The effect of Sirt1 deficiency on Ca2+ and Na+ regulation in mouse ventricular myocytes

Abstract This study addressed the hypothesis that cardiac Sirtuin 1 (Sirt1) deficiency alters cardiomyocyte Ca2+ and Na+ regulation, leading to cardiac dysfunction and arrhythmogenesis. We used mice with cardiac‐specific Sirt1 knockout (Sirt1−/−). Sirt1flox/flox mice were served as control. Sirt1−/− mice showed impaired cardiac ejection fraction with increased ventricular spontaneous activity and burst firing compared with those in control mice. The arrhythmic events were suppressed by KN93 and ranolazine. Reduction in Ca2+ transient amplitudes and sarcoplasmic reticulum (SR) Ca2+ stores, and increased SR Ca2+ leak were shown in the Sirt1−/− mice. Electrophysiological measurements were performed using patch‐clamp method. While L‐type Ca2+ current (I Ca, L) was smaller in Sirt1−/− myocytes, reverse‐mode Na+/Ca2+ exchanger (NCX) current was larger compared with those in control myocytes. Late Na+ current (I Na, L) was enhanced in the Sirt1−/− mice, alongside with elevated cytosolic Na+ level. Increased cytosolic and mitochondrial reactive oxygen species (ROS) were shown in Sirt1−/− mice. Sirt1−/− cardiomyocytes showed down‐regulation of L‐type Ca2+ channel α1c subunit (Cav1.2) and sarcoplasmic/endoplasmic reticulum Ca2+ ATPase 2a (SERCA2a), but up‐regulation of Ca2+/calmodulin‐dependent protein kinase II and NCX. In conclusions, these findings suggest that deficiency of Sirt1 impairs the regulation of intracellular Ca2+ and Na+ in cardiomyocytes, thereby provoking cardiac dysfunction and arrhythmogenesis.


| INTRODUC TI ON
Sirtuins are nicotinamide adenine dinucleotide-dependent class III histone deacetylases that are involved in ageing, gene silencing and DNA damage repair. [1][2][3] Sirtuin 1 (Sirt1), one of the sirtuins, deacetylates a variety of substrates and modulates angiogenesis and vascular tone, thereby may provide protective effect on atherosclerosis, cardiac ischaemic/reperfusion injury, and catecholamine-induced cardiomyopathy. [4][5][6][7] Sirt1 activator resveratrol was shown to reduce the degree of cardiac dysfunction and hypertrophy in spontaneously hypertensive rats. 8 However, the underlying mechanism employed by Sirt1 in protecting cardiac health remains to be elucidate. 9 Sirt1 deficiency may be proarrhythmogenic. It has recently been reported that Sirt1-deficient mice exhibit abnormalities in cardiac conduction and arrhythmia-induced premature death that leads to the hyperacetylation of Na + channels. 10 Furthermore, Sirt1 modulates intracellular Ca 2+ homeostasis. [11][12][13] Recently a study reported deficiency of Sirt1 enhances the acetylation and alters the function of sarcoplasmic/endoplasmic reticulum Ca 2+ ATPase (SERCA). 14 However, the correlation between Sirt1 and Ca 2+ handling and the effect of Sirt1 deficiency on the homeostasis of intracellular Ca 2+ and Na + in cardiomyocytes are complex and remain poorly understood.
The aim of the present study was to assess whether Sirt1 deficiency in the heart of mice alters intracellular Ca 2+ and Na + regulation, resulting in cardiac dysfunction and predisposition to arrhythmia.
Here, we used cardiac-specific Sirt1 knockout mice to identify the Ca 2+ and Na + regulatory mechanisms affected. We assessed a variety of indices that mandate cardiac function and ionic regulation to establish a clear picture of pathological cellular Ca 2+ and Na + homeostasis in the heart with Sirt1 deficiency. Our results suggest deficiency of cardiac Sirt1 promote dysregulation of Ca 2+ and Na + , leading to contractile dysfunction and providing proarrhythmic substrates. Schneider, Imperial College London) and are currently in use in the laboratory. 15   The average was calculated from measurements taken from three consecutive cardiac cycles.

| Preparation of ventricle tissues for electromechanical and pharmacological analyses
Mice were anesthetized by intraperitoneal injections of Zoletil 50 (5 mg/kg) and xylazine (5 mg/kg) with isoflurane inhalation (5% in oxygen) in a vaporizer. The hearts were harvested from the mice by performing a midline thoracotomy as described previously. 16 The ventricular tissues were separated from the atria at the atrioventricular groove in normal Tyrode's (NT) solution. The ventricular tissue preparation was pinned with needles onto the bottom of a tissue bath. The other end part of the preparation was connected to a Grass FT03C force transducer with silk thread. The preparations were superfused with a solution composed (in mM) of 137 NaCl, 4 KCl, 15 NaHCO 3 , 0.5 NaH 2 PO 4 , 0.5 MgCl 2 , 2.7 CaCl 2 and 11 dextrose at a constant rate (3 ml/min), saturated with a 97% O 2 -3% CO 2 gas mixture. The bath temperature was maintained at 37°C.
Before the electrophysiological assessments, the preparations were allowed to equilibrate in the bath for 1 h.
Transmembrane action potentials (APs) were recorded using 3M KCl-filled glass microelectrodes connected to a WPI Duo 773 electrometer as described previously. 17 Signals were recorded digitally using a data acquisition system with a cut-off frequency of 10-kHz low-pass filter and a 16-bit accuracy at a rate of 125 kHz.
Pulse stimulation with 1-ms duration was provided by a Grass S48 stimulator through a Grass SIU5B stimulus unit. The AP durations (APDs) were measured in ventricle preparations under 2 Hz pulse stimulation. The AP amplitude (APA) was determined by the difference between the peak potential of depolarization and the resting membrane potential (RMP). The repolarization extents of 20%, 50% and 90% of the APA were denoted as the APD 20 , APD 50 and APD 90 . Spontaneous electrical activity and arrhythmia, including burst firing, delayed after depolarizations (DADs), and ventricular tachycardia were recorded and analysed. Ventricle preparations were perfused with KN93, a calmodulin-dependent protein kinase II (CaMKII) inhibitor, (1 μmol/L) or ranolazine, 18 a selective late Na + current (I Na,L ) inhibitor, (10 μmol/L) at a constant rate to determine pharmacological responses.

| Cardiomyocyte isolation
Ventricular myocytes were enzymatically dissociated as previously described with modifications. 19 Briefly, mice were killed using a mixture of Zoletil 50 and xylazine, and the hearts were procured and cannulated via the aorta to a Langendorff perfusion system at 37°C. The heart was firstly perfused with normal Tyrode's (NT) solution for 10 minutes and digested with Ca 2+ -free solution con-

| External solution
The

| Intracellular Ca 2+ monitoring
Cardiomyocytes from control and Sirt1 −/− mice were loaded with Ca 2+ dye (10 μmol/L Fluo-3 AM) at room temperature for 30 minutes and imaged as previously described method. 20  (10 mmol/L NiCl 2 )-sensitive currents as previously described. 21 The I Na was elicited during potential steps from a holding potential of −120 mV to testing potentials from −80 to 0 mV in 10-mV steps for 40 ms at a frequency of 3 Hz. I Na,L was measured using a step/ramp protocol as described below: start with a potential of −100 mV stepping to +20 mV for 100 ms afterwards ramp back to −100 mV for 100 ms The I Na,L was determined as tetrodotoxin (30 µmol/L TTX)-sensitive current obtained when the potential was ramped back to −100 mV.

| Measurement of ROS and cytosolic Na + level
Ventricular myocytes were incubated in NT solution with 10 μmol/L CellROX green and 2 μmol/L MitoSOX Red (Life Technologies) to assess cytosolic and mitochondria reactive oxygen species (ROS) production, respectively. Myocytes incubated with 5 μmol/L Asante NaTRIUM Green-2 AM (Teflabs) was used to measure the cytosolic Na + level. Experiments were conducted using an inverted laser-scanning confocal microscope (Zeiss LSM 510, Carl Zeiss) with a 63x1.25 objective as previously described. 24 Excitation light with wavelength of 488 nm was used, and emission fluorescence was detected at wavelengths over 505 nm in the XY mode of the confocal microscope system. Cardiac myocytes were paced at 1 Hz in the experiment. Images were analysed using ImageJ as described previously. 25

| Western blot analysis
The protein extraction buffer contained 100 mmol/L Tris-HCl (pH secondary IgG antibodies at a dilution of 1:10 000. Immunoreactive proteins were detected by enhanced chemiluminescence (GE Healthcare, Chicago, USA) and quantified using the ImageJ software.

| Acquisition systems and statistical analysis
Continuous values have been expressed as mean ± SEM. Student's t test, or Pearson's chi-square test were used to compare the differences. The SigmaPlot version 12 (Systat Software Inc., San Jose, CA, USA) was used for statistical comparisons. The 'n' stands for the total cells from the total number of hearts (n = cells/hearts) and the 'N' is the animal numbers. Statistical significance was represented as *, **, and *** for P < 0.05, P < 0.01 and P < 0.005, respectively.

| In Vivo M-mode echocardiography
Sirt1 −/− mice possessed larger LVIDs than those in the control group ( Figure 1B). FS and EF decreased in the Sirt1 −/− mice as compared to those in the control mice ( Figure 1B).

| Ventricle electrical activity
The ventricles in Sirt1 −/− mice showed faster rates of spontaneous activity as compared with those in the control mice ( Figure 1C). The APD 20 , APD 50 , APD 90 , APA and RMP showed no difference between the Sirt1 −/− and control mice ( Figure 1D). The increase in the rate of spontaneous activity in the ventricles of Sirt1 −/− mice was suppressed by KN93, and ranolazine (Figure 2A,B). Furthermore, Sirt1 −/− ventricles showed an increased incidence of burst firing compared with that in the control mice; this phenotype was abrogated upon treatment with KN93 or ranolazine ( Figure 2C,D).

| Ca 2+ transient amplitudes, SR Ca 2+ stores and SR Ca 2+ leak
Steady-state and caffeine-induced Ca 2+ transient amplitudes in cardiomyocytes were 26% and 23% lesser in Sirt1 −/− mice as compared with those in the control mice, respectively ( Figure 3A,B).
Sarcoplasmic reticulum Ca 2+ content, obtained by integrating the caffeine-induced inward NCX current, was 39% less in the Sirt1 −/− mice than that in the control mice ( Figure 3C). Sirt1 −/− cardiomyocytes had 55% larger SR Ca 2+ leak compared with that in the control cardiomyocytes ( Figure 3D).

| L-type Ca 2+ current and nickel-sensitive NCX current
The density of I Ca,L in the Sirt1 −/− myocytes was smaller compared with those in the control myocytes ( Figure 4A). Moreover, Sirt1 −/− ventricular myocytes showed larger reverse-mode of nickel-sensitive NCX current compared with that in the control ventricular myocytes ( Figure 4B).

| Expression of intracellular Ca 2+ regulatory proteins
We determined the protein expressions associated with intracellular Ca 2+ regulation in the cardiomyocytes of control and Sirt1 −/− mice using Western blotting ( Figure 7A). The L-type Ca 2+ channel subunit α1c was down-regulated in Sirt1 −/− ventricles compared with the control ventricles ( Figure 7B). While the protein levels of SERCA2a was reduced in Sirt1 −/− mice, NCX and CaMKII in Sirt1 −/− ventricles were up-regulated compared with the control mice ( Figure 7B).

F I G U R E 2
Electrocardiographic changes of ventricles and burst firing in control and Sirt1 −/− mice. A and B, Sirt1 −/− ventricles showed faster rates of spontaneous activity as compared to those in the control mice that were suppressed upon treatment with KN93 (Control N = 6 and Sirt1 −/− N = 6; *P < 0.05) and ranolazine (Control N = 6 and Sirt1 −/− N = 7; **P < 0.01, ***P < 0.005). C and D, Sirt1 −/− ventricles showed an increased incidence of burst firing that was inhibited by KN93 and ranolazine (Sirt1 −/− N = 7; *P < 0.05, ***P < 0.005) SR. 26 Calmodulin-dependent protein kinase II phosphorylates phospholamban at Ser-10 to decrease the function of SERCA2a. 27    In conclusion, Sirt1 deficiency in the cardiac tissues resulted in detrimental effects on Ca 2+ and Na + regulation in mice cardiomyocytes. Dysregulated Ca 2+ handling and Na + regulation leads to a higher frequency of ventricular arrhythmia and cardiac dysfunction. I Na,L was enhanced alongside with increased cytosolic Na + level. ROS production and expression of CaMKII were higher in the Sirt1 −/− mice. The CaMKII inhibitor KN93 and ranolazine prevented arrhythmia. These findings suggest that the deficiency of Sirt1 in the cardiomyocytes leads to dysregulation of intracellular Ca 2+ and Na + that provide proarrhythmic substrates.  Taiwan (TSGH-C108-025, ATSGH-C107-200, TSGH-C108-008-S06, and MAB-108-082). The authors are grateful to Dr Shih-Che Hsu for expertise with in vivo mouse echocardiography. We thank Elsevier Author Services (https://websh op.elsev ier.

ACK N OWLED G EM ENTS
com/langu age-editi ng-servi ces/) for English editing service.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data of the present study are available from the corresponding authors following reasonable request.