Integrated Simian Virus 40 DNA: Nucleotide Sequences at Cell-Virus Recombinant Junctions

(1)

0022-538X/81/050671-09$02.00/0 Vol.38,No. 2

Integrated

Simian

Virus

40 DNA:

Nucleotide

Sequences

at

Cell-Virus

Recombinant

Junctions

JAMESR. STRINGER

Cold SpringHarbor Laboratory, Cold Spring Harbor,New York11724 Received 15 December1980/Accepted13February1981

DNA fragments containing the integrated viral DNA present in the simian virus 40 (SV40)-transformed rat celllines SVRE9 and SVRE17 were cloned in

procaryotic vectors, and the DNAsequences linking SV40 and cell DNA were

determined. Comparison of the DNA sequences atthe SV40-cell junctions in SVRE9and SVRE17 cells with those ofapreviouslycharacterized viral insertion

from SV14B cells shows thatnospecific viralorcellularsequences occurat

SV40-celljunctions and that the cellular DNAsequencesadjacenttointegrated SV40

DNAdonotdisplay the directrepeat structurecharacteristicoftransposonsand retrovirus proviruses.

Transformation of cells by simian virus 40

(SV40) isstabilized by integration of all orpart of the viralgenome into that of the cell. Previous

studies (4,5, 7, 14),inwhich the sizes of restric-tionendonucleasefragmentsbearingSV40DNA were determined by Southern blot analysis of DNAextracted fromavariety of rat and mouse transformed celllines,have shown that integra-tionofSV40 DNA is nonspecific with respect to thepositionontheSV40genome at whichviral

DNAis linked tocellular DNA. These studies also suggested that SV40 DNA does not

inte-grate at specific sites in the cellular genome.

However, this conclusion is subject to two ca-veats. First, short stretches of specific cellular

DNA sequencecould borderviral insertions and

not be detected by restriction endonuclease analysis. Second, cellularDNAcould bealtered

uponSV40 DNAintegration. Thus, integration

at the same site in the cellular genome could give risetorestrictionfragmentsofvarious sizes if different alterations occurred in the surround-ing cellular DNA in different cell lines. This

possibility gains credence from recent

experi-ments in which molecular cloningwas used to

isolateaDNAfragmentwhichcontainstheviral insertion andflanking cellularsequences incell line SV14B(3, 21). Analysis oftheclonedDNA

fragment revealed that in SV14B cells a rear-rangement of cellularsequenceshadoccurredat the site ofSV40 insertion.

Inthestudyreported here,Icloned theSV40

DNA insertions present in the genomes oftwo

transformedratcelllines,SVRE9 andSVRE17,

and haveanalyzedthese insertionsbyrestriction

endonuclease digestion and DNA sequencing.

Theresults of theseexperiments,combined with thoseobtained with the cell line SV14B referred

to above, show that integrated SV40DNA can

belinkedtocellularDNA atnumerouspositions onthe viral genome without sequence

specific-ity. The cellular DNA sequences adjacent to

integrated SV40DNA are also nonspecificand do notdisplaythe repeat structures

character-isticoftransposons orretrovirusproviruses.

MATERIALS AND METHODS Cell lines. The SVRE9 and SVRE17 cell lineswere

first isolated by Pollacketal. (19) after infection of primary cultures of Fisherratembryocells with

viri-onsofathrice plaque-purified stock ofSV40 strain

776.SV14Bcellswereobtainedbyinfectinga

contin-uousline of Fisherratcells withSV40DNA(4). DNAcloning.ThelambdaphageAgtWESABhas been described byLeder et al. (15). The phagewas

modified to accept SstI fragments as follows. XgtWESABDNA wasdigestedwith restriction endo-nucleaseEcoRI, and thelargeterminalfragments (A-arms)werepreparedbysucrosedensitycentrifugation as described by Maniatis et al. (16). A short DNA

fragment containing A DNA from the SstI siteat53.9

mapunitstotheEcoRI siteat54.3mapunits of the A genomewasisolatedbypolyacrylamide gel electro-phoresis and used as a linker between the EcoRI terinini of theA-armsandDNA fragmentsproduced bycleavage withSstI. A mixture of WESB A-arms, SstI-RI linkers, andSstI-digested Charon 4Awas li-gated, and theligation productswerepackedin vitro asdescribedbyBlattneretal. (2)and Maniatisetal.

(16). A resultant phage was plaque purified,

propa-gated, and usedtoprepareA-armswithSstI termini

by the method ofManiatisetal.(16). CellularDNA

waspreparedasdescribed previously (4).DNA

frag-mentsbearing SV40sequenceswereclonedfrom

trans-formed cell DNA by ligation ofSstI digestsof total cell DNAtoSstI-terminalA-arms,followedby

pack-aginginvitroandscreeningforSV40bythemethod

of Benton and Davis(1).

Restriction endonucleaseanalysis. Restriction

671

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

endonucleaseswerepurchasedfrom either Bethesda ResearchLaboratories, Inc., Rockville, Md., orNew EnglandBiolabs, Inc.,Beverly, Mass.,and DNA diges-tionsweredone under the conditionsspecifiedbythe enzymesupplier.Samples ofdigested DNAwere elec-trophoresed onagarosegelsandtransferredto nitro-cellulose by the method of Southern(24).SV40 DNA sequencesweredetectedby hybridizationto radioac-tive SV40 DNA followedbyradioautography as de-scribedpreviously(4).

DNA sequencing. DNA sequences were deter-minedby the method of either Maxam and Gilbert (18) orSangeretal.(22). Both SV40-celljunctionsin SVRE9 were analyzed by the Maxamn and Gilbert method. Thejunctionatnucleotide 2320 (seeFig.3) was'sequenced from theSV40HaeIII siteatnucleotide 2242(SV40 sequence numbersarebyBuchmanetal.;

seereference 25). TheSVRE9junctionatnucleotide 475wassequenced fromacellularHindIIIsite within 15base pairs ofthejoint. The SVRE17 junction at

nucleotide1502inFig.3 wassequencedbythe Maxam and Gilberttechnique. DNAwaslabeled at aHinfl site in thecell DNA. The otherjunction in SVRE17

was sequenced by the chain' termination method of Sangeret al. (22). XSst-17 DNA wasdigested with exonuclease III and annealedto anSV40 DNAprimer, consisting of SV40 sequences fromnucleotide418 to

nucleotide453.Theprimerwasextendedby the Kle-nowfragment of DNApolymeraseI fromEscherichia coli in the presence of different dideoxynucleotide triphosphates, and the products were separated by thin-gel electrophoresis.

RESULTS

Molecularcloning oftheviral insertions

present inSVRE9 and SVRE17 cells. Both SVRE9 andSVRE17 containasingle insertion ofSV40 DNA. ASouthern blot analysis ofthe

DNAfragments produced by digestion oftotal

DNA from these lines with the restriction

en-donuclease SstI is shown in Fig. 1.Ineach case,

asingleDNAfragment hybridizedtoradioactive SV40 DNA. The SstI DNA fragments which

contain the viralinsertionswereclonedby ligat-ingSstI-digestedsamples of totalcellularDNAs to a modified XgtWESXB phage vector which would accept SstI fragments. Recombinant phage were screened forthe presence ofSV40

DNAsequencesbyhybridization with

radioac-tive SV40 DNA. The cloned DNA -fragments obtained from the recombinant phage comi-grated onagarosegels with theSV40-containing

SstI fragments detected in DNA from SVRE9 andSVRE17cells(Fig. 1).

Arrangement of the integrated SV40

DNA in SVRE9 cells and SVRE17 cells.

Physical maps of the DNA fragments cloned

fromSVRE9andSVRE17cells weredetermined

by restriction endonuclease digestion, followed

by agarose gel electrophoresis, Southern

blot-ting, and hybridization to radioactive SV40

FIG. 1. Detection ofDNA fragments containing SV40sequences in cell lines SVRE17 and SVRE9 and theiridentity withfragmentsclonedfromthese cells. Ten micrograms of total cellular DNA was digested with SstI. The fragments wereseparated by electrophoresis through 0.7% agarose, and the viral sequences were detected by blotting and hybridiza-tion asdescribedby Southern (24). A 100-pg amount of recombinant X-phage DNAs bearing the SV40 DNAsequences cloned fromcell lines SVRE17 and SVRE9 weredigested with SstI andrunin parallel withcellular DNA samples. The cloned DNA frag-mentscomigrated with thefragments in total cellular DNAwhichhybridize to SV40 DNA. The low mobility bands in theA-phagereconstructionmigrate with the X vector arm fragments and arepartial digestion products. Tracks: (1) SVREI7 DNA digested with SstI; (2) SstIdigest of a recombinantphagebearing the SV40 DNA insertion from SVRE17 cells; (3) SVRE9 DNA digestedwithSstI; (4)SstIdigestofa recombinantphagebearing the SV40 DNA insertion from SVRE9 cells.

DNA.Theresults ofsomeoftheseexperiments

are shownin Fig. 2, andstructuresof the viral insertions are diagrammed in Fig. 3. All

num-bering of SV40sequencesisbyBuchman etal.;

seereference 25.

TheSV40 insertioninSVRE9 cellsis apartial

tandemrepeat of viral sequences. This structure was deduced from the following observations. Digestionof the DNA from theX-clonecarrying

SVRE9 sequences (XSst-9) with EcoRI

pro-duced three DNA fragments which hybridized

to SV40 DNA (Fig. 2A). The largest of these

http://jvi.asm.org/

[image:2.500.312.404.78.319.2]

(3)

SEQUENCE ANALYSIS OF INTEGRATED SV40 DNA 673

FIG. 2. Electrophoretic patterns of DNA fragmentsproduced by restriction endonuclease digestion of recombinantlambdaphagesXSst-9andXSst-17. (A) DNAfromASst-9andSV40 DNA were digested with the indicated enzymes,electrophoresedon1.0%agarose,stained withethidium bromide, transferred to nitrocel-lulose andhybridized toradioactiveSV40DNAtoreveal theDNAfragmentscontainingSV40 sequences. SV40DNAmarkersarein tracks 3,5, 7, 11, and 13. White on black gels are ethidium stained. Black on white gelsareautoradiograms ofSouthern blotsofthe same gels. (B) DNAfromarecombinantpBR322 plasmid

containingthelargestandsmallest EcoRIfragmentspresentinXSst-9DNA(A; track 2).Tracks1, 3, 5, 7, and 9aredigestsofthe recombinantplasmid DNA, and tracks 2, 4, 6, 8, and 10 are SV40 DNAfragments. (C) DNAfromarecombinantpBR322 which contains the second largest EcoRIfragmentinXSst-9DNA. Track IisSV40DNA;track2isrecombinantplasmidDNA.(D)DNAfromarecombinantpBR322 which contains thesmallest EcoRIfragmentinXSst-9DNA. Track1 isplasmid DNA, and track2isSV40DNA.(E)DNA from XSst-17 and SV40 DNAwere digested with theindicated enzymes and treatedasdescribedfor (A). Tracks1,2,4, 6, 8, and10 areXSst-17 DNA,and tracks 3, 5, 7, 9, and11 areSV40 DNA.

fragments comigrated with formIIISV40DNA, suggestingthat the EcoRI site incircular SV40

DNA waspresent twice in the linear array of

SV40DNA inXSst-9. The SV40 form II DNA

markerinFig. 2A,track3migratedalittlefaster than the largest SV40-containing band from

XSst-9 DNA shown in track 2. This apparent

size differenceisnotreal,but isprobablydueto

thelargeamountofX-phagevectorDNApresent in theXSst-9DNAsample. Whenamixture of

forn IIISV40 DNA andEcoRI-digestedXSst-9 DNA was electrophoresed through an agarose

gel,there was nodifference in the mobilities of form III SV40 DNA and thelargest

SV40-con-taining EcoRI fragment in XSst-9 DNA (data

not shown). Corroborating evidence for a tan-dem repeat arrangement was the presence (in XSst-9)of all theSV40DNAfragmentsproduced

byHindIII, PvulI, HpaI, and Hhal (Fig. 2A).

Confirmation that the largest EcoRI fragment

VOL. 38,1981

http://jvi.asm.org/

[image:3.500.49.445.76.427.2]

(4)

674 STRINGER

(A)

SV40 Restriction Enzyme Map

(B)

EcoRI 2320

9r

17I'

BgI

I

HpaI

BglI 518 1

BamHI EcoRI

.1

I I

HpoI HpaI

475

HpaI

BomHI EcoRI

I- 11502

I 1I1

HpoI HpoI HpoI

BglI aomHI

850 1 1650

148 1 .) C

HoelI HpoI Hpaw

[image:4.500.119.407.61.460.2]

2244 857

FIG. 3. Physical maps of SV40 DNA andoftheDNAfragments clonedfromSVRE9, SVREI 7, and SV14B cells. (A) restriction endonuclease map of SV40DNA. (B)DNA.fragments clonedfrom transformedcells. Open bars representSV40sequencesandaredrawntoscale.Lines representflankingcellular sequences and

are notdrawntoscale. The numbersatthe SV40-celljunctionsindicate the nucleotide numbers in theSV40 DNA sequence atwhich the viral genomesjoincellular DNA. Thenumberingis byBuchman etal. (see reference 25). Thestructureofthe SV14B viral insertion isreproducedfromBotchan etal. (3). The viral insertion in SV14BconsistsoftwosegmentsofviralDNA whichareseparated by40nucleotidesofnonviral sequence. Thepositionofthisjunctionisindicatedbytheindentation in the open bar. Theendpoints ofviral sequences separated by the 40 nucleotides ofnonviral DNA are indicated by the numbers below the indentation.

from XSst-9is identicaltoform IIISV40 DNA

was obtained by subcloning thisfragment into the plasmid pBR322. Double digestionsof this

plasmid with EcoRIand eitherHpaI,PuuII, or

HindIII showed the identity of the subcloned

fragment with SV40DNAlinearized withEcoRI

(Fig. 2B).TheplasmidDNA used in the exper-iments shown in Fig. 2B contained both the

largestand smallestEcoRIfragments

from'XSst-9DNA.ThesmallestXSst-9 EcoRI fragmentis

visible in each of the tracks containing recom-binant plasmid DNA. This fragment was not

digested with either HpaI, PvuII, or HindIII,

and its presence didnotinterferewiththe inter-pretationof theseexperiments.

Theextent towhichtheviral insertionextends

beyondthetwoEcoRI sites inASst-9DNAwas

estimated by restriction endonuclease analysis

of the two EcoRI fragments which migrated

faster than form III SV40 DNA (see Fig. 2A,

http://jvi.asm.org/

(5)

SEQUENCE ANALYSIS OF INTEGRATED SV40 DNA 675

track2). Eachofthese twofragmentswas

sub-cloned in pBR322 and analyzedasshown in Fig. 2C and D. Digestion of the larger subcloned EcoRI fragment with RI and HpaI together generatedasinglefragment which hybridizedto

SV40DNA andcomigratedwith the

1,263-base-pairDNA fragment,correspondingtoviral DNA

betweenthe HpaI site atnucleotide502and the

EcoRI site at nucleotide 1765 (Fig. 20). The

absence ofaseconddetectableSV40-containing DNAfragmentofnovelsizeindicatedthat viral

sequencesdidnotcontinuemuch past the HpaI

siteatnucleotide 502. Therefore, one endpoint of the viral insertion in XSst-9DNA is close to

the HpaI site-atnucleotide502.Thattheother endpoint of the viral insertion in XSst-9 extends beyond the Pstsite atnucleotide 1971 was

in-dicated by the fact that digestion of the small EcoRI subclonedfragmentwithEcoRI andPst

togetherproducedtwoDNAfragments contain-ing the SV40 sequence: one which comigrated with the SV40 marker DNA corresponding to

the viral sequence from the EcoRI site to the

Pstsite at nucleotide 1971 and another larger fragmentof novel size (Fig. 2D). The viral

inser-tionmustend before the BamHI siteat nucleo-tide2516becausedigestion of XSst-9 DNA with

BamHIdidnotproduce afragmentthe size of linearSV40 DNA(Fig.2A,track8).

The arrangement of theviral sequences in the DNA

fragment

cloned from SVRE17cells

(XSst-17)wasdeduced from the data showninFig. 2E. Digestion ofXSst-17DNAwithEcoRIproduced

two SV40-containing DNA fragments ofnovel size,indicating thatasingle EcoRI site is present intheviral insertion. Thelargest stretch of SV40 DNA sequence present inXSst-17 DNA is the

3,790base pairs between the HpaII site at

nu-cleotide 329 and the EcoRI site at nucleotide

1765.DigestionswithHhaI, HpaI,andHindIII localized the endpointsof theviral insertionas

follows.ThepresenceoftheSV40 HindIII

frag-ment Aand absence of the HindIII fragments C, E, and F (Fig. 2E) indicated that viral

se-quencescontinuepasttheEcoRI siteatleastas far as the HindIII site at nucleotide 1691 but

endbefore the next HindIII site atnucleotide 1476.The otherend of the SV40 insertion was

positionedbetweentheHhaIsiteatnucleotide

816andthe HpaIsiteat482.Asshown inFig.

2E, XSst-17 does not contain the small HhaI fragment of SV40 but does contain the viral

HpaIfragmentB.

DNA sequenceslinkingSV40 andcellular

genomesin SVRE9-andSVRE17 cells. The

DNA sequences which link viral and cellular

DNAweredetermined either

by

themethodof MaxamandGilbert(17) or

by

the chain

termi-nationtechniqueofSangeret al.(22). Each viral

insertion endsat a differentpoint on the SV40 map. In Fig. 3 the SV40nucleotide numbers at

whichviralsequencesarejoinedtocellularDNA areshown.Thesequencesat thefour

virus-cell

junctions in SVRE9 and SVRE17 cells are

shown in Fig. 4 together with the

SV40-cell

junctionsinline SV14B,aspreviously described by Botchanet al. (3). In Fig. 4A is shownthe

sequence ofone DNA strandof each insertion running lefttorightfrom

cell

sequenceintoviral

sequence andbackout to cellsequence in 5' to 3'polarity. The

cellular

DNAsequences which

abut the SV40 insertions are all different, and

theyarenotrepeatedineitherdirect orinverse orientation. Fig. 4A also shows that different SV40 sequencesoccur at the end of each inser-tion. This ismorevisibleinFig.4B, in which

all

ofthevirus-celljunctionsaredisplayedin

align-ment and in the same polarity. Fourjunction sequences for celllineSV14BareshowninFig.

4B. Thefirsttwo lines of SV14Bsequence

cor-respond to those at the ends ofthe complete viral insertion. The next two lines of SV14B

sequencearethosefound interposedatthe junc-tion of thetwotractsof SV40DNAwhich

com-prise the complete viral insertionin these cells

(seethelegendtoFig.3andreference 3). If the sequences linking viral and cellular DNAin these

cell

lines are thedirectresult of theprimaryintegrationevent, thenSV40 DNA doesnotintegrateat aspecific sequenceineither the viral or cellular DNA. The question arises as towhether there are other, lessspecific

se-quencefactors, suchasunusualnucleotidebase

composition or repetitious sequence

organiza-tion, which are common to the sequences of

virus-cell

recombinantjoints.Thecompositions

of the first12nucleotide baseson either side of recombinantjunctionsareshown inFig.4C. The

average basecompositionofSV40 DNAis59.3% adenineplusthymine (A+T) (11).A+T nucleo-tideresiduesoccurmorefrequentlythanthisin

the SV40 sequences adjacent to recombinant jointsinfive of sixcases.However, A+T richness

is

generally

not afeatureof theadjoining cellular

sequences or the SV40 sequences which may haveparticipatedin therecombinationevent.

Two virus-cell junctions show

strikingly

un-usualDNAsequenceorganization.The

junction

sequencesshown in lines 1and4of

Fig.

4Bare

bothcomposedofa

simplified

setof nucleotides.

TheSVRE17junctionshown in line4of

Fig.

4B has 27 cytosines and thymidines in a row; the

SVRE9junctionshown in line 1 of

Fig.

4B has 22consecutivecytosinesandadenosines. Within

these stretches of

simplified

base

composition

there are directsequence repeats. In

SVRE17,

VOL. 38,1981

http://jvi.asm.org/

(6)

(A)

5' CELL SV40 SV40

AAGCTTGTAGAGCACAGAAAGGTTAA .TTCTGGAACACA

GAAAGAAGAGGAAGAAGAAAAGGAAG ...ACTGGTAAGTTT

TAGGATGTCTGGCTACTAAAAAATCT...TTTAAATCCTCA

(B) 5' SV40

-TTCTGGAACACA TTAACCTTTCTG

CEL.T., 3'

CACACACACACCAAC SVRE 9

CCAGCTTCTGGCTAT SVRE17

CTTCAGGGGTAAGAG

SJ14B

CELL

CACACACACACCAACAGCCTTGCA TGCTCTACAAGCTT

3'

9

ACTGGTAAGTTTCCAGCTTCTGGCTATTATAAATAA

17

CTTCCTTTTCTT CTTCCTCTTCTTTCTAAATTATTG AGATTTTTTAGTAGCCAGACATCCTAGAATGTGAAG

TTTAAATCCTCACTTCAGGGGTAAGAGTTTCATTCT 14B

A A C T T T G TG A TCITTTG C A A T C C C A C A G ATTG C A G C A C T T T GC A C A A G GIAC A G G C A AG C C T G G T A T G C T G C A A

SV40

5' s

TTCTGGAACACA

TTCTGGAACACA CACACACACACCAACAGCCTTGCAGCAGTGGAAGGGACTTCCCAGATA

TTAACCTTTCTGITGCTCTACAAGCTT

TTAACCTTTCTGGTTTTTTGCGTTTC

67 A C TG GTAAGT T TCCAGCT TCTGGCTATTA TAAATA A 33

ACTGGTAAGTTTAGTCTTTTTGTCTTTTATTTCAGG 67

17

67 C TTCCTTTTCTT CTTCCTCTTCTTTCTAAATTATTG 58

CTTCCTTTTCTT TTTGTTGGGGCCATCTTCATAAGC 42

85 AGATTTTTTAGT AGCCAGACATCCTAGAATGTGAAG 42

AGATTTTTTAGTGACTGGGATCACAAAGTTTCTACT 33

14B 75 T TTAAATCCTCA CT TCAGG G GTAAG AGT TT CATTCT 58

TTTAAATCCTCAAATGGGCAATCCTGATGAACATCA 67

FIG. 4. DNAsequenceslinking SV40 DNAtocellular DNA. (A)Sequences ofonestrandof DNA from each

insertionrunning from cellularsequencesinto the viral insertion, throughthe viralsequencesand backinto

cell DNA in 5'to 3' orientation. (B) SixSV40-cell junctions written in the samepolarity. Thereare two junctions from SVRE9, twofrom SVRE17, and four from SV14B. The top two SV14Bjunctionsare those

whichboundthe complete viral insertion; thebottomtwolines of SV14Bsequence arethoseatthejunction of

the two segments of SV40 DNA which comprise the whole SV14B insertion (see Fig. 3 andreference 3).

Sequences overscored with linesaredirectly repeatedoneitherside of theSV40-celljunction. (C) Sequences

atSV40-celljunctions compared withSV40sequenceswhichwould have participated in the integrationevent. Thetoplineof each doublet is thesequenceoftheSV40-celljunction; the bottom linesarethe parentalSV40

sequences. Possibleremnantsofparental SV40sequences areoverscoredwith lines and theputative donor sequences areunderscored. Thepercentage ofadenosines and thymidines in the parentalSV40sequences

and the cell sequences adjacent to SV40 insertions are indicated in the margins. These numbers were

calculatedfor the12proximal residuesoneither side ofeach junction.

(C)

% A+T

58

67

3' %Ad

42

9 50 50

http://jvi.asm.org/

[image:6.500.54.440.52.559.2]

(7)

SEQUENCE

the terminal SV40 sequences, CTTCCT and

TTCTT, are directly repeated in the adjacent cellular DNA. In SVRE9, the sequence ACACA

isrepeatedoneither side of thevirus-cell

junc-tion. Whetheror notthese repeatstructuresare

significantisunknown,but it isinterestingthat

inimmunoglobulingenerearrangements, recom-binationoccurs near blocksofrepeated hepta-mers(20).

The question of the role of base pairing in

SV40 integration isnotdirectly addressed by the data presented here because the sequences of

parental cellular DNAs which took part in

re-combination arestill unknown. However,some

cluesmay be found bycomparing the cellular

sequences whichflank SV40 insertions with the SV40 sequences which couldhavetakenpart in

partially mismatched basepairingatthe siteof integration.Itispossiblethatduringintegration, cellular and viral DNA sequences sharingpartial sequencehomology formpartiallymismatched

heteroduplexes.Mismatches in theheteroduplex

regioncould be repaired randomly, sometimes conserving DNA sequences of SV40origin and sometimes copying cellular sequences. This

processwould beexpectedtoleaveremnantsof SV40 sequences interspersed with nonviral

se-quences inthe DNA proximalto the cell-virus

junction. Sequences which may be vestigial

SV40 DNA are present in various amounts in the DNA adjacent to several virus-cell joints

(Fig. 40). However, the presence of such puta-tive remnants of viral DNA sequence is not common to all virus-cell DNA junctions

se-quenced.

DISCUSSION

My analysis of DNA fragments cloned from

cell lines SVRE9 andSVRE17, coupledwiththe work of Botchanetal.(3)onSV14Bcells,shows

thatnospecificviralorcellular sequencesoccur at the SV40-cell junctions of three different SV40 insertions. These findings extend to the

level of DNA sequence the conclusion thatSV40

integration is not site specific on the viral ge-nome and provide the first unambiguous evi-dence thatnosingle DNAsequence inthe cel-lulargenome servesas anobligatory

integration

site for SV40 DNA. The DNAsequenceswhich occur at the recombinant junctions between

SV40 andcell DNA showstructural similarities

insome cases, but there isnostructuralfeature common to all ofthe junctions.There is some

preference for A+T richness in the SV40

se-quences at the termini of viral insertions, but

adjacentcell sequencesare notgenerallyrichin A+T residues. Two SV40-cell junctions are

strikingin that DNAsequencesarerepeatedon

either side of the joint. However, these repeat structures are not present at every

SV40-cell

junctionsequenced.

SV40-cell

junctions are quite nondescript comparedwiththejunctions ofretrovirus

pro-viruses (8, 23), copia elements in Drosophila species(9),andTyl elementsinyeastcells(10, 12). These elements all have specific sites in

theirDNAsequenceatwhich they attach to

cell

DNA, whereas SV40 insertions show no such

specificity. Retroviruses, copia elements, and Tyl elementscaninsertatmultiplesites in

cell

DNA, but their insertion is accompanied by

duplication ofashortsequencewhichpreexisted

inthecellDNA attheintegrationsite, generat-ingashortrepeatofhostsequences flankingthe

DNA insertion. Integrated SV40 DNA is not

flanked by direct repeats of cell DNA. Trans-position and retrovirus integration do not

per-turb hostDNAsequences beyondgenerationof

the shortduplicationsatthe siteofinsertion(6).

Integration of SV40DNA canresultinextensive changes in the hostgenome. Botchanet al. (3)

have shownthat arearrangement ofcellDNA occurs atthe site of SV40 insertion in SV14B

cells.

Preliminaryresults indicatethat the same

is true in SVRE9 and SVRE17 cells (J. R.

Stringer, manuscript inpreparation).

Clearly, the mechanism of SV40 integrationis

quitedifferentfrom thoseacting intransposition andinintegration of retrovirusproviruses. Itis

probable that SV40 DNA integrates via the

function

of some generalized cellular recombi-nationsystemwhichcanratherindiscriminantly integrateheterologousDNAs.Culturedcells

ap-parentlypossessamechanism whichcan

incor-porateanyforeign DNA introducedtothe cell

(26). Because integration is a general

phenom-ena notuniquetoviral genomes, thefunctionof SV40geneproductsisprobablynotrequired for integration, although SV40 proteins may

en-hance the action of a cellular recombination

system. Itis ofcoursepossiblethat the nucleo-tide sequences of the cell-virus recombinant

junctionsanalyzed in this workaretheresult of

secondaryrearrangementsafterintegrationand thatSV40integrationmayproceedviaa

trans-position-like mechanism. For this to be true,

SV40 insertions wouldhaveto(i)be very unsta-ble after the original insertion eventand then,

after rearrangement, becomeverystable struc-tures and (ii) still be different from movable genetic elements and retroviruses by virtue of this temporarystructuralinstability.

Whatever themechanism ofSV40integration,

itissimilar to that which mediates theformation

ofdefectivevariantsofSV40whichare substi-VOL. 38,1981

http://jvi.asm.org/

(8)

tuted with cellular DNA (13). The

recombina-tionsites in thesevariants are similarto those foundlinking integrated SV40DNA tothe cel-lulargenome in that (i) theyare notsequence

specific; (ii) they often occur in regions of the SV40 genomewhich are rich inadenosine and thymidine; and(iii) putative remnantsofSV40

sequence occur incellularsequencesofsomeof the recombinants, suggesting that the DNA strandsparticipatingin therecombinationevent sometimessharepatches ofsequencehomology. These characteristics are suggestive of a

mechanism of recombination which does not

require perfect DNA sequence homologies

be-tween parental DNA strands. Ofcourse to

re-solve fully whether or not DNA sequence

ho-mologies contribute to integrative recombina-tion, it will be necessary to use the flanking cellularsequences in the DNAclones described

here asprobes toclonecellularDNAfragments containingtheviralintegration sitesin

untrans-formedratcells.Thesequencesof both parental DNAs, viral andcellular,could then bedirectly compared with each other and with the DNA

sequencesatthevirus-cell junctions. Two stud-ies ofother SV40-recombinants have been

re-ported in which both parental DNA sequences

were determined and compared with those of

therecombinant. Zain and Roberts(27)analyzed

anadenovirus-SV40 recombinant andfoundno sequencehomology between the regions of

ade-novirus DNA and SV40DNA involved in the

recombination event.Gutai studied the nucleo-tide sequences atviral-viral recombinantjoints in naturally arising defective variants ofSV40

andfound some instancesofpatchyhomology between parental DNA sequences, but other

recombinants showedlittleor no suchhomology. (M.W.Gutai, Virology,inpress).

The studies cited above and others have

shown that eucaryotic cellshave the abilityto

recombine DNAillegitirnately, i.e.,without the

mediation of extensive DNA sequence homol-ogy. Itwill beparticularly interestingto deter-mine whether DNA sequence homology is in-volved in integration in SVRE9 and SVRE17

cells. Thedirectly repeated DNAsequenceson either side ofthevirus-cellDNA jointsinthese twoinsertions(Fig. 4B) must havepresentedthe cellwithregionsofhomology which couldhave beenused in therecombination reactionifDNA

homologywere importantinthemechanism. If

SV40DNAdid integratein thesecellsvia ille-gitimate recombination, it was not for lack of

sequencehomologies between proximalcellular

and viralDNAs.

ACKNOWLEDGMENTS

IthankLaurelGarbarini and JeanneLanzarone for

excel-lent technical assistance and Patti Pope,MikeOckler, and

Cindy Carpenterforpreparationof thismanuscript.

Iwassupportedbyfellowshipsfrom the American Cancer

Society (PF-1565) and Public Health Service grant 1 F32 CA06460-01 from the National Institutes of Health.

LITERATURE CITED

1. Benton,W.D.,andR. Davis.1977.ScreeningXgt recom-binant clonesby hybridizationtosingleplaques.Science

196:180-182.

2. Blattner,F.R., A. E.Blechl,K.Penniston-Thomson,

H. E.Faber,J. E.Richards,J. L,Slightom,P. W.

Tucker,andD.Smithies. 1978.Cloninghuman fetal yglobinand mousea-type globinDNA: preparation

andscreening.Science 202:1279.

3. Botchan,M.,J.Stringer,T.Mitchison,and J. Sam-brook. 1980.Integrationand excision ofSV40 DNA from thechromosome ofatransformed rat cell.Cell 20: 143-152.

4. Botchan, M.,W.Topp,and J.Sambrook. 1976. The arrangement of simian virus 40 sequences in the DNA oftransformedcells.Cell 9:269-287.

5. Botchan,M.R.,W. C.Topp,and J.Sambrook. 1979. Studiesonsimian virus 40 excision from cellular chro-mosomes.ColdSpringHarborSymp. Quant.Biol. 43: 709-719.

6. Calos, M. P., and J. H. Miller. 1980. Transposable

elements.Cell20:579-595.

7. Chepelinsky,A.B.,R.Seif,and R. G.Martin. 1980.

Integrationofthesimianvirus40genome intocellular

DNA in temperature-sensitive (N) and temperature-insensitive (A)transformants of 3T3 rat andChinese hamsterlungcells. J.Virol. 35:184-193.

8. Dhar, R.,W. L.McClements, L. W. Enquist, and G. F.Van deWoude. 1980.Nucleotide sequences of

in-tegrated Moloneysarcoma provirus long terminal re-peats and their host and viral junctions. Proc. Natl. Acad. Sci.U.S.A.77:3937-3941.

9. Dunsmuir, P.,W. J.Brorein,M.A.Simon, and G. M. Rubin. 1980.Insertion of thedrosophila transposable elementcopiageneratesa5basepair duplication.Cell 21:575-579.

10. Farabaugh, P. J., and G. R. Fink.1980.Insertion of the eucaryotic transposable element Ty 1 creates a 5-base

pairduplication. Nature (London) 286:352-356. 11. Fiers, W., R. Contreras, G.Haegeman,R.Rogiers,A.

van deVoorde, J. VanHerreweghe, G. Volckaert, and M.Ysebaert. 1978. Completenucleotide sequence ofSV40 DNA. Nature (London)273:113-120. 12. Gafner,J., and P. Philipsen. 1980. The yeast transposon

Ty 1generatesduplications of target DNA on insertion. Nature(London) 286:414-418.

13. Gutai, M. W., and D. Nathans. 1978.Evolutionary var-iants of simian virus 40: cellular DNAsequences and sequences at recombinant joints of substituted variants. J.Mol. Biol.126:275-288.

14. Ketner, G.,andT.Kelly, Jr. 1976. Integrated simian virus 40 sequences intransformed cell DNA: Analysis using restriction endonucleases. Proc. Natl. Acad. Sci. U.S.A.73:1102-1106.

15. Leder, P., D. Tiemier, and L. Enquist. 1977. EK2 derivatives ofbacteriophage lambda useful in the clon-ingof DNA from higherorganisms: The XgtWES sys-tem.Science 196:175-177.

16.Maniatis, T., R. C. Hardison, E. Lacy, J. Lauer, C. O'Connell, D. Quon, G. K. Sim, and A. Efstratiadis. 1978.The isolation of structural genes from libraries of eucaryotic DNA. Cell15:687-701.

17.Maxam, A. M., and A.Gilbert. 1977. A new method for sequencing DNA. Proc. Natl. Acad. Sci. U.S.A. 74:560-564.

18.Maxam, A. M., and A. Gilbert. 1980.Sequencing

http://jvi.asm.org/

(9)

labeled DNA with base-specific chemical cleavages. Methods Enzymol. 65:499-560.

19. Pollack, R., R. Risser, S. Conlon, and R. Rifkin.1974. Plasminogen activator productionaccompanieslossof

anchorage regulationintransformation ofprimaryrat

embryocells by simian virus40.Proc. Natl.Acad.Sci. U.S.A. 71:4792-4796.

20.Sakano, H.,R.Maki,Y.Kurosawa,W.Roeder, and S.Tonegawa. 1980.Twotypesof somatic recombina-tion arenecessaryfor thegeneration of complete im-munoglobulin heavy-chain genes. Nature (London) 286:676-683.

21. Sambrook, J., R.Greene, J. Stringer, T. Mitchison, S.LHu, and M. R. Botchan. 1980.Analysis of the

sites ofintegration of viral DNAsequences in cells transformedby adenovirus2and SV40. Cold Spring HarborSymp. Quant. Biol.43:569-584.

22. Sanger,R.,S.Nicklen,and A.R.Coulson. 1977. DNA

sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. U.S.A. 74:5463-5467.

23. Shimotohno,K.,S.Mizutani,and H.M. Temin.1980. Nucleotide sequences flanking retrovirus proviruses. Nature(London) 285:550-554.

24. Southern, E. 1975. Detection of specificsequences among DNA fragments separated by gelelectrophoresis. J. Mol. Biol. 98:503-517.

25.Tooze, J.1980. The molecularbiologyoftumorviruses, 2nd ed. ColdSpringHarborLaboratory,Cold Spring Harbor,N.Y.

26. Wigler, M., R. Sweet,G. K.Sim,B.Wold,A.Pellicer, E.Lacy, T.Maniatis, S. Silverstein, and R. Axel. 1979. Transformation ofmammal cellswithgenes

fromprocaryotesandeucaryotes.Cell 16:777-785. 27. Zain,B.S.,andR. Roberts.1978. Characterization and

sequence analysis ofa recombination site in hybrid

virusAd2+ND1.J.Mol. Biol. 120:13-32.

http://jvi.asm.org/