Principles of protein targeting to the nucleolus

Full Te ms & Condi ions o access and use can be ound a h ps://www. and online.com/ac ion/jou nalIn o ma ion?jou nalCode=kncl20 Nucleus ISSN: 1949-1034 (P in ) 1949-1042 (Online) Jou nal homepage: h ps://www. and online.com/loi/kncl20 P inciples o p o ein a ge ing o he nucleolus Robe M Ma in, Goha Te -A e isyan, Hen y D He ce, Anne K Ludwig, Gisela Lä ig-Tünnemann & M C is ina Ca doso To ci e his a icle: Robe M Ma in, Goha Te -A e isyan, Hen y D He ce, Anne K Ludwig, Gisela Lä ig-Tünnemann & M C is ina Ca doso (2015) P inciples o p o ein a ge ing o he nucleolus, Nucleus, 6:4, 314-325, DOI: 10.1080/19491034.2015.1079680 To link o his a icle: h ps://doi.o g/10.1080/19491034.2015.1079680 © 2015 The Au ho (s). Published wi h license by Taylo & F ancis G oup, LLC© Robe M Ma in, Goha Te -A e isyan, Hen y D He ce, Anne K Ludwig, Gisela Lä ig- Tünnemann, and M C is ina Ca doso View supplemen a y ma e ial Published online: 01 Sep 2015. Submi you a icle o his jou nal A icle iews: 2742 View ela ed a icles View C ossma k da a Ci ing a icles: 31 View ci ing a icles P inciples o p o ein a ge ing o he nucleolus Robe M Ma in 1 , Goha Te -A e isyan 2 , Hen y D He ce 3 , Anne K Ludwig 3 , Gisela L€ a ig-T€ unnemann 4 , and M C is ina Ca doso 3, * 1 Ins i u o de Medicina Molecula ; Faculdade de Medicina; Uni e sidade de Lisboa; Lisboa, Po ugal; 2 Max Delb ueck Cen e o Molecula Medicine; Be lin, Ge many; 3 Depa men o Biology; Technische Uni e si € a Da ms ad ; Da ms ad , Ge many; 4 Cen e o S oke Resea ch Be lin Cha i  e; Uni e si € a smedizin Be lin; Campus Mi e; Be lin, Ge many Keywo ds: luo escence mic oscopy, GFP, nucleolus, nucleola localiza ion sequence, p o ein a ge ing Abb e ia ions: aa, amino acid; CPP, cell pene a ing pep ide; DDC, dense ib illa compa men ; FC, ib llla cen e s; FITC, luo es- cein iso hiocyana e; GBP, GFP binding p o ein; GC, g anula compa men ; GFP, g een luo escen p o ein; NoLS, nucleola locali- za ion sequence; NLS, nuclea localiza ion sequence; PI, p opidium iodide; TAMRA, 5-ca boxy e ame hyl hodamine; TAT, ansac i a o o ansc ip ion. The nucleolus is he hallma k o nuclea compa men aliza ion and has been shown o exe mul iple oles in cellula me abolism besides i s main unc ion as he place o RNA syn hesis and assembly o ibosomes. Nucleola p o eins dynamically localize and accumula e in his nuclea compa men ela i e o he su ounding nucleoplasm. In his s udy, we ha e assessed he molecula equi emen s ha a e necessa y and su ficien o he localiza ion and accumula ion o pep ides and p o eins inside he nucleoli o li ing cells. The da a showed ha posi i ely cha ged pep ide en i ies composed o a ginines alone and wi h an isoelec ic poin a and abo e 12.6 a e necessa y and su ficien o media ing significan nucleola accumula ion. A h eshold o 6 a ginines is necessa y o pep ides o accumula e in nucleoli, bu al eady 4 a ginines a e su ficien when used wi hin 15 amino acid esidues o a nuclea localiza ion signal o a p o ein. Using a pH sensi i e dye, we ound ha he nucleola compa men is pa icula ly acidic when compa ed o he su ounding nucleoplasm and, hence, p o ides he ideal elec ochemical en i onmen o bind poly-a ginine con aining p o eins. In ac , we ound ha oligo-a ginine pep ides and GFP usions bind RNA in i o. Consis en wi h RNA being he main binding pa ne o a ginines in he nucleolus, we ound ha he same p inciples apply o cells om insec s o man, indica ing ha his mechanism is highly conse ed h oughou e olu ion. In oduc ion P o eins inside he highly compa men alized cell nucleus o en localize in co ela ion wi h hei unc ion o disc e e subnu- clea egions also e e ed o as subnuclea bodies. 1,2 These nuclea subs uc u es we e iden i ied as si es whe e speci ic bio- chemical eac ions ake place, e.g., DNA eplica ion oci, o as si es o s o age and modi ica ion o di e en p o eins like in Cajal bodies and nuclea speckles. 3,4 The localiza ion o p o eins in he nucleus is o en media ed by specialized signals, a ge ing o in e ac ion mo i s o di e en leng h and can change o e ime, cell cycle and wi h ex e nal s imuli. 5-8 The mos p ominen subnuclea s uc u e is he nucleolus, which appea s as a da k egion in phase con as mic oscopy due o i s high densi y. 9-11 The nucleolus is he place o ansc ip ion and s o age o he DNA as well as he assembly and ma u a ion si e o ibosomes and, in ansla ionally ac i e cells, illed wi h p e- ibosomal pa - icles and componen s. The nucleola s uc u e is e y dynamic and o ms a ound DNA loci. 12,13 I has u he subs uc u es, which a e connec ed o he di e en s eps in ibosome biogenesis and can be bes iden i ied by an ibody labeling o elec on mic oscopy. The ib illa cen e s (FC) ha bo he DNA in he nucleola cen e and a e su ounded by he dense ib illa compo- nen s (DFC) con aining he nascen RNA ansc ip s. The g an- ula componen (GC) is illed wi h p e ibosomal pa icles composed o RNA and p o eins, ex ends in o he nucleus and is su ounded by ch oma in. 11,14,15 An ea ly s udy has shown ha he nuclea impo o se - e al p o eins is media ed by sho sequences o 6–10 basic amino acid (aa) e med nuclea localiza ion sequences (NLS), some o which also se e as nucleola a ge ing sequence (NoLS). 16 Ano he nucleola a ge ing sequence (NoLS) was desc ibed in 1990 and con ains a minimum o 5 basic aa, o en lysines and a ginines. 17 The ecen sys ema ic analysis o 46 NoLS sequences o hei amino acid composi ion and sequence ea u es has shown g ea e sequence di e si y © Robe M Ma in, Goha Te -A e isyan, Hen y D He ce, Anne K Ludwig, Gisela L€ a ig-T€ unnemann, and M C is ina Ca doso *Co espondence o: M C is ina Ca doso; Email: [email p o ec ed] Submi ed: 06/24/2015; Re ised: 07/29/2015; Accep ed: 07/31/2015 h p://dx.doi.o g/10.1080/19491034.2015.1079680 This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-Non-Comme cial License (h p://c ea i ecommons.o g/licenses/ by-nc/3.0/), which pe mi s un es ic ed non-comme cial 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. The mo al igh s o he named au ho (s) ha e been asse ed. 314 Volume 6 Issue 4Nucleus Nucleus 6:4, 314--325; July/Augus 2015; Published wi h license by Taylo and F ancis G oup, LLC RESEARCH PAPER including bipa i e sequences as well as o e lap wi h NLS. The s udy de e mined also a high con en o basic amino acids (48%) and a p edominan loca ion in he e mini o p o eins. 18 P o eins smalle han 60 kDa o up o a diame e o 9 nm can passi ely di use in o he nucleus 19 bu his p o- cess becomes mo e and mo e ine icien wi h inc easing size ( e iewed in 20,21 ). The nucleus is sepa a ed om he cy o- plasm by a memb ane sys em and in ol es an ac i e impo mechanism. 22 In con as , he localiza ion o p o eins o he nucleolus has ne e been shown o in ol e ac i e anspo mechanisms. Nuclea p o eins wi hou nucleola unc ion a e o en less concen a ed o comple ely excluded om he nucleoli while nucleola p o eins a e highly en iched in his nuclea compa men . The localiza ion o exclusion o p o- eins om he nucleolus is subjec o d ama ic changes upon cellula s imuli, s ess and comple e eo ganiza ion du ing cell cycle p og ession. 23-25 In a p e ious s udy, we ha e shown ha a epo e p o ein a e ses nucleoli apidly wi h less obs uc ion compa ed o nucleoplasmic egions ich in ch oma in. 26 In ha s udy, we desc ibed a s uc u al ea u e o he nucleolus wi h channel like egions oid o nucleic acids and p o eins ha acili a es a as mo emen o non-nucleola p o eins h ough his compa - men . Con as ingly, he neighbo ing densely packed s uc u es in he nucleolus educe he a ailable space o he localiza ion o non-nucleola p o eins. 26 S ill i is unclea , wha a e he essen ial p ope ies o equi emen s o p o eins and pep ides o accumula e in he nucleoli and wha a e he ea u es o hese molecules ha p e en nucleola accumula ion. The sys ema ic iden i ica ion o NoLS sequences by an a i icial neu al ne wo k ained wi h con i med NoLS sequences has ound housands o po en ial sequences ha could se e o media e nucleola a - ge ing. Se e al p e iously unknown NoLS ha e been expe i- men ally con i med he ea e . 18,27 Howe e , a conside ably alse posi i e p edic ion a e especially among sequences om nuclea p o eins aises he ques ion o how NLS and NoLS can be dis inguished. The di e ence in he composi ion o a s an- da d NLS and he pep ides ha accumula e in nucleoli is a he sub le, wi h he NLS being a ew basic cha ged amino acids sho e han, e.g., he nucleola ma ke pep ide o 10 a ginines desc ibed by us. 16,28 This obse a ion led o he ol- lowing 2 ques ions: (a) wha a e he speci ic equi emen s in amino acid composi ion o pep ides and p o eins o localize o he nucleoli and (b) wha a e he key di e ences be ween NoLS and NLS sequences? In his s udy, we sys ema ically in es iga ed he molecula p ope ies o pep ides and p o eins ha esul in di e en dis i- bu ion le els be ween nucleoplasm and nucleolus. Fo his pu - pose, we es ed and analyzed he dis ibu ion o a se ies o luo escen ly labeled sho pep ides wi h di e en cha ge and iso- elec ic p ope ies be ween he cy oplasm, nucleoplasm and hei accumula ion in he nucleolus. The same pep ide sequences we e gene ically used o he 50end o he open eading ame coding o he ace p o ein GFP, wi h no in acellula binding si es, o es he e ec o he same sho pep ide sequences on he localiza- ion o p o eins. Ma e ials and Me hods Pep ides, plasmids and cell lines All pep ides we e syn hesized om L-amino acids excep deca- a ginine (R10), which was made as 2 e sions one om L- and one om D-amino acids by Pep ide Special y Labo a o ies GmbH (Heidelbe g, Ge many). The pep ides we e coupled di ec ly o luo escein a he N- e minus, con ain a C- e minal amide-g oup and we e pu i ied by HPLC. The TAT pep ide was syn he ized wi h D-amino acids and conjuga ed wi h TAMRA as desc ibed. 29 I can be excluded ha di e en amoun s o luo es- cen labeling on he di e en pep ides play a ole in he in acel- lula dis ibu ion, since he in i o syn hesis o pep ides coupled o a esin ma e ial allows only in line addi ion o amino acids and he luo opho e. The pI alues o he pep ides and p o eins we e calcula ed using he P o Pa am ool on h p://web.expasy. o g/p o pa am/. The GFP based mammalian exp ession cons uc s we e gene - a ed by using oligonucleo ides encoding he espec i e poly-D, poly-G and poly-R sequences (supplemen a y Table 1). The oli- gos wi h o e hanging 30and 50AgeI si es we e annealed and liga ed in o he AgeI si e o an NLS-GFP cons uc based on he pEVRF ec o . 30 This ec o is iden ical o he one desc ibed be o e o he CMV-d i en exp ession o GFP- agged PCNA. 31 In addi ion, he pEGFP-C1 ec o (Clon ech, Heidelbe g, Ge - many) wi hou a NLS sequence was used. Human HeLa and HEK 293-EBNA cells, as well as mouse C2C12 myoblas s we e cul u ed as desc ibed be o e. 32,33 Pac2 zeb a ish cells we e g own a 28C in Leibo i z L-15 medium supplemen ed wi h 15% e al cal se um in p esence o 1% penicillin-s ep omycin. S 9 insec cells we e main ained in EX- CELL TM 420 Insec Se um F ee (SAFC) medium wi h L-glu a- mine (Biosciences) and 10% e al cal se um and cul u ed wi h shaking a 100 pm and 28C. The L40ccua yeas s ain used o he cu en s udy has he ollowing gene ic backg ound: (MATa his3_200 p1–901 leu2–3,112 LYS2::(lexAop)4-HIS3 u a3:: (lexAop)8- lacZ ADE2::(lexAop)8-URA3 GAL4 gal80 can1 cyh2). 34 Yeas cells we e g own a 30C in yeas ex ac /pep one/ dex ose (YPD) and we e supplemen ed wi h 2% glucose as a ca - bon sou ce. Li e cell pep ide loading and ans ec ions All pep ide expe imen s we e pe o med exclusi ely wi h li - ing cells pla ed on 8-well Ibidi chambe s. The di e en poly- lysine (poly-K) and poly-a ginine (poly-R) pep ides we e sc a ch loaded in o li ing cells. Fo he sc a ch loading o pep ides we used sy inges and sc a ched o e he chambe bo om in di e en di ec ions. 35,36 The size o he pep ides pe mi ed passi e di u- sion in o he cells and nuclei a e deli e y. A e he sc a ch load- ing he cells we e allowed o eco e o abou hal an hou a 37C be o e exchanging he medium o emo e unloaded pep i- des. Mic oscopy and image acquisi ion was s a ed immedia ely a e he p ocedu es desc ibed abo e o ci cum en pep ide sho - ening by p o eases and, hence, mislocaliza ion. Pep ide concen- a ions om 10 o 100 mm we e ini ially es ed and no appa en changes in cell iabili y we e obse ed du ing he cou se o he www. and online.com 315Nucleus expe imen s. As low concen a ions o 20 mM we e su icien o quan i a i e imaging, hese condi ions we e used o he subse- quen quan i ica ion. Sc a ch loaded cells did no show d ama ic di e ences in he le el o in acellula luo escence, likely because he pep ide concen a ion in he medium was he same in all cases. Only cells wi h no mal cellula physiology compa able o non-sc a ch loaded cells we e imaged. The c i e ia o he selec- ion o cells ha epai ed a e sc a ch loading we e: i) cells a e comple ely a ached and s e ched ou on he chambe bo om su ace; ii) no memb ane lesions o leakage o cy oplasmic con- s i uen s isible in phase con as ; iii) ound o o al shaped nuclei wi h smoo h ou line and nucleoli isible in phase con as . The ans ec ion o cells wi h plasmid cons uc s was pe - o med using he CaPO 4 -DNA co-p ecipi a ion p o ocol as desc ibed be o e, 37 excep o HEK 293-EBNA cells, which we e ans ec ed using poly-e hylenimine (1 mg/mL in ddH 2 O, pH 7; Sigma-Ald ich, S . Louis, MO, USA) as desc ibed be o e. 38 Cell ixa ion and an ibody s aining Cells g own on glass co e slips o 24 h we e incuba ed wi h 10 mM FITC-R10 o 1 h. A e washing wi h 1xPBS, cells we e ixed in 3.7% o maldehyde in PBS o 10 min and pe meabi- lized wi h 0.25% T i on X100 in PBS o 10 min. Immunos ain- ing wi h an i-B23 mouse monoclonal an ibody (clone FC82291; Sigma) was ollowed by de ec ion wi h donkey an i-mouse IgG an ibody conjuga ed wi h TexasRed (Jackson). Co e slips we e moun ed wi h Mo iol (Sigma). Mic oscopy, image acquisi ion and analysis Li e cell mic oscopy was pe o med wi h a Zeiss LSM510Me a con ocal se up moun ed on an Axio e 200 M in e ed mic oscope using a 63x phase con as plan-apoch oma oil objec i e NA1.4. The mic oscope was placed in an incuba ion chambe hea ed o 37C o main ain he cell incuba ion condi- ions (Okolab, I aly). Fo he measu emen s o plasmid ans- ec ed cells, a Leica TCS SP con ocal mic oscope was used. Fo all acquisi ion se ings he main beam spli e was HFT UV/488/ 543/633. FITC was exci ed wi h he 488 nm line o an a gon lase and luo escence de ec ed wi h a band-pass il e 500– 530 nm. Phase con as o di e en ial in e e ence con as images we e eco ded simul aneously wi h FITC luo escence in he ansmission channel. The li e-cell DNA dye DRAQ5 was used and imaged as desc ibed. 32 Cells loaded wi h pep ides we e chosen o imaging by ha ing low bu easily de ec able in acellula luo escence in ensi y. The same c i e ia we e applied o cells ans ec ed wi h plasmid con- s uc s. Imaging o li e cells was pe o med using as much as pos- sible simila se ings pe ins umen . The SNARF 4F cell-pe mean dye (Sigma) was loaded in o cells by incuba ing o 30 min a a inal concen a ion o 5 mM in g ow h medium. Fluo escence was exci ed a 561 nm and emission de ec ed a 587 nm and 640 nm. Fo image analysis, he nuclei we e iden i ied in he co e- sponding phase con as images by means o he nuclea mem- b ane delinea ion and nucleoli co espond o he da k dense s uc u es inside he cell nucleus. Iden i ica ion o nucleoli was con i med in ea lie expe imen s using a nucleola ma ke o li - ing cells. 28 The selec ion o egions (cy oplasm, nucleus, nucleo- lus) and measu emen s o mean signal in ensi y we e pe o med in ImageJ. Measu emen s o mean luo escence in ensi ies we e pe o med by excluding non-in e es a eas om he measu e- men , e.g. he nucleoli a eas excluded om nucleoplasm and, co - espondingly, o all o he egions in ensi ies we e de e mined. In each expe imen he a e age was calcula ed o 10 cells o each pep ide and p o ein. Da a we e no malized o 1.0 (100%) wi h espec o he nucleoplasmic luo escence (excluding he nucleoli) o di ec ly illus a e he ela i e accumula ion o luo escence in he nucleoli. Cy oplasmic luo escence le els we e no malized co espondingly (excluding nucleus and nucleoli). Da a analysis and s a is ical es s we e pe o med and displayed wi h O igin 7 so wa e. The comple e lis o p edic ed nucleola localiza ion sequences was downloaded om h p://www.compbio.dundee.ac.uk/www- nod/ and analyzed o mean leng h, mean numbe s o amino acids and equencies in Excel and O igin 7. P o ein p epa a ion GFP and NLS-R7-GFP we e p epa ed om HEK cells 48 h pos ans ec ion using an ice-cold bu e composed o 20 mM T is-HCl, 1.5 mM MgCl2, 0.2 mM EDTA, 1 M NaCl and 0.4% NP40. A e homogeniza ion by mechanical shea ing h ough a 23 gauge sy inge, lysa es we e incuba ed o 10 min on ice. Following cen i uga ion (14,000 pm o 12 min a 4C) supe na an s we e incuba ed o one hou wi h GFP binding p o- ein (GBP) coupled o sepha ose beads a 4C on a o a y shake as desc ibed. 39 GBP bound p o eins we e washed 3 imes wi h ice-cold HEPES bu e (140 mM NaCl, 2.5 mM KCl, 5 mM HEPES, 5 mM glycine, pH 7.4) and used o he in i o RNA pull-down assay. All bu e s we e supplemen ed wi h 1 mM o p o ease inhibi o PMSF (Ca l Ro h, Ka ls uhe, Ge many). RNA p epa a ion To al RNA was isola ed om HEK cells using he RNeasy Mini Ki (Qiagen, Hilden, Ge many) acco ding o he man- u ac u e ’s ins uc ions. To emo e aces o genomic DNA, RNA was ea ed wi h RNase- ee, ecombinan DNaseI (Mache ey Nagel, Due en, Ge many) o 30 min a 37Cand u he pu i ied wi h he Qiagen RNeasy Mini Ki . To assess he concen a ion and pu i y o RNA, he a io o abso bance a 260 nm and 280 nm was measu ed on a TECAN in ini e M200 pla e eade (Tecan G oup L d., Maennedo , Swi ze land). To u he e i y RNA quali y, o al RNA was dena u ed o 5 minu es a 99C, sepa a ed on a 1.5% T is-ace a e-EDTA/aga- ose gel supplemen ed wi h 0.05 mL/mL Ro i-Sa e GelS ain (Ca l Ro h, Ka ls uhe, Ge many) by elec opho esis and imaged on an Ame sham Image 600 (GE Heal hca e, F eibu g, Ge many). In i o RNA binding assays Fo RNA slo blo s, o al RNA was mixed wi h RNase- ee ddH 2 O a a inal concen a ion o 500 ng/mL and blo ed on p e-equilib a ed (5 min me hanol, 5 min 20x saline sodium 316 Volume 6 Issue 4Nucleus ci a e) PVDF memb anes (0.45 mm po e size; Pall GmbH, D eieich, Ge many). A e ai -d ying, memb anes we e blocked o 30 min wi h 5% milk in PBS supplemen ed wi h 1x P o- ec RNA RNase inhibi o (Sigma-Ald ich, S . Louis, MO, USA) and incuba ed o 1 h wi h ei he TAT, D-R10, o L-R10 (all luo escen ly labeled), espec i ely, on a o a y shake . Following h ee washing s eps (5 min each) wi h HEPES bu e , emaining luo escen signals we e de ec ed on an Ame sham Image 600 (GE Heal hca e, F eibu g, Ge many). Quan i ies o blo ed RNA we e moni o ed in pa allel by me hylene blue (0.02%) s aining (Ca l Ro h, Ka ls uhe, Ge many) in 0.3 M sodium ace- a e (Me ck, Da ms ad , Ge many) pH 5.5. Fo he pulldown assay, immobilized GFP and NLS-R7-GFP, espec i ely, we e incuba ed wi h equal amoun s (3.4 mg) o o al RNA in 20 mL HEPES bu e o 1.5 h a 4C on a o a y shake . A e h ee washing s eps wi h HEPES bu e , RNA was labeled wi h 3.3 mg/mL p opidium iodide (Sigma-Ald ich, S . Louis, MO, USA). Following h ee washing s eps in HEPES bu e , emaining luo escen signals we e measu ed on a TECAN in ini e M200 pla e eade (Tecan G oup L d., Maen- nedo , Swi ze land) using exci a ion/emission a 488/555 nm and 515/617 nm o GFP and p opidium iodide, espec i ely. To con ol o he amoun o GFP and NLS-R7-GFP, luo es- cen signals o p opidium iodide we e no malized o GFP signals. In addi ion, he same beads we e imaged in an Ul aVIEW VoX spinning disc con ocal sys em (Pe kinElme ), moun ed on a Nikon TI mic oscope. Images we e aken wi h a 20x/0.7 NA objec i e. GFP and p opidium iodide o TAMRA we e imaged wi h 488 and 561 nm lase exci a ion and 527 §55 and 612 § 70 nm emission il e s, espec i ely. Resul s Sho speci ic sequence mo i s in p o eins a e well known o play a key ole in localizing p o eins o speci ic cellula and sub- cellula compa men s. We ha e shown ea lie ha sho pep ides o 10 a ginines wi h a s ong basic cha ge ha e he po en ial o accumula e inside he nucleoli o li ing cells and can be used o label his subnuclea compa men . 28 Now, we wished o in es i- ga e he speci ic molecula equi emen s o pep ides and p o eins ha a e necessa y and su icien o each and accumula e inside o he nucleolus. The e o e, ou i s aim was o de e mine he p ope ies o cha ge, leng h and composi ion o basic cha ged pep ides neces- sa y o nucleola a ge ing and accumula ion. We used sho syn he ic pep ides labeled wi h FITC o isualiza ion, which we e sc a ch loaded in o li ing C2C12 mouse myoblas cells. The pep ides, ei he composed o a ginines (R) o lysines (K) wi h a leng h a ying om 5 o 12 amino acids, we e allowed o en e he cells and dis ibu e in he cy oplasm and nucleus. The obse a ion o he cells by con ocal mic oscopy s a ed a e a eco e y ime o 30 min a e sc a ching loading. Fo each pep- ide a leas 10 cells we e selec ed and imaged wi h espec o sim- ila in acellula luo escence in ensi y le els. The mic oscopic obse a ion o he cells shows o all pep ides a homogeneous dis ibu ion in he cy oplasm, whe e only esicula s uc u es we e ee o labeled pep ides. Simila ly, in he nucleoplasm he pep ides dis ibu ed wi hou en iching a any oci. All L-amino acid pep ides es ed he e, K5 o K12 and R5 o R12 accumula ed p edominan ly in he nucleus compa ed o he cy oplasm, as shown by he luo escence in ensi y in mouse C2C12 cells (Fig. 1A and B). The pep ides R6 o R12 clea ly showed highe luo escence in ensi y in a eas inside he cell nucleus, which we e iden i ied as nucleoli co ela ing wi h he da k dense s uc u es isible in he phase con as images (Fig. 1A). The pep ide R5 did no accumula e in nucleoli (Fig. 1A), which was also he case o all poly-K pep ides (Fig. 1B). This indica es ha cha ge alone is no he only de e minan bu a he he isoelec ic p ope ies, which di e be ween K and R polyme s. To quan i a i ely ana- lyze he in ensi y and dis ibu ion pa e ns, we de e mined o 10 cells he mean luo escence in ensi y in he cy oplasm, nucleus (excluding nucleoli) and in he nucleoli. The egions o luo es- cence in ensi y measu emen s we e chosen wi h e e ence o he phase con as images and o e layed wi h he luo escence image as shown in Figu e 1C. Da a we e no malized o he nucleoplas- mic le els o e lec he accumula ion po en ial o molecules in he nucleolus o e he nucleoplasmic le els (Fig. 1C). The nume ical alues a e gi en in supplemen a y Tables 2 and 3. The compa ison o he mean luo escence in ensi y shows, o all poly-K pep ides, an a e age cy oplasmic le el o 0.74 old less compa ed o he nucleoplasm. Fo poly-R pep ides he cy o- plasmic le el is on a e age 0.68 old less han in he nucleoplasm (Fig. 1A, B and supplemen a y Table 2). These da a al eady demons a e ha poly-R pep ides ha e he po en ial o s onge accumula ion in he nucleus han poly-K pep ides. Al hough all pep ides es ed a e small enough o passi ely di use in and ou o he nucleus and should dis ibu e e enly, all o hem accumula e in he nucleoplasm. S ikingly, only he pep ides R6 o R12 show a u he isible accumula ion inside nucleoli wi h an a e age o 1.16- old o e he nucleoplasmic le el (Fig. 1C). The s a is ical analysis wi h a non-pa ame ic K uskal-Wallis ANOVA es shows signi icance o he accumula ion o R6 o R12 in he nucleolus. The poly-K pep ides show only sligh ly highe le el in he nucleoli wi h al- ues be ween 1.02 and 1.06 old o e he nucleoplasm. An inc ease o he pep ide leng h om R6 o R12 and hus he posi- i e cha ge and isoelec ic p ope ies does no linea ly inc ease he nucleola accumula ion le el o he pep ides. The highes le el o nucleola accumula ion was ound o he R9 pep ide wi h 1.22 old o e he nucleoplasmic le el, while he mean accu- mula ion o R11 and R12 a he dec eases sligh ly o 1.12 and 1.14- old espec i ely (Fig. 1A, B and supplemen a y Table 2). This sugges s ha he e is a sa u a ion/pla eau le el a and abo e 9 a ginine esidues. F om yeas o human cells, main ea u es o he nucleolus s uc u e and unc ion a e conse ed as i o ms a ound clus- e s o RNA genes and is he si es o ibosome biogenesis. We we e now in e es ed i he nucleola localiza ion and accumu- la ion o p o eins and pep ides is simila ly conse ed. The e- o e, we es ed he nucleola accumula ion po en ial o he R10 pep ide in cul u ed cells o di e en species. We sc a ch www. and online.com 317Nucleus loaded he FITC labeled R10 in o cells om yeas (S. ce e isiae), insec s (D. mela- nogas e ), ish (D. e io), mouse (M.musculus)andhuman(H. sapiens) cells and analyzed he localiza ion o he pep ides by con ocal mic oscopy (Fig. 1D). In yeas cells he nucleus and nucleolus a e mo e di icul o iden i y and clea accumula ions in hecellsa eno as isibleas o hecellso highe euka - yo es. Regions wi h less pep ide inside he yeas cells could co - espond o he acuole. In he cells om insec , ish, mouse and human, on he o he hand, he nucleoli a e isible in he phase con as images and show clea ly an accumula ion o he R10 pep ide compa ed o he nucleus and cy oplasm o he cells (Fig. 1D). To cla i y he localiza ion o he FITC-R10 in yeas we pe o med a co-label- ing o yeas cells wi h he li e cell DNA dye DRAQ5. 32 In his expe imen he yeas nucleus is isible as in ense labeled ound compa men in he cells. In some cells we could iden i y a c escen shaped egion in he pe iphe y o he nucleus wi h highe FITC-R10 in ensi y compa ed o he nuclea in e io and he cy o- plasm. A co esponding luo es- cence in ensi y linescan analysis con i ms an accumula ion o he pep ide in a c escen shaped egion in he nuclea pe iphe y (Fig. 1E). Ea ly s udies on he composi ion o yeas nuclei ha e shown ha , depending on he o ien a ion a ound o c es- cen shaped pe iphe al nuclea compa men ha bo s he nucle- olus. 40-42 Allcell ypeso mul i- cellula o ganisms es ed he e show a simila en ichmen o he labeled poly-R pep ides in he nucleoli. Some yeas cells show a simila accumula ion in he nuclea pe iphe y. The accumula ion o pep ides like R10 inside nucleoli is, hus, e o- lu iona y conse ed a leas om insec s o humans. No ably, HeLa cells in me aphase ha ing no cell nucleus and wi h he nucleoli disassembled lack also he nucleola accumula ion o pep ides ound in in e phase cells. Thus, he accumula ion o Figu e 1. Dis ibu ion o poly-(R)and poly-(K)pep ides in li ing cells. In acellula dis ibu ion o poly-R pep i- des in (A) and poly-K pep ides (B) in li ing C2C12 mouse cells. In each panel, he fluo escence image is on op o he co esponding phase con as image. Nucleoli a e clea ly isible as da k ound s uc u es wi hin he nuclei in he phase con as images. The ba diag ams in (C) show he quan ifica ion o he poly-K and poly-R pep ide mean fluo escence in cy oplasm, nucleoplasm and nucleoli a e aged o 10 cells om 2 independen expe imen s. The nucleoplasmic alues we e used o no maliza ion. A eas o quan ifica ion we e defined as desc ibed in me hods and o e layed wi h fluo escence images as shown. In acellula dis ibu ion o R10 pep- ide in li ing cells o di e en species as indica ed is shown in (D). In (E) li ing yeas cells we e u he s ained wi h DRAQ5 o be e isualiza ion o he nucleus and a line in ensi y p ofile in a bi a y uni s (a.u.) o bo h DNA and pep ide is shown. The in acellula dis ibu ion o (D) and L-R10 pep ides in li ing C2C12 cells and he co esponding quan ifica ion o mean fluo escence in ensi ies is shown in (F). Scaleba s: 5 mm. 318 Volume 6 Issue 4Nucleus poly-R pep ides is di ec ly linked o he s uc u e and unc ion o he nucleolus (supplemen a y Fig. 1D). All amino acids inco po a ed in o p o eins by ansla ion in li ing o ganisms a e L-enan iome s and only pos - ansla ional enzyma ic eac ions in a ious o ganisms con e some o D- amino acids. 43,44 Besides ha , ee D-amino acids we e disco e ed only in b ain issue. 45 We a gue ha a poly D-a ginine pep ide may no exhibi he speci ic binding o he same cellula a ge componen s, e.g., p o eins o nucleic acids, as he L-enan iome . In con as , he pep ide cha ge and isoelec ic p ope ies a e no al e ed when composed o D-amino acids. This phenomenon is exploi ed in mi o image phage displays used in he sea ch o he apeu ic pep ides wi h no el a ge s and imp o ed p ope - ies. 46,47 The la e exploi s he s abili y o he D-amino acid pep- ides o p o eoly ic deg ada ion. Hence, we compa ed he dis ibu ion o FITC-R10 composed o D and L amino acids (Fig. 1F). Inside li ing cells bo h D- and L-R10 pep ides dis ib- u e homogenously in he cy oplasm and accumula ed in he nucleoli. The main di e ence in he dis ibu ion is ha D-R10 had a much lowe concen a ion in he nucleoplasm compa ed o he L-R10, he e o e he measu emen o ela i e accumula ion o D-R10 o e he nucleoplasmic le el appea s much s onge (Fig. 1F). Howe e , he absolu e de ec ed luo escence in ensi ies we e compa able be ween bo h ypes o pep ides using he same mic oscopic de ec ion se ings. This inding indica es ei he a binding o he L-pep ide o a nucleoplasmic componen , which is no he case o he D-pep ide, and/o he lack o deg ada ion o he D-R10. In e es ingly hough, he mean in ensi y o luo es- cence in he nucleoli is a simila le els o bo h ypes o pep ides. Nex , we es ed he ole o a ious pep ides in he a ge ing o p o eins inside he cell nucleus. We used pep ides used o he neu al ace p o ein GFP o de e mine localiza ion pa e ns. Fo his pu pose, we used GFP al eady used o a NLS sequence (SV40 T an igen de i ed; PKKKRKV 48 ) and added downs eam u he basic, acidic and neu al pep ides o di e en leng hs ol- lowed by he enhanced GFP coding sequence. The di e en GFP usions we e exp essed in C2C12 mouse myoblas cells (Fig. 2A) and in human HeLa cells (supplemen a y Fig. 1) and imaged wi h a con ocal mic oscope. We also es ed he dis ibu ion o an uncoupled GFP and he NLS-GFP alone as compa ison and as con ols o cy oplasmic and nucleoplasmic localiza ion wi h no expec ed nucleola localiza ion and accumula ion. The mic oscopic images o he GFP e sions in Figu e 2A show o he GFP wi hou NLS a homogeneous dis ibu ion h oughou he cell wi h simila le els in cy oplasm and nucleus wi h signi ican educ ion in he nucleoli (p D0.01 see supplemen a y Table 2). The sligh ly lowe mean in ensi y in he cy oplasm is he esul o memb anous compa men s de oid o GFP p o ein (Fig. 2A and C; supplemen a y Table 2). The homogeneous dis ibu ion be ween nucleus and cy oplasm is he esul o passi e di usion in o he nucleus. Molecules wi h adiame e o up o»9nma e capable o en e ing he cell nucleus by passi e di usion, which has been measu ed o luo escen molecules he size o GFP o occu wi hin a ew minu es. 19,49 The NLS coupled GFP, hough, shows mo e han 2- old accumula ion in he nucleus o e he cy oplasmic le el bu wi h a simila educ ion in he nucleoli, isible as ound ele a ed objec s in he DIC images. This dis ibu ion pa e n is simila o all NLS-GFP a - ian s wi h addi ional acidic (aspa a e) and neu al (glycine) amino acid se ies (Fig. 2A and C). The nucleola le el o he di e en p o eins in Figu e 2A is signi ican ly educed in he ange o 0.7 o 0.5- old compa ed o he nucleoplasmic le el as de e - mined by measu ing he mean luo escence in ensi y (Fig. 2C and supplemen a y Table 2). Figu e 2. In acellula dis ibu ion o pep ide agged GFP ace p o eins in li ing cells. (A) Rep esen a i e mic oscopic images o li e C2C12 cells ans ec ed wi h GFP, NLS-GFP and a usion o NLS-GFP wi h 8 glycines (8G) as example o he addi ion o neu al amino acids is shown in he fi s panel. The second panel shows images o cells wi h usions o an inc easing numbe o aspa a es as examples o acidic amino acid usions o NLS-GFP. The GFP fluo escence is depic ed in he uppe ow wi h he co esponding di e en ial in e e - ence con as (DIC) images below. (B) Mic oscopic Images o he dis ibu ion o he NLS-GFP used o 4 o 7 addi ional a ginines. The g aphs in (C) ep esen he mean dis ibu ion o he GFP cons uc s depic ed in (A) and (B) in cy oplasm and nucleoli in ela ion o he nucleoplasm o an a e age o 10 cells om 2 indepen- den expe imen s. Scaleba s: 5 mm. www. and online.com 319Nucleus Con as ingly, all NLS-GFP e sions coupled o poly-R pep i- des om R4 o R7 show a clea accumula ion in he nucleoli a ound 2 o 3- old o e he nucleoplasmic le el. The nucleola accumula ion is lowes o NLS-R4-GFP a 1.7- old and inc eases wi h he g owing numbe o a ginine esidues up o 2.7- old o NLS-R7-GFP (Fig. 2B and C and supplemen a y Table 2). In e es ingly, he compa ison o he da a o pep ides and p o eins shows ha in con as o he FITC-R5 pep ide, we see al eady o he R4 used o NLS-GFP a nucleola accumula- ion. This may e lec he p o eoly ic deg ada ion o he L-amino acids con aining pep ides e sus he mo e s able co esponding GFP usion p o eins. On he o he hand, he addi ional a ginine wi hin he NLS sequence could accoun o his appa en di e ence. No ewo hy, pep ides wi hou nucleola accumula ion like K5 and R5 show no educed concen a ion inside nucleoli. Al eady in an ea lie s udy es ing unc ional pep ides no accumula ing in nucleoli, we obse ed a simila homogenous nuclea and nucleola dis ibu ion o a luo opho e labeled pep ide. 50 These obse a ions exempli y he dense s uc u e o he nucleoli esul - ing in a size dependen sie ing e ec o molecules wi hou po en ial o accumula ion he ein. La ge molecules like GFP (wi h dimensions o a ound 2.4 £4.2 nm) 51 a e signi ican ly excluded compa ed o much smalle pep ides. In e es ingly, he R5 o R7 coupled NLS-GFP molecules show in addi ion an e en highe e iciency o nuclea impo wi h le els up o 10- old o e he cy oplasm compa ed o NLS-GFP alone (Fig. 2B, C and supplemen a y Table 2). This obse a ion shows ha he addi- ion o a ginine esidues o a minimal NLS inc eases he e i- ciency o nuclea impo la gely exceeding he passi e leakage e lux o small p o ein molecules h ough nuclea po es. Al e na- i ely, hei ac i e e en ion in he nucleus by binding, e.g., o he nucleolus, p e en s hei exi om he nucleus. The da a p esen ed he e so a demons a e ha pep ides and p o- eins wi h a s ong posi i e cha ge and high isoelec ic poin ha e he po en ial o accumula e in he nucleolus o li e cells. The h esh- old o s ong nucleola accumula- ion in pep ides is R6, while in p o eins bea ing an NLS (wi h one addi ional R) al eady 4 addi ional consecu i e a ginines a e su icien . In addi ion, he nuclea impo is enhanced by he addi ion o a leas 5 a ginine esidues in p oxim- i y o a minimal NLS. We con- clude ha he cha ge and isoelec ic p ope ies o he pep i- des and p o ein domains esponsi- ble o nucleola accumula ion is key in his p ocess because i s , he di e ence be ween nuclea impo and nucleola accumula ion has ne e been shown o ely on an ac i e mechanism. Second, e en pep ides syn hesized wi h D-amino acids show nucleola accumula ion, which may ail o speci ically bind o he same cellula a ge s as hei L-enan iome coun e pa s. The e idence o a cha ge and isoelec ic poin dependency o pep ide and p o ein accumula ion in nucleoli led us o pos ula e he p esence o an elec ochemical componen . Elec ochemical in e ac ions ha would accumula e posi i ely cha ged basic pep- ides should be media ed by highly abundan o s ong nega i ely cha ged acidic nucleola componen s. These ha e o be con ined o nucleoli as a e he accumula ed basic pep ides and p o eins. Such molecules could be nucleic acids, which a e highly abun- dan in he nucleoli in o m o RNA. The e o e, we wan ed o de e mine he pH dis ibu ion inside he nucleus in li ing cells. We used he li e cell pe meable luo escen dye SNARF-4F, which has 2 pH dependen a iome ic emission peaks a a ound 587 and 640 nm. The calcula ion o he a io o he luo escence in ensi y a he 2 emission peaks gi es in o ma ion on he pH o subcellula s uc u es. The a iome ic images o HeLa cells show in he luo escence channel iew as well as in he in ensi y a io iew di e en cellula s uc u es highligh ed (Fig. 3A op ow). The magni ied image de ail shows s uc u es a he nuclea pe iphe y and in he cen e o he nucleus highligh ed as mo e acidic (Fig. 3A bo om ow). Al hough he cy oplasm is gene ally conside ed o ha e a neu al pH, he mo e acidic egions in he cy oplasm do mos likely ep esen pa s o he Golgi ne wo k and o ganelles like pe oxisomes. 52 The cen al nuclea egion wi h mo e acidic pH is eminiscen o he nucleoli in HeLa cells shown in he phase con as images in Figu e 1D as well as he DIC images in Figu e 2 and in supplemen a y Figu e 1. Almos Figu e 3. In acellula pH landscape by a iome ic fluo escence mic oscopy owa d he gene a ion o an in acellula pH landscape. The figu e in (A) displays fluo escence mic oscopy images o he pH sensi i e dye SNARF-4F in li e HeLa cells (colo ). The g een channel shows he emission a 587 and he ed channel he emission o he dye a 640 nm du ing exci a ion wi h he same wa eleng h. The esul ing in ensi y a io image (g ay scale) shows a iome ic di e ences be ween he in ensi ies o he 2 fluo escence emission peaks o he dye due o pH a ia ions in subcellula s uc u es. The fluo escence colo scale and he a io scale in g ay indica e he ela i e ange om mo e acidic o mo e basic pH. Panel (B) illus a es a map o he in acellula pH landscape by measu ing he pH in a ious cellula compa men s and subs uc u es like nucleoplasm and nucleolus (modified om 53 ). Scaleba s: 5 mm. 320 Volume 6 Issue 4Nucleus e e y cell in he o e iew image has he nucleoli highligh ed as mo e acidic compa ed o he nucleoplasmic su ounding (Fig. 3A op ow). Al hough p ecise measu emen s o subcellula pH alues a e no possible wi h he SNARF-4F dye, i allows us o conclude ha he nucleola en i onmen is mo e acidic han he neu al pH o he nucleoplasm. The subcellula pH landscape was measu ed in a ious s udies be o e and is e iewed in. 53 Figu e 3B summa izes he e iewed da a and was adap ed o now inco po a e he di e en ia ed neu al pH o he nucleo- plasm and mo e acidic pH o he nucleolus ob ained om ou measu emen s. Nex , we wan ed o es whe he he pep ides and p o eins wi h he sequence equi emen s su icien o accumula e in nucleoli do bind RNA. To es ou hypo hesis, we pu i ied o al RNA (Fig. 4A) and pe o med slo blo analysis (Fig. 4B). We blo ed inc easing amoun s o o al RNA on a PVDF mem- b ane ollowed by incuba ion wi h ei he luo escen ly labeled L-R10, o D-R10, espec i ely (Fig. 4B). The luo escen ly agged D- and L-R10 we e only de ec ed on slo s spo ed wi h RNA (Fig. 4B, ows 4 and 6) bu no in con ol slo s wi hou RNA (Fig. 4B, ows 3 and 5). As a loading con ol, memb anes we e s ained in pa allel wi h me hylene blue (Fig. 4B, ows 1 and 2). To es he RNA binding o basic pep ides used o GFP, we pe o med an in i o RNA pulldown assay using NLS-R7- GFP immobilized o sepha ose beads ia he GFP binding p o- ein (GBP) 39 (Fig. 4C, scheme). This assay allowed us o obse e (Fig. 4C, con ocal image) and quan i y (Fig. 4C, ba plo ), he binding o RNA o NLS-R7- GFP. The accumula ion o pep i- des in nucleoli is a common ea- u e o a ginine- ich cell pene a ing pep ides (CPPs). 54 Consis en ly, we ound ha he TAT pep ide, one o he bes known CPPs de i ed om HIV- 1 TAT p o ein, 55 also binds o RNA (supplemen a y Fig. 2). De ining expe imen ally he equi emen s o nucleola pep ide and p o ein accumula ion has shown ha se e al a ginines in sequence a e equi ed. A leas 6 a ginine esidues in a pep ide and 4 a ginines in combina ion wi h an NLS p o ide su icien local- ized su ace cha ge o nucleola accumula ion. Impo an ly, cha ge alone canno explain he di e ence be ween poly-R and poly-K, which may addi ionally ely on he highe numbe o unc ional amino g oups in a ginines s. lysines. Now we asked, how ou indings ela e o NoLS sequences ound in he human p o eome? A p e ious comp ehensi e s udy o nucleola a ge ing sequences has sys ema ically e alua ed 46 nucleola a ge ing sequences cu a ed om he li e a u e. Based on his analysis an algo i hm was de eloped o p edic nucleola a ge ing based on pep ide amino acid composi ion. The algo i hm was hen un agains he In e na ional P o ein Index da abase (IPI 3.4) and p edic ed >10,000 pu a i e nucleola a ge ing sequences. We ha e used his lis o pe o m u he analysis o he amino acid composi ion o p edic ed nucleola a ge ing sequences. An o e - all s a is ics shows ha he p edic ed NoLS ha e an a e age leng h o 27 amino acids wi h a mos equen leng h o 20 amino acids. The as majo i y (83%) o p edic ed NoLS sequen- ces has a leng h o 30 amino acids o less. Fo hese NoLS he mean a io o he numbe o basic (lysine, a ginine and his idine) e sus he o al numbe o amino acids is 0.34. The sho es and mos equen ly p edic ed NoLS sequences o 20 amino acids he e o e con ain on a e age a leas 7 basic amino acids. The 15 Figu e 4. Analyses o RNA binding o poly-(R)pep ides and GFP usions. (A) To al RNA used o in i o RNA binding assays sepa a ed by size showing he absence o genomic DNA con amina ion, as well as cha ac e - is ic bands o he 28S and 18S RNA, which a e p ima ily syn hesized and localized in nucleoli. The slo blo in (B) ( ep esen a i e om 2 independen expe imen s), ows 1 and 2, show a me hylene blue s ained PDVF memb ane in he absence (1) and p esence (2) o di e en amoun s o o al RNA (0.5 mg and 1 mg) and we e used as a loading con ol. The es o he blo shows he binding o fluo escen ly agged L-R10 (3 and 4) and D-R10 (5 and 6) in he absence (3 and 5) and p esence (4 and 6) o o al RNA. Slo s lacking RNA ( ows 5 and 6) do no show binding o L- and D-R10, espec i ely. In con as , slo s p obed wi h inc easing amoun s o RNA show inc easing L- and D-R10 binding. (C)In i o RNA pulldown assay using NLS-R7-GFP immobilized o sepha ose beads ia he GFP binding p o ein (GBP) (scheme). RNA was s ained using p opidium iodide (PI) and GFP alone was used as a nega i e con ol. RNA binding was measu ed on a fluo escen pla e eade (ba plo ) and imaged by con ocal mic oscopy (mic oscopic images). Plo s ep esen he a e age plus s anda d de ia ion o 2 independen expe imen s. Scaleba : 100 mm. www. and online.com 321Nucleus