Separation and characterization of herpes simplex virus type 1 immediate-early mRNA's.

(1)

JOURNAL OF VIROLOGY, July1979, p. 42-52 Vol.31, No. 1 0022-538X/79/07-0042/1 1$02.00/0

Separation

and Characterization of

Herpes

Simplex

Virus

Type 1

Immediate-Early

mRNA's

ROGER J.WATSON,t CHRISTOPHER M. PRESTON,ANDJ. BARKLIE CLEMENTS* Departmentof Virologyand Medical Research CouncilVirology Unit, University of Glasgow,Church

Street, Glasgow Gll 5JR, Scotland

Polyadenylated immediate-early transcripts of herpes simplex virus type 1,

made in BHK cells infected and maintained in the presence ofcycloheximide,

have been separated on denaturing agarose gels containing methyl mercuric

hydroxide. Threevirus-specific mRNA bands of estimated sizes 4.7, 3.0, and 2.0

kilobases (kb) were detected, and these mRNA's were mapped on the virus genome and also used to direct protein synthesis in vitro. The 4.7- and 3.0-kb mRNA'shybridizedpredominantly tocertain DNAfragmentswhicharelocated in the shortandlong repetitive regionsof the genome,respectively,whereas the 2.0-kb mRNA's mapped to three discrete regions of the virus DNA. In vitro

translationof theseseparatedmRNA sizeclasses indicated that the 3.0-kb mRNA

specifiedthesynthesisofviruspolypeptideVmw 110,whereas the 2.0-kb mRNA's

specified Vmw 68, 63, and 12. The synthesis of small amounts ofVmw 175 was

specified bythe 4.7-kb mRNA. Incontrastwiththe mRNA's whichspecify these

otherimmediate-early polypeptides, that specifyingVmw 12 is much larger than required for itscodingsequences.

The locations on the genome of transcripts

which accumulate in cells infected with herpes

simplex virus type 1 (HSV-1) at various times

postinfection,in the presenceorabsence of DNA

or protein synthesis inhibitors, have been

ana-lyzedbyhybridizationtothefragmentsof

HSV-1DNAgenerated by digestion withanumber of restrictionendonucleases (3, 10). These, and pre-vious studies (6, 22), have indicated that

tran-scription of the HSV-1 genome is subject to

temporalregulation.

Inhibition of virus protein synthesis with cy-cloheximidewasfoundtoresult inaccumulation ofRNA sequences which hybridized to certain

DNAfragmentsonly (3, 10), and these restricted

RNA sequences, transcribedbyapreexisting

a-amanitin-sensitive host RNA polymerase (4),

have been defined as immediate-early (IE) RNA. IE RNAsaccumulate also in cells infected at the nonpermissive temperature with certain HSV-1 temperature-sensitive (ts) mutants, and aretheonlyvirus transcripts detectable in ts K-infected cells (24).Release of the cycloheximide block (3), orshift-down ofcells infected with ts Ktothepermissive temperature (24), was found toallowtranscription from additional regions of HSV-1 DNA which map throughout the virus genome. These latter regions were also tran-scribed during the normal replicative cycle at early and late times postinfection (3), that is,

t Present address: Laboratory of Molecular Virology, Na-tionalCancerInstitute, Bethesda,Md. 20205.

before and after the onset ofvirus DNA

repli-cation.

The polypeptides specified by IE RNA, the

IEpolypeptides, formasubset of the virus

pro-teins made during productive infection (8, 19).

In ourlaboratory,theIEpolypeptidesidentified have been designated by their apparent

molec-ularweightand include Vmw 175, 136, 110, 87, 68,

63, and 12. The IE polypeptidesinclude the

a-polypeptides class as defined by Honess and

Roizman (8). These authors have shown that

functional a-polypeptides synthesis is required

for the subsequent appearance oflater

classes,

,BandX-polypeptidesin theirterminology,which

are synthesized in the absence of virus DNA

replication (8).

Theobservations detailed above suggest that IE RNA codes for one or more virus

polypep-tides whichregulateamajoreventin virus tran-scription; thatis,the transition from IEtoearly

RNA synthesis. It is as yet unknown how IE

polypeptides functioninthisregulation,butit is knownthatthesepolypeptidesaremodified and

preferentially accumulated in the nucleus (17)

and have DNA binding properties in vitro (R. T. Hay and J. Hay, manuscript submitted for publication).

We have previously mappedthe locationson the HSV-1 genome ofcytoplasmic and nuclear HSV-1 IE RNAs (3) and found that these tran-scripts hybridized predominantlyto DNA frag-ments containing, or adjacent to, the inverted

42

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

repeat regions of the virus genome. We report here the results ofexperiments designedto char-acterize the IEmRNA's. IE mRNA's havebeen resolved byelectrophoresisthrough denaturing

gels, and the separated size classes have both

been mapped against HSV-1 DNA fragments

andusedtodirect invitroprotein synthesis in a

cell-free translationsystem.

MATERIALS AND METHODS

Cells andvirus.BHK-21 (C13)cells were grown asmonolayers in 80-oz. (ca 2.37-liter)rollerbottles and infected with HSV-1 (Glasgow strain 17) at a multi-plicity of infection of50PFU/cellas described previ-ously (3).

Labeling RNA.Cellswere pretreated with 200ug ofcycloheximide per ml for1hand either infectedor mock infected (MI) with HSV-1. Cells were labeled with[nP]orthophosphatefrom0 to 6hpostinfection in the presence ofcycloheximide as previously de-scribed (3).

Labeling DNA. HSV-1 DNA was 32P-labeled in vitro bynick translation as previously described (3). Cell fractionation and RNAextraction. Cyto-plasmicRNAwaspreparedbylysisof cells with Non-idetP-40,removal of nucleiby low-speed centrifuga-tion, and phenol-chloroform extraction of the cyto-plasmic fractionaspreviouslydescribed(18).

SelectionofpolyadenylatedRNA.Cytoplasmic RNAwasrecoveredbylow-speedcentrifugation and dissolved in2ml ofbindingbuffer(0.5 MKCI,25mM Tris-hydrochloride, pH 7.9). The RNAwasapplied to acolumn ofoligo(dT)-cellulose (obtainedfrom Collab-orativeResearchInc.)and washed with3volumes of binding buffer. Polyadenylated [poly(A)] RNA was eluted with distilled water, made0.4M with respect tosodium acetate, andprecipitated with 3 volumes of ethanol by overnight storage at -20°C. RNA was recoveredbylow-speed centrifugation.

CH3HgOHagarosegelelectrophoresis. Prepar-ative and analytical RNA electrophoresis was per-formed through 1.5% agarose gels containing 5 mM

methyl mercuric hydroxide (CH3HgOH) basically by the method of Bailey and Davidson (1). Pelleted poly(A) RNAwasredissolved in50plof Ebuffer (50 mM boric acid, 5 mM Na2B407-1OH20, 10 mM Na2SO4,and1mMEDTA,pH 8.2), containing25mM CH3HgOH, 10%glycerol, and sufficientbromophenol blueto act as a tracking dye. RNA waskeptfor 10 minat roomtemperatureimmediately preceding elec-trophoresis.

Preparative RNAelectrophoresis (approximately5

iLgof RNA pergel) wasperformedontubegels(12by 0.6 cm)at 6mA/gel for5h at roomtemperature. Gels werethen soaked for1 hin 10mM ,B-mercaptoetha-nol-20 mM Tris-hydrochloride (pH 7.9) to remove

CH3HgOHboundtoRNA.Gelswerescanned witha

Joyce-Loebel UVscanner to locate the endogenous, contaminating 288 and 188 rRNA's. Gels were cut into 2-mm slices with a Mickle gel slicer, andeach slicewasmacerated in 1 ml ofelutionbuffer(1 mM EDTA,10mMTris, pH 7.9,containing10lOg of yeast tRNA perml), usingaDouncehomogenizer.RNAwas

eluted by shaking at 4°C overnight. The agarose was pelleted by low-speed centrifugation, and the super-natantfractions were retained. A10-plsample of each fraction wascounted in Aquasol to estimate the dis-tribution oflabeled RNA in the gel.

AnalyticalRNA electrophoresis was performed on slab gels (20 cm long). Electrophoresis was at 30 V overnight, and upon completion, gels were soaked in 0.5Mammonium acetate containing 1 ug of ethidium bromide per ml for 1 h. RNA bands were visualized by UVfluorescence,and thegels were dried and subjected toautoradiography by using Kodirex film.

Hybridization of RNA to total HSV-1 DNA. Hybridizationof[nP]RNAto HSV-1 DNA bound to nitrocellulosefilterswasperformed as described pre-viously (9).Individual gel fractions were hybridized to onefiltercontaining2,ugofHSV-1 DNA and one filter containing 2 utg of bacteriophage A DNA. Counts bound to A DNAfilterswere subtracted from those boundtoHSV-1 DNA filterstoestimate HSV-1-spe-cific RNA.

Quantitative estimates of the percentage of virus-specific labeled RNA in the total cytoplasmic poly(A) RNA and in each ofthe pooled fractions were made asabove, except that filters contained 5,g of DNA and reactionswereallowedtoproceedfor40h. The efficiency of these reactionswasestimated inaparallel reaction, using HSV-1 cRNA, made with Escherichia coli RNApolymeraseasdescribedpreviously(3). Un-dertheseconditions,itwasfound thatapproximately 45% of theHSV-1 cRNAinput hybridizedtothe HSV-1DNA filters.

Preselection of RNAagainstHSV-1 DNA

frag-ments.Poly(A)RNAin 70% formamide-2xSSC(lx SSC is 0.15 M NaClplus0.015M sodiumcitrate)was incubatedfor20hat37°C, with10-,Agequivalents of the individualHSV-1 DNAfragments boundto nitro-cellulose disks. Filterswereextensivelywashed in2x

SSC, either withorwithout the addition of25 tLg of RNase A per ml. The filters werethen incubated in iodoacetate solution (0.15 M sodium iodoacetate,0.1 Msodiumacetate in2x SSC, pH5.2)at54°Cfor40 mintodestroy any residual RNaseactivity (23),and further washed in 2x SSC. Hybridized RNA was elutedfrom the filtersby boilingin0.lx SSC for10 imm in the presence of10,ugof yeasttRNA carrier per ml.

Hybridizationtoblots.[ P]RNAwashybridized toHSV-1 DNAfragmentsand immobilized to nitro-cellulosemembraneby theSouthern blot technique

(21),asdescribedpreviously (3).Hybridizationto

frag-ments wasdetectedbyfluorography, using Kodak X-Omat H film and DuPont CronexLightningPlus in-tensifyingscreens.Filmswereallowedtoexposefor up to 3weeksat-70°C.

Invitro translation. Thepooled RNAfractions weremade0.5M withrespecttoKCI and reselected onoligo(dT)-cellulose columns,asdescribedabove,to remove residual agarose. The RNAeluted from the columnswasethanolprecipitatedin thepresenceof 5 ytg ofE. coli rRNA carrier. RNAwasrecovered by low-speedcentrifugation, dissolvedindistilled water, and used to direct cell-free translation inthe rabbit reticulocyte lysate system describedpreviously (18).

http://jvi.asm.org/

(3)

44 WATSON, PRESTON, AND CLEMENTS Endogenous mRNAwasdestroyed by preincubation with micrococcal nuclease (16). The products of in vitrotranslationwereseparated bypolyacrylamidegel electrophoresis (PAGE) as described by Marsdenet al. (11), and were detected by autoradiography by usingKodirex film.

Preparation ofin vivo-labeled IE polypeptides. Cellswereinfected with 50 PFU of HSV-1 per cell in medium containing 200 ,ug ofcycloheximide per ml and incubated for 5 h in the presence of the drug. Cultures were then washed three times with pre-warmed mediumcontaining5jigofactinomycinD per ml and incubated for 1 h with phosphate-buffered saline containing 100,uCiof[3S]methionineper ml. Cellswere harvested and disrupted asdescribed by Marsden etal. (11).

RESULTS

Analytical RNA electrophoresis.

Cyto-plasmic RNAswereisolated from infected (IE)

and uninfected (MI) cells labeled for 6 h with

[32P]orthophosphate in the presence of

cyclo-heximide, and thepoly(A) RNAswereselected

by chromatography on

oligo(dT)-cellulose

col-umns.Thisprocedure didnotresult in the

com-plete removal of ribosomal RNA, but thetrace

amounts of endogenous 28S and 18S rRNA's

provided useful size markersonsubsequent

anal-yses.

IE and MI poly(A) RNA's were denatured

with25 mM CH3HgOHandelectrophoresed in

parallelon anagaroseslabgel containing5mM

CH3HgOH. After electrophoresis, the gel was

stained with ethidium bromide, and the RNA

was visualized by UV fluorescence (Fig. 1). In

addition tothe 28S and 18S rRNA bandsseen

in the MI RNA track, there were three

induced bands in the IE RNA track. These virus-inducedbandswerealso detectedby

autoradiog-raphy of the gel (Fig. 2). By using the rRNA

species asmarkers assumedtobe5.3 kilobases

(kb) and 2.1 kb inlength (12), thesizes of the

virus-induced bands were estimated to be 4.7,

3.0,and2.0kb.Henceforth,these RNAs will be

referredtobytheir sizedesignation.

PreparativeRNAelectrophoresis.IE and

MI poly(A) RNA preparations (approximately

5 ,ug each) were electrophoresed in parallel

throughagarosetubegels containing

CH3HgOH.

Afterelectrophoresis,the gels weresoakedin 10

mM 8-mercaptoethanol to remove CH3HgOH

(1) andUV scannedto detect the rRNAmarkers

andthe threevirus-induced bands.

Thegelswerethencut into 2-mm slices,the

slices were homogenized, and the RNA was

elutedovernightinlow-salt buffer. Mostof the

agarose was removed by low-speed

centrifuga-tion, and a sample of each supernatant was

counted to determine the total distribution of

labeled RNA in the gel. This is shown for IE

[image:3.508.294.437.63.283.2]

RNA inFig.3.

FIG. 1. Poly(A) cytoplasmic RNAs isolated from HSV-l-infected (HSV)andMI BHK cells were sepa-ratedby electrophoresisthroughCH3HgOHagarose gels anddetected byUV illuminationafterethidium bromide staining. The positions of the endogenous 28Sand18SrRNA's,the4.7-, 3.0-,and2.0-kb virus-induced RNA species, and the bromophenol blue (BPB) dye frontareindicated.

Part of each elutedgel fractionwasthen

hy-bridized to total filter-bound HSV-1 DNA to

determine the distribution ofvirus-specific RNA

(Fig. 3). The distribution ofvirus-specific RNA

closely paralleled that of total RNA, and the

peaks exactly coincided. This experiment did

notallow accurateestimation of the percentage

of totalcountsthatwerevirusspecific, sincethe

eluate volumes varied somewhatand the

hybrid-izationreactions didnot go tocompletion. More

quantitative estimates were made by using, in

parallel, the hybridization of HSV-1 cRNA

(madewithE. coliRNApolymerase)tomeasure

theefficiency of the hybridization reactions; this

showed that total labeledIERNAwas

approx-imately 70% virus specific. These proportions for

the virus-specific peak fractions were greater

than 80%.

Theportions of the IE RNAfractions

remain-ingwerepooled as indicatedin Fig. 3 (fraction

1 contained the 4.7-kb species). The MI RNA

fractionsofequivalentsizewerelikewisepooled,

using theendogenousrRNAspeciestoalignthe

two gels. The pooled IE RNA fractions were

divided;onehalfwashybridizedtoHSV-1DNA

fragments and theother wasusedtodirect

pro-tein synthesis in vitro. The pooled MI RNA

fractionsweresimilarly translatedinvitro.

Hybridization ofgel fractions to HSV-1

DNAfragments. ThepooledIERNA fractions

J. VIROL.

http://jvi.asm.org/

(4)

HSV

Ml

18S---ated by digestion with theenzymes used in these

analyses have beenobtained (25; A. J. Davison

and N. M. Wilkie, personal communication).

Thesemaps areshownforone orientation of the

unique regions (Fig. 4). Fluorographs of the

pooled IE RNA fractions hybridized to the

BamHI, HindIll, and HpaI digests of HSV-1

DNAareshown inFig.5,6,and7,respectively,

andarecomparedtothe total fragmentpatterns

(revealed by hybridization of 32P-labeled HSV-1 DNA to the blotstrips) and the hybridization

patterns of unresolvedpolyadenylated

cytoplas-mic IE RNAs.

It is apparent from hybridization to the

BamHI digest (Fig. 5) thatgood separation of

RNAsequenceshas been obtained.Hence,with

peak fraction 5 there was abundant

hybridiza-tiontoBamHI-n, -x, and-z, ascomparedtolittle

or nohybridizationtothesefragments with

frac-li

'o

nw E

rz

z _

(A ._ o

FIG. 2. Autoradiograph ofthegel shown inFig.1.

were each hybridized to blot strips containing

the DNA fragments generated by digestions

with theBamHI,HindIII,andHpaIrestriction

endonucleases,andhybridizationwasvisualized by fluorography. Interpretation of these RNA

mapping datarequiresaknowledge of the

HSV-1 genome structure.

Asfirst describedby SheldrickandBerthelot

(20), thegenomehasanunusualstructure,

con-sistingoftwouniqueDNAregions (ULandUs),

each flanked byinvertedrepeat regions (TRL/

IRL,TRs/IRs) andjoined atIRL-IRs. Four ge-nomearrangements,resultingfrom thepossible

combinations of inversions of the two unique

regions, are present in DNA preparations in

approximately equal amounts (2, 7). A further

consequence of thisstructure is that a number

offragments produced by digestionwith a

re-striction endonucleasemay contain sequences in

common (25). Hence, hybridization ofRNA to a fragment containing any part of the repeat

regionsdoesnotnecessarilyimplythat this RNA

was transcribed from that particular fragment.

Physicalmaps for the DNAfragments

gener-18S 28S

51

3 1

A

4L,,

2

Fraction Number

FIG. 3. Distribution of32P-labeled total and virus-specific RNAafterresolutionof IE mRNA's on pre-parativegels. The positions of endogenous 28S and 18S rRNA'sareindicated,as arethe positions of the eluted RNAfractions(underlined numbers1through 9), which werepooledand used insubsequent anal-yses.

Map Units

0 01 02 0.3 04 0-5 06 07 0.8 0.9 1

Hind III

o h j kI d imn g

Hpa

mo pkuj n t b h f e qsvrml c g

I

Im

i 1 I

Bam HI

s e c o m t pu gv rw d h o f b sq nJzxyq

aifi'debj g c'

a=cm

d=g.m

k=s-.q

FIG. 4. Physical mapsforthefragmentsofHSV-1 DNA generated by restriction endonucleases HindIII, HpaI,and BamHI.

http://jvi.asm.org/

[image:4.508.79.224.72.398.2] [image:4.508.258.449.276.415.2] [image:4.508.256.450.508.640.2]

(5)

IE 1 2 3 4 5 6 7 8 9 DNA *e

41.KI ..

k ;

4-nvi

0

n ... .

S

FIG. 5. Fluorographs ofthepooledIERNAfractionsindicated inFig.3(fractionsl1through9)hybridized

totheBamHIfragments ofHSV-1I DNAarecomparedtothehybridizationpatternsofunresolvedcytoplasmic poly(A)IEG RNAs(IEG)and HSV-1DNAlabeledbynick translation(DNA).Fractions1,3, and5containedthe 4.7-,the3.0-, and the 2.0-kb RNA sizeclasses,respectively.

IE 1 2 3 4 5 6 7 8 9 DNA

[image:5.508.107.405.54.383.2]

his ... .

FIG. 6.

Fluorographs

ofthepooledIERNAfractions(i through9)hybridizedtothe

HindrII

fragments of HSV-1 DNA are

compared

to thehybridizationpatternsofunresolvedcytoplasmic poly(A) RNAs(IE)and HSV-1DNA (DNA).

46

http://jvi.asm.org/

[image:5.508.122.405.441.636.2]

(6)

[image:6.508.109.396.73.305.2]

FIG. 7. Fluorographs ofthepooledIERNAfractions (1 through 9) hybridized to the HpaI fragments of HSV-I DNA arecomparedtothehybridization patterns of unresolved cytoplasmic poly(A) RNAs (IE) and HSV-IDNA(DNA).

tions 3 and 1. A smallamount of material

hy-bridizing toBamHI-j, -k, and -qappears to be

distributedthroughout the gel, and this

presum-ably reflects considerable sizeheterogeneity of

RNAspecies containing thesesequences. Little

RNAcomplementarytoBamHI-owasdetected

afterresolutionof IE RNAsonthesegels, and it

is possible that these sequences were resolved

outside thegel fractions analyzed.

The RNAmapping data for fractions1,3,and

5 (corresponding to the 4.7-, 3.0-, and 2.0-kb RNA species) have beensummarized in Fig.8.

Inmaking this summary,onlyabundant

hybrid-ization tofragments has been considered. This

analysis may locate only the major sequences

present in the separated RNAs. Hence, small noncontiguous orleader sequences may fail to

bedetected.

Fraction 1 hybridized predominantly to

BamHI-yand alsoto-kand-q.Fragmentsyand

q are adjacent and map wholly within TRs/

IRs (Fig. 4),whereas k is a fusionfragmentof q

and s. Because there was no hybridization to

fragments, the 4.7-kb RNAappears tohybridize

exclusivelytopart ofthe TRs/IRs region.

Fraction 3 (the 3.0-kb RNA) hybridized

pre-dominantly toBamHI-b and -e, and also to -s.

BamHI-slies entirelywithinTRL/IRL,whereas

fragmentsb and e contain the same sequences

fromTRL/IRLbut differ incontainingsequences

from opposite ends of UL. That hybridization

was to the repetitive rather than the unique

DNA sequences of BamHI-b and -e may be

deduced from the HpaI digest (Fig. 7), where

there was hybridization to HpaI-m, but not to

HpaI-sor toanyotherfragmentwhichdoesnot

containsequences presentinHpaI-m. The

3.0-kb RNA, therefore, hybridized only to part of TRL/IRL.

Fraction 5 (the 2.0-kb RNA) hybridized

pre-dominantlytothree distinctregions of thevirus

genome. There washybridization toBamHI-b,

-n, -x, and -z (Fig. 5). Fragments x and z are

adjacent and lie at the rightend of Us (inthe

orientationgiven inFig. 4). Fragmentzcontains

only Ussequences, whereas x contains part of

TRs/IRsinaddition to Ussequences. A priori,

hybridization to BamHI-n could, therefore, be duetosequencesshared withx.However,since

there was hybridization to HindIII-n (Fig. 6),

which maps entirely within the left end ofUs,

fraction 5 must contain sequences

homologous

to each end of Us. Evidence that these

tran-scriptsarein partcomplementarytoTRs/IRsis

presented below.

Thethird regionto which fraction 5

hybrid-izedmaps inBamHI-b.The

HpaI digest

(Fig.

7)

indicates thatthis

hybridization

wastothe UL

sequencesof

BamHI-b,

sincethere was

hybrid-izationtoHpaI-s,butnot toHpaI-mor-r.

a c.d.o.

9.h

5 ... ....:

http://jvi.asm.org/

(7)

48 WATSON, PRESTON, AND CLEMENTS

Map Units

0 01 02 03 04 0-5 06 0-7

Total IE RNA

Ci

Fraction 1

addition, RNA sequences homologous to BamHI-z. These sequencesweresusceptible to 0:8 0:9 1 RNase treatment and arethus nonhomologous to fragment n. The RNase-resistant sequences whichhybridizedtofragment x,therefore,must be complementary to part of TRs/IRs, since these sequences alone are common to both -n

<=:2 and -x.

The RNase-resistant sequences selected against BamHI-z hybridized only back to BamHI-z (Fig. 9). Undigested RNA selected against thisfragment hybridized toboth-xand -n, in addition to-z. Thisimplies that there isa

DNA

RNA

ase

L

TRL U

J

be6 S 0

IRL IRSUs TRs

{

1--FIG. 8. Summary oftranscriptmappingdata.Map locations ofthe IE mRNA's contained in resolved fractions 1, 3,and 5(the4.7-, 3.0-,and 2.0-kbRNAs, respectively) are compared to those ofunresolved cytoplasmicpoly(A)IE mRNA'sbythefollowing cri-teria: (i) regions to which an abundance ofRNA hybridizedareindicatedbyadouble continuousline;

(ii) clearly detectable hybridization to a region is representedbyasinglecontinuousline;and(iii) low

levels ofhybridizationareindicatedbya

discontin-uousline.

Selection of transcripts which overlap thejunction between Us and TRs/IRs. To provide evidence that the 2.0-kb transcripts whichmap at eachend ofUsare also

homolo-gous in part to TRs/lRs, total cytoplasmic poly(A) IE RNAs werehybridized atlow tem-perature to duplicate filters containing either BamHI-n or -z. After hybridization, the filters

werewashedextensively. One of each duplicate

was then treated withRNase under conditions

inwhichRNA in an RNA/DNA hybrid is

pro-tected, but single-stranded RNA is susceptible

to digestion. This treatment, therefore, elim-inates tails ofnonhomologous RNA sequences

covalently linked to the hybridized RNA

se-quences. RNA sequences from the

RNase-treated anduntreatedfilterswerehybridizedto thetotalBamHI DNAfragments (Fig.9).RNA selected against BamHI-n, both from treated and untreated filters, was homologous to

BamHI-n and-x.Nondigested filters selected,in

.!..

~ ~

no:.-:'

x

z

n

z

FIG. 9. IE mRNA's selected by hybridization to filters containing HSV-1 DNAfragments BamHI-n

or-zwereeither treated withRNase(+)oruntreated

(-), andthe hybridizedRNAs wereeluted. Fluoro-graphs of the elutedRNAsampleshybridizedtoblot stripscontainingtheBamHI DNAfragments(right)

are compared to the total fragment pattern (left) revealedbyhybridization of32P-labeled HSV- I DNA tothefragments.

Froction 3

E::-Froction 5

J. VIROL.

JL

http://jvi.asm.org/

[image:7.508.67.258.75.346.2] [image:7.508.317.413.216.581.2]

(8)

transcript whichspansthe BamHIcleavagesite

between-zand-xand which containssequences

presentinTRs/IRs.

In vitrotranslation ofpooledRNA

frac-tions. The pooled IE and MI RNA fractions

wereusedtodirectprotein synthesisinarabbit

reticulocyte cell-free translation system. For

comparison, unfractionated IE and MI RNAs

were translated in parallel. The products of in

vitro translation were analyzed by PAGE, and

theresultantautoradiographsareshown inFig.

10and 11. Also shown areIE and MI

polypep-tideslabeledinvivo.

Anumberofvirus-specific polypeptideswere

made when the cell-freetranslationsystemwas

programmed with total IERNA, and attention

is drawntopolypeptides Vmw 175, 136, 110, 87,

63,and 12(Fig. 10).Thesynthesis,both in vivo

and in vitro, ofa virus-induced polypeptide of

estimated molecular weight 38,000 was als6

noted(Fig. 10and11).Thispolypeptidehasnot

been classifiedanIEpolypeptide, since its

syn-thesis isnotinvariablyobserved.

The fractionated IE RNApreparations,

like-wise,directed thesynthesisofanumber of virus

in vivo

MlIGE Ml IE 1 175- .._.

136-110-r

87-68-_ 63-_

38::::0s_

polypeptides. Fraction 3 (the 3.0-kb mRNA

band) specified primarily Vmw 110, although

smallamountsof Vmw 87were also synthesized

(Fig. 10). Synthesis of this latter polypeptide

was also directed by fraction 4 RNA, which indicates that the major IE mRNA contained by

fraction3specified Vmw110.

Fraction5 (the 2.0-kb mRNA's) specifiedthe

invitro synthesis ofVmw68, 63,and12(Fig. 10).

Thesynthesisof small amounts ofVmw 63 and

reducedamountsof Vmw12 wasalso directed by

fraction4.Thevirus-inducedpolypeptide of

es-timated molecular weight38,000 wasprimarily specified byfraction7.

Few products of translation of fraction 1

mRNA (the 4.7-kb size class) were apparent

(Fig. 10),and thispresumably reflectsthe diffi-cultyofisolatinglargemRNA's intact. Alonger

exposure ofthegel shown in Fig. 10 (datanot

shown) indicated thatsmallamountsofVmw175 weredirectedbyfraction1.Becausesynthesisof

other high-molecular-weight polypeptides was

also apparent at such an exposure,

demonstra-tionthat the 4.7-kb virus mRNAspecified Vmw

175alone isequivocal.

2 3 4 5 6 7 8 9 -RNA

:':.:!'.: :,::Zi,B;

. .,

__

12-

w-_

VW v *

in

vitro

translation

FIG. 10. In vitro translation ofresolved IE mRNA's. Comparison by PAGE oftheproducts of in vitro translationoftheresolved IE mRNA's(fractions I through9)with those directed bytotal IE mRNA's and total MI mRNA's translated both in vivo and in vitro. Also shown are the products oftranslation of

endogenous mRNA whenthecell-freesystem wasincubatedinthe absenceofadded RNA(-RNA). Virus-specificpolypeptidesareindicatedtotheleftoftheappropriatetrack(0).

http://jvi.asm.org/

[image:8.508.106.398.335.615.2]

(9)

50 WATSON, PRESTON, AND CLEMENTS V.iVQ1YvIII vitro

zE Ml IE MI 1 2 3 4 5 6 7 8 q -PNA

FIG. 11. In vitro translation ofresolved MImRNA's. Comparison by PAGE oftheproducts of in vitro translationoftheresolved MI mRNA's(fractions 1 through 9)with thosedirectedbytotal MImRNA'sand totalIEmRNA's translated both in vivo and in vitro. Virus-specificpolypeptidesareindicatedtotheright of theappropriate track (0).

DISCUSSION

Three distinctvirus-specificRNAbandswere

observed when IE RNAwasseparated by

elec-trophoresis through CH:,HgOH agarose gels.

Thesehadestimated sizes of 4.7, 3.0, and2.0kb. Thetwolarger specieswererelatively

homoge-neous asdetermined by in vitro translationand genomelocation,whereasthe2.0-kbcomponent

washeterogeneous. Thus, the 4.7-kb RNA

hy-bridizedtopartof theTRs/IRsregion(between

map coordinates0.830and0.866,and0.964and 1.000) andspecified small amounts of Vmw 175, whereasthe 3.0-kb RNAhybridized to part of

TRL/IRL (between map coordinates 0.000 and

0.038, and 0.792and0.830)and specified predom-inantlyVmw 110.The2.0-kbRNA hybridizedto threeregions(defined bymapcoordinates0.741

and 0.754, 0.865and 0.898,and0.938and 0.964) andspecifiedthesynthesisofthreepolypeptides (Vmw68, 63,and12).

Thesevirus RNAspeciesweresized by using

28S and 18S rRNA'sasmarkers, andestimates

depend on theaccuracy with which the molec-ularweightsof these markers have been deter-mined. The rRNA molecular weight estimates used herewereobtainedby using denaturinggel systems (12), but similar studies have given somewhat different sizes (1).Nonetheless, with the notable exceptionofthe mRNAcodingfor

Vmw 12,the mRNA'scodingfor IEpolypeptides

are consistent with the sizes required for their

protein codingsequences.

Figure 8 shows that the three virus-specific RNA bandshybridized to restricted regions of the virusgenome,and theseweremostlylocated

within or adjacent to the repetitive DNA

se-quences. Unfractionated IE mRNA's were

shown (Fig. 8) tohybridize to a number of ad-ditional loci mapping in the unique DNA

re-gions,and these RNAsweremostlyof low abun-dance (Fig. 8).IERNAs of low abundancewere

not considered whenmaking the summaries of the DNAregionsrepresented bythethree virus mRNA size classes. Furthermore, certain

tran-scripts, i.e., those hybridizing to BamHI-j (Fig.

.1 ".

;36--%810 7-

87-J. VIROL.

http://jvi.asm.org/

[image:9.508.116.401.73.385.2]

(10)

5), were heterodisperse in size and were not

resolved by using this gel system. Therefore,

these analyses allow characterization of the

abundantIEmRNA'sonly.

HSV-1 IEpolypeptideshavepreviously been

mapped by analyses of the virus genomes and

thepolypeptidesexpressed by intertypic

recom-binants between HSV types 1 and 2 (13, 19).

Theseanalysesdependeduponsmall differences

inmobilityonPAGE between equivalentHSV

types 1 and2polypeptides which allowed

iden-tificationof the type-specific proteins expressed byrecombinants.Strictly, this methodmapsthe

genome regions affecting polypeptide mobility

rather than delineating the polypeptide coding

regions. Thetechniqueused heredirectlymaps

themRNA's coding for thesepolypeptides. The

combination of both sets of data has allowed

someof the IE polypeptides tobe mapped on

theHSV-1 genomemoreprecisely.

Intertypic recombinant analyses have mapped

Vmw 175, which is equivalent to ICP 4 (8), in

TRs/IRs (13, 19) andVmw 110,which is

equiva-lent to ICP 0 (8), in TRL/IRL. Recombinants

whichareheterotypicfor TRL/IRLexpressboth the HSVtype 1Vmw110and its Vmw118type 2

equivalent (19).Because bothTRL and IRLare

expressedeveninsuch recombinants which have

afixed,noninvertingorientation of theLregion

(19), this indicates that Vmw 110-coding

se-quences lieentirely within repetitive DNA.This

location is consistent with our data.

Recombi-nants heterotypic for TRs/IRs made both the

HSV-1ICP4polypeptideand itsHSV-2

equiv-alent (13). This suggests that these coding

se-quencesmapentirelywithinTRs/IRs.Our data

are consistent with this location and indicate

thatthe mRNA is homologous toonly partof

TRs/IRs(therewas nohybridization of the

4.7-kb mRNAtoBamHI-nor-x).

Previous analyses ofintertypic recombinants

in ourlaboratory havemappedVmw63between

coordinates0.562and0.790, andVmw68andVmw

12in theS component(19). Correlation of these

analyseswiththe data shown heregivesa more

precise maplocation for Vw63 (betweenmap

coordinates0.741 and0.754) and placesVmw68

and 12,respectively, betweencoordinates0.865

and 0.898, and0.938and0.964.Similaranalyses

of intertypic recombinants reported by other

workers(13)mapped ICP27(anIEpolypeptide probably equivalent to Vmw 63) between map

coordinates 0.790 and 0.830, which is part of

TRL/IRL. The reason for the discrepancy be-tween these data and our map location is un-clear.However,ouranalyseswereperformed by

labelingtheIEproteinssynthesizeduponrelease

of a cycloheximide block, whereas those of Morseetal. (13)wereperformedby

pulse-label-ing thepolypeptidesmadeatvarioustimes

post-infection. Since ICP 27 waslabeled when

cells

werepulsed from9 to 10hpostinfection,a

time

atwhich IE polypeptide synthesisis negligible

(8), it is

unlikely

that the protein mappedwas

equivalent to the IE polypeptide Vmw 63. It is

also

pertinentthatrecombinants heterotypicfor

TRL/IRL made only one type-specific ICP 27

(13).

The results presented here indicate that the

two2.0-kbtranscripts whichmap in theS region

overlapthejunctionbetween Usand both TRs

andIRs.The reasons forthis conclusionare that

only the 2.0-kbRNA (fraction 5) hybridized to

fragments (BamHI-n and -x) containing these

junctions, and that IE transcripts homologous

to BamHI-n are also, in part, homologous to

BamHI-x. Data recently obtained in our

labo-ratory (J. B. Clements and D. J. McGeoch,

manuscript submitted for publication),by

map-ping

cDNA made tothe 3' end ofIE mRNA's

against HSV-1 DNAfragments, have indicated

thatthe 3' ends of both these mRNA's liein Us.

This

implies

thatthe 5' ends of these mRNA's

mapwithinTRs/IRs,and itseemsreasonableto

presume that thetwotranscriptsaretranscribed

from

locationally

distinct promoters of identical

sequence. Two such mRNA's would contain

identical5'ends, and distinct 3' ends.

Vmw 12 is synthesized from a much larger

mRNA

than isrequired for its coding sequences.

Suchasituation has been described for the small

T-antigen of simian virus40 (14), which

shares

N-terminal

peptides

with the large

T-antigen

(15)and isterminatedatacodon which appears

tobe excised from the large T-antigen

rnRNA

(5). Polypeptides with identical N-terminal

re-gionscould

arise

withHSV-1 as aconsequence

of the genome structure. Hence, the mRNA's

which code for Vmw 68 and

Vmw

12 and which

may map at either end of Us could code for proteins with identical N-terminals (codedfor byTRs/IRssequences)butdistinctC-terminals (coded for by Us

sequences).

These two

poly-peptides chains would be terminated by noni-denticalterminator codons

(encoded

byunique

DNA sequences from

opposite

ends ofUs),and

the

maxinux

length encoded bytheTRs/IRs

DNAwould be that

required

to

specify Vmw

12.

ACKNOWLEDGMENTS

Wethank J.H.Subak-Sharpefor hisadvice andinterest in thiswork.

R. J. Watsonwas the recipient ofa Medical Research Council grant fortraininginresearchtechniques.

LITERATURE CITED

1. Bailey,J.M.,and N. Davidson.1976.Methylmercury as areversibledenaturingagent for agarosegel electro-phoresis.Anal. Biochem.70:75-85.

VOL. 31, 1979

http://jvi.asm.org/

(11)

52 WATSON, PRESTON, AND CLEMENTS 2. Clements,J.B.,R. Cortini,and N. M. Wilkie. 1976. Analysis ofherpesvirus DNA substructure bymeansof restrictionendonucleases. J. Gen. Virol. 30:243-256. 3. Clements, J. B.,R. J.Watson, andN.M.Wilkie.1977.

Temporal regulation ofherpes simplex virus type 1 transcription: location oftranscripts on the viral ge-nome. Cell 12:275-285.

4. Costanzo,F.,G.Campadelli-Fiume,L.Foa-Tomasi, and E. Cassai. 1977. Evidence that herpes simplex virusDNAistranscribed bycellularRNApolymerase B. J. Virol. 21:996-1001.

5. Crawford, L.V., C. N. Cole, A. E.Smith,E.Paucha, P.Tegtmeyer, K. Rundell, and P.Berg. 1978. Or-ganization andexpressionofearlygenesof simian virus 40.Proc. Natl.Acad. Sci. U.S.A.75:117-121. 6. Frenkel, N., and B.Roizman. 1972. Ribonucleic acid

synthesis in cells infected with herpes simplex virus: controlsoftranscriptionand of RNAabundance.Proc. Natl. Acad.Sci.U.S.A. 69:2654-2658.

7. Hayward, G.S.,R. J.Jacob,S. C.Wadsworth,and B. Roizman. 1975.Anatomy ofherpes simplex virus DNA:evidenceforfourpopulationsofmoleculesthat differ in the relativeorientations of theirlong and short components. Proc. Natl. Acad. Sci. U.S.A. 72:4243-4247.

8. Honess, R.W., and B.Roizman. 1975. Regulation of herpesvirusmacromolecular synthesis: sequential tran-sition ofpolypeptidesynthesis requires functional viral polypeptides. Proc. Natl. Acad. Sci. U.S.A. 72:1276-1280.

9. Jamieson, A. T.,J.C. M.Macnab, B. Perbal, andJ. B.Clements.1976.Virusspecifiedenzymeactivityand RNAspeciesinherpes simplexvirus type 1transformed mousecells. J. Gen.Virol.32:493-508.

10. Jones,P.C.,G. S.Hayward,andB.Roizman. 1977. Anatomyofherpes simplexvirus DNA. VII. a RNA is homologoustononcontiguous sitesin boththeL and S components of viral DNA. J. Virol. 21:268-276. 11. Marsden, H. S., I. K. Crombie, andJ. H.

Subak-Sharpe. 1976.Controlof proteinsynthesisin herpes-virusinfected cells-analysisof thepolypeptides induced by wildtypeand sixteen temperature-sensitive mutants ofHSV strain17.J.Gen. Virol. 31:347-372.

12. McMaster, G. K., and G. G. Carmichael.1977.Analysis ofsingle-and double-stranded nucleic acidson poly-acrylamide and agarose gelsbyusingglyoxaland acri-dineorange. Proc.Natl.Acad. Sci.U.S.A. 74:4835-4838.

13. Morse, L. S., L.Pereira, B.Roizman, and P. A. Schaf-fer. 1978. Anatomy of herpes simplex virus (HSV) DNA. X.Mappingof viral genesbyanalysisof polypep-tidesand functions specified by HSV-1 xHSV-2 recom-binants.J. Virol. 26:389-410.

14. Paucha, E.,R.Harvey,and A. E.Smith.1978.Cell-free synthesis of simian virus 40 Tantigen.J.Virol. 28:154-170.

15. Paucha, R., A.Mellor, R.Harvey, A. E. Smith, R. M. Hewick, and M. D. Waterfield.1978.Large and small tumour antigensfromsimian virus40 have identical aminoterminimappingat0.65mapunits.Proc. Natl. Acad. Sci.U.S.A. 75:2165-2169.

16. Pelham, H. R. B., and R. J. Jackson. 1976.Anefficient mRNA-dependenttranslationsystemfromreticulocyte lysates.Eur.J.Biochem. 67:247-256.

17. Pereira, L., M.H.Wolff,M.Fenwick,and B. Roizman. 1977. Regulation ofherpesvirus macromolecular syn-thesis V.Propertiesofpolypeptides madeinHSV-1and HSV-2 infected cells.Virology 77:733-749.

18. Preston, C.M. 1977. Thecell-freesynthesisof herpesvi-rusinduced polypeptides.Virology 78:349-353. 19. Preston,V.G.,A. J.Davidson,H. S.Marsden,M. C.

Timbury, J. H.Subak-Sharpe, and N.M.Wilkie. 1978.Recombinantsbetweenherpessimplexvirustypes 1and 2:analysesofgenome structuresandexpression ofimmediateearlypolypeptides. J. Virol. 28:499-517. 20. Sheldrick, P., and N. Berthelot. 1974.Inverted repeti-tionsinthe chromosomeofherpes simplexvirus.Cold SpringHarborSymp. Quant. Biol. 39:667-678. 21. Southern, E. M. 1975. Detection ofspecificsequences

amongDNAfragments separated by gel electrophore-sis. J.Mol. Biol. 98:503-533.

22. Swanstrom,R.I.,and E. K.Wagner.1974.Regulation of synthesis of herpes simplex type 1 virus mRNA during productiveinfection.Virology60:522-533. 23. Taylor, K.,Z. Hradecna, and W.Szybalski. 1967. A

symmetric distribution of thetranscribing regionson thecomplementary strandsofcoliphageADNA.Proc. Natl. Acad.Sci.U.S.A.57:1618-1625.

24. Watson,R. J., andJ. B.Clements. 1978. Characteri-zation oftranscription-deficient temperature-sensitive mutantsofherpessimplexvirus type 1.Virology 91: 364-379.

25. Wilkie, N. M. 1976. Physicalmaps forherpes simplex virustype 1DNA for restrictionendonucleasesHindIII, Hpa-1, and X. bad. J. Virol. 20:222-233.

J. VIROL.

http://jvi.asm.org/