Heart Failure After Myocardial Infarction
Altered Excitation-Contraction Coupling
Background— Heart failure (HF) frequently follows the occurrence of myocardial infarction (MI). Questions about how HF develops and what cellular defects contribute to this dysfunction led to this study.
Methods and Results— MI was induced in rats by coronary artery ligation. Clinical examination of the post-MI (PMI) surviving animals indicated that they were in overt HF by all measures. Cellular examination of the cardiomyocytes by patch-clamp and confocal [Ca2+]i imaging methods indicated that cellular function was significantly compromised. At the single-cell level, [Ca2+]i transient amplitudes were reduced and contractions were decreased and slowed, although Ca2+ current (ICa) remained unchanged. The excitation-contraction coupling (ECC) gain function measured as Δ[Ca2+]i/ICa was significantly decreased. Ouabain, a cardiotonic steroid that blocks the Na+,K+-ATPase and activates Ca2+ entry via cardiac Na+ channels, largely alleviated this defect.
Conclusions— After MI, ICa becomes less able to trigger release of Ca2+ from the sarcoplasmic reticulum. This failure of ECC is a major factor contributing to the development of contractile dysfunction and HF in PMI animals. The improved ECC gain, enhanced Ca2+ entry, and augmented Ca2+ signaling due to cardiotonic steroids contribute to the beneficial effects of these agents.
Received January 24, 2001; revision received April 11, 2001; accepted April 17, 2001.
Myocardial infarction (MI) is one of the leading causes of cardiac mortality, morbidity, and dysfunction.1 The time course of ventricular dysfunction is arbitrarily divided into 2 periods. MI produces cellular death and thus leads to complete loss of contractile function in the infarcted region (acute phase), whereas there is no necessary change in contractile function in other regions. In survivors, there is a remodeling of the noninfarcted area that underlies the second phase of dysfunction. Heart failure (HF) is frequently seen during and after this period of remodeling.2
Dysfunction of ventricular myocytes develops in many cardiovascular diseases. In hypertension, there are changes in contractile protein isoforms,3,4 ionic currents,5,6 ion transporters,7 and 2Ca2+ signaling,8 and cellular hypertrophy is observed.2 Ca2+-signaling alterations appear to go through 2 phases, 1 associated with early components of hypertension3,4 and 1 revealed as HF develops.8 A similar array of molecular and cellular changes is produced by other insults that lead to HF.9
This report presents a cellular investigation of HF in a rat model of post-MI (PMI) remodeling after coronary artery ligation.10 The work identifies specific cellular changes that are observed in cardiac myocytes after MI and shows how a controversial therapy (cardiotonic steroids) can act to improve cellular function.
Induction of MI
Male Wistar rats (180 to 200 g) were used to produce MI as previously described.10 Age-matched rats were used as controls.
Some of the animals were used for hemodynamic measurements (Table 1). Furthermore, before they were euthanized, animals underwent blood pressure measurements or echocardiography (Table 1). Blood pressure was measured in conscious animals as described earlier.8 M-mode echocardiography was performed in anesthetized animals (ketamine 50 mg/kg and xylazine 10 mg/kg).
Cell Isolation and Recording Techniques
Ventricular cardiac myocytes were isolated by standard enzymatic techniques.8 After isolation, cells were stored in DMEM (1.8 mmol/L CaCl2); cells from the left ventricular free wall were used for the study.11
Whole-cell current was monitored with an Axopatch 200A amplifier and recorded with pCLAMP-6.01 (Axon Instruments). Series resistance was electronically compensated to 40% to 60%.
Calcium current (ICa) was elicited by 100-ms step depolarizations according to a protocol previously described.8 Sarcoplasmic reticulum (SR) Ca2+ content was normalized by a loading protocol of 4 steps from −80 to 0 mV at 1 Hz before each test depolarization. External solution contained (in mmol/L) NaCl 135, MgCl2 1, CsCl 20, glucose 10, HEPES 10, 4-aminopyridine 3, and CaCl2 1.8; pH 7.4. Patch pipettes contained (in mmol/L) CsCl 130, fluo-3 0.05, HEPES 10, MgCl2 0.33, TEA-Cl 20, and Mg-ATP 4; pH 7.2.
For action potential (AP) measurements, the current-clamp mode of the patch-clamp technique was used. Cells were stimulated at 1 Hz with current pulses 1.5 times threshold. External solution contained (in mmol/L) NaCl 140, KCl 5.0, NaH2PO4 0.33, MgCl2 1, HEPES 5, and glucose 5.5; pH 7.4. Internal solution was the same as for ICa recordings, but Cs+ was replaced with K+, and no TEA-Cl was added. Averaged APs were used as the voltage command with Cs+-containing solutions. Temperature was maintained at 35°C to 37°C.
[Ca2+]i transients and Ca2+ sparks were imaged by confocal microscopy and analyzed as previously described.8
Data are presented as mean±SEM. Two-sample comparison was made by paired or independent t test, as appropriate. A level of P<0.05 was assumed to be statistically significant.
MI produces a dramatic change in cardiac output and may be lethal. In the animal model used here, ≈40% of the animals died during the first 2 days. We waited 6 months because in our hands, this is the time when consistent HF is presented in PMI rats. Visual inspection revealed large scars on the left ventricular free wall. Table 1 characterizes the cardiovascular properties of the surviving animals, indicating that PMI animals were suffering from HF. Table 2 shows that there was cellular hypertrophy.
To investigate whether excitation-contraction coupling (ECC) was altered after MI, isolated cardiomyocytes were examined by electrical and optical methods. A loading protocol (see Methods) was used to normalize SR Ca2+ content. The similarity of the SR Ca2+ load was verified by a caffeine-release method.8 This procedure was carried out because SR Ca2+-ATPase (SERCA) levels are decreased in HF7,12 and may affect our results and interpretation.
Figure 1 shows representative examples of depolarization-activated signals from control and PMI ventricular myocytes. The amplitudes of both the [Ca2+]i transient and contractions were reduced in PMI cells. Cell shortening kinetics were significantly slowed in PMI myocytes (time to peak: 79.5±6.4 ms in 5 control versus 109.8±7.3 ms in 12 PMI cells, P<0.05). Whereas the slow time constant of ICa decay was similar in control and PMI myocytes, the fast time constant increased (at +10 mV, 6.3±0.9 ms in 9 control myocytes versus 9.0±0.7 ms in 21 PMI cells, P<0.05). This finding is consistent with the slowing of ICa inactivation observed in pressure overload—induced HF.8
Averaged data are plotted as a function of voltage in Figure 2. The decreases in cell shortening (Figure 2A) and [Ca2+]i transient (Figure 2B) observed for PMI cells are statistically significant compared with control cells over a wide range of membrane potentials. ICa density was maintained (Figure 2C), however, suggesting that changes in ICa density cannot account for the contractile dysfunction observed in PMI cells. Additional explanations include the “myosin isoform shift hypothesis”4 and the “Ca2+-signaling hypothesis”8 (Figure 2D and Discussion).
The reduction of the [Ca2+]i transient for PMI cells shown in Figure 2 would not have been predicted if the only significant cause of the contractile dysfunction were changes in the contractile proteins.4 The Ca2+-signaling hypothesis is consistent with the data in Figure 2 but requires further testing. To investigate the importance of Ca2+ signaling, an inotropic agent was applied. The cardiotonic steroid ouabain was chosen because it has no known direct actions on contractile proteins and has established actions on Ca2+ signaling (see below). Figure 1C shows records obtained in the same PMI myocyte as displayed in Figure 1B, 1 minute after 100 nmol/L ouabain application. Ouabain clearly increased the [Ca2+]i transient and contraction. Averaged data obtained from PMI myocytes after ouabain perfusion are shown in Figure 2. Cell shortening and [Ca2+]i transients were restored to normal over voltage steps negative to +20 mV. No significant changes in ICa density were seen.
The efficacy of ICa to activate SR Ca2+ release is measured as the ECC gain: [Ca2+]i transient normalized to ICa [(F/F0)/ICa] (Figure 2D). It is best to measure gain at fairly negative potentials because the opening of a single L-type Ca2+ channel at these negative potentials can trigger a single Ca2+ spark.13 The decrease in gain seen in PMI cells was restored by ouabain application.
Because the decrease in gain appears to play a role in the contractile dysfunction in PMI cells, we were interested in determining why the [Ca2+]i transient was reduced. The Ca2+ influx (ICa) activates Ca2+ sparks that sum to produce the [Ca2+]i transient.14 To investigate whether alterations in Ca2+-spark properties may lead to the reduction of [Ca2+]i transient, Ca2+ sparks were analyzed. Signal-averaged Ca2+ sparks from control and PMI cells look alike (Figure 3A and 3B). They have similar magnitudes (F/F0: 1.70±0.02 versus 1.72±0.02 in 141 Ca2+ sparks in control versus 338 Ca2+ sparks in PMI myocytes) and decay kinetics (τ, obtained by fitting the decay phase to a single exponential: 21.87±1.30 versus 24.20±0.85 ms in control versus PMI Ca2+ sparks). Counting the number of Ca2+ sparks triggered at different membrane potentials in the presence of a submaximal concentration of nifedipine13 revealed that fewer Ca2+ sparks were activated in PMI cells even though ICa was similar. Because Ca2+-spark characteristics in control and PMI cells are similar, these results suggest that the reduction in [Ca2+]i transient is due to a decreased ability of Ca2+ current to trigger Ca2+ sparks.
Changes in contractile protein isoforms or properties could also contribute to the observed dysfunction. To investigate this, the relationships between the [Ca2+]i transient and cell shortening were examined at different voltages (Figure 4A). The shapes of these plots (phase-plane loops) reflect the kinetic response of the contractile proteins to changes in [Ca2+]i.15 During the rapid [Ca2+]i rise, there is little contractile response. Contraction begins in earnest after [Ca2+]i is near its peak and continues even as [Ca2+]i falls. Finally, both [Ca2+]i and contraction decline together. [Ca2+]i transient reduction in PMI cells leads to small contraction and different loop trajectory (Figure 4B) that is restored back to control by ouabain (Figure 4C). There is, however, a small residual rounding of these trajectories that may reflect alterations in the contractile proteins (see Discussion). Figure 4D shows plots of peak contraction versus peak [Ca2+]i transient and is consistent with data shown in Figure 4A through 4C.
In vivo, the decreases of [Ca2+]i transients in PMI cells will be influenced by changes in action potential duration (APD), which is prolonged in HF. APs were recorded in control and PMI myocytes and used to control voltage. Figure 5 shows that the PMI cells had smaller contractions and [Ca2+]i transients than control cells despite their longer APD. In addition, ouabain increased both contraction and [Ca2+]i transient to normal. Blockade of Na+ channel by tetrodotoxin (TTX) (10 μmol/L) reduced the [Ca2+]i transient back to the level of PMI cells before the ouabain addition (Figure 5B). Importantly, Santana et al16 showed in normal rat cells that TTX has no immediate effect on the [Ca2+]i transient in the absence of ouabain, that ouabain acutely increases the [Ca2+]i transient, and that TTX can block the immediate ouabain-dependent increase in the [Ca2+]i transient.
Changes that contribute to contractile dysfunction in ventricular myocytes after MI were examined in this article. The reduction of SERCA7,12 must play an important role in the altered Ca2+ signaling observed in this model of HF. Data presented here, however, suggest that when SR Ca2+ content is normalized, PMI cells still present a significant defect in Ca2+ signaling: the ability of Ca2+ influx (through L-type Ca2+ channels) to trigger SR Ca2+ release is reduced. Under our experimental conditions, changes in contractile proteins4 do not appear to be a major factor. The importance of defective Ca2+ signaling in the cellular pathophysiology of PMI myocytes shown here has many implications for medical therapy and challenges our understanding of HF.
The contractile dysfunction and altered Ca2+ signaling that develop after MI are similar to those observed in other models of HF.8,9,15,17 To date, 2 molecular interventions that alleviate HF have been identified: restoration of β-adrenergic receptor function15,18 and augmented SERCA function.19,20 These approaches restore or improve Ca2+ signaling. Three issues are raised by studies of cellular Ca2+-signaling defects in HF: (1) How do the molecular/drug therapies work? (2) What are the cellular defects contributing to the altered Ca2+ signaling? (3) What is the instigating cause of the Ca2+-signaling defect?
How can 100 nmol/L ouabain improve the defective Ca2+ signaling in PMI cells? There are at least 4 possible explanations. (1) Inhibition of the Na+ pump (Na+,K+-ATPase) decreases Na+ extrusion and, through the Na+/Ca2+ exchanger, increases [Ca2+]i. We cannot rule out a contribution by this mechanism to our findings if the changes in SR Ca2+ content were small (within the statistical error). These effects require time (minutes), however, and do not occur within a single heartbeat. (2) There may be sufficient elevation of subcellular [Na+]i during a single depolarization for Na+/Ca2+ exchanger to trigger Ca2+ release (see Leblanc and Hume21). This possibility is examined in Figure 6. (3) Ca2+ flux through Na+ channels occurs (ie, through slip-mode conductance). This does not apply to the data in Figures 1 to 3⇑⇑ because INa is inactivated, but it probably applies in part to Figure 5, because the effect is blocked by TTX. (4) There may be a novel action of ouabain to directly activate the ryanodine receptors (RyRs).22
Leblanc and Hume21 suggested that INa could provide sufficient Na+ influx to activate Ca2+ entry through the Na+/Ca2+ exchanger even during a single AP. To enable the brief Na+ influx to significantly elevate [Na+]i, a subsarcolemmal region around Na+ channels with reduced diffusion, the “fuzzy space,” was hypothesized.23 To test this possibility, we examined the actions of instantaneous Na+-pump blockade on [Ca2+]i transients and contraction on control cardiac myocytes. We produced a maximal Na+ influx by applying an AP clamp to activate INa after removal of extracellular Cs+ (in K+-free solutions) to completely block the Na+ pump24 (Figure 6). If the mechanisms of Leblanc and Hume21 and of Arnon et al25 were relevant to the enhanced contractions produced by ouabain (Figure 5), then an increase in [Ca2+]i transients and contraction should be seen after complete Na+-pump inhibition in control cells. There was no statistical difference, however, in the measured [Ca2+]i transients (Figure 6B and 6C). This finding is similar to that reported by Su et al26 under similar conditions. From this and the fact that the effects in Figure 5 are blocked by TTX, we conclude that the positive inotropic effects of ouabain occur at least in part through the activation of slip-mode conductance of the Na+ channel. When [Na+]i is normal to high (10 to 15 mmol/L), however, in addition to slip-mode—dependent Ca2+ entry, 2 mechanisms will contribute to cardiotonic sterioids-increased Ca2+ signaling: the mechanism of Leblanc and Hume21 and the increase in SR Ca2+ content after Na+-pump inhibition.26
Cellular Defects That May Contribute to the Altered Ca2+ Signaling
The efficacy of ICa to activate SR Ca2+ release is reduced in PMI cells. In the absence of changes in other elements in the Ca2+-signaling cascade, a spatial change in the organization of the ECC elements is the simplest explanation. If the L-type Ca2+ channels, on average, were more distant from the RyRs, then the probability that the opening of a Ca2+ channel would activate Ca2+ release would decrease.8 There are several topologically equivalent changes in cellular structures that could lead to such changes: reorganizing the distribution of the L-type calcium channels or the RyRs with respect to each other (“mismatch”), increasing the separation of the transverse tubules from the SR (“gap”), or T-tubule remodeling (“orphan”) (Figure 7). Some recent evidence favors the third possibility.27,28 It is thus possible that the ECC defect reflects T-tubule remodeling that cuts off, or orphans, SR from the triggering Ca2+ signal.
During the early phase of such hypothesized cellular remodeling, the increased distance between triggering Ca2+ influx and RyRs makes activation of SR Ca2+ release less likely. By itself, however, such a change would lead to only transient changes in [Ca2+]i transient amplitude, because higher than normal levels of SR Ca2+ would result.29 Indeed, the orphaned SR phenomenon may also explain the large Ca2+ sparks reported in the early phases of hypertensive cardiomyopathy.30 An agent that increases Ca2+ entry or the sensitivity of the RyRs to be triggered by Ca2+ would effectively reverse or reduce this effect, as has been observed with β-adrenergic receptor activation or ouabain application.
What Is the Initiating Cause of the Ca2+-Signaling Defect?
It is clear from the discussion above that ≥2 defects are involved in the development of the Ca2+-signaling defect in HF: SERCA is expressed at a lower level, and the efficacy of calcium-induced calcium release is reduced. It is not clear which of these 2 events precedes the other or whether there is any causal relationship.
We conclude that PMI cells have a contractile defect that involves altered Ca2+ signaling as a major component. Changes in AP shape may serve as a means of slightly increasing Ca2+ entry and hence contractility, but compensation is not complete. Increased contractions with the cardiotonic steroid ouabain confirm the importance of Ca2+-signaling defects in the contractile dysfunction of the PMI heart. This also suggests that activation of slip-mode conductance of the cardiac Na+ channel by ouabain can improve the Ca2+ signaling in PMI cells and that Na+-pump inhibition produced by ouabain will provide a further beneficial effect that depends on [Na+]i and increased SR Ca2+ content.
This study was supported by the National Institutes of Health, Centre National de la Recherche Scientifique (Dr Gómez), Institut National de la Santé et la Recherche Medicale (INSERM), and Fondation pour la Recherche Medicale (Dr Gómez). We thank Cecilia Frederick for technical support, Sandra Bromble and Michèle Benoit for administrative assistance, and Dr Eric A. Sobie for comments on the manuscript. We thank Mohamed Lemallam and Patrice Bideaux for animal surgery and Dr Jean-Michel Rauzier for echocardiography.
Willerson JT, Cohn JN. Cardiovascular Medicine. 2nd ed. New York, NY: Churchill Livingstone; 2000.
Swynghedauw B. Molecular mechanisms of myocardial remodeling. Physiol Rev. 1999; 79: 215–262.
Lompre AM, Schwartz K, d’Albis A, et al. Myosin isoenzyme redistribution in chronic heart overload. Nature. 1979; 282: 105–107.
Perez NG, Hashimoto K, McCune S, et al. Origin of contractile dysfunction in heart failure: calcium cycling versus myofilaments. Circulation. 1999; 99: 1077–1083.
Benitah JP, Gomez AM, Bailly P, et al. Heterogeneity of the early outward current in ventricular cells isolated from normal and hypertrophied rat hearts. J Physiol. 1993; 469: 111–138.
Benitah JP, Gomez AM, Delgado C, et al. A chloride current component induced by hypertrophy in rat ventricular myocytes. Am J Physiol. 1997; 272: H2500–H2506.
Qi M, Shannon TR, Euler DE, et al. Downregulation of sarcoplasmic reticulum Ca2+-ATPase during progression of left ventricular hypertrophy. Am J Physiol. 1997; 272: H2416–H2424.
Gomez AM, Valdivia HH, Cheng H, et al. Defective excitation-contraction coupling in experimental cardiac hypertrophy and heart failure. Science. 1997; 276: 800–806.
Wessely R, Klingel K, Santana LF, et al. Transgenic expression of replication-restricted enteroviral genomes in heart muscle induces defective excitation-contraction coupling and dilated cardiomyopathy. J Clin Invest. 1998; 102: 1444–1453.
Alvarez JL, Aimond F, Lorente P, et al. Late post-myocardial infarction induces a tetrodotoxin-resistant Na+ current in rat cardiomyocytes. J Mol Cell Cardiol. 2000; 32: 1169–1179.
Gomez AM, Benitah JP, Henzel D, et al. Modulation of electrical heterogeneity by compensated hypertrophy in rat left ventricle. Am J Physiol. 1997; 272: H1078–H1086.
Zarain-Herzberg A, Afzal N, Elimban V, et al. Decreased expression of cardiac sarcoplasmic reticulum Ca2+-pump ATPase in congestive heart failure due to myocardial infarction. Mol Cell Biochem. 1996; 163: 285–290.
Santana LF, Cheng H, Gomez AM, et al. Relation between the sarcolemmal Ca2+ current and Ca2+ sparks and local control theories for cardiac excitation-contraction coupling. Circ Res. 1996; 78: 166–171.
Cheng H, Lederer WJ, Cannell MB. Calcium sparks: elementary events underlying excitation-contraction coupling in heart muscle. Science. 1993; 262: 740–744.
Esposito G, Santana LF, Dilly K, et al. Cellular and functional defects in a mouse model of heart failure. Am J Physiol. 2000; 279: H3101–H3112.
Santana LF, Gomez AM, Lederer WJ. Ca2+ flux through promiscuous cardiac Na+ channels: slip-mode conductance. Science. 1998; 279: 1027–1033.
Arber S, Hunter JJ, Ross J Jr, et al. MLP-deficient mice exhibit a disruption of cardiac cytoarchitectural organization, dilated cardiomyopathy, and heart failure. Cell. 1997; 88: 393–403.
White DC, Hata JA, Shah AS, et al. Preservation of myocardial beta-adrenergic receptor signaling delays the development of heart failure after myocardial infarction. Proc Natl Acad Sci U S A. 2000; 97: 5428–5433.
Minamisawa S, Hoshijima M, Chu G, et al. Chronic phospholamban-sarcoplasmic reticulum calcium ATPase interaction is the critical calcium cycling defect in dilated cardiomyopathy. Cell. 1999; 99: 313–322.
Miyamoto MI, del Monte F, Schmidt U, et al. Adenoviral gene transfer of SERCA2a improves left-ventricular function in aortic-banded rats in transition to heart failure. Proc Natl Acad Sci U S A. 2000; 97: 793–798.
Leblanc N, Hume JR. Sodium current-induced release of calcium from cardiac sarcoplasmic reticulum. Science. 1990; 248: 372–376.
McGarry SJ, Williams AJ. Digoxin activates sarcoplasmic reticulum Ca2+-release channels: a possible role in cardiac inotropy. Br J Pharmacol. 1993; 108: 1043–1050.
Lederer WJ, Niggli E, Hadley RW. Sodium-calcium exchange in excitable cells: fuzzy space. Science. 1990; 248: 283.
Eisner DA, Lederer WJ. The role of the sodium pump in the effects of potassium-depleted solutions on mammalian cardiac muscle. J Physiol. 1979; 294: 279–301.
Arnon A, Hamlyn JM, Blaustein MP. Ouabain augments Ca2+ transients in arterial smooth muscle without raising cytosolic Na+. Am J Physiol. 2000; 279: H679–H691.
Su Z, Sugishita K, Ritter M, et al. The sodium pump modulates the influence of I(Na) on [Ca2+]i transients in mouse ventricular myocytes. Biophys J. 2001; 80: 1230–1237.
Kaprielian RR, Stevenson S, Rothery SM, et al. Distinct patterns of dystrophin organization in myocyte sarcolemma and transverse tubules of normal and diseased human myocardium. Circulation. 2000; 101: 2586–2594.
He J, Conklin MW, Foell JD, et al. Reduction in density of transverse tubules and L-type Ca2+ channels in canine tachycardia-induced heart failure. Cardiovasc Res. 2001; 49: 298–307.
Overend CL, O’Neill SC, Eisner DA. The effect of tetracaine on stimulated contractions, sarcoplasmic reticulum Ca2+ content and membrane current in isolated rat ventricular myocytes. J Physiol. 1998; 507: 759–769.
Shorofsky SR, Aggarwal R, Corretti M, et al. Cellular mechanisms of altered contractility in the hypertrophied heart: big hearts, big sparks. Circ Res. 1999; 84: 424–434.