Translation of encephalomyocarditis virus RNA in reticulocyte lysates: kinetic analysis of the formation of virion proteins and a protein required for processing.

JOURNAL OFVIROLOGY,May1979, p. 472-480 Vol. 30, No.2 0022-538X/79/05-0472/09$02.00/0

Translation

of

Encephalomyocarditis

Virus RNA in

Reticulocyte

Lysates: Kinetic

Analysis

of

the Formation

of

Virion

Proteins

and

a

Protein

Required

for

Processing

DINGS. SHIH,* C. T.SHIH,DAVID ZIMMERN, R. R. RUECKERT, AND PAUL KAESBERG Biophysics Laboratory of the Graduate School andDepartment of Biochemistry of the College of

AgriculturalandLifeSciences, University ofWisconsin,Madison, Wisconsin 53706 Receivedfor publication 11 December 1978

In cell-free extracts derived from rabbit reticulocytes, encephalomyocarditis RNAcanbe translatedcompletely,and theproductscanbeprocessed extensively to give encephalomyocarditis virion proteins and several nonvirion proteins, includingagenome-coded protein required for processing. The latter is probably

a protease. Translation is very efficient. Under typical conditions, each EMC

RNA istranslatedapproximately eight times duringa3-hperiod. Kinetic analyses (time-course experiments, pulse-chase experiments, and pulse-stopexperiments) have been usedtodetermine thetime ofappearanceofmajor products, andthese times have been correlated withmappositions. Thegeneforthe putativeprotease is located nearthe middle of the genome downstream from the virionprotein

genes.Ribosomescantravelthelength of encephalomyocarditisRNA within 30

min, but there isadelayintheirprogressalongthe RNA at some point soon after

theytraversetheregion coding for virion proteinprecursors.Thisdelayresults in

theaccumulation ofprecursors foraperiod of about 10minbeforethe putative

proteaseis madeand virion proteins(e,a,andy) arereleased byproteolysis.

Encephalomyocarditis (EMC)virus isa mem-ber of the picornavirus group. In vivo studies indicate that EMC RNA translation starts mostly at a single initiation site and proceeds throughalmostthe entirelength of thegenome. The synthesized protein is processed, at the

nascentchainstage, intoatleast threeprimary

products:amongthem,acapsid protein precur-sor,A, andthe noncapsid proteinsFand C. The intact polyprotein, containing A, F, and C, is detectableonly under conditions inwhich cleav-age of the synthesized protein isinhibited (3). The capsid proteinprecursoris cleavedto give proteinsa, E, and y. In thefinal stage of virion

maturation,afterpackaging of the viral RNA,e is cleaved into proteins ,B and S. F is a stable protein. Protein C is processed to yield protein

D andthenprotein E (2, 4, 13). Fig. 1 shows the

cleavagemap (3).

Early attempts to translate picornavirus

RNAs in vitro were not successful. The

transla-tion products in most casesconsisted of a

spec-trumof peptides of various lengths whose

exist-ence wasattributed to premature peptide chain

termination.The first definitiveidentification of

viral proteins synthesized in a cell-free system

wasbyVilla-Komaroffet al. (17). They showed

thatin extracts derived from HeLa cells,

polio-viral RNAdirects the synthesis of proteins la,

X, lb, 2, and 4

(corresponding

to EMC viral proteins A, F, C, D, and E,

respectively).

The

capsid protein la wasthe

major

product

under all conditionsreported.

Studiesof

picomaviral

RNAtranslation have beengreatly facilitatedbythe

procedure

of Pel-hamand Jackson fortreatingcell-free extracts

with nuclease to remove

endogenous

mRNA's (12). Svitkin and Agol, using nuclease-treated Krebs-2 ascites cellextracts,showed thata

num-berofproteins

synthesized

intheirsystem had

electrophoreticmobilityidenticaltothat of pro-teins made in EMC-infected cells

(15).

They

showed that insyntheses carriedto

completion

(2.5h),products

co-electrophoresing

with virion proteinsaandyweredetectable.However,ina

laterstudyin whichthey analyzed productsas

afunction oftime,no virionproteinswere evi-dent duringa 2-hperiod (16). Pelham showed thatEMC RNAcanbetranslated

completely

in

rabbitreticulocyteextracts, andthatprocessing

of theseproductscanbefollowedwithtime(11).

Hefoundthatsomeoftheprocessingiseffected byaproteolyticactivityresultingfrom the

trans-lation.In thisstudyalso,onlyasmall

proportion

ofprecursorswas cleaved.Recentlyweshowed

thatpolioRNAcanbetranslated

completely

in rabbit reticulocyte extracts and that the prod-uctsareextensively processed (14).Inthesame

472

on November 10, 2019 by guest

http://jvi.asm.org/

TRANSLATION OF EMC VIRUS RNA 473

Al

A

B

DI

F C

D E

a

FIG. 1. Cleavagemap of the EMC viral proteins

(modified slightly fromreference 3toinclude protein Al,amajor proteinfound in in vitro translation).

studywereportedbriefly that EMC viralRNA canbetranslated accurately in reticulocyte

ex-tracts,and,furthermore, the processing into

vir-ionproteinsa, y,andeand nonvirion proteinsD and E is extensive.

The translation efficiency of EMC RNA is about fourfold higher than that ofpolio RNA

(14), and this fact along with the completeness ofprocessing of the synthesized EMC products prompted us to study translation of its RNA

further with the objective ofgaining a better

understanding of the posttranslational cleavage reactions.Inthispaper,wewilldemonstrate the

progression of translation and of cleavage

result-ingin the appearance ofprecursors, a

genome-codedproteolytic activity, intermediatecleavage products, and final cleavage products. We also demonstrate anapparentdelay in theprogress of ribosome along the RNA at a region after

completionof the virionproteinprecursorsbut

beforecompletionofprotein F and defineamap

location for thegenome-coded proteolytic activ-ity downstream from thisobstruction. This

pro-vides experimental evidence foranovel

mecha-nism by which EMC can temporally regulate

expressionofitsgenes.

MATERIALS AND METHODS

Virus and RNA. Propagationof EMC virus and

RNA isolationwere aspreviouslydescribed(6, 18). Preparation of rabbitreticulocytelysate and

conditions forprotein synthesis.Reticulocyte

ly-sateswerepreparedaccording to theprocedure pre-viously described (14). Thelysate wasmade mRNA dependent by treatingwith micrococcal nuclease

ac-cording toa method slightly modified from that of Pelhamand Jackson(12).After thelysatewasthawed, creatine kinase (50 pg/ml),hemin (50 puM),

Tris-hy-drochloride (pH8.2) (20 mM), potassiumacetate (10 mM), staphylococcalnuclease(75 U/ml),and CaCl(1 mM)wereaddedto the final concentrations indicated.

Atypicalnucleasedigestion mixture contained 183.5

p1oflysate,4 A1 of1MTris-hydrochloride,1,ulof2M

potassium acetate,2p1 of5-mg/mlcreatine kinase(in

50%glycerol),2.5 p1of 4 mM hemin(in90%ethylene

glycol),2p1of 100mMCaCl,and 1plof micrococcal nucleaseat 15,000 U/ml.The mixturewasincubated atroomtemperaturefor 10minand thenucleasewas inactivatedbyadding4 p1of 100 mM

ethyleneglycol-bis(B-aminoethyl ether)-N,N-tetraacetic acid (neu-tralizedwithKOH).

Astandard protein synthesismixture (30

Au)

con-tained 17.5

I1

of the nuclease-treated lysate plus the following assay components: 10 mM creatine phos-phate, 90,uMeachof 19 unlabeled amino acids, 58,ug

ofcalf liver tRNA per ml, 5 mM dithiothreitol, 0.6 mM magnesium acetate, 100 mM potassium acetate (in-cluding that addedtothe lysate), and [3S]methionine (600 to 750 Ci/mmol) which was 0.8 pM for most experiments. Incubation was at 30°C for 3 h unless stated otherwise.

Electrophoresis. Conditions for treatment of sam-ples and for slab gel electrophoresis were as described (14). Normally, 5-plsamples of the original reaction mixtures wereanalyzed. Afterelectrophoresis, the gel was soaked in 400 ml of a solution containing 25% methanol, 10%acetic acid, and 1% glycerol for 4 h with twochanges of solution. The gel was dried and exposed

toaKodak NS 54T no-screen X-ray film for 1 or 2 days.

Materials. [iS]methionine was obtained from Amersham Corp. Staphylococcal nuclease and calf liver tRNA were obtained from Boehringer Mannheim

Biochemicals.Sparsomycinwas agift of John Douros,

Develpmental Therapeutics Program, Division of CancerTreatment, National Cancer Institute.

RESULTS

General properties ofthe incorporation reaction. Translation of EMC RNA proceeds exceptionally well in rabbit reticulocyte lysates. Upon addition of EMC RNA, amino acid incor-poration isreadily detectable within 5min and continues forabout 3h. Intherange of0 to 30 pg/ml, incorporation is proportional to RNA concentration andlevels offthereafter.

Maximumincorporationoccurs when

magne-sium acetate and potassium acetate are added

tolevels ofapproximately0.6 mM and 100mM, respectively,anddependsslightly onthe batch ofreticulocyte lysate. We estimate an endoge-nous contributionof 0.2 mM

Mg2e

and 50 mM

K+ so that maximum incorporation occurs at

approximately 0.8 mM Mg2+ and 150 mM K+.

Incorporation is roughly halved when the

con-centrations of these ionsareone-thirdortwice

theoptimumvalues.

From isotope dilution experiments we have determined that the concentration of methio-nine in our reaction mixture is about 10 puM.

Undertypical conditions,about45% of this

me-thionine is incorporated in 180 min when the concentration of EMCRNA (molecularweight

= 2.6 x 106) is 30 ,ug/ml, i.e., 11.5 x 10-3 pM.

Thus, approximately 4.5/(11.5 x 10-3) = 390

methionine residues areincorporatedperEMC

RNA molecule. We estimatethat the EMC

po-lyproteincontains about 49methionineresidues.

Thus each EMC RNA is translated

approxi-mately 390/49=8times.

Enumeration of theproducts of transla-tion andprocessing.Time-courseanalyses.

VOL. 30, 1979

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.497.53.245.51.125.2]

474 SHIH ET AL.

We have madeanalysessimilar to those of

Pel-ham(11)and confirm hisobservation that trans-lation of EMC RNA in rabbit reticulocyte

ex-tractsresults in theappearanceofEMC-specific

proteins, most ofwhich can be identified with

proteins synthesized in EMC-infected HeLa

cells (2, 14). Both in vivoand in vitro most,if

notall, of these proteinsresult from translation

beginning

ata

single

initiationsite,followed

by

a series ofproteolytic cleavagesofthe

polypro-tein

product.

We have extended our analyses to establish

thetime ofappearanceof thesynthesizedEMC

proteins, in particular the virion proteins e, a,

andyand theviralRNA-coded protease.

Figure

2 shows a radioautograph of a

polyacrylamide

.e_ P_ _ _qt st ;tv

e: -.^

m

_

*...

s

t._W__

r

i:w

1W:-gSfBt.

...

S

... *

e:S.2

tu_w

...

...i

w. * 3E ...},% S

_^.,

... ... c4W a;.

icr iugj _MS

-FIG. 2. Sequential synthesis ofvariousEMC viralproteins.Proteinsynthesiswascarriedoutunder normal

incorporationconditions.Incorporationwasstoppedattheindicated timesby addingEDTA andpancreatic

ribonuclease. Thesampleswereanalyzedonsodiumdodecylsulfate-polyacrylamide gels.Lanesathroughi

correspond to samples incubated for5, 10, 15, 30, 60, 90, 120, 180, and 300min, respectively. Lane i is a

diagramidentifyingthema]orband. Lane kcorrespondstoasampleincubatedfor16h.

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.497.105.392.156.617.2]

VOL. 30, 1979

TABLE 1. Apparent molecular weights calculated

fromband positions in Fig. 2a

Band A

Al .... .... ...

A ... ...

B ...

C ...

D ...

Dl

E. .... ... ..

E ...

F

a...

'y... Gl .... .... ...

G ...

H ... ...

J...

Lpparentmolwt

108,000 100,000 92,700 87,500 78,600 69,600 60,200 40,600 36,900 34,800 25,000 (15,000) (14,500) (11,600) (10,300) (9,700)

'True andapparentmolecularweightsmaydiffer

substantially for proteinscorrespondingtobandsG1,

G,H,I,and J.

gel pattern ofproteins synthesized (i.e.,

trans-lated andprocessed) during various times after

the addition of EMC RNAtoreticulocyte lysates

(lanes ato i), adiagram identifying the major

bands (lane j), and a gel pattern after a 16-h

incubationtoallownearcompletion of

process-ing (lane k). Table 1 lists the major bands

to-gether with their apparent molecular weight,

calculated from their mobility assuming that

protein A and proteinyhave molecularweights

of100,000 and 25,000, respectively (9). The

pro-teinorproteins called preA by Pelham (11)we

haveredesignated Altoconforn toour

nomen-clature.

Returningto Fig. 2, it maybeseen thatthe

time of appearance and disappearance of the

various bands is consistent with themapofFig.

1. Laneashows that within 5min of the addition

ofmessenger,bands A and Barealready visible,

althoughfaint. Band Al appearswithin 10min

and until 30minis thestrongestband. Bands F

and C are visible within 30 min. Most other

bands appearthereafter andchange

character-istically with time. After 16 h the only strong

bands that remainareE, F, G, H,and I and the

bandsofEMCvirionproteinsE,a,and-y.Clearly,

proteolytic processingisalmostcomplete.

It is evident thatinitialprocessingoccurs on

the nascent polyprotein to give Al, F, and C.

From theirearlyappearanceit isprobablethat

A and B are also derived directly from the

nascentpolyprotein. The intensityand time of appearanceof bandsD and Esuggestthatthey may also derive directly from thepolyprotein,

but the eventual disappearance of C and D

suggests that the processingrouteC-+D-+E

exists.

Pulse-chase analyses. Although time

TRANSLATION OF EMC VIRUS RNA 475

course analyses are useful for determination of

the quantity of a product as afunctionoftime, they donot serve wellin determiningthe pro-gression of processing, because intermediate products are augmented by protein synthesis while they are depleted by the processing reac-tions. Processing and identification of interme-diates can be followed more readily by pulse-chase analyses. Because of the highrate of trans-lation, labeling times as short as 15 min give

ampleradioactivity for detailed studies. During the first 15 min, translation will be predomi-nantly from the leftmost portion of the map of

Fig. 1. ThismaybeseeninFig. 3,lanesa to e,

which exhibits the proteins translated during that time and their processingover aperiod of 3 h. Incorporation proceeded for 15 min, at which timealargeexcessof nonradioactive me-thioninewasadded and incorporation and proc-essingwerecontinued. Since thechanging inten-sities of the bands reflect redistribution of the label incorporatedinthe initial15 min, process-ing canbe followed independent ofnew incor-poration. Lane a shows radioactive incorpora-tion for 15 min followed by 15 min of further incubation. Comparison with lane c of Fig. 2 showsthat theadditionalstrongbandsDl, Gl, and H andmoderatelystrongbanda arepresent. Presumably thesearecleavage products of pro-teinsAl, A,and/or B. Band Dl isstrongerthan its complement band a because protein Dl is richer in methionine than isproteina(H. Yosh-imura, Ph.D. thesis, University of Wisconsin,

Madison, 1976). Lanes bto eshowclearly that,

with increased time ofprocessing, proteins Al, A, and B decline inamountwhile virionproteins

E,

a,and yandproteinG increase. ProteinsDl andGl behaveasintermediates,firstincreasing

and then decreasing. The largeamount ofDl

remainingafter3hsuggeststhatitsdegradation

isrelativelyslow.

ProteinFisalso visiblethough weak, andits

amountremainsconstant. Probablytranslation into this region of the cistron hasbegunin the first 15 min and has continued with nonradio-activemethionineduringthe chaseperiod.

Lanes fto k (Fig. 3) show a corresponding experimentin which translation proceeded for 30min in thepresenceof radioactive methionine andthereafter for various times withexcess

non-radioactive methionine. It may be seen that

band F is nowstrong and, as before, invariant

withtine.BandsC, D,and Earepresent, indi-cating that translationinto theirportionof the

mRNA hasbegun withinthe first 30min. The

declineinamountofCand D with time of chase isconsistentwith the notion of theirprocessing

into E, whose amount increases with time of

chase.

Pulse-stopanalyses. Althoughprecursors of

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.497.53.246.84.254.2]

476 SHIH ET AL.

... ..

6&

^

u

4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~...

'....:...X:.'.

:..'..8}

..~~~~~~~Vl . .b,o:::Ig;bt@fo%W.:P:., ..:.

-

.. o_

*. *..* .S~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.

. ...'... . .. ...i;

-v~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

...

...

....

FIG. 3. Pulse-chase analyses of theformationof EMC viral proteins. The viral proteinswerelabeledwith

[35S]methionine for15min(lanesa toe) orfor30min(lanes ftok). Nonradioactivemethioninewasthen added tothe reaction mixtures (60,ul) to2 mM, and sampleswere withdrawn atthe indicated times for

analyses.

theEMC virionproteinsareevidentinthefirst 10minoftranslation,theirprocessingdoes not

commence until some 10 minlater, suggesting

that translation beyond these precursors is

re-quired fortheirprocessing.Thisfeatureof proc-essing can be followed by terminating protein synthesis (aswell aslabeling) at various times

so that processing can occur only bymeans of

enzymes already present in the extracts. Such

an experimentis shown in Fig. 4, which shows

gelpatterns where translation has been

terrni-natedatvarioustimes with theelongation

inhib-itorsparsomycin butproteolyticprocessing has

been allowed to proceed duringa

long

further

incubationperiod(lanesatog).Fortranslation

times shorter than 22 min, only bands Al, A,

and B are produced, and these remain after

prolonged further incubation. At 22 min the pattern changes abruptly (lane c) with the

ap-pearance of virion proteinprecursor Dl, virion

protein a, and protein Gi. Another processing intermediate, El, whose map position is

un-known, also appears. Also barely visible are

bands E, y,F, and G, which by 24 min are strong

bands. Thusanewproteolytic activity appears

after about 20 min oftranslation.

Figure 4 shows the emergence of band E at

about 26 min, a weak band D at 30min, but a

minor band in the position of band C even at

relatively earlytimes. We suggest that

transla-J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.497.105.394.68.450.2]

TRANSLATION OF EMC VIRUS RNA 477

15

20

22 24 26 28 30

min

r-B

+...i~~~~~~~~...,

Wem

p q

--B

D:

_~~~-^.-E

!a.

e2

~~ ~ ~ ~ ~ . . \z

'11.7

b r d e "f

*4_

h

9

FIG. 4. Pulse-stop analyses showing the effect of thelengthofincorporationonthe viral protein cleavage

reactions.Sparsomycin (200 uM)wasaddedtothe reaction mixturesattheindicated timestostopprotein synthesis. Lanesatog:incubationwascontinued foratotalof3h afterthe sparsomycin addition. Lanes h andi:sampleswerewithdrawnattheindicated timeswithoutfurtherincubation.

tion proceedsrapidlytogenerate B, A,andAl

very early, within 5 to 10 min of the time of messengeraddition; then translation slows down markedly nearF, traversingitatabout 20min,

Eatabout26min,and D at about 30min.The

slowdown in the F region is consistent with

studies we have in progress which show that

translationtraversingthe Fregion requires

ad-diton of tRNA'stothe reactionmixture,whereas

translation to thatregion is much less affected

bythe tRNA addition.

The weak band D indicatesthatproteinEcan

beproduceddirectlyfromthenascent

polypro-teinpriortotranslation farenoughtocomplete

proteinD. Thepresenceof minor band Cwhen

band D hasnot yet appeared requires further

study. Possibly itsmapposition isincorrect;that

is, in vitro band C may represent a protein

different from in vivo band C.Possiblyasmall

amountofrapid translation throughthe

obstruc-tion in region F is pernitted, enoughto see a

traceof band C.

Lanes h and i show thegelpatternsofsamples

takenright after theadditionofsparsomycin (at

30m.r 2C'! Mr

w VOL. 30, 1979

..1

4'9

..ON--ft ON" .60,4

K",

-,-l.

W-OW

"'m

.11. W-Iqw

I", go

so

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.497.107.399.69.481.2]

478 SHIH ET AL.

30 and 20 min, respectively) without further

incubation. It canbeseen from lane ithat the intensitiesof bands Al, A, and Bareverysimilar tothe intensities of thecorrespondingbands in lane b (20-min translationplus aprolonged in-cubation). Thisresult indicatesthat, priortothe formation of the putative proteolytic enzyme activity,proteins Al, A, andB arestable in the reticulocyte lysate andprocessinginthemanner

Al -e A-4Bdoesnot occur.Comparisonof the

patterns of lanes h and g (30-min translation

plusprolongedincubation)represents adifferent situation. Here the patterns differ greatly; the differentpatternfollowingprolongedincubation isattributabletothesynthesizedproteolytic ac-tivity, which inducesprocessing of theprecursor proteinsduring the incubationperiod.

EMC genome-coded proteolytic activity.

Figure4showed that termination of translation after 20 min, followed by long incubation to

allowprocessing, resulted inapatterncomprised

ofonly major bands Al, A, and B.If, however,

asecond reactionmixture,translated for60min, isadded, band Aldisappears,bandsAandB are diminished, and strong bands E, a, and y are evident (Fig. 5, lanesa to c). Translation pro-ceededinthepresenceof radioactive methionine for20min,sparsomycinwasadded,a nonradio-activetranslation mixturewasadded,and incu-bationwascarriedoutfor 2, 4, and16h. Lane d shows the corresponding 4-h experiment in whichanonradioactive reactionmixture devoid ofEMC RNA wasadded. Lane eshows a cor-responding experiment where the nonradioac-tive reaction mixture containedpolioRNAand its translation products. It may be seen that neither the mock reaction mixturenorthepolio products caused processing. A parallel experi-ment showed that the polio reaction mixture hadtranslated andprocessedpolio proteins

ef-fectively. We conclude that completely trans-lated and processed EMC extracts contain a proteolytic activity thatcangenerateEMC vir-ion proteins from precursors. Presumably this activity is due to a genome-encoded protease,

but it is alsopossible that thesynthesized

pro-tein in some way modifies a protease already

existingin thereticulocyteextracts.Inany event ourresults showthat thesynthesized protein is required for processing.

fr. ,

A

-I..., -

[image:7.497.270.439.70.383.2]

_.-_ Sa --E.

FIG. 5. Cleavageof the capsid protein precursors

bya virion-coded protease. For preparing

[35S]me-thionine-labeled capsid protein precursors, incorpo-rationreaction mixtures were incubated for 20min,

followed bythe additionof sparsomycin to stop

fur-ther protein synthesis. For preparing nonradioactive EMC or polio viral RNA translation products,

[36Slmethioninewasomittedfromthenormal reac-tion mixture andincubation was for 1 h. For the mock reaction mixture, both the[35S]methionineand theviral RNA were omitted. Lanes a to c:Ivolume ofthe radioactive precursor-containing reaction

mix-turewasmixedwith 1 volume of the reaction mixture

containing nonradioactive EMC viral RNA products andincubated for2h(lane a), 4 h (lane b), and 16 h (lane c). Lanes dande: theradioactive precursors

weremixed with the mock reaction mixture (lane d)

orthepolioviralprotein-containing reaction mixture

(lane e) and incubated for 4 h.

DISCUSSION

Ourresults suggestthat,inrabbitreticulocyte

extracts, translation of EMCRNAproceeds as

follows. Translation ofthelefthalfof theRNA

proceedsrapidly. This portion ofthe resulting

nascent protein has three sites susceptible to

proteolytic cleavage by reticulocyteenzymes to

yieldtheproteins Al,A, and B. Theseproteins

representtranslationfrom the beginningof the

cistron to the three susceptible sites and thus

differfromeach otheronlyin their

carboxyter-minalregion. Oncegenerated,theseproteinsare

resistanttofurthercleavagebyreticulocyte

en-zymes so that conversion of Al to A and B is

notdetectable, evenafterprolongedincubation

J. VIROL.

IP9

DI

on November 10, 2019 by guest

http://jvi.asm.org/

VOL. 30, 1979

(Fig. 4). Translation then markedly slows in traversing the middle portion of the RNA.The existence ofsome kind ofmechanism delaying

polypeptidechainelongation inaspecific region downstreamof Al seemsclearontwogrounds. First,weobserve completion of Al (108,000 dal-tons) within10min, whereascompletion of sig-nificantamounts of F (37,000 daltons)requires muchlongerthan theadditional4minpredicted ifelongation beyond the end ofAl wereequally fast.Second,poliovirusRNA, which isvery sim-ilar in size to EMC and directs synthesis of

analogous products, can be completely trans-lated under identical conditions within about10 min (14). Comparison of the results of polio RNA translation with EMC RNA translationas presented hereseems toindicate that the block-ingmechanism, whatever itis, isnotuniversal

among picornaviruses. We do not understand

why thesetwoapparentlysimilar viruses should differ in this respect, but the ability to make relatively pure uncleaved capsid precursors in thecaseof EMC isexperimentallyvery conven-ient and has allowedus to make amuchmore detailed analysis of the cleavage process than waspossible with polio.

Once translationproceeds beyond this block, protein Fappearsand also the members of the

C, D,Efamily. Proteolytic cleavageof thecapsid protein precursors Al, A, and B is correlated with the appearance of the proteins coded by theright-handhalf of thegenome, oneof which isclearly responsible for the proteolytic activity. Ourresults do notidentifywhich of these it is.

Preliminary experiments (unpublished) in col-laboration withAnn Palmenberg indicate that theputativeproteaseisnotproteinForprotein

G butmaywell be a proteinwe designate P22 thatcomigrateswith virionproteinyunderour conditions ofelectrophoresis (4). This is in

par-tialagreementwith thefinding of Lawrence and

Thach that virionproteinyhasproteolytic

ac-tivity (7).

We should reiterate atthis point one obser-vation thatmay notfit the schemejust outlined,

namely,theearlyappearanceinsmall amounts ofaband in thepositionweidentifywithprotein

C. Work is inprogresstoexaminethe

relation-ship of this early-appearing "C" band to the othersbytrypticpeptidemapping.

It canbe seen from the datathat,in contrast

tothestability ofAl,A, andBin the absence of

the viral protease, label can always be chased

from C and D into E on prolongedincubation evenafter translationhas beenstoppedsothat

onlythe processing of released polypeptides is being observed. Either the reticulocyte or the putative viralproteasecould beresponsiblefor these cleavages, since both are presentby the

TRANSLATION OF EMC VIRUS RNA 479

time C issynthesized.Smallfragments expected as by-products in all these cleavage reactions remain unidentified.Therefore we cannot deter-mine, for example, whether Al, A, and B may all becleavedtogiveDl and a and, if so, what is leftoverin eachcase.

On surveying the entireprocess as weobserve it, one is particularly struck by the longdelay encountered by ribosomes downstreamof poly-peptideAl. Onewould liketoknow, first,does thishappen in vivo, and, ifso, howis it accom-plished and what, ifany, purposedoes it serve? Ifit happens in vivo, it should leadto a transient overproduction of capsid proteins andtheir

pre-cursors midway through exponential growth,

compared to the equimolar yields of products expected if all ribosomes have run off the mRNA. There have indeed been somereports thatcapsid proteins aremade inexcess in car-diovirus-infected cells (8, 10), although these observationsmaybepartly duetouneven label-ing of the various products (11) or to under-counting the translation frequency of regions downstream from the coat protein (13). With regardtomechanism, wehavepreliminary evi-dence indicating thatan addition ofexogenous tRNA's to the reaction mixture increases syn-thesis past the block. Therefore a cluster of codonsrequiring minor isoaccepting tRNA spe-cies is one candidate for the block, possibly together withahighlystructuredregionof the RNA. Weshouldpointoutthat thepolycytidylic

acid tract known to exist in EMC RNA (1)

cannotberesponsiblesince it istooclosetothe

5'end of the RNAtobelocated downstream of Al (5). Finally, asfarasfunction is concerned,

it isprobablytooearlytomakegood suggestions. However, the long lag period in which capsid

precursors are stable before processing begins

does raise the possibility that in synchronous

infection the capsid protein precursor has an

earlyfunctionseparatefrom itslater structural role.

ACKNOWLEDGMENT1S

This researchwassupported bygrants01466 and AI-21942from the National Institute ofAllergyand Infectious

Diseases, bygrantCA-15613 from the National Cancer Insti-tute,andbygrantVC-26 from the American CancerSociety.

We thankMichael Janda forexcellenttechnical assistance.

LITERATURE CIMD

1. Brown, F.,J. F.E.Newman,J.Scott,A.Porter,D. Frisby, C. Newton, N.Carey,andP.Fellner.1974. Poly(C) in animal viral RNAs. Nature(London) 251: 342-344.

2. Butterworth,B.E., L.Hall,C. M.Stoltzfus,andR. R.

Rueckert. 1971.Virus-specificproteinssynthesizedin

encephalomyocarditisvirus-infected HeLa cells. Proc. Natl. Acad. Sci.U.S.A.68:3083-3087.

3. Butterworth, B. E., and B. D. Korant.1974. Charac-terizationof thelarge picornaviralpolypeptides pro-ducedin thepresence ofzincion. J. Virol.14:282-291.

on November 10, 2019 by guest

http://jvi.asm.org/

480 SHIH ET AL.

4. Butterworth,B.E., and R. R. Rueckert.1972.Kinetics ofsynthesis andcleavage ofencephalomyocarditis vi-rus-specificpolypeptides.Virology50:535-549. 5. Chumakov, K.M.,and V. I.Agol. 1976.Poly(C)

se-quence is located near the 5'-end of

encephalomyocar-ditis virus RNA. Biochem.Biophys.Res. Commun.71: 551-557.

6. Hall,L., and R. R. Rueckert. 1971. Infection ofmouse fibroblastsbycardioviruses: prematureuncoatingand itsprevention by elevatedpHandmagnesium chloride. Virology 43:152-165.

7. Lawrence,C.,and R. E. Thach. 1975.Identification of aviralproteininvolved inpost-translationalmaturation ofthe encephalomyocarditis viruscapsid precursor. J. Virol. 15:918-928.

8. Lucas-Lenard,J.1974.Cleavageofmengovirus

polypro-teins in vivo. J. Virol.14:261-269.

9. McGregor, S., L. Hall, and R. R. Rueckert 1975. Evidence for the existence of protomers in theassembly

ofencephalomyocarditis virus. J.Virol. 15:1107-1120. 10. Paucha, E.,J.Seehafer,and J.S.Colter.1974. Syn-thesis ofviral-specific polypeptidesinmengo-infected L-cells: evidence for asymmetric translation of the viral genome.Virology61:315-326.

11. Pelham, H. R. B. 1978. Translation of

encephalomyocar-ditis virus RNA in vitroyieldsanactive proteolytic

processing enzyme. Eur. J. Biochem. 85:457-482. 12. Pelham,H.R.B., and R. J. Jackson.1976.Anefficient

mRNA-dependenttranslation system from reticulocyte lysates. Eur. J. Biochem.67:247-256.

13. Rueckert, R.R. 1976. On the structure and morphogen-esis ofpicornaviruses, p. 131-213. In H. Fraenkel-Conrat and R. R. Wagner (ed.), Comprehensive virology. PlenumPublishing Co., New York.

14. Shih, D. S., C. T. Shih, 0. Kew, M. Pallansch, R. Rueckert, and P.Kaesberg. 1978. Cell-free synthesis andprocessing of the proteins of poliovirus. Proc. Natl. Acad. Sci. U.S.A. 75:5807-5811.

15. Svitkin,Y.V., and V. I. Agol. 1978. Complete transla-tion ofencephalomyocarditis virus RNA and faithful

cleavage ofvirus-specific proteinsincell-free system from Krebs-2cells. FEBS Lett. 87:7-11.

16. Svitkin, Y. V., V. N. Lyapustin, V. A. Lashkevich, and V. I.Agol.1978.Acomparative study on transla-tion offlavivirusand picornavirus RNAs in vitro. FEBS Lett. 96:211-215.

17. Vilia-Komaroff,L, N.Guttman, D. Baltimore, and H.F. Lodish.1975.Completetranslation ofpoliovirus RNA inaeukaryotic cell-free system. Proc. Natl. Acad. Sci. U.S.A. 72:4157-4161.

18. Zimmern,D., and P.Kaesberg. 1978.3'-Terminal nu-cleotide sequence ofencephalomyocarditis virus RNA determined by reverse transcriptase and chain-termi-nating inhibitors. Proc. Natl. Acad.Sci. U.S.A. 75:4257-4261.

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/