Diagnosis and individual treatment of cardiovascular diseases: targeting vascular oxidative stress

Abstract

Cardiovascular diseases are the leading cause of death and disability worldwide, yet we do not fully understand their underlying causes to reliably identify and treat, let alone prevent, these diseases. The majority of therapeutic approaches are symptom orientated, and current practice often follows a ‘one-fits-all’ approach. New strategies are needed, which harness the potential of individualized medicine with its three major pillars: in vitro diagnostics for early identification of individuals at risk and monitoring of drug efficacy; molecular imaging for disease localization and monitoring; and innovative, mechanism-based drugs. One so far untargeted mechanism of cardiovascular disease is oxidative stress, that is, the increased occurrence of reactive oxygen species in the vascular wall that leads to endothelial dysfunction. We outline why previous antioxidant supplements do not work and suggest an alternative approach targeting the enzymatic sources of oxidative stress and using emerging biomarkers of oxidative stress. These and similar approaches may be applied to fewer patients but in a much more individualized, effective and cost-saving manner.

Keywords::

• cardiovascular diseases
• individualized medicine
• NADPH oxidase
• nitric oxide
• oxidative stress
• prevention
• soluble guanylate cyclase
• VASP

Figure 1. Reactive oxygen and nitrogen species.

Oxygen can be activated in several reductive steps, resulting in the formation of the superoxide anion radical (O₂•^-) or hydrogen peroxide (H₂O₂). The second reduction can occur spontaneously or be catalyzed by SOD. Hydrogen peroxide is detoxified by Cat into oxygen (not shown) and water. Superoxide or hydrogen peroxide can also react with nitric oxide (NO•) or NO₂^-, respectively, to generate the very reactive ONOO^-. The iron salt-dependent decomposition of hydrogen peroxide generates the highly reactive hydroxyl radical (OH•; Fenton reaction). The hydroxyl radical initiates lipid peroxidation of cholesterol (not shown) and polyunsaturated fatty acids (L-H), resulting in the formation of a fatty acid radical (L•). Fatty acid radicals are unstable and readily react with molecular oxygen (not shown) resulting in a peroxyl fatty radical, which is unstable too and can react with another free fatty acid (chain reaction). Lipid peroxdation can occur enzymatically and nonenzymatically and results in a variety of products (for details see [73]).

Cat: Catalase; NO₂^-: Nitrite; ONOO: Peroxynitrite; SOD: Superoxide dismutase.

Modified from [74].

Figure 2. Enzymatic sources of reactive oxygen species and nitric oxide, their effects and pharmacological modulation.

NOS generate NO from the amino acid L-Arg. Part of the NO-dependent signal transduction occurs through protein S-nitrosylation (not shown) with roles in physiology and pathopysiology, the latter correlating with hypo- or hyper-S-nitrosylation. In addition, NO activates its receptor sGC. Upon activation, sGC generates the second messenger cGMP, which activates cGMP-dependent protein kinase (not shown) mediating many of the vasoprotective effects of NO. cGMP is degraded to GMP by PDEs. Ang II induces oxidative stress by activating NOX. ROS generated by NOX mediate a variety of physiological and pathophysiolgical effects. For example, ROS mediate the accumulation of the ADMA. ADMA is an endogenous inhibitor of NOS. ROS also oxidize the NOS cofactor tetrahydrobiopterin, resulting in the uncoupling of NOS. ucNOS generates ROS instead of NO. In addition, ROS oxidize sGC. In the first step, the heme iron of sGC is oxidized (ox-sGC), and in the second step, the heme group is released from sGC, resulting in heme-free apo-sGC. Ox-sGC and apo-sGC cannot be activated by NO. apo-sGC can be re-activated by sGC activators, whereas sGC stimulators activate only the reduced form of sGC. sGC stimulators allow maximum sGC activation even at reduced NO concentrations. By inhibiting PDEs and thus the breakdown of cGMP, PDE inhibitors result in increased cGMP concentrations. Antioxidants scavenge ROS. Some of the pleiotrophic effects of statins are mediated by inhibition of NOX. Drugs affecting the renin–angiotensin system mediate an inhibition of NOX via Ang II. eNOS enhancer and flavonoids from red wine and dark chocolate increase the expression of NOS, at least in vitro. Nitrate(s) either from drugs or vegetables (e.g., beetroot) are converted to NO. The processes within the white area occur endogenously, the compounds outside the white area indicate exogenous factors or drugs.

ACE: Angiotensin-converting enzyme; ADMA: Asymmetric dimethyl arginine; Ang II: Anigotensin II; apo-sGC: Heme-free sGC; AT1R: Angiotensin II type 1 receptor; cGMP: Cyclic GMP; eNOS: Endothelial NOS; L-Arg: L-Arginine; NO: Nitric oxide; NOS: Nitric oxide synthases; NOX: NADPH oxidases; ox-sGC: Heme-bound sGC; PDE: Phosphodiesterase; ROS: Reactive oxygen species; sGC: Soluble guanylyl cylcase; uc-NOS: Uncoupled NOS.

Modified from [74].

Connecting oxidative stress, endothelial dysfunction & cardiovascular diseases

Cardiovascular diseases (CVDs) are the major cause of death and disability worldwide. A functional impairment of the vascular endothelium, caused by a disruption of the protective nitric oxide (NO) signaling pathway [1], results in decreased relaxation responses of blood vessels, vascular wall thickening and a prothrombotic state [2]. This state is termed ‘endothelial dysfunction’ and represents a hallmark of CVD. Indeed, development and prognosis of CVD are often associated with endothelial dysfunction.

Nitric oxide can be generated enzymatically by NO synthases (NOSs) or nonenzymatically from nitrite (NO₂^-). Enzymatically, NO is synthesized from the amino acid L-Arginine in a two-step process via formation of N-hydroxy-L-arginine. Three NOS isoforms exist:

• • Neuronal NOS (nNOS; NOS1);

• • Inducible NOS (iNOS; NOS2);

• • Endothelial NOS (eNOS; NOS3).

In the vasculature, NO derived from eNOS plays a critical role in the regulation of vascular tone, growth, remodeling and the thrombotic balance [3]. In the nervous system, NO synthesized by nNOS acts as a neurotransmitter. In contrast to eNOS and nNOS, iNOS is an inducible isoform and is upregulated during inflammatory responses. In animal studies, induction of iNOS leads to the production of large quantities of NO that can have detrimental effects. Indeed, it has been proposed that the induction of iNOS is involved in the pathophysiology of sepsis. In this scenario, inhibition of iNOS might represent a logical therapeutic approach. However, respective clinical studies have shown negative results [4–6].

Nitric oxide production and NO bioavailability are compromised under conditions of oxidative stress, that is, an imbalance between the production and removal of reactive oxygen species (ROS; Figure 1). Owing to the nature of their chemistry, ROS and free radicals are highly reactive and overproduction can lead to oxidative damage of cellular components such as DNA and proteins [7]. However, ROS also have important physiological functions. For example, they are essential in immune defense as well as in the regulation of cellular growth and gene expression [7].

The link between endothelial dysfunction and the reduced activity as well as bioavailability of NO caused by oxidative stress may be caused by several mechanisms:

• • Reactive oxygen species directly inactivate NO. NO and ROS can react with each other generating peroxynitrite (ONOO^-), the most reactive compound of all ROS. It can oxidize and nitrite proteins, lipids and nucleic acids [8];

• • Reactive oxygen species uncouple eNOS. ROS oxidize the essential NOS cofactor tetrahydrobiopterin (BH₄) leading to ‘uncoupling’ of eNOS. Uncoupled NOS produces ROS itself instead of NO [1];

• • Oxidative stress alters the NO receptor soluble guanylate cyclase (sGC), making it insensitive to the activation by NO;

• • Oxidative stress interferes with the activity of enzymes involved in the formation (protein arginine N-methyltransferase [PRMT]) and degradation (dimethylarginine dimethyl-aminohydrolases [DDAH]) of the endogenous NOS inhibitor, asymmetric dimethyl-L-arginine (ADMA). ADMA inhibits eNOS by competing with the substrate L-Arginine for NOS [9].

Inter- and intra-cellular signaling of NO involves post-translational modifications and functional regulation of proteins. In this article we focus on the NO–sGC–cyclic GMP (cGMP) signaling pathway and cGMP-dependent protein kinase-mediated protein phosphorylation (see later and Figure 2). Nevertheless, S-nitrosylation of cysteine residues may also mediate part of the effects of NO including both hypo- and hyper-S-nitrosylation in different disease states, including CVD. For more information on S-nitrosylation we refer the reader elsewhere [10].

Since disruption of the protective NO signaling pathway by ROS [1] leads to a functional impairment of vascular endothelial function, therapeutic strategies that can preserve or re-constitute endothelial function by reducing oxidative stress hold promise for the prevention or treatment of certain forms of CVD, respectively.

The ‘one-fits-all’ approach

It is important to critically acknowledge – despite success – where cardiovascular medicine currently stands: patients at risk are not identified early enough and a heart attack or stroke may be the first sign of disease, which is often fatal. Since we do not know the mechanisms of CVD, drugs often target symptoms, but not the underlying mechanisms. Therapies cannot be monitored efficiently and, therefore, treatments are not tailored to the individual patient. Instead, current therapies are administered uniformly across a heterogeneous spectrum of disease etiology (e.g., arterial hypertension).

On an individual level, the available measures of risk are too crude, too late and too imprecise. The key for improving health outcomes is the identification of individuals who are disease free but at high risk for developing CVD later in life. There is a need to reliably detect subclinical disease states in order to establish individually tailored preventive measures and the monitoring of therapies. In addition, segmentation of complex diseases into subclassifications through the use of novel biomarkers and their profiles are likely to improve our understanding of the diseases and guide the way towards new mechanism-based therapies. Currently, no technology is available in the clinic to reliably measure early markers of disease. Should they become available, the identification of patients at risk would be revolutionized towards a much more individualized and effective form of medicine that has the potential to prevent many cardiovascular events.

Traditional risk factors, such as high low-density lipoprotein levels and arterial hypertension, can provide a reasonable risk prediction on a population level. Indeed, approximately 50% of strokes and ischemic heart disease events may be attributed to high blood pressure. However, not every individual with these risk factors will develop CVD. How can we identify those individuals that are truly at risk? What causes the rise in blood pressure in the first place? There are a variety of causes, which may need different treatment strategies. Inhibitors targeting the renin–angiotensin system, β-adrenoceptors and diuretics are effective in some, but do not work for all patients. There will always be a group resistant to current medications.

The recently suggested polypill is a ‘one-fits-all’ approach to treating CVD. The advantage of such a pill is that levels of compliance may be increased as patients now take only one pill instead of several. The feasibility of this approach has been shown recently [11]. However, there is the danger that the pill could replace lifestyle interventions. Moreover, while some risk factors are reduced, no data on outcomes are available, nor is the full long-term safety profile explored (patients would take the pill for many years).

Therefore, the one-fits-all concept needs a shift to the paradigm of personalized medicine (PM), that is, using information about a person’s biochemical makeup to individually tailor treatment strategies, ensuring ‘the right drug for the right patient at the right dose and time’. While only large-scale clinical trials will provide information about the accuracy of new biomarkers and their pattern as well as novel molecular imaging techniques in the general population, therapy should no longer rely on generalizations from large patient cohorts where a statistical significance cannot be traced back to an individual. Instead, it should be possible to monitor, and tailor therapies based on relevant predictors of outcomes, to improve therapeutic efficacy and ideally complete event prevention. In that respect, recent publications have questioned long-standing dogmas of relevant treatment targets [12,13]. In addition, adverse drug reactions may be predicted or monitored more closely.

Personalized medicine: the three pillars

New in vitro diagnostics

Under current practice, when a typical patient is diagnosed with CVD, the disease has often been progressing for many years. Therefore, there is a need to identify individuals at risk for CVD earlier. Several novel biochemical markers for cardiovascular risk stratification have been identified recently, for example, inflammatory markers such as C-reactive protein. However, none of the emerging biomarkers appear to be ready for clinical use. Even though some of these have been reported to be associated with increased relative risks of cardiovascular events independently of established risk factors, when added to traditional risk factors they did not substantially improve individual risk stratification [14].

In addition, the use of a pattern of blood biomarkers would be a major step forward. Thus, harnessing proteomic and metabolomic technologies such as antibody-based microarrays to assess variations in proteins and metabolites has great potential. It can also be applied for the identification of orthological biomarkers (-pattern), which are unlikely to be found by focusing on well-studied pathways [15]. The feasibility of such an approach has been provided for Alzheimer’s disease – Ray et al. established a highly specific plasma biomarker phenotype that can characterize Alzheimer’s disease years before a clinical diagnosis can be made [16].

With these advances, we have the opportunity to expand our armamentarium of diagnostic tests to detect subtle and early signs of vascular diseases. Early detection of patients at risk before symptoms occur has the potential to provide major opportunities for the prevention of many cardiovascular events, which will allow a more effective use of resources and reduction of disease burdens.

Molecular imaging

Current cardiovascular imaging technologies visualize end-stage morphology rather than early disease – we can detect plaques, but we cannot differentiate stable from vulnerable plaques. We can detect an obliterated blood vessel but we cannot detect the process that will lead to blood vessel wall thickening several years down the track. The development of molecular imaging techniques that localize an individual’s risk to specific vascular beds, at a stage where no morphological changes can be observed but where altered cellular or vascular functions or metabolites are present, would revolutionize disease diagnosis. Molecular imaging may also be used for monitoring a vascular segment’s response to drug therapies. Indeed, besides plasma values and a calculated risk score, an image of an atherosclerotic vascular wall or a labile plaque in a vessel could motivate patients to change their lifestyles.

Mechanism-based drugs

Mechanism-based drugs are likely to allow a shift from symptom-based therapy to therapies that target disease mechanisms. In addition, therapeutic monitoring using in vitro diagnostics and molecular imaging could be applied to ensure that any therapy is optimized for each patient. This approach has already resulted in major breakthroughs for cancer therapy [17,18]. However, cardiovascular medicine is lagging behind in embracing this concept. In future, it might be possible to monitor as well as adapt preventative and curative, anti-remodeling therapies based on relevant predictors of outcome. This may then lead to improved therapeutic efficacy, ideally to complete event prevention. Therefore, future cardiovascular medications may not target standard parameters but the pathomechanisms that are relevant for each individual patient.

Personalized medicine: proof of concept

In oncology, PM is, in part, already in clinical practice with significant outcomes [19]. Molecular approaches to discover the underlying causes of oncogenesis already allow the identification and classification of subgroups of cancer patients based on the underlying molecular mutations, gene and protein expression profiles of the tumor. For example, Gleevec^® (imatinib) is a tyrosine kinase inhibitor that inhibits kinases such as c-kit and c-abl. It is used in the treatment of chronic myeloid leukemia as patients with this disease have a mutant, constitutively active BCR–ABL kinase [20]. In breast cancer, identification of patients with HER2 positive tumors allows for the selection of the best treatment as only women with the HER2 marker will respond to trastuzumab, a HER2-specific monoclonal antibody [17].

With respect to imaging, PET/computed tomography or single photon-emission computed tomography (SPECT)/computed tomography imaging visualizes the effectiveness of therapeutic interventions – not only the localization and size of the malignancy, but also the tumor metabolism, tumor hypoxia and gene/protein expression and activity with radiolabeled probes [21]. This has important prognostic implications as it may indicate the response of the tumor to therapy prior to evidence observed clinically.

While the area of cancer is leading the field of PM, we believe that CVD will follow soon. One area, among others, which certainly has the potential provide new strategies for diagnosis followed by individualized and mechanism-based treatment of CVD is oxidative stress.

Why pharmacological approaches using antioxidant supplements were unsuccessful

Experts forecast a worldwide epidemic of CVD, not only in industrialized, but also in developing countries. Therefore, in parallel with optimized CVD treatment, a considerable amount of effort should be directed towards preventive strategies.

Oxidative stress is likely to play a major role in the development and progression of many diseases and thus an attractive target to treat and prevent these diseases. The strategy to counteract oxidative stress and thus prevent CVD by supplementing antioxidants, that is, attempting to scavenge ROS after their production, has been pursued for many years. However, respective clinical studies using antioxidant supplements have not resulted in beneficial outcomes [22]. By contrast, antioxidants in high doses, such as vitamin E or β-carotene, may even be harmful and increase overall mortality [22,23].

There are several plausible explanations for these observations. Oxidative stress is rarely systemically evenly distributed. Rather, it is localized in individual organs, tissues and cells, or subcellular compartments. However, antioxidants are likely to be distributed more systemically – at least not targeted to the precise localizations where ROS concentrations are elevated. The bioavailability and distribution of antioxidants within different tissues is unknown, and the rate of reaction between antioxidants and ROS are vastly lower than the rates between ROS and their targets. For example, the rate of reaction between superoxide and NO is 6.7 × 10⁹ M^-1s^-1[24], while the rate of reaction between vitamin E and superoxide is 4.9 × 10³ M^-1s^-1[25]. After their reaction with ROS, antioxidants can become radicals themselves, initiating new radical chain reactions, and thus causing harm. In addition, untargeted scavenging of all ROS by antioxidants is likely to interfere with the physiological functions of ROS [26]. These functions differ for different types of ROS, and counteracting specific types of ROS may be more successful. Moderate or acute oxidative stress may even be good, as it may trigger the upregulation of endogenous antioxidative defense mechanisms as a reflective response [27], which is similar to a vaccination. Antioxidants are likely to prevent this adaptive response [27]. For example, the health-promoting effects of exercise can be partially negated by supplementation with vitamins C and E [28]. Prophylactic or therapeutic supplementation with, for example, vitamin E can therefore no longer be justified [22]. However, it may be that individuals with confirmed localized or systemic oxidative stress may benefit from vitamin E. Nevertheless, this leaves us with the problem of accurately measuring oxidative stress in vivo.

Are current biomarkers of oxidative stress reliable?

Oxidative stress cannot be defined in universal terms. Different diseases or their subclasses may be associated with different types of oxidative stress, for example, different types of ROS that need different markers or indices. Currently, several methods are used to quantify ‘oxidative stress’ in humans. These include assays that measure the concentrations of oxidation products of lipids, proteins and DNA, assays for evaluation of the oxidative and reductive capacity of biological fluids, including determination of antioxidants, protein-SH-groups, and assays for evaluation of the ex vivo susceptibility of lipids to oxidation [29]. Accurate markers for the quantification of the redox balance in organisms or single organs do not exist [22,29]. In particular, the commonly used criteria based on lipid peroxidation cannot be regarded as a general estimate of the individual ‘oxidative stress’, and no available method is an appropriate criterion for defining oxidative stress [29]. Isoprostanes, for example, are often referred to as the ‘gold standard’ for the measurement of oxidative stress. While they can be accurately measured in biological fluids, some concerns have been raised. Often, single spot measurements are performed, although the timing of sampling is crucial. Also, the kinetics of isoprostane occurrence in plasma and urine differs, and there is no agreement regarding whether results should be standardized (and if so, to what parameter) [30]. Therefore, accurate diagnostics and, in particular, molecular imaging technologies for subtle classifications and the precise localization of oxidative stress would be a major advantage. Here too, a one-fits-all approach is unlikely to lead to advances. It is more likely that a combination of methods and biomarkers are needed for each individual, and different types and localizations of oxidative stress should be evaluated by different indices.

Novel therapeutic strategies: targeting mechanisms of CVD

Inhibiting the sources of reactive oxygen species

Oxidative stress is often caused by an overproduction of ROS, less often by their reduced breakdown [7]. Therefore, inhibition of ROS production to prevent oxidative stress or to reverse it may have great potential for future therapies of certain CVDs. Only one enzyme family has been identified whose only function is to generate ROS – NADPH oxidases [7]. Other enzymatic sources of ROS, for example, xanthine oxidases, cyclooxygenases, lipoxygenases, uncoupled NOS, cytochrome P450 enzymes and enzymes of the mitochondrial respiratory chain, have different functions and catalytic products as well. They rather generate ROS as a ‘metabolic accident’, as a side product or when they are converted into a dysfunctional state. Thus, they are unlikely to represent primary sources of oxidative stress. NADPH oxidases were discovered in phagocytic cells where they are responsible for the respiratory burst [7,31]. In addition, several nonphagocytic NADPH oxidases have been identified. NADPH oxidases are present in most, if not all, organs and tissues. They have various physiological functions, which are not yet fully understood. These include signal transduction, regulation of gene expression, cell proliferation and cell differentiation [7]. NADPH oxidases are enzyme complexes composed of different subunits. The catalytic activity is conferred by the membrane-spanning NOX subunit. NOX are flavin- and heme-containing proteins that transfer electrons from NADPH to oxygen, releasing ROS. Five NOX isoforms exist (NOX1–5), of which NOX1, NOX2, NOX4 and NOX5 seem to be relevant in the vasculature [31,32]. One of them, NOX5, is not expressed in rodents and is active independently of other subunits [7]. Since NOX5 is directly regulated by calcium, it may be the link between calcium overload of blood vessels and oxidative stress [7,31]. It was shown that NOX5 protein levels are increased in the coronary arteries of patients with coronary heart disease [33]. Interestingly, NADPH oxidases produce ‘kindling’ radicals that can uncouple eNOS and upregulate xanthine oxidases, thus promoting further generation of ROS [34]. NADPH oxidases have been involved in a variety of cardiovascular pathologies, including, for example, hypertension [7,35], myocardial infarction [7], heart failure [7] and stroke [7,36]. Currently, the development of specific pharmacological inhibitors for NADPH oxidases is still in its infancy [37]. Thus, any associated hopes are limited to the future.

Vascular NADPH oxidases are activated by different pathological stimuli, such as angiotensin II, glucose and oxidized low-density lipoprotein (Figure 2)[7]. Therefore, angiotensin-converting enzyme inhibitors and angiotensin II receptor antagonists may indirectly inhibit NADPH oxidases and thus reduce oxidative stress [7]. In line with this, angiotensin II-induced hypertension is reduced in NOX1-knockout mice [7]. Inhibition of NADPH oxidases may also explain some pleiotropic effects of statins, which inhibit isoprenylation of the cytosolic NADPH oxidase subunit Rac [38].

Increasing the bioavailability of vascular NO

One strategy for increasing vascular NO, which was pursued for a couple of years, is to increase plasma L-Arginine concentration by oral supplementation of the NOS substrate L-Arginine. Although these studies may have seemed improbable from the beginning, as the Michaelis–Menten constant Km of the NOS enzymes is approximately 5 µM while the intracellular concentration of L-Arginine is 0.3–1 mM [39], some clinical trials showed beneficial effects of L-Arginine administered in doses of at least 3 g/day, for example, improving the walking capacity of patients with peripheral artery disease [40]. This phenomenon that L-Arginine improves NO-mediated vascular function in vivo, although its baseline plasma concentration is approximately 25- to 30-fold higher than the Km of isolated, purified eNOS in vitro is called the ‘L-arginine paradox’. The most likely explanation is the occurrence of the endogenous eNOS inhibitor ADMA (Figure 2). Other explanations for this paradox are the upregulation of arginase activities and that neither extracellular nor intracellular L-Arginine concentrations determine NOS activity but rather the L-Arginine amount transported across the plasma membrane. Furthermore, subcellular concentrations may play a role, since local concentrations in the vicinity of eNOS may be lower than whole-cell concentrations [41]. However, most positive studies were short in duration. In contrast to short-term supplementation, L-Arginine may not be beneficial if administered to chronic patients [42], but this study (Vascular Interaction with Age in Myocardial Infarction [VINTAGE-MI]) has been criticized for serious flaws [41]. A recent meta-analysis concluded that short-term L-Arginine treatment only improved endothelial function in those individuals with endothelial dysfunction, as determined with flow-mediated vasodilation (FMD) [43]. Therefore, studies on L-Arginine therapy should include an assessment of endothelial function and of the L-Arginine:ADMA ratio [41]. As these measures may be a marker to predict the clinical effectiveness of L-Arginine individually, this is an example of a therapy that should be personalized. Current methods to assess endothelial function include FMD, flow-mediated MRI, pulse wave analysis and pulse contour analysis. Of these methods, FMD is the most commonly used. Its disadvantages include cost, operator dependency, large variations between populations (possibly due to nonstandardization of the technique), time needed for the assessment of FMD and influence by events such as fluctuations in the menstrual cycle. flow-mediated MRI is less operator dependent, but is more expensive, less widely used than FMD and has limitations associated with all forms of MRI [44]. Pulse wave analysis and pulse contour analysis have been shown to produce more variability than FMD [45]. Unfortunately, no biochemical marker exists to precisely measure the functional state of the endothelium, which would make assessment of endothelial function more practicable.

However, other options to increase vascular NO are to treat patients with the NOS cofactor BH₄[46], or to increase eNOS expression, the so-called ‘NOS enhancers’. The latter strategy has not yet reached the clinics. Nevertheless, it is promising that the eNOS enhancer AVE 9488 protected against ischemia-reperfusion damage in a mouse model [47].

Stimulating & activating soluble guanylate cyclase

Soluble guanylate cyclase is the receptor of NO. sGC is a heme containing enzyme that generates the intracellular second messenger cyclic GMP from GTP when NO binds to sGC [48]. This pathway mediates many physiological functions of NO. Interestingly, this pathway is targeted by a therapy for the acute treatment of angina pectoris and heart failure, which is over 100 years old, that is, organic nitrates that release NO. Owing to the development of nitrate tolerance, the use of most organic nitrates is restricted to short treatment intervals. Nevertheless, specific nitrates may be used for chronic treatments, as shown in the African–American Heart Failure Trial (AFeHT), in which chronic nitrates used in combination with hydralazine increased survival by 43% in an African–American subgroup with advanced heart failure [49]. Interestingly, nitrates may activate NADPH oxidases, resulting in enhanced ROS production and uncoupling of eNOS [50].

Recently, novel strategies to activate sGC have been introduced: sGC stimulators and sGC activators, both of which bind to sGC. sGC stimulators require a reduced sGC heme moiety and potentiate the effects of NO on sGC [48]. This results in maximal stimulation of sGC at reduced NO concentrations and synergism between NO and the sGC stimulator. One orally available sGC stimulator, riociguat (BAY 63-2521), is in clinical development for pulmonary hypertension. It was already demonstrated in a Phase II study that riociguat improved the exercise capacity, the stroke volume of the heart, and lowered the resistance of pulmonary vessels [51].

In contrast to sGC stimulators, sGC activators activate sGC independently of NO, by only interacting with NO-insensitive apo- (i.e., heme-free) sGC. The heme of sGC is released after its oxidation by ROS. An intermediate state is formed, sGC with the oxidized heme bound to it (ox-sGC). Interestingly, ox-sGC and heme-free sGC (apo-sGC) are increased in CVD that are associated with oxidative stress [48].

In addition to being a novel therapeutic target, apo-sGC may also become a biomarker to individualize the treatment with and predict the effectiveness of sGC activators, since these compounds may selectively act on heme-free sGC. It was indeed demonstrated that vascular apo-sGC is increased under pathological conditions such as diabetes [52]. Phospho-fingerprints of cGMP-dependent protein kinase (cGK) substrates may prove to become valuable biomarkers for the sGC signaling state. sGC activators are in clinical development (Phase II) for heart failure, so far with promising results [53]. An already established biomarker of the NO–sGC signaling cascade is the phosphorylation of the cGK substrate vasodilator-stimulated phosphoprotein (VASP) at Ser-239 [2], which is used to monitor the efficacy of treatments with antiplatelet drugs [54].

The importance of lifestyle for prevention

What can be recommended until such mechanism-based drugs and biomarkers become available? Eating a lifelong diversified diet rich in fruits and vegetables that contain many protective factors is a major recommendation [55]. Interestingly, studies demonstrated that some foods interact with the NO signaling pathway. For example, nitrate from vegetables is transformed into nitrite and finally NO [56,57]. Indeed, beetroot juice acutely lowered blood pressure in healthy subjects, prevented endothelial dysfunction induced by acute ischemia of the forearm and reduced platelet aggregation [57]. This may be one mechanism that mediates the cardioprotective effects of vegetables [57,58]. Dark, flavonoid-rich chocolate also lowers blood pressure, which is at least in part mediated by enhanced NO bioavailability [59,60]. The extent of the blood pressure-lowering effect of cocoa-containing foods is comparable with that of monotherapy with a β-blocker or angiotensin-converting enzyme inhibitor [61]. Only 30 g/day of dark chocolate is needed to achieve this [59]. However, one should keep in mind that chocolate is a confectionary that reflects on bodyweight. Polyphenols from red grapes stimulate the production of NO and inhibit NADPH oxidases, at least in animal and in vitro models. These effects may explain the cardioprotective effect of moderate red wine consumption [60]. However, it may never be possible to identify which of the myriad of compounds in foods confer their beneficial effects, and complex interactions of numerous food ingredients cannot be replaced by supplements.

There is accumulating evidence that exercise is an effective lifestyle feature for preserving and restoring endothelial function [62]. It may also have therapeutic implications. For example, in patients with coronary artery disease, exercise training reduced the expression of NOX2, NOX4 and p22phox in the coronary arteries. ROS production in aortic rings was also reduced, and endothelial function improved, showing the potential of exercise to influence the antioxidant capacity [63]. In mice, voluntary exercise restored age-associated endothelial dysfunction. This effect was mediated by reduced oxidative stress via stimulation of superoxide dismutase antioxidant activity and NADPH oxidase inhibition. eNOS protein and activation also increased [64]. Interestingly, the health-promoting effects of exercise can be partially negated by vitamin C supplementation. It was suggested that exercise-induced oxidative stress causes an adaptive response promoting endogenous antioxidant defense capacity. Supplementation with antioxidants may preclude these health-promoting effects of exercise in humans [28].

Smoking is an obvious risk factor for the development of CVD. In the context of oxidative stress, smoking increases the expression of NADPH oxidases, and increases superoxide and hydrogen peroxide production in the vasculature [65]. Cigarette smoke itself contains ROS and free radicals of organic compounds. There is only one intervention to prevent the deleterious effects of cigarette smoke and that is cessation.

High salt intake is associated with an increased incidence of strokes and total cardiovascular events [66]. Indeed, evidence linking sodium chloride intake with elevated blood pressure is without dispute [67]. Our current consumption of salt is a major factor for increasing blood pressure in salt-sensitive individuals. Modest reductions in salt intake result in a fall of blood pressure equivalent to single drug therapy in hypertensive individuals, and this is additive to antihypertensive drug treatments. A significant effect on blood pressure is also observed in people with normal blood pressure [68]. Importantly, there is evidence that patients with drug treatment-resistant hypertension are extremely sensitive to the blood pressure-lowering effect of sodium reduction that are equivalent to adding two antihypertensive medications [69]. This shows that lifestyle modifications are effective in combination with current drug treatments and sodium reduction may result in diminished need for the number of antihypertensive drugs in patients with resistant hypertension. Interestingly, it was suggested that a low-sodium diet also decreases oxidative stress and improves vascular function in salt-sensitive subjects [70].

Thus, to prevent CVD and other chronic diseases, a healthy lifestyle is a key element [71]. However, currently used healthy lifestyle promotions do little to encourage these lifestyles as many patients are not responsive to such general recommendations. Precise diagnostics for monitoring the results of lifestyle changes would help in passing the preventive message to an individual [55]. These will allow for individual optimization of interventions and may be the same as those used for screening diagnostics. As with drugs, lifestyle programs will be more successful if targeted to individuals.

Expert commentary

Novel therapeutic strategies targeting endothelial dysfunction by reducing oxidative stress hold promise for the prevention and treatment of CVD. This is despite the failure of clinical trials using antioxidant supplements to reduce the burden of CVD, as inhibiting the sources of oxidative stress is likely to be superior compared with scavenging ROS after they have been produced. Furthermore, emerging therapies that re-activate oxidatively damaged enzymes such as uncoupled NOS and ox/apo-sGC are likely to advance cardiovascular therapies.

Currently, most cardiovascular therapies are not tailored to individual patients and the identification of patients at risk occurs too late to prevent onset of often disabling or even fatal CVD. However, personalizing cardiovascular medicine, including the application of novel in vitro diagnostics, molecular imaging and mechanism-based drugs, has the potential to prevent many cases of CVD and advance their therapies in the future, which cannot be achieved by current one-fits-all approaches.

Five-year view

Soluble guanylate cyclase stimulators and activators are novel approaches to treat CVD that are likely to improve clinical practice within the next 5 years. NADPH oxidases are another attractive therapeutic target for the prevention and therapy of CVD associated with oxidative stress. However, clinical proof is still several years away – the first clinical trials might not even commence within 5 years. Nevertheless, future focus for therapy might not be on symptoms but on disease-triggering mechanisms. Identifying pathomechanisms that are relevant for the individual patient and treating these in a targeted manner is the concept of individualized medicine – new drugs, combined with novel diagnostic tests and imaging technologies would be able to treat the pathomechanisms that are relevant for the individual patient and thus increase the chance of therapeutic success. This therapeutic concept is in contrast to current one-fits-all therapies.

While full implementation into clinical use may not be possible within 5 years, individual elements are likely to reach application. For example, the first novel validated biomarkers may be used for conducting clinical trials, allowing for prospective patient selection, improving success rate and reducing trial costs. Since drug development costs have to be passed on through the drug price, this might also result in less expensive drug therapies. Tailored treatment may also reduce side effects with an associated further cost reduction. Nevertheless, the overall costs or savings associated with PM are hardly predictable. It may initially cost more to implement and deliver this strategy. However, the long-term savings due to resulting health benefits might result in a net reduction of healthcare burden. Importantly, today an enormous part of healthcare budgets goes towards expensive tests and treatments that produce little health gain per dollar [72]. Despite important economic aspects, health is a non-monetary value that cannot be measured in cost–effectiveness analysis. Nevertheless, the success of such a paradigm shift towards PM depends on the cooperation of industry, government bodies, clinicians and patients. In summary, PM has the potential to produce great health benefits, although possibly not within the next 5 years. Nevertheless, the proof of concept for the effectiveness of cardiovascular PM is warranted.

Key issues

• • Individualized medicine with its three pillars of in vitro diagnostics, molecular imaging technologies and mechanism-based drugs has the potential to revolutionize the way we diagnose and treat cardiovascular diseases (CVD).

• • Innovative, mechanism-based drugs that are tailored to the individual are needed to advance the therapy of CVD.

• • Reactive oxygen species are associated with CVD. However, they also have physiological functions. This might explain why untargeted supplementation with antioxidant confers harm rather than benefit.

• • In contrast to antioxidants, NADPH oxidases inhibitors prevent the formation of reactive oxygen species in the first place; however, these have not yet reached clinical development.

• • Soluble guanylate cyclase stimulators and soluble guanylate cyclase activators are novel vasodilators that are in clinical development and may improve CVD therapy within the next few years.

• • The authors’ view is that no drug is more beneficial for preventing CVD than a healthy lifestyle.

Financial & competing interests disclosure

The authors have not received payment for preparation of this manuscript. Harald Schmidt and Kirstin Wingler receive research funding from Servier and Bayer Schering. Kirstin Wingler is a former employee of Vasopharm GmbH, which develops NADPH oxidase inhibitors. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.