Viral protein synthesis in Friend erythroleukemia cell lines.

(1)

JOURNALOF VIROLOGY, Jan. 1977, p. 328-337

Viral Protein

Synthesis

in

Friend

Erythroleukemia

Cell

Lines

JANIS RACEVSKIS* AND GEBHARD KOCH

Roche Institute ofMolecular Biology,Nutley, New Jersey 07110

Receivedfor publication 2June 1976

Viral proteinsynthesis wasstudiedintwoFriendvirus-induced erythroleuke-mia cell lines(Ostertag celllines FSD1-F4and B8) by thetechniqueof immuno-precipitation withmonospecific antisera to the majorenvelope glycoproteingp7O andmajorcoreprotein p30. One of the cell lines(F4) releases active Friend virus complex to the growthmedium, where release of virus from the other cell line (B8) isbarely ornondetectable. It was foundthat in thenonproducercell line B8, alarge-molecular-weightprotein of about 65,000containingp30antigenic deter-minants issynthesized, yetnop30 is produceduponprolonged incubation and chase, suggesting that this might be the actual lesion that prevents mature virus production by these cells. In both cell lines, the predominant protein species that is immunoprecipitated with monospecific anti-gp7O serum is a protein of55,000 to60,000daltonsthat is labeled withglucosamineto amuch lesser extent thangp70 and appears to becomeheterogeneouswithtime. Large amounts of gp7O can be detected in the cell-free medium, but none of the unstable species of 55,000 to 60,000 molecular weight.

Friendvirus-transformed mouse erythroleu-kemic cell lines have attainedwidespread use as a model system for the in vitro study of

differentiation (5, 16).The Friend cells are

mor-phologically similar to proerythroblasts (19) and are presumably arrested at this stage of

differentiation by the virus-induced transform-ing event;furthermore,manyof thesecell lines

releasevirusparticles into the growth medium (14, 19, 25).

The cell line FSD1/F4 (F4) was established from the spleens ofDBA/2 mice infected with NB tropic BALB/c-adapted Friend virus

com-plex byOstertaget al. (16). Whilegrowing in

culture, these cells release biologically active Friend viruscomplex; thelymphaticleukemia

helper virus and the replication-defective erythroid cell-transforming spleen focus-form-ing virus (3). In those studies, the released virus was detected and characterized by elec-tron microscopy and by the XC assay (3). The presence of thespleenfocus-forming virus and helper virus was measured by spleen focus for-mation in DBA/2 and BALB/c mice (N andB type,respectively) after injection of filtered tis-sueculture supernatants.

A5-bromodeoxyuridine-resistantand thymi-dine kinase-negative (TK-) subclone of line FSD1/F4, clone B8, was found byOstertag et al. (17) to release 1,000- to 100,000-fold reduced amountsof Friendvirus, ascomparedwiththe TK+parental cell clone F4.Interestingly,

mor-phological studies by electron microscopy

re-vealed that the nonproducer B8 cells contained

agreat many intracisternal A particles,many

moresothan the producer F4 cells (14). Inspite

ofthese differences, however, both cell clones do share that characteristic property ofmost Friend cell lines: the capacity to be induced to differentiateuponexposure todimethyl sulfox-ide in the growth medium (3).

Wepresent inthisreport a characterization

of viral protein synthesis in the producer cell line F4 andthe nonproducer line B8, arrivedat by the application ofimmunoprecipitation tech-niques using monospecific antiserato purified viral structural proteins.

MATERIALS AND METHODS

Cell cultures. Frienderythroleukemia cell lines FSD1/F4 (F4) and B8 (3), weregenerouslyprovided

by W. Ostertag, Max-Planck Institut fuer

Experi-mentelleMedizin, Goettingen, West Germany. The

cells were grown in suspension cultures in Eagle

minimal essential medium (GIBCO, Grand Island,

N.Y.) supplemented with 10% fetal calf serum

(GIBCO), 2x (the concentration found in Eagle

minimal essential medium) glutamine, 1 x vita-mins, 1x nonessential amino acids, penicillin (100 U/ml), streptomycin (100 ,ug/ml), and gentamicin (25a.g/ml). The cells were found to be free of myco-plasma contamination as determined by the method of Hayflick (7).

Rauscher leukemia virus-infectedmouse3T3cells

werekindly provided byJ. A. Bilello, Albert Ein-stein College of Medicine, Bronx, N.Y. The cells were propagated in Dulbecco-modified Eagle me-328

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

dium supplemented with 10% fetal calf serum

(GIBCO), penicillin (100 U/ml), and streptomycin

(100 ,ugIml).

Reverse transcriptase assay for virus production. Culture medium was cleared of cells and debris by centrifugation at 2,000 xg for 10 min.Virions were concentrated bycentrifugationthrough a 20%

glyc-erol layer in 50 mM Tris (pH 7.8)-0.1 M KCl at

100,000 xgfor 90 min.Pellets were resuspended as 100-fold concentrates in 50 mM Tris-0.1 M NaCl-1

mM dithiothreitol. Just before assay, the concen-trate isdiluted, made 0.1% with Triton X-100, and sonicated. The assay is carried out in 0. 1-ml-volume reaction mixturescontaining 50 mM Tris-hydrochlo-ride buffer (pH 8.0), 50 mMKCl,8 mMMnCl2,5 mM

dithiothreitol, 0.02absorbance units of primer

tem-plate complex at 260 nm, oligo(dT)1218'poly (A),

0.1% Triton X-100, and 5 ,Ci of [3H]TTP (specific

activity, 53 Ci/mM, Schwarz/Mann). Reactions are initiated by the addition of[3H]dTTPand are run for 30min at 37°C, atthe end of which time the reaction isstopped by the addition of 0.5 ml of a 0.1 M sodium

pyrophosphateand then 0.5 ml of cold 25%

trichloro-acetic acid. The mixture is kept at 4°C for 30 min, andthen the precipitates are collected on glass-fiber

filters (Whatman GF/C)andwashed.The filters are

dried and then counted in a liquid scintillation

counter.

Labeling of cells. Cells in logarithmic growth were spundownat 500xgfor 5 min, suspended at a

density of6 x 106 cells/ml in 3 ml of serum-free

mediumbuffered with 25 mM HEPES

(N-2-hydrox-yethylpiperazine-N'-2-ethanesulfonic acid)and

con-taining one-twentieth the normal concentration of methionine, and incubatedwith shaking for 15 min at37°C. Atthe endof thispreincubation, 200,uCiof

[35S]methionine per ml (250 Ci/mmol, Amersham/

Searle)wasadded,and incubation wascontinuedfor

45 min. Atthe endof the labeling period,aportionof thecell suspension wasremovedfor extraction, and the remaining cells were chased forvaryingperiods

oftimeby the addition ofa10,000-foldexcessof cold

L-methionine.Ten-microliter aliquotsof cell

suspen-sion were removed at the end ofeach incubation

periodforquantitation of[3S]methionine

incorpora-tion intoprotein(12). Cellswerewashed once with

Earle balanced salt solution, and then extracted

accordingto the procedure ofShapiro and August

(Biochim.Biophys. Acta,inpress),which consists of

lysing the cells by the addition of an extraction

buffer containing 5 mMTris-hydrochloride (pH9.2),

1mMEDTA,400mMKCl,1%TritonX-100,1mM

L-1-tosylamido-2-phenylethylchloromethyl ketone,

and 1 mM phenylmethylsulfonyl fluoride (PMSF)

(protease inhibitors, Sigma Chemical Co., were added to the extraction buffer just before use from a 100 mMstock solutionindimethyl sulfoxide). The suspensions were spun down at 25,000 x g for 10

min, and the pellets were reextracted with buffer containing no KCl. The pooled supernatants were

dialyzedagainst TEN buffer and spunat100,000xg

for60minbefore immuneprecipitation.Thedialysis

wasperformedtoget rid of some of the Triton X-100

andthereby facilitateimmunoprecipitation;no pro-teolytic degradation ofproteins was observed

result-ing from thedialysis step. Insoluble proteins were

removed by the high-speed spin and nonspecific

background was reduced.

Labeling of Rauscher virus-infected 3T3 mono-layer cells was performed in75-cm2 T-flasks using similar conditions as described above for the Friend cells, except that the preincubation was done for 60 min in serum-free medium lacking methionine, and the cells were lysed by scraping them into 2 ml of the extractionbuffer.

For 24-h labeling of cells with [3Hlglucosamine,

20 uCi ofD-[6-3H]glucosamine (Amersham/Searle)

was added per ml of a cell culture growing in stan-dard growth media at a density of 8 x 105 cells/ml. After the labeling period, the cells were spun down,

washed, and extracted, and the cell-free medium

was concentrated with a collodion bag (Schleicher

and Schuell, Inc., Keene, N.H.) in TEN buffer. Three-hour [3H]glucosamine labeling was

per-formed in cell suspension at a density of 6 x 106

cells/ml instandard growth media incubated with

shaking in the presence of 100 ,Ci of

D-[6-3H]glucosamineper ml.

Immunoprecipitation.Monospecificgoat antisera

againstpurifiedRauschervirus coreprotein p30and

envelope glycoprotein gp 69/71 were the generous

gift of T. August and M. Strand, AlbertEinstein

College of Medicine, Bronx, N.Y.Aliquots, 0.5 ml,

ofdialyzedandcentrifugedcellextracts were

incu-bated with 2 ,lI of antisera, for aperiod of2 hor

longer at 4°C. After thisperiod, pig anti-goat immu-noglobulin G antiserum wasadded,andprecipitates

were allowed to form at 4°C for periods of 6 h or

longer, and then washed three times with TEN buffer (20 mM Tris-hydrochloride, pH 7.4, 1 mM EDTA, and 100 mM NaCl), and then once with acetone.Thedriedpelletsweredissolved in 75,ulof

electrophoresis samplebuffer (10)byincubatingat

70°Cfor 10 min, and thenat100°Cfor2min.

Electro-phoresis was carried out in 20-cm-long 5 to 20%

exponential gradient,sodiumdodecylsulfate

(SDS)-polyacrylamide slab gels using the discontinuous

Tris-glycinebuffer system ofLaemmli (10).

Autora-diographyofdriedgelswasperformed with Kodak

Blue medical X-ray film (BB-54). For detection of

3H-labeledproteins,gelsweretreatedbythe method

ofBonnerandLaskey (2) andexposedtoKodakRR

RoyalX-Omat film at -70°C.

RESULTS

Detectionof virusrelease. Virusproduction

was measured by assaying for reverse

tran-scriptase activity inthe culture medium. Ali-quotsofcell suspensionweretakenatvarious timesafterseeding;celldensitywasmeasured, and reverse transcriptase activity in the cell-freemediumwasdeterminedasoutlined under Materials and Methods. As shown in Fig. 1,

virus release from producer F4 clone cells is

directly related to cell growth. As the cells reach a saturation density of about 2 x 106 cells/ml, virus release levels off. Media har-vested from growing clone B8 cell cultures,

http://jvi.asm.org/

(3)

330 RACEVSKIS AND KOCH

E 6/

,A

0 24 48 72 96

TIME (hours)

FIG. 1.Friend leukemiavirusproductio suredbyreversetranscriptaseactivity in th medium compared with cell growth. Rev scriptase activity releasedfromF4cells (A transcriptase activity released fromB8cells ble cellspermilliliter in F4 cell culture ( cellspermilliliter inB8cellculture (0).

however, shownodetectablereverset tase activity above background level, assay,indicatingthatnosignificantar

virusarereleased by these cells,as p

reported by Ostertagetal. (17).Itmay

that B8 cells have to be seeded at s

higher densities, and grow more slo'

theF4cells; both cell linesreach simi ration densities. After96hof growth, dropinviable cellspermilliliter intk

culture(not shown), justasinthecase

culture after72h (Fig. 1).

Immunoprecipitation ofviral pro

geta moredetailedunderstanding ofN

ductionby these cells, we undertook

viral proteinsynthesis byutilizingim cipitation techniques. Monospecific prepared against purifiedRauschermi

kemiavirusproteinsweremade availl

for this purpose. It has been shc

Rauscher and Friend viruses are ver

related andthat their structuralprote

many of the sameantigenicdetermin

24). Labeled viral proteins were pre

from Friend cellextractsby theseant

ing eitherthe director indirect tech]

immunoprecipitation. In using th method ofimmunoprecipitation,simil

were obtainedifeitherFriend orRau

rus carrier was added, indicating the

virusproteinswereboundjustaseffic

Rauscher virusproteinsundertheseco

2.4 To make the most use ofour limitedquantities of antisera, the indirect method of immunopre-cipitation was utilized throughout these

stud-2.0 ies.

Findings from several laboratories indicate '9 thatthe structural proteins of RNA tumor vi-1.6 o ruses are first synthesized as high-molecular-x weight precursors (9, 26, 27), which are subse-1.2 - quently cleaved to yieldthe maturevirion

pro-E teins. As shown in Fig. 2, polyacrylamide gel

_J electrophoretic analysis of

[35S]methionine-la-0.8 , beled Friend cellextracts, immunoprecipitated with monospecific antisera to the major enve-lope glycoprotein gp70and the major core

pro-0.4

tein

p30,

reveals several molecular species

spe-cifically precipitated with each antiserum. As 0 molecular-weight markers, we have used

[35S]methionine-labeled

lysates

ofmouse

plas-macytoma

(MPC-11)

cells infectedwith

vesicu-)nas mea- lar stomatitis virus

(13)

(Fig.

2a) and BHK-21

ge

cell-free cells infected with reovirus (21)

(Fig.

2o).

Coe-terse tran- lectrophoresis of labeled protein standards with

i),reverse unknowns is a more convenient method for

3(A).Via- comparison purposes on autoradiograms than

*), viable thecommonly used procedure of running

unla-beled standards and stainingthem in the gel.

The use of the indirect technique of immuno-transcrip- precipitationinthesestudies results inafairly

sforthis high backgroundinthegels duetononspecific mounts of adhesion oflabeled cellular proteins to the rela-ireviously tivelylarge precipitates. Someveryprominent

rbe noted contaminatingbands areseen inthe regionof omewhat about 50,000molecularweight;theseare proba-wly than bly actin-type proteins, which are synthesized

ilar satu- inlarge amountsby tissue culturecells; sirmilar thereis a observationshave beenreported by other work-ieB8cell ers (4, 8). Arelatively high degree of radioac-of theF4 tivelabelingmustbe achieved in these cellsto detect viral proteins, for they only constitute teins. To approximately 0.1% of total cellular protein viruspro- synthesis (unpublished observation).

tostudy Figures 2c, d, and e show

[35S]methionine-munopre- labeled F4 cell extractsprecipitated with anti-antisera gp7Oserum. Inall threechannels,45-min label-urineleu- ing (c), 60-minchase (d),and 120-minchase(e), able to us it can be seen that most of the gp70-specific

)wn that label ispresentina broadband,withapparent ry closely molecular weight between 55,000 and 60,000, ,insshare which seems to migrate faster with increasing

iants (22, chase time. There is a band in the expected

'cipitated

region of about 70,000daltons, and apossible

tisera us- precursorof80,000-molecular-weightwhich dis-niques of appearsafter a 60-min chase.A low-molecular-Le direct weight proteinof less than20,000isalso

specifi-arresults callyprecipitated inall three extracts. Figures

ischervi- 2f,g, andhshowaliquotsofthesameextracts

at Friend

precipitated

withanti-p30 serum. Ashas been

cientlyas reportedforRauschervirus-producingcells (9, )nditions. 26), p30-specific label is initially incorporated J. VIROL.

http://jvi.asm.org/

[image:3.505.71.252.46.256.2]

(4)

b c

d e

f g h i

i

k

I

mn

o

A. C.*b

H. e

BSA

IgG(H)

N

---

+fg

SirS.

5E'

_ 4

+ ..

t :

iki

';

t ,,.

1F

*:rt 3s

bsj_.jjj..

i38 1S1i

i iit ww

.t.-

iSl 1

M+X''s..E'

.s

., Aii

$,_,;wsx;S..

ii;t' v_!

:-; ,,,jl

:sss

-wss

--P<t-X,j,ji s:M:S

[image:4.505.48.449.43.511.2]

r

.|:$> At

ttAWi*

**+ SX9f9_

si80

---

70 --

60 -

50 --40

-_

30

FIG. 2. Autoradiograph ofSDS-gelelectrophoresis of immunoprecipitatesfromFriendcell extracts

pulse-labeled with[35S]methionine. A scale ofapproximate molecular weightsin kilodaltons hasbeen included

alongsidethegel fororientationpurposes. The approximate positionofmigrationofunlabeled bovineserum

albumin (BSA) (molecular weight, 68,000) has also been drawn in. (a) Molecular-weight markers of

[35S]methionine-labeled lysateof mouseplasmacytoma(MPC-11)cellsinfectedwith vesicular stomatitis virus

(VSV), showingthefollowingproteins with corresponding molecular weights: L, 195,000; G, 66,000;MPC-11

immunoglobulinG(IgG) (H),53,000;N,48,000;M,29,000; and MPC-11IgG(L),22,000. (b) Control

45-min

[PS]methionine-labeled,

F4 cell extract precipitated with nonimmune goat serum. (c) A 45-min

[5S]methionine-labeledF4 cell extractimmunoprecipitated with anti-gp7O serum.(d)Same as (c), butchased

60min with10,000xexcess cold L-methionine. (e) Same as (c), but chased 120 min.

(t)

Same as(c), except

precipitatedwith anti-p30 serum. (g) Same as(f),but chased 60 min. (h)Same as (f), but chased 120 min. (i)

A45-min[35S]methionine-labeledB8cell extract immunoprecipitated with anti-p30 serum. (j) Same as(i),

butchased60min.(k) Same as(i),butchased120min.(1)Sameas(i), exceptimmunoprecipitatedwith

anti-gp70serum.(m)Sameas(1),butchased60min.(n)Sameas(1),butchased120min.(o)Molecular-weight

markersof[35S]methionine-labeled lysateof BHK-21 cells infected with reovirus, showing the three protein groups with thefollowing molecular weight ranges: X,143,000 to 153,000;

g,

72,000 to 79,000; O.,43,000 to 54,000.

331

a

L~

L2200

x

-

150

.,, ,.S .

wl.41

f...I 44.

http://jvi.asm.org/

(5)

intoalarger-sizedprecursor of 65,000 to 70,000 molecularweight. Aftera 45-minpulse,there is very little label inthe p30 region, but after the 60- and 120-min chase periods we can see a decrease oflabelintheprecursoranda concom-itant increase in p30. An examination of la-beled nonproducer B8 cell extracts (Fig. 2i,j,

andk), reveals that, similarly, p30-specific la-bel is incorporated into a high-molecular-weight precursor which, however, migrates

faster thanthecorrespondingprecursorprotein of the F4 cells, andthat no p30 appears even

after a 120-min chase. Labeled gp7O-specific

proteins in B8cell extracts (Fig. 21, m, and n) appear to be qualitatively and quantitatively

the same as those ofF4cells, with mostofthe

label appearing in anapparently unstable spe-ciesof 55,000 to 60,000 initial molecularweight, which migrates faster with increasing chase

time.

Comparison with Rauscher virus proteins.

The unexpected observation of the unstable 55,000- to60,000-molecular-weightprotein

pre-cipitated with anti-gp7O serum necessitated a similar analysis of the better-characterized

Rauscher virus system for comparison pur-poses. Figure 3c showsthat,inRauscher virus-infected 3T3 cell extracts precipitated with anti-p30serum, mostof the label aftera45-min pulse is foundina65,000- to70,000-dalton pre-cursorthat comigrates with theanalogous pre-cursorinalabeledFriend cell(F4)extract(Fig. 3b). As shownpreviously forthe Friend cells,

aftera 60-minchase,mostof thelabel islostin

the precursor region and has increased in p30

(Fig. 3d). Precipitation of the Rauscher cell ex-tractwithanti-gp7Oserum (Fig. 3f)shows that aftera45-minpulse,mostof thelabel is inan

80,000-daltonproteinand that noneis detecta-bleintheregioncorrespondingtotheunstable

species of the Friend cells. Aftera60-minchase (Fig. 3g),therehas been a loss of labelfromthe

80,000-molecular-weight protein and appear-ance oflabel inadiffusebandthat is presum-ably gp7O. As in the case of the gp7O-specific proteins of the Friend cells, there is also precip-itation of a small-molecular-weight protein (<20,000) in the Rauscher virus-producing cells. The 80,000-dalton precursorglycoprotein

seemstobe more stable in the Rauscher virus

cells thaninthe Friend cells, for in the latter the correspondingprotein is no longer detecta-bleafter a 60-min chase.

Long-termlabelingoffriend cells. Cell cul-tures were labeled for a period of24 h with

[15S]methionine and virus concentrated from the cell-free medium. Inspection of the gp7O-specific protein bands in the 24-h-labeled cell extract (Fig. 4c) reveals that the 80,000- and

70,000-dalton protein bandsappearquitesharp,

whereas the 55,000- to60,000-dalton species is noticeably more diffuse and faster migrating than in the 45-min-labeled extract (Fig. 4b). Mostof thelabelintheviralfractionof the24-h labeling (Fig. 4d) is ingp7O, althoughthere is some at 80,000 molecular weight and some at 60,000. Interestingly, thegp7O-specific low-mo-lecular-weightprotein (<20,000)fromthe viral fraction(Fig.4d)migratesfasterthan thesmall

species from the 45-min-labeled cell extract (Fig. 4b). Anti-p30 precipitated proteins from the 24-h-labeled cells are shown in Fig. 4m; there was asimilarquantity oflabeled p30 in the viralfraction, with no otherspecificbands

visible (notshown).

Labeling of viral glycoproteins. Cells were

labeled with[3H]glucosaminetodeterminethe

degreeofglycosylationinthe different protein species specifically precipitated with anti-gp7O

serum. Two labeling periodswere used, 3and 24 h, and both showed essentially the same

profiles. As seen in Fig. 4eandh, the glucosa-mine is incorporated into four species, with most of it in a broad band corresponding to

gp7O,someinahigher-molecular-weight

mate-rialthat is bestdiscernible afterthe 3-h

label-ing(Fig. 4e) and is the presumed precursor of

gp7O.The unstable55,000-to60,000-dalton ma-terial is glycosylated as well, andjust as was seen after 24-h labeling with

VFS]methionine

(Fig. 4c), ittravels as a much broader peak in the 24-h[3H]glucosamine samplethaninthe 3-h sample, indicating greater heterogeneity in the former. There also appearstobe an

inter-mediate-sized glucosamine-containing protein of about 65,000daltonsinboth extracts,andno label is evidentinthe region of the small pep-tide of less than 20,000 daltons that isobserved

in

[35S]methionine-labeled

extracts precipitated withanti-gp7Oserum.

Theconcentrated supernatantof the culture medium was immunoprecipitated afterthe la-beling periods, and Fig. 4fand i reveal that there is a large amount of

[3H]glucosamine-labeled gp7O in the culture medium,

presum-ably released as part of the budding virions.

FIG. 3. (a)VSV-infectedMPC-11 cellextract molecular-weight marker. (b) p30-specific proteins of a 45-min[35S]methionine-pulsed Friend cell (F4) extract. (c) p30-specific proteins of a45-min [35S]methionine-pulsed Rauscher virus-producing 3T3 cell extract. (d) Same as (c) but chased 60 min. (e) Same as (b) except immunoprecipitated withanti-gp7Oserum.(f)Same as (c) exceptimmunoprecipitated with anti-gp7O serum. (g) Same as(f) but chased 60 min.

J. VIROL.

http://jvi.asm.org/

(6)

a

:is,!

bc

def

200 4----150

---pre

gp80

gp70

pre

p65

g55-60

__~i_I

60~~~~~~~e.

50~~~~~~~~~~~~~~~~~~....

40 3Q-e-

,_

g

80 -1

70-p30

333

http://jvi.asm.org/

(7)

.- c 0 e

f g

h

j

k

I

m

200 150-

OR

80--70-

i

..tS

60-50

-*

iw4a

*1.&

S^

40-_pre

qP

70 c

p

-

-w

pre

P

6

55-60

.,

I

.., -

i

30

30 -s-

_

FIG. 4. Autoradiograph of a dimethyl sulfoxide-,2,5-diphenyloxazole-treatedgelaccording to the method

ofBonner and Laskey (2). (a) VSV-infected MPC-11 cell extract molecular-weight marker. (b) A 45-min

[35S]methionine-labeledFriend F4 cell extractimmunoprecipitated with anti-gp7O serum. (c) gp7O-specific

proteinsof a 24-h[35S]methionine-labeledF4cellextract. (d)gp7O-specific proteins from the viral fraction of cell-free medium obtained from the 24-h[35S]methionine-labeledF4cells. (e) Extract of F4 cells labeled 3h

with[3H]glucosamineand immunoprecipitated with anti-gp7O serum. (f)gp7O-specific proteins of

concen-tratedcell-free medium of same cells as in (e). (g) Extract of F4 cells labeled 24 h with[3H]glucosamine treated with control nonimmune goat serum. (h)gp7O-specificproteins of 24 h [3H]glucosamine-labeled F4 cell extract. (i)gp7O-specificproteins ofconcentrated cell-free medium of same cells as in (h). (j) Reovirus-infected BHK-21 cell extract molecular-weight marker. (k) Rauscher virus-producing 45-min

[35S]methionine-labeled3T3 cell extracts precipitated withanti-gp7O serum. (1) Same as (k), but chased60

min.(m) A24-h[35S]methionine-labeledFriend F4 cell extractimmunoprecipitated with anti-p30 serum. J. VIROL.

http://jvi.asm.org/

[image:7.505.74.463.72.527.2]

(8)

Moststriking,however,isthefactthatnoneof the unstable 55,000- to60,000-dalton species is

observed in the media, whereas some of the

intermediate-sizedmaterial of 65,000 molecular weight is present.

DISCUSSION

The observation that the nonproducer B8

cellssynthesizeap30-specificprecursor protein

that is not eventually processed to form p30

suggests thatthis might be the actual lesion

thatprevents mature virus production by these

cells. A similar observationhas been reported

by Stephensonetal. (23) in the case of mutants ofRauschervirus, where therestrictionto rep-lication was shown to be associated with the intracellular accumulation at the restrictive temperature ofa 70,000-dalton precursor con-taining antigenic reactivities of three viral structural proteins:p30,p15,and p12. Stephen-son etal. (23) alsoreportedthat theprecursor was rapidly cleaved when infected cells were

shifted downtothe permissive temperature in

theabsence of furtherprotein synthesis. They

speculated that the temperature sensitivity may be a property of the precursor, such as a reversibleconformational defect. In the case of

theB8cells, it is clear thatthe

high-molecular-weightprecursor itself is altered, since it mi-grates faster than thecorresponding molecule in theproducer F4cells, suggesting a deletion type ofdefect or an improper cleavage in an

earliermaturation step. Restriction to

replica-tionthatiscausedby ablockinprecursor cleav-age has alsobeen shown inavian

virus-trans-formedhamstercells (4). Inthatsystem,

how-ever, Eisenmanetal. (4)suggestedthat,onthe basis oftheir observations, themostlikely

ex-planation forthe blockwasthat aspecific pro-tease was eitherpresent or virus inducible in avian cells and absent from the avian

virus-transformed hamster cells.

Of possible relevance is the observation by

Ostertag et al. (14) that, even though the B8 cellsproducebarely detectableamountsoftype Cparticles,they do displaymanyintracisternal virus-like A particles when viewed under the electron microscope. Evidence presented in a recent reportby Robertsonet al. (18)indicates that theintracisternalAparticles and

extracel-lular type C particles produced by a mouse myeloma cell line are closely related, and is suggestive ofaprecursor-product relationship

between the twotypes ofparticles.On the basis of thesereports, it is temptingtospeculatethat the A particles observed in B8 cells are imma-ture virions arrested atthat stage of matura-tionby the uncleaved precursor protein.

InthetwoFriendcell linesstudied,themost abundant protein that is precipitated with anti-gp70serum is a speciesof55,000 to 60,000

mo-lecular weight, which migrates faster and be-comes moreheterogeneous withincreasing in-cubation time. It is not an artifact resulting

from some experimental procedure used in these studies, since an identical analysis of a Rauscher virus-producing cell line reveals no

analogous peptide. This protein incorporates

glucosamine to a muchlesserextentthangp70 anddoes not seem to be released from the cell. Some workers have previously reported the presence ofa very minor glycoprotein compo-nent inFriend virus of approximate molecular

weightof60,000 or 45,000 (1, 11). Whetherthis is the same species that was observed in our study cannot be determined from the data given in those reports. Ofpossible interest is the observation by Bolognesi et al. (1) that treatment ofintact Friend virus preparations with glycosidases removed about 80% of the

glucosamine-labeled material from gp7O, and that the remaining material migrated as a

broader peak to a position of about 60,000 to 65,000molecularweight. In allstudiesdealing

withcharacterization ofglycoproteins, it must bekept in mind thatglycoproteins displayan anomalous migration on SDS-polyacrylamide gelelectrophoresisbecauseof a decreased bind-ing ofSDS to the oligosaccharide side chains relativeto thepolypeptide backbone (20). It is

possible that the protein moietyof the

faster-migrating species in these cells isidentical to that ofgp7O; the large difference inmigration

may be accounted for by its lesser degree of

glycosylation, which wouldresult in two addi-tive effects: a higher relative binding ofSDS and asmalleractualmolecularsize.The

broad-ening ofthe peak, loss of label seen after

ex-tended incubation periods, and absence from

the culture medium isindicative of

intracellu-lardegradation of thematerial, whichsuggests

that thecompletecarbohydrate side chainsare necessary to protectthe viralglycoproteinfrom

degradation.

All three cell lines analyzed in this study,

Friend leukemia cell lines F4 and B8 and Rauscher virus-infected 3T3 cells, showed the

presenceofasmall-molecular-weightproteinof

less than 20,000being specificallyprecipitated with monospecificanti-gp70 serum. Asimilar,

butfaster migrating, protein was also present in viruspelleted from cell-free media of an F4 cell line culture labeled for24h.Ongoing stud-ies have revealed that these proteins migrate slower thanp12on anSDSgeland donot cross-react with antisera toeitherp12orp15

http://jvi.asm.org/

(9)

skis and Koch, unpublished observation). The

fact thatthese peptides share antigenic deter-minantswith gp7O suggests that they might be breakdown products of gp7O. In support of our

observation is a very recent reportby Naso et al. (12), which describes the presence of two small nonglycosylated proteins (termed pl5E and pl2E) in Rauschervirus-infectedJLS-V16 cells, which share tryptic peptide sequences with the major glycoprotein and its precursor. Furthermore, Naso et al. suggest a precursor-product relationship between pi5E and pl2E, which is in accord with our observation that the smaller species are found in the

long-term-la-belingvirus preparation. Morein-depth studies will be required to elucidate the role of these

peptidesinthe virallifecycle.

Inaddition to the two cell lines described in this report, wehave analyzed four more Friend

erythroleukemia cell lines established by W.

Ostertag, and one line established by C. Friend, CL-745, and have observed in all of

them the same characteristic pattern of viral proteins, with the prominent 55,000- to

60,000-daltonspecies being precipitated with anti-gp7O serum (Racevskis and Koch, unpublished

ob-servation). Whether this pattern of viral pro-teinsynthesisischaracteristicof all Friend cell lines, and is an integral part of their unique properties, will have to awaitfurther study.

ACKNOWLEDGMENTS

WethankW.Ostertag forproviding us with the cell lines usedinthis studyand M.Strand andJ. T.August for their generousgiftof theantisera. We arealso indebtedtoJ. A. Bilelloand S. Z. Shapiro of Albert Einstein College of Medi-cinefor their aid andadviceintheapplicationofthe tech-niquesusedinthese studies.

LITERATURE CITED

1. Bolognesi, D. P., J. J. Collins, J. P. Leis, V. Moennig, W.Schaefer,and P. H. Atkinson. 1975. Role of carbo-hydrateindetermining the immunochemical proper-ties of the majorglycoprotein(gp71)ofFriend murine leukemia virus. J. Virol. 16:1453-1463.

2. Bonner, W. M., and R. A. Laskey. 1974. A film detec-tion methodfor tritium-labeled proteins and nucleic acidsinpolyacrylamidegels. Eur. J. Biochem. 46:83-88.

3. Dube, S. K.,I. B.Pragnell, N. Kluge, G.Gaedicke, G. Steinheider,andW.Ostertag. 1975.Induction of en-dogenous and of spleen focus-forming viruses during dimethylsulfoxide-induced differentiation of mouse erythroleukemia cells transformed by spleen focus-formingvirus. Proc.Natl. Acad. Sci. U.S.A. 72:1863-1867.

4. Eisenman, R.,V. M. Vogt, and H. Diggelman. 1975. Synthesis ofavian RNAtumorvirus structural pro-teins. Cold Spring Harbor Symp. Quant. Biol. 39:1067-1075.

5. Friend, C., W. Scher, J. G. Holland, and T. Sato. 1971. Hemoglobin synthesis in murine virus-induced leu-kemic cellsinvitro:stimulation oferythroid differen-tiationby dimethyl sulfoxide. Proc. Natl. Acad. Sci. U.S.A.68:378-382.

6. Halpern, M. S., D. P. Bolognesi, and L. J. Lewan-dowski. 1974. Isolation of the major viral glycoprotein and a putative precursorfrom cells transformed by aviansarcoma viruses.Proc. Natl. Acad. Sci. U.S.A. 71:2342-2346.

7. Hayflick, L. 1965. Tissue cultures and mycoplasmas. Tex. Rep.Biol.Med. 23(Suppl. 1):285-303.

8. Ikeda, H., W.Hardy, Jr.,E.Tress, and E. Fleissner. 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.

9. Jamjoom, G., W. L.Karshin,R. B. Naso,L.J. Arce-ment, and R. B. Arlinghaus. 1975. Proteins of Rauscher murine leukemia virus: resolution of a 70,000dalton, nonglycosylated polypeptide contain-ingp30 peptide sequences. Virology 68:135-145. 10. Laemmli, U. K. 1970. Cleavage of structural proteins

during theassemblyof the head ofbacteriophageT4. Nature(London) 227:680-682.

11. Moenning, C., H.Frank,G.Hunsmann, I.Schneider, and W. Schaefer.1974.Properties ofmouseleukemia viruses. VII.The major viral glycoprotein ofFriend leukemia virus.Isolation andphysicochemical prop-erties.Virology 61:100-111.

12. Naso, R. B., L. J. Arcement, W. L. Karshin, G. A. Jamjoom, and R. B.Arlinghaus.1976.A fucose-defi-cientglycoprotein precursor to Rauscher leukemia virus gp69/71. Proc.Natl. Acad. Sci. U.S.A. 73:2326-2330.

13. Nuss, D. L.,andG. Koch.1976.Differential inhibition of vesicular stomatitis viruspolypeptide synthesis by hypertonic initiation block.J. Virol.17:283-286. 14. Ostertag,W., T.Cole,T.Crozier,G.Gaedicke,J.Kind,

N. Kluge, J. C.Krieg,G.Roesler, G.Steinheider,B. J. Weimann, and S. K.Dube.1974.Viral involvement inthedifferentiation oferythroleukemicmouseand humancells, p. 493-520. In Proceedingsof the 4th Symposium Princess Takamatsu Cancer Research Fund. Differentiation and control ofmalignancy in tumorcells. University ofTokyo Press, Tokyo. 15. Ostertag,W., T.Crozier,N.Kluge,H.Melderis,and S.

Dube. 1973. Action of5-bromodeoxyuridine on the induction ofhaemoglobinsynthesisin mouse leuke-miacells resistantto5-BUdR. Nature(London)New Biol. 243:203-205.

16. Ostertag,W., H.Melderis,G. Steinheider, N. Kluge, and S. Dube. 1972.Synthesisofmousehaemoglobin andglobin mRNAinleukaemic cell cultures. Nature (London) New Biol. 239:231-234.

17. Ostertag,W., G.Roesler, C. J.Krieg,J.Kind,T.Cole, T. Crozier, G.Gaedicke, G.Steinheider, N. Kluge, and S. Dube.1974.Induction ofendogenousvirusand thymidine kinasebybromodeoxyuridineincell cul-turestransformedbyFriend virus. Proc. Natl. Acad. Sci. U.S.A. 71:4980-4985.

18. Robertson, D.L.,P.Yau,D. C.Dobbertin,T. K. Swee-ney, S. S.Thach,T.Brendler,andR. E.Thach.1976. Relationships between intracisternal typeAand ex-tracellular oncornavirus-like particles produced in murine MOPC-460 myelomacells. J. Virol. 18:344-355.

19. Sato, T., C.Friend,andE.DeHarven.1971. Ultrastruc-tural changes in Friend erythroleukemia cells treatedwithdimethyl sulfoxide. Cancer Res. 31:1402-1417.

20. Segrest, J. P.,R. L.Jackson,E.P.Andrews, andV. T. Marchesi. 1971. Humanerythrocyte membrane gly-coprotein:are-evaluation of themolecularweightas determinedbySDSpolyacrylamide gel electrophore-sis.Biochem. Biophys.Res. Commun.44:390-395. 21. Shatkin,A.J.,and G. W.Both. 1976. Reovirus mRNA:

transcriptionand translation. Cell7:305-313.

J. VIROL.

http://jvi.asm.org/

(10)

22. Steeves, R. A., M. Strand, and J. T. August. 1974. Structural proteins of mammalian oncogenic RNA virues: murineleukemia virus neutralization by

anti-seraprepared against purified envelope glycoprotein.

J. Virol. 14:187-189.

23. Stephenson, J. R., S. R.Tronick, and S. A. Aaronson. 1975. Murineleukemiavirusmutantswith

tempera-ture-sensitivedefectsinprecursorpolypeptide

cleav-age.Cell 6:543-548.

24. Strand, M., and J.T. August. 1974.Structuralproteins ofmammalian oncogenic RNAviruses:multiple anti-genicdeterminants of themajorinternal proteinand

envelope glycoprotein. J. Virol.13:171-180. 25. Ussery, M. A., R. Ramirez.Mitchell, and B. A.

Har-desty. 1976. Inhibition ofFriend murine leukemia virus production by low-ionic-strength medium. J. Virol. 17:453-461.

26. Van Zaane, D., A. L. J. Gielkins, M. J. A. Dekker-Michielson, and H. P. J. Bloemers. 1975. Virus-specificprecursorpolypeptidesincells infected with Rauscher leukemia virus. Virology 67:544-552. 27. Vogt, V. M.,and R. Eisenman. 1973.Identification ofa

large polypeptide precursor of avian oncornavirus proteins. Proc.Natl. Acad.Sci. U.S.A. 70:1734-1738.

http://jvi.asm.org/