Isolation and characterization of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate-early regulatory protein ICP4.

0022-538X/85/110558-13$02.00/0

Copyright© 1985,AmericanSociety for Microbiology

Isolation and Characterization of Deletion Mutants of Herpes

Simplex Virus Type 1 in the Gene Encoding Immediate-Early

Regulatory Protein ICP4

NEAL A. DELUCA, ALICE M. McCARTHY, AND PRISCILLA A. SCHAFFER*

Laboratory of Tumor Virus Genetics, Dana-FarberCancer Institute, andDepartmentof Microbiology and Molecular Genetics, Harvard MedicalSchool, Boston, Massachusetts 02115

Received 3 June 1985/Accepted 7 August 1985

Using Vero cellstranformedwiththe wild-typegenefor ICP4asthe permissive host cell,weisolatedherpes

simplex virustype 1(HSV-1)mutantscontaining deletions in both copies of the ICP4gene.Themutants,d120

and d202, contained deletions of 4.1 and 0.5 kilobases, respectively, in each copy ofICP4. ICP4 mRNA

synthesized in d202-infected Vero cellswas0.5kilobasessmaller than that synthesized in cellsinfected withthe

wild-typevirus. No ICP4 mRNAwasdetectedind120-infectedVero cells.d120andd202 specifiedpolypeptides

that reacted with ICP4 antiserum and were smaller than the wild-type ICP4 polypeptide by 130 and 30

kilodaltons, respectively. The only other HSV-1geneproducts detectableoninfection ofVero cells with d120 and d202 were ICP6 (of the early kinetic class of HSV-1 polypeptides), ICPO (immediate early), ICP22 (immediate early), and ICP27 (immediate early). Immediate-early polypeptides ICPO and ICP27 were

expressedtoahigherlevel inVerocellsinfectedwithanICP4 temperature-sensitive(ts)mutant(tsB32)at39°C, indicating immediate-early stimulatory activity associated with the ts ICP4 polypeptide. In addition, the patternsof complementation of d120, d202,andtsB32inICP4-transformed cells alsodemonstratedinhibitory activityassociatedwith the tsformof theICP4 polypeptide.

The viralgenes expressed initially afterinfection of sus-ceptible cells withherpessimplex virustype 1(HSV-1)have been

designated

immediate-early (or a) genes (3, 19). These genes are operationally defined as those transcribed in the absence of prior viral protein synthesis (3, 19). Five

imme-diate-early

mRNA species are translated into the polypeptides ICP0, ICP4, ICP22, ICP27,andICP47(19, 50). Functional immediate-early polypeptides are required for efficient expression ofearly

(3)

orlate (y) viralgenes(20).

The functional roles of individual immediate-early gene products in the regulated expression of HSV-1 genes have been examined by several methods. First,

temnperature-sensitive (ts)mutantsof HSV-1 containing mutations in the genesfor ICP4 and

1CP27

havebeenisolatedand character-ized (11, 35, 37, 40). Phenotypic analysis ofthese mutants hasshown thatICP4andICP27perform essentialreplicative functionsand are involvedspecificallyinmodulatingHSV-1 gene expression (11, 36, 40, 49). Second, reconstruction experiments involvingcotransfection of plasmids containing the immediate-early genes for ICP4 and ICP0 and suitable

testgenesindicate that both stimulateexpression ofHSV-1

early

genesin trans(12, 30,39). Finally, deletionmutantsin the gene for ICP22 have been shown to be capable of growing in Vero cells, a cell line which is routinely usedas a

laboratoryhost for

HSV.

Theinvolvement ofthe immediate-early polypeptides ICP22and ICP47 in HSV-1 gene expres-sion, however, remains obscure.

The most extensively studied immediate-early gene of HSV-1 is thatencodingthe

175-kilodalton

polypeptideICP4. Alarge number

of

temperature-sensitivemutantsin the gene forICP4 exist.Collectively, theydefine HSV-1 complemen-tationgroup

1-2

(43). Acommon characteristic ofts mutants

inthisgroup isthattheyinducedramatic overproduction of immediate-early polypeptides atthe nonpermissive

temper-*Correspondingauthor.

ature. Based onthis observation and the resultsof temper-ature-shift experiments, it has beenpostulated thatICP4 is

required for negative regulationorattenuated expression of immediate-early genes (11, 36). Other features of cells in-fected with most ICP4 ts mutants at the nonpermissive temperatureareunderrepresentationorabsenceof early and late polypeptides and absence ofviral DNA (11, 36). The basis for thesephenotypic propertiesis thatwild-type ICP4 is

required

for the accumulation ofstable transcripts from early and late genes, and the products of early genes are

required for viralDNA synthesis. Asubset oftsmutantsin group1-2inducethesynthesis of appreciable levels of early polypeptides and viral DNAatthe nonpermissive

tempera-ture(8). The absence of detectable infectious virus in such mutantsis attributable to theabsence of late polypeptides. Basedonthecumulativeevidence, itisthoughtthatICP4is (i) required for the expression ofearly and late genes, (ii) autoregulatory,and(iii)involvedinthe inhibition of

expres-sionofotherimmediate-earlygenes.Furthermore,it appears thatthemechanismof actionof ICP4 involves modulationof gene expression at the level of accumulation of stable transcripts.

Asjust stated, ourcurrentunderstanding oftheactivities of andrequirementsfor

wild-type

ICP4 is basedprimarilyon

the phenotypic properties of ICP4 ts mutants at the nonpermissive temperature. The conclusions drawn from studies ofthesemutants rest in partonthe

assumption

that the temperature-sensitive forms of the ICP4 polypeptide havenoinfluenceonHSV-1 geneexpression.This may not, infact, be thecase. Recently, it has been shown that ICP0

can actintrans tostimulateexpressionfrom

early

promoters inacotransfectionassay(30, 39).

This

presehts an interest-ing paradox in that most ICP4 ts mutants overproduce immediate-earlypolypeptides (including

ICPO)

and yet failto expressearly

polypeptides.

Therefore,eithercotransfection experiments donotreflectan eventoccurring in viral infec-558

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.612.60.298.86.180.2]

HSV-1 DELETION MUTANTS IN THE ICP4 GENE 559 TABLE 1. Growth of HSV-1 ts mutants on 2-6 and Vero cells

Complement- Defective Virus grownona

Vlrusationgroup gene 2-6 Cells Vero cells

KOS 8.0 x 106 3.7 x 106

tsB32 1-2 ICP4 7.8 x 105 2.1 x 102

tsB21 1-2 ICP4 6.0 x 104 5.0 x 100

tsL14 1-2 ICP4 1.1 x 105 1.7 x 102

tsY46 1-5 ICP27 2.5 x 101 1.0 x 101

tsJ12 1-9 gB 4.0 x 102 3.6 x 102

tsA24 1-1 ICP8 3.0 x 101 3.0 x 101

a2-6 and Vero cellswereinfected with the indicated virusmultiplicity of infection of 1.0 PFU per cell and incubated at39'C for 18 h. The cultures were then harvested and assayed on Vero cells at34'C.

tion or, more interestingly, the tsform oftheICP4 polypep-tide has aninhibitoryeffect onearly gene expressionor the transactivity of ICPOorboth. Inaddition,theisolation ofts mutants in the structural gene for ICP4 which exhibit dis-tinctly different regulatory phenotypes (8) suggests that the ts form(s) of ICP4 may exhibit activities capable of modu-lating gene expression to a greater or lesser degree.

To betterdefinetheroleofICP4in the lifecycle ofHSV-1 and to answer questions regarding the possible regulatory activities oftsforms ofthisprotein on HSV-1 gene expres-sion, we isolated deletion mutants ofHSV-1 which induce the synthesis of truncated forms ofthe ICP4 polypeptide. BecauseICP4performsanessentialreplicative function,we developed a system for screening and propagating ICP4 deletion mutants by constructing cell lines expressing the wild-type ICP4 gene. Such a system has proven invaluable forthestudy of deletionmutantsin themajortranscriptional regulatory protein of adenovirus, EIA (22). Although cells transformedwith HSV-1 genes have been shown to comple-ment ts mutants(7, 41), suchcellshave notbeenusedtodate to isolate deletion mutantsofHSV-1. In the present study, twoICP4 deletionmutants wereisolated andused to address thefollowing questions. (i)What is thepolypeptide pheno-typeof cells infected witha mutantlackingafull-length ICP4 polypeptide, and (ii) doesthisphenotype differ fromthat of ts mutant-infected cellsinwhichafull-sizeyet temperature-sensitiveform oftheICP4 polypeptideis produced?

(kb) Saclfragment (Fig. 1) to yield pKX2-ASacI. pKX2-P4 was constructedby insertion of the 220-base-pair EcoRI-to-SmaI fragment(-110 to -330[Fig. 1]) into the EcoRI site of pKX2 in the proper orientation. TheSmaI site at -330 was modified with EcoRI linkers. DNA-modifying enzymes and EcoRI linkers were obtained from New England BioLabs, Inc. (Beverly, Mass.) and used as prescribed by the manu-facturer.

Nucleic acid isolation. Purified HSV-1 (14) and bacterial plasmid DNAs were isolated as described previously (8). Cytoplasmic RNA wasisolated from Nonidet P-40 extracts of Vero cells free ofnuclei,as previously described (24).

Transformation of Vero cells. pSV2neo confers resistance to the antibiotic G418 (geneticin; GIBCO Laboratories, Grand Island,N.Y.)whenintegrated intocultured mamma-lian cells (46).Vero cells were cotransformed with pSV2neo and HSV-1 ICP4 as follows. pSV2neo (0.5 p.g) was coprecipitatedwith 5.0 ,ug ofpKXCin thepresence of 15 ,ug ofsalmon testes DNA(SigmaChemicalCo.,St. Louis,Mo.) in a total volumeof1.0 mlbytheprocedure ofGraham and van der Eb (16). Freshly trypsinized Vero cells (4 x

106)

were added to the precipitate and incubated at 370C with continuous shakingfor 30 min. The cell-DNA mixture was suspended in20mlof culturemedium andtransferred to two 85-mm petri dishes. After a 4-h incubation at 37°C, the

adsorbed cells were subjected to glycerol (15%) shock and incubated for an additional 2 days at 37°C. The cells were then trypsinized and seeded into 85-mm petri dishes at a densityof 5 x

103

cellsper cm2. After the cells had adsorbed to the dish, themedium was removed and medium contain-ing1.0mgof G418per ml wasadded.After 5 to 10days,the G418concentrationwaslowered to 400,ug/ml.After 12 to 18 days, individual G418-resistant colonies were isolated, am-plified, and subsequently screened for ability to support replication of tsB32 at 39°C in a plaque assay. Approxi-mately20% oftheG418-resistantclonesexhibited detectable complementing activity. One cellline, 2-6,was retained for further study.

Cell lines 1-2, 1-5, and 4-2 were generated in a similar manner, except that pKX2-P4 rather than pKXC and d120 and d202 were used to generate and screen G418-resistant clones, respectively. Among

pKX2-P4-transformed

cells,

MATERIALS ANDMETHODS

Cells and viruses. Procedures for the growth and maintenance of African green monkey kidney cells (Vero, ATCC P130; CV-1, ATCC P27) were previously described (51). Nero cells are an unclonedpopulation of G418-resistant Vero cells generated bytransformation withpSV2neo.

Thewild-type KOS strain ofHSV-1 and theKOS-derived temperature-sensitive mutants (Table 1) were propagated and assayed as previously described (8). tsB32, tsB21, and tsL14 are mutants in the structural gene for ICP4 (8, 11).

Plasmids. pSV2neo contains the bacterial gene for neomycin resistance under control of the simian virus 40 early promoter (46). pKXC (Fig. 1) contains the XhoI C fragment of KOS and was constructed as

described

previ-ously(10). pKBZandpKX2(pKEB-X2) are diagrammedin Fig. 1and were constructed asdescribedpreviously (8).

Plasmids pKX2-ASacI and pKX2-P4 were constructed from pKX2 in the following manner. pKX2 was partially digested with Sacl, religated with T4 DNA ligase, and transformed into Escherichia coli. Ampicillin-resistant clones were screened only for deletion of the 0.5-kilobase

0 81 0 82 083 084 085 086 087

1 1

11 LI1 1 T

0 2 3 4 5 6 7 8 9 10

X X

m u

-kIb

E pKXZP4

-X pKX2 -pKX2-&SocO

-pK8Z

E

C C E

8 83

ICP4 mRNA EMORIs

FIG. 1. HSV-1 DNA inserts inplasmidsused in this study.The regionof the HSV-1 genome in theIsorientation(17) between map units(m.u.)0.806 and 0.877containingoris(47),the ICP4upstream regulatory sequences(closed box) (23),theICP4message(50),and thea sequence (27) is shown with respectto the HSV-1 inserts in pKXC, pKX2-P4, pKX2, pKX2-ASacI,andpKBZ.Thederivation of these plasmids is described in Materials and Methods. The relevantrestriction sites shownareEcoRI (E), SmaI (M), BamHI (B), SacI (C), andXhoI (X). The dashed line represents the L-S junctionof the viral genome.

R7s -8 1- "

a

VOL.56, 1985

DKXC

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.612.319.560.528.626.2]

approximately 40% of the G418-resistant clones demon-strated complementing activity.

CAT assays. The transfection of HSV-1-chloramphenicol acetyltransferase (CAT) chimeras and in vitro assays for CATwereperformed byamodification (10) of the procedure of Gorman et al. (15). The construction of pIE3CAT, ptkCAT, and p5CATwas described previously (10). These plasmids contain the CAT gene under control of HSV-1 immediate-early, early, and late promoters, respectively.

Electrophoresis. Lysates of [35S]methionine-labeled in-fected cell cultures were prepared and electrophoresed on sodium dodecyl sulfate (SDS)-polyacrylamide gels as de-scribedpreviously (8). DNA fragments generated by restric-tion endonuclease digestion were separated on horizontal agarosegels. Cytoplasmic RNA was separated by electro-phoresis on agarose gels containing 2.2 Mformaldehyde by the procedure of Goldberg (13).

Blot hybridization. The treatment of DNA restriction fragments inagarosegels, subsequenttransferto nitrocellu-lose, and hybridization were performed by the method of Southern (45). Formaldehyde gels containing separated RNA werewashedwithwater, soakedfor30 minin 50 mM NaOH-10 mM NaCl, soaked for 30 min in 0.1 M Tris hydrochloride (pH 7.5), and then soakedfor 30 min in 20x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate). RNA wasthentransferred tonitrocellulose by the procedure of Southern (45). RNA blots were hybridized to denatured probe in50%formamide-5x SSC-10 mM Tris hydrochloride (pH

7.5)-i

x Denhardtsolution-25 ,ug of salmon testes DNA per ml-10%dextran sulfate at42°C for 18 h. ForNorthern blot hybridization, 32P-labeled probe was NaOH (0.5 M) denatured in the presence of 1 mgofsalmon testes DNA. The solution was neutralized by addition of 0.2 M Tris hydrochloride (pH 7.5)-0.5 M HCI. Sequential washes of 3 and 0.1x SSC were performed at 50°C. Probes were pre-pared by nick translation (25) in the presence of[32P]dGTP and

[32P]dCTP

(AmershamCorp., Arlington Heights, Ill.).

Western blot analysis. Western blot analysis of gel-separated SDS peptides was conducted as previously de-scribed (2, 48), with the following modification. Before reaction with antibody, the nitrocellulose was incubated overnightat4°C with BLOTTO(21). Theantibody reaction

was performed in 10 ml ofBLOTTO-0.1 ml of ICP4 rabbit antiserum. The filter was then stained with I251-labeled protein A(Amersham). Rabbit ICP4 antiserum (6)was the generous gift of

Richard

Courtney (University ofTennessee, Knoxville).

RESULTS

Derivationof a celllineexpressing ICP4. ICP4performsan

essential replicative function. To propagate viral mutants

deleted in the ICP4 gene, we constructed cell lines which express thewild-typegenefor ICP4. Viable cellsexpressing this gene have been isolated andcharacterizedpreviously(7, 32),ruling outthepossibility of inherentcytotoxicity of the resident ICP4 gene or itsproduct.

Verocellswerecotransfected withplasmids pSV2neo and pKXC (Fig. 1). After2weeks, G418-resistantcolonieswere

isolated, amplified, and screenedforabilityto support rep-lication of tsB32 at 39°C. tsB32 is a temperture-sensitive

mutant in the structural gene for ICP4 (11). Approximately 20% of the G418-resistant lines isolated were capable of complementing tsB32. The cell line designated 2-6 demon-strated the greatest complementing activity and therefore

waschosenfor further study. As wellasall other cell linesto

BamHIZ _ _

I

*

--4~~~~~..

A

_

to_A_

P5 P20 C I

2-6 0

3 10 30

FIG. 2. Southern blot analysis of HSV-1 DNA in 2-6 cells. Electrophoretically separated BamHI digests of2-6 cellDNA (10 ,ug)(frompassages5 and20)andVerocellDNAweretransferredto nitrocellulose paper and probed with the 32P-labeled BamHI Z fragment(1.84 kb) frompKBZ(Fig. 1). We alsoincluded BamHI-digestedpKBZ (7.8kb)tovisualize1(25 pg), 3 (75 pg), 10 (250 pg), and 30(750pg)copies of viralDNAper 3 x 109basepairsofcellular DNA.

be described in this study, 2-6 cells are indistinguishable fromVerocellswith respecttomorphologyandgrowthrate.

Todemonstrate the presenceof ICP4codingsequencesin 2-6cells andtodetermine theapproximatenumberof copies of the ICP4genepresent, wecompared DNA from 2-6 cells with Vero cell DNA bySouthern blot analysis (Fig. 2). The probe used in this experiment was the purified BamHI Z fragment obtained from plasmid pKBZ (Fig. 1). This

frag-mentcontainsapproximately50% of thecodingsequencefor ICP4. The standards present on the right half of the gel indicate that 2-6cells contain approximately five copies of the BamHI Zfragmentper3 x 109basepairs.Moreover, 2-6 cells appear stable with respect to the integrity of these sequences over 15 passages(Fig. 2).

2-6 cellsexpressbiologically active ICP4. 2-6 cells contain approximately5copies ofthe BamHI Zfragmentperhaploid equivalent or 10 copies per cell. This does not necessarily imply thatall five copies serve astemplatesfor the produc-tion of ICP4. The measure ofICP4 expressionused in this studywasthecomplementationof ICP4 ts mutants at39°C. Table 1 shows thegrowth characteristics ofvarious HSV-1 ts mutants on2-6 and Vero cellsat 390C. tsB21,tsB32, and

tsL14 each specifyathermolabile ICP4polypeptide (8, 11). tsA24, tsJ12, andtsY46contain mutations in ICP8(51), gB (9), andICP27 (40),respectively. Inthisexperiment,2-6and Verocellswereinfected with1PFUof theindicatedmutant

per cell at 39°C. The cultures were harvested at 18 h

postinfectionandassayedon Verocell

monolayers

at34°C. Onlymutantsdefective inICP4grewto substantially higher levels (3to4ordersof

magnitude)

in 2-6 than in Verocells. However, eithercomplementation was not

complete

or the

mutantforms of ICP4 wereabletointerfere with the action of thecomplementing wild-typeICP4polypeptide,sincethe

yieldof mutant virus was 1 to 2orders of

magnitude

below

9 9

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.612.351.521.71.298.2]

HSV-1 DELETION MUTANTS IN THE ICP4 GENE 561

Vero

0 1 3 10 K K

2-6

_

__"_F

~..md

fI

e. >* v;X -i $.

4

5

6 0

8

11

-.S

am aw.- 25

_, &a-* 27

[image:4.612.91.255.71.410.2]

*6 meMO

FIG. 3. Synthesis ofviral polypeptides in tsB32-infected Vero and2-6 cells.Monolayers of 2-6and Vero cellswereinfected atthe indicatedmultiplicities ofinfection with tsB32 at39°C. Cells were

also infected with KOS ata multiplicity of infection of 3 PFUper cell(K).[35S]methioninewaspresentduringincubation from 5to16 h postinfection. At 16 h postinfection, total cell lysates were

preparedfor SDS-PAGE. The infected-cellpolypeptides were elec-trophoretically separatedona 9% polyacrylamide gelcross-linked with N,N'-dicyclohexylcarbodiimide (28). Infected-cell polypeptides representative of the x (ICP4, ICPO, and ICP27), 1B

(ICP6, ICP8, ICP11,andICP36),and-y(ICP5, ICP19-20,andICP25)

kinetic classes arelabeledfor reference (28).

that of KOS in 2-6 cells. The diminished yield of ICP4 ts

mutants in 2-6 cells at 39°C relative to wild-type virus was not significantly reflected in the 39-34°C plating efficiencies of the mutants on2-6 cells. The plating efficiency (39-34°C)

ofwild-type viruson2-6 cells wasnear1.0, whereas that of themutants ranged from0.3to 0.8. The plaque sizes of the mutants on 2-6 cells at 39°C, however, were substantially

smaller thanthatof KOS (datanot shown).

An alternative means of assessing the extent of comple-mentation of ICP4 ts mutantsin 2-6 cellsisto compare the

synthesis ofviralpolypeptidesin2-6 and Vero cellsat39°C.

InVero and other cell types, tsB32 characteristically

over-produces immediate-early polypeptides and fails to

synthe-sizeearlyand latepolypeptidesat39°C (11).This is

presum-ablydue tothe lack offunctional ICP4 at thistemperature. The complementation of tsB32 by wild-type ICP4

synthe-sized in 2-6 cells, asreflected bythe synthesis of early and late viralgeneproducts, wasboth substantial and

multiplic-ity dependent (Fig. 3).Atamultiplicityof 1 PFUpercell,the

polypeptide profileof tsB32 in2-6 cells resembledthat ofthe wild-type virus.Atmultiplicitiesgreaterthan 1 PFU per cell, however, characteristics of the mutant phenotype were clearly evident and becamemore pronounced with increas-ing multiplicity. A similar multiplicity effect has also been observed forts mutants inthe HSV-1DNA-binding protein in transformed cells (41). 2-6 cellscontainapproximately10 copies of the BamHI Z fragment per cell, yet this number

was notsufficient to restore completely the wild-type phe-notype atthe multiplicities of infection tested. Three possi-ble explanations existfor this observation. (i) Only a small fractionof theintegratedICP4 sequencesserve astemplates for theexpressionof ICP4.(ii)ICP4 mRNA isnot

expressed

asefficientlywhen its templateisintegratedinto the cellular genome as in the viral genome. (iii) The mutant ICP4 polypeptide interferes with the action of wild-type ICP4 present in transformed cells. Data

presented

below support the third

possibility.

Derivation ofa cell lineexpressing ICP4but lacking oris. The fragment of HSV-1 DNA used to generate 2-6 cells

contains,

in the

following order,

the

cis-acting

packaging

sequences(a)(27)derived from the

joint region

orterminiof the viral genome, the gene for ICP4 (26,

28, 38),

the regulatory sequencesfor

expression

of ICP4 mRNA

(5, 23),

andanoriginof viral DNA

synthesis (oris) (47) (Fig. 1).

As shown previously (7), passage of ICP4 ts mutants in such

cells resulted in the

generation

of ts+ recombinants. In the present study, the number of recombinants

generated

in

complementation

tests after

amplification

ofa

single

tsB32 plaque isolate in 2-6 cells was

approximately

1%

(data

not

shown). Because this level was well below the

yield

ofts mutantvirus

generated by

complementation

in these

cells,

it

was felt that wild-type recombinants would not

seriously

interfere with the isolation ofdeletion mutants.

Therefore,

2-6cells wereusedtogenerateandscreen

putative

deletion

mutants in

ICP4,

as described below. To obtain recombi-nant-free stocks of the ICP4 deletion mutants,

however,

it

was necessary toisolate ICP4-transformed cell lines which would not give rise to

wild-type

recombinants. Conse-quently, the deletionmutants

generated

as described below

were used to screen G418-resistant cell lines isolated after cotransfection of

pSV2neo

and

pKX2-P4

(Fig. 1). pKX2-P4

containstheasequences,

ICP4,

and the

regulatory

elements for ICP4

expression

but lacks

oris.

It was

anticipated

that deleting

oris

would

reduce,

ifnot

eliminate,

the

generation

of wild-type recombinants.Recently,it has beendemonstrated thatthis isindeedthecase

(7). Among

the celllines isolated after cotransfection with

plasmids pKX2-P4

and

pSV2neo,

approximately

40% were able to support

growth

of the deletion mutants. Three were retained for further

analysis.

The three

lines,

1-2, 1-5,and4-2,contained

three, three,

and five copies, respectively, of the ICP4

coding

sequence per

haploid equivalent

of DNA

(data

not

shown).

Isolation of deletion mutants in 2-6 cells. A

plasmid

con-taining

a 0.5-kb deletion in the ICP4

coding

sequences was constructed from

pKX2

by partial

Sacl

cleavage

and

religa-tion

(Fig.

1). The deleted

ICP4-containing

plasmid

pKX2-ASacI was first assayed for abilityto express ICP4activity before attempts to introduce the deletion into the viral genome. This was done

by

cotranfecting

CV-1 cells with either pKX2 or

pKX2-ASacI

and HSV-1-CAT chimeric genes and

measuring

the levels of

expression

of the HSV-1-CATchimeras induced

by wild-type

ICP4 and the deleted form of the

protein.

In this way the effect of the 0.5-kb deletionon ICP4

activity

could be assessed. We and others have shown that ICP4 alone can suppress

immediate-early

VOL.56, 1985

'A ii ii&Ai lmw ".0.0.w

v

'.c. 1.

-M,41

414 I

on November 10, 2019 by guest

http://jvi.asm.org/

CAT chimeric gene expression (10) and simultaneously

enhance early andsomelategeneexpression intrans(10, 12, 30). Here, immediate-early-CAT (pIE3Cat) expression was

suppressed whereas early-CAT (ptkCat) and late-CAT (p5Cat) expressions wereenhanced bywild-type ICP4 (Fig.

4). The CATsignal generated fromextractsof cells cotrans-fected with the deleted ICP4 plasmid resembled that derived fromextractsof cells cotransfected with pBR322. Therefore, the0.5-kb internal deletion abolished the trans-acting activ-ity ofICP4.

pKX2-ASacI wasthenlinearized andcoprecipitated with

infectious KOS DNA, and 2-6 cellswerecotransfected with

the mixture. Individual progenyplaques from this

transfec-tionwereisolated and screened for ability toreplicate in 2-6 cells and not in Vero cells. Of 224 plaque isolates, 2, designated d120 and d202, exhibited impaired ability to replicate inVero cells.

Stocks of d120 and d202 generated in 2-6 cells always contained 1 to 5% ([titer of Vero cells/titer of 2-6 cells] x

100) wild-type recombinants which produced plaques in Vero cells. Such recombinants, when isolated and plated in Vero and 2-6cells, always yielded platingefficienciesof 1.0. When individual plaque isolates of d120 and d202 were

propagated in 1-2, 1-5, or4-2 cells, wild-type recombinants wererarely detected. Occasionally, however, when d120or

d202was amplified through several rounds of growth in

ori-ICP4+ cell lines, wild-type recombinantsweregeneratedata

frequency approaching

10-5.

The growth characteristics of d120 and d202 on control Nero (G418-resistant Vero cells

[seeMaterials and Methods]), 1-2, 4-2, 1-5, and2-6 cellsare

shown in Table 2. In thisexperiment, wild-type virus was not present in lysates obtained from ori- cells. Recombi-nant-free stocks of d120 and d202 grown in 4-2 cells were

usedfor further analysis.

Characterization of deletions in d120 and d202. d120 and

d202 aretheprogeny ofcotransfection of pKX2-ASacI and

KOS DNA. To determine whetherthese isolates contained the deletion in pKX2-ASacI, we analyzed viral DNA from

d120 and d202by restriction enzymedigestion.

Electropho-retically separated restriction fragments of KOS, d120, and d202 were transferred to nitrocellulose paper and probed with theindicated 32P-labeled fragment (Fig. 5).

The isolate d202 contained the expected 0.5-kb Sacl deletion in bothcopies of ICP4. The 3.2-kb HincIIfragment ofpKX2 comigrated withthat ofKOS, whereas theHincIl

plE3Cat ptkCat p5Cat

04 X'R 41 X X0 4

('I~~~~~~~~~~~~~~~~~~

(a4 C'o4 C r

C') X XII IL X4

j-('4 x

* *

b' a' c'

ICPO +ICP 4

X2

[image:5.612.354.524.72.350.2]

X2Asac1

FIG. 4. trans-Inducing activities of wild-type and deleted forms of ICP4. CV-1 cells were cotransfected with 2 ,ug each of the

indicated CAT plasmid and the indicated test plasmid. At 40 h posttransfection, cellextractswerepreparedand analyzed forCAT activity in vitroas previously described (15). Shownare the auto-radiographic images of the chromatographically separated input,

unacetylatedchloramphenicol (lowerspot),andthe acetylated prod-ucts (upper spots). Alsodiagrammedarethetestplasmidsusedin the cotransfection with respectto the genes forthe infected cell

polypeptides ICP4 and ICPO. b', a', and c'representtheelementsof the HSV-1joint region(17).

fragment of pKX2-ASacI comigrated with that of d202. Furthermore, theSaclfragments of d202homologoustothe 3.2-kb HincIl probe were identicaltothose of KOSexcept

for the absence of the 0.5-kb deletion inpKX2-ASacI. The HincIl and Sacl digests of d120 DNA indicated a

TABLE 2. Growth ofICP4mutants ontransformed cells Virusgrown ona

Virus

Nerocells 4-2Cells 2-6 Cells 1-5Cells 1-2Cells

KOS 1.9 x 108b 2.2 x 108 1.4 x 108 2.0 x 108 9.0 x 107

1.8 x 108c 1.5 x 108 1.1 x 108 1.7 x 108 1.0 x 108

d120 2.5 x 104b 4.0x 107 4.9 x 107 5.2 x 107 4.2 x 107

<5c <102 1.7 x 105

<102

<102

d202 7.2 x 104b 6.5 x 107 3.7 x 107 6.5 x 107 3.3 x 107

<5C <102 2.2 x 105 <102 <102

tsB32 6.2 x 102d 1.0x 104 3.0 x 104 2.0 x 103 2.0x 103

<5e <5 2.0 x 102 <5 <5

aNero,4-2, 2-6, 1-5, and1-2cellswereinfectedwith theappropriatevirusunderconditions identicaltothose stated inFig.9.

bInfectedcultureswereharvestedat18 hpostinfectionandassayedat34°Con1-2cells.

' Same cultureasthatdescribedinfootnote bassayedsimultaneouslyat34°ConNero cells.

dInfected cultureswereharvested at 18 hpostinfectionandassayedat34°ConNerocells.

eSame cultureasthatdescribedinfootnotedassayedsimultaneouslyat39°ConNerocells.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.612.68.564.546.679.2]

HSV-1 DELETION MUTANTS IN THE ICP4 GENE

Hunc II; Sacl EcoRI BamHI Sal I PvuII

io o X VI cm

a t4~x£~ 0 0

x C a9 xgx

-lb

40-.

4

4.4

00 a.

0

00 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8r 0.3 1.0

I I I I I 1 I I I, 1I

o b

>~~

b'a"c' c a

-

..

... ... ... ...

0.81 082

os

-

-0.84

0.85 0.86 \

\ f ..

I I I 1 1 I

I}

|

x

-?1

2 3 14 5 6

SI

8 a

t-_

I,__C

I--1C

C I -_,. HCCc P S/

Probe

k. b. E (Hl

r

--b a'o c'I

rCPAM

c a

KOS

crzz=

d202

ar-

d120

FIG. 5. Genomestructuresofd120 andd202. Viral (KOS, d120, and d202) and plasmid (pKX2 and pKX2-ASacI) DNAsweredigestedwith

theindicated restriction enzymes and electrophoretically separated on a1.2% agarose gel. Separated DNAfragments wereanalyzed as

described in thelegendtoFig. 2 with the indicated 3.2-kb HincIl fragmentasprobe. Forcomparison of KOS and d120 fragments (right half

of thegel), terminal (0) and joint (0) fragmentsareindicated. Shown below the autoradiogram in descending orderaretheKOSgenome,an

expansion of thejoint region of thegenomebetween 0.806 and 0.867mapunits(m.u.),thelocations ofrelevantrestriction sites,thelocation

ofthefragment usedasprobe, and the deducedstructuresof d120 and d202. Theopenand shaded boxesin thestructuresofd120andd202 indicate deletedsequences andregions of uncertainty, respectively. The boundaries between deleted sequencesandregions of uncertainty ind120DNAwerededuced fromthefollowing observations. (i)d120containsa4.1-kbdeletion. (ii) The PvuII site ispresentatm.u.0.863.

(iii) The functionalasequences arepresent atthejoint,asdemonstrated by evidence of inversionof L and S. (iv) The restriction sites spanned by theopenboxind202areabsent in d120. The right portions of the KOS, d120, andd202genomes arediagrammedin the lower portion of thefiguretoillustratetheeffects of deletions in both oftheinvertedrepeats(C and C')onthesizesof the shortcomponentsof thegenomes.

Adashedline is drawntoindicatetheL-Sjunction. The relevant restrictionsitesarethesameasin Fig. 1, with the addition ofHinclI (H),

PvuII (P), andSall (S).

-5.3.

.3.2

.41.84

-.cl24 -a1.O .c0.95

.40s8

{0.30

-d 120

d202

MD

I

11

VOL.56, 1985 ₅₆₃

I I I

I I

i

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.612.155.456.70.605.2]

more substantial deletion. The structure of d120 was ana-lyzed further by Southern blot hybridization. The right half ofFig. 5 is an example of one such experiment. The 5.3-kb terminal EcoRI K fragment of the KOS short component was absent in d120 DNA, and a new 1.24-kb band was present. The deletion in d120 spanned the twoBamHI sites located near 0.853 map units (Fig. 5) in KOS, and a new 1.1-kb band was present in d120. The 5.1-kb terminal Sall fragment of KOS was replaced by a 0.95-kb fragment in d120. Likewise, the4.7-kbterminal PvuII fragment of KOS wasreplaced by a 0.58-kb fragment in d120. The size of the novel terminal fragments in d120 was consistent with the separation between, and order of, the EcoRI, BamHI, SalI, and PvuII sites near the 5' terminus of the ICP4 gene in wild-type HSV-1 (Fig. 5). The sizes of the reductions in the abovementioned terminal fragments (Fig. 5) and in the joint fragments (Fig. 5; data not shown) are consistent with the existenceof a 4.1-kb deletion in both copies of ICP4.

Based on other Southern blots (data not shown), restric-tion sites in the long component of the genome did not appear to be affected in either d120 or d202, and no se-quenceshomologous to pBR322 were detectable in the DNA of eithervirus. In addition, the expected patterns generated by genomeisomerizationwerereflected inEcoRIandPvuII

44.3

- 3.8

< 2.8

b' ca0' c'

r1

1:=

ICPO ICP 4

X2

[image:7.612.98.267.338.627.2]

X2 Sac1

FIG. 6. ICP4 mRNA synthesized in d120 and d202. Vero cell monolayers were preincubated in thepresence of 75 ,ugof cyclo-heximideperml for 1 hat37°C. Monolayerswerethen infected with 10 PFU ofKOS,d120,ord202percell in the presenceof 75 ,ugof cycloheximideperml. After 6 h of incubationat37°C, the cultures

wereharvested andcytoplasmic RNAwasanalyzedasdescribed in

the text. The locations of the ICPO and ICP4 mRNAs and of the

deletioninpKX2ASacIusedtogenerate d202areshown beneath the autoradiogram. The numbersontherightindicate size inkilobases.

b', a', and c'aredescribed in thelegend toFig.4.

!,

4 "w4v49

4

K64

infected asdescribed in the legendto Fig. 6. At 6 hpostinfection, cycloheximide was washed outwith medium containing 10 ~±g of actinomycinDperml.Incubationwascontinuedforanadditional 3 h in the presence of 10 ~tg of actinomycin D and 50 p.Ci of [35S]methionine per ml. Cell lysates were then prepared and

ana-lyzed bySDS-PAGEasdescribed in thelegendtoFig.3. Numbers

onthe leftaredescribed in thelegend toFig. 3.

digests (Fig. 5; data notshown). Thus, because the presence of a sequences at both thejoint and molecular termini are required for isomerization (33), their presence in d120 in a functionalconfigurationis implied.

ICP4 mRNAsynthesis by d202 and d120. d202 contains the 0.5-kb deletion introducedbypKX2-ASacI.Therefore,d202

was expectedto synthesize an ICP4 mRNA 0.5 kb smaller than that of the wild-type virus. d120 contains a large deletion (4.1kb)andthereforewasexpectedtosynthesizea

muchsmaller messageor none atall. Northern blotanalysis ofcytoplasmic RNA from KOS-, d120-, and d202-infected

Vero cells verified that thiswas the case (Fig. 6). Infected Verocellswereincubated in the presence ofcycloheximide for 6 h. Cytoplasmic RNAwasthenisolated, electrophoret-ically separated, transferred to

nitrocellulose,

and probed with nick-translatedpKX2 DNA.pKX2 contains sequences homologous to the immediate-early mRNA for ICPO and ICP4. KOS induced the synthesis ofa4.3-kb mRNA speci-fyingICP4 anda2.8-kb mRNAspecifyingICPO(50) (Fig.6). d120 synthesized only the 2.8-kb message, whereas d202 synthesizeda3.8- anda2.8-kb message.

Therefore,

although ICP0 mRNA was synthesized in cells infected with the deletion mutants, as it was in KOS-infected cells, ICP4

t,

0

0

40

"40

on November 10, 2019 by guest

http://jvi.asm.org/

HSV-1 DELETION MUTANTS IN THE ICP4 GENE 565

Mock

d120 d202

KOS

.Avmm

170

140

&~~~~~~~~~~~~~~~~~~~~~~~

_

1~~40

FIG. 8. Western blot analysis of d120- and d202-infected Vero cells. Non-ICP4-containing cells were infected withthe indicated

virusatamultiplicity of10 PFUpercell.At5hpostinfection, SDS

lysateswerepreparedandrunon a9%SDS-polyacrylamide gel.The SDSpeptideswereelectrophoretically transferredtonitrocellulose andanalyzedasdescribedin Materials and Methods. Numberson

theright indicateapparentmolecularweight.

mRNA was 0.5 kb smaller in d202-infected cells and not detectableby thisprocedure in d120-infected cells.

Immediate-early polypeptide synthesis in d120- and

d202-infected Vero cells. To determine what immediate-early polypeptides where synthesizedin d120- and d202-infected Vero cells, we performed an experiment similar to that shown inFig. 6, except thatat6 h postinfection,

cyclohex-imidewas washedoutand incubationwascontinued for 3 h in the presence of actinomycin D and [35S]methionine. Imnmediate-early polypeptides ICPO, ICP22,andICP27were

clearlyvisible inextracts ofdl20-, d202-,andKOS-infected cells on SDS-polyacrylamide gels (Fig. 7). As in other

cycloheximide reversal experiments, small quantities of ICP6were also observed. The ICP4polypeptide, however, was detectable only in extracts of KOS-infected cells.

Whereas one might not expect an ICP4 polypeptide to be

synthesized in d120-infected cells, it is possible that a deletedform of ICP4wassynthesizedind202-infectedcells.

Although this is not evident inFig. 7, the possibility ofan ICP4-like peptide comigrating with a peptide ofa different

electrophoretic mobility than ICP4 exists. This possibility

was investigated furtherby Western blot analysis. Electro-phoretically separatedSDS peptides preparedfrom infected

anduninfected Nero cells were transferred to

nitrocellulose,

probed with anti-ICP4 rabbit serum, and visualized by staining with

'25I-labeled

protein A. The anti-ICP4 serum used was generated from rabbits inoculated with a gel-eluted SDS ICP4polypeptide (6). Such polyclonal antibody would be expected to react efficiently with truncated SDS ICP4 peptides. Cells infected with d202 contained a polypeptide shorter than KOS by about30kb (Fig. 8). This polypeptide probably comigrated withICP6 orICPO (Fig. 7). The d202 formof ICP4appeared relatively unstable with respect to the wild-typeform of ICP4, asreflectedby the descending smear and several specific lower-molecular-weight bands. The or-igin of the lower-molecular-weight band in KOS is unclear and may represent a

degradatiqn

product of the wild-type ICP4polypeptide. Itisnot,inall likelihood,acellular band

or another HSV-1 immediate-early peptide, since it was absent from mock-infected and d120-infected cell profiles. d120 appears to specify a peptide of approximately 40 kilodaltons, which is reactive with ICP4 antiserum. This peptide represents the amino terminusofICP4, since only the 5' end of the ICP4 genewasretained in d120 (Fig. 5).

Otherphenotypicproperties of d120andd202. Thegrowth propertiesandpolypeptide phenotypesofd120, d202,tsB32, and KOS were compared in Nero and ICP4-transformed cells. Nero cells and cells of four ICP4-transformed lines

wereinfectedat amultiplicity of10 PFU percell and labeled with [35S]methionine from 5 to 16 h postinfection. All in-fected cultureswereincubatedat34WC,exceptthoseinfected with tsB32, which we incubated at 39°C to visualize the

mutant phenotype.

Duplicate

cultures were harvested for simultaneous infectivity determinations.

Thepolypeptideprofiles ofd120 and d202differed drasti-callyfrom that of KOS in Nero cells (Fig. 9). Qualitatively, the profiles of d120- and d202-infected cells were quite similartoeach other andtothatoftsB32-infected cells,with the exception of ICP4, which was evident only in tsB32-infected cells. Thepredominant viral polypeptides detected in Nerocells infectedwith thedeletionmutants were ICP6, ICPO, and ICP27. The truncated forms of ICP4 are not

evident inFig.9.In ourhands, ICP22wasdiscernibleonly in cycloheximide reversalexperiments. Both ICPO and ICP27

are immediate-early polypeptides (28), whereas ICP6 is

generally notregarded as amember ofthis class of HSV-1 genes. It isclear, however,that theexpressionof ICP6 does

notrequire functional ICP4 (Fig. 9).

The polypeptide profiles of d120 and d202 in ICP4-transformed cells resembled that of the wild-type virus, indicating efficient complementation by resident, wild-type ICP4. Enhanced expression ofearly polypeptides in tsB32-infected ICP4-transformed cells relative to tsB32-infected Nero cells (i.e., ICP8, ICP11, and

ICP36)

wasalso evident. Aspects of the mutant phenotype characteristic of tsB32, however, were evident in all ICP4-transformed cell linesat

this multiplicity. This observation is consistent with the results shown in Fig. 3, in which the ts mutant was

ineffi-ciently complemented

in ICP4-transformed cells at high multiplicity of infection. The fact that early and late polypeptideswereexpressedrelatively inefficientlyin ICP4-transformed cells infected with thetsmutant, whereasthose of the deletionmutants wereefficiently expressed, indicates that the ts mutantform of the ICP4 polypeptideexhibitsan

interferingeffectongeneexpression,thusinhibiting comple-mentation.

Thegrowthcharacteristics ofKOS,d120, d202,and tsB32

are summarized in Table 2. The following statements

con-cerningthese viruses can be made. (i) The growthofd120,

VOL.56, 1985

on November 10, 2019 by guest

http://jvi.asm.org/

[image:8.612.93.261.74.415.2]

Nero 2-6

I

1-5

I

1-2 12

*-U4 U-_

-*~ _ U_O8-

--M KOS 120 202 832 K05 120 202 19 20

_m 25 IMM

_ 27 -N_ *-..O ---ku ow -'1

, t ~. ;w,4 i*.

36 wom w_> 0 ,

U c:

44 _

48

-832 KOS 120 202 832 KOS 120 202 B32 KOS 120 202 832

FIG. 9. Viralpolypeptides synthesized inKOS-,d120-,d202-,and tsB32-infected Nero andICP4-transformedcells. Confluentmonolayers of each celltype wereinfectedasdescribedinthe text andanalyzed by SDS-PAGEasdescribed inthelegendto Fig. 3. All infectionswere performedat34°C,except thosewithtsB32,whichwereperformedat39°C.AnuninfectedNero cell culture was maintained for visualization of cellularbands (M). Theuninfected cell polypeptide profiles ofNero,4-2, 2-6, 1-5, and 1-2 cells were identical (data not shown). Numbers in the center aredescribed in the legend to Fig. 3.

d202, and tsB32 (at 39°C) in Nero cells was severely re-stricted. (ii) All ICP4-transformed cell lines complemented

d120 and d202 efficiently. (iii) ICP4-transformed cell lines complementedthegrowthof tsB32poorlyatamultiplicityof infection of 10 PFU percell. Only in2-6 and 4-2 cellswere

significant levels of complementation demonstrated. (iv) Wild-type recombinants were observed only when ICP4 deletionortsmutantswerepropagatedin2-6 cells. Thus, the yields ofmutant and wild-type viruses incell lines

express-ing ICP4 supportthe results ofpolypeptide analysis.

The experimental results shown in Fig. 9 illustrate a

qualitative comparison of polypeptide synthesis by d120, d202, and tsB32 (at 39°C) in Nero and ICP4-transformed cells. When all of the incubations werecarriedoutat 39°C,

the same pattern of polypeptide synthesis was observed in ICP4-transformed cell lines. The polypeptide profiles of d120- and d202-infected 4-2 cells were similar to that of

KOS, whereas the profile of tsB32-infected 4-2 cells was morerestricted,asshownbythe relativeunderproductionof ICP5 andICP25 (Fig. 10).Also clear is the factthat,inNero

cells, d120 and d202 synthesized reduced levels of

immedi-ate-early polypeptides ICPO and ICP27 relative to tsB32. ICP6 was also reduced in d120- and d202-infected cells relative to tsB32-infected cells. To determine more

specifi-cally whether the rates of synthesis of ICPO, ICP27, and ICP6 in d120- and d202-infected cells differ from that in tsB32-infected cells and to examine the possible effects of

proteinturnoveronpolypeptide phenotype,welabeled Nero cells infected with these mutants for 0.5-h intervals at

selected times from 0to8 hpostinfectionat39°C. Therates

ofsynthesisofICP0, ICP27, and

ICP6

incells infected with the deletion mhutants were slower than that seen in tsB32-infectedcells(Fig. 11). This differencemay be attributedto adirect orindirect effect ofthetsB32form of ICP4. Inthis case, it would be postulated that the presence of the ts mutantform of the ICP4

polypeptide

resultedinexaggerated expression of immediate-early polypeptides.

DISCUSSION

The genome of HVS-1 is capable ofcoding for approxi-mately 100 polypeptides. Despite the large size of its genome,HSV-1ishighly conducivetogenetic analysis,thus pertnittingtheidentification ofspecificHSV-1-encoded func-tions. ts mutants, in particular, have

proven

invaluable in thisregard (1, 42) in that, todate, complementation groups representing approximately 30 essential replicative genes have beenlocalized ontheHSV-1 genome. Inadditionto ts

mutants, mutants resistantto avariety ofanti-herpesdrugs (4),mutantswhich render cells resistanttoimmune

cytolysis

(31), and mutantsresistanttoneutralization with monoclonal antibodies (18) have proven useful in

identifying

and char-acterizingthegenomiclocations andfunctions ofindividual HSV-1 geneproducts.

Using cloned HSV-1 genes and selections for both the thymidine

kinase-negative

(tk-) and

-positive (tk+)

pheno-types, Post andRoizman (34) have developeda method to deletenonessential genesof HSV-1. Thisapproachledtothe isolation of viable deletion mutants in the

immediate-early

on November 10, 2019 by guest

http://jvi.asm.org/

[image:9.612.141.479.75.373.2]

HSV-1 DELETION MUTANTS IN THE ICP4 GENE 567

Nero 4-2

o X X o o

-3-c co

OA__

I₀

£

4 5 6

11

25 27

_ _ _ _

FIG. 10. Viral polypeptides synthesized in KOS-, d120-, d202-, and tsB32-infected Nero and 4-2 cells at 39°C. Nero and 4-2 cells were infected, labeled, and analyzed asdescribed in the legend to Fig. 8,exceptthat allincubationswerecarriedout at39°C.Numbers ontherightaredescribedin thelegendtoFig.3.

ICP4 deletion mutants d120 and d202 were isolated from the progenyof a cotransfection of pKX2-ASacI (Fig. 1) and intact KOS (wild-type) DNA. pKX2-ASacI contains a 0.5-kb deletion in the coding sequence for ICP4. In the absence of sequenceinformation in this region of the genome, it was of interest to determine if the ICP4 gene in pKX2-ASacI retained any ICP4 activity. The transient expression assay shown in Fig. 4 indicates that pKX2-ASacI does not specify ICP4-like activity. This experiment enabled us to ascertain what the consequences of such a deletion would be before introducing the deleted gene intowild-type virus. As noted above, adeletionmutant in ICP4 hasbeen isolatedrecently which does not inactivate the gene product (44).

d202contains theexpected0.5-kb deletion in both copies of ICP4. Consistent with the presence ofa 0.5-kb deletion internal to the ICP4-codingsequences wasthe synthesis of anmRNAwhich was smaller than itswild-typecounterpart by 0.5 kb and the synthesisofa correspondingly truncated form of ICP4. The structureofd120, ontheotherhand, was

quiteunexpected. Thismutant containsa4.1-kbdeletion in both copies of ICP4 and failed to express detectable cyto-plasmic ICP4 mRNA. A severely foreshortened form of ICP4 was, however, detectedby Western blot analysisand

KOS d 120 d 202 B32

A AAA A A Ai AA A AAAA A

In An if in; InAM kmem I e,in il VW VW

(a)geneforICP22, indicatingthenonessentiality of thisgene for virus replication in the cell lines tested. Moreover, a

viable deletion mutant has recently been isolated in a non-essentialregion of ICP4(44). In thisstudy,wegeneratedcell lines transformed with the gene for the HSV-1 immediate-earlygene encodingICP4.

ICP4-containing

cells were used

toisolateand propagate twoHSV-1 deletionmutantsin the geneforICP4.Thephenotypesofthe twodeletionmutants, d120 and d202, were then analyzed on cells which do not

contain the ICP4 gene.

Isolation ofICP4 deletionmutants. ICP4-transformedcells

were generated by cotransformation of Vero cells with pKXCandpSV2neo. ICP4-containing2-6cellswereused to

screen for ICP4 deletion mutants. Two ICP4 deletion mu-tants were isolated in this manner. Consistent with the findings of Davidson and Stow (7), passage of ICP4 ts or deletion mutants on 2-6 cells resulted in the generation of wild-type (ts+ for ts mutants; host cell independent for deletion mutants) virus in the progeny. This presumably

occurs in the presence ofhelper functions provided by the superinfecting virus by replication, mobilization, and prop-agation ofwild-type sequences containing ICP4 in 2-6 cells. Amplifiedsequences arepackagedasdefective genomes (7) which are then able to recombine with the superinfecting

mutanthelpervirustoproducewild-type virus. Cells gener-atedby transformation with pKX2-P4 lack

oris

and did not result in thegenerationofwild-typevirus afterinfectionwith

ts ordeletion mutants (Table 2).

g m".I

A"

me b

"4000 4

11

25

5 w Sn lo _27

*-***

~~~~~~36

FIG. 11. Rates of synthesis of viral polypeptides at 39°C in KOS-,d120-, d202-,andtsB32-infectedNero cells. Nerocellswere infected with the indicated viruses and incubated at 39°C as de-scribedinthelegendtoFig.8. ["5S]methionine(100p.Ci)wasadded duringtheindicated timeintervals. At the endofthelabeling period, monolayers werewashed and celllysates were preparedand ana-lyzedbySDS-PAGE asdescribed in thelegendtoFig.3. Numbers ontherightaredescribed in thelegendtoFig. 3.

VOL.56, 1985

.f,.

"I Z.

44, 2Lk-.

4"

.:

4

p

on November 10, 2019 by guest

http://jvi.asm.org/

[image:10.612.98.263.75.367.2] [image:10.612.338.520.331.653.2]

may represent expression of the amino terminus of ICP4. One wouldpredictthatanotmRNAwouldinitiate fromthe

ICP4 mRNA start site ind120, sincethisregion ofthe ICP4 gene is intact. Repeated efforts to detect an ICP4-specific mRNA in d120-infected cells have failed, however. As anticipated, the quantity and mobility of the ICP0 mRNA wereidenticaltothe analogous mRNAsynthesized in KOS-andd202-infected, cycloheximide-treated cells.

Despite differences in the size and structure of the ICP4 gene and the size of the ICP4 mRNA synthesized in d120-and d202-infected cells,

the

immediate-early polypeptides expressed in cells infected with the two mutants were identicalin a cycloheximide reversal experiment (Fig. 7). In fact, withthe exceptionof ICP4, which was notdetected by standardpolyacrylamide gelelectrophoresis (PAGE) analy-sis in extracts of deletion mutant-infected cells, the quanti-ties and mobilities of ICP6, ICP0, ICP22, and ICP27 were identicaltothose observed in cells infectedwith KOS. The experiments of Fig. 7 and 8 established two points. (i) Full-lengthICP4polypeptidesare notexpressedind120-and d202-infected Nero cells. Independentofwhether the dele-tion ind202is inframe,apeptideantigenicallyreactive with ICP4 antiserumwith an apparent molecularweightreduction of30,000 with respect to the wild-type protein was synthe-sized. Anantigenicallyreactivepeptideofapproximately40 kilodaltonswassynthesized ind120-infectedcells. Thesizes of theshortened peptides were consistent with the sizes of thedeletions ind120andd202(Fig.5).(ii) Expressionof the other immediate-early polypeptides in cells infected with d120 and d202 was identical to that seen in wild-type virus-infectedcells inacycloheximide reversalexperiment. Viral geneexpression inthe absence ofICP4. Theabsence of functional ICP4 in d120- and d202-infected cells had profound effects on the expression of other viral genes relative to that seen in wild-type

infection.

Qualitatively, with theexceptionofICP4,thepeptides

synthesized

in d120-andd202-infectedNerocellswerethesamespeciesasthose expressed in tsB32-infected cells at

39°C.

ICPO and ICP27

were produced in moderate quantities in d120- and d202-infectedNero

cells

(Fig. 9, 10, and 11). Interestingly, ICP6

was the most abundantly expressed polypeptide in cells infectedwith thedeletionmutants. Honess

and

Roizman(19) havepreviouslyshown thatICP6 is

inefficiently

expressedin theabsence ofpriorviralprotein

synthesis, placing

ICP6in the early

(1)

class of HSV-1 genes. They have also shown thatICP6 is synthesized to some

degree

in the presence of the amino acid analog canavanine (20). In our

hands,

re-ducedquantities ofICP6 relative toICPO were

synthesized

inacycloheximide reversal experiment(Fig. 7).This ICP6-ICP0 ratio was reversed in d120- and d202-infected Nero cells

(Fig.

11). The synthesis of ICP6in this

genetic

back-ground indicates that ICP4 is not

required

for efficient expression of the ICP6 gene. This observation and the

minimal expression of ICP6 in a cyclohexiinide reversal experiment may well reflect the involvement ofan a gene other thanICP4

(i.e., ICP0,

ICP22, ICP27,

or

ICP47)

in ICP6 gene expression.

Recently it

has

been shown bya number of

investigators

that the

immediate-early

geneproduct ICP0can actintrans in transient expression systems to enhance

expression

of early

8ISV-1

genes (30, 39). With the

exception

of

ICP6,

early genes were underexpressed to the same or a greater

extentindeletion mutant-infectedcells than in cellsinfected with an ICP4 ts mutant at 39°C.

Therefore,

the lack of expression ofearlygenesinICP4tsmutant-infected cells at

390Ccannotbe due

solely

to

inhibitory

effects manifested

by

the

ts form of the ICP4 polypeptide. Studies of the ICP4 deletionmutantsreflect theinsufficiencyofICPOto actalone tostimulate

early

gene

expression.

The tsform of ICP4affectsgeneexpression. Thetsformof ICP4 intsB32-infected cellswasapparentlyinhibitorytothe activity of the wild-type form of ICP4 in all of the

trans-formed cellstested.Inhibitionwas moreprominentat

higher

multiplicities (Fig. 3).Atamultiplicityof10 PFU percell, all of the cell linescomplementedd120 and d202efficiently,yet tsB32wascomplemented relatively inefficientlyat39°C

(Fig.

9 and10;Table2).Infact, only2-6 and 4-2cells

significantly

enhancedtheyield oftsB32at 39°Cabove theyieldin Nero cells. Possible explanations for this inhibition include the following. (i)Thetsform ofICP4bindstightlytositesbound by wild-type ICP4, inhibiting the activity of the latter by competition. (ii)Thetsform ofICP4 formsastable

complex

with the wild-type form of the protein in the cytoplasm, thereby inhibitingeither transportof thelatter to the nucleus

or itsactivity oncein thenucleus. (iii) Thetsform of ICP4 alters cellular metabolism in such a way as to render the wild-type formof ICP4 less effective.

Another property of the ts form of ICP4, as indicated by comparisonwiththedeletion mutants, is itsabilitytoinduce overproduction of immediate-early polypeptides. This was

evident from continuous labeling experiments and by pulse-labelingat 39°C (Fig. 9, 10, and 11). The rates ofsynthesis and accumulation of ICPO and ICP27 were decreased in d120- and d202-infected Nero cells relativetotsB32-infected cells. Consistent with these findings isthe observation that

an ICP4-containing clone derived from tsB32 can induce CATactivityat39°C when CAT geneexpression is under the control of an immediate-early promoter (10). A possible explanation for this finding is thatthe ts form of ICP4 retains

some of the activity associated with that of the wild-type polypeptide.

Analysis of ICP4 deletion mutants has bothconfirmed and extended the findings previously obtained with ICP4 ts

mutants(11, 36). ICP4 is required for the expression ofearly and late genes. In the absence of functional ICP4, the remaining immediate-early genes are either insufficient or

insufficiently expressed to induce significant levels ofearly genes, with thepossible exception of ICP6. The phenotypes ofd120 andd202 also indicate that the exaggerated levels of immediate-early polypeptides induced by ICP4 ts mutants

are due, in part, to activities associated with the ts ICP4 polypeptides themselves.

Many of the contentions put forth in this studyrestin part ontheassumption that the deleted forms of ICP4 have little or no effect on gene expression or on the activity of the wild-type ICP4 in any of the transformed cell lines tested. The 40-kilodalton peptide synthesized ind120-infected cells has noeffect ontherestoration of thewild-typepolypeptide phenotype in d120-infected ICP4-containing cells. On the other hand, the slightly reduced intensities of some late polypeptides on gel profiles of d202-infected ICP4-containing cells (Fig. 9 and 10) suggest a small inhibitory effect. Althoughwe cannotformally ruleout anactivityassociated with the d202 ICP4-related polypeptide, it should be noted that the plasmid clone used to generated202 had noactivity in a transient assay (Fig. 4). It remains of interest to

determine the molecular basis for the apparent inhibitory effect of the ts ICP4 polypeptide and possibly that of the d202 peptide. An attractive hypothesis is that the mutant ICP4 polypeptides (tsB32 and possibly d202) are, for the mostpart, nonfunctional and yet may still bindtothe normal site ofaction, thereby competing with any wild-type ICP4

on November 10, 2019 by guest

http://jvi.asm.org/

HSV-1 DELETION MUTANTS IN THE ICP4 GENE 569 present. Furtherstudiesareunderway to test this

hypothe-sis.

Severalotherobservationsconcerningthegrowthof d120 andd202 in ICP4-negative cells suggest

interesting

avenues

forfurther study. Unlike the ts mutants in ICP4 (29), d120 and d202 do not induce certain cellular stress proteins on infection of HELcells (datanotshown), norarecytopathic effectsevident after 24 hat a

multiplicity

of 10PFUper cell, suggesting additional differences between the deletion mu-tants and the ts mutants. Also of interest is the genesis of d120. Efforts are in progress to determine whether the deletion ind120 was inducedas aresultof markertransferor

whether deletions such as that found in d120 arise

sponta-neously.

ACKNOWLEDGMENTS

We thank Maryellen Polvino-Bodnar forexpert assistance with Western blot experiments. We also thank Don Coen for useful comments on the manuscript and Madelon Datz for manuscript

preparation.

Thisinvestigationwas supported byPublic Health Servicegrant CA20260 from the National Cancer Institute.N.A.D.is therecipient

ofPublic HealthServicepostdoctoralfellowshipAI06790 from the National Institute ofAllergyandInfectious Diseases.

LITERATURECITED

1. Brown, S. M., D. A. Ritchie, and J. H. Subak-Sharpe. 1973. Genetic studieswithherpessimplex virustype 1. Theisolation oftemperature-sensitive mutants, theirarrangement into com-plementation groups and recombination analysis leading to a linkagemap. J.Gen. Virol. 13:329-346.

2. Burnette,W. N.1981. Westernblotting: electrophoretictransfer ofproteinsfrom sodiumdodecylsulfate-polyacrylamidegelsto unmodified nitrocellulose and radiographic detection with an antibody and radioiodinated protein A. Anal. Biochem. 112:195-203.

3. Clements,J. B.,R.J.Watson,and N. M. Wilkie.1977.Temporal regulationofherpessimplexvirustype 1transcription: localiza-tion oftranscriptsontheviralgenome. Cell 12:275-285. 4. Coen,D.M.,and P. A.Schaffer. 1980.Twodistinctlociconfer

resistance to acycloguanosine in herpes simplex virus type 1. Proc. Natl. Acad. Sci. USA 77:2265-2269.

5. Cordingly,M.G.,M. E. M.Campbell,andC. M. Preston. 1983. Functionalanalysis ofaherpes simplexvirustype 1 promoter: identification of far-upstream regulatory sequences. Nucleic Acids Res. 11:2347-2365.

6. Courtney, R. J., and M. Benyesh-Melnick. 1974. Isolation and characterization of a large molecular-weight polypeptide of

herpes simplexvirustype 1. Virology62:539-551.

7. Davidson, I.,and N. Stow. 1985. Expression ofan immediate

early polypeptide and activation of a viral origin of DNA replicationin cellscontainingafragmentofherpessimplexvirus DNA. Virology141:77-88.

8. DeLuca,N. A.,M. A.Courtney,and P.A. Schaffer. 1984.Tem-perature-sensitivemutantsinherpes simplexvirustype 1ICP4

permissiveforearlygeneexpression. J.Virol. 52:767-776. 9. DeLuca, N., S. Person, D. J. Bzik, and W. Snipes. 1984.

Location oftemperaturesensitivemutantsinglycoprotein gB of

herpes simplexvirustype 1. Virology137:382-389.

10. DeLuca,N.A.,and P. A.Schaffer.1985. Activationof

immedi-ate-early, early, and late promoters by temperature-sensitive and wild-type forms of herpes simplex virus type 1 protein ICP4. Mol. Cell. Biol. 5:1997-2008.

11. Dixon, R. A. F., and P. A. Schaffer. 1980. Fine-structure

mapping and functionalanalysis oftemperature-sensitive mu-tants in the gene encoding the herpes simplex virus type 1 immediateearlyprotein VP175.J. Virol.36:189-203.

12. Everett, R. D.1984. Transactivation oftranscription by herpes virus product: requirement for two HSV-1 immediate-early

polypeptidesfor maximumactivity. EMBO J. 3:3135-3141. 13. Goldberg, D. A. 1980. Isolation and partial characterization of

theDrosophilaalcoholdehydrogenasegene. Proc. Nati.Acad. Sci. USA77:5794-5798.

14. Goldin, A. L., R. M. Sandri-Goldin, M. Levine, and J.C. Glorioso. 1981. Cloning of herpes simplex virus type 1 se-quencesrepresenting the wholegenome.J. Virol. 38:50-58. 15. Gorman, C. M., L. F. Moffat, and B. H. Howard. 1982.

Recombinant genomeswhich express chloramphenicol acetyl-transferase in mammalian cells. Mol. Cell. Biol.2:1044-1051. 16. Graham,F.L.,and A.J.vanderEb.1973. A newtechniquefor

theassayofinfectivityofhuman adenovirus 5 DNA. Virology 52:456-467.

17. Hayward, G.,R.J.Jacob,S. C. Wadsworth,andB.Roizman. 1975.Anatomyofherpes simplexvirus DNA: evidence for four populationsofmolecules that differ in the relativeorientations of their longand short segments. Proc. Natl. Acad. Sci. USA 72:4243-4247.

18. Holland,T.C.,S. D.Marlin,M. Levine,andJ.Glorioso. 1983. Antigenicvariants ofherpes simplex virusselected with

glyco-protein-specific monoclonalantibodies. J. Virol.45:672-682. 19. Honess,R.W.,and B.Roizman.1974.Regulationofherpesvirus

macromolecular synthesis.I.Cascaderegulationofthe synthe-sis ofthreegroups of viralproteins.J. Virol. 14:8-19. 20. Honess,R.W.,and B.Roizman.1975.Regulationofherpesvirus

macromolecularsynthesis: sequential transition ofpolypeptide

synthesis requires functional viral polypeptides. Proc. Natl. Acad. Sci. USA72:1276-1280.

21. Johnson, D. A., J. W. Gautsch, J. R. Sportsman, and J. H. Elder. 1984. Improved technique utilizing nonfat dry milk for analysisofproteins and nucleic acids transferred to nitrocellu-lose. Gene Anal.Tech. 1:3-8.

22. Jones, N.,and T.Shenk.1979. Anadenovirustype 5early gene function regulates expressionofother viral genes. Proc. Natl. Acad. Sci. USA76:3665-3669.

23. Mackem, S., and B. Roizman. 1982. Structural features ofthe herpes simplex virus a. gene 4, 0, and 27 promoter-regulatory sequences which confer a regulation of chimeric thymidine kinase genes. J. Virol.44:939-949.

24. Maniatis, T.,E. F. Fritsch,andJ. Sambrook. 1982. Molecular cloning: alaboratory manual, p. 191-193. Cold SpringHarbor Laboratory,ColdSpring Harbor,N.Y.

25. Maniatis, T., A. Jeffrey, and D. G. Kleid. 1975. Nucleotide sequence of the rightward operator of phage X. Proc. Natl. Acad. Sci. USA72:1184-1188.

26. Marsden,H.S.,N. D.Stow,V.G. Preston, M. C.Timbury,and N. M.Wilkie. 1978. Physical mappingofherpes simplex virus-inducedpolypeptides. J. Virol.28:624-642.

27. Mocarski,E.S.,and B. Roizman.1982.Structureand roleofthe herpes simplexvirus DNA termini ininversion,circularization, andgeneration ofvirion DNA. Cell 31:89-97.

28. Morse, L.S.,L.Pereira,B.Roizman,andP. A.Schaffer. 1978. Anatomy ofherpes simplex virus(HSV) DNA X. Mappingof viralgenes byanalysis of polypeptides andfunctions specified by HSV-1 x HSV-2recombinants. J. Virol. 26:389-410. 29. Notarianni, E. L., and C. M. Preston. 1982. Activation of

cellular stressprotein genes by herpes simplex virus tempera-ture sensitive mutants which overproduce immediate-early polypeptides. Virology 123:113-122.

30. O'Hare, P.,andG. S.Hayward.1985.Evidencefor adirectrole forboth the175,000- and 110,000-molecular-weight immediate-early proteins ofherpessimplex virusin the transactivationof delayed-earlypromoters. J.Virol. 53:751-760.

31. Pancake, B. A., D. P. Aschman, and P. A. Schaffer. 1983. Genetic andphenotypic analysis of herpessimplexvirustype 1 mutants conditionally resistant to immune cytolysis. J. Virol. 47:568-585.

32. Persson, R. H., S. Bacchetti, and J. R. Smiley. 1985. Cellsthat constitutivelyexpresstheherpes simplex virus immediate-early protein ICP4 allow efficient activation of viral delayed-early genesintrans.J. Virol. 54:414-421.

33. Poffenberger, K. L., E. Tabares, and B. Roizman. 1983. Char-acterization ofa viable, noninverting herpes simplex virus 1 genomederived by insertion and deletion of sequencesat the junction ofcomponents L and S. Proc. Natl. Acad. Sci. USA VOL.56, 1985

on November 10, 2019 by guest

http://jvi.asm.org/

80:2690-2694.

34. Post,L. E., and B.Roizman. 1981. A generalized technique for deletion of specific genes in large genomes: a gene 22 of herpes simplex virus type 1 is not essential for growth. Cell 25:227-232. 35. Preston, C. M. 1979.Abnormal properties ofanimmediateearly polypeptide in cells infected with herpes simplex virus type 1 mutant tsK. J. Virol.32:357-369.

36. Preston, C. M. 1979. Control of herpes simplex virus type 1 mRNA synthesis in cells infected with wild-type virus or the temperature-sensitive mutant tsK. J. Virol. 29:275-284. 37. Preston, V. G. 1981. Fine-structure mapping of herpes simplex

virus type 1 temperature-sensitive mutations within the short repeat region of thegenome. J. Virol. 39:150-161.

38. Preston, V. G., A. J. Davison, H. S. Marsden, M. C. Timbury, J. H. Subak-Sharpe, and N. M. Wilkie. 1978. Recombinants between herpes simplex virus types1 and 2:analyses of genome structures and expression of immediate early polypeptides. J. Virol.28:499-517.

39. Quinlan,M.P., and D.M.Knipe. 1985.Stimulation ofaherpes simplex virus DNA-binding protein by two viral functions. Mol. Cell. Biol. 5:957-963.

40. Sacks, W. R., C. C. Greene,D. P.Aschman, and P. A. Schaffer. 1985.Herpessimplex virus type 1ICP27 isanessential regula-toryprotein. J.Virol. 55:796-805.

41. Sandri-Goldin, R. M., A. L. Goldin, L. E. Holland, J. C. Glorioso, and M. Levine. 1983. Expression ofherpes simplex virus p and y genes integrated in mammalian cells and their induction by anageneproduct. Mol. Cell. Biol. 3:2028-2044. 42. Schaffer, P. A., G. M. Aron, N. Biswal, and M.

Benyesh-Melnick. 1973. Temperature-sensitive mutants of herpes simplex virus type 1: isolation, complementation and partial characterization. Virology 52:57-71.

43. Schaffer, P. A., V. C. Carter, and M. C. Timbury. 1978. Collaborative complementation study oftemperature-sensitive mutants of herpes simplex virus types 1 and 2. J. Virol. 27:490-504.

44. Schroder,C. H.,J.DeZazzo, K.W.Knopf,H. C.Kaerner,M. Levine, andJ. Glorioso. 1985. A herpes simplex virus type 1 mutant withadeletion in thepolypeptide-codingsequences of theICP4 gene. J.Gen. Virol. 66:1589-1593.

45. Southern, E. M. 1975. Detection ofspecificsequences among DNAfragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517.

46. Southern,P.J.,and P.Berg. 1982. Transformation of mamma-lian cells to antibiotic resistance with abacterial gene under control ofthe SV40early region promoter.J.Mol.Appl. Genet. 1:327-341.

47. Stow,N. D.,andC. McMonagle. 1983. Characterization ofthe TRS/IRsorigin ofDNAreplication of herpes simplex virus type 1.Virology130:427-438.

48. Towbin,H.,T.Staehelen,andJ.Gordon.1979.Electrophoretic

transfer of proteins from polyacrylamide gels to nitrocellulose sheets:procedure andsomeapplications. Proc. Natl. Acad. Sci. USA 76:4350-4354.

49. Watson, R. J., and J. B. Clements. 1978. Characterization of transcription-deficient temperature-sensitive mutantsofherpes simplexvirustype 1.Virology91:364-379.

50. Watson, R. J., C. M. Preston, and J. B. Clements. 1979. Separation and characterization of herpes simplex virustype1 immediate-early mRNA's.J.Virol. 31:42-52.

51. Weller, S. K., K. J. Lee, D.J. Sabourin, and P. A. Schaffer. 1983.Genetic analysis oftemperature-sensitive mutants which define the gene for the major herpes simplex virus type 1 DNA-binding protein.J. Virol. 45:354-366.

on November 10, 2019 by guest

http://jvi.asm.org/