Poly(ADP-ribose)polymerases inhibitors prevent early mitochondrial fragmentation and hepatocyte cell death induced by H2O2

RESEARCH ARTICLE Poly(ADP- ibose)polyme ases inhibi o s p e en ea ly mi ochond ial agmen a ion and hepa ocy e cell dea h induced by H 2 O 2 Sand a M. Ma ı ´n-Gue e o 1 , Jose ´A. Muñoz-Ga ´mez 2 , Ma ı ´a-Ca men Ca asco 1 , Ja ie Salme o ´n 2 , Ma ı ´a Ma ı ´n-Es ebane ´ 1 , Miguel A. Cuad os 1 , Julio Na ascue ´s 1 , Da id Ma ı ´n-Oli a 1 * 1Depa amen o de Biologı ´a Celula , Facul ad de Ciencias, Uni e sidad de G anada, G anada, Spain, 2Ins i u o de In es igacio ´n Biome ´dica (ibsG anada), Hospi al Uni e si a io San Cecilio, G anada, Spain *[email p o ec ed] Abs ac Poly(ADP- ibose)polyme ases (PARPs) a e a amily o NAD + consuming enzymes ha play a c ucial ole in many cellula p ocesses, mos clea ly in main aining genome in eg i y. He e, we p esen an ex ensi e analysis o he al e a ion o mi ochond ial mo phology and he ela- ionship o PARPs ac i i y a e oxida i e s ess using an in i o model o human hepa ic cells. The ollowing ou comes we e obse ed: eac i e oxygen species (ROS) induced by oxida i e ea men quickly s imula ed PARPs ac i a ion, p omo ed changes in mi ochon- d ial mo phology associa ed wi h ea ly mi ochond ial agmen a ion and ene gy dys unc ion and inally igge ed apop o ic cell dea h. Pha macological ea men wi h speci ic PARP-1 ( he majo NAD + consuming poly(ADP- ibose)polyme ases) and PARP-1/PARP-2 inhibi o s a e he oxidan insul eco e ed no mal mi ochond ial mo phology and, hence, inc eased he iabili y o human hepa ic cells. As he PARP-1 and PARP-1/PARP-2 inhibi o s achie ed simila ou comes, we conclude ha mos o he PARPs e ec s we e due o PARP-1 ac i a- ion. NAD + supplemen a ion had simila e ec s o hose o he PARPs inhibi o s. The e o e, PARPs ac i a ion and he subsequen NAD + deple ion a e c ucial e en s in dec eased cell su i al (and inc eased apop osis) in hepa ic cells subjec ed o oxida i e s ess. These esul s sugges ha he al e a ions in mi ochond ial mo phology and unc ion seem o be ela ed o NAD + deple ion, and show o he i s ime ha PARPs inhibi ion ab oga es mi o- chond ial agmen a ion. In conclusion, he inhibi ion o PARPs may be a aluable he apeu- ic app oach o ea ing li e diseases, by educing he cell dea h associa ed wi h oxida i e s ess. In oduc ion The li e is a i al o gan ha plays a decisi e ole in de oxi ica ion, and he e o e hepa ic dam- age is equen ly he cause o se e e pa hologies. One o he main ac o s p o oking hepa ocy e degene a ion (and consequen ly li e damage) is oxida i e s ess, which is o en associa ed PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 1 / 26 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Ci a ion: Ma ı ´n-Gue e o SM, Muñoz-Ga ´mez JA, Ca asco M-C, Salme o ´n J, Ma ı ´n-Es ebane ´M, Cuad os MA, e al. (2017) Poly(ADP- ibose) polyme ases inhibi o s p e en ea ly mi ochond ial agmen a ion and hepa ocy e cell dea h induced by H 2 O 2 . PLoS ONE 12(10): e0187130. h ps://doi. o g/10.1371/jou nal.pone.0187130 Edi o : Aami Ahmad, Uni e si y o Sou h Alabama Mi chell Cance Ins i u e, UNITED STATES Recei ed: July 21, 2017 Accep ed: Oc obe 13, 2017 Published: Oc obe 26, 2017 Copy igh : ©2017 Ma ı ´n-Gue e o e al. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Da a A ailabili y S a emen : All ele an da a a e wi hin he pape and i s Suppo ing In o ma ion iles. Funding: This wo k was suppo ed by g an s om Minis e io de Economia y Compe i i idad, Spain (BFU2010-19981), CEI BioTic G anada, Spain (BS9-2015) and Minis e io de Educacio ´n, Cul u a y Depo e, Spain (FPU14/02219). The unde s had no ole in s udy design, da a collec ion and wi h he de oxi ica ion unc ion o he li e . Oxida i e s ess in a cell de elops when he e is an imbalance be ween he amoun o eac i e oxygen species (ROS) p esen and he abili y o he cell o elimina e i o o epai he damage esul ing om he ac ion o ROS. Oxida i e s ess leads o mul iple ypes o cell damage, including DNA b eaks, p o ein modi ica ions, lipid pe - oxida ion, dis up ion o calcium homeos asis, mi ochond ial ailu e, impai men o he ene gy me abolism and NAD + deple ion [1]. Oxida i e s ess is appa en ly a he o igin o mos li e diseases, such as hose caused by alcohol consump ion [2,3], hepa o oxic d ugs [4,5], en i on- men al pollu an s [6] and o he ac o s [7,8]. Mo eo e , oxida i e s ess in pa ien s su e ing non-alcoholic a y li e disease (NAFLD) is signi ican ly g ea e han in heal hy con ols [9]. ROS is a collec i e e m ha includes oxygen ee adicals (such as supe oxide anion, hyd oxyl and hyd ope oxyl adical) and non adical agen s (such as hyd ogen pe oxide [H 2 O 2 ], single oxygen and pe oxyni i e) wi h oxidising capaci y [10]. ROS a e p oduced as a consequence o oxida i e p ocesses ha ake place in a ious ypes o memb ane o ganelles, especially in he mi ochond ia du ing ae obic me abolism [11]. As well as being c ucial o ene gy p oduc ion, he mi ochond ia also pa icipa e in o he aspec s o cell ac i i y such as apop osis and he biosyn hesis o ce ain molecules. Consequen ly, when ROS a e p esen in excess, a ec ing mi ochond ial unc ion, hey may o igina e bioene ge ic and me abolic laws ha unde lie a he e ogeneous g oup o human diseases [12,13]. In pa icula , mi ochond ial dys unc ion and oxida i e s ess ha e been documen ed in he p og ession o a ious li e dis- eases, including NAFLD and non-alcoholic s ea ohepa i is [14,15]. Mi ochond ia a e dynamic o ganelles ha p esen di e en mo phologies, anging om small sphe ical o ms o long in e connec ed ubula ne wo ks, depending on he cell ype and he physiological condi ions. In ac , hese mo phologies may be a ec ed by he al e a ion o mi ochond ial unc ion. This beha iou is gene a ed by mi ochond ial ission, which ag- men s he mi ochond ia, p oducing small sphe ical o ms, and also by usion p ocesses, which gi e ise o ubula ne wo ks o in e connec ed mi ochond ia [16,17]. Inc eased ROS le els induce mi ochond ial agmen a ion in di e en cell lines [16], an ou come ha seems o be a hallma k o apop osis; i is widesp ead in apop o ic cell dea h and blocking mi ochond ial is- sion delays apop osis [18]. Poly(ADP- ibose)polyme ases (PARPs) a e a amily o nuclea p o eins ha consume NAD + o modi y a ge p o eins. In humans, PARPs amily membe s a e encoded by 17 genes and he main membe o his amily is he enzyme poly(ADP- ibose)polyme ase-1 (PARP-1), which is esponsible o 85–90% o PARPs ac i i y (and NAD + consump ion), ollowed by PARP-2, which accoun s o he emaining 10–15% o ac i i y [19]. Only hese wo membe s o he PARPs amily ha e been implica ed in DNA damage sensing and epai ing [20]. Thus, ROS exposu e, inducing s and b eaks in DNA, inc eases he ac i a ion o PARP-1 and PARP- 2 (he ea e e e ed o, join ly, as PARPs). Ac i a ed PARPs clea e NAD + subs a es o o m poly-ADP- ibose (PAR) polyme s ha a e added o hemsel es and o o he p o ein accep o s. Many PAR-modi ied p o eins a e ela ed o DNA epai [21]. O e ac i a ion o PARPs deple es he NAD + om he cell and consequen ly induces a d ama ic educ ion in ATP le els [22]. Thus, ex ensi e PARPs ac i a ion could esul in a global ene gy ailu e, causing nec osis [23], which is implica ed in many in lamma o y diseases [24,25]. In he p esen s udy, an in i o model o human hepa ic cells exposed o sho imes o high le els o a p o-oxidan agen was used o de e mine whe he PARPs inhibi ion exe s a p o ec i e ole agains mi ochond ial dys unc ion and cell dea h. Ou esul s show ha PARPs inhibi o s inc ease he iabili y o hepa ocy es and enable he eco e y o no mal mi ochon- d ial mo phology a e he oxidan insul . Thus, PARPs inhibi ion could be a no el a ge o supp ess bo h he cell dea h associa ed wi h oxida i e s ess and he pa hological ou come o ela ed hepa ic diseases. PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 2 / 26 analysis, decision o publish, o p epa a ion o he manusc ip . Compe ing in e es s: The au ho s ha e decla ed ha no compe ing in e es s exis . Ma e ials and me hods Cell cul u e and ea men s WRL68 human li e cells (hepa ic cells) we e pu chased by he Scien i ic Ins umen a ion Se - ices (CIC) o he Uni e si y o G anada om he Eu opean Collec ion o Cell Cul u es (ca # 89121403). Hepa ic cells we e main ained as an adhe en cul u ed monolaye in DMEM (Sigma, S . Louis, MO) added wi h 10% oe al bo ine se um (Sigma), L-glu amine solu ion (2 mM; Sigma) con aining s ep omycin (100 μg/ml) and penicillin (100 U/ml), and incuba ed a 37˚C in 5% CO 2 . In o de o induce ROS p oduc ion and oxida i e insul , hepa ic cell cul u es we e ea ed wi h a single dose o H 2 O 2 ( om a 30% s ock solu ion, Sigma) dilu ed o concen a ions ang- ing om 0.25 o 5 mM o 0–30 min. The medium was hen eplaced wi h esh g ow h medium, and he cells we e u he incuba ed o 2, 4 o 24 h depending on he expe imen (Fig 1). PJ34 (Enzo Li e Sciences, San Diego, CA) and AG14361 (Selleck Chemicals, Hous on, TX) we e used as PARPs inhibi o s. PJ34 is a wa e -soluble cell-pe meable phenan h idinone de i - a i e which selec i ely inhibi s he ca aly ic ac i i y o PARP-1 and PARP-2 (EC 50 = 20 nM) [26]; while AG14361, a icyclic benzimidazole, is an ex emely po en inhibi o speci ic o PARP-1 (K i <5 nM) [27]. The cells we e p e-incuba ed ( his is e med he Inhibi o p e- ea - men s ep in Fig 1) wi h he PJ34 o AG14361 a a concen a ion o 1 μM o 16 h be o e he oxida i e ea men ; he same PARPs inhibi o and concen a ion we e also p esen in he medium o H 2 O 2 incuba ion and in he esh medium added pos - ea men ( e med he Pos - incuba ion ime s ep in Fig 1). To a oid NAD + deple ion, exogenous NAD + (0.25 mM, Sigma) was added o he cell cul- u e du ing he H 2 O 2 ea men and o he esh medium du ing pos - ea men . Some cell cul- u es we e ea ed wi h 4 mM o sodium py u a e (Sigma), a ROS sca enge , du ing he H 2 O 2 ea men and pos - ea men . Fig 1. Schema ic summa y o expe imen al p ocedu es. Oxida i e ea men was applied as a single dose o H 2 O 2 o 30 min. PARPs ac i i y was analysed du ing oxida i e ea men and 15 minu es a e wa ds. ROS p oduc ion was analysed du ing H 2 O 2 ea men and a e 30 min o pos -incuba ion. Cell iabili y and cha ac e isa ion o cell dea h we e de e mined a e 24 h o pos -incuba ion. Mi ochond ial mo phology was s udied a 2 h o 4 h a e H 2 O 2 exposu e. Cellula ATP con en was s udied a 2 h a e he end o he H 2 O 2 ea men . PARPs inhibi o s we e added o cul u e media 16 h be o e oxida i e ea men , du ing he ea men and du ing pos -incuba ion. Media we e supplemen ed wi h NAD + du ing H 2 O 2 ea men and du ing pos -incuba ion. h ps://doi.o g/10.1371/jou nal.pone.0187130.g001 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 3 / 26 Measu emen s o in acellula ROS The hepa ic cells we e seeded in 6-well cul u e pla es a a densi y o 2.0 x 10 5 cells/well. Be o e he H 2 O 2 ea men , he cells we e incuba ed wi h 20 μM o dichlo odihyd o luo escein diace- a e (H 2 DCFDA) o 30 min. This p obe is cell-pe meable and is hyd olysed in acellula ly o he DCFH ca boxyla e anion, which is e ained in he cell. Oxida ion o DCFH by in acellula ROS esul s in he o ma ion o he luo escen p oduc dichlo o luo escein (DCF) [10]. A e H 2 DCFDA incuba ion, he medium was emo ed and he cells we e ea ed wi h H 2 O 2 o 0–30 min (Fig 1). The cells we e hen de ached wi h ypsin-EDTA solu ion (Sigma), cen i- uged a 300 g and washed wi h ice-cold PBS. Finally, he luo escen s aining was de ec ed by low cy ome y analysis in a Bec on Dickinson FACSA ia III cy ome e using FACSDi a so - wa e (BD Biosciences, E embodegem, Belgium). The numbe o luo escen posi i e-cells was exp essed as a pe cen age o he o al numbe o cells. I is impo an o no e ha DCFH does no di ec ly eac wi h H 2 O 2 o o m he luo escen p oduc DCF [10]. Immuno luo escence and PARPs ac i i y Hepa ic cells seeded on co e slips in 6-well cul u e pla es we e ea ed o 0–30 min wi h H 2 O 2 and ixed wi hin 15 min wi h ice-cold me hanol-ace one 1:1 o 10 min o de e mine he ime- cou se o maximum ac i i y o PARPs (Fig 1). The co e slips we e incuba ed o e nigh a 4˚C wi h he p ima y mouse monoclonal an i-PAR an ibody. Two an ibodies we e indis inc ly used o hese immuno luo escence s udies: ca # 4335-MC-100 (T e igen, Gai he sbu g, MD) and ca # ALX-804-220-R100 (Enzo Li e Sciences, San Diego, CA) [28], bo h a a dilu ion o 1:400. Then, he co e slips we e incuba ed wi h he seconda y an ibody (Alexa luo 488-con- juga ed goa an i-mouse IgG; Molecula P obes, Eugene, OR; ca # A-11001; dilu ion 1:1000) a oom empe a u e o 2 h and cell nuclei we e coun e s ained wi h he dye Hoechs 33342 (Sigma) o 2 min. Finally, he co e slips we e moun ed on slides using Fluo omoun G moun ing medium (Sou he n Bio ech, Bi mingham, AL), and analysed using an Axiopho mic oscope (Zeiss, Obe kochen, Ge many). To de e mine he le el o PARPs ac i a ion, he numbe o PAR-posi i e nuclei de ec ed a 400X magni ica ion was coun ed in a o al o 200 nuclei o each condi ion, om h ee sepa a e expe imen s. De e mina ion o cell iabili y and cell dea h Cell iabili y was de e mined using he MTT (3-[4,5-dime hyl hiazol-2-yl]-2,5-diphenyl e azo- lium b omide, Sigma) me hod [29]. The cells we e seeded in 96-well cul u e pla es a an ini ial densi y o 8.0 x 10 3 cells/well. A e 30 min o H 2 O 2 ea men , he cells we e incuba ed o 24 h (pos -incuba ion ime) in esh medium o eco e y, and 5 mg/ml MTT was hen added o 3 h o measu e cell iabili y (Fig 1). A e his, he cul u e medium was emo ed and 100 μl/well o DMSO (Sigma) we e added o dissol e he o mazan c ys als. The abso bance o each well a 595 nm was measu ed in a mic opla e spec opho ome e eade (Mul iskan Ascen , The mo Scien i ic, Rock o d, lL). Cell iabili y in each well was exp essed as a pe cen age o he abso - bance o each expe imen al condi ion in ela ion o he con ol wells (un ea ed cells), assum- ing ha un ea ed cells ep esen ed 100% iabili y. Apop o ic cell dea h was measu ed using an Annexin-V-FITC ki wi h p opidium iodide solu ion (InmunoS ep, Salamanca, Spain; ca # ANXVF-200T) o de ec phospha idylse ine exposu e [30]. Hepa ic cells, seeded in 6-well cul u e pla es, we e ea ed o 30 min wi h H 2 O 2 , and a e 24 h o pos -incuba ion ime (Fig 1) we e de ached, labelled ollowing he manu ac u e ’s p o ocol and analysed wi h a Bec on Dickinson FACSA ia III cy ome e using FACSDi a so wa e (BD Biosciences). Cells ea ed wi h 6 μM o s au ospo ine (Sigma) o 24 h se ed as posi i e con ols o apop osis [31]. PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 4 / 26 Nec o ic cell dea h was e alua ed by examining he elease o lac a e dehyd ogenase (LDH) [32]. In his me hod, he cells we e seeded in 96-well cul u e pla es a an ini ial densi y o 12.0 x 10 3 cells/well and cul u ed o e nigh a 37˚C be o e H 2 O 2 ea men (Fig 1); a e 24 h o pos -incuba ion, LDH elease was assessed by measu ing LDH ac i i y in cul u e medium using a LDH Cy o oxici y Assay Ki (The mo Scien i ic; ca # 88953) acco ding o he manu- ac u e ’s ins uc ions. The abso bance (Abs) a 490 nm o cell-cul u e medium om each expe imen al condi ion (Abs exp) was measu ed in he mic opla e spec opho ome e eade (Mul iskan Ascen , The mo Scien i ic); o de e mine he Abs a ia ion due o he expe imen al condi ions, he backg ound Abs alues we e sub ac ed in each Abs exp. Some cul u es we e ea ed wi h 1% T i on (Sigma) o 10 min as a posi i e con ol o nec osis: abso bance co e- sponding o he spon aneous LDH elease (Abs spon ) was de e mined in un ea ed cells, while ha o he T i on- ea ed cells co esponded o he maximum LDH elease (Abs max). The pe cen ages (%) o LDH elease we e de e mined as LDH elease (%) = (Abs exp—Abs spon ) / (Abs max–Abs spon ). Mi ochond ial mo phology quan i ica ion and ul as uc u al s udy Changes in mi ochond ial mo phology and ul as uc u e we e s udied in hepa ic cells seeded on co e slips in 6-well cul u e pla es subjec ed o 30 min o hyd ogen pe oxide ea men , and pos -incuba ed in esh medium o 2 o 4 h (Fig 1). Two s a egies we e employed in his s udy: i s , he luo escen mi ochond ial p obe Mi o- acke Red CMXROS (Cell Signaling Technology, Dan e s, MA) was added a a concen a- ion o 125 nM o he cells du ing he las 45 min o pos -incuba ion. Then, he cells we e ixed wi h ice-cold me hanol-ace one 1:1 o 10 min, moun ed on slides using Fluo omoun G (Sou he nBio ech), and analysed wi h a con ocal Leica TCS-SP mic oscope (Leica, We zla , Ge many). The mi ochond ia we e analysed and classi ied using Mic oP so wa e, a use ul ool ha has been used p e iously o s udy mi ochond ial agmen a ion [33]. Two majo ypes o mi ochond ia we e conside ed, in e ms o he mo phology obse ed: a) Type I, small mi ochond ia wi h globula mo phology; b) Type II, mi ochond ia wi h ubula mo phologies (including b anched, wis ed o s aigh ubules). App oxima ely 7,000 mi ochond ia we e analysed in 10 high-powe ields pe condi ion in each sepa a e expe imen a 600x magni ica- ion. The elonga ion index and he mi ochond iala ea we e also measu ed wi h Mic oP so - wa e. The elonga ion index was calcula ed as he leng h o he mi ochond ia di ided by hei wid h. A la ge elonga ion index co esponded o ubula mi ochond ia mo phology. In he second s a egy, he cells we e p ocessed o ansmission elec on mic oscopy (TEM) by ixing hem in 2% glu a aldehyde in 0.05 M cacodyla e bu e (pH 7.4) supple- men ed wi h 2 mM Cl 2 Mg and 0.03 g/L suc ose o 2 h, pos ixed in 1% osmium e oxide o 1 h, dehyd a ed in g aded se ies o e hanol, and embedded in epoxy esin. Ul a hin sec ions (50–70 nm) we e moun ed on coppe g ids and examined unde a Zeiss Lib a 120 EDX elec- on mic oscope (Zeiss, Obe kochen, Ge many). Cellula ATP assay WRL68 cells we e seeded in a 25 cm 2 lask a a densi y o 1.0 x 10 6 cells/ lask and allowed o g ow o 24 h. The cells we e hen ea ed wi h H 2 O 2 o 30 min and pos -incuba ed o 2 h in esh medium (Fig 1). A e his, he cells we e eco e ed, cen i uged a 300 g o 10 min and washed wi h PBS. In acellula ATP was ex ac ed using he boiling wa e p ocedu e desc ibed by Yang and cowo ke s [34]: 400 μl o boiling wa e was added o each cellula pelle , which was subsequen ly o exed and incuba ed o 5 min; he esul ing suspension was cen i uged a 12000 g o 5 min. ATP con en in he supe na an was de e mined using he Adenosine 5’- PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 5 / 26 iphospha e (ATP) Bioluminescen Assay Ki (Sigma, ca # FLAA-1KT) acco ding o he man- u ac u e ’s ins uc ions. Luminescence was measu ed om 96-well pla es using a mic opla e eade (TRIAD se ies mul imode de ec o , Dynex Technologies, Chan illy, VA) and he quan- i y o ATP was calcula ed using a s anda d cu e. P o eins we e measu ed using he B ad o d me hod (Bio ad). ATP con en was exp essed as nmol pe mg o o al p o ein. S a is ical analysis Da a a e exp essed as mean ±SEM om a leas h ee independen expe imen s. S uden ’s - es was used o de e mine signi ican di e ences. S a is ical analyses we e pe o med using IBM-SPSS S a is ics e sion 19.0 so wa e (IBM Co p., A monk, NY). A alue o P<0.05 was conside ed s a is ically signi ican . Resul s A single sho exposu e o an oxidizing agen educe hepa ic cell iabili y The esponse o hepa ic cells o expe imen al oxida i e s ess was assessed by de e mining he iabili y o WRL68 cells a 24 h a e ea men o 30 min wi h a single dose o H 2 O 2 anging om 0.25 o 5 mM. The MTT assay showed ha cell iabili y dec eased in a dose-dependen manne a e oxida i e ea men . Pa icula ly, concen a ions o less han 2.5 mM o H 2 O 2 educed cell iabili y by 30–40%, compa ed o he con ol (Fig 2A; 29.4 ±2.7%, 32.8 ±2.8% and 37.7 ±1.1% educ ion o 0.25, 0.5 and 1 mM, espec i ely). Howe e , ea men s a a con- cen a ion exceeding 2.5 mM o H 2 O 2 p o oked a dec ease in cell su i al o o e 60% (Fig 2A; 63.2 ±2.5% and 90.3 ±0.2% educ ion o 3.5 and 5 mM, espec i ely; 43.4 ±4.4% educ- ion o 2.5 mM). To e i y he gene a ion o ROS in hepa ic cells du ing H 2 O 2 ea men , we analysed he con e sion o H 2 DCFDA o DCF (see Me hods). Time-cou se analysis o his oxi- da i e ea men showed ha he la ges amoun s o cells showing in acellula ROS we e de ec ed as ea ly as 15 min a e incuba ion wi h 3.5 mM o H 2 O 2 and ha hese p opo ions o labelled cells we e main ained un il 30 min o H 2 O 2 incuba ion, when he oxida i e ea - men concluded (Fig 2B). The numbe o cells wi h in acellula ROS d ama ically dec eased jus a ew minu es a e he oxida i e ea men . I is o no e ha he signi ican inc ease in cells wi h in acellula ROS was seen only a concen a ions o H 2 O 2 equal o o highe han 2.5 mM (Fig 2C; 84.6 ±7.6% and 96.8 ±4.7% inc ease o 2.5 and 3.5 mM, espec i ely). The e- o e, he inc eased p oduc ion o ROS obse ed a e ea men wi h H 2 O 2 (concen a ion 2.5 mM) co ela es wi h a s ong dec ease in cell iabili y wi hin 24 h. ROS induced by oxida i e ea men ac i a es PARPs Oxida i e s ess can ac i a e PARPs due o i s abili y o cause DNA b eaks [35]. The e o e, we s udied he le els o PARPs ac i a ion, measu ed by PAR polyme o ma ion using immuno- luo escence analysis, when ROS we e p esen (Fig 3). We con i med ha mos PAR polyme o ma ion had aken place a 15 min om he beginning o H 2 O 2 (Fig 3A), coinciding wi h he p e iously desc ibed ea ly in acellula ROS p oduc ion; hence, PARPs ac i a ion is an ea ly e en in hese oxida i e ea men . Pa icula ly, PARPs ac i a ion inc eased as H 2 O 2 concen- a ion augmen ed (Fig 3B and 3C; 20.7 ±5.5%, 22.0 ±4.6%, 42.3 ±3.5%, 45.3 ±4.7%, 50.6 ± 4.0% and 72.6 ±5.7% inc ease in PAR-posi i e cells o 0.25, 0.5, 1, 2.5, 3.5 and 5 mM H 2 O 2 ea men compa ed o un ea ed con ol cells, espec i ely). So, ea men wi h H 2 O 2 3.5 mM inc eased ea ly ROS p oduc ion (96.8 ±4.7% inc ease; Fig 2C), PARPs ac i a ion (50.6 ±4.0% inc ease in PAR-posi i e cells; Fig 3) and s ongly educed cell iabili y a 24 h a e he ea - men (63.2 ±2.5% educ ion; Fig 2A). PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 6 / 26 Fig 2. Dec eased cell iabili y and ROS p oduc ion induced by oxida i e ea men . (A) Twen y- ou hou s a e exposu e o di e se concen a ions o H 2 O 2 o 30 min, he iabili y o WRL68 cells was analysed wi h he MTT assay. Cell iabili y was exp essed as he pe cen age o iable cells wi h espec o hose de e mined in he con ols (H 2 O 2 non- ea ed cells), which was conside ed o be 100%. Ba g aph shows he mean ±SEM o h ee independen expe imen s. Signi ican di e ences wi h espec o he con ols (H 2 O 2 non- ea ed cells): *P<0.05, **P<0.01 and ***P<0.001. (B) ROS p oduc ion by WRL68 cells was e alua ed by incuba ing he cells wi h H 2 DCFDA o 30 min, be o e exposing hem o H 2 O 2 . The le el o DCF luo escence was e alua ed by low cy ome y a 0, 5, 15 and 30 min du ing exposu e and a 15 and 30 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 7 / 26 PARPs inhibi o s inc ease cell iabili y and dec ease he a e o cell dea h a e oxida i e ea men We show abo e ha cell iabili y s ongly dec eases wi h ROS induc ion (Fig 2) and PARPs ac i a ion (Fig 3) a e 3.5 mM o ea men wi h H 2 O 2 . The e o e, we will use his concen a- ion o H 2 O 2 o s udy he ole o PARPs ac i a ion on he hepa ocy es su e ing an oxida i e insul . In i s place we de e mined whe he PARPs inhibi o s would ab oga e he dec ease in cell iabili y (Fig 4A). To do so, we used AG14361 (a po en and selec i e inhibi o o PARP-1 ac i i y), which is known o inhibi o e 90% o PARP-1 ac i i y in human cells [36], and PJ34 (an inhibi o o PARP-1 and PARP-2 ac i i y) a 1 μM concen a ion; highe concen a ions we e ound o be cy o oxic in his cell line (S1 Fig). Bo h PARPs inhibi o s p oduced a signi i- can inc ease in he a e o cell su i al a e H 2 O 2 ea men , wi h 27.3 ±6.1% and 32.7 ±5.5% inc ease in cell iabili y o PJ34 and AG14361 espec i ely (Fig 4A). Immuno luo escence echniques showed ha PAR polyme o ma ion was s ongly diminished in hepa ic cells ea ed wi h H 2 O 2 and PARPs inhibi o s (S2 Fig). We specula ed ha PARPs inhibi o s migh inc ease cell su i al a es due o he di ec neu alisa ion o ROS o , al e na i ely, by p e en ing he deple ion o NAD + and hence he educ ion in ATP, as a esul o PARPs ac i a ion. To decide his ques ion, we i s examined whe he ROS we e o med in he p esence o PARPs inhibi o s. We ound ha ROS con inued o be p oduced a no mal le els despi e he p esence o he inhibi o s o PARPs ac i i y (Fig 4B); he e o e bo h PJ34 and AG14361 ha e no sca enge capaci y pe se and, hence, hei e ec in eco e ing cell iabili y a e oxida i e damage is de i ed exclusi ely om he inhibi- ion o PARPs ac i i y. As a second app oach, we conside ed whe he NAD + supplemen a ion p e en ed a dec ease in cell iabili y a e he oxida i e damage. To de e mine his poin , hepa ic cells we e exposed o 3.5 mM H 2 O 2 o 30 min, oge he wi h NAD + incuba ion (0.25 mM), ollowed by 24 h o NAD + pos -incuba ion. NAD + supplemen a ion was ound o signi - ican ly inc ease he cell su i al a e o hepa ic cells exposed o H 2 O 2 ea men (27.2 ±4.8% inc ease; Fig 4C), o a deg ee simila o ha p esen ed by he PARPs inhibi o s. The dec ease in cell iabili y o hepa ic cells exposed o H 2 O 2 ea men was mainly co e- la ed wi h a signi ican inc ease in Annexin V-posi i e cells (a ma ke o apop osis; 60.5 ± 10.9% inc ease; Fig 5A and 5B). This apop o ic iden i y was co obo a ed by using s au ospo - ine, a well-known apop osis induce [31], which ma kedly inc eased he numbe o Annexin V-posi i e cells (29.6 ±4.7% inc ease; Fig 5B). In con as , he e was no such inc ease in LDH elease, conside ed o indica e nec osis, a e H 2 O 2 ea men (6.4 ±2.5% inc ease; Fig 5C). This sugges s ha a cell dea h mechanism simila o apop osis (iden i y by he phospha idylse - ine exposu e de ec ed wi h Annexin V and he low LDH elease) may ha e occu ed in hepa ic cells when oxida i e s ess was induced. In e es ingly, he PJ34 inhibi o signi ican ly sup- p essed he inc ease in Annexin V-posi i e cells induced by H 2 O 2 (52.9 ±8.8% educ ion; Fig 5A and 5B) bu did no a ec he LDH elease a e he oxida i e insul (Fig 5C). Taken oge he , hese esul s show ha a c i ical ole is played by PARPs ac i a ion and NAD + deple ion in dec easing he su i al a es o hepa ic cells subjec ed o oxida i e s ess. min a e 30 min exposu e o H 2 O 2 ( ime poin o 45 min and 60 min, espec i ely, in he igu e). The linea g aph shows he pe cen age o luo escen cells (DCF posi i e cells) a each ime poin . Mean ±SEM o h ee independen expe imen s. Signi ican di e ences be ween each ea men and hei espec i e con ols (H 2 O 2 non- ea ed cells): *P<0.05, **P<0.01, and ***P<0.001. (C) Ba g aph showing he p opo ion o DCF- luo escen cells (indica ing ROS p oduc ion) a 15 minu es o exposu e o di e en doses o H 2 O 2 ea men . Ba s ep esen he mean ±SEM o h ee independen expe imen s. Signi ican di e ences wi h espec o H 2 O 2 non- ea ed cells: ***P<0.001. h ps://doi.o g/10.1371/jou nal.pone.0187130.g002 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 8 / 26 PARPs inhibi o s p e en he ea ly al e a ions in mi ochond ial o m induced by ROS Oxida i e s ess induced by ROS o e p oduc ion has been ela ed wi h mi ochond ial damage and cell dea h [37,38]. In ou s udy, he apid onse o ROS gene a ion ollowing exogenous Fig 3. PARPs ac i a ion in esponse o oxida i e ea men . (A) PARPs ac i i y was measu ed by de ec ing PAR p oduc ion by means o immuno luo escence. Cells we e ea ed wi h H 2 O 2 and PAR p oduc ion was de ec ed a 0, 15 and 30 min o exposu e, and 15 minu es a e 30 min o exposu e o H 2 O 2 ( ime poin o 45 min in he g aph). The linea g aph shows he pe cen age o PAR posi i e nuclei (a leas 200 cells we e coun ed o each poin and condi ion) a di e en ime poin s; he highes le el o PARPs ac i a ion was a 15 min o H 2 O 2 ea men . Mean ±SEM o h ee independen expe imen s. Signi ican di e ences be ween each H 2 O 2 ea men and hei espec i e con ols: **P<0.01 and ***P<0.001. (B) Ba g aph e ealing he e ec o di e en concen a ions o H 2 O 2 on he pe cen age o PAR posi i e nuclei de ec ed a 15 min o H 2 O 2 exposu e; a leas 200 cells pe condi ion we e coun ed. Ba s ep esen he mean ±SEM o h ee independen expe imen s. Signi ican di e ences wi h espec o H 2 O 2 non- ea ed cells: *P<0.05, **P<0.01, and ***P<0.001. (C) Rep esen a i e immuno luo escence images showing he inc ease in PAR polyme (g een) when WRL68 cells we e ea ed wi h di e en doses o H 2 O 2 o 15 min. Nuclei we e s ained wi h Hoechs (blue). Scale ba : 50 μm. h ps://doi.o g/10.1371/jou nal.pone.0187130.g003 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 9 / 26 In summa y, he abo e esul s clea ly sugges ha PARPs inhibi o s p e en he ea ly changes in mi ochond ial and ul as uc u al mo phologies associa ed wi h oxida i e s ess induced by a single sho exposu e o H 2 O 2 . Fig 9. Analysis o mi ochond ial ul as uc u e. T ansmission elec on mic oscope pho omic og aphs co esponding o hin sec ions o WRL68 cells a 4000x (A) and 8000x magni ica ion (B). Cells we e p e- incuba ed, ea ed wi h H 2 O 2 o 30 min and hen pos -incuba ed o 2 h (in he p esence o absence o PJ34 inhibi o ) be o e being analysed by elec on mic oscopy. The H 2 O 2 ea ed cells, in he absence o PJ34 inhibi o (lowe le pho omic og aphs), show mi ochond ia wi h small g anula mo phology and sca ce ans e sal c is ae, while he p esence o he inhibi o in oxida i e ea men (lowe igh pho omic og aphs) es o es he ubula -like mo phology and ans e sal c is ae dis ibu ion. Scale ba : 1 μm in (A) and 0.5 μm in (B). h ps://doi.o g/10.1371/jou nal.pone.0187130.g009 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 16 / 26 AG14361 inhibi o coun e ac s he dec ease in ATP induced by ROS Mi ochond ial agmen a ion and he appa en educ ion in mi ochond ial c is ae obse ed in H 2 O 2 ea ed cells ha e been closely ela ed o de ec s in ene gy p oduc ion and, hence, o mi ochond ial dys unc ion. In o de o de e mine he impai men o mi ochond ial unc ion, we nex analysed cellula ATP con en , using an ATP bioluminescence assay pe o med on in ac cells ea ed o 30 min wi h H 2 O 2 ei he in he absence o p esence o PARPs inhibi o s (see Me hods). A signi ican dec ease in ATP con en was obse ed a 2 h a e he H 2 O 2 ea - men (a 4.7 ±1.4 nmol/mg educ ion compa ed wi h non- ea ed cells; Fig 10A), which was signi ican ly neu alised wi h he AG14361 inhibi o (2.6 ±0.2 nmol/mg inc ease in ATP con- en compa ed wi h H 2 O 2 ea men ; Fig 10A). Conco dan ly, NAD + supplemen a ion in oxi- da i e ea men also caused an inc ease in ATP con en simila o hose obse ed o he AG14361 inhibi o (2.4 ±0.6 nmol/mg inc ease compa ed wi h H 2 O 2 ea men ; Fig 10B), sug- ges ing ha NAD + deple ion caused by PARPs ac i a ion con ibu es o mi ochond ial ene - ge ic dys unc ion. Discussion The p esen s udy in es iga es he e ec o he PARPs inhibi o s PJ34 (which inhibi s PARP-1 and PARP-2) and AG14361 (which speci ically inhibi s PARP-1) on cell su i al and dea h a e oxida i e insul . The oxida i e insul o WRL68 cells, used as an in i o hepa ic model [43], consis ed in a sho ea men wi h H 2 O 2 . As PARP-1 and PARP-2, whose o e ac i a ion p oduces he deple ion o cellula NAD + , a e in ol ed in he sensing and epai o DNA, hei e ec on cell dea h and su i al was analysed by inhibi ing hei ac i i y. The su i al o WRL68 cells ollowing H 2 O 2 ea men inc eased when he PARPs we e inhibi ed. Mo eo e , since bo h inhibi o s eco e ed simila amoun s o cell iabili y, we concluded ha his e ec Fig 10. PARPs inhibi ion o NAD + supplemen a ion inc eases he ATP cell con en a e H 2 O 2 ea men . (A) The ba g aph shows he ATP con en (in nmol pe mg o o al p o ein) ob ained in soluble ex ac s om WRL68 cells in each expe imen al condi ion. Hepa ic cells we e p e-incuba ed, ea ed wi h H 2 O 2 o 30 min and pos -incuba ed o 2 h (always in he p esence o AG14361 inhibi o ) be o e being p ocessed o ob ain he cell ex ac s. (B) NAD + was supplemen ed du ing oxida i e ea men (30 min) and du ing he pos - incuba ion ime (2h). Then, cell ex ac s we e ob ained and p ocessed o measu e ATP con en (in nmol pe mg o o al p o ein). Ba s ep esen he mean ±SEM o ou (A) and h ee (B) independen expe imen s. Signi ican di e ences: *P<0.05 and **P<0.01 e sus con ol (non- ea ed cells); # P<0.05 and ## P<0.01 e sus 3.5 mM H 2 O 2 . h ps://doi.o g/10.1371/jou nal.pone.0187130.g010 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 17 / 26 is essen ially due o he inhibi ion o PARP-1 ac i i y. In addi ion, we demons a e ha he inc eased cell su i al could be due o p e en ing he NAD + deple ion p oduced a e he PARPs ac i a ion induced by oxida i e s ess. Simila esul s ha e been ob ained in chick emb yo li e and o he ypes o cells cha ac e ised by PARP-1 o e ac i a ion and supple- men ed wi h NAD + [44–46]. As he deple ion o NAD + has been ela ed o a loss o abili y o egene a e ATP wi hin he cells, i s es o a ion o he p e en ion o i s deple ion by inhibi ing PARPs enzymes could a oid ene gy ailu e in hepa ic cells. We used millimola concen a ions o H 2 O 2 o induce oxida i e s ess in ou in i o model, which a e highe han physiological and mos pa hological concen a ions. We made his choice a e inding ha WRL68 hepa ic cells we e esis an o cell dea h induced by mic omola H 2 O 2 concen a ions. This esis ance may be ela ed o he elabo a e de oxi ica ion and an ioxidan sys ems o hepa ic cells, likely de eloped o main ain cellula homeos asis in he li e [47]. In con as , o he cell ypes a e mo e sensi i e o H 2 O 2 ea men , and concen a ions abo e 50 μM ha e been desc ibed as cy o oxic o a wide ange o animal cells in cul u e[48]. Hence, he H 2 O 2 dosage mus be adap ed o each cell ype. In summa y, he concen a ion o H 2 O 2 selec ed (3.5 mM concen a ion o 30 min) esul ed in a s ong educ ion in cell iabili y (Fig 2A), ROS accumula ion in all cells (Fig 2C), an inc ease in PARPs ac i a ion (Fig 3) and mi o- chond ial al e a ions (Fig 6). These mi ochond ial al e a ions we e simila o hose documen ed in a ious li e diseases[14,15,49,50]. The p esen s udy p esen s a possible expe imen al model o he s udy o hese changes and he ela ionships among ROS p oduc ion, PARPs ac i- a ion and mi ochond ial al e a ion. Fu he esea ch is wa an ed on he po en ial ole o PARPs in he supp ession o oxida i e s ess-associa ed cell dea h in hepa ic diseases ha de elop wi h ea ly mi ochond ial changes. The inhibi ion o PARPs is o clinical in e es . These inhibi o s ha e been desc ibed as a powe ul he apy o pa ien s wi h he edi a y b eas o o a ian cance con aining mu a ions in he BRCA-1 and BRCA-2 genes ha a e essen ial o epai ing DNA lesions by he homolo- gous ecombina ion pa hway [51,52]. The inhibi ion o PARPs in hese gene ic condi ions enhances he o ma ion o non- epai ed double s and b eaks, p oducing cy o oxic lesions ha induce cell dea h in he cance cells [53]. Addi ionally, PARPs inhibi o s ha e been ela ed o po en ial he apeu ic e ec s in non-oncological diseases [54]; in an i-in lamma o y he apy [22,24,55] hey educe he NAD + deple ion ha igge s nec o ic cell dea h and an agonise he ansc ip ion o p o-in lamma o y genes p omo ed by PARPs [24,25]. Ou indings, om using WRL68 cells (as an in i o model o hepa ic cells), co obo a e he he apeu ic possibili- ies o PARPs inhibi o s in diso de s linked o oxida i e s ess as a means o educing NAD + and ATP deple ion. The e o e, ou esul s ag ee wi h ha o o he au ho s, ha he pha maco- logical inhibi ion o PARP-1 is a p omising he apy o hepa ic cells su e ing oxida i e s ess, as well as o in lamma o y diso de s and bioene ge ic dys unc ion [56–58]. The esul o PARP-1 o e ac i a ion is cell dea h. Deple ion o NAD + and/o ATP a e PARPs ac i a ion may o igina e nec o ic cell dea h [23]. In addi ion, PAR is in ol ed in he elease o AIF om mi ochond ia o he nucleus ha induces apop osis [59]. In ou s udy, 30 min o exposu e o 3.5 mM H 2 O 2 ( e e ed o in he ex as sho -exposu e ime) caused dea h in WRL68 li e cells ha was cha ac e ised as apop osis, as he cells exp essed biochemical ma ke s o apop osis, such as phospha idylse ine exposu e on he ou e lea le o he plasma memb ane; howe e , ansloca ion o he AIF o he nucleus was no de ec ed (da a no shown). Acco ding o a s udy in which he same hepa ic cell line was ea ed o 3 h wi h 0.3 mM H 2 O 2 , he PJ34 inhibi o inc eases Ak ac i a ion, which in u n inhibi s he p oapo o ic molecule BAD [60]. I should be no ed ha bo h he la e s udy and ou own eco ded an inc ease in he iabili y (e alua ed by he MTT assay) o cells ea ed wi h a PARPs inhibi o , al hough we did no obse e a ela ion be ween he ac i a ion o Ak and cell su i al (da a no PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 18 / 26 shown). Finally, a hi d s udy [56], which used mouse p ima y hepa ocy es ea ed wi h 3 mM H 2 O 2 o 16 h (long exposu e ime), desc ibed an inc ease in nec osis ha was educed by PARPs inhibi o s. Hence, i seems ha he mechanisms implica ed in H 2 O 2 induced cell dea h in hepa ic cells depend on he concen a ion and du a ion o he oxida i e ea men . Oxida i e s ess induced by exogenous H 2 O 2 ea men leads o mul iple cell damage, p o- ducing an inc ease in ROS and in i s eac ion wi h cellula componen s such as p o ein, lipids and nucleic acids. Hence, oxida i e s ess and ROS can ha e a ious biological consequences, which include non-su i al o su i al acco ding o he ex en o damage and he cellula con- ex [61]. The esul an DNA oxida i e damage can lead o mild ac i a ion o o e ac i a ion o PARPs. In he i s ins ance, PARPs ac i a ion induced by mode a e oxida i e damage does no cause deple ion o NAD + , and he cell dea h p oduced would no depend on PARPs ac i a- ion. Howe e , excessi e oxida i e s ess can elici ex ensi e DNA damage associa ed wi h a s ong NAD + deple ion gene a ed by PARPs o e ac i a ion, and he ensuing cell dea h would be PARPs-dependen [22,62,63]. The p esen esul s a e consis en wi h his p oposi ion, gi en ha lowe doses o H 2 O 2 (0.25 mM) we e associa ed wi h low ROS gene a ion (Fig 2C), mode a e PARPs ac i a ion (Fig 3B) and no mi ochond ia al e a ion (da a no shown), p o- ducing a lesse dec ease in cell iabili y (Fig 2A), whe eas highe H 2 O 2 concen a ions (3.5 mM; Fig 2A) ma kedly educed he p opo ion o iable cells, gene a ing ele a ed ROS p o- duc ion (Fig 2C), PARPs o e ac i a ion (Fig 3B) and mi ochond ial dys unc ion (Fig 6). These da a sugges ha he dec eased cell iabili y a low H 2 O 2 concen a ions is independen o PARPs ac i a ion, while he inc eased cell dea h a highe H 2 O 2 concen a ions esul s om PARPs o e ac i a ion. The e o e, PARPs a e no in ol ed in all o he cell dea h phenomena, as indica ed by he ac ha PARPs inhibi o s did no escue all cell dea h, and PARPs-inde- penden cell dea h mechanisms would also play a ole a e oxida i e damage. Ne e heless, PARPs inhibi o s con e ed p o ec ion agains cell dea h o a subs an ial pe - cen age o cells, and esea ch is wa an ed on i s use ulness in he apies in which hepa ic cells a e exposed o high ROS concen a ions. A majo poin o ou indings is he ela ionship obse ed be ween al e a ions in mi ochon- d ial mo phology and he ene ge ic dys unc ion de ec ed a e oxida i e ea men . In i o obse a ions sugges ha mi ochond ial dys unc ion is in ima ely linked o he bioene ge ics s a us in me abolic diseases o hepa ic cells [15]. Mi ochond ial dys unc ion is ela ed o changes in he mo phology o mi ochond ia, which anges om long in e connec ed ubules o small globula o ms, depending on he physiological condi ion [17], wi h he small globula o ms being ela ed o he p ocess o mi ochond ial agmen a ion. Ou ex ensi e analysis o he mi ochond ial al e a ions induced by oxida i e s ess and ROS in hepa ic cells shows ha mi ochond ial agmen a ion and loss o c is ae a e an ea ly and p og essi e e en a e oxida- i e s ess, and ha hese al e a ions a e coun e ac ed by PARPs inhibi ion. Mi ochond ial ission lowe s espi a o y ac i i y, ATP cell con en , mi ochond ial dys unc- ion and cell dea h [16]. P e ious in i o s udies e ealed ha mi ochond ial agmen a ion is enhanced by oxida i e s ess and ROS [64–68]. In e es ingly, we show, oo, ha he mi ochon- d ial agmen a ion and ene ge ic dys unc ion (loss o ATP con en ) induced by ROS p oduc- ion a e H 2 O 2 ea men a e inhibi ed by PARPs inhibi o s in an in i o model o hepa ic cells. In sho , ROS o ma ion, PARPs ac i a ion, mi ochond ial agmen a ion and cell dea h appea o be causally ela ed. A ecen wo k using e inal cells epo ed a simila co ela ion [69]. PARPs ac i a ion appa en ly impai s mi ochond ial unc ion [70–74], bu he mechanism unde lying his e ec has no been elucida ed and se e al possibili ies ha e been p oposed. In he i s place, as he NAD + coenzyme is essen ial o glycolysis, he ica boxylic acid cycle and mi ochond ial espi a ion, i s deple ion a e PARPs ac i a ion would esul in an PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 19 / 26 ene ge ic ca as ophe wi hin he cell, and, hence, in mi ochond ial dys unc ion. Secondly, some au ho s ha e p oposed ha , in addi ion o nuclea ac i i y, he e exis s a mi ochond ial PARPs ac i i y ha could p omo e mi ochond ial dys unc ion by he poly-ADP- ibosyla ion o mi ochond ial p o eins [73,74]. Thi d, we also hypo hesise ha mi ochond ial impai men Fig 11. Scheme showing he p oposed mechanism by which PARPs inhibi o s p e en mi ochond ial agmen a ion and dec ease cell dea h induced by ROS in hepa ic cells. Reac i e oxygen species (ROS) induced by oxida i e ea men (H 2 O 2 ) s imula ed PARPs ac i a ion, gene a ing NAD + deple ion and causing an al e a ion in he NAD + /NADH pool. This al e a ion p oduces changes in mi ochond ial mo phology, esul ing in mi ochond ial agmen a ion and cell dea h as a consequence. The inhibi ion o PARPs ac i i y by PARPs inhibi o s a oid NAD + deple ion and he e o e p e en s mi ochond ial agmen a ion and cell dea h. h ps://doi.o g/10.1371/jou nal.pone.0187130.g011 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 20 / 26 a e PARPs ac i a ion may be ela ed o al e a ions in he NAD + /NADH pool, as NAD + enhances he biogenesis o mi ochond ial espi a o y chain complexes [75] in addi ion o play- ing a ole in mi ochond ial ene gy p oduc ion [76]. NAD + is educed o NADH, and NADH, in u n, enhances he impo and biogenesis o espi a o y chain complexes; he e o e, he educ ion o NADH a e NAD + deple ion would a ec he mi ochond ia. Finally, PARPs ac i- a ion-induced NAD + deple ion may in e e e wi h NAD + -dependen deace ylase si uin-3 (SIRT3), which is a membe o he si uin enzyme amily loca ed in he mi ochond ia. SIRT3 is highly exp essed in he li e and is c ucial o he main enance o mi ochond ial unc ions [77]. SIRT3 unc ion in he mi ochond ia is essen ial o egula ion o he ac i i ies o espi a- o y complexes, he an ioxidan enzyme MnSOD and o he enzymes in ol ed in egene a ing educed co ac o s in o de o main ain a p ope glu a hione edox s a us[77]. SIRT3 he e o e has a c ucial ole in ene gy p oduc ion and ROS de oxi ica ion. SIRT3 p obably compe es wi h PARPs o NAD + , and PARPs o e ac i a ion has been ound o co ela e wi h si uin ac i i y down egula ion[78].Con e sely, PARP-1 inhibi ion es o es he ac i i y o SIRT3 and a ious mi ochond ial an ioxidan enzymes[79]. The e o e, he mechanism by which PARPs inhibi- ion could p e en mi ochond ial dys unc ion and agmen a ion migh be ela ed o he pa - icipa ion o SIRT3 in main aining mi ochond ial unc ions. Fu he s udies a e necessa y o es ablish he pa icula ole played by PARPs ac i a ion in he agmen a ion o mi ochond ia and in ene ge ic dys unc ion. In conclusion, his s udy p esen s some no el and in e es ing da a ega ding he change in mi ochond ial mo phology a e PARPs inhibi ion and i s ela ionship wi h he dec ease o hepa ic cell dea h induced by ROS, which a e summa ized in Fig 11. We epo ha mi ochon- d ial agmen a ion is an ea ly e en in hepa ic cells su e ing oxida i e s ess causing cell dea h. PARPs inhibi o s educe his cell dea h, by p o ec ing agains mi ochond ial agmen a- ion and NAD + deple ion, and es o e mi ochond ial unc ion. Suppo ing in o ma ion S1 Fig. S udy o cy o oxic e ec s o PARPs inhibi o s in WRL68 cells. E ec s o PARPs inhibi o s on he cell iabili y o WRL68 cells assessed by he MTT me hod. Cell iabili y was de e mined in WRL68 cells incuba ed wi h PJ34 o AG14361. To do so, hepa ic cells we e exposed o 40 h (16 h o p e- ea men and 24 h o pos -incuba ion ime) o di e en concen- a ions (0.1, 0.5, 1, 10 mM) o PARPs inhibi o s as indica ed. Cell iabili y was exp essed as pe cen ages o he con ol, which was conside ed o be 100%. Mean ±SEM o h ee indepen- den expe imen s. Signi ican di e ences wi h espec o he con ol (non- ea ed cells): P<0.05. (TIF) S2 Fig. PARPs inhibi o s block cellula PAR polyme o ma ion. Immuno luo escence de ec ion o PAR polyme (g een) in WRL68 cells ea ed wi h H 2 O 2 o 15 min in he absence o p esence o he PJ34 inhibi o . Immuno luo escence images show ha PAR polyme o ma- ion is blocked when PARPs inhibi o s we e used in oxida i e ea men . Nuclei we e coun e - s ained wi h Hoechs (blue). Rep esen a i e images om h ee independen expe imen s. Scale ba : 50 μm. (TIF) S3 Fig. Mo phology o mi ochond ia 4 h a e H 2 O 2 ea men . (A) Pe cen age o ype I (small globula ; le g aph) and ype II ( ubula including linea , wis ed, b anched and looped o ms; igh g aph) o mi ochond ial mo phology. Cells we e ea ed wi h H 2 O 2 o 30 min and hen pos -incuba ed o 4 h p io o quan i ying mi ochond ial mo phology wi h Mic oP PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 21 / 26 so wa e. (B) Mi ochond ial elonga ion index ( ela ion be ween majo and mino axis leng hs). (C) Mi ochond ial a ea. Ba s ep esen he mean ±SEM o ou independen expe imen s; a leas 7,000 mi ochond ia we e analysed in each condi ion and in each expe imen . Signi ican di e ences wi h espec o non- ea ed cells: P<0.001. (TIF) S4 Fig. PARP inhibi ion es o es mi ochond ial mo phology 4 h a e H 2 O 2 ea men . (A) Pe cen age o ype I (small globula ; le g aph) and ype II ( ubula including linea , wis ed, b anched and looped o ms; igh g aph) o mi ochond ial mo phology p io o quan i ying mi ochond ial mo phology wi h Mic oP so wa e. WRL68 cells p e-incuba ed 16 h wi h AG14361 we e ea ed wi h H 2 O 2 o 30 min and hen pos -incuba ed o 4 h. (B) Elonga ion index o mi ochond ia. (C) A ea o mi ochond ia. A leas 7000 mi ochond ia we e analysed wi h Mic oP so wa e in each condi ion and in each expe imen . Ba s ep esen he mean ± SEM o h ee independen expe imen s. Signi ican di e ences: P<0.01 wi h espec o he con ol (non- ea ed cells); # P<0.05 wi h espec o 3.5 mM H 2 O 2 . (TIF) Acknowledgmen s We hank Concepcio ´n He na ´ndez and Da id Po cel (CIC, Uni e sidad de G anada) o hei aluable assis ance in elec onic mic oscope echniques, Rosa io Sepu ´l eda (Uni e sidad de G anada) o e ising he manusc ip and p o iding c i ical commen s, and Glenn Ha ding and Richa d Da ies o p oo eading he English-language manusc ip . Au ho Con ibu ions Concep ualiza ion: Da id Ma ı ´n-Oli a. Fo mal analysis: Sand a M. Ma ı ´n-Gue e o, Da id Ma ı ´n-Oli a. Funding acquisi ion: Jose ´A. Muñoz-Ga ´mez, Ma ı ´a-Ca men Ca asco, Julio Na ascue ´s. In es iga ion: Sand a M. Ma ı ´n-Gue e o, Jose ´A. Muñoz-Ga ´mez, Ma ı ´a-Ca men Ca asco, Ma ı ´a Ma ı ´n-Es ebane ´, Da id Ma ı ´n-Oli a. P ojec adminis a ion: Miguel A. Cuad os, Da id Ma ı ´n-Oli a. Resou ces: Jose ´A. Muñoz-Ga ´mez, Ja ie Salme o ´n. Visualiza ion: Sand a M. Ma ı ´n-Gue e o, Miguel A. Cuad os, Da id Ma ı ´n-Oli a. W i ing – o iginal d a : Sand a M. Ma ı ´n-Gue e o, Da id Ma ı ´n-Oli a. W i ing – e iew & edi ing: Miguel A. Cuad os, Julio Na ascue ´s, Da id Ma ı ´n-Oli a. Re e ences 1. A e y SV. Molecula a ge s o oxida i e s ess. Biochem J. 2011; 434(2):201–10. h ps://doi.o g/10. 1042/BJ20101695 PMID: 21309749 2. Bane jee P, Jana S, Chak abo y S, Swa naka S. In lamma ion and MMPs in alcohol-induced li e dis- eases and p o ec i e ac ion o an ioxidan s. Indian J Biochem Biophys. 2013; 50(5):377–86. PMID: 24772959 3. Gao B, Ba alle R. Alcoholic li e disease: pa hogenesis and new he apeu ic a ge s. Gas oen e ology. 2011; 141(5):1572–85. h ps://doi.o g/10.1053/j.gas o.2011.09.002 PMID: 21920463 4. Pieniazek A, Czepas J, Piasecka-Zelga J, Gwozdzinski K, Koce a-Chyla A. Oxida i e s ess induced in a li e by an icance d ugs doxo ubicin, pacli axel and doce axel. Ad Med Sci. 2013; 58(1):104–11. h ps://doi.o g/10.2478/ 10039-012-0063-1 PMID: 23612702 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 22 / 26 5. Sama ghandian S, A sha i R, Fa khondeh T. E ec o long- e m ea men o mo phine on enzymes, oxida i e s ess indices and an ioxidan s a us in male a li e . In J Clin Exp Med. 2014; 7(5):1449–53. PMID: 24995110 6. Bando I, Reus MI, And es D, Cascales M. Endogenous an ioxidan de ence sys em in a li e ollowing me cu y chlo ide o al in oxica ion. J Biochem Mol Toxicol. 2005; 19(3):154–61. h ps://doi.o g/10.1002/ jb .20067 PMID: 15977196 7. Ozgu E, Gule G, Seyhan N. Mobile phone adia ion-induced ee adical damage in he li e is inhibi ed by he an ioxidan s N-ace yl cys eine and epigalloca echin-galla e. In J Radia Biol. 2010; 86(11):935– 45. h ps://doi.o g/10.3109/09553002.2010.496029 PMID: 20807176 8. Yildi im NC, Yu ekli M, Yildi im N. In es iga ion o Some An ioxidan Enzymes Ac i i ies Depending on Ad enomedullin T ea men and Cold S ess in Ra Li e Tissue. Tu k J Biochem. 2010; 35(2):138–42. 9. Madan K, Bha dwaj P, Tha eja S, Gup a SD, Sa aya A. Oxidan s ess and an ioxidan s a us among pa ien s wi h nonalcoholic a y li e disease (NAFLD). J Clin Gas oen e ol. 2006; 40(10):930–5. h ps://doi.o g/10.1097/01.mcg.0000212608.59090.08 PMID: 17063114 10. Halliwell B, Whi eman M. Measu ing eac i e species and oxida i e damage in i o and in cell cul u e: how should you do i and wha do he esul s mean? B J Pha macol. 2004; 142(2):231–55. h ps://doi. o g/10.1038/sj.bjp.0705776 PMID: 15155533 11. Mu phy MP. How mi ochond ia p oduce eac i e oxygen species. Biochem J. 2009; 417(1):1–13. h ps://doi.o g/10.1042/BJ20081386 PMID: 19061483 12. Yu-Wai-Man P, Chinne y PF. Dys unc ional mi ochond ial main enance: wha b eaks he ci cle o li e? B ain. 2012; 135(P 1):9–11. h ps://doi.o g/10.1093/b ain/aw 352 PMID: 22271662 13. Pica d M, Wallace DC, Bu elle Y. The ise o mi ochond ia in medicine. Mi ochond ion. 2016; 30:105– 16. h ps://doi.o g/10.1016/j.mi o.2016.07.003 PMID: 27423788 14. Co ez-Pin o H, Camilo ME, Bap is a A, De Oli ei a AG, De Mou a MC. Non-alcoholic a y li e : ano he ea u e o he me abolic synd ome? Clin Nu . 1999; 18(6):353–8. h ps://doi.o g/10.1054/clnu.1999. 0047 PMID: 10634920 15. Pe ez-Ca e as M, Del Hoyo P, Ma in MA, Rubio JC, Ma in A, Cas ellano G, e al. De ec i e hepa ic mi ochond ial espi a o y chain in pa ien s wi h nonalcoholic s ea ohepa i is. Hepa ology. 2003; 38 (4):999–1007. h ps://doi.o g/10.1053/jhep.2003.50398 PMID: 14512887 16. Pica d M, Shi ihai OS, Gen il BJ, Bu elle Y. Mi ochond ial mo phology ansi ions and unc ions: implica- ions o e og ade signaling? Am J Physiol Regul In eg Comp Physiol. 2013; 304(6):R393–406. h ps://doi.o g/10.1152/ajp egu.00584.2012 PMID: 23364527 17. Wes e mann B. Bioene ge ic ole o mi ochond ial usion and ission Biochim Biophys Ac a. 2012; 1817 (10):1833–8. h ps://doi.o g/10.1016/j.bbabio.2012.02.033 PMID: 22409868 18. F ank S, Gaume B, Be gmann-Lei ne ES, Lei ne WW, Robe EG, Ca ez F, e al. The ole o dynamin- ela ed p o ein 1, a media o o mi ochond ial ission, in apop osis. De Cell. 2001; 1(4):515–25. PMID: 11703942 19. Sch eibe V, Dan ze F, Ame JC, de Mu cia G. Poly(ADP- ibose): no el unc ions o an old molecule. Na Re Mol Cell Biol. 2006; 7(7):517–28. h ps://doi.o g/10.1038/n m1963 PMID: 16829982 20. Yelamos J, Fa es J, Llacuna L, Ampu danes C, Ma in-Caballe o J. PARP-1 and PARP-2: New playe s in umou de elopmen . Am J Cance Res. 2011; 1(3):328–46. PMID: 21968702 21. De Vos M, Sch eibe V, Dan ze F. The di e se oles and clinical ele ance o PARPs in DNA damage epai : cu en s a e o he a . Biochem Pha macol. 2012; 84(2):137–46. h ps://doi.o g/10.1016/j.bcp. 2012.03.018 PMID: 22469522 22. Vi ag L, Szabo C. The he apeu ic po en ial o poly(ADP- ibose) polyme ase inhibi o s. Pha macol Re . 2002; 54(3):375–429. PMID: 12223530 23. Ha HC, Snyde SH. Poly(ADP- ibose) polyme ase is a media o o nec o ic cell dea h by ATP deple ion. P oc Na l Acad Sci U S A. 1999; 96(24):13978–82. h ps://doi.o g/10.1073/pnas.96.24.13978 PMID: 10570184 24. Ba X, Ga g NJ. Signaling Mechanism o Poly(ADP-Ribose) Polyme ase-1 (PARP-1) in In lamma o y Diseases. Am J Pa hol. 2011; 178(3):946–55. h ps://doi.o g/10.1016/j.ajpa h.2010.12.004 PMID: 21356345 25. Pe al a-Leal A, Rod iguez-Va gas JM, Aguila -Quesada R, Rod iguez MI, Lina es JL, de Almodo a MR, e al. PARP inhibi o s: New pa ne s in he he apy o cance and in lamma o y diseases. F ee Radic Biol Med. 2009; 47(1):13–26. h ps://doi.o g/10.1016/j. ee adbiomed.2009.04.008 PMID: 19362586 26. Pelliccia i R, Camaioni E, Cos an ino G, Fo men ini L, Sabba ini P, Ven u oni F, e al. On he way o selec i e PARP-2 inhibi o s. Design, syn hesis, and p elimina y e alua ion o a se ies o isoquinolinone PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 23 / 26 de i a i es. ChemMedChem. 2008; 3(6):914–23. h ps://doi.o g/10.1002/cmdc.200800010 PMID: 18409175 27. Skali zky DJ, Ma ako i s JT, Maegley KA, Ekke A, Yu XH, Hos omsky Z, e al. T icyclic benzimidazoles as po en poly(ADP- ibose) polyme ase-1 inhibi o s. J Med Chem. 2003; 46(2):210–3. h ps://doi.o g/ 10.1021/jm0255769 PMID: 12519059 28. Ma in-Oli a D, Fe e -Ma in RM, San os AM, Ca asco MC, Sie a A, Ma in-Te a JL, e al. Simul a- neous cell dea h and up egula ion o poly(ADP- ibose) polyme ase-1 exp ession in ea ly pos na al mouse e ina. In es Oph halmol Vis Sci. 2011; 52(10):7445–54. h ps://doi.o g/10.1167/io s.11-7222 PMID: 21705688 29. Be idge M. TA, McCoy K., Wang R. The Biochemical and Cellula Basis o Cell P oli e a ion Assays ha Use Te azolium Sal s. Biochemica. 1996; 4:14–9. 30. Wlodkowic D, Skomme J, Da zynkiewicz Z. Flow Cy ome y-Based Apop osis De ec ion. Me hods Mol Biol. 2009; 559:19–32. h ps://doi.o g/10.1007/978-1-60327-017-5_2 PMID: 19609746 31. Feng G, Kaplowi z N. Mechanism o s au ospo ine-induced apop osis in mu ine hepa ocy es. Am J Physiol Gas oin es Li e Physiol. 2002; 282(5):G825–34. h ps://doi.o g/10.1152/ajpgi.00467.2001 PMID: 11960779 32. Chan FK-M, Mo iwaki K, De Rosa MJ. De ec ion o nec osis by elease o lac a e dehyd ogenase ac i - i y. Me hods Mol Biol. 2013; 979:65–70. h ps://doi.o g/10.1007/978-1-62703-290-2_7 PMID: 23397389 33. Peng JY, Lin CC, Chen YJ, Kao LS, Liu YC, Chou CC, e al. Au oma ic mo phological sub yping e eals new oles o caspases in mi ochond ial dynamics. PLoS Compu Biol. 2011; 7(10):e1002212. h ps:// doi.o g/10.1371/jou nal.pcbi.1002212 PMID: 21998575 34. Yang NC, Ho WM, Chen YH, Hu ML. A con enien one-s ep ex ac ion o cellula ATP using boiling wa e o he luci e in-luci e ase assay o ATP. Anal Biochem. 2002; 306(2):323–7. h ps://doi.o g/10. 1006/abio.2002.5698 PMID: 12123672 35. Hegedus C, Vi ag L. Inpu s and ou pu s o poly(ADP- ibosyl)a ion: Rele ance o oxida i e s ess. Redox Biol. 2014; 2:978–82. h ps://doi.o g/10.1016/j. edox.2014.08.003 PMID: 25460733 36. Calab ese CR, Almassy R, Ba on S, Ba ey MA, Cal e AH, Canan-Koch S, e al. An icance chemo- sensi iza ion and adiosensi iza ion by he no el poly(ADP- ibose) polyme ase-1 inhibi o AG14361. J Na l Cance Ins . 2004; 96(1):56–67. PMID: 14709739 37. Slimen IB, Naja T, Gh am A, Dabbebi H, Ben M ad M, Abd abbah M. Reac i e oxygen species, hea s ess and oxida i e-induced mi ochond ial damage. A e iew. In J Hype he . 2014; 30(7):513–23. h ps://doi.o g/10.3109/02656736.2014.971446. PMID: 25354680 38. Guo CY, Sun L, Chen XP, Zhang DS. Oxida i e s ess, mi ochond ial damage and neu odegene a i e diseases. Neu al Regen Res. 2013; 8(21):2003–14. h ps://doi.o g/10.3969/j.issn.1673-5374.2013.21. 009 PMID: 25206509 39. Galloway CA, Yoon YS. Wha comes i s , misshape o dys unc ion? The iew om me abolic excess. J Gen Physiol. 2012; 139(6):455–63. h ps://doi.o g/10.1085/jgp.201210771 PMID: 22641640 40. Hinoi E, Taka ada T, Tsuchihashi Y, Fujimo i S, Mo iguchi N, Wang L, e al. A molecula mechanism o py u a e p o ec ion agains cy o oxici y o eac i e oxygen species in os eoblas s. Mol Pha macol. 2006; 70(3):925–35. h ps://doi.o g/10.1124/mol.106.024398 PMID: 16766717 41. Kladna A, Ma chlewicz M, Piechowska T, K uk I, Aboul-Enein HY. Reac i i y o py u ic acid and i s de i a i es owa ds eac i e oxygen species. Luminescence. 2015; 30(7):1153–8. h ps://doi.o g/10. 1002/bio.2879 PMID: 25754627 42. Wang X, Pe ez E, Liu R, Yan LJ, Malle RT, Yang SH. Py u a e p o ec s mi ochond ia om oxida i e s ess in human neu oblas oma SK-N-SH cells. B ain Res. 2007; 1132(1):1–9. h ps://doi.o g/10.1016/j. b ain es.2006.11.032 PMID: 17174285 43. Gu ie ez-Ruiz MC, Bucio L, Souza V, Gomez JJ, Campos C, Ca abez A. Exp ession o some hepa o- cy e-like unc ional p ope ies o WRL-68 cells in cul u e. In Vi o Cell De Biol Anim. 1994; 30A(6):366– 71. PMID: 7522099 44. Alano CC, Ga nie P, Ying W, Higashi Y, Kauppinen TM, Swanson RA. NAD+ deple ion is necessa y and su icien o poly(ADP- ibose) polyme ase-1-media ed neu onal dea h. The Jou nal o neu osci- ence: he o icial jou nal o he Socie y o Neu oscience. 2010; 30(8):2967–78. Epub 2010/02/26. PubMed h ps://doi.o g/10.1523/JNEUROSCI.5552-09.2010 PMID: 20181594; PubMed Cen al PMCID: PMC2864043. 45. Zhu Y, Zhao KK, Tong Y, Zhou YL, Wang YX, Zhao PQ, e al. Exogenous NAD(+) dec eases oxida i e s ess and p o ec s H2O2- ea ed RPE cells agains nec o ic dea h h ough he up- egula ion o au op- hagy. Sci Rep. 2016; 6:26322. h ps://doi.o g/10.1038/s ep26322 PMID: 27240523 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 24 / 26 46. Diani-Moo e S, Shoo s J, Singh R, Zuk JB, Ri kind AB. NAD(+) loss, a new playe in AhR biology: p e- en ion o hymus a ophy and hepa os ea osis by NAD(+) eple ion. Sci Rep. 2017; 7(1):2268. h ps:// doi.o g/10.1038/s41598-017-02332-9 PMID: 28536482 47. Li S, Tan HY, Wang N, Zhang ZJ, Lao L, Wong CW, e al. The Role o Oxida i e S ess and An ioxidan s in Li e Diseases. In J Mol Sci. 2015; 16(11):26087–124. h ps://doi.o g/10.3390/ijms161125942 PMID: 26540040 48. Halliwell B, Clemen MV, Long LH. Hyd ogen pe oxide in he human body. FEBS Le . 2000; 486(1):10– 3. PMID: 11108833 49. Man ena SK, King AL, And inga KK, Eccles on HB, Bailey SM. Mi ochond ial dys unc ion and oxida i e s ess in he pa hogenesis o alcohol- and obesi y-induced a y li e diseases. F ee Radic Biol Med. 2008; 44(7):1259–72. h ps://doi.o g/10.1016/j. ee adbiomed.2007.12.029 PMID: 18242193 50. Pessay e D, Be son A, F omen y B, Mansou i A. Mi ochond ia in s ea ohepa i is. Semin Li e Dis. 2001; 21(1):57–69. PMID: 11296697 51. Balmana J, Tung NM, Isako SJ, G ana B, Ryan PD, Sau a C, e al. Phase I ial o olapa ib in combina- ion wi h cispla in o he ea men o pa ien s wi h ad anced b eas , o a ian and o he solid umo s. Ann Oncol. 2014; 25(8):1656–63. h ps://doi.o g/10.1093/annonc/mdu187 PMID: 24827126 52. Del Con e G, Sessa C, on Moos R, Vigano L, Digena T, Loca elli A, e al. Phase I s udy o olapa ib in combina ion wi h liposomal doxo ubicin in pa ien s wi h ad anced solid umou s. B J Cance . 2014; 111(4):651–9. h ps://doi.o g/10.1038/bjc.2014.345 PMID: 25025963 53. Helleday T. The unde lying mechanism o he PARP and BRCA syn he ic le hali y: clea ing up he mis- unde s andings. Mol Oncol. 2011; 5(4):387–93. h ps://doi.o g/10.1016/j.molonc.2011.07.001 PMID: 21821475 54. Be ge NA, Besson VC, Boula es AH, Bu kle A, Chia ugi A, Cla k RS, e al. Oppo uni ies o epu pos- ing o poly(ADP- ibose) polyme ase (PARP) inhibi o s o he he apy o non-oncological diseases. B J Pha macol. 2017. h ps://doi.o g/10.1111/bph.13748 PMID: 28213892 55. Decke P, Mulle S. Modula ing poly (ADP- ibose) polyme ase ac i i y: po en ial o he p e en ion and he apy o pa hogenic si ua ions in ol ing DNA damage and oxida i e s ess. Cu Pha m Bio echnol. 2002; 3(3):275–83. PMID: 12164482 56. Mukhopadhyay P, Rajesh M, Cao Z, Ho a h B, Pa k O, Wang H, e al. Poly (ADP- ibose) polyme ase- 1 is a key media o o li e in lamma ion and ib osis. Hepa ology. 2014; 59(5):1998–2009. h ps://doi. o g/10.1002/hep.26763 PMID: 24089324 57. Huang K, Du M, Tan X, Yang L, Li X, Jiang Y, e al. PARP1-media ed PPARalpha poly(ADP- ibosyl) a ion supp esses a y acid oxida ion in non-alcoholic a y li e disease. J Hepa ol. 2016; S0168-8278 (16):30697–3. h ps://doi.o g/10.1016/j.jhep.2016.11.020 PMID: 27979751 58. Ga iani K, Ryu D, Menzies KJ, Yi HS, S ein S, Zhang H, e al. Inhibi ing poly ADP- ibosyla ion inc eases a y acid oxida ion and p o ec s agains a y li e disease. J Hepa ol. 2017; 66(1):132–41. h ps://doi. o g/10.1016/j.jhep.2016.08.024 PMID: 27663419 59. Yu S-W, And abi SA, Wang H, Kim NS, Poi ie GG, Dawson TM, e al. Apop osis-inducing ac o medi- a es poly(ADP- ibose) (PAR) polyme -induced cell dea h. P oc Na l Acad Sci U S A. 2006; 103 (48):18314–9. h ps://doi.o g/10.1073/pnas.0606528103 PMID: 17116881 60. Tapodi A, Deb eceni B, Han o K, Bogna Z, Wi mann I, Gallyas F J ., e al. Pi o al ole o Ak ac i a ion in mi ochond ial p o ec ion and cell su i al by poly(ADP- ibose)polyme ase-1 inhibi ion in oxida i e s ess. J Biol Chem. 2005; 280(42):35767–75. h ps://doi.o g/10.1074/jbc.M507075200 PMID: 16115861 61. Huang Q, Shen HM. To die o o li e: he dual ole o poly(ADP- ibose) polyme ase-1 in au ophagy and nec osis unde oxida i e s ess and DNA damage. Au ophagy. 2009; 5(2):273–6. PMID: 19139632 62. Bai P. Biology o Poly(ADP-Ribose) Polyme ases: The Fac o ums o Cell Main enance. Mol Cell. 2015; 58(6):947–58. h ps://doi.o g/10.1016/j.molcel.2015.01.034 PMID: 26091343 63. Sodhi RK, Singh N, Jaggi AS. Poly(ADP- ibose) polyme ase-1 (PARP-1) and i s he apeu ic implica- ions. Vascul Pha macol. 2010; 53(3–4):77–87. h ps://doi.o g/10.1016/j. ph.2010.06.003 PMID: 20633699 64. Ba soum MJ, Yuan H, Ge encse AA, Lio G, Kushna e a YE, G abe S, e al. Ni ic oxide-induced mi o- chond ial ission is egula ed by dynamin- ela ed GTPases in neu ons. Embo J. 2006; 25(16):3900–11. h ps://doi.o g/10.1038/sj.emboj.7601253 PMID: 16874299 65. Jend ach M, Mai S, Pohl S, Vo h M, Be ei e -Hahn J. Sho - and long- e m al e a ions o mi ochond ial mo phology, dynamics and m DNA a e ansien oxida i e s ess. Mi ochond ion. 2008; 8(4):293–304. h ps://doi.o g/10.1016/j.mi o.2008.06.001 PMID: 18602028 PARPs inhibi o s educe mi ochond ial agmen a ion and cell dea h induced by ROS PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187130 Oc obe 26, 2017 25 / 26