Detection and characterization of multiple forms of simian virus 40 large T antigen.

0022-538X/81/010092-1 1$02.00/0

Detection and

Characterization of

Multiple

Forms of

Simian

Virus 40 Large T Antigen

E. FANNING,* B. NOWAK, AND C. BURGER

Fakultatfur Biologie, Universitat Konstanz, D- 7750Konstanz, Federal Republic of Germany

Subclasses of simian virus 40 large T antigen in simian virus 40-transformed

and -infected cellsseparated by zone velocity sedimentation in sucrose density

gradients have been characterized. Three forms oflargeT antigen were

distin-guished: a 5 to 6S form, a 14 to 16Sform, anda 23 to 25S form. These forms appeared to differ biochemically and biologically. Differential labeling

experi-mentssuggestedthat the 5to6S form waslesshighly phosphorylatedthan the

faster-sedimenting forms. The 23to25S form whichwas

complexed

withone or

morehostphosphoproteins, asreported recently (D.P. Laneand L. V.Crawford,

Nature[London] 268:261-263, 1979; F. McCormickand E. Harlow,J.Virol. 34:

213-224, 1980), was prominent in extracts of transformed cells, but was also

detected inproductively infected cells. Pulse-chase experiments suggested that the5 to6SlargeTantigenisaprecursorof themorestable, faster-sedimenting

forms of Tantigen. Monkeycells infected withatsAmutantof simian virus 40 at

410C

containedonly5 to6SlargeTantigen,implyingthat thisform is not active

in the initiation of simian virus 40 DNA replication. In pulse-chase, shift-down

experiments, DNA

replication

resumed, and the 5 to 6S

large

Tantigenwhich hadaccumulated at

410C

was

partially

converted at33°C to afast-sedimenting form. However, shift-up experiments demonstrated that the fast-sedimenting largeTantigen,once

formed,

remainedstableat

410C, although

itwasunableto

function in initiation. These experimentssuggest that different biological func-tions oflargeTantigenmaybe carriedoutby different subclasses ofthisprotein.

Simian virus 40

(SV40)

DNA codes for two early proteins, the large and small tumor (T) antigens (22,30, 33), whichcanbe

immunopre-cipitated from extractsof cells

productively

in-fectedortransformed

by SV40,

using

anti-SV40

tumor serum (38).Althoughthefunctionof the

small T antigenisstill uncertain (23),multiple functionsinwhich thelargeT antigenis

essen-tial have been demonstrated, among them the

initiation of each round of viral DNA

replication,

the stimulation of late gene transcription and attenuation ofearly gene

transcription

in

pro-ductively infected cells, and the initiation and

maintenanceoftransformation (reviewedin

ref-erence12). It is not at all clear whether thelarge

T antigen

performs

these varied functions

di-rectlyorwhether it must interact with host cell factors to fulfill its functions.

Recently,another class ofpolypeptides with

molecularweightsof50,000 to 55,000 (-50 to 55K)

has been specifically immunoprecipitated from

extracts of SV40-transformed (7, 9, 25) and

-infected cells (25, 26). An increasing body of

evidence suggests that theseproteinsare

prob-ablycodedby thehostgenome(5, 15, 18, 20, 25,

26, 35) and may be induced or stabilized in

transformed cells (9, 20, 21). The interaction

between these host

proteins

and the

large

T

antigen couldconceivably modifytheproperties of the host proteins or those of the T antigen itself.

Anotherpossibility which might beconsidered

is thatthe

multiple

functionsin whichthelarge

Tantigen isinvolved may really be performed

by different subclasses oflarge Tantigen. This

communicationdescribestheidentification and

characterization of several such subclasses of

large T antigen and their possible correlation

with some of thefunctionsof this protein.

(Apreliminary account of this work was

pre-sented at the Imperial Cancer Research Fund DNATumorVirus Meeting in Cambridge, Eng-land, in July 1979.)

MATERIALS AND METHODS

Cells and virus. SV40-transformed mouseSVT2 (1),hamsterTHK-1, (3), monkeyC2 (10), and human SV80 (39) cell lines have beendescribed previously. SubconfluentCV-1P monkey cells were infected with

1to 5PFUofSV40,strainSVS, per cell or with the

mutant tsA58asdescribed previously (2, 8). All cell lines were cultured in Dulbecco-modified Eagle me-dium containing 10% calfserum (Boehringer, Mann-heim,Germany).

Antisera.Initialgiftsofhamster anti-SV40tumor

serum wereobtained from H.Fischer,German Cancer Research Center, Heidelberg, Germany, and J.

92

on November 10, 2019 by guest

http://jvi.asm.org/

HETEROGENEITY OF SV40 LARGE T ANTIGEN 93

Gruber, National Institutes of Health, Bethesda, Md. Larger amountsof tumor serum were prepared from hamsters carrying tumors induced by subcutaneous injection of THK-1, cells and from BALB/c mice injectedsubcutaneously or intraperitoneally or both with SVT2 cells. Normal sera were obtained from healthy adult mice and hamsters. Rabbit antiserum against isolated, denatured large T antigen was a gift fromW.Deppert,University of Ulm, Ulm, Germany. Monkeyanti-U serum (19) was obtained from A. M. Lewis, Jr., National Institutes of Health, Bethesda, Md.

Labeling and extraction of proteins. Confluent monolayers were starved for1 hinmethionine- and serum-free medium and then labeledwith 100,iCi of

[0S]methionine (Amersham Buchler, Braunschweig,

Germany) per ml in methionine-free medium. Alter-natively, the cells were incubated in phosphate- and serum-free medium for 1hand labeled for 2hwth

100yCi of

'PO4

(Amersham Buchler) per ml in phos-phate-free medium. In some experiments, the cells werelabeled for14 to 15hwith20to 30,uCi of[3H]

leucine per ml and40 to 50t,Ci of0"PO4 per ml in leucine- andphosphate-free medium. At the endof thelabelingperiod, the cellswerewashed threetofive times withcold Hanks balanced saltsolution, scraped from the plates, and pelleted. The cells were

sus-pended(106 cells perml)in50mMTris-hydrochloride

(pH 8)-120 mM NaCl-0.5% Nonidet P-40 (34), ex-tracted for1h at0°C,andpelletedagain. The super-natant wasuseddirectly for immunoprecipitation or storedat-70°Cin smallsamples.Reextraction of the nuclear pellet accompanied by ultrasonic treatment failed to yield additional T antigen in significant amounts.

Immunoprecipitation. Immunoprecipitation with anti-SV40 tumorserum and normal serum was per-formedasdescribedpreviously (34), except calf serum wassubstituted for sheepserum.Formaldehyde-fixed Staphylococcus aureus, strain Cowan I, was prepared asdescribed previously (13) and stored at -20°C in smallsamples.

SDS-polyacrylamide gel electrophoresis. So-diumdodecyl sulfate (SDS)-polyacrylamide gel elec-trophoresis was carried out as described previously (17).Immunecomplexeswereeluted from the washed bacterialpelletsbyboilingfor3min in20 to 50 LI of

double-strength sample buffer. Electrophoresis was

carriedoutat40mApergelfor2.5to3hwithcooling. The gelswere stained for2 h with 0.1% Coomassie brilliant blueinmethanol-water-acetic acid (20:17:3),

destained, dried, and autoradiographed or

fluoro-graphedonKodak XR-5 film. Proteins with the fol-lowing known molecularweightswereusedtocalibrate

thegels:,8-galactosidase, 130K;Escherichia coliRNA

polymerase ,B',f,, and a, 165, 155, and 39K, respec-tively;phosphorylasea,94K; fructose-6-phosphate ki-nase, 81K; bovine serum albumin, 68K; glutamate dehydrogenase, 53K; alcoholdehydrogenase, 37K;and soybeantrypsin inhibitor,21.5K.

Zone velocity sedimentation of nuclear ex-tracts. Cellmonolayerswerelabeledand washedas

describedabove,rinsedoncewith10mMHEPES

(N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid;

pH 7.8)-5mMKCI-0.5 mMMgCl2-1mM dithiothre-itol-1 mM phenylmethylsulfonyl fluoride (bufferA),

and drained. Cells were scraped from the plates in bufferAand disrupted in a Dounce homogenizer by 10strokes with a tight-fitting Teflon pestle. The nuclei werepelleted, resuspended in buffer A(10'nuclei per ml), and incubated for 1 h at 0°C. The nuclei were pelleted, and the supernatant extract (0.2 ml) was layered onto a5 to 30% sucrose density gradient in bufferA containing5 mM KCI. Similar results were obtainedwhen the gradients contained0.5% Nonidet P-40 or 150 mM KCI instead of 5 mM KCL. The gradients were centrifuged in an AH650 (Sorvall) or SW55 rotor (Beckman) for 180 to 190 min at 45,000 rpm and0°C.SV40 form I DNA (20S) and ,8-galacto-sidase(16S) were run inparallelgradients as markers. Fractions werecollected from the bottom. Acid-pre-cipitable radioactivity was determined ina

10-pl

sam-ple of each fraction, and the rest of the material was analyzed by immunoprecipitation and SDS-polyacryl-amide gel electrophoresis. In many experiments,

50-pl

samples of the extract loaded onto the gradients wereimmunoprecipitated with tumor serum and nor-mal serum and run on the samegels.

Zone velocity sedimentation of whole-cell ex-tract.After the cellswerelabeledasindicated in each figure legend, theywerewashed three times with cold Hanks balanced salt solution, scraped from the plates,

andpelleted.Theyweresuspended (10' cellsperml)

in50mMTris-hydrochloride (pH 8)-120 mM

NaCl-0.5%Nonidet P-40 (34), extracted for 1 hat0°C,and

pelleted again. Thesupernatantwasloaded directly

onto sucrosegradients (0.2 ml per gradient)and ana-lyzedasdescribed abovefornuclearextracts.

RESULTS

Detection oftumor

antigens

in infected and transformed cells used in this

study.

SV40-transformed human, mouse, and monkey cells,aswellasinfected and uninfectedmonkey cells, were labeled with

[:3S]methionine.

Pro-teins extracted from eachcell line were

immu-noprecipitatedwithanti-SV40tumorserumand

normalserumand

separated by

SDS-polyacryl-amidegel electrophoresis. ThelargeT

antigen,

about 90K (34, 38), and the small T antigen, about 17K (30), were detected in each of the

transformed andinfectedcellsexcept,

curiously,

this strain ofSVT2

(Fig.

1).The

large

T

antigen

from SVT2 cells was slightly larger than that

from the othercells, as shownpreviously

(31).

Limited

degradation

of

large

T

antigen

(34)

could notbe eliminated

entirely

andwas

espe-cially

evidentinextractsof

productively

infected

cells (Fig. 1).

Hostproteinsof 50to55K(5, 9, 15, 18, 20, 21,

25,35)were

specifically

precipitated

from these

transformedcells

by

tumor serum.Theirrelative amounts and apparent molecular

weights

de-pended upon the animal

species

of the

trans-formed cell, as

reported

recently

(7, 20, 25).

These proteins were not observed in infected

cells in this experiment (Fig. 1), but were

de-tectable in extracts of

'2P04-labeled

infected

VOL. 37,1981

on November 10, 2019 by guest

http://jvi.asm.org/

94 FANNING, NOWAK,

SV40O

SVT2

SVBO

C2

CV-1P

CV-1P

m

1

m

m

II

m

T

N T N T N

T

N

T N

FIG. 1. SDS-polyacrylamide gel fluorograms of

[35S]methionine-labeled T antigens from

SV40-in-fected and-transformedcells. Proteins were immu-noprecipitatedfrom whole-cell extractsof SV40-in-fected (48 hafter infection),SVT2,SV80,and C2 cells

and uninfected CV-1P cells with normal hamster

serum (N) and anti-SV40tumorserum (T)as indi-cated. Thepositions ofmarker proteins of known molecular weights are shownon the right. Thegel contained 12.5%acrylamide.

cells (seeFig.30). Others have also found small

amountsof 54Kproteininnontransformed cells

(11, 20,26).

Characterization of hamstertumorsera.

Some anti-SV40tumorserahavebeen shownto

reactwith55Khostproteins specifically (5, 20,

21), whereas othertumor serawhichareunable

to recognize the host

proteins

precipitate them

as acomplex with the

large

Tantigen(18). The

tumor serum used in thisstudywas

character-izedin twoways:

(i)

its

capacity

to

re-immuno-precipitate denatured hostproteinswas tested,

and (ii) an attempt was made to differentially

dissociate the host proteins and the large T

antigen from immune complexes.

Whole-cell extracts of

:12P04-labeled

SV80

cells were immunoprecipitated with anti-SV40

hamster tumor serum. The immune complexes

were dissociated from the bacteria by boiling

first

insample buffercontaining 0.1% SDS(Fig.

2b) and then reboiling the bacterial pellet in

samplebuffercontaining 1% SDS (Fig. 2a) or by

boilingthesampleasusual(Fig. 2c). Treatment with sample buffer as usual completely

disso-ciated the immune complex, freeing both the

hostproteinsand thelargeTantigen.Treatment

with 0.1% SDS, on the other hand, dissociated

only the host proteins from the immune

com-plex. The large T antigen was recovered after

treatmentwith 1% SDS.Apreparation

contain-130

ing both host proteins and thelarge Tantigen

-94

was diluted and immunoprecipitated a second

:81

time with thesametumorserum.Thedenatured

68

large

T

antigen

was

precipitated,

but the host

-53 proteins failed to react with the tumor serum

-37 (Fig.2d).Theseresultsindicatethatthe tumor

serausedinthisstudy hadastrongaffinity for

the large T antigen but only weak

affinity,

if any,for55Kproteins.

Nuclear extracts were preparedinhypotonic buffer from SV80 cells labeled with

"2P0,4

and

-215

analyzed byzone velocity sedimentation in

su-crosedensitygradients. The gradientswere

frac-tionated, andacid-precipitable radioactivitywas

[image:3.496.62.252.77.292.2]

ab c d

FIG. 2. Separation ofhostproteins and large T antigenfrom SV80cellsby differentialdissociation ofimmunecomplexes. Whole-cellextractsof

32pO4-labeled SV80 cells were immunoprecipitated with hamstertumor serum as describedin the text. The immune complexes were dissociated by boiling in samplebuffer (c)orinsample buffer containing0.1% SDS (b), followed by reboiling in 1% SDS (a). A duplicate sample (c)without2-mercaptoethanolwas diluted in 50 mM Tris-hydrochloride (pH 7.5)-150

mMNaCI-5mMEDTA-0.05% Nonidet P-40to0.02% SDS andre-immunoprecipitated withfreshhamster tumor serum(d). Theimmunoprecipitateswere

ana-lyzed byelectrophoresisinSDS-polyacrylamide gels

andautoradiography.

53-

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.496.298.418.273.515.2]

HETEROGENEITY OF SV40 LARGE T ANTIGEN 95

determined in a sample of each fraction (Fig.

3A). The proteins immunoprecipitated from

each fraction with hamstertumor serum were electrophoresed on SDS-polyacrylamide gels; the gels were stained, dried, and autoradi-ographed. Two peaks of large T antigen were prominent, one at about 14 to 16S (fractions 15

to 17ofFig. 3B) and another at about 23 to 25S

(fractions 12 to 13 of Fig. 3B), which cosedi-mented with most of the 55K protein. Neither the 55K protein nor the large T antigen was immunoprecipitated with normal hamster

se-rum.These twopeaks contained the bulk of the

largeTantigen in the gradients, as indicated by

Coomassie brilliant blue staining and

comple-ment fixation (28; A. Dorn and E. Fanning,

unpublished data), and were also demonstrated

by using extracts from

[:5S]methionine-labeled

SV80 cells and from

32P04-labeled

SV40-trans-formed monkeycells (C2) (data not shown). A

complex from transformed mouse cells

analo-gous to the 23 to25S form has been described

recently by McCormickandHarlow (24).

When the nuclear extracts prepared from

SV40-infected CV-1P cells labeled with

132P04

wereanalyzed in thesameway,onlyonepeak of largeTantigen wasfound (Fig.30). Thispeak sedimented at about 14 to 16S and also

con-tained mostof the 90K

protein

in the

gradient,

as indicated by complement fixation (data not

shown). The 55Kprotein was detected in the 23

to 25S region of the gradient, apparently

co-sedimenting withaheavyshoulder of the main

peak of the large Tantigen (Fig.

30).

At later

timesafterinfection,the 23 to 25S peak

contain-ing T antigen and 55K protein was more

prom-inent andcouldbeclearly separated from the 14

to 16S peak (E. Fanning, C.

Bu

rger, and D.

Corlin, manuscript inpreparation).

Theseresults, together with the fact that the

tumor serum used did not recognize the 55K

protein (Fig.2),suggested thatproductively

in-fected cellsmay also contain a 55K protein-T

antigen

complex analogous

tothatinSV80and

C2cells. Other antisera which donotreactwith 55K protein, such as rabbit antiserum to the

SDS-denatured

large

T antigen or a monkey anti-Userum(18),gaveresultsindistinguishable from thoseinFig. 3(datanotshown).

These results indicate that both transformed

cells andproductivelyinfectedcells containtwo

forms ofphosphorylated

large

T

antigen,

the 23

to 25S

complex

between the 55K

protein

and

the largeT

antigen,

as described

recently

(24),

anda 14 to 16S form.

SV40 large T

antigen

may be

differen-tially

phosphorylated.

In additiontothetwo

forms of

phosphorylated large

T

antigen

in

SV40-transformed and -infected cells

(Fig. 3),

A

0

I

U~

FRACTION NUMBER

B 5

swXww

10 15 20 r N

--- vK

K

:

E t

f ts i *4F

e =

C FRACTION NUMBER

S 1Q w @ " T

J _ _ _ _ wr.,.wv-M- s

-:T-f.,.<4t-...<Ri;:Hia's:,,iff-.:...i

0X.Xt f._ 3 it _

b*.-+%.e i--<m w

.L. ::

-0-; -i.

FIG. 3. Zonevelocitysedimentation ofan extract

from transformedandproductively infectedcells. An

extractfrom32PO4-labeledSV80 nucleiwasanalyzed

ina5to30%sucrosedensity gradientasdescribedin

thetext(AH650, 45,000rpm,0°C,190min). (A)

Acid-precipitable radioactivity was determined in 10-,il samples ofeach fraction. The arrow indicates the

position of,B-galactosidaseruninaparallel gradient asamarker. (B)Eachfractionwas

immunoprecipi-tated with anti-SV40tumorserumandanalyzed by SDS-polyacrylamide gel electrophoresisand

autora-diography.An extractfrom5x whole cellswas

immunoprecipitatedwith normal hamsterserum(N)

and anti-SV40tumorserum(T)andrunonthesame gel. (C)A nuclear extract ofSV40-infected CV-1P cells labeledwith32pO4from 45to47.5 hafter infec-tionwasanalyzedon asucrosedensity gradientas

describedfor (A)and(B). Acid-precipitable

radioac-tivity in

10-t1I

samples ofeachfraction was

deter-mined(datanotshown),and therestofeachfraction

wasimmunoprecipitatedwith anti-SV40tumorserum

andanalyzed by SDS-polyacrylamide gel electropho-resis andautoradiography. The,B-galactosidase16S

peakwasinfraction14. The directionof

sedimenta-tion isfrom righttoleft.

-94 -68

-53 -37

- 1.5

13D -9'

-e

- 53

- 37

215 VOL. 37,1981

on November 10, 2019 by guest

http://jvi.asm.org/

anotherapparentlyless-phosphorylatedform of

T antigen was distinguished by zone velocity

sedimentation. SV40-infected cellswerelabeled

with[ sH]leucine and;2PO4from24 to39h after

infection, and whole-cellextracts wereprepared

and centrifuged in sucrose density gradients.

Eachfraction wasanalyzed by

immunoprecipi-tation with anti-SV40 tumor serum and

SDS-polyacrylamide gel electrophoresis. After the

large Tantigen bands were located by autora-diography,each bandwasexcised from thegel, solubilized, andcounted (Fig. 4A). Thepeakof

highly phosphorylated Tantigenwasobserved

at 14 to 16S, with a fast-sedimenting shoulder

(Fig.4A). AnewspeciesoflargeTantigenwhich

sedimented at 5 to 6S appeared to be

under-phosphorylatedrelativetothe14 to16Sspecies.

Bycalculating the:32P/:3H ratio in eachpeakof

large Tantigen, itcanbe estimatedthatthe 14

to 16S species contained about three times as

muchphosphatepermoleculeasdidthe5 to6S Tantigen. Nodifferencewasobservedbetween the:32P/:3H ratiosof the 14 to 16Speakand the

faster-sedimenting shoulder.

When extracts of SV40-transformed cells (SV80) labeledwith[3H]leucineand;2PO4were analyzed in the same fashion, a small peak of

underphosphorylated large T antigen was

de-tected at 5 to 6S (Fig. 4B). This form of T

antigenrepresentedonlyaminorfractionofthe

total largeTantigen inSV80 cells (10 to15%),

whereas in productively infected cells, it

ac-counted for about 25 to 30% of the total 3H-labeled large T antigen at 39 h after infection.

The ratio ofthe;H-labeled5 to 6S form to the

;H-labeled 14 to 16S form was approximately

thesameinSV80cellsandinfectedcells. Using

a different procedure, McCormick andHarlow

havealso foundevidenceforan underphospho-rylated5to6S form of Tantigenintransformed mousecells (24).

Labeling kinetics of different forms of

SV40

large

T

antigen.

Pulse-chase

experi-ments were performed with SV80 cells labeled

for 15 minwith [35S]methionine or labeled for

15min and thenincubated for 12hin medium

containinga100-foldexcessofunlabeled

methi-onine.Whole-cell extracts were analyzed by zone

velocity sedimentation in sucrose density

gra-dients, followed by immunoprecipitation with

tumor serum, SDS-polyacrylamide gel electro-phoresis, and autoradiography (Fig. 5).

Whole-cell extracts from pulse-labeled and

pulse-chasedcultures were immunoprecipitated with

tumorserumand normalserumand

electropho-resedonthesamegels. Most of the pulse-labeled

largeTantigenwasdetected in the 5 to 6Sform

(Fig.5A).Since the50 to55Khost proteins were

obscured bynonspecificallyprecipitated bands,

5- 4.-aoN

3-xx

o

o2-I

1_1

10 15 20

FRACTION NUMBER

x

"I

a-10 15 20

FRACTION NUMBER

FIG. 4. Zone velocity sedimentation of [3H]leu-cine- and32P04-doubly labeled extractfrom SV40-infected and -transformed cells. (A) SV40-infected CV-1P cellswerelabeled with 30 uCiof [3H]leucine

perml and40,uCi of32pO4permlfrom24 to39 h

afterinfection asdescribed in thetext.Awhole-cell

extractfromabout107cellswaslayeredon a5 to30%o

sucrosedensity gradientin10 mM HEPES(pH 7.8)-5mMKCI-1 mM dithiothreitol-0.5 mMMgCl2-1mM phenylmethylsulfonyl fluorideandcentrifugedinan

SW55rotorfor3 h at45,000rpmand0°C.Fractions

werecollectedfromthe bottom andanalyzed by im-munoprecipitationwithanti-SV40tumorserum,

SDS-polyacrylamide gel electrophoresis, and autoradiog-raphyasdescribed in thetext.Thepeak ofSV40 form IDNAwasinfractions12to13.ThelargeTantigen bands (84to90K)wereexcisedfromthegel,

solubi-lized,and counted.(B) SubconfluentSV80cellswere

labeledfor15hwith 20luCiof[3HJleucineperml and

50 tCi of32pO4per ml. A whole-cell extract from about 107 cells wasanalyzed as in (A), except the sucrosegradientcontained150mMKCI. Thepeak of

SV40formI DNAwasinfractions11to12.Symbols: O, 3H; ,32P.

it was not clear whetherthey were associated with the 5to6Sform. Aftera 12-hchase,most of the labeled Tantigen appearedinthe

faster-sedimentingforms(Fig. 5B). The50to55K host

proteins could now be easily detected, both in

immunoprecipitated whole-cell extract and in

the 23to25Sregionofthesucrosegradient (Fig. A

TOP

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.496.264.454.70.368.2]

HETEROGENEITY OF SV40 LARGE T ANTIGEN

FRACTION NUMBER

5 10 15

. I

FRACTION NUMBER

10 15 20 N T

Il

FIG. 5. Pulse-chaseexperiment:zonevelocity sedimentationofwhole-cellextractsfrom

[3'Slmethionine-labeledSV40-transformedcells.SubconfluentSV80 cellswerepulse-labeledwith 100,iCiof[3"SJmethionine permlfor15min(A)orpulse-labeledandthen incubatedwithfreshmediumcontaining1.5mgofunlabeled

methioninepermlfor12 h (B). Whole-cellextractswereanalyzed byzonevelocity sedimentation,

immuno-precipitationwith anti-SV40tumorserum,SDS-polyacrylamidegel electrophoresis, and autoradiographyas

described in thetext.The directionofsedimentation isfrom righttoleft. The /3-galactosidase16Speakwas

infraction15.NotethatsomelargeTantigenappearedas an84Kdegradation product (34, 37),presumably duetothe small volume usedtoextractTantigeninconcentratedform beforesedimentation.Thetotal acid-precipitable radioactivity appliedtothegradientswas4.8x10'cpm(A)and 3.0x10 cpm(B).

5B). The small T antigen, however, was ob-scured by nonspecific bands in both

autoradi-ograms.Thisexperimentdemonstratesaflow of label in the large T antigen from the 5 to 6S

form to the fast-sedimenting forms, suggesting

aprecursor-product relationshipbetweenthese

forms.

Pulse-chaseexperimentswerealsocarried out

with SV40-infected CV-1P cells. Infected cells

were pulse-labeled with [35S]methionine for 15

min or pulse-labeled and then incubated with

fresh medium containing excessunlabeled me-thionine for2or7h.Whole-cellextractsof each cultureweresubjectedtozonevelocity

sedimen-tation, immunoprecipisedimen-tation, and

SDS-poly-acrylamide gel electrophoresis. The bands of

large T antigen (84 to 90K) were located by

fluorography,excised from thegels,solubilized,

and counted (Fig. 6). Pulse-labeled largeT

an-tigen appeared mainly in the 5 to 6S form,

whereas aftera2-hchase,mostofitwasdetected inabroadpeakbetween 15 and 25S. This broad

peakwasstillpresent aftera7-hchase, although

asmallamountof materialwasobservedat5 to

6S(Fig. 6).These resultssuggestthat in

produc-tively infected cells, as well as in transformed

cells(Fig. 5),the5 to 6S formoflargeTantigen

mayserve asaprecursortothemorestable

fast-sedimentingforms.

Correlation between form and function

oflargeTantigenintsA-infected cells.

Mu-tants ofSV40 with a temperature-sensitive le-sion in geneA are unable to carryoutseveral functions at a nonpermissive temperature, for

example, the initiation ofnew rounds of viral

DNA replication (36) and regulation of early

gene transcription (32, 38). Analysis of these

mutantsthus offers thepossibilityofcorrelating

T

-94

68

F 53

-37

215

Tv7T:.{rT7^:7T754...s-.7

_a3aE=sSS

-v

0Egi0Lrrat0sv;-;.t^S-0-r0wJ-.

v-4 +-^ 0 d Xe- r

iisEi-_

m wWJW---03

;-00 ,0 00 0,,,9'001' 9 874 t.<-,,X,

D

+

=

. ts** S"I

00ii;>..if.:<i:,*-0.-*<a>->

-ss04>-+*Mesf0'j0t0<gt-0t0

<^0".4.fop:B.aiES...A.::,00.0'05:00.00:o:

_-..f.AT4

:- i: @:

;<>>v1ide*>;<^>V;'ex/<2.|"t'jB;00+¢;-f

S.S,:.-,.E:g:XS:w':it

<0

ti,9 re>trlw eV-3 03 lL $; ^^WE s

0} f 'SX ;w X,;f ffW 00 t+.+,+,0*1e,#

5

:

o,tb wlt

7S fv A08;s E

- 94

- 68

-53

- 37

-21.5

VOL. 37,1981 97

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.496.103.393.73.374.2]

FRACTION NUMBER

FIG. 6. Pulse-chase experiment:zonevelocity

sed-imentationof whole-cellextractsof

[3"Smethionine-labeledSV40-infectedCV-1P cells.SV40-infected

CV-IP cellswerelabeled with1X) MCiof[35S]methionine

permlfor15minat42hafter infectionorlabeled

and then incubatedfor another 2 or 7 h in fresh medium containing 1.5mgof unlabeled methionine

per ml. Whole-cell extracts were analyzed as de-scribed in thelegendtoFig. 4B. The16Speakof

13-galactosidasewasinfraction15. The total acid-pre-cipitableradioactivity appliedtothegradientswas asfollows:0, 1.2x 10' cpm(15-min label); A, 1.1 x

107cpm(2-hchase); and0, 7.8x10';cpm(7-hchase).

the various formsofthelargeTantigen with the

functions of thisprotein.

CV-1P cells infected with the mutant tsA58

were labeledatthepermissive temperaturefor

2 h with

[:35S]methionine.

Whole-cellextractsof

each culture were analyzed by zone velocity

sedimentation,immunoprecipitation,

SDS-poly-acrylamide gel electrophoresis, and

fluorogra-phy. At 330C, the labeledlarge T antigen was

observed in both the5 to6S andthe 14to 16S

forms (Fig. 7), as might beexpected from the

results obtained withwild-type SV40 (Fig.4A).

Whencellsinfectedwith tsA58werelabeled for

2 h at 330C with

[:"5S]methionine

and then

shifted to 410C for a chase, about 70% ofthe

fast-sedimentinglarge Tantigenlabeledat330C

wasstillpresentintheseformsafter3hat410C,

about thesame as thefraction of

acid-precipi-tableradioactivity remainingintotal cellprotein

after the chase (75%) (Fig. 7). Thus, the 14to

16Sform,oncemadeatthepermissive

temper-ature, seems to be stable at thenonpermissive temperature.Most of the5to6Slarge Tantigen

was no longer present after the chase, but it

cannot be determined from these results

whether the large T antigen was converted to

the 14to 16S form duringshift up ordegraded

completely at410C.

In asimilar fashion,T antigenwasextracted

fromtsA58-infected cells labeled with

[3'S]me-thionine for 2 h at 410C and analyzed as

de-scribed above. Most of the labeled large T anti-genappeared in the 5 to 6S form (Fig.8A) (16, 28),thusresemblingthe15-minpulse-labeled T

antigen from wild-type-infected cells (Fig. 6).

These results suggested that the 5 to 6S T

anti-gen was not converted to the fast-sedimenting

form at the nonpermissive temperature and, thus, accumulated. To test whether the 5 to 6S

largeT antigen made at 410C could be converted

to the 14 to 16S form, shift-down experiments

were carried out and analyzed as described

above. Mutant-infected cells were labeled with

[35S]methionine for 2 h at410Cand thenchased

in fresh medium containing excess unlabeled

methionine for 2 h at330C (Fig. 8B). About half

ofthe large Tantigen made at the nonpermissive

temperature had been converted to the 14 to

16S form after 2 h at the permissive

tempera-ture.

Toinvestigate the correlation between SV40

DNA synthesis and the different forms of T

antigen, viral DNA synthesis in tsA58-infected

cells was monitored at thepermissiveand

non-permissive temperatures (Table 1). Little SV40

DNA was detected in cells labeled with

[3H]-thymidine for 14 h at 41°C, whereas at

33°C,

viral DNA was made in large amounts. When

cells infected at330Cwereshifted to41°Cfor 30

min and then labeled with[3H]thymidinefor 60

min at410C, againlittlelabeled SV40 DNA was

observed. When infected cells held for 30 min at

41°C were shifted down to 330C and

immedi-5 10 1'5 2'0 TOP

FRACTIlON NUMBER

FIG. 7. Stability of14 to 16S large Tantigen in

tsA58-infectedCV-IP cellsatthenonpermissive

tem-perature. CV-IP cells infected with tsA58 at 33°C

werelabeled with 100yCiof [35S]methionineperml

firom

71 to 73h after infectionorlabeledand then

incubatedat41°Cwith

firesh

mediumcontaining1.5 mgofunlabeled methionine per mlforanother 3h. Whole-cellextractsfrombothcultureswereanalyzed

byzone velocity sedimentation as describedin the

legendtoFig.4B, The totalacid-precipitable

radio-activity appliedtothesucrosegradientswas as

fol-lows:*,2.4x 10';cpm(labelat33°C); andO, 1.8x

10' cpm(chaseat41°C).

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.496.58.248.62.218.2] [image:7.496.283.432.410.537.2]

HETEROGENEITY OF SV40 LARGE T ANTIGEN 99

1 5

10

15

20

-130

~~~81

-68

-37 21.5

10W~~~~~~~~~~~~

B

F=RACTION NUMBER

10

15

20

T N

a .

-130

-:

_94

-81 -68

-37

F-21.5

FIG. 8. Conversionof5 to6SlargeTantigento a14to16SformintsA58-infected cells after a shift to the

permissivetemperature.CV-IP cellsinfectedwithtsA58at41°Cwerelabeled with 100jLCiof

[35S]methionine

permlfrom44 to 46hafter infectionorlabeled and then incubatedat330Cinfresh medium containing excessunlabeled methionineforanother2h. Whole-cellextractsfrom bothcultures were analyzed by zone velocitysedimentationasdescribed in thelegendtoFig.4B.(A) 2-h labelat41°C; (B) 2-h chase at 33°C. The directionofsedimentation isfrom righttoleft.Thepeak ofSV40formI DNAwasinfraction 13. The total

acid-precipitableradioactivity appliedtothegradientswas 4.1x10"cpm(A) and 3.9x 10"cpm (B).

ately labeled with

[3H]thymidine,

some viral

DNAsynthesiswasdetected. When

[3H]thymi-dinewasadded later after the shiftdown,more

labeled viral DNA was found. These

experi-mentsindicate that intsA58-infected cells,SV40

DNAsynthesis is severely reducedsoonafter a

shifttothe

nonpermissive

temperatureand that itbeginsagain afterashort timeatthe

permis-sivetemperature, inagreementwith earlier

re-ports(36).

DISCUSSION

SV40

large

T

antigen

from

productively

in-fected and transformed cells occursin at least

threesubclasses

separable by

zone

velocity

sed-imentationof crudecellextractsinsucrose

den-sity

gradients.

Such subclasses have been

ob-served

previously (4,

16,

28, 29),

but

biological

and

biochemical

differences among them have

remained, for the most part, obscure. The

ap-parentsedimentation coefficients of these three

subclassessuggestthat differentstatesof

aggre-gationof T antigen exist. The 5 to6Sform of

largeTantigen could

correspond

to afree80to

90K monomer, whereas the 14to 16Ssubclass

couldrepresent atetrameror,

possibly,

a com-plex withunlabeled hostproteins.Inagreement

withrecentobservations(24), the23to25S form

consists of

large

T

antigen

andone ormore 50 to 55Khost proteins (5-7, 9, 15, 18, 20, 25, 26,

35). Theseproteinsare tightly complexed with

thelargeTantigen;the host

proteins

from SV80

cells

cosediment with the

large

T antigen and

co-immunoprecipitate with it even after

sedi-mentation in 1 M KCI

(unpublished

data). A

comparison of the 55K

proteins

from infected and transformed cells

by

two-dimensional

gel

electrophoresis and V8

peptide mapping

sug-VOL. 37,1981

on November 10, 2019 by guest

http://jvi.asm.org/

[image:8.496.110.395.72.389.2]

TABLE 1. SV40 DNAsynthesis intsA58-infected

CV-IP cells

Acid-pre-cipitable 'H-la-Expt Labeling protocol beled SV40 DNA (cpm)

1" 330C, 14 h 152,841

41°C, 14h 7,144

2b 410C, 30 min after shift up 529 33°C,immediately after shift 2,933

down

33°C,45minafter shift down 8,405 33°C,60minafter shift down 10,345

33°C,90minaftershift down 11,876

"Infected cellswerelabeledwith 5,uCiof[

'H]thy-midine per ml from 76to 90 hafterinfection. SV40 DNA wasextracted asdescribed in the text and ana-lyzed by zonevelocitysedimentation in sucrose density gradients (14). Acid-precipitable radioactivity under thepeaks ofSV40 DNA was summed.

bCells infectedat33°Cwereshifted to410Cat72 h afterinfection for30min and thenlabeled for 60 min at410C with100LCi of[;'Hthymidineper ml.Parallel cultures were shifted to 410C for 30 min and then shifted back down to 33°C and labeled for 60 min, eitherimmediatelyorafterincubation for 45, 60, or 90 minat33°C. Acid-precipitable radioactivity in SV40 DNA wasdetermined as in experiment 1.

gests that theseproteinsareverysimilar

(Fan-ning, Burger,andNowak,manuscriptin

prepa-ration). Inaddition to the5 to6S and 23to25S

forms described recently (24), we have

consis-tently observed a 23 to 25S form in extracts of

infected cells anda 14 to 16Sform inextractsof

infectedandtransformedcells;thesedifferences

couldpossiblyresult fromdifferencesin

extrac-tion condiextrac-tions. It is interesting that three

im-munologically distinguishable subclassesoflarge

Tantigen,one ofwhichcorrespondstothe 23 to 25Sform,havealsobeen found inboth infected

and transformed cells with monoclonal

anti-bodies (11).

The different forms of

large

T antigen are

apparentlydifferentially phosphorylated (Fig. 4

and reference24).The5to6Sform of Tantigen

containslesslabeledphosphatenotonlyafter a

long label(Fig. 4), but also after apulse-labelof

15min(E. Fanningand B.Bukau, unpublished data). These experiments do not exclude the

possibility that the apparent underphospho-rylationcould result in part from

cosedimenta-tion ofthe 5to6S formwith

phosphatase

activ-ity.Furthermore,it isnotclear whether some of

the 5 to 6S T antigen is fully phosphorylated

and the rest is not

phosphorylated

at all or

whether all of the5 to6S T

antigen

is modified

inonly oneor some ofseveral possiblesites of

phosphorylation. In agreement with the second

possibility,

T

antigen

from tsA-infectedcellshas

been

reported

to be underphosphorylated in a

particular

tryptic peptide atthe nonpermissive

temperature (G. Walter and J. P. Flory, Cold

Spring

Harbor

Symp.

Quant. Biol., in press).

Edwardset al.

(7)

havealsoproposed that two

populations

ofphosphorylated residuesmay

ex-ist,

basedonthe kinetics of the32PO4 turnover

in Tantigen.

Pulse-chase experiments indicate that the

subclassesoflargeTantigenmaybetemporally

related. Newly synthesized Tantigen

pulse-la-beled with methionine sedimented at 5 to 6S.

Aftera shortchase, the 5 to 6Sform had been

converted tofaster-sedimenting forms, both 14

to 16S and 23 to 25S. The label persisted in

these formsevenafter chase periods of 7 to 12h,

indicatingthat thefast-sedimentingTantigen is

quite stable. The 55K host proteins also

ap-pearedtobe stable in transformed cells aftera

12-h chase(Fig.5B).

We have attempted to correlate thesubclasses

oflarge T antigen with some of the functions attributedtothisprotein throughthe use of tsA

mutants. Since tsA-infected cells at the

nonper-missive temperature contain only 5 to 6S T

antigen,this form isunlikelytobe functional in

the initiation of viral DNA replication or the

regulation of transcription. The fact that

fast-sedimentingforms of T antigen, once formed at thepermissivetemperature, persist at41°C,

al-though theinitiation of new rounds of

replica-tion isinhibited almostimmediatelyuponshift

up(reference 36 and Table 1), means either that

the fast-sedimenting forms are alsounlikely to

be functional in the initiation ofreplication or

that several forms of Tantigencosedimentat 14

to 16S, one or more of which is temperature

sensitivefor initiation.

Alternatively,since neither the

fast-sediment-ing nor the 5 to 6S form of T antigen was

correlated with the initiation ofreplication,

per-haps the transition process itself or a transitory

intermediateform of T antigen is essential for

initiation. Consistent with this interpretation,

both DNA replication and the conversion of 5 to

6ST antigen to the 14 to 16S formresumeafter

ashift down to thepermissive temperature. The

observation that newly synthesized T antigen

binds more tightly to DNA than does older T

antigen (27) is alsoconsistentwiththis idea.

Theamount ofthefast-sedimenting formsof

Tantigenmay alsoplayarole intheattenuation

ofearlytranscription. Thisfunction oflargeT

antigenislost, not

immediately,

butgradually,

after a shift to the nonpermissive temperature

(32), just as the amount offast-sedimenting T

antigen graduallydeclines atthenonpermissive

on November 10, 2019 by guest

http://jvi.asm.org/

HETEROGENEITY OF SV40 LARGE T ANTIGEN 101

temperature(Fig. 7). In thisway,early

transcrip-tionwould bemaximal until sufficient

fast-sed-imenting T antigen accumulated.

In summary, the large T antigen exists in

multiple forms in different relative amounts in

SV40-infected and -transformedcells. Evidence

has been presentedthat these forms differ from

each other biochemicallyand that theymaybe

correlated withsomeof the biologicalfunctions

of T antigen. Thistypeof analysismay proveto

be useful in elucidating the mechanisms by

which T antigen carriesoutits varied functions.

ACKNOWLEDGMENTS

WethankA.J. Levine,A.M.Lewis, Jr., F. J.O'Neill,G. Sauer,andE.Winocour forseed stocksoftransformedcells, P. Tegtmeyer for a seed stock ofthe t.A58 mutant, W. Deppert, H. Fischer, J.Gruber, and A. M. Lewis, Jr., for samples of antiserum,G.Walterforcommunicatinghisresults before publication, and F. Gruberfor animal care. Weare

especially gratefultoD. Corlinforexcellenttechnical

assist-ance,M.Hornbergerfor the constructionoflaboratory

equip-ment,R.Mettke for the preparation of culture media, I. Luhr andR.Phildiusforhelp with the manuscript, and R. Knippers for hiscontinued interestandsupport.

This workwassupported by the Deutsche

Forschungsge-meinschaft throughSFB138/B4. LITERATURE CITED

1. Aaronson, S.A.,andG. J. Todaro. 1968.Development of3T3-likelinesfromBalb/cmouseembryo cultures:

transformation susceptibility toSV40. J. Cell.Physiol. 72:141-148.

2. Baumgartner, I., C. Kuhn, and E. Fanning. 1979. Identificationandcharacterizationoffast-sedimenting SV40 nucleoprotein complexes. Virology96:54-63. 3. Black,P.H.,and W. Rowe. 1963. Ananalysis of

SV40-induced transformation of hamster kidney tissue in vitro. I. General characteristics. Proc. Natl. Acad. Sci. U.S.A.50:606-613.

4. Carroll,R.B.,L.Hager,and R. Dulbecco. 1974. Simian

virus40TantigenbindstoDNA.Proc.Natl.Acad.Sci. U.S.A. 71:3754-3757.

5. Chang, C.,D.T.Simmons,M. A.Martin,and P. T. Mora. 1979. Identification andpartialcharacterization of new antigens from simian virus 40-transformed

mousecells. J. Virol. 31:463-471.

6. DeLeo,A.B.,G.Jay,E.Appella,G. C.Dubois,L. W. Law, andL.J. Old. 1979. Detection ofa

transforma-tion-related antigen in chemically induced sarcomas

and other transformed cells of themouse. Proc.Natl. Acad. Sci. U.S.A. 76:2420-2424.

7. Edwards,C.A.F.,G.Khoury,and R. G.Martin. 1979.

PhosphorylationofT-antigenandcontrolofT-antigen expression incellstransformed by wild-typeand tsA mutantsof simian virus 40. J. Virol. 29:753-762. 8. Fanning, E.,andI.Baumgartner. 1980. RoleofSV40

nucleoprotein complexes in virusassembly. Virology 102:1-12.

9. Gaudray,P.,M.Rassoulzadegan,andF.Cuzin. 1979.

Expressionof simian virus40earlygenesintransformed ratcells iscorrelated with maintenance of the trans-formedphenotype. Proc. Natl. Acad. Sci. U.S.A. 75: 4987-4991.

10. Gluzman,Y.,J.Davison,M.Oren,and E. Winocour. 1977.Propertiesofpermissive monkeycellstransformed byUV-irradiatedsimian virus 40. J. Virol. 22:256-266. 11. Gurney,E.G.,R. 0.Harrison,and J. Fenno. 1980.

Monoclonalantibodiesagainstsimian virus 40 T

anti-gens:evidence fordistinct subclasses oflargeTantigen

and for similarities among nonviral T antigens. J. Virol. 34:752-763.

12. Kelly, T. J., Jr., and D. Nathans. 1977. The genome of simian virus 40. Adv. Virus Res. 21:85-173.

13. Kessler, S. W. 1975. Rapid isolation of antigens from cells with aStaphylococcal protein A-antibody adsorbent: parametersof theinteraction ofantibody-antigen com-plexeswithprotein A. J.Immunol. 115:1617-1624. 14. Klempnauer, K.-H., E. Fanning, B. Otto, and R.

Knip-pers. 1980.Maturation of newly replicated chromatin ofsimian virus 40 and its host cell. J. Mol. Biol. 136: 359-374.

15.Kress,M., E. May, R. Cassingena, and P. May. 1979. Simian virus40-transformed cells express new species ofprotein precipitable by anti-simian virus 40 tumor serum. J.Virol.31:472-483.

16. Kuchino,T., and N. Yamaguchi. 1975. Characterization of Tantigen incells infected with a temperature-sensi-tivemutant ofsimian virus 40. J. Virol.15:1302-1307. 17. Laemmli, U. K. 1970.Cleavage of structural proteins

during the assembly of the head ofbacteriophage T4. Nature(London) 227:680-685.

18. Lane,D.P., and LV.Crawford. 1979. Tantigen is bound to a host protein in SV40-transformed cells. Nature(London) 278:261-263.

19. Lewis, A. M., Jr., and W. P. Rowe. 1971. Studies on nondefectiveadenovirus-simian virus 40 hybrid viruses. I. A newly characterized simian virus 40 antigen induced by the Ad2+ND, virus. J. Virol. 7:189-197.

20. Linzer,D.I.H., andA. J.Levine.1979.Characterization of a54Kdaltoncellular SV40 tumor antigen present in SV40-transformed cells and uninfected embryonal car-cinomacells.Cell 17:43-52.

21. Linzer, D. I. H., W. Maltzman, and A. J. Levine. 1979. The SV40 geneAproduct isrequired for the production of a54,000 MW cellular tumor antigen. Virology 98: 308-318.

22. Mann,K., T. Hunter, G. Walter, and H.Linke.1977. Evidence forsimianvirus40(SV40) coding of SV40 T-antigen and the SV40-specific proteins in HeLa cells infectedwith nondefectiveadenovirus type 2-SV40 hy-bridviruses. J.Virol. 24:151-169.

23. Martin, R.G.,V. P.Setlow, C. A. F. Edwards, and D. Vembu. 1979.Theroles of the simian virus40tumor

antigensintransformation ofChinese hamsterovary cells.Cell 17:635-643.

24. McCormick, F.,and E.Harlow. 1980.Association of a murine53,000-dalton phosphoproteinwith simianvirus 40large-T antigen in transformed cells. J. Virol. 34: 213-224.

25. Melero, J.A.,D. T.Stitt, W.F.Mangel,and R. B. Carroll.1979.Identification ofa newpolypeptide spe-cies(48-55K)immunoprecipitable byantiserumto pur-ifiedlargeTantigen and presentinSV40-infectedand -transformedcells.Virology93:466-480.

26. Melero, J.A.,S.Tur, and R. B. Carroll. 1980.Host nuclear proteinsexpressed in simian virus 40-trans-formedand-infected cells.Proc.Natl.Acad.Sci.U.S.A. 77:97-101.

27. Oren, M.,E.Winocour,and C. Prives.1980. Differen-tial affinitiesofsimian virus40largetumorantigenfor DNA. Proc. Natl. Acad. Sci.U.S.A.77:220-224. 28. Osborn, M., and K. Weber. 1975. T antigen, the A

functionandtransformation. ColdSpringHarborSymp.

Quant.Biol. 39:267-276.

29. Potter, C.W.,B. C.McLaughlin,andJ.S. Oxford. 1969.Simian virus40-inducedTand tumorantigens.J. Virol. 4:574-579.

30. Prives,C., E. Gilboa, M. Revel, and E. Winocour. 1977.Cell-free translationof simianvirus 40early

mes-senger RNA coding for viral T antigen. Proc. Natl. Acad. Sci. U.S.A. 74:457-461.

31. Prives,C.,Y.Gluzman,andE. Winocour. 1978. Cel-VOL. 37,1981

on November 10, 2019 by guest

http://jvi.asm.org/

lular and cell-free synthesis of simian virus 40

T-anti-gensin permissiveand transformed cells. J. Virol. 25:

587-595.

32. Reed, S. I., G. R. Stark, and J. C. Alwine. 1976.

Autoregulation ofsimian virus 40geneAbyTantigen. Proc. Natl. Acad. Sci.U.S.A. 73:3083-3087.

33. Rundell, K., J. K.Collins,P.Tegtmeyer,H.LOzer, C.-J. Lai, and D. Nathans. 1977. Identification of simian virus40proteinA. J. Virol. 21:636-646. 34.Smith,A.E.,R.Smith,andE. Paucha. 1978.Extraction

andfingerprint analysisof simianvirus 40large and small T-antigens. J.Virol.28:140-153.

35. Smith,A.E.,R.Smith,and E. Paucha. 1979.

Charac-terization of differenttumor antigenspresent incells transformed by simian virus40.Cell 18:335-346. 36.Tegtmeyer, P. 1972.Simian virus 40deoxyribonucleic

acidsynthesis:the viral replicon. J. Virol.10:591-598.

37. Tegtmeyer, P., K. Rundell,and J.K.Collins. 1977. Modification ofsimian virus 40 protein A. J. Virol.21: 647-657.

38.Tegtmeyer, P., M.Schwartz,J. K. Collins,and K.

Rundell. 1975. Regulationoftumorantigen synthesis by simian virus40geneA.J. Virol. 16:168-178. 39.Todaro, G.J., H. Green, andM. R.Swift. 1966.

Sus-ceptibility of human diploidfibroblast strainsto

trans-formation by SV40virus.Science153:1252-1254.

on November 10, 2019 by guest

http://jvi.asm.org/