Replication process of the parvovirus H-1. X. Isolation of a mutant defective in replicative-form DNA replication.

JOURNALOFVIROLOGY, Jan.1978,p.215-223 Copyright ©1978 AmericanSociety for Microbiology

Vol.25, No. 1 Printedin U.S.A.

Replication Process of the Parvovirus H-1

X.

Isolation

of a Mutant

Defective

in

Replicative-Form

DNA

Replication

SOLON L. RHODE III

PutnamMemorialHospital Institute for Medical Research, Bennington, Vermont 05201

Received forpublication26 May 1977

A

temperature-sensitive

mutant of H-1, ts14, that is

partially

defective in

replicative-forn

(RF)

DNA

synthesis

hasbeen isolated. ts14 H-1 is

characterized

byadecrease in

plaque-forming ability

and

production

ofinfectious virusatthe

restrictive

temperature of

39.50C.

RF DNA

synthesis

ofts14isreduced to 3 to

7% of that of

wild-type

H-1ateither the restrictive or thepermissive temperature.

A complementation

analysis

ofRF synthesis of ts14 and a viable defective H-1

virus,

DI-1,

or

wild-type

H-3indicates that the defective RF DNA synthesis of

ts14 is

cis-acting.

ts14, unlike

wild-type H-1,

causes a

multiplicity-dependent

inhibition

ofDI-1or

H-3,

butnot

LuII,

RF DNA

synthesis.

Mixedinfections of

cells withtwo

parvoviruses

also exhibited a cross-interference for viral

protein

synthesis

thatwas

multiplicity dependent.

ts14inhibitedinfectious virus

produc-tion of H-1 or

H-3,

but not

LuIlI.

LuIII- or

H-3-pseudotype particles

were

produced by coinfection with H-1. H-3 and H-1 showed similar interactions with

ts14, and H-3 DNAwas more

homologous

toH-1 thanwasLuIII

by

comparative

physical

mapping

studies. Theresults suggest that ts14 is a mutant with a defect

ina

regulatory

sequence ofits DNA that

influences

RF DNA

replication.

The

parvovirus H-1,

which hasagenome of

linear

single-stranded

DNA, produces

a

double-stranded

replicative-form

(RF)

DNA

during

in-fection (5,

6).

This RF DNA

replicates

and

pro-vides the

template

for the

synthesis

ofprogeny

single-stranded

DNAfor

incorporation

into

vir-ions

(6-8). Progeny

DNA

synthesis

requires

one

orboth of the virion

capsid proteins

(71.

It has

been shown that RF DNA

replication requires

aviral gene

product

(S.

L.

Rhode,

in D. Ward

and

P.

Tattersall,

ed., Replicationn

of

Mam-malian

Parvoviruses,

in

press).

To

analyze

this

requirement,

I have searched for

temperature-sensitive mutants defective in RF DNA

repli-cation. In this

study,

the

temperature-sensitive

mutant ts14 will be shown to be defective in

RF DNA

replication (RF

rep-),

but not

condi-tionally

with the temperature.

Plaque-forming

ability

and virus

production

arethe

only

param-eters ofts14

replication

that have been found tobe heatlabile.

Thenature of theRFrep-defectofts14was

explored by

complementation

studies between

ts14 and RF

rep'

viruses of

H-1, H-3,

andLuIII. The RF rep-

phenotype

was found to be

cis-acting.

ts14also

produced

a

multiplicity-depend-ent inhibition of RF DNA syntheses of the

helper

virusesH-1, H-3,andH-T, butnotLuIII.

A

multiplicity-dependent

cross-interference for

viral

protein (hemagglutinin

[HA])

synthesis

was also defined for H-1, H-3, or LuII. H-3

DNA was

found

to be more homologous than

LuIII DNA to

H-1

by physical

mapping

tech-niques.

These studies suggest that ts14 has a

mutation ina

regulatory region

of itsDNA that

influences RF DNA

replication

and that ts14

proteins

are notdefective for RF DNA

replica-tion.

MATERIALS AND METHODS

CeUs andvirus strains. The cells used in this study are human newborn kidney cells transformed bysimianvirus 40 (NBcells).The H-1 mutants ts14 through-18 wereisolatedaspreviously described(3, 7) from a virus stock mutagenized with N-methyl- N'-nitro-N-nitrosoguanidine, obtained from Aldrich ChemicalCo., Inc., Metuchen, N.J. (8). Mutageniza-tionwas carriedout by treatinga parasynchronous culture of hamsterembryocellsinfected with H-1at amultiplicityof infection of5with mediumcontaining N-methyl-N'-nitro-N-nitrosoguanidine,25

iLg/ml,

from 12 to 14 hpostinfection.Themutagenwasremoved, and the cultureswerewashed twice withHanks bal-ancedsalts and refed withregularmedium. The cul-tures wereharvestedat56hpostinfection.Theyield ofplaque-formingviruswasreduced 90%bytreatment with thisdosage ofmutagen. The H-1 mutant DI-1 wasisolatedfroimapreparation of defective interfering (DI) viruses obtainedby serial passage at high multi-215

on November 10, 2019 by guest

http://jvi.asm.org/

216

plicity with wild-type (wt) H-1 helper virus (manu-script inpreparation).

Virusand HA neutralization. Virus preparations in tissueculture mediumwereadjustedtopH8.5with 0.5NNaOH and frozen and thawed three times. Equal volumes of virusand antiserumweremixed and incu-bated for 1 h at 37°C. After neutralization, twofold dilutions of the sample weremade with phosphate-buffered saline for titration of HA with guinea pig erythrocytes. Other samples were used directly to inoculate cover glasscultures of NB cells for deter-mination ofproductionof viralantigenby immunoflu-orescent (FA) staining. Cells with nucleipositivefor parvovirusantigenswerecounted withanocular grid at X400.

Preparation of radionuclide-labeled viral DNA. Parasynchronous cultures of NB cells were infected with H-1 and incubated inalow-phosphate medium containing [32P]orthophosphate at 20 to 50

,uCi/ml and/or

[3H]thymidine

in the presence of

5-fluorodeoxyuridine,0.5,ug/ml, from12 to 16h postin-fection. Viral DNAwasextracted bythe method of Hirt andpurifiedby sedimentation in neutralsucrose

gradientsaspreviouslydescribed(7). Hirtextracts to beanalyzeddirectly bygel electrophoresiswere con-centratedbyprecipitationwith ethanol. The

precipi-tates were collected by centrifugation and dried in vacuo. The DNA was redissolved in 50 mM Tris-hydrochloride (pH7.5)-imMEDTA anddigested for 30min at roomtemperature withpancreatic RNase (treatedfor20minat80°C)at50ug/ml.

Gelelectrophoresis.Slabgel electrophoresiswas conducted in an EC470 apparatus (E-C Apparatus Corp.,Philadelphia, Pa.), using gels (0.3 by 12by 16

cm)of1% agarose.Afterelectrophoresis,thegelswere dried invacuoandexposedtoKodakRoyalX-Omat RP/R2for upto2weeksatroomtemperature. Acryl-amide gels with acrylamide concentrations greater than 7.5%werepartially dehydrated byimmersion in 70% ethanol plus30% 0.15 MNaCl-50 mM Tris-hy-drochloride (pH 7.5)for1to2hat4°Cbeforedrying. Thisprocedure resulted insomeshrinkageofthe gel butprevented cracking duringthedryingprocess.

Quantitationof 32Por3H indried agarosegels was done byexcising the DNA-containing region of the gelandincubatingthe slice in 0.15 ml of50mM Tris-hydrochloride (pH 7.5)-0.5 mMMgCl2-2 ,ug of pan-creatic DNase for 12 to 16 h at 37°C. The sample

was then treated with 0.6 ml of NCS (Amer-sham/Searle,Chicago, Ill.)and counted with 10 ml of toluene-basedscintillation fluid inaliquidscintillation spectrometer. Acrylamide gelslices weresolubilized with perchloric acid and hydrogen peroxide (4) and counted withAquasol (NewEngland NuclearCorp., Boston,Mass.).

Restriction endonucleasedigestionof H-1 RF DNA. The bacterial restriction endonucleases used here were purchased from Miles Laboratories, Inc., Elkhart, Ind.Digestionswereperformedaspreviously described (9).

RESULTS

Isolation andcharacterization ofts14. All

H-1isolates

showing

temperature

sensitivity

for

plaque formationwere

propagated

totiters of5

x 107or higher. Theisolate ts14 had a titer of

less than 5 x

103

PFU/mnl

at the restrictive

temperature of 39.5°C and a titer of 3 x

108

PFU/mlatthepermissivetemperatureof33°C.

Theplaquesatthepermissive temperature

ap-peared

smaller than those ofwtH-1.All ofthe

mutantsisolatedwerescreened for reduced RF

DNA

replication

atthe restrictive temperature

by

infecting

with amutant one 100-mmdish of

NB

cells

synchronized with methotrexate and

labeling the viral DNA with

[32P]orthophos-phate from 10 to 14 h postinfection at39.5°C.

The viral DNAs were extracted and analyzed

by

electrophoresis

in a

slab

gel of 1% agarose,

as described in Materials and Methods. One

isolate, ts14, was selected for furtherstudy

be-causeit

synthesized considerably

less32P-labeled

RF DNA monomer or dimer than didwtH-1.

ts14isthe

only

mutantisolatedtodate that has

been defective in RF DNA

synthesis

at the

restrictivetemperatureof39.5°C.

ts14 and wt H-1 DNA syntheses were

com-pared undera

variety

ofcondtionsto

determine

if the

synthesis

ofts14 RF DNA was

tempera-ture

labile

and if itwas rapidly inhibited by a

shift from the

permissive

tothe restrictive

tem-perature. In this

experiment,

viral DNA was

labeled for the indicated times with

[3H]-thy-midine and fractionated in a slab

gel

of 1%

agarosein thepresenceofmarkers of

32P-labeled

H-1 DNAs. Themonomerand dimer RF DNA

and viral DNA bands of each

sample

were

ex-cised from the dried

gel,

and the

3H

was

mea-sured

by

liquid

scintillation spectrometry. The

results

(Table

1) indicate that ts14 monomer

and dimer RF DNA

synthesis

are defectiveat

both39.5and33°C. The

synthesis

of viral DNA

was also

decreased

at both temperatures, but

not as

profoundly

as that of RF DNA

(72

to

87%

compared

with92to

96%).

Since the diminished

synthesis

ofts14 viral

DNAwas nottemperature

conditional,

the

pro-duction of infectious viruswasexaminedatthe

restrictive andpermissivetemperatures.The

re-sults indicate thatts14 infectious virus

produc-tion at

39.50C

is defective, with yields about

10-1

to

10-2

of those of ts14 at 33°C or wt at either temperature (Table 2). The production ofts14 at33°Cissurprisingly high, considering

thedefectivenessofts14RF DNAsynthesis at

330C.

The parameters of ts14

replication

that

are

thermolabile

are

plaque-forming

ability

and

infectiousvirusproduction.I was notsuccessful

indemonstratingcomplementation betweentsl,

ts2,

orts18andts14forinfectiousvirus

produc-tion at

39.50C.

tsl,

ts2,

and ts18 are RF DNA

rep' (7),butnoneoffiveH-1mutantstested to

date have been complemented by each other

even though

they

demonstrated differences in

J.

on November 10, 2019 by guest

http://jvi.asm.org/

phenotypeatthe restrictive temperature. There-fore, negative complementation has not been

proventobesignificant for conditionalmutants

of H-1. Theonlyviruses for which complemen-tation has been demonstratedare DI particles of H-1 (Rhode,inpress).

Complementation

oftsl4and RF

rep'

H-1.Aputative missensemutation in a viral

pro-teinmightbe difficult to identify directly, so a

genetic approachwas usedtodetermine ifts14 has amutation in astructural gene foraviral proteinrequired for RF DNA replication. If ts14 produces atrans-acting defective protein, then

coinfectionwithanRF

rep'

helper virusshould

rescue the ts14 RF DNA and greatly increase itsreplicationrate.Alternatively,ats14protein

may beinhibitory to RF DNAreplication and willactin atrans-dominant fashion, inhibiting

the replication of an RF

rep'

DNA. Also, the mutationmaybe inaregulatory region of ts14

DNA and, consequently, behaveas a

cis-domi-nant mutation. Complementation experiments with H-1weremadepossible by the isolation of

DI-1, aviable defectiveH-1 virus with normal

levels of RF DNA replication whose RF DNA

can be identified by virtue of an addition of about65 base pairs in the HindII-D region of the H-1 physical map (unpublished results).

Theseexperimentsweredoneby comparing the

yields of 32P-labeled RF DNA of ts14 at 370C in thepresenceand absence ofcoinfection with

TABLE 1. Synthesis ofwtandtsl4RFDNAand viral DNAat39.5or330Ca

[3H]thymidineincorporationb Incuba- ts14(%)

teionp

Di- Mono-

tsl4/wt

tep mer mer Viral (total

RF RF DNA cpm)

DNA DNA

33°C 4 4 13 2,623/70,275

39.50C 7 6 28 2,970/43,432

33-39.50Cc 5 4 16 5,391/122,207

aReplicate cultures of parasynchronous NB cellswere in-fectedwithts14orwtH-1atamultiplicityofinfection of 25

to50. Viral DNAwaslabeledby incubating the cultures in prewarmed medium containing 5-fluorodeoxyuridine, 0.5

ag/ml,and [3H]thymidine, 5 puCi/ml, 10' M. Thelabeling periodswere:33.C,18 to 20 hpostinfection;39.50C,12 to 14 hpostinfection;and33 to 39.50C, 19 to 20 hpostinfection.

Theincorporationof[3H]thymidineintoH-1 RF DNAand

viral DNAwasquantitatedasdescribed in thetext.

bThevalues shownarethe relativeamountof 3H in each type of DNA for ts14 compared with wt, expressed as a

percentage and the total countsper minute (dimer RF+

monomerRF + viral DNA) for each virus. Monomer RF

DNA represents about 80% of the total, dimer RF DNA

represents about 15 to 18% ofthe total, and viral DNA

represents the remainder.

cIncubatedat330C until18hpostinfection, then shifted

[image:3.505.262.451.62.191.2]

to39.50C.

TABLE 2. Production of tsl4 at the restrictive and permissivetemperaturesa

Virus production H-1 Group Incuba- (PFU/ml) type no. tiontemp

1hp.i. 48hp.i.

tsl4 1 330C 4x 106 1 x 10l

39.50C 1x 107 1 X 107 2 39.50C 1 x105 9 x106

wt 1 330C 7 x106 6x10

39.50C

1x107 1X109 2 39.50C 2 x106 3 x105

aNB

cell

cultures were inoculated withts14 or wt

H-1 at a multiplicity of infection of 5 to 10 and ad-sorbed for 60 min at their respective temperatures. The cultures were washed twice with Hanksbalanced salt solution and incubated at theirrespective temper-atures.Cultures were harvested and virus infectivity was titratedby plaque assay at 33°C at 1 and 48 h postinfection (p.i.).

DI-1. The total incorporation of 32p into RF DNA from each infection was determined by

gel

electrophoresis

or by sucrose gradient

cen-trifugation. The yield of 32P-labeled RF DNA

of each virus inamixedinfectionwascalculated,

using the relativeamountof each viral DNA in

the mixture determined by the amounts of its

marker

fragments

(HindII-D or -D'). A

repre-sentative

electropherogram

ofa HindII digest

of the RF DNA ofts14,DI-1, andamixture of

thetwois shown in

Fig.

1.Parasynchronous NB

cultures were infected with ts14, DI-1, or ts14 and DI-1 at various multiplicities of infection

(Table

3). Itshould be noted that the

multiplic-ities of infection ofts14 and DI-1 quoted here

are based on the infectivity titrations of each

virus by plaque assay, and the efficiencies of

thisassayfor thetwovirusesmaydiffer. When

ts14 had a

multiplicity

of infection

advantage

overDI-1

(experiment

1,

Table

3), it inhibited

the RF DNA

synthesis

of DI-1 andreduced the

combined

yield

of RF DNA for DI-1 andts14

by

65%

compared

with DI-1 alone. In other

words,

thets14 RFrep-

phenotype

was

trans-acting.

Since a

multiplicity-dependent

cross-in-terference for viral

protein synthesis

doesoccur

in mixed

infections,

asdiscussed below,

experi-ments 2and3 weredone with DI-1atthesame

multiplicity

asts14orwitha

multiplicity

advan-tageof3:1overts14.The results showthateven when DI-1 was the majority virus and should

have produced most ofthe viral protein, ts14

RF DNA

synthesis

wasonlymodestly increased. Because ts14 RF DNA incorporated relatively lownumbers of

32p

countsper minute and be-cause thisincreasedidnotapproacha wtlevel

of

incorporation,

it was notconsideredasa

sig-nificant complementation of the defective RF

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.505.53.247.406.514.2]

DNAreplication. Thus, the ts14 mutation was

cis-actingatlowmultiplicities of ts14 in comple-mentation with DI-1. DI-1 RF DNAsynthesis

was reduced by the trans-dominant effect of

ts14 ts14

+ +

ts14 DI-I Dl-l

D-i-1. w --D

FIG. 1. Gelelectrophoresis pattem ofHindIIand HindII-plus-HpaII digestion products oftsl4and

DI-1RFDNA. The32P-labeledRFDNAoftsl4-,DI-l-,

and tsl4-plus-DI-1-infected cultures wereprepared

byHirt extraction and sedimentation in a neutral

sucrosegradientaspreviouslydescribed(7). Samples ofeach were digested in 20-IAl volumes for4 h at

37°Caspreviouslydescribed(9).(1)40%ofthetotal

yield ofts14RF DNAdigested byHindII;(2)20%of the totalyield ofts14plusDI-1digested byHindII; (3)asin(2), digested byHindIIandHpaII; (4)10%

oftheyield of DI-1RF DNAdigested by HindII.

ts14inproportiontothe ratio of ts14toDI-1 in the inoculum.

RF DNA synthesis in heterotypic infec-tions oftsl4 H-1 with H-3 or LuLI. In the

complementation tests with ts14 and DI-1, it

was not possible to determine if DI-1 proteins were synthesized, since ts14 and DI-1 proteins

areantigenically similar. Therefore,

complemen-tation experiments were carried out with ts14 and other nondefectiveparvoviruses capable of infecting NB cells. In preliminary experiments,

ts14 H-1 caused a reduction in total RF DNA

synthesis on coinfectionwith wtH-1, H-3, and

H-T,aspreviously described for coinfection with

DI-1. In contrast,LuIII RF DNA synthesiswas

not inhibited. Therefore, H-3 and LuIII were

selected for further study. First, itwas necessary

toobtain markers for H-3andLuHI RF DNAs,

so that synthesis of their respective RF DNAs

could be measured in the presence ofH-1 RF

DNA. Cleavage maps of H-3 and LuIII were

determined for the restriction endonucleases EcoRI,HindIII, HindII,HaeIII, and HpaII(Fig. 2). The assignment ofthe origin of replication and the 5' terminushasbeenmade only for H-1; H-3 and LuIIIare oriented in the figure on

the basis ofhomologous cleavage sites and the variations instructureofthemolecular endsas

describedfor H-1(9).TheRFDNAsofts14 H-1, H-3, or LuIII alone or of H-3 or LuIII in

mixed infection with ts14 were labeled with

[32P]orthophosphate and prepared for digestion with HaeH as in Table 3. Replicate cultures wereharvested for HA neutralization assaysto

measure the effect of mixed infection on viral

protein synthesis. Cover glass cultureswerealso

infected with the virus mixtures in the same

manner and stained for the antigens of both viruses oreach virusseparately by the indirect

FAstaining method. The FA staining in the H-3 experiment showed that 70% of the cells (300

TABLE 3. InhibitionofDI-1RF DNA synthesis bycoinfection withtsl4a

MOjb RF DNAyieldc

Exptno.

DI-1 ts14 ts14 DI-1 tsl4/DI-1 Total(cpm/107cells)

1 10 20 0.9 0.1 1.6 115,540

2 5 5 0.8 0.4 0.4 205,065

3 15 5 1.5 0.8 0.2 144,683

aTheyields ofts14 RFDNA orDI-1 RF DNAafter double infection at 37°C with ts14 and DI-1 were

calculated from thecountsper minuteof3P recovered in theirrespective marker fragments, HinduI Dfor ts14andHindII D' forDI-1,asillustrated inFig.1. The valueswereadjusted for relative fragment sizes and quantities appliedtothegel. The total RF DNAyieldsweremeasured by eithergel electrophoresis in0.8% agarose gels or Cerenkov counting ofthe preparative neutral sucrose gradients. The specific activities of

[32P]orthophosphate

in eachexperimentarenot thesame.

b

MOI,

Multiplicity of infection

(PFU/cell).

cValues shownare:theyield of each virus RFDNAinthe

mixed

infectionnormalizedto theyield of that

virus alone(ts14, DI-1),the ratioof ts14 RFDNA tothat of DI-1 in the mixture (ts14/DI-1), andthetotal countsperminute inmonomerRF DNAper107cells.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.505.66.259.123.374.2]

counted) were

positive

foreither H-1 orH-3or both (incubated with anti-H-1 and anti-H-3), with 78% positive for H-1 only and 62% positive

for H-3

alone.

The

result

for

the

LuIII

experi-ment was 75% positive for H-1 or LuIII, 71%

for H-1, and 51%positive for LuIII. Therefore,

the majority of the infected cells produced

an-tigens of both viruses in the mixed infections.

HO- 3'r-0

V

1 2 3

Kilobase Pairs

c 0

I i

zo II

Origin

4 5

11

t

0 a_II

ILI

0 10a

O 00.

Q- B

l, m0 c().cS

Ix

I II I

FIG. 2. Comparison ofthephysical mapsof H-1,

H-3,andLuIII.ThecleavagemapsofH-3 andLuIII

to the bacterial restriction endonucleases endo R EcoRi, endo R HaeII, endo R HindII, endo R HindIII, and endo R HpaII were determined with native RFDNA asforH-1 (9). Cleavagesites thatarehomologousbetweenH-1 and H-3orLuIII

are indicated by the long solid lines, and nearly homologous cleavagesitesaremarkedbythe dashed lines.CleavagesitesonH-3orLuIIIwithout similar sites on H-1 aremarked by short solid lines. The clusterof HpaIIcleavagesites atthe rightendof H-3,indicatedbythe braces,isverysimilar, butnot identical, tothatattherightendofH-1. Thereare no additional cleavagesitesfor HindIIwithin the brackets markedHindII. TheorientationoftheH-3 andLuIII cleavagemapswithrespect tothe5'-PO4 terminusofthe viral strand andorigins ofreplication

werenotdeterminedandare assignedhere strictly

on the basis oftheirhomology toH-1 (13, 14). The totallength ofH-I has beenadjusted upwardto5,400 basepairs, based on more recent estimates ofthe

sizeof

OMK74

DNA(11).

The yields of RF DNA for each virus were

determined, using

their distinct HaeII-A

frag-ments

(Table

4). Coinfection

with ts14

H-1

re-duced

H-3 RF DNA

synthesis

about 50%

and

did notinhibit LuIII RF DNA

synthesis.

The

level ofts14RFDNA

synthesis

was not

changed

by H-3.On the other

hand,

inthe culture

coin-fected with

LuIII,

ts14 RFDNA

synthesis

was

markedly reduced,

whereasLuIII RFDNA

syn-thesis was

unchanged.

It will be shown in the

next section that in the mixed infections with

ts14 at alower

multiplicity

than the

helper virus,

as used

here,

the

majority

ofthe HAs

synthe-sized were H-3 or

LuIII,

not H-1. When ts14

and

H-3wereused inamixed infection with a

multiplicity

of 3:1 in favor of ts14, the

major

portion of the HAwasH-1 and the

yield

of

"P-labeled RF DNA of H-3 was reduced at least

70%(datanot

shown).

Thus,

ts14inhibited H-3

RF DNA

synthesis

ina

multiplicity-dependent

manner, as found

previously

with DI-1 H-1.

With wt H-1 and H-3 at

multiplicities

of 10

PFU/cell, the

bulk

of theHAwasH-1

(c.f.

Table

5, experiment 3) and the RFDNA

synthesis

of

each viruswas notreduced.

ts14 inhibtion of HIA

synthesis

of other

parvoviruses.

The

parvoviruses H-1, H-3,

and

LuIII have similar hostrangesin that

they

all

infect many of the same human and hamster

cell

lines (1, 14).

They

also showatleast some

sequence

homology

as determined

by physical

mapping

with bacterial restriction

endonucle-ases

(Fig. 2).

Itwill be shown here that

coinfec-tion ofH-3-orLuIII- infected NB cultures with

ts14 H-1 exhibits a cross-interference for the

synthesis

of their HAs

(capsid proteins).

The inhibition of DI-1 H-1 or H-3 RF DNA

synthesis by

ts14 H-1 could result from the

inhibition of

synthesis

of a

putative

RF

rep'

gene

product.

For

LuIII,

whose RF DNA

syn-thesis

was not

repressed by

ts14,

itshould be

determined

whether

LuIII

inhibitedts14

protein

TABLE4. Effect of

tsi4

or wt

H-i

coinfection on the RF DNA synthesis of H-3 and

LuIIIa

MOIb RF DNAyielda

Exptno. Viruses H-3or

H-1i LIII H-1 H-3orLuIII Total cpm

1 ts14+H-3 6 12 1.07 0.47 384,717

2 ts14+LuIII 6 12 0.04 1.23 143,921

3 wt +H-3 10 10 1.34 1.15 1,585,713

aThe yields of32P-labeledH-1 and H-3 orLuIII RF DNA after single or mixed infection at 37°C were

calculated by determining the relativeamountof eachvirus in themixture and measuring the total yield of RF DNAof each infectionby gel electrophoresis. Intheseexperiments, the HaeIIA fragments of each virus wereusedasmarkerfragmentsand thecountsperminute recoveredwerecorrected for differences in size of therespectiveHaeIIAfragments. Experiments1and2weredoneatthesametime.

bMOI, Multiplicity of infection

(PFU/cell).

cValues shownaretheyield of each virusRF DNAin the mixed infection normalized to the yield of that

virusalone and total counts perminute in monomer RF DNA. H-1

Lu

mE

219

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.505.54.246.160.280.2] [image:5.505.51.448.514.584.2]

220

synthesis. For H-3 and LuIII, whichare

antigen-ically distinct from H-1, these possibilities can

be testedby measuringtheeffect ofts14 on

H-3 or LuIII HA synthesis. This was done by

infecting cultures with ts14 H-1, H-3, or LuIII

alone orincombination (and similarly with wt

H-1 and H-3) and comparing the yield of HA after neutralization with the various antisera (Table 5). Itwasfound that ts14 inhibited H-3

orLuIIIHAsynthesis and thatwtH-1inhibited

H-3 protein synthesis as well. This inhibition

wasmostmarked whents14wasusedatahigher

multiplicitythan that of H-3. FAstaining of NB

cellsinfectedwith ts14and H-3 (as inTable 5, experiment 1) revealed about 60% of the cells

tobepositive for H-3 intranuclear antigen,

sug-gesting thatmostcellsmayhaveproduced H-3

protein in reduced amounts. By giving H-3 a

multiplicity advantageof2:1 (experiment 4, Ta-ble 5), theproportionofH-3 HAwasincreased

to an amount greater than that of H-1. This

was the same experiment used for analysis of H-3RF DNAsynthesis in thepresence ofts14

in Table 4. Thus, it isunlikely that inhibition of H-3proteinsynthesis accountedforthe inhi-bition of H-3 RF DNA synthesis caused by coinfectionwithts14. The failure ofts14to in-hibitLuIII RF DNAsynthesiswas not due to LuIII inhibition ofts14 protein synthesis since LuIII RF DNAwas not found to be inhibited forexperiment5 (c.f. experiment 2, Table 4) or

forexperiment6 (comparativedata notshown) of Table5,wherets14proteinwaspredominant.

Since the hemagglutinating units after each neutralizationwere not additive tothecontrol,

perhaps dueto phenotypic mixing ofcoat

pro-teins, thesedatacannot beinterpretedinstrictly quantitative terms. They do indicate that H-1,

H-3,andLuIIIcompeteforsomelimitingfactor

concerned with viralprotein synthesisandthat H-1 tends to dominate these mixed infections, especially whenit hasanadvantage in

multiplic-ity.

Effect of ts14 H-1onproduction of

infec-tious virus inmixed infections with other parvoviruses. The effect of coinfection with ts14 H-1 on the production ofinfectious virus

ofwt H-1, H-3,andLuIIIwas examined.Since

ts14producesnoplaquesat39.5°C,theproduct of mixed infectionscanbe plaquedat39.5°Cto

determine theyield ofwtvirus. The results of these experiments are shown in Table 6. ts14 inhibited wt H-1, but not LuIII. The increase in titer for LuIII (as determined by PFU at

39.5°C) wassurprising, since there appearedto

benoLuIII HAproducedinthesecultures

(Ta-ble 7).The possibility of phenotypic mixing in whichLuIIIDNAisencapsidated with H-1

pro-tein (completely or in part) was tested by a

neutralization assay. Particles containing LuIII

DNA were assayed by FA staining for LuIII

antigen after the infectious particleswere

neu-tralizedbytreatment with H-1 antiserum. The LuIII from the mixed infection of ts14 and LuIII produced 7.3 1.2 (mean of 10 microscopic fields at x200 95% confidence limit) LuIlI FApositive cells, and, after neutralization with H-1 antiserum, this dropped to 0.5 ± 0.3. The

control standard LuIII gave 19.3 ± 2.3 LuIII FA-positive cells and 16.3 ± 1.3 after neutrali-zation with anti-H-i. Thus, a large portion of

the virus (from the mixed infection) that

pro-duced LuIII FA antigen could be neutralized with H-1 antiserum. This implies that LuIII DNA was encapsidated wholly or partly with

H-1protein.

In mixed infection with H-3 at 39.5°C, ts14 inhibited H-3production. However, the

replica-TABLE5. Effectofmixedinfectiononthesynthesis ofts14H-i,wtH-i,H-3, and LuIIIHAsa

MOIb HA titer

Exptno. Viruses H-3or Anti-H-3or

H-1i -30r Control Anti-H-I -LuIII

1 ts14+H-3 15 5 17 6 16

2 ts14+H-3 10 10 18 11 15

3 wt+H-3 10 10 17 10 15

4 ts14+H-3 6 12 18 13 11

5 ts14+LuIII 6 12 15 14 12

6 ts14+LuIII 20 40 17 11 16

aNB culturesweresynchronizedwith methotrexateand infected withts14H-1, wtH-1, H-3,and

LuIlI

or combinations of H-1 and H-3orLuIIIat37°C.Thecultureswereharvestedat18hpostinfection andfrozen and thawedthreetimes.Lysates (0.1ml)weretreatedwith0.1mlofspecificantisera for1hat37°Cand then dilutedto0.4ml withbuffer,andtheHAwastitrated withtwofold dilutions. TheH-1, H-3, andLuIIIantisera neutralized allof theHAproduced bytheirrespectiveviruses insingle infectionstotitersofless than4 and showednocross-reactivity.Controlsweretitrateddirectly,withouttreatmentwithserum.Similarresultswere obtainedwhen controlsweretitrated aftertreatmentwith nonimmuneserum.

bMOI,Multiplicityofinfection

(PFU/cell).

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.505.65.460.488.583.2]

[image:7.505.54.449.88.277.2]

TABLE 6. Effect of coinfection withtsl4 on production of infectious virus by wt H-i-, H-3-, orLulII-infected NB cellsa

Virus yield(PFU/ml)

Expt Virus MOIb 39.50C 330C

lhp.i. 48hp.i. lhp.i. 48hp.i.

1 (asyn) wt 4 2x104 3x107 6x 105 4x 108

ts14 20 0 0 2x106 1X 107

wt+ts14 4+20 2x 104 9X 104 2X106 1X107

2 (syn) LuIII 4 8x 104 9X106 lx 6 6x107

ts14 20 0 0 7x106 4x107

LuIII+ts14 4+20 3x 104 5 x 6 3 x106 5x107

3 (asyn) H-3 4 2 x 104 7 x 105 8 x 105 1X 107

ts14 20 0 0 1x106 3 x106

H-3+ts14 4+20 3 x104 3x 104 3 x106 3x106

4 (syn) H-3 10 4 x104 2 x106 1 x 6 2 x107

ts14 10 0 0 2 x 105 4 x106

H-3+ts14 10+10 6x104 1x 105 9x 105 2x 107

aNB

cells

wereinfected withwtH-1,LuIII,orts14and with each virusplusts14 at39.5°Cforexperiments

1through3andat37°Cforexperiment4.Yields of infectious virusweredetermined byplaqueassayat 1 and 48h postinfection (p.i.).Theplaque assays were done atboth 39.5 and 33°C. Since ts14does not produce plaques at39.5°C, wt H-1, LuIIl,and H-3can betitrated in the presence ofts14. Inexperiment 2, the NB cellsweresynchronized(syn) with methotrexate (0.5

jig/ml)

at37°C.Themethotrexate-treated cultureswere infected with LuIIIormockinfectedat37°Cin the presence of thedrug12hafterinceptionof thetreatment. Theywereinfected withts14 ormock infected4hlater and shiftedto39.5°C, with the methotrexate replaced bythymidine(10-' M). For the other viruses (experiments 1, 2, and 4) the infectionsweredone simultaneously withoutmethotrexatesynchronization (asyn)orwith methotrexatesynchronization.

bThemultiplicity of infection (MOI, PFU/cell) is based on the quantity of virus in the inoculum, and adsorptionisapproximately50to70%.

cValues shownarePFU permilliliterasassayedat 39.5or33°C.

TABLE 7. Effect of mixedinfectiononthe synthesis of ts14 and LuIII

HAG

HAtiter"

Virus MOIb Control

anti-H-i antiLullI

1hp.i. 48 hp.i.

ts14 20 6 17 <4 17

LuIII 4 0 15 16 12

ts14+LuIII 20+4 7 16 <4 16

aNBcultureswereinfected with ts14, LuIII,orts14plusLuIIIasinTable6.Thecultureswereharvested andsubjected tofreezingandthawingthree times. Samples (0.1 ml) at48hpostinfection (p.i.) weretreated with0.1ml ofspecificantisera for1hat37°Cand then dilutedto 0.4ml with buffer, and the HAwastitrated. Controlsweretitrateddirectly,withouttreatmentwithserum.

IMOI,

multiplicity

of infection(PFU/cell).

cValues shownarethe

log2

HA titers.

tionof H-3 anditsinfectivity titrationby plaque

assay appear to be more sensitive to the high

temperature than do those of H-1. Therefore, thisexperimentwasrepeatedat37°C and with synchronized NBcells.Evenunderthesemore

favorableconditions, theproduction of H-3was

reduced by about 90% by coinfection with ts14 H-1. Since H-3 RF DNA synthesis was also

reduced by ts14 H-1, this resultwas not

unex-pected.

The H-3 virus

produced

in mixed infection

withts14

H-1

in

experiment

4,

Table

6,

orwith wtH-1atthesame

multiplicity

ofinfectionwas

tested for

phenotypic mixing

of H-3 DNA with

H-1

capsid protein

asabove for

LuIII.

The H-3 virus

(i.e.,

H-3

DNA)

of both virus

preparations

was

partially

neutralized

by

anti-H-i

antiserum,

andpure H-3wasnot. The number of H-3

FA-positive

nucleiper

microscopic

fieldforts14 H-1

plus

H-3 virus was 10.6 + 1.1

(mean

of 10

221

on November 10, 2019 by guest

http://jvi.asm.org/

222

RHODE

areas +95%confidence

limit)

withnonimmune serum and 2.9 ± 1.5 after anti-H-1. The same figuresfor thewtH-1plus H-3were 15.1 ± 2.2

for nonimmuneserumand4.2 ± 1.2for

anti-H-1. Therefore, H-3 virus

pseudotypes

with H-1

antigen in

their capsids

were formed in mixed

infectionwith eitherwt orts14H-1.

DISCUSSION

Current

knowledge

of the structure and

rep-lication of

parvoviruses

suggests that

they

are

monocistronic

and thatthis cistron

produces

the

coat

proteins

(10). Studies ofcellsinfected with

H-1 orminute virus of mice have detected the

synthesis

of

only

two

polypeptides

induced by

viralinfection, the

capsid proteins

VP1and VP2'

(termed

A and B in minute virus of mice

[2,

15]). Recent studies of the

peptide

maps of the

Aand B

proteins

of minutevirus of mice indicate

that these

polypeptides

arecoded forbya

com-monregion of the minute virus of micegenome

(16). These findings

imply

that the

complex

processes of viral DNA synthesis have a high

degree of

dependency

onhostcell

proteins.

The

synthesis

of viral DNAcanbe divided intothree

operational

stages:

(i)

synthesis

of the

parental

RF DNA,

(ii) replication

of RF DNA, and

(iii)

synthesis

of

single-stranded

virionDNA.

Anal-ysis

ofinfection

by

a

temperature-sensitive

mu-tantof

H-1,

tsl,has shown that

single-stranded

virion DNA

synthesis requires

a function

con-trolled

by

one orboth

capsid proteins (7).

Stud-ies of(DI) viruses of H-1 have established that

aviral

protein

is

required

for RF DNA

replica-tion

(Rhodes,

in

press),

and

temperature-sensi-tivemutants ofH-1 defective in RF DNA

rep-lication,

i.e.,

RFrep-,are

being sought.

Inthis

report, Ihave described the isolation and

char-acterization of ts14, the

only

mutantwith

defec-tive RF DNA

replication

of 20 H-1 mutants

characterized

todate.

ts14 was isolated as a

temperature-sensitive

mutant witha thermolabile

ability

to

produce

plaquesatthe restrictivetemperatureof

39.5°C.

Virus

production

was also diminished when

in-fectionwascarriedout attherestrictive temper-ature. An

analysis

of ts14 RF DNA

synthesis

revealedit tobedefectiveatboth the restrictive

and the permissive temperatures. The

uptake

of

[3H]thymidine

into ts14 RF DNA was 3 to

7% ofthat ofwt H-1.

Thus,

the RF rep-

phe-notype ofts14 was nottemperature

dependent

undertheseconditions.

Anumberofmechanismscanbe

proposed

for the action of the RF rep- mutation. A choice among themwas

sought by

conducting

comple-mentation tests for RF DNA

replication

be-tween ts14 and RF

rep'

viruses

H-1, H-3,

or

LuIII. It has been shown previously that H-1

expresses a

trans-acting

RF rep gene product required for RF DNA replication in NB cells

(Rhode, in press). Thiswasestablished by

dem-onstrating that nonviable H-1 DI particles

re-quire coinfection with an RF

rep' helper

virus

to replicate their RF DNA. H-3

complemented

the H-1 DI

particles

for thisfunction, but LuIII

didnot.

Theresults obtained inthecomplementation

tests with ts14and H-1 DI-1 or H-3 indicated

that the RF rep- phenotype of ts14 was

cis-acting andwasnotrescuedby coinfection with

these RF

rep'

viruses. This result for the

com-plementation

tests with H-3 was validated by

confirming that H-3 proteins weresynthesized

in the double-infected

cells

by FAstaining and

byanalysis of H-3 HA synthesis. Thiswas not

possible

for the mixed infections with H-1 DI-1,

sincets14andDI-1

antigens

are

indistinguisha-ble.Since the H-1 RFrep geneproduct is

trans-acting (Rhode, in press), these findings

imply

that the RFrep-phenotype ofts14is not caused

byamutation in the viral protein required for

RF DNAreplication. Insupportofthis

interpre-tation,ts14 proteinswerefoundtocomplement

the RF rep- DI

particles

described above for

RF DNA synthesis as well as wt H-1 (Rhode,

in press). The most likely explanation for the

ts14 mutation

affecting

RF DNA

replication

is

that it is a change ina regulatory sequence of

ts14DNA.

Itisnotcertainwhether the RFrep-mutation

isthesamemutation thatprevents ts14plaque

formation and reduces infectious virus

produc-tion at 39.5°C. Three revertants of ts14 were

isolated

through

their

ability

toformplaquesat

39.5°C, and all three had lost theRF rep-

phe-notype. Since H-1 has high particle/infectivity

ratios

(unpublished

results), it is possible that

themethod of selection forrevertants salvaged

some

"subliminal"

contaminants intheoriginal

ts14 preparation. This seems unlikely, because

each of the three revertants differed from wt

H-1 in either

plaque

morphology or plaquing

efficiency

at the restrictive temperature.

Be-causethe revertants werenot wt virusand be-cause noneof themwereRFrep-butnot tem-perature sensitive,thequestionofwhetherts14 has one or more than one mutation has not

beenanswered.

It was observed in mixed infections of ts14

with DI-1, H-3, or LuIII that the RF DNA

synthesis of H-1 or H-3, but not LuIII, was

inhibited

by

ts14 in a

multiplicity-dependent

manner. wt H-1 did not inhibit H-3 RF DNA

synthesisunder thesameconditions.Since both

H-3 and ts14

produce

RF

rep'

proteins (Rhode, in press;

unpublished results),

this inhibition

on November 10, 2019 by guest

http://jvi.asm.org/

maybedirectlymediatedbyts14 DNA andnot

by theproteinsitcodes for.Themechanism by

which thisoccursisunknown.

Another feature of heterotypic infections be-tween wt H-1 or ts14 and H-3 or LuIII was a

multiplicity-dependent cross-interference for

HA synthesis. It is likely that LuIII inhibited

ts14RF DNAsynthesis (Table 4) by inhibiting

ts14 RF rep protein synthesis, since its own

proteins do not support H-1 (ts14) RF DNA

synthesis (Rhode, in press). Asimilar

cross-in-terference for the production of infectious

vi-rus wasobtained inheterotypicinfectionsexcept

that the magnitude of the inhibition was less

severethanthatfor HAsynthesis. Thisgreater

survivalof H-3orLuIIIinfectiousvirus

produc-tion during coinfection with H-1 was found to

be due, in part, to the production of H-3 or

LuIIIpseudotype particles. These virions have

H-3orLuIII DNAina capsid that contains

H-1protein.

The complex interrelationships amongthese

parvovirusesshouldbereflectedin the

similari-ties and differences in the base sequences of

theirrespectivegenomes.Thephysicalmapsof

H-1,H-3,andLulIRFDNAsfor the restriction

endonucleases Hindll, Hpall,

HindIII,

HaeII,

andEcoRI were compared. Using the method

described by Upholt (17), it can be estimated

from these data that H-1 and H-3 have about

95% homologyand thatH-1 andLuIIIare only

about80%homologous.These estimatesdo not

consideradditionsordeletions,and the

65-base-pair differenceinsize of theH-1 and H-3HindII

fragments at the 95% position (0 is at the left

endinFig. 2) maybeanaddition similarto the

oneinH-1 DI-1.The degreeofDNAsequence

homology between H-1 and H-3 or LuIII as

determinedbythesemappingstudiescorrelates

with the interaction of theseviruses at thelevel of RF DNA replication. But a more detailed

knowledge of the nucleotide sequences in the

regionsencompassingthe replication originsof

these viruses and at their promoters for

tran-,scription

is needed to better understand the

interactionsexhibitedbythem.

ACKNOWLEDGMENTS

This researchwassupported by theU.S. Public Health

ServicegrantCA07826-12from theNational Cancer Institute.

The parvovirus antisera used in this studywerekindly provided byHeleneToolan. Igratefully acknowledgeJessica

Bratton for expert technical assistanceandVirginia Haas and JaneenPrattforsecretarial duties.

LITERATURE CITED

1. Hallauer, C., G.Siegl, and G. Kronauer. 1972. Parvo-viruses as contaminants ofpermanent humancelllines. Arch.Gesamte Virusforsch.38:366-382.

2. Kongsvik,J.R., J. F. Gierthy, and S. L. Rhode mH. 1974.Replicationprocess of the parvovirus1. IV. H-1-specificproteins synthesized insynchronized human NBkidneycells.J.Virol.14:1600-1603.

3. Ledinko,N. 1967.Plaque assay of the effects of cytosine arabinosideand5-iodo-2'-deoxyuridine on the synthesis of H-1 virusparticles. Nature (London) 214:1346-1347. 4. Mahin, D. T., and R. T. Lofberg. 1966. A simplified method of sample preparation for determination of tritium"Cor3Sinblood or tissue by liquidscintillation counting. Anal. Biochem. 16:500-509.

5. Rhode,S.L.,EI. 1974.Replication process of the par-vovirus H-1. II. Isolation and characterization ofH-1 replicativeformDNA. J.Virol. 13:400-410.

6. Rhode.S.L., Im. 1974.Replication process of the par-vovirus H-1. III. Factors affectingH-1 RF DNA syn-thesis. J.Virol. 14:791-801.

7. Rhode, S.L.,m. 1976.Replicationprocess of the par-vovirusH-1. V. Isolation and characterization of

tem-perature-sensitive H-1 mutants defective in progeny DNAsynthesis.J. Virol. 17:659-667.

8. Rhode, S. L.,IH. 1977.Replication process of the par-vovirus H-1. VI. Characterization of a replication ter-minusof H-1replicative-form DNA. J. Virol. 21:694-712. 9. Rhode,S. L.,m. 1977.Replication process of the par-vovirus H-1. IX.Physical mapping studies of the H-1 genome. J. Virol. 22:446-458.

10. Rose, J. 1974. Parvovirus reproduction, p. 1-61. In H. Fraenkel-Conratand R. R.Wagner(ed.), Comprehen-sive virology, vol. 3. Plenum PublishingCorp., New York.

11. Sanger,F.,G. M.Air, B. G. Barrell, N. L. Brown, A. R. Coulson,J.C. Fiddes, C.A.HutchisonIH, P. M. Slocombe, and M. Smith. 1977. Nucleotide se-quence ofbacteriophage4X174DNA.Nature (London) 265:687-695.

12. Siegl, G. 1976. The parvoviruses. In S. Gard and C. Hallauer(ed.),Virology monographs.Springer-Verlag NewYork, Inc.,NewYork.

13. Singer, L. I., and S. L. Rhode MI. 1977. Replication process of theparvovirus H-1. VII. Electron microscopy ofreplicative-formDNAsynthesis. J. Virol. 21:713-723.

14.Singer, L. I., and S. L. Rhode HI. 1977. Replication process of theparvovirusH-1. VIII. Partial denatura-tionmappingandlocalization of the replication origin ofH-1replicative-formDNAwith electron microscopy. J. Virol. 21:724-731.

13.Tattersall, P.,P. J.Cawte,A. J. Shatkin, and D. C. Ward. 1976.Three structuralpolypeptides coded for byminutevirusofmice,aparvovirus.J.Virol.20:273-289.

16. Tattersall, P.,A. J.Shatkin, and D.C.Ward. 1977. Sequence homology between the structural polypep-tides of minute virus of mice. J. Mol. Biol.111:375-394.

17.Upholt,W. B.1977.Estimation of DNAsequence diver-gence fromcomparisonof restrictionendonuclease di-gests.NucleicAcidsRes.4:1257-1265.

VOL. 25, 1978

on November 10, 2019 by guest

http://jvi.asm.org/