHS-treated PV cardiomyocytes had lower cytosol ROS than did control PV cardiomyocytes. Similarly, previous study has shown that H 2 S reduces oxidative stress in mouse model. 42 These findings suggested that oxidative stress does not underlie the effects of H 2 S on cardiomyocytes, and H 2 S may activate PKC through its direct chemical effects, leading to the increases in I NCX and I KATP.
The effects of H 2 S has been extensively studied as an environmental pollutant. 43,44 Although H 2 S has been widely recognized as a cardioprotective agent for majority of cardiac disorders such as F I G U R E 6 Effects of NaHS on I NCX and I KATP in atrial cardiomyocytes. A, The tracings and current-voltage relationship of I NCX in left atrial (LA, n = 8) and right atrial (RA, n = 9) cardiomyocytes before and after NaHS (100 lmol/L). B, The tracings and current-voltage relationship of I KATP in LA (n = 9) and RA (n = 8) cardiomyocytes before and after NaHS (100 lmol/L). The insets in the current traces show the various clamp protocols. *P < .05, **P < .01 vs baseline myocardial infarction/reperfusion injury, cardiac hypertrophy, myocardial fibrosis and heart failure, 45 acute exposures of H 2 S may cause cardiac arrhythmia. 46 Inhaled H 2 S induces sinus bradycardia and sinus arrest. 47 Circulating halogen reactants cause cardiac injury by damaging important intracellular Ca 2+ regulators. 48 Although we do not provide a direct relationship between H 2 S and AF, our works suggest that H 2 S increases PV and SAN arrhythmogenesis and regulates atrial electrophysiology which contribute to AF.
This study should be interpreted with cautions due to the potential limitations. Air pollutants may trigger the occurrence of AF via direct and/or indirect effects on the atrial myocardium. 49 In this study, we found that H 2 S has direct electrophysiological effects on AF substrates and triggers, supporting that H 2 S may contribute to air pollution-induced AF at least in part. However, simply investigating H 2 S may not fully uncover the mechanisms of polluted airinduced PV and atrial arrhythmogenesis as air pollution contains multiple pollutants in addition to H 2 S. In addition, this study found that chelerythrine or rottlerin had inhibitory effects on NaHS-induced arrhythmogenesis, and NaHS-treated atrium had larger PKC activity than those without treatment, suggesting that PKC pathway plays a crucial role in H 2 S-mediated arrhythmogenesis. Nevertheless, chelerythrine is an inhibitor with multiple functions, it is also an antagonist of G-protein-coupled CB 1 receptors. 50 Studies have suggested that activation of CB 1 receptors promotes activation of mitogen-activated protein kinases p38 and JNK. 51 Mitogen-activated protein kinase is known to be functional in cardiomyocytes and is activated in response to stress, reactive oxygen species and inflammation. 52 Rottlerin, a PKC d inhibitor, 53 is also an uncoupler of mitochondrial oxidative phosphorylation. 54 Therefore, the electrophysiological data in this study did not exclude the possibility that several signalling pathways may involve the effects of H 2 S-mediated arrhythmogenesis. The precise signalling underlying the effects of H 2 S may not be fully elucidated.

| CONCLUSION
Hydrogen sulphide increases the arrhythmogenesis of PVs and SANs and differentially regulates the cardiac electrophysiology of the LA and RA. The activation of PKC signalling and increases in the I KATP and I NCX induced by H 2 S in PV and SAN cardiomyocytes may contribute to air pollution-induced AF.

CONFLI CTS OF INTEREST
The authors have no conflict of interest to disclose. F I G U R E 7 Effects of NaHS on membrane translocation of PKC isoforms, PKC activity and intracellular reactive oxygen species (ROS). A, Representative immunoblot and average data of subcellular distribution of PKC e (n = 3) and PKC a (n = 5) from control and NaHS-treated left atrial tissue preparations. B, The average data of relative PKC activity in control and NaHS-treated left atrial tissue preparations (n = 3). C, An example and average data of intracellular ROS in control (n = 13) and NaHS-treated (n = 13) PV cardiomyocytes. *P < .05, ***P < .005 vs control