Reprinted from the PROUEEDISCS OF NATIONAL .\CADEN>- OF SCIRUCES Vol. 54, No. 2. pp. 579-587. .\ugrlut, 1965. SI'EC'IFIC TEMPL.4 TI3 REQUIREMENTS OF RNA REI'LIC.4 SE8* By I. HARUNA AND S. SPIEGELMAN DEPARI`ZlEh r OF MICROBIOLOGY, VAIVERSITY OF ILLINOIS, TTHH 4V 4 Communicated June 21, 1W5 We have previously' reported the isolation of an RNA-dependent RNA polyni- erase (termed a "repiicase" for brevity) from E. coli infected with the RKA hac- teriophage ;\IS-2. The purified enzyme showed a mandatory requirement for added RNA and, furthermore, exhibited a unique preference for its homologous RNA. Ribosomal and sRNA of the host could not substitute RS a template arid neither of these cellular RNA types showed any ability to interfere with the tem- plate function of the viral RNA. We pointed out' that the ability of the replicase to discriminatc solved a crucial problem for an RXA virus attempting to direct its own duplication in an environ- ment replete with other RNA molecules. By produring a polymerase which ignores the mass of pre-existent cellular RNA, a guarantee is provided that replication is focused on the single strand of incoming viral RNA, the ultimate origin of progeny. It seems worth noting that sequence recognition by the enzyme can be of value not only to the virus but also to the investigator. The search for viral RNA rep. licases must perforce be carried out in the midst of a variety of highly active cellular polymerases capable of synthesizing polyribonucleotides. If the enzyme finally isolated possesses t,he appropriate template requirement, a coniforting assurance is furnished that the effort expended, and t)he information obtained, are indeed rele- vant to an understanding of viral replication. Operationally, t,his view deniands that viral RNA be used at all fract'ionation stages in assaying for polynierase ac- tivity. Our line of reasoning Ied to the expectat#iori that RNA replicases induced by other RNA viruses would show a similar preference for t8heir homologous templates. An opport,uriit'y to test' the validity of this predict>ion came with t,he isolation of a new and unrelated3! RNA bacteriophage (Qp) in the laboratlory of Professor I. Wata- iiabe. A modification of t,he procedure employed to isolate the AIS-2 replicase suf- ficed to yield a highly purified and active &p replicase. It is the primary purpose of the present paper t,o det,nil the pertinent properties of tJhe purified enzyme. Comparison of t,he AIS-2 and Qp enzymes isolated from the same host coilfirms t,he predicted requirement' for homology. Each replicase recognizes t>he RKA genonie of its origin arid requires it :is II template for iiornial synthetic activity. The baci erial viriises employed :ire 11s-2 (originally obtained from Dr. 9. J. Clark) and Qp (kindly provided by Professor Watanabe). It is essential for any laboratory working with both RIS-2 and Qp to monitor contin\ially for contamination of one by the other. Fortriiiately there is virtually no serological cross react ion between the so that the appropriate antisera can be itsed for identification and piirity checks. The host and assay organism is a mutant Hfr strain of E. coli (Q-13) isolated in the laboratory of W. Gilbert by Diane 1-argo. It, has (he convenient property of lacking6 riboniiclease I and RNA phosphorylase. Preparation of viriis stocks and pririfirtl RIA followed the methods of Doi and 8piegelman.2 The basic mediiim employed for growing infected cells and prodiicing virus contained the followirlg in grams per liter: XILCl, 1 ; 31gF0,. 7 H20, 0.06; gela- t,in 1 x casamino acids (vitamin-free), 15; glycerol, 30; to this is added after separate aiito- claving 7 ml of 0.1 M CaCh and 10 ml containing 4 gm of Na2HP04. 7 H20 and 0.9 grn KH,P04. Lysates in lit,er qiianlitien are first prepared to be iised for infection of larger volumes of cell siis- pensions. These are obtained by infecting log phase ciiltures (Olletia of 0.25) with a prirified phage preparation at a multiplic*ity of about 3. is complete and then monitored for titer and purit>y of t,he phage. Siich lysates ran be stored frozen at -17°C indefinitely and thawed just prior to use. In geiierd, 35-liter quantities of cells are grown up in carboys to an Ohti0 of bet,ween 0.275 and 0.290. The temperatiire in the carboys is 34T, while the temperature of the water bath in which t,hey are immersed is maintained at 37°C. When the cells reach an O&O of 0.275, they are infected with virus at a niidtiplicity of between 10 and 50 and allowed to aerate for mixing for 1 min. The aerat>ion is interriipted for 10 min for absorption, reinstit rrted, and the incubation continiied. At 25 min sufficient sucrose arid may- nesium are added to give final concentrations of 18% and 0.01 M, respectively. hfter another 5 min the process is terminated by the addition of crushed ice. The cells are harvested in a Sharples centrifuge and stored at - 14"C, at which temperatlure ability to yield active enzyme is retained for periods exceeding 6 months. Vninfected cells are prepared and stored in the same manner. To provide uniform preparations for enzynie isolation, the cells are thawed sometime prior to iise and resuspended (20 gm of packed cells in 100 ml) in a solation containing 0.01 M Tris buffer p€I 7.4, 0.001 M MgC12, and 0.0005 M mercaptoethanol and 5 Hg/nil of DNase. After thorough resuspension with a magnetic stirrer at 4"C, the suspension is divided into convenient aliqiiots in plastic tubes, frozen, and stored at - 14°C. P32-laheled ribonucleoside monophosphates were synthesized as described previously. The labeled mononucleotides were converted enzymatically to the corre- sponding ribonucleoside triphosphates by a kinase preparation isolated6 from E. coli. Unlabeled riboside triphosphates were from P-L Bio- Materials and Alrthods.--( 1) Hact~rict and z~irnses: (2) Preparation of infected cells: They are incubated while shaking at 37°C rintil 1 (3) Radioactive substrates: (4) Chemical and biological reagents: chmicals, Inc., Milwaukee, Wisconsin. IINase, 2 X recrystallized, was from Worthiiigtori Bio- chemical Company, Freehold, New Jersey; it was further purified on DEAE columns to remove contaminating ribonuclease. Phosphoenolpyruvate (PEP) and the corresponding kinase (PEP kinase) were purchased from Calbiochem, he., Ims Arigeles, California; lysozyme from Armour and Company, Kankakee, Illinois; aiid protamine siilfate from Eli Idly, Indianapolis, Indiana. The turnip-yellow-mosaic-viriis (TY MV-RKA) and the "satellite virus" of the tobacco-necrosis- virus (STNV-RNA) were both kindly provided by T>r. E. Reichmann of the Botany Department at the IJniversity of Illinois. The following procedure is described for 20 gm of packed cells. The frozen cell suspension (120 ml) is thawed and to this is added 0.5 mg/ml of lysozyme, follow- ing which the mixtrrre is frozen and thawed twice, using methanol and dry ice as the freezing mix- 1,iire. To the lysate are added 0.9 ml of 1 M MgC1, and 2.5 pg/ml DNase, and the resiiltirig mix- tnre is incubated for 10 miri in an ice bath. The extract is then centrifuged for 20 min at 30,000 X g and the supernate removed. The pellet is transferred to a prechilled mortar, ground for 5 min, and then resiispended in 30 nil of the same biiffer as used for the cell suspension except that the magnesium concentration is raised t>o 0.01 Ai` lo increase t,he effectiveness of the DNase digestion. The extract is t,hen cerit~rif\iged at 30,000 X g for 20 miii arid the two siipernates are combined, adjusted to 0.01 M ElITA (previoiisly broright t>o p11 7.4), and incubated at 0°C for 5 min. In- soluble proteins appear aiid are rernoved by ceritrifrigaliori :it, 30,000 X g for 20 min. At t,his stage, a t,ypical active infected ext,rac>t, has an Ol),," of between 150 :md 180. 1,ower valiies commonly sigrial a poor iiifeciiori with :I resiill itig low yield of enzyme. TO the c:le:ired sriperiiatant fluid is added 0.01 mg of protarnine srilfate for e:tcli 01)260 ririit. After IO mill the precipitate, coritairiirig virtually all the enzyme activity, is collec.1ed by c~entrifiigai~ioti at 12,000 x g for 10 min. It is dissolved in 12 nil of "st:tridartl I)iiffer" (0.01 tris briffer, plI 7.4; 0.005 hi hlgCla; 0.0005 M rnercaptoeltiariol), adjiisled to 0.4 izf (N t I4)?SO4, :&rid allowed to stjarid overnight, at, 0°C. This period of waiting is irriportarit for the siit)seqiierit, frmcl ioriatiori sirice complete disaggregation was foiind t.o be esseiil ial for :~cceplal)le separat ion of 1 he rep1ic:tse from transcript ase. The extract is diluted wit,h 24 ml of slatitl:trd biiffer, and after 20 iriiri is c.erilrifriged at, 30,000 X g for 20 min; for each 40 in1 of siiprrii:tt:int are :tdtled 12 nil of :i 0.5u/, soliitioti of protmiirie siilfwte. The pre- cipitate which forms coti1:iiris vir1 iially dl of t tie 1)NA-depeiidenl IlXh polymerase along with an RXA-iritleperitleiit RNA polytiieriziiig :ivk ivil y. `lY~e RNA repliwse remains in l,he sriper- n:tlaiil and begiris to show good tlepeiidenc~ oii :itlded 1tNA. (Note: This is OW of the critical steps in the frwtiotiatioii, :mcl :tiiy v;triatioti in tiost, tnetliiirii, titoe, or Ietiiperatrire of irifectioii modifies the :tiiioiiiit of 1)roI:imiiie reqiiired to :tcliieve sep:tr:it,ion. It is often safer to titrate small aliqiiots and determiire the :trnoiitit of protariiiiie needed by appropri:Lte assays.] After 10 miri the preripitate is removed by cetitrifiigittioii at 12,000 x g for 10 inin. To the resiilt,ing siipernate is added an eqiial voliime of s:ttiiriited itniniotiiiim siilfate (satriraled at 0°C arid adjiisted to pH 7.0 with ainmoniiini liydrosicle). Afler 10 miri :it 0°C the precipitate is collected by centrifiigation at 12,000 X g for 10 min :md dissolvetl in 4 rill of stitridard briffer c:ontaiiiirig 0.4 M ammonium siilfate. The dialyzed fmctioii is atljiisied lo 0.05 .If anitiioiiiriiii siilfale with sl:iridard hiiffer and passed through a TIEAE coliinin (1.2 X IO ciii) whicati is w:ished with 100 in1 of standard biiffer jrist prior to use. After loading the protein, the c~oliinin is washed with 40 1111 (Jf siandard biiffer containing 0.12 SaCl which removes proimiiiic, :I p~ly-A syiit,hetase,' and residual K-dependent ribonidease. The enzyme is then clrited with 35 ml of standard biiffer containing 0.20 M XaC1. To fractions possessing enzyme act ivity, sat rir:ttcd ammonium srilfate is added to make the final soliition 10% satiiration. At, this stage, the enzyme prep:tration has aii OI)~~O/OD~~O ratio of 1.35 and risiially contains 1 irig of protein per ml. I-iitler the ionic (-oritlitions specified, no loss in activity is oh- served over a month of storage at 0°C. (6) The stuntkirtl ci,ssci!i-cis.sci!j of rnz!lnie cictiuity by iricoi~porution of rcctliocictLur nuc.1eotide.s: `I'he standard re:tc+ioti volrirne is 0.25 nil and, iinless specified differently, coritairis the following in pnioles: tris H(3 pII 7.4, 21 ; riiugtiesiriiii (.tiloride, 3.2 (when iiicliided, manganese (ahloride, 0.2); C`I'P, ATP, IV", arid (;TP, 0.2 each. The erizyme is iisiially ass:tyed at a level of 40 pg of pro- tein in t>he presence of 1 pg of RNA template. The reaction is rim for 20 min at 35°C arid ter- minated in an ice bat'h by the addition of 0.15 ml of neutralized sat,iirated pyrophosphate, 0.1.5 ml of iieiitralized saturated ort,hophosphate, and 0.1 ml of 80% trichloroacetic acid. The pre- (5) Preparation of enzyme: `Hie resiillirig soliilioti is tlicri dialyzed :igitiiist 1 liter of st,:tndsrd biifler for 1.5 hr. OD 50 40 30 -40 -30 -20 -1 0 - - - m -100 b n W 4 a 8 -50 5 V x a U FIG. 1.-Chromatography on 1lE:AIl; of second protaniiiie siiperriatarit. Jiist hefore collectioii of fract,ioii no. 1, 35 In1 of 0.2 AI NaCl iri staiidartl brifier was placed on the column. Prior to this, the cwlriiiiii liatl been washed with 0.15 M as desc:ril-ied in Mrhods, (5). It will be rioted that the peak of enzyme ac.tivity is foiiiid iii the drsc.riitlirig shoiilder of the 01) profile. I~rctrr.otri~ito#I.itl,hq. yirltfs ~~>iiii~iclc~ric~t~ id the two. cipitate is transferred to a inetnbruie filter :ind w:tshecl soveti times with 5 in1 of cold 1076 'I'CA. The membraiie is then dried :tiid coiiiitetl in a licliiitl s~:ititill~~tji~~i coiiriter as described previoiisly.' This washing provediire briiigs zero time ~~iiiit~ ~IJ less than SO (ym with inprit coruits of 1 x 106 rpm. The spwific activit ics of the 1:hek:d triphosphates :dtled were adjristed so that with the eficieiicy eniidoved, 1 X IOG ('1)iii c~orrcspirids to 0.2 ~IMJIW of the corresporidirig triphosphate. E'igurc: 1 compares the elution profiles of protein arid c~rizytric act ivi1,y of ptqxwatiotw [sw Meth.ods, (5) ] derived froiii iiifcctetl and twiiiiifvct(d cvlls. Infected prep:mttioiis c?tliihitj :L 1)olyiiimiw :w(,ivit,y whicli elut (:s with 0.2 Ill NaCI, responds exc:ctllently to :tdd(td QP-IZNA, :~nd is devoid of thr L)n'A-d(.l,eridoiit l P*F) 3938 4.5 I!) 96 8 QP DNase (5 pg) QS RNase (1 pa) QP __ - Conditions of assay are those describer1 in Methods, (GI, evcept that RIn++ was iiirhiiled at 0.2 prnides per 0.25 ml. - 37 - Assay mixture Complete -QS-HNA - GTP -Mg, -Ah 4142 75 None QS - Evcept for the additions noted, tile assays uere ca~ ried out under the standard conditions described in Wethods, (6) Table 1 shoi~s that after thc DEAE step the enzyme is completely dcvoid of ribonuclease I, phosphorylase, and the IC+-stimulated ribonuclease. The presence of the energy-generating system (PEP + PEP kinase) has little effect on the reac- tion (Table 2), indicating freedom from intcrfcririg enzymes v hich can desti.oy ri- boside tripliosphat cs. Firially, DNase h:iS no influence on the rcactioii, xiliereas the presenre of even sm~lll amounts of pancrnitic~ riboriuclease completely c1iiiiiii;ites the net synthesis of polyriborwcleotidc. The enzyme system requires not oiily tciiiplate brit, in addition, all four riboside triphosphates (Table 3). The replicxrc. has iiri aLwlut e requirement for divalent ions, magnesium being thc 1)refcrrc.d ion with homologous RNA. llangunese sub- stitutes partially (10%) arid induces interesting rh:mges in the nature of thc reac- tion, the details of which will be described e1~ewhere.l~ Figure 2 shon s t hc kirictics observed in a read ion mixture containing saturating amounts of template (1 pg RNA per 40 pg of protein). Continued synthesis is observed at 35°C for periods exceeding 5 hr. It will be rioted that in 2 hr the amount of RNA synthesized correspoiids to 5 times the input template. By variation in the amourit of RNA added and the time permitted for synthesis, vir- tually any desired amount of increase of the starting material has been achieved. The cessation of synthesis within 5-10 min reported by otherss-ll for presumably similar preparations has been observed by us only in the early stages of purification. Figure 3 examines the effect of added amounts of protein at a fixed level of template (5 pg RNA/0.25 ml). It 1s evident that the reaction responds linearly, indicating the absence of interfering contaminants in the purified enzyme. We now turn our attention to the primary question whicbh the present study sought to resolve. Table 4 records the abi1itic.s of vwious IiNh molecules to stimulate the Qp replicase to synthetic. activity at the saturation concentration (1 pg) of homologous RNA and twice this level. The response of the Qp replicase is in accord with that, reported' for the AIS-2 replicase, the preference being clearly for its OWI~ template. The only heter- ologous RNA showing detectable activity is TYAIV and, at the 2-pg level, it supports a synthesis corresponding to 6 per ccrit of that observed with the homologous Qp- RNA. Both of the heterologous viral RXL',4's, AIS-2, and STNV are completely inactive and, again, so are the ribosomal arid transfer RNA species of the host cell (E. coli &-1R). ed, bulk RNA from infected cells shows some (B) Specific inizplatr rrqii iwiwcnts of the replicases: As might he exp 20 40 60 PROTEIN FIG. 3.-lZesponse lo added protein. As- says were rim for 20 min at 35°C under coil- ditions specified in Methods, (6). Again, the incorporation of 4,000 cpm corresporids to the synthesis of 1 pg of RNA. 30 60 90 120 MINUTES- FIG. 2.-Kinetics of replicase activity. Each 0.25 ml contained 40 pg of protein and 1 pg of Qp-RNA. All other conditions are as specified in Methods., (6). The specific ac- tivity of the LTiTP32 was such that the in- (norparation of 4,000 cpm corresponds to the synthesis of 1 pg of KNA. templatirig activity which iricreaqes as the infection is allowed to progre is no detectable DXA-dependent RNA polymerase activity. To permit a definitive cornparkon of' the two replicases derived from the same host, the AIS-2 enzyme was isolated from appropriately infected Q-13. The puri- fication of the AIS-2 replicase followed precisely the same protocol as described for the QP erizyme [Methods, (,5)] except that 0.22 ;\I NaCl is used for elution from thr DEAE column. The results of the comparison between the two replicases are shown in Table 5 and they arc satisfyingly clear-cut. The AB2 replicase shows no evidence of ma- repting the Qp-RNA as a template at either level of RSA input. Similarly, the Qp replicase completely ignores the AIS-2 RIVA while func~tiorling quite well on it5 own template. It would appear from Ihese daia that the 1)redic.tion of' tcni1)latv specificity IS completely confirmed. It is possit)lc\ to isolate comparatively pure replicase from suitably infected wild-type Hfr strains However, the use of the mutant Q-13 offered an obvious advantage for the purifica- tion of the replicase since the crude extract was already free of ribonuclease I arid phosphorylase. The two remaining interfering activities were due to the DNA- dependent RNA polymerase (transcriptase) and a potassium-stimulated ribonu- clease. It is important to emphasize that the complrte removal of the transcriptase is Dzsrussion.-(l) State of the enzyme and the method of purtjication: I:UsruNsu ow QS l~P;Pl,1~~.4sP: 'I'O IhFFERENT 'hMI'L.\'PES Input Lerds of RN:\--- 4929 4945 146 312 3 5 26 MS-2 ltibosonial RNA 45 9 sRNA 15 57 Bulk RNA from in- fected cells 146 263 Satellite virus 61 51 I)NA (10 pg) :;G Template 1 Pg 2 %lm Conditions of assay are those sperified in Jleth- ods, (6). Ho\\ever, as in all cases, assay for DNA- dependent activity is carried out at 10 pg of DNA per 0.25 ml of reaction mixture. Control reactions containing no template yielded an average of ZO epm. Conditions of assay are those specified In Methods, (a), with h'In++ present at 0.2 pmoles per 0 25 ml. ctssent.i;tl if yuestioiis of iiicchanisni and replicase specificity are to be answered wit'h- out ambiguity. The t,ranscriptasc can employ any RNA as templates for RNA synthesi~'~~ l4 and, in the process, forms a ribonuclease-resistaiit structure Con- sequent~ly, the use of DKase or act'irioniycin D does not ensure against confusion wit,h t,ranscriptase activity. 111 the fractionation procedure described [Methods, (5) 1, most of the transcriptase is removed in the prec:ipit,at.e fraction of the second protamine step. The remainder is left behind as a late component in the DEAE column. The potassium-dependent nuclease is tightly bound to cell membrane fragments which are discarded in the low-speed fraction by our coniparatively gent,le freeze-t,haw niet,hod of cell rupture. The small amount of ribonuclease that does contamiiit1t.e the extract is removed from the DERE column by washing with 0.12 llf NaC1 which at the same hie eliminates protamine and a poly-h-synthesiz- ing en~ynie.~ The resulting freedom from iiit,erfering and confounding activities makes it possible t,o study the replicase in a simple mixture coiitxinirig only the re- quired ions, substrat'es, and templat,e. We may summarize the distinc- tive propert,ies of the purified replicases described above as follows : (a) coniplet,c dependence on added RNA; (0) competence for prolonged (more than 5 hr) syn- thesis of RNA; (c) ability to synt'hesize many times the input ternplate; (d) satura- tion at low levels of RSA (1 pg RNA per 40 pg protein) ; and (e) virtually exclusive requirement for homologous ternplate uiidcr optimal ionic conditions. In view of the very dift'erent st.ates of purity, it is difficult to interpret the differ- ences in the properties observed with the enzymes reported here as compared with those det,ected15 or isolated8-12 by ot>hers. suggests that the purificat'ion has riot achieved removal of contamiriatirig RNA and, in any case, precludes examination of template specificity. August et a1.10-12 isolated an enzyme from 13'. coli infected with a mutant of f2 which is stimulated by a variety of RNA species, including host ribosomal and sRNA. It is conceivable that the f2-replicase is nonspecific. However, t'he August prel~arat'ion requires 20 pg of RNA for each pg of protein, whereas the enzymes described here are fully saturated at 0.025 pg per pg of enzyme protein. The authors point out," that t,he iiiordinat'ely large amounts of RSA required may be due to t>he detectable mn- (2) Conzpayison with other viral-induced enzymes: Activity independent of added RNA8. 586 BIOCHEMISTRY: HARU-ITA AsD SPIEGELII.IAS PROC. N. A. S. lanunalio~i of thcir prq)i!ral hi with ribo~iric~l~~iist~. Uiidcr t hc ('ir('~ililsliiiic(is, oirc should withhold judgnierit on llic sigiiificaric-e of thc appnrcmt lack of spedicity since it is open as ye1 to :t rathcr irivial cxplariiitiori. The coinparison of the two purified replicases reported establishes that each requires its homologous template. The experiment with the satellite-RNA (STNV) was a particularly interesting challenge. ReichmannlG showed that the satellite virus contains only enough RNA to code for its own coat protein, which suggcsts that it must employ the replicase of the companion virus (TNV) for its inu1liplic:it ion. This inil)lies citlier that the satellite is rclatcd in scc~uc~~ic.c to th~ TNV virus, or that it poss inittirig it to employ any vim1 RNA rq)Ii(*a~e. The fact that STKV-RNA did not serve as a template for either oiic of Ilic two purified replicases implies that the answer will be found in at least parlial sequencc homology betwcen STNV and TNV genomes, a prediction opcii to ~~spcrinicrrtal test. The spevificity relatioiis cxhibitctl raise the question of t Ire i~re~~h:~~risiri used by the replicase to distinguislt its tcmplal(~ from othcr HNA inolecules. The irivolve- merit of a heginnirig scyucnw is an obvious possihility. Howrver, :ts will be sho~vn,~~ the rerognition niechanisin is evcn more subtle, bcirig designed to avoid replication of fragments of its own geiionic even if they c~mtain the beginning sequence. It should be evident that the replicases are approaching a state of purity permitting the performalice of unanibiguous experiments which can hope to illu- minate the inecha~iism'~ of the RNA replicative process. Summary.-Two RKA replicases induced in the same host by unrelated RNA bacteriophages have been purified and their responses lo various RNA molecules examined. Under optimal ionic conditions both are virtually inactive with heter- ologous RNA, including ribosomal and sRNA of the host. Neither replicase can function with the other's RNA. Each recognizes the RXA genome of its origin and requires it as a temptate for synthetic activity. This discriminating selectivity of its replicase is of obvious advantage to a virus attempting to direct its own dupli- cation in a cellular environment replete with other RNA molecules. (3) ImpZications of template specihcity requirement: We should like to express our deep appreciation to Ilr. I. Watanabe for making available to us the Qp virus which made possible the informative comparison reported. Similarly our profound thanks are due to Drs. W. Gilbert and J. 11. Watson and to Miss Iliane Targo for providing the mutant Q-13. Finally, we are grateful to Rlrs. Louise Jordan for skillful assistance in the per- formance of the experiments. * This investigation was supported by 1J.S. Public Bealtli Service research grant, CA-01094 Naruna, I., K. Nozu, Y. Ohtaka, arid H. Spiegelman, these PROCEEDINGS, 50, 90.5~91 I (1963). Doi, Roy TI., and S. Spiegelman, these PROCEEDI~XGS, 49, :355-360 (1963). Wat,anabe, I., Nihon Rinsho, 22, 243 (1964). Overby, L., G. H. Barlow, R. H. Doi, M. Jacob, and S. Spiegelman, J. Kacteriol., in press. Gesteland, R. F., Federation Proc., 24, 293 (1965). Bresler, A. E., Biochim. Bwphys. Acta, 61, 29-33 (1962). August, J. T., P. J. Artiz, and J. Hurwit,z, J. Hiol. (:hem., 237, 3786-3793 (1'362). Weissmann, C., P. Burst, R. H. Burdon, M. A. Billeter, and S. Ochoa, these I'RO(:EEI)IN(+S, 51, from the National Cancer Institute and a grant from the National Science Foundation. (Q-13 is a derivative of A-19, a RNase negative mutant reported here.) 682 (1964). 9 Weissmann, C., TJ. Simon, and S. Ovhoa, these I'R~CEE~INGS, 49, 407 41 4 (1 963). VOI.. 54, 1!65 HIO('HB:I1IIS!L1KY: HtllZUNA AND SPIBGELMAN 5x7 Io Aiigitst, J. T., S. Cooper, 1,. Shapiro, and N. 11. Zinder, in Cold Spring Harbor Symposia "August, J. T., L. Shapiro, and L. Eoyang, J. Mol. Bwl., 11, 257-271 (1965). l2 Shapiro, L., and J. T. August, J. Mol. Biol., 11, 272-284 (1965). l3 Krakow, J. S., and S. Ochoa, these PROCEEDINGS, 49, 88-94 (1963). l4 Kakamoto, T., and S. B. Weiss, these PROCEEDINGS, 48, 880-887 (1962). on Quantitative Biology, vol. 28 (1963), p. 99. Baltimore, D., and R. &I. Franklin, Biochem. Biophys. Res. C'ommun., 9, 388-392 (1963). Reichmann, M. E., these PROCEEDINGS, 52, 1009-1017 (1964). Haruna, I., and S. Spiegelman, manuscript in preparation.