Comparative structural analysis of glycoprotein gD of herpes simplex virus types 1 and 2.

0022-538X/80/08-0428/08$02.00/0 Vol.35,No.

Comparative

Structural

Analysis

of

Glycoprotein

gD of

Herpes Simplex

Virus

Types

1

and

2

ROSELYN J.EISENBERG,`* MANUEL PONCE DELEON,2 AND GARY H. COHEN'

Department ofPathobiology, SchoolofVeterinary Medicine,1 andDepartment ofMicrobiologyandCenter for OralHealthResearch,SchoolofDentalMedicine,2University ofPennsylvania, Philadelphia,

Pennsylvania19104

We studied thesynthesisandprocessingofthetype-common glycoprotein gD inherpes simplexvirus type 2 (HSV-2) andcompareditstructurallyto glycopro-tein gDofherpes simplexvirustype 1 (HSV-1). We demonstratedthat in HSV-2, gD undergoes posttranslational processing from a lower-molecular-weight precursor(pgD51) toahigher-molecular-weightproduct (gD56). Tryptic peptide analysis bycation-exchange chromatography indicated thatthisprocessingstep alteredneither the methionine northe argininetryptic peptide profile ofgD of HSV-2. Comparative tryptic peptideanalysis of gD of HSV-1 andHSV-2showed that the methionine and arginine tryptic peptide profiles ofthese two proteins wereverysimilar,but notidentical. Some of the resolvedpeptidescoelutedfrom

the cation-exchange column, suggesting that someamino acid sequences ofthe

two proteins might be very similar. However, each protein also appeared to possess several type-specific tryptic peptides. The structural similarityofthese

twoglycoproteins correlateswell with theirantigenic cross-reactivitysince

mon-oprecipitinantibodyto gDofHSV-1 alsoimmunoprecipitatesgD ofHSV-2and

neutralizes theinfectivityofbothvirusestoapproximatelythesameextent.

Herpes simplex virus (HSV) appearsto

con-tain five

glycoproteins

which have been desig-nated gA, gB, gC, gD,and gE (1). These

mole-cules differin

electrophoretic

mobilityin sodium

dodecyl sulfate (SDS)

polyacrylamide

gels (3,8,

14, 16, 19, 22,26), inbiological properties (1, 13, 23, 28),and ingenetic mappositions (14,22).

Previously,wedescribedanHSV-specific

an-tigen, CP-1 (5, 7), purified from HSV type 1

(HSV-1)-infected

cells andhavingtheproperties

ofa glycoprotein. This glycoprotein stimulated

production ofhightitersoftype-common

neu-tralizing activity. Recently(5, 8), weshowedthat

CP-1 is associated in infected cells with two components, a 52,000-molecular-weight (52K) precursor (pgD 52) and a 59K product (gD59).

Furthermore,werecentlyestablished by tryptic

peptideanalysis that pgD52 and gD59 of HSV-1 shared methionine and arginine tryptic pep-tides (8). In addition, gD found in cytoplasmic

extracts of HSV-1-infected cells has the same methionine peptide profile as gD isolated from the virion.ProcessingofpgD52 togD59involved addition of carbohydrate and especially sialic acid residues to theprotein, and this latter ad-dition altered the charge and the molecular weight of themolecule (8; manuscript in prepa-ration).

AlthoughHSV-1differs from HSV-2 in a

num-ber of biological, biochemical, and biophysical

properties

(15,

20),

the two types also exhibit

similarities.DNA-DNA

hybridization

studies

in-dicate that the DNAs of HSV-1 and HSV-2

exhibit

approximately

50%

homology

with 85%

matching

of base

pairs

of the

homologous

re-gions

(12).

Whereas the

glycoproteins

ofHSV-2 appear in the same general

molecular-weight

range inSDSpolyacrylamide gel

electrophoresis

(PAGE)

asthose of

HSV-1,

what appearstobe

glycoprotein gD

ofHSV-2 isslightly smaller in molecular

weight

thanglycoprotein gDof

HSV-1

(3,

14, 19, 22, 25).

Studies with

intertypic

recombinantssuggest that

gD

fromHSV-1 and

gD

from HSV-2 map in approximatelythesame

position

onthe

unique

short segment of the viral genome

(14, 22).

In

addition,

we found that

monoprecipitin antibody to gD ofHSV-1

neu-tralizes the infectivity ofHSV-1 and HSV-2 to

approximately

thesameextent

(5).

Almost all of the workestablishing

precursor-product

relationshipsamongHSV

glycoproteins

has been conducted on HSV-1. In the present

study,

weextended this type ofanalysisto HSV-2. The purpose of the present studieswas two-fold:

(i)

toestablish whetheraprecursor-product

relationship

similartothatfound forgDof

HSV-1 exists for

gD

ofHSV-2,and (ii) todetermine whether

gD

ofHSV-2 is structurally similarto

gD

ofHSV-1.

We presentevidence that the precursor

pgD51

428

on November 10, 2019 by guest

http://jvi.asm.org/

andproduct gD56 forms of gD of HSV-2 share

methionine andarginine trypticpeptides.

Proc-essingofgDofHSV-2 does not appear toinvolve

changes in thepolypeptideportion of the mole-cule. Detailedtryptic peptideanalysesof gD of

HSV-1andgDofHSV-2 indicate thatthe

mol-ecules are verysimilar, thoughnotidentical,and appear tohaveseveraltryptic peptidesin

com-mon.Each proteinalso appears to possess

sev-eraltype-specifictryptic peptides.

MATERIALS AND METHODS

Cell cultures. Conditions for the growth and

main-tenanceof BHK cells have beenpreviouslydescribed (7).

Viruspreparationand titration. The procedures used for the preparation ofvirus stocks of HSV-1 (strainHF) and HSV-2 (Savage strain) and the plaque assayweredescribed previously (4, 7). Forinfection,

unless otherwise noted, an input multiplicity of 20 PFU of HSV-1 or 10 PFU of HSV-2 per cell was

employed.

Pulse-chaseexperiments.Previous studies of this kind using HSV-1 were done with KB cells (5, 8).

However, theSavagestrainof HSV-2 doesnotyield

high titersof virus in KB cells. In other studies (8,17), weestablished that the methionine andarginine tryp-ticfingerprintprofilesofgDfrom HSV-1arethesame

in BHKorKB cells.Nevertheless,toavoid any

prob-lems in comparing gDof HSV-1 withgD of HSV-2 thatmight have arisenas aresult of differences in cell type,weperformedallpulse-chaseexperimentsinthis study on BHK cells by a modification of methods

previouslydescribed forKB cells (5, 8).To increase theincorporation ofisotopicallylabeled methionineor arginine, the cells(35-mmplates)wereoverlaid after infection withEagleminimal mediumcontaining 1/10

thenormal concentration of methionine orarginine. Pulse-labelingwith methionineorargininewascarried

out by incubating infected cells in0.5 ml of Hanks salts containing one of the following radioisotopes:

[3S]methionine (specific activity,600Ci/mmol), 200

,uCi;[methyl-3H]methionine (specific activity,100Ci/

mmol),250,Ci;[U-'4C]arginine (specificactivity,336 mCi/mmol), 125

jLCi;

[2,3-3H]arginine (specific

activ-ity,15Ci/mmol),125,uCi.Forincorporationof labeled mannose,infectedcellswerepulse-labeledin0.5mlof

Eagleminimal medium minusserumandglucosewith

500

l.Ci

of [2-3H]mannose (specific activity, 18 Ci/

mmol).Aftera15-min pulse,thelabelwasremoved,

and the monolayers were either washed with iced saline and immediately frozen at -70°C (pulse) or

incubated foranadditional4h inprewarmedcomplete medium(chase).After thepulseandchase,thecells

werelysed, andcytoplasmicextracts werepreparedas described previously (5, 8). Extracts were stored at -70°C. All radioisotopes were purchasedfrom New

England NuclearCorp.

Virus purification. HSV-1 was purified by the method ofSpearand Roizman (27). Cellswere har-vested for viruspurificationat18 hpostinfection(p.i.). TosolubilizetheHSVenvelope, purifiedvirionswere

suspended in0.02M Trisbuffer (pH7.5) containing

0.15 M NaCl andNonidet P-40(NP-40) and incubated

at25°C for 1 h. L-1-Tosylamide-2-phenethyl chloro-methyl ketone and N-2-p-tosyl-L-lysine chlorochloro-methyl ketone hydrochloride were added, each at a concen-trationof 0.1 mM, to inhibit proteolytic activity. Nu-cleocapsids were removed by centrifugation at 100,000

xgfor2h. The resulting pellet was reextracted with NP-40and recentrifuged. The supernatants from both extractions were pooled and frozen at -70°C.

Immunological studies.The preparation of

anti-seraagainst HSV-1 used in these studies was previ-ously described (5, 7). Anti-ENV-1 serum was pre-paredagainst an NP-40 extract of the virion envelope ofHSV-1. Anti-CP-1 serum was prepared against a purified preparation of gD of HSV-1 (5). Anti-ENV-2 serum wasprepared against an NP-40 extract of virion envelopes of HSV-2 by the regimen previously de-scribed (7). Anti-ENV-2 serum had a serum neutrali-zation titer (50% reduction of PFU [18]) of 1,024 againstHSV-2 and 256 against HSV-1.

For antibody precipitation, Staphylococcus aureus Cowanstrain I(IgSorb; New England Enzyme Center) wasemployed to collect antigen-antibody complexes (11, 17, 24). Each antiserum was titrated to ensure that the maximal amount of antigen was precipitated. The precipitates were washed, and the antigen-anti-body complexes were dissociated as described previ-ously (17).

SDS-PAGE. SDS-PAGE wascarried out in slabs of 10%acrylamide cross-linked with 0.4% N,N'-diallyl-tartardiamide by essentially the same method de-scribed bySpear (25). After electrophoresis, the gels werestained withCoomassiebrilliant blue (5), dried

onfilter paper, and exposed to Kodak X-Omat R (XR-5) film. To locate tritium-labeled bands for tryptic peptideanalysis, the dried gelswereexposed to LKB Ultrafilm (LKB Instruments, Inc.). For fluorography, the procedure of Bonner and Laskey (2) was followed. Protein standardsranging from 15,000to 130,000

dal-tons were run oneach gel (5).

Preparation of samples for tryptic peptide

analysis. N,N'-diallyltartardiamide gel slices were dissolved in 2% periodic acid by the method of Gibson (9) and prepared for trypsinization by methods de-scribed previously (8). Trypsinization and ion-ex-change chromatographywerecarriedoutaspreviously described (8, 29), except that in certain cases,a modi-ficationwasmade inthe buffer system, whichhelped

toresolvepeptides that elutedatthemorebasic end ofthepH gradient. For purposes of clarity, the buffer system describedpreviously (8) will be called buffer system 1,andthe modified buffer system will be called buffer system2.Inbuffer system 1, buffer A consisted of 280ml of aceticacid,4mlofpyridine,and716ml of

water(8). Inbuffer system 2, buffer A consisted of310

ml ofaceticacid,4mlofpyridine, and716ml ofwater.

Inboth buffer systems, bufferB consistedof143mlof aceticacid, 161 ml ofpyridine,and 696ml ofwater.

Three mixing chambers were used to generate the

gradient(8);chambers1and 2each contained180ml

of bufferA, and chamber3contained60mlof buffer

Aplus120ml ofbufferB.Thearrows ontheFig.2, 3, 5, and 6indicate where elutionwithbuffer B alone (basicwash) started.

on November 10, 2019 by guest

http://jvi.asm.org/

430 EISENBERG, PONCE DE LEON, AND COHEN

RESULTS

SDS-PAGE of HSV-2 glycoproteins. In

previous experiments with HSV-1 infection of

KBcells, weestablished that synthesis and

proc-essing ofglycoproteins occurred optimally

be-tween5 and 8 hp.i. (unpublisheddata). Inthe current experiments employing BHK cells, we found that this timewasalsooptimal for

HSV-1orHSV-2glycoprotein synthesis and

process-ing. Morespecifically,synthesisandprocessing

ofpgD51 of HSV-2 could bedetectedasearlyas 2hp.i., and maximalsynthesisofthisprecursor

occurred by4 to6hp.i. Therefore, wechose5 h p.i. for pulse-labeling of both HSV-1- and HSV-2-infected BHKcells.

Figure 1shows theelectrophoretic pattern of

[3H]arginine-labeledproteins (tracks 1, 2, 5, and

6) or [2-3H]mannose-labeled proteins (tracks 3

and 4) immunoprecipitated from cytoplasmic

extractsofHSV-2-infected cells aftera 15-min pulseat5 hp.i.oraftera15-minpulseat5 hp.i. followedbya4-h chase. Twoantiserawereused in this experiment: a homologous antiserum, anti-ENV-2, prepared againstanNP-40extract of virionenvelope of HSV-2(tracks1through 4)

1

2

3

4

5

6

95K_

67

K-60K

49K_-_

36K

P--

25KP--

15K_-FIG. 1. SDS-PAGE analysis of HSV-2 glycopro-teins.Fluorogram ofa 10%

N,N'-diallyltartardiam-ide cross-linkedpolyacrylamidegel of immune

pre-cipitates obtained from lysates of HSV-2-infected BHKcells. Track1,[3HJarginine pulse, anti-ENV-2

serum;track2, [3HJarginine chase, anti-ENV-2

se-rum; track 3, [2-3Himannosepulse, anti-ENV-2

se-rum; track4, [2-3Himannose chase, anti-ENV-2

se-rum; track5, [3H]arginine pulse, anti-CP-1 serum; track6,[3H]arginine chase, anti-CP-1serum.Tracks

Iand 2wereexposedtoKodak X-Omat R (XR-5) film forIday. Tracks 3through6wereexposedtofilm for

3days.

anda heterologousserum, anti-CP-1, prepared

against gD of HSV-1 (5-8)

(tracks

5 and 6).

Tracks 1, 3,and5show thata51K

protein

was

immunoprecipitated

fromthe

pulse-labeled

cy-toplasmic extractby eitherserum. Tracks 2, 4,

and 6 show that thesesamesera

immunoprecip-itated a

larger-molecular-weight

form

(56K)

from the

cytoplasmic

extract of the chase.

Es-sentially

the same results were obtained when

[3H]methionine

or

[3S]methionine

wasused as

the label. Since [3H]mannosewas

incorporated

intothe 51K and 56K forms

(tracks

3and4), it

appearsthat both of thesemolecules were

gly-cosylated. Furthermore,

monoprecipitin

antise-rum togD of HSV-1 reacted with both the 51K

and 56K formsof gD of HSV-2(tracks5and 6).

Taken together, the results of this

experiment

show that the 51K and 56K

proteins

of HSV-2

are

antigenically

relatedand

probably

represent precursor(pgD51) and product (gD56) forms of thesame

glycoprotein.

Furthermore

pgD51

and gD56 are both

antigenically

related to gD of

HSV-1 (tracks5and 6). Itshould also be noted

that anti-ENV-2 serum immunoprecipitated

several other bands from the pulse-labeled

ex-tract with molecular weights of

approximately

63K, 65K, 105K, and 127K. These bands

ap-peared

tobe replaced in the chase bya setof

bands of73 to 97K and 110K. Since mannose

wasalsoincorporated into these polypeptides,it

appearsthatall of these bandsrepresent

glyco-proteins. These polypeptides were not

investi-gated further.

Tryptic peptide

analysis

of

pgD and gD of HSV-2. To establish that pgD51 and gD56 of

HSV-2were

structurally

related, we compared

thetwopolypeptides by tryptic peptide analysis.

Cytoplasmicextractslabeled with

[3S]methi-onine or

[3H]methionine

were

immunoprecipi-tated andanalyzed by SDS-PAGE. The bands

correspondingtopgD51 and gD56wereeluted,

trypsinized, and cochromatographedon a

Chro-mobeads P (Technicon Corp.) column. Figure2

shows that the elution profiles forpgD51 and

gD56 were identical. For both polypeptides,

sevendistinctmethionine-labeled peptide peaks

wereeluted from the column; one peak was the

flow-through (fractions 1 through6), three

mi-norpeakseluted in the acidic region (fractions

30through 50), two majorpeaks elutedin the

middle portion of the column, and one peak

elutedwith the basicwash. These results

indi-catethat pgD51 and gD56 ofHSV-2 are

struc-turally

related.

To obtain a more complete structural

com-parison ofpgD51 and gD56, we examined the

arginine

profiles

of the two proteins. Figure 3

shows that pgD51and gD56shared

essentially

all resolved argininepeptides. Moreover, since a

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.504.66.256.341.534.2]

2 z

2₀

z

2

C) In

in

2 0.

1600

iL

z

z

0

z

w

V)

a.)

0 10 20 30 40 50 60 70 60 90 100 110 120 130 140 ISO 960 1701610 1so FRACTION NUMBER

FIG. 2. Tryptic fingerprint analysis of glycoprotein gD of HSV-2.pgD51 ( )isolated by immunoprecip-itation and SDS-PAGE from [3H]methioninepulse-labeled extracts (15 min at 5hp.i.) was cochromato-graphed on acolumn of Chromobeads P with[3S]methionine-labeled gD56 (---) isolated in a similar fashion from the chase (15-min pulse at5h p.i.,followed by a 4-h chase). Fraction size 2.2 ml, buffer systemI

(see the text).

In :

I

Wz w

2

I_

i

0.r," U'i

0 l0 20 30 40 50 60 70 80 90 900 110 120 130 14 150 0 170 10 190 200 210 220

FRACTION NUMBER +

FIG. 3. Trypticfingerprint analysis of glycoprotein gD of HSV-2. pgD51 ( )isolated by immunoprecip-itation andSDS-PAGEfrom[14C]argininepulse-labeled extracts (15min at 5hp.i)wascochromatographed

on acolumnof Chromobeads P with

[3HJarginine-labeled

gD56(---) isolated in a similar fashionfromthe chase(15-minpulseat5hp.i., followed by a 4-hchase).Fraction size 2 ml, buffer system 2 (see the text).

largenumber ofargininepeptideswereresolved,

it is

likely

thata

fairly

substantial

portion

of the

amino acid sequence in the two

polypeptides

wasidentical.Thus,weconcluded that the

proc-essing of

pgD51

to gD56 in HSV-2 does not

entail any

major

alteration in the

peptide

por-tion of themolecule.

Comparison

ofgD of HSV-1 and HSV-2. (i)

Immunological

similarities.Weshowed in

Fig. 1 that monoprecipitin antibody to gD of

HSV-1 (5, 8) also acted as a

monoprecipitin

antibodytopgD51 and

gD56

ofHSV-2.

Figure

4 compares pgD51 of HSV-2

(track 1)

with

pgD52ofHSV-1

(track

2)by SDS-PAGE.These

results showed

directly

that the type1precursor

of gD has a

slightly higher

molecular

weight

than the type 2precursor. Thisdifference has

been noted

previously

(3, 14, 22,

26)

and has

beenusedasthe basis for

distinguishing

thetwo

proteinsin

intertypic

recombinants(14,

22).

Pre-viously (5), we showed that

anti-CP-i

serum

neutralized theinfectivityof HSV-2to

approxi-mately the same extent as

HSV-1,

suggesting

that some of the sites on

pgD

involved in the initial stages of infection were similar in struc-ture.

(ii)

Structural similarities. The precursor

forms of

glycoprotein gD

ofHSV-1 andHSV-2 wereisolated bySDS-PAGE,

oxidized,

trypsin-ized, and

cochromatographed

on a column of

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.504.57.448.47.240.2] [image:4.504.57.449.309.462.2]

i.. 2

FIG. 4. SDS-PAGEanalysis of glycoprotein gD of HSV-1 and HSV-2. Fluorogram ofa 10%o N,N'-dial-lyltartardiamide cross-linked polyacrylamide gel of immuneprecipitates obtained from pulse-labeled ex-tracts ofHSV-1 and HSV-2. Track 1, [3H]arginine

pulse, HSV-2-infected cells,anti-CP-1serum;track2,

[3H]argininepulse, HSV-1-infected cells, anti-CP-1 serum.

Chromobeads P.Figure5 showstheelution

pro-fileof[3H]methionine-labeled pgD52 of HSV-1 cochromatographed with

[35S]meuhionine-la-beled pgD51 of HSV-2. The results show that the gD glycoproteins of HSV-1 and HSV-2 shared one major methionine-labeled peptide

peak (fraction 107, pH 3.33). The second major methionine-labeled peak seen for pgD51 of HSV-2 (fraction 90) wasabsentfrom pgD52 of HSV-1. ThepatternforpgD52 of HSV-1 agrees

withprevious results (8, 17). Two type-specific peaks, onefor pgDof HSV-1 (fraction 35) and one for pgD of HSV-2 (fraction 50), were

re-solved in this experiment. We concluded that

thegDglycoproteinsofHSV-1 and HSV-2 have

onemethionine-containing peptide peak in com-mon(pH 3.33) and three thataredifferent.

In the next experiment (Fig. 6), pgD51 of HSV-2 labeled with [3H]arginine was

cochro-matographed with pgD52 of HSV-1 labeledwith

['4C]arginine. It can be seen that the elution profiles ofthe tryptic peptides of the two

pro-teinswereverysimilarthroughoutthe gradient.

Insome cases,thepeaks overlapped completely

(e.g., peaks at fractions 99, 115, 131, 167, 175,

181,and 185). In otherinstances, argininepeaks

for the two virus types eluted very closely but

did not overlap completely. An interesting

ob-servation about these similar but

nonoverlap-ping peaks is that in every case, the peak for HSV-2 eluted at a slightly more acidic pH (to

the left) than did the

corresponding

peak

for

HSV-1. In a few cases, notably atfractions 30,

40, 126, and possibly 140, there were peaks

re-solved for pgD52 of HSV-1 that were missing

fromtheprofileofpgD51of HSV-2. These four

peaks thus appear to represent

HSV-1-specific

arginine peptides.

Our conclusionfrom these

experiments

is that

themethionineandarginine peptide profilesof

gDofHSV-1 andHSV-2are verysimilar.

Sev-eral tryptic peptides

coeluted,

suggesting

that

they may bestructurally similar. However, the

two glycoproteins are not identical since each

contained severaltype-specificpeptides.

DISCUSSION

The studiespresentedin this paperestablish

two properties of

glycoprotein

gD. First, the

protein undergoes posttranslational processing

inHSV-2-infectedcells froma

lower-molecular-weightprecursor(pgD51)to a

higher-molecular-weightproduct (gD56) inamannerwhich does not appear to alter its methionine or arginine

trypticpeptide profile. Theprocessing ofgDof HSV-2 thus closely resembles that of gD of HSV-1 (8). The second property of gD estab-lished in this paper is that the methionine and

argininetryptic peptide profilesofgDof HSV-1

and HSV-2 appear to be very similar but not

identical. Someof the resolved tryptic peptides

ofthetwoproteinscoeluted during

chromatog-raphy, suggesting that their amino acid

se-quencesarequitesimilaroridentical.Thus, the

intertypic structural similarities correlate well

with the immunological cross-reactivityof anti-CP-1 serum (5). Otherinvestigatorshave noted

a similar cross-reactivity (10, 25). It might be interesting to determine whether the antigenic sites responsible for inducing neutralizing anti-body are structurally identical in the two pro-teins and whether, in fact, they correspond to

one or more of the resolved tryptic peptides. Assuming that gD participates in the absorption

phaseof virus entry,assuggested bySpear et al.

(28), studies of the structure of the specific pep-tides involved in this process in HSV-1 and HSV-2 may shed some light on the nature of the cell receptor.

Although our studies suggest that there may be considerable intertypic structural homology of gD, they do not allow us to ascertain the

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.504.111.204.68.321.2]

1000r

240-bjLh Z Z 200

0 0

FE~ 160

I1Jw 1120

I soS_

40 0.

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 1S 190

FRACTION NUMBER

FIG. 5. Comparative tryptic fingerprint analysis of glycoprotein gD ofHSV-1 andHSV-2. pgD52 isolated byimmunoprecipitation and SDS-PAGE from

['Himethionine

pulse-labeled extracts ofHSV-1-infected cells

( ) was cochromatographed on Chromobeads P withpgD51 isolated in a similar fashion from

['5S]-methionine pulse-labeled extracts of HSV-2-infected cells (-). Fraction size 2.2 ml, buffer system1 (see the

text).

200

170

1500

rxF-0 10 20 30 40 5o 60 70 80 90 100 110 120 130 140 ISO 160 170 180 190 200 210 FRACTION NUMBER

FIG. 6. Comparativetryptic fingerprint analysisofglycoproteingD ofHSV-1 andHSV-2.pgD52isolated

from['Hiargininepulse-labeledextracts ofHSV-1-infected cells ( ) was cochromatographedon

Chro-mobeads PwithpgD51 isolatedfrom

['4CJarginine

pulse-labeled extractsofHSV-2-infectedcells ( Fraction size2.0ml, buffersystem2(seethetext).

extent of this

homology.

Other properties of

these

glycoproteins

offersome additional clues

about the extent of structural relatedness. For

example, studies with intertypic recombinants

have shown that HSV-1 strains are infectious whentheycontain a type 2 gD and vice versa

(14, 22). This meansthat thetwoglycoproteins

arefunctionallysimilar and that any structural

differences that do exist, even in the site for

absorption,are notimportant enoughtoprevent

infectionbytherecombinant strain. In this

re-gard, some estimates of the number of

type-commonandtype-specific antigenicsites present

in gD weremade by Halliburtonet al. (10). In

thosestudies,monoprecipitin antibodytoband II (corresponding to gD) was made HSV-1 spe-cificbyabsorptionwith the heterologousvirus. This serum was capable ofneutralizing type 1

parents and recombinants containing gD from

HSV-1butnottype2parentsorrecombinants.

The data of Halliburtonetal.

imply

thatatleast

one of the antigenic sites involved in

inducing

neutralizing antibody is not identical in gD of

HSV-1 and HSV-2.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.504.53.448.65.260.2] [image:6.504.57.445.327.493.2]

Inthecourseofourpulse-chase experiments,

we noted that the apparent difference in the molecular weight of the product glycoproteins

(gD56andgD59)wasgreaterthan the difference inthe molecularweight of the precursors(pgD51

and pgD52). Since therewas noapparent alter-ationof thepolypeptidechainduring

processing

of either glycoprotein, it is possible that the greater apparent molecular weight change in

processing of gD of HSV-1 might be due to

addition ofmore ordifferentcarbohydrates dur-ing processing from the precursor, pgD, to the product,gD. We hadpreviouslyshown (8) that processing ofgD59 in HSV-1 involved addition

ofan 1,800-molecular-weight core

oligosaccha-ride (high-mannose type) to the core

polypep-tide; possibly a 50K molecule (17), to form pgD52. This was followed by another step of

processingin which thecoreoligosaccharidewas

modified by addition of other carbohydrates,

including sialic acid, to form a complex-type

oligosaccharide. A similar processing pattern

probablyoccursforgDofHSV-2. One

explana-tion for the smallermolecularweightincrease in

the processing ofgD of HSV-2 is thatsome of

theoligosaccharideside chains of this molecule

are not converted to the complex type. For closely related murine leukemia viruses, slight

differences in protein structure apparently de-termine whetherornot aparticular

oligosaccha-ride side chain willundergoprocessingfrom the high-mannose type to the complex type (21). Thus,slight differences in amino acid sequence might accountfor thedifferenceinthe apparent

molecular weightofgD56 and gD59.

This explanation would not account for the smaller molecular weight difference of the two precursorspgD51 andpgD52. It ispossible that gD of HSV-1 is8 to 10 amino acidslarger than

gDof HSV-2.Alternatively,thisdifferencecould be due to the number of attached oligosaccha-ride chains. For instance, gD of HSV-2 might contain oneless asparagine-linked oligosaccha-ride than gD of HSV-1. This might be due to substitution of anotheramino acid for aspara-gine. In fact, if gD ofHSV-2 contained fewer

glycosylation sites, there would be fewer sites

available forsialylation. Processing of pgD51 to gD56 would therefore involve asmaller molec-ularweight change than processing of pgD52 to gD59 even if all of the oligosaccharides of the

twoproteins were processed similarly.

These differences in oligosaccharides are all consistent with the idea that gD glycoproteins ofHSV-1 and HSV-2 are very similar in amino acid sequence. Preliminary evidence suggests that thearginine peptides present in gD of HSV-1, but not in gD of HSV-2, can be partially

accounted forbytype-specific lysine peptidesin gDof HSV-2. Thus, some of theintertypic

dif-ferences in thearginine peptide profilesshown

in Fig. 6could arise as theresultofasinglebase change, e.g., AAA (lysine) toAGA(arginine) or

AAG (lysine) to AGG (arginine). Ofcourse, all of this is speculative, and the actual extent of

structural homologycanonlybedetermined by

moredetailedanalysisof both thecarbohydrate

and peptide portionsof theproteins.

Ourpreviousstudies of thecapsid proteins of

HSV-1 and HSV-2 ledustoconclude that none of the analogous structural proteins ofHSV-1

and HSV-2wereidentical, although two of them (themajor capsid protein NC-1 and thesmallest

capsid proteinNC-7) appearedtobe verysimilar by tryptic peptide analysis (6). gD represents thesixth HSVstructural protein examined for intertypicsimilarities by tryptic peptide analysis (6) and thethirdonewhich appears tobe highly

conservedinthetwovirus types.

ACKNOWLEDGMENTS

This work wassupported byPublic Health Service grant DE-02623 from the National Institute for Dental Research and by a special grant to R.J.E. from the University of Pennsylvania.

Wethank Wesley Wilcoxand William C. Lawrence for help in preparation of the manuscript and Deborah Long and MadelineCohen for excellent technical assistance.

LITERATURE CITED

1. Baucke,R. B., andP.G.Spear. 1979.Membrane pro-teinsspecified by herpes simplex viruses.V. Identifica-tion of anFc-bindingglycoprotein.J.Virol.32:779-789. 2.Bonner, W. M., and R. A.Laskey.1974.Afilm detection method fortritium-labeled proteinsandnucleicacids in polyacrylamidegels. Eur.J.Biochem.46:83-88. 3.Cassai, E.,M.Sarmiento,and P.G.Spear.1975.

Com-parisonof the virionproteins specified by herpes sim-plexvirus types 1 and 2. J.Virol.16:1327-1331. 4. Cohen, G. H.,M.N.Factor, andM.PoncedeLeon.

1974.Inhibition ofherpessimplexvirus type 2 replica-tionbythymidine.J.Virol. 14:20-25.

5.Cohen, G. H.,M.Katze, C.Hydrean-Stern,and R. J. Eisenberg.1978.Type-commonCP-1antigenofherpes simplex virus is associated with a 59,000-molecular-weightenvelopeglycoprotein. J. Virol. 27:172-181. 6. Cohen, G. H.,M. Ponce deLeon,H.Diggelmann,W.

C.Lawrence,S. K.Vernon,and R. J.Eisenberg. 1980.Structuralanalysisof thecapsid polypeptidesof herpessimplexvirus types 1and 2. J. Virol. 34:521-531. 7. Cohen, G. H.,M. PoncedeLeon,andC. Nichols.1972. Isolation of a herpes simplex virus-specific antigenic fraction whichstimulates theproductionofneutralizing antibody.J.Virol. 10:1021-1030.

8. Eisenberg,R.J.,C.Hydrean-Stern,and G. H.Cohen. 1979.Structuralanalysisof precursorand productforms oftype-commonenvelope glycoprotein D (CP-1 anti-gen) ofherpessimplexvirus type 1. J. Virol. 31:608-620.

9. Gibson,W. 1974.Polyomavirusproteins.Adescription of the structuralproteinsof thevirion based on poly-acrylamide gel electrophoresis and peptide analysis. Virology 62:319-336.

on November 10, 2019 by guest

http://jvi.asm.org/

10. Halliburton,I. W., R. E.Randall, R. A.Killington, and D. H.Watson. 1977. Some propertiesof

recom-binants between type 1 and type 2 herpes simplex

viruses.J. Gen. Virol. 36:471-484.

11. Kessler,S. W. 1975.Rapidisolation ofantigensfrom cells withastaphylococcusproteinA-antibody adsorbent:

parametersof the interaction ofantibody-antigen

com-plexeswithproteinA.J. Immunol.115:1617-1624. 12. Kieff, E., B. Hoyer,S.Bachenheimer, and B.

Roiz-man. 1972. Geneticrelatedness oftype 1andtype 2

herpessimplexviruses.J.Virol.9:738-745.

13. Manservigi, R.,P. G.Spear,and A.Buchan. 1977.Cell fusion inducedbyherpes simplexvirus ispromotedand suppressed by differentviralglycoproteins.Proc. Natl. Acad. Sci.U.S.A. 74:3913-3917.

14. Marsden, H. S., N. D. Stow, V. G. Preston,M. C. Timbury, and N.M.Wilkie.1978.Physicalmapping ofherpes simplexvirus-inducedpolypeptides.J. Virol. 28:624-642.

15. Morse,L.S.,L.Pereira,and B. Roizman. 1978. The use of intertypic recombinants for analysis of gene organizationinherpes simplexvirus,p.41-61. InG. de-The, W. Henle, and F. Rapp (ed.), Oncogenesis and herpesvirusesIII. Part 1.InternationalAgencyfor Re-searchonCancer, Lyon.

16. Norrild, B.,S. L. Shore, and A. J. Nahmias. 1979. Herpessimplexvirusglycoproteins:participationof in-dividualherpes simplexvirustype1glycoprotein

anti-gensinimmunocytolysisand their correlation with

pre-viously identifiedglycopeptides.J. Virol.32:741-748. 17. Pizer,L. I.,G. H.Cohen,and R. J.Eisenberg. 1980.

Effectoftunicamycinonherpes simplexvirus glycopro-teins and infectious virusproduction.J. Virol. 34:142-153.

18. Ponce deLeon, M.,H.Hessle,and G. H. Cohen. 1973. Separationofherpes simplexvirus-inducedantigensby concanavalin Aaffinity chromatography.J.Virol. 12: 766-774.

19. Powell,K.L.,R.Mirkovic,and R. J.Courtney.1977. Comparative analysisofpolypeptidesinducedbytype

1andtype2 strains ofherpessimplexvirus.

Intervirol-ogy8:18-29.

20. Roizman, B. 1978.Theherpesviruses,p.769-848.InD.

P.Nayak (ed.), Themolecular biology of animal viruses, vol.2.Marcel Dekker, Inc., New York.

21. Rosner, M. R., L. S. Grinna,and P. W. Robbins.1980.

Differencesinglycosylationpatternsofcloselyrelated

murineleukemia viruses. Proc. Natl.Acad.Sci.U.S.A.

77:67-71.

22. Ruyechan, W. T., L. S. Morse, D. M. Knipe, and B. Roizman. 1979. Molecular geneticsofherpessimplex virus. II. Mapping of the major viral glycoproteins and of the genetic loci specifying the social behavior of infectedcells. J. Virol. 29:677-697.

23. Sarmiento, M., M. Haffey, and P. G. Spear. 1979.

Membrane proteins specifiedbyherpessimplex viruses. III. Role of glycoprotein VP7(B2)in virion infectivity. J. Virol.29:1149-1158.

24. Smith, A.E.,R.Smith, andE.Paucha.1978.Extraction andfingerprint analysis of simianvirus 40 largeand smallT-antigens. J. Virol.28:140-153.

25. Spear, P. G. 1975. Glycoproteins specified by herpes simplex virus type 1: theirsynthesis, processingand antigenic relatednesstoHSV-2 glycoproteins,p.49-61.

InG.de-The,M. A.Epstein, and H. Zur Hausen (ed.),

Oncogenesis and herpesvirusesII.Part2.International AgencyforResearchonCancer,Lyon.

26. Spear,P. G.1976.Membrane proteins specifiedby herpes simplex viruses. I. Identification of four glycoprotein

precursorsand theirproductsintype1-infectedcells.J. Virol.17:991-1008.

27. Spear, P. G.,and B.Roizman.1972.Proteinsspecified by herpes simplexvirus.V.Purification and structural proteinsof theherpesvirion. J. Virol.9:143-159. 28. Spear, P. G., M. Sarmiento, and R. Manservigi. 1978.

Thestructuralproteins andglycoproteinsof

herpesvi-ruses: areview,p.157-167.InG.de-The,W.Henle,and F.Rapp(ed.), OncogenesisandherpesvirusesIII. Part I. InternationalAgency forResearchonCancer, Lyon.

29. Vogt, V.M.,R.Eisenman,and H.Diggelmann.1975. Generation of avianmyeloblastosisvirus structural pro-teinsby proteolyticcleavageofaprecursorpolypeptide. J. Mol. Biol. 96:471-493.

on November 10, 2019 by guest

http://jvi.asm.org/