Nitric Oxide/Reactive Oxygen Species Generation and Nitroso/Redox Imbalance in Heart Failure: From Molecular Mechanisms to Therapeutic Implications

A. Sources and species of ROS

Growing experimental evidence documents that oxidative stress mediated by ROS plays a key role in the pathogenesis of HF (15, 63, 74, 102). ROS include free radicals, such as superoxide (O₂ ^•−) and hydroxyl (^·OH), and nonradical species such as hydrogen peroxide (H₂O₂). ROS generation occurs because the complete reduction of oxygen (O₂) into two molecules of H₂O requires four electrons and O₂ accepts only one electron at a time. The addition of the first electron yields O₂ ^•− that becomes H₂O₂ when one additional electron is added. ^·OH could arise from electron exchange between O₂ ^•− and H₂O₂ via the Haber–Weiss reaction. In addition, ^·OH is also generated by the reduction of H₂O₂ in the presence of endogenous iron by means of the Fenton reaction. The cellular sources of ROS generation within the heart include cardiac myocytes, endothelial cells, vascular smooth muscles, fibroblasts, and infiltrating inflammatory cells of different types. In these cells ROS can be generated by mitochondria, NADPH oxidases, xanthine oxidase (XO), and uncoupled NO synthases (NOSs) (Fig. 2). Increasing evidence supports the view that NADPH oxidases play important roles as initiators and integrators of redox signaling via cross-talk with other ROS-producing systems (53). For example, exposure of endothelial cells to oscillatory shear stress leads to an NADPH oxidase-dependent activation of XO (252), whereas angiotensin II (AT-II) stimulation results in mitochondrial ROS production that is downstream of Nox activation (106). Endothelial (e)NOS uncoupling has also been shown to be a direct result of NADPH oxidase activation (212), and, in a recent review, a cross-talk between mitochondrial and NADPH oxidase-derived ROS is discussed (93). Amplification of ROS production may thus occur and may be important in affecting different signaling pathways. Such a novel aspect may have significant implication for understanding disease mechanisms and for designing rational therapeutic interventions aimed at restoring and resetting redox homeostasis.

OPEN IN VIEWER

FIG. 2.

Cross-talk between enzymatic sources of ROS within the heart. NADPH oxidases (Nox2 is represented in the figure) play important roles as initiators and integrators of redox signaling because they are capable of modulating ROS production by other enzymes. For example, xanthine dehydrogenase is converted to xanthine oxidase, generating O₂ ^•− through thiol oxidation; therefore, depletion of BH4 can result in the functional uncoupling of NOS. Uncoupled NOS generates more ROS and less NO, shifting the nitroso-redox balance and affecting adversely the cardiovascular system. BH4, tetrahydrobiopterin; NO, nitric oxide; NOS, nitric oxide synthase; O₂ ^•−, superoxide; XDH, xanthine dehydrogenase; XO, xanthine oxidase; XOR, xanthine oxidoreductase.

1. Mitochondria

Cardiac myocytes have the highest density of mitochondria compared to most other cells to meet the demand for synthesis of ATP by oxidative metabolism (376). In cardiac myocytes, a large amount of ROS stems as a relevant by-product of electron flow through the respiratory chain. By using electron spin resonance spectroscopy, Ide et al. (173, 174) directly demonstrated that the inhibition of electron transport at the sites of complex I and III resulted in an enhanced generation of ROS in mitochondria from the failing myocardium. Electrons from complex I and III can convert O₂ to O₂ ^•−, which is then converted to H₂O₂ by spontaneous or enzymatic dismutation mediated by mitochondrial isoenzyme of superoxide dismutase. H₂O₂ can further react to form highly reactive ^·OH radicals (Fig. 3). The generation of ^·OH implies a pathophysiological significance of ROS in HF because ^·OH radicals are the predominant oxidant species causing cellular injury (44). Further, the high reactivity of the ^·OH and its extremely short physiological half-life of 10⁻⁹ s restrict its damage to a small radius from its origin since it is too short-lived to diffuse a considerable distance. Mitochondrial ROS generation is potentially significant in the setting of insulin resistance (279) and ischemia–reperfusion (140). In the latter context mitochondrial ROS production appears to be a dual effector responsible for both ischemia–reperfusion injury and cardioprotection (89). Ischemia causes immediate disturbance of mitochondrial function, including failure of ATP synthesis and drop in membrane potential that is accompanied by changes in cytosolic composition, like increased Ca²⁺, phosphate and fatty acids. This altered state is met during reperfusion by a large increase in ROS originating from the respiratory chain (111, 377). These factors promote opening of the mitochondrial permeability transition (MPT), a high-conductance pore in the inner mitochondrial membrane, which is the main cause of necrotic cell death in ischemia–reperfusion injury (103, 156, 391). It follows that any effort to protect the heart from these consequences must ultimately involve the prevention of MPT opening (391). The heart possesses self-defensive mechanisms able to reduce cell death and functional impairment after prolonged episodes of ischemia–reperfusion. Cardioprotective procedures include ischemic preconditioning, in which one or more periods of brief ischemia precede the index ischemia (270). Ischemic preconditioning protocol requires transient opening of the mitochondrial ATP-regulated potassium channel (mitoKATP) that is believed to be essential cardioprotective signal transduction against ischemia–reperfusion injury (Fig. 3) (89). In fact, mitoKATP opening induces a moderate increase in ROS production from Complex I; ROS produced by mitoKATP activity diffuses and inhibits MPT, thus reducing cell necrosis and infarct size (135).

OPEN IN VIEWER

FIG. 3.

Source of ROS and NO production and their implications in the mitochondrion. O₂ ^•−, generated at complex I and III, is converted to H₂O₂ by spontaneous or enzymatic dismutation mediated by superoxide dismutase (mtMn-SOD). H₂O₂, also generated by monoamino oxidase that catabolizes serotonin and NE, can further form reactive hydroxyl radicals through Haber-Weiss and, in the presence of Fe²⁺ or Cu²⁺, Fenton reactions. Excessive mitochondrial ROS generation may lead to mitochondrial DNA damage and, of consequence, to mitochondrial functional decline with further increase of ROS production and oxidative damage. ^•NO production by nitric oxide synthase (mtNOS) can reversibly inhibit cytochrome-c oxidase (complex IV), or react with O₂ ^•− to generate peroxynitrite (ONOO−), which cause irreversible injury to the mitochondria. On the contrary, in the ischemic heart, mitoKATP channel opening triggers a moderate increase of ROS production from complex I that are involved in cardioprotective signaling pathways inhibiting the mitochondrial permeability transition, and thus necrosis (see also Fig. 5A). H₂O₂, hydrogen peroxide; MAO, monoamine oxidase; ^•OH, hydroxyl radicals; OH⁻, hydroxyl anion.

ROS are also produced within mitochondria at sites other than the inner mitochondrial membrane, for example, by monoamine oxidase (MAOs) activity (104, 258) (Fig. 3). MAOs are flavoenzymes located within the outer mitochondrial membrane, which catalyze oxidative deamination of catecholamines and biogenic amines such as serotonin. During this process, they generate H₂O₂ and thus can potentially be a source of oxidative stress in the heart, particularly under stress conditions. MAOs exist in two isoforms, MAO-A and -B, with a distinct substrate and inhibitor sensitivity (113). MAO-A, in particular, principally catabolizes serotonin, norepinephrine, and epinephrine and is inhibited by low concentrations of clorgyline. All of these neurotransmitters have major functional implications in the heart, especially in the modulation of cardiac inotropy. In contrast, MAO-B has a higher affinity for phenylethylamine and benzylamine, and is inhibited by selegiline (348, 411). Yet, although the role of MAOs in terminating neurotransmitter signaling in the brain is well established, little is known about its modulation of cardiac morphology and function. A tight regulation of serotonin to maintain normal cardiovascular activity has been demonstrated in different experimental models (148, 255). Pharmacological agents acting through serotonin-related pathways have been associated with a number of significant cardiovascular adverse effects, including neonatal death (20). Receptor-independent effects of serotonin, have been described recently. H₂O₂ produced by MAO-A dependent oxidative deamination of serotonin exerts hypertophic effects or aggravate ischemia/reperfusion injuries (41). A recent study of Kaludercic et al. (187) showed that in addition to serotonin, norepinephrine catabolism by MAO-A plays a prominent role in hypertrophy in vitro and in its progression toward HF in vivo through enhanced H₂O₂ production. Therefore, owing to metabolism of these amine, MAO-A activity may represent a check-point for the sympathetic/serotoninergic system but also a local producer of ROS, thus ascribing a role for MAO-A dependent catalysis in cell signaling (305). In cardiomyocytes catecholamines and serotonin also play a metabolic role; in fact, they show insulin-like effects by increasing glucose uptake with different mechanisms (129, 199). In this context, a modulated increase MAO-A activity might be of some beneficial on the metabolic de-arrangement of the failing heart (see below). To date, no evidence exists on the role of MAO in human HF. Preliminary results from our laboratory showed that in human end-stage failing hearts secondary to ischemic and dilated cardiomyopathies both MAO-A and MAO-B isoforms were present with total MAO activity (A+B) 10 times higher in the ischemic than in nonischemic cardiomyopathy. Moreover, activity of the two isoforms appeared different in the two ventricles with respect of the cardiomyopathy. Our results indicate that the evaluation of MAO activity, in particular MAO-A, might help to discriminate between nonischemic and ischemic cardiomyopathy associated with HF.

Although recently the question has become controversial, several earlier studies suggested the presence of NOS in mitochondria (mtNOS) (210). Local NO generates can decrease respiration and potentially trigger apoptosis or cell necrosis or both by inhibiting cytochrome oxidase (54). Moreover, in the presence of superoxide NO reacts very rapidly, generating peroxynitrite (ONOO_) radicals that can impair proteins involved in electron transport resulting in a further increase of ROS production (290) (Fig. 3).

The mitochondria would not only be the origin, but also the target of oxidative/nitrosative stress. The mitochondrial DNA (mtDNA) could be a major target for ROS-mediated damage for several reasons (Fig. 3). First, mitochondria do not have a complex chromatin organization consisting of histone proteins, which may serve as a protective barrier against ROS. Second, mtDNA has a limited repair activity against DNA damage. Third, a large part of O₂ ^•−, which is formed inside the mitochondria, cannot pass through the membranes; hence, ROS damage may be contained largely within the mitochondria (376). In fact, mtDNA accumulates significantly higher levels of the DNA oxidation product (8-hydroxydeoxyguanosine) than nuclear DNA (141). A number of pathogenic mtDNA base substitution mutations, such as missense mutations and mtDNA rearrangement mutations (deletions and insertions), have been identified in patients with mitochondrial diseases (190, 226, 248, 390).

In HF, the centrality of the mitochondrion is not related only to its ROS generation and signaling, but also its role in bioenergetics' production, in particular at ATP levels that are critical for myocardial contractility and electrophysiology (see section VI. B).

2. NADPH oxidases

Another important source of ROS in the heart is the NADPH oxidase family of enzymes that generate O₂ ^•− in a highly regulated manner by catalyzing electron transfer from NADPH onto molecular O₂. NADPH oxidases appear to be the only enzymes whose primary function is ROS generation; they also appear to be especially important for redox signaling (211). The NADPH oxidase is a multicomponent enzyme complex, initially discovered in neutrophils, consisting of a membrane-bound flavocytochrome (which mediates the electron transfer process), a catalytic Nox subunit and a lower molecular weight p22 ^phox subunit. Five distinct Nox subunits (Nox 1–5,) each encoded by a separate gene and expressed in a tissue-specific pattern, have been identified (53). In the cardiovascular system, Nox1 is expressed mainly in ventricular smooth muscle cells (216). Nox2 (or gp91 ^phox , the original neutrophil isoform) is expressed in endothelial cells (144, 223), cardiomyocytes (34, 162, 402), fibroblasts (292), and some ventricular smooth muscle cells (375). Nox4 is expressed in endothelial cells (2), ventricular smooth muscle cells (159), cardiomyocytes (62, 224), and fibroblasts (91). Nox3 is not expressed in cardiovascular cells, while Nox5 has been reported in human endothelial cells and smooth muscle cells but is not found in rodents (33, 181). Multiple Nox subunits are expressed within a single cell at different levels and in different locations supporting the concept that individual Nox subunits have specifically delineated functions within the cell. Nox2 and Nox4 are the predominant isoforms in cardiomyocytes. These isoforms can be activated by a variety of stimuli, including agonists such as AT-II norepinephrine, endothelin, and insulin, inflammatory cytokines and growth factor, mechanical stretch, or hypoxia/reoxygenation (Fig. 4). Byrne et al. (62) reported a differential response of the Nox isoforms to pressure versus AT-II induced hypertrophy in that Nox2 was critical for the development of cardiac hypertrophy in response to AT-II, but pressure overload induced hypertrophy from other sources of ROS (possibly Nox4 oxidase) may be involved. A very recent study indicates that Nox2 and Nox4 exhibit distinct patterns of agonist-induced activation and downstream kinase activation, which could be attributable to specific compartmentation of redox signaling (15). Activation mechanisms for Nox2 involve complex formation with regulatory cytosolic subunits, namely, p47 ^phox , p67 ^phox , p40 ^phox , and Rac1 (7). Although oxidase activation is primarily driven by interactions with p67 ^phox and Rac1 subunit, other subunits (in particular p47 ^phox ) are important for translocation of the activating cytosolic subunits to the flavocytochrome and overall organization of the activated enzyme complex (32, 74). Posttranslational modifications of these subunits control the key steps, notably phosphorylation of p47 ^phox and geranyl-geranylation of Rac1, which are promoted by the above-described stimuli (223). However, Nox4 appears to be quite distinct in that the only subunit that appears to be necessary for its activity is p22 ^phox (12, 115, 246) (Fig. 4). Nox4 is an inducible Nox, and its activity is proportional to Nox4 protein expression alone. Heymes et al. (158) provided the first evidence that NADPH oxidase is expressed in human myocardium and specifically in cardiomyocytes. They also showed that NADPH oxidase activity is increased in end-stage failing human heart and that this activation was due to p47 ^phox translocation rather than increases in oxidase subunit expression. In line with this finding we (278) observed a coordinated increase of NADPH oxidase activity in the right (RV) and the left (LV) from failing end-stage hearts irrespective of the etiology of the disease (ischemic or dilated cardiomyopathy). We also found that increased Nox2 expression was not a prerequisite for increased Nox-dependent ROS production and, on the other hand, we observed both overexpression and membrane translocation of p47 ^phox the regulatory subunit of Nox2. The relative contribution of the two regulatory mechanisms to NADPH oxidase activation appeared different in the two ventricles (see below). The hypothesis that p47 ^phox translocation was a requirement for enzyme activation was reinforced by the finding of Borchi et al. (47) in which in human failing hearts Nox4 was not significantly increased with respect to the control ones. Moreover Maak et al. (238) showed that in patients with both ischemic and dilated cardiomyopathy, in addition to overexpression of p47 ^phox protein expression and translocation, an upregulation of Rac1 as well as increased rac1-GTPase activity was associated with increased NADPH oxidase. Treatment of HF patients with statins decreased Rac-1 activity in myocardium from these patients, possibly via statin effects on ROS activity (see below). In fact, statins, in addition to inhibiting cholesterol synthesis, downregulate Rac1-GTPase activity by reducing isoprenylation and translocation of Rac1 to the cell membrane (178).

OPEN IN VIEWER

FIG. 4.

Schematic structure and main activator of Nox2 and Nox4, the predominant isoforms of NADPH oxidase in the heart. A different range of stimuli activate the Nox's, including G-protein-coupled receptor agonists, mechanical forces, ischemia-associated factors, metabolic factors, growth factors, and cytokines. Notably, the same agonist can stimulate different isoforms in different cells to mediate a distinct cellular effect (see text for details) (dashed line indicates preferential stimuli for Nox4). AGE, advanced glycation end-products; GPCR, G-protein coupled receptor; LDL, low density lipoproteins; NEFA, nonesterified fatty acids; TNF-α, TNF-β, tumor necrosis factor α and β; VEGF, vascular endothelial growth factor.

B. NO and RNS in HF

Accumulating evidence indicates that NO and its derivative RNS may also have important effects in the developing of HF (287, 290). NO is a lipophilic diatomic gas with a relatively small Stoke's radius, and this, in combination with its neutral charge, facilitates rapid membrane diffusion (143). The presence of an unpaired electron in NO supports its high reactivity with O₂, O₂ ^•−, transition metals, and thiols, which largely shape its cellular functions within the cell (287).

Biosynthesis of NO is dependent on enzymatic activity of NOSs, which catalyze the conversion of the amino acid L-arginine to L-citrulline in a reaction that requires O₂ and a cofactor like as tetrahydrobiopterin (BH4), which is essential for the catalytic activity of all NOS isoforms. In fact, due to its oxidation and/or reduced synthesis BH4 depletion, can result in functional uncoupling of NOS (249, 357). Uncoupled NOS generates more ROS and less NO, shifting the nitroso-redox balance and having adverse consequences on the cardiovascular system (262).

In the heart, NO is produced by constitutive NOS, eNOS, localized in endothelium and caveolae of cardiomyocytes, by neuronal NOS (nNOS), colocalized with ryanodine receptor (RyR) in cardiac sarcoplasmic reticulum (SR), and under pathological situations by inducible NOS (iNOS) in the sarcoplasm (339). The isoforms nNOS and eNOS are regulated by a variety of different stimuli in different cell types via Ca^2+/calmodulin activation, as well as de novo synthesis (14, 51, 96). In addition to Ca^2+/calmodulin activation, phosphorylation is also involved in the regulation of eNOS activity (171). In contrast, the activity of iNOS is regulated by expression in a Ca²⁺-independent manner (340). One of the major controversies surrounding NO in the heart derives from the observation that in HF and in cardiac injury due to myocardial infarction, NOS isoforms have been ascribed both protective and detrimental roles (327). The findings that eNOS-deficient mice develop more severe LV dysfunction and remodeling after myocardial infarction than do wild-type mice (330) and, vice versa, that endothelial overexpression of eNOS attenuates LV dysfunction in mice after myocardial infarction (183) suggest that NO was beneficial in HF. However, this effect is lost by the coexistence of oxidative stress from eNOS uncoupling that stimulates cardiac pathologic remodeling from chronic pressure load. In fact, in a transgenic eNOS knockout model with low ROS production, severely pressure-overloaded hearts developed only modest concentric hypertrophy with little fibrosis and without LV cavity dilation (366), indicating that eNOS activation becomes detrimental rather than beneficial in LV remodeling when ROS coexists and eNOS uncoupling occurs. Importantly, eNOS activation has been also proposed as a key mechanism for cardioprotection mediated by early ischemic preconditioning. In this setting, rise of intracellular Ca²⁺ stimulates eNOS function through activation of Ca²⁺/calmodulin-dependent protein kinase II (CAMK-II). Resultant increase in eNOS-derived NO promotes posttranslational modifications of proteins in cardiomyocytes as described in Figure 5A. The subsequent activation of cyclic guanosine 5′-triphosphate (cGMP)-dependent protein kinase and PKC-ɛ promotes opening of mitoKATP channels (see section II, A1) that inhibit MPT. The activation of PKC-ɛ also activates nuclear factor kappa B (NF-κB) and upregulates transcription of iNOS, which is an obligatory mediator of the late phase of ischemic preconditioning, and of manganese superoxide dismutase (Mn-SOD) (45). Recent evidence demonstrates that cardiomyocyte-restricted expression of iNOS is sufficient to confer chronic cardioprotection and that transgenic upregulation of cardiac iNOS decreases reperfusion-free radical generation, which in turn prevents MPT and chronically protects the heart against ischemia–reperfusion injury (394). It is an intriguing finding because numerous experimental and human studies have demonstrated overexpression and increased activity of iNOS in the myocardium of animals and patients with various forms of HF and benefits of iNOS inhibition on cardiac function (125, 204, 259) (see section II, D). Thus, the role of iNOS and NO in the development and progression of heart disease is a subject of recent debate, the preeminent hypothesis being that increased iNOS activity contributes to cardiac maladaptation. Nevertheless, it is tempting to speculate that increasing bioavailability of iNOS-derived NO concomitant to simultaneous inhibition of oxidative stress may become a novel strategy to treat HF failure (45, 110, 266).

OPEN IN VIEWER

FIG. 5.

Interplay between ROS- and NO-regulated pathways in the cardiomyocyte. (A) When the level of NO and O₂ ^•− production are balanced within the cell, the posttranslational modification of effector proteins is facilitated. For instance, NO binds protein thiols (−SH), causing S nitrosylation that regulates the activity of several proteins, including the cardiac L-type calcium channel, and ryanodine receptor. Alternatively, NO binds hemoproteins generating nitrosyl–iron complex with the resultant activation of GC, which generates cGMP that activates cGMP-dependent protein kinase involved in cardioprotection through mitoKATP opening (see the text). NO in the presence of O₂ ^•− generates peroxynitrite that at low concentration causes tyrosine nitration of certain proteins also involved in cardioprotective signal transduction such as PKC-ɛ and Src. All the cardioprotective signal transductions promoted by NO converge on inhibition of MPTP opening. On the other hand, ROS bind −SH, inducing a sequential generation of reversible (sulfenic derivate, −SOH) or irreversible oxidation products (sulfinic [−SO₂H] and sulfonic [−SO₃H] derivatives). Additionally, ROS can induce disulfide formation (−S−S−). This latter modification prevents further and irreversible oxidation. Moreover, SH oxidation leads to reversible inactivation of PTP that results in activation in protein tyrosin kinase (i.e., Src) or of other adaptive redox signaling pathways (Ras-Raf-MEK/ERK). The concomitant action of direct Src oxidation (−S−S−) and PTP inhibition actives the Src-PKC-PI3K signal transduction pathways involved in the cardioprotective effect mediated by ischemic preconditioning. (B) NO/O2 imbalance (characteristic of HF) favors oxidation reactions disrupting protein S-nitrosylation signaling and promoting peroxynitrite formation. This latter, through nitration of several critical proteins (SERCA, voltage-gated K+ channel, desmin, etc.), has been proposed as a major mechanism of cardiomyocyte dysfunction. Peroxynitrite also binds many components of the MRC, thereby inhibiting the mitochondria function and ATP generation, contributing to cardiomyocyte injury. An excess of ROS promotes irreversible oxidation products that irreversibly inhibits protein phosphatases, leading to an overactivation of maladaptive signaling pathways (ASK-1, JNK/SAPK, etc.). ASK-1, apoptosis-signal regulating kinase-1; cGMP, cyclic guanosine 5′-triphosphate; ERK, extracellular signal-regulated protein kinase p38; GC, guanylyl cyclase; JNK/SAPK, c-Jun NH2-terminal kinase/stress-activated protein kinase; MAPKs, mitogen activated protein kinases; Mn-SOD, superoxide dismutase; MPTP, mitochondrial permeability transition pore; MRC, mitochondrial respiratory chain; P3K/Akt, phosphoinositide 3-kinase; PKC, protein kinase C; PKG, cGMP-dependent protein kinase; PTP, protein tyrosin phosphatases; RyR, ryanodine receptor; SERCA, sarco/endoplasmic reticulum calcium ATPase.

At variance with eNOS and iNOS, the contribution of nNOS to myocardial pathophysiology has been undervalued for years. However, recent experimental studies point to an important role for myocardial constitutive NO production through nNOS in the regulation of basal and adrenergic-modulated cardiac function (58, 72, 228). The location of nNOS in the SR and/or sarcolemma suggests that ion channels and transporters involved in the regulation of Ca²⁺ cycling in the myocyte may be obvious targets for NO downstream signaling, as extensively described later. Emerging data showing colocalization of XOR and nNOS in the SR of rodents, and increased XOR activity in the nNOS^−/− mice myocardium suggest that nNOS gene deletion may have wider implications on the myocardial redox state (196, 198).

In general, the documented nNOS-mediated inhibition of Ca²⁺ entry through L-type Ca²⁺ channels (337) is a likely mechanism through which nNOS-derived NO regulates Ca²⁺ cycling within intracellular Ca²⁺ stores and therefore excitation–contraction coupling mechanisms (see later on). Altogether, available evidence suggests that NO produced by the neuronal isoform may exert a negative feedback regulation on Ca²⁺ influx: any increase in intracellular Ca²⁺ would stimulate nNOS-mediated synthesis of NO, which in turn attenuates L-type Ca²⁺ current.

The discovery that eNOS and nNOS are expressed in distinct subcellular compartments in the cardiomyocyte (124, 396, 403) suggests that the two isoforms might couple to distinct effector molecules and elicit quite different results following enzyme activation. In fact, NO diffusion within cardiac myocytes is likely to be limited to a local environment by both a high cytoplasmic concentration of myoglobin (which has a high affinity for NO and acts as a scavenger) and, particularly in disease states, by an abundance of superoxide anions (which can react with NO to limit its bioavailability) (130). Although the NO effect is mainly compartmentalized, emerging evidence suggests that NO generated at one compartment can be converted to inert NO metabolites such as nitrite and nitrate and transferred to other compartment and then reduced back to NO in the presence of myoglobin (101). This finding has relevant consequences, since several studies in animal models demonstrated the ability of nitrite to provide potent cytoprotection against cardiac injury of the heart (101). For example, Hendgen-Cotta et al. (157) recently reported that during hypoxia nitrite can enter cardiomyocytes and be reduced to NO by endogenous myoglobin nitrite reductase activity, to inhibit mitochondrial respiration potentially by binding cytochrome c oxidase. The reversible slowing in the rate of mitochondrial respiration may allow preserving oxygen gradients during ischemia and favor a more controlled resumption of respiration during reoxygenation, minimizing free radical injury. Although the mechanisms of this phenomenon are still waiting complete characterization, the reproducibility of this effect in multiple animal models of ischemia–reperfusion suggests nitrite as a novel potential therapy for human ischemic diseases. Summarizing, in HF oxidant-producing enzymes are upregulated and NO-producing enzymes are altered in either their abundance or spatial localization. A relative NO deficiency may further promote oxidase activities, which suggests that NO may be a global modulator of O2–/ROS production (155). In particular, NADPH oxidase activities are increased in the failing hearts, at least in part due to increased levels of AT-II, which indicates a link between neurohormonal activation and NO/redox disequilibrium (263). Moreover, increased XO activity that directly reflected a dysregulation of NO signaling contributes to vasoconstriction and depressed cardiac function (324).

C. Signals activating ROS production

Among the stimuli that generate ROS are the following: peptide and amine hormones, such as AT-II, norepinephrine, endothelin, serotonin, and thrombin; cytokines, such as interleukin-1β, interleukin-6, and tumor necrosis factor-α; and mitogens, such as fibroblast growth factor and epidermal growth factor. These agonists bind to receptors of different classes that are coupled to multiple signal transduction cascades, including tyrosine kinases, mitogen-activated protein kinases (MAPKs), PKC, calcineurin, phosphoinositide 3-kinase (PI3-kinase)/Akt, and NF-κB (371). Mechanical stress is another important stimulus that generates ROS production (6, 418) as well as hypoxia-ischemia/reoxygenation (48).

AT-II, a small peptide and main component of the renin–angiotensin system (RAS), plays an important role in cardiovascular homeostasis (227), cell metabolism, and pathogenesis of cardiovascular diseases, including HF (19). Not only does AT-II bind to its cell surface receptors (angiotensin type 1 receptor antagonists [AT1] and AT2) and exert multiple effects by activating a number of intracellular signal transduction pathways through heterotrimeric G proteins, but it also interacts with receptor tyrosine kinases and cytokine receptors (139, 253, 381). It is well known that AT-II elicits in cardiomyocytes the generation of ROS, mainly due to NADPH oxidase, and that locally produced ROS acts as an initial mediator of AT-II-induced signal transduction, gene regulation, and phenotypic modulation in cardiomyocytes (9, 272, 349). During HF, cross-talk between some of these stimuli may affect many aspects of the adverse remodeling. For instance, in dilated cardiomyopathy, ventricular hypertrophy or dilatation elicits mechanical stress with concurrent changes of extracellular matrix and cytoskeleton (214, 329). By disarraying myocardial contraction and impairing cell metabolism, ischemia switches several ionic channels on/off, thus perturbing intracellular (and extracellular) ionic milieu (30, 280).

Mechanical stress also triggers the release of AT-II and endothelin, which in turn increases ROS production, stimulating the activity of NADPH oxidase (64, 97) via a PKC-mediated phosphorylation of the oxidase's p47 ^phox subunit, leading to its translocation to the membrane-bound cytochrome (32, 149, 263, 271). In ventricular myocytes, stretching (195, 406) and AT-II mediated pathways rapidly converge toward Rac1-GTPase activation (17), which likely plays an early role in mechano-transduction and neurohumoral signaling. Stretching depolarizes cell membrane during diastole changes the action potential profile and predisposes to arrhythmias, effects mediated at least in part by stretch-activated ion channels (170). Stretch-activated ion channels are mechano-sensitive molecules localized to specific regions of the cell such as the Z-disc, thus sensing changes in mechanical forces imposed by the hemodynamic load. Channel opening and consequent sodium/calcium flowing into the cell can influence electrophysiological properties and also trigger mechano-sensitive signaling cascades (355).

D. Downstream effectors and targets

The biological effects of ROS/RNS depend upon the specific moiety generated, its localization, and the relative balance between levels generated and the activity of antioxidant systems that reduce ROS levels. In the setting of NO/ROS imbalance where large amounts of radicals are generated, these may induce oxidation and damage of macromolecules, membranes, and DNA and thus be detrimental for cellular function and viability (155).

Accumulating evidence suggests that ROS/RNS are not only injurious but also essential participants in cell signaling and regulation (327, 339). The term nitroso/redox signaling describes a process by which milder physiological levels of ROS/RNS induce modifications to proteins that are discrete, site specific, and reversible. Nitroso/redox signals may abrogate or enhance activity of the target protein, and have been implicated in physiological signaling processes that include kinase signaling, channel function, apoptotic proteolysis, and regulation of transcription. The tenet of redox sensing is that certain proteins undergo reversible chemical changes in response to changes in local redox potential. Cysteine residues are the typical targets of redox modification, as they can adopt multiple oxidation states (133). The free thiol group of cysteine can undergo reversible or irreversible covalent modifications by ROS that can induce the sequential generation of reversible or irreversible oxidation products, such as sulfenic, sulfinic, and sulfonic derivatives. Additionally, ROS can induce disulphide formation. Blockade of cysteine -sulfenic derivate by disulphide formation (either with intramolecular cystine or with glutathione [GSH] binding) is the common mechanism of all redox regulated protein tyrosine phosphatases. This process leads to reversible inhibition of protein tyrosine phosphatases and prevents their further and irreversible oxidation. The same mechanism, on the contrary, induces activation of protein tyrosine kinases like Src (82).

At a molecular level, NO and RNS exert their biological actions by chemical modification of target molecules, preferentially interacting with transition metals, free radicals, and, similarly to ROS, with thiol groups (287). One well-known signaling pathway of NO is initiated by its reaction with heme iron of soluble GC, leading to increased cGMP accumulation and kinase activation. Binding of NO to other heme proteins, such as cytochrome c oxidase competes with that of oxygen and regulates oxygen consumption by the mitochondria (290) Another target is protein tyrosine residues, which can be modified to stable 3-nitrotyrosines. Posttranslational modification of tyrosine residues has been shown to play an important role in modulating the activity of several PKC isozymes, including PKC-ɛ and Src, which has consistently been implicated in the cardioprotective signal transduction (287), as described above. On the other hand, tyrosine nitration of several critical proteins (Ca²⁺-ATPase pump, sarco/endoplasmic reticulum calcium ATPase [SERCA], voltage-gated K+ channel, desmin, etc.) has been proposed as a major mechanism contributing to cardiomyocyte dysfunction (290). Another important signaling pathway of NO and RNS is represented by the reaction with thiols. Protein thiols and GSH react with NO derivatives to produce a range of products, including S-nitrosothiols (312). S-nitrosation of protein thiols has gained particular interest because it may modulate a fast reversible regulation of protein functions, including the cardiac L-type Ca²⁺ channel (168) and the RyR2 (228) as described in section V. Importantly, this process is sensitive to disruption by redox milieu. In this context, oxidative stress disrupts a physiologic signaling process driven by S-nitrosation and mediated by modification of effector proteins. In other words, the relative flux of NO and O₂, depending on abundance and location of both NOS's and oxidases in cardiomyocytes, determines the chemical fate of their interactions. In fact, during physiologic situations when NO is higher than O₂, S-nitrosylation is favored; instead, NO/O₂ imbalance, characteristic of HF, favors oxidation reactions (155).

Intracellular GSH has a central role in cellular redox balance (250). Glutathionylation of proteins, through the formation of a mixed disulphide between one cysteine of glutathione and one cysteine of the other protein, constitutes an efficient mechanism to protect proteins from irreversible modifications. GSH also provides cellular protection against oxidative damage by reacting with RNS to form S-nitrosoglutathione. GSH thereby serves as an efficient endogenous scavenger of peroxynitrite and plays a major role in the cellular defense against this species (290). Accordingly, the susceptibility of cells to peroxynitrite toxicity largely depends on the amount of intracellular GSH.

Cellular GSH abundance, combined with the ready conversion of S-nitroso and sulfenic acid derivatives into S-glutathione mixed disulfides, strongly suggests that reversible protein S-glutathionylation is a central mechanism of redox signal transduction (342). The possible impact of S-nitrosation in mediating cardiac effects of NO in a situation of nitrosative stress is largely unknown. In circumstances of acute ischemia, sepsis, or HF, iNOS abundance may increase, leading to nitrosative stress, a pathophysiologic situation characterized by accumulation of S-nitrosylated proteins to hazardous levels (amount and/or spatio-temporal distribution) (154). Nitrosative stress may be exacerbated in situ by oxidants (oxidative stress); stimuli that lead to iNOS induction may also upregulate oxidases, and concomitant elevations in NO/RNS and O₂–/ROS may lead to formation of higher amounts of peroxynitrite (291) This condition favors polynitrosylation and oxidation of cysteine thiols as well as nitration of tyrosines in proteins as reported above. This situation is relevant to the failing heart, in which the loss of spatial confinement of nNOS, which redistributes from SR to the plasma membrane, may be a proximate cause of oxidative/nitrosative stress both by relieving the local (SR) control of XO (196), and by altering NO/redox balance at the sarcolemma. At the molecular level, poly–S-nitrosylation, oxidation and nitration of the SERCA2a and RyR2 may ensue with adverse effects on Ca²⁺ homeostasis (404). Thus, a central pathophysiologic consequence of redox disequilibrium is the disruption of NO signaling by alteration of the occurrence or nature of the posttranslational modifications.

Multiple enzymatic systems regulate the functional changes induced by ROS/RNS, including glutaredoxin, thioredoxin (4), and glutathione S-transferase. Glutaredoxin acts as a specific and efficient catalyst of protein de-glutathionylation reactions (342). In addition to the glutaredoxins, thioredoxin can also reduce reversibly modified thiols. Some isoenzymes of the super family of glutathione S-transferases can also regulate MAPKs or facilitate protein S-glutathionylation (251).

Among the ROS, H₂O₂ is more stable than O₂ ^•− and capable of crossing biological membranes. It makes H₂O₂ especially suitable as a signaling molecule. H₂O₂ is tightly regulated biologically by catalase (CAT), glutathione peroxidase (GPx), and peroxiredoxins, which convert H₂O₂ to water and other metabolites (53).

As illustrated in Figure 5 ROS/RNS may modulate diverse redox signaling pathways by activation of kinases and/or oxidative inactivation of protein phosphatases (82) resulting in increasing tyrosine and serine/threonine phosphorylation signaling events. These events are converged and integrated to induce various redox-sensitive transcriptional factors that are involved in regulating redox-sensitive gene expression, leading to various physiological and pathophysiological responses. Among the several downstream signaling pathways activated in cardiomyocyte the extracellular signal-regulated kinase, Jun N-terminal kinase (JNK), apoptosis signal-regulating kinase 1, and PKC (21, 179, 233), cyclic AMP-dependent protein kinase (PKA) (334), cGMP-dependent protein kinase (57), and Ca2+/calmodulin-dependent protein kinase II (118) are well established and characterized. Nonreceptor tyrosine kinases c-Abl and Abl-related gene (Arg), belonging to the so-called Abl family (313), have recently been added to the list on the basis of the latest observations by Borchi et al. (48), being able to modulate the activity of some antioxidant enzymes via posttranslational modification. Another important signaling pathways is phosphoinositide 3-kinase (PI3K)/Akt that plays an important role in myocardial metabolism (88, 207, 282), as central mediator of insulin's effects (see Section VI), and mechanotransduction (288, 361).

A. Antioxidant enzymatic system

B. Antioxidant enzymatic defense in HF

In various animal models of HF, enhanced oxidative stress may be combined with a decrease in scavenging enzyme activity, either as a cause or as a result. It remains unclear whether human endstage HF is also associated with altered myocardial antioxidant reserve (105). To date, a comprehensive view of changes in oxidative stress mechanisms and antioxidant enzyme activity in human HF is far from being achieved. Moreover, the currently available information is somehow contradictory. For example, CAT activity in the failing human ventricle was reported to be either decreased (29), upregulated (105) or unchanged (325) in different studies. Sam et al. (325) found a decreased Mn-SOD activity though in the presence of upregulated mRNA levels in human failing hearts, whereas Baumer et al. (29) showed Mn-SOD mRNA, protein levels, and activity unchanged in human hearts with idiopathic dilated cardiomyopathy, in accordance to Dieterich et al. (105). As for GPx, all these studies agreed for no change in activity of this enzyme. During pathophysiological conditions, several cellular ROS sources are activated which alter intracellular oxidative–antioxidant homeostasis in favor of the former. As an adaptive response, antioxidant defense systems are upregulated in response to increased ROS. Most antioxidant enzymes contain gene regulatory sequences in their promoter and intron regions that can interact with redox-sensitive transcription factors to trigger upregulation of gene expression (9). Thus, upregulation of gene expression via redox-sensitive signaling pathway becomes a necessary strategy to restore homeostasis. Mn-SOD is a well-known target gene for NF-κB activation and its expression has been shown to be upregulated by acute stimuli (166). A redox-dependent posttranslational modification, including protein phosphorylation/oxidation is another key signaling response.

A contribution to a possible mutual relationship between oxidant and antioxidant activities in human failing hearts was reported by our recent study (47). We found that despite a marked increase of CAT and GPx activities, mRNA expression and protein levels were unchanged for both enzymes. This finding supports the hypothesis that dissociation occurs between mRNA/protein expression and activity and suggests that the adaptive response to increased oxidative stress does not lead to increased transcription of these antioxidant enzymes but may reflect posttranslational modifications. In fact, a significant CAT and GPx Tyr phosphorylation was found in failing hearts when compared to the control ones, and these changes in Tyr phosphorylation go hand-in-hand with enzyme activities. Although clear evidence for a cause–effect relationship was prevented in end-stage hearts by obvious limitations, on the basis of Cao et al. (66 –68) and our (47) findings, we speculated that increased NADPH oxidase-mediated ROS production may activate c-Abl and Arg kinases, with subsequent stimulation of the CAT and GPx catalytic activities, a mechanism that may contribute to normalize the level of oxidative stress (Fig. 6).

OPEN IN VIEWER

FIG. 6.

Hypothetical mechanism responsible for the increase of CAT and GPx activities in end-stage failing heart. Increased NADPH-oxidase mediated ROS production may activate c-Abl and Arg kinases, with subsequent stimulation of the catalytic activities of CAT and GPx via c-Abl/Arg-mediated tyrosine phosphorylation (A). (B) Representative immunoblot showing tyrosine phosphorylation of CAT and GPx in right and left ventricles from nonfailing and failing hearts. Modified from Borchi et al. (58). Arg, Abl related gene; c-Abl, Abelson family nonreceptor tyrosine kinase; CAT, catalase; F, failing heart; GPx, glutathione peroxidase; LV, left ventricle; NF, nonfailing heart; RV, right ventricle; SOD, superoxide dismutase.

A. NO/ROS and calcium homeostasis

A hallmark of HF is the diminished contractile capacity. Many culprits account for the malfunction: loss of viable myocytes due to necrosis and apoptosis (see above), their replacement by noncontractile substrate (fibroblasts and collagen), local release of mediators affecting inotropism (natriuretic peptides, purines), or desensitization to positive inotropic agents (e.g., loss of function of the β-adrenergic signaling pathway) and, not least, intrinsic cardiomyocyte remodeling. Indeed, even when isolated from cardiac tissue by enzymatic digestion, failing myocytes persistently exhibit depressed inotropic responses and altered intracellular Ca²⁺ homeostasis. To get insight into the mechanisms and role of redox imbalance in determining this maladaptive behavior, it is worthwhile to recall the key features of excitation–contraction coupling mechanisms (Fig. 7). Special attention will be paid to proteins, which are sensitive to the NO/redox environment: as above described, most of proteins contain free thiol groups of cysteine, making them susceptible to the redox action and nitrosylation and subsequent alterations in their properties. In Figure 7, proteins undergoing NO/redox control are marked by “*”: it is evident that most of Ca²⁺ handling proteins are potential substrates.

OPEN IN VIEWER

FIG. 7.

Schematic representation of the excitation–contraction coupling system in the cardiac cell. Calcium handling proteins undergoing redox control are marked by “*.” During the action potential, Ca²⁺ enters the cells through the L-type Ca²⁺ channels (Cav1.2), which are mainly localized in the T-tubule system, generating the rapid and maintained component of the L-type Ca²⁺ current (I_Ca,L). The rapid influx of Ca²⁺ via the T tubules induces the release of Ca²⁺ from a subsarcolemmal compartment of the terminal SR (CICR) through the RyR2 SR Ca²⁺ release channels. RyR2 interact with a complex of proteins at the luminal side of the terminal SR, to form a protein complex with regulatory proteins such as calsequestrin2, triadin, and junctin. RyR2 also associates with calstabin2 (from Ca²⁺ release channel-stabilizing protein); dissociation, induced, for example, by PKA-mediated phosphorylation, increases activity and open probability of RyR2. Ca²⁺ is released from SR activating the contractile filaments to contract. Relaxation follows because the cytosolic Ca²⁺ is sequestered again in an uptake compartment of the SR by the SR Ca²⁺ ATPase (SERCA2), which is modulated by PLN. Part of cytosolic Ca²⁺ is extruded through the cell membrane by the NCX and by the low-capacity high-affinity Ca²⁺ pump (ATP). The process of Na⁺/Ca²⁺ exchange is electrogenic so that Ca²⁺ extrusion through the exchanger leads to a depolarizing current (see also Fig. 8). The force of contraction is thus determined by the circulation of Ca²⁺ from the SR to the myofilaments and back to the SR, and by the amount of Ca²⁺ that has entered during the preceding action potential. The relaxation rate of the twitch depends on the rate of Ca²⁺ dissociation from the myofilaments and on the rates of Ca²⁺ sequestration and extrusion. Both relaxation and contraction are severely impaired in failing cardiomyocytes. CICR, calcium induced calcium release; NCX, Na⁺–Ca²⁺ exchanger; PLN, phospholamban; SR, sarcoplasmic reticulum.

During the action potential plateau, L-type Ca²⁺ channels activation triggers a subsarcolemmal rise in Ca²⁺, which initiates the release of Ca²⁺ from SR mainly via the RyR2; a process termed Ca²⁺-induced Ca²⁺ release (CICR). The free Ca²⁺ in the cytosol increases from nanomolar to micromolar levels and binds to troponin C, thus producing a conformational change in the troponin I–tropomyosin complex, which permits the initiation of myosin-actin cross-bridge cycling and contraction. Recovery of intracellular Ca²⁺ concentration to diastolic level is mainly due to reuptake into SR via SERCA2 and, in order of relevance, to extracellular extrusion via sarcolemmal Na⁺–Ca²⁺exchanger (NCX), sarcolemmal Ca²⁺-ATPase, and mitochondrial Ca²⁺ uniporter (37). Amplitude of Ca²⁺-transients (e.g., as measured by Ca²⁺-sensitive dyes) is substantially decreased in HF and its kinetics slowed down (Fig. 7).

Redox-sensitive thiols are present in Ca²⁺ channel (65), which may account for the abrupt decrease in L-type Ca²⁺ current observed in isolated cardiomyocytes exposed to various ROS sources (75). The α1C subunit of the L-type Ca²⁺ channel contains 48 cysteine residues that are potential thiol redox-sensitive sites of the channel. While some of these residues are not available for oxidation or reduction, being within the plasma-membrane itself or already involved in disulphide bonds, others are likely target of redox or nitrosylation modification (168). However, Ca²⁺ current amplitude is not significantly diminished in failing myocytes, thus suggesting that downstream abnormalities are likely responsible, such as defective coupling between L-type Ca²⁺ channels and RyR2 and/or the contractile machinery. Underlying mechanisms, whose detailed description is beyond the scope of this review, have been extensively reviewed (42).

Reduced SR Ca²⁺ release is observed in end-stage HF as a consequence of diminished Ca²⁺ content (299); abnormalities in both release (e.g., RyR2) and uptake (SERCA) mechanisms cooperate to empty SR. Malfunction is caused by dysregulation of several pathways, namely, phosphorylation and redox pathways. RyR2 hyperphosphorylation by PKA (245) or Ca2+/calmodulin-dependent protein kinase II (5) in HF has been consistently observed; the consequence is an SR leak of Ca²⁺ during diastole because of enhanced Ca²⁺ sensitivity of the RyR2. Recently, the role of redox mechanisms as RyR2 regulator has emerged. Susceptibility of the RyR2 to reducing-oxidizing modifications is suggested by hyperactive cysteine residues, which may form disulfide bonds, thus altering RyR2 structure and function, including sensitivity to luminal or cytoplasmic concentration (123). In failing canine cardiomyocytes, altered redox balance was paralleled by reduced Ca²⁺ transients and enhanced leakage of Ca²⁺ from SR; treatment of HF myocytes with reducing agents (dithiothreitol and 2-mercaptopropionylglycine) led to normalization of Ca²⁺ cycling. At the same time, oxidizing agents converted normal cardiomyocytes to a failing phenotype (370).

Hyper-responsiveness to [Ca²⁺]_i was also observed in dystrophic cardiomyocytes, a condition characterized by high ROS production, and could be normalized by the reducing agent mercaptopropionyl-glycine (378). A second modulator pathway consists in S-nitrosylation of RyR2, that is, the covalent attachment of a NO moiety to a reactive thiol side chain on a cysteine residue (228). Exogenous NO donors (S-nitroso-N-acetyl penicillamine, SNAP) increases RyR2 opening reversibly; on the other side, sulfhydryl reducing agents promote channel closure (356). This evidence, obtained in synthetic planar lipid bilayers, was confirmed in intact cardiomyocytes by Ziolo et al. (421); the picture emerging from these studies is a complex modulation of RyR2 by NO, which depends not only on NO concentration but also on the level of channel phosphorylation by PKA. Recent data indicate that the 1-electron reduction product of NO, nitroxyl, can potently activate RyR2 at picomolar concentrations (81).

Contrary to the diseased state, NO exerts a physiological role as endogenous and constitutive modulator of RyR2, as briefly mentioned in the previous sections (II, B). The nNOS immunoprecipitates with RyR2 (35, 94). Experiments in cardiomyocytes from nNOS knockout mice suggest that nNOS-derived NO modulate basic SR function by reducing release and accelerating reuptake, an effect that translates into a slight negative inotropic and positive lusitropic effect [as reviewed by Lim et al. (228)]. Finally, S-nitrosylation of several effectors of excitation contraction coupling mechanism (L-type Ca²⁺ channels, SERCA2) has been detected during ischemic preconditioning and appears to be involved in cardioprotection (363). Therefore, S-nitrosylation can be viewed as a physiological mechanism controlling beat-to-beat cardiomyocyte function; its rapid and reversible activity on several relevant effectors of excitation–contraction coupling is important for adjusting the contractile machinery to neurohumoral (e.g., beta-adrenergic stimulation) or metabolic state (e.g., oxygen supply). Likely, in chronic HF most of these regulatory and cardioprotective mechanisms are largely ruled out by irreversible post-transductional modifications, alterations in cell size, and compartmentation of subcellular structures.

Recent evidence points at the interplay between ROS and RNS species as a signaling axis (155). As reported above, NO and O₂ ⁻ interact by forming the highly reactive species peroxynitrite, a reaction which—on one side—exhausts NO and—on the other side—scavenges O₂ ^•−. Therefore, unscavenged O₂ ⁻ could result in an RyR2 gain of function by irreversible activation consequent to oxidative stress and of RyR2 (404).

Being a highly reactive species, NO effect is extremely compartmentalized; that said, immunoprecipitation of RyR2 and nNOS points at a spatially circumscribed interaction. In HF, compartmentalization is grossly altered by multiple factors, such as changes in volume/surface ratio of hypertrophied cells and loss of T-tubules. nNOS does not colocalize with caveolin-3 in normal myocytes, since its sarcolemmal expression is very low. However, nNOS/cav3 complexes increase in HF suggest a redistribution of nNOS from SR to plasma membrane (35, 94), which might further lessen NO-mediated modulation of RyR2 in cardiac disease. Combined to increased NO-mediated depression of sarcolemmal L-type Ca²⁺ channels (337), this phenomenon may contribute to Ca²⁺-depletion in the SR. Inhibition of SERCA2 by ROS is another crucial checkpoint of Ca²⁺ homeostasis. The cysteine residue near the ATP binding site of SERCA2 represents a possible target for ·OH-induced loss of function (84). In addition, ·OH may directly affect alter Ca²⁺ sensitivity of sarcomeric proteins, as demonstrated for skeletal muscle (268). Altogether, disruption of the constitutive ROS/RNS control of SR function is likely to contribute to the phenotype of HF.

B. ROS and the arrhythmogenic hazard in HF

Besides impairing contractility, altered Ca²⁺ cycling and excitation–contraction coupling triggers a cascade of events such as nonuniform mechanical dysfunction, cardiomyocyte overstretching, and cell-to-cell uncoupling, which may predispose to arrhythmias; the interlink between Ca²⁺ and arrhythmogenesis has been reviewed elsewhere (369). However, ROS may directly affect electrophysiological properties of cardiomyocytes by modulating relevant channels. Figure 8 shows a schematic action potential and underlying ion currents, with emphasis on those modulated by radical species. Besides Ca²⁺ channels, already discussed, several other ion channels and pumps may be affected.

OPEN IN VIEWER

FIG. 8.

Schematic representation of the human ventricular action potential ( upper ) and major underlying ionic currents and transporters ( bottom ). Ion currents likely undergoing redox control are marked by “*.” Abbreviations (left): I_Na, sodium current; I_CaL, L-type calcium current; I_f, hyperpolarization-activated pacemaker or funny current; I_NCX, sodium–calcium exchanger current; I_to, transient outward potassium current; I_Ks, delayed rectifier potassium current (slow); I_Kr, delayed rectifier potassium current (rapid); I_K1, inward rectifier potassium current. Abbreviations (right): Nav1.5, also known as SCN5A, sodium channel protein, type 5, subunit α; Cav1.2, voltage-dependent L-type calcium channel subunit α-1C; HCN, hyperpolarization-activated cyclic nucleotide-gated cation channel; Kv4.x, potassium voltage-gated channel family of genes; KChIP, Kv-channel interacting proteins; KvLQT1, also known as KCNQ1, potassium voltage-gated channel KQT-like subfamily, member 1; minK, also known as KCNE1, potassium voltage-gated channel, subfamily E, member 1; HERG, also known as KCNH2, potassium voltage-gated channel, subfamily H [eag-related]; MiRP-1, potassium voltage-gated channel, subfamily E, member 2; Kir2.x, inward rectifier potassium channel family of genes.

One of the first evidence of electrophysiological remodeling in human HF was the marked loss of function for the Ca²⁺-independent transient outward current (I_to). Channel activation/inactivation generates a fast, transient outward current that is responsible for the rapid repolarization phase of the ventricular action potential (phase 1) following the upstroke. The α- subunit is coded by a family of isoforms termed Kv1.4, Kv4.2, and Kv4.3 with different distribution among species. Downregulation of I_to expression and function has been documented in several animal models of myocardial hypertrophy and failure [see Nass et al. (274) for a review] and in diabetic cardiomyopathy (306). Examples of currents recorded in rat ventricular cardiomyocytes from control and hypertrophied or diabetic hearts are shown in Figure 9; current was restored by in-vivo treatment with AT-II receptor antagonists, thus suggesting a major role of redox-mediated mechanisms associated to RAS overactivation in this phenomenon (76, 306). A relevant transcriptional regulation has been proved, consisting in a reduction of mRNA levels and proteins coding for the fast Kv 4.2 and Kv 4.3 α subunits). On top of that, channel β subunits such as Kvβ1.2, nonenzymatic homologs of aldo-keto reductases (392), may confer redox-state sensitivity to the channel (296). Evidence of the redox control of potassium channels function comes from the pioneering work of Rozanski (318, 319). The reversible oxidative modification of proteins forming the TO-channel involves the reactivity of the free thiol group of cysteine residues, and may be viewed as a protective mechanism preventing other irreversible oxidative changes. Blunted expression (protein level) and function (current density) observed in rats with chronic myocardial infarction and diabetic cardiomyopathy were recovered by extracellular GSH or N-acetylcysteine, a putative GSH precursor (319, 405). In a rat model of impaired oxido-reductase due to inhibition of the thioredoxin/glutaredoxin systems, insulin significantly recovered I_to, an effect likely due to signaling control of intracellular redox state by the hormone (225). In general, the positive correlation between current density and protein level, more than with mRNA levels (225), suggests that redox mechanisms control sarcolemmal density of functional I_to channels at a posttranslational level, such as redox-mediated increase in the rate of channel degradation or decreased levels of chaperone β-subunits (K⁺ channel interacting protein 2). Notwithstanding the mechanisms, I_to reduction in HF is considered a proarrhyhmogenic mechanism that (i) destabilizes ventricular repolarization and, especially combined to chronic/acute fall of other potassium currents, predisposes to early after-depolarizations and torsade de pointes, (ii) alters the timing of early plateau phase and the kinetics of L-type Ca²⁺ current (25), (iii) affects crucial physiological processes such as action potential duration gradient within the ventricular wall (25) and the so-called cardiac memory (182).

OPEN IN VIEWER

FIG. 9.

I_to is reduced in hypertrophied ( upper panels ) and diabetic ( bottom panels ) rat ventricular cardiomyocytes. Each panel shows a family of currents evoked by depolarizing steps to increasing potentials, from −40 to 70 mV, as a function of time; current is expressed as density, that is, normalized with respect to membrane capacitance (in pF), which is an index of cell size. Normalization allows for comparing current amplitude in cells with different size, such as normal versus hypertrophied cardiomyocytes. Typically, upon depolarization I_to peaks in a few milliseconds and then inactivates; a residual steady state current persists throughout depolarization and depends on the activation of other potassium channels. Both diabetes and ventricular hypertrophy are associated with a marked reduction in peak outward current (middle panels) with respect to control (left panel). In vivo treatment of rats with losartan restores I_to amplitude to control values (right panels). Modified from refs. (229, 230).

Recently, redox-related modulation of the inwardly rectifying potassium channel I_K1 has been demonstrated, with an unexpected high affinity of the channel for this gaseous mediator (142). Specific S-nitrosylation of Cys76 of Kir2.1, one of the proteins coding for the α-subunit of the channel, increases the opening frequency and therefore current amplitude. This feature may represent a relevant mechanism by which NO participates in the control of cardiac excitability under physiological conditions. In fact, I_K1 is a key player for setting and maintaining diastolic membrane potential; therefore, any disturbance (i.e., impairment of NO signaling) would have profound effects on cardiac excitability and arrhythmogenesis.

Emerging evidence points to the sodium channel as a redox target leading to acute and chronic electrophysiological remodeling in HF. The sodium channel is responsible for the upstroke (phase 0) of the action potential; it generates a fast transient inward current (peak, I_Na,P) that inactivates almost completely within a few milliseconds. However, a noninactivating current persists and accounts for a net sodium influx during the plateau (phase 2).

Disruption of Nav1.5 (also known as SCN5A, sodium channel protein, type 5) inactivation increases the persistent Na⁺ current; well-known examples are mutations in the III–IV linker (Fig. 10) associated to the inherited LQT3 syndrome. The COOH terminal of the protein has binding sites for numerous regulatory proteins implicated in the modulation of Nav1.5 inactivation. The sodium channel is a highly modulated protein undergoing PKA and PKC phosphorylation, which causes, respectively, current gain and loss of function (419). In this respect, by activating PKC, ROS may exert an inhibitory yet indirect, activity on the channel. Recently, mutations in gene coding for glycerol-3-phosphate dehydrogenase 1–like have been shown to associate with the clinical phenotype of the Brugada Syndrome, a channelopathy usually due to mutations within the sodium channel (234). Being glycerol-3-phosphate dehydrogenase involved in NAD-dependent energy metabolism, this result was interpreted as a NADH-mediated regulation of human cardiac sodium channels (Nav1.5) whose dysregulation may predispose to arrhythmias (232). In the same line, increased intracellular NADH levels in HF may contribute to the reduced I_Na,p in this condition. Besides affecting the peak sodium current, redox modifications of the gating mode of the channel may lead to increased persistent or late sodium current flowing during the plateau phase (late sodium current [I_Na,L], Fig. 9). Increased I_Na,L might result from multiple reopening of a small fraction of noninactivating sodium channels (240) and is promoted by several factors such as hypoxia (184), and NO (3) H₂O₂ and ditiothreitol have opposite effects on peak and late sodium current, by respectively decreasing or increasing I_Na,p/I_Na,L ratio (237).

OPEN IN VIEWER

FIG. 10.

Cardiac sodium channel structure and (dys) function. (Upper panel) Schematic representation of the α-subunit of Nav1.5 and interacting proteins. The α subunit Nav1.5 consists of four homologous domains termed I–IV, each domain contains six transmembrane segments termed S1–S6; the S4 helices serve as voltage sensors, and between S5 and S6 there is a hairpin-like P-loop that comprises part of the channel pore. The amino and carboxy termini are intracellular. The intracellular loop between D-III and D-IV contains IFM residues, which are key amino acids for fast inactivation gating. Several proteins are reported to interact with Nav1.5, among which ankirin, calmodulin, and syntrophin, here represented schematically with their approximate binding sites. By disrupting the fast inactivation gating process—either directly or indirectly—ROS/RNS can enhance the so-called late sodium current (I_NaL), which causes a persistent net influx of sodium during the plateau of the action potential. In turn, intracellular sodium overload negatively modulates the NCX activity (see also Fig. 7) and causes cytoplasmic calcium overload. IFM, isoleucine-phenylalanine-methionine; RNS, reactive nitrogen species.

The role of I_Na,L in the pathogenesis of HF and associated drawbacks (arrhythmias and systolic/diastolic dysfunction) has emerged soundly in recent literature, and a comprehensive description is given elsewhere (242). Briefly, I_Na,L generates a net Na+ influx during the plateau phase with two major consequences: (i) unbalance between repolarizing and depolarizing currents, and (ii) altered intracellular Na+ ([Na+]i) level. The latter process is intimately linked to Ca²⁺ homeostasis, as increased [Na+]i negatively modulates the NCX activity (38) and causes cytoplasmic Ca²⁺ overload (Fig. 7). Indeed, increased I_NaL reportedly occurs in human ventricular cardiomyocytes from diseased explanted hearts (241) as well as in animal models of HF (380). In vitro models, such as H9c2 myocytes, showed that a 48-h exposure to AT-II increased I_Na,L, an effect that could be prevented by ROS scavengers or Nox inhibitors (341).

A. Energy substrate metabolism: the cytoplasmic branch

At least three main important steps can be identified in energy substrate metabolism: (i) intracellular transport, (ii) entering into mitochondria, and (iii) final oxidation. Each part of these steps cross-talks, but intracellular transport represents the limiting one.

Fatty acids are the main energy-producing substrates of the adult healthy heart. Due to their lipophylic nature, they cannot enter inside cells by simple diffusion but need to be transported. Proteins devoted to this affair constitute a family among which the fatty acid translocase (FAT/CD36) (78). FAT/CD36 is the only transporter sensitive to insulin and to contraction (236), and its transcription is stimulated by peroxisome proliferator-activated receptor α (PPARα), a key player in cardiac fatty acid metabolism (407). Both insulin and contraction, with nonexcluding mechanisms, induce FAT/CD36 translocation from an intracellular depot to the plasmalemma and its trafficking toward mitochondria, where it controls also fatty acid oxidation (77). Once having entered inside the cell and having been activated by acyl- CoA synthetase, fatty acids undergo two major fates: esterification, through glycerophosphate acyltransferase to triglycerides, or conversion by carnitine palmitoyltransferase I (CPT-I) into acylcarnitine (235). In the contracting heart, the bulk of long chain-CoA esters is taken up into the mitochondria via the carnitine shuttle and subsequently degraded in the β-oxidation pathway (353). The CPT-I represents a shuttle system for driving fatty acids from the outer to the inner membrane of mitochondria, but it does not represent the main limiting step in long chain fatty acid oxidation (46, 236, 336).

Glucose is the only substrate that can make energy without oxygen consumption; this peculiar feature makes glucose the preferential choice in conditions of reduced oxygen tension. Glucose enters the heart via two facilitative glucose transporters (GLUTs), GLUT1 and GLUT4. GLUT1 is a major mediator of basal cardiac glucose uptake and normally accounts for ∼30% of the total cardiac glucose transport in the adult heart. This contribution is higher in late fetal life and early neonatal life, whereas GLUT4 expression is typical of the adult heart which is, therefore, more insulin sensitive. Complete glucose oxidation ensures a huge amount of ATP through the glycolytic flux, which consists of a cytoplasmic (partial glucose oxidation) and a mitochondrial branch. Signaling downstream the insulin receptor is crucial to maintain glucose homeostasis.

Once activated and phosphorylated, insulin receptor stimulates a series of effectors, including the insulin receptor substrate-1 and Shc. This recruitment leads to the activation of two main pathways, the PI3K and the MAPK pathway, respectively. PI3K is a lipid kinase considered the main player of the metabolic action of insulin, whereas the MAPK pathway is involved in cell growth and differentiation. The increase in phosphoinositols at the plasma membrane induces the recruitment and colocalization of the phosphoinositide-dependent kinase 1 (PDK1), which activates protein kinase B (PKB)/Akt. Insulin is a very potent PKB/Akt activator in the heart (40). By this mechanism, insulin regulates GLUT4, translocation, and trafficking toward plasma membrane and activates the cardiac 6-phosphofructo-2-kinase isoform, the main enzyme-regulating glycolytic flux (314). Downstream to 6-phosphofructo-2-kinase is the glyceraldehyde-3-phosphate dehydrogenase activity, which catalyzes the conversion of glyceraldehyde 3-phosphate to 1,3-diphosphoglycerate, the major regulatory step in the glycolytic pathway. In fact, this enzyme is regulated by NADH/NAD⁺ and activity increases when NAD⁺ is high. Since, glycolytic enzymes are clustered near the sarcoplasmic reticulum and sarcolemma (117), it has been hypothesized that oscillations in this pathway might reflect a stress-induced need of good energy, as it occurs during acute myocardial ischemia when ATP/ADP ratio limited. Moreover, glycolytic-derived ATP would signal to mitoK_ATP triggering their opening (408), which is a typical survival mechanism against ischemic injuries. From this, it is not just a speculation the importance of maintaining a correct glycolytic flux to protect the heart against ischemia.

Once glycolysis has produced pyruvate, this substrate has to enter mitochondria, where it is phosphorylated by the pyruvate kinase, whose activity depends on NADH⁺ and acetyl-CoA coming from fatty acid β-oxidation. The activity of pyruvate dehydrogenase is regulated by a kinase whose phosphorylation reduces kinase activation. Experimental evidence suggests that activation of pyruvate dehydrogenase kinase is a cause of RV dysfunction and that reduction of such enzyme activity not only restores RV function but also enhances glucose oxidation (300). Insulin inhibits the pyruvate kinase activity. The relative pyruvate dehydrogenase/kinase activity can either determine or reflect fuel preference (carbohydrate vs. fat), occupies a pivotal position in fuel cross-talk permitting glucose oxidation, and allows the formation of mitochondrial intermediates (e.g., malonyl-CoA and citrate) that reflects fuel abundance. Fatty acid oxidation suppresses pyruvate dehydrogenase activity stimulating the pyruvate dehydrogenase kinases, which are regulated differentially by metabolite effectors with an NADH-dependent mechanism. By maintaining acyl-CoA removal by β-oxidation, upregulation of pyruvate dehydrogenase kinase facilitates the continuous uptake of long-chain fatty acyl-CoA into the mitochondria for oxidation, preventing accumulation in the cytoplasm, where it would exert deleterious effects on cardiac function (359).

In conclusion, any activator of fatty acid flux toward correct β-oxidation keeps the pyruvate dehydrogenase/kinase in homeostatic conditions. In this respect, a pivotal role might be played by PPARα (400) (Fig. 11).

FAT/CD36 is the only fatty acid transporter sensitive to insulin and to contraction (236) and its transcription is stimulated by PPARα, a key player in cardiac fatty acid metabolism (407). Both insulin and contraction, with not excluding mechanisms, induce FAT/CD36 translocation from an intracellular depot to the plasmalemma and its trafficking toward mitochondria, where it controls also fatty acid oxidation (77). In myocytes, contraction induces trafficking of vesicles containing GLUT4 or FAT/CD36 with a PI3/Akt-independent mechanism involving the activation of the AMP-activated protein kinase (AMPK) (150), retards GLUT4 endocytosis, and increases calcium, which activates the CAMK. This latter mechanism is a rapid response to energy stresses and a key reaction to stimuli requiring rapid energy supply (200). Events downstream of AMPK and CAMK lead to activation of various transcription factors. These include the nuclear respiratory factors 1 and 2, which promote expression of proteins of the respiratory chain; the metabolic regulator peroxisome proliferator-activated receptor-γ coactivator (PGC) PPAR-α, which upregulates the levels of enzymes of β oxidation; the mitochondrial transcription factor A, which activates expression of the mitochondrial genome; and the myocyte-enhancing factor 2A, the transcription factor that regulates GLUT4 expression (285). Activation of AMPK has also been shown to increase the phosphorylation and activity of endothelial NOS (eNOS) (171).

AMPK-dependent activation of GLUT4 appears to acts in synergism or in alternative to insulin, a finding of practical utility in type 2 diabetes, where physical exercise results as a potent stimulus counteracting insulin resistance. General strategies aimed at activating AMPK represent a promising approach for the treatment of HF being an energy-deficient state that gives rise to contractile dysfunction. Recent studies using metformin support this hypothesis (146) (see section VII).

B. Energy substrate metabolism: the mitochondrial branch

Mitochondrial metabolism of fatty acids accounts for approximately 60%–90% of total energy production (in the form of ATP), with carbohydrates contributing to the remaining 10%–40%. However, fatty acid metabolism requires approximately 10% more oxygen than glycolysis to produce an equivalent amount of ATP than glycolysis (59, 352). Despite its obvious benefits, β-oxidation has several drawbacks: (i) fatty acid oxidation requires more oxygen for producing energy than glycolysis; (ii) fatty acids uncouple the respiratory cycle increasing mitochondrial uncoupling proteins (UCP), making the respiratory cycle more ineffective (i.e., requiring more and more oxygen); and (iii) fatty acids can damage the plasma membrane and disturb ion channel behavior, thus favoring arrhythmogenesis (353). The NADH/NAD and NADPH/NADP from glycolysis and pentose phosphate shunt regulate the mitochondrial transmembrane potential (87). This looks of particular importance in HF, where oxygen supply is often limited and uncoupling protein, among which the cardio-specific UCP3 (126), levels change and fatty acid oxidation increases (80, 269).

Mitochondria present specific and autonomous DNA that allows self-generation biogenesis of proteins, a mechanism regulated by the function of DNA-binding transcriptional regulators often mediated by coactivators, among which the PPAR-γ coactivator-1α (PGC-1α) (220, 229, 401), which plays a pivotal role in fatty acid metabolism (activating PPAR-α, thus exerting a control on enzymes of fatty acid oxidation). Mitochondrial biogenesis and respiration are stimulated by PGC-1α through powerful induction of nuclear respiratory factor-1 and 2 gene expression. Moreover, PGC-1α increases nuclear respiratory factor transcriptional activities on the promoter of mitochondrial DNA maintenance factor, thus synchronizing activities of mitochondrial and nuclear genomes. Hearts of mice with cardiomyocyte-specific overexpression of PGC-1 α (myosin heavy chain-PGC-1 α) exhibit a marked increase in cardiomyocyte mitochondrial biogenesis and, ultimately, develop HF (219).

Recently, our knowledge of the control of some transcription factors has been further improved by the discovery of a novel system controlling their activation/deactivation (323). This system includes sirtuin1 (SIRT1), whose low SIRT1 mRNA expression in insulin-sensitive tissues reflects impaired regulation of mitochondrial function associated with insulin resistance in humans (323).

In HF, there is a downregulation of genes controlling mitochondrial biogenesis (257, 338) and of enzymes involved in fatty acid oxidation (136, 219). In our recent study (338) we documented that in different cardiomyopathies there is a specific metabolic gene expression profile caused by a different derangement of the mitochondrial energy production pathway and that in mitochondrial cardiomyopathy the metabolic gene expression profile is similar to that developing in diabetic cardiomyopathy (55, 127, 338). In this condition, increased circulating free fatty acids may lead to activation of a PPARα/PGC-1α complex, which in turn causes induction of mitochondrial biogenesis and fatty acid oxidation. Accordingly, we observed in mitochondrial cardiomyopathy, the induction of genes involved in fatty acid metabolism, glucose transport, and mitochondrial biogenesis. The latter change was consistent with a dramatic increase in mtDNA content per cell, as well as with histologic and ultrastructural features of marked mitochondrial proliferation. In contrast, downregulation of the PPARα/PGC-1α complex and its targets, as well as reduced expression of glucose transporters, was observed failing hearts with dilatative and ischemic cardiomyopathy, confirming previous observations (27, 309) (Fig. 12).

OPEN IN VIEWER

FIG. 12.

Expression of genes involved in mitochondrial biogenesis (A) and fatty acid oxidation (B) in cardiomyopathies with different etiology. For interpretation of results see text. Modified from Sebastiani et al. (302). *p < 0.001 versus NF. DCM, dilated cardiomyopathy; IHD, ischemic heart disease; LCAD, long-chain acyl-CoA dehydrogenasey; MCAD, medium-chain acyl-CoA dehydrogenase; MIC, mitochondrial cardiomyopathy; NRF, nuclear respiratory factor; PGC-1α, peroxisome proliferator activated receptor γ coactivator 1α; Tfam, transcription and mitochondrial deoxyribonucleic acid maintenance factor.

In the same study, we reported that in dilated and ischemic failing hearts an increased myocardial ROS production, mainly due to an increase of NADPH oxidase activity, was associated to a decrease of UCP expression. On the contrary, in mitochondrial cardiomyopathy, sharing similarities with diabetic hearts, a more marked increase in ROS production was observed compared to that found in cardiomyopathies with different etiology. Thus, in addition to NADPH oxidase, mitochondrial dysfunction may contribute to the increased oxidative stress both by mitochondrial-derived O₂ ⁻ and by a reduction of ATP/AMP (adenosine monophosphate) ratio, which in turn activates the XO pathways. In mitochondrial cardiomyopathy, a high level of ROS was associated to increased expression of UCP2 and UCP3 that may lead to increased mitochondrial oxygen consumption, thus amplifying energy dysfunction. Moreover, overexpression of UCP in mitochondrial cardiomyopathy is insufficient to decrease ROS production through the mechanism of uncoupling respiration, and therefore high levels of ROS are maintained (338).

In summary, complete fatty acid oxidation supplies the acyl-CoA that enters in the citric acid cycle, thus linking fatty acid oxidation (β-oxidation) to complete glucose oxidation. A correct fatty acid oxidation throughout the respiratory chain supplies the intermediary products FADH2 and NADH linking electrons to CoQ and cytochrome-b producing ATP (Fig. 13).

OPEN IN VIEWER

FIG. 13.

Main steps of fatty acid metabolism. LCFAs enter cells, thanks to the activity of transporters, namely, the insulin-sensitive transporter FAT/CD36. Once inside, LCFAs undergo activation by acetylation (Acetyl-CoA). Only the acetylated fatty acids can be transported by CPT-1 inside the mitochondrion, where they undergo β-oxidation dehydrogenation, which produces NADH and FADH2. A correct flux of fatty acids through oxidation pathways ensures coupling with Krebs's cycle. FABP, fatty acid binding protein; LCFA, long chain fatty acid; TCA, tricarboxylic acid.

C. Insulin resistance impacts glucose and fatty acid metabolism

Insulin signal dysregulation affects glucose and fatty acid metabolism deeply. It remains to be addressed whether insulin is the primer or the consequence of the shift in the energy substrate metabolism occurring in HF. In fact, on one hand, insulin resistance can be induced by a well-known pattern of inflammatory neurohumoral factors (409); on the other, by substrate excess as an increased availability of cardiac fatty acids (298), a condition that also induces the typical electrophysiological alterations observed in diabetic cardiomyocytes (153).

Insulin resistance is commonly defined as the inability of the hormone to activate its receptor-dependent cascade, which governs glucose and fatty acids intracellular metabolism in insulin-sensitive tissues. In this context chronic and/or increased production of ROS/RNS or a reduced capacity for elimination can cause the activation of multiple stress-sensitive serine/threonine kinase signaling cascades, which led to dysregulation in intracellular signaling, ultimately resulting in insulin resistance (121). The insulin signaling pathway offers a number of potential targets (substrates) of these activated kinases, including the insulin receptor and the family of insulin receptor subsstrate (IRS) proteins. As illustrated in Figure 11 activation of multiple stress-sensitive serine/threonine kinase signaling cascades, such as JNK/stress-activated protein kinase (SAPK) activity, increases the level of serine/threonine phosphorylation of the insulin receptor and IRS-1/2 and decreases the extent of tyrosine phosphorylation, resulting in the attenuation of insulin action. The serine/threonine phosphorylated forms of IRS molecules are less able to associate with the insulin receptor and downstream target molecules, especially PI 3-kinase, resulting in impaired insulin action, including PKB activation and glucose transport. More in general, the interruption of the main axes of the insulin receptor signals at the IRS-1-PI3K/Akt reducing GLUT4/GLUT1 and FAT/CD36 trafficking from endosomes to plasmalemma, producing a net but dynamic intracellular retention of GLUT4 and stabilization at plasmalemma of FAT/CD36. These modifications are responsible, at least in part, for the glucose and fatty acid metabolism impairment found in experimental (345) and human diabetes (298). On the reverse, the relationship between GLUT-mediated glucose transporters and cardiac phenotype is stressed further by results obtained in mice with cardiac knockout for GLUT4 or overexpressing GLUT1. While GLUT4 null mice exhibit cardiac hypertrophy and failure and show higher mortality (194) than wild-type mice, overexpression of GLUT1, enhancing glucose uptake and oxidation, in insulin-independent manner, reduces cardiac metabolic flexibility and makes the heart susceptible to contractile dysfunction. In particular, overexpression of GLUT1 does not allow upregulation of myocardial fatty acid oxidation when mice were fed a high-fat diet, resulting in increased oxidative stress, activation of p38 mitogen-activated protein kinase, and, more importantly, contractile dysfunction (Yan et al., 2009). It is well known that, in humans, insulin resistance at GLUT4 is associated with cardiac hypertrophy (295).

In HF there is a reexpression of fetal phenotype that includes overexpression of GLUT1 (309). A pivotal role in reexpression of GLUT1 might derive from altered expression/function of PPARα (27, 412). GLUT1 trafficking from recycling endosomes is increased by contraction and, to a lesser extent, by insulin (395). Experimental evidence confirms a role of GLUT1 as the bench warmer since its expression/activity increases when GLUT4 activity is reduced (194). Moreover, enhanced expression of GLUT1 has been associated with failure of postinfarcted rat hearts (354).

When insulin resistance emerges, glycolytic anaerobic flux interrupts at the cytoplasmic branch with accumulation of intermediary products and consequent activation of extra glycolytic pathways (i.e., polyols, pentose phosphates, and aldose reductase) (49) leading to modification of the lactate/pyruvate ratio (a measure of NADH/NAD+) and consequent reduction of protection against ischemic injuries (172).

Recent evidence suggests that activation of alternative glucose pathways also occurs in ischemic cardiomyopathy (Fig. 14). In particular, in the failing left ventricle of patients with ischemic cardiomyopathy, the activity of glucose-6-phosphate dehydrogenase was found higher than in normal donor hearts, thus enhancing NADPH levels, NADPH oxidase activity, and oxidative stress (147). Accordingly, inhibition of glucose-6-phosphate dehydrogenase activity might represent a further target of the metabolic therapy of HF (147). Another NADPH-dependent alternative pathway for glucose-6P oxidation is represented by the polyol pathway regulated by the activity of aldose reductase, which diverges glycolysis from to glyceraldehyde-3-phosphate transformation. Aldose reductase is a multifunctional enzyme with a broad substrate specificity, which overlaps with that of ubiquitous, structurally related aldehyde reductase, transforming aldehydes to the corresponding alcohol. Aldose reductase can lower intracellular NADPH levels, when it reduces glucose to sorbitol while increasing NADH (when sorbitol is oxidized to fructose). One of the main consequences of these changes is the depletion of key intracellular antioxidants (such as GSH), a finding that may expose cells to oxidative and nitrosative injury (92). The production of sorbitol from glucose increases osmotic stress, which has been implicated in the pathophysiology of diabetic complications, including cardiac disease. In this line, pharmacological inhibition of the aldose reductase exerts significant benefit against the development of various diabetic complications in animal models (10, 283). Instead, in virtue of its antioxidant features, from scavenging toxic aldehydes such as hydroxynonenale, aldose reductase activities considered a key protective component of late preconditioning against ischemia in nondiabetic rodents (343). Hence, a different role for aldose reductase activity is envisaged in normo- and hyperglycemic conditions.

OPEN IN VIEWER

FIG. 14.

Main steps generating redox imbalance in anaerobic glycolysis. Interruption of the glycolytic flux in mitochondria produces accumulation of upstream substrates and activation of alternative glucose metabolism pathways: aldose reductase and its counterpart, the sorbitol dehydrogenase, in terms of redox balance. Moreover, the activation of the pentose phosphate shunt, whose limiting step is the glucose-6-phosphate dehydrogenase, also produces NADPH which, in turn, fuels the NADPH oxidase. AR, aldose reductase; G6PDH, glucose-6-phosphate dehydrogenase; SDH, sorbitol dehydrogenase.

By accumulating upstream glycolytic products, insulin resistance reduces glycolytic flux of glucose toward its mitochondrial oxidation (Fig. 11), thus impacting on cardiomyocytes oxidative/antioxidant status substrate phosphorylative oxidation other than mitochondrial biogenesis.

As mentioned above, the mitochondrial pyruvate dehydrogenase complex catalyzing the oxidative decarboxylation of pyruvate links glycolysis to the tricarboxylic acid cycle and mitochondrial ATP production. Adequate flux through the complex is important in tissues with a high ATP requirement as the heart, in lipogenic tissues (since it provides cytosolic acetyl-CoA for fatty acid synthesis), and in generating cytosolic malonyl-CoA, a potent inhibitor of CPT-I. Conversely, suppression of pyruvate dehydrogenase complex activity is crucial for glucose conservation when glucose is scarce. Insulin resistance upregulates PDK activity (360) in the heart. This defect in pyruvate metabolism triggers mitochondrial dysfunction by multiple mechanisms.

Boudina et al. (50) have recently demonstrated that a defect of cardiac insulin signal results in reduction of the pyruvate dehydrogenase activity, in stimulation of fatty acid, and in an increase of O₂ ^{• −} ions production. All these effects disarrange respiration and trigger ROS-mediated mitochondrial dysfunction These results confirm that insulin signal is essential to maintain correct cardiac mitochondria function, which is, in turn, crucial for the management of pyruvate and fatty acid entering in the respective mitochondrial metabolism to supply ATP for sustaining inotropism (50). In addition, these experimental results sustain that mitochondrial dysfunction would be secondary and not a primary cause to insulin resistance and that mitochondrial defects could add to hyperglycemia precipitating to failure.

Moreover, at mitochondria, insulin resistance impairs intracellular Ca²⁺ homeostasis (398) and belates the activation of protective mechanisms against ischemia-induced arrhythmogenesis (177), increasing heart susceptibility to ischemia (408). As mentioned above, modification of SERCA function is an early, crucial factor leading to diastolic dysfunction in type 1 and type 2 diabetic patients (164, 410).

A defective insulin signal greatly impacts on FAT/CD36 trafficking, by stabilizing the transporter at plasma membrane level and diverting it from mitochondria. We have previously demonstrated that exposure of immortalized cardiomyocytes, HL-1 cells, to AT-II increases oxidative stress and raises insulin resistance at glucose but also at fatty acids, by stabilizing FAT/CD36 at plasma membrane (8). Either oxidative stress or insulin resistance may be ameliorated by cell pretreatment with type 1 AT-II receptor antagonists. These results seem to confirm that FAT/CD36 is likely a target of AT-II-induced oxidative stress. An increased fatty acid uptake following insulin resistance at FAT/CD36 joined to their inefficient detoxification triggers the so-called lipotoxic effect (78), which provides the link between insulin resistance and loss of mitochondrial function (298). The pathophysiology of insulin resistance is, at least in part, related to the inability of mitochondria to oxidize fatty acids.

When long chain fatty acid oxidation increases, electron leakage from mitochondria also increases, favoring ROS production. Moreover, while a correct short, medium, and long chain fatty acid acyl-CoA dehydrogenase activity supplies mitochondria of reduced nucleotides (FADH₂ and NADH, Fig. 12), their levels decrease in insulin resistant mitochondria, opening the way to ROS-induced mitochondrial dysfunction (279). Evidence demonstrates that UCP3 expression and activity are downregulated in hypertrophied and failing hearts (136, 219) but also in type 2 diabetic patients, suggesting that in the insulin-resistant heart, mitochondria are more exposed to ROS attack (80).

In conclusion, defects of the coordinated regulation of all metabolic pathways or of several pathways simultaneously could ultimately contribute in a meaningful way to the development of metabolic abnormalities, a base on which insulin resistant cardiomyopathy may develop with its pro-oxidant ground. Nevertheless, despite the aforementioned association between high fatty acid oxidation and cardiac diseases, interventions to sustain myocardial glucose oxidation and/or to prevent high fatty acid oxidation did not fulfill all the expectations in HF patients.

A. The pharmacological management of HF includes indirect antioxidants

Conventional medical therapy for HF has been outlined extensively and several classes of agents are used, among which diuretics play a fundamental role in the relief of volume overload symptoms besides positively impacting the survival of patients with HF (301). Therefore, tolerance to loop diuretics may develop with a concomitant loss of function (308), making diuretic dosage and maintenance critical for HF control and management.

Even if some data sustain a certain degree of antioxidant capacity for furosemide (346), no clinica data, up to now, encourage the identification of furosemide antioxidant features as a part of its mechanism of action in controlling HF symptoms.

1. Angiotensin-converting enzyme inhibitors

Angiotensin-converting enzyme inhibitors (ACEi) represent the most beneficial, common, and recommended therapy for HF since their use is associated with reduced renal disease risk, and cardiovascular mortality and morbidity in various at-risk patient populations (1). Several studies have definitively demonstrated that the use of ACEi exerts beneficial effects on the cardiovascular system, which goes beyond the control of high systolic pressure. In fact, ACEi, by reducing AT-II production (local and systemic) opposite to cardiac hypertrophic remodeling, reduces catecholamine overload, thus exerting protective effects on cardiomyocyte metabolism and, directly and indirectly, on their redox state (Fig. 15). In fact, animal studies suggest that most of the deleterious effect of AT-II, including cardiomyocyte apoptosis (128), are ROS mediated (28). Moreover, the beneficial effects of ACEi on cardiomyocyte metabolism include the increase of insulin sensibility by several mechanisms: (i) the reduction of local AT-II production promotes FAT/CD36 recycling from plasma membrane and decreases AT-II-dependent activation of NADPHox and (ii) the increase of bradykinin tissue levels exerts positive effects on GLUT4 activity (302). For all these reasons ACEi are the drugs of choice in the prevention of target-organ damage (for example, cardiomyopathy and nephropathy) associated with diabetes, in reduction in the risk of new onset diabetes (243), the prevention of coronary artery disease in risk patients (The PEACE Trial Investigators, 2004), and in preventing the progression of coronary artery disease toward failure (The ONTARGET investigators, 2008).

OPEN IN VIEWER

FIG. 15.

Main pharmacological targets. Membrane-bound AT1 are the target of angiotensin II, which activates NADPH oxidase. By reducing the AT-II effect or availability, respectively, AT1 antagonists (as losartan) or angiotensin converting enzyme inhibitors are effective therapeutic strategies in HF. β₁ adrenoceptors activate an intracellular cascade coupling to Gs stimulation, which increases contractility but also favors remodeling due to chronic overactivation. On the contrary, β₃ adrenoceptors are coupled to Gi that activates endothelial/neuronal NOS downstream, producing relaxation. Nebivolol blocks β₁ and activates β₃ receptors in cardiomyocytes and the vascular bed, thus reducing heart workload and counteracting ventricular remodeling. BH4 is the cofactor of NOS: increased turnover of this cofactor may cause NO dysfunction, while supplementation could amend NO deficiency. Among metabolic drugs is etomoxir, an inhibitor of CPT-1, the shuttle system for fatty acids to the mitochondrion. A correct delivery system is essential for maintaining β-oxidation and glucose complete oxidation. In insulin-resistant cardiomyocytes, this system is dysregulated. By reducing CPT-1 activity, etomoxir improves glucose oxidation. White arrows indicate stimulation, black arrows inhibition. ACEi, angiotensin converting enzyme inhibitors; AT1, angiotensin type 1 receptor.

However, the developing resistance to ACEi or the occurrence of typical adverse effects is increasingly recognized as an emerging clinical problem that reduces their effectiveness. In such a case the use of AT1 antagonists or renin inhibitors (aliskiren) may represent a suitable alternative.

B. Metabolically active antioxidants: from a failure new successes?