Replication of simian virus 40 DNA in normal human fibroblasts and in fibroblasts from xeroderma pigmentosum.

(1)

0022-538X/81/080481-09$02.00/0 Vol.39,No.2

Replication of Simian

Virus 40 DNA in Normal Human

Fibroblasts

and in

Fibroblasts

from

Xeroderma

Pigmentosum

HARVEY L. OZER,* MARTINL. SLATER,t JAMES J. DERMODY, AND MORTON MANDEI4

Department of Biological Sciences, Hunter College, City University of New York, New York, New York 10021,* andWorcester Foundation for Experimental Biology, Shrewsbury,Massachusetts01545

Received 10 November 1980/Accepted 29 April 1981

Simian virus40infectionofsemipernissivehumandiploidfibroblasts(HF), at

earlypassagein cell culture,was comparedwith that ofpermissive established monkeycell lines.Viral DNA can bereadily detectedat24 to 48 hpostinfection

at370Cwithahigh multiplicityofinfection, approaching10% ofthat of monkey

cells(TC7). The length of time necessaryforreplicationofan averagemolecule of viral DNA wasfound to be indistinguishable in HF and TC7 cells. Strand

elongation plus terminationwereassessed byfollowingtheaccumulationofDNA

I at40°C from replicative intermediates oftsA30 prelabeled at

330C,

obviating isotope pool problems. Combined initiation and elongation of wild-type viral

DNAwasmeasured by densityshiftexperimentsinvolving a5-bromodeoxyuridine chase ofprelabeled [3H]thymidine-labeled viral DNA. Determination of

accu-mulation of viralTandVantigens supports the conclusion that themostlikely basis for the reduced virus yield in HF cellsresults from the inefficiency ofan

earlystage in virus infection, before or during uncoating. Similar results were

obtained in fibroblasts derived from patients with xeroderma pigmentosum, suggesting that enzymes ofUV repair arenot required in unirradiated simian virus40DNAsynthesis.

Papovaviruses, including simian virus 40

(SV40), initiate infection in a wide variety of cells from different species (for a review see

reference 33). Typically, twotypes ofinfection

may occur:permissive ornonpermissive. In the

firstcase,fullvirusreplicationresults in synthe-sis of progeny infectious viral

particles.

In the latter case, there isa partial expression of the viral genome, miniimalor noviral DNA synthe-sis,andnoprogenyvirusproduced.Thiscourse may either be abortive or associated with

per-sistence of the viral genome, typically in an

integratedmanner.InthecaseofSV40, permis-sive infection characteristically occursin

mon-key cells andnonpermissiveinfectionoccursin rodentcells.Athirdtype of infection has been terned "semipermissive." In human cells, for example,SV40 infectionresultsinappearanceof virus-encoded early proteins (T

antigens)

and onlylimitedprogenyvirus(2,4,9,21,22).There

islittleor noinformationontherateandextent

of viralDNAsynthesis. We haveundertakena

detailed study of viral DNA synthesis in early

passage normaldiploidhuman fibroblasts inan

effort to understand better the basis for this

tPresent address: Division of Research Grants, National Institutes ofHealth, Bethesda,MD 20205.

*Present address:DepartmentofBiocheniistry,University ofHawaii School ofMedicine, Honolulu,HI96822.

semipermissivity; most notably, to determine whether such semipermissivity represents a

quantitatively or qualitatively different

virus-cell interaction than the typical permissive

in-fection. Suchastudy takesonadded interestat

the present time for two reasons: first, there

have beenincreasinglymoreinvestigations into the biochemistry of humanpapovaviruses and theirrelationshiptoSV40(27, 33).Second,there

arecurrently available several cell lines of fibro-blasts derived frompatients withhereditary dis-ordersaffecting cellular DNArepairand, possi-bly, DNAreplication (11, 28). The latter

repre-sent apotentialsource ofcellmutants tostudy

theprofound cellular contributiontoviral DNA replication.

We have foundthatSV40efficientlyreplicates

its DNA in human fibroblasts. Moreover, the

rate of viral DNA strand elongation, which is

totallydependentoncellularfunctions,is indis-tinguishable from that of permissive monkey

cells.Finally,weobserved nodifferenceamong

several complementation groups of fibroblasts

derived from patientswith xeroderma

pigmen-tosum(XP),adisorderassociatedwithadefect

inrepairof DNAdamagedbyUVlight,

indicat-ing that such an enzyme(s) defective in this

disorder isnotrequiredforreplicationof

unir-radiatedviralDNA.

481

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

MATERIALS AND METHODS

Virus. Wild-type SV40 (strain SV-S) and tsA30 werepropagatedat 37and330C,respectively,at alow multiplicity of infection (MOI) in established green monkey kidney (GMK) cellsaspreviouslydescribed (7,19).Plaque assayswereperformedonTC7or CV-1Pmonolayers.

Cell lines. Cell lineswerepropagated in the Dul-becco-Vogt modification of Eagle medium (DME) (Microbiological Associates) with4.5gofglucoseper liter and 10% fetal calf serum (complete medium). GMKcell lineswerepassagedaspreviouslydescribed (13) and were obtained from the following sources: TC7 from J.Robb and J.Kaplan,respectively,CV-1P from D.Jackson, and Vero from the AmericanType Culture Collection(ATCC).Human fibroblast cultures wereobtained at passage3 to 6from cellrepositories andpassagedaminimal number of timesat 1:3 sub-cultures. HS lineswereobtained fromthe Cell Culture Laboratory, University of California School of Public Health.Their propertiesaredescribedby Smithetal. (30) and Zouziasetal. (34). CRL cell lines were ob-tainedfrom theATCC. Theproperties of those from XPpatients are summarized by Robbins etal. (23). Thefamilial relationships ofpatient9 (XP9BE) are reported by Lynch et al. (16). All cell lines were verified to befree of mycoplasma by cultivation pro-cedures, except for CRL 1199and CRL 1200which werepositive asreported by the ATCC. The tissue origins and XP phenotypesaresummarized in Table 1.

Chemicals and solutions. 5-Bromodeoxyuridine (BUdR) and deoxycytidine (CdR)wereobtained from

Sigma Chemical Co. (A grade). Fluorodeoxyuridine

(FUdR), thymidine (TdR), and

cytosine-l-fl-D-arabi-nofuranoside-HCl (araC) were obtained from

Calbi-ochem.Solutionswereprepared in distilled water and filtersterilized.

[image:2.503.57.247.449.634.2]

Viral DNAsynthesis. Subconfluentorconfluent TABLE 1. Human fibroblastic celllinesa

Cellline XP designation(genetic Tissueoforigin

group)

~~~tient

HS27 None(normal) Foreskin,

new-born HS74 None (normal) Bonemarrow,

fetal CRL 1161 XP9BE (C) Skin, 12 yr CRL 1162 XP4BE (variant) Skin,27 yr CRL1165 None (parent of Skin,54 yr

XP9BE)

CRL 1166 XP2BE(C;siblingof Skin, 22 yr XP9BE)

CRL1167 None (parent of Skin, 54 yr XP9BE)

CRL 1170 XP1BE(C) Skin,27 yr CRL1199 XPIlBE (B) Skin, 28yr CRL1200 XP7BE (D) Skin, 11 yr

CRL1204 XP1OBE(C) Skin, 16 yr

CRL 1223 XP12BE (A) Skin, 7yr

aOriginsand properties of

cell

lines are described

inthetext.

cultures in 60-mmpetri dishes were infected with 0.5 ml of an unpurified virus preparation as previously described (20). At appropriate times postinfection (p.i.), cultureswereradiolabeled for 1-to4-h periods with [3H]TdR at 10,tCi/ml(specific activity, 50Ci/ mmol, Schwarz/Mann). Viral DNAwasextractedby the Hirtprocedure (12) and quantitated asforms I and II(18to21S) by sedimentationonneutralsucrose gradients (NSG), using purified [14C]DNA I as an internal marker as described previously (29). Incor-porationwasdetermined afterprecipitationwith 5to 10% trichloroacetic acid.

Viral DNA strandelongation. Confluent cultures in T25 flasks were infected with 0.2 ml of virus as described above. After 3to7daysat33°C (depending on the cell line and viruspool),triplicate cultures were radiolabeled with20,uCiof[3H]TdR per ml for 20min (TC7 cells)or30min(human fibroblasts)at33°C and then shifted rapidly to a water bathequilibrated at 40.5°C. Incubations were terminated atthe time of shiftto40.5°Candat10-minintervalsbyplungingthe flask intoice, rapid washing (twice) by cold phosphate-bufferedsaline, and Hirt extraction. Thepooled Hirt supernatantwaslayeredontoNSG, and viral DNA I wasdetermined. In thoseexperimentsinvolving infec-tion of CRL1161and CRL1166bytsA30, the region of the NSG containing DNA Iwasidentifiedonthe basis of the[14C]DNA.Thepeak fractionswerepooled, dialyzed against 0.15 M NaCl-0.01 M EDTA-0.01 M Tris-hydrochloride (pH 7.6), andcentrifugedto equi-librium in CsCl(p =1.56g/cm)containing 100,ugof ethidium bromide perml. Quantitation of[3H]DNA wascorrected for recovery ofadmixed ['4C]DNA I in thiscase.

Kineticsof viral DNA reentry. Theprocedure

usedwas amodification of that employed by Roman and Dulbecco (24, 25) for polyoma virus-infected mousecells. Confluent cells were infected in 60-mm dishes withwild-typeSV40. At 24or40hp.i. [3H]TdR wasaddedat10,uCi/mlfor60min,followed by i05 M TdRforanadditional60min.The medium was then removed, the culturewaswashed once in DME, and BUdR mediumwasadded (5 x 10-5 M BUdR, 2 x lo-5 M FUdR, 1 x 1i-5 M deoxycytidine [CdR] in complete medium). Viral DNA was extracted after various durations of chase in BUdRby the Hirt pro-cedure. DNA I was isolated from pooled triplicate cultures aftersedimentationonalkaline sucrose gra-dients. After neutralization, HL (DNA substituted withBUdR inonestrand) and LL (no BUdR substi-tution) moleculeswereseparated by equilibrium cen-trifugation in CsCl, p=1.72g/cm. The percentage of 3Hcpm found inHL DNA [HL/(HL+ LL) x 100] wasdetermined.Incorporation of[3H]CdR(1,uCi/ml

in10' M) into DNA Iwaslinearininfected TC7 or HS27 cells over a 6-h period in BUdR medium in control experiments; the rate of incorporation of BUdR into either viral or cellular DNA was approxi-mately60to 70% of that observed for TdR in compa-rable medium.

Virus production in human fibroblasts. Con-fluent60-mmpetri disheswereinfected with 0.2 ml of virus. After adsorption for 2 h, the inoculum was removed, the monolayers were washed five times with DME,5ml ofcomplete medium wasadded, and the

http://jvi.asm.org/

(3)

SV40 REPLICATION IN HUMAN FIBROBLASTS 483 cultureswere harvested by freezing and thawing at

appropriate times.

Serologicalassays.Microcomplement fixationfor SV40-specific antigens were performed as described

previously (20). T antigen(s)wasassayed witha

ham-ster antiserum to a transplantable SV40 tumor. V antigen was assayed with a hyperimmune monkey

antiserum prepared against viralcapsids, which was

nonreactive with T+ V- cells. Infected extractswere

prepared from duplicate or triplicate 100-mm petri

dishesinfectedwith1mlof wild-type virus,undiluted

or after dilution in DME containing 1% fetal calf serum.

RESULTS

Viral DNA synthesis in human

fibro-blasts.Whensubconfluentorconfluentcultures

ofhuman fibroblastsareinfected with SV40 at

multiplicities of infection (MOIs) ofgreaterthan

10 at370C, miniimal cytopathic effects are

ob-served. Virusyield (in PFU) has beenreported

tovaryover awiderangefromminimallyabove

theabsorbed levelto1 to3PFUpercell (4, 22)

incontrast to greaterthan 100PFUpercell(109

to 1010 perculture) commonly observed for

per-missive monkey cell lines (33). A number of

earlypassagehumanfibroblast cell strainswere

tested for viral DNAsynthesisat2days

postin-fection with SV40. The resultsaresummarized

in Table 2. Different cell strains vary in their

susceptibilities, but readily detectible levels of

viral DNAareobserved inthecaseoffibroblasts

from bothpresumablynormal donors and those

frompatients with the hereditary disorder XP.

[image:3.503.260.454.68.219.2]

Table 3 showsacomparison between the human

TABLE 2. Synthesis of SV40DNA in human fibroblastsa

Viral DNAC

Celllineb % oftotal

cpm perculture counts

HS 27 40,560 9.4

HS 74 .3,150d c1.4

CRL1161 10,900 4.3

CRL 1162 7,920 2.0

CRL 1199 31,700 6.8

CRL 1200 39,600 7.7

CRL 1223 56,700 7.7

aConfluentmonolayersin60-mmpetridisheswere

infected withwild-type SV40at anMOI of 20to40 PFUpercell andlabeled for 2 h with[3H]TdRat 48 hp.i.asdescribed in thetext.

bCelllinesarelisted in Table1.

cViral DNAwas quantitated byNSG analysis of the supernatant fraction by the Hirt procedure as

described inthe text. Extracts from uninfected cells processed by the same method averaged approxi-mately 2,400cpmand 1.5% ofthetotalcounts.

[image:3.503.57.249.435.551.2]

dNodistinctpeakwasobservedonNSG.

TABLE 3. Synthesis ofSV40DNA inmonkeycellsa

ViralDNA'at: 24 hp.i. 48 hp.i. Cellline

cpmpercul- % ofto- cpm percul %of

to-ture %

ftal

ture %

ftal

counts counts

T C7 3.8X 105 24 2.6X 106 75 CV-1P 3.5x105 34 >1.3 x 105c 43 Vero 6.4X104 9.3 1.7X105 37

HS27 1.0x104 1.9 4.1 X104 9.4

a Confluent monolayers were infected as described inTable 2 except for Vero cells in which the MOI was approximately 5 PFU per cell.

'ViralDNAwasdetermined as in Table 2.

c

Underestimate

of viral DNA due to evident viral

cytopathicity.

fibroblasts HS27 and three permissive monkey

cell lines. Dueto the multiplevariables among

thedifferent cultures, these resultscan only be

considered approximations. Nonetheless, it is

clear thatconsiderable viralDNAsynthesiscan

occurin humanfibroblastsat alevel

approach-ing 10% of that of permissive cells. A more

detailed comparison between HS27 and TC7 cellswasundertakenin anefforttounderstand

thebasis for thisapparentreducedrateofviral

DNAsynthesisinhuman fibroblasts.

DNA strand elongation was determined p.i.

withwild-type SV40or atemperature-sensitive

mutant(tsA30) ofSV40 defectiveininitiationof

viral DNAsynthesisattherestrictive

tempera-ture. Cultureswere infected at

330C

and

incu-bated until viral DNA synthesis was readily measurable. Tritiated TdRwasadded for20 to 30 min to permit labeling of the intracellular nucleotide andreplicatingviral DNApools.

Cul-tureswerethen shiftedto

400C,

and

radioactiv-ity in progeny viral DNA I was determined immediately andat10-min intervals. Wewould

expect accumulation ofradioactivityin DNA I

to be a direct reflection of strand elongation since ithadpreviouslybeenreportedthat initi-ation ofnewrounds of viral DNA synthesis in

monkey cellsisinhibited within 10minforthis

mutant. Figure 1A shows the results for TC7,

andFig. 1Bshowstheresults for HS27.Incells infected withtsA30, radioactivity in DNA I

in-creases in monkey cells for

approximately

20

min as expected, which is consistent with the

inactivation of A function (-10 min) and the

timenecessarytocompleteanalready

initiation-replicative intermediate (-15 min) (15, 30). A

similar result was obtained for HS27 cells

in-fectedwith tsA30. The markeddecrease in

ac-cumulation of DNA I after 20 min at

400C

cannotbeattributedtotoxicityassociatedwith

VOL.39,1981

http://jvi.asm.org/

(4)

thevirus infectionorradiolabeling procedureor

both since the parallel cultures infected with wild-type SV40 showprogressiveaccumulation of DNA I foratleast 60minafterashiftto400C. Direct comparison of the level of viral DNA synthesis between thetwocelllines isnot

pos-sibleinview of the different conditions of infec-tion,etc. (seefigure legends).

Wenextinvestigatedthe

efficiency

of reentry of viralDNAinthetwocelllinesbythe method of Roman and Dulbecco(25).Cellswereinfected withwild-type SV40, and the replicatingDNA poolwaslabeled with[3H]TdR.Subsequent

rep-lication was determined by incorporation of BUdR. Theproportion ofDNA Iofintermediate density was determined at 24 and 40 h p.i. in

monkey cells and40 hp.i.inhuman cellssince

there was insufficient DNA replication at the

earlier time p.i. to obtain accurate

determina-tions. Aspreviously observed forpolyoma

virus-infectedmousecellsandSV40-infected monkey cells,usingasomewhat different labeling regi-men, reentryof viral DNAmolecules in permis-sivemonkeycellsisincomplete, withaplateau

value of less than 100% (10, 24, 25). Although

theplateau valuecanvaryconsiderably with the time p.i., the period during which reentry is

observed isreasonably constant,being

approxi-mately4h afteraddition ofBUdR(Fig. 2). The results with theinfected human fibroblastsare

similar (e.g., within experimental error consid-ering the number of manipulations involved),

25

I0

x

0

20

15

10

5

6

5

4

0

X 3

aL 0

2

a..

0

Xl a

IL

0 20 40 60 0 20 40 60

MINUTES AT40.5 C

FIG. 1. ViralDNA strandelongation. Confluent cultures wereinfected bytsA30 (0) orwild-typeSV40

(0) at 33°C, pulse-labeled with[3H]TdR, and shifted to40.50C in the continuouspresence of isotopefor quantitationof accumulationof viral DNA IonNSGasdescribedin thetext.(A) Monkey cells(TC7)infected for3days; (B) human cells(HS27) infected for4days.The MOIwascomparableforbothcelllines (20to40

forwildtypeand 10to20fortsA30).

+iV.

sj 0. x

3 4 5 6 7 2 3 4 5 6 7 2 3 4 5 6 7

[image:4.503.135.380.250.422.2]

HOURS IN BUDR

FIG. 2. Kineticsofreentry ofviralDNA synthesis. Confluent cultures infectedwith wild-type SV40were

pulse-labeled with[3H]TdR, followedbyachase in BUdRmediumasdescribed in thetext. Theproportion

ofviral[3H]DNA Iconvertedtointermediate HLdensitywascalculated after purificationby Hirt

fraction-ation andalkalinesucrosegradientandcesium chloride-ethidium bromide equilibrium sedimentations. (A) Monkey cells (TC7)at24h p.i.; (B)monkey cells(TC7)at40h p.i.; (C) humancells(HS27) at40h p.i.

A

http://jvi.asm.org/

[image:4.503.135.377.487.603.2]

(5)

SV40 REPLICATION IN HUMAN FIBROBLASTS 485

particularlywhen compared with those of

mon-key cells infected for the same length of time

(compare Fig. 2B and C). This procedure

mea-suresall stagesofviral DNA replication.

Inas-muchaswehaveshown inthepreceding section

that thosesteps afterinitiation donotdiffer in

thetwoinfected celllines,we canconclude that

nomajordifference exists in reinitiation of viral

DNAsynthesisaswell.

Accumulation of viral antigens in

in-fectedcells. Theseresultsonthesimilarity in

theratesof viral DNA synthesisbetween

semi-permissivehuman cellsand permissivemonkey

cellssuggestthat thereduced level of viral DNA

inthe former isareflection ofadefect inastep

earlier in infection. Consistent with thismodel,

other investigators have reported that not all

human cells accumulate detectable Tantigen by

fluorescent antibody p.i. by amoderately high

MOI (i.e., sufficient to induce T antigen in

greaterthan 90% ofmonkey cells) (2, 3, 21). We similarly observed that accumulation of T

anti-genby complement fixation at 24 h p.i. is

re-ducedin HS27cellsascomparedwith TC7cells

when thesameinoculum is usedtoinfectparallel

cultures (Table 4). Cultures were harvested at

24 h p.i., and an inhibitor of DNA synthesis

(araC,1 x

10'"

M)wasincluded in the medium

tominimize differences secondarytothe

differ-entlevelsof DNAsynthesisin the twocell lines.

(No differencewasobserved inpreliminary

ex-periments withand withoutaraC.)However,as

the virusinputisincreasedfrom 10 to 20 to 50

to100 PFUpercell, there isonlyaslight

incre-ment in Tantigen in monkey cells, whereas T

antigencontinuestoincreaseproportionatelyin

HS27 cells. At lowMOI,there is alsoamarked

differencein Tantigenaccumulation. Theeffect

of this differencebetween the twopopulationsis

[image:5.503.261.452.399.478.2]

further indicated by the results in Table 5 in

TABLE 4. AccumulationofTantigen by complement fixationa

Virusinput T-antigentiter for:

(PFUx10-6) TC7 HS27

200 640 170

40 160-480 20-43

13 80-160 10-13

7 80 <3

aConfluent 100-mm petri disheswere infected as

describedformicrocomplementfixation in thetext. A

singlestock ofwild-typeviruswasdiluted in medium with 1%serumforparallelinfections. Two dishes of infected TC7 cellswereextractedin 0.5ml,and three dishes ofinfected HS27cells wereextracted in 0.25 ml for antigen titration. Data are presented as antigen

perinfected petridish. Thereweresimilarcell

num-bersin cultures of bothcelllines.

which accumulationofvirus particles as V

anti-gen was determinedby complement fixation at

a high MOI. There is a fivefold-lower level in

HS27 cellssimilartothe reduction in the level

ofviral DNA. Fluorescent-antibody studiesfor

viral capsidproteinsconfirmthata minority of

human cells are infected at alltime points in

contrastto>90% of themonkeycells (datanot

shown). There is also a comparable reduced

yield of virusatdifferenttimesp.i. (Table 6).A

humanfibroblast(HS74)cell linewhich showed

minimal viral DNAsynthesis in Table 2 has a

stillfurther reduced virus yield.

Viral DNA synthesis in human fibro-blastsfrom XP. The studies described above indicated that thesemipermissive nature of

in-fection ofearlypassagehumanfibroblastswith

SV40 is consistent withareadily detectable level of viral DNAsynthesis in those cells whichare

efficiently infected. Therefore,wesoughtto de-terminewhether viral DNA replication occurred efficiently in cells derived from patients with hereditary defects in DNAsynthesis orrepair.

We chose to examine fibroblasts frompatients withXP,adiseasecharacterizedby skintumors

and defects inrepair ofdamagetoDNAby UV light. Multiple complementation groups have beenreported,someof whichmightlogicallybe expectedtobe defectiveinfunctions associated with synthesis of unirradiated DNA (one or TABLE 5. AccumulationofT and Vantigensby

complement fixationa

TC7 HS 27

Timep.i.

(h) Tanti- Vantigen Tanti- V

anti-gen gen gen

24 320 640 128 128

48 640 6400 256 1280

72 320 12,800 256 640

aMicrocomplementfixation ofinfected

cell

extracts

wasperformedasdescribedinTable4.2x 108PFUof SV40wereused forinfection inallcases.

TABLE 6. Virusyieldin humanfibroblastsa Timep.i. PFU per ml in:

(h) CV-1P HS27 HS74

2 <1.0 x 108" <1.0 x105b 3.0x105

24 1.0X106 1.0X105 1.0X105 48 6.4x107 3.6x106 1.0x 105 72 1.1xi08 6.2x 106 3.8x106

120 NTc 1.2x 107 3.4x106

168 NT 1.5x 107 3.8x106

a

Infected

cultureswereharvested forvirusas

de-scribed in thetextandassayedforplaqueformation onCV-1Pmonolayers.

bNoplaque observedatlowest dilution tested.

cNT,Not tested.

39,1981

http://jvi.asm.org/

[image:5.503.54.252.487.572.2]

(6)

morepolymerases, ligase, etc.) aswellasthose

stepsuniquetoUV-damaged DNA,suchas

py-rimidine dimer excision. As shown in Table 2, representatives of three of fivecomplementation

groups synthesize viral DNA comparable to

HS27(CRL 1223, CRL 1199, CRL 1200)at48h p.i. and earlier times (data not shown). CRL

1162hadareducedlevel; however, incorporation intoviralDNA wasconsistent with several other

celllinesinthesameexperimentandappears to

reflectexperimental conditions rather than an

intrinsic defect inthe cellline (see Table7for comparison). CRL 1161,onthe otherhand,

con-sistently showed areduced incorporation rela-tivetoothercell lines tested in thesame

exper-iment under different conditions of infection (e.g., earlier timesp.i.and when confluent

cul-tures were infected and maintained in medium

putatively depletedforgrowthfactors).Further analysis of CRL1161andgeneticallyrelated cell lines indicated that the defect was mostlikely intrinsictothecellline andgeneticinnaturebut probably unrelated to the XP phenotype or a

defect in DNA strand elongation. In Table 7,

viral DNAsynthesisin CRL 1161 is compared with that ofanumber of additional phenotypi-callyorgenotypically related cell lines. Whereas

CRL 1161 and CRL 1166, derived from an

af-fected(XP)sibling,showdiminishedviralDNA synthesis,twootherrepresentatives of thesame

complementationgroup (CRL 1170, CRL 1204)

have considerably higher levels. Both parents

(CRL 1165 and CRL 1167) also show higher levels. DNA strandelongation wasdetermined

TABLE 7. SynthesisofSV40 DNA in human fibroblasts relatedtoCRL1161a

cpm ofDNA perculturec Celllineb

Infected Infected-um-_fecteddun

CRL1161 6,420 2,910

CRL 1165 8,720 6,500

CRL 1166 3,055 1,720

CRL 1167 6,945 4,145

CRL1170 12,270 11,050

CRL 1204 11,235 9,865

aConfluent monolayers were infected or the

me-dium waschanged without infection (uninfected) as describedinTable2.

b

Cell

linesandtheir genetic relatednessare as listed

inTable 1.

'Viral DNA in the Hirt

supernatant

was quanti-tated asdescribed in Table2.

dViral

DNA

was corrected by subtraction of

com-parableNSG fractionsobtained fromparallelextracts from uninfected cells. Trichloroacetic acid-precipi-tatedcountsperminute in the totalcellextract aver-aged2.9 x 105 for infectedcultures and 2.5 x 105for uninfected cultures.

in CRL 1161 and CRL 1166 p.i. withtsA30 at

330C

and shiftto

400C

(Fig. 3). Theresultsare

similarto those in Fig. 1 formonkeycells and

HS27. Itshould be noted that because of thelow

level ofviral DNAbeing measured,it was

nec-essarytopurifyfurther theappropriatefractions

containing DNA I to free it of contaminating

cellular DNA. Therefore, the data have been

corrected for recovery of added

["4C]DNA

I.

Variabilitywasalsoobservedamongtheseveral

experiments performed. It was not possible to

determine rates ofreentryofviralDNA due to

insufficient incorporation. The results suggest

that CRL 1161 and CRL 1166 arelikely to be examples of a reduced efficiency of infection

relatedtoviruspenetrationoruncoatingorboth

aspreviously described (1). Consistent withthis

interpretation,we haveobserved reduced

accu-mulation of Tantigen bycomplementfixation. DISCUSSION

SV40 infection of human fibroblasts has been foundtoresult inreadily detectable replication

ofviral DNA andproduction of infectious prog-eny.We determined therateofviralDNAstrand elongation directly and initiation indirectly through kinetics ofreentryinanormal fibroblast and found no significant difference from simi-larly infected permissivemonkeycells. Compa-rablestrandelongationrateswerealso observed in two cell strains derived from patients with XP. Similarresultswereobtained under condi-tions of varied effectiveMOIs (compareFig.1B, and3AandC). Thesemeasurementsshould best be considered approximations. Our analysis of strandelongationassumesthat the periodbeing measured ispredominantlyareflection of events throughout strandelongation ratherthanthose

at a single step. The distribution of viral repli-cativeintermediates has beenreported as

non-0 20 40 0 20 40 0

MINUTES AT 40.5*C

15

7

10;

[image:6.503.59.249.438.534.2]

5

FIG. 3. Viral DNAstrand elongationinXP

fibro-blasts.Cultureswereinfected with tsA30at33°Cand analyzedasinFig. 1. Viral DNA Iwasquantitated

after Hirt extraction andNSG and cesium chloride-ethidium bromide equilibrium sedimentations. (A) CRL1161 cellinfected for 4 daysatanMOI of20to 40; (B) CRL 1166 cell infected for 4 daysatanMOI

of20to40;(C) CRL 1161cellinfected for7daysatan

MOIof 1to5.

X. B C

0

;5

http://jvi.asm.org/

[image:6.503.262.454.483.570.2]

(7)

SV40

random with accumulate of late replicative

in-termediates inSV40-infected monkeycellsand

it is unclear to what extent that phenomenon

would distort the expected results (26, 31). It

should also bestressed thatallanalyses on DNA

synthesis disproportionately involve those

in-fected fibroblasts which most efficiently

repli-cateviralDNA since the contribution of others

in the incorporation of radioactive precursors

wouldlogically be insignificant by comparison.

Despite these limitations, the conclusion

ap-pearswarranted thatatleastasubpopulation of

cells in earlypassage line of human fibroblast canperform thestepsofSV40DNAreplication

asefficientlyasmonkeycells. This would imply that thecell factors responsible forpermissivity

thatoperate atthelevel of DNA initiation and

elongation are equivalent. Previous studies on

papovavirus DNA replicationinvitro had shown

that nonpermissive cells could provide one or morefactors (e.g., cytosol fractioncomponents)

whichfacilitatedDNAstrandelongation in

nu-clear or subnuclear systems (6, 8). It has not

been possible thus farto measure initiation of viralreplicons inacell-freesystem.

The basis for the semipermissivity of human fibroblasts appears to reside predominantly at an earlierstageof infection. It has been previ-ously reported that SV40 infection of human cellswaslessefficient than that ofmonkeycells,

asdetermined byaccumulation of Tantigenin

the nucleus(2-4, 21). Cell lines have been iden-tified with varied susceptibility for virus-medi-ated transformation and accumulation of T

an-tigenby immunofluorescenceassay (designated

assusceptible, normal,andresistant) (2).

How-ever, in all cases, there was a relatively low

percentageofpositivecells.

Even at ahigh MOI when the number ofT

antigen-positive cells reached maximal values, less than 30% of themosthighly susceptiblecell strain culture is T antigen positive. Others are

considerably reduced and have notyet shown evidence ofplateauvalues(0.2to

4%).

The

slopes

ofcurvesfor thepercentageofTantigen-positive

cellsas afunction ofinputof virusareconsistent with a single-hit event (2). Our data on

accu-mulation of Tantigen(s)at24hp.i. by

comple-ment fixation show a similar phenomenon. T

antigen levels are reducedas compared with6

monkey cellsat anequallylowtomoderateMOI and approachthe latterat ahighMOI, due to

already maximal levels in monkey cells at a

lower MOI. Wewould assume thatthe human

fibroblast HS27inourstudyis inthenormalor, more likely, susceptible category of Aaronson

and Todaro

(2),

whereas CRL 1161 cells arein

the resistantclass. It appears thattwo

phenom-ena are operative: a reduced efficiency of T

antigen expression and amixedcell population,

although the less likelypossibility that the latter

couldrepresent an extreme case of the former

hasnot been ruled out.

Thebasis for reduced T antigen levels appears

to be areflection of inefficient steps before or

during uncoating of virus. Aaronson (1) found that T antigen induction p.i. with viral DNA

(with DEAE-dextran as facilitator) was as

effi-cient in normal and susceptible fibroblasts.

HS74, whichshows reduced viral DNAsynthesis

orinfectiousvirus as compared with HS27, upon virion infection in this study is capable of highly

efficient viral DNA replication and T antigen

synthesis from intracellular SV40 DNA as we

have previously demonstrated in a clone

trans-formed byafragmentof theSV40genome

bear-ing the early region and the origin of DNA

replication (33). It should be noted, however, thatCarp and Sokol (5)saw norelative

improve-ment in the efficiency of T antigen induction with viral DNA in acomparison between pri-maryAfrican GMK cells and the human fibro-blast cell line W1-38; GMKwasapproximately 50-fold better with virionorDNA.

Therefore, it would appear most likely that semipermissive infection of human fibroblasts with virions containing nondefective viral

ge-nomesispredominantlythe summationof

nor-mal infection in a minority of the cells. Even these cells have alower efficiency of infection, such that the effective MOI being employed is moderately ormarkedly reduced as compared with that of GMK cells which were used to

titrate the virus preparation. ViralDNA

repli-cation andsubsequent eventsappear tofollow

in afashion consistent with those ofpermissive cells. Viral capsids (as V antigen assayed by complement fixation) and infectious virions

ac-cumulateat alevelroughly

proportional

tothe levelof viralDNAsynthesisathighinputMOI

asshown in thisstudy.Immunofluorescence as-saysforTand Vantigenhavepreviouslyshown concordance between the number of antigen-positivecellsinsome (2) butnotallstudies(4). In ourexperiments andthe study described in reference 2, conditions were used which have

been shown to minimize accumulation of viral

particles with defective genomes

(wild-type

strainSV-SatlowMOI).

As aninitialefforttowardthepossible

exploi-tationofhereditaryhumandisordersto

identify

hostfunctioninvolved inSV40-cellinteraction,

weinvestigatedtheefficiencyof viral DNA

syn-thesis in human fibroblasts derived from XP.

Representatives of multiple

complementation

groups involved in excisionrepair and a

http://jvi.asm.org/

(8)

iant"classpresumablydefective in postreplica-tion repair were included (29). The enzymatic bases for these defects inrepairof UV damage havenotbeenunambiguouslydetermined.

Cur-rentmodels ofrepair basedonprocaryotic sys-tems invoke in large part the involvement of multiple enzymatic functions which could be relevant insome cases tosemiconservative viral

DNAreplication aswell. The data thatwe ob-tainedfailtodemonstrateanydeficiencyin viral

DNA synthesis. In the tworelated cell strains which showed reduced viralsynthesisupon

pre-liminary testing (CRL 1161 and CRL 1166),

DNA strand elongation was not abnormal in

contrast tothe prediction if theputative

enzy-matic defect in the XPphenotypewere

respon-sible. Two incidental positivefindingsare that early passage fibroblasts from adult donors (CRL) support viral DNAsynthesis aswell as

doHS27 cells derived fromanewborn and that

accurateDNAstrandelongationmeasurements

can be perforned even under ratherpoor effi-ciency of infectionasinCRL 1161,emphasizing the possible utility of the approaches to other poorlypermissive virus-cellsystems.

It should be noted that this model for the

basis ofsemipermissive infectioninhuman cells need not apply to othersemipermissive SV40-cellinteractions;mostnotably, with rodentcells

asthose ofthe hamster (14, 22).Alternatively, SV40 infection of human cells has been reported in which extensive viral cytopathic effects are

observed (17). That SV40-cell interaction also hastheunusual property ofaveryhigh propor-tion ofdefective viral genomes,evenafter infec-tionatlow MOI (18).

ACKNOWLEDGMENTS

We thankourcolleaguesforprovidingcelllinesandthe ATCCCellRepositoryand theCell CultureLaboratory at the NavalBiochemicalResearch Laboratory, Oakland, Calif., un-dercontractE-73-2001-NO1withintheSpecial Virus-Cancer Program, National Institutes of Health. This investigation was supportedby Public Health Service research grant CA-23002 from the National Cancer Institute to H.L.O.M.L.S. was supportedby a Public Health Service research fellowship from theNational CancerInstitute.

LITERATURE CMD

1. Aaronson,S. A. 1970.Susceptibility of humancellstrains

totransformation by SV40 and SV40 DNA. J.Virol. 6: 470-475.

2.Aaronson, S. A., and G. J. Todaro. 1968. SV40 T antigen induction and transformation in human fibro-blastcellstrains.Virology 36:254-261.

3. Blattner,W.A., A. S.Lubiniecki,J. J.Mulvihill,P. Lalley, and J.F.Fraumeni, Jr. 1978. Genetics of SV40T-antigen expression: studies of twins, heritable syndromes,and cancerfamilies. Int. J. Cancer 22:231-238, 1978.

4. Carp,R.I., and R. V. Gilden. 1966. Acomparison of the replication cycles of SV40 in human diploid and African greenmonkey kidneycells.Virology 28:150-162.

5. Carp,R.I.,and F.Sokol. 1969. Further studiesonthe differences in the interaction ofSV40withAfrican green monkeykidneyandhumandiploidcells. J. Gen. Virol. 5:433-436.

6. DePamphilis, M. L., and P. Berg. 1975. Requirement of

acytoplasmicfraction forsynthesisofSV40DNA in isolated nuclei. J. Biol. Chem.250:4348-4354. 7.Dooley,D., R. Ryzlak, and H. L. Ozer. 1976.

Endonu-cleases andSV40 virions:originsofavirion-associated endonuclease. J. Virol. 17:352-362.

8.Francke, B., and T. Hunter. 1975. In vitro polyoma DNAsynthesis: requirementforcytoplasmic factors. J. Virol. 15:97-107.

9. Girardi,A.J.,F. C.Jensen,and H.Koprowski. 1965. SV40-induced transformation of human diploid cells: crisis andrecovery. J. Cell.Comp.Physiol.65:69-84. 10.Green,M.H.,andT. L.Brooks. 1978.Recently

repli-catedSV40 DNA isapreferential template for tran-scriptionandreplication.J.Virol.26:325-334.

11.Hand, R.,andJ. German. 1975. A retarded rate of DNA chaingrowth in Bloom's syndrome. Proc. Natl. Acad. Sci. U.S.A.72:758-762.

12.Hirt,B.1967.Selective extraction of polyoma DNA from infectedmousecellcultures. J. Mol. Biol.26:365-369.

13. Kidwell, W., R. Saral, R. G. Martin, and H. L. Ozer. 1972. Characterization ofanendonuclease associated withSV40 virions. J. Virol. 10:410-416.

14. Lavialle, C., H. G. Suarez, A. G.Morris,S.Estrade, J.Stevenet,andR.Cassingena. 1976.SV40-chinese hamsterkidney cell interaction. II. The semipermissiv-ity of the cell system. Arch. Virol. 50:137-146. 15. Levine, A. J., H. S.Kang, and F. E.Billheimer.1970.

DNAreplication inSV40 infected cells. I. Analysis of replicatingSV40 DNA. J. Mol. Biol. 50:549-568. 16.Lynch, H. T., D. E. Anderson, J. L. Smith, J. B.

Howell,and A. J.Krush. 1967.Xeroderma pigmen-tosum,malignantmelonoma,andcongenitalichthyosis. Arch.Dermatol. 96:625-635.

17. O'Neill, F. J. 1976.Propagation of SV40 inahuman cell line.Virology 72:287-289.

18.O'Neill, F. J., and D. Carroll. 1978. Appearance of defective SV40following infection of cultured human glioblastoma cells.Virology87:109-119.

19.Ozer, H. L. 1972. Synthesis andassembly of SV40. I. Differential synthesis of intact and emptyshells. J. Virol. 9:41-51.

20. Ozer, H.L., and K. K. Takemoto. 1969. Site of host restriction of Simian virus 40 mutants in an established African greenmonkeykidneycellline. J. Virol. 4:408-415.

21. Potter, C.E., A. M.Potter,and J. S. Oxford. 1970. Comparison oftransformationand T antigeninduction inhuman cell lines. J. Virol. 5:293-298.

22. Rakusanova, T.,W.P.Smales, J. C. Kaplan, and P. H. Black. 1979. Effects ofmitomycin C and 'Co y-irradiation on the replication of SV40 in cell lines of varyingpermissivity for SV40replication.J.Gen. Virol. 43:235-239.

23. Robbins, J. H., K. H.Kraemer,M. A.Lutzner,B. W. Festoff,and H.G.Coon. 1974. Xeroderma

pigmento-sum: aninherited disease with sun sensitivity, multiple cutaneousneoplasms, andabnormal DNA repair. Ann. Intern. Med. 80:221-248.

24. Roman, A. 1979. Kinetics of reentry of polyomaformI DNA into replication asafunction oftime postinfection. Virology 96:660-663.

25. Roman, A., and R. Dulbecco. 1975. Fate of polyoma formIDNAduringreplication. J. Virol. 16:70-74. 26. Seidman,M.M., and N. Salzman. 1979. Late replicative

intermediates are accumulated during SV40 DNA rep-licationinvivo and invitro. J.Virol. 30:600-609.

27. Seif,I., G.Khoury,andR.Dhar.1979.Thegenomeof

humanpapovavirus BKV.Cell 18:963-977.

http://jvi.asm.org/

(9)

VOL. 39, 1981 SV40 REPLICATION IN HUMAN FIBROBLASTS 489 28. Setlow,R. B. 1978.Repairdeficient human disorders and uousDNAreplication:accumulation ofSV40 DNA at cancer.Nature (London) 271:713-717. specific stages in itsreplication.J. Mol.Biol. 120:401-29. Slater, M. L, and H. L. Ozer. 1976. Temperature-sen- 422.

sitive mutantsofBalb/3T3 cells. II. Description of a 32. Tegtmeyer, P. 1972. SV40 DNA synthesis: the viral mutantaffected incellular and polyoma DNA synthesis. replicon. J. Virol. 10:591-598.

Cell 7:289-295. 33. Tooze, J. (ed.). 1980. The molecularbiology of tumor 30.Smith,H.S.,R. B.Owens,A. J.Huller,W. A. Nelson- viruses,2nd.ed.,part 2. ColdSpring Harbor Laboratory,

Rees,and J. 0. Johnston. 1976. The biology of human ColdSpring Harbor, N.Y.

cellsin tissueculture I. characterization of cells derived 34. Zouzias, D., K. K. Jha, C. Mulder, C. Basilico, and H. fromosteogenic sarcomas. Int. J. Cancer 17:219-234. L. Ozer. 1980. Transformation of human fibroblasts by 31.Tapper, D.P.,and M.L. DePamphilis. 1978. Discontin- SV40 DNA. Virology 104:439-453.

http://jvi.asm.org/