Regulation of Ca2+ Homeostasis by Atypical Na+ Currents in Cultured Human Coronary Myocytes
Abstract—Primary cultured human coronary myocytes (HCMs) derived from ischemic human hearts express an atypical voltage-gated tetrodotoxin (TTX)-sensitive sodium current (INa). The whole-cell patch-clamp technique was used to study the properties of INa in HCMs. The variations of intracellular calcium ([Ca2+]i) and sodium ([Na+]i) were monitored in non–voltage-clamped cells loaded with Fura-2 or benzofuran isophthalate, respectively, using microspectrofluorimetry. The activation and steady-state inactivation properties of INa determined a “window” current between −50 and −10 mV suggestive of a steady-state Na+ influx at the cell resting membrane potential. Consistent with this hypothesis, the resting [Na+]i was decreased by TTX (1 μmol/L). In contrast, it was increased by Na+ channel agonists that also promoted a large rise in [Ca2+]i. Veratridine (10 μmol/L), toxin V from Anemonia sulcata (0.1 μmol/L), and N-bromoacetamide (300 μmol/L) increased [Ca2+]i by 7- to 15-fold. This increase was prevented by prior application of TTX or lidocaine (10 μmol/L) and by the use of Na+-free or Ca2+-free external solutions. The Ca2+-channel antagonist nicardipine (5 μmol/L) blocked the effect of veratridine on [Ca2+]i only partially. The residual component disappeared when external Na+ was replaced by Li+ known to block the Na+/Ca2+ exchanger. The resting [Ca2+]i was decreased by TTX in some cells. In conclusion, INa regulates [Ca2+]i in primary cultured HCMs. This regulation, effective at baseline, involves a tonic control of Ca2+ influx via depolarization-gated Ca2+ channels and, to a lesser extent, via a Na+/Ca2+ exchanger working in the reverse mode.
Ca2+ ions play a central role in the regulation of arterial smooth muscle tone.1 2 The changes in the cytosolic free calcium concentration ([Ca2+]i) mediate the effect of various vasoconstrictor and vasodilator agents, including hormones, neurotransmitters, and drugs that bind to specific receptors. Both direct mobilization of intracellular Ca2+ and transmembrane Ca2+ entry can activate contraction of smooth muscle cells.2 3 The relationship between smooth muscle membrane potential and arterial tone is quite steep.3 Membrane depolarization results in activation of voltage-gated Ca2+ channels, thereby leading to Ca2+ influx. In contrast, membrane hyperpolarization through activation of potassium (K+) channels is an effective mechanism to inactivate Ca2+ channels and thus dilate arteries.3
Voltage-gated Na+ channels are generally responsible for the fast depolarizing phase of the action potential in most electrically excitable tissues. In response to moderate depolarization, these channels open for a few milliseconds and, thereby, generate a rapid and massive influx of Na+ ions promoting the large depolarization that gates other voltage-activated ion channels. Na+ channels are present in a wide variety of cells, including neurons and heart and skeletal myocytes.4 5 6 In contrast, Na+ channels have only recently been detected in arterial smooth muscle cells.7 8 9 10 11 We have recently identified a tetrodotoxin (TTX)–sensitive INa in primary cultured human coronary myocytes (HCMs) derived from end-stage failing hearts of transplanted patients with an ischemic cardiopathy.10 This current exhibits quite unusual properties, including the presence of a sustained component resulting from very slow inactivation. The major aim of the present work was to investigate whether, and how, this current plays a role in the regulation of Ca2+ homeostasis in HCMs.
Materials and Methods
Myocyte Isolation and Culture
Tissue samples were obtained from the left descending coronary artery of 6 male patients (age range, 44 to 58 years) undergoing heart transplantation. Those patients had end-stage heart failure (New York Heart Association, classes III and IV) caused by an ischemic disease. Cells were enzymatically isolated and grown in culture as described previously.12 13 Briefly, arteries were collected aseptically after cardiectomy, and myocytes were then dispersed in Hanks’ solution containing 0.6 mg/mL collagenase (Worthington) and 1 mg/mL elastase (Boehringer Mannheim) for at least 50 minutes at 37°C. Next, the cells were centrifuged (100g, 7 minutes); resuspended in Ham’s F-10 (Eurobio) supplemented with 10% human serum (Institut Jacques Boy SA), 2 mmol/L glutamine, and antibiotics (Eurobio); and plated (10 000 cells per mL) in 35-mm disposable Petri dishes (Falcon) that were placed at 37°C in an air/CO2 incubator. The start medium was changed every day during the first week. Thereafter, the cells were transferred to maintenance medium (DMEM and Ham’s F-10; 1:1 vol/vol; Eurobio) containing human serum (5%) and 5% FBS (Myoclone Super Plus; Life Technology, Cergy Pontoise, France), which was changed every 2 to 3 days. Confluence was obtained 15 to 20 days after plating.
The whole-cell patch-clamp technique14 was performed at room temperature (20°C to 22°C). For whole-cell recordings, pipettes (borosilicate glass, Sutter) were filled with the following (in mmol/L): CsCl 130, EGTA 10, HEPES 25, Mg-ATP 3, Na-GTP 0.5, glucose 10, succinic acid 5, and aspartic acid 5, with pH adjusted to 7.3 with CH3SO3H. The bathing solution contained the following (in mmol/L): NaCl 140, MgCl2 2, CaCl2 0.02, HEPES 10, 4-aminopyridine 5, and glucose 10, pH adjusted to 7.3 with CH3SO3H. Junction potential was zeroed before seal formation. Voltage errors resulting from the residual, uncompensated series resistance (≤1 MΩ) were estimated to be ≤2 mV (INa<2 nA). Leak and capacitive currents were subtracted using a 5-subpulse method. Sampling frequencies ranged from 0.16 to 2 kHz, and current signals were filtered at 3 to 5 kHz before digitization and storage. The holding potential (HP; −100 mV), test potentials, and rate of stimulation (0.1 Hz) were controlled with an IBM personal computer connected to the electrophysiological equipment (Axopatch 200A amplifier; Axon Instruments). Data acquisition and analysis were performed using the pCLAMP software (version 6.02).
Solutions for [Ca2+]i and [Na+]i Measurement Experiments
Standard normal Locke buffer was used. The buffer contained the following (in mmol/L): NaCl 140, KCl 5, KH2PO4 1.2, MgSO4 1.2, CaCl2 1.8, glucose 10, and HEPES 10, with pH 7.2 adjusted with NaOH. A Na+-free solution was prepared by replacing Na+ with N-methyl-d-glucamine chloride. The high K+ solution contained the following (in mmol/L): NaCl 90, KCl 50, KH2PO4 1.2, MgSO4 1.2, CaCl2 1.8, glucose 10, and HEPES-NaOH 10, pH 7.2. The osmolarity of all of the solutions ranged between 298 and 303 mosm/L.
Dye Loading and Measurement of [Ca2+]i and [Na+]i
The [Ca2+]i in single cells was measured as described previously.15 Briefly, the cells were loaded by incubation with 2.5 μmol/L Ca2+-sensitive dye fura-2-acetoxymethyl ester (AM) (dissolved in DMSO) plus 0.02% Pluronic F-127, a surfactant polyol (dissolved in water; Molecular Probes Inc), in Locke buffer. Dye loading was carried out for 40 minutes at 37°C in a humidified air atmosphere. The loaded cells were subsequently rinsed several times in Locke buffer and then mounted on the microscope stage. Fluorescence measurements of [Ca2+]i were performed with the Zeiss Microscope Photometer System (FFP, Zeiss), based on an inverted microscope (Axiovert 100, Zeiss) equipped for epifluorescence (objective, Plan-Neofluar 100×/1.30 oil immersion). With fluorescence values corrected for background and dark current, the [Ca2+]i were calculated from the ratio of 340-/380-nm recordings, in accordance with the equation given by Grynkiewicz et al.16 Fura-2 calibration was performed as described previously.17
The [Na+]i was measured in individual cells that had been loaded with the Na+ indicator benzofuran isophthalate (SBFI, 5 μmol/L; Sigma). To load with SBFI-AM, cells were incubated in Locke buffer containing 5 μmol/L SBFI-AM, 0.02% wt/vol Pluronic F-127 for 60 minutes at 37°C in a humidified air atmosphere. After loading, coverslips were rinsed with Locke buffer and mounted in a recording chamber. All recordings were made at room temperature (21°C to 23°C) using fast fluorescence photometry equipment previously described for [Ca2+]i measurements. For [Na+]i measurements, fluorescence ratios obtained from background-subtracted fluorescence signal at each wavelength were converted to [Na+]i using the following equation described by Grynkiewicz et al16 : [Na+]i=K×(Ratio−Rmin)/(Rmax−Ratio), where K is a constant describing the apparent affinity of the dye for Na+ and is related to the dissociation constant (Kd) of SBFI (18 mmol/L). Rmin and Rmax are the fluorescence ratios measured in the nominal absence and in the presence of saturating amounts of ion, respectively. K, Rmin, and Rmax were obtained from standard curves. Because the spectral characteristics of SBFI in the cytosol are different from those of SBFI in solution, the [Na+]i standard curve was constructed from fluorescence ratios obtained in situ on exposing cells loaded with SBFI-AM to perfusion solutions containing known concentrations of Na+ (0, 50, 100, and 140 mmol/L) and 5 μmol/L gramicidin D.18 In our experimental conditions, when measuring the fluorescence ratio for these concentrations of Na+, the relationship between the fluorescence ratio and [Na+]i was linear and the levels of [Na+]i happened to fall within the most linear part of the equation. Experiments were performed after a resting period of 15 minutes from the end of the incubation. Transient variations of [Na+]i and [Ca2+]i were measured at their maximal amplitude.
TTX, lidocaine, N-bromoacetamide, toxin V from the sea anemone Anemonia sulcata (a generous gift of Dr H. Schweitz and Prof M. Lazdunski, Nice, France), and nicardipine (Sandoz) were prepared as stock solutions (1 mmol/L for TTX in 0.1% acetic acid, 0.1 mmol/L each lidocaine and N-bromoacetamide in double-distilled H2O, and 1 mmol/L nicardipine in dry DMSO), stored at −20°C, and subsequently diluted at the desired working concentrations in test solutions. The lipid-soluble plant alkaloid toxin veratridine (Sigma) was prepared extemporaneously (0.01 mol/L stock solutions in 0.1N HCl). The control and test solutions were applied using a multiple capillary perfusion system (200 μm inner diameter tubing, flow rate 100 μL/min) placed to the proximity of each cell tested (<0.5 mm). Each capillary was fed by a reservoir 50 cm above the bath. After each application, the cells were washed with Locke buffer. Incubations (5 minutes) with inhibitory substances were carried out in a 500-μL bath containing inhibitors diluted in Locke buffer.
The results were analyzed using the Student t test. Unless otherwise stated, all inhibitors used here showed a “significant” effect at P<0.01. The results are expressed as mean±SEM.
Window Current and Slow Inactivating Component
INa in HCMs activates at voltages positive to −50 mV and peaks between −10 and 0 mV.10 This is illustrated in Figure 1A⇓. Another surprising feature is the presence of a slow inactivating current component with relatively large amplitude (up to 30% of total peak amplitude in some cells), which is totally blocked by 1 μmol/L TTX (Figure 1B⇓). The steady-state inactivation curve of INa, determined by applying various conditioning potentials for 5 seconds before the test depolarization, shows that a significant fraction of INa can still activate from depolarized membrane potentials (Figure 1C⇓). The membrane potential at which 50% of the channels are still available for opening (V0.5) is around −45 mV (Figure 1D⇓). Furthermore, the overlap between the conductance and inactivation curves determines a “window” between −50 mV and −10 mV (Figure 1D⇓), suggesting that a fraction of the Na+ channels are open and generate a steady-state Na+ influx in this range of potentials. At the minimum voltage required for activation (−40 mV), a substantial noninactivating Na+ influx was indeed observed (Figure 1E⇓). However, a persistent component was also observed at all potentials of the I-V relationship (Figure 1A⇓), ie, at potentials far higher than those determining the window current, indicating that a slow inactivating current is also generated by particular gating properties of the channels. Therefore, in addition to the overlap in the voltage-dependent activation/inactivation properties, the slow decay kinetics of INa constitute an intrinsic mechanism to promote massive Na+ influx into the cells. Interestingly, the sustained component of INa (INa, sus) was less sensitive to voltage-dependent inactivation than the peak current (INa, peak; Figure 1D⇓), implying that its amplitude also contributes to determine the amplitude of the steady-state window current (eg, at −30 mV; Figure 1D⇓).
Effect of Na+ Channel Agonists
To increase Na+ channel activity in non–voltage-clamped cells, HCMs were subjected to pharmacological agents known to be potent Na+ channel agonists. Although INa in HCMs has particular properties, it can be blocked or enhanced by various specific agents and toxins10 that bind on receptor sites of the vertebrate Na+ channel family. The water-soluble heterocyclic guanidine TTX binds to site 1 and block INa in HCMs (Figure 1B⇑). Toxin binding on the other receptor sites generally increases Na+ influx, although there are substantial mechanistic differences among the effects of various activators on INa. This was also the case in HCMs for veratridine, which binds to site 2, and for toxin V from A sulcata, which binds to site 3.4 5 19 The effect of veratridine is complex, as illustrated in Figure 2A⇓. Veratridine (10 μmol/L) decreased the peak amplitude of INa by suppressing completely the fast inactivating component in all cells tested (n=12). However, it promoted at the same time the appearance of both a large sustained component during the test depolarization and a large, slowly deactivating tail current at repolarization.10 The effect of toxin V from A sulcata (100 nmol/L) was different. This toxin increased consistently the peak amplitude of INa (33±10%; n=6) and slowed significantly its decay, with no major effect on current deactivation (Figure 2B⇓). However, the agonistic effect of both toxins was blocked by addition of 1 μmol/L of TTX (Figure 2A⇓ and 2B⇓) in all cells tested (n=4 for each). Veratridine and toxin V from A sulcata had no significant effect on the activation threshold of INa in none of the four cells tested. In addition to these two natural toxins, we also used N-bromoacetamide, a chemical known to prolong the open time of Na+ channels and to slow the kinetics of macroscopic INa.4 In HCMs, we found that N-bromoacetamide (300 μmol/L) had no significant effect on current amplitude (−5±8% decrease; n=4) but prevented the inactivation of INa with no or only minor effect on the deactivating tail current. N-Bromoacetamide had no effect on the activation threshold of INa (data not shown). Therefore, we expected that, despite clear mechanistic differences in their effects on Na+ channels gating in HCMs, veratridine, toxin V from A sulcata and N-bromoacetamide could be used to increase Na+ influx through drug-modified channels into non–voltage-clamped myocytes.
Na+ Channel-Dependent Increase of [Na+]i and [Ca2+]i
We investigated whether the modulation of Na+ channel activity results in modulation of [Ca2+]i in resting (non–voltage-clamped) HCMs. To increase Na+ channel activity, the cells were subjected to veratridine (10 μmol/L). First, we investigated the effect of veratridine on [Na+]i in SBFI-loaded cells (see Materials and Methods). In control cells, the mean fluorescence intensity ratio was 1.81±0.26 (n=4), which corresponds to 8.6±0.2 mmol/L [Na+]i. The resting value remained stable for up to 1 hour under superfusion of standard Locke solution. Brief applications of veratridine induced large increases of [Na+]i (Figure 3A⇓ and 3B⇓). On average, the resting [Na+]i was augmented to 25.9±2.9 mmol/L [Na+]i (≈3-fold increase), suggesting that veratridine could be used to increase Na+ channel activity.
The next experiments were performed in fura-2–loaded cells (see Materials and Methods). First, we found that addition of the depolarizing agent K+ (KCl 50 mmol/L) induced a transient increase of [Ca2+]i from 41±7 to 447±70 nmol/L (data not shown) in all cells tested (n=8), suggesting the involvement of depolarization-activated Ca2+ channels. Then we assessed the effects of Na+ channel antagonists and agonists without the help of high extracellular K+. Figure 4A⇓ through 4C shows that N-bromoacetamide, veratridine, and toxin V from A sulcata induced large rises in [Ca2+]i. Figure 4A⇓a shows the effects of N-bromoacetamide at 30 and 300 μmol/L. On average, N-bromoacetamide (300 μmol/L) increased [Ca2+]i from 47±4 to 368±40 nmol/L (n=5) (Figure 4A⇓a and 4Ab), veratridine (10 μmol/L) increased [Ca2+]i from 78±6 to 611±44 nmol/L (n=12) (Figure 4B⇓a and 4Bb), and toxin V from A sulcata (100 nmol/L) increased [Ca2+]i from 59±5 to 588±42 nmol/L (n=11) (Figure 4C⇓a and 4Cb). The increase of [Ca2+]i was greater in amplitude with veratridine and toxin V from A sulcata than with N-bromoacetamide. Biphasic responses (even oscillatory activity) were sometimes observed, possibly reflecting release of Ca2+ from intracellular stores. In total, >80% of the cells responded to Na+ channel agonists.
The rise in [Ca2+]i was strictly dependent on the presence of external Na+, consistent with our previous observation that INa is abolished in Na+-free conditions.10 The increase in [Ca2+]i induced by veratridine and toxin V from A sulcata was indeed abolished when extracellular Na+ was substituted with the nonpermeating N-methyl-d-glucamine ion (Figure 5A⇓a). On average, veratridine, which increased [Ca2+]i from 64±9 to 639±36 nmol/L (n=4) in the presence of external Na+, had no effect after preincubation of the cells in the Na+-free solution (Figure 5A⇓b). The [Ca2+]i remained stable at 73±12 nmol/L. Figure 5B⇓a shows that the rise in [Ca2+]i induced by toxin V from A sulcata was also abolished when extracellular Ca2+ was removed. On average, toxin V from A sulcata, which increased [Ca2+]i from 66±9 to 467±88 nmol/L (n=5) in control conditions, had no significant effect after external Ca2+ was removed (Figure 5B⇓a). The [Ca2+]i was 86±6 nmol/L (n=5) in free Ca2+ conditions (Figure 5B⇓b). Taken together, these results suggest that transmembrane influxes of both Na+ and Ca2+ ions are required to observe any rise in [Ca2+]i after activation by veratridine and by toxin V from A sulcata.
The effects of veratridine and toxin V from A sulcata were prevented in the presence of 1 μmol/L TTX (Figure 6A⇓ and 6B⇓) and of 10 μmol/L lidocaine (Figure 6C⇓). The effects of agonists were recovered on washout of the antagonists (eg, see Figure 6A⇓). These experiments confirmed that the effects induced by the agonists reflect a genuine increase of Na+ channel activity. Furthermore, using the whole-cell technique (see Materials and Methods), we found that veratridine and toxin V from A sulcata were unable to promote any current in all of 6 cells with no detectable basal macroscopic INa, suggesting that these agonists do not activate a silent Na+ channel (as has been reported for veratridine in rat aorta),20 but rather modulate the activity of channels active at baseline. It was also unlikely that veratridine and toxin V cause any Ca2+ influx, eg, by increasing the permeability of Na+ channels for Ca2+ or by direct activation of Ca2+ channels (INa is decreased by external Ca2+; S.R., J.F.Q., unpublished results, 1997).
Involvement of Ca2+ Channels and Na+/Ca2+ Exchanger
We assessed whether the rise in [Ca2+]i induced by veratridine or toxin V from A sulcata involved depolarization-activated Ca2+ channels. The rise in [Ca2+]i observed after K+ depolarization was abolished by a saturating concentration of nicardipine (5 μmol/L; data not shown), which blocks all of the L-type Ca2+ current in these cells.12 Veratridine, which increased [Ca2+]i from 70±6 to 782±71 nmol/L (n=5) in control conditions, had a much smaller effect in the presence of nicardipine (177±16 nmol/L) (Figure 7A⇓a and 7Ab), suggesting that voltage-gated Ca2+ channels are the main pathway involved in the response. Nevertheless, a residual nicardipine-insensitive rise in [Ca2+]i could still be observed consistently (Figure 7A⇓a and 7Ab). But, when extracellular Na+ was replaced by Li+ on an equimolar basis (Na+-free medium), this residual component was abolished (Figure 7B⇓a). Similar results were observed in all cells tested (n=4). Because Li+ permeates through Na+ channels but is not taken by the Na+/Ca2+ exchanger, these data suggest that the dihydropyridine-insensitive Ca2+ influx is provided by the exchanger working in reverse mode. Consistent with this observation, we found that, although Li+ and Na+ have similar permeating properties through the Na+ channels and, presumably, equivalent depolarizing effects, veratridine increased [Ca2+]i from 34±7 to only 389±34 nmol/L (n=4) when Li+ was used as the permeating ion (in absence of Ca2+ channel blocker) (Figure 7B⇓b). Therefore, the rise was significantly lower than when Na+ was used as the permeating ion (611±44 nmol/L; n=12), confirming that the dihydropyridine-sensitive pathway is not the only route for Ca2+ entry. This latter result was also consistent with the participation of the Na+/Ca2+ exchanger in the Ca2+ influx activated after Na+ channel activation.
Na+ Channel-Dependent Decrease of Resting [Na+]i and [Ca2+]i
The fact that veratridine induces a rise in [Na+]i suggests that the Na+ channels are active at baseline, given that this substance (and toxin V from A sulcata as well) would not work on closed channels. Indeed, the electrophysiological characterization is consistent with some Na+ channels being open in HCMs with membrane potentials between −50 and −10 mV (Figure 1D⇑). We further assessed the existence of such a steady-state Na+ influx in resting (non–voltage-clamped) SBFI-loaded cells. Brief applications of 1 μmol/L TTX induced a decrease in the fluorescence ratio (Figure 8A⇓) from 1.93±0.14 to 0.58±0.06, corresponding to a decrease in [Na+]i from 10.6±0.8 to 3.2±0.3 mmol/L (n=4). This decrease occurred within seconds after application of TTX and lasted as long as the blocker was applied. We next addressed the question of whether such a steady-state Na+ influx could eventually control the resting [Ca2+]i. Figure 8B⇓ illustrates the effect of TTX on the resting [Ca2+]i as observed in 3 different fura-2–loaded cells. No significant effect was observed in 5 other cells tested. The cells that responded had a high resting [Ca2+]i level (>200 nmol/L), whereas the other had a resting [Ca2+]i level <100 nmol/L (data not shown).
We recently reported the presence of voltage-activated INa in cultured HMCs.10 Because of particular voltage- and time-dependent properties (compared with most types of INa), INa here is likely to exert a unique type of regulation on the resting [Ca2+]i. In the present work, we show that INa actually regulates [Na+]i and [Ca2+]i. This type of regulation seems well adapted to vascular physiology.
First, we demonstrated that an enhancement of Na+ channel activity using various agonists (N-bromoacetamide, veratridine, and toxin V from A sulcata) increases [Ca2+]i in the vast majority of HCMs. Clear evidence, such as prevention of the increase in [Ca2+]i by external Na+ depletion and by TTX or lidocaine, showed that this rise reflects a genuine effect on Na+ channels.
Second, we showed that the rise in [Ca2+]i induced by application of Na+ channel agonists occurs mainly as a consequence of the opening of voltage-activated Ca2+ channels. The presence of Ca2+ channels in the HCMs,12 the requirement of extracellular Ca2+ to observe the effects of INa on [Ca2+]i, and the fact that the rise in [Ca2+]i is largely antagonized by nicardipine all suggest that voltage-gated Ca2+ channels are a major route for the transmembrane Ca2+ influx and a key step between Na+ channel activation and elevation of [Ca2+]i. Therefore, our proposed cascade is the following: (1) the Na+ channel agonists enhance Na+ channel activity and, thereby, generate a Na+ influx, and (2) this Na+ influx, in turn, promotes the membrane depolarization that gates the Ca2+ channels and thus produces the Ca2+ influx. Although the cultured HCMs express both T- and L-type Ca2+ channels,12 the effects described here concern only L-type channels, because T-type Ca2+ channels are completely inactivated at the voltages required to open the Na+ channels (>−50 mV).
Third, part of the transmembrane Ca2+ influx that contributes to the rise in [Ca2+]i induced by Na+ channel agonists is not mediated by Ca2+ channels. This contribution, which is smaller than that of Ca2+ channels, is independent of the depolarizing effect induced by Na+ (or Li+) influx through Na+ channels on membrane potential. It seems rather related to the nature of the permeating ion, because it is not observed when Li+ is used as the depolarizing permeating ion. Consistently, the nicardipine-insensitive rise in [Ca2+]i is suppressed when external Na+ is replaced by Li+, which blocks the Na+/Ca2+ exchanger. Therefore, enhanced activity of the Na+/Ca2+ exchanger working in the reverse mode (entry of Ca2+ against Na+ extrusion) provides an additional route for transmembrane Ca2+ influx after activation of Na+ channels in HCMs.
Fourth, we showed that, because of their particular electrophysiological properties, Na+ channels may play a role in the control of the resting [Na+]i and, thereby, of the basal [Ca2+]i in HCMs. Two factors contribute to the development of a steady-state Na+ influx. One is the sustained current predicted by the Hodgkin-Huxley analysis that occurs as a result of the voltage dependence of INa activation and inactivation overlap. The other is the presence of slowly inactivating Na+ currents, which could be observed even at positive potentials, raising the possibility that these currents are generated by an unusual subtype of Na+ channels. However, whether the transient and the sustained component are related to one or two channels is not completely clear.
The fact that Na+ channels activators increase [Ca2+]i as a result of the opening of voltage-gated Ca2+ channels could be considered as relatively implicit, although this occurs at potentials relevant to vascular physiology. It could also be argued that the use of exogenous substances leads to unphysiological increases in [Na+]i. We actually used these agonists to identify clearly the mechanisms involved downstream. Another important result from these experiments is that some Na+ channels stay open at baseline potentials. Indeed, N-bromoacetamide, veratridine, and toxin V from A sulcata, the primary effect of which is to cause Na+ channels to open more easily and/or to stay open longer than normal,4 5 induced large rises in [Ca2+]i without the help of the usual depolarizing agent, K+. Consistently, we observed a TTX-induced decrease of [Na+]i, confirming the contribution of a basal steady-state Na+ influx to the resting [Na+]i in non–voltage-clamped cells. Therefore, the regenerative mechanism probably involved in the effect of agonists would lead to the following sequence of events: (1) the Na+ channel agonist prolongs the open time of channels already open and may also help opening some others; (2) this leads to increased Na+ influx; and (iii) as channels become activated, the resulting depolarization recruits additional Na+ channels, which, ultimately, leads to activation of voltage-gated Ca2+ channels.
The prolonged activity of the HCM Na+ channels during sustained depolarization, their availability for opening from relatively depolarized membrane potentials, and the existence of a substantial window current at a wide range of potentials, are likely to confer a potential physiological role to these Na+ channels in the vascular myocytes. In a variety of neurons, such noninactivating INa act physiologically to amplify synaptic potentials and set repetitive action potentials to enhance endogenous rhythmicity.21 22 23 24 25 In cardiac cells, it is involved in the regulation of action potential duration and resting membrane potential.26 27 28 29 In HCMs, INa may contribute to a tonic control of Ca2+ channel activity, possibly counterbalancing membrane hyperpolarization through activation of K+ channels, and as a result of [Ca2+]i. Interestingly, we detected a TTX-sensitive decrease in [Ca2+]i in some cells with a high resting [Ca2+]i, which may occur because of the overlap of the window currents generated by both Na+ channels and L-type Ca2+ channels. Because ICa starts to activate at ≈−30 mV,12 this tonic control of Ca2+ channel activity can probably be observed only in depolarized cells. The steady-state Na+ influx may therefore ensure a fine graded regulation of the steady-state Ca2+ influx. It is certainly an effective mechanism to induce and maintain cell depolarization, to turn on Ca2+ channel activity and promote sustained Ca2+ influx and, in addition, to maintain elevated [Na+]i.
Because INa was observed in primary cultured cells, it is difficult to speculate on any physiological function in vivo at the moment. Direct extrapolation from in vitro observations to in vivo physiology must be considered with caution, specially because expression of INa seems to be related to cell dedifferentiation.10 Nevertheless, it is conceptually interesting to note that the repertoire of ion currents of HCMs is changing (eg, T-type Ca2+ currents) during phenotypic modulation of the cells in vitro.12 Interestingly, vascular smooth muscle cells in culture undergo many changes resembling those occurring in diseased vessels. Indeed, dedifferentiation and proliferative disorders play a major part in coronary artery diseases, including atherosclerosis, neointimal formation after endothelial injury, restenosis after angioplasty, and also hypertension. Therefore, the possibility that INa is expressed under certain pathophysiological circumstances and helps regulate the basal arterial tonus, or other Ca2+- or Na+-dependent function(s), is worth considering and will be explored in the near future.
This work was supported by Fondation pour la Recherche Médicalé (Languedoc Roussillon), Fondation de France (Grant 97003982 to S.R.), and by a Société Française d’Hypertension Artérielle grant (to C.C.). We thank Dr Hélène Widmer (INSERM Montpellier), Dr Gerald Zamponi (University of Calgary, Alberta, Canada), and Dr Colin Brown (University of Edinburgh, UK) for helpful comments and critical reading of the manuscript.
↵1 Both authors contributed equally to the study.
- Received October 12, 1998.
- Accepted July 16, 1999.
- © 1999 American Heart Association, Inc.
Filo RS, Bohr DF, Rüegg JC. Glycerinated skeletal and smooth muscle: calcium and magnesium dependence. Science. 1965;147:1581–1583.
Nelson MT. Ca2+-activated potassium channels and ATP-sensitive potassium channels as modulators of vascular tone. Trends Cardiovasc Med. 1993;3:54–60.
Hille B. Ionic Channels of Excitable Membranes. 2nd ed. Sunderland, Mass: Sinauer Associates Inc; 1992.
Catterall WA. Molecular analysis of voltage-gated sodium channels in the heart and other tissues. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Philadelphia, Pa: W. B. Saunders Co, Inc; 1995:1–11.
Fozzard HA, Hanck DA. Structure and function of voltage-dependent sodium channels: comparison of brain and cardiac isoforms. Physiol Rev. 1996;76:887–926.
Okabe K, Kitamura K, Kuriyama H. The existence of a highly tetrodotoxin sensitive Na channel in freshly dispersed smooth muscle cells of the rabbit main pulmonary artery. Pflügers Arch. 1988;411:423–428.
James AF, Okada T, Horie M. A fast transient outward current in cultured cells from human pulmonary artery smooth muscle. Am J Physiol. 1995;268:H2358–H2365.
Quignard JF, Ryckwaert F, Albat B, Nargeot J, Richard S. A novel tetrodotoxin-sensitive Na+ current in cultured human coronary myocytes. Circ Res. 1997;80:377–382.
Dayanithi G, Widmer H, Richard P. Vasopressin-induced intracellular Ca2+ increase in isolated rat supraoptic cells. J Physiol. 1996;490 (pt 3):713–727.
Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985;260:3440–3450.
Harootunian A, Kao JPY, Eckert BK, Tsien RY. Fluorescent ratio imaging of cytosolic free Na in individual fibroblasts and lymphocytes. J Biol Chem. 1989;264:19458–19467.
Schweitz H, Vincent JP, Barhanin J, Frelin C, Linden G, Hugues M, Lazdunski M. Purification and pharmacological properties of eight sea anemone toxins from Anemonia sulcata, Anthopleura xanthogrammica, Stoichactis giganteus, and Actinodendron plumosum. Biochemistry. 1981;20:5245–5252.
French CR, Sah P, Buckett KJ, Gage PW. A voltage-dependent persistent sodium current in mammalian hippocampal neurons. J Gen Physiol. 1990;95:1139–1157.
Coraboeuf E, Deroubaix E, Coulombe A. Effect of tetrodotoxin on action potentials of the conducting system in the dog heart. Am J Physiol. 1979;236:H561–H567.
Kiyosue T, Arita M. Late sodium current and its contribution to action potential configuration in guinea pig ventricular myocytes. Circ Res. 1989;64:389–397.