Replication process of the parvovirus H-1 V. Isolation and characterization of temperature-sensitive H-1 mutants.

Copyright01976 AmericanSocietyfor Microbiology PrintedinU.S.A.

Replication Process

of

the Parvovirus

H-1

V. Isolation

and

Characterization of

Temperature-Sensitive H-1 Mutants

Defective in Progeny DNA

Synthesis

SOLON L. RHODE III

Putnam Memorial Hospital Institute for Medical Research,Bennington, Vermont05201 Receivedforpublication25August 1975

Twotemperature-sensitive mutants of the parvovirus H-1 have been isolated

and characterized. Thesemutantsaredistinguishable

by

the immunofluorescent

staining of cells infected by them and by the thermolability of their

hemag-glutinins. Under restrictive conditions both mutants synthesize a missense

capsid protein defectiveinhemagglutination. Synthesisofthe viral DNA strand

was shown tobe diminished for both wild-type virus andthese mutants at the

restrictive temperature of 39.5C, but themutants were more defective than the

wild type at the less restrictive temperature of 38 C. Replicative form DNA

replicationwasnot decreased inthesemutants. It is proposed that H-1 requires

one ofthecapsid proteinsforthesynthesisofthe single-strandedprogeny DNA.

Replication ofthesingle-stranded DNA

par-vovirus H-1 involves the synthesis and

replica-tionof adouble-strandedreplicativeform (RF)

DNA(8).Progenyviral strand (V) DNAis then

synthesizedwith thecomplementarystrand(C)

DNA in the RF as template(8, 10).

Viral hemagglutinin, antigen, and the viral

proteins VP1 and VP2 have been shown to

dependonpriorDNAsynthesis, termed

hemag-glutinating (HA)-DNA synthesis,anevent that

takesplace ataspecific time nearthe end of S

phase of the infected cell (5, 9). Viral protein

synthesis and theinitiation ofRF DNA

replica-tion occur concomitantly shortly after

HA-DNAsynthesis (10). Although the initiation of

RF DNA replication was shown to require

proteinsynthesis, it was notdetermined ifviral

protein synthesis was required for RF DNA

replication or for progeny DNA synthesis (10).

The present study describes the isolation and

characterization of two temperature-sensitive

(ts) mutants ofH-i. The mutations occur in a

capsid

protein, a protein which isrequired for

progeny DNAsynthesis butnotRFDNA

repli-cation.

MATERIALS AND METHODS

Cell culture. The cells used inthisstudy are the SV40 transformed human newbornkidney (NB)cells routinely used for the plaque assayofH-1orhamster embryo cellsaspreviously described (6, 9). Thecells werecultivated inEagle minimum essential medium (FlowLaboratories, Autopowminimum essential me-diumEagle, modified) supplemented with10% heat-inactivated calf serum, glutamine, penicillin (10

Isg/

ml), streptomycin (5

,g/ml),

and sodiumbicarbonate

topH 7.6. Parasynchronous cultures, used for some purposes,werepreparedby incubationinculture me-diumMAG containing methotrexate (0.5gg/ml), ade-nine (13

gg/ml),

and Eagle minimum essential me-dium nonessential amino acidsfor 16 to 20hat37C. The methotrexate block was reversedby removing the medium, washing once with Hanks balanced salt solution, and refeeding with culture medium con-tainingthymidine (TdR) at 10-iM.

Virus. Wild-type H-1 virus was propagated in hamster embryo cells and purified as previously described (9).

Infection of cells. NB cultures were infected by inoculating the monolayer with 0.1 or 0.2 ml of virus stock per60-mmpetri dish and adsorbing it at 33 C for 30 min. The cultures were washed twice with Hanks balanced salt solution, medium was added, and the cultures were incubated at the desired tem-perature.

Plaque assay. The plaque method ofvirusassay has been described (6). This method was modifiedin thelengthofincubationbefore the neutral red over-lay, both at thepermissive temperature (33 C) and

nonpermissive temperature (39.5 C) used here to

define the conditional mutants ofH-1. The overlay with neutral redwasmadeonday9 forthe assayat 33 C andon day6 forthe assay at 39C. Plaqueswere counted 16 to 24h aftertheoverlay with neutral red. Wild-typevirususually titeredasefficientlyat 39.5C as at 33C, but modest reductionsintiterrangingas highas 75% of the titer at 33 Cwerecommon.

Mutagenesis. H-1 virus (4 x 109PFU/ml), puri-fiedaspreviously described (9),wastreated with0.25 Mhydroxylamine asoutlined byTessman for bacte-riophage S13 (11). The reaction mixture contained onepart H-1 inH,O, twoparts0.1 MP04(pH 6.0),

and one part1 Mhydroxylamine(made pH6.0with NaOH) plus10-3MEDTA. Thereaction wasstopped

659

on November 10, 2019 by guest

http://jvi.asm.org/

660 RHODE

by diluting an aliquot 100-fold in regular medium

with 10% fetal calf serum, 10% tryptose phosphate broth, and 2% acetone at 4 C. In the absence of hydroxylamine, H-1 infectivity decreased only 20% after 4 h at 37 C. Hydroxylamine inactivated H-1 infectivityexponentially ata rate of 10-1/hof treat-ment to a final surviving fraction of 10- the last tested.

Isolation of ts mutants. A preparation of virus inactivatedtoasurviving fraction of10 wasusedas

the sourceofmutagenized virus.Thisvirus

prepara-tionwaspropagatedinhamsterembryo cells infected

atamultiplicity of infection of 0.1 for 48hat33Cto

dilute nonconditional mutants. With this

mutagen-izedvirusstock,diluted togive approximately 1 to5 plaques per 60-mm petri dish, individual plaques

were isolated from assays at 33 C with a Pasteur

pipette. The plaque isolate was stored in 0.5 ml of

N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic

acid-buffered Eagle minimal essential medium and subsequently propagatedinNB cellsina35-mmpetri

dishor ina microtiterplate for 5daysat33C. The virus and cellswerethen harvested anddilutedtoa

final volume of 3 ml in N-2-hydroxyethylpiperazine-N'-2-ethanesulfonicacid-buffered medium. Viruswas

released from cells by freezing and thawing three times. Aliquots of 0.2 ml ofthe isolates were

inocu-latedonduplicate plates andassayedatthe permis-sive andnonpermissivetemperaturesforplaque

pro-duction. Isolates which showed a 100-fold orgreater titerat 33C thanat39.5Cwereselected for further

study. Prepared in this manner, wild-type isolates

generally produced20to100PFUat39.5C and 40or

moreat33C with undilutedinocula. Isolates

appear-ingtobetemperature sensitiveby thescreening test

were assayed again at the two temperaturesand in appropriate dilutions forquantitative determination oftiters.Insome cases,ts mutantswererepurifiedby

one or more plaque purifications. Mutants were

propagated in NB cells at 33 C once or twice to

producevirus stocks with titersof107or moreatthe permissive temperature.

RESULTS

Replication of wild-type H-1at33 and 39.5

C. The kinetics ofhemagglutinin and infectious

virus production in parasynchronous NB cells

for wild-type H-1 at33 and 39.5C were

exam-ined. The growth curves for HA synthesis and

infectious virus production were similarat-the

two temperatures (Fig. 1, Table 1). If the time

axis for HAsynthesisat33C is transformed by

a factorof0.5, the curves are nearly identical.

Theyieldof infectious viruswasreduced about

50% at39.5C comparedto33C.

Isolation ofH-1 mutantstsl and ts2.

Puri-fied wild-type H-1 was mutagenized with

hy-droxylamine to a survival of 10-1, propagated

onceatlowmultiplicity in NB cellsat33C, and

plaqueswereisolated and screened for

tempera-ture sensitivityasdescribed above. Isolates190

16

12

~ ~ C

33C

39.5 C/ C

<~ ~ ~ 0

4 / 00

0 8 16 24 32

HoursPI

FIG. 1. H-1 HA synthesis in parasynchronous NB cultures infectedwithwild-typeH-i atamultiplicity

ofinfection of5to 10and incubatedat39.5 or33C

from I h p.i. to the time of harvest. Cultures were

harvested with their mediumatvarious timesp.i.and the HA titers weredetermined in the usualmanner

[image:2.491.261.450.58.247.2]

withguinea pig erythrocytes.

TABLE 1. Replication ofH-i attherestrictive and permissive temperatures

Time PFU/culture

(hp.i.) 33C 39.5C

1/2

6.6x

106

6.6x

106

6 2.6x 106 1.6x 106

12 2.4x 106 2.0 x106

24 1.4x 107 1.6x 107

48 5.4x 107 2.0 x 107

120 5.8 x 10' 2.2x 107

aReplicate cultures of NB cells in exponential growth were inoculated with wild-type H-1 at a multiplicity of infection of 5 to 10. After a 30-min adsorption at 33 C, the cultures were washed twice with Hanks balanced salts, and medium was added andincubatedat 39.5 or 33Cforthetimesindicated. Cultureswere harvested, subjected to freeze-thawing three times, and H-1 infectivity was determined by plaque assayat 33C.

and 389, subsequently termed tsl and ts2,

respectively, produced higher titers atthe

per-missive than the nonpermissive temperatures,

asreported inTable 2. Subsequentpassages of

the virus stock maintained this difference. The

ratio of titer at 33C to that at 39.5 C did decline

with some propagations at 33 C, indicating a

possible selective disadvantage for the mutant

virus. It hasnot been determinedasyet ifthe

plaques appearing at 39.5 C are produced by

J.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.491.257.450.344.442.2]

[image:3.491.45.240.72.260.2]

TABLE 2. Isolation ofH-1 mutants tsl and ts2a Virus 33C 39.5C 33C/39.5

PFU/ml PFU/ml C(PFU) tsl

Original-screening CL 0

Original I x 103 0

P2 0

P3 5xlO 1.7x103 2.9.A 104

P4 3.4 x10 4.2x10' 8.1 X 102 Plaque purified two 4.1 x10' 1.2 x 104 3.4 < 103

times P3

Plaquepurified two 3.8 x10' 1.1 X104 3.5x 103 times P3

ts2

Original-screening CL 5

Original 2.5x103 0

P2 5.3 x10' 9.8 x10' 5.6X 10' P3 2.5x 10' 8.5 x 103 2.9x 104 Plaquepurified two 3.1 x10' 1x10' 3.1 x 103

timesP3

aThe infectivity of tsl and ts2 H-1 mutants at the permissive and nonpermissive temperatures at various stages oftheir propagation are presented. Assays used innocula of 0.2ml. but titers are given as PFU/milliliter. CL indicates a complete lysisof the cell layer. The screening test is an assay of the plaque isolate undiluted after its first passage. A passage(P)represents a propagation of the virus for 5 days at 33CinNBcell cultures.

wild-type virus or the mutant, due to a

diffi-cultyinrecovering virus from these plaques.

Characteristicsof thereplicationof tsland

ts2. Anumberof parameters ofH-1replication,

including synthesis of HA-DNA, HA, and H-1

antigen, were examined withtsl and ts2atthe

permissive and nonpermissive temperatures.

Both tsl and ts2 were indistinguishable from

wild-type H-1 for these parameters at 33 C

(Table3).

At 39.5 C, tsl andts2produced little HAby

16hpostinfection (p.i.) inparasynchronous NB

cells (Table 4). When cultures of tsl and ts2

were shifted down to 33 C for various periods,

the HA titer increased at least 16-fold andwas

maximal in2h. The appearance of the HA took

place at 33 C, even when DNA or protein

synthesis was inhibited with arabinofuranosyl

cytosine (araC) (10

Ag/ml)

orcycloheximide (50

Ag/ml),

respectively, during

the shiftdown and

the 15 min preceding the temperature shift.

Sincecycloheximide

probably

does notachieve

a complete inhibition ofprotein synthesis and

viralproteinsynthesisaccountsforonlyasmall

percentage of totalcellprotein

synthesis,

itwas

necessary to confirmthat viral HAsynthesisis

sensitive toinhibition by the drug. To do

this,

the inhibition of wild-type HA synthesis by

cycloheximide (50

gg/ml)

at 37 C in the same

cell system wascompared to that of

actinomy-TABLE 3. Synthesis of HA-DNA, HA, and FA by H-I ts mutants

HA-DNA HA FA

Virus

33 39.5 33 39.5 33 39.5

Wildtype + + + + + +

tsl + + + _ + +a

ts2 + + + - +

-aReplicate parasynchronous NB cultures were in-fected with wild-type H-1, tsl, or ts2 and incubated for 16 h at 39.5 C or 20 h at 33 C. Cultures were harvested for HA determination or fixed for FA staining. HA-DNA synthesis at39.5 C for tsl orts2 wasdetermined by adding araC (10/Ag/ml)tocultures at 39.5 Cat 16 hp.i. for 15 min and then shifting the cultures to 33Cfor 3hin the presence ofthedrug.If theHA titerincreasedtothe samedegreeas acontrol culture receiving no araC, then HA-DNA synthesis had occurredat39.5C preceding the addition of araC.

'Nuclei contained FA material in an abnormal pattern asdescribed inFig. 3.

TABLE4. HAsynthesisby tsl, ts2, and wild-typeH-il Log2HA titer

Time tsl ts2 Wild

type

1. '2hp.i. 4 4 4

2. 39.5C,16hp.i. 8 8 15

3. 39.5C, 16 h p.i. + 1h, 33 C 12 10 15 4. 39.5C, 16 hp.i. + 2h,33 C 14 12 14 5. 39.5C, 16hp.i. + 3h, 33 C 14 12 14 6. 39.5C, 16hp.i.--3h,33C+ 14 12 14

araC

7. 39.5C, 16hp.i. +3h,33C + 13 11 12 Cx

8. 33C, 72hp.i. 16 15 15

aReplicate dishes ofparasynchronous NB cultures

were infected with tsl, ts2, or wild-type H-1 and placedat 33 or39.5C. At the times indicated, cultures wereharvestedforHA determinations or shifted from 39.5to 33C for various times before harvest (cultures 3through7). Cultures 6 and 7 were treated with araC (10 ug/ml) or cycloheximide (Cx) (50 gg/ml) for 15 minbefore andduring the temperatureshiftto 33C. All valuesare the log2 HA titer.

cin D at 5

qg/ml

(1 ug/ml was previously

found to be a potent inhibitor of viral HA

synthesis [9]). The actinomycin D control was

necessary to establish that viral mRNA was

present at the time of addition of

cyclohexi-mide, thus allowing the inference that any

increment in inhibition by

cycloheximide

com-paredtoactinomycinDwasdueto aninhibition

oftranslation. It was found that cycloheximide

caused immediate inhibition ofthe

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.491.248.443.77.161.2] [image:3.491.249.441.307.446.2]

662 RHODE

tion of HA ateither10.5 or12hp.i. On the other

hand, the total HAunitsper culture increased

by eight- and

fourfold,

respectively, by

16hp.i.

inthe presenceof 5

,g

of

actinomycin

Dperml.

Thus HA synthesis was sensitive toinhibition

by cycloheximide. This indicates that

HA-INA, viral mRNA, and the hemagglutinin

protein had been

synthesized

before the

shift-down with tsl and ts2. It appears, therefore,

that the temperature-sensitive

phenotype

isthe

result of an altered polypeptide that does not

assume its active structure at 39.5C.

Thestabilityofthe HAtotemperature

inacti-vation wasdetermined

by

using thevirusstocks

oftslandts2(produced at33C) and wild type

atpH7.2.The results showeda

sharp

tempera-ture transition forthe inactivationofthe HAat

72 to 78 C. The temperature

completely

inac-tivating tsl was reduced about 3 to 4 C

compared to ts2 and

wild-type

virus

(Fig.

2).

The temperature for activation of the HA for

the wild-type H-1 agrees with

previously

pub-lished data (2).

To summarize, we have found that the

con-+

0

-I

-4

1--E

7

-8 2

ts

II

.11

1I

oll

I I

I

6I

5 Is

II II

I I

I t

3 4 %

II % II %

7 11 1.

II % II

t

I'll.I

I

..a

4 68 70 72 74 76 78

[image:4.491.54.244.315.570.2]

0C

FIG. 2. ThermalstabilityofH-i hemagglutininfor wildtype, tsl,andts2.Aliquots(1 ml) of virus stocks diluted 1:10 in Tris-buffered saline, pH 7.2, were

incubated for20minatvarioustemperatures,chilled in ice water, and the hemagglutination titers were

determined. The values shown are the log, of the

surviving fraction of thehemagglutinin relative toa

controlmaintainedat0C.

version ofthe mutant viral

hemagglutinin

pro-teins to an active form is deficient at 39.5 C.

Once the virus is

formed,

the ts2 HA hasa heat

stability

similar to that of the

wild-type

virus

and tsl shows a

slightly

reduced transition

temperature.

Immunofluorescence of tsl- and

ts2-in-fectedcells. H-1

antigen synthesis

wasstudied

inasynchronousNBcultures incubatedat 33or

39.5 C for 24h afterinfectionwithtsl, ts2, and

wild-type

H-1. At 39.5

C,

tsl produced

strongly

positive

fluorescent nuclei (fluorescent

antibody

[FA])

with an abnormal

globular

pattern

of

antigen,

illustratedinFig.3B.Cultures infected

with ts2 at 39.5 C produced little or no FA

staining (Fig. 3D),

asin theuninfected control

culture

(Fig. 3G).

However, infectionwith ts2at

33 C resulted in many FA nuclei with the

wild-type

FA staining pattern (Fig. 3A,

C).

Cultures infected with ts2 for 24 h and then

shifted down to 33 C for 1 h, with and without

araC or cycloheximide as above, showed a

gradual transitiontoFAnuclei withawild-type

staining

pattern.

Sincetsl and ts2 aredistinctly differentwith

regard

to FA staining at the

nonpermissive

temperature and the sensitivity oftheir HAto

heat inactivation, they must contain different

mutations.

It should be noted that antisera

prepared

using

guinea pigswith sodiumdodecyl

sulfate-disrupted purified H-1 as the antigen by the

procedure ofJohnson et al. (3), as well as the

very high-titer antisera obtained from

chroni-cally infected hamsters inoculated with

suble-thal doses of H-1 at birth, produced brightly

positive stainingofts2-infectedcellsincubated

at 39.5 C. Thus ts2 capsid proteins have been

transported to the nucleus under restrictive

conditions. We also observed that nuclei

in-fectedwith tsl underrestrictiveconditionswere

stained in thepatternofwild-type-infectedcells

by

thesodium dodecylsulfate-disrupted

whole-virusantisera andintheabnormal pattern (Fig.

3B) by antisera to purified VP2 (4). Our

anti-VP1serumwas tooweak toestablishitsstaining

patternsintheseexperiments. Thus these

find-ings cannot be usedtodeterminetheidentityof

themutantpolypeptide.

Viral DNA synthesis. The effects of the

mutations of tsl andts2onviralDNAsynthesis

were examined. As described above, HA-DNA

synthesis occurred undernonpermissive

condi-tions for both mutants. It was of interest to

determine ifRF DNA replication and progeny

DNA

synthesis

were affected by these muta-tions.

Intheabsenceof apractical, direct,

quantita---1

J. VIROL.

I

on November 10, 2019 by guest

http://jvi.asm.org/

FIG. 3. Immunofluorescentstaining of wild-typeH-i and tsmutant-infected NB cells. NB cells, grown on coverslips, wereinfected withH-i in the usual mannerand incubated at the restrictive (39.5 C) or permissive temperature(33C).Atthe termination of incubationat 24hp.i., the cover slips were quickly rinsed in Hanks balanced salts andfixedwithacetone 15to20minat 4C. The dried cover slips were stained by the indirect immunofluorescent method using anti-H-i hamster sera (hemagglutination inhibition titer of 210) and

fluorescein-conjugated rabbit anti-hamster immunoglobin. (A) tsl, 33 C; (B) tsl, 39.5 C; (C) ts2, 33 C; (D) ts2,

39.5C; (E)wild type,33C;(F)wild type, 39.5 C;(G) uninfected, 39.5 C. (Magnification x460).

tive measureofviralsingle-strandedDNA

syn-thesis, the extent of incorporation of

[3H]bromodeoxyuridine (BUdR) into the V

strandand itsCstrandwasusedtoestimate the

proportion of V strand to C strands newly

synthesized in RF DNA. Thisprovidesa

quali-tative measure ofthe extent of progeny DNA

synthesis. The method employsashort period of

BUdR substitution for TdR to label the newly

synthesized

DNA andtoallow

separation

ofthe

Vand C strands.

5-Fluorodeoxyuridine

(FUdR)

is used to prevent fluctuations in the

specific

activity ofthe BUdR

incorporated

with respect

toTdR.Vstrand

synthesis

inexcessofC strand

synthesis

cannot be ascribed to

semiconserva-tive replication of RF DNA and is thus a

qualitative

measure ofthe presence of progeny

DNA synthesis (8). Thismethod does have the

VOL. 17, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.491.38.434.57.483.2]

664 RHODE

advantage over a more directmeasure of

prog-eny DNAaccumulation inthat itdetects

prog-enyDNAsynthesis, evenif the progeny DNA is

rapidly degraded for some reason, such as

defective encapsidation.

Infected cultures were density labeled with

BUdR forsubsequentseparation oftheVandC

strands of RF DNA. The relative rates of

synthesis ofeach strandforwildtype, tsl, and

ts2could then be measuredfor comparison. In

hamsterembryo cells infectedat37C, RF DNA

incorporated BUdR asymmetrically with a

higherrate ofsynthesisfortheVstrand thanfor

the C strand (8, 10). This was the expected

result for anRF DNA pool engaginginprogeny

DNAsynthesisbyastranddisplacement

mech-anism in addition to semiconservativeRFDNA

replication. If a mutant is defective inprogeny

DNAsynthesis butnot in RF DNA replication,

the RF DNA will incorporate BUdR intoequal

numbers of V and C strands andinproportion

to their TdR contents

(V-TdR/C-TdR

= 1.15)

(7). A mutant defective in RF replication will

shownoincorporation intoRFDNA.The

exper-iment wascarriedoutby infecting

parasynchro-nous NB cultures withtsl,

ts2,

orwildtype at

the restrictive temperature. At 16 h p.i. the

cultureswere treated with FUdRfor 30min to

exhaust the cell poolsofTdR and then labeled

for 1h with [3H

]BUdR

inthepresence ofFUdR.

Viral DNA was extracted by the Hirt method

andanalyzedbyzonalcentrifugationinneutral

sucrose. The results (Fig. 4) showed increased

labelingofRFDNAfor tsl (andts2,notshown)

relative to the wild-type virus. In a similar

experiment, the region of the sucrose gradient

that contains the RFDNA contained lessthan

1% of the radioactivity for a mock-infected

control culture as that recovered from a

tsl-infected culture andnopeak waspresent.Thus

tsl (and ts2) does not show any defect in RF

DNA replication. In addition, there is some

change inthe sedimentationpatternconsisting

of a decrease in the material sedimenting at

fractions 13 and 14, the expected position of

progeny DNA (8). A DNA peak, which is

usually more prominent than shown here, is

commonly seen at this position in

sedimenta-tion patterns of Hirt extracts of H-1-infected

NBcells asopposed to infectedhamsterembryo

cells,whetherBUdRorTdR is used forlabeling.

Todeterminetherelativerates ofsynthesisof

Vstrands versusC strands bymeasuring

incor-poration ofBUdR into the V and C strands of

the RFDNA,the peakfractions of the RF DNAs

(Fig. 4) werepooled, precipitatedwithethanol,

and subjected to equilibrium centrifugation in

Cs2SO4

after heat denaturation as previously

0~~~~~~~~

3

ts

2

0'

1 5 10 15 20 25

FrOction No. 9

FIG. 4. Zonal sedimentation in neutralsucrose gra-dients ofwild-type H-i and tsl viral DNA synthesized atthe restrictive temperature(39.5 C). Wild type and tsl wereused to infect separate replicate culturesof synchronized NB cellsat amultiplicity of infectionof

5; cultures were treated with FUdR (0.5Ag/ml)16to 17hp.i. and thenlabeled with [3H]BUdR (1 Ci/10-9 mol) 17to18hp.i. Viral DNA wasextractedby the Hirt method.Sedimentationwascarriedoutin

gradi-entsof 5 to 20% sucrose(wt/vol) in50mM Tris-hydro-chloride, pH 8.0,1MNaCI,1mMEDTA, and 0.15% Sarkosyl in the SW25 Beckmanrotorfor 18 h at25,000 rpm at 4 C.Aliquots were measured for radioactivity byliquid scintillationspectrometry aspreviously (8). Theexpectedsedimentation position of virion DNA is indicated by the arrow labeled SS (8).

described (8, 10). The accumulations of

[3H ]BUdRwerenearlyequalforboth strands of

RF DNA for tsl and for ts2 (notshown) at39.5

C. The wild-type virus incorporated

considera-bly less BUdR into the V strand relative to the C

strandin NB cells at 39.5 C than waspreviously

found for wild-type virus in hamster embryo

cells at 37 C (V/C = 3.5; reference 8). Since

wild-type virus shows a twofold reduction in

PFUproductionwhenpropagatedat 39.5C,the

partial defectiveness of V strand synthesis at

the restrictive temperature may account for

this.

To establish that a mutation in tsl affects

progeny DNA synthesis, theabove experiment

wasrepeated at the lessrestrictivetemperature

of 38 C in hamsterembryo cells which permit

more virally related DNA synthesis (8).

Para-synchronous hamsterembryo cultures were

in-fected withwild type or tsl andincubatedat 38

C, from 30 min p.i. The cultures were then

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.491.255.442.63.271.2]

labeled with [14C]TdR (1 pCi/ml, 62 M

Ci/

mmol) in the presence of FUdR (0.5 ug/ml),

from 14 to 17 h p.i. At 17 h p.i., the cultures

were washed twice with warm Hanks balanced

salts and incubated from 17 to 18 h p.i. with

FUdRand [3H]BUdR (5

gCi/ml,

2.5Ci/mmol).

Viral DNA was extracted by the Hirt method

with Pronase digestion (4) and RF DNA was

preparedby zonal centrifugation as above. The

pooled RF DNA fractions were precipitated

withethanol, redissolvedin50 mM

Tris-hydro-chloride (pH 8.0), 10 mM EDTA, and heat

denatured

by

treating 6 min in a boiling-water

bath.Thetubeswerequenched inice water and

subjected to equilibrium centrifugation in

Cs2SO4 (Fig. 5).

Wild-type

virus shows a clear

disproportionate incorporation of [3H]BUdR

into theV strand ascomparedto the C strand.

The mutant tsl synthesized nearly equal

amounts of V and C strands even at 38 C,

indicating that it is relatively defective in V

strand synthesis in comparison to wild-type

H-1. The RF DNA labeled

by

[14C]TdR

pro-vides a reference for light DNA and it can be

seenthatmostofthisDNA was not converted to

hybrid

density

in the period 17 to 18 hp.i.

DISCUSSION

Temperature-sensitive mutants of the

par-vovirus H-1 have been isolated and partially

characterized. In this report, the properties of

two mutants are described. Both of these

mu-tants aredefectiveinplaque productionat39.5

C. Theyalso showinhibitionof viral

hemagglu-tininsynthesisandalterationsintheir

immuno-fluorescent antigen productionat therestrictive

temperature. It was found that for both of the

mutants thehemagglutinin, and for onemutant

(ts2), theantigen, appearedininfected cultures

after shiftdown to thepermissive temperature,

even when protein synthesis was inhibited by

cycloheximide. Thus these mutations are

caus-ing the synthesisof amissense capsid

polypep-tide with an alteredstructure atthe restrictive

temperature. This is the expected case for a

temperature-sensitive change in phenotype. In

further supportofthis,the hemagglutininoftsl

was found to have an increased sensitivity to

thermal inactivation.

We have previously shownthat the capsid of

H-1 consists oftwo polypeptides, VP1 and VP2

(5). Which of these two polypeptides contains

the mutation is under study. An earlierreport

8

!

C H-H RF

Hybrid

RF

AI

6

/v\

4~~~~~~~~

2~~~~~~~

OO\b~~0 ~ ~ ~ ~ ~ V

0

5

BI

15 0

~15

20 25 30 35 4

Fraction

No.

FIG. 5. Isopycnic gradientcentrifugationofheat-denatured [3H]BUdR-substituted wild-type RF DNA(Fig.

5A,and tsl RF DNA,Fig. 5B).Parasynchronoushamsterembryocultures wereinfectedwithwild-typeH-1or tsW asdonepreviously (4) and incubatedat38Cfrom 1hp.i. From 14 to17hp.i.thecultureswereincubated withFUdR (0.5 Ag/ml)and ['4C]TdR (1

ACi/ml,

62mCi/mmol).At 17 hp.i. the cultureswerewashed twicein Hanks balanced salt solution and incubatedfrom17to18hp.i.in mediumcontaining FUdR and [3H]BUdR(5 MCi/ml,2.5Ci/mmol).At18h,theviral DNAwasextractedandRFDNAwasisolatedasinFig.4.RFDNAwas heatdenatured andsubjectedtoequilibrium centrifugation inCs2SO4. The centrifugation wasconducted in

CsS04, 25mMTris-hydrochloride (pH8.0),5mMEDTA, and0.15%Sarkosyl for48hat35,000rpm at4 C

in the type 40 fixed-angle rotor. The density positions of the viral strand DNA is indicated by V, the complementary strandby C; theheavy,fullysubstitutedRF DNA andhybriddensity RF DNAaresolabeled. Gradientswereprocessedaspreviously (8). Symbols: (0) 14C, (0) 3H.

VOL. 17,

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.491.106.388.341.544.2]

666 RHODE

has identified thehemagglutininandviral

anti-gen ofthe closelyrelated parvovirusKRV with

the major capsid component, protein

B,

which

suggests VP2 may be the affectedprotein (11).

The effect ofthese capsid mutations on viral

DNA synthesis was also examined. H-1 DNA

synthesis has been classified into three

cate-gories: (i) parental RF DNA synthesis, (ii) RF

DNA replication, and (iii) progeny DNA

syn-thesis (8, 10). A DNA synthetic event upon

which subsequent VP1 and VP2 synthesis is

dependent andwhich occurs near the end of S

phase in the infected cell was termed HA-DNA

synthesis (9). It has been shown here that tsl

and ts2synthesize viralproteins atthe

restric-tive temperatureand thus HA-DNAsynthesisis

not defective. If RF replication required viral

protein synthesis, which in turn requires

HA-DNA synthesis (5), then HA-DNA synthesis

probably represents parental RF DNA

synthe-sis.

Similarly, tsl and ts2 produced normal or

increased levels of RF DNA. By using BUdR

substitution for TdR, it is possible to separate

the V strand and C strand of RF DNA and

compare the relative amounts of the density

label incorporated intoeach strand (8). Inthis

manner it was found that tsl and ts2 show no

evidence of asymmetric synthesis of V strand

DNA as a result of progeny DNA synthesis at

39.5C.Wild-typeH-1 was alsorelatively

defec-tive in V strand synthesis versus C strand

synthesis at that restrictive temperature. By

repeatingthecomparison atthe lessrestrictive

temperature of 38

C,

it was shown that the

mutation in tsl rendered V strand DNA

synthe-sismore sensitive tothermal inactivation than

that ofwild-typevirus.Thusacapsidprotein of

H-1, apparently thehemagglutinin, isrequired

fortheasymmetricsynthesisof Vstrand DNA.

A DNA moiety was observed consistently in

neutralsucrosegradients ofthe Hirt extracts of

NB cells infected with wild-type H-1 thatwas

decreased or absent with tsl infection. This

DNAsedimentedata ratesimilar tothatofH-1

viralDNA, butit has not yetbeenidentified as

H-1progeny DNA. The sedimentation patterns

of viral DNA extracted from hamster embryo

cells infected with wild-typeH-1 do not show a

peakatthat position (8). If that DNA proves to

be viral progeny DNA, then there may be

cellular factors that play a role in the virion

assembly and various cell types may exhibit

different efficiencies for this process. In the case

of adeno-associated virus, a similar form of

defectiveassemblyhas been suggested. Thus, in

thepresence of the partial helper virus, herpes

simplex virus, adeno-associated virus protein,

and DNA synthesis occur, but a late step in

infectious virus production is defective and

requires an adenovirus helper function (1). However, it has not been shown herpesvirus

infection is not inhibitory to adeno-associated

assembly. Minute virus ofmice, another

par-vovirus, has been shown to produce a pool of

viral DNA

readily

extractable by the Hirt

procedure (12). For both minute virus of mice

and H-1, the virions are not

lysed

by sodium

dodecyl sulfate at room temperature, so that a

Hirtextraction without Pronase does not

liber-ate DNA fromfully assembled virions. In

recon-struction experiments, H-1 viral DNA was

added to the sodium dodecyl sulfate lysing

bufferand was recovered in the final

superna-tant with a 50 to 60% efficiency. Therefore, it

canbe concluded thatnosignificantpooloffree

viral DNA is present in infected hamster

em-bryocells. Whetherthe rate of incorporation of

progeny

viral DNA into virions isdependenton

certain cell functions assuggested will require

furtherstudy.

ACKNOWLEDGMENTS

This research was supported by Public Service grants CA13414-03andCA07826-10fromthe National Cancer Insti-tute.

The guinea pig and hamster antisera used in thisstudy werekindlyprovided by Helene Toolan. The author gratefully acknowledgesK.Ellem andH.Toolanforhelpful criticisms of the manuscripts, Jessica Bratton for expert technical assistance, andVirginiaHaasand Janeen Prattforsecretarial duties.

LITERATURE CITED

1. Boucher, D. W.,J. L.Melnick, andH. D. Mayor. 1971. Nonencapsidated infectious DNA of adeno-satellite virus in cells coinfected with herpesvirus. Science 173:1243-1245.

2. Greene, E.L. 1965.Physical and chemicalproperties of H-1 virus. I. pH and heat stability of the hemag-glutinating property. Proc. Soc. Exp. Biol. Med. 118:973-975.

3. Hirt, B.1967.Selectiveextraction ofpolyoma DNAfrom infected mouse cellcultures. J. Mol. Biol.26:365-369. 4. Johnson, F. B., N. R. Blacklow, and M. D. Hoggan.1972.

Immunological reactivityofantisera preparedagainst the sodiumdodecylsulfate-treated structural polypep-tides of adenovirus-associated virus. J. Virol. 9:1017-1026.

5. Kongsvik, J.,J.Gierthy,and S. L. Rhode. 1974. Replica-tion process of the parvovirus H-1. IV. H-1 specific proteins synthesized in human NB kidney cells. J. Virol.14:1600-1603.

6. Ledinko, N. 1967.Plaqueassay oftheeffects of cytosine arabinosideand 5-iodo-2'deoxyuridineonthe synthesis ofH-1virusparticles. Nature(London) 214:1346-1347. 7. McGeoch, D. J.,L. V. Crawford,andE.A.Follett. 1970. The DNA'softhreeparvoviruses. J. Gen. Virol. 6:33-40 8. Rhode, S. L. 1974.Replication process of the parvovirus H-1. II. Isolation and characterizationofH-1 replica-tive formDNA.J. Virol. 13:400-410.

9. Rhode, S. L.1973.Replicationprocess of the parvovirus H-1. I. Kinetics in aparasynchronous cell system. J. J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

VOL. 17, 1976 Virol. 11:856-861.

10. Rhode, S. L. 1974. Replicationprocessoftheparvovirus H-1. III. Factorsaffecting H-1 RF DNA synthesis. J. Virol. 14:791-801.

11. Salzman, L. A., and W. L. White. 1970.Structural pro-teins of Kilham rat virus. Biochem. Biophys. Res. Commun. 41:1551-1556.

REPLICATION OFPARVOVIRUS H-1

667

12. Tattersall, P., L. V.Crawford, and A. J. Shatkin. 1973. Replication ofthe parvovirus MVM. II. Isolationand characterization ofintermediates in thereplicationof the viraldeoxyribonucleic acid.J. Virol. 12:1446-1456. 13. Tessman, I. 1968. Mutagenic treatment ofdouble- and

single-strandedDNAphagesT4 andS13with hydroxyl-amine. Virology 35:330-333.

on November 10, 2019 by guest

http://jvi.asm.org/