• No results found

Viral DNA in bursal lymphomas induced by avian leukosis viruses.

N/A
N/A
Protected

Academic year: 2019

Share "Viral DNA in bursal lymphomas induced by avian leukosis viruses."

Copied!
9
0
0

Loading.... (view fulltext now)

Full text

(1)

0022-538X/80/04-0178/09$02.00/0 34, No. 1

Viral DNA in Bursal Lymphomas Induced by Avian Leukosis

Viruses

PAULNEIMAN,13* L. N.PAYNE,2AND ROBIN A.

WEISS'

ImperialCancer Research Fund Laboratories, Lincoln's Inn Fields, London WC2A 3PX,England';

Houghton Poultry ResearchStation, Houghton,Huntingdon, Cambridgeshire,England2;and Fred

HutchinsonCancer Research Center, Seattle, Washington98144

Avian leukosis viruses (ALV) induce malignant lymphoma of the bursa of Fabricius. Viral DNA in tumors and normal tissuesfrom infected birds were analyzed by using restriction endonucleases. ViralDNA fragments diagnostic of the exogenous ALV were easily detected in tumors, uninvolved bursal

tissue,

kidney, and erythrocyte nuclei. Exogenous viralDNA was moredifficulttodetect in liver. Usingarestrictionendonuclease (SacI)which cleaves linear unintegrated ALVDNA in asingle site to define integration sitesin DNA from thevarious tissues, we were able todetect ALV DNA only in tumor tissue. Weconcluded thatthe proviral DNA detected in the variousnontumortissuemust beintegrated in multiple sites. The appearance ofALVintegration sitesuniquely in tumors suggests that they are clonal growths. Furthermore, the data suggested the presenceofasingleexogenousintegration site for theALV provirus in each of six

early

neoplastic bursal nodules. This provirus appearedtoretain theorganization

of EcoRI and BamHI recognitionsequencespresentin thegenomeofvirus used to infect the birds. TheALV integration site appeareddifferent in each ofthe tumorsstudied.In awidespread metastaticlymphoma, multipleALVintegration siteswere found as wellas structuralalterations in atleastsome copies ofthe ALVprovirus.

Malignant

lymphomas

of the bursa of

Fabri-cius

(lymphoid leukosis)

are common

neoplasms

in domestic chickens. The induction of these tumors is

closely

associated with infection by avian leukosis viruses

(ALV).

The genome of thisclass of avian retroviruses encodes thegenes known to be

required

forviral

replication,

but no viral gene(s)

mediating

induction ofbursal lymphomas or the short list of otherless

com-mon

neoplastic

diseases (e.g.,

nephroblastomas

and

erythroleukemia)

associated with these

agentshas been identified. The mechanisms

un-derlying

tumor induction

by

ALV remain

ob-scure. The genome of ALV is

closely

related

(about

85%

homology)

to that of

genetically

transmitted

endogenous

retrovirusesof chickens

exemplified

by

Rous-associated virus type 0

(RAV-0) (14).

RAV-0, however,

appearstolack

the

oncogenic

potential

of ALV

(10).

The

major-ity ofsequence

divergence

that does exist be-tweenALV andRAV-0isconcentrated in the 3' end of the genomesof theseagents

(3, 13).

The

significance,

ifany,of this

region

of

divergence

with respect to the

biological

differences be-tween these viruses remains tobe established.

Inthe

field,

ALV is often

spread

by

congenital

infection of

embryos

by

viremic hens

(2).

Exper-imentally, newly

hatched

susceptible

chickscan

be infected by intraperitoneal injection. Viral replication has been observed in many organs afterinfection, butthedevelopment of

lympho-masis

absolutely dependent

uponthepresence

of the bursa of Fabricius (5, 15). The earliest recognizable change in the bursa is the accu-mulation of

lymphoblasts

withinindividual fol-licles (5). We have observed such transformed follicles in themajority of infected birdsasearly as 30 days after hatching and infection (P. E.

Neiman,

L. N. Payne, andR. A.Weiss, in Virus

in

Naturally

Occurring Cancer, in press).Inthe

study

cited,

wefound that thevastmajority of

the 10 to 100 suchtransformed follicles per in-fected bursa didnotappearcapable of progres-sivegrowth.Infact, thesestructuresdisappeared with thephysiological regression of the bursa5 to 6 months after hatching. A much smaller number (zeroto twoperbursa) ofdiscrete but progressively growing neoplastic nodules ap-pearedbeginning 14to 15weeks after infection. These tumorswere composedof thesametype ofpyroninophilic lymphoblastswhich

appeared

in transformed follicles. That the

progressively

growing nodulesarise from the

larger population

oftransformed folliclesseems areasonable spec-ulation. Thesenodulesgrowrather

slowly

fora

periodof weeks.

They

presumably

arethe

pre-178

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

cursorsof therapidly growing metastatic bursal lymphomas which kill most susceptible birds between 20 and 40 weeks after hatching and

infection. Thusatleast three distinct stagescan

be observedin the development of bursal lym-phomas.

In a previous study, hybridization reactions betweengenomicviral RNA and cellular DNA

in solution were used to detect ALV-specific proviral DNA sequences in bursas and bursal tumors from infected chickens(14). On the basis of the kinetics of hybridization, there are an

average of one to two copies of ALV proviral DNA perhaploid genomeinbursal lymphoma cells. In the present study, we have extracted

DNA from early bursal nodules, from wide-spread bursal tumors,and from a series of other

tissues from infected chickens. Viral DNA was analyzed, afterdigestionwith aseries of restric-tion endonucleases, by the Southern transfer technique (17). The dataprovidefurther infor-mation on the distribution of ALV DNA in infected birds, the clonal origin ofbursal nod-ules,and thestructureof the ALVprovirusand adjacent host sequences in bursal lymphomas.

MATERIALS AND METHODS

Viruses and chickens. The virus used for this experiment wasALV 5938, afield isolate which has been characterized in detail intermsof both its onco-genicity and its extensive sequence homology with laboratory strains of ALV suchasRAV and transfor-mation-defective (td) deletion mutants of Rous sar-comavirus (RSV) (11, 14). We used 104interference units of ALV to infect by intraperitoneal injection

newly hatched chicks fromamatingbetween suscep-tible (to both infection andleukosis) inbred line 15I males and outbred brown leghorn hens. Birdswere sacrificed atintervals and examined for the appear-ance ofbursalnodules, bursallymphomas,and other neoplasms such as nephroblastomas (which were found intwoinstances). Thecourseofdevelopmentof neoplastic change in this experiment,asmonitored by histological examination of serial sections of bursae, is describedelsewhere(Neimanetal., in press). Individ-ual neoplastic bursal nodules were dissected free of surroundingnormalbursal tissue. Bursal nodules, lym-phomas,andnephroblastomas and normalbursa, kid-ney,erythrocytes (RBC), and liverwereallcollected forDNAextractions.

DNA extraction. DNA wasprepared from RBC nucleiontheday ofcollection.RBCfromabout 2 ml ofbloodwerewashed in TNE (0.01 M Tris-hydrochlo-ride [pH 7.4]-0.1 MNaCl-0.001 M EDTA) and sus-pendedin 0.01 MTris-hydrochloride (pH 7.4)-0.01 M

NaCl-0.005 M MgCl2-1% Nonidet P-40. After brief blendingin aVortex mixer, nuclei werepelletedand washedtwoto fourtimesinTNE until the hemoglobin was removed. The nuclei were then suspended in about 5 ml of0.01 M Tris-hydrochloride (pH 7.4)-0.001 M EDTA (TE). Proteinase K and sodium do-decyl sulfate (SDS),werethen addedto200ug/ml and

0.5%, respectively, andthe mixturewasincubatedat 37°C for about 2h,oruntil the solutionwasclear. The solutionwasthenextracted withanequalvolume of

chloroform-isoamyl alcohol (24:1), and the aqueous phasewasobtained aftercentrifugationat10,000xg for 10 min. After precipitation with 2 volumes of ethanol, the DNAwasspooledon aglassrod,dissolved inTE, anddigestedfor30min with 20,gof RNase A per ml (Worthington Biochemicals Corp.) at 37°C.

Thiswasfollowedbyanother 2-hdigestionwith pro-teinase Kasbefore,extraction withanequalvolume ofchloroform-phenol (1:1), extraction withanequal

volume ofchloroform-isoamyl alcohol, and dialysis with threechanges of2liters of TE. If necessary, DNA wasconcentratedtoabout1mg/ml by respoolingand dissolution in TE.

DNA from sources other than RBC was usually extracted from tissues frozen and stored at -20°C.

Thiswasdonebyslicingthe tissuewhile still frozen into smallslivers withascalpelanddispersingthe cells inasmall volume of cold saline-EDTA(0.1M NaCl-0.1MEDTA,pH7.4),usingasinglestroke inachilled Douncehomogenizer. This suspensionwasaddedto saline-EDTAcontaining 1%SDS preheatedto 60°C

(totalvolume,5ml/goforiginal tissue)andincubated for 10 min. The mixture was then cooled to room temperature, made 1 M in sodium perchlorate, and extracted withchloroform-isoamyl alcoholasfor the RBC nuclear DNApreparations. The remainder of theprocedurewasalso thesame asthat used for the RBC DNA.

Preparation ofunintegratedALV 5938 DNA. Twentysubconfluentplatesofchickenembryo fibro-blasts were infected with ALV at a multiplicity of infection of about 1.At40hafterinfection,thecells layers wererinsed with coldTNE,scraped into cen-trifuge tubes with a rubber policeman, and washed twicemorewith cold TNE.Cellswerelysed with SDS and fractioned by precipitation in the cold in the presence of 1 M NaCl according to the procedure describedby Hirt (6). Thesupernatantfraction (Hirt

supernatant) containingunintegrated viral DNAwas digested with RNase Aand proteinase K, extracted withchloroform-phenol,andprecipitated with ethanol aspreviouslydescribed (13). Whenanalyzedon0.7% agarosegelsbySoutherntransfer and hybridization to

32P-labeledviralcomplementaryDNA(cDNA) as de-scribedlater, thesepreparations contained predomi-nantly a 5.1 x 106-daltonlinear viral DNA molecule and a small amount (roughtly0.1 of the amount of linear form) ofasupercoiled viralDNAmolecule (data notshown).Whenportions representing 1 x 106to 2 x10'fibroblasts in the original infected cultures were usedinsubsequent experiments (e.g., Fig. 1A), nearly allofthedetected viralDNAwascontributed by the linear molecule.

Restriction endonuclease digestion and

elec-trophoresis ofviral DNA. Restriction endonucle-asesEcoRI, SacI, BamHI, andHpaI were obtained fromNewEngland Biolabs. Digestions, carried out in 50

pl

of therecommended buffersolution, contained either10MigofcellularDNApreparation or a portion of theHirtsupernatantDNApreparation representing about2 x 106 infectedfibroblasts along with 1 to 5 U oftheappropriate restriction endonuclease. One mi-VOL.

on November 10, 2019 by guest

http://jvi.asm.org/

(3)

180 NEIMAN, PAYNE, AND WEISS

crogramof lambda phage DNA was dissolved in

5-p

(0.1 volume) portions of the complete reaction mix-ture, and both mixtures wereincubated for 2 to 4 h at 37°C. The small reaction containing the lambda DNA was analyzed by agarose gel electrophoresis (as de-scribedbelow) to confirm that the reaction had gone to completion. Only limit digests were analyzed fur-ther. The main reaction mixture was reduced in vol-umebyevaporation inavacuum,redissolved in 50

pd

of 0.01 M Tris-hydrochloride (pH 7.4)-0.001 M EDTA-0.1%SDS-50% glycerol containing bromophe-nolblue, andloaded on a 0.4-cm-thick 1.1% horizontal agarose(Seakem)gel.Electrophoresis was at 15 V/cm forabout24h.

Transfer to nitrocellulose membranes and hy-bridization. DNAwas transferred to nitrocellulose

membranes (0.1

jam;

Sartorius) by the blotting tech-nique described by Southern (17). Membranes were dried ina vacuum oven at80°C for2 hand prehy-bridized for16hat420Cin5xSSC(SSC=0.15NaCl plus 0.015sodium citrate), 50% formamide containing Denhardt solution (0.02% albumin, Ficoll, and polyvi-nylpyrrolidone), 20,ugofsingle-stranded calf thymus DNA perml, and 50,ug of yeast RNA perml. The membrane wasthen transferred to a second plastic envelope containing the same buffer plus 106 cpm of a

32P-labeledviral cDNAprobe. The latter was prepared by in vitro reversetranscriptase reactions using

puri-fied ALV 3958 35Sgenomic RNAastemplate prepared as described previously (12) and limit DNase I-di-gested calf thymus DNA oligomers as primer (19). Reactionconditions and isolation of cDNA were

es-sentiallyasdescribedpreviously(1978).Hybridization

was carried out for 48 to 72 h. The membrane was thenwashed in200ml of2xSCCin Denhardtsolution for20minat roomtemperaturefollowed by200ml of

0.lx SSC-0.1% SDS at 500C with shaking for 1 h. Final washing was with four to six rapid changes of 0.lx SSC followed by air drying. Viral DNA bands werethenvisualized by radioautography in the pres-enceof calcium tungstateintensifyingscreens.

RESULTS

Digestion of viral and cellular DNA with EcoRI. Shanketal. (16) have

recently

reported that EcoRI cleaves the linear forms of both

parental

and td RSV

unintegrated

DNA very

closetoboth the3'and5'endsof themolecule owingtothe presence ofarecognitionsitewithin a300-nucleotide

repeated

sequenceateach

ter-minus. Two additional internal sitesare recog-nized, givingacharacteristic three-bandpattern of internal fragments released by digestion of RSV andtdRSVDNAwith EcoRI. Given the extensive sequence

homology

between ALV 5938and td RSV (11), itwas not

surprising

that

digestion

ofthe linearform of

unintegrated

ALV

DNAyieldedthree

fragments:

2.5 x

106,

1.65 x

106,

and 0.9 x 106 daltons (Fig. 1A, lane HS). These sizes are

reasonably

close to

predicted

values from the EcoRI mapoftdRSV(16).We assumed,therefore, that these three fragments

represented three internal fragments of ALV, theterminal fragments beingtoosmallto detect bythis method.

Figure 1alsoshows the results of digestion of a seriesof tissues from tumor-bearingbirds with EcoRI. In Fig. 1A, digests of DNA from two individualbursal nodules

(T6

and

Tb),

from non-tumorousportions of the bursa, and fromRBC and theliver ofonebird(2989) are shown. Viral DNAfragmentsof2.5 x

106,

1.65x

106,

and 0.9 x 106daltons were easily detected inall tissues excepttheliver, where only the largestfragment waseasily observed. This fragmentis commonly observed in EcoRI digests of DNA from unin-fected chicken cells (e.g.,Fig. 1,bird2407), and it, along with the other higher-molecular-weight viral DNA-containing fragments, were consid-ered to representendogenous ALV-related se-quences. We concluded that the liverwas not extensively infected with ALV, whereas all of the other tissues contained easily detectable quantities of

exogenously

introduced ALV DNA. Furthermore, comparison of the tissue-derived viral DNAbands with those of linearDNAfrom ALV 5938 (lane HS) showed thatmostof pro-viral DNAin infected tissues, including the tu-mors, had thesameEcoRImap as theoriginal infecting virus.

Figure 1Bdemonstrates theresults of EcoRI digestion of DNA from the tissues of threemore birds.RBC and bursal DNA fromanuninfected control bird (2407) lacked exogenous ALV EcoRI 1.65 x

106-

and 0.9 x 106-dalton frag-ments, whereas DNA from

RBC,

bursa, and bursal nodule (Tf), but not from the liver, of infected bird2999contained the standard exog-enous ALVEcoRI fragments. This figure also shows theresults ofdigestion ofDNAfrom RBC andan advancedwidespread bursal

lymphoma

whichappearedinchicken2902.Inthiscase,in addition to the standard viral EcoRI

bands,

a

newviralDNA

containing fragment

of2.3x

106

daltons is

clearly

seen in the EcoRI

digest

of tumor DNA, but not in the RBC DNA. This observation with the

metastasizing

tumors was

notrepeatedin

analysis

of DNA from six bursal nodules from this

experiment

where unusual EcoRI fragments didnotappear. For

example,

anEcoRI

digest

ofbursalnodules T'andT and other tissuesfrom bird2993

yielded

the standard ALV-specific intemal fragments

(Fig.

10).

In this case, we

probed

the

organization

of ALV DNAin the tumors and other tissues further

by

digestionwithasecond restriction

endonuclease,

BamHI. In all

tissues,

two small viral DNA fragmentsof1.1 x 106and1.5 x 106daltonswere

detected which corresponded to the

expected

internal fragments released from ALV linear

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

(4)

81A-"

2-407

RC Bu

6> - 4*

Us.R--,.

6...4w

4.'._

2902 2999 RC T9 RC L

M ;:...%..

c2X3-`2 Ecor .rr,

-B:B.u

FIG. 1. Analysis of viral DNA with EcoRI and BamHI. DNApreparations from various tissuesof individual ALV-infectedbirdsweredigestedwiththe restrictionendonuclease, subjectedtoelectrophoresis on agarose gels, and transferred to nitrocellulose membranes. ViralDNA-containing fragments were

located by hybridizationto 32P-labeledALVcDNA probesandradioautography.(A)Outerlanesare32p

labeledsize markers (x 106daltons) of adenovirus

type2DNAdigestedwith eitherXbaI(left)orEcoRI

(right).HS isanEcoRI digest of AL V linear

uninte-grated DNA from a Hirt supernatant fraction of infected fibroblasts.Inbird 2989, DNAwasanalyzed

from twodistinct neoplastic tumornodules (Taand

Tb),normaluninvolved bursa (Bu),RBC nuclei (RC), and liver (L). (B) Sizemarkers used wereHindIII fragments ofphage DNA(notshown). Bird2407was

an uninfectedcontrol. Bird 2902was infected and

hadametastasizing malignant lymphoma(Tw). Bird

2999 hadasinglebursalnodule(Tf). Theremainder

DNA by BamHI digestion (data not shown). Therefore, the three BamHI sites and thefour EcoRIsites presentin the genome of the ALV usedfor infectionappear tobepreserved in the ALV DNA detected in the various infected tis-suesandtumorsstudied in thisexperiment.

Digestionof viral and cellular DNA with Sacl. We observed that Sacl(and its isoschizo-mer SstI) cleaved linear ALV DNA at asingle site,yieldingtwofragments of4.8 x 106 and0.25 x 106daltons.Therefore, Sacldigestion of inte-gratedproviral DNA which is coextensive with cytoplasmic linearviral molecules should yield only fragments

covalently

linked to host cell sequences. Figure 2 (panels B,

C,

and D, lane HS) shows the

larger SacI-generated fragment

of unintegrated linear ALV fragment (the smalleronemigratedbeyond the bottom of the illustration and required verylong exposure of the radioautogram in order to visualize). The position of this bandcanbecompared with viral

DNA-containing fragments

observedinSacI

di-gestsofDNA fromthesametissuesdepictedin Fig. 1except 2989RC (Fig. 2B), where theDNA was not digested by Sacl. In contrast to the results obtained with EcoRI

digests,

no bands were observed in either bursa or RBC DNA whichmigrated atthe sameposition asthat of the SacIfragment of unintegrated ALV DNA. In fact,

only

endogenous viral DNA

fragments

weredetected in normal tissues. Weinterpreted this observation toindicatethattheviralDNA molecules detected in EcoRI

digests

of DNA from bursa, RBC, andkidney (not shown) are probably integrated, butinsuchalarge number ofsites thattheycannotbevisualized.Inbursal nodules T', Tb(Fig. 2B),

Tc, Tr

(Fig. 2C), and

Tf

(Fig. 2A, bird 2999), SacI-generated fragments containing viralDNA weredetected whichwere absent in thedigests of both linear unintegrated DNA and DNA from infected but untrans-formed tissues. A pair of such fragments was detected in four of thesenodulesasindicatedby the data in thevariousfigures; onlyonedistinct tumor-specific fragment was identified in

Tf.

These fragmentsmay represent both halves of a single SacI-digested ALV provirus linked to flanking host cellsequences, orthey may repre-sent distinct integration sites of ALV. It also seemsreasonable to assume that we were able tovisualize theseintegration sitesin tumors, as

of the lane designations are as for (A). The dot denotesaunique tumor-specificfragment. (C) Anal-ysis withboth EcoRIandBamHIofDNAfrom tissues

from bird2993. There were twodistinct bursal nod-ulesin this bird, TC andTd. The markers and lane designations are as before. The dots denote tumor-specificfragments.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.508.55.247.81.474.2]
(5)

A;

z07

2902

2999

- C

Bu

RC

Tg

RC

L

Bu

T

42

.-

...

26-

1.4--2993

HS

TC

Td

Bu

RC

i3.5-_

12.5

5.9--

2.8-

2.36-

i.8-'--1.4-

-

1.2--

0.92-KD)

13.5

125

5.9

-_

L

2989

HS

To

Tb

Bu

2892

HS

Tbb

K

RC

L

Bu

2.8---

2.36---1..8

--1.2-...

0.92-FIG. 2. Analysis ofviral DNAwithSac. DNAfrom the tissuesof six different birdsweredigested with

SacIandanalyzedasdescribedinFig.1.(A)DNAfrom tissues from birds2407, 2902,and2999with markers and lanedesignationscorrespondingtoFig. lB. The dots again denotetumor-specific fragments. (B) DNA

fromtissuesofbird 2989correspondingtoEcoRIdigests depicted inFig. IA. RBC DNA failedtodigestwith

Sac.(C)Sameanalysis forDNAfrombird 2993asinFig. I C. (D) Analysis ofDNAfrom tissuesofabird2892

witha large nephroblastoma

(?*b).

Lane K isfrom uninvolvedkidney. The remainderof the designations correspondtoourfigures.

opposed to infected but untransformed bursal tissues,because theintegrationsite of ALV

pro-virus isidenticalin eachcellinthe tumor. The

simplest explanationfor thisfindingis that these nodulesarecomposedof theprogenyofasingle cell (that is,areclonalgrowths).

Figure 2A also shows SacI digests of DNA from tissues ofanuninfected controlbird, 2407,

and of DNA from RBC and tumor tissue from the bird(2902)withwidspreadmetastaticbursal

lymphoma.Inthe lattercase,therewere alarge

number(aboutsix oftumor-specificALV

DNA-containing fragments. Similarly, in bird 2892,

which developed alarge nephroblastoma, four

viralDNA-containing fragmentswereobserved

(Tbb

,Fig. 2D) whichwere clearly absent inthe

uninvolved kidneyand othertissues. The stan-dardEcoRIviral DNAfragmentswereobserved

inallthese tissues(not shown).Therefore other

virus-induced neoplasms,suchas

nephroblasto-mas,areprobablyclonalgrowths. Furthermore,

thepossibilityis raised that inaggressive metas-tatic lymphomas like Tg, there is an amplified

number of integrated ALV proviruses. Since these tumors alsoyieldeda newEcoRIfragment (Fig. 1B), atleast one of these provirusesmay

RC

L

a..j,/I

i3.5-

52.5-

5.9---

2.8---2.36

i.8--1.2

0.92---J. VIROL.

Al- .4

ow

A*W

,:.il:.

4m 40

40.

AbAr

.-.-.-..- f

& 40 40M 40 :,a

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.508.64.457.77.453.2]
(6)

bestructurallyaltered.

Finally, fragments of ALV containing DNA wereobserved of about 3.4 x 106daltonswhich were variably present in SacI digests of DNA from different tissues. For example, in Fig. 2, thisspecificfragmentappearsinbursalnodules and normal bursa, but not in other tissues in digests of DNA from birds2989, 2993, and2892 (Fig. 2B to D). In contrast, a fragment of this

sizewasdetected in alltissues of bird 2999 and

notissues from bird2407(Fig. 2A),aswellasin

other infected birds (not shown). The signifi-canceof thisfindingis,atpresent,unclear.

The data from SacI digests of DNA from

tissue of six different birds with bursal nodules

ortumors are summarized in Table 1. All viral DNA-containing fragmentsobservedwere clas-sified into two broad groups. Those fragments

appearing in allnormal andneoplastic tissuesof

anygiven birdwere assignedtothe categoryof endogenous viral information. Some of these fragments were common to all of these birds studied and presumably came from the inbred whiteleghornparent.Forexample,bands of13.0

x 106, 6.0 x 106, and 3.9x106daltons have been

characterized in line 15I (1). Other fragments werevariablypresentindifferent birds andwere probably segregating within the brown leghorn flock. The 3.4 x 106-dalton variable fragment wasassigned tothis endogenous category. The remainder of the observedSacI-generated viral DNA-containingfragmentsweretumorspecific. The table lists the molecular weights of these fragments for six distinct bursal nodules, one metastasizing bursal lymphoma, and two ne-phroblastomas. If these fragments define the

TABLE 1. Analysisof viral DNA withSacE

Endogenous' TumorspecificC

Birdno.

AR,t

Bursalnodulesd Bursal

Nephro-no. Bursal lymphoma' blastoma

All tissues Bursa__

a b c d e f g aa bb

2989 13.0 8.5 Same 6.0 7.3

6.0 plus 2.1 4.6

3.9 3.4

2993 13.0 8.5 Same 7.3 5.1

6.0 2.1 plus 4.8 4.0

3.9 1.4 3.4

3000 13.0 8.5 Same 4.55 5.0

6.0 2.1 4.50 4.5

3.9 1.4

2999 13.0 1.1 Same 5.5

6.0 8.5 3.9 2.1 3.4

2902 13.0 8.5 NAf Multiple

6.0 2.1 5.5-4.8

3.9 1.45

2892 13.0 2.1 Same 7.0

6.0 1.4 plus 5.4

3.9 3.4 4.5

3.4

aData

indicate the molecular weight (x 10-6) of SacI-generated viral DNA-containing fragments. All fragments couldbeclassedin twocategories, endogenous or tumor specific. Most of the data were derived from theexperiments depictedinFig.2.

'Definedasbandsobserved inDNA from all tissues (tumors, bursa, liver, kidney, RBC nuclei) in a particular bird.

CDetectedonlyin DNA from tumor tissue.

d Individualsmallneoplastic bursal nodules dissected free ofsurroundingnormal bursaltissue. Designations correspondtoFig.2.

'Main massofawidespreadbursal lymphoma.

fNA, Notapplicable.

on November 10, 2019 by guest

http://jvi.asm.org/

(7)

integration site of ALV in the neoplastic cells, then their different sizes suggestdifferent inte-gration sites in each tumor. This observation appearstohold both for fragments comparedon

thesame gel (see Fig. 2) and for distincttumor

nodules arising in thesamebursa (e.g., nodules T5 and Tb in bird 2989 as well asTr and Td in bird2993).

Analysis with other restriction endonu-cleases. One of thepotential conclusions arising from the preceding analysis withSacI is that the pair of tumor-specific fragments observed rep-resented a single exogenously introduced ALV provirus in each bursal nodule examined. This conclusionmight be challenged onthe grounds that thesmall(0.25x

10'

daltons)SacI fragment of linear ALV DNAmightnotbereadily detect-ableevenwhencovalently attachedtoadjacent host sequences, using our probe. If that were

true, then all of the SacI tumor-specific

frag-mentsmight contain only the larger (4.8 x 106 daltons) viral segment, indicating at least two

proviral copies per cell in each of the tumor

nodules. Table 1andFig.2show several exam-ples of tumor-specific fragments which are smaller than 4.8 x 106 daltons and therefore

cannotbecomposedofaviral DNA segment of that size plus additional host cell sequences.

Either these fragments must represent the smaller viral segment linked to flanking host

sequences, orelsesomealtered arrangementof ALV DNA sequences must be present. An al-tered arrangement was not detected in the EcoRIdigests.To resolve thistissue,DNA from

the tissues of bird 2993 was digested with two

additionalenzymes:BamHI, which should yield both junction fragments and internals, and

HpaI, which cuts linear ALV DNA once near

the middle of the molecule (data not shown).

The results with BamHI (Fig. 1C) and HpaI (not shown) are summarized in Table 2. With bothenzymes, a distinct pair of tumor-specific fragmentswasdetected in each of thetwobursal

nodules (TC and T) present in the bird. This result is also most consistent with the notion thata single provirus ispresentin each of the cellsmakingupthe bursaltumornodules.

It was also of interest to determine whether DNA from the metastatic tumor (T9) which appeared to contain several ALV proviruses when analyzed with Sacl contained multiple tumor-specific bands when analyzed with other enzymes. Digestion with HpaI also generated multiple tumor-specific viral DNA-containing fragments (Table 2). We therefore conclude that this metastatic neoplasm is also probably a clonal growth carrying about three exogenous ALVproviruses integrated in distinct sites in the host cell genome. The simple explanation that

the multiple copiesare the result of the fusion ofseveral bursal nodules in the formation of the

tumormight be challenged on the groupsthat theSacI-generated tumor-specific fragmentsin

Tgshown inFig. 2A (aswellasthoseproduced by HpaI listedinTable2)appeartoberoughly of the same intensity as the endogenous viral bands in the same DNA. We would have ex-pected that amixture ofcellsoriginating from different tumornodules would have resultedin

dilution of the tumor-specific bands in

compar-TABLE 2. Analysis with other restrictionendonucleasesa

Tumorspecific Exogenous

Endonuclease Endogenous

intemnals

2993DNA

HpaI 13.0 5.5 3.9 3.3

11.0 4.2 2.8 2.5

8.0 2.4

BamHI 5.1 2.4 12.5 15.5 1.5

3.9 1.4

3.4 1.0 6.4 6.6 1.1

2.9

2902DNA

HpaI 13.0 4.8 14.0

11.0 3.6 7.5

6.5 2.0 5.5

4.0

3.3 2.4

aMolecular weight (x

10-6)

of viral

DNA-containing

fragments

from tissuesinbirds 2993 and 2902after

digestionwithHpaI and/orBamHI.Designations correspondtothose inFig.2 and Table 1.

'Correspond

tothefragmentsreleased fromlinearALV DNA.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.508.64.459.450.634.2]
(8)

ison with the endogenous band present in all cells in thetumor.

DISCUSSION

These observations have led ustoa number of conclusions. First, the provirus of ALV can

easily be detected in the target tissues (RBC, kidney, bursa) of infected, susceptiblebirds.

In-fection of the liver appearsto be less efficient

than that of othertissues. Second,theprovirus

appearstobeintegratedindistinct sites in

virus-induced tumors

arising

in these tissues. The ALV integration sites

appeared

different for each of nine

neoplasms,

six bursal

nodules,

one

metastatic bursal

lymphoma,

and two

nephro-blastomasweexamined.These

findings

are sim-ilar to the results of studies carried out with other retrovirus-induced

neoplasms

such as

mammaryadenocarcinomas inducedby murine mammary tumorvirus (4) and

thymic

lympho-masinducedbymurineleukemia virus(18).We share with these authors the conclusion that such observations suggest a clonal origin for these relatively

long-latency-type

tumors in-ducedby virusesapparently

lacking

(at leaston the basis ofpresent

knowledge)

unique trans-forminggenes.

Aided by the fact that the

endogenous

viral

DNA-containing

fragments generated by SacI

(orSstI) from whiteleghorn chickens have been extensively catalogued (1), we were able to clearly separate all of the SacI viral bands ob-servedintoeitherendogenous ortumor-specific categories (Table1).This,inturn,allowedus to count theapparentintegration sitesfor ALV in eachtumor.Thereappeared tobeonlyasingle suchsiteperlocusineachofthe bursalnodules. Furthermore, onthe basis of the analysis with EcoRI andBamHI,the ALVprovirus occupying this site appears to have the same restriction map as the genome ofbulk of thevirusparticles inthe stock used toinfect the chickens. Inter-estingly, the metastasizing bursal lymphoma studied in this experiment appeared to have several proviral copies integrated in different sites in the host genome. Futhermore, EcoRI digestionsuggested a mutation in the EcoRI site present inproviral DNA in this tumor. Such a mutationmight involve either the addition of an EcoRIsite withintheproviral DNA segment or possiblytheloss of arecognition site. The pos-sibility is thus raised that

this

amplification of viral DNAcopies, and perhaps other changes in provirusstructure, isassociatedwiththe acqui-sition ofmetastatic capability by the neoplastic bursalnodules.

We mentioned in the introduction that the mechanismsunderlyinginduction of bursal lym-phomas (or eithertumor) by ALV remain

ob-scure. The conclusions reached in this study may be examined withrespect to theirimpact on at leastsome classes of mechanisms which might be involved in viral oncogenesis in this system.

Forexample,theidea that ALVmight induce neoplasticchange byintegratingat aspecific site intarget cells iscompromised

by

thelackof any evidence ofasimilarity in theintegration site of ALV in the various tumors examined in this study (as well as the evidence cited in other retrovirus systems).

Furthermore,

the presence ofa single exogenously introduced provirus for ALV in each of the bursal nodules with an apparently orthodox restrictionmapfor EcoRI andBamHI wouldseem toexcludethe possibil-ityof alarge host cellsequenceinsert within the provirus of the kind identified with sarcoma viruses anddefective acuteleukemiaviruses,at leastatthestageof formation of bursal nodules. Thesituationinmetastasizinglymphomasmay be different. In this study, multiple ALV ge-nomes were detected as well as alterations in-volving the EcoRIrecognition site. Although the consistency ofthis phenomenon remains tobe

established,

it is of interest that similar

obser-vations of viral gene amplification have been made in Moloney leukemia virus-induced thymic lymphomas in inbred BALB/Mo mice (7-9).If,indeed, metastasizing lymphomas con-taining several viral genomes arise from less malignant clonaltumors with single exogenous proviruses, then itseemsreasonabletopropose that tumorprogression inthis system involves one or more rounds of clonal proliferation of altered cells beyond the initial clonal expan-sion(s) responsible for the formation of bursal nodules. The additional proviral copies may arisesimply by superinfection proceding clonal overgrowth.It isunclearatpresent whether this apparent viral gene amplification is a cause of tumorprogression orsimplyamarkerofclonal evolution. Continued examination ofthe struc-ture of the proviral DNA in these tumors at variousstages inthe developmentofbursal lym-phomas, coupled with assessment of the viral RNA and protein gene productsin these cells, shouldcontinue to narrowdownthelistof pos-sibilities and provide further insight into the mechanisms ofoncogenesis by these agents.

ACKNOWLEDGMENTS

Wethank L. Jordan and D.Macdonnellforexcellent tech-nical assistanceandM.Rossfor typingthe manuscript.

P.E.N.is ascholarof theLeukemia SocietyofAmericaand was supported byafacultyscholarship from the Josiah P. MacyJr.Foundation. Thiswork was alsosupported byPublic Health Servicegrant CA 12895 from the National Cancer Institute.

LITERATURE CITED

1. Astrin, S.M. 1978.Endogenous viralgenes ofthewhite

on November 10, 2019 by guest

http://jvi.asm.org/

(9)

leghorn chicken: common site of residence and sites associated withspecific phenotypes of viral gene expres-sion.Proc. Natl. Acad. Sci. U.S.A.75:5941-5946. 2. Burmester, B. R., and N. F. Waters. 1955. The role of

the infectedegg in thetransmission of visceral lympho-matosis.Poultry Sci. 34:1415-1429.

3.Coffin, J. M., M. Champion, and F. Chabot. 1978. Nucleotide sequence relationships between the ge-nomes of an endogenous and an exogenous avian tumor virus. J.Virol. 28:972-991.

4. Cohen, J.C., P. R. Shank, V. L Morris, R.Cardiff, and H. E. Varmus. 1979.Integrationof the DNA of mousemammary tumor virus in virus infectednormal andneoplastic tissue of themouse.Cell 16:333-341. 5. Cooper, M.D., L. N.Payne,P. B.Dent,B. R.

Burmes-ter, and R. A. Good. 1978. Pathogenesis of avian lymphoid leukosis. I.Histogenesis. J. Natl. Cancer Inst. 41:373-389.

6. Hirt,B. 1967. Selective extraction ofpolyomeDNAfrom infectedmousecultures. J.Mol. Biol. 26:365-369. 7.Jaeni8h,R. 1976.Germ lineintegrationandMendelian

transmission of exogenous Moloney leukemia virus. Proc. Natl. Acad.Sci.U.S.A. 73:1260-1264.

8. Jaenish,R. 1979.Moloneyleukemia virus gene expres-sion and gene amplffication inpreleukemic and leu-kemicBalb/Momice.Virology93:80-90.

9.Jahner,D.,H.Stuhlman,andR.Jaenish. 1980. Con-formation of free and ofintegrated Moloneyleukemia virus proviral DNA in preleukemic and leukemic BALB/Mo mice.Virology, in press.

10. Motta,J.V., L.B.Crittenden,H.G.Purchase,H. A. Stone, and P. L. Witter.1975.Lowoncogenic poten-tial ofendogenousRNAtumorvirus infectionor expres-sion. J. Natl. Cancer Inst.55:685-689.

11.Neiman, P. E.1978.Mappingby competitive

hybridiza-tion of sequenceswhich differ between endogenous and exogenous chicken leukosis viruses. Virology85:9-16. 12.Neiman, P. E., S. Das, D. Macdonnell, and C. Mc-Millin-Helsel. 1977. Organization of shared and un-shared sequences in the genomes of chicken endogenous and sarcoma viruses.Cell 11:321-329.

13. Neiman,P.E.,C.McMillin-Helsel,andG. M.Cooper.

1978.Specific restriction of avian sarcomaviruses by a line oftransformedlymphoid cells.Virology 89:360-371.

14. Neiman,P.E.,H.G.Purchase,and W.Okazaki.1975. Chicken leukosis virusgenomesequencesinDNA from normal chick cells and virus-induced bursal lymphomas. Cell 4:311-319.

15. Peterson, R. D. A., B.R. Burmester, T. M. Frederick-son, H. G.Purchase, and R. A. Good. 1964. Effect of bursectomy and thymectomy on the development of viscerallymphomatosis in the chicken. J. Natl. Cancer Inst. 32:1343-1354.

16. Shank,P.R., S. H.Hughes,H. J.King,J. E.Majors, N.Quintrell, R. V.Guntaka, J. M. Bishop, and H. E.Varmus. 1978.Mapping unintegrated avian sarcoma virus DNA:terminioflinear DNA bear 300 nucleotides present once ortwice in two species of circular DNA. Cell 15:1383-1395.

17. Southern, E. M. 1975. Detection ofspecific sequences among DNAfragments separated by gel electrophore-sis. J.Mol. Biol. 98:503-517.

18. Steffen, D., and R. A.Weinberg. 1978. The integrated genome ofmurineleukemia avirus. Cell 15:1003-1012. 19. Taylor, J. M., R. Illmenee, and J. Summer. 1976. Efficient transcription of RNA into DNA by avian sarcomaviruspolymerase. Biochim. Biophys. Acta 442: 324-330.

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

Figure

Fig. 1resultswas except 2989 RC (Fig. 2B), where the DNA not digested by Sacl. In contrast to the obtained with EcoRI digests, no bands
FIG. 2.fromSacISac.andwithcorrespond Analysis of viral DNA with Sac. DNA from the tissues of six different birds were digested with and analyzed as described in Fig
TABLE 2. Analysis with other restriction endonucleasesa specific

References

Related documents

The growth rate and volume fraction of the major age hardening phase (i.e. γʺ), is strongly dependant on the cooling rate from the solution treatment temperature

A GSFR model has been developed to demonstrate that reduced system inertia will lead to a faster frequency drop and lower frequency nadir for the same amount of

Food systems could combine the goals related to diet-related health, environment and economy if they involved farmers, entrepreneurs, small- and medium-sized enterprises and

The convergence of these Krylov subspace methods depends on the spectrum of the coefficient matrix which for the Poisson and Stokes problems, as well as other elliptic

The effect of minimum tip clearance gap, which is an input to the grid generation methodology and whose dependency with the angular position has been so far neglected, was

4)29 and 4)15 genomes by bacterial restriction endonucle- ases, EcoRI and HpaI. Unusual base sequence arrangement in phage 4)29 DNA. EcoRI analysis of bacteriophage P22 DNA

(A and B) Alternative physical maps of restriction endonuclease PstI sites (0) in RIII MuMTV proviral DNA isolated from FeI/C6 cells and CCL64 mink lung cells infected with

14 Young people reported a wide range of services were available in custody such as youth work,. groupwork, addictions support, employability services, social work, and support