Stateofthe Rat Cells Transformed Polyoma Virusjvi.asm.org/content/18/2/436.full.pdfwas detected. The yield of similarly infected 3T3mousecultures was1.6 x 106 and107 PFU/ culture at

JOURNAL OF VIROLOGY, May 1976, p. 436-444Copyright © 1976 American Society for Microbiology

Vol. 18, No. 2Printed in U.S.A.

State of the Viral DNA in Rat Cells Transformed byPolyoma Virus

I. Virus Rescue and the Presence of Nonintegrated ViralDNA Molecules

ISHWARI PRASAD, DIMITRIS ZOUZIAS, AND CLAUDIO BASILICO*Department ofPathology, New York University School of Medicine, New York, New York 10016

Received for publication 24 December 1975

The interaction of polyoma virus with a continuous line of rat cells wasstudied. Infection of these cells with polyoma did not cause virus multiplicationbut induced transformation. Transformed cells did not produce infectious virus,but in all clones tested virus was rescuable upon fusion with permissive mousecells. Transformed rat cells contained, in addition to integrated viral genomes,20 to 50 copies of nonintegrated viral DNA equivalents per cell (average). "Free"viral DNA molecules were also found in cells transformed by the ts-a and ts-8polyoma mutants and kept at 33 C. This was not due to a virus carrier state,since the number of nonintegrated viral DNA molecules was found to be un-changed when cells were grown in the presence of antipolyoma serum. Reclon-ing of the transformed cell lines produced subclones, which also contained freeviral DNA. Most of these molecules were supercoiled and were found in thenuclei of the transformed cells. The nonintegrated viral DNA is infectious. Itsspecific infectivity is, however, about 100-fold lower than that of polyoma DNAextracted from productively infected cells, suggesting that these moleculescontain a large proportion of defectives.

Infection with the oncogenic DNA virusespolyoma and simian virus 40 causes neoplastictransformation in cells that are nonpermissiveto viral multiplication. Transformed cells donot generally produce infectious virus but canbe shown to contain the viral DNA integratedinto the host genome (20). Fusion of nonpermis-sive simian virus 40-transformed cells with per-missive cells results in virus rescue (13, 22). Inthe case of polyoma. However, the most widelyused host for transformation, hamster cells,usually fails to yield infectious virus, even afterfusion with permissive mouse cells (2, 7, 22).This is not due to the fact that the heterokar-yons are nonpermissive for viral multiplication(2). It is possible that the polyoma DNA fails tobecome excised or that a small degree of per-missiveness in hamster cells causes selectionfor transformation by defective polyoma ge-nomes. Hamster cells transformed by the ts-amutant ofpolyoma virus at 33 C and kept at thenonpermissive temperature (39 C) can, in fact,produce virus upon fusion when shifted to per-missive conditions (7).

Previous reports (5, 6) have shown that aclonal line of rat myoblasts transformed bylarge-plaque polyoma often produces a smallamount of infectious virus. A large increase in

virus yield is obtained when these cells aretreated with several chemical and physicalagents (6). Moreover, "virus-free" subclones ofthis line yield virus upon fusion with mousecells (5). More recently, Kimura et al. (12) haveshown that,l in another line of rat cells trans-formed by polyoma virus, the virus can gener-ally be rescued by fusion with mouse cells.We have investigated in detail the interac-

tion of polyoma virus with a continuous line ofFischer rat cells, F2408 (9). In this paper wereport that virus can be rescued by fusion fromall polyoma-transformed F2408 rat cells tested.These transformed cells do not produce infec-tious virus spontaneously but contain, in addi-tion to integrated viral genomes, "free" viralDNA molecules.

MATERIALS AND METHODSCell lines. Swiss 3T3 mouse cells (clone D) and rat

cells of the F2408 established line (9) were used.F2408 rat cells were kindly provided by G. Di-Mayorca. Primary rat cell cultures were preparedfrom Fischer's rat embryos. Cells were grown inDulbecco-modified Eagle medium containing 10%calf serum.

Virus. Wild-type, small-plaque polyoma and thetemperature-sensitive, large-plaque polyoma mu-tants ts-a (10) and ts-8 (4) were used. Wild-type

436

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from

POLYOMA VIRUS-TRANSFORMED RAT CELLS 437

virus was grown at 37 C, and the ts mutants weregrown at 33 C. Viruses were extracted, purified bycesium chloride density gradient centrifugation,and titered by plaque assay on monolayers of 3T3cells.

Transformation. Transformation of rat cells bypolyoma virus was determined by the ability oftransformed cells to grow in suspension in soft agar,following the technique of MacPherson and Montag-nier (14). Cells were infected with the virus at differ-ent multiplicites in TD buffer (0.8% NaCl, 0.038%KCI, 0.01% Na2HPO4, and 0.3% Tris-hydrochloride,pH 7.2). Adsorption was carried out in suspensionfor 1 h at room temperature. The final agar concen-tration in the medium was 0.34%. Wild-type poly-oma-infected cells were incubated at 37 C and ts-a-and ts-8-infected cells were incubated at 33 C. After2 to 4 weeks of incubation, transformed colonieswere isolated. Uninfected cells never grew in agarmedium, even when plated at a high concentration.

Virus rescue. Transformed cells (106) were mixedwith mouse 3T3 cells (106) and exposed to 8-propi-olactone-inactivated Sendai virus as previously de-scribed (17). The fused cells were cultured for 24 h.The medium was changed, and the cells were incu-bated at 37 C for 4 days or at 32 C for 5 days. For co-cultivation, the same procedure was used, but thecells were not exposed to the Sendai virus.

Infectivity of viral DNA. Confluent monolayers of3T3 cells in 60-mm plates were washed with TDbuffer. A 200-Ag amount of DEAE-dextran (16) in0-.2 ml of TD buffer was spread over the cells. After10 min, 0.1 ml of a solution containing DNA in lxSSC (0.15 M NaCl and 0.015 M sodium citrate) wasadded to the cells. Twenty minutes later, the cellswere gently washed with TD buffer and then over-laid with 7 ml of medium containing 0.9% agar.Cells were stained with neutral red at day 7 to 10and incubated until the plaques were clearly count-able.

V-antigen. The presence of polyoma V-antigenwas determined by immunofluorescence, as previ-ously described (1).

Chromosomes. The method for determining chro-mosomes has been described previously (1).

Autoradiography. Cells were grown on glasscover slips in petri dishes. After labeling with[3H]thymidine, they were fixed with ethanol-aceticacid (9:1). After washing with 70% ethanol anddrying, the cover slips were mounted on slides withPermount and dipped in nuclear track emulsion(NTB-2, Kodak). After the appropriate time of expo-sure the slides were developed and stained withGiemsa, prior to counting.

Preparation of polyoma viral DNA. 3T3D mono-layers were infected with polyoma at 50 PFU/cell.When the cells were partially lysed, low-molecular-weight DNA was extracted following the Hirt proce-dure (11). The Hirt supernatant was extracted withsaturated phenol and then with chloroform-isoamylalcohol. Nucleic acid in the aqueous phase was pre-cipitated by the addition of 2 volumes of ethanol.After 24 h at -20 C, the precipitate was centrifugedand suspended in 1 x SSC. Form I DNA (covalentlyclosed circular duplex DNA) was isolated after CsCl-

ethidium bromide gradient centrifugation. For thepreparation of [32P]polyoma DNA, infected 3T3 cellswere incubated in phosphate-free medium contain-ing [32P]orthophosphate (100 ,Ci/ml) for about 50 h.Viral DNA was extracted-and purified by equilib-rium density centrifugation in cesium chloride-ethidium bromide, followed by velocity sedimenta-tion through neutral sucrose gradients. [32P]poly-oma DNA with a specific activity of 2 x 106 to 3 x 106counts/min per ,ug was obtained.

Determination of the number of "free" viral DNAequivalents. Free viral DNA equivalents were esti-mated by measuring the effect of low-molecular-weight DNA preparations from transformed ratcells on the rate of reassociation of 32P-labeled poly-oma DNA. Cells were lysed in 0.6% sodium dodecylsulfate-10-2 M EDTA, and low-molecular-weightDNA was extracted according to Hirt (11). The Hirtsupernatant was extracted twice with phenol andonce with chloroform-isoamyl alcohol (24:1). DNAwas precipitated with ethanol at - 20 C. The precipi-tate was dried and dissolved in a small volume of0.01 M phosphate-0.001 M EDTA (pH 6.8) and dia-lyzed extensively against the same buffer. Beforehybridization, DNA was fragmented by boilingtogether with the [32P]polyoma DNA probe for 10min in 0.3 M NaOH. DNA-DNA reassociation ki-netics and hydroxyapatite chromatography weredone according to Sharp et al. (18). Nonintegratedpolyoma DNA equivalents per cell were calculatedfrom the formula: number of polyoma DNA equiva-lents/cell = a X 2 x 1011/A, where a = micro-grams of viral DNA in the preparations, A = thenumber of cells, and 2 x 1011 = the number ofpolyoma DNA molecules per microgram.

Determination of the number of viral DNA equiv-alents associated with the cellular DNA. High-mo-lecular-weight DNA was first resolved from low-molecular-weight DNA by Hirt extraction. The pel-let was washed twice with 1 x SSC at 4 C and DNAwas extracted according to Marmur (15), with minormodifications. The DNA was further purified byneutral sucrose gradient centrifugation (10 to 30%sucrose in 1 x SSC). Fractions containing DNA sedi-menting faster than 50S were pooled, dialyzedagainst 1 x SSC, and precipitated with ethanol. ForDNA-DNA reassociation experiments the DNA wasincubated in 0.3 M NaOH for 7 to 10 h at roomtemperature, neutralized with HCl, adjusted to 0.1M NaCl, and precipitated with ethanol. The precipi-tate was dissolved in 0.01 M phosphate-0.001 MEDTA, pH 6.8, and dialyzed extensively against thesame buffer. Before hybridization, cellular DNAwas fragmented by sonication and subsequentlyboiled together with the [32P]polyoma DNA probefor 10 min in 0.3 M NaOH. Samples were removedfrom the mixtures at intervals, and the fraction of32P-labeled single-stranded DNA (f.) was deter-mined by chromatography on hydroxyapatite (18).

RESULTSCharacteristics of the cells. Fischer rat fi-

broblasts derived from the established lineF2408 were propagated in medium containing

VOL. 18, 1976

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from

438 PRASAD, ZOUZIAS, AND BASILICO

10% calf serum. The mean doubling times ofthese cells were 16, 18, and 24 h at 40, 37, and33 C, respectively. At saturation, these cellsreached densities of about 1.5 x 106 cells/60-mmplate. The colony-forming ability of these cellsat 37 and 33 C was -45%, whereas at 40 C itwas 25 to 30%. Karyological studies showedthat these cells contained 42 chromosomes, in-cluding 12 telocentric, 14 metacentric, and 16acrocentric. It appears that F2408 rat cells con-tain a diploid number of chromosomes.Response of F2408 rat cells to polyoma in-

fection. Rat cells are generally nonpermissiveto polyoma multiplication but can be trans-formed by the virus (20). To determine the levelof permissiveness of F2408 cells, semiconfluentcultures were infected with polyoma virus at amultiplicity of infection (MOI) of 50 PFU/cell.After adsorption infected cultures were kept inmedium containing antipolyoma serum for 12 hto inactivate unadsorbed viral particles. After 3days of incubation, no cytopathic effect wasobserved in the cultures. When virus was ex-tracted and titered, there was no increase invirus titer above that determined at 0 time.Similar experiments were also done using in-fection with purified polyoma DNA. DNA infec-tion (104 PFU/culture) was carried out as de-scribed in Materials and Methods. After 3 and 5days of incubation virus was extracted and ti-tered. In one experiment, DNA-infected ratcells failed to yield any infectious virus. In asecond experiment, a yield of 102 PFU/culturewas detected. The yield of similarly infected3T3 mouse cultures was 1.6 x 106 and 107 PFU/culture at 3 and 5 days, respectively.When infected rat cells were tested for the

presence of polyoma V-antigen by immunofluo-rescence, we did not find any positive cells outof - 105 examined. If the MOI was raised to-500 PFU/cell, a few (-0.1%) positive cellswere observed. These results show that F2408cells are nonpermissive for polyoma multiplica-tion, although the use of very high MOI canlead to virus production in a small proportion ofthe infected cells.

Since polyoma induces cellular DNA synthe-sis in resting cells, it was of interest to deter-mine whether it caused a similar effect on ratF2408 cells. Confluent cultures of rat cells wereinfected and the cultures were labeled with[3H]dT at 12-h intervals. The frequency of theDNA-synthesizing cells was determined by au-toradiography. In cultures infected at 200 PFU/cell, the frequency of DNA-synthesizing cellsincreased considerably with time, whereas itdecreased gradually in uninfected cultures (Ta-ble 1).Polyoma transformation. The technique

used to transform rat cells with polyoma virushas been described in Materials and Methods.Figure 1 shows that the transformation fre-quency obeys one-hit kinetics. At an MOI of1,000 PFU/cell, 1.7% of the infected cells weretransformed. In comparison to wild-type virus,the ts-a and ts-8 mutants induced transforma-tion at 33 C at a slightly reduced frequency. Atthe nonpermissive temperature (39 C) the poly-oma ts-a mutant failed to transform rat cells(10).Rat cells were transformed at low and high

MOI using wild-type and mutant viruses.Transformed colonies were isolated. Cells frompractically all of these colonies grew in a criss-

TABLE 1. Induction of cellular DNA synthesis in ratF2408 cells after infection with polyoma virusa

% Cells synthesizing DNAMOI

0-12 hb 12-24 h 24-36 h 36-48 h

0 4.5 5 3.3 4.350 5.4 4.2 5.7 8.4

200 4.4 5.3 9.8 27.4a Cells were grown on cover slips. When they

reached confluence, fresh medium containing 10%calf serum was added. After 2 days, the cells wereinfected with polyoma virus and then received theold medium diluted 1:1 with fresh serum-free me-dium. Cells were labeled with [3H]thymidine (2 IACi,0.6 ug/ml) for the times indicated. They were thenfixed and processed for autoradiography.

b Hours after infection.

0-0

z 1.0LuI

0Lu

IL 0.1z0

0.010U)z< 0.001H~ 1 10 100 1,000

M.O.I. (PFU/CELL)FIG. 1. Relationship between the multiplicity of

infection by polyoma virus (wild type) and the trans-formation frequency of rat F2408 cells. The experi-mental procedure has been described in Materialsand Methods.

J. VIROL.

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from

POLYOMA VIRUS-TRANSFORMED RAT CELLS 439

cross manner to high saturation density (about10 to 15 times higher than the normal cells). Inaddition, transformed cells were capable of sus-tained multiplication in medium containing alow serum concentration (0.5%), whereas nor-mal cells ceased to grow.Spontaneous yield of virus from the poly-

oma transformants. Cells of all the trans-formed lines tested did not usually produce in-fectious virus spontaneously (Table 2). Trans-formed clones polyoma Rat-11, polyoma Rat-12,and polyoma Rat-13 were serially passaged fourtimes. Virus was extracted each time from 4 x106 cells in 10 ml of culture fluid and titered. Atpassage 4 some virus (10 PFU/ml) was detectedin the extracts from polyoma Rat-11 cells. Cellsof the ts-a Rat-21, ts-a Rat-22, ts-a Rat-23, andts-a Rat-24 lines did not produce infectious vi-rus. However, when ts-a Rat-23 cells weremaintained in culture for about 8 weeks, somevirus was produced (15 PFU/ml). It thus ap-pears that virus production in these cells is avery exceptional event.On one occasion we detected in about 105 cells

of the ts-a Rat-23 line, by immunofluorescence,

TABLE 2. Viral production in rat cells transformedby polyoma (Py) virus

PFU/cultureClone Spontane- Co-cultiva-

Oue" tioeb FusioncPyRat-11 Od 5x103 2x105Py Rat-12 0 5 x 102 5 x 104Py Rat-13 0 6.5x 103 1 x 105Py Rat-14 0 2.5 x 103 1.5 x 105PyRat-15 0 1x104 2.8x106Py Rat-16 NDe ND 4.6 x 106Py Rat-17 ND ND 8.5 x 104Py Rat-18 ND ND 1.5 x 104Py Rat-19 ND ND 1 x 104ts-a Rat-21 0 ND 3.5 x 103ts-a Rat-22 0 ND 5 x 10ts-a Rat-23 0 ND 7.5 x 103ts-a Rat-24 0 ND 2.5 x 104a Cells from the Py rat and ts-a rat lines were

grown at 37 and 33 C, respectively. Virus was ex-tracted from confluent cultures (4 x 106 cells) andtitered at 37 C for wild-type Py and at 33 C for ts-a.

b Transformed cells and 3T3D cells were mixedand then cultured for 4 days at 37 C. Virus wasextracted and titered.

c The same procedure as in b except that the cellswere fused by means of inactivated Sendai virus.Fused cells were incubated at 37 C for 4 days for thewild-type Py-transformed lines and at 32 C for 5days for the ts-a-transformed lines. ts-a virus wastitered at 33 C.

d 0, <10 PFU/culture.e ND, Not done.

two cells positive for V-antigen. Cells of otherlines tested, such as polyoma Rat-12, polyomaRat-13, and ts-a Rat-11, were always negative,and on a different occasion no positive cellswere found in the ts-a Rat-23 line.

Virus rescue. It was of interest to determinewhether infectious virus could be rescued fromthe polyoma-transformed rat cells. Rescue wasattempted using two methods: (i) cells werefused with permissive 3T3D cells by means ofinactivated Sendai virus; (ii) cells were co-culti-vated with permissive cells. After 3 to 5 daysthe cells were extracted and the extracts wereexamined for infectious virus. Table 2 showsthe results. In all the clones tested the viruswas rescuable. The yield was 20- to 100-foldhigher when the transformed cells were fusedby Sendai virus as compared to co-cultivationalone (Table 2). It appears that the formation ofheterokaryons with permissive cells is neces-sary to allow virus rescue.

Free viral DNA molecules in polyoma-transformed rat cells. We determined whethernonintegrated viral DNA molecules were byany chance present in the transformed F2408rat cells. Low-molecular-weight DNA was ex-tracted by the method of Hirt (11) and purified.The number of viral DNA equivalents was thendetermined by DNA-DNA reassociation kinet-ics. Table 3 shows the results. Cells of all theclones tested contained free viral DNA mole-cules.Polyoma Rat-11, polyoma Rat-12, and poly-

oma Rat-13 were independently isolated trans-formants from cells infected at 1,000 PFU withwild-type virus. Rat cells transformed by the ts-a mutant at 1 PFU (ts-a Rat-13) and at 200 PFU(ts-a Rat-21) also contained free viral DNA.When polyoma Rat-12 and ts-a Rat-13 cellswere kept in culture for over 2 months, thenumber of free viral DNA molecules did notchange appreciably. Thus, it appears thattransformed rat cells contained nonintegratedpolyoma DNA, regardless ofthe type ofvirus orthe multiplicity used for transformation.

Origin of the free viral DNA in rat trans-formed cells. Since all polyoma-transformedrat cells contained free viral DNA molecules, itwas important to verify whether or not the cellscarried over some infectious virus from the ini-tial infection or whether a few cells in thepopulation made a small amount of virus whichconstantly caused reinfection, producing freeviral DNA molecules. To rule out the possibil-ity of a virus "carrier state," polyoma Rat-12cells were grown in the presence or absence ofantipolyoma serum for 8 days. (The concentra-tion of antiserum that we used was found to beable to reduce virus titer by 4 logs in 60 min at

VOL. 18, 1976

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from

440 PRASAD, ZOUZIAS, AND BASILICO

TABLE 3. Free viral DNA molecules in rat cellstransformed by polyoma (Py) virus

Transformed line" Viral DNA equivalents/

Py Rat-il 11Py Rat-lib' 8Py Rat-11cc 22Py Rat-11fc 21Py Rat-12 22Py Rat-13 11ts-a Rat-13 52ts-a Rat-21 20ts-8 Rat-il 14Py Rat PJd 18.5

a Wild-type Py-transformed lines were grown at37 C, ts-a- and ts-8-transformed lines at 33 C.

b Determined as described in Materials and Meth-ods.

' Subclones of Py Rat-li.d Derived from primary rat embryo cultures.

37 C). Low-molecular-weight DNA was iso-lated. The number of viral DNA equivalentswas then determined and found to be 33 and 39copies/cell (Fig. 2). It therefore appears thatreinfection did not play a significant role indetermining the presence of nonintegratedviral DNA in polyoma-transformed rat cells.

Since we used a large number of cells in all ofthese experiments, the possibility existed thatproduction of free viral DNA was a function ofonly a specific minority of the cell population.To answer this question, cells of polyoma Rat-11 were recloned in the presence of antipolyomaserum and three clones were tested. Cells ofeach clone contained free viral DNA (Table 3).In summary, the presence of nonintegratedviral DNA molecules in these cells is not due toa virus carrier state, and every cell in the popu-lation appears capable of producing or carryingfree viral DNA.

Localization of the free viral DNA. Nuclearand cytoplasmic fractions were prepared frompolyoma Rat-12 cells as previously described(19). The number of viral DNA molecules pres-ent in the Hirt supernatant of these fractionswas determined. We found that -13 copies ofviral DNA equivalents were present in the nu-clear fraction and -4 copies were present in thecytoplasmic fraction. It is likely that duringfractionation some nuclei might have been bro-ken, thus contaminating the cytoplasmic frac-tion. We therefore interpret these results asindicating that the bulk of free viral DNA inthese cells is nuclear.

Physical state of the free viral DNA, Low-molecular-weight DNA extracted from poly-oma-transformed rat cells was subjected toethidium bromide-cesium chloride density gra-

J. VIROL.

dient centrifugation in the presence of simianvirus 40 DNA, which was added as a marker.Polyoma viral DNA sequences were then de-tected by DNA-DNA reassociation in the frac-tions containing supercoiled DNA and in thefractions containing linear and open circularDNA. Ninety percent of the polyoma DNA se-quences were found in the supercoiled fraction(Fig. 3). Thus it appears that most of the freepolyoma DNA in transformed rat cells is pres-ent as supercoiled circular molecules. This ex-periment also shows that the appearance ofviral DNA in the Hirt supernatant was mostlikely not due to the fact that integrated viralDNA molecules were randomly excised duringextraction.

Infectivity of the free viral DNA. The spe-cific infectivity of the free viral DNA moleculeswas determined by plaque assay on 3T3 cells.The viral DNA isolated from the polyoma Rat-12 and ts-a Rat-13 lines was infectious butformed plaques at a much lower efficiency thanpolyoma DNA extracted from infected 3T3 cells(Table 4).

7.0

6.(1-~5.fss

4.0

3.0

2.0

1.0 'O, _ _ _ _ _ - - -.4

2 4 6 8 10 12 14 16

COP t X103 (ug -hrs.-ml-1)FIG. 2. Reassociation kinetics of [3'2P]polyoma

DNA in the presence of low-molecular-weight DNAfrom polyoma Rat-li cells grown in the presence orabsence of antipolyoma serum for 8 days. Medium,containing fresh antiserum, was changed every 3days. Hybridization was carried out as described inMaterials and Methods. The data are plotted as 1/faversus C0p x t, where f8, represents the fraction of 32p_labeled single-stranded DNA, C0,, the input ofprobein micrograms per milliliter, and t the time of sam-pling in hours. Symbols: (-) renaturation of132'P]polyoma DNA in the presence of low-molecular-weight DNA from untransformed rat cells; (A) DNAfrom polyoma Rat-li cells grown in the absence ofantiserum; (A) DNA from polyoma Rat-11 cellsgrown in the presence of antiserum.

-~I

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from

POLYOMA VIRUS-TRANSFORMED RAT CELLS 441

Viral genomes associated with chromo-somal DNA. To determine whether the poly-oma-transformed rat cells also contained viralDNA molecules associated with the chromo-somal DNA and therefore presumably inte-grated, purified high-molecular-weight DNAfrom transformed cells was analyzed for itsability to influence the rate of reassociation ofP2P]polyoma DNA (Fig. 4). On the basis of thereassociation kinetics, polyoma Rat-13 and ts-aRat-23 contained 6 and 10 copies of viral DNAequivalents associated with the host genome,respectively. It is important to note that thevalue for ts-a Rat-23 cells was obtained withDNA extracted from cells grown at 39.5 C.Under these conditions these cells did not con-

1

fss

2 4 6 8 10 12cop t X103 (,ug-hrs.- ml-1)

FIG. 3. Reassociation kinetics of [32P]polyomaDNA in the presence ofform I and form II (linear or

nicked circular) low-molecular-weight DNA frompolyoma Rat-12 cells. Hybridization was done as

described in Materials and Methods and the data are

expressed as in Figure 2. Symbols: (0) reassociationof t32P]polyoma DNA by itself (control); (A) reasso-

ciation of[32P]polyoma DNA in the presence offormII DNA; (A) reassociation of [32P]polyoma DNA inthe presence ofform I DNA.

tain detectable free viral DNA (manuscript inpreparation).

Virus rescue from cells containing inte-grated and free viral DNA molecules. Wewanted to determine whether the virus rescuedfrom transformed rat cells upon fusion withmouse cells originated from free, integrated, or

both types of viral DNA molecules.The thermosensitive polyoma mutant, ts-a,

does not multiply at the nonpermissive temper-ature (40 C) but does multiply at the permis-sive temperature (33 C) (10). ts-a-transformedrat cells at 33 C contain approximately 20 cop-

ies of free viral DNA molecules per cell. Whenthese cells were incubated at the nonpermissivetemperature, viral DNA disappeared from theHirt supernatant, but 10 copies of viral DNAwere still found associated with the host DNA(manuscript in preparation). ts-a-transformedcells were incubated for 10 days at the nonper-missive temperature. When these cells were

shifted to the permissive temperature and incu-bated for 3 days, free viral DNA moleculesreappeared (20 copies/cell).

Cells from the ts-a Rat-23 and ts-a Rat-21lines grown at 33 and 40 C were fused sepa-

rately with 3T3 cells in the presence of Sendaivirus. Cells were then incubated at the permis-sive temperature. Virus was extracted and ti-tered. If the presence of free viral DNA hadbeen necessary for virus rescue, the viral yieldfrom the fusion using cells from 33 C shouldhave been higher and earlier than that fromthe fusion using cells from 40 C, since the lattercells did not contain free viral DNA at the timeof fusion. However, similar amounts of viruswere rescued from the cells grown at 33 and40 C, either after 3 or 5 days (Table 5).The fact that the virus yield from cells that

contained or did not contain measurableamounts of free viral DNA molecules at thetime of fusion is about the same suggests thatthe presence of nonintegrated viral DNA does

TABLE 4. Infectivity of the "'free" viral DNA extracted from polyoma (Py)-transformed rat cells

Expt DNA sourcea Concnb (pig/ml) (PFU/ml) Specific infectiv-

I 3T3D lyrically infected with wild-type Py 0.2 3.6 x 104 1.8 x 105(control)

Py Rat-12 0.011 20 1.8 x 103ts-a Rat-13 0.013 20 1.6 x 103

II 3T3D lyrically infected with wild-type Py 12 1.5 x 106 1.2 x 105Py Rat-12 0.04 30 7.5 x 102ts-a Rat-13 0.07 50 7.1 x 102

a ts-a Rat-13 cells were maintained at 33 C; the other cells were maintained at 37 C.b Determined by DNA-DNA reassociation kinetics.e DNA from polyoma wild type-infected or -transformed cells was titered at 37 C, DNA from polyoma ts-a-

transformed cells at 33 C.

VOL. 18, 1976

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from

442 PRASAD, ZOUZIAS, AND BASILICO

2.5

1

fss2.0

1.5-

1.0~~~~~~~~A

1.0 AS"2 4 6 8 10 12

cop t X 103 (/ug-hrs.-ml-1)FIG. 4. Renaturation of [32P]polyoma DNA in the

presence of high-molecular-weight chromosomalDNA from polyoma-transformed rat cells. Each reac-

tion mixture contained 500 ,ug of cellular DNA perml and 2.5 x 10-3 to 3.0 x 10-3 pug of [32P]polyomaDNA per ml (specific activity, 2.3 x 106 counts/minper pg). The data are plotted as 1 If,8 versus Cop x t,where Cop represents the input of probe in micro-grams per milliliter and t the time of sampling inhours. Symbols: (@) renaturation of [32P]polyomaDNA in the presence ofDNA from untransformed ratcells; (A) DNA from polyoma Rat-13 cells; (A) DNAfrom ts-a Rat-23 cells (growing at 39.5 C).

not significantly increase the efficiency of virusrescue by fusion.

Free viral DNA molecules in rat embryocells transformed by polyoma. Primary ratcultures were prepared from Fischer's rat em-

bryos. Cells of secondary cultures were trans-formed by polyoma at an MOT of 500 PFU in softagar. After 5 to 6 weeks of incubation, trans-formed colonies were isolated. Transformedcells grew slowly, yet attained high saturationdensity. When these cells were inoculated intosoft agar, they again formed colonies. Cellswere analyzed for the presence of free viralDNA molecules. The Hirt supernatant fromthese cells (polyoma Rat P-1) contained 18.5copies of free viral DNA molecules/cell (aver-age) (Table 3).

DISCUSSION

The system of polyoma-transformed rat cellsdescribed in this paper has a number of inter-esting features. F2408 cells grow easily in cul-ture, have a good efficiency of plating and adiploid chromosome number, and easily with-stand high incubation temperatures. They can

be transformed with good efficiency by polyomavirus. Untransformed cells do not grow at all inagar medium, thus making it very easy to de-tect even very low frequencies of transformantsin a transformation assay.The rat cells used are for all practical pur-

poses nonpermissive for polyoma virus multi-plication, independent of whether they are in-fected with virus or viral DNA. The nonpermis-siveness is, however, not absolute, as infectionat a very high MOI (>1,000 PFU/cell) results inthe appearance of a few (-0.1%) virus-produc-ing cells. This situation is very similar to whatis found in BHK-21 hamster cells (1, 7, 8). Incontrast to BHK-21 cells, however, F2408 ratcells always produce virus upon fusion withmouse 3T3 cells. The transformed cells per se donot generally produce virus, and in our caseonly an extensive search for virus productionrevealed that occasionally these cells sponta-neously produce small amounts of infectiousvirus.

This situation superficially resembles that ofthe LPT line of polyoma-transformed rat myo-blasts described by Fogel and Sachs (5). Thereare, however, two major differences. (i) In theircase only cells transformed by large-plaque pol-yoma virus, and only one clone in particular,could be induced to make virus. In our case, alltransformed clones tested produced virus uponfusion, irrespective of the type of virus used forinfection. (ii) Spontaneous virus production didnot generally occur in our transformed lines orwas, at best, a very exceptional event.

It should be noted that Kimura et al. (12)reported that most polyoma-transformed ratcells they isolated were virus-free and producedvirus upon fusion with mouse cells. The situa-

TABLE 5. Rescue of virus from ts-a-transformed ratcells by fusion with mouse 3T3 cells

Time of in-

Clonee cubation of Yield of virusthe fused (PFU/culture)

cells (days)

ts-a Rat-21 (40 C) 5 1 x 103ts-a Rat-21 (33 C) 5 1.8 x 103ts-a Rat-21 (40 C) 3 0"ts-a Rat-21 (33 C) 3 0ts-a Rat-23 (40 C) 5 7.5 x 104ts-a Rat-23 (33 C) 5 6.5 x 104ts-a Rat-23 (40 C) 3 2 x 103ts-a Rat-23 (33 C) 3 1 x 103

a Cells from ts-a Rat-21 and ts-a Rat-23 lineswhich had been growing at 33 or 40 C for 10 dayswere fused with 3T3 cells by means of inactivatedSendai virus. Fused cells were incubated at 33 C.Virus was extracted and titered at 33 C.

b <10 PFU/culture.

J. VIROL.

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from

POLYOMA VIRUS-TRANSFORMED RAT CELLS 443

tion appears to be very similar to what wefound in rat F2408 cells.The most interesting finding in our study

was that, in all cases tested, transformed ratcells contained, in addition to presumably inte-grated viral DNA, a small number of noninte-grated viral DNA molecules. This number re-mains rather constant even with long times inculture, and the variation among differentclones is not very high.Although no infectious virus can generally be

found in these cultures, it was still important torule out the possibility of a virus-carrier state(below detectable levels) being responsible forthis phenomenon. This was conclusively ex-cluded by the fact that incubation in the pres-ence of antipolyoma serum did not affect thepresence of free viral DNA. In addition, sub-cloning in the presence of antipolyoma serumproduced cell lines that still contained free viralDNA.The latter experiment is also important in

that it showed that every cell in the trans-formed population was capable of carrying or atleast producing free viral DNA molecules. It isnot clear at present whether all cells in thetransformed population contain a limited num-ber of nonintegrated polyoma DNA molecules,or whether only a minority of the cell popula-tion contains a large number of viral DNAmolecules per cell at any given time. The for-mer situation would be analogous to that of acell carrying a plasmid DNA (3), whereas thelatter would be likely to be caused by somemechanism of induction of viral DNA replica-tion continuously occurring with a low but con-stant probability in every cell. The free viralDNA would, in the latter case, originate fromthe integrated one and replicate to some extentwithout concomitant virus production. Work isin progress to distinguish between these twohypotheses.With regard to the origin of the free viral

DNA, the data presented in this paper suggestthat it originates from an integrated template.Thus, we were not able to establish the pres-ence of free viral DNA in a majority of cells at 1week after infection, and, more importantly,cells transformed at an MOI ranging from 1 to1,000 PFU/cell contained approximately thesame number of free viral DNA molecules. In apaper now in preparation we will show conclu-sively that free viral DNA can originate fromintegrated genomes.

It is interesting to note that the presence offree viral DNA does not seem to influence theefficiency of the virus rescue by fusion. In ouropinion there are two possible explanations ofthis finding. (i) Fusion of the transformed cells

with permissive mouse cells causes immediateexcision of the integrated viral DNA. Thus,cells that contain nonintegrated viral DNA orcells that do not contain it would be in practicein the same state shortly after fusion. (ii) Thenumber of cells that contain free viral DNAcould be quite low (e.g., <0.1%). Since fusionprobably does not involve more than 10% of thecells, most of which do not become virus pro-ducers anyway (21, 22), the contribution ofthese cells to the total virus production wouldbe negligible.A final point of interest resides in the nature

of the free viral DNA. As shown in this paper,it is generally present as nuclear, supercoiledform I DNA, which is infectious. Its specificinfectivity is, however, very low. This findingstrongly suggests that these molecules containa large proportion of detectives. It is temptingto speculate that this results from improperexcision of integrated viral DNA.

ACKNOWLEDGMENTS

We wish to thank Ilene Kievman for her excellent tech-nical assistance.

This work was supported by Public Health Servicegrants CA 11893 and CA 16239 from the National CancerInstitute. I. P. was a Special Fellow and C. B. is a scholar ofthe Leukemia Society.

LITERATURE CITED

1. Basilico, C., Y. Matsuya, and H. Green. 1970. Theinteraction of polyoma virus with mouse-hamster so-matic hybrid cells. Virology 41:295-305.

2. Basilico, C., and R. Wang. 1971. Susceptibility to su-perinfection of hybrids between polyoma transformedBHK and normal 3T3 cells. Nature (London) NewBiol. 230:105-107.

3. Clowes, R. 1972. Molecular structure of bacterial plas-mids. Bacteriol. Rev. 36:361-405.

4. DiMayorca, G., J. Callender, G. Marin, and R. Gior-dano. 1969. Temperature sensitive mutants of poly-oma virus. Virology 38:126-133.

5. Fogel, M., and L. Sachs. 1969. The activation of virussynthesis in polyoma transformed cells. Virology37:327-334.

6. Fogel, M. 1972. Inducation of virus synthesis in poly-oma transformed cells by DNA antimetabolite and byirradiation after pretreatment with 5-bromodeoxyuri-dine. Virology 49:12-22.

7. Folk, W. R. 1973. Induction of virus synthesis in poly-oma-transformed BHK-21 cells. J. Virol. 11:424-431.

8. Fraser, K. B., and M. Gharpure. 1962. Immunofluores-cent tracing of polyoma virus in transformation ex-periments with BHK-21 cells. Virology 18:505-507.

9. Freeman, A. E., R. V. Gilden, M. L. Vernon, R. G.Wolford, P. E. Hugunin, and R. J. Huebner. 1973. 5-Bromo-2-deoxyuridine potentiation of transformationof rat embryo cells induced in vitro by 3-methylchol-anthrene: induction of rat leukemia virus gs antigenin transformed cells. Proc. Natl. Acad. Sci. U.S.A.70:2415-2419.

10. Fried, M. 1965. Cell-transforming ability of a tempera-ture sensitive mutant of polyoma virus. Proc. Natl.Acad. Sci. U.S.A. 53:486-491.

11. Hirt, B. 1967. Selective extraction of polyoma DNAfrom infected mouse cell cultures. J. Mol. Biol.

VOL. 18, 1976

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from

444 PRASAD, ZOUZIAS, AND BASILICO

26:365-369.12. Kimura, G., A. Itagaki, and J. Summers. 1975. Rat cell

line 3YI and its virogenic polyoma and SV40-trans-formed derivatives. Int. J. Cancer 15:694-706.

13. Koprowski, H., F. C. Jensen, and Z. Steplewski. 1967.Activation of production of infectious tumor virusSV40 in heterokaryon cultures. Proc. Natl. Acad. Sci.U.S.A. 58:127-133.

14. MacPherson, I., and L. Montagnier. 1964. Agar sus-pension culture for the selective assay of cells trans-formed by polyoma virus. Virology 23:291-294.

15. Marmur, J. 1961. A procedure for the isolation of deoxy-ribonucleic acid from microorganisms. J. Mol. Biol.3:208-218.

16. Pagano, J., J. H. McCutchan, and A. Vaheri. 1967.Factors influencing the enhancement of the infectiv-ity of polio virus ribonucleic acid by diethylamino-ethyl-dextran. J. Virol. 1:891-897.

17. Prasad, I., D. Zouzias, and C. Basilico. 1975. Simianvirus 40 integration sites in the genome of virus-

J. VIROL.

transformed mouse cells. J. Virol. 16:897-904.18. Sharp, P. A., U. Petterson, and J. Sambrook. 1974.

Viral DNA in transformed cells. I. A study of thesequences of adenovirus 2 DNA in a line of trans-formed rat cells using specific fragments of the viralgenome. J. Mol. Biol. 86:709-726.

19. Toniolo, D., H. K. Meis, and C. Basilico. 1973. Atemperature sensitive mutation affecting 28S ribo-somal RNA production in mammalian cells. Proc.Natl. Acad. Sci. U.S.A. 70:1273-1277.

20. Tooze, J. 1973. Transformation by polyoma virus andSV40, p. 350402. In The molecular biology of tumorviruses. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

21. Watkins, J. F. 1974. The SV40 rescue problem. ColdSpring Harbor Symp. Quant. Biol. 39:355-362.

22. Watkins, J. F., and R. Dulbecco. 1967. Production ofSV40 virus in heterokaryons of transformed and sus-ceptible cells. Proc. Natl. Acad. Sci. U.S.A. 58:1396-1403.

on June 7, 2018 by guesthttp://jvi.asm

.org/D

ownloaded from