Replication of human cytomegalovirus DNA: lack of dependence on cell DNA synthesis.

(1)

Replication of

Human

Cytomegalovirus DNA: Lack of

Dependence

on

Cell

DNA

Synthesis

JEAN M. DEMARCHI AND ALBERT S. KAPLAN*

DepartmentofMicrobiology, Vanderbilt University Schoolof Medicine, Nashville, Tennessee 37232

Received for publication29December1975

Human cytomegalovirus (CMV)DNA synthesiswasstudied in5-fluorouracil

(FU)-treated and untreated human embryonic lung cells, which differ greatly with respectto the number of cells in the culture synthesizing cellular DNA.

CMV DNAsynthesis proceededatthesamerateinFU-treated andinuntreated

cells. CMV infection also reversed the inhibitory effects ofFU and activated cellular DNA synthesisinsomeof the cellsintheFU-treatedculture.

Autoradi-ographic studies showed that more than 20% of the cells in the infected

FU-treated culture synthesized viral DNA when less than 1% had synthesized

cellularDNA, indicating that the synthesis of viral macromolecules proceeds in cells that do not synthesize cellular DNA from the time of infection, and that viral DNA synthesis proceeds independently of the host cell DNA synthesis. Combined autoradiographic and immunofluorescence studies of both the FU-treated and unFU-treated infected cells showedthat, whereas 20% of the cells in the

cultures synthesizeviral DNA and viral antigens, only about 3 to 6%of those

cells thatsynthesize cellular DNA also synthesize viral antigen. Thus, produc-tive infection wasdelayed or inhibited in those cells that werestimulated by

CMVinfectiontosynthesize cellular DNA.

Human cytomegalovirus (CMV)caninfecta varietyof celltypes,butproductiveinfection is

limitedtohumanfibroblasts (22). Both

produc-tive andnonproductive CMV infection can

re-sultin astimulation ofcellularDNAsynthesis

(7, 16, 17; J. DeMarchiand A. Kaplan,

unpub-lishedobservations), and it has been proposed

thatacorrelation existsbetweenthesynthesis

ofcellularDNA and thatof CMV DNA (S. C.

St. Jeor, R. Hutt, andF. Rapp, Abstr. Annu.

Meet. Am. Soc. Microbiol. 1975,

S153,

p. 239).

The ability to stimulate cellular DNA after

infection has been documented for oncogenic

DNAviruses, inparticularsimian virus 40and

polyoma virus (5,8). Withthese viruses,

repli-cationof viral DNAisrelatedto

specific

events

of thehost cell

cycle

andoccurs

only

after the

cells have entered S

phase

(2, 3, 14, 19). A

similar

dependence

oncellular functions would

be unusual forCMV, since it contains a

large

genome

(-108

daltons) and can, in

principle,

code for all the proteins necessary for replica-tion. However, another herpesvirus, equine abortionvirus, which hasagenome of

approxi-mately the same size, is dependent on the S

phase when grownincertaincelllines (13). The experimentsdescribedinthis paper deal

with the synthesis ofCMV DNA in cells that

have been treated with5-fluorouracil

(FU)

and

thymidine (TdR), a

procedure

that suppresses

theirability to synthesize cellular DNA (12) but does not affect viral replication. The aim of

these experiments was to determine whether

CMVstimulatescellularDNAsynthesisin

FU-treatedcells and whether CMV DNA synthesis occurspreferentiallyinthecellsinwhich cellu-lar DNA synthesis has beenactivated.

MATERIALS AND METHODS

Cell culture and virus. Primary cultures of

hu-man embryonic lung (HEL) cells were purchased from Flow Laboratories and, after reaching con-fluency, were passaged further according to Hay-flick and Moorhead (9). The cells were grownin

90-mmglass petri dishes and maintained at 37 Cina humidified atmosphere of 5%CO2. Growthmedium consisted of Eagle minimal essential medium (MEM) containing 0.2% NaHCO3, 100Uof penicil-lin, 100 ,ug of streptomycin, and 50 ug of chlortetra-cycline per ml, with 5% fetal bovine serum. For maintenance ofcultures and for the experiments, theserum inthemediumwasreducedto 2%.

Stockcultures of the Davis strain of human CMV

(20) were grown in HEL cells and stored in 10%

glycerol at -70C asinfected cellsuspensions. For

inoculationonto HEL monolayers, thecell

suspen-sions were firstdisruptedinaRaytheonsonic oscil-lator (1 min at maximum setting) andclarifiedby

centrifugation. The supernatantwasdiluted appro-priatelytogivethedesiredmultiplicityofinfection andwasusedasinoculum. After1hofadsorptionat 37C, the inoculumwasremoved andreplacedwith 1063

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

1064 DEMARCHI AND KAPLAN

medium. In allexperiments, the multiplicity of

in-fection used wasapproximately0.1 to0.2PFU/cell. With the virusstocksused, these conditions of infec-tion resulted in thebeststimulation of cellular DNA synthesis wehaveobserved.

Assay for infectious virus. The procedure of Wentworth and French (21) was used, except that adsorption was for 1h at 37 C and theplateswere

overlayed with MEM containing 1% agarose (Sea-Kem). Plaques werecounted2weeks later.

Chemicals andradiochemicals. Cesium chloride, optical grade, was purchased from the Harshaw Chemical Company; FU and TdR were purchased

from Calbiochem; 5-iodo-2'-deoxyuridine (IUdR)

was obtained from Schwarz Bioresearch. [Methyl-3H]TdR (specific activity, 55 Ci/mmol), [2-'4C]TdR (specific activity, 57 mCi/mmol), and

[5-3H]deoxycytidine ([3H]CdR; specific activity, 25 Ci/ mmol) were purchased fromSchwarz/Mann.

FU treatment ofcell cultures. Cultureswere

in-cubated with FU (10

jug/ml)

and TdR (1 ,ug/ml) (12) fora minimum of 16 h before inoculation. Control cultures were treated with TdR (1

jug/ml).

FU and TdR were maintained in the treated cultures, and TdR was maintained in thecontrol cultures, during the course of experiments. TdR wasomitted when radiolabeled TdRwasadded.

Analysis of radiolabeled nucleotide incorpora-tion into DNA. CellsweresuspendedinRSB (0.01M

Tris-hydrochloride [pH7.4],0.01 MKCl, and0.0015 MMgCl2) and lysed by the addition of sodium dode-cyl sulfate (SDS) (final concentration, 1%). Portions

were precipitated with 7% trichloroacetic acid at

0 C. The precipitates were collected on 0.45-,um pore size membrane filters (Millipore Corp.) and washed with cold 7% trichloroacetic acid. After drying, the filters were counted in a Packard liquid scintillation spectrometer. The amount of label in-corporated into viral and cellular DNA was deter-mined after isopycnic banding, as described previ-ously (11). Alkali stability of labeled DNA was de-termined bytreatment with 1 NNaOH for 16 h at

37C, according to the method of Schmidt and Thannhauser (15). The samples were then neutral-ized and precipitated as described above.

Antisera. Human convalescent serum containing antibodies to CMV, human CMV-negative serum, and fluorescein isothiocyanate-conjugated rabbit antiserum to humanglobulin (immunoglobulin G)

weresuppliedcourtesy of Thomas D. Flanagan, Erie County Virology Laboratory, Buffalo, N.Y. The CMV-positive serumwas reactive with early CMV antigen (as determined with cytosine arabinoside-treated, CMV-infected cells) at a dilution of 1:320. It reacted with late antigen (CMV-infected cells at 5 days postinfection) at a dilution of 1:5,120. It did not react withuninfectedcells at a dilution of 1:10. In all experiments described in this paper, the serum was usedatadilution of1:20.

Autoradiographyand immunofluorescence. HEL cells grownoncoverslips in 90-mm petri dishes were either infected or mock-infected and labeled with

[PH]TdR (1 ,uCi/ml). At the end of the labeling

pe-riod, the cover slips were rinsed in phosphate-buffered saline (PBS) (4), fixed with cold acetone (3

J. VIROL.

min),air-dried, and mountedontoglass slides(cell sideup). The slides were coated with Ilford Nuclear Research emulsion (typeL-4) diluted withanequal amount ofdistilledwaterand stored in adesiccator underpartial vacuum at 4C. Afteranappropriate exposure time, the autoradiograms weredeveloped with Kodak D-19 developer, stained with 0.03% aqueous methylene blue, and viewed by light mi-croscopy.

When immunofluorescence was combined with autoradiography, the fixedcoverslip cultureswere

first incubated (30min, 37 C) with CMV antiserum diluted in PBS, then rinsed in PBS (10 min, room

temperature), and incubated in the same manner

withconjugated antiserum. Afterafinal PBS rinse (30min), thecoverslipswereimmersed inwater(2 min) and then in 95% ethanol (20 min) to fix the conjugate(18). The cover slipswerethencoated with emulsion. Afteran appropriate exposure timethey weredeveloped and mounted (cell side down) witha

solution of90%glycerol,10%PBS. Thepreparations wereviewed witha Zeissmicroscopeequipped with

adark fieldcondenser, UV light source, and a 470-nmbarrier filter. Fluorescence wasobserved witha

BG-12 exciter filter, and autoradiographic grains were viewed witha combination of blue glass and NG-9 (neutralgray) filters.

RESULTS

Incorporation of

[3H]TdR

into CMV-in-fected and mock-inCMV-in-fected FU-treated and un-treated cultures.Figure 1 shows the amount of

[3H]TdR that is incorporated into the DNA of

infected andmock-infected untreated and

FU-treatedHELcells. In untreatedcells, a

stimu-lation of incorporation of [3H]TdR into DNA wasdetectable both atearlytimes(2days)and late times (4and5days) postinfection (Fig. 1A). The stimulation of[3H]TdRincorporation into

DNAbyCMV was much moremarkedin cells

inwhich cellularDNAsynthesishadbeen

sup-pressed due to pretreatment with FU (Fig. 1B). By 4and 5 days postinfection, FU-treated and untreated cellsincorporatedapproximately the same amountof[3H]TdR.

Todetermine the type of DNA (viral or cellu-lar) synthesized by FU-treated and untreated cells, the DNAfrom the infected cultures that hadbeenincubated with [3:H]TdRbetween 24 to

48 h and 72 to 96h after infection was

centri-fugedinCsCl density gradients. (Human CMV DNA has a buoyant density of 1.716 g/cm3 in CsCl [10] and can readily be separatedby this method from cellular DNA, which has adensity of1.699g/cm3.) Figure 2 shows that most ofthe

[3H]TdR taken upby the cells from 24 to 48 h after infection in both FU-treated and

un-treated cells was incorporated into cellular DNA.Between 72 and 96 h after infection, how-ever, [3H]TdR was incorporated in large part into viral DNA.

on November 10, 2019 by guest

http://jvi.asm.org/

(3)

B

12

A 8,,

4

2 3 4 5 2 3 4 D

TIME AFTER INFECTION (DAYS)

FIG. 1. Incorporation of[PH]TdR intoDNA ofFU-treatedanduntreatedcells.FU-treated and untreated cultures of HEL cells were either mock-infectedor infected with CMV (approximately 0.1 PFUlcell). The

cultures were incubated with[:'H]TdR (10 pfCilml) forsequential 24-hperiodspostinfection. At theendof eachlabeling period,the culture fluidswereremoved and the cellswereharvestedinRSBplus1%SDS,as described in Materials and Methods. Each point in the figure represents the amount of radioactivity incorporated by the cells incubatedwith label for24 hpriortothe indicated day of harvest. Symbols: (0)

CMV-infected cultures;(0)mock-infectedcultures. (A) Untreated; (B)FU-treated.

UNTREATED FU-TREATED

30-3,0

20- 200

10 L

10

0~~~~~~~~~~~~~~~~~~~~

o

12 UNTREATED

12-72-96h 72 -96h 2

8-1

0.8

2 3 4 2 3 4

ML

FIG. 2. CsCl density gradient analysis of DNA synthesized after CMV infection in FU-treated and

untreated cultures. PortionsofCMV-infected,FU-treatedand untreatedcultures, harvested at 2 and 4 days

after infection intheexperimentpresentedinFig. 1,wereanalyzedin CsClbuoyant density gradients. The

samples analyzedinthegradientswerenotnecessarilyequalin volume. '4C-labeledcellularDNAwasadded

as amarker.

1065

a

16

K 12

o0

8

4 lol

2

Q-u

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.503.133.369.48.195.2] [image:3.503.112.396.278.617.2]

(4)

The amount of L3H]TdR incorporated into viral and cellular DNA in FU-treated and un-treated cellsatvarious times after infection is summarized in Table 1. The following points

emergefrom these experiments. (i) Under the

conditions used in these experiments, only a small amountof CMVDNA is synthesized be-fore 48 hpostinfection andisdetectableonly in FU-treated infected cultures. Labeled viral

DNA was not detected at that time in

un-treated, infected cells, probably because of the high background of the cellular DNA synthe-sized. (ii) The amountof[3H]TdRincorporated into viral DNA in FU-treated and untreated, infected cultures was approximately the same

throughout infection. (iii) Infection with CMV partially reversed the FU-induced inhibition of TdR incorporation into cellular DNA. The

amountof radioactivity incorporated into cellu-lar DNA at 2, 3, and 4 days after inoculation

was approximately threefold greater than the amount incorporated into the DNA of mock-infected cultures. (iv) Infection of untreated cells stimulated incorporation of TdR into cellu-lar DNAatearlystagesof infection (24to48h). This stimulation, though slight, is significant andwasobservedrepeatedly.

Activation of cellular DNA synthesis by CMV infection in FU-treated cells. Table 1

shows thatincorporation of[3H]TdR into cellu-lar DNA is stimulated after infection of FU-treated cells. That these resultsare

reproduci-ble is shown by the data in Tareproduci-ble 2. In this experiment, infected and mock-infected FU-treated cultures were pulse-labeled at times when the background of label incorporated into

TABLE 1. [3H]TdR incorporationintoviral and cellular DNA inCMV-infectedcells'

Incorporation (counts/min) Labeling period Infected

(h)

Mock-in-Viral Cellular fected

FU-treated cells

0-24 0 1,359b 1,413

24-48 114 1,962 691

48-72 3,501 1,666 555

72-96 9,215 1,768 635

Untreated cells

0-24 0 17,467 17,983

24-48 0 16,819 14,629

48-72 2,972 6,436 10,036

72-96 8,908 3,162 5,924

aThe experiment wasperformed asdescribed for FU-treatedand untreated culturesinthe legendtoFig. 1.The amountofradioactivityincorporated into viral and cellular DNA,isolated byisopycnic centrifugation, wasdetermined

asdescribedinMaterials andMethods.

bAverageof duplicate samples.

J. VIROL.

TABLE 2. Incorporation of[1H]TdR into cellular DNA of CMV-infected and mock-infected HEL cells

treated withFU"'

Incorporation

(counts/min) Infected/

Exptno.

inod

(h)

mock-in-Mock-infected Infected fected

1 16-40 1,070b 436 2.5

16-64 1,837 619 3.0

2 16-40 957 338 2.8

16-64 2,065 800 2.6

aFU-treated cultureswerepreparedasdescribedin Ma-terials and Methods, inoculatedwith CMV(approximately 0.1PFU/cell),and incubatedfurther for 16 h in thepresence of FU.The cultures were then incubated with[3H]TdR(10

jCi/ml)and harvested 24 and 48 h later. The DNA was bandedinCsCldensity gradients,viral DNA wasseparated fromcellular DNA, and theamountofradioactivity associ-ated with each DNA was determined. In allcases, more than90% of theradioactivitywasassociatedwith cellular DNA.

bAverage of duplicate samples.

cellular DNA of uninfected cells would be as

low aspossible, and when appreciable amounts

of viralDNAwouldnotyetbesynthesized.The

results (Table 2) show

that,

between 16 to 40

and16to 64hafter

inoculation, nearly

a three-foldincrease intheamountof labeledprecursor

incorporated into DNA in FU-treated cells

could be detected after infection. An

experi-mentsimilartothe one

summarized

inTable2

showed that incorporation of[:H]CdRinto the

DNAof infectedFU-treated cells was also three

timesgreater than that into the DNAof unin-fected, FU-treated cells.

That the increase in TdR incorporation into

cellularDNAof infected cellsisrelatedtoCMV

infection was shown by two experiments. (i)

Uninfected cellextractsdid not stimulate

cellu-larDNAsynthesis. (ii) Antisera against CMV

prevented thevirus inoculum from stimulating

cellularDNAsynthesis (Table 3).

Theincreased incorporation of labeled

deoxy-ribonucleosides into cellular DNA may reflect

(i) enhanced utilization of TdRorCdRas a

re-sult ofincreased kinase activity, (ii) increased repair ofcellular DNA, or (iii) enhanced

syn-thesis of cellular DNA. The first possibility

canbe eliminatedby thedata presented below

(Table

4),

which show a threefold

increase,

after CMVinfection, inthe number of cells in FU-treated cultures in which cellular DNA

synthesisoccurred.To distinguish between the

two last possibilities, the cells were labeled

with [3H]CdR in medium containing IUdR. If

theincrease in incorporation into cellular DNA inFU-treated infected cells were due to repair ofpreexistingcellular DNA, most of the label

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.503.60.254.453.604.2]

(5)

HUMAN CMV DNA REPLICATION 1067

TABLE 3. Failure of serum-neutralized CMV to stimulate cellular DNA synthesis"

Activity (counts/min)with HEL cells antiserum:

CMV-negative CMV-positive

Mock-infected 1,101b 992

CMV-infected 4,121 1,052

a Human CMV-positive and CMV-negative sera wereincubated ata1:10 dilution with CMV or with MEM (30 min, 37 C) prior to inoculation of FU-treated HEL cells. Between 24 and 48 h after inoc-ulation, the cells were incubated with medium con-taining [3H]TdR (10 ,uCi/ml). The cells were then lysed with SDS (1%), the DNA was banded in CsCl density gradients, and the radioactivity incorpo-rated into cellular DNA was determined.

bAverage of duplicate samples.

would be associated with a peak cosedimenting with marker 14C-labeled cellular DNA in a

CsCl density gradient. If, on the other hand,

the increasedincorporation is due to a

stimula-tionof cellularDNA synthesis, thelabel should

be associated with DNAbanding at a position

of higherbuoyant densityas aresultof

substi-tutionof IUdR for TdRinoneof the strands.

Figure 3 shows that practically all the label

incorporated into DNA of both infected and

mock-infectedcells was found withinthe heavy

peak of IUdR-substituted DNA. Figure 3C

il-lustrates the results obtained when IUdRwas

omitted and serves as a control to show that,

under the labeling conditions used,

incorpora-tion ofCdR by the infected cells was mainly

into cellular DNA. Thus, infection with CMV

inFU-treated cells stimulates the synthesis of

cellular DNA. These results confirm the

find-ings of St. Jeor et al. (17), who showed that

stimulation of cellular DNA

synthesis

occurs

after CMV infection ofpermissive cultures in

which DNA synthesis was suppressed by pre-treatmentwithiododeoxyuridine.

Viral DNAsynthesis is

independent

of

cel-lular DNA synthesis. The experiments

de-scribed below were undertaken to determine

whether only those infected cells which had

enteredS

phase

afterinfection

synthesize

viral

DNA. Ifthiswerethecase,thenumber of cells in the infected cultures whichsynthesizeviral

DNA should not exceed the number of cells

whichsynthesize cellular DNA. The cells that

have synthesized cellular DNAandthose that

have synthesized viral DNA can be easily

dif-ferentiated from each other. A cell that has

synthesized cellular DNA

during

a 24-h

label-ing period incorporates much more [3H]TdR

than does a cell that has

synthesized

viral

DNA. By autoradiography, one can therefore detect, after a short exposure time, only those cells in the culture which have synthesized cel-lular DNA. The nuclei of these cells are covered evenly with grains and look identical to the nuclei of cells that have synthesized DNA in uninfected cultures labeled under identical con-ditions. The cells that have synthesized viral DNA can be detected by autoradiography only

afterlongerexposure times.These cellscontain

inclusion-like bodies covered sparsely with grains(after a time of exposurethatcompletely blackens the emulsion over cellsthat have

syn-thesized cellularDNA). These cells also contain

viral antigens. Thedifferencebetweenthe

ap-pearance ofthe cells thatsynthesized cellular DNA and those that synthesized viral DNA is

illustrated in Fig. 4A.

Using the criteria described above, we

de-teminedthe numberof cells synthesizing

cellu-lar or viral DNA at various times after infec-tion. The results are summarized in Table 4.

The following conclusions can be drawn from

this experiment. (i)Thenumber of cells

synthe-sizingcellularDNA isincreasedafter infection

ofboth FU-treated and untreated cells. During

four successive 24-h labelingperiods, an aver-age of 0.1% of the nuclei of uninfected

FU-treated cells synthesized cellular DNA. This

increased to an average of 0.5% inthe infected

cells. Inuntreatedcells, approximately 26% of

the infected cells, as opposed to 20% of the

mock-infected cells, synthesized cellular DNA

between 24 and 48 h postinfection. This

in-crease, though small, is significant and has

TABLE 4. Percentageof nuclei thatincorporated

[3H]TdRinto viral and cellular DNA inFU-treated anduntreatedcellsa

%Nuclei incorporating[3H]TdRin:

Labeling period Infected

Mock-in-(h) Mock_____in___

Viral Cellular fected FU-treated cells

0-24 0 0.2b 0.1

24-48 5.3 0.7 0.2

48-72 15.0 0.6 0.1

72-96 20.2 0.5 0.1

Untreated cells

24-48 1.5 25.9 20.1

72-96 20.3 9.9 14.8

a FU-treated and untreated cells (see Materials and Methods) grown on cover slips were infected with CMV (approximately0.2PFU/cell).At various timesafter infec-tion,the cultureswereincubatedinthepresenceof[3H]TdR

(1 uCi/ml). At theend of eachlabelingperiod, thecover

slipswerefixed andpreparedforautoradiography, as de-scribedinMaterials and Methods.

bPercentage values were calculated aftercounting at

least 1,500 nuclei.

VOL. 18, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.503.254.449.461.590.2]

(6)

I

a-u

(-)

I?

[image:6.503.142.376.73.456.2]

ML

FIG. 3. CellularDNA synthesis in CMV-infected FU-treated cultures. FU-treated cultures were either mock-infectedorinfected with CMV(approximately0.1PFUlcell).Between16and64 hafter infection, the cultures were incubated with medium containing[:3H]CdR(10

,uCiIml)

and IUdR(100

mg/ml).

IUdR was

omittedfroma thirdculture ofFU-treatedinfected cells. At64h, the cultureswereharvestedinRSBplus1% SDS. Equal volumes from eachof the cultures wereanalyzed in CsCldensitygradients, with '4C-labeled cellular DNA of normaldensity added as marker. (A) Mock-infected cellsplusIUdR;(B)CMV-infectedcells plus IUdR; (C) CMV-infected cells, no IUdR.

been observed repeatedly. (ii) The number of

cells synthesizing viral DNA is similarin

FU-treated and unFU-treatedcultures, confirming the results obtained by CsCl density gradient anal-ysis (see Table 1). The percentage of nuclei

bearinggrainsassociated withintranuclear

in-clusions inFU-treated cultures increased

dur-ing the course of the experiment (Table 4).

Whereas a small number of cells in

CMV-in-fected, FU-treatedcultures synthesized cellular

DNA afterinfection,many morecellswereable

to synthesize viral DNA.

Furthermore,

the

number ofcells synthesizing viral DNA was

thesameinFU-treated and untreated infected cultures, despite a great difference in the number of cellssynthesizing cellular DNA. We conclude, therefore, that viral DNA synthesis does not require prior synthesis of cellular DNA.

Whereas cellular DNA synthesis does not

seem tobe required forviral DNAsynthesis,it seemedpossible that viral growth mightoccur

preferentially in those cells that synthesize cel-lular DNA. To ascertain whether this is the case, we used atechniquedescribed by Tennant

etal. (18),by which the same cell can be exam-J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

(7)

HUMAN CMV DNA REPLICATION 1069

inedby bothautoradiographyand immunofluo-rescence. By meansof thistechnique, as

illus-trated in Fig. 4, we found that viral antigen

synthesis is less likely to occur in the cells in

which cellular DNA was synthesized after

in-fection. Table 5summarizes the resultsof this

experiment. In the FU-treated infected mono-layer, 0.7% of the cells incorporated [3H]TdR into cellular DNA, and 23% synthesized viral antigen. However, only0.02% of the cells

(ap-proximately 3% ofthose that had synthesized

cellular DNA) synthesized both viral antigen

and cellular DNA,indicating that cells

synthe-sizing cellular DNA were less, rather than more, likely to synthesize viral antigen.

Simi-larly,inuntreatedcultures, whereas 23% of the

cellssynthesized cellularDNAand24%

synthe-sized viralantigen, 1.4%ofthecells (about 6%

of those that had synthesized cellular DNA)

also synthesized viral antigen. These findings

support the conclusion, based on the experi-mentsdescribedabove, that viral DNA synthe-sis is notdependent on the Sphase of the cell

growthcycle.

DISCUSSION

CMVinfection resultsinanactivationof

cel-lular DNA synthesis in some of the cells in

culturesin which DNAsynthesis has been

ar-rested bytreatment with FU andTdR. This is

reflectedby a threefold increase in the amount

of[3H]TdR incorporated into cellular DNA, as

determined by equilibrium centrifugation in

CsCl, and a similar increase in the number of

cells synthesizingcellular DNA after infection,

asdetermined byautoradiography. A

stimula-tion ofcellular DNA synthesis after infection of

untreatedcells also wasdetected. This

stimula-tion, though slight, was significant and was represented by an increase in the amount of

TdRincorporatedinto cellular DNA, as well as

byanincrease in the number of cells

synthesiz-ing cellular DNA. Activation ofcellular DNA

synthesiswasdetectableinuntreated cells only

during thefirst48 hpostinfection.

Although cellular DNA synthesis is acti-vated after infection, this activation is

appar-entlynotessential for viral DNA orviral

anti-gensynthesis. Less than 1%of the cellsin

FU-treated cultures synthesized cellular DNA dur-ing the first 48 hafterinoculation. However, a muchgreater number of cellssynthesizedviral DNA and viral antigens during this period of

time (Tables4and5). Furthermore, viral

anti-gensynthesisappears tobedelayedorinhibited

in thosecells that do synthesize cellularDNA

after infection. In both FU-treated and

un-treated cultures, 23% of the total number of

cells in the culture, but only 3 to 6% of those

cells that had synthesized cellular DNA from

FIG. 4. Absence of viralantigenproductionincells whichsynthesize cellularDNA.FU-treatedculturesof

HELcellson coverslipswerelabeled with[3H]TdR (1

/CiIml)

afterCMVinfection.At 48h,thecoverslips

werefixedinacetone, stainedby indirectimmunofluorescence,andcoatedwithphotographicemulsion. (A) Preparations viewed with neutralgrayandblueglass filters.Autoradiographicgrainsareassociatedmainly

with cellular chromatin (arrows). (B) Same cells viewed with a BG-12 exciter filter, which allows the visualizationofthefluorescence.Note that the cellswhich hadincorporatedlabel into cellular DNA did not fluoresce.

VOL. 18, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.503.59.449.389.590.2]

(8)

[image:8.503.60.250.75.200.2]

TABLE 5. Percentageof nuclei thatincorporated [3H]TdR intocellularDNAand exhibited

CMV-specificfluorescenceafter infection"

%Nucleiincorporating [3H]TdRin: Mock-in- CMV-infected

fected Prepn

Cellular Cellular Cellular Viral DNA and

DNA DNA antigens viral an-tigens FU-treated 0.Job 0.71 23.42 0.02 Untreated 20.60 23.48 24.48 1.38

aFU-treated and untreated cellsgrown on cover slips wereinfectedwithCMV(approximately0.2PFU/cell)and incubated for48hwith[3H]TdR(1 ,Ci/ml). After acetone fixation,thecoverslipswerestainedby indirect immuno-fluorescence and coated withphotographic emulsion, and the percentage ofcellsintheculturewhich hadsynthesized cellular DNA, hadsynthesized viral antigens,orhad syn-thesized bothwasdetermined.

IPercentage values were calculated after counting at

least7,000nuclei.

the time of infection, synthesized viral anti-gens. Since the antiserumusedinthese experi-ments detects both early- and late-type anti-gens (see Materials and Methods), it is clear that most viral functions were lesslikely to be expressed in those cells that had synthesized cellular DNA after infection than in cells in

which stimulation of cellular DNA synthesis was detected. Since the induction of cellular DNA synthesis is a direct result of infection with the virus (see Table 3), the interesting

possibility arisesthat onlyafew specificviral

functions are expressed in the cells that are

stimulated to synthesize cellular DNA. This

finding may have some bearing on the

trans-forming potential of CMV (1). Therefore, the

elucidation of the conditions that govern the activationofcellularDNAsynthesisincertain cells and those that allow for the synthesis of viral antigens in others isofconsiderable inter-est.

ACKNOWLEDGMENTS

Thiswork wassupported byPublicHealth Service grant AI-10947from the NationalInstituteofAllergyand Infec-tiousDiseases. J.M.DeMarchiwassupportedby a fellow-shipfrom theNational Instituteof Allergy andInfectious Diseases(5 F02 AI 55816).

LITERATURE CITED

1. Albrecht,T.,andF.Rapp.1973.Malignant transforma-tionofhamster embryo fibroblasts following exposure toultraviolet-irradiated human cytomegalovirus. Vi-rology55:53-61.

2. Ben-Porat, T., and A. S. Kaplan. 1967. Correlation betweenreplicationanddegradationofcellularDNA

inpolyomavirus-infected cells. Virology 32:457-464. 3. Ben-Porat,T.,A.S. Kaplan,andR.W.Tennant.1967. Effectof5-fluorouracilonthemultiplication ofa viru-lentvirus(pseudorabies) andanoncogenicvirus (pol-yoma).Virology 32:445-456.

4. Dulbecco, R.,andM.Vogt.1954.Plaque formationand isolation ofpure lines with poliomyelitisviruses. J. Exp.Med. 99:167-182.

5. Dulbecco, R.,L.H.Hartwell, and M.Vogt. 1965. In-duction ofcellular DNAsynthesisby polyomavirus. Proc. Natl. Acad.Sci. U.S.A. 53:403-410.

6. Fioretti,A., T. Furukawa,D.Santoli,and S. A. Plot-kin.1973.Nonproductiveinfection ofguinea pigcells with humancytomegalovirus. J. Virol. 11:998-1003. 7. Furukawa, T., S. Tanaka, and S. A. Plotkin. 1975. Stimulationofmacromolecular synthesisin guinea pigcells by human CMV. Proc.Soc.Exp.Biol. Med. 148:211-214.

8. Gershon, D., L. Sachs, and E. Winocour. 1966. The induction of cellularDNAsynthesisbysimian virus 40 incontact-inhibited and X-irradiated cells. Proc. Natl.Acad.Sci. U.S.A. 56:918-925.

9. Hayflick, L., and P. S. Moorhead. 1961. The serial cultivationofhuman diploid cellstrains. Exp. Cell Res. 25:585-621.

10. Huang, E-S., S-T. Chen, andJ.S. Pagano. 1973. Hu-mancytomegalovirus. I. Purification and characteri-zationofviralDNA. J.Virol. 12:1473-1481. 11. Kaplan, A. S. 1969. Isopycnicbanding of viral DNAin

cesiumchloride, p. 487-495. In K.Habel andN. P. Salzman(ed.), Fundamental techniquesinvirology. AcademicPress Inc., New York.

12. Kaplan,A.S., and T. Ben-Porat. 1961.Theactionof 5-fluorouracilonthenucleic acid metabolismof pseu-dorabiesvirus-infected and noninfected rabbitkidney cells. Virology 13:78-92.

13. Lawrence, W. C. 1971. Evidence forarelationship be-tween equine abortion (herpes) virus deoxyribonu-cleic acid synthesis andthe S phase of theKBcell mitoticcycle.J.Virol. 7:736-748.

14. Pages, J.,S. Manteuil, D. Stehelin, M. Fiszman,M. Marx, and M. Girard. 1973. Relationship between replication of simian virus 40 DNA and specific eventsof the host cellcycle. J. Virol. 12:99-107. 15. Schmidt, G., and S. J. Thannhauser.1945.Amethodfor

the determinationofdesoxyribonucleic acid, ribonu-cleicacid and phosphoproteins inanimaltissues. J. Biol. Chem. 161:83-89.

16. St. Jeor, S., andF.Rapp. 1973. Cytomegalovirus: con-versionofnonpermissive cellstoapermissivestate forvirusreplication. Science 181:1060-1061. 17. St.Jeor, S. C., T. B. Albrecht, F. D. Funk, and F.

Rapp.1974.StimulationofcellularDNA synthesis by humancytomegalovirus.J.Virol. 13:353-362. 18. Tennant,R. W., K. R.Layman, andR. E. Hand.1969.

Effect of cell physiologicalstate on infection byrat virus.J.Virol. 4:872-878.

19. Thorne, H. V.1973.Cyclic variation insusceptibilityof Balb-c 3T3 cells to polyoma virus. J. Gen. Virol. 18:163-169.

20. Weller, T. H., J. C. Macauley, J. M. Craig, and P. Wirth. 1957. Isolation ofintranuclear inclusion pro-ducingagentsfrom infantswith illnessesresembling cytomegalicinclusiondisease. Proc. Soc. Exp. Biol. Med.94:4-12.

21. Wentworth, B. B., and L. French. 1970.Plaque assay ofcytomegalovirus strains of human origin. Proc. Soc.Exp.Biol. Med. 135:253-258.

22. Wright, H. T.1973. Cytomegaloviruses, p. 353-388. In A. S. Kaplan (ed.), The herpesviruses. Academic PressInc., New York.

J. VIROL.