Outline
Reactive Oxygen Species and Antioxidant Systems, 115
Markers of Oxidative Stress in Human Heart Failure, 116
Lipid Oxidation Products, 116
Glycoprotein and DNA Products, 117
Other Oxidative Stress Markers, 117
Mechanisms of Increased Oxidative Stress in Heart Failure, 119
Oxidases, 119
Mitochondria, 120
Nitric Oxide Synthase, 120
Decreased Antioxidant Activity, 120
Nonenzymatic Auto-Oxidation, 121
Oxidative Stress and Antioxidant Therapy in Animal Models of Heart Failure, 121
Hemodynamic Overload, 121
NAD(P)H Oxidase Transgenic Mice, 121
NOS3 Knockout and Tetrahydrobiopterin, 121
Thioredoxin and N -2-Mercaptopropionyl Glycine, 121
Peroxisomal Catalase, 121
Vitamins E and C, and Resveratrol, 121
Mitochondrial Antioxidants, 122
Heart Failure in Gαq-Transgenic Mice, 122
Anthracycline-Induced Cardiomyopathy, 122
Models of Diastolic Dysfunction, 122
Other Experimental Heart Failure Models, 123
Human Trials of Antioxidant Therapies, 123
Mechanisms of ROS-Mediated Left Ventricular Remodeling and Heart Failure, 123
Left Ventricular Remodeling, 123
Molecular and Cellular Mechanisms of Reactive Oxygen Species-Induced Heart Failure, 124
Mitogen-Activated Protein Kinases, 125
Mitochondrial Signaling and Energetics, 125
Endoplasmic Reticulum Stress Response, 125
Calcium Handling, 125
Impaired Energetics, 125
Summary and Future Directions, 126
Oxidative stress is increased in heart failure (HF), and experimental studies suggest that this contributes to structural and functional changes in the heart that are central to both cardiac dysfunction and disease progression.
Reactive Oxygen Species and Antioxidant Systems ( Fig. 8.1 )
Reactive Oxygen Species
Reactive oxygen species (ROS) are a by-product of aerobic metabolism, and so the highly metabolically active myocardium is rich in ROS ( see Chapter 17 ). As in all tissues, ROS are handled in the myocardium by both soluble and enzymatic antioxidant systems. However, differences in subcellular and tissue compartmentalization of ROS, as well as levels of ROS, contribute to the downstream effects. It is now clear that ROS signaling pathways are complex and in many cases essential for normal signal transduction and cardiovascular physiology. “Oxidative stress” occurs when the production of ROS exceeds the capacity of antioxidant defense systems. In general, ROS cascades begin with the formation of superoxide anion (O 2 − ) by either enzymatic or nonenzymatic one-electron reduction of molecular oxygen. The unpaired electron in O 2 − is an unstable free radical that reacts with itself and other oxygen-containing species, and directly or indirectly with organic molecules—including lipids, nucleic acids, and proteins—ultimately leading to regulation or disruption of cellular functions. All aerobic organisms, from bacteria to man, have evolved a complex antioxidant defense system of enzymatic and nonenzymatic components to defend against the unavoidable formation of ROS. In parallel, there has been the evolution of specific ROS-generating systems that are used both in the immune system, where the toxicity of ROS is exploited to fight infectious organisms, as well as in all cell types, where ROS act as signaling intermediates for the purpose of triggering specific intracellular processes.
Primary Antioxidant Systems
Primary antioxidant enzymes, defined here as those that directly interact with ROS, include superoxide dismutase (SOD), catalase, and other peroxidases. These enzymes work in parallel with nonenzymatic antioxidants to protect cells and tissues from ROS. The mitochondrial enzymes manganese superoxide dismutase (MnSOD) and glutathione peroxidase (GPx) appear to be the most important in controlling myocardial levels of O 2 − and H 2 O 2 . Approximately 70% of the SOD activity in the heart, and 90% of that in the cardiac myocyte, is attributable to MnSOD (SOD2). The remainder consists of cytosolic Cu/ZnSOD (SOD1) , with less than 1% contributed by extracellular SOD (ECSOD, SOD3). This is in contrast to other organs, where Cu/ZnSOD plays a greater role. The relative importance of MnSOD in the regulation of oxidative stress in the myocardium is highlighted by the demonstration that mice deficient in MnSOD die soon after birth with dilated cardiomyopathy. In contrast, mice deficient in CuZnSOD or ECSOD have no overt myocardial phenotype. As the only SOD located in the mitochondria, MnSOD plays a critical role in the control of mitochondrial ROS generated during normal oxidative phosphorylation (see later discussion). The phenotype of the MnSOD knockout mouse therefore also underscores the importance of the mitochondria as a source of ROS in the myocardium.
H 2 O 2 , the product of SOD, is handled by catalase and/or one of several GPxs. Catalase is expressed in the cytosol, where it is located primarily in peroxisomes (pCAT) and in the mitochondria (mCAT). Catalase contains four porphyrin heme groups that interact with H 2 O 2 to facilitate its decomposition to water and oxygen. Transgenic expression of either pCAT or mCAT exerts beneficial effects on cardiac structure and function in a variety of animal HF models, suggesting that H 2 O 2 is an important oxidant species in the failing heart. Compared with O 2 − , H 2 O 2 is longer lasting and able to cross cellular membranes.
GPx are a family of selenium-containing enzymes that catalyze the removal of H 2 O 2 through oxidation of reduced glutathione (GSH), which is recycled from oxidized glutathione (GSSG) by the nicotinamide adenine dinucleotide phosphate (NADPH)–dependent glutathione reductase (GRed). GPx-1 is encoded on the nuclear genome but localizes both to the cytosol and mitochondria.
Ancillary Antioxidant Mechanisms
Several enzymes and molecules play a supporting role in regulating oxidative stress and/or the effects of oxidants on targets. The activity of GPx requires stoichiometric quantities of glutathione (GSH), and therefore decreased levels of GSH inhibit the activity of GPx. GRed requires NAD(P)H as a reductant to recycle GSSG to GSH. Glutathione is also a direct scavenger of reactive oxygen and nitrogen species. Cells replenish GSH both by de novo GSH synthesis and through the action of glutathione reductase on GSSG. In this context, enzymes in the pentose phosphate pathway and glucose 6-phosphate dehydrogenase (G6PD), the rate-limiting enzyme in this pathway, are ancillary antioxidant enzymes that are critical to cellular antioxidant defenses. In mice, chronic GSH depletion causes worse LV function, fibrosis, and survival in response to pressure overload, and GSH supplementation ameliorates the phenotype. Increased oxidative stress lowers the GSH/GSSG ratio, which may thus be used as a measure of myocardial oxidative stress. Unlike the levels of lipid peroxidation products, which can reflect oxidative stress inside and/or outside of the cell, the GSH/GSSG ratio specifically measures intracellular oxidative stress. Other thiol-containing proteins, such as metallothionein , have significant antioxidant functions through direct scavenging of ROS and may play a role in the failing heart. Overexpression of metallothionein suppresses mitochondrial oxidative stress, cardiac apoptosis, and the development of diabetic cardiomyopathy (see also Chapter 17 ).
Several thiol-containing proteins modulate the effects of ROS. The thioredoxin (Trx) system consists of Trx reductase, Trx, peroxiredoxin, and NADPH ( Fig. 8.2 ). This system, like HO-1 and metallothionein, is induced by a variety of oxidative stresses, and Trx levels are increased in the myocardium and blood of patients with HF. Trx exerts an antioxidant effect by scavenging ROS, but also interacts directly with multiple cellular signal transduction pathways and activates HO-1 and Bcl-2. Thioredoxin-interacting protein (Txnip) interacts with Trx via a disulfide bond, decreasing thioredoxin activity and thus increasing oxidative stress.
The glutaredoxin (Grx) system is similar in organization and function to the Trx system, but has a much higher selectivity for cysteine thiols that are glutathiolated, and thus may play an important role in the regulation of protein function via glutathionylation.
The peroxiredoxins (Prx) are a family of antioxidant enzymes that reduce H 2 O 2 and which may also interact with Trx and Grx, from which they accept electrons in order to be active. Although Prxs are abundant, their precise role in the heart remains to be determined.
Heme oxygenase-1 (HO-1) is induced by oxidative stress and serves a cytoprotective function through the breakdown of pro-oxidant heme into equimolar amounts of carbon monoxide, biliverdin/bilirubin, and free ferrous iron. Carbon monoxide and bilirubin exert direct cardioprotective effects via their respective anti inflammatory and antioxidant actions.
Vitamin antioxidants play a role in the control of ROS cascades and the prevention of free radical chain reactions. Vitamin E (α-tocopherol) and vitamin C (ascorbic acid) prevent lipid peroxidation and membrane breakdown. α-Tocopherol is a fat-soluble vitamin that concentrates in cellular membranes. Through the aromatic ring head group, α-tocopherol is able to form a “stable” tocopheryl radical when it reacts with ROS and lipid peroxy radicals. Ascorbic acid reacts with tocopheryl radicals, thereby converting them back to tocopherol. Circulating and tissue levels of α-tocopherol have been used as measures of both antioxidant capacity and oxidative stress.
Markers of Oxidative Stress in Human Heart Failure
Oxidation products of several organic molecules, including lipids, proteins, and nucleic acids, have been used to assess oxidative stress in HF ( Table 8.1 ). While these methods can individually be criticized for their relative lack of specificity, collectively the data support the conclusion that there is increased systemic and myocardial oxidative stress in patients with HF.
Markers | Studies | Findings |
---|---|---|
MDA, Thiols, SOD, GPx | McMurray et al. | Increased plasma MDA, decreased plasma thiols in patients with CAD and LV dysfunction |
Belch et al. | Increased plasma MDA, decreased plasma thiols in patients with CHF; weak correlation between decreasing thiols and worsening LV function | |
McMurray et al. | Increased MDA, decreased plasma thiols in both CAD and non-CAD patients with HF | |
Ghatak et al. | Increased MDA, and reduced erythrocyte SOD and GPx activities in CHF; weak correlation with LV function; improved by vitamin E administration | |
Radovanovic et al. | Increased MDA in CHF; MDA predicted death | |
Keith et al. | Correlated oxidative stress with NYHA clinical class, as well as levels of soluble TNF receptor levels as marker of prognosis | |
Mak et al. | Plasma unsaturated aldehydes including 4-OH-nonenal were elevated in association with impaired LV contractile function in HF patients | |
Polidori et al. | Plasma MDA was higher in patients with severe CHF than those with moderate disease | |
Campolo et al. | Reduced cysteines and MDA were increased in CHF patients | |
MPO | Tang et al. | MPO was associated with an increased likelihood of more advanced HF, and predictive of worse long-term clinical outcomes |
Tang et al. | MPO predicted development of HF in aging subjects | |
Reichlin et al. | MPO predicted higher mortality in acute decompensated HF | |
8-Isoprostanes | Mallat et al. | Pericardial 8-isoprostanes in patients undergoing open-heart surgery correlated with preoperative NYHA functional class |
Kameda et al. | Pericardial 8-isoprostanes correlated with LV end-diastolic volume, and activities of MMP-2 and MMP-9 and gelatinolysis in CAD patients | |
Polidori et al. | Plasma F2 isoprostanes correlated with antioxidant status and NYHA class | |
Nonaka-Sarukawa et al. | Urinary 15-Ft-isoprostanes increased in proportion to the severity of CHF, and correlated with plasma BNP and serum IL-6 | |
PON-1 | Kim et al. | PON-1 activity was reduced in HF |
Tang et al. | Indirect measure of PON-1 activity independently predicted high risk of cardiac events in systolic HF | |
Glutathione S-transferase | Andrukhova et al. | Serum levels of glutathione S-transferase are associated with EF and NYHA class |
Breath pentane | Sobotka et al. | Increased exhaled pentane in CHF compared with controls; pentane levels reduced by captopril therapy |
Pentosidine | Koyama et al. | Serum pentosidine was an independent risk factor for cardiac events in patients with HF |
Plasma carbonyls | Amir et al. | Serum oxidative stress levels increased with NYHA functional class and were associated with higher CRP and BNP |
Radovanovic et al. | Increased plasma carbonyls in CHF; carbonyls correlated with echocardiographic remodeling indices | |
8-Hydroxy-2’-deoxyguanosine | Watanabe et al. | Correlated with NYHA functional class, left atrial diameter, LV end-diastolic diameters, LV end-systolic diameters, and plasma BNP |
Kono et al. | Increased in both the serum and myocardium of patients with HF; reduced with carvedilol therapy | |
Pignatelli et al. | Serum levels increased in CHF progressively from class I-II to class III-IV; correlated with TNF-α and sCD40L | |
Kobayashi et al. | Urinary levels correlated with symptoms and functional severity of HF | |
Nagayoshi et al. | Higher levels in CAD vs. non-CAD HF patients despite same NYHA class | |
Uric acid | Cicoira et al. | Elevated serum uric acid levels correlated with parameters of diastolic dysfunction in HF |
Anker et al. | Uric acid predicted mortality in patients with moderate to severe CHF | |
Kojima et al. | Correlated with Killip classification and mortality in patients with acute MI | |
Sakai et al. | Positive transcardiac gradient increased with the severity of HF, and inversely correlated with LV ejection fraction | |
Kittleson et al. | High uric acid was associated with increased cardiac filling pressure and reduced cardiac index; correlated with NT-proBNP | |
Ioachimescu et al. | An independent predictor of death in patients at high risk of cardiovascular disease | |
Manzano et al. | Independent predictor of mortality in the SENIORS trial patients with systolic and diastolic HF | |
Bishu et al. | Elevated levels in acute decompensated HF | |
Oxidized LDLs | Charach et al. | Plasma levels predicted mortality and morbidity in HF patients |
Phospholipid transfer protein | Chen et al. | Higher levels associated with LV systolic function |
Biopyrrins | Hokamaki et al. | Urinary biopyrrin levels were elevated, and correlated with blood BNP and severity of HF |
Ceruloplasmin | Dadu et al. Saunders et al. | Higher levels correlate with incident HF even after adjustment for other biomarkers |
Serum free thiols | Koning et al. | Higher levels correlate with lower levels of NT-proBNP, decreased rehospitalization rate, and increased patient survival |
Lipid Oxidation Products
Lipid peroxidation products, such as malondialdehyde (MDA) and 4-hydroxy-nonenal , are increased, and total thiol levels are decreased in patients with ischemic and nonischemic cardiomyopathy compared with subjects without HF. MDA is an independent predictor of death in patients with chronic HF. Exhaled pentane, a volatile lipid peroxidation product, is increased in patients with HF. The 8-isoprostanes (8-iso-prostaglandin F 2α ) are a family of prostaglandin F 2α isomers formed by the peroxidation of arachidonic acid through a non-cyclooxygenase-mediated reaction catalyzed by free radicals. In contrast to reactive aldehydes, lipid hydroperoxides, and conjugated dienes, 8-isoprostanes are more stable products of lipid peroxidation and thus may be a more useful indicator of oxidative stress. 8-Isoprostanes measured in the pericardial fluid of patients with HF undergoing open heart surgery correlated with increasing New York Heart Association (NYHA) functional class. Plasma and urinary isoprostanes correlate with the clinical severity of HF, antioxidant status, and blood B-type natriuretic peptide (BNP). Plasma oxidized LDL (oxLDL), a marker of oxidative stress, is elevated in HF patients and predicts mortality and morbidity independent of conventional factors.
Glycoprotein and DNA Products
Oxidation products of glycoproteins and nuclear DNA are increased in patients with HF. Serum pentosidine , an advanced glycation end product, is an independent risk factor for cardiac events in patients with HF. Plasma carbonyls are increased in patients with more symptomatic HF, and associated with higher C-reactive protein (CRP) and BNP, as well as progressive LV remodeling. 8-Hydroxy-2-deoxyguanosine (8-OHdG), formed when DNA is oxidatively damaged, is increased in myocardium of animals with cardiac hypertrophy and tachycardia-induced cardiomyopathy, and in both the serum and myocardium of patients with HF. Plasma levels of 8.OHdG correlate with NYHA functional class, cardiac function, and several other biomarkers such as plasma BNP, tumor necrosis factor-α (TNF-α), and sCD40L, while urinary 8-OHdG levels correlate with symptoms and functional severity of HF.
Other Oxidative Stress Markers
Uric acid, produced by the ubiquitous ROS-generating xanthine oxidase, is released from the failing human heart in inverse relation to LV ejection fraction. Increased serum uric acid levels are associated with worse hemodynamic function and correlate with plasma NT-proBNP in patients with HF. Uric acid was an independent predictor of mortality in the SENIORS trial, which included patients with both systolic and diastolic HF. Uric acid is elevated in patients presenting with acute decompensated HF. However, in high-risk HF patients with reduced ejection fraction and elevated uric acid levels, xanthine oxidase inhibition with allopurinol failed to improve clinical status, exercise capacity, quality of life, or left ventricular ejection fraction (LVEF).
Biopyrrins are oxidized metabolites of bilirubin, which are increased in HF, probably secondary to hepatic dysfunction and/or increased HO-1 activity. Urinary biopyrrin levels are elevated and correlate with blood BNP and the severity of HF.
Free thiols , which are a part of the antioxidant machinery, are oxidized in the presence of excess ROS, leading to depletion of their plasma levels. This phenomenon has been observed in patients with HF.
Myeloperoxidase (MPO), a peroxidase enzyme abundant in granulocytes, is increased in the circulating blood of patients with HF and is an independent predictor of death, heart transplantation, or HF hospitalization. Among patients with acutely decompensated HF, elevated MPO concentrations are associated with a higher 1-year mortality, even when adjusted for BNP levels. MPO predicts the risk of developing HF in aging individuals.
Ceruloplasmin , an acute-phase reactant and protein involved in cooper transport, has been shown to have ferroxidase I activity that is responsible for the conversion of reactive Fe 2+ into Fe 3+ , thereby preventing Fe 2+ from participating in the generation of hydroxyl radicals, GPx activity, and the ability to inhibit MPO. Elevated plasma ceruloplasmin levels are associated with incident HF, even after adjusting for other biomarkers, such as BNP, troponin, and CRP. While ceruloplasmin, per se, was not associated with mortality in HF, low serum ferroxidase I activity, presumably as a result of ceruloplasmin nitration, was a predictor of HF mortality.
Glutathione S-transferase P1 participates in the detoxification of ROS and maintenance of the cellular redox state, and is elevated in the plasma of patients with HF in proportion to systolic dysfunction and functional class.
Thioredoxin 1 (Trx1) , part of the thioredoxin antioxidant defense system described previously, decreases oxidative stress by reducing peroxiredoxin, which in turn reduces H 2 O 2 . Trx1 was shown to be an independent predictor of cardiac events survival in HF patients.
Paraoxonase-1 (PON-1) is a high-density lipoprotein (HDL)-associated glycoprotein that contributes to the systemic antioxidant activities of HDL. PON-1 is decreased in HF patients ; decreased serum arylesterase activity, a measure of diminished antioxidant properties of PON-1, predicts higher risk of long-term adverse cardiac events in patients with systolic HF.
Phospholipid transfer protein (PLTP) modulates lipoprotein metabolism and plays a role in inflammation and oxidative stress. A higher PLTP activity is associated with depressed LV systolic function.
Mechanisms of Increased Oxidative Stress in Heart Failure ( Fig. 8.3 )
Increased oxidative stress may occur as a result of the increased generation of ROS, decreased clearance of ROS by various antioxidant systems, or both. Both mitochondria and several enzyme systems that generate O 2 − may produce pathophysiologic amounts of O 2 − in the failing heart.
Oxidases
The NAD(P)H oxidase (NOX) family consists of at least five transmembrane enzymes that mediate electron transfer from NAD(P)H to molecular oxygen to generate O 2 − . The NOX-2 isoform was first described in the neutrophil, where it is responsible for the oxidative burst, which produces large amounts of cytotoxic ROS. Other isoforms produce lower levels of ROS that can act as signaling intermediates. NOX-2 and NOX-4 are the predominant isoforms in cardiac myocytes, and both have been implicated in mediating hypertrophy. NOX-2 is located in the plasma membrane, whereas the localization of NOX-4 is less certain but appears to include the endoplasmic reticulum (ER) and/or mitochondrial membranes. NOX-2 mediates LV hypertrophy and failure in response to activation of the renin-angiotensin system, but not in response to pressure overload. NOX-4 also appears to contribute to oxidative stress and to be involved in mediating cardiac hypertrophy and failure in response to hemodynamic stress, though the data are conflicting.
Xanthine oxidoreductase consists of two interconvertible forms, xanthine dehydrogenase and xanthine oxidase, both of which are involved in the conversion of hypoxanthine and xanthine to uric acid. The constitutive xanthine dehydrogenase uses NAD+ primarily as an electron acceptor, whereas the inducible xanthine oxidase transfers electrons to molecular oxygen, yielding ROS. In addition, xanthine oxidoreductase can generate O 2 − via NADH oxidase activity and produce NO via nitrate and nitrite reductase activities. Thus activation of xanthine oxidoreductase may cause both oxidative and nitrosative stress. The expression of xanthine oxidase is increased in the hearts of rats with HF and patients with dilated cardiomyopathy. In animal models of HF, xanthine oxidase inhibitors (e.g., allopurinol, oxypurinol, febuxostat) attenuate the production of ROS, improve cardiac function, decrease LV size, improve β-adrenergic receptor sensitivity, improve myocardial energetic coupling, inhibit fetal gene expression, and improve Ca 2+ handling. In rats with pressure overload, the magnitude of improvement in cardiac function with oxypurinol is related to the initial level of xanthine oxidase activity.
In patients with HF, allopurinol reduced plasma MDA, improved endothelium-dependent flow-mediated response, reduced myocardial oxygen consumption, and improved myocardial efficiency. In acute and short-term human studies, oxypurinol increased LV ejection fraction and reduced LV end-diastolic volume. However, in a dedicated randomized controlled study of HF patients with elevated uric acid levels (EXACT-HF study), chronic allopurinol treatment failed to lead to an improvement in clinical status, exercise capacity, quality of life, or LVEF.
Mitochondria
Mitochondria are an important source of myocardial ROS in the failing heart. Under normal circumstances, a small fraction of the electrons entering the electron transport chain “leak” to molecular oxygen forming O 2 − . Under pathologic conditions, the leakage may increase, overwhelming the capacity of the mitochondrial antioxidant system. In dogs with rapid pacing-induced HF, electron paramagnetic resonance (EPR) spectroscopy showed a 2.8-fold increase in the rate of O 2 − formation in a mitochondrial fraction of the heart, thus providing direct evidence for increased mitochondrial ROS generation. Increased mitochondrial H 2 O 2 production was directly demonstrated in mitochondria isolated from db/db mice, a model of obesity, and type 2 diabetes with a characteristic cardiomyopathy. Increased ROS production was associated with ETC uncoupling, and appeared to be derived from complex I and other complexes. Indirect support for mitochondrial ROS production comes from studies in transgenic mice that overexpress MnSOD or mCAT in mitochondria. In mice with type 1 diabetes, cross-breeding with MnSOD transgenic mice improved mitochondrial respiration and mass and ameliorated abnormalities in cardiac structure and function. Likewise, as discussed later, transgenic expression of mCAT improved several aspects of cardiac structure and function in mice with pathologic cardiac remodeling and/or HF due to pressure overload (aortic constriction), systolic failure due to overexpression of Gαq or angiotensin infusion, and ameliorated cardiac aging. Deletion of mitochondrial thioredoxin reductase 2, a component of antioxidant defense, also leads in increased mitochondrial ROS production, dysregulated autophagy, and cardiomyopathy.
The mitochondrial enzyme, monoamine oxidase-A (MAO-A), has recently emerged as a potent generator of ROS in the heart—in particular, H 2 O 2 . Cardiac MAO-A expression is increased in models of HF and cardiac aging. Overexpression of MAO-A in the heart is associated with aging and HF, resulting in significant mitochondrial and lysosomal dysfunction and cardiac damage. This is thought to be mediated by ROS-dependent inhibition of nuclear translocation of transcription factor-EB, a master regulator of autophagy and lysosome biogenesis.
A pathologic role of mitochondrial ROS has also been confirmed in obesity-mediated cardiomyopathy. Wild-type mice fed a high-fat, high-sucrose diet develop a cardiomyopathy characterized by LV hypertrophy and diastolic dysfunction. In this model, there is an increase in mitochondrial ROS production, particularly from complexes I and II, that precedes the development of obesity. Importantly, overexpression of mCAT prevents both mitochondrial and nonmitochondrial abnormalities in this model and restores a near normal cardiac phenotype, suggesting a major mechanistic role of mitochondrial ROS in the pathogenesis of obesity-mediated cardiomyopathy. Although these studies show that mitochondrial ROS is a key mediator of cardiac dysfunction, they do not absolutely establish the source of the ROS. For example, ROS from nonmitochondrial sources may cause mitochondrial dysfunction and may include ETC electron leakage, leading to further ROS generation (i.e., ROS-mediated ROS release).
Nitric Oxide Synthase
Nitric oxide (NO) is a free radical that can modify the myocardial response to oxidative stress both directly and indirectly. NO is synthesized in the conversion of l -arginine to l -citrulline by a family of nitric oxide synthases (NOS). NO, a free radical gas, is buffered in the cell by reactions with glutathione, and reacts reversibly with sulfhydryl groups of proteins forming S -nitrosothiols that can alter protein function. Through chemical reactions with ROS, NO can either decrease or increase the oxidative stress in a cell or tissue. Under normal circumstances, myocardial NO is produced at low levels by endothelial NOS (eNOS or NOS3) . NOS3, which is present in virtually all cell types in the myocardium, including myocytes, fibroblasts, and endothelial cells, is regulated by a calcium-sensitive interaction with calmodulin. Inducible nitric oxide synthase (NOS2) is not regulated by Ca 2+ and when induced is capable of producing high levels of NO. Though NOS2 is expressed minimally in the normal myocardium, it is induced by exposure to cytokines, hypoxia, and other stimuli in both myocytes and nonmyocytes, leading to a marked increase in the production of NO. NOS2 can catalyze the formation of O 2 − , particularly in the setting of arginine depletion, and may contribute directly to the formation of ROS. The expression and activity of NOS2 are increased in the myocardium of patients with idiopathic and ischemic dilated cardiomyopathies.
Low levels of NO, as are formed by NOS3, may decrease the level of oxidative stress by decreasing the production of O 2 − through inhibition of oxidative enzymes. Through the activation of guanylate cyclase, NO can inhibit signaling and transcription factors that modify myocyte hypertrophic and apoptotic signaling. Mice lacking NOS3 have worse ventricular function late after MI, consistent with the notion that NOS3-derived NO is beneficial for the failing heart. Higher levels of NO increase oxidative stress by reacting with O 2 − to generate peroxynitrite (ONOO − ) , a free radical that is toxic and longer-lived than either NO or O 2 − . ONOO − can react with many cell constituents, including tyrosine residues of susceptible proteins such as MnSOD, causing irreversible inactivation. Based on the relative rate constants for the reaction of O 2 − with SOD versus NO, the formation of ONOO − is favored when the levels of O 2 − and/or NO are high or the level of SOD is low.
High concentrations of NO can have direct toxic effects on cardiac myocytes in vitro. The cytotoxic effect of cytokines on cardiac myocytes in culture is inhibited by the NOS inhibitor l -NMMA. Cytokine-induced apoptosis can be prevented by either an inhibitor of NOS2 or an SOD-mimetic, thus implicating ONOO − . In mice, late post-MI NOS2 expression is increased in the remote myocardium, and NOS2 knockout mice have less myocyte apoptosis, improved contractile function, and increased survival. Likewise, in animal models of autoimmune and viral myocarditis, the amount of myocardial injury was reduced by aminoguanidine, an inhibitor of NOS2. Treatment with aminoguanidine also decreased the amount of O 2 − anion formed, suggesting the presence of NOS2-dependent O 2 − production or the inactivation of MnSOD by ONOO − . The divergent effects of NO were revealed when mice with viral myocarditis were treated with l -NAME. A low dose improved survival, HF, and myocardial necrosis, whereas the highest dose worsened survival. Similarly, stretch-induced myocyte apoptosis is inhibited by an NO donor. Potential mechanisms for a protective effect of low NO levels include inhibition of enzymes in the programmed cell death pathway and decreased mitochondrial ROS production.
NOS can be uncoupled due to oxidation of the essential cofactor tetrahydrobiopterin, leading to the generation of O 2 − . ROS production from uncoupled eNOS contributes to HF and diastolic dysfunction, and the cardiac phenotype is improved by supplementation with tetrahydrobiopterin. Uncoupling of NOS may be caused by ROS from other sources (e.g., mitochondria or other oxidases), thus providing a mechanism for amplification of ROS.
Decreased Antioxidant Activity
ROS are exquisitely regulated by a large number of interacting and, to some extent, redundant antioxidant systems. Impaired function of one or more of these antioxidant systems can lead to an increase in oxidative stress. In guinea pigs with pressure overload due to aortic banding, both SOD and GPx activity decreased during the progression to HF, in association with a decrease in the ratio of GSH/GSSG, indicating an increase in myocardial oxidative stress. Similar changes in antioxidant capacity and oxidative stress occur in the rat heart late after myocardial infarction (MI). Decreased antioxidant enzyme capacity and depletion of vitamin E occurred in a large animal model of volume overload–induced HF secondary to mitral regurgitation. Two studies found no decrease in SOD or GPx activity in pathologic samples from explanted human hearts at the time of cardiac transplant, whereas in another study, MnSOD activity was reduced in the failing human heart, apparently due to a posttranscriptional level mechanism because mRNA expression was not decreased.
Sirtuins, a seven-member family of NAD + -dependent histone deacetylases, have recently emerged as major regulators of multiple native antioxidant defense systems. Sirt1, which is predominantly localized to the nucleus, can deacetylate and activate FoxO1 to upregulate expression of antioxidants, including MnSOD, catalase, and Trx1, as well as a number of antiapoptotic factors. Deacetylation of p66Shc, a master regulator of ROS, by Sirt1 has been shown to be a key protective mechanism against diabetes-medicated vascular oxidative stress. In addition to its ability to regulate redox stress, the activity of Sirt1 in turn can be redox regulated and decreased by oxidative posttranslational modifications (OPTM) including S-glutathiolation. Similarly, Sirt3, another well-studied member of sirtuin family that is located in mitochondria, has been shown to block cardiac hypertrophy by decreasing ROS production via activation of FoxO3-dependent antioxidant systems, including MnSOD and catalase, as well as via suppression of Ras activation and inhibition of the MAPK-ERK and PI3K–Akt signaling pathways. While Sirt5, Sirt6, and Sirt7 play protective roles against ROS-induced injury, their role in the heart is not well documented.
Nonenzymatic Auto-Oxidation
Reactions involving several organic molecules may contribute to the formation of ROS in vivo. Oxidation of norepinephrine and epinephrine to adrenochrome and O 2 − has been proposed as a mechanism for myocardial injury in the presence of chronic adrenergic stimulation. Thiol compounds, including cysteine and GSH, can auto-oxidize to form O 2 − , particularly in the presence of transition metals such as iron. The cardiotoxicity of iron overload is likely a combination of this plus Fenton chemistry to generate hydroxyl radicals. Myoglobin can autoxidize from oxymyoglobin to metmyoglobin with the release of O 2 − .
Oxidative Stress and Antioxidant Therapy in Animal Models of Heart Failure
Work in animal models has allowed greater insight regarding the sources and consequences of oxidative stress in HF. Studies have been performed in a variety of models of HF, with a range of antioxidant strategies.
Hemodynamic Overload
The aortic constriction model of HF occurs in two phases. In the initial days after constriction there is LV hypertrophy, whereas over the ensuing weeks there is progressive chamber dilation, wall thinning, and systolic failure. Several types of antioxidant intervention have been shown to ameliorate one or both phases of this model.
NAD(P)H Oxidase Transgenic Mice
The NOX-2 and NOX-4 isoforms of NAD(P)H oxidase are predominant in cardiac myocytes. In mice, NOX-2 knockout prevented cardiac hypertrophy and fibrosis caused by chronic sub-pressor infusion of angiotensin, suggesting a role in the response to neurohormones. However, NOX-2 knockout did not inhibit the hypertrophic response to pressure overload with aortic constriction. NOX-4 appears to modulate the hypertrophic response to pressure overload, although the exact role remains unclear. In one study, NOX-4 knockout mice did better after aortic constriction, and NOX-4 transgenic mice did worse. NOX-4 overexpression was associated with increased mitochondrial ROS and worse mitochondrial function. The latter was supported by another study where NOX-4 transgenic mice had significantly higher production of ROS, and manifested cardiac interstitial fibrosis. In that study, the mechanism seemed to be related to the upregulation of cardiac fetal genes and activation of the Akt-mTOR and NFκB signaling pathways. However, another group using a similar aortic constriction model found that NOX-4 knockout mice developed worse LV failure with aortic constriction, whereas NOX-4 overexpressing animals were protected. These experiments illustrate the complexity of redox systems, which on the one hand can mediate pathophysiologic processes and on the other are essential to normal cellular and defense mechanisms. For example, knockout of both NOX-2 and NOX-4 exacerbated ischemia-reperfusion injury in the heart, suggesting that low-level ROS production by the NOX system is required for the activation of protective mechanisms.
NOS3 Knockout and Tetrahydrobiopterin
Aortic constriction causes less LV hypertrophy and fibrosis in NOS3 knockout mice, suggesting that uncoupled NOS3 contributes to myocardial oxidative stress. Likewise, treatment with oral tetrahydrobiopterin (BH4), a cofactor necessary for NOS coupling, recoupled NOS3, decreased oxidative stress, and decreased both LV hypertrophy and fibrosis.
Thioredoxin and N -2-Mercaptopropionyl Glycine
N -2-mercaptopropionyl glycine (MPG) is a low-molecular-weight antioxidant analogue of glutathione, which scavenges H 2 O 2 . Administration of MPG to mice decreased markers of myocardial oxidative stress and myocardial hypertrophy 7 days after aortic constriction. In transgenic mice, cardiac-specific overexpression of a dominant-negative Trx-1 worsened, and overexpression of wild-type Trx-1 ameliorated the response to aortic constriction. Consistent with a protective effect of Trx-1, cardiomyocyte-specific knockout of Txnip, which inhibits Trx, protected from early, but not late, pressure overload–induced cardiac dysfunction.
Peroxisomal Catalase
Left ventricular (LV) hypertrophy and failure associated with a generalized increase in myocardial oxidative stress and specific oxidative modifications of SERCA at cysteine 674 and tyrosine 294/5 due to ascending aortic constriction is ameliorated by myocyte-specific overexpression of pCAT. Furthermore, SERCA cysteine modifications were also prevented. This is consistent with the ability of pCAT to reverse aging-associated myocardial dysfunction via decrease in cysteine OPTM.
Vitamins E and C, and Resveratrol
In guinea pigs subjected to aortic banding, vitamin E (α-tocopherol) had no effect on LV hypertrophy early after banding, but at late time points prevented the transition to HF. In the same model, vitamin C (ascorbic acid) administration to guinea pigs with aortic banding inhibited LVH. Resveratrol, a naturally occurring polyphenol antioxidant present in grapes, decreased LV hypertrophy and improved diastolic relaxation early after aortic constriction in the rat model.
Mitochondrial Antioxidants
Transgenic overexpression of mCAT or administration of an antioxidant peptide targeted to mitochondria exerts broad beneficial effects in mice with aortic constriction, resulting in less LV hypertrophy and progression to failure. The effects of mCAT were associated with correction of the mitochondrial proteome. Euk-8 is a potent SOD-mimetic, catalase-mimetic, and free radical scavenger. In mice with aortic constriction, Euk-8 decreased LV hypertrophy and progression to failure and improved survival. Elamipretide, a cell-permeable and mitochondrial inner membrane-targeting peptide, is known to protect mitochondrial cristae structure, reduce oxidative stress and improve ATP synthesis. Elamipretide has been shown to ameliorate LV hypertrophy/cardiac failure in a model of angiotensin II-induced cardiomyopathy and pressure overload, likely via a decrease in mitochondrial ROS generation.
Heart Failure in Gαq-Transgenic Mice
The G protein Gαq mediates many of the effects of neurohormones and mechanical strain in cardiac myocytes, and its overexpression in mice leads to pathologic myocardial remodeling that progresses to a severe dilated cardiomyopathy. Myocyte-specific overexpression of catalase that is expressed primarily in the peroxisomes (pCAT) had no effect on the early phenotype of LV dilation and contractile dysfunction at the time of weaning, but prevented the subsequent progression to HF. A key mechanism for improvement in contractile function appeared to be prevention of OPTM of SERCA at cysteine 674. In another study, overexpression of catalase targeted to the mitochondria (mCAT) prevented progressive adverse remodeling and failure. In the same study, mCAT also alleviated adverse remodeling in response to chronic angiotensin infusion, and the effects in both models were mimicked by an orally administered antioxidant peptide targeted to the mitochondria, thus highlighting the importance of mitochondrial ROS and demonstrating the potential therapeutic utility of this approach.
Anthracycline-Induced Cardiomyopathy ( Fig. 8.4 )
Anthracycline causes cardiotoxicity via several mechanisms including excess free radical production induced by the quinone group; modulation in topoisomerase 2β activity, leading to DNA strand breaks; alterations in the multidrug-resistant (MDR) efflux proteins, leading to higher levels of intracardiac levels of anthracycline; and decreased progenitor cell population. The anthracyclines induce myocyte apoptosis and necrosis in a dose-dependent manner. Both NOX-2 and mitochondria-derived ROS have been implicated as potential mediators of these effects. The apoptosis, but not necrosis, could be inhibited by the addition of the iron chelator dexrazoxane, which has shown promise clinically in the prevention of anthracycline cardiotoxicity. These results implicate hydroxyl radicals formed via the Fenton reaction in anthracycline-induced apoptosis. Anthracycline cardiotoxicity can be acute, occurring within days, or chronic, developing over months or even years. Both acute and chronic doxorubicin (Dox) cardiac toxicity and oxidative stress in mice were improved by transgenic overexpression of a variety of antioxidant proteins including metallothionein, pCAT, sirtuin 3, and thioredoxin. In animal models of Dox-induced cardiomyopathy, vitamin C administration ameliorated both oxidative stress and Dox-induced cardiac dysfunction.