Encephalomyocarditis virus RNA. II. Polyadenylic acid requirement for efficient translation.

(1)

Encephalomyocarditis

Virus

RNA

II. Polyadenylic Acid Requirement

for

Efficient Translation

DENNIS E. HRUBY* AND WALDEN K. ROBERTS

Department of Microbiology and Immunology, University ofColorado Medical Center, Denver,

Colorado 80262

Receivedforpublication 7 March 1977

Differentiallypolyadenylated subpopulations ofencephalomyocarditis (EMC) viral RNA were isolated by affinity chromatography on oligodeoxythymidylic acid-cellulose. Translation of these RNA fractions in several in vitro protein-synthesizingsystems, isolated from Ehrlich ascites tumorcells, demonstrated thatpoly(A)+ EMC viral RNA wastranslatedtwo to three times moreefficiently than poly(A)- EMC viral RNA. Sodium dodecyl sulfate-polyacrylamide gel electrophoretic analysis ofthepolypeptidessynthesized by theinvitrosystemin response to the different RNAs showednodetectable differences in the size or relative amounts of the translational products. mRNA saturation curves indi-cated that the in vitro systems were stimulated maximally by equivalent amounts ofRNA, whether it be poly(A)- or poly(A)+ EMC viral RNA. Time course experiments showed that the differences in translatability were more pronounced late in the reaction when reinitiation was required, and that by eliminating reinitiation with high salt theapparenteffect ofpoly(A)on transla-tion wasdiminished. Together, these results suggest that poly(A) may be re-quired for efficient initiation and reinitiation ofprotein synthesis in the cell-freesystems. This interpretation is discussed relative to earlier data.

Many investigators have soughttoestablish acytoplasmic functions) forpolyadenylic acid [poly(A)], which is found covalently linked to the3'-terminusofmostRNAmolecules capable ofbeing translated byeucaryotic cells (4). Much ofthis work has been carried out in various eucaryotic cell-free protein-synthesizing sys-temsutilizing de-adenylatedmRNA as aprobe. These experiments have failed to discern any stringent requirement for

poly(A)

inthe

trans-lation ofmRNA (2, 15, 27). This may, in part, be due to the

inefficiency

of the in vitro sys-tems being used or the conditions used for translation. Huez et al. (10) have utilized the Xenopus oocyte system,which has beenshown tocarry out translation of exogenously injected mRNA for long periods oftime, to show that poly(A) is important in maintaining the func-tional stability of globin mRNA. Also, recent careful analysis of ovalbumin message transla-tionhasshown that poly(A) facilitates the reu-tilization of this mRNA incell-free systems (6). Whether these effects are due to a protection of polyadenylated mRNA from nuclease degrada-tion, assuggestedby Hieter et al. (8), or rather a moredirect roleinproteinsynthesisis notyet clear.

Wehave been interested in determining the role(s), ifany, of poly(A) in the replication of encephalomyocarditis (EMC) virus. To this

end, differentially polyadenylated fractions of EMC viral RNA have been isolatedand trans-lated in cell-freeextracts from Ehrlich ascites tumor cells. This approach has the advantage ofutilizing naturally occurring populations of homogeneous mRNAinwhich the only appar-ent differenceisthe lengthof the poly(A) moi-ety and translating these RNAs in a cell-free extract derived from cells that translate this message in vivo. The resultsreported here indi-catethat the presence of apoly(A)region stim-ulates the overall translation of EMC viral RNA two- tothreefold. Thisis inapparent con-trast toearlierwork onthe translation of polio-virus RNA (25)but is explicable dueto differ-ences inexperimental methods.

MATERIALS AND METHODS

Materials. [8-3H]adenosine (33 Ci/mmol),

[5-3H]uridine (28 Ci/mmol), and a 14C-labeled amino

acid mixture (algal profile) were obtained from

Schwarz/Mann; [32P]orthophosphate and

[35S]-methionine (525 Ci/mmol) were from Amersham/

Searle; oligodeoxythymidylic acid

[oligo(dT)]cellu-lose, T-3 grade, was from Collaborative Research,

Inc.; polyacrylamide gel electrophoresis materials

wereofelectrophoresis purity,purchasedfrom

Bio-Rad; Seakem agarose wasfromBauschand Lomb;

X-ray film and photochemicals were bought from

Eastman Kodak.

Virus growth. EMC viruswasgrown in

suspen-338

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

sions of Ehrlichascitestumorcells andpurifiedas

previously described (9). The viral RNAwas

radio-actively labeled andextractedasbefore (9).

RNA analysis. Oligo(dT)-cellulose

chromatogra-phywasperformedessentiallyasdescribed by Aviv

and Leder(1), with certainmodifications(9).Intact

35S EMC viralRNAwasisolatedby centrifugation

through 11.5mlof5 to 20% sodiumdodecyl sulfate

(SDS)-sucrose gradients that were spun at 35,000

rpmfor4.75hat240CinaBeckmanSW41rotor(9).

Gel electrophoresis. (i) Agarose-acrylamide gels

(for RNA). Composite gelswere runaccordingtothe

methods ofLoening (14). Agarose(0.5%)-acrylamide

(1.8%)gelswerepolymerizedfor1h inacid-washed

glass tubes (15by 0.8 cm). Thegelsweresubjected

toelectrophoresis at5.5mA/gelfor10minin0.4 M

Tris-hydrochloride, 1 mM EDTA, 0.02 M sodium

acetate, 0.1% SDS (pH 5.5). RNA samples to be

analyzedweredissolvedin 251u of1mMEDTA,10

mM Tris-hydrochloride (pH 7.5), 0.2% SDS and

thendiluted1:1with50% sucrose-0.5%bromophenol

blue. Sampleswere heated at 850C for 2 min and

then applied immediatelytothe gel. Electrophoresis

was at 5.5 mA/gel for 3 to 5 h in running buffer.

After electrophoresis, gels were fractionated on a

Gilson gel slicerinto 2-mmslices. Sliceswere

incu-bated in0.5ml ofconcentrated ammonia overnight

andthencounted in10ml of Bray solution(5).

(ii) Polyacrylamide gel electrophoresis [for

poly(A)]. Polyacrylamide gels weresetupand run

accordingtoPeacock and Dingman(17). RNA

sam-plesweresubjectedtoelectrophoresison3-mm10%

polyacrylamide slab gelsat200Vat40C for4h inpH

8.3buffer. After electrophoresis, the gelwasfixedin

1 Macetic acid, stained with methylene blue, and

destained with water to locate nonradioactive

markers. The gelwasthenimpregnated with PPO

(2,5-diphenyloxazole) (3), driedundervacuum,and

putonKodakRoyal X-Omatfilm at-70°C for2to4

weeks. Theresulting autoradiogramwastraced

us-ingtheOrtec model 4310densitometer.

(iii) SDS-polyacrylamide gel electrophoresis (for

protein). Ten-microliter samples of cell-free protein

synthesis reactions were acetone-precipitated,

dried, dissolved in25

Al

of sample buffer, and boiled

for 5min. Sampleswerethenputon a10%, 1-mm,

0.1% SDS-polyacrylamide slab gel and subjectedto

electrophoresis at 100 V for 6 h (12). The gel was

thenstainedwith Coomassie blue anddestained in

methanol-acetic acid (5:7.5%,respectively). The gel

was then dried under vacuum and put on Kodak

Blue Brand film for 1to2weeks.

Cell-freeprotein synthesis. (i) Fractionated

sys-tem. An mRNA-dependentin vitroprotein

synthe-sizing system, consisting ofpurified ribosomes and

anammonium sulfatefraction, wasprepared from

Ehrlich ascites tumorcellsas previouslydescribed

by Sharma et al. (22). Reactions contained, in a

volume of 100 ,l: 20 mM Tris-hydrochloride (pH

7.5), 1mMATP, 0.2mMGTP,6mMcreatine

phos-phate (all adjusted to pH 7.5), 20 Mg of creatine

kinase,0.5 MiCiofa "C-labeledaminoacidmixture,

0.05mMeach of thesixnonradioactiveaminoacids

missing from the "4C-labeledaminoacidmixture, 6

mM /3-mercaptoethanol, 5 mM Mg(Ac)2, 120 mM

KCl,0.07opticaldensity(at260 nm)unitof

ammo-nium sulfate fraction, and 0.33 optical density (at

260 nm) unit ofpurified ribosomes. Followingthe

reaction, 0.2 mlof 0.1 N KOH was added, and the

tubes were incubated at370C for 20 min. A 1.0-ml

amountof cold 10% trichloroaceticacidwasadded,

and theprecipitate wascollectedonWhatmanGF/A

glassfiber filters.

(ii) S10-1 system (endogenous protein synthesis

reduced by preincubationof the cell-free extract).

Ehrlichascites cells from8-day tumors were washed

withphosphate-buffered salineto remove

erythro-cytes (20) andsuspendedto 5% inFischer medium

for leukemic cellsof mice (GrandIsland Biological

Co.),and the suspensionwasrotatedgentlyat370C

for 1 h to restore proteinsynthesis.The ascites cells

were then removed by centrifugation and used to

prepare an S10protein-synthesizing system, using

themethodpreviously described for L-cells (11).

Re-actions werecarriedout in atotal volume of 25 ul,

whichcontained:2.5 pulof S10 extract, 0.125

/Ci

of

the amino acid mixtureplus the six nonradioactive

amino acids to 0.05 mM, 108 mM KCI, 5 mM

Mg(Ac)2, 20 mM HEPES(pH 7.6),7.2mM

,-mercap-toethanol, 4

,Ag

of creatine phosphokinase, 10mM

ATP, 1 mM GTP,6 mM CTP, and14mM creatine

phosphate. Following the reaction, 10-,ul samples

from each tube were spotted on Whatman 3 MM

filter squares andplunged into cold5%

trichloroace-ticacid. Filters were heated to 900C for 15 min and

washed twice with cold 5% trichloroacetic acid,

twice with cold ethanol, and twice with acetone.

Filters were dried and counted byliquid

scintilla-tion.

(iii) S10-2 system (endogenousproteinsynthesis

reduced by cellstarvation). We have found that by

incubating ascitescellsinnutrient-free mediumwe

canobtain an active S10 protein-synthesizing

sys-temwith low endogenous amino acid incorporation

without the usualpreincubation of the S10

superna-tant fraction. Ehrlich ascitescells were washed as

above andsuspendedto 3%in146mMNaCl-35 mM

Tris-hydrochloride (pH 7.5). The suspensionwas

ro-tated for 90 min at370C,and the cellswerecollected

by centrifugation. An S10 systemwas preparedas

above except that thepreincubation of theextract

was omitted. Assays were performed as with the

S10-1 system.

(iv) S10-3 system (endogenousproteinsynthesis

reduced bypreincubation withstaphylococcal

nu-clease). The S10-3 system wasprepared usingthe

S10-1 procedure, exceptthatinstead of

preincubat-ing the S10 extract at370C for45min, the extract

was preincubated briefly with staphylococcal

nu-cleaseaccording to the method of Pelham and

Jack-son(18).The S10-3 system wasassayedasabove and

showed low background incorporation and good

stimulation with EMC viral RNA.

RESULTS

Oligo(dT)-cellulose

chromatography.

We have previously shownthataffinity chromatog-raphyof EMC viral RNA onoligo(dT)-cellulose columns separates the RNA into three peak fractions, which differin bothpoly(A) content andbiologicalactivity (9). These threefractions 23, 1977

on November 10, 2019 by guest

http://jvi.asm.org/

(3)

weredenotedaspeak1[poly(A)-] RNA, peak2

[oligo(A)] RNA, and peak 3 [poly(A)+] RNA andwereshowntocontainRNA molecules with

an averageof 16, 25, and 74nucleotides,

respec-tively, in their poly(A) tracts (9). In this

com-munication we will continue to use peaks 1

through3 asthedesignation for these

differen-tially polyadenylated subpopulations of EMC RNA.

The EMC viral RNA is notdegraded during fractionation on the oligo(dT)-cellulose

col-umns. The RNAs from peaks 1 through 3

mi-grate as a single band in agarose-acrylamide

gels (Fig. 1), thus corroborating the earlier SDS-sucrose analysis, which showed that the RNA from these fractions consisted of unde-gradedgenomic-length EMC RNA(9).

For this study we wanted to determine

whether peak 1EMC RNA contains auniform

population of RNAs with short or missing

poly(A) moieties,orwhether this RNA consists

ofpoly(A)- RNA together withaminor

contam-ination of poly(A)+ RNA sufficient to account

for the averagepoly(A) tractsizeof 16 nucleo-tides. Accordingly, [3H]adenosine-labeled peak

1 and3 EMC viral RNAswere treated witha

combination of RNases A and T1, and the

re-sistant fraction was isolated and run on 10%

polyacrylamide gel electrophoresis. Peak 3 EMC viral RNA poly(A)(Fig. 2B)ran as a

rela-tively homogeneous peak, with a mobility

slightly faster than that of 4S RNA. This is similartotheelectrophoretic mobility of polio-viruspoly(A) regions(24). Virtuallynopoly(A)

sequences weredetected in RNase-treated peak

1 EMC viral RNA (Fig. 2A). This eliminates thepossibility that peak1EMCviral RNA

con-mx0

DISTANCE(mm)

FIG. 1. Agarose-acrylamide gel electrophoresis of EMC viral RNA. Approximately100,000cpmofpeak

332P-labeled EMC viral RNA wassubjectedto elec-trophoresis through 0.5% agarose-1.8% acrylamide composite gels. Gels werefractionated, and the

ra-dioactivity ineach slicewasdetermined. The position

of the markerswassimilarly determinedon a

paral-lelgel using rRNA from Ehrlichascitestumorcells. Identicalpatterns were observed withpeak 1 and2

EMCviral RNA.

tainsafew longpoly(A) regions, whichaccount

for its residual infectivity, but does not prove

that this RNA is truly poly(A)-. One of the

stepsusedtoisolate the poly(A) regions is

bind-ingtobenzoylated cellulose (21), and it is possi-ble thatveryshortoligo(A)tractsfrom thepeak

1RNA donotbindtobenzoylated celluloseand

escape detection by this procedure.

Translation of oligo(dT)-cellulose-fraction-ated EMC viral RNA. Unlabeled EMC viral RNAwasfractionatedonoligo(dT)-celluloseas

usual. Peaks 1 through 3 were then run on

SDS-sucrose gradients, andtheRNA in the35S peakwasisolatedby ethanolprecipitation. The

three RNA fractions wereassayed by in vitro

protein-synthesizing systems prepared from Ehrlich ascitestumorcellsusing several

differ-entmethods. These different systemswere

uti-lized to investigate the possible effects of changes inpreincubation conditions, ions, and relative concentrations of translational

compo-nents on the translation of the EMC viral

RNAs.Reactionswerecarriedoutatsaturating

RNAconcentrations: 3.0 ,ug/assay for the frac-tionated system; 0.3 ,ug/assay for each ofthe

S10 systems. Regardless of the method of

sys-tempreincubation, reaction temperature, ionic conditions (Mg2+-spermidine), or presence of

hemin, peak 3 EMC viral RNA always stimu-latedtwo- tothreefoldmore radioactive amino

acid incorporation than peak 1 EMC viral RNA. Peak2EMCviral RNAwasusually 90to

E

c

0

0 v) LU u

z

0c

co

.5 .4 .3 .2

5S 4S a

.+ + a

I

20 4'0 6o 80 +

[image:3.499.290.437.405.560.2]

DISTANCE (mm)

FIG. 2. Gel electrophoresis of EMC viral RNA poly(A) regions. [3H]adenosine-labeled poly(A)

re-gions were isolated from peak1 and 3 EMC viral

RNAby RNasetreatment,followed by chromatogra-phy on benzoylated cellulose (21). The poly(A)

re-gionswerethensubjectedtoelectrophoresison a10%

polyacrylamide gel along with 4S and 5S RNA markersfrom Ehrlich ascitestumorcells. Above, the densitometertracings of the autoradiogram are

dis-played. (a) Peak1EMC viral RNA; (b) peak 3 EMC

viralRNA.

48 28S 18S

4 4

36-

24-12

20 40 60 80 1

5 .4 .3 2 .I

IV 1.6

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.499.63.253.459.566.2]

(4)

POLY(A) REQUIREMENT FOR EMC RNA TRANSLATION 341

100% as effective as peak 3 EMC viral RNA.

The data in Table 1 give the results ofseveral representative experiments. Anumber of other experiments were performed in which the

aforementioned variables were modulated,

in-cludingtranslation in the S10-2 and S10-3

sys-tems, and although the overall level of protein

synthesis was affected, the relative

translata-bilities of peak 1, 2, and 3 EMC viral RNA remained constant. This effect has been ob-served withEMC viral RNAs isolated from five

separate viruspreparations.

The nature of the polypeptides synthesized

by the S10 in vitrosysteminresponsetopeak 1, 2, and 3 EMC viral RNAwas investigated by

subjecting the 14C-labeled proteins to electro-phoresison a10%SDS-polyacrylamide gel. The

resultsof thisareshown in Fig.3.Little, ifany,

difference in the relative amounts or size of

products was detectable. This is in agreement

with earlier studies on the translation of

de-adenylated polioviral RNA (25).

Concentrations of RNA required for opti-mal translation. EMC viral RNA saturation experimentswerecarriedout toseeifthe

differ-encesintranslatabilities could beovercome by

simply addingmorepeak 1EMCviral RNAto

the cell-free system. As canbe seen in Fig. 4,

this was not the case. Radioactive amino acid

incorporation using 2.5 ,ul of S10 was

stimu-lated maximally by 0.3 ,ug of EMC viral RNA (12 ,ug/ml), regardless of whether itwaspeak1

orpeak 3 EMC viral RNA. Similar experiments werecarriedoutusing5 -lIof S10 (or the frac-tionated system), with identical results, the only difference being in the amount of RNA

requiredto reach saturation (5

Al

of S10 = 40

pg/ml; fractionated system = 30 /Lg/ml).

At this point, it seemed possible that the lowered translatability of peak 1 EMC viral RNA wasdue tocontaminating host cell RNA

species, which alsosedimentedat35S.Tocheck this, EMC viruswasgrowninthepresenceof5

,tg

of actinomycin Dperml, which inhibitshost

cell RNA synthesis, and labeled with [3H]uridine. The RNA was fractionated as

usual and thenrun on SDS-sucrose gradients.

Figure 5 shows that the radioactivity and ab-sorbancyat260nmprofiles of the 35S RNA in

peak 1, 2, and 3 EMC viral RNA were

coinci-dent. Theaveragespecific activities of the peak

fractions were virtually identical: peak 1 =

1,140cpm/gg,peak2= 1,170cpm/gg,and peak

3 = 1,190cpm/,g. Therefore, it appeared that

therewas little or no contamination of peak 1

EMC viral RNA by unlabeled host RNA

spe-cies.

Kinetics of translation. The incorporation of 14C-labeled amino acids into proteins in

re-sponse to peak 1 and 3 EMC viral RNA was

monitored with time (Fig. 6a). Translation of the two RNAswas foundto be similar during

the first 30 min of incubation; however, after this period the translation of peak 1 RNA slowedconsiderably, whereas the translationof

peak 3 RNA continued ata linear rate for an

additional 60 min, resulting in the eventual

two- to threefold difference in the extents of

translation. Similar kinetic patterns for the translation ofpoly(A)- ovalbumin RNA have been obtained by Doel and Carey (6). When

translation of the EMC viral RNAs was

initi-TABLE 1. Invitrotranslation ofoligo(dT)-cellulose-fractionatedEMC viral RNAa

In vitro Reaction conditions EMC viral RNA 14Cincorporated Stimulation

system

aC

Ions (mM) Fraction /19 (cpm) (fold)

Fractionated 37 Mg2+, 5 0 5,100

K+, 120 Peak1 3 46,800 9.2

Peak2 3 128,700 25.4

Peak3 3 133,400 26.3

S10-1 30 Mg2+, 2 0 3,800

Spd, 0.3 Peak1 0.3 14,400 3.9

K+, 108 Peak2 0.3 26,500 7.2

Peak 3 0.3 29,100 7.9

S10-1 30 Mg2+, 5 0 400

K+,108 Peak 1 0.3 10,700 25.7

Peak2 0.3 22.800 54.6

Peak 3 0.3 23,400 55.9

a Oligo(dT)-cellulose-fractionated EMC viral RNAs were furtherpurified by zonal centrifugation and

translatedinseveraldifferentinvitro systemsprepared fromEhrlich ascitescells.Reactions wereallowed

toproceedfor 120 min, andthen14C-labeledaminoacidincorporation wasmeasuredasdescribedinthetext.

VOL. 23, 1977

on November 10, 2019 by guest

http://jvi.asm.org/

(5)

342 HRUBY AND

a

b

c

m

u

57-

45-EMC VIRAL RNA ADDED(Mg)

FIG. 4. RNA saturationcurves. Varyingamounts

ofpeak1 or 3EMC viral RNAwereaddedto25-piin

vitroprotein-synthesizingassays which contained 2.5

pIof S10-2. The reactionswerecarriedoutat5 mM

Mg2+-l08 mM K+ and30'C, for120 min. Symbols:

(0),peak3EMC viralRNA; (a), peak1EMC viral

RNA.

34-

23-FIG. 3. SDS-polyacrylamide gel electrophoresis

analysis of viral polypeptides synthesized in vitro. Peak 1 (a), peak 2 (b), or peak 3 (c) EMC viral RNA

(0.3pg) was added to a 25-

gl

protein synthesis assay

which contained 2.5 pl of S10-1. After synthesis,

approximately 10,000 cpm of each reaction mixture

wasanalyzed by10%SDS-polyacrylamide gel

elec-trophoresis. Above is displayed the autoradiograph of the polypeptides synthesized in response to the added viral RNA. Numbers refer to the molecular

weights of marker proteins (xlO-3). Note that the

largest newly synthesized viral protein corresponds to

anapproximate molecular weight of 140,000.

ated asusual, followedbyshiftingto high-salt reaction conditionsto permit polypeptide elon-gation but prevent further initiation and

reini-tiation events (26), much ofthis difference in

theextentsoftranslation disappeared (Fig. 6b).

DISCUSSION

We have never failed to observe a two- to

threefold difference inefficiency of translation

25-

20-

15-

10-

5-

75-

60-01

C

45-X.

30-I

15-

50-

40-

30-

20-10.

.1.5

.1.0

.0.5

.0

_*t

<

> z

.1.5 > a

.1.0 <

0

CLQ

.1.5

1.0

o.0

5 10 15 20 25

FRACTION NUMBER

FIG. 5. Specific activity of [3H]uridine-labeled

EMC viral RNA oligo(dT)-cellulose fractions.

[3H]uridine-labeled EMC viral RNA was

fraction-atedon oligo(dT)-cellulose. Eachof the three

frac-tions wasthenanalyzed onSDS-sucrosegradients.

Each gradient was analyzed for absorbance at 260

nm

(A26&)

and trichloroaceticacid-precipitable

radio-activity.Specificactivities werecalculatedassuming

1A260unitequals40pgofEMC viral RNA.(a)Peak

1 EMC viral RNA; (b) peak2EMCviralRNA; (c)

peak3EMC viral RNA.

betweenpoly(A)+ andpoly(A)- RNAs isolated from five different preparations of EMCvirus.

This differencewasobservedunderavarietyof ionic and temperature conditions used for

.-o

E .1_-%o_ f~

.8-C4

.2 -

--2-L

__

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.499.265.460.62.223.2] [image:5.499.83.231.65.433.2] [image:5.499.268.455.300.511.2]

(6)

REQUIREMENT FOR EMC RNA TRANSLATION 343

0

x

I 0r

0.6

b)

4

2

0-03

INCUBATION TIME (h) FIG. 6. Timecourse of EMC viral RNA

transla-tion. A 0.6-pgamountof peak1 (a)orpeak3 (0)

EMCviral RNA was addedto a50-p/X S10-1 assay

containing 5 piof S10. At theindicated times,

3-t1

samples were removed and processed to determine

hottrichloroacetic acid-precipitableradioactivity. (a)

Reactionrun at90 mMKCl; (b) reactionstartedat

90 mMKCl. At 15 min the KCl concentration was

shiftedto155 mM.

translation (Table 1). From the mRNA

satura-tioncurvesandspecific activity analysis itwas

evident that this difference could not be

ex-plained by thepresenceofcontaminatingRNA

species in the peak 1EMC viral RNA fraction (Fig.4and 5). Also,electrophoresis under dena-turing conditions showed the RNAstobe unde-graded (Fig. 1).

Thereare atleasttwopossible explanations

forthe reducedtranslatability ofpoly(A)-EMC

viral RNA. First, itmay be that

poly(A)-defi-cient RNAis moresusceptibletonuclease deg-radation (8) and therefore hasalowered

proba-bility of remaining intact and interacting

pro-ductively with the protein synthetic

machin-ery. Inpreliminary experiments,wehave been

unabletodetectanyincreasedsusceptibility of

thepoly(A)- RNAtoexonucleolytic or

endonu-cleolytic attack during the protein synthesis

incubation. However, a more subtle role of the poly(A) moiety, such asaffecting the degrada-tion ormodification of the5'initiation region of the viral RNA, cannotbe ruled out. A second possible explanation for the reduced translata-bility ofthe poly(A)- RNA is that poly(A) di-rectly influences one of the protein synthetic reactionsrequiredforinitiation, elongation, or termination. Since the patternsof polypeptides synthesizedinresponse topeak 1, 2, and 3 EMC viral RNA are identical (Fig. 3), this would indicate that the termination process is not af-fected.Also, thedecreased differentialbetween thetranslationsof poly(A)+and poly(A)-RNAs at early times or under high-salt conditions (Fig. 6) suggests that elongation is notinvolved inthe discrimination between the RNAs (13). It seems most likely that the reduced transla-tional capacityof poly(A)- EMC viral RNA is due to some impairmentwith the initiation or reinitiation process (6). Thisimpairmentcould resultinanincreasedfunctional inactivationof

the

poly(A)-

RNA wheneverit is not engaged

with ribosomesinproteinsynthesis, leadingto an accumulating inhibition of protein synthe-sis, which becomes increasingly apparent late intranslation.

There are several possible reasons why we have been able to demonstrate a reduction in the translational capacity of a poly(A)- mes-sagewhen similar analyses by others(2, 25, 27) have not. One possibility is that the Ehrlich ascites tumor cell-free system carries out pro-tein synthesis for long periods oftime in re-sponse to EMC viral RNA. Incorporation of radioactive amino acids continues for more than2h at 30'C (Fig. 6). This is in contrast to 30 to 60 min of linear incorporation that has beenmonitoredinother studies (2, 15). Huez et al. (10)have been themostsuccessfulin estab-lishinga role forpoly(A) inthe translation of mRNA,and thisisprobably duetothe fact that their studies have been done in the Xenopus oocyte system, which carries out manyrounds of translation.Thisis verydifferent frommany cell-free systems, such as the wheat germ em-bryosystem, which are veryinefficientat reini-tiating new rounds ofsynthesis (19). The Ehr-lichascites systemusedhere continues to reini-tiate newrounds ofsynthesis formorethan90 min (D. E. Hruby, unpublished data), which would facilitate detection ofaneffectofpoly(A) on the accumulating inactivation of mRNA. This is in contrast to themethodsofSpector et al. (25), who allowed initiation of poliovirus RNA toproceedforonly 15 min before shifting toahigher potassiumconcentration, which per-mits elongation but inhibits initiation. Using theirprocedure,we toocandetect little effect of

VOL. 23, 1977

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.499.63.216.53.367.2]

(7)

poly(A) on the translation of viral RNA (Fig. 6b).

Inall of the reactions where it wasassayed, peak 2 EMC viral RNA was nearly as active as peak 3 EMC viral RNA. This was true even if peak2waschromatographed asecond time. It may be, as suggested for globin mRNA (16), that thereis a critical sizeforpoly(A)function. For globin mRNA this was estimated to be 32 nucleotides. This agrees fairly well with our data, aspeak 2 EMC viral RNA has previously been estimated to contain an average of 26 nucleotides in its poly(A) region (9).

Lowered infectivity of poly(A)- picornaviral RNA molecules has previously been observed by us (9) and others (7, 23). It is not known whether this effect of the poly(A) moiety on infectivity is at the level of penetration, protec-tion against nucleases, translation, or tran-scription. The loweredefficiency of translation ofpeak1EMC viral RNAmay, atleastinpart, explain its lowered infectivity. Currently, re-search is in progress to find conditions that potentiate the reducedtranslationalcapacityof peak1EMCviral RNAinordertobetterassess therole ofpoly(A) intranslation and infection.

ACKNOWLEDGMENTS

We thank0. K. Sharma for help in preparing the frac-tionated cell-free system. We also thank J. L. Brown for use ofthe Gilson gel crusher.

This work was aided by grant no NP-200 from the Ameri-canCancer Society.

LITERATURE CITED

1. Aviv, H.,and P. Leder. 1972.Purification ofbiologically activeglobin messenger RNAbychromatographyon

oligothymidylicacid-cellulose. Proc. Natl. Acad. Sci. U.S.A. 69:1408-1412.

2. Bard, E., D. Efron, A. Marcus, and R. P. Perry.1974. Translational capacity of deadenylated messenger RNA. Cell 1:101-106.

3. Bonner, W. M., andR. A.Laskey. 1974. A film detec-tionmethod for tritium-labeled proteinsand nucleic acidsinpolyacrylamide gels. Eur.J.Biochem. 46:83-88.

4. Brawerman, G.1976. Characteristics andsignificance ofthe polyadenylate sequence inmammalian mes-sengerRNA.Prog. Nucleic Acids Res.17:118-148. 5. Bray, G. A. 1960. A simple efficientliquidscintillator

forcounting aqueous solutions inaliquid scintilla-tioncounter. Anal.Biochem. 1:279-285.

6. Doel, M. T., and N. H. Carey. 1976.The translational capacityofdeadenylated ovalbumin messengerRNA. Cell 8:51-58.

7. Goldstein, N.O., I.U. Pardoe, andA.T.H. Burness. 1976.Requirement of aadenylic acid-rich segmentfor theinfectivityofencephalomyocarditisvirus RNA. J. Gen.Virol. 31:271-276.

8. Hieter, P.A., S. M.LeGendre, and C. C. Levy. 1976. Stabilization of an RNA molecule by 3'-terminal poly(A)-inducedinhibition ofRNaseactivity.J. Biol. Chem.251:3287-3293.

J. VIROL.

9. Hruby, D. E., and W. K. Roberts.1976.

Encephalomyo-carditisvirusRNA: variations inpolyadenylic acid contentand biologicalactivity. J. Virol. 19:325-330. 10. Huez, G., G. Marbaix, E. Hubert, M. Leclercq, M. Nudel, H. Soreq, R. Salomon, B.Lebleu,M.Revel, and M. Z. Littauer. 1974. Role of thepolyadenylate segmentinthe translation ofglobinmessengerRNA inXenopus oocytes. Proc. Natl. Acad. Sci. U.S.A. 71:3143-3146.

11. Kerr, I. M., R. E. Brown, M. J. Clemens, andC. S. Gilbert. 1976. Interferon-mediated inhibition of cell-free-protein synthesisinresponsetodouble-stranded RNA. Eur. J. Biochem. 69:551-561.

12. Laemmli, M. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685.

13. Lodish, H. F. 1976. Translational control ofprotein synthesis. Annu. Rev. Biochem. 45:39-72.

14. Loening, U. K.1967.Thefractionationof high-molecu-lar-weight ribonucleic acid by polyacrylamide-gel electrophoresis. Biochem. J. 102:251-257.

15. Muller, W. E. G., G. Seibert, R. Steffen, and R. K. Zahn. 1976. Endoribonuclease IV.2.Further investi-gation on the specificity. Eur. J. Biochem. 70:249-258.

16. Nudel, U., H. Soreq, U. Z. Littauer, G. Marbaix, G. Huez, M. Leclercq, E. Hubert, and H. Chautrenne. 1976. Globin mRNA species containing poly(A) seg-ments of differentlengths:theirfunctional stability inXenopus oocytes. Eur. J. Biochem. 64:115-121. 17. Peacock, A. C., and C.W.Dingman.1967.Resolution

ofmultiple ribonucleicacidspeciesbypolyacrylamide gelelectrophoresis.Biochemistry 6:1818-1827. 18. Pelham, H. R. B., and R. J. Jackson.1976. Anefficient

mRNA-dependent translation system from reticulo-cytelysates. Eur. J. Biochem. 67:247-256.

19. Roberts, B. E., and B. M. Patterson. 1973. Efficient translationof tobacco mosaic virus RNA and rabbit globin 9SRNA in acell-freesystem fromcommercial wheat germ. Proc. Natl. Acad. Sci.U.S.A. 70:2330-2334.

20. Roberts, W. K.1965.StudiesonRNAsynthesisin Ehr-lich ascites cells: extraction and purificationof la-beled RNA. Biochim.Biophys.Acta108:474-487. 21. Roberts, W. K. 1974. Use ofbenzoylated cellulose

col-umns for theisolationofpoly(adenylic acid) contain-ingRNA and otherpolynucleotideswith little second-ary structure. Biochemistry 13:3677-3682.

22. Sharma, 0. K., W. K. Roberts, D.N.Beezley, andE. Borek.1975. AtransferRNA-dependent protein syn-thesizing system from Ehrlich ascites extracts. Biochim. Biophys. Acta 390:327-331.

23. Spector, D. H., and D. Baltimore. 1974. Requirement for3-terminalpoly(adenylicacid)for theinfectivity ofpoliovirus RNA. Proc. Natl. Acad. Sci. U.S.A. 71:2983-2987.

24. Spector, D. H., and D. Baltimore. 1975. Polyadenylic acid onpoliovirusRNA. II.Poly(A)onintracellular RNAs.J. Virol. 15:1418-1431.

25. Spector, D. H., L. Villa-Komaroff, and D. Baltimore. 1975. Studies on the function of polyadenylic acidon poliovirus RNA.Cell 6:41-44.

26. Villa-Komaroff, L., N. Guttman, D.Baltimore, and H. F. Lodish. 1975. Complete translationofpoliovirus RNA in a eukaryotic cell-free system. Proc. Natl. Acad. Sci. U.S.A. 72:4157-4161.

27. Williamson, R.,J. Crossley, and S. Humphries.1974. Translation ofmouse globin messengerribonucleic acid from whichthepoly(adenylicacid) sequencehas been removed.Biochemistry13:703-707.