Introduction

Atherosclerosis is the underlying condition in most cardiovascular diseases (CVDs), which are the leading cause of death in Western societies. It is a chronic systemic disease affecting the entire arterial tree, associated with impaired inflammatory status [1, 2]. Moreover, endothelial dysfunction (ED) due to decreased nitric oxide (NO) bioavailability and activity plays a pivotal role in the initiation and progression of atherosclerosis [3]. Accordingly, ED is now considered an important early event in the development of atherosclerosis as it seems to precede atherosclerotic lesions in coronary vessels, and even occurs in offspring with a positive history for CVDs [4].

More specifically, NO is a key molecule managing endothelial function and vascular homeostasis [5], while reduced production of NO or increased production of reactive oxygen species (ROS) promotes ED. Thus, oxidative stress plays a crucial role in the pathogenesis and development of CVD [6].

Consequently, a range of antioxidant strategies have been tested with the aim to improve ED [7, 8]. Novel antioxidants have shown encouraging results. However, the available data are inadequate. Recently, a growing body of evidence has indicated the role of dietary modification. Particularly, dietary flavonoids appear to have the potential to restore endothelial function by decreasing oxidative stress status, though the suggested mechanisms mediating their effects are not fully evaluated [9, 10].

In the present review article, we aim to provide an overview of the physiological pathways involved in oxidative stress and in the pathophysiology of atherosclerosis. Further, we will focus on the available therapeutic strategies for targeting redox signaling in vascular endothelium.

Vascular Endothelium

The vascular endothelium consists of a thin semipermeable layer of cells covering the internal surface of vessels and forming a boundary between the vessel wall and blood flow. Its structural and functional integrity is vital for the protection of the vessel wall and circulatory function. Endothelial cells (ECs) have several functions, exerting significant autocrine, paracrine and endocrine actions and influencing smooth muscle cells (SMCs), platelets and circulating leucocytes [11].

It is worth noting that there is a considerable phenotypic deviation between ECs in different parts of the vascular system, expressing different surface antigens and receptors and in turn generating different responses to the same stimulus. Similarly, ECs’ in vivo responses may differ from in vitro responses seen in cultures of ECs’ lines used in many studies [12].

Over the last decade, the vascular endothelium has emerged mostly as a paracrine organ responsible for the secretion of several beneficial substances with anti-atherogenic effects. It regulates body homeostasis and affects thrombosis, thrombolysis, platelet adherence, vascular tone and blood flow. It is involved in many disease processes, including atherosclerosis, arterial and pulmonary hypertension, sepsis and inflammatory syndromes which are related to endothelial injury, dysfunction and activation [13].

NO bioavailability is well-known to exert multiple actions on the vascular endothelium (Table 1). It is capable of reversing constrictive effects of acetylcholine, leading to vasorelaxation and maintaining the balance against various endothelium-derived contracting factors, such as endothelin-1 and thromboxane A2, thereby modulating vascular tone. Beyond this role, decreased NO bioavailability crucially participates in atherothrombosis, given the NO antithrombotic, antiapoptotic, anti-inflammatory and antioxidant effects. Thus, several pathophysiological conditions such as accumulation of ROS, oxidative stress, inflammation, increased adhesion molecules, insulin resistance, decreased shear stress etc., participate in the induction of ED, mainly through a decrease in endothelium-dependent vasodilation [14, 16, 17].

Table 1 Physiological role of nitric oxide [14, 15]
Full size table

Indisputably, loss of the functional integrity of the endothelium and of its anti-atherogenic effects causes a shift of the action of the endothelium towards reduced vasodilation, inflammation and thrombosis [18] (Fig. 1) and plays a major role in all stages of atherosclerosis, from plaque formation to rupture [19, 20].

Fig. 1
figure 1

Pathophysiological mechanisms contributing to atherosclerosis. Abbreviations: NADPH, Nicotinamide adenine dinucleotide phosphate-oxidase; ROS, Reactive oxygen species; BH4, Tetrahydrobiopterin; NOS, Nitric oxide synthase; NO, Nitric oxide

Full size image

Furthermore, several studies have extensively explored ED with respect to its clinical utility as a biomarker in risk prediction [21, 22]. In particular, ED as estimated by the existing invasive and non-invasive techniques could serve as a useful diagnostic and potentially prognostic tool in coronary artery disease (CAD) [23]. Regarding non-invasive techniques, peripheral vascular endothelial function can be assessed by strain gauge venous plethysmography, which evaluates forearm blood flow changes in response to direct intra-arterial administration of agonists or during reactive hyperemia [24]. However, another technique assessing endothelium-dependent flow-mediated dilation (FMD) of the brachial artery using high resolution ultrasound in the brachial circulation, has been widely accepted as a low-cost and easily to perform test [25]. FMD has been found to predict cardiac events in subjects with stable CAD, peripheral arterial disease and in post-myocardial infarction (MI) patients [26–28].

Oxidative Stress and Atherosclerosis Pathophysiology

Atherosclerosis is well known as a disease of the large and medium-sized arteries. It develops progressively, starting at the beginning of life and leading to the formation of atherosclerotic plaques [29]. ED, which is considered an early step in the progression of atherosclerosis, is linked to low NO bioavailability. Low NO bioavailability is due either to decreased production by the endothelium or increased NO inhibition by ROS [30].

The second step in the pathophysiology of atherosclerosis is the passage of plasma LDL into the arterial wall. The strong association between hyperlipidemia and atherosclerosis has been previously recognized and lipid-laden macrophages or foam cells are documented as a hallmark of the disease [31]. More specifically, plasma LDL is transported across the endothelium and is trapped in the subendothelial space where it is oxidized to produce highly oxidized LDL. These molecules are potent inducers of inflammatory molecules which stimulate inflammatory signaling by ECs, through releasing chemotactic proteins and growth factors, and by further recruiting monocytes into the arterial wall [32].

Concurrently, monocytes entering into the arterial wall differentiate into macrophages incorporating cholesterol from lipoproteins and remain in the subendothelial space. The oxidized LDL particles are taken up by macrophages which then can induce a local inflammatory response [32]. Chemokines, which are produced in response to oxidized lipoproteins, participate in the transendothelial migration of adherent monocytes. Accordingly, oxidative stress status alters the expression of adhesion molecules, such as vascular cell adhesion molecule-1 (VCAM-1). Moreover, the expression of a number of proinflammatory cytokines and other inflammation-induced molecules like CD40 ligand are strongly related to redox-sensitive factors such as the nuclear factor-kB [33, 34].

The final consequence of the evolution of atherosclerosis is the formation of plaques with thin fibrous caps which are the result of increased collagen breakdown by matrix metalloproteinases and decreased collagen synthesis by dysfunctional or apoptotic SMCs [35]. Finally, the significant contribution of macrophages to the rupture of the thin fibrous cap has been well-recognized, as plaques are likely to rupture at sites of increased macrophage content [36].

Obviously, these observations are of major importance, as they have provided not only mechanistic links between lipoproteins and cell biology of atherosclerosis, but have also provided concepts for potential therapeutic interventions.

Therapeutic Interventions Targeting Oxidative Stress

As it is widely established that ED leads to atherosclerosis, significant effort has been devoted to improving clinical outcome by modulating vascular redox state. There are various therapeutic strategies intending to improve or restore ED targeting oxidative stress. Some of them are effective while others are promising or under investigation (Table 2).

Table 2 Established and novel treatments of oxidative stress in atherosclerosis
Full size table

Angiotensin Converting Enzyme Inhibitors and Angiotensin Receptor Blockers

According to clinical and experimental data, angiotensin converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) seem to exert beneficial anti-atherosclerotic and anti-ischemic effects [44–46]. Growing data indicate that ACE inhibitors and ARBs are capable of reversing ED, atherosclerosis and vascular inflammation and may decrease the burden of CVDs [47–50] . Moreover, from data based on vascular response to acetylcholine, on B-mode ultrasound evaluation of atherosclerotic progression in the carotid artery and on the progression of CAD using quantitative coronary angiography, it is evident that ACE inhibitors and ARBs may increase plaque stability and reduce cardiac events [51–53].

Disparity exists concerning the endothelial protective effects of different anti-hypertensive regimes. In particular, calcium channel antagonists improve endothelial function in the microcirculation while ACE inhibitors and ARBs mostly do so in conduit arteries [54]. Interestingly, a sulfhydryl group confers to ACE inhibitors some additional properties to those of ARBs, such as scavenging of superoxide anion, greater protection against LDL oxidation and nuclear factor kappa-light-chain-enhancer of activated B cells activation, leading to more pronounced favorable effects on NO balance in vascular ECs [55].

More specifically, zofenoprilat (zofenopril prodrug) and to a less significant extent captopril, have been demonstrated to decrease generation of ROS induced by tumor necrosis-alpha (TNF-α) in human umbilical vein ECs [37].

Moreover, ACE inhibitors, beyond their decreasing effect on angiotensin II levels, also decrease the degradation of bradykinin. Elevated bradykinin levels oppose the negative actions of angiotensin II with parallel antiapoptotic actions. As Oeseburh et al. have shown, by adding bradykinin in cultured bovine aortic endothelial cells, bradykinin protects against oxidative stress-induced endothelial cell senescence [56]. The endothelial protective and anti-oxidative role of bradykinin was further confirmed by Kobayashi et al. who found that treatment of hypertensive rats with quinapril and a bradykinin B2 receptor antagonist did not achieve up-regulation of eNOS and down regulation of lectin-like oxidized LDL receptor-1 as was the case in rats treated only with quinapril [57].

Importantly, evidence from the EUROPA study in CAD patients has shown that blood pressure reduction with perindopril by itself does not explain the observed benefits [58]. In addition, treatment with perindopril resulted in up regulation of eNOS protein expression and activity [59]. Similar results concerning the role of perindopril on bradykinin increase and in endothelial function improvement were also reported from sub-studies of EUROPA [60, 61]. Interestingly, we have to notice that the effect of ACE inhibitors on bradykinin may vary between ACE inhibitors and may depend on the degree of tissue affinity of specific ACE inhibitors, with perindopril having the highest selectivity for the bradykinin binding sites [62, 63].

Olmesartan, an ARB, has also exhibited beneficial properties regarding progression of coronary atherosclerosis (OLIVUS) [52, 53], reduction of inflammatory biomarkers in hypertension (EUTOPIA) [64] and improvement of intima-media thickness (IMT) in patients with diagnosed atherosclerosis [65, 66]. Of note, valsartan compared to amlodipine has exhibited an enhanced vasodilatory response after co-infusion of acetylcholine and L-NG-monomethyl Arginine in a double-blind, crossover trial [67]. Even though the underlying physiological mechanisms have not been elucidated yet, this trial supported that valsartan reversed ED through both NO-dependent and independent pathways, while amlodipine had only a partial effect on NO bioactivity.

Conclusively, although ARBs appear to have some vasoprotective effects, in clinical trials only ACE inhibitors have proven their efficacy in reducing total mortality in subjects with heart failure [68], vascular disease [48], or diabetes [69]. However, not all clinical trials have demonstrated that ACE inhibitors reduce all-cause mortality in patients with atherosclerosis and preserved left ventricular function [70]. ARBs can be used instead of ACE inhibitors but no clinical outcome studies have shown a beneficial effect of ARB in stable CAD [71].

Renin Inhibitors

Aliskiren is the only direct renin inhibitor used in clinical practice for its antihypertensive properties. Animal studies suggest that aliskiren down-regulates pro-atherogenic cells and reduces atherogenesis and aortic plaque areas in atherogenic mice partly through attenuation of vascular nicotinamide adenine dinucleotide phosphate (NADPH) oxidase activity [38]. In line with this, Yamamoto et al. reported that in NOS deficient mice the combination of aliskiren and valsartan exerts synergistic organ-protective effects through synergistic attenuation of oxidative stress [72]. Despite the theoretical advantages of aliskiren in experimental studies, which have raised expectations, the neutral or disappointing results regarding clinical outcome [73] warrant careful extrapolations and further investigation.

Aldosterone Antagonists

Mineralocorticoid receptor blockers are mostly used in hypertension and heart failure. Aldosterone, apart from regulating salt homeostasis, also enhances fibrosis, inflammation and oxidative stress. Recently, numerous experimental studies have focused on the antioxidative potential of mineralocorticoid receptor blockers. Indeed in post-myocardial infarction rats, spironolactone and angiotensin II increased NADPH oxidase-dependent and mitochondrial superoxide production in myocytes, and the combination of an angiotensin II receptor blocker and spironolactone resulted in a synergistic attenuation of cardiac oxidative stress [40]. Moreover, in rats after himdlimb ischemia, eplerenone significantly decreased the NADPH oxidases in parallel with an improvement in the proliferation and function of endothelial progenitor cells, introducing a possible novel and effective therapeutic strategy for the repair of atherosclerotic cardiovascular diseases [39].

Beta-Blockers

Third-generation beta-blockers have been shown not only to avoid increased peripheral resistance, but also to stimulate vasodilation via various mechanisms. In particular, nebivolol which has the highest beta(1)-receptor affinity among beta-blockers, has been associated with improvement of endothelial function via its strong stimulatory effects on the activity of eNOS and its antioxidative effects on cellular superoxide dismutase and dimethylarginine levels [74, 75]. In particular, it has been found that nebivolol, beyond improving blood pressure levels, improved FMD and increased erythrocyte cellular superoxide dismutase levels indirectly, suggesting a beneficial effect mediated by increased NO bioavailability [76]. Furthermore, nebivolol has been shown to significantly affect the amount of ROS released from human ECs under oxidative stress. Particularly this effect may be mediated, at least in part, by the inhibition of endothelial NADPH oxidase and also by the direct ROS scavenging effect of the drug [41, 76, 77]. Interestingly, in patients with slow coronary flow nebivolol treatment for 6 months reduced the levels of malondialdehyde, superoxide dismutase (SOD) and erythrocyte catalase [78]. Furthermore, recent studies have documented the beneficial effects of nebivolol on arterial stiffness to a greater extent than previous agents, which may have significant clinical implications for its use in the treatment of CVD [55].

Statins

The 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA) inhibitors or statins are the most commonly used lipid-lowering agents. Recent compelling evidence suggests that their beneficial effects may include not only cholesterol lowering effects, but also cholesterol-independent or so-called pleiotropic effects. These effects are mediated by concomitant inhibition of protein isoprenylation, a process responsible for a variety of cellular responses downstream the mevalonate pathway [17]. Importantly, ex vivo experiments in internal mammary arteries incubated with either atorvastatin or atorvastatin with mevalonate have shown that in the presence of mevalonate atorvastatin failed to achieve a reduction in oxygen free radicals confirming the key role of mevalonate inhibition in the restoration of arterial redox state achieved with statins [17]. Through these pleiotropic effects, statins are directly involved in restoring or improving endothelial function, attenuating vascular remodeling, inhibiting vascular inflammatory response, and, perhaps, stabilizing atherosclerotic plaques [79, 80]. Also, they may ameliorate ED through their antioxidant properties, since they seem to attenuate Ang-II-induced free radical production in SMCs by inhibiting Rac1-mediated NADPH oxidase activity and down-regulating Ang-II type 1 receptor expression [81]. Moreover, there is evidence that fluvastatin may exert a superoxide or hydroxyl radical scavenging activity and diminish susceptibility to oxidation, while atorvastatin has been shown to reduce vascular mRNA expression of essential NADPH oxidase subunits p22phox and NADPH oxidase 1, increasing at the same time catalase expression both in vitro and in vivo [55, 82]. Further, in 12 dyslipidemic patients with ischemic heart disease who had received organic nitrates for a long period, fluvastatin reduced anti-oxidized LDL antibody titer and serum 8-hydroxydeoxyguanosine levels, while it attenuated nitrate tolerance in subjects under organic nitrites for a long-term period [83].

In addition, evidence suggests that statins exhibit beneficial properties in endothelial function and arterial stiffness in CVD [84]. Importantly, it has been shown recently that intensive therapy with rosuvastatin delayed progression of the mean-IMT within 12 month in hypercholesterolemic patients [85]. Likewise, dual lipid-lowering therapy has been linked with atherosclerosis regression, even though it has not been clarified whether it contributes to significant changes in plaque composition [86].

Thus, it seems reasonable that several data from observational and randomized studies have indicated favorable effects of statin therapy early after the onset of acute coronary syndrome (ACS). In secondary prevention of CVD, the MIRACL (Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering) study, randomly assigned 3,086 patients with unstable angina or non-Q-wave acute MI to atorvastatin 80 mg or placebo for 16 weeks. Lipid-lowering therapy reduced recurrent ischemic events, and mostly recurrent symptomatic ischemia requiring rehospitalization [87]. Of note, recent studies have provided evidence that statins may be administered as adjunctive therapy in ACS and may also improve clinical outcomes [88, 89].

Peroxisome Proliferator-Activated Receptors Agonists

Peroxisome proliferator-activated receptors (PPARs) agonists (fibrates, thiazolidinediones or glitazones) have been shown to exert a broad spectrum of antiatherogenic effects in vitro, in animal models of atherosclerosis and in humans. These agents, which are widely used in the clinical setting, antagonise Ang-II effects and have been shown to exert antioxidant and anti-inflammatory effects [90]. Interestingly, emerging data have suggested that they may augment endothelial NO release [91], decline NADPH-dependent O2 − production in human umbilical vein endothelial cells and also reduce relative mRNA levels of the NADPH oxidase subunits. Further, PPAR-γ ligands induce both activity and expression of Cu/Zn-SOD [92]. Moreover, in hypertensive rats pioglitazone reduces vascular ROS production and Nox1 levels while it does not affect eNOS expression [42]. Interestingly, novel thiazolidinediones (SF23) under investigation offer superior antioxidant effects compared to rosiglitazone, preventing ROS generation and the expression of NADPH oxidase subunits, Nox1 and Nox2 [93]. Unlike rosiglitazone, these effects are independent of nuclear factor erythroid 2-related factor 2. Although more data are required on the effects of PPARγ agonist on cardiovascular events, due to their interference with key processes of atherogenesis, they present additional properties to improve cardiovascular risk beyond glycemic control in patients with DM [94].

Xanthine Oxidase Inhibitors

Xanthine oxidase catalyzes the oxidation of hypoxanthine to xanthine and finally to uric acid. Xanthine oxidase is a source of ROS with detrimental effects on the progression of atherosclerosis. Experimental studies have also shown that xanthine oxidase binds to endothelial cells and inactivates NO [95].

Taking into consideration the physiological role of xanthine oxidase, inhibition of its activity may be beneficial for endothelial function. Several studies have examined the impact of xanthine oxidase inhibitors on endothelial function [96]. Xanthine oxidase inhibition has been tested in patients at increased cardiovascular risk [97], in patients with CAD [98] and with heart failure [99], with favorable effects on endothelial function. Moreover, Yiginer et al. concluded that in patients with metabolic syndrome allopurinol reduces oxidative stress, improves endothelial function and ameliorates myeloperoxidase, further confirming the pathophysiology of improvement of endothelial function. Although there is a lack of a large randomized trials, a recent meta-analysis supported the positive impact of such an approach [96].

Strategies to Increase Tetrahydrobiopterin

Tetrahydrobiopterin (BH4) is a pteridine that is known to have a plethora of cofactor and antioxidant roles in pathological conditions associated with cardiovascular and endothelial dysfunction, monoamine neurotransmitter formation, the immune response, and pain sensitivity [100]. In order for eNOS to transport electrons to L-arginine and synthesize NO, it has to be bound to BH4 (“coupling” of eNOS) [101]. Decreased levels of BH4 in endothelial cells can result in eNOS “uncoupling” and generation of oxygen radicals causing oxidative damage [102, 103].

Given the key pathophysiological role of BH4 in vascular function it is not surprising that redox disequilibrium can be achieved by increasing BH4 levels. Experimentally, incubation of isolated vessels (human coronary arteries) with BH4 results in an improvement of endothelial function [104]. Moreover, oral administration of BH4 to ensure continuous BH4 availability in hypercholesterolemic apolipoprotein E-knockout mice improves endothelial dysfunction and attenuates increased mRNA expression of NADPH oxidase components [105]. Similarly, in hypercholesterolaemic patients endothelial dysfunction and oxidative stress can be reversed by chronic oral treatment with BH4. Thus, oral supplementation with BH4 may provide a rational therapeutic approach to maintain NO synthesis and low levels of free oxygen radicals and to prevent cardiovascular disease [106]. Interestingly, statins can also increase BH4 levels. In patients with CAD atorvastatin not only increases BH4 levels with a parallel improvement in NO bioavailability, but also decreases oxygen radicals [17]. Moreover, in diabetic rats telmisartan prevented down regulation of BH4, eNOS uncoupling and the increase in NADPH oxidase and ROS with a parallel restoration of endothelial function. These findings provide evidence how telmisartan can restore endothelial dysfunction in diabetic models through up-regulation of BH4 [107].

Antioxidant Vitamins

Even though oxidative stress plays a vital role in the evolution of atherosclerosis, and antioxidants have been enthusiastically used in the treatment and prevention of cardiovascular disease, the role of antioxidant vitamins is a matter of long debate, as the results of prospective, randomized and clinical studies have not been as encouraging as expected [108]. Several data have illustrated positive responses of vitamin C and vitamin E. Specifically, they were considered as inhibitors of LDL oxidation by ROS scavenging and as mediators that increase NO bioavailability. Moreover, vitamin C has been shown to improve endothelial NO synthase (eNOS) coupling by scavenging ROS, while vitamin E suppresses eNOS expression (protein kinase C- dependent) [109, 110]. Moreover, several data have demonstrated positive responses of these vitamins in hypertensive patients, as combined treatment significantly improved ED and arterial stiffness, effects which were associated with changes in plasma markers of oxidative stress [111, 112]. Additionally, favorable effects of combined administration of vitamins C and E on endothelial function, inflammatory process and thrombosis/fibrinolysis have also been revealed in chronic smokers [113, 114]. In terms of CVD endpoints, a few studies using different combinations of antioxidant vitamins have reported encouraging results [108].

However, a large meta-analysis including over 77,000 subjects has shown neutral effects concerning the clinical outcome [115]. Similarly, vitamin supplementation has also been unconvincing in the progression of atherosclerosis [116]. Moreover, neither vitamin E nor vitamin C supplementation reduced the risk of major CVDs in the long-term [117], and likewise, in females at high risk for CVDs, no beneficial effects have been observed for ascorbic acid or β-carotene on cardiovascular events [118]. Thus, despite the encouraging data from basic-science studies using several vitamin supplements, antioxidant treatment with vitamins has not been proven an ideal strategy to reduce cardiovascular risk. Further, the effects of vitamins in some occasions have turned to be even detrimental, given that oral vitamin E supplementation may actually have pro-oxidant effects [109].

Folic Acid

Folic acid and its circulating metabolite 5-methyltetrahydrofolate (5-MTHF) have been shown to exert effects on vascular function, independently of the effects on oxidized or reduced plasma homocysteine [119, 120]. 5-MTHF, via improved eNOS activity and coupling as well as scavenging of peroxynitrite radicals, has been considered to improve endothelial NO bioavailability and decrease vascular superoxide production both in vivo and ex vivo, leading to a remarkable improvement of vascular tetrahydrobiopterin bioavailability [121, 122]. However, large randomized data derived from prospective folate supplementation revealed no beneficial effects of folic acid, B6 and B12 vitamins on cardiovascular risk [119, 123].

Omega-3 Polyunsaturated Fatty Acids

The omega-3 polyunsaturated fatty acids (omega-3 PUFAs) eicosapentaenoic acid and docosahexaenoic acid are present mainly in oily fish and commercially available supplements, which are available either over the counter (as fish oils) or as concentrated pharmaceutical preparations. The use of omega-3 PUFAs has been associated with substantial cardiovascular benefits in subjects with diabetes mellitus, heart failure and cardiovascular risk factors [124–126]. Omega-3 PUFAs also improve endothelial function and arterial elastic properties in subjects with metabolic syndrome and in healthy smokers [127, 128]. The beneficial effects of omega-3 PUFAs on endothelial function are attributed to reduced production of inflammatory cytokines, reduced levels of adhesion molecules and suppression of thromboxane production [129, 130]. Experimental data have also shown that omega-3 PUFAs exert favorable effects on lipid metabolism and on the oxidant/antioxidant status of offspring of diabetic rats [131]. Similarly, supplementation of human aortic endothelial cells with omega three fatty acids decreases ROS formation. Therefore it seems that this fatty acid acts as an indirect anti-oxidant in vascular endothelial cells, hence diminishing inflammation and, in turn, the risk of atherosclerosis and cardiovascular disease [132]. Moreover, 2 month treatment with omega-3 PUFAs in patients with atherosclerotic lesions significantly decreased malondialdehyde [133]. Recently, eicosapentaenoic acid and docosahexaenoic acid supplementation was shown to lower plasma lipoperoxide concentrations in mild cognitive impairment patients, further establishing the antioxidant role of omega-3 PUFAs [134].

Polyphenols-Flavonoids

It is evident that lifestyle factors such as smoking, alcohol consumption and obesity are risk factors for the development of impaired endothelial function. There is significant evidence that emphasizes the importance of a variety of strategies to use lifestyle modification to improve endothelial function and cardiovascular status. In this direction, many epidemiological research studies have shown that dietary factors, such as red wine, Greek coffee and tea consumption, rich in polyphenols, have been associated with a reduced risk of CVDs and beneficial effects on endothelial function [135, 136]. Polyphenols consist of a large number of molecules, further divided according to their chemical structure into the subcategories of phenolic acids, flavonoids, stilbenes, and lignans [137].

Even though the mechanisms by which polyphenols act as antioxidants are not fully elucidated, it seems that they are reducing agents or hydrogen atom-donating molecules, thus playing a role as free radical scavengers. In particular, several free hydroxyl groups participate in the scavenging effects of flavonoids, while hydroxyl groups in an ortho position of ring B is a key structure [138]. It has become evident that the most effective antioxidant actions of flavonoids are dependent on indirect mechanisms that inhibit enzymes such as xanthine oxidase, lipoxygenases, and NADPH oxidases which produce ROS [139]. Also, polyphenols exert antioxidant activities via inhibition of NADPH oxidase, 15-lipoxygenase, cytochrome p450 and myeloperoxidase, while further evidence advocates that they may prevent LDL oxidation by their binding to LDL, or act via alterations in gene expression resulting in changes in cell communication [140].

Many in vivo trials have assessed the effects of a variety of fruit polyphenols on different CVD risk factors, including ED [141]. Interestingly, the majority of data have indicated that grape products might produce hypotensive, hypolipidemic and anti-atherosclerotic effects, potentially through ameliorating oxidative stress as measured in terms of oxidation biomarkers and maintenance of endothelial function [10, 142, 143]. Randomized intervention studies in humans have presented additional evidence of a causal relationship between vascular health outcomes and flavonoid intake. A meta-analysis which included 133 human studies with polyphenol interventions showed that cocoa significantly increased acute and chronic FMD of the brachial artery [144]. The intake of polyphenol-rich sources (red wine, cocoa, green tea and berries) has been shown to favor cardiovascular health via an improved lipid profile, anti-atherosclerotic, anti-hypertensive and anti-inflammatory effects, as well as direct actions on ECs [145]. Indeed, green tea supplementation decreased several cardiovascular risk factors, including body composition, dyslipidemia, inflammatory status, and antioxidant capacity, in rats fed an atherogenic diet [146]. More evidence concerning the favorable effect of tea in endothelial function is provided by a recent meta-analysis of nine studies [147]. The effect of tea on FMD is constant and independent of the population and type of tea and can be attributed to the effect of dietary flavonoids in reducing SOD mediated NO breakdown [148] and the inhibition of NADPH oxidase activity [149].

Another source of phenolic compounds is olive oil, with hydroxytyrosol and oleuropein as the most representative [150]. The Mediterranean diet and olive oil consumption were associated with decreased risk factors and lower LDL cholesterol levels [151]. Moreover, olive oil consumption was associated with improved endothelial function [152, 153]. The ATTICA study documented that adherence to Mediterranean diet and increased olive oil consumption is associated with increased total antioxidant capacity in healthy subjects [154]. Further insights into the mechanisms of atherosclerosis protection by olive oil consumption are provided by the documentation of the inhibitory effect of hydroxytyrosol in superoxide anion and F2-isoprostanes production [155, 156]. Oleuropein inhibits also oxidation of LDL and superoxide production [157].

Despite the ample experimental and epidemiological data, there are only a few studies examining the effect of flavonoids on clinical endpoints in atherosclerosis [158].

Coenzyme Q10

Coenzyme Q10 is essential for mitochondrial oxidative phosphorylation and adenosine triphosphate production and is located mostly in the mitochondria but also in lysozomes, Golgi and plasma. It exerts antioxidant activities either by directly reacting with free radicals or by regenerating tocopherol and ascorbate from their oxidized state. Coenzyme Q10 is either synthesized in tissue or is obtained from diet [159–161]. In human umbilical vein endothelial cell cultures treated with oxidized LDL, coenzyme Q10 protects endothelial cells from oxidative stress-induced injury by up-regulation of eNOS and down-regulation of inducible nitric oxide synthase [43]. The antioxidant properties of coenzyme Q10 can possibly explain the improvement in endothelial function of subjects at risk, as documented in a recent metanalysis of five trials [162].

Endocannabinoid System- Cannabinoid Receptors Antagonists

Recent data suggests that the endocannabinoid system contributes to the progression of atherosclerosis and can modulate oxidative stress and endothelial dysfunction. Despite the fact that cannabinoid receptors antagonists are not in use in cardiovascular practice, several experimental studies have documented favorable effects. Tiyerilli et al. concluded that inhibition of the cannabinoid receptor one with rimonabant in apolipoprotein E deficient mice leads to decreased NADPH oxidase activity and ROS production together with endothelial protective effects [163]. Beneficial effects were also reported with rimonabant in subjects with metabolic risk factors and early atherosclerosis raising expectations for a novel and efficacious treatment of patients at risk [164]. Nevertheless, adverse psychiatric effects reported in clinical studies prevent cannabinoid receptors antagonists from being used in clinical practice.

NADPH Oxidase Inhibitors

As previously described, NADPH oxidases are key molecules in vascular oxidative stress. Consequently, approaches to inhibit their action have gained attention, as potential modification of NADPH oxidase can restore endothelial function and ameliorate the progression of atherosclerosis.

Diphenylene iodonium was one of the first oxidase inhibitors [165]. Data suggest that diphenylene iodonium abolished NADPH oxidase-mediated ROS formation, but also inhibited other flavo-enzymes such as NO synthase (NOS) and xanthine oxidase which may induce toxicity [7] .

Apocynin, is another direct NADPH oxidase inhibitor which has only minor efficacy due to the fact that it is a pro-drug requiring metabolic activation [7] . Nevertheless, Kinkade et al., in an atherosclerotic mice model, concluded that apocynin attenuates the progression of atherosclerosis, potentially by its ability to inhibit generation of superoxide by NADPH oxidase through its ability to inhibit translocation of the subunit p47phox (subunit of the NADPH oxidase) [166]. Liu et al. also found that in mice inhibition of NADPH oxidase by apocynin reduced endotheial NOS activation and platelet adhesion, which are likely responsible for the arrest of plaque growth and improvement of vascular mechanical properties [167]. Similarly, 7 days treatment with apocynin reduced endothelial cell adhesion molecule expression in atherosclerotic mice but without a detectable change in oxidative burden [168].

Several other direct NADPH inhibitors have also been studied with limited efficacy and specificity [169]. Favorable results have also been reported with selective NADPH oxidase inhibitors which selectively bind to the p47phox subunit of NADPH oxidase inhibited Nox1 and Nox2 but not Nox4 [170].

Interestingly, the novel NADPH oxidase inhibitor VAS3947, used in low micromolar concentrations, consistently inhibited NADPH oxidase activity, but did not inhibit xanthine oxidase or endothelial NOS activities. Nevertheless, the mechanisms of actions of triazolo pyrimidines such as VAS3947 are unclear and we cannot exclude the possibility that VAS3947, in addition to inhibiting NADPH oxidases, also interferes with alternative sources of ROS that have not yet been elucidated, such as the mitochondrial electron chain [7].

Despite the preliminary positive results regarding the use of direct NADPH oxidase inhibitors we have to notice that data are based on experimental studies. There are no clinical data yet and a major efforts have to be made before the possible use of this agents in clinical practice.

Conclusions

The endothelium is considered of major importance in maintaining vascular homeostasis while oxidative stress has a central role in the progression of endothelial dysfunction. Several established treatments, with proven cardiovascular utility, such as angiotensin converting enzyme inhibitors and angiotensin receptor blockers, beta-blockers and statins, beyond their main action, have antioxidant effects and can restore endothelial function. Additionally, novel strategies such as antioxidant supplementation, peroxisome proliferator-activated receptors agonists and dietary flavonoids may also have a role against oxidative stress and endothelial dysfunction, exerting anti-inflammatory and anti-thrombotic actions at the same time. Nevertheless, their impact on clinical outcome has not been established yet and further studies are required to establish their clinical usefulness.