Characterization of Multiple Ion Channels in Cultured Human Cardiac Fibroblasts

Background Although fibroblast-to-myocyte electrical coupling is experimentally suggested, electrophysiology of cardiac fibroblasts is not as well established as contractile cardiac myocytes. The present study was therefore designed to characterize ion channels in cultured human cardiac fibroblasts. Methods and Findings A whole-cell patch voltage clamp technique and RT-PCR were employed to determine ion channels expression and their molecular identities. We found that multiple ion channels were heterogeneously expressed in human cardiac fibroblasts. These include a big conductance Ca2+-activated K+ current (BKCa) in most (88%) human cardiac fibroblasts, a delayed rectifier K+ current (IKDR) and a transient outward K+ current (Ito) in a small population (15 and 14%, respectively) of cells, an inwardly-rectifying K+ current (IKir) in 24% of cells, and a chloride current (ICl) in 7% of cells under isotonic conditions. In addition, two types of voltage-gated Na+ currents (INa) with distinct properties were present in most (61%) human cardiac fibroblasts. One was a slowly inactivated current with a persistent component, sensitive to tetrodotoxin (TTX) inhibition (INa.TTX, IC50 = 7.8 nM), the other was a rapidly inactivated current, relatively resistant to TTX (INa.TTXR, IC50 = 1.8 µM). RT-PCR revealed the molecular identities (mRNAs) of these ion channels in human cardiac fibroblasts, including KCa.1.1 (responsible for BKCa), Kv1.5, Kv1.6 (responsible for IKDR), Kv4.2, Kv4.3 (responsible for Ito), Kir2.1, Kir2.3 (for IKir), Clnc3 (for ICl), NaV1.2, NaV1.3, NaV1.6, NaV1.7 (for INa.TTX), and NaV1.5 (for INa.TTXR). Conclusions These results provide the first information that multiple ion channels are present in cultured human cardiac fibroblasts, and suggest the potential contribution of these ion channels to fibroblast-myocytes electrical coupling.


Introduction
It is generally recognized that cardiac myocytes and fibroblasts form extensive networks in the heart, with numerous anatomical contacts between cells [1]. Cardiac fibroblasts play a central role in the maintenance of extra-cellular matrix in the normal heart and act as mediators of inflammatory and fibrotic myocardial remodeling in the injured heart, e.g. ischemic, hypertensive, hypertrophic, and dilated cardiomyopathies, and heart failure [2,3]. The cardiac myocyte network, coupled with gap junctions, is generally believed to be electrically isolated from fibroblasts in vivo. However, in the co-culture of cardiac myocytes and fibroblasts, the heterogeneous cell types form functional gap junctions, providing a substrate for electrical coupling of distant myocytes, interconnected by fibroblasts. In addition to the evidence of fibroblast-to-myocyte electrical coupling in the rabbit SA node [4], fibroblasts have been shown to be coupled electrotonically with myocytes in vitro [1,[5][6][7][8]. Moreover, there is increasing evidence that implicates potential heterocellular electrical coupling in the diseased myocardium with arrhythmogenesis [9,10]; therefore, the cardiac fibroblasts are considered to be potential targets in managing cardiac disorders including hypertrophy, heart failure and arrhythmias [2,3,9,10].
Ion channels and their functions are well studied in cardiomyocytes; however, the ion channel expression and their physiological roles are not fully understood in cardiac fibroblasts. An inward rectifier K + current (I Kir ), a delayed rectifier K + current (IK DR ), and a nonselective cation channel current were recently reported in rat ventricular fibroblasts [11][12][13]. Although a Ca 2+ -activated big conductance K + current (BK Ca ) was described in human cardiac fibroblasts [14], it is unknown whether other types of ion channel currents are present in human cardiac fibroblasts. The present study was designed to employ the approaches of whole-cell patch voltage clamp and RT-PCR to examine the functional ion channels in human cardiac fibroblasts. Using these techniques, we identified multiple ion channels expressed in cultured human cardiac fibroblasts.

Cell cultures
Human cardiac fibroblasts (adult ventrical, Catalog# 6310) were purchased from ScienCell Research Laboratory (San Diego, CA). The cells were cultured as monolayers in completed DMEM containing 10% fetal bovine serum (Invitrogen, Hong Kong) and antibiotics (100 U/ml penicillin G and 100 mg/ml streptomycin) at 37uC in a humidified atmosphere of 95% air, 5% CO 2 . No difference in cell growth and ion channel expression were observed with either our culture medium or the medium from ScienCell Research Laboratory. Cells used in this study were from the early passages 2 to 6 to limit the possible variations in functional ion channel currents and gene expression. The cells were harvested for electrophysiological recording and RT-PCR determination via trypsinization [15].

Electrophysiology
A small aliquot of the solution containing the cardiac fibroblasts was placed in an open perfusion chamber (1 ml) mounted on the stage of an inverted microscope. The cells were allowed to adhere to the bottom of the chamber for 10-20 min, and then superfused at 2-3 ml/min with Tyrode solution. The studies were conducted at room temperature (22-24uC).
The membrane ionic currents were recorded with a whole-cell patch-clamp technique as described previously [16]. Borosilicate glass electrodes (1.2 mm OD) were pulled with a Brown-Flaming puller (Model P-97, Sutter Instrument Co. Novato, CA), and had tip resistances of 2,3 MV when filled with pipette solution. The tip potentials were compensated before the pipette touched the cell. 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 MV. The cell membrane capacitance (49.6612.1 pF) was electrically compensated with the Pulse software. The series resistance (R s , 3-5 MV) was compensated by 50-70% to minimize voltage errors. Membrane currents were elicited with voltage protocols as described in the following Results section for individual different current recording. Data were acquired with an EPC10 amplifier (Heka, Lambrecht, Germany). The membrane currents were lowpass filtered at 5 kHz and stored on the hard disk of an IBM compatible computer.

Messenger RNA determination
The messenger RNA was examined using RT-PCR technique using Table 1 primers as described previously [15]. Total RNA was extracted from human cardiac fibroblasts using Trizol reagent (Invitrogen), and further treated with DNase I (GE Healthcare, Hong Kong) for 30 min at 37uC, then heated to 75uC for 5 min and finally cooled to 4uC [17]. Reverse transcription was performed using a RT system (Promega, Madison, WI) in a 20 ml reaction mixture. A total of 2 mg RNA was used in the reaction and a random hexamer primer was used for the initiation of cDNA synthesis. After the RT procedure, the reaction mixture (cDNA) was used for PCR.
PCR was performed with thermal cycling conditions of 94uC for 2 min followed by 35 cycles at 94uC for 45 s, 55-58uC for 45 s, and 72uC for 1 min using a Promega PCR kit and oligonucleotide primers as shown in Table 1. This was followed by a final extension at 72uC (10 min) to ensure complete product extension. The PCR products were electrophoresed through 1.5% agarose gels and visualized under a UV transilluminator (BioRad, Hercules, CA) after staining with ethidium bromide.

Statistical analysis
Results are presented as means 6 SEM. Paired and/or unpaired Student's t-tests were used as appropriate to evaluate the statistical significance of differences between two group means, and analysis of variance was used for multiple groups. Values of P,0.05 were considered to indicate statistical significance. Figure 1 illustrates the families of membrane currents recorded in human cardiac fibroblasts using a standard pipette solution. Five types of membrane currents were observed in human cardiac fibroblasts (in a total of 265 cells). One current was activated at depolarization voltages between 270 and +60 from a holding potential of 280 mV (0.2 Hz), and showed an outward current with noisy oscillation between +20 and +60 mV (Fig. 1A). These features suggest that this current is likely a big conductance Ca 2+activated K + current (BK Ca ) [15]. The noisy oscillatory BK Ca was present with other currents in most (88%, 233 of 265) of fibroblasts. Another current activated by the same protocol was a transient outward current (Fig. 1B), and presented in 15% (40 of 265) of cells. Third current was an inward component activated by hyperpolarization voltage steps a holding potential of 240 mV and co-existed with the noisy oscillatory current activated by depolarization voltage steps. This inward component exhibited the properties similar to inward rectifier K + current (I Kir ) (Fig. 1C). I Kir was observed in 24% (64 of 265) of cells. Fourth current was elicited by voltage steps between 2120 and +60 from a holding potential of 240 mV, showing a very small inward component and a large outward current with outward rectification (Fig. 1D). This current was observed in 7% (19 of 265) of cells. Moreover, an inward current coexists with the oscillatory current in 61% (167 of 265) of human cardiac fibroblasts ( Fig. 1E and 1F). Interestingly, the inward current exhibits either a fast inactivation (Fig. 1E) or a slow (Fig. 1F) inactivation.

Families of membrane ionic currents in human cardiac fibroblasts
Ca 2+ -activated noisy oscillatory current Figure 2A displays the noisy oscillatory BK Ca reversibly suppressed by the BK Ca blocker paxilline (1 mM, 5 min exposure) in a representative fibroblast. Current-voltage (I-V) curves recorded with a 2-s voltage ramp (280 to +80 mV from a holding potential of 240 mV) in the absence of paxilline showed outward rectification (control) in another cell. The outwardly-rectifying current was remarkably reduced by paxilline (Fig. 2B). The current at +60 mV was reduced from 29.865.3 pA/pF of control to 3.962.1 pA/pF with 1 mM paxilline (n = 35, P,0.01 vs control).

Transient outward K + current
The transient outward K + current I to was present in 15% of cardiac fibroblasts. I to in human cardiac myocytes was sensitive to inhibition by 4-AP [18], therefore we determined whether I to in human fibroblasts could be decreased by 4-AP. Figure 3A shows the I to traces recorded in a typical experiment in the absence and presence of 5 mM 4-AP. I to (+60 mV) was substantially inhibited by 4-AP to 11.961.4 pA/pF from 36.562.6 pA/pF (n = 7, P,0.01). Figure 3B illustrates the mean values of voltage-dependent activation (g/g max ) and inactivation (availability, I/I max ) of I to . The g/g max was determined from the I-V relationship of each cell as previously described [19]. The I/I max was determined with the protocol as shown in the left inset (with 1-s conditioning pulses from voltages between 2100 and 210 mV followed by a 300-ms test pulse to +60 mV). Data were fitted to a Boltzmann distribution to obtain the half activation or availability voltage (V 0.5 ) and the slope factor (S). The V 0.5 s of activation and availability of I to were 11.260.4 mV (n = 7) and 240.661.5 mV (n = 9), and the S was 11.161.1 and 28.461.3, respectively. Fig. 3C shows the time course of the mean values of I to recovery from inactivation, determined with a paired-pulse protocol as shown in the inset. I to recovery was complete within 900 ms and fitted to a mono-exponential function with time constant (t) of 257.465.9 ms (n = 7). These properties of I to in human cardiac fibroblasts, i.e. 4-AP sensitivity, voltage-dependent activation and availability, and recovery from inactivation, are similar to those observed in human cardiac myocytes [19] and mesenchymal stem cells [15], though there are differences in the values of the recovery time constant and the V 0.5 s of voltage-dependent activation and availability.

Inward rectifier K + current
It is generally believed that inwardly-rectifying K + channels are sensitive to inhibition by Ba 2+ [20], therefore we determined the effect of Ba 2+ on I Kir in human cardiac fibroblasts. Figure 4 shows the current traces recorded in a representative cell with the voltage protocol as shown in the inset in the absence (control) and presence of Ba 2+ . Ba 2+ (0.5 mM) reversibly reduced I Kir . Figure 4B displays that the increase of external K + (K + o , from 5 to 20 mM) enhanced I Kir conductance. Basal I Kir and the high K + o -induced current were suppressed by Ba 2+ . Figure 4C illustrates the I-V relationships of I Kir recorded in a representative cell with a 2-s ramp protocol (2120 to 0 mV from 240 mV) in solution containing 5 mM K + (control) or 20 mM K + , and after application of 0.5 mM Ba 2+ in bath solution. Ba 2+ strongly inhibited I Kir . Ba 2+ -sensitive current was obtained by digitally subtracting currents before and after application of Ba 2+ (Fig. 4D). The I-V relationships of Ba 2+sensitive I Kir in 5 and 20 mM K + o exhibited a strong inward rectification, typical of an inwardly-rectifying K + current. Similar results were obtained in 5 other cells.

Volume-sensitive chloride current in human cardiac fibroblasts
The current with outward rectification shown in Fig. 1D was insensitive to inhibition of K + channel blockers including 5 mM tetraethylammonium (TEA), 5 mM 4-AP, or 0.5 mM Ba 2+ (n = 426), suggesting that the outwardly-rectifying current is not carried by K + ion. We then employed the Cl 2 channel inhibitor DIDS to determine whether the current is carried by chloride ions. Figure 5A shows the current traces recorded in a representative cell with the protocol shown in the inset; DIDS (150 mM) suppressed the current. The I-V relationship (Fig. 5B) of the DIDS-sensitive current obtained by subtracting control currents by the current recorded after DIDS application displayed outward rectification and had a reversal potential at 235 mV, which is close to Cl 2 equilibrium potential (E Cl , 246.8 mV). Similar results were obtained in a total of 6 cells. This result suggests that the recorded current under isotonic conditions is carried by Cl 2 ions.
To investigate whether the Cl 2 channel is volume sensitive in human cardiac fibroblasts, we employed a 0.7T tonic solution and recorded membrane current using a K + -free pipette solution, symmetrical Cl 2 ion in pipette and bath medium as described in the Methods section. The membrane conductance was remarkably enhanced by exposure to 0.7T (20 min), and the increased current was highly suppressed by the Cl 2 channel blocker NPPB (Fig. 5C). The I-V relationship of 0.7T-induced Cl 2 current is linear under symmetrical Cl 2 conditions Fig. 5D), similar to the previous report [21]. These results indicate that volume-sensitive Cl 2 channel (I Cl.vol ) is present in human cardiac fibroblasts.

Inward Na + currents in human cardiac fibroblasts
The depolarization-elicited inward currents (Fig. 1E and 1F) were studied under K + -free conditions. Figure 6 illustrates two types of  inward currents recorded in human cardiac fibroblasts with voltage steps (50 ms) to between 260 and +70 mV from 280 mV (inset) in 10-mV increments at 0.2 Hz. One of these currents exhibited an incomplete inactivation (or a persistent component) during 50 ms depolarization (control of Fig. 6A, 6B), similar to L-type Ca 2+ current (I Ca.L ) in human cardiac myocytes [22]. However, this current was insensitive to inhibition by a high concentration of the I Ca.L blocker nifedipine (10 mM), in contrast with human cardiac I Ca.L , which is fully suppressed by nifedipine [22]. Interestingly, the current was abolished by replacing bath Na + (Na + o ) with equimolar choline, and recovered upon restoration of Na + o (Fig. 6A, n = 6). In addition, this current is sensitive to inhibition by 10 and 100 nM tetrodotoxin (TTX), and the effect was reversed by washout (n = 6). These results suggest that this inward current is likely a TTXsensitive I Na (I Na.TTX ) with a persistent component.
Another inward current exhibited a complete inactivation (control of Fig. 6C & 6D). This current had no response to 10 nM TTX; however, nifedipine (10 mM) reversibly reduced the current (Fig. 6C, n = 7). Replacement of Na + o with equimolar choline reversibly abolished this inward current, and the current required a high concentration (10 mM) of TTX for a substantial suppression (Fig. 6D, n = 6). These results suggest that this inward current is likely a TTX-resistant Na + current (I Na.TTXR ).
The concentration-dependent inhibitory effects of TTX on I Na.TTX and I Na.TTXR at 0 mV are illustrated in Fig. 6E. The IC 50 (50% inhibitory concentration) of TTX for inhibiting I Na.TTX was 7.8 nM with a coefficient of 0.94, while the IC 50 of TTX for inhibiting I Na.TTXR was 1.8 mM with a Hill coefficient of 0.58. The I Ca.L blocker nifedipine had no significant inhibitory effect on I Na.TTX , whereas it inhibited I Na.TTXR with an IC 50 of 56.2 mM and a Hill coefficient of 0.59 (Fig. 6F).
The I-V relationships for the peak current of I Na.TTX and I Na.TTXR are illustrated in Fig. 7A. I Na.TTX had a threshold potential of 240 mV and peaked at +10 mV, while I Na.TTXR had a threshold potential of 250 mV and peaked at 0 mV. Inactivation of I Na.TTX and I Na.TTXR was fitted to a monoexponential function with time constant (t) as shown in the left panel of Fig. 7B. The inactivation process of I Na.TTX was slower than that of I Na.TTXR (Fig. 7B, n = 12, P,0.01 at 220 to +60 mV). Normalized mean values of voltage-dependent availability (I/I max ) and activation conductance (g/ g max ) of I to were fitted to the Boltzmann function: y = 1/ {1+exp[(V m 2V 0.5 )/S]}, where V m is membrane potential, V 0.5 is the estimated midpoint, and S is the slope factor. C. Normalized I to (I 2 /I 1 ) plotted vs. P 1 2P 2 interval. The recovery curve was fitted to a monoexponential function. The I to was measured from the current peak to the 'quasi'-steady-state level. doi:10.1371/journal.pone.0007307.g003  Figure 7C illustrates the mean values of the steady-state voltage dependent activation (g/g max ) and inactivation (availability, I/I max ) for both I Na.TTX and I Na.TTXR . The g/g max was determined from the I-V relationships of each cell in Fig. 7A as previously described [23]. The I/ I max was determined with the protocol as shown in the left inset (with 1-s conditioning pulses from voltages between 2120 and 210 mV then to a 50-ms test pulse to 0 mV). Data were fitted to a Boltzmann equation. The V 0.5 s of g/g max and I/I max for I Na.TTX were 27.261.1 mV (n = 9) and 261.461.6 mV (n = 10), and the S was 8.761.2 and 210.861.2, respectively. While the V 0.5 s of g/g max and I/I max for I Na.TTXR were 224.761.4 (n = 7) and 272.361.5 (n = 9) mV, and the S was 7.561.1 and 28.561.3, respectively. The V 0.5 s of g/g max and I/I max were more positive in I Na.TTX than those in I Na.TTXR (P,0.01). Figure 7E shows the time course of mean values of recovery of I Na.TTX or I Na.TTXR from inactivation, which was determined using a paired-pulse protocol shown in the inset as described previously [23]. The recovery of I Na.TTX and I Na.TTXR from inactivation was complete within 150 ms, and the curves were fitted to a mono-exponential function. The time constant (t) was 14.362.1 ms for I Na.TTX (n = 11) and 21.462.9 ms for I Na.TTXR (n = 9). The recovery of I Na.TTXR from inactivation was slower than that of I Na.TTX (P,0.05). These results indicate that two types of Na + channels with distinct TTX-sensitivity and kinetics are present in human cardiac fibroblasts.

Messenger RNAs of functional ion channels
To explore the molecular identities of the functional ionic currents, we examined gene expression of various ionic channels in human cardiac fibroblasts with RT-PCR using the specific primers targeting human genes for KCa, Kv, Kir, Clcn, and Na V channel families as shown in Table 1. Figure 8A displays the significant gene expression of KCa1.1 (responsible for BK Ca ), Kv1.5, Kv1.6 (responsible for IK DR ), Kv4.2, Kv4.3 (responsible for I to ), Kir2.1, Kir2.3 (for I Kir ), Clcn3 (for I Cl.vol ), Na V 1.2, Na V 1.3, Na V 1.6 and Na V 1.7 (for I Na.TTX ), and Na V 1.5 (for I Na.TTXR ) in human cardiac fibroblasts. In addition, Clcn2 was also significantly expressed in human cardiac fibroblasts. When RNA was directly amplified by PCR without reverse transcription, the bands for these positive genes disappeared (Fig. 8B), suggesting that the genes detected were not false-positive signals from genomic DNA contamination.

Discussion
In the present study, we have demonstrated that multiple ionic currents (BK Ca , IK DR , I to , I Kir , I Cl.vol , and I Na.TTX and I Na.TTXR ) are present in human cardiac fibroblasts. BK Ca was inhibited by paxilline, IK DR and I to were inhibited by 4-AP. I Kir was blocked by Ba 2+ and I Cl.vol was inhibited by DIDS or NPPB, while I Na.TTX and I Na.TTXR were suppressed by different concentrations of TTX. The channel genes corresponding to the functional currents (KCa1.1 for BK Ca , Kv1.5/Kv1.6 for IK DR , Kv4.2/Kv4.3 for I to , Kir2.1/Kir2.3 for I Kir , Clcn3 for I Cl.vol , Na V 1.2/Na V 1.3/Na V 1.6/ Na V 1.7 for I Na.TTX , and Na V 1.5 for I Na.TTXR ) were confirmed by RT-PCR.
Recent studies demonstrated that an inward rectifier K + current (I Kir ), a delayed rectifier K + current (IK DR ), and a non-selective cation channel current were present in rat ventricular fibroblasts [11][12][13]. Only BK Ca was described in human cardiac fibroblasts [14]. The present study provides novel information that multiple ion channels are heterogeneously expressed in human cardiac fibroblasts. In addition to BK Ca as previously reported by Wang and colleagues [14], IK DR , I to , I Kir , I Cl.vol , I Na.TTX , and I Na.TTXR were present with BK Ca in different populations of human cardiac fibroblasts (Fig. 1). These currents have different distribution and properties compared to those in human cardiomyocytes [18,[24][25][26][27].
Earlier studies have demonstrated that I Cl.vol are present in human cardiac myocytes [31,32], and the current is only recorded when the hypotonic insult is applied [31,32]. Nonetheless, I Cl.vol is recorded in a small population (7%) of human cardiac fibroblasts without hypotonic exposure (Fig. 1), and it is activated in almost all fibroblasts with hypotonic exposure (Fig. 5). I Cl.vol is believed to play a role in arrhythmogenesis, myocardial injury, preconditioning, and apoptosis of myocytes [33]. Nonetheless, physiological function of I Cl.vol in human cardiac fibroblasts remains to be studied in the future.
It is well recognized that I Na channels expressed in cardiomyocytes (mainly encoded by Na V 1.5) play an important role in controlling excitation-contraction and impulse conduction in the hearts. I Na has been also found to participate in regulating sinus node pacemaker function [34]. In the present study, we found that I Na was expressed in most (61%) human cardiac fibroblasts (Fig. 1). The I Na.TTX in human cardiac fibroblasts (Figs. 6 & 7) shares some properties with neuronal A. An inward current with a persistent component (arrow) recorded in a representative cell under K + -free conditions using the voltage steps as shown in the inset. Nifedipine (10 mM) had no effect on the current, while the current disappeared when Na + o was replaced with equimolar choline, and recovered as restoration of Na + o . B. Similar inward current with persistent component (arrow) recorded in another cell was highly sensitive to inhibition by low concentrations of TTX. C. An inward current with fast inactivation recorded using the same voltage protocol as shown in the inset of A. The current was not affected by 10 nM TTX, but reversibly inhibited by 10 mM nifedipine. D. Similar current recorded in another cell disappeared with Na + o removal, and recovered as restoration of Na + o . The current was suppressed by a high concentration of TTX (10 mM). E. Concentration-dependent response of two types of inward currents to TTX. The data were fitted to the Hill equation: E = E max /[1+(IC 50 /C) b ], where E is the percentage inhibition of current at concentration C, E max is the maximum inhibition, IC 50 is the concentration for a half inhibitory effect, and b is the Hill coefficient. The IC 50 of TTX for inhibiting TTX-sensitive I Na was 7.8 nM (n = 529 for each concentration), the Hill coefficient was 0.94. The IC 50 of TTX for inhibiting TTX-resistant I Na was 1.8 mM (n = 629 cell for each concentration), the Hill coefficient was 0.58. F. Concentration-dependent relationships of I Na.TTX and I Na.TTXR to nifedipine. The IC 50 of nifedipine for inhibiting I Na.TTXR was 56.2 mM (n = 427 cells for each concentration) with a Hill coefficient of 0.59. doi:10.1371/journal.pone.0007307.g006 I Na , e.g. a transient inward current followed by a persistent component, sensitive to inhibition by nanomolar TTX, and likely encoded by Na V 1.2, Na V 1.3, Na V 1.6, and Na V 1.7 [35,36].
The I Na.TTXR in human cardiac fibroblasts (Figs. 6 & 7) shares some features with I Na in cardiomyocytes (e.g. inhibited by micromolar TTX and encoded by Na V 1.5) [35,37]. Some properties of I Na.TTXR in human cardiac fibroblasts are not identical to those of I Na in human cardiomyocytes [27,37,38], e.g. more positive V 0.5 s of activation (225 mV vs 239 mV) and availability (272 mV vs 295 mV) and more positive peak current potential (0 mV vs 235 mV), compared to I Na in human cardiomyocytes [27,37]. In addition, I Na.TTXR in cardiac fibroblasts, as Na V 1.5-encoded I Na.TTXR in gastric epithelial cells [39], was inhibited by high concentrations of the I Ca.L blocker nifedipine (Fig. 6). Nonetheless, no report is available in the literature regarding the information whether I Na of cardiomyocytes is sensitive to a high concentration of nifedipine. Moreover, it is unknown how I Na.TTX and I Na.TTXR participate in cellular function of cardiac fibroblasts.
It has been recognized that cardiac fibroblasts are electrically unexcitable, but they contribute to the electrophysiology of myocytes in various ways, such as electrical coupling of fibroblasts and myocytes [40]. The electrical coupling between fibroblasts and myocytes was observed at cellular and tissue level as well as in cell cultures [5,8,[40][41][42]. Coupling between fibroblasts and myocytes was demonstrated to be via Cx43 gap junctions in sheep ventricles and Cx45 in rabbit sinoatrial node cells [4,6] and in sheep ventricular scars [43]. The cardiac fibroblasts are therefore believed to maintain electrical contact with myocytes. Our results of multiple ion channels in human cardiac fibroblasts likely provide a basis for understanding of the potential contribution of these ion channels to fibroblast-myocytes electrical coupling under physiological conditions, and also for future studies on the potential mechanism how cardiac fibroblasts participate in regulating cardiac electrophysiology.
In proliferative cells, ion channels play a role in cell cycle progression [44,45]. The activity of BK Ca (i.e. KCa1.1) channels was regulated by the spontaneous Ca 2+ oscillations, resulting in Right panel: mean values of voltage dependence of inactivation of I Na.TTX (n = 8) and I Na.TTXR (n = 10). P,0.05 or P,0.01 at 220 to +60 mV. C. Voltagedependent availability (I/I max ) of I Na was determined with the protocol as shown in the left inset (with 1-s conditioning pulses from voltages between 2120 and 210 mV then a 50-ms test pulse to 0 mV). Curves of I/I max and activation conductance (g/g max ) were fitted to a Boltzmann equation. E. Recovery curves of I Na.TTX and I Na.TTXR from inactivation were fitted to a monoexponential function. doi:10.1371/journal.pone.0007307.g007 Figure 8. RT-PCR for detecting ion channels expressed in human cardiac fibroblasts. A. Images of RT-PCR products corresponding to significant gene expression of KCa1.1 (BK Ca ), Kv1.5 (IK DR ), Kv4.3 (I to ), and Kir2.1 (I Kir ) and Clcn3 (I Cl.vol ), and Na V 1.2, Na V 1.3, Na V 1.5, Na V 1.6 and Na V 1.7 in human cardiac fibroblasts. A weak expression of Kv4.2, Kir2.3, Clcn2 and Na V 1.1 was also found in human cardiac fibroblasts. B. No significant bands were observed in the PCR experiment when RT product was replaced by total RNA. doi:10.1371/journal.pone.0007307.g008 fluctuations of membrane currents and potentials. BK Ca was reported to play a role in regulating proliferation of human preadipocytes [46], endothelial cells [47], and breast cancer cells [48]. I Kir was found to participate in regulating the proliferation of human hematopoietic progenitor cells [49]. Although the underlying mechanisms of ion channels in cell proliferation regulation remain elusive, the involvement of K + channels in cell proliferation was well established [44,45,49]. Further exploration is required to find out whether these ion channels contribute to human cardiac fibroblast proliferation.
Clcn3 channel is regarded as one of the candidate channels for volume regulated anion channels and has been shown to play an important role in cell proliferation and apoptosis [45]. Blockade or disruption of Clcn3 channel resulted in arrest of cell cycle and prevention of cell proliferation in several cell types [21,50]. The present observation demonstrated that functional chloride current encoded by Clcn3, sensitive to cell volume, was observed in human cardiac fibroblasts (Fig. 5). Whether this I Cl current would contribute to human cardiac fibroblast proliferation remains to be studied in the future.
One of limitations of the present study was that ion channels, BK Ca , I to , I Kir and I Cl.vol , and I Na.TTX , and I Na.TTXR , were heterogeneously expressed within the same species of cultured human cardiac fibroblasts. This could result from heterogeneous cell population of the fibroblasts. An earlier study demonstrated that myofibroblast could differentiate from fibroblasts when plated at low density and could revert back to fibroblasts at higher density [51]. Consequently, a subpopulation of human cardiac fibroblasts may display different patterns of ion channel expression.
In summary, the present study provides the first information that multiple ion channel currents are present in cultured human cardiac fibroblasts, the patterns and properties of these ion channel currents differ from those observed in human cardiac myocytes. The information obtained form the present study provides a basis for future study how ion channels participate in regulating cardiac electrophysiology.