Cell surface expression of the env gene polyprotein of dual-tropic mink cell focus-forming murine leukemia virus.

(1)

0022-538X/79/06-0720/09$02.00/0

Cell Surface

Expression

of the

env

Gene

Polyprotein

of

Dual-Tropic Mink

Cell

Focus-Forming

Murine

Leukemia Virus

NANCY G. FAMULARI* AND KAREN JELALIAN MemorialSloan-KetteringCancerCenter,NewYork,New York10021

Receivedforpublication20November1978

Differenceshave been observed in the kinetics ofprocessingofthe env gene

polyprotein of ecotropic, xenotropic, and dual-tropic mink cell focus-forming

(MCF)murine leukemia virus. Inpulse-chase experiments,theenvgene

polypro-tein of thedual-tropicMCF virusexhibitsamarked increase instabilityrelative to that of either ecotropic or xenotropic virus. A comparison of cell surface expression ofenvgene products ofecotropic,xenotropic, and dual-tropic MCF

murineleukemia virus has been made.Onlygp7O is accessibleto

lactoperoxidase-catalyzed radioiodination of fibroblastsinfected byecotropic orxenotropic virus, whereas both gp7O and the env genepolyproteinare expressedonthesurface of

dual-tropicMCF virus-infected cells.

The dual-tropic MCF (mink cell

focus-form-ing) class of murine leukemia virus (MuLV),

described by Hartleyand co-workers (13), has been shown tohave several interesting proper-ties attributable to the envelope (env) gene.

Boththe biologicalandbiochemical

character-istics of this virus class suggestan etiology in-volving recombination betweenanecotropicand

axenotropic MuLV (7, 9, 13, 26, 29), although

the occurrence of this recombinationwithin the

lifetimeofasinglemouse or as anevolutionary

event is not established. For example,

dual-tropic MCF viruses exhibit the host range of

bothN-ecotropicandxenotropicMuLV,are

in-terfered withbyboth virusclasses,andare

neu-tralized byantiseraspecifictoeither class (13).

Byserological analysis, it has been shown that

two gp7O-related cell surface antigens,

GRADA1

(inducedbyN-ecotropicvirusinfection; 23)and

GERLD

(induced by xenotropic

virus

infection;

29;

Y. Obata, personal communication), are

ex-pressed on fibroblasts infected by many

dual-tropic MCF isolates (29; P. V. O'Donnell,

per-sonalcommunication).

Analysis of the tryptic peptides of the gp7O species (9)andenvelope antigens (7, 29) of

dual-tropicMCFviruses has shownconsiderable

het-erogeneity among various isolates, particularly

among those dual-tropic MCF viruses isolated from different mouse strains. (The gag gene products, on the other hand, appearto beclosely related to those of N-ecotropic virus by both

tryptic peptide mapping [9]andantigenic

anal-ysis [7; P. V.O'Donnell, E.Stockert,Y. Obata,

A. B.DeLeo, H. W. Snyder, Jr.,and L. J. Old,

personalcommunication).

Characterization of RNase Tl-resistant oli-gonucleotides of certain isolates has demon-strated thatwhereasthe 5' portions of the

dual-tropicMCF virus genomes are very similar and

related to the N-ecotropic viruses Akv-1 and Akv-2, the 3' portions appear to be only partially sharedamong the dual-tropic MCF viruses and tobe different in large part from those of the Akv-1 and Akv-2 viruses (26).

Thesignificanceofthese envgene properties

inrelationshiptothebiological properties of the

dual-tropic MCF class of virus in vivo is

un-known. It is clear that certain isolates derived from late preleukemic and leukemic thymuses of AKR mice will induceamplifedexpression of MuLV-related cell surface antigens (16, 17) in

thethymusof young AKR mice (E. Stockert, J.

W. Hartley, P. V. O'Donnell, W. P. Rowe, Y.

Obata,and L. J. Old, personal communication)

aswellas acceleratingthedevelopment of

leu-kemia in AKR mice (7, 22; M. Cloyd, J. W. Hartley, W. P. Rowe, E. Stockert, P. V.

O'Donnell, andL. J.Old,personal

communica-tion). We have undertaken to examine the virus-hostinteraction of MCF247virus infibroblasts

of both mouse and mink origin to ascertain

whether any unusual features of protein proc-essing exist. Emphasis has been placed on the processing of the env gene products.

MATERIALS AND METHODS

Cells and viruses. Producer lines of MCF 247

virusisolatedfrom thethymusofa6-month-oldAKR

mousebyHartleyet al. (13)were constructedin the

SC-1mousecell line(12)andinthe CCL64minklung cellline(14)byP.V.O'Donnellof thisCenter.

720

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

The ecotropicvirus 69E5andthe xenotropic virus

69X9 usedinthis studywerebothisolated from the

thymusof the same6-month-oldAKR mouse,cloned, and usedtoinfectSC-1 cells and CCL64 mink cells, respectively (P. V. O'Donnell, personal communica-tion). Cultures were maintained as previously de-scribed(24).

Pulse-chase labeling of cell cultures. Pulse-chaselabeling of cell cultureswascarriedout as pre-viously described (10) using 40,Ci of "C-labeled

amino acid mixture(NewEnglandNuclearCorp.)per ml of minimal essential medium containing 10% of normal amino acid concentrations. Cellextracts were

preparedby scraping monolayers intocoldlysisbuffer (0.02 MTris[pH 7.4], 0.05 M NaCl, 0.5% Nonidet

P-40[ShellOilCo.], and 0.5% deoxycholate [34]),

blend-ing thecelllysate withaVortex mixer for1min, and centrifuging the lysateat1,500xgfor20minat4°C. The clarifiedsupernatantswereusedascellextracts.

Virus was purified by isopycnic centrifugation as pre-viously described (11).

Radiolabelingof cell cultures with

[3H]gluco-samine. Confluentmonolayerswereincubated for16

h inDulbecco-modified Eagle minimalessential me-diumcontaining 5% fetal calfserum and70,Ci of

D-[6-3H]glucosamine (New England Nuclear Corp.) per

ml. Cell extracts and purified virus were prepared as described above.

Radiolabelingof cell surface withiodine-125.

Confluentmonolayers werewashed oncewith phos-phate-buffered saline (calcium and magnesium-free)

andscrapedinto 0.05% EDTA inphosphate-buffered

saline. Cells were pelleted at 300 x g, washed once with cold phosphate-buffered saline, and resuspended incold phosphate-buffered salineat a concentration of15 x 106 cells per ml for SC-1 cells and 35 x 106

cellsper ml for CCL64 mink cells. Cellsurface

iodi-nationby the lactoperoxidase methodwascarriedout ina1-ml reaction volume using2mCi of1251I(30).Cell

extracts werepreparedasdescribedabove.

Serological procedures. Goat

anti-Rauscher-MuLV (R-MuLV) gp7O serum was provided by R.

Wilsnack, HuntingtonResearchCenter. Rabbit

anti-R-MuLVp15(E) serum wasprovided by E. Fleissner of this Center (15). This serum has been shown to

havereactivity to the virus structural proteinp15(E) (15) and to the cellular protein LETS (A. Pinter,

personal communication),butnot togp7O(25).

Immunoprecipitations of cellextractswerecarried

out aspreviously described (10) with the modification thatcellextracts weremade 0.5% for sodiumdodecyl sulfate (SDS) immediately prior to addition of the antiviralserum.

SDS-PAGE. SDS-polyacrylamide gel electropho-resis(PAGE)analysis of virus and immunoprecipitates

wascarried out using modifications of the procedure ofLaemmli (18). Electrophoresis using 7.5% acrylam-ide slab gels has been previously described (10). Step gradient acrylamide slab gels (1.5 mm thick) were prepared by layering 8 ml of 12.5% acrylamide, 8 ml of 10%acrylamide, and 6 ml of 7.5% acrylamide (a ratio of 30:1.04, acrylamide to bisacrylamide, was used). Stacking gels of 4.5% acrylamide were 1 cm long. Electrophoresis was carried out at 25 mA per gel until

acytochromecmarker reached the bottom of thegel.

Slabgelswerefixed witha50%methanol-10%acetic acid solutionfor20minanda5% methanol-10% acetic acidsolution for30min(6). Thefluorographic proce-dure of Bonner and Laskey (5) wasapplied to gels containing `C- and3H-labeled proteins. Autoradiog-raphyondriedgelswascarriedoutwith Kodak XR-2 X-rayfilmat-70°C.Gels containing'25I-labeled pro-teinswereautoradiographedwith Cronexintensifying screens,alsoat-70°C.

Peptide mapping by limited proteolysis and

analysis bySDS-PAGE. Radiolabeled proteins

iso-lated from cell extractsby immunoprecipitation and SDS-PAGE (see above) were subjected to limited

proteolysiswithStaphylococcus aureusV8protease

(Miles Laboratories) accordingtoClevelandetal. (6). Protease peptides were analyzedwith modified La-emmli (18) acrylamideslab gels (mapping gels)

con-sisting ofa15%acrylamide separating gel (acrylamide

to bisacrylamide, 30:0.8) of 15-ml volume (1.5 mm

thick; approximately6 cmlong)anda4.5%acrylamide

stackinggel (acrylamide to bisacrylamide, 30:0.8) of 10-ml volume (1.5 mm thick; approximately 3 cm

long).Both thestackingandseparatinggelscontained

1mMEDTA.

Immunoprecipitated, radiolabeled proteins were

visualized byfluorography (5) for l4C-or3H-labeled proteins,orby autoradiography withaCronex inten-sifying screenfor '"I-labeledproteins, and cut from dried polyacrylamideslabgels. (Isolationofproteins fromgelswhichhad beensubjectedtothe Bonner and

Laskey method [5] didnot affect the results of the

mapping procedure [unpublished data]). Acrylamide sliceswererehydratedbyshakingat room

tempera-turein10mlof 0.125 M Tris(pH6.8), 0.1%SDS,and

1mMEDTA, and loaded into samplewells of mapping gel.Sliceswereoverlaid with10plof buffercontaining

0.125MTris(pH 6.8), 0.1%SDS, 1mMEDTA,0.5%

thioglycolate, and 20%glycerol and 10ul of solution

containing 21 U ofS. aureusV8 protease per ml in

0.125M Tris (pH 6.8), 0.1%SDS, 1mMEDTA, and

10% glycerol. Electrophoresis wascarried out at

25-mA/gel constant current until a bromophenol blue

markerwasapproximately2 cminto thestacking gel. Currentwasthen turnedoff for30min,after which

electrophoresis was resumed until a cytochrome c

marker(loadedinablankwell)reached the bottomof

theseparatinggel.Gelswerefixedasdescribed above

andsubjectedtofluorography (5)for14C-or3H-labeled

peptides,orautoradiographywith Cronexintensifying

screensfor "25I-labeledpeptides.

RESULTS

Acomparison ofthekineticsof

cleavage

andmobilityinSDS-PAGEof theenvgene

products of ecotropic, xenotropic, and

dual-tropic MCF viruses. Differences were

observed in the kinetics ofcleavageaswellasin

the mobility in SDS-PAGE of the env gene

productsofecotropicvirus69E5, xenotropic vi-rus69X9,andMCF247virus. env geneproducts

were examined from extractsofcultures

http://jvi.asm.org/

(3)

722 FAMULARI AND JELALIAN

labeled with"C-aminoacid mixture (New Eng-land NuclearCorp.) ina15-minpulseorin a 15-minpulse followedbya3-h chase. Goat

anti-R-MuLVgp70serum wasusedtoidentifyenvgene

products by the doubleantibody immunoprecip-itation procedure (Fig. 1). In pulse-labeled

ex-tractsof cells infected withany oneof thethree

viruses, only theenv gene polyprotein (PrENV protein) was recognized (Fig. 1A, track4; Fig. 1B,tracks 1, 3,5).[Rabbit anti-R-MuLV p15(E)

serum also recognized this species (data not

shown).] During the 3-h chase period, some

cleavage of the PrENVproteintothegp7O

spe-cies tookplace in all the infectedcultures, but

the extent ofcleavage varied amongthe three

virus classes. The PrENV protein of the

xeno-tropic virus 69X9 was almostcompletely

proc-essed duringa 3-h chase (Fig. 1B, tracks5and 6). Scanning densitometer tracingsof autoradi-ographs ofgels fromtwoexperiments, including

the one shown in Fig. 1, indicate cleavage of

approximately 90% of PrENV of 69X9 in a 3-h chase. Cleavage ofapproximately 70 to 75% of PrENV of 69E5 occurred during a 3-h chase (Fig. 1B, tracks 1 and 2), consistentwith previ-ously reported data (10). However, little

cleav-age oftheMCF 247 virusPrENV protein took

place during a 3-h chase; approximately 30 to

40% of PrENV of MCF 247virus-infectedSC-1 cells (Fig. 1A, tracks4 and5) and 15 to20% of MCF 247 virus-infected mink cells (Fig. 1B,

tracks 3 and 4) were cleaved duringthischase

period.

U

-J. VIROL.

Themobilitiesof the env gene products of the

three virustypes areclearly distinguishable by

the gelsystemdescribed inFig. 1. The PrENV

protein (Pr80env) and gp7Oof69E5have higher

apparent molecular weights than the PrENV

proteinsorgp7O's of69X9 and MCF247virus. The mobilities of the env gene polyprotein of

the latter two viruses are not distinguishable, whereas the gp7O species of MCF 247virus

mi-gratesslightly faster than that of thexenotropic

virus. (The PrENV protein of 69X9 and MCF

247 virus will be referred to as Pr76env since it

has an apparent molecular weight of 76,000.)

Whether the differences observed in the appar-entmolecular weights of the PrENV andgp7O

proteinspeciesare areflectionofdifferencesin

thelength of theirpolypeptidechainor a reflec-tion of thedegree of glycosylation isnotknown.

To analyze further the stability of Pr76env of MCF247virus, apulse-chase experiment using chase periods of 2, 4, 6, 8, 10, and 12 h was

conducted. The data presented in Fig. 2A are

the result of analysis by immunoprecipitation with goat anti-R-MuLV gp7O serum of SC-1

cellsinfectedby MCF247 virus; similar results

were obtained when mink cells infected by this viruswereanalyzed (datanotshown). For

com-parison,Fig.2B shows ananalysis of SC-1 cells

infected by 69E5 virus. When several chase

pe-riods were examined, the PrENV protein of

MCF 247 virus appeared to be considerably

morestablethan that of 69E5virus,confirming the results presented in Fig. 1. However, the

.

FIG. 1. SDS-PAGEanalysis of immunoprecipitates ofpulse-chase-labeled MCF, ecotropic, and xenotropic virus-infected cells. Cytoplasmic extracts werepreparedfrom cultures pulse-labeled for 15min with 14C-labeled amino acidsorpulsed-labeledandchasedfor3h.Equalquantities ofprotein fromeach extract were reactedwithgoatanti-R-MuLVgp7Oserum, andequal quantitiesoftheresultingprecipitatesweresubjected toelectrophoresis in 7.5% acrylamide slabgels at25 mApergel (constant current). Electrophoresis was terminated1hafter bromophenol blueranoffthegel. (A) Track1: Uninfected SC-I cells pulse-labeled for15 min; track 2: uninfected mink cellspulse-labeled for15 min; track 3: MCF247virus labeledwith

[3H]-glucosamine;track4:MCF247-infectedSC-I cells pulse-labeled for15min;track 5:MCF247-infectedSC-I

cells chasedfor3h.(B) Track1:69E5-infected SC-1 cells pulse-labeled for 15 min;track 2:69E5-infected

SC-1 cells chasedfor3h; track 3: MCF247-infectedmink cellspulse-labeled for 15min; track 4: MCF

247-infected mink cells chasedfor3h;track5:69X9-infectedmink cellspulse-labeledfor15min;track 6:

69X9-infected minkcells chasedfor3h.

http://jvi.asm.org/

[image:3.505.71.458.425.553.2]

(4)

2 3 4 5 6 7 8 9

lw1 4

'Sm,do -ww

- :- _F~ _Wm

2 33 4 5 6

*7c. I,L

s...-cpC-M,-'-F

247I

_m

am

.__..

_fm "'. ::

7-869v

6 0-gp7 -69E5

FIG. 2. SDS-PAGE analysis ofimmunoprecipitatesofpulse-chase-labeledMCF247and69E5virus-infected

SC-I cells.Cytoplasmicextractswereprepared from infectedoruninfectedSC-I cellspulse-labeled for15min

with"4C-aminoacids and chasedforvarious timeperiods. Equal quantities ofprotein fromeachextractwere reacted withgoatanti-R-MuLVgp70serum,andequal quantities oftheresulting precipitateweresubjected

to electrophoresis in 7.5% acrylamide slabgels at25 mApergel (constant current). Electrophoresis was

terminated I hafter bromophenolblueranoffthegel. (A)Track 1: Uninfectedcellspulse-labeled for15min;

track 2: MCF247-infectedcellspulse-labeled for15min;track 3:MCF247-infectedcells chasedfor2h;track 4:MCF247-infectedcells chasedfor4h;track 5: MCF247-infectedcells chasedfor6h;track 6: MCF

247-infectedcells chasedfor8h;track 7: MCF247-infectedcells chasedfor10h;track 8: MCF247-infectedcells chasedfor12h;track11: MCF 247 virus labeled with[3H]glucosamine. (B)Track 1: Uninfectedcells

pulse-labeledfor15min; track 2:69E5-infectedcellspulse-labeled for15min;track 3: 69E5-infectedcells chased

for2h;track 4:69E5-infectedcells chasedfor4h;track 5:69E5-infectedcells chasedfor6h;track 6:

69E5-infectedcells chasedfor12h;track 7: 69E5 virus labeled with[3H]glucosamine.

kinetics ofappearanceand disappearance ofthe

gp7Ospecies of MCF 247 and of 69E5werevery

similar. Thehighest intracellular concentration

ofthis cleavage product was found in the 2-h

chase, witharoughly linear decay insubsequent

chaseperiods. (gp7O of bothMCF247and69E5

is present in a 1-h chase, the shortest chase

period examined [data not shown].) The fact

that thestability of Pr76env ofMCF247virus is

not accompanied byadelay in the appearance

ofthe gp7Ospeciessuggeststhatonlya portion

of the PrENVproteinmaybeavailable for

cleav-ageinto the viralenvelope glycoproteinand that

that fractionofthePrENVprotein poolis

proc-essed with kinetics similar to those governing

thecleavageof the PrENVproteinofecotropic

virus. Examination of the ratio of the PrENV

protein to its cleavage product shows that

Pr8env and gp7O of 69E5 virus are in roughly

equivalentamountsinthe 2-hchase,withgp7O

becoming the majorintracellularenvgene

prod-uct in subsequent chases, whereas Pr76env of

MCF247 virus is thepredominantintracellular

envgeneproductinall chaseperiodsexamined.

A

B

7

http://jvi.asm.org/

[image:4.505.98.390.74.399.2]

(5)

Attemptstomeasurethekinetics of

incorpo-ration ofgp70 ofMCF 247 into released virus

wereunsuccessful becausesucrose

density-puri-fied MCF virus exhibits erratic retention ofthis

molecule. However, analysisof theincorporation ofp15(E) into releasedvirusindicatedthat this

cleavageproduct of the PrENV proteinis

assem-bled into virus with kinetics similar to those

governing the incorporationofp15(E)into 69E5

virus. [It has been shownpreviously thatgp7O andp15(E) ofthe N-ecotropicvirusWN1802N

areincorporatedinto virions with thesame

ki-netics[10].)

Identification of the PrENV protein of

MCF 247 virusonthecellsurface.Envelope geneexpressiononthe

cil

surface of MCF 247

virus-infected SC-1 andmink cellswasexamined

by themethod oflactoperoxidase-catalyzed

ra-dioiodination. SDS-PAGE analysisof

immuno-precipitates of cell extracts reacted with goat

anti-R-MuLV gp7Oserumwascarriedout(Fig. 3). Goat anti-R-MuLV gp7O serumrecognized

twovirus-specific speciesin extracts ofMCF 247

virus-infected mouse orminkcells: gp7O anda

higher-molecular-weight species (Fig. 3, tracks

3 and 4), whereas the same serum recognized

only gp70 in extracts of radioiodinated 69E5 virus-infected SC-1 cells and 69X9 virus-infected

minkcells (Fig.3,tracks 1and2).

Inan efforttoidentify the secondenv

gene-related cellsurface protein found onMCF 247

virus-infectedcells,immunoprecipitatesof

radi-oiodinatedcells weremadewith rabbit

anti-R-MuLVp15(E)serum(Fig. 4). Itwasfound that

this serum recognized the

higher-molecular-i A

weight species, indicating that this protein is Pr76envoracloselyrelated species (Fig.4,tracks

2 and 4). Precipitation of a minor amount of

gp7O by this serum is due to trapping. The

protein near the top of the gel labeled "h" is found in infected and uninfected cells (Fig. 4,

tracks 2, 4, 6) and isprobably LETS protein (see Materials and Methods).

Analysis of Pr76efv of MCF 247 virus by partialproteasedigestion and SDS-PAGE

analysis. To determine the relationship

be-tween the env gene polyprotein of MCF 247

virus identified by pulse-labeling with radiola-beled amino acids and theenvgenepolyprotein

identified by radioiodination, the partial

pro-teasedigestmappingprocedure of Clevelandet

al. (6) wasused. The PrENVprotein and gp7O

of MCF 247 virus were isolated from infected

minkcellsby SDS-PAGE analysis of

immuno-precipitatesandtreatedwith S. aureasV8

pro-tease. Figure 5A shows representativemaps of

Pr76env radiolabeledina15-min pulse witha

14C-amino acidmixture (track 1) or radioiodinated

onintactcellsbythelactoperoxidase procedure

(track 2); amapofradioiodinatedgp7Oisolated

from a cellextract is included for comparison (track 3). Themapsofbiosynthetically labeled orradioiodinated PrENV proteinareessentially

identical anddistinguishable from those ofgp7O.

[Similarresults have beenobtained bymapping

theenvelopegeneproductsofMCF247isolated from SC-1 cellsand by mapping envelopegene

products isolated by immunoprecipitation with rabbit anti-R-MuLV p15(E) serum; data not

shown.] These data indicate thatthere areno

5 6

Vk:

FIG. 3. SDS-PAGEanalysisofgp7Orelatedspeciesonthecellsurface ofradioiodinated MCF 247, 69E5,

and69X9virus-infectedcells.Cytoplasmicextracts werepreparedfromintact cells which had beensubjected

to lactoperoxidase-catalyzed radioiodination. Extracts werereacted with goat anti-R-MuLVgp7O serum,

and theresultingprecipitatewasanalyzedon astepgradientpolyacrylamideslabgel(7.5%,10%, and12.5%

acrylamide).Electrophoresiswascarried outat25mA pergel(constant current) until acytochromecmarker reached the bottomof the gel. Track1:69E5-infectedSC-Icells; track2:69X9-infected mink cells; track3:

MCF247-infectedminkcells;track4:MCF247-infectedSC-Icells; track5:uninfectedminkcells; track6:

uninfectedSC-I ceUs. Note that the stepgradientslabgel doesnotclearly resolve themigrationdifferences

inthegp7Ospecies of 69E5, 69X9, and MCF247virusesdemonstrated in Fig. 1(7.5% slab gel).

http://jvi.asm.org/

[image:5.505.120.415.452.581.2]

(6)

1 2 3 4

NWOp_ 5 6

-h

Pr76ep

gp7O-MCF 247

FIG. 4. SDS-PAGEanalysisofgp7Oandp15(E) relatedspeciesonthesurface of radioiodinatedMCF 247 virus-infected cells. Cytoplasmic extracts wereprepared from intact cells which had been subjected to

lactoperoxidase-catalyzedradioiodination. Extractswerereacted with eithergoatanti-R-MuLVgp70serum

or rabbit anti-R-MuLVp15(E) serum, and the resultingprecipitates were analyzed on a stepgradient

polyacrylamideslabgel (7.5%, 10%, 12.5%acrylamide). Electrophoresis was carriedout at25mApergel

(constant current)until acytochromecmarker reached the bottomofthegel. Track1:MCF247-infected

SC-Icellsreacted withanti-gp7Oserum; track 2:MCF247-infected SC-Icells reacted withanti-pl5(E)serum;

track 3:MCF247-infected mink cells reacted withanti-gp70 serum; track4:MCF247-infectedmink cells reacted with anti-p15(E) serum; track 5: uninfected SC-I cells reacted with anti-pi5(E) serum; track 6:

uninfectedminkcells reacted withanti-pl5(E)serum.SeeFig.3for anti-gp7Oserumreactionwithuninfected

SC-Iandminkcells.

majordifferences between thenewly synthesized

PrENV protein and the env gene polyprotein

expressedonthe cell surface. To examine further

thequestion of possible modification ofPr76env

inthe process of itsbeing expressed on thecell

surface, a comparison was made between the

maps ofPr76elvisolated from infectedcellsafter

a 15-min pulse and Pr76env isolated from cells

which had been pulse-labeled for 15 min and chased for 5 h (Fig. 5B, tracks 1 and 2). No

differences in the maps of the PrENV protein

radiolabeled by these two protocols were ob-served. Furthermore, the map of

Pr76ent

isolated

fromcellsafter a 16-h exposure to

[3H]glucosa-mine (Fig.5B, track 3) was very similar to that

of

"4C-amino

acid-labeled or radioiodinated

Pr76env. The major difference is the absence of

bands "a" and "b" in maps of

[3H]glucosamine-labeled

Pr76env,

indicating thatthese fragments do not containglucosamine. Since the majority

offragments analyzed by this method are

gly-cosylated, it islikely that modifications of the

PrENVprotein involving changes in the degree

ofglycosylation would result in changes in the

electrophoretic migration of some ofthe peptide fragments. Such altered migration has not been observed.

DISCUSSION

Identification ofthe PrENVprotein of MCF

247 virus on the cell surface ofinfected fibro-blasts indicates that the primary translational productofthe env gene is capable ofserving two

functions: (i)astheprecursortothe viral

struc-tural components gp7O andp15(E) (10,21, 35),

and (ii) as avirus-induced cell surface compo-nent. Dual functioning of an MuLV precursor

proteinisnotuniquetotheenvgenepolyprotein;

the gag gene polyprotein has previously been

shown to be the precursor to the virion core

proteins (1, 2, 27, 33), aswell as to anintegral

cell surface protein (19, 20, 28, 31, 32). The

controlling elements which allow the env and

gag genepolyproteinstofunctionasprecursors

tothevirion structuralproteinsaswellas

inte-gral cell surface proteinsare notyetunderstood,

butwefeel that it islikelythatdifferent mech-anisms govern the compartmentalization of thesetwogeneproducts. Inthe caseof the gag

gene product, the major precursor species,

Pr65Va9, is a nonglycosylated, cytoplasmic

pro-tein, the cleavage products of which are not

found external to theplasmamembrane (or

vir-ionenvelope) (1, 2, 27, 33). The cell surface gag

polyproteins, gP85Ya and gP95Ya, have been

shown to be glycosylated (20, 28, 31), a fact

reflectedin their estimatedmolecularweightsof

85,000 and 95,000.Specification ofaproteinas

cytoplasmic, ontheonehand, or as anintegral

membraneproteinorasecretoryprotein,onthe

other, has been hypothesized to reside in the

aminoacidscodedforbythe 5' end of a

partic-ular mRNA(signal hypothesis[3,4]).Thusit is

possible that the regulation of expression of

thesetwo forms ofthe gag gene product is at thelevel ofmRNAprocessing.

http://jvi.asm.org/

[image:6.505.112.404.80.214.2]

(7)

w'

i

2

...

2s

[image:7.505.64.257.76.499.2]

34

FIG. 5. Partial digestmapsofPr76envandgp7O of

MCF 247 virus. S. aureus V8protease was usedto

generatepartial digestmapsofPr76en"andgp70 of

MCF 247virus; proteinswereisolatedfrom infected

minkcellsbySDS-PAGEanalysis of

immunoprecip-itates made with goat anti-R-MuLVgp7O serum.

Partialcleavagepeptideswereresolvedon15%

acryl-amide slab gels run at 25 mA per gel (constant current) until a cytochrome c marker reached the

bottomofthegel. (A) Track 1:Pr76NWV radiolabeled

with "4C-amino acids in a 15-min pulse; track 2:

Pr76en' radioiodinatedonintact minkcells;track 3: gp7Oradioiodinatedonintact mink cells. (B) Track

1:Pr76env radiolabeled with14C-aminoacids ina

15-minpulse; track 2: Pr76eIn radiolabeled with

14C-amino acids in a 15-min pulse followed by a 5-h chase;track 3: Pr76env radiolabeled with

[3H]gluco-samine for 16 h; track 4: gp7O radiolabeled with [3H]glucosamine for16 h.

In the case of the env gene products, the PrENVprotein is eitherprocessedinto the viral membrane proteins gp70 and p15(E) or, as in the case of the PrENVproteinofMCF247virus,

is itself an integral membrane protein. (The

PrENV proteinofMCF247 virus does not ap-pear tobeincorporatedinto virus[Fig. 1A, track 3; Fig. 2A, track 9; unpublished data].) The mechanismresponsibleforsparingthecell

sur-face PrENVproteinfromcleavageintogp7O and p15(E) is unclear, but differentiation between cytoplasmic and cell surface localization is ob-viously not involved. We have been unable to

detect any modification of the PrENV species

usingpartial digest peptide mappingtocompare

pulse-labeled and pulse-chase-labeled PrENV,

orpulse-labeledand cellsurface-labeledPrENV.

Since the majority of the peptide fragments

examined by partial digest mappingare

glyco-sylated, it is likely thatwewoulddetect

migra-tion differencesin some fragmentsif

compart-mentalizationof the functional precursor

mole-cule and the cell surfacespeciesinvolvedmajor

differences inglycosylation.Itisinteresting that,

as is the case for the env gene products of

ecotropicMuLV(21),thegp7OofMCF247virus

contains fucose whereas the PrENVspeciesdoes

not (unpublished data). It is possible that the

intracellular stabilityof the PrENVspecies, i.e.,

its resistance to cleavageintogp7Oandp15(E),

mightbeareflection of the recombinantnature

of the env gene of theMCFclass of viruses. For

example, the Pr76env of MCF 247 virus might

representan unfavorable substrate for the

en-zyme(s) responsible forprocessing thisspecies.

The significance ofcellsurface expression of

PrENVproteinorthe gag genepolyprotein,for thatmatter, is notknown. gP95Vg andgP859`9 are known to carry the determinants of the

GCSA antigen (19, 28), a specificity to which

mice can respond. Whether or not the PrENV

protein carries antigenic determinants unique

from those expressed by its cleavage products,

gp7O and p15(E), is not known. It is known,

however, that leukemias of the AKR mouse

expressamoleculewhich, onpreliminary

anal-ysis, is very similar to the PrENV protein of

MCF 247 virus. Examination ofanin vitro-cul-tured AKR leukemia AK2D (8; J. L. Biedler,

personal communication) and an in

vivo-pas-saged AKR leukemia AKSL2 (29) have shown

expression ofaprotein ofapproximately 80,000

daltonswhich isimmunoprecipitated byantisera

recognizinggp7Oorp15(E).Partialdigest map-ping of this protein isolated from theAK2Dcell

line shows ittobevery similar butnotidentical

to that expressed on MCF 247-infected fibro-blasts (unpublished data).A similarspecies has

beenidentified onspontaneousAKRleukemias

A

E

a...

L.

J. VIROL.

http://jvi.asm.org/

(8)

(J.-S. Tung and E. Fleissner, personal

commu-nication).

An important question that remains to be

addressedisthat of the distribution of the phe-notypeof cell surface expression of the PrENV species amongvariousMuLVisolates. We have

examined other dual-tropic MCF virus isolates

derived from preleukemic and leukemic AKR

thymus tissue;these isolatesarepositivefor cell

surfaceexpression of PrENV (N. Famulari and

P. V.O'Donnell, unpublisheddata). However, it remains to examine dual-tropic MCF isolates derivedfrom mouse strains other than theAKR,

aswellas avarietyofecotropic,xenotropic, and

amphotropic MuLV from different sources, to

determine whether cell surfaceexpressionofthis

protein is MCF virus specific and whether its

expressioncorrelates with any knownbiological

propertiesof MuLV such ashost range or

leu-kemogenicity.

ACKNOWLEDGMENTS

WethankP.O'Donnell,E.Fleissner,E.Stockert,H.

Sny-der, andE.Tres formanyhelpful discussionsand for their criticalreadings of thismanuscript.

Thisworkwassupported byPublic HealthServicegrant

CA-15297from the National Cancer Institute.

LITERATURE CITED

1. Arcement, L J., W. L Karshin, R. B. Naso, G.

Jamjoom,andR. B.Arlinghaus.1976.Biosynthesis of Rauscher leukemia viralproteins: presence ofp30

andenvelope p15sequences in precursorpolypeptides. Virology69:763-774.

2. Barbacid,M.,J. R.Stephenson,andS. A. Aaronson.

1976.gag geneof mammaliantype-CRNA tumor vi-ruses.Nature(London)262:554-559.

3. Blobel G.,and B.Dobberstein.1975.Transfer of pro-teinsacrossmembranes.I.Presenceofproteolytically processed and unprocessed nascent immunoglobulin lightchainsonmembrane-boundribosomesof murine

myeloma.J.CellBiol. 67:835-851.

4. Blobel,G.,and B.Dobberstein.1975.Transfer of

pro-teinsacrossmembranes. II.Reconstitution of functional

rough microsomes fromheterologouscomponents. J. CellBiol.67:852462.

5. Bonner,W.ML, andR. A.Laskey.1974.Afilm detection method for tritium-labeledproteinsand nucleicacids in

polyacrylamide gels.Eur. J. Biochem.46:8348.

6. Cleveland, D., S. G. Fischer,M.W.Kirschner, and U. K.Laemmli.1977.Peptidemapping by limited

prote-olysis in sodiumdodecyl sulfate and analysis by gel electrophoresis.J.Biol. Chem.252:1102-1106.

7. Devare,S.G.,U. R. Rapp, G.J.Todaro, and J. R. Stephenson.1978.Acquisitionofoncogenicity by

en-dogenousmousetypeCviruses:effectsof variations in envandgaggenes. J. Virol. 28:457-465.

8. Dofuku,R.,J. LBiedler,B. A.Spengler, and L J. Old.1975.Trisomyof chromosome 15 inspontaneous leukemia of AKR mice. Proc. Natl. Acad. Sci. U.S.A. 72:1515-1517.

9. Elder, J. IL, J. W. Gautsch, F. C. Jensen, R. A.

Lerner,J. W.Hartley, and W. P. Rowe. 1977. Bio-chemicalevidence that MCF murine leukemia viruses areenvelope (env)gene recombinants. Proc. Natl.Acad. Sci.U.S.A. 74:4676-4680.

10.Famulari,N.G.,D.LBuchhagen,H.-D. Klenk, and E.Fleissner. 1976. Presence of murine leukemia virus

envelope protein gp70 and p15(E) in a common

poly-proteinofinfectedcells.J.Virol. 20:501-508. 11. Fleisner, E. 1971. Chromatographic separation and

an-tigenic analysis of proteins of the oncornaviruses. I. Avianleukemia-sarcoma viruses. J. Virol. 8:778-785. 12. Hartley, J. W, and W. P. Rowe. 1975. Clonal cell lines

from a feral mouse embryo which lack host-range

re-strictionsfor murine leukemia viruses. Virology 65:128-134.

13. Hartley,J.W., N. K. Wolford, L J. Old, and W. P. Rowe. 1977. A new class of murine leukemia virus

associatedwithdevelopment ofspontaneous lympho-mas.Proc.Natl.Acad. Sci. U.S.A.74:789-792. 14. Henderson,L.C.,M. Ll[ieber,and G. J. Todaro. 1974.

Minkcellline Mvl Lu (CCL64). Focus formation and

the generationof "nonproducer" transformed cell lines with murine and feline sarcoma viruses. Virology 60: 282-287.

15. Ikeda, H.,W. D.Hardy,Jr., E.Tress,and E.

Fleis-sner. 1975.Chromatographic separation and antigenic analysis of proteins of the oncornaviruses. V. Identifi-cation of a new murine viral protein,p15(E). J. Virol. 16:53-61.

16. Kawashima,K., H.Ikeda, J. W. Hartley, E. Stockert, W. P. Rowe,andL J.Old.1976.Changes in expres-sion of murineleukemiavirusantigens and production ofxenotropic virus in the late preleukemic period in AKR mice.Proc. Natl.Acad. Sci. U.S.A. 73:4680-4684. 17.Kawashima,K., H.Ikeda,E.Stockert, T.Takahashi, andLJ.Old.1976.Age-relatedchanges in cell surface antigens of preleukemicAKR thymocytes. J. Exp. Med. 14:193-208.

18. Laemmli,U. K. 1970.Cleavage of viral precursor protein in vivoand in vitro. J. Virol. 10:751-759.

19. Ledbetter, J., andR.C. Nowinski.1977.Identification of the grosscellsurface antigenassociatedwith murine

leukemia virus-infectedcells. J. Virol. 23:315-322. 20. Ledbetter, J., R.C.Nowinski, and S.Emery. 1977.

Viralproteins expressed on the surface of murine leu-kemiacells.J.Virol. 22:65-73.

21. Naso, R.B.,L J. Arcement, W. L Karshin, G.A.

Jamjoom, and R. B.Arlinghaus. 1976. A fucose-deficientglycoproteinprecursor to Rauscher leukemia virusgp69/71. Proc. Natl. Acad. Sci.U.S.A. 73:2326-2330.

22. Nowinaki,R.C.,and E. F.Hays.1978.Oncogenicityof AKRendogenousleukemiaviruses. J. Virol. 27:13-18. 23. Obata, Y.,E.Stockert,P. V.O'Donnell,S.Okubo,H. W.Snyder,Jr.,andL J.Old.1978.G(ADA1) a new cellsurface antigen of mouse leukemiadefinedby

nat-urally occurringantibody and its relationship to murine

leukemiavirus. J. Exp.Med. 147:1089-1105.

24. O'Donnell,P.V.,and E. Stockert. 1976. Induction of GlxantigenandGrosscell surfaceantigenafter infec-tionby ecotropicandxenotropic murineleukemia vi-ruses in vitro. J.Virol. 20:545-554.

25.Pinter, A., and E. Fleissner. 1977. The presence of

disulfide-linkedgp7-plS(E)complexesin AKRmurine

leukemia virus. Virology83:222-227.

26. Rommelaere,J., D. V.Faller, andN.Hopkins.1978.

Characterization andmapping ofRNaseT,-resistant oligonucleotidesderivedfromthegenomesofAKV and MCFmurineleukemia viruses. Proc. Natl. Acad. Sci. U.S.A. 75:495-499.

27. Shapiro, S. Z., M. Strand, and J. T. August. 1976. High molecular weightprecursorpolypeptides to structural

proteinsof Rauscher murine leukemiavirus. J. Mol. Biol. 107:459-477.

28. Snyder, H. W., Jr., E. Stockert, and E. Fleissner.

1977. Characterization ofmolecular species carrying Gross cell surfaceantigen. J. Virol. 23:302-314. 29. Stockert,E., A. B.Deleo, P. V. O'Donnell, Y. Obata,

andL. J.Old.G(AKS2):a newcellsurfaceantigen of the mouse related to the dual tropic MCF class of

http://jvi.asm.org/

(9)

728

MuLV detected by naturallyoccurringantibody. J. Exp.

Med. 149:200-215.

30. Tung, J.-S., P. V. O'Donnell, E. Fleissner, and E. A. Boyse. 1978. Relationshipsofgp7Oof MuLVenvelopes togp7Ocomponentsofmouselymphocyte plasma

mem-branes. J. Exp. Med. 147:1280-1284.

31.Tung, J.-S., A. Pinter, and E. Fleissner. 1977. Two

species oftypeC viralcorepolyproteinonAKRmouse

leukemia cells. J. Virol. 23:430-435.

32. Tung, J.-S., T. Yoshiki, and E. Fleissner. 1976. Acore

polyprotein ofmurineleukemia virusonthesurface of

mouseleukemiacells.Cell 9:573-578.

33. vanZaane,D., A. Gielkens, M. Dekker-Michielsen, and H. Bloemers.1975.Virus-specificprecursor poly-peptides in cells infected with Rauscher leukemia virus. Virology 67:544-552.

34. Vogt, V., and R. Eisenman. 1973. Identification ofa

large polypeptideprecursorof avianoncornavirus

pro-teins. Proc.Natl. Acad. Sci. U.S.A. 70:1734-1738. 35.Witte,0.N.,A.Tsukamoto-Adey,andL L

Weiss-man. 1977. Cellular maturation ofoncornavirus glyco-proteins: topological arrangement of precursor and product formsincellular membranes.Virology 76:539-553.

http://jvi.asm.org/