Cardiovascular Toxicology

, Volume 10, Issue 2, pp 73–86

Protective Role of Antioxidants in Diabetes-Induced Cardiac Dysfunction


  • Guy Vassort
    • INSERM U-637, Physiopathologie Cardiovasculaire, CHU Arnaud de Villeneuve
    • Department of Biophysics, Faculty of MedicineAnkara University

DOI: 10.1007/s12012-010-9064-0

Cite this article as:
Vassort, G. & Turan, B. Cardiovasc Toxicol (2010) 10: 73. doi:10.1007/s12012-010-9064-0


Cardiac dysfunction occurs during type 1 and type 2 diabetes and results from multiple parameters including glucotoxicity, lipotoxicity, fibrosis and mitochondrial uncoupling. Oxidative stress arises from an imbalance between the production of ROS and the biological system’s ability to readily detoxify the reactive intermediates. It is involved in the etiology of diabetes-induced downregulation of heart function. Several studies have reported beneficial effects of a therapy with antioxidant agents, including trace elements and other antioxidants, against the cardiovascular system consequences of diabetes. Antioxidants act through one of three mechanisms to prevent oxidant-induced cell damages. They can reduce the generation of ROS, scavenge ROS, or interfere with ROS-induced alterations. Modulating mitochondrial activity is an important possibility to control ROS production. Hence, the use of PPARα agonist to reduce fatty acid oxidation and of trace elements such as zinc and selenium as antioxidants, and physical exercise to induce mitochondrial adaptation, contribute to the prevention of diabetes-induced cardiac dysfunction. The paradigm that inhibiting the overproduction of superoxides and peroxides would prevent cardiac dysfunction in diabetes has been difficult to verify using conventional antioxidants like vitamin E. That led to use of catalytic antioxidants such as SOD/CAT mimetics. Moreover, increases in ROS trigger a cascade of pathological events, including activation of MMPs, PPARs and protein O-GlcNAcation. Multiple tools have been developed to counteract these alterations. Hence, well-tuned, balanced and responsive antioxidant defense systems are vital for proper prevention against diabetic damage. This review aims to summarize our present knowledge on various strategies to control oxidative stress and antagonize cardiac dysfunction during diabetes.


Oxidative stressReactive oxygen speciesReactive nitrogen speciesSeleniumDoxycyclineHyperglycemiaIntracellular calcium ionHeart function



6-Phosphogluconate dehydrogenase


Advanced glycation end products


Angiotensin converting enzyme


Angiotensin II


Aldose reductase


Angiotensin II type 1 receptor




Endothelial nitric oxide synthase


Glucose-6-phosphate dehydrogenase


Glutathione reductase




Glutathione peroxidase


Oxidized glutathione


Reduced glutathione




Hydroxyl radical


Hydrogen peroxide


Inducible nitric oxide synthase


Left ventricular


Matrix metalloproteinases


Manganese superoxide dismutase






Nicotinamide adenine dinucleotide phosphate


Nitric oxide radical


Nitric oxide synthase


Nicotinamide adenine dinucleotide phosphate oxidase


Nuclear factor-kappa B


Oxygen ion


Superoxide radical


Poly(ADP-ribose) polymerase


PPARγ coactivator-1α


Protein kinase A


Protein kinase C


Peroxisome proliferator–activated receptors

PPARα, γ

Peroxisome proliferator–activated receptor α, γ


Reactive oxygen species


Reactive nitrogen species


Renin–angiotensin system


Ryanodine receptor type 2


Sarcoplasmic reticulum


Sarco/endoplasmic reticulum Ca2+ ATPase


Superoxide dismutase




Thiobarbituric acid-reactive substances




Thioredoxine reductase


Thioredoxine peroxidase


Uncoupling protein


The prevalence of diabetes mellitus is growing rapidly from 135 million in 1995 to an estimated 330–380 million in 2025. Among the 3.8 million deaths each year, about 2/3 are attributable to cardiovascular complications associated with the disease ( Diabetes is recognized as a potent and prevalent risk factor for ischemic heart disease. For long, little was known as to whether diabetes causes an altered cardiac phenotype independent of coronary atherosclerosis, as early inferred from hyperglycemia-induced microvascular damage leading to glomerosclerosis and retinopathy. A specific diabetic cardiomyopathy, distinct from coronary arteriosclerosis was first proposed by Rubler et al. [1]. The proportion of cardiovascular disease morbidity and mortality attributable to diabetes mellitus has increased over the past 50 years according to the Framingham heart study implying that diabetes impairs ventricular function independent of other risk factors, coronary artery disease or hypertension [2]. Diabetic cardiomyopathy has been associated with both type 1 (insulino-deficient) and type 2 (insulino-resistant) diabetes and is characterized by both early-onset diastolic and late-onset systolic dysfunctions [3].

Diabetic cardiomyopathy was initially classified as a dilated cardiomyopathy with prominent left ventricular enlargement and depressed systolic function. Over the last two decades, however, diastolic left ventricular (LV) dysfunction was identified as an early manifestation of diabetic cardiomyopathy [4, 5]. Cardiac dysfunction occurring during type 1 and type 2 diabetes results from multiple parameters including glucotoxicity, lipotoxicity, fibrosis and mitochondrial uncoupling. However, type 2 diabetes carries additional risk factors compared to type 1, including insulin resistance, obesity and dyslipidemia. It is also a far more common health problem and it is a major risk factor for congestive heart failure.

It is a well-known fact that hyperglycemia increases the production of reactive oxygen species (ROS), alters the cellular redox status and causes rapid changes in membrane function, followed by contractile dysfunction within weeks in the diabetic heart [616]. Significant increases in oxidants trigger a cascade of pathological events, including contractile dysfunction [8, 9, 16]. Oxidative stress, being an imbalance between endogenous ROS and antioxidant systems in favor of the former, is involved in the etiology of diabetes-induced downregulation of heart function. Moreover, there is a close relationship between impaired insulin signaling and alteration in heart function via depressed endogenous antioxidant defense mechanism [8, 9, 1214, 17, 18].

Several major general reviews have appeared [1012]. Some are more specifically related to the effects of antioxidants since alterations in the redox status appear to be a common link. However, most reviews are devoted to endothelial and vascular aspects [1215]. Therefore, the present review aims to summarize our present knowledge on the effects of antioxidants to ameliorate cardiac dysfunction during diabetes.

Redox Status in Diabetic Cardiomyopathy and Its Alterations

The pathogenesis of diabetic cardiomyopathy is undoubtedly multifactorial and complex. It includes alterations in cardiac energy metabolism showing a reduced glucose uptake and an increased free fatty acid oxidation related to mitochondrial uncoupling, as well as impaired Ca2+ homeostasis, both leading to a deficient contractile activity. More specifically, in diabetes, the heart is becoming almost solely dependent on the metabolism of fatty acids. Such an increase in the myocardial uptake of fatty acids implies an increase in fatty acid oxidation and reduction in glucose oxidation resulting in a decrease in ATP production per mole of oxygen and an increase in mitochondrial uncoupling leading to an unfavorable energetic state together with an overproduction of ROS. This occurs through activation of peroxisome proliferator–activated receptors, PPARs, a superfamily of nuclear, ligand-activated transcriptor factors that play an important role in the transcriptional regulation of genes coding for proteins involved in lipid utilization, lipoprotein metabolism and insulin action. Furthermore, when compared with glucose, oxidation of fatty acid consumes about 10% more oxygen. This reduction in cardiac efficiency and increase in oxygen demand makes the heart especially vulnerable to damage following increased workload or ischemia. Augmented fatty acid oxidation results in augmented ROS production [16].

In isolated ventricular myocytes, diabetes has long been reported to rapidly induce contractile dysfunctions associated with altered Ca2+ handling [1921]. These effects are mostly attributed to reduced Ca2+ current [22], reduced sarcoplasmic reticulum (SR)-Ca2+ load in relation to anomalous SR pump activity and altered cardiac ryanodine receptors (RyR2) due to formation of disulfide bonds between adjacent sulfhydryl groups, increased advanced glycation end products (AGEs) and protein kinase A (PKA)-dependent hyperphosphorylation of RyR2 and FKBP12 leading to unbinding of the latter and Ca2+ leakage through the RyR2 channels, besides the reduced number of RyR2 channels [23, 24]. Of note, these effects are minimized in female rats [25]. Ca2+ handling is also impaired in the mitochondria of an animal model of obesity and type-2 diabetes, the ob/ob mice [26]. However, there is no general consensus in results from studies on intracellular Ca2+ homeostasis. It is agreed that the slower AP and reduced cardiac SR Ca2+ ATPase (SERCA2a) [27] can account for the slower Ca2+ transient, while in some reports, the contractile deficit should be attributed to LV remodeling [28]. Redox status is also known to impair contractile machinery [2931] that may account for the known depressed cardiac myofilament function observed in human diabetes.

Hyperglycemia is a major etiological component in the development of diabetic cardiomyopathy and is known to promote the production of ROS and reactive nitrogen species (RNS) and/or to deplete antioxidant mechanisms in many cell types. A growing amount of evidence indicates that oxidative stress contributes markedly to the alterations observed during diabetes [13]. ROS, such as free oxygen, oxygen ions (O2), hydroxyl radicals (OH), superoxide radicals (O2) and peroxides (H2O2), are continuously produced in most cells under physiological conditions. Their levels are regulated by a number of enzymes and physiological antioxidants such as superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase (CAT) as well as by other nonenzymatic antioxidants. Four enzyme systems are of major importance to produce ROS, nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, xanthine oxidase, the enzymes of the mitochondrial respiratory chain and a dysfunctional endothelial nitric oxide synthase (eNOS). NADPH oxidases are multicomponent enzymes localized in caveolae with specific tissular distribution with NOX2 and NOX4 found in the heart. They are involved in several cardiovascular diseases [32], particularly under the activation of AT1 receptors by angiotensin II, Ang II [33]. Mitochondria are constant source of superoxide radicals (O2) with about 1% of oxygen consumed by mitochondria reduced to a single electron at two sites of the respiratory chain, NADPH dehydrogenase in complex 1 and ubiquinone–cytochrome bc in complex III. However, the amount of superoxide released by mitochondria depends on the activity of the manganese superoxide dismutase (SOD2 or MnSOD) located in the matrix. Recent results also demonstrate that a large spectrum of cytoprotective proteins such as NAD(P)H: quinone oxidoreductase, NQO1, and antioxidant proteins such as heme oxygenase 1, HO1, are induced upon activation of the nuclear factor erythroid 2-related factor, NrF2, through the antioxidant response element-dependent pathway [34]. Normalizing the level of mitochondrial ROS prevents glucose-induced activation of protein kinase C (PKC), formation of AGEs, as well as increases in hexosamine pathway flux and polyol pathway flux and nuclear factor-kappa B (NF-κB) activation [11, 12, 35].

Thiol groups are kept in a reduced state at a concentration of 5 mM in animal cells. In effect, glutathione (GSH) reduces any disulfide bond form within cytoplasmic proteins to cysteines by acting as an electron donor (Fig. 1). In the process, GSH is converted to its oxidized form glutathione disulfide (GSSG). GSH is found almost exclusively in its reduced form, because of glutathione reductase (GR), which can revert it from its oxidized form, is constitutively active and inducible upon oxidative stress. In fact, the ratio of reduced glutathione (GSSH) to oxidized glutathione within cells is often used scientifically as a measure of cellular toxicity. In diabetic heart, GSSG is markedly increased without marked changes in GSH content. Thiol antioxidants act through a variety of mechanisms including as components of the general thiol/disulfide redox buffer, as metal chelators, radical quenchers, substrates for specific redox reactions (GSH) and as specific reductants of individual protein disulfate bonds (thioredoxin). A variety of thiol-related compounds have been used such as GSH and its derivatives, cysteine and N-acetyl-l-cysteine (NAC), dithiols such as lipoic acid, and prothiol compounds such as ornithine carbamoyltransferase, OTC.
Fig. 1

A pathway underlying the scavenging mechanism of free radicals by some glutathione (GSH) and nicotinamide adenine dinucleotide phosphate (NADPH)-dependent enzymes. GSSH, reduced glutathione; GSSG, oxidized glutathione; Trx-S2, oxidized thioredoxin; Trx-(SH)2, reduced thioredoxin; O2, superoxide radical; H2O2, hydrogen peroxide; ROH, a hydroxylated product; ROOH, alkyl hydroperoxide

The ubiquitously expressed thiol-reducing systems include the thioredoxin, Trx1, glutaredoxin and glutathione systems. The thioredoxin system (thioredoxin, thioredoxine reductase and NADPH) reduces oxidized cysteine groups on proteins through an interaction with the redox-active center of Trx (Cys–Gly–Pro–Cys) to form a disulfide bond that in turn can be reduced by thioredoxine reductase (Trx-R) and NADPH. Thioredoxin appears to exert most of its antioxidant, ROS-scavenging properties through thioredoxine-peroxidase (Trx-P). Thioredoxin plays an essential role in cell function by limiting oxidative stress directly by antioxidant effects and indirectly by protein–protein interaction with key signaling molecules such as Trx-interacting protein, TXNIP. Hyperglycemia and diabetes induce TXNIP and decrease Trx activity [36].

The activities of enzymes that play important roles in antioxidant defense mechanisms of the cells, including GR, GSHPx, glutathione-S-transferase, GST, glucose-6-phosphate dehydrogenase (G6PD) and TRX-P, decreased significantly in diabetic rat heart [37]. It is to note that some previous studies reported that both the levels and activities of the antioxidant enzymes G6PD, 6-phosphogluconate dehydrogenase (6PGD), GR, GSH-Px and CAT but not GST are increased in diabetic heart [38, 39]. In addition, an elevated G6PD level was reported which may result in an increased NADPH production required for GR activity and thus GSH production [40]. Recent proteomic studies in STZ-diabetic rats indicate that 12 of the 24 altered proteins were localized in mitochondria, associated with a downregulation of antioxidant (aldehyde dehydrogenase, ALDH-2, and 3-mercaptopyruvate sulfurtransferase, MST) and antiapoptotic proteins [41].

Diabetes alters the activity and expression of nitric oxide synthase (NOS) [42]. NOS alteration could also result from deficiency in tetrahydrobiopterin (BH4), a critical cofactor of eNOS [43]. Furthermore, it was recently reported that an increase in ROS activity can upregulate NOS expression in vitro and also in vivo. This effect appears to be, in part, mediated by limiting the availability of nitric oxide (NO), thereby exerting a negative feedback influence on NOS expression through activation of NF-κB [44]. The toxicity of NO is believed to be associated with its interaction with iron–sulfur-centered enzyme of the respiratory cycle, such as to control mitochondrial-free oxygen consumption and limit ATP production. This is further complicated by superoxide. Both nitric oxide (NO) and superoxide radicals react to generate peroxinitrite (ONOO), which rapidly decomposes to highly oxidant species such as nitronium ion (NO2+). Like NO, peroxinitrite has been associated with both deleterious and beneficial effects (Fig. 2). It is particularly toxic due to its remarkable stability as an anion at alkaline pH. Peroxynitrous acid is a strong oxidant that reacts with many biological molecules such as iron–sulfur centers, zinc fingers and protein thiols. Peroxinitrite is toxic by direct oxidative mechanisms whose chemistry is poorly known. The cardiotoxicity of peroxinitrite generally reported in crystalloid cardioplegia as opposed to cardioprotection observed in blood cardioplegia suggests a dependency on the environment in which the anion is present. One interpretation is that glutathione, a major component of blood, reacts with peroxinitrite to form NO, nitroglutathione or nitrothiol [45]. Moreover, the potent cytotoxin peroxinitrite leads to lipid peroxidation and to the nitration of tyrosine residues to give 3-nitrotyrosine, a reaction catalyzed by SOD. Protein tyrosine-nitration impaired mitochondria functions [46] and was shown to alter many enzymes in various cardiovascular pathology [47]. Increased oxidative and nitrosative stress also activates the nuclear enzyme poly(ADP-ribose) polymerase (PARP), which importantly contributes to the pathogenesis of cardiac dysfunction associated with diabetes [48].
Fig. 2

Generation of reactive species in diabetic heart under hyperglycemia. Oxygen is converted to O2 via the activation of several enzymatic pathways and then dismutated to H2O2 by Cu- and Mn-SOD in the cytosol and the mitochondria, respectively (and see Fig. 1). In addition, O2 reacts rapidly with nitric oxide radical (NO) to form ONOO triggering the formation of advanced glycation end products (AGEs) and hexoamine, lipid peroxidation, nitrotyrosine and activates the polyol pathway, which all may contribute to diabetic cardiomyopathy. Some inhibitory compounds highlighted in gray are named closed to their site of action

In the last decade, evidence indicating upregulation of the cardiac renin–angiotensin system (RAS) in diabetic heart is repeatedly accumulating [4951]. Ang II, via AT1 receptor stimulation, is also suggested to increase the production of free oxygen radicals [5255] and to activate NAD(P)H oxidase enzymes that may generate superoxide anions such as O2 in various tissues including heart [5254, 56, 57]. AT1 receptors also activate PKC as was found in diabetic heart [50, 58]. Fiordaliso et al. [49] also observed an elevated myocyte death associated with upregulation of RAS in diabetic heart.

Although results from a Cochrane review suggest no beneficial effects of antioxidant supplements, such as vitamin A, C and E and β-carotene, for prevention of mortality in healthy participants and in patients with various diseases [17], several studies have reported beneficial effect of antioxidant therapy against the cardiovascular system consequences of diabetes. Antioxidants act through one of three mechanisms to prevent oxidant-induced cell damages. They can reduce the generation of ROS, scavenge ROS, or interfere with ROS-induced alterations.

Reducing ROS Generation

Modulating mitochondrial activity is an important possibility to control ROS production. BM 17.0744, a PPARα agonist reduces fatty acid oxidation with a concomitant increase in glucose oxidation in db/db mice in which the fatty acid-induced mitochondrial uncoupling is mediated by the uncoupling proteins, UCP leading to impaired oxidative capacity with increased hydrogen peroxide production. However, this compound did not improve left ventricular contractile function [59]. Furthermore, several animal studies reported on various tissues that thiazolidinediones, initially known as insulin-sensitizing agents, also binds to PPARγ and so reduces ROS production [60]. Indeed recently, rosiglitazone and pioglitazone have both been reported in patients to improve myocardial dysfunction in association with reduced oxidative stress [61, 62]. However, in the later study, the functional changes were not associated with significant changes in the high-energy phosphate metabolism.

Chronic physical exercise appears to be a physiological stimulus able to induce mitochondrial adaptation that can counteract the adverse effects of diabetes. Exercise is shown to increase the expression of several proteins including PGC-1 and restore the low UCP3 expression at least in skeletal muscle [63].

Trace elements such as selenium and zinc might act as antioxidants. Selenium has beneficial effects on glucose metabolism [64]. In STZ-induced diabetic rats, sodium selenite protected ultrastructural alterations and restored the altered electrical and mechanical dysfunctions of diabetic rat heart partially by restoring the cell glutathione redox cycle [8, 9, 65, 66]. More recent studies demonstrate that sodium selenate administration for 4 weeks also reduced the oxidized protein sulfhydryl and nitrite concentrations via reducing MMP-2 activation and therefore reducing the degradation of two of its target proteins, troponin I (TnI) and α-actinin [67].

Zinc is a constituent of more than 300 catalytic active Zn2+ metalloproteins and Zn2+-dependent transcription factors. Zn2+ homeostasis in the cells is strictly controlled [68]. Several studies in neurons have reported that elevated intracellular Zn2+ is correlated with mitochondrial dysfunction that leads to reduced ATP and increased ROS productions following DNA damage and the activation of PARP-1 [69, 70]. In isolated cardiomyocytes, oxidants are known to trigger increased intracellular free Zn2+ levels ([Zn2+]i) [71]. More particularly, diabetic cardiomyocytes exhibited significantly increased [Zn2+]i and [Ca2+]i [9] but see [72]. This was accompanied by decreased levels of metallothioneins (MTs) and GSSH, increased levels of lipid peroxidation and NO products and decreased activities of SOD, GR and GSH-Px. Thus, an increase in [Zn2+]i may contribute to oxidant-induced alterations of excitation–contraction coupling in diabetes. Treatment of diabetic rats with sodium selenite prevented these Zn2+-induced cellular defects [9].

On the other hand, Zn2+ deficiency was found to be a risk factor for cardiac oxidative damage, while Zn2+ supplementation provides a significant prevention. Zinc is a potent antioxidant whose deficiency causes multiple damages in the heart [73]. Zn2+ deficiency has been considered as a risk factor for the development of diabetes. Serum Zn2+ concentrations, but not Cu2+, are significantly lower in patients with diabetes with beneficial effects of Zn2+ supplementation [74, 75]. Zn2+ supplementation also prevented the development of cardiomyopathy in STZ-induced diabetic mice [72]. Zinc functions as a complex antioxidant through participation in SOD and Trx enzymatic and chelator activities as well as inhibiting lipid peroxidation [76]. Zn2+ deficiency also accelerates the peroxinitrite-induced eNOS uncoupling [77]. Furthermore, Zn2+ is a potent inducer of MTs that are very efficient in scavenging free radicals and hydrogen peroxide [72, 78]. Indeed, the prevention of cardiomyopathy by Zn2+ supplementation in STZ-induced diabetic mice is predominantly mediated by an increase in cardiac MT [79].

The pathogenic role of nitrosative stress and peroxinitrite, and the downstream mechanisms including PAPR-1 activation could be antagonized by a metalloporphyrin decomposition catalyst, FP15, in murine models of diabetic cardiovascular complications [80].

Activation of the RAS, and subsequent signaling through the AT1 receptors, appears to contribute to the development of diabetic cardiomyopathy [81]. Tissue concentrations of Ang II and AT1 density increase in diabetic rat myocardium [82]. Ang II is long known to increase vascular smooth muscle O2 production by activation of a membrane-bound NAD(P)H oxidase. Angiotensin converting enzyme (ACE) inhibitors improve the outcome of heart failure in patients with diabetes, even to a greater extent than in nondiabetic ones [83]. AT1 blockade, as well as NADPH oxidase inhibition, protects from the enhanced ROS production induced by high glucose applied to rat cardiomyocytes. Thus, it has been suggested that diabetes-related oxidative stress attenuates K+ currents through Ang II-generated increased superoxide ion levels [84]. AT1 blockade restores action potential duration, transient outward K+ current, Ito amplitude, Ca2+ homeostasis including Ca2+ transients kinetics, SR-Ca2+ load, spatio-temporal properties of Ca2+ sparks, and basal Ca2+ level, as well as the β-adrenergic-mediated enhancement of glucose uptake [58, 85]. The later report [58] further described that these effects were associated with a reduction in the increased PKC level and oxidized protein thiol level in membrane fraction of diabetic rat heart. In a recent study, we also reported that the AT-1 blocker candesartan cilexetil restores lower levels of G6PD, 6PGD, GR, GSH-PX, CAT, with no effect on GST [86]. ACE inhibitors and Ang II-receptor blockers improve cardiovascular and all-cause mortality outcomes in patients with diabetes to a greater degree than in nondiabetics [87]. It was also demonstrated that overexpression of the growth factor IGF-1 protects diabetic myocardium by depressing the synthesis of Ang II [88]. Besides the nitrosative damage induced by Ang II, apoptosis consequent to the membrane translocation and activation of NOX isoform p47phox was prevented in cardiac-specific MT and CAT-overexpressing transgenic mice [89]. Interestingly, also in STZ-induced diabetic rat cardiomyocytes, bosentan, an endothelin-1 receptor inhibitor, induced similar recovery effects on K+ current and action potential duration [90] as well as did spironolactone, a mineralocorticoid antagonist in male but not in female diabetic rats, there reported associated with a reduction in the aldosterone-induced oxidative stress [91].

In patients who had just experienced myocardial infarction, glucose values in excess of 0.11–14 g/l are associated with a threefold increase in mortality and a higher risk of cardiac failure [92]. High glucose is known to enhance inducible nitric oxide synthase (iNOS) expression. Although increased NO may decrease vascular resistance, high NO may also depress myocardial contractility, and through the formation of peroxinitrite may cause myocardial damage. After ischemia and reperfusion, both iNOS+/+ and iNOS−/− diabetic mice had myocardial infarct size greater than their respective nondiabetic littermates. The diabetic iNOS+/+ showed a greater infarct size associated with the greatest levels of nitrotyrosine and proinflammatory cytokines. The beneficial effect of iNOS in modulating defensive responses against ischemia–reperfusion seems to be abolished in diabetic mice [93].

Scavenging ROS

The paradigm that interrupting the overproduction of superoxide radicals and hydrogen peroxides would normalize most alterations that contribute to cardiac dysfunction has been difficult to accomplish using conventional antioxidants. Conventional scavenging antioxidants like vitamin E act on a one-to-one basis, while hyperglycemia-induced overproduction of superoxide radicals is continuous. That led to the use of catalytic antioxidant such as SOD/CAT mimetics.

Vitamin E as a lipid-soluble antioxidant mainly scavenges oxidized products that result from damaged molecules by hydroxyl radicals and peroxinitrite. However, it does not provide protection against the damage of these molecules (key enzymes, lipids and DNA). In randomized clinical trials, vitamin E does not bring significant benefit [94, 95].

In the first of these studies, (n-3)polyunsaturated fatty acids, omega-3 was shown to lower the cardiovascular death significantly. Increasing evidence related with increased consumption of PUFA benefits subjects without adversely affecting glucose control that has prompted the American Diabetes Association to recommend the consumption of fish. Experimentally, a regimen including omega-3 together with vitamin E given to STZ-induced diabetic rats caused almost complete normalization of the antioxidant defense enzymes. This regimen caused also significant recovery of the left ventricular-developed pressure by normalizing the lengthening of relaxation in male rats, while it had little effect in females [37].

α-Lipoic acid is a disulfide compound that functions as a coenzyme in pyruvate dehydrogenase and α-ketoglutarate dehydrogenase mitochondrial reactions, leading to the production of cellular energy. α-Lipoic acid and its reduced form, dihydrolipoic acid, reduce oxidative stress by scavenging a number of free radicals by chelating transition metals in biological systems, by preventing membrane lipid peroxidation and protein damage through the redox regeneration of other antioxidants such as vitamin C and E and by increasing intracellular glutathione. Specifically, in STZ-induced diabetic female rats, α-lipoic acid normalized the reduced SOD activity in heart after 14 days of treatment [96].

Cardiac-specific overexpression of wild-type Trx1 in mice did not show cardiac modification at baseline but demonstrated reduced levels of hypertrophy and oxidative stress in response to cardiac overload. Despite their markers with significantly increased oxidative stress, dominant negative Trx1-mutant mice exhibited cardiac hypertrophy with maintained cardiac function at baseline; however, aortic banding then caused greater alterations [97].

Aldose reductase, AR, has been implicated in the pathogenesis of various diabetic complications. It catalyzes the reduction of aldehydes, including the aldehyde form of glucose using NADPH as a cofactor, initiating the polyol pathway that elicits an increased NADH/NAD+ ratio. At early stage of hyperglycemia in STZ-induced diabetic mouse model, AR activity was markedly increased. AR possesses a redox-sensitive cysteine residue that modulates the enzyme activity. Thus, NAC, a precursor of GSH, increases the GSSH/GSSG ratio along with a decrease in thiobarbituric acid-reactive substances, TBARS [98]. In fact, the levels of GSH in STZ-induced or insulin-dependent diabetic rats are elevated in the heart, suggesting that the cardiac tissue exposed to ambient high glucose may initiate a defense response by enhancing its antioxidant systems. Augmented activities of antioxidative enzymes such as SOD, CAT and GSH-Px were also documented at early stage of diabetes without any changes in TBARS [99]. More recently, this was also shown for the protein expression of Cu–Zn-SOD, heme oxygenase-1, and the total SOD activity despite manganese superoxide dismutase (MnSOD or SOD2) activity was clearly reduced [100, 101]. NAC treatment prevented the increased expression of Cu–Zn-SOD and HO-1, while the reduced left ventricular developed pressure and rate of relaxation were only partially compensated.

Diabetic animals have up to a threefold elevated cytosolic redox ratio NADH/NAD+, and this redox imbalance is probably due to elevated polyol pathway flux. The increased polyol pathway activity may inhibit glycolysis due to a decrease in the NAD+-dependent flux from glyceraldehyde 3-phosphate to 1,3-bisphosphoglycerate. Inhibiting aldose reductase with zopolrestat preserves high-energy phosphates, maintains a lower cytosolic NADH/NAD+ ratio and markedly protects both diabetic and nondiabetic hearts during ischemia and reperfusion [102].

Mitochondria respiration acts as a major source of oxidative stress in diabetic complications. The overgeneration of superoxide radicals as a by-product of electron transport in diabetes appears to be antagonized in transgenic mice overexpressing the mitochondrial antioxidant protein MnSOD. Such transgene protects cardiac morphology and completely normalizes contractility of diabetic cardiomyocytes [103]. To some extent, SOD mimetics such as M40403, a manganese-containing biscyclohexylpyridine that removes superoxide anions without interfering with other reactive species, were shown to prevent myocardial injury induced by acute hyperglycemia in perfused heart [104].

Taurine has been shown to be unable to directly scavenge the classic ROS, superoxide radical, hydroxyl radical or hydrogen peroxide. Rather, taurine upregulates the cellular antioxidant defense enzymes in various tissues, and it reduces oxidant generation by preventing Ca2+ overload and thus mitochondria defective activity. Another recent proposal is the presence of taurine-conjugated tRNAs in the mitochondria whose deficiency reduces the expression of some respiratory chain components [105].

Oxidatively stressed cells respond by induction of protective enzymes, particularly they overexpressed CAT. CAT transgene in both OVE26 and agouti mice, respectively type 1 and type 2 diabetic models, preserves cardiac morphology and prevents most contractile defects although it does not reverse the slow Ca2+ decay [106]. That CAT was effective in both types of diabetes reinforces the view that increased ROS production induced by hyperglycemia is an essential cause of diabetic cardiomyopathy.

MT is a cysteine-rich protein that can bind heavy metal ions such as Cu2+ and Zn2+. MT also has a strong effect in scavenging free radicals because of its high thiol content and was shown to scavenge hydroxyl radicals in vitro, both several hundredfold more than GSH [107]. It is highly inducible by a variety of agents including zinc so it is clinically potential as an antioxidant by Zn2+ supplementation. Unlike other antioxidants that protect against specific species of ROS such as SOD against superoxide radicals and CAT and GSH-Px against hydrogen peroxide, MT protects against a wide range of free radicals [107]. Despite the MT content was the lowest in heart compared to other organs suggesting a high susceptibility to oxidative stress, cross-breeding transgenic mice overexpressing MT specifically in the heart with OVE26 diabetic mice benefits diabetic heart morphology and whole-heart contractility [108]. In STZ-induced diabetic mice, MT-transgene expression significantly alleviates cardiomyocyte contractile dysfunctions, while it reduces the increased expression of AT1 receptor, NADPH oxidase and PARP, increases the levels of SERCA2 and Na+/Ca2+ exchanger and antagonizes the switch to heavy chain myosinβ [109]. Besides its ROS-scavenging properties, MT might have further effects, beneficial or detrimential, in diabetic heart through its zinc-binding properties since under oxidative stress, Zn2+ could be released from MT.

Interfering with ROS-Induced Alterations

Significant increases in ROS trigger a cascade of pathological events, including activation of matrix metalloproteinases, MMPs, activation of peroxisome proliferator–activated receptors, PPARs, and protein O-GlcNAcation.

Matrix metalloproteinases are zinc-endopeptidases best known for their actions in remodeling the extracellular matrix during development and diseases [110]. Enhanced oxidative stress activates MMPs of which MMP-2 is found in heart. MMPs have been reported to contribute the acute contractile dysfunction during ischemia–reperfusion injury in part by activation of MMP-2 that induces proteolytic cleavage of susceptible intracellular targets such as TnI while inhibitors of MMP activity prevent cardiac dysfunction. Diabetes caused a significant decrease in myocardial MMP2 activity, in the level of tissue inhibitor of metalloproteinase, TIMP-4, as well as a significant loss of TnI. Inhibition of MMPs by doxycycline improved cardiac function, restored K+ currents, and as well normalized MMP-2, TIMP-4 and TnI levels in the diabetic rat heart [111]. However, only antioxidants such as selenate and polyunsaturated fatty acids but not doxycycline could restore the depressed β-adrenergic response of the diabetic rat heart by restoring the density of β1-receptors and the β1-receptors-G-protein coupling [112].

Posttranslational modification of proteins is a common mechanism for the modulation of protein function, of which phosphorylation is the most widely studied, besides acylation, ubiquitylation, methylation, acetylation, thiolation, nitration and glycosylation. Hyperglycemia increases glucose flux through the hexosamine pathway, resulting in overproduction of UDP-GlcNAc, donor of O-GlcNAc. Protein O-GlcNAcation is a key regulator of critical biological processes. The intersection of redox and O-GlcNAcylation has been stressed out in the regulation of signaling pathways [113]. Specifically, the altered levels of O-GlcNAc contribute to the adverse effects of diabetes on Ca2+ cycling [114]. Conversely, O-GlcNAcase overexpression reversed the effects of increased O-GlcNAc improving Ca2+ handling by increased SERCA and phosphorylated phospholamban expressions [115].


To conclude, we like to stress out a few topics that require further attention.

Positive or Negative, Preventive or Deleterious Effects?

Zn2+ appears to be a complex antioxidant and a potent inducer of MT. Zn2+ deficiency is a risk factor for the development of diabetes with Zn2+ supplementation having beneficial effects and preventing the development of cardiomyopathy in STZ-induced diabetic mice. On the other hand, diabetic hearts exhibit significantly increased [Zn2+]i levels accompanied by decreased contents of MT and reduced glutathione, with sodium selenite preventing these Zn2+-induced defects. It is also known that a selenium deficiency induces cardiomyopathy while its application in vitro, through alterations of cellular thiol redox status, induced a dual, positive then negative, effect on muscle contraction that can be imputed to combined actions on Ca2+ channels, Ca2+ transporters and contractile proteins.

Increased levels of O-GlcNAc have been implicated as a pathogenic contributor to glucose toxicity and insulin resistance, which are both major hallmarks of diabetes mellitus and diabetes-related cardiovascular complications. Conversely, there is a growing body of data demonstrating that the acute activation of O-GlcNAc levels is an endogenous stress response designed to enhance cell survival [113].

ROS generation can be helpful. The generation of superoxide has been argued to be an important element of signal transduction of preconditioning, and in isolated cardiomyocytes, hypoxia has been shown to increase superoxide generation and initiate preconditioning while preconditioning protects by attenuating oxidant generation during subsequent ischemia–reperfusion [116118]. The myocardium from patients with poor left ventricular function can be protected by the mito-KATP opener diazoxide, whereas the diabetic myocardium cannot. Possibly in diabetes, dysfunctional mito-KATP channels that lead to mitochondrial dysfunction including impaired depolarization, mitochondrial swelling and superoxide production are responsible for the inability to respond to preconditioning [119]. Modulating mito-KATP channel activity in diabetes by other factors such as Mg2+ could induce cardioprotective effects as shown in control heart during hypoxia [120].

Gender Effect

In the Framingham Heart Study, diabetes increased the prevalence of congestive HF in women more than in men and with a greater difference in younger than in 65-year-old patients [121]. Reduced antioxidant activity and increased oxidative stress occur early after the diagnosis of type 1 diabetes, specifically in women accounting for an increased susceptibility of diabetic women to cardiovascular complications [122]. The greater effect of diabetes in women compared to men as a risk factor for congestive heart failure is in agreement with a greater contribution of diabetes to arteriosclerosis in women [123]. However, at the cardiomyocyte level, gender difference indicates an increased Ang II level only in male rat with a consequent more marked reduction in K+ currents [124, 125]. Also, diabetes-induced alterations in RyR2 phosphorylation and FKBP12 unbinding, total and free sulfhydryl group reduction, and PKC increase were less marked in female than in male rats [25]. It is to note that antioxidant treatment with omega-3 induces significant recovery in depressed left ventricular function and its rates of changes in males while it further lengthens the diabetes-induced increase in time to peak of the developed pressure in females although, however, restoration in the altered antioxidant enzyme activities occurs without significant gender differences [37].

Diastolic Dysfunction as a Consequence of Altered PCr/ATP Metabolism

Patients with diabetes demonstrate in the early stage an impaired diastolic LV distensibility. This diastolic dysfunction was thought to result from AGEs causing strong collagen cross-links in the extracellular matrix [126] to SH oxidation by peroxinitrite inducing disulfide bonds and altering myofilament activity [127] or to altered Ca2+ homeostasis [24] such that a slower recovery to an increased basal Ca2+ level sustains some cross-bridges. Several other mechanisms have been suggested. Thus, ramipril, an ACE inhibitor, exhibits beneficial effects on myocardial diastolic dysfunction as investigated by pulsed tissue Doppler in type 2 patients with diabetes with normal LV systolic pressure and without coronary artery disease. The effects were attributed to ROS-dependent changes in endothelial function [128]. Similarly, pravastatin, but not atorvastatin, prevented the onset of diabetes and diabetic dysfunction in a rat model. This could be related to a larger eNOS upregulation by pravastatin [129, 130]. Furthermore, it could be anticipated that O-GlcNAcation by analogy with protein phosphorylation modulates diastolic tension as does titin phosphorylation [131]. Diastolic dysfunction in diabetic heart may also result from mitochondrial dysfunction. The latter was attributable to a reduced pyruvate dehydrogenase activity leading to lesser ATP production consequent to reduced Ca2+ uptake [132]. Rather than damage to the Ca2+ uptake machinery, the reduced mitochondrial Ca2+ uptake is related to an enhanced susceptibility to the permeability transition as indicated by the depressed oxygen consumption during the phosphorylated state [132, 133]. Although their cardiac morphology, mass and function appeared normal, asymptomatic, normotensive patients with diabetes had significant lower phosphocreatine (PCr)/ATP ratio compared to healthy volunteers [134, 135]. The lower PCr content leads to lower ATP available at the sarcomeres, slowing cross-bridges detachment and consequently altering diastolic distensibility. Furthermore, rigor or slow detaching cross-bridges may also result from a reduced available ATP concentration at the myofilament level consequent to a deficient myofibrillar creatine kinase after tyrosine-nitration by peroxinitrite [136]. Of note, mitochondrial biogenesis occurs early in the development of diabetic dysfunction through a transcriptional regulatory circuit that involves activation of PGC-1 gene expression by the fatty acid-activated nuclear receptor PPARα [137].

It is trivial to say that the multiplicity of the proposed therapies is an index of their poor individual efficacy. This may have several causes that include the fact that besides our relatively poor knowledge of the exact mechanisms involved in alterations occurring during in diabetes and particularly in the redox status, several compounds act both as a pro-or antioxidant depending on their concentrations and, more difficult to handle, on the physiopathological status of the cells. This lack of efficacy might also result from the fact that we are pharmacologically acting much too late in the signaling cascade, at a time at which initiation of the alterations in the redox status are already processing. More knowledge is needed to better understand the initial steps so that to be able to counteract the very first trigger of imbalance between oxidant and antioxidant systems into cardiomyocytes induced by diabetes.

Copyright information

© Springer Science+Business Media, LLC 2010