Toward Biologically Targeted Therapy of Calcium Cycling Defects in Heart Failure

Yasuhiro Ikeda, Masahiko Hoshijima, Kenneth R. Chien


A growing body of evidence indicates that heart failure progression is tightly associated with dysregulation of phosphorylation of Ca2+ regulators localized in the sub-cellular microdomain of the sarcoplasmic reticulum. Chemical or genetic correction of abnormalities in cardiac phosphorylation cascades is emerging as a potential target in the treatment of heart failure. Here, we review how specific kinases and phosphatases finely tune Ca2+ cycling and regulate excitation-contraction (E-C) coupling in cardiomyocytes.

Calcium cycling defects have emerged as a critical form of molecular dysfunction that directly contributes to contractile dysfunction and associated malignant arrhythmia in failing hearts (2022, 45, 49, 108). Periodic cytosolic Ca2+ oscillations, named Ca2+ transients, fundamentally govern contraction and relaxation in cardiomyocytes. Briefly, membrane depolarization triggers Ca2+ influx through L-type Ca2+ channels (LTCC), followed by a Ca2+-induced Ca2+ release (CICR) through the ryanodine receptors (RyR) (6). Although the LTCC is being rapidly inactivated by the local elevation of Ca2+ and sustained membrane depolarization, transiently induced cytosolic Ca2+ binds to the thin filament troponin C and activates the contractile machinery. As sarcoplasmic retiulum (SR) Ca2+ content is being depleted, RyRs are inactivated, switching the cardiomyocyte from a systolic to diastolic mode. During the diastolic period, cytosolic Ca2+ ions are resequestered into the SR by SERCA2a or pumped out from the cell by the sarcolemmal Na+-Ca2+ exchanger (NCX). The relative contribution of SERCA2a and NCX to Ca2+ efflux depends on the animal species. Ca2+ uptake through the SERCA pump is negatively regulated by phospholamban (PLN). Ca2+-dependent intracellular Ca2+ homeostasis through these nodal components has been extensively reviewed elsewhere (6).

“Yin-Yang” Regulation of Ca2+ Cycling by Kinases and Phosphatases

The amplitude and velocity of the Ca2+ transient is regulated by phosphorylation of Ca2+ cycling regulators in cardiomyocytes. In this regard, two kinases, namely protein kinase A (PKA) and Ca2+/calmodulin-dependent protein kinase II (CaMK II), play a major role in dynamic balance with phosphatases to control local phosphorylation of specific targets with the sarcoplasmic reticulum (FIGURE 1).


Regulation of Ca2+ cycling in cardiomyocytes
The amplitude and velocity of the Ca2+ transient is regulated by phosphorylation of nodal Ca2+ cycling regulators in cardiomyocytes. Two kinases, namely protein kinase A (PKA) and Ca2+/calmodulin-dependent protein kinase II (CaMK II), play a major role along with phosphatases to control local phosphoreactions. Protein phosphatase 1 and 2A dephosphorylate target substrates, thereby negatively regulating Ca2+ cycling. β-Adrenergic receptor signaling activates PKA and CaMK signaling, whereas Gq-receptor-coupled PKCα signaling controls PP1-associated regulatory signaling.

The upstream activator of PKA in the cardiac signal cascade is cyclic AMP (cAMP). Briefly, an increase in adrenalin activates the β-adrenergic receptor, which results in coupling of the α subunit of Gs protein to adenylate cyclase (AC) and in turn increases AC activity, followed by an increase in cAMP and the activation of PKA in subcellular microdomains. Sustained activation of the β-receptor results in receptor phosphorylation by β-adrenergic kinase and is processed toward inactivation (desensitization) (30). PKA phosphorylates at least three key Ca2+ regulators, i.e., LTCC, RyR, and PLN, which enhances Ca2+ influx, increases the velocity and the amplitude SR Ca2+ release, and augments SERCA pump activity, respectively. PKA, thus, is a major regulator of cardiac contractile and relaxation functions (108). Intriguingly, these three PKA targets of Ca2+ cycling regulation are also major cardiac substrates of CaMK II. The major isoform of cardiac CaMK II is IIδ (31) splice variants of which shares distinctive subcellular localizations; i.e., IIδC is in the cytosol, whereas IIδB localizes in the nuclei due to its nuclear localization signal near its amino terminal (31). CaMK II regulates cardiac Ca2+ cycling throughout the phosporylation of LTCC, RyR, and PLN; CaMK II-dependent phosphorylation of the sodium channel (95) and the potassium channel (92) indirectly modulates cardiomyocyte excitation-contraction (E-C) coupling. Presumably, CaMK II is regulated by Ca2+ oscillation itself (65, 114); however, it is not entirely clear which component of Ca2+ regulates CaMK II activities in cardiomyocytes and how.

Protein phosphorylation is a dynamic and reversible process of protein modification. To maintain cellular homeostasis, proteins phosphorylated by PKA or CaMK II are actively dephosphorylated by phosphatases, and Ca2+ cycling returns to the baseline state. Although the physiological roles of protein kinases have been extensively characterized, details of cardiac protein phosphatases have been poorly understood until recently. The major isotypes of cardiac phosphatases are type 1 (PP1) and 2A (PP2A), both of which together comprise more than 90% of the protein phosphatases (63) in cardiomyocytes. Both types of phosphatases affect E-C coupling. Although PP1 and PP2A share highly similar sequences in the catalytic domain, PP1 appears to have a higher affinity for the dephosphorylation of Ca2+ cycling regulators including LTCC, RyR, and PLN (64), whereas PP2A has been mainly attributed to the dephosphorylation of myofibrillar proteins such as troponin I and myosin binding protein C (69).

In the failing hearts, it has increasingly become clear that altered phosphorylation level of these phosphoproteins by kinases/phosphatases is initially driven by a compensatory mechanism to adjust cardiac output to meet metabolic demands of the body, and it may eventually become maladaptive and further lead to deleterious Ca2+ homeostasis (45, 108). Therefore, molecular targeting of these phospho-regulatory proteins as well as these kinase/phosphatases can be potentially therapeutic targets for heart failure, as described below.

Compartmentation of the PKA Signal Cascade

Phosphorylation of Ca2+ cycling in cardiomyocytes is governed by a complex intracellular spatial and temporal set of signals, and defects in their compartmentation seem to play a significant role in cardiac disease. PKA signaling is compartmented at multiple levels in a cascade, i.e., through isoform-specific subcellular localization of G-protein coupling (30), adenylate cyclases (35), PKA anchoring proteins (27), and cyclic nucleotide phosphodiesterases (35, 111). Compartmentation of cardiac membrane receptor signaling and its physiological significance are becoming ever more recognizable (52, 91). A typical example of receptor compartmentation has been found in the β1–AR and the β2-AR (15, 103). Both β1-AR and β2-AR couple with the Gαs -cAMP-PKA cascade, and it has been well documented, during progression of heart failure, that expression patterns and ratios of β 1-AR and β2-AR change. Intriguingly, β1-AR activation leads to inotropic and lusitropic responses; However, β2-AR activation enhances the amplitude of the Ca2+ transient and twitch in cardiomyocytes without an acceleration in relaxation. β2-AR activation indeed fails to induce phosphorylation of PLN. Such β1-AR and β2-AR distinction has been explained, at least in part, by their spatial compartmentation; at rest β2-AR is confined in caveolae, where this receptor is closely associated with LTCC, whereas the β1-AR distributes globally on the sarcolemma, and the vast majority are excluded from caveolae and indirectly associated with LTCC activation via the lateral diffusion of cAMP (15). Indeed, excessive β1-AR activation has been considered cardiotoxic, partially due to the activation of CaMK II (118), whereas β2-AR activates cell protective signals, including Akt/PKB (18), where receptor compartmentation may also be critical for their downstream signaling events. Cyclic AMP, the significant second messenger downstream of Gs-coupled receptors, is also compartmented to transfer the signal to two integrated effecter pathways, PKA and Epac signaling, the latter of which functions as a guanine-nucleotide-exchange factor for small GTPase protein Rap1 (24). Muscle-specific AKAP is found to be involved with these two integrated pathways (27, 28). Compartmentation of cAMP is achieved by the special limitation of both its production and degradation (76, 85). Studies have shown that adenylate cyclase type 5 and 6 differentially couple to the cAMP signaling in the distinctive cardiomyocyte subcellular domain, whereas the family of cyclic nucleotide phosphodiesterases (PDE) regulates the local concentration of cAMP (and/or cGMP) by its catalytic activity. Among five families of PDEs, PDEs 2–5 have been identified in cardiomyocytes, and studies using relatively specific chemical inhibitors have been utilized as molecular tools (35). A leak of such signal may cause cell dysfunctions, e.g., through excessive ICER (inducible cAMP early repressor) induction that is followed by a reduction in the anti-apoptotic protein, Bcl-2 (26, 106). For details of cardiomyocyte microdomain signaling via PDEs, see the recent series of reviews (27, 30, 35, 111).

Altered CaMK II Expression, Its Physiological Regulation, and Its Dysregulation

CaMK II activation in heart failure is one of the current topics of great interest in the field of Ca2+ cycling. Several studies have demonstrated that an increase in CaMK II activity closely correlates with the left ventricular ejection fraction in patients with heart failure (42, 56), so increased CaMK II activity was initially thought to be a compensatory mechanism of the heart to augment PLN phosphorylation-dependent upregulation of SR Ca2+ uptake. However, excessive activation has been suggested to be detrimental as reported by Zhang et al. (116), in that transgenic overexpression of CaMK IIδC, the most dominant cytosolic isoform in cardiomyocytes, induces severe contractile dysfunction and heart failure along with impaired intracellular Ca2+ handling. Presumably, CaMK II is activated by Ca2+ oscillation in cardiomyocytes; its activation is both frequency and Ca2+ concentration dependent on Ca2+ changes in subcellular microdomains where CaMK II are segregated. Such local Ca2+ changes are indeed hard to measure using micro-electrodes and fluorescent Ca2+ indicator technologies. Furthermore, as mentioned, CaMK II activation has been directly linked to β-adrenergic stimulation (118), and isoproterenol-induced cardiomyocyte apoptosis is inhibited by pharmacological CaMK II inhibition. In addition, the adaptor molecule that localizes CaMK II at the corresponding subcellular microdomain has not been well characterized, whereas that of PKA signaling is demonstrated to be a family of AKAP proteins. For example, RyR and CaMK II are known to make macro-molecular protein complexes along with PKA, PP1, and PP2A, but CaMK II anchoring protein to RyR has not yet been identified. The significance of the local activation of CaMK II in cardiac hypertrophy and heart failure was shown in a study on the role of the nuclear δB isoform in the induction of a pattern of hypertrophic gene expression (115), whereas the cytoplasmic δC isoform significantly affected E-C coupling through phosphorylation of the RyR (116). Although the importance of CaMK II regulation in the failing heart was confirmed by a study by Anderson and colleagues (113), which showed that the inhibition of CaMK II in infarct models can prevent the progression of adverse cardiac remodeling, although the study did not identify what fraction of CaMK II was involved in this treatment.

Phosphatases Locally Regulate Ca2+ Cycling

Subcellular localization of PP1 and its functional consequence are critically regulated by its specific targeting subunits. A regulatory subunit RGL (5) targets the PP1 catalytic subunit to the longitudinal SR and leads to PLN dephosphorylation. On the other hand, RyR, located in the junctional SR, is a component of a macromolecular complex, including PKA, CaMK II, PDE4D3, PP1, and PP2A, creating a phosphorylation regulatory network. The PP1 is anchored to RyR through the regulatory subunit spinophilin (66), whereas PP2A and PKA are docked to RyR via PR130 and the muscle-specific A-kinase anchoring protein (mAKAP), respectively. Cytosolic PP1 activity is regulated by endogenous inhibitors, such as inhibitor-1(INH-1) (33) and inhibitor-2 (INH-2) (38); INH-1 undergoes bidirectional control by kinases for PP1 inhibitory activity. Although it is not completely clear whether the membrane-associated PP1 is similarly regulated by associating peptides, PKA phosphorylated INH-1 works as a PP1 inhibitor and acts as an amplifier of β-adrenergic signaling (33). In contrast, PKCα-phosphorylated INH-1 is thought to function as a PP1 activator (12, 80, 81), which induces PLN dephosphorylation and subsequent inhibition of SR Ca2+ pump activity. INH-2, a constitutively active type of endogenous PP1 inhibitor and known to be a chaperone protein for the PP1 catalytic subunit (96), also regulates cardiac PP1 activity in cardiomyocytes (38). We assume that an imbalance of these kinases and phosphatase-dependent regulation of Ca2+ cycling in a local environment is critically involved in the progression of heart failure and associated arrhythmia, as discussed below.

“Further validation and optimization of Ca2+ uptake targeting therapy in heart failure will be needed. ”

Protein Phosphatase 1 is Abnormally Regulated in Heart Failure

In heart failure, the increase in global and SR-associated PP1 activity has been linked to depressed SR Ca2+ pump activity (37, 48). Herein, depressed phosphorylation levels of PLN may play a role. Although details have yet to be clarified, the mechanism of augmented PP1 activity is in part explained by either loss of endogenous inhibitory subunits or upregulation of the PP1 catalytic subunit. Namely, decreased expression level of INH-1 (38), decreased phosphorylation level of INH-1 at Thr35 (32), decreased activity of INH-2(38), and increased expression of the PP1 catalytic subunit associated with SR preparation (104) are reportedly involved. As shown in FIGURE 2, PP1 activity progressively increases both in the cytosol and SR preparations and correlates well with development of cardiac dysfunction (104). In addition, the role of PP1 hyper-activity in heart failure is supported by the finding of Carr et al. that the overexpression of the PP1 catalytic subunit α in the mouse heart caused dilated cardiomyopathy (13). In the PP1 trangenic hearts, there is a progressive decrease in the SR Ca2+ content, a decrease in the Ca2+ transient amplitude, and an increase in the diastolic cytosolic Ca2+ concentration. This leads to Ca2+ overload-induced cellular damage [e.g., via mitochondrial Ca2+ overload (117) or calpain activation (17)], followed by progressive loss of cardiomyocytes. These changes are consistent with the general features of the terminal stage of heart failure. In contrast, overexpression of INH-1 or constitutively active INH-1 has been shown to augment Ca2+ cycling and cell contraction/relaxation in vitro in isolated cardiomyocytes (33) and in vivo aorta-constricted rat hearts (78). In addition, cardiac-specific overexpression of INH-2 increased cardiac contractility by augmenting Ca2+ cycling in a transgenic mouse model (55), and we demonstrated that in vivo induction of INH-2 via a gene delivery model in hamsters were beneficial (104). Taken together, these findings suggest that increased global and SR-associated PP1 activities are detrimental factors that aggravate SR Ca2+ cycling and confer a heart failure phenotype. Indeed, Braz et al. (12) reported that PP1 activity is potentially enhanced in heart failure by excessive neurohumoral stimuli involving noradrenalin, angiotensin II, and endothelin, all of which are initially adaptive but become maladaptive as heart failure progresses. These stimuli activate Gq-coupled receptors and in turn cause activation of protein kinase Cα (12, 41), which phosphorylates INH-1 at its Ser67 residue, thereby activating PP1 (12, 80), leading to dephosphorylation of PLN.


Disease-dependent increase in PP1 activity in the heart failure animal model
As cardiac dysfunction progresses in the cardiomyopathic hamster afflicted with chamber dilation (A), PP1 activity increases both in the cytosolic (D) and SR preparations (E) (104). LVDd, left-ventricular end-diastolic diameter; %FS, percent fractional shortening of the left ventricle; LVSP&EDP, left-ventricular systolic function and end-diastolic pressure.

Notably, however, increased global or SR-associated PP1 activity does not always cause hypophosphorylation of Ca2+ cycling regulators in failing hearts. Marks and colleagues found that RyR is hyperphosphorylated in the animal and human failing hearts. According to their study, the RyR-associated PP1 protein level is decreased compared with that of PKA; therefore, they speculated that the local balance of kinase and phosphatases is more critical for causing hyperphosphorylation of RyR compared with the global increase in PP1 activity (60, 67). This notion was further supported by the finding by Yano et al. (109) showing that “hypo”phosphorylation of PLN and “hyper”phosphorylation of RyR coexist within the same failing cardiomyocytes (see FIGURE 3 for heterogeneous phosphorylation balance in the failing heart SR). These findings suggest that phosphatase activity in subcellular microdomains is critical in failing cardiomyocytes. Indeed, LTCC (16) and NCX (99) were also found to be hyperphosphorylated in failing hearts, further supporting heterogeneous micro-domain regulation of PP1, perhaps via dysregulation of targeting subunits.


Schematic image of sarcoplasmic reticulum microenvironment
In the cardiomyocyte, SR, a Ca2+ storage organelle, is anatomically divided into two basic components: junctional SR and longitudinal SR. The junctional SR includes RyR and its associated proteins and regulates Ca2+ release during systolic period, whereas the longitudinal SR function is a Ca2+ retriever from the cytosol by SERCA pump-rich structure (top). On β-adrenergic stimulation, it increases Ca2+ transient amplitude and augments contractility (middle). In heart failure, heterogeneous phosphorylation balancing occurs between the RyR and PLN, which induces decreased SR Ca2+ content and deteriorated Ca2+ transient ( bottom ).

With relation to these changes of PP activities in the failing heart, it has been suggested that increased PP1 and PP2A activities are the compensatory mechanism to facilitate myofilament contraction against PKC-mediated hyperphophorylation of myofilament phosphoproteins, which cause decrease in Ca2+ sensitivity (4, 77). Apparently, in this regard, further investigation regarding increased protein phosphatase regulation remains to be performed.

RyR Hyperphosphorylation by PKA and CaMK II: Cause of Leaky SR?

Dysregulation of the RyR by hyperphosphorylation in heart failure is characterized by improper gating of SR Ca2+ release and the diastolic Ca2+ leak. This concept has been pioneered by Marks’s group (67). Marks and colleagues reported that hyperactivation of PKA causes increased phosphorylation at Ser2809 of RyR, thereby inducing dissociation of FKBP12.6 from RyR, followed by instability of the RyR channels (67). In this regard, studies from Marks’ group and Yano et al. (110) found that altered stoichiometry between FKBP12.6 and RyR leads to a SR Ca2+ leak and cardiac contractile dysfunction. Nevertheless, the working model that RyR-FKBL12.6 coupling is regulated by RyR phosphorylation has been contradicted by several studies so that additional studies are needed for final determination. For instance, Stange et al. (90) found that a PKA-phosphorylation mimicking mutant (RyR S2809D mutant) did not have any effects on RyR gating. Li et al. (61) found that the onset of a SR Ca2+ leak by PKA phosphorylation is totally dependent on PLN phosphorylation and the subsequent increase in SR Ca2+ content but not RyR phosphorylation. Carter et al. (14) pointed out the importance of the extent of RyR phosphorylation by PKA as a critical event for changing the RyR gating properties; 100% phosphorylation of RyR, but not 75%, is required to increases the probability of the RyRs being in the open state.

Interestingly, the growing interest in RyR phosphorylation had led several investigators (2, 36, 57) to propose that CaMK II, instead of PKA, is critical for RyR phosphorylation and the associated SR Ca2+ leak. Curran et al. (23) found that activated CaMK II through β-adrenergic stimulation is responsible for the SR Ca2+ leak, but not by PKA activation. The frequent Ca2+ sparks and decrease in SR Ca2+ content found in CaMK II transgenic mice support this leak mechanism. However, this is still an emerging research area, and more work is needed to untangle the complexity of CaMK II-dependent RyR phosphorylation. For example, the CaMK II-mediated hyper-phosphorylation effect on RyR has been questioned by a recent study (107) that showed that CaMK II-dependent phosphorylation suppressed Ca2+ sparks and Ca2+ waves in cultured rat cardiomyocytes. However, the conflicting results among these studies might be explained by differences in the experimental settings, including distinct molecular functions between cell types and animal species, a difference between intact and permeabilized sarcolemma, which may affect subcellular compartmentalization of cyclic nucleotides, and the artificial substrates selected for various experimental strategies to overexpress molecules (88, 105). Additional complexity has been found in the ongoing debates with regard to phosphorylation sites of RyR that are critical for the RyR Ca2+ leak in the failing heart. Other than the originally reported Ser2808/2809 (shared animo acids with a putative PKA site), there are several candidate CaMK II phosphorylation sites in RyR, e.g., Ser2815 (a newly identified CaMK II site) (98, 102).

Other Molecular Players Regulating RyR Phosphorylation

As reported by Marks’s group and other investigators, the SR Ca2+-releasing channel is composed of a megaprotein complex (7) that includes the RyR tetramer, FKBP12.6 (calstabin2), PKA, PP1, PP2A, calmodulin, CaMK II, and phosphodiesterase (PDE)4D3, and this molecular conglomerate physically associates with other structural proteins including sorcin (34, 70), homer (101), triadin, junction, calse-questrin (112), and histidine-rich Ca2+ binding protein (59). Lehnart et al. (60) found that a lack of PDE4D3 by gene targeting caused adult-onset progressive cardiomyopathy in mice. In PDE4D3-deficient mice, the RyR displayed hyperphosphorylation at Ser2808, and RyR channels were found to be leaky. Since PDE4D3 expression is downregulated in human failing hearts, Lehnart et al. suggested that local cAMP accumulation in the vicinity of RyR was critical for RyR hyperphosphorylation and heart failure development. Nevertheless, PDE4D3 is identified at other cellular compartments such as the Golgi complex (94) and a perinuclear region (29) in cardiomyocytes, so the physiological role of PDE4D3 directly associated with RyR will have to be determined in future studies. On the other hand, a recent report by Yamaguchi et al. showed that calmodulin directly regulates RyR and thereby SR Ca2+ release and cardiac contractility (105). A RyR mutant mouse lacking CaM binding sites exhibits early onset progressive cardiac hypertrophy, reduced contractility, and premature death. CaM mutant RyR is characterized by a lack of termination of SR Ca2+ release in cardiomyocytes, suggesting the essential role of CaM binding to RyR for SR Ca2+ inactivation. Notably, calmodulin-dependent dual regulation (i.e., CaMK II-depedent and direct modulations) has also been noted in LTCC (9); thus Ca2+ apparently regulates E-C coupling utilizing multiple pathways. Perhaps the duration and specific distribution of modulation differ between direct regulation and regulation through kinase activity.

Regulation of Ca2+ Uptake and Its Link to Heart Failure

In addition to the reduced expression of the SERCA2a protein found in failing hearts (42, 62, 79, 84), phosphorylation of PLN clearly affects SR Ca2+ uptake. Human genetic studies strongly support this notion. The R9C mutation (86) disrupts PKA phosphorylation in PLN, suppresses oligomerized wild-type PLN phosphorylation, and causes dilated cardiomyopathy. The R14del mutation, on the other hand, is super-inhibitory, only partially responding to PKA phosphorylation, and leads to severely dilated cardiomyopathy and heart failure (39). PLN was found to be hypophosphorylated at the Ser16 residue, a PKA phosphorylation site, in several studies that tested failing hearts, in clear contrast to hyperphosphrylated RyR by PKA (73, 83, 87, 109). As such, PLN is generally dephosphorylated and inactive to stimulate SERCA2a, probably due to the increased PP1 activity, although several animal models failed to reproduce these reported changes (10, 75). The Thr17 residue of PLN is reported to be reduced in heart failure (74, 75), although cardiac CaMK IIδ expression and activity are found to be increased (43, 56).

Interacting proteins with the SERCA/PLN complex include sarcolipin (SLN) (8), S100A1 (54), RGL (5), sorcin (68), and myotonic dystrophy protein kinase (DMPK) (51). Among these, sorcin, a penta EF hand family protein, is regulated by PKA and PP1, and altered in the hamster heart failure animal model (68). Reduction of S100A1 in myocardial infarction and heart failure was found to be associated with depressed SR Ca2+ cycling in heart failure (8). An upregulation of SLN (3, 72), the transmembrane protein that has highly homologous sequence with PLN, suppresses SERCA2a activity. Kaliman et al. reported that loss of DMPK led to PLN dephosphorylation and reduced Ca2+ uptake (51). The phospho-regulation of S100A1 and SLN has not been established.

Wei et al. (100) reported that NCX activity is also regulated by phosphorylation through PKA and PP1. In heart failure, NCX is found to be hyperphosphorylated due to reduction of NCX-associated PP1 (100). Initially, NCX functions as a compensatory mechanism and keeps cellular Ca2+ level constant by upregulation of Ca2+ efflux in association with decreased SERCA pump activity. However, this compensatory mechanism eventually becomes maladaptive because decreased SR Ca2+ content decreases the amplitude of Ca2+ transient and cardiac contractility. In addition, NCX activity is indirectly linked to Na-K-ATPase that has an accessory protein, phospholemman (PLM), a trans-sarcolemmal protein that has a single phosphorylated peptide residue, which has long been known as a major PKA substrate in the cardiac membrane fraction. PLM expression is found to be decreased and hyperphosphorylated at its Ser68 residue in heart failure, thereby inducing less inhibition of Na-K-ATPase and indirectly affecting NCX (11).

Therapeutic Approach and Clinical Implication: Pump Up or Seal the Leak?

Calcium cycling defects serve as potential therapeutic molecular targets for heart failure. However, to date, none of the pharmaceutical agents (small molecular compounds) has sufficient target specificity to ameliorate this abnormal Ca2+ cycling in heart failure (19). PDE inhibitors have been extensively used to treat acute cardiac decompensation; however, several large-scale clinical trials failed to support its chronic therapeutic efficacy, which is likely due to its spatio-temporal nonspecificity in modulating the various PKA targets. Accordingly, many efforts have been made to correct defective Ca2+ cycling in the settings of experimental failing hearts by genetically engineered techniques or a vector-mediated gene transfer, although the clinically relevant therapeutic framework has not been established yet. Overexpression of specific subtypes of adenylyl cyclase to enhance cardiac cAMP production has been tested in small and large animal models of heart failure (58, 82). Intriguingly, therapeutic benefit of adenylyl cyclase VI was linked to selective phosphorylation of PLN in murine model of heart failure (58). Intriguingly, RyR phosphorylation was suppressed in adenylyl cyclase VI cardiac transgenic mice.

Direct benefits of a selective enhancement of PLN phosphorylation to treat heart failure have been supported by the PLN pseudo-phosphorylated mutant (S16EPLN) therapy from our group (44, 46, 47, 50, 71). The therapeutic efficacy of this pseudo-phosphorylation therapy has been apparent in cardiomyopathic hamsters, postmyocardial infarct rat hearts with chronic heart failure, and failing hearts resulting from continuous rapid pacing in sheep (53). Ongoing clinical studies employing intracoronary AAV directed delivery to overexpress SERCA2 in patients with advanced heart failure are currently in progress (1) and, depending on the results, could represent a major breakthrough in biologically targeted therapy for advanced heart failure.

Correction of abnormal phosphatase regulation is another therapeutic approach for upregulating SR Ca2+ uptake in heart failure (FIGURE 4). Our group reported that INH-2, an endogenous PP1 inhibitory protein, can selectively decrease the SR microvesicle-associated PP1 activity, perhaps by translocating the PP1 catalytic subunit from the SR to the cytosol, thereby increasing PLN phoshsphorylation and upregulating SR Ca2+ uptake in the cardiomyopathic hamster heart failure model (104). In addition, it is important that RyR phosphorylation was unaltered in the INH-2 delivered heart failure model. Consequently, overexpression of INH-2, a cytosolic PP1 inhibitor, may correct an abnormal subcellular PP1 environment and restore defective SR Ca2+ cycling.


Effect of INH-2 gene delivery (PP1 modifying) on the progression of heart failure
AAV-mediated INH-2 gene delivery improved cardiac function for over 3 mo (A–C), suppressed SR associated PP1 activity (D), and thereby increased PLN phosphorylation ( E ) (104). This inotropic therapy was effective for preventing cardiac fibrosis (F) and for extending animal survival (G) in studies of PLN inhibitory therapies (44, 46, 50, 53, 71).

The amelioration of SR Ca2+ uptake activity not only improves the speed of Ca2+ removal during relaxation but also increases the SR Ca2+ content, and therefore the amount of released Ca2+ contributes to the enhancement of cardiac contraction, which is somehow contradictory of the general perception of the inappropriateness of using inotropic therapies to treat heart failure. Indeed, there is a controversy regarding the functional role of upregulating SR Ca2+ flux. Indeed, in certain forms of genetic cardiac hypertrophy (25, 89), PLN ablation failed to rescue cardiac dysfunction, although cellular contractility showed significant improvement (89). In addition, hyperactivation of SR Ca2+ uptake by ablating PLN and substituting SERCA2a with SERCA2b, an isoform with higher Ca2+ affinity compared with SERCA2a, caused pathological hypertrophy and reduced lifespan (93), and endogenous PLN seems to play a protective role on cardiac injury in that model. With relation to this, one of the mutations of PLN, Leu39stop, proposed as PLN null-equivalency (40) due to the lack of trans-membrane domain, has been linked to human dilated cardiomyopathy. Although mechanisms and associated experimental conditions of these negative phenotypes need to be further considered in future experiments, these data suggest that overactivation of Ca2+ flux far beyond the normal Ca2+ homeostasis throughout the lifespan is associated with cardiac injury in certain situations. Also, it seems clear that upregulation of SR Ca2+ uptake cannot compensate pathophysiological signaling, which primarily causes cardiac hypertrophy. Further validation and optimization of Ca2+ uptake targeting therapy in heart failure will be needed.

RyR stabilization is another framework to ameliorate impaired SR Ca2+ handling. As RyR is being hyperphosphorylated by PKA or CaMK II in the failing heart, the channel becomes unstable and leaky. A recently identified chemical compound, JTV519, reduces SR Ca 2+ leak and improves cardiac function in a canine pacing-induced heart failure model (109). Intriguingly, JTV516 is apparently not a direct phosphorylation modifier but rather a channel stabilizer based on its benzodiazepine-like structure, suggesting that the local environment in the immediate vicinity of the RyR complex is critical for proper RyR phospho-regulation and normal channel gating. Furthermore, JTV is effective for preventing fatal arrhythmia induced by excessive β-adrenergic stimulation in the FKBP12.6-deficient mice (97), which mimics genetic arrhythmia-susceptible heart conditions.

CaMK II inhibition also is receiving attention as a new treatment strategy for preventing heart failure progression. Because CaMK II hyperactivation is directly linked to improper phospho-regulation of RyR and leaky SR, modification of this kinase may also be useful to seal the leaky Ca2+ releasing channels. Indeed, Anderson and colleagues documented that chemical or genetic inhibition of CaMK II prevents adverse cardiac remodeling in isoproterenol-induced mouse cardiomyopathy (113). Further study will be needed for its clinical application.

In summary, Ca2+ cycling defects in heart failure are characterized by impaired SR Ca2+ release and associated Ca2+ leak, reduced SR Ca2+ uptake, and reduced Ca2+ transients. Molecular targeting approaches to correct these abnormalities hold promise as a new therapeutic modality in the advanced end-stage heart failure patients who are less suited to the currently available medical therapy due to the severely impaired cardiac contractility and associated low blood pressure. Further progress in understanding of Ca2+ cycling defects with relevant application in the clinical setting are keenly anticipated.


View Abstract