Numerous studies have supported the concept that the mitochondria induce nonphysiological spatiotemporal heterogeneity in the cardiac action potential and predispose the heart to reentrant dysrhythmia. The influence of mitochondrial energetic status on the sarcolemmal action potential is mediated in large part by energy-sensing ATP-sensitive potassium channels (sarcK_ATP) in the sarcolemmal membrane. A significant amount of attention has been devoted to the role that sarcK_ATP channels may play in inducing action potential heterogeneity, leading to cardiac dysrhythmias [7, 8]. First discovered by Noma and coworkers in the early 1980s [9], myocardial sarcK_ATP channels are heteromultimers composed of four pore-forming subunits and four accessory subunits, the sulfonylurea receptors, that bind to ATP. Inhibited by intracellular ATP and activated by ADP, Pi, Mg, and/or pH, sarcK_ATP channels open in the face of oxidative stress to produce an inwardly rectifying background current, usually observed within the first 10 min of ischemia [10]. SarcK_ATP channels are among the most densely populated ion channels in cardiac myocardium, and the opening of even 1% of the total amount of channels in the sarcolemma can significantly shorten the cardiac action potential [11].

The opening of sarcK_ATP channels may be an endogenous protective mechanism of the myocardial tissue, where channel opening signaled by inadequate ATP supply decreases calcium-mediated cardiac energy demand. As the population of sarcK_ATP opens, the cardiac action potential shortens and reduces the calcium transient. Since calcium overload can lead to necrotic and apoptotic cell death, sarcK_ATP channel opening is believed to be cytoprotective by decreasing the extent contracture by the myofilaments and blunting mitochondrial calcium overload. Several lines of evidence indicate that the expression of functional sarcK_ATP channels is vital to cellular survival in the face of oxidative stress. First, increased sarcK_ATP protein expression is correlated with protection against ischemia/reperfusion injury in female (vs. male) animals [12–14] or following exercise training [12, 15]. Second, pharmacological blockade of the sarcK_ATP channel population increased cell death in hearts exposed to ischemia/reperfusion [13, 15, 16], with the block during ischemia being the critical period leading to increased injury [13]. Third, gene knockout of the sarcK_ATP channel pore-forming subunit produced animals that were severely intolerant to exercise and displayed enhanced sensitivity to calcium overload [17–19]. Taken all together, it appears that there is a physiological role for sarcK_ATP opening in attenuating cell death during ischemia. Supporting this concept are observations in humans where diabetic patients taking oral sulfonylureas to control type II diabetes were at a higher disposition for cardiac injury following ischemia [20].

While the opening of sarcK_ATP channels appears to be protective of the viability of ischemic cardiac myocytes, the consequence of increasing potassium conductance to the whole organ predisposes to electrical dysfunction and in some cases the generation of fatal dysrhythmia [7, 21–23]. With such a high density of channels in the sarcolemmal membrane, the opening of sarcK_ATP channels can significantly shorten the action potential and, if enough channels open, can render cells inexcitable by holding the membrane potential very close to potassium’s Nernst equilibrium potential. This creates a significant current sink for the propagating depolarization wave, and dysrhythmias may be favored when there are local regions where the open probability of sarcK_ATP channels is high, i.e., where the energetic status of the cell has been compromised, a phenomenon that has been termed “metabolic sinks” [24, 25]. Presence of metabolic sinks potentially enhances propensity for dysrhythmia by influencing the effective refractory period (ERP) of the myocardium, resulting in a shortened excitation wavelength. Pathological heterogeneity in action potential duration increases the dispersion of refractoriness within the tissue and is known to promote reentry [26]. SarcK_ATP opening abbreviates the action potential duration and shortens ERP. SarcK_ATP channel openers [27] and blockers [28] decrease and increase ERP, respectively. ERP is also prolonged after knockout of sarcK_ATP pore-forming subunit suggesting that these features are intrinsic to the myocardium and that K_ATP channels in the myocardium may have an important role in protecting the heart from lethal dysrhythmias and adaptation to stress situations [29]. However, other factors may come into play during ischemia which alters the relationship between action potential duration and ERP. For example, although ischemia activates sarcK_ATP and shortens the action potential, a prolonged ERP may occur due to postrepolarization refractoriness [30], presumably owing to alterations in Na channel availability.

A dysrhythmogenic role for sarcK_ATP has been confirmed in studies using either glibenclamide which blocks both the mitochondrial and sarcolemmal isoforms of the K_ATP channels, or the sarcolemmal-specific HMR 1833 compounds, or HMR 1098, the sodium salt of HMR 1883. Blocking sarcK_ATP channels with HMR1883 decreased the incidence of ventricular dysrhythmia in rat, pig, and dog [22, 31, 32]. These findings from animal models are confirmed in clinical studies where sarcK_ATP channel blockers have been shown to reduce the incidence of ventricular fibrillation in humans [33, 34]. While it is conceivable that the prevention of dysrhythmias with sarcK_ATP blockers is due to direct inhibition of sarcK_ATP currents, sarcK_ATP blockers could theoretically also prevent dysrhythmias by indirect mechanisms. Specifically, by inhibiting sarcK_ATP currents and preventing action potential shortening, the ensuing cellular calcium overload may promote gap junction closure, thereby blocking reentrant wave fronts via cellular uncoupling [35].

The cardiac mitochondrial network produces over 95% cellular ATP and accounts for 20–30% of myocardial volume in many species ranging from mouse to man [36]. According to classic chemiosmotic theory [37], mitochondria create a proton motive force by pumping protons out of the mitochondrial matrix and then use this proton electrochemical gradient to liberate the energy needed to phosphorylate ADP to ATP by the F₁F_o-ATPase. The majority of the proton motive force is comprised of the mitochondrial membrane potential (Δψ_m), with the magnitude of Δψ_m being around150 mV in energized mitochondria. Decreases in Δψ_m diminish the amount of free energy available to generate ATP, with mitochondria shifting to ATP hydrolysis under pathophysiological conditions when the Δψ_m collapses substantially (Fig. 19.1).

The dynamic relationship between K_ATP current and the metabolic status of heart cells was first reported by O’Rourke and coworkers [38]. Inducing metabolic stress via substrate deprivation or in response to increased ADP levels, glibenclamide-sensitive current oscillations were observed in cardiomyocytes. Oscillating sarcK_ATP currents were observed in phase with NADH fluctuations and were not influenced by changing cytosolic calcium concentrations. The vacillating sarcK_ATP currents directly influenced cardiac repolarization and introduced significant lability in the length of the action potential waveform [38]. The energy-driven oscillations in potassium currents produced cyclical changes in the cardiac action potential and thus may contribute to the genesis of dysrhythmias during metabolic compromise. Subsequent studies confirmed the initial observation of oscillatory sarcK_ATP currents and action potential duration in cardiac myocytes under conditions of metabolic stress [39, 40]. The fluctuations in sarcK_ATP currents and consequently action potential duration are linked to the behavior of the mitochondria. Collapses in Δψ_m have been observed in a few studies where the myocardium is subjected to oxidative stress and with sarcK_ATP current increasing in phase with losses of Δψ_m [39].

Using cationic lipophilic rhodamine fluorescent probes, several studies have noted reversible collapses in Δψ_m in isolated cells subjected to oxidative stress via substrate deprivation [39, 41], ATP depletion [40], generation of reactive oxygen species (ROS) [39], the thiol oxidant diamide [42], as well as inhibition of respiration [40].

Recent evidence using two-photon microscopy has confirmed cellular data as reversible collapses in Δψ_m were seen in intact hearts exposed to global ischemia/reperfusion or diamide [43]. In addition to nucleotide-dependent activation of sarcK_ATP currents after loss of Δψ_m, the collapse of bioenergetics might also activate sarcK_ATP currents through mechanical stretch. In this scenario, the loss of mitochondrial function would then quickly preclude development of tension and result in paradoxical segment lengthening of the ischemic ventricular tissue. Given that both ischemia and stretch activate sarcK_ATP channels [44, 45], bulging of the myocardium may also contribute to the activation of sarcK_ATP channels.

Several distinct energy-dissipating ion channels in the inner membrane have been proposed to be involved in the Δψ_m collapse, thereby contributing to the generation of dysrhythmia. The first of these channels is the inner membrane anion channel (IMAC).

Anion flux across the inner mitochondrial membrane was first observed 50 years ago [63]. Earlier studies have focused primarily on the contribution of anion movement on mitochondrial volume regulation. Since the first observations, the IMAC has been characterized in several biologic tissues and is believed to play an important role in anion efflux from energized mitochondria [64]. Although the exact structure of the IMAC is not known, the sensitivity of the anion channel to regulation by benzodiazepine ­compounds [65] suggests that the molecular composition consists of an anion channel subunit that associates with a peripheral benzodiazepine receptor in the outer membrane.

Insights into the factors mediating the collapse in Δψ_m have focused on the production of ROS by the mitochondria. ROS-dependent oscillations in Δψ_m were first noted by Sollott and coworkers [66]. These investigators noted that local generation of ROS produced by laser flash elicited synchronous collapses in Δψ_m that were prevented by a ROS scavenger. There is growing evidence that the collapse in Δψ_m may be mediated by superoxide anion, leading to depolarizations in the myocardium through a process termed “ROS-induced ROS release” [66, 67]. According to this ­theory, ROS produced at the level of a single mitochondrion can stimulate superoxide-mediated depolarization of neighboring mitochondria. This spatiotemporal behavior among the mitochondrial network led others to conclude that ­mitochondria are arranged in a percolation matrix. According to empirical data, the increase in ROS under conditions of oxidative stress can reach a critical level, after which cell-wide Δψ_m oscillations in the mitochondrial network are observed [68].

The importance of IMAC in influencing the Δψ_m was first noted when several distinct ligands to IMAC were found to prevent loss of Δψ_m observed in isolated cardiac myocytes. Aon et al. used a laser flash to induce a local burst of mitochondrial ROS, which causes cell-wide increases in ROS production and oscillations in Δψ_m [39]. The reversible collapses in Δψ_m, and the cell-wide ROS accumulation, could be prevented with the addition of PK11195, 4′-chlorodiazepam, or DIDS, three distinct compounds that have previously been shown to block the activity of IMAC [69]. Furthermore, blocking the reversible collapses in Δψ_m by targeting the IMAC terminated the oscillations in action potential duration [39], providing further cellular evidence that targeting the IMAC may be effective in preventing dysrhythmias by stopping ROS-induced ROS release.

A series of studies whereby inhibiting the IMAC prevented dysrhythmias in intact mammalian hearts further confirms the involvement of IMAC [24, 62]. To examine whether diamide, a glutathione oxidant, caused mitochondrial depolarization and promoted dysrhythmias in normoxic isolated perfused guinea pig hearts and to investigate whether stabilization of Δψ_m with a ligand of the mitochondrial benzodiazepine receptor (4′-chlorodiazepam; 4-ClDzp) prevented the formation of metabolic sinks and consequently precluded dysrhythmias, oxidation of the glutathione (GSH) pool was initiated by treatment with 200 μM diamide for 35 min followed by washout. This treatment increased oxidized (GSSG) and decreased both total GSH and the GSH/GSSG ratio. All hearts receiving diamide transitioned from sinus rhythm into ventricular tachycardia during the diamide exposure. These dysrhythmias and impaired LV function were significantly inhibited by coadministration of 4-ClDzp (Fig. 19.2) [70].

Optical mapping of the epicardial surface of guinea pig hearts revealed that blocking IMAC decreased ischemia-induced action potential shortening and was accompanied by a lack of a development of ventricular dysrhythmias at the onset of reperfusion [24]. Cardioprotection evoked by blocking the IMAC was also observed in isolated rabbit heart and was accompanied by significantly improved left ventricular developed pressure 74. In both studies the reperfusion dysrhythmias were prevented when the IMAC was blocked only at the onset of reperfusion as opposed to as a pretreatment [24, 62].

Evidence for a mitochondrial ATP-sensitive potassium (mitoK_ATP) channel was first observed in the rat liver mitochondria [61, 71] and later seen in the heart as well [72]. The opening of mitoK_ATP channels may mediate the protective interventions administered before the onset of ischemia by partially dissipating the Δψ_m, reducing the driving force for calcium into the mitochondria, and while cellular respiration secondary to mild swelling of the matrix [8, 73, 74].

Most studies that have examined the potential cardioprotective effect of mitoK_ATP opening have focused on examining the role of mitoK_ATP in reducing infarct size elicited by a single preconditioning stimulus [73, 74]. In most but not necessarily all of these studies, blocking the mitoK_ATP with 5-hydroxydecanoate (5-HD) abolished the benefit of reduction in infarct size. While single episodes of preconditioning may provide mechanistic insight, when repetitive stimuli are administered, mitoK_ATP blockade does not reduce the evoked protection, as evidenced by the lack of effect of 5-HD in abolishing the infarct-reducing effects of repetitive ischemic preconditioning [75] or of chronic exercise [15]. On the other hand, few studies have examined the activity of mitoK_ATP channels in cardiac dysrhythmia. A role for mitoK_ATP in protecting against dysrhythmia is apparent where mitoK_ATP blockers abolished the antidysrhythmic phenotype provided by a preconditioning stimulus such as ischemic preconditioning [76], adenosine [77], δ-opioid agonists [78], estrogen [79], 3-nitropropionic acid [80], nitroglycerin [81], norepinephrine [82], or the endothelin receptor agonists. Although mitoK_ATP channels appear to be important in mediating the antiarrhythmic effects in some preconditioning models, their activity cannot be attributed to all models of preconditioning. For instance, blocking the mitoK_ATP during preconditioning from bradykinin [83], low-flow ischemia [83], peroxynitrite [84], or estradiol [85] failed to attenuate the antiarrhythmic protection. Protection against dysrhythmias via direct activation of mitoK_ATP channels prior to ischemia has also yielded opposing results, with some investigators showing protection from dysrhythmia [77, 86] while others showing no beneficial effect at all [75]. One putative explanation for the discordant findings is that the pharmacological agents used to open mitoK_ATP were different between these studies (minoxidil, diazoxide, and/or BMS-191095), and some of these compounds may be limited by their nonspecific actions.

While literature on preconditioning provides useful mechanistic insight regarding antidysrhythmic strategies administered before index ischemia, the clinical relevance of these strategies remain to be determined. To the clinician, attenuation of dysrhythmias must often be attempted after the onset of ischemia. Targeting the mitoK_ATP channels after the onset of metabolic stress seemed promising based on cellular experiments, where administration of mitoK_ATP terminated oscillations in Δψ_m evoked by halting respiration [40], and mitoK_ATP opening with diazoxide improved cellular survival and mitochondrial integrity during cellular reoxygenation [87]. Despite these encouraging cellular data, postischemic administration of mitoK_ATP openers does not decrease dysrhythmias [86], and effects of postconditioning interventions have been shown to be independent of the activity of mitoK_ATP channels [61]. Indeed, the same investigators that observed beneficial effects of diazoxide on isolated cells [87] found that the cytoprotective properties of the drug were independent of mitochondrial potassium flux [88]. The nonspecificity of commonly used mitoK_ATP openers, such as diazoxide and blockers, such as 5-HD is well known and has been reviewed previously [15, 74, 89, 90].

The most recent study of mitoK_ATP channel inhibition has involved the ischemia-exercised heart [91]. Male Sprague-Dawley rats were randomly assigned to cardioprotective treadmill exercise or sedentary conditions before ischemia/reperfusion (ischemia 20 min, reperfusion 30 min) in vivo. Subsets of exercised animals received pharmacological inhibitors for mitoK_ATP (5-hydroxydecanoate) or sarc K_ATP (HMR1098) before ischemia/reperfusion. Analysis of digital ECG tracings revealed that mitoK_ATP inhibition blunted the antiarrhythmic effects of exercise, while sarc K_ATP inhibition did not (Fig. 19.2). Endogenous antioxidant enzyme activities for total, CuZn, and Mn superoxide dismutase, catalase, and glutathione peroxidase from ischemic and perfused ventricular tissue were not mitigated by IR, although oxidative stress was elevated in sedentary and mitoK_ATP-inhibited hearts from exercised animals. These findings suggest that the mitoK_ATP channel provides antidysrhythmic protection as part of exercise-mediated cardioprotection against ischemia/reperfusion, and the observed antidysrhythmic protection may be associated with preservation of redox balance in exercised hearts.

The redox status of heart cells directly influences the cellular excitability. An oxidative shift in the cellular redox potential can promote action potential heterogeneity by modulating several different ion channels. Increased oxidation has been shown to directly activate sarcK_ATP channels [105], alter the inactivation kinetics of L-type calcium channels via increased calcium leak from the ryanodine receptor [106], and influence the state of channels on the mitochondrial inner membrane.

Bursts of ROS are observed within the first few minutes of reperfusion, when the propensity for dysrhythmia is extremely high [107]. In anesthetized rat, the prior administration of folic acid solution, amflutizole, superoxide dismutase, catalase, and superoxide dismutase plus catalase reduced the incidence of reperfusion-induced dysrhythmias and resultant mortality, caused by reperfusion after a transient period of coronary artery occlusion. Prior administration of soybean trypsin inhibitor significantly reduced mortality. In an isolated, perfused rat heart preparation with temporary coronary artery occlusion, addition of xanthine oxidase–hypoxanthine to the perfusion medium increased the incidence of reperfusion dysrhythmias and decreased the total duration of sinus rhythm during reperfusion. Further addition of superoxide dismutase or l-methionine increased significantly the total duration of sinus rhythm, suggesting that in the rat heart, xanthine oxidase may be involved in the genesis of reperfusion-induced dysrhythmias [108]. Several experiments have induced ventricular dysrhythmias under normoxic conditions with delivery of ROS bursts [109], and attempts to scavenge ROS with superoxide dismutase mimetics [110] or mitochondrial-targeted antioxidant peptides [111] were successful in decreasing the incidence of dysrhythmia. Future experiments that optimize effective delivery of ROS-scavenging agents to mitochondria clearly have significant potential in abrogating electrical dysfunction.