High-resolution characterization of herpes simplex virus type 1 transcripts encoding alkaline exonuclease and a 50,000-dalton protein tentatively identified as a capsid protein.

0022-538X/83/120591-13$02.00/0

Copyright © 1983,American Society for Microbiology

High-Resolution Characterization of Herpes Simplex Virus

Type

1

Transcripts Encoding Alkaline Exonuclease

and

a

50,000-Dalton Protein Tentatively Identified

as a

Capsid

Protein

R. H.COSTA,1 K. G.DRAPER,1 L. BANKS,2 K. L. POWELL,2 G. COHEN,3R. EISENBERG,4 AND E. K.WAGNER'*

Department of Molecular Biology and Biochemistry, University of California, Irvine, California 92717'; Department ofMicrobiology (Virology), University of Leeds, Leeds LS2 9JT, England2; and Department of

Microbiology and Center for Oral Health Research, DentalMedicine,3andDepartmentof Pathobiology, Schoolof Veterinary Medicine,4 University of Pennsylvania, Philadelphia, Pennsylvania 19104

Received 27 June 1983/Accepted 9 September1983

Fourpartially overlapping mRNAs (1.9, 2.3, 3.9, and 4.5 kilobases [kb])were

located between 0.16 and 0.19 map units on the herpes simplex virus type 1

genome. Their direction oftranscription was foundto be fromrightto left. The

2.3-kbmRNAwasfoundtobeearly(1),whereas the otherswerelate

(P3y).

Partial

sequence analysisof the DNA encoding thesegenes indicated that the promoter

for the2.3-kb mRNA shares structural features with otherearly(1)promoters. In

vitrotranslation of hybrid-selectedmRNA indicated thatamongtheproteinsthese

mRNAs encodeare an82,000-dalton (d) polypeptidereactive withamonoclonal

antibodyagainst herpes simplexvirustype 2 alkaline exonucleaseanda50,000-d polypeptide weakly reactive with apolyclonal antibodymade against thecapsid protein VP19C. Furtherexperiments suggested that the 2.3-kb mRNA encodes the82,000-d polypeptide,whereasone(or both)of thelargermRNAsencodes the

50,000-dprotein. Anovelfindingwasthat the 1.9-kb mRNAappearstosharepart

of the translationalreadingframe for alkalineexonuclease,butanypolypeptideit

encodes does not reactwith the monoclonalantibody to thisenzyme.

We have established in a number of recent publications (reviewedinreference 35)that her-pes simplex virustype 1 (HSV-1) mRNAs map relatively simply on the viral genome. Further-more, the low frequency and relatively short spanofthecharacterized HSV-1 intronssuggest that the correlation ofspecific transcripts with genetic markers and known viralproteins willbe generallystraightforward. Thus, HSV transcrip-tion maps add a valuable further dimension to high-resolution genetic marker localization.

As described in the most recent available genetic mapping report (S. K. Weller, W. R. Sacks, D. M. Coen, and P. A. Schaffer, Virolo-gy,in press), five complementation groups map in the region between 0.1 and 0.2 mapunits on

theParrangementof the HSV-1 genome. Details ofthe datamappingthese groupsarepresented in thatpublication. Twocharacterized viral pro-teins have been located in thisregion. The first is alkaline exonuclease, an important enzyme of HSV-induced DNA replication (10, 16, 25, 26, 33). Preston andCordingley (29) mapped this by measuringenzyme levels of in vitro translation productsofhybrid-selected mRNAandby

per-forming hybrid-arrestedtranslation. The second viral protein characterized isa50,000-dalton (d) capsid protein (VP19C) located byLemasterand Roizman (20),using intertypic recombinants.

Recent work (L. Banks, D. J. M. Purifoy, R. A. Killington, and K. L. Powell, J. Gen. Virol., in press) has led to the isolation of several monoclonal antibodies against HSV-2 alkalineexonuclease, some of whichare

cross-reactivewith theHSV-1-inducedenzyme. Previ-ously reported work (5) described the prepara-tion ofpolyclonalrabbit antiserareactivewitha

50,000-d capsid protein (NC-2 in that publica-tion), which can be inferredto be VP19C. This protein appears to be located on the capsid vertices (34).

Inthe presentstudy,weused theseantiserato

identifythe mRNAencodingthealkaline

exonu-cleaseas a 1 (early) one and, tentatively, a

Py

(late) mRNA as encoding the capsid protein. High-resolution mapping of these mRNAs indi-cated thatall are partially colinear. Correlation oftranscript locationwith nucleotide sequence data demonstrated that the 5' end of the , mRNA lies downstream of a 120-base region

591

on November 10, 2019 by guest

http://jvi.asm.org/

Downloaded from

on November 10, 2019 by guest

http://jvi.asm.org/

Downloaded from

on November 10, 2019 by guest

http://jvi.asm.org/

sharingcertain features with promoters for other ,B HSV-1 mRNAs.Another mRNAwasfoundto

be colinear with the 3' portion of the alkaline exonuclease mRNA. It appeared to have its own independent 5' end and was temporally a

3-y

mRNA. Preliminary data suggested that this mRNA can be translated in vitro to

yield

a polypeptide smaller than alkaline exonuclease. Such apolypeptide was not reactive with theQl antibody, but partial sequence analysis suggest-ed that it is encoded by the same translation frame as the alkaline exonuclease.

MATERIALSANDMETHODS

Cellsandvirus. ForRNApreparation,

plaque-puri-fied isolates of the KOS strain of HSV-1 were usedto

infect HeLa cells. Monolayer cultures of HeLa cells

weregrown at 37°C in Eagle minimal essential medium

containing 10% calf serum, penicillin, and

streptomy-cin. HSV-2alkaline exonuclease was prepared from

human epidermoid carcinoma no. 2 (HEp-2) cells

grownin thesamemedium. HSV-2wasthe 186strain.

Enzymes. All restriction enzymes were obtained

from Bethesda Research Laboratories. Digestions

were carried out in buffers recommended by the

supplier. Phage T4 polynucleotide kinase (Bethesda

ResearchLaboratories)wasusedfor 5' endlabelingas

described byMaxamand Gilbert(23). Escherichiacoli

DNA polymerase I (Klenow fragment,

Boehringer-Mannheim) was used to generate 3'-end-labeled

re-strictedDNAbyprocedures described by Maniatiset

al(21).

Isolation, labeling,andsizefractionation of

polyribo-somal RNA.Monolayer cultures ofHeLacells(2x107

cells perflask)wereinfected for30minat a

multiplic-ityof10PFUof virus per cell inphosphate-buffered

saline containing 0.1% glucose and 1.0% fetal calf

serum. Polyribosomes were isolated from the

cyto-plasm ofHSV-1-infected cells by themagnesium

pre-cipitation method of Palmiter (27). Polyadenylic

acid-containing [poly(A)] mRNA was isolated from total

rRNAbyoligodeoxythymidylicacid-cellulose

(Collab-orative Research, Inc.) chromatography. This is

re-ferred to as HSV poly(A) mRNA. Details of this

procedurewere presented elsewhere (13). RNA was

isolated at 6 h postinfection except when early (,)

RNA wasrequired. Then cellswerepreincubatedfor1

h andtreated for 5 h with 1.5 x

1i-'

M adenosine

arabinoside and 3.7 x 106 M pentostatin to inhibit viral DNA replication, as described previously (15). RNA wassizefractionatedbyelectrophoresison1.4%

agarose gels containing 10 mM methylmercury

hy-droxide(3)as previously described (1, 2).

Recombinant DNA. All recombinant DNA clones

described in this paper were derived from either

HindIIIfragment IO(map units0.082-0.182),HindlIl

fragment J (0.182-0.261). XhoI-HindIII fragment

C'-10 (0.171-0.182), BamHI-HindIII fragment A-IO

(0.152-0.182), or HindIII-SalI fragment J-D

(0.182-0.195), cloned in pBR322. Procedures for cloning

HSV-1 DNA fragments in the pBR322 vector were

described previously (1, 6). Cloned DNA fragments

were named asdescribed previously andlocated by

their map coordinates on the P arrangement ofthe

HSV-1 genome(6).

In situ Northern RNAblots. As described

previous-ly, samples (5 ,ug) of HSV poly(A) mRNA were

fractionated on methylmercury gels and dried onto

Whatman3mm paper with vacuum (15). The agarose

film was floated off the paper in water and hybridized

with appropriate nick-translated, 32P-labeled DNA

probesin50% formamide containing 0.4MNa+, 0.1 M

HEPES

(N-2-hydroxyethylpiperazine-N'-2-ethanesul-fonic acid; pH 8.0), 0.005 M EDTA, and Denhardt

solution (7)at50°C for 36 h. Blots were rinsed at 50°C:

thefirst two rinses were in50%

formamide-2x

SSC

(1x SSC=0.15 MNaCl plus 0.015 M sodium

citrate)-0.1% sodium dodecyl sulfate (SDS). The last rinse was

in 0.1x SSC-0.1% SDS. Autoradiography was on

Kodak XRPfilm withorwithout intensifying screens

asneeded.

Invitro32P-labeledDNA wasmade by

nick-translat-ing appropriate DNA clones with DNA polymerase I,

DNase I (Boehringer-Mannheim), and 50 ,uCi of

[a-32P]dCTP(3,000Ci/mmol, Amersham).

Isolation of restrictionfragment-specific mRNA.

Re-strictionfragment-specific mRNA was isolated from

HSV poly(A)mRNAby preparative hybridization to

theappropriate DNA covalently coupled to cellulose.

Details ofcouplingofDNA tocellulose and

prepara-tivehybridizationwere asdescribed previously (2, 6).

Nuclease mappingofHSV-1mRNA.S1nuclease and

exonuclease VII analysis of RNA was carried out

essentiallyasdescribedby Berk and Sharp (4) andas

describedpreviously (1, 6, 8, 11-13, 15). Appropriate

HSV-1 DNA clones (10 ,ug) were cleaved at the

desired site with the appropriate restriction enzyme.

TheDNAthenwas5' labeled with[-y-32P]ATP(3,000

Ci/mmol, ICN), usingpolynucleotidekinase (Bethesda

ResearchLaboratories), toaspecific activity of 4,000

to 10,000 cpm/,ug of DNA. Alternatively, the DNA

was 3' labeled tothesame specific activity by using

DNA polymerase I-Klenow fragment

(Boehringer-Mannheim).

The DNAfragments thenweredenatured and strand

separated on 5% acrylamide gels as described by

Maxam and Gilbert (23). The strand-separated DNA

(from 10 ,ug of cloned DNA)washybridized with 10

jig

of infected-cell mRNA in 0.1 M Na+-0.1 M HEPES

(pH8.0)-0.01MEDTAat65°C for6 to16hina30-,u

volume.Hybrids were subjected toS1nuclease

(Boeh-ringer-Mannheim)or exonuclease VII (Bethesda

Re-search Laboratories) digestion as described (11, 12).

Materialwasfractionatedon adenaturing 5%

acryla-midegel with 5'-end-labeled, HaeII-orHinfI-digested

pBR322DNAfragmentsas sizestandards. When the

nuclease-protected materialwas tobe fractionatedon

sequenceladders, mung beannuclease(PL

Biochemi-cals)wasused insteadofS1 nuclease. Allprocedures

were basedon thoseofMaxamandGilbert(23).

Nucleotidesequencing. As describedpreviously (8,

13), nucleotide sequenceanalysis was carriedoutby

theprocedures ofMaxamand Gilbert(23).

Invitrotranslation.Translation of size-fractionated

viral mRNA was carried out in vitro by using a

micrococcal nuclease-treated rabbit reticulocyte

sys-tem(NewEnglandNuclearCorp.)with

[35S]methion-ine (>800 Ci/mmol) as the radioactive amino acid.

Detailsof theprocedureandfractionation of

polypep-tides in SDS-acrylamide gels by the method of

Laemmli(19) weredescribed in several previous

pa-pers. Gelswere treatedwithEn3Hance(NewEngland

on November 10, 2019 by guest

http://jvi.asm.org/

593

Nuclear Corp.) and dried with vacuum at60°C, and

radioactive bands were localized by autoradiography

withKodak XRP film. Exposure was for 3 to 5 days at

-700C.

Immunoprecipitation of in vitro translation products

wasperformed as described previously (13). One half

of the RNase-treated translation product (14 ,u) was

diluted with an equal volume of 2x lysis buffer and

incubated with 4 ,ul of polyclonal NC-2 serum, a

polyvalent antibody to the HSV-1 50,000-d capsid

protein (VP19C), or with 1 ,ul of Ql, a monoclonal

antibody made against purified HSV-2 alkaline

exonu-clease. Lysis buffer is 20 mM Tris-hydrochloride (pH

7.4)-S50 mMNaCl-10 mM methionine-0.5% Nonidet P-40-0.5% sodium desoxycholate-0.1% SDS. After 1

h onice, 50 ,ul of a10%suspension of Pro-ASepharose

beads (Pharmacia) in 50 mM Tris-hydrochloride (pH

7.5)-150mMNaCl-5mM disodiumEDTA-0.2%

sodi-um azide wasadded, and the suspension was

incubat-ed for a further 30 min on ice with frequent mixing.

ThePro-ASepharose beads with the immune complex

absorbedwerethendeposited by1 minof

centrifuga-tion in an Eppendorfmicrocentrifuge, and the pellet

wassuspended in 200p.loflysisbuffer and centrifuged

for 5 minthrougha1-mlpad oflysis buffer containing 1

M sucrose. TheSepharose was then washed by

sus-pension in 0.5 ml of 10 mM Tris-hydrochloride (pH

7.4)-150mMNaCI-10mMmethionine-0.2% Nonidet

P40-0.1%SDS and recentrifugation. After five wash-es, the Sepharosepellet wassuspended in 40 p.lofa

buffer containing 0.075 M Tris-hydrochloride (pH

6.8)-3% SDS-10% P-mercaptoethanol. The

suspen-sion was heated to 950C for 2 min, and the Pro-A

Sepharose was pelleted bycentrifugation. After this the supernatant, containing released immunoglobulin

andanytranslation product with which it had reacted,

wasloaded onto SDS-acrylamide gels for size

fraction-ation.

Inonesetofexperiments, we used washed,

Formal-in-killed Staphylococcus aureus (IgGsorb, The En-zymeCenter, Inc., Boston, Mass.) toprecipitate the immune complex. This method resulted in a

considera-blylargeramountofproteinprecipitatedbuta much

higherbackground of nonspecific radioactive proteins.

This procedure, however, yielded interpretable

re-sults.

Preparation of radioactive protein markers.

["S]methionine-labeled,

cytoplasmic infected-cell

protein was isolated by the basic procedure of

Ed-wards and Fan (9). Cultures (6 x 106cells) of HeLa

cellswereincubated from6 to8 hpostinfection with

200 p.Ci of

[lS]methionine

(400 Ci/mmol; New

En-gland NuclearCorp.) in4 mlofmethionine-free Eagle

minimal essential medium.

In vitro 125I labeling of purified HSV-2 alkaline

exonuclease was carriedoutby themethodof

Green-wood et al. (14). Samples of the purified protein (0.05

p.g)werelabeled with carrier-free1251(17Ci/ml; New

England NuclearCorp.).

Preparationandcharacterizationofmonoclonal

anti-bodies against HSV alkaline exonuclease. Details of

monoclonalantibody preparation are described in the

paper by Banks et al. (in press). Briefly, HSV-2

enzymewaspurified from high-salt extracts of

HSV-2-infected Vero cells (28). Steps included

DEAE-cellu-lose, phosphocellulose, and DNA-cellulose

chroma-tography. Enzyme activity was measured by the

method of Purifoy and Powell (30). Purity can be

assessed from the data in Fig. 3A. .

Monoclonal antibodies werepreparedby two

inocu-lationsofpurified enzyme (2 to 3 p.g) into the footpads

ofBALB/c mice during a 2-week interval. Cell fusion

was by the method of Kennett et al. (17).Hybrid cell

colonies were grownin microtiter plates (Linbro) and

tested for production of antibody with an

enzyme-linkedimmunosorbent assay (horseradish

peroxidase-conjugated anti-mouse immunoglobulin G; DAKO,

Denmark) using detergent extracts of HSV-2-infected

cells as antigen. Peroxidase assay was done with

0.01% orthophenylenediamine and 0.003% hydrogen

peroxide.

Theantibody used in this paper(Q1)wasshown to

be able toimmunoprecipitate and neutralize the

HSV-2alkaline exonuclease.Furthermore, it has been used

in an immunosorbent column topurifythe enzyme. It

is strongly cross-reactive with the HSV-1-induced

enzyme (K.Powell, unpublisheddata).

RESULTS

Four partially colinear mRNA species map between 0.16 and 0.19 on the HSV-1 genome. We used a number ofcDNA probes made to small (1.0- to 1.5-kilobase [kb]) fragments ofHSV-1 DNA mappingbetween 0.16 and 0.19 map units todetermine the number and time of appearance of abundant mRNAs. Four readily detectable mRNA species of approximately 2.1, 2.5, 4.1, and4.7 kb were seen inNorthern (RNA) blots of HSV-1 mRNAisolated from cells 6 h postinfec-tion. Wecould inferthattheactual mRNA sizes were 1.9, 2.3, 3.9, and 4.5 kb, since the average poly(A) tail length ofHSV-1 mRNA is ca. 200 bases(32). Nick-translated DNAprobesmapping betweenca. 0.16 and 0.174 map units hybridized to all mRNA species. Probes containing DNA mapping betweenca. 0.174 and 0.19 hybridized to only the three larger species. Examples of suchexperiments are shown in Fig. 1.

ANorthern blotof viral mRNA isolated from cells in which DNA synthesis was inhibited is

shown in Fig. 1A. Here the RNA blot was hybridized with32P-labeledDNA madeby nick-translating cloned BgIII-XhoI fragment O-G (0.164-0.171 mapunits).The2.3-kbmRNA was a majorband early, whereas the 1.9-, 3.9-, and 4.5-kbmRNAs were presentinlow abundance. The same probe revealed the 2.3-kb mRNA in equivalent abundance in Northernblots of late viral mRNA(Fig. IA). In this RNA, however, the other species were present in relatively greater abundance. This relativeabundancy dif-ference with time after infection suggested that the2.3-kbmRNAis an early

(p)

mRNA, where-asthe1.9-, 3.9-,and 4.5-kb speciesarelate(P,y) species. Criteria for temporal classification of specific mRNA specieshave been described in several recent reviews (31, 35, 36).

Weused two nick-translated probes to deter-mine how far the 1.9-kb mRNA extended

be-48, 1983

on November 10, 2019 by guest

http://jvi.asm.org/

mRNAsare

considerably

lessabundant than the

two smaller mRNAs.

E:zrl

t.+

e

High-resolution

localization of the 5' and 3'

endsof the overlapping mRNA species. We used 5- and3'-end-labeled single-stranded DNAs as

k*

;r5kr

1(

lhybridization

probes for precise S1 and

exonu-clease VII mapping of the four mRNAslocated in Fig. 1. Basic methods have been reviewed recently(35) and are outlined above. These data

FIG. 1. In situ RNA

(Northern)

blots of HSV-1 are summarized in Fig. 2.

mRNA encodedby DNA mapping between map units The 2.85-kb

BamHI-Xhol

fragment A-G 0.164 and0.190.Samples of HSV poly(A) mRNA from (0.152-0.171mapunits) was 3' end labeled at the cellsinwhich DNAsynthesis was inhibited (early) or XhoI site at map unit 0.171. Strand-separated allowedtoproceed (late) were fractionated on methyl- probe was hybridized with viral mRNA, the mercury-containing agarose gels and immobilized by hybrids were digested with S1 nuclease, and drying in vacuo. The region-specific RNA was detect- protected DNA was fractionated on a denaturing

edbyhybridization with nick-translated DNA probes

acrylamide gel.

The

major protected

species

was

as indicated. Sizes shown were determined by the

position of HeLa cell rRNA markers (not shown) as 1,600 bases long (Fig. 2A, trackS),

indlcating

a describedpreviously (1, 6). (A) mRNA species hybrid- major mRNA 3' end 1,600 bases to the left of the izing to a cDNA probe fromBglII-XhoIfragment O-G XhoI site at map unit 0.171. Minor amounts of (0.164-0.171 map units) (B) mRNA species

hybridiz-

other bands were seen with long exposures. ing to a probe from the 490 bases between theXhoI However, none contained more than 1 to 2% of

site at 0.171 andaDdeI siteindicated in Fig. 3 and4 the total radioactivity seen. Furthermore,

simi-(below). (C) mRNA species hybridizing to a probe lar experiments (not shown) using DNA 3'-made from HindIII-EcoRI fragment J-D (0.182-0.190 labeled at the BglII site at map unit 0.163 gave a

map units). major S1 protected band 400 bases long. These

data, whentaken with experimentsbelow, indi-catedthat thefour mRNAs shareacoterminal3'

yond

the XhoI site at map unit 0.171. A probe end ca. 400bases tothe leftof theBglII site at

extending

toaDdeI site490 bases to theright of 0.163. Wedidnotexcludethe presenceofsmall theXhoI site(seesequencedatabelow) hybrid- amounts of other RNA species terminating at izedtoallfourmRNAspecies(Fig.

IB).

A DNA othersites.

probe extending

from this DdeI site

rightward

to The 5' ends of the 1.9- and 2.3-kb mRNA theHindIII siteat0.182 didnothybridizedetect-

species

werelocatedby hybridizing viralmRNA

ably

with the 1.9-kb mRNA(not shown). with strand-separated XhoI-HindIII fragment

The two larger

P-y

mRNAs were

readily

C'-IO DNA (0.171-0.182 map units) 5' end la-showntoextend

beyond

theHindIII siteatmap beledattheXhoI siteatmapunit0.171. Exonu-unit 0.182 since a

32P-labeled

DNA probe made cleaseVII

digestion

ofhybrids (Fig. 2B, track X) fromHindIII-EcoRI fragmentJ-D (0.182-0.190 gave three protected species. One was 1,650 map

units) hybridized

tothem inNorthernblots bases

long, corresponding

to the full length of of late mRNA

(Fig.

1C). As mentioned

above,

theDNA. Thiswas dueto the

protection

ofthe other

probes

were used to show the

precise

3.9- and 4.5-kbmRNAs(see

below).

Thesecond

extent ofthese mRNAs. Heavier exposures of protected

fragment

was700baseslong,

suggest-Northern blots hybridized with HindIII-EcoRI

ing

that the 2.3-kbmRNAextends700 bases to

fragment

J-D (0.182-0.190 map

units)

also re- the

right

of the XhoI site at 0.171. The third vealed a relatively low-abundancy mRNA ca. species was 270 to 280 bases in

length.

This 2.7kb in size. This mRNA extendedtothe

right

suggested that the 5' end of the contiguous oftheareaof interestpresentedinthispaper and portion ofthe 1.9-kb mRNA extendsthisfarto

is not characterized here. the

right

of the XhoI site. In the exonuclease VII The dataofthenextsectiondemonstrated that

digestion

trackof

Fig.

2B (track X), someother the 3' end ofthefourmRNA

species

liestothe

(faint)

intermediate-sized bands were seen.

left of the

BgIII

site at map unit 0.164. There- However,thesewere notconsistentlypresent in

fore,

the

BglII-XhoI fragment

O-G (0.164-0.171 several repeat

experiments.

We suggest these map units) DNA probe could be expected to were

digestion

intermediates of the nuclease hybridizewith thefourmRNAscolinear through reaction.

Si

nuclease digestion of hybrids be-that

region

in amounts reflecting their relative tween viral mRNA and

strand-separated

XhoI-abundance. The data of Fig. 1A, then, indicate

HindIll

fragment

C'-IO DNA (0.171-0.182map that the 2.3-kb

(p)

mRNAis the most abundant units)5' endlabeled at theXhoI siteyieldedthe mRNA in thisregion,whereas the1.9-kbmRNA samethreefragments (Fig.

2B,

track S). These is somewhat less so and the 3.9- and 4.5-kb mapping data,theclosecorrelation ofthesizeof

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.490.54.241.71.158.2]

1900 n

4500n 3900 n

2300n

Bam HI BglI XhoI HindM EcoRI Sol!

-i

0.152 0.163 0.171 0.182 0.190

0.164 0195

II A. .. .. j . In

A.0.152-0.171* B. 0.17i*-D0182

S D M D XM2 M D S M S X D M

1600-3 _167V,_{_.,.;}50- _ 165C

40 -- 70Cr' --_-'250-2K

--6r,--- ~2?C

-287 _

28zt5U f

270-280

-FIG. 2. Localization of the mRNAs by Si nuclease and exonuclease VII mapping. Single-strandedDNA5'or

3'labeledatspecific restriction siteswashybridized with HSV poly(A) mRNA, and the DNA protected from

nuclease digestion was size fractionated ondenaturing acrylamide gels asdescribed previously (13, 15). (A)

BamHI-XhoI fragment A-G DNA (0.1524.171 mapunits) 3'labeledatthe XhoI site.S, S1 nuclease-resistant

material, D,Undigested DNA,M,Size marker of Hinfi-digested pBR322 DNA; sizesare(in nucleotides from

bottom): 220/221, 298, 345, 396, 506/517, 600, and 999. (B) XhoI-HindIII fragment C'-IO DNA (0.1714.182map

units) 5' labeledattheXhoI site. D, Undigested DNA. X, Exonuclease VII-digested material. M2, Marker of

HaeII-digested pBR322 DNA; sizesare(in nucleotides frombottom): 227, 280 (partial digestion fragment), 370,

430/439, 622,and1876. S, S1 nuclease-digested material. M, Sizemarkersasdescribed for(A). (C) HindIII-SaII

fragment J-DDNA(0.1824.195mapunits) 5' labeledatthe HindlIl site.M,Size markerasdescribed for(A);S,

S1-digested material; X, exonuclease VII-digested material; D, undigestedDNA.

the Si-resistant hybrid (ca. 1,880 nucleotides) compared with the total size of the mRNA (2to

2.1 kb), and the lack of hybridization of DNA probestotheright of the DdeI sitetothe 1.9-kb mRNAall stronglysuggestthat this mRNA has

a distinct 5' end. Certainly any intron must be less than50 bases longin total length and must

be unableto hybridize efficientlywiththeDNA

underourstandard conditions.

The 5' endsof the generally colinear 3.9- and 4.5-kb mRNAs were located by usingas a

hy-bridization probe strand-separated HindIII-SalI fragment J-D (0.182-0.195 map units) 5' end labeled at the HindIII site at map unit 0.182. Here late viral mRNAprotected fragments 650 and1,250 baseslong from both S1 nuclease and exonuclease VIIdigestion (Fig. 2C, tracks S and X). Thus, the 5' ends of these mRNAsmap650

and1,250 basestotheright of the HindIlI siteat mapunit0.182. The fact that the ratio of

radioac-tivities in the two bands is similar with both nucleasedigestion regimens suggested that each mRNA has itsowndiscrete 5' end.

Furthercharacterization of the regions

contain-ing the 5' end ofthe 2.3-kb ,B and 1.9-kb I-y

mRNAs.The 1,000-base nucleotidesequence of

the DNAfromca.30bases 3' of the XhoI siteat map unit 0.171 tojust beyond the AvaI site at

0.177 is shown in Fig. 3. A summary of the results of the following experiments and the

sequencestrategy areshown in Fig. 4.

Weprecisely located the 5' end of the 2.3-kb mRNA by doing nuclease protection

experi-ments and fractionating the mRNA-protected DNA fragmenton a DNA sequencing gel. This

procedure has been documented in several

pre-vious publications (8, 12, 13). It was done by

hybridizing HSV poly(A) mRNA with single-stranded DNA, spanning the twoHinfl sites at

nucleotides 625to780, whichwas5' end labeled at nucleotide 625. Hybrids were digested with mungbeannuclease. Protected DNA was frac-tionated against a sequence ladder ofDNA 5'

labeled at the same Hinfl site (Fig. SA). Note

that the sequence ladder is of the DNA strand

complementarytothe mRNA. The5'endof the

mRNA fell within the sequence GTATC, and

thus the mRNA begins within the region be-tweennucleotides 273 and 277.

The 5'end of this mRNA fellca.28to30bases

3' tothe sequence TATAAATTA, anexcellent

TATA box. The 120or so bases5' upstream of

this 5'end share several features withpromoters

for other HSV-1 P mRNAs (reviewed in

refer-ence 35). These include a TATA box about 30

bases upstream of the transcript and an AC

C.0.182*-0195

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.490.105.390.71.267.2]

1 0 20 30 40 50 60 70 80

CCGCGAGTCCACCATACCCGCCTCCTCGAGGACCACGGCCAGGGAACACAGATiTCCAGGCGGGCCCAGAGGGGACCGA GGCGCTCAGGTGGTATGGGCGGAGGAGCTCCTGGTGCCGGTCCCTTGTGTCTATTAGGTCCGCCCGGGTCTCCCCTGGCT

Xhol

90 100 110 120 130 140 150 160

TGGCCACAGGGGCGCGGACGCCGCGCAGCAACCCGCGCAGGTGGCGCTCGAACGTCTCGGCTAGTATATGGGAGGGCAGC

ACCGGTCTCCCCGCGCCTGCGGCGCGTCGTTGGGCGCGTCCACCGCGAGCTTGCAGAGCCGATCATATACCCTCCCGTCG

170 180 190 200 210 220 230 240

GCGTTGGGGATCACCGACGCCGACCACATAGAGTCAAGGTCCGGGGAGTCGGGATCGGCGTCCGGGTCGCGGGCGTGGGT CGCAACCCCTAGTGGCTGCGGCTGGTGTATCTCAGTTCCAGGCCCCTCAGC:CCTAGCCCGCAGGCCC:GCGCCCGCACCCA

250 260 270 280 290 300 310 320

GCCCCCAGGAGATAGCGGAATGTCCGGGGTCGGAGGCCCGGAGGCGTCACAAAGTGCCGGCCGACGCGGCCCGGGGCTTTT CGGGGCGTC.CTCTATCGCCTTACAGGCCCCAGCCTCCGGG.CCTCCGCAGTCTTTCACGGCCGCTGCGCCgGGCCCCGAAA

5 -1.9kb*'

330 340 350 360 370 3C0 390 400

CGTCTGCGGTGTCGGTGGCGTGCTGATCACGTGGGGGGTTAACGGGC GA6iTGGGAGCTCGGGTCCAChGCTGACCGTCGTC

GCAGACGCCACAGCCAC CGCACGACTAGTGCACCCCCCAATTGCCCGCTTACCCTCGAGC:CCAGGTGTC GECTGC'AGCAG Aval

410 420 430 440 450 460 470 480

TG0GGG0TGGGGGGGCAGGGGAC0GGAAGGTGGTTGTCAGCGCAAGACTGTTG3GGCCGGGGGCGCTTGGGGGGGCTGTCGGG

ACCCCACCCCCCCCGTCCCCTGCCTTCCACCAACAGTCGCCTTCTGACAATCCCGCCCCCGCGAACCCCCCCGACAGCCC

490 500 510 520 530C 540 550 560

GCCACGAGGGGTGTCCTCGGCCAGGGCCCAGGAACGCTTAGTCACGGTCCOTCCCGG3GGACATGCTGGW3C-CTCCCGTG G

COGGTGCTCCCC:ACAGGAGCCGGTCCCGGGTC.CTTGCGAATCAGTGCCACGC.AGGGCC.TCCTGTCCGACCCCGGAGGGCACC

Ddel

570 580 590 600 610 620 630 640

ACTCCATTTCCGAGACGACGTGGGGGA0GC GGTGGTTGAGCGCGCCGCCGGGTGAACGCTGATTCTCACGACAGCGCGTG TGAGGTAA.GGCTCTGCTGCACCCCCCTCGCCACCAACTCGCGCGGCGGCCCACTTGCGACTAAGAGTGCTGTCGCGCAC

:4- Hinf I

650 660 670 680 690 700 710 720

CCGC:GCGCACGGGTTGGTGTGACACAGGCGGGACACCAGCACCAGGAGAGGCTTAAGCTCGGG3AGGC6GCGCCACCGACG

GGCGCGCGTGCCCAACCACACTGTGTCCGCCCTGTGGTCGTGGTCCTCTCCGAATTCGAGCCCTC CGTTCCGGTGGCTGC Ava I

730 740 750 760 770 780 790 800

ACAGTATCGCCTTGTGT6TGTGCTGGTAATTTATACACCGATCCGTAA3C GCGCGCCGAATCTTGGCiATTGCGGAGGTGG TGTCATAGCGGAACACACACACGACCATTAAATATGTGGCTAGGCATTTGCGCGCGGCTTA0AACCCTAACGCCTCCACC

°5- 2.3kb HinfI

810 020 830 840 850 860 870 880

CGCCGGATGCCCTCTGGGACGTCATACGCCAGGCCGTGGGTGTTGGTCTCGGCCGAGTTGACAAACAGGGCTGGGTGCAG GCGGCCTACGGGAGACCCTGCAGTATGCGGTCCGGCACCCACAACCAGAGCCGGCTCAACTGTTTGTCCCGACCCACGTC

890 900 910 920 930 940 950 960

CACGCAGCGATAGCGCAGCAGGGCCAGGGCGAAGTCC0TCGACAGCTGGTGTTTT0ATACTGGTAACCGG CCGGG GTGCGTCGCTATCGCGTCGTCCCGGTCCCGCTTCAGGCAGCTGTCGACCAACAACTTTATGACCATTGGCCCTTTGOCCC

970 980 990 1000

TCACGGGTACGCCCAGGCTCGGGGCOACGTACACGCTAAC AGTGCCCATGCGGGTCCGAGCCCCGCTGCATGTGCGATTG

FIG. 3. Nucleotide sequence ofDNA betweentheXhoIsite at map unit 0.171 and the AvaIsite at 0.177.

Sequencing

wasbythe method of Maxam andGilbert(23), and the strategy is shown in Fig. 4. The locations of

the 2.3- and 1.9-kb mRNA5'ends

(*-O),

in-phase translation initiators( 3),and several other features of the

sequenceareindicated.

0 (100) 200 300 400 500 600 700 800 900 1000

Hinf I AvaI DdeI AvaI Pvull

Xhol HinfI Smal Pvull HinflHinfl Hinfl AvaI

-, -....

-2.3

kb

J13)

mRNA 3.9 kb

(13)1

mRNA 4.5kb

(IY)

mRNA

1.9kb

(or)1

mRNA Translation Frames

-3 -3

-3-

9:43

IK 1

FIG. 4. Sequencing strategyand summary. High-resolution restriction maps forXhol, Hinfl,

Sinal,

Aval, Pvull,andDdel sites areshown, aswellasthe lengthofsequence ladders from DNA 5' labeledat suchsites

Thelocations ofthe 5' ends of the 2.3-kb(p)and the 1.9 kb

(P3y)

mRNAsand of thetranslationalstart

E)andstop (Z)codonsare shown.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.490.106.391.69.376.2] [image:6.490.110.391.470.625.2]

HSV mRNA 597

S~~~~~~~~~~~~~~~~~~~~~~~~~~L , .._

.... 3D. #;.

-_

.

FIG. 5. (A)Precise localization of the

2.3-kb alkaline exonuclease mRNA. I

mRNA washybridized with single-strar

end labeledattheHinfl siteatnucleotid

andextendingtotheHinfl siteatnucleo

was then digested with mung beannucli

protected DNA wasfractionated agains

ladderof DNA 5' labeledatnucleotide 62

localization of the 5' end of the 1.9-kbr

spanning from the XhoI siteatnucleotide

the Ddel siteatnucleotide 520 was 5'e

base 28 and strandseparated. Itwashy

viral mRNA, thehybridsweredigestedv

ase, andprotected DNAwasfractionati

turingacrylamide gel, usingHaeII-cut I

fragments as asize standard.

string110 to 120bases upstream. T son of the 120 bases upstream o characterized ,B mRNAsis shownin comparison gives us reasonable cc

concluding

thatthis regionupstream4

of the2.3-kb mRNA is, indeed, a, ]

-120

EARLY (BETA) HSV PROMOTORS

-90

The sequence data in Fig. 3wereanalyzedfor potential translational initiation and termination codons. The 5' end of the 2.3-kb (1) mRNA is 160bases upstream of the canonical eucaryotic translation initiation sequence AAATGG (nucle-otides 432 to 437), as described by Kozak (18). This translation frame (frame 1 of Fig. 4) stays openthroughout the region sequenced and con-tainsmethionine codons 379, 553, and 643 bases downstream. It is a reasonable assumption that this frame defines the N-terminal region of a protein. A second ATG codon isseeninreading frame 3 at position 457-459, but this frame is terminated24 bases downstream with the TAA 5' end of the triplet atposition 481-483.

HSV poly(A) All reading frames contain multiple

termina-nded

DNA 5 tors upstream ofthepotential translation frame le622(Fig. 3) for the 2.3-kbmRNA, so anyprotein encoded by

lease

and

the

the 3.9- or4.5-kb mRNAs would notshare any

st

a

sequence

translational frames with a

protein

encoded

by

2.(B) Precise the 2.3-kb mRNA.

mRNA. DNA We hybridized single-stranded DNA, 5' end 28(Fig. 3) to labeled at the XhoI site (base 28; Fig. 3) and nd labeled at extending to the DdeI site at base 520, with viral 'bridized with mRNA and

carried

out Si nuclease analysis.

with

S1

nucle- Twoprotected specieswere seen. One migrated

ed on adena- with asize of 290 bases, and one migrated along

pBR322

DNA with undigested DNA (Fig.

SB).

The shorter DNA species corresponded to the 270- to 280-base bandseenin Fig.2B (tracks S andX). The full-length band indicated that no other major 'he compari- species was seen in the region extending up to

f five well- 520 basesupstream of the XhoI site at 0.271. ^Fig.6.This Control experiments (not shown) indicated )nfidence in that S1 nuclease-protected DNA was entirely of the 5' end dependent on the presence of viral mRNA in the promoter. hybridization. We precisely located the mRNA

-60 -30

Alkaline Exonuclease (2.5kb)

Transcribed from rightto left with 5' end at 0.175

AGACCAACACCCACGGCCTGGCGTATGACG TCCCAGAGGGCATCCGGTGCCACCTCCGCA ATCCCAAGATTCGGCGCGCGTTTACGGATC GGTGTATAAATTACCAGCACACACACAAGA

Thymidine kinase-type 1 (1.5 kb)

Transcribedright toleft with5' end at 0.315

CTATGATGACACAAACCCCGCCCAGCGTCT TGTCATTGGCGAATTCGAACAGGCAGATGC AGTCGGGGCGGCGCGGTCCGAGGTCCACTT CGCATATTAAGGTGACGCGTGTGGCCTCGA

5.2 kb mRNAencodingHSV-1 140k protein Transcribed left to rightwith 5' end at0.565

CATGGAAGGAACACACCCCCGTGACTCAGG ACATCGGCGTGTCCTTTTGGGTTTCACTGA AACTGGTCCGCGCCCCACCGCTGCGCGATG TGGATAAAAAGCCAGCGCGGGTGGTTTGGG

1.5 kbmRNAencoding HSV-1 38k protein

Transcribed left torightwith 5' endat0.590 CGACCATAGCCAATCCATGACCCTGTATG TCACGGAGAAGGCGGACGGGACCCTCCCAG

1.5kb mRNA encoding 39k protein

Transcribed left to rightwith 5' end at 0.699

;CATATAAGCGCGGACTAAAAACAGGGATGT

CCC CTCACCCCACACAGGGCGGGTTCAG GCCTGCCCGGCAGCCAGTAGCCTCTGGCAGiATCTGACAGAC(iTGTGCGATAATACACACG CCCATCGAGGCCATGCCTACATAAAAGGGC

FIG. 6. Sequence comparison for the promoters of HSV-1 ,B mRNAs. The 2.3-kb mRNA promoter region

from the data of Fig. 3 is shown compared with data for other HSV-11 mRNAs. The data for HSV-1 thymidine

kinasearefrom McKnight (24), those for the 5.2-kb mRNA are fromFrink et al. (13), those for the 1.2-kb mRNA

arefrom Draper et al. (8), and those for the 1.3-kb mRNA are from Hall et al. (15).

VOL.48, 1983

7

c-.

,,:-'L

,.4--r - ;%

I " "

i

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.490.61.229.73.205.2]

598

end (seen above) to around base 320 by using high-resolution sequencing gels against se-quence ladders (not shown). This putative cap site does not lie 25 to 30 bases downstream of any exceptionally appealing TATA box se-quence. However, the sequence between bases 350 and 345 (TGATC) could conceivably serve as an equivalent feature. The cap site does lie 131 bases to the right of the ATG triplet at position 187-189. This could serve as a transla-tional initiator. If so, the polypeptide encoded should share the same reading frame as that encoded by the 2.3-kb mRNA.

Evidence that the 2.3-kb earlymRNA encodes

alkalineexonuclease. We used a monoclonal anti-body against HSV-2 alkaline exonuclease to demonstratethat the 2.3-kb mRNA encodes the cross-reactive HSV-1 protein. Denaturing gel

electrophoresis showed that this antibody (Q1) specifically binds to purified '251-labeled HSV-2

alkaline exonuclease (82,000 molecular weight) (Fig. 7, track Q1). Another monoclonal antibody (T2/Ti), which is directed against other specific

sitesonthe HSV-2 enzyme, did not precipitate it

(Fig. 7A, track

T2/T1).

This demonstrated the specificity of the Ql precipitation. Ql reacted

with the 82,000-d HSV-1 enzyme from

[35S]methionine-labeled, HSV-1-infected cell extracts (Fig. 7A, track Ql-I.P.). Other faint protein bands were also visualized. No specific protein band could be seen when Ql was reacted with uninfected cell extract (data not shown). Therefore, it is not known whether these

repre-sent nonspecific binding or binding of antigeni-cally related, infected-cell proteins orboth.

Wepurified mRNA encoded in this region of HSV-1 DNA by using BamHI-HindIll fragment A-IO (0.152-0.182 map units) and XhoI-HindIII fragment C'-IO (0.171-0.182 map units) bound

to cellulose. Both fragments hybridized to all mRNAs (see Fig. 1), but the former fragment also encodes the 3' region of another 2-kb mRNAwhose 5' end maps near 0.147 and whose 3' end maps near 0.160 (data not shown). We subjected a sample of purified mRNA (ca. 0.1 p.g

oftotal mRNA) to in vitro translation by using

[35S]methionine

and incubated the total transla-tion mix with Ql. Reactive protein was size fractionated on a denaturing SDS-acrylamide gel (Fig. 7B, tracks C'-IO and A-IO). Both mRNA preparations translated the same 82,000-d poly-peptide. In vitro translation of HSV-1 mRNA purified by using DNA fragments from other regions of the genome did not yield any detect-able protein reactive with

Qi

antibody (data not shown).

Inanother experiment,wepurified mRNA by usingXhoI-HindIIIfragment C'-IO (0.171-0.182 mapunits) andtranslated it invitro. Translation products were reacted with Ql antibody. This

FIG. 7. In vitro translation and identification of

HSV-1 alkaline exonuclease. All autoradiographs are

ofdenaturing SDS-acrylamide gels(19). Sizes of

pro-teins (X103 d) were determined by comigration with

adenovirus mRNA translation products (not always

shown) asdescribedpreviously(1). (A)Ql,

Immuno-precipitation of125I-labeled HSV-2 alkaline

exonucle-ase with monoclonal antibody Ql.

T,/Tj,

Lack of

precipitation with monoclonal antibody TJ/T1. En-zyme, Authentic enzyme. Ql-I.P.,

Immunoprecipita-tion ofcytoplasmic, 35S-labeled, infected-cell protein

with monoclonal antibody Ql. I.P., Cytoplasmic

in-fected-cell protein. (B)Immunoreactivities of in vitro

translationproducts.Cl10-Ql,Productsoftranslation

of Xhol-Hindlll fragment C'-IO DNA (0.171-0.182

map units) specific mRNA immunoprecipitated with

Ql. AIO-Ql, The same immunoprecipitation, except

thatBamHI-HindlllfragmentA-10DNA(0.152-0.182

map units) specific mRNA was used for the in vitro

translation.Ad, In vitro translationproductsof

adeno-virusmRNA. (C) ComigrationofHSV-2 enzymeand

HSV-1 in vitro translation product. HSV, HSV-1

poly(A) mRNA (1

p.g)

in vitro translation products. Enzyme-Ql, 125I-labeled HSV-2 alkaline exonuclease

precipitated with Ql. Cl10-Ql, In vitro translation

productofXhol-Hindlllfragment C'-4ODNA

(0.171-0.182mapunits)specificmRNAprecipitatedwithQl.

(D) Translation of size-fractionated HSV poly(A)

mRNAandprecipitation withQl. Sizes ofmRNAare

indicated above the appropriate tracks. I.p.,

35S-la-beled, cytoplasmicinfected-cell protein.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:8.490.256.442.98.365.2]

HSV mRNA 599

material was then subjected to denaturing gel electrophoresis along withpurified, 125I-labeled HSV-2 alkaline exonuclease which had also beenimmunoprecipitatedwith Ql antibody (Fig. 7C). This experiment confirmed the essential comigration of the HSV-1 in vitro translation product with authentic HSV-2 alkaline exonu-clease.

The experiments described above established thefact that oneof the fourmRNAs located in the region between 0.160 and 0.190 map units encodes HSV-1 alkaline exonuclease. We trans-lated size-fractionated total HSV-1 mRNA and used Ql antibodytoobtainaprecise size for the mRNAencoding theenzyme. We size fraction-ated 10,ugof HSV poly(A)mRNA on a denatur-ingagarose gel. Four mRNA pools were taken, migrating with sizes between 5.2 and 4 kb, 3.7 and 3.1 kb, 2.9 and 2.5 kb, and 2.3 and 2 kb. Each was eluted, reselected on oligodeoxythy-midylic acid-cellulose, and subjected to in vitro translation. The translation products were then reactedwith Ql antibody. It was clear(Fig.7D) that major synthesis of the 82,000-d alkaline exonuclease is mediated by mRNA in the size range of2.9 to 2.5 kb. These data demonstrate thatthe 2.3-kb ,B mRNA does, indeed, encode the enzyme. [Note: as in Fig. 1, this mRNA migrates with a rate corresponding to a size of 2.5kbbecause of the poly(A) tail.] The radioac-tivity migratingat-50,000 d in each track is an endogenousband from the reticulocyte system. Its presence here is duetothefact thatweused IgGsorb instead of Pro-A Sepharose to react

with the immune complex because of the

former'sgreaterabilitytobindtoimmune com-plexes.

Identificationof a potential translation product forthe 1.9-kb mRNA.Partialsequencedata(Fig. 3)indicated that the 1.9-kbmRNAcolinear with the 3'region ofthe2.3-kbmRNAcould encodea polypeptide about 130 amino acids shorter than

the alkaline exonuclease. We used

BamHI-HindIlI fragment

A-TO

(0.152-0.182 mapunits) bound to cellulose to isolate all the mRNAs encoded in the region of interest. We size frac-tionatedtheseregion-specificmRNAs and trans-lated both the 2.3-kb and 1.9-kb mRNAs in vitro. We used methods described in detail in several recentpublications (see reference 13) for the size fractionation of the hybrid-selected mRNA.Translation of the 2.3-kb mRNA yielded anumberofpolypeptides, butonly the 82,000-d alkaline exonuclease was reactive with Ql anti-body (Fig. 8,tracks T-1 and Q1-1).

The mRNA isolated as the 1.9-kb species actuallycontains both this mRNAand a 1.9-kb mRNA whichis homologous to the other DNA strand and has a 3' region homologous to the region between 0.152 and 0.160 map units (see

P'>

mF;;-w23bc- r

.s _

z

f7 A

"._

-.

FIG. 8. In vitro translation of the 1.9-kb mRNA.

The 2.3-kb and 1.9-kb hybrids, selected by using

BamHI-HindIII fragment A-IO (0.152-0.182 map units) bound to cellulose, were fractionated by

electro-phoresis on a denaturing agarose gelcontaining

meth-ylmercury hydroxide. Each was translated in vitro;

halfof each translation product was fractionated

with-outfurther treatment (tracks T), andhalf was reacted

with Ql monoclonal antibody (tracks Q1). Details of the translation and fractionation of products are as

described for Fig. 7. A track of the translationproduct

of rabbit globin mRNA is includedas acontrol.

the preceding section). Translation of this mRNA mixture consistently gave two polypep-tide products distinct from endogenous bands found in the translation system. The larger mi-grated witha rate

corresponding

toamolecular weight of60,000, and the smallermigratedwitha

rate correspondingto54,000(Fig. 8,trackT-2). We suggest that oneof these

polypeptides

is the translationproduct of the 1.9-kbmRNA in ques-tion. Since, however, neither product reacted withQl antibody (Fig.8,trackQ1-2),our identi-fication of a translation product for the 1.9-kb mRNAdescribed in thisreportisonlytentative. Tentative evidence that a50,000-d HSV-1 cap-sidprotein is encoded by either the 3.9- or 4.5-kb mRNA. We used apolyclonal rabbit antiserum made against the SDS gel-purified 50,000-d HSV-1 capsid protein (VP19C) to tentatively identify itas atranslationproduct of the 3.9- or 4.5-kb mRNA.This antiserum(NC-2) specifical-lyreactswitha50,000-d protein from [35S]meth-ionine-labeled extracts from HSV-1-infected cells. Adenaturingacrylamide gelsize fraction-ationofinfected-cell proteinreactive with NC-2 is shown in Fig. 9A (track NC2-I.P.). Some other proteins are also detectable, especially with long exposures (ca. 2 weeks). Aprominent oneis the155,000-dcapsid protein, ICP5,which may have been a contaminant of the original size-fractionated protein antigen.

We prepared purified

region-specific

viral

mRNA, using BamHI-HindIII fragment A-TO

VOL.48, 1983

.-',i

el"..,"

a o.,;,_tD.1..-.-n!'il. -

-.4

.:7, 4-ll.- -j

on November 10, 2019 by guest

http://jvi.asm.org/

[image:9.490.289.410.65.220.2]

[image:10.490.48.242.54.189.2]

FIG. 9. Immunoprecipitationofa50,000-d protein with antiserum NC-2. Details are asfor Fig. 7. (A)

Comigration ofproteinprecipitablefromcytoplasmic

infected-cell proteinand in vitro translation product.

I.P., 35S-labeled cytoplasmic infected-cell protein.

NC2-I.P., Cytoplasmicinfected-cellprotein

precipitat-ed withNC-2 antiserum. NC2-A-IO, In vitro

transla-tion products of BamHI-HindIIIfragmentA-IO DNA

(0.152-0.182 map units) specific mRNA precipitated

with NC-2 antiserum. (B) Antiserum NC-2

precipita-tionofa50,000-dtranslation productofHindIII-SaIl

fragment J-D (0.182-0.195mapunits) specificmRNA.

I.P., Cytoplasmic infected-cell protein. NC-2 No

RNA, Material precipitable with NC-2 and IgGsorb

fromthetranslationproductsresultingfromnoadded

RNA.NC-2HS-JD, Proteinsprecipitated usingNC-2

andIgGsorb fromthe translation products of

HindIll-SallfragmentJ-Dspecific mRNA. The bandmigrating

at -65,000d(?)wasnotidentified.

(0.152-0.182 map units) and XhoI-HindIII frag-mentC'-IO(0.171-0.182mapunits) DNA bound tocelluloseasdescribed in the previous section.

We used this mRNA asa template for in vitro

translation and used the NC-2 antiserum as a reagent to detect the 50,000-d capsid protein. Both fragments consistently gave positive

re-sults. However, the amount ofradioactivity in the 50,000-d protein isolated was considerably

lessthanthat in the82,000-d protein isolated by using Ql antibody against alkaline exonuclease. This conclusion was based on the fact that

autoradiographs required three to five times

moreexposuretoobtainequivalentexposuresof

the 50,000-d protein compared to the 82,000-d

one. Atypical experimentusing BamHI-HindIII

fragment A-IO (0.152-0.182mapunits) DNAfor

hybrid selection is represented in Fig. 9A (track NC2-AIO).

Thebandofradioactivitymigratingat82,000 d is alkaline exonuclease, which is also translated fromthis mixedmRNApreparation (seeabove). We could notdetermine whether this ability of NC-2 antiserum to react with this enzyme was

due to specific immunological cross-reactivity.

Long exposures of immunoprecipitates of

35S-labeled infected-cell protein extracts, such as

shown in Fig. 9, did reveal a band of protein

migrating witha rate correspondingto 82,000 d in size. However, nonimmune rabbit serum

boundpurified

251I-labeled

alkaline exonuclease unless

large

amounts of carrier protein were

present. Therefore, the precipitation ofthe

en-zymeseen withtheinvitrotranslation couldbe nonspecific binding. Ql did not appear toreact

with the50,000-dproteinsynthesizedin vitroas

judged byverylongexposures(3months)of the autoradiographs seen inFig. 3.

Therelativespecificityof theNC-2

antiserum,

aswell as thecomigration of immunoprecipitat-ed infectimmunoprecipitat-ed-cell 50,000-d capsid protein and the in vitro translation product (compare Fig.

9A,

tracks NC2-I.P. and NC2-AIO), suggested that the mRNA for thisproteiniseitherthe3.9-kbor

the 4.5-kb

(P3y)

mRNA mapped in Fig. 2. This identification must be regarded as tentative, however, since the amount ofradioactivity iso-lated in the 50,000-d translation product preclud-ed tryptic peptide comparison with thepurified capsid protein. The relatively small amount of protein recovered from the in vitro translation product could be due to any combination of three factors. First, total translation product contains arelatively small amount of the 50,000-dprotein, equivalent to endogenous products of translation (data not shown). This reflects the fact that neither of the two mRNAs is highly abundant. Second, the methionine content of the proteinis unknown and may be low. Third, the NC-2 antiserum was made against denatured 50,000-d capsid protein and may not react effi-ciently with protein translated in vitro.

We confirmed our inference that either the 3.9-kb or4.5-kb

(B-y)

mRNAencodeda50,000-d

polypeptide in two ways. First, we used

HindIII-Sall fragment J-D (0.182-0.192 map units) DNA bound tocellulose topurifythe 3.9-kb and 4.5-kb mRNAs away from the 2.3-kb mRNA for alkaline exonuclease (see Fig. 1). Suchhybrid purifiedmRNA, when translated in vitro, yielded the 50,000-d protein immunoreac-tive with NC-2 antiserum (Fig. 4B). Here, the use of IgGsorb led to the recovery of other bands due to endogenous translation products, but thepresence of the 50,000-d protein specific for the viral mRNA is clear. Second, we used size-fractionated total HSV poly(A) mRNA to translatethisprotein. Thisexperiment(data not shown) wasdone as describedin the preceding section, except that mRNA ranging from a size of 5.0 to 4.5 kb wastranslated and then reacted with NC-2 antiserum. Here otherprotein bands were also seen, but the 50,000-d protein was clearly translated by mRNA of the size range between 5 and 4 kb. Larger and smaller size fractionsdid not translate thisprotein.

DISCUSSION

The moderate-resolution transcription map presented inFig.2isentirelyconsistent with the

VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

HSV mRNA 601

picture of HSV-1 gene expression and gene packaging developed in the last several years (reviewed in reference 35). The fact that the mRNAs characterized can be correlated with specific viral genes demonstrates, again, the genetic resolution available through HSV tran-scription mapping.Weshouldnote attheoutset, however, that none ofthe datapresented here precludes the presence of small amounts of othertranscripts in this

region

being

generated via splicing from other promoters orby use of otherpolyadenylation sites.

Ourconfidence in the identification of the 2.3-kba mRNA asencoding alkaline exonuclease is quite high. The immunereactivity with the high-ly specific monoclonal antibody Ql is, itself, strong evidence, especially when migration of theinvitro translation product is compared with that ofauthentic enzyme and with that seenin the infected cell (Fig. 7). The fact that Preston andCordingley(29)also located thegenefor this important enzyme in this region adds to the alreadystrong case. It should benoted, howev-er, that these workers estimated themRNAfor the enzyme to be in the size range of 3.6 kb, considerably greaterthan that seen here. Their result could be due to the procedures used to fractionate the RNA (sucrose

gradients).

It could bedue,however,tothe

amphibian

oocyte system they used for translation

being

able to initiate translation in the interior of the partially colinear 3.9-and 4.5-kb mRNAs. Basedon the present data, we conclude that these mRNAs contain all the information for the alkaline

exo-nucleasedownstream of their translational read-ing frames. Such patterns of long untranslated regions ofsomelarge HSV-1 mRNAs have been described byus

previously.

Our confidence in the conclusion that the 50,000-d

capsid

protein

VP19C is encoded

by

either the 3.9- or4.5-kb mRNA is not as high. Certainly, these mRNAs encode a protein mi-grating at 50,000 d, but the relatively weak reaction between the NC-2 antiserum and thein vitro translation products must indicate some caution. It is conceivable (although

unlikely)

that the

precipitation

of the translation product by this antiserum is nonspecific. The fact that the capsid protein has been mapped in this general location (20)appears to make this possi-bilityevenlesslikely.

Thepartial sequence analysis presented here, and further sequence analysis currently under way in this laboratory, are useful in defining potential promoters and translational reading frames. Ourlocation of the

5'

end of the 2.3-kb alkaline exonuclease mRNA is quite solid and places thismRNAimmediately downstreamof a 120-basesequence containing features of HSV-1 1 promoters. We first noted the similarity

be-tweenanother,Bpromoter(forthe5.2-kbmRNA mapping in HindIII fragment K) and that for HSV-1 tk several years ago (12). Subsequent studies have tended to confirm thegeneral fea-turesnoted. These studies have beenreviewed, but the data of Fig. 6 are the most complete comparison available at this time.

It is clear from the analysis ofthe potential reading framessummarized in Fig.4that thereis noapparentsharedreading framefor the 82,000-dalkalineexonuclease and the 50,000-dVP19C. This isconsistent with results from other regions of the genome containing partially colinear mRNAs. We cannot be absolutely certain that translation frame1of Fig. 4 doesindeed encode the alkaline exonuclease, nor that frame 3 (which initiates further upstream) encodes VP19C. We arecertain, however,ofourfinding oftranslational stop signals in all three frames before the startsignal of frame 1. Furthermore, sequence data not shown indicate that frame 1 stays open for at least several hundred bases downstream of the data shown and that frame 3 staysopenforatleastthesame amountofbases upstream.

Given the size of 2.3 kb as the actual tran-script size for alkaline exonuclease and transla-tion frame 1 as its properframe, we can infer that the total mRNA coding capacity is on the order of2,100 nucleotides. Such would encodea protein of 700 amino acids, somewhat small for the measured size of alkaline exonuclease (82,000d). However, thereare37prolines inthe first 236 positions in the predicted amino acid sequence. Suchahigh prolinecontent(ca. 15%) would lead to a protein migrating

.more

slowly than predicted from its amino acid residue weight. We havecitedthisas the reasonforthe discrepancy between the predicted amino acid residue molecular weight of HSV-1 gC and its migration rate (60,000dversus 69,000 d;13).

Partial sequence data (not presented here) around the 5'ends ofthe 3.9- and 4.5-kbmRNAs suggest that the larger of the two mRNAs en-codesashortreading frame which terminates in the leader of the 3.9-kb mRNA. Furthermore, datasuggest that the reading frame seen in the 3.9-kbmRNAis open forasignificant stretchof nucleotides. Such data indicate that the 3.9-kb mRNA,then, actually encodes the 50,000-d pro-tein that we have suggested is VP19C. This situation wouldbesimilartothatseenbyHall et al. (15) for two partially colinear and comple-mentaryoverlappingmRNAsin theregionabout map unit 0.7.

The mostdisturbingaspect of ourconclusion that the 50,000-d translation productis actually VP19C is the lowabundanceof the mRNA. We should state, however, that the amount of

[35S]methionine-labeled

VP19C frompurified

vi-VOL.48, 1983

on November 10, 2019 by guest

http://jvi.asm.org/

rions is very much less than that of VP5 (5). Also, we have found that the total amount of VP19Crecoverablefrominfected-cellextracts is very much less than that of VP5 by using im-mune precipitation (Costa and Wagner, unpub-lisheddata). Therefore, theamountof VP19C in theinfected cell maybe low and the abundance of the mRNA

encoding

itmay also be low.

A striking result in terms of HSV-1 mRNA expression ofourpresent study is the

finding

of a smaller mRNA underlying the 3' three-quar-tersof the 2.3-kbmRNAandapparentlysharing atranslational reading frame with it. Other ex-amples of shared translational reading frames arefound in the HSV-1thymidine kinasegene as shownby Marsdenetal. (22)and,potentially, in the gCgene (13).

The sequence data suggest that any protein encodedby the 1.9-kbmRNAshould contain the 570C-terminal amino acids of alkaline exonucle-ase. Thisprediction is certainly consistent with the size for either in vitro translation product seen in Fig. 8. We can assume that the type-commonepitoperecognized by the Ql antibody is dependent upon the N-terminal 130 amino acids of the alkaline exonuclease. Theproperties andpossiblebiological function ofanytruncated form of the alkaline exonuclease which would lead to its being required at late times after infection are unknown. Indeed, it should be

emphasized

that such a

protein

has not been describedas

being

presentin normal

lytic

infec-tionby HSV. Furthersequenceanalysis,aswell

asfurthermRNAand

protein

fractionation pro-cedures

currently

being

undertaken in our

vari-ous

laboratories,

may lead to more

specific

predictions

regarding

thisputative protein. The datadosuggest,however, that thepartial

colin-earity

ofmRNAsunderindependent temporalor promoter control can be a result of related features of theproteins encoded.

ACKNOWLEDGMENTS

We thank L. Hall, M. Rice, and R. Frinkfor help and discussion.

Workwas supportedby Public Health Service grant CA-11861from the National Cancer Institute andgrantMV-159 from the American CancerSociety toE.W. and by Public

HealthServicegrantsDE-02623(NationalInstitute of Dental

Research) and AI-18289 (National Institute of Allergy and InfectiousDisease)toR.E.and G.C. R.C. isatrainee under Public Health Service Molecular and CellularBiology training

grantGM-07311.

LITERATURECITED

1. Anderson,K.P.,R.J.Frink.G.B.Devi, B. H.Gaylord,

R. H.Costa,andE.K.Wagner.1981. Detailed

character-ization of the mRNA mapping theHindlIl fragment K regionof theherpessimplexvirus type1genome.J. Virol. 37:1011-1027.

2. Anderson, K.P., J. Stringer, L. Holland, and E.Wagner. 1979. Isolation and localization ofherpes simplex virus type 1 mRNA. J. Virol. 30:805-820.

3. Bailey, J. M., and N. Davidson. 1976.Methylmercury asa reversible denaturing agent for agarose gel electrophore-sis.Anal. Biochem. 70:75-85.

4. Berk, A.J., and P. A. Sharp. 1977. Sizing and mapping of early adenovirus mRNAs by gel electrophoresis ofS1 endonuclease-digested hybrids. Cell 12:721-732. 5. Cohen, G. H., M. Ponce de Leon, H.Diggelmen, W. C.

Lawrence, S. K. Vernon, and R. J. Eisenberg. 1980. Structuralanalysis of the capsid polypeptides of herpes simplex virus types 1 and 2. J. Virol. 34:521-531. 6. Costa, R. H., B. G. Devi, K. P. Anderson, B. H. Gaylord,

and E.K. Wagner. 1981. Characterization of a major late herpes simplex virus type 1 mRNA. J. Virol. 38:483-4%. 7. Denhardt, D. T. 1966. A membrane-filter technique for the detection ofcomplementary DNA. Biochem. Biophys. Res.Commun. 23:641-646.

8. Draper, K. G., R. J. Frink, and E. K. Wagner. 1982. Detailedcharacterization of an apparently unspliced 1 herpes simplex virus type 1 gene mapping in the interior of another. J. Virol. 43:1123-1128.

9. Edwards,S.A.,andH.Fan.1979.gag-Related polypro-teins ofMoloney murine leukemia virus: evidence for independent synthesisofglycosylatedandunglycosylated forms. J. Virol. 30:551-563.

10. Franke, B., H. Moss, M. C.Timbury, and J. Hay. 1978. Alkaline DNaseactivityin cellsinfected witha tempera-ture-sensitivemutantofherpes-simplex virus type 2. J. Virol. 26:209-213.

11. Frink. R.J., K. P. Anderson, and E. K. Wagner. 1981. Herpessimplex virus type 1HindlII fragment L encodes spliced and complementary mRNA species. J. Virol. 39:559-572.

12. Frink. R. J., K. G. Draper, and E. K. Wagner. 1981. Uninfected cell polymerase efficiently transcribes early but not late herpes simplexvirus type 1 mRNA. Proc. Natl. Acad. Sci. U.S.A. 78:6139-6143.

13. Frink,R.J., R.Eisenberg,G.Cohen,andE.K.Wagner. 1983. Detailed analysis of the portion of the herpes simplexvirus type 1 genomeencoding glycoproteinC. J. Virol. 45:634-647.

14. Greenwood, F., W. Hunter, and J. Glover. 1963. The preparationof125I-labeledhumangrowthhormoneofhigh specificradioactivity.Biochem. J. 89:114-123. 15. Hall, L.M.,K.G.Draper,R.J.Frink,R.H.Costa, and

E. K.Wagner.1982.Herpes simplexvirus mRNA species mappinginEcoRIfragmentI.J. Virol. 43:594-607. 16. Hoffman, P.,andY.Cheng.1977. Thedeoxyribonuclease

induced after infection of KB cells by herpes simplex virus type1 ortype 2. J.Biol. Chem. 253:3557-3562. 17. Kennett, R. H., K. A. Denis, A. S. Tung, and N. R.

Klinman. 1978.Hybridplasmacytoma production:fusion with adultspleencells, spleenfragments,neonatalspleen cells and human spleen cells. Top. Microbiol. Immun. 81:77-91.

18. Kozak, M.1981. Possiblerole offlankingnucleotides in recognition ofthe AUG initiator codon by eukaryotic ribosomes. Nucleic Acids Res. 9:5233-5252.

19. Laemmil, U. K. 1970. Cleavage of structural proteins

during the assembly of the head ofbacteriophage T4.

Nature(London)277:680-685.

20. Lemaster, S.,andB.Roizman. 1980.Herpes simplex virus phosphoproteins. II.Characterization of the virion pro-tein kinase and ofthepolypeptidesphosphorylated inthe virion. J.Virol. 35:798-811.

21. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring HarborLaboratory,ColdSpring Harbor,N.Y. 22. Marsden, H. S., L. Haarr, and C. M. Preston. 1983.

Processingofherpessimplexvirusproteinsandevidence thattranslationofthymidinekinasemRNAis initiatedat

threeseparateAUG codons. J.Virol. 46:434-445.

23. Maxam,A.,and W.Gilbert. 1980.Sequencingend-labeled J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

DNA with base-specific chemical cleavages. Methods Enzymol. 65:499-559.

24. McKnight, S. L. 1980. The nucleotide sequence and transcript map of the herpes simplex virus thymidine kinasegene. Nucleic Acids Res. 8:5949-5964.

25. Morrison, J., and H. Keir. 1968. A new DNA exonuclease in cells infected with herpesvirus: partial purification and properties of the enzyme. J.Gen. Virol. 3:337-347. 26. Moss,H., P. Chartrand, M. C. Timbury, and J. Hay. 1979.

Mutantof herpessimplex virus type2with temperature-sensitive lesions affecting virion thermostability and DNaseactivity: identification of the lethal mutation and physical mapping of thenuc- lesion. J. Virol. 32:140-146. 27. Palmiter, R. D. 1974. Mg++ precipitation of ribonucleo-protein complexes. Expedient techniques for the isolation of undegraded polysomes and messenger ribonucleic acid. Biochemistry 13:3606-3614.

28. Powell,K. L., and D. J. M. Purifoy. 1977. Nonstructural proteinsof herpes simplex virus. I. Purification of the induced DNA polymerase. J. Virol. 24:618-626. 29. Preston,C. M., and M. G. Cordingley. 1982. mRNA- and

DNA-directed synthesis of herpes simplex virus-coded exonuclease in Xenopuslaevis oocytes. J. Virol. 43:386-394.

30. Purifoy, D.J. M.,andK. K.Powell.1976.DNA-binding

proteins inducedby herpes simplex virus type 2 inHEp-2 cells. J. Virol. 19:717-731.

31. Spear, P., and B. Roizman. 1980. Herpes simples virus, p. 615-746. In J. Tooze (ed.), Molecularbiology of tumor viruses: DNA tumor viruses, part 2, 2nd ed. ColdSpring Harbor Laboratory, Cold Spring Harbor, N.Y. 32. Stringer,J., L. Holland,R.Swanstrom, K. Pivo,andE.

Wagner.1977.Quantitation of herpes simplex virustype 1

RNA in infected HeLa cells. J. Virol. 21:889-901. 33. Strobel-Fidler, M.,and B. Franke.1980. Alkaline

deoxyri-bonuclease induced by HSV type 1: composition and properties of the purified enzyme. Virology 103:493-501. 34. Vernon, S. K., M. Ponce de Leon, G. H.Cohen, R. J. Eisenberg, and B. A. Rubin. 1981. Morphological

compo-nentsof herpesvirus. III. Localization of herpessimplex

virus type 1 nucleocapsid polypeptides byimmune elec-tronmicroscopy. J. Gen. Virol. 54:39-46.

35. Wagner, E. 1983. Transcription patterns in HSV infec-tions, p. 239-270. In G. Klein (ed.), Advances in viral oncology, Vol. 3. Raven Press, New York.

36. Wagner, E., K. Anderson,R.Costa, G.Devi,B.Gaylord, L.Holland, J. Stringer, andL. Tribble. 1981. Isolationand characterization of HSV-1mRNA, p. 45-67. InY. Becker (ed.), Developments in molecular virology, vol. 1: Her-pesvirus DNA. MartinusNijhoff,B.V., The Hague.

on November 10, 2019 by guest

http://jvi.asm.org/

ERRATUM

High-Resolution Characterization of

Herpes

Simplex Virus

Type 1 Transcripts

Encoding Alkaline Exonuclease

and

a

50,000-Dalton

Protein

Tentatively

Identified

as a

Capsid

Protein

R. H.COSTA, K. G.DRAPER, L. BANKS, K. L. POWELL. G.COHEN, R. EISENBERG, ANDE. K. WAGNER

Department of MolecularBiologyand Biochemistry, Universit'ofCalifornia, Irvine. Californiia92717;DepacrtmentofMicrobiology (Virology), Universityof Leeds, LeedsLS2 9JT, England:andDepairtnentofMicrobiologyatndCenterforOrial HealthResearch, Dental

Medicine, and Departmentof Pathobiologv. SchoolofVeterinaryMedicine, Universityof Pennsylvania, Phildelpliti, Pensylvani 19104

Volume 48, no. 3, p. 591-603: In this paper the authors incorrectly quoted Lemaster and Roizman (S. Lemaster and B. Roizman, J. Virol. 35:798-811, 1980 [20]) as locating VP-19C in the region in question. VP-19Chas been located elsewhere on the HSV-1 genome (D. K. Braun, W. Batterson,and B. Roizman, J. Virol., in press). The protein located by Lemaster and Roizman is VP-18.8.