Cell-free translation of purified virion-associated high-molecular-weight RNA synthesized in vitro by vaccinia virus.

(1)

0022-538X/78/0028-0905$02.00/0

Cell-Free Translation of Purified Virion-Associated

High-Molecular-Weight RNA Synthesized In Vitro by

Vaccinia

Virus

WALTER BOSSART,t ENZO PAOLETTI, AND DONALD L. NUSS*

Division of Laboratories and Research, New York State Department of Health, Albany, New York 12201

Received forpublication11May 1978

Virion-associated high-molecular-weight (HMW) RNA synthesized in vitro by

purified vaccinia virus particles has been translated in a wheat germ cell-free

proteinsynthesizing system. Purified HMW RNA directs the synthesis of

trans-lation products whichareidentical to the translation products made in response

toinvitro-synthesized,virion-released 8 to12SmRNA. The translation of HMW

RNA proceeds exclusively through a 5'-terminal cap-mediated initiation step.

Furthermore, only onecoding sequence is translated per HMW RNAmolecule,

and that sequence is probably located near the 5' end ofthe molecule. These

conclusions are based on the following results. (i) Sodium dodecyl

sulfate-polyacrylamide gel electrophoresispatternsoftranslation products

syn-thesized in response to HMW RNA and in response to 8 to 12S mRNAwere

qualitatively identical. (ii) Onanequalweightbasis, HMW RNA was 25 to 30%

as active as 8 to 12S mRNA in stimulating in vitro protein synthesis. (iii)

Unmethylated HMW RNAwastranslatedat10% theefficiency of themethylated

form of this RNA.

(iv)

m7pG

inhibited the translationof fullymethylated HMW

RNAby 90%. (v) After the initiationstepoftranslationwasblockedby

aurintri-carboxylic acid, theratewithwhichaminoacidswereincorporatedintoindividual

polypeptides decreased in a similar mannerfor the translation of both HMW

RNAand8 to 12SmRNA. Virion-released8 to 12S mRNAderived from

virion-associated HMW RNA during a chase in the presence ofATP, GTP, and

S-adenosylmethionine

wasalso translated. Atlow RNAconcentrations, the derived

RNA appearedtostimulate amino acid incorporation more efficiently than the

HMWRNAprecursor. However,athigherconcentrations of this RNA,protein

synthesiswasseverely inhibited.

RNA synthesized and extruded by vaccinia

virusparticlesin vitro has an average

sedimen-tation value of between 8 and 12S (15, 35),

contains a

7-methylguanosine

cap at the 5'

ter-minus (36),and ispolyadenylated atthe3' end

(16,24). Underappropriate conditions, vaccinia

virus particles

preferentially

synthesize a

high-molecular-weight (HMW), 20 to 30S form of

RNA(26,29a). Theselarge transcriptsarefound

only

inassociation with the viruscoresandare

not released from the virion as HMW RNA

molecules (26, 27, 29a). HMW RNA is also

cappedatthe5'end(25),but it isnotextensively

polyadenylated at the 3' end (27, 29a). This

virion-associated HMW RNA can be chased

withinthe virionparticletoRNA in the8 to12S range, some of which is released from the virus (27). Thiscleavageprocess is notdependenton tPresentaddress: Institute forMicrobiologyandHygiene, University ofBasel, CH-4000, Basel, Switzerland.

continued RNA synthesis but does require

ri-bonucleosidetriphosphates and is

prevented by

theaddition of ethidium bromide

(27).

Recently,

an endoribonuclease whichappears

specific

for

the

cleavage

of the virion-associated HMW

RNA has been solubilized from vaccinia virus

particles

(29). Furthermore, the

transcriptional

complexity of virion-released 8 to 12S mRNA

and virion-associated HMW RNA is identical

(28). These combined results suggesta

precur-sor-product

relationship

between

virion-associ-ated HMW RNA and virion-released 8 to 12S

mRNA.

Recently both we (7) and H. Pelham (30)

described the effect of UV irradiation on the

expression of vaccinia virus gene

products

syn-thesizedinacell-freesystem

coupling

transcrip-tion and translation. Although different

trans-lation systems were used in these two

studies,

similar resultswereobtained. These results

in-905

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

906 BOSSART, PAOLETTI, AND NUSS

dicated that each translationally functional mRNA species produced in vitro by vaccinia virus cores is synthesized from individual pro-motersites.However,wecouldnotruleout the possibility that the polypeptides expressed in the coupled system are the products of se-quences found very near the 5' end of several long transcriptional units, whereas other se-quences that are expressed only after normal processingoftheselongtranscripts are transla-tionally silentin the invitro translationsystem. To investigate this possibility and to further characterize the virion-associated HMWRNA, we have determined a number of parameters with respecttothetranslation of this RNA in a wheat germ cell-free protein-synthesizing sys-tem.

MATERIALS AND METHODS

Synthesis and purification of vaccinia RNA. Vaccinia virus (WR)was grown in andpurifiedfrom

HeLa cells as described by Joklik (13). RNA was

synthesized in a standard in vitro RNA polymerase reactionconsistingof 50 mMTris-hydrochloride (pH 8.4), 10mMMgCl2, 10mMdithiothreitol,0.05%

Non-idetP-40, 2 mM eachATP, GTP, and CTP, and0.2 mM [:3H]UTP, (0.025 Ci/mmol). Purified virus was

addedto afinal concentration of4 to5optical density

units at 260nmperml.Incubationwasat37°C. For the preparation ofthemethylatedand unmeth-ylated forms of virion-released 8 to 12SmRNA, the

RNApolymerase reactionwasincubatedat37°C for

30minin thepresenceofeither10jimof S-adenosyl-methionine (AdoMet)or 100jim of S-adenosylhomo-cysteine (AdoHcy). Viruswas removed from the

re-action mixture atthe end of theincubation time by centrifugation at 15,000 rpm for 30 min. RNA was

phenol extracted from the virus-freereactionmixture and purified on sodium dodecyl sulfate

(SDS)-sucrose gradientsaspreviously described(25, 27).

For thepreparationofthemethylated and

unmeth-ylated forms ofHMW RNA, reaction mixtures

con-taining eitherAdoMetorAdoHcywerepreincubated for20min in the absence ofUTPtoreduce theATP

concentration(29a), andsynthesiswasthen startedby

theaddition of[3H]UTP (0.025Ci/mmol) to afinal concentration of0.2 mM. Afteranincubation period

of40min,RNA wasextracted from thevirusparticles asdescribedpreviously (26, 27, 29a), and then purified

by sedimentation through SDS-sucrose gradients. Bothtypesof RNA wererecovered from thegradients

byethanolprecipitation.Threeadditional ethanol

pre-cipitationswereperformedto removeall traces of SDS before the addition of RNA to thecell-free translation system.

Cell-free translation. A cell-free

protein-synthe-sizingsystem was prepared from Niblack brand wheat germ by the method of Roberts and Paterson (32),

excluding the preincubation step. Reaction mixtures for cell-free translation contained wheat germ extract

at a finalconcentration of 25optical density units at

260 nm perml;20 mMN-2-hydroxyethyl piperazine-N'-ethanesulfuric acid (pH 7.5); 135 mM KOAc; 3.5

mMMg(OAc)2; 1.25mMATP;0.25mMGTP;15mM

creatine phosphate (Sigma Chemical Co.); 75 jg of creatine phosphokinase (Sigma) perml.; 3.6mM di-thiothreitol; 0.3 mM spermidine-hydrochloride; 50

,uMeach amino acid minus either leucine or methio-nine; and either 200jiCi of [3H]leucine perml (54.6 Ci/mmol; NewEngland NuclearCorp.) or500jiCiof

[35S]methionine perml (573 Ci/mmol;NewEngland Nuclear). Insomereactions, AdoHcywasincludedat aconcentrationof 0.3 mMtoinhibitmethyltransferase activitypresent in the wheat germextract.

RNA samples were heated at60°C for 1 min and subsequently quenchedonicejust before additionto the translation system to reduce any RNA-RNA aggregation. Protein synthesis was monitored as de-scribed byMansand Novelli(21).

Polyacrylamide gel electrophoresis. Samples from translation reaction mixtureswerepreparedand

analyzed by electrophoresis on 15% polyacrylamide gels followed byautoradiographyas described previ-ously (7). Radioactivepolypeptideswere detectedon

somegels bythefluorographicmethodofBonnerand Laskey (4).

RESULTS

Stimulation ofprotein synthesis by

vir-ion-associatedHMW RNA. Purified vaccinia

virion-associated HMW RNA, synthesized

un-derrecentlyestablishedconditionswhich permit

the preferential synthesis of these large

tran-scripts (29a),wastested for theabilityto stim-ulateprotein synthesis inawheat germcell-free translation system. The level of amino acid in-corporation into protein which was directed by virion-released 8 to 12S mRNA or virion-asso-ciated HMW RNA is shown in Fig. 1A and B, respectively, as a function of RNA concentra-tion. Notice that both RNA populations stimu-late protein synthesistothe same extent at the concentration of RNA which issaturating. How-ever,the concentration of RNA whichgivesthis maximal stimulation is considerably lower for the 8 to 12S mRNAthan for the HMW RNA. The concentration of 8 to 12S mRNA which gave maximal stimulation ofprotein synthesis

wasconsistently foundtobe between three- and fourfold lower, on an equal weight basis, than the concentration of HMW RNA which gave maximal protein synthesis under identical con-ditions. In this regard, sedimentation analysis has shown that the HMW RNA is

approxi-mately three- to fourfold larger than 8 to 12S

mRNA (25-27). Furthermore, a comparison of

theincorporationofmethyl groupsatthe5' end

with theincorporation of UMPatinternal sites of these RNAs indicatedthat,on anequalweight

basis, 8 to 12S mRNA has three- to fourfold morecapped 5'-terminal ends thanHMWRNA

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

(3)

VACCINIA

synthesized under identical conditions (25).

Analysis by SDS-gel electrophoresis of

poly-peptidessynthesized inresponse to

virion-asso-ciated HMW RNA andvirion-released 8 to 12S

mRNA revealed thatthese two populations of

RNA directed the synthesis ofsimilar

transla-tionproducts. (Compare lanesAand B in Fig.

2.)Thereare someobservabledifferencesin the

relative synthesis of the larger (greater than

50,000 molecular weight) polypeptides when

comparing thetranslation products coded forby

HMWRNA with thetranslation products

pro-grammed by 8 to 12S mRNA; the larger

poly-peptides are moreprominent in translation

re-actions programmed by HMW RNA. Longer

exposure of the 8 to 12S mRNA translation

products doesreveal,however, that the

synthe-sis ofpolypeptides aslarge as90,000daltons is

directedby this population of RNA. These

quan-titative differencesareprobably duetothe sizing

ofthese RNAs on sucrose gradients as partof

the

purification

scheme. Itis importantto note

thatthesynthesis of large polyproteins didnot

occurduring the translation of the HMW RNA.

This result suggests that normal termination

E250

LO) ' 150

0

E

Z 50

0

o 25

0 0

z

LZ 15

z

0 I

LJ 5

C

Po C 10 20 30 0 30 60 90

RNA (,ug/ml)

FIG. 1. Stimulation ofpolypeptide synthesis in a

wheatgermcell-freetranslation systeminresponse tothemethylated andunmethylated forms ofHMW RNA and 8to12S mRNA. Themethylatedand

un-methylated forms ofHMWRNA and 8to12SmRNA

were synthesized andpurified as described in the text. Translation was in the presence of0.3 mM

AdoHcy for30minat22°C. The levelofamino acid

incorporation directed byeachtypeofRNA is pre-sentedasafunction ofRNAconcentration.(A)

Meth-ylated8to12S mRNA; (B)methylatedHMWRNA; (C) unmethylated8to12SmRNA; (D) unmethylated

HMW RNA.

N

sm1m

A

B

C

:_ 66,000

o

47,000

:_ 30,000

AMaf

40wfk

_IvA

i_p 40

..

4=m

17,

000

FIG. 2. Autoradiograph of an SDS-polyacryl-amidegel after electrophoresis oftranslation

prod-ucts synthesizedin acoupled transcription-transla-tion system or in response to exogenous 8 to 12S mRNA or HMW RNA. [3S]methionine-labeled translationproducts synthesized in thecoupled

sys-tem(laneC)aspreviouslydescribed(3)orsynthesized under standard translation conditions in responseto purified8 to12S mRNA (lane A) orpurifiedHMW RNA(laneB) wereanalyzedby SDS-gel electropho-resis. Each well received 500,000 acid-precipitable cpm. Themigrationpositions ofmarkerproteins of

known molecular weight are indicated on the left margin. Theexposure timewas 4days.

signals arerecognized byribosomestranslating

thelarge transcripts.

We recently described a coupled

transcrip-tion-translationsystem in which we determined

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.505.253.447.83.466.2] [image:3.505.76.218.350.536.2]

(4)

theeffect of UV irradiationontheexpressionof

vacciniageneproductsinawheatgermcell-free system (7). The results of thatstudy suggested that eachtranslationallyfunctional mRNA

spe-cies producedin vitrobyvaccinia virus coresis synthesized from individual promoter sites. A comparison of the translation products synthe-sizedinthe coupledsystem(Fig. 2,laneC)with the translation products directed by 8 to 12S mRNA (lane A) and HMW RNA (lane B) also showed some quantitative differences in the

translation products synthesized in all three

cases. However, close examination of the gel

patterns revealed no qualitative differences, at this level ofsensitivity, between the translation products synthesizedin thecoupledsystemand thosesynthesizedinthe translation system

pro-grammed by either HMW RNA or 8 to 12S

mRNA.

Requirementofacapped5' terminus for

efficient translation of HMW RNA. It is known that virion-associated HMW RNA is cappedatthe5'-terminal end(25).Todetermine whether the translation of this RNA proceeds exclusivelyviaa5'-terminalcap-mediated

initi-ationstep, thefollowing experimentswere

per-formed. Virion-released 8 to 12S mRNA and virion-associated HMW RNA were purified

from invitrotranscriptionreactions which

con-tained AdoHcy to prevent methylation of the transcription products. The stimulation of

pro-teinsynthesiswith respecttothe RNA

concen-trationwasthendeterminedfor thefully meth-ylatedandunmethylatedformsof both 8 to 12S mRNA and HMW RNA. In thisexperiment,the wheatgermtranslation systemwasalso

supple-mented with 0.3 mMAdoHcytoprevent meth-ylation ofanyunmethylated RNA by enzymes presentinthe cell-freesystem. Themethylated andunmethylated forms of each type ofRNA exhibited maximal stimulation of protein

syn-thesisatthesameRNAconcentration, i.e.,12to

17 .tgofbothmethylated and unmethylated8to

12S mRNAperml(Fig. 1A and C)and55to65 tigof bothforms of theHMWRNAperml(Fig.

1Band D).However,the unmethylated form of

both 8 to 12S mRNA and HMW RNA

stimu-latedprotein synthesis only 10%asefficientlyas

the methylated forms of these RNAs. The

re-quirement ofa 5'-terminal capfor the efficient

translation ofvaccinia8to12S mRNAhas

pre-viously been reported (34). The data presented inFig. 1suggestthat the efficienttranslation of HMW RNA also depends onthe presence ofa

5'-terminalcap.

The translation ofmRNA's which normally requireacapforefficient translationis inhibited

bytheaddition ofm7pG (11). The translationof

mRNA's which do not require the cap for effi-cienttranslation, e.g., EMC RNA (33) and, pre-sumably, the translation of RNA sequences which is initiated at internal initiation sites, is not affectedby the addition of m7pG at a con-centration which inhibits the translation of capped mRNA (14). The effect of m7pG on the translation of themethylated andunmethylated forms of both 8 to 12S and HMW RNA was tested (Table 1). The effect of m7pG on the translation of both the 8 to 12S mRNAand the HMW RNA populations was identical. The translation of the methylated forms of both RNAs was inhibited by 90% when 1 mMm7pG wasadded tothe cell-free system, whereas this concentration of

m7pG

had no effect on the translation of the unmethylated forms of these twoRNAs. Indetermining the effect of m7pGon thetranslation ofunmethylated RNA,onemust becognizant of the observation that the addition of anyRNA, e.g., rRNA (Table 1), results ina stimulation ofendogenous wheat germ protein synthesis in translation systems which are not preincubated. Thisendogenous protein synthe-sis is completely inhibited by the addition of

TABLE 1. Effect ofm7pGonthe translationof the methylated andunmethylated forms of HMW RNA

and 8 to12SmRNA

["S]methionineincorporation Reactionmixture' after

(min):'

0 10 20 30 60

-RNA

-m7pG 8.7 11.4 13.9 16.1 19.5

+m7pG 8.5 8.1 8.5 8.3 7.1

rRNA(10jig/ml)

-m7pG 7.9 15.7 22.1 25.3 33.5

+m7pG 7.6 9.2 9.2 10.0 9.1

Methylated 8 to 12S mRNA (10jig/ml)

-m7pG 70.4 163.9 230.6 276.8

+m7pG 6.3 12.3 16.7 22.7

Unmethylated8 to12S mRNA (10 jig/ml)

-m7pG 6.4 13.2 18.4 21.5

+m7pG 6.7 14.5 20.4 22.9

Methylated HMW RNA(40jug/ml)

-m7pG 89.7 186.9 247.0 280.0

+m7pG 7.3 15.1 20.6 23.3

Unmethylated HMW RNA(40jg/ml)

-m'pG 3.7 9.5 13.1 19.4

+m7pG 4.4 10.9 19.2 22.9

'Themethylatedandunmethylatedforms of HMW RNA and 8 to 12S mRNA were translated under standard conditions in the presence or absence of 1 mM m7pG. AdoHcy was present at a concentration of 0.3 mM in all reaction mixtures topreventmethylationofunmethylatedRNAbythe wheat germmethyltransferase.

bResults are presented as countsper minute x 10-" of

[C3S]methionine

incorporationinto hotacid-precipitable

radio-activityper 5jilof reaction mixture.

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

(5)

m7pG.

Thus,the stimulationofendogenous

pro-tein synthesis by the addition of unmethylated

RNA and theinhibition of that endogenous

pro-tein synthesisby

m7pG

mustbe taken into

ac-count when calculating the effect of m7pG on

the actual translation of unmethylated RNA.

Theresults presented here withrespect to the

relative translation of the methylated and

un-methylated forms of the 8 to 12S mRNA and

the HMW RNA, aswell asthe effect of

m7pG

on the translation of these RNAs, are in total

agreementwithpreviously published results for

8 to 12S mRNA (34) and stronglysuggest that

the translation of HMW RNA also proceeds

exclusively

through

a5'-terminal cap-mediated

initiationstep.

Asafurthertestof this conclusion, the

trans-lationproductsdirected by both methylated and

unmethylated HMW RNA and8 to12SmRNA

in thepresenceand absence of

m7pG

were

com-pared by

SDS-gel

electrophoresis. Ifanyof the

HMW RNA sequences are translated from an

internal initiation site, then the corresponding

polypeptide would beveryprominent inthegel

analysis of translation products directedbythe

unmethylated form of the HMW RNA. Also, the

synthesis of these polypeptides would not be

inhibited after the additionof

m7pG.

Asshown

inFig. 3, the

gel

profiles

of the translation

prod-uctsof8 to 12SmRNA, which are synthesized

only from5'-terminal initiation sites,are

similar

tothe

proffies

of HMW RNA translation

prod-ucts synthesized under each set of conditions.

Thus, we can conclude that the translation of

HMW RNAproceeds througha5'-terminal

cap-mediated initiation step and can rule out the

possibility

that there is a

significant

level of

initiation of HMW RNA translation atinternal

initiation sites.

Number and probable location of

trans-lated sequences on the

BMW

RNA

mole-cule.

Although

the data

presented

in

Fig.

1and

2andinTable1ruleoutthepossibility that the

translationofHMW RNA starts atinternal

ini-tiation

sites,

it is

possible

that multiple

se-quenceswell downstream from the5'end of the

HMW RNA are translated after translation is

initiated atthe 5' end of the molecule, as has

been determined for the translation of some

coliphage

mRNA (18). Toinvestigate this

pos-sibility,

thefollowing experimentwas

performed.

Translationof8 to12S mRNA and HMW RNA

wasallowed to initiate andproceedfor 4minin theabsence ofisotope, atwhichtimethe

inhib-itorofpolypeptidechain initiation,

aurintricar-boxylicacid(ATA),wasaddedtoafinal

concen-tration of 75 ,M. The ATA-inhibited reaction mixtures were then supplementedwith the

la-beled amino acid either at the time of ATA

additionorat1-min intervals after the addition

of ATA, andelongation wasallowedtoproceed

for a totalof30 min.Thisexperiment measures

the time it takes for ribosomes to finish the

translation of the mRNA molecules on which

they had initiated beforeATAaddition, i.e.,the

runoff

time.

Arunoff

time

for HMWRNA

last-ing three to four times longer than the runoff

time for 8 to 12S mRNA would suggest that

sequences all along the HMW RNA molecule

arebeingtranslated after initiationatthe 5' end.

Identical runoff times for HMW RNA and8 to

12S

mRNA

wouldsuggestthat onlyonecoding

sequence per HMW RNA wastranslated. The

runoff kinetics wereidentical for both 8 to 12S

mRNAand HMWRNA (Fig. 4).

As afinaltest,thetranslation products labeled

during the runoff of both8 to 12S mRNA and

HMWRNAwereanalyzed bySDS-gel

electro-phoresis. Thetime required for runoff,as

deter-mined by the decrease in the level of

radioactiv-ity incorporated into polypeptides with

time

after ATAaddition, increases

progressively

with

increasing molecular weight of thepolypeptides

coded forby the8 to12SmRNA(Fig. 5A).From

these measurements, aribosometransittime of

approximately

40codonspermin wascalculated.

Acomparison of thegelpatternsinFig. 5A and

Breveals that the individualpolypeptides coded

forby both the8 to 12SmRNA andtheHMW

RNA have identicalrunoff kinetics.

Witharibosome transittime of40codonsper

min

(calculated

from data inFig. 5),aribosome

which initiates at zero time can proceed

only

about160codons from the ribosomebinding site

by4min,the time of ATA

addition;

160codons

representsabout 70% of the

coding region

ofan

mRNA coding for a

polypeptide

of molecular

weight 30,000. Since the initiation of HMW RNA

translation,

like the initiation of8to 12S

mRNA translation, proceeds through a

5'-ter-minal

cap-mediated

step, the similar runoff

ki-netics forpolypeptides codedforby HMWRNA

and 8 to 12S mRNA suggests that the one

se-quencetranslated from eachHMW RNA

mole-culeislocatednearthe 5' end ofthatmolecule.

However,itisconceivable, though

unlikely,

that

thesecondary structure of the

large transcripts

is such that the 5' terminus is in an unusually

close functional association with the initiator codon for acoding sequence nearthe 3' end of

the HMWRNA.

Inpreliminary experimentswedeterminedthe

stability of both the HMW RNA and8 to 12S

mRNA in the wheat germ system

by

sucrose

gradientanalysis of theRNA's recovered from

the translation system with

increasing

time of 28,1978

on November 10, 2019 by guest

http://jvi.asm.org/

(6)

1 2 3 4

J. VIROL.

5

6

7 8

Gum.

Nsa_

Na..

[image:6.505.118.407.69.458.2]

M_..

FIG. 3. Autoradiograph ofaSDS-polyacrylamidegel afterelectrophoresis of translationproducts synthe-sized inresponsetomethylatedandunmethylatedHMWRNA and8 to12S mRNA in the presenceorabsence ofImMm7pG.Each well received120,000

acid-precipitable

cpm.Samples werenormalized with respectto

protein concentration byadding appropriate amounts of non-radioactive wheat germextract. (Lanes1 and 2)Methylated8to12S mRNA in the absence and presenceofm7pG,respectively; (lanes3and4)unmethylated

8 to12S mRNA in the absence and presenceof m7pG respectively; (lanes5and6)methylatedHMW RNA in the absence and presenceof m7pG, respectively, (lanes7and8)unmethylatedHMWRNA in the absence and presence ofm7pG,respectively. The positions of molecular weight markersare shown in themargins.Exposure

timeforthisautoradiographwas 3weeks.

incubation. Both HMW RNA and 8 to 12S mRNA'sweredegradedatsimilar rates, and the sourceofthenuclease activitywastracedtothe creatine phosphokinase preparation. Several lots of creatinephosphokinaseweretested,and

allwerefoundtocontainnucleaseactivity(12).

Althoughsomedegradationof HMW RNAand

8to12S mRNAwasobserved in the translation system,over50%of the HMW RNAsedimented

atgreaterthan 18Swithin15minofincubation.

The time for initiation and runoff after ATA additionwas10min(Fig. 4); consequently,under these conditions if thereweretranslation of in-ternal sequence near the 3' end of the HMW RNAafter the initiation of translation occursat the 5' end it would be reflected in both the kinetics of runoff(Fig. 4) and in thegel analysis

ofpolypeptides labeled duringthe runoff (Fig.

5).Furthermnore,allpolypeptideswhichare

nor-mally synthesized in response to HMW RNA

on November 10, 2019 by guest

http://jvi.asm.org/

(7)

TRANSLATION TRANSCRIPTS

z w

w

40-Q

20-0 5 10 20 30

[image:7.505.63.218.60.264.2]

MINUTES

FIG. 4. Incorporation ofamino acidsintoprotein

with time afteraddition ofATAto translation sys-temsprogrammed byeither HMW RNA or8to 12S mRNA. The translation of either HMW RNA (40

pg/ml)or8to 12S (10pg/ml) mRNA wasinitiated and allowedtoproceed for4minin the absence of added leucine, atwhichtime ATA wasaddedtoa

finalconcentration of75pM. [3HJleucinewasthen

addedtoportions oftheATA-inhibited reactionsat zerotime,i.e., with the ATA,oratthetimesindicated

in the graph afterATA addition. Elongation was allowedto continuefor a total of30 min after the addition of[3H]leucine for each timepoint.

[3H]-leucineincorporationintoproteinwasdeterminedby the methodofMans and Novelli (21). The levelof [3H]leucine incorporation resulting when

[3H]leu-cinewasaddedwithATA (zero time) isconsidered 100%o incorporation aspresented on the ordinate. Symbols: (0) runoffkineticsfor8to12S mRNA; (0) runoffkineticsforHMWRNA.

when initiation of translation isnotlimited are

also synthesized when the initiation step is in-hibited4min after thestartoftranslation (Fig. 5B). Given the kinetics ofrunoff (Fig. 4), this result strongly suggests that none of the

ex-pressed coding sequences in the population of HMW RNA arelocated exclusively atinternal

sites ofapolycistronicRNA molecule. From the

combined results, we conclude that only one

codingsequenceistranslated from each HMW RNAandthat thatsequenceisprobablylocated

near the 5'-terminal end of the HMW RNA molecule.

Translation of8 to12SRNAderived from BMWRNA. Since normallyreleased 8to 12S mRNAwasfoundtobeapproximatelythree- to fourfoldmoreefficienton aweightbasis in stim-ulating protein synthesis than was purified

HMW RNA, it was considered important to

determinewhether 8to12SRNA derived from

corescontaining HMWRNA during incubation

in the presence of ATP, GTP, andAdoMet is

translated

with a similar efficiency as HMW RNA or asnormallyreleased 8 to 12S mRNA.

Virion-released8 to 12S mRNA and

virion-as-sociated HMW RNA were synthesized as

de-scribed above. Cores containing HMW RNA

synthesizedat limiting ATPconcentrations were

washed twice, and the RNA in one-half of the

cores was chased for60min in the presence of

ATP, GTP, and AdoMet. The three RNA

spe-cies were then prepared for translation by

phenol extraction, chromatography on G-25

Sephadex to remove nucleoside triphosphates

and,finally, ethanol precipitation and

resuspen-sion inwater.Itmust be stressed that

prepara-tionof these RNAs didnotincludeapurification

step on sucrose gradients to remove traces of

virion-associated 8 to 12S RNA in the HMW

RNA preparation as was done for the other

experiments reported in this paper. Since the

contaminating 8 to 12S RNA is three to four

times more efficiently translated than true

HMW RNA, even a 10% contamination will

significantly

decrease theapparentdifferencein

translational efficiencies of thesetwoRNA

pop-ulations. Sedimentation patterns of the three

RNA

preparations

wereexaminedtodetermine

whether the chase was efficient. There was a

slight contamination of the HMW RNA with

RNA whichsedimentsinthe

region

ofreleased

8to12SmRNA(Fig. 6).Asinmostexperiments,

HMW RNA was chasedto 8 to 12S RNA and

released from the virion with an efficiency of

approximately 65%.

Shown inFig. 7A, B, and Carethesaturation

curves for the translation of the three RNA

species.Asexpected, the difference in saturation

curvesforHMWRNA andnormally released8

to 12S mRNA is reduced from that shown in

Fig. 1. Atlow concentrations ofRNA, the level

of

protein

synthesis

stimulated

by

derived8to

12S RNA issimilartothat of

normally

released

8 to 12S mRNA rather than the

virion-associ-ated HMW RNA.Thisincreased

specific

trans-latability

suggeststheprocessing of HMWRNA

intoactive 8 to12S mRNAmolecules. The

sat-uration curve is unusual, however, in that the

level ofproteinsynthesis atthe apparent

satu-ration levelof RNA is reducedfrom thelevels

observed for both the

normally

released 8to12S mRNA and HMW RNA.Furthermore,athigher

RNA concentrations, the level ofprotein

syn-thesis isradicallyreduced.

This reduction in proteinsynthesisathigher

RNAconcentrationssuggeststhepresenceofan

inhibitor.

Recently,

the

synthesis

ofa

polyade-nylated and capped small

(4S)

RNA

species

VOL. 28,1978

on November 10, 2019 by guest

http://jvi.asm.org/

(8)

0 1 2 3 4 5 6 7

_

* ,zbF

.S . u.. ..

w X yX

rs

1E

illiL

Fs-s--e

k _E 39 =. .b.; :..

l'S | s ':'

_ Xb |

.kw. w ,,

i ,

^g

..

4 4

>

A

B

FIG. 5. Autoradiograph ofaSDS-polyacrylamide gel afterelectrophoresis ofpolypeptide labeled duringa

runoff experiment. Conditions for the runoff experimentwereidenticalto those describedfor Fig.4, except

that [3H]leucine wasreplaced by [35S]methionine. Each well received anequal volume of the respective

reaction mixture, with thezero time sample (lane1) containing80,000cpm. (A)8to12S mRNA; (B) HMW

RNA. The numbersatthetopofeachlaneindicate the timeafterATAaddition thatthe labeled amino acid

wasadded. Thefluorographic procedure ofBonner andLaskey (4)wasusedfortheprocessing of these gels.

(polyadenylated leader sequences [PALS]) by

vacciniacoresunderavariety of conditions has

been observed (Paoletti and Lipinskas,

manu-script in preparation). If the distortion of the saturation curve forthe derived 8to 12S RNA

results from the presence ofthese small RNA

species actingasinhibitors oftranslation, then

addition ofa purified fraction of this RNA to

released 8to12S mRNA shouldgeneratea curve

similartothatobserved inFig.7C. As shown in Fig. 7D, E, and F, addition of increasingamounts ofPALS did indeed result inanalteration of the

saturation curve for 8 to 12S RNA to curves

approaching that shown in Fig.7C.

PALS are synthesized in the ATP-depleted

reaction butarereleasedonly after restoration

of the ATPconcentration in the reaction mix-ture, as wasdone for the chase in Fig. 6.Thus,

PALS are nornallypresent in the virion-asso-ciated HMW RNApopulationas well asinthe

population of derived 8 to 12S RNA yet, inex-plicably, high concentrations of HMW RNA do

notshow theinhibition ofprotein synthesis ob-served withhigh concentrations of derived 8to 12S RNA.

Analysis of the translation products synthe-sized inresponseto the derived 8to 12S RNA showed the appearance of no new bands, no

increase in thesynthesis ofasubset of

polypep-tides also synthesized in responseto virion-as-sociated HMW RNA, andno synthesis of

low-molecular-weight protein duetothe translation ofPALS(datanotshown). The translation prod-uct directed by virion-associated RNA and de-rived 8 to 12S RNA were identical and again

similartotheproduct synthesized inresponseto normally released8to 12S mRNA and synthe-sizedinthecoupledsystem.

DISCUSSION

Purified vaccinia virion-associated HMW RNA directs polypeptide synthesis in a wheat

germ cell-free translation system. Translation

Ns_

N_

J. VIROL.

6-.ML. .%-in.

Ga--*.

.w

_X.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:8.505.70.463.72.373.2]

(9)

TRANSLATION OF LARGE TRANSCRIPTS OF VACCINIA

C-'

0 5 10 15 20 25

[image:9.505.63.214.53.263.2]

FRACTION NUMBER

FIG. 6. Sedimentation profiles of normally

re-leased8to 12SmRNA, virion-associated RNA

syn-thesizedatlimiting concentrationsof ATP, and re-leased 8to12S RNA derivedfrom the

virion-associ-ated RNA duringa chase in thepresenceof ATP,

GTP, and AdoMet. Virion-released 8 to 12S RNA (0) waspurified fromavirus-freesupernatantafter

40min of synthesis. Virion-associated HMW RNA (0) wassynthesizedinanATP-depleted reactionas

described(29a) from virus pelleted fromapolymerase reaction.Released8to12S mRNA (a) derived from the virion-associated RNA during a chase in the

presence of ATP, GTP, and AdoMet waspurified

fromavirus-freesupernatantafter60minof chase.

/2HJUTPspecific activity wasmaintainedconstant

inallreactionstomonitor thesynthesis of RNA. productssynthesized inresponsetoHMWRNA

are qualitatively identical and quantitatively

similar to the translation products synthesized in responseto virion-released 8to 12S mRNA. Efficient initiation of HMW RNA translation requires that these largetranscripts be capped at the 5' terminus. Initiation of translation at internal initiation sitesonthe HMWRNA

mol-ecules seems not to occur. Finally, only one

codingsequenceoneachHMW RNAmolecule

istranslated, and thatsequence isprobably

lo-catedverynearthe 5'-end of themolecule.

Aninterpretationof theseresults intermsof

theorigin andbiological significance of the

vac-cinia HMW RNA is somewhat limited by the

sheer complexity of this large DNA virus. The size of the vacciniagenomeis estimated at123

X 106daltons (10). Basedonthecalculation that

vaccinia mRNA hasan averagelength of 1,100

nucleotides (5) and the assumption that only

non-complementary regions of the DNA are

transcribed, there isaminiimal potentialwithin

the vaccinia genome for the expression of

be-.2 A D

100 /

o50 ** r

50

Q,o 20 E0o lo 2

0

w

10 _E_

U 0

0 20 40 0 2

RNA (,ug/ml)

FIG. 7. Stimulation ofpolypeptide synthesis in a wheatgerm cell-free translation system in response to normallyreleased 8 to 12S mRNA, virion-associ-atedRNA made at limiting concentration of ATP, and released 8 to 12SRNA derived from the virion-associated RNA during a chase in the presence of ATP, GTP, and AdoMet. Translation of each species of RNAsynthesized as described in the legend to Fig. 6 was performed asdescribed for Fig. 1. (A) normally released 8 to 12S mRNA; (B) virion-associated RNA made at limiting ATPconcentrations; (C) released 8 to 12S RNA derived from virion-associated RNA during a chase in the presence of ATP, GTP, and AdoMet. (D), (E), and (F) show the effect of adding increasing concentrations ofsmallpolyadenylated and cappedRNAs (PALS) on the translation of nor-mallyreleased 8 to 12S mRNA. The final concentra-tions of PALS were 0, 5, and 10 pg/ml, respectively, for (D), (E), and (F). PALS were synthesized in the presence of2Opgof actinomycin Dper ml andpurified byphenol extraction and selection on poly(U)-Seph-arose. Details of thesynthesis and characterization of PALS will be presented in afuture communication.

tween 160 and180 differentgene products.

Al-though transcriptional complexity studies

indi-cate that essentially all of the viral genome is

transcribedinvitro (6, 28),abundancy

measure-ments do suggestthat asmall proportion of the

genome is represented bythe majority of the in

vitro-synthesized RNA (6). In this regard, the

translationproductssynthesizedin the coupled

transcription-translation system programmed

byvaccinia virusparticles (7, 30, 31) or in

cell-free systems programmed by purified 8 to 12S mRNA (7) (Fig.2)havebeenresolved into 30 to

40 discrete polypeptides on one-dimensional

polyacrylamidegels and approximately 60

poly-peptidesontwo-dimensional gels.

Considerable evidence hasaccumulatedwhich 913

on November 10, 2019 by guest

http://jvi.asm.org/

[image:9.505.277.429.59.263.2]

(10)

isconsistent with aprecursor-product relation-ship between the vaccinia virion-associated HMW RNA and the virion-released 8 to 12S mRNA(25-29a). Nevertheless,thereareseveral observations which are, at present, difficult to reconcile with thishypothesis.Forexample, only

transcripts having5'-di-ortriphosphate termini,

butnot5'-hydroxylor-monophosphate termini,

whichwould beproducedbycleavageofaHMW RNA precursor, are capped by the virion-asso-ciatedguanylytransferase(2, 9, 22,23). Although the elucidation of the mechanismby which caps are formed allows the elimination of several possible pathways which could be envisioned for processing ofalarge transcript, several alterna-tive pathways, e.g., a mechanism analogous to

gene splicing inadenovirus (3, 8, 17) orsimian

virus 40 (1) have not yet beenruledout. Asecond observation which isdifficultto un-derstand in terms of a processing mechanism comesfromrecent studies whichexamined the

effect of UV irradiation on the expression of

vaccinia gene products in a coupled transcrip-tion-translation system (7, 30). The results of these studies indicate that eachtranslationally

functionalmRNA producedinvitroby vaccinia

cores is synthesized from individual promotor sites and isnotderived fromlarge polycistronic

transcripts. In this case, it is conceivable that

thepolypeptidessynthesized in thecoupled

sys-tem arecodedforbyRNA sequenceslocated at ornearthe 5' terminus ofafewlarge transcrip-tionalunitsand that sequencesdownsteamfrom the 5' end of these large transcripts are not

expressed because of a lesion in the normal

processingmechanism under in vitro conditions.

Consistentwiththispossibilityarethe data

pre-sented inFig.2, 4,and5which suggest thatonly onesequence perHMW RNAmoleculeis

trans-lated, that this sequence is probably locatedat

ornearthe5' end of the HMW RNAmolecule, and that the sequences expressed in response to HMW RNA are the same as those expressed both in the cell-free translation system in re-sponse to 8 to 12S mRNA and in the coupled system.However, ifoneconsiders the possibility that.virion-released 8 to 12S mRNA isderived fromasetofpolycistronic HMW RNAs by an in vitro processing mechanism in which only the sequences derived from the 5' end of the large transcript are translationally functional, whereas the processed sequences derived from internal regions of the HMW RNA precursor

are translationally nonfunctional, then one

wouldpredict that the translational activity of the HMWRNA and the derived 8 to 12S RNA would be identical on anequal weight basis. As shown inFig. 1 and in Table 1, this is not the

case; 8to 12S mRNA is approximately three- to fourfold more active instimulating protein syn-thesis in the cell-freesystem than acomparable amount of HMW RNA. Thus, there is no evi-dence for the buildup oftranslationally inactive RNA sequence as a result offaulty processing. Furthermore, if only the 5'-terminal sequence of apolycistronic HMW RNA precursorwere proc-essed in vitro to yield a translationally active molecule, but several translationally active mRNA moleculeswerederivedfrom each HMW RNA precursor in the infectedcell, where proc-essing would proceed normally, then there should be major differences in the

polyacryl-amide gel patterns of the translation products synthesized in a cell-free system in response to eitherpolysomal mRNA from vaccinia-infected cells or in response to invitro-synthesized8 to 12SmRNA. Infact,polyacrylamide gelpatterns of the translationproducts directedbyeither in vitro-synthesized vaccinia mRNA or RNA iso-lated frompolysomes ofvaccinia-infected cells are quite similar (J. Cooper and B. Moss, per-sonalcommunication).

Perhaps each HMW RNA is the precursor of only one 8 to 12S mRNA molecule and the remainder of the HMW RNA sequences are eitherdegraded orremain virion associated. In

thisregard, the results ofexperimentsin which

HMW RNA was preferentially (27) or

exclu-sively (29a)labeled and thencleavedtoRNA in

the8 to12S size range suggeststhat themajority

of this labeled HMWRNA is conserved during

cleavage;i.e.,there isnoevidencesuggesting the

extensivedegradationofaportionof the HMW

RNA during the cleavage step. Indeed,

tran-scriptional complexity studies have indicated

thatthesamesequencesarerepresented in both

the virion-released 8 to 12S mRNA and the

virion-associated HMW RNA (28).

Given the difference in efficiencywith which

equal weights of released 8 to 12S RNA and

purified virion-associatedHMW RNA direct in

vitro proteinsynthesis (Fig. 1), a determination

of whether HMW RNA isprocessed into func-tional messenger RNA should be possible by comparing the efficiency of translation of de-rived8 to12S RNA withvirion-associated HMW RNA from which it was derived and normally released8 to 12S mRNA. However, this

experi-ment is plagued by two problems. First, the

virion-associated RNA synthesized under

limit-ing ATP concentrations is contaminated with some 8 to12S RNAwhich is released upon chase in the presence of ATP and thusbecomes part of the population of derived 8 to 12S RNA. While this virion-associated 8 to 12S RNA makes uponlyafraction of the totalbulk ofthe

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

(11)

915

virion-associated RNA, it contributes more to

the translational efficiency of the total RNA

than toitsbulkbyafactor of3 to 4.Thus, the

differenceinefficiency of the normally released

8 to12S mRNAand the unpurifiedtotal

virion-associated RNAwill be on the order of 1.5- to

2.5-fold rather than 3- to 4-fold. Second, the

presenceofinhibitory PALS released under the

chase conditions (Paoletti and Lipinskas,

man-uscriptinpreparation) addsanadditional

com-plication to the interpretation of these data.

However, based on the level of incorporation

observed atlow concentrations ofderived 8 to

12S RNA, it wouldseemthat thetranslational

efficiency of the virion-associatedRNAdoes

in-creaseduring the chase. This result is consistent

with theproposal that the presynthesized HMW

RNA isprocessedquantitatively within the

vir-iontofunctional mRNA.

An alternative explanation for the origin of

the HMW RNA is that this RNAresults froma

lesion in the termination oftranscription, i.e.,a

read-through mechanism. It is possible that

faulty termination, which mightoccur atlimiting

ATP concentrationsinvitro (19,20,29a), could

result inthe synthesis of large transcripts with

properties similar to those described for the

vaccinia virion-associated HMW RNA. Such

large transcriptswould be cappedat the 5' end

(25). The ratio of molecules having either

m7G(5')ppp(5')Gm

or

m7G(5')ppp(5')Am

termini

would bethe same for the largetranscripts

re-sulting from read through and for the correctly

terminated8 to 12S mRNA (25). The 8 to 12S

mRNA and the

large

read-through transcripts

would have identicalsequencecomplexity (28).

Translation products synthesized in cell-free

translation systems inresponse tothe

large

tran-scripts would be identical to the translation

products synthesized in response to 8 to 12S

mRNA (Fig. 2), and the sequences that are

translatedonthelarge transcripts would be

lo-catedatthe 5' endofthose molecules(Fig.4and

5).

Finally,

themechanismofcapformation and

theobserved effectof UVirradiationonvaccinia

gene expression in the coupled

transcription-translation systems arenotinconsistent with a

read-through mechanism. At the same time, a

read-through mechanism for the generation of

HMW RNA does not

explain

the consistent

observation thatprelabeled HMWRNA can be

cleavedwithin thevirion core, underappropriate

conditions,toRNA in the size range of 8 to 12S

(27, 29a) which appears to be

translationally

functional (Fig. 7). Also,the detection ofa vac-cinia virion-associated endoribonucleaseactivity

(29) whichacts

specifically

on the HMW RNA but notthe8 to 12S mRNAsupportsa

process-ing mechanism rather than a read-through

mechanism. Additional studies arerequired to

determine whether the vaccinia virion-associ-atedHMWtranscriptsresult from faulty termi-nationof transcription or, a more exciting pros-pect,theyrepresent precursortranscripts to

ma-turevaccinia mRNA.

Independent of the mechanismbywhich

vac-cinia HMW RNA isgenerated, it is likely that

these large transcripts contain more than one

gene sequence

and,

consequently,

morethanone

potential ribosome binding sitepermolecule. It

isinteresting, therefore, that translation of these

large transcripts is initiated only through a

5'-terminal cap-mediated step and that initiation

at internal sites does not occur. These large

transcripts shouldproveuseful for studies aimed

at understanding the molecular basis of

site-specific recognition of mRNA by eucaryotic ri-bosomes.

ACKNOWLEDGMENTS

We aregratefultoBernard Lipinskasand Andrew Peterson forexcellenttechnical assistance.

This work wassupported in partby Public Health Service grant1R01 GM23853-01 from the National Institute of Gen-eralMedical Sciences. W.B.was arecipientofaSwiss National Science Foundationpostdoctorialfellowship.

LITERATURE CITED

1. Aloni, Y., R.Dhar, 0. Laub, M. Horowitz, and G. Khoury. 1977. Novel mechanism for RNA maturation: theleader sequences of simian virus40mRNAare not transcribed adjacent to the coding sequences. Proc. Natl. Acad. Sci. U.S.A. 74:3686-3690.

2. Ball, L. A., and C. N. White.1976.Order of transcription of genes of vesicular stomatitis virus. Proc. Natl. Acad. Sci. U.S.A. 73:442-446.

3. Berget, S.M., C. Moore, and P. A. Sharp. 1977.Spliced

segmentsatthe 5'-terminus of adenovirus2late mRNA. Proc.Natl. Acad. Sci. U.S.A. 74:3171-3175.

4. Bonner, W. M., and R. A.Laskey.1974. Afilmdetection method fortritium-labeledproteins and nucleic acids in

polyacrylamide gels.Eur. J. Biochem.46:83-88. 5. Boone, R.F., and B. Moss.1977.Methylated 5'-terminal

sequences ofvaccinia virus mRNA species made invivo at early and late times after infection. Virology 79:67-80.

6. Boone,R.F., and B. Moss.1978.Sequencecomplexity andrelative abundance of vacciniavirusmnRNA's syn-thesized in vivo and in vitro. J. Virol. 26:554-569. 7.Bossart,W.,D.L.Nuss,and E. Paoletti.1978.Effect

of UV irradiationon theexpressionofvaccinia virus geneproductssynthesizedinacell-free systemcoupling

transcription and translation. J. Virol. 26:673-680. 8. Chow,L.T.,R.E.Gelinas,T. R.Broker, and R. J.

Roberts. 1977. An amazing sequence arrangementat the 5'-ends of adenovirus 2 messenger RNA. Cell 12:1-8.

9. Ensinger, M. J., S.A. Martin, E. Paoletti, and B. Moss. 1975. Modification of the5'-terminusof mRNA by solubleguanylylandmethyltransferasesfrom vac-ciniavirus.Proc.Natl.Acad.Sci.U.S.A. 72:2525-2529. 10.Geshelin, P., and K. L. Berns. 1974. Characterization andlocalization ofthenaturallyoccurring cross-linksin vaccinia virus DNA. J.Mol.Biol. 88:785-796.

VOL. 28,1978

on November 10, 2019 by guest

http://jvi.asm.org/

(12)

11. Hickey, E. D., L. A. Weber, and C.Baglioni. 1976. Inhibition of initiation of protein synthesis by 7-meth-ylguanosine-5'-monophosphate. Proc. Natl. Acad. Sci. U.S.A. 73:19-23.

12. Hickey, E. D., L. A. Weber, and C.Baglioni. 1978. Nuclease activity in preparations of creatine phospho-kinase: effect on mRNA stability. Biochem. Biophys. Res. Commun. 80:377-383.

13. Joklik, W. K. 1962. The preparation and characteristics of highly purified radioactivity labeled poxvirus, Biochim. Biophys. Acta 61:290-301.

14. Kaempfer, R., H. Rosen, and R. Israeli. 1978. Trans-lational control: recognition of the methylated 5' end andaninternal sequence in eukaryotic mRNA by the initiation factor that bindsmethionyl-tRNAf"e'. Proc. Natl. Acad. Sci. U.S.A. 75:650-654.

15. Kates, J., and J. Beeson. 1970.Ribonucleic acid synthe-sis invaccinia virus. I. The mechanism of synthesis and release of RNA in vaccinia cores. J. Mol. Biol. 50:1-18. 16.Kates, J.,and J. Beeson.1970.Ribonucleic acid synthe-sis invaccinia virus. II. Synthesis of polyriboadenylic acid. J.Mol. Biol. 50:19-33.

17. Klessig, D. F. 1977. Two adenovirus mRNAs have a common5'-terminalleader sequence encoded at least 10Kb upstream from their main coding regions. Cell 12:9-21.

18. Lodish, H.F. 1976. Translational control of protein syn-thesis. Annu. Rev. Biochem. 45:39-72.

19.Lowery,C., and J. P. Richardson. 1977. Characteriza-tionof the nucleosidetriphosphate phosphohydrolase (ATPase) activity of RNA synthesis termination factor P.I. Enzymatic properties and effects of inhibitors. J. Biol. Chem. 252:1375-1380.

20. Lowery,C., and J. P. Richardson. 1977. Characteriza-tion ofthe nucleoside triphosphate phosphohydrolase (ATPase). Activity of RNA synthesis termination factor P.II.Influence of synthetic RNA homopolymersand random copolymers on the reaction. J. Biol. Chem. 252:1381-1385.

21. Mans, J. R.,and G. D. Novelli. 1961. Measurement of theincorporation of radioactive amino acids into pro-tein by a filter-paper disc method. Arch. Biochem. Biophys.,94:48-53.

22. Martin,S. A., and B. Moss. 1975. Modification of RNA by mRNA guanylyltransferase and mRNA (guanine-7-) methyltransferase from vaccinia virions. J. Biol. Chem. 250:9322-9329.

23. Moss, B., A. Gershowitz, C. M. Wei, and R. Boone. 1976.Formation ofthe guanylylated and methylated 5'-terminus ofvaccinia virus mRNA. Virology 72:341-351. 24. Nevins, J. R., and W. K. Joklik. 1975. Poly(A)

se-quences ofvacciniavirusmessenger RNA: nature, mode ofaddition and functionduring translation in vitro and in vivo.Virology 63:1-14.

25. Nuss, D.L., and E.Paoletti.1977.Methylgroupanalysis

ofvirion-associatedhigh-molecular-weight RNA syn-thesized in vitro by purified vaccinia virus. J. Virol. 23:110-116.

26. Paoletti,E.1977.Invitrosynthesisofa

high-molecular-weight virion-associated RNA by vaccinia. J. Biol. Chem. 252:866-871.

27. Paoletti, E. 1977.The high-molecular-weight virion-as-sociated RNA of vaccinia:apossibleprecursorto8-12S mRNA. J. Biol. Chem. 252:872-877.

28. Paoletti, E., and L. J.Grady. 1977. Transcriptional

complexity of vaccinia virus in vivo and in vitro. J. Virol. 23:608-615.

29. Paoletti, E.,and B. R.Lipinskas. 1978.Soluble endo-ribonucleaseactivity from vaccinia virus: specific cleav-ageofvirion-associatedhigh-molecular-weightRNA. J. Virol. 26:822-824.

29a.Paoletti, E.,and B. R.Lipinskas.1978.Therole of ATP in the biogenesis of vaccinia virus mRNA in vitro. Virology 87:317-325.

30. Pelham, H. R. B.1977.Use ofcoupled transcriptionand translation to study mRNA production by vaccinia cores.Nature(London) 269:532-534.

31. Pelham,H. R.B., J. M. M.Sykes,and T.Hunt.1978. Characteristics ofacoupledcell-freetranscriptionand translation system directed by vacciniacores.Eur.J. Biochem. 82:199-209.

32. Roberts, B. E., and B. M. Paterson. 1973. Efficient translation of tobacco mosaic virus RNA and rabbit globin 9S RNA inacell-free system from commercial wheatgerm.Proc.Natl.Acad.Sci. U.S.A. 70:2330-2334. 33. Shafritz,D.A.,J.A.Weinstein,B.Safer,W.Merrick,

L.A.Weber,E. D.Hickey,andC.Baglioni.1976. Evidence for the role ofm7G-phosphategroupin rec-ognition ofeukaryoticmRNAbyinitiation factor IF-M3. Nature(London)261:291-294.

34. Weber,L.A.,E.D.Hickey,D.L.Nuss,andC.

Bag-Honi.1977.5'-terminal7-methylguanosineand mRNA function: influence ofpotassiumconcentrationon trans-lation in vitro. Proc. Natl. Acad. Sci. U.S.A. 74:3254-3258.

35. Wei, C. M., and B. Moss. 1974.Methylationofnewly synthesized viral messenger RNA by an enzyme in vaccinia virus. Proc. Natl. Acad. Sci. U.S.A. 71:3014-3018.

36. Wei,C. M., and B. Moss.1975.Methylatednucleotides block 5'-terminus of vaccinia virus messenger RNA. Proc.Natl.Acad. Sci. U.S.A.72:318-322.

J. VIROL.