CIP2A is a candidate therapeutic target in clinically challenging prostate cancer cell populations

Onco a ge 19661 www.impac jou nals.com/onco a ge www.impac jou nals.com/onco a ge / Onco a ge , Vol. 6, No. 23 CIP2A is a candida e he apeu ic a ge in clinically challenging p os a e cance cell popula ions Anchi Khanna1,2, Jayan K. Rane3, Ka i K. Ki inummi1,4, Al onso U banucci1,5,6, Me ja A. Helenius1, Teemu T. Tolonen1, Ou i R. Sa amäki1, Leena La onen1, Visa Manni1, John E. Pimanda2, No man J. Mai land3, Jukka Wes e ma ck7,8 and Tapio Visako pi1 1 P os a e Cance Resea ch Cen e (PCRC), Ins i u e o Biosciences and Medical Technology (BioMediTech), Uni e si y o Tampe e and Tampe e Uni e si y Hospi al, Tampe e, Finland 2 Adul Cance P og am, The P ince o Wales Clinical School, Lowy Cance Resea ch Cen e, UNSW Medicine, Uni e si y o New Sou h Wales, Sydney, Aus alia 3 YCR Cance Resea ch Uni , Depa men o Biology, The Uni e si y o Yo k, Hesling on, Uni ed Kingdom 4 Depa men o Signal P ocessing, Tampe e Uni e si y o Technology, Tampe e, Finland 5 Cen e o Molecula Medicine No way (NCMM), No dic EMBL Pa ne ship, Uni e si y o Oslo and Oslo Uni e si y Hospi al, Oslo, No way 6 Depa men o Cance P e en ion, Ins i u e o Cance Resea ch, Oslo Uni e si y Hospi al, Oslo, No way 7 Tu ku Cen e o Bio echnology, Uni e si y o Tu ku and Åbo Akademi Uni e si y, Tu ku, Finland 8 Depa men o Pa hology, Uni e si y o Tu ku, Tu ku, Finland Co espondence o: Tapio Visako pi, email: [email p o ec ed] Co espondence o: Anchi Khanna, email: [email p o ec ed] Keywo ds: CIP2A, and ogen ecep o , cas a ion- esis an p os a e cance , cance s em-like cells Recei ed: Decembe 30, 2014 Accep ed: Ap il 03, 2015 Published: Ap il 19, 2015 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. ABSTRACT Residual and ogen ecep o (AR)-signaling and p esence o cance s em-like cells (SCs) a e he wo eme ging pa adigms o clinically challenging cas a ion- esis an p os a e cance (CRPC). The e o e, iden i ica ion o AR- a ge p o eins ha a e also o e exp essed in he cance SC popula ion would be an a ac i e he apeu ic app oach. Ou analysis o o e h ee hund ed clinical samples and pa ien -de i ed p os a e epi helial cul u es (PPECs), e ealed Cance ous inhibi o o p o ein phospha ase 2A (CIP2A) as one such a ge . CIP2A is signi ican ly o e exp essed in bo h ho mone- naï e p os a e cance (HN-PC) and CRPC pa ien s . CIP2A is also o e exp essed, by 3- and 30- old, in HN-PC and CRPC SCs espec i ely. In i o binding o he AR o he in onic egion o CIP2A and i s unc ionali y in he AR-mode a e and AR-high exp essing LNCaP cell-model sys ems is also demons a ed. Fu he , we show ha AR posi i ely egula es CIP2A exp ession, bo h a he mRNA and p o ein le el. Finally, CIP2A deple ion educed cell iabili y and colony o ming e iciency o AR-independen PPECs as well as AR- esponsi e LNCaP cells, in which ancho age-independen g ow h is also impai ed. These indings iden i y CIP2A as a common denomina o o AR-signaling and cance SC unc ionali y, highligh ing i s po en ial he apeu ic signi icance in he mos clinically challenging p os a e pa hology: cas a ion- esis an p os a e cance . INTRODUCTION Since AR is ins umen al in he p og ession o p os a e cance (PC) [1], i has been he main ocus o he apeu ics. And ogen abla ion, ei he chemical o su gical, has been he mains ay o ea men o Onco a ge 19662 www.impac jou nals.com/onco a ge ad anced PC. Howe e due o subsequen esis ance o and ogen abla ion, his app oach ails o achie e a cu e. The esul ing ecu en umo s a e commonly known as cas a ion- esis an p os a e cance s (CRPCs). I has now been demons a ed ha hese cells emain and ogen sensi i e [2]. One hi d o he CRPCs ha bo ampli ica ion o he AR gene [1] . In addi ion, ea angemen s o AR, esul ing in ligand independen ac i a ion o he ecep o has been epo ed [2]. Also, in an iand ogen ea ed umo s, poin mu a ions ab oga ing he e ec o he d ugs ha e been iden i ied [3] and i has been shown ha CRPC cells can p oduce low le els o and ogens endogenously [4]. These adap a ions make CRPCs esponsi e o he second line AR-signaling blocking d ugs, such as abi a e one and enzalu amide. Bu e en hese second line d ugs a e e ec i e only in a p opo ion o pa ien s and ail o comple ely cu e esponsi e pa ien s [5]. On he o he hand, he exis ence o cance s em cells (SCs) in ad anced s age PC and CRPCs has been sugges ed as ano he po en ial mechanism o esis ance o d ugs [6]. Se e al ecen s udies ha e sugges ed ha cance SCs a e and ogen-independen [7, 8]. Toge he hese esul s indica e ha e icien he apy o ad anced s age PC would equi e simul aneous a ge ing o wo ypes o cance cell popula ions: hose ha a e dependen on AR-signaling and he AR-independen (cells ha a e non- esponsi e o and ogens) cance s em-like cells. CIP2A, an endogenous inhibi o o p o ein phospha ase 2A (PP2A), is a ecen ly iden i ied oncogene, which is o e exp essed in se e al umou ypes a a e y high equency [9-14]. In ac , i s o e exp ession is eme ging as one o he mos equen al e a ions in human cance s [13]. MYC, E2F1, AKT and mTOR ha e so a been iden i ied as downs eam oncogenic e ec o s o CIP2A [13, 15]. The clinical ele ance o CIP2A, in he o m o a p ognos ic ma ke , has been es ablished o di e en human cance s [13]. No ably, in a s udy in ol ing small numbe o cases, CIP2A o e exp ession was obse ed in ho mone-naï e p os a e cance (HN- PC) specimens and i s exp ession was associa ed wi h poo ly di e en ia ed and high- isk PCs [16]. Howe e , i s unc ional ole in p os a e cance cell biology s ill emains elusi e. RESULTS To s udy he ole o CIP2A in p os a e cance , we i s in es iga ed he exp ession and and ogen egula ion o his gene in clinical issue samples, p os a e cance cell lines and di e en cell popula ions om benign p os a e hype ophy (BPH), HN-PC and CRPC pa ien samples. CIP2A was o e exp essed in HN-PC and e en mo e in CRPC compa ed o BPH a he mRNA le el (Figu e 1A). Also immunos aining showed highe p o ein le els in CRPC compa ed o HN-PC (Figu e 1B, 1C). No ably, 99% o CRPC pa ien samples we e posi i e o CIP2A exp ession (Figu e 1C). CIP2A p o ein exp ession was also demons a ed in AR-posi i e PC cell lines (Figu e 1D). Nex , we assessed he exp ession o CIP2A in he di e en popula ions o cells om biopsy samples o BPH, HN-PC and CRPC pa ien s. We isola ed he s em-like cell (SC), ansien ampli ying cell (TA) and commi ed basal cell (CB) popula ions using CD133, CD44 and α 2 β 1 cell su ace ma ke s as shown in Figu e S1A. We ha e p e iously demons a ed ha p ima y HN-PC issue and SC cells o m se ially ansplan able subcu aneous umo s in Rag2 -/- γC -/- immune comp omised mice, which can o m dis an me as asis and mul iple in ap os a ic umo s in nude mice when g a ed o ho opically in a ma igel plug con aining human p os a ic s oma [17]. As shown in Figu e 2A, he exp ession o CIP2A in he cance s em-like cell (SC) popula ion, was inc eased by abou h ee- old in HN-PC-SCs and by abou hi y- old in CRPC-SCs compa ed o BPH-SCs. In con as , CIP2A exp ession was dec eased in TA and CB cell popula ions in cance (Figu e S1B, S1C). Func ionally, deple ion o CIP2A in he p ima y HN-PC PPECs cells wi h wo di e en siRNAs (Figu e 2B) esul ed in dec ease in bo h cell iabili y (Figu e 2C) and colony o ming capaci y (Figu e 2D). Pa icula ly, he colony o ming capaci y in he BPH PPECs was no signi ican ly al e ed on CIP2A deple ion (Figu e 2D). Toge he hese esul s indica e ha CIP2A is impo an o he su i al o HN-PC s em-like cell (SC) popula ions. Ha ing con i med unc ional signi icance, we nex in es iga ed he egula ion o CIP2A exp ession. Al hough, o e exp ession and a unc ional signi icance o CIP2A in and ogen-independen cance SCs implied AR-independen e ec s o CIP2A, we also ha e s ong e idence o AR-media ed CIP2A egula ion in AR- esponsi e cells. A s a is ically signi ican associa ion o CIP2A immunoposi i i y wi h AR nuclea s aining was obse ed in p ima y HN-PC samples (Figu e S2A;[18]). ChIP-Seq da a in p os a e cance cell lines om p e iously published s udies e ealed an AR binding si e in CIP2A in onic egion (Figu e S2B). The in i o unc ionali y o his si e was demons a ed by an inc ease in AR binding a his locus upon dihyd o es os e one (DHT) induc ion in LNCaP cells using ChIP-qPCR (Figu e 3A). The LNCaP cells modi ied o exp essed 2-4 (ARmo) and 6- (ARhi) old highe le els o AR compa ed o con ol (pcDNA3.1) LNCaP cells [19] showed s onge binding (Figu e 3A) upon DHT ea men . In addi ion, an inc ease in CIP2A mRNA exp ession was seen in LNCaP-ARhi compa ed o LNCaP-pcDNA3.1 cells a e ea men wi h e en modes doses o DHT (Figu e 3B and S3A). Simila ly, highe CIP2A p o ein le els we e obse ed in LNCaP-ARhi cells in compa ison o LNCaP-pcDNA3.1 cells wi h low DHT s imula ion (Figu e 3C). Also, immuno luo escene s aining demons a ed highe CIP2A p o ein le els in LNCaP-ARhi compa ed o LNCaP-pcDNA3.1 cells a e Onco a ge 19663 www.impac jou nals.com/onco a ge 1nM DHT (24h) ea men (Figu e S3B). These indings a e in line wi h ou p e ious obse a ions ha and ogen- egula ed genes a e induced in cells o e exp essing AR, e en in lowe ligand concen a ions [19]. Nex , an independen siRNA based app oach o e i y he ole o and ogen ecep o in CIP2A egula ion was used. We ans ec ed wo known AR-posi i e p os a e cance cell lines, LNCaP and VCaP (Figu e 3D), wi h siRNAs agains AR and es ima ed he CIP2A p o ein le els. As shown in Figu e 3D, AR deple ion esul ed in dec eased CIP2A p o ein exp ession in bo h he AR-posi i e p os a e cance cell lines. To assess whe he he AR-media ed posi i e egula ion o CIP2A exp ession has any unc ional consequence, we ans ec ed LNCaP-pcDNA3.1 and LNCaP-ARhi cells wi h CIP2A siRNAs and pe o med he unc ional assays. As demons a ed in Figu e 4A, abou 4 o 5 old dec ease in cell iabili y was obse ed in bo h LNCaP-pcDNA3.1 and LNCaP-ARhi cells. Fu he , we demons a ed ha CIP2A p omo es clonogenici y (Figu e 4B) and cellula ans o ma ion po en ial (Figu e 4C) o LNCaP-pcDNA3.1 and -ARhi cells using monolaye clonogenic and ancho age-independen so aga assays, espec i ely. These unc ional s udies clea ly iden i y AR as an ups eam egula o o CIP2A, which in u n posi i ely modula es cell su i al. Figu e 1: CIP2A is o e exp essed in p os a e cance . A. O e exp ession o CIP2A mRNA in cas a ion esis an p os a e cance (CRPC; n = 15) ho mone-naï e p ima y p os a e cance (HN-PC; n = 27) compa ed o benign p os a ic hype plasia (BPH; n = 8). * = n < 0.05, *** = p alue < 0.001 using he K uskal-Wallis es wi h Dunn’s mul iple compa ison pos - es . B. CIP2A p o ein exp ession in p os a e cance specimens as de ec ed by immunohis ochemis y. A sco e o 0 (nega i e), 1 (weak), 2 (mode a e) and 3 (high) was gi en based on he cy oplasmic CIP2A immunoposi i i y. C. S a is ical analysis o CIP2A immunoposi i i y in bo h HN-PC and CRPC cases. D. Wes e n blo showing CIP2A p o ein le els in benign and malignan p os a e cance (PC) cells. Onco a ge 19664 www.impac jou nals.com/onco a ge DISCUSSION O e exp ession o AR has been commonly obse ed no only in p ima y PCs bu also in CRPC cases [1, 2, 4]. I s ole in sensi izing umo cells o low le els o and ogens has been a c i ical de e en o e ec i e he apies in p os a e cance . Addi ionally, p esence o cance s em cells adds ano he laye o complexi y o e ec i e ea men , especially in CRPC cases [6]. The e o e, new he apeu ic app oaches ha o e come hese hu dles a e u gen ly equi ed o an e ec i e long- e m he apy agains p os a e cance . This can be achie ed by ei he using d ug combina ions ha a ge AR-signaling and cance SC-signaling simul aneously, o by iden i ying and a ge ing e ec o p o eins common o hese signaling cascades. In his s udy using pa ien -de i ed p os a e epi helial cul u es (PPECs) and o e h ee hund ed clinical samples, we iden i y CIP2A o be a common denomina o o AR and cance SC-signaling unc ionali y. We demons a e ha CIP2A is o e exp essed in PC and CRPC cases and p omo es he iabili y and clonogenici y o AR- esponsi e p os a e cance cells. No ably, he di e ence in g ow h in he LNCaP-pcDNA3.1 and LNCaP-ARhi cells was no clea ly appa en while doing he siRNA based unc ional assays (Figu e 4A-4C) despi e di e en AR le els in hese wo popula ions. This could be ei he be due o blun ing o he induc ion o CIP2A p o ein le els in ARhi cells compa ed o he pa en al LNCaP cells (1.6 old V’s 3.4 old) as shown in Figu e 3C, o due o usage o no mal se um medium o ca y ou he knock-down assays as he Figu e 2: CIP2A is o e exp essed and p omo es g ow h and iabili y o cance s em-like cell (SC) popula ion om pa ien -de i ed p os a e epi helial cell (PPECs) cul u es. A. O e exp ession o CIP2A mRNA in he s em-like cell (SC) popula ion o bo h p ima y HN-PC and CRPC pa ien samples in compa ison o BPH samples. No e he y-axis scale change be ween HN-PC and CRPC panels. B. Wes e n blo showing he e ec o wo di e en CIP2A siRNAs on CIP2A p o ein le els in p ima y HN-PC-SC cells in compa ison o he Sc ambled (Sc .) o con ol ans ec ed cells. C. E ec o CIP2A deple ion, using wo di e en siRNAs agains CIP2A, on cell iabili y o HN-PC-SC cells om pa ien s in compa ison o he con ol/Sc . ans ec ed cells a 3 and 6 day ime poin . ( * = p- alue < 0.05 using s uden - es ) D. E ec o CIP2A deple ion on colony o ming capaci y o p ima y cells om BPH and HN-PC pa ien s ( o al cell popula ions) in compa ison o he con ol/Sc . ans ec ed cells exp essed as 100% o each. Shown a e he e ec s on bo h small (less han 32 cells) and big colonies (mo e han 32 cells ; * = p- alue < 0.05 using s uden - es ) Onco a ge 19665 www.impac jou nals.com/onco a ge s ess on he cells, due o he combina ion o ans ec ion eagen and p esence o s ipped medium o low DHT condi ions, is oxic o he cells. Fu he , we highligh ed he ole o CIP2A p o ein in p omo ing he su i al o he PC s em-like cell (SC) popula ions. Impo an ly, se e al ecen s udies ha e demons a ed he umo supp essi e ole o PP2A complex in p os a e cance [20-22]. Based on hese indings, we sugges he apeu ic a ge ing o CIP2A, an es ablished endogenous inhibi o o PP2A [10, 11, 14] as a no el app oach o simul aneous a ge bo h he luminal ype mass o p os a e cance , as well as he umo ini ia ing capaci y o he SC popula ions (Figu e 4D). Howe e , addi ional mouse p e-clinical s udies a e equi ed o u he s eng hen CIP2A’s ole in he apy agains p os a e cance . Al oge he , as CIP2A deple ion in many o he cance models ha e been demons a ed o esul in obus an i umo e ec s [13, 23] and inhibi ion o ac i i y o se e al oncogenic d i e pa hways [13, 15, 23], hese no el esul s highligh he po en ial emedial ole o CIP2A in he mo e esis an and he apeu ically challenging p os a e cance cases. MATERIALS AND METHODS Clinical samples o mRNA exp ession F eshly ozen p ima y p os a e cance samples om p os a ec omies (8 BPH and 27 un ea ed o HNPC in addi ion o 15 CRPC specimens om ansu e h al esec ion o he p os a e- ea ed pa ien s) we e used in he s udy. The samples we e snap ozen in liquid ni ogen and o al RNA was isola ed wi h T izol-Reagen (In i ogen Inc.) acco ding o he manu ac u e ’s ins uc ions. Tumo samples con ained, a leas , 70% cance cells. The use o he clinical ma e ial has been app o ed by he e hical commi ee o he Tampe e Uni e si y Hospi al. Figu e 3: And ogen ecep o (AR) binds o CIP2A and i s ac i i y and exp ession posi i ely egula es CIP2A exp ession in p os a e cance . A. AR binding on CIP2A in onic egion. ChIP-qPCR on LNCaP-pcDNA3.1, -ARmo, -ARhi cells, ho mone s a ed o 4 days ollowed by ea men wi h 1nM and 100nM DHT o 2h o E hanol was pe o med o assess he AR ec ui men on he CIP2A in onic egion. B. E ec on CIP2A mRNA exp ession in LNCaP-pcDNA3.1, -ARhi cells on DHT (1nM, 10nM, 100nM) s imula ion a 24h ime poin . C. E ec on CIP2A p o ein exp ession in LNCaP-pcDNA3.1, -ARhi cells on DHT (1nM) s imula ion a 24h ime poin acco ding o Wes e n blo ing. D. E ec o wo independen AR siRNAs on CIP2A and AR p o ein exp ession in LNCaP cells 72h pos - ans ec ion and e ec o AR and CIP2A siRNAs on CIP2A and AR p o ein exp ession in VCaP cells 72h pos - ans ec ion. Onco a ge 19666 www.impac jou nals.com/onco a ge Figu e 4: CIP2A p omo es iabili y, clonogenici y and ans o ma ion po en ial o AR-posi i e p os a e cance cell lines. A. E ec o wo independen CIP2A siRNAs on cell iabili y o LNCaP-pcDNA3.1, -ARhi cells 3 and 6 days pos - ans ec ion. S uden ’s - es was used o ob ain he p alues. B. E ec o CIP2A siRNA on clonogenici y o LNCaP cells 6 days pos - ans ec ion. S uden ’s - es was used o ob ain he p alues. C. E ec o wo di e en CIP2A siRNAs on ancho age-independen g ow h. He e, LNCaP cells we e ans ec ed wi h CIP2A.1, CIP2A.2 o Sc . siRNA and, a 48 hou s pos ans ec ion, pla ed a low densi y in so aga medium. S uden ’s - es was used o ob ain he p alues. D. Schema ic ep esen a ion o he he apeu ic po en ial o a ge ing CIP2A. Since CIP2A posi i i y is common o bo h AR-dependen and cance SC popula ions o p os a e cance , i su aces as an a ac i e he apeu ic a ge , especially in clinically challenging cas a ion- esis an p os a e cance s. Onco a ge 19667 www.impac jou nals.com/onco a ge Tissue mic oa ays (TMAs) Tissue mic oa ays con ained 185 o malin- ixed pa a in-embedded p os a ec omy and 92 CRPC ( ansu e h al esec ion o he p os a e) specimens ob ained om Tampe e Uni e si y Hospi al. Fo he p os a ec omy- ea ed pa ien s, de ec able PSA alues ≥0.5 ng/ml) in wo consecu i e measu emen s o he eme gence o me as asis we e conside ed as signs o p og ession. The use o issue mic oa ays has been app o ed by he e hical commi ee o Tampe e Uni e si y Hospi al and he Na ional Au ho i y o Medicolegal A ai s. CIP2A immunoposi i i y was g aded in one o h ee umou co es o each pa ien based on he in ensi y o he cy oplasmic immuno eac i i y in he cance cells, ha is, 3 was s ong, 2 mode a e, 1 weak, and 0 nega i e. All samples we e sco ed by, an expe ienced u opa hologis (T. Tolonen), wi hou knowledge o clinical s a us and ou come da a. Es ablishmen o p ima y p os a e epi helial cul u es and en ichmen o p os a e epi helial sub-popula ions The epi helial cul u es om p ima y p os a e samples we e es ablished acco ding o an es ablished p o ocol [24]. The pos ope a i e p os a e samples we e ob ained wi h ull e hical consen and we e chopped in o ine pieces, which we e hen subjec ed o o e nigh collagenase (Lo ne Labo a o ies) ea men a 37 o C. Then, he s oma was sepa a ed by di e en ial cen i uga ion and p os a e epi helial acini we e u he dis up ed by mechanical and enzyma ic sepa a ion me hods. A his poin , CD24 posi i e luminal cells we e emo ed and en iched by employing CD24 MACS immunomagne ic beads (Mil enyi Bio ec). The es o he p os a e epi helial cells we e hen g own wi h i adia ed STOs on collagen I coa ed plas ic pla es in a s em cell media a 37oC wi h 5% CO2. Cells we e sub-cul u ed ypically when hey we e 80-90% con luen by spli ing hem in o 1:3 o 1:4 p opo ions. A passage wo; cul u ed epi helial cells we e selec ed on he basis o exp ession o α2β1-in eg in, CD44, and CD133 exp ession by employing apid collagen adhesion and CD133 MACS ac iona ion (Figu e S1A). Cell lines and cell cul u e LNCaP cells we e pu chased om Ame ican Type Cul u e Collec ion (ATCC) and well es ablished LNCaP- pcDNA3.1, LNCaP-ARmo and LNCaP-ARhi sub cell lines [19] exp essing low (endogenous AR le els), mode a e ( wo o ou old) and high AR ( ou o six old) AR exp ession (26) espec i ely we e also g own as pe he ecommended condi ions. VCaP cells used we e a kind gi by D . Jack Schalken- Radboud Uni e si y, Nijmegen Medical Cen e , and we e g own unde he ecommended condi ions. Wi h espec o p ima y cells p os a e cance di ec ed co e biopsies om h ee HN-PC pa ien s had Gleason sco e o 4+3 while CRPCs we e Gleason 9 (4+5, one sample) and 10 (5+5, 2 samples) we e p ocessed o es ablish cul u es in s em cell medium (SCM) as desc ibed be o e [24]. Quan i a i e (q) RT-PCR Cells we e collec ed om dishes and hei o al RNA was ex ac ed using T izol (In i ogen) acco ding o he manu ac u e ’s ins uc ions. Fi s -s and cDNA syn hesis was ca ied ou om o al RNA using AMV e e se ansc ip ase (Finnzymes) acco ding o he manu ac u e ’s ins uc ions. The p ime s o Q-RT-PCR we e designed wi h P ime 3 p og am. SYBR G een II-Fas S a ki (Roche Diagnos ics) and Ligh Cycle appa a us (Roche Diagnos ics) we e used o Q-RT-PCR. TBP (TATA box binding p o ein) and Ac in mRNA was used as a e e ence. The speci ici y o he eac ions was con i med, in addi ion o he mel ing cu e analysis, wi h 1.5% aga ose gel elec opho esis. Ac in F- 5´- ACTTCACATCACAGCTCCCC -3´ Ac in R- 5´- GAATATAATCCCAAGCGGTTTG -3´ TBP – F - 5´- CATAGGAATCCTTCTGACCCATG-3´ TBP R- 5´- CGAGCACAGAGCCTCGCCTTTGC-3´ CIP2A F- 5′-CTGGTGAGATAATCAGCAATTT-3′ CIP2A R- 5′-CGAAACATTCATCAGACTTTTCA-3′ Small in e e ing RNA (siRNA) ans ec ions The siRNAs o inhibi CIP2A exp ession we e ob ained om Eu o ins MWG ope on (Ebe sbe g, Ge many). The ollowing double-s anded oligonucleo ides we e used: CIP2A.1-5′-CUGUGGUUGUGUUUGCACUTT-3′ CIP2A.2-5′-ACCAUUGAUAUCCUU AGAATT-3′ AR.1(sense)-CCCAAGAUCCUUUCUGGGA+UU AR.2 (sense)-GUGGAAGAUUCAGCCAAG C+UU Sc ambled-5′-UAACAAUGAGAGCACGGCTT-3 ′ Cells a 60%con luency we e ans ec ed wi h 100nM in 3 mL o an ibio ic ee medium (RPMI-1640 medium supplemen ed wi h 10% FCS) in each well wi h he espec i e siRNAs using Lipo ec amine 2000 Reagen (Li e Technologies, Ca lsbad, CA), acco ding o he manu ac u e ’s ins uc ions. Fo he p ima y PC cells (pa ien -de i ed p os a e cance cells, Gl-4+3), Cells a 70% con luency we e ans ec ed wi h 100nM o espec i e siRNA in 500 μL o Op i-MEM se um ee medium (Li e Technology) in each o he wells o Onco a ge 19668 www.impac jou nals.com/onco a ge a 6-well pla e using Oligo ec amine (Li e Technology). 2mL o SCM was added a e 4 hou s in each well. A e 4 mo e hou s, wells we e washed wi h PBS wice and 2 mL o esh SCM was added. Cells we e ha es ed a 72 hou s a e ans ec ion o de e mine he e ec on p o ein exp ession. Immuno lou escence LNCaP-pcDNA3.1 and LNCaP-ARhi cells we e seeded on s e ilized co e slips placed in each well o a 6-well pla e o e nigh and he cells we e hen ea ed wi h 1nM DHT o 24hou s. Subsequen ly he co e slips we e placed in esh s e ilized wells and hen washed wi h PBS be o e being ixed using 4% pa a o maldehyde o 10 min a 37° C. Then a e washing wi h PBS wice, cells we e ea ed wi h 0.5% NP-40 o 5min a oom empe a u e. This was ollowed by blocking he cells wi h 0.5% bo ine se um albumin o 30 min. Cells we e hen exposed o he ollowing p ima y an ibody dilu ions, CIP2A; 1: 250 (San a C uz Bio echnology, San a C uz, CA). A e which he cells we e washed wi h PBS 3 imes and incuba ed in FITC-conjuga ed goa an i Rabbi IgG (In i ogen– Molecula P obes, Eugene, O egon, USA) o 1h a 37°C. Nex cells we e hen s ained wi h he Vec ashield an i ade solu ion (Vec o Labo a o ies) con aining 0.001 mol/L 4′,6-diamidino-2- phenylindole as coun e s ain and cells obse ed unde he mic oscope and he images aken. Ch oma in immunop ecipi a ion (ChIP) CIP2A ARBS was iden i ied in p e iously published ChIP-seq da ase and ChIP was ca ied ou as desc ibed be o e [25]. Ch oma in was immunop ecipi a ed wi h 10 µg o no mal abbi IgG, 10 µl o an i-AR polyclonal an ibody (AR3). P ime sequences used o he CIP2A ARBS ampli ied ia qPCR examined in his wo k a e – CIP2A Fo wa d -5’ – TTGTTCCTGGTCTTCCCTGT-3’, CIP2A Re e se -5’ – GCAAAGTTTGATTGGTTTTTCC-3’ Immunohis ochemis y Mouse An i-CIP2A (No us Biosciences.) was used wi h Powe Vision+ Poly-HRP IHC ki (ImmunoVision Technologies Co., Bu lingame, CA, USA) acco ding o he manu ac u e ’s ins uc ions. The p o ocol has p e iously been desc ibed [18, 26]. Wes e n blo ing P o eins we e ex ac ed in ho Laemmli sample bu e and subjec ed o Wes e n blo analysis. 30 µg o al p o ein ex ac s we e sepa a ed by 12% SDS-PAGE and ans e ed o ni ocellulose memb anes. Memb anes we e blocked wi h 5% non- a milk in TBS-0.1%-NP40 and hen incuba ed wi h mouse monoclonal AR (AR441, NeoMa ke s-Lab Vision Co po a ion, F emon , CA, USA), goa polyclonal an i-β-Ac in (San a C uz, Bio echnology, San a C uz, CA, USA) o wi h mouse monoclonal an i- CIP2A (San a C uz Bio echnology, San a C uz, CA). Fo p ima y PC cells Cy oBus e (Me ck Millipo e) bu e was used o lyse eshly cul u ed cells o ex ac p o eins. Cell iabili y One day be o e ans ec ion o CIP2A.1 o CIP2A.2 siRNAs, LNCaP and LNCaP-ARhi cells we e seeded in RPMI-1640 medium supplemen ed wi h 10% FCS a a densi y o 1 × 103 o 2 × 103 cells pe well in 96-well pla es. The cells we e ans ec ed wi h he ollowing -Lipo ec amine 2000 eagen , 33 nM sc ambled siRNA o 33 nM CIP2A.1 o CIP2A.2. Cells we e hen cul u ed o 3 and 6 days pos - ans ec ion be o e ela i e numbe s o iable cells we e measu ed by luo escence a he 544 and 590 nm wa eleng hs in a FLUOs a OPTIMA Mic opla e Reade (BMG Lab ech, Inc., Du ham, NC), using he esazu in-based CellTi e -Blue Assay (P omega Co po a ion) acco ding o he manu ac u e ’s ins uc ions. In case o p ima y PC cells, 20µl o cell suspension was added o an equal olume o 0.4% T ypan Blue S ain (Sigma-Ald ich). The li e cells we e coun ed using modi ied Neubaue ’s haemocy ome e . So aga and monolaye clonogenic assays A 48 hou s a e ans ec ion, LNCaP (1 × 104 pe dish) we e suspended in 1 mL o 0.25% aga ose (GellyPho ; Eu oClone Spa, Pe o, I aly) supplemen ed wi h 2 mL comple e RPMI-1640 cul u e medium (changed e e y hi d day). This suspension was laye ed o e 1 mL o a base laye o 0.5% aga ose in comple e medium in six-well pla es. A e 8 o 12 days in aga ose, cells we e s ained wi h Giemsa 1:20 in wa e (Sigma- Ald ich), he wells we e scanned using he Su eyo So wa e (Objec i e Imaging L d.) wi h came a (Imaging Inc., Canada) a ached o he Olympus IX71 (Olympus, Tokyo, Japan) mic oscope and colonies we e coun ed by analysis wi h ImageJ So wa e (Wayne Rasband, Na ional Ins i u es o Heal h, Be heseda, MD). Cell g oupings ha we e g ea e han 1200 pixels in diame e wi h 3.2× enla gemen s we e coun ed as colonies. In case o p ima y PC cells, CIP2A and sc ambled siRNAs we e ans ec ed in o hese cells o a day, a e which hey we e ypsynised and 250 cells we e pla ed in each well o he 6-well Collagen-I coa ed pla es (BD Biosciences). Cells we e pla ed in iplica es wi h 2 ml o s em cell medium (SCM) and 500 µl o i adia ed STO eede cells pe well. Onco a ge 19669 www.impac jou nals.com/onco a ge The media was changed e e y 2-3 days. Clonal colonies we e coun ed a he end o 12 days unde a s anda d ligh mic oscope. Fo isualiza ion, colonies we e s ained wi h c ys al iole solu ion (Sigma-Ald ich). S a is ical analysis The associa ion be ween he CIP2A posi i i y in p ima y HN-PC, CRPC umo s and wi h AR exp ession was done using chi-squa e es . K uskal-Wallis and s uden - es s we e used o o he expe imen s. All s a is ical es s we e wo-sided. FINANCIAL SUPPORT This wo k was suppo ed by he Academy o Finland, he Cance Socie y o Finland, he Sig id Juselius Founda ion, and he Medical Resea ch Fund o Tampe e Uni e si y Hospi al. AK is a Na ional Heal h and Medical Resea ch Council (Pe e -Dohe y) Resea ch Fellow, Aus alia. CONFLICTS OF INTEREST None. Abb e ia ions And ogen Recep o (AR); Cance ous Inhibi o o P o ein phospha ase 2A (CIP2A); cas a ion- esis an p os a e cance (CRPC); p os a e cance (PC); ho mone- naï e p os a e cance (HN-PC); benign p os a e hype ophy (BPH); pa ien -de i ed p os a e epi helial cul u es (PPECs) ; dihyd o es os e one (DHT) ; s em-like cell (SC); ansien ampli ying cell (TA); commi ed basal cell (CB). REFERENCES 1. Visako pi T, Hyy inen E, Koi is o P, Tanne M, Keinanen R, Palmbe g C, Palo ie A, Tammela T, Isola J and Kallioniemi OP. In i o ampli ica ion o he and ogen ecep o gene and p og ession o human p os a e cance . Na u e gene ics. 1995; 9:401-406. 2. Wal e ing KK, U banucci A and Visako pi T. And ogen ecep o (AR) abe a ions in cas a ion- esis an p os a e cance . Molecula and cellula endoc inology. 2012; 360:38-43. 3. Ha a T, Miyazaki J, A aki H, Yamaoka M, Kanzaki N, Kusaka M and Miyamo o M. No el mu a ions o and ogen ecep o : a possible mechanism o bicalu amide wi hd awal synd ome. Cance esea ch. 2003; 63:149-153. 4. Visako pi T. No el endoc ine aspec s o p os a e cance . Molecula and cellula endoc inology. 2012; 360:1-2. 5. Rane JK, Pellacani D and Mai land NJ. Ad anced p os a e cance --a case o adju an di e en ia ion he apy. Na u e e iews U ology. 2012; 9:595-602. 6. Mai land NJ and Collins AT. P os a e cance s em cells: a new a ge o he apy. Jou nal o clinical oncology : o icial jou nal o he Ame ican Socie y o Clinical Oncology. 2008; 26:2862-2870. 7. Old idge EE, Pellacani D, Collins AT and Mai land NJ. P os a e cance s em cells: a e hey and ogen- esponsi e? Molecula and cellula endoc inology. 2012; 360:14-24. 8. Qin J, Liu X, La in B, Chen X, Choy G, Je e CR, Calhoun- Da is T, Li H, Palapa u GS, Pang S, Lin K, Huang J, I ano I, Li W, Su aneni MV and Tang DG. The PSA(-/ lo) p os a e cance cell popula ion ha bo s sel - enewing long- e m umo -p opaga ing cells ha esis cas a ion. Cell s em cell. 2012; 10:556-569. 9. Khanna A. Regula ion o Cance ous inhibi o o PP2A (CIP2A) by small molecule inhibi o o c-Jun NH 2-Te minal Kinases (JNKs), SP600125, in Human Fib osa coma (HT1080) cells. F1000Resea ch. 2013; 2. 10. Khanna A, Bockelman C, Hemmes A, Jun ila MR, Wiks en JP, Lundin M, Junnila S, Mu phy DJ, E an GI, Haglund C, Wes e ma ck J and Ris imaki A. MYC-dependen egula ion and p ognos ic ole o CIP2A in gas ic cance . J Na l Cance Ins . 2009; 101:793-805. 11. Khanna A, Kauko O, Bockelman C, Laine A, Sch eck I, Pa anen JI, Szwajda A, Bo mann S, Bilgen T, Helenius M, Pokha el YR, Pimanda J, Russel MR, Haglund C, Cole KA, Kle s om J, e al. Chk1 a ge ing eac i a es PP2A umo supp esso ac i i y in cance cells. Cance Res. 2013; 73:6757-6769. 12. Khanna A, Okke i J, Bilgen T, Tii ikka T, Vihinen M, Visako pi T and Wes e ma ck J. ETS1 media es MEK1/2- dependen o e exp ession o cance ous inhibi o o p o ein phospha ase 2A (CIP2A) in human cance cells. PLoS One. 2011; 6:e17979. 13. Khanna A, Pimanda JE and Wes e ma ck J. Cance ous Inhibi o o P o ein Phospha ase 2A, an Eme ging Human Oncop o ein and a Po en ial Cance The apy Ta ge . Cance esea ch. 2013. 14. Jun ila MR, Puus inen P, Niemela M, Ahola R, A nold H, Bo zauw T, Ala-aho R, Nielsen C, I aska J, Taya Y, Lu SL, Lin S, Chan EK, Wang XJ, G enman R, Kas J, e al. CIP2A inhibi s PP2A in human malignancies. Cell. 2007; 130:51-62. 15. Puus inen P, Ry e A, Mo ensen M, Kohonen P, Mo ei a JM and Jaa ela M. CIP2A oncop o ein con ols cell g ow h and au ophagy h ough mTORC1 ac i a ion. J Cell Biol. 2014; 204:713-727. 16. Vaa ala MH, Vaisanen MR and Ris imaki A. CIP2A exp ession is inc eased in p os a e cance . J Exp Clin Cance Res. 2010; 29:136. 17. K oon P, Be y PA, S owe MJ, Rod igues G, Mann VM, Simms M, Bhasin D, Che ia S, Li C, Li PK, Mai land