Abnormal LTCC function has been implicated in arrhythmogenesis. Gain of function mutations of Cav1.2α1c, as well as loss of function mutation of CaM (reduced Ca²⁺ sensitivity) were linked to hereditary Long QT syndrome type 8 and 14 (Venetucci et al., 2012; Crotti et al., 2013; Marsman et al., 2014). Changes in activation and inactivation parameters leading to widening of so called “window” current were linked to enhanced propensity of reactivation during late phases of AP and thereby generation of early after depolarizations (EADs) (Weiss et al., 2010). Reduction in LTCC expression levels is thought to promote arrhythmogenic Ca²⁺ alternans via reduced fidelity of channel coupling with RyR2s (Harvey and Hell, 2013). Interestingly, reduced LTCC expression levels in disease states are not always reflected by reduced current. For example, in ventricular cardiomyocytes from human failing hearts I_Ca was similar to controls, despite of a significant decrease in α1c expression levels, likely due to enhanced phosphorylation by PKA (Chen et al., 2002). Also, fidelity of LTCC-RyR2 coupling can be reduced due to structural remodeling and loss of T-tubules as in hypertrophy, myocardial infarct and HF (Wei et al., 2010).

The majority of studies using various models of diabetes did not find statistically significant changes in I_Ca with a few exceptions (Pereira et al., 2006; Lu et al., 2007). Pereira et al. (2006) showed that in db/db mice (Type 2), the decrease in I_Ca originates from a reduced number of channels in the sarcolemma. Similar results were obtained in the Akita mouse model (Type 1, Lu et al., 2007). In both models, steady state activation of I_Ca was shifted to more positive voltages which is expected to reduce ‘window current.’ However, in the latter model steady state inactivation was found shifted even further to the right, resulting in larger “window current.” The information as to whether LTCCs in diabetes undergo posttranslational modifications, or if their distribution with regard to RyR2s is altered, is scarce. While Shao et al. (2012) saw no T-tubular remodeling in STZ-diabetic rats, diminished T-tubular density was observed in db/db mice (Stølen et al., 2009). These findings, along with changes in LTCC in HF, are summarized in Table 1.

Under pathophysiological conditions, diastolic Ca²⁺ leak from the SR is increased, thus exceeding the critical threshold level and increasing RyR2 channel activation, which in turn activates other channels and results in proarrhythmic diastolic Ca²⁺ waves (Cheng et al., 1996). This subsequently activates NCX1, driving a net inward current that gives rise to delayed after depolarizations (DADs) and arrhythmias (Ferrier et al., 1973; Mechmann and Pott, 1986; Landstrom et al., 2017). Evidence suggests that increased refractory period shortening, thus increased RyR2-mediated SR Ca²⁺ leak, promotes Ca²⁺-dependent arrhythmias in failing hearts (Belevych et al., 2012; Brunello et al., 2013; Cooper et al., 2013).

Abnormal Ca²⁺ release is the driving force for arrhythmia observed in patients with catecholaminergic polymorphic ventricular tachycardia (CPVT), a condition characterized by pathogenic mutations in RyR2 (Priori et al., 2002; Priori and Chen, 2011), as well as in associated accessory proteins including CSQ (Terentyev et al., 2006; Nyegaard et al., 2012; Roux-Buisson et al., 2012). The condition presents under conditions of enhanced catecholaminergic drive, in the absence of any structural defects in the heart (Priori et al., 2002; Tester et al., 2005). Premature ventricular contractions (PVCs) that often manifest during exercise or periods of stress can degenerate into polymorphic or bidirectional ventricular tachycardia (VT) or fibrillation (VF) and subsequently lead to SCD. Mutations in RyR2 associated with CPVT usually cause increased SR Ca²⁺ leak that is exacerbated by β-adrenergic stimulation, but both gain- and loss-of-function mutations have been reported (Wehrens et al., 2003; Loaiza et al., 2013; Zhao et al., 2015; Landstrom et al., 2017; Uehara et al., 2017). Mutations in CSQ cause reduction or complete loss of protein expression, and without CSQ-mediated Ca²⁺ buffering within the SR, RyR2 channels are prone to spontaneous activation, offering a trigger for arrhythmia (Yano and Zarain-Herzberg, 1994; Lahat et al., 2001; Knollmann et al., 2006; Faggioni et al., 2012).

The significance of altered PKA-mediated phosphorylation of RyR2 function in the pathogenesis of cardiac arrhythmia remains controversial. Hyperphosphorylation of RyR2 at S2808 was originally hypothesized to cause dissociation of FKBP12.6, destabilizing the channel and thus increasing open channel probability (Marx et al., 2000). However, the significance of both phosphorylation at this site and the role of PKA-mediated phosphorylation on the function of RyR2 remains a disputed subject (Jiang et al., 2002; Benkusky et al., 2007; Curran et al., 2007; Belevych et al., 2011b; Bovo et al., 2017). The role of CaMKII-mediated Ry2 phosphorylation in modulating channel function is more strongly supported, with consensus that activation of CaMKII as opposed to PKA increases SR Ca²⁺ leak (Ai et al., 2005; Curran et al., 2007; Niggli et al., 2013), although this is not universal (Wehrens et al., 2006; Yang et al., 2007). It is also well established that chronic CaMKII activity in cardiac disease is a major regulator of RyR2 function with arrhythmogenic consequences (Ai et al., 2005; Zhang et al., 2005; Terentyev et al., 2009; Belevych et al., 2012; Respress et al., 2012; Uchinoumi et al., 2016). Additionally, oxidation as a posttranslational modification may alter RyR2 function in cardiac disease. In the healthy heart, oxidation may serve to transiently enhance Ca²⁺ release, increasing cardiac output (Niggli et al., 2013). However, in conditions of severe oxidative stress such as HF, increased RyR2 oxidation by ROS can lead to RyR2 activation and increased proarrhythmic SR Ca²⁺ leak (Mochizuki et al., 2007; Belevych et al., 2011a,b). Modulation of RyR2 activity by ROS can also be indirect, via the oxidation of CaMKII and subsequent increased CaMKII-mediated phosphorylation of the channel (Chelu et al., 2009; Anderson, 2015).

The SR Ca²⁺ load is thought to be a critical factor of cardiac alternans, a repetitive beat-to-beat fluctuation in cellular repolarization at a constant heart rate that closely linked to development of ventricular arrhythmias (Edwards and Blatter, 2014). Any impairment in the refractoriness of RyR2-mediated Ca²⁺ release or the recovery after channel inactivation may facilitate the onset of alternans, due to a reduction in the subsequent Ca²⁺ transient amplitude (Edwards and Blatter, 2014). These properties of RyR2 have been shown to be a major component underlying generation of alternans in both computational and experimental studies (Sobie et al., 2006; Nivala and Qu, 2012; Shkryl et al., 2012; Alvarez-Lacalle et al., 2013; Sun et al., 2018).

It is well established that increased SR Ca²⁺ leak due to enhanced RyR2 channel activity significantly contributes to arrhythmogenic potential in diabetic cardiomyopathy.

Injection of STZ destroys insulin-producing β cells and has long been used for the generation of Type 1 diabetes phenotypes. Early studies of RyR2-mediated Ca²⁺ release in diabetes utilized STZ-diabetic rat models. Work of Yu et al. (1994) showed that cardiomyocytes from STZ-diabetic rats had reduced maximum rates of shortening and relengthening, as well as depressed SR Ca²⁺ content. Using [³H] ryanodine binding assay as an indication of RyR2 channel functionality, isolated SR membranes from diabetic rats showed reduced high-affinity binding sites. Bidasee et al. (2001) also reported decreased [³H] ryanodine binding in 6-week post STZ injection, but posited this was due to dysfunctional RyR2 rather than a decrease in protein expression or mRNA level as observed in Teshima et al. (2000) and Choi et al. (2002). Interestingly, Zhao et al. (2014) found that Ca²⁺ spark frequency showed a gradual decline in correlation with progression of STZ-diabetes, with significant differences between 4-week and 12-week post-injection groups.

Yaras et al. (2005) importantly showed increased Ca²⁺ spark frequency in cardiomyocytes from STZ-diabetic rat, with a reduced Ca²⁺ transient amplitude and depressed SR Ca²⁺ loading. This was accompanied by significantly increased phosphorylation of RyR2 at S2808 and a ∼40% decrease in FKBP12.6 association. Similar findings were reported by other groups, where these phenomena were suggested to underscore depressed SR Ca²⁺ release in STZ-diabetic cardiomyocytes (Shao et al., 2007, 2009; Tuncay et al., 2014), indicative that Ca²⁺ leak via hyperactive RyR2 contributes to the disease phenotype. Later work showed that exercise training for 4 weeks could attenuate this, reducing S2808 phosphorylation and increase levels of FKBP12.6 expression (Shao et al., 2009). However, the functional role of PKA-mediated RyR2 phosphorylation remains controversial and many studies have shown phosphorylation at the S2808 site does not modulate channel activity in other cardiac disease states (Xiao et al., 2004; Guo et al., 2010). An increase in endogenous CaMKII-mediated phosphorylation of RyR2 has also been implicated in the aberrant Ca²⁺ handling observed in STZ-diabetic rats (Netticadan et al., 2001) and db/db mice (Stølen et al., 2009). Conversely, Tian et al. (2011) posited that gain-of-function changes in RyR2 function observed in single channel recordings were independent of phosphorylation at either S2808 or S2814 sites. Instead, the increase in open channel probability and 20% reduction in conductance was attributed to increased responsiveness to cytoplasmic activators including Ca²⁺, alterations in the threshold for activation by luminal Ca²⁺, and a blunted response to physiological inhibitors.

Other posttranslational modifications of RyR2 in Type 1 diabetes have also been suggested to underscore channel dysfunction. Bidasee et al. (2003a) suggested RyR2 dysfunction, evidenced by decreased [³H] ryanodine binding in STZ-diabetic rats, was in part due to formation of disulfide bonds between adjacent sulfhydryl groups. The same group also showed that non-cross-linking advanced glycation end products (AGEs) on RyR2 are significantly increased in diabetic heart tissue, and this increase could be partially attenuated with insulin treatment (Bidasee et al., 2003b). Extensive carbonylation of RyR2 by increased reactive carbonyl species (RCS) in STZ- diabetic rats was suggested to reduce the responsiveness of RyR2 to cytoplasmic Ca²⁺. While expression levels of RyR2 remained unchanged, there was an increase in non-functional RyR2 channels, but also an increase in the activity of others. The increased heterogeneity of RyR2 channels was posited to increase spontaneous and dyssynchronous SR Ca²⁺ release in isolated cardiomyocytes, thus providing a trigger for arrhythmia. Treatment of diabetic rats with RCS scavengers attenuated spontaneous SR Ca²⁺ release, reduced RyR2 carbonylation and normalized channel functionality.

Fewer studies have investigated changes in RyR2-mediated Ca²⁺ handling in models of Type 2 diabetes. In the non-failing myocardium of type 2 diabetic patients RyR2 protein expression was decreased, while mRNA levels were decreased in the Goto-Kakizaki model (Reuter et al., 2008; Gaber et al., 2014). In a prediabetic model of metabolic syndrome, whereby dogs were chronically fed a high fat diet (HFD), phosphorylation of RyR2 at S2808 was significantly elevated and the channel’s ability to bind [³H] ryanodine significantly depressed in the ventricles compared to healthy controls, while no changes in RyR2 mRNA or protein expression were observed (Dinçer et al., 2006). Okatan et al. (2016) also observed increased RyR2 phosphorylation at S2808 in rats with high sucrose diet-induced metabolic syndrome, accompanied by reduced FKBP12.6 expression. Through studies of electric-field stimulated intracellular Ca²⁺ handling, cardiomyocytes isolated from rats with metabolic syndrome showed significantly increased SR Ca²⁺ leak, depressed SR Ca²⁺ loading and reduced Ca²⁺ transient amplitude vs. controls. This data is suggestive that alterations in RyR2 function and SR Ca²⁺ release may be an important mechanism of early cardiac dysfunction in insulin resistance and diabetes development.

Abnormal intracellular lipid concentration is a hallmark of both obesity and diabetes, and Joseph et al. (2016) recently studied Ca²⁺ handling in cardiomyocytes from a transgenic model of cardiac lipid overload, with peroxisome proliferator-activated receptor-γ (PPARg) overexpression. This revealed increased Ca²⁺ spark activity compared to controls that could be reduced by application of antioxidant mitoTEMPO. A significant increase in mitochondrial oxidative stress was suggested to increase RyR2 oxidation and subsequent SR Ca²⁺ release. In 8-week mice with HFD-induced obesity, Sánchez et al. (2018) reported a shift in the distribution of single RyR2 channel responsiveness to activating cytosolic [Ca²⁺], whereby channels were much more active to those isolated from control mice. No changes were observed in RyR2 expression levels or phosphorylation status at S2808 or S2814 sites. Instead, this phenomenon was attributed to significantly increased RyR2 oxidation in HFD-mice, implicating the diabetes-related increase in oxidative stress in abnormal Ca²⁺ handling. Changes in RyR2 in HF and diabetes are summarized in Table 2.

In HF, impaired SERCa2a expression and activity blunts Ca²⁺ transient amplitude and rate of decay (Winslow et al., 1999). Reduced sequestration of Ca²⁺ into the SR drives Ca²⁺ extrusion from the cardiomyocyte via NCX1. This generates a net inward depolarizing current that can prolong action potential and thus facilitate triggered activity (O’Rourke et al., 1999; Bers et al., 2002). Diminished SR Ca²⁺ uptake by SERCa2a is also associated with the initiation of cardiac alternans (Merchant and Armoundas, 2012; Nivala and Qu, 2012).

While it may seem counterintuitive to increase SR Ca²⁺ uptake as a therapeutic strategy, given that SR Ca²⁺ overload may exacerbate SR Ca²⁺ leak through RyR2, accumulated evidence suggests the contrary. Upregulation of SERCa2a in small and large animal studies has been shown to protect against development of arrhythmias, improve contractile function and normalize intracellular Ca²⁺ handling (Meyer and Dillmann, 1998; del Monte et al., 2001, 2004; Suarez et al., 2004; Prunier et al., 2008; Lyon et al., 2011; Fernandez-Tenorio and Niggli, 2018). Multicenter SERCa2a gene therapy trials in humans with HF have been completed, but with limited success (Jaski et al., 2009; Jessup et al., 2011; Zsebo et al., 2014; Greenberg et al., 2016). Increased SERCa2a activity also suppressed cardiac alternans in both computational models and experimental studies (Cutler et al., 2009; Stary et al., 2016).

Diminished SR Ca²⁺ uptake has been identified as a primary mechanism for decreased cardiac contractility observed in diabetic cardiomyopathy (Ganguly et al., 1983). Dysfunction of SERCa2a has been established at early stages of type 1 diabetes development and is primarily ascribed to decreased mRNA levels or expression of the protein, resulting in reduced Ca²⁺ transient amplitudes and a slower rate of transient decay (Ganguly et al., 1983; Trost et al., 2002; Bidasee et al., 2004; Hu et al., 2005; Lacombe et al., 2007). Increased oxidative stress and intracellular ROS concentrations in diabetic hearts can reduce SERCa2a activity by oxidizing Cysteine 674, as well as interfering with the ATP binding site (Xu et al., 1997; Ying et al., 2008). Other molecular mechanisms suggested to drive SERCa2a downregulation include increased carbonylation, glycation and O-GlcNAcylation (Bidasee et al., 2004; Hu et al., 2005; Shao et al., 2011). Changes in PLB expression and phosphorylation have also been reported in STZ-diabetic rats, but this finding is not universal (Wold et al., 2005). Gene therapy with recombinant PLB antibody, which could mimic PLB phosphorylation thus activate SERCa2a, was also shown to increase the rate of whole heart relaxation, contraction and pressure development in a diabetic mouse model and cardiomyopathic hamsters (Meyer et al., 2004; Dieterle et al., 2005).

The pathophysiological role of SERCa2a in cardiomyopathy observed in type 2 diabetes remains to be fully elucidated, although activity is mostly reduced in various models (Zarain-Herzberg et al., 2014). Decreased SERCa2a mRNA levels were observed in the obese fa/fa rats while levels of protein, but not mRNA, were reduced in Otsuka Long-Evans Tokushima Fatty (OLETF) rats. No changes in protein expression were observed in the db/db mouse, but rather an increased PLB expression was posited to account for diminished SR Ca²⁺ uptake in this model. Stølen et al. (2009) reported reduced SERCa2a protein expression, increased PLB phosphorylation and overall reduced SERCa2a Ca²⁺ uptake in db/db mice. Conversely, increased expression of SERCa2a and reduced PLB mRNA was observed in ZDF rats, which could be further increased with insulin treatment in a concentration-dependent manner (Fredersdorf et al., 2012). Upregulation of SR Ca²⁺ uptake via SERCa2a may offer protection in early phases of disease development, countering volume overload and impaired relaxation (Fredersdorf et al., 2012; Zarain-Herzberg et al., 2014). Changes in SERCa2a expression and function in both HF and diabetes are summarized in Table 3.

Enhanced expression and activity of NCX1 is thought to be one of the major causes of increased arrhythmogenesis in HF (Pogwizd et al., 1999; Bers et al., 2002). NCX1 translates intracellular [Ca²⁺] during spontaneous Ca²⁺ waves into depolarizations, i.e., DADs, which can lead to activation of Na²⁺ channels and extrasystolic action potentials. During systole enhanced NCX1 prolongs APD allowing LTCCs to reactivate thereby contributing to generation of EADs. Increased NCX1-mediated Ca²⁺ influx during reverse mode due to increased intracellular [Na²⁺] characteristic of HF or ischemia may increase cytosolic Ca²⁺ and thereby activity of RyR2s (Satoh et al., 2000; Szepesi et al., 2015). Pharmacological inhibition of NCX1 substantially reduced triggered activity in various models of ventricular arrhythmia with perturbed Ca²⁺ homeostasis (Bourgonje et al., 2013; Jost et al., 2013; Nagy et al., 2014; Zhong et al., 2018). Experiments using mouse ventricular myocytes with reduced NCX1 demonstrated that genetic inhibition of NCX1 suppressed arrhythmogenic after depolarizations (Bögeholz et al., 2015), while transgenic overexpression promoted generation of EADs due to prolonged repolarization and spontaneous action potentials (Pott et al., 2012).

The levels and activity of NCX1 vary in different diabetes models. Alloxan or STZ- injected mice with diabetes type 1 showed reduction in NCX1 activity, reduced expression and mRNA levels (Makino et al., 1987; Pierce et al., 1990; Golfman et al., 1998; Hattori et al., 2000). A number of studies shows increased intracellular [Na⁺] in myocytes from diabetic hearts which enhances NCX1-mediated Ca²⁺ influx and reduces extrusion (Doliba et al., 2000; Wickley et al., 2007), but reduced NCX1 expression levels have also been reported (Bilginoglu et al., 2013). Belke et al. (2004) using the diabetes type 2 db/db mouse model showed no difference in NCX1 activity. In db/db mice, Stølen et al. (2009) reported increased NCX1 function. Insulin-resistant sucrose-fed rats with diastolic dysfunction showed normal expression and function of NCX1 (Wold et al., 2005). No changes in I_NCX were reported in high fat diet mouse model (Ricci et al., 2006). An increase in mRNA levels of NCX1 was shown in human patients with diabetes type 2 (Ashrafi et al., 2017). A similar increase was demonstrated in the mouse model with lipid overload (Joseph et al., 2016) and diabetes type 1 Akita(ins2) mouse model of cardiomyopathy (LaRocca et al., 2012). Alterations in NCX1 expression and function in HF and diabetes are summarized in Table 4.

Strategies to manage diabetes include insulin replacement and aerobic exercise regimes (Winnick et al., 2008). Insulin elicits positive inotropic effects on the myocardium, inducing SERCa2a activity (Pierce et al., 1985), while endurance exercise has been shown to improve intracellular Ca²⁺ cycling and protect against oxidative stress in diabetic animal models (Kim et al., 1996; Shao et al., 2009; Stølen et al., 2009). Combined therapy was shown to increase expression of Ca²⁺ handling proteins, restore Ca²⁺ handling and improve basal cardiac function in type 1 diabetic rats (Le Douairon Lahaye et al., 2012; da Silva et al., 2015). However, due to the multifaceted nature of the disease, diabetes remains a prime risk factor for cardiovascular disease despite beneficial effects of these frontline treatments (Leon and Maddox, 2015).

During the progression of diabetic cardiomyopathy, hyperglycemia is known to increase activity of the renin-angiotensin system (RAS) (Sechi et al., 1994; Ohishi, 2018). Angiotensin II has been shown to have direct effects on cardiomyocytes, with activation of its receptor (AT1) increasing generation of oxidants and increasing intracellular [Ca²⁺], as well as activating protein kinase C (PKC) and PKA (Malhotra et al., 1997; de Lannoy et al., 1998; Dostal, 2000; Raimondi et al., 2004). Angiotensin-converting enzyme (ACE) inhibitors and AT1 blockers are used to prevent hypertension and cardiovascular disease in diabetic patients (Mancia et al., 2013; Singh et al., 2018). Antagonism of angiotensin II was demonstrated to alleviate diminished SERCa2a activity in HF models (Okuda et al., 2004; Gassanov et al., 2006), while administration of AT1 blockers to STZ-diabetic rat cardiomyocytes reduced cellular oxidative stress as well as phosphorylation levels of RyR2, improving Ca²⁺ homeostasis (Privratsky et al., 2003; Yaras et al., 2007; Ozdemir et al., 2009).

Current and conventional treatments of diabetes are limited in terms of preventing ventricular arrhythmias and SCD. Blockade of the β-adrenergic stimulation cascade remains a primary treatment of HF in a bid to reduce arrhythmogenic responses, including those of Ca²⁺ handling proteins (Klapholz, 2009; Rehsia and Dhalla, 2010). However, use of antagonists in diabetic patients remains controversial, given β-blockers have been associated with increased risk for cardiovascular events in diabetic patients and can have hypoglycemic side effects (Casiglia and Tikhonoff, 2017; Tsujimoto et al., 2017).

As it stands, inhibition of LTCC or NCX1 does not appear to be an appropriate strategy in ameliorating Ca²⁺ mishandling in diabetic hearts. Current density of LTCC is reported as unaltered or reduced in diabetic models (Pereira et al., 2006; Lu et al., 2007). A reduction of I_Ca in this setting would diminish the trigger for RyR2-mediated Ca²⁺ release and thereby may exacerbate systolic dysfunction and increase propensity to Ca²⁺ alternans. Pharmacological inhibition of NCX1 is generally viewed as beneficial since it reduces Ca²⁺ influx during reverse mode limiting Ca²⁺ overload and attenuates depolarization during forward mode reducing triggered activity (Antoons et al., 2012). However, if NCX1 is already diminished as shown in many models of diabetes (Makino et al., 1987; Pierce et al., 1990; Golfman et al., 1998; Hattori et al., 2000) additional inhibition was proven to be detrimental leading to adverse accumulation of Ca²⁺ in cytosol and cell death (Bögeholz et al., 2017).

SERCa2a overexpression and/or enhancement may be a more suitable approach to ameliorate reduced contractility in diabetic cardiomyopathy, given the majority of studies report diminished pump activity. Insulin treatment has been shown to restore SERCa2a expression levels and improve Ca²⁺ homeostasis in obese type 2 diabetic rats (Fredersdorf et al., 2012). Transgenic overexpression of SERCa2a (or SERCa1a) in diabetic models increased SR Ca²⁺ uptake and attenuated diminished contractile function (Trost et al., 2002; Vetter et al., 2002; Waller et al., 2015). As previously discussed, studies using adenoviral mediated SERCa2a gene transfer in both small and large animal models of HF demonstrated similar results (del Monte et al., 2001, 2004; Lyon et al., 2011). While gene transfer via adeno-associated virus showed promise in the first human trial, it has shown limited success in others. Improvements and advances in gene therapy technology are likely to facilitate efficient and effective strategies to treat cardiac disease in the future (Ishikawa and Hajjar, 2017).

To enhance SERCa2a activity one could also modulate the inhibitory action of the accessory protein PLB. Ablation or knockdown of PLB has been demonstrated to suppress pro-arrhythmic Ca²⁺ waves generation in a model of CPVT (Bai et al., 2013), improve mortality rates in CSQ-transgenic mice [severe HF model, (Kaneko et al., 2016)] and importantly, improve contractile function in failing human cardiomyocytes (del Monte et al., 2002). However, ablation has not alleviated HF development in all models and a mutant form of PLB unable to inhibit SERCa has been linked to lethal dilated cardiomyopathy in humans (Haghighi et al., 2003; Song et al., 2003; Sipido and Vangheluwe, 2010; Zhang et al., 2010).

Abnormally high activity of RyR2 is the most universal finding demonstrated across several models of diabetes. Compounds thought to modulate RyR2 including JTV-519 (K201), carvedilol, dantrolene and tetracaine analogs have previously been tested for therapeutic potential (Kaneko et al., 1997; Wehrens et al., 2004a; Kobayashi et al., 2010; Zhou et al., 2011; Zhang et al., 2015). However, there remains a need for drugs without off-target effects and significant effort is currently being made to identify small novel modulators of the channel that will prevent arrhythmogenic Ca²⁺ leak (Li et al., 2017; Rebbeck et al., 2017).

Stabilization of RyR2-mediated Ca release can be achieved indirectly by targeting CaMKII, given chronic activity in cardiac disease has been linked to RyR2 channel dysfunction (Ai et al., 2005; Uchinoumi et al., 2016; Zhang, 2017). It has been demonstrated that blockade of CaMKII and inhibition of RyR2 phosphorylation in cardiac disease improves intracellular Ca²⁺ homeostasis and attenuates arrhythmogenesis (Ather et al., 2013; Tzimas et al., 2015; Uchinoumi et al., 2016), including in a rat model of type 2 diabetes (Sommese et al., 2016), but this is also not a universal finding (Chakraborty et al., 2014). Alternatively, stabilization could be achieved by a reduction of RyR2 oxidation. In mouse and rat models of type 2 diabetes, it was demonstrated that treatment with ROS scavengers protects against spontaneous Ca²⁺ release events, blunting diastolic dysfunction and arrhythmogenesis in vivo (Shao et al., 2011; Joseph et al., 2016; Sommese et al., 2016; Sánchez et al., 2018). Antioxidants may also reduce ROS-dependent CaMKII activation, hence reduce RyR2 phosphorylation and elevated SR Ca²⁺ leak (Luczak and Anderson, 2014; Sommese et al., 2016; Uchinoumi et al., 2016). While increased oxidative stress and ROS concentrations are a hallmark of multiple cardiac disease states including diabetes, usage of ROS scavengers as a therapeutic strategy is not straightforward because certain levels of intracellular ROS are essential for many physiological processes and the ability to target antioxidants to specific subcellular compartments remains limited (Zima et al., 2014; Dietl and Maack, 2017).

Targeting accessory proteins of RyR2 also offers therapeutic potential. Adenoviral overexpression of sorcin in the hearts of diabetic mice improved contractile function and increased the Ca²⁺ transient amplitude of isolated rat cardiomyocytes (Suarez et al., 2004). In recent work of Liu et al. (2018) it was demonstrated that gene transfer of modified CaM prolonged refractoriness of RyR2-mediated SR Ca²⁺ release, abolishing ventricular arrhythmias observed in a mouse model of CPVT.