INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE DEVELOPMENT OF CONGESTIVE heart failure (CHF) is one of the most serious outcomes of myocardial infarction (MI). CHF is also the end stage of a variety of other conditions, such as hypertension, valvular disease, and cardiomyopathies almost invariably associated with cardiac hypertrophy. It is unclear whether hypertrophy and CHF are associated with reduced cell shortening. Most studies, some of which are based on quite heterogeneous human material, have demonstrated a decrease in the magnitude of contractions and intracellular Ca²⁺ transients as well as a slowing in time to peak and relaxation of the mechanical and Ca²⁺ signals (6, 11, 14, 15). However, these changes in mechanical function are most obvious at elevated Ca²⁺ concentrations and heart rates (11). Other in vitro studies of heart failure do not show reduced contractility, including one of myocardial strips from explanted human hearts (34) and another on isolated cells from rats with MI (1). The Ca²⁺ current (I_Ca) through L-type Ca²⁺ channels seems to be unchanged in failing human (5, 6) and pacing-induced hypertrophy and CHF in dog (31) myocytes. One might therefore speculate that the coupling between I_Ca and Ca²⁺ release from the sarcoplasmic reticulum (SR) is highly variable depending on the specific disease conditions and choice of experimental models. Indeed, a recent report suggests a reduced efficiency of the coupling between the sarcolemmal trigger and the Ca²⁺ release response of the SR during hypertrophy and CHF in a strain of rats with salt-sensitive hypertension (14).

Other mechanisms might also be involved in alterations in excitation-contraction (E-C) coupling in hypertrophy and CHF. It has been proposed that Na⁺/Ca²⁺ exchange, working in its reverse (Ca²⁺ influx) mode, can trigger Ca²⁺ release from the SR (4, 23, 27, 32). The sarcolemmal Na⁺/Ca²⁺ exchange protein and mRNA have been reported to be upregulated in CHF (35, 40). In line with this, enhanced Na⁺/Ca²⁺ exchange activity was observed in a genetic model of cardiomyopathy (17). However, in an MI model in which direct electrophysiological measurements were used, it was reported that Na⁺/Ca²⁺ exchange current (I_NaCaX) is decreased in CHF (47).

To examine these possible alterations in E-C coupling, we investigated changes in the contribution of I_Ca to E-C coupling in a well-controlled model of hypertrophy and CHF after MI in normal Wistar rats. The extent of the MI was such that acute cardiac insufficiency was avoided, but already after 1 wk there were signs of hypertrophy and failure (41) closely mimicking the disease state in humans. Furthermore, experimental conditions were chosen to ensure that normal temperature and transmembrane K⁺ gradient were maintained, since recent results have indicated that lower temperatures and substitution of Cs⁺ for K⁺ might have direct influences on E-C coupling (25, 42).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Induction of MI. Methods for induction of MI and hypertrophy have been published elsewhere (41), as have the characteristics of the resulting cardiac disease (38). Extensive MI was induced by ligature of the left coronary artery in male Wistar rats weighing ~300 g under halothane anesthesia according to a protocol approved by the Norwegian National Committee for Animal Care and Use conforming to the Norwegian Animal Welfare Act and National Institutes of Health guidelines. Sham animals were treated identically, except no ligature was placed on the left coronary artery. Animals were kept for 6 wk before isolation of myocytes, during which time they developed a well-defined and organized scar area. The surviving myocardium showed no inflammatory reaction or collagen deposit (38). At this time the animals were again anesthetized with halothane, and left ventricular end-diastolic pressure (LVEDP) was measured by means of a 2-F micromanometer-tipped catheter (model SPR-407, Millar Instruments) inserted through the right carotid artery before the hearts were rapidly removed for further processing. Only MI animals with LVEDP >15 mmHg were included in the study. Compared with the gradual development of CHF in humans, 6 wk is a short period. However, the acute inflammatory reaction has subsided at this time and the clinical picture resembles the human condition quite closely, in that the hearts are hypertrophied and increased LVEDP is required to maintain cardiac output. Thus this experimental model produces reliable eccentric hypertrophy and volume overload of the left ventricle.

Measurements of ionic current and cell shortening. Cells were isolated using a Langendorff setup with retrograde perfusion of collagenase solution as described elsewhere (18, 41). The perfusion solution was based on a modified Joklik's medium (catalog no. 22300, Life Technologies). Only cells from the surviving part of the left ventricle were suspended in medium 199 (catalog no. M7528, Sigma Chemical), and >70% were rod shaped. Isolated cells were placed on laminated coverslips until use within 24 h of isolation. The coverslip was used as the floor of the experimental chamber on the stage of a Nikon Diaphot inverted microscope. The chamber (200 µl volume) was superfused at 36 ± 1°C at a rate of 2-3 ml/min. Myocyte contraction was measured with a video edge detector.

Northern blot analysis of mRNA for the Na⁺/Ca²⁺ exchanger. Poly(A)⁺ RNA was extracted from homogenized left ventricular tissue from MI and Sham rats. The infarct area was carefully removed, together with the right ventricle and the atria, leaving only viable tissue from the left ventricle. Oligo(dT)-conjugated paramagnetic beads were used to extract poly(A)⁺ RNA according to the manufacturer's instructions (Dynal, Oslo, Norway). The poly(A)⁺ RNA was size fractionated on a formaldehyde-agarose gel by use of 10 µg of poly(A)⁺ RNA per lane, transferred to a Biotrans nylon membrane (ICN Biomedicals), and hybridized at 42°C first for 3 days with a cDNA probe for the NCX1 (500 bp, kindly provided by Dr. K. D. Phillipson, University of California at Los Angeles) and after stripping for 1 day with a probe for glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 1,300 bp, kindly provided by Dr. H. Prydz, University of Oslo, Norway). After hybridization, the Na⁺/Ca²⁺ exchanger and GAPDH membranes were washed four times at 35°C in 0.1× standard saline-sodium citrate (SSC)-0.1% SDS and 2× saline-sodium citrate-0.1% SDS at room temperature for 10 and 15 min and then washed twice for 15 min at 60°C. cDNA probes were randomly primed with [³²P]dCTP and [³²P]dATP, and stripping was carried out at 65°C for 1 h in a buffer containing 10 mM NaH₂PO₄ and 50% formamide at pH 6.5. The membranes were scanned and analyzed using PhosphorImager and ImageQuant software (both from Molecular Dynamics, Queensland, Australia). To estimate the Na⁺/Ca²⁺ exchanger mRNA tissue level, the Na⁺/Ca²⁺ exchanger transcript level-to-GAPDH transcript level ratios were calculated in each sample. For more precise quantification and to ensure linearity of transcript signal, a slot-blot analysis was carried out by applying 0.5, 1, and 2 µg of poly(A)⁺ RNA to a nylon membrane with use of a filtration manifold (Minifold II, Schleicher & Schuell, Dassel, Germany). The slot-blot membranes were subjected to the same procedure as the Northern membranes, and the Na⁺/Ca²⁺ exchanger transcript signal was normalized to the signal of an oligo(dT)₁₈ probe (Eurogentec, Seraing, Belgium) for each sample.

Western blot analysis. A Western blot analysis was performed on a crude membrane fraction isolated from homogenates of left ventricles from Sham and MI hearts, with care taken to exclude the infarct area (16, 30, 38). Protein concentration in the membrane preparation was determined by a micro-bicinchoninic acid protein assay (Pierce Chemical, Rockford, IL) with BSA as standard. The proteins were size fractionated by SDS-PAGE on 8% polyacrylamide gels as described by Laemmli (22) and electrophoretically transferred to nitrocellulose membranes. After they were blocked with 10% nonfat dry milk in Tris-buffered saline with 0.1% Tween 20 (TBS-T buffer) for 1 h at room temperature, the membranes were incubated for 1 h with a rabbit anti-Na⁺/Ca²⁺ exchanger polyclonal antibody (code 11-13, Swant, Bellinzona, Switzerland) diluted 1:1,000 in blocking solution for 1 h. After repeated washing in TBS-T buffer, the membranes were incubated with anti-rabbit IgG conjugated to horseradish peroxidase (catalog no. NA 934, Amersham, Oakland, ON, Canada) as secondary antibody. Again, repeated washing was carried out in TBS-T buffer. The immunoreactive bands were detected by the enhanced chemiluminescence method (RPN 2106, Amersham). The image was scanned and staining density was quantified with ImageQuant.

Statistics. Values are means ± SE. Comparisons between means were made using Student's t-test with the Bonferroni correction for multiple comparisons when appropriate. A minimum of three animals was used in each group. Comparisons of averaged I_NaCaX were accomplished with multivariate ANOVA followed by Scheffé's test. Differences between sample means were considered significant when P < 0.05 unless specified otherwise.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characteristics of the rat MI model. Six weeks after ligation of the left coronary artery, heart weight was 74% higher in MI than in Sham rats (Table 1). Heart weight-to-body weight ratios were also significantly increased, showing that a significant hypertrophy had developed in the remaining viable part of the left ventricle mainly localized to the septal region (Table 1). Left ventricular systolic pressure was significantly lower and LVEDP significantly higher in MI than in Sham rats. Together with an increased lung weight (data not shown), this indicates that the animals had developed CHF.

Voltage dependence of contraction in Sham and MI. The characteristics of contraction and changes in E-C coupling after MI were investigated under voltage-clamp conditions. Contractions were elicited by voltage steps to V_t of 30 to +100 mV from a V_h of 40 mV, at which cell length was stable and not different from the length at 80 mV throughout the 1-s period between the last prepulse and the test pulse. Maximum shortening was greater at a V_t of +80 than 0 mV by 47 ± 10 and 37 ± 11% in Sham and MI, respectively (P < 0.05 in both groups). This is exemplified by the two experiments shown in Fig. 1 (cf. A and C with B and D), and Table 2 gives mean data on fractional shortening at these two voltages. There was no significant difference between Sham and MI cells at any V_t. However, contractions were significantly slower in MI than in Sham, since time to peak contraction and time to 50% relaxation were increased at both V_t (Table 2).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Transmembrane voltage, net ionic current, and cell shortening during depolarizations from a holding potential of 40 mV to test potentials (Vt) of 0 mV (A and C) and +80 mV (B and D) in a Sham cell (A and B) and in a cell 6 wk after myocardial infarction (MI, C and D) in control, in the presence of verapamil (20 µM), and after addition of Ni2+ to the superfusate. Recordings were made 2 min after the addition of each drug. Inset: prepulse protocol used in all experiments unless otherwise indicated.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Summary of voltage dependence of cell shortening in Sham (A, n = 7) and MI (B, n = 10) cells in the absence and presence of verapamil (20 µM). Cell shortening was normalized to 100% in control at 0 mV. C: verapamil-sensitive component of contraction in Sham () and MI () cardiomyocytes. In contrast to Sham cells, cell shortening in MI cells was not significantly different from zero, except in the voltage range of maximal Ca2+ current (* P < 0.05).

Verapamil sensitivity of contraction. The effects of verapamil and Ni²⁺ on voltage dependence of contraction in Sham and MI cells are illustrated by the two experiments in Fig. 1. Verapamil (20 µM) reduced the contraction at a V_t of 0 mV by 66% and at +80 mV by 48% in a Sham cell (Fig. 1, A and B). Subsequent addition of Ni²⁺ (4 mM) abolished the remaining verapamil-insensitive contraction at both voltages. Exposure of an MI cell to verapamil reduced contraction by 26% at 0 mV and by 19% at +80 mV (Fig. 1, C and D). Thus a high concentration of verapamil caused much less block of contraction at both V_t in the MI than in the Sham cell. Again, exposure to Ni²⁺ abolished the verapamil-insensitive component of contraction. Exposure to ryanodine (5 µM) blocked all verapamil-insensitive contraction in both cell types (n = 5; data not shown).

Effect of verapamil on I_Ca. The efficacy of Ca²⁺ channel blockade by verapamil was confirmed in each cell type in a separate series of experiments. Figure 3A shows the effects of verapamil (20 µM) on Sham and MI cells. The I_Ca density was nearly the same in the two cell types under control conditions. These values from rats are almost three times greater than in normal and failing human hearts (5, 31). Verapamil blocked inward current in both treatment groups, as shown by the two inset current traces (V_t = 0 mV) and by the summarized mean data for all V_t. Thus it is likely that I_Ca was blocked to the same extent in both cell types, so that differential sensitivity to the Ca²⁺ channel antagonist in the two cell types cannot underlie the resistance of MI cells to the negative inotropic effects produced in Sham cells.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   A: effects of verapamil on Ca2+ current measured in Cs+-containing internal and external solutions after three 50-ms prepulses to 0 mV at 1 Hz. Insets: original current recordings at Vt = 0 mV (vertical bar is 1 nA). Ca2+ current is shown under control conditions in Sham (n = 11) and MI (n = 7) and during exposure to verapamil (Vp). B: relationship between the magnitude of Ca2+ current (data from A) normalized to 100% at maximal current at 0 mV and cell shortening (data from Fig. 2, A and B). Arrow, direction of increasing depolarization.

Relationship between I_Ca and contraction. Figure 3B summarizes the relationship between normalized current (from Fig. 3A) and normalized cell shortening under these conditions (data from Fig. 2, A and B). There was a close association between I_Ca and contraction amplitude over the voltage range of current activation (arrow) in the Sham cells until maximal current was reached (at 0 mV). Current declined with additional depolarization as the reversal potential for I_Ca was approached, but contraction continued to increase. Similar results were obtained in MI cells, demonstrating a close correlation between activation of I_Ca and contraction only up to the voltage range of peak current (~0 mV), above which current and contraction amplitudes become negatively correlated.

Measurements of I_NaCaX. One possible contributor to activation of contraction in the absence of I_Ca could be reverse-mode Na⁺/Ca²⁺ exchange. We measured I_NaCaX in Sham and MI cells to determine any changes induced by MI. Under control conditions, successive depolarization ranging from 40 to +100 mV caused a progressively larger outward current, as seen in a Sham cell in Fig. 4A. Addition of Ni²⁺ (10 mM) to the superfusate reduced the magnitude of the outward currents at all V_t (Fig. 4B). The Ni²⁺-sensitive current, which is a measure of I_NaCaX, is shown in Fig. 4C. When the identical protocol was applied to an MI cell (Fig. 4, D and E), I_NaCaX was greater in the MI cell, as evidenced by the magnitude of the Ni²⁺-sensitive current (Fig. 4F), whereas Ni²⁺-insensitive currents were nearly the same. In control experiments with 0 mM Na⁺ in the pipette, Na⁺/Ca²⁺ exchange was inactive and Ni²⁺ had no detectable effect on current in either cell type.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Na+/Ca2+ exchange current (INaCaX) in Sham and MI cells. Currents were measured in Cs+-containing solutions with stepping from a holding potential of 40 mV to Vt ranging from 40 to +80 mV in a Sham (A-C) and an MI (D-F) cell. Test potentials were applied after 3 conditioning pulses to 0 mV (50 ms). Currents were recorded before (A and D) and after (B and E) exposure to Ni2+ (10 mM), and Ni2+-sensitive currents were calculated (C and F).

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Summary of INaCaX measurements in Sham and MI cells. Ni2+-sensitive currents normalized for cell capacitance are summarized for all voltages in Sham (n = 9) and MI (n = 9) cells (* P < 0.05, for each data point). Open symbols show corresponding control measurements made in the absence of INaCaX (0 mM intracellular Na+ and 0.5 mM extracellular Ca2+).

Quantification of the Na⁺/Ca²⁺ exchanger mRNA signal and protein. An important question is whether this increased outward I_NaCaX represents an increase in the exchanger protein or whether exchange kinetics are altered. Figure 6A shows a representative Northern blot. A strong signal was detected at 7 kb, and weaker signals appeared at ~4.7 and 1.5 kb in accordance with previous reports (9). When normalized against the GAPDH signal, the transcript signal from MI (n = 13) hearts exceeded the Sham (n = 7) signal by 30-40% (P < 0.05). Figure 6A also shows a strong transcript signal for the Na⁺/Ca²⁺ exchanger in the infarct area. Figure 6B shows the summarized results from a separate slot-blot analysis. Again, the signal from MI hearts (n = 13) exceeded that from the Sham hearts (n = 7, P < 0.05) by 37 ± 10%. The signal in the infarct area was more than twice as strong as in Sham hearts.

View larger version (34K): [in this window] [in a new window]   Fig. 6.   Northern blot of Na+/Ca2+ exchange (NCX) expression in Sham and MI hearts. A: mRNA was isolated from the tissue surviving the infarct (MI) and from the infarct area (IA). GAPDH, glyceraldehyde-3-phosphate dehydrogenase; SH, Sham. B: quantitative slot-blot results. mRNA (0.5, 1.0, and 2.0 pg) extracted from 7 Sham and 13 MI hearts and 4 infarct areas were analyzed.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   A: Western blot with a crude membrane preparation from whole heart tissue by use of a Na+/Ca2+ exchange antibody. B: quantitative results from Sham (n = 8) and MI (n = 12) hearts normalized to the mean values for Sham samples (* P < 0.05).

Cell contractions in the absence of Na⁺. To confirm that I_Ca can serve as an effective trigger for E-C coupling in the absence of Na⁺/Ca²⁺ exchange, we performed a separate series of experiments in the absence of intracellular and extracellular Na⁺. Figure 8 shows the results of experiments performed in a Sham and MI cell in the absence of Na⁺. The magnitude of net outward current in the Na⁺-free solutions increased with depolarization in the Sham cells; the magnitude of the outward current was dramatically increased at all V_t in the MI cells. The nature of this outward current was not examined further, but it could be a chloride current; K⁺ currents are reduced in cardiac hypertrophy and failure (7, 36), whereas there seems to be sustained activation of chloride channels (3, 10).

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Activation of contraction in Na+-free conditions. Transmembrane potential, ionic current, and cell shortening in a Sham cell and in an MI cell in response to test voltage steps to 20, 0, and +50 mV are shown after 3 prepulses to 0 mV for 30 ms at 1 Hz.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Summary of cell shortening in Na+-free solutions. A: composite results of cell contractions normalized to maximal contraction at 0 mV for all Vt in Sham cells (, n = 6) and MI cells (, n = 6 *P < 0.05 vs. Sham cells) before and after (open symbols) addition of verapamil. B: relationship between Ca2+ current and contraction, both normalized to 100% of their respective values at 0 mV (cf. Fig. 3B), in Sham and MI cells in Na+-free solutions. Ca2+ current values were obtained from Fig. 3B in Sham and MI cells (Cs+-containing solutions), and data for contraction were taken from A. Arrows, trajectory described by increasing depolarization.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We found that, in a model of experimentally induced MI with resulting hypertrophy and failure in rat hearts, isolated ventricular myocytes surviving the infarct show altered structure, physiological function, and pharmacological responses. Most importantly, the sensitivity of contraction to L-type Ca²⁺ channel blockade with verapamil was greatly reduced in MI compared with Sham cells, and the verapamil-insensitive contractions were blocked by Ni²⁺. Finally, there was an increased I_NaCaX, which might be responsible for a more prominent role of reverse-mode Na⁺/Ca²⁺ exchange in triggering contraction in MI, thus explaining the relative insensitivity of MI cells to the effects of verapamil.

Voltage dependence of E-C coupling after MI. E-C coupling has been studied extensively in isolated cardiac myocytes and has revealed that Ca²⁺-induced Ca²⁺ release is activated by Ca²⁺ influx through the L-type Ca²⁺ channel (2, 44). The resulting voltage dependence demonstrates a clear reliance of E-C coupling on I_Ca, giving a characteristic bell shape in which intracellular Ca²⁺ transients and contraction follow the voltage dependence of I_Ca. Thus it is somewhat surprising that contractions from Sham and MI cardiomyocytes demonstrate a sigmoidal dependence on V_t. The explanation for this result is likely to derive from recent reports suggesting that voltage-dependent activation of the reverse (Ca²⁺ influx) mode of Na⁺/Ca²⁺ exchange might contribute to triggering in a number of mammalian species including the rat (21, 23, 24, 26, 42). However, in normal cardiomyocytes the importance of the reverse-mode exchange for normal contraction has been questioned (39). The difference between these and many other reports demonstrating an exclusive reliance of activation on I_Ca is most likely the result of differences in experimental conditions; substitution of the primary physiological monovalent cation K⁺ with Cs⁺ is usually used to block K⁺ currents and thus isolate I_Ca. However, low temperatures and Cs⁺ replacement appear to suppress the apparent contribution of the exchanger to activation of contraction (25, 42, 43). The net result is that when the normal K⁺ gradient and physiological temperatures are used, the voltage dependence of contraction is indeed sigmoidal. The present study reveals that this is true for Sham and MI cells, suggesting that the exchanger contributes to triggering in normal myocytes and in myocytes surviving MI.

Altered sensitivity to verapamil in MI cells. The most striking observation in this study was that MI cells showed reduced sensitivity to block of contraction by verapamil compared with Sham cells, despite the demonstration that the same concentration of verapamil was highly effective in blocking I_Ca in both cell types.

Changes in I_NaCaX after MI. There have been several recent but conflicting reports of measurements of I_NaCaX in cells isolated from animal models of hypertrophy and CHF. One recent report found a decrease in I_NaCaX in a similar model of MI in rat heart (47). The cells were studied 3 wk after the MI, selection of animals was not made on the basis of LVEDP, the cellular hypertrophy was significantly less than in the present study, and the cells were studied at a lower temperature. Thus the data are not directly comparable to the present study. Litwin and Bridge (28) found that I_NaCaX was increased in a model of MI in rabbit heart. In addition, O'Rourke and co-workers (33, 46) found that a 28% reduction in sarco(endo)plasmic reticulum Ca²⁺-ATPase subunit 2a (SERCA2a) and phospholamban in the pacing-induced model of heart failure in dog was accompanied by an increase in Na⁺/Ca²⁺ exchange protein of 100%. Schwinger et al. (37) recently reported no change in exchanger protein levels, despite evidence of an increase in activity in failing human heart (35).