Antioxidants: Positive or Negative Actors?

biomolecules Re iew An ioxidan s: Posi i e o Nega i e Ac o s? Baha e Salehi 1,2 , Miquel Ma o ell 3, Jack L. A bise 4,*, An oni Su eda 5, Na ália Ma ins 6,7,* , Pawan Kuma Mau ya 8, Mehdi Sha i i-Rad 9, P adeep Kuma 10,* and Ja ad Sha i i-Rad 11,12,* 1Medical E hics and Law Resea ch Cen e , Shahid Behesh i Uni e si y o Medical Sciences, Teh an 88777539, I an; baha [email p o ec ed] 2S uden Resea ch Commi ee, Shahid Behesh i Uni e si y o Medical Sciences, Teh an 22439789, I an 3Depa men o Nu i ion and Die e ics, Facul y o Pha macy, Uni e si y o Concepcion, Concepcion 4070386, Chile; [email p o ec ed] 4Depa men o De ma ology, Emo y Uni e si y School o Medicine, A lan a Ve e ans Adminis a ion Medical Cen e , Winship Cance Ins i u e, A lan a, GA 30322, USA 5 Resea ch G oup on Communi y Nu i ion and Oxida i e S ess (NUCOX) and CIBEROBN (Physiopa hology o Obesi y and Nu i ion CB12/03/30038), Uni e si y o Balea ic Islands, E-07122 Palma de Mallo ca, Balea ic Islands, Spain; [email p o ec ed] 6Facul y o Medicine, Uni e si y o Po o, Alameda P o . He nâni Mon ei o, 4200-319 Po o, Po ugal 7Ins i u e o Resea ch and Inno a ion in Heal h (i3S), Uni e si y o Po o, 4200-135 Po o, Po ugal 8Ami y Ins i u e o Bio echnology, Ami y Uni e si y, U a P adesh, Noida 201303, India; kuma [email p o ec ed] 9Zabol Medicinal Plan s Resea ch Cen e , Zabol Uni e si y o Medical Sciences, Zabol 61615585, I an; [email p o ec ed] 10 Depa men o Fo es y, No h Eas e n Regional Ins i u e o Science and Technology, Ni juli 791109, A unachal P adesh, India 11 Phy ochemis y Resea ch Cen e , Shahid Behesh i Uni e si y o Medical Sciences, Teh an 11369, I an 12 Depa men o Chemis y, Richa dson College o he En i onmen al Science Complex, The Uni e si y o Winnipeg, Winnipeg, MB R3B 2G3, Canada *Co espondence: ja bise@emo y.edu (J.L.A.); [email p o ec ed] (N.M.); [email p o ec ed] (P.K.); [email p o ec ed] (J.S.-R.); Tel.: +351-225-512-100 (N.M.); +91-76-2884-1479 (P.K.); +98-21-8820-0104 (J.S.-R.) Recei ed: 14 Sep embe 2018; Accep ed: 17 Oc obe 2018; Published: 25 Oc obe 2018   Abs ac : The e m “an ioxidan ” is one o he mos con using de ini ions in biological/medical sciences. In chemis y, “an ioxidan ” is simply concei ed “a compound ha emo es eac i e species, mainly hose oxygen-de i ed”, while in a cell con ex , he concep ual de ini ion o an an ioxidan is poo ly unde s ood. Indeed, non-clinically ecommended an ioxidan s a e o en consumed in la ge amoun s by he global popula ion, based on he belie ha cance , in lamma ion and degene a i e diseases a e igge ed by high oxygen le els (o eac i e oxygen species) and ha h ough blocking eac i e species p oduc ion, o ganic unbalances/diso de s can be p e en ed and/o e en ea ed. The popula i y o hese chemicals a ises in pa om he widesp ead public mis us o allopa hic medicine. In ac , eac i e oxygen species play a dual ole in dealing wi h di e en diso de s, since hey may con ibu e o disease onse and/o p og ession bu may also play a key ole in disease p e en ion. Fu he , he abili y o he mos commonly used supplemen s, such as i amins C, E, selenium, and he bal supplemen s o dec ease pa hologic eac i e oxygen species is no clea ly es ablished. Hence, he p esen e iew aims o p o ide a nuanced unde s anding o whe e cu en knowledge is and whe e i should go. Keywo ds: an ioxidan s; cance ; in lamma ion; na u al p oduc s; eac i e species Biomolecules 2018,8, 124; doi:10.3390/biom8040124 www.mdpi.com/jou nal/biomolecules Biomolecules 2018,8, 124 2 o 11 1. In oduc ion Ae obic cells p oduce eac i e oxygen species (ROS) as a me abolic p ocess byp oduc . ROS cause mac omolecules-oxida i e damages when body an ioxidan de enses a e o e whelmed. Jones ede ined oxida i e s ess as a dis up ion o edox signaling and con ol [ 1 ]. Howe e , a ce ain amoun o oxida i e damage akes place e en unde no mal condi ions, besides o be obse ed, a ise in damage a e wi h aging and disease p ocesses, since an ioxidan and epai mechanisms e iciency dec eases [ 2 ]. On he o he hand, oxida i e/ni osa i e s ess (O and NS) has also been ma kedly implica ed in he pa hophysiology o many diso de s [ 2 ]. B ie ly, O and NS a e de ined as an imbalance be ween ROS p oduc ion and ROS neu alizing/coun e ac ing o ganic abili y, h ough an ioxidan and de oxi ying mechanisms. Reac i e oxygen and ni ogen species (ROS/RNS), such as he supe oxide adical (O 2 −. ), hyd ogen pe oxide (H 2 O 2 ), he hyd oxyl adical ( . OH), ni ic oxide (NO) and pe oxyni i e (ONOO − ) a e na u ally p oduced by all ae obic o ganisms and exis in cells in homeos asis wi h an ioxidan molecules and enzymes. An ioxidan s a e compounds ha inhibi oxida ion. Oxida ion is a chemical eac ion ha can p oduce ee adicals, he eby leading o chain eac ions ha can damage he cells o o ganisms. An ioxidan s, such as hiols o asco bic acid ( i amin C) end hese chain eac ions. To balance he oxida i e s a e, plan s and animals main ain complex sys ems o o e lapping an ioxidan s, such as glu a hione and enzymes (e.g., ca alase and supe oxide dismu ase), p oduced in e nally, o he die a y an ioxidan s i amins C and E [3]. An ioxidan de enses induc ion o endogenous ROS/RNS le els educ ion is a apid and clea oxida i e s ess indica o . Indeed, ROS/RNS p oduc ion and accumula ion is a common denomina o in many diso de s and en i onmen al insul s, a he same ime ha can cause se ious cell damage leading o physiological dys unc ion and cell dea h in almos all ae obes [ 4 ]. An ioxidan he apy has long been in es iga ed as a means o educing he ex en o inju y esul ing om an ischemic s oke wi h a ying deg ees o success [5]. Enzyma ic and non-enzyma ic an ioxidan sys ems in he body, including supe oxide dismu ase (SOD), ca alase (CAT), glu a hione pe oxidase (GPX), lipid-soluble i amin E, ca o enes, and wa e -soluble i amin C, egula e he balance be ween ROS and an ioxidan s. On he o he hand, die a y an ioxidan s, mos ly ob ained om ui and ege able consump ion, has also been associa ed wi h a g ea balance be ween ee adicals and an ioxidan s a us, which helps o minimize oxida i e s ess and educe he isk o cance , ca dio ascula diseases and aging [ 6 ]. Indeed, die a y an ioxidan s comp ise a widesp ead g oup o colo ed molecules esponsible o mul iple bioac i e e ec s, b oadly co esponding o di e en colo s o dis inc biomolecules classes. Full consump ion ad an ages and heal h bene i s can also be eached h ough hei mixing since hey ac syne gis ically. Fo example, some an ioxidan s display a mo e impo an ole in p e en ing ce ain diseases, such as cance , while o he ones wo k be e igh ing degene a i e diseases. In ac , housands o an ioxidan s a e p esen in die a y pa e ns and ha some o hem may ha e s onge an ioxidan e ec s. Cance s induce oxida i e s ess by changing wi h p og ession and he esul ing an ioxidan s a us di e s om pa ien o pa ien . In “ea ly” lung cance s, seconda y eac i e oxygen species (SOSs) a e compa able o in e io o ha o heal hy people. The ole o i amins a ies wi h gende , ype o cance and me as asis [7]. 2. An ioxidan : Func ional De ini ion An ioxidan s en e ed in public’s nu i ional ocabula y in he 1990s, a a ime when esea che s we e slowly disco e ing how oxygen- igge ed ee adical eac ions in he body play a key ole in aging-associa ed ch onic diseases [ 8 ]. Cu en ly, an an ioxidan is de ined as any subs ance able o elimina e ROS and de i a i es (RNS, o eac i e sul u species, RSS), di ec ly o indi ec ly, ac ing as an an ioxidan de ense egula o , o eac i e species p oduc ion inhibi o [ 9 ]. ROS a e a molecules g oup p oduced o e cellula me abolism, due o mi ochond ial oxidases ac ion o by o he cellula compa men s, being his p oduc ion aised wi h mi ochond ial damage. RNS and RSS esul , espec i ely, om he eac ion be ween ROS and ni ic oxide and hiols [ 9 ]. ROS/RNS can cause cell damages by co alen joining wi h o he molecules and by s imula ing abno mal cell g ow h, o e en Biomolecules 2018,8, 124 3 o 11 senescence induc ion, which may lead o pe sis en cells popula ion ha p oduce in lamma o y cy okines a la ge amoun s. Acco ding o he common belie , so-called an ioxidan s may block eac i e species p oduc ion-dele e ious e ec s and, he e o e, block aging, in lamma ion and cance . Thei unc ion can be classi ied in o dis inc de ense lines, acco ding o hei mechanisms o ac ion: (a) p e en a i e agen s ha supp ess new adicals o ma ion (which includes enzymes, such as SOD, CAT and GPX, p o eins ha bind me als, like e i in and ce uloplasmin, and mine als such as selenium (Se), coppe (Cu), and zinc (Zn)); (b) adical sca enging agen s ha inhibi chain ini ia ion and/o p opaga ion, which includes glu a hione, albumin, i amins C and E, ca o enoids, and la onoids; (c) epai and de no o enzymes ha epai and econs i u e cell memb anes, which includes lipases, p o eases, DNA epai enzymes, ans e ases, and me hionine-sul oxide educ ases; and (d) adap a ion agen s ha gene a e app op ia e an ioxidan enzymes and ans e hem o essen ial si e o ac ion [ 9 , 10 ]. Redox balance is essen ial in heal hy cellula mic oen i onmen main enance. 3. An ioxidan Role in Redox Imbalance P e en ion: Gaps o Knowledge Redox imbalance is caused by a balance al e a ion be ween ROS and i s de i a i es and o he cell an ioxidan de ense sys em e iciency [ 10 ]. Cells and issues a e con inuously being exposed o eac i e species de i ed om me abolism o ex e nal ac o s, like smoking, pollu ion, pes icides, mic obes, alle gens, and ul a iole and gamma adia ion, which gene a e ee adicals, all o hem associa ed wi h bo h aging and o he diseases [ 11 ]. In addi ion, eac i e species p oduc ion can o e come body abili y o elimina e o o con ain hem. The an ioxidan s ield has been hampe ed by he ollowing gaps o knowledge: (1) Wha a e he co ec an ioxidan s doses? The e is ample e idence ha mo e is no be e and may be wo se [ 12 ]. Vi ually, all so-called an ioxidan s do no demons a e a classic dose- esponse bu may ha e opposi e e ec s wi h changing doses. (2) How a e an ioxidan s abso bed? Li le is known on an ioxidan s gu mic obial me abolism, as well as whe he an ioxidan s a ec gu mic o lo a in a p o- o an i-in lamma o y way. I is also poo ly unde s ood whe he an ioxidan s a e abso bed unchanged o me abolized o comple ely di e en compounds. One example is ellagic acid, a poo ly soluble compound ha is me abolized in o mo e soluble me aboli es, ha can media e i s an i-in lamma o y ac i i y [ 13 , 14 ]. (3) An ioxidan s na u ally occu ing may be abso bed as complexes o wo o mo e compounds and complexes may ha e e y di e en biological ac i i ies han isola ed compounds alone. (4) E en wi h ideal abso p ion and pene a ion in o umo s, umo s may espond wi h compensa o y pa hways. Fo example, ROS inhibi ion may lead o compensa o y mi ogen-ac i a ed p o ein (MAP) kinase ac i a ion, ha could mani es as mo e apid umo g ow h, despi e he ac ha hese umo s migh be de icien in nuclea ac o kappa B (NF-κB) ac i a ion and mo e suscep ible o chemo he apy o adia ion [15] (Figu e 1). 4. Ad e se E ec s o An ioxidan s The mos popula “an ioxidan s” o ms include i amins, such as i amin A ( e inol, e inoic acid), i amin C (L-asco bic acid, asco bic acid, asco ba e), i amin E ( α - ocophe ol), β -ca o ene, mine als, like Se, and na u ally-occu ing polyphenols, each one has a di e en e ec on body cells. Vi amins and β -ca o ene ha e conjuga ed double bonds and key unc ional g oups esponsible o hei an ioxidan ole and quali y as pigmen s in se e al oods, like ui s and ege ables. Below, we b ie ly summa ize he ad e se e ec s o hese popula an ioxidan s, o en consumed as supplemen s a much highe doses han hose ound in oods u s. While hei ad e se e ec s a e known in he medical communi y, hey a e no well-known among he popula ion, who belie e ha na u al p oduc s canno be oxic. Cze nichow in es iga ed he e ec o an ioxidan supplemen a ion o 7.5 yea s on me abolic synd ome (Me S) incidence and e en he epidemiologic associa ion be ween baseline se um an ioxidan concen a ions and Me S p ospec i e isk [ 16 ]. No bene icial e ec s o an ioxidan supplemen a ion we e obse ed in a gene ally well-nou ished popula ion. Baseline se um an ioxidan concen a ions o β -ca o ene and i amin C, howe e , we e nega i ely associa ed wi h Me S isk. Baseline se um zinc concen a ions we e posi i ely associa ed wi h he isk o de eloping Me S [16]. Pa k [17] ound ha Biomolecules 2018,8, 124 4 o 11 he e was no associa ion be ween die a y in akes o i amins A, C, and E and colon cance isk in his pooled analysis o hi een p ospec i e coho s udies. Howe e , o al i amins A, C, and E in akes we e each one in e sely associa ed wi h colon cance isk. Mul i i amin use, pa icula ly in combina ion wi h a single i amin A, C and/o E supplemen s use, was in e sely associa ed wi h colon cance isk. A low die a y in ake o an ioxidan i amins and mine als aises he incidence o ca dio ascula diseases and cance [17]. A e 7.5 yea s, low-dose an ioxidan supplemen a ion lowe ed o al cance incidence and all-cause mo ali y in men bu no in women. In ac , supplemen a ion may be e ec i e in men only because o hei low basal s a us o ce ain an ioxidan s, especially o β-ca o ene [18]. Biomolecules 2018, 8, x FOR PEER REVIEW 4 o 11 in akes o i amins A, C, and E and colon cance isk in his pooled analysis o hi een p ospec i e coho s udies. Howe e , o al i amins A, C, and E in akes we e each one in e sely associa ed wi h colon cance isk. Mul i i amin use, pa icula ly in combina ion wi h a single i amin A, C and/o E supplemen s use, was in e sely associa ed wi h colon cance isk. A low die a y in ake o an ioxidan i amins and mine als aises he incidence o ca dio ascula diseases and cance [17]. A e 7.5 yea s, low-dose an ioxidan supplemen a ion lowe ed o al cance incidence and all-cause mo ali y in men bu no in women. In ac , supplemen a ion may be e ec i e in men only because o hei low basal s a us o ce ain an ioxidan s, especially o β-ca o ene [18]. Figu e 1. O e iew o eac i e oxygen signaling. Supe oxide, which can a ise om NADPH oxidases ac ion o mi ochond ial leak, oxida i ely inac i a e p53, PTEN, and IkB, leading o Ak and NF-κB ac i a ion. Reac i e oxygen inhibi ion wi h NADPH oxidase inhibi o s can e e se his pheno ype. Glu a hione o ma ion h ough N 2 ac i a ion can lead o eac i e oxygen educ ion and, hus, possibly an NF-κB dec ease bu also may eac wi h chemo he apy and adia ion gene a ed species, hus, p o ec ing umo cells. Resea che s epo ed he exis ence o inc eased e a ogenici y isk o bi h de ec s among babies bo n om women who ook mo e han 10,000 IU i amin A pe day in he o m o supplemen s [19]. Indeed, excessi e die a y i amin A in ake has been associa ed wi h bi h de ec s in humans in se e al s udies epo ed in pas yea s [20–22]. A con olled clinical ial ound ha people who ook 25,000 IU o i amin A pe day o a median o 3.8 yea s had 11% inc ease in iglyce ides, 3% inc ease in o al choles e ol and 1% dec ease in high-densi y lipop o ein (HDL) choles e ol, unlike hose who did no ake i amin A [23]. In a ecen case epo , a 4-yea -old boy p esen ed se e al bone pains due o i amin A oxici y (600,000 IU e e y day o mo e han 3 mon hs) [2]. In ac , i has been epo ed ha excessi e i amin A in ake can accele a e bone loss and isk o hip ac u e, possibly due o i amin A-induced os eoclas s s imula ion, besides i inhibi s new bone o ma ion, inc easing os eopo osis isk [24]. On he o he hand, i amin C can be me abolized o oxala e and migh inc ease kidney oxala e exc e ion. Se e al s udies sugges ha i amin C supplemen s may inc ease u ina y oxala e Figu e 1. O e iew o eac i e oxygen signaling. Supe oxide, which can a ise om NADPH oxidases ac ion o mi ochond ial leak, oxida i ely inac i a e p53, PTEN, and IkB, leading o Ak and NF- κ B ac i a ion. Reac i e oxygen inhibi ion wi h NADPH oxidase inhibi o s can e e se his pheno ype. Glu a hione o ma ion h ough N 2 ac i a ion can lead o eac i e oxygen educ ion and, hus, possibly an NF- κ B dec ease bu also may eac wi h chemo he apy and adia ion gene a ed species, hus, p o ec ing umo cells. Resea che s epo ed he exis ence o inc eased e a ogenici y isk o bi h de ec s among babies bo n om women who ook mo e han 10,000 IU i amin A pe day in he o m o supplemen s [ 19 ]. Indeed, excessi e die a y i amin A in ake has been associa ed wi h bi h de ec s in humans in se e al s udies epo ed in pas yea s [ 20 – 22 ]. A con olled clinical ial ound ha people who ook 25,000 IU o i amin A pe day o a median o 3.8 yea s had 11% inc ease in iglyce ides, 3% inc ease in o al choles e ol and 1% dec ease in high-densi y lipop o ein (HDL) choles e ol, unlike hose who did no ake i amin A [ 23 ]. In a ecen case epo , a 4-yea -old boy p esen ed se e al bone pains due o i amin A oxici y (600,000 IU e e y day o mo e han 3 mon hs) [ 2 ]. In ac , i has been epo ed ha excessi e i amin A in ake can accele a e bone loss and isk o hip ac u e, possibly due o i amin A-induced os eoclas s s imula ion, besides i inhibi s new bone o ma ion, inc easing os eopo osis isk [24]. Biomolecules 2018,8, 124 5 o 11 On he o he hand, i amin C can be me abolized o oxala e and migh inc ease kidney oxala e exc e ion. Se e al s udies sugges ha i amin C supplemen s may inc ease u ina y oxala e concen a ions, doubling he isk o calcium oxala e kidney s ones [ 25 – 27 ]. A s udy de ined ha high i amin C in ake om supplemen s is associa ed wi h a ise in ca dio ascula disease mo ali y in pos menopausal women wi h diabe es bu his has ne e been con i med [ 28 ]. Theo e ically, i amin C may cause oo much i on abso p ion bu his is likely o be signi ican only in pe sons who ha e high i on s o es o in pa ien s wi h i on o e load, such as he edi a y hemoch oma osis, whe e an inc easing i on oxici y isk may exis [ 29 ]. Pa lo ou [ 30 ] e alua ed ee oxygen adicals (FORT) and ee oxygen adicals de ense (FORD) le els in pa ien s wi h newly diagnosed ype 2 DM pa ien s. The au ho s ound ha FORT le els we e inc eased in diabe ic pa ien s compa ed o con ols; howe e , FORD le els we e lowe in diabe ic pa ien s compa ed o con ols. A s udy epo ed ha die a y i amin E supplemen a ion signi ican ly inc eases p os a e cance isk among heal hy men [ 31 ]. A me a-analysis ende s mo e e idence o i amin E ad e se e ec s on s oke sub ypes. Indeed, he s udy de ined a 22% inc eased hemo hagic s oke isk and a 10% dec eased ischemic s oke isk wi h i amin E supplemen a ion, al hough he absolu e e ec s a e mino [ 32 ]. S ill, a s udy unde lined ha 22–30 mg/day o i amin E in human p egnancy may be associa ed wi h bi h weigh dec ease [33]. Scien is s ha e epo ed ha β -ca o ene supplemen a ion (and no he in ake o ege ables ich in β -ca o ene) has ac ually inc eased he isk o dea h om lung cance o hea disease in smoke s, a he han educing cance incidence [ 34 – 37 ]. In ac , we belie e ha o he an ioxidan side e ec s may no be epo ed and o he ones will be disco e ed in he u u e. None heless, he s udies pe o med so a ha e se e ely impai ed he epu a ion o an ioxidan s in gene al. Howe e , in an ideal wo ld, we could go back o his s udy, h ough examining issues p e- and pos - ea men , and de e mine whe he in e en ions had a signaling e ec in he umo s. Based on p e ious s udies o ebselen, an o ganoselenium compound wi h b oad an ioxidan p ope ies [ 38 ], i is likely ha ROS educ ion in umo s, o ins ance in Bu ki s lymphoma, esul s in MAP kinase signaling ac i a ion and an inc eased umo g ow h a e [ 15 ]. In addi ion, mo e apidly g owing umo s may be mo e suscep ible o subsequen chemo he apy and adia ion. Thus, he cu en s a e o a ai s sugges s ha he mos common an ioxidan s a e, a bes , ine ec i e and po en ially ha m ul. 5. A e An ioxidan s Bene i s Mo e Appa en Than Real? Pa o he p oblem ega ding an ioxidan s is ying o impose an an ioxidan chemical de ini ion on a biochemical sys em. Ancien man used ca bon as an an ioxidan o educe i on o e o i on, h ough emo ing oxygen in i on o e wi h ca bon and d i ing o oxygen as ca bon dioxide. This de ini ion, while applicable o me al e ining, does no conside he complexi y o biological sys ems. Fo example, nuclea ac o e y h oid 2-like 2 (N 2) ansc ip ion is one o he mos s udied an ioxidan sys ems in biology, being no mally N 2 complexed wi h Kelch-like ECH-associa ed p o ein 1 (Keap 1). So, cells ea ed wi h N 2 induce s leads o N 2 nuclea ansloca ion [ 39 , 40 ]. Many N 2 induce s bind o N 2 h ough a Michael addi ion, oxidizing N 2 sul hyd yl g oups. Thus, many N 2 induce s a e oxidan s and, he e o e, an an ioxidan esponse is an oxidan esponse [ 41 – 43 ]. Fu he mo e, educed glu a hione, he main small an ioxidan molecule in mammalian cells, is a p oduc o se e al N 2downs eam a ge genes, coun e balancing mi ochond ial ROS p oduc ion [ 44 ]. None heless, he e is an open ques ion whe he he an ioxidan e ec o glu a hione syn hesis exe s an ioxidan ac i i y beyond co ec ing o iginal oxida i e s esso o ins ead exe s mul iple unc ions, beyond educing in acellula milieu main enance [ 44 ]. Finally, in glu a hione, ee sul hyd yl g oups (-SH g oups) could exe an ioxidan e ec s bu , a he same ime, bind o chemo he apy eac i e in e media es and block e ec i e umo cells killing [ 45 – 47 ]. This ac may explain why glu a hione induc ion/supplemen a ion has no been an as e ec i e s a egy agains cance as o iginally implied. Tumo s wi h mu an p53 demons a e high N 2 ac i a ion le els and his could p o ec umo cells agains eac i e adduc induced DNA damage [48]. Biomolecules 2018,8, 124 6 o 11 In a ecen s udy, N 2 was ound o be induced by oncogenic Ras and N 2 loss was ound o be p o ec i e in ansgenic models o ca cinogenesis. In his s udy, endogenous oncogenic Ras was ound o dec ease ROS h ough N 2induc ion, while exogenous oncogenic Ras o e exp ession inc eases ROS le els. Indeed, N 2 loss dec eased p eneoplas ic panc ea ic cells exp ession and ea u es o senescence we e obse ed in N 2 de icien umo cells [ 49 ]. Thus, hese s udies sugges ha ROS may play a physiologic ole in p e en ing p eneoplas ic cells p opaga ion wi h Ras d i e mu a ions (Figu e 2). Biomolecules 2018, 8, x FOR PEER REVIEW 6 o 11 ound o dec ease ROS h ough N 2induc ion, while exogenous oncogenic Ras o e exp ession inc eases ROS le els. Indeed, N 2 loss dec eased p eneoplas ic panc ea ic cells exp ession and ea u es o senescence we e obse ed in N 2 de icien umo cells [49]. Thus, hese s udies sugges ha ROS may play a physiologic ole in p e en ing p eneoplas ic cells p opaga ion wi h Ras d i e mu a ions (Figu e 2). Figu e 2. Po en ial ou comes o an ioxidan s a di e ing umo igenesis s ages. The p esence o a d i e mu a ion in a p ima y cell leads o eac i e oxygen-media ed endoplasmic e iculum (ER) s ess. No mally, his could lead o senescence, in which P16ink4a and FOXO4 senescence ma ke s a e ele a ed, as well as NF-κB ac i a ion. This leads o a pe sis en senescen pheno ype which canno e-en e he cell cycle. On he o he hand, ER s ess elie wi h an an ioxidan , o umo supp esso (i.e., p16ink4a) loss leads o clonal cells expansion wi h d i e mu a ions. This may lead o ca cinogenesis. In ad anced umo s, NF-κB migh be ac i a ed by ROS and a ROS blockade may lead o NF-κB ac i a ion dec ease and chemo he apy and adia ion sensi i i ies inc ease. T ea men wi h an ioxidan N-ace yl cys eine elie ed he block, which is consis en wi h: (a) Cells wi h oncogenic d i e mu a ions exhibi endoplasmic e iculum s ess, which migh be in pa media ed by high ROS le els, and he e o e high ROS le els elie may allow p eneoplas ic cells p oli e a ion and ankly malignan cells con e sion and (b) o he obse a ions ha an ioxidan s pa adoxically p omo e cance [9,12,48,50–52]. ROS ha e dis inc umo -p omo ing e ec s, which include DNA me hyl ans e ase 1 (DNMT) induc ion [53,54], oxida i ely umo supp esso s p53 inac i a ion, kB (IkB), phospha ase and ensin homolog (PTEN) inhibi ion [55]. This leads o p o umo igenic NF-κB and Ak ac i a ion and, he e o e, ea men wi h a supe oxide p oduc ion inhibi o (NADPH oxidase inhibi o ) leads o p53 eac i a ion, and NF-κB and Ak inac i a ion. This would be bene icial as NF-κB and Ak inhibi ion p imes o inc ease chemo he apy and adia ion suscep ibili y h ough he down egula ion o a ge genes, such as mul id ug esis ance p o ein 1 (Md 1) and DNA epai genes. On he o he hand, sul u supplemen a ion in e ms o inc easing educed glu a hione, o N 2 induc ion, could ha e dele e ious e ec s on chemo he apy and adia ion by p o ec ing umo DNA. As i has been exposed, supplemen al an ioxidan adminis a ion du ing chemo he apy and adia ion he apy is cu en ly con o e sial [9,10,12,56,57]. So a , i has been di icul o de e mine which an ioxidan s may exe a bene icial impac on cance ea men ou comes o which may con ibu e o ea men ad e se e ec s’ amelio a ion [9,57]. Fo his eason, du ing chemo he apy Figu e 2. Po en ial ou comes o an ioxidan s a di e ing umo igenesis s ages. The p esence o a d i e mu a ion in a p ima y cell leads o eac i e oxygen-media ed endoplasmic e iculum (ER) s ess. No mally, his could lead o senescence, in which P16ink4a and FOXO4 senescence ma ke s a e ele a ed, as well as NF- κ B ac i a ion. This leads o a pe sis en senescen pheno ype which canno e-en e he cell cycle. On he o he hand, ER s ess elie wi h an an ioxidan , o umo supp esso (i.e., p16ink4a) loss leads o clonal cells expansion wi h d i e mu a ions. This may lead o ca cinogenesis. In ad anced umo s, NF- κ B migh be ac i a ed by ROS and a ROS blockade may lead o NF- κ B ac i a ion dec ease and chemo he apy and adia ion sensi i i ies inc ease. T ea men wi h an ioxidan N-ace yl cys eine elie ed he block, which is consis en wi h: (a) Cells wi h oncogenic d i e mu a ions exhibi endoplasmic e iculum s ess, which migh be in pa media ed by high ROS le els, and he e o e high ROS le els elie may allow p eneoplas ic cells p oli e a ion and ankly malignan cells con e sion and (b) o he obse a ions ha an ioxidan s pa adoxically p omo e cance [9,12,48,50–52]. ROS ha e dis inc umo -p omo ing e ec s, which include DNA me hyl ans e ase 1 (DNMT) induc ion [ 53 , 54 ], oxida i ely umo supp esso s p53 inac i a ion, kB (IkB), phospha ase and ensin homolog (PTEN) inhibi ion [ 55 ]. This leads o p o umo igenic NF- κ B and Ak ac i a ion and, he e o e, ea men wi h a supe oxide p oduc ion inhibi o (NADPH oxidase inhibi o ) leads o p53 eac i a ion, and NF- κ B and Ak inac i a ion. This would be bene icial as NF- κ B and Ak inhibi ion p imes o inc ease chemo he apy and adia ion suscep ibili y h ough he down egula ion o a ge genes, such as mul id ug esis ance p o ein 1 (Md 1) and DNA epai genes. On he o he hand, sul u supplemen a ion in e ms o inc easing educed glu a hione, o N 2 induc ion, could ha e dele e ious e ec s on chemo he apy and adia ion by p o ec ing umo DNA. As i has been exposed, supplemen al an ioxidan adminis a ion du ing chemo he apy and adia ion he apy is cu en ly con o e sial [ 9 , 10 , 12 , 56 , 57 ]. So a , i has been di icul o de e mine Biomolecules 2018,8, 124 7 o 11 which an ioxidan s may exe a bene icial impac on cance ea men ou comes o which may con ibu e o ea men ad e se e ec s’ amelio a ion [ 9 , 57 ]. Fo his eason, du ing chemo he apy and adia ion he apy, an an ioxidan p esc ip ion is con using and consequen ly, his should conside he ype o cance , backg ound and s a e o pa ien , an i umo he apy, d ugs mechanism o ac ion and d ugs used in ea men , as also he an ioxidan ype and dosage [9,57]. 6. Redox Imbalance Posi i e and An ioxidan E ec Nega i e I is now well-es ablished ha eac i e species and a basal edox imbalance le el a e essen ial o cell su i al [58]. Concomi an ly, i is also well-known ha while se e e edox imbalance o en leads o widesp ead oxida i e and ni osa i e damage and cell dea h, a mode a e edox imbalance le el, induced by wide s esso s a ie y, can yield g ea bene icial e ec s on adap i e cellula esponses, such inc eased endogenous an ioxidan de ense sys ems le els [12,58]. Ho mesis was de ined as a cellula adap i e esponse o s esso s ha esul s in a biphasic dose- esponse ela ionship, such ha low-dose s imula ion esul s in a bene icial adap a ion, whe eas a high-dose esul s in a oxic e ec [ 59 ]. In his con ex , low ROS doses, p oduced du ing exe cise aining, a e equi ed o he exe cise-induced aining esponse in skele al muscle. Thus, ROS a e equi ed o exe cise adap i e esponse and a e essen ial o enhance spo s pe o mance [ 59 ]. In his sense, low cell s esso doses, such as chemicals, oxins, adia ion and mode a e exe cise esul s in an adap i e esponse, inc ease he an ioxidan capaci y o cells. In ac , exe cise i sel can be conside ed an an ioxidan , since aining inc eases classical an ioxidan enzymes exp ession, like SOD and GPX, while, in gene al, an ioxidan supplemen s may no be a good s a egy when aining because hey elimina e ROS p oduc ion ha ac s o s imula e endogenous an ioxidan enzymes [ 60 ]. Mo eo e , mi ochond iogenesis is egula ed by many edox-sensi i e enzymes, in ol ing MAP kinases, NF- κ B, p53, hea shock ac o , pe oxisome p oli e a o -ac i a ed ecep o gamma coac i a o 1-alpha (PGC-1 α ), and o he s in ol ed in modula ing muscle adap a ion o muscle [60]. On he o he hand, i is belie ed ha an ioxidan s can p e en cance de elopmen a ec ing he cell cycle, in lamma ion, umo p oli e a ion and in asi eness, apop osis and de ox mechanisms. So, he an i umo e ec s o se e al an ioxidan s (i.e., ca echins, iso la ones, lignans, la anones, es e a ol, ellagic acid, que ce in, and cu cumin) ha e been ex ensi ely s udied [ 9 ]. Howe e , an ioxidan supplemen a ion may block endogenous an ioxidan s aise and o he cell adap a ion mechanisms, such as be e ene ge ic me abolism. Indeed, a basal edox imbalance le el is c ucial o cell adap a ion. The ques ion is wha eac i e species concen a ion ange is bene icial and wha is ha m ul? In his con ex , he co ela ion be ween ciga e e smoke and lung-cance is well-es ablished bu will he e be a bene icial low-dose? (Ha dly an e hics commi ee would app o e his kind o expe imen al s udy). 7. Wha An ioxidan s Can Do o Imp o e Heal h o Wha They Canno Do? Despi e billions o dolla s spen on an ioxidan supplemen s yea ly, he modi ica ion o a ch onic disease cou se emains elusi e. Nume ous agen s ha e demons a ed chemop e en i e e ec s in mu ine models bu hese ha e no ye been ansla ed o human diseases. Cu en ly, we ha e enough clinical e idence ha an ioxidan s a ailable as supplemen s ha e a bes li le alue in p e en ing o modi ying a ch onic disease cou se. Inhe en di e ences in oden s e sus human’s biology, high-dose ea men in oden s and non-deli e y o human issues a e p obably behind he cu en ailu e eco d. Howe e , he e is oom o op imism bu i can only come om a sophis ica ed unde s anding o ch onic disease biology. In he case o cance , cu en ly a ailable an ioxidan s could be ha m ul in p e en ing physiologic senescence in pa ien s who ha e d i e mu a ions lesions p edominance. Physiologic senescence p e en ion could expand mu a ed cells popula ion ha can unde go ca cinogenesis. Howe e , in ad anced umo s, ROS down egula ion agen s inducing NF- κ B could be use ul in chemo he apy and adia ion combina ion o igge umo cell apop osis. Biomolecules 2018,8, 124 8 o 11 The e is nuance in his as well, as an agen ha inc eases sul hyd yl’s, ei he h ough exogenous sul hyd yl supply o N 2 induc ion, could p o ec umo DNA and inac i a e chemo he apeu ic agen s. Wha is needed is a “sma an ioxidan ”, i.e., one ha could cause a edox imbalance in umo cells bu no in no mal issues. A po en ial ca ego y o compounds is si uin 3 ac i a o s. Indeed, si uin 3 is a majo mi ochond ial NAD + -dependen deace ylase ha plays a c i ical ole in mi ochond ial p o eins ac i a ion, is in ol ed in ene gy me abolism, and changes in i s exp ession a e associa ed wi h excessi e ROS p oduc ion [ 61 ]. The e is e idence ha some polyphenolic compounds ound in na u e could ac as “sma an ioxidan ” ia si uin 3 ac i a ion [ 62 – 64 ]. Such an agen could po en ially dec ease NF- κ B a he same ime as no inducing N 2. This would ip he balance and could be use ul in bo h oxida i e p ecu so lesions des uc ion wi h d i e mu a ions as well as sensi izing ad anced cance o chemo he apy and adia ion. 8. Conclusions and Upcoming Pe spec i es In conclusion, o minimize ch onic edox imbalance damages, i is bes o ollow a balanced and a ied die , including in i s composi ion many g ains, legumes, ui s and ege ables o di e en colo s. In addi ion, heal hy li es yle habi s should be included, such as exe cise aining on a egula basis o a oid obesi y, no smoking and educing alcoholic be e ages in ake. The in ake o an ioxidan supplemen s would only make sense in a case o de ici s, ying o no malize hei le els, bu no as a usual in ake. In addi ion, an ioxidan s he apeu ic use ulness agains cance s ill has many open on s ha should be in es iga ed in he u u e. Au ho Con ibu ions: Au ho s B.S., M.M., J.L.A., A.S., N.M., P.K.M., M.S.-R., P.K. and J.S.-R. con ibu ed equally in he w i ing o his wo k. J.L.A., N.M., P.K., and J.S.-R. c i ically e iewed he manusc ip . All he au ho s ead and app o ed he inal manusc ip . Funding: The APC was unded by N.M. Acknowledgmen s: An oni Su eda acknowledges he suppo o Ins i u e o Heal h Ca los III (P ojec CIBEROBN CB12/03/30038). Na ália Ma ins hank o Po uguese Founda ion o Science and Technology (FCT–Po ugal) o he S a egic p ojec e . UID/BIM/04293/2013 and “NORTE2020-P og ama Ope acional Regional do No e” (NORTE-01-0145-FEDER-000012). Con lic s o In e es : The au ho s decla e no con lic o in e es . Re e ences 1. Jones, D.P. Rede ining oxida i e s ess. An ioxid. Redox Signal. 2006,8, 1865–1879. [C ossRe ] [PubMed] 2. Baineni, R.; Gula i, R.; Delhi, C.K. Vi amin A oxici y p esen ing as bone pain. A ch. Dis. Child. 2016 ,102, 556–558. [C ossRe ] [PubMed] 3. Pawa , R.K.; Bhagu e, G.R.; Cha an, R.P. An ioxidan s and hei ole in nu u e human li e and indus y: A e iew. In . J. Chem. S ud. 2016,4, 22–26. 4. Maes, M.; Boni acio, K.L.; Mo elli, N.R.; Va gas, H.O.; Ba bosa, D.S.; Ca alho, A.F.; Nunes, S.O.V. Majo di e ences in neu ooxida i e and neu oni osa i e s ess pa hways be ween majo dep essi e diso de and ypes i and ii bipola diso de . Mol. Neu obiol. 2018,55, 1–16. [C ossRe ] [PubMed] 5. Shi ley, R.; O d, E.N.J.; Wo k, L.M. Oxida i e S ess and he Use o An ioxidan s in S oke. An ioxidan s 2014 , 3, 472–501. [C ossRe ] [PubMed] 6. Kuma , P.; Chand, S.; Mau ya, P.K. Que ce in-modula ed e y h ocy e memb ane sodium-hyd ogen exchange du ing human aging: Co ela ion wi h a pase’s. A ch. Physiol. Biochem. 2016 ,122, 141–147. [C ossRe ] [PubMed] 7. Pa ice, T.; Rozec, B.; Sido o , A.; Blanloeil, Y.; Despins, P.; Pe igaud, C. In luence o i amins on seconda y eac i e oxygenspecies p oduc ion in se a o pa ien s wi h esec able NSCLC. Diseases 2016 ,4, 25. [C ossRe ] [PubMed] 8. Halliwell, B.; Gu e idge, J.M. The an ioxidan s o human ex acellula luids. A ch. Biochem. Biophys. 1990 , 280, 1–8. [C ossRe ] Biomolecules 2018,8, 124 9 o 11 9. Mu -Salud, N.; Ál a ez, P.J.; Ga ido, J.M.; Ca asco, E.; A ánega, A.; Rod íguez-Se ano, F. An ioxidan in ake and an i umo he apy: Towa d nu i ional ecommenda ions o op imal esul s. Oxid. Med. Cell. Longe . 2016,2016, 6719534. [C ossRe ] [PubMed] 10. Lawenda, B.D.; Kelly, K.M.; Ladas, E.J.; Saga , S.M.; Vicke s, A.; Blumbe g, J.B. Should supplemen al an ioxidan adminis a ion be a oided du ing chemo he apy and adia ion he apy? J. Na l. Cance Ins . 2008,100, 773–783. [C ossRe ] [PubMed] 11. Hekimi, S.; Lapoin e, J.; Wen, Y. Taking a “good” look a ee adicals in he aging p ocess. T ends Cell Biol. 2011,21, 569–576. [C ossRe ] [PubMed] 12. Halliwell, B. The an ioxidan pa adox: Less pa adoxical now? B . J. Clin. Pha macol. 2013 ,75, 637–644. [C ossRe ] [PubMed] 13. La osa, M.; González-Sa ías, A.; Ga cía-Conesa, M.T.; Tomás-Ba be án, F.A.; Espín, J.C. U oli hins, ellagic acid-de i ed me aboli es p oduced by human colonic mic o lo a, exhibi es ogenic and an ies ogenic ac i i ies. J. Ag ic. Food Chem. 2006,54, 1611–1620. [C ossRe ] [PubMed] 14. González-Sa ías, A.; Giménez-Bas ida, J.A.; Ga cía-Conesa, M.T.; Gómez-Sánchez, M.B.; Ga cía-Tala e a, N.V.; Gil-Izquie do, A.; Sánchez-Ál a ez, C.; Fon ana-Compiano, L.O.; Mo ga-Egea, J.P.; Pas o -Qui an e, F.A. Occu ence o u oli hins, gu mic obio a ellagic acid me aboli es and p oli e a ion ma ke s exp ession esponse in he human p os a e gland upon consump ion o walnu s and pomeg ana e juice. Mol. Nu . Food Res. 2010,54, 311–322. [C ossRe ] [PubMed] 15. Ce imele, F.; Ba le, T.; Lynch, R.; F ank, D.A.; Mu ad, E.; Cohen, C.; Maca on, N.; Sixbey, J.; Smi h, K.; Wa nick, R.S. Reac i e oxygen signaling and MAPK ac i a ion dis inguish eps ein–ba i us (EBV)-posi i e e sus EBV-nega i e bu ki ’s lymphoma. P oc. Na l. Acad. Sci. USA 2005 ,102, 175–179. [C ossRe ] [PubMed] 16. Cze nichow, S.; Ve gnaud, A.C.; Galan, P.; A naud, J.; Fa ie , A.; Fau e, H.; Huxley, R.; He cbe g, S.; Ahluwalia, N. E ec s o long- e m an ioxidan supplemen a ion and associa ion o se um an ioxidan concen a ions wi h isk o me abolic synd ome in adul s. Am. J. Clin. Nu . 2009 ,90, 329–335. [C ossRe ] [PubMed] 17. Pa k, Y.; Spiegelman, D.; Hun e , D.J.; Albanes, D.; Be gk is , L.; Bu ing, J.E.; F eudenheim, J.L.; Gio annucci, E.; Goldbohm, R.A.; Ha nack, L.; e al. In akes o i amins A, C, and E and use o mul iple i amin supplemen s and isk o colon cance : A pooled analysis o p ospec i e coho s udies. Cance Causes Con ol 2010,21, 745–757. [C ossRe ] [PubMed] 18. He cbe g, S.; Galan, P.; P eziosi, P.; Be ais, S.; Mennen, L.; Mal y, D.; Roussel, A.M.; Fa ie , A.; B iançon, S. The SU.VI.MAX S udy: A andomized, placebo-con olled ial o he heal h e ec s o an ioxidan i amins and mine als. A ch. In e n. Med. 2004,164, 2335–2342. [C ossRe ] [PubMed] 19. Ro hman, K.J.; Moo e, L.L.; Singe , M.R.; Nguyen, U.-S.D.; Mannino, S.; Milunsky, A. Te a ogenici y o high i amin A in ake. N. Engl. J. Med. 1995,333, 1369–1373. [C ossRe ] [PubMed] 20. S a g o e, M.B.; T easu e, J.; McKee, D.L. He b, Nu ien , and D ug In e ac ions: Clinical Implica ions and The apeu ic S a egies; Else ie Heal h Sciences: Ams e dam, The Ne he lands, 2008. 21. Biesalski, H.K. Compa a i e assessmen o he oxicology o i amin A and e inoids in man. Toxicology 1989 , 57, 117–161. [C ossRe ] 22. Azaïs-B aesco, V.; Pascal, G. Vi amin A in p egnancy: Requi emen s and sa e y limi s. Am. J. Clin. Nu . 2000,71, 1325S–1333S. [C ossRe ] [PubMed] 23. Ca mel, B.; Moon, T.E.; Le ine, N. E ec s o long- e m in ake o e inol on selec ed clinical and labo a o y indexes. Am. J. Clin. Nu . 1999,69, 937–943. [C ossRe ] [PubMed] 24. Binkley, N.; K uege , D. Hype i aminosis a and bone. Nu . Re . 2000,58, 138–144. [C ossRe ] [PubMed] 25. Baxmann, A.C.; De OG Mendonca, C.; Heilbe g, I.P. E ec o i amin C supplemen s on u ina y oxala e and ph in calcium s one- o ming pa ien s. Kidney In . 2003,63, 1066–1071. [C ossRe ] [PubMed] 26. Moyad, M.; Lee, J. The Supplemen Handbook: A T us ed Expe ’s Guide o Wha Wo ks & Wha ’s Wo hless o Mo e Than 100 Condi ions; Rodale Books: Emmaus, PA, USA, 2014. 27. Clayman, R.V. E ec o i amin C supplemen s on u ina y oxala e and pH in calcium s one- o ming pa ien s. J. U ol. 2005,173, 478–479. [C ossRe ] 28. Lee, D.-H.; Folsom, A.R.; Ha nack, L.; Halliwell, B.; Jacobs, D.R., J . Does supplemen al i amin C inc ease ca dio ascula disease isk in women wi h diabe es? Am. J. Clin. Nu . 2004 ,80, 1194–1200. [C ossRe ] [PubMed]