INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

IN CARDIAC CELLS, direct measurements of intracellular Cl activity ([Cl]_i) using ion-selective microelectrodes have shown that [Cl]_i has a range of 10-30 mM, which is about three times higher than would be expected if the ions were distributed passively according to the membrane potential (V_m) (3, 5, 33, 36, 37). Because of this nonpassive distribution of Cl, the Cl equilibrium potential (E_Cl) in many areas is more positive than the resting membrane potential (15). To maintain this higher level of [Cl]_i, it was suggested that Cl/HCO₃ exchange would be a main mechanism in Purkinje fibers in the steady state in quiescent cardiac cells (36, 37), whereas activation of the Na⁺-K⁺-Cl cotransport was also proposed as playing an important role in maintaining the level of [Cl]_i in cultured chick heart (2, 24) and rat and rabbit ventricular cells (7, 12).

During cardiac ischemia, [Cl] homeostasis may play a role in maintaining the resting membrane potential or regulating action potential firing. In fact, recent studies showed that modulation of anion homeostasis by substitution of extracellular Cl with equimolar NO₃ prevented ischemia- and reperfusion-induced ventricular fibrillation in rats and rabbits in vitro (30). The antiarrhythmic activity of an external Cl-free environment was thought to be a consequence of alterations in membrane permeability of the anion (10, 11). Furthermore, Cl blockers such as anthracene-9-carboxylic acid and SITS were shown to exert protective effects against myocardial ischemia-reperfusion damage in coronary-perfused guinea pig ventricular preparations (34). Our previous work (23) using pH-selective microelectrodes to measure intracellular pH (pH_i) in ischemic guinea pig ventricular muscles also showed that the application of SITS and DIDS suppressed the rapid onset of ischemia-induced intracellular acidosis and delayed the onset of ischemia-induced deterioration of action potentials in cardiac muscles. Under external Cl-free conditions, the time to cessation of action potentials caused by ischemia was significantly delayed, and the development of intracellular acidosis during ischemia was attenuated. These results indicate that activation of the Cl/HCO₃ exchange system may result in countertransport of intracellular HCO₃ and extracellular Cl and, therefore, is partly involved in the development of intracellular acidosis during the early phase of cardiac ischemia (23).

One can, therefore, presume that ischemia might affect the homeostasis of [Cl]_i and result in an accumulation of Cl within the intracellular space. Although it has been reported that ischemia might have induced a time-dependent change in E_Cl during global ischemia in perfused rat heart (29), it is still not clear whether ischemia induces any changes in [Cl]_i, and the relative contribution of intracellular Cl to ischemia and reperfusion injury has not yet been fully investigated. To answer this question, direct measurements of the [Cl]_i in ischemic hearts are essential. In the present study, we have successfully utilized the doubled-barreled Cl-selective microelectrode technique to measure [Cl]_i in ischemic guinea pig ventricular papillary muscle in vitro. Our results provide direct evidence that ischemia induces a dramatic increase in [Cl]_i in guinea pig ventricular papillary muscle and suggest that the activation of the Cl/HCO₃ exchanger plays a significant role in accumulation of [Cl]_i during early-phase ischemia.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Preparation and solutions. Papillary muscles of guinea pig (250-300 g) right ventricle were used in the present experiments. The heart was taken from the animal and rapidly immersed in cold Tyrode solution, and papillary muscles 3-5 mm in length and ~0.5 mm in diameter were excised and mounted in a flow chamber (a bath chamber of ~2-ml volume). The NaHCO₃-buffered Tyrode solution equilibrated with a 95%O₂-5%CO₂ gas mixture was introduced to the flow chamber continuously at ~5 ml/min. The NaHCO₃-buffered Tyrode solution contained (in mM) 127 NaCl, 4 KCl, 0.25 MgCl₂, 1.4 CaCl₂, and 5.5 glucose and was buffered with 20 mM NaHCO₃. A Cl-free solution was made by replacing Cl with equimolar gluconate (8, 9, 31). The pH of the NaHCO₃-buffered solution was adjusted to 7.4 by addition of HCl, and the pH of the HCO₃-buffered Cl-free solutions was adjusted to 7.4 by adding acetic acid. The temperature of the solutions was monitored using a thermocouple-type thermoprobe (PTC-201, Unique Medical, Tokyo, Japan) placed in the bath chamber and was controlled at 36 ± 0.5°C. The preparations were equilibrated without stimulation for 1 h before the start of the experiments.

Experimental procedure. To see the effects of simulated ischemia on [Cl]_i in guinea pig ventricular muscle, double-barreled Cl-selective microelectrodes were used to monitor [Cl]_i and V_m. Six preparations obtained from different animals were impaled with double-barreled Cl-selective microelectrodes for 15 min to record control levels of [Cl]_i and V_m. The muscles were then subjected to simulated ischemia for 15 min followed by reperfusion for 25 min. To see the effects of SITS or Cl-free conditions on ischemia-induced changes in [Cl]_i, five preparations and six other muscles were treated with 0.5 mM SITS or Cl-free solution, respectively, and received the same procedures as described above.

Simulated ischemia of guinea pig ventricular papillary muscle. Isolated guinea pig papillary muscles were superfused in vitro at 36°C, and, at regular time intervals, were subjected to simulated ischemia as described in previous studies (23, 35). Briefly, the ischemic situation was mimicked by first arresting the normal superfusion and removing the Tyrode solution from the experimental chamber, whereby a thick layer of prewarmed paraffin oil (36°C, ~1.5 ml), which had been present on top of the superfusate, was lowered and settled around the whole muscle. In this way, all ions and metabolites were trapped within the interstitial space and within the thin, stagnant layer of Tyrode solution (0.3-0.5 ml of volume) around the muscle. The thickness of the Tyrode film, which was determined as described by Vanheel et al. (35), was ~70 µm (Fig. 1A). Cooling was prevented by pumping warm water (36°C) through a 0.2- to 0.3-mm-outer diameter polyethylene tube around the muscle.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   A: model of simulated ischemia in isolated guinea pig ventricular papillary muscles. Amp., high-input impedance amplifier (WPI FD223-D); Stim. electrode, stimulus electrodes. B: response of a double-barreled Cl-selective electrode to different Cl concentrations at 0.5, 5, 50, or 142 mM. The cationic concentration was adjusted to 142 mM. C: calibration curve of Cl electrode after Cl was replaced by gluconate (data were obtained from B). VCl, Cl ion output (mV); Vm, membrane potential (mV); [Cl], Cl concentration (mM).

Double-barreled ion-selective microelectrodes. The double-barreled ion-selective microelectrodes were constructed as described previously (22, 23), using liquid sensor cocktails (Cl ionophore I cocktail A, model 24902, Fluka Chemika-Biochemika, Buchs, Switzerland) for Cl. Because reports showed that Cl electrodes are insensitive to gluconate (9), gluconate was used as the replacement for Cl in the perfu- sion solution. Each double-barreled Cl-selective electrode was calibrated in Tyrode solution or 142 mM NaCl solution in which NaCl was replaced by sodium gluconate to become a set of four solutions (NaCl + Na gluconate) (21). The concentration of Cl in the solutions was 0.5, 5, 50, or 142 mM, and the cationic concentration was adjusted to 142 mM (Fig. 1, B and C). In each Cl electrode, the response to the above test solutions showed a sensitivity of >55 mV/decade in the concentration change between 5 and 0.5 mM or 5 and 50 mM of Cl. The Cl electrodes were not sensitive to bicarbonate or acetic acid used in the present study.

Drugs and statistics. All salts used to prepare solutions and SITS were obtained from Sigma (St. Louis, MO) and dissolved directly in Tyrode solution at the desired concentration. Data were analyzed by basic statistical methods including ANOVA followed by paired-comparison or repeated-measure tests. Data are expressed as arithmetic means ± SE, and significance was established at P < 0.05.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effects of ischemia on [Cl]_i with or without SITS. To determine [Cl]_i, we measured [Cl]_i in six impalements (n = 6 muscles from 6 animals) in quiescent guinea pig ventricular papillary muscle superfused with NaHCO₃-buffered Tyrode solution. The average [Cl]_i was 18.7 ± 3.5 mM, and the value was similar to the levels reported in previous studies in cat papillary (6), rat ventricle (28), and sheep Purkinje fibers (36, 37). In six nonstimulated guinea pig ventricular papillary muscles, we measured [Cl]_i during simulated ischemia produced by stopping perfusion and immersing ventricular muscle in warmed paraffin oil. Figure 2 shows a representative example from a guinea pig ventricular papillary muscle subjected to simulated ischemia and reperfusion. The top trace in Fig. 2A shows changes of [Cl]_i, and the bottom trace shows V_m recorded with the reference electrode. Before the impalement of the electrode into the cell, the potential recorded with the reference electrode was set to 0 mV, and the extracellular Cl concentration was later calculated based on the calibration data. When the electrode impaled the cell, V_m was about 85 mV and [Cl]_i became steady at about 17 mM. After being recording for another 15 min as the control, the muscle was subjected to simulated ischemia for 15 min followed by reperfusion for 25 min. Ischemia induced a marked increase in [Cl]_i and progressive membrane depolarization. At 5 min after the onset of ischemia, V_m became about 55 mV and [Cl]_i became steady (45 mM) and remained at the same level for 10 min (peak of [Cl]_i was 54 mM). After reperfusion was started, [Cl]_i and V_m recovered gradually to control levels. Successful measurements of [Cl]_i were also obtained in five other cells of isolated guinea pig ventricular papillary muscle (from 5 animals) subjected to simulated ischemia. The peak value of [Cl]_i during ischemia was 55.3 ± 2.5 mM, which was significantly higher than the control value.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effects of ischemia on intracellular Cl activity ([Cl]i) and Vm with or without SITS (0.5 mM) in guinea pig ventricular papillary muscle. A: representative example of a recording for [Cl]i in guinea pig ventricular muscle subjected to simulated ischemia. Dashed line indicates control level of [Cl]i. Open bars, time course of ischemia; horizontal solid line bar, application of 0.5 mM SITS. Pretreatment with 0.5 mM SITS delayed [Cl]i elevation and depolarization of Vm induced by ischemia. B: effects of SITS on [Cl]i elevation induced by ischemia. , Response of [Cl]i to simulated ischemia in Tyrode solution without SITS; , changes of [Cl]i during simulated ischemia with SITS pretreatment.

View this table: [in this window] [in a new window]   Table 1.   Effects of 0.5 mM SITS on ischemia-induced changes in [Cl]i and membrane potentials in guinea pig ventricular muscles

Effects of external Cl-free conditions on [Cl]_i during ischemia. The effect of external Cl-free conditions on ischemia-induced elevation of [Cl]_i was examined in six preparations obtained from different animals. Gluconate was used as a Cl substitute because it is barely sensed by the Cl electrode (9). Figure 3A shows the changes in [Cl]_i in external Cl-free conditions during continuous impalement with the Cl-selective electrode. In external Cl-free conditions (Fig. 3A), [Cl]_i gradually decreased until it reached a stable level (~12 mM) after 20 min of perfusion of Cl-free solution. The papillary muscle was then exposed to paraffin oil. During this period, simulated ischemia produced only a slight, transient increase of [Cl]_i ([Cl]_i = 18.5 mM). The results of similar experiments are summarized in Fig. 3B. In Cl-free conditions, the [Cl]_i peak was, on average, 7.3 ± 0.4 mM before ischemia, and 20.3 ± 1.7 mM during ischemia (n = 6 cells from 6 animals). The increment of [Cl]_i induced by ischemia was significantly suppressed by the use of Cl-free solution.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Effects of external Cl-free conditions on [Cl]i in ischemia. A: effects of Cl-free conditions on [Cl]i during ischemia. [Cl]i in control was ~18 mM. After superfusate was switched to a Cl-free solution, [Cl]i gradually began to decrease to ~12 mM at 20 min of exposure. Preparation was then immersed in paraffin oil. Elevation of [Cl]i induced by simulated ischemia was significantly suppressed. [Cl]o, extracellular [Cl]. B: changes of [Cl]i during ischemia in normal-Cl conditions and in external Cl-free conditions; n, no. of muscles used. Open bars, control; hatched bars, values obtained during ischemia. Under Cl-free conditions, [Cl]i increase during ischemia was suppressed. * Significant differences from value obtained at control (P < 0.05); # significant differences between normal-Cl conditions and Cl-free conditions (P < 0.05).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The model of simulated ischemia used in the present study was first described by Vanheel et al. (35). An ischemic condition was induced by arresting the normal superfusion and immersing the preparation in mineral oil, thus limiting the substrate supply to the interstitial fluid and affecting the clearance of metabolites. The model is therefore considered to be similar to situations of myocardial ischemia in situ, which most commonly arise from a partial restriction of the coronary blood flow, leading to a decrease in the O₂ supply and impairment of the removal of metabolites such as lactate, CO₂, and protons. Furthermore, this method allows for direct microelectrode measurements of cardiac intracellular H⁺ and other ions by stable cell impalements as described previously (23). It has been reported that the simulated ischemia made by this method in previous studies exerted a greater effect on contractility, action potentials, and intracellular pH than did hypoxic superfusion (35). Our earlier studies (23) also showed that the depolarization in V_m and shortening of action potential duration induced by ischemia were in agreement with previous results obtained in global ischemia for guinea pig (19), pig (20), and dog (18) heart. Even though the intracellular acidification induced by simulated ischemia (see Ref. 23) was greater than that reported by previous authors, the magnitude of intracellular acidification observed in our study was within the range reported for ischemic myocardium, from normal values of ~7.2 to values as low as 6.3 (26) or even 5.7 (17). Thus, although we are aware of the technical limitations of the present method in simulating in situ myocardial ischemia, the method adopted here is considered to reflect, at least in part, some aspects of the pathophysiological changes that occur in cardiac ischemia in situ.

The main goal of the present study was to investigate whether ischemia induces any change in [Cl]_i in guinea pig ventricular muscle. To our knowledge, this study is the first to measure [Cl]_i directly in ischemic ventricular muscles. Our results showed that [Cl]_i markedly increased during ischemia and that this increase of [Cl]_i was suppressed by both external Cl-free conditions and the presence of SITS.

The mechanisms involving the [Cl]_i increment in ventricular muscle cells during ischemia are complex and unclear. However, the concentration of SITS used in the present study (0.5 mM) is known to completely inhibit the Cl/HCO₃ exchanger in ventricular cells (36-38) and red blood cells (25). The suppressing effect of SITS on the amplitude of ischemia-induced [Cl]_i elevation at the earlier period of ischemia, therefore, may be a result of its ability to block the Cl/HCO₃ exchanger in ventricular cells. The suppression during the early phase of ischemia (5 min after induction of ischemia) was ~45% ([Cl]_i in the control was ~48 mM at 5 min after inducing ischemia and 26 mM in the presence of SITS), similar to that in a previous study by Ramasamy et al. (29). In their study using ¹⁹F NMR to measure Cl potentials in perfused rat hearts, they found that SITS inhibited ~48 ± 3% of trifluoroacetate uptake, which is thought to reflect the E_Cl. Furthermore, in our previous study, it is interesting that SITS at the same concentration (0.5 mM) also inhibited ~46% of ischemia-induced acidosis at 5 min after ischemia was induced (see Fig. 6 in Ref. 23). These results suggest that a certain amount of H⁺ is continuously increased at an intracellular site via activation of the Cl/HCO₃ exchanger, resulting in an increase in [Cl]_i during the early phase of ischemia. However, when ischemia is prolonged, the Cl/HCO₃ exchanger would become inactivated by acidosis, as has been reported in previous studies. This explains why SITS only suppressed the rate of increase of the [Cl]_i during the initial phase of ischemia but did not affect the peak magnitude of [Cl]_i elevation. On the other hand, the increase in [Cl]_i induced by ischemia became SITS insensitive when ischemia was sustained >10 min. There are, however, some possible mechanisms to account for the SITS-insensitive [Cl]_i increase during ischemia, such as Na⁺-K⁺-Cl cotransport and others that were proposed to explain the SITS-insensitive [Cl]_i increases observed in normal rat cardiac ventricular cells (7) or in response to elevation of extracellular K⁺ concentration (27). However, the mechanism contributing to those SITS-insensitive [Cl]_i increases during simulated ischemia in the present study is still not clear, and additional experiments are required.

Another characteristic of the ischemia-induced elevation in [Cl]_i was its dependence on external Cl. In light of this, in addition to stimulation of anion transports by ischemia, the activation of Cl channels should also be taken into consideration. In fact, there have been reports of ischemia-induced myocardial swelling and accelerated repolarization by activation of Cl channels (1). Recent electrophysiological and molecular studies have also provided evidence, suggesting that as many as six different Cl conductances can be identified in the sarcolemma of cardiac myocytes (16). However, because the E_Cl is known to show a time-dependent change more positive than expected for passive distribution during ischemic conditions (29), activation of Cl channels may be not be involved in Cl influx but rather in Cl efflux to maintain Cl homeostasis.

A mechanism for maintaining high levels of [Cl]_i during ischemia may involve the release of Cl from sites within the intracellular space, because an external Cl-free solution did not completely block the increase in [Cl]_i. Even though the magnitude of the elevated [Cl]_i was significantly smaller than that in normal Tyrode solution and the increase was transient, [Cl]_i still significantly increased during ischemia. One explanation for this phenomenon is that ischemia and other pathological conditions may stimulate the release of intracellular Cl from intracellular compartments or intracellular stores of Cl such as those reported in isolated sheep heart mitochondria (14) and other tissues (4, 13, 32). If this is true, the external Cl-independent transient increase in [Cl]_i (~38.5%) may be an important contributing factor in the ischemia-induced increase in [Cl]_i in cardiac cells. In fact, the mitochondrial metabolism decreases when preparations are exposed to simulated ischemia, and the increase of [Cl]_i during ischemia in Cl-free conditions is similar to metabolic inhibitor-induced releases of Cl from intracellular Cl in rat lactotroph cells (13). The transient increase in [Cl]_i in Cl-free conditions may indicate a gradual depletion of the intracellular Cl reserve.

Although we found that the [Cl]_i in perfused muscles stimulated with 1 Hz was similar to the value obtained from quiescent muscles (data not shown), we could not obtain successful impalements for recording [Cl]_i in stimulated muscles during ischemia. Because of the technical limitations of the present method, we could not explore the possible difference in response between quiescent and contracting muscles. However, our recent studies using conventional and ion-selective microelectrodes showed that SITS and the external Cl-free solution not only suppressed the development of intracellular acidosis in noncontracting muscles but also delayed the onset of ischemia-induced deterioration of action potentials and prolonged the time to cessation of action potentials (23). This result implies that SITS and Cl-free conditions would have effects on stimulated preparations similar to those on quiescent muscles during ischemia. The activation of the Cl/HCO₃ exchanger would contribute, in part, to the acidification of pH_i and the accumulation of [Cl]_i during early-phase ischemia. On the other hand, the present study showed that 0.5 mM SITS significantly suppressed the ischemia-induced depolarization during ischemia. This result is similar to our previous data obtained by conventional microelectrodes (see Fig. 4 in Ref. 23) and suggests that inhibition of the Cl/HCO₃ exchanger by SITS may allow the modulation of V_m of ischemic ventricular muscles. If the attenuation of depolarization in V_m by SITS is related to its suppression of ischemia-induced elevation of [Cl]_i, the application of stilbene derivatives would be effective to prevent the ischemia-induced deterioration of the V_m by its decrease of [Cl]_i. However, at this moment the ionic mechanism involved is still unclear, and further investigations are necessary.

In conclusion, this is the first report to show that simulated ischemia induces a marked increase of [Cl]_i in the quiescent guinea pig ventricular papillary muscle with direct measurement of [Cl]_i using double-barreled ion-selective microelectrodes. Our results provide supporting evidence for activation of the Cl/HCO₃ exchanger involved in elevation of [Cl]_i during ischemia.