Clinical Science (2012) 123, 73–91 (Printed in Great Britain) doi:10.1042/CS20110562 73

REVIEW Free radical biology of the cardiovascular system

Alex F. CHEN∗, Dan-Dan CHEN∗, Andreas DAIBER†, Frank M. FARACI‡, Huige LI§, Christopher M. REMBOLD and Ismail LAHER¶ ∗Department of Surgery, McGowan Institute of Regenerative Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, U.S.A., †2nd Medical Clinic, Molecular Cardiology, Medical Center of the Johannes Gutenberg University, Mainz, Germany, ‡Departments of Internal Medicine and Pharmacology, Carver College of Medicine, University of Iowa, Iowa City, IA 52242, U.S.A., §Department of Pharmacology, University Medical Center, Johannes Gutenberg University, Mainz, Germany, Cardiovascular Division, Department of Internal Medicine, University of Virginia, Charlottesville, VA 22908, U.S.A., and ¶Department of Pharmacology and Therapeutics, Faculty of Medicine, University of British Columbia, 2176 Health Sciences Mall, Vancouver, BC, Canada V6T 1Z3

ABSTRACT

Most cardiovascular diseases (CVDs), as well as age-related cardiovascular alterations, are accompanied by increases in oxidative stress, usually due to increased generation and/or decreased metabolism of ROS (reactive oxygen species; for example superoxide radicals) and RNS (reactive nitrogen species; for example peroxynitrite). The superoxide anion is generated by several enzymatic reactions, including a variety of NADPH oxidases and uncoupled eNOS (endothelial NO synthase). To relieve the burden caused by this generation of free radicals, which also occurs as part of normal physiological processes, such as mitochondrial respiratory chain activity, mammalian systems have developed endogenous antioxidant enzymes. There is an increased usage of exogenous antioxidants such as vitamins C and E by many patients and the general public, ostensibly in an attempt to supplement intrinsic antioxidant activity. Unfortunately, the results of large-scale trails do not generate much enthusiasm for the continued use of antioxidants to mitigate free-radical-induced changes in the cardiovascular system. In the present paper, we review the clinical use of antioxidants by providing the rationale for their use and describe the outcomes of several large-scale trails that largely display negative outcomes. We also describe the emerging understanding of the detailed regulation of superoxide generation by an uncoupled eNOS and efforts to reverse eNOS uncoupling. SIRT1 (sirtuin 1), which regulates the expression and activity of multiple pro- and anti-oxidant enzymes, could be considered a candidate molecule for a ‘molecular switch’.

− Key words: cardiovascular disease, endothelial progenitor cell, free radical, oxidative stress, sirtuin, superoxide anion (O2 ). Abbreviations: ACE, angiotensin-converting enzyme; ADMA, asymmetric ω-NG,NG-dimethylarginine; AGE, advanced glycation end-product; AngII, angiotensin II; ApoE, apolipoprotein E; ARB, angiotensin receptor blocker; AT1 receptor, AngII type 1 receptor; BH4, tetrahydrobiopterin; CAD, coronary artery disease; CVD, cardiovascular disease; EPC, endothelial progenitor cell; FOXO, forkhead box O; GCH, GTP cyclohydrolase; GPx, glutathione peroxidase; GTN, nitroglycerine; KO, knockout; LDL, low- density lipoprotein; l-NAME, NG-nitro-l-arginine methyl ester; MI, myocardial infarction; mitoQ, mitoquinone; MMP, matrix metalloproteinase; mPTP, mitochondrial permeability transition pore; mtKATP channel, mitochondrial ATP-sensitive potassium − − channel; NOS, NO synthase; eNOS, endothelial NOS; O2 , superoxide anion; ONOO , peroxynitrite; PARP, poly(ADP- ribose) polymerase; PETN, pentaerithrityl tetranitrate; PKC, protein kinase C; PPAR, peroxisome- proliferator-activated receptor; RAS, renin–angiotensin system; RNS, reactive nitrogen species; RONS, reactive oxygen and nitrogen species; ROS, reactive oxygen species; siRNA, small interfering RNA; SIRT1, sirtuin 1; SOD, superoxide dismutase; Cu/Zn-SOD, copper/zinc SOD; Mn-SOD, manganese SOD; UAG, unacylated ghrelin; VEGF, vascular endothelial growth factor. Correspondence: Dr Ismail Laher (email [email protected]).

C The Authors Journal compilation C 2012 Biochemical Society 74 A. F. Chen and others

INTRODUCTION Cu/Zn-SOD (copper/zinc SOD; SOD1), Mn-SOD (manganese SOD; SOD2) and ec-SOD (extracellular There is much interest in the role of free radical oxidative SOD; SOD3). GPx proteins consume reduced gluta-

damage in human diseases. The first description of free thione to convert H2O2 into water and lipid peroxides radicals was by Gomberg in 1900 at the University to their respective alcohols, with GPx1 being the most of Michigan; the term free radicals refers to atoms, abundant selenoperoxidase and a key antioxidant enzyme molecules or ions with unpaired electrons in their in many cell types. Catalase is an intracellular antioxidant outer shells, making them highly reactive in biological enzyme that is mainly located in cellular peroxisomes and systems. This unstable configuration provides energy that to some extent in the cytosol, which catalyses the reaction

is released through reactions with adjacent molecules of H2O2 into water and molecular oxygen [9]. Non- such as proteins, lipids, carbohydrates and nucleic acids. enzymatic antioxidants include vitamins E (tocopherol) As a result, oxidative stress is implicated in several and C (ascorbic acid), thiol antioxidants (glutathione, human diseases, including hypertension, atherosclerosis thioredoxin and lipoic acid), melatonin, carotenoids, and vascular diabetic complications, as well as in the natural flavonoids and other compounds [10]. aging process [1]. The majority of free radicals that The growing interest in the possible beneficial roles damage biological systems are oxygen free radicals. of antioxidant supplements in the treatment of CVD Oxygen free radicals or, more generally, ROS (reactive (cardiovascular disease) has contributed to a debate about oxygen species), as well as RNS (reactive nitrogen their value in providing complementary therapies aimed species), are products of normal cellular metabolism at improving standard therapy. There are numerous [2]. The most common RONS (reactive oxygen and explanations for the failure to observe noticeable benefits − nitrogen species) include O2 (superoxide anions), H2O2 with the use of currently available antioxidants. It is • (hydrogen peroxide), OH (hydroxyl radicals), carbon- possible that the agents examined are ineffective and • centred peroxides and peroxyl radicals, NO (NO non-specific, and that the dosing regimens and the • − radicals), NO2 (nitrogen dioxide radicals), ONOO duration of therapy are insufficient. Vitamins C and E − (peroxynitrite), ClO (hypochlorite) and others [3–5]. may have pro-oxidant properties with harmful and − O2 can be formed from different sources such as deleterious interactions. It is also possible that orally xanthine oxidase, NADPH oxidases, uncoupled NOSs administered antioxidants may be inaccessible to the (NO synthases), enzymes of the mitochondrial respir- source of free radicals, particularly if ROS are generated in atory chain and cytochrome P450 mono-oxygenases [1]. intracellular compartments and organelles. Furthermore,

Among these enzymes, NADPH oxidases and uncoupled antioxidant vitamins do not scavenge H2O2, which may − eNOS (endothelial NOS) are crucial ROS sources under be more important than O2 in CVD [11]. In the present pathological conditions [6]. Oxidative stress defines a review, we will discuss strategies for the design of newer state with either increased (uncontrolled) formation of antioxidants. For example, it may be more promising to RONS and/or impaired cellular antioxidant defence develop antioxidants that accumulate at the site of oxidant system (e.g. down-regulation of important antioxidant formation (e.g. mitochondrial-targeted antioxidants) or proteins) with subsequent depletion of low-molecular- to inhibit the sources of oxidants (e.g. NADPH oxidase mass antioxidants and a shift in the cellular redox balance inhibitors). Another strategy would be the synthetic [7]. production of intrinsic antioxidant enzymes (e.g. SOD- Organisms have evolved a number of antioxidant mimetic antioxidants). enzymes to combat the burden of increased oxidative stress. An antioxidant refers to any molecule capable of stabilizing or deactivating free radicals before they OXIDATIVE STRESS IN VASCULAR AGING attach to cells. Humans have evolved highly complex AND DISEASE antioxidant systems (enzymatic and non-enzymatic) that work synergistically, and in combination with each Aging produces complex changes in both structure and other, to protect cells and organ systems of the body function throughout the vascular tree [12]. Mechanisms against free radical damage. Antioxidants can be either that underlie vascular homoeostasis and maintain oxygen endogenously produced substances or obtained from delivery often become impaired with aging. For example, exogenous sources, e.g. as a part of a diet or as dietary resting blood flow and most vasodilator responses decline supplements. Some dietary compounds that do not with age. Complex responses including neurovascular neutralize free radicals, but enhance endogenous activity, coupling and endothelium-dependent regulation of can also be classified as antioxidants [8]. vascular tone exhibit age-related dysfunction in both The most efficient enzymatic antioxidants involve GPx experimental animals and humans [13]. Vascular or (glutathione peroxidase), catalase and SOD (superoxide endothelial dysfunction are terms used to describe − dismutase) [9]. SOD catalyses the dismutation of O2 abnormalities that collectively increase vascular tone, into H2O2. There are three isoforms of mammalian SOD: promote thrombosis and alter vascular growth.

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 75

Role of oxidative stress In addition to having direct effects on vascular tone, ROS impair vasodilatory responses mediated by other − mechanisms [14]. For example, O2 reacts extremely efficiently with NO (the major endothelium-derived relaxing factor) [15], decreasing the bioavailability of NO. The rate of reaction of superoxide with SOD is 2 × 109 M − 1 · s − 1, while the rate of reaction between NO and superoxide is (6–10) × 109 M − 1 · s − 1, meaning that, when superoxide levels are in excess, NO will preferentially react with superoxide. Since NO plays a major role in vascular biology, loss of normal NO-mediated signalling has broad implications with both short- and long-term consequences [14,16,17]. In addition to impairing NO-mediated signalling, the − interaction of NO with O2 results in formation of ONOO − . Increases in ONOO − occur with age in blood Figure 1 Selected mechanisms thought to promote and vessels [7] and have several effects including activation protect against vascular disease during aging of nuclear PARPs [poly(ADP-ribose) polymerases] [18]. Both oxidative stress and inflammation play a major role in producing changes Excessive activation of PARP depletes energy stores and that occur within vascular cells which ultimately lead to increased vascular tone, thus produces cellular dysfunction [18]. Several studies the onset and progression of atherosclerosis, reduced blood flow and increased κ κ suggest that activation of PARP contributes to vascular risk for cardiovascular events. Nox, NADPH oxidase; NF- B, nuclear factor B; dysfunction with aging [19]. Acute effects of ROS include AT1-R, AT1 receptor. reductions in resting blood flow and impairment of vasodilator responses. Many chronic effects of ROS also arise, including effects on the ECM (extracellular matrix) Activation of these receptors is functionally important and vascular structure via activation of metalloproteina- because oxidative stress and endothelial dysfunction does ses and other mechanisms. not occur in old animals lacking AT receptors compared Other studies have focused on the role of NADPH 1 to age-matched controls [19]. Such findings provide direct oxidase in promoting vascular disease [20]. During aging, − evidence for a fundamental role for AngII and AT1 increased O2 occurs in vascular cells and is functionally receptors in age-related vascular dysfunction [19]. In this important, because impaired vascular responses with regard, it is noteworthy that AngII is a major stimulus for aging can be restored to normal with scavengers of − activation of NAPDH oxidase [21], implicated previously O2 , inhibitors of NADPH oxidase or following genetic in promoting oxidative stress during aging [7]. deletion of the catalytic subunit of NADPH oxidase [20]. − Thus a key source of O2 in blood vessels during aging is NADPH oxidase (Figure 1). Antioxidants and vascular aging Studies utilizing mice genetically deficient in individual RAS (renin–angiotensin system) forms of SOD have implicated key roles for these The RAS plays important roles in vascular biology and is enzymes in protecting the vasculature during aging. − implicated in a variety of pathophysiological conditions For example, increases in vascular O2 with age including hypertension [14]. Most of the detrimental are augmented in mice partially deficient in SOD1 effects of this system are mediated by AngII (angiotensin [heterozygous KO (knockout) mice] [24]. Endothelial

II) acting on AT1 receptors (AngII type 1 receptors). function is markedly impaired in old SOD1-deficient One of the major effects of AngII within vascular cells mice at an age when these responses are still normal in is AT1 receptor-dependent activation of pathways that wild-type mice [24]. Thus oxidative stress and vascular promote oxidative stress. In addition, these receptors dysfunction with aging is accelerated in the face of SOD1 activate components of the inflammatory response that deficiency. Similarly, SOD2 haploinsufficiency increases contribute to vascular disease as well [14] (Figure 1). oxidative stress and augments endothelial dysfunction Studies have focused on the impact of AngII in vascular with aging [25]. aging [21]. Expression of the enzyme that produces Klotho is a gene known to inhibit age-related AngII [ACE (angiotensin-converting enzyme)], tissue symptoms and extend life span [26]. In relation levels of AngII and AT1 receptors all increase with age to vascular disease, klotho suppresses pro-oxidant − in blood vessels [22]. Chronic inhibition of the RAS mechanisms including AngII-induced O2 formation in − or genetic deletion of AT1 receptors suppress ONOO endothelial cells [27]. In a genetic model of spontaneous formation and inhibit vascular growth during aging [23]. hypertension, overexpression of klotho reduces the

C The Authors Journal compilation C 2012 Biochemical Society 76 A. F. Chen and others

− expression and activity of NADPH oxidase and O2 THERAPEUTIC USE OF ANTIOXIDANTS IN levels in the vasculature [28]. Although potentially CVD important, the functional importance of klotho within the vessel wall during normal aging has not been Evidence for a beneficial effect of antioxidants such defined. as vitamin C on cardiovascular events is based Possible therapies for vascular aging have emerged on observations that patients with a pronounced recently [13]. For example, voluntary exercise decreases improvement of endothelial function (measured by oxidative stress and improves endothelial function in old acetylcholine-induced forearm dilation) in response to mice [29]. These beneficial effects appear to result from vitamin C infusion were at higher risk for cardiovascular a combination of increased SOD activity and reduced events as compared with those patients displaying only expression of NADPH oxidase [29]. Short-term caloric minor improvement of endothelial function in response restriction also reduces oxidative stress and enhances to vitamin C infusion [37]. The main conclusion of vascular function during aging [30]. Such findings are this study was that patients with increased vascular important because they provide evidence that vascular oxidative stress (making them more responsive to dysfunction with age is not entirely fixed and may vitamin C) displayed a higher incidence of cardiovascular be reversed or lessened with appropriate treatment or events. In addition, molecular proof for the involvement intervention. of oxidative stress in cardiovascular complications is based on animal experimental models with genetic deletion of NADPH oxidase subunits or of antioxidant enzymes. Related to this, only two examples are PPAR (peroxisome-proliferator-activated mentioned. In mice with MI (myocardial infarction), receptor) γ the genetic deletion of the NADPH oxidase subunit PPARγ is a ligand-activated transcription factor active p47phox almost normalized vascular NO• bioavailability, in vascular cells [31], where it regulates expression of reduced ROS formation and improved heart function, multiple target genes including genes that promote or as well as the survival rate by 20% after MI [38]. In suppress increases in ROS. Antioxidant effects of PPARγ contrast, deletion of mitochondrial SOD increased age- occur in blood vessels. For example, genetic interference dependent mitochondrial oxidative stress and endothelial to decrease the expression of PPARγ in adult mice dysfunction [39]. Some animal experimental models − increases vascular O2 and impairs vascular function, as for CVDs that are associated with oxidative stress are well as producing hypertrophy and inward remodelling shown in Figure 2. Given the importance of oxidative in the microcirculation [32]. Activators of PPARγ stress in almost all cardiovascular, inflammatory and prevent vascular remodelling during hypertension, neurodegenerative diseases, pharmacological control of

decrease expression of AT1 receptors and NADPH the sources of oxidative stress may be of great clinical oxidase, while increasing the expression of SOD1 [31] interest. However, considering the fact that some RONS (Figure 1). have important functions in immune defence and fulfil There is an emerging appreciation of a role for functions in redox signalling, this therapeutic strategy PPARγ in aging, with the findings that the expression requires cautious fine tuning and spatial selectivity. and activity of PPARγ decreases with age [33] and γ that decreases in PPAR expression reduce lifespan Clinical trials of antioxidants in CVD in mice [34]. Expression of klotho increases following In patients with or at higher risk of CVD, there is pharmacological activation of PPARγ [35] and thus may both basic science and clinical trial evidence for an contribute to the protective effects of PPARγ [34]. In improvement in atherosclerosis with statins [40], niacin heterozygous KO mice expressing a human dominant- [41], fibrates [42], aspirin, n − 3 fatty acids [43] and negative mutation in the ligand-binding domain of a Mediterranean diet [44]. Unfortunately, such dual PPARγ (which inhibits transcriptional activity of wild- evidence does not exist for antioxidants. Despite very type PPARγ ), age-related vascular dysfunction occurred promising basic science findings [45], the clinical trial earlier and to a greater extent compared with age-matched results with various antioxidants have been disappointing controls. Oxidative stress appears to underlie this and the possible reasons for the divergent results are accelerated age-induced vascular dysfunction, because − reviewed below. a scavenger of O2 prevented these changes. These preliminary findings suggest a fundamental role for PPARγ in protection against age-induced oxidative stress Clinical trial evidence and vascular dysfunction. Consistent with these results, β-Carotene (60–200 mg, daily) was studied in 138113 vascular dysfunction in an aged model of Alzheimer’s subjects in four large primary and four large secondary disease improves in response to a synthetic activator of prevention trials [46]. Six trials individually showed no PPARγ [36]. benefit and two trials individually showed increased

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 77

Figure 2 Examples of increased vascular oxidative stress in vessels from animal experimental models of cardiovascular complications, as well as human mammary arteries from bypass surgery of GTN-treated patients − − The images show vascular RONS (most probably O2 ) formation in aortic cryosections that were stained with dihydroethidine (DHE). O2 and other RONS oxidize DHE producing highly red fluorescent products. The green fluorescence represents the autofluorescence of the laminae. The right-hand panels in each column represent diseased tissue and the left-hand panels are the respective control tissue (Ctr). E, endothelial; M, media; A, adventitia; wt, wild-type; WHHL, Watanabe heritable hyperlipidaemic. Reproduced from Daiber A, Munzel¨ T. Oxidativer Stress, Redoxregulation und NO-Bioverfugbarkeit¨ - experimentelle und klinische Aspekte. Steinkopff Verlag Darmstadt, 2006; Darmstadt, Germany, with permission.

mortality. A meta-analysis showed a significant increase the combined end point of cardiovascular death, MI in total mortality (absolute risk increase 0.4%, P = 0.003). and revascularization (P < 0.05). However, in PQRST Vitamin E (300–800 units, daily) was studied in 81788 (Probucol Quantitative Regression Swedish Trial) with subjects in two large primary and five large secondary 276 subjects, probucol (500 mg) did not slow the prevention trials [46]. No trial individually showed an progression of femoral intimal medial thickness [48]. effect on mortality. A meta-analysis showed no effect on Probucol was removed from the market because it total mortality (absolute risk increase 0.2%, P < 0.42). increased the QTc interval. One secondary prevention trial [CHAOS (Cambridge Succinobucol (300 mg, daily) was studied in 6144 Heart Antioxidant Study)] showed a significant reduction subjects in the ARISE (Aggressive Reduction of in MI (absolute risk decrease 2.6%, P = 0.01); however, Inflammation Stops Events) trial [49]. There was no effect meta-analysis of all of the trials revealed no effect on MI on the primary end point of cardiovascular death, cardiac (absolute risk unchanged, P = 0.93). arrest, MI, stroke, angina or revascularization (P = 0.96). Probucol (500 mg, daily) was studied in 246 subjects In this trial, there were divergent effects on the end in FAST (Fukuoka Atherosclerosis Trial) [47]. Probucol points: succinobucol significantly reduced cardiovascular reduced carotid initimal medial thickness and decreased death, cardiac arrest, MI and stroke (P = 0.03), while

C The Authors Journal compilation C 2012 Biochemical Society 78 A. F. Chen and others

significantly increasing angina and revascularization. It of coffee intake on CVD [54] and may indeed have a is important to note that succinobucol was not marketed. protective effect on dementia [55]. A possible scenario A combination of vitamin E (600 mg), vitamin C (250 mg) is that administration of a mild pro-oxidant could induce and β-carotene (20 mg) in the Heart Protection Trial endogenous antioxidant production, which is cardio- and (20536 subjects) had no significant effect on mortality, vasculo-protective. MI or stroke [50]. These findings suggest that we do not have data The Mediterranean diet used in the Lyon trial supporting oral administration of single or combination [44] showed a very large 5.2% first year and large antioxidant supplements at present. However, they do 1.6% subsequent year absolute risk reduction in MI not rule out a benefit from antioxidants in food given the and cardiovascular death [51]. This occurred despite large benefits observed with a Mediterranean diet, which a relatively small reduction in LDL (low-density includes both food-based antioxidants, high ω − 3and liporotein)-cholesterol from 176 to 162 mg/dl. Some of ω − 9 fatty acids, and low ω − 6 and saturated fats. this reduction in MI and cardiovascular death may have − resulted from the change in diet from saturated and n 6 Antioxidant treatment: oral compared with infusion, and fats to n − 3andn − 9 fats [44]. However, it is also possible that antioxidants present in the Mediterranean chronic compared with acute As noted above, large clinical trials with antioxidants such diet also had some effect on MI and cardiovascular death. as vitamins C and E did not show any benefit for oral Further clinical trials on the effect of antioxidants in the antioxidant treatment on the incidence and prognosis Mediterranean diet are indicated. of CVDs. In contrast, the first controlled studies such Some potential explanations for negative clinical trials as HOPE (Heart Outcome Prevention Evaluation) and HOPE-TOO (Heart Outcome Prevention Evaluation – It is possible that the antioxidant doses used in these The Ongoing Outcomes), as well as a prospective clinical trials may have been inadequate. Roberts et al. study with vitamin C in post-menopausal women [52] found that a daily dose of 1600 or 3200 mg of with diabetes mellitus, reported an increased incidence vitamin E (RRR-α-tocopherol) in humans was required of cardiomyopathy, left heart decompensation and to induce significant reductions in plasma F2 isoprostanes. cardiovascular mortality in the group with antioxidant This suggests that the doses of vitamin E (300–800 mg) treatment [56]. These large-scale trials with oral treatment used in the clinical trials described above may have been are in contrast with a great number of animal experimental insufficient. In addition, the following are also possible. studies [57], as well as human studies with rather small (i) Some antioxidant preparations are more protective numbers of patients and acute infusion of antioxidants than others, for example α- compared with γ -tocopherol. [58], indicating that antioxidants may be highly beneficial (ii) Antioxidant therapy may be beneficial in the early in improving endothelial dysfunction. The main reasons phases of atherosclerosis, although arguing against this for the failure of chronic oral antioxidant therapy could hypothesis is the lack of benefit in the primary prevention be as follows: vitamins C and E act as pro-oxidants, the trials described above. (iii) Endogenous antioxidants CAD (coronary artery disease) of the patients included is are more important than exogenous antioxidants. For already irreversible, the CAD patients are already being example, Collins et al. [53] found that LDLR (LDL − − treated with drugs displaying antioxidant properties receptor) / mice fed on a high-fat diet had more aortic [for example ACE inhibitors and ARBs (angiotensin atherosclerosis, more oxidative stress (higher plasma F2a − receptor blockers)], chronic antioxidant therapy inhibits isoprostane and higher vascular O2 ) and lower levels of intrinsic ischaemic preconditioning, which relies on endogenous antioxidants [lower SOD, catalase, GPx-1, RONS formation, or oral vitamin treatment does not GPx-4, NADPH dehydrogenase quinone 1 and upstream result in high enough concentrations of the antioxidants at mediators of antioxidants including FOXO (forkhead the place of oxidative stress. Some of these reasons would box O) 1 and FOXO4]. It is possible that the increase in favour the acute infusion of vitamin C in accordance with atherosclerosis observed in these mice could have resulted respective observations. from decreased levels of endogenous antioxidants. (iv) The route of administration of antioxidants is important. Oral administration could potentially allow oxidation Direct RONS scavengers by gastrointestinal bacteria, a process that could limit Vitamin C reacts with free radical species (for example antioxidant benefits. The effects of oral oxidants can be OH•) in two subsequent one-electron-oxidation steps, quite complex. For example, coffee drinking in humans yielding ascorbyl radicals as an intermediate that

increased urinary oxidant H2O2 3–10-fold; incubation either undergo disproportionation to ascorbic acid and of coffee in a neutral medium also generates 120– dehydroascorbate or are reduced by glutathione. The

420 μMH2O2 [54]. Theoretically, these findings would two-electron oxidation product dehydroascorbate is suggest that coffee drinking should be pro-oxidant and reduced (restored) by dehydroascorbate reductase. It is therefore harmful; however, meta-analyses show no effect important to note that these classical antioxidants would

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 79 be depleted without an associated reduction system. to transplanted organs. Unfortunately, ebselen displayed Other important examples of this class are vitamin E, severe liver toxicity, which could have been related to thiol-based compounds (for example glutathione and its high reactivity with all thiols (for example ebselen cysteine), thioethers (for example methionine), bilirubin is a potent inhibitor of alcohol dehydrogenase and and many more. In the presence of OH•, almost any other thiol-dependent enzymes) [62]. Another group biomolecule may act as an antioxidant. The other class of synthetic antioxidants comprise the soluble metal of classical antioxidants acts catalytically and comprises pophyrins {FeTMPyP [iron (III) tetrakis(N-methyl-4- enzymes with a metal centre or selenocysteine: SODs pyridyl)porphyrin], FeTMPS [iron(III) meso-tetra(2,4,6- − that catalyse the breakdown of O2 , catalases to detoxify trimethyl-3,5-disulfonato)porphyrin chloride] and MnT- − H2O2 and GPxs to scavenge H2O2, ONOO and BAP [Mn(III)tetrakis(4-benzoic acid)porphyrin]} acting lipid peroxides. These considerations already indicate the as SOD mimetics [63], whose development and pre- requirements for a synthetic antioxidant: if it is consumed clinical tests were spearheaded by the pharmaceutical during the antioxidant reaction, it should use a cellular company Monsanto [64]. Although these compounds reduction system in order to be restored. A catalytically are clearly catalytic and display high reaction rates with − − acting synthetic antioxidant seems even more promising. O2 and ONOO , they never reached clinical utility, Another question is which kind of oxidant should be most probably due to their toxicity when used at higher scavenged by the synthetic antioxidant. Since ascorbic concentrations or possibly because they can also be pro- − acid and tocopherol are poor scavengers of O2 ,the oxidant under some conditions (for example by acting as − structure of a synthetic O2 scavenger should be a catalyst of phenol nitration) [59]. different. The reactivity of a RONS may change upon − protonation: O2 are much less reactive as compared Inhibitors of RONS sources • with HOO (hydroperoxyl radicals), which show a It may be more promising to inhibit the formation considerable reaction rate with ascorbic acid, tocopherol of RONS instead of scavenging these harmful species − and unsaturated fatty acids. ONOO is rather after their emergence. Controlling mitochondrial RONS unreactive, whereas its conjugated acid (ONOOH) or formation will be addressed later in the present review. − its CO2 adduct (ONOOCO2 ) or, more precisely, the Inhibition of xanthine oxidase is a well-recognized free radicals derived during their decomposition display strategy to improve cardiovascular complications in high reactivity towards a broad range of biomolecules. several CADs, as supported by animal experimental Accordingly, each of these species requires specially studies using allopurinol or tungsten treatment [65], designed antioxidants. For example, to inhibit the as well as clinical studies in patients with Type 2 nitration of tyrosine by ONOOH, ascorbic acid or diabetes and idiopathic dilated cardiomyopathy [66]. uric acid is more efficient compared with glutathione or Most published data on the beneficial effects of xanthine methionine, since they specifically react with ONOOH- oxidase inhibition are related to ischaemia/reperfusion derived free radicals (up to 30% based on the ONOOH damage after MI [67]. concentration) which mediate nitration. Glutathione Another therapeutic target of interest is the NADPH − and methionine react with the ONOO until they oxidases, since their involvement in most CVDs has been − are consumed and the remaining ONOO undergoes demonstrated sufficiently in the literature [68]. These protonation with subsequent radical-mediated nitration multi-protein complexes are of special pharmacological of tyrosine. This results in low IC50 values for ascorbic interest, since there are many regulatory pathways and acid and uric acid and high IC50 values for glutathione accordingly many sites for pharmacological intervention. − and methionine. In contrast, prevention of ONOO - The following inhibitory mechanisms apply for the mediated oxidation of the zinc–sulfur active site of phagocytic NADPH oxidase isoform (Nox2, gp91phox): alcohol dehydrogenase is conferred by low IC50 values apocynin or more precisely its peroxidase-activated for glutathione and methionine, whereas ascorbic acid dimer inhibits the assembly of NADPH oxidase [69], and uric acid require high concentrations to show any the PKC (protein kinase C) inhibitor chelerythrine effect, and vitamin C even had some pro-oxidative effects blocks phosphorylation and translocation of the cytosolic in this system. These considerations have been reviewed p47phox subunit, gliotoxin and PAO (phenyl arsenoxide) elsewhere [59,60]. target essential thiol groups at the site of assembly Some promising synthetic antioxidants used previ- of cytosolic subunits and thereby suppress activation, ously were mimetics of antioxidant enzymes. Ebselen [2- the third-generation β-blocker nebivolol interferes phenyl-1,2-benzoisoselenazol-3(2H)-one] was designed with the assembly but even may dislocate already as- as a GPx mimetic to scavenge H2O2 and lipid peroxides sembled cytosolic subunits, and DPI (diphenyliodonium) [61]. This compound was among the fastest reacting syn- inhibits the electron flow in NADPH oxidases (as it − thetic ONOO scavengers. The therapeutic indication does in all flavin-dependent oxidoreductases) and inhibits − of ebselen was in the field of transplantation surgery and O2 formation. All of these NADPH oxidase inhibitors was intended to prevent ischaemia/reperfusion damage have been demonstrated to be highly efficient in animal

C The Authors Journal compilation C 2012 Biochemical Society 80 A. F. Chen and others

models of CVD, but to date have not been not employed Cu/Zn-SOD survived [78]. Previous and recent findings in patients due to their toxicity, with the exception of have suggested that mitochondrial RONS are involved the β-blocker nebivolol, which has yet to be tested for in many cardiovascular complications (for example in its effects on NADPH oxidase in patients. There are the setting of diabetes, hypertension and MI) and may more than 20 registered patents for NADPH oxidase even act as amplifiers of cytosolic oxidative stress, inhibition in the U.S. Patent Collection database, and and thus inducing vicious cycles of RONS-mediated there is a great effort to develop (isoform-) specific effects [79]. The hypothesis behind this amplification NADPH oxidase inhibitors [70]. As studies suggest mechanism is that mitochondrial RONS engages in − the important roles for O2 and H2O2 in cellular cross-talk with cytosolic RONS sources (and vice versa) signalling, proliferation, migration and differentiation by mechanisms involving mitochondrial channels, such [71], all therapeutic strategies to inhibit NADPH oxidases as the mPTP (mitochondrial permeability transition

(Nox1, Nox2 and Nox4) require cautious and tedious pore) and mtKATP channels (mitochondrial ATP-sensitive evaluation in cell culture and animal models. potassium channels). Therefore one strategy to control Another source of RONS formation may be an mitochondrial RONS formation could be ‘fixing a hole’, uncoupled NOS. Under certain conditions, for example as suggested by Di Lisa and Bernardi [80]. There are great

depletion of the cofactor BH4 (tetrahydrobiopterin), opportunities for the pharmacological manipulation of oxidation of the zinc–sulfur complex at the dimer- these mitochondrial channels, which have so far been binding site, excess of ADMA (asymmetric ω-NG,NG- somewhat disregarded by the pharmaceutical industry. dimethylarginine), S-glutathionylation of eNOS and A study by Piot et al. [81] has demonstrated a adverse (PKC-dependent) phosphorylation at Thr495,the significant reduction in MI-induced cardiac damage in − function of NOS can transform from an NO- to a O2 - patients receiving a post-infarct infusion of the mPTP producing enzyme [72]. There is ample evidence for a blocker cyclosporin A [81]. Numerous studies indicate − contribution of eNOS-dependent O2 formation and a a beneficial effect of mPTP blockade to reduce oxidative loss of NO synthesis due to cardiovascular complications damage in response to ischaemia/reperfusion, and that

[72]. One therapeutic option for eNOS ‘recoupling’ is inhibition of the mtKATP channel may thus represent supplementation with BH4 or its precursors sepiapterin an attractive therapeutic target, since drugs such as and folic acid, which are cheaper and more stable. glibenclamide or 5-hydroxydecanoate have beneficial

Supplementation with BH4 has proven highly beneficial effects in many CAD-related models. However, errant on endothelial function in many studies, such as manipulation of this susceptible system may have improvement in diabetic patients or chronic smokers disastrous outcomes, since these mitochondrial channels [73]. Overexpression or activity enhancement of GCH are not only important gates for cellular compounds, but (GTP cyclohydrolase) represents another pathway to are also intimately involved in ischaemic preconditioning,

increase BH4 levels as demonstrated in diabetic mice a process that is believed to have essential implications for [74]. Under certain conditions, the levels of ADMA are priming the cell for future (oxidative) stress events and increased in endothelial cells, which may contribute to the increasing their adaptation to altered redox conditions. uncoupling of eNOS, and supplementation with a high Another option to suppress mitochondrial RONS dose of l-arginine may be of great benefit by replacing formation is to target antioxidants to the mitochondria. ADMA or even support its export from endothelial cells There are numerous examples of this strategy, but by driving the cationic amino acid transporters. This one of the first successes was with using mitoQ may be of clinical importance since ADMA is a valid (mitoquinone) [82], a quinone that is coupled to predictor of cardiovascular events and is associated with a triphenylphosphonium group to facilitate mitoQ increased cardiovascular mortality [75]. Besides these accumulation in mitochondria by up to 10000-fold. directly eNOS-targeting therapeutic interventions, any GTN (nitroglycerine)-induced vascular dysfunction was

cardiovascular therapy that reduces oxidative stress will improved by inhibition of mtKATP channels or the contribute to ‘recoupling’ of eNOS, as demonstrated for mPTP [83], and MnSOD deletion aggravated the adverse ACE inhibition, ARB therapy, treatment with statins effects of GTN on vascular function and oxidative or antihyperglycaemic drugs and, to a lesser extent, stress [84], whereas another study reports that GTN treatment with calcium antagonists or β-blockers [76]. induced oxidative stress and nitrate tolerance was normalized by mitoQ [85]. Previous findings have suggested that mitoTEMPO is another highly effective Mitochondria-targeted antioxidants or drug able to suppress mitochondrial RONS formation ‘fixingahole’ and stop cellular vicious cycles of RONS generation. Mitochondria are important sources of oxidative stress, MitoTEMPO was able to completely normalize AngII- − especially O2 formation. This was documented by the induced endothelial dysfunction and increase in blood lethality of MnSOD deletion, the mitochondrial SOD pressure in mice [86]. Overexpression of MnSOD showed [77], whereas mice with a deficiency in extracellular similar protective effects under these conditions.

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 81

Endothelium-targeted antioxidants GTN-induced nitrate tolerance [92] and may thus also Administration of extracellular SOD, which directly beneficially influence isosorbide-5-mononitrate-induced binds to the endothelium, is highly protective in many endothelial dysfunction [93]. CVD states [87]. Likewise, deletion of extracellular Another interesting example of a cardiovascular drug SOD impairs vascular function and haemodynamics with pleiotropic antioxidant effects is PETN. After it [87]. Therefore the targeting of extracellular SOD to was developed in the U.S.A., it was abandoned, but the endothelium is a promising strategy to prevent then used for many years in the former East Germany. endothelial (vascular) dysfunction. Shuvaev et al. [88] After the reunion of Germany, PETN use seized the have demonstrated that targeting of SOD, but not nitrate market and is now the best selling nitrate in catalase, to the endothelium reversed AngII-induced Germany. PETN is the only organic nitrate in clinical endothelial dysfunction. The antioxidant enzymes SOD use that is devoid of nitrate tolerance and endothelial and catalase were conjugated with an antibody against dysfunction [94]. On the basis of earlier studies, PETN PECAM-1 (platelet/endothelial cell adhesion molecule- is a potent inducer of the intrinsic antioxidant HO-1 1) to ensure endothelial binding. The results of that study (haem oxygenase-1) system [95] and extracellular SOD − demonstrate that O2 is a more harmful species in the [96]. Even more surprisingly, a gene array study has vascular system than H2O2. revealed that PETN, in contrast with GTN, induces a The coupling of heparin to 4-amino-TEMPO to create number of cardioprotective genes, whereas GTN induces a spin-labelled amide of heparin has been used to study cardiotoxic genes [97]. Although both compounds are the heparin–anti-thrombin complex [89]. This strategy organic nitrates and regarded as NO donors, they have appears to be highly attractive, since the endothelium distinctly different gene regulation profiles, indicating has heparin-binding sites. In addition, most synthetic or that organic nitrates are not a homogenous class of drugs biological low- or high-molecular-mass molecules can be [98]. Recent findings suggest that PETN also improves conjugated to heparin. Moreover, heparin is easier to use vascular complications in an animal model of Type 1 compared with, for example, an antibody. Despite these diabetes mellitus [98a]. advantages, to the best of our knowledge, there are no industry-based targeted approaches using this rationale, with only modest interest by academic researchers to pursue this hypothesis and create new heparin-bound ANTIOXIDANTS AND EPCs (ENDOTHELIAL antioxidants. PROGENITOR CELLS)

EPCs have been described as a circulating CD34- Cardiovascular drugs with pleiotropic positive cell population derived from bone marrow antioxidant effects and homing to sites of ischaemia where they exert Many cardiovascular drugs (for example ACE inhibitors, their effects on endothelial repair and angiogenesis by ARBs and statins) have pleiotropic antioxidant proper- various means: (i) EPCs directly integrate into blood ties; as such they cause decreases in NADPH oxidase vessels and physically participate in endothelial repair activity and a ‘recoupling’ of eNOS. Hydralazine and [99], (ii) EPCs deliver angiogenic growth factors to PETN (pentaerithrityl tetranitrate) are two examples of ischaemic tissues and contribute to angiogenesis via drugs that are regarded as rather antiquarian, but are paracrine effects [100], and (iii) EPCs also release MMPs now recognized to have additional properties that can be (matrix metalloproteinases) to promote a concomitant exploited for therapeutic benefits. Hydralazine was one increase in matrix degradation that enables endothelial of the first antihypertensive drugs used, but is now mainly cell migration and vascular remodelling [101] (Figure 3). used for treating pre-eclampsia. However, it experienced The past 12 years has provided growing evidence in a ‘revival’ when the company NitroMed introduced support of EPC-based cell therapy as a promising strategy their combination drug BiDil containing hydralazine and to restore deficient endothelial function and angiogenesis isosorbide dinitrate. This combination therapy showed in CVDs [102]. However, several clinical trials have an impressive decrease in mortality in Afro-Americans revealed that reduced EPC number is associated with with severe heart failure and who responded poorly to diabetes mellitus [103], hypertension [104], aging [105] ACE inhibitors and other standard medication (the study and hypercholesterolaemia [103], in which EPCs are design was based on data from V-HeFT (Vasodilator- continuously exposed to ROS. EPCs from diabetic, Heart Failure Trial) and A-HeFT (African-American hypertensive and aged patients (and also from animal Heart Failure Trial) [90]. We can only speculate as to why models) exhibit impaired proliferation, adhesion and the combination of hydralazine and isosorbide dinitrate incorporation into vascular structures [106]. In addition, has beneficial effects, but previous findings indicate EPC senescence is positively correlated with the severity that hydralazine is a highly efficient RONS scavenger of hypertension and reduced telomerase activity [107]. [91], which in part contributes to the improvement in The alterations of EPC biology in diseased conditions

C The Authors Journal compilation C 2012 Biochemical Society 82 A. F. Chen and others

Figure 3 Schematic diagram of EPC mobilization and participation in angiogenesis EPCs are derived from bone marrow, spleen, adipose tissue and liver. Mobilized EPCs circulate in peripheral blood as specific heterogeneous population co-expressing stem cell markers (CD133 and CD34) and endothelial cell markers [CD31, vWF (won willebrand factor) and VE-cadherin]. EPCs can home to ischaemic sites, where they participate in angiogenesis in several patterns. (a) EPCs directly integrate into blood vessels and physically anticipate endothelium repair. (b) EPCs deliver angiogenic growth factors to ischaemic tissues and contribute to angiogenesis by paracrine effects. EPCs secrete the angiogenic growth factors VEGF, IGF (insulin-like growth factor), FGF (fibroblast growth factor) and GM-CSFs (granulocyte/macrophage colony-stimulating factors). The insoluble factors recruit circulating endothelial cells, fibroblasts and macrophages to ischaemic sites promoting angiogenesis and tissue repair in injured sites. (c) EPCs also release MMPs, such as MMP-2 and MMP-9, to promote a concomitant increase in matrix degradation that enables endothelial cell migration and vascular remodelling. EC, endothelial cell.

raise significant concerns about its clinical applications. Normal EPCs, compared with mature endothelial Reduced EPC pro-angiogenic ability associated with cells, are more resistant to oxidative stress due to their cardiovascular risks, including diabetes, hypertension and intrinsically high expression of intracellular antioxidant age, limits their therapeutic applications in these patients. enzymes (such as catalase, Mn-SOD and GPx-1), Furthermore, the relatively low level of circulating which facilitate their active participation in repair at EPCs and their finite proliferative potential limits their ischaemic sites [108]. Thus inhibition of these enzymes expansion to sufficient numbers for some therapeutic results in abundant ROS accumulation in EPCs, and applications. Thus understanding the pathogenic role of leads to EPC apoptosis and impaired migration [109]. oxidative stress in EPC dysfunction will identify potential However, there are several CVD-related risk factors, targets to improve the efficiency of EPC therapy in such as hyperglycaemia, hyperlipidaemia, AngII and CVDs. AGEs (advanced glycation end-products), which can also

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 83 significantly augment intracellular ROS accumulation stimulate endogenous EPC mobilization and improve and thus lead to poor hindlimb blood reperfusion and their function. Physical exercise is an important decreased Matrigel plug tube formation [110]. Com- non-pharmacological intervention to increase EPC pelling evidence indicates that EPC dysfunction may mobilization [118], and PPARγ agonists (for example be one possible reason for impaired angiogenesis and piaglitazone) increase the number and migration of subsequent vasculopathies in diabetes, hypertension and EPCs in patients with CAD and abnormal glucose aged subjects [111]. It has been reported that EPCs exhibit tolerance by reducing NADPH oxidase activity [119]. increased apoptosis and diminished tube-forming ability Systemic administration of UAG (unacylated ghrelin) ex vivo and in vivo in response to oxidative stress, which prevented diabetes-induced EPC damage (by modulating was directly linked to the activation of a redox-dependent the NADPH oxidase regulatory protein Rac1) and also Ask1 (apoptosis signal-regulating kinase 1) [111]. This improved the angiogenic potential in patients with Type 2 leads to the speculation that ROS accumulation exceeds diabetes and in diabetic mice [120]. In addition, UAG the intrinsic antioxidant capacity of EPCs and/or that the facilitates the recovery of bone marrow EPC mobilization antioxidant defences of EPCs are damaged under these by rescuing eNOS activity [120]. Some studies have diseased conditions, consequently retarding angiogenesis. used factors such as SDF-1 (stromal-cell-derived factor- NADPH oxidase is a major ROS generator in vascular 1), VEGF (vascular endothelial growth factor) and systems and its activation plays an important role in SHH (Sonic hedgehog) to recruit EPCs from the bone EPC senescence through elevating oxidative stress and marrow to the peripheral blood, thereby leading to the inactivating telomerase in AngII-treated EPCs [107]. formation of new vessels at ischaemic sites [102]. Other eNOS uncoupling is another well-known pathogenic potential strategies involve the ex vivo manipulation process impairing EPC mobilization, which is associated of impaired EPCs before autologous cell therapy. For with increased intracellular ROS generation [112]. It has example, restoration of Mn-SOD activity in diabetic been demonstrated that eNOS critically regulates EPC EPCs by ex vivo Mn-SOD gene transfer successfully mobilization and function, but is uncoupled in salt- improves their therapeutic efficiency in diabetic wound sensitive hypertension because of the reduced availability healing [121]. Constitutive hTERT (human telomerase of the cofactor BH4 [113]. Overexpression of GCH1, the reverse transcriptase) transduction protects EPCs from rate-limiting enzyme of BH4 de novo synthesis, protects ROS-induced senescence and improves their tube- EPC function in salt-sensitive hypertensive rats, partially forming ability [122]. Pre-treatment of EPCs with an via suppressing TSP-1 (thrombospondin-1; a potent eNOS enhancer, which significantly up-regulates eNOS angiogenesis inhibitor) secretion and oxidative stress expression in EPCs, enhances the therapeutic effects of [113]. In addition, accumulating evidence shows that cell therapy [123]. Of note is a recent study reporting that levels of antioxidant enzymes are diminished in vascular suppressing SIRT1 (sirtuin 1) rescued mir-34a-induced diseases. For example, Mn-SOD levels are significantly EPC senescence [124], indicating that microRNAs may decreased in EPCs in both Type 1 and Type 2 diabetes, present a novel therapeutic target for enhancing the which can be restored by activation of AMPK (AMP- regenerative capacity of EPCs. Thus strategies protecting activated protein kinase) [114]. Other studies report that EPCs from excessive oxidative stress would ensure that p66ShcA, a subunit of the adaptor protein ShcA, modulates they maintained their highly proliferative and angiogenic oxidative stress and survival of EPCs in response to potential in ischaemic sites. high glucose through inhibiting Mn-SOD expression There is emerging a clear appreciation that oxidative [115]. Levels of the normally abundant antioxidant stress impairs EPC function and mobilization in a enzyme GPx-1 are decreased in aged EPCs [116], whereas number of vascular diseases. The evidence from basic and CRP (C-reactive protein; a marker of CAD) promotes clinical research provides the basis for the development EPC sensitivity towards oxidant-mediated apoptosis of novel targeting to regulate the redox status of and telomerase inactivation through diminished GPx- EPCs and improve their utilization in treating vascular 1 activity [117]. Taken together, these findings suggest diseases. Currently, findings on the utility of EPCs are that enhanced ROS generation and impaired antioxidant derived from animal models or small-scale clinical trials. capacity contributes to excessive ROS accumulation in Significant advances will be derived from multi-centre EPCs, damaging endogenous EPC function and resulting clinical trials aimed at further evaluating EPC therapy in in poor endothelial repair and angiogenesis (Figure 4). revascularization in various vascular diseases. Interventions that inhibit excessive ROS generation or restore antioxidant capacity in EPCs could potentially improve the outcomes of EPC therapy in clinical REDUCING CARDIOVASCULAR OXIDATIVE applications. Several strategies have been developed to STRESS BY TARGETING SIRT1 improve the homing, survival and therapeutic potential of EPCs and these are aimed at counteracting EPC Reducing ROS formation (by down-regulating ROS- dysfunction. Some strategies attempt to systemically producing enzymes) and enhancing ROS inactivation

C The Authors Journal compilation C 2012 Biochemical Society 84 A. F. Chen and others

Figure 4 Schematic representation of the effects of CVD risk factors on EPCs and potential intervention targets Impaired EPC function and poor angiogenesis are associated with CVD risk factors, including diabetes, hypertension and aging. Many CVD-related risk factors, such as hyperglycaemia, hyperlipidaemia, AngII and AGEs, significantly augmented intracellular ROS accumulation via multiple mechanisms: activation of NADPH oxidase, eNOS uncoupling, inactivation of antioxidant enzymes and telomerase, which consequently lead to the developments of poor angiogenesis. Strategies to inhibit excessive ROS generation or restore antioxidant capacity would be prospective ways to overcome the EPC defects in disease circumstances and improve the outcomes of EPC therapy in clinical applications. The strategies include the following. (i) Physical exercise and pharmacological intervention with primary disease treatment to systemically

stimulate endogenous EPC mobilization and improve their function. A PPARγ agonist or a supplement with exogenous BH4 blunts ROS accumulation and restore EPC dysfunction. (ii) Ex vivo gene manipulation of impaired EPCs before autologous cell therapy. Constitutive hTERT (human telomerase reverse transcriptase) transduction, microRNA-regulated post-transcription or overexpression of antioxidant enzymes, such as Mn-SOD and GPx-1, are options to increase EPC antioxidant defence, which would protect cells from oxidative stress damages and maintain their highly proliferative and angiogenic potentials in the ischaemic sites.

(by up-regulating the endogenous antioxidant systems) SIRT1 protein levels in several tissues are increased should be more effective than using non-specific following calorie restriction [125]. antioxidant scavengers to ameliorate oxidative stress. The SIRT1 has also emerged as a promising new therapeutic optimal approach would be to modulate the expression target to treat Type 2 diabetes [125]. SIRT1 activators or activity of multiple redox genes by targeting a single improve insulin sensitivity, increase mitochondrial molecule, analogous to a ‘master switch’. It is not yet content and prolong the survival of mice fed a high-fat known whether such a ‘master switch’ exists, but there is high-calorie diet. A potent and selective SIRT1 activator, intriguing evidence that the class III histone deacetylase SRT1720, improves glucose homoeostasis, increases SIRT1 modulates several endogenous pro- and anti- insulin sensitivity in the liver, skeletal muscle and fat, oxidative enzymes concurrently and could thus be a and increases mitochondrial function in rodent models of candidate molecule. Type 2 diabetes. Importantly, transgenic overexpression SIRT1 catalyses the NAD + -dependent deacetylation of SIRT1 in mice recapitulates many of the findings of ε-amino-acetylated lysine residues from protein produced by pharmacological activation of SIRT1 [127]. substrates [125]. It regulates a variety of cellular functions, Moreover, SIRT1 is probably also a novel target in such as genome maintenance, longevity and metabolism preventing atherosclerosis. TIMP3 (tissue inhibitor of [126]. SIRT1 has been implicated as a key mediator of metalloproteinases 3) levels are reduced in atherosclerotic the pathways downstream of calorie restriction, a dietary plaques from subjects with Type 2 diabetes and are

regimen shown to extend the lifespan of many species, increased by SIRT1 [128]. SIRT1 down-regulates AT1 including yeast, worms, flies, mice, dogs and rhesus receptor expression in vascular smooth muscle cells [129]. monkeys. Humans on long-term calorie restriction have Endothelium-specific overexpression of SIRT1 decreases lower levels of inflammation, show fewer signs of atherosclerosis in ApoE (apolipoprotein E)-KO mice cardiovascular aging and better memory performance. [130].

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 85

Figure 5 Possible role of SIRT1 as a molecular switch for cardiovascular oxidative stress

Through the up-regulation of GCH1 expression, SIRT1 enhances BH4 biosynthesis and reverses eNOS uncoupling. Increased expression of SOD1, SOD2, GPx1 and catalases by SIRT1 enhances ROS inactivation in cardiovascular tissues.

There is evidence that SIRT1 also modulates of ApoE-KO mice with the SIRT1 activator resveratrol cardiovascular oxidative stress. Moderate overexpression results in enhanced expression of GCH1 and elevated − of SIRT1 (up to 7.5-fold) enhances the expression of levels of BH4 [134]. At the same time, cardiac O2 catalase and protects the heart from oxidative stress production in resveratrol-treated mice is markedly induced by paraquat [131]. A very high level of SIRT1 reduced to a level that cannot be lowered any further (12.5-fold), however, increases apoptosis/fibrosis and by l-NAME. These data suggest that eNOS is no − stimulates cardiomyopathy. The latter may be due to longer producing O2 in resveratrol-treated ApoE- mitochondrial dysfunction and NAD + depletion [131]. KO mice, i.e. resveratrol can reverse eNOS uncoupling SIRT1 is a key regulator of oxidative stress in [134]. Resveratrol has both SIRT1-dependent and - the cardiovascular system by targeting multiple redox independent effects. The up-regulation of GCH1 by genes simultaneously. Under a number of pathological resveratrol is probably mediated by SIRT1 as this conditions, the enzymatic reduction of molecular oxygen effect can be reduced by the SIRT1 inhibitor sirtinol by eNOS is no longer coupled to l-arginine oxidation, or by siRNA (small interfering RNA)-mediated SIRT1 − resulting in the production of O2 rather than NO. knockdown [134]. This phenomenon is referred to as eNOS uncoupling Treatment of ApoE-KO mice with resveratrol also (see above) [132]. Evidence for eNOS uncoupling has causes a significant down-regulation of Nox2 and Nox4 been obtained in animal models of CVD, as well as in cardiac tissue [134]. Similar effects are also observed in in patients with endothelial dysfunction resulting from cultured endothelial cells. Nox4 is the most predominant hypercholesterolaemia, diabetes mellitus or essential Nox isoform in human EA.hy 926 endothelial cells [135]. hypertension, and in chronic smokers [76]. Treatment of these cells with resveratrol results in a The hypercholesterolaemic atherosclerosis-prone dose-dependent down-regulation of Nox4 [135]. siRNA- ApoE-KO mice can be used as a model of oxidative mediated knockdown of SIRT1 has no effect on the Nox4 stress and eNOS uncoupling. The major cause down-regulation by resveratrol, indicating that the effects of eNOS uncoupling is a deficiency of the essential of resveratrol on Nox4 is SIRT1-independent. eNOS cofactor BH4 due to oxidative-stress-mediated Resveratrol enhances the expression of all three SOD oxidation of the molecule. BH4 supplementation is isoforms, GPx1 and catalase in both ApoE-KO mice capable of correcting eNOS dysfunction in several and cultured human EA.hy 926 endothelial cells [134]. types of pathophysiology [76]. Untreated ApoE-KO Inhibition of SIRT1 with sirtinol or knockdown of SIRT1 mice show increased oxidative degradation of BH4 [133] reduces the effects of resveratrol on SOD1, SOD2 and and significant ROS production in their aortas GPx1, but not that of SOD3 or catalase in EA.hy 926 [133] and hearts [134]. Both aortic [133] and cardiac [134] endothelial cells [134]. These results indicate that the − O2 production can be reduced by the NOS inhibitor effects of resveratrol on SOD1, SOD2 and GPx1 are l-NAME (NG-nitro-l-arginine methyl ester), indicating SIRT1-dependent, whereas its effects on the SOD3 and that eNOS is in an uncoupled state and that it produces catalase are probably SIRT1-independent in this cell type. ROS in this pathological model. In contrast with these results obtained in human EA.hy

BH4 is synthesized from GTP via a de novo pathway, 926 endothelial cells, SIRT1 overexpression has been with GCH1 acting as the rate-limiting enzyme. Treatment shown to induce catalase in cardiac tissues [131].

C The Authors Journal compilation C 2012 Biochemical Society 86 A. F. Chen and others

Figure 6 Summary of the mechanisms underlying vascular (endothelial) dysfunction by oxidative stress Known cardiovascular risk factors (for example smoking, hypertension, hyperlipidaemia and diabetes) activate the RAAS (renin–angiotensin–aldosterone system), leading − to elevated AngII levels as well as increased endothelial and smooth muscle O2 formation from NADPH oxidase activation by PKC and from the mitochondria. − • • − − O2 reacts with NO , thereby decreasing NO bioavailability in favour of ONOO formation. ONOO causes the uncoupling of eNOS due to oxidation of BH4 to

BH2 (dihydrobiopterin) and nitration/inactivation of PGI2S[PGI2 (prostacyclin) synthase]. Direct proteasome-dependent degradation of the BH4 synthase GCH further − • contributes to eNOS uncoupling. Uncoupled NOS produces O2 instead of NO and nitrated PGI2S produces no PGI2, but activated COX-2 (cyclo-oxygenase-2) − (due to increased peroxide tone) generates vasoconstrictive PGH2 (prostaglandin H2). Inhibition of smooth muscle sGC (soluble guanylate cyclase) by O2 and ONOO − contributes to vascular dysfunction, as well as increased inactivation of cGMP by PDEs (phosphodiesterases) and oxidative stress increases the sensitivity to

vasoconstrictors such as ET-1 (endothelin-1). Mitochondrial ROS formation is modulated by oxidative activation of mtKATP, leading to altered mitochondrial membrane potential and permeability. Upon uncoupling of mitochondrial respiratory complexes, the mPTP may be oxidatively opened allowing mtROS (mitochondrial ROS) to escape into the cytosol activating the PKC/NADPH oxidase (NADPH-Ox) system. This Figure is modified from Thomas Munzel,¨ Andreas Daiber and Alexander Mulsch, Explaining the phenomenon of nitrate tolerance. Circulation Research, 97 (7), 618–628, with permission.

The SIRT1-dependent up-regulation of antioxidant Given the pivotal role of oxidative stress in pathology, enzymes also occurs in cardiovascular tissues in other SIRT1 is a promising target for the development of new disease models. TO-2 hamsters have a genetic defect in drugs to treat CVDs. the δ-sarcoglycan gene and spontaneously develop dilated cardiomyopathy. Oral treatment of TO-2 hamsters with the SIRT1 activator resveratrol increases SOD2 levels OUTLOOK AND CLINICAL IMPLICATIONS in cardiomyocytes, suppresses fibrosis, preserves cardiac function and significantly improves survival. In C2C12 Figure 6 provides a summary of the oxidant mechanisms myoblasts, the up-regulation of SOD2 by resveratrol that contribute to vascular dysfunction. Possible depends on the level of nuclear SIRT1. siRNAs against antioxidant therapeutic interventions are also provided SOD2 or SIRT1 [136] can block the cell-protective within the scheme. Despite the fact that most large effects of resveratrol. Thus SIRT1 appears to act as a clinical trials do not support a beneficial role of oral molecular switch able to modulate oxidative stress in antioxidant therapy for CVDs, there is ample evidence − the cardiovascular system (Figure 5). It prevents O2 that antioxidant therapy may be protective when directed production from uncoupled eNOS and enhances ROS to specific sites and administrated acutely. In addition, detoxification by SOD1, SOD2, GPx1 and catalase. some of the most promising therapeutic targets have

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 87 not been exploited thus far (e.g. mitochondrial pores or Bugher Foundation Award in Stroke from the American NADPH oxidases), and there is a poor recognition of the Heart Association [grant number 0575092N] (to F.M.F.); pleiotropic antioxidant properties of established drugs. the Deutsche Forschungsgemeinschaft (DFG), Bonn, In addition to the strategies aimed at developing new Germany [grant number LI-1042/1-1] (to H.L.); and synthetic therapeutics that have been reviewed, there is the Canadian Society of Pharmacology and Therapeutics also the possibility of gene therapy with microRNAs and (to I.L.). antagomirs, which will undoubtedly create exciting new avenues for future strategies for antioxidant therapy, for example by silencing of oxidant-producing systems or REFERENCES inducing antioxidant systems. 1 Forstermann, U. (2008) Oxidative stress in vascular disease: causes, defense mechanisms and potential therapies. Nat. Clin. Pract. Cardiovasc. Med. 5, 338–349 CONCLUSIONS 2 Rahman, K. (2007) Studies on free radicals, antioxidants, and co-factors. Clin. Interv. Aging 2, 219–236 3 Rhee, S. G. (1999) Redox signaling: hydrogen peroxide Despite extensive basic science evidence for the benefit of as intracellular messenger. Exp. Mol. Med. 31, 53–59 antioxidants, clinical trials show little benefit from oral 4 Stamler, J. S. (1994) Redox signaling: nitrosylation and related target interactions of nitric oxide. Cell 78, antioxidant supplements. We have reviewed details of the 931–936 clinical trials and have discussed possible reasons that oral 5 Ullrich, V. and Kissner, R. (2006) Redox signaling: bioinorganic chemistry at its best. J. Inorg. Biochem. antioxidant supplements have been ineffective as either 100, 2079–2086 preventative or therapeutic agents in CVDs. Vascular 6 Brandes, R. P. and Kreuzer, J. (2005) Vascular NADPH dysfunction routinely occurs with aging. Activation of oxidases: molecular mechanisms of activation. Cardiovasc. Res. 65, 16–27 AT1 receptors by local or circulating AngII appears to be 7 Rodriguez-Manas, L., El-Assar, M., Vallejo, S., a key element for activation of pathways that increase Lopez-D´ origa,´ P., Solıs,´ J., Petidier, R., Montes, M., Nevado, J., Castro, M., Gomez-Guerrero,´ C. et al. the production of ROS and promote oxidative stress. (2009) Endothelial dysfunction in aged humans is related Protective molecules within vascular cells, including with oxidative stress and vascular inflammation. Aging SOD1, SOD2, and PPARγ , may function normally to Cell 8, 226–238 8 Golbidi, S. and Laher, I. (2010) Antioxidant therapy in suppress vascular aging. An emerging area of importance human endocrine disorders. Med. Sci. Monit. 16, is the negative role of oxidative stress in the functioning RA9–R24 9 Paravicini, T. M. and Touyz, R. M. (2008) NADPH and mobilization of EPCs. Thus increasing the intrinsic oxidases, reactive oxygen species, and hypertension: antioxidant activities of EPCs would bode well in the clinical implications and therapeutic possibilities. ongoing efforts to repair the damaged intimal lining of Diabetes Care 2, S170–S180 10 Valko, M., Leibfritz, D., Moncol, J., Cronin, M. T., blood vessels in diabetes, hypertension etc. It is likely Mazur, M. and Telser, J. (2007) Free radicals and that targeting the sources that generate free radicals, antioxidants in normal physiological functions and human disease. Int. J. Biochem. Cell. Biol. 39, 44–84 such as uncoupled eNOS, will make great strides. An 11 Touyz, R. M. (2004) Reactive oxygen species, vascular additional approach would be the modulation of SIRT1, oxidative stress, and redox signaling in hypertension: which can be considered a molecular switch that is able what is the clinical significance? Hypertension 44, − 248–252 to inhibit O2 production from uncoupled eNOS, while 12 Folkow, B. and Svanborg, A. (1993) Physiology of also augmenting free radical metabolism by intracellular cardiovascular aging. Physiol. Rev. 73, 725–764 13 Ungvari, Z., Kaley, G., de Cabo, R., Sonntag, W. E. and enzymes. Csiszar, A. (2010) Mechanisms of vascular aging: new perspectives. J. Gerontol. A Biol. Sci. Med. Sci. 65, 1028–1041 14 Faraci, F.M (2010) Cerebral vascular dysfunction with ACKNOWLEDGEMENTS aging. In Handbook of the Biology of Aging, 7th edn (Masoro, E. J and Austad, S. N., eds), pp. 405–419, The graphical assistance of Margot Neuser and Thilo Academic Press, New York 15 Chrissobolis, S. and Faraci, F. M. (2008) The role of Weckmuller¨ is gratefully acknowledged. oxidative stress and NADPH oxidase in cerebrovascular disease. Trends Mol. Med. 14, 495–502 16 Faraci, F. M. and Heistad, D. D. (1998) Regulation of the cerebral circulation: role of endothelium and potassium FUNDING channels. Physiol. Rev. 78, 53–97 17 Chen, J., Zacharek, A., Zhang, C., Jiang, H., Li, Y., Our own work was supported the National Institutes Roberts, C., Lu, M., Kapke, A. and Chopp, M. (2005) Endothelial nitric oxide synthase regulates brain-derived of Health [grant number GM077352], Veterans Affairs neurotrophic factor expression and neurogenesis after RR&D Merit Award [grant number I0RX000244] and the stroke in mice. J. Neurosci. 25, 2366–2375 American Diabetes Association [grant number 7-11-BS- 18 Jagtap, P. and Szabo, C. (2005) Poly(ADP-ribose) polymerase and the therapeutic effects of its inhibitors. 23] (to A.F.C.); the Johannes Gutenberg University and Nat. Rev. Drug Discovery 4, 421–440 Medical Center Mainz (MAIFOR and Forschungsfonds 19 Modrick, M. L., Didion, S. P., Sigmund, C. D. and Faraci, F. M. (2009) Role of oxidative stress and AT1 grants) (to A.D.); the National Institutes of Health receptors in cerebral vascular dysfunction with aging. [grant numbers HL-38901, HL-62984, NS-24621], a Am. J. Physiol. Heart Circ. Physiol. 296, H1914–H1919

C The Authors Journal compilation C 2012 Biochemical Society 88 A. F. Chen and others

20 Park, L., Anrather, J., Girouard, H., Zhou, P. and 38 Doerries, C., Grote, K., Hilfiker-Kleiner, D., Luchtefeld, Iadecola, C. (2007) Nox2-derived reactive oxygen M., Schaefer, A., Holland, S. M., Sorrentino, S., Manes, species mediate neurovascular dysregulation in the aging C., Schieffer, B., Drexler, H. and Landmesser, U. (2007) mouse brain. J. Cereb. Blood Flow Metab. 27, 1908–1918 Critical role of the NAD(P)H oxidase subunit p47phox 21 Benigni, A., Cassis, P. and Remuzzi, G. (2010) for left ventricular remodeling/dysfunction and survival Angiotensin II revisited: new roles in inflammation, after myocardial infarction. Circ. Res. 100, 894–903 immunology and aging. EMBO Mol. Med. 2, 247–257 39 Wenzel, P., Schuhmacher, S., Kienhofer, J., Muller,¨ J., 22 Wang, M., Takagi, G., Asai, K., Resuello, R. G., Hortmann, M., Oelze, M., Schulz, E., Treiber, N., Natividad, F. F., Vatner, D. E., Vatner, S. F. and Lakatta, Kawamoto, T., Scharffetter-Kochanek, K. et al. (2008) E. G. (2003) Aging increases aortic MMP-2 activity and Manganese superoxide dismutase and aldehyde angiotensin II in nonhuman primates. Hypertension 41, dehydrogenase deficiency increase mitochondrial 1308–1316 oxidative stress and aggravate age-dependent vascular 23 Basso, N., Cini, R., Pietrelli, A., Ferder, L., Terragno, dysfunction. Cardiovasc. Res. 80, 280–289 N. A. and Inserra, F. (2007) Protective effect of 40 Heart Protection Study Collaborative Group long-term angiotensin II inhibition. Am. J. Physiol. MRC/BHF Heart Protection Study of cholesterol Heart Circ. Physiol. 293, H1351–H1358 lowering with simvastatin in 20536 high-risk individuals: 24 Didion, S. P., Kinzenbaw, D. A., Schrader, L. I. and a randomised placebo-controlled trial (2002) Lancet 360, Faraci, F. M. (2006) Heterozygous CuZn superoxide 7–22 dismutase deficiency produces a vascular phenotype 41 The Coronary Drug Project Group (1975) Clofibrate with aging. Hypertension 48, 1072–1079 and niacin in coronary heart disease. JAMA, J. Am. Med. 25 Brown, K. A., Didion, S. P., Andresen, J. J. and Faraci, Assoc. 231, 360–381 F. M. (2007) Effect of aging, MnSOD deficiency, and 42 Rubins, H. B., Robins, S. J., Collins, D., Fye, C. L., genetic background on endothelial function: evidence for Anderson, J. W., Elam, M. B., Faas, F. H., Linares, E., MnSOD haploinsufficiency. Arterioscler. Thromb. Vasc. Schaefer, E. J., Schectman, G. et al. (1999) Gemfibrozil Biol. 27, 1941–1946 for the secondary prevention of coronary heart disease in 26 Kuro-o, M. (2010) Klotho. Pflugers Arch. 459, 333–343 men with low levels of high-density lipoprotein 27 Rakugi, H., Matsukawa, N., Ishikawa, K., Yang, J., Imai, cholesterol. N. Engl. J. Med. 341, 410–418 M., Ikushima, M., Maekawa, Y., Kida, I., Miyazaki, J. 43 GISSI-Prevenzione Investigators (1999) Dietary and Ogihara, T. (2007) Anti-oxidative effect of Klotho supplementation with n-3 polyunsaturated fatty acids on endothelial cells through cAMP activation. Endocrine and vitamin E after myocardial infarction: results of the 31, 82–87 GISSI-Prevenzione trial Lancet 354, 447–455 28 Wang, Y. and Sun, Z. (2009) Klotho gene delivery 44 de Lorgeril, M., Salen, P., Martin, J. L., Monjaud, I., prevents the progression of spontaneous hypertension Delaye, J. and Mamelle, N. (1999) Mediterranean diet, and renal damage. Hypertension 54, 810–817 traditional risk factors, and the rate of cardiovascular 29 Durrant, J. R., Seals, D. R., Connell, M. L., Russell, M. J., complications after myocardial infarction: final report of Lawson, B. R., Folian, B. J., Donato, A. J. and the Lyon Diet Heart Study. Circulation. 99, 779–785 Lesniewski, L. A. (2009) Voluntary wheel running 45 Touyz, R. M. and Schiffrin, E. L. (2004) Reactive oxygen restores endothelial function in conduit arteries of old species in vascular biology: implications in hypertension. mice: direct evidence for reduced oxidative stress, Histochem. Cell Biol. 122, 339–352 increased superoxide dismutase activity and 46 Vivekananthan, D. P., Penn, M. S., Sapp, S. K., Hsu, A. down-regulation of NADPH oxidase. J. Physiol. 587, and Topol, E. J. (2003) Use of antioxidant vitamins for 3271–3285 the prevention of cardiovascular disease: meta-analysis 30 Rippe, C., Lesniewski, L., Connell, M., LaRocca, T., of randomised trials. Lancet 361, 2017–2023 Donato, A. and Seals, D. (2010) Short-term calorie 47 Sawayama, Y., Shimizu, C., Maeda, N., Tatsukawa, M., restriction reverses vascular endothelial dysfunction in Kinukawa, N., Koyanagi, S., Kashiwagi, S. and Hayashi, old mice by increasing nitric oxide and reducing J. (2002) Effects of probucol and pravastatin on common oxidative stress. Aging Cell 9, 304–312 carotid atherosclerosis in patients with asymptomatic 31 Ketsawatsomkron, P., Pelham, C. J., Groh, S., Keen, H. hypercholesterolemia. Fukuoka Atherosclerosis Trial L., Faraci, F. M. and Sigmund, C. D. (2010) Does (FAST). J. Am. Coll. Cardiol. 39, 610–616 peroxisome proliferator-activated receptor-γ (PPARγ ) 48 Walldius, G., Erikson, U., Olsson, A. G., Bergstrand, L., protect from hypertension directly through effects in the Hadell,˚ K., Johansson, J., Kaijser, L., Lassvik, C., vasculature? J. Biol. Chem. 285, 9311–9316 Molgaard,¨ J., Nilsson, S. et al. (1994) The effect of 32 Beyer, A. M., Baumbach, G. L., Halabi, C. M., Modrick, probucol on femoral atherosclerosis: the Probucol M. L., Lynch, C. M., Gerhold, T. D., Ghoneim, S. M., de Quantitative Regression Swedish Trial (PQRST). Am. J. Lange, W. J., Keen, H. L., Tsai, Y. S. et al. (2008) Cardiol. 74, 875–883 Interference with PPARγ signaling causes cerebral 49 Tardif, J. C., McMurray, J. J., Klug, E., Small, R., Schumi, vascular dysfunction, hypertrophy, and remodeling. J., Choi, J., Cooper, J., Scott, R., Lewis, E. F., L’Allier, P. Hypertension 51, 867–871 L. and Pfeffer, M. A. (2008) Effects of succinobucol 33 Sung, B., Park, S., Yu, B. P. and Chung, H. Y. (2004) (AGI-1067) after an acute coronary syndrome: a Modulation of PPAR in aging, inflammation, and calorie randomised, double-blind, placebo-controlled trial. restriction. J. Gerontol. A Biol. Sci. Med. Sci. 59, Lancet 371, 1761–1768 997–1006 50 Heart Protection Study Collaborative Group 34 Argmann, C., Dobrin, R., Heikkinen, S., Auburtin, A., MRC/BHF Heart Protection Study of antioxidant Pouilly, L., Cock, T. A., Koutnikova, H., Zhu, J., Schadt, vitamin supplementation in 20536 high-risk individuals: E. E., Auwerx, J. et al. (2009) Pparγ 2isakeydriverof a randomised placebo-controlled trial (2002) Lancet 360, longevity in the mouse. PLoS Genet. 5, e1000752 23–33 35 Zhang, H., Li, Y., Fan, Y., Wu, J., Zhao, B., Guan, Y., 51 Rembold, C. M. (2007) To statin or to non-statin in Chien, S. and Wang, N. (2008) Klotho is a target gene of coronary disease – considering absolute risk is the PPAR-γ . Kidney Int. 74, 732–739 answer. Atherosclerosis 195, 1–6 36 Nicolakakis, N., Aboulkassim, T., Ongali, B., Lecrux, 52 Roberts, II, L. J., Oates, J. A., Linton, M. F., Fazio, S., C., Fernandes, P., Rosa-Neto, P., Tong, X. K. and Hamel, Meador, B. P., Gross, M. D., Shyr, Y. and Morrow, J. D. E. (2008) Complete rescue of cerebrovascular function in (2007) The relationship between dose of vitamin E and aged Alzheimer’s disease transgenic mice by antioxidants suppression of oxidative stress in humans. Free Radical and pioglitazone, a peroxisome proliferator-activated Biol. Med. 43, 1388–1393 receptor γ agonist. J. Neurosci. 28, 9287–9296 53 Collins, A. R., Lyon, C. J., Xia, X., Liu, J. Z., Tangirala, 37 Heitzer, T., Schlinzig, T., Krohn, K., Meinertz, T. and R. K., Yin, F., Boyadjian, R., Bikineyeva, A., Pratico,` D., Munzel, T. (2001) Endothelial dysfunction, oxidative Harrison, D. G. and Hsueh, W. A. (2009) Age- stress, and risk of cardiovascular events in patients with accelerated atherosclerosis correlates with failure to coronary artery disease. Circulation 104, 2673–2678 upregulate antioxidant genes. Circ. Res. 104, e42–e54

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 89

54 Sofi, F., Conti, A. A., Gori, A. M., Eliana Luisi, M. L., 73 Schulz, E., Jansen, T., Wenzel, P., Daiber, A. and Munzel, Casini, A., Abbate, R. and Gensini, G. F. (2007) Coffee T. (2008) Nitric oxide, tetrahydrobiopterin, oxidative consumption and risk of coronary heart disease: a stress, and endothelial dysfunction in hypertension. meta-analysis. Nutr. Metab. Cardiovasc. Dis. 17, 209–223 Antioxid. Redox Signaling 10, 1115–1126 55 Santos, C., Costa, J., Santos, J., Vaz-Carneiro, A. and 74 Alp, N. J., Mussa, S., Khoo, J., Cai, S., Guzik, T., Lunet, N. (2010) Caffeine intake and dementia: Jefferson, A., Goh, N., Rockett, K. A. and Channon, K. systematic review and meta-analysis. J. Alzheimers Dis. M. (2003) Tetrahydrobiopterin-dependent preservation 20, S187–S204 of nitric oxide-mediated endothelial function in diabetes 56 Lee, D. H., Folsom, A. R., Harnack, L., Halliwell, B. and by targeted transgenic GTP-cyclohydrolase I Jacobs, D.R.J. (2004) Does supplemental vitamin C overexpression. J. Clin. Invest. 112, 725–735 increase cardiovascular disease risk in women with 75 Schnabel, R., Blankenberg, S., Lubos, E., Lackner, K. J., diabetes? Am. J. Clin. Nutr. 80, 1194–1200 Rupprecht, H. J., Espinola-Klein, C., Jachmann, N., 57 Elhaimeur, F., Courderot-Masuyer, C., Nicod, L., Post, F., Peetz, D., Bickel, C. et al. (2005) Asymmetric Guyon, C., Richert, L. and Berthelot, A. (2002) Dietary dimethylarginine and the risk of cardiovascular events vitamin C supplementation decreases blood pressure in and death in patients with coronary artery disease: DOCA-salt hypertensive male Sprague–Dawley rats and results from the AtheroGene Study. Circ. Res. 97, this is associated with increased liver oxidative stress. e53–e59 Mol. Cell. Biochem. 237, 77–83 76 Li, H. and Forstermann, U. (2009) Prevention of 58 Heitzer, T., Just, H. and Munzel, T. (1996) Antioxidant atherosclerosis by interference with the vascular nitric vitamin C improves endothelial dysfunction in chronic oxide system. Curr. Pharm. Des. 15, 3133–3145 smokers. Circulation 94, 6–9 77 Li, Y., Huang, T. T., Carlson, E. J., Melov, S., Ursell, P. 59 Daiber, A. and Ullrich, V. (2002) Radikalchemie im C., Olson, J. L., Noble, L. J., Yoshimura, M. P., Berger, Organismus: Stickstoffmonoxid, Superoxid und C., Chan, P. H. et al. (1995) Dilated cardiomyopathy and Peroxynitrit. Chem. Unserer Zeit 36, 366–375 neonatal lethality in mutant mice lacking manganese 60 Daiber, A. and Bachschmid, M. (2007) Enzyme superoxide dismutase. Nat. Genet. 11, 376–381 inhibition by peroxynitrite-mediated tyrosine nitration 78 Jung, O., Marklund, S. L., Geiger, H., Pedrazzini, T., and thiol oxidation. Curr. Enzyme Inhib. 3, 103–117 Busse, R. and Brandes, R. P. (2003) Extracellular 61 Wendel, A., Fausel, M., Safayhi, H., Tiegs, G. and Otter, superoxide dismutase is a major determinant of nitric R. (1984) A novel biologically active seleno-organic oxide bioavailability: in vivo and ex vivo evidence from compound–II. Activity of PZ 51 in relation to ecSOD-deficient mice. Circ. Res. 93, 622–629 glutathione peroxidase. Biochem. Pharmacol. 33, 79 Daiber, A. (2010) Redox signaling (cross-talk) from and 3241–3245 to mitochondria involves mitochondrial pores and 62 Daiber, A., Zou, M. H., Bachschmid, M. and Ullrich, V. reactive oxygen species. Biochim. Biophys. Acta 1797, (2000) Ebselen as a peroxynitrite scavenger in vitro and 897–906 ex vivo. Biochem. Pharmacol. 59, 153–160 80 Di Lisa, F. and Bernardi, P. (2006) Mitochondria and 63 Pasternack, R. F. and Skowronek, W.R.J. (1979) Catalysis ischemia-reperfusion injury of the heart: fixing a hole. of the disproportionation of superoxide by Cardiovasc. Res. 70, 191–199 metalloporphyrins. J. Inorg. Biochem. 11, 261–267 81 Piot, C., Croisille, P., Staat, P., Thibault, H., Rioufol, G., 64 Misko, T. P., Highkin, M. K., Veenhuizen, A. W., Mewton, N., Elbelghiti, R., Cung, T. T., Bonnefoy, E., Manning, P. T., Stern, M. K., Currie, M. G. and Angoulvant, D. et al. (2008) Effect of cyclosporine on Salvemini, D. (1998) Characterization of the reperfusion injury in acute myocardial infarction. N. cytoprotective action of peroxynitrite decomposition Engl. J. Med. 359, 473–481 catalysts. J. Biol. Chem. 273, 15646–15653 82 Koopman, W. J., Verkaart, S., Visch, H. J., van der 65 Suzuki, H., DeLano, F. A., Parks, D. A., Jamshidi, N., Westhuizen, F. H., Murphy, M. P., van den Heuvel, L. Granger, D. N., Ishii, H., Suematsu, M., Zweifach, B. W. and Schmid-Schonbein,¨ G. W. (1998) Xanthine oxidase W., Smeitink, J. A. and Willems, P. H. (2005) Inhibition of complex I of the electron transport chain causes activity associated with arterial blood pressure in −• spontaneously hypertensive rats. Proc. Natl. Acad. Sci. O2 -mediated mitochondrial outgrowth. Am. J. U.S.A. 95, 4754–4759 Physiol. Cell Physiol. 288, C1440–C1450 66 Butler, R., Morris, A. D., Belch, J. J., Hill, A. and 83 Wenzel, P., Mollnau, H., Oelze, M., Schulz, E., Struthers, A. D. (2000) Allopurinol normalizes Wickramanayake, J. M., Muller,¨ J., Schuhmacher, S., endothelial dysfunction in type 2 diabetics with mild Hortmann, M., Baldus, S., Gori, T. et al. (2008) First hypertension. Hypertension 35, 746–751 evidence for a crosstalk between mitochondrial and 67 Lee, B. E., Toledo, A. H., Anaya-Prado, R., Roach, R. R. NADPH oxidase-derived reactive oxygen species in and Toledo-Pereyra, L. H. (2009) Allopurinol, xanthine nitroglycerin-triggered vascular dysfunction. Antioxid. oxidase, and cardiac ischemia. J. Investig. Med. 57, Redox Signaling 10, 1435–1447 902–909 84 Daiber, A., Oelze, M., Sulyok, S., Coldewey, M., Schulz, 68 Griendling, K. K. and FitzGerald, G. A. (2003) E., Treiber, N., Hink, U., Mulsch,¨ A., Oxidative stress and cardiovascular injury: part I: basic Scharffetter-Kochanek, K. and Munzel,¨ T. (2005) mechanisms and in vivo monitoring of ROS. Circulation Heterozygous deficiency of manganese superoxide + / − 108, 1912–1916 dismutase in mice (Mn-SOD ): a novel approach to 69 Heumuller, S., Wind, S., Barbosa-Sicard, E., Schmidt, H. assess the role of oxidative stress for the development of H., Busse, R., Schroder,¨ K. and Brandes, R. P. (2008) nitrate tolerance. Mol. Pharmacol. 68, 579–588 Apocynin is not an inhibitor of vascular NADPH 85 Esplugues, J. V., Rocha, M., Nunez, C., Bosca, I., Ibiza, oxidases but an antioxidant. Hypertension 51, 211–217 S., Herance, J. R., Ortega, A., Serrador, J. M., D’Ocon, P. 70 Jaquet, V., Scapozza, L., Clark, R. A., Krause, K. H. and and Victor, V. M. (2006) Complex I dysfunction and Lambeth, J. D. (2009) Small-molecule NOX inhibitors: tolerance to nitroglycerin: an approach based on ROS-generating NADPH oxidases as therapeutic mitochondrial-targeted antioxidants. Circ. Res. 99, targets. Antioxid. Redox Signaling 11, 2535–2552 1067–1075 71 Brandes, R. P. and Schroder, K. (2008) Differential 86 Dikalova, A. E., Bikineyeva, A. T., Budzyn, K., Budzyn, vascular functions of Nox family NADPH oxidases. K., Nazarewicz, R. R., McCann, L., Lewis, W., Harrison, Curr. Opin. Lipidol. 19, 513–518 D. G. and Dikalov, S. I. (2010) Therapeutic targeting of 72 Munzel, T., Daiber, A., Ullrich, V. and Mulsch, A. (2005) mitochondrial superoxide in hypertension. Circ. Res. Vascular consequences of endothelial nitric oxide 107, 106–116 synthase uncoupling for the activity and expression of 87 Fukai, T., Folz, R. J., Landmesser, U. and Harrison, the soluble guanylyl cyclase and the cGMP-dependent D. G. (2002) Extracellular superoxide dismutase and protein kinase. Arterioscler. Thromb. Vasc. Biol. 25, cardiovascular disease. Cardiovasc. Res. 55, 1551–1557 239–249

C The Authors Journal compilation C 2012 Biochemical Society 90 A. F. Chen and others

88 Shuvaev, V. V., Christofidou-Solomidou, M., Bhora, F., 103 Tepper, O. M., Galiano, R. D., Capla, J. M., Kalka, C., Laude, K., Cai, H., Dikalov, S., Arguiri, E., Solomides, Gagne, P. J., Jacobowitz, G. R., Levine, J. P. and Gurtner, C. C., Albelda, S. M., Harrison, D. G. and Muzykantov, G. C. (2002) Human endothelial progenitor cells from V. R. (2009) Targeted detoxification of selected reactive type II diabetics exhibit impaired proliferation, adhesion, oxygen species in the vascular endothelium. and incorporation into vascular structures. Circulation J. Pharmacol. Exp. Ther. 331, 404–411 106, 2781–2786 89 Fernandez, C., Hattan, C. M. and Kerns, R. J. (2006) 104 Penn, M. S. and Topol, E. J. (2001) Tissue factor, the Semi-synthetic heparin derivatives: chemical emerging link between inflammation, thrombosis, and modifications of heparin beyond chain length, sulfate vascular remodeling. Circ. Res. 89, 1–2 substitution pattern and N-sulfo/N-acetyl groups. 105 Scheubel, R. J., Zorn, H., Silber, R. E., Kuss, O., Carbohydr. Res. 341, 1253–1265 90 Taylor, A. L., Ziesche, S., Yancy, C., Carson, P., Morawietz, H., Holtz, J. and Simm, A. (2003) D’Agostino, Jr, R., Ferdinand, K., Taylor, M., Adams, K., Age-dependent depression in circulating endothelial Sabolinski, M., Worcel, M. and Cohn, J. N. (2004) progenitor cells in patients undergoing coronary artery Combination of isosorbide dinitrate and hydralazine in bypass grafting. J. Am. Coll. Cardiol. 42, 2073–2080 blacks with heart failure. N. Engl. J. Med. 351, 2049–2057 106 Tepper, O. M., Galiano, R. D., Capla, J. M., Kalka, C., 91 Daiber, A., Oelze, M., Coldewey, M., Huth, C., Gagne, P. J., Jacobowitz, G. R., Levine, J. P. and Gurtner, Schildknecht, S., Bachschmid, M., Nazirisadeh, Y., G. C. (2002) Human endothelial progenitor cells from Ullrich, V., Mulsch,¨ A., Munzel,¨ T. and Tsilimingas, N. type II diabetics exhibit impaired proliferation, adhesion, (2005) Hydralazine is a powerful inhibitor of and incorporation into vascular structures. Circulation peroxynitrite formation as a possible explanation for its 106, 2781–2786 beneficial effects on prognosis in patients with 107 Imanishi, T., Moriwaki, C., Hano, T. and Nishio, I. congestive heart failure. Biochem. Biophys. Res. (2005) Endothelial progenitor cell senescence is Commun. 338, 1865–1874 accelerated in both experimental hypertensive rats and 92 Munzel, T., Kurz, S., Rajagopalan, S., Thoenes, M., patients with essential hypertension. J. Hypertens. 23, Berrington, W. R., Thompson, J. A., Freeman, B. A. and 1831–1837 Harrison, D. G. (1996) Hydralazine prevents 108 Chen, D. D. and Chen, A. F. (2006) CuZn superoxide nitroglycerin tolerance by inhibiting activation of a dismutase deficiency: culprit of accelerated vascular membrane-bound NADH oxidase. A new action for an aging process. Hypertension 48, 1026–1028 old drug. J. Clin. Invest. 98, 1465–1470 109 Keswani, S. G., Katz, A. B., Lim, F. Y., Zoltick, P., Radu, 93 Thomas, G. R., DiFabio, J. M., Gori, T. and Parker, J. D. (2007) Once daily therapy with isosorbide-5- A., Alaee, D., Herlyn, M. and Crombleholme, T. M. mononitrate causes endothelial dysfunction in humans: (2004) Adenoviral mediated gene transfer of PDGF-B evidence of a free-radical-mediated mechanism. J. Am. enhances wound healing in type I and type II diabetic Coll. Cardiol. 49, 1289–1295 wounds. Wound Repair Regen. 12, 497–504 94 Munzel, T., Daiber, A. and Mulsch, A. (2005) Explaining 110 Imanishi, T., Kobayashi, K., Hano, T. and Nishio, I. the phenomenon of nitrate tolerance. Circ. Res. 97, (2005) Effect of estrogen on differentiation and 618–628 senescence in endothelial progenitor cells derived from 95 Schuhmacher, S., Wenzel, P., Schulz, E., Oelze, M., bone marrow in spontaneously hypertensive rats. Mang, C., Kamuf, J., Gori, T., Jansen, T., Knorr, M., Hypertens Res. 28, 763–772 Karbach, S. et al. (2010) Pentaerythritol tetranitrate 111 Asai, J., Takenaka, H., Kusano, K. F., Ii, M., Luedemann, improves angiotensin II-induced vascular dysfunction C., Curry, C., Eaton, E., Iwakura, A., Tsutsumi, Y., via induction of heme oxygenase-1. Hypertension 55, Hamada, H. et al. (2006) Topical sonic hedgehog gene 897–904 therapy accelerates wound healing in diabetes by 96 Oppermann,M.,Balz,V.,Adams,V.,Dao,V.T.,Bas,M., enhancing endothelial progenitor cell-mediated Suvorava, T. and Kojda, G. (2009) Pharmacological microvascular remodeling. Circulation 113, 2413–2424 induction of vascular extracellular superoxide dismutase 112 Thum, T., Fraccarollo, D., Schultheiss, M., Froese, S., expression in vivo. J. Cell. Mol. Med. 13, 1271–1278 Galuppo, P., Widder, J. D., Tsikas, D., Ertl, G. and 97 Pautz, A., Rauschkolb, P., Schmidt, N., Art, J., Oelze, Bauersachs, J. (2007) Endothelial nitric oxide synthase M., Wenzel, P., Forstermann,¨ U., Daiber, A. and Kleinert, H. (2009) Effects of nitroglycerin or uncoupling impairs endothelial progenitor cell pentaerithrityl tetranitrate treatment on the gene mobilization and function in diabetes. Diabetes 56, expression in rat hearts: evidence for cardiotoxic and 666–674 cardioprotective effects. Physiol. Genomics. 38, 176–185 113 Xie, H. H., Zhou, S., Chen, D. D., Channon, K. M., Su, 98 Gori, T. and Daiber, A. (2009) Non-hemodynamic D. F. and Chen, A. F. (2010) GTP cyclohydrolase I/BH4 effects of organic nitrates and the distinctive pathway protects EPCs via suppressing oxidative stress characteristics of pentaerithrityl tetranitrate. Am. J. and thrombospondin-1 in salt-sensitive hypertension. Cardiovasc. Drugs 9, 7–15 Hypertension 56, 1137–1144 98a Schuhmacher, S., Oelze, M., Bollmann, F., Kleinert, H., 114 Wang, X. R., Zhang, M. W., Chen, D. D., Zhang, Y. and Otto, C., Heeren, T., Steven, S., Hausding, M., Knorr, Chen, A. F. (2011) AMP-activated protein kinase rescues M., Pautz, A. et al. (2011) Vascular dysfunction in the angiogenic functions of endothelial progenitor cells experimental diabetes is improved by pentaerithrityl via manganese superoxide dismutase induction in type 1 tetranitrate but not isosorbide-5-mononitrate therapy. diabetes. Am. J. Physiol. Endocrinol. Metab. 300, Diabetes 60, 2608–2616 E1135–E1145 99 Asahara, T., Murohara, T., Sullivan, A., Silver, M., van 115 Di Stefano, V., Cencioni, C., Zaccagnini, G., Magenta, der Zee, R., Li, T., Witzenbichler, B., Schatteman, G. and A., Capogrossi, M. C. and Martelli, F. (2009) p66ShcA Isner, J. M. (1997) Isolation of putative progenitor modulates oxidative stress and survival of endothelial endothelial cells for angiogenesis. Science 275, 964–967 progenitor cells in response to high glucose. Cardiovasc. 100 Rehman, J., Li, J., Orschell, C. M. and March, K. L. Res. 82, 421–429 (2003) Peripheral blood ‘endothelial progenitor cells’ are derived from monocyte/macrophages and secrete 116 He, T., Joyner, M. J. and Katusic, Z. S. (2009) Aging angiogenic growth factors. Circulation 107, 1164–1169 decreases expression and activity of glutathione 101 Fadini, G. P. and Avogaro, A. (2010) Potential peroxidase-1 in human endothelial progenitor cells. manipulation of endothelial progenitor cells in diabetes Microvasc. Res. 78, 447–452 and its complications. Diabetes Obes. Metab. 12, 117 Fujii, H., Li, S. H., Szmitko, P. E., Fedak, P. W. and 570–583 Verma, S. (2006) C-reactive protein alters antioxidant 102 Jarajapu, Y. P. and Grant, M. B. (2010) The promise of defenses and promotes apoptosis in endothelial cell-based therapies for diabetic complications: progenitor cells. Arterioscler. Thromb. Vasc. Biol. 26, challenges and solutions. Circ. Res. 106, 854–869 2476–2482

C The Authors Journal compilation C 2012 Biochemical Society Free radical biology of the cardiovascular system 91

118 Laufs, U., Werner, N., Link, A., Endres, M., Wassmann, 128 Cardellini, M., Menghini, R., Martelli, E., Casagrande, S., Jurgens,¨ K., Miche, E., Bohm,¨ M. and Nickenig, G. V., Marino, A., Rizza, S., Porzio, O., Mauriello, A., (2004) Physical training increases endothelial progenitor Solini, A., Ippoliti, A. et al. (2009) TIMP3 is reduced in cells, inhibits neointima formation, and enhances atherosclerotic plaques from subjects with type 2 angiogenesis. Circulation 109, 220–226 diabetes and increased by SirT1. Diabetes 58, 2396–2401 119 Werner, C., Kamani, C. H., Gensch, C., Bohm, M. and 129 Miyazaki, R., Ichiki, T., Hashimoto, T., Inanaga, K., Laufs, U. (2007) The peroxisome proliferator-activated Imayama, I., Sadoshima, J. and Sunagawa, K. (2008) receptor-γ agonist pioglitazone increases number and SIRT1, a longevity gene, downregulates angiotensin II function of endothelial progenitor cells in patients with type 1 receptor expression in vascular smooth muscle coronary artery disease and normal glucose tolerance. cells. Arterioscler. Thromb. Vasc. Biol. 28, 1263–1269 Diabetes 56, 2609–2615 130 Zhang, Q. J., Wang, Z., Chen, H. Z., Zhou, S., Zheng, 120 Togliatto, G., Trombetta, A., Dentelli, P., Baragli, A., W., Liu, G., Wei, Y. S., Cai, H., Liu, D. P. and Liang, C. Rosso, A., Granata, R., Ghigo, D., Pegoraro, L., Ghigo, C. (2008) Endothelium-specific overexpression of class E. and Brizzi, M. F. (2010) Unacylated ghrelin rescues III deacetylase SIRT1 decreases atherosclerosis in endothelial progenitor cell function in individuals with apolipoprotein E-deficient mice. (2008) Cardiovasc. Res. type 2 diabetes. Diabetes 59, 1016–1025 80, 191–199 121 Marrotte, E. J., Chen, D. D., Hakim, J. S. and Chen, A. 131 Alcendor, R. R., Gao, S., Zhai, P., Zablocki, D., Holle, E., F. (2010) Manganese superoxide dismutase expression in Yu, X., Tian, B., Wagner, T., Vatner, S. F. and Sadoshima, endothelial progenitor cells accelerates wound J. (2007) Sirt1 regulates aging and resistance to oxidative healing in diabetic mice. J. Clin. Invest. 120, stress in the heart. Circ. Res. 100, 1512–1521 4207–4219 132 Li, H., Wallerath, T., Munzel, T. and Forstermann, U. 122 Murasawa, S., Llevadot, J., Silver, M., Isner, J. M., (2002) Regulation of endothelial-type NO synthase Losordo, D. W. and Asahara, T. (2002) Constitutive expression in pathophysiology and in response to drugs. human telomerase reverse transcriptase expression Nitric Oxide 7, 149–164 enhances regenerative properties of endothelial 133 Alp, N. J., McAteer, M. A., Khoo, J., Choudhury, R. P. progenitor cells. Circulation 106, 1133–1139 and Channon, K. M. (2004) Increased endothelial 123 Sasaki, K., Heeschen, C., Aicher, A., Ziebart, T., Honold, tetrahydrobiopterin synthesis by targeted transgenic J., Urbich, C., Rossig, L., Koehl, U., Koyanagi, M., GTP-cyclohydrolase I overexpression reduces Mohamed, A. et al. (2006) Ex vivo pretreatment of bone endothelial dysfunction and atherosclerosis in marrow mononuclear cells with endothelial NO ApoE-knockout mice. Arterioscler. Thromb. Vasc. Biol. synthase enhancer AVE9488 enhances their functional 24, 445–450 activity for cell therapy. Proc. Natl. Acad. Sci. U.S.A. 134 Xia, N., Daiber, A., Habermeier, A., Closs, E. I., Thum, 103, 14537–14541 T., Spanier, G., Lu, Q., Oelze, M., Torzewski, M., 124 Zhao, T., Li, J. and Chen, A. F. (2010) MicroRNA-34a Lackner, K. J. et al. (2010) Resveratrol reverses induces endothelial progenitor cell senescence and endothelial nitric-oxide synthase uncoupling in impedes its angiogenesis via suppressing silent apolipoprotein E knockout mice. J. Pharmacol. Exp. information regulator 1. (2010) Am. J. Physiol. Ther. 335, 149–154 Endocrinol. Metab. 299, E110–E116 135 Xu, H., Goettsch, C., Xia, N., Horke, S., Morawietz, H., 125 Milne, J. C. and Denu, J. M. (2008) The Sirtuin family: Forstermann,¨ U. and Li, H. (2008) Differential roles of therapeutic targets to treat diseases of aging. Curr. Opin. PKCα and PKCε in controlling the gene expression of Chem. Biol. 12, 11–17 Nox4 in human endothelial cells. Free Radical Biol. Med. 126 Michan, S. and Sinclair, D. (2007) Sirtuins in mammals: 44, 1656–1667 insights into their biological function. Biochem. J. 404, 136 Tanno, M., Kuno, A., Yano, T., Miura, T., Hisahara, S., 1–13 Ishikawa, S., Shimamoto, K. and Horio, Y. (2010) 127 Pfluger, P. T., Herranz, D., Velasco-Miguel, S., Serrano, Induction of manganese superoxide dismutase by M. and Tschop, M. H. (2008) Sirt1 protects against nuclear translocation and activation of SIRT1 promotes high-fat diet-induced metabolic damage. Proc. Natl. cell survival in chronic heart failure. J. Biol. Chem. 285, Acad. Sci. U.S.A. 105, 9793–9798 8375–8382

Received 8 November 2011/21 December 2011; accepted 6 January 2012 Published on the Internet 23 March 2012, doi:10.1042/CS20110562

C The Authors Journal compilation C 2012 Biochemical Society