• No results found

Purification and translation of murine mammary tumor virus mRNA's.

N/A
N/A
Protected

Academic year: 2019

Share "Purification and translation of murine mammary tumor virus mRNA's."

Copied!
12
0
0

Loading.... (view fulltext now)

Full text

(1)

JOURNAL OFVIROLOGY, July1981,p.207-218 Vol. 39, No. 1 0022-538X/81/070207-12$02.00/0

Purification

and

Translation of Murine Mammary Tumor

Virus

mRNA's

JAQUELIN P. DUDLEYt* AND HAROLD E. VARMUS

DepartmentofMicrobiology and Immunology, University of California, San Francisco, California 94143 Received 14January 1981/Accepted 15 April 1981

Wehave studied thefunctionsof the intracellular RNAs of mouse mammary tumor virus (MMTV) by purification and translation in vitro. Two major size classesof MMTV RNA, 35S and 24S RNA, were isolated from MMTV-infected rat(XC)cells and culturedmammary tumorcellsby preparative hybridization of

wholecellorpolyadenylated RNA to cloned MMTV DNA covalently bound to

chemicallyactivated paper disks (diazobenzyloxymethyl paper). Genomic-length

(35S) RNA was prepared free of 24S RNA by annealing to a 4-kilobase PstI

fragment apparently deficient in sequences homologous to 24S RNA; the 24S

species was separated from 35S RNA by rate zonal sedimentation in sucrose

gradients. Experiments using[3H]uridine-labeledcellular RNAindicated that the

preparative annealing methodwas highly specific and capable of effecting a

300-foldenrichmentfor viralRNA;the recovered RNA appeared to be intact under

denaturing conditions and directed synthesis offull-length gag and env

polypep-tides in vitro. Theproducts of in vitro translation were identified by gel mobility,

immunoprecipitationtestswithantisera against gag and env products, and partial

digestion with Staphylococcus V8 protease. The 35S RNA species directed

synthesis of several gag-related polypeptides, includingthreepreviously reported

in extracts of infected cells; 24S RNA directed synthesis oftwo polypeptides

closelyrelatedto envproteinsfrom infected cells.Therefore, 35S RNA includes

mRNA's forgag and gag-pol, whereas 24S RNA is the mRNA for env. These results help establish the position of env on the physical map of the MMTV genomeand bearuponthe codingpotentialofthe genome.

Theunusual features ofthe mouse mammary tumorvirus (MMTV), its regulation by

gluco-corticoid hormones (19,42) and its capacity to

induce mammarycarcinomas (19), command a

detailed understanding of the organization and

modeof expression ofthe viral genome.

How-ever,comparedwith otherretroviruses, MMTV isdifficulttopropagateinculture,andthere is

no convenient biological assay for infection of cultured cells. Since the definition ofgenesby classical methods hasnot beenpossible,

inves-tigatorshavereliedprincipallyuponmorerecent

biochemical methodsto obtainatentative pic-ture of the viralgenomeand of thestrategyfor its expression. Evidencetodate,gatheredfrom

analysisof MMTVproteinssynthesizedinvivo

and in vitro (5, 7, 26, 41) and of intracellular

speciesof viral RNA(16,38,43), conformstothe view thatthe genome ofMMTV is similar to that of otherreplication-competentretroviruses, with three essential genes, gag, pol, and env,

positioned from5' to3' onthe viral 35S RNA t Present address: McArdle Laboratory for Cancer Re-search, UniversityofWisconsin,Madison,WI 53706.

subunit. Furthermore, the mechanism of

MMTV gene expression appears to resemble that of otherretroviruses, with thegag andpol

genesbeingexpressedviacleavablepolyproteins

translated frommessengerRNAsof subunit size and theenvgenebeingexpressedviaa

polypro-tein, later modified, translated from a spliced,

subgenomic mRNA (24S RNA). Several

addi-tional RNA species have been described in MMTV-producing cells(16,38,43),butno func-tion hasbeenassignedtothem.

To test rigorously these prevailing views in

the absenceofgeneticanalysis,wehave

purified

intracellular MMTV RNAsbypreparative

hy-bridization techniques and subjected them to

functional analysis by translation in vitro.

By

usingcloned MMTV DNAlinkedtochemically

activated diazobenzyloxymethyl (DBM) paper disks(1,48,55)toselectvirus-specific RNA,the 35S and 24Sspecies,in

apparently

purified

and

intactforn,wereshowntoprogram the

synthe-sisofgag-relatedandenv-related

polypeptides,

respectively, inanin vitrotranslationalsystem.

These results support the current model for

organization and expression of the MMTV ge-207

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

208 DUDLEY AND VARMUS

nome.Inaddition, by using subgenomic restric-tion fragments of MMTV DNA, we have placed constraints upon theposition of env within the viral genome. Our inability to find other virus-specific RNAsin this studyargues against the existence of heretoforeunidentified genes (e.g., for nonstructural proteins) in the MMTV

ge-nome.

MATERIALS AND METHODS Cell lines and virus. Theratcell lineXC, derived fromatumorinduced inaWistarratby the Prague strain of Roussarcomavirus (49)wasobtained from J. Levy (University of California, San Francisco). MMTV-infectedXCcells (960) andaderivativeclonal line (D108) were obtained from D. Robertson (Uni-versity ofCalifornia, San Francisco). C3H (line H-1) (42) and BALB/cfC3H (39) mousemammarytumor cells were kindly provided by J. S. Butel (Baylor College of Medicine) andN.H.Sarkar

(Sloan-Ketter-ing CancerCenter), respectively. Cell lines were prop-agatedat37°CinDulbecco minimal essential medium

(GIBCOLaboratories)containing10% fetal calfserum or2%fetal calfserum and 8%calfserum (GIBCOor Irving Scientific Co.), 50Mug ofgentamicin (Schering

Corp.) per ml,and 125 ng ofamphotericin B (E. R.

Squibb&Sons)perml.

XC cellswere infected with MMTV as described

previously for mink lung cells (38). Virus used for infections and virion RNApreparationswaspurified from the C3H mouse mammary tumor cell line Mm5mt/cl by the Frederick Cancer Research Center (12).

Isolation of RNA. Total virionRNAwasprepared as describedpreviously (9), using asodium dodecyl sulfate (SDS)-proteinaseKdigestion, followed by ex-traction with phenol-chloroform. MMTV 70S RNA was isolated by sedimentation in 15 to30% (wt/wt) sucrose gradients containing10mM Tris-hydrochlo-ride,pH 7.5,0.1MNaCl, L mM EDTA, and 0.2% SDS. Gradients werecentrifuged at40,000 rpmfor3 hat 20°C inanSW41rotor. The70S fraction was

subse-quentlyheat-denatured in1mMEDTAat100°C for 1to 2minandlayeredonto a15 to30% (wt/wt)SW41 gradient containing 10 mM Tris-hydrochloride, pH 7.5,and10mMEDTA. Gradientswerecentrifuged at 35,000 rpmfor17hat4°C, and the RNA sedimenting atvalues greater than 16S(relativetorRNAmarkers) waspooled and used for complementary DNA (cDNA) synthesis andinvitro translations.

Isolation of totalcellularRNA and oligodeoxythy-midylate [oligo(dT)]-cellulose chromatography were performedasdescribedpreviously (9), with the

follow-ing exceptions. Total RNA was heated to 80 to 90°C in 1 mM EDTAbeforeitwas loaded onto

oligo(dT)-cellulose columns; the columns were then washed with threesequential Tris buffers containing either 0.5M, 0.1M,or noKCl,in additionto1mM EDTA.

Isolation andpurification of cloned DNAs. Un-integrated MMTV DNA from XCcellsinfected with the C3H strain of MMTV was isolated by the Hirt procedureasdescribedpreviously (44), cleaved with the restriction enzymePstI, and cloned into the single PstI site of pBR322 (J. Majors, unpublished data).

Thep7-lAandp2-lAcloneswerederivedbycleaving

MMTVsupercoiledDNAatthesingleEcoRI site and cloninginthevectorXgtWESXB.DNA from

recom-binantphagewasextracted,cleaved withEcoRI,and

subclonedinto the RI site ofpBR322togenerate

p7-1A andp2-lA.Byendonuclease restrictionmapping,

p7-lAappearedtolack sequencescorrespondingtoa

portion of the PstI-C and Dfragments and most of thePstI-Efragmentasdescribedonaphysicalmap of

unintegratedMMTV DNAbyShanketal.(44) (Fig.

1).p2-lA hasa moreextensive deletion than p7-lA,

includingpart of the PstI-Afragment.

Recombinant plasmids grown in Escherichia coli HB101 were amplified inthe presence of 170 ug of

chloramphenicol (Sigma Chemical Co.) per ml, and the DNA waspurified as described previously (18).

AfterchromatographyonBio-RadA50,the excluded material waspooled, and the supercoiled DNA was

recoveredafterequilibriumsedimentationonCsCl-PI2

gradients(37).

SynthesisofcDNA.MMTV cDNA's which rep-resentthe 3' terminus of viral RNA (cDNA3r)orthe entire genome(cDNArep)weresynthesizedin reactions using MMTV virion RNA andoligo(dT)12-18and calf thymus oligonucleotides as primers, respectively, as

describedpreviously (38, 44).

Labeled DNAscomplementary to cloned MMTV DNAweresynthesizedasfollows.SupercoiledMMTV DNAcloned inpBR322wasdigestedwith the restric-tion enzymePstI(obtainedfrom J.Majors)in 20 mM

Tris-hydrochloride,pH7.5,10mMMgCl2,and50mM

(NH4)2SO4 at 37°C for 1 h, extracted twice with

phenol-chloroform, and ethanol precipitated. The DNAwasdilutedto 100ug/ml,heatedto100°Cfor5

p C 4,J

LS3.

E

P P

A I84i D

E S(A)n3' 35S RNA

n^l 'an. 35 N

909 Poi env

5' ID-en-(A) 3' 24S RNA

env

FIG. 1. Physical and genetic maps ofMMTVDNA and RNA. The positions of restriction endonuclease sitesfor Pst (P), EcoRI (E), and SacI (S) on the linear, unintegratedform of DNA from the C3H strain of MMTVare illustrated in relation to the probable positions of the accepted coding domains in the MMTV genome, gag, pol, and env. Sequences from the3'end(open boxes) and the 5' end (cross-hatched boxes) ofviralRNAare present at both ends of viral DNAinastructurereferred to as the longterminal

repeat. The major species of MMTV RNA, 35S and 24S RNA,aredrawnbelow the DNA, with the box at the5'endof both RNAs, indicative of sequencesfrom

the5'terminusof viral RNA joined tosequences on the 5'side of the env gene in 24S RNA, as suggested by work presented here and elsewhere (38). Hybridi-zation reagents used in this paper includecDNA, transcribed randomly from genomic (35S) RNA; clonedPstIfragments A (4.0 kb), B (1.8kb), and C (1.3kb); and cloned incomplete MMTV DNA mole-culeswhichincludeallof the DNA on theright side ofthemap from at least the EcoRI site to theSacI

site(p7-lAandp2-lA;J. Majors,unpublished data).

Annn

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.493.255.450.387.447.2]
(3)

VOL. 39, 1981

min, andquicklycooled onice. Twohundred

nano-gramsof thisDNAwasaddedtoamixturecontaining

250 MCi of[aL-3P]dCTP (300 Ci/mmol;NewEngland NuclearCorp.), 120 MM each of dGTP, dATP, and dTTP, 100 Mug of calf thymus oligonucleotideprimers,

62.5 mMTris-hydrochloride, pH 8.1, 10mM MgCl2,

50mMKCI,2.5mMdithiothreitol, and200U of avian myeloblastosisvirusDNA polymerase (providedby J. Beard, Life Sciences,Inc.)permlinavolume of200

pi.

Thereactionwas incubated at370Cfor 1 h, ad-justed to 20 mM EDTA, 1% SDS, and 100 Mg of

proteinaseKperml, and theincubationwascontinued

for 30 min. Approximately 40 Mg ofsingle-stranded

salmonspermDNAwasadded,and the mixturewas

extracted withphenol-chloroform. Subsequently, the reactionwasadjustedto0.3N NaOH, incubated for1 hat500C, and neutralizedwith 1.2 NHCI.The unin-corporated radioisotopewasremovedbyG-50 column

chromatography. Specific activities were

approxi-mately108to2x108cpm/MLg.

Agarosegelelectrophoresis of RNAand trans-fer to diazotized paper. Total cellular RNA was

subjectedtoelectrophoresison1.2%agarosegels

con-taining 10mMmethylmercury hydroxide (Alfa Divi-sion, Ventron Chemical) as described previously (2, 38). Preparation and activation of DBM paper and

alkali treatment of methylmercury gels were

per-formedasdescribedby Alwineetal. (1). AfterRNA transfer,DBMpaper wasprehybridizedforatleast5 h at410C in a solution containing 50% formamide,

0.1%polyvinyl pyrrolidone (PVP-360; SigmaChemical Co.),0.1% Ficoll(Sigma), 0.1% bovineserumalbumin,

0.1%SDS, 10 mM

N-2-hydroxyethylpiperazine-N'-2-ethanesulfonicacid(HEPES) buffer, pH7.4,1mgof

yeast RNAper ml, 0.45 M NaCl, 45mM trisodium citrate, 100 Mgofsingle-stranded salmonspermDNA per ml, and 1%glycine. TheRNA-DBM paper was

then hybridized to 2 x 106 to 5 x 106 cpm of

[32P]cDNAintheabovesolution lacking glycinewith the addition of 10% (wt/vol) dextran sulfate (Phar-maciaFineChemicals, Inc.)(52) for6to18hat410C. Hybridizationwas terminated bywashing thepaper

several times in0.lxSSC(0.15mMNaCland1.5mM trisodium citrate) at roomtemperature and in 0.1x SSC and 0.1% SDS for1 hat530C. The paper was

then dried andsubjected toautoradiography as

de-scribedby Robertson and Varmus(38).

PreparationofDNA-DBMpaperdisks. Purifi-cationof viral mRNA'swasachievedbyusing DBM

paperessentiallyasdescribedbyAlwineetal.(1) and modifiedby others (35, 48, 55).Purified supercoiled recombinantplasmidswerelinearized withan

appro-priaterestrictionendonuclease(eitherPstIorEcoRI),

extracted withphenol-chloroform,andethanol precip-itated. In thecaseofp7-lA,plasmidDNAwas

sepa-rated from MMTVsequencesbycentrifugationon10 to 40%(wt/vol) sucrose gradients containing10mM Tris-hydrochloride, pH 7.5, 0.1 MNaCl, and 1 mM EDTA in a Beckman SW27 rotor at 25,000rpm at 200Cfor 30 h. The DNApreparations (including

ap-proximately 10,000 to 20,000 cpm of MMTV [nP]

cDNA,,p) wereresuspendedin 25mM sodium phos-phatebuffer, pH 6.0, at aconcentration of5 mg/ml and heated for 2minat80 to 1000C. Subsequently, theDNAwascooledonice, dilutedto 1mg/mlwith

MMTV mRNA's 209 dimethylsulfoxide, and addedto1-cm2activated DBM paperdisks. The diskswereallowedtostandovernight at roomtemperatureand then washed with1literof water percm2, fourtimeswith 0.4 N NaOH (1 ml/

disk)at370Cfor10min, andagainwith threechanges

of water. The paper disks were subsequently moni-tored for DNA binding by Cerenkov counting; the percentageof DNA boundrangedbetween15and 25% of the cDNAadded,and 6to11,ugofMMTV DNA wasboundtoeach disk. Diskswerestoredat40Cin 50%formamide containing20mM1,4-piperazine die-thanesulfonic acid(PIPES) buffer, pH6.4.Disks have been used for up to 17 consecutivepreparative an-nealings without apparentdeterioration.

DNA-filterhybridizations.Approximately 10to 50 Lgofpolyadenylicacid[poly(A)]-containingRNA was suspended in 50Au ofa buffer containing 50% deionizedformamide, 20mMPIPES buffer, pH6.4, 0.4 M NaCl, 5 mM EDTA, 0.4% SDS, 250 jg of

polyriboadenylicacid[poly(rA)],2xDenhardt's buffer (6) (diethylpyrocarbonate treated), and 170 iLg of chicken liver tRNA and subsequentlyhybridized to DBM-DNA-containing disks at 410C for 5 to 10h. DNA disks were briefly alkali-treated with 0.4 N NaOH and then neutralized with 20 mM PIPES buffer,pH 6.4, beforehybridization.

Hybridizationwasterminatedbywashingthe disks four times in0.1xSSCat roomtemperatureand five

timesin0.1x SSC and 0.1% SDS for5 min each at

530C. Purified RNA was eluted in 100,u of buffer

containing 95% deionizedformamide,10mMHEPES,

pH 7.4,3mMEDTA,and 0.1%SDSat680Cfor1to 2 min. The elution was repeated once with 50 ,lI of formamide buffer; the eluates werepooled and ethanol precipitated.

RNAlabeling.Cell cultureswereexposedto 10-5 Mdexamethasone in growth medium for 24 to 48 h, to labeling medium(growth medium containing 10-5M dexamethasone and 2% dialyzed fetal calf serum as a substitute for other sera) for 12 h, and to labeling mediumcontaining0.5 mCi of[3H]uridine (25 to 30

Ci/mmol) permlfor24h.Cellswerethenextracted for RNAasdescribed above.

In vitro translation- and polyacrylamide gel

electrophoresis.Translationswereperformedas de-scribed by Weiss et al. (53). Reactions contained

poly(A)-containing RNAat concentrations of50

pg/

mlorless,75mMKCI, 1.7mMMgCl2,50,MMamino acids exceptmethionine,20mMHEPES, pH 7.8,10 mM creatinephosphate,1mMATP,2mMGTP,0.3 mMspermidine,50

Mg

of creatinephosphokinaseper

ml,approximately 500MCiof[3S]methionine (500to 1,200Ci/mmol)perml, and 76% micrococcal nuclease-treated rabbitreticulocytelysatesinatotal volume of 20

pl.

Reticulocyte lysateswere nuclease-treated by

themethod of Pelham and Jackson(33).RNAsamples

purified byhybridization to DNA-DBM paperwere washedonce in66%ethanol and67mMsodium ace-tate, pH 5.0, before translation. Translations were incubatedat280Cfor90 to 120minandtheneither used for immunoprecipitation or diluted in 2x gel samplebuffertofinalconcentrations of62.5mM Tris-hydrochloride, pH 6.8,2% SDS, 10%glycerol, 5%

,B-mercaptoethanol,and0.001%bromophenolblue

(sam-plebuffer).

on November 10, 2019 by guest

http://jvi.asm.org/

(4)

DUDLEY AND VARMUS

Polyacrylamide gel electrophoresis was performed according to Laemmli(22), withmodifications by Op-permannetal.(27). Afterfixation, gelsweresoaked in 1 M sodium salicylate for 30min, rinsed briefly in water,dried, and exposed to preflashed Kodak Royal X-Omat filmat-70°C.

Antisera and immunoprecipitations. Antisera against the MMTV structural proteinswereprepared bymultiple injections ofpurified proteins into rabbits andwerekindly suppliedby D. Robertson. Immuno-precipitations were performed essentially as described by Oppermann et al. (28), using Formalin-fixed Staphylococcus aureus to precipitate immune com-plexes (20).

Immunoprecipitates of MMTV structural polypep-tide precursors were obtained by incubating dexa-methasone-treated H-1cells with 500,uCiof [3S]me-thionine per ml for45min before lysis of thecells. D. Robertson supplied immunoprecipitates of proteins from MMTV-infected rat cellstreated with 1jig of tunicamycin per ml.

Proteasemapping.Gel slices foranalysis by par-tial digestion with protease (4)werecutfrom the gel, washed in waterfor 30 min, and then washed in0.1x gel samplebuffercontaining 10% glycerol for10min. Sliceswereplacedinsamplewells and overlaid with 50to 75plof 4

jig

ofStaphylococcus V8 protease per ml inthe sample buffer mentioned above. Samples wereanalyzed bypolyacrylamide gel electrophoresis at 15 mA through a 4.5% stacking gel and a 14% separating gel. Gelswere subjected tofluorography, dried, and exposedtofilmaspreviously described.

RESULTS

Identification and mapping of MMTV

RNA species by

hybridization

to RNA

bound to DBM paper. Before

attempting

to

purifyputativemRNA'sfor MMTVproteins by

preparative hybridization, the majorspeciesof

virus-specific intracellular RNAweredefinedby

using analytical techniques. Whole cell RNA was isolated from MMTV-infected rat cellsor mousemammarytumorcells aftergrowthin the presenceofdexamethasone,fractionated

by

aga-rose gel electrophoresis under denaturing

con-ditions, transferredtochemicallyactivated

cel-lulose (DBMpaper),anddetectedwith various

32P-labeledreagentsderived fromMMTV DNA cloned in procaryotic hosts (Fig. 2 andabove). Twospeciesof MMTV RNAwere observed in these experiments. The larger species, ca. 8.0

kilobases(kb) (referredto as35SRNA),wasthe

same length as a subunit of the viral genome

and, as expected,annealed toall ofthe probes

tested (PstI-A [4.0kb],PstI-B[1.8 kb],PstI-C

[1.3

kb],

p7-lA,

and

cDNArp).

Forseveral

rea-sons, RNA of this size is predicted to direct

synthesisof gag andpol proteins (5, 25, 27).The

smaller species, ca. 3.1 kb (referred to as 24S RNA), annealed to probes representing the 3' one-third of viral RNA, PstI-B (1.8 kb), and

FIG. 2. Characterization ofintracellular MMTV RNA byhybridization with cloned PstIfragments.

Total cellular RNA was isolatedfrom MMTV-in-fected XC cells(line D108)or mousemammarytumor cell linesafter growth for24 to72h inthe presenceof

dexamethasone. RNA was fractionated by electro-phoresis in 1.2% agarosegels containing

methylmer-curyhydroxide, transferredtoDBM paper, and de-tectedbyannealingwith 5x 106cpmof[32P]PstI-B

(1.8kb) (lanes1through 4)orPstI-A(4.0kb) (lanes5 through8).Autoradiographic exposureswerefor12 h(lanes3,4, 7and8)or13days(lanes1,2,5and6). Thearrowsindicate thepositionsofviralspecies of 35S and24SRNA; the blanchedareasofthefilms

correspondtothepositionsof28S and18S rRNA's in thegel. Lanes I and5:960cellularRNA;lanes2and 6: GR cellularRNA; lanes3and 7: BALB/cfC3H cellular RNA; lanes 4 and 8: H-1 (C3H) cellular RNA.

PstI-C (1.3 kb), but not to the PstI-A (4.0 kb)

fragment derived from thecentral portion of the

viralgenome (seeFig.2and unpublished data).

Identical results were obtained byusing RNA

derived fromGR, C3H,or BALB/cfC3H tumor

cell lines or from MMTV-infected XC cells.

These results conform to previous suggestions

that24S RNA isasubgenomicspecies

respon-sibleforsynthesis of proteins encoded near the

3'end of viral RNA(16,38,43). However, due to

high backgrounds observed with the PstI-A probe and its probable content of 24S RNA sequences (see below), low levels of hybridiza-tion to 24S RNA may have been overlooked. None ofthe other previously described species of MMTV RNA (16, 38, 43) were detected in these andsubsequent analyses; the reasons for this areconsideredbelow.

The results presented in Fig. 2indicated that mostif notallof the env gene maylieon the 3' side of the PstI site which divides PstI-A (4.0

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.493.259.446.70.257.2]
(5)

MMTV mRNA's 211

kb) from PstI-B (1.8 kb) and that itmight be

possibletoseparate35S RNA from 24SRNA by

preparativehybridizationtothePstI-A (4.0 kb) fragmentimmobilizedon DBMpaper.

Characterization of the preparative

hy-bridization method for purification of

MMTVRNA. Other laboratories havereported theuseof cloned DNA linkedtoDBMpaperfor the preparative isolation of defined species of RNA (46, 55).The specificity of this procedure

wasexamined forourpurposes bydetermining the proportion of RNA which binds adventi-tiously to filters containing nonhomologous

DNA and by measuring the enrichment

achievedbyannealing RNAtofilterscontaining viralDNA.Approximately3.3%of

[3H]uridine-labeled, polyadenylated

[poly(A)+]RNA

and

0.4% of poly(A) RNA [RNA not binding to

oligo(dT)-cellulose] from a line of MMTV-in-fected XC cells (line D108) annealed to filters bearing DNA fromahybrid plasmid composed of pBR322 sequences and mostof the MMTV genome (p7-lA) (Table 1). These results are

consistent with previousobservations that about

0.1 to0.5% of total cellular RNA is virusspecific in dexamethasone-treated cells of this lineage and that a 10- to 20-fold enrichment for viral RNA can be effected by selection of poly(A)+ RNA (38). Only 0.01% of thesameRNA

prepa-ration boundtofilters containing pBR313DNA

alone, indicating that the annealing step can

provide a ca. 300-fold enrichment for viral

se-quences.

[3H]uridine-labeled

poly(A)+ [and

poly(A)-]

RNA from uninfected XC cells also bound inefficiently (0.01 to 0.02%) to filters carrying either pBR313orp7-lA DNA. Similar

values for"background" adherencetoDBM

pa-TABLE 1. Quantitationof[H]RNA hybridization toDNA-DBM paperdisksa

DAo Source

DNAon ofla- Poly(A) cpm cpmhy- %RNA DBM aeddfraction added bridized hybrid. dikb bldfato (XI106) (Xl103) ized

RNAc

p7-1A D108 + 1.49 48.60 3.26 p7-1A D108 - 3.05 11.60 0.38

p7-lA XC + 1.49 0.30 0.02

p7-1A XC - 1.59 0.07 0.01

pBR313 D108 + 1.43 0.17 0.01

pBR313 XC + 1.49 0.13 0.01

aLabeled RNA was hybridizedto DNA disksin a50% formamidehybridizationbuffer(seethetext)for6.5hat41°C. Diskswerewashed,and thehybridizedRNAwaseluted and counted inSml ofTritosolscintillation cocktail(13).

DNA(9to 11gg)fromp7-lA (apBR322hybridplasmid containingmostof the MMTVgenome)orfrompBR313was

boundtoDBM paper disksasdescribed in thetext.

'MMTV-infectedXCcells(lineD108) oruninfected XC

cellswerelabeled for24hwith[3H]uridine(500yCi/ml)before preparation of whole cell RNA and fractionationby oligo(dT)-cellulosechromatography (seethetext).

perhave beenreported by Stark and Williams

(48).

Results shown in Table 1 wereconfirmedby

analysis of the size and sequence

specificity

of theselectedspeciesof RNAinexperimentsalso designed to determine whether intact MMTV RNA survived the selection

procedure (Fig.

3).

[3H]uridine-labeled

RNA from D108 cells was

divided into poly(A)+ and poly(A)- fractions,

and eachwas annealed

preparatively

tocloned

MMTV DNA (p2-lA) bound to DBM paper.

FIG. 3. Analysis ofRNA fromMMTV-infectedXC cellsafter hybridization to MMTV DNA-containing DBM paper. MMTV-infected XC cells (line D108) werelabeledfor24hwith[3H]uridine(500yCi/ml),

andpoly(A)+ andpoly(A)J RNAfractionswere pre-pared by oligo(dT)-cellulose chromatography. After incubationofeachfractionwith MMTVp2-IA DBM paperdisks, unbound and bound components were concentrated and subjected to electrophoresis in a 1.2% agarosegel containing methylmercury hydrox-ide. [Approximately 1.52% (3.0 X 104 cpm) of the

poly(A)' RNA and 0.06% (7.6 x 103 cpm) of the poly(A)-RNAwereboundtotheDBM disks in this

experiment,andonly20 to50%of each was recovered

after ethanol precipitation.] After electrophoresis,

one-half ofthegel(panel A) waspreparedfor fluo-rography and exposedtofilmfor24 days; RNA in theotherhalf (panel B)wastransferredto asheet of DBMpaperandhybridizedwith[32PJcDNA,,p(2 x

10i cpm). Panel B showsanautoradiogramexposed

for19days. Lane1:25%ofthepoly(A)+ RNA which annealedtoDBMdisks;lane2:0.3%(panel A)and 5%(panelB)ofpoly(A)J RNAnotannealedtoDBM disks; lane 3: 25% ofthepoly(A)- RNA which an-nealedtoDBMdisks;lane4:0.15%(panelA) and 5% (panel B) of poly(A)- RNA not annealed to DBM disks.

VOL. 39, 1981

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.493.256.449.193.410.2]
(6)

212 DUDLEY AND VARMUS

Portions of RNA whichhybridized orfailed to

hybridize in this step were then subjected to agarosegelelectrophoresis and examined either by fluorography (panel A) or by hybridization with[32P]MMTVcDNArepandautoradiography

(panel B). Thetestswith

cDNArep

showed that

most (ifnot all) of the intact MMTV 35S and 24S RNAwasin thepoly(A)+fractions (lanes1

and2) and that atleast 20% of the viral RNA had been selectedby annealing toDBMfilters under these conditions.(Thiswasdeduced from the similar intensities of bands obtained with one-fourth of theselectedRNA[approximately

1.5% of total RNA] [lane 1, panel B] and an

aliquot [5%] of the unselected RNA [lane 2, panel B]). The marked enrichment for virus-specific RNApredictedfrom theresultsinTable

1 was evident from the fluorogram. Thus, the poly(A)+ RNAwhichfailedto annealto DBM

filters contains manyspecies, amongwhichthe expected viralspeciescannotbeunambiguously

identified (lane 2), whereas the poly(A)+ RNA which did annealtofiltersbearingMMTV DNA consists of only two discrete species (lane 1,

panel A) comigrating with the 35S and 24S species detected with

[32P]cDNAr,,p

(lane 1,

panel B). The poly(A)- RNA which did not

adheretothe DBM diskswascomposedchiefly

of28S and 18S ribosomalspecies (lane4,panel

A), andnodiscretespecieswerediscerned in the poly(A)- RNA which boundtothe disks(lane3, panel A).

Purification and fractionation of MMTV

RNA bypreparative hybridization to PstI

fragments.Preliminarytestsof thepreparative

hybridization method(Table1,Fig.3)employed

clonedMMTV DNA representing most of the viral genome. Basedupon results presentedin

Fig. 2, it appeared that preparative annealing with clonedPstIfragments wouldallow separa-tion of the two major species ofviral RNA as

wellasmarked enrichment forviralRNA. This prediction was confirmed in the experiments showninFig.4.[3H]poly(A)+RNA wasisolated from MMTV-infected XC cells(line 960)orfrom

culturedC3H mammary tumorcells (line H-1)

andannealedpreparativelyto DBM disks

bear-ing p7-lA, PstI-B (1.8 kb), or PstI-A (4.0 kb)

DNA.Hybridizationtothep7-lAorPstI-B (1.8 kb) resulted in selection of both major species of viral RNA (lanes 1 and 2, panels A and B), whereashybridizationto thelargerPstfragment

selected only the 35S RNA. This result

con-firmed earlier evidence (Fig. 2) that 24S RNA

containslittleornosequence homology with the PstI-A(4.0kb) fragmentandalso indicated that

the 35S species could be separated from 24S

RNAwithout fractionation bysize.

Products of translation of MMTV RNA

FIG. 4. SelectionofMMTV intracellularRNAsby

preparative annealing to cloned PstI fragments. MMTV-infectedXC cells(line 960; panelA)or virus-producingC3H mammarytumorcells(panelB)were grownforca. 72h with dexamethasone and labeled with [3H]uridinefor24 h at500,uCi/ml. Poly(A)+

RNAwaspreparedfromwhole cells and annealedto DBMpaperdisksbearingeitherp7-lADNA(lane 1),

PstI-B (1.8kb)DNA(lane 2),orPstI-A(4.0kb)DNA (lane3)for9hat41'C.Afterelutionofthehybridized

RNA, sampleswere analyzed byelectrophoresis in 1.2% agarosegelscontainingmethylmercury

hydrox-idebeforefluorography. The bandsofslowly

migrat-ing material inpanelA were shown to represent labeled cellular DNA and could be eliminated by

pretreatmentofthesamplewith3Msodium acetate, pH6.0(9) (cf. panelB) orDNase(datanotshown).

Thegelswerecalibratedbyinclusionof28S and 18S rRNA inparallel lanes. Autoradiograms were ex-posedtofilm for2days(panelB)or3days(panel A).

selected by preparative hybridization. To

assessthe coding potential of the 35S and 24S

RNAspecies obtainedby preparative annealing toDNAboundtoDBMpaper disks,C3Htumor cell RNAselected bythisprocedure was

trans-latedinan invitroprotein-synthesizingsystem prepared from rabbit reticulocytes. Products wereexaminedinpolyacrylamidegels and

com-pared with products oftranslation of

subunit-sized virion RNA (Fig. 5,panel A); in addition,

the productswere analyzed by immunoprecipi-tationwithantisera directed against products of

thegag and env genes (p27 and gp52,

respec-tively) (Fig. 5, panel B).

Virion RNAand RNA selected by annealing to either PstI-A (4.0 kb) or PstI-B (1.8 kb)

directed synthesis of a series of polypeptides

immunoprecipitable with anti-p27 antisera, as well as severalsmallerpolypeptides not precip-itablewith thisantiserum. Although these

gag-related products migrated as doublets for

un-known reasons, most of them appeared to be

similar in size (ca. 75,000, 110,000, and 160,000 Mr) to previously described gag and gag-pol

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.493.280.422.70.218.2]
(7)

MMTV mRNA's 213

FIG. 5. Translationproducts ofMMTVRNAselectedby hybridizationtoPstIfr-agments.

[MSJmethionine-labeledpolypeptides

synthesizedinanuckease-treatedrabbitreticulocytesystem(seethetext)wereanalyzed

byekectrophoresisin 7.5%discontinuouspolyacrylamide gels before(panel A)orafter (panel B)precipitation

with indicated sera. Panel A: Productsofsynthesis withnoadded RNA (lane 1); with0.5pgof26 to40S MMTV virion RNA(lane2);withpoly(A) RNAfr-omC3Hmammary tumorcellsafterselectionby annealing

toDBM diskscontainingPstI A(4.0kb)DNA(lane3);and withpoly(A)'RNAfrom C3Htumorcellsafter

selectionbyannealingtoPstI B(1.8kb)DNA(lane 4).Panel B:Portionsofthe translationproductsshown inpanelAwereprecipitatedwithnormnal rabbitserum(lane1),anti-p27"'9serum(lanes2,4, and6),or

anti-gp52Vserum (lanes 3,5, and7).

Polypeptides

inlanes1through3weref-omthesample analyzedin lane2

ofpanelA;polypeptidesinlanes4and5werefr-omthesampleanalyzedinlane3ofpanelA;andpolypeptides

in lanes6and 7werefrom thesample analyzedin lane4ofpanelA. PhosphorylaseB (94,000Mr), bovine

serumalbumin (67,000)Mr),ovalbumin(43000

Md)

and carbonicanhydrase(30,000Mr)servedasmolecular

weightmarkers.

translationproductsofgenomic andintracellular 35S RNA (5, 26, 43). In addition, we observed gag-related products ofca.60,000 and 35,000 Mr

in all three samples but have not investigated these further. These resultsconfirm earlier

con-tentions that the 35S RNA species includes the mRNA'sforgag and gag-polproteins (38).

Experiments presented in previous sections demonstrated that RNA selected byannealing

tothePstI-B(1.8 kb) fragmentcontains the24S species, but RNA annealed to the PstI-A (4.0

kb) fragmentdoes not. Translationproducts of the RNAselectedwith thePstI-B(1.8 kb) frag-mentincludedpolypeptidesof68,000 and 70,000

Mr(panel A,lane4)notevidentamongproducts

of the other translations (panel A, lanes 1-3).

These polypeptides were precipitated by

anti-gp52 serum (panel B, lane 7), whereaslittle or noproteinwasprecipitated from theother

trans-lation mixtures bythis serum (panel B, lanes3

and5). Suchresultsindicate that the 24Sspecies of MMTV RNA directs the synthesis of env

polypeptides, confirming the predictions out-lined in Fig. 1.

Translation of 24S RNA purified after separation from 35S RNA. Evidence in the

precedingsection for thecodingfunctionof 24S

RNA was based upon translation of 24S RNA

copurifiedwith35S RNA. Since the sequences in24S RNA appear tooverlap completelywith VOL. 39, 1981

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.493.130.366.70.373.2]
(8)

214 DUDLEY AND VARMUS

thecontentof 35SRNA,separationof 24S RNA

from 35S RNA required fractionation by size. RNA fromcultured C3Hmammary tumorcells

was subjected to rate zonal sedimentation in

sucrose,pooledinfractionsappropriatefor35S and 24SRNA, concentrated,and annealed

pre-parativelyto DBMpaperdisksbearing PstI-A

(4.0 kb) andPstI-B (1.8kb), respectively. The recovered RNAwasthen translated invitro,and thelabeled translationproductswereexamined bypolyacrylamide gelelectrophoresis with and without prior immunoprecipitation (Fig. 6). A

second experiment was conducted in

parallel

with RNA from MMTV-infected XC cells and yielded similarresults (datanotshown).

Translation of the purified 24S RNA

gener-ated two polypeptides with sizes (68,000 and 70,000Mr) similar to those of the env-specific

proteinsobserved in the previous experiment,as well as several additional polypeptides (Fig. 6,

panel A, lane 2); the two major proteins and several of the minorproductswereprecipitated by anti-gp52 serum but not by p27 antiserum (panel B, lanes1and2). The mobilities of these polypeptides differed from those synthesized by 35S RNA (panel A, lane 3); moreover, these polypeptides were precipitated by anti-p27 (panel B, lane 4) butnotbyanti-gp52sera(panel B, lane 3). These results strongly suggest that 24S RNA, composed principally of sequences

[image:8.493.135.371.247.540.2]

from the PstI-B (1.8 kb) and PstI-C (1.3 kb) fragments, is the mRNA forenvelope proteins

FIG. 6. Translationproducts ofsize-fractionated24S and 35S RNA. Wholecell RNA waspreparedfrom

dexamethasone-treated C3H mammary tumor cells,selected twice on oligo(dT)-cellulose, and sedimented in ratezonal sucrose gradients as described in the text. Fractions containing 24S and35SRNA were concen-tratedby ethanol precipitation, and the RNA was annealed to DBM paper disks bearing either PstI-B (1.8 kb) (for 24S RNA)orPstI-A (4.0 kb)(for35S RNA). After elution, the hybridized RNAs were translated in vitro(cf. Fig.5and thetext) and the

[?S]methionine-labeledproducts

were subjected directly toelectrophoresis ig 7.5%polyacrylamide gels (panel A) or immunoprecipitated with the indicated antiserum before electropho-resis (panel B). Panel A. Products of translation with: no added RNA (lane1); with 24S RNA (lane2);and with 35S RNA(lane 3). Film exposure was for 6 days for all lanes. Panel B. Products of translation with 24S RNAafter immunoprecipitation with anti-gp52 serum (lane 1) or anti-p27gw serum (lane 2); with 35S RNA after precipitation with anti-gp52 serum (lane 3) or anti-p27 serum (lane 4); with 16-26S virion RNA after precipitation with anti-gp52 serum (lane 5). Film exposure was for 6 days for all lanes.

on November 10, 2019 by guest

http://jvi.asm.org/

(9)

VOL. 39, 1981

and that theenv gene mustliewithinca.3 kb of

the 3' terminus of viral RNA (see Fig. 1). In agreementwith this prediction, the synthesisof envproteinswas also observed frompoly(A)+16 to 26S RNA prepared from virions (panel B,

lane5).

Identificationof translationproducts by

mappingwithStaphylococcusV8protease.

Todocument theidentity of putative env

poly-peptides synthesized in vitro from purified

FIG. 7. Partial proteasedigestion ofpolypeptides

synthesized from purifiedmRNA.Polypeptides were

excisedfrom 7.5%polyacrylamide gels anddigested withStaphylococcus V8protease (seethetext). Prod-ucts were separated on a 14%polyacrylamide gel. Lane 1, 63,000-Mr polypeptide immunoprecipitated with gp52 antiserum from tunicamycin-treated MMTV-infectedXCcells;lane2,68,000-Mr polypep-tideimmunoprecipitated withanti-gp52serumfrom translatesof16to26S virionRNA;lane3, 68,000-Mr polypeptide fromtranslatesofpurified24Smammary tumorcellRNA;lane4, 70,000-Mrpolypeptides from purified24S RNAtranslates;lane5, 73,000-Mr poly-peptide immunoprecipitated from mammary tumor cells withgp52antiserum;lane6,77,000-Mrgag poly-peptide from translates ofpurified 35S mammary tumorcellRNA; lane 7, 77,000-Mrgagpolypeptide fromtranslatesof26to40SvirionRNA. Film

expo-sure wasfor22days for aUlanes.

MMTV mRNA's 215

MMTV 24S mRNA, translation productswere

subjectedtopartial protease digestion (Fig. 7).

Current evidence suggests that the primary

productof env isprocessed in cells byremoval of amino-terminal amino acids, presumably a

"signal peptide" (7; D. Robertson and H. E.

Varmus,submittedforpublication),and by

gly-cosylationatmultiple sites (7).For thisreason, the Staphylococcus V8 protease products of both the68,000 and 70,000 Mr products of

intra-cellular 24SRNA(lanes3 and 4)werecompared with the products of a 63,000-Mr env protein synthesized inMMTV-infectedXCcells in the

presence of tunicamycin, an inhibitor of

glyco-sylation (50) (lane 1), with the products of a

68,000Mr env polypeptide synthesized in vitro by 16 to 26S virion RNA (lane2),and with the products of a 73,000-Mr env polypeptide

(gPr73env)

synthesizedininfectedcells(lane 5).

In addition, the products were compared to

digestionproducts of a 77,000-Mrgag polypep-tidesynthesized in vitro fromvirion 35S RNA (lane6) orpurifiedintracellular 35SRNA (lane

7).

Digestion of the68,000Mr polypeptides pro-grammedby 24StmRNAandsubgenomic-length

virion RNA yielded identical patterns which were verysimilartothepatternof theproduct

of the 63,000-Mrpolypeptide from tunicamycin-treated cells; we assumethe difference reflects

the presenceofthe uncleaved "signal"sequence

inthe in vitroproducts. The 70,000-Mrproduct of24S RNA produced a pattem more closely

related to that seen with gPr73env, suggesting that themoreslowly migratingofthe env

pro-teins synthesized in vitro may have been

par-tially glycosylated. Although glycosylation has not generally beenobserved during translation in the rabbitreticulocytesystem (34),this

inter-pretation was supported by the elimination of

the more slowly migrating env species after

digestion with endoglycosidase H (data not

shown). The V8 protease products of thegag

protein appeared unrelatedinsize to the prod-uctsofenv proteins, whereas theV8 digestion

products of virion and intracellular 35S RNA

translateswereindistinguishable. DISCUSSION

We haveusedpreparativemolecular hybridi-zationtocloned viral DNA fixedtoDBMpaper disks to isolate the mRNA's of MMTV and

assesstheir functionsbyin vitrotranslation.The

products oftranslationwere then identifiedby

electrophoretic mobility, antigenic reactivity, and size ofproteasedigestionproducts.Twosize classes ofintracellularMMTV RNA have been

studied with these procedures. The 35S RNA

on November 10, 2019 by guest

http://jvi.asm.org/

[image:9.493.62.229.184.478.2]
(10)

216 DUDLEY AND VARMUS

programs the synthesis of gag (and presumably

pol)polypeptides,whereas 24S RNA directs

syn-thesis ofenv polypeptides, indicating that the

threeknown genes involved in the replication of

MMTV areexpressed in the manner illustrated

forotherretroviruses (15, 20, 22, 25, 27, 29, 31,

53)andpreviously proposed for MMTV (16, 38,

43).

This study isunique in that both species of retroviral RNA (35S and 24S) detected

previ-ouslybyhybridization werepurified and

trans-lated inacell-freesystem.Inthe singleinstance

inwhich retroviral RNAs have been both

puri-fied and translated (3), onlytwo viral

polypep-tidesof76,000and 60,000 Mr were synthesized.

Although it wassuggested that these products were gagandsrcspecific, nodefinite

identifica-tion was made, and no envelope-specific prod-ucts weredetected.

The use of preparative hybridization for

mRNA purification has beenfacilitated by the

availability of abundantamounts of viral DNA

cloned in procaryotic host-vector systems and

by the development ofan efficient method for

fixingDNA tofilterpaper (1, 36). The procedure

adopted here hasanacceptably lowbackground

(Table 1),permitsup to 300-foldenrichment in asingle cycle of annealing (Table1),yields RNA

whichappearsphysicallyintact, evenunder

de-naturing conditions (Fig. 3 and 4), and can be

translated intolarge polypeptidesin vitro (Fig. 5and6).This solid-phasesystem also offers the

advantage of filterreuse (up to 17times in our

hands withoutapparent loss of quality). In

com-bination with other simple isolation methods,

oligo(dT)-cellulose chromatography and rate

zonalsedimentation, thepreparative annealing step appears to permitvirtually complete

puri-fication of the virus-specificspecies studied here, asjudged by gel electrophoresisoflabeled RNA

(Fig.3and4) or by products of in vitro

transla-tion (Fig. 5 and 6). These methods may prove

useful in the analysis ofother species of

virus-relatedRNAs which are present at much lower

concentrations.

By using restriction endonuclease fragments to select viral RNA it is possible to separate various species and to locate their sequences on thephysical map of the genome. In thisstudy, thelargestPstI fragment failed to anneal with MMTV 24SRNA (Fig. 2 to 4) in either

analyt-ical or preparative procedures. Since isolated 24S RNAprograms the synthesis of env proteins

ofnormalsize and complexity (Fig. 5 to 7), most

or all ofenv mustreside onthe 3' side of the PstI site which separates PstI-A (4.0kb) from PstI-B (1.8kb), and few of thesequences(coding ornoncoding) fromPstI-A (4.0 kb) can be pres-ent in 24S RNA (see Fig. 1). We have shown

elsewhere that 24S RNA also containssequences from within 135 bases of the 5' terminus of35S

RNA and from the region adjacent to the poly(A)tractatthe 3' end(38). Theposition of

env, asdeduced here fromhybridization

experi-ments (see Fig. 1), is in reasonable accord with the position as deduced from nucleotide and amino acidsequencing studies. The nucleotide

sequencewhich encodes the amino terminus of

gp52 (the major virion glycoprotein, derived from the amino-terminalportion of theenv pre-cursor [40; G. Schochetman, personal

commu-nication]) begins about 115 basepairs to the 3' side of the PstI siteseparating PstI-A (4.0 kb) from PstI-B (1.8 kb) (J. Majors, unpublished data). The primary product of env has been estimatedtobe50 to90 amino acidslongerthan theglycosylated precursor to gp52(gPr73env) (7; Robertson and Varnus, submitted for publica-tion), and it is assumed that the additional length is encoded at the 5' end of the gene.

Nucleotide sequencing of this region reveals

three AUG codons in a single reading frame,

withoutinterrupting terminationcodons,ca.40,

75,and 180 basepairstothe 5' side of the PstI

site (J. Majors, unpublisheddata). Althoughit

isnotyet knownwhich of these threeservesas

the initiation site for translation, the acceptor

splice point for env mRNA appears to reside about one base pair upstream from the first AUG in thissequence, indicatingthatatotal of about 180nucleotides from the PstI-A (4.0 kb)

fragment are represented in mRNAenv (J. Ma-jors,unpublisheddata).Wepresumethat failure

todetect24S RNAbyannealingwith32P-labeled PstI-A (4.0 kb) (see Fig. 2) reflects the high

background observed with the 4.0-kb probe or thesmallpercentageof the probe (ca.4%)

rep-resentedinthe mRNA. We cannotexplain our

inabilityto select 24S RNA

preparatively

with

the PstI-A (4.0-kb) fragment (Fig. 3), but it is possible that selection ofmRNAwithDNAfixed

toDBM paperrequiresagreater regionof ho-mology than does DNA fixed to nitrocellulose

paper. In the latter case, 50 to 100 base pairs wereclaimedtosufficefor mRNA selection(36). Our data donotdirectlyidentifythe 3'

bound-aryofenv,butextrapolationfromthesize of the

primary gene product suggests that the gene ends about 100to 200basepairstothe 5' sideof thePstIsite which separates thePstI-B and D

fragments. This impliesthatthe long terminal

redundancysequence, which beginsca. 10base

pairs to the 5' side of this PstI site, does not

contributetoenv,althoughthere isanextensive openreadingframe in thelongterminal

redun-dancy (8;Dickson andPeters, personal

commu-nication;J.Majors,unpublisheddata).

Inthisstudy,onlytwosize classes of MMTV

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

(11)

MMTV mRNA's 217

RNA, 35S and 24S RNA, the functional and

structural correlates of the gag (and gag-pol)

and env mRNA's described for other

retrovi-ruses (3, 10, 11, 14, 17, 23, 30-32, 45, 47, 51, 54), have been observed. However, we and others havepreviously reportedavariety of other

spe-ciesnotseenin thepresent studies.These other

species include adiscrete 13S RNA detectable only withMMTVcDNA3 (38);aheterogeneous

groupof13 to15S RNAs(43); aspecies slightly

smaller than genomic length (16); and a 20S speciesannealingtoprobes from boththe3'and

5'termini of viral RNA (RobertsonandVarmus, submitted for publication). Althoughwecannot

explain ourfailuretodetectsomeofthese

spe-cies,wehave recently found that the 13SRNA

is nonviral in origin, hence not capable of

an-nealing to the cloned viral DNAs used here;

previous detection of this species apparently de-pendedupon thepresence of anabundant

spe-cies ofcellularpoly(A)+ RNAinthe virionsused as asourceoftemplate for synthesis ofcDNA3

(Robertson and Varmus, submitted for

publica-tion).The20S RNA has been foundtransiently, generally 4 to 8 h after dexamethasone treat-mentofcultured cells, andwasnot likely to be present in appreciable amounts under the

growth conditions used here. Current findings tentatively indicate that the MMTV genome maydirectsynthesis only of structural proteins, encoded ingag,pol, andenv, ineither infected heterologous cells or mouse mammary tumor

cells. This conclusion has obvious implications for mechanisms ofoncogenesis by thisvirus.

ACKNOWLEDGMENTS

We areindebted to D. Robertson, J. Majors, H. Opper-mann,and B. Baker for reagents and adviceconcerningthese experiments.

J.P.D. was the recipient of a fellowship from the National Institutes ofHealth, and the work was supported by Public Health Service grant CA-19287 from the National Institutes of Health.

LITERATURE CITED

1. Alwine, J. C., D. J.Kemp, and G. R. Stark. 1977. Method for detection ofspecific RNAs in agarose gels by transfertodiazobenzyloxymethyl-paperand hybrid-ization with DNAprobes. Proc. Natl. Acad. Sci. U.S.A. 74:5350-5354.

2. Bailey,J.M., and N. Davidson. 1976.Methylmercury

as areversibledenaturing agent foragarose-gel electro-phoresis. Anal. Biochem. 70:75-85.

3. Bromley,P.A., P.-F.Spahr, andJ.-L. Darlix. 1979. New procedure forisolation of Rous sarcoma virus-specific RNA from infected cells.J. Virol.31:86-93. 4. Cleveland,D.W., S. G. Fischer, M. W.Kirschner,

and U.K.Laemmnli.1977.Peptidemapping by limited proteolysis insodiumdodecyl sulfate and analysis by gel electrophoresis. J. Biol. Chem.252:1102-1106. 5. Dahl,H. H.M.,and C. Dickson.1979.Cell-freesynthesis

of mousemammarytumorviruspr77from virion and

intracellularmRNA.J. Virol. 29:1131-1141.

6. Denhardt,D.T.1966.Amembrane-filtertechnique for thedetectionofcomplementaryDNA. Biochem.

Bio-phys. Res.Commun.23:641-646.

7. Dickson, C., and M. Atterwill. 1980. Structure and processing of the mousemaummarytumor virus glyco-protein precursorPr73'e@.J.Virol.35:349-361. 8. Donehower, L A., A.L. Huang, andG. L. Hager.

1981. Regulatory and coding potential of the mouse mammary tumor viruslong terminal redundancy. J. Virol. 37:226-238.

9. Dudley, J. P., J. S. Butel, S. H. Socher, and J. M. Rosen. 1978. Detection of mouse mammary tumor virus RNA in BALB/c tumor cell lines of nonviral etiologies. J. Virol. 28:743-752.

10.Faller,D. V., J.Rommelaere, and N. Hopkins. 1978. LargeTi oligonucleotidesofMoloney leukemia virus missing in an env generecombinant, HIX, are present onan intracellular 21S Moloney viral RNA species. Proc.Natl. Acad. Sci. U.S.A.75:2964-2968.

11. Fan,H., and D. Baltimore. 1973. RNA metabolism of murineleukemia virus: detection of virus-specific RNA sequences in infectedand uninfectedcellsand identifi-cation ofvirus-specific messenger RNA. J. Mol. Biol. 80:93-117.

12. Fine, D. L., L 0. Arthur, J. K. Plowman, E. A. Hillman,and F.Klein.1974.In vitro systemfor pro-duction of mousemammarytumorvirus.Appl. Micro-biol. 28:1040-1046.

13.Fricke, U. 1975. Tritosol: a new scintillation cocktail based on Triton X-100.Anal. Biochem. 63:555-558. 14.Gielkens, A.L.J., M. H. LSalden, and H.

Bloemen-dal.1974. Virus-specificmessengerRNAonfree and membrane-bound polyribosomes from cells infected withRauscherleukemia virus. Proc. Natl. Acad. Sci. U.S.A.71:1093-1097.

15.Gielkens,A. L J., D. Van Zaane, H. P. J. Bloemers, andH. Bloemendal.1976.SynthesisofRauscher mu-rineleukemiavirus-specificpolypeptides in vitro. Proc. Natl.Acad. Sci.U.S.A. 73:356-360.

16. Groner, B., N. E. Hynes,andH.Diggelmann. 1979. Identificationof mousemammarytumorvirus-specific mRNA. J. Virol. 30:417-420.

17.Hayward,W.S.1977.Size and genetic content of viral RNAs in avianoncovirus-infectedcells.J.Virol. 24:47-63.

18. Hershfield, V., H. W. Boyer, C. Yanofsky, M. A. Lovett, and D. R. Helinski. 1974. Plasmid ColElasa

molecular vehicle forcloning and amplifcation of DNA. Proc. Natl.Acad. Sci. U.S.A. 71:3455-3459.

19.Hilgers,J., and P. Bentvelzen. 1978. Interaction be-tween viral and genetic factors in murinemammary cancer, p.143-189.InG. Klein and S. Weinhouse(ed.), Advances incancerresearch, vol.26.AcademicPress, Inc.,London.

20. Kerr, I.M.,U.Olshevsky,H. F.Lodish,and D. Bal-timore. 1976. Translation of murine leukemia virus RNAincell-free systems from animal cells. J. Virol.18: 627-635.

21. Kessler,S. W.1975.Rapidisolation ofantigensfromcells with a staphylococcal protein A-antibody adsorbent: parameters of theinteraction ofantibody-antigen com-plexes with protein A. J. Immunol.115:1617-1624.

22. Laemmli, U. K. 1970. Cleavage ofstructural proteins during theassemblyof the head ofbacteriophageT4. Nature(London) 227:680-685.

23. Mellon, P., and P. H.Duesberg. 1977. Subgenomic, cellularRous sarcoma virusRNAs contain

oligonucle-otides fromthe3' half andthe 5' terminus ofvirion RNA. Nature(London)270:631-634.

24. Murphy,E.C.,Jr.,D.CamposHI,andR. B.

Arling-haus. 1979. Cell-free synthesis of Rauscher murine leukemia virus "gag" and "env" gene productsfrom separatecellularmRNAspecies.Virology93:293-302.

25. Murphy,E.C., Jr., J.J.Kopchick, K.F.Watson,and R. B.Arlinghaus 1978.Cell-free synthesisofa pre-cursorpolyprotein containing both gag andpolgene VOL. 39,1981

on November 10, 2019 by guest

http://jvi.asm.org/

(12)

218 DUDLEY AND VARMUS

products byRauscher murine leukemiavirus 35SRNA. Cell 13:359-369.

26.Nusse, R., F. A. M. Asselbergs, M. H. L. Salden,R. J. A. M. Michalides, andH.Bloemendal. 1978. Trans-lationofmousemammarytumorvirusRNA:precursor

polypeptidesarephosphorylated during processing.

Vi-rology91:106-115.

27. OppermaDn, H., J. M. Bishop,H.E.Varmus, and L. Levintow. 1977. A joint product of thegenesgagand pol of avian sarcoma virus: a possible precursor of reversetranscriptase. Cell12:993-1005.

28. Oppermann, H.,A. D.Levinson,H. E.Varmus, L. Levintow,and J. M.Bishop. 1979.Uninfected

ver-tebratecellscontainaproteinthatisclosely relatedto

theproduct of the aviansarcomavirustransforming

gene(src). Proc. Natl.Acad. Sci. U.S.A. 76:1804-1808. 29. Papkoff, J., T. Hunter, and K.Beemon. 1980.Invitro translation of virion RNA from Moloney murine sar-comavirus. Virology 101:91-103.

30. Parsons, J.T.,P.Lewis, andP.Dierks.1978. Purifi-cation of virus-specific RNAfrom chicken cellsinfected with avian sarcoma virus: identification of

genome-length andsubgenome-length viral RNAs. J. Virol. 27: 227-238.

31.Pawson,T., R. Harvey, andA.E.Smith. 1977. The sizeof RoussarcomavirusmRNAsactive incell-free translation. Nature(London)268:416-420.

32. Pawson, T., P. Mellon, P. H. Duesberg, and G.S. Martin.1980.envgeneof Roussarcomavirus: identi-fication of thegeneproduct by cell-free translation.J. Virol. 33:993-1003.

33.Pelham,H.R.B.,and R. J. Jackson. 1976. An efficient mRNA-dependent translationsystemfromreticulocyte lysates.Eur.J.Biochem. 67:247-256.

34. Ploegh, H. L., L. E. Cannon, and J. L. Strominger. 1979.Cell-free translation ofthemRNAsfor theheavy andlight chainsofHLA-A andHLA-Bantigens.Proc. Natl. Acad. Sci. U.S.A. 76:2273-2277.

35. Reiser,J.,J.Renart,and G. R. Stark. 1978. Transfer ofsmallDNA fragments frompolyacrylamide gelsto

diazobenzyloxymethyl-paperanddetectionby hybridi-zationwith DNAprobes. Biochem. Biophys. Res.

Com-mun.85:1104-1112.

36. Ricciardi, R. P., J. S. Miller, and B. E. Roberts.1979. Purification and mapping ofspecific mRNAs by hybrid-ization-selection and cell-free translation. Proc. Natl. Acad. Sci. U.S.A. 76:4927-4931.

37. Ringold, G. M.,K. R.Yamamoto,P. R.Shank, and H. E. Varmus. 1977. Mousemammarytumorvirus DNA ininfectedratcells: characterization ofunintegrated forms.Cell10:19-26.

38. Robertson, D.L.,andH.E. Varmus.1979.Structural analysis of the intracellular RNAs of murinemammary

tumorvirus. J. Virol.30:576-589.

39. Sarkar,N.H.,A.A.Pomenti, andA.S. Dion. 1977. Replication ofmouse mammarytumorvirusintissue culture.I. Establishment ofamouse mammarytumor

cell line, virus characterization, and quantitation of

virusproduction. Virology77:12-30.

40.Schochetman, G., S. Oroszlan, L. Arthur, and D. Fine. 1977. Geneorderof themouse mammarytumor

virusglycoproteins. Virology83:72-83.

41. Schochetman,G., and J. Schlom. 1976. Independent polypeptide chain initiation sites for the synthesisof differentclasses of proteinsforanRNAtumorvirus:

mouse mammary tumor virus. Virology 73:431-441. 42. Scolnick, E.M., H. A. Young, and W. P. Parks.1976.

Biochemical andphysiologicalmechanisms in glucocor-ticoid hormone induction of mouse mammary tumor virus. Virology 69:148-156.

43. Sen, G. C., S. W. Smith, S. L. Marcus, and N. H. Sarkar. 1979.Identification of the messenger RNAs coding for the gag andenvgeneproducts of the murine mammary tumor virus. Proc. Natl. Acad. Sci. U.S.A. 76:1736-1740.

44. Shank, P. R., J. C. Cohen, H. E. Varmus, K. R. Yamamoto, and G. M.Ringold. 1978. Mapping of linear and circular forms of mouse mammary tumor virus DNA withrestriction endonucleases: evidence for a largespecific deletion occurring at high frequency duringcircularization. Proc. Natl. Acad. Sci.U.S.A. 75: 2112-2116.

45.Shanmugam,G.,S.Bhaduri,and M. Green. 1974. The virus-specific RNA species in free and membrane-boundpolyribosomes of transformed cellsreplicating murinesarcoma-leukemiaviruses. Biochem. Biophys. Res.Commun.56:697-702.

46. Smith,D.F.,P.F.Searle,and J.G.Williams.1979. Characterization of bacterial clones containing DNA sequences derived from Xenopus laevis vitellogenin mRNA. Nucleic Acids Res. 6:487-506.

47. Stacey, D. W.1979.Messengeractivityof virion RNA for avian leukosis viralenvelope glycoprotein.J. Virol. 29: 949-956.

48.Stark,G. R., and J. G.Williams. 1979. Quantitative analysis ofspecific labeled RNA's using DNA cova-lently linked todiazobenzyloxymethyl-paper. Nucleic Acids Res. 6:195-203.

49.Svoboda, J., P. P.Chyle,D.Simkovic,and I.Hilgert. 1963.Demonstration of the absence of infectious Rous virus in rat tumorXC, whosestructurallyintactcells produce Rous sarcoma when transferred to chicks. Folia Biol.(Prague) 9:77-81.

50. Tkacz, J.S., and J. 0. Lampen.1975. Tunicamycin inhibition ofpolyisoprenyl N-acetylglucosaminyl pyro-phosphate formation in calf-liver microsomes. Biochem. Biophys.Res.Commun. 65:248-257.

51. VanZaane,D., A. L. J.Gielkens,W.G.Hesselink, and H. P. J. Bloemers. 1977. Identification of Rauscher murine leukemiavirus-specific mRNAs for thesynthesisofgag-and env-gene products. Proc. Natl. Acad.Sci. U.S.A. 74:1855-1859.

52. Wahl, G. M., M.Stern,andG. R. Stark. 1979. Efficient transfer oflargeDNAfragments from agarose gels to diazobenzyloxymethyl-paperandrapid hybridization by using dextran sulfate. Proc. Natl. Acad. Sci. U.S.A. 76: 3683-3687.

53. Weiss,S.R., P. B. Hackett, H.Oppermann,A. Ull-rich, L.Levintow,and J. M.Bishop. 1978. Cell-free translation of aviansarcomavirusRNA:suppression of the gag termination codondoes not augment synthesis of thejointgag/polproduct. Cell 15:607-614. 54. Weiss, S. R., H.E.Varmus, and J. M. Bishop. 1977.

Thesize andgenetic composition of virus-specific RNAs inthecytoplasmofcellsproducing avian sarcoma-leu-kosis viruses.Cell12:983-992.

55.Wrninam,J.G.,M.M.Lloyd,and J. M. Devine. 1979. Characterizationandtranscription analysis of a cloned sequence derived from a majordevelopmentally regu-lated mRNA of D.discoideum.Cell17:903-913.

on November 10, 2019 by guest

http://jvi.asm.org/

Figure

FIG.envpositions24SMMTVMMTVDNArepeat.transcribedclonedsitesofsiteandcules(1.3zationunintegratedthethethethebyboxes) the work 1
FIG. 2.phoresisfected35SRNAhdexamethasone.the6:correspondcellularthroughcurycelltected(1.8TheTotal (lanes GR Characterization of intracellular MMTV by hybridization with cloned PstI fragments
FIG. 3.poly(A)poly(A)'paperparedforone-half5%DBMDBMrographyexperiment,afterthedisks;nealedconcentratedannealedide.andincubationwere1.2%cells(panel10i cpm)
FIG. 4.preparativeproducingpHpretreatmentposedgrownMMTV-infectedRNADBMpaperRNA,PstI-BwithinglabeledrRNA(lane1.2%ideThe Selection ofMMTV intracellular RNAs by annealing to cloned PstI fragments
+4

References

Related documents

A simplistic analysis would merely point to what is idiomatic in this music: the quasi- duple (notationally 4/4) metre, 10 syncopated (short-long-short) rhythm in

This suggests that academics responsible for submitting impact case studies to the ‘Social Work and Social Policy’ unit of assessment in REF2014 believed (like several of

Doing social good and contributing to the well-being of society can thus be described as a more complex and higher differentiated goal than generating profit and contributing to

A closer reading of the ontological turn in anthropology and a conversation with the eco-philosophy of Serres has shown us the need to urgently return to the laws of

If the high Reynolds number behavior of the global properties of the mixing zone is the main point of interest then the ILES approach is able to estimate this with significantly

To solve this problem, in this paper, a 3-D global model (as it shown in Fig.1) was developed using finite element method to simulate the whole building behavior of

An examination of mammary tumor and normal, nonmammary tissue DNAs with BamHI supported the idea that the cNIV MuMTV is identical to the C3Hf MuMTV and demonstrated that these

The inspired creative power, the original Promethean gift, original in its continuous power of perhaps ex nihilo contribution to history, which drives the human affair along,