The multifunctional Ca2+/calmodulin-dependent protein kinase II (CaMKII) regulates a rich variety of downstream targets in heart. Ca2+ homeostatic proteins are important CaMKII targets that support myocardial excitation-contraction coupling. Under stress conditions, excessive CaMKII activity promotes heart failure and arrhythmias, in part through actions at Ca2+ homeostatic proteins. Here, we briefly review the molecular and cellular physiology of CaMKII in myocardium.
Calmodulin kinase II is activated by intracellular Ca2+ and regulates key aspects of myocardial Ca2+ handling.
There is a growing interest in the role of protein kinases and phosphatases in cardiovascular disease (6). At the same time, loss of normal intracellular Ca2+ homeostasis has emerged as a central event initiating and perpetuating myocardial dysfunction, electrical instability, and arrhythmias (45). The multifunctional Ca2+ and calmodulin-dependent protein kinase II (CaMKII) is a serine-threonine kinase that is activated by increases in cellular Ca2+. CaMKII regulates many of the Ca2+ homeostatic proteins in myocardium. CaMKII is now a validated signal for causing heart failure and arrhythmias based on work in cells and animal models and the finding that CaMKII expression is increased in patients with structural heart disease and arrhythmias. This review will focus on the molecular physiology of CaMKII and present a conceptual framework for understanding how CaMKII can participate in myocardial physiology and disease.
CaMKII: Structure Determines Function
The multimeric nature of CaMKII allows its activation to be graded by changes in intracellular calcium. The four known isoforms of CaMKII (α, β, γ, δ) are encoded by separate genes. α and β-CaMKII are the predominant neuronal isoforms, whereas the δ and γ isoforms are expressed in diverse tissues including the heart (62). In heart, the predominant isoform is δ (23). Although these different isoforms have similar core structures, they differ subtly in their response to Ca2+ activation in vitro (17). This suggests that each isoform has evolved to decode Ca2+ oscillations differently, depending on in which cell system they reside. Different splice variants of these isoforms add further diversity to CaMKII input-output characteristics. One splice variant of the δ isoform (δB or δ3) has a nuclear localization signal in the hypervariable region (between the regulatory and association domains), resulting in nuclear targeting (60). Subcellular targeting of CaMKII is likely to be important for assignment of signal function.
The CaMKII monomer consists of an NH2-terminal catalytic domain, a centrally located regulatory domain, and a COOH-terminal association domain (FIGURE 1A⇓). The catalytic domain is sufficient to catalyze the transfer of the γ phosphate from ATP to serine or threonines embedded within a CaMKII consensus motif (12). Predicted CaMKII phosphorylation consensus motifs generally follow the form R-X-X-S/T. The regulatory domain contains a pseudosubstrate sequence that, under basal conditions, binds and constrains the catalytic domain. Also located in the regulatory domain is a calmodulin (CaM) binding domain. The pseudosubstrate sequence is built around an activating “autophosphorylation” site at Thr286/287 (the precise numbering is isoform dependent) (56) and a recently identified activating oxidation site (Met281/Met282) (24). Phosphorylation of Thr286 or oxidation of Met281 and Met282 prevents re-association of the catalytic and regulatory domains even after dissociation of CaM, thereby converting CaMKII into a Ca2+/CaM-independent signal. An inhibitory autophosphorylation site, Thr 306/307, overlaps with the Ca2+-CaM binding sequence. Phosphorylation of Thr306 is inhibitory because it prevents CaM binding (14). Oxidation of a lone Met at position 308 also abrogates CaM trapping (24), suggesting that adductive modifications in the COOH-terminal region of the regulatory domain reduce the avidity of CaM binding. Under resting conditions, basal phosphorylation occurs preferentially at Thr306, preventing Ca2+/CaM binding, resulting in decreased CaMKII activity. This inhibitory autophosphorylation provides a form of feedback regulation of the kinase that is dependent on basal/resting intracellular Ca2+ levels (14) and perhaps the intracellular redox state (24). Posttranscriptional mechanisms for regulating CaMKII activity are further discussed below. The association domain is responsible for the assembly of CaMKII monomers into the holoenzyme. A variable/hypervariable region, the product of alternative splicing, is located between the regulatory and association domain and may include a nuclear localization sequence (60).
The partial structure of the CaMKII holoenzyme has been determined by crystallography studies. The ~600 kDa holoenzyme consists of two central rings stacked on top of each other, with a diameter of ~145 Å, formed by interaction of multiple association domains. This central ring serves as the scaffold from which an outer ring, consisting of the regulatory and catalytic domains, arises (33). The holoenzyme is either a dodecameric (40, 47) or a tetradecameric (33) structure. This arrangement allows for a high concentration of catalytic domains that can interact with target proteins, including adjacent CaMKII monomers within the holoenzyme. CaMKII autophosphorylation (at Thr286/287) critically depends on the arrangement of monomers within the holoenzyme. In vitro studies revealed that the holoenzyme structure is critical for autophosphorylation graded activity responses related to the frequency and duration of intracellular Ca2+ transients (31). Cardiac myocytes are unique compared with other excitable cells in that they encode Ca2+ information by changes in both the frequency and duration of intracellular Ca2+ transients. However, physiological ramifications of the holoenzyme structure in cardiomyocytes have not been directly tested. Novel approaches are needed to test the role of the holoenzyme structure in CaMKII-mediated integration of Ca2+ signaling responses.
CaMKII Activation: Molecular Mechanisms
CaMKII requires Ca2+/CaM for activation (FIGURE 1B⇑). CaM is a bilobed intracellular protein (ubiquitous in vertebrate cells) that contains four Ca2+-binding EF hands (2 EF hands/lobe). The NH2 and COOH lobes of CaM are not functionally equivalent (3). CaM binding activates CaMKII preferentially by the COOH-terminal lobe (57). Under basal conditions, CaMKII is inactive because of intramolecular binding of the catalytic domain to the regulatory domain. Under conditions of low Ca2+/CaM concentration, most CaMKII is inactive, because of intramolecular binding of the catalytic domain to the regulatory domain. However, a fraction of CaMKII is active even under “basal” conditions of Ca2+/CaM, because autophosphorylated Thr286/287 is detectable in quiescent cells (36). This inhibitory interaction between the catalytic and regulatory domains prevents substrate and ATP binding (15, 38, 58). Calcified CaM binds to the regulatory domain, inducing a conformational change that frees the catalytic domain from its pseudosubtrate (43). The allosteric rearrangement of CaMKII on binding Ca2+/CaM allows ATP access to the ATP binding pocket, which in turn allows CaMKII to catalyze the transfer of a phosphate donor group to downstream targets, including itself (i.e., autophosphorylation). Autophosphorylation, which occurs by an intra-holoenzyme reaction (41), has several important implications for CaMKII activity. First, Thr286/287 autophosphorylation results in a 1,000-fold increase in the affinity of CaM-CaMKII binding, a property known as “CaM trapping”(46). Second, autophosphorylation results in the ability of the kinase to maintain catalytic activity even in the absence of CaM binding (56). Phosphorylation at Thr306 provides a negative regulatory mechanism for titering CaMKII activity. Under resting conditions, phosphorylation occurs preferentially at Thr306, preventing Ca2+/CaM binding, which in turn results in decreased CaMKII activity. This inhibitory autophosphorylation provides a form of feedback regulation of the kinase that is dependent on basal/resting intracellular Ca2+ levels (14).
Recently, a novel form of reactive oxygen species (ROS)-mediated CaMKII activation has been described. Treating Jurkat cells with H202 resulted in activation of CaMKII by a ROS-mediated mechanism in the apparent absence of Ca2+ flux (34). Similarly, adult ventricular myocytes treated with H202 demonstrated increased CaMKII-mediated phosphorylation of phospholamban (PLN), suggesting increased activity of CaMKII under oxidant stress conditions (90). These data suggested that CaMKII activation may be more complex than previously realized, potentially indicating that CaMKII is a nodal signal for integrating cellular Ca2+ and ROS into downstream responses.
Our group recently mapped a pathway for activation and conversion of CaMKIIδ into a Ca2+/CaM-independent species by oxidation of paired Met residues (Met281/282) in the regulatory domain (24). These Met residues are oxidized by angiotensin II stimulation and generation of ROS by NADPH oxidase in myocardium. Met oxidation prevents inhibitory reassociation of the catalytic and regulatory domains, much like Thr286/287 autophosphorylation, but Met281/282 oxidation and Thr286/287 autophosphorylation are independent events. In contrast to autophosphorylation, Met oxidation does not result in CaM trapping due to simultaneous oxidation of Met308 located near the negative regulatory residue Thr306. A Met308Val CaMKIIδ mutant was activated by oxidation and exhibited CaM trapping similar to Thr286/287 autophosphorylation. These data suggest that activating and inhibitory Met residues are readily available to oxidation. Met oxidation is specifically reversed by Met sulfoxide reductase A (MsrA) (68). Mice null for MsrA showed higher levels of oxidized CaMKII in myocardium after angiotensin II stimulation or after myocardial infarction compared with controls. MsrA−/−mice had worse myocardial dysfunction and increased mortality after pathological stress. Taken together, these findings show that CaMKII is dually activated by “upstream” Ca2+ and oxidative signals and that both phosphorylation and Met oxidation are dynamically reversible events.
CaMKII Activation: Cellular Mechanisms
The cardiac action potential plays an important role in mediating CaMKII activation (FIGURE 2⇓). The ventricular cardiomyocyte action potential duration (APD) lasts several hundred milliseconds, approximately two orders of magnitude longer than neuronal action potentials. The prolonged cardiac APD is due to a long plateau phase (so called phases 2 and 3) that is a balance of inward currents (mostly L-type Ca2+ current) and repolarizing outward currents (mostly K+). This prolonged plateau phase of the cardiac action potential is critical for allowing Ca2+ entry, mainly through voltage-gated L-type Ca2+ current (ICa), and limiting cell membrane excitability by imposing a prolonged electrical refractory period. CaMKII activation is enhanced by APD prolongation because the intracellular Ca2+ transient lengthens in step with APD prolongation (74). In cardiomyocytes, CaMKII activity appears to be increased by APD prolongation (5) and faster heart rates (67). Langendorff-perfused hearts treated with the K+ channel blocker clofilium exhibit APD prolongation and increased Ca2+/CaM-independent activity compared with control hearts paced at the same frequency (5). Increasing the pacing frequency of Langendorff-perfused hearts has also been demonstrated to result in increased Thr287 autophosphorylation and CaMKII activity (67). In contrast, the gradation of CaMKII activity is dependent primarily on changes in stimulation frequency in neurons where the dynamic range of APD is more limited.
The dependence of CaMKII activation on pulse duration and frequency has been demonstrated in vitro. When CaMKII is stimulated by shorter Ca2+ pulse durations (80 ms), it requires higher frequency oscillations (10 Hz) to achieve increased Ca2+-autonomous activity. On the other hand, longer Ca2+ pulse durations (1,000 ms) respond to lower stimulation frequencies (0.8 Hz) to achieve the same level of Ca2+ and CaM-autonomous activation (17). These in vitro findings not only have implications for the normal physiological effects of APD on CaMKII duration but also suggest a mechanism of excess CaMKII activation in genetic (long QT syndrome) and structural heart disease with APD prolongation (5, 73). Although stimulation frequency and pulse duration are the best understood upstream regulators of CaMKII activation (17), other factors such as redox potential, CaMKII isoform distribution, phosphatase activity, and intracellular localization also likely participate in determining CaMKII activity in cardiomyocytes.
In heart failure, abnormal Ca2+ handling can in part be explained by prolongation of APD due to remodeling of ionic currents (10, 48). Electrical remodeling in heart failure may be initially adaptive by favoring cellular Ca2+ entry and restoration of contractility. However, these changes are ultimately maladaptive and promote dysregulation of Ca2+ homeostasis (11) and worsening of cardiac function (86) that are mediated at least in part by excess CaMKII activation.
Excess CaMKII activity itself can promote APD prolongation in part through a posttranslational mechanism. CaMKII induces L-type Ca2+ channels (LTCCs) to enter a highly active gating mode marked by frequent, long openings (20). CaMKII is responsible for dynamically increasing ICa, a phenomenon termed facilitation (4, 77, 82). CaMKII can also increase APD by a phosphorylation-dependent increase in a non-inactivating component of cardiac Na current (INa)(64). Both ICa and INa increase net inward current during the AP plateau leading to APD prolongation. CaMKII overexpression leads to cardiomyopathy and heart failure with APD prolongation that may be at least partially due to reduced expression of repolarizing K+ currents (44). Mice with CaMKIV overexpression have coordinate increases in CaMKII expression and activity and show APD prolongation due to reduced repolarizing K+ currents (76). In contrast, CaMKII inhibition shortens the APD and cellular Ca2+ transient in surviving myocardium after infarction (85) and leads to APD shortening by augmenting the fast component of the transient outward and the inwardly rectifying K+ currents by an incompletely characterized nontranscriptional mechanism (42). Taken together, these findings show that CaMKII sculpts the cardiac APD by a number of mechanisms that appear to be important for the proarrhythmic electrical remodeling phenotype that is a common feature in myocardial hypertrophy and heart failure.
Excessive APD prolongation and increased CaMKII activity predispose to early afterdepolarizations (EADs) by ICa facilitation and high LTCC activity (73, 76, 83). APD prolongation leads to CaMKII activation (5), increased sarcoplasmic reticulum Ca2+ loading, and activation of the delayed afterdepolarization (DAD)-initiating transient inward current (74). EADs and DADs are depolarizing oscillations in the cell membrane potential that are the probable initiating cause for many arrhythmias, including atrial fibrillation and ventricular arrhythmias in heart failure (53) (FIGURE 3⇓).
The structure of CaMKII not only serves to decode Ca2+ frequency, it also serves to target the kinase directly to key regulatory proteins involved in Ca2+ homeostasis. Proper targeting of kinases and phosphatases to protein substrates is essential for maintaining signaling fidelity (51). Other Ser/Thr protein kinases, including PKA and PKC, utilize anchoring proteins for targeting protein substrates (19, 27, 37), but similar adapter proteins suitable for localizing CaMKII to substrate targets have not been identified. Despite an absence of CaMKII adapter proteins, CaMKII is selectively enriched along the Z band of ventricular myocytes where it co-localizes with LTCCs and ryanodine receptors (21, 73). CaMKII binds directly to the NMDA receptor NR2B subunit after phosphorylation of Thr286 (8, 61). In adrenal cells, CaMKII directly binds with the intracellular loop of T-type Ca2+ channels, regulating channel function by phosphorylation of Ser1198 (81). CaMKII also binds the LTCC (CaV1.2) α1C subunit at multiple sites, including the COOH terminus (35). Thus CaMKII selectively adapts to protein targets in a regulated manner in cardiomyocytes.
We recently identified a CaMKII binding sequence in an auxiliary LTCC β subunit (β2a) (29). CaMKII binding to the β subunit and to NR2B occurs by way of a sequence that resembles the CaMKII regulatory domain. Thr286 autophosphorylation enables CaMKII binding to the β subunit and NR2B. On the other hand, phosphorylation of T498 on the β subunit reduces CaMKII-β subunit binding. The identification of a CaMKII-specific binding sequence in these important Ca2+ homeostatic proteins shows that CaMKII adapter sequences (CaMkaps) are embedded in at least some CaMKII substrate targets rather than existing as distinct proteins. These findings suggest that CaMKII targeting may work by a parsimonious (i.e., no requirement for adapter proteins) and novel mechanism. CaMKII inhibitory peptides that mimic the autoinhibitory sequence (59, 85) are thus also anticipated to prevent normal targeting of CaMKII to substrates and so may lead to suppression of CaMKII activity by two distinct mechanisms.
CaMKII and Excitation Contraction Coupling
The excitation-contraction coupling (ECC) apparatus is the mechanism for myocardial mechanical function. CaMKII regulates many of the key proteins for ECC. There is an increasing awareness that disorders of ECC are also the basis for many arrhythmias and can lead to cardiac hypertrophy and heart failure. The joint dependence of myocardial mechanical and electrical functions on ECC provides a rationale for understanding why patients with heart failure (i.e., a clinical condition of inadequate mechanical cardiac function) are also at most risk for arrhythmias and sudden death. CaMKII is now known to contribute to cardiac hypertrophy, heart failure, and arrhythmias, and CaMKII activity and expression are increased in failing human hearts and animal models of structural heart disease (87). The following text will briefly discuss the role of CaMKII at some key ECC proteins.
L-type Ca2+ channels
LTCCs serve as the primary entry point for Ca2+ into cardiomyocytes. The predominant ventricular LTCC α subunit (CaV1.2) belongs to a family of six membrane-spanning, voltage-gated channels. Activation of the channel occurs by cell membrane depolarization However, gradation of ICa is mediated by CaM and CaMKII signaling events. Ca2+/CaM can act directly to accelerate channel inactivation. Ca2+/CaM, in turn, can activate CaMKII, resulting in augmentation of ICa, a process termed ICa facilitation (72).
CaMKII binds the cytoplasmic COOH terminus of Cav1.2 (35), as well as the β subunit (29), which associates with Cav1.2 at the intracellular I–II linker domain (84). Early studies demonstrated ablation of ICa facilitation using pharmacological and peptide inhibitors of CaMKII (4, 77, 82). Treatment with constitutively active CaMKII results in phosphorylation of the LTCC complex, inducing high activity (mode 2) gating that is characterized by long, frequent openings (20). CaMKII anchors to and phosphorylates regions along the CaV1.2 COOH terminus. Mutations to a putative COOH terminus binding site abolish ICa facilitation (35). Recently, CaMKII was shown to bind the β subunit and phosphorylate Thr498, a residue that is conserved in all the major cardiac β subunit isoforms. Mutation of Thr498 to Ala results in ablation of CaMKII-mediated ICa facilitation in cardiomyocytes (29). Interestingly, the CaMKII site (T498) that increases ICa is in close proximity to a protein kinase G (PKG) site (S494) that results in decreased ICa (79). It is unknown whether these sites interact. CaV1.3 is an LTCC that is present in atrium, cardiac pacemaker cells, and the specialized conduction system. CaMKII may increase CaV1.3 by phosphorylating a Ser residue that is present in an intracellular COOH terminal EF hand domain (26). Both CaV1.2 and CaV1.3 share a homologous EF hand, but the Ser and predicted CaMKII consensus motif are only present in CaV1.3. The potential physiological and disease consequences of CaMKII regulation of CaV1.3 in heart are unexplored.
CaMKII co-localizes with the ryanodine receptor (RyR) and the LTCC, poising it to mediate Ca2+ induced Ca2+ release (73). The RyR, located on the sarcoplasmic reticulum (SR) membrane, is juxtaposed with the sarcolemmal LTCC in highly organized SR Ca2+ release units known as couplons (25). Calcium entry via the LTCC results in activation and release of SR Ca2+ from the RyR, a process termed Ca2+-induced Ca2+ release. In human and animal models of heart failure, the RyR becomes hyperphosphorylated (2, 45), resulting in inappropriate diastolic SR Ca2+ release (Ca2+ sparks), which may contribute to impaired contractility, promote DADs, and activate cardiomyopathic gene programs. The RyR is a 2.5-MDa intracellular Ca2+ release channel that consists of four identical subunits. The majority of this channel serves as a cytoplasmic scaffold for myriad accessory proteins, including CaMKII. CaMKII directly binds and phosphorylates RyR at Ser2809 and Ser2815 (16, 67, 70). However, phosphorylation studies suggest CaMKII may phosphorylate additional sites, known and unknown (54).
Under different experimental conditions, CaMKII has been shown to either increase or decrease Ca2+ release via the RyR (39, 44, 67, 71, 78). Isolated RyRs placed in lipid bilayers demonstrate increased CaMKII-mediated phosphorylation (Ser2815) and channel activity at increasing heart rates, which are associated with increased CaMKII autophosphorylation (67). In transgenic mice overexpressing CaMKII, RyRs show increased Ca2+ spark frequency, which can be reduced by CaMKII inhibition (44). In a rabbit heart failure model of left ventricular pressure and volume overload, there is increased CaMKII phosphorylation of RyRs. Inhibiting CaMKII in this model reduced SR leak and enhanced SR Ca2+ content (2). Adenovirus-mediated overexpression of CaMKII in cultured adult ventricular myocytes increases (39) or inhibits (78) RyR function. Overall, these data support an important regulatory role of CaMKII in RyR function but leave open many questions about how CaMKII acts on RyR and why CaMKII appears to have both inhibitory and stimulatory actions in various models. In heart disease, it appears that CaMKII effects on RyR contribute to the phenotype of heart failure, which includes depleted SR Ca2+ stores as well as increased arrhythmia-initiating DADs.
CaMKII regulates Ca2+ reuptake from the cytoplasm by the sarcoplasmic/endoplasmic reticulum Ca ATPase (SERCA), exclusively or in large part, via its effects on phospholamban (PLN). SERCA is located on the membrane of the longitudinal SR where it functions to remove intracellular Ca2+, resulting in cardiomyocyte relaxation. In humans and rabbits, SERCA accounts for 70% of cytoplasmic Ca2+ removal, whereas in rats and mice it accounts for ~90% (9). Most of the remaining cytoplasmic Ca2+ is removed via the sodium-potassium exchanger (NCX). SERCA is tonically inhibited by PLN, which is a 52-amino acid transmembrane protein. Association of PLN with SERCA lowers the affinity for Ca2+ ions. Phosphorylation of PLN by either PKA (at Ser16) or CaMKII (at Thr17) relieves SERCA of PLN’s inhibitory effects, accelerating intracellular Ca2+ removal and enhancing SR Ca2+ content (30, 63).
Frequency-dependent acceleration of relaxation (FDAR) is a property of cardiomyocytes caused by the acceleration of intracellular Ca2+ removal by SERCA at increasing heart rates. From a physiological perspective, FDAR promotes acceleration of myocardial relaxation (diastolic function) for the maintenance of efficient ventricular filing at higher heart rates, despite a decreased diastolic time interval. The ability of CaMKII to detect changes in intracellular Ca2+ as well as to promote SERCA activity by phosphorylating PLN makes it a strong candidate for mediating FDAR.
In isolated ventricular myocytes, it has been shown that Thr17 phosphorylation and FDAR increase with increasing stimulation frequency, independent of Ser16 phosphorylation (30). It also has been demonstrated that Thr17Ala mutation attenuates FDAR at higher pacing frequencies compared with wild-type and Ser16Ala mutation mouse cardiomyocytes (89). CaMKII inhibition suppresses FDAR (7, 18). In contrast to these findings, PLN knockout mice still exhibit FDAR, which can be ablated with CaMKII inhibition (18). Furthermore, other studies have demonstrated that Ser16 is sufficient for mediating β-adrenergic receptor-mediated FDAR (13). These data support CaMKII as an important mediator of FDAR but do not resolve the controversy about the identity of the key proteins that underlie FDAR. The effects of CaMKII and PLN appear to wane at high (physiological) steady-state frequencies (69), suggesting that modulatory effects of CaMKII on PLN (or alternative FADR mediators) may be most important at slower, irregular stimulation frequencies.
CaMKII inhibition in transgenic mice has been shown to be cardioprotective against chronic β-adrenergic receptor (βAR) stimulation, in part by protecting against SR Ca2+ overload (85). This protective effect and the protection against SR Ca2+ overload are dependent on decreased CaMKII-mediated phosphorylation of PLN(75). Crossing of CaMKII inhibitory mice with PLN knockouts abolished the resistance to ischemic stress-induced apoptosis by CaMKII inhibition (80). These findings suggest that PLN is an important molecular determinant for the anti-apoptotic actions of CaMKII inhibition.
CaMKII Inhibition as a Therapeutic Target
In the United States, 280,000 patients die of heart failure annually (55). With a risk of sudden cardiac death six to nine times that of the general population, about half of heart failure patients die of ventricular arrhythmias. Although drugs that inhibit the adrenergic and renin-angiotensin aldosterone systems have improved survival (1, 1, 49, 52), deaths from heart failure increased 28% from 1994 to 2004. Treating arrhythmias in patients with structural heart disease by ion channel antagonist drugs does not reduce mortality (22, 65). These data support the importance of finding suitable agents for treating both heart failure and arrhythmias.
Pharmacological approaches that inhibit G protein-coupled receptors (GPCR) have produced impressive therapeutic successes, but they also produce unwanted side effects, in part by targeting diverse intracellular signals linked to GPCRs. Recently, targeting of protein kinases has been considered as an alternative and potentially more biologically selective therapeutic strategy compared with GPCRs (6). CaMKII is an appealing potential target for pharmacological inhibition. CaMKII activity is upregulated in hypertrophy and heart failure (32, 88). CaMKII overexpression in transgenic mice results in impaired Ca2+ handling, APD prolongation, heart failure, and arrhythmias (44, 76, 88), whereas CaMKII inhibition protects against development of heart failure and arrhythmias (76, 85).
A potential benefit of direct CaMKII inhibition may include better specificity of inhibition that reduces unwanted global effects of GPCR inhibition. For example, β-adrenergic receptor inhibition is associated with improved outcomes, but negative inotropic actions limit the use of these drugs in severe (class III to IV) heart failure (28, 50). In contrast, CaMKII inhibition prevents maladaptive remodeling from excessive intermittent β-adrenergic receptor stimulation and myocardial infarction with preservation of baseline cardiac function and contractile reserve (85). On the other hand, continuous exposure to excessive β-adrenergic stimulation may induce cultured cardiomyocytes to rely on CaMKII for contractility (66), suggesting the possibility that CaMKII inhibition could reduce mechanical function under some conditions.
There are many potential methods for inhibiting CaMKII function including developing drugs that prevent ATP binding at the catalytic domain, drugs that result in allosteric inhibitory conformational changes to CaMKII, as well as drugs that specifically prevent CaMKII binding to protein targets. Therapies that affect the CaMKII holoenzyme architecture or prevent the conformation changes necessary for autophosphorylation and Ca2+-independent activity may prevent excess CaMKII activity that results in heart failure and arrhythmias.
CaMKII is a remarkable signaling molecule that is abundant in myocardium and other excitable tissues. CaMKII targets a wide range of proteins in heart but is especially important for regulating proteins necessary for ECC. The CaMKII holoenzyme is uniquely adapted to integrate and transduce cellular Ca2+ signals into physiological responses in heart. Under pathological stresses, CaMKII appears to be an important disease signal. The actions of CaMKII on electrical and Ca2+ handling proteins help to explain how CaMKII can simultaneously favor heart failure and arrhythmias. Recent findings that CaMKII inhibition reduces heart failure and suppresses arrhythmias suggest that developing CaMKII inhibitory drugs may be a new therapeutic approach to these diseases.
This work was funded by National Heart, Blood, and Lung Institute Grants R01 HL-079031, R01 HL-62494, and R01 HL-70250 and the University of Iowa Research Foundation. Dr. Couchonnal was funded by the University of Iowa Cardiovascular Interdisciplinary Research Fellowship, HL-007121. Mark Anderson is a named inventor on patents claiming to treat heart failure and arrhythmias by CaMKII inhibition.
- © 2008 Int. Union Physiol. Sci./Am. Physiol. Soc.