Characterization of endogenous viral Loci in five lines of white leghorn chickens

Characterization of endogenous viral Loci in five lines of white leghorn chickens

VIROLOGY 135.125-138 (1934) Characterization of Endogenous Viral Loci in Five Lines of White Leghorn Chickens ERIC H. HUMPHRIES,’ MARTHA L. DANHOF...

2MB Sizes 0 Downloads 21 Views

VIROLOGY

135.125-138

(1934)

Characterization of Endogenous Viral Loci in Five Lines of White Leghorn Chickens ERIC H. HUMPHRIES,’

MARTHA L. DANHOF,

AND

IV0 HLOZANEK*

Department of Microbiobgg, University of Texas Health Science Center at Da&s, 5X23Harry Hines Blvd, Da&w, Texas 75225,and *Czechx&avak Academy of L%ences, Institutes of MdeMclar Genetics, Flemingovo nam 2, Prague 6, Czechoslovakia Received December 19, 1983; accepted February 27, 198.4 Five lines of chickens have been examined for the presence of DNA sequences related to the endogenous avian retrovirus. Five new loci have been identified, based upon analysis with the restriction endonucleases Sac1 and BamHI. One locus has been associated with the production of infectious endogenous virus. Restriction endonuclease mapping suggested a limited similarity between the flanking cellular sequences of two of these loci, ~-1’7 and m-18, and several endogenous loci, including eu-1, already characterized. The data suggested that these two loci might have been generated by chromosomal duplication. Hybridization analysis with a probe containing the cellular sequences that flank or-l, however, revealed that these flanking sequences shared no detectable homology with the cellular sequences that surround eo-17, e-u-18, or nine other endogenous loci that were examined. These results are consistent with the hypothesis that several of the endogenous viral loci resulted either from independent infections of the germ line or from virus transpositions.

and Payne, 1971; Hanafusa et cd, 1974). The presence of the endogenous loci can All naturally breeding chickens that also be detected by hybridization of viralhave been analyzed contain germ line genes specific nucleic acid probes to cellular DNA that are related to the retroviruses of the isolated from uninfected chicken cells. avian leukosis-sarcoma complex (Astrin, Early studies estimated that several copies 1978; Robinson, 1978). These genes are of viral genes were present in the genomes present in all the cells of an organism and of a variety of chicken fibroblasts (Rosenare inherited in a stable Mendelian fashion. thal et a& 1971; Baluda, 1972; Varmus et The presence of these genes can be ob- a& 1972; Neiman, 1973). More recently, reserved in two ways. A variety of chicken striction enzyme and hybridization analfibroblasts produce low levels of an infec- ysis has identified 16 different endogenous tious retrovirus that has a characteristic viral loci (m-1, ev-2, etc.) (Astrin, 1978; Assubgroup E envelope glycoprotein (Vogt trin et aL, 1980; Hughes et u& 1981). The and Friis, 1971; Crittenden et al, 1973; physical size and character of several loci Robinson, et a& 1976). Some fibroblasts, have been defined, and their expression into while not producing infectious virus, syn- RNA transcripts has been examined (Asthesize the major internal virion structural trin et ak, 1980; Hayward et d, 1980; proteins, the group-specific antigens, and Hughes et u& 1981). Furthermore, genetic the subgroup E envelope glycoprotein, the studies have utilized segregation analysis chicken-helper factor, in either a coordi- to assign biological phenotypes to particnate or noncoordinate fashion (Payne and ular physical loci (Astrin and Robinson, Chubb, 1968; Hanafusa et al, 1970; Weiss 1979; Astrin et aL, 1980; Crittenden and Astrin, 1981; Smith and Crittenden, 1981; Crittenden et &, 1983). Several loci, en-2, i To whom correspondence should be addressed. tn.+7,ev-10, ev-11, and ev-12, have been asINTRODUCTION

125

0042-6822B.4 $3.00 Copyright All rights

8 1984 by Academic Press, Inc. of reproduction in any form reserved.

126

HUMPHRIES,

DANHOF,

sociated with the production of infectious virus. In contrast, ev-3, ev-6, and ev-9 produce viral proteins in the absence of infectious virus. In situ hybridization has been employed to localize several of the endogenous loci to specific chromosomes. Ev-1, ev-4, ev-5, ev-6, ev-8, and ev-13 are all found on chromosome 1. Ev-2 and ev7 have been mapped to chromosome 2 and chromosome Z, respectively (Tereba and Astrin, 1980; Tereba et d, 1981; Tereba and Astrin, 1982). Chickens that lack any endogenous loci have been prepared by selective breeding (Astrin et d, 1979). These chickens have been reported to be normal, healthy, and fertile (Rovigatti and Astrin, 1983). While the endogenous viral loci may be nonessential for the development of healthy chickens, some selective advantage may result from the presence of eu-3 (Crittenden et d, 1982). The presence of multiple endogenous viral loci in the germ line of chickens may be the result of separate infection events, viral transpositions, or gene duplications. The presence of six loci on chromosome 1 suggests that the process of gene duplication may have been involved in the establishment of some of these endogenous loci (Tereba, 1981). It seems likely, however, based upon the structure of the long terminal repeat and the six-base duplication created at the site of residence for m-1 (Hishinuma et aL, 1981), that, in at least one instance, viral transposition or viral infection was responsible for the creation of an endogenous viral locus. We have examined flocks of five lines of chickens for the presence of endogenous viral loci, and report the identification of five new loci. One locus was responsible for the production of infectious endogenous virus. Virus production from a second locus was detected but the locus has not been identified. We have examined 12 loci for the presence of flanking cellular sequences that share homology with the cellular sequences that flank ev-1. Our results indicate that no such homology is present, and suggest that gene duplication of ev-l was not involved in thegeneration ofthe other ev loci examined.

AND

HLOZANEK MATERIALS

AND

METHODS

Cells and viruses. All cells were cultured in plastic dishes (Nunc, Denmark) with Dulbecco’s modified Eagle’s medium (Flow Laboratories, Rocksville, Md.), containing 10% tryptose phosphate broth (Difco Laboratories, Detroit, Mich.) supplemented with calf serum and fetal bovine serum (Sterile Systems, Logan, Utah) in a humidified atmosphere at 38” containing 5-10% coz. Chicken embryo fibroblasts (CEF) prepared from a number of inbred lines and closed breeding flocks of chickens have been used in this laboratory previously (Humphries and Glover, 1981; Humphries et al, 1981). CEF from a SPAFAS embryo (SPAFAS, Storrs, Conn.) contain only m-1. CEF from RPRL (Regional Poultry Research Laboratory, East Lansing, Mich.) line 6s, RPRL line ?‘a,and RPRL line 15s were used to examine m-3 (line 63), ev-2 (line ‘72),and ev-7 (line 15s). CEF derived from the Reaseheath C line were prepared from embryonated eggs obtained from the Houghton Research Laboratory, Huntingdon, England. These Reaseheath C line CEF will be referred to as RC-H. Other Reaseheath C line embryos were obtained from the flocks maintained at the Institute of Molecular Genetics, Prague, Czechoslovakia. The CEF derived from these chickens will be referred to as RC-P. Embryonated eggs were also obtained from the Reaseheath I and W lines maintained at the Institute of Molecular Genetics (Hala et a& 1966) and from the FP and SC lines from Hyline International (Dallas Center, Iowa). The Schmidt-Ruppin strain of avian sarcoma virus, subgroups A, D, and E, and the Prague strain of avian sarcoma virus, subgroups A, B, C, and E, were all maintained as cloned stocks and checked for plating efficiency on CEF from inbred lines of chicken possessing differential sensitivity to the subgroups of the ALSV complex. Cells that were C/E and chicken-helper factor positive (see below) were reexamined for sensitivity to subgroup E virus infection. Cells were first washed with 50 tiglycine-HCl (pH 2.2) prior to infection with virus (Humphries and Glover, 1981;

ENDOGENOUS

VIRAL

LOCI

IN WHITE

Robinson et aL, 1981), so that interference by the envelope glycoprotein could be reduced. The presence of the group-specific antigens was determined by indirect immunofluorescence using a murine monoclonal antibody against ~27 (Richert, 1982). The presence of chicken-helper factor was determined by fusion with the 16Q cell line as described previously (Humphries and Glover, 1981). Avian leukosis virus was not observed in any of the cells examined in this study. Spontaneous release of the endogenous virus was detected using culture supernatant from the test cell to infect permissive (Gr+) [IL& X K(-)] X KB CEF (Robinson, 1976). Virus produced from this infection was characterized for infection and replication on Gr- C/E and Gr- C/O CEF after prolonged cultivation, as determined by the presence of reverse transcriptase activity (Humphries et aL, 1981). Preparation of cellular DNA, restriction enzyme analysis, southern DNA transfer, and nucleic acid h&-idizaticm, Cellular DNA was prepared as previously described (Humphries et aL, 1981). All restriction enzymes were purchased from New England Biolabs (Beverly, Mass.). The digestion conditions were as defined in the product specifications. Digested cellular DNA was phenol extracted and precipitated prior to electrophoresis, as previously described

LEGHORN

CHICKENS

127

(Humphries et aL, 1981). DNA samples were analyzed by electrophoresis using 0.6 to 1% agarose gels (Marine Colloids, Rockland, Me.). In analysis of &a-specific DNA fragments, electrophoresis was continued until the bromphenol blue marker had migrated approximately 25 cm. Southern DNA transfer was carried out as described previously (Humphries et aL, 1981), except that some analyses required mild acid hydrolysis in 0.1 M HCl for 20 min in order to transfer high-molecular-weight DNA efficiently. Hybridizations were at 68” in 6X SSC, with 0.005 M EDTA, 0.5% SDS, and 0.1% bovine serum albumin. Some hybridizations were carried out in the same conditions but at 37”. All hybridizations contained 20 pg/ml of purified calf thymus DNA that had been boiled in 0.3 M KOH for 10 min. Five DNA probes were used in this study. Figure 1 illustrates the location of the sequences in relationship to the m-1 locus that these probes detect. A RAV-0 probe, consisting of an equimolar mixture of two subclones of RAV-0 provided by S. Hughes (Hughes, 1982), contained greater than 95% of the viral genome. These subclones, RS-HR-13 and R8-HR-16, were derived from R8 by Hind111 and EcoRI codigestion. A probe specific for the 5’ end of the endogenous virus was prepared from the mo-

x BAM HI n EC0 RI . HIND III

pHRl3 PHR I6 pss I4 pBB I2 P53

-

pGdIII

FIG. 1. Location of the sequences contained in the six DNA clones used as probes in this study. The endogenous viral locus eu-1 is represented as viral sequences (-), flanked by two LTRs (m), and located within cellular sequences (-----). The location of BmI-II (X), EcoRI (m), and Hind111 (0) restriction endonuclease cleavage sites are indicated.

128

HUMPHRIES,

DANHOF,

lecular clone, pSRA-2, of the SchmidtRuppin A strain of avian sarcoma virus (DeLorbe et al, 1980). The BamHI 1300-bp fragment was subcloned into the BumHI site of pBR322. This probe will be referred to as pBB-12. A probe specific for the 3’ BumHI fragment of RAV-0 was also subcloned from pSRA-2. A SW& DNA fragment of approximately 300 bp was subcloned into the SmaI site of pUC8. The fragment hybridizes to the 3.75kb EcoRI fragment and the 3.9-kb BarnHI fragment derived from RAV-0, and is located approximately 6.2 kb from the 5’ end of the virus. This probe will be referred to as pss-14. ~53, a clone from the Prague B strain of Rous sarcoma virus, contains the long terminal repeat and approximately 600 bases into the 3’ end of the src gene (Neiman et al, 1981). This plasmid was obtained from K. Beemon, and used as a probe to detect the 5’ and 3’ ends of the endogenous loci. pGdll1 was obtained from A. Skalka. This plasmid contains 3.3 kb of the cellular sequences that surround ev-I, and was prepared by cloning the unoccupied site of residence of ev-I from a bird that lacked endogenous loci (Hishinuma et al, 1981). These plasmids were labeled by nicktranslations to a specific activity of approximately 2 X lOa cpm/pg (Rigby et aL, 1977). They were prepared as probes by boiling in 0.3 MNaOH for 10 min, followed by neutralization. They were used immediately for hybridization at 10 rig/ml. Autoradiography was generally for 20 hr using RX film with DuPont Cronex intensifiers at -80”. Molecular size markers. A number of molecular size determinations were calculated in this study. The following values were used for DNA fragments that were used as standards throughout the analysis. These values were selected based upon the best agreement between a consensus for the different published values and our experimentally determined values. Several gels that contained all the DNA fragments provided the following set of DNA fragment standards: &cl-specific fragments ev-1, 9.5 kb; ev-2, 6.0 kb; ev-3, 6.3 kb; ev-7, 13 kb; and ew-10, 21 kb. EcoRI-specific

AND

HLOZANEK

fragments the 5’ ev-1 junction, 8.7 kb; eu1 internal, 3.75 kb; and the 3’ ev-1 junction, 17.5 kb. BarnHI-specific fragments the 3’ ev-1 junction, 5.2 kb; the 3’ ev-2 junction, 8.2 kb; the 3’ ev-3 junction, 7.3 kb; the 3’ ev-7 junction, 7.6 kb; and eu-1 internals, 1.35 and 1.65 kb. All molecular size determinations were based upon marker endogenous viral DNA fragments that were electrophoresed, transferred, and detected by hybridization on the same gel used to analyze the DNA fragment being characterized. RESULTS

Expression of Endogewus Viral Loci in the RC-P, RI, RW, FP and SC Lines of Chickens Cultured chicken embryo fibroblasts were tested for the expression of the avian group-specific (gs) antigens and chickenhelper factor (chf), the subgroup E envelope glycoprotein, as well as for the production of endogenous virus (Table 1). The three Reaseheath lines (RC-P, RI, and RW) are maintained as inbred lines, respectively B12Bi2, B’B’, and B15B15,and showed no variation in the phenotype of the four embryos tested. The data are consistent with those already published for the RC and RI lines (Payne and Chubb, 1968; Robinson, 1978). It should be noted, however, that the RI line produced an endogenous virus that had not been previously identified. Spontaneous production of endogenous virus was observed in all embryos tested for the RC-P, RI, and RW lines. The sensitivity of the RI CEF to subgroup E virus is masked by the production of the subgroup E envelope glycoprotein. Infection of these CEF with a subgroup E avian sarcoma virus required the use of the pH 2.2 glycine-HCl wash prior to adsorption of the virus (Robinson et a.& 1981). The phenotypes of these three inbred lines are, therefore, RC-P (C/ AE, gs-chf, V-E+), RI (C/O, gs+chf+, V-E+), and RW (C/E, gs+chft, V-E+). These results demonstrate that the RC-P line held in Prague, Czechoslovakia, has the same phenotype as that held in Huntington, England (RC-H). The SC line (BzB2) is maintained as an F1 line prepared from two parental inbred

ENDOGENOUS

VIRAL

LOCI

IN WHITE TABLE

LEGHORN

129

CHICKENS

1

PHENOTYPES OF CULTURED FIBROBLAST~ FROM FIVE CHICKEN LINES EXAMINED FOR ENDOGENOUS LOCI”

Chicken lineb

Susceptibility to subgroups A to E of the ALSV

Reaseheath C-P” Reaseheath I Reaseheath W Hyline FP Hyline SC

C/AE c/o C/E C/E c/o

Presence of gs antigen + + + -

Chicken-helper factor activity lo-10ed 103-10” 103-10’ 108-10’ 0

Production of subgroup E virus + + + -

a Fibroblast cultures were derived from individual embryonated eggs. The cells were characterized for the expression of endogenous loci and for susceptibility to infection by the subgroups of ALSV, as described under Materials and Methods. b Four different embryos were tested for each of the Reaseheath C, I and W lines. Ten embryos of each of the SC and FP Hyline chickens were tested. “The Reaseheath C line chickens examined in this study were obtained from the flocks maintained in Prague, Czechoslovakia, and are designated RC-P. d The low level of activity present in these embryos was probably due to the spontaneous production of subgroup E virus.

We have analyzed cellular DNA isolated from fibroblasts prepared from individual embryos of the RC-P, RI, RW, FP, and SC chicken lines. The results of the analysis with Sac1 endonuclease are presented in Fig. 2. The molecular sizes of the endogenous loci identified in the experiment are presented in Table 3. Ev-1 was present in all five DNA samples and has been omitted from these data. To characterize these loci further, BamHI endonuclease digestion was used to identify the 3’end of each locus. The BamHI 3’ junction fragments were identified by their hybridization to the pSS14 probe (Table 3). The BamHI 5’ junction fragment failed to hybridize to this probe Characterization of the Erubgw Viral but did hybridize to the p53 probe. The Loci Present in the RC-P, RI, R W, FP, Sac1 analysis, combined with the identiand SC Chicken Lines fication of the BamHI 3’ end of each locus, The work of Astrin (1978) and others identified five loci containing sequences re(Astrin et d, 1980; Hughes et al, 1981; Ter- lated to the endogenous avian retrovirus eba, 1981) has identified 16 different loci that have not been described previously in the chicken that contain DNA sequences (Tables 2 and 3). All four embryos of the Reaseheath homologous to the endogenous avian retrovirus. These loci have been characterized (Prague) C line, RC-P, contained the three loci ev-1, m-7, and ev-IO previously obby restriction enzyme analysis and/or served in RC-H birds (Astrin et al, 1980). chromosomal localization. The 16 different loci that have been identified have recently They also contained an additional locus of been reviewed (Rovigatti and Astrin, 1983), 11 kb that has not been seen in RC-H CEF, and that we have designated ev-1’7. The and the properties required to identify phenotype of this locus has not been dethem have been summarized (Table 2).

lines. The FP (B15Bz1)line is maintained as a cross between the dam of one inbred line (B21B21)and the sire of an F1 line (B15B15) prepared from two additional inbred lines. Because we expected heterogeneity in the FP and SC lines, we examined 10 embryos of each line. Despite the genetic background of these two lines, no heterogeneity was observed either in the expression of the endogenous loci present in these birds or in their susceptibility to the different subgroups of the ALSV complex. Their observed phenotypes were FP (C/E, gs+chft, V-E-) and SC (C/O, gs-chf-, V-E-).

130

HUMPHRIES, TABLE

DANHOF,

2

RESTRICTION ENZYME-SPECIFIC DNA FRAGMENTS THAT IDENTIFY THE DIFFERENT ev LOCI”

ev-1 en-2 w-3 eu-4 o-r-5 ev-6 ev-7 eu-8 m-9 ev-10 en-11 en-12 @J-14* ev-15 en-16

Size of the major Sac1 DNA fragment* (kb)

Size of the 3’ specific BomHI DNA fragmentc (kb)

Phenotype

9.5 6.0 6.3 8.7 19 21 13 18 2.3 21 13 8.1 9.5 4.2 5.4

5.2 8.2 7.3 7.3 13 4.4 7.6 23 11 14 NP NId 15 NDf ND’

None V+ gs+ chf + None None gs- chf + V+ None gB- ehf + V+ V+ V+ V+ None None

AND

HLOZANEK

or 4.4 kb (Table 2). One (or both) of these loci was probably responsible for the spontaneous production of an endogenous virus. While ev-7 is present in these cells, the virus produced from this locus is noninfectious unless it undergoes recombination (Robinson et aL, 1979). Most CEF containing ev-I and ev-7 are phenotypically gs-chf and do not produce infectious virus (Robinson et al, 1976; Crittenden et aL, 1977). Since infectious virus was consistently isolated from these cells, the cells are properly designated V-E+. Restriction enzyme analysis of cellular DNA from CEF of the FP and SC birds revealed that the distribution of ev loci in

RW

FP

RC RI

SC

17.5 ‘The data presented have been collected from several sonrees (Hughes et aL, 1981;Tereba et d, 1981;Rovigatti and Astrin, 1983). I%-13 was identified by in situ hybridization only, and no additional information is available (Tereba, 1981). ‘Several loci are cleaved by SacI. The size given is the size of the largest fragment. ‘This fragment contains the 3’ end of the locus together with the adjacent cellular sequences. d Not identified. The EamHI fragment of ev-11 apparently con&rates with that of m-1, m-6, or m-10. The BarnHI fragment of m-12 apparently corn&rates with that of ev-1, en3, or eu-6. ’ The molecular sizes are estimates based upon data from Tereba et aL (1981). fNot determined (Hughes do&, 1981).These loci may eontsin only long terminal repeats.

termined but it did not express gs antigens or chf (Table 1). All four embryos of the Reaseheath I line contained ev-1 and ev-3 as predicted (Payne and Chubb, 1968; Astrin and Robinson, 1979). They also contained a locus of 10.5 kb designated here as ev-18. While it has not been demonstrated by segregation analysis, this locus was associated with the spontaneous production of virus (V-E’) (Table 1). The four embryos of the Reaseheath W line contained ev-I, ev-3, and ev-7, as well as two loci, ev-19 (7.6 kb) and ev-20 (8.1 kb), previously unidentified. B-12 is also 8.1 kb, but the BumHI fragment is either 5.2,7.3,

13.5

- e_v-1 8.7 z5

5.2

FIG. 2. Sac1 endonucleaze identification of the endogenous viral loci present in five lines of chickens. Cellular DNA isolated from CEF of the Reazeheath (Prague) C line (RC), I line (RI), and W line (RW), and the Hyline FP and SC lines was digested with SacI endonuclease and analyzed by electrophoresis on a 0.7% agarose gel, Southern DNA transfer, and hybridization with a RAV-0 DNA probe. The markers are in kilobases. B-1, 9.5 kb, is indicated. Autoradiography waz for 20 hr.

ENDOGENOUS VIRAL

LOCI IN WHITE LEGHORN

131

CHICKENS

TABLE 3 ENDOGENOUS LOCI PRESENT IN THE RC-P, RI, RW, FP, AND SC LINES’

Chicken line

Size of the major Sac1 DNA fragment

Size of the 3’ specific BarnHI fragment

Loci previously identifiedb

New locus designation”

RC-P

11 13 21

NId 7.6 14

f?V-7 ev-10

ev-17

RI

6.3 10.5

7.3 25

ev-3 -

6.3 7.6 8.1 13

7.3 9.8 or 18e 18 or 9.8 7.6

ev-3 en-7

en-19 ev-20

6.3 9.2 22

7.3 20 4.5

en-3 ev-6

ev-21

8.8

7.2

ev-4

RW

FP

SC

e-u-18

a Cellular DNA from CEF of the RC-P, RI, RW, FP, and SC chickens was analyzed by restriction enzyme digestion, electrophoresis, Southern DNA transfer, and hybridization with viral specific DNA probes. All DNA fragments detected except for e-u-l are listed. The %&specific DNA fragments were identified by hybridization with the RAV-0 probe. The BornHI-specific DNA fragments were identified by hybridization both to the RAV-6 probe and the pSS-14 probe. bBased upon the Sac1 and BarnHI analyses, these loci are indistinguishable from loci already described (see Table 2). ‘By Sac1 and BurnHI analyses, these loci have not been described previously and have been assigned a locus number. dNI, not identified. The 3’ end of this locus appears to comigrate with the 3’ end of either m-1, en-7, or ev-10. ’ Since these loci did not segregate in the embryos we examined, it was not possible to determine which BornHI DNA fragment was derived.from which locus.

these two flocks was heterogeneous. Con- Virus-CeU Junction Fragments of Ev17 and sequently, 35 FP and 42 SC birds were anEv-18 alyzed by restriction enzyme analysis for the presence of eu loci (Table 4). While all Six restriction endonucleases were used of the SC birds contained ev-1, only 50% to identify the 5’ and 3’ virus-cell junction contained ev-4. All of the FP birds con- fragments of 12 ev loci. BarnHI, &$I, BglII, tained ev-1, ev-3, and a locus previously EcoRI, HindIII, and XbaI were used to anunidentified and designated here as eu-21. alyze m-1, ev-2, ev-3, ev-4, ev-6, ev-7, evThe FP birds were segregating at eu-6. Two 10, en-17, ev-13, m-19, ev-20, and ev-21. other loci were present in 10 to 20% of the While there were exceptions, enzyme-spebirds. The SacI-specific DNA fragments cific 3’ and 5’ DNA junction fragments containing these loci were 6.0 and 10.5 kb. characteristic of the different loci were The BumHI-specific 3’ junction fragments identified in most instances (data not of these loci have not been identified. The shown). These data were consistent with phenotype of eu-21 is unknown but it was the hypothesis that each locus was .situated not associated with spontaneous virus pro- in a unique region of cellular DNA. Howduction- (Table 1). ever, in several analyses, the junction

132

HUMPHRIES,

DANHOF,

AND

TABLE

HLOZANEK

4

FREQUENCY OF ENDOGENOUS LOCI IN FP AND SC CHICKENS Number

of birds containing

specific loci

Line

Number of birds examined

G-V-1

ev-3

ev-4

ev-6

ev-21

6.0 kbb

10.5 kbb

FP SC

35 42

35 42

35 0

0 19

28 0

35 0

4 0

6 0

’ 35 FP birds and 42 SC birds were examined for the presence of ev loci by analysis with the restriction endonuclease SacI. b The BanzHI-specific 3’ end of the eu loci characterized by the 6.0- and 10.5-kb Sac1 DNA fragments have not been identified.

fragments of ev-17 and ev-I8 either could not be identified or were observed to be identical to the junction fragments of other loci. Since CEF containing only m-17 or ev-18 were not available, the identification of junction fragments was indirect. Identification of the junction fragments of ev17 required comparison of the DNA fragments produced by enzyme digestion of DNA from cells containing either en-l, ev-7 and ev-10 or ev-I, ev-7, ev-10, and ev -17. This was done by comparing DNA from CEF from the birds of the two RC flocks kept at Houghton (RC-H) and Prague (RC-P), respectively. Similarly, ev18 junction fragments were characterized by comparing DNA from line 63, which contains ev-1 and ev-3, and RI CEF, which contain ev-I, ev-3, and m-18. The results from this study provided two types of information (Table 5). First, some junction fragments were identical in size. The 5’3.1kb junction fragment produced by EcoRI digestion of ev-17 was identical to the EcoRI-specific 5’ junction fragment of ev18. BgZI digestion of DNA from cells containing m-18 produces a 21.5-kb DNA fragment that appears identical in size to the BglI-specific DNA fragment containing the m-10 locus (Fig. 3). BgZI does not appear to cut within the viral-specific sequences of any of the ev loci except ev-3 (Humphries, unpublished data). Second, the junction fragments of a-17 and ev-18 were not always identified. BamHI endonuclease fails to identify either a 3’ or 5’ junction fragment specific for ev-17. Similarly, Hind111 and EcoRI both fail to iden-

tify 3’ junction fragments specific for .ev18. In the context of all of the data which we obtained during the restriction enzyme analysis of these loci, the easiest explanation for the failure to identify these junction fragments was that they comigrated with junction fragments derived from another locus in the same cell. For ev-17, the second locus could have been ev1, eu-7, or m-10 (Table 5). In the case of ev-18, this locus could have been ev-1 or ev-3. A similar observation has been reTABLE

5

VIRUS-CELL DNA JLJNC~~ONFRAGMENTS COMMON TO m-17 OR ev-18 AND A SECOND ENDOGENOUS Locus

Locus

Restriction endonuclease/ characteristic fragment’

Second locus

m-17 eV-17 f?V-17 ev-18 tW-17 er-18 e-u-18

EceRI/5’, 3.1 kb EcoRI/3’, 1.4 kb HindIII/3’, 1.6 kb BgW21.5 kb’ BamHI/3’ (NI)d, 5’ (NI) HindIW3’ (NI) EcoRI/3’ (NI)

ev-18 eu-1 (2”)b f?u-1(27 @J-l0 ev-1, ev-7, or eu-10 et?-1or w-3 en-1 or ev-3

‘Restriction enzyme digestion of either ~1’7 or en-18 identified a junction fragment identical to one associated with a second ev lccus. The BglI, EceRI, and iYindII1 junction fragments were identified as either 5’ or 3’ by hybridization with the pBB-12 and p53 probes. BernHI junction fragments were identified by hybridization with the pSS-14 and p53 probes. bThis is a secondary (2”) cleavage site (see text). ‘This BglI fragment hybridized pBB-12, pSS-14, and ~53, and represents the entire lecus. ’ NI, not identified. This junction fragment appears te cemigrate with one derived from one of the loci present in the same cell.

ENDOGENOUS

SP

RI

VIRAL

LOCI

IN WHITE

RC

27.021.516.5-

- e_v-

LEGHORN

133

CHICKENS

(Figs. 4 and 5). The site responsible for producing the 1.6-kb fragment is located 5’ to the normal site and is just 0.4 kb from the end of the 3’ long terminal repeat (LTR). In several experiments, a 3’-specific 1.4-kb fragment was observed following EcoRI digestion of ev-1 that was identical in size to the EcoRI-specific 3’ junction fragment of ~~-1’7 (data not shown). The 3’ junction fragment normally observed after EcoRI analysis is 17.5 kb. Again, the site, responsible for producing the 1.4-kb EcoRI-specific DNA fragment is located approximately 0.2 kb 3’ to the end of the SP

L6

L15

RC

L7

5.1 -

FIG. 3. BglI endonuclease analysis of er-10 and er-18. Cellular DNA isolated from CEF derived from a SPAFAS embryo (SP), a Reaseheath I line (RI) embryo, and a Reaseheath (Houghton) C line (RC) embryo was digested with BglI, and analyzed by electrophoresis on a 0.7% agarose gel, Southern DNA transfer, and hybridization with a RAV-O DNA probe. The SPAFAS DNA contains only en-1 (16.5 kb), while RI DNA contains eu-1, m-3 (10.0 kb), and eu-18 (21.5 kb). RC (Houghton) DNA contains eu-1, eu-7 (27 kb), and eu-10 (21.5 kb). The 16.5-kb DNA fragment migrated approximately 8 cm on this gel.

ported for BamHI analysis of ev-11 and ev-12 (Table 2). Finally, a frequent, but inconsistent, observation was the identification of an additional viral-specific DNA fragment following either Hind111 or EcoRI digestion of m-1. In many experiments, Hind111 digestion of ev-1 produced a 1.6-kb fragment that was identical in size to the HindIII-specific 3’ junction fragment of eu17 (Fig. 4). This fragment hybridized with the p53 probe but not the pBB-12 probe. The 3’ junction fragment normally observed after Hind111 analysis is 1.9 kb

3.45-

1.9 1.6-

FIG. 4. Hind111 endonuclease analysis of m-1 and eu-1’7. Cellular DNA was isolated from SPAFAS (SP), line 6* (L6), line 15s (L15), Reaseheath (Prague) C line (RC), and line ‘7x(L7) CEF, and analyzed for DNA fragments that hybridize to a RAV-0 DNA probe following Hind111 digestion. The 5.1-, 3.&i-, and 1.9-kb DNA fragments are the 5’ junction, internal, and 3’ junction DNA fragments normally observed following Hind111 digestion of m-1. The 1.6-kb DNA fragment is a 3’-specific junction of m-1 that is occasionally observed and that migrates with the 3’ junction of eu-17 present in RC (Prague) DNA.

HUMPHRIES,

134

l

Hind III

-

DANHOF,

0.5 kb

. EcoRI

FIG. 5. Location of EcoRI and Hind111 cleavage sites in the 3’ region of en-l. The 3’ portion of m-1, 2 kb of viral sequences (I), and the 3’ flanking cellular sequences (-) are shown. EcoRI (B) and Hind111 (0) cleavage sites are marked. The two secondary sites are within parentheses. These secondary sites define the EcoRI- and X&III-specific 3’ junction fragments of eu-17. The 3’-DNA junction fragment normally produced by EcoRI digestion is 17.5 kb.

3’ LTR (Fig. 5). Both of these experiments suggested that there were two sites in the flanking 3’ region of ev-1 that were identical to the sites that define two 3’junction fragments of m-17. It is necessary to emphasize that the 1.4-kb EcoRI-specific DNA fragment and the 1.6-kb HindIII-specific DNA fragment were detected inconsistently following analysis of ev-1. These fragments have been detected using both a cloned RAV-0 DNA probe and a tdPRB 35 S genomic RNA probe. These fragments have not been reported in other studies. While methylation should not influence the availability of either EcoRI or Hind111 sites, it is possible that variable secondary structure influences the cleavage of these sites. Since both of these sites appear to be recognized inconsistently, we refer to them as secondary cleavage sites that flank ev-1. Cellular Sequences that Flunk Ev-1 do not Hybridize to the Sequences that Flunk the Other Ev Loci The data presented in the previous sections suggested two somewhat contradictory findings. First, the extensive mapping of the 12 ev loci listed earlier indicated they were all located within distinct cellular sequences.This situation was not only predicted from the Sac1data (Tables 2 and 3), but also from nearly all the results of other investigations (Astrin et al, 1980; Hughes et a& 1981; Rovigatti and Astrin,

AND

HLOZANEK

1983). In contrast, analysis of m-17 and ev18 frequently identified junction fragments that were characteristic of a second locus or that presumably corn&grated with a second junction fragment (Table 5). The ev1 locus was identified twice directly (and, potentially, three times indirectly) as the second locus. It was possible, therefore, that some degree of relatedness existed between the flanking sequences of ev-1 and those of ev-1’7 or ev-18. To test this hypothesis, a DNA probe containing the unoccupied site of ev-1, pGdll1, was hybridized to the E&I-specific DNA fragments that contain ev-17 and ew-18 (Fig. 6). The pGdll1 probe contains 2.3 kb of cellular DNA immediately 5’ to m-1 and 1.0 kb of DNA immediately 3’ to ev-1 (Fig. 1). If the SacI-specific DNA fragments containing ev-17 and ev-18 have flanking cellular sequences similar to those flanking ev-1, the pGdll1 probe would be expected to hybridize to these fragments. As seen in Fig. 6, the probe hybridized only to eu-1. In fact, it failed to hybridize to any of the SacIspecific DNA fragments that contain ev FP

SC

M

SP

RC

M

RI

RW - 17.5

-9.5

@-II

--8.7

FIG. 6. Hybridization of en-1 flanking cellular sequences to DNA of different chicken lines. Cellular DNA was isolated from CEF of the Hyline FP and SC chickens, a SPAFAS embryo (SP), and the Reaseheath (Prague) C line (RC), I line (RI), and W line (RW). After digestion with SacI, the DNA was analyzed by electrophoresis on a 0.7% gel, Southern DNA transfer, and hybridization to the pGdll1 DNA probe. The markers (M) are SPAFAS DNA, containing only en-l, digested with EcoRI. Two DNA fragments of 8.7 and 17.5 kb, containing the 5’ and 3’ halves of eu1, are produced.

ENDOGENOUS

VIRAL

LOCI

IN WHITE

LEGHORN

CHICKENS

135

supported the general hypothesis that these loci, and others (ev-1, ev-2, ev-3, c-v4, ev-6, ev-7, and ev-10) were, in general, characterized by different flanking cellular sequences. Some of the ev loci have been localized by in situ hybridization, either to different chromosomes or to different regions on the same chromosome (Tereba, 1981; Tereba et aL, 1981). The data, therefore, indicate that the ev loci contain viral sequences with the gene order present in RAV-0 (with or without deletions). These viral sequences are located within cellular sequences that are unique with respect to each other, and that may be present on different chromosomes or in different regions within the same chromosome. Restriction enzyme analysis of m-17 and ev-18 indicated that the cellular sequences that flank the viral DNA of these loci have several enzyme sites at distances from DISCUSSION their 3’ and 5’ long terminal repeats that correspond to similar sites that flank other Using the BumHI and Sac1 restriction endonucleases, we have analyzed the en- ev loci (Table 5). While these data sugdogenous loci present in five lines of chick- gested a similarity in the flanking cellular ens. Five of the loci we observed have not sequences around these loci, data from been described previously. These loci have analysis with other enzymes, as already been designated ev-17, ev-18, ev-19, m-20, discussed, demonstrated that significant and ev-21. Ev-18, found in the Reaseheath differences also existed. There was a sugI line, was associated indirectly with the gestion, therefore, that a limited similarity spontaneous production of infectious en- existed between several loci, one of which dogenous virus in all four of the RI embryos was ev-1. Similar data obtained by analysis that we studied. The four RW embryos that of other loci have been reported by others we examined also released infectious en- (Table 2). (i) The BamHI-specific 3’ ends dogenous virus spontaneously. The pres- of ev-3 and ev-4 are identical in size. (ii) ence of five loci in the RW birds prevented The BamHI 3’ ends of ev-11 and ev-I2 apdefinitive association of the V-E+ pheno- parently comigrate with the 3’ ends of type with a particular locus. Cells con- other ev loci, perhaps ev-1, present in the taining ev-7 are V-15+ and can produce in- same cell. Similar observations have been fectious virus, but recombination is nec- noted following analysis of additional loci essary and induction is frequently required (Humphries, unpublished data). It is likely (Robinson et aZ.1976; 1979). However, since that some of these apparent identities are CEF containing ev-1 and m-7 are pheno- coincidental in nature. The frequent identypically gs-chf-, it is likely that either (or tification of ev-1, however, as similar, or both) ev-19 or ev-20 are responsible for the potentially similar, to many of the loci production of virus. Eu-17 does not express studied suggested that some distant regs antigens or chf, and m-21 is not asso- lationship might exist between m-1 and these loci. Such a relationship has been ciated with production of virus. The restriction enzymes used to define proposed for ev-I and ev-4, ev-6, m-8, evthese five loci, Sac1 and BamHI, charac- 9, and m-13 (Tereba, 1981) based upon (i) terized these loci as different from the the presence of common structural features other loci that had been described (Table within the viral sequences and (ii) their 2). Further, restriction enzyme analysis location on chromosome 1. Such a possi-

loci, including ev-2, m-3, ev-4, ev-6, ev-7, eu-IO, ev-17, ev-18, ev-19, ev-20, and ev-21 (Fig. 6). The position of ev-1 in this gel is located by the SPAFAS DNA sample, since ev-1 is the only ev locus in this DNA. The same cellular DNA preparations examined in Fig. 6 were reexamined by digestion with BglI and hybridization to the pGdll1 probe. Hybridization was detected only at the location of the 16.5kb DNA fragment containing m-1 (data not shown). The hybridization described in Fig. 6 was done at 68” in 6X SSC. It was repeated at 37” in 6X SSC. Again, hybridization to loci other than ev-1 was not observed (data not shown). The results of these experiments indicate that the cellular sequences that flank ev-1 were not associated with the other ev loci examined.

136

HUMPHRIES,

DANHOF,

bility is made more likely since ev-1 is present in 99% of the chickens examined, and may represent the earliest introduction of viral DNA into the chicken germ line. To evaluate this hypothesis, a cloned probe, pGdll1, containing the cell sequences that flank ev-1 (but not eu-1 itself), was hybridized to SacI- or BglI-digested DNA from several different lines of chickens. Both of these enzymes identify the different ev loci as large DNA fragments in association with adjacent cellular DNA sequences. The pGdll1 probe hybridized only to DNA fragments that contain ev-1. No hybridization to DNA fragments containing ev-2, ev-3, m-4, ev-6, ev-7, ev-10, ev-17, ev-18, ev-19, ~~-20, or m-21 was detected (Fig. 6 and data not shown). The failure to detect hybridization indicates that either no homology exists between the flanking sequences of ev-1 and the flanking sequences of the other loci or that the homology is too weak and/or too dispersed to hybridize the probe. Four different mechanisms have been proposed for the generation of the endogenous viral loci of the chicken. These include (i) independent viral infection of the germ line, (ii) gene duplication (including duplication of both viral sequences and flanking cellular sequences), (iii) viral transposition, and (iv) intracellular replication and integration via an RNA intermediate (Hughes, et al, 1981; Tereba, 1981; Rovigatti and Astrin, 1983). The sixnucleotide repeat surrounding ev-1 suggests that m-1 was generated either by viral infection of the germ line, by viral transposition from another locus, or by intracellular replication and integration. Based upon the presence of 5’-terminal deletions in ev-4, ev-5, and ev-6 (Hayward et CL&1980; Hughes, et al, 1981) and the location of the six loci on chromosome 1, gene duplication and viral transposition have been suggested as two mechanisms by which ev-4, ev-5, ev-6, ev-8, and ev-13 could have been derived from ev-1. Our characterization of m-17 and ev-18 suggested that, though they were distinct, these loci shared a limited homology with the flanking sequences of ev-1. Such a sit-

AND

HLOZANEK

uation might have been created by a gene duplication associated with extensive nucleotide alterations in the associated flanking regions. The lack of hybridization with pGdll1 demonstrated that either no gene duplication occurred and the restriction site similarities were coincidental, or extensive alteration has occurred since the duplication and no significant homology remains. Sequence data from the flanking cellular sequences5’ to ev-2, while revealing the presence of purine-rich regions similar to those observed in flanking sequences 5’ to m-1, also demonstrated a lack of homology between these cellular sequences and, therefore, support our conclusion (Scholl et aL, 1983). Our data further suggest that the proposed generation of ev-4 and ev-6 by gene duplication of ev-1 would require that a similar alteration of the flanking sequences of ev-4 and ev-6 have occurred since the duplication. It is possible that gene duplication has gone unrecognized by hybridization with the pGdll1 probe either because other loci have been duplicated or because extensive modification and/or elimination of flanking sequences occurred after duplication. Testing these hypotheses would require additional molecular cloning and extensive sequence analysis of flanking cellular DNA. The data, therefore, appear most consistent with the creation of endogenous loci within the germ line either by independent viral infections or by viral translocations. Some evidence for independent infections comes from the location of ev-2, ev-7, and ev-14, which code for complete virus, on separate chromosomes. It is possible that ev-10 and ev-11, which also code for complete virus and segregate independently, are located randomly on other chromosomes. However, in the absence of gene duplication, the presence of ev-4, ev-5, ev6, ev-8, and ev-13 on chromosome 1, three of which have deletions of the 5’ long terminal repeat, is curious. It is possible that there are DNA sequences on chromosome 1 that interact preferentially with viral DNA and that promote faulty insertion in this chromosome. Alternatively, they may represent the reintegration of defective viral genomes produced by faulty viral ex-

ENDOGENOUS VIRAL LOCI IN WHITE LEGHORN

CHICKENS

137

cisions associated with transpositions. BALUDA, M. A. (1972). Widespread presence in chickens of DNA complementary to the RNA genome of Such viral excisions, involving proviruses avian leukosis viruses. Proc Nat1 Acad SC% USA of the murine ecotropic endogenous virus, 69,576-580. have been postulated to explain reversion of mutations in the mouse that leave a sin- COPELAND,N. G., HUTCHISON,K. W., and JENKINS, N. A. (1983).Excision of the DBA ecotropic provirus gle copy of the long terminal repeat in the in dilute coat-color revertants of mice occurs by reversion site (Copeland, et u.?.,1983). Enhomologous recombination involving viral LTRs. dogenous viral loci that appear to contain Cell 33, 379-387. a single long terminal repeat have also CRITTENDEN,L. B., WENDEL, E. J., and MOTTA, J. V. been observed in the chicken (Hughes et (1973). Interaction of genes controlling resistance CJA,1981). Understanding the process(es) to RSV (RAV-0). V&logy 52,373-384. by which the endogenous loci of the chicken CRITTENDEN,L. B., MOT~A, J. V., and SMITH, E. J. (1977). Genetic control of RAV-0 production in were generated may help to develop a more chickens. virology 76, 90-97. complete appreciation of the importance of insertional mutagenesis and its influence CRITTENDEN,L. B., and ASTRIN, S. M. (1981). Independent segregation of ev 2 and ev 10, genetic loci on the expression of cellular genes. for spontaneous production of endogenous avian ACKNOWLEDGMENTS We thank Caroline Glover for excellent technical assistance in the preparation of the subclones of pSRA-2 (pBB-12 and pSS-14). We also thank Karen Beeman, Stephen Hughes, and Ann Skalka for generously providing the molecular clones ~53, R&HR13, Rf3-HR-16, and pGdll1; Brad Ozanne for helpful discussions and critical evaluation of the manuscript; and Daisi Tucker for preparation of the manuscript. This work was supported by Public Health Service Award CA-32295 from the National Institutes of Health and Grant I-962 from the Robert T. Welch Foundation. REFERENCES ASTRIN, S. M. (1978). Endogenous viral genes of the White Leghorn chicken: Common site of residence and sites associated with specific phenotypes of viral gene expression. Proc Natl Acad Sci USA 75,59415945.

ASTRIN,S. M., Buss, E. G., and HAYWARD,W. S.. (1979). Endogenous viral genes are non-essential in the chicken, Nature (Lo&m) 282, 339-341. ASTRIN, S. M., and ROBINSON,H. L. (1979). Gs, an allele of chickens for endogenous avian leukosis viral antigens, segregates with en-3, a genetic locus that contains structural genes for virus. J. vird 31.420-425. ASTRIN, S. M., CRITPENDEN,L. B., and Buss, E. G. (1980). eu 2, a genetic locus containing structural genes for endogenous virus, codes for Rous-associated virus type 0 produced by line 7z chickens. J. ViTOl

33, 250-255.

ASTRIN, S. M., ROBINSON,H. L., CRIYTENDEN,L. B., Buss, E. G., WYBAN,J., and HAYWARD,W. S. (1980). Ten genetic loci in the chicken that contain structural genes for endogenous avian leukosis viruses. Cold Spring Harbor Symp. Quad. BioL 44, 11051109.

retroviruses. fir110, X20-127. CRIY~ENDEN,L. B., FADLEY, A. M., and SMITH, E. J. (1982). Effect of endogenous leukosis virus genes on response to infection with avian leukosis and reticuloendotheliosis viruses. Avian Dis 26, 279294.

CRITTENDEN,L. B., ASTRIN, S. M., and SMITH, E. J. (1983). Independent segregation of m-10 and eu-11, genetic loci for spontaneous production of endogenous avian retroviruses. Fir129, 514-516. DELORBE,W. J., LUCIW,P., GOODMAN,H. M., VARMUS, H. E., and BISHOP,J. M. (1980). Molecular cloning and characterization of avian sarcoma virus circular DNA molecules. J. fir01 36, 50-61. HALA, K., HASEK, M., HLOZANEK,I., HORT, J., KNIZETOVA,F., and MERVARTOVA,H. (1966). Syngeneic lines of chickens. II. Inbreeding and selection within the M, W and I lines and crosses between the C, M and W lines. Folia Bid (Prah.a) 12, 407-422. HANAFUSA, H., MIYAMOTO, T. and HANAFUSA, T. (1970). A cell-associated factor essential for formation of an infectious form of Rous sarcoma virus. Proc. Nat1 Ad Sci USA 66, 314-321. HANAFUSA, H., HANAFUSA, T, KAWAI, S., and LuGENBUHL,R. E. (1974). Genetic control of expression of endogenous virus genes in chicken cells. fir58, 439-448.

HAYWARD, W. S., BRAVERMAN,S. B., and ASTRIN, S. M. (1980). Transcriptional products and DNA structure of endogenous avian proviruses. Cold, Spring Harbor Symla Quant. Bid 44,1111-1121. HISHINUMA, F., DEBONA, P. J., ASTRIN, S. M., and SKALKA, A. M. (1981). Nucleotide sequence of acceptor site and termini of integrated avian endogenous provirus en 1: integration creates a 6 bp repeat of host DNA. Cell 23,155-X4. HUGHES, S. H., BISHOP, J. M., and VARMUS, H. E. (1981). Organization of the endogenous proviruses of chickens: Implications for origin and expression. Virology 108, 189-207.

138

HUMPHRIES,

DANHOF, AND HLOZANEK

HUGHES,S. H. (1982). Sequence of the long terminal repeat and adjacent segments of the endogenous avian virus Rous-associated virus-o. J. vird 43, 191-200. HUMPHRIES,E. H., ALLEN, R., and GLOVER,C. (1981). Clonal analysis of the integration and expression of endogenous avian retroviral DNA acquired by exogenous viral infection. J. viral 39, 584-596. HUMPHRIES,E. H., and Glover, C. (1981). Independent expression of avian sarcoma virus in doubly infected chicken embryo fibroblasts. J. vird 37,721-729. NEIMAN,P. E. (1973) Measurement of endogenous leukosis virus nucleotide sequences in the DNA of normal avian embryos by RNA-DNA hybridization. Virw 53, 196-204. NEIMAN, P., BEEMON,K., and LUCE, J. A. (1981). Independent recombination between avian leukosis virus terminal sequences and host DNA in virusinduced proliferative disease. Proc. Nati Ad Sci USA 78, 1896-1999. PAYNE,L. N., and CHUBB,R. C. (1968). Studies on the natural and genetic control of an antigen in normal chick embryos which reacts in the COFAL Test. J. Gfm ViroL 3, 379-391. RICHERT,N. D. (1982). Monoclonal antibody specific for avian sarcoma virus structural protein ~27. J. Gen ViroL 62.385-392. RIGBY,P. W. J., DIECKMANN,M., RHODES,C., and BERG, P. (1977). Labelling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. MoL Bid 113, 237-251. ROBINSON,H. L. (1976). Intracellular restriction on the growth of induced subgroup E avian type C viruses in chicken cells. J. ViroL 18, 856-866. ROBINSON,H. L., SWANSON,C. A., HRUSKA,J. F., and CRITPENDEN,L. B. (1976). The production of unique C type viruses by chicken cells grown in bromodeoxyuridine. virology 69, 63-74. ROBINSON,H. L. (1978). Inheritance and expression of chicken genes that are related to avian leukosis sarcoma virus genes. Curr. Top Microbid ImmwwL 83, l-36. ROBINSON, H. L., EISENMAN,R., SENIOR,A., and RIPLEY, S. (1979).Low-frequency production of recombinant subgroup E avian leukosis viruses by uninfected V-15s chicken cells. l%o&v 99, 21-30. ROBINSON,H. L., ASTRIN, S. M., SENIOR,A. M., and SALAZAR,F. H. (1981). Host susceptibility to en-

dogenous viruses: Defective glycoprotein-expressing proviruses interfere with infections. J. ViroL 40, 745-751. ROSENTHAL,P. W., ROBINSON,H. L., ROBINSON,W. S., HANAFUSA,T., and HANAFUSA, H. (1971). DNA in uninfected and virus-infected cells complementary to avian tumor virus RNA. Proc Nat1 Ad Sci USA 68,2336-2340.

ROVIGATTI,U. G., and ASTRIN, S. M. (1983). Avian endogenous viral genes. Curr. Top. Microbid ImmunoL 103,1-22. SCHOLL,D. R., KAHN, S., MALAVARCA,R., ASTRIN, S., and SKALKA, A. M. (1983). Nucleotide sequence of the long terminal repeat and flanking cellular sequences of avian endogenous retrovirus ev-2: Variation in Rous-associated virus-0 expression cannot be explained by differences in primary sequence. J. ViroL 45, 868-871. SMITH,E. J., and C~~ITENDEN,L. B. (1981).Segregation of chicken endogenous viral loci eu 7 and a, 12 with the expression of infectious subgroup E avian leukosis viruses. ViroloQl 112,370-373. TEREBA,A. (1981).5’-Terminal deletions are a common feature of endogenous retrovirus loci located on chromosome 1 of White Leghorn chickens. J. vird 40,920-926. TEREBA,A., and ASTRIN, S. M. (1989). Cbromosomal localization of eu-1, a frequently occurring endogenous retrovirus locus in White Leghorn chickens, by in situ hybridization. J. ViroL 35.888-894. TEREBA, A., CRI~ENDEN, L. B., and ASTRIN, S. M. (1981). Chromosomal localization of three endogenous retrovirus loci associated with virus production in White Leghorn chickens. J. Viral 39, 282-289. TEREBA,A., and ASTRIN, S. M. (1982). Chromosomal clustering of five defined endogenous retrovirus loci in White Leghorn chickens. J. ViroL 43. 737-740. VARMUS,H. E., WEISS,R. A., FRIIS, R. R., LEVINSON, W., and BISHOP,J. M. (1972) Detection of avian tumor virus-specific nucleotide sequences in avian cell DNAs. Proc NatL Ad Sci USA 69,20+X VOGT,P. K., and FRIIS, R. R. (1971). An avian leukosis virus related to RSV(0): properties and evidence for helper activity. virdogy 43, 223-234. WEISS,R. A., and PAYNE, L. N. (1971). The heritable nature of the factor in chicken cells which acts as a helper virus for Rous sarcoma virus. lrirology 45, 508-515.