Oxidative stress in early metabolic syndrome impairs cardiac RyR2 and SERCA2a activity and modifies the interplay of these proteins during Ca2+ waves

Abstract We investigated how oxidative stress (OS) alters Ca2+ handling in ventricular myocytes in early metabolic syndrome (MetS) in sucrose-fed rats. The effects of N-acetyl cysteine (NAC) or dl-Dithiothreitol (DTT) on systolic Ca2+ transients (SCaTs), diastolic Ca2+ sparks (CaS) and Ca2+ waves (CaW), recorded by confocal techniques, and L-type Ca2+ current (ICa), assessed by whole-cell patch clamp, were evaluated in MetS and Control cells. MetS myocytes exhibited decreased SCaTs and CaS frequency but unaffected CaW propagation. In Control cells, NAC/DTT reduced RyR2/SERCA2a activity blunting SCaTs, CaS frequency and CaW propagation, suggesting that basal ROS optimised Ca2+ signalling by maintaining RyR2/SERCA2a function and that these proteins facilitate CaW propagation. Conversely, NAC/DTT in MetS recovered RyR2/SERCA2a function, improving SCaTs and CaS frequency, but unexpectedly decreasing CaW propagation. We hypothesised that OS decreases RyR2/SERCA2a activity at early MetS, and while decreased SERCA2a favours CaW propagation, diminished RyR2 restrains it.


Introduction
Under a variety of pathological conditions, imbalance between reactive oxygen species (ROS) production and antioxidant cellular systems leads to oxidative stress (OS).ROS modify the structure and function of key Ca 2þ handling proteins, such as the sarcoplasmic reticulum (SR) Ca 2þ release/ ryanodine receptors (RyR2), L-type Ca 2þ channels (LCC) and the sarco/endoplasmic reticulum Ca 2þ ATPase (SERCA2a), via a variety of oxidative posttranslational modifications (OPTMs) (Zima and Blatter 2006, Santos et al. 2011, Perj es et al. 2012, Kholer et al. 2014).ROS can induce a diversity of functional outcomes on RyR2, LCC and SERCA2a, and thus, on cardiac performance, which depend on the cellular redox state, the specific amino-acid residues affected, the type and stoichiometry of OPTMs and the presence of other posttranslational modifications (PTMs), such as phosphorylation (Zima and Blatter 2006, Santos et al. 2011, Perj es et al. 2012, Kholer et al. 2014).In this sense, studies to establish the physiological and pathological relevance of ROS on the excitationcontraction coupling (ECC) have yielded controversial data, since it has been reported that ROS can inhibit, stimulate or have no effect on RyR2, SERCA2a or LCC function (Zima and Blatter 2006, Santos et al. 2011, Perj es et al. 2012, Kholer et al. 2014).For example, recent evidence suggests that physiological production of ROS and nitric oxide-derived reactive nitrogen species (RNS) can enhance the activity of RyR2 and SERCA2a to improve cardiac performance (Zima and Blatter 2006, Tocchetti et al. 2007, Santos et al. 2011, Perj es et al. 2012).Nevertheless, prolonged exposure to increased ROS levels may decrease SERCA2a activity (Zima andBlatter 2006, Perj es et al. 2012) and lead to irreversible RyR2 inhibition (Yan et al. 2008).
Under pathological conditions, excessive and prolonged ROS accumulation induces irreversible OPTMs, which result in a detrimental increase in LCC-mediated Ca 2þ current (I Ca ) (Muralidharan et al. 2017).OS also increases diastolic RyR2 activity (Terentyev et al. 2008, Gonzalez et al. 2010, Sommese et al. 2016, Popescu et al. 2019), and impairs SERCA2a function (Qin et al. 2013), both of which contribute to decreased SR Ca 2þ load, lethal arrhythmias, and contractile dysfunction.Nonetheless, most of these studies have been done in animal models of aging or advanced chronic cardiac diseases, such as type 2 diabetic cardiomyopathy (T2DC) (Sommese et al. 2016, Popescu et al. 2019), ischaemic heart disease (Muralidharan et al. 2017) and heart failure (HF) (Terentyev et al. 2008, Gonzalez et al. 2010); conditions that promote the presence of other PTMs in Ca 2þ handling proteins, such as hyperphosphorylation, which can obscure the effect of ROS on their activity, and thus might not be representative of what happens at early stages of cardiac disease (Niggli et al. 2013).For instance, the OS-mediated RyR2 over-activation in T2DC (Sommese et al. 2016, Popescu et al. 2019) and HF (Terentyev et al. 2008, Gonzalez et al. 2010) animal models, present both hyperphosphorylation and oxidation, suggesting a synergism between these conditions (Niggli et al. 2013).In the context of T2DC however, clinical and experimental evidence demonstrated that OS is established early during pre-diabetic stages, and although OS has been implicated in the development of subclinical diastolic and systolic dysfunction (Grundy et al. 2004, Abel et al. 2008, Hopps et al. 2010, Carvajal et al. 2014), its consequences on cardiac Ca 2þ handling proteins during pre-diabetes have not been completely investigated.Studies in animal models of metabolic syndrome (MetS) induced by high-fat or high-carbohydrate diets showed that pre-diabetic diastolic dysfunction is characterised by prolonged cell relaxation and systolic Ca 2þ transients (SCaTs); alterations attributed to decreased SERCA2a activity, resulting from its oxidation (Dong et al. 2006, Li et al. 2006, Balderas-Villalobos et al. 2013, Carvajal et al. 2014) (Dutta et al. 2001, Li et al. 2006, Vasanji et al. 2006, Fern andez-Miranda et al. 2019).Notably, we and others have reported that at early stages of MetS, these alterations are not linked to a decreased SR Ca 2þ load (Wold et al. 2005, Balderas-Villalobos et al. 2013, Fern andez-Miranda et al. 2019), therefore, smaller SCaTs cannot be due to decreased SR Ca 2þ gradient for release, however, it might be related to LCC or RyR2 dysfunction, or T-tubules disarrangement.Furthermore, we showed that decreased SCaTs amplitude could be increased upon acute exposure to the ROS scavenger N-acetyl cysteine (NAC), which diminished OS in MetS (Balderas-Villalobos et al. 2013).
In order to better understand the underlying mechanism of altered systolic Ca 2þ signalling in isolated ventricular myocytes of the MetS model, we assessed the effects of either NAC or the potent reducing agent of thiol groups, DL-dithio- threitol (DTT) on the oxidation level and function of key Ca 2þ handling proteins, and on the SCaTs properties under field stimulation, as well as on the spontaneous SR Ca 2þ release, in the form of Ca 2þ sparks (CaS) and Ca 2þ waves (CaW).Properties of I Ca were assessed under whole cell patch clamp conditions.These studies would help to have a better understanding of the role of ROS on altered systolic Ca 2þ handling in physiological conditions and at early MetS.

MetS in sucrose-fed rats
All animal procedures were approved by the Scientific Board Committees from the Instituto Nacional de Pediatr ıa (protocol number INP11/048) and performed according to the NIH guidelines.MetS was induced in male Wistar rats (250 ± 10 g) that received 30% sucrose in their drinking water for 16-18 weeks, as previously described (Balderas-Villalobos et al. 2013).Sex and age matched animals receiving tap water were used as control.All animals were housed under controlled laboratory conditions (25 C, 55-60% humidity and 12:12 h light-dark cycle) and allowed ad libitum access to commercial rat chow (Purina Lab Diet 5008).
At the end of the treatment, MetS was assessed by determining the retroperitoneal adipose tissue weight, plasma levels of glucose, triglycerides and insulin, and systemic OS, as described (El Hafidi andBaños 1997, Balderas-Villalobos et al. 2013).Briefly, animals were fasted overnight, anaesthetised (pentobarbital sodium, 60 mg/kg, i.p.), and blood samples were collected from the femoral artery.Triglycerides, insulin and glucose were determined using TR0100 commercial kit (TR0100, Sigma-Aldrich), ELISA kit (ALPCO) and Free-style glucose strip kit (Abbot), respectively.Insulin resistance was assessed by the homeostatic model assessment (HOMA) index.Systemic OS was evaluated by measuring plasma catalase and superoxide dismutase (SOD) activities using commercial kits (Enzo Life Sciences, Inc), and by quantifying malondialdehyde (MDA) equivalents using the TBARS technique (El Hafidi and Baños 1997).

Confocal Ca 21 recordings
SCaTs, CaS and CaW were recorded using confocal microscopy (Balderas-Villalobos et al. 2013).Fluo-4 loaded cells were placed in a recording chamber on top of an inverted microscope equipped with a confocal system (Olympus Fluoview FV1000) and perfused with Tyrode's solution.Fluo-4 was excited at 488 nm using a multi-line argon laser and emission was collected at > 505 nm in the line scan mode.Sampling rate was 20 ms/pixel for SCaTs and CaW and 4 ms/ pixel for CaS.SCaTs were evoked by a train of electrical fieldstimulation pulses (0.3 Hz, 20-25 V, 10 ms) to steady state.After electrical stimulation, SR Ca 2þ load was estimated upon rapid caffeine application (10 mM, 1 s).CaS and spontaneous CaW were recorded in resting ventricular myocytes.Where indicated, CaS were recorded under Ca 2þ overload conditions at 10 mM CaCl 2 .Control and MetS cells were pretreated with NAC (1 mM, 30 min) or DTT (1 mM, 30 min), and their effects on Ca 2þ handling were assessed.While it is expected that NAC and DTT cause a decrease in ROS levels (Yan et al. 2008, Balderas-Villalobos et al. 2013), DTT has also been reported to reverse OPTMs of RyR2 (Terentyev et al. 2008).Nonetheless, we did not find differences in their effects on Ca 2þ signals.Notably, at the concentration and exposure time used, DTT can reverse the effects of oxidation on Ca 2þ handling without toxic side effects (Yan et al. 2008).
Confocal microscopy data are presented as fluorescence intensity normalised to basal fluorescence (F/F 0 ).Properties of SCaTs and CaW were determined using ImageJ and Sigma plot software.The parameters determined for SCaTs were: peak amplitude (DF/F 0 ), maximal SR Ca 2þ release rate (d(F/ F 0 )/dt) and time to 50% decay (DT 50 ).CaW parameters were analysed from three equidistant sites (1 mm width) within a 20 mm segment along the cell length, as described (Salazar-Cant u et al. 2016), and included: peak amplitude (DF/F 0 ), propagation velocity (t wave ), local maximal SR Ca 2þ release rate (dF/dt, normalised to peak amplitude), and rate of decay (j decay ).SR Ca 2þ load was estimated as peak amplitude (DF/ F 0 ) of the caffeine-induced Ca 2þ release signal.CaS analysis was done using the SparkMaster plugin for ImageJ (Picht et al. 2007).

RyR2 oxidation level
Content of free thiol groups in RyR2 was assessed using the monobromobimane (mBB) fluorescence method previously described (Terentyev et al. 2008) in SR microsomes from ventricular tissue.After microsomes incubation with mBB, proteins were precipitated with cold acetone and subjected to SDS-PAGE (6%) in the dark.mBB fluorescence was normalised to total RyR2 determined using Coomassie Blue staining in a gel run in parallel.Fluorescence was measured using a ChemiDoc XRS System and analysed with Quantity One software (Bio-rad).

Confocal imaging of T-tubules
Ventricular myocytes were incubated with the lipophilic membrane marker Di-8-ANEPPS (10 lM in Tyrod es solution) for 20 min.Cells were then perfused with CaCl 2 -free Tyrod es solution, supplemented with 1 mM EGTA, to prevent spontaneous contractions.X-Y images were scanned at 0.5 mm intervals in the Z-axis (confocal Z-stack) and confocal planes at the centre of each cell were selected for analysis.Cellular area and relative T-tubules density were assessed after background subtraction using ImageJ.Transversal organisation index (TTpower) and T-tubules longitudinal spacing were estimated using the TTorg algorithm (Pasqualin et al. 2015).Cellular area and the ratio of heart weight to tibia length (HW/TL) were used to assess hypertrophy.

Statistical analysis
Data are shown as mean ± SEM. "N" represents the number of cells or rats studied.Statistical analyses were performed using unpaired Student's t-tests or two-way ANOVAs followed by Student Newman-Keuls post hoc tests, when appropriate.Differences were considered significant if p < .05.Data analysis was performed using Sigma Stat.

MetS development in sucrose-fed rats
Table 1 shows that sucrose diet caused a significant increase in intra-abdominal fat, plasma levels of triglycerides and insulin, as well as HOMA index, confirming the presence of some components of MetS, including central obesity, hypertriglyceridaemia, hyperinsulinemia and insulin resistance (Carvajal et al. 1999, Baños et al. 2005, Balderas-Villalobos et al. 2013).Furthermore, sucrose diet induced a decrease in plasma SOD activity, augmented catalase activity and increased lipid peroxidation (as measured as MDA), ratifying systemic OS (El Hafidi andBaños 1997, Baños et al. 2005).Nevertheless, sucrose diet did not produce significant changes in body weight or glucose, suggestive of an early stage of MetS (Carvajal andBaños 2002, Balderas-Villalobos et al. 2013).Alterations in the density and organisation of T-tubules linked to cardiac hypertrophy have been reported to contribute to impaired systolic SR Ca 2þ release in HF and T2DC (Stolen et al. 2009, Wei et al. 2010).Therefore, we assessed whether blunted SCaTs in MetS would be caused by defective ECC due to altered T-tubules.Nonetheless, as confocal images with Di-8-ANEPPS show (Supplementary Figure S2(A)), T-tubules in MetS myocytes maintained the characteristic striated pattern, indicative of well-organised distribution along z-lines, as in Control cells.In fact, T-tubules density (Supplementary Figure S2(B)), TT power (Supplementary Figure S2(C)) and T-tubules spacing (1.79 ± 0.01 mm, n ¼ 11 for Control and 1.80 ± 0.01 mm, n ¼ 13 for MetS) were not altered in MetS myocytes.Moreover, HW/TL ratio and the area of ventricular myocytes were unchanged in MetS (Supplementary Figures S2(D,E)).These data ruled out the contribution of hypertrophy and major structural cellular remodelling to impaired SCaTs and confirmed that our model represents an early stage in the development of MetS-related cardiomyopathy (Carvajal andBaños 2002, Fern andez-Miranda et al. 2019).

Effects of acute NAC challenge on Ca 21 handling in control and MetS myocytes
We previously demonstrated increased basal ROS levels in isolated MeTs myocytes, which decreased upon acute exposure to the ROS scavenger NAC (Balderas-Villalobos et al. 2013).OS in MetS could affect systolic SR Ca 2þ release, as it does on SR Ca 2þ reuptake (Balderas-Villalobos et al. 2013).Therefore, we aimed to assess the effects of acute exposure of NAC on SR Ca 2þ release and SCaTs.As Supplementary Table S1 shows, NAC in Control myocytes significantly decreased peak SCaTs amplitude and SR Ca 2þ release rate, and a small, although no significant increase in TD 50 .Conversely, as we previously reported (Balderas-Villalobos et al. 2013), in MetS myocytes NAC induced a partial but significant increase in the amplitude and in the SR Ca 2þ release rate, while TD 50 decreased to values close to those of Control myocytes in basal conditions.These distinct effects of NAC on systolic Ca 2þ handling in both Control and MetS myocytes cannot be attributed to alterations in SR Ca 2þ load, because the amplitude of the caffeine-induced Ca 2þ transients was not significantly affected by NAC in any experimental group.Others and we have previously shown that impaired cytosolic Ca 2þ removal in MetS is associated with OPTMs in SERCA (Dong et al. 2006, Li et al. 2006, Balderas-Villalobos et al. 2013).Since LCC and RyR2 are also affected by ROS (Zima and Blatter 2006, Perj es et al. 2012, Kholer et al. 2014), we investigated their functional status to establish their contribution to depressed SCaTs.

Effects of acute NAC challenge on LCC I ca in control and MetS myocytes
During ECC, I Ca triggers and grades the synchronised activation of thousands of CaS, whose temporal and spatial summation underlie SCaTs (Cheng andLederer 2008, Eisner et al. 2017).To determine the possible contribution of dysfunctional LCC to altered SR Ca 2þ release in MetS myocytes we characterised the biophysical properties of I Ca , under basal conditions and upon NAC exposure.Figure 1(A) shows representative I Ca traces elicited at þ20 mV in Control and MetS myocytes in the absence and presence of NAC pre-treatment.The average I-V curves (Figure 1(B)) and summary peak I Ca density evoked at þ10 mV (Figure 1(C)) show that I Ca density in MetS myocytes was unchanged.In addition, maximal conductance (G max ), voltage dependence of activation (V 1/2 ) and slope factor (k) were not modified in MetS myocytes (Supplementary Table S2).Furthermore, NAC did not affect I Ca density, G max , V 1/2 or k in any group (Figure 1 and Supplementary Table S2).It is thought that ROS affects RyR2 function by OPTMs in the protein, and it has been demonstrated that DTT not only reduced ROS in cardiac cells, but also can reverse some kind of OPTM involved in RyR2 dysregulation in HF (Terentyev et al. 2008).Therefore, for these experiments we pre-treated Control and MetS myocytes with DTT.We found that DTT, significantly reduced CaS frequency in Control, while caused a partial but significant increase in MetS myocytes (Figure 2(B)).Other properties of CaS, such as amplitude and duration were not affected by DTT in any group (Figure 2(C,D)).Furthermore, the effects of DTT on RyR2 activity in Control and MetS myocytes were not associated with changes in SR Ca 2þ load (Figure 2(E)).Importantly, we repeated the experiments using NAC and obtained similar results (data not shown).

Effects of acute
It is well known that CaS frequency steeply depends on SR Ca 2þ load (Gy€ orke et al. 2002, Cheng andLederer 2008), therefore, decreased CaS frequency in MetS myocytes, despite normal SR Ca 2þ load, may indicate that RyR2 sensitivity to luminal Ca 2þ is diminished.Thus, we assessed CaS frequency under Ca 2þ overload conditions, induced by high extracellular Ca 2þ (10 mM; Figure 3(A)).As expected, we found a significant increase in CaS frequency, in both Control and MetS myocytes (Figure 3(B) and Supplementary Table S3).Nonetheless, compared with Ca 2þ -overloaded Control cells, CaS frequency was still significantly smaller in Ca 2þ -overloaded MetS myocytes (Figure 3(B) and Supplementary Table S3), suggesting that responsiveness of RyR2 to luminal Ca 2þ is decreased in MetS.Similarly, Ca 2þ overload resulted in a decrease in amplitude and duration of CaS in both groups, compared with basal conditions (Supplementary Table S3).Nevertheless, the amplitude of these events was still significantly smaller in MetS compared with Control (Figure 3(C)).Decreased amplitude of CaS under Ca 2þ overload conditions, was associated with a decrease in SR Ca 2þ load in both groups (Supplementary Table S3), most likely caused by the large increase in CaS frequency ($3 and $4 fold increase for Control and MetS, respectively; p < .05),and also in CaW frequency ($9 and $3.5 fold increase, for Control and MetS, respectively; p < .05)(Supplementary Table S3).DTT pre-treatment of Ca 2þ -overloaded Control myocytes decreased CaS frequency by $30%, although the change was not significant (Figure 3 MetS myocytes.Furthermore, SR Ca 2þ content in Ca 2þ -overloaded cells was not affected by DTT in both experimental groups (Figure 3(E)).These data suggest that diastolic RyR2 sensitivity to luminal Ca 2þ in MetS might be decreased, and that it can increase upon DTT exposure.
Several studies have shown that ROS can regulate the sensitivity of RyR2 to cytosolic and luminal Ca 2þ by inducing OPTMs on specific cysteine residues (Zima and Blatter 2006, Zissimopoulos and Lai 2006, Terentyev et al. 2008, Donoso et al. 2011).Accordingly, we evaluated the oxidation extent of RyR2 in ventricular myocardium by labelling free reactive thiols with mBB and how it changed upon NAC exposure.Labelled RyR2 signal was significantly decreased in MetS compared to Control hearts (Figure 4(A)), without changes in RyR2 protein expression (Figure 4(B)), indicating augmentation of oxidised reactive cysteines.Moreover, NAC exposure significantly decreased oxidation of RyR2 in MetS, suggesting that decreased RyR2 spontaneous activity and responsiveness and blunted SCaTs could be associated with the increased oxidated state of the channel.Conversely, NAC exposure did not modify RyR2 free reactive thiols signal in Control hearts (Figure 4(A)).S3).

OS and cardiac dysfunction in sucrose-fed rats with early MetS
As previously reported, high-sucrose diet in rats for 16-18 weeks produces an early stage of MetS associated with the development of central obesity, hypertriglyceridaemia, hyperinsulinemia and insulin resistance (Table 1) (Carvajal et al. 1999, Baños et al. 2005), but where hyperglycaemia (  abnormalities in SCaTs (Supplementary Table S1) could be linked to OS-induced changes in Ca 2þ handling proteins.As previously reported, and confirmed in this report, hearts from sucrose-fed rats present systemic OS, revealed by enhanced lipid peroxidation and decreased SOD activity (Table 1) (Baños et al. 2005, Balderas-Villalobos et al. 2013).Moreover, we previously reported that the ROS scavenger NAC reverts excessive ROS accumulation in isolated cardiac myocytes from sucrose-fed rats (Balderas-Villalobos et al.

2013
).Here, we confirmed that either NAC increased SCaTs amplitude, SR Ca 2þ release rate and reduced TD 50 (Supplementary Table S1), supporting the notion that in MetS myocytes OS-mediated OPTMs in Ca 2þ handling proteins affect SR Ca 2þ reuptake and systolic and diastolic SR Ca 2þ release.ROS, concentration, duration of exposure and type of OPTMs), as well as other PTMs that might occur in parallel, such as phosphorylation (Zima and Blatter 2006, Perj es et al. 2012, Niggli et al. 2013, Kholer et al. 2014, Fern andez-Miranda et al. 2019).For instance, the effects of ROS on LCC are controversial, with studies showing a decrease (Lacampagne et al. 1995), an increase (Muralidharan et al. 2017) or not effects (Greensmith et al. 2010) on I Ca .Furthermore, evidence suggests that S-glutathionylation increases, while S-nitrosylation decreases LCC function (Gonzalez et al. 2009, Muralidharan et al. 2017).Here, we found that NAC did not alter LCC function in Control nor MetS myocytes, since I Ca density voltage dependence were unchanged (Figure 1 and Supplementary Table S2).
Therefore, we conclude that neither physiological ROS nor OS, at least at this stage of MetS development, affect I Ca .Therefore, blunted systolic Ca 2þ -induced Ca 2þ release (CICR) must be related to RyR2 dysfunction, although other alterations such as action potential remodelling, as described in other pathologies (Harris et al. 2005), deserve further investigation.
Regarding the RyR2 function, we observed that NAC exposure had differential effects on systolic SR Ca 2þ release depending on the physiological or pathological context.In Control myocytes, NAC diminished SCaT amplitude and whole cell SR Ca 2þ release rate (Supplementary Table S1), suggesting that basal ROS and RyR2 OPTMs have a stimulatory effect on systolic SR Ca 2þ release, probably by optimising RyR2 responsiveness to unchanged I Ca .These data agree with published evidence showing that the positive regulation of ROS on RyR2 acts as a physiological mechanism of finetuning contractile cardiac function (Zima and Blatter 2006, Yan et al. 2008, Prosser et al. 2013, Kholer et al. 2014), and it has been shown that nitric oxide-derived RNS improve cardiac contractility by stimulating RyR2 and increasing SCaT amplitude (Tocchetti et al. 2007).
On the other hand, in MetS, exposure to NAC caused partial increase in SCaT amplitude and SR Ca 2þ release rate (Supplementary Table S1), suggesting that OS has an inhibitory effect on systolic SR Ca 2þ release.Moreover, since these effects were not due to changes in SR Ca 2þ load (Supplementary Table S1), LCC function (Figure 1) nor in Ttubules (Supplementary Figure S2 ), blunted SR Ca 2þ release could be associated with diminished RyR2 sensitivity to either luminal SR Ca 2þ or trigger I Ca , resulting from increased oxidated state (see below).
Given the results suggesting that alteration in RyR2 due to OS could explain the alteration on SCaTs, we used DTT to evaluate the diastolic activity of this channel, alternatively to NAC, since DTT not only decreases ROS, but also reverts OPTM in the channel (Terentyev et al. 2008, Yan, et al. 2008), and we obtained similar results with either agent.Decreased RyR2 sensitivity to luminal Ca 2þ in MetS is supported by our experimental results in CaS properties.Indeed, increased oxidation of RyR2 in MetS (Figure 4(A)) concurred with a reduced CaS frequency under basal conditions (Figure 2(B)) and lower increase in response to SR Ca 2þ overload (Figure 3(B)), which was largely reverted upon DTT exposure (Figures 2(B) and 3(B)).These effects were accompanied by a decrease on RyR2 oxidation level induced by NAC (Figure 4(A)).The mechanisms by which OS decreases RyR2 sensitivity to I Ca and luminal Ca 2þ in MetS might be associated with OPTMs that can alter the RyR2 subunits interaction with SR regulatory proteins, such as triadin and calsequestrin (Zima and Blatter 2006) or cytosolic regulators, such as FKBP12.6, as it occurs when other PTM are present (Shao et al. 2007).Conversely, in a basal redox state of RyR2 in Control myocytes, although no further reduction of the channel was found (Figure 4 Altogether, our results support the notion that ROS levels might have a differential effect on systolic and diastolic SR Ca 2þ release by enhancing RyR2 activity in physiological conditions but inhibiting it in the pathological settings of early MetS.This dual ROS-dependent effect might be supported by previous reports under a variety of conditions.Studies in lipid bilayers and isolated ventricular myocytes have shown that ROS can either, increase or decrease RyR2 open probability and CaS frequency, in a concentration and time dependent fashion by affecting their sensitivity to Ca 2þ (Zima and Blatter 2006, Zissimopoulos and Lai 2006, Yan et al. 2008, Kholer et al. 2014).
Nevertheless, the inhibitory effect of OS on RyR2 in early MetS, as described here diverges from studies in advanced MetS and T2D showing over-activation of RyR2, attributed not only to increased oxidation state, but also to hyper-phosphorylation mediated by OS-dependent activation of CaMKII (Roe et al. 2013, Sommese et al. 2016, Popescu et al. 2019).These divergent results might reflect different levels of CaMKII activation during the progression of cardiac disease.Indeed, increased function of CaMKII caused cardiac hypertrophy (Maier 2012), which occurs in parallel with OS-mediated RyR2 hyperphosphorylation, in models of early and late diabetes (Sommese et al. 2016, Popescu et al. 2019).Nonetheless, cardiac hypertrophy was absent in our model of early MetS (Supplementary Figure S2), and we previously reported that CaMKII-mediated phosphorylation of phospholamban is unchanged (Balderas-Villalobos et al. 2013).Moreover, a recent report in the same model of early MetS, depressed SCaT amplitude was associated with decreased RyR2 activity due to reduced CaMKII-dependent RyR2 phosphorylation (Fern andez-Miranda et al. 2019).Thus, it appears that at early MetS, OS might decrease RyR2 activity by increasing their oxidation state, instead of leading them hyperactive by promoting their phosphorylation in late stages.Although we did not address the specific OPTMs that results in RyR2 stimulation or inhibition, an issue that requires further research, evidence suggest that S-nitrosylation might exert opposite functional effects (Gonzalez et al. 2009, 2010, Cutler et al. 2012) and modify CaMKII-mediated phosphorylation (Cutler et al. 2012).On the other hand, CaS in MetS myocytes have smaller amplitude, nonetheless, the lack of effect of DTT on this parameter in MetS and Control (Figures 2(C) and 3(C)) suggests that this alteration is not produced by ROS, in agreement with reports showing that ROS do not affect RyR2 conductance (Boraso and Williams 1994).
Steady state SR Ca 2þ load is maintained by the balance between SERCA2a Ca 2þ re-uptake and SR Ca 2þ leak (Gy€ orke et al. 2002).Since there are not apparent alterations in SR Ca 2þ load in Control and MetS, and NAC/DTT did not change it in both conditions (Figure 2(E), Supplementary Table S1) despite significant effects on RyR2 activity (Figure 2(B)), it seems that, at the low frequency of stimulation of these experiments, SERCA2a is able maintain SR Ca 2þ content close to normal levels.By measuring j decay during CaW, we estimated SERCA2a function (Bassani et al 1994, Ginsburg 1998), considering that this pump handles $92% of cytosolic Ca 2þ removal during the decline of the ventricular cell Ca 2þ transient in rat, while NCX is only responsible for $7% (Bassani et al 1994).Our estimates of j decay in MetS myocytes suggested that SERCA2a function decreased by $30%, a similar reduction to that reported in the same rat model measuring the rate of Ca 2þ uptake in SR-enriched microsomes, which directly reflects SERCA2a activity (Fern andez-Miranda et al. 2019).Slow SERCA2a function would underlie diastolic contractile dysfunction and might even favour CaW propagation (see below).We previously reported that impaired SERCA activity in MetS myocytes recovered upon NAC exposure, and here we showed that similar results could be obtained with DTT (Figure 5(F)).Furthermore, in Control myocytes, DTT slowed SERCA2a (decreased j decay ; Figure 5(F)), suggesting that, similar to RyR2, ROS evoke a differential effect on SERCA2a function, depending on the cell overall condition.In Control myocytes ROS sustain efficient SR Ca 2þ pump activity, while it is decreased under OS in MetS.Indeed, evidence suggests that low concentrations of ROS or RNS enhance SERCA2a activity, while OS leads to its inhibition (Zima and Blatter 2006, Lancel et al. 2009, Perj es et al. 2012).

Effects of DTT on CaW propagation in control and MetS myocytes
CaW propagation in cardiac myocytes has been explained by the "fire-diffuse-fire" mechanisms, which states that propagation occurs when Ca 2þ released by an spontaneously activated RyR2 cluster diffuses within the cytosol to trigger activation of neighbouring inactive RyR2 clusters via CICR (Keizer et al. 1998).More recently, the possibility that SERCA2a might affect CaW propagation has gained attention, since local cytosolic Ca 2þ removal by SERCA2a reuptake might either decrease cytosolic Ca 2þ available for diffusion (the trigger) to neighbouring sites, and thereby decreasing CICR efficacy and t wave , or locally increase luminal SR Ca 2þ ahead of the wave front, sensitising RyR2 to cytosolic Ca 2þ and enhancing t wave (Lukyanenko et al. 1999, Keller et al. 2007, Venetucci et al. 2008, Sobie and Lederer 2012).Nevertheless, experimental evidence is conflicting.Taking together all our data in both cell types, where whole cell SR Ca 2þ was constant but changes in both SERCA2a and RyR2 were induced by DTT, we propose that local SERCA2a contribution to CaW dynamics would depend on cellular conditions.Under physiological conditions (Control myocytes) with normal ROS levels, the combined optimal SERCA2a and RyR2 functions and steady state SR Ca 2þ load may facilitate CaW propagation and local SR Ca 2þ release rate.Though, in MetS myocytes, where ROS production are elevated, notwithstanding decreased function of both RyR2 and SERCA2a (Figures 2(B) and 5(F)), t wave and SR Ca 2þ release rate were similar to Control myocytes (Figure 5(D,E)).Our interpretation of these results is that a new balance between the two altered Ca 2þ handling proteins arises to keep t wave unchanged, where lower RyR2 sensitivity might act as a restrainer of CaW propagation, while slower SERCA2a might facilitate cytosolic Ca 2þ diffusion to act as trigger for CaW propagation.This premise is supported by results from DTT experiments, since the effects of these reagents on SERCA2a and RyR2 in both cell types are different, but t wave and SR Ca 2þ release rate are decreased to a similar extent.Indeed, DTT enhanced activity of SERCA2a and RyR2 in MetS myocytes (Figure 5(F)), which resulted in a reduction of t wave and local SR Ca 2þ release rate (Figure 5(D,E)), suggesting that enhanced local cytosolic Ca 2þ buffering by SERCA2a overcomes the partial increase in RyR2 sensitivity.On the other hand, in Control myocytes, DTT exposure caused a decrease in both SERCA2a and RyR2 activities (Figures 2(B) and 5(F)), which yielded slower t wave and SR Ca 2þ release rate (Figure 5(D,E)), suggesting that the decrease in RyR2 sensitivity outweighs the increase in locally available cytosolic Ca 2þ .
Thus, it seems that, although OS-mediated decrease in SERCA2a activity at early MetS might be pro-arrhythmogenic, decreased diastolic RyR2 sensitivity to luminal Ca 2þ might restrain further CaW acceleration of propagation, preventing arrhythmias.Although studying arrhythmogenesis was not the main objective in this work, and more research is required to fully understand the molecular details, our observations might have important clinical implications given the pivotal role of CaW in triggering lethal arrhythmias in diabetic cardiomyopathy and HF, where over-activation of RyR2 and decreased SERCA2a activity are present (Terentyev et al. 2008, Sommese et al. 2016, Popescu et al. 2019).In this sense, it would be relevant to establish the effect of b-adrenergic which would help to exacerbate the cellular dysfunction due to altered RyR2 and SERCA2a.

Conclusions
This work supports the pivotal role of ROS in regulating normal cardiac function, by maintaining redox state of Ca 2þ handling proteins and thus cardiac performance, however when unbalanced, dysfunctional ventricular myocyte performance arises.At early MetS, increased oxidation of RyR2 and SERCA2a affects their activity during ECC, spontaneous CaS and their interplay during CaW, contributing to subclinical diastolic and systolic cardiac dysfunction.These proteins are control spots for opportune treatment in early Mets to prevent further cardiac diseases.
NAC/DTT challenge on diastolic Ca 21 sparks in control and MetS myocytes Diastolic CaS are thought to reflect the spontaneous open probability of RyR2 and thus, are generally used as tool to study resting RyR2 function in isolated myocytes (Cheng and Lederer 2008), and this becomes dysregulated under a variety of pathological conditions, including OS.To determine whether abnormal activity of RyR2 might contribute to the defective SCaTs in MetS (Fern andez-Miranda et al. 2019), we evaluated the frequency and spatiotemporal properties of diastolic CaS in resting myocytes.Figure 2 shows that CaS in MetS myocytes displayed a significant reduction in frequency (by $72%, Figure 2(B)) and amplitude (by $25%, Figure 2(C)), while duration (Figure 2(D)) and width (not shown) remained unchanged, when compared with Control.Decreased amplitude and CaS frequency were not associated with changes in SR Ca 2þ load (Figure 2(E)), suggesting defective RyR2 function in MetS.

Figure 1 .
Figure 1.Effects of NAC on electrophysiological properties of I Ca in Control and MetS myocytes.A. Representative I Ca traces elicited at þ20 mV in Control and MetS myocytes under basal conditions and after treatment with NAC.B. Current-voltage curves for I Ca density in Control and MetS myocytes under basal conditions or after treatment with NAC. C. Peak I Ca density evoked at þ10 mV in Control and MetS myocytes, under basal conditions or after treatment with NAC.Data represent the mean ± SEM, N ¼ 14-17 cells from 3-4 animals.

Figure 2 .
Figure 2. Effects of DTT on frequency and properties of diastolic CaS in Control and MetS myocytes.A. Representative confocal line-scan images and corresponding time-dependent profiles of CaS recorded in Control and MetS myocytes under basal conditions.Arrow heads indicate the regions where CaS profiles were obtained.Pooled data for B. Frequency, C. Amplitude, and D. Duration of CaS, from Control and MetS myocytes under basal conditions or after treatment with DTT.E. SR Ca 2þ load determined under basal conditions or after treatment with DTT in Control and MetS myocytes.Data represent the mean ± SEM.N ¼ 49-60 cells for CaS frequency, 22-170 events for CaS properties and 37-46 cells for SR Ca 2þ load, from 3-5 animals.Ã p < .05,two-way ANOVA followed by Student Newman-Keuls post hoc test.

Figure 3 .
Figure 3. Effects of DTT on frequency and properties of CaS in Ca 2þ -overloaded Control and MetS myocytes.Ca 2þ overload was induced by the incubation of the cells with Tyrod es solution containing 10 mM CaCl 2 .A. Representative confocal line-scan images, and corresponding time-dependent profiles, of CaS recorded in Control and MetS myocytes.Arrow heads indicate the regions where CaS profiles were obtained.Pooled data for B. Frequency, C. Amplitude, and D. Duration of CaS from Control and MetS myocytes in the absence and after treatment with DTT.E. SR Ca 2þ load determined in the absence and after treatment with DTT in Ca 2þ -overloaded Control and MetS myocytes.Data represent the mean ± SEM.N ¼ 19-26 cells for CaS frequency, 34-107 events for CaS properties or 16-22 cells for SR Ca 2þ load, from 3-4 animals.Ã p < .05,two-way ANOVA followed by Student Newman-Keuls post hoc test.

Figure 4 .
Figure 4. Oxidation level and RyR2 expression in Control and MetS myocytes.A. Representative gel of mBB RyR2 labelling (top) and densitometric analysis (bottom) obtained in Control and MetS ventricular homogenates under basal conditions and after treatment with NAC.B. Representative immunoblot of total RyR2 protein (top) and densitometric analysis (bottom) in ventricular homogenates obtained from Control and MetS hearts.Data represent the mean ± SEM, N ¼ 4 hearts.Ã p < .05,two-way ANOVA followed by Student Newman-Keuls post hoc test or unpaired Student's t-tests.CT ¼ control.

Figure 5 .
Figure 5. Effects of DTT on properties of CaW in Control and MetS myocytes.A. Representative confocal line-scan images and corresponding time-dependent profiles of CaW recorded in Control and MetS myocytes in basal conditions.Arrow heads indicate the regions where CaW profiles were obtained.Pooled data for B. Frequency, C. Amplitude, D. t wave , E. dF/dt and F. j decay of CaW in Control and MetS myocytes, in the absence and after treatment with DTT.Data represent the mean ± SEM, N ¼ 8-17 cells from 3-5 animals.Ã p < .05,two-way ANOVA followed by Student Newman-Keuls post hoc test.
(A)), DTT decreased diastolic RyR2 spontaneous activity (Figure 2(B)) and the cell response to Ca 2þ overload (Figure 3(B)), suggesting that physiological ROS might be modulating Ca 2þ sensitivity by indirect mechanisms that have to be established.

Table 1 .
Characteristics of MetS rats.
Representative confocal line-scan images and corresponding time-dependent profiles of CaS recorded in Control and MetS myocytes under basal conditions.Arrow heads indicate the regions where CaS profiles were obtained.Pooled data for B. Frequency, C. Amplitude, and D. Duration of CaS, from Control and MetS myocytes under basal conditions or after treatment with DTT.E. SR Ca 2þ load determined under basal conditions or after treatment with DTT in Control and MetS myocytes.Data represent the mean ± SEM.N ¼ 49-60 cells for CaS frequency, 22-170 events for CaS properties and 37-46 cells for SR Ca 2þ load, from 3-5 animals.Ã p < .05,two-way ANOVA followed by Student Newman-Keuls post hoc test.
(Venetucci et al. 2008, Swietach et al. 2010 triggered by self-regenerative spontaneous CaW, which propagate within the myocyte by sequential recruitment of quiescent neighbouring RyR2 clusters, and their incidence and propagation depend, among other factors, on RyR2 and SERCA2a function(Venetucci et al. 2008, Swietach et al. 2010, Sobie and Lederer 2012, Salazar-Cant u et al. 2016).During our recordings of diastolic Ca 2þ signalling, at physiological extracellular