PARP‐1 regulates DNA repair factor availability

Resea ch A icle PARP-1 egula es DNA epai ac o a ailabili y Ma hew J Schiewe 1,2,* , Amy C Mandigo 1,2 , Nicolas Go don 1,2 , Fangjin Huang 3 , Sanchaika Gau 3 , Renée de Leeuw 1,2 , Shuang G Zhao 4 , Joseph E ans 4 , Sumin Han 4 , Theodo e Pa sons 2,5 , Ru h Bi be 6 , Pe e McCue 2,5 , Ch is ophe McNai 1,2 , Saswa i N Chand 1,2 , Ylenia Cendon-Flo ez 1,2 , Pe e Gallaghe 1,2 , Jenni e J McCann 1,2 , Nee mala Poudel Neupane 1,2 , Ayesha A Sha i 1,2 , Emanuela Dylgje i 1,2 , Lucas J B and 1,2 , Tapio Visako pi 7 , Ganesh V Raj 8 , Cos as D Lallas 2,9 , Edoua d J T abulsi 2,9 , Leona d G Gomella 2,9 , Adam P Dicke 2,10 , Wm. Ke in Kelly 2,11 , Benjamin E Leiby 2,12 , Bea ice Knudsen 3 , Felix Y Feng 13 & Ka en E Knudsen 1,2,9,10,11 Abs ac PARP-1holds majo unc ions on ch oma in, DNA damage epai and ansc ip ional egula ion, bo h o which a e ele an in he con ex o cance . He e, unbiased ansc ip ional p o iling e ealed he downs eam ansc ip ional p o ile o PARP-1enzyma ic ac i - i y. Fu he in es iga ion o he PARP-1- egula ed ansc ip ome and seconda y s a egies o assessing PARP-1ac i i y in pa ien issues e ealed ha PARP-1ac i i y was unexpec edly en iched as a unc ion o disease p og ession and was associa ed wi h poo ou come independen o DNA double-s and b eaks, sugges ing ha enhanced PARP-1ac i i y may p omo e agg essi e pheno- ypes. Mechanis ic in es iga ion e ealed ha ac i e PARP-1 se ed o enhance E2F1 ansc ip ion ac o ac i i y, and speci i- cally p omo ed E2F1-media ed induc ion o DNA epai ac o s in ol ed in homologous ecombina ion (HR). Con e sely, PARP-1 inhibi ion educed HR ac o a ailabili y and hus ac ed o induce o enhance “BRCA-ness”. These obse a ions b ing new unde - s anding o PARP-1 unc ion in cance and ha e signi ican ami i- ca ions on p edic ing PARP-1inhibi o unc ion in he clinical se ing. Keywo ds DNA epai ; E2F1; PARP; ansc ip ion Subjec Ca ego y Cance DOI 10.15252/emmm.201708816 | Recei ed 22 Decembe 2017 | Re ised 10 Oc obe 2018 | Accep ed 25 Oc obe 2018 | Published online 21 No embe 2018 EMBO Mol Med (2018)10:e8816 In oduc ion Poly(ADP- ibose) polyme ase 1 (PARP-1) is a mul i unc ional p o ein o impo ance in cance . PARP-1 is an abundan ly exp essed nuclea enzyme which uses NAD + as a subs a e o poly(ADP- ibose)-yla e (PARyla e) nuclea p o eins, including au omodi i- ca ion o PARP-1 i sel (D’Amou s e al, 1999; K ishnakuma & K aus, 2010). PARP-1 plays a key ole in se e al key biological p ocesses: eplica ion o k s abili y (B yan e al, 2009), cell dea h (Yu e al, 2002), DNA epai and genomic s abili y (Du kacz e al, 1980), elome e main enance (Beneke e al, 2008), ch oma in o ga- niza ion (Poi ie e al, 1982), and ansc ip ional egula ion (K aus & Lis, 2003; Schiewe & Knudsen, 2014). The DNA epai unc ions o PARP-1 ha e been a ge ed o an i- cance e ec s h ough use o pha macological PARP inhibi o s (PARPi; Lo d & Ashwo h, 2008), which ha e been app o ed o o a ian cance , and a e unde clinical in es iga ion in a numbe o o he umo ypes, including p os a e cance (PCa). I is hough ha PARPi a e especially e ec i e in umo s ha lack homologous ecombina ion (HR) capaci y h ough loss-o - unc ion mu a ions in BRCA1 o BRCA2, in a phenomenon e med syn he ic le hali y (McCabe e al, 2006; Lo d & Ashwo h, 2017). Howe e , clinical ial da a in BRCA1/2mu an -selec ed umo s indica e ha objec i e esponse a es a e only ~40%, sugges ing ha BRCA1/2mu a ion is no su icien o PARPi esponse (Fong e al, 2009; Audeh e al, 2010; Gelmon e al, 2011; Kaye e al, 2012; Sandhu e al, 2013; Coleman e al, 2015). Addi ionally, a ecen ly published clinical ial 1Depa men o Cance Biology, Thomas Je e son Uni e si y, Philadelphia, PA, USA 2Sidney Kimmel Cance Cen e , Thomas Je e son Uni e si y, Philadelphia, PA, USA 3Ceda s-Sinai Medical Cen e , Los Angeles, CA, USA 4Depa men o Radia ion Oncology, Uni e si y o Michigan, Ann A bo , MI, USA 5Depa men o Pa hology, Thomas Je e son Uni e si y, Philadelphia, PA, USA 6Coope Uni e si y Heal h, Camden, NJ, USA 7Uni e si y o Tampe e, Tampe e, Finland 8UT Sou hwes e n, Dallas, TX, USA 9Depa men o U ology, Thomas Je e son Uni e si y, Philadelphia, PA, USA 10 Depa men o Radia ion Oncology, Thomas Je e son Uni e si y, Philadelphia, PA, USA 11 Depa men o Medical Oncology, Thomas Je e son Uni e si y, Philadelphia, PA, USA 12 Depa men o Pha macology and Expe imen al The apeu ics, Thomas Je e son Uni e si y, Philadelphia, PA, USA 13 Depa men s o Radia ion Oncology, U ology, and Medicine, Uni e si y o Cali o nia, San F ancisco, San F ancisco, CA, USA *Co esponding au ho . Tel: +1 215 503 8574; E-mail: [email p o ec ed]u ª2018 The Au ho s. Published unde he e ms o he CC BY 4.0license EMBO Molecula Medicine 10:e8816 |2018 1o 20 Published online: No embe 21, 2018 combining PARPi and and ogen ecep o (AR)-di ec ed he apy in pa ien s wi h ad anced PCa demons a ed clinical bene i , i espec- i e o HR s a us (Cla ke e al, 2018). Fu he mo e, he TO-PARP ial (Ma eo e al, 2015) led o FDA B eak h ough S a us o pa ien s wi h BRCA2 o ATM mu an cas a ion- esis an p os a e cance (CRPC). Olapa ib esponde s we e en iched o de ec s in DNA epai genes, such as biallelic loss o BRCA2 and ATM. Howe e , while mos esponde s (14/16) in his ial we e ca ego ized as bioma ke posi i e o HR de iciency, he bioma ke sui e included single copy loss o DNA epai ac o s, as well as al e a ions o HDAC2, which is in ol ed in ansc ip ional ep ession (Roun ee e al, 2000). While hese s udies ha no all PARPi esponde s wi h PCa ha bo HR-de ec i e umo s, and no all PCa umo s ha exhibi abe an DNA epai a e PARPi esponsi e, he e is clinical e idence ha PARPi esis ance is associa ed wi h es o ed HR unc ion in mul iple umo ypes (Edwa ds e al, 2008; Ba be e al, 2013; Ch is ie e al, 2017; Kond asho a e al, 2017; Pish aian e al, 2017; Weigel e al, 2017), including PCa (Goodall e al, 2017; Quigley e al, 2017). Addi ionally, PARPi esis ance has been associa ed wi h di e en ial DNA damage esponse (DDR) ne wo k unc ioning (Jaspe s e al, 2013; Johnson e al, 2013; Gogola e al, 2018). These mechanisms o esis ance o PARPi indica e ha o hese umo s, DDR de ec s likely led o PARPi esponses. These clinical indings indica e ha u he mechanis ic unde s anding o PARP-1 unc ions is needed o de elop use ul clinical bioma ke s o esponse o PARPi. Gi en he po en ial implica ions o PARP-1-media ed unc ions in human malignancies, and he need o bioma ke s o PARPi esponse, i was impe a i e o disce n he molecula basis o PARP- 1 unc ion and ac i i y in he con ex o BRCA1/2wild- ype PCa, and de e mine he con ibu ion o PARP-1-media ed ansc ip ional e en s on umo pheno ypes. Resul s PARP-1enzyma ic ac i i y is inc eased as a unc ion o disease p og ession and is associa ed wi h poo ou come To asce ain he impac o PARP-1 unc ion on agg essi e umo beha io , PCa was u ilized as a disease sys em. In his umo ype, he ole o PARP-1 in ansc ip ional egula ion o key ansc ip ion ac o s o PCa ele ance has been demons a ed (ETS ansc ip ion ac o s and and ogen ecep o (AR); B enne e al, 2011; Schiewe e al, 2012), and AR is a key d i e o PCa ini ia ion and p og es- sion. Fu he mo e, PARPi has gene a ed p omising clinical ial da a in ad anced PCa (Ma eo e al, 2015). Ini ially, human issues om p ima y, ho mone he apy (HT)-sensi i e PCa, and me as a ic CRPC (mCRPC) we e que ied o PARP-1 enzyma ic ac i i y ia immuno- his ochemis y (IHC) o PAR (Poly(ADP- ibose), he p oduc o PARP-1 enzyma ic ac i i y; Fig 1A). PARP-1 enzyma ic ac i i y was ele a ed in mCRPC when compa ed o p ima y PCa (Fig 1B). These da a gi e con i ma ion o p edic ions om p eclinical models which showed ele a ed PARP-1 enzyma ic ac i i y in CRPC cell lines (in- cluding C4-2 and LNCaP-abl) compa ed o ho mone he apy (HT)- sensi i e cell lines (including LNCaP, LAPC4, and VCaP; Schiewe e al, 2012). To que y he impac o ele a ed PARP-1 enzyma ic ac i i y on clinical ou comes, PARP-1 ac i i y was assessed as a unc ion o p oli e a i e indices (Appendix Fig S1A) and cT s age a p ima y diagnosis (Appendix Fig S1B). No co ela ion was obse ed, indica ing ha highe PARP-1 ac i a ion s a us is no simply due o inc eased cell p oli e a ion o la ge olume umo . Fu he mo e, he e we e no co ela ions be ween PARP-1 enzyma ic ac i i y and molecula al e a ions ha a e equen in PCa, including TMPRSS2:ERG usion s a us (Appendix Fig S1C), PTEN sco e (Appendix Fig S1D), o AR copy numbe (Appendix Fig S1E). Howe e , enhanced PARP-1 ac i i y was signi ican ly associa ed wi h dec eased p og ession- ee su i al (PFS; Fig 1C). These da a indica e ha PARP-1 enzyma ic unc ion is no only ele a ed in CRPC, bu also p edic i e o PFS, which is associa ed wi h disease- speci ic mo ali y. To expand upon hese da a, mul iplexed quan i iable immuno lu- o escen IHC was pe o med on non-neoplas ic p os a e issue, p ima y PCa, and mCRPC (Fig 1D op le , highe magni ica ion a igh ). As measu ed h ough quan i ica ion o PAR immuno eac i - i y, PARP-1 enzyma ic ac i i y was ele a ed in p ima y PCa (median alue 62.03) as compa ed o non-neoplas ic p os a e issue (median alue 51.52), and highes in mCRPC issue (median alue 69.10; Fig 1D, bo om le ). Howe e , he obse ed inc ease in PARyla ion du ing disease p og ession canno be simply a ibu ed o o al PARP-1 p o ein exp ession, as he a ios o PARP-1 and PAR exp es- sion le els di e ed ac oss disease s a es (Fig 1D, bo om middle; medians o PAR alues 51.67, 54.29, and 47.81 o non-neoplas ic, p ima y PCa, and mCRPC, espec i ely). Being in ica ely in ol ed in DNA damage epai , PARP-1 enzy- ma ic ac i i y is induced by DNA damage (Du kacz e al, 1980). To de e mine whe he he ele a ed PARP-1 enzyma ic ac i i y in mCRPC obse ed abo e was associa ed wi h DNA damage epai , immuno eac i i y o cH2AX, a measu e o epai o DNA double- s and b eaks (DSBs; Podho ecka e al, 2010), was pe o med. This analysis indica ed ha PARP-1 enzyma ic ac i i y as a unc ion o disease p og ession was no associa ed wi h epai o DSBs (Fig 1D, bo om igh ; median alues 44.20, 51.80, and 46.20 o non- neoplas ic, p ima y PCa, and mCRPC, espec i ely), sugges ing ha PARP-1 ac i i y is egula ed by o he ac o s in addi ion o DNA damage. Dual assessmen o DSB epai and PARP-1 ac i i y in each specimen e ealed a posi i e co ela ion be ween PAR and cH2AX in non-neoplas ic p os a e issues ( =0.2853), and p ima y PCa issues ( =0.3573), bu his associa ion is lacking in mCRPC issues ( =0.03825; Fig 1E), u he indica ing ha ele a ed PARP-1 enzyma ic unc ion in mCRPC is no a ibu able o inc eased DNA DSB epai . Toge he , hese da a demons a e ha PARP-1 enzy- ma ic ac i i y is he e ogeneous, inc eases as a unc ion o PCa p og ession, is no associa ed wi h le els o ei he PARP-1 p o ein exp ession o o DNA damage epai in mCRPC, and may p edic poo ou come in PCa. Iden i ica ion o he PARP-1- egula ed ansc ip ome and ele ance o disease p og ession As demons a ed abo e, PARP-1 enzyma ic ac i i y is ele a ed as a unc ion o PCa p og ession independen o DNA DSB epai . As such, o he PARP-1 unc ions we e analyzed. To assess PARP-1- media ed ansc ip ional egula ion in he con ex o and ogen signaling, ho mone he apy-sensi i e (HT-sensi i e) PCa cells we e dep i ed o s e oids o 72 h, hen ea ed wi h PARP-1 inhibi ion 2o 20 EMBO Molecula Medicine 10:e8816 |2018 ª2018 The Au ho s EMBO Molecula Medicine PARP-1impinges on DDR gene exp ession Ma hew J Schiewe e al Published online: No embe 21, 2018 A C D E B Figu e 1. ª2018 The Au ho s EMBO Molecula Medicine 10:e8816 |2018 3o 20 Ma hew J Schiewe e al PARP-1impinges on DDR gene exp ession EMBO Molecula Medicine Published online: No embe 21, 2018 (o con ol) ollowed by 16 h dihyd o es os e one (DHT) s imula- ion as depic ed in Fig 2A (le ) o assess he impac o PARP-1 supp ession in he p esence and absence o AR ac i i y. As expec ed, DHT al e ed he exp ession o a la ge numbe o an- sc ip s (n=1358), and PARP-1 supp ession esul ed in di e en ial ansc ip exp ession when compa ed o DHT (n=877; Fig 2A, igh ), consis en wi h p e ious epo s ha PARPi al e s he an- sc ip ional e ec s o and ogen signaling (Schiewe e al, 2012). This was u he con i med using a p e iously cha ac e ized se o AR/ DHT- esponsi e a ge genes, he majo i y o hese genes a e oppo- si ely egula ed by DHT and PARPi in LNCaP (Appendix Fig S2A). PARP-1 has also been ound o egula e cas a ion- esis an AR unc- ion (Schiewe e al, 2012). To assess he o e all ansc ip ional e ec s o PARP-1 in an unbiased manne in he con ex o CRPC, C4-2 cells we e dep i ed o s e oids o 72 h, hen we e ei he ea ed PARPi (o con ol) as depic ed in Fig 2B (le ) o 16 h. In o al, 2011 ansc ip s we e di e en ially egula ed upon PARPi ea men when compa ed o con ol in CRPC cells, hus de ining a PARP-1- egula ed ansc ip ome in CRPC. The o e lap o di e en- ially egula ed genes in HT-sensi i e s. CRPC cells was de i ed, and he da a indica e he e a e bo h o e lapping and dis inc an- sc ip ional changes elici ed by each condi ion and in he indi idual cell lines (Appendix Fig S2B). Gene lis s a e included in Da ase EV1. These da a indica e ha he e may be a co e ansc ip ional p og am egula ed by PARP-1 in PCa cells, which includes a la ge numbe o DHT- esponsi e genes (n=169), bu he ansi ion o cas a ion esis ance likely expands he ele ance o PARP-1- egu- la ed ansc ip ion, gi en he la ge numbe o ansc ip s ha a e al e ed upon PARPi (n=1,810 unique genes egula ed by PARP-1). Impo an ly, he ansc ip s associa ed wi h ac i e PARP-1 (down- egula ed by PARPi) in bo h HT-sensi i e and CRPC cells signi i- can ly inc eased in exp ession om benign issues, o p ima y PCa, o PCa me as ases (Fig 2C) when hese ansc ip s we e que ied agains a publically a ailable da a se (G asso e al, 2012). Fu he - mo e, hese da a we e alida ed using o he publically a ailable da a se s (Lapoin e e al, 2004; Taylo e al, 2010; Yu e al, 2007; Appendix Fig S3), hus indica ing ha he PARP-1- esponsi e an- sc ip ome is ele a ed as a unc ion o PCa p og ession. Toge he wi h immunohis ochemical PARP-1 ac i i y assessmen (Fig 1), hese collec i e da a indica e ha bo h PARP-1 enzyma ic ac i i y and PARP-1-sensi i e ansc ip ional e en s a e enhanced as a unc- ion o disease p og ession. PARP-1 egula es p o-oncogenic ansc ip ion ac o signaling To assess he po en ial biological consequences o he obse ed ansc ip ional enhancemen o PARP-1, Gene Se En ichmen Anal- ysis (GSEA; Moo ha e al, 2003; Sub amanian e al, 2005) Molecula Signa u es Da abase (MSigDB) analyses we e pe o med using he unbiased da a gene a ed as desc ibed abo e. U ilizing he gene aliz- able KEG MSigDB demons a ed an en ichmen o cell cycle- ela ed and DNA damage epai -associa ed pa hways (including homolo- gous ecombina ion; Fig 3A, le ). Analyses using he mo e speci ic Hallma ks MSigDB con i med p e ious s udies, in ha he And ogen Response hallma k was en iched in and supp essed in CRPC cells (NES =2.54; Fig 3A, igh bo om). The s a is ically highes en iched MSigDB hallma k was E2F Ta ge s (HT-sensi i e NES =1.51, CRPC NES =3.31; Fig 3A, igh op), which has canonical oles in he egula ion o bo h he cell cycle and DNA damage epai (Biswas & Johnson, 2012). These da a indica e ha in addi ion o playing a key ole in AR ansc ip ional ac i i y, PARP-1 ansc ip ionally egula es p ocesses associa ed wi h he cell cycle and DNA damage epai . The E2F amily o ansc ip ion ac o s egula e c i ical p ocesses o impo ance in cance , including cell cycle egula ion, DNA epai (Biswas & Johnson, 2012), mi ochond ial unc ion (Go o e al, 2006), cell dea h (Polage & Ginsbe g, 2009), umo p og ession and me as a ic de elopmen (Alla e al, 2010), s emness (Chen e al, 2008, 2009), and angiogenesis (Qin e al, 2006). E2F1 is equen ly de egula ed in PCa (Sha ma e al, 2010), and de egula ed E2F1 ac i i y is associa ed wi h agg essi e disease (McNai e al, in p ess, JCI). Fo alida ion, bo h HT-sensi i e and CRPC cells we e ea ed as depic ed in Fig 2A and B abo e, RNA was ex ac ed, and subjec ed o qPCR o canonical E2F1 a ge genes (E2F1,PCNA, MCM7, and CCNA2). As shown, each o hese ansc ip s was dimin- ished by ea men wi h he PARPi elipa ib by 40–60% in bo h he con ex o HT-sensi i e (Fig 3B, op) and CRPC cells (Fig 3B, bo om). Con i ma ion ha hese genes a e E2F1 a ge genes was conduc ed by ansien ly knocking down E2F1, and subsequen gene exp ession analyses (Appendix Fig S4A). To explo e he impac o exogenous E2F1 exp ession on PARP-1- egula ed E2F1 ac i i y, models o exogenous E2F1 we e gene a ed. Upon examina ion o E2F1 a ge gene exp ession a e PARP inhibi ion (Appendix Fig S4B), i was de e mined ha E2F1 a ge gene exp ession is no longe unde he con ol o PARP-1. These da a indica e ha ◀Figu e 1. PARP-1enzyma ic ac i i y is inc eased as a unc ion o disease p og ession and is associa ed wi h poo ou come. A Tissue mic oa ays (TMAs) om p ima y PCa (n=132) and CRPC (n=148) we e s ained ia immunohis ochemis y o poly(ADP- ibose; PAR), and sco ed by a clinical pa hologis (T. Pa sons) o in ensi y (0–3) and pe cen age (0–3). B PAR sco e was gene a ed ia he equa ion: (in ensi y ×1) + (pe cen age ×2). PAR sco es we e compa ed be ween p ima y and CRPC. ****P alue <0.0001 by Chi-squa e es . C Manual PAR sco es we e di ided in o qua iles and hen we e compa ed o p og ession- ee su i al in he CRPC TMAs. *P<0.05, ns = no s a is ically signi ican by Log- ank (Man el-Cox). 1 s qua ile s. 2 nd qua ile, P=0.1482;1 s qua ile s. 3 d qua ile, P=0.5794;1 s qua ile s. 4 h qua ile, P=0.0160;2 nd qua ile s. 3 d qua ile, P=0.3869;2 nd qua ile s. 4 h qua ile, P=0.2110;3 d s. 4 h qua ile, P=0.0201.1 s qua ile (n=24); 2 nd qua ile (n=22); 3 d qua ile (n=27); 4 h qua ile (n=26). D Top le : Rep esen a i e image o one TMA co e a e mul iplex luo escen IHC o cH2AX (g een), PAR ( ed), PARP-1(pu ple), wi h DNA (blue). Top igh : Inse s o pa en image on he le . Numbe s abo e inse columns coincide wi h numbe s on image a le ha we e chosen o u he magni ica ion and ep esen a ion (boxed a eas). Bo om le : Pe cen posi i e s aining o PAR o he en i e y o each TMA coho . Bo om middle: Pe cen posi i e s aining o PARP-1. Bo om igh : cH2AX o he en i e y o each TMA coho . Da a we e conside ed a e a median in ensi y cu o and analyzed o s a is ical signi icance using wo- ailed S uden ’s - es o PAR, PARP-1, and cH2AX, espec i ely. Exac P alues a e indica ed. Ho izon al lines a e median. Box limi s a e 25% and 75% pe cen iles, and whiske s a e min o max. E Two- ailed Spea man co ela ion es be ween PAR and cH2AX (% posi i e wi h a median in ensi y cu o ). Exac P alues a e indica ed when a ailable. Sou ce da a a e a ailable online o his igu e. 4o 20 EMBO Molecula Medicine 10:e8816 |2018 ª2018 The Au ho s EMBO Molecula Medicine PARP-1impinges on DDR gene exp ession Ma hew J Schiewe e al Published online: No embe 21, 2018 exogenous exp ession o E2F1 esul s in loss o E2F1 egula ion by PARP-1. As such, ampli ied E2F1 may se e as exclusion c i e ia in u u e clinical in es iga ion o PARPi in PCa. These da a indica e ha canonical E2F1 a ge gene exp ession is sensi i e o PARP-1 unc ion. To assess he impac o PARP-1 on E2F1 unc ion, ch oma in immunop ecipi a ion (ChIP) analyses we e pe o med. In condi ions ha we e iden ical o hose u ilized o he ansc ip ome analyses in Fig 2, hese ChIP analyses indica e ha PARP-1 supp ession esul ed in diminished E2F1 a he E2F1 locus by ~40% (Fig 3C, op le ). This is impo an , gi en ha E2F1 is a egula o o E2F1 gene exp ession. Addi ionally, PARP-1 was ound a he E2F1 locus, and PARP-1 esidency a his locus was educed ~50% in esponse o PARPi (Fig 3C, op igh ). Fu he mo e, RNA polyme ase II esi- dency was educed by ~50%, as was he ac i e ansc ip ional ma k, ace yla ed his one H4 by ~66% (Fig 3C, bo om). These da a indica e ha PARP-1 enzyma ic ac i i y is in ol ed in he biochemi- cal egula ion o E2F1 ansc ip ional unc ion on ch oma in. A B C Figu e 2. Iden i ica ion o he PARP-1- egula ed ansc ip ome and ele ance o disease p og ession. A Le : Schema ic ep esen ing he condi ions u ilized o ansc ip omic analyses (n=2) o HT-sensi i e LNCaP cells. Cells we e dep i ed o ho mones o 72 h, ollowed by ei he ea men wi h 2.5lM elipa ib (PARPi) o ehicle con ol (DMSO) o 1h, hen subsequen ly ea ed wi h ei he 1nM DHT o ehicle con ol (E OH) o 16 h. Middle: Immunoblo wi h he indica ed an ise a. Righ : Volcano plo s o ansc ip s ound o be di e en ially egula ed by DHT s. E OH (le ) o DHT s. PARPi ollowed by DHT ( igh ). Red do s indica e ansc ip s ha we e bo h s a is ically signi ican ly al e ed (P<0.05) and mo e han 1.5- old changed. B Le : Schema ic ep esen ing he condi ions u ilized o ansc ip omic analyses (n=2) o CRPC C4-2cells. Cells we e dep i ed o ho mones o 72 h, ollowed by ei he ea men wi h 2.5lM elipa ib (PARPi) o ehicle con ol (DMSO) o 16 h. Middle: Immunoblo wi h he indica ed an ise a. Righ : Volcano plo s o ansc ip s ound o be di e en ially egula ed PARPi s. ehicle con ol. Red do s indica e ansc ip s ha we e bo h s a is ically signi ican ly al e ed (P<0.05) and mo e han 1.5- old changed. C Genes ound o be down- egula ed by PARPi as desc ibed abo e (P alue <0.05,1.5- old change) in ei he HT-sensi i e cells (le ) o CRPC cells ( igh ) we e que ied agains he exp ession o hese genes in he G asso e al da a se in Oncomine. Benign = g ay, p ima y PCa = blue, me as ases = o ange. Boxplo was gene a ed using he mean exp ession o he PARPi down- egula ed genes in he indica ed da a se s. S a is ical signi icance de e mined by wo- ailed S uden ’s - es . Box plo s a e median and uppe and lowe qua iles. Whiske s a e min and max. Fo he G asso e al da a se , n=28 benign p os a e issues, n=59 localized p os a e cance , and n=35 me as a ic cas a ion esis an . ª2018 The Au ho s EMBO Molecula Medicine 10:e8816 |2018 5o 20 Ma hew J Schiewe e al PARP-1impinges on DDR gene exp ession EMBO Molecula Medicine Published online: No embe 21, 2018 A BC DE Figu e 3. 6o 20 EMBO Molecula Medicine 10:e8816 |2018 ª2018 The Au ho s EMBO Molecula Medicine PARP-1impinges on DDR gene exp ession Ma hew J Schiewe e al Published online: No embe 21, 2018 To assess he impac o PARP-1 on E2F1 unc ion in i o, CRPC (C4-2) xenog a s we e gene a ed in cas a ed, immunocom- p omised mice. Tumo -bea ing mice we e hen ea ed wi h he PARPi elipa ib o 72 h, sac i iced, and umo s we e excised. As shown, he exp ession o canonical E2F1 a ge genes (E2F1,PCNA, MCM7, and CCNA2) was diminished in i o upon PARP-1 supp es- sion (Fig 3D). To u he alida e hese indings, human issues we e u ilized o an explan p o ocol ha has been p e iously desc ibed (Cen ene a e al, 2012, 2013; Schiewe e al, 2012; Coms ock e al, 2013; Goodwin e al, 2015; de Leeuw e al, 2015; Ha sough e al, 2018). B ie ly, esh human PCa samples a e ob ained a he ime o su gical esec ion, subdi ided, and cul u ed ex i o unde condi ions ha e ain he glandula a chi ec u e, s o- mal con en , and clinicopa hologic ea u es o he o iginal umo . Explan s we e exposed o PARPi (o con ol), and he exp ession o canonical E2F1 a ge genes (E2F1,PCNA,MCM7, and CCNA2) was assessed. As shown, he esponse was he e ogeneous, bu hese pa ien issues demons a ed signi ican ly diminished E2F1 a ge gene exp ession in esponse o PARPi (Fig 3E). These collec i e da a iden i y PARP-1 as a majo e ec o o E2F1 unc ion in i o, in i o, and in human PCa issues. PARP-1e ec s on E2F signaling a e independen o cell cycle phase and dis inc om hose elici ed by CDK4/6inhibi ion To assess he impac o cell cycle phase on PARP-1-media ed E2F1 egula ion, HT-sensi i e and CRPC cells we e ea ed using condi ions iden ical o hose desc ibed in Fig 2, and subjec ed o a B dU pulse and FACS analyses. As shown, he e was no change in DNA eplica ion a an ea ly ime poin (3 h) o a he ime poin a which he ansc ip ional e ec s o PARP-1 we e assessed (16 h; Fig 4A), indica ing ha cell cycle phase canno explain he dec ease in E2F1 unc ion a e PARP-1 supp ession, al hough a la e ime poin s, DNA eplica ion is diminished upon PARPi. While E2F1 i sel canno cu en ly be he apeu ically a ge ed, he ups eam kinases ha posi i ely egula e E2F1 unc ion (cyclin- dependen kinases 4 and 6, CDK4/6) can be inhibi ed (O’Lea y e al, 2016), and CDK4/6 inhibi o s (CDK4/6i) a e unde clinical in es iga ion o a numbe o umo ypes, including PCa (NCT02905318, NCT02494921, NCT02555189). The analyses abo e indica e ha E2F unc ion is unde he con ol o PARP-1, and hus, i was necessa y o compa e he ansc ip ional e ec s o CDK4/6i o PARPi o disce n he ansc ip ional e ec s o PARP-1. To accomplish his, unbiased ansc ip omic da a gene - a ed in HT-sensi i e cells ea ed wi h ei he he CDK4/6i palboci- clib o he PARPi elipa ib we e compa ed. As shown in Fig 4B, le , he e was no signi ican o e lap in he genes up- egula ed by CDK4/6i and PARPi (n=1), and minimal o e lap in he genes down- egula ed by each ea men (n=45). Howe e , hese anal- yses indica e ha he genes speci ically down- egula ed by PARPi we e no only he mos abundan (n=157), bu GSEA MSigDB analyses indica e his gene se was en iched o DNA epai p ocesses, including HR (Fig 4B, igh ). These da a indica e ha PARP-1 egula es a cell cycle-independen E2F1 unc ion, dis inc om he ansc ip ional gene egula ion by E2F associa ed wi h cell cycle con ol. PARP-1con ols o HR ac o a ailabili y a e associa ed wi h modula ion o he ch oma in con ex o E2F1 unc ion As he da a abo e iden i y PARP-1 as a posi i e egula o o E2F1 ac i i y and subsequen exp ession o genes con olling HR, he impac o PARP-1 inhibi ion was compa ed o ha o HR de iciency. U ilizing he HR gene se o gene a e hea maps om he unbiased da a de i ed abo e in Fig 2, i was de e mined ha whe he he compa a o was DHT in HT-sensi i e cells, o ehicle con ol in CRPC cells, he majo i y o HR gene exp ession was diminished wi h PARPi (Fig 5A, le ). In ac , he majo i y o genes in ol ed in mos DNA epai pa hways declined a e PARPi ea men (Appendix Fig S5). Fu he mo e, compa ison o he unbiased da a gene a ed abo e wi h a p e iously de eloped HR de iciency ansc ip ional signa u e (Peng e al, 2014) demons a ed a signi ican o e lap in bo h down- egula ed (n=104/151) and up- egula ed (n=44/89) genes (Fig 5A, middle). This signa u e was gene a ed by independen ly silencing BRCA1,RAD51,o BRIT1, ollowed by unbiased ansc ip- omic p o iling. The in e sec ion o hese condi ions se es as he HR de iciency ansc ip ional signa u e. This in e sec ion p o ed o be s a is ically signi ican using GSEA analyses (Fig 5A, igh ). These da a sugges PARP-1 supp ession educes a ailabili y o HR ac o s by ansc ip ional egula ion. ◀Figu e 3. PARP-1 egula es p o-oncogenic ansc ip ion ac o signaling. A Le : Da a gene a ed as desc ibed abo e in Fig 2we e u ilized o Gene Se En ichmen Analysis (GSEA) Molecula Signa u e Da aBases (MSigDB) KEGG analyses. Cu o o epo ing was a alse disco e y a e q alue o <0.25, and no malized en ichmen sco es (NES) a e shown, wi h da ke colo s indica ing mo e en ichmen . Middle: Da a gene a ed as desc ibed abo e in Fig 2we e u ilized o Gene Se En ichmen Analysis (GSEA) Molecula Signa u e Da aBases (MSigDB) KEGG analyses. Cu o o epo ing was a alse disco e y a e q alue o <0.25, and no malized en ichmen sco es (NES) a e shown, wi h da ke colo s indica ing mo e en ichmen . Open ci cles indica e cell cycle- ela ed hallma ks, and closed ci cles indica e DNA damage epai - ela ed hallma ks. Righ : Selec ed GSEA MSigDB Hallma ks pa hways a e shown wi h NES and alse disco e y a e (FDR). B Indica ed cell lines we e ea ed as depic ed in Fig 2. Da a a e depic ed as mean s anda d de ia ion o h ee independen biological expe imen s. S a is ical signi icance was de e mine by wo- ailed S uden ’s - es whe e *P<0.05,**P<0.01, ***P<0.001, ****P<0.0001. LNCaP: E2F1,P=0.0159; PCNA, P=0.0217; MCM7,P=4.0936e-6; CCNA2,P=0.0005.C4-2:E2F1,P=0.0074;PCNA, P=0.1258; MCM7,P=3.7471e-5; CCNA2,P=0.0031. C ChIP-qPCR a e C4-2cells we e ea ed as depic ed in Fig 2. Da a a e depic ed as mean s anda d de ia ion o h ee independen biological expe imen s. S a is ical signi icance was de e mined by wo- ailed S uden ’s - es whe e *P<0.05, ****P<0.0001.E2F1ChIP, P=0.4610; PARP-1ChIP, P=0.1773; Pol II ChIP, P=0.0305; AcH4ChIP, P=7.4261e-5. D A hymic nude mice we e injec ed wi h C4-2cell mixed wi h ma igel. Once umo s became 100 mm 3 , mice we e ea ed wi h ei he ehicle con ol o elipa ib. Se en y- wo hou s la e , umo s we e ha es ed, RNA was isola ed and used o qPCR quan i ica ion o he indica ed ansc ip s. Da a a e depic ed as log2absolu e gene egula ion o elipa ib samples compa ed o con ol samples, s anda d de ia ion o h ee independen xenog a umo s. E P os a ec omy issue (n=6) was cul u ed as p e iously desc ibed, and ea ed wi h ei he ehicle con ol o elipa ib o 6days. RNA was hen ha es ed om he issues and used o qPCR quan i ica ion o he indica ed ansc ip s. Da a a e depic ed as log2absolu e gene egula ion o elipa ib samples compa ed o con ol samples. Each indi idual issue is depic ed by a sepa a e ba colo . S a is ical analyses we e pe o med by Wilcoxon signed ank es . ª2018 The Au ho s EMBO Molecula Medicine 10:e8816 |2018 7o 20 Ma hew J Schiewe e al PARP-1impinges on DDR gene exp ession EMBO Molecula Medicine Published online: No embe 21, 2018 A B Figu e 4. 8o 20 EMBO Molecula Medicine 10:e8816 |2018 ª2018 The Au ho s EMBO Molecula Medicine PARP-1impinges on DDR gene exp ession Ma hew J Schiewe e al Published online: No embe 21, 2018 The impac o PARP-1 ac i i y on he exp ession o HR genes was alida ed a he ansc ip le el (~20–50% educ ion; Fig 5B, le ) and a he p o ein le el (~15–80% educ ion) in i o (Fig 5B igh ). Valida ion ha hese HR genes a e E2F1- egula ed was accomplished by ansien ly knocking down E2F1 and examining HR gene exp ession (Appendix Fig S6A). T ansc ip ional egula ion o HR gene exp ession was ound o be conse ed ac oss all PCa/ CRPC models es ed (Appendix Fig S6B). Fu he mo e, he depen- dence o HR gene exp ession on PARP-1 enzyma ic ac i i y was ali- da ed in i o (Fig 5C). Addi ionally, u ilizing he pa ien issue explan p ocess desc ibed in Fig 3 in which p os a ec omy issues a e cul u ed in he labo a o y, he eliance o HR gene exp ession on PARP-1 enzyma ic unc ion could be u he explo ed. PARPi hus elici ed a mo e obus and signi ican dec ease o HR gene exp es- sion, han canonical E2F1 a ge genes as desc ibed abo e, bu s ill wi h pa ien he e ogenei y o esponse (Fig 5D). Toge he , hese da a indica e ha PARP-1 inhibi ion educes exp ession o many genes in ol ed in DNA epai (especially HR), sugges ing ha inhibi ing PARP-1 enzyma ic unc ion may ansc ip ionally induce a s a e o “BRCA-ness”, o ela i e HR de iciency. To de ine po en ial mechanism(s) by which PARP-1 egula es HR gene exp ession, ChIP-qPCR expe imen s we e pe o med a egula- o y loci o HR genes known o be egula ed by E2F1. While he e was no clea pa e n o al e ed E2F1 esidency a h ee HR gene loci (BRCA2,RAD51, and TOP2A) a e PARPi (Fig 5E, op le g aphs), in each case, PARP-1 was ound o eside a each locus, and his esidency was diminished upon PARPi by ~60–83% (Fig 5E, op igh g aphs). Thus, PARPi des abilizes PARP-1 unc ion a HR gene egula o y loci, likely comp omising E2F1 ac i i y. As would be expec ed, RNA polyme ase II and ace yla ed his one H4 le els we e diminished a hese HR gene loci in esponse o PARPi by 40–80 and 28–60%, espec i ely (Fig 5E, bo om le and igh g aphs, espec- i ely). Fu he mo e, i was de e mined ha PARPi al e s he ac i a- ion s a us he endogenous inhibi o o E2F1 unc ion, he e inoblas oma umo supp esso (RB; Fig 5F), whe ein PARPi esul ed in en ichmen o hypophospho yla ed (ac i e) RB, sugges - ing ha he unc ions o PARP-1 supp ession may be pleio opic. Addi ionally, based on he obse ed dec ease in he ac i e ace y- la ed his one H4 ma k upon PARPi in Fig 5E, i was de e mined ha his was associa ed wi h educed CBP ch oma in occupancy (Fig 5G). These cong uous da a a e impo an , as CBP is a key his one ace yl ans e ase wi h known unc ions in PCa (San e e al, 2011; Ianculescu e al, 2012). Combined, hese da a indica e ha PARP-1 no only esides a HR gene egula o y loci and is dimin- ished upon PARPi, bu PARP-1 enzyma ic ac i i y appea s o suppo E2F1 in he con ex o a coac i a o , whose unc ions include modula ion o RB. Al e ed HR ac o exp ession is p e alen in human PCa and is en iched du ing disease p og ession Da a he ein indica e ha PARP-1 posi i ely egula es E2F1-media ed HR gene exp ession in cance , and ha supp ession o his ac i i y can po en ially induce a “BRCA-ness” pheno ype. Gi en ha PARP- 1 ac i i y is enhanced as a unc ion o agg essi e disease, pa e ns o HR gene exp ession we e que ied in human cance . An assess- men o he TCGA da a se (Cance Genome A las Resea ch N, 2015), which includes only p ima y PCa, demons a ed ha when bo h RNA and DNA al e a ions a e aken in o accoun , 50.45% o umo s in his da a se ha bo ed al e ed HR gene RNA o DNA (Fig 6A, le ). The mos equen HR gene al e a ion ound in p ima y disease was mRNA up- egula ion (65%), while mu a ions only occu ed in 6% (Fig 6A, igh ). U ilizing he po ion o he TCGA da a se (Cance Genome A las Resea ch N, 2015) wi h ma ched no mal and p ima y PCa issues, se e al HR genes we e signi ican ly up- egula ed in cance (Fig 6B), sugges ing ha de egu- la ion o HR gene exp ession occu s du ing PCa umo igenesis. Genes ha ei he did no pass he cu o o s a is ical signi icance o did no inc ease a e shown in Appendix Fig S7. In he PCF-SU2C da a se o ad anced mCRPC umo s (Robinson e al, 2015), he e was an inc eased occu ence o HR gene de ec s, wi h 68% o umo s ha bo ing ei he DNA o mRNA al e a ions (Fig 6C, le ). The mos equen al e a ion was mRNA up- egula ion (71%), while only 10% o hese umo s ha bo ed mu a ions in hese HR genes (Fig 6C, igh ). These obse a ions we e suppo ed by wo o he , independen da a se s (Kuma e al, 2016; Taylo e al, 2010; 94.74% HR gene al e a ion (Appendix Fig S8, op le ), 76% o which was mRNA up- egula ion (Appendix Fig S8, op igh ); 67.65% HR gene al e a ion (Appendix Fig S8, bo om le ), 26% o which was mRNA up- egula ion (Appendix Fig S8, op igh ), espec i ely). Howe e , he mos equen gene al e a ion in he second da a se was gene ampli ica ion, no mu a ion, u he sugges ing ha HR gene up- egula ion is he p edominan al e a ion p esen in human PCa. Assessmen o indi idual umo -le el da a indica es ha HR al e a ions a e no mu ually exclusi e, and he mos equen ly al e ed HR gene is NBN (22%), while BRCA1 and BRCA2 a e al e ed in ~7 and 8% o hese umo s, espec i ely (Fig 6D). Se e al s udies ha e indica ed ha he equency o DNA epai gene mu a ions is ele a ed in ad anced PCa when compa ed o p ima y disease (G asso e al, 2012; Robinson e al, 2015; P i cha d e al, 2016). Da a p esen ed he ein con i m his and also indica e ha HR gene exp ession is also inc eased as a unc ion o PCa p og ession. Combined, hese da a no only ei e a e ha HR gene de ec s occu a a highe equency in ad anced PCa, bu he mos equen HR gene abe a ion is mRNA up- egula ion, a he ◀Figu e 4. PARP-1e ec s on E2F signaling a e independen o cell cycle phase and dis inc om hose elici ed by CDK4/6inhibi ion. A Indica ed cell lines we e ea ed as depic ed in Fig 2, and labeled wi h b omodeoxyu idine (B dU), ha es ed a indica ed ime poin s and u ilized o FACS analyses. Da a a e depic ed as mean s anda d de ia ion o h ee independen biological expe imen s. *P<0.05 as de e mined by wo- ailed S uden ’s - es . LNCaP: 3h, P=0.9838;16 h, P=0.2197,24 h, P=0.0207.C4-2:3h, P=0.4520;16 h, P=0.9446;24 h, P=0.4025;48 h, P=0.3431. B Top: Da a gene a ed as desc ibed abo e in Fig 2we e compa ed o a sepa a e mic oa ay analysis in which he same cell line was exposed o 1lM palbociclib ins ead o elipa ib. Cu o s o compa ison we e a P alue <0.05, and old change o 1.5.Venn diag ams shows he o e lapping and non-o e lapping genes o bo h down- ( op) and up- egula ed (bo om) genes in esponse o ei he ea men modali y. S a is ical signi icance was de e mined using he Chi-squa ed s a is ical es . Bo om: Genes ound o be exclusi ely egula ed by palbociclib, commonly egula ed by palbociclib and elipa ib, o exclusi ely egula ed by elipa ib we e used o Gene Se En ichmen (GSEA) KEGG pa hway analyses. Da a indica e bo h FDR q alue, whe e he da ke colo s indica e highe con idence (lowe q). Numbe s indica e q alues. Blue a ow highligh s he Homologous Recombina ion KEGG pa hway. ª2018 The Au ho s EMBO Molecula Medicine 10:e8816 |2018 9o 20 Ma hew J Schiewe e al PARP-1impinges on DDR gene exp ession EMBO Molecula Medicine Published online: No embe 21, 2018 NEL794B001KT). Immuno luo escen de ec ion o pcH2AX(Se 139; CST #2577) was ca ied ou wi h he i s using a 1:200 dilu ion, ollowed by PARP-1 (Ac i e Mo i #39559) using a 1:100 dilu ion and PAR (T e igen Inc., 4335-AMC-050) using a 1:200 dilu ion. The TMA slide was i s blocked wi h 3% H 2 O 2 o 10 min, hen ea ed wi h animal- ee p o ein blocke (Vec o Labo a o ies ca . no. SP- 5030) o 15 min, and hen incuba ed o e nigh a 4°C wi h he pcH2AX p ima y an ibody dilu ed in An ibody Dilu ion Bu e (Ven- ana Medical Sys ems ca . no. ADB250). The nex day, he TMA slide was incuba ed wi h EnVision+Sys em—HRP labeled polyme goa an i- abbi seconda y an ibody (Dako ca . no. K4003) o 30 min a oom empe a u e ollowed by incuba ion wi h OPAL- FITC luo opho e o 10 min. Nex , he slide was loaded on o he Ven ana au os aine using he Ven ana eagen s o he machine. The pcH2AX an ibody was comple ely emo ed using hea e ie al wi h CC2 bu e , only lea - ing he FITC luo opho e behind ha was c osslinked o he issue. The PARP-1 an ibody was applied manually, ollowed by manual applica ion o he OPLA-Cy3 eagen . Nex , he PARP-1 an ibody was comple ely emo ed om he slide, lea ing he Cy3 luo opho e behind as i was c osslinked o he issue. The inal incuba ion occu ed wi h he PAR an ibody and he OPAL-Cy5 luo opho e. The slide was incuba ed wi h DAPI, washed, and co e slipped using p olong gold as he moun ing medium. No c oss- eac i i y in signals was obse ed be ween an ibodies, demons a ing ha he emo al o he an ibodies be ween s aining cycles was comple e. Indi idual co es we e imaged on he Vec a TM 2 quan i a i e slide imaging sys em. Non-neoplas ic and cance a eas we e anno a ed by a pa hologis esul ing in 156 non-neoplas ic a eas, 277 p ima y cance a eas, and 159 mCRPC a eas. Missing co es and co es wi h- ou glands we e excluded om he anno a ion. The InFo m TM so - wa e was used o ob ain he g ay-le el s aining images o indi idual luo opho es. The amoun o nuclea s aining in indi idual nuclei was measu ed o all ou luo opho es (DAPI, FITC, Cy3, Cy5), and in ensi y le els we e no malized ac oss he ou TMA slides. No malized in ensi ies o each luo opho e we e dicho omized in o posi i e o nega i e using as a cu o he median in ensi y ac oss all nuclei wi hin he TMA. The pe cen o posi i e nuclei o e e y an i- body was coun ed in benign and neoplas ic glands. Al e na i ely, he a e age exp ession o each luo opho e ac oss all nuclei in each anno a ed egion was de e mined. Cell cul u e and ea men s LNCaP and C4-2 cells we e main ained in minimum essen ial media (IMEM) supplemen ed wi h 5% FBS (hea -inac i a ed e al bo ine se um). 22R 1 cells we e main ained in Dulbecco’s modi ied Eagle’s media supplemen ed wi h 10% FBS. All media we e supplemen ed wi h 2 mmol/l o L-glu amine and 100 uni s/ml penicillin-s ep o- mycin. Velipa ib was ob ained om Enzo Li e Sciences (Fa ming- dale, NY, USA) and dissol ed in DMSO and used a indica ed concen a ions. Fo s e oid-deple ed condi ions, cells we e pla ed in app op ia e phenol ed- ee media supplemen ed wi h 5 o 10% cha coal dex an- ea ed FBS (CDT) as app op ia e. DHT was dissol ed in E OH and used a indica ed concen a ions. Cell lines we e no cul u ed o longe han 6 mon hs a e eceip om hei o iginal sou ce, o no longe han 45 passages. Cell lines a e au hen- ica ed by ATCC annually. Mic oa ay analysis Cells we e seeded a equal densi y in s e oid-deple ed (CDT) condi- ions hen ea ed as indica ed wi h as speci ied o 16 h; RNA was isola ed using TRIzol (In i ogen) acco ding o manu ac u e ’s spec- i ica ions, and submi ed o mic oa ay analysis o he Sidney Kimmel Cance Cen e Cance Genomics Sha ed Resou ce. Gene exp ession was p o iled using he A yme ix Human Gene 1.0 ST mic oa ay (San a Cla a, CA, USA), wi h hyb idiza ion pe o med using he GeneChip Hyb idiza ion O en 645, ollowed by scanning on A yme ix Gene Chip Scanne 3000. Da a p ep ocessing was pe o med in A yme ix Exp ession Console 1.1 using i e PLIER summa iza ion wi h PM-GCBG backg ound co ec ion and quan ile no maliza ion. Gene exp ession analysis Cells we e seeded a equal densi y in s e oid-deple ed (CDT) condi- ions and we e ea ed as speci ied; RNA was isola ed using TRIzol and cDNA gene a ed using Supe Sc ip III (In i ogen). Quan i a i e PCR was conduc ed wi h p ime s desc ibed in Appendix Table S1 and wi h an ABI S epOne machine and Powe Syb in acco dance wi h he manu ac u e ’s speci ica ions. ChIP analysis Cells we e cul u ed in media con aining CDT o 72 h and ea ed as indica ed. ChIP analyses and qPCR we e conduc ed as p e iously desc ibed (60), using p ime s desc ibed in Appendix Table S1. Xenog a analysis Fou -week-old male BALB/c nu/nu mice we e pu chased om Cha les Ri e , Inc. C4-2 (2 ×10 6 cells) we e esuspended in 100 ll o saline wi h 50% Ma igel (BD Biosciences) and we e implan ed subcu aneously in o he lank o he mice. All umo s we e s aged o 4 weeks be o e s a ing he d ug ea men . Fo assessmen o in i o gene exp ession, umo s om mice we e ea ed wi h a single dose o elipa ib (100 mg/kg ia o al ga age) and ha es ed 72 h a e ea men . Tissue was ha es ed a indica ed a e 6 days RNA was isola ed using TRIzol. No s a is ical me hods we e used o animal sample size es ima e, and no blinding was done. Animals we e andomized in o he wo ea men egimens ia coin lip. Mice we e housed in s anda d condi ions. All animal wo k was done in compliance wi h he egula ions se o h by he Je e son Uni e si y IACUC. Ch oma in e he ing assays C4-2 cells we e ea ed wi h ei he 2.5uM elipa ib o ehicle con ol, hen ha es ed and p ocessed 24 h la e as p e iously desc ibed (Schiewe e al, 2012). Human p os a e umo ex i o cul u e Human p os a e ex i o explan cul u es we e conduc ed as p e i- ously desc ibed (de Leeuw e al, 2015). B ie ly, esh issue was ob ained om a pa hologis immedia ely ollowing adical 16 o 20 EMBO Molecula Medicine 10:e8816 |2018 ª2018 The Au ho s EMBO Molecula Medicine PARP-1impinges on DDR gene exp ession Ma hew J Schiewe e al Published online: No embe 21, 2018 p os a ec omy. The de-iden i ied specimens we e p ocessed unde a lamina low hood, using s e ile echnique, and anspo ed o he lab in IMEM on ice. The Thomas Je e son Uni e si y Ins i u ional Re iew Boa d has e iewed his p ocu emen p o ocol and de e - mined his esea ch o be in compliance wi h ede al egula ions go e ning esea ch on de-iden i ied specimens and/o clinical da a [45 CFR 46.102( )]. The ollowing p ocedu es we e conduc ed unde s e ile issue-cul u e condi ions. Ve e ina y den al sponges (No a is Ca . #96002) we e placed in 12-well pla es and soaked in 500 ml media (IMEM supplemen ed wi h 5% hea -inac i a ed FBS, hyd o- co isone, insulin om bo ine panc eas, and 100 uni s/ml peni- cillin-s ep omycin) and app op ia e ea men (ei he ehicle con ol o 2.5 lM elipa ib) o 5–10 min a 37°C. Tissue was placed in o he lid o a 10-cm pla e and dissec ed in o 1-mm 3 pieces wi h a scalpel. Th ee pieces o issue we e placed on each sponge, using s e ile weeze s o o ceps. Pla es we e placed in an incuba o a 37°C and 5% CO 2 . Media we e eplaced e e y day wi h app op i- a e ea men . Tissue was ha es ed a indica ed a e 6 days RNA was isola ed using TRIzol. While he e was no clinical in es iga ion epo ed in his s udy, in o med consen was ob ained om all subjec s and ha he expe - imen s con o med o he p inciples se ou in he WMA Decla a ion o Helsinki and he Depa men o Heal h and Human Se ices Belmon Repo . Cell g ow h assays Cells we e seeded a equal densi ies, ea ed as indica ed, and ha es ed a 96 h. A he ime o ha es , cell numbe was de e - mined using ypan blue exclusion and a hemocy ome e . An ibodies and immunoblo ing P o ein isola ion and immunoblo ing we e conduc ed as p e iously desc ibed (Knudsen e al, 1998), using an ise a desc ibed in Appendix Table S1. Da a a ailabili y The da a se s p oduced in his s udy a e a ailable in he ollowing da abases: Mic oa ay da a: Gene Exp ession Omnibus GSE118222 (h ps://www.ncbi.nlm.nih.go /geo/que y/acc.cgi?acc=GSE118222). Expanded View o his a icle is a ailable online. Acknowledgemen s We g a e ully hank all he membe s o he Knudsen labo a o y o hei in el- lec ual and echnical suppo . Addi ionally, we hank he ollowing ins i u ions ha suppo ed his wo k: he NIH/NCI g an s o KEK (R01 CA176401,R01 CA182569,R01 CA217329,P30 CA056036) and he Sidney Kimmel Cance Cen e (5P30CA056036), he P os a e Cance Founda ion ( o MJS and KEK), and he T ansla ional Pa hology and Me aOmics co e acili ies a SKCC. Au ho con ibu ions Concep ualiza ion: MJS and KEK; me hodology: MJS, BK, and KEK; in es iga ion and da a analysis: MJS, ACM, NG, FH, SG, RdL, SGZ, JE, SH, TP, RB, PMcC, CMcN, SNC, YC-F, PG, JJMcC, NPN, AAS, ED, LJB; bios a is ics: BEL; bioin o ma ics: CMcN; w i ing—o iginal d a : MJS and KEK; w i ing— e iew and edi ing: all au ho s; unding acquisi ion: MJS and KEK; esou ces: TV, GVR, CDL, EJT, LGG, APD, WKK, FYF. All au ho s ead and app o ed he manusc ip . Con lic o in e es The au ho s decla e ha hey ha e no con lic o in e es . Re e ences Alla V, Engelmann D, Nieme z A, Pahnke J, Schmid A, Kunz M, Emm ich S, S ede M, Koczan D, Pu ze BM (2010)E2F1in melanoma p og ession and me as asis. J Na l Cance Ins 102:127 –133 Audeh MW, Ca michael J, Penson RT, F iedlande M, Powell B, Bell-McGuinn KM, Sco C, Wei zel JN, Oaknin A, Loman N e al (2010) O al poly(ADP- ibose) polyme ase inhibi o olapa ib in pa ien s wi h BRCA1o BRCA2 mu a ions and ecu en o a ian cance : a p oo -o -concep ial. Lance 376:245 –251 Ba be LJ, Sandhu S, Chen L, Campbell J, Koza ewa I, Fenwick K, Assio is I, Rod igues DN, Reis Filho JS, Mo eno V e al (2013) Seconda y mu a ions in BRCA2associa ed wi h clinical esis ance o a PARP inhibi o . J Pa hol 229: 422 –429 Beneke S, Cohausz O, Malanga M, Boukamp P, Al haus F, Bu kle A (2008) Rapid egula ion o elome e leng h is media ed by poly(ADP- ibose) polyme ase-1.Nucleic Acids Res 36:6309 –6317 The pape explained P oblem While he oles poly(ADP- ibose) polyme ase-1(PARP-1) pe o ms in esponse o DNA damage a e inc easingly well unde s ood, as a e he oles o PARP-1in o he aspec s o genome in eg i y ( elome ic main- enance and eplica ion o k s abili y), he o he ch oma in-associa ed unc ion o PARP-1( ansc ip ion) has no been ully explo ed as a means o egula e DNA epai . Resul s Cen al indings a e as ollows: (i) PARP-1enzyma ic ac i i y is inc eased as a unc ion o disease p og ession and is associa ed wi h poo ou come. (ii) Ele a ed PARP-1enzyma ic unc ion in ad anced disease is no a ibu able o inc eased DNA DSB epai . (iii) Iden i i- ca ion o he PARP-1- egula ed ansc ip ome e eals ele ance o disease p og ession. (i ) PARP-1 egula es p o-oncogenic ansc ip ion ac o signaling, including E2F1. ( ) PARP-1e ec s on E2F signaling a e independen o cell cycle phase and dis inc om hose elici ed by CDK4/6inhibi ion. ( i) PARP-1 egula es homologous ecombina ion (HR) ac o a ailabili y ia modula ing ch oma in a E2F1binding si es.( i) Al e ed HR ac o exp ession is p e alen in human PCa and is en iched du ing disease p og ession.( iii) PARP-1 egula es DNA epai ac o a ailabili y and DNA epai compe ency. Impac These da a es ablish h ee essen ial poin s. Fi s , hey p o ide he i s e idence ha PARP-1enzyma ic and ansc ip ional unc ions a e ele a ed as a unc ion o disease p og ession, i espec i e o DNA epai . Second, he da a es ablish an unexpec ed ole o PARP-1in con olling DNA epai gene exp ession, and e eal a new pa adigm o PARP-1 o unc ion as an enhance o induce o “BRCA-ness”. Finally, hese da a demons a e he clinical ele ance o PARP-1- egu- la ed E2F1-d i en exp ession o HR ac o s and p o ide s iking new e idence o no el bioma ke s o human disease. ª2018 The Au ho s EMBO Molecula Medicine 10:e8816 |2018 17 o 20 Ma hew J Schiewe e al PARP-1impinges on DDR gene exp ession EMBO Molecula Medicine Published online: No embe 21, 2018 Bi FF, Li D, Yang Q (2013) Hypome hyla ion o ETS ansc ip ion ac o binding si es and up egula ion o PARP1exp ession in endome ial cance . Biomed Res In 2013:946268 Bind a RS, Gibson SL, Meng A, Wes e ma k U, Jasin M, Pie ce AJ, B is ow RG, Classon MK, Glaze PM (2005) Hypoxia-induced down- egula ion o BRCA1 exp ession by E2Fs. Cance Res 65:11597 –11604 Biswas AK, Johnson DG (2012) T ansc ip ional and non ansc ip ional unc ions o E2F1in esponse o DNA damage. Cance Res 72:13 –17 B enne JC, A eeq B, Li Y, Yocum AK, Cao Q, Asangani IA, Pa el S, Wang X, Liang H, Yu J e al (2011) Mechanis ic a ionale o inhibi ion o poly(ADP- ibose) polyme ase in ETS gene usion-posi i e p os a e cance . Cance Cell 19:664 –678 B yan HE, Pe e mann E, Schul z N, Jem h AS, Lose a O, Issae a N, Johansson F, Fe nandez S, McGlynn P, Helleday T (2009) PARP is ac i a ed a s alled o ks o media e M e11-dependen eplica ion es a and ecombina ion. EMBO J 28:2601 –2615 Bye s LA, Wang J, Nilsson MB, Fujimo o J, Sain igny P, Yo dy J, Gi i U, Pey on M, Fan YH, Diao L e al (2012) P o eomic p o iling iden i ies dys egula ed pa hways in small cell lung cance and no el he apeu ic a ge s including PARP1.Cance Disco 2:798 –811 Cance Genome A las Resea ch N (2015) The molecula axonomy o p ima y p os a e cance . Cell 163:1011 –1025 Cen ene a MM, Gillis JL, Hanson AR, Jindal S, Taylo RA, Risb idge GP, Su he land PD, Sche HI, Raj GV, Knudsen KE e al (2012) E idence o e icacy o new Hsp90 inhibi o s e ealed by ex i o cul u e o human p os a e umo s. Clin Cance Res 18:3562 –3570 Cen ene a MM, Raj GV, Knudsen KE, Tilley WD, Bu le LM (2013)Ex i o cul u e o human p os a e issue and d ug de elopmen . Na Re U ol 10: 483 –487 Chand SN, Za ei M, Schiewe MJ, Kama h AR, Romeo C, Lal S, Cozzi o o JA, Ne le A, Scola o L, Londin E e al (2017) Pos ansc ip ional egula ion o PARG mRNA by HuR acili a es DNA epai and esis ance o PARP inhibi o s. Cance Res 77:5011 –5025 Chao OS, Goodman OB J (2014) Syne gis ic loss o p os a e cance cell iabili y by coinhibi ion o HDAC and PARP. Mol Cance Res 12:1755 –1766 Chen X, Xu H, Yuan P, Fang F, Huss M, Vega VB, Wong E, O lo YL, Zhang W, Jiang J e al (2008) In eg a ion o ex e nal signaling pa hways wi h he co e ansc ip ional ne wo k in emb yonic s em cells. Cell 133:1106 –1117 Chen D, Pacal M, Wenzel P, Knoep le PS, Leone G, B emne R (2009) Di ision and apop osis o E2 -de icien e inal p ogeni o s. Na u e 462:925 –929 Ch is ie EL, Fe eday S, Doig K, Pa naik S, Dawson SJ, Bow ell DDL (2017) Re e sion o BRCA1/2ge mline mu a ions de ec ed in ci cula ing umo DNA om pa ien s wi h high-g ade se ous o a ian cance . J Clin Oncol 35: 1274 –1280 Cla ke N, Wiechno P, Aleksee B, Sala N, Jones R, Kocak I, Chiu i VE, Jassem J, Flechon A, Red e n C e al (2018) Olapa ib combined wi h abi a e one in pa ien s wi h me as a ic cas a ion- esis an p os a e cance : a andomised, double-blind, placebo-con olled, phase 2 ial. Lance Oncol 19:975 –986 Coleman RL, Sill MW, Bell-McGuinn K, Aghajanian C, G ay HJ, Tewa i KS, Rubin SC, Ru he o d TJ, Chan JK, Chen A e al (2015) A phase II e alua ion o he po en , highly selec i e PARP inhibi o elipa ib in he ea men o pe sis en o ecu en epi helial o a ian, allopian ube, o p ima y pe i oneal cance in pa ien s who ca y a ge mline BRCA1o BRCA2 mu a ion –an NRG Oncology/Gynecologic Oncology G oup s udy. Gynecol Oncol 137:386 –391 Coms ock CE, Augello MA, Goodwin JF, de Leeuw R, Schiewe MJ, Os ande WF J , Bu kha RA, McClendon AK, McCue PA, T abulsi EJ e al (2013) Ta ge ing cell cycle and ho mone ecep o pa hways in cance . Oncogene 32:5481 –5491 D’Amou s D, Desnoye s S, D’Sil a I, Poi ie GG (1999) Poly(ADP- ibosyl)a ion eac ions in he egula ion o nuclea unc ions. Biochem J 342(P 2): 249 –268 Du kacz BW, Omidiji O, G ay DA, Shall S (1980) (ADP- ibose)n pa icipa es in DNA excision epai . Na u e 283:593 –596 Dziaman T, Ludwiczak H, Ciesla JM, Banaszkiewicz Z, Winczu a A, Chmiela czyk M, Wisniewska E, Ma szalek A, Tudek B, Olinski R (2014) PARP-1exp ession is inc eased in colon adenoma and ca cinoma and co ela es wi h OGG1.PLoS One 9:e115558 Edwa ds SL, B ough R, Lo d CJ, Na ajan R, Va che a R, Le ine DA, Boyd J, Reis-Filho JS, Ashwo h A (2008) Resis ance o he apy caused by in agenic dele ion in BRCA2.Na u e 451:1111 –1115 Fong PC, Boss DS, Yap TA, Tu A, Wu P, Me gui-Roel ink M, Mo ime P, Swaisland H, Lau A, O’Conno MJ e al (2009) Inhibi ion o poly(ADP- ibose) polyme ase in umo s om BRCA mu a ion ca ie s. N Engl J Med 361:123 –134 Gan A, G een AR, Nolan CC, Ma in S, Deen S (2013) Poly(adenosine diphospha e- ibose) polyme ase exp ession in BRCA-p o icien o a ian high-g ade se ous ca cinoma; associa ion wi h pa ien su i al. Hum Pa hol 44:1638 –1647 Ga emann A, B edeho s R, Wielckens K, S a ling WH, Hilz H (1981) Mono- and poly-ADP- ibosyla ion o p o eins in mouse kidney a e cas a ion and es os e one ea men . Biochem J 198:37 –44 Gelmon KA, Tischkowi z M, Mackay H, Swene on K, Robidoux A, Tonkin K, Hi e H, Hun sman D, Clemons M, Gilks B e al (2011) Olapa ib in pa ien s wi h ecu en high-g ade se ous o poo ly di e en ia ed o a ian ca cinoma o iple-nega i e b eas cance : a phase 2, mul icen e, open- label, non- andomised s udy. Lance Oncol 12:852 –861 Godoy H, Mhawech-Fauceglia P, Beck A, Mille A, Lele S, Odunsi K (2011) Exp ession o poly (adenosine diphospha e- ibose) polyme ase and p53 in epi helial o a ian cance and hei ole in p ognosis and disease ou come. In J Gynecol Pa hol 30:139 –144 Gogola E, Dua e AA, de Rui e JR, Wiegan WW, Schmid JA, de B uijn R, James DI, Gue e o Llobe S, Vis DJ, Annunzia o S e al (2018) Selec i e loss o PARG es o es PARyla ion and coun e ac s PARP inhibi o -media ed syn he ic le hali y. Cance Cell 33:1078 –1093 e12 Goodall J, Ma eo J, Yuan W, Mossop H, Po a N, Mi anda S, Pe ez-Lopez R, Dolling D, Robinson DR, Sandhu S e al (2017) Ci cula ing cell- ee DNA o guide p os a e cance ea men wi h PARP inhibi ion. Cance Disco 7: 1006 –1017 Goodwin JF, Ko ha i V, D ake JM, Zhao S, Dylgje i E, Dean JL, Schiewe MJ, McNai C, Jones JK, Ay es A e al (2015) DNA-PKcs-media ed ansc ip ional egula ion d i es p os a e cance p og ession and me as asis. Cance Cell 28:97 –113 Go o Y, Hayashi R, Kang D, Yoshida K (2006) Acu e loss o ansc ip ion ac o E2F1induces mi ochond ial biogenesis in HeLa cells. J Cell Physiol 209: 923 –934 G asso CS, Wu YM, Robinson DR, Cao X, Dhanaseka an SM, Khan AP, Quis MJ, Jing X, Lonig o RJ, B enne JC e al (2012)The mu a ional landscape o le hal cas a ion- esis an p os a e cance . Na u e 487:239 –243 Ha sough EJ, Kugel CH, Vido MJ, Be ge AC, Pu win TJ, Goldbe g A, Da ies MA, Schiewe MJ, Knudsen KE, Bollag G e al (2018) Response and esis ance o pa adox b eaking BRAF inhibi o in melanomas in i o and ex i o.Mol Cance The 17:84 –95 Ianculescu I, Wu DY, Siegmund KD, S allcup MR (2012) Selec i e oles o cAMP esponse elemen -binding p o ein binding p o ein and p300 p o ein 18 o 20 EMBO Molecula Medicine 10:e8816 |2018 ª2018 The Au ho s EMBO Molecula Medicine PARP-1impinges on DDR gene exp ession Ma hew J Schiewe e al Published online: No embe 21, 2018 as co egula o s o and ogen- egula ed gene exp ession in ad anced p os a e cance cells. J Biol Chem 287:4000 –4013 Jaspe s JE, Ke sbe gen A, Boon U, Sol W, an Deem e L, Zande SA, D os R, Wien jens E, Ji J, Aly A e al (2013) Loss o 53BP1causes PARP inhibi o esis ance in B ca1-mu a ed mouse mamma y umo s. Cance Disco 3: 68 –81 Johnson N, Johnson SF, Yao W, Li YC, Choi YE, Be nha dy AJ, Wang Y, Capelle i M, Sa osiek KA, Mo eau LA e al (2013) S abiliza ion o mu an BRCA1p o ein con e s PARP inhibi o and pla inum esis ance. P oc Na l Acad Sci USA 110:17041 –17046 Kaye SB, Lubinski J, Ma ulonis U, Ang JE, Gou ley C, Ka lan BY, Amnon A, Bell- McGuinn KM, Chen LM, F iedlande M e al (2012) Phase II, open-label, andomized, mul icen e s udy compa ing he e icacy and sa e y o olapa ib, a poly (ADP- ibose) polyme ase inhibi o , and pegyla ed liposomal doxo ubicin in pa ien s wi h BRCA1o BRCA2mu a ions and ecu en o a ian cance . J Clin Oncol 30:372 –379 Kim HC, Song JS, Lee JC, Lee DH, Kim SW, Lee JS, Kim WS, Rho JK, Kim SY, Choi CM (2014) Clinical signi icance o NQO1polymo phism and exp ession o p53, SOD2, PARP1in limi ed-s age small cell lung cance . In J Clin Exp Pa hol 7:6743 –6751 Kim KM, Moon YJ, Pa k SH, Pa k HJ, Wang SI, Pa k HS, Lee H, Kwon KS, Moon WS, Lee DG e al (2016) Indi idual and combined exp ession o DNA damage esponse molecules PARP1, gammaH2AX, BRCA1, and BRCA2 p edic sho e su i al o so issue sa coma pa ien s. PLoS One 11: e0163193 Knudsen KE, A den KC, Ca enee WK (1998)Mul iple G1 egula o y elemen s con ol he and ogen-dependen p oli e a ion o p os a ic ca cinoma cells. J Biol Chem 273:20213 –20222 Kond asho a O, Nguyen M, Shield-A in K, Tinke AV, Teng NNH, Ha ell MI, Kuipe MJ, Ho GY, Ba ke H, Jasin M e al (2017) Seconda y soma ic mu a ions es o ing RAD51C and RAD51D associa ed wi h acqui ed esis ance o he PARP inhibi o ucapa ib in high-g ade o a ian ca cinoma. Cance Disco 7:984 –998 K aus WL, Lis JT (2003) PARP goes ansc ip ion. Cell 113:677 –683 K ishnakuma R, K aus WL (2010) The PARP side o he nucleus: molecula ac ions, physiological ou comes, and clinical a ge s. Mol Cell 39:8–24 Kuma A, Coleman I, Mo issey C, Zhang X, T ue LD, Gula i R, E zioni R, Bolou i H, Mon gome y B, Whi e T e al (2016) Subs an ial in e indi idual and limi ed in aindi idual genomic di e si y among umo s om men wi h me as a ic p os a e cance . Na Med 22:369 –378 Kuma i A, Iwasaki T, Pyndiah S, Cassime e EK, Palani CD, Sakamu o D (2015) Regula ion o E2F1-induced apop osis by poly(ADP- ibosyl)a ion. Cell Dea h Di e 22:311 –322 Lapoin e J, Li C, Higgins JP, an de Rijn M, Bai E, Mon gome y K, Fe a i M, Ege ad L, Ray o d W, Be ge heim U e al (2004) Gene exp ession p o iling iden i ies clinically ele an sub ypes o p os a e cance . P oc Na l Acad Sci USA 101:811 –816 de Leeuw R, Be man-Boo y LD, Schiewe MJ, Cimen SJ, Den RB, Dicke AP, Kelly WK, T abulsi EJ, Lallas CD, Gomella LG e al (2015) No el ac ions o nex -gene a ion axanes bene i ad anced s ages o p os a e cance . Clin Cance Res 21:795 –807 Lege K, Hopp AK, Fey M, Ho ige MO (2016) ARTD1 egula es cyclin E exp ession and consequen ly cell-cycle e-en y and G1/S p og ession in T24 bladde ca cinoma cells. Cell Cycle 15:2042 –2052 Li Z, Yan X, Sun Y, Yang X (2016) Exp ession o ADP- ibosyl ans e ase 1is associa ed wi h poo p ognosis o glioma pa ien s. Tohoku J Exp Med 239: 269 –278 Liu L, Zhou W, Cheng CT, Ren X, Somlo G, Fong MY, Chin AR, Li H, Yu Y, Xu Y e al (2014) TGFbe a induces “BRCAness”and sensi i i y o PARP inhibi ion in b eas cance by egula ing DNA- epai genes. Mol Cance Res 12:1597 –1609 Lo d CJ, Ashwo h A (2008) Ta ge ed he apy o cance using PARP inhibi o s. Cu Opin Pha macol 8:363 –369 Lo d CJ, Ashwo h A (2017) PARP inhibi o s: syn he ic le hali y in he clinic. Science 355:1152 –1158 Mascolo M, Ila di G, Romano MF, Cele i A, Siano M, Romano S, Luise C, Me olla F, Rocco A, Vecchione ML e al (2012) O e exp ession o ch oma in assembly ac o -1p60, poly(ADP- ibose) polyme ase 1and nes in p edic s me as asizing beha iou o o al cance . His opa hology 61: 1089 –1105 Ma eo J, Ca ei a S, Sandhu S, Mi anda S, Mossop H, Pe ez-Lopez R, Na a Rod igues D, Robinson D, Omlin A, Tuna iu N e al (2015) DNA- epai de ec s and olapa ib in me as a ic p os a e cance . N Engl J Med 373: 1697 –1708 McCabe N, Tu ne NC, Lo d CJ, Kluzek K, Bialkowska A, Swi S, Gia a a S, O’Conno MJ, Tu AN, Zdzienicka MZ e al (2006) De iciency in he epai o DNA damage by homologous ecombina ion and sensi i i y o poly (ADP- ibose) polyme ase inhibi ion. Can Res 66:8109 –8115 Mi za MR, Monk BJ, He s ed J, Oza AM, Mahne S, Redondo A, Fabb o M, Lede mann JA, Lo usso D, Ve go e I e al (2016) Ni apa ib main enance he apy in pla inum-sensi i e, ecu en o a ian cance . N Engl J Med 375: 2154 –2164 Moo ha VK, Lindg en CM, E iksson KF, Sub amanian A, Sihag S, Leha J, Puigse e P, Ca lsson E, Ridde s ale M, Lau ila E e al (2003) PGC- 1alpha- esponsi e genes in ol ed in oxida i e phospho yla ion a e coo dina ely down egula ed in human diabe es. Na Gene 34:267 –273 O’Lea y B, Finn RS, Tu ne NC (2016) T ea ing cance wi h selec i e CDK4/6 inhibi o s. Na Re Clin Oncol 13:417 –430 Pa k SH, Jang KY, Kim MJ, Yoon S, Jo Y, Kwon SM, Kim KM, Kwon KS, Kim CY, Woo HG (2015) Tumo supp essi e e ec o PARP1and FOXO3A in gas ic cance s and i s clinical implica ions. Onco a ge 6:44819 –44831 Peng G, Chun-Jen Lin C, Mo W, Dai H, Pa k YY, Kim SM, Peng Y, Mo Q, Siwko S, Hu R e al (2014) Genome-wide ansc ip ome p o iling o homologous ecombina ion DNA epai . Na Commun 5:3361 Pish aian MJ, Biankin AV, Bailey P, Chang DK, Lahe u D, Wol gang CL, B ody JR (2017) BRCA2seconda y mu a ion-media ed esis ance o pla inum and PARP inhibi o -based he apy in panc ea ic cance . B J Cance 116: 1021 –1026 Podho ecka M, Skladanowski A, Bozko P (2010)H2AX phospho yla ion: i s ole in DNA damage esponse and cance he apy. J Nucleic Acids 2010:1–9 Poi ie GG, de Mu cia G, Jongs a-Bilen J, Niede gang C, Mandel P (1982) Poly (ADP- ibosyl)a ion o polynucleosomes causes elaxa ion o ch oma in s uc u e. P oc Na l Acad Sci USA 79:3423 –3427 Polage S, Ginsbe g D (2009)p53 and E2 : pa ne s in li e and dea h. Na Re Cance 9:738 –748 P ensne JR, Chen W, Iye MK, Cao Q, Ma T, Han S, Sahu A, Malik R, Wilde - Romans K, Na one N e al (2014) PCAT-1, a long noncoding RNA, egula es BRCA2and con ols homologous ecombina ion in cance . Cance Res 74:1651 –1660 P i cha d CC, Ma eo J, Walsh MF, De Sa ka N, Abida W, Bel an H, Ga o alo A, Gula i R, Ca ei a S, Eeles R e al (2016) Inhe i ed DNA- epai gene mu a ions in men wi h me as a ic p os a e cance . N Engl J Med 375: 443 –453 Qin G, Kisho e R, Dolan CM, Sil e M, Wecke A, Luedemann CN, Tho ne T, Hanley A, Cu y C, Heyd L e al (2006) Cell cycle egula o E2F1modula es ª2018 The Au ho s EMBO Molecula Medicine 10:e8816 |2018 19 o 20 Ma hew J Schiewe e al PARP-1impinges on DDR gene exp ession EMBO Molecula Medicine Published online: No embe 21, 2018 angiogenesis ia p53-dependen ansc ip ional con ol o VEGF. P oc Na l Acad Sci USA 103:11015 –11020 Quigley D, Alumkal JJ, Wya AW, Ko ha i V, Foye A, Lloyd P, Agga wal R, Kim W, Lu E, Schwa zman J e al (2017) Analysis o ci cula ing cell- ee DNA iden i ies mul iclonal he e ogenei y o BRCA2 e e sion mu a ions associa ed wi h esis ance o PARP inhibi o s. Cance Disco 7:999 –1005 Robinson D, Van Allen EM, Wu YM, Schul z N, Lonig o RJ, Mosque a JM, Mon gome y B, Taplin ME, P i cha d CC, A a d G e al (2015) In eg a i e clinical genomics o ad anced p os a e cance . Cell 161:1215 –1228 Rojo F, Ga cia-Pa a J, Zazo S, Tusque s I, Fe e -Lozano J, Menendez S, E oles P, Chamizo C, Se i ja S, Rami ez-Me ino N e al (2012) Nuclea PARP-1 p o ein o e exp ession is associa ed wi h poo o e all su i al in ea ly b eas cance . Ann Oncol 23:1156 –1164 Roun ee MR, Bachman KE, Baylin SB (2000) DNMT1binds HDAC2and a new co- ep esso , DMAP1, o o m a complex a eplica ion oci. Na Gene 25: 269 –277 Salemi M, Galia A, F agge a F, La Co e C, Pepe P, La Vigne a S, Imp o a G, Bosco P, Caloge o AE (2013) Poly (ADP- ibose) polyme ase 1p o ein exp ession in no mal and neoplas ic p os a ic issue. Eu J His ochem 57:e13 Sandhu SK, Schelman WR, Wilding G, Mo eno V, Bai d RD, Mi anda S, Hylands L, Riisnaes R, Fo s e M, Omlin A e al (2013) The poly(ADP- ibose) polyme ase inhibi o ni apa ib (MK4827) in BRCA mu a ion ca ie s and pa ien s wi h spo adic cance : a phase 1dose-escala ion ial. Lance Oncol 14:882 –892 San e FR, Hoschele PP, Oh SJ, E b HH, Bouchal J, Ca a e a IT, Pa son W, Meye s DJ, Cole PA, Culig Z (2011)Inhibi ion o he ace yl ans e ases p300 and CBP e eals a a ge able unc ion o p300 in he su i al and in asion pa hways o p os a e cance cell lines. Mol Cance The 10: 1644 –1655 Schiewe MJ, Goodwin JF, Han S, B enne JC, Augello MA, Dean JL, Liu F, Planck JL, Ra ind ana han P, Chinnaiyan AM e al (2012) Dual oles o PARP-1p omo e cance g ow h and p og ession. Cance Disco 2: 1134 –1149 Schiewe MJ, Knudsen KE (2014) T ansc ip ional oles o PARP1in cance . Mol Cance Res 12:1069 –1080 Sha ma A, Yeow WS, E el A, Coleman I, Clegg N, Thanga el C, Mo issey C, Zhang X, Coms ock CE, Wi kiewicz AK e al (2010) The e inoblas oma umo supp esso con ols and ogen signaling and human p os a e cance p og ession. J Clin In es 120:4478 –4492 Simbulan-Rosen hal CM, Rosen hal DS, Boula es AH, Hickey RJ, Malkas LH, Coll JM, Smulson ME (1998) Regula ion o he exp ession o ec ui men o componen s o he DNA syn hesome by poly(ADP- ibose) polyme ase. Biochemis y 37:9363 –9370 Simbulan-Rosen hal CM, Rosen hal DS, Luo R, Sama a R, Espinoza LA, Hassa PO, Ho ige MO, Smulson ME (2003) PARP-1binds E2F-1independen ly o i s DNA binding and ca aly ic domains, and ac s as a no el coac i a o o E2F-1-media ed ansc ip ion du ing e-en y o quiescen cells in o S phase. Oncogene 22:8460 –8471 Sub amanian A, Tamayo P, Moo ha VK, Mukhe jee S, Ebe BL, Gille e MA, Paulo ich A, Pome oy SL, Golub TR, Lande ES e al (2005) Gene se en ichmen analysis: a knowledge-based app oach o in e p e ing genome-wide exp ession p o iles. P oc Na l Acad Sci USA 102: 15545 –15550 Taylo BS, Schul z N, Hie onymus H, Gopalan A, Xiao Y, Ca e BS, A o a VK, Kaushik P, Ce ami E, Re a B e al (2010) In eg a i e genomic p o iling o human p os a e cance . Cance Cell 18:11 –22 Veskimae K, S a S, G onholm A, Pesu M, Laaksonen M, Nyk e M, Isola J, Maenpaa J (2016) Assessmen o PARP p o ein exp ession in epi helial o a ian cance by ELISA pha macodynamic assay and immunohis ochemis y. Tumou Biol 37:11991 –11999 Wal e RF, Vollb ech C, We ne R, Mai inge T, Schmelle J, Flom E, Wohlschlaege J, Ba be akis N, Paliou as D, Cha zinikolaou F e al (2016) Sc eening o pleu al meso heliomas o DNA-damage epai playe s by digi al gene exp ession analysis can enhance clinical managemen o pa ien s ecei ing pla in-based chemo he apy. J Cance 7:1915 –1925 Weigel B, Comino-Mendez I, de B uijn I, Tian L, Meisel JL, Ga cia-Mu illas I, F ibbens C, Cu s R, Ma elo o LG, Ng CKY e al (2017) Di e se BRCA1and BRCA2 e e sion mu a ions in ci cula ing cell- ee DNA o he apy- esis an b eas o o a ian cance . Clin Cance Res 23:6708 –6720 Wiegmans AP, Yap PY, Wa d A, Lim YC, Khanna KK (2015) Di e ences in exp ession o key DNA damage epai genes a e epigene ic-induced BRCAness dic a e syn he ic le hali y wi h PARP1inhibi ion. Mol Cance The 14:2321 –2331 Wu W, Zhu H, Liang Y, Kong Z, Duan X, Li S, Zhao Z, Yang D, Zeng G (2014) Exp ession o PARP-1and i s ac i e polyme PAR in p os a e cance and benign p os a ic hype plasia in Chinese pa ien s. In U ol Neph ol 46: 1345 –1349 Yu SW, Wang H, Poi as MF, Coombs C, Bowe s WJ, Fede o HJ, Poi ie GG, Dawson TM, Dawson VL (2002) Media ion o poly(ADP- ibose) polyme ase-1-dependen cell dea h by apop osis-inducing ac o . Science 297:259 –263 Yu J, Cao Q, Meh a R, Laxman B, Yu J, Tomlins SA, C eigh on CJ, Dhanaseka an SM, Shen R, Chen G e al (2007) In eg a i e genomics analysis e eals silencing o be a-ad ene gic signaling by polycomb in p os a e cance . Cance Cell 12:419 –431 Zhai L, Li S, Li H, Zheng Y, Lang R, Fan Y, Gu F, Guo X, Zhang X, Fu L (2015) Polymo phisms in poly (ADP- ibose) polyme ase-1(PARP1) p omo e and 30un ansla ed egion and hei associa ion wi h PARP1exp ession in b eas cance pa ien s. In J Clin Exp Pa hol 8:7059 –7071 Zhang T, Be ocal JG, Yao J, DuMond ME, K ishnakuma R, Ruhl DD, Ryu KW, Gamble MJ, K aus WL (2012) Regula ion o poly(ADP- ibose) polyme ase-1- dependen gene exp ession h ough p omo e -di ec ed ec ui men o a nuclea NAD + syn hase. J Biol Chem 287:12405 –12416 License: This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion 4.0 License, which pe mi s use, dis ibu ion and ep oduc- ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. 20 o 20 EMBO Molecula Medicine 10:e8816 |2018 ª2018 The Au ho s EMBO Molecula Medicine PARP-1impinges on DDR gene exp ession Ma hew J Schiewe e al Published online: No embe 21, 2018