Research and Development Unit, National Heart Centre Singapore, Singapore, Singapore
Duke-NUS, Graduate Medical School, Singapore, Singapore
Correspondence to: Winston SHIM, Research and Development Unit, National Heart Centre, 9 Hospital Drive, School of Nursing, #03-02, Block C, SingHealth Research Facilities, Singapore 169612, Singapore.
Cardiac fibroblasts are crucial in pathophysiology of the myocardium whereby their aberrant proliferation has significant impact on cardiac function. Hydrogen sulphide (H2S) is a gaseous modulator of potassium channels on cardiomyocytes and has been reported to attenuate cardiac fibrosis. Yet, the mechanism of H2S in modulating proliferation of cardiac fibroblasts remains poorly understood. We hypothesized that H2S inhibits proliferative response of atrial fibroblasts through modulation of potassium channels. Biophysical property of potassium channels in human atrial fibroblasts was examined by whole-cell patch clamp technique and their cellular proliferation in response to H2S was assessed by BrdU assay. Large conductance Ca2+-activated K+ current (BKCa), transient outward K+ current (Ito) and inwardly rectifying K+ current (IKir) were found in human atrial fibroblasts. Current density of BKCa (IC50 = 69.4 μM; n = 6), Ito (IC50 = 55.1 μM; n = 6) and IKir (IC50 = 78.9 μM; n = 6) was significantly decreased (P < 0.05) by acute exposure to NaHS (a H2S donor) in atrial fibroblasts. Furthermore, NaHS (100–500 μM) inhibited fibroblast proliferation induced by transforming growth factor-β1 (TGF-β1; 1 ng/ml), Ang II (100 nM) or 20% FBS. Pre-conditioning of fibroblasts with NaHS decreased basal expression of Kv4.3 (encode Ito), but not KCa1.1 (encode BKCa) and Kir2.1 (encode IKir). Furthermore, H2S significantly attenuated TGF-β1–stimulated Kv4.3 and α-smooth muscle actin expression, which coincided with its inhibition of TGF-β–induced myofibroblast transformation. Our results show that H2S attenuates atrial fibroblast proliferation via suppression of K+ channel activity and moderates their differentiation towards myofibroblasts.
Cardiac fibroblasts are fundamentally involved in cardiac remodelling in normal ageing heart  and in damaged myocardium . Aberrant proliferation of fibroblasts and their transformation to myofibroblasts is a hallmark of cardiac fibrosis, which is characterized by excessive extracellular matrix built-up leading to loss of tissue compliance [3, 4]. Because of their wide-ranging participation in myocardial pathophysiology, cardiac fibroblasts represent an attractive target in managing cardiac disorders, including cardiac hypertrophy, heart failure and arrhythmias . Indeed, atrial fibrosis has been closely associated with atrial fibrillation [6, 7] and sinus node dysfunction .
Hydrogen sulphide (H2S) is an endogenously generated gaseous transmitter that has been reported to attenuate cardiac fibrosis . It is known to mediate its effects by modulating ion channel activity in many cellular systems . Hydrogen sulphide was the first opener of KATP channel identified in vascular smooth muscle cells . Through activation of KATP channels, H2S lowers blood pressure, protects heart from ischaemia and reperfusion injury [12, 13]. We have recently reported that H2S inhibited delayed rectifier potassium channels in human iPS–derived cardiomyocytes . Yet, effect of H2S on cardiac fibroblasts remains poorly understood. We hypothesized that H2S inhibits proliferation of atrial fibroblasts by inhibiting functioning of potassium channels. We present supporting data that H2S may potentially modulate cardiac fibrosis by inhibiting BKCa, Ito and IKir, independent of KATP channels, leading to decreased proliferation and suppression of transforming growth factor-β1 (TGF-β1)–induced myofibroblast transformation of atrial fibroblasts.
Materials and methods
Patients undergoing mitral valve repair and coronary bypass surgery (n = 10) were recruited after informed consent in protocol approved by institutional review board of Singapore General Hospital that conformed to the Declaration of Helsinki. Atrial appendages were collected as surgical by-product. Human atrial fibroblasts were isolated by mincing the appendages to less than 1 mm3 and followed by 0.1% trypsin digestion for 20 min. before plating onto tissue culture–treated 60-mm dishes to produce fibroblastic outgrowth from minced tissue pieces. The isolated fibroblasts were confirmed with expression of collagen I (1/20; Southern Biotech, Birmingham, AL, USA) and anti-human fibroblast (1/1000; Sigma-Aldrich, St. Louis, MO, USA) antibodies (Fig. S1). Atrial fibroblasts were passaged as monolayer in 10% foetal bovine serum–supplemented DMEM. Fibroblasts between passage 1 and 3 were used for subsequent experiments.
Cell were placed on the stage of a Nikon Diaphot inverted microscope and superfused continuously at 36 ± 1°C with Tyrode solution containing (in mM) 140 NaCl, 5.4 KCl, 1.8 CaCl2, 1 MgCl2, 10 HEPES and 10 Glucose (pH adjusted to 7.4 with NaOH). The patch-clamped cell was superfused by means of a temperature-controlled micro-superfusor (TC-324B, Warner Instruments, Hamden, CT, USA). Patch pipettes were made from borosilicate glass shanks (Sutter Instrument, Novato, CA, USA) and pulled with a Brown–Flaming puller (Model P-97; Sutter Instrument Co), and had tip resistances of 2–3 MΩ when filled with pipette solution. Pipette tips were polished (Microforge MF830; Narishige, Tokyo, Japan). These patch pipettes were filled with a standard solution containing (in mM) 140 KCl, 1.2 MgCl2, 0.05 EGTA, 10 HEPES, 0.1 GTP and 5.0 Mg ATP (pH adjusted to 7.2 with KOH). For Na+ current recording, the patch pipettes were filled with (in mM) 35 NaCl, 105 CsF, 0.1 EGTA and 10 HEPES (pH adjusted to 7.4 with CsOH). After a gigaohm seal was obtained by negative pressure suction, the cell membrane was ruptured by a gentle suction to establish whole-cell configuration with a seal resistance >800 MΩ. The cell membrane capacitance (40.27 ± 8.2 pF) was electrically compensated with the pulse software. The series resistance (Rs, 3–5 MΩ) was compensated by 50–70% to minimize voltage errors. Currents were elicited with voltage protocols as described in the following results section for different individual current recordings. Whole-cell voltage-clamp experiments were performed with an Axopatch 200B amplifier (Axon Instruments, Foster City, CA, USA) interfaced to a Digidata 1322A data acquisition system controlled by Clampex version 8.1 software (Axon Instruments). Data were analysed with pCLAMP software (Version 10.0; Axon Instrument) and Origin 8.0 (OriginLab, Northampton, MA, USA).
Cell proliferation and apoptosis assay
Cell proliferation assay was performed with BrdU kit (Roche, Basel, Switzerland). Briefly, cells were plated on 96-well plate at a density of 3000/well and cultured for 24 hrs. After 4 hrs of serum starvation, cells were incubated for 24 hrs with medium containing ion channel blockers, NaHS or growth factors. BrdU labelling solution (100 μM) diluted 10 times in DMEM (0.1% FBS) was added to each well and the plates were incubated at 37°C for an additional 2 hrs. Incorporated BrdU was detected by an anti-BrdU antibody for 90 min. and colorimetric development proceeded for 15 min. before analysis by ELISA plate reader (SpectraMax, Molecular Device, Sunnyvale, CA, USA). Cellular apoptosis assay was performed with Caspase-3 Fluorescence Assay kit as instructed (Cayman Chemical, Ann Arbor, MI, USA). Briefly, cells were plated on 96-well plate at a density of 104/well and cultured for 24 hrs. After 4 hrs of serum starvation, cells were incubated for 24 hrs with medium containing NaHS. Fluorescent intensity was obtained with ELISA plate reader (SpectraMax, Molecular Device) at 485 nm excitation and 535 nm emission wavelengths.
RNA isolation and RT-PCR
Total RNA was extracted from human atrial fibroblasts with Trizol reagent (Life Technologies, Carlsbad, CA, USA) after 12 hrs of treatment. RT-PCR was performed with one-step kit (Invitrogen) where 1 μg RNA and random hexamer primer were used for the initiation of cDNA synthesis. Gene-specific primers for the BKCa (KCa1.1): forward 5′- GGAGGATGCCTCGAATATCA-3′; reverse 5′-AGCTCGGGATGTTTAGCAGA-3′; Ito (Kv4.3): forward 5′-CTGGACAAGAACCAGCGACAGTGCG-3′; reverse 5′-ATCACGATCAGGAGGGCCACATAGGG-3′ and IKir (Kir2.1): forward, 5′-TTGAGACCCAGACAACCATAGGCTATGG-3′; reverse 5′-TGGCCATGACTGCGCCAATGATG-3′; α-SMA: forward 5′-CATCACCAACTGGGACGACA-3′; reverse 5′-GTGGGTGACACCATCTCCAG- 3′; CSE: forward 5′-TCCGGATGGAGAAACACTTC-3′; reverse 5′-GCTGCCTTTAAAGCTTGACC-3′; KATP (Kir6.2): forward 5′-GACCCTCATCTTCAGCAAGC-3′; reverse 5′-GGTGTTGCCAAACTTGGAGT-3′; β-actin: forward 5′-TTTGAGACCTTCAACACCCC-3′; reverse 5′-TTTCGTGGATGCCACAGGA-3′. PCR products were fractionated on 2% agarose gel electrophoresis. Data were expressed as values of optical density (OD) standardized to those of β-actin.
Atrial fibroblasts cultured on LabTek chamber slides (Nunc; Thermo Fisher Scientific, Waltham, MA, USA) were fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton-X100 and blocked with 2% BSA. Cells were incubated overnight with antibodies against α-smooth muscle actin (1/2000; α-SMA; Sigma-Aldrich) to identify myofibroblasts, against anti-Kv1.1 (1/1000; Abcam, Cambridge, UK), anti-Kv4.3 (1/500; Abcam) and anti-Kir2.1 (1/1000; Abcam) to identify BKCa, Ito and IKir channels, respectively (Fig. S2), before incubating with Alexa Fluor 488 or 555 secondary antibody (Life Technologies) and mounted in Vectashield mounting media containing DAPI for nuclear counterstain.
Data were expressed as mean ± SE. Statistical significance of the difference between groups was determined with Student's t-test. A value of P < 0.05 was considered statistically significant.
Hydrogen sulphide suppresses ion currents in human atrial fibroblasts
Multiple ionic channels are reported to be expressed in human cardiac ventricular fibroblasts , ionic channels in our atrial fibroblasts were activated by depolarization voltage between −70 and +60 mV from a holding potential of −80 mV (0.2 Hz) to elicit total outward K+ currents. Activated currents that were sensitive to paxilline (1 mM), a specific BKCa inhibitor, were significantly suppressed at +60 mV, confirming the presence of BKCa current (52%; 163/309 cells) in human atrial fibroblasts (Fig. 1A). Under identical voltage-clamp condition, exposure to 100 μM NaHS (as a donor of H2S) similarly reduced the peak current density of BKCa (Fig. 1B). The inhibitory effects observed could not be washed out (Fig. 1C). The presence of NaHS resulted in a voltage-dependent suppression of the I–V curve from 10.5 ± 1.2 pA/pF to 6.8 ± 0.9 pA/pF at +40 mV (P < 0.01; n = 6) (Fig. 1D) and a dose-dependent inhibition of BKCa peak current density with an IC50 of 69.4 μM (Fig. 1E).
To verify the specificity of H2S inhibition on BKCa, we assessed its effect in the presence of naringenin (10 μM), a specific opener of BKCa . BKCa currents were elicited with clamp pulses at +40 mV from a holding potential of −80 mV under control condition (Fig. 1F). Compared with baseline (9.9 ± 0.8 pA/pF), naringenin increased BKCa current significantly (14.1 ± 0.5 pA/pF; P < 0.01; n = 6), but addition of NaHS returned naringenin-induced current to baseline (9.2 ± 0.4 pA/pF; P < 0.05; n = 6) (Fig. 1G). The rising phase of the BKCa currents at 50 mV with activation τ (τact) at baseline (15.4 ± 0.1 ms) was lowered significantly by naringenin (8.6 ± 0.2 ms; P < 0.01; n = 6), but reversed to baseline after addition of NaHS (15.4 ± 0.2 ms; P < 0.05; n = 6), which confirmed its modulation of BKCa channel kinetics (Fig. 1H).
Similarly, under conditions to elicit total outward K+ currents, a 4-aminopyridine (4-AP; 0.5 mM)–sensitive current was detected, indicating the presence of transient outward currents, Ito (34%; 104/309 cells) in the atrial fibroblasts (Fig. 2A). Under identical voltage-clamp condition, exposure of fibroblasts to 100 μM NaHS reduced the peak current density of Ito (Fig. 2B). The inhibitory effects occurred within 1 min., reached saturation by 10 min. and could not be washed out (Fig. 2C). Addition of NaHS showed a voltage-dependent suppression of the Ito current in the I–V curve from 18.2 ± 1.5 pA/pF to 12.7 ± 1.7 pA/pF at +40 mV (P < 0.05; n = 6) (Fig. 2D) and demonstrated a dose-dependent inhibition of peak current density with an IC50 of 55.1 μM (Fig. 2E).
Steady-state activation of Ito was unaffected by NaHS (Fig. 2F). [The curves were fitted by the Boltzman equation: G/Gmax=1/[1 + exp(VT − V1/2/κ)], where G/Gmax represents a ratio of conductance to the maximum conductance, and VT represents the values of the depolarizing pulses]. The half-maximum activation voltage (V1/2) and slope factor under control condition were 17.2 ±1.5 mV and 19.3 ± 1.3, respectively, which were not significantly different from those in the presence of NaHS (V1/2: 18.3 ± 1.2 mV, slope factor 20.2 ± 1.2) (P = NS; n = 6). In contrast, NaHS significantly influenced the steady-state inactivation of Ito (Fig. 2G). When fitted to Boltzman function, I/Imax=1/[1 + exp(VT − V1/2/κ)], the half-maximum inactivation voltage (V1/2-inact) and slope factor under control condition were −53.6 ± 1.2 mV and 9.08 ± 1.1, respectively, which were significantly different from those in the presence of NaHS (V1/2-inact: −71.1 ± 3.1 mV, slope factor 14.7 ± 2.4) (P < 0.05; n = 6). Furthermore, recovery of Ito from inactivation was analysed by delivering two identical 500 ms depolarizing pulses from −80 to +60 mV and varying the interpulse from 50 to 3500 ms. Addition of NaHS shifted the curve right and increased the half-recovery time of Ito from of 461.7 ± 57 to 1218.2 ± 49 ms. (P < 0.01; n = 6) (Fig. 2H), confirming inhibition of NaHS on the kinetic property of Ito channel recovery. Furthermore, these properties of Ito were similar to those reported in human ventricular fibroblasts .
The inhibitory effect of NaHS on Ito was further confirmed in the presence of NS5806 (10 μM), a specific opener of Ito  (Fig. 2I). The Ito currents were elicited with clamp pulses at +40 mV from a holding potential of −80 mV. Compared with baseline (18.8 ± 0.85 pA/pF), peak current density significantly increased (24.9 ± 1.5 pA/pF; P < 0.05; n = 6) after the addition of NS5806, but additional presence of NaHS (100 μM) returned the NS5806-stimulated currents to baseline levels (18.6 ± 0.6 pA/pF; P < 0.01; n = 6) (Fig. 2J). After exposure to NS5806 (10 μM), inactivation of Ito was significantly subdued, as reflected by an expansion in time constant (τ, from 8.6 ± 0.2 to 13.6 ± 0.7 ms at +30 mV, P < 0.05; n = 6). However, addition of 100 μM NaHS returned the time constant to 10.1 ± 0.9 ms at +30 mV in the presence of 10 μM NS5806 (Fig. 2K), confirming inhibition of H2S on Ito current.
Besides BKCa and Ito currents, an inward rectifier current activated by hyperpolarization voltage steps on a holding potential of −40 mV that was sensitive to Ba2+ (0.5 mM) was found, indicating the presence of IKir inward current (28%; 28/309 cells) in the atrial fibroblasts (Fig. 3A). Exposure of atrial fibroblasts to 100 μM NaHS reduced the peak current density of IKir (Fig. 3B). The inhibitory effects occurred within 1 min., reached saturation at 10 min. and could not be washed out (Fig. 3C). NaHS showed a voltage-dependent suppression of the IKir current on the I–V curve from −4.4 ± 0.1 pA/pF to −3.0 ± 0.1 pA/pF at −110 mV (P < 0.05; n = 6) (Fig. 3D) and a dose-dependent inhibition of peak current density with an IC50 of 78.9 μM (Fig. 3E).
A minority of the atrial fibroblasts (1%; 1/54 cells) were found to exhibit inward currents with 50 ms voltage steps between −60 and +70 mV from −80 mV holding potential in 10 mV increments that resembled sodium current, indicating that K+ currents represent the major ionic species in human atrial fibroblasts.
H2S inhibits proliferation of atrial fibroblasts via suppression of Ito currents and gene expression
Inhibition of BKCa channel by paxilline, but not Na channel, has been reported to suppress proliferation of ventricular fibroblasts previously . We investigated whether inhibition of the major K+ currents of BKCa and Ito by H2S similarly affected atrial fibroblast proliferation. Cell proliferation was found to be dose-dependently suppressed by paxilline (BKCa inhibitor), 4-AP (Ito inhibitor) and Ba2+ (IKir inhibitor) (Fig. 4). Similarly, NaHS at 100, 300, 500 μM reduced cell proliferation by 33.1 ± 4.2%, 43.7 ± 3.1%, 58.4 ± 6.2%, respectively (*P < 0.05; **P < 0.01 versus vehicle control; n = 10) without significant apoptotic effect observed at 300 μM (Fig. 4B). While naringenin (100 μM) had no effect on cellular proliferation, NS5806 (100 μM) enhanced fibroblast proliferation by 9.1 ± 5.0% (P < 0.05; n = 10). However, NaHS (100 μM) reduced cellular proliferation by 29.1 ± 5.8% (P < 0.01; n = 10) and 23.1 ± 4.8% (P < 0.05; n = 10) in the presence of naringenin (100 μM) and NS5806 (100 μM), respectively, confirming additive inhibitory effects of H2S on BKCa and Ito currents in reducing cellular proliferation (Fig. 4C and D).
KATP channel has been reported to affect cellular proliferation . However, modulation of KATP channel (30%; 22/73 cells) (Fig. 5A and B) and Kir6.2 (responsible for KATP) gene expression (Fig. 5C) by H2S while confirming its role in enhancing current density, failed to show any appreciable effect on proliferation of our atrial fibroblasts. The KATP currents were elicited from voltage-clamped at the holding potential of −40 mV, voltage ramps were applied every 9 sec. from −120 mV to +60 mV at 20 mV/sec. and subsequently ramps to −40 mV at −100 mV/sec. Consistently, activation of the KATP channel by 30 μM pinacidil (specific channel enhancer) or its inhibition by 100 μM glibenclamide (specific channel inhibitor) did not significantly affect cellular proliferation despite the observed drastic modulation of current density (Fig. 5D and E). Down-regulation of fibroblast growth was observed only in the additional presence of NaHS with glibenclamide (19.9 ± 2.9% reduction versus control; P < 0.01; n = 4) or NaHS with pinacidil (22.5 ± 4.2% reduction versus control; P < 0.05; n = 4), suggesting that H2S inhibition of proliferation was independent of its modulating role of KATP channel in atrial fibroblasts.
Gene expression showed that H2S reduced the mRNA level of KCa1.1 (responsible for BKCa), Kv4.3 (responsible for Ito), Kir2.1 (responsible for IKir) in TGF-β1–stimulated fibroblasts by 42.3 ± 5.1% (P < 0.05; n = 4), 76.9 ± 3.5% (P < 0.01; n = 4), 90.8 ± 4.7% (P < 0.01; n = 4), respectively, at 12 hrs after addition of NaHS (Fig. 6A–D). Furthermore, pre-treatment with NaHS decreased mRNA level of Kv4.3 by 21.6 ± 2.2% (n = 4; P < 0.05 versus basal levels), but not that of KCa1.1 and Kir2.1. Furthermore, NaHS enhanced production of endogenous H2S by enhancing cystathionine γ–lyase (CSE) mRNA levels and maintaining its expression even in the presence of D,L-propargylgylcine (PPG), a potent inhibitor of CSE (Fig. 6E and F). These results indicated that H2S inhibited fibroblast proliferation by regulating Kv4.3 mRNA expression and inhibiting Ito current, possibly via an autocrine feedback mechanism.
H2S inhibits TGF-β1–induced differentiation of atrial fibroblasts to myofibroblasts
Transforming growth factor-β1 and Angiotensin II (Ang II) as the major mediators of fibroblast proliferation and their differentiation towards myofibroblasts in atrial fibrosis [20, 21] were consistently shown to promote proliferation of atrial fibroblasts in our study (Fig. 7A). Additional presence of NaHS decreased TGF-β1– (1 ng/ml), Ang II- (100 nM) and 20% FBS-induced fibroblast proliferation by 50.1 ± 4.3% (P < 0.01; n = 10), 42.1 ± 5.7% (P < 0.01; n = 10) and 21.2 ± 3.4% (P < 0.05; n = 10), respectively, which suggested H2S as a potent inhibitor of cytokine-mediated fibroblast proliferation. Furthermore, NaHS (100 μM) decreased TGF-β1-(1 ng/ml)–induced fibroblast transformation into myofibroblasts whereby mRNA expression of α-SMA, a hallmark of fibroblast differentiation, was significantly down-regulated (34.1 ± 7.1% reduction versus TGF-β1 alone; P < 0.05) (Fig. 7B), which was confirmed by reduced immunocytochemical α-SMA staining (percentage of α-SMA–positive cells, 47 ± 6% versus 90 ± 7%; P < 0.01; n = 4) (Fig. 7C and D). Nevertheless, no significant change in α-SMA–containing stress fibres was observed after NaHS treatment alone (percentage of α-SMA–positive cells, 33 ± 4%; n = 4) as compared with standard cultured atrial fibroblasts (32 ± 7%; n = 4) in 10% FBS.
Multiple potassium channels are known to express in cardiac ventricular fibroblasts  and inhibition of BKCa current resulted in suppression of fibroblast proliferation . Transient outward K+ current, Ito, is present in neonatal rat cardiac fibroblasts (encoded by Kv1.4)  and human ventricular fibroblasts (encoded by Kv4.3) . Similarly, Ba2+-sensitive inward rectifier K+ current (encoded by Kir2.1/Kir2.3) is present in human ventricular fibroblasts  and rat ventricular fibroblasts  whereby its modulation may have major significance in cardiac fibrosis. However, their roles in atrial fibroblasts which are more actively participating in cardiac fibrosis , are relatively not well understood.
We demonstrated that H2S dose-dependently inhibited BKCa, Ito and IKir in human atrial fibroblasts within minutes, suggesting an acute modulation of H2S on such channels. The inhibitory effect of H2S on BKCa, Ito and IKir was observed at 25–400 μM. The physiological levels of plasma H2S have been reported to be 50–160 μM in human brain  and 50–100 μM in human serum . As NaHS dissolved in saline, one-third of the H2S exists as an undissociated gas, and the remaining two-third as the HS− anion . Therefore, the physiologically relevant concentration of H2S (25–400 μM) used in this study, which effectively blocked BKCa, Ito and IKirin vitro, is likely to be attainable in vivo.
We found that NaHS attenuated naringenin-induced BKCa activation and decelerated the transition from closed to open state of the channel, suggesting a role for H2S in regulating BKCa channel kinetic and voltage sensitivity. However, NaHS had no effect on the half-maximum voltage activation, but shifted the steady-state inactivation curve to the left, indicating that the voltage-dependent steady-state inactivation kinetics of Ito channel were altered. Furthermore, NaHS markedly shifted the recovery curve of Ito to the right, indicating that H2S attenuated the recovery of Ito from inactivation. These results indicated that H2S inhibited Ito through facilitation of steady-state inactivation and attenuation of recovery from inactivation. In contrast to reported presence of sodium current in ventricular fibroblasts (61%) , we found relatively few cells (1%) with detectable sodium current. This is consistent with previous reported presence of fast sodium current only in atrial myofibroblasts, but not in undifferentiated fibroblasts  like those used in our study.
BKCa channels (encoded by KCa1.1) have been demonstrated to regulate proliferation of human cardiac ventricular fibroblasts  and endothelial cells . Furthermore, inhibition of IKir current suppressed proliferation of endothelial cells . Similarly, inhibition of BKCa (by paxilline), Ito (by 4-AP) and IKir (by Ba2+) currents resulted in a significant reduction in fibroblast proliferation in our study. Consistently, suppression of the K+ currents by NaHS inhibited atrial fibroblast proliferation in a dose-dependent manner. Furthermore, suppression of proliferation by NaHS in the presence of naringenin (channel opener of BKCa) or NS5806 (channel opener of Ito) suggested an additive inhibitory effect of H2S on BKCa and Ito channels in proliferation of atrial fibroblasts. Consistent with KATP channel–activating effect of H2S [11, 31], addition of NaHS recovered KATP channel activity from glibenclamide inhibition. Nevertheless, suppression of cellular proliferation by NaHS in the presence of glibenclamide (specific inhibitor of KATP) or pinacidil (specific enhancer of KATP) indicated that KATP channel was unlikely to be involved in proliferation of atrial fibroblasts. Consistently, H2S inhibition of lung fibroblast proliferation has been reported to be independent of KATP channel .
Consistent with electrophysiological findings on the presence of BKCa, Ito and IKir potassium currents, RT-PCR confirmed expression of KCa1.1, Kv4.3 and Kir2.1 in atrial fibroblasts. Furthermore, H2S decreased Kv4.3 expression and significantly moderated TGF-β1–mediated enhanced expression of Kv4.3 as well as KCa1.1 and Kir2.1. Effect of NaHS (exogenous donor of H2S) on expression of cystathionine γ-lyase (CSE) that produces endogenous H2S is controversial, with reports of no effect in human airway smooth muscle cells  to inhibitory effect in mouse aortic smooth muscle cells . However, in concordance with other reports [27, 35], our results showed that NaHS enhanced CSE expression and further sustained its expression in the presence of DL-PPG that strongly inhibited expression of CSE.
Myofibroblasts characterized by increased α-SMA expression are abundant in cardiac fibrosis  that has been associated with TGF-β-mediated  and Ang II-mediated  atrial fibrillation. Preventing myofibroblast differentiation from proliferating fibroblasts has been an attractive target in limiting cardiac fibrosis. Inhibition of TGF-β1 function by anti–TGF-β1 antibodies reduced myofibroblasts and lessened fibrosis . Hydrogen sulphide was found to inhibit TGF-β–induced transformation of MRC5 lung fibroblasts to myofibroblasts . Consistently, our results showed that NaHS effectively reduced proliferation of atrial fibroblasts in response to TGF-β1, Ang II or FBS. Furthermore, NaHS ameliorated transformation towards myofibroblasts whereby α-SMA expression and their stress fibres were significantly suppressed, although causal role of potassium channels in such transformation remained to be ascertained.
In summary, our study provides evidence of major K+ channels in human atrial fibroblasts that share similar heterogenous expression as in human ventricular fibroblasts . Hydrogen sulphide inhibits fibroblast proliferation probably through a combined modulation of BKCa, Ito, IKir, but not KATP, channels. Although roles of MAPK and ERK pathways in our atrial fibroblasts remain to be determined, they were implicated in H2S-mediated suppression of proliferation of vascular smooth muscle cells  and lung fibroblasts . Both kinase pathways were linked to cell cycle progression in lung fibroblasts , which, in turn were reportedly regulated by BkCa in human ventricular fibroblasts . However, KATP was found to play no significant role in ERK-inhibiting effect of H2S , which may explain our observation in this study. Consistent with the observed beneficial effects of H2S on cardiac fibrosis in vivo [12, 13], our results suggested that such effects may be partly mediated via selective inhibition of K+ channels in atrial fibroblasts and suppression of their transformation to myofibroblasts. Such regulating role of H2S in atrial fibroblasts may have clinical value in targeting atrial fibrillation, which invariably linked to atrial fibrosis.
This study was supported by funding from the National Research Foundation Singapore (NRF-003-CRP-002), the Goh Foundation (Duke-NUS-GCR/2013/008) and Biomedical Research Council Singapore (BMRC 13/1/96/19/686) to W.S.
Conflicts of interest
The authors confirm that there are no conflicts of interest.