Control of simian virus 40 gene expression at the levels of RNA synthesis and processing: thermally induced changes in the ratio of the simian virus 40 early mRNA's and proteins.

(1)

0022-538X/80/07-0157/08$02.00/0

Control

of

Simian

Virus

40

Gene

Expression

at

the

Levels

of

RNA

Synthesis

and

Processing:

Thermally

Induced

Changes

in

the

Ratio of the

Simian

Virus 40

Early mRNA's and

Proteins

JAMES C. ALWINE'* ANDGEORGEKHOURY2

DepartmentofMicrobiology, School of Medicine, University of Pennsylvania, Philadelphia,Pennsylvania

19104,'andLaboratory ofMolecularVirology,NationalCancerInstitute,National InstitutesofHealth,

Bethesda, Maryland202052

Examination of thesimian virus40early mRNA's frominfectedAGMKor

CV-1 cells showed that the ratio of large T- tosmall t-antigen mRNA's increased

with an increased incubation temperature. In tsA58 mutant-infected cells, an

increased incubationtemperatureresulted in the overproductionofearly RNAs; however, the ratio of the early mRNA's was the same,at any temperature, in bothwild-type-andtsA58-infected cells.Thus, thethermallyinduced alteration

intheearlymRNA ratioswasapparentlynotaffected by the tsA mutationorby the overproduction of early RNA in tsA mutant-infected cells. Time course

studies atvarious temperatures showed that, although the ratio of large T- to

small t-antigen mRNA's increased with temperature,atany one temperatureit

was constantfromearlytolate times of infection. Furthernore, the ratio of the

early mRNA's adjusted intemperatureshiftexperiments. Thus, the ratio of the early mRNA's appearedtobeintrinsictothe thermodynamic environment of the cell. Thethermally induced alterationsintheearly mRNA'swerereflectedatthe protein level byparallelchanges in the ratio of largeT- tosmallt-antigens. These

datasuggest alevel ofgeneexpressioncontrol whichmayfunctionatthestageof

sphlcing.

Thesmall DNAtumorviruses haveprovided excellent modelsystemsforstudying the control ofgene expression. Inthisregard,simian virus

40 (SV40;see Fig. 1) hasbeen particularly val-uable for theanalysisoftranscriptionaland

post-transcriptional regulatory mechanisms. The temporal expression of early andlateSV40genes

appearstobe controlledatthelevel of

transcrip-tion (2, 5, 15,21). These studies showed thata

primary factor in determining the transcrip-tional activity ofthe SV40 genes isone of the early gene products, large T-antigen, which is thoughttohavearepressor-likeactivity in mod-ulating early transcription. These conclusions

are based, in part, onstudies with early SV40 temperature-sensitive mutants (tsA mutants). Atelevatedtemperaturesthesemutantsbecome

defective inlarge T-antigenfunction(la, 19, 28,

30) andoverproduceearlyviralRNA (2, 15,21).

The primary early and late transcripts are

modifiedsubsequent totheirsynthesis by

cap-ping,polyadenylation,internalmethylation, and

splicing.All of these modificationsare alsofound inthebiosynthesisofeucaryoticmRNA's. What

appears, thusfar,tobe unique forviral mRNA's

is the observation that one primary transcript

may be the precursor for different mRNA's

throughalternatesplicingevents. Forexample,

a single primary transcript of the SV40 early coding region may be the precursor for the mRNA coding for either small t- or large T-antigen,dependingonthe location of thesplicing

event(4,6). If this is thecase,thenthe selection

of the splicing eventprovidesapotential

regu-latory step inearly SV40 gene expression,and factors influencing the relative frequency of eithersplice will control the relative abundance of the two early SV40 proteins, large T- and smallt-antigens.

The study reportedhere wasinitially under-takentodeterminewhetherlarge T-antigen

reg-ulatesnotonlytherateofsynthesisofprimary early transcripts, but also the relative amounts oflarge T- and small t-antigen mRNA's. Our

results indicatedamuch more interesting

phe-nomenon;wefound inbothwild-type (WT)-and

tsA mutant-infectedcellsthat theratio oflarge

T- to small t-antigen mRNA's increased and

decreasedwith incubation temperature. These

thermally induced alterations in the early

mRNA ratios werereflected at theproteinlevel

by similaralterations in thelargeT-tosmall t-157

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

158 ALWINE AND KHOURY

antigen ratio. Thus,thethermal(or thegeneral thermodynamic) environmentof thecellalters the expression of the early genes, indicating a

level ofgene

expression

controlwhich may func-tionatthe level of splicing.

MATERIALS AND METHODS

Cellsand viruses. Primary African green monkey

kidneycells(AGMK) and theestablishedAGMK line CV-1 were cultivated in minimal essential medium plus 10% fetal calf serum. The virus strains used were the WTstrain VA4554 (25)anda

temperature-sensi-tive mutantderived from it, tsA58 (26, 27).

Infectionof cells andpreparation of cytoplas-niicRNA. Confluent monolayers of cells in 150-cm2 Costarplasticbottles were infected at a multiplicity of

10 PFU/cell in 5 ml ofminimal essential medium

containing 2%fetal calf serum. After rocking at room temperature for 1.5h, 20 ml of medium was added, and thecells were incubated at320Cfor 40 h. At this point, cellswereharvested (32°C samples) orshifted to 41°C for5h (shifted samples) before harvesting. Othercultures wereinfected for24hat41°C before harvesting (41°C samples). Cytoplasmic RNA from infected cells was prepared as previously described (13).PolyadenylatedRNA wasselected oncolumns of

oligodeoxythymidylic acid-cellulose(3).

Preparationof52P-labeledSV40 DNA

hybrid-izationprobes.SV40 DNAwaslabeled in vivo and

purifiedaspreviouslydescribed (16). The labeled form

ISV40 DNA wasdigested with appropriate restriction endonucleases (New England BioLabs, Waltham, Mass.), using conditions indicated by the company. The reaction mixtures were phenol extracted and ethanolprecipitated,and the strands of thefragments

wereseparatedbythemethod of Hayward (10).

Sep-aratedstrands were electroeluted,self-annealed,and

chromatographed on hydroxyapatite. The resulting

purified singlestrandsweredialyzedinto10mM

Tris-hydrochloride(pH7.5)-imM EDTA.

Hybridizationofsingle-strandedDNA probes

and cytoplasmicRNA. To maintain conditionsof DNAexcess, the amountof RNAused in each hybrid-ization wasvaried according to the abundance in the

sample. DNA excessconditionswere establishedby

titration of differentsamples withseparated strand probes containing the early coding region. Samples

contained0.5to10

,g

ofpolyadenylatedRNA or5 to

70jugofnonpolyadenylatedRNA. Foreach reaction,

10ng (10,000 cpm) of DNAprobe was added. The RNAandprobewereethanolprecipitatedtogether in silane-treated tubes. The pelletsweredried ina vac-uumand thendissolved in30plof 50%formamide-0.4

M NaCl-0.04 M PIPES

[piperazine-N,N'-bis(2-eth-anesulfonic acid)](pH6.5). Thesampleswere

hybrid-izedat370Cfor24to 72h. Longhybridizationtimes wereusedtoinsure that the reactionswerecomplete;

this varied with thesample and had beenpreviously

determined.

Analysis ofhybrids bynucleaseSIdigestion

and alkalinegelelectrophoresis.After hybridiza-tion, thesamplesweretreated with nucleaseS1 and

electrophoresedon 1.4 or1.8%alkaline agarosegelsas

described by Berk and Sharp (4). Thegelsweredried onDEAE paper andautoradiographedat-70°C,

us-ingKodak XR-1 X-ray film and a Du Pont Cronex Lightning-plusintensifyingscreen.

Analysis of

[3'Sjmethionine-labeled

viral pro-teins. For theanalysisof the ratio oflargeT-tosmall

t-antigens,confluent monolayers of AGMK or CV-1 cells, grown in 50-mm plates (three plates per set), wereinfected with VA4554 or tsA58 at a multiplicity of 10. The infectedcells wereincubated for40h at

320C.

Atthis point, thecellswerepulse-labeledfor1 h at

320C

with

[nS]methionine

orshiftedto41°C for 4hbeforepulse-labelingfor1hat

410C.

Athirdsetof cultures wasinfected entirely at

410C

for 24 hand

then pulse-labeled for 1 h. The pulse-labeling was

accomplishedby washing the cells threetimes with

appropriately warmed (32 or

410C)

methionine-free

minimalessentialmediumcontaining 2%dialyzedfetal bovine serumand then labelingeach platewith 1ml ofprewarmedmediumcontaining200giCiof [3S]me-thionine (300 Ci/mmol; Amersham Corp., Arlington

Heights,Ill.).Afterlabeling, theplateswereputonice and washed three times with cold analysis buffer

(Tis-bufferedsaline,pH8,containing1mMdithiothreitol,

300,ugofphenylmethylsulfonyl fluoride perml, and 150Agoftosylamid-2-phenylethylchloromethylketone perml).Thecellswerescrapedinto1.5ml ofanalysis

buffer,sedimentedat3,000 rpm for 10 min,suspended

inanalysisbuffercontaining0.5% NonidetP-40,

incu-bated onice for30min, andvigorouslyblended ina Vortex mixer every 5 min. The lysates were sedi-mentedat15,000 rpm for60min, and thesupernatant fractionwaswithdrawn andsaved.Samples(0.3 ml)

of the supernatants were immunoprecipitated with

heat-inactivated Staphylococcus aureus and

antise-rum (directed against the SV40 early proteins) as

described by Jay et al. (12). The serum had been titratedpreviouslytodetermineoptimalconditions for thesesamples.Both theserumand S. aureuswerethe gift ofGilbert Jay.Immunoprecipitates (in15- to 20-plvolumes)wereanalyzedbyelectrophoresison10% sodiumdodecylsulfate-polyacrylamideslabgels (11).

Radiolabeledproteinsweredetectedbyfluorography

(17). After location ofspecificbandsonthegels,the driedgelsegmentscontainingthemwereexcised and

rehydratedinwaterfor20minat roomtemperature.

Thegelsegmentswerethenequilibratedindimethyl

sulfoxideto removethediphenyloxazoleused in

fluo-rography(fivewashes,2ml ofdimethylsulfoxideeach,

for20minatroomtemperature).Thegelsliceswere

equilibrated inwater and thentreated with 1 ml of 0.8% ammonium bicarbonatecontaining60

pLg

of tryp-sinfor12hat

370C.

Themixturewasthencountedin 10ml ofAquasol(NewEnglandNuclearCorp., Bos-ton,Mass.).

RESULTS

Experimental

approach.Totalcytoplasmic

or polyadenylated cytoplasmic RNA (both of

which gave identical results in the following

experiments) was isolated from SV40-infected monkeykidney cells under avariety of experi-mentalconditions. TheseRNAswereannealed

with 3P-labeled, separated, single-stranded

DNAprobes containingtheentireknownearly coding regions oftheSV40genome.Thehybrid

http://jvi.asm.org/

(3)

moleculeswereassayed by theSi nuclease

tech-niqueofBerk and Sharp (4), followed by alkaline

agarose gel electrophoresis and

autoradiogra-phy. For the quantitation of the amounts of

mRNAspeciesfrom autoradiographic band

in-tensities, hybridization reactions were

per-forned in DNA excess, and incubation times

wereprolongedtoinsure that reactions hadgone tocompletion.

Synthesis of large T- and small t-antigen mRNA's. Previous experiments showed that earlySV40 RNA is overproduced in the absence ofafunctional large T-antigen (2, 15, 21). We wanted todetermine whether thelarge

T-anti-gen wasinvolved(i) only in the control of

initi-ation ofanyRNAsynthesis from the early cod-ing region, presumably by operatcod-ingonthe

pro-moter, or(ii) inregulation of theamountof large

T-antigen mRNA relative to small t-antigen mRNA. Parallel cultures of AGMK cells were

infected withanearlySV40

temperature-sensi-tivemutant(tsA58)atthe permissive

tempera-ture (32°0) for40h. At that time, one culture

washeldat32°C (320Csamples), and the other

was shifted to the nonpermissive temperature

(410C) for 5 h (shifted samples). In a control experiment, similar RNAsampleswereprepared

from parallel cultures infected with WT SV40. Samples of the RNAs purified from these

cul-tures were annealed with a 3P-labeled SV40

DNAproberepresentingthe earlystrand ofan HpaII-BamHIfragment (0.14to0.72SV40 map unit

[m.u.]

[Fig. 1]) which contains the early coding region (input amounts of RNA varied due to overproduction in the tsA-shifted sam-ples; see legendto Fig. 2). The analysisof the RNAsfrom these cultures is shown inFig.2.In

each track, there are three prominent bands (2,000,580,and320nucleotides

[n])

which

rep-resentthe common early RNA body (2,000 n)

and theleaders for smallt-antigen mRNA (580 n) and largeT-antigen mRNA (320 n) (4; Fig. 1).In addition, a 1,900-n bandwas oftenseen,

which arises dueto the 3'overlapofearlyand lateSV40mRNA's whichcausesprobe

displace-ment (1). UnderDNA probe excess hybridiza-tion conditions, the relative band intensities of

the 580- and 320-n segments corresponded to

EcoRI

0.14 BamHI

0.17

Hpa110.73-P

Orep

Q.67

BglI

TaqI

FIG. 1. Map of the SV40 genome. The heavy outer circle represents the viral DNA with the restriction endonucleasecleavage sites indicated by arrows.

O,,p

denotes the origin ofDNA replication, the site at which large T-antigenbinds(20, 31) and the presumedsiteofitsregulatory control on early RNA synthesis (2, 15, 21). Thetriangles mark the positions of RNA endpoints (0.15 and 0.17m.u)and splice junctions(0.59,0.54,

0.76,and 0.94m.u.).The counterclockwise arrows indicate the map locations of the two major polyadenylated

cytoplasmic mRNA's found early in viral infection, i.e., before the onset of viral DNA synthesis. The early RNA coding for large T-antigen has a leader which extends from 0.660 to0.600 m.u., spliced to a "body" extendingfrom0.533to 0.154 m.u. The other early RNA, coding forsmall t-antigen, consists of a larger leader

(0.660to0.546m.u.),spliced to the same body as the largeT-antigen mRNA. The clockwise arrows indicate

themap locationsof the additional polyadenylated cytoplasmic mRNA 's found late ininfection(i.e., after the onsetof viral DNAreplication), the 19S and168RNAs, whichcode for the viral structural proteins. Map positionsarederivedfrom references 4, 7, 9,16, 18, and 19.

http://jvi.asm.org/

[image:3.508.128.374.329.546.2]

(4)

cv-1

A W-T A58 A58 WT M 32 Shift 32 Shift 32

AGMK

WT A58 A58 Shift 32 Shift 2145

1867

1171 1087 828 528

447X

[image:4.508.270.458.509.567.2]

367 .LX .i

FIG. 2. Nuclease SI analysis of SV40 cytoplasmic

earlymRNA. TotalcytoplasmicRNA isolated from CV-1orAGMK cellsinfectedwith WTVA4554(WT)

ortsA58 virus washybridizedto theearly (minus) strand ofthe larger fragment of 32P-labeled SV40 DNAgeneratedby digestionwith therestriction

en-donucleases HpaIIand BamHI (0.14 to 0.73 m.u.

[Fig. 1]). Samples were hybridized, nuclease SI treated, electrophoresed, and autoradiographed as

described in the text. The 32°Csamples (32) were

harvested after40 hofinfectionat32'C. Theshifted samples(shift) wereharvestedaftera5-hshift from 32to41'C.TheinputamountsofRNAvariedamong samplestomaintainSV40 DNAexcesshybridization

conditions.InputRNAamountswere asfollows: 64

and 54pg,respectively, for WT-infectedCV-1 cellsat

32'Candshift;22 and 5 pg,respectively, for

tsA58-infected CV-1 cellsat32°Candshift; 14and 9pg,

respectively, for WT-infectedAGMKcells at 32°C andshift;and 7 and2.5pgfortsA58-infectedAGMK cellsat32'C andshift.LaneM containsSV40DNA fragmentmarkers. Similar resultswereobtained with

polyadenylated cytoplasmic RNAs. Itwill be noted that insomeofthesamples, the2,000-nband isnot as intense as would be expectedfrom comparison with the 580- and3(0-n bands. Thisoccurs insome samplesduetothelonghybridizationtimesnecessary

to obtain complete hybridization reactions. This causes some breakdown ofthe RNA, having the greatest effect on the detection oflargersegments.

This didnotoccurin most ofthe samplesused to

determine the ratio oflarge T- to small t-antigen

mRNA's. The data abovewerechosenfor reproduc-tion in thispaperbecausethey clearlyshow allofthe

results discussedon onegel.

the relativeamountsofthesmallt-andlarge

T-antigenmRNA's,respectively. Inspectionof the

relative intensitiesofthesebands inFig.2

sug-gested thatafterashift to410C,the amount of

large T-antigenmRNA relative tosmall

t-anti-genmRNA increasedsignificantly, both in cells

infected by tsA58and in cells infected by WT

SV40.A similar shift towardproductionoflarge

T-antigenmRNA at thehigher temperaturewas

seen inCV-1 cellsinfected eitherbyWT SV40

orbytsA58(Fig. 2). For thequantitation of the

relative amounts of the early SV40mRNA's,

autoradiogramssimilarto those shown in Fig.2

wereanalyzed bymicrodensitometry (Table 1). Although therewas two to three times asmuch large T-antigen mRNA as small t-antigen

mRNA at320C, this ratioincreasedto between

five andseven timesduring shift-upconditions

forbothWT- andtsA58-infectedcultures. When

infected AGMK cells were incubated

continu-ouslyat410C (41°C samples),theratio of large

T-to smallt-antigenmRNA's was even greater

(8-to11-fold).

The experiments aboveshow that there was asubstantial increase in theratio oflargeT- to

smallt-antigen mRNA's withanincreasein the

temperatureofincubation. Itcan be argued that,

during the course of a lytic infection at any

temperature, the ratio oftheearly mRNA's may change with time. Thus, in the shift from 32to

410Cthe lytic process may have been

acceler-ated to a later phase with ahigher ratio of large

T- tosmall t-antigen mRNA's. Likewise,the

24-hinfectionat410Cmayrepresent an even later

phase of infection. To determine whether this

wasthesourceof the ratio changes,AGMK cells

wereinfected withWT SV40 for various times

at 32, 37, and 410C; the ratios of cytoplasmic early mRNA's were then determined as de-scribed above. Table 2 showsthat the specific early mRNA ratio for anyof thetemperatures

remained the same from early tolate times of infection atthat temperature. Inaddition, the ratio ofearly mRNA's adjustedtothe expected values in temperature shift-up and shift-down experiments (datanotshown). Thus, the ratio oflargeT- tosmallt-antigenmRNA's appeared

to be an intrinsic property determined by the

thermal environment of theinfectedcell. Since

the temperature shift experiments (Table 1)

TABLE 1. Ratiooflarge T-tosmall t-antigen

mRNA'sin WT- andtsA58-infectedcells under various temperatureconditionsa

Infecting AGMKcells CV-1cells

virus 320°Cb Shiftc 41oCd 32oCb Shifte 41 oCd

VA4554 2.1 5.7 8.8 2.4 5.2 ND'

(WT)

tsA58 2.7 6.6 10.7 2.8 6.3 ND

aRatiosweredeterminedbymicrodensitometertracingof the 580- and 320-n bandsofautoradiographssuchasthose shown inFig.2.The valuesshownarethe average of at least twodeterminations andaremultiplied by1.72 tocorrectfor the difference in sizesof the leadersegmentsforlargeT-and

smallt-antigenmRNA'sdetermined from nucleotide sequence

data(18, 19).Allnumbers haveastandarderrorof±0.20.

bInfectionat32'C for40h.

'Infectionat32'C for 40 h witha5-hshift to 41'C.

dInfection doneentirelyat41°Cfor 24h.

'ND,Not determined. Bzses

J. VIROL.

ffT

http://jvi.asm.org/

(5)

showed that theearly mRNA ratios varied with

temperature identically in WT- and

tsA58-in-fected cells,wealsoconclude thatlarge

T-anti-gen (atleastthe portion affected bythe tsA58

mutation) and overproduction ofearlymRNA have no significant effect on the earlymRNA

ratio.

Effect ofthermally induced alterationsof early mRNA ratiosonthesynthesisoflarge T- and small t-antigens. We nextwanted to

determine whether the temperature-dependent alterations in the early mRNA ratio were

re-flected atthe proteinlevel. To determine this, AGMKorCV-1 cellswereinfected with WTor

tsA58 at (i) 320C for 40 h; (ii) 320C for40 h, followedbya5-hshiftto410C;and(iii)410Cfor

24 h. During the last hour of each infection period, the cells were pulse-labeled with [3S]-methionine. After extraction and immunopre-cipitationwithanti-T-serum,theearly proteins

were analyzed by sodiumdodecyl sulfate-poly-acrylamide gel electrophoresis (Fig. 3). As

ex-pected from a previous study (26, 29), tsA58-infected cells synthesized morelarge T-antigen after ashiftto

410C.

Small t-antigen synthesis

wasalso increased in these cells. In theexposure

of the gel shown in Fig. 3, the small t-antigen

bandswere notvisible in the WT-infected CV-1 cellsamples; however, theywereeasily detected

uponlongerexposures (datanotshown). Fora

determination of theratio oflargeT- tosmall

t-antigens inthesesamples, the bands containing

the pulse-labeled polypeptides were excised from the gel and counted as described above. Table 3shows theratio of large T- to small

t-antigens determined from thecountsin the gel bands; these ratios have been corrected forthe

difference in the number of methionines in the

twoproteinsasdetermined from thenucleotide

sequence(9, 19).Theratios of theearly proteins

[image:5.508.53.244.510.640.2]

atthe variousincubation temperatures variedin a fashion similartothe ratio of their mRNA's.

TABLE 2. Ratios of large T- tosmallt-antigen mRNA's inWT-infectedAGMK cellsfromearly to

latetimesofinfection at 32, 37, and41C°Ca

320C 370C 410C

Time Time Time

postin- Rto postin- Rto

postin-Rai

fection

ftfection

fection

(h) (h) (h)

12 2.1 8 4.3 6 8.8

24 2.4 21 4.4 12 8.5

48 2.3 40 4.2 24 8.1

aRatiosweredetermined asdescribed in Table 1, footnotea, and thetext. AGMK cellswere infected withWTVA4554 at 32, 37, and 41°C at 10PFU/cell. Atthe notedtimespostinfection,thetotalcytoplasmic RNA washarvestedandanalyzed.

Thus,theresults show that thethermaleffects ontheearlymRNA ratios were reflected at the

proteinlevel.

DISCUSSION

The absoluteamountsofSV40 earlymRNA's

in WT- andtsA58 mutant-infected cells at

dif-ferent temperaturesarenot thesame(2, 15,21); however, the relative ratio oflarge T-tosmall

t-antigen mRNA'svariedsimilarlywith

tempera-ture in both WT- and tsA58 mutant-infected

cells (Table 1). Thus, it appears that large

T-antigen doesnotinfluence the determinationof

the ratio of thetwoearly mRNA's.Although it

isconceivable that the tsA mutantlarge T-an-tigen maybetemperature insensitive for sucha

regulatory function, this appearsunlikely. For example,Tegtmeyeretal. (29) have shown that

the tsA large T-antigen is rapidly degraded at

the nonpermissive temperature. It seems im-probable thatafragment of themutantprotein

could maintain WTactivityforthis function.In

addition, Handa and Sharp (personal

commu-nication) have shown that the ratio of thetwo

early SV40 mRNA's at 370C isnot altered by

the presenceofinhibitors such as emitine and

cycloheximidewhichblock the synthesis of the earlyproteins. Weconclude, therefore,thatlarge T-antigen controlsearlygeneexpression primar-ily by generalregulation of allearly RNA

tran-scription,presumably byfunctioningonthe

pro-moter. The relativeproportions oflargeT-and

small t-antigen mRNA's appear to be

deter-mined by otherfactors whichareunaffectedby thetsA mutationortheoverproduction of early

RNAin tsAmutantinfections.

Thethermally induced variationsintheearly mRNA (andprotein) ratiosmay occur atseveral levels. Recent sequencing studies (18) have showntwodifferent5'terminifor theearly SV40 mRNA's(within7 nof eachother).Althoughno

evidencesuggeststhat either terminusisspecific for one of the early mRNA's, these data may indicate that there are two early promoters.

Thus, thermal factors may alter the early

mRNAratiobyinfluencingpromoterselection. Anotherpossible mechanism foraltering the ratio of theearly mRNA's derives from the fact thatbothearly mRNA's share thesamegenomic

sequences,differing onlyintheirsplice positions.

Thus, the thermally induced alterationsin the

mRNA ratios mayinvolve the relatively small

sequencedifferences whichdetermine large

T-and small t-antigen mRNA's (in the 0.534 to

0.60-m.u. region [Fig. 1]).Forexample, thermal

effects could alter theactivityof anuclease(s),

thus differentially affecting the early mRNA

stabilities, or alter the transport mechanism

such that the rate oftransport of specific mRNA

http://jvi.asm.org/

(6)

CV-1 WST 32Shift 41

CV-1

A58 32 Shift 41

AGMK WT 32Shift41

AGMK A58 32Shift 41

T

VP1 __ _ FW

Mock CV-1 AGMK

MW X10)-3 92.5 69

46

30

t _,_ _;_

12.3

FIG. 3. Examinationofpulse-labeledSV40earlyproteins. Cultures of CV-1 or AGMK cells were infected with WTVA4554 (WT)ortsA58virus. The infection conditions were as follows: (i) 32°C for 40 h (32);(ii) 32°C for40h,followed by a 5-h shift to 41 °C (shift); (iii) 41 °C for 24 h (41). During the last hour of infection at the various temperatures, the proteins were pulse-labeled with [35SJmethionine asdescribed in the text. The labeledproteinswereextracted, immunoprecipitated and analyzed on a sodium dodecyl sulfate-polyacryl-amide gel (see text) with appropriate markers. Thepositionsof large T-antigen (T) andsmall t-antigen (t) are indicated. Inaddition, the serum used in this experiment precipitated the viral coat proteins (predominantly VP1). Note that the 41 °C samples from cells infected with tsA58 contain little or no VP1, as expected for a tsA mutantinfection.

TABLE 3. Ratiosof large T- tosmall t-antigen proteins in WT- andtsA58-infectedcells atvarious

incubationtemperaturesa

Infecting AGMK cells CV-1cells

Virus 320C Shift 410C 32°C Shift 410C

VA4554 2.5 4.6 5.5 1.8 3.5 4.9 (WT)

tsA58 2.1 8.6 10.6 1.8 4.7 6.0

aRatiosweredetermined fromcountsinpulse-labeled im-munoprecipitated large T- and small t-antigens after

separa-tionon asodiumdodecylsulfate-polyacrylamide gel (Fig.3). Theratiosarecorrected fordifferences innumbers of methi-onines between the two proteins asdetermined from the nucleotidesequence(9, 19).

species changes. It seems likely that the two

early mRNA'sare derived froma common

pri-mary transcript of the early coding region by

different splicing events. With this

considera-tion, the thermal effects on the early mRNA

ratiosmayresult fromanalterationofsequence

recognition by a splicing enzyme(s) such that

therelativefrequency ofsplicingeventschanges

orfromanalteration ofthestability ofsecondary

structuresintheprimary RNA transcript which

arerequiredforthesplicingevents,thus

chang-ing theequilibriumofsplicereactions.

Theselattertwopossibilitiesaresupported by previous studies (2, 21) whichindicate that the total amount of early RNA (or precursor) is

relatively equal in the cytoplasm and nuclei of WT- infected cellsat320C and aftera shift to 410C.Thissuggeststhattheratiochangesoccur

by qualitative variation within a fixedamount

ofprecursor. Inaddition, examination of deletion

mutants in the 0.535- to 0.60-m.u. region, the

unique codingregionforsmallt-antigen(Fig. 1), hasyielded information aboutpossiblesequence

requirements for thesplicing ofearly mRNA's

(14; G. Khoury,J. C.Alwine,R.Dhar,P.Gruss, C.-J. Lai, S. Segal, and I. Seif, Cold Spring

Harbor Symp. Quant. Biol., in press). None of

theseviable deletionmutantsaffects theability

to produce the large T-antigen mRNA splice.

Since these mutants include deletions which

removethe smallt-antigenmRNAsplicedonor

sequenceat0.546 m.u.(18,24),itisunlikelythat

thesmallt-antigenmRNAspliceisa

prerequi-site forlargeT-antigenmRNAsynthesis.Onthe

otherhand, specific deletionswithinthisregion

which do not involvesuspected splicejunctions

willgreatlydecrease the amount ofstable small

J. VIROL.

I.i,, 1: I

PI

ammm

41A R4:%:::--:..,'

http://jvi.asm.org/

[image:6.508.122.406.80.307.2] [image:6.508.64.257.415.494.2]

(7)

t-antigen mRNA produced. This indicates that specificsequences,well removed from thesplice

junctions, maybeimportantfor this splice (14;

Khoury et al., in press). It has been suggested

fromthese data that these deletions mayalter conformations inprimarytranscripts whichare

required for the splice. If this is the case, it is possible that thermal (or thermodynamic) ef-fects whichalter the stabilities of these confor-mationschange theratesofsplicngeventsand, hence, the ratios oftheearly mRNA's. Another indicationof theimportance ofsequence in

splic-ing isnoted in theobservationthatmanyRNA's

are spliced identically in severaldifferent cells (Khoury et al., in press). Our examination of early mRNA's in SV40 (VA4554)-transformed hamstercells demonstrates thatnotonlyarethe splices thesame, but alsothe ratios of theearly mRNA'svarywithtemperatureinthesame way as in lytically infected, monkey cells (data not

shown).

Although these data indicate thatinformation within the sequence may be involved in the

determination oftheearly mRNA ratio,we

can-not rule out a high degree of conservation of

splicing factors between species. In this regard, therequirement for factors other than primary

RNAsequenceisimplied by the demonstration

ofcomplete unspliced early transcripts in SV40-infectedmouseteratocarcinomacellsbefore

dif-ferentiation; after the induction ofcell differen-tiation by retinoic acid, correctly spliced early

mRNA's are found (22, 23). This implies that

factors may be induced during differentiation which directly or indirectly affect splicing of

RNAmolecules.

Thus, there are several possible levels at

whichtheenvironment ofthecellmay influence

the early mRNA ratios. Whatever the

mecha-nism, these changesinearlymRNA ratios

indi-cate apotentially importantlevel forcontrol of

geneexpression.In ageneral model,this control may allow aDNA sequence which encodes

in-formation formore than oneproductto

differ-entially express the products by altering the

processing which produces the final mRNA's.

Thisdifferentialexpression may be determined

bytemperature or otherthermodynamic

varia-bles(pHandionicstrength) of the environment ofthecell.

LITERATURE CITED

1. Alwine, J. C., R. Dhar, and G. R. Khoury. 1980.A

small RNA induced lateinSV40lytic infection can associate with early viral mRNAs. Proc. Natl. Acad. Sci. U.S.A. 77:1379-1383.

la.Alwine, J. C., S. I. Reed, J. Ferguson, and G. R.

Stark. 1975.Propertiesof Tantigens inducedby wild typeSV40 andtsAmutants in lytic infection.Cell6: 529-533.

2. Alwine, J. C., S. L. Reed, and G. R. Stark. 1977. Characterization of the autoregulation of simian virus 40gene A. J.Virol. 24:22-27.

3. Aviv, H.,and P.Leder. 1972. Purification of biologically active globinmessenger RNA by chromatography on oligothymidylic-acid-cellulose. Proc. Natl. Acad. Sci. U.S.A. 69:1408-1412.

4. Berk, A.J., and P. A. Sharp. 1978. Spliced early mRNAs of simian virus 40. Proc. Natl. Acad. Sci. U.S.A. 75: 1274-1278.

5. Cowan, K., P. Tegtmeyer, and P. P.Antony. 1973. Relationship of replication and transcription of simian virus 40 DNA. Proc. Natl. Acad.Sci. U.S.A. 70:1927-1930.

6. Crawford, L. V., C. N. Cole, A. E. Smith, E.Paucha, P.Tegtmeyer, K. Rundell, and P. Berg. 1978. Or-ganization and ezpression ofearly genes of simian virus 40.Proc.Natl. Acad. Sci. U.S.A. 75:117-121. 7. Dhar, R., K. N. Subramanian, J. Pan, and S. M.

Weissman. 1977. Nucleotide sequence of afragment of SV40 DNA that contains the origin of DNA replication andspecifies the 5'-ends ofearly and late viral RNA. J. Biol.Chem. 252:368-376.

8. Ferdinand, F.-J., M. Brown,and G. Khoury. 1977. Synthesis and characterization of late lytic simian virus 40RNA from transcriptionalcomplexes. Virology 78: 150-161.

9. Fiers, W., R.Contreras, G. Haegeman, R. Rogiers, A. Vande Voorde, H. VanHeuverawyn,J. Van

Her-reweghe, G. Volckaert, and M. Ysebaert. 1978. Complete nucleotide sequence of SV40 DNA. Nature (London) 273:113-120.

10. Hayward,G. 1972. Gel electrophoretic separation of the complementary strands of bacteriophage DNA. Virol-ogy 49:342-347.

11.Jay, G.,F. T.Jay,R. M.Friedman,andA. S. Levine. 1977. Simian virus 40-specific ribosome-binding pro-teins induced by a nondefective adenovirus2-simian virus 40hybrid. J. Virol. 23:692-699.

12. Jay,G.,R. P.C.Shiu,F. T.Jay, A. J.Levine, and I. Pastan. 1978.Identification of a transformed specific protein induced byaRous sarcoma virus.Cell 13:527-534.

13. Khoury,G.,B. J.Carter,F.-J.Ferdinand,P. M. How-ley, M. Brown, and M. A. Martin. 1976. Genome localization of simian virus 40 RNA species. J. Virol. 17:832-40.

14. Khoury, G.,P.Gruss,R.Dhar,and C.-J. Lai. 1979. Processing andexpressionofSV40mRNA: theeffect of

earlydeletions.Cell18:85-92.

15.Khoury,G.,and E.May.1977.Regulationofearlyand late simian virus40 transcription: overproduction of

earlyviral RNA in the absence ofafunctional T-anti-gen. J.Virol. 23:167-176.

16.Lai, C.-J., R.Dhar,andG.Khoury.1978.Mappingthe

splicedandunsplicedlatelytic SV40RNAs.Cell 14: 971-982.

17.Laskey,R.A.,and A. D.Mills.1975.Quantitativefilm detection of 3H and"4Cinpolyacrylamidegelsby fluo-rography. Eur. J. Biochem. 56:335-341.

18.Reddy,V.B.,P.K.Ghosh,P.Lebowitz,M.Piatak,

and S. M. Weissman. 1979. Simian virus40 early

mRNA's.I.Genomiclocalization of 3' and 5'terminiof two majorsplices in mRNA from transformed and lytically infectedcells. J. Virol. 30:279-296.

19.Reddy,V.B.,B.Thimmappaya,R.Dhar,K. N. Sub-ramanian, B. S.Zain,J.Pan,P. K.Ghosh,M.L. Celma,andS.M.Weissman. 1978.Thegenomeof simian virus 40.Science206:494-502.

20. Reed,S.I.,J.Ferguson,R.Davis,andG. R. Stark. 1975.T-antigen bindstosimianvirus 40 DNAatthe originofreplication. Proc. Natl. Acad. Sci. U.S.A. 72: 1605-1609.

http://jvi.asm.org/

(8)

21. Reed, S. I., G. R. Stark, and J. C. Alwine. 1976. Autoregulation of simian virus40geneA byT-antigen.

Proc. Natl. Acad. Sci. U.S.A. 73:3083-3087.

22. Segal, S., and G. Khoury. 1979. Differentiation as a

requirement forSV40geneexpression in F-9 embryonal

carcinomacells. Proc. Natl. Acad.Sci. U.S.A. 76:1746-1751.

23. Segal, S., A. J. Levine, and G. Khoury.1979.Evidence fornon-spliced SV40 RNA in undifferentiated murine teratocarcinomastemcells.Nature(London) 280:335-338.

24. Seif, I.,G. Khoury, and R. Dhar. 1979. BKVsplice

sequencesbased on analysis of preferred donor and

acceptorsites. Nucleic Acids Res. 6:3387-3398. 25.Tegtmeyer, P. 1972. Simian virus40 deoxyribonucleic

acidsynthesis: the viral replicon. J. Virol. 10:591-598. 26.Tegtmeyer, P. 1974. Alteredpatternsofprotein synthesis

in infections by SV40mutants. Cold Spring Harbor

Symp. Quant. Biol. 39:9-16.

27. Tegtmeyer, P., C. Dohan, and C. Reznikoff. 1970. Inactivating and mutagenic effects of nitrosoguanidine

onsimian virus40.Proc.Natl. Acad. Sci. U.S.A.66:

745-752.

28. Tegtmeyer, P., K.Rundell, and J. K. Collins. 1977. Modification of simian virus40proteinA.J. Virol.21: 647-657.

29. Tegtmeyer, P., M. Schwartz, J. K. Collins, and K. Rundell.1975.Regulation oftumorantigen synthesis of simian virus40geneA. J. Virol. 16:168-178. 30. Tenen, D. G., P. Baygel, and J. Livingston. 1975.

Thermolabile T (tumor) antigen from cells transformed byatemperaturesensitivemutantof simian virus40. Proc. Natl. Acad. Sci. U.S.A.72:43514355.

31.ITian,R. 1978. Thebinding siteonSV40 DNA foraT antigen-related protein. Cell 13:165-179.

http://jvi.asm.org/