Electrophysiological characteristics of pressure overload-induced cardiac hypertrophy and its influence on ventricular arrhythmias

Objective To explore the cardiac electrophysiological characteristics of cardiac hypertrophy and its influence on the occurrence of ventricular tachyarrhythmias. Methods Adult C57BL6 mice were randomly divided into a surgery group and a control group. Thoracic aortic constriction was performed on mice in the surgery group, and cardiac anatomical and ultrasonic evaluations were performed to confirm the success of the cardiac hypertrophy model 4 weeks after the operation. Using the Langendorff method of isolated heart perfusion, monophasic action potentials (MAPs) and the effective refractory period (ERP) at different parts of the heart (including the epi- and endo-myocardium of the left and right ventricles) were measured, and the induction rate of ventricular tachyarrhythmias was observed under programmed electrical stimulus (PES) and burst stimulus. Whole-cell patch-clamp was used to obtain the I-V characteristics of voltage-gated potassium channels in cardiomyocytes of different parts of the heart (including the epi- and endo-myocardium of the left and right ventricles) as well as the channels’ properties of steady-state inactivation and recovery from inactivation. Results The ratio of heart weight to body weight and the ratio of left ventricular weight to body weight in the surgery group were significantly higher than those in the control group (P < 0.05). Ultrasonic evaluation revealed that both interventricular septal diameter (IVSD) and left ventricle posterior wall diameter (LVPWD) in the surgery group were significantly larger than those in the control group (P < 0.05). Under PES and burst stimuli, the induction rates of arrhythmias in the surgery group significantly increased, reaching 41.2% and 23.5%, respectively. Both the QT interval and action potential duration (APD) in the surgery group were significantly longer than in the control group (P<0.01), and the changes showed obvious spatial heterogeneity. Whole-cell patch-clamp recordings demonstrated that the surgery group had significantly lower potassium current densities (IPeak, Ito, IKur, Iss, and IK1) at different parts of the heart than the control group (P < 0.01), and there were significant differences in the half-inactivation voltage (V1/2) and the inactivation-recovery time constant (τ) of Ito and IKur at different parts of the heart (P < 0.01) between the surgery group and the control group. In addition, the surgery group had significantly lower densities of IPeak, Ito, and IKur in cells of the endo-myocardium (P < 0.05), and the changes showed obvious spatial heterogeneity. Conclusion Changes in the current density and function of potassium channels contributed to irregular repolarization in cardiac hypertrophy, and the spatially heterogeneous changes of the channels may increase the occurrence of ventricular arrhythmias that accompany cardiac hypertrophy.


Results
The ratio of heart weight to body weight and the ratio of left ventricular weight to body weight in the surgery group were significantly higher than those in the control group (P < 0.05). Ultrasonic evaluation revealed that both interventricular septal diameter (IVSD) and left ventricle posterior wall diameter (LVPWD) in the surgery group were significantly larger than those in the control group (P < 0.05). Under PES and burst stimuli, the induction rates of arrhythmias in the surgery group significantly increased, reaching 41.2% and 23.5%, respectively. Both the QT interval and action potential duration (APD) in the surgery group were significantly longer than in the control group (P<0.01), and the changes showed obvious spatial heterogeneity. PLOS

Introduction
Cardiac hypertrophy, which is regarded as an adaptive process under the influence of continuous haemodynamic overload, is a common complication of hypertension, ischemic cardiomyopathy, valvular diseases, and many other cardiovascular diseases [1]. Without early intervention, cardiac hypertrophy leads to cardiac enlargement, congestive heart failure, and malignant arrhythmias, which are very common complications of cardiac hypertrophy, ultimately resulting in sudden cardiac death [2]. Although great progress has been achieved regarding the use of medication to treat this disease, its morbidity remains high at approximately 43% among people over 70 and more than 50% among severe hypertension patients [3]. Therefore, research has focused on the pathogenesis of ventricular arrhythmias, which occur during the occurrence and progression of cardiac hypertrophy, to identify effective control measures. Electrophysiological changes (electrophysiological remodelling) during the occurrence and progression of cardiac hypertrophy primarily manifest as repolarization delays, including extensions of the QT interval and APD, and a resulting increase in the dispersion of repolarization [4][5][6]. A previous study revealed that APD prolongation is attributable to alterations in multiple potassium channels [7] and that potassium channels that function in key ion transport during the repolarization phase in cardiomyocytes played a critical role in the occurrence of arrhythmias in a model of cardiac hypertrophy [8]. Studies have revealed that potassium current densities are decreased in patients with left ventricular hypertrophy [9]. It is not clear whether the heterogeneous distribution of channels is associated with the occurrence of arrhythmias in cardiac hypertrophy. With the establishment of a cardiac hypertrophy model, changes in the potassium-ion channel density and kinetics in different regions of the heart were detected in this study, and the results may provide a theoretical basis for the pathogenesis of arrhythmias that accompany cardiac hypertrophy.

Materials and methods
All protocols conformed to the Guide for the Care

Animal model
All C57BL/6 mice for research (male, 10-12 weeks old) were provided by the animal experimental centre. The animals were reared in the room at constant temperature and humidity. All mice were anaesthetized with 3% pentobarbital sodium (50 mg/kg). The cardiac hypertrophy model was established by performing thoracic aortic constriction on the mice in the surgery group. Thoracotomy was performed on animals in the control group (sham group), but no treatment was performed on their aortas. A cardiac ultrasonic evaluation was carried out to confirm the success of the model 12 days after the operation.

ECG recording
Three percent pentobarbital sodium (prepared with normal saline) was administered to mice through an intraperitoneal injection at a dose of 0.03 ml/g. After 10-20 min, mice were placed on a constant-temperature heating plate (37˚C). Ag/AgCl electrodes (Linton Instruments, Harvard Apparatus) were placed in the subcutaneous tissue of the four limbs and connected to an ECG amplifier (AD Instruments, Australia). The low-pass filter and high-pass filter were adjusted to 0.05 Hz and 1 kHz, respectively. After stabilization for 5 min, lead II ECG recording from the limbs was performed for 30 min.

Isolated heart preparation
A total of 100 IU of heparin sodium was administered to mice through intravenous and intraperitoneal injections to inhibit coagulation. Ten minutes later, the mice were killed, and their hearts were removed immediately and placed into 100% oxygen-saturated normal saline (4˚C) to remove the pericardium and surrounding adipose tissue. A 22-gauge needle was inserted through the aorta under a microscope in a retrograde manner and then connected to a Langendorff heart perfusion apparatus (AD Instruments, Australia). Retrograde trans-aortic heart perfusion with 100% oxygen-saturated Tyrode's-HEPES solution (mM: NaCl 130; KCl 5.4; CaCl 2 1.8; MgCl 2 1; Na 2 HPO 4 0.3; HEPES 10; glucose 10; the pH value was adjusted to 7.4 by adding NaOH; at a constant temperature of 37˚C) was carried out at 2-2.5 ml/min, and the perfusion pressure was kept within the range of 60~80 mmHg. The entire procedure from removing the heart to achieving perfusion and heart resuscitation must be completed within 5 min, and recording should be started after stabilization for 20 min. Samples that were found with obvious ischaemia or arrhythmias within 20 min after perfusion were excluded.

MAP and bipolar ECG recording
A pair of platinum stimulating electrodes (at an interval of 1 mm) was placed on the outer myocardium at the right ventricular base for cardiac pacing. The frequency was 8 Hz, pulse width was 1 ms, and the stimulating voltage was 2 times the diastolic pacing threshold. MAP electrodes (Linton Instruments, Harvard Apparatus) were used to record the MAP of left and right ventricles. By using silver chloride bipolar electrodes (Linton Instruments, Harvard Apparatus), which were 0.25 mm in diameter and coated with Teflon, a bipolar ECG recording was carried out to record the waveforms of ventricular arrhythmias. All signals recorded using a MAP amplifier (AD Instruments, Australia), and the filter range was 0.3 Hz~1 kHz.

Programmed electrical stimulus.
The bipolar electrodes and MAP electrodes were placed on the ventricles to record the MAPs and bipolar ECGs at baseline; 10 min later, PES (8 S1 stimuli plus a single extra-stimulus, S2) was delivered. The S2 was delivered from 125 ms, and reverse scanning was performed at a step of -1 ms. Changes in the AP and bipolar ECG were observed. The ventricular effective refractory periods, types of ventricular arrhythmias, durations, and frequencies were recorded.

Burst stimulus.
The stimulating electrodes were used to deliver burst stimuli at the right ventricular base at a frequency of 50 Hz for 2 s. The burst frequency was 0.2 Hz, and the total stimulation time was less than 3 min [10]. The types of ventricular arrhythmias, durations, and frequencies were recorded. The effective refractory period was defined as the longest S1-S2 interval that failed to result in an excitation wave. The durations of ventricular tachyarrhythmia (VT) were divided into sections of <10 s, 10~30 s, and >30 s. VTs with a duration of >30 s were defined as sustained ventricular tachyarrhythmias.

Separation of ventricular myocytes
A total of 100 U heparin was administered to the mice via intraperitoneal injection. Ten minutes later, the mice were decapitated, and their hearts were immediately removed and placed in a Langendorff recirculating perfusion apparatus after aortic intubation. Cardiomyocytes were isolated via the step-by-step perfusion method: 1) Tyrode's-HEPES solution (mM: NaCl 130; KCl 5.4; CaCl 2 1.8; MgCl 2 1; Na 2 HPO 4 0.3; HEPES 10; glucose 10; the pH value was adjusted to 7.4 by adding NaOH) for 5 min; 2) calcium-free Tyrode's-HEPES solution for 5 min; 3) enzyme-containing digestive fluid (30 ml calcium-free Tyrode's solution; 0.6 mg/ml collagenase II; 0.1% BSA; 20 mM taurine; 30 μM CaCl 2 ) for 15 min; 4) KB solution (mM: taurine 10; glutamic acid 70; KCl 25; KH 2 PO 4 10; glucose 22; EGTA 0.5; the pH value was adjusted to 7.2 by adding KOH) for 5 min. The ventricles were cut off after perfusion. Under a microscope, iris scissors were used to separate the ventricular wall between the left and right ventricles as well as the inner and outer myocardium of the left ventricular free wall [11]. All separation procedures were carried out at room temperature (25˚C). The separated cardiomyocytes were placed into KB solution and stored in a refrigerator at 4˚C.

Recording of I-V characteristics of potassium channels in cardiomyocytes
An EPC-9 amplifier was used in the whole-cell patch-clamp technique, and Pulse was used for data recording and analysis. Cells were placed in a perfusion chamber and treated with continuous constant-temperature perfusion of calcium-free Tyrode's solution at 2 ml/min. The electrode impedance was 2.5-5 MΩ, and intracellular fluid (mM: K-aspartate 110, KCl 20, NaCl 8, MgC1 2 1, CaC1 2 1, MgATP 4, EGTA 0.1 and 10 HEPES; the pH value was adjusted to 7.2 by adding KOH) was perfused. The series resistance was kept within the range of 4-8 MΩ and at room temperature within the range of 22-25˚C. 1) For the I-V characteristics of the total potassium current (I peak ), the current was derived by the test pulses (500 ms, -40 mV-+60 mV, at a step of 10 mV), the clamping voltage was -80 mV, and the pulse frequency was 0.1 Hz. 2) For recording fast-activated transient outward rectifying potassium current (I Kur ), the current was derived by first delivering a pre-stimulation (100 ms, -40 mV) to inactivate the transient outward potassium current (I to,f ) and then by delivering the test pulses (500 ms, -40 mV-+60 mV, at a step of 10 mV). 3) For recording steady-state outward potassium current (I SS ), the current was derived by adding 100 μM 4-aminopyridine (4-AP) after recording I Kur . 4) For recording I to,f , the current was derived under the settings of holding potential: -80 mV; testing potential: -40 mV-+60 mV; step: 10 mV; and clamping time: 400 ms. 5) For recording inward rectifying potassium current (I K1 ), the current was derived under the settings of holding potential: -80 mV; testing potential: -120 mV--40 mV; step: 10 mV; and clamping time: 350 ms.

Recording of I to,f and I Ku channel kinetics
1) The steady-state inactivation properties of I to and I Kur were obtained by using double-pulse stimuli. The cell membrane potential was kept at -80 mV; extra-stimuli (-110 mV--10 mV, at a step of 10 mV) were delivered for 1 s, and then, the inactivation properties of I to were recorded under the testing voltage of +30 mV for 1 s. The cell membrane potential was kept at -80 mV; extra-stimuli (-110 mV--10 mV, at a step of 10 mV) were delivered for 5 s, and then, the prestimulation (100 ms, -40 mV) was delivered to inactivate I to . The inactivation properties of I to were then recorded under the test pulses (5 s, +30 mV). Standardized ratios (I/I max ) were calculated by dividing the test current amplitudes under all voltages by the maximum amplitude. Linear fitting of the data was performed using the Boltzmann equation, and inactivation curves of the channels were then obtained. 2) Recovery properties of I to after the time-dependent inactivation: First, the cell membrane potential was kept at -80 mV; clamping pulses (+30 mV, 500 ms) were delivered to inactivate the I to channel; clamping pulses (+30 mV, 500 ms) were delivered at different time intervals, and peak current amplitudes of I to at the time points within 10 ms-500 ms were recorded. Recovery properties of I Kur after the time-dependent inactivation: Immediately after sending the clamping pulses (+30 mV, 1.5 s), pre-stimulation (100 ms, -40 mV) was delivered to inactivate I to , and clamping pulses (+30 mV, 500 ms) were delivered at different time intervals (10 ms-3000 ms). Standardized ratios (P2/P1) were calculated by dividing the test current peaks at the time points by the current peaks derived by the inactivation pulses. Exponential curve fitting of the data with respect to the time interval was performed.

Statistical analysis
Lab chart 7.0 was used to collect and measure the AP graphs. Data analysis was carried out with SPSS16.0. Pulsefit and Origin 8.0 were used for current curve drawing and analysis. The data are expressed in the form of the mean ± standard deviation. Differences with P < 0.05 were considered statistically significant.

Assessment of the cardiac hypertrophy model via anatomical parameters and echocardiography
Cardiac ultrasonic examination was performed four weeks after the cardiac hypertrophy model was established, and the results showed that LVEDD, IVSD, LVPWD, and FS in the surgery group were all significantly higher than in the control group. After animals were sacrificed, their hearts were removed and weighed. The results showed that both HW/BW and LVW/BW in the surgery group were significantly higher than in the control group (see Table 1). This finding demonstrated that the cardiac hypertrophy model, induced by thoracic aortic constriction, was successfully established.

Assessment of repolarization phase by surface ECG and MAP
Analysis of lead II ECGs revealed no significant difference in heart rate between the surgery group and the control group  1).
During the delivery of the stimuli at S1-S1 cycle lengths of 200 ms, 150 ms, 125 ms, and 100 ms, the ventricular ERPs at different frequencies in the surgery group were all higher than those in the control group: 45.0±12

Induction rate of ventricular arrhythmias
Ventricular tachyarrhythmias were induced under both PES and burst stimuli, and both induction rates in the surgery group (41.2% and 23.5%, respectively) were higher than those in the control group (12.5% and 12.5%, respectively). The results also indicated that the hearts of mice in the surgery group were more sensitive to PES. Furthermore, the duration of the arrhythmias in the surgery group was longer than that in the control group. The induction rate of sustained ventricular arrhythmias (>30 s) was 5.9% in the surgery group, and the rate was zero in the control group (See Fig 2). The surgery group had a longer cycle length of arrhythmia waves. In addition, regarding the types of arrhythmias, polymorphic ventricular tachyarrhythmias were observed in the surgery group (5.9%) (See Table 2).

Changes in I-V characteristics of potassium currents
As revealed by the whole-cell recording of total outward potassium currents (I Peak ) in cardiomyocytes of different parts (including the endo-and epi-myocardium of the left ventricle and right ventricle), changes in the I Peak density in cardiomyocytes of mice in the surgery group with voltage were significantly lower than those in the control group (see Fig 3). The peak current densities (+60 mV) in the epi-myocardium, endo-myocardium, and right ventricle were 26.6±6.1 pA/pF vs. 48.1±10.3 pA/pF (P<0.05), 21.4±4.7 pA/pF vs. 46.1±8.5 pA/pF (P<0.05), and 30.8±3.8 pA/pF vs. 39.7±5.5 pA/pF (P<0.05), respectively. The total outward potassium currents included many components, and the transient outward potassium current (I to,f ), ultra-rapid delayed rectifier potassium current (I Kur ), and steady-state potassium current (I ss ) played dominant roles. The I-V characteristics of these three outward potassium currents in the surgery group were significantly different from those in the control group, the changes in the endo-myocardium of the left ventricle were particularly significant, and the peak currents of all components (+60 mV) were significantly different compared to the control group (see Table 3). In addition, the potassium current densities in cardiomyocytes of the left ventricle in the surgery group showed obvious transmural heterogeneity; the densities of I Peak , I to,f , and I Kur in the endo-myocardium of the left ventricle were significantly lower than those in the epi-myocardium of the left ventricle (P < 0.05). However, this heterogeneity was not observed in the control group. The I K1 density in cardiomyocytes of the endo-and epi-myocardia of the left ventricle in the surgery group was significantly lower than that in the control group (P<0.05) (see Fig 3); the peak current densities (-120 mV) in the epi-myocardium and endo-myocardium were 14.3 ±3.2 pA/pF vs. 18.0±4.3 pA/pF (P<0.05) and 15.3±3.6 pA/pF vs. 18.7±1.3 pA/pF (P<0.05), respectively. However, no significant differences between the two groups were found in cardiomyocytes of the right ventricle. In addition, I K1 density in the surgery group did not show obvious transmural heterogeneity (see Table 3).

Changes in the kinetics of I to and I Kur channels
To further verify whether the decreases in the potassium current densities in cardiac hypertrophy were related to channel kinetics, the kinetic characteristics of I to,f and I Kur were analysed in this study. The results showed that compared to the control group, the steady-state inactivation curves of I to,f in the epi-myocardium, the endo-myocardium, and the right ventricle in the surgery group noticeably shifted to the left (direction of hyperpolarization). However, the inactivation recovery curves of I to,f in the surgery group shifted to the right, and the time constant (τ) was significant longer than in the control group (P<0.01) (Fig 4 and Table 4). In addition, kinetics of I Kur in the surgery group showed the same changing characteristics as I to,f . V 1/ Previous studies have shown that alterations in I to expression are linked to cardiac hypertrophy [16]. Additionally, the subunits and densities of K + channels, including I to,f (Kv4.2 and Kv4.3), I ss (Kv2.1), and I kur (Kv1.5), contribute to ventricular repolarization [11,17]. However, these previous studies only showed that the pathogenesis of ventricular arrhythmias in cardiac hypertrophy or heart failure is caused by a prolonged QT interval [18]. The present study not only indicated that the extension of the repolarization phase was an important cause of ventricular arrhythmias induced by cardiac hypertrophy but also demonstrated that the repolarizing heterogeneity of cardiac hypertrophy was caused by density and kinetic heterogenous changes in Kv channels. In the present study, the prolonged repolarization was associated with marginally increasing ERP and predominant prolongation of APD 50 -ADP 90 . Thus, the relative refractory period and supranormal period after ERP were prolonged. This change increased the window of vulnerability and facilitated the formation of trigger activity by premature stimuli.
Under normal conditions, K + channels are heterogeneously distributed in the ventricular walls in animals. The subunits of K + channels, including I to,f , I kur , and Iss, contribute to heterogeneities in ventricular repolarization, and the density of K + currents was more prominent in the epicardium than in the endocardium of mice [11,17]. Consistently, the control group in our study also showed the same profile of heterogeneity in APD and K + channel density between the endo-and epicardium, and this heterogeneity was markedly enhanced in cardiac hypertrophy. Furthermore, kinetic heterogeneity of multiple K + channels were also demonstrated in the present study; these alterations have not been mentioned by previous studies. Notably, the increased dispersion of V 1/2 and τ of I to and I Kur resulted in heterogeneity of depolarization and repolarization in the ventricles. Thus, the formation of a re-entry mechanism was greatly promoted, and arrhythmia was easily induced by electrical stimulation in cardiac hypertrophy.
Previous studies have indicated that in pressure overload-induced cardiac hypertrophy [14,19,20], the histomorphology of the myocardium is characterized by obvious heterogeneity. Namely, cardiomyocytes in the endo-myocardium of the left ventricle have a higher degree of   Table 3. Peak densities of K + currents in myocytes of epicardium, endocardium and the right ventricle.

Control group Surgery group
Epi (n = 13) Endo (n = 10) RV (n = 11) Epi (n = 10) Endo (n = 10) RV (n = 10) Cardiac hypertrophy and ventricular arrhythmias hypertrophy than cardiomyocytes in the epi-myocardium [14]. This may be because the endomyocardium has a larger mechanical tension than the epi-myocardium under a high-pressure load. As a result, the increase in the volume of cardiomyocytes in the endo-myocardium is more significant than in the epi-myocardium, and cardiomyocytes in the endo-myocardium have a significantly lower ion channel density. Notably, this study found that cardiomyocytes of the endomyocardium had more significant APD prolongation, which was associated with the spatial heterogenous decrease in potassium current densities. The changes in the cardiomyocytes of the endomyocardium were particularly significant and directly influenced the transmural electrophysiological heterogeneity of the ventricular myocardium. Transmural heterogeneity has been found in the hearts of many animals and humans, and the cause of this phenomenon is considered to be the regional differences in the density and type of repolarization K + currents in the myocardium Cardiac hypertrophy and ventricular arrhythmias   [21]. Several studies have proposed that there is a transmural gradient of I to between the endomyocardium and epi-myocardium of mice, which is caused by the heterogeneity of expression of the α subunit (Kv4.3) and the calcium-binding auxiliary subunit (KChIP2) [22,23]. However, whether there are physiologic transmural differences in outward rectifier currents (I Kur, I Kr , and I Ks ) remains controversial. Brunet et al. noted that cardiomyocytes in the epi-myocardium of C57BL6 mice had a higher I Ks density than those in the endo-myocardium. Because the inactivation time constant of I Ks in the endo-myocardium is significantly higher than in the epi-myocardium, they speculated that different types of I Ks channels were distributed on the membrane of cardiomyocytes in the endo-and epi-myocardia. However, the expression levels of Kv1.5 and Kv2.1 fail to indicate a transmural gradient [11]. Marionneau et al. found that in aortic coarctation-induced cardiac hypertrophy, there were no significant differences in the amplitudes of I to , I K,slow , and I K1 between cardiomyocytes in the endo-myocardium and those in the epi-myocardium, and there was no transmural spatial heterogeneity in the expression of channel proteins, I to (Kv4.2, Kv4.3, KChIP2), I K1 (Kir2.1, Kir2.2), and I ss (TASK1, TASK2). However, the membrane capacitance (C m ) increases significantly, thus changing the transmural spatial heterogeneity of potassium currents in the myocardial tissue. This finding indicates that the transmural spatial heterogeneity in cardiac hypertrophy does not depend on changes in potassium channel protein expression and ion currents but depends on heterogeneous changes in the cell volume [14]. The results of our present study also demonstrate that the I to,f and I Kur densities change more significantly in cardiomyocytes of the endo-myocardium and that the changes are consistent with changes in the membrane capacitance.

Conclusion
The non-uniform attenuation of potassium currents in cardiomyocytes contributes to irregular repolarization in cardiac hypertrophy, and the spatially heterogeneous changes of the channels may directly increase the occurrence of ventricular arrhythmias that accompany cardiac hypertrophy.