Affinity chromatography with an immobilized RNA enzyme

P oc. Na l. Acad. Sci. USA Vol. 83, pp. 5904-5908, Augus 1986 Biochemis y A ini y ch oma og aphy wi h an immobilized RNA enzyme (RNase P/ca aly ic RNA/ca aly ic subuni Ml RNA/C5 p o ein) AGUSTIN VIOQUE AND SIDNEY ALTMAN Depa men o Biology, Yale Uni e si y, New Ha en, CT 06520 Communica ed by Clemen L. Ma ke , May 5, 1986 ABSTRACT Ml RNA, he ca aly ic subuni o Esche ichia coli RNase P, has been co alen ly linked a i s 3' e minus o aga ose beads. Unlike Ml RNA, which is ac i e in solu ion in he absence o he p o ein componen (C5) o RNase P, he RNA linked o he beads is ac i e only in he p esence o C5 p o ein. A ini y ch oma og aphy o c ude ex ac s o E. coli on a column p epa ed om he beads o which he RNA has been c osslinked esul s in he pu i ica ion o C5 p o ein in a single s ep. The p o ein has been pu i ied in his manne om cells ha con ain a plasmid, pINIIIR20, which includes he gene ha codes o C5 p o ein. A 6- old ampli ica ion o he exp ession o C5 p o ein is ound in hese cells a e induc ion as compa ed o cells ha do no ha bo he plasmid. A ini y ch oma og aphy has been used o pu i y a a ie y o biologically impo an molecules such as enzymes, co ac o s, and ho mone ecep o s (1, 2). Immobilized nucleic acids ha e been used o pu i y o iden i y p o eins, bo h ca aly ic and nonca aly ic, wi h which hey in e ac . Fo example, ibo- somal p o eins ha in e ac wi h 5S and 5.8S RNA ha e been iden i ied by use o columns made wi h hese RNAs (3-5). We now epo he success ul p epa a ion o a column made wi h an immobilized ca aly ic RNA, which has been used success ully o pu i y he p o ein ha , oge he wi h he RNA, makes up Esche ichia coli RNase P in i o. RNase P is he enzyme esponsible o ma u a ion o he 5' e mini o RNA p ecu so molecules in i o (6). In all o ganisms in es iga ed hus a , i has been ound ha he enzyme is composed o wo kinds o subuni s, one o which is RNA and one o which is p o ein, and bo h o which a e essen ial o ac i i y in i o (7). The RNA componen (Ml RNA) o E. coli RNase P and o some o he bac e ia (8, 9) can, by i sel , ca y ou he ca aly ic eac ion unde ce ain condi ions in i o. While ample quan i ies o hese RNA molecules a e a ailable, s udies o he holoenzyme complex om E. coli ha e been hampe ed by he di icul ies in he pu i ica ion o su icien quan i ies o he pu e p o ein com- ponen (C5 p o ein) o biochemical expe imen a ion. The gene o his p o ein ( npA) in E. coli has been cloned (10) bu does no o e exp ess he gene p oduc o a signi ican deg ee. We ha e inc eased ou yield and acili a ed p epa a ion o C5 p o ein by using a ini y ch oma og aphy, wi h immobilized Ml RNA, o cell ex ac s p epa ed om E. coli cells ha ha bo a no el subclone o he cis on and con ain he npA gene adjacen o an inducible, s ong p omo e . In addi ion o CS p o ein (M 13,700), we ha e iden i ied a second p o ein (M 27,000) ha binds s ongly o Ml RNA. METHODS Bac e ial S ains and Plasmids. E. coli s ain HB101 (F-, hsdS20, ecA13, a a14, p oAZ, lacYI, galk2, psL20, xyl-5, m l-i, supE44, X-) was used as he hos cell in he cloning expe imen s. E. coli s ain MRE600 was used as a sou ce o RNase P. Cells we e g own in he ollowing media acco ding o he equi ed condi ions o each expe imen : LB medium, sup- plemen ed when necessa y wi h ampicillin (50 ug/ml); M9 medium supplemen ed wi h glucose (2 mg/ml), leucine (50 pug/ml), p oline (100 gg/ml), h eonine (60 ,.g/ml), hiamine (1 /ig/ml), and, when necessa y, ampicillin (25 ug/ml). Plasmids pINIIIompAl [gi o M. Inouye (11); a pBR322 de i a i e ha con ains he ipp p omo e and he inducible lac p omo e -ope a o egion ups eam om he ompA gene] and pFHC1008 [dona ed by F. Hansen (10)] we e used in he cons uc ion o he subclones. Plasmid pRR1, ca ying he Ml RNA gene, has been desc ibed (12). DNA Manipula ions. Plasmid isola ion and ou ine DNA manipula ions we e pe o med as desc ibed elsewhe e (13). Res ic ion enzymes, Klenow agmen o E. coli DNA polyme ase, and T4 DNA ligase we e pu chased om New England Biolabs. Cal in es inal phospha ase was pu chased om Boeh inge Mannheim. Pu i ica ion o Ml RNA and RNase P. Ml RNA was pu i ied om p epa a i e polyac ylamide gels o o al RNA om E. coli HB101 ha con ained he plasmid pRR1 as desc ibed ea lie (12). Pa ially pu i ied RNase P was ob ained om cellula ex ac s o s ain MRE600 a e ch oma og aphy on DEAE- Sephadex and p ecipi a ion wi h (NH4)2SO4 as desc ibed (14). P epa a ion o P o ein Labeled wi h 35S. Cells we e g own o e nigh in M9 medium a 370C, washed wice wi h M9 medium om which sul a e ions we e omi ed, and incuba ed o 1 h a 370C in he sul a e- ee medium. [35S]Me hionine (10 ,uCi/ml; 1 Ci = 37 GBq) was hen added, and incuba ion was con inued o 6 mo e h . Fo s udies o he induc ion o cells ha ca y he cons uc ed pINIIIR20 exp ession plas- mid, 2 mM isop opyl 3-D- hiogalac oside (iP SGal) was added a he same ime as he label. Immobiliza ion o Ml RNA. P io o immobiliza ion, an aliquo o Ml RNA was dephospho yla ed and labeled a he 5' end wi h [y-32P]ATP and polynucleo ide kinase (P omega Bio ec, Madison, WI) as desc ibed (15). The labeled RNA (0.8 mg) was hen oxidized wi h 10 mM sodium pe ioda e (16). To examine he e iciency o oxida ion, he abili y o he pe ioda e- ea ed Ml RNA o ac as a subs a e in an RNA ligase condensa ion wi h cy idine 3',5'-[5'-32P]bisphospha e was es ed (17). We concluded ha >95% o he Ml RNA was oxidized. The 5'-end-labeled and oxidized Ml RNA (0.8 mg; 4 x 106 cpm, Ce enko adia ion) was coupled o 0.5 ml o aga ose- adipic acid dihyd azide (P-L Biochemicals) as desc ibed (4), washed wi h 2 M NaCl un il no u he adioac i i y was eleased, and packed in a Pas eu pipe e. Abou 90% o he RNA was ac ually coupled o he gel as es ima ed om he adioac i i y eco e ed in he washes. Abb e ia ions: iP SGal, isop opyl /-D- hiogalac oside; bp, base pai (s). 5904 The publica ion cos s o his a icle we e de ayed in pa by page cha ge paymen . This a icle mus he e o e be he eby ma ked "ad e isemen " in acco dance wi h 18 U.S.C. §1734 solely o indica e his ac . P oc. Na l. Acad. Sci. USA 83 (1986) 5905 Subcloning o he npA Gene ha Codes o C5 P o ein. The exp ession plasmid pINIIIR20 was cons uc ed o inc ease he p oduc ion o C5 p o ein. The inducible exp ession ec o pINIIIompA1 (11) was cu a he unique BamHI si e, and he s agge ed ends we e illed in by polyme iza ion wi h he Klenow agmen o E. coli polyme ase I and dephospho yl- a ed wi h cal in es inal phospha ase. Plasmid pFHC1008 (10), which con ains pa o he pmH ope on and includes he comple e sequence o pmH ( he gene o ibosomal p o ein L34) and npA ( he gene o C5 p o ein), was cu a he unique BssHII and A a I si es, and a 776-base-pai (bp) agmen ha con ained he pmH and npA genes was pu i ied. This agmen was cu wi h Rsa I o gi e ou agmen s (538, 142, 67, and 33 bp). The 538-bp agmen con ained mos o he pmH gene and he comple e npA gene. The s agge ed ends we e illed wi h he Klenow agmen , and he mix u e o agmen s was liga ed o he linea ized ec o wi h T4 DNA ligase. T ans o man s we e selec ed by ampicillin esis ance, and hose ca ying he 538-bp agmen in he igh o ien a ion we e iden i ied by es ic ion analysis. In his cons uc he pmH gene is in phase wi h he coding sequence o he signal pep ide o he OmpA p o ein (11). When selec ion o ans o man s was ca ied ou in LB medium, all o he clones ha con ained he 538-bp agmen had he agmen inse ed in he w ong o ien a ion. Howe e , when selec ion was ca ied ou in M9 medium, he igh cons uc ion was ob ained. This esul sugges s ha he exp ession o he ca boxyl- e minal-modi ied L34 p o ein does no pe mi iabili y. Leaky induc ion o he lac gene p omo e by chemical con aminan s in LB medium (18) is a oided in minimal medium. No e also ha E. coli s ain HB101 ca ying pINIIIR20 canno g ow in he p esence o 2 mM iP SGal. Pu i ica ion o C5 P o ein. Ex ac ion o RNAs. The RNAs p esen in S30 ex ac s o in RNase P, pa ially pu i ied a leas h ough he DEAE-Sephadex s ep (14), we e ex ac ed wi h LiCl/EDTA by he me hod o Dijk and Li lechild (19) as used o p epa a ion o ac i e yeas ibosomal p o eins (20). (We also ha e used LiCl/EDTA o elu e basic p o eins om immobilized RNA as is desc ibed below.) One olume o ex ac was mixed wi h 2 olumes o 10.5 M LiCl, and EDTA (pH 8.0, adjus ed wi h NaOH) was added o a concen a ion o 10 mM. The mix was gen ly agi a ed o e nigh a 4°C and cen i uged a 100,000 x g o 10 h . The supe na an , con aining he p o eins, was dialyzed agains bu e A (50 mM T is HCl, pH 7.4/10 mM MgCl2/100 mM NH4Cl/10 mM 2-me cap oe hanol). A ini y ch oma og aphy on immobilized MI RNA. The dialyzed supe na an om he p e ious s ep was loaded on he a ini y column, and he sample was eci cula ed h ough he column o e nigh by a pe is al ic pump. The column was hen washed and elu ed wi h bu e A ha con ained di e en concen a ions o NH4Cl as indica ed in he legends o Figs. 2 and 3. C5 p o ein was elu ed in he inal s ep wi h 7 M LiCl/10 mM EDTA, pH 8.0. Elec opho esis o P o eins. P o eins we e analyzed on 0.1% NaDodSO4/4 M u ea/12.5% ac ylamide gels by he me hod o Laemmli (21). To al cellula p o eins we e ana- lyzed om a cell pelle ( om 0.5 ml o cul u e) ha was boiled o 3 min in loading bu e and hen loaded di ec ly on he gel. O he p o ein ac ions we e p ecipi a ed wi h ichlo oace ic acid in he p esence o 15 ,ug o bo ine se um albumin, when necessa y, be o e boiling in loading bu e and loading. P o ein concen a ions we e de e mined by he me hod o Low y e al. (22) wi h bo ine se um albumin as a s anda d. Assays o RNase P Ac i i y. Recons i u ion assays we e pe o med by he di ec mixing me hod (8) in 50 mM T is HCl, pH 7.4/10 mM MgCl2/100 mM NH4Cl. PEG (5% w / ol) was added whe e indica ed. Ml RNA ac i i y was assayed in 50 mM T is HCl, pH 7.4/100 mM MgCl2/100 mM NH4Cl/5% PEG. The p ecu so o E. coli RNATY was used as subs a e (23) in all o he expe imen s epo ed he e. All incuba ions we e ca ied ou a 37°C o he imes indica ed. The eac ion p oduc s we e analyzed on 10% (w / ol) poly- ac ylamide gels as desc ibed elsewhe e (8), and he amoun o p oduc o med was es ima ed by scanning densi ome y o au o adiog aphs. When immobilized Ml RNA was assayed, i s concen a ion was es ima ed om he adioac i i y o i s 5'-end label, de e mined by scin illa ion coun ing. RESULTS P ope ies o he Ml RNA Column. The 3' e minus o Ml RNA can be al e ed in a ious ways wi h li le o no e ec on he ca aly ic ac i i y o he RNA (9, 24). We e i ied (da a no shown) ha Ml RNA om E. coli, oxidized wi h pe ioda e o c ea e 2' and 3' ke o g oups, has simila ac i i y o ha o in ac Ml RNA, jus as Ma sh and Pace (17) ound o he RNA componen o Bacillus sub ilis RNase P. The oxidized Ml RNA was coupled h ough a dihyd azide adipic acid space o aga ose beads as desc ibed, and a slu y o he esul ing ma e ial was es ed o RNase P ac i i y. We chose he pa icula beads and space o educe p oblems o s e ic hind ance. When he Ml RNA-bead complex was assayed o ac i i y (Fig. 1), he ca aly ic ac i i y was almos ully eco e ed in he econs i u ion assay wi h C5 p o ein (lanes 5 and 6) bu was almos comple ely los when es ed unde condi ions whe e he ee RNA exhibi ed ac i i y by i sel (lanes 2 and 3). No ac i i y was obse ed when C5 p o ein o Ml RNA was assayed alone in bu e con aining 10 mM MgCl2. Fu he mo e, es ima es o ela i e a es o clea age we e made om mo e ex ensi e kine ic s udies (da a no shown). When ee and immobilized Ml RNA we e assayed in he p esence o C5 p o ein, hey had abou he same ac i i y. Bu abou 5% o he expec ed ac i i y o ee Ml RNA was eco e ed when he slu y o immobilized RNA was assayed in he absence o C5 p o ein. Howe e , when he slu y was analyzed by gel elec opho esis, abou 5% o he RNA an as ee Ml RNA in he gel, while he es did no en e he gel. F om his esul , we in e ha a small amoun o Ml RNA was leached o he beads and could accoun o he ca aly ic ac i i y de ec ed in he absence o p o ein. Howe e , we canno ule ou he possibili y ha his small amoun o 1 2 3 5 -* 4 5 6 _ -p RNA- IeRNAS FIG. 1. Compa ison o he ac i i y o immobilized Ml RNA wi h ha o ee oxidized Ml RNA. Assays we e cam ed ou as desc ibed in he p esence o 5% PEG. p RNA, RNA, and 5' indica e he posi ions o he subs a e RNA p ecu so molecule and he wo clea age p oduc s, espec i ely. Lanes: 1 and 4, no addi ion; 2, 10 ng o oxidized Ml RNA assayed o 15 min in bu e ha con ained 100 mM MgCl2; 3, 10 ng o immobilized Ml RNA assayed as in lane 2; 5, 10 ng o oxidized Ml RNA assayed o 5 min in bu e ha con ained 10 mM MgCl2 and an excess o pu i ied C5 p o ein; 6, 10 ng o immobilized Ml RNA assayed as in lane 5. Biochemis y: Vioque and Al man 5906 Biochemis y: Vioque and Al man ac i i y was, in ac , due o he immobilized RNA. Analysis o he slu y a e he econs i u ion assay indica ed ha he p esence o he C5 p o ein did no induce he elease o RNA om he beads (da a no shown). The e o e, he ca aly ic ac i i y de ec ed in he assay in he p esence o C5 p o ein was due, in ac , o he coupled RNA-bead complex. We conclude ha immobiliza ion o he 3' e minus o Ml RNA in e e es wi h a c i ical s ep in he ac i a ion o he enzyme, in he absence o he p o ein co ac o . Howe e , he RNA coupled o he beads appea s o e ain he desi ed speci ici y o he binding o C5 p o ein so ha he column can be used o he success ul pu i ica ion o C5 p o ein. To analyze he condi ions ha allow he pu i ica ion o C5 p o ein on he a ini y column, we used a p o ein mix u e p epa ed om RNase P ha had been pa ially pu i ied h ough he DEAE-Sephadex s ep and (NH4)2SO4 p ecipi a- ion as desc ibed by S a k e al. (14). The RNAs p esen in his p epa a ion we e ex ac ed by ea men wi h 7 M LiCl/10 mM EDTA. The esul ing p o ein mix u e was essen ially ee o RNA (<1%, as de e mined by do ing a sample on aga ose gel ha con ains e hidium b omide and isualizing he ma e ial wi h a UV ansillumina o ). No RNase P ac i i y was mani es ed by his ma e ial unless exogenous Ml RNA was added. The C5 p o ein in he p epa a ion did no bind o DEAE-Sephadex (da a no shown) in con as o he holoenzyme, which does bind as a consequence o i s RNA con en . These esul s con i med ha he LiCI/EDTA ex ac was ee o Ml RNA, and he C5 p o ein con ained in i was s ill ac i e in he RNase P econs i u ion assay. Al hough p o ein ex ac s o B. sub ilis RNase P ha ga e RNase P ac i i y in he econs i u ion assay ha e been made using ace ic acid o emo e RNA (25), his me hod has no yielded p epa a ions o C5 p o ein om E. coli RNase P ha a e ac i e in econs i u ion assays. A dialyzed ex ac o RNase P om which he RNA had been emo ed by ea men wi h LiCl/EDTA was loaded on he a ini y column, and he p o eins ha we e elu ed om he immobilized RNA a e shown in Fig. 2. The posi ion o C5 p o ein in he PAGE analysis o he elua es is shown by he a ow adjacen o lane 16. (No e ha C5 p o ein has an anomolous elec opho e ic mobili y. I mig a es mo e slowly han expec ed om i s molecula weigh .) Op imized condi- ions o he binding o his p o ein o he column include ecycling o he load sample and he use o PEG in he loading bu e . PEG inc eases binding e iciency by enhancing he e ec i e local concen a ion o he p o ein. The p o ein composi ion o he load sample and he low- h ough ac ion a e shown in Fig. 2, lanes 2 and 3. The e is some Msmea ing o hese samples because o he p esence o PEG in he bu e . P o eins in subsequen elua es om he column a e shown in lanes 5-15. Many p o eins we e bound o he column, bu mos we e elu ed wi h bu e ha con ained 0.37 M NH4Cl (lanes 5-8). A p o ein o appa en M 27,000 bound s ongly o he column bu could be sepa a ed om CS p o ein by elu ion wi h bu e ha con ained 0.7 M NH4Cl (lanes 11-13). A e his elu ion, only C5 p o ein emained bound o he column, bu i could be elu ed along wi h almos all ma e ial ha is ac i e in he RNase P econs i u ion assay by 7 M LiCl/10 mM EDTA (lanes 14 and 15). This inal elu ion s ep has been adap ed success ully om he me hod used o sepa a e ibosomal p o eins om RNA (19). Each ac ion was assayed o RNase P ac i i y in he absence and p esence o Ml RNA in bu e ha con ained 10 mM MgCl2. The only column ac ion exhibi ing signi ican ac i i y in he p esence o added Ml RNA was he one co esponding o ha seen in lane 14 o Fig. 2 (LiCI/EDTA wash). The p o ein species elu ed om he column ha has been labeled as "C5 p o ein" comig a es wi h ac i e C5 p o ein pu i ied by a di e en (and less e icien ) echnique (8). This 1 2 3 4 5 6 7 8 9 10 11 12 13 14 ,5 16 i. I i 'a .. ., Al_ Fi2a ',................. 4. 4 9 FIG. 2. Elec opho e ic analysis o he ac ions elu ed om he a ini y column. P o ein ex ac ed wi h LiC1/EDTA om pa ially pu i ied RNase P (see ex ) was placed in bu e A ha con ained 5% PEG, and 12 ml (6 mg) was loaded on a 0.5-ml column con aining 700 ,4g o bound Ml RNA. The sample (lane 2) was eci cula ed on he column o 15 h a a low a e o 6 ml/h . The low- h ough ac ion was collec ed (lane 3), and he column was washed in succession wi h 15 ml o bu e A ha con ained 5% PEG (no shown), 2 ml o bu e A (lane 4), h ee imes wi h 0.6 ml o 0.37 M NH4Cl in bu e A (lanes 5-7), h ee imes wi h 0.6 ml o 0.5 M NH4Cl in bu e A (lanes 8-10), h ee imes wi h 0.6 ml o 0.7 M NH4Cl in bu e A (lanes 11-13), and wice wi h 0.4 ml o 7 M LiCl/10 mM EDTA (lanes 14-15). Lane 1 con ains myoglobin (M 17,200) and a smalle con aminan . Lane 16 con ains he ollowing molecula weigh ma ke s: phospho ylase B (M 92,500), bo ine se um albumin (M 66,200), o albumin (M 45,000), ca bonic anhyd ase (M 31,000), soybean ypsin inhibi o (M 21,500), and lysozyme (M 14,400). The uppe mos dense band in lanes 2-15 is he bo ine se um albumin used as ca ie in he p epa a ion o he sample. Lanes 2 and 3 con ained one- i h as much sample as lanes 4-15. The a ow nex o lane 16 indica es he posi ion o C5 p o ein. The gel was s ained wi h Coomassie b illian blue. species, elu ed in he inal s ep (Fig. 2, lane 14), was ac i e in he econs i u ion assay o RNase P ac i i y, and so we concluded ha i is, indeed, C5 p o ein. Howe e , a species wi h simila mobili y in he gels, was elu ed a many o he lowe concen a ions o NH4Cl bu did no mani es ac i i y in he econs i u ion assay wi h Ml RNA. The e we e many basic p o eins in addi ion o CS p o ein in hese elua es ha could compe e o binding si es on Ml RNA. I is also possible ha he C5 p o ein in hese ac ions was i e e sibly dena u ed. We ha e no eco e ed ac i e CS p o ein om he p o ein elu ed a a lowe concen a ion o NH4Cl, by dialysis agains bu e A, o by ea men o hese ac ions wi h LiCl/EDTA p io o dialysis. When ac ions ha we e elu ed a 0.37 M NH4Cl we e dialyzed and ecycled h ough he column, some o he M 27,000 p o ein and he appa en C5 p o ein we e elu ed a highe sal concen a ions, hough mos o he ma e ial s ill was elu ed a 0.37 M. Vi ually no ac i e CS p o ein was eco e ed. These da a indica e ha he appa en C5 p o ein ha was elu ed a he lowe sal con- cen a ions may indeed be dena u ed o may be inhibi ed in he econs i u ion issay by he concen a ions o o he basic p o eins, including he M 27,000 p o ein, which a e ela i ely high compa ed o hose ound in c ude ex ac s. Twen y g ams o E. coli yielded abou 5.4 mg o LiCl ex ac and, inally, abou 5 ,g o abou 15% o he o al C5 ac i e p o ein in he o iginal 7 M LiCI/10 mM EDTA ex ac . P oc. Na l. Acad Sci. USA 83 (1986) P oc. Na l. Acad. Sci. USA 83 (1986) 5907 The eco e y o ac i e C5 p o ein wi h his me hod was a leas h ee imes be e han ha ob ained by p epa a i e elec opho esis o he same amoun o s a ing ma e ial (14), and he pu i ica ion was achie ed in less han hal he ime needed o he la e me hod. I is di icul o es ima e accu a ely he ex en o eco e y o C5 p o ein because econs i u ed RNase P ac i i y depended on he concen a- ion o added C5 p o ein in a nonlinea ashion (unpublished da a), and pu e C5 p o ein was inac i a ed on dilu ion. The la e p oblem could be o e come, in pa , by inclusion o PEG in he dilu ion bu e . Thus, ou es ima es o o al eco e ed ac i i y may be accu a e only wi hin a ac o o 2. I one assumes ha a leas one C5 p o ein molecule can bind o one Ml RNA molecule, he capaci y o he Ml RNA column is 0.11 ,ug o C5 p o ein pe Ag o Ml RNA a ailable o binding. In he case we desc ibe, he heo e ical capaci y is 77 Ag o C5 p o ein. The only limi a ion in scaling up his p ocedu e is he a ailabili y o Ml RNA, and ha is no longe a se ious obs acle. Ampli ica ion o C5 P o ein in Vi o. To enhance he yield o C5 p o ein om c ude ex ac s o cells, we subcloned, nex o a s ong, inducible p omo e in pINIIlompAl, he gene ha codes o C5 p o ein ( npA) and he adjacen gene ha codes o he ibosomal p o ein L34 ( pmH) om pFHC1008. C ude ex ac s we e made om cells ha ha bo ed he new ec o , pINIIIR20, and om he pa en cells wi h no ec o . These ex ac s we e bo h subjec ed o elec opho esis o analysis o he p o ein con en and also we e passed o e he Ml RNA column. The p oduc ion o p o eins in cells wi h pINIIIR20, a e induc ion wi h iP SGal, is shown in Fig. 3, lane 2. Lane 1 shows p o eins om an ex ac o uninduced cells. No e he massi e inc ease in he p oduc ion o a p o ein ha co e- sponds in mobili y o L34, he gene ha is adjacen o and ups eam om npA, he gene ha codes o C5 p o ein. The uppe a ow in lane 2 poin s o a ain mino band, unseen in 1 23 A-I 0- 4 5 6 7 8 91011121314 45 31 - 21.5 -1 i.: 14.4 .- $ FIG. 3. Au o adiog am o he elec opho e ic analysis o p o eins o [5S]me hionine-labeled E. coli s ain HB101 ha ca ies pINIIIR20. Lanes: 1, uninduced cells; 2, cells induced wi h 2 mM iP SGal; 3, S30 ex ac p epa ed om induced cells; 4-14, elu ion pa e ns om he immobilized MI RNA column o he S30 ex ac shown in lane 3 a e ex ac ion o RNAs wi h 7 M LiCl/10 mM EDTA (elu ion was done as in Fig. 2, bu he 0.7 M NH4Cl washes we e omi ed); 4, low- h ough ac ion; 5, bu e A con aining 5% PEG wash; 6, bu e A wash; 7-9, 0.37 M NH4Cl washes; 10-12, 0.5 M NH4Cl washes; 13-14,7 M LiCl/10 mM EDTA washes. Gels we e d ied and exposed o 24 h (lanes 1-2), 1 wk (lane 3), o 4 wk (lanes 4-14). The a ows nex o lane 2 indica e he posi ions o he L34 usion p o ein (lowe a ow) and C5 p o ein (uppe a ow). The ha ch ma ks nex o lanes 2, 3, and 4 show he posi ions o he molecula weigh (X 10-3) ma ke s be ween lanes 3 and 4. The a ow nex o lane 14 indica es he posi ion o he C5 p o ein. No e ha ibosomal p o ein L34 is no p esen in he p epa a ion o he S30 ex ac (lane 3) om he o al cell p o ein (lane 2). he uninduced cells, which has he mobili y o C5 p o ein. This analysis did no allow us o es ima e he amoun o ampli ica ion o p oduc ion o C5 p o ein. The e o e, we ea ed he c ude ex ac o pINIIIR20 wi h 7 M LiCl/10 mM EDTA o emo e RNA and used he esul ing p o ein in econs i u ion assays o RNase P ac i i y. The esul an ac i i y o RNase P om he c ude ex ac s o E. coli s ain HB101 ha bo ing plasmid pINIIIR20 induced wi h iP SGal is abou 6- old highe han he ac i i y in he ex ac om uninduced cells. We p esumed, he e o e, ha he p oduc- ion o C5 p o ein in he induced cells is a leas 6- old g ea e han in uninduced cells, and his p esump ion was e i ied by expe imen s wi h adioac i ely labeled ex ac s. When an S30 cell ex ac , om which RNA had been emo ed wi h 7 M LiCl/10 mM EDTA (Fig. 3, lane 3), o induced cells ha bo ing pINIIIR20 was passed o e he immobilized Ml RNA column, >80% o he p o ein ound in he column ac ion elu ed wi h 7 M LiCl/10 mM EDTA (Fig. 3, lane 13) was C5 p o ein, as judged by au o adiog aphy o ma e ial labeled wi h 35S. The yield o C5 p o ein in his ac ion is 4- o 6- old g ea e han ha ound in ex ac s o uninduced cells as de e mined by compa isons o he eco - e y o 35S-labeled C5 p o ein. The M 27,000 p o ein was no appa en in he ma e ial elu ed wi h 7 M LiCl om he column ha was de i ed om he induced cells. Ou esul s demon- s a e ha i is possible o pu i y a e y mino p o ein om E. coli (abou 250 copies pe cell) in one s ep on he immobilized Ml RNA column. DISCUSSION A ini y ch oma og aphy o p o eins on immobilized RNAs has been used as an analy ical ool o de e mine which p o eins in e ac in a speci ic manne wi h di e en RNAs (3-5). We ha e used his me hod, no only o pu i y p o eins ha bind s ongly o Ml RNA, a ca aly ic mac omolecule, bu also o p epa e hese p o eins in amoun s use ul o biochemical expe imen a ion. In ac , we ha e been able o pu i y C5, he p o ein componen o RNase P, in a single ch oma og aphic s ep om a c ude ex ac o E. coli. We also ha e ound ano he p o ein o M 27,000 ha binds s ongly o Ml RNA. In addi ion, he me hod we ha e de eloped allows p o ein pu i ica ion o be achie ed wi hou he use o dena u ing agen s such as NaDodSO4, u ea, o ace ic acid o o lyophiliza ion. Al hough he high concen a ions o sal s we used o sepa a e RNA om p o ein in he ini ial c ude ex ac and in he column elu ion bu e s in gene al will a ec he e ia y s uc u e o p o eins, he eco e ed p o ein is in a "na i e" s a e (see e . 19). The LiCl/EDTA me hod has been used o ex ac yeas ibosomal p o eins ha a e subsequen ly able o econs i u e unc ional 60S yeas ibo- somal subuni s (20). The C5 p o ein we pu i ied is ac i e in he econs i u ion assay o RNase P ac i i y and, hus, is highly sui able o u he s udies o he mechanism o ac ion o RNase P. A p esen we do no know i he M 27,000 p o ein is ela ed o he unc ion o Ml RNA o he p ocessing o he p ecu so o Ml RNA in i o. Al hough he size o he M 27,000 p o ein may be compa ible wi h ha o a dime o C5 p o ein, i is no likely ha such a dime su i es boiling in NaDodSO4 o elec opho esis h ough NaDodSO4 and u ea, bo h p ocedu es being pa o ou p o ocols. In addi ion, he M 27,000 p o ein is no seen in he inal column ac ions o 35S-labeled ex ac s o HB101 cells ha do o do no ha bo pINIIIR20, al hough la ge amoun s o C5 p o ein a e ob- se ed. This obse a ion indica es ha he M 27,000 p o ein lacks me hionine esidues and, he e o e, is un ela ed o C5 p o ein. The elu ion om he a ini y column o C5 p o ein o e a wide ange o sal concen a ions is no unexpec ed because Biochemis y: Vioque and Al man 5908 Biochemis y: Vioque and Al man we know ha Ml RNA can exis in a a ie y o con o ma- ions, some o which may p o ide highe speci ici y o binding o C5 p o ein (and M 27,000 p o ein) han o he s. Howe e , he amoun o ac i e C5 p o ein eco e ed in hese ac ions is no well co ela ed wi h he appa en amoun o p o ein, mos o he ac i i y being eco e ed in he inal wash wi h 7 M LiCl/10 mM EDTA. A his poin we canno say o ce ain whe he he species ha comig a es wi h C5 p o ein is inac i e o modi ied C5 p o ein, which binds less s ongly o Ml RNA han does na i e C5 p o ein, no do we know whe he i s ac i i y is simply inhibi ed by he p esence o basic p o eins ha copu i y wi h i . The amoun o C5 p o ein in cellula ex ac s has been inc eased by subcloning a cis on ha con ains he gene coding o his p o ein ( npA) om he plasmid pFHC1008 o a si e adjacen o an inducible, s ong p omo e in he plasmid pINIIIompAl. No mally, npA is co ansc ibed wi h he gene o ibosomal p o ein L34 ( pmH) and wo o he p o eins (10), bu no signi ican o e p oduc ion o he npA gene p oduc is obse ed in cells ha bo ing pFHC1008, al hough L34 is ampli ied (10). When hese wo genes a e placed adjacen o he lpp and lac p omo e s in pINIIIR20, an =6- old ampli ica ion o C5 p o ein is obse ed, as judged by he amoun o C5 eco e ed a e a ini y ch oma og aphy. Thus, a combina ion o he no el column echnique and gene ic enginee ing has led o a signi ican imp o emen in he yield and e iciency o pu i ica ion o C5 p o ein. In cons uc ing he immobilized RNA column, we ook ad an age o he ac ha he 3' e minus o Ml RNA can be al e ed in a ious ways wi hou signi ican ly diminishing he unc ion o he molecule (24). Fu he mo e, he leng h o he b idge o he aga ose beads was chosen o minimize s e ic p oblems and, hus, we an icipa ed ha he immobilized RNA would s ill e ain i s ca aly ic unc ion. Howe e , ca aly ic ac i i y a he same a e as shown by he equi alen amoun o ee RNA is de ec ed only in he p esence o he p o ein co ac o . This esul p omp s he specula ion ha , o he RNA o unc ion by i sel , i mus ha e some deg ee o con o ma ional lexibili y a i s 3' e minus ha is absen when i is co alen ly linked o he beads. Con o ma ional change, such as dime iza ion (26), as a p elude o ca aly ic unc ion may be slowed o p e en ed when Ml RNA is linked o he beads. Indeed, some immobilized p o ein enzymes ha e been epo ed o unde go con o ma ional changes 10_-105 imes mo e slowly han do he ee p o eins (27). Fo he immobilized Ml RNA o become ac i e, h ough wha - e e con o ma ional change may be needed, C5 p o ein mus be p esen as an essen ial co ac o . The equi emen o a p o ein co ac o in his eac ion bea s on specula ion ega d- ing ea ly biochemical sys ems. I he i s enzymes we e composed o RNA and had p ope ies simila o hose o Ml RNA, immobiliza ion on he in e io o p imi i e cell mem- b anes o in simple o ganelles would ha e c ea ed immedia e selec i e p essu e o he pa icipa ion o p o eins in enzy- ma ic eac ions. The p epa a ion o enzyma ically ac i e immobilized RNA may also be o p ac ical signi icance since ca aly ically ac i e immobilized p o eins ha e ound wide- sp ead applica ions (28). Finally, we no e ha ac i e RNase P can be econs i u ed om Ml RNA and he app op ia e p o ein componen s om HeLa cells (29) and B. sub ilis (8). These obse a ions aise he possibili y ha he a ini y-ch oma og aphy p ocedu e we desc ibe he e may be use ul in pu i ying he p o ein componen s o RNase P om a a ie y o o ganisms. The p ocedu e also may be adap ed o he pu i ica ion o p o ein co ac o s ha s imula e o he eac ions known o be ca alyzed by RNA (30) o in which ibonucleop o eins a e in ima ely in ol ed (31). We hank Donna Wesolowski o pu i ying Ml RNA and ou colleagues o aluable ad ice and gi s o ma e ials. A.V. is he ecipien o a long- e m ellowship om he Eu opean Molecula Biology O ganiza ion. This wo k was suppo ed by g an s om he Uni ed S a es Public Heal h Se ice and he Na ional Science Founda ion o S.A. 1. Le man, L. S. (1953) P oc. Na i. Acad. Sci. USA 39, 232-236. 2. Wilchek, M., Mi on, T. & Kohn, J. (1984) Me hods Enzymol. 104, 3-55. 3. Bu ell, H. R. & Ho owi z, J. (1977) Eu . J. Biochem. 75, 533-544. 4. Ulb ich, N. & Wool, I. G. (1978) J. Biol. Chem. 253, 9049-9052. 5. Ulb ich, N., Lin, A. & Wool, I. G. (1979) J. Biol. Chem. 254, 8641-8645. 6. Al man, S., Gue ie -Takada, C., F ank o , H. & Robe son, H. D. (1982) in Nucleases, eds. Linn, S. & Robe s, R. (Cold Sp ing Ha bo Labo a o y, Cold Sp ing Ha bo , NY), pp. 243-274. 7. Kole, R., Bae , M., S a k, B. C. & Al man, S. (1980) Cell 19, 881-887. 8. Gue ie -Takada, C., Ga dine , K., Ma sh, T., Pace, N. & Al man, S. (1983) Cell 35, 849-857. 9. Gue ie -Takada, C. & Al man, S. (1984) Science 223, 285-286. 10. Hansen, F. G., Hansen, E. B. & A lung, T. (1985) Gene 38, 85-93. 11. Gh ayeb, J., Kimu a, H., Takaha a, M., Hsiung, H., Masui, Y. & Inouye, M. (1984) EMBO J. 3, 2437-2442. 12. Reed, R. E., Bae , M. F., Gue ie -Takada, C., Donis-Kelle , H. & Al man, S. (1982) Cell 30, 627-636. 13. Mania is, T., F i sch, E. F. & Samb ook, J. (1982) Molecula Cloning: A Labo a o y Manual (Cold Sp ing Ha bo Labo a- o y, Cold Sp ing Ha bo , NY). 14. S a k, B. C., Kole, R., Bowman, E. J. & Al man, S. (1978) P oc. Na l. Acad. Sci. USA 75, 3717-3721. 15. Gue ie -Takada, C. & Al man, S. (1984) Biochemis y 23, 6327-6334. 16. Fahnes ock, S. R. & Nomu a, M. (1972) P oc. Na l. Acad. Sci. USA 69, 363-365. 17. Ma sh, T. L. & Pace, N. R. (1985) Science 229, 79-81. 18. Takaha a, M., Hible , D. W., Ba , P. J., Ge l , J. A. & Inouye, M. (1985) J. Biol. Chem. 260, 2670-2674. 19. Dijk, J. & Li lechild, J. (1979) Me hods Enzymol. 59, 481-502. 20. Vioque, A. & Palacian, E. (1985) Mol. Cell. Biochem. 66, 55-60. 21. Laemmli, U. K. (1970) Na u e (London) 227, 680-685. 22. Low y, 0. H., Roseb ough, N. J., Fa , A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, 265-275. 23. Robe son, H. D., Al man, S. & Smi h, J. D. (1972) J. Biol. Chem. 247, 5243-5251. 24. Gue ie -Takada, C. & Al man, S. (1986) Cell 45, 177-183. 25. Ga dine , K. J., Ma sh, T. L. & Pace, N. (1985) J. Biol. Chem. 260, 5415-5419. 26. Gue ie -Takada, C., Haydock, K., Allen, L. & Al man, S. (1986) Biochemis y 25, 1509-1515. 27. Da ydo , R. M., Ves e man, B. G., Genkin, M. V., Blumen eld, L. A. & K ylo , 0. V. (1981) Dokl. Akad. Nauk. SSSR 256, 491-494. 28. Mosbach, K., ed. (1976) Me hods Enzymol. 44, 491-936. 29. Gold, H. A. & Al man, S. (1986) Cell 44, 243-249. 30. K uge , K., G abowski, P. J., Zaug, A. J., Sands, J., Go sch- ling, D. E. & Cech, T. R. (1982) Cell 31, 147-157. 31. Cech, T. R. & Bass, B. L. (1986) Annu. Re . Biochem. 55, 599-629. P oc. Na l. Acad. Sci. USA 83 (1986)