Picornaviral VPg sequences are contained in the replicase precursor.

0022-538X/80/08-0414/06$02.00/0

Picornaviral

VPg Sequences

Are Contained in the

Replicase

Precursor

MARK A.PALLANSCH,'* OLEN M.

KEW,'t

ANN C. PALMENBERG,' FREDGOLINI,2t ECKARD WIMMER,2ANDROLAND R. RUECKERT'

BiophysicsLaboratory ofthe Graduate School andDepartment ofBiochemistry oftheCollege of AgriculturalandLifeSciences, University of Wisconsin, Madison, Wisconsin53706,1andDepartment of

Microbiology,StateUniversity ofNew York atStony Brook, Stony Brook,New York117942 Ithas

previously

been shown that the RNAreplicaseof

encephalomyocarditis

virus contains two virus-coded proteins, D and E, which are produced in two

successiveproteolytic steps: (i)C-- D+?;and(ii)D -*p22+E. It is here shown

(i)

that virus protein H (molecular weight, 12,000) is the previously

unidentified

product of the firststepand(ii) thatVPg,aprotein linkedcovalentlyto thevirion

RNA,

yields twotryptic peptides foundinprotein C but not in proteinD.The resultssuggestthat

VPg

isderived

by

cleavage

ofproteinC and thatproteinH

maybeanintermediate.

Preliminary

experimentswith VPgsequencesin polio-viralnoncapsid protein lb, the counterpartofencephalomyocarditisviralprotein C,wereinconclusive.

Inall

picornaviruses

examined there isasmall protein (VPg) covalentlyattached to the 5'-ter-minal residueof thegenomeRNA

(8,

10, 13,

21). Circumstantial evidence suggests that

VPg

is

encoded bythe viral genome (8, 13);

however,

its function is

presently

unknown.BecauseVPg has been found linked to nascent RNA strands of the

replicative

intermediate

(7,

16),

it has

been

proposed

thatit isinvolved in RNA

repli-cation,

possibly

atthe

initiation

step

(7,

13, 16,

18).

An

unexpected result,

however,

was the

observation that

free

VPg

cannot

be detected

in extracts of

poliovirus-infected

cells

(Golini

and

Wimmer, unpublished

data).

This led

to the

suggestion

that

VPg

is

generated

by proteolytic

cleavage

from a

larger

precursor

protein during

initiation

of

RNA synthesis

(16,

25), possibly

from

noncapsid

viral

protein

2

(NCVP

2; refer-ence

6).

(Nomenclature for polioviral protein

is that ofSummers etal.

[24]

as

modified by

But-terworth

[1],

and that for

encephalomyocarditis

[EMC] viral proteins

is

that of

Butterworth

et al.[2].)

Alternatively,

it has beensuggested that VPg may represent the

product

of a less

fre-quently

translated second cistron(11). There is,

however,

nodirectevidence

supporting

either of theseproposals.

We now report

evidence,

based on tryptic

mapping,

thatVPg is

indeed

a

virus-coded

pro-tein and that it is derived from

protein

C of EMCvirus,

possibly

via an

intermediate protein,

H.

tPresent address:DivisionofVirology, CenterforDisease Control,Atlanta, GA 30333.

fPresentaddress: Department of Cellular Biology, Scripps Clinic and Research Foundation,La Jolla, CA 92037.

MATERIALS AND METHODS Virus and cells. The EMC virus and thepoliovirus

type 1(Mahoney) used have been describedpreviously

(9, 19). Virus was propagated in H-HeLa cells which were grown in medium B (15) containing 10% calf serum (K. C. Biologicals, Inc., Lenexa, Kans.) except whereotherwise noted.

Preparation of radiolabeled EMC VPg. Infec-tion ofcells insuspension culture followed the proce-dure ofMcGregoretal.(14), with the exceptionofa 30-minvirus attachmentperiod. Virionswerelabeled by addition ofeitherL-[4,5-3H]lysine (5.0 mCi to 4 x 107 cells) or L-[4,5-3H]leucine (10.0 mCito 4 x 105 cells) at2.5 hpostinfection. Incubation (37°C) in the presence of label continued until 6 h postinfection,at which time theinfected cell culture was subjected to threefreeze-thaw cycles. The released virus was puri-fied by pelleting,followed by banding in an isopycnic cesiumchloride gradient as previously described (15). The virus was removedfrom cesium chloride by pel-leting through a sucrose cushion. The RNA was ex-tracted from virions and precipitated as described previously (22). The VPg-containing RNA was di-gested for 60 minat37°C with a mixture of pancreatic RNase A (10,ug/ml),RNaseT,(10

jg/ml),

andRNase T2 (5U/ml) in 0.05Msodium acetate (pH 4.5). VPg was dissolved in 0.01% sodium dodecyl sulfate by heating for5min at100°C.

Preparation ofradiolabeled pollovirus VPg. Poliovirus VPg labeled withL-[4,5-3H]lysine was pre-paredasdescribed previously (19).

Synthesis of protein in cell-free extracts. In vitro EMCRNA-dependent protein synthesis was car-ried outexactly aspreviously described (22, 23). Trans-lation

(90-pl

reaction mixture) was in the presenceof

L-[4,5-3H]leucine (24,LCi) for 60 min, at which time 5

,ul

of pancreatic RNase A (1 mg/ml) and 10,Ld of cycloheximide (1mg/ml) wereadded. After an addi-tional 60min ofincubation, the sample was acetone 414

on November 10, 2019 by guest

http://jvi.asm.org/

VPg IN REPLICASE PRECURSOR

415

precipitated as previously described (17), dissolved in

electrophoresissample buffer, and dialyzed as previ-ously described (2).

Synthesis of protein in infected cells. Lysates wereprepared from 2x 107cellsinfected at a multi-plicityof 100 PFU/cell. To obtain proteins D or NCVP 2,cells were exposed at 240 min postinfection to either

L-[4,5-3H]lysine or L-[4,5-3H]leucine for 15 min. To obtainproteins C or NCVP lb, cells were infected in the presence of0.05%fetalcalf serum and treated with 0.91 mM (EMC virus) or 0.70 mM(poliovirus) ZnCl2

at204minpostinfection as described (3). Either

L-[U-'4C]lysine

or

L-[U-'4C]leucine

wasadded from 210min to 310 min postinfection. In both cases, cells were lysed at the end of the labeling period without washing, using hot sodium dodecyl sulfate, and the samples weredialyzed (2).

Preparative electrophoresis. Preparation of samples for electrophoresis has been described (15). Preparativeelectrophoresis for infected cell lysates to isolate proteinsC, D, NCVP lb, or NCVP2was the sameaspreviously described (22) except that the gel composition was 4.9% acrylamide, 0.2%

N,N'-methyl-enebisacrylamide, 0.1% sodium dodecyl sulfate, 0.1 M sodium phosphate at pH 7.2, 0.1%

N,N,N',N'-tetra-methylenediamine, 0.5M urea, andammonium per-sulfateat afinalconcentration of 0.05%. Preparative electrophoresis of in vitro reaction mixtures to isolate protein H wasperformed on 10% acrylamide gels con-taining8Murea(12).

Tryptic mapping. Proteinsweremixed,reduced,

alkylated,acidprecipitated,digested with trypsin, and analyzed by cation-exchange chromatographyexactly as

previously

described(12).

RESULTS

Evidence that

VPg

sequences

of EMC

vi-rus

reside in

protein

C. Evidence that

VPg

sequences are

contained

in

protein C

was

ob-tained

by

digesting

a

mixture of the

proteins

which had been

differentially

labeled

by

incor-poration of radiolabeled leucine.

To

obtain

la-beled

VPg,

purified

RNA,

extracted from

virions

grown

in

the

presence

of

[3H]leucine,

was

di-gested

for

1

h

with

a

mixture of

RNases

using

conditions which had

previously

been

shownto

release the

covalently

attached

protein

of

polio-virus

in the form

VPg-pUp (7, 13).

Protein

C

was

isolated

by

preparative electrophoresis

of

lysates from cells which had been

exposed

to

['4C]leucine

from

3.5to 4.5h

postinfection. Zinc

chloride

(0.91

mM)

was

included

during

the

incorporation period

to enhance the

yield

of EMC viral

protein C (3),

which is otherwise present in

only

small amounts. The mixture

of

VPg

and

protein C

was

alkylated,

digested

with

trypsin,

and

chromatographed

on a

cation-ex-change

column

to

resolve

the

peptides (Fig. 1A).

VPg

yielded

a

single

leucine-containing

peak

which

emerged

atfraction 11

(Fig. 1A,

arrow)

just behind

the

flow-through

peak (fraction

5).

The

digest

from

protein

C

also

exhibited

a

peak

in this position which

was,

however, somewhat

broader

on

the

right

shoulder.

That this broader

peak is

a

compound

one is

shown

by

comparing

the

tryptic profile

ofprotein C with that of its

cleavage product

(2), protein D (Fig.

1B).

This

profile shows that

the

tryptic

peak from VPg is missing from

protein

D. There

remains,

how-ever, a

smaller peak

in Dwhich accounts for the

noncoincidence

of the peaks from

protein

C and

VPg

(seen

in

Fig. 1A, arrow). The

difference

between the profiles of

proteins C and D was,

however,

not

completely accounted

for by the

tryptic peptide of VPg.

Thus one sees in the

profile

of protein C noncoincidences

atfractions 27,62, 93,

and

98.

As

seen

in

Fig. 1C, the

tryptic profile of protein H

accounts

for

all

of these

differences

between

C and

D,

including

the tryptic peak

emerging

with that in

the VPg profile (arrow). These data

indicate that

H is the

previously unidentified

cleavage

complement

of

D

(i.e.,

C -. H

+

D).

The data also

strongly

suggest that VPg is part

of the H sequences

contained

in

protein

C but not

in

protein D.

The

latter conclusion

isgreatly

strengthened

by

a

second experiment modified by substituting

radiolabeled

lysine for radiolabeled

leucine (Fig.

2A).

The

lysine-labeled

peak coincides with

one

found in

protein C but

not

in

protein

D

(Fig.

2B), and this

peak

is

different from the

leucine-labeled

peak

in

Fig.

1. Thus

VPg exhibits

two

tryptic

peaks,

one

containing

lysine

but

no

leu-cine

and

a

second that contains leucine but

no

lysine, which coincide with

tryptic

peaks

from

protein

C but

not

from

protein

D.

VPg in

polioviral

proteins.

The

polioviral

homolog of

protein

C is NCVP

lb,

which is

cleaved

to

NCVP

2

plus

a

still

unidentified

pro-tein.

Based

uponan

amino acid

analysis

of

VPg,

the

RNA

sequence

coding

for

VPg

has been

determined (N.

Kitamura, C.

Adler,

J.

Martinko,

S. G.

Nathenson,

and E.

Wimmer,

submitted for

publication).

This RNA

sequence is

located

ap-proximately

2,000

nucleotides from the 3' end of

the

poliovirus

genome,

that

is,

within

a

segment

of RNA

coding

for

NCVP lb.

Moreover,

the

amino

acid sequence of

VPg,

as

deduced

from the

RNA sequence,

predicts

that

digestion

of

VPg with

trypsin

should

yield

three

peptides.

Thesethree

peptides

have

indeed

been

observed

(Fig.

3A).

To

determine

whether the

VPg

sequences of

poliovirus could

be

demonstrated

inNCVP

lb,

the

proteins,

differentially

labeled with

lysine,

were

compared

as

before.

In

this

case

polioviral

VPg

exhibited

three

tryptic

peaks,

a,

b,

and

c

(Fig. 3A).

None

coincided

exactly

withthe

major

peaks

in

NCVP

lb;

moreover,

peak

afrom

VPg

appeared

tobe

entirely

absentfrom

NCVP

lb.

VOL. 35, 1980

on November 10, 2019 by guest

http://jvi.asm.org/

416 PALLANSCH ET AL.

4000-

A

2000- N

E

la

CL

50 1 00 150 200

Fraction Number

FIG. 1. Tryptic profiles ofleucine-labeled EMCproteinsC, D,H, andVPg.Differentially labeled proteins were mixed, digestedwithtrypsin, chromatographedon acation-exchange column, andassayed for radio-activityasdescribed in thetext.(A)3H-labeledVPg( )versus

"4C-labeled

proteinC(---). (B)3H-labeled proteinD( )versus"C-labeledproteinC(---). (C)3H-labeledproteinH( )versus"C-labeledprotein C(---).ProteinHwasisolatedfromacell-freetranslationproduct (seethetext).Scaleswereadjustedsuch that theratiooftheirheightsequaledthespecificactivityoftheproteins (disintegrationsperminute[dpm]

perdalton).

These results do not exclude the possibility,

however, thatVPg peaksb andc aremembers

ofcompound tryptic peaksfromNCVP lb,and that VPg peak a represents the new tryptic peptideproducedwhen NCVP lb iscleavedby

the viralprocessingenzyme.Alternatively, peak

acould beatryptic peptidelinkedto-pUp.This

nucleotidyl peptidewould have no counterpart inNCVP lb.

The possibility of compoundpeaks in NCVP lbwasfurtherexaminedbycomparingthe tryp-ticprofilesofNCVP lband its cleavage product

NCVP2 (Fig.3B). The resultsobtained in this

experimentwereinconclusive. Thiswas due in part to inability to resolve the VPg peptides

definitively withinthe largenumber of peptides found in NCVP lb and 2. A second difficulty was the presence of an unrelated protein, de-tectedbyanalytical electrophoresisinslab gels, in ourpurified preparations of NCVP lb. This

contaminantwasalsomanifestedbytryptic

pro-files which became significantly simpler when

NCVPlbwasenriched by treating infectedcells

with zinc (results not shown). When VPg and NCVP 2 were compared directly, two of the three VPgpeptides were clearly resolved from

anyinNCVP 2,suggesting the absence ofVPg

sequences in this protein (not shown). Other

experiments with lysine-labeled VPg excluded thepresenceofVPgsequencesinpolioviral

cap-sid precursor NCVP la, in NCVP5b, which is relatedtostableproteinX(20),and in thesmall unmapped proteins 9a, 9b,and 10.These

exper-imentswererepeated usingleucineasthe

radi-olabel, but the results again failed to establish conclusive evidence forVPgsequencesin lb.

DISCUSSION

Thedatapresented hereindicate (i) that pro-tein H is the previously unidentified product

50 100 150 200

-6 0 0

-400

-20 0

E

C)

)00

-600

-300

I

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.504.119.407.62.392.2]

VPg

IN REPLICASE PRECURSOR 417

2000-1

000-E 0Q

2:

A

4

:4

.4.': ::

1: :*

-800

-400

E

0-p400

t

1*

-20 0

50 100 1 50 200

Fraction Number

FIG. 2. Trypticprofiles oflysine-labeled EMC viral proteins C, D, and VPg. (A) 'H-labeled VPg ( )

versus

"C-labeled

protein C (---). (B)3H-labeledprotein D ( )versus"C-labeledprotein C

(---)-

2000-1

000-E

0--o

-1 50

-100

50

E

-o

[image:4.504.103.398.68.291.2]

(-Fraction Number

FIG. 3. Tryptic profiles oflysine-labeled poliovirusproteinsNCVPlb, NCVP2, andVPg. (A)3H-labeled VPg( ) versus '4C-labeled NCVP lb (---). (B) 3H-labeled NCVP2 ( ) versus"C-labeledNCVP lb

(-- ).

obtained when EMC viralprotein Cis cleaved to formprotein D, and (ii) thatVPgis partof

protein H. These experiments are consistent

with models ofpicornavirus replication proposed previously (7, 13, 16, 18, 25) which suggest in-volvement ofVPgininitiation of RNAsynthesis.

They arealsoconsistentwith theproposal (17)

that thepolymeraseprecursorC donatesVPgto the nascent RNA strand during initiation of

synthesis.ThefindingofVPgsequencesin

pro-tein H suggests that generation of VPg from

protein C involves more than one proteolytic

50 1 00 1 50 200

A

b4

a C

4 *:A:

4

A ::

.i1

hAI

50 100 1 50 200

u

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.504.102.392.329.552.2]

CAPSID

.I

NON -CAPSID

C(83)

A(loo) [ H(12) D (75) ]

B(90)

VPl8)([p22(20)

E(56) ]

DI (65) a(34)

C(40) Y(23)

G016) 1(11)

i(9)()

FIG. 4. ProcessingmapofEMC viralproteins re-visedto includeproteinsH andVPg. Brackets en-closeproteins whose order is still unclear. Box en-closesunmapped proteins.Electrophoretically deter-mined molecular masses(parentheses) aregiven in kilodaltons (17). Theelectrophoretically determined molecularmassof VPgis8,000to10,000daltons(8); however, sequencing studiesonpolioviral- VPg sug-gestamasslessthan3,000.

step.It remainstobeclarified,however,whether

H is an intermediate in the pathway of VPg additiontoRNA.

Thenewdata have beenincorporated into the proteolytic processing mapof

EMC

virus (Fig. 4). The position of protein H on the amino-terminal side ofprotein D is basedon pactamy-cin mapping (4); however, noconclusioncanbe drawn from the tryptic dataon the position of VPg within protein H.

IfEMC andpoliovirus follow thesameprotein cleavage pattern, then the presence ofVPg in protein H could explain why polio VPg is not

found free in the infectedcell(25). Because it is known that the number of translations of the viral genome significantly exceeds the number ofnewviral RNA strands made inaninfection (reviewed in N. Kitamura, C. Adler, and E. Wimmer, Ann. N. Y. Acad. Sci., in press, and

reference5),there should beasignificantexcess ofVPgsequences.If thesequencescontainedin

proteins C and H areonly released when RNA synthesisoccurs,freeVPgmaynotbe detectable ininfected cells. In otherwords, protein Hmay

representtheexcesssynthesis of

VPg,

although its role in RNAreplication ispresently unclear. Themajor cleavagestepsin the processing of replicase-related proteins of poliovirus and EMC virusareanalogous (lb-- 2+? is analogousto

C

-- D + H; reference 20). Accordingly, we

anticipate that VPg of poliovirus is containedin

NCVP lb andnotin NCVP2.Theresults

pre-sentedherearecompatible with this model,but inconclusive.

ACKNOWLEDGMENTS

WethankDaniel Omilianowski forexcellenttechnical

as-sistance in running the tryptic peptide column and Jane Hubertz for assistance inthepreparationofVPg.

This workwassupported bygrantMV-33Ifrom the

Amer-icanCancerSociety and by Public Health Service grants CA-08662,CA-16879, and AI-15122 from the National Institutes of Health. M.A.P. is a National Science Foundation Predoc-toralFellow.

LITERATURE CiTED

1. Butterworth, B. E. 1973. A comparison of the virus-specific polypeptides ofencephalomyocarditis virus, hu-manrhinovirus-lA, andpoliovirus. Virology 56:439-453.

2. Butterworth,B.E., L.Hall,C. M.Stoltzfus,and R. R. Rueckert.1971.Virus-specific proteinssynthesizedin encephalomyocarditis virus-infected HeLa cells.Proc. Natl. Acad. Sci. U.S.A. 68:3083-3087.

3. Butterworth, B. E., and B. D. Korant.1974. Charac-terization of the large picornaviral polypeptides pro-ducedin thepresence of zinc ion. J.Virol. 14:282-291. 4. Butterworth, B.E., and R. R. Rueckert. 1972.Gene

order ofencephalomyocarditis virusasdeterminedby studies withpactamycin.J.Virol. 9:823-828.

5. Cooper,P.D.,A. Steinor-Pryor,and P. J.Wright. 1973.Aproposed regulator for poliovirus: the equestron. Intervirology1:1-10.

6. Dasgupta,A.,M.H.Baron,and D. Baltimore.1979. Poliovirusreplicase: asoluble enzyme able toinitiate copying of poliovirus RNA. Proc. Natl. Acad. Sci. U.S.A. 76:2679-2683.

7. Flanegan, J. B., R. F. Pettersson, V.Ambros,M.J. Hewlett, and D. Baltimore.1977.Covalent linkageof aproteinto adefined nucleotide sequenceatthe 5'-terminus of virionandreplicative intermediate RNAs ofpoliovirus. Proc. Natl. Acad. Sci. U.S.A.74:961-965. 8. Golini, F.,A. Nomoto, and E. Wimmer. 1978. The genome-linked protein of picomaviruses. IV. Difference in theVPg's ofencephalomyocarditis virus and polio-virus as evidence that the genome-linkedproteinsare virus-coded.Virology 89:112-118.

9. Hall, L., and R. R. Rueckert. 1971. Infection ofmouse fibroblastsbycardioviruses: prematureuncoating and itsprevention by elevated pH and magnesium chloride. Virology 43:152-165.

10. Hruby,D.E.,and W. K. Roberts.1978. Encephalomy-ocarditis virus RNA. III. Presence ofa genome-associ-atedprotein. J. Virol.25:413-415.

11. Humphries, S.,F.Knauert, and E. Ehrenfeld. 1979. Capsidprotein precursor is one of two initiated products oftranslation ofpoliovirus RNA in vitro. J. Virol. 30: 481-488.

12.Kew,0.M., M. A.Pallansch, D. R.Omilianowski, and R. R. Rueckert.1980.Changes in three of the four coatproteinsof oralpoliovaccine strainderived from type1poliovirus.J.Virol.33:256-263.

13. Lee,Y.F.,A.Nomoto,B. M.Detjen,and E. Wimmer. 1977.Aprotein covalently linkedtopoliovirus genome RNA. Proc. Natl. Acad. Sci. U.S.A. 74:59-63. 14.McGregor, S., L. Hall, and R. R. Rueckert. 1975.

Evidence for theexistence ofprotomersinthe assembly ofencephalomyocarditisvirus. J. Virol. 15:1107-1120. 15.Medappa,K.C.,C.McLean,and R. R. Rueckert.1971.

Onthe structureof rhinovirusIA.Virology 44:259-270. 16.Nomoto, A.,B.Detjen,R.Pozzatti, and E. Wimmer. 1977.The location of thepolio genome protein in viral RNAs and itsimplication for RNA synthesis. Nature (London) 268:208-213.

17. Palmenberg,A.C.,M. A.Pallansch, and R.R. Rueck-ert. 1979.Protease required forprocessing of picorna-viral coatprotein residesinthe viralreplicase gene. J. Virol.32:770-778.

18. Pettersson, R. F., J. B. Flanegan, J.K. Rose, and D. Baltimore. 1977.5'-Terminal nucleotide sequences of poliovirus polyribosomal RNA and virion RNA are identical.Nature(London)268:270-272.

19.Rothberg,P.G.,T. J. R.Harris, A. Nomoto, and E.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.504.68.259.47.166.2]

Wimmer. 1978. Genome-linked protein of

picornavi-ruses.5.04-(5'-uridylyl) tyrosine isbond between

ge-nome-linkedprotein and RNA of poliovirus. Proc. Natl. Acad.Sci. U.S.A.75:4868-4872.

20. Rueckert, R. R., T. J. Matthews, 0. M. Kew, M.

Pallansch, C. McLean, andD.Omilianowski.1979. Synthesis and processing of picornaviralpolyprotein,p.

113-125.InR.Perez-Bercoff (ed.), The molecular

biol-ogyofpicornaviruses. Plenum Press, New York. 21. Sangar, D. V.,D. J.Rowlands,T.J.R.Harris, andF.

Brown. 1977. Protein covalently linked to foot-and-mouth diseasevirus RNA. Nature (London) 268:648-650.

22. Shih,D.S.,C. T.Shih,0.Kew,M.Pallansch,R.R.

Rueckert,and P.Kaesberg.1978.Cell-freesynthesis

andprocessingoftheproteinsofpoliovirus.Proc.Natl. Acad.Sci. U.S.A. 75:5807-5811.

23. Shih, D.S., C. T. Shih, D. Zimmern, R. R. Rueckert, and P.Kaesberg. 1979. Translation of encephalomy-ocarditis virus RNA inreticulocyte lysates: kinetic anal-ysis of the formation of virion proteins andaprotein

required for processing. J.Virol.30:472-480. 24. Summers, D. F., J. V. Maizel, Jr., and J. E. Darnell,

Jr. 1965.Evidence forvirus-specific noncapsid proteins inpoliovirus-infected HeLacells.Proc.Natl. Acad. Sci.

U.S.A. 54:505-513.

25. Wimmer. E. 1979.Thegenome-linkedproteinof picor-naviruses:discovery, properties and possible functions,

p. 175-190. InR. Perez-Bercoff (ed.), The molecular biology of picornaviruses. Plenum Press,NewYork.

on November 10, 2019 by guest

http://jvi.asm.org/