Abelson murine leukemia virus mutants with alterations in the virus-specific P120 molecule.

(1)

0022-538X/80/01-0340/09$02.00/0

Abelson

Murine Leukemia Virus Mutants with Alterations in

the

Virus-Specific

P120 Molecule

NAOMIROSENBERG`* AND OWENN.WITTE2

Cancer Research Center, Tufts University School of Medicine, Boston, Massachusetts 02111,1 and Cancer ResearchCenter,Massachusetts Instituteof

Technology, Cambridge,

Massachusetts021392

Abelson murine leukemia virus (A-MuLV) is a replication-defective virus that

transformsbothfibroblastsand hematopoieticcellsin vitro. The virus encodes a

120,000-molecular-weight protein (P120) that is composed of Moloney murine leukemiavirus-derived gag gene sequences and A-MuLV-specific sequences. This

protein is the only A-MuLV-encoded protein that has been detected, and thus P120 is a candidate for the transforming protein of A-MuLV. We now report isolation and characterization of three new A-MuLV isolates that do not synthe-size P120 but do produce analogous proteins of larger (160,000 molecular weight)

and smaller (100,000 and 90,000 molecular weight) size. All of these A-MuLV isolates transform fibroblasts and lymphoid cells in vitro. Because the different A-MuLV proteins vary in the A-MuLV-specific region of the molecule, these variants may set a maximum limit on the size of the A-MuLV transforming

protein.

Abelson murine leukemia virus (A-MuLV) is areplication-defective, transforming retrovirus that arose in asteroid-treated BALB/c mouse

injected with Moloney murine leukemia virus (M-MuLV) (1). The virus induces a fatal

non-thymiclymphosarcoma3 to 4weeks after injec-tion into neonatal mice and transforms both fibroblasts and hematopoieticcellsin vitro (18,

21). A-MuLV stocks always contain a replica-tion-competent virus (such as M-MuLV) that provides the defectiveA-MuLV withenzymatic functions andstructuralproteins (9, 21).

The defective nature ofA-MuLV was dem-onstrated directly by Scher and Siegler (21)

when they isolated an A-MuLV-transformed nonproducer cell line called ANN-1. A-MuLV rescued from ANN-1 cells with cloned helper virus inducedtypical Abelson lymphosarcomas,

indistinguishable from the disease induced by

the uncloned virus prepared from tumor ex-tracts.ANN-i-derivedvirusthusbecamewidely

used and served as the prototype cloned

A-MuLV strain.

Heteroduplex mappingandSinuclease

map-ping studies using ANN-i-derived A-MuLV

show that the 5.6-kilobase A-MuLV RNA ge-nomehas tworegionsofprecise homologywith M-MuLV:a 730-baseregionatthe 3' end anda

1,320-baseregionatthe 5' end(D.Baltimore,A

Shields, G. Otto,S. Goff, P. Besmer, 0. Witte,

and N. Rosenberg, Cold Spring Harbor Symp. Quant. Biol., in press; A. Shields, G. Otto, S.

Goff, M. Paskind,and D. Baltimore, submitted forpublication). The 3.5-kilobase regioninthe

center ofthe A-MuLV genome is not

homolo-gous to M-MuLV, and hybridization studies

show that this region is not present in other replication-defective murine retroviruses(8;

Bal-timore et al., in press). These A-MuLV se-quences are present innormalmouseDNAand

probably represent a portion of the genome of the BALB/c mouse from which A-MuLV was

originally isolated.Because these sequences

dis-tinguishA-MuLVfromits M-MuLV parent and

other murineretroviruses, theymay be

respon-sible for the unique biological properties of

A-MuLV.

Analysisof theproteins expressed inANN-1

cells and A-MuLV-transformed cellsderived us-ing virus recovered directly or indirectly from ANN-1 cells has identified a single protein of

120,000molecularweight (P120)that isencoded by the 5.6-kilobase genomic RNA of A-MuLV (18, 19, 26). Thisprotein contains some of the determinants of theM-MuLV gag geneproducts

(p15, p12,andapart ofp30),and theremaining

90,000 molecularweightis made up of

A-MuLV-specificsequences(25;Baltimoreetal.,inpress).

These A-MuLV-specific sequences share

sero-logical determinants with a normal mouse

cel-lularproteinof150,000 molecularweight

(NCP-150) that is present inspleen, thymus,and bone marrow(25a).

The P120 protein is the only A-MuLV-en-codedproteinthatwehavedetected by usinga

varietyofserologicalreagentsincludingan anti-Abelson tumor (anti-AbT) serum. In addition,

this molecule isexpressedin all of the

A-MuLV-340

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

transformedcells wehave examined(26;

Rosen-berg and Witte, unpublished data). Thus, P120 isacandidate for the transforming protein of

A-MuLV. However, mutants expressing altered

P120 are needed to directly assess the role of

this molecule transformation.

Wenowreportthe isolationofthreeA-MuLV

strainsvarying in the size of the A-MuLV-spe-cificprotein. Although all three of these variants transform fibroblasts andlymphoidcellsinvitro,

the variants set a maximum size for the

A-MuLV-specific sequences thatmay be respon-sible for cellular transformation. One of the A-MuLV variants expresses alarger,

160,000-mo-lecular-weight (P160) protein. This A-MuLV strainappearstobeanaturallyoccurring variant that ispresentina secondnon-ANN-1-related

A-MuLV viruspool maintainedatthe National Institutes of Health. The P160 molecule has also been observedby R. Risser and co-workers using these virus stocks (D. Grunwald and R. Risser, personal communication). The other two A-MuLV variants produce smaller proteins, 100,000and 90,000 molecular weight (P100, P90), andwereisolated fromAb-NRK cells (8).

Pro-teins madeby these variants are

nonglycosyla-tedphosphoproteins andexpressthe samegag

genedeterminantsasP120. Thenewisolatesare

stable upon cloning and transform both

fibro-blast andlymphoid cells. The lymphoid

trans-formants aresimilar morphologically and with

respect to expression of the lymphoid-specific markers immunoglobulin and terminal deoxy-nucleotidyl transferase (TdT).

MATERIALS AND METHODS Cells and viruses. Routine maintenance of cell cultures hasbeen described elsewhere (16, 17). Ab-NRK cells were originallya giftof E. Scolnick,

Na-tional Institutes of Health. ABPC-22 cells (9)were a

gift of M. Potter, National Institutes ofHealth, and

RAW309Crcells(11)were agiftof P.Ralph,Memorial

Sloan-Kettering Institute. Lymphoid cells

trans-formedbythe different A-MuLVisolateswerederived

fromindividual foci ofinfected bonemarrow cells by usingtheagartransformationassay (16)and adapted togrowinliquidcultureaspreviously described (17).

Clones ofA-MuLV-transformed fibroblastswere

de-rivedby platingthe infected cellsinmicrotiterwells

aftera 1.5-hadsorption period. Thetrypsinized cells

were plated at aconcentration of0.3to 0.4cell per well.

All virus stocks were prepared from 24-h culture

fluids and passed through0.45-,um filters beforeuse.

The titers offocus-forming virus in A-MuLV stocks

weredeterminedusingNIH/3T3cells (21).Titers of helpervirusweredeterminedbyassayonS+L- cells

(2)orbythe XCplaqueassay(20).

A-MuLV(P120) (i.e., A-MuLV producinga 120,000-molecular-weightA-MuLV-specific protein)was

pre-pared using ANN-1 cellssuperinfectedwith M-MuLV clone 1 (5). A-MuLV(P160) was prepared using RAW309Cr cells or NIH/3T3 cells transformed by filtered culture fluids from RAW309Cr. The RAW309Cr cells used in theseexperiments release A-MuLVincombination withanNB-tropic, XC plaque-positivehelper virus(datanotshown) that is probably related tothe M-MuLV in theoriginalA-MuLVtumor

extracts.A-MuLV(P100) and A-MuLV(P90)were

de-rived from clones of Ab-NRK cells that had been treated with 10,ug offluorodeoxyuridine per ml for 4 days beforecloning. The A-MuLVwasrescued from thesenonproducerclonesby usingtheamphototropic

virusisolate292 (12) andpassedontoNIH/3T3 cells. Theresultingtransformed cultures ofNIH/3T3cells wereused for routine virusproduction.

Proteinlabeling studies. All radiochemicalswere purchased from NewEnglandNuclearCorp., Boston,

Mass.Techniques forlabeling,immunoprecipitation,

andgelelectrophoresis have been described in detail elsewhere (22, 24, 25).Briefly, rapidly growing cells werelabeled with50,uCiof[3S]methionineper ml in serum-free medium lacking methionine. Cells were labeledincomplete medium with[3H]glucosamineat

100,uCi/ml.Endoglycosidase Hreactionswere as de-scribed (27).Acellextract wasprepared from107 cells by usinglysis buffer (10mMNaPO4, pH 7.5,100mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1%sodiumdodecyl sulfate [SDS])at0°Cand clari-fiedat 100,000xg for2.5 h. Antisera were added to the extract, and the immunoprecipitates were col-lected with Staphylococcus aureus (6) and analyzed onSDS-polyacrylamide gels (7) developed by fluorog-raphy (3).

Antisera used were rabbit anti-M-MuLV reverse transcriptase/p15, rabbit anti-M-MuLV gp7O, goat anti-Rauscher MuLV p12, goat anti-Rauscher plO, goat anti-M-MuLV virion, mouse anti-AbT, rabbit anti-mouse immunoglobulin (a gift of V. Sato), and rabbit anti-calf TdT serum. The characterization of these reagents has been describedelsewhere (22, 24-26; A. Silverstone, L. Sun,0. N. Witte, and D. Balti-more, submitted forpublication). Virus proteins and TdTwereexaminedby electrophoresis of immunopre-cipitatesthrough 10% polyacrylamide gels; immuno-globulin molecules were examined in the presence of 2-mercaptoethanol on 12.5% gels and in the absence of2-mercaptoethanolon 5to 20% gradient gels.

RESULTS

Detectionofa strainof A-MuLV produc-ing a larger form of P120. The A-MuLV-encoded P120 proteinhas been expressed in all of the more than 100 different lymphoid and fibroblast transformed cells we have examined (26; Rosenberg and Witte, unpublished data). These cell lines were all derived in vitro in a similarfashion, either by cloninginfected cells

atlimiting dilution,in the case of thefibroblasts, or by using the agarose transformation assay (16) for lymphoid cells.In the case of lymphoid

cells, allof the celllines expressed asimilarset

http://jvi.asm.org/

(3)

342 ROSENBERG AND WITTE

ofdifferentiation markers and are probably anal-ogous to primitivecells related to B lymphocytes (22, 23). The A-MuLV used in all of these

ex-periments was derived either directly or

indi-rectlyfrom ANN-1 cells.

Toinvestigate the universality of P120

expres-sion, we examined two A-MuLV-transformed

cell linesisolated from tumor tissue by two other laboratories. Two celllinesexpressing different

phenotypic markers were used: the ABPC-22

cell line, a typical immunoglobulin-secreting

plasmacytoma (9), and RAW39OCr, a

macro-phage line (11). [35S]methionine-labeledcell ex-tractsfromthesecelllines were immunoprecip-itated withantiseraspecific for M-MuLV and

A-MuLVproteinsandanalyzed by electrophoresis

(Fig. 1). The previously described lymphoblas-toid 2M3 cell line (17, 26) produced P120 that reacted with anti-p15, anti-p30, and anti-AbT sera(Fig.1C).However, the ABPC-22 myeloma

B

!

2

3

4 5 6

4_:a

cell line (Fig. 1D) and the macrophage cell line

RAW309Cr(Fig.1A)both did notproduce P120, butdidproduce a 160,000-molecular-weight

pro-tein (P160) that was precipitated withanti-p15, anti-p30,and anti-AbT sera. The weak

precipi-tation of P160 with anti-p30 serum was not

observed consistently.Atypical lymphoblastoid

cell line isolated by infecting spleen cellswith virus fromRAW309Cr,the R5cellline(Fig.1B), alsoproduced P160. Thus, P160 expression was notrelated tothe highly differentiated pheno-types expressed by RAW309Cr and ABPC-22

cells.

A more complete survey of the serological

determinantspresent on P160 is shown in Fig. 2.

Immunoprecipitation of P160 from the R5 spleen cell line showed that the protein

con-tained determinants of p15, p12, and p30, but notplO. Anti-AbTserum(lane 7) but not normal mouseserum(lane6) alsoprecipitated P160. In

2 3 4 56

-P160-

-P120-..

2

3

4 5 6,

*

aw-Pr80e

;K-am

-Pr65-q-"--

TdT--)1jt TdT 2)anti.-.pIl

5)normal mouse serum 6)ant-AbT

FIG. 1. Analysis of virus-specificproteinsinA-MuLV-transformed celllines. Cell linesRAW309(A),R5

(B),2M3(C),and ABPC22(D)werelabeled with[35S]methionine (100 .Ci/107cells,1h,37QC)and extracted

aspreviouslydescribed (24).A volumeof106cells wasimmunoprecipitatedwith 5,ul ofanti-TdT(lane 1),

anti-M-MuLVp15reversetranscriptase (lane 2), anti-M-MuLVgp7O (lane 3), anti-M-MuLVp 30 (lane 4),

normal mouse serum (lane 5), and mouse anti-Abelson MuLV tumor regressor serum (AbT) (lane 6).

Precipitates were collected with S. aureas, denatured, and analyzed on a 10%o SDS-polyacrylamide gel

developed byfluorography (3, 6, 7).

A

- A >4 6i

I"s.x...R

L,F-3)

ant g p

http://jvi.asm.org/

[image:3.504.70.462.280.591.2]

(4)

1 2345

6 7

-P160O-

t

~

Pr8Oenv-_

* *

*

Pr659"-

-FIG. 2. Further serologicalanalysis ofP160. The R5celllinewas labeled,extracted, immunoprecipi-tated, and analyzedasdescribed inFig. 1. Lane1,

Anti-M-MuLVp15/reverse transcriptase; lane 2, anti-M-MuLVgp70; lane3,anti-M-MuLVp30;lane 4, anti-Rauscher MuLVplO; lane5, anti-Rauscher MuLVp 12;lane6, normalmouse serum;lane7, anti-AbT.

addition to thegaggene-related determinants, the P160molecule alsocontained A-MuLV-spe-cificdeterminants, becauseprecipitation of P160 by anti-AbT was not blocked by excess

M-MuLV virionprotein (datanotshown). No

reac-tivity with anti-gp70 (Fig. 1 and 2) orwith re-verse transcriptase determinants was detected (data not shown). In addition, P160 was

effi-ciently labeled with inorganicphosphate butnot

with[3H]glucosamine, and the size of the mole-culewasnotalteredbytreatmentwith endogly-cosidase H(datanotshown).

A-MuLV strains synthesizing smaller forms of P120. In another series of experi-ments,weattemptedtoisolateP120mutantsby

selecting for flat variants ofthe Ab-NRK cells

(a P120 producer) (8, 26). Ab-NRK cells were

treatedwith10,ug of fluorodeoxyuridineperml

for4days andclonedatlimiting dilution. After 3weeks,twocloneswerechosen forstudy. One of the clones, AbNRK-1, was morphologically

distinct,beingcomposedofelongated,refractile cells that formedan incomplete cell layer, one

cellinthickness. The secondclone,Ab-NRK-3,

wassimilarinmorphologytotheparentAbNRK cell line, being composed ofcells with a more

regular epithelioid morphology that formed denselypackedlayers.

Todetermine whether themorphologyofthe

two clones was controlled bythe A-MuLV ge-nome,theviruswasrescuedfrom themwiththe

amphotropic wild mouse virusisolate 292 (12)

and passed to NIH/3T3 cells. A-MuLV that transformed NIH/3T3 cells efficiently was

re-covereJ

from both clones. The titers of focus-forming virus, measured in the NIH/3T3 cell

transformationassay(21),wereabout thesame:

105focus-forming units for virus recovered from Ab-NRK-3 and 2 x 105 focus-formingunits for

virus recovered from Ab-NRK-l. In addition,

the morphology of the NIH/3T3 cells

trans-formedby both virusstockswas

indistinguisha-ble and typical of A-MuLV-transformed

fibro-blasts.

Focusisolates of cells transformed withvirus

derived from thetwoAb-NRKcell cloneswere

examined forexpressionof the

A-MuLV-specific

protein. All of the clones derived from virus rescued from AbNRK-1 synthesized, instead of P120, a smaller, 100,000-molecular-weight

(P100) protein. The clones derived from virus

rescued from AbNRK-3 fell intotwogroups,one

synthesizing P120 and another synthesizing a

90,000-molecular(P90)protein.Serological

anal-ysis of the P100 andP90molecules is shown in Fig. 3. Extracts from a nonproducer cell line

synthesizingP90wereprecipitated with the

M-MuLVgag gene reagentsanti-p15, anti-p12,and

anti-p30sera (Fig.3A, lanes 2, 3, and 4), butnot

byanti-plOserum(Fig.3A, lane5).P90wasalso

precipitated by anti-AbT serum (Fig. 3A, lane

6). A similar pattern of precipitation was

ob-tained whenextracts from a producer cell line synthesizingP100 wereexamined(Fig. 3B).This

molecule reacted with anti-p15, anti-p30, and

anti-p12sera(Fig.3B, lanes 2, 3, and 4) and

anti-AbT (Fig. 3B, lane 6), but not with anti-plO

serum (Fig. 3B, lane 5). Bothproteins reacted with anti-M-MuLV virion serum (lane 1), and neither reacted with anti-gp7Oserum (lane 7).

The presence of A-MuLV-specific

determi-nants in P90 was demonstrated in a blocking experiment (Fig. 4)whichshowed thatP90was

precipitated with anti-AbT (Fig. 4B, lane 1)

alone or withanti-AbTin thepresence of60 or 180,ug of M-MuLV virion proteins (lanes 2and

3). P90precipitation via gag gene determinants with anti-M-MuLV serum (Fig.4A, lane 1)was

blocked by the presence of60and 180

jig

of

http://jvi.asm.org/

[image:4.504.59.231.72.359.2]

(5)

MuLV virion protein (Fig. 4A, lanes 2 and 3). P100 molecules also had A-MuLV-specific

de-terminants that couldbedemonstratedin

block-ingexperimentswithanti-AbTserum (data not

shown).

Retrospective examinationofAbNRK-1 cells

and the culture ofNIH/3T3 cellstransformed

by mass infection with AbNRK-1-derived

A-MuLVshowedthat all ofthesecells synthesized

P100. The NIH/3T3 culture transformed by mass infection with Ab-NRK-3-derived A-MuLVsynthesizedboth P120 and P90, but only P120 could be detected in Ab-NRK-3 cells.

Thus, althoughastable changeintheA-MuLV genome in the fluorodeoxyuridine-treated cells

doesnotappear relatedtothealtered

morphol-ogyofsomeof thesecells,thisscreening proce-duredid resultin therecoveryofaltered

P120-like proteins.

Stability of the A-MuLV variants. Virus

stocks fromA-MuLV-transformedcells

synthe-sizingthe various sizes ofproteinswereusedto

transform aseries ofnew clones to determine

whether the size of theA-MuLV-specificprotein

was a stable,heritable trait. Clones of both fi-broblasts and bone marrow cells derivedusing

A-MuLV(P160) and A-MuLV(P100)

synthe-sized theappropriateproteins (Table 1). In ad-dition,anewseries of ANN-1-derived A-MuLV clones were also examined, and all of these clonessynthesized P120 (Table 1).Mostof the clonesderivedusingthe mixed P90 and P120

A-MuLVstock

expressed

onlyP90. Thehigh

pro-A

A

!~

;_..J

-w

_ < 5 5

..i

40

[image:5.504.318.410.70.251.2]

P100 P90 PrB0 Pr65 <

FIG. 3. Analysis ofA-MuLVP90 and P100

pro-teins. Bone marrow cells transformed with

A-MuLV(P90) (A) andA-MuLV(PIOO) (B) were pre-paredandanalyzedasdescribedin Fig. 1.Lane1,

Anti-M-MuLVvirions; lane 2,anti-pl5/reverse

tran-scriptase; lane3, anti-p12; lane 4, anti-p30; lane5, anti-plO;lane6, anti-Abt; lane 7,anti-gp70.

A

3 *

t-

-P90

FIG. 4. P90contains serologicaldeterminants re-active with anti-AbT serum.

A-MuLV(P90)-trans-formedbonemarrowcell line234-9waslabeled with

[35S]methionineandprepared for

immunoprecipita-tionanalysisasdescribed in Fig.1. (A) Extractfrom 106cellsimmunoprecipitatedwith

5Mul

ofapolyvalent goatanti-M-MuLVvirion antiserum in the absence (lane 1)orpresenceof50pg(lane 2)or 180pg(lane 3)

ofunlabeledM-MuLVproteins. (B) Equalsamples

immunoprecipitatedwith5ulofmouseanti-AbT

se-rum asdescribed above.Precipitates were analyzed

asdescribed in Fig.1.

portion of the P90 producers may reflect the

ratio of A-MuLV(P90) and A-MuLV(P120) in

the uncloned, transformedNIH/3T3 cellsused

toprepare thevirusstock. Theamountsof P90

andP120inthiscelllinedetectedby metabolic

labeling indicate that P90 is the

predominant

species (datanotshown).

The P160 molecule is the largest

A-MuLV-specific protein we have found; this A-MuLV

strainmaybe the

parental

A-MuLVisolate.To

determine whetherP120was achanceisolatein

the ANN-1nonproduceror

represented

asecond

majorvariant ofA-MuLV,we

analyzed

the size

ofthe A-MuLV-specific

protein

in a series of

clones derived froma tumorextract similarto

thoseused in theoriginalScherandSiegler (21)

experimentsthatledtotheisolation ofANN-1.

TheA-MuLVinthistumorextract (a

gift

ofR.

Siegler) descendedfrom

original

material

given

toDr. SieglerbyH. T.Abelson and L. S.

Rab-steinandwasmaintainedby animalpassagein Boston for several years prior to the in vitro

studies. Seven clones of

A-MuLV-transformed

fibroblasts and

lymphoid

cellswereisolated

us-ing this tumorstock, andall seven clones

syn-thesized P120 (Table 1). Therefore, the P120

strain ofA-MuLVwasthemajor componentof the Boston lineage of

A-MuLV,

at leastatthe

http://jvi.asm.org/

[image:5.504.71.262.415.584.2]

(6)

TABLE 1. Stabilityofthe A-MuLV-specificprotein

uponcloninga

Protein ex- Proteinexpressed in clon-Source of pressedby ally derived transformants A-MuLV parentcell

line P90 P100 P120 P160 ANN-1 P120 0/50 0/50 50/50 0/50 RAW309Cr P160 0/21 0/21 0/21 21/21 Ab-NRK-1 P100 0/12 12/12 0/12 0/12 Ab-NRK-3 P90,P120 12/16 0/16 4/16 0/16 Tumor extract 0/7 0/7 7/7 0/7

aFiltered A-MuLVstockswereusedtoinfect both bone marrowcells andNIH/3T3 fibroblasts.Afteradsorption,the infectedcellswereclonedby usinglimitingdilution forNIH/

3T3cellsand the agarose transformation assay(16)for bone

marrowcells. Transformed cloneswereisolated,and the size of theA-MuLV-specificproteinwasdeterminedbySDS-gel electrophoresis of [nSlmethionine-labeled immunoprecipi-tates.

timein passage atwhich this extractwas

pre-pared.

Lymphoid-specificmarkersexpressed by

A-MuLV-transformedclones. Previous work

has established the presence of

lymphoid-spe-cific gene markers in A-MuLV-transformed lymphoid cells (4, 9,11,22,23). Direct biochem-ical and immunological assays have demon-strated the presence of the markers TdT and intracellularimmunoglobulin inmany cell lines (22, 23). Thefrequencyofexpression of immu-noglobulin and TdTwasmonitored in A-MuLV-transformed bone marrow clonesto determine whether thetransformed cells derived using the different A-MuLV strainswere similar. Clones

derived using A-MuLV(P90) and A-MuLV-(P160)wereanalyzedalong witha newseriesof

A-MuLV(P120)clones. Bonemarrowcellsfrom

C57BL/6J, C57L/J,andBALB/cAn micewere

used for the transformation studies. The

pres-ence of immunoglobulin and TdT was

deter-mined by SDS-polyacrylamide gel analysis of

[3S]methionine-labeled immunoprecipitates (Fig. 5). Consistent with observationsdescribed

elsewhere, the anti-calf TdT serum used here precipitated a 60,000-molecular-weight protein

from murine cell lines synthesizing TdT

(Silver-stone etal.,submitted forpublication). The anti-mouseimmunoglobulinserumusedhere

precip-itated immunoglobulin M (IgM) and IgG from

cell linessynthesizing these proteins(22).

Screening of a large number of lymphoid

clones showed that most of them synthesized TdT(Table 2).About60% oftheclones synthe-sized immunoglobulin in the form of ,u heavy chain (Table 1). No 7 to8S IgM molecules or freelight-chain moleculesweredetected in any ofthe clones analyzed here. The frequency of

immunoglobulin-positive clones and the pres-ence of, heavy chain in the absence of light

chain is consistentwithearlier observations(22). As shown in Table 2, no strain-specific

differ-ences were seenwhen the frequency of

immu-noglobulin and TdT expression was tabulated

accordingtotheA-MuLV isolate used to

trans-form the clones. In addition, when these data

werereanalyzed with respecttothemousestrain

A

12 34

B

12

34 W

~H

[image:6.504.44.237.75.176.2]

TdT

FIG. 5. Immunoprecipitationanalysisfor intracel-lularimmunoglobulinand TdT in clonal

A-MuLV-transformedlymphoidcell lines. Individual clonesof

A-MuLV-transformedcells(lanes1to4)werelabeled

andpreparedasdescribed inFig. 1, then

immuno-precipitatedwith5,ulofrabbit anti-TdT(A) anda polyvalent rabbit anti-mouse immunoglobulin (B).

Sampleswereprocessedasdescribed inFig. 1. TdT

expression was analyzed on a 10%

SDS-polyacryl-amidegel,andimmunoglobulinexpression was an-alyzedona12.5%SDS-polyacrylamidegel.

TABLE 2. Frequency of A-MuLV-transformedbone

marrowclonesexpressinglymphoid markersa

Frequency in A-MuLV strain: Marker

P160 P120 P90

IL 15/22 6/11 2/3

TdT 20/22 9/11 2/3

aSingle-focus isolates of A-MuLV bone marrow

cells were adapted to grow in liquid medium and examined forexpression of immunoglobulinandTdT by SDS-gel electrophoretic analysis of

[nS]methio-nine-labeled immunoprecipitates(22, 23). The size of the A-MuLV-specificprotein was confirmed in each of the clones by electrophoretic analysis of labeled immunoprecipitates.

http://jvi.asm.org/

[image:6.504.247.438.153.392.2] [image:6.504.248.438.531.583.2]

(7)

usedasa sourceofbone marrowcells,nomouse

strain-specific differences in the frequency of

marker expression were evident (data not

shown).

DISCUSSION

Allof the A-MuLV-transformed cell lineswe

have examined synthesizeanA-MuLV-specific protein that contains M-MuLV gag

gene-de-rivedsequences.Allofthemoleculesare

nongly-cosylated phosphoproteins, and the size ofthe

different proteins is stable upon cloning. Al-though these proteinsvary in sizefrom 160,000 to 90,000molecular weight, they aresimilar in

theirstructure,beingmadeupofthe

M-MuLV-derivedp15, p12, andpartof thep30 protein in

additionto anA-MuLV-specific region. Because

theproteinscontain thesamegaggene-derived

determinants, theyprobablyvary inthe size of theirspecific region. If the

A-MuLV-specific regionof thesepolyproteinscontains the

transforming sequences, the A-MuLV(P90)

strainplacesamaximum limitonthe size ofthis

region. Because P90 resembles P120in structure,

about 30,000molecularweight of P90 is probably derived from the M-MuLV gag geneproducts

p15,p12, and p30 (26; Baltimoreetal.,inpress). Thus, theremaining 60,000 molecularweightof

P90, the A-MuLV-specific sequences, is

suffi-cient toencode for anytransformingfunctions thesesequencesdirect.Inaddition,the similar-ity of lymphoid markers expressed by clones transformed with all of these isolates shows that the addition ofsequences inA-MuLV(P160)and the loss ofsequencesinA-MuLV(P90) doesnot

affecttargetcell selection.

The P160 strain ofA-MuLV appears tobea

naturallyoccurring isolate thatprobablyis the

predominant component of A-MuLV stocks maintainedattheNational Institutes ofHealth bypassageoftumorextracts. R. Risser and D.

Grunwald, usingA-MuLV derived from the

Na-tional Institutes of Health pool ofvirus, have

identified a 160,000-molecular-weight protein

thatexpressesp12determinantsinsomeof their A-MuLV-induced tumors (R. Risser, personal

communication). This protein is probably the

same asthe P160 describedhere.

The P160 protein probably contains about

130,000 molecular weight of A-MuLV-specific

sequences;basedontheabilityofA-MuLV(P90)

totransformcells,overhalfofthesesequences

must notberequiredtotransformcellsin vitro.

Whetherthese sequences are beneficial to the

virusgrowthandoncogenicityinvivo isunclear.

However, the predominance ofA-MuLV(P120)

in another tumor extract virus pool indicates thatA-MuLV(P160) is notalways selected byin

vivo passage.Whether differences in the passage history of the Boston and National Institutes of

Health pools of A-MuLV are important in the

derivation of two A-MuLV strains cannot be ascertained at thistime.

The genomes of the new A-MuLV strains have notbeenfullycharacterized. However, the

3.6-kilobase central region of the A-MuLV(P120) genome is large enough to encode 130,000

mo-lecular weight of protein. Thus, if the

A-MuLV(P160)genome is the same size as the

A-MuLV(P120) genome, P160 would account for

the entire codingcapacity of the genome.

The origin of theshorter A-MuLVproteins is

unclear. Thepossible role of fluorodeoxyuridine

treatment in their origin must be considered. although neither P100 or P90 appears to be directly relatedtothemorphological variations observedin thefluorodeoxyuridine-treated cells,

20 random Ab-NRK clones did not synthesize smaller forms of P120 (0. Witte, unpublished

data). The Ab-NRK cells were originally iso-lated using virus from ANN-1 cells. We have

shown that this strain of A-MuLV is stable through repeatedcloningin mousecells(26) (see above).

Thelymphoidmarker studies indicate that all

thestrains of A-MuLV transform similartypes

of cells. In the case of A-MuLV(P160), which

was originally detected in cell lines ofpeculiar

differentiationtypeforA-MuLV,it isclear that

the P160 protein is not responsible for these

alteredphenotypes. Our dataareconsistent with thoseof Raschkeetal. (11),whoshowed that

A-MuLV produced byRAW309Cr, amacrophage

cellline, produced typical Abelsonlymphomas.

Thepercentageof A-MuLV clonesexpressing

,uchain is similartothefrequency reported by

Sidenetal.(22)usingthesamemethodtoisolate

thetransformed clones. The

IL-only

phenotypeis the most common immunoglobulin phenotype

expressed by A-MuLV-transformed lymphoid

cellsisolated in vitrousingfocal transformation

assays.Therelationships,ifany,of these,u-only cellstonormallymphocytesis unclear. No

nor-mal ,u-only cell has been detected, although

suggestiveevidence of,u-onlycells exists (10;E.

Siden and D. Baltimore, personal

communica-tion). Sucha cellmightbe theprecursor of the

cytoplasmic IgM-positive pre-B cells. The

abil-itytostimulatecomplete IgMsynthesisinsome

A-only

A-MuLV-transformed lymphoid cell lines byusing the B lymphocytemitogen

lipo-polysaccharide (19) suggests that such a

path-way may exist.

TdT expression by A-MuLV-transformed

cellshasbeenpreviouslyreported(4,23),butno

prospective surveys have beenconducted. This

J. VIROL.

http://jvi.asm.org/

(8)

study shows that almost all

A-MuLV-trans-formed cells express TdT. Many of the

TdT-positive clones also synthesized,i chain. These results support thehypothesis suggested by Sil-vertone et al. (23), that TdT expression is not

limited to committed pre-T-lymphocytes, but may characterize alltypesofimmature

lymph-oidcells.

The lymphoid marker survey also indicates that thecellssusceptibletoA-MuLV in the bone marrowofBALB/cAn, C57BL/6J, andC57L/J

mice areprobably similar. Thisresult is

signifi-cant because these three strains differ in their

susceptibility to A-MuLV-induced tumors, BALB/c being susceptible and C57BL/6J and

C57Lbeingresistant (15).Whereasprevious

re-sultshave shown that bone marrow cells from these three strainsare susceptible to A-MuLV in vitro (16), the results describedhere demon-strate that similar A-MuLV-sensitive target

cells are present in all three strains. Thus, the resistance ofC57BL/6J and C57Ladult miceto

A-MuLV is probably not mediated by altera-tions in target cellpopulation.

ACKNOWLEDGMENTS

We aregratefultoDanClarkforexcellent technical assist-ance and to David Baltimore for his support and helpful discussions.

Thisworkwassupported by Public Health Service CA-24220andProgramProjectgrantCA-24530from theNational CancerInstitute (N.R.),bygrantVC-4Jfrom the American CancerSociety (to DavidBaltimore),andby Public Health Service grantCA-14051 from the National Cancer Institute (coregrant toS. E.Luria).N.R.is arecipientofanAmerican CancerSocietyResearch Scholar Award(Massachusetts Di-vision). O.N.W.is aHelenHayWhitney Postdoctoral Fellow.

LITERATURE CITED

1. Abelson, H.T.,andL.S. Rabstein.1970.Influence of prednisoloneonMoloney leukemogenicvirus inBALB/ cmice.CancerRes. 30:2208-2212.

2. Bassin, R. H., N.Tuttle, and P. J.Fischinger. 1971. Rapid cell cultureassay technique for murine leukemia viruses. Nature (London) 269:564-566.

3. Bonner,W.M., andR. A.Laskey.1974. A film detection methodfortritium-labeled proteinsandnucleicacids in polyacrylamide gels. Eur.J.Biochem. 46:83-88. 4. Boss, M., M. Greaves, and N. Teich. 1979. Abelson

virus-transformed haematopoieticcell lines with pre-B cell characteristics. Nature (London) 278:551-553. 5. Fan,H., and M. Paskind. 1974. Measurement of the

sequence complexity of clonedMoloneymurine leuke-mia virus 60 to 70 S RNA: evidence for a haploid genome. J. Virol.14:421-429.

6. Kessler, S.W. 1975.Rapidisolation of antigens from cells with a staphylococcal protein-A-antibody adsorbent: parameters of the interaction ofantibody-antigen com-plexeswith protein A. J. Immunol.115:1617-1624. 7. Laemmli, U.K. 1970. Cleavage of structural proteins

duringthe assembly of the head ofbacteriophageT4. Nature(London) 227:680-685.

8. Parks,W.P., R. S.Hawk,A.Anisowicz, andE. M. Scolnick. 1976. Deletion mapping ofMoloneytype C

virus:polypeptideand nucleic acid expressionin differ-enttransforming virus isolates. J.Virol. 18:491-503. 9. Potter,M.,M. D.Sklar,and W. P. Rowe.1973.Rapid

viral induction of plasmacytomas in pristane-primed BALB/c mice. Science 182:592-594.

10. Raff,M. C., M. Megson,J. J. T. Owen, and M. D. Cooper.1976.Early productionof intracellularIgMby B-lymphocyte precursors in mouse. Nature (London) 259:224-226.

11. Raschke,W.C., S.Baird,P.Ralph, andI.Nakoinz. 1978.Functionalmacrophage cell lines transformedby Abelsonleukemia virus. Cell 15:261-267.

12.Rasheed, S.,M.B.Gardner, andE.Chan.1976. Am-photropichost rangeof naturally occurringwild mouse leukemiaviruses.J. Virol.19:13-18.

13. Reynolds,F.H., Jr.,T.L.Sacks,D. N.Deobaghar, andJ. R. Stephenson. 1978. Cells nonproductively transformed byAbelson murine leukemia virus express ahighmolecularweight polyprotein containing struc-turalandnon-structuralcomponents. Proc. Natl.Acad. Sci. U.S.A.75:3974-3978.

14. Reynolds, R. K., W. J. M. van deVen, and J. R. Stephenson. 1978.Translation oftypeC viral RNAs inXenopuslaevis oocytes: evidencethat the 120,000-molecular-weight polyproteins expressed in Abelson leukemiavirustransformed cellsis virus coded. J. Virol. 28:665-669.

15. Risser,R.,M.Potter, andW. P.Rowe.1978.Abelson virus-inducedlymphomagenesis in mice. J.Exp.Med. 148:714-726.

16. Rosenberg, N., andD.Baltimore.1976. Aquantitative assay fortransformationof bone marrow cellsby Abel-sonmurine leukemiavirus. J.Exp. Med.143:1453-1463. 17. Rosenberg,N., andD.Baltimore. 1978.Theeffect of helpervirus onAbelson virus-induced transformation oflymphoid cells.J.Exp.Med. 147:1126-1141. 18. Rosenberg, N.,D.Baltimore, and C.D.Scher. 1975.

In vitrotransformation oflymphoid cells by Abelson murine leukemia virus. Proc. Natl. Acad. Sci. U.S.A. 72:1932-1936.

19. Rosenberg, N., E. Siden, and D. Baltimore. 1979. Synthesis of,u chains by abelson virus-transformed cells and inductionoflight chainsynthesiswith lipopolysac-charide,p. 379-386. In M.Cooper,D.Mosier,J.Scher, and E. Vitetta (ed.), Lymphocytes and the immune response. Elsevier, North Holland Biomedical Press, New York.

20. Rowe,W.P.,W. E.Pugh,andJ. W.Hartley. 1970. Plaqueassaytechniquesfor murineleukemiaviruses. Virology42:1136-1144.

21. Scher,C.D.,and R.Siegler.1975.Direct transformation

of 3T3cellsby Abelson murineleukemia virus. Nature (London)253:729-731.

22. Siden,E.J.,D.Baltimore,D.Clark,and N.

Rosen-berg. 1979. Immunoglobulin synthesis by lymphoid cells transformedinvitroby Abelsonmurineleukemia virus.Cell16:389-396.

23. Silvertone, A.E., N. Rosenberg, V.L. Sato,M. P.

Scheid,E.A.Boyse, andD.Baltimore. 1978. Cor-relating terminaldeoxynucleotidyltransferase andcell surface markersin thepathway of lymphocyte ontog-eny, p. 433-453. In B. Clarkson,J. E. Till,and P. A. Marks(ed.), Differentiation of normal and neoplastic hematopoietic cells. Cold Spring Harbor Laboratory, ColdSpring Harbor,New York.

24. Witte,0.N.,and D. Baltimore. 1978.Relationship of retrovirus polyprotein cleavagestovirion maturation studied withtemperature-sensitivemurine leukemia

vi-rus mutants.J. Virol. 26:750-761.

25. Witte,0.N.,N.Rosenberg, andD.Baltimore. 1979. Preparationofsyngeneictumorregressor serum reac-tive with the uniquedeterminants of the murine

http://jvi.asm.org/

(9)

kemiavirus-encoded P120proteinatthe cell surface. J. Virol. 31:776-784.

25a.Witte,D.N.,N.Rosenberg,and D.Baltimore. 1979. Identification ofanormal cellular protein cross-reactive

tothe major Abelson murine leukemia virusgene

prod-uct.Nature(London)281:396-398.

26. Witte,0.N.,N.Rosenberg,M.Paskind, A. Shields,

andD.Baltimore. 1978.IdentificationofanAbelson murine leukemia virus-encoded protein present in transformed fibroblast and lymphoid cells. Proc. Natl. Acad. Sci. U.S.A. 75:2488-2492.

27. Witte, 0. N., and D. Wirth. 1979. Structure of the

murine leukemia virusenvelope glycoproteinprecursor.

J.Virol.29:735-743.

http://jvi.asm.org/