Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

ACE inhibitors have now been widely shown to be clinically beneficial in a number of cardiac conditions, including patients with asymptomatic left ventricular dysfunction following acute myocardial infarction (Pfeffer et al., 1992). It seems likely that a component of the beneficial effects of ACE inhibitors on the myocardium are indirect, resulting from the inhibition of the deleterious peripheral vasoconstrictive effects of A-II, which increase afterload. However, there is now mounting evidence that A-II may exert direct effects on the myocardium.

Recent work from our laboratory (Ikenouchi et al., 1994) has demonstrated that A-II induces an intracellular alkalosis in adult rabbit myocytes, and that this effect is a consequence of stimulation of Na⁺/H⁺ exchange (Matsui et al., 1995). Intracellular acidification develops during ischemia/hypoxia, and increased Na⁺ loading as a consequence of Na⁺/H⁺ exchange appears causally related to ischemic/hypoxic injury in association with Ca⁺⁺ overload (Tani and Neely, 1990; Renlund et al., 1984; Haigney et al., 1992; Barry, 1991). We suspected that A-II may therefore directly enhance ischemic/hypoxic myocardial injury by augmenting Na⁺ overloading (via increased Na⁺/H⁺ exchange) which would then lead to Ca⁺⁺ overloading (via reverse sarcolemmal Na⁺/Ca⁺⁺ exchange). Our study was designed to test this hypothesis, in resting isolated adult rabbit ventricular myocytes. This preparation allows examination of the effects of A-II on cell injury and [Ca⁺⁺]_i without the influence of vascular effects, the positive inotropic effect of A-II which could affect energy metabolism of myocytes or alterations in loading conditions that might complicate interpretation of the results. We demonstrate that A-II increases [Ca⁺⁺]_i and enhances myocyte contracture caused by CN exposure. These effects are augmented by partial inhibition of the Na⁺ pump and reversed by Na⁺/H⁺ exchange inhibitors.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Myocyte preparation. Adult rabbit ventricular myocyte isolation was performed by a modification of previously reported method (Ikenouchi et al., 1994). Briefly, hearts were removed from albino rabbits (2-3 kg) anesthetized with sodium pentobarbital (65 mg/kg, i.v.). The heart was immediately attached to an aortic cannula, and continuous retrograde coronary artery perfusion at 37°C by a pump (Masterflex, Cole-Parmer Instrument Co., Chicago, IL) was initiated at a coronary perfusion pressure of 60 mm Hg. The heart was first perfused with nominally Ca⁺⁺-free modified MKRBB solution for 5 min, immediately followed by 20 min of recirculating perfusion with the same solutions containing 0.28 mg/ml collagenase (class II, Worthington Biochemical, Freehold, NJ), 0.4 mg/ml hyaluronidase (type I-S, Sigma Chemical Co., St. Louis, MO) and 50 µM CaCl₂. The heart was then detached from the cannula, and the left ventricle was minced and transferred to a 50-ml conical tube with the same solution containing 0.28 mg/ml collagenase, 2 mg/dl trypsin and 50 µM Ca⁺⁺ and incubated for 10 min for further digestion. The minced tissues were continuously agitated by gassing the solution with 5% CO₂ and 95% O₂ to help release isolated myocytes. The resulting suspension was transferred to another conical tube with the same volume of the same cell isolation solution containing 2.4 mg/dl of trypsin inhibitor (Sigma) and 12% heat-inactivated fetal calf serum. The cell suspension solution was centrifuged at 300 rpm for 5 min. The supernatant was discarded and cells were resuspended in solution with higher Ca⁺⁺ concentration and incubated 15 min in the CO₂ incubator to settle down cells. The same procedure was repeated twice to slowly bring up Ca⁺⁺ concentration (200 then 1000 µM). Calcium step up solutions were made up from MKRBB with 2% albumin, 50 µM CaCl₂ and 1 mg/dl insulin mixed with the appropriate amount of MEM (Gibco Laboratories, Grand Island, NY). The cells were then suspended in MEM containing 2% albumin to decrease cell-to-cell adherence.

Solutions. For flow cytometric analysis, a HEPES-buffered balanced salt solution was used containing (in mM): NaCl 126, KCl 4.4 or 0.8, CaCl₂ 1.08, MgCl₂ 0.5, HEPES 24 (pH 7.4, adjusted with NaOH) and glucose 5. For myocyte contracture experiments, bicarbonate-buffered Tyrode's solution equilibrated with 95% air-5% CO₂ (pH 7.4, 37°C) was used containing (in mM): 126 NaCl, 4.4 KCl, 1.0 MgCl₂, 0.9 CaCl₂, 18 NaHCO₃, 5 glucose. For experiments using low K⁺, 0.8 mM KCl was substituted for 4.4 mM KCl. For metabolic inhibition, 2 mM CN or 2 mM CN + 20 mM 2DG, was added to the above solutions and glucose was removed. pH was adjusted to 7.4 with HCl. To these solutions, one or more of the following were added to make the respective study solutions: A-II (100 nM), amiloride (1 mM), EIPA (10 µM) (Sigma). Amiloride and EIPA were first dissolved into DMSO before being added into solution. The final concentration of DMSO was always less than 0.5%.

Assessment of [Ca⁺⁺]_i. Intracellular calcium measurements were made with the Ca⁺⁺ sensitive fluorescent probe, fluo-3 (Molecular Probes, Eugene, OR), using a flow cytometer (FACScan, Becton-Dickenson). The cells were exposed to 4 or 10 µM fluo-3 for 30 min at room temperature. After 30 to 60 min of wash, the cells from a dissociation were separated into aliquots and exposed to their respective solutions for 90 min at 37°C. Propidium iodide (25 µM, Molecular Probes), was added just before data acquisition. PI is an impermeant ion that is fluorescent when bound to DNA, and is therefore a marker for nonviability. During flow cytometry, the cells were exposed to an argon laser (flow cell 0.18 × 0.43 × 2.2 mm, excitation wavelength, 488 nm). Side and forward scattering characteristics were used to separate individual cells from cell clumps and debris. Approximately 10⁴ myocytes were then analyzed for emission fluorescence intensity in each sample over 1 to 5 min. Data were collected for emission intensity at wavelengths of 530 nm for fluo-3 and 670 nm for PI and plotted simultaneously. Only those cells with the lowest fluorescent intensity at 670 nm (propidium iodide "negative" or viable cells) were included in the comparative analysis of [Ca⁺⁺]_i (fluo-3 fluorescent intensity, 530 nm). In most experiments, the fluo-3 fluorescence intensity was not calibrated, but was recorded as arbitrary units above background (unloaded cell) fluorescence intensity. In some experiments, calibrated [Ca⁺⁺]_i values were measured by exposing cells to 10 µM ionomycin in the presence of 10 mM MnCl₂, as recently described (Yao et al., 1997). Average Fmax, or Ca⁺⁺-saturated fluorescence, was estimated as 5× FMn, and average Fmin, fluorescence in the absence of Ca⁺⁺, as 1/40 Fmax as described by Kao et al. (1989). Average [Ca⁺⁺]_i was then calculated as [Ca⁺⁺]_i = K_D (F  Fmin/Fmax  F) using a value of 864 nM for K_d (Merritt et al., 1990), after all F values were corrected for background fluorescence. Probenecid 0.5 mM was present in loading, wash and protocol solutions to prevent loss of fluo-3 via the anion transporter (DeVirgilio et al., 1988). Fluo-3 fluorescence is relatively insensitive to intracellular pH in the range of 6.6 to 7.4 (Eberhard and Erne, 1989; Kao et al., 1989).

Assessment of cell contracture. The severity of cell injury was assessed by microscopic examination of cell morphology. Cells were plated on coverslips with cell adhesive (Laminin, Sigma) and were incubated for about 1 hr in a CO₂ incubator before experimental manipulation. On every coverslip, three small circled areas were marked, which allowed us to follow the same cells in the circles under the microscope periodically. To change media for metabolic inhibition, coverslips were taken out from petri dishes and then put into other petri dishes that contained a washing solution (substrate-free Tyrode's solution). Then coverslips were taken out from the petri dishes again and put into other petri dishes containing substrate-free Tyrode's solution with CN. During this procedure, most of the dead cells were washed away. Therefore, more than 95% of the cells attached to the coverslips at the start of metabolic inhibition had a rod-shaped morphology. The dishes containing cells were placed in a CO₂ incubator at 37°C during metabolic inhibition and taken out of the incubator briefly for counting rod-shaped cell number every 30 min. Cells were classified as either rod-shaped cells (length/width >3), or contracted cells (length/width <3). Loss of normal rod shape morphology was used as an index of myocyte injury.

Measurement of the degree of inhibition of Na/Ca exchange. The exchange current was measured by means of a whole-cell voltage clamp technique (Chin et al., 1993). Myocytes were voltage clamped with single suction pipettes and a discontinuous voltage clamp circuit (Axopatch 200A, Axon Instruments Inc., Foster City, CA). Pipettes were made from borosilicate glass tubing (Corning 7052, 1.65 mm o.d., 1.2 mm i.d., A-M Systems Inc., Everett, WA) and had initial resistances of 1.5 to 3 M. The cells were held at a potential of 40 mV. The pipette contained (mM) NaCl 20, MgCl₂ 0.3, EGTA 14.0, MgATP 3.0, dextrose 5.5 and HEPES 10. Calcium (3.9 mM) was added as H₂CaEGTA. The free Ca⁺⁺ was estimated to be 100 nM. The solution pH was adjusted to 7.1 with CsOH. CsCl was added to give a final Cs⁺ concentration of 130 mM. Voltage-clamped cells were superfused in a microstream containing (mM) NaCl 138.0, MgCl₂ 1.0, CaCl₂ 1.0, dextrose 11.0 and HEPES 12. The pH was adjusted to 7.4 with NaOH which gives a final Na⁺ concentration of 145 mM. Outward exchange current was activated when the cell was abruptly exposed to an adjacent microstream of solution in which Li⁺ replaced Na⁺. These rapid solution changes were accomplished with a modified version of the switching device whose characteristics and design have previously been described (Yao et al., 1997). For these experiments the two adjacent microstreams simultaneously flowed from two square glass tubes (200 µm) separated by a 70-µm glass septum. Currents were measured in the presence and absence of amiloride 1 mM, and EIPA 10 µM.

Statistical analysis. All values are expressed as means ± S.E.M. A paired t test was used to assess significance between the intervention groups. For contracture studies, the test was performed at the last time point. Thus there was no need to adjust for multiple comparisons, because only one time point was used.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Flow cytometry analysis of [Ca⁺⁺]_i and viability. Use of flow cytometry for measurement of [Ca⁺⁺]_i and viability in adult ventricular myocytes has not been previously reported. We therefore devoted some effort to the validation of these measurements. PI proved to be a reliable marker for cell viability. In preliminary experiments with an epifluorescence microscope, we observed that essentially all freshly dissociated rod-shaped cells excluded PI, and all "balled up" cells showed PI fluorescence. As shown in figure 1, we demonstrated a very good correlation between percent PI negative cells assessed by flow cytometry, and the percent rod-shaped (presumably viable) myocytes by manual microscopic count of freshly dissociated cells from nine different dissociations.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   The % PI positivity and the % rod cells were compared in samples of cells from nine dissociations. There was a high degree of correlation between flow cytometric analysis of viability by PI fluorescence and viability by cell morphology in control cell populations.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   A, Auto fluorescence at 530 and 670 nm of unloaded cells. B, There was no significant change relative to auto fluorescence at 530 nm in PI-only loaded cells. C, Fluorescence was detected at 670 nm in fluo-3 loaded cells, indicating some "spillover" of fluo-3 fluorescence into the PI detector. See text for details.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Examples of a flow cytometry dot plot of PI and fluo-3 fluorescence intensities in control cells (left), and in cells exposed to 2 mM CN and 0 glucose for 90 min (right). A line separates the PI negative (viable) cells from the PI positive (nonviable) cells. The line is skewed to account for detection of higher intensity fluo-3 fluorescence by the PI channel (670 nm). Note that [Ca++]i (fluo-3 fluorescence) increased in viable cells during CN exposure, but that % viability was not decreased.

Effects of A-II and partial inhibition of the Na pump on fluo-3 fluorescence during CN exposure. We next examined whether stimulation of Na⁺/H⁺ exchange by A-II might increase cytosolic calcium during CN exposure in adult rabbit ventricular myocytes. Na⁺ pump function is impaired in intact myocardium during ischemia, and impairment of the Na⁺ pump might be expected to enhance any effect of stimulation of Na⁺/H⁺ exchange by causing a resultant greater increase in intracellular Na, and therefore more marked alteration of Na⁺/Ca⁺⁺ exchange. However, in isolated myocytes, exposure to CN does not rapidly or completely inhibit the Na pump (Hasin and Barry, 1984). Therefore, we investigated the effects of partial inhibition of the Na⁺ pump. In preliminary experiments, we used voltage clamp techniques (Shattock and Matsuura, 1993) to quantitate the Na⁺ pump current. We found that exposure to K_o⁺ = 0.8 mM decreased the pump current by approximately 50%. Figure 4 shows an example of the effects of exposure to CN on fluo-3 fluorescence in the presence of a normal (4.4 nM) K_o⁺ and in the presence of impaired function of the Na⁺ pump produced by 0.8 mM K_o⁺. A-II induced a slight increase in fluo-3 fluorescence during CN exposure. Exposure to 0.8 mM K caused a slight increase in resting fluo-3 fluorescence and exacerbated both the rise in fluo-3 fluorescence caused by CN exposure, and the increment in this increase associated with simultaneous exposure to 100 mM A-II. Figure 5A shows normalized average values from eight experiments. Exposure to A-II caused a statistically significant increase in fluo-3 fluorescence in myocytes inhibited in the presence of partial inhibition of the Na⁺ pump. As shown in figure 5B, average cell viability was not affected by MI, or MI plus A-II in the presence or absence of partial pump inhibition.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Example of effects of angiotensin II (Ang II) on [Ca++]i in PI negative (viable) cells during exposure to CN, in the presence of normal Ko+, and during partial inhibition of the Na+ pump by reduction of Ko+ to 0.8 mM. The mean fluo-3 fluorescence intensity values are indicated to the right of each fluorescence distribution plot.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   A, This histogram demonstrates the average changes in fluo-3 fluorescence during exposure to CN (MI) and the effects of exposure to Angiotensin II (Ang II), in normal Ko+ and 0.8 mM Ko+. Ang II significantly increased fluo-3 fluorescence during MI in the presence of partial Na pump inhibition. Values are normalized, with the fluorescence in CN and normal K = 100. B shows there was no change in viability under these conditions.

View larger version (10K): [in this window] [in a new window]   Fig. 6.   This histogram demonstrates the effects of CN and 2DG (MI) on calibrated [Ca++]i, and the effects of 400 nM Ang II and 10 µM EIPA. Bars indicate means ± S.E.M. for n = 6 experiments Ang II increased [Ca++]i (P = .02) and EIPA decreased [Ca++]i (P = .02) during MI. In the presence of EIPA Ang II did not affect [Ca++]i during MI.

Effects of A-II on cell contracture caused by metabolic inhibition exposure. To determine if effects of A-II during metabolic inhibition are of physiological significance, we examined alterations in cell morphology. Figure 7A shows the effects of A-II on loss of rod shape morphology during CN exposure in the presence of normal extracellular K⁺ ([K⁺]_o = 4.4 mM). Cells were exposed to substrate free Tyrode's solution containing 2 mM CN in the presence or absence of 100 nM A-II. During CN exposure, rod-shaped cells were gradually decreased in number over a 3-hr period. A-II significantly enhanced loss of rod shape morphology during CN exposure, but the difference was small. As discussed, it appears that the Na⁺ pump can function in cardiac myocytes quite well when glycolysis is not inhibited (Hasin and Barry, 1984). We therefore also examined the effects of partial inhibition of the Na⁺ pump by exposure to low [K⁺]_o on cell morphology under these conditions. Figure 7B shows effects of low [K⁺]_o (0.8 mM) in the presence or absence of 100 nM A-II on cell morphology. These quiescent cells could tolerate exposure to 0.8 mM [K⁺]_o and retained rod shape morphology under these conditions for more than 3 hr. In this experiment, in the presence of low K⁺, A-II more markedly enhanced cell contracture, assessed by loss of rod shape morphology, during CN exposure. To examine if the enhanced cell contracture caused by A-II is mediated by stimulation of Na⁺/H⁺ exchange, the effect of a Na⁺/H⁺ inhibitor was examined. As shown in figure 7C, 1 mM amiloride completely abolished the effects of A-II. We have previously shown that amiloride and EIPA have similar inhibitory effects on Na⁺/H⁺ exchange in these cells (Matsui et al., 1995). Comparison with the effects of amiloride in the absence of A-II during CN exposure (fig. 7D) indicated that A-II has little effect on cell morphology in the presence of amiloride during Na⁺/H⁺ exchange inhibition. Inhibitors of Na/H exchange may also inhibit Na/Ca exchange (Kleyman and Cragoe, 1988). To investigate whether any of the effects of amiloride and EIPA noted in our experiments could be due to inhibition of Na⁺/Ca⁺⁺ exchange, we examined their effects on the Na⁺/Ca⁺⁺ exchange current magnitude in adult rabbit ventricular myocytes. The control I_Na/Ca was 0.82 ± 0.04 pA/pF; in EIPA (10 µm) it was 0.83 ± 0.07; and in amiloride (1 mM) it was 0.58 ± 0.05 (means ± S.E.M., n = 5 to 12). Thus, the 28% inhibition of Na/Ca exchange induced by the concentration of amiloride we have used could have contributed in part to its effects on contracture, but inhibition of Na/Ca exchanger does not contribute to the effects of EIPA on changes in [Ca⁺⁺]_i during metabolic inhibition. Taken together, these results suggest that the augmentation of cell contracture caused by A-II is mediated by stimulation of Na⁺/H⁺ exchange, and that an increase in [Ca⁺⁺]_i resulting from altered Na⁺/Ca⁺⁺ exchange may be involved.

View larger version (34K): [in this window] [in a new window]   Fig. 7.   A, Effects of A-II (100 nM) on loss of rod shape morphology during exposure to 2 mM CN in the presence of normal [K+]o (4.4 mM). Numbers of rod-shaped cells were expressed as percentages relative to initial numbers of rod-shaped cells of each group. Means ± S.E.M. for eight experiments are plotted for control cells (squares), CN cells (diamonds) and CN + A-II cells (circles). The curve for CN+A-II as a whole was borderline significantly different from CN (P = .06) but at 150 and 180 min values were different at the P = .01 level. B, Effects of 100 nM A-II on loss of rod shape morphology during exposure to 2 mM CN in the presence of low [K+]o (0.8 mM). Means ± S.E.M. for eight experiments are plotted for low [K+]o cells (squares), low [K+]o + CN cells (diamonds) and low [K+]o + CN + A-II cells (circles). The low [K+]o + CN + A-II curve differed significantly from the low [K+]o + CN curve (P = .0003). C, Effects of 1 mM amiloride on the effects of A-II during exposure to 2 mM CN and 0.8 mM [K+]o. Means ± S.E.M. for six experiments are plotted for low [K+]o cells (square), low [K+]o + CN cells (diamonds) and low [K+]o + CN + A-II + amiloride cells (circles). Loss of rod cell morphology was inhibited by amiloride. D, The CN + amiloride plot (circles) was similar to the CN + amiloride + A-II curve in C, indicating a lack of effect of A-II in the presence of amiloride.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Possible importance of direct effects of A-II on myocardial ischemic injury/dysfunction. A number of studies have demonstrated that angiotensin converting enzyme inhibitors can have a beneficial effect in patients with ischemic heart disease (see Lonn et al., 1994 for review). Although a decrease in afterload seems likely to account for a component of the benefit of ACE inhibition, it has been recognized because the work of Koch-Weser (1964) that A-II can have direct myocardial effects. More recent studies (see Lindpaintner and Ganten, 1991) have also demonstrated that A-II may be produced directly within the myocardium by a local renin-angiotensin system. Thus some of the beneficial effects of converting enzyme inhibition in the presence of myocardial ischemia could be due to a decrease in a direct effect of circulating and/or locally produced A-II on the cardiac myocyte.

Importance of Na⁺/H⁺ in ischemic injury/dysfunction. Murphy et al. (1991) have demonstrated in an MRI study that the Na⁺/H⁺ exchange inhibitor amiloride reduced the rise in intracellular Na⁺ during global ischemia in the rat heart, and markedly delayed the increase in cytosolic calcium. Amiloride also decreased the rise in resting tension as estimated by isovolumic left ventricular diastolic pressure during ischemia, but had no effect on the rate or extent of ATP depletion, or intracellular pH. This important observation has resulted in additional studies that have confirmed beneficial effects of Na⁺/H⁺ exchange inhibitors on ventricular contractile function (Moffat and Karmazyn, 1993), and electrophysiologic stability (Scholz et al., 1992) during ischemia. Recent work has suggested that a reduction in Ca⁺⁺ overload of blood perfused hearts might be responsible for the beneficial effects of Na⁺/H⁺ exchange inhibition on post ischemic function (Hendrikx et al., 1994). It has also been shown that protein kinase-C activation aggravates hypoxic myocardial entry by stimulating Na⁺/H⁺ exchange (Ikeda et al., 1988).

Possible mechanisms of A-II effects during metabolic inhibition. Recent work from our laboratory (Ikenouchi et al., 1994; Matsui et al., 1995; Kohmoto et al., 1993; Ito et al., 1997) has demonstrated that A-II stimulates Na⁺/H⁺ exchange in adult rabbit and rat ventricular myocytes, and induces an intracellular alkalosis. This effect appears to be mediated by activation of protein kinase C (Kohmoto et al., 1993). Thus, it seemed possible that A-II might exacerbate ischemic injury/dysfunction via a PKC-mediated stimulation of Na⁺/H⁺ exchange.

Relevance of these findings to other species. The extent to which these findings can be extrapolated to species other than the rabbit is uncertain at this point. Ishihata and Endoh (1995) have emphasized the species-related differences in inotropic effects of A-II in mammalian ventricular muscle. Dog, rat and ferret myocardium was less sensitive to the effects of A-II than rabbit. Moravec et al. (1990) have reported that A-II has positive inotropic effects in human myocardium, both atrium and ventricle, whereas Lefroy et al. (1996) have reported that A-II has no effect on contraction of isolated myocytes from guinea pig, rat and human ventricle, and from human atrium. Holubarsch et al. (1993) have reported that A-II has inotropic effects in atrial but not ventricular human myocardium. In addition, work from our laboratory has emphasized that the effects of A-II on myocardial cells may vary with development (Kohmoto et al., 1993) and hypertrophy (Ito et al., 1997). Thus the extent to which our findings can be related to beneficial effects of angiotensin converting enzyme inhibition in a clinical situation where both acute and chronic myocardial effects can be expected is uncertain.

Usefulness of flow cytometry for measurement of [Ca⁺⁺]_i and viability in adult ventricular myocytes. To our knowledge, this is the first report that has described the application of flow cytometry for the measurement of changes in cytosolic Ca⁺⁺ and viability simultaneously in adult ventricular myocytes. Flow cytometry, of course, has been widely applied for the measurement of a variety of fluorescent-based parameters in nonmyocyte cells, but the large size and partial viability of acutely dissociated adult ventricular myocytes has complicated the methodology. By the use of PI and fluo-3 we were clearly able to separate viable (non-PI staining) cells from nonviable cells, and to measure cytosolic Ca⁺⁺ in resting myocytes. This approach allows the detection of alterations in Ca⁺⁺ during metabolic inhibition induced by drugs, and extends previously reported use of flow cytometric analysis to study microtubular fluorescence in cardiac myocytes (Armstrong and Ganote, 1992). This promises to be a powerful technique that will facilitate the detection of subtle changes in cytosolic Ca⁺⁺ in large numbers of viable isolated myocytes.