Integration of the simian virus 40 genome into cellular DNA in temperature-sensitive (N) and temperature-insensitive (A) transformants of 3T3 rat and Chinese hamster lung cells.

0022-538X/80/07-0184/10$02.00/0

Integration of the Simian Virus

40

Genome into

Cellular

DNA

in Temperature-Sensitive

(N) and

Temperature-Insensitive

(A) Transformants of 3T3

Rat

and Chinese Hamster

Lung

Cells

ANA BERTA CHEPELINSKY,* ROLAND SEIF, AND ROBERT G. MARTIN

Laboratory of Molecular Biology,National Institute ofArthritis,Metabolism and DigestiveDiseases,

NationalInstitutesofHealth, Bethesda, Maryland20205

We studied thepattern ofintegration of the simian virus 40 (SV40) genome into the cellular DNA of N-transformants (temperature sensitive) and

A-trans-formants (temperature insensitive) derived from 3T3-Fisher rat and Chinese hamster lungcells.The SV40DNAwascovalently linkedtothecellular DNA in

both typesof transformants. In therat cells, most N-transformants contained

SV40sequencesintegratedatasingle site;mostA-transformants contained SV40

sequences integrated at two to five sites. In the Chinese hamster cells, no significant correlation between the number of integration sites and the phenotype of the transformantwasfound; onetothree integration siteswereobserved for both the N- and A-transformants. Single copies and tandem repeats of SV40

sequences were observed in A- and N-transfornants derived from rat cells. A-transformants arise neither by amplification of the SV40genome norby integra-tionataunique site.

Infection of mammalian cells by tsA mutants of simian virus 40 (SV40) and polyoma virus

gives risetotransformantsthat areeither tem-perature sensitive(N) or temperature insensitive

(A) (7-10, 12, 13, 17, 20-23, 26-28, 32). The

relative proportions of the two types of trans-formants seem todependupon several parame-ters, including thegrowth state ofthe cells for

thefirst few

days postinfection

(23, 26, 27) and the ability of the virus to express the small-t

antigen (28).Although several models have been proposed (12, 14,23,26-28), the molecular

mech-anism(s)

responsible

forthe difference between

the two types of transformants has not been

found yet. In an attempt to gainsome insight

into these mechanisms, we examined the

inte-grationpatternsoftheSV40genome into

cellu-lar DNAin anumberof N- andA-transformed

rat and hamstercell lines.

Ithas beenreportedthat the numberof

inte-grationsitesandnumber ofcopies of theSV40

genome integrated per site varied from one

transformed cell line to another (5, 16). This

studywasundertakentodetermine

what,

ifany,

correlations existbetween the temperature sen-sitivityof the tsAmutant-transformedcelllines

andthe number ofcopies of the SV40genome

integrated. Wefoundthat the SV40genome is

integratedinto host sequences in N-transform-ants aswell as in A-transformants and thatno specialpattem ofintegrationisassociated with the difference intemperaturesensitivity.

MATERIALS AND METHODS

Transformed cells. The N- and A-type tsA209,

SV40-transformed rat cells and the tsA209 and tsA209 viable deletion double-mutant-transformed Chinese hamsterlung (CHL) cellshave been described

previ-ously (20, 27, 28).

ViralDNA. TheSV40form I DNA used for nick translation and also toobtain form II and III DNAs thatwereusedasstandardswaspurchasedfrom Be-thesda Research Laboratories.

Isolation of high-molecular-weight cellular DNA.Ratcells orCHL cellsweregrowntoconfluence

in Dulbecco-Vogt modified Eagle medium

supple-mented with 10% fetal calf serum. After cells were

washed with phosphate-buffered saline, they were

treated with lysisbuffer(10mM Tris-hydrochloride,

pH 7.9,1 mM, EDTA,0.5%sodiumdodecylsulfate),

incubated with 0.5mgofproteinase K (Boehringer)

perml overnight at 370C, and extracted twice with

phenol andoncewithchloroform-isoamylalcohol(24:

1,vol/vol).Theextractswerethendialyzed,incubated

with 0.1 mgof RNase A (Boehringer) per ml, incu-bated again with proteinase K (0.1 mg/ml),

reex-tracted, anddialyzed,asdescribedbyBotchanet al.

(4). The molecularweightof the DNAwaschecked byagarosegel electrophoresis.

Restriction enzymes, digestion, and agarose

gels.Cellular DNA(25 ,Lg)wasdigestedfor 2to5 h

with restrictionenzymesby usingtheincubation

me-dia specified by New England Biolabsor Bethesda

ResearchLaboratories, dependingonthesourceofthe enzyme. Afterincubation, 20mM EDTA and 0.1to

0.2M NaCl (final concentrations) were added. The

samples were extracted once with phenol and once

withchloroform-isoamylalcohol (24:1). Sodium

ace-184

on November 10, 2019 by guest

http://jvi.asm.org/

INTEGRATION OF SV40 GENOME INTO CELLULAR DNA 185

tate(pH 6.0;finalconcentration,0.2M)and 2volumes

of absoluteethanolwereadded,and the solutionswere

stored overnightat-20°Cand then centrifugedina

Brinkmancentrifugefor15min.Theprecipitateswere

suspendedinasolutioncontaining10 mM

Tris-hydro-chloride(pH 7.8),6 mMEDTA,0.2mgof bromophe-nol blue per ml, and 7% glycerol and resolved by

electrophoresis in a 0.7% agarose (Marine Coiloids)

horizontal slab gel in a solution containing 40 mM

Tris-acetate (pH 7.8), 5 mM sodiumacetate, and 1 mM EDTA (31) for 16 to 18 h at 40 V at room

temperature. LambdaphageDNAHindIIIfragments

labeledwith32P attheir 5' ends(BethesdaResearch

Laboratories) were used as DNA molecular weight

standards.

Blotting.Transfer of DNA from thegelsto

nitro-cellulosepaperwasperformedbythe Southern

tech-nique(30)withsomemodifications (5, 16).Gelswere

stained with 2.5 pgof ethidium bromideperml,and

picturesweretaken underUVillumination. Thegels

werethen treated with 0.2 M NaOH-0.6MNaCl for 45 to 60min, neutalizedwith 1 MTris-hydrochloride

(pH7.3)-0.6MNaCl for 45min,andequilibratedwith

6xSSC (lxSSC is 0.15 M NaClplus0.015M sodium

citrate)for 30 to 45minwithgentlerotation.Blotting

onnitroceliulose membrane filterpaper(0.45pm;type

HAWP; Millipore Corp.) was performed downward

(gelatthetop)forapproximately3h andupwardfor

approximately20h, soakingthe 3 MMpaperbridgein

6x SSC. The nitrocellulose paper wasrinsedfor 15

minin 6x SSC and air dried. Itwasbaked for2 h at

800Cunderavacuumandkeptatroomtemperature until itwastobe used.

RadioactivelabelingofSV40DNA. Nick trans-lationwasperformedbythe method of Maniatisetal.

(18),with modificationssuggested byShoshanaSegall

(personal communication). The incubation mixture

contained [a-3P]dATP [tetra(triethylammonium)

salt]and[a-32P]dCTP[tetra(triethylammonium) salt]

(300to500Ci/mmol [New EnglandNuclearCorp.]or

approximately 2,000 to 3,000 Ci/mmol [Amersham

Corp.] previouslydried),each ataconcentration of 3.6

pM; the mixture also contained 3.6 pM unlabeled

dGTP, 3.6 pMTTP,50mMTris-hydrochloride (pH

7.8),3 mMMgCl2,10 mM,B-mercaptoethanol,50 pgof

bovineserumalbuminperml,and 10 pgof SV40 form I DNA(BethesdaResearchLaboratories)perml. The mixturewaspreincubatedfor 15minat150C.

Esche-richia coli DNApolymeraseI(40U/ml;Boehringer)

and 1.2 ng of DNase I (Worthington Biochemicals

Corp.) per ml were thenadded, andthe incubation

wascontinued for 110to120min (untilaplateauwas

reached in theincorporationof3p into trichloroacetic

acid-precipitable material). A solutioncontaining 10

mMTris-hydrochloride (pH 7.5),0.1MNaCl,10mM

EDTA, and 0.5% sodiumdodecylsulfate wasadded;

the sample was extracted with phenol and passed

through a Sephadex G-50 fine column equilibrated

with 10 mMTris-hydrochloride (pH 7.5)-0.1M

NaCl-1 mM EDTA, and the fractions that contained the

labeled DNAwerepooled.Anapproximate yield of1.5 x 10Wto6x108cpm/pugofDNAwasobtained.

Hybridization.Blottednitrocellulosepapers were

soaked in 3xSSC forapproximately45minatroom

temperature.Theywerethenpreincubatedfor 3 h in

3xSSC-10x Denhardt solution (lx Denhardt

solu-tion contains 0.02% bovine serum albumin, 0.02%

Fi-coll, and 0.02%polyvinylpyrrolidone)at68°C and later

with a solution containing 3x SSC, 1OX Denhardt

solution, 0.1% sodiumdodecyl sulfate, 10 pg of

poly-adenylic acid per ml, and 100pg of sheared and

dena-turedsalmon sperm DNA per ml for1hat680C. The

hybridization mixture contained 3x SSC, lOx

Den-hardt solution, 0.1% sodium dodecyl sulfate,10 pg of

polyadenylic acid per ml, 100 pig of sheared and

dena-tured salmon sperm DNA perml,and 0.9x 107to1.2

X 107cpm of denatured32P-labeledSV40 DNA

(ob-tainedby nicktranslation; concentration, 20 to 90 ng/

ml) per ml in a total volume of 7 to 14 ml; the

hybridizationwasinsealedbagsat680C for

approxi-mately 40 h. Afterhybridization, therewere four to

fivewashings withasolutioncontaining3xSSC,1Ox

Denhardtsolution, 0.1% sodium dodecylsulfate, and

50pg of sheared and denatured salmon sperm DNA

per ml and onewashing with a solution containing

0.lx SSC, 50 pg of denatured and sheared salmon

sperm DNAperml,and 0.1% sodiumdodecylsulfate

at680C; the filter paperwasthen rinsedoncewith3x

SSCat roomtemperature(15). Milliporefilterpapers

wereairdried andexposedtoKodak X-Omat R film

with Max IPickerscreens at-70°Cfor2 to 9days.

RESULTS

Integration of theSV40genomein N- and A-type tsA209 and tsA2O9 viable deletion

double-mutant-transformed CHL celis. High-molecular-weight cellular DNA from CHL cellstransformed bythe tsA209and tsA209 vi-able deletion double mutants ofSV40 was di-gested with restriction enzymes that cut the cellular DNA but not SV40 DNA. One of the limitations of the Southern technique is that the transfer efficiency of pieces of DNA

approxi-mately20,000 base pairs long or longer is lower than that of smaller pieces (1, 5) and may go undetected. Therefore, several restriction en-zymes were used, and the highest number of bandsobtainedwith any of them was considered the minimumnumberofintegration sites. The enzymes used were BglII,SacI,SstI (isoschizo-merofSacI),and XbaI.

Figure 1shows the patterns of integration of SV40 into the cellular DNA of several N- and

A-type CHL cell lines transformed either by

tsA209 or by one of several double mutants. Sincehamstercells are semipermissive forSV40, the cells were grown at

410C

in orderto avoid the presence of free SV40 DNA thatoccurs at

330C (19). No form I or II SV40 DNA was observed.

Table 1shows theminiimnum number of

inte-grationsites and the sizes of thedifferent bands

obtained with each restriction enzyme for the tsA209 N-andA-transformants. Fromthe sizes oftherestrictionfragments itispossible to

cal-culate the maximum number of copies of the SV40 genome possible per integration site; i.e.,

VOL. 35,1980

on November 10, 2019 by guest

http://jvi.asm.org/

186 CHEPELINSKY, SEIF, AND MARTIN

A 2 3 4 5 6 7 8 9

lo

11

} 3~~~~~~~~~~~~~~~~~~

B _{| 2 3 4 .5}

_ W*W

* .:.*

FS-... p>. ^

..:.eF...

.... ^ * .. r.: .;. ,¢

X,

..

*.

...

i.. i z :..6..*, ..:.

:f~~~~~~~~~~~~:

I ! -Oo

[image:3.508.63.460.62.333.2]

-s4-0

FIG. 1. HybridizationpatternsofCHLtransformed ceUlineswithSV40DNAno-cutenzymes.(A)Cellular

DNAwasdigested withBglII,runingels,and hybridizedasindicated in thetext.Lane1,32P-labeledADNA

HindIII fragments (numbers ofbasepairsperbandareindicatedbyarrows;I, II, and III,SV40 form I, II,

andIII DNA);lane2, CHLA209L4;lane 3,CHLA209L5; lane4,CHLA209L38; lane5,CHLA209L49;lane6,

CHLA209L62; lane 7, CHLA209L51; lane 8, CHLA209A289L5; lane 9, CHLA209A289L4; lane 10,

CHLA209A88L3;lane11,SV40forn II,III,andIDNAs. (B) XbaI digestion ofCHL cellular DNA. Lane1,

CHLA209L51; lane2, CHLA209L38; lane3, CHLA209A29OL4; lane4, standardsasin (A), lane 1; lane5,

CHLA209A292L14.

CHLA209L4andCHLA209L5must have fewer

than two full-size SV40 DNA copies each,

whereasCHLA209L38 andCHLA209L62 must

have fewerthanthreecopiesandCHLA209L49 musthave fewer than fourcopies.One to three

integration siteswereobserved in the different N-and A-transformed celllines, andthepatterns varied both between the N- and A-types and amongthe differentN-and A-transformants.

N-and A-transformantsofCHLcells induced

by tsA209 viable deletion double mutants (12)

werealso examined.Thenumber ofintegration

sitesvariedfromonetothree,and themaXimum number ofcopies of the SV40genome per

mte-gration site varied from fewer than two to

fewer than six (Table 2). In cell lines

CHLA209,6292L14,

CHLA209A287L5, and

CHLA209A292L19 bands of similar molecular

weights(insidethelimits oferror)wereobserved after digestion with XbaI. When these results

werecombinedwith the results withthe tsA and

double-mutant transformants, 4 of 12

N-trans-formants appearedto have arisen from the

in-sertion of SV40 ata single site, and 2 of10

A-transformantsarosesimilarly. Itmaybe

signifi-cantthatof11celllinesestablishedat multiplic-itiesofinfection from 0.065to0.48(Tables1and 2), 7 appeared to have SV40 integrated at a single site, whereas of12celllinesestablishedat

multiplicities of infection from 0.7to3,nonehad SV40integratedatonlyasingle site.

In some of thecelllinesthatappeartohave SV40 DNAintegratedatonlyonesite, the cel-lular DNAwastreated with restrictionenzymes

that cut the SV40 genome once (e.g., BglI,

TaqI, BamHI, EcoRI). Thepresence oflinear, full-length SV40 DNA (formn III) after this treat-mentwould indicatearepetition of the relevant site. The flanking sequences bound to cellular DNAmay ormaynotbedetected accordingto

their sizes and/or the size of the SV40 region attached (Table 3). No evidence formorethan asingle full-length insertionwasfoundfor sev-eral N-type cell lines (e.g., CHLA209L5 and CHLA209L62),whereasrepetition of the TaqI, BamHI, and EcoRI sequences wasobserved in CHLA209L49,anA-typeline.

IntegrationoftheSV40genomein N-and

J. VIROL.

-4-0

J:io

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.508.53.447.76.379.2]

INTEGRATION OF SV40 GENOME INTO CELLULAR DNA 187

TABLE 1. Number of SV40 genome integration sites intsA209-transformedCHLcellsa

No. of

Cellline Molb Type of_formanttrans- integra-_{tion containing SV40}No. of base pairs per_sequences'band Restriction_enzyme sites

CHLA209L5 0.48 N 1 7,900 BglII

1 NDd SstI

CHLA209L62 0.065 N 1 11,400 BgllI

1 13,800 XbaI

CHLA209L51 0.065 N 2 8,000,8,800 XbaI

3 7,200, 8,400, 9,700 BglII

CHLA209L61 0.065 N 1 26,800 XbaI

2 20,900, 25,100 BgIII

CHLA209L56 0.065 N 1 15,800 XbaI

1 10,100 BgllI

CHLA209L4 0.48 N or A 1 7,600 BgII

1 16,000 XbaI

1 ND SstI

CHLA209L49 0.065 A 1 18,100 BglII

1 15,900 XbaI

CHLA209L38 0.15 A 1 17,400 BglII

1 11,900 XbaI

1 13,700 SstI

CHLA209L34 0.15 A 2 9,500, 11,200 XbaI

3 7,900,9,500, 13,100 BglII

aDNA fromCHLtsA209-transformed

cell lines

wasincubated with restriction enzymesasdescribed in the

text.

bMOI,Multiplicityofinfection(see reference 20).

'The number of basepairswasdeterminedby using3P-labeled A DNA

HindIII

fragmentsasstandards(see

Fig. 1). Theerrorinthedetermination of basepairs,whichwasbasedondataobtained in differentexperiments,

wasapproximately200basepairs forfragments smaller than 9,500 base pairs; however, forfragments10,000to

30,000 basepairslong itwasonthe order of 1,000 basepairs insome casesduetothehigherslopeof thecurve

ofmolecularweightversusmobilityin thatregion(2, 5).

d ND, Not

determined.

A-type tsA209-transformed rat celis. The

integrationof theSV40genomeintothecellular

DNAs of N- and A-transformed rat cells was alsostudied. Someof the patterns ofintegration obtainedwiththe rattransformantsare shown in Fig. 2. Asexpected, nofreeSV40DNAwas observed since rat cells are nonpermissive for SV40.Three of theA-typecell lines had similar

integration patterns withtwo different

restric-tionendonucleasesthatdo not cutSV40 DNA.

Cell lines A209-FR3T3-A30 and

A209-FR3T3-A37wereobtainedfromthesameflask;it cannot

berigorously excluded theyarenotderivedfrom

the same transformed cell. However,

A209-FR3T3-A43 was isolated in a separate

experi-ment and must therefore have arisen by an independentevent.

Table4 showstheresults obtained with

sev-eral restriction enzymes that do not cut SV40 DNA.Amongthe

N-transformants,

one to three

integrationsites were observed, and among the

A-transformants one to five integration sites were observed. For the N-transformants there was a predominance of single-integration-site

patterns (5 of8), whereasfor the A-transform-antsmultipleintegration sites predominated (9

of10).

Insertions of sizes smaller than full-length

SV40 DNA have been observed, as in

A209-FR3T3-A99. In a long fihn exposure of A209-FR3T3-A43 digested with anSV40DNA no-cut enzyme, a light band of 4,000 base pairs was

observed,whichmay have been due tothe pres-ence of a small piece of SV40 DNA in that

fragment.

A number ofthecell lineswere examined after

VOL. 35,1980

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.508.66.455.94.510.2]

TABLE 2. Numberof SV40 genome integration sites in viable deletion tsA209double-mutant-transformed

CHL cells

Cellline CHLA209A288L3

CHLA209A289L14

CHLA209A29OL5

CHLA209A29OL12

CHLA209A291L4

CHLA209A292L14

CHLA209A292L16

CHLA209A287L5

CHLA209A287L6

CHLA209A289L5

CHLA209A290L4

CHLA209A291L2

CHLA209A291L13

CHLA209A292L19

Type of No.of in-MOja transform- tegration ant sites

2.3 N 1

2 2

3 N 2

2

1 N 2

3

0.14 N 1

1

1.6 N 1

1 2

2.9 N 2

3

2.9 N 2

2

2 A 2

3

0.9 A 2

2

2 A 2

0.7 A 3

3

2.4 A 2

3

0.3 A 1

2

No. of base pairs per band con-tainingSV40sequencesb

16,100 19,500, 23,000 12,900, 16,500

6,600, 13,200 9,400, 18,000

NDC

7,900, 20,000, 28,100

19,000 19,200

12,200 28,600 10,700, 17,000

ND

10,400, 15,500, 21,000

15,300, 19,700 17,300, 25,000

19,800, 23,000 10,800, 15,500, 23,700

19,500, 24,500 17,000, 24,000

9,100, 11,600

7,100, 14,500, 23,700 10,500, 13,800, 17,000

11,300, 21,500 ND

8,000 ND

0.9 A 2 ND

3 10,800,15,200, 22,900

aMOI, Multiplicity of infection (see reference20).

'See

Table 1,footnotec.

'ND, Notdetermined.

digestionwithrestrictionenzymesthatcutSV40

DNAonce.Figure3shows thepatternsobtained

after digestion ofcellular DNA from somecell

lines with two of these enzymes, and Table 5

summarizes all of theseresults. Among the

N-transformants withoneintegration site,the

fol-lowingobservations weremade. (i) One (A209-FR3T3-N94)contained fewerthan 1.5SV40

ge-nomic equivalents (sequence repetitions of the

BglI,BamHI,andTaqIsiteswerenotpresent).

(ii) Three transformants contained partial

re-peats of theSV40genome,asdemonstratedby

thefactthatsomerestriction enzymesthat cut

SV40DNAoncegaverise tofull-lengthformIII

DNA,whereas others did not. And(iii)no

trans-formant contained a complete duplication. In

theonlyA-transformantcontainingone

integra-tion site(A209-FR3T3-A12)fewerthan 1.4

cop-ies of SV40 DNA were present (BglI, TaqI, HpaII, and BamHI did not release form III

SV40 DNA). In the cell lines withmore than

oneintegration site,sevenof ninecontained at Restriction

enzyme

SstI XbaI BglII

BglIl XbaI

Bgll

SstI

XbaI

BgllI

BglII

XbaI SstI

BgllI

XbaI

BglII

XbaI

BglII XbaI

XbaI

BglII

BglII

XbaI

BglII

XbaI

BglII

XbaI

BglII

BglII

XbaI

on November 10, 2019 by guest

http://jvi.asm.org/

INTEGRATION OF SV40 GENOME INTO CELLULAR DNA 189

leastsomepartial repetition.

notdistinguish from which

DNAwasderived.

DISCUSSII

Integration of SV40 D genome of transformed CHL and 3T3 rat transforn

TABLE 3. Presence ofsingi

repeats ofSV40 sequencesii genome oftransforrm

Type of Forn III Cellline trans- SV40

formant DNA" CHLA209L5 N

CHLA209L62 N _

CHLA209L4 Nor A

CHLA209LA9 A +

'Presence (+) or absence (-) released after digestion with the e table.

A I _{2 3 4} B I

VW

i _ ...p

FIG. 2. Hybridizationpattei celllines digested with SV40 (A) XbaIdigestion of 3T3rat tr

1,A209-FR3T3-A30; lane2,A2

A209-FR3T3-A43; lane 4, SV4 DNAs. (B) SacI digestion of cells. Lane1,A209-FR3T3-A37

A30; lane 3,A209-FR3T3-A43; larweight standards (see Fig.

However,wecould contained fromone tofiveintegration sites. For

of the sitesform III the CHL transformed lines, most contained SV40 DNA integrated atmultiple sites.

How-ON ever, among those lines generated at initial mul-tiplicities of infection of <0.48 PFU/cell, the

)NA into the

cell

majority had SV40 sequences integrated at a

cells.

Among the single site (7 of 11), whereasallofthose

gener-nants examined, all ated at multiplicitiesofinfectionof>0.7 PFU/

cellhadSV40 sequencesatmore than one site lecopies or tandem (12 of12). For the 3T3 rat transformants, which

ztegrated

inthecell weregenerated at initial multiplicities of

infec-ed CHLcells tionof from 5 to100PFU/cell

(i.e.,

at multiplic-ities of infection higherthanusedwith the ham-Restrictionenzymes ster cells), most contained SV40 integrated at

multiplesites (12of18),but asignificantnumber

TaqI,BamHI, EcoRI, hadSV40 integratedatasingle site. (Botchanet BglI al. [5] alsoobserved apredominanceofmultiple

TaqI,BamHI, EcoRI

integration

sites in

SV40

rat

transformants.)

TaqI,_BgtIBamHI, EcoRI, Thus,it appears that in rat

cells

single-site in-TaqI,BamHI,

EcoRI

tegrations (6 of 18) are more common at high

ofSV40

forn

III DNA multiplicities of infection than they are in

ham--nzymesindicated on the stercells(0 of

12).

This differencemayrelateto

the fact that hamster cells are

semipermissive

andratcells are

nonpermissive.

Among thecell 2 3 4 linesthat

appeared

to contain SV40integrated

,, at asingle site,somemaycontain onlya

single

copy of the SV40 genome

(CHLA209L4,

CHLA209L5,

CHLA209L62, A209-FR3T3-N94,

A209-FR3T3-A12), whereas others clearly

con-s-

_ >? tained

partial repetition

of the SV40 genome

ti7

(A209-FR3T3-N11,

A209-FR3T3-N53,

A209-WE

wJ *FR3T3-N30) and still others maycontain

mul-'AM

9-

500 tiple repetitions of the SV40 genome

*^0*

(CHLA209L49).

0

'f Some

explanation

is

required

for the fact that

4-_, transformed celllinesthatcontain

papovavirus

-l sequences at asingle site frequentlyhavepartial

v

4o

300repetition

of the

papovavirus

genome.Tandem

repeats of viral sequences have beenobserved before in

polyoma

virus-transformedratcells

(a

iS _~-

semipermissive

system) (2)

and in

SV40-trans-formed rat cells (a

nonpermissive system) (5).

Weobserved tandemrepeatsofSV40sequences *

4- 330

in oneof four CHL lines (a

semipermissive

sys-.

4-'°0°tem)

andinthreeof fiveratlines(a

nonpermis-sive system). At least one mechanism for the generation of tandemrepeatsisvia

rolling

circle

replication. Rolling

circle

replication

in

permis-sive cells has been

demonstrated

for

polyoma

virus (3) and SV40 (R. G. Martin and V. P.

rns

of

transformed

rat Setlow,Cell, in press) andsuggestedin

nonper-DNA

no-cut

enzymes-

missive cells for

SV40

(W.

Chia and

P.

Rigby,

ansformedcells. Lane personalcommunication).We

speculate

thatthe

0 form II

III7

and

I

generation

of

lnear

moleculesisprerequisite to

3T3 rat

transformed

integration

within

the host genome and that l-lane2, A209-FR3T3-

rolling

circle

replication

is the

preferred

manner

lane4, DNAmolecu- for

generating

linear molecules. Other

mecha-1). nisms forgenerating tandemrepeats have been

VOL. 35,1980

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.508.54.249.153.568.2]

T Cellline tra A209-FR3T3-Nllb

A209-FR3T3-N53b

A209-FR3T3-N55b

A209-FR3T3-N30b

A209-FR3T3-N94C

A209-FR3T3-N82C

A209-FR3T3-N87c

A209-FR3T3-N91c

A209-FR3T3-A12b

A209-FR3T3-A24b

A209-FR3T3-A22b

A209-FR3T3-A30b

A209-FR3T3-A37b

A209-FR3T3-A43b

A209-FR3T3-A16b

A209-FR3T3-A81C

A209-FR3T3-A84c

A209-FR3T3-A99C

Iype of No. of in-nsform- tegration ant sites

N 1

1

N 1

1

N 1

N 1

1 1

N 1

1 1

N 2

3

N 2

1

N 2

2 3

A 1

1 1

A 1

3

A 1

2 1

A 1

2 5

A 2

5

A 1

2 5

A 1

2

A 2

2

A 1

2

A 2

3

No. of base pairs per band containingSV40 sequences'

8,900 23,500

14,500 17,500

14,800

8,400 14,000 17,200

16,300 10,500 10,300

10,800,27,000 15,300,21,200,25,500

8,900,27,000 24,800

12,400,18,800 14,400,19,800

9,100,13,800,21,000

7,900 10,500 11,800

NDd

7,500,10,200,26,000

17,800 11,500,18,000 27,000

15,000 7,000, 7,700

8,000, 9,800, 13,100, 14,800, 18,200

7,000, 7,700

8,100, 9,800, 13,100, 15,000, 18,200

ND 7,000, 7,700

8,000, 9,600, 13,000, 15,000, 18,000

24,500 8,400,18,500

9,000,10,300 11,200,21,500

15,400 16,100,25,500

4,100,18,000 4,500, 17,200, 23,700

Restriction enzyme

XbaI SacI

SacI XbaI

XbaI

BglHl XbaI SacI

XbaI Sacl SstI

XbaI SacI

XbaI SacI

BglII SacI XbaI

XbaI SstI SacI

BglII XbaI

XbaI Sacl SstI

BglII XbaI SacI

XbaI Sacl

BglII XbaI Sacl

SacI XbaI

XbaI SacI

XbaI SacI

XbaI Sacl

aSee Table 1,footnotec.

bThese

cell lines

arethesame asthosedescribedinreference27,but withthe letterL,M,orHinstead ofN

orAbefore the lasttwodigits.

cThese celllineswereisolatedasdescribed in reference28.

dND,Notdetermined.

190

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.508.77.459.52.608.2]

INTEGRATION OF SV40 GENOME INTO CELLULAR DNA 191

A 1 2 3 4

Y :::::

s:::A;'

*.;s...

... ,,l *: .. pffl Z ..Ak

..W, S p.

w4 *4hy._{7r;fs >}

w

-*:b.

4

F

* _F *A,. W,2..

!I_

_ b

* .4,,.5

..:

i-.

**.:.o::.#

_*

_

5 6

r1

-:

0

B 1

FIG. 3. Hybridizationpatternsof

transft

cell linesdigestedwithSV40 DNAone-cut

(A) BglIandTaqI digestions of3T3ratcell

A209-FR3T3-N53TaqI digestion; lane2,.

lecularweight standards(see Fig. 1); lan4 FR3T3-A22TaqIdigestion;lane4,A209-Fi

BglI digestion; lane5, A209-FR3T3-A22 B

tion; lane 6,DNAmolecular weightstanc

Lane 1, SV40form IIIDNA;lane 2, A20

N30 TaqI digestion; lane3, A209-FR3T3-,

digestion.Arrows indicateposition of form

DNA.

proposed (2, 6).

Theresultsobtained with theviablE doublemutantsindicate thatsmall-te

notrequired for integration of the SV4i into thecellularDNA. Furthermore,sI

tigen does not seem to affect the pa integration; i.e., both single-site and siteintegrationswerefoundwhether c viruswas capable of expressing thesxi

tigen.

The conclusion thatSV40 DNA do(

tegrate atspecific sites in the cell gen

basedontheobservationof differentpi integration in different cell lines(5).Hz is important to point out that integ

specific sequences cannot be ruled possibletogetdifferent restrictionpal

spite integration in thesame geneif (i)

amounts ofhost DNA are deleted ur

integration, (ii) differentamountsofSI areintegrated in thesamegene, or(iii

tionoccursatdifferent portions(restric

ments) of thesamegene. Furthermor

2 3 DNAcontainshighly repeatedsequenceswhich exist repeated in tandem or interspersed with unrelatedsequences(29). TheSV40genome can recombine with thesehighly repeatedsequences (24, 25).Differentpatternsof integration of the

viral DNA in different cell lines may be

ex-plained by integration in a specific highly

re-peatedsequencethatisinterspersed with unre-latedsequences.

Alternatively,

ifthe viral DNA

integrated in a sequence repeated in tandem,

j* identical bands would be observedin different

* W cell

lines,

even when

integration

had

actually

occurred at different sites. Three rat

trans-formed

cell

lines (at least two of which were

obtainedindependently) presented thesame

in-*

tegration

pattern withtwodifferent SV40no-cut

enzymes. It remains to be established whether

multiple

bands in a

single

cell line arise

by

independent

events.

Integration

of the SV40 genomeinto the

cellular

DNA ofN- and A-transformants. There was some suggestion that

N-transform-ants were more

likely

than

A-transformnants

to

contain SV40 at a single insertion site. Taken

ornmed

rat together,9of20

N-transformants

and3of20

A-IenLanmes.

transformants contained SV40 DNAintegrated DNA mo- at a

single

site. For the CHL cells 4 of 12

N-e3,

A209.

transformants

and 2 of 10

A-transformants

con-R3T3-N53 tained

single-site insertions;

for theratcells 5 of gilI diges- 8N-transformants and 1of10A-transformants lards. (B) contained single-site insertions.

)9-FR3T3- Among the lines with only one integration site

A30

TaqI

we found

A-transformants

with fewer than 1.4

III

SV40

copies (A209-FR3T3-A12) andwithmore than

1.5copies

(CHLA209L49)

and

N-transformants

with fewer than 1.5 copies (CHLA209L5, CHLA209L62, A209-FR3T3-N94) and with

adeletion more than 1.7

copies

but fewer than 2

copies

antigen is

(A209-FR3T3-N30).

The maximum and

mini-0genome mum numbersarebased on the

assumption

of mall-tan- tandem repetition of the integrated sequences

Ltterns

of andcomefrom the

ability

orfailure of the

SV40

multiple- DNAone-cutenzymetorelease

full-length SV40

)r notthe DNA. No correlationbetweencopynumber and nall-tan-

phenotype

isapparent, since of those testedtwo CHL N-transformants have fewer thantwo cop-es notin- iesandoneA-transformantmayhavemorethan

iomewas two

copies,

whereas for the rat cells three

N-atterns

of

transformants

have fewer thantwo

copies

and

owever, it one may have a

single

copy, but the one

A-ration at transformantalso may havea

single

copy. out. It is Several conclusionscanbe drawn

concerning

tternsde- the

relationship

between SV40 genome copy different number and the temperature

sensitivity

ofthe

2on SV40

resulting

transformants.

(i) N-transformants

do

V40 DNA not arise from the failure of

SV40

to become

)integra-

integrated.

(ii)

A-transformantsdonotarise

by

tionfrag- gene

amplification

or

by

a

simple

gene

dosage

e,cellular effect. And

(iii)

neither A-norN-transformants

VOL. 35,1980

4" ft"

on November 10, 2019 by guest

http://jvi.asm.org/

[image:8.508.47.245.74.309.2]

TABLE 5. Presenceof singlecopiesortandem repeatsofSV40 sequences integrated in the cell genome of transformed rat cells

Minimum

CeRline Typeof Form III Retito

nyeno.

of

inte-Cell line transformant SV40DNA0 Restriction enzyme gration sites

A209-FR3T3-N11 N + BglI 1

- BamHI

A209-FR3T3-N53 N - TaqI 1

+

BglI,

HpaII

A209-FR3T3-N30 N + TaqI, BamHI,HpaII 1

-

EcoRI

A209-FR3T3-N94 N - BglI, BamHI, TaqI 1

A209-FR3T3-N82 N + Bgll, BamHI 3

A209-FR3T3-N87 N + BglI, TaqI, HpaII, BamHI 2

A209-FR3T3-N91 N + BglI 3

- BamHI

A209-FR3T3-A12 A - BglI, TaqI, HpaII, BamHI 1

A209-FR3T3-A22 A + TaqI, BglI 2

A209-FR3T3-A30 A - TaqI, BamHI 5

+

EcoRI,

HpaII

A209-FR3T3-A16 A - Bgll, BamHI, TaqI 2

A209-FR3T3-A81 A - Bgll,BamHI 2

A209-FR3T3-A84 A + Bgll, BamHI 2

- HpaII

A209-FR3T3-A99 A + Bgll, BamHI 3

aSeeTable3andFig.3.

haveunique integration patterns.

One

possible

mechanism to explain the

N-phenotypewasthatSV40

replicated

as aplasmid

in N-transformants. This possibility is

elimi-nated by the finding that all N-transformants containedintegrated SV40sequences.

Asecondpossible explanationfor the

temper-ature-insensitivephenotype ofthe

A-transform-antsis thatT-antigenisrequired for the main-tenance of transformation in all transformed cells but that high levelsof thethermolabile

T-antigen resultin"leaky"expression of the

trans-formed phenotypeattherestrictivetemperature

(7,11,33).Itisstillpossible thatin A-transform-ants T-antigen accumulates more than in N-transformants. However, we haveeliminatedthe

possibilitythatsuchleakinessinvariablyresults

from the integration of multiple copies of the

SV40genome. It cannot, forexample,beargued

that all

A-transformants

contain multiple

re-peatsoftheearly regionofSV40 andthateven

those A-transformants with SV40 sequences at

asingle site containmultiple repeats of theSV40

genome. Cell line A209-FR3T3-A12 contains

fewer thantwo copies of theSV40early region,

and cell lines A209-FR3T3-A16 and

A209-FR3T3-A81 contain a maximum of two

func-tional early regions. Except for CHLA209L49,

wehave not found anA-transformant that

re-leasedunit-lengthformIII SV40DNA withall

of the SV40 DNA one-cut restriction enzymes

assayed. Thisallows us to rule out the possibility of more than five or six copies of the SV40

genome tandemly integrated (gone undetected

by SV40DNA no-cutrestrictionenzymes due to thelimitations of the Southern technique). We conclude that gene dosage alone cannot explain theA-transformants.

Another

possible explanation

for the A-trans-formants isthattheyall result from mutationof

agrowth controlgeneby

integration

of the viral genome into it.T-antigenwould not berequired

for the maintenance of transformationin these

cells, in contrast to N-transfornants, inwhich

on November 10, 2019 by guest

http://jvi.asm.org/

[image:9.508.66.461.99.444.2]

INTEGRATION OF SV40 GENOME INTO CELLULAR DNA 193

therole ofintegrationof theSV40genomeis to

permanently provide large T-antigen (27, 28).

Certainly with thisapproach we could discern nouniquepatternofintegrationin

A-transform-ants which distinguished them from N-trans-formants. Furthermore, there was no readily

discemable common element to either the A-typeorN-typeintegrationpattern.We have not eliminatedthepossibilitythatA-transformants

result from integration of the viral genome at

any of a small number of control genes. Rat transformants withasingleintegrationsitewere

frequent among N-transformants but rare amongA-transformants. Ifgrowthcontrolgenes

representaminority of the cellulargenome, (i)

A-transformants would arisepreferentiallyfrom

cells in which the viral genome was able to

integrate atseveralsites, and (ii)

N-transform-antswould arise more frequentlyfrom cellsin

which the SV40genomewasabletointegrateat onlyonesite.

LITERATURE CITED

1. Basilico, C., S. Gattoni, D. Zouzias, and G. Della Valle.1979.Loss ofintegrated viral DNAsequencesin polyoma-transformedcellsisassociatedwithanactive viralAfunction. Cell 17:645-659.

2. Birg, F., R. Dulbecco, MLFried,andR. Kamen.1979. State and organization ofpolyoma virus DNA se-quencesintransformedratcelllines.J.Virol.

29:633-648.

3. Bjursell, G.1978. Effectsof2'-deoxy-2'-azidocytidineon

polyomavirus DNAreplication: evidence for rolling circle-type mechanism. J. Virol.26:136-142.

4. Botchan, M., G. McKenna, and P. A. Sharp. 1973. Cleavage ofmouseDNAbyarestrictionenzyme as a

cluetothearrangementofgenes.ColdSpring Harbor Symp. Quant.Biol. 38:383-395.

5. Botchan,M.,W.Topp,and J. Sambrook. 1976.The arrangementof simianvirus40sequencesin the DNA

oftransformedcells.Cell9:269-287.

6. Botchan, M., W. Topp, and J. Sambrook. 1978. Studies

onsimianvirus 40excision fromcellular chromosomes. ColdSpring Harbor Symp. Quant. Biol. 43:709-719. 7. Brockman,W. W. 1978. Transformation ofBALB/c-3T3

cellsby tsAmutantsofsimian virus40- temperature sensitivity of the transformed phenotype and

retran-formationby wild-type virus. J.Virol. 25:860-870. 8. Brugge, J. S., and J. S.Butel.1975.Role of simianvirus

40geneAfunction inmaintenance of transformation.

J.Virol.15:619-635.

9. DiMayorca,G., J. Callender, G.Marin, and R. Gior-dano. 1969.Temperature-sensitivemutantsofpolyoma virus.Virology 38:126-133.

10. Eckhart, W.1969.Complementation and transformation by temperature-sensitive mutants of polyoma virus. Virology38:120-125.

11. Edwards, C. A. F., G. Khoury, and R. G. Martin.1979. Phosphorylation of T-antigen and control of T-antigen expression in cellstransformed bywild-type andtsA mutantsof simian virus40.J. Virol.29:753-762. 12.Fluck, M.M., and T. L Benjamin.1979.Comparisons

oftwoearlygenefunctions essential fortransformation inpolyomavirusandSV40. Virology 96:205-228. 13.Fried, M.1965.Celltransformation ability ofa

tempera-ture-sensitive mutant of polyoma virus. Proc. Natl. Acad.Sci. U.S.A.53:486-491.

14. Gaudray, P.,M.Rassoulzadegan, and F. Cuzin.1978.

Expressionof simian virus 40earlygenes intransformed ratcells is correlated with maintenance of the trans-formedphenotype. Proc. Natl. Acad. Sci. U.S.A. 75: 4987-4991.

15.Jeffreys,A.J., and R. A. Flavell.1977.Aphysicalmap of the DNAregionsflankingthe rabbit,B-globingene. Cell 12:429-439.

16.Ketner,G., and T. J.Kelly,Jr.1976.Integratedsimian virus 40 sequences intransformed cell DNA:analysis usingrestriction endonucleases. Proc. Natl. Acad. Sci. U.S.A. 73:1102-1106.

17.Kimura,G.,and A.Itagaki.1975.Initiation and main-tenance of celltransformationbysimian virus 40:.aviral genetic property. Proc. Natl.Acad.Sci. U.S.A. 72:673-677.

18.Maniatis,T., S. G. Kee, A. Efstratiadis, and F. C. Kafatos.1976.Amplificationand characterization ofa

,8-globingenesynthesizedinvitro. Cell8:163-182.

19.Martin,R.G., andJ. Y.Chou.1975.Simian virus40 functionsrequired for the establishment and mainte-nanceofmalignanttrnsformation.J. Virol. 15:599-612.

20.Martin, R.G.,V. P.Setlow, and C. A. F. Edwards. 1979.Rolesof simianvirus 40 tumor antigens in trans-formationof Chinesehamsterlungcells: studies with simianvinrs40double mutants. J.Virol.31:596-07. 21. Osborn, M.,and K.Weber. 1975. Simian virus 40 gene

Afunction and maintenance oftrnsformation.J.Virol. 15:636-644.

22. Rassoulzadegan,M., B.Perbal, and F.Cuzin.1978. Growthcontrol in simian virus 40-transformedratcells: temperature-independentexpression of the transformed phenotype intsAtransformants derived by agar selec-tion. J. Virol. 28:1-5.

23. Rassoulzadegan,M., R.Seif,and F.Cuzin.1978. Con-ditionsleadingtotheestablishment of the N (a gene-dependent) and A (agene-independent) transformed statesafterpolyoma virus infection of rat fibroblasts. J. Virol. 28:421-426.

24. Rosenberg, M., S.Segal,E.ILKuff, and M. F. Singer. 1977. The nucleotide sequence of repetitive monkey DNAfound in defectivesimian virus 40. Cell 11:845-857.

25. Rosenberg, H.,M. Singer, and M. Rosenberg. 1978. Highlyreiteratedsequencesof .sim

-iansimian.Science200:394-402.

26. Seif, R., and F. Cuzin. 1977. Temperature-sensitive

growthregulation in one type oftransformedratcells inducedby the tsa mutant of polyomavius.J.Virol. 24:721-728.

27. Seif,R.,and R.G.Martin. 1979. Growth state of thecell

early after infection with simian virus 40 determines whether the maintenance oftransformation willbe A-genedependentorindependent. J. Virol. 31:350-359.

28. Seif,R., and R. G.Martin. 1979.Simian virus 40small

t-antigen is notrequired for themaintenanceof trans-formation but may act as apromoter (cocarcinogen) duringestablishment of transformation inresting rat cells.J. Virol. 32:979-988.

29. Singer,D. S. 1979.Arrangement of a highly repeated DNAsequence in the genome and chromatin of the African greenmonkey. J. Biol. Chem. 264:5506-5514. 30. Southern, E. M. 1975. Detection of specific sequences

among DNAfragmentsseparated by gel electrophore-sis.J.Mol. Biol. 98:503-517.

31. Sugden,B., B. Detroy, R. J. Roberta,and J. Sam-brook. 1975. Agaroseslab-gel electrophoresis equip-ment.Anal.Biochem.68:36-46.

32. Tegtmeyer, P. 1975. Functions of simian virus 40 gene A intransforming infection. J. Virol.15:613-618. 33. Tenen, D. G., R. G.Martin,J.Anderson, and D. M.

Livingston. 1977.Biologicaland biochemical studies of cells transformed by simian virus40 temperature-sensitivegeneAmutants andAmutant revertants.J. Virol.22:210-218.

VOL. 35,1980

on November 10, 2019 by guest

http://jvi.asm.org/