PLCδ1 Protein Rescues Ischemia-reperfused Heart by the Regulation of Calcium Homeostasis

Abstract

Myocardial Ca²⁺ overload induced by ischemia/reperfusion (I/R) is a major element of myocardial dysfunction in heart failure. Phospholipase C (PLC) plays important roles in the regulation of the phosphoinositol pathway and Ca²⁺ homeostasis in various types of cells. Here, we investigated the protective role of PLCδ1 against myocardial I/R injury through the regulation of Ca²⁺ homeostasis. To investigate its role, PLCδ1 was fused to Hph1, a cell-permeable protein transduction domain (PTD), and treated into rat neonatal cardiomyocytes and rat hearts under respective hypoxia-reoxygenation (H/R) and ischemia-reperfusion conditions. Treatment with Hph1-PLCδ1 significantly inhibited intracellular Ca²⁺ overload, reactive oxygen species generation, mitochondrial permeability transition pore opening, and mitochondrial membrane potential elevation in H/R neonatal cardiomyocytes, resulting in the inhibition of apoptosis. Intravenous injections of Hph1-PLCδ1 in rats with I/R-injured myocardium caused significant reductions in infarct size and apoptosis and also improved systolic and diastolic cardiac functioning. Furthermore, a small ions profile obtained using time-of-flight secondary ion mass spectrometry showed that treatment with Hph1-PLCδ1 leads to significant recovery of calcium-related ions toward normal levels in I/R-injured myocardium. These results suggest that Hph1-PLCδ1 may manifest as a promising cardioprotective drug due to its inhibition of the mitochondrial apoptotic pathway in cells suffering from I/R injury.

Introduction

The formation of reactive oxygen species (ROS) and the development of Ca²⁺ overload following ischemia/reperfusion (I/R) injuries lead to apoptosis in cardiomyocytes.¹ Ca²⁺ overload in reperfusion is a key factor in inducing the opening of mitochondrial permeability transition pores (mPTP), principally by ROS production, which sequentially causes the activation of numerous cytosolic proteins, phospholipases, protein kinases, proteases, and endonucleases.² Calcium-activating proteases destroy the proteins that regulate intracellular calcium levels and thereby decrease calcium regulatory activity.³ It is well established that the opening of mPTPs can induce irreversible reperfusion injury, with its process eventually translating to hypertrophy, heart failure, and myocardial infarction.⁴ Therefore, calcium regulation, which is considered to be a major initiation factor for cardiovascular symptoms, is often managed using ion channel blockers—particularly, calcium channel blockers, which have been used widely in the treatment of heart disease. However, these drugs have many severe side effects such as nausea, headache, edema, low blood pressure, dizziness, and high mortality rates over extended periods of use.^5,6

Phospholipase C (PLC) plays an important role in the Ca²⁺ regulation in many cells including cardiomyocytes.⁷ PLC activation hydrolyzes its substrate, phosphatidylinositol 4,5-biphosphate, into two second messengers, inositol 1,4,5-triphosphate [I(1,4,5)P₃] and 1,2-diacyglycerol,⁸ and regulate calcium by activating the IP3 receptor and protein kinase C (PKC). PLCδ1 is the most abundant isoform of PLC expressed in normal rat hearts⁹ and responds to Ca²⁺ increase and costimulation by Gα_h (known as transglutaminase II).¹⁰ The several negatively charged residues within the catalytic domain of the isoform make it highly sensitive to Ca²⁺.¹¹ In addition, PLCδ1 is found in both the cellular and mitochondrial membranes of liver cells and is associated with Ca²⁺ regulation through the mPTP.¹² When calcium-mediated proteases are activated by increasing calcium levels, PLCδ1 can be selectively degraded⁹ among other PLC isoforms, and thus, its regulatory function for both calcium and mPTP may be lost. Therefore, PLCδ1 may be a new therapeutic target in the prognosis for ischemic heart disease.

mPTP is a voltage-dependent channel complex composed of a voltage-dependent anion channel on the outer membrane, an adenine nucleotide translocator (ANT) on the inner membrane, cyclophilin D (CypD), and other molecules.¹³ In an experiment with transgenic mice lacking CypD, there was a marked reduction in the rate of apoptosis after exposure to I/R.¹⁴ According to recent studies, CypD is regulated by ERK/GSK-3, and CypD phosphorylation can increase the possibility of mPTP opening.¹⁵

By taking advantage of the specificity of proteins and the ability of small molecules to regulate intracellular signaling, we can develop new drugs that are safer and more efficient. In this study, we develop a new therapeutic protein drug that can directly penetrate the cell membrane with quick delivery, high efficiency, and minimal side effects. Protein transduction domains (PTDs) are small protein domains that are essential for viral replication. Fusion proteins, composed of PTDs, are powerful tools for the delivery of therapeutic proteins into eukaryotes.¹⁶ PTD fusion proteins have many important advantages including high transduction efficiency, rapid cellular uptake, and low toxicity compared to previously used gene delivery methods in primary nondividing cells.

In the present study, we found that PLCδ1 plays an important role in the regulation of calcium and mPTP of cardiomyocytes under H/R or I/R conditions. Therefore, we fused PLCδ1 with PTDs to represent a promising new therapeutic strategy and to investigate PLCδ1 as a possible target for cardioprotective therapy against myocardial I/R injury through the regulation of Ca²⁺ homeostasis.

Results

Protective roles of PLCδ1 against ROS-induced Ca²⁺ overload in hypoxia/reoxygenated neonatal cardiomyocytes

To investigate whether PLCδ1 may offer a cardioprotective function in hearts suffering ischemic injury, we fused PLCδ1 with an Hph-1 PTD (Supplementary Figure S1), which was then isolated (Supplementary Figure S2) and treated into cardiomyocytes. PLCδ1 was first detected at 5 minutes and remained for more than 12 hours (Supplementary Figure S3). We also monitored ROS production which was significantly increased in H/R (Supplementary Figure S4). We next investigated H/R-induced changes in intracellular Ca²⁺ levels and effects of PLCδ1 on Ca²⁺ overload. Hph1-PLCδ1 was added concurrently with reoxygenation treatment (Figure 1a). PLCδ1 maintained intracellular Ca²⁺ levels similar to that of the H/R control. Because Ca²⁺ overload causes mitochondrial ROS generation, we presupposed that PLCδ1 can regulate intracellular ROS production through regulation of Ca²⁺ levels. Appropriately, PLCδ1-treated neonatal cardiomyocytes showed a significant reduction in ROS generation (Figure 1b). As cells hire several ion regulatory proteins to handle Ca²⁺ homeostasis, we investigated whether PLCδ1 controls Ca²⁺ channels and Ca²⁺ pumps related to Ca²⁺ overload in H/R neonatal cardiomyocytes (Figure 1c). The expression levels of the Na⁺-Ca²⁺ exchanger 1 (NCX1) and ryanodine receptor 2 (RyR2) Ca²⁺ release channel, known to be major regulators of intracellular calcium levels in cardiomyocytes,¹⁷ as well as other important Ca²⁺ regulators were investigated. H/R clearly increased expressions of both NCX1 and RyR2, but decreased expressions of type 3 inositol-1,4,5-triphosphate receptor (IP3R) channel, sarco-endoplasmic Ca²⁺ reticulum ATPase, voltage-dependent L-type Ca²⁺ channel, and plasmalemmal Ca²⁺ ATPase. Thereof, PLCδ1 showed to significantly recover expression levels of these channels under H/R.

Figure 1.

Hph1-PLCδ1 protects cardiomyocytes from apoptosis induced by H/R through the regulation of Ca²⁺ overload. (a) The expression levels of PLCδ1 and intracellular Ca²⁺ overload in H/R cardiomyocytes with or without treatment with Hph1-PLCδ1. PLCδ1 was detected by western blotting. Fluorescent images were obtained using fluo-4 AM, and the fluorescent intensities were quantified. (b) Reactive oxygen species (ROS) production in H/R cardiomyocytes. ROS was assessed by DCF. (c) The effects of PLCδ1 on the altered expression of calcium channels in H/R cardiomyocytes. Expression levels of RyR2, IP3R-3, sarco-endoplasmic Ca²⁺ reticulum ATPase, NCX1, L-type Ca²⁺ channel, and plasmalemmal Ca²⁺ ATPase were estimated by reverse transcription polymerase chain reaction in H/R cardiomyocytes with or without 0.1 µmol/l Hph1-PLCδ1. P value was calculated by one-way analysis of variance followed by a Bonferroni test: *P < 0.05 versus H/R cardiomyocytes, **P < 0.001. All data are presented as the means ± SEM from at least three independent experiments. CON, control; DCF, 2′,7′-dichlorofluorescein.

Inhibition of apoptosis through blocking of mPTP

To investigate the correlation between PLCδ1 and apoptosis through the mPTP in neonatal cardiomyocytes, we first examined H₂O₂-induced mPTP opening (Supplementary Figure S5). PLCδ1 caused a significant reduction in mPTP opening in H₂O₂-treated neonatal cardiomyocytes (Figure 2a). Because changes in mitochondrial function resulting from the opening of mPTPs and the following loss of mitochondrial membrane potential can initiate apoptosis, we observed changes in mitochondrial membrane potential. Normal neonatal cardiomyocytes showed red-orange (FL1) mitochondrial staining under JC-1, indicating normal high membrane potentials. In contrast, H/R-induced neonatal cardiomyocytes showed an increased rate of green fluorescence (FL2), indicating a loss of mitochondrial membrane potential. Pretreatment with PLCδ1 significantly inhibited the loss of red fluorescence in a dose-dependent manner, indicating the prevention of ROS-induced mPTP opening (Figure 2b). To examine whether PLCδ1 on the mitochondrial membrane inhibits apoptosis through the suppression of mPTP opening in H/R cardiomyocytes, apoptotic proteins targeting mitochondria were investigated. PLCδ1 significantly inhibited cytochrome c release from mitochondria into the cytosolic fraction, restored the Bax/Bcl-2 protein ratio to control levels (Figure 2c), and downregulated caspase-3 activity (Figure 2d).

Figure 2.

Antiapoptotic effects of Hph1-PLCδ1 in H/R through the regulation of mitochondrial permeability transition pores (mPTP) opening. (a) The effects of PLCδ1 on opening of the mPTP. P value was calculated by one-way analysis of variance (ANOVA) followed by a Bonferroni test: *P < 0.05. (b) Flow cytometric analysis for mitochondrial membrane potential in H/R cardiomyocytes with or without Hph1-PLCδ1. (c) Bcl-2/Bax ratio and cytochrome c release. P value was calculated by one-way ANOVA followed by a Bonferroni test: *P < 0.05 versus CON. CON, control. (d) Caspase-3 activity in H/R cardiomyocytes with or without Hph1-PLCδ1. P value was calculated by one-way ANOVA followed by a Bonferroni test: *P < 0.05 versus CON. CON, control. All data are presented as the means ± SEM from at least three independent experiments.

A regulation mechanism of PLCδ1 for the mPTP opening in hypoxia/reoxygenated neonatal cardiomyocytes

As PLCδ1 regulated the opening of mPTPs, we suggested that PLCδ1 might regulate CypD. Indeed, we showed a high possibility that PLCδ1 may regulate mPTP opening through regulation of the PKC-MEK-ERK-GSK3β-CypD signaling pathway. CypD phosphorylation was not investigated in this experiment, but based on published literatures, we assumed that activated GSK3β led to CypD phosphorylation resulting in the opening of mPTPs.¹⁶ Protein kinase B (AKT) is also considered an important regulator of GSK3β in mitochondria. We thus investigated whether AKT can be phosphorylated in H/R. Our following results indicate that AKT was phosphorylated under H/R stimulation but did not show any significant changes under PLCδ1 treatment. However, reduced ERK activation still caused GSK3β to be activated to open mPTP (Figure 3). This result may suggest that ERK inhibition has a stronger influence than AKT activation on GSK3β regulation under H/R, and that PLCδ1 can maintain the phosphorylation level of GSK3β (inactive form) through upregulation of PKC-MEK-ERK.

Figure 3.

Mechanisms involved in PLCδ1-mediated mitochondrial permeability transition pores (mPTP) opening. (a) Western blots on whole cell lysates of H/R cardiomyocytes with or without Hph1-PLCδ1. AKT, protein kinase B; CypD, cyclophilin D; ERK, extracellular signal-regulated kinase; GSK, glycogen synthesis kinase; MEK, mitogen-activated protein kinase; PI3K, phosphatidylinositol-3-kinase; PKC, protein kinase C; PLCδ1, phospholipase C-δ1. P value was calculated by one-way ANOVA followed by a Bonferroni test: *P < 0.05, **P < 0.001; ns, nonsignificant. All data are presented as the means ± SEM from at least three independent experiments. (b) A proposed model of CypD regulation by PLCδ1 in normal condition and H/R.

Cardiac protective role of PLCδ1 in ischemia/reperfused heart

To test whether PLCδ1 could be delivered into reperfused regions, we isolated hearts 3 hours after injection and investigated the corresponding PLCδ1 levels in the reperfused areas. PLCδ1 expression which was reduced in I/R hearts showed greater levels in PLCδ1-treated I/R hearts, and a similar tendency in the phosphorylation levels of PKC was observed (Figure 4a). To investigate the therapeutic effects of PLCδ1, we evaluated histological analyses 2 weeks after injection. Thinner left walls of I/R-induced hearts were protected by PLCδ1 injection (Supplementary Figure S6). Area at risk was similar in both groups, whereas infarct size was reduced in the PLCδ1-treated I/R group (9.5 ± 1.5%) compared to the I/R group (15.5 ± 2%) (Figure 4b). Extensive interstitial fibrosis was induced in I/R hearts (25 ± 3%) compared to the controls (1.2 ± 0.5%); meanwhile, 36 and 108 nmol/l treatment of PLCδ1 significantly decreased interstitial fibrosis to 9 ± 3.5% and 7.5 ± 4%, respectively (Figure 4c). To further examine the protective role of PLCδ1, selected proteins including connexin-43 (cn-43), n-cadherin, and α-smooth muscle actin were examined. Cn-43 and n-cadherin are very important for maintaining the cardiomyocyte function of cell-to-cell interactions.¹⁸ Dephosphorylated cn-43 was detected in I/R and I/R+Hph1, while phosphorylated cn-43 was examined in I/R+ PLCδ1 as well as the control (Supplementary Figure S7a). N-cadherin, which was disorganized in I/R-induced heart, exhibited more organized staining patterns around the border of the infarct area in I/R-induced heart treated with PLCδ1 (Supplementary Figure S7b). α-Smooth muscle actin–positive vessels in the border area were also significantly increased in the PLCδ1-treated I/R-induced hearts (Supplementary Figure S7c). The terminal deoxynucleotidyl transferase–mediated dUTP nick-end labeling (TUNEL) assay showed that I/R-induced apoptosis was attenuated in myocardial tissue by PLCδ1 treatment (Figure 4d).

Figure 4.

Cardioprotective effects of Hph1-PLCδ1 in ischemia/reperfusion (I/R)-injured rat myocardium. (a) Western blots showing the levels of PLCδ1 and activated protein kinase C in infarcted myocardium after intravenous injection of 750 ng/kg (108 nmol/l) Hph1-PLCδ1 fusion protein. (b) Representative heart sections after Evans blue perfusion and 2, 3, 5-triphenyltetrazolium chloride staining, and (right) staining data expressed as the percent of area at risk and infarction site (MI) sizes to left ventricle (LV) (left). P value was calculated by one-way analysis of variance (ANOVA) followed by a Bonferroni test: *P < 0.05 versus MI. (c) Masson's trichrome images in infarcted myocardium. P value was calculated by one-way ANOVA followed by a Bonferroni test: **P < 0.001 versus I/R+saline. (d) Terminal deoxynucleotidyl transferase–mediated dUTP nick-end labeling assay showing the amount of apoptotic death in infarcted myocardium. P value was calculated by one-way ANOVA followed by a Bonferroni test: **P < 0.001 versus I/R+saline. (e) Representative left ventricular pressure–volume loops in infarcted myocardium. End-systolic pressure–volume relationships during preload reduction are indicated by dashed lines and heart rates were estimated accordingly. 36 nmol/l Hph1-PLCδ1 fusion protein was injected in the infarcted myocardium. P value was calculated by one-way ANOVA followed by a Bonferroni test: *P < 0.05, **P < 0.001.

Heart function studies

We estimated heart function using echocardiographic analysis. Echocardiography results are shown in Supplementary Figure S8, and left ventricular (LV) function and remodeling indices are summarized in Table 1. Left ventricular ejection fraction and fractional shortening were significantly improved in the I/R+PLCδ1 group compared to the I/R group. Radial strain (S_rad) and circumferential strain (S_cir) in the I/R group showed much lower values than in the control group, but both were improved by PLCδ1 treatment in the global and infarction regions. We further estimated pressure–volume because of the limitations in echocardiography, in which the injection of PLCδ1 resulted in a better catheterization-determined ejection fraction (data not shown), a less steep end-systolic pressure–volume relationship slope, and a significantly reduced heart rate, suggesting that PLCδ1 treatment protected cardiac function from I/R injury (Figure 4e). We also tested for behavioral abnormalities and the cytotoxic effect. Behavioral tests were performed in male and female mice, and no significant differences were observed between the sexes under treatment with Hph1-PLCδ1 (data not shown). Human ether-a-go-go-related gene (hERG) potassium channels are important for regulating membrane potential.¹⁹ However, several drugs such as terfenadine and bertosamil have been recently removed from the market because of their severe cardiac toxicity originating from the blocking of hERG potassium channels.²⁰ So, we tested the toxicity from the blocking of hERG potassium channels using a patch clamp technique. PLCδ1-treated samples showed over 10 μmol/l of IC50 for the inhibition of hERG potassium channels, which means that PLCδ1 did not affect the inhibition of hERG potassium channels (Supplementary Figure S9).

Table 1. Echo-data for untreated control, I/R, and I/R-PLCδ1-treated rats.

Role of PLCδ1 in metabolic stabilization related to Ca²⁺ regulation in ischemia/reperfused heart

To examine the cardioprotective effects of Hph1-PLCδ1 on infarcted myocardium at the systemic level, we performed a global small ion profiling of three myocardium groups using time-of-flight secondary ion mass spectrometry (TOF-SIMS) analysis and applied a partial least square-discriminant analysis. Upon our hypothesis that PLCδ1 can regulate Ca²⁺ homeostasis through mPTP and prevent damage from I/R injury, we further speculated the possibility of PLCδ1 affecting the population of small ions which influences heart function. One hundred and sixty ions that differed in abundance among the three groups were analyzed (Figure 5a and Supplementary Figure S10).²¹ We then performed a clustering analysis in which the ions among the three groups were clustered into four types of differential expressions: (i) the ion abundances in C1 did not change in I/R but were increased by PLCδ1 treatment; (ii) ion abundances in C2 were increased by I/R but were rescued by PLCδ1 treatment; (iii) ion abundances in C3 were decreased by I/R but were not altered by PLCδ1 treatment; and (iv) ion abundances in C4 were decreased by I/R but were then significantly rescued by PLCδ1 treatment. These results suggest that the ions seen in C2 and C4 might be associated with the cardioprotective effects of PLCδ1. Interestingly, ion abundances in C4 were significantly anticorrelated (P < 0.05) with the intracellular Ca²⁺ levels (Figure 5b, left). Among the ions analyzed in C2 and C4, several ions (Ca²⁺, Na⁺, H₂O₂, nitric oxide, and histamine) were found to be closely associated with the PLCδ1 pathway in I/R (Figure 5b, right), supporting our hypothesis that PLCδ1 treatment induced a functional loss of Ca²⁺ regulation in I/R resulting in H₂O₂ production and cellular damage.

Figure 5.

Global small ion profiling using TOF-SIMS analysis. (a) Clustering analysis of the selected ions. The results show the four types of differential expression among the three sample groups. The colors represent the increase (red) and decrease (blue) in ion abundances, compared to the mean levels of the individual ions. (b) Distributions of correlation coefficients of the ion abundances with the intracellular Ca²⁺ levels in C2 and C4 (left). The representative P values in C2 and C4 were computed for the median correlation coefficients in both clusters. A hypothetical model generated by integrating molecular data and TOF-SIMS data (right). The colors represent the ions belonging to C2 (blue) and C4 (orange) in I/R. All data are the means ± SEM from six independent experiments. ER, endoplasmic reticulum.

Discussion

Our results indicate that increases in ROS production and Ca²⁺ levels in H/R or I/R lead to mitochondrial membrane transition, which results in the release of intermembrane proteins. In turn, the activation of Ca²⁺-activated proteases leads to the degradation of PLCδ1, which causes PLCδ1 to lose its regulatory function for the PKC-MEK-ERK-GSK3β signaling pathway to control mPTP opening through CypD. PKC also becomes unable to regulate calcium channels and pumps such as L-type Ca²⁺ channel and plasmalemmal Ca²⁺ ATPase.²² Therefore, myocardial apoptosis is initiated, and heart function is impaired. However, PLCδ1 fusion protein leads to a significant increase of cell survival through the regulation of mPTP opening in H/R and inhibits apoptosis of cardiomyocytes in I/R (Figure 6).

Figure 6.

A hypothetical model in which PLCδ1 regulates intracellular Ca²⁺ in normal and ischemia/reperfusion conditions. In the normal adult heart (right), PLCδ1 increases intracellular Ca²⁺ levels through activation of IP₃R, leading to contraction. Then, cytosolic Ca²⁺ is quickly removed mainly through SERCA to the ER/SR or through NCX to the extracellular space for relaxation. At the same time, PLCδ1 inhibits mitochondrial permeability transition pores (mPTP) opening. However, under pathophysiological conditions such as early reperfusion and heart failure, NCX allows the Ca²⁺ influx through the reverse mode into the cytosol and mitochondria⁴⁹; subsequently, Ca²⁺-activated proteases degrade PLCδ1 resulting in the loss of inhibition of mPTP opening. mPTP opening is stimulated further by accumulated matrix Ca²⁺ and induces caspase activation and apoptotic cell death.⁵⁰ However, transduced PLCδ1 may have a cardioprotective effect through inhibition of mPTP opening. ER/SR, endoplasmic reticulum/sarcoplasmic reticulum.

In this study, we showed that PLCδ1 prevents intracellular Ca²⁺ overload in I/R and may be closely related to suppression of cardiomyocyte apoptosis in H/R or I/R. Correspondingly, a previous report indicated that PLCδ1 is decreased in ischemia.⁹ Our study showed that reduced levels of PLCδ1 expression and phosphorylation of PKC are restored by PLCδ1 injection, which means that injected PLCδ1 targets the injured heart region and recovers biological activity. As PKC, a downstream target of PLCδ1, is a serine/threonine kinase, it has been implicated in a number of diseases, including ischemic heart disease and congestive heart failure.²³ The activation of PKC is reported to be associated with cardiac protection in I/R.²⁴ In PLCδ1 KO mice, PKCs and other downstream signals of PLCδ1 are remarkably decreased.²⁵ These results are consistent with the enhanced cardiac survival following PKC activation induced by PLCδ1 overexpression in this study. Another study showed the contribution of PLCδ1 in cardiac function, in which selective inhibition of PLCδ1 was shown to cause severe cardiac dysfunction under adriamycin-treated mouse, and suggested a new upstream regulator for PLCδ1.²⁶ This result correlates with our data because adriamycin is known to induce cardiac apoptosis due to oxidative stress and calcium overload by mitochondrial dysfunction.²⁷ Therefore, previous studies support our hypothesis that PLCδ1 has a therapeutic potential for ischemic heart disease.

To transduce PLCδ1 into neonatal cardiomyocytes and adult hearts, we harnessed them to PTDs as a therapeutic vehicle. Our data indicated that Hph-1 conjugated with PCLδ1 did not exacerbate ROS-induced cell death. Furthermore, although it was not specifically investigated in cardiomyocytes, Lee et al.¹⁶ reported that Hph-1 did not affect cell viability, and moreover, present any cytotoxic effects or behavioral abnormalities in vitro and in vivo, even with around a 1,000-fold higher concentration (10 mmol/l) than the effective concentration. Based on these, we speculate that the cytotoxicity of Hph-1 itself would be minimal if there is any. However, a Hph-1 control was not used to verify our speculation, and instead, saline was used as a vehicle for most of our in vivo experiments as commonly practiced.²⁸ Nevertheless, the toxicity produced by the PTD alone was not directly investigated here and it remains to be one of the limitations of the present study.

In the normal adult heart, RyR and NCX have been known to play critical roles in the increase of intracellular Ca²⁺ levels under H/R or I/R, in which activities of RyR and reverse mode of NCX are increased.²⁹ Aleksey et al. reviewed that oxidizing reagents increased RyR and NCX activity but decreased L-type Ca²⁺ channel, SERCA, and plasmalemmal Ca²⁺ ATPase activity, which is consistent with our results.³⁰ As we did not investigate the levels of protein expression or function of Ca²⁺-related proteins (Figure 1c) in this study, these data are not enough to prove that PLCδ1 directly alters Ca²⁺-related protein function as proposed. An in-depth explanation of the roles of PLCδ1 cannot be discerned in this study because neonatal cardiomyocytes have a different predominant Ca²⁺ release system, in which RyRs in the adult mammalian heart and IP3Rs in the early postnatal heart act dominantly.³¹ Although this is the first attempt to investigate the effect of PLCδ1 concerning calcium channels, further study is necessary. Nevertheless, there are several studies to support the relation between calcium channel expression and activity. Decreased Ca²⁺ uptake and contractile functions were observed under diminished SERCA expression by adenoviral SERCA gene delivery method,³² and enhanced NCX functional activity was seen in increased gene expression from hypertension.³³ Therefore, our data suggest the possibility that PLCδ1 can modulate intracellular Ca²⁺ through the regulation of pumps and channels in the sarcoplasmic reticulum, mitochondria, or cytosolic membrane.

It is important to demonstrate a regulatory mechanism of PLCδ1 in the opening of mPTPs. Since mPTP opening during reperfusion has been known to cause cell death, molecular mechanisms and regulators for mPTP have been published for therapeutic applications.³⁴ p53 can interact with CypD, which acts as a trigger for mPTP opening in stroke pathology,³⁵ and heat shock protein 60 can protect against CypD-dependent cell death in tumor cells.³⁶ However, molecular mechanisms related to PLCδ1 are unknown. In addition to PLCδ1, AKT is also known as an important regulator for CypD by regulating GSK3β. Our results showed that AKT is activated in H/R but failed to show a protective effect in neonatal cardiomyocytes (Figure 3a). Several studies have showed that only additional activation of AKT, not just by I/R but by ischemic postconditioning or constitutive activation, provides a cardiac-protective effect.^37,38 At the same time, ERK phosphorylation was decreased in the H/R of our experiments. This phenomenon is evident in the research on rat kidneys following I/R.³⁹ Therefore, we inferred the role of AKT based on our results. Although AKT is important for cell survival against H/R and I/R injury, the increased phosphorylation levels of AKT under H/R does not have a significant effect on cell survival. We also investigated the relationship between AKT and PLCδ1. PLCδ1 was not able to significantly activate AKT within H/R. This result is also supported by the report, in which PKC (a downstream target of PLCδ1) inhibitor cannot modulate AKT activation.⁴⁰ Therefore, Figures 2 and 3 suggest the high possibility that PLCδ1 may regulate mPTP opening through the GSK3β-CypD signaling pathway.

With echocardiography, we might confirm that the indices related to heart function were restored by PLCδ1 injection in I/R hearts. Especially, S_cir and S_rad showed great improvements both globally and in the infarction region. By using 2D echocardiography, Zoran et al.⁴¹ showed that S_cir and S_rad are associated with the segmental LV dysfunction following myocardial infarction and are also highly related to scar size. These results are consistent with our data on average S_cir and S_rad, and infarct size. In addition to echocardiography, we estimated pressure–volume to further measure heart function because some limitations of echocardiography have been reported in many patient cases after thoracic surgery or with pulmonary obstructive disease.⁴² These results using a Millar catheter showed improved heart functions as in the echocardiography of rats with PLCδ1 injection.

Partial least square-discriminant analysis showed that some ion abundances have significantly anticorrelated relationships with intracellular Ca²⁺ levels, which might indicate the cardioprotective effects of PLCδ1 (Figure 5). There are no direct references between these ions and PLCδ1, but some papers show the relationships between PKC, a downstream regulator of PLCδ1, and histamine or nitric oxide. It is known that the relaxing action of histamine is dependent on the calcium concentration in the ventricular muscle fiber,⁴³ and Belevych et al.⁴⁴ showed that PKC can regulate the histamine receptor that is coupled with the β1-adrenergic receptor in the heart. However, there are some conflicting researches about the effects of histamine in cardiac protection. It is suggested that cardiac histamine induced by doxorubicin may cause cardiotoxicity.⁴⁵ In our results, we suggested that nitric oxide increased by PLCδ1 might be related to its protective role in I/R. Although there are controversial arguments, our results are supported by several researches that demonstrate a protective role of nitric oxide in ischemia or ischemia-reperfused conditions.⁴⁶

Taken as a whole, our studies showed the benefits of acute treatment using PTD and suggested the possibility of PLCδ1 as a strong therapeutic candidate for I/R or I/R-derived heart disease. Thereof, Hph1-PLCδ1 may prove to be a promising cardioprotective drug with its ability to inhibit the mitochondrial apoptotic pathway in cells suffering from I/R injury.

Materials and Methods

Purification of Hph1-PLCδ1 proteins. The expression plasmid pHph1-PLCδ1 was transformed into BL21-DE3 cells (ATCC No. 53863) using the heat shock transformation method. The transformed bacteria were grown at OD₆₀₀ 0.7 in lysogeny broth medium. Protein expression was induced with 1 mmol/l isopropyl β-D-1-thiogalactopyranoside (Life Technologies, Grand Island, NY) for 4 hours at 37 °C. The cells were harvested by centrifugation at 6,000 rpm for 20 minutes, resuspension of the pellet in binding buffer (50 NaH₂PO₄, 300 NaCl, 10 imidazole, mmol/l, pH 8.0), and sonication of the bacteria using intervals of 6 seconds on/off for a total time of 8 minutes. After removal of the cell debris by centrifugation, 0.5 ml of 50% Ni²⁺-NTA agarose bead (Qiagen, Valencia, CA) was added to the clarified cell extract. Binding on the agarose beads was performed at 4 °C. The extract was loaded on to a poly-Prep chromatography column (0.8 × 4, Bio-Rad Laboratories, Hercules, CA). The column was then washed with wash buffer (20 Tris–HCl, 500 NaCl, 20 imidazole, mmol/l, pH 7.9), eluted with 1 ml each of elution buffers 1 (50 NaH₂PO₄, 300 NaCl, 250 imidazole, mmol/l, pH 8.0) and 2 (50 NaH₂PO₄, 300 NaCl, 500 imidazole, mmol/l, pH 8.0), then loaded onto a PD-10 desalting column (GE Healthcare Biosciences, Pittsburgh, PA).

Isolation of neonatal rat cardiomyocytes. The hearts of 1–2-day-old Sprague-Dawley rat pups were dissected and washed with pH 7.4 Dulbecco's phosphate-buffered saline (D-PBS), which was free of Ca²⁺ and Mg²⁺. The hearts were then minced to about 1 mm³ and treated with 5 ml of collagenase II (0.5 mg/ml, 262 units/mg, Life Technologies) for 5 minutes at 37 °C. After the supernatant was removed and combined with fresh collagenase II solution for an additional 5 minutes, the cells in the supernatant were transferred to a new tube and suspended with α-MEM containing 10% fetal bovine serum. The tubes were centrifuged at 1,200 rpm for 3 minutes at room temperature, and the cell pellet was resuspended in 5 ml of cell culture medium. This digestion process was repeated seven to nine times until very little tissue remained. Cell suspensions were collected and incubated in 100-mm tissue culture dishes for 2 hours to preattach the fibroblasts. Nonadherent cells were collected and seeded at a final concentration of 1 × 10⁶ cells/ml in 0.1 μmol/l BrdU (Sigma-Aldrich, St Louis, MO). Finally, cells were incubated in 5% CO₂ at 37 °C.

Measurement of mitochondrial membrane potential. To measure mitochondrial membrane potential (ψ_m), cardiomyocytes were exposed to hypoxia and reoxygenation for 12 hours and 1 hour, respectively. Cells were then trypsinized, washed with PBS, and incubated with 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-benzamidazolocarbocyanine iodide (JC-1) (Immunochemistry Technologies, Bloomington, MN) at 37 °C for 10 minutes in the dark. Flow cytometric analysis was performed using a FACS Calibur system (Becton Dickinson, Franklin Lakes, NJ) using CellQuest software with 10,000 events recorded for each sample. Data were acquired in a single parameter histogram with appropriate particle size and light scatter gating.

Measurement of mPTP opening in mitochondria. mPTP opening was measured by the Petronilli method with minor modifications.⁴⁷ 1 × 10⁵ cells were cultured for 1 day in Lumitrac 600 on a 96-well plate (Greiner Bio-One, Frickenhausen, Germany) coated with 1.5% gelatin. After attachment, Hph1-PLCδ1 was added to the cells for 30 minutes before calcein loading. Cells were washed twice with D-PBS, and 1 µmol/l calcein was then loaded for 10 minutes. To quench the cytosolic and nuclear calcein fluorescence, 1 mmol/l CoCl₂ was loaded for 15 minutes. Cells were washed with D-PBS and then incubated for 30 minutes with 10 µmol/l cyclosporine A as an mPTP opening inhibitor. Following treatment with 500 µmol/l H₂O₂, calcein fluorescence was measured using a Perkin-Elmer LS-5 fluorescence spectrophotometer (Santa Clara, CA). When calcein was released into the cytosol through the mPTP opening stimulated by H₂O₂, cytosolic calcein was quenched by CoCl₂ and lower fluorescence intensity was detected.

Hypoxia/reoxygenation in neonatal cardiomyocytes and treatment with Hph1-PLCδ1. Neonatal cardiomyocytes were incubated in 5% CO₂ at 37 °C. Medium was exchanged with deoxygenated α-MEM without fetal bovine serum in an anaerobic chamber (Thermo Scientific, Waltham, MA). After 12 hours of incubation, reoxygenation was carried out using 10% α-MEM maintained in 5% CO₂ at 37 °C for 1 hour. Hph1-PLCδ1 was added at the time of reoxygenation.

Measurement of intracellular reactive oxygen species generation. Neonatal cardiomyocytes were labeled with 10 μmol/l 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA; Life Technologies). The H₂DCFDA probe entered into the cells, after which the acetate group on H₂DCFDA was cleaved by cellular esterases and caused the nonfluorescent 2′,7′-dichlorofluorescin to be trapped inside. Subsequent oxidation by reactive oxygen species yielded the fluorescent product DCF (2′,7′-dichlorofluorescein). The dye, when exposed to an excitation wavelength of 480 nm, emits a light of 535 nm only after oxidization. Labeled cells were examined using a luminescence spectrophotometer.

Cytosolic calcium measurement. Measurement of the concentration of free cytosolic Ca²⁺ was performed by confocal microscopic analysis (Carl Zeiss, Stuttgart, Germany). Neonatal cardiomyocytes were plated on 4-well slide chambers coated with 1.5% gelatin for 1 day in α-MEM containing 10% fetal bovine serum and 0.1 µmol/l bromodeoxyuridine. After incubation, the cells were washed with modified Tyrode's solution containing 0.265 g/l CaCl₂, 0.214 g/l MgCl₂, 0.2 g/l KCl, 8.0 g/l NaCl, 1 g/l glucose, 0.05 g/l NaH₂PO₄, and 1.0 g/l NaHCO₃ at a pH of 7.4. Cells were then loaded with 10 µmol/l acetoxymethyl ester of fluo-4 (Fluo-4 AM; Life Technologies) for 20 minutes at 37 °C in the dark. Fluorescence images were collected using a confocal microscope under excitation by a 488-nm line argon laser and emission by a 510–560-nm band-pass filter. Relative data on intracellular Ca²⁺ were determined by measuring the fluorescent intensity.

Cell fractionation and western blotting. To quantify cytochrome c release, analyses of mitochondrial and cytosolic protein fractionations were performed. After harvesting 5 × 10⁶ neonatal cardiomyocytes, the cell pellets were resuspended in 100 µl of buffer (20 mmol/l hydroxyethyl piperazineethanesulfonic acid, pH 7.5, 1.5 mmol/l MgCl₂, 10 mmol/l KCl, 1 mmol/l ethylenediaminetetraacetic acid, 1 mmol/l ethylene glycol tetraacetic acid, 250 mmol/l sucrose, 0.1 mmol/l phenylmethylsulfonyl fluoride, 1 mmol/l dithiothreitol, 4 μg/ml pepstatin, 4 μg/ml leupeptin, 5 μg/ml aprotinin). After 10 minutes of incubation on ice, cells were centrifuged at 750 ×g for 10 minutes at 4 °C, and the supernatant was further centrifuged at 10,000 ×g for 30 minutes at 4 °C. The mitochondrial pellets were resuspended in buffer, and the supernatant (cytosolic protein) was saved to perform a western blot and caspase-3 assay. The protein concentration of each fraction was determined by bicinchoninic acid (Thermo Scientific). Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis using 12–15% polyacrylamide gels and then electrotransferred to methanol-treated polyvinylidene difluoride membranes. The blotted membranes were washed twice with water and blocked by incubation with 10% nonfat dried milk in PBS buffer (8.0 g NaCl, 0.2 g KCl, 1.5 g NaH₂PO₄, 0.2 g K₂HPO₄ per liter). The membranes were probed with PLCδ1 (Santa Cruz Biotechnology, Santa Cruz, CA), phosphor-PKC (Cell Signaling Technology, Danvers, MA), PKC (Santa Cruz Biotechnology), phosphor-AKT (Cell Signaling Technology), AKT (Cell Signaling Technology), phosphor-ERK1/2 (Cell Signaling Technology), ERK1/2 (Cell Signaling Technology), phosphor-GSK-3β (Cell Signaling Technology), GSK-3β (Cell Signaling Technology), phosphor-MEK1/2 (Cell Signaling Technology), MEK1/2 (Cell Signaling Technology), cytochrome c (Santa Cruz Biotechnology), Bcl-2 (Santa Cruz Biotechnology), Bax (Enzo Life Sciences, Farmingdale, NY) and β-actin (Abcam, Cambridge, UK), then with goat anti-mouse and goat anti-rabbit IgG-peroxidases. Blots were detected using enhanced chemiluminescence kits (GE Healthcare Biosciences).

Measurement of caspase-3 activity. Relative caspase-3 activity was determined using an ApopTarget Capase-3 Colorimetric Protease Assay used according to the manufacturer's instructions (Life Technologies). This assay is based on the generation of free DEVD-pNA chromophores when the provided substrate is cleaved by caspase-3. Upon cleavage of the substrate by caspase-3, free pNA light absorbance can be quantified using a microplate reader at 405 nm. Briefly, 2 × 10⁶ neonatal cardiomyocytes were harvested in lysis buffer within 1 mol/l dithiothreitol after different treatments, and the cell extracts were centrifuged to eliminate cellular debris. Aliquots (50 µl) of the cell extracts were incubated at 37 °C for 2 hours in the presence of the chromophore substrate. Free DEVD-pNA was determined colorimetrically. The comparison of the absorbance of pNA from the apoptotic sample with an uninduced control allowed the determination of the amount of increase in capase-3 activity.

Reverse transcription polymerase chain reaction analysis. The expression levels of mRNA were analyzed using the reverse transcription polymerase chain reaction. Total RNA was prepared using an Ultraspect-II RNA system (Biotecx Laboratories, Houston, TX), and single-stranded cDNA was synthesized from isolated total RNA by avian myeloblastosis virus reverse transcriptase. A 20-µl reverse transcription reaction mixture containing 1 µg of total RNA, 1× reverse transcription buffer (10 mmol/l Tris–HCl, pH 9.0, 50 mmol/l KCl, 0.1% Triton X-100), 1 mmol/l deoxynucleoside triphosphates, 0.5 units of RNase inhibitor, 0.5 µg of oligo(dT)₁₅, and 15 units of avian myeloblastosis virus reverse transcriptase was incubated at 42 °C for 15 minutes, heated at 99 °C for 5 minutes, and then incubated at 0–5 °C for 5 minutes. The primer sequences were as follows: ryanodine receptor 2 (RyR2), sense: 5′-CCAACATGCCAGACCCTACT-3′ and anti-sense: 5′-TTTCTCCATCCTCTCCCTCA-3′; Na⁺-Ca²⁺ exchanger 1 (NCX), sense: 5′-TGTCTGCG ATTGCTTGTCTC-3′ and anti-sense: 5′-TCACTCATCTCCACCAGACG-3′; IP3R-3, sense: 5′-GAGAAGCTGTGCGTGAAG-3′ and anti-sense: 5′-CGGCCCATCCTGCGCTTC-3′; sarco-endoplasmic Ca²⁺reticulum ATPase (Serca-2A), sense: 5′-TCCATCTGCCTGTCCAT-3′ and anti-sense: 5′-GCGGTTACTCCAGTATTG-3′; plasmalemmal Ca²⁺ ATPase, sense: 5′-TGCCTTGTTGGGATTTCTCT-3′ and anti-sense: 5′-CACTCTGGTTCTGGCTCTCC-3′; L-type Ca²⁺ channel, sense: 5′-TGTC ACGGTTGGGTAGTGAA-3′ and anti-sense: 5′-TTGAGGTGGAAGGGACTTTG-3′. GAPDH was used as the internal standard, and the signal intensity of the amplification product was normalized to its respective glyceraldehyde 3-phosphate dehydrogenase signal intensity.

Myocardial I/R and injection of Hph1-PLCδ1. Experiments were conducted in accordance with the International Guide for the Care and Use of Laboratory Animals. The experimental protocol was approved by the Animal Research Committee of the Yonsei University College of Medicine. The myocardial infarction animal model was created according to the method developed by Lipsic et al.⁴⁸ with minor modifications. Under general anesthesia, 8-week-old Sprague-Dawley male rats (230 ± 10 g) were ventilated with positive-pressure (180 ml/minute) using a Harvard ventilator (Harvard Apparatus, Holliston, MA). The rat hearts were exposed through a 2 cm incision in the left lateral costal rib. The proximal portion of the left coronary artery beneath the left atrium was ligated with a 6-0 silk suture for 1 hour. After occlusion, the ligature was removed for reperfusion. Successful reperfusion was indicated by the restoration of redness. The skin was sutured and the thorax was closed under negative pressure. Hph1-PLCδ1 was injected intravenously and concurrently at the time of reperfusion.

Determination of myocardial infarct size and area at risk. Two weeks after the reperfusion, the left coronary artery, which was the site of previous occlusion, was ligated again and hearts were perfused with 250 µl of Evans blue solution (1%) over 5 minutes to measure area at risk and infarct size. Then, hearts were exposed through the costal ribs and excised. The hearts were sectioned transversely into 2-mm slices from the apex. Slices were incubated in 20 ml of 1% 2, 3, 5-triphenyltetrazolium chloride solution for 30 minutes at 37 °C. The heart slices were fixed in 10% formalin solution for 24 hours at 4 °C. Digital images were obtained using a microscopy (MICROS computerized Real Time Digital Microscope, Direct industry, Austria). Area at risk is shown in bright red color, and infarct size (MI) is expressed by a 2, 3, 5-triphenyltetrazolium chloride- and Evans blue-unstained area. After fixation, a paraffin block was also prepared, and a 2-µm slide was prepared using hematoxylin–eosin staining and Masson's Trichrome staining. Total infarct size and digital images were measured with MetaMorph software version 4.6 (Molecular Devices, LLC., Sunnyvale, CA) for the control (n = 6), MI+saline (n = 6), and MI+Hph1-PLCδ1 (n = 6) groups and was expressed as a percentage of the total left ventricle.

Immunohistochemistry. Histological analysis was performed according to the instructions of the manufacturer (R.T.U VECTASTAIN Universal Quick kit; Vector Laboratories, Peterborough, UK). In brief, the excised heart tissues were fixed in 3.7% buffered formaldehyde and embedded in paraffin. Tissue sections, 5 mm thick, were deparaffinized, rehydrated, and rinsed with PBS. Sodium citrate antigen retrieval was performed with 10 mmol/l sodium citrate (pH 6.0) in a microwave for 10 minutes. Sections were incubated in 3% H₂O₂ in order to quench endogenous peroxidase. The tissue sections were blocked in 2.5% normal horse serum and then incubated with primary monoclonal antibody (total connexin 43, 1:100 dilution, Cell signaling Technology; Phosphor-connexin 43, 1:100 dilution, Zymed Laboratories, South San Francisco, CA; Dephosphor-connexin 43, 1:100 dilution, Zymed Laboratories; N-cadherin, 1:100 dilution, Santa Cruz Biotechnology; α-smooth muscle actin, 1:100 dilution, Abcam). The samples were then incubated in fluorescein isothiocyanate–conjugated goat anti-mouse IgG (Jackson Immuno Research Lab, West Grove, PA) or Texas red-conjugated goat anti-rabbit IgG (Jackson Immuno Research Lab) secondary antibody, and followed by dehydration with 100% N-butanol, ethanol, and xylene. For the evaluation of α-smooth muscle actin reaction, six slices per group were prepared, and 10 view fields from each sample slice were randomly chosen by which the area of the positive reaction in relation to the whole view field was calculated.

TUNEL assay. The TUNEL assay was performed according to the manufacturer's instruction (EMD Millipore, Billerica, MA). In brief, the excised heart tissues were fixed in 3.7% buffered formaldehyde and embedded in paraffin. Five-micrometer-thick tissue sections were deparaffinized, rehydrated, and rinsed with PBS. A positive control sample was prepared from a normal heart section by treatment with DNase I (10 U/ml, 10 minutes at room temperature). The sections were pretreated with 3.0% H₂O₂, subjected to reaction with TdT enzyme at 37 °C for 1 hour, and incubated in a digoxigenin-conjugated nucleotide substrate at 37 °C for 30 minutes. Nuclei exhibiting DNA fragmentation were stained with 3,3-diamino benzidine (Vector Laboratories, Burlingame, CA) for 5 minutes. The nuclei of the apoptotic cardiomyocytes were stained dark brown. The sections were counterstained with methyl green and then cover slipped. The sections were observed by light microscopy. Six slices per group were prepared, and 10 different regions were observed in each slice (×400).

Rat echocardiography. Rats were sedated with zoletil (50 mg/kg) and xylazine (5 mg/kg) by intraperitoneal injection. Imaging was performed at 15 MHz with a linear transducer interfaced with an ultrasound system (Vivid 7, GE Vingmed Ultrasound, GE Healthcare). Two-dimensional guided M-mode and two-dimensional echocardiographic studies were performed at the mid-papillary muscle level, and all data were recorded and subsequently analyzed at the end of the study. For each animal, LV end systolic diameter and LV end diastolic diameter dimensions were measured from the M-mode tracings, and the LV fractional shortening, ejection fraction, end-systolic circumferential strain (S_circ), and radial strain (S_rad) were calculated.

TOF-SIMS analysis. Rat myocardium samples were cut to a thickness of 12 µm at −20 °C using a cryostat (Leica Microsystems, Buffalo Grove, IL). The tissue sections were deposited onto a silicon wafer and then directly analyzed using a TOF-SIMS V instrument (ION-TOF, Münster, Germany) equipped with a bismuth liquid metal ion gun. Bi₃⁺ primary ions at 25 kV and a repetition rate of 5 kHz in the low-current bunched mode were used to obtain positive and negative spectra. An analysis area of 250 × 250 µm was randomly rastered by the primary ions with a spatial resolution of 1 µm. Glass-slide samples were charge compensated using low-energy electron flooding. The primary ion dose density was maintained below 10¹² ions/cm² to ensure static SIMS conditions. Mass resolution was higher than 7,000 at m/z < 500 in both positive and negative modes. Positive and negative ion spectra were internally calibrated using H⁺, H₂⁺, CH₃⁺, C₂H₃⁺, C₃H₅⁺, and C₇H₇⁺ peaks, and H⁻, C⁻, CH⁻, C₂H⁻, and C₄H⁻ peaks, respectively. The ion peaks were identified in each TOF-SIMS dataset using Ionspec software (version 4.1.0.1). The abundance of each ion peak was then estimated using the same software. In each group, the ion abundance data from six myocardium samples obtained from different rats were analyzed to measure the abundance of a total of 980 ions (469 ions from positive mode and 511 from negative mode), and those that had variable importance in projection from partial least square-discriminant analysis larger than 1 were selected for further analysis.

LV catheterization. For invasive hemodynamics, a LV catheterization was performed 3 weeks after I/R. After anesthetization of the specimen, a Millar Mikro-tip 2F pressure transducer (model SPR-838; Millar Instruments, Houston, TX) was introduced into the left ventricle via the right carotid artery. Real-time pressure–volume loops were recorded by a blinded investigator, and all data were analyzed off-line with PVAN 3.5 software (Millar Instruments, Houston, TX). Data are represented as a graph of end-systolic elastance (end-systolic pressure volume relation).

Statistical analysis. Data are expressed as means ± SEM. Statistical analyses of two groups were carried out using Student's t-test. Analysis involving more than two groups was done using one-way analysis of variance and Bonferroni test (SigmaPlot; Systat Software, San Jose, CA). P < 0.05 was considered significant.

SUPPLEMENTARY MATERIAL Figure S1. Structure of the Hph1-PLCδ1 conjugated fusion proteins. Figure S2. Identification of pure recombinant PLCδ1 by HPLC and Coomassie Brilliant Blue staining. Figure S3. 0.1 µmol/L Hph1-PLCδ1 was incubated and collected during 24 hours. Figure S4. Reactive oxygen species (ROS) production in H/R cardiomyocytes. Figure S5. mPTP opening induced by H2O2 in a concentration-dependent manner. Figure S6. Left ventricular wall thickening estimated by hematoxylin and eosin staining images from histological sections (magnification: x1.6). Figure S7. Representative confocal images of (a) connexin-43, (b) n-cadherin, and (c) a-SMA from transverse sections of I/R rat ventricles (magnification: x400). Figure S8. Two-dimensional-echocardiography and two-dimensional-speckle-tracking imaging of left ventricle. Figure S9. hERGK+ Channel assay by patch clamp. Figure S10. ToF-SIMS analysis of metabolites in infarcted myocardium.