INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IT IS A WELL-KNOWN CLINICAL observation that the Q-T interval of the electrocardiogram is generally longer in women than in men (1, 6, 11, 21) and that there is also a greater risk of women developing drug-related torsades de pointes with a prolonged Q-T interval (11). Furthermore, it has been shown that sex hormones prolong the Q-T interval and downregulate K⁺ channel expression (6). These facts suggest that sex hormones may have a direct and an indirect effect on cardiac repolarization. Although 17-estradiol is known to affect cardiovascular function (19, 27, 30), its effects on cardiac membrane potentials have not been fully elucidated.

The repolarization phase of the cardiac action potential is formed by several ionic currents, including inward Ca²⁺ current (I_CaL), transient outward K⁺ current (I_to), and delayed outward K⁺ current (I_K), which overlap each other with similar time courses (4, 16). Previous studies demonstrated that 17-estradiol inhibited I_CaL and shortened the action potential duration (APD) in guinea pig ventricular muscles and myocytes (7, 9, 10). Although these findings are in line with the reported negative inotropic effect of 17-estradiol on cardiac preparations (20, 26), they do not explain the clinical observations of prolonged Q-T interval and high prevalence of torsades de pointes in women. Recently, 17-estradiol was found to prolong the APD due to inhibition of I_to in rat ventricular myocytes (2). Because guinea pig and rat ventricular myocytes have different components of repolarizing K⁺ currents with a less developed I_to in the former and a prominent I_to in the latter, different effects of estrogen may arise from different components of the repolarizing currents. We thus investigated the effects of 17-estradiol on action potential and ionic currents responsible for the repolarization phase of the action potential in guinea pig ventricular myocytes.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Preparation of guinea pig ventricular myocytes. Single ventricular myocytes from guinea pig hearts were prepared by a previously described enzymatic dissociation procedure (8). We used mainly female guinea pigs weighing 300-400 g, unless otherwise stated. Briefly, animals were anesthetized with pentobarbital sodium (15-20 mg/kg ip). The chest was opened under artificial respiration, and the aorta was cannulated in situ before the heart was removed. By use of a Langendorff apparatus, the excised heart was first perfused with normal Tyrode solution and then with nominally Ca²⁺-free Tyrode solution for 5 min. Subsequently, Ca²⁺-free Tyrode solution with collagenase (0.6 mg/ml, type II; Worthington Biochemical, Lakewood, NJ) was perfused through the heart for ~20 min. The temperature of all perfusates was kept constant at 36-37°C. Single cells were obtained by gentle agitation of small pieces of ventricular tissue in high-K⁺, low-Cl solution. Harvested cells were stored in the high-K⁺, low-Cl solution, kept for 60 min at 4°C, and then transferred to normal Tyrode solution at room temperature before use.

Solutions. The normal Tyrode solution contained (in mM) 144 NaCl, 4.0 KCl, 4.0 CaCl₂, 0.53 MgCl₂, 0.33 NaH₂PO₄, 5.5 glucose, and 5.0 HEPES, and the pH was adjusted to 7.4 with NaOH. The nominally Ca²⁺-free Tyrode solution was prepared by omitting CaCl₂ from the normal Tyrode solution. The high-K⁺, low Cl solution contained (in mM) 70 glutamic acid, 15 taurine, 30 KCl, 10 NaH₂PO₄, 10 HEPES, 0.5 MgCl₂, 11 glucose, and 0.5 EGTA, and the pH was adjusted to 7.3 with KOH. The standard external bath solution was normal Tyrode solution. The internal solution contained (in mM) 100 potassium aspartate, 20 KCl, 0.02 CaCl₂, 5.0 Mg²⁺-ATP, 5.0 potassium creatine phosphate, 0.05 EGTA, and 5.0 HEPES, and the pH was adjusted to 7.25 with KOH. For measurement of I_K, 2 µM nisoldipine was added to the standard bath external solution to block I_CaL. To record the isolated I_CaL, the external bath solution contained (in mM) 140 tetraethylammonium chloride, 2.0 CaCl₂, 0.53 MgCl₂, 10 glucose, and 10 HEPES, and the pH was adjusted to 7.4 with tetraethylammonium hydroxide. The internal solution contained (in mM) 130 CsCl, 2.0 MgCl₂, 5.0 Na⁺-ATP, 20 tetraethylammonium chloride, 10 EGTA, and 10 HEPES, and the pH was adjusted to 7.25 with CsOH. All experiments were carried out at 35-36°C.

Drugs. 17-Estradiol (Sigma Chemical, St. Louis, MO) was dissolved in ethanol to give a stock solution of 50 mM. The final concentration of 17-estradiol was obtained by diluting the stock solution into the bath solution. The same amount of ethanol (1:2,000 vol/vol) was also added to normal Tyrode solution for use as the control. Nisoldipine (a gift from Bayer Pharmaceutical, Osaka, Japan) was dissolved in DMSO to give a stock solution of 10 mM.

Electrical recording. Myocytes were placed in the tissue bath on the stage of an inverted microscope (TMD, Nikon, Tokyo, Japan). Oxygenated Tyrode solution was continuously perfused through the bath at an average speed of 1.5 ml/min by gravity, and exchange of the bath solution was completed within 10 s. The whole cell configuration of the patch-clamp technique was applied to record membrane potentials and currents by using a patch-clamp amplifier (Axopatch 1C, Axon Instrument, Foster City, CA). To make the suction pipettes, borosilicate glass capillaries with inner filaments (Clark Electromedical Instruments, Pangbourne, UK) were heated and pulled by two steps with a microelectrode puller (model PA-91, Narishige, Tokyo, Japan). Resistance of a typical electrode was 2-4 M when the pipette was filled with the internal solution. At the start of each experiment the junction potential was adjusted to zero by adjusting the compensation circuit in the external bath solution; it was also checked at the end of each experiment. If the difference between the two measurements was >2 mV, the values were corrected accordingly. Membrane potential and current signals were monitored by a storage oscilloscope (model VC10, Nihon Koden, Tokyo, Japan). The stability of current amplitude in the control state was checked 5 min before drug application. The analog signals were digitized using an analog-to-digital converter (Digidata 1200, Axon Instruments) at a sampling frequency of 2 kHz and stored in a personal computer (Deskpro 4/66i, Compaq, Houston, TX) for later analysis. pCLAMP software (version 5.5.1, 6.0.4, Axon Instruments) was used to generate voltage pulse protocols, data acquisition, and analysis.

Statistical analysis. Values are means ± SE. Student's t-test for paired samples was used for statistical analysis. P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of 17-estradiol on action potential parameters of ventricular myocytes. Effects of 17-estradiol (3-30 µM) on action potentials were examined by a current-clamp mode (Fig. 1, A and B). After exposure to 3 µM 17-estradiol, no significant changes in action potential characteristics were observed for 15 min. Application of 10 µM 17-estradiol caused a significant prolongation of the APD at 20% (APD₂₀) and 90% repolarization (APD₉₀). Estradiol prolonged APD₂₀ by 16.4 ± 4.8% of the control and APD₉₀ by 25.2 ± 5.8% (n = 5, P < 0.05). Application of 30 µM 17-estradiol prolonged APD₂₀ by 46.2 ± 17.1% of the control and APD₉₀ by 63.4 ± 11.7% (n = 5, P < 0.05). The prolongation induced by 17-estradiol developed rapidly and reached a steady state 5 min after start of estradiol superfusion. Effects of estrogen were almost reversible during 10 min of washout with estrogen-free solution. After washout, APD₂₀ recovered to 102.3 ± 9.6% and APD₉₀ to 106.9 ± 7.8% of the control. The resting potential and the amplitude of the action potential were unaffected even at the highest concentration (30 µM) of 17-estradiol.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Effects of 17-estradiol on transmembrane action potential of a guinea pig ventricular myocyte. A: superimposed traces of action potential in control and in presence of 10 and 30 µM 17-estradiol. B: average values of action potential duration at 20 and 90% repolarization (APD20 and APD90) in control and in presence of 3, 10, and 30 µM 17-estradiol. C and D: effect of 30 µM 17-estradiol on APD20 and APD90, respectively, at different stimulation rates. Values are means ± SE; n = 5 in each group. * Significantly different from controls (P < 0.05).

Effects of 17-estradiol on membrane currents. To examine the effects of 17-estradiol on membrane currents, 1-s test pulses to voltages between 100 and +50 mV were applied from a holding potential of 40 mV. Figure 2 shows the results of a typical experiment. Application of 30 µM 17-estradiol had little effect on membrane currents at voltages negative to 50 mV and induced a slight decrease at 40 mV. At potentials positive to 30 mV, 17-estradiol mildly suppressed initial inward current on depolarization and decreased the late outward currents at test voltages positive to 20 mV. Results similar to those shown in Fig. 2 were confirmed in five myocytes. At 100 mV the current was 6.46 ± 0.8 pA/pF in the control and 6.33 ± 1.0 pA/pF (P = NS) after application of 30 µM 17-estradiol. At 0 mV, initial inward current was 7.10 ± 0.8 pA/pF in the control and 5.60 ± 0.9 pA/pF after 17-estradiol (P < 0.05). At the test potential of 50 mV, the late currents were 6.77 ± 1.3 pA/pF in the control and 5.32 ± 1.1 pA/pF after estradiol (P < 0.05).

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Effects of 30 µM 17-estradiol on membrane currents. Insets: protocol for A and B and for C and D. A and B: current traces induced by 1-s test pulses between 30 and +50 mV with 10-mV steps from a holding potential of 40 mV at a rate of 0.1 Hz. C and D: current traces induced by test voltages between 100 and 50 mV by 10-mV steps. E: current-voltage (I-V) relationships. Outward currents were measured at end of a 1-s test pulse. Inward currents were measured at initial peak relative to holding current level during test voltages. * Significantly different from respective controls (P < 0.05).

Effect of 17-estradiol on I_CaL. Suppression of the initial inward current on depolarization suggested a decrease in I_CaL after application of estradiol. Therefore, using the solutions described in METHODS, we examined the effects of 17-estradiol on isolated I_CaL. The currents were recorded during 200-ms test pulses between 30 and +50 mV in 10-mV steps applied at 10-s intervals after 200-ms conditioning steps to 40 mV from a holding potential of 80 mV. Figure 3A demonstrates that 30 µM 17-estradiol decreased the peak amplitude of I_CaL. The inhibition occurred rather quickly to reach a steady level within 3-5 min of 17-estradiol application. The inhibition of I_CaL was reversible after 2-5 min of washout. The peak I_CaL was decreased by 30 µM 17-estradiol to 80.1 ± 2.5% of the control (measured at 0 mV current; n = 5, P < 0.05). After washout, I_CaL recovered to 93.6 ± 2.2% of the control (n = 5, P < 0.05 vs. estradiol). Figure 3B shows the current-voltage (I-V) relationships of I_CaL in the absence and presence of 30 µM 17-estradiol and after washout. The shape of the I-V curve was not affected by estradiol.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of 30 µM 17-estradiol on L-type Ca2+ current (ICaL). A: superimposed current traces. Inset: pulse protocol. B: I-V relationships in control, in presence of 30 µM 17-estradiol, and after washout (n = 5). * Significantly different from controls (P < 0.05).

Effects of 17-estradiol on I_K. I_K is mainly composed of two components: the rapidly activating component (I_Kr) and the slowly activating component (I_Ks) (25). The estrogen-induced block development of I_K was examined by the envelope of tails test. Membrane potential was held at 40 mV and pulsed to +40 mV for a variable time, from 50 to 3,000 ms. Sufficient time (>12 s) between test pulses was allowed for full deactivation of tail currents before the application of another depolarizing pulse. In six experiments the envelope of tails test was performed in the same cell before and after application of 30 µM 17-estradiol. In the presence of 30 µM 17-estradiol, tail currents were suppressed at variable test durations (Fig. 4A). At the short pulse (50 ms), tail current was decreased by 30 µM 17-estradiol to 71.8 ± 6% (n = 6, P < 0.05) of the control, whereas at the long pulse (3,000 ms), the current was decreased to 53.3 ± 4% (n = 6, P < 0.05). To ensure inhibition of the two I_K components, we used 5 µM E-4031, which specifically blocked I_Kr. In the presence of 5 µM E-4031, 30 µM 17-estradiol also decreased the tail current. At the long pulse (3,000 ms), 17-estradiol inhibited the tail current to 60.4 ± 8% (n = 5, P < 0.05) of the control. After washout of 17-estradiol, tail I_K (I_Ktail) recovered to 81.8% (Fig. 4B). Thus 17-estradiol reversibly inhibited I_Ks.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Effect of 17-estradiol on envelope of delayed outward K+ tail current. A: current traces in control and in presence of 30 µM 17-estradiol. Inset: voltage protocol. B: current traces recorded in presence of 5 µM E-4031 (control), 30 µM 17-estradiol, and washout of 17-estradiol.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Fully activated I-V relationships for slowly activating component of delayed outward K+ current (IKs). A: superimposed current traces in control and in presence of 17-estradiol. B: fully activated I-V relationships for IKs in control and in presence of 30 µM 17-estradiol. Inset: voltage protocol. * Significantly different from controls (P < 0.05).

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Effect of 17-estradiol on IK. Current ratio is shown as a function of pulse duration. Inset: procedure used to determine ratio of tail current to time-dependent current (IKtail/IK). A: current ratio in control (n = 6), in presence of 30 µM 17-estradiol (n = 6), and in presence of 5 µM E-4031 (n = 5). Values are means ± SE. For most data points, SE bars are smaller than symbol size. B: ratio of 17-estradiol-sensitive current.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   Effect of 17-estradiol on IKtail. A: experimental protocol of envelope of tails tests with different drug treatments and tail current components with each treatment. B and C: percent inhibition of rapidly and slowly activating components of IK, respectively, by 30 µM 17-estradiol with use of protocol in A. IKE2, tail current after addition of 17-estradiol; IK(E2 + E-4031), tail current after addition of 17-estradiol + E-4031; IKcont, control tail current. Abscissa, pulse duration to elicit tail current components.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study we demonstrated that application of 10 µM 17-estradiol prolonged the APD in guinea pig ventricular myocytes. This hormone was also shown to inhibit three important currents forming the repolarization of the ventricular action potential: I_CaL, I_Kr, and I_Ks. It is thus likely that the prolongation of APD by 17-estradiol is mainly caused by the inhibition of I_Kr and I_Ks.

Electrophysiological studies exploring the direct actions of estrogen on cardiac membrane have not offered any explanation for the clinical observations of the prolonged Q-T interval observed in women compared with that in men (1, 6, 11, 21) but, rather, have presented conflicting results. Action potentials were shortened by 17-estradiol in guinea pig ventricular muscles and myocytes at a frequency of 1.0 Hz, with inhibition of I_CaL (7, 9, 10). In contrast, Berger et al. (2) observed that 17-estradiol caused prolonged APD at a frequency of 0.05-5.0 Hz by mainly inhibiting I_to in rat ventricular myocytes. Our results in guinea pig preparations agree with the observed changes in action potential in rat myocytes demonstrating APD prolongation in a reverse use dependent manner at a wide range of stimulation rates, although different currents were affected. Therefore, different results in the previous reports could not be explained by the species difference or different stimulation rates. We are unable to explain why in previous studies opposite effects on action potentials were observed in guinea pig preparations.

With regard to the effects of estrogen on membrane currents, most studies demonstrated the inhibition of I_CaL in various cardiac preparations (2, 7, 9, 10, 15, 18). These results correspond to reports of a negative inotropic effect by estrogen (20, 26). The inhibition of I_CaL has also been noted in other tissue preparations, including GH₃ cells (29), neurons (12), myometrial cells (31), and vascular smooth muscle cells (5, 15, 18). In the present experiments, application of 30 µM 17-estradiol caused a reduction of I_CaL by 20% with respect to the control without a shift in voltage dependence. Other studies also reported a 60-70% reduction of I_CaL compared with the control after application of 30 µM 17-estradiol (7, 9, 10). Because I_CaL has a tendency to decrease with time during the whole cell recording (rundown), the reduction after treatment may be somewhat overestimated. Our measurements of I_CaL returned to 94% of the control value on washout of 17-estradiol. This may indicate that the 20% reduction with respect to the control represents a real effect caused by 30 µM 17-estradiol.

The hormone affected two components of I_K, I_Kr and I_Ks, without reduction of the inward rectifier K⁺ current, while it was mildly inhibited in rat myocytes with 30 µM 17-estradiol (2). Although the late currents at potentials negative to 50 mV were not affected, the current at 40 mV was significantly decreased by 17-estradiol (Fig. 2). Because the inward currents at negative voltages to the E_rev were not depressed, we judged that the depression of the current at 40 mV was not due to the I_K1 inhibition but to the effect on I_K. The differential effects on different components of K⁺ currents exclude a nonspecific action of estrogen but indicate channel-specific action. Its effects on I_Kr and I_Ks were demonstrated by the envelope of tails test. The degree of inhibition of I_Kr and I_Ks was nearly equal, i.e., a 30-40% decrease with respect to the control. The voltage-dependent activation of both components was unaffected. The fully activated I-V relationship of I_Ks was depressed by 17-estradiol, indicating that the number of functional channels was decreased or the single-channel current amplitude was reduced. At 30 µM 17-estradiol, the maximum amplitudes of the fast and slow components of I_to in rat myocytes were decreased to 50 and 43%, respectively (2). Therefore, the degrees of inhibition of I_to and I_K appear to be similar. In previous studies the effects of 17-estradiol on K⁺ channels seemed to depend on the tissues. For example, estradiol had no significant effect on the outward K⁺ current in vascular smooth muscle cells (15, 18), whereas 17-estradiol stimulated the Ca²⁺- and voltage-activated K⁺ channels in aortic endothelial cells and coronary myocytes (24, 30).

The inward I_CaL and outward I_Kr and I_Ks are in delicate equilibrium during the plateau, and their net effects determine the repolarization phase of the action potential (4, 16). It is difficult to quantify the contribution of each current component to the action potential prolongation induced by 17-estradiol. Despite this uncertainty, the reduction of the two outward current components by estrogen was comparable to or higher than that of I_CaL. Therefore, it seems reasonable to assume that the former effects can overcome the shortening effect by the latter to prolong APD.

The concentration of 17-estradiol (30 µM) that was found to cause inhibition of I_CaL, I_Kr, and I_Ks is much higher than the in vivo concentration of 17-estradiol. Normal plasma concentrations of 17-estradiol have been shown to be <10 nM in various species, including the guinea pig (14). Maximal plasma concentrations in humans are 0.14 nM in men and 1.4 nM in women during the preovulatory period. During pregnancy the 17-estradiol concentration increases up to a maximum of 0.1 µM by the end of the third trimester (23). Recent evidence has indicated that the acute effective concentration of steroid hormone accumulated by target cells may far exceed plasma levels (17). Nearly all circulating estrogen (95-98%) is bound to plasma proteins, i.e., albumin and sex hormone-binding globulin. Interestingly, accumulation of estrogen in target cells is greater in the presence of sex hormone-binding globulin than in the presence of free hormone alone. Although the physiological solutions used in the present and previous experiments did not contain plasma proteins, micromolar range concentrations of estrogen are often required to produce consistent and maximal responses of target cells in vitro. Further experiments are necessary to determine the effective steroid concentrations accumulated by target cells, including cardiac cells in vivo.

Action potential prolongation by estrogen was observed in myocytes from males as well as from females. Although the nuclear estrogen receptor was not expressed in rat ventricle of either gender (28), estrogen affected membrane currents of ventricular myocytes (2). Therefore, 17-estradiol seems to exert its effect in rat ventricle via a nongenomic pathway. In other species, including the guinea pig, the estrogen membrane receptor has not been identified in cardiac tissue. In our study, 17-estradiol prolonged APD and inhibited I_CaL, I_Kr, and I_Ks within 5 min. This rather rapid effect of 17-estradiol and the above features are inconsistent with its action being mediated via conventional slow-acting nuclear receptors. Recently, Meyer et al. (13) reported that the reduction of I_CaL by 17-estradiol developed with a time constant of 3-4 s. This is consistent with the presence of a cell surface receptor that could also affect I_K. It is not clear from the present study how the inhibitory effects of 17-estradiol are exerted on the three channels, and further studies are necessary to prove the mechanism of action on the different membrane currents.

Our results show that 17-estradiol prolonged APD by inhibition of I_K, which may contribute to the high prevalence of the incidence of torsades de pointes with Q-T interval prolongation. This result may not necessarily indicate that estrogen always exhibits proarrhythmic potential. In the previous as well as the present studies, estrogen has been shown to reduce L-type Ca²⁺ channel activity in vitro and to cause relaxation in arterial smooth muscles and cardiac myocytes (7, 9, 15, 18). Clinical observations have demonstrated that estrogen replacement therapy is associated with a reduced incidence of cardiac arrhythmias in postmenopausal women (3), and cyclic increases in estrogen at the premenopausal stage abolish the appearance of supraventricular tachycardia (22). Therefore, estrogen may exhibit proarrhythmic as well as antiarrhythmic effects, depending on the clinical situation. Large-scale prospective studies to investigate the clinical effects of estrogen are necessary to clarify this problem.