Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Nitrovasodilators are frequently used for treatment of coronary heart disease and heart failure. By releasing NO either spontaneously (e.g., sodium nitroprusside) or after both enzymic and nonenzymic metabolism (e.g., isosorbide mononitrate, isosorbide dinitrate and molsidomine), these agents induce relaxation of coronaries and other arteries (Taniguchi et al., 1993; Khan et al., 1993; Archer et al., 1994). The precise mechanism by which NO and other nitrovasodilators cause relaxation of smooth muscle remains to be defined and probably involves multiple mechanisms. Membrane hyperpolarization has been invoked as an important mechanism for the relaxation produced by nitrates in some, but not all, arterial beds (Tare et al., 1990). K⁺ channel activity is the main determinant of membrane potential in smooth muscle cells, and K⁺ efflux resulting from K⁺ channel opening causes hyperpolarization, inhibits voltage-dependent Ca⁺⁺ channels and promotes relaxation (Nelson et al., 1990; Nelson and Quayle, 1995; Gollasch et al., 1992). Recent evidence suggests that NO as well as other nitrovasodilators can activate large-conductance K_Ca channels (Robertson et al., 1993; Taniguchi et al., 1993; Miyoshi and Nakaya, 1994), which may contribute to vessel relaxation (Williams et al., 1988; Taniguchi et al., 1993; Hecker et al., 1995).

K_Ca currents and STOCs have been identified in many types of smooth muscle, including human coronary artery vascular smooth muscle cells (Gollasch et al., 1996). These currents are carried by K_Ca channels. K_Ca currents are activated by submicromolar Ca⁺⁺ as well as by membrane depolarization and are blocked by external tetraethylammonium ions and iberiotoxin (Golasch et al., 1996). STOCs are generated by spontaneous Ca⁺⁺ (calcium sparks) released through ryanodine-sensitive Ca⁺⁺ channels of the sarcoplasmic reticulum (Nelson et al., 1995). Recently, we were able to show that STOCs are present in human coronary arteries and that Ca⁺⁺ entry into the cell through reverse mode Na⁺/Ca⁺⁺ exchanger determines calcium store refilling, which in turn regulates the generation of STOCs in human coronary vascular smooth muscle cells (Bychkov et al., 1997). Whether or not nitrovasodilators affect K_Ca current and STOCs in human coronary smooth muscle is unclear. We present the first direct evidence that nitrovasodilators can activate K_Ca currents and STOCs in human coronaries. We show that activation of K_Ca channels contributes to the vasorelaxing action of these drugs.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Coronary preparations. Human coronary arteries were obtained from patients with dilatative cardiomyopathy, but without significant atherosclerosis, after orthotopic heart transplantation. The tissue was immediately placed in cold (8°C) Hanks' solution (119 NaCl, 4.7 KCl, 1.2 KH₂PO₄, 25.0 NaHCO₃, 1.2 MgSO₄, 11.1 glucose, 0.026 EDTA, 2.5 CaCl₂ mM, 5% CO₂- 95% O₂) during transportation to the laboratory for further dissection. Branches from left, right and circumflex coronary arteries (diameter ~1.5 mm) were dissected and cleansed of adhering tissue and fat in the Hanks' solution.

Contraction recordings. Isometric contractions of coronary artery segments 4 to 5 mm long were measured using a vessel myograph as previously described (Gollasch et al., 1995). Small stainless steel wires (diameter 0.6 mm) were gently inserted into the lumen of the arterial segments under a microscope. The vessels were then transferred into an organ bath (volume, 20 ml) containing Hanks' solution (119 NaCl, 4.7 (or 8.7 or 13.7) KCl, 1.2 KH₂PO₄, 25.0 NaHCO₃, 1.2 MgSO₄, 11.1 glucose, 0.026 EDTA, 2.5 CaCl₂ mM, 5% CO₂- 95% O₂). One of the two wires was connected to a F-30 force transducer (Hugo Sachs, Freiburg, FRG) for isometric tension recordings. The output from the transducer was displayed on a strip chart recorder. The arterial segments were stretched in a stepwise manner to preloads of approximately 2 g. The organ baths were continuously bubbled with carbogen (5% CO₂- 95% O₂) to provide oxygenation and pH of 7.4. The temperature was maintained at 37°C. After equilibration for 1 h, the isometric contraction was measured. The contractile capacity of the arterial segments was assessed by changing the bath solution to an isotonic 50 mM K⁺-Hanks' solution with the following composition (in mM): 75.0 NaCl, 48.8 KCl, 1.2 KH₂PO₄, 25.0 NaHCO₃, 1.2 MgSO₄, 11.1 glucose, 0.026 EDTA, 2.5 CaCl₂ (5% CO₂- 95% O₂). In some experiments, the endothelium was removed by gentle scrubbing of the lumen with a stainless steel rod (Gollasch et al., 1995).

Isolation of smooth muscle cells. Vascular smooth muscle cells were isolated as previously described (Gollasch et al., 1995, 1996). The vessels were cut into small segments (about 3 mm in length) and placed in a Ca⁺⁺-free Hanks' solution containing (in mM) 137 NaCl, 5.4 KCl, 0.44 KH₂PO₄, 0.42 NaH₂PO₄, 2 MgCl₂, 0.05 Ca⁺⁺, 11.11 glucose, 10 HEPES; pH adjusted to 7.4 with NaOH) for 2 to 10 min at room temperature (20°C-24°C). The segments were then placed in the Ca⁺⁺-free solution containing 2 mg/ml collagenase (Sigma type IA; Sigma, Deisenhofen, FRG), 10 mg/ml bovine serum albumin (BSA) and 0.5 mg/ml elastase (Sigma type IIA) and were incubated for 40 min with gentle agitation at 36°C. After the digestion was complete, single cells were dispersed by gentle agitation in the Ca⁺⁺-free Hanks' solution.

K⁺ current recordings. Whole-cell K⁺ currents were measured according to the conventional patch-clamp method of Hamill et al. (1981) (for details see Gollasch et al., 1991, 1993) or using the perforated patch method with nystatin (Gollasch et al., 1996). Cells were held at 80 mV, and linear voltage-ramp pulses at 0.67 V/s from 100 mV to +100 mV or 500-ms depolarizing step pulses to different voltages were applied (stimulation frequency, 0.3 Hz). The membrane capacity was 37 ± 3.8 pF (mean ± S.E.M., n = 16). The external solution E1 contained (in mM) 140 NaCl, 1.8 CaCl₂, 1 MgCl₂, 5.4 KCl, 0.1 CdCl₂, 10 glucose and 10 Na-HEPES (pH 7.4). The patch pipette (resistance, 4-8 MOhm) was filled with a solution I1 containing (in mM) 80 K-aspartate, 50 KCl, 1 MgCl₂, 3 Mg-ATP, 10 EGTA, 5 K-HEPES (pH 7.4). If not otherwise indicated, experiments were done at room temperature (20°C-24°C). Nystatin (Sigma, Deisenhofen, FRG) was dissolved in DMSO and diluted into the pipette solution to give a final concentration ranging from 50 to 100 µg/ml. Whole-cell access was achieved by nystatin within 10 to 20 min of seal formation. Whole-cell currents were recorded using a List EPC-7 or an Axopatch 200A amplifier, digitized at 10 kHz using a CED1401 interface (Cambridge Electronic Design Limited, Cambridge, UK) and analyzed using CED Patch and Voltage Clamp Software Version 6.08.

Materials. Iberiotoxin and DEA-NO were obtained from RBI (Natick, MA). PTIO was purchased from (Sigma, FRG). Sodium nitroprusside was obtained from Sigma (Deisenhofen, FRG). Isosorbide mononitrate was a gift from Astra GmbH (Wedel, FRG). Stock (10 mM) solutions of PTIO were made using DMSO as the solvent.

Statistical analysis. All values are given as mean ± S.E.M.; n represents the number of arterial rings or cells tested. The Wilcoxon rank sum test or the Mann-Whitney-Wilcoxon test was used to determine significant differences. Comparisons of dependent samples were done using one-way analysis of variance and Bonferroni's inequality (Wallenstein et al., 1980). A value of P < .05 was considered significant. The terms increase and decrease are employed only when the results were statistically significant. All contraction experiments examining the effects of iberiotoxin and PTIO on nitrovasodilator relaxation were conducted on coronary arteries from different patients.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Relaxant effects of SNP, DEA-NO and IMN on human coronary arteries. The effects of the nitrovasodilators SNP, DEA-NO and IMN on human coronary artery rings are shown in figure 1. Serotonin 5 µM was given in a sustained fashion over 15 min. The characteristic sustained contractions of human vessels by serotonin were observed. The same serotonin concentrations were found to constrict intact human coronary arteries (McFadden et al., 1991). In addition, we have previously shown that 5 µM serotonin induced sustained contractions, mediated primarily by Ca⁺⁺ influx through voltage-dependent Ca⁺⁺ channels. These contractions did not decrease within 30 to 40 min (Gollasch et al., 1995) in all investigated arteries. SNP, DEA-NO and IMN were added at concentrations ranging from 10 nM to 100 µM. All three nitrovasodilators induced a dose-dependent decrease in vascular tone. Figure 1A shows a preconstricted human coronary artery exposed to increasing doses of SNP. The stepwise relaxation is apparent. Half-maximal relaxation obtained by fit was about 0.72 ± 0.09 µM SNP (IC₅₀; n = 8; fig. 1B). The Hill coefficient (n_H) was 0.95 ± 0.05. The SNP effect was completely reversed with washout. DEA-NO and IMN induced half-maximal relaxation of human coronary arteries at 35.1 ± 5.0 µM (n = 5) and 17.9 ± 3.0 µM (n = 5), respectively (fig. 1C). The Hill coefficients of DEA-NO and IMN dependent relaxation were 1.52 ± 0.23 and 0.80 ± 0.11, respectively. We then repeated these experiments (n = 5) with the endothelium removed from human coronary arteries. Half-maximal relaxation was observed at 0.67 ± 0.08 µM SNP, which was not different from when the endothelium was present. The Hill coefficient was 0.90 ± 0.06. We next studied the effects of NO neutralization. PTIO is known to neutralize NO in biological systems specifically and directly via a unique radical-radical reaction with NO (Miyoshi and Nakaya, 1994). The effects of PTIO on relaxation of human coronary arteries by SNP are presented in figure 1B. After pretreatment with PTIO, SNP produced relaxation at significantly higher doses, with an IC₅₀ of 4.04 ± 0.20 µM (n = 4). The Hill coefficient was 0.95 ± 0.03. 

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Dose-dependent relaxation of human coronary arteries with SNP, DEA-NO and IMN in vessels precontracted with serotonin. Effects of PTIO, iberiotoxin and tetraethylammonium ions on SNP relaxation. A) Dose-dependent relaxation of human coronary arteries with SNP in vessels without (upper trace) and with pretreatment with iberiotoxin (lower trace). The presence of serotonin (5-HT, 5 µM), iberiotoxin (IBTX, 100 nM) and SNP (cumulative doses) in the bath is indicated by horizontal bars. Doses of half-maximal relaxation (IC50) and Hill coefficients (nH) were calculated by fitting the data with the equation where T is tension in response to SNP, DEA-NO or IMN; Be is maximum response induced by SNP, DEA-NO or IMN; B0, is a constant; IC50 is the dose of SNP, DEA-NO or IMN that elicits a half-maximal response and D is the dose of SNP, DEA-NO or IMN. The symbols represent means ± S.E.M. Tension is expressed as a percentage of the steady-state tension (100%) obtained with 5 µM serotonin before administration of the drugs. Panel B) Cumulative dose-response curve for relaxation induced by SNP and by SNP plus PTIO (SNP + PTIO). IC50 and nH were obtained by fit and were IC50 = 0.72 ± 0.09 µM, nH = 0.95 ± 0.05 (n = 8) and IC50 = 4.04 ± 0.20 µM, nH = 0.95 ± 0.03 (n = 4) for SNP and SNP+PTIO, respectively. Cumulative dose-response curve for relaxation induced by SNP in the presence of 100 nM iberiotoxin (SNP-IBTX) or 1 mM tetraethyammonium chloride (SNP-TEA). Tension is expressed as a percentage of the steady-state tension (100%) obtained with 5 µM serotonin plus 100 nM iberiotoxin or with 5 µM serotonin plus 1 mM TEA before administration of SNP. In arteries pretreated with IBTX or TEA, SNP was added cumulatively. IC50 and nH were obtained by fit and were IC50 = 7.20 ± 0.30 µM, nH = 0.91 ± 0.06 (n = 8) and IC50 = 7.09 ± 0.21 µM, nH = 0.89 ± 0.08 (n = 4) for iberiotoxin- and TEA-pretreated coronary rings, respectively. C) Cumulative dose-response curve for relaxation induced by DEA-NO and IMN. Tension is expressed as a percentage of the steady-state tension (100%) obtained with 5 µM serotonin. IC50 and nH were obtained by fit and were IC50 = 35.1 ± 5.0 µM, nH = 1.50 ± 0.16 (n = 5) and IC50 = 17.9 ± 3.0 µM, nH = 0.80 ± 0.09 (n = 5) for DEA-NO and IMN, respectively.

Effect of K_Ca channel blockers on relaxation. Figure 1 shows the effect of the K_Ca channel blocker iberiotoxin on SNP-dependent relaxation in 5 µM serotonin-preconstricted human coronary arteries with intact endothelium. Iberiotoxin 100 nM is known to block K_Ca channels completely in human coronary artery vascular smooth muscle cells (Gollasch et al., 1996; Bychkov et al., 1997). Iberiotoxin elevated the sustained phase of serotonin-induced contraction. After pretreatment with 100 nM iberiotoxin, SNP relaxed rings preconstricted with 5 µM serotonin but produced half-maximal relaxation of arteries at significantly higher doses than without the presence of the K_Ca channel blocker. Half-maximal relaxation at 7.20 ± 0.30 µM SNP (n = 8) was observed in the presence of iberiotoxin (100 nM) in human rings. We next administered tetraethylammonium ions, which block K_Ca channels in human coronary artery smooth muscle cells (concentration of half-block K_i, 0.2 mM; Gollasch et al., 1996). Tetraethylammonium decreased the cumulative relaxation to SNP. In the presence of 1 mM tetraethylammonium, half-maximal relaxation of human coronary arteries was observed at 7.09 ± 0.21 µM SNP (n = 4; fig. 1B).

SNP-induced stimulation of K_Ca current in coronary myocytes. To provide direct evidence that nitrovasodilators open K⁺ channels and hyperpolarize human coronary arterial myocytes, we measured transmembrane K⁺ currents with the patch-clamp technique on single smooth muscle cells from human coronary arteries. The currents were recorded using a high-K⁺ dialyzing pipette solution (I1). Interfering currents through voltage-dependent Ca⁺⁺ channels were blocked by 100 µM Ca⁺⁺. The current-voltage relationships (I-V curve) of outward currents were investigated using step depolarizing pulses in the whole-cell (wc) configuration or perforated patch (pp) configuration with nystatin. Depolarizing step pulses of voltage (duration, 400 ms) were applied from a holding potential of 80 mV as shown in figure 2. The first detectable outward current was observed when the voltage ramp reached approximately 40 mV. For voltages positive to this value, the magnitude of the outward current increased, and at very positive potentials (> +40 mV), the current became very noisy. We have previously shown that the total outward currents were due to K⁺ currents through both K_dr channels and K_Ca channels (Gollasch et al., 1996).

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Effects of SNP on K+ currents in smooth muscle cells isolated from human coronary arteries. A) The plot shows the superimposed current traces recorded during step pulses to potentials between 40 and +80 mV with an increment of 10 mV before (Con) and after application of 1 µM SNP. Holding potential, 80 mV. Con, control. B) Shown are the corresponding current-voltage relationships (I-V curves) (n = 5). The cells responded to SNP with the induction of a voltage-dependent IK. The currents were recorded at 37°C. C and D) IK (shown in panel C) were induced by linear voltage ramp from 100 mV to 0 mV. The currents were filtered at 2 kHz and digitized at 1 kHz. The digitized points are plotted against the corresponding voltage. The SNP-induced current (IK,SNP  IK,Con) (shown in the panel D) was fitted by a Boltzmann equation: where IK,Con is current recorded before SNP, IK,SNP is current recorded in the presence of SNP and V is membrane potential in milli volts.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Inactivation of SNP-induced K+ currents. A) The relationships between peak outward currents recorded at the test potential of +50 mV and membrane potential (preconditioning potential) before and after application of 1 µM SNP (n = 4). In these experiments, membrane potential was held at 80 mV for 60 s. Then it was stepped to preconditioning voltage (10-s prepulse) between 80 mV and +80 mV (in 5-mV increments) and finally to depolarized test potentials (400-ms test pulse) of +50 mV. The fitted curves described the Boltzmann equation with midpoint of inactivation (V0.5) of 28.7 mV and 29.1 mV before and after application of SNP, respectively. The steepness factor (k) was 8 mV before and 8.2 mV after application of SNP, respectively. The Boltzmann equation is: where k is the steepness factor, V0.5 is the midpoint of inactivation and V is the preconditioning potential (all in millivolts). B) The panel shows superimposed current traces recorded during step pulses to test potentials with an increment of preconditioning voltage pulse of 5 mV before (Con) and after application of 1 µM SNP.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Stimulation of KCa currents by SNP in human coronary artery vascular smooth muscle cells. A) KCa current sensitive to IBTX. Shown are superimposed current recordings before (Con) and after iberiotoxin (100 nM, IBTX). B) Stimulation of KCa current by 1 µM SNP. Blockade of the SNP-induced K+ current by 100 nM IBTX. C) Stimulation of KCa current by 1 µM SNP. Glibenclamide (3 µM) did not block SNP-induced potassium currents. Shown are superimposed control currents (Con) and current recordings in the presence of 1 µM SNP, in the presence of 1 µM SNP and 100 nM IBTX (SNP + IBTX) and in the presence of 1 µM SNP and 3 µM glibenclamide (SNP + Glib). In these experiments, membrane potential was held at 80 mV and then stepped to preconditioning voltage (15-s prepulse) of 0 mV (to inactivate Kdr current). This maneuver was followed by depolarized test potentials (400-ms test pulse) of 50 mV for 400 ms.

View larger version (32K): [in this window] [in a new window]   Fig. 5.   Stimulation of STOCs by SNP in human coronary artery vascular smooth muscle cells. A) Single and complex STOCs recorded under steady-state potential held at 20 mV. B) Recording of STOCs under control conditions, after application of SNP (1 µM) and after additional application of IBTX (IBTX, 100 nM).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

We provide the first direct evidence that both K_Ca currents and STOCs in human coronary artery smooth muscle cells can be opened by nitrovasodilators. Furthermore, we demonstrate that opening of these channels is essential to relaxation by nitrovasodilators. We therefore propose the following cascade of events leading to nitrate-induced relaxation of human coronary arteries: 1) activation of STOCs and K_Ca currents in smooth muscle cells, 2) increase in K⁺ efflux, 3) membrane hyperpolarization, 4) closure of voltage-dependent Ca⁺⁺ channels and 5) decrease in Ca⁺⁺ entry and vasorelaxation.

SNP, DEA-NO and IMN are thought to cause relaxation by liberating NO in smooth muscle cells. However, SNP was found to be the most potent nitrodilator for human coronary strip muscle rings. We observed that PTIO, which neutralizes NO (Miyoshi and Nakaya, 1994), significantly inhibited the human coronary vasorelaxations induced by SNP. The vasodilatory response to SNP was not influenced by removal of the endothelium. These studies demonstrate the functional significance of the NO-signaling pathway in dilating human coronary arteries.

In patch-clamp experiments on freshly isolated smooth muscle cells, we observed stimulation of iberiotoxin-sensitive K_Ca currents and STOCs by SNP at concentrations that induce coronary vasorelaxation. Furthermore, we observed that iberiotoxin and tetraethylammonium ions inhibited the dose-dependent relaxations of human coronary arteries by SNP. These results indicate that K_Ca channels are involved in the coronary vascular relaxation by NO in humans. The data provide functional support for the previous observations that NO and nitrovasodilators produce a K⁺ channel-mediated hyperpolarization in arterial beds of several animal species (Tare et al., 1990). The data also support the hypothesis that K⁺ channels integrate a variety of vasoactive signals to dilate coronary arteries through membrane hyperpolarization in coronary artery smooth muscle cells. We have previously shown that opening of K_ATP channels by exogenous or endogenous agonists, e.g., by pinacidil or by pituitary adenylate cyclase peptides, leads to vasorelaxation of human coronary arteries (Gollasch et al., 1995; Bruch et al., 1997; Gollasch et al., 1996). Furthermore, the data from the present study provide an important support for the hypothesis, first presented by Williams et al. (1988), that nitrovasodilators are potent activators of vascular smooth muscle K_Ca channels. Recent studies have reported two pathways of K_Ca channel activation by NO and nitrovasodilators in smooth muscle. Whereas Bolotina et al. (1994) suggested that this class of channels can be activated directly in vascular smooth muscle, other investigators provided patch-clamp data showing that NO can stimulate K_Ca via cyclic GMP-dependent protein kinase (Taniguchi et al., 1993; Robertson et al., 1993; Archer et al., 1994; Koh et al., 1995). NO-induced stimulation of STOCs has been reported in a previous study using SNP and pulmonary arterial smooth muscle cells (Clapp and Gurney, 1991) Suggesting activation of Ca sparks.

In conclusion, these are the first results showing that nitrovasodilators have an effect on K_Ca channels in human vascular smooth muscle cells. Furthermore, the data provide evidence for the modulation of this channel by NO and suggest that these channels play an important role in mediating the therapeutic responses of nitrovasodilators. We suggest that just as the K_ATP channels have been shown to play an important role in modulating human coronary artery relaxation during hypoxia and in response to drugs such as pinacidil (Gollasch et al., 1995), the present study shows that K_Ca channels may play a similar role in the regulation of vascular tone by nitrates. These findings may have clinical significance for the development of antianginal and antihypertensive drugs that selectively target K⁺ channels and calcium sparks.