Altered Spatiotemporal Dynamics of the Mitochondrial Membrane Potential in the Hypertrophied Heart

Abstract

Chronically elevated levels of oxidative stress resulting from increased production and/or impaired scavenging of reactive oxygen species are a hallmark of mitochondrial dysfunction in left ventricular hypertrophy. Recently, oscillations of the mitochondrial membrane potential (ΔΨ_m) were mechanistically linked to changes in cellular excitability under conditions of acute oxidative stress produced by laser-induced photooxidation of cardiac myocytes in vitro. Here, we investigate the spatiotemporal dynamics of ΔΨ_m within the intact heart during ischemia-reperfusion injury. We hypothesize that altered metabolic properties in left ventricular hypertrophy modulate ΔΨ_m spatiotemporal properties and arrhythmia propensity.

Introduction

Left ventricular hypertrophy (LVH) in response to elevated systemic pressure forms an adaptive and reversible mechanism that allows the heart to maintain cardiac output in the face of external stress. However, because LVH is associated with complex electrophysiological, structural, molecular, mechanical, and metabolic remodeling, it ultimately becomes a leading risk factor for heart failure, myocardial infarction, and ventricular arrhythmias (1).

Changes in mitochondrial energetics are a hallmark of hypertrophied and failing hearts (2–6). Numerous studies have shown that mitochondria are indeed central to the pathophysiology of the hypertrophied heart. For example, a marked shift in substrate utilization (free fatty acid to glucose) for energy production has been documented in various animal models and humans with hypertrophy (7). In addition, LVH is associated with mitochondrial structural deformities (8), including increased mitochondrial volume and disrupted mitochondrial network architecture. These fundamental changes, which alter the intricate regulation of myocardial bioenergetics, disrupt the balance between pro- and antioxidant pathways, and promote the activation of cell death pathways, contribute significantly to the progression of cardiac dysfunction and remodeling.

The mitochondrial membrane potential (ΔΨ_m) is a key indicator of cellular viability because it reflects the capacity of the electron transport chain to pump hydrogen ions across the inner membrane, a requirement for oxidative phosphorylation and ATP generation (9,10). Chronic alterations in ΔΨ_m in various metabolic disorders cause cell death. Dynamic oscillations of ΔΨ_m produced by artificial conditions of oxidative stress in isolated cardiac myocytes can result in metabolic and electrophysiological oscillations that significantly impact myocyte function and lead to inexcitability at the cellular level by activating ATP-sensitive potassium channels (11–13). Furthermore, stabilization of ΔΨ_m using pharmacological interventions that inhibit the mitochondrial benzodiazepine receptor can lead to action potential (AP) stabilization and prevention of arrhythmias in ex vivo perfused hearts subjected to ischemia-reperfusion (IR) injury (11,14). As such, changes in ΔΨ_m are thought to modulate cellular electrophysiological changes that can scale to the level of the intact organ and affect excitability and arrhythmia propensity via a mechanism that Akar et al. (11) termed metabolic sink/block.

Despite correlative findings linking interventions that alter mitochondrial function in vitro to functional changes in vivo, a direct investigation of dynamic changes in ΔΨ_m at the level of the intact organ remains lacking because of technical challenges in measuring the spatiotemporal kinetics of ΔΨ_m across the intact heart. In this work, we developed a novel (to our knowledge), quantitative imaging technique for measuring the spatiotemporal dynamics of ΔΨ_m with subcellular resolution at the organ level. We then assessed dynamic changes in ΔΨ_m gradients during IR injury in normal and chronically diseased LVH hearts.

Materials and Methods

Pressure overload hypertrophy model

All procedures involving the handling of animals were approved by the Animal Care and Use Committee of the Mount Sinai School of Medicine and adhered to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health. As previously described in detail (15,16), pressure overload LVH was produced in male Sprague-Dawley rats (200–250 g) by ascending aortic banding. Briefly, 14 male rats underwent a right lateral thoracotomy to allow the placement of a 1-mm-diameter metallic clip around the ascending aorta. The animals were allowed to survive for 1 month after the procedure, which established an artificial aortic stenosis. We recently characterized the electrophysiological, mechanical, and structural changes that develop in this model of concentric hypertrophy (15). Thirteen normal rat hearts served as the control group.

Echocardiographic measurements

Echocardiography was performed under ketamine sedation (intraperitoneal, 60–100 mg/kg). Long- and short-axis views were obtained using a 14 MHz GE-i13L probe as recommended by the American Society of Echocardiography. Short-axis views were obtained as transverse sections of the LV at the mid-papillary level, and echocardiographic windows were selected to obtain circular images. The transducer position was carefully adjusted until the short-axis image of the LV cavity appeared to be circular, indicating perpendicular intersection of the LV by the ultrasonic beam. A 12MHz GE-10S probe was used to obtain apical four-chamber views. LV wall thickness and cavity dimensions were measured in systole and diastole with M-mode echocardiography. Additional measurements of the LV end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), ejection fraction (EF), and heart rate (HR) were also performed in all animals at baseline and after 4 weeks (time of sacrifice) of aortic banding or sham operation.

High-resolution optical ΔΨ_m imaging in ex vivo perfused rat hearts

The animals (n = 27) were anesthetized with pentobarbital sodium (30 mg/kg, IP). Their hearts were rapidly excised, washed with ice-cold cardioplegic solution, transferred to a Langendorff apparatus, and retrogradely perfused through the aorta with oxygenated (95% O₂/5% CO₂) Tyrodes solution containing (in mM) NaCl 121.7, NaHCO₃ 25, KCl 4.81, MgSO₄ 2.74, CaCl₂ 2.5, and dextrose 5.0, pH = 7.40, at a temperature of 36 ± 1°C. Perfusion pressure was maintained at ∼60 mmHg by adjusting perfusion flow. Volume-conducted electrocardiograms were recorded for rhythm analysis and arrhythmia scoring using noncontact silver electrodes (Grass Instruments) placed within the chamber. ECG signals, which were recorded continuously throughout the entire ex vivo perfusion protocol, were amplified (ECG100-MP150 amplifier; Biopac Systems, Goleta, CA) and displayed in real time using the AcqKnowledge 3.9 software package (Biopac Systems). The hearts were positioned such that the mapping field was centered over a 4 × 4 mm region of LV epicardium, midway between the apex and base (Fig. 1 A). These preparations remained stable for >4 h of perfusion. The ex vivo experimental protocol was typically completed in <2 h.

Figure 1.

(A) Three representative traces (A–C) from a total of 6400 pixels are shown indicating the measurement of background fluorescence used to monitor relative changes in ΔΨ_m. Also shown is the background fluorescence during and after (for up to 300 min) the dye-loading and washout procedures. These data demonstrate the stability of TMRM fluorescence during steady-state perfusion. The steady-state level of TMRM fluorescence for each pixel is used for normalization purposes. (B) Representative ECG images of a sham-operated and an aortic-banded LVH heart. H&E-stained histological sections from sham-operated and LVH hearts indicate marked concentric hypertrophy in aortic banded animals. Also shown are representative optical AP traces measured in representative di-4-ANEPPS-stained normal and LVH hearts during steady-state pacing.

We developed a novel (to our knowledge), semiquantitative imaging technique of optical ΔΨ_m mapping in ex vivo perfused guinea pig (17) and rat (18,19) hearts using the ΔΨ_m-sensitive dye TMRM. This method allows mitochondrial function to be assessed at a subcellular resolution within the intact organ (18,19). Briefly, after cannulation, the animals' hearts were allowed to stabilize for 20 min at physiological temperature (36 ± 1°C). They were then stained with TMRM (250 nM; Molecular Probes) mixed in a 500 mL volume of Tyrodes solution (dye-loading phase) for 20 min. This was followed by a 20–30 min dye washout phase during which perfusion was switched back to dye-free Tyrodes solution. TMRM background fluorescence intensity was measured periodically (in 1 min intervals) throughout the dye-staining and washout phases using a 6400 pixel CCD-based optical imaging approach that allowed the measurement of normalized ΔΨ_m with subcellular resolution (50μm) over a 4 × 4 mm window of the epicardial surface. To measure TMRM background fluorescence, the hearts were excited with filtered light (525 ± 20 nm) emitted from a quartz tungsten halogen lamp (Newport Corp., Goleta, CA). Emitted fluorescence was filtered (585 ± 20 nm for TMRM) and focused onto the high-resolution CCD camera (Scimeasure, Decatur, GA). Background fluorescence intensity was measured as the amplitude difference before and after excitation (Fig. 1 A).

During the dye-loading procedure, a marked rise in background fluorescence was verified for multiple pixels to confirm adequate dye staining of the heart (Fig. 1 A). During dye washout, the stability of TMRM background fluorescence was evaluated from multiple pixels in real time, as this baseline level served for normalization purposes during the IR protocol. In all experiments, the dye washout phase was associated with stable signal intensity, which was required to normalize subsequent changes in TMRM fluorescence caused by IR injury. In preliminary experiments, we ensured the stability of background fluorescence for >3 h of steady-state perfusion (Fig. 1 A).

High-throughput analysis of optical signals was performed using the CardioPlex software v.8.3.3 (RedShirt Imaging, Decatur, GA). The peak emitted TMRM fluorescence signal from each of 6400 pixels was measured before and after excitation achieved by a computer-automated filter shutter switch. TMRM background fluorescence was baseline corrected by subtracting fluorescence levels before dye staining (<0.1%) for each pixel. Background-corrected TMRM fluorescence (ΔΨ_m) during the IR protocol was then normalized to the value of steady-state TMRM fluorescence achieved during the dye washout phase for each of the 6400 individual pixels.

Normalized ΔΨ_m measurements during IR injury across the imaged 4 × 4 mm region of the heart were plotted as contour maps using Delta Graph 5.6 (Red Rock Software, Salt Lake City, UT). These maps served to illustrate the dynamic changes in the spatial distribution of ΔΨ_m.

IR injury in normal and hypertrophied hearts

IR injury promotes mitochondrial dysfunction and oxidative stress (13,20,21). Therefore, we used a protocol of global no-flow ischemia (for 7 min), followed by reperfusion in normal and LVH hearts. In preliminary experiments, we determined that this moderate protocol (i.e., 7 min of ischemia) produced sustained reperfusion arrhythmias in <50% of control hearts. Therefore, this protocol allowed us to determine whether LVH was associated with a heightened susceptibility to IR-mediated arrhythmias. ΔΨ_m recordings were performed in 10-s intervals throughout the first minute of reperfusion, when rapid changes were most likely to occur, and at 1-min intervals throughout the remaining time.

The propensity for spontaneous arrhythmias was scored according to the guideline described by Walker et al. (22) using the Lambeth Conventions. Briefly, the following criteria were used to generate an arrhythmia score (AS) for each unpaced control and LVH heart: 0 = <50 ventricular premature beats; 1 = 50–499 ventricular premature beats; 2 = >500 ventricular premature beats and/or one episode of spontaneously reverting VT/VF; 3 = more than one episode of spontaneously reverting VT/VF (<1 min total combined duration); 4 = 1–2 min of total combined VT/VF; and 5 = >2 min of VT/VF. AS was measured during ischemia and reperfusion in all normal and LVH hearts.

In addition, we used a series of control (n = 4) and LVH (n = 4) animals to investigate potential differences in the incidence of complex arrhythmias (VT/VF, AS ≥ 3) occurring in hearts that had undergone ex vivo steady-state pacing (4 Hz) during IR injury.

Results

Structural, mechanical, and electrophysiological remodeling in LVH

Four weeks of pressure overload led to marked changes in structural and functional parameters (Fig. 1 B and Table 1), consistent with concentric LVH. Shown in Fig. 1 B are representative echocardiograms (right) and hematoxylin and eosin (H&E)-stained histological cross sections (left) from representative sham-operated and LVH animals, indicating the presence of marked concentric hypertrophy in banded animals. As shown in Table 1, standard metrics of wall thickness measured in systole (interventricular septal width in systole (IVSs) and left ventricular posterior wall in systole (LVPWs)) and diastole (interventricular septal width in diastole (IVSd) and left ventricular posterior wall in diastole (LVPWd)) were significantly (p < 0.001) increased by 50–75% over baseline values. Fractional shortening (FS) increased by 29% (p < 0.001), whereas ESV decreased by 67% (p < 0.001). HR, SV, and cardiac output remained stable. Also consistent with hypertrophic remodeling, optical AP mapping in a subset of di-4-ANEPPS-stained control (n = 4) and LVH (n = 4) hearts revealed a significant increase in APD in LVH. Representative AP traces from control (black) and LVH (red) hearts during steady-state pacing are shown in Fig. 1 B (left).

Table 1.

Results of statistical analyses of structural and mechanical indices of echocardiographic measurements at baseline and after 4 weeks of LVH

	Preoperative		1 Month postoperative		% Change	Paired t-test
	Mean	SD	Mean	SD
HR (bpm)	427.29	55.90	395.44	36.36	−6.25	p = 0.0694
IVSD (cm)∗	0.17	0.02	0.29	0.04	75.95	p < 0.0001
IVSS (cm)∗	0.32	0.03	0.52	0.04	63.86	p < 0.0001
LVIDd (cm)	0.63	0.05	0.64	0.10	0.82	p = 0.82
LVIDs (cm)∗	0.26	0.05	0.16	0.08	−40.23	p < 0.0001
LVPWd (cm)∗	0.18	0.05	0.29	0.04	64.07	p < 0.0001
LVPWs (cm)∗	0.32	0.04	0.48	0.08	49.55	p < 0.0001
EDV (mL)	0.60	0.13	0.64	0.25	5.21	p = 0.39
ESV (mL)∗	0.05	0.03	0.02	0.02	−67.40	p < 0.0001
EF (%)∗	91.75	3.30	97.62	2.07	6.50	p < 0.0001
SV (mL)	0.54	0.11	0.62	0.25	12.45	p = 0.15
FS (%)∗	59.04	5.91	76.02	10.19	29.83	p = 0.0002
CO (mL/min)	427.29	55.90	395.44	36.36	−6.25	p = 0.64

LVIDd: left ventricular internal dimension in diastole; LVIDs: left ventricular internal dimension in systole.

^∗Indicates statistical significance p < 0.05.

Metabolic remodeling in LVH: paradoxical stability of ΔΨ_m during ischemia

As shown in Fig. 2 A, ischemia caused a rapid and sustained decline (depolarization) of normalized ΔΨ_m in control hearts. Within 1 min of ischemia, normalized ΔΨ_m was reduced by 26% compared to preischemic baseline levels. ΔΨ_m was further decreased to ∼50% of baseline by 4 min of ischemia and was sustained at that level throughout the remaining ischemic episode. Of interest, reperfusion led to a rapid and sustained repolarization of ΔΨ_m. In a subset of studies, we found that reperfusion after longer (15 min) episodes of ischemia could only transiently repolarize ΔΨ_m (not shown) (19). Since these hearts exhibited reperfusion VF in almost all cases, we chose a milder IR injury protocol to study potential differences between control and LVH hearts.

Figure 2.

(A) Average and SD of ΔΨ_m response to IR injury in normal and LVH hearts. Clearly, LVH prevents ischemia-induced depolarization of ΔΨ_m. (B) Average ΔΨ_m response during the entire episode of ischemia (over 7 min) and reperfusion in normal and LVH hearts. LVH is associated with a significantly more polarized ΔΨ_m during both ischemia and reperfusion.

In sharp contrast, LVH hearts were resistant to ischemia-induced ΔΨ_m depolarization (Fig. 2 A, black symbols). On average, ΔΨ_m was significantly increased (by 17%, p = 0.0082) rather than decreased during ischemia compared to baseline levels (Fig. 2 A). Furthermore, the paradoxical increase in ΔΨ_m observed during ischemia was sustained after reperfusion in LVH (Fig. 2 A). The marked differences in average ΔΨ_m between control and LVH hearts during ischemia and reperfusion are also quantified in Fig. 2 B. Clearly, average normalized ΔΨ_m was significantly (p < 0.001) greater during both ischemia and reperfusion compared to control hearts.

In addition to the average ΔΨ_m, we measured the spatiotemporal patterns of ΔΨ_m during ischemia and reperfusion. Shown in Fig. 3 are contour maps recorded at various time points during the course of ischemia (top row) and reperfusion (bottom row) in control (Fig. 3 A) and LVH (Fig. 3 B) hearts. It is evident that ischemia led to a rapid (within 1 min) depolarization of ΔΨ_m in the control heart. Of interest, these contour maps reveal the formation of an organized wavefront of ΔΨ_m depolarization that actively propagated during the course of ischemia (arrow) across the control heart. As mentioned above, reperfusion led to a rapid and sustained repolarization of ΔΨ_m across the entire mapping field. Although on average, ΔΨ_m failed to depolarize during ischemia in LVH, an examination of the spatiotemporal changes in ΔΨ_m revealed the presence of complex neighboring regions of increased (green) and decreased (red) ΔΨ_m. These regions of opposite ΔΨ_m polarity compared to baseline were relatively static and did not exhibit major fluctuations throughout the entire ischemic episode (Fig. 3 B). This was in contrast to control hearts, in which ischemia led to the active propagation of a wave of ΔΨ_m depolarization that progressively captured more myocardium (Fig. 3 A).

Figure 3.

Contour maps showing the spatiotemporal distribution of ΔΨ_m during the course of ischemia (top row) and reperfusion (bottom row) in representative normal (A) and LVH (B) hearts.

Spatial heterogeneity of ΔΨ_m in control and LVH hearts

Because LVH hearts were distinguished by the presence of spatially adjacent regions of increased and decreased ΔΨ_m, we systematically investigated the degree of ΔΨ_m heterogeneity in all control and LVH hearts during the course of IR injury. Shown in Fig. 4 are representative ΔΨ_m contour maps recorded after 7 min of ischemia, and select ΔΨ_m traces measured during the course of IR injury (sites A–H). Clearly, the control hearts exhibited distinct ΔΨ_m depolarization during ischemia (Fig. 4 A, sites A–H). In contrast, LVH was marked by heterogeneous ΔΨ_m responses with depolarizing and hyperpolarizing sites (Fig. 4 B). The extent of ΔΨ_m heterogeneity was quantified as the SD of normalized ΔΨ_m measured from 6400 pixels during IR injury in control (n = 6) and LVH (n = 11) hearts (Fig. 5). At all time points during the IR protocol, LVH was associated with increased ΔΨ_m heterogeneity compared to control hearts (Fig. 5 A). In fact, on average, ΔΨ_m heterogeneity was significantly greater during ischemia (by 80.8%, p < 0.0001) and reperfusion (144.9%, p < 0.001) in LVH compared to control hearts (Fig. 5 B).

Figure 4.

ΔΨ_m contour maps recorded after 7 min of ischemia in representative normal (A) and LVH (B) hearts. (Right) Multiple traces recorded from the same normal and LVH hearts during the course of IR injury.

Figure 5.

(A) Spatial heterogeneity indexed by the SD of ΔΨ_m in normal and LVH hearts. (B) Average and SD of ΔΨ_m heterogeneity measured during ischemia and reperfusion in normal and LVH hearts.

Enhanced susceptibility of hypertrophied hearts to reperfusion-mediated arrhythmias

We then investigated potential differences in the propensity of control and LVH hearts to IR-mediated arrhythmias. We first calculated the AS values from the continuous ECG recordings performed in all hearts. Shown in Fig. 6 are average AS values during ischemia and reperfusion of control and LVH hearts (top) and representative ECG traces indicating various levels of arrhythmia complexity, and reflecting the range of AS values from normal rhythm (AS = 0) to sustained VT/VF (AS = 5) (bottom). Of note, the average AS values were comparable during ischemia in both unpaced control and LVH hearts (p = NS). Reperfusion, on the other hand, caused a significant increase in AS in LVH (p < 0.001) but not control (p = 0.082) hearts.

Figure 6.

AS as an index of spontaneous arrhythmias measured during ischemia and reperfusion from all unpaced normal and LVH hearts. Representative ECG traces indicate progressively more complex arrhythmias, ranging from AS 0 to 5. AS: arrhythmia score.

An additional series of eight hearts (four control and four LVH) were used to investigate potential differences in the incidence of complex arrhythmias during ex vivo pacing (at 4 Hz) and IR injury. Shown in Fig. 7 are representative volume-conducted ECG traces from control (left) and LVH (right) hearts. Of interest, steady-state pacing during IR injury revealed the presence of arrhythmias in LVH (2/4) but not (0/4) control hearts (Fig. 7). Finally, 7 min of ischemia resulted in rapid shortening of APD₉₀ in LVH but not control hearts (Fig. 7, bottom).

Figure 7.

Incidence of VT/VF in paced hearts during IR injury. Representative volume-conducted ECG traces measured at baseline, during early and late ischemia, and after reperfusion in control (left) and LVH (right) hearts. VT/VF was observed exclusively in paced LVH hearts. Representative AP traces from control and LVH hearts demonstrate rapid shortening of APD in LVH but not control hearts in response to 7 min of ischemia (bottom).

Discussion

Here, we demonstrate the feasibility of measuring the spatiotemporal dynamics of ΔΨ_m across the intact organ during IR injury. The major findings of this study are as follows: 1) Ischemia generated a wave of active ΔΨ_m depolarization across control but not hypertrophied hearts. 2) Complex areas of ΔΨ_m depolarization and hyperpolarization coexisted across the hypertrophied heart during the entire course of ischemia. 3) The spatial heterogeneity of ΔΨ_m was significantly greater in LVH than in control hearts during both ischemia and reperfusion. 4) Hypertrophied hearts were more sensitive to IR-mediated arrhythmias compared to control hearts.

Disrupted ΔΨ_m depolarization in LVH during IR injury

One of the main contributions of this study is the identification of spatiotemporal changes in ΔΨ_m at a subcellular resolution within the intact organ. We found that ΔΨ_m, measured in native myocardium, underwent marked changes during IR injury. Our finding of ΔΨ_m depolarization over a 4 × 4 mm area during ischemia is consistent with the elegant findings of Matsumoto-Ida et al. (23), who used two-photon microscopy to image a 50 μm region. More recently, Slodzinski et al. (24) showed heterogeneous ΔΨ_m depolarization within the intact guinea pig heart that was associated with increased reactive oxygen species (ROS) production and depletion of intracellular glutathione levels. Remarkably, our findings indicate that reperfusion after 7 min of global no-flow ischemia in the rat results in complete and sustained recovery of ΔΨ_m after reperfusion.

Another major finding of this work is that LVH paradoxically disrupts progressive ΔΨ_m depolarization observed during the course of ischemia in the control heart. Our findings regarding disrupted ΔΨ_m polarization in LVH are consistent with earlier reports indicating that hypertrophy is associated with marked baseline alterations of ΔΨ_m at the cellular level. Previous measurements of ΔΨ_m in hypertrophy have yielded disparate results. Notably, Sharma et al. (25) demonstrated that LVH in guinea pigs is associated with a basal reduction of ΔΨ_m. In contrast, Nagendran and colleagues (26) reported an increase of ΔΨ_m in phenylephrine-treated rat neonatal cardiomyocytes, adult rat hearts from animals with monocrotaline induced pulmonary hypertension, and human tissue samples from patients with right ventricular hypertrophy undergoing cardiac surgery (26). Both groups compared the total fluorescence of mitochondrial dyes (TMRM or JC1) during steady state, whereas our approach allowed the measurement of dynamic spatiotemporal changes in ΔΨ_m relative to baseline during IR injury. It is worth noting that in the study by Sharma et al. (25), animals were examined at a more advanced stage of the disease (10 weeks after aortic banding), which reflected the transition point from cardiac hypertrophy to heart failure.

Nagendran and colleagues (26) linked the increase in ΔΨ_m to changes in pyruvate metabolism, whereas Sharma et al. (25) emphasized the activation of apoptotic pathways through ROS. A major consequence of ROS accumulation involves alterations in the function of mitochondrial ion channels, such as the permeability transition pore (PTP) and the inner membrane mitochondrial ion channel (IMAC). Aon and colleagues (27) elegantly showed that the glutathione redox couple (GSH/GSSG) of the cardiac cell governs the sequential opening of these channels, with reversible IMAC activation occuring before irreversible PTP opening. Of interest, PTP opening during calcium stress is more likely in hypertrophied compared to control hearts (28). Others have described changes in ion fluxes that may protect against the loss of ΔΨ_m. The activity of the newly discovered mitochondrial calcium channels (mCa1 and mCa2) was shown to be significantly reduced in end-stage human heart failure, resulting in lower mitochondrial calcium influx and stabilized ΔΨ_m (29). Moreover, the development of hypertrophy is associated with marked changes in the expression of uncoupling proteins, which are known to regulate ΔΨ_m and ROS generation. In fact, our finding of ΔΨ_m hyperpolarization in LVH is consistent with the down-regulation of uncoupling proteins that has been observed in animal models and humans with hypertrophy (30).

In this study, we demonstrate how mitochondria within their native environment can react differentially to the same acute stressor (IR injury) in chronically diseased compared to control hearts. The altered polarization patterns of ΔΨ_m in hypertrophy could represent an adaptive mechanism with beneficial or deleterious physiological consequences. Both elevations and depressions of ΔΨ_m can lead to increased ROS production (31). An increase in ΔΨ_m, for example, may provide a protective mechanism against apoptosis, a process that is considered to be central in the transition from compensated hypertrophy to end-stage heart failure. On the other hand, increased ΔΨ_m may also interfere with mitochondrial substrate transport and respiration, and thus may underlie some of the adverse metabolic remodeling observed in hypertrophy.

Increased ΔΨ_m heterogeneity in LVH

In addition to allowing the measurement of average changes in ΔΨ_m, our technique is also well suited for detailed investigations of the spatiotemporal dynamics of ΔΨ_m across a relatively large (4 × 4 mm) region of the epicardial surface. We observed a spatially complex distribution of polarized and depolarized regions of ΔΨ_m in hypertrophied hearts that were not present in control hearts during ischemia.

Since the mitochondrial function of individual cells is highly influenced by network properties, it is critical to investigate subcellular mitochondrial function within the milieu of the intact heart. Specifically, our findings indicate a heterogeneous metabolic substrate across the heart during IR injury. These heterogeneities may ultimately modulate myocardial excitability and contribute to the formation of zones of conduction block by heterogeneous activation of surface ATP-sensitive potassium channels (11,32). As such, these metabolic heterogeneities are consistent with the notion of metabolic sink/block recently suggested by Akar and colleagues (11).

Heightened susceptibility of LVH hearts to reperfusion-mediated arrhythmias

Sustained arrhythmias caused by oxidative stress associated with IR injury are known to occur during early reperfusion. Indeed, we found that spontaneously arising arrhythmic events (quantified by an AS index) were significantly greater during reperfusion compared to ischemia in LVH but not control hearts, implicating oxidative stress in the mechanism of these arrhythmias. Moreover, our finding that complex arrhythmias (VT/VF) in paced hearts were exclusively present in LVH but not control hearts undergoing IR injury further highlights the susceptibility of LVH hearts to arrhythmias, even when HR is controlled. Although hypertrophy is characterized by complex structural, mechanical, and electrophysiological alterations that render the heart more vulnerable to arrhythmias (15), our findings highlight a potentially critical role for metabolic remodeling in general, and altered ΔΨ_m properties in particular, in the mechanism of these arrhythmias.

Finally, we previously showed that partial blockade of IMAC, which in isolated myocytes can stabilize ΔΨ_m, is an effective tool for preventing reperfusion-mediated VT/VF in guinea pigs (11). Our current findings indicate that increased ΔΨ_m during ischemia in LVH promotes rather than prevents arrhythmias upon reperfusion. These apparently contradictory findings suggest that there might indeed be a delicate balance of ΔΨ_m changes that can either promote or prevent arrhythmias. Of interest, the highest concentration of IMAC blocker (4′-Cl-DZP) in our previous study (100 μM), which effectively prevented the formation of inexcitability (electrical silence) throughout the entire ischemic period, was not effective in suppressing reperfusion-mediated arrhythmias, in contrast to lower concentrations (64 μM) (11). This is consistent with our present findings that complete abrogation of ΔΨ_m depolarization during metabolic stress associated with ischemic injury is indeed pro-arrhythmic.

Limitations

Ischemic injury results in highly complex electrophysiological changes that ultimately render the heart prone to arrhythmias. In this work, we focused on a key metric of mitochondrial function (ΔΨ_m) that drives the electron transport chain by setting the electromotive force. In humans and animal models, IR-related arrhythmias are dependent on multifactorial mechanisms such as changes in ion gradients (hyperkalemia, sodium and calcium overload), acidosis, and gap junction uncoupling. Therefore, one should not expect that a single metric of cardiac function will fully predict arrhythmias in this setting. However, our present findings strongly implicate mitochondrial dysfunction in general, and disrupted ΔΨ_m kinetics during ischemia in particular, as an important mechanism of arrhythmias in LVH.