MARK-AGE biomarkers of ageing

Mechanisms o Ageing and De elopmen 151 (2015) 2–12 Con en s lis s a ailable a ScienceDi ec Mechanisms o Ageing and De elopmen jou nal homepage:www.else ie .com/loca e/mechagede MARK-AGE bioma ke s o ageing Alexande Bü klea,∗, Ma ía Mo eno-Villanue aa, Jü gen Be nha db, Ma ía Blascoc, Ge ben Zondagd, Jan H.J. Hoeijmake se, Oli ie Toussain , Bea ix G ubeck-Loebens eing, Eugenio Mocchegianih, Sebas iano Collinoi, E s a hios S. Gonosj, Ewa Siko ak, Daniela G adina ul, Ma ijn Dollém, Michel Salmonn, Pe e K is enseno, Helen R. G i fi hsp, Claude Libe q, Tilman G une ,s, Nicolle B eusing , And eas Simm , Claudio F anceschiu, Mi iam Cap iu, Duncan Talbo , Paola Caia aw, Be and F igue x, P. Eline Slagboomy, An i He onenz, Mikko Hu mez, Richa d AspinallA aMolecula Toxicology G oup, Depa men o Biology, Box 628, Uni e si y o Kons anz, 78457 Kons anz, Ge many bBioTeSys GmbH, 73728 Esslingen, Ge many cSpanish Na ional Cance Resea ch Cen e (CNIO), 3 Melcho Fe nandez Almag o, 28029 Mad id, Spain dDNage BV1, Leiden, The Ne he lands eDepa men o Gene ics, E asmus Uni e si y Medical Cen e , P.O. Box 1738, 3000 DR Ro e dam, The Ne he lands Uni e si y o Namu , Resea ch Uni on Cellula Biology, Rue de B uxelles, 61, Namu B-5000, Belgium gResea ch Ins i u e o Biomedical Aging Resea ch, Uni e si y o Innsb uck, Rennweg, 10, 6020 Innsb uck, Aus ia hT ansla ional Resea ch Cen e o Nu i ion and Ageing, IRCCS-INRCA, Via Bi a elli 8, 60121 Ancona, I aly iNes lé Ins i u e o Heal h Sciences SA, Molecula Bioma ke s, EPFL Inno a ion Pa k, 1015 Lausanne, Swi ze land jNa ional Hellenic Resea ch Founda ion, Ins i u e o Biology, Medicinal Chemis y and Bio echnology, A hens, G eece kLabo a o y o he Molecula Bases o Ageing, Nencki Ins i u e o Expe imen al Biology, Polish Academy o Sciences, 3 Pas eu s ee , 02-093 Wa saw, Poland lAna Aslan – Na ional Ins i u e o Ge on ology and Ge ia ics, Bucha es , Romania mNa ional Ins i u e o Public Heal h and he En i onmen (RIVM), Cen e o P e en ion and Heal h Se ices Resea ch, P.O. Box 1, 3720 BA Bil ho en, The Ne he lands nS a icell, Science Pa k C ealys, Rue Jean Sone 10, 5032 Les Isnes, Belgium oDepa men o Enginee ing – BCE P o ein Enginee ing, Gus a Wieds ej 10, 8000 Aa hus, Denma k pLi e and Heal h Sciences, As on Resea ch Cen e o Heal hy Ageing, As on Uni e si y, Bi mingham, UK qDepa men o Molecula Biomedical Resea ch, VIB, Ghen , Belgium Ins i u e o Nu i ional Medicine, Uni e si y o Hohenheim, 70593 S u ga , Ge many sDepa men o Nu i ional Toxicology, F ied ich Schille Uni e si y Jena, Do nbu ge S . 24, 07743 Jena, Ge many Depa men o Ca dio ho acic Su ge y, Uni e si y Hospi al Halle, E ns -G ube S . 40, 06120 Halle (Saale), Ge many uCIG-In e depa men al Cen e “L.Gal ani”, Alma Ma e S udio um, Uni e si y o Bologna, 40126 Bologna, I aly Unile e Co po a e Resea ch, Sha nb ook, UK wDepa men o Cellula Bio echnologies and Hema ology, Facul y o Pha macy and Medicine, “Sapienza” Uni e si y Rome, V.le Regina Elena 324, 00161 Rome, I aly xSo bonne Uni e si és, UPMC Uni Pa is 06, UMR UPMC CNRS 8256, Biological adap a ion and ageing – IBPS, INSERM U1164, F-75005 Pa is, F ance yDepa men o Molecula Epidemiology, Leiden Uni e si y Medical Cen e, Leiden, The Ne he lands zMedical School, Uni e si y o Tampe e, 33014 Tampe e, Finland ARegene a i e Medicine G oup, C anfield Heal h, C anfield, UK a i c l e i n o A icle his o y: Recei ed 2 Decembe 2014 Recei ed in e ised o m 19 Ma ch 2015 Accep ed 21 Ma ch 2015 A ailable online 24 Ma ch 2015 Keywo ds: Ageing bioma ke s Human s udies MARK-AGE a b s a c Many candida e bioma ke s o human ageing ha e been p oposed in he scien ific li e a u e bu in all cases hei a iabili y in c oss-sec ional s udies is conside able, and he e o e no single measu emen has p o en o se e a use ul ma ke o de e mine, on i s own, biological age. A plausible eason o his is he in insic mul i-causal and mul i-sys em na u e o he ageing p ocess. The ecen ly comple ed MARK-AGE s udy was a la ge-scale in eg a ed p ojec suppo ed by he Eu opean Commission. The majo aim o his p ojec was o conduc a popula ion s udy comp ising abou 3200 subjec s in o de o iden i y a se o bioma ke s o ageing which, as a combina ion o pa ame e s wi h app op ia e weigh ing, would measu e biological age be e han any ma ke in isola ion. © 2015 The Au ho s. Published by Else ie I eland L d. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). ∗Co esponding au ho . Tel.: +49 7531 884035; ax: +49 7531 884033. E-mail add ess: [email p o ec ed] (A. Bü kle). h p://dx.doi.o g/10.1016/j.mad.2015.03.006 0047-6374/© 2015 The Au ho s. Published by Else ie I eland L d. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/ licenses/by-nc-nd/4.0/). A. Bü kle e al. / Mechanisms o Ageing and De elopmen 151 (2015) 2–12 3 1. In oduc ion Ageing has been defined as he ime-dependen decline o unc- ional capaci y and s ess esis ance, associa ed wi h inc eased isk o mo bidi y and mo ali y. Ageing is a p ocess ha a ec s mos i no all issues and o gans o he body. Mo eo e , c oss- alk can occu be ween mul iple physiological sys ems, e.g. me abolic sys ems may influence he ageing o he immune sys em. The mechanisms unde lying he ageing p ocess a e beginning o be un a elled a he molecula le el (López-O ín e al., 2013), ye he e is clea e idence ha he a e o ageing di e s significan ly be ween membe s o he same animal species, including humans. In o he wo ds, he “biological age” may di e om he ch onological age. The classical quan i a i e assessmen o “ he a e o ageing” elies on he analysis o mo ali y cu es (Gompe z unc ion) o popula ions. In o he wo ds, indi iduals ha e o be ollowed up un il he end o hei li es in o de o de e mine hei “biologi- cal age” a any ime poin du ing li e. The e o e, a he le el o a li ing indi idual, a eliable o assessmen o he s a e o ageing, i.e. he s a e o he abo e-men ioned unc ional decline, and a p edic- ion o he isk o he onse o mo bidi y and he esidual indi idual li e expec ancy a e no possible wi h his me hod. One s a egy o sol e his p oblem is he iden ifica ion o (an) age- ela ed change(s) in body unc ion o composi ion ha could se e as a measu e o “biological” age and p edic he u u e onse o age- ela ed diseases and/o esidual li e ime mo e accu a ely han ch onological age. Such pa ame e s a e e med “bioma ke s o age- ing” (Bake and Sp o , 1988). This e m has been coined in analogy o bioma ke s o specific ch onic diseases, such as HIV in ec ion, o bioma ke s o exposu e o oxins. The Ame ican Fede a ion o Aging Resea ch has p oposed he ollowing c i e ia o a bioma ke o ageing: 1. I mus p edic he a e o ageing. In o he wo ds, i would ell exac ly whe e a pe son is in hei o al li e span. I mus be a be e p edic o o li e span han ch onological age. 2. I mus moni o a basic p ocess ha unde lies he ageing p ocess, no he e ec s o disease. 3. I mus be able o be es ed epea edly wi hou ha ming he pe son, o example, a blood es o an imaging echnique. 4. I mus be some hing ha wo ks in humans and in labo a o y animals, such as mice. This is so ha i can be es ed in lab animals be o e being alida ed in humans. The ou h o he abo e c i e ia may, howe e , be ques ioned as he e a e ce ainly some pa ame e s whose impo ance o he ageing p ocess may di e be ween mammalian species. One exam- ple would be elome e sho ening in humans and in labo a o y mice: While in human soma ic issues elome e sho ening can eadily be de ec ed, his is no he case in wild- ype labo a o y mouse s ains owing o hei much g ea e o e all leng h o elom- e es. The e o e elimina ing some candida e pa ame e s jus based on hei lack o ele ance in some model sys ems may lead o an exclusion o pa ame e s ha a e po en ially in e es ing o he human sys em. I should be no ed ha many candida e bioma ke s o human ageing ha e been p oposed in he scien ific li e a u e bu in all cases hei a iabili y in c oss-sec ional s udies is conside able, and he e o e no single measu emen has p o en o se e a use ul ma ke o de e mine, on i s own, biological age. A plausible eason o his is he in insic mul i-causal (Holliday, 2006) and mul i- sys em na u e o he ageing p ocess. MARK-AGE was a la ge-scale in eg a ed p ojec suppo ed by he Eu opean Commission. The majo aim o his p ojec was o conduc a popula ion s udy com- p ising abou 3200 subjec s in o de o iden i y a se o bioma ke s o ageing which, as a combina ion o pa ame e s wi h app op ia e weigh ing, would measu e biological age be e han any ma ke in isola ion. 2. MARK-AGE conso ium In o de o ackle he scien ific p oblem o es ablishing powe - ul bioma ke s o human ageing, he MARK-AGE conso ium, which consis ed o 26 Pa ne s comp ising 21 non-p ofi o ganisa ions (uni e si ies and public esea ch ins i u es), 3 small and medium sized en e p ises (SMEs), and 2 la ge companies, was o med. The scien ific g oups in ol ed a e a he o e on in he field o ageing esea ch, and some Pa ne s a e in e na ional leade s e en in wide fields such as Gene ics. The MARK-AGE conso ium was cha ac- e ised by a high deg ee o in e disciplina i y: The a ay o expe ise anged om Ge ia ics, Epidemiology and Human Gene ics o Clin- ical Chemis y, Biochemis y, Cell Biology, Immunology, Molecula Gene ics, Bioin o ma ics and Ma hema ical Modelling. Such a le el o in e disciplina i y is essen ial o he success o a p ojec o his la ge scale. The lead esea che s a e he au ho s on his documen . 3. The MARK-AGE s a egy In he La ge-Scale In eg a ing P ojec MARK-AGE, he Pa ne s p oposed o pe o m a comp ehensi e and cohe en Eu ope- wide popula ion s udy aiming a he iden ifica ion o powe ul bioma ke s o human ageing ac oss a ange o physiological Fig. 1. Schema ic ep esen a ion o he managemen s uc u e o he MARK-AGE p ojec . 4 A. Bü kle e al. / Mechanisms o Ageing and De elopmen 151 (2015) 2–12 Table 1 MARK-AGE wo k packages. Wo k package numbe Wo k package i le 1 Rec ui men o p obands and physiological ma ke s 2 DNA-based ma ke s 3 Ma ke s based on p o eins and hei modifica ions 4 Immunological ma ke s 5 Clinical chemis y, ho mones and ma ke s o me abolism 6 Oxida i e s ess ma ke s 7 Eme gen bioma ke s o ageing om model sys ems and no el me hodological app oaches 8 Da a analysis and bioin o ma ics 9 Dissemina ion and aining 10 P ojec managemen and e hical issues sys ems. The s udy popula ion comp ised o abou 3200 subjec s and ep esen ed se e al di e en geog aphical egions o Eu ope. The s udy popula ion co e ed he age ange o 35–74 yea s, as his is he ime span du ing which p ophylaxis / in e en ion o coun e age- ela ed diseases may be possible and p omising. A wide ange o candida e bioma ke s was es ed, including (1) “classical” ones o which da a om se e al smalle s udies ha e been published; (2) “new” ones, based on p elimina y da a ob ained in small-scale s udies, as well as (3) “no el” ones, based on ecen esea ch on mechanis ic aspec s o ageing, conduc ed by p ojec Pa ne s. I is easonable o assume ha a combina ion o se e al bioma ke s will p o ide a much be e ool o measu e biologi- cal age han any single bioma ke in isola ion. I has o be aken in o accoun , hough, ha no all bioma ke s a e o equal weigh . The e o e a e aging all possible candida e bioma ke s may no app op ia e. A majo ask o his p ojec was, he e o e, o op i- mise he weigh ing o he di e en ma ke s, by using mul i- a ia e analysis, wi h he aim o educing a iance and o de i e a ma he- ma ical o mula ha will yield a “biological age sco e”. I should be men ioned ha wo k pe o med in he con ex o he “MacA hu s udies o success ul aging” on a coho o 171 adul s aged 70–79 had al eady p o ided p oo -o -concep by showing ha an “allo- s a ic load sco e”, inco po a ing 10 biological ma ke s, can be p edic i e o mo ali y isk (Seeman e al., 1995). The MARK-AGE p ojec p o ided a sys ema ic app oach: The Conso ium es ablished S anda d Ope a ing P ocedu es o he ec ui men o subjec s and p ocessing o samples (see Mo eno- Villanue a and Cap i e al., his issue), as well as quali y con ol measu es (Jansen e al., his issue). I was deemed essen ial o ec ui a new popula ion o subjec s, since p e ious ec ui men e o s pe o med in many Eu opean coun ies nei he ha e co - e ed he age ange o in e es no ha e hey p o ided he biological ma e ials o be s udied, including c yop ese ed blood cells. The ac i i ies wi hin he MARK-AGE p ojec we e dis ibu ed in Wo k Packages (Table 1) 3.1. WP1: ec ui men o subjec s and assessmen o physiological ma ke s Two la ge g oups o subjec s we e ec ui ed. A e exclusion o 138 hepa i is posi i e subjec s, he fi s g oup comp ised o 2262 andomly ec ui ed age-s a ified indi iduals om he gen- e al popula ion (RASIG) om se e al di e en geog aphical egions o Eu ope. Equal numbe s o men and women we e en olled, com- p ising simila numbe s o indi iduals in he ollowing age classes: 35–39 y s, 40–44 y s, 45–49 y s, 50–54 y s, 55–59 y s, 60–64 y s, 65–69 y s, 70–74 y s. This g oup was hough o be display he “a e age popula ion ageing a e”. The second g oup comp ised o subjec s bo n om a long-li ing pa en belonging o a amily wi h long-li ing sibling(s), such as he “90+ sibpai s” ec ui ed wi hin he amewo k o he EU In eg a ed P ojec GEHA, and hence o h designa ed GEHA o sp ing (GO) (528 subjec s). GO co e he age ange o 55–74 yea s. Acco ding o da a om he ecen li e a u e, indica ing ha o sp ing o long-li ing pa en s age in a “be e ” way han con ols bo n om non long- li ing pa en s, GO a e p edic ed o age a a slowe a e han he a e age popula ion. Wi hin he MARK-AGE p ojec , he GO subjec s we e he e o e compa ed wi h hei spouses, hence o h designa ed spouses o GEHA o sp ing (SGO) (305 subjec s). Sys ema ic compa ison o GO and SGO coho s should p o ide a unique oppo uni y o a fi s al- ida ion o he bioma ke s iden ified in he c oss-sec ional s udy o he RASIG subjec s. I is expec ed ha GO display a lowe biological age han SGO. The p ojec also ook ad an age o he ac ha some ela i ely a e ‘segmen al p oge oid synd omes’ p esen cha ac e is ics o d ama ically accele a ed ageing and p ema u e dea h om ypi- cal ageing-associa ed diseases. This is he case o subjec s a ec ed by Down’s synd ome (DS) o We ne ’s synd ome (WS). Due o he (ex eme) a i y o hese synd omes, only a small numbe o DS subjec s we e ec ui ed, while biological ma e ial (se um, plasma, u ine and blood) om WS pa ien s was s o ed in he MARK-AGE biobank. The ageing p ocess o DS and WS subjec s is being com- pa ed wi h RASIG and GO/SGO subjec s. I is expec ed ha hei bioma ke s indica e a highe biological age, and so his compa ison is expec ed o p o ide an addi ional alida ion o he bioma ke s iden ified in he c oss-sec ional analysis o RASIG. In o de o asce ain he biological and analy ical obus ness o he measu emen s o candida e bioma ke s, 97 dono s om he whole s udy popula ion ha e been e-sampled wi hin 3–6 mon hs. In such a sho ime pe iod, no significan change in he biological age s a us o he subjec s is expec ed; he e o e an ideal bioma ke essen ially should yield he same esul s. Finally, a limi ed andom sample o subjec s was ollowed-up in a small longi udinal s udy, compa ible wi h he ime and budge a y cons ain s o he p ojec . We e- es ed 12% o he ec ui ed sub- jec s RASIG, GO, SGO (389 subjec s in o al) a e 3 yea s. I was expec ed ha hose subjec s whose bioma ke p ofile indica ed an ad anced biological age a baseline should display a simila o accele a ed pa e n a he 3-yea ollow-up, and ice e sa o he biologically younge indi iduals. F om all subjec s en olled, an h opome ic, clinical and demo- g aphic da a ha e been collec ed in a s anda dised ashion. Upon w i en in o med consen , he ollowing se o in o ma ion was ob ained by using a s anda dised ques ionnai e: •Demog aphic in o ma ion: amily composi ion, ma i al s a us, educa ion, occupa ion, and housing condi ions. •Li es yle: use o obacco and alcohol, daily ac i i ies. •Func ional s a us: Ac i i ies o Daily Li ing (ADL) and No on Scale. •Cogni i e s a us: STROOP es , 15-pic u e lea ning es . •Heal h s a us: p esen and pas diseases, sel -pe cei ed heal h, numbe and ype o p esc ibed d ugs. •Mood: ZUNG dep ession scale. A physical examina ion o all subjec s comp ised measu emen o he ollowing “classical” candida e bioma ke s: •Body mass index. •Wais and hip ci cum e ence. •Blood p essu e a es . •Hea a e a es . A. Bü kle e al. / Mechanisms o Ageing and De elopmen 151 (2015) 2–12 5 •Lung unc ion – o ced expi a o y olume in 1 s (FEV1). •Lung unc ion – o ced i al capaci y (FVC). •Fi e- imes chai s anding. •Handg ip s eng h. All subjec s we e asked o dona e blood (55 ml) by enipunc u e a e o e nigh as ing. The blood sample was p ocessed o ob ain plasma, se um and pe iphe al blood mononuclea cells (PBMC). PBMC we e c yop ese ed and all he o he componen s we e ozen down. Buccal mucosal cells we e also collec ed (using a ki ) as well as spo u ine samples (see Mo eno-Villanue a and Cap i e al., his issue). 3.2. WP 2: DNA-based ma ke s The in eg i y o he nuclea genome and he epigenome is o i al impo ance o he p ope unc ion o cells, issues and o gans. The e is, howe e , a cons an a ack by exogenous and endogenous compounds and agen s (including eac i e oxygen species [ROS]) ha can damage DNA and/o dis u b epigene ic egula ion. Possible consequences a e mu a ion and dys egula ion o gene exp ession, which ei he can lead o cell dea h o cellula senescence o o malignan ans o ma ion o he cells ul ima ely esul ing in can- ce . P o ec ion and main enance sys ems ha e e ol ed ha help main ain a sus ainable s eady-s a e le el o molecula damage, and hese include DNA epai sys ems and DNA me hyl ans e ases. Fu he mo e elome ic DNA unde goes a i ion wi h each eplica- ion cycle and also as a consequence o DNA damage. A c i ical loss o elome e epea sequences has been shown o p e en u he cell p oli e a ion and in some cell ypes induces cellula senescence. Mi ochond ial DNA is an especially ulne able a ge o mu a- genesis, in iew o he high local le els o endogenous ROS in mi ochond ia. Damage and mu a ion o mi ochond ial DNA is iewed as a majo mechanism d i ing he ageing p ocess (Niemi e al., 2003; Wong e al., 2009; Al ilia e al., 2012). Ne e he- less D-loop egion con ains le el o he e oplasmy associa ed wi h longe i y, as p e iously iden ified (Rose e al., 2007), sugges ing also m DNA a ian s- based mechanisms o p o ec ion (Raule e al., 2014). Fu he , APOE geno ype, which is conside ed a gold s an- da d o he gene ics o longe i y and was ecen ly e-confi med (Deelen e al., 2014), was aken in o accoun wi h he basic idea o iden i y possible subg oups o indi iduals wi h bes o wo s heal h condi ions (ex eme pheno ypes). Ou o e a ching hypo hesis was ha he p esence o p ofi- cien sys ems o p e en / epai damage and mu a ion (Beneke and Bu kle, 2007) o he nuclea genome (Reale e al., 2005; Caia a and Zampie i, 2005; Za do e al., 2002), including elome e sho ening (Canela e al., 2007; Bene i e al., 2007; Flo es e al., 2005; Gonzalo e al., 2006), and o he mi ochond ial genome (Bellizzi e al., 2006; De Benedic is e al., 1999) should help e a d he ageing p ocess in many i no all issues. The e o e hese cellula unc ions ha e a po en ial o se e as bioma ke s o ageing. The ollowing esea ch asks ha e specifically been add essed: •We ha e s udied he main enance o he epigenome by analysing gene exp ession pa e ns in PBMC and cy osine me hyla ion s a- us. DNA me hyla ion was o be co ela ed wi h he possible age-dependen exp ession le el o some genes whose exp ession has been associa ed wi h ageing o longe i y. •The ageing-dependen decline o DNA epai was e alua ed by unc ional analysis o he epai o DNA s and b eaks induced by X- ays and in s udies on cellula poly(ADP- ibosyl)a ion capaci y and PARP-1 exp ession le els. •A en ion was also di ec ed owa ds elome e leng h, which is being co ela ed wi h modifica ions o sub elome ic DNA me hy- la ion pa e n. •The ques ion o an age- ela ed accumula ion o mu a ions in m DNA was o be add essed by quan i ying he le el o he e o- plasmy. Special a en ion was paid o he e oplasmy o he m DNA con ol egion. •Dono s we e s a ified o hei APOE geno ype o co ela e his wi h he ype o ageing, i.e. success ul o unsuccess ul ageing. 3.3. WP 3: ma ke s based on p o eins and hei modifica ions One impo an physiological pos ansla ional modifica ion o sec e ed p o eins is addi ion o N-linked oligosaccha ides (N- glycans). Since mos N-glycans a e on he ou e su ace o cellula and sec e ed mac omolecules, hey can modula e o media e a wide a ie y o e en s in cell-cell and cell-ma ix in e ac ions c ucial o he de elopmen and unc ion o complex mul icellula o ganisms. Because he biosyn hesis o glycans is no con olled by in e ac- ion wi h a empla e bu depends on he complica ed conce ed ac ion o glycosyl ans e ases, he s uc u es o glycans a e much mo e a iable han hose o p o eins and nucleic acids, and hey can be easily al e ed by he physiological condi ions o he cells. Acco dingly, s udying age- ela ed al e a ions o he glycans could be ele an o unde s anding he complex physiological changes in ageing indi iduals (Vanhoo en e al., 2010). The objec i e o his sub- ask was o de e mine he changes in he blood N-glycome du - ing human ageing o heal hy humans and o de elop me hodology o p ofiling u ine N-glycans (Vanhoo en e al., 2007). The apolipop o ein J/Clus e in (ApoJ/CLU) is a highly conse ed mul i unc ional glycop o ein. Amongs i s mul iple physiological unc ions, his p o ein is a chape one ha s abilizes s essed p o- eins in a olding-compe en s a e. P e ious wo k had shown ha ApoJ/CLU is associa ed wi h human ageing and wi h ageing o human cells in i o, and ha i s le el is inc eased in pa ien s wi h ype II diabe es, co ona y hea disease, and myoca dial in a c ion. The e o e ApoJ/CLU may ep esen a aluable ageing bioma ke . Non-enzyma ic p o ein glyca ion is a common pos ansla ional modifica ion o p o eins in i o, esul ing om eac ions be ween glucose and amino g oups on p o eins; his p ocess is e med “Mail- la d eac ion” and leads o he o ma ion o Ad anced Glyca ion Endp oduc s (AGEs). Du ing no mal ageing, he e is accumula ion o AGEs o long-li ed p o eins such as collagens and ca ilage. AGEs, ei he di ec ly o h ough in e ac ions wi h hei ecep o s, a e in ol ed in he pa hophysiology o nume ous age- ela ed diseases (Simm e al., 2014), such as ca dio ascula and enal diseases and neu odegene a ion. Howe e , in a coho s udy on human ageing, he co ela ion o AGEs wi h human age emained deba able. By analysing o e all pa ame e s o AGEs as well as specific AGEs, i was o be de e mined i hese modifica ions co ela e wi h age indepen- den ly o disease and i he e a e gende di e ences (Scheubel e al., 2006; Simm e al., 2004). I is well known ha le els o oxidised p o eins inc ease wi h age, due o inc eased p o ein damage induced by ROS, dec eased elimina ion o oxidised p o ein (i.e. epai and deg ada ion), o a combina ion o he wo. Since he p o easome is in cha ge o bo h gene al p o ein u no e and he selec i e emo al o oxidized p o- ein, i s a e du ing ageing has ecei ed conside able a en ion, and e idence has been p o ided o an impai men o p o easome unc ion wi h age in di e en cellula sys ems (Chond ogianni and Gonos, 2010; Ba aiba and F igue , 2013), including human PBMC (Chond ogianni e al., 2003, 2005; Ca a d e al., 2003; F igue , 2006). Apa om o being deg aded, ce ain oxidised p o eins can be epai ed. Howe e , epai is limi ed o he e e sion o a ew oxida i e modifica ions o sulphu -con aining amino acids, such as he educ ion o me hionine sul oxide by he me hionine sul oxide educ ase (Ms ) sys em. Ms ac i i y is known o be impai ed du - ing ageing and eplica i e senescence. The e o e, he s a us o bo h p o easome and Ms s A and B in human PBMC om he ec ui ed 6 A. Bü kle e al. / Mechanisms o Ageing and De elopmen 151 (2015) 2–12 dono s o di e en ages was o be assessed. These pa ame e s had p e iously been shown o be key playe s in oxidised p o ein deg a- da ion and epai and o exhibi a declining ac i i y wi h age (Pico e al., 2004, 2007). These p o ein main enance sys ems, iewed as po en ial bioma ke s o ageing, we e o be moni o ed a he le els o enzyma ic ac i i y, p o ein exp ession, and RNA exp ession. The ollowing esea ch asks ha e specifically been add essed: •Analysiso he N-glycomic changes in glycop o eins om blood o all dono s. U ine glycop o ein changes we e o be s udied in a subse o subjec s. •ApoJ/CLU le els in se um om all dono s. •AGEs in plasma by fluo escence spec oscopy and by immuno- logical analysis o ca boxy-me hyllysine, pen osidine, a g- py imidine and imidazolone. •P o ein damage in blood a di e en le els, i.e. ac i i ies o p o- easome and me hionine sul oxide educ ases in PBMC lysa es; RNA le els o ep esen a i e p o easome subuni s (ca aly ic and egula o y) and me hionine sul oxide educ ases A and B; and p o ein le els o ep esen a i e p o easome subuni s and me hi- onine sul oxide educ ases A and B in PBMC lysa es. 3.4. WP 4: immunological ma ke s Thymic ou pu is known o decline wi h age; u he mo e he a e o decline is dependen on gende , wi h g ea e hymic ou - pu o longe in emales compa ed wi h males (Aspinall e al., 2007). Females o en de elop be e immune esponses han males, which may ela e o hei longe li espan. Wi hin MARK-AGE, signal join T-cell ecep o ea angemen excision ci cles (sjTREC) we e assessed as a candida e bioma ke o ageing and hymic in olu ion, by analysing sjTREC le els in indi iduals a a ious ages (Aspinall e al., 2007). A obus immunological memo y is known o be a gua an o o heal h in adul s and in pa icula in elde ly pe sons, while ch onic la en in ec ions, such as CMV in ec ion, ha e been shown o be associa ed wi h sho e li e expec ancy. Au o-immune esponses may also es ic he di e si y o immune esponsi eness o o - eign an igens. We he e o e e alua ed long- e m and sho - e m immunological memo y and au oimmune esponses as po en ial bioma ke s o ageing (Almanza e al., 2005; Ko aiou e al., 2007; Weinbe ge e al., 2007; He ndle -B ands e e e al., 2005). In i o, wo ypes o senescence ha e been desc ibed. One is elome e-dependen eplica i e senescence and he second is s ess-induced p ema u e senescence (SIPS) (Siko a e al., 2014). In iew o p e ious esul s we hypo hesised ha du ing age- ing, ch onic an igenic load as well as oxida i e s ess may cause dec eased lymphocy e suscep ibili y o Damage-Induced Cell Dea h (DICD) and, on he o he hand, inc eased suscep ibili y o Ac i a ion-Induced Cell Dea h (AICD). As an in ac equilib ium be ween su i al and elimina ion o immune cells may be deci- si e o in ac immune unc ion and heal h we s udied DICD and AICD in T cells om dono samples (see Siko a e al., his issue). The ollowing esea ch asks ha e specifically been add essed: •Analysis o o al IgG, IgE, IgM and IgA; se um/plasma concen- a ions o 14 cy okines; blood coun s and di e en ial blood coun s (pe o med by he ec ui e s locally); and pheno yping o T cells, B cells NK cells and monocy es by immunofluo escence in p oband samples. •Analysis o he numbe o sjTRECs; i was o pa icula in e es o analyse whe he sjTREC concen a ions di e in pe sons wi h and wi hou la en i al in ec ions such as CMV, HHV-6 and HHV-7. •Analysis o an ibodies and cellula immuni y (IFN gamma p o- duc ion by Elispo ) agains measles and mumps i us ( ypically childhood exposu e) in o de o assess long- e m immunological memo y. •Analysis o an ibodies and cellula immuni y o highly con- se ed p o eins o he influenza i us (NP and M p o eins) as well as e anus (an agen agains which mos adul pe sons a e accina ed a egula in e als) in o de o assess sho - e m immunological memo y. •Analysis o immune esponses agains CMV, in o de o assess he e ec o la en i al in ec ion. •Analysis o au oan ibodies agains hy oglobulin (as an exam- ple o a issue-specific an igen) and an inuclea an ibodies (as example o a sys emic immune esponse). •Analysis o suscep ibili y o Damage-Induced Cell Dea h (DICD) and Ac i a ion-Induced Cell Dea h (AICD), espec i ely, by using apop osis ma ke s. 3.5. WP 5: clinical chemis y, ho mones and ma ke s o me abolism In he li e a u e a ple ho a o classical clinical chemis y pa am- e e s ha e been p oposed as po en ial bioma ke s o ageing. P ominen examples a e pa ame e s o ca bohyd a e and lipid me abolism o ho mones. We ha e selec ed he mos p omising ones o inclusion in he MARK-AGE p ojec and we ha e added se - e al new po en ial bioma ke s ela ed wi h me abolism ha ha e eme ged in he ecen wo k o some Pa ne s (Al-Delaimy e al., 2006; Rezzi e al., 2007a,b; Kochha e al., 2006; Heijmans e al., 2006; Mooijaa e al., 2006; Hu me e al., 2005, 2007; Ron u e al., 2006; Leh imaki e al., 2007). The ollowing candida e bioma ke s ha e specifically been add essed: Sys emic me abolism and oxici y pa ame e s •Blood u ea ni ogen and c ea inine, used o he e alua ion o enal unc ion. •Me al binding p o eins including ans e in, e i in, ␣2- mac oglobulin and ce uloplasmin, in o de o complemen me al ion de e mina ions (see below). •Fas ing glucose and as ing insulin as a measu e o glucose homeos asis, insulin esis ance and diabe ic condi ions. •Glycosyla ed hemoglobin (A1C) as a measu e o he long- e m sys emic glucose load, in o de o de ec (p e) diabe ic condi ions. •Some basic/ e e ence pa ame e s, including albumin and se um p o ein concen a ion. Fa y acid and choles e ol me abolism pa ame e s •Fas ing iglyce ides and ee a y acids we e measu ed o de ec me abolic diso de s in lipid me abolism. •To al choles e ol, HDL and LDL-choles e ol we e measu ed ( oge he wi h iglyce ides) o isk assessmen o ca dio ascula diseases. •Concen a ions o lipop o ein pa icle size classes by NMR. Sys emic inflamma ion pa ame e s •C- eac i e p o ein (CRP), homocys ein, u ic acid and fib inogen a e inflamma o y ma ke s associa ed ca dio ascula disease and hype ension. •Se um amyloid A and P, and pen axin 3 we e amongs he acu e phase p o eins s udied. •Adiponec in is co ela ed wi h an an i-inflamma o y s a e and supp esses me abolic de angemen s ha may esul in ype II diabe es, obesi y, a he oscle osis and non-alcoholic a y li e disease. A. Bü kle e al. / Mechanisms o Ageing and De elopmen 151 (2015) 2–12 7 Addi ional candida e bioma ke s •Tes os e one, he p incipal male sex ho mone whose le els a e known o decline g adually wi h age in males. •P os a e specific an igen (PSA) was measu ed o he de ec ion o (p e) neoplas ic p ocesses in he p os a e and p os a e cance in pa icula . •Vi amin D, a ecen ly iden ified p omising candida e ela ed o ageing and se e al ch onic diseases, was also s udied. •Dehyd oepiand os e one sul a e (DHEAS) is known o decline wi h age and is a classical candida e bioma ke o human ageing (Lane e al., 1997). No el bioma ke s o be de i ed om me abonomics analysis •Nuclea Magne ic Resonance (NMR)-based me abolic p ofiling o se um/plasma samples and u ine samples om p obands. NMR p ofiles display a se o esonances a ising om majo low- molecula weigh molecules, such as ke one bodies, o ganic acids, amino acids, and a oma ic me aboli es (Oos endo p e al., 2006; Rezzi e al., 2007a,b; Ramadan e al., 2007) 3.6. WP 6: oxida i e s ess ma ke s I has been pos ula ed ha oxida i e s ess is causal o he age- ing p ocess. Oxida i e s ess e e s o an imbalance be ween ROS o ma ion and an ioxidan de ence. In human beings la ge amoun s o oxidan s a e o med in a ious physiological me abolic eac ions and e en a wide a ie y o pa hophysiological condi ions. The body is able o espond o such enhanced oxidan o ma ion wi h com- pensa o y an ioxidan eac ions. These eac ions a e also complex and a la ge a ie y o di e en enzymes a e in ol ed. I an ioxi- dan p o ec ion is insu ficien , oxida i e s ess wi h an enhanced o ma ion o oxida i e s ess ma ke s occu s. Oxida i e damage accumula ion in mac omolecules has been conside ed causa i e o cellula damage and pa hology. Such dam- age seems o be closely ela ed o he ageing p ocess. Al hough he ela ionship be ween oxida i e damage and he ageing p ocess has been es ablished in a ious model sys ems, only ew s udies epo ed a sys ema ic analysis o oxida i e s ess pa ame e s in heal hy humans ela ed o age o indi iduals (Pandey and Riz i, 2010; Gil e al., 2006). The pu pose o his wo k package was o analyse a se o pa am- e e s o oxida i e s ess pa ame e s (Gil e al., 2006), i amins and ace elemen s (Mazza i e al., 2007; Mala ol a e al., 2006; Mocchegiani e al., 2006) in human blood, se um, u ine and buc- cal mucosa cells. P e e ence was gi en o new echnologies o he assessmen o oxida ion ma ke s and o ma ke s al eady es ab- lished and sui able o adap a ion o high- h oughpu o ma s. The ollowing candida e bioma ke s ha e specifically been add essed: •Malondialdehyde. •Ca bonyla ed and ni a ed p o eins. •Oxida ion o LDL. •NO me abolic-pa hway p oduc s (NOx)x. •Isop os anes. •Cellula glu a hione. •Vi amin con en (␣- ocophe ol, ␣-ca o ene and asco ba e) o se um and buccal mucosal cells. •T ace elemen s (Zn, Cu, Se and Fe) in blood/se um. 3.7. WP 7: eme gen bioma ke s o ageing om model sys ems and no el me hodological app oaches Whils con en ional bioma ke s o disease ha e been es ab- lished by hypo hesis-d i en app oaches based on an unde lying knowledge o he disease p ocess o se endipi y, s udies which ocus on iden ifica ion o bioma ke s o heal hy ageing a e con- s ained in hei abili y o ollow indi iduals o e p olonged pe iods o ime un il hei ch onological age de ia es om hei biological age. To o e come his p oblem, we adop ed pa allel, sys ema ic app oaches o in es iga e pu a i e bioma ke s in specific ageing coho s (as defined in WP1) alongside he s udy o models o accel- e a ed ageing, such as p oge oid synd omes (in humans and mice) and induced senescence in leukocy es om subjec s o di e en ages. We used bo h es ablished and no el app oaches o sea ch o bioma ke s o ageing in an i e a i e p ocess, whe e ma ke s de i ed om models would in o m in i o bioma ke sea ches. (a) Model Sys ems P oge oid mouse models P oge oid synd omes a e a e diso de s wi h p ema u e ageing and a sho ened li e expec ancy. These condi ions a e cha ac e - ized by ex emely accele a ed ageing, showing many hallma ks o no mal ageing including cessa ion o g ow h, li e , kidney and bone abno mali ies, e inopa hy, hea ing loss, sa copenia, neu- odegene a ion, sensi i i y o UV ligh , and a p ema u e aged appea ance due o kyphosis, baldness, loss o subcu aneous a , and d y w inkled skin. Cockayne synd ome (CS), Hu chinson- Gil o d synd ome, We ne ’s synd ome (WS), Bloom’s synd ome, and icho hiodys ophy a e all au osomal ecessi e diso de s wi h p oge oid symp oms. Al hough some di e ences exis in he pa hology o hese condi ions i is s iking ha he causal ac- o o all hese synd omes lies in impai ed genome main enance due o DNA epai deficiencies o genome ins abili y. One aim o he p ojec was he iden ifica ion o bioma ke s o ageing in CS, a a e human diso de , in which pa ien s su e om segmen- al bu bona fide accele a ed ageing. The mean age a dea h o CS pa ien s is 12.5 yea s. The e is cu en ly no ea men o CS and ela ed diso de s a ailable and he clinical managemen o pa ien s is pu ely suppo i e. CS is an au osomal ecessi e diso de caused by mu a ions in he CSA o CSB genes, which a e in ol ed in ansc ip ion-coupled nucleo ide and base excision epai (TCER). The TCER sub-pa hway selec i ely emo es lesions om he an- sc ibed s and ha ac ually block ansc ip ion. As such his p ocess is impo an o p omo ing eco e y o he i al p ocess o an- sc ip ion and hus cellula su i al om ansc ip ion-blocking DNA lesions. One o he objec i es o he p ojec was o apply he bioma ke s s udied in Wo k Packages 2 h ough 6 o p oge oid pa ien s, he eby de e mining o which ex en p ema u e and no - mal ageing esemble each o he and hei sui abili y o iden i y ageing ea u es independen o ch onological age. To unde s and he ae iology o CS and o he DNA epai dis- o de s such as icho hiodys ophy (TTD) and he cance -p one condi ion xe ode ma pigmen osum (XP), an ex ensi e collec ion o DNA epai -deficien ansgenic mice had p e iously been gene - a ed, many o which mimic he hallma ks o he co esponding human epai synd ome. These mice display ei he a s ong can- ce p edisposi ion (XP-like) o many ea u es o p ema u e ageing, o a combina ion. Al hough he Csbm/m mouse model eliably eflec s he epai de ec and UV-sensi i e pheno ype o he pa ien , animals show ela i ely mild g ow h e a da ion and neu ological abno mali ies, accompanied by age- ela ed e inal degene a ion. In e es ingly, when TCER-de ec i e Csbm/m mice we e c ossed wi h comple ely NER-deficien Xpa−/−animals double mu an s phenocopy human CS su p isingly well, including i s age- ela ed 8 A. Bü kle e al. / Mechanisms o Ageing and De elopmen 151 (2015) 2–12 pa hology. Al hough Csbm/m/Xpa−/−pups a e de oid o any o e emb yonic de elopmen al pheno ype, hey display se e e pos - na al g ow h e a da ion, impai ed psychomo o de elopmen , a axia, p og essi e cachexia, and kyphosis. Loss o e inal pho- o ecep o s is also u he accele a ed, as compa ed o Csbm/m animals. Mo eo e , mos Csbm/m/Xpa−/−newbo ns die du ing o sho ly a e bi h, whe eas animals ha su i e bi h s ess do no su i e beyond h ee weeks. Exploi ing he gene ic and en i- onmen ally ully defined mouse sys em hese and o he mouse mu an s o XP, XP/CS and TTD p o ided a con enien ool o deduce specific bioma ke s in a ious o gans/ issues including se um. Wi hin MARK-AGE a se o mouse bioma ke s o ageing was de eloped and e alua ed, using ansc ip omics, immuno- his ochemis y and se um/u ine ma ke s in p ema u ely ageing mouse models wi h di e en li e span and ageing- ela ed pa hol- ogy o deli e uni e sal ma ke s o ageing. Biological ma e ial om gene ically and en i onmen ally con olled, (his o) pa hologically well-defined coho s udies wi h NER-deficien mouse models, se ed o iden i y pa ame e s ha we e expec ed epo on he biological age o he animals and/o he onse and p og ession o ageing- ela ed pa hology in a ious issues (e.g. li e , b ain) ( an de Pluijm e al., 2007; de Boe e al., 2002; Rossi e al., 2007; Niede nho e e al., 2006). S ess induced p ema u e senescence The e a e se e al pa hways ac i a ing cellula senescence; hese include elome e uncapping, DNA damage, oxida i e s ess and oncogene, amongs o he s (Ben-Po a h and Weinbe g, 2005). No - mal human diploid fib oblas s cul u ed in i o i e e sibly s op di iding a e a ce ain numbe o cumula i e popula ion doublings in a p ocess known as eplica i e senescence (Hayflick, 1965). This limi ed p oli e a i e li e span has been obse ed in many o he euka yo ic cell ypes and has been in e p e ed as a mani es a ion o cellula ageing. Random me abolic modifica ions appea wi hin hese cells o e ime, leading o andom damage o he cellula com- ponen s. These damaged cellula componen s a e no comple ely elimina ed o epai ed and he e o e accumula e wi h ime and p og essi ely impai cellula unc ions. Cellula senescence can be also ega ded as a pe manen ly main ained DNA damage esponse s a e ( on Zglinicki e al., 2005). ROS a e impo an con ibu o s o he ageing p ocess and we ha e confi med he simila i ies be ween eplica i e senescence and s ess induced p ema u e senescence (SIPS) (Dumon e al., 2000; Die ick e al., 2002; Pascal e al., 2005). In MARK-AGE, we used SIPS o “age” T and B cells om he subjec s ec ui ed in WP1 and sea ched o no el bioma ke s using genomic a ay and p o eomic app oaches desc ibed below (Debacq-Chainiaux e al., 2005; F ippia e al., 2001). •No el me hodological app oaches Analysis o miRNAs Mic oRNAs (miRNAs) a e small, abundan non-coding RNA molecules o abou 21–23 nucleo ides ha ha e been shown o a ec a b oad spec um o biological ac i i ies. In e es ingly, he e is e idence ha a ema kably la ge p opo ion o genes (>30%) a e subjec o mic oRNA-media ed egula ion. In gene al, miR- NAs unc ion pos - ansc ip ionally by inhibi ing ansla ion om specific a ge mRNAs. Up o now, abou 600 miRNAs ha e been cha ac e ized in humans. These small RNA molecules we e hough o con ibu e o ageing o C. elegans. I has been p e iously shown ha miRNA cause a gene al educ ion o message-specific ansla- ional inhibi ion du ing ageing. Reducing he ac i i y o a specific miRNA lin-4 sho ened li e span and accele a ed issue aging, whe eas o e exp essing lin-4 o educing he ac i i y o lin-14 ex ended li e span o C. elegans, equen ly used as a model sys- em o mammalian ageing (Boehm and Slack, 2005). S udies on miRNA exp ession le els in issues o young and old mice showed he di e en ial and clea ly issue specific exp ession o some miR- NAs (Smi h-Vikos and Slack, 2012). On he le el o cells, i was also shown ha such di e en ial exp ession can di ec ly influ- ence cellula ageing. miR-21 was ound up- egula ed by eplica i e and s ess-induced senescence in human endo helial cells. miR- 21 o e -exp ession educes he eplica i e li espan, while s able knock-down ex ends he eplica i e li espan o hese endo helial cells (Dellago e al., 2013). On he o gan le el, i is clea ha no all miRNAs ha a e up- o down- egula ed du ing ageing neces- sa ily play c ucial oles du ing ageing. As no “key egula o ” on issue ageing was iden ified ye in mammalians, one has o cla i y which miRNAs a e ac i a ed o ep essed especially in degene a- i e disease con ex s and which a e eally associa ed wi h aging pe se. The e o e we a e e alua ing miRNA exp ession as a “no el” bioma ke o ageing using he leukocy es o subjec s ec ui ed in WP1. Phage an ibodies agains no el ma ke s o endo helial and T cell ageing The phage display an ibody lib a y echnology has been ound o be a use ul me hod o isola e an igen-specific an ibody ag- men s, since he epe oi e o an ibody specifici ies is b oad and since i bypasses he need o immuniza ion. When applied as a disco e y ool, he phage display echnology can be conside ed a complemen a ion o adi ional p o eomic app oaches using 2- D gel elec opho esis and mass spec ome y, which qui e o en ha e p oblems in iden i ying p o eins which a e e y hyd ophobic (Gonzalez-Dosal e al., 2006; Jensen e al., 2003). Du ing ageing, bo h in i o and in i o, changes in he p o eomic p ofile a e obse ed. By pe o ming sub ac i e selec ion o ecombinan an i- bodies binding o e.g. endo helial cells allowed o age in i o, whe e young cul u ed endo helial cells is applied as compe i o , an ibod- ies binding po en ial bioma ke s o ageing can be ob ained (Boisen e al., 2010; Boisen and K is ensen, 2010) Such app oaches ha e enabled he de elopmen o a panel o an ibodies ecognizing in i o ageing human endo helial cells du - ing he p e ious EU p ojec P o eomage, in pa icula he sec e ome o endo helial cells. Wi h age he e is a dec ease in he abili y o o m new blood essels, which is a bioma ke o ageing. The pu - pose o he wo k in he MARK-AGE p ojec was o e alua e he in i o significance o endo helial sec e ome bioma ke s iden i- fied om in i o models o hei in i o ele ance by sc eening plasma om subjec s ec ui ed in WP1 o hese ma ke s. Mo e specifically i was ound ha he in e media e filamen p o ein, imen in, is ound in he se um. As s udies wi hin he EU unded p ojec , P o eomage, es ablished ha ex acellula imen in can exe unc ional changes o he abili y o o m new blood essels, i has been o pa icula in e es o see i he e is an age specific co - ela ion wi h he amoun o imen in in se um. This, in pa , migh explain why in gene al olde people exhibi dec eased abili y o o m new blood essels. Using a ba e y o an ibodies aised agains endo helial p ogeni o cells, i was u he p oposed o e alua e whe he he numbe o endo helial p ogeni o cells in se um qual- ifies as a bioma ke o ageing (Be elsen e al., 2014; Williamson e al., 2012). This again could ha e implica ions o he gene ally dec eased abili y o olde subjec s o o m new blood essels. Mic oa ay and p o eomics Genomics and p o eomics o e he oppo uni y o an unbiased sys ema ic disco e y ou e o no el bioma ke s and a e becom- ing inc easingly popula (G i fi hs e al., 2002; G i fi hs e al., 2006; Ald ed e al., 2006; G an e al., 2007). Ne e heless, only ew g oups ha e unde aken p o eomic s udies o ei he plasma p o eins o mononuclea cells in heal hy human ageing. The fi s o hese, published by Thambie y e al. in 2010, desc ibed a di e en ial plasma p o ein pa e n be ween 57 olde adul s wi h and wi hou amyloid deposi ion in he b ain (Thambise y e al., 2010). Subse- A. Bü kle e al. / Mechanisms o Ageing and De elopmen 151 (2015) 2–12 9 quen ly, di e en ial exp ession o ApoE and an ioxidan p o eins was obse ed in he plasma o 10 Japanese supe cen ena ians com- pa ed wi h 10 young people (Miu a e al., 2011). In 2012, some o us desc ibed al e a ions in ans e in glycosyla ion du ing heal hy ageing (Duns on e al., 2012). The ad an ages o applying such an app oach in he MARKAGE popula ion is he g ea e powe o e alu- a e he alidi y o no el bioma ke s disco e ed h ough p o eomics when compa ed wi h small sample size disco e y p og ammes. A e he comple ion o MARKAGE, one o he s udy has analysed by ELISA he le els o p o ein bioma ke s ha we e no disco e ed using p o eomics in plasma du ing heal hy ageing in compa ison wi h olde adul s ha de elop ail y synd omes. These au ho s showed ha highe le els o ans e in fib inogen and in e leukin- 6 we e associa ed wi h ail y s a us and ail y sco e (Da in e al., 2014). E e since he concep o MARK-AGE has been de eloped mic oa ay has been used ex ensi ely o cha ac e ise ageing- ela ed changes in gene exp ession. Indeed, he sys ema ic analysis o miR, single nucleo ide polymo phisms and deep sequencing app oaches ha e led o u he insigh in o exp ession changes in specific cell ypes (Lau ie e al., 2012; Smigielska-Czepiel e al., 2014; an de B ug e al., 2010). The la e a e by necessi y c oss-sec ional s udies. The app oach aken by MARK-AGE was o in es iga e exp ession changes wi hin unique accele a ed mod- els o ageing he conso ium had access o. Sys ema ic mic oa ay analysis o p oge oid mu an s has al eady yielded new po en ial bioma ke s: using ull-genome mic oa ay analysis some o us ha e ecen ly iden ified a ‘su i al’ esponse in he p oge oid mouse models di ec ed by down- egula ion o he IGF1 soma o ophic axis ha boos s an ioxidan de ence, down- egula es me abolism and edi ec s ene gy esou ces om g ow h and de elopmen o p o ec ion, main enance and epai . This adap i e swi ch aims o slow down ageing- ela ed pa hology and pos pones dea h, he eby p omo ing success ul ageing. I is cons i u i ely u ned on in he epai -comp omised mouse mu an s as a u ile a emp o ex end li espan and explains hei dwa pheno ype. In no mal mice he same p incipally beneficial esponse can be ansien ly igge ed by ch onic exposu e o DNA-damaging agen s and ROS-p oducing compounds. We hypo hesise ha no mal ageing is also associ- a ed wi h a simila esponse due o age-dependen accumula ion o damage. The exis ence o such a esponse allows p edic ions o shi s in le els o specific p o eins, ac i i ies, pa hways and me aboli es ha could se e as bioma ke s. As he p o eome is a mo e ex ensi e han he ansc ip ome, i o e s a iche sou ce o po en ial bioma ke s bu also poses inc eased p oblems in e ms o dynamic ange, pa icula ly in plasma. This is being add essed using quali y-assu ed sub ac iona ion s eps and es ic ed IPG ange in he fi s dimension. In MARK-AGE, p o eomics was o be adop ed in acco dance wi h HUPO guidelines, in he sea ch o bioma k- e s in plasma om subjec s ec ui ed in WP1, in CS/p oge ia and in T and B cells subjec ed o SIPS in subjec s ec ui ed in WP1. The iden ifica ion o pu a i e ma ke s was o be confi med by sequence analysis and hei alida ion as bioma ke s o ageing con- fi med by al e na i e app oaches such as ELISA whe e possible. In iew o he s ong pa allels be ween he mouse mu an s and he human synd omes, down- egula ion o he IGF1 soma o ophic axis bioma ke s is likely o be ins umen al o iden ifica ion o co - esponding ma ke s in human pa ien s and e en no mal ageing and will be e alua ed in WP1 subjec s. 3.8. WP 8: da a analysis and bioin o ma ics In iew o he la ge amoun s o clinical and biochemical da a, which ha e been collec ed in he amewo k o he MARK-AGE p ojec an app op ia e and cohe en s a egy o da a analysis and model building is manda o y. In o de o ex ac a obus se o bioma ke s o human ageing and o de i e a model o heal hy ageing, he ollowing asks we e pe o med: •Da ap ope y analysis. Pa ial knowledge abou suspec ed co - ela ions be ween measu emen s is a ailable and being used o judge he noise a io wi hin some o hese measu emen s using classical s a is ical echniques. We also applied co ela ion measu es o iden i y addi ional ela ionships be ween measu e- men s. Repea ed sampling, i.e. ob aining samples a e 6 mon hs om 97 subjec s, was done o u he in es iga e biological and analy ical a iabili y in he measu emen s. •Modelling. Bo h, s a is ical models as well as machine lea n- ing/da a mining me hods we e o be used in o de o build models aiming o p edic biological age om he a ailable measu e- men s. We used classical echniques, such as eg ession analysis bu also aim o in oduce addi ional knowledge (mono onici y) o imp o e hose models. Neu al Ne wo ks and Decision/Reg ession T ees we e o be used o find mo e local ela ionships wi hin he da a, o ins ance e ealing p ope ies ha a e ele an only o a subse o he chosen popula ion. •Va iance educ ion. Th ough dimensionali y educ ion ech- niques (p incipal componen analysis and o he s) we aim a educing he numbe o equi ed measu emen s while, a he same ime, educing he a iance in he p edic ions gene a ed. Machine Lea ning o e s ensembles o models o his pu pose, which allows combining di e en , di e se p edic o s o gene a e models wi h lowe a iance. •Clus e ing/ isualisa ion. We expec o disco e p e iously unknown o unexpec ed ela ionships in he da a ha define success ul ageing. Da a isualisa ion echniques and in e ac i e me hods such as isual clus e ing models may help unco e some o hese ela ionships. I is expec ed ha some measu emen s will ha e highe co ela ions wi h he biological age han o he s in pa s o he popula ion. Finding such clus e s is u he helping educe a iance in he gene a ed models since we will be able o be e model cha ac e is ics o subg oups. 3.9. WP 9 and WP 10 WP 9 and WP 10 we e dedica ed o dissemina ion, aining, p ojec managemen (Fig. 1) and e hical issues. 4. Discussion Bioma ke s o human ageing a e u gen ly needed o a i- e y o easons, including he iden ifica ion o indi iduals a high isk o de eloping age-associa ed disease o disabili y. This would p omp a ge ed ollow-up examina ions and, i a ailable, p o- phylac ic in e en ion (e.g. changes in li es yle) o ea ly-s age ea men o age- ela ed disease. Fu he mo e, he a ailabili y o powe ul bioma ke s would allow he assessmen o he e ficacy o o hcoming pha macological and o he in e en ions (including op imisa ion o mic onu ien in ake and o he die a y componen s o physical ac i i y) cu en ly being de eloped wi h he aim o lowe he isk o age-associa ed disease e en in indi iduals wi hou accele a ed ageing. In iew o he apidly inc easing a e age li e expec ancy o human beings wo ld-wide, he p e alence o age- ela ed diseases is likely o inc ease as well. This necessi a es e ec i e new s a egies o p e en ion and ea ly diagnosis o such condi ions. I should be no ed ha di e en ypes o bioma ke s ha e been en isaged: (1) “neu al” ma ke s o age (also called ma ke s o “ch onological ageing”) possibly lacking he powe o di ec ly p e- dic ing disease isk, as he unde lying physiological change may pe 10 A. Bü kle e al. / Mechanisms o Ageing and De elopmen 151 (2015) 2–12 se be ha mless and wi hou unc ional comp omise; (2) ma ke s o (global) isk o age- ela ed diseases. Those in ca ego y #2 a e o ob ious and di ec in e es . Ne e heless, he use ulness o hose in ca ego y #1, especially as a combina ion o ma ke s, should no be dismissed, as hey could eflec he a e o ageing ha has p e ailed o e a ce ain ime pe iod in he pas and he cumula i e change i has p oduced in he body. The de ec ion o a p e ious as e - han- no mal a e o ageing in a gi en indi idual should also be ala ming and call o addi ional diagnos ic and p e en i e measu es. Disclaime Unile e PLC con ibu ed only o he human wo k – wi hin Wo k Packages 5 (Clinical chemis y, ho mones and ma ke s o me abolism) and 6 (Oxida i e s ess) as well as analysis o human s udy da a wi hin Wo k Package 8 (Da a analysis and bioin o ma - ics). Acknowledgemen s We wish o hank he Eu opean Commission o financial sup- po h ough he FP7 la ge-scale in eg a ing p ojec “Eu opean S udy o Es ablish Bioma ke s o Human Ageing” (MARK-AGE; g an ag eemen no.: 200880). We a e e y g a e ul o D . Bea - ice Luca oni (Eu opean Commission Scien ific O fice in cha ge o he MARK-AGE p ojec ) o he excellen suppo and ad ice du ing all phases o he p ojec . Re e ences Al-Delaimy, W.K., Jansen, E.H., Pee e s, P.H., an de Laan, J.D., an Noo d, P.A., Boshuizen, H.C., an de Schouw, Y.T., Jenab, M., Fe a i, P., Bueno-de-Mesqui a, H.B., 2006. Reliabili y o bioma ke s o i on s a us, blood lipids, oxida i e s ess, i amin D, C- eac i e p o ein and uc osamine in wo Du ch coho s. Bioma ke s 11 (4), 370–382. Ald ed, S., Sozzi, T., Mudway, I., G an , M.M., Neube , H., Kelly, F.J., G i fi hs, H.R., 2006. Alpha ocophe ol supplemen a ion ele a es plasma apolipop o ein A1 iso o ms in no mal heal hy subjec s. P o eomics 6 (5), 1695–1703. Almanza , G., Schwaige , S., Jenewein, B., Kelle , M., He ndle -B ands e e , D., Wu zne , R., Schoni ze , D., G ubeck-Loebens ein, B., 2005. Long- e m cy omegalo i us in ec ion leads o significan changes in he composi ion o he CD8+ T-cell epe oi e, which may be he basis o an imbalance in he cy okine p oduc ion p ofile in elde ly pe sons. J. Vi ol. 79 (6), 3675–3683. Al ilia, S., San o o, A., Malagoli, D., Lanza ini, C., Balles e os Al a ez, J.A., Galazzo, G., Po e , D.C., C occo, P., Rose, G., Passa ino, G., Roninson, I.B., F anceschi, C., Sal ioli, S., 2012. TP53 codon 72 polymo phism a ec s accumula ion o m DNA damage in human cells. Aging (Albany N.Y.) 4 (1), 28–39. Aspinall, R., Pido-Lopez, J., Imami, N., Henson, S.M., Ngom, P.T., Mo e, M., Niphuis, H., Rema que, E., Rosenwi h, B., Heeney, J.L., 2007. Old hesus macaques ea ed wi h in e leukin-7 show inc eased TREC le els and espond well o influenza accina ion. Reju ena ion Res. 10 (1), 5–17. Bake 3 d, G.T., Sp o , R.L., 1988. Bioma ke s o aging. Exp. Ge on ol. 23 (4–5), 223–239. Ba aiba , M.A., F igue , B., 2013. Oxida i e p o eome modifica ions a ge specific cellula pa hways du ing oxida i e s ess, cellula senescence and aging. Exp. Ge on ol. 48 (7), 620–625. Bellizzi, D., Ca alcan e, P., Ta e na, D., Rose, G., Passa ino, G., Sal ioli, S., F anceschi, C., De Benedic is, G., 2006. Gene exp ession o cy okines and cy okine ecep o s is modula ed by he common a iabili y o he mi ochond ial DNA in cyb id cell lines. Genes Cells 11 (8), 883–891. Ben-Po a h, I., Weinbe g, R.A., 2005. The signals and pa hways ac i a ing cellula senescence. In . J. Biochem. Cell Biol. 37 (5), 961–976. Beneke, S., Bu kle, A., 2007. Poly(ADP- ibosyl) a ion in mammalian ageing. Nucleic Acids Res. 35 (22), 7456–7465. Bene i, R., Ga cia-Cao, M., Blasco, M.A., 2007. Telome e leng h egula es he epigene ic s a us o mammalian elome es and sub elome es. Na . Gene . 39 (2), 243–250. Be elsen, L.B., Bohn, A.B., Smi h, M., Molgaa d, B., Molle , B., S odkilde-Jo gensen, H., K is ensen, P., 2014. A e endo helial ou g ow h cells a po en ial sou ce o u u e e- ascula iza ion he apy? Exp. Ge on ol. 58, 132–138. Boehm, M., Slack, F., 2005. A de elopmen al iming mic oRNA and i s a ge egula e li e span in C. elegans. Science 310 (5756), 1954–1957. Boisen, L., K is ensen, P., 2010. Con on ing cellula he e ogenei y in s udies o p o ein me abolism and homeos asis in aging esea ch. Ad . Exp. Med. Biol. 694, 234–244. Boisen, L., D asbek, K.R., Pede sen, A.S., K is ensen, P., 2010. E alua ion o endo helial cell cul u e as a model sys em o ascula ageing. Exp. Ge on ol. 45 (10), 779–787. Caia a, P., Zampie i, M., 2005. DNA me hyla ion and ch oma in s uc u e: he puzzling CpG islands. J. Cell Biochem. 94 (2), 257–265. Canela, A., Ve a, E., Kla , P., Blasco, M.A., 2007. High- h oughpu elome e leng h quan ifica ion by FISH and i s applica ion o human popula ion s udies. P oc. Na l. Acad. Sci. U. S. A. 104 (13), 5300–5305. Ca a d, G., Dieu, M., Raes, M., Toussain , O., F igue , B., 2003. Impac o ageing on p o easome s uc u e and unc ion in human lymphocy es. In . J. Biochem. Cell Biol. 35 (5), 728–739. Chond ogianni, N., Gonos, E.S., 2010. P o easome unc ion de e mines cellula homeos asis and he a e o aging. Ad . Exp. Med. Biol. 694, 38–46. Chond ogianni, N., S a o d, F.L., T ougakos, I.P., F igue , B., Ri e , A.J., Gonos, E.S., 2003. Cen al ole o he p o easome in senescence and su i al o human fib oblas s: induc ion o a senescence-like pheno ype upon i s inhibi ion and esis ance o s ess upon i s ac i a ion. J. Biol. Chem. 278 (30), 28026–28037. Chond ogianni, N., Tza elas, C., Pembe on, A.J., Nezis, I.P., Ri e , A.J., Gonos, E.S., 2005. O e exp ession o p o easome be a5 assembled subuni inc eases he amoun o p o easome and con e s amelio a ed esponse o oxida i e s ess and highe su i al a es. J. Biol. Chem. 280 (12), 11840–11850. Da in, K., Randolph, A., O alles, S., Halade, D., B eeding, L., Richa dson, A., Espinoza, S.E., 2014. Plasma p o ein bioma ke s o he ge ia ic synd ome o ail y. J. Ge on ol. A Biol. Sci. Med. Sci. 69 (2), 182–186. De Benedic is, G., Rose, G., Ca ie i, G., De Luca, M., Falcone, E., Passa ino, G., Bona e, M., Mon i, D., Baggio, G., Be olini, S., Ma i, D., Ma ace, R., F anceschi, C., 1999. Mi ochond ial DNA inhe i ed a ian s a e associa ed wi h success ul aging and longe i y in humans. FASEB J. 13 (12), 1532–1536. de Boe , J., And essoo, J.O., de Wi , J., Huijmans, J., Beems, R.B., an S eeg, H., Weeda, G., an de Ho s , G.T., an Leeuwen, W., Themmen, A.P., Me adji, M., Hoeijmake s, J.H., 2002. P ema u e aging in mice deficien in DNA epai and ansc ip ion. Science 296 (5571), 1276–1279. Debacq-Chainiaux, F., Bo lon, C., Pascal, T., Roye , V., Eliae s, F., Ninane, N., Ca a d, G., F igue , B., de Longue ille, F., Bo e, S., Remacle, J., Toussain , O., 2005. Repea ed exposu e o human skin fib oblas s o UVB a subcy o oxic le el igge s p ema u e senescence h ough he TGF-be a1 signaling pa hway. J. Cell Sci. 118 (P 4), 743–758. Deelen, J., Beekman, M., Uh, H.W., B oe , L., Aye s, K.L., Tan, Q., Kama ani, Y., Benne , A.M., Tamm, R., T ompe , S., Gue hbja sson, D.F., Flachsba , F., Rose, G., Vik o in, A., Fische , K., Nygaa d, M., Co dell, H.J., C occo, P., an den Akke , E.B., Boh inge , S., Helme , Q., Nelson, C.P., Saunde s, G.I., Al e , M., Ande sen-Ranbe g, K., B een, M.E., an de B eggen, R., Caliebe, A., Cap i, M., Ce enini, E., Colle on, J.C., Da o, S., Da ies, K., Fo d, I., Gampe, J., Ga agnani, P., de Geus, E.J., Ha ow, J., an Heems , D., Heijmans, B.T., Heinsen, F.A., Ho enga, J.J., Ho man, A., Jeune, B., Jonsson, P.V., La h op, M., Lechne , D., Ma in-Ruiz, C., McNe lan, S.E., Mihailo , E., Mon esan o, A., Mooijaa , S.P., Mu phy, A., Noh , E.A., Pa e nos e , L., Pos mus, I., Ri adenei a, F., Ross, O.A., Sal ioli, S., Sa a , N., Sch eibe , S., S e ansson, H., S o , D.J., Tiemeie , H., Ui e linden, A.G., Wes endo p, R.G., Willemsen, G., Samani, N.J., Galan, P., So ensen, T.I., Boomsma, D.I., Jukema, J.W., Rea, I.M., Passa ino, G., de C aen, A.J., Ch is ensen, K., Nebel, A., S e ansson, K., Me spalu, A., Magnusson, P., Blanche, H., Ch is iansen, L., Ki kwood, T.B., an Duijn, C.M., F anceschi, C., Houwing-Duis e maa , J.J., Slagboom, P.E., 2014. Genome-wide associa ion me a-analysis o human longe i y iden ifies a no el locus con e ing su i al beyond 90 yea s o age. Hum. Mol. Gene . 23 (16), 4420–4432. Dellago, H., P eschi z-Kamme ho e , B., Te lecki-Zaniewicz, L., Sch eine , C., Fo schegge , K., Chang, M.W., Hackl, M., Mon e o e, R., Kuhnel, H., Schosse e , M., G ube , F., Tschachle , E., Scheidele , M., G illa i-Voglaue , R., G illa i, J., Wiese , M., 2013. High le els o oncomiR-21 con ibu e o he senescence-induced g ow h a es in no mal human cells and i s knock-down inc eases he eplica i e li espan. Aging Cell 12 (3), 446–458. Die ick, J.F., Kalume, D.E., Wende s, F., Salmon, M., Dieu, M., Raes, M., Roeps o , P., Toussain , O., 2002. Iden ifica ion o 30 p o ein species in ol ed in eplica i e senescence and s ess-induced p ema u e senescence. FEBS Le . 531 (3), 499–504. Dumon , P., Bu on, M., Chen, Q.M., Gonos, E.S., F ippia , C., Maza a i, J.B., Eliae s, F., Remacle, J., Toussain , O., 2000. Induc ion o eplica i e senescence bioma ke s by suble hal oxida i e s esses in no mal human fib oblas . F ee Radic. Biol. Med. 28 (3), 361–373. Duns on, C.R., Choudhu y, K., G i fi hs, H.R., 2012. Te minal galac ose esidues on ans e in a e inc eased in midli e adul s compa ed o young adul s. P o eomics 12 (21), 3147–3153. Flo es, I., Cayuela, M.L., Blasco, M.A., 2005. E ec s o elome ase and elome e leng h on epide mal s em cell beha io . Science 309 (5738), 1253–1256. F igue , B., 2006. Oxidized p o ein deg ada ion and epai in ageing and oxida i e s ess. FEBS Le . 580 (12), 2910–2916. F ippia , C., Chen, Q.M., Zdano , S., Magalhaes, J.P., Remacle, J., Toussain , O., 2001. Subcy o oxic H2O2s ess igge s a elease o ans o ming g ow h ac o -be a 1, which induces bioma ke s o cellula senescence o human diploid fib oblas s. J. Biol. Chem. 276 (4), 2531–2537. Gil, L., Siems, W., Mazu ek, B., G oss, J., Sch oede , P., Voss, P., G une, T., 2006. Age-associa ed analysis o oxida i e s ess pa ame e s in human plasma and e y h ocy es. F ee Radic. Res. 40 (5), 495–505. Gonzalez-Dosal, R., So ensen, M.D., Cla k, B.F., Ra an, S.I., K is ensen, P., 2006. Phage-displayed an ibodies o he de ec ion o glyca ed p o easome in aging cells. Ann. N. Y. Acad. Sci. 1067, 474–478.