Nucleosome-like structural subunits of intranuclear parental adenovirus type 2 DNA.

(1)

0022-538X/79/03-0888/11$02.00/0

Nucleosome-Like Structural

Subunits of Intranuclear Parental

Adenovirus

Type 2 DNA

ALAINSERGEANT,t MICHAELA. TIGGES,ANDHESCHELJ.RASKAS*

Department of Pathology and Department of Microbiology and Immunology, Washington University, School

of

Medicine,St.Louis,Missouri 63110

Received for publication 17 February 1979

The intranuclear structure ofparental adenovirus 2 DNAwas studied using

digestion with micrococcal nucleaseas aprobe.When cultureswereinfected with 32P-labeled virions, at a multiplicity of 3,000 particles per cell, 14 to 21% of parental DNA penetrated thecelland reached the nucleus. Of thisparentalDNA, 60% could besolubilized byextensivedigestion with micrococcal nuclease. The nuclease-resistant fraction contained viraldeoxyribonucleoproteinmonomersand

oligomers.Thesenucleosome-likestructurescontained DNAfragmentswhichare

integral multiples ofa unit-length DNA ofapproximately 185 base pairs. The monomeric DNA is similar inlengthto

the4unit-length

DNAcontained in cellular nucleosomes.However, the viraloligomersareslightlysmallerthantheircellular

counterparts. DNA-DNAhybridization demonstrated that all segments of the viralgenome,including those expressedasmRNAonlyatlatetimes, are

repre-sented in the nucleosomal viral DNA. The amount ofearly intranuclear viral chromatin was proportional to multiplicity of infection up to multiplicities of

4,000particlespercell.However,viraltranscriptionalactivitydidnotincrease in directproportiontotheamount ofviral chromatin. Maximum accumulation of intranuclear viral chromatinwasachievedby3hafter infection.The intranuclear parental viral chromatin remained resistant to nuclease digestion even atlate times ininfection,after viral DNAreplicationhadbegun.

Adenovirustype 2replicationrequiresaseries of steps which result in the transfer of virion DNAtothe nucleus of theinfected cell. Studies of virus penetration anduncoating have dem-onstrated that the final steps in this process occur in the nucleus, approximately 1 to 2 h

after infection (24). Although purified virion DNA can be infectious (13, 27) and thevirion DNA has beenreportedtobe free of viral

pro-teinsafteruncoating (24),itseemsunlikelythat naked viral DNA normally serves astemplate

fortranscription. Itisnowwell established that

eucaryotic genes are packaged in chromatin

structures identified as nucleosomes (28).

Nu-cleosomeoligomersand monomers canbe gen-eratedby controlleddigestion of nucleior chro-matin by micrococcal nuclease (18). Quantita-tiveanalysishasdemonstrated thateach

nucleo-some contains about 185 to 200 base pairs of DNAconsisting of a nuclease-resistant core of 140 base pairs wrapped around an octomer of

the four histonesH2A, H2B, H3, andH4. This

coreis connectedtotheadjacentcorebyaDNA

tPresent address:U 102deVirologie,InstitutNational de la Sante etde la Recherche M6dicale, 59045Lille Ledex, France.

segment of15to 100basepairs (18), the linker DNA, which is more susceptible to digestion thancoreDNA.

Similarnucleosome structureshave been de-tected inpapovavirus DNA, both in virions and inintracellularstructurespresentlate in produc-tiveinfection (15, 25, 35). A similarstructurehas beenproposed for the virioncoresofadenovirus

2DNA(7).Ithas been suggested that the virion DNA(23x 106daltons)consists of 180

"nucleo-somes," each containing about200base pairs of DNAwrapped around six copies of the arginine-rich virionpolypeptide VII. Additionaldata sug-gestthat late inproductiveinfection newly

rep-licated viral DNA is also packaged in a

chro-matinstructure (7).

Wehave studied thestructureof intranuclear

parentaladenovirus2DNA incultures infected

withvirions containing radioactive viralDNA.

Using micrococcal nucleasetostudyDNA struc-ture, we have demonstrated that the parental

intranuclear DNAis in achromatin-like struc-ture. The relationship of these structures to

multiplicityofinfection (MOI),thetimecourse

of infection, and viral transcription has been

examined.

888

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

ADENOVIRUS 2 DNA INTRANUCLEAR STRUCTURE 889 MATERIALS AND METHODS

Cell culture and virus purification. KB cells

weregrowninsuspensionculturesasdescribed

previ-ously (10). Stocks of3P-labeled adenovirus 2 were

preparedasdescribed (8), and viruswaspurifiedby

cesium chloride centrifugation (10). Viruswas used

within14days after purification. The concentration of 32P-labeled adenovirus 2 preparationswasdetermined

from the optical density at 260 nm based on the

equation, 1optical densityunitat260nm- 1 x 1012

virus particles (24).

Isolation of nuclei. Suspension culture KB cells

wereinfectedataMOI of3,000particlespercellwith

radioactive virusor100PFU/cellwithnonradioactive virus,exceptwhere otherwiseindicated. After 1 h of adsorption,cellswerediluted to 3x105cellspermlin Joklik's minimum essential medium supplemented with 5% horseserum.Unless indicatedotherwise, cul-tureswereusuallyharvested3 h after thebeginningof adsorption, and thecells werewashed twice with cold phosphate-bufferedsaline.

All steps inharvestingcellsorpurifyingnucleiwere

performedat0 to4°C.Thefollowing threemethods

wereusedtopreparenuclei.

(i) Isotonic buffer.Washedcellswereresuspended

ataconcentration of 3x 10' cellsperml in 0.01 M Tris-hydrochloride, pH 7.5, 0.15 M NaCl, 0.0015 M MgCl2, and0.05% Triton X-100(SigmaChemicalCo.). After 15 min at 0°C, cells were disrupted with 10 strokesofatightlyfittedpestleinaDounce homoge-nizer. The nucleiwerepelletedat600xgfor 5minat 40C.

(ii) Hypotonic buffer. Washed cellswere

resus-pendedatadensityof 3 x 106cellsperml in 0.01M

Tris-hydrochloride, pH 7.4, 0.01 M NaCl, 0.0015 M MgCl2, and 0.05%TritonX-100. After 5minat40C, theswollencellswerelysedinaDouncehomogenizer with 20 strokes ofaloosely fitted pestle. The nuclei

werepelletedat600x gfor 5minat4°C.

(iii) Sucrose buffer. Washed cells were

resus-pendedataconcentration of107cellspermlinlysis buffer(0.3Msucrose,0.01 M Tris-hydrochloride, pH 7.9, 0.001 M CaCl2, and 0.5% Nonidet P-40 [Shell Chemical Co.]) (30). Cells were lysed at 4°C in a

Dounce homogenizer as in method ii. Nuclei were

collectedbylow-speedcentrifugation andwashed two times withlysisbuffer and twice withdigestionbuffer (0.3 Msucrose, 0.001 MTris-hydrochloride, pH 7.9, and0.1 mMCaCl2).Insomeexperimentswith isolated nucleidigestionbuffer contained1.0 mMCaCl2,which increasedthe rate ofdigestion bymicrococcal nuclease but didnot alter thepatternoffiagmentrelease with respect todegreeofsolubility.

Procedure iii was used routinely for isolation of nuclei before micrococcal nuclease digestion. Each batch of nucleiprepared bythe above methodswas

foundintact andessentiallyfree ofcytoplasmic

con-taminationby phase-contrastmicroscopy.All nuclear

preparationswereusedimmediatelyfor further

exper-iments.

Isolationofchromatin KBcellslabeledfor72h with[8H]thymidine (final concentration, 0.2 pCi/ml)

were washed with phosphate-buffered salineas

de-scribedabove (method iii). Chromatin was prepared fromthe nuclearpelletasdescribedby Hancock (17), except that 10mM Tris-hydrochloride, pH 8.0, re-placed phosphate buffer. The chromatin pellet was washed twice with 10 ml of 1 mM

Tris-hydrochloride-0.5 mM EDTA (pH 8.0) and twice with digestion buffer. After the final wash the chromatin pellet was disbursed in digestion buffer byshearingin a Dounce homogenizer fitted with a tight pestle. The DNA con-centration was adjusted to 300 to 500

Ag/ml

with digest buffer.

Micrococcal nuclease digestion of nuclei. Nu-clei from cultures infected with32P-labeled viruswere prepared by method iii and then resuspended in diges-tionbuffer to give a DNAconcentrationof about 300

pg/ml, asdetermined by the optical density reading (at 260nm) of an aliquot diluted in 5 M urea and 2 M NaCl (30). Thedigestion was carried out at370Cwith micrococcal nuclease (Worthington Biochemicals

Corp.;0.1 U/pg of DNA). Aliquots were removed at

various times,and the amount of labeled DNA ren-dered acid soluble was determined by precipitation with 10% trichloroacetic acid. In some experiments solubilization in perchloric acid was determined as described byLohr et al. (22).

WhenDNAwasto beanalyzed byelectrophoresis, thedigestion was stopped by addition of EDTA, NaCl, and pancreatic RNase A (2). The material was di-gested for 30 min at 370C and then treated with lauryl

sarcosine (0.5% final concentration) and 160 pg of

proteinase K per ml,followedby phenol extraction

(33).When deoxyribonucleoproteins (DNP) were to be analyzed by electrophoresis, the digestion was halted by the addition of EDTA to a 10 mM final concentration. Thenuclear suspension was held on ice for 60 min,sheared,andclearedbycentrifugationfor 10min at 10,000 rpm(SorvallR2B,SS34 rotor). The

supernatantwasused as asource ofnucleosome mon-omersandoligomers(3).

Gelelectrophoresis ofDNA and DNP. Nuclear DNApurifiedafter micrococcal nucleasedigestionand

adenovirus 2DNAdigested withendo R.SmaI (26) wereresolved on either 1.4% agarose slab gels (0.3 by 18by 16 cm) or2.0% agarose tube gels (0.7 by 12 cm) asdescribedpreviously(10).Electrophoresiswas per-formedovernightat 50V at room temperature for slab

gels and 8 h at 1.5 mA pertube forcylindricalgels.

32P-labeled viral DNA wasvisualized by

autoradiog-raphyofgelscovered withplasticwrap and exposed onKodak X-omat R film (XR-5).Theslices (1 mm) fromcylindrical gelswerehydrolyzed at80°C in 0.4 mlof 30% H202containing 1% concentratedNH4OH foratleast4h,neutralized with 1 M acetic acid, and counted in 5 ml ofFormula-963 (New England

Nu-clear). DNP was analyzed in 3.5% acrylamide-0.5% agarose gels (34). Electrophoresis was conducted in thecoldat 20mAfor6h. Afterelectrophoresis, gels were prepared for fluorography (4, 20). 32P-labeled

DNA fagments generated by micrococcal nuclease

digestionof nucleifrom infected cells werehybridized toSmaIadenovirus 2 DNAfragmentsaccording to a method described previously (29), and the hybrids

werevisualizedbyautoradiography.

Invitro RNA synthesisbyisolated nuclei. Pu-rified nucleiwereassayed for RNA syntheticcapacity Voi. 29,1979

http://jvi.asm.org/

(3)

890 SERGEANT, TIGGES, AND RASKAS as previously described (37, 38), using 500 ,uCi of

[3H]UTP (New England Nuclear, [6,5-3H]UTP) per ml (35 Ci/mmol). After 15min of incubation at 37°C, the mixture was diluted with an equal volume of 0.05 M Tris-hydrochloride (pH 7.9), 0.2 M NaCl, 0.01 M EDTA, 2% N-laurylsarcosine, 200,ug of dextran sulfate perml, 0.002 M UTP, and 300 ,ug of proteinase K (E. Merck; fungal, chromatographically purified and ly-ophilized) per ml. The suspension was pipetted gently, using a large-bore pipette, andincubated overnight at 37°C. RNAs were purified by phenol extraction (39). RNA-DNAhybridization experiments wereperformed with adenovirus 2 DNAimmobilized on 6.5-mm cel-lulose nitrate membranes (type B6, Schleicher & Schuell Co.) (10).

RESULTS

Preparation of nuclei containing

paren-tal viral DNA.Threemethods of nuclear

prep-arationwerecompared (Table 1)before

analyz-ing the physical state of parental viral DNA

earlyininfection. Forallthreemethods, nuclei

weretreated with nonionicdetergentsto remove

cytoplasmic material. Nuclei prepared in iso-tonic(i)orhypotonic (ii) buffers,usedroutinely

forpreparation ofnuclei for in vitro RNA

syn-thesis (9, 38,40), contained the same amountof parental DNA as did nuclei prepared in

su-crose-calcium buffer for micrococcal nuclease digestion (iii). Comparable amounts of viral

RNA were synthesized in vitro with the three

typesof nuclearpreparations (Table 1).

Since nuclei prepared for micrococcal

nu-clease digestion exhibited typical amounts of viral transcriptional activity, we used this methodtostudythe fateandstructureof

paren-tal viral DNA. The subcellular distribution of

parental

32P-labeled

adenovirus2DNA3hafter

infection is shown in Table2.Thepercentageof

parentalDNAfound in nucleirangedfrom14to

21%. To determine if the 32P-labeled DNA in nuclei3hafter infectionrepresentedviralDNA,

nuclear DNAwasanalyzedbyCsClequilibrium

density gradient centrifugation. The results of

such an analysis are shown in Fig. 1. Greater than95% of the32P-labeledDNA had the buoy-antdensity of viralDNA(1.714g/cm3).Asimilar result was obtained when parentalviral DNA

was examined 18 h after infection (data not

shown).Thus, 5%is the upperlimitof viral DNA

moleculesthatmight be covalently attachedto

cellularDNA(16)ortrappedwithcellularDNA, or,alternatively,whichweredegradedand

rein-corporatedintocellularDNA.

Kinetics ofdigestion of intranuclear ad-enovirus 2 DNA by micrococcal nuclease.

The rate at which DNA is renderedacidsoluble wheneucaryoticcell nucleiaredigested by

mi-crococcal nuclease islinear earlyinthe course

TABLE 1. Recovery of32P-labeledadenovirus2 DNA and viraltranscriptional activity in nuclei prepared for micrococcal nucleasedigestion (iii), as

comparedtocrude nuclearpreparations (i,ii)a Intranuclear Viral tran-adenovirus 2 Total scriptional DNA [3H]- activity (% UTP in- hybridiza-Nuclear prepn Ge- corpo- tion of

nomes %of rated [3H]RNA per input (cpm/108) to

adenovi-nu- cells rus 2

cleus DNA)

Isotonic-Tri- 500 15 423,000 1.90 ton (i)

RSB-Triton(ii)b 700 22 490,000 2.60 Sucrose-calci- 600 19 327,000 3.00

um-Noni-det P-40(iii)

aKB

cells

wereinfected with 3x 103

particles

per

cellusing32P-labeledadenovirus 2. After 3 h of infec-tion, nucleiwereisolatedbythe indicated methods,

the amount of intranuclear32P-labeledadenovirus 2 DNA was determined,and the nuclei were incubated invitro for RNA synthesis as described in the text.

[3H]RNAwasextracted from nuclei and hybridized to 10

jig

of adenovirus 2 DNA.

RSB,Reticulocyte standard buffer.

TABLE 2. Subcellular distributionofparental

adenovirus2DNA in culturesharvested early in infectiona

% of infective 'P-la-Fraction beledadenovirus2b

CelIsc. 37-58 (47.5)

Postnuclearsupernatant.16-33 (24.5)

Nuclearwashesd . 1.6-4 (2.8)

Nuclei ... 14-21 (17.5)

aKB

cells

wereinfectedwith

32P-labeled

adenovirus 2atanMOI of 3,000particles per cell. Cultures were harvested 3 h afterinfection,and nucleiwere fraction-atedbythesucrose-calcium method.

bThe numbers give the range of values obtained from sixexperiments. Numbers inparentheses repre-sentaveragevalues.

cKBcells after two washes with cold phosphate-buffered saline.

dNuclear washes include two washes with Nonidet P-40 buffer and two washes with the micrococcal of digestion and gradually reaches a plateau when 50 to 75% of the DNA isdigested (1, 6, 30).

Digestionoftheintranuclear parental

adenovi-rus2 DNAbymicrococcalnuclease in a digest

buffercontaining1.0 mMCaCl2 yielded kinetics of degradation similar to those described for

eucaryotic chromatin; a plateau was reached when about 60% of the viral DNA wasrendered

acid soluble (Fig. 2). The digestion kinetics of

purified adenovirus 2 DNA mixed with

http://jvi.asm.org/

[image:3.503.272.463.94.240.2] [image:3.503.274.465.356.434.2]

(4)

ADENOVIRUS 2 DNA INTRANUCLEAR STRUCTURE 891

0.2

0

`0

oi

b

I

U

0L

(N4

0.1 -10

-1-68

-5

-1.66

10 20 30 40 50

FRACTION NUMBER

FIG. 1. CsCIdensitygradientanalysis of intranu-clear32P-labeledparentaladenovirus2DNA. Nuclei

containing30 pg ofDNA wereprepared 3 h after infection. The nuclear DNAwascentrifugedto

equi-librium in 10-mlCsClgradients (32). Centrifugation

was at100,000Xgfor69hat15°CinaSpinco50Ti rotor.Gradientfractions(0.2 ml) werecollected,and

densitywasdeterminedby refractiveindex.Symbols: *, 32P-labeledadenovirus 2DNA;0, opticaldensity at 260 nm(OD260)ofcellularDNA; , CsCldensity.

Under theseconditions, the desitiesofadenovirus2 DNA and KB nuclear DNA are,respectively, 1.714 and1.698g/cm3.

infected KBcell nuclei weresignificantly

differ-ent. After30 s ofdigestion, as muchas 55% of the

32P-labeled

viral DNA wasdigested, as com-pared to 4% of the intranuclear parental viral

DNA (Fig. 2). The digestion of purified viral DNA reached a plateau when about 90% had

been rendered acid soluble. As afurthercontrol

purified

32P-labeled

viral DNA wasmixedwith

H-labeledKB cell chromatin andsubjected to digestion. Under these conditions viral DNA was degraded at least as fast as in the absence of chromatin. Forexample,after 1min ofdigestion (micrococcalnuclease, 1

U/,ug

ofDNA), 18%of the

[3H]chromatin

countsperminute were

sol-ubilized as compared to 70% of the

3"P-labeled

viralDNA.

Relationship betweenMOI and the

intra-nuclear concentration of nuclease-resist-antviral DNA. To determine the effect of MOI on intranuclear viral DNA, three parameters

weremeasuredatvarious multiplicities(Fig.3). The amountof intranuclear viral DNA increased linearly up to 4,000 particles per cell. As

meas-uredin isolated nuclei, the amount of viral

tran-scriptional activity increased uptoMOIs of 3,000 to 4,000 particles per cell (Fig. 3A). However, the increase was not directly proportional to MOI.

Nucleipurified from cultures infected at var-iousmultiplicities were digested with micrococ-calnuclease (Fig. 3B). The kinetics of digestion were the same irrespective of the intranuclear concentration ofparental viral DNA. We con-clude that asimilar percentage of the viral DNA is in anuclease-resistant structure irrespective of theMOI.

Nuclease resistance of parental viral DNA atvarious times during the course of infection. The fate ofparental viral DNA dur-ing the course of infection was analyzed by har-vestingcultures andpurifying nuclei at various times in infection (Fig. 4). Parental viral DNA

was detectable in nuclei by 1 hafterinfection,

Z 100 In E 80 0 z

c 60

-D 40

6

e 20

u

4

0 10 20 30 40 50

TIME OF DIGESTION (Min) 60

FIG. 2. Kineticsofmicrococcalnuclease digestion of32P-labeledparentalviralDNA inpurified KBcell

nuclei. Nuclei were digested with micrococcal

nu-cleaseasdescribed in thetext.Aliquotswereremoved fromthedigestattheindicatedtimes, and theamount

of32P-labeledadenovirus 2DNArenderedacid sol-uble in 10% trichloroacetic acid was determined.

Symbols: 0, infectious intranuclear3P-labeled

pa-rental viralDNA;0,deproteinized 3P-labeled

ade-novirus 2 DNA mixed with mock-infected KB cell nuclei.

I I a

.

a

I

VOL. 29,1979

IF

4

9

http://jvi.asm.org/

[image:4.503.49.234.52.378.2] [image:4.503.252.446.373.545.2]

(5)

892 SERGEANT, TIGGES, AND RASKAS

10

8

6

0

x

c-uJ z

o

4 - -4 ae

j2I

0 2 4 10

ADENOVIRUS-2 PARTICLES/ CELLx10-3

4c

uJ )Z Zo 4c

a C~4

'.-I;

ZD)

uJ>

,£O

i

_JuJ

00

U)< 0

47

be

Adenovirus-2 Particles/Cell

o400 * 3,000 O 4,000 a10,000

0 20 40 60 80 100

[image:5.503.122.406.70.247.2]

TIME OF DIGESTION(Min)

FIG. 3. EffectofincreasingMOIontheamountofintranuclear viralDNA, virus-specificRNAsynthesis in

vitro,and nuclease-resistant viral DNA. Nucleiwerepurified from32P-labeledadenovirus2-infectedcellsby methodiii. The MOIwaseither400, 3,000,4,0X0,or10,000adenovirus 2 virions percell. (A)*, Recoveryof intranuclear3P-labeledadenovirus 2DNA;0,synthesis ofvirus-specificRNAs in nuclei incubated in vitro.

(B)Kinetics of micrococcalnucleasedigestionofintranuclear adenovirus 2 DNA atdifferentMOIs.

1 2 4 6 8

HOURS AFTER INFECTION

x z

0

F-C~4VI

10 0

>z 80

CI-A

400

20 z2

20 ZL

0

FIG. 4.Micrococcal nuclease digestion of intra-nuclearparental adenovirus2DNAatdifferenttimes duringproductive infection.Nucleiwereisolatedby the sucrose-calciummethodfrom KB cellsinfected with 32P-labeled adenovirus2particles.Cultureswere

harvestedattheindicatedtimesduring the produc-tiveinfection. Symbols: 0, Recoveryofintranuclear

3P-labeledparental adenovirus 2 DNA; 0, limit

digestion of intranuclear3P-labeled parental ade-novirus2DNAby micrococcalnuclease.

although maximal transfer to nuclei was not

achieved until4h. Theamountofintranuclear viral DNA remained constant until the latest timesexamined, 10 hafter infection. For each time sample the percentage ofparental 3P-la-beled DNA resistant to micrococcal nuclease digestion (after limit digestion)wasdetermined. Forallsamples tested thepercentwasconstant

(Fig. 4), and the kinetics of digestion were the same (not shown).

Electrophoresis of DNA fragments

gen-erated by micrococcal nuclease digestion.

When nuclei from humancellsaredigested with

micrococcal nuclease, the DNA fragments are

integralmultiplesofamonomeric unit of about

185basepairs, consisting ofacoreof

nuclease-resistant

materialabout140base pairs in length and a heterogeneous nuclease-sensitive linker withameanlength of about45base pairs (19,

21, 23). The sizes of cellular and viral DNAs

resistanttodigestionwere compared. KB cells

werelabeled with[3H]thymidinefor72hbefore

infection with3P-labeled adenovirus 2. At3h

afterinfectionnucleiwerepurified and prepared

for micrococcal nuclease digestion. Samples

weretakenfrom the digestion mixtureatvarious

times, and the DNA was extracted and then

fractionatedon2%agarosecylindrical gels (Fig. 5).

Asthe digestion proceeded, thesizes of both

cellularandviraloligomersdecreased.The

high-molecular-weight fragments diminished, and

DNA shiftedtothe mobilitiesofmonomers and

material smaller than 140base pairs.The viral

oligomers were somewhat smaller than

corre-spondingcellular units. Also, the viral nucleo-somes shiftedto smaller sizes at a more rapid

rate than the cellular chromatin. The most

prominent difference between cellularand viral patterns was the rate of accumulation of sub-monomerfragmentsattheexpenseofmonomer.

These submonomers migrated faster than the dyefront (slice 80). Higher-resolutiongels (not

0

K

U)

LU

.-z

U)

uJ

0

0 4

03)

Inv. >..J

z X

0 Z ILL

c

http://jvi.asm.org/

[image:5.503.70.262.308.475.2]

(6)

m A

a I Mon.

i~~~~~~~

w

.

to nuclease digestion were also displayed as

DNP. Nucleiwereprepared fromunlabeled KB

8.0 cells infected for3h with [3H]thymidinelabeled adenovirus 2. A sample was prepared after 10 6.0 minof digestion; electrophoresiswasperformed

before deproteinization (Fig.6).Aswasthecase

4.0 with viralDNA from infectednuclei,viral DNP

dimers and trimers migrated in the gel more

2.0 rapidly than cellular forms, reflecting the

re-duced DNAcontentofthesestructures (34).

Viral 'nucleosomes" are not formed by association of nuclear proteins with DNA

during digestion. To evaluate the possibility

°.0 that cellular nuclear proteins had exchanged

withviral genomes during the course of diges-5.0 tion, 3H-labeled cellular chromatin wasmixed

4.0 2.0 °

x

0-8.0

6.0

-

TC

Tv

X

4.0

2.0

_-Dc

in D

35 15mi.

8.0

3.0-x 2.56.

~,2.0

1.5 ~~~~~~~~~4.0

1.01 2.0

0.5-20 40 60 80 100

SLICE NUMBER

FIG. 5. Micrococcal nucleasedigestionpatternsof

DNA from3H-labeled KB cells 3 h after infection

with 32P-labeledadenovirus2. NucleifromKB cells

infectedwith 3,000virusparticlespercellwere

pre-pared and digested with micrococcal nuclease as

described. Atthe indicated times, sampleswere re-movedfor phenol extraction andelectrophoresis in 2% agarosecylindrical gels.(A)1-mindigest,1%acid soluble;(B)3-mindigest,10% acidsoluble;(C)5-min digest, 15% acidsoluble;(D)15-mindigest,20%acid soluble.

shown) showedthatthese fragments ranged in

size from 20 to 80basepairs.

The parental viral DNA fragments resistant

[image:6.503.52.224.54.525.2]

Mq-

X-Mc

FIG. 6. Electrophoretic analysis ofDNP released fromadenovirus2-infectednucleidigestedwith mi-crococcal nuclease. KB cells were infected with [3H]thymidine-labeledadenovirus 2(3,000 particles

per cell). Nuclei were isolated anddigestedas de-scribed in thetext.Electrophoresiswasinacomposite

3.5% acrylamide-0.5% agarose gel. M,, Mc:

Nucleo-somemonomers,viralandcellular;D,T:nucleosome

dimersand trimers. 6.0

4.0 2.0

8.0 0

E 6.0 I 4.0

1

2.0

8.0

6.0

4.0

2.0

-I

-

1-VOL. 29,1979

http://jvi.asm.org/

[image:6.503.251.438.248.578.2]

(7)

withphenol-extracted, "2P-labeledadenovirus 2

DNA and then digested with micrococcal

nu-clease. The DNA was extracted andsubjected

toelectrophoresis. To maximize the possibility of detecting resistant DNA, the ratio of 32p-labeled DNAto 3H-labeled DNA was 20times

greaterthan thatpresentinisolated nuclei. The results areshown in Fig. 7A through C. Atno

time during the course of digestion did there

appear 32P-labeled DNA fragments of discrete sizes in thegelregionscorrespondingto nucleo-somal DNA.In a separateexperiment(Fig.7D), purified 3H-labeled cellular DNA was mixed withpurified 32P-labeled adenovirus2 DNA. A

similar resultwasobtained.Thus,it is clear that

thenucleosomemonomersandoligomersof viral

DNA inthenucleusare notformedbyexchange

duringdigestion.

Allregions of the adenovirusgenome are

represented in viralnucleosomes. To deter-mine ifall regions of the adenovirusgenomeare

represented in viral nucleosomes, sequences in

nucleosomes were analyzed by DNA-DNA

hy-bridization.Amicrococcalnuclease digestof

in-fected nuclei containing 32P-labeled parental DNA was fractionated on an agarose slab gel. Ethidium bromide staining of the gel showed the typical pattern of cellular DNAfragments

migrating asintegral multiplesofamonomeric unit (Fig. 8, top). The viral nucleosomal DNA was then hybridized tothe setofadenovirus2 DNAfragments producedbycleavage wth endo R-SmaI. The nonradioactive viral DNA frag-ments werefractionatedon aslabgel, denatured,

and then transferred to a nitrocellulose sheet

(29).Thefractionatednucleosomal DNA in the slab gelwas annealed directly tothe DNA on

thenitrocellulose sheet. Thepattern of

hybridi-zation wasdisplayed by autoradiography.

Intra-nuclear viral DNA present as monomers and

monomerrepeatsaslargeaspentamers hybrid-ized toalltheSmaIfragments.The same results were obtainedaftera limitdigestion of the

nu-clei, when the DNA resistant to micrococcal nuclease migrated as a single monomer band

(notshown).

Parental viral DNA retains a chromatin-like structure after viral DNA synthesis begins.It was ofinteresttodetermine whether

the chromatin-like structure of parental viral

DNA wasalso observed after the onset of viral DNA replication. To examine this possibility, 3H-labeled KB cells infected with 32P-labeled adenovirus 2 were continued in cultureuntil 18 hafterinfection;the cellswerethenharvested,

and nuclei were prepared and digested. The DNA wasextractedandsubjectedto

electropho-resis (Fig. 9). As withnuclei 3 h after infection

(Fig.5), achromatin-like structureofviral DNA

J. VIROL.

[image:7.503.286.452.51.537.2]

20 40 60 80 100 SLICE NUMBER

FIG. 7. Micrococcal nuclease digestpatterns of DNAfrom 3H-labeled KBcell chromatin mixed with phenol-extracted,32P-labeledadenovirus 2 DNA. The isolation of chromatin anddigestion by micrococcal nuclease are described in the text. Electrophoresis wasin2%oagarosecylindrical gels. (A)1-mindigest, 3Hcpm,0.1%acid soluble;32pcpm,22%oacid soluble. (B)3-mindigest, 3Hcpm,3%acidsoluble; 32pcpm, 37% acidsoluble. (C) 5-mindigest,3Hcpm, 8% acid soluble; 32P cpm, 47% acid soluble. (D) Phenol ex-tracted, 3H-labeled KB cell DNA mixed with

32p-labeled adenovirus2and digested to 25%acid-soluble 3Hcpm and 65%32pcpm.

http://jvi.asm.org/

(8)

Direction of Migration

MDT Multimers

I I

Ethidiurm Bromide Staining

I Hybridization to DNA fragments

M D T

Smo I Adenovirus-2

[image:8.503.102.392.63.469.2]

_W,..

FIG. 8. Hybridizationtransferof DNA in nucleosomestonitrocellulose-bound endo R.SmaI fragmentsof adenovirus 2 DNA. Nuclei isolatedfrom3P-labeledadenovirus 2-infected KB cells (3 h after infection, 2,000 particlespercell)weredigested for5min withmicrococcal nuclease(500pgof DNAperml; 30Uofenzyme

perml). DNA fragments purified from the digestwere run on a1.4%agaroseslab gel (top). 'P-labeledDNA

intheslabgelwasthenhybridizedtononradioactive adenovirus2SmaI fragments (A-J) which previously hadbeenseparatedon a1%agaroseslab gel andtransferredtonitrocellulose (20). For the hybridizationstep

theradioactivegelwasorientedsuch that thedirection ofmigrationwasatright anglestothe direction of migration of the DNA which had been transferredtonitrocellulose. (Bottom)Pattern of theautoradiogram after the hybridizationstep.M, D, and T designate nucleosomemonomers,dimers,andtrimers, respectively.

wasobserved.Inthesepreparationsalso,a

sub-monomerfractionwasobserved. DISCUSSION

Basedonthe studiespresentedhere,we con-cludethatparentaladenovirus DNA molecules

enteringthe nucleusearlyininfection (3 h) are organizedinapatternsimilar to the nucleosomal

structureofcellular chromatin(Fig. 5); parental

DNA found in the nucleusatlate timesalsohas achromatin-likestructure (Fig. 9). Digestion of nuclei from infected cellswithmicrococcal nu-clease releasesviraland cellularDNPcontaining

DNAsequenceswhicharemultiples ofa

mono-mericunit. It is clear that at least 40%ofthe

viral DNA exists in thisorganization; after 15 min of digestion 40% of the 32P-labeled viral

countsperminute aresmaller thanfull-length

Sma I

Fragments

A _

B C

D

--E

-F

F

G-H VOL. 29,1979

http://jvi.asm.org/

(9)

T0

x

I

SLICE NUMBER

FIG. 9.Micrococcalnucleasedigestionpatternsof DNAfrom 'H-labeled KB cells 18h after infection with3P-labeled adenovirus2. Theinfection, diges-tion,andelectrophoresisconditionsaredescribedin thelegendtoFig.5.(A) 1-min digest, 4% acidsoluble; (B) 3-min digest, 15%acidsoluble; (C)5-min digest, 17% acidsoluble;(D) 15-min digest,20% acid soluble.

adenovirus 2 DNA but equaltoorlarger than

cellularmonomers. Anintracellular adenovirus chromatin structure was notobserved in

elec-tron microscopy studies of nuclear replicative DNAs extractedlate ininfection (18). However,

ourstudy has focusedexclusivelyonthe fate of parental DNA, which is a minor component at late times and might have a structure different from that of themajority of the DNA molecules. The viral nucleosomal products do differ in size from the cellular nucleosomes. Viral

oligo-merDNAs are smaller than the corresponding cellular DNAs and submonomer fragments ac-cumulate more rapidly, suggesting that viral chromatin is moresensitive tomicrococcal

nu-clease both in the linker regionand withinthe

core (Fig. 5 and9). It isperhapsnotsurprising

that viral DNA sequencesare moresensitiveto

nuclease than bulk cellular chromatin. It has

beenshown thattranscriptionallyactiveregions

of hen oviduct chromatinareselectively released

bymicrococcal nuclease(3),and,intermsofthe proportion of the genome engaged in transcrip-tion, viral sequencesaremuchmoreactiveboth

earlyand late than cellulargenes.

DNA inthe submonomerregionwasalso gen-erated when deproteinized DNA was digested (Fig. 7D). Because DNA fragment mobilities tended to be compressed inthis region of the

gels, itisimpossibletodistinguishbetweentwo

alternativestructuresfor the submonomer

frac-tion. Discretefragmentsthat aremultiplesof 10

basepairsaregenerated bymicrococcal nuclease

digestion ofchromatin (31).Alternatively,

ran-domly sized products would be expected from

thedigestionofprotein-freeDNA.Experiments

to characterize further the submonomer DNA arein progress.

Thequantitation of viral chromatin at

differ-ent MOIs has suggested differences between

amounts of chromatin and viraltranscriptional

activity (Fig. 3). Theamountof viral chromatin increased in proportion to MOI up to at least 4,000particlespercell. Viraltranscriptional

ac-tivity, asmeasured in isolatednuclei, appeared

toreachmaximal levelsatsomewhatlower

mul-tiplicities. Thus, quantitation of chromatin

structureisnotnecessarilyadirect reflection of viraltranscription. Viral templates may exist in chromatinstructuresbutnotbeutilized for viral

transcription.Infact,it may bethatonlyasmall

fraction of the viral chromatin isactively

tran-scribed. Although limitations in cellular RNA polymerase concentrations might be responsible forlimitedtemplateutilization,previous studies suggest that other factors may influence

tem-plate-enzyme interaction and thereby determine thetemplates actively transcribed (37).

Analysis of viral chromatin has provided

ad-ditionalevidence for the utilizationofparental genomes in virusreplication. Maximum transfer

of parental DNA to the nucleus required 3 h

from thebeginning ofvirusadsorption, in

http://jvi.asm.org/

[image:9.503.67.254.46.523.2]

(10)

mentwithpreviousstudiesofviralpenetration

(24).Theviral chromatinwasshowntobe pres-ent in nuclei as late as 18 h after infection; at all timesexamined, a constantproportion of paren-tal DNA was resistant to nuclease digestion. Thus, the chromatin-like structure ofparental

DNA waspreserved at least 11 to12hafter viral DNA replication had begun. An earlier study

haddemonstratedthepreservationofa

substan-tial amount ofparentalDNAuntillate in infec-tion(36).Since thelatephase ofviral

transcrip-tion is initiated at the time DNA replication

begins (14), ourfindings demonstrate that this

late transcriptional patterndoesnotrequireor

coincide with theeliminationofchromatin struc-turescontainingparentalDNA strands.

Although allregionsof the viral genomeare

represented inchromatin before the beginning

of viralDNAreplication(Fig. 8),it iscertainly

possiblethatregions containingearlygeneshave

astructuredifferent from those genesexpressed onlyatlate times(11,12,27).Thepolypeptides

responsible forthechromatin-likeorganization

ofparental viral DNA in the nucleus havenot

been identified. Structural analysis of virions

suggestsarole fortheinternal

proteins

VIIand

V (7). However, the ability toinfect cells with

nakedDNA (13, 25) indicates that virion

pro-teins are notobligatory forearlyexpression of

the viralgenome. Thus, arole for cellular

his-tones in the early viral chromatin cannot be

excluded.

ACKNOWLEDGMENTS

Weappreciatethetechnicalassistanceof MarkTelle. Com-puter time fordataanalysiawasgenerously provided byM. McDaniel.

Thisstudywaasupported byPublic HealthServicegrant CA-16007 from theNational Cancer Institute and American CancerSocietygrantVC-94.Cell culture mediawereprepared

in a Cancer Centerfacilityfundedbythe National Cancer Institutes. Thisstudywasalso supported bythe following companies Brown&WillismnTobaccoCorp.;Larus and BrotherCompany, Inc.;LiggettandMyers,Inc.; Lorillard,a

Division ofLoewsTheatres,Inc.; Philip Morris, Inc.; RJ. Reynolds Tobacco Co.; UnitedStates TobaccoCo.;and To-bacco Associates, Inc. Thisinvestigation wassupportedin partbyNationalInstitutesof HealthResearchService Award GM 07067,TrainingPrograminCellular and Molecular Biol-ogy,fromtheNationalInstitute ofGeneral MedicalSciences.

LITERATURE CITED

1. Axel,R. 1975.Cleavageof DNA innuclei and chromatin withstaphylococcal nuclease. Biochemistry 14:2921-2925.

2. Axel,R.,W.Melchior,Jr.,B.Sollner-Webb, andG. Felsenfeld.1974.Specificsites of interaction between histonesand DNA in chromatin. Proc.Natl.Acad.Sci. U.S.A. 71:4101-4105.

3. Bloom,K.S.and J. N. Anderson.1978.Fractionation of hen oviduct chromatin intotranscriptionallyactive and inactiveregionsafter selectivemicrococcalnuclease digestion.Cell 15:141-150.

4.Bonner, W.M*,and A.Laskey.1974. Afilmdetection

method for tritium-labeledproteins and nucleic acids in polyacrylamidegels. Eur.J.Biochim.46:83-88. 5. Compton, J. L,AL Bellard,and P. Chambon. 1976.

Biochemicalevidenceofvariabilityin the DNA repeat lengthin the chromatinofhighereukaryotes. Proc. NatLAcad.Sci.U.SA73:4382-4386.

6.Compton,J. I., R.Hancok,P.Oudet,and P. Cham-bon. 1976. Biochemical andelectron-microscopic evi-dence that thesubunitstructure of Chinese-hamster-ovary interphase chromatin is conserved in mitotic chromosomes.Eur.J.Biochem.70:555-68.

7. Corden,J.,HLM.Engelking,and G. D.Pearson. 1976. Chromatin-like organization of the adenovirus chro-mosome. Proc.Natl. Acad. Sci.U.SA.73:401-404. 8.Craig,E.A.,andH. J.Raskas. 1974. Two clasesof

cytoplasmic viralRNAsynthesizedearlyinproductive infection with adenovirus2. J.Virol.14:751-757. 9.Craig,E.A.,andE.J.Raskas.1976. Nucleartrmncripts

largerthan the cytoplasmic mRNAs are specified by segments of the adenovirus genecodingforearly func-tions. Cell8:205-215.

10. Craig, E. A., S. Zinmer, and H. J. Raskas. 1975. Analysis of early adenovirus2 RNAusingEcoR-R1 viralDNAfragments J.Virol.15:1202-1213. 11.Flint, S. J., andH. AL Weintraub. 1977. An altered

subunit configuration associated withtheactively tran-scribedDNAofintegrated adenovirus genes. Cell 12: 783-794.

12. Garel, A., and R. Axel. 1976. Selective digestion of tanscriptionallyactive ovalbumin genesfromoviduct nucleiProc. Natl.Acad. Sci.U.S.A.73:3966-3970. 13. Graham, F.L,and A.J. Van der Eb.1973. A new

techniquefor the assay ofinfectivity of human adeno-virus5DNA.Virology52:456-467.

14. Green,ML,J.T. Parsons,ML Pina,K. Fujinaga,H. Caffier,andLLandraf-Leurs 1970.Transcription of adenovusgenes in productivelyinfectedand intrans formedcells. ColdSpringHarborSymp. Quant. Biol. 35:803-818.

15. Griffith,J.D. 1975.Chromatin structure: deduced from aminichromosome. Science187:1202-1203.

16. Groneberg, J.,D. T. Brown, and W.Doefler. 1975. Uptake and fate of the DNA ofadenovirus type 2 in KBcells. Virology 64:115-131.

17. Hancock,R. 1974.Interphase chromosomal deoxyribo-nucleoproteinisolatedas adiscrete structure from cul-tured cells. J. Mol. Biol.86:649-663.

18 Keer, C.,0. Brian,F.Perrin, and J.Wilhelm. 1978.Structuralanalysisofviralreplicative intermedi-atesisolated from adenovirustype 2-infected HeLa cell nuclei. J. Virol.26:364-379.

19.Kornberg, R. D. 1977. Structureof chromatin. Annu. Rev.Biochem.46:931-954.

20. Laskey,R.A., andD. Mills. 1975.Quantitative film detection of3H and 4Cinpolyacrylamidegels by fluo-rography.Eur.J. Biochem.56:335-341.

21.Lohr,D.,J.Corden,K.Tatcheli R. T.Kovacic, and K. E. VanHolde. 1977.Comparative subunit structure of HeLa, yeast and chicken erythrocyte chromatin. Proc.Natl.Acad. Sci.U.S.A.74:79-83.

22.Lohr, D.,R. T.Kovacic,and K. E.VanHolde. 1977. Quantitatinganalysisof the digestion ofyeast chroma-tinbystaphylococcalnuclease. Biochemistry 16:463-471.

23. Lohr, D.,K.Tatchel,and K. E. Van Holde.1977. On the occurrence of nucleosomephasing in chromatin. Cell12:829-836.

24. Lonberg-Holm, K., and L. Philipson. 1969. Early eventsof vimscellinteraction in an adenovirussystem. J.Virol.4:323-338.

25. Louie, A. J. 1974. The organization ofproteinsinpolyoma and cellular chromatin. Cold Spring Harbor Symp. Quant. Biol.39:259-266.

VOL. 29,1979

http://jvi.asm.org/

(11)

898 SERGEANT, TIGGES, AND RASKAS 26. McGrogan, M., andH. J.Raskas.1977.Species

identi-fication andgenomemapping of cytoplasmicadenovirus type2RNAs synthesizedlate in infection. J. Virol.23:

240-249.

27.Nicolson, M.O., and R.M.McAllister.1972.Infectivity ofhuman adenovirus1DNA. Virology 48:14-21. 28. Oudet, P., M.Gross-Bellard,and P. Chambon.1975.

Electron microscopic and biochemical evidence that chromatinstructureisarepeating unit. Cell4:281-300. 29. Sato, S., C. A. Hutchinson m,and J.I.Harris.1977.

Athermostable sequencespecific endonuclease from thermusaquaticus. Proc. Natl. Acad. Sci. U.S.A.74:

542-546.

30. Sollner-Webb, B., and G. Felsenfeld. 1975. A

compar-ison ofthedigestion of nuclei and chromatin by staph-ylococcal nuclease. Biochemistry 14:2915-2920. 31. Sollner-Webb, B., W. Melchior, Jr., and G.

Felsen-feld. 1978. DNAse I, DNAse II and staphylococcal

nucleasecut atdifferent,yetsymetrically located, sites

onthenucleosomecore.Cell14:611-627.

32. Tibbets, C., K. Johanson, and L Philipson. 1973. Hydroxyapatite chromatography and formamide dena-turationofadenovirus DNA. J. Virol. 12:218-225. 33.Tibbets, C., V. Pettersson, K. Johansoon, and L

Philipson. 1974. Relationship of mRNA from produc-tively infected celistothecomplementary strands of adenovirustype2 DNA. J.Virol. 13:370-377.

34.Todd, R. D.,and W. T. Garrard. 1977. Two dimensional electrophoresis analysis of polynucleosomes. J. Biol. Chem. 252:4729-4738.

35. Varshavsky,A.J.,V. V.Bakayev,P. M.Chumackov, and G. P.Georgiev.1976. Minichromosome of simian

virus40:presenceof histoneHi.NucleicAcids Res. 3: 2101-2113.

36.Vlak,J.M.,T. H.Rozijn,and F.Spies.1976. Replica-tion ofadenovirus type5DNA inKBcells:localization and fate ofparental DNA during replication. Virology 72:99-109.

37. Weinman,R.,J. A.Jaehning,H.J.Raskas,and R. G. Roeder.1976. ViralRNAsynthesis and levels of DNA-dependent RNApolymerases during replication ofadenovirus2.J.Virol. 17:114-126.

38. Weinmann,R.,H.Raskas,and R. Roeder. 1974. Role ofDNA-dependent RNA polymerase II andIII in of the adenovirusgenomelateinproductiveinfection. Proc. Natl. Acad. Sci. U.S.A. 71:3426-3430.

39. Wilhelm,J.,0.Brisoni,C.Kedinger, and P.

Chan-bon. 1976. Characterizationofadenovirus type2 tran-scriptional complexesisolated frominfectedHeLacell

nuclei. J.Virol. 19:61-81.

40. Zimmer,S.G.,C. J.Goldenberg,D.P.Carlson,E. A.

Craig, and H. J. Raskas. 1978. Size distribution of polyadenylatedadenovirus 2 RNAsynthesizedin iso-lated nuclei. Biochemistry 17:4207-4213.

http://jvi.asm.org/