Vaccinia Virus Replication I. Requirement for the Host-Cell Nucleus

(1)

0022-538X/79/02-0705/11$02.00/0

Vaccinia

Virus

Replication

I. Requirement for the

Host-Cell Nucleus

DENNIS E. HRUBY,l*LINDA A. GUARINO,2 ANDJOSEPH R. KATES'

DepartmentsofMicrobiology'andPharmacology,2HealthSciences Center, State University of New York,

Stony Brook,New York 11794

Received for publication 29 August 1978

Using cytochalasin B-induced enucleation techniques,weexamined the ability ofvaccinia virus toreplicate in the absence ofthe host-cellnucleus in several

mammaliancell lines.Itwasfound that virus-infected enucleatedcells(cytoplasts) prepared fromBSC-40, CVC, and Lcellswereincapable ofproducing infectious

progeny virus. The natureof this apparent nuclearinvolvementwas studied in

detail in BSC-40 cells. Modulations designedtomaximize cytoplastintegrity and longevity, suchasreductionof thegrowth temperatureand initialmultiplicityof

infection, did notimprove virus growth in cytoplasts. Sodium dodecyl

sulfate-polyacrylamide gel analysis of the[35S]methionine pulse-labeledproteins

synthe-sized in vaccinia virus-infected cytoplasts demonstratedthat bothearly and late

viral gene products were being expressed at high levels and with the proper

temporalsequence. Vaccinia virus cytoplasmic DNA synthesis, asmeasured by [3H]thymidine incorporation, peaked at 3hpostinfection andwas 70 to90% of control levels in cytoplasts. However, in the cytoplasts this DNA was not

convertedto aDNase-resistant form late in infection, whichwasconsistent with the failure to isolate physical particles from infected cytoplasts. Treatment of vaccinia virus-infectedcells with100

,jg

of rifampin/ml from0 to 8h toincrease thepools of viralprecursors,followed by subsequent removal of the drug, resulted inathreefold increase virusyield.This treatmenthadnoeffectonvirus-infected cytoplasts. Finally, vaccinia virus morphogenesis wasstudiedunderanelectron microscope in thin sections of virus-infectedcellsandcytoplasts which hadbeen

preparedatvarious timesduringasingle-step growthcycle. Itwasapparentthat, although early virus morphogenetic forms appeared, there was no subsequent DNAcondensationorparticle maturation in thecytoplasts. These resultssuggest

that vaccinia virus requiressomefactororfunction from thehost-cellnucleus in ordertomatureproperly and produce infectiousprogenyvirus.

Vacciniavirus,alargecomplex DNA-contain-ingpoxvirus, has beenthoughttoreplicatesolely in the cytoplasm ofsusceptible host cells (23). This ideawasbasedonseveral lines of evidence: vaccinia virus containsorcodes formany ofthe

enzymes necessary for DNA synthesis,

tran-scription, and modification (18); virus replicative factoriesarelocatedexclusivelyinthecytoplasm (11); and DNAsynthesis,asmeasuredby

auto-radiography, occurs in vaccinia virus-infected enucleated L cells (24). However,other

experi-ments have shown that pretreatment of host

cellswith eithermitomycin CoractinomycinD

toinhibit hostnuclear functionscauses amarked

diminutioninsubsequentvaccinia virusgrowth

(15, 26). Also, there have been indications that some virus-specific DNA synthesis may occur

within the nucleus ofpoxvirus-infected cells(9,

13,29). PenningtonandFollett(22) studied the

replication ofvaccinia virus inenucleated

BSC-1 cells and found that the production of infec-tious progeny virus was completely blocked. Thiswasaccompanied byaconsiderable reduc-tion in the levels ofvirus-specificDNAsynthesis, protein synthesis, and mature particle forma-tion, and also in the size and number ofvirus

factories. Similar observations have been made concerningtheinabilityofaniridovirus,African swine fevervirus,toreplicatein Verocytoplasts (20). It should bepointedoutthatreplicationof cytoplasmic RNA viruses suchasvesicular

sto-matitus virus and SemlikiForestvirus is

essen-tiallynormal inenucleated cells(8).

With therecentevidencedemonstratingthat

anothergroupofcytoplasmicDNAviruses,the

icosahedralvirusessuchasfrogvirus3,require

thenucleus forviral DNAreplicationand

tran-scription(10),itwasof interesttoreexamine the

so-calledcytoplasmicmode ofpoxvirus

replica-tionand therelationshipofthe apparent nuclear

705

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

706

involvement. Intheexperiments reportedhere,

wehave carefullyexaminedtheability of

vacci-niavirustoreplicateinthe absence of the host-cellnucleus. Our resultsconfirmand extend the

previous observationthatvaccinia virus willnot

grow in enucleated cells. Furthermore,

experi-mental conditions have been defined under

which the levels of early and late viral gene

expression in enucleatedcellswerenearly

equiv-alent tothose in controlcells, yetno infectious

virus was produced. This apparently resulted

fromaberrantpackagingand maturation of virus

particles late in the infectious cycle, indicating

that some host-cell nuclear function was

re-quiredtoobtainpropervaccinia virus

morpho-genesis.

MATERIALS AND METHODS

Radioisotopes and chemicals. [3S]methionine

(586.4 Ci/mmol) and [methyl-3H]thymidine (55.6

Ci/mmol)werepurchasedfromNewEnglandNuclear

Corp. Cytosine arabinoside and rifampin were from

Sigma Chemical Co. Cytochalasin B was obtained

from Aldrich Chemical Co. All gel electrophoresis

reagentscamefrom Bio-Rad Laboratories.

Cells and virus. BSC-40 cells,obtained from M.

Ensinger,whoselectedforaderivative of BSC-1cells

which grew well at 400C, were maintained in Eagle

minimalessential medium(MEM) plus 10%

heat-in-activatedfetal calfserum (AFCS).Lcells(I. Tamm)

were grown in MEMplus 5% AFCS. CVC cells (P.

Tegtmeyer) weremaintained in basal Earle medium

(BME) plus10%AFCS.

Vacciniavirus,WRstrain, wasobtained from the

American Type CultureCollection and twiceplaque

purifiedonLcells.Purifiedvesicularstomatitisvirus,

Indianastrain,wasdonatedbyL.A. Ball.

Growth and purification of vaccinia virus.

Crude stocks of vaccinia virus were maintained by

low-multiplicitypassageonL cells.Crude virus stock

wastrysinizedwithanequalvolumeof 0.25%trypsin

(GIBCO) at37°Cfor15minwithfrequent shakingin

a Vortex mixer and then was diluted with ice-cold

adsorptionmedium(1part Puck'ssaline,1partMEM

plus5% AFCS). Confluentmonolayersof L cells (107

to3 x 107cells/100- by20-mmdish)wereinfected at

amultiplicityofinfection(MOI)of 0.01to0.1 at250C

for 3 h. The virus inoculumwasthenremoved,MEM

plus5%AFCS wasadded,and the disheswere

incu-batedat370Cfor 2to4days(oruntil themicroscopic

cytopathiceffect[CPE]wascomplete).Infectedcells

wereharvestedbycentrifugationand thenwere

resus-pended in 1 ml of MEM (minus serum)/dish and

storedat-70°C.The titer of crude stockspreparedin

thisway was109to3x 109PFU/ml.

Purified vaccinia virus stocks were prepared by

infecting log-phase spinner L cells (about 5 x 105

cells/ml) ataninitialMOI of1with crude virus stock

and thenincubatingthe infected culture at370Cfor 48 h. Although one mightnot expect allcells to be

initially infectedat thisMOI, itin fact consistently

gavethehighestviralyields.Viruswaspurifiedfrom

the infected cellsby two cycles ofsucrose gradient

centrifugation as previously described (1) except that

virusdispersion by sonic treatment wastotally

elimi-nated andreplaced with gentlehomogenization in a

Duallhomogenizer. Assuming1unit of optical density

at 260 nm equals 1.2 x 1010particles, purified viral

pelletswereresuspendedin1mMTris-hydrochloride

(pH 9) to a concentration of1mg/ml(2x 1011

parti-cles/ml) andstored insmall portionsat-70°C. Virus

yields from this procedure were 150 to 300 PFU/cell,

and the resultantpurifiedvirushad a PFU-to-particle

ratio of 1:8 to 1:20.

Plaque assay. Infectious vaccinia virus was

as-sayed induplicateonconfluent60-by 15-mmdishes

of BSC-40 cells. Ten-fold serial dilutions of purified

virus were made inphosphate-buffered saline

contain-ing1 mMMg2".Crude virus stocksweretrypsinized

with anequal volume of 0.25% trypsin at370Cfor 15

min and then diluted with adsorption medium. In

either case,washed-cellmonolayers were infected with

0.5ml of virus inoculum for3h at25°C with occasional

rocking to prevent drying and to obtain a uniform

distribution of virus. Virus inocula were removed after

3h andreplaced with MEMplus 5% AFCS and 1%

agar (Noble). The assays were incubated at 37°C

under 5%C02for2 to3daysand then were stained

with 0.01% neutral redfor2 to 4h to visualize and

enumerateplaques.

Enucleation ofcell monolayers. Cell monolayers

wereenucleated on 60-by 15-mm tissue culture dishes

byuseof thedrugcytochalasinB (3). Slight

modifi-cations ofpreviously published procedures (25)were

necessary to achieve high-yield mass enucleation of

the variouscelllines(specific proceduresfor eachcell

line aregivenbelow). Subconfluent monolayers ofcells

wereinvertedasepticallyinbottles of 37°C

cytochal-asin B-containing medium plus serum and

preincu-bated at 37°C. The bottles were then placed in a

prewarmed GSA rotor and centrifuged in a Sorvall

RC2-B centrifuge at 37°C. After centrifugation, the

cytochalasin B-containing medium was replaced with

mediumplus serum, and the monolayers were put at

370C

untilthey returned to their normal morphology.

Control cellsweretreatedsimilarly except that they

were notcentrifuged. The efficacy of enucleation was

estimated by Giemsa staining the monolayers and

counting the percentage ofcellswhichstillretained a

nucleus. Thecellloss was estimated by both

micro-scopicinspectionandcountingcells,before and after

enucleation, byuseofahemocytometer. Specific

de-tails of theprocedure used for each cellline are as

follows: for L cells, MEM + 5% AFCS + 10 jAg of

cytochalasin B/ml, 15-minpreincubation,

centrifuga-tionat8,500 rpm for45min,1-hrecovery; for BSC-40

cells,MEM+10%AFCS+10

Mlg

ofcytochalasin B/ml,

nopreincubation, centrifugation at 8,500 rpm for 20

min, 1-h recovery; for CVCcells, BME + 10%AFCS

+50,ugofcytochalasin B/ml, 1-hpreincubation,

cen-trifugationat10,500 rpm for45min,30-minrecovery.

Single-step growth experiments. After the

re-covery interval of the enucleation procedure,

mono-layerswerewashed twice with

370C

medium(minus

serum) and then infected with 0.5 ml ofanappropriate

dilution of purified vaccinia virus in

phosphate-buffered salinecontaining 1mMMg2'.After 30 min

at 25°C, the virusinoculum was removed, and the

http://jvi.asm.org/

(3)

monolayers were washed twice with 370C medium

(minusserum).Mediumplusserum(4ml)wasadded,

and thedisheswere put at370Cand 5%C02.Atthe

indicatedtimesduplicatedishes ofvirus-infectedcells

andcytoplasts were put at -70°C to stop further virus

development. The following day the dishes ofcells

were thawed, andalldebris was loosened with the aid

of a rubber policeman and an additional 1 ml of

medium. The entireinfected-cell lysate wascollected

and refrozen. Infectious virus titers were then deter-mined as described earlier. Experiments measuring virus macromolecular synthesis were carried out

sim-ilarlyexceptthat themonolayerswerewashed twice

with medium(lacking serum and the appropriate

me-tabolite) prior tobeing labeled with the radioactive

precursor.

SDS-PAGE. Sodium dodecyl

sulfate-polyacryl-amidegel electrophoresis (SDS-PAGE) performed on

12.5%slabgelsby the methods of Studier (28). Small

samples ofradioactively labeled infected cellswere

acetone precipitated, dried, resuspended in sample

buffer,boiled for 5min,and thenanalyzedon

1-mm-thick slab gels. Autoradiography of dried gels was

carriedoutwith Kodak RPRoyalOmat medical

X-rayfilm,and the films were later developed in a Kodak

Rapid-Processautomaticdevelopingunit.

Electronmicroscopy. Multipledishesof BSC-40

cells andcytoplasts (106 cells/dish,98.6%enucleation,

2% cell loss) were infected with vaccinia virus at a

MOI of 10. At 0, 3, 6, 9, and 12 hpostinfection,cells

wereremoved from the dishes withtrypsin-EDTA and

harvestedby low-speed centrifugation.Thecellswere

suspended in 1 ml of 2.5%glutaraldehyde in 0.15 M

cacodylate buffer (pH 7.4) containing 2 mM CaCl2.

After 2 min at250C,thesuspensionwascentrifuged,

the supernatant fluid was decanted, and the

undis-turbedpelletswerefixed in the aboveglutaraldehyde

for30minat250C.Thepelletswerepostfixedin 1%

osmium tetroxide in 0.15 Mcacodylate (pH7.4) for 30

minand then stained en bloc with 0.5%uranylacetate

in acetatebuffer (pH 5.0) for 30 min,dehydratedin

acetone, and embedded inEpon812(14).Thin sections

were cut on a Porter-Blum MT-2 ultramicrotome,

stained withuranylacetateand leadcitrate(27),and

examinedin a JEOL lOOBelectronmicroscopeat 60

kV.

RESULTS

Lackofvaccinia virusreplicationin

enu-cleated mammalian cells.Cytochalasin B-in-ducedmassenucleation ofmonolayersof animal cells isawidelyusedtechniquewhichhasproven

useful in the study of various virus-host cell interactions (25). We have utilized this meth-odologytoenucleate several different cell lines

to testtheeffects of thisprocedureontheability

of thecellstosupportasubsequent infectionby vacciniavirus. Figure 1 shows the results

ob-tained by infecting CVC, L, and BSC-40 cells

andcytoplastswith vaccinia virus at low multi-plicity and monitoring infectious virus growth for 24 h postinfection. Low-multiplicity infec-tions, 5 to 10 PFU/cell, were chosen for two

I

U)

5

0

0-j

0 8 16 24

[image:3.505.277.449.64.521.2]

HOURS P. I.

FIG. 1. Replication ofvaccinia virus in control

and enucleated mammalian cells. Subconfluent

monolayersofintactcells(0)orcytoplasts(0)were

infectedwithvacciniavirus at a MOIof 5. Duplicate

dishes were removed at theindicated times,and virus

washarvested and titered on BSC-40 monolayers as

described. (a) CVC cells (9 x 105 cells/dish, 99.8%

enucleation, 5% cell loss); (b) L cells (5 x 106

cells/dish,99.5%enucleation, 8% cell loss); (c)

BSC-40cells(1.5x 10'cells/dish,98.4%enucleation, 3%

cell loss).

reasons.First,itwasfoundthat theseconditions

gave the highest relative virus yield in 24 h.

Second, high-multiplicity infections elicitedan

acute destructive CPEwhich cytoplasts might

http://jvi.asm.org/

(4)

708 HRUBY, GUARINO, AND KATES

have beenless abletotolerate andrepair than wholecells. It is obvious inFig. 1that,although vaccinia virusgrewquite well inallthreecontrol cells (80to 105PFU/cell),itgrewpoorly ifatall

in thecognate cytoplasts (0.4to 1.2 PFU/cell).

The small increment of increase in infectious virus titer inthecytoplastpreparationscould be accounted for by the 1 to 3% remaining

nu-cleated cells,soit isdoubtful whether enucleated cells are capable of producing any infectious

progenyvaccinia virus. Also, noplaquesorfoci

of infection could be observedonmonolayersof cytoplastsinfected with50to100PFU of virus

perdish andstained with neutral redatvarious timespostinfection. As acontrol,the above

ex-periments wererepeatedwith vesicular

stoma-titis virus whichwasabletogrowto30to100% ofcontrol titers, depending on initial MOI, in each of the variouscytoplasts,inagreementwith previous work (7). One interesting experiment would have been to enucleate vaccinia virus-infected cells at various times postinfection to

ascertain when the nucleuswasrequiredin the viralgrowthcycle.Unfortunately, the combina-tion of virus-induced CPE and centrifugation caused the cellstofall off the dishduring

enu-cleationattimes after3 h,and thus the above experimentwas nottechnicallyfeasible.

Furtherstudies into theapparentnuclear

re-quirement for vacciniavirusgrowthwerecarried

outwithBSC-40 cells. Thissystemwas chosen foravarietyofreasons:itaffordedease,rapidity, andreproducibilityof theenucleationprotocol; the biosynthetic capabilities of BSC-40

cyto-plasts were very high, as will become evident later; BSC-40 monolayers demonstrated the highest plaquing efficiency of several cell lines

thatwere tested, soexperiments would be

car-ried out and titered on homologous cells; and, finally,results could becomparedtoearlier work whichusedasimilar cellline,BSC-1 (22).

Experiments designedtoruleoutpossible

ar-tifactual results duetoside effects from thedrug cytochalasinBonvacciniavirusreplicationwere

carriedout.Pretreatmentof BSC-40monolayers

with 10 ,ug ofcytochalasin B/mlorpresence of

thedrugthroughoutthe courseof infectiononly reduced thevirusyieldby 21 and32%,

respec-tively.Thisagrees withearlierstudies in which 4 ,ug of cytochalasin B/ml was used (6). Note thatthe1-hpretreatmentregimenwasthe nor-mal control used in most of the experiments reportedhere. Underamicroscopeitwas

appar-ent that BSC-40 cells rapidly extruded their

nucleusinresponsetocytochalasinBto apoint

whereitwasconnectedtothe mainbodyof the

cellbyonlyaslender threadofcytoplasm.Even

in this grossmorphological state, thecells

sup-ported vacciniavirusgrowth,whereasbreakage

of the cytoplasmic stalk withcentrifugal force

during the enucleation procedure resulted in a

total loss of infectious virus production (Fig. 1).

Onetrivial explanation for the failure of

vac-cinia virus to replicate in enucleated BSC-40

cells is that during the period of timerequired

for virusdevelopment (9to 10h) the cytoplasts deteriorate too rapidly for mature virus to be

formed. Several lines of evidenceargue against

this. First of all, other viruses with generation

timessimilartothatofvaccinia virus growwell inenucleated BSC-1cells(8),althoughthe

com-plex morphogenetic pathway ofvaccinia virus may complicate this sort of analogy. Second, BSC-40cytoplastsareabletotakeup and retain

the vitaldye neutral red for 36 to 48 h. Third,

infectingBSC-40cellsand cytoplastsatseveral

MOI (1, 10, or 100) to minimize or maximize vaccinia virus-inducedcellularCPEdidnot raise orlower the ability of cytoplaststosupport virus

growth, whereas there was aninverse relation-ship between multiplicity and virus yield in

whole cells. Microscopic observations

demon-strated that whatever virus function elicited CPE (rounding up and clumping), it was

ex-pressed with the same timing and severity in

cytoplastsasin wholecells.Finally, carryingout

experiments at

310C,

acondition knownto

pro-long the integrity and pro-longevity of cytoplasts

(25), didnotallow vaccinia virusreproduction in

cytoplasts but it did lower the virus yield in

wholecells50% (datanotshown).

Expression of both early and late virus functions in enucleated cells. The ability of vaccinia virustodirect itsownmacromolecular synthesis andtoinhibit host-specificevents was next examined. Figure 2 shows the results of

pulse-labeling vaccinia virus-infected BSC-40

cellsandcytoplasts with

[3H]thymidine

to mea-sureDNAsynthesis. Incorporation of the radio-active label showed virtually identical kinetics

in bothcases,withthemaximumrateoccurring at 3 h postinfection. At this time, the virus-infected cytoplastsweresupporting synthesis of

vaccinia virus DNA at about 80% of control

levels,althoughthiscapability dropped off more

rapidlyatlatertimes(3.5 to 5 h) in the cytoplast

preparations. To ensure that what was being

measuredwas, infact,vaccinia virus DNA

syn-thesis, an identical experiment wascarried out

except that cytoplasmic extracts were made at

each time point andcounted. The results were

identical to thosein Fig. 2, indicating that

vir-tually all of the radioactive counts were being incorporated in the cytoplasmic fraction, i.e., into viral DNA. The fact that vaccinia virus

DNA wassynthesizedincytoplastsstrongly

im-plies that the virus-coded early functions for DNAbiosynthetic enzymes and core uncoating

http://jvi.asm.org/

(5)

12 0

8

100

0

4I 1

I

-J 0

I-O 2 4 6

HOURS P.I.

FIG. 2. Vaccinia virus DNAsynthesis ininfected

BSC-40 cells. Intact BSC-40 cells (0) orcytoplasts

(0) (2.75x 106cells/dish, 98.7%enucleation,2%cell

loss) wereinfectedwith vacciniavirusat aMOIof

10.Atthe indicated times, disheswerepulsedwith

[3H]thymidine (2,uCi/ml) for 10min, and then the

trichloroacetic acid-precipitable radioactivity per

dishwasdetermined.(a)Pulse-labeled vaccinia virus

DNAsynthesis.(b)Cumulativeplot ofvaccinia virus

DNA synthesis. Note: at 3 h, dishes of uninfected

BSC-40 cells andcytoplasts incorporated154,000and

1,200 cpm,respectively.

had been expressed (24). The question ofthe

newlysynthesized DNAbeing transcriptionally active in enucleates was next examined. RNA

synthesis was not studied directly, but rather

theviruspolypeptides beingcoded forwere

re-solved by SDS-PAGE. In the absence of any

stringent translationalcontrol,theviralproteins being translated should reflect theamountand

types of vaccinia virus mRNA species being

transcribedatvarioustimespostinfection. Figure3shows the results of[35S]methionine pulse-labeling of BSC-40 cells and cytoplasts

infected athigh multiplicity(50PFU/cell) with

vaccinia virus. The top panel exhibits some of

the kineticconsiderations. The first point to be

made is that BSC-40 cytoplasts, whether

in-fected ornot,initiallysynthesized protein at 60

to 70%of the rate of controlcellsand then lost

their protein synthetic capabilities slowly

throughout the duration of the experiment. In both cells and cytoplasts, a rapid inhibition of

host-cellprotein synthesis occurred during the

first1.5h, but thiswasrapidly obscuredas

virus-specific proteins began to be made (3 to 8 h).

This effect was even more dramatic when the

FIG. 3. Vaccinia virus protein synthesis in BSC-40

cells. IntactBSC-40 cells(0)or cytoplasts(0) (2.4 x

106 cells/dish, 98.0% enucleation, 2% cell loss) were

infectedwith vaccinia virusataMOIof 50. At the

indicatedtimes,disheswerepulsedfor20minwith 5

,uCiof[3S]methionine/ml. Cells were harvested into

0.5mlof ice-cold 1 mMTris-hydrochloride (pH 9).

Uninfected cellsandcytoplastswerealso monitored

(dashed lines). (a)Samples of100Il were processed

todeterminehottrichloroacetic acid-precipitable

ra-dioactivity. (b)Amountsof50,ul of the infected

sam-pleswere analyzed by SDS-PAGE andsubsequent

autoradiography (lanes 1-5, vaccinia virus-infected

BSC-40 cells at 0, 1.5, 3, 5, and 8 hpostinfection;

lanes6-10,vacciniavirus-infected cytoplastsat0, 1.5,

3, 5,and 8 hpostinfection).

http://jvi.asm.org/

[image:5.505.73.218.66.331.2] [image:5.505.252.447.132.522.2]

(6)

infectionswerecarriedout at alow multiplicity

(5 PFU/cell), in which case virus protein

syn-thesis was only a smallpercentage of the total.

Figure 3b is an autoradiograph of an

SDS-polyacrylamide gel analysisofthe

[35S]methio-nine-labeledproteins synthesizedinthe

experi-ment shown in Fig. 3a. Invirus-infected whole

cells (lanes 1-5) and enucleated cells (lanes

6-10), the patternofgeneexpressionwas

iden-tical:at 0hprimarily host proteinswerelabeled,

at 1.5 hearly vacciniavirus proteins appeared,

at 3h (coincident withpeak viralDNA

synthe-sis) therewerebothearlyandlate vaccinia virus proteins beingmade, andat 5and8hlate viral proteins predominated.Also, itwasobviousthat

the same proteins in the same relative ratios werebeing synthesizedinthe enucleatedcellsas

in the whole cells, albeit in somewhat reduced

amounts. The early andlate vaccinia proteins

wereidentifiedby comparisonto standard vac-cinia virus proteins labeled in the presence or

absence of vaccinia virus DNA synthesis (21;

Fig. 4). It has previously been shownthat

vac-cinia virusproteinsundergo normalcleavagesin

enucleated cells (22), andwe were also ableto

substantiate this (datanot shown). The status

of otherproteinmodifications in vaccinia virus-infected enucleatedcells, suchasacetylationor

phosphorylation, isnotyetknown.

Itwasof interest to seehowlong enucleated cellswerecapable ofsynthesizing vacciniavirus

proteins. Therefore, infectionswerecarried out

in the presence of cytosine arabinoside, which

inhibits DNA synthesis and hence late gene

expression. Figure 4 showsthat, although host protein synthesiswasrapidly shut off, early

vac-cinia proteins continued to be synthesized in

enucleated and whole cellsformore than24h.

Thus, it wasobvious that enucleatedcellshave

thecapabilityofmaking largeamountsof virus protein forlongperiods of time. It was

interest-ing to note that in the absence of late gene

expression there was a great reduction in

ob-served morphological CPE, thereby implying that this effect was the result of a late gene function.

DNA packaging and viral

morphogene-sis.Theassemblyof vaccinia virus constituents

into matureinfectiousparticlesis acomplicated

multistepprocess which has beenonlypartially

characterized(17). It waspossiblethat,although

vaccinia virus-infected enucleated cells

pro-duced all of the necessary viral components,

criticalcatalytic concentrations of one or more

may not have beenreached, thereby inhibiting

morphogenesis. To address this question, we

used the drugrifampin, which has been shown

to completely block vaccinia virus assembly

while allowing continued expression of both

.. X \.

HOU.8 P

R E.

H0iP' PT

b...

FIG. 4. Vaccinia virus protein synthesis in the

presenceof cytosine arabinoside(CAR).BSC-40 cells

(0) or cytoplasts (0) (2.3 x 106 cells/dish, 97.6%

enucleation, 3%o cellloss)wereinfected with vaccinia

virusataMOIof50.CAR waspresentat25pg/ml

throughout the infection. Disheswerepulsed with2

,ICiof[35S]methionine/mlatthe indicatedtimes and

analyzed as described in the legendofFig. 3. (a)

Vaccinia virus protein synthesis. (b) SDS-PAGE

analysis of [35S/methionine-labeled vaccinia virus

proteins (lanes 1-5, 50,ul ofvirus-infected BSC-40

cellsat0, 4,8, 12, and24hpostinfection; 6-10, 100

1lofvirus-infected cytoplastsat0, 4, 8, 12, and24h

postinfection.

early and late virus functions (19). Upon drug

removal this effect israpidly reversed, and

infec-tiousparticles arequicklyformed. Table 1 shows

the results ofsuch a treatmentonvaccinia

virus-infected BSC-40cells and cytoplasts. The rifam-pin block-reversal step elicited a 3.5-fold increase

of virusgrowthinwhole cells. No similar

ampli-fication was observed in the cytoplast

prepara-tions.

Earlier it was demonstrated that vaccinia

vi-rus-specific DNA synthesis occurred at a high

rateininfectedcytoplasts (Fig. 2),butthe

sub-sequent fate of this DNA was not examined.

Table 2 shows the results obtained when the

viral DNAsynthesized from2to4h

postinfec-tion was labeled with [3H]thymidine and then

J. VIROL.

http://jvi.asm.org/

[image:6.505.269.458.69.378.2]

(7)

NUCLEAR REQUIREMENT FOR VACCINIA VIRUS GROWTH

TABLE 1. Effectonrifampinonthe yield of

vaccinia virus from infected BSC-40 cellsa Timeof Control Enucleatedcells rifampin

removal PFU/dish PFU/cell PFU/dish PFU/cell (h)

0 3.4 x 107 50.3 3.5 x 105 0.5

8 1.2 x108 177.8 3.7x 105 0.6

24 3.6 x105 0.5 3.3 x 105 0.5

aBSC-40

cells

and cytoplasts (7 x 105

cells/dish,

99.5% enucleation, 1% cell loss) were infected with

vaccinia virus at a MOI of5inthe presence of 100,tg

of rifampin/ml. Drug-containing medium was

moved at the indicated times postinfection and

re-placed with medium alone. At 24 h, infections were

stopped and titers were determined as usual. All assays

weredone induplicate.

evaluatedatvarious timesforpackaging,as

as-sayed by conversion of the labeled DNA froma

DNase-sensitivetoaDNase-resistant form(19).

Whereasmorethan50% of theviralDNAmade

inwhole cells became resistanttotheactionof

DNase, therewas no change in the DNase

sen-sitivity of cytoplast viral DNA in 20 h. This

observation was supported by the failure ofus

andothers (22)toisolatephysical particles from

vaccinia virus-infectedenucleated cells. Vaccinia

virus infections of BSC-40 cells and cytoplasts

were carried outin thepresence of

[3H]thymi-dine. The labeledlysates, plusunlabeledcarrier

virus, werethensubjectedtothevirus

purifica-tion scheme described in Materials and

Meth-ods. Less than 5% of the radioactivity banding

with maturevirus from controlcellswasfound

ata corresponding position from the cytoplast

preparation (data notshown). Taken together,

thesedataimplythat viral DNAsynthesizedin

the absence of a host-cell nucleus was either

neverpackaged orwas packagedin an

osmoti-callyfragile form thatwasdisrupted during the

DNase assays and particle purification

proce-dures.The exact nature ofmorphogeneticevents

happening in vaccinia virus-infected cells and

cytoplasts was then examined in an electron

microscope.

Figure 5 displays typical examples of virus

developmentalforms observed in vaccinia

virus-infected BSC-40 cells during a single growth

cycle experiment.Fromuncoatingtoproduction

ofnewinfectious progenyvirus 9 to 12 hlater,

an orderly set of morphogenetic steps can be

seen,similartothosepreviously described (17).

Vacciniaviruswastakenupanduncoated(Fig.

5A), the cores subsequently broke down and

became "viroplasm" or "virus factories" (Fig.

5B), membraneswithprominent spicules

enve-loped "viroplasm" (Fig. 5C), andDNA

conden-sationbeganatoneend(Fig. 5D) andproceeded

to the formation ofmature particles (Fig. 5E).

Figure 6 demonstrates that thiswasnot true

in virus-infectedenucleatedcells.Although the

earlystepsofmorphogenesisuptoformation of

immature particles (Fig. 6A and B) appeared

normal, beyond this development went awry.

DNA condensation, ifseenatall,wasaberrant,

giving a mottled appearance to the immature

particles (Fig. 60). At later times (6 to 12 h

postinfection), membrane aberrations and bi-zarreviral formswereapparent(Fig. 6D). Virus

development seemed to be arrested ata point

analogoustotheclass Jmutants of Dalesetal.

(5).Also, virus factories of similar size and

den-sitywereobserved in both whole and enucleated

cells.

DISCUSSION

Pennington and Follett originally observed

that vaccinia virus wouldnotgrowinenucleated

BSC-1 cells (22). These experiments were

car-riedout atahigh MOI under conditions where

viral yields in control cells wererelatively low

andoverallviralgeneexpression and DNA

rep-lication in enucleated cells were depressed. It

could beargued that this resultwasduetosome

factors other than enucleation per se. For

ex-ample, the high MOI could cause severe CPE

whichwould result in aninhibition ofthe

pro-duction of critical concentrations of viral

com-ponentsrequired forpropermaturation in

BSC-1cells.However,wehavebeen abletoshowthat

inseveralcell lineslow-multiplicity vaccinia

vi-rus infectionsresulting in good viral growthin

cellsdidnot elicit any viralgrowth in the

cor-responding cytoplasts (Fig. 1). Byallthe criteria

tested, both early and late viral functionswere

expressed properly andatrelatively high levels

in enucleated BSC-40 cells (Fig. 2 and 3), but

TABLE 2. Conversionofvaccinia viruscytoplasmic

DNAtoaDNase-resistantforma

Timepostinfec- Percent DNase-resistant cpm tion

(h) Control Enucleated

4 7.6 7.5

8 42.4 5.5

20 52.0 8.6

aBSC-40

cells

andcytoplasts (3.5x 106

cells/dish,

98%enucleation,3%cellloss)wereinfected with

vac-ciniavirus at aMOI of10.From2 to 4hpostinfection,

thedisheswerelabeledwith 2,uCiof[3H]thymidine.

Atthe indicatedtimes,duplicate plateswereremoved,

cells were harvested, and acytoplasmic extract was

prepared (19). One half of theextractwasprecipitated

withtrichloroacetic acid and counted; theother half

wasmade50yg/mlinpancreaticDNaseand incubated

at37°C for30niinpriortotrichloroacetic acid

precip-itation.

71

VOL. 29,1979

http://jvi.asm.org/

[image:7.505.263.454.511.572.2]

(8)

*~~~~'F- F# = m

FIG. 5. Electron micrographs ofthin sectionsofvacciniavirus-infectedBSC-40 cells. Displayed are the

typicalvaccinia virusmorphogenetic formsthatwereobservedattheindicated timespostinfection. (A)Virus

particle shortlyafterpenetration, Oh. (B)Virusfactory,3h.(C)Immatureparticlesintheprocess of enveloping

DNA, 6h. (D)Partiallycondensed DNAwithin immatureparticle,9h. (E)Mature virusparticle, 12h. Bar

=0.1

ALM.

the viral DNA synthesized in enucleated cells the maturation of the viral components into

wasnotconvertedtoaDNase-resistant formas matureprogenyparticleswasaberrantorabsent

itwasin controlcels(Table 2).Amoredetailed inenucleatedcells (Fig. 5and6). Incontrast to

electron microscopic study demonstrated that PenningtonandFollett (22),wefoundno

differ-J. VIROL.

http://jvi.asm.org/

[image:8.505.92.438.78.572.2]

(9)

B

A

&

i-,-4.. ,,ll.. t

-A. I 1

FIG. 6. Electronmicrographs of thin sections of vaccinia virus-infected BSC-40 cytoplasts. The

predomi-nantforms observedatthe indicated timespostinfectionare shown. (A)Immature particle enveloping DNA,

3h. (B) Completely enveloped particle,6h.(C) Immature particle exhibiting faultyDNAcondensation, 12 h.

(D)Immatureparticleslateininfection withaberrant membranes and noapparent DNAcondensation,12 h. Bar=0.1,im.

encein the number or size of virus factories in

the enucleated cells. From the data reported

here and from other studies of virus-host cell

interactionsusingenucleatedcells(25), it is

un-likelythat this result is due to the enucleation

procedure itself. Therefore,ourdatasupport and

extend earlier work (22) and indicate that the

hostnucleusplays some importantrole in

http://jvi.asm.org/

[image:9.505.81.422.74.571.2]

(10)

cinia virus growth. This also agreeswith other studiesusing metabolic inhibitors (15, 26). Thus, it would seen that, like frog virus 3, vaccinia viruscan nolonger be consideredacytoplasmic

virus in thestrictsenseof the word (23).

Ifone acceptsthe notion that vaccinia virus

requiresthe host-cell nucleus in orderto

repli-cate, then several interesting questions are

raised. First, does vaccinia virus merely require the physical environment of the nuclear body,

or mustthenucleus be active? Previous studies

with phytohemagglutinin-activated leukocytes suggested an active nuclear requirement (16). However, because of the quiescentstate of the protein synthetic machinery in resting

leuko-cytes,theseresultsmustbe viewed with caution.

Asecondquestionis whether theneeded nuclear functionisacomponent or an enzymeand, sim-ilarly, whether this function is normally

ex-pressed inuninfected cellsoris inducedby the vaccinia virusinfection. Onefinding whichmay

bepertinent is that of LaColla and Weissbach (13), whoshowed thatsomevacciniavirus DNA replication occursin the nucleus. Itis possible that replication of vaccinia virus DNA in the nuclear environment is required for expression of the needed nuclear function. However, the observed absence of UV-enhanced reactivation ofsubsequent vaccinia virus growth under

con-ditions which stimulate the growth of a bona fide nuclear virus such asherpes simplex virus (2) questions this hypothesis.

Whatever itsnature,therequirement for

nu-clear function becomes apparentduring the

as-sembly of vaccinia virus. Itmaybe asituation analogoustothegroElocus of Escherichiacoli, where a host-coded function is required for

properphage morphogenesis (4). Itisalso

pos-sible thatone or moreof the virioncomponents

(DNA,RNA,orprotein)aresynthesizedor mod-ifiedimproperly in the absence of the nucleus. Experimentally, thismay notbe detected until

this component is unable to interact

produc-tively during assembly,hencefaulty morphogen-esis.

It is interesting to consider how these data

compare with earlierresearch on the effectsof

theanti-vaccinia virusdrugVirazole orribavirin

(1-fi-D-ribofuranosyl-

1,2,4-triazole-3-carboxam-ide) (12). This drug allows expression ofboth

earlyand late viralfunctionsbutgreatlyinhibits

production of infectious progeny. It has been

speculated that Virazole inhibits some specific

late viral function. An alternate explanation is

that the drug inhibitstherequiredhostnuclear

function.

Obviously, much more work remains in the

studyof vaccinia virusreplication and howthe

virus interacts with both the cytoplasmic and

nuclear compartments of the cell. Current

re-search efforts in this laboratoryarecenteredon

using intact cells to investigate the molecular

nature of the apparent nuclear requirement.

Thisareaof research wouldseem tobe

impor-tant in understanding vaccinia virus-host cell interactions and possibly in elucidating previ-ously undescribed host-cellfunctions.

ACKNOWLEDGMENTS

Thiswork wassupported by Public Health Service training grantCA-09176-04 from the National Cancer Institute and by Public Health Service grantAI-11839-01 from the National Institute of Allergyand Infectious Diseases to J.R.K.

Wethank D. L.Lynn for excellent technical assistance and R. C. Condit for many valuable discussions and suggestions. We alsoacknowledge the members of J. Lucas's laboratory forsharing theirexpertisein cellenucleation procedures.

LITERATURE CITED

1.Becker, Y., and W. K.Joklik.1964.Messenger RNA in cells infected with vaccinia virus. Proc. Natl. Acad. Sci. U.S.A.51:577-585.

2. Bockstahler, L.E., andC. D.Lytle. 1977. Radiation enhanced reactivation of nuclearreplicating mamma-lian viruses. Photochem. Photobiol.25:477-482. 3. Carter, S. B. 1967. Effects of cytochalasins on mammalian

ceLs.Nature(London)213:261-264.

4.Casjens, S.,and J.King.1975.Virus assembly. Annu. Rev. Biochem. 44:55-611.

5. Dales, S., V. Milovanovitch, B. G. T. Pogo, S. B.

Weintraub, T. Haima, S. Wilton, and G.

Mc-Fadden.1978.Biogenesisof vaccinia: isolation of con-ditional lethalmutantsand electronmicroscopic char-acterizationof theirphenotypicallyexpressed defects. Virology 84:403-428.

6. Deitch,A.D.,S.G.Sawicki,G. C.Godman, and S. W.Tanenbaum.1973.Enhancement of viral infectiv-ity bycytochalasin. Virology 56:417-428.

7. Foilett,E.A.C.,C. R.Pringle,W.H. Hunner, and J. J. Skehel. 1974.Virus replication in enucleatecells:

vesicular stomatitis virus and influenza virus. J. Virol. 13:394-399.

8. Follett,E.A.C., C.R.Pringle,and T. H.Pennington. 1975.Virusdevelopment in enucleate cells: echovirus,

poliovirus, pseudorabies virus,reovirus, respiratory syn-cytial virus and semliki forest virus. J. Gen. Virol. 26: 183-196.

9. Gafford,L.G.,andC. C. Randall.1976.Virus-specific

RNAand DNA in nuclei of cells infected with fowlpox virus.Virology 69:1-14.

10.Goorha, R.,D. B.Willis,and A.Granoff.1977. Mac-romolecularsynthesisin cells infected by frog virus3. VI.Frog virus 3replicationisdependentonthecell

nucleus. J. Virol. 21:802-805.

11.Joklik,W.K.,and Y. Becker.1964.Thereplicationand uncoating of vacciniaDNA.J. Mol. Biol.10:452-474. 12.Katz,E.,E.Margalith,and B. Winer.1976.Inhibition of vaccinia virusgrowth bythenucleosideanalogue

1-,8-D-ribofuranosyl-1,2,4-triazole-3-carboxamide

(vira-zole, ribaviran).J. Gen. Virol.32:327-330.

13.LaColla, P., and A. Weissbach. 1975. Vacciniavirus infection of HeLa cells.I.Synthesis ofvaccinia DNA in hostcellnuclei. J. Virol.15:305-315.

14.Luft, J. H. 1961.Improvementsinexpoxyresin embed-dingmethods. J.Biophys. Biochem. Cytol. 9:409-412. 15.Magee, W. E., and 0. V. Miller. 1968. Initiation of vaccinia virusinfection in actinomycin D-pretreated

cells.J. Virol.2:678-685.

16. Miller,G., andJ.F. Enders.1968.Vaccinia virus

http://jvi.asm.org/

(11)

cationand cytopathic effect in cultures of phytohem-agglutin-treated human peripheral bloodleukocytes. J. Virol.2:787-792.

17. Morgan, C. 1976.Vaccinia virus reexamined: develop-mentandrelease. Virology73:43-58.

18.Moss, B. 1974.Reproduction of poxviruses,p.405-474.In

H.Fraenkel-Conrat and R. R. Wagner (ed.), Compre-hensivevirology, vol. 3. Plenum Press, New York. 19. Moss, B., E. N.Rosenblum, E. Katz, and P. M.

Grim-ley. 1969.Rifampicin: aspecificinhibitor ofvaccinia virus assembly. Nature(London) 224:1280-1284. 20. Ortin, J., and E. Vinuela. 1977. Requirement of thecell

nucleus for African swine fever virusreplication in Vero cells. J.Virol. 21:902-905.

21. Pelham, H. R. B., J. M. M.Sykes, and T. Hunt.1978. Characteristics ofacoupled cell-free transcription and

translationsystemdirectedby vaccinia cores.Eur. J.

Biochem. 82:199-209.

22. Pennington, T. H., and E. A. C. Follett. 1974. Vaccinia virusreplication in enucleate BSC-1cells:particle

pro-duction andsynthesis of viral DNA and proteins. J.

Virol. 13:488-493.

23. Portner, A., and C. Pridgen. 1975. Viral replication,p.

27-42.In A. Notkins (ed.), Viral immunology and im-munopathy. AcademicPress Inc., New York. 24. Prescott,D.M.,J.Kates,and J.B. Kirkpatrick. 1971.

Replication of vaccinia virus DNA in enucleated L-cells. J. Mol. Biol. 69:505-508.

25. Pringle, C. R. 1977. Enucleationas atechnique in the

study of virus-host interactions. Curr. Top. Microbiol. Immunol. 76:49-82.

26. Reich, E., and R. M. Franklin.1961.Effect ofmitomycin C onthe growth ofsomeanimalviruses.Proc.Natl.

Acad. Sci.U.S.A. 47:1212-1217.

27. Reynolds, E. S. 1963. Theuseof lead citrateathigh pH

as anelectronopaquestain in electronmicroscopy. J.

Cell Biol. 17:208-213.

28. Studier, F. W.1973.Analysis of bacteriophage T7 early RNAs and proteins on slab gels. J. Mol. Biol. 79:

237-248.

29. Walen, K. H. 1971. Nuclear involvement in poxvirus infection. Proc. Natl. Acad. Sci. U.S.A. 68:165-168.

http://jvi.asm.org/