Introduction

Heart failure (HF) remains an unsolved public health problem. A failing heart is unable to efficiently supply oxygenated blood to the body, resulting in insufficient oxygen supply and nutrients to the body. Cardiovascular-related diseases, including chronic cardiac overload or injury (e.g., high blood pressure, valvular heart disease), myocardial infarction or ischemia, cardiac remodelling, functional abnormalities, and genetic disorders, can eventually lead to HF [1]. Response of the heart can initially be compensatory to additional load or heart damage by increasing its size and contractility [2]. The increase in heart size and mass is thought to be accompanied by biochemical, molecular, structural, and metabolic changes that maintain the function of an enlarged heart. However, chronic stress or disease can lead to dilation of the ventricles and decline of the systolic function, eventually progressing to HF [3]. The most obvious functional change in HF is the decline of diastolic and systolic function, an important cause of which is Ca2+ dysfunction in the myocardium.

Ca2+ dysfunction is associated with abnormal activation or inactivation of key kinases and phosphatases, which can cause phosphorylation and dephosphorylation imbalance in HF (Fig. 1). Protein kinase A (PKA) is a central regulator of cardiac function and morphology [4]. The typical PKA signaling pathway is essential for cardiac activity, especially catecholamines, including norepinephrine secreted from cardiac sympathetic nerve terminals in the heart and epinephrine released from the adrenal medulla [5]. Acute PKA activation improves cardiac performance and is associated with enhanced myocardial contractility, but chronic PKA activation or inhibition can lead to HF. Constitutive PKA activation induces hyperphosphorylation of phospholamban (PLN) and ryanodine (RyR2) of the sarcoplasmic reticulum (SR), leading to reduced contractility and dilated cardiomyopathy [6]. Protein kinase C (PKC) is another member of the serine-threonine kinase family. The increased expression and activity of PKC in HF are closely related to the activation of PKC-α [7, 8]. PKC-α is a fundamental regulator of cardiac contractility and Ca2+ processing in cardiomyocytes. Regulation of PKC-α activity affects dephosphorylation of the SR Ca2+ ATPase-2 pump (SERCA-2) inhibiting PLN and altering the SERCA-2 Ca2+ load and transients. Other kinases, such as protein kinase D (PKD), calmodulin kinase 2 (CaMKII), etc., have also been shown to be activated in HF, while protein kinase G (PKG) activity is reduced [9, 10]. The activation mechanism of PKD involves PKC-mediated PKD phosphorylation, which can be attenuated by PKC inhibition [11]. And neurohormonal stimulation of PKD activity may be enhanced under conditions where PKA activity is down-regulated [12], perhaps allowing PKD-mediated pathways to assume greater significance in the acute regulation of contractile function in HF [13]. PKD-mediated myofilament phosphorylation may have physiological significance in the neurohormonal regulation of myocardial contractile function. PKG can be activated by PKA, and play a role in inhibiting adrenaline, which is related to anti-myocardial hypertrophy [14]. PKG also phosphorylates many PKA-related sites, including the L-type Ca2+ channel (LTCC), PLN, troponin I, myosin-binding protein C (cMyBP-C), and titin. CaMKII is a highly validated signal associated with a variety of diseases, especially those of the cardiovascular system [15].

Fig. 1
figure 1

Calcium function in normal myocardium and failing myocardium. In normal hearts, appropriate phosphorylation and dephosphorylation of LTCC, ER, and myofilament are associated with faster Ca2+ release and recovery and lower Ca2+-sensitivity; while in failing hearts, abnormal phosphorylation and dephosphorylation of LTCC, ER, and myofilament is associated with faster Ca2+ release but weakened recovery ability and higher myofilament Ca2+-sensitivity

Dephosphorylation in the heart related to systolic and diastolic function is mainly mediated by protein phosphatase 1(PP1), protein phosphatase 2A (PP2A), and phosphodiesterase (PDE). PP1 is a serine-threaminophosphatase phosphatase that primarily targets the PLN of SR [16]. Activity of PP1 has also been shown to increase in failing hearts and is associated with reduced Ca2+ recovery in ER [17, 18]. PP2A is another major phosphatase in the heart that regulates Ca2+ processing [19]. PP2A coordinates the excitation and contraction of the heart. The importance of PP2A in the heart lies in its ability to antagonize the effects of β adrenergic receptor (β-AR) stimulation, reducing Ca2+ transient amplitude while increasing the Ca2+ sensitivity of myofilaments in force development. PP2A is the main phosphatase of LTCC. PP1 and PP2A form complexes on the RyR and have regulatory effects on the RyR. In HF, PP2A expression and activity are dysregulated [19]. PDE superfamily consists of several distinct subtypes that regulate the strength and duration of cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) signaling in discrete compartments of cardiomyocytes [14], with the PDE4 and PDE3 subtypes controlling Ca2+ release and reuptake in the sarcoplasmic reticulum via RyR2 and SERCA-2 [20,21,22,23]. In pathological hypertrophy and HF, levels of PDE1, PDE2, PDE5, PDE9, and PDE10 are elevated. Overall, decreased activity of PDE3 and PDE4 amplified catecholamine toxicity [14].

Changes in the expression and phosphorylation / dephosphorylation imbalance are associated with changes in Ca2+ activity in cardiomyocytes. Studies on changes in phosphorylation levels have been quite mature. It has been recognized that three main amino acids can be phosphorylated: serine (Ser), threonine (Thr), and tyrosine (Tyr), which are characterized by active hydroxyl groups that can be negatively charged by binding to phosphate groups [24,25,26,27,28,29,30]. Animal experiments were conducted to study the effects of phosphorylation in different regions of cardiomyocytes on Ca2+ function by replacing amino acid targets with negatively charged aspartic acid to simulate continuous phosphorylation and positively charged alanine to simulate continuous dephosphorylation. In this review, we summarize the current research and findings of the relationship between intracellular Ca2+ dysfunction and phosphorylation/dephosphorylation imbalance in cardiomyocytes and clarify potential therapeutic directions.

Receptor activation and related-phosphorylation

One of the most prominent features of the progression of cardiac hypertrophy is the persistent activation of β-AR. The rapid positive inotropic action of β-AR activation is dependent on the activation of PKA and its downstream target phosphorylation. Epinephrine signaling directly contributes to PKA activation and stimulates downstream phosphorylation by activating β-AR. In cardiomyocytes, the major targets are LTCC, RyR2, and PLN [31]. This phosphorylation is thought to be beneficial during early hypertrophy. The phosphorylation of LTCC facilitates external Ca2+ uptake, thereby activating more ER Ca2+ release. The phosphorylation of RyR2 is beneficial to the release of Ca2+, the increase of cytosolic Ca2+ concentration, and the increase of contractility during systole. The phosphorylation of PLN enhances the function of the Ca2+ pump, which is conducive to the rapid reduction of intracellular Ca2+ concentration, the rapid completion of the relaxation and contraction cycle, and the improvement of myocardial function.

In addition to PKA activation, the downstream activator of β-AR is PKCε, which is not activated through the cAMP pathway. Studies have shown that epinephrine can activate PKCε via β-AR, independent of PKA activation [32]. The activation was not mediated by the cAMP signaling pathway but by the classical PLC/PKC pathway. In cardiomyocytes, the major phosphorylation targets of PKC are located in myofilaments and are associated with decreased myofilament contractile function. In addition to this, PKC also phosphorylated Thr286 of CaMKII [33], which was related to the decrease of CaMKII activity. Loss of PKCε promotes the occurrence and development of HF [8] indicating that PKCε plays an important role in preventing the occurrence of HF.

Epinephrine signaling activates not only β-AR but also α adrenergic receptor (α-AR). α-AR also shows positive inotropic effects. During the development of physiological hypertrophy or a stress state, epinephrine mainly activates β-AR, triggering a strong positive inotropic effect. However, the α-AR mediated positive inotropic response is predominant in HF rats, which is related to the inhibition of β-AR phosphorylation [34]. Activation of α-AR can promote the phosphorylation of the myosin light chain (MLC) by promoting the activation and expression of Ras homolog gene family (Rho) kinase, which is related to the increase of myofilament Ca2+-sensitivity [35].

In normal conditions, neurotransmitters can inhibit the β effect of epinephrine by activating muscarinic 2 receptors (M2-R). In the progression of cardiac hypertrophy, the activation of epinephrine inhibits the M2-R. When β-AR is continuously activated, parasympathetic nerves are activated in a feedback manner. Undergoing hyperphosphorylation leads to the inhibition of β-AR, and activation of M2-R can be manifested. In cardiomyocytes with suppressed β-AR, M2-R activation promotes the activation and expression of MIC kinase and Rho kinase, leading to MLC-2 phosphorylation and increased myofilament Ca2+-sensitivity [36], suggesting that the activation of these two kinases is independent of cAMP. This would act as a compensatory effect for the loss of β-AR-related inotropic effects. In addition to the partial positive inotropic effect caused by M2-R activation, the overall performance of myocardial contractility is still reduced due to the inhibition of β-AR phosphorylation and the activation of the dephosphorylation signal.

Abnormal Ca2+ and phosphorylation/dephosphorylation

LTCC and abnormal phosphorylation/dephosphorylation

LTCC, commonly referred to as dihydropyridine receptor (DHPR), is sensitive to various 1,4-dihydropyridines [37]. LTCCs in the cardiomyocyte are composed of four different polypeptide subunits (a1, b, a2, d), and the pore-forming subunit a1 is the most important part of the channel, which forms the channel pore for ion flow.

LTCC is the main entry point for Ca2+ influx (ICa2+) into cardiomyocytes and determines the activity of the entire heart [38,39,40]. The main pathway of Ca2+ channel activation is through PKA-mediated phosphorylation (Fig. 2A), which is activated by the second messenger cAMP. This process is also regulated by phosphorylation of LTCC and intracellular Ca2+ concentration [38, 41,42,43]. However, the inhibition of adenylyl cyclase (AC) activity is one of the most common pathways to interrupt PKA-dependent LTCC stimulation [44]. Another way to reduce PKA-dependent channel phosphorylation is the activation of PDEs, which hydrolyze cAMP and cGMP and reduce their intracellular concentrations [42, 45] and play an important role as PKC in LTCC regulation. PKC has a biphasic effect on ICa2+. PKCβ isoform stimulates ICa2+, whereas PKCε inhibits ICa2+ [46, 47].

Fig. 2
figure 2

Phosphorylation and dephosphorylation in normal myocardium and failing myocardium. A In normal conditions, PKA is the main kinase enhancing myocardial contractility when receiving adrenaline stimulation through the β-AR. PDE mainly coordinates with the function of PKA. PP2A and PP1 are responsible for LTCC and ER dephosphorylation, facilitating myocardial dilation. B But in HF, with a weakened β-AR signal and increased PDE function, the M2-R and α-AR remain functional. Phosphorylation caused by PKC, CaMKII, Rho, and MLCK is associated with Ca2+ leakage and enhanced myofilament Ca2+ sensitivity. Dephosphorylation caused by PP1 can reduce Ca2+ recovery into ER

Previous studies have found that it may not be the number and activity of LTCC that are reduced in HF, but mainly the ability of LTCC to stimulate the release of Ca2+ in ER, which is related to the excitation-contraction coupling defect [48]. Basal ICa2+ of LTCC was also found to be weakened in HF (Fig. 1), which may be related to the Ca2+ concentration in the cytoplasm, in which the Ca2+ binding protein calmodulin (CaM) plays a key role [49]. The balance of phosphorylation and dephosphorylation of LTCC is related to their activity, and the phosphorylated form of LTCC may be related to maintaining channel activity and enhancing the ability to stimulate ER Ca2+ release. It was previously thought that PKA and PP2A regulate the phosphorylation level of Ser1928 and thereby determine channel activity [50]. However, later experiments found that the phosphorylation level of Ser1928 is not related to LTCC activity, but the distal carboxyl terminus of α1C is the required factor for the β-AR stimulation of LTCC in cardiomyocytes [51]. Another interesting phosphorylation site is the Ser1700 site of the a-subunit. Although one study demonstrated that PKA-mediated phosphorylation at Ser1700 did not have a major effect on the enhancement of ICa2+ [52], other experiments found that ICa2+ was reduced and cardiac hypertrophy developed when phosphorylation at Ser1700 was absent [53], while Ser1700 and Thr1704 double mutations accelerate cardiac hypertrophy and HF [54]. Another experiment showed that these conserved consensus PKA phosphorylation sites (in addition to those mentioned above), including Ser1512 and Ser1570 (CaMKII-mediated phosphorylation) in α1, Ser459, Ser478, and Ser479 in β2, were not responsible for elevated LTCC activity when phosphorylated [55]. In addition, the phosphorylation of LTCC by PKA is also associated with A kinase anchoring protein (AKAP). PKA without AKAP15 was ineffective in regulating LTCC in cardiomyocytes when the corresponding β-AR pathway was stimulated [56]. Genetic disruption of AKAP150 in mice significantly reduces the co-immunoprecipitation of PKA with LTCC and prevents phosphorylation of Ser1928 upon β-AR stimulation in vivo [57]. Although different experimental results have emerged, it is suggested that PKA is important for the maintenance of LTCC activity. It has been observed that the dysfunction of LTCC is consistent with β-AR depression, and the normal function of β-AR may be the basis for the maintenance of LTCC activity.

SR and abnormal phosphorylation/dephosphorylation

SR is a Ca2+ store in cardiomyocytes, which is divided into longitudinal SR (LSR) parallel to myofibrils and junctional SR (JSR) in contact with the transverse tube. The coupling reaction of JSR and LTCC on the transverse tube is the trigger mechanism of myocardial contraction. The major phosphorylation sites of the SR are located at RyR2 and PLNs (located at the Ca2+ pump/SERCA-2), and their phosphorylation status is related to intracellular Ca2+ release and recycling [58].

RyR2 and abnormal phosphorylation/dephosphorylation

RyR2, a calcium channel in the SR in cardiomyocytes, is the most important component of myofilament contraction triggered by Ca2+ release during contraction [59]. Local regulation of RyR2 channels by PKA phosphorylation is an effective mechanism for regulating SR Ca2+ release. RyR2 is a tetramer composed of four 565,000 Dalton RyR2 peptides and four 12,000 Dalton FK-506 binding proteins (FKBP12.6). FKBP12s, which stabilizes RyR channel function [60] and facilitates coupled gating between adjacent RyR channels [61], are packaged into dense arrays of special regions of the SR that release intracellular stores of Ca2+ to trigger muscle contraction. One FKBP12 molecule binds to each RyR subunit, and dissociation of FKBP12 significantly changes the biophysical properties of the channel, resulting in the appearance of subconductance states and an increase in P0 due to increased sensitivity to Ca2+-dependent activation [60, 62], while dissociation of FKBP12 from RyR channels inhibits coupling gating, resulting in random gating of the channel rather than fusion [61].

Phosphorylation of RyR2 enhances Ca2+ release, and CaMKII-mediated hyperphosphorylation promotes the occurrence of HF. Previous studies have found that hyperphosphorylation of RyR2, specifically mediated by PKA, is present in HF accompanied by decreased phosphatase activity, resulting in the increasing activity of LTCC in the diastolic period and Ca2+ leakage through inhibition of FKBP12.6 binding [59]. Moreover, it was found that not only PKA but also CaMKII was involved in the hyperphosphorylation of RyR2 (Fig. 2A, B), leading to the enhancement of RyR2 activity. The phosphorylation sites included Ser2808, Ser2814, and Ser2815 [33, 63, 64]. However, ablation of the PKA-mediated phosphorylation at Ser2808 failed to prevent the progression of cardiac dysfunction [65, 66]. CaMKII-mediated ablation at Ser2814 prevented the progression of HF [63]. In animal experiments and human hearts, both PKA and CaMKII are involved in RyR2 phosphorylation in hypertrophic hearts. However, in failing hearts, RyR2 phosphorylation is mainly mediated by CaMKII and is accompanied by a higher level of Ca2+ leakage [63, 67]. Meanwhile, lower activity of PDE4D3 has been detected in HF cardiomyocytes, which also contributes to RyR2 hyperphosphorylation and HF progression [63]. It can be speculated that the progression of HF is mainly related to late CaMKII-mediated hyperphosphorylation and reduced PDE4D3 activity. The early PKA-mediated phosphorylation of RyR2 enhances the activity of LTCC and facilitates the release of more Ca2+ to adapt to the higher strength of muscle contraction, but this requires the cooperation of the enhanced function of SERCA-2. The CaMKII-mediated hyperphosphorylation of RyR2 results in Ca2+ overload and decreases Ca2+ transients.

SERCA-2 and phosphorylation/dephosphorylation signals

SERCA-2, located in the SR, is the most important component of Ca2+ recycling during diastole. SERCA-2a is the most important subtype in adult cardiomyocytes [68]. SERCA-2a can transport cytosolic Ca2+ into the SR through ATPase activity, which keeps the cardiomyocyte’s low Ca2+ concentration in the diastolic state and provides the necessary conditions for myocardial contraction [69]. PLN, a small, reversibly phosphorylated transmembrane protein located in the cardiac SR [68], was identified as a major substrate of cAMP-dependent kinases and a regulator of the SERCA-2 [69,70,71,72,73,74,75,76], which can be phosphorylated by PKA and CaMKII (Fig. 2A, B). The main effect of dephosphorylated PLN association with SERCA-2a is to reduce the apparent affinity of SERCA-2a for Ca2+. Alleviation of SERCA-2a inhibition by PLN is a major contributor to the positive inotropic and exotropic effects of β-agonists [77,78,79,80,81].

PLN phosphorylation enhances SERCA-2a activity, and CaMKII-mediated PLN hyperphosphorylation promotes the progression of HF. The phosphorylation of PLN at Ser16 and Thr17 mediated by PKA and CaMKII has been shown to inhibit PLN activity and enhance SERCA-2 activity, which is conducive to the rapid recycling of cytosolic Ca2+, shortening the relaxation [12] and contraction cycles to adjust to the cytosolic hypercalcemia environment. The dephosphorylated form of the PLN inhibits SERCA-2a activity [12, 69, 70]. Although some animal studies support that PLN phosphorylation is reduced in HF [82], others have found that PLN phosphorylation is increased [83, 84], which may be related to the type of animal studied and the stage of HF. The major site of increased phosphorylation in animal models of HF is located at Thr17, and the increase of Thr17 phosphorylation in the HF group is negatively correlated with myocardial contractility. It was also found that 24-hour continuous induction of Ca2+ transient also promoted Thr17 phosphorylation and decreased Ser16 phosphorylation and inotropic drug response [85]. Consistent with findings in human HF samples [68, 86], decreased phosphorylation was found at Ser16 in animal models of HF [82]. This suggests that CaMKII-mediated PLN phosphorylation, but not PKA, is associated with HF progression. Meanwhile, PP1β knockdown could increase the phosphorylation of PLN [87], and increase PP1 activity in HF (Fig. 2B) [17, 18] probably reducing PLN phosphorylation and impaired Ca2+ pump function.

Myofilament and abnormal phosphorylation/dephosphorylation

The sarcomere is the basic unit of myofibril. The sarcomere consists of three different myofilament systems. The components of the thick filament system are myosin and cMyBP-C. The thin myofilament system is composed of monomers of myosin (attached to myosin), tropomyosin, and troponin. Concomitant myofibrin and titin maintain the physical structure of sarcomeres. According to the sliding filament theory, the combination of troponin and Ca2+ affects tropomyosin, exposing the binding sites of actin and myosin, and then actin slides along the myosin. Myofilament is also an important part susceptible to phosphorylation in HF.

Myosin and abnormal phosphorylation/dephosphorylation

Myosin is the main component of the thick filaments, consisting of two heavy chains and four light chains, which have ATPase activity. In failing human hearts, there was no difference in MLC-1 phosphorylation levels compared to normal hearts, while MLC-2 phosphorylation levels were significantly reduced, which is associated with increased Ca2+-sensitivity (Fig. 1) [88]. The maximum tension does not alter, although with increased Ca2+-sensitivity [88]. It was discovered that phosphorylation of myosin regulatory light chain (RLC) and phosphorylation of cRLC enhanced the Ca2+-sensitivity [89, 90], which is related to increased contractility [91].

Located between myosin and actin, cMyBP-C can respond to PKC, PKA, and CaMKII (Fig. 2A, B). Phosphorylation of the CaMKII signal is attributed to enhanced filament contractility. Previous studies have shown that healthy human hearts have higher levels of cMyBP-C phosphorylation, whereas failing human hearts have lower levels of phosphorylation [92]. Independent of cTnI, PKA phosphorylation of cMyBP-C accelerated cross-bridge kinetics [93]. The cMyBP-C can be dephosphorylated in response to cholinergic signaling in HF, which is related to calcineurin overexpression [94]. The rate of force development and filament activation were found to be inhibited by the dephosphorylation form of the cMyBP-C phosphorylation sites Ser273, Ser282, and Ser302 [95], with Ser282 possibly having the greatest impact [96].

Troponin and abnormal phosphorylation/dephosphorylation

Troponin includes three subunits: troponin T (cTnT), troponin C (cTnC), and troponin I (cTnI), of which cTnT is a tropomyosin-binding subunit and cTnC is a Ca2+-binding subunit. The role of cTnI is to prevent myosin and actin from binding to one another. Experiments showed that cTnT was phosphorylated at the same level in cardiomyocytes of human failing and normal hearts, while cTnI was phosphorylated at a lower level in the failing hearts [88, 97].

Human cTnI contains 209 amino acids, including 12 Ser residues, 8 Thr residues, and 3 Tyr residues, and the phosphorylation of cTnI is mediated by PKA and PKC (Fig. 2A, B). The PKA-mediated phosphorylation of cTnI is associated with increased length-dependence and decreased Ca2+-sensitivity [98], while the PKC-mediated phosphorylation may be associated with cardiac disorder [99]. The percentage of cTnI dephosphorylated states was found to be elevated in human failing hearts compared to normal hearts [88]. A study found that in human failing hearts, phosphorylation decreased in some sites of cTnI including Ser5, Ser6, Ser5/Ser6 duplex, Ser23, Ser24, Ser23 /Ser24, Tyr26, while increased in Ser42, Ser44, Thr51, Ser77, Thr78, Ser77/Thr78, Thr143, Ser166, Thr181, and Ser199 [100]. It was also found that pseudo-phosphorylation at both Ser42/44 and Ser23/24 reduced myofilament Ca2+-sensitivity [101]. Double phosphorylation at Ser23/24 is essential for reducing Ca2+-sensitivity, whereas phosphorylation at a single Ser23 or 24 is not [102]. However, a later experiment found that a single Ser23 or 24 phosphorylation was sufficient to reduce Ca2+-sensitivity [103]. Pseudo-phosphorylation of Ser42/44, the PKC phosphorylation sites, is linked to a greater decrease in myofilament Ca2+-sensitivity [101]. Meanwhile, the pseudo-phosphorylation of Ser42/44 weakened the length dependence and blunted the length dependence mediated by PKA, while the pseudo-phosphorylation at Ser23/24 enhanced this length dependence and could be reinforced by PKA [101]. In another experiment, PKCα and phosphorylation levels at Ser44 were found to increase in human and rat failing hearts, which decreased following the implantation of a ventricular assist device [104]. Increased phosphorylation of Ser23/24 and Ser150 was found in the ischemic myocardium [105]. At neutral PH (PH = 7), cardiomyocytes with phosphorylation at Ser150 showed higher Ca2+-sensitivity, whereas phosphorylation at Ser23/24 showed lower Ca2+-sensitivity. However, co-phosphorylation of Ser23/24/150 alleviated the low Ca2+-sensitivity caused by Ser23/24 at neutral PH (PH = 7) [105, 106]. The presence of a phosphorylated acidic environment was shown to attenuate myofilament Ca2+-sensitivity. At acidic PH (PH = 6.5), a greater decrease of Ca2+-sensitivity was found when Ser23/24 and Ser23/24/150 were phosphorylated than single phosphorylation of Ser150, indicating that phosphorylation at Ser150 enhanced the tolerance of cardiomyocytes to an acidic environment [105]. It was also found that phosphorylation of Ser23/24 accelerated the rate of Ca2+ dissociation from troponin, whereas phosphorylation of Ser150 blunted this increase. Independent of the acidic environment, the presence of Ser150 phosphorylation slowed the speed of Ca2+ dissociation from troponin [105]. Another study found that, similar to phosphorylation at Ser23/24, phosphorylation at Tyr26 reduced Ca2+-sensitivity while accelerating the rate of dissociation of Ca2+ from troponin. It’s also has been found that co-phosphorylation of Ser23/24 and Tyr26 did not further reduce Ca2+-sensitivity but further accelerated the rate of dissociation of Ca2+ from troponin [27]. Compared with non-failing myocardium, Ser199 phosphorylation was increased in end-stage HF [100]. Ser199 was found to be phosphorylated mainly mediated by PKC [107], and its elevated phosphorylation increased myofilament Ca2+-sensitivity without affecting its length dependence [108]. Phosphorylation at Thr143 is also mediated by PKC, and pseudo-phosphorylation of Thr143 increases Ca2+-sensitivity but does not alter length dependence [109].

Phosphorylation of cTnT can be mediated by PKC, among which PKCa has four phosphorylation sites on cTnT including Thr197, Ser201, Thr206, and Thr287, and phosphorylation of these sites is associated with decreased myofilament Ca2+ sensitivity [110, 111]. The functional consequences of the phosphorylation of Thr144 were unknown [111]. Although PKC phosphorylation can play different roles in cTnT/cTnI, it is mainly related to elevated Ca2+-sensitivity [107].

Titin and abnormal phosphorylation/dephosphorylation

The primary function of titin is to maintain the integrity and stability of myofibrils. HF is often accompanied by increases in myocardial stiffness-based titin [112, 113].

The increased phosphorylation of Ser4043 and Ser12884, which can be phosphorylated by CaMKII, was found in the failing heart [114], suggesting that excessive activation of CaMKII is an adaptive response in HF patients and HF animal models [114, 115]. Titin is also a substrate of PKG, of which PKG1α is the major isoform expressed in the myocardium and involved in the phosphorylation of several cardiac target proteins, playing a key role in the sarcomole [116]. PKG has been shown to phosphorylate the N2Bus region in titin, thereby reducing titin-based myocardial stiffness [112]. As for the phosphorylation modification of Titin, in addition to the two major kinases, other kinases are also involved in the post-translational modification of Titin. In HFrEF rats, the kinase PKCα showed increased activity in the ventricle and was shown to phosphorylate the PEVK element of titin [117]. PKCα-dependent phosphorylation at Ser11878 in the PEVK-titin fragment was found to be hyperphosphorylated in the HFpEF animal model [118], enhancing the passive tension of titin. PKD, through its mediated phosphorylation, regulates heat shock protein 27 to alleviate titin aggregation, thereby inhibiting titin-dependent cardiomyocyte stiffness [118]. PKA and extracellular regulating kinases 2 (ERK2) were shown to phosphorylate titin springs at specific sites within heart-specific N2-Bus elements, and this modification alters the molecular stiffness of N2-Bus [118].

Treatment directions focusing on phosphorylation/dephosphorylation

Restoration of the balance of key kinases and phosphatases is important to the normal Ca2+ function in cardiomyocytes, which is highly related to the phosphorylation-dephosphorylation balance. Here are some treatments based on the phosphorylation-dephosphorylation disorder to treat HF (Table 1).

Table 1 Drugs and potential treatments for HF based on abnormal phosphorylation /dephosphorylation in cardiomyocyte

The increase in PP1 activity observed in HF was associated with decreased Inhibitor-1 phosphorylation as well as increased I-2 phosphorylation [119,120,121]. Inhibitor-1 (I-1) is the first putative endogenous inhibitor of PP1 (Table 1) [122], and I-1-deficient mice exhibit increased PP1 activity, decreased cardiac function, blunted β-AR response, and reduced PLN phosphorylation (Table 1) [119]. PP1 can be phosphorylated after treatment with isoproterenol [123,124,125], resulting in reduced PP1 activity; PP1 can also be dephosphorylated by PP2A and PP2B, which allows restoration of function to basal levels by relieving PP1 inhibition [126]. When PKA is phosphorylated at Thr35, I-1 efficiently inhibits PP1 activity [122, 127, 128]. However, the phosphorylation of PKCα at Ser67 and Thr75 of I-1 was associated with increased PP1 activity and decreased contractility in vivo [129, 130]. Inhibitor-2 (I-2) is a heat-stable phosphoprotein similar to I-1 [122]. Expression of I-2 resulted in reduced PP1 activity is associated with enhanced contractile parameters and increased instantaneous kinetics of Ca2+, which shows that it manifested by increased phosphorylation at Ser16 of PLN. But not at Thr17, suggesting that the PP1c/I-2 complex may preferentially dephosphorylate the PKA sites in the PLN [131, 132]. Studies on the external application of PP1 inhibitors are also ongoing. It was found that adenovirus-mediated I-35 in the truncated form of I-1c expression which lacks Ser67 and Thr75 enhances the contractile response and Ca2+ dynamics in human failing myocytes [133]. It was also found to attenuate the progression of HF in experimental mice, which was characterized by a reduced degree of cardiac hypertrophy. Similar results were found in I-2 as a therapeutic modality [18]. These beneficial effects are mediated by the enhanced phosphorylation of PLN, while the phosphorylation level of RyR2 remains unchanged. This may be important at the therapeutic level, as increased RyR phosphorylation may potentially lead to diastolic leakage and arrhythmogenic activity [134, 135].

Changes in PDE in HF are diverse, as mentioned above, levels of PDE1, PDE2, PDE5, PDE9, and PDE10 are increased in pathological hypertrophy and HF. However, the PDE3 and PDE4 changes are diverse, and their overall reduced activity is correlated with hypertrophy. Treatment options specific to PDE are also constantly emerging. It has been clinically demonstrated that inhibitors of PDE3 and PDE5 are ineffective [136]. PDE1 enzymes bind to and hydrolyze cAMP and cGMP in a mutually competitive manner. PDE1A regulates the unique cAMP and cGMP pools, predominantly in the perinuclear and nuclear regions of cardiac fibroblasts. Inhibitors of PDE1 or PDE1A gene silencing have been shown to inhibit the adrenalin-induced reduction in PKG activity [137]. PDE1A is also upregulated in cardiac fibroblasts activated by profibrotic stimuli, and inhibition or silencing of PDE1A was shown to limit myofibroblast transformation and the synthesis of extracellular matrix [138]. The pan-PDE1 inhibitor vinpocetine prevented cardiomyocyte hypertrophy and fibroblast activation, thereby blunting pathological remodeling induced by angiotensin II (Table 1) [139]. ITI214 is a drug directed against the PDE1 inhibitor, which can produce acute inotropic and lusitropic effects by promoting a cAMP pool independent of β-AR signaling and increasing ICa2+ (Table 1). However, it was less associated with Ca2+ transients and myofilament phosphorylation [140]. ITI-214 augmented cardiac inotropy, cardiac output, and heart rate (Table 1) [141]. PDE2 has a low affinity for cAMP and cGMP but a high hydrolytic capacity. Increased expression of PDE2 and the hydrolysis activation of cAMP were associated with a diminished β-AR response [142,143,144], suggesting a possible association with the progression of HF [144]. An antihypertrophic effect of PDE2 inhibition was reported in isolated cardiomyocytes [145, 146]. Inhibition of PDE2 was also shown to counteract cardiac hypertrophy and pathological remodeling (Table 1), particularly in fibrosis. In contrast, cardiomyocyte activation by PDE2 may favor ischemic HF to improve Ca2+ homeostasis, limit systolic dysfunction, and prevent arrhythmia [147, 148]. PDE2 cardiomyocytes may be beneficial to counteract the pressure overload caused by pathological remodeling [145,146,147, 149]. Isoforms of PDE4 and PDE3 were found to control Ca2+ release and reuptake in the SR by RyR2 and SERCA-2 respectively [20,21,22,23], or localized to the myofilaments [150, 151] or the nuclear envelope [23, 152, 153]. PDE3 activity is present in the cytosolic and microsomal fractions and constitutes the majority of cAMP-hydrolysing activity in the latter [154]. PDE3 regulates ICa2+ [155,156,157,158,159] and Ca2+ uptake in the SR by modulating cAMP-PKA [160, 161]. The role of its expression in HF was different in different experiments, with some experiments finding that its activity is unchanged in human failing hearts [162, 163], while others have found that its activity is reduced [164]. Some animal experiments reported reduced PDE 3 activity in HF [164,165,166,167,168,169], while others found elevated PDE 3 activity and expression [170,171,172,173]. However, PDE3 inhibition was shown to increase the incidence of arrhythmias in patients [174, 175]. PDE4D3 localizes within the RyR2 macromolecular complex, whose activity occurs mainly in the nuclear membrane. This localized pool of PDE4 also controls the integration of β-AR to AMP-PKA signaling in the nucleus. PDE4 activity can be regulated by PKA phosphorylation or by MAPK1 [176, 177], which is a major negative regulator of β-AR responses in healthy rat cardiomyocytes [156, 178], and the conversion from PDE4 to PDE3 occurs in cardiac hypertrophy and HF [166, 179, 180]. Reduced PDE4 activity in HF increased RyR2 phosphorylation and promoted Ca2+ leakage [22]. On the other hand, this promotes the phosphorylation of SERCA-2, contributes to Ca2+ uptake, and may be beneficial for HF [82, 86, 181,182,183]. PDE5 is localized at the Z-band [184], which is cGMP-activated and specifically hydrolyzes cGMP. PDE5 expression is elevated in animal and human failing hearts [185,186,187,188,189], although some experiments found it to be reduced [190]. PDE5 inhibition provided cardioprotection by promoting cGMP-PKG signaling to prevent cardiomyocyte hypertrophy and cardiac dysfunction [191]. PDE5 inhibition was also shown to attenuate diastolic dysfunction and decrease fibrosis and collagen type I deposition [192, 193]. The cGMP-PKG pathway under the control of PDE5 counteracts the effects of adverse cardiac remodeling (Table 1). PDE8 is a high-affinity, cAMP-specific enzyme [194, 195]. PDE8A was found to regulate excitation-contraction coupling by controlling a specific pool of cAMP involved in β-AR regulation of Ca2+ homeostasis [196]. PDE9 is highly specific for cGMP hydrolysis and is mainly located on the membrane, transducing the np-coupled signal. PDE9 inhibition may lead to improved diastolic function and impaired systolic function (Table 1) [197]. Animal studies have demonstrated that the inhibition of PDE 9 contributes to cardiac output [198]. PDE10 is upregulated in failing hearts in animals and humans [199], and PDE10 was shown to reduce epinephrine-induced cardiac hypertrophy [199]. However, other experiments simultaneously suggested that PDE overexpression exerted protective effects on the heart and reduced cardiac hypertrophy and cardiac hypertrophy caused by β-AR stimulation [143, 147], which can be associated with decreased phosphorylation of RyR2 and reduced ER Ca2+ leaks in early stages of HF or ischemic heart disease, preventing the occurrence of arrhythmia.

PP2A can dephosphorylate many sites on LTCC [200,201,202,203], RyR2 [204, 205], as well as TnI, TnT, and MyCP-c [206,207,208,209], which regulates the relaxation and contraction capacity of cardiomyocytes. The relationship between reduced LTCC activity and altered PP2A activity is unclear at present, and the treatment of PP2A is mainly in the cancer field. Whether PP2A inhibition is helpful in the treatment of HF still needs further experiments to be proven.

PKA is important in the progression of HF, and the restoration of its activity depends on the normal function of β-AR. However, for PKA, whether promoting the activity of PKA alone is helpful in the treatment of HF, more animal experiments are needed. PKG is another important kinase that is related to the phosphorylation of MLC [210]. PKG decrease is associated with HFpEF (with normal systolic function but impaired diastolic function, because PKG facilitates Ca2+ recovery into ER) [211]. PKG activation has broad prospects [212] (Table 1), and PKG may be the best alternative in the absence of normal β-AR function. PKC and CaMKII are activated in HF. Experiments proved that infusion of the oral PKCα/β/γ inhibitor ruboxistaurin increased contractility in wild-type and PKCβγ (- / -) mice, but not in PKCα (- / -) mice, which showed that the inhibitory effect of PKCα improved HF (Table 1) [8]. The same effects were found in other animals [213], providing a new direction for HF treatment. Hesperadin is a CaMKII inhibitor, and its application ameliorates cardiomyocyte injury and HF (Table 1) [214]. CaMKII activation is mainly associated with ER dysfunction, and its inhibitory effect may help to restore ER function. CaMKII inhibitors have broad prospects for the treatment of HF.

As shown in Fig. 3, a comprehensive treatment plan targeted abnormal kinase and phosphatase can be started by PKG activation and PP1 inhibition, for they are benefical to restoring the function of SERCA-2. CaMKII inhibition is also important because former studies indicated a relationships between activation of CaMKII and ER dysfunction and myofilament dysfunction. Although PKCα is activated in HF, it is mainly related to a rising Ca2+ sensitivity. PKCα inhibition can be considered. PDE inhibitiors should be used with caution, because they may cause arrhythmias. Future experiments are expected to prove the effect of the potential treatment plan.

Fig. 3
figure 3

Potential treatment plan targeted on abnormal kinase and phosphatase. Conventional treatments including diuresis, angiotensin inhibition, and β-AR inhibition. Potential treatment targeted abnormal kinase and phosphatase should be sequential. PKG activation and PP1 inhibition can be firstly used for they are responsible for Ca2+ recovery. CaMKII inhibition is also important for it is beneficial to normal ER function. PKC inhibition is the last for it is mainly related to Ca2+ sensitivity recovery (a higher Ca2+ sensitivity may play a compensation role with relatively low intracellular Ca2+)

Conclusions

Calcium dysfunction because of abnormal phosphorylation and dephosphorylation is directly related to impaired myocardial systolic and diastolic functions. β-AR recovery is the end point of HF. Considering the relationships between Ca2+ and CaMKII and PKC, recovery of SR (Ca2+ release and recovery) is the most important, and PKG activation and PP1 inhibition may play a great role in the absence of normal β-AR function. Other potential therapeutic directions in HF can be focused on CaMKII and PKCα inhibitions with the removal of HF-leading risks, which can facilitate a normal Ca2+ function and have benign circular effects, and also improve the function of myofilament. Also, myofilament dysfunction is related to α-AR and M2-R activation. Inhibition of α-AR and M2-R inhibition can be considered, but it can cause many side effects. Treatment plans targeted on abnormal kinase/phosphatase should be sequential, for sudden Ca2+ decrease may cause a more serious systolic disorder. Future experiments are expected to prove the combined therapeutic effect and establish an appropriate dosage and course of treatment.