Alterations of Phospholamban Function Can Exhibit Cardiotoxic Effects Independent of Excessive Sarcoplasmic Reticulum Ca2+-ATPase Inhibition
Background— Low activity of the sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA2a) resulting from strong inhibition by phospholamban (PLN) can depress cardiac contractility and lead to dilated cardiomyopathy and heart failure. Here, we investigated whether PLN exhibits cardiotoxic effects via mechanisms other than chronic inhibition of SERCA2a by studying a PLN mutant, PLNR9C, that triggers cardiac failure in humans and mice.
Methods and Results— Because PLNR9C inhibits SERCA2a mainly by preventing deactivation of wild-type PLN, SERCA2a activity could be increased stepwise by generating mice that carry a PLNR9C transgene and 2, 1, or 0 endogenous PLN alleles (PLN+/++TgPLNR9C, PLN+/−+TgPLNR9C, and PLN−/−+TgPLNR9C, respectively). PLN−/− +TgPLNR9C hearts demonstrated accelerated sarcoplasmic reticulum Ca2+ uptake rates and improved hemodynamics compared with PLN+/++TgPLNR9C mice but still responded poorly to β-adrenergic stimulation because PLNR9C impairs protein kinase A–mediated phosphorylation of both wild-type and mutant PLN. PLN+/++TgPLNR9C mice died of heart failure at 21±6 weeks, whereas heterozygous PLN+/−+TgPLNR9C mice survived to 48±11 weeks, PLN−/−+TgPLNR9C mice to 66±19 weeks, and wild-type mice to 94±27 weeks (P<0.001). Although Ca2+ reuptake kinetics in young PLN−/−+TgPLNR9C mice exceeded those measured in wild-type control animals, this parameter alone was not sufficient to prevent the eventual development of dilated cardiomyopathy.
Conclusions— The data demonstrate an association between the dose-dependent inhibition of SERCA2a activity by PLNwt and the time of onset of heart failure and show that a weak inhibitor of SERCA2a, PLNR9C, which is diminished in its ability to modify the level of SERCA2a activity, leads to heart failure despite fast sarcoplasmic reticulum Ca2+ reuptake.
Received March 31, 2008; accepted November 7, 2008.
Dilated cardiomyopathy (DCM), characterized by left ventricular dilatation and systolic dysfunction, is a leading cause of cardiovascular morbidity and mortality. In ≈30% of all cases, direct relatives of affected individuals also show symptoms of DCM, suggesting a genetic origin of the disease. Most of the disease-causing mutations are located in genes that encode cytoskeletal proteins, but mutations in the sarcomere and nuclear lamin proteins also have been reported. Studies of genetically engineered animal models have identified defects in force transmission, endoplasmic reticulum stress response, apoptosis, and biomechanical stress as important players in the pathogenesis of DCM.1–3
Clinical Perspective p 444
Previously, we described a pathway leading to DCM, heart failure, and premature death that is caused by specific changes in myocyte Ca2+ reuptake resulting from the heterozygous substitution of cysteine for arginine 9 (R9C) in phospholamban (PLN).4 PLN is a small transmembrane protein located in the sarcoplasmic reticulum (SR) that regulates the activity of the sarco(endo)plasmic reticulum Ca2+-ATPase isoform 2a (SERCA2a). Ca2+ transport into the SR by SERCA2a accounts for at least 70% of the removal of Ca2+ from the cytosol during cardiac relaxation. PLN binds to SERCA2a and inhibits its activity by lowering its apparent affinity for Ca2+. As a result, PLN determines myocardial relaxation rates and influences cardiac contractility.5 The inhibitory effect of PLN on SERCA2a is relieved on phosphorylation of PLN at serine 16 by protein kinase A (PKA).6,7 The PLNR9C mutant protein inhibits PKA-mediated PLN phosphorylation of both mutant (PLNR9C) and wild-type (PLNwt) PLN in heterozygous humans and mice, leading to a situation in which more PLN remains in its active inhibitory state, thereby reducing SR Ca2+ uptake rates.4
Depressed SR Ca2+ cycling is commonly observed in failing human myocardia.8 When mimicked in mouse models, for example, by overexpression of PLN, the contractile performance of the heart also is impaired.9 By contrast, PLN-null mice have a nearly normal life span without overt cardiac problems.10,11 A partial or complete reduction in PLN inhibitory activity, which enhances SR Ca2+ uptake and increases cardiac contractility, has been shown to rescue certain mouse models of heart failure.10,12,13 These findings indicate that chronic inhibition or superinhibition of SERCA2a activity by mutant forms of PLN causes heart failure, whereas reduced PLN inhibitory activity may even prevent it. However, the identification of a human PLNL39stop mutation as a causal factor in recessively inherited DCM suggests that loss of PLN may be lethal in humans because the mutation led to a truncated form of PLN that was equivalent to a null allele.14 This discrepancy between humans and mice can be rationalized by the differences in human and mouse cardiac physiology in that slow-beating human hearts may be more dependent on PLN inhibitory function than mouse hearts, the latter constantly operating close to their maximal heart rate.14 A second possible explanation could be that parameters of PLN function other than high PLN inhibitory activity may contribute to the development of heart failure.
The goal of the present study was to further define the relations between PLN malfunction and the development of heart failure. The level of PLN inhibitory activity was decreased stepwise in transgenic mice expressing the PLNR9C mutation by ablation of 1 or both PLNwt alleles. We report here that the time of onset of lethal DCM depends on the gene dosage of the PLNwt gene and the rate of SR Ca2+ uptake in hearts expressing PLNR9C. However, the PLNR9C defect is more complicated because detrimental remodeling processes were triggered, albeit at a later stage, in the absence of wild-type PLN and in the presence of enhanced Ca2+ reuptake rates. Analyses in vitro and in vivo suggest that the inability to modulate the activity of the Ca2+-ATPase appropriately may provide an independent stimulus for the development of heart failure.
Mice expressing PLNR9C under the control of the cardiac α-myosin heavy chain promoter (PLN+/++TgPLNR9C mice)4 were crossed with PLN-ablated mice (PLN−/− mice; a generous gift from Dr Evangelia G. Kranias, Department of Pharmacology and Cell Biophysics, University of Cincinnati, Cincinnati, Ohio)10 to generate transgenic mice that carry 1 (PLN+/−+TgPLNR9C) or 0 (PLN−/−+TgPLNR9C) endogenous PLN alleles. Contractile properties of mouse hearts were investigated by in vivo left ventricular catheterization and echocardiography.15 Ca2+ kinetics and sarcomere mechanics were measured in isolated myocytes.16 Ca2+ transport assays were carried out in microsomal preparations of transfected HEK-293 cells.17 See the online-only Data Supplement for an expanded description.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Ablation of Endogenous PLN Prevents DCM and Heart Failure in PLNR9C Transgenic Mice
Mice overexpressing a PLNR9C transgene under the control of the cardiac α-myosin heavy chain promoter (PLN+/+ +TgPLNR9C) progressively developed DCM and terminal heart failure.4 Twenty weeks after birth, these hearts showed massive dilatation of ventricular chambers and enlarged atria (Figure 1A). Cardiac histology was characterized by irregularly shaped nuclei, myocyte enlargement, and substantial interstitial collagen depositions. Lungs and livers of PLN+/+ +TgPLNR9C mice were enlarged and dark red, and histology revealed signs of vascular congestion, indicating failure of the left and right ventricles (Figure 1B). In contrast, the size and histology of the heart, lungs, and liver were regular in PLN−/− +TgPLNR9C mice at 20 weeks of age (Figure 1A and 1B).
Twenty-week-old PLN+/++TgPLNR9C and PLN−/− +TgPLNR9C mice were further evaluated for hemodynamic function using left ventricular catheterization (Figure 1C). Rates of pressure change (dP/dtmax and dP/dtmin) and maximum left ventricular pressures were distinctly depressed in PLN+/++TgPLNR9C hearts compared with wild-type hearts, indicating heart failure. However, the performance of PLN−/− +TgPLNR9C hearts closely resembled that of wild-type hearts. PLN−/−+TgPLNR9C hearts exhibited significantly higher rates of pressure change than PLN+/++TgPLNR9C hearts. Heart rates were similar in all mouse lines.
Development of Left Ventricular Dilatation and Contractile Dysfunction Resulting From PLNR9C Depends on the Gene Dosage of PLNwt
We also analyzed mice bearing the PLNR9C transgene and only 1 PLNwt allele (PLN+/−+TgPLNR9C) to further investigate the relation between PLNwt gene dosage and the development of DCM and decreased cardiac contractile function. As analyzed by echocardiography of 26-week-old mice, fractional shortening was lower by 35% in PLN+/++TgPLNR9C hearts (P<0.001), by 33% in PLN+/−+TgPLNR9C hearts (P<0.001), and by 18% in PLN−/−+TgPLNR9C hearts (P<0.001) compared with wild-type mice (Figure 2). Left ventricular end-systolic diameter had increased 3.4-fold in PLN+/+ +TgPLNR9C (P<0.01), 2.8-fold in PLN+/−+TgPLNR9C mice (P<0.001), and 1.9-fold in PLN−/−+TgPLNR9C mice (P<0.01). Apparently, all mouse lines bearing the PLNR9C transgene had developed contractile dysfunction and dilatation of left ventricles at 26 weeks of age, the extent correlating with the dosage of PLNwt. Fractional shortening was moderately decreased even in PLN−/−+TgPLNR9C compared with wild-type mice as a result of mild left ventricular dilatation. In no case was left ventricular wall thickness increased over wild-type values, indicating a direct progression to pure DCM (see also Reference 4).
Protein levels of SERCA2a and other Ca2+ handling proteins were shown to be equal in all mouse lines at 10 weeks of age. PLNR9C expression was estimated to be similar or slightly higher than PLNwt expression in PLN+/+ +TgPLNR9C hearts (Figure IA of the online-only Data Supplement). Phosphorylation of PLNwt was ≈30% lower in PLN+/++TgPLNR9C and PLN+/−+TgPLNR9C mice than in wild-type mice, demonstrating the inhibitory effect of PLNR9C on phosphorylation of PLNwt. The phosphorylation status of PKA target proteins other than PLN such as the ryanodine receptor (RyR2) and troponin I did not change significantly (Figure IA, online-only Data Supplement).
Enhanced SR Ca2+ Uptake Rates and Regular Myocyte Mechanics in PLN−/−+TgPLNR9C Mice
In view of decreased PLNwt phosphorylation in the presence of PLNR9C, which reduces SR Ca2+-ATPase activity,4 we wondered whether the abnormal Ca2+ kinetics of myocytes expressing PLNR9C were normalized when PLNwt was ablated. Therefore, SR Ca2+ transients were measured in isolated ventricular myocytes taken from 8- to 10-week-old mice and electrically stimulated at 0.5 Hz (Figure 3A). In our experiments, the rate of SR Ca2+ release (T80% peak) and the amplitude of the Ca2+ transient were not significantly different among PLN+/++TgPLNR9C, PLN−/−, PLN−/−+TgPLNR9C, and wild-type cells. SR Ca2+ contents were estimated from the amount of Ca2+ release on application of 10 mmol/L caffeine. Measurements showed a trend but no significant difference for higher SR Ca2+ loads in the absence of PLNwt. The return of Ca2+ to the SR via SERCA2a (T50% baseline and T90% baseline) was slower in PLN+/++TgPLNR9C cells than in wild-type cells (P<0.01). However, in PLN−/− myocytes, SR Ca2+ uptake was accelerated in both the presence and absence of PLNR9C (P<0.01). Intriguingly, in PLN−/−+TgPLNR9C myocytes, Ca2+ uptake rates were faster than in wild-type myocytes (P<0.01). We conclude that SR Ca2+ uptake rates are slow in myocytes expressing PLNR9C only in the presence of PLNwt.
In parallel with measurements of SR Ca2+ transients, the rates of contraction and relaxation were recorded with the edge detection method to investigate the mechanical consequences of altered Ca2+ kinetics (Figure 3B). Cell length was similar in all groups. However, the PLNR9C transgene decelerated myocyte relaxation compared with wild-type myocytes (P<0.05), but in the absence of endogenous PLN (PLN−/−+TgPLNR9C), myocyte relaxation was not significantly different from that of wild-type cells. Contractile speed was equal in wild-type and PLN+/++TgPLNR9C myocytes but slightly faster in PLN−/− and PLN−/−+TgPLNR9C myocytes (P<0.01), presumably because of enhanced restocking of cellular Ca2+ stores in the PLN-null background. The data show impaired relaxation rates of PLNR9C myocytes, which are rescued by ablation of PLNwt (PLN−/−+TgPLNR9C).
Ca2+ Transport Rates of HEK-293 Cells Transfected With PLNR9C Depend on the Amount of Coexpressed PLNwt
Ca2+ uptake rates also were measured in microsomal preparations from HEK-293 cells transfected with SERCA2a, PLNwt, and PLNR9C. These preparations permitted selective analysis of the inhibitory activities of PLNR9C and PLNwt in an unphosphorylated state. At various Ca2+ concentrations <10 μmol/L, the activity of SERCA2a was higher in cells expressing SERCA2a and PLNR9C than in cells expressing SERCA2a and PLNwt (Figure 3C). EC50 values were 200±13 nmol/L for SERCA2a alone and 532±37 nmol/L for SERCA2a+PLNwt but only 248±12 nmol/L for SERCA2a+PLNR9C (P<0.001). On cotransfection of equal amounts of PLNR9C and PLNwt cDNA, the Ca2+ affinity of SERCA2a was almost equal to that found for cotransfection with PLNwt alone (Figure 3D). The results indicate that PLNR9C alone exhibits little inhibitory effect on SERCA2a or, in the absence of kinases, on PLN-dependent SERCA2a regulation.
Next, HEK-293 cells were transfected with SERCA2a and varying ratios of PLNwt and PLNR9C (Figure 3E). For expression of SERCA2a alone, the EC50 for CA2+ uptake was 6.7 pCa units but was reduced by 0.42 pCa units in the presence of PLNwt. In the presence of PLNR9C, EC50 was 6.61, for a decrease of only 0.09 pCa units. If the ratio of PLNwt to PLNR9C was 2:6, EC50 was reduced by 0.24 pCa units, and if the ratio of PLNwt to PLNR9C was 4:4, EC50 was reduced by 0.387 pCa units. We conclude that the inhibitory effects of unphosphorylated PLNR9C and PLNwt are additive and depend largely on the amount of coexpressed PLNwt. By extrapolation, in the presence of equal amounts of PLNwt and PLNR9C, as occurs in heterozygous humans with DCM, PLNwt function is likely to predominate and, in an unphosphorylated state, would be almost as strong an inhibitor in the presence of an equal amount of PLNR9C as in its absence (inhibitory effect, −0.38 pCa units versus −0.42 pCa units in HEK-293 cells). As further illustrated in Figure 3E, relative loss of function resulting from the R9C mutation was equivalent in murine and human PLN, the latter carrying an asparagine instead of a lysine in position 27 and known to exhibit a stronger inhibitory effect on SERCA2a than PLN of other species.18
Late-Onset DCM, Decreased Cardiac Contractile Function, and Death in PLN−/−+TgPLNR9C Mice
The fast Ca2+ reuptake kinetics in the absence of PLNwt provided a molecular explanation for the improved heart phenotype of PLN−/−+TgPLNR9C mice. Intriguingly, at 60 weeks of age, all cardiac chambers of these mice were bigger than in age-matched wild-type hearts (Figure 4A). Histological sections revealed enlarged myocytes, nuclei with irregular size and shape, and interstitial fibrosis, indicating that although PLNR9C promotes DCM significantly via PLNwt, PLNR9C on its own also can trigger adverse cardiac remodeling processes.
The direct contribution of PLNR9C to the structural and functional decline was further analyzed by echocardiographic comparison of PLN−/−+TgPLNR9C and PLN−/− mice (Figure 4B). Fractional shortening of PLN−/−+TgPLNR9C hearts at 60 weeks of age was only 46% compared with 69% in PLN−/− hearts (P=0.01). Left ventricular dimensions of PLN−/− +TgPLNR9C hearts were higher than in PLN−/− mice at both end diastole (34% higher; P<0.01) and end systole (135% higher; P<0.01). Compared with 26-week-old mice (Figure 2), left ventricular end-diastolic and end-systolic diameters of PLN−/−+TgPLNR9C hearts had increased by 13% (P<0.05) and 73% (P<0.01), respectively, and fractional shortening had declined by 29% (P<0.01), indicating the development of left ventricular dilatation and contractile dysfunction.
Invasive measurements of left ventricular hemodynamics demonstrated normal left ventricular maximal pressures in 1-year-old PLN−/−+TgPLNR9C mice (Figure 4C). However, rates of pressure rise and pressure fall (dP/dtmax and dP/dtmin) were depressed compared with age-matched PLN−/− mice; differences in dP/dtmax and heart rate became significant only after administration of dobutamine (P<0.05). This depression of cardiac hemodynamics, along with a poor response to β-adrenergic stimulation, indicates that PLN−/− +TgPLNR9C hearts, but not PLN−/− hearts, are failing at 1 year of age.
PLN−/−+TgPLNR9C mice died at an average age of 66±17 weeks (average±SD) (Figure 5). On autopsy, chests were filled with clear fluid, lungs were dark red, and hearts showed massive dilatation consistent with symptoms of terminal heart failure. PLN−/−+TgPLNR9C mice lived ≈3 times longer than PLN+/++TgPLNR9C mice (23±6 weeks) but clearly shorter than wild-type FVB mice (94±27 weeks; P<0.001). Mice carrying the PLNR9C transgene and only 1 endogenous PLN allele (PLN+/−+TgPLNR9C) lived for 48±11 weeks. This was shorter than the mean life span of PLN−/−+TgPLNR9C mice but longer than that of PLN+/++TgPLNR9C mice (P<0.001).
PLN−/−+TgPLNR9C Mice Do Not Respond to β-Adrenergic Stimulation
In the absence of PLNwt, PLNR9C still causes late-onset heart failure, despite fast Ca2+ reuptake kinetics. Therefore, PLNR9C alone must damage the myocardium by mechanisms that are independent of PLNwt and independent of a high degree of SERCA2a inhibition. Because the R9C mutation inhibits PKA-mediated phosphorylation of serine 16, we further studied the physiological effects of β-adrenergic stimulation on PLN−/−+TgPLNR9C myocytes compared with wild-type cells (see Figure 3A and 3B for basal cardiomyocyte function). β-Adrenergic agonists normally lead to phosphorylation of PKA target proteins such as L-type Ca2+ channels, RyR2, and PLN. Consequently, myocyte contraction and relaxation are accelerated by enhanced intracellular Ca2+ kinetics. Ca2+ transients of isolated myocytes were measured after addition of isoproterenol to the perfusion buffer compared with unstimulated cells. In our experiments, the increases on isoproterenol stimulation in both the amount (Δamplitude [iso−basal]) and the rate (ΔT80% peak [iso−basal]) of SR Ca2+ release were not different in PLN−/−+TgPLNR9C and wild-type cells (Figure 6A). Notably, the speed of SR Ca2+ reuptake was accelerated significantly in wild-type cells on adrenergic activation (decrease in T90% baseline from 192±7 to 140±14 ms; P<0.01) and not in PLN−/−+TgPLNR9C myocytes (decrease in T90% baseline from 130±6 to 126±6 ms; P=NS).
Accordingly, isoproterenol was able to enhance myocyte relaxation rates, which depend largely on the level of PLN activity, in wild-type cells (P<0.01; Figure 6B) but not in PLN−/−+TgPLNR9C myocytes (P=NS). The amplitude and speed of myocyte contraction increased in all cells on isoproterenol stimulation. These results would most likely be due to the effects of PKA phosphorylation on L-type Ca2+ channels or RyR2 and might reflect an increased size of the Ca2+ store resulting from enhanced SERCA2a activity on stimulation. The amplitude and speed of myocyte contraction were not significantly different in wild-type and PLN−/− +TgPLNR9C myocytes, suggesting regular phosphorylation of Ca2+ release channels.
The response to β-adrenergic stimulation was further evaluated in vivo by invasive left ventricular catheterization of 20-week-old PLN−/−+TgPLNR9C mice compared with wild-type mice (Figure 6C). At this age, PLN−/−+TgPLNR9C hearts showed no signs of heart failure at baseline (Figure 1). However, PLN−/−+TgPLNR9C hearts failed to accelerate the speed of pressure decay (dP/dtmin) during infusion of dobutamine (1.5 μg/min): dP/dtmin fell by 24±10% in wild-type hearts (P<0.01) but only by 5±3% in PLN−/−+TgPLNR9C hearts (P=NS). The limited acceleration of SR Ca2+ uptake rates in PLN−/−+TgPLNR9C myocytes on stimulation with isoproterenol (Figure 6A) and the lack of a lusitropic effect in PLN−/−+TgPLNR9C myocytes and whole hearts (Figure 6B and 6C) are explained by the resistance of PLNR9C to PKA-mediated phosphorylation and by the limited inhibitory effect of PLNR9C itself on the Ca2+-ATPase.
We have shown previously that PLNR9C leads to DCM, heart failure, and premature death in heterozygous humans and transgenic mice (PLN+/++TgPLNR9C).4 PLNR9C prevents phosphorylation of coexpressed PLNwt and thereby decelerates diastolic Ca2+ transport into the SR via SERCA2a (Figure 3A). Unlike the nearly normal inhibitory effect that is measured in transfected HEK-293 cells when PLNR9C is present together with PLNwt, SERCA2a inhibition by PLNR9C alone is weak (Figure 3C and 3E). Thus, cardiomyocyte expression of PLNR9C alone (PLN−/−+TgPLNR9C) results in SR Ca2+ uptake rates that are faster than those in wild-type myocytes and almost as fast as those in PLN−/− myocytes (Figure 3A and 3C). The acceleration of SR Ca2+ reuptake kinetics as a result of PLNwt ablation was accompanied by an improved morphological and functional phenotype of the hearts (Figures 1 and 2⇑); lower expression of atrial natriuretic peptide, brain natriuretic peptide, and β-myosin heavy chain (Figure IC of the online-only Data Supplement); and a 3-fold-longer survival of PLN−/−+TgPLNR9C mice compared with PLN+/++TgPLNR9C mice (Figure 5). These data provide evidence that the dominant effect of PLNR9C on PLNwt plays an important role in the progression of heart failure when both forms are present.
The pronounced benefit to the life span of PLNR9C transgenic animals that arises from ablation of PLN confirms that the ratio of SERCA2a to PLN activity is a promising target for the prevention of DCM and heart failure. PLN ablation and SERCA2a overexpression have been reported to improve the structural and/or functional phenotype in several experimental models of heart failure. Examples are mice lacking the muscle-specific LIM protein and mice with transgenic overexpression of the Ca2+ storage protein calsequestrin or the β1-adrenergic receptor.12,13,19 On the other hand, PLN ablation had little or no effect on left ventricular dysfunction in myosin-binding protein C–deficient mice or in mice overexpressing the sarcomeric protein tropomodulin.20,21 In our mouse model, PLN ablation delayed both structural and functional signs of heart failure by several months and tripled the survival time, on average, from ≈21 weeks to ≈66 weeks, suggesting that PLN ablation is a highly efficient intervention for delaying the onset of heart failure that arise in full or in part from Ca2+ dysregulation. Importantly, even a partial mending of the underlying Ca2+ cycling defect, as in PLNR9C transgenic mice with only 1 PLNwt allele, largely improved the phenotype (Figure 5).
PLN−/−+TgPLNR9C mice demonstrated faster Ca2+ uptake rates than wild-type mice (Figure 3A and 3C). According to current theories, fast Ca2+ reuptake does not lead to heart failure but may even prevent it.12,22 However, PLN−/− +TgPLNR9C mice progressively developed DCM during adulthood and eventually died of heart failure at 66±19 weeks of age (Figures 4 and 5⇑). We conclude that the PLNR9C mutation promotes heart failure via at least 2 distinct mechanisms. One is chronic slow SR Ca2+ uptake, which is crucial to heterozygous humans and PLN+/++TgPLNR9C mice. A second pathway that causes heart failure in PLN−/− +TgPLNR9C mice is independent of SR Ca2+ uptake rates. A mechanism that is independent from SERCA2a function cannot be ruled out but appears unlikely because PLN specifically interacts with SERCA2a and the phosphorylation defect induced by PLNR9C seems to be confined to PLN (Figure IA, online-only Data Supplement).
In PLN−/−+TgPLNR9C mice, PLNR9C inhibits SR Ca2+ reuptake less than PLNwt, but PLNR9C also blocks the capacity to alter the inhibitory activity of PLNwt and PLNR9C via PKA-mediated phosphorylation (Figure 6). Consequently, cardiac relaxation, which largely determines diastolic filling and cardiac stroke volume, remains fixed at a certain rate. Therefore, we propose that the resulting inability of PLN−/− +TgPLNR9C hearts to adjust adequately to the changing requirements during low and high activity of the heart may act as an independent trigger for heart failure. As illustrated in our working model, diagrammed in Figure 7, healthy myocytes alter Ca2+ transport velocities into the SR by phosphorylation and dephosphorylation of PLN (Figure 7A). In contrast, this regulatory mechanism is impaired in myocytes harboring PLNR9C because PLNR9C is not phosphorylated and suppresses the phosphorylation of PLNwt in a dominant fashion. Thus, myocyte Ca2+ reuptake will always be slow in PLN+/++TgPLNR9C hearts (Figure 7B). In unstimulated PLN−/−+TgPLNR9C hearts, Ca2+ reuptake rates are faster, lying midway between wild-type and PLN−/− rates (Figure 3), and they are not stimulated by isoproterenol (Figure 6). These abnormalities in Ca2+ uptake lead to dilated cardiomyopathy (Figure 7C). In a hypothetical scenario, intracellular Ca2+ sensors such as calmodulin would detect abnormalities of cytosolic Ca2+ and subsequently activate target proteins, including the calcium/calmodulin-dependent protein kinase, the myosin light-chain kinase, the protein kinase C, and the phosphatase calcineurin. These enzymes control classic downstream remodeling pathways that, on long-term stimulation, can produce phenotypic features of DCM and heart failure, as observed in PLN−/−+TgPLNR9C mice.
Murine PLN-null hearts (PLN−/−) cannot regulate Ca2+ uptake rates but nevertheless do not display overt heart failure and are even hypercontractile as a result of enhanced SR Ca2+ transport. Given the related Ca2+ kinetics of PLN−/− and PLN−/−+TgPLNR9C mice (Figure 3A), similar phenotypes would be expected for mice with either genotype. The fact that they are not identical suggests that PLNR9C adds a dimension of abnormal regulation that we do not yet fully understand. This type of regulation might be linked to problems that occur in humans with nonfunctional PLN. In 2 large human families, a heterozygous PLN39stop/+ mutation produced a truncated and unstable protein and showed no SERCA2a inhibition in transfected HEK293 cells or adult rat cardiomyocytes.14 One third of the older family members bearing this loss-of-function mutation developed left ventricular hypertrophy, and the inheritance of 2 mutant copies led to DCM and heart failure in 2 teenage siblings.
We have shown that the rate of diastolic Ca2+ kinetics, as determined by the ratio of SERCA2a to PLN activity, is a crucial feature in the maintenance of a healthy heart. However, chronic alterations of PLN function can also foster the development of heart failure at high Ca2+ transport rates. According to our current understanding, heart failure patients may benefit from strategies that enhance SR Ca2+ uptake. Nevertheless, mere inactivation of PLN to achieve this goal may be a double-edged sword because complete lack of SERCA2a regulation might eventually promote cardiac failure. Our model proposes that therapeutic concepts aimed at myocyte Ca2+ kinetics also need to restore the capability of the myocardium to regulate the speed of SR Ca2+ uptake—both up and down—as is accomplished by a properly functioning PLN that is responsive to phosphorylation.
We thank M. Babl and B. Thur for technical assistance with hemodynamic measurements and genotyping of mice.
Sources of Funding
This work was supported by the Henrietta and Frederick Bugher Fund, the Howard Hughes Medical Institute, the National Institutes of Health, the Deutsche Stiftung für Herzforschung, grant T-5042 from the Heart and Stroke Foundation of Ontario, and grant MT-12545 from the Canadian Institutes of Health Research.
Schmitt JP, Debold EP, Ahmad F, Armstrong A, Frederico A, Conner DA, Mende U, Lohse MJ, Warshaw D, Seidman CE, Seidman JG. Cardiac myosin missense mutations cause dilated cardiomyopathy in mouse models and depress molecular motor function. Proc Natl Acad Sci U S A. 2006; 103: 14525–14530.
Gramolini AO, Kislinger T, Alikhani-Koopaei R, Fong V, Thompson NJ, Isserlin R, Sharma P, Oudit GY, Trivieri MG, Fagan A, Kannan A, Higgins DG, Huedig H, Hess G, Arab S, Seidman JG, Seidman CE, Frey B, Perry M, Backx PH, Liu PP, MacLennan DH, Emili A. Comparative proteomics profiling of a phospholamban mutant mouse model of dilated cardiomyopathy reveals progressive intracellular stress responses. Mol Cell Proteomics. 2008; 7: 519–533.
Schmitt JP, Kamisago M, Asahi M, Li GH, Ahmad F, Mende U, Kranias EG, MacLennan DH, Seidman JG, Seidman CE. Dilated cardiomyopathy and heart failure caused by a mutation in phospholamban. Science. 2003; 299: 1410–1413.
Luo W, Chu G, Sato Y, Zhou Z, Kadambi VJ, Kranias EG. Transgenic approaches to define the functional left ventricular role of dual site phospholamban phosphorylation. J Biol Chem. 1998; 273: 4734–4739.
Luo W, Grupp IL, Harrer J, Ponniah S, Grupp G, Duffy JJ, Doetschman T, Kranias EG. Targeted ablation of the phospholamban gene is associated with markedly enhanced myocardial contractility and loss of beta-agonist stimulation. Circ Res. 1994; 75: 401–409.
Minamisawa S, Hoshijima M, Chu G, Ward CA, Frank K, Gu Y, Martone ME, Wang Y, Ross J Jr, Kranias EG, Giles WR, Chien KR. Chronic phospholamban-sarcoplasmic reticulum calcium ATPase interaction is the critical calcium cycling defect in dilated cardiomyopathy. Cell. 1999; 99: 313–322.
Sato Y, Kiriazis H, Yatani A, Schmidt AG, Hahn H, Ferguson DG, Sako H, Mitarai S, Honda R, Mesnard-Rouiller L, Frank KF, Beyermann B, Wu G, Fujimori K, Dorn GW 2nd, Kranias EG. Rescue of contractile parameters and myocyte hypertrophy in calsequestrin overexpressing myocardium by phospholamban ablation. J Biol Chem. 2001; 276: 9392–9399.
Haghighi K, Kolokathis F, Pater L, Lynch RA, Asahi M, Gramolini AO, Fan GC, Tsiapras D, Hahn HS, Adamopoulos S, Liggett SB, Dorn GW 2nd, MacLennan DH, Kremastinos DT, Kranias EG. Human phospholamban null results in lethal dilated cardiomyopathy revealing a critical difference between mouse and human. J Clin Invest. 2003; 111: 869–876.
McConnell BK, Fatkin D, Semsarian C, Jones KA, Georgakopoulos D, Maguire CT, Healey MJ, Mudd JO, Moskowitz IP, Conner DA, Giewat M, Wakimoto H, Berul CI, Schoen FJ, Kass DA, Seidman CE, Seidman JG. Comparison of two murine models of familial hypertrophic cardiomyopathy. Circ Res. 2001; 88: 383–389.
Toyofuku T, Kurzydlowski K, Tada M, MacLennan DH. Amino acids Glu2 to Ile18 in the cytoplasmic domain of phospholamban are essential for functional association with the Ca(2+)-ATPase of sarcoplasmic reticulum. J Biol Chem. 1994; 269: 3088–3094.
Zhao W, Yuan Q, Qian J, Waggoner JR, Pathak A, Chu G, Mitton B, Sun X, Jin J, Braz JC, Hahn HS, Marreez Y, Syed F, Pollesello P, Annila A, Wang HS, Schultz Jel J, Molkentin JD, Liggett SB, Dorn GW 2nd, Kranias EG. The presence of Lys27 instead of Asn27 in human phospholamban promotes sarcoplasmic reticulum Ca(2+)-ATPase superinhibition and cardiac remodeling. Circulation. 2006; 113: 995–1004.
Engelhardt S, Hein L, Dyachenkow V, Kranias EG, Isenberg G, Lohse MJ. Altered calcium handling is critically involved in the cardiotoxic effects of chronic beta-adrenergic stimulation. Circulation. 2004; 109: 1154–1160.
Song Q, Schmidt AG, Hahn HS, Carr AN, Frank B, Pater L, Gerst M, Young K, Hoit BD, McConnell BK, Haghighi K, Seidman CE, Seidman JG, Dorn GW 2nd, Kranias EG. Rescue of cardiomyocyte dysfunction by phospholamban ablation does not prevent ventricular failure in genetic hypertrophy. J Clin Invest. 2003; 111: 859–867.
Iwanaga Y, Hoshijima M, Gu Y, Iwatate M, Dieterle T, Ikeda Y, Date MO, Chrast J, Matsuzaki M, Peterson KL, Chien KR, Ross J Jr. Chronic phospholamban inhibition prevents progressive cardiac dysfunction and pathological remodeling after infarction in rats. J Clin Invest. 2004; 113: 727–736.
This study shows that correcting abnormal myocyte Ca2+ flow can serve as a powerful tool against heart failure. Intracellular Ca2+ cycling, which enables the heart muscle to contract and relax, has been known for many years to be disrupted in heart failure. Although it has been unclear whether the disturbance is contributing to cardiac dysfunction or a secondary consequence, the identification of a rare phospholamban mutation (PLNR9C) provided the first evidence that altered Ca2+ flow alone is sufficient to incite cardiac remodeling processes that lead to dilated cardiomyopathy and terminal heart failure in humans. Transgenic mice with cardiac expression of PLNR9C also develop dilated cardiomyopathy and die prematurely; counteracting the slowing of sarcoplasmic reticulum Ca2+ uptake in such mice by releasing sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA2a) via ablation of endogenous PLN (while maintaining PLNR9C expression) dose dependently delays the onset of heart failure and prolongs survival up to 3-fold. However, even sarcoplasmic reticulum Ca2+ uptake rates that are faster than normal are not sufficient to prevent the development of late-onset heart failure in these mice. Thus, 2 independent principles of PLN dysfunction can trigger the development of heart failure. One is excessive SERCA2a inhibition. Because PLNR9C is phosphorylation deficient, the second may be the inability to turn SERCA2a activity up and down by phosphorylation and dephosphorylation of PLN. Apparently, increasing the ratio of SERCA2a to PLN is a potent target for the treatment of heart failure that is caused by Ca2+ defects, but such strategies may be most successful if they allow efficient regulation of PLN activity.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.108.783506/DC1.
Guest Editor for this article was Mark A. Sussman, PhD.