Membrane proteins specified by herpes simplex viruses. V. Identification of an Fc-binding glycoprotein.

(1)

0022-538X/79/12-0779/11$02.00/0 Vol. 32, No. 3

Membrane Proteins Specified by Herpes

Simplex

Viruses

V. Identification of an Fc-Binding Glycoprotein

ROBERT B. BAUCKE ANDPATRICIA G. SPEAR*

Department ofMicrobiology,The University of Chicago, Chicago, Illinois 60637 Received for publication 7 June 1979

Aglycoprotein with affinity for the Fc region ofimmunoglobulinwas isolated from extracts of cultured cells infected with herpes simplex virus type 1, and

experimentsweredone to characterize its properties and to investigate whether

it could account for the Fc-binding activity previously demonstrated on the

surfaces of intact herpes simplexvirus-infected cells. The technique of affinity

chromatographywasusedtoidentifyandisolate the Fc-bindingglycoproteinand

to demonstrate the specificity of its interaction with

inmunoglobulin

G-Fc.

Although threeelectrophoreticallydistinguishable Fc-binding polypeptides were

identified by affinity chromatography, these three species appear to be different

forms of the same translation product, based on comparisons of proteolytic

digestionproducts andon the kinetics ofappearance of each form after a brief

pulse with radioactive amino acids. The results suggest that one polypeptide,

designated pE, is processedtoyield

gEj,

which is in turn processed to yieldgE2.

Both

gE,

and gE2 areglycosylatedmembrane proteins and both can be labeled

by the lactoperoxidase-catalyzedradioiodination of intact infectedcells,indicating

thepresenceof these proteins in surfacemembranes of the cells. Increases in the amountsofgE,and gE2at thecellsurface were found toparallelthe increase in

Fc-binding activity of intact infectedcells.

Cells infected with either serotypeofherpes

simplex virus (HSVtype 1 [HSV-1]or HSV-2)

display membrane receptors thathave affinity

for the FcregionofimmunoglobulinG(IgG) (11,

19, 26, 28, 29, 31,32). These viruses induce the

expressionof thereceptorsinavarietyof differ-entcelltypesthatnormallydonotdisplaysuch

bindingcharacteristics, i.e., culturedepithelioid

orfibroblastic cells. The increase innumberof

receptorsparallels theproduction of HSV

pro-teinsand,infact, has been showntodependon

viralgeneexpression (9,29,31).Morerecently,

ithasbeenreported thattwoother human

her-pesviruses, cytomegalovirus (15, 21, 23, 30) and

varicella-zoster virus

[M.

OgataandS. Shigeta,

Abstr. Annu. Meet. Am. Soc. Microbiol. 1979,

S(H)59,p.304],alsoinducereceptorscapableof

bindingtothe FcregionofIgG.

Immunoglobulins from several mammalian

species,includingman, have been showntobind

toHSV-infected cells(31, 32).Assumingthatall

theseimmunoglobulinsbind via their Fcregions

tothe same virus-induced receptor (whichhas notbeen proven in everycase),one couldinfer thatthe receptor hasaffinity forahighly

con-served portion of the antibody molecule. The fact thatidentification and characterization of the virus-induced receptor hassofardepended

on its affinity for IgG should not obscure the

possibility,however, that itmayhave evolvedto

interact with some other molecule, related in

structure toIgG.

Speculationsaboutthefunction(s)of this

her-pesvirus-inducedcell surfacereceptorhave been

basedon theassumptionthatbindingofIgGis

pertinenttoitsroleinviralreplicationor

path-ogenesis. Ithas beensuggested thatbindingof

IgGtothereceptors onthesurfaces of infected

cells could interfere with immune cytolysis (7,

16), and evidence has recentlybeen presented

thataggregated rabbitIgGcaninhibit both

an-tibody-dependent, complement-mediated and

antibody-independent,cell-mediated

cytolysis

of

HSV-infected cells in vitro (1). In addition, it

has beensuggested that the binding ofIgG to

theherpesvirus-inducedreceptors mayinfluence theexpressionofviralgeneproductsin infected

cellsandcouldperhapsleadtomalignant

trans-formationortheestablishment oflatency (8, 16,

31).Withregardtothishypothesisit isof

inter-est that incubation of HSV-infected cells with

relatively high concentrations of nonimmune

rabbit IgG can reduce the yield of infectious progeny recovered from thecells(8).Itremains to be determined, however, whether these

ob-served effects of

immunoglobulin

are

directly

779

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

relevanttothe primary function of the herpes-virus-inducedreceptor.

Regardless of the function of thisreceptor,its affinity for IgGcanbeexploited for the isolation

and characterization of themolecule(s)

respon-sible for theFc-binding activity. Sakuma etal. (23) recently presented evidence that

cytomeg-alovirus receptor-IgG complexescanbe

precipi-tatedby anti-IgG and that thecytomegalovirus Fc-binding receptor is a glycoprotein with an apparentmolecularweight of 43,000.

In this report we describe the identification

andisolation,by affinity chromatography, ofan

HSV-1-induced Fc-binding glycoprotein and also provide evidence that it could accountfor

atleastsome,andpossiblyall, of the Fc-binding

activity detectedonthe surfaces ofinfectedcells.

This Fc-binding glycoprotein accumulates in

two electrophoretically distinguishable forms, whoseapparentmolecularweightsare

approxi-mately 80,000 and 65,000, and is distinct from the HSV-1glycoproteins that have been

previ-ously characterized (6, 12, 17, 22, 24, 25, 27).

MATERIALS AND METHODS Cells and viruses. The cel lines used in these studieswere HEp-2andBHK-21,clone 13, obtained from Flow laboratories (Rockville, Md.). Both cell types were grown in the Dulbecco modification of Eagle minimal essential medium supplemented with 10% heat-inactivated fetal bovine serum. The virus

strains used were HSV-1(HFEM)syn, a syncytial

plaquevariant thatweisolated froma nonmutagen-ized stock ofHSV-1(HFEM) providedbyA.Buchan (University of Birmingham, Birmingham, England), and HSV-2(GP), asyncytial plaque variant isolated

fromHSV-2(G) by Cassaietal. (4). Although HSV-l(HFEM)syncausesextensive fusionof BHKcells,it exhibits much lessfusion-inducing activityinHEp-2 cells. HSV-2(GP),onthe otherhand,induces exten-sive fusion ofHEp-2 cells. Both virusstrains were

propagatedinHEp-2cellsand,fromtheir initial iso-lation, have been passaged onlyafew times at low

multiplicitiesof infection.

Chemicals and radioactive precursors. Re-agentsfor electrophoresis, including acrylamide and the cross-linker N,N-diaVlyltartardiamide, were ob-tained from Bio-Rad Laboratories(Richmond, Calif.). The nonionic detergent Nonidet P-40 (NP40) was

obtained from Gallard-Schlesinger Manufacturing Corp. (Carle Place, N.Y.);cyanogenbromide-activated Sepharose 4B was from Pharmacia Fine Chemicals

(Piscataway, N.J.); bovine serum albumin (BSA),

chickeneggalbumin(OV),andproteaseV8 of Staph-ylococcusaureus werefromMilesLaboratories

(Elk-hart, Ind.); rabbit anti-BSA and anti-OV IgG (each

containing4 to6mgofspecific antibodyperml)were

fromCappelLaboratories (Downington, Pa.);pepsin, glucose oxidase, and lactoperoxidase were from

Worthington Biochemicals Corp. (Freehold, N.J.); chloramine-Twasfrom Sigma (St. Louis, Mo.); and

[35S]methionine (500 to 600Ci/mmol),

['4C]glucosa-mine(45to 60mCi/mmol),and[125I]Na (carrier-free,

17 Ci/mg) were from New England Nuclear Corp.

(Boston, Mass.).

Infectionormock infection of cells and incu-bation with radioactive precursors. Monolayer

cultures of cellswereinfected with virusat aninput

multiplicity of5 to 10PFU/cell.Afteradsorptionfor 2 h at370C,theinoculumwasremoved,and the cells wereincubatedat37°C in medium199supplemented

with1% fetal bovine serum (199-V). When continu-ously labeledextracts weredesired, medium contain-ing[3S]methionine(3,uCi/mlin199-Vpreparedwith one-tenth the usual concentration of unlabeled methi-onine) or['4C]glucosamine (1 ,uCi/ml in 199-V) was addedat 3hafterinfection,andincubationwas con-tinuedat370Cuntil 18 h afterinfection,atwhich time the cellswereharvested. Forpulse-labeled extracts, infected cellswere incubated for 5 minat 6h after infection with [nS]methionine (10l,iCi/ml in 199-V lacking unlabeled methionine), and the cells were

either harvested immediatelyor after continued in-cubation in nonradioactive 199-V.

Cellsurface iodination.Cellsurfaceproteinswere

iodinated with "2I accordingto the

lactoperoxidase-catalyzed procedure describedby Hubbard and Cohn (13).Briefly, cellmonolayerswereoverlaid with

phos-phate-buffered saline(PBS;10mMNa2HPO4,1.5 mM KH2PO4, 0.14 M NaCl, 3 mM KCI,0.5mM MgCl2.

6H20,1mMCaCl2), pH7.4,containing5mMglucose

and [251I]Naat400,uCi/ml.Lactoperoxidaseand

glu-coseoxidasewereaddedto give final concentrations of 20ug/mland0.1U/ml, respectively. After10min atroomtemperatureon ashaker, unreacted125Iwas

pouredoff and the cellswerewashed andharvestedas usual.

Preparation of cell extracts. After cell mono-layerswerewashed withPBS,thecellswerescraped witharubberpoliceman and collected by centrifuga-tion. For extraction ofproteins the cells were

sus-pendedinPBS containing1%NP40 (1.0ml/2 x 107 cells) andkeptonice,with occasionalmixing, for15 min.Nucleiwerethen removedby low-speed

centrif-ugation,and thecytoplasmicextracts werestoredat

-70°C. Immediately before usefor affinity

chroma-tography,theextractswerethawed andcentrifugedat 25,000 rpm in an SW 27.1 rotor for 2 h to remove insoluble material.

Preparation and use of Fc affinity columns.

BSA orOV wascovalently linkedtocyanogen bro-mide-activatedSepharose4Baccordingtoprocedures supplied byPharmacia. Typically, 15g ofcyanogen bromide-activatedSepharose4B wasrehydratedand washed with1mMHCIonasintered-glass filter and then equilibrated with coupling buffer (0.1 M NaHCO3, pH 8.3, containing 0.5 MNaCl).BSAor OV dissolved incoupling buffer was added to give a final concentration of 5 to 10 mg ofprotein per ml, and couplingwasallowedtoproceed at roomtemperature for2hwithshaking ofthemixture. Unreactedprotein wasthen washed out with coupling buffer, and the remaining active groups on the Sepharose were blocked with 1Methanolamine, pH 8.5 (2 h at room temperature). The Sepharose was washed several times with coupling buffer, alternating with 0.1 M sodiumacetatebuffer(pH 4.0) containing0.5 MNaCl,

http://jvi.asm.org/

(3)

toremoveproteins that were adsorbed but not cova-lentlycoupled to the beads.Finally,the

protein-con-jugated Sepharose was equilibrated with PBS and stored at 4°C with 0.02% sodium azide.

Affinity columns were prepared by pouring suffi-cient BSA- or OV-conjugated Sepharose into 5-ml polypropylene columns toyield a bed volume of

ap-proximately1 ml. Thiswaswashed andequilibrated

with PBS-0.5% NP40 by passingatleast 25 ml of this bufferslowly through thecolumn,and then0.4 to 0.6 mgofspecific anti-BSAoranti-OVIgG (1.0to 3.0mg of totalIgG) in1.0ml of PBS-0.5%NP40wasadded. Thissolution wasallowed to enter the bed slowly over aperiodof1hat roomtemperature,andthe unbound

immunoglobulin was then washed out with excess

PBS-0.5% NP40. The columns were equilibrated at

370C, and NP40-solubilizedcellextractswere added

(3 ml/column). Every 15 min approximately 0.5ml wasallowedto enterthebeduntil all had beenapplied,

after which the columnswerewashedextensivelywith PBS-0.5% NP40.Finally,column-bound proteins

(in-cludingimmunoglobulin)wereeluted with 2to3mlof 3 Mpotassiumthiocyanate inPBS-0.5%NP40. The eluted macromoleculeswereprecipitatedwith 5% tri-chloroacetic acid, and theprecipitates were washed

sequentiallywith ethanol andacetoneand then solu-bilized foranalysis bysodiumdodecylsulfate

(SDS)-acrylamidegel electrophoresis (12).Greater than 70% of trichloroacetic acid-precipitable radioactivity was recovered for theanalysis.

Preparation of

F(ab')2

fragments from IgG. Rabbitanti-BSAIgGwastreated withpepsintoyield F(ab')2 fragments according to the procedure de-scribedbyNisonoff (20).Afterdialysisof theIgG (7 mg/ml) againstseveralchangesof0.1Msodium ace-tate,pH7.8, thepHwasadjustedtopH4.5with1M acetic acid andpepsinwasaddedtoyielda 1:50(wt/ wt)ratioofpepsintoIgG.This mixturewasincubated

overnightat370Candthenneutralized withNaOHto inactivatethepepsin.Asacontrol,anequalamount ofIgG wastreatedidenticallyexcept thatnopepsin

wasadded("mock-treated"IgG).Ouchterlony

double-diffusionprecipitationtestsindicated that the pepsin-treated, mock-treated, and untreated IgG had com-parable precipitating activity with BSA (data not shown).

Preparation and use of radiodinated

immu-noglobulin. RabbitIgG wasiodinated with "I ac-cordingtothe chloramine-Tmethoddescribedby Byrt

and Ada (3)and modifiedbyJensenius and Williams (14). A

10-Pd

amountof

["lI]Na

(5mCi)wasaddedto 5mlofchilled rabbitIgGsolution (1.5mg/mlin0.05 M sodium phosphate buffer, pH7), followedbythe addition andrapidmixingof100

pl

ofchloramine-T solution (1mg/ml).After10minonice the reaction wasstopped byaddition ofexcesstyrosine (100piat 0.4 mg/ml).TheIgG wasisolated from thereaction mixture bychromatography on Sephadex G-50 and hadaspecificactivityofapproxiimately 108 cpm/mg of IgG. This preparation was diluted 10-2 in PBS containing 5% heat-inactivated fetal bovineserumfor incubationwith intactcellmonolayerstomonitorthe appearance ofFc-bindingreceptors at varioustimes after infection. Incubationwasfor2hwithshakingat

370C,

afterwhich the

cells

were washed extensively

with PBS-5% heat-inactivated fetal bovine serum, sol-ubilized in 1%SDS,and sonicated briefly; small sam-ples were taken for quantitation of bound radioactiv-ity.

Isolation and partial proteolysis of the Fc-bindingpolypeptides.Pulse-labeled or continuously labeled forms of the Fc-binding polypeptides were isolated byaffinity chromatography, followed by pre-parative acrylamide gel electrophoresis, for partial

proteolysis and analysis of the peptides. For each analysis extracts were prepared from approximately 1.2x108cells and applied to a single affinity column. Thepolypeptides eluted from the column were con-centrated by trichloroacetic acid precipitation as de-scribed above andfractionated byelectrophoresison

SDS-acrylamide gelslabs(1.5mmthick). Wide sample wells (4 cm) were used to accommodate the large amountofmaterial. Both thefast-migratingandmore

slowlymigrating Fc-binding polypeptides were local-ized in the gel by either punching out segments of the

gelforquantitationofradioactivityorestimatingtheir positions relative to BSA (the location of BSAwas determinedbystaining one ortwo vertical strips of the gel; the remainder of thegelwasnotstained or fixed).

Ahorizontal strip of gel containing the desired Fc-binding polypeptide was cut from the gel slab and divided into 5-mm segments for the peptide-mapping procedure described by Cleveland et al. (5). Briefly, the gel segments were placed in sample wells of a second acrylamide gel (15% separating gel prepared with aN,N-diallyltartardiamide cross-linker instead ofbisacrylamide) andoverlaid with buffer containing variousconcentrations of the staphylococcal V8 pro-tease.Theresultingproteolyticfragmentswere sepa-ratedbyelectrophoresis at constant current (12mA/ gel) for about 6h. The gels were fixed and stained according to the procedure of Fairbanks et al. (10), and the radioactive bandswerelocatedby fluorogra-phy (2).

RESULTS

Detection of an IgG-binding protein in

extracts from infected celils. The technique

ofaffinity chromatography was applied to the

isolation ofamolecule(s) responsible for

HSV-1-inducedFc-binding activity. The affinity

col-umnsconsistedof rabbitantibodyboundto

an-tigen which itself was covalently coupled to

Sepharose 4B.Inallexperimentsreportedhere,

the antigen was either OV or BSA and the

antibodies had been prepared against OV or

BSA,respectively. Inthe experimentshown in Fig. 1, extractspreparedfrom

[3S]methionine-labeled infected cells, byuse of the detergent

NP40 as described in Materials and Methods,

were applied to a pair of OV-conjugated

col-umnns, onlyoneof whichcontained boundIgG.

After being extensively washedto remove

un-bound proteins, the columnswere sequentially

elutedwith2-mlaliquotsofpotassium

thiocya-natesolution,ranginginconcentration from0.25

http://jvi.asm.org/

(4)

782 BAUCKE AND SPEAR

A

FIG. 1. Elutionofpolypeptides bound Sepharose (A) orto controlOV-Sepharo

tractswereprepared from HEp-2cells in HSV-1(HFEM)synandincubated with[

ninefrom3to 18 hafter infection. Sam.

extractswereappliedtocolumns A and extensivewashingto removeunboundpi

columnswereelutedsequentially with 2-i ofthe solutionsindicated, asdescribed

The labeledpolypeptidespresentin each4 tionwereanalyzed by electrophoresison

amidegel slabs, autoradiograms ofwhich to 4 M, and then with 1% SDS. Fi Sepharosewassuspended inasecond

1% SDS and heatedto1000C for2mi

supernatant was recovered for ana

elutedproteins in each of these fraci precipitated by trichloroacetic acid,s

andanalyzed by SDS-acrylamide ele sis. Theautoradiogram shown in Fig.

stratesthespecific binding ofatleast peptide (indicated by the filled

triani

IgG-OV-Sepharose column(panel A) butnot to

I - the control OV-Sepharose column

(panel

B).

SDS SDS Thisproteinwas

optimally

elutedby

potassium

thiocyanate in therangeof 1.0to2.5 M. Small

quantities ofaproteinwith similar

electropho-reticmobilityhad boundtothecontrol column

(indicated by the arrow), but its elution at a

lower molarity of potassium thiocyanate

sug-geststhat it is eitheradifferentprotein of

ap-proximately the same molecularweight or the

sameprotein capable of interacting inadifferent

fashion with the control column.

_mww

An experiment was done to determine

whether theIgG-binding protein wasproduced

only after infection orcould also be detected in

extracts from uninfected cells. Extracts were prepared from HSV-1-infected and from

mock-infected cells labeled with [3S]methionine for

15h(from3 to 18hafterinfection),andaliquots

wereappliedeithertoIgG-BSA-Sepharose col-umns or to control BSA-Sepharose columns.

After

being

washed,

thecolumn-bound

proteins

wereelutedin asinglestepwith3Mpotassium

thiocyanate and prepared for electrophoresis.

Figure 2 shows that the IgG-binding protein

(indicated by the filled triangle) could be

iso-lated byaffinitychromatography fromextracts

of the infectedcells butnot from those of the

uninfected cells. It should be noted that this

IgG-bindingproteinisaminorcomponentof the

infectedcellextractsand isbarelydetectable in

electropherograms of unfractionated extracts.

For theelectrophoretic analysisshown in Fig.2,

theIgG-binding protein eluted from the affinity

column andappliedtothegelwasobtained from

- approximately80timesasmuchextract aswas alsoappliedtothegelforcomparative purposes. toIgG-OV- Results

indistinguishable

from those shown in se

(B).

Ex.

Fig. 2 were obtained whether OV or BSA was

tfectedwith

35simethio-

usedasthe

covalently

coupled

antigen

(data

not

wples

ofthe

shown),

indicating

that interaction of the

IgG-,

and,

after

binding protein

with the

affinity

column was

roteins, the

independent

of the

antigen

used.

mlaliquots Inpreliminary studies,wehave also identified inthetext. an

IgG-binding

protein induced after infec-elutedfrac- tion ofHEp-2 cells by an HSV-2 strain

[HSV-SDS-acryl-

2(GP)].

The

results

obtained by affinity

chro-are

shown.

matographywerevery

similar

tothoseshown in

Fig. 2, and theIgG-binding protein inducedby

inally, the HSV-2(GP) was indistinguishable by

electro-aliquotof phoretic mobility from theprotein induced by

in,and the HSV-1(HFEM)syn.

lysis. The Glycosylationof the IgG-bindingprotein.

tions were To determinewhether the IgG-binding protein

;olubilized, is glycosylated,extractswere also preparedfrom

ctrophore- HSV-1-infected cellslabeled with

["4C]glucosa-1demon- mine. Column-bound proteins from these

ex-tonepoly- tracts are compared with

[3S]methionine-la-gle) tothe beled, column-bound proteins in Fig. 3. The

http://jvi.asm.org/

[image:4.504.63.252.70.441.2]

(5)

Uninfected Infected

Column-bound proteins

Ext C IgG IgG C Ext

- -130 K

gB-gA,

4C-glu-NH2 35S-meth

Column-bound proteins

Ext C IgG IgG C Ext

S.F

S 130 K

[image:5.504.248.451.60.265.2] [image:5.504.54.245.74.290.2]

1 - 94 K

-° I D | - 68 K

...M 4

- 94 K

- 68 K gD- °

.-...4...

..- 43 K *5W

FIG. 2. Specific bindingtoIgG-BSA-Sepharoseof

a polypeptide produced by

HSV-1(HFEM)syn-in-fected HEp-2 cells, butnot byuninfected cells.

Ex-tractswereprepared from infectedormock-infected

HEp-2 cellsthathad beenincubated with

[rS]me-thioninefrom 3to 18 haftertheaddition ofvirus. Thisautoradiogram ofanacrylamide gelslabshows

the labeledpolypeptides that were bound to and elutedfromeitherIgG-BSA-Sepharose(IgG)or

con-trolBSA-Sepharose (C)andalsothelabeled

polypep-tidespresent in unfractionated extracts (Ext) from the uninfectedor infected cells (the column-bound

proteinswereobtainedfromapproximately80times

asmuchextractas waspresentintheunfractionated sample appliedtotheacrylamide gel).Inthis andall

subsequentfigureselutionwasachieved inasingle

stepbytheapplicationof 3Mpotassium thiocyanate, after extensive washing ofthe affinity columns to

remove unbound proteins. The numbers along the

right edge indicate the positions of unlabeled poly-peptidesusedasmolecularweightstandards (/3-ga-lactosidase, 130,000; phosphorylase b, 94,000; BSA, 68,000; OV,43,000).

results show thataglucosamine-containing

poly-peptide binds specifically to

IgG-BSA-Sepha-rosecolumns and thatits

electrophoretic

mobil-ity is the same as thatof the [

S]methionine-labeled, IgG-binding protein (filled triangles). Another glycosylated polypeptide of lower ap-parent molecular weight (open triangles) also

appearstobindspecifically tothe IgG-contain-ing column. Variable quantities of this species

were detected in replicate experiments.

Evi-dence will be presented below which suggests

thatthesetwoIgG-binding speciesarerelated in aminoacidsequence.

Demonstration that theIgG-binding

pro-teininteracts with theFcregionofIgG.To

FIG. 3. Presenceof radiolabelfrom

['4C]glucosa-mine, as well as from [35Slmethionine, in HSV-1-inducedpolypeptidesthat boundspecificallyto

IgG-BSA-Sepharose. This autoradiogram of an

SDS-acrylamide gelslabshows thelabeledpolypeptides

presentinunfractionatedextracts (Ext)from

HSV-1(HFEM)syn-infected HEp-2cells and the

polypep-tidesboundtoandelutedfromeither

IgG-BSA-Seph-arose(IgG)orcontrolBSA-Sepharose (C). The

poly-peptideindicatedbythefilled triangleshas thesame

electrophoretic mobility as the major IgG-binding

species shown in Fig. 1 and2. In different experi-ments, variable quantities ofthepolypeptide indi-catedby the opentriangles werealsofoundtobind

specificallyto the IgG-containing affinitycolumns. Thesymbolsshownontheleft edgearedesignations ofthemajorHSV-1glycoproteins, which have

previ-ously beendescribed (6, 27).

determine whether binding of the glycosylated

polypeptidestoIgGdependsupon anintact Fc

region, rabbit anti-BSA IgG was treated with

pepsinto yield F(ab')2 fragments, as described

in Materials and Methods. The F(ab')2

frag-ments were demonstrated tohaveretained

an-tigen-bindingactivityanddivalency by

Ouchter-lony gel diffusion tests with BSA (data not

shown). Extracts from[3S]methionine-labeled,

HSV-1-infected cells were applied to F(ab')2-BSA-Sepharose columns. As controls, equal

amounts ofextract were also applied to

BSA-Sepharosecolumns andto

IgG-BSA-Sepharose

columns which had been prepared with either

untreated IgGorIgG subjected tothe reaction

conditionsused forpepsindigestion (inthe

ab-senceofprotease). The resultsshown inFig.4

illustratethatmarkedlylessof theIgG-binding

proteinwasretainedonthe

F(ab')2-BSA-Seph-arose column compared with retention by the

IgG-BSA-Sepharosecolumn.Thissuggests that

the bindingis mediated through the Fcregion

oftheIgGmolecule.

VOL. 32,1979

http://jvi.asm.org/

(6)

784 AND SPEAR

.c>,

.> e i ..

9g

l.

i 1 3C K

-Q4 K

68K

FIG. 4. Reduction in binding of the

HSV-1-in-ducedpolypeptidestotheaffinitycolumnsbyremoval

of the Fcregion ofIgG.Affinity columnswere pre-pared by the additiontoBSA-Sepharoseofuntreated

anti-BSA (IgG) orofanti-BSA which had been

di-gestedbypepsintoyieldF(ab')2 fragments[F(ab')2]

orhad beensubjectedtothesameconditions usedfor

proteolysis in the absenceofpepsin(IgG°).Anextract waspreparedfromHSV-I(HFEM)syn-infected

HEp-2cells that had been incubatedwith[35S]methionine from 3 to 18 h after infection, and aliquots were appliedtothe threeaffinitycolumnsandtoacontrol BSA-Sepharose (C)column. Thisautoradiogramof

anSDS-acrylamide gelslab shows thepolypeptides

boundtoand elutedfromeach columnalongwith the

polypeptidespresentin a sampleofthe

unfraction-ated extract (Ext). The filled and open triangles

indicate thepositionsoftwo IgG-binding

polypep-tides,asdescribed in thelegendtoFig.3.

Newly synthesized and stable forms of

the Fe-binding glycoprotein. Affinity

chro-matographyofextractsfrompulse-labeled

HSV-1-infected cellswasdone todetermine whether

newlysynthesized forms of the Fc-binding

poly-peptide(s) could bind to the IgG-antigen

col-umns and might perhaps differ in

electropho-retic mobility from theprocessed, stable forms.

Infectedcellswerepulse-labeledfor 5 min with

[3S]methionineat 6 hafter infection and then

harvested immediately or after 15, 60, or 120 min of continued incubation in nonradioactive medium. BothHEp-2and BHK cells were used for thisexperimentinorder to compare the

Fc-binding polypeptides produced bydifferentcells.

The results shown in Fig. 5 demonstrate that

three electrophoretically distinct polypeptides bound specifically to the IgG-BSA-Sepharose columns (control columns without IgG not shown).

Two of these forms, designated pEand gE1,

were detected immediately after the

pulse.

[image:6.504.264.456.71.262.2]

p~

FIG. 5. Newlysynthesizedandprocessed forms of

Fc-bindingpolypeptidesproducedinHEp-2cells and

BHKcells infected with HSV-1(HFEM)syn.At 6h after infection the cells were incubated with [35S

methioninefor 5min, and then extracts were pre-pared immediately (0) or afteran additional incu-bationofthe cellsin nonradioactive medium(for15, 60, or 120min). The extracts were applied to

IgG-BSA-Sepharosecolumns andtocontrol

BSA-Sepha-rosecolumns, and the boundproteinswereelutedfor electrophoretic analysis. The designations pE, E1,

andE2 indicate thepositionsofpolypeptides bound

specifically to theIgG-BSA-Sepharose columns but not tothecontrolBSA-Sepharosecolumns(controls

notshown).

Within 15minafter thepulseathirdform,gE2,

wasdetectable.By60 minpE had disappeared,

and by 120 min very little of gE1 remained.

These resultsareconsistentwiththepossibility

thatallthreeelectrophoretic formsarederived

from thesametranslationproduct and that

post-translationalprocessingoccurswith the

conver-sion ofpE to

gEI

and

gEI

togE2. The results

obtained with infected HEp-2 and BHK cells weresimilar, althoughpEand

gE,

producedin the BHK cellswere more easily recognized as distinct species. Because the apparent conver-sion ofgEltogE2is notcompleteeven after2 h, it seems likely that extracts of infected cells

labeled forlong periods of time wouldcontain

both ofthesespeciesand thatgEIisthe

faster-migrating Fc-binding glycoproteinseeninFig.3

and4(marked bytheopentriangles).

To testthepossibilitythat gE2 has amino acid sequences in common with pE and gE1, these

[3S]methionine-labeledspecieswereisolated by

affinity chromatography and preparative

elec-trophoresisandsubjectedtopartial proteolysis

by the procedure of Cleveland et al. (5). The

diffemoo

lo. k:...:..

4800ma so

http://jvi.asm.org/

[image:6.504.87.231.74.252.2]

(7)

partial proteolytic products obtained with

var-ious concentrations of the staphylococcal

pro-teaseV8areshown inFig.6. It canbe seen that,

at allconcentrations of enzyme tested, the

elec-trophoretic profile of peptides from the

pulse-labeled material (pE plus

gEj)

issimilar to that

ofpeptides from the moreslowlymigrating

sta-ble material (gE2), except that the profiles are

displaced to the extent that the uncleaved

poly-peptides differ in mobility. This suggested the

possibility that all the [3S]methionine-labeled

peptides obtained fromgE2 contained the

mod-ification responsible for the slower

electropho-reticmobility of intact gE2. If this were true, and

ifthemodification includes glycosylation, then

all themethionine-containingpeptides from gE2

should containcarbohydrateand would

proba-blyincorporate label fromradioactive

glucosa-mine. Theresults shown inFig.7demonstrate,

infact, thatallthe[3S]methionine-labeled

pep-tides derived from gE2 also incorporated label

from

["4C]glucosamine.

Expression of the Fc-binding

glycopro-tein on thesurfaces ofinfected cells. Other

investigators have quantitated the binding of

125I-labeled

IgG to HSV-infected cells to

inves-tigate the kinetics of appearance ofFc-binding

activity in the infected cell surface (18, 31). Sim-ilarexperimentswereperformedinour

labora-tory(results shown in Fig. 8). Fc-binding activity

became detectable bythis assaybetween3 to6

hafter infection and reachedamaximumat 24

h. Todetermine whether the Fc-binding

glyco-protein, detected by affinity chromatography,

appeared in the cell surface withsimilarkinetics,

the following experiment was done. Intact

in-fected cellswereiodinated with

"2I

bya

lacto-peroxidase-catalyzed reaction at 3, 12, or 24 h

afterinfection.Detergentextractsof these

iodi-nated cells were then preparedand applied to

bothIgG-BSA-Sepharose columns and control

BSA-Sepharose columns. Figure 9 shows an

electropherogram of the

"MI-labeled

polypep-tides that were boundto and eluted from the

IgG-containing columns; the triangles identify

twospecies that boundspecificallyto the

IgG-containing columns butnot to the control

col-umns(controlsnotshown).Thesetwo

polypep-tides, whoseelectrophoretic mobilitiesare

indis-tinguishable from those of

gE,

and gE2

(Fig.

5

and 9), werevirtually undetectable in extracts

prepared from cells at 3 h after infection,

whereas significant quantities were present in

the extractspreparedat 12 and24h.Asis also

evident fromFig.9,sampleseluted from theIgG

affinity columns sometimes contained

hetero-geneous labeled material whose

electrophoretic

mobilitywasgreater than that of themajor

Fc-binding species.Therelationshipof this material

togE remainstobedetermined.

Peptides from

Fc-binding

protein

1

5 P

C

P C

44Ih~ ~ ~ ~ ~ ~~~4

FIG. 6. Peptidesobtainedfr-omnewly synthesized (P)andprocessed (C) forms oftheFc-bindingglycoprotein afterpartial proteolysiswith various concentrations(0 to125 pg/sample) ofS.aureusproteaseV8. The pulse-labeledforms ofthe Fe-binding glycoprotein (P)arethespecies designated pEandElin the kegendtoFig.5, andthepulse-chatsedform (C) isE2. Thesepolypeptideswere isolatedforthisanalysisasdescribed in the

text.

0

P C

25 P

C

125 P

C

http://jvi.asm.org/

[image:7.504.107.390.392.609.2]

(8)

786 BAUCKE AND SPEAR

Peptides

from

Fc-binding

prore,

:.Fg aci g

4 _l

w-"I

FIG. 7. Peptidesobtainedfr-om[35S]methionine-labeled(aa)and[14C]glucosamine-labeled (g) Fc-binding glycoproteinE2afterpartial proteolysiswith various concentrations(0 to250pg/sample) ofS.aureusprotease

V8.

30

25[

x

20

0

-o0

02 15 U0

10

[

5

6 12 18 24 30

Hoursafterinfection

FIG. 8. Bindingof[125IJIgGto HSV-1(HFEM)syn-infected (filled circles) anduninfected (open circles) HEp-2 cells,as afunction oftimeafter infection or

mockinfection.Radioactivitywasquantitatedby

liq-uid scintillationspectrometry; the valuesgiven are

countsperminute boundper4x 1(1 cells.

The previously identified HSV-1 glycopro-teins gC, gB, and gD werereadily detected as

125I-labeled

bands in electropherograms of the unfractionated extractsprepared from the iodi-nated infected cells. The results presented in

Fig.9indicate thatsignificantlygreater

quanti-ties of theseproteinswerepresent inthesurfaces ofinfected cells at 12 and 24 h afterinfection than at 3h,as wasalso foundfortheFc-binding

polypeptides. Several other polypeptides

de-tected in the unfractionated extracts were

la-beledtoapproximatelythesameextent atboth

early and late times after infection andare

prob-ably cellular proteins, as they comigrate with

5I-labeled polypeptides from uninfected cells

(datanotshown).Two of theselabeled

polypep-tides (indicated bythe small arrows in Fig. 9)

are similar in electrophoretic mobility to the

virus-induced Fc-binding polypeptides gE, and

gE2 but are probably unrelated to them. The

contribution of

l2II-labeled

Fc-binding

polypep-tidestoradioactivityin theextractsis too

sniall

tobedetected in electropherograms of the

un-fractionatedextracts.

DISCUSSION

Thisreport describes the isolation and char-acterization of an Fc-binding glycoprotein

pro-duced in HSV-1-infected cells. Several

proper-ties ofthisglycoprotein, designated gE, suggest

that itisresponsibleforthepreviouslydescribed

Fc-binding activity ofintactinfectedcells(9, 11,

18, 19, 26,28, 29, 31, 32). (This Fc-binding

gly-coprotein is thefifth antigenically or

function-allydistinct HSV-1-induced glycoproteintobe described. Assignment of thealphabetic desig-nationgEconforms tothe nomenclature agreed uponbyagroup ofparticipantsatthe Herpes-virus Workshop held in August 1978 in

Cam-bridge,England.)First, gEbinds toIgGandthis

0 0

/Z

/.

./

O O. O

http://jvi.asm.org/

[image:8.504.123.409.75.274.2] [image:8.504.68.263.317.551.2]

(9)

HERPESVIRUS Fc-BINDING GLYCOPROTEIN 787

NP4O Column-bound

35S-

pany posttranslational modifications (6,27).The

extracts proteins meth finding that newly synthesizedaswellasstable 3 12 24 3 12 24 CBP forms ofgE could bind to the affinitycolumns

indicates that Fc-binding activitydoes not

de-pend on complete processing or complete gly-cosylationof theprotein.It remainstobe

deter-mined whether allforms of theFc-binding

pro-gC-

" > -130K tein,

including

the two forms detected on cell

gB- surfaces,interact thesamewaywithIgG.

- 94 K

Although

the

possibility

existsthat

gE

is an

_U-A ."_

**4

induced cellular gene product, it seems more

M-=

= < D ^-

68K

likely

that the

polypeptide moiety

of this

Fc-gD binding glycoprotein will prove to be of viral

43K genetic origin. First, gEisapparently produced

-43 K onlyafter infection of the cell linesanalyzedin

thisstudyand issynthesizedat atimewhen all

or mosthostprotein synthesisisinhibited.

Sec--- ond,theelectrophoreticmobilities of the various

FIG. 9. Presence and kineticsofappearanceofthe gE formsproducedinHEp-2cellsaresimilaror

Fc-binding glycoproteinon thesurfaces of infected identical to those produced in BHK cells.

Fi-cells. HEp-2 cells infected with HSV-1(HFEM)syn nally, recent studies (M. Para, R. Baucke, and

were labeled with

125I,

by aprocedure whichlabels p. G. Spear manuscript in preparation) have

only proteins exposedtothe extracellularfluid soP

thatg

is a

*constin

pof

thenvirio

<¢v¢vsw- * r *

_rd,

at 3 shown that

gE

iS a constituent of the virion

12, or24 hafter infection. Extracts were then

pre-pared andapplied toIgG-BSA-Sepharosecolumns envelope as well as of the

mfected

cell surface.

and to control BSA-Sepharose columns. This auto- It can be concluded from the publications

radiogramofanSDS-acrylamidegelslab showsthe cited in the firstparagraphof this section that

'25I-labeledpolypeptides bound to andeluted from a number ofdifferent HSV strains caninduce

IgG-BSA-Sepharose,aswellasthelabeledpolypep- Fc-bindingreceptors,suggestingthatexpression

tidespresentinsmallsamples oftheunfractionated of this receptor is not a peculiarity of a few

extracts.Also shown isaprofileof[3Slmethionine- strains. There are differences among HSV

labeledpolypeptideswhich bound to and were eluted strains, however, in the number of receptors

from

anIgG-BSA-Sepharosecolumn. detectable by "2I-labeled IgG binding to the

surfaces of infected cells. We choseto use

HSV-binding depends

upon

integrity

of theFcregion. l(HFEM)syn for the studies reported here, on

Second,

gE

is a constituent of theinfected cell the basis ofscreening several 1 and

HSV-surface.

Finally,

the time of its appearance in 2strains for

maximal

expressionof cell surface

the cellsurface correlates with the time ofap- receptors.Theresultsof this screening suggested

pearance of Fc receptors on intact cells. The that syncytial plaque variants of eitherHSV-1

association of

Fc-binding

receptoractivitywith orHSV-2expressedlarger numbers of

Fc-bind-amembrane

glycoprotein

producedafter infec- ing receptors than did wild-type strains (R.

tion is consistent with

trypsin

sensitive and that itsexpres- remainsto be determined whether the

expres-sion is

prevented

by deoxyglucose

andbyinhib- sion of Fc-binding activity is related to

virus-itorsof

protein synthesis

(9, 18, 31). induced cell fusion and whether the apparent

Although

Fc-binding activity

wasfoundtobe differences in numbers ofreceptors aredue

pri-associated with three

electrophoretically

distinct

manly

tovariabilityin theamountofFc-binding

polypeptides,

ranging

in apparent molecular glycoprotein produced, its intracellular

distri-weight

from 65,000 to 80,000, it appears that bution,oritsbindingcharacteristics. In

prelim-these

polypeptides

shareamino acidsequences inary studieswehavefound that theFc-binding

in common.Both the kinetics of appearance of glycoproteininducedby HSV-2(GP)is indistin-each

Fc-binding polypeptide

after a pulse of guishable from theHSV-1(HFEM)syn glycopro-radioactive amino acid and theresults ofcom- teinbytwocriteria,electrophoretic mobilityof

parisons by partial

proteolysis suggest that a intactSDS-solubilizedproteinsand mobilities of

single

translation

product

yieldsallthree forms peptides obtained afterdigestion with V8

pro-by sequential

stages of

posttranslational

proc- tease (BauckeandSpear, unpublished data).

essing.

Previous studies revealed that theHSV- The molecule reportedto beresponsible for

1

glycoproteins gB,

gC, and gD are processed cytomegalovirus-induced Fc-binding activity similarly to

gE,

at least with respect to the (23) differs in at least two respects from the

changes

in

electrophoretic mobility

thataccom- HSV-1-inducedFc-binding gE. First,the

appar-32,1979

http://jvi.asm.org/

[image:9.504.47.242.76.255.2]

(10)

ent molecular weight of the

cytomegalovirus-inducedFc-bindingglycopolypeptide is43,000 as

comparedwith65,000to80,000forthe various

forms of gE. Second, the

cytomegalovirus-in-duced glycoprotein was isolated from cellular

extractsprepared without the use ofdetergents (and therefore may not have been membrane

bound), whereasHSV-1gEcouldnotbe

solubi-lized withoutdetergents.Itwill be of interestto

determine whether detergents can solubilize

other forms of the cytomegalovirus-induced

Fc-binding glycoprotein and to compare the

propertiesof theFc-binding proteinsinducedby

HSV,cytomegalovirus,andvaricella-zostervirus

(see referencecitations inIntroduction).

The function of the herpesvirus-induced

Fc-binding glycoprotein and its role in viral

repli-cation orpathogenesis,orinboth,remaintobe

determined. Thefinding thatFc-bindingactivity

canbe detectedonthe surfaces ofcells infected

by three different human herpesviruses

under-scores the possible importance of this activity.

The fact that detection ofthe binding activity

hassofar reliedontheuseofIgG,however, does

not necessarily imply that the purpose of the

virus-induced receptoris to interact with IgG.

Anunderstandingof the function of this

recep-tor,which in the caseof HSV-1 is expressedon

thesurfaces of both virions(Paraetal., in

prep-aration) and infected cells, will depend upon

additional experimentation, including

explora-tion of thepossibilitythat the receptor mayhave

affinity for molecules other than IgG.

ACKNOWLEDGMENTS

This workwassupported bygrant VC125 from the Amer-ican Cancer Society. P.G.S. is recipient of Public Health Service Research CareerDevelopmentAward5K04-CA00035 from the National Cancer Institute.

ADDENDUM IN PROOF

AttheHerpesvirusWorkshopheld in August1979

(ColdSpringHarborLaboratory,ColdSpring Harbor, N.Y.),agroupof theparticipants agreedupona con-ventionfor thenamingof the variousprocessed forms ofasingle HSVglycoprotein.Accordingto this con-vention,pEshould bedesignated pgE(64),gE,should bepgE(66), andgE2 should be gE. The numbers within parenthesesareapparentmolecularweights (x10-3),

usedtodifferentiate amongmultiple precursor forms.

LITERATURE CITED

1. Adler, R., J. C.Glorioso,J.Cossman,and M. Levine. 1978.Possible role ofFcreceptors oncellsinfectedand transformed by herpesvirus: escape from immune cy-tolysis. Infect.Immun.21:442-447.

2. Bonner, W.M., and R. A. Laskey. 1974. A film detection method fortritium-labeled proteinsand nucleic acids in polyacrylamidegels.Eur.J.Biochem. 46:83-88. 3. Byrt, P., and G. L. Ada. 1969. An in vitro reaction

betweenlabeledflagellinorhaemocyaninand

lympho-cyte-like cells from normalanimals.Immunology17: 503-516.

4. Cassai,E., R.Manservigi,A.Corallini,and M.Terni. 1975/76. Plaquedissociation of herpes simplex viruses: biochemical andbiologicalcharacters ofthe viral var-iants.Intervirology 6:212-223.

5.Cleveland, D. W., S. G. Fischer, M. W. Kirschner, andU. K. Laemmli. 1977.Peptide mapping bylimited proteolysis insodium dodecyl sulfate and analysis by gel electrophoresis. J. Biol. Chem.252:1102-1106. 6. 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-weight envelopeglycoprotein. J. Virol.27:172-181. 7. Costa, J.C.,and A. S.Rabson. 1975.Roleof Fc

recep-torsinherpessimplexvirusinfection. Lanceti:77-78. 8. Costa, J., A. S.Rabson,C. Yee, and T. S.Tralka. 1977.Immunoglobulin binding to herpes virus-induced Fcreceptors inhibits virus growth. Nature (London) 269:251-252.

9. Costa,J., C. Yee, Y.Nakamura, and A. Rabson. 1978. Characteristics of the Fc receptor induced by herpes simplex virus.Intervirology 10:32-39.

10.Fairbanks, G.,T.L.Steck,and D. F. H.Wallach.1971. Electrophoreticanalysisof themajorpolypeptideof the humanerythrocytemembrane.Biochemistry 10:2606-2617.

11.Feorino,P.M.,S. L Shore,and C. B. Reimer. 1977. Detectionby indirectimmunofluorescenceof Fc recep-tors incells acutely infected with herpes simplex virus. Int. Arch.AllergyAppl.Immunol.53:222-233. 12.Heine,J.W.,R. W.Honess,E.Cassai,and B.

Roiz-man.1974.Proteinsspecified by herpes simplexvirus. XII. The virionpolypeptidesoftype 1 strains. J. Virol. 14:640-651.

13.Hubbard, A. L., and Z. A. Cohn. 1972. The enzymatic iodination of the red cell membrane. J.CellBiol.55: 390-405.

14.Jensenius, J.C.,and A. F.Williams.1974.The binding ofanti-immunoglobulin antibodies to rat thymocytes and thoracic ductlymphocytes.Eur.J. Immunol. 4:91-97.

15. Keller, R., R.Peitchel,J. N.Goldman,and M. Gold-man. 1976.AnIgG-Fcreceptorinducedin cytomega-lovirus-infected human fibroblasts. J. Immunol. 116: 772-777.

16. Lehner, T.,J. M. A.Wilton,and E. J.Shillitoe.1975. Immunological basis for latency, recurrences, and pu-tative oncogenicity ofherpessimplexvirus. Lancet ii: 60-62.

17. Manservigi,R., P. G.Spear,and A.Buchan. 1977. Cell fusion inducedbyherpessimplexvirus ispromoted and suppressed bydifferentviralglycoproteins. Proc. Natl. Acad. Sci. U.S.A. 14:3913-3917.

18. McTaggart, S. P.,W. H.Burns,D.0.White,and D. C. Jackson. 1978. Fc receptors induced by herpes simplexvirus. I.Biologic and biochemical properties. J. Immunol. 121:726-730.

19. Nakamura, Y., J.Costa,T.S. Tralka, C. L. Yee, and A.S. Rabson.1978.Propertiesofthecellsurface Fc-receptor inducedby herpessimplexvirus.J.Immunol. 121:1128-1131.

20. Nisonoff, A.1964.Enzymatic digestion of rabbit gamma globulin andantibody and chromatography of digestion products. Methods Med. Res. 10:134-141.

21. Rahman, A. A., M. Teschner, K. K. Sethi, andH. Brandis. 1976.AppearanceofIgG(Fc)receptor(s) on culturedhumanfibroblasts infected with human cyto-megalovirus. J. Immunol. 117:253-258.

22. Ruyechan,W.T., L. S. Morse, D. M. Knipe, and B. Roizman. 1979.Molecular genetics ofherpes simplex virus.II.Mapping ofthemajor viralglycoproteins and ofthe genetic loci specifying the social behavior of

http://jvi.asm.org/

(11)

789

infectedcells. J. Virol. 29:677-697.

23. Sakuma, S., T. Furukawa, and S. A.Plotkin. 1977. Thecharacterization of IgGreceptor induced by human cytomegalovirus. Proc. Soc. Exp. Biol. Med. 155:168-172.

24. Sarmiento, M., M. Haffey, and P. G. Spear. 1979. Membrane proteins specified by herpes simplex viruses. III. Role of glycoprotein VP7(B2) in virion infectivity. J. Virol.29:1149-1158.

25. Sarmiento, M., and P. G. Spear. 1979. Membrane pro-teinsspecifiedbyherpessimplex viruses. IV. Confor-mationof the virion glycoprotein designatedVP7(B2). J. Virol. 29:1159-1167.

26. Shimizu, Y. 1971. Modification of host cell membrane after herpessimplex virus infection. Arch. Gesamte Virusforsch.33:338-346.

27. Spear, P.G. 1976. Membrane proteins specified by herpes simplexvirues.I. Identification of fourglycoprotein

precursors and their products in type1-infectedcells.J. Virol.17:991-1008.

28. Watkins,J. F. 1964.Adsorption of sensitizedsheep eryth-rocytes to HeLacells infected with herpessimplexvirus. Nature(London) 202:1364-1365.

29.Watkins, J. F. 1965. The relation ofherpes simplex haemadsorptionphenomenontothe virusgrowth cycle. Virology26:746-753.

30. Westmoreland,D., S. St. Jeor, and F.Rapp.1976.The development bycytomegalovirus-infectedcells of bind-ing affinity fornormalhumanimmunoglobulin.J. Im-munol. 116:1566-1570.

31. Westmoreland, D.,and J.F. Watkins. 1974. TheIgG receptor induced by herpessimplexvirus: studiesusing radioiodinatedIgG.J. Gen. Virol. 24:167-178. 32. Yasuda, J.,and F.Milgrom. 1968.Hemadsorptionby

herpessimplexinfectedcellcultures. Int. Arch.Allergy

33:151-170. VOL. 32,1979

http://jvi.asm.org/