Nucleotide sequence analysis of viable deletion mutants lacking segments of the simian virus 40 genome coding for small t antigen.

0022-538X/79/06-0668/06$02.00/0

Nucleotide

Sequence Analysis of Viable Deletion Mutants

Lacking

Segments

of the

Simian Virus

40

Genome

Coding

for

Small

t

Antigen

BAYAR THIMMAPPAYA AND THOMAS SHENK*

Department of Microbiology, University of Connecticut HealthCenter,Farmington,Connecticut06032

Receivedforpublication27September 1978

The deletions in nine viable simian virus 40 mutants have been mapped by direct DNA sequence analysis. The mutant DNAs lack small segments ofthe early region of the viral chromosome (between0.535 and 0.600 map unit). The deletionsarealllocated in the region which isremovedfrom thelargeTantigen transcript bysplicing. No onedeletion removes this entire region, but no partof thissegment isconservedin allof themutantsexcept for severalnucleotidesnear thesplice points ofthetranscript. Although the deletions do not alter theregion coding for the largeTpolypeptide,they dodeleteportionsof thesegmentcoding for the C-terminal half of the small tpolypeptide.

Twotranscripts have beenidentified insimian virus40 (SV40)-infectedcellswhich map in the early region of the viral genome (Fig. 1). Both transcripts extend (5' -+ 3') from near 0.66 to

0.16 onthe SV40map,andboth arespliced (1,

2). One transcript is believed to code for the SV40 small t antigen, a polypeptide of about 17,000 daltons. Thispolypeptidehasbeen iden-tified in bothlyticallyinfectedmonkeycellsand

transformed rodent cells (3, 10, 17, 18). The

second transcript codes for large T antigen, a

polypeptideof about 90,000 daltonswhichisalso

present in both lytically infected and

trans-formed cells (10, 13).

The T and t polypeptides are structurally

related. Most of the

[35S]methionine-labeled

trypticpolypeptides ofsmall tare alsopresent in large T antigen (10), and Paucha et al. (9) have shown the N-terminal amino acid se-quences of these polypeptides to be identical. Thisisconsistent with thestructureof the tran-scripts coding forthe Tandtpolypeptides.The

transcripts share the same 5'-nucleotide

se-quence, coding for the N-terminal portions of

thepolypeptides.But,asaresultofsplicing,the

mRNA coding for large T antigen lacks the sequence codingforboththeC-terminalportion ofsmall tantigen and a series oftranslational terminationcodons (0.534to0.600mapunit) (1, 12, 20).

Aseriesof deletion mutants has been isolated

withlesionslocatedbetween 0.535 and0.600 on

theSV40 map(16,18).Thisregionisnotpresent

in thelargeTtranscript,but iscontainedin the

smalltantigentranscript (Fig. 1).Aspredicted,

these deletions donotaffect the size oflargeT

antigen but do alter small t antigen. The

mu-tants areviable,growingat aslightly slowerrate

than wild-type virus (16, 18),butare markedly impaired in their ability to transform rodent

cells(2, 18). This suggests thatatleast portions

of the C-terminus ofsmalltantigenare dispen-sable for lytic growth of SV40, but an intact

small t antigen is required for efficient viral

transformation.

We havedetermined by directDNAsequence analysis the nucleotide sequence whichnine of these deletion mutants lack. All mutants lack segments of theregionbetween 0.535 and0.600 on theSV40 chromosome. Althoughno mutant lacks this entire region,there is noportionofthe sequence which is conserved in all ofthe

mu-tants, except small segments near the splice

points ofthelargeTantigentranscript.

MATERIALS AND METHODS

Cells and viruses. The origin andproceduresfor growth of the CV-1P line ofAfrican green monkey cells have beendescribedby Mertz and Berg (8). The wild-typeSV40(designatedwt830) is aplaque-purified derivative oftheSVS strain (19).Thedeletionmutants

d1883, d1884, d1890,andd1891werederived fromwt830

(16). Mutantd12006waskindlyprovided by William Topp (18). Mutants d11410 andd11440 werederived by expansion ofthedeletion ind1884, using the pro-cedure ofShenk(14).Mutantd11410containsasecond deletion at 0.70 to 0.71 map unit (15). Thedeletion in

d1891wasexpandedtoproduced11441andd11442.

DNA andenzymes. SV40DNAwasextracted(5) from CV-1P cells infected at a multiplicity of<0.05 PFU/cellwhen >90% of the cellsshowedcytopathic effect. Closed circularviral DNA[SV40(I)]was puri-fieddirectlyfrom thesupernatant byaddingCsCl to 668

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.505.55.246.61.253.2]

FIG. 1. DiagramoftheSV40chromosomerelating

themappositionof viable deletionmutants(0.60to

0.535 mapunit)tothelocationofthetwoknownearly

transcripts. Theportions ofthe transcriptsbelieved

tocodefor earlyproteinsareindicatedbythe blocked

segments (t=smalltantigen; T=largeTantigen).

Regions removed from transcripts by splicing are

represented by dottedlines. ORI = origin ofDNA replication.

1.56 g/cm3 andethidium bromide to 200,ug/ml and centrifuging toequilibrium.The band ofSV40(I) DNA

wascollected,and the ethidium bromide was removed

by passingthe DNAthrough AG50W-X8resin

(Bio-Rad Laboratories) (11). Restriction endonucleases

were either prepared in the laboratory or purchased from New England Biolabs and used according to

published protocols. Bacterial alkaline phosphatase

was purchasedfrom WorthingtonBiochemicals Corp., and T4 polynucleotide kinase was from Boehringer Mannheim Corp.

DNA sequence analysis. Restriction

endonucle-ase-generated fragmentswereendlabeled bythe

pro-cedure of Maniatiset al. (6). Sequence analysis of end-labeled fragments was perfonned as described by MaxamandGilbert(7), using 12%polyacrylamidegels (0.7 mmthick,40 cmlong).

RESULTS

Toformulateasimple strategy for the

nucleo-tide sequenceanalysisof alterations present in the deletion mutants, it was firstnecessary to determine whichrestriction endonuclease

cleav-agesiteswerepresent andwhich were missing

in each mutant DNA. Accordingly, each DNA was cleaved with several enzymes which cut SV40DNAbetween 0.53 and 0.60 map unit. The data from theseanalyses aresummarized in Fig. 2.Mutantd1884DNA, for example, retained the Hinfendonucleasecleavagesite at 0.533 and the

HaeIIIendonucleasecleavage site at 0.595 map

unit, but lost all of the cleavage sites assayed

between these points (Fig. 2). Mutant dl890

DNAretainedall of the cleavage sites assayed, but its HaeIII and MboI endonuclease A frag-ments were bothshortened, indicating that the deletion was contained between theHaeIII and MboI endonuclease cleavage sitesat0.589 and 0.572 mapunit, respectively (Fig. 2).

Utilizingtheinformation inFig. 2, appropriate restriction endonuclease-generated fragments from each mutant were prepared. The fragments

were end labeled (6) and finally cleaved with

anotherrestriction endonuclease togenerate a fragmentlabeledwith 32P at one 5' end. Mutant d1890DNA, forexample,was cleaved with the Hinf and MboIendonucleases, andthe 0.572 to 0.642 map unit fragment was end labeled and then cleaved with HaeIII endonuclease. This produced a fragment with a labeled 5' end at 0.572 map unit and anunlabeled 5' end at 0.589 mapunit for sequenceanalysis. The end-labeled

fragments prepared from the deletion mutant

DNAs are diagramedinFig. 3.This figure also indicates the sequence which was read (solid lines) on either side of the deletions (dashed

lines)aftersubjectingtheend-labeledfragments

tonucleotide sequenceanalysis by themethod

of Maxam andGilbert(7).

Anexampleof a sequencinggel (dl890DNA)

and thenucleotidesequencedetermined from it areshown inFig.4.The sequence starts from a

pointnearthelabeled, MboI

endonuclease-gen-erated 5' end(complementarytonucleotide 457/ 4704inFig. 5) and isreadfrom bottom to top. The locationofthe deletion isindicatedby an arrowwhere the sequence jumps ahead 27

nu-cleotides as compared with the wild-type

se-quence.

to 'Q o0 U) Uo<

0 I& .00 %t%..

0-0-c

=LLJ =

883 884 890 891 2006 1410 1440 1441 1442 FIG. 2. Location ofviable deletion mutations de-terminedbyrestrictionendonucleasecleavageof

mu-tantDNAs. The mappositionsatwhich the enzymes cleave weredetermined bylocatingthe known

rec-ognition sequencesoftheenzymes in theSV40

nu-cleotidesequence of Reddyetal.(12).Thesolid bars are bounded by restriction endonuclease cleavage

sitespresent inmutant DNAs and extend through

cleavage sites which themutantslack.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.505.263.452.461.580.2]

p. o

o q .0 o

cli

I') 0 0 .0X.

"IC)I

co

c c

I I

A.

-_--. 883

--- -_ 884

- -_ 890

- -- -. 890

891

_--

-- - .* 2006

- 1410

-- _ .1440

1441

-1442

FIG. 3. Strategyfor nucleotide sequence analysis

of mutantDNAs. The solidcircles represent 32P-la-beledends;the solid lines representregionsin which the sequence wasread;thedashedlines indicate the segmentsdeleted inmutantDNAs.

ThenucleotidesequenceofSV40intheregion

codingfor smalltantigenis

reproduced

inFig.

5 (taken from the data of Fiers et al.

[4]

and

Reddyetal. [12]).The segments which various

deletionmutantslackareindicatedbybrackets.

The points at which Reddy et al.

(12)

have

determined splicesto occur inthe

large

T and smallttranscriptsarealsoindicated. All of the

mutantslackportionsof thenucleotide segment

which are not present in the

spliced

large

T transcript.

DISCUSSION

Itwas relatively easy to determine the

loca-tion ofthesedeletionsintheSV40genome since

the nucleotide sequence of the viral DNA is

known(4, 12). Ingeneral, thedeletionscouldbe

unambiguouslyfixed in thenucleotidesequence

byanalyzing only onestrand ofthe viral DNA

in the region ofinterest. Volckaert, Fiers, and Berg (personalcommunication) have also deter-mined the nucleotide sequence alterations in

d1883,d1884,d1890,andd1891.Theirresultsare

incomplete agreement with ours.

Thedeletionsrangein sizefrom25to253base

pairs.Theyareall

loeated

inthe region removed from thelarge Tantigentranscript bysplicing

(Fig. 5). None removes this entire region (346

nucleotides), but nopart ofthis segment is re-tained in allofthe mutants except forseveral nucleotidesnearthe splicepoints.Although the deletions donotalterthe regioncodingforlarge Tantigen, theydodelete portionsofthe segment

codingfor thesmall tpolypeptide. Thisis

con-sistentwithreports that themutants synthesize normallarge Tbut alteredsmalltpolypeptides

(3,16, 18;G.Khoury,personalcommunication).

The alterations predicted by the nucleotide

sequence agreewellwith thesize of the mutant

i__-_t,.

.g

S-! \

s ZF'

tv .v.

x,

',,61 ..6..

r

:

l.;t.

..^R

.:s

.,

..7F

S,

R

U 1

... .4

.2--,. A

r

G

A

.:. .-I

mmrnm~

FIG. 4. Autoradiograph ofa sequencinggel

con-tainingdl89 DNA. The sequence, which is readfrom

bottomtotop,beginsattheresiduecomplementaryto

position 457/4704 (Fig. 5) and reads through the

deleted segment(indicated byarrows).The Cresidue,

which is six residuesfrom the top ofthe sequence listedbeside thegel,isartifactual. It shouldbea T. WefindanA residue when the opposite strand is

analyzedaspredictedbythedataofFiersetal. (4)

andReddyetal. (12).

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.505.65.257.54.210.2] [image:3.505.264.453.64.562.2]

-MOW ATGG T

9561@AT010

GAG

GAA

TCr

TTGCGC

G

CITMG

MTGC TGG GMMT 80/5081 90/5071 100/5061 110/5051 120/5041 130/5031 140/5021 150/5011

MET ASPLYS VAL LEUASN ARG GLU GLU SER LEU GLN LEUMETASP LEULEU GLY LEUGLUARG SERALA TRP GLY ASN

ATTCCT C AGKA A TAT

T1bA

AM TGC AA GAG m CAT

CU[fALM

GGA GCA

GALCA

GM M4 AM

E

/5001 170/4991 180/4981 190/4971 200/4961 210/4951 220/4941 230/4931 240/4921 ILE PRO LEU MET ARG LYS ALA TYR LEU LYS LYS CYS LYS GLU PHE HIS PRO ASP LYS GLY GLY ASP GLU GLU LYS PET LYS LYSPiT

T SPLICE

AT

ATCTG TACMG A ATG G GATGGAG TAT GCT CTCMCC m MA a TTCTGG GATG

I

GTAm

250/4911 260/4901 270/4891 280/4881 290/4871 300/4861 310/4851 320/4841 330/ ASN THR LEUTYR LYS LYS PETGLU ASP GLYVALLYSTYR ALA HIS GLN PRO ASP PHE GLY GLYPHETRP ASP ALATRGLUVALPHE

r1441

F1442'2006

1441

r1440

r1410

88

GCT TCT TCC TAMT CCT GT

GLAT

GCA ATGTAC TGCAMCM

CGLG

TGT GCA

AG

MA

ATGTCT

GCIMC

TGC ATA TGC

340/4821 350/4811 360/4801 370/4791 380/4781 390/4771 400/4761 410/4751 ALA SER SER LEU ASN PROGLYVAL ASP ALAMET TYRCYS LYSGLN TRPPROGLU CYS ALA LYS LYS PETSER ALA ASN CYS ILECYS

890i

14421

TTGCTGTGC TTA

CI

AG

a&ACA-

AI-

AM TTA TAC MG AM GAT CCA MTGTG TGG

GL

T aGTAC TGC TTC GAT /4741 430/4731 440/4721 450/4711 460/4701 470/4691 480/4681 490/4671 500/4661 LEU LEU CYS LEU LEU ARG MET LYS HIS GLU ASN ARG LYS LEU TYRARG LYSASP PRO LEU VAL TRP VAL ASP CYS T CYS PHE ASP

r883

TGC

TUlMA

ATG TGG M WACTT M TGT

%A GGA AC

TTA

MT

CTG

TGG

TL

AATT

MA CM AU

ACC

TACNA

GAT

510/4651 520/4641 530/4631 540/4621 550/4611 560/4601 570/4591 580/4581 590/ CYS PHEARGMETTRP PHE GLY LEUASP LEU CYSGLUGLYTHlR LEU LEULEUTRP CYS ASP ILE ILEGLY GLN THR

TM

YR ARG ASP

2006_it SPLICE_F-i 883

₁₈₃

88,44

_W01

1

T+t

SPLICE

[image:4.505.63.458.121.382.2]

-TILAPG

CTCJ,GG3T TAIA1

TMTGTAL,TGTGTm

ACTACATr

CI&TTGM

GTATA1ATTMACC

TATGGCTG-600/4561 610/4551 620/4541 630/4531 640/4521 650/4511 660/4501 610/4491 680/4481 690/4471 LEULYS LEU

FIG. 5. Nucleotidesequence of the portion of the SV40 genome coding for small t antigen. The region

extends fromabout 0.65 to 0.53 map unit. The sequence isfromFiers et al. (4) andReddy et al. (12), and the

nunberscorrespond to those assigned by Fiers etal./Reddyetal.Thesplicepointspresentin the large T and

small ttranscriptsarefrom Reddy et al. (12). The segments which variousdeletionmutants lack are indicated

by brackets.

t proteins which have been observed. Mutant

d1884lacks a DNA segment coding for 63 of the

174 amino acids which the wild-type sequence

predicts comprise the little tpolypeptide (Fig.

5). Three amino acids not found in the

wild-type

proteinshould be added to the C-terminal end

of the dl884polypeptide beforereaching a ter-mination codon (UAG) at position 669/4492

(Fig. 5). This predicts that the d1884 genome

should code for a 114-amino acid little t

poly-peptide.Sleigh et al. (18) and Khoury (personal

communication) haveobserved a 12,000-dalton

polypeptide in dl884-infected cells. Mutant

dl890,whose DNA lacks 27 base pairs, should

synthesize a 165-amino acid small t protein. Khoury (personalcommunication) has found a 16,000-dalton polypeptide in cells infected with

thismutant.Finally,therelativelysmall deletion

(25 base pairs) in d1891should alter the trans-lational reading frame (the deletion is not a

multiple of3 base pairs) toproduce a

dramati-cally shortened polypeptide. Synthesis of this

mutantpolypeptideshould

terminate

at aUGA

codon (Fig. 5, position 432/4729) after adding 19 amino acids in a new reading frame. Thus, the sequencepredictsa109-amino acidpolypeptide.

Khoury (personalcommunication) observes an

11,000-daltonpolypeptide.

Mutant d1884, among others, removes one

splicepointutilizedfor production of thesmall

t antigen transcript (12; Fig. 5, position 604/ 4557). It follows that this mutant cannot synthe-sizeasmall t antigen transcript which isspliced atthe normal position; yetd1884-infected cells

containa12,000-daltonsmalltpolypeptide (18;

Khoury, personalcommunication). Possibly, the

mutant polypeptide is produced utilizing an

unspliced transcript. Alternatively, a suitable

transcript may be produced which is spliced at

a newlocation.Analysis of the early viral

tran-scripts present in the cytoplasm of

d1884-in-fected cells should distinguish between these possibilities.

Several of the deletions described in this

on November 10, 2019 by guest

http://jvi.asm.org/

THIMMAPPAYA AND SHENK

portextend to positions very nearsplice points in both the large T andsmall t transcripts. Using such mutants, we can ask how manynucleotides, ifany, adjacent to a splice point must remain intact if a mature, spliced transcript is to be

generated.The d12006 deletion extends towithin

2 base pairs of thesplice point for the small t transcript at position 604/4557 (Fig. 5). If this mutant can produce a bona fide small t tran-script, then no more than two nucleotides on oneside of thesplicepoint are required by the splicing machinery, unless, of course, the nucleo-tides at the far end of the deletion which are fused to the region of interest can fortuitously

performtherequiredfunction. Ifd12006-infected

cells do not containa smallttranscript with a

normal splicepoint, itshouldbesafe toconclude

thatacritical segment has been deleted.

The d1884 deletion extends through position

659/4502 (Fig. 5) and isonly10basepairs from

the splice point (position 670/4991) utilized in

both the large T and small t transcripts. The mutant makes a normal large Tantigen (3, 16,

18) and,therefore, ispresumed tosynthesize a

properly spliced large T transcript. Thed1884 deletion was expanded by the procedure of

Shenk (14) to determine whether any of the

remaining 10 base pairs between the deletion

and the splice point could be removed while maintaining viability (and, thus, the ability to produce a functional large T transcript). Two expanded mutants were produced, d11410 and

d11440 (Fig. 5). Strikingly, neither deletion

ex-tended closer tothesplicepoint, but thed11410

and d11440 deletions extended 6 and 23 base

pairs in the opposite direction, respectively. Since the deletions did not move further toward

thesplicepoint, it is temptingtospeculate that

all or most of the 10 nucleotides between the ends of the deletions and the splice point are

required to produce a functional (i.e., spliced)

large T transcript. We are presently isolating additional expandeddeletion mutants, both vi-able and defective, to test this notion. Mutant

d11440hasacquiredtwoT/A base pairs at

po-sitions658-659/4502-4503 (Fig. 5) which are not presentinitsparent(d1884).Wecannot offer an explanation for this observation. Nucleotide se-quenceanalysisof additionalexpanded deletion mutants will tell whether this is a common event.

Mutant d1891 lacks a segment extending

through position 352/4809, 26 base pairs from thesecondlarge T transcript splice point (Fig. 5, position 325/4836). This deletion was ex-panded to generate d11441 and d11442. The

d11442deletion has notextendedfurther toward

the splice point, but the

d11441

deletion has

moved anadditional 8 base pairs in this direction (Fig. 5). Thus, we set themaximum number of base pairsrequired on the deletable side of this

splice point at 18. Further experiments are in

progress to refine thisestimate.

ACKNOWLEDGMENTS

We acknowledge the competent technical assistance of BonnieMassey.

Thisworkwassupportedby Public Health Service research grantCA-19151 fromtheNationalCancer Institute and was

carriedoutduringthe tenureofanEstablished Investigator-ship of the American Heart Association to T.S.

We thank P. Berg, W.Fiers, G. Khoury, and G.Volckaert forcommunicatingtheirunpublisheddata to us.

LITERATURE CMD

1. Berk,A.J., andP. A.Sharp.1978.Spliced earlymRNAs of SV40. Proc. Natl. Acad. Sci.U.S.A.75:1274-1278.

2. Bouck, N., N. Beales, T. Shenk, P. Berg, and G.

diMayorca. 1978. New region of the SV40 genome required for efficient viraltransformation. Proc. Natl. Acad. Sci. U.S.A.75:2473-2477.

3. Crawford, L.V.,C. N.Cole,A. E.Smith,E.Pancha, P.Tegtmeyer,K.Rundell, andP. Berg. 1978.

Or-ganizationandexpressionofearlygenes ofSV40. Proc.

Natl.Acad. Sci. U.S.A.75:117-121.

4. Fiers,W.,R.Contreras, G. Haegeman,R.Rogiers,A.

VandeVoorde, H. VanHeuverswyn,J. Van

Her-reweghe, G. Volckaert, and M. Ysebaert. 1978.

Completenucleotide sequence of SV40 DNA. Nature

(London)273:113-120.

5. Hirt,B. 1967.SelectiveextractionofpolyomaDNA from infected mousecellcultures.J. Mol. Biol. 26:365-369. 6. Maniatis, T., A. Jeffrey,and D.Kleid.1975.Nucleotide sequenceof therightward operator ofphage A.Proc. Natl. Acad.Sci.U.S.A. 72:1184-1188.

7. Maxam,A., andW.Gilbert. 1977. Anewmethod for

sequencingDNA. Proc. Natl. Acad. Sci. U.S.A. 74:560-564.

8. Mertz,J.E., andP.Berg.1974.DefectiveSV40genomes: isolation andgrowthofindividualclones.Virology62: 112-124.

9. Paucha, E.,A.Mellor,R.Harvey,A. E.Smith,R. M.

Hewick, andM. D.Waterfield.1978.Largeandsmall

tumorantigensfromSV40 have identical aminotermini

mapping at 0.65 map units. Proc. Natl.Acad.Sci. U.S.A. 75:2165-2169.

10. Prives, C., E. Gilboa,M.Revel, andE. Winocour. 1977. Cell-free translation of SV40 early messenger RNAcoding for viralT-antigen.Proc.Natl. Acad. Sci.

U.S.A.74:457-461.

11. Radloff, R.,W.Bauer, andJ.Vinograd.1967.A dye-buoyantdensity methodfor the detection and isolation of closed circularduplexDNA: the closedcircularDNA in HeLacells. Proc. Natl. Acad. Sci. U.S.A. 57:1514-1521.

12.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.Thegenome of SV40. Science 200:494-502.

13.Rundell, K.,J.K.Collins,P.Tegtmeyer,H.L.Ozer,

C.-J. Lai,and D. Nathans. 1977. Identification of simian virus40proteinA.J.Virol.21:636-646. 14. Shenk,T.1977. Abiochemical method forincreasingthe

sizeof deletion mutationsinSV40 DNA. J. Mol. Biol. 113:503-515.

15. Shenk,T. 1978. Construction ofa viableSV40 variant

containingtwofunctionaloriginsof DNAreplication. Cell13:791-798.

on November 10, 2019 by guest

http://jvi.asm.org/

SV40 DELETION

16. Shenk, T.E., J. Carbon, andP.Berg.1975. Construc-tion andanalysis of viable deletionmutantsofsimian virus 40.J. Virol. 18:664-671.

17.Simmona, D. T., and M. A. Martin. 1978. Common

methionine-tryptic peptidesnearthe amino terminal

end ofprimate papovavirustumorantigens. Proc. Natl. Acad. Sci. U.S.A. 75:1131-1135.

18.Sleigh, M.J.,W. C.Topp,R.Hanich,and J. F. Sam-brook. 1978. Mutants of SV40withanalteredsmallt

proteinarereduced intheir abilityto transformcells.

Cell 14:79-88.

19. Takemoto, K. K., R.L Kirchstein, and K. Habel. 1966.Mutants of simian virus40differing in plaquesize, oncogenicity, and heatsensitivity. J. Bacteriol. 92:990-994.

20. Thimmappaya, B., and S. M. Weissman. 1977. The early region ofSV40 DNAmayhavemorethanone gene.Cell 11:837-843.

on November 10, 2019 by guest

http://jvi.asm.org/