- Email: [email protected]
Generation of packaging-defective dna molecules of equine adenovirus
VIROLOGY
(1986)
151,66-76
Generation of Packaging-Defective
DNA Molecules of Equine Adenovirus
TOSHIRO ISHIYAMA,* MORIKAZU SHINAGAWA,* GIHEI KEI FUJINAGA,? AND RADHA PADMANABHAN$’
SATO,*
*Department of Veterinary Public Health, School of Veterinary and Veterinury Medicine, Obihiro 080, Japan; tsapparo
Medicine, Obihiro University of Agriculture Medical Colkge, Sapporo 060, Japan; and $Llqnwtment of Biochemistry, University of Kansas Medical Center, 39th and Rainbow Boulevard, Kansas City, Kansas 66103 Received May 6, 1985; accepted January
90, 1986
Equine adenovirus (EAd) DNA prepared from infected bovine kidney (MDBK) cells contained additional sequences of about 100 to ‘700 bp at the left-hand end of the genome. These aberrant viral genomes were produced even after the first passage of the wild type EAd in MDBK cells and their relative amounts did not change significantly during serial passage. The left terminal fragments of two defective viral DNAs were cloned into the plasmid vector pBR322 and the nucleotide sequences of their terminal regions were analyzed. The data indicate that one viral DNA contained a duplication of the inverted terminal repetition (ITR) and the other contained 2’70 bp of additional sequences derived from the right-terminal region of EAd genome added to the left-terminal, ITR. While the former DNA was packaged into virions, the latter was not, presumably due to the alteration of the distance from the left terminus to the putative DNA packaging signal, reported to be located between 290 and 390 bp (Hammarskjold and Winberg, 1980). The possible mechanism for the generation of these defective DNAs is discussed. 0 1986 Academic PMS. IDC. INTRODUCTION
at each terminus. Electron microscopic analysis of viral replicative intermediates isolated from cells infected with adenovirus type 2 revealed two basic types of intermediates. Type I molecules consisted of unit length linear duplexes with one or more single-stranded branches. Type II replicating molecules consisted of unit length, unbranched linear molecules that contained both double-stranded and singlestranded regions (Ellens et al, 1974; Lechner and Kelly, 1977). Type I intermediates represent the displacement mechanism for the parental strand and type II intermediates represent the replication process of the displaced parental strand which could arise from the putative panhandle intermediate (Daniell, 1976). Evidence for the interaction of the two molecular ends during DNA replication was provided by Stow (1982). Deletions of different lengths, shorter than the unit length of the ITR, introduced at the left end were repaired during replication of the viral genome with an intact ITR at the
Adenovirus (Ad) genomes contain linear double-stranded DNA with a molecular weight of 20-30 X lo6 (Green and Pina, 1964; van der Eb and van Kesteren, 1966; Norrby et aL, 1976). Ad DNA possesses a unique inverted terminal repetition (ITR) of the type abc - - -. c’b’a’, which permits the formation of single-stranded circular DNA after denaturation and reannealing of the two strands (Garon et aL, 1972; Wolfson and Dressler, 1972). An additional novel feature of Ad genome is that the 5’ end of each DNA strand is covalently linked to a protein of 55 kDa in size (Robinson et aL, 1973; Sharp et al, 1976; Rekosh et aL, 1977; Padmanabhan and Padmanabhan, 1977; Shinagawa et aL, 1979). Although the biological function of the ITR is not completely understood, it plays an important role in DNA replication in having an origin ‘To whom dressed.
requests
for
reprints
0042-6822186 $3.00 Copyright All rights
0 1986 by Academic Press, Inc. of reproduction in any form reserved.
should
be ad-
66
PACKAGING-DEFECTIVE
EAd
DNA
67
right end. Deletions larger than the length genome. In this paper, we describe the characterization of these variant moleof ITR proved to be defective in replication. Hammarskjold and Winberg (1980) re- cules, which form a different class from ported that a DNA sequence located be- those characterized in other laboratories in which terminal rearrangements and dutween 290 and 390 bp from the left terminus of Ad16 contained recognition signals plications occurred at the right end. We for encapsidation of the viral DNA. The find that the variant EAd DNA molecules location was determined by analyzing a se- which arose by the acquisition of about 100 ries of spontaneous mutants of Ad16 which bp at the left end was packaged into viarose by serial passage in HeLa cells and rions, whereas those which had 270 bp or carried duplications of 200 to 500 bp of left more were not. terminal sequences at the right end of the genome. A duplication of about 390 bp seMATERIALS AND METHODS quences from the left end at the right end enabled the subgenomic defective viral Virus and cells. T-l strain of EAd type 1 DNAs containing both molecular ends to be encapsidated. However, the incomplete (Kamada et al, 1977) was obtained from M. virus particles containing the left end se- Kamada, Equine Research Institute, Japan quences, of less than 290 bp duplicated at Racing Association. The virus had been right end did not promote encapsidation of passaged three times in MDCK cells, and right-terminal sequences. six times in MDBK cells at the time it was Serial passage of Ad12 in Vero cells of used in this study. MDCK and MDBK cells African green monkey kidney cells (Wer- were grown in Eagle’s minimum essential ner and zur Hausen, 1978) resulted in the medium (MEM) containing 5% calf serum. insertions of 190-280 bp at the right end After infection with the virus, the cells and deletions of 70 bp at the left end of the were maintained in MEM containing 1% genome. These variants were selected be- fetal calf serum. Preparation of viral DNA. The cell culcause of their efficient growth properties in Vero cells. Similar alterations of the tures on monolayer were prepared in 55 viral genome were observed when the virus X 105 mm bottles and infected with the was serially passaged in human cervical EAd stock. The viral DNA was extracted carcinoma cells (Kruczek et ah, 1981) and from the purified virions as described by in human melanoma cells (Schwarz et al, Green and Pina (1964). Isolation of viral DNA from the infected cells was according 1982). These cell lines do not generally support an efficient growth of the virus. to Shinagawa et al. (198313). Enzymes. Restriction endonucleases However, serial passage in these cell lines resulted in the generation of these host BumHI, EcoRI, HueIII, HincII, H&fI, P&I, range mutants with duplication of the ITR SuZI, and SmaI, T4 DNA polymerase, T4 and its adjacent sequences of the left end polynucleotide kinase, and T4 DNA ligase were purchased from Takara-Shuzo (Kyoto, at the right end of the viral genome. Equine Ad (EAd) propagates efficiently Japan) and used as specified by the supin the cells derived from the natural host plier. Bacterial alkaline phosphatase was such as primary or low-passaged equine from Sigma (St. Louis, MO.) and was used kidney cells. It was also found that EAd in the presence of 0.1% SDS (Shinagawa was able to grow efficiently in canine kid- and Padmanabhan, 1979). Cloning of EAd Since a plaque assay for ney cells (MDCK) and bovine kidney cells (MDBK). When the EAd DNA prepared EAd in MDCK or MDBK cells is not yet from stock virus propagated in MDBK cells successfully developed, limiting dilution was analyzed, we discovered the presence technique was employed for cloning of EAd of altered viral DNAs in addition to the and its variants. Monolayers of MDCK and MDBK cells in five culture tubes were inwild type molecules and these variants seemed to arise from the acquisition of ad- fected with 2 ml of EAd viral stock serially ditional sequences at the left end of the diluted from 10e5 to lo-* with MEM con-
68
ISHIYAMA
taining 1% fetal calf serum. The infected cells showing the cytopathic effect using the most or the second most serially diluted virus were frozen and thawed to prepare the virus stock. The lysates were used to infect bottle cultures to obtain sufficient viral DNA for restriction endonuclease analysis and for molecular cloning. Physical mapping of EAd DNA. Restriction endonuclease cleavage maps of EAd DNA for BamHI, EcoRI, and Hind111 were constructed using the combination of partial digestion and the conventional double digestion methods. The terminal fragments were identified using the EAd DNA labeled at the 3’ end with 9 (Challberg and Englund, 1980). By convention, the A-T rich terminal fragment is designated as the right terminal fragment (Doerfler and Kleinschmidt, 1970; Mulder et al, 1974; Anonymous, 1977). In order to identify the A-T rich end fragment, Hind111 and BamHI digests of EAd was melted in 50% formamide for 15 min at different temperatures (49-54’) and then quickly chilled in ice and electrophoresed on an agarose gel (Zsak and Kisary, 1981). The terminal fragment that melts at a lower temperature than the other is designated as the right-terminal fragment. Southern hybridization. DNA fragments fractionated on an agarose gel by electrophoresis were transferred to nitrocellulose filters according to the method of Southern (1975). The probe for hybridization was prepared by digesting the plasmids containing BamHI-D or -D3 with SalI. The SalI site was filled in by repair synthesis catalyzed by DNA polymerase (Klenow fragment) in the presence of [a-32P]dCTP and unlabeled dTTP. The labeled DNA was treated with BamHI for pOHEAdBD or HpaII for pOHEAdBD3 plasmid. The filters containing DNA were hybridized to a =P-labeled probe at 40” for 20 hr, washed, dried, and exposed to x-ray films as described earlier (Fujinaga et al, 1979). The 32P-labeled probes were prepared by nick translation (Mackey et al, 1977). Molecular cloning of BamHI-terminal fragments of EAd The BamHI terminal fragments (D, Dl, and D3 in Fig. 2) of EAd DNA, prepared from infected MDBK cells
ET AL.
were purified by electrophoresis on agarose gels and treated with 0.5 M piperidine at 37” for 3 hr (Stillman et aL, 1982) to remove the residual peptides covalently bound to the original 5’ end of EAd DNA (Roninson and Padmanabhan, 1980). The DNA fragments were renatured in 1 M NaCl in 10 mm Tris-HCl, pH 7.5, and 1 mM EDTA at 65” for 6 hr. The vector plasmid, pBR322 DNA, was linearized by digestion with SalI and the 5’-protruding ends were made blunt by treatment with T4 DNA polymerase in the presence of 4 deoxynucleoside triphosphates. The DNA fragments were subsequently digested with BamHI and the 5’ ends were dephosphorylated by treatment with bacterial alkaline phosphatase. After three extractions with buffer-saturated phenol, the DNA was precipitated by the addition of two volumes of ethanol. The vector plasmid and the BamHI terminal fragments were mixed and ligated using T4 DNA ligase at 16” for 15 hr (Maniatis et al, 1982). For transformation, the Escherichia coli strain SGll, a ret A mutant of SG8 (Shinagawa et al, 1982) was used as the host (Cohen et al, 1973). Transformants were selected on heart infusion agar plates containing 25 pg/ml of ampicillin. The plasmid DNAs were isolated and analyzed by digestion with restriction endonucleases, BamHI and SalI. DNA sequence analysis. The plasmid DNA containing an EAd-BamHI terminal fragment (D, Dl, or D3) was digested with SalI and treated with bacterial alkaline phosphatase. The DNA was labeled at the 3’ end using DNA polymerase (Klenow) and [cy-32P]dTTP and digested with PstI. The two labeled fragments were fractionated by electrophoresis on an agarose gel. The labeled terminal fragment was subjected to the chemical degradation procedure of Maxam and Gilbert (1977) for DNA sequence analysis. RESULTS
Characterization of variant EAds possessing elongated terminal BamHI-D fragmerits. We discovered that EAd DNA prepared from the virions propagated in MDBK cells contained defective DNA mol-
PACKAGING-DEFECTIVE
ecules resulting from insertions of DNA sequences at the left terminus as shown in Fig. 1A (BarnHI-Dl). However, when the DNAs from 208 viral isolates from MDCK cells were examined by BumHI digestion, the patterns were all identical to each other in having only five fragments as shown in Fig. lB-a and the BumHI-Dl was not present. When the DNA from MDBK cells, in-
A
EAd
fected by 45 isolates of viral stocks, were analyzed by BumHI digestion, the pattern showed the presence of the terminal BumHI-D fragment and, in addition, the elongated fragments, Dl, D2, D3, and so on up to at least D’7 (Fig. lB-b). The sizes of these terminal fragments, BumHI-D, -Dl, -D2, and -D3 were estimated to be about 17’70,1870,1920, and 2030 bp, respectively.
B
C a' b'
a b
VA IL B ‘-c
69
DNA
o
b
A-
/A
BC-
=;
FIG. 1. Restriction endonuclease BamHI cleavage pattern of EAd DNA. (A) EAd DNA, prepared from purified virions which were propagated in MDBK cells, was digested with BarnHI and electrophoresed on an agarose gel. (B) EAd was cloned by limiting dilution technique in MDCK or MDBK cells as described in the text. The isolates were grown in the respective cells. The viral DNA extracted from the infected cells was digested with BamHI and electrophoresed on an agarose gel. DNA prepared from MDCK cells is shown in B-a, and from MDBK cells in B-b. The DNA fragments in the gel were examined by the Southern hybridization (1975) using q-labeled BarnHI-D of EAd DNA as a probe (B-a’, B-b’). (C) EAd DNAs prepared from virions and from infected cells were compared. MDBK cells were infected with EAd virus stock which was grown in MDBK cells. The infected cells were divided into two portions. The viral DNA was prepared directly from one portion of the infected cells, while the virions were purified from the other before the extraction of DNA. Both DNAs were digested with BarnHI and analyzed by agarose gel electrophoresis. Lane a contained DNA from the purified virions, and b, DNA from the infected cells.
70
ISHIYAMA
To confirm that the elongated fragments were derived from BumHI-D, the sequence homology between these fragments and BumHI-D was examined by Southern hybridization using 32P-labeled BamHI-D as a probe. As shown in Fig. lB-b’, the probe hydridized to all of these elongated DNA fragments, indicating that these fragments were derived from BamHI-D. Interestingly, the hybridization of the probe to the DNA isolated from MDCK infected cells showed only the presence of the terminal BumHI-D fragments (Fig. lB-a’). In Fig. lC, the EAd DNA was prepared from one portion of the infected cells, while the virions were purified from the other portion and subsequently, the viral DNA was extracted. Both DNAs were digested with BamHI and the digests were analyzed by electrophoresis on agarose gel. The results were the same as shown in Figs. 1A and B. These results indicate that the variant EAd DNAs containing BamHI-D2, -D3, and further elongated fragments were deficient in DNA packaging. E#ect of serial passage of EAd wild type virus in MDBK cells on the generation of elongated DNA molecules. We investigated the effect of serial passage of EAd in MDBK cells on the generation of defective genomes. Wild type EAd stock, obtained by cloning in MDCK cells was serially passaged in MDBK cells and the viral DNAs synthesized in the infected cells after second, third, sixth, and ninth passage were analyzed by BamHI digestion. As shown in Fig. 2A, BarnHI-Dl, -D2, -D3, and the further elongated fragments already appeared after second passage and the relative amounts seemed to be unaffected by subsequent serial passage of the virus. The results were reproducible as shown in Figs. 2B and C in which the virus was passaged three times. It is interesting to note that the replication of the virus in MDBK cells even during the first passage (Figs. 2B and C) produced the elongated DNA molecules. Location of BamHI-D fragment in the EAd genome. Although it was reported that BamHI-D fragment of EAd DNA was one of the terminal fragments (Shinagawa et al., 1983a), its location at the right or the
ET AL. A
B
C
FIG. 2. BumHI digestion of EAd DNA after serial passage in MDBK cells. Wild type EAd was serially paasaged in MDBK cells. The viral DNA was prepared from infected cells at the indicated passage level, digested with BumHI, and electrophoresed on an agarosegel. Passage numbers are shown above each lane. (B) and (C) represent duplicate experiments.
left molecular end of the genome was unknown. In order to determine the orientation of the BamHI terminal fragments of EAd, we made use of the observation of earlier investigators that the A-T rich half of the viral DNA is positioned to the right of the adenovirus genome in general (Doerfler and Kleinschmidt, 1970; Mulder et al, 1974). It would mean that the end fragment that melted at a lower temperature than the other was considered to be located at the right end of the DNA molecule (Anonymous, 1977). To find the A-T rich half of EAd DNA, BumHI digest of the DNA was heated in 50% formamide at various temperatures between 45 and 52”, chilled in ice rapidly, and then electrophoresed on an agarose gel. As shown in Fig. 3B, one of the terminal fragments, BumHIB melted at 51°, but the BamHI-D fragment was intact at that temperature, but melted at 52”, indicating that BumHI-B is A-T rich and hence, by convention, is located at the right end. Similar results were obtained when Hind111 digest was exam-
PACKAGING-DEFECTIVE
EAd
DNA
A DE
1 11
C
4:9-7.0 E
I
25.0 B
11 6.‘2
A
A
1 34.6’
I
I
c
BamHI
7911 A
0 53.;
II
I 51.4
,D
10
I I
G42’7C
37.1
B
EcoRI 60.2 E
F
B
100
I II 6f.6
71.9-77.2
Hind111
,E
10
FIG. 3. Restriction endonuclease cleavage maps of EAd DNA. (A) Cleavage maps of EAd DNA for BarnHI, EcoRI, and Hind111 were constructed as in the text. (B) To define the left-to-right orientation of the cleavage maps, BarnHI digest of EAd DNA in 50% formamide was heated at the indicated temperature for 15 min, chilled quickly, and electrophoresed on an agarose gel (Zsak and Kisary, 1981).
ined. HindIII-B fragment melted at 49”, whereas HindIII-A was intact at that temperature indicating that HindIII-B is A-T rich (data not shown). Therefore, BumHID fragment, which is a subfragment of HindIII-A is located at the left end and the elongation of EAd DNA occurred at the left end of the genome. Physical mapping and DNA sequence analysis of the elongated segments of EAd. In order to study these variants, arising due to the acquisition of additional sequences at the left end, at the molecular level, we cloned the BamHI-D, -Dl, and -D3 fragments into pBR322 vector as described under Materials and Methods. The recombinant plasmids containing the respective terminal fragments, -D, -Dl, and -D3 were designated as pOHEAdBD, pOHEAdBD1, and pOHEAdBD3, respectively (Fig. 4A). Since the regeneration of the Sal1 site was used for screening the re-
combinants resulting from the blunt end ligation of the left terminal BamHI fragments of EAd variants to the SalI-cut and filled-in site of the vector, only those molecules containing the dC at the 5’ left terminus such as Dl and D3 were cloned. D2 variant was not seen among the recombinants. The elongated segments of the left terminal BumHI-D, BamHI-Dl, and -D3 were analyzed by mapping the restriction endonuclease cleavage sites on the cloned fragments, pOHEAdBD, pOHEAdBD1, and pOHEAdBD3 for HincII, PstI, S’aZI, and SmaI (Fig. 4A). Among the enzymes used, H&c11 gave rise to the smallest fragment, 237 bp long that contained the original left end of EAd DNA in pOHEAdBD. In pOHEAdBD1 and pOHEAdBD3, the HincII fragments were 338 and 506 bp, respectively, in length due to the addition of DNA sequences at the left end during
ISHIYAMA
72
ET AL.
A POHEAdBD
Hincll cmtl
I
Him11 BMlHI
B D
5'-CATCATCMT MTATACAGG ACACACGGGCATGGGGCCM GAAAGGGGAGGAGTTGAGW GWGCGGGAG GCGGGGGCGG AGGCGGGGCG GCGGGCGG~ G ITR
Dl
amCTT---
5'4ATCATCAAT MTATACAGG ACACACGGGCATGGGGCCAAGMAGGGGAGGAGTTGAGGCGTGGCGGGAG TCATCAA IAATATACAG GACACACFCGCATGGGGGCA GCGGGGGCGG AGGCGGGGCG G4Y3XCGGG.A AGAAAGGGA GGAGTTGAGGCGTGGCGGGA GAGGCGGGGC GGCGGGCGGG A AAAAATTC---
D3
5'-cG4XCTCCCI TATGCGACTC CT~~~M~~C~~AGTAGGTTGILGGCCGTTGACCICCCCCG CCGCMGGAA TGGTGCATGCAAGGAGATGGCGGCCAACAGTCCCCCGGCCACGGGGCCTGCCACCATACC CACWXAAA CAAGCGCEA TGAGCCCGM GTGGCGAGCC CCXST’IWC CATCGGIGAT GTCGGCGATA TAGGCGCCAGCAACCGCACC TGTGGCGCCGGTGATGCCGGCACGATGCGTTCGGCTAG4G CAICAATAAT ATACAGGA--
hTR
FIG. 4. Restriction site mapping and nucleotide sequence analysis of the variants at the left end. (A) Restriction endonuclease cleavage maps of the cloned left-terminal fragments, BarnHI-D, -Dl, and -D3 of EAd DNA. The thick lines indicate the regions derived from EAd DNA and the thin lines indicate the region from pBR322. Fine maps of the extreme left-terminal H&c11 fragments for Hue111cleavage sites are shown in the expanded regions under each restriction map. The numbers indicate length (bp) of the fragments. (B) Nucleotide sequences at the extreme left end region of the cloned EAd DNAs, BornHI-D, -Dl, and -D3. The nucleotide sequences were determined using the chemical method of Maxam and Gilbert (1977) as described under Materials and Methods. The end point of ITR sequences are shown with bracketed arrow. In Dl, the second copy of ITR starts at nucleotide No. 3 and ends with a dG instead of the normal dC residue. In D3, the extra sequences of 270 nucleotides are fused to the 4th nucleotide of ITR.
the growth of EAd in MDBK cells (Figs. 4A and B). The fine mapping of H&c11 fragments of pOHEAdBD, pOHEAdBD1,
and pOHEAdBD3 was carried out using Hue111 (Fig. 4A). The H&II fragments of 86 and 115 bp were conserved in both
PACKAGING-DEFECTIVE
pOHEAdBD1 and pOHEAdBD3 in the same orientation as in pOHEAdBD (Fig. 4A) indicating that the additional sequences are added to the 36-bp Hue111 fragment at the original left end of the genome. We determined the nucleotide sequences of the wild type and the variant DNAs at the original left end of EAd (from SalI site toward the H&c11 site in Fig. 4A) and the sequence data are presented in Fig. 4B. The length of ITR is 103 bp in EAd D in Fig. 4B. In Dl, the sequence data showed that there is a tandem duplication of ITR at the left end except that there is a deletion of two nucleotides from the second copy of ITR. However, in D3, three nucleotides from the left end ITR of wild type were replaced by a stretch of 270 nucleotides and the nucleotide sequence data revealed that, unlike the case of Dl, the region was not a tandem duplication of ITR region. Nucleotide sequence analysis at the right terminal end using BumHI-B fragment of EAd DNA prepared from the infected MDBK cells indicated that there was no alteration to the EAd genome at the right terminus. Origin of the additional sequences at the left end of pOHEAdBD3. In order to determine whether the sequences added to the left end of the viral DNA in the molecular clone, pOHEAdBD3 is of cellular or viral origin, the DNA from the virions isolated from the infected MDCK cells was digested with BamHI (Fig. 5) and fractionated by electrophoresis on an agarose gel. The DNA fragments were denatured, neutralized, and transferred to a nitrocellulose filter. The probe DNA to be used for Southern hybridization should consist of sequences only from the newly acquired region at the left-hand end and not include ITR sequences which are common to both molecular ends. The probe was prepared from pOHEAdBD3 as described under Materials and Methods using HpuII as the second enzyme, which cleaves at a site 117 nucleotides away from the left-hand terminus of the variant, D3, (see Figs. 4A and B) within the newly acquired sequences, but 150 bp upstream from the ITR region. The ter-
EAd
DNA
73
FIG. 5. The origin of the extra sequences added to the left end of EAd. EAd DNA, purified from the virions isolated from the infected MDCK cells was digested with BamHI and electrophoresed on an agarose gel. The DNA fragments were transferred to a sheet of nitrocellulose filter and hybridized with the 120bp-long HpoII fragment as the probe which represents the left terminal region of the EAdBD3 DNA which contains the extra sequences added during replication. The labeled HpuII was isolated by digesting the plasmid pOHEAdBD3 with SalI and repairing the cohesive ends with DNA polymerase (Klenow fragment) in the presence of [o-“PJdCTP and unlabeled d’ITP. The labeled DNA was digested with H&I and fractionated by agarose gel electrophoresis. The hybridization was carried out according to Southern (1975). The internal BarnHI-E fragment ran out of the gel in this experiment.
minally labeled HpaII fragment was used as a probe for Southern hybridization of a BamHI digest of EAd DNA and the results are shown in Fig. 5. BumHI-B fragment hybridized with the probe indicating that these additional sequences are of viral origin and are derived from the right end of the viral genome.
ISHIYAMA
74
ET AL.
and different segments of the adjacent unique sequences. Growth of EAd in We have shown that EAd prepared di- MDBK cells differ from the growth of Ad12 rectly from the infected MDBK cells con- in Vero cells or human melanoma cells in tained defective DNA molecules in addition that the variants were generated even at to the wild type. These variants arose due the first passage due to the insertion of seto a duplication of ITR (EAdBDl) or due quences at the left end and the relative to the addition of sequences which showed amounts of these variants and the wild no homology to ITR (EAdBD3). Of all these type seemed to remain constant during sevariants, only EAdBDl was packaged into rial passage (Fig. 2). This observation inthe virions, as the increase in the length dicates that these variants do not seem to of the genome at the left end is only 101 have any improved growth properties, unbp. The other defective DNAs are found like Ad12 variants (Werner and zur Hauonly in the infected MDBK cells and are sen, 1978; Schwarz et al, 1982). In addition probably deficient in DNA packaging. The to the Ad12 variants, others that were deficiency of these molecules (EAdBD2, characterized (Hammerskjold and Win3, etc.) in DNA packaging is unlikely due berg, 1980; Garon et aL, 1982; Brusca and to their larger size of the genomes be- Chinnadurai, 1983) resulted from the recause the Ad capsid has the capacity to duplication of the left-end ITR and the adpackage Ad DNA which is up to 105% in jacent unique sequences at the right end. length (Hammarskjold and Winberg, 1980; However, when EAd replicates in MDBK Schwarz et al, 1982; Garon et al., 1982; cells, reduplications of ITR (EAdBDl) as Brusca and Chinnadurai, 1983). Daniel1 well as other sequences which map within (1976) and Tibbetts (19’7’7)reported that the right-hand terminal BumHI-B fragsequences from the left end of the genome ment (as in the case of EAdBD3) occurred are preferentially incorporated into in- at the left end of the genome. The reducomplete particles of Ad3 and Ad7. Anal- plication of 101 bp of ITR at the left end ysis of the DNA isolated from the incom- of EAd did not block the viral DNA to be plete particles prepared from the vari- packaged into the virions. However, the ants of Ad16 in which the sequences unique variant with a longer insertion of 270 bp to the left end were added to the right at the left end was not packaged. It is posend enabled Hammarskjold and Winberg sible that the distance from the left end of (1980) to conclude that there is a recogni- the genome to the DNA packaging signal tion signal for the encapsidation of viral can be varied up to approximately 101 bp DNA located about 390-400 bp from the for the DNA to get packaged but not up to left end and the reduplication of this signal 270 bp. at the right end enabled the packaging of The mechanism by which these aberrant the right terminal DNA into incomplete molecules of EAd are generated when EAd particles. undergoes replication in MDBK cells but A different type of variant was charac- not in MDCK cells is unknown at present. terized by Schwarz et al. (1982) who re- Ad DNA replicates in permissive cells ported that after serial passage of Ad12 in through a strand displacement model as cells of human melanoma cell line, variants originally proposed by J. Sambrook (see with enhanced growth properties were iso- Daniell, 1976). The elongation of the lated which carry additional sequences of daughter strands toward the opposite end regularly increasing size at the right end results in the progressive displacement of of the genome. Sequence analysis of these one of the parental strands (which were variants indicated that these extra se- seen as type I molecules under the elecquences were of viral origin. They origitron microscope by Lechner and Kelly, nated from either the extreme right or left 197’7). The self-complementary terminal end of the genome consisting of the ITR sequences of the completely displaced DISCUSSION
PACKAGING-DEFECTIVE
strands, probably hybridizes to each other to form the putative “panhandle” intermediate (Daniell, 1976) before replication to form the second daughter molecule in a normal permissive cell. The experimental evidence for the interaction of the two molecular ends during the replication of Ad DNA was reported by Stow (1982). It is interesting to note that replication of viral genome containing an additional 18 dG residues at the left end of Ad2 gave rise to the progeny viral genomes that contained perfectly normal ITR, suggesting that the interaction between the two ITRs and subsequent cellular repair might be involved in the restoration of normal ends (Stow, 1982). The model proposed by Daniel1 (1976) suggests that the displaced strand is susceptible to breakage and its 3’ end could fold back on itself at a region of partial homology and continue its synthesis until it reaches the 5’ end. The double-stranded DNA molecule could then be formed via a type II mechanism. If this model is correct, the length of reduplication of DNA sequences from one terminus to the other would depend on the location of the foldback region for the DNA synthesis. This model would require, for example, the displaced parental 1 strand of EAd which probably lost two or three nucleotides from the left end ITR could then transiently hybridize to a region adjacent to the righthand ITR and undergo DNA synthesis to generate an ITR reduplication variant. These molecules have the potential to undergo homologous recombination between the internal ITR and the terminal ITR either on the same or on a different molecule. The variants, EAdBD2,3, etc., might be the products of such recombination events, although a rather high frequency is required to produce these molecules which seems rather unlikely. According to this model, the reduplication of DNA sequences is not expected to be a specific event at either terminus since both displaced parental strands (r and 1) would participate in the generation of these mutants.._.. However, - our studies indicated that the addition of extra
EAd
DNA
75
sequences seems to be specific to the left end. The reason for this specificity is not clear. Further work is necessary to map the precise location of the DNA sequences within the right-terminal fragment which were duplicated at the left end during replication. ACKNOWLEDGMENT This work was partly supported by a grant-in-aid for Cancer Research from the Ministry of Education, Science and Culture, Japan, and by Grant 56480067, from the same Ministry (to M.S.) and from the National Cancer Institute (CA 33099 to R.P.). We thank Linda Hicks for typing this manuscript. REFERENCES ANONYMOUS (19’77). Adenovirus strand nomenclature: A proposal. J. V+oL 22,830~831. BRUSCA, J. S., and CHINNADURAI, G. (1983). Structure and reversion of an adenovirus type 2 mutant containing a duplication of the left end of the genome at the right end. virology 129,381-392. CHALLBERG, M. D., and ENGLUND, P. T. (1980). Specific labeling of 3’ termini with T4 DNA polymerase. In “Methods in Enzymology” (L. Grossman and K. Moldave, eds.), Vol. 65, pp. 39-43. Academic Press, Orlando, Fla. COHEN, S. N., CHANG, A. C. Y., BOYER, H. W., and HELLING, R. B. (1973). Construction of biologically functional bacterial plasmid in vitro. Proc Nat1 Acad Sci USA 79,3240-3244. DANIELL,, E. (1976). Genome structure of incomplete particles of adenovirus. J. viral 19.685-708. DOERFLER, W., and KLEINSCHMIDT (1970). Denaturation pattern of the DNA of adenovirus type 2 as determined by electron microscopy. J. Mel Bill 50, 579-593. ELLENS, D. J., SUSSENBACH, J. S., and JANSZ, H. S. (1974). Studies of the mechanism of replication of adenovirus DNA. III. Electron microscopy of replicating DNA. V+oEogy 61,427-442. FUJINAGA, K., SAWADA, Y., and SEKIKAWA, K. (1979). Three different classes of human adenovirus transforming DNA sequence. virdogy 93,578-581. GARON, C. F,, BERRY, K. W., and ROSE, J. A. (1972). A unique form of terminal redundancy in adenovirus DNA molecules. PTIX. Nat1 Acud Sci USA 69,24552459. GREEN, M., and PINA, M. (1964). Biochemical studies on adenovirus multiplication. VI. Properties of highly purified tumorigenic human adenoviruses and their DNA’s. Proc NatL Acad Sci USA 51.12511259.
76
ISHIYAMA
HAMMARSKJOLD,M. L., and WINBERG,G. (1980). Encapsidation of adenovirus 16 DNA is directed by a small DNA sequence at the left end of the genome. Cell 20,787-795. KAMADA, M., AKIYAMA, Y., SATO,K., and KODERA,S. (1977). Isolation of adenovirus from adult thoroughbred horses. Japan. J. Vet. Sci. 39,661-664. KRUCZE~,I., SCHWARZ,E., and ZURHAUSEN,H. (1981). Mutants of adenovirus type 12 after adaptation of growth in tumor cell lines. II. Reproducible acquisition of additional sequences after adaptation to a human cervical cancer line. Znt. J. Cancer 27,139143. LECHNER, R. L., and KEUY, T. J., JR. (1977). The structure of replicating adenovirus 2 DNA molecules. Cell 12,1007-1020. MACKEY,J. K., BRACKMANN,K. H., GREEN,M. R., and GREEN,M. (1977). Preparation and characterization of highly radioactive in vitro labeled adenovirus DNA restriction fragments. Biochemistry 16,44784483. MANIATIS,T., FRITSCH,E. F., and SAMBROOK, J. (1982). “Molecular Cloning: A Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. MAXAM, A. M., and GILBERT, W. (1977). A new method for sequencing DNA. Proc. Nat1 Acnd Sci USA 74, 560-564. MULDER,C., ARRAND, J. R., DELIUS, H., KELLER, W., PETTERSSON, U., ROBERTS,R. J., and SHARP,P. A. (1974). Cleavage mapping of DNA from adenovirus type 2 and 5 by restriction endonucleases, EcoRI and HpaI. Cold Spring Harbor Symp. Quad Bid
ET AL. SCHWARZ,E., REINKE, C., YAMAMOTO,N., and ZUR HAUSEN,H. (1982).Terminal rearrangements in the genome of adenovirus type 12 mutants adapted to growth in two human tumor cell lines. virology 116, 284-296. SHARP,P. A., MOORE,C., and HAVERTY,J. (1976). The infectivity of adenovirus 5-DNA protein complex. Virology 75,442-456.
SHINAGAWA,M., ISHIYAMA, T., PADMANABHAN,R., FUJINAGA,K., KAMADA, M., and SATO, G. (1983a). Comparative sequence analysis of the inverted terminal repetition in the genomes of animal and avian adenoviruses. Virology 125,4X-495. SHINAGAWA,M., MAKINO, S., HIRATO, T., ISHIGURO, N., and SATO, G. (1982). Comparison of DNA sequences required for the function of citrate utilization among different citrate utilization plasmids. J Bacterial. 151,1046-1050. SHINAGAWA,M., MATSUDA, A., ISHIYAMA, T., GOTO, H., and SATO,G. (1983b). A rapid and simple method for preparation of adenovirus DNA from infected cells. Mic~obiol. ZmmunoL 27,817-822. SHINAGAWA,M., and PADMANABHAN,R. (1979). Inhibition of a nuclease contaminant in the commercial preparations of Escherichia coli alkaline phosphatase. Anal. B&hem. 95,458-464. SHINAGAWA,M., PADMANABHAN,R. V., and PADMANABHAN,R. (1979). Isolation and characterization of a homogeneous DNA-protein complex from adenovirus type 2 virion. J. Gen. Viral. 45, 519-525. SOUTHERN, E. M. (1975).Detection of specific sequences among DNA fragments separated by gel electrophoresis. .Z Mol. Biol 98,503-517. STILLMAN,B. W., TOPP,W. C., and ENGLER,J. A. (1982). 39,397-400. Conserved sequences at the origin of adenovirus NORRBY,E., BARTHA, A., BOULANGER,P., DREIZIN, DNA replication. J. viral 44,530-537. R. S., GINSBERG,H. S., KALTER, S. S., KAWAMURA, STOW,N. D. (1982). The infectivity of adenovirus geH., ROWE,W. P., RUSSELL,W. C., SCHLESINGER, R. nomes lacking DNA sequences from their left-hand W., and WIGAND, R. (1976). Adenoviridae. Inter&termini. Nucleic Acids Rex 10, 5105-5119. rologv 7,117-125. TIBBETTS,C. (1977). Viral DNA sequences from inPADMANABHAN,R., and PADMANABHAN,R. V. (1977). complete particles of human adenovirus type 7. Cell Specific interaction of protein(s) at or near the ter12,243-249. mini of adenovirus 2 DNA. B&hem Biophys. Res. VAN DER EB, A. J., and VAN KESTEREN,L. W. (1966). Cmmun 75,955-%X. Structure and molecular weight of the DNA of adREKOSH,D. M. K., RUSSELL,W. C., BELLETT,A. J. D., enovirus type 5. B&him Biophys. Acta 129,441. and ROBINSON,A. J. (1977). Identification of a proWERNER,G., and ZURHAUSEN,H. (1978). Deletion and tein linked to the ends of adenovirus DNA. CelE11, insertion in adenovirus type 12 DNA after viral 283-295. replication in Vero cells. Virology 86, 66-77. ROBINSON,A. J., YOUNGHUSBAND, H. B., and BELLET~, WOLFSON,J., and DRESSLER,D. (1972). Adenovirus 2 A. J. D. (1973). A circular DNA-protein complex DNA contains an inverted terminal repetition. Pnw: from adenoviruses. Virology 56,54-69. Natl. Acad Sci. USA 69,3054-3057. RONINSON,I., and PADMANABHAN,R. (1980). Studies ZSAK, L., and KISARY, J. (1981). Studies on egg drop on the nature of the linkage between the terminal syndrome (EDS) and chick embryo lethal orphan protein and the adenovirus DNA. B&k Biophys. (CELO) avian adenovirus DNAs by restriction enRes. Commun 94,398-405. donucleases. J. Gen firol 57.87-95.
(1986)
151,66-76
Generation of Packaging-Defective
DNA Molecules of Equine Adenovirus
TOSHIRO ISHIYAMA,* MORIKAZU SHINAGAWA,* GIHEI KEI FUJINAGA,? AND RADHA PADMANABHAN$’
SATO,*
*Department of Veterinary Public Health, School of Veterinary and Veterinury Medicine, Obihiro 080, Japan; tsapparo
Medicine, Obihiro University of Agriculture Medical Colkge, Sapporo 060, Japan; and $Llqnwtment of Biochemistry, University of Kansas Medical Center, 39th and Rainbow Boulevard, Kansas City, Kansas 66103 Received May 6, 1985; accepted January
90, 1986
Equine adenovirus (EAd) DNA prepared from infected bovine kidney (MDBK) cells contained additional sequences of about 100 to ‘700 bp at the left-hand end of the genome. These aberrant viral genomes were produced even after the first passage of the wild type EAd in MDBK cells and their relative amounts did not change significantly during serial passage. The left terminal fragments of two defective viral DNAs were cloned into the plasmid vector pBR322 and the nucleotide sequences of their terminal regions were analyzed. The data indicate that one viral DNA contained a duplication of the inverted terminal repetition (ITR) and the other contained 2’70 bp of additional sequences derived from the right-terminal region of EAd genome added to the left-terminal, ITR. While the former DNA was packaged into virions, the latter was not, presumably due to the alteration of the distance from the left terminus to the putative DNA packaging signal, reported to be located between 290 and 390 bp (Hammarskjold and Winberg, 1980). The possible mechanism for the generation of these defective DNAs is discussed. 0 1986 Academic PMS. IDC. INTRODUCTION
at each terminus. Electron microscopic analysis of viral replicative intermediates isolated from cells infected with adenovirus type 2 revealed two basic types of intermediates. Type I molecules consisted of unit length linear duplexes with one or more single-stranded branches. Type II replicating molecules consisted of unit length, unbranched linear molecules that contained both double-stranded and singlestranded regions (Ellens et al, 1974; Lechner and Kelly, 1977). Type I intermediates represent the displacement mechanism for the parental strand and type II intermediates represent the replication process of the displaced parental strand which could arise from the putative panhandle intermediate (Daniell, 1976). Evidence for the interaction of the two molecular ends during DNA replication was provided by Stow (1982). Deletions of different lengths, shorter than the unit length of the ITR, introduced at the left end were repaired during replication of the viral genome with an intact ITR at the
Adenovirus (Ad) genomes contain linear double-stranded DNA with a molecular weight of 20-30 X lo6 (Green and Pina, 1964; van der Eb and van Kesteren, 1966; Norrby et aL, 1976). Ad DNA possesses a unique inverted terminal repetition (ITR) of the type abc - - -. c’b’a’, which permits the formation of single-stranded circular DNA after denaturation and reannealing of the two strands (Garon et aL, 1972; Wolfson and Dressler, 1972). An additional novel feature of Ad genome is that the 5’ end of each DNA strand is covalently linked to a protein of 55 kDa in size (Robinson et aL, 1973; Sharp et al, 1976; Rekosh et aL, 1977; Padmanabhan and Padmanabhan, 1977; Shinagawa et aL, 1979). Although the biological function of the ITR is not completely understood, it plays an important role in DNA replication in having an origin ‘To whom dressed.
requests
for
reprints
0042-6822186 $3.00 Copyright All rights
0 1986 by Academic Press, Inc. of reproduction in any form reserved.
should
be ad-
66
PACKAGING-DEFECTIVE
EAd
DNA
67
right end. Deletions larger than the length genome. In this paper, we describe the characterization of these variant moleof ITR proved to be defective in replication. Hammarskjold and Winberg (1980) re- cules, which form a different class from ported that a DNA sequence located be- those characterized in other laboratories in which terminal rearrangements and dutween 290 and 390 bp from the left terminus of Ad16 contained recognition signals plications occurred at the right end. We for encapsidation of the viral DNA. The find that the variant EAd DNA molecules location was determined by analyzing a se- which arose by the acquisition of about 100 ries of spontaneous mutants of Ad16 which bp at the left end was packaged into viarose by serial passage in HeLa cells and rions, whereas those which had 270 bp or carried duplications of 200 to 500 bp of left more were not. terminal sequences at the right end of the genome. A duplication of about 390 bp seMATERIALS AND METHODS quences from the left end at the right end enabled the subgenomic defective viral Virus and cells. T-l strain of EAd type 1 DNAs containing both molecular ends to be encapsidated. However, the incomplete (Kamada et al, 1977) was obtained from M. virus particles containing the left end se- Kamada, Equine Research Institute, Japan quences, of less than 290 bp duplicated at Racing Association. The virus had been right end did not promote encapsidation of passaged three times in MDCK cells, and right-terminal sequences. six times in MDBK cells at the time it was Serial passage of Ad12 in Vero cells of used in this study. MDCK and MDBK cells African green monkey kidney cells (Wer- were grown in Eagle’s minimum essential ner and zur Hausen, 1978) resulted in the medium (MEM) containing 5% calf serum. insertions of 190-280 bp at the right end After infection with the virus, the cells and deletions of 70 bp at the left end of the were maintained in MEM containing 1% genome. These variants were selected be- fetal calf serum. Preparation of viral DNA. The cell culcause of their efficient growth properties in Vero cells. Similar alterations of the tures on monolayer were prepared in 55 viral genome were observed when the virus X 105 mm bottles and infected with the was serially passaged in human cervical EAd stock. The viral DNA was extracted carcinoma cells (Kruczek et ah, 1981) and from the purified virions as described by in human melanoma cells (Schwarz et al, Green and Pina (1964). Isolation of viral DNA from the infected cells was according 1982). These cell lines do not generally support an efficient growth of the virus. to Shinagawa et al. (198313). Enzymes. Restriction endonucleases However, serial passage in these cell lines resulted in the generation of these host BumHI, EcoRI, HueIII, HincII, H&fI, P&I, range mutants with duplication of the ITR SuZI, and SmaI, T4 DNA polymerase, T4 and its adjacent sequences of the left end polynucleotide kinase, and T4 DNA ligase were purchased from Takara-Shuzo (Kyoto, at the right end of the viral genome. Equine Ad (EAd) propagates efficiently Japan) and used as specified by the supin the cells derived from the natural host plier. Bacterial alkaline phosphatase was such as primary or low-passaged equine from Sigma (St. Louis, MO.) and was used kidney cells. It was also found that EAd in the presence of 0.1% SDS (Shinagawa was able to grow efficiently in canine kid- and Padmanabhan, 1979). Cloning of EAd Since a plaque assay for ney cells (MDCK) and bovine kidney cells (MDBK). When the EAd DNA prepared EAd in MDCK or MDBK cells is not yet from stock virus propagated in MDBK cells successfully developed, limiting dilution was analyzed, we discovered the presence technique was employed for cloning of EAd of altered viral DNAs in addition to the and its variants. Monolayers of MDCK and MDBK cells in five culture tubes were inwild type molecules and these variants seemed to arise from the acquisition of ad- fected with 2 ml of EAd viral stock serially ditional sequences at the left end of the diluted from 10e5 to lo-* with MEM con-
68
ISHIYAMA
taining 1% fetal calf serum. The infected cells showing the cytopathic effect using the most or the second most serially diluted virus were frozen and thawed to prepare the virus stock. The lysates were used to infect bottle cultures to obtain sufficient viral DNA for restriction endonuclease analysis and for molecular cloning. Physical mapping of EAd DNA. Restriction endonuclease cleavage maps of EAd DNA for BamHI, EcoRI, and Hind111 were constructed using the combination of partial digestion and the conventional double digestion methods. The terminal fragments were identified using the EAd DNA labeled at the 3’ end with 9 (Challberg and Englund, 1980). By convention, the A-T rich terminal fragment is designated as the right terminal fragment (Doerfler and Kleinschmidt, 1970; Mulder et al, 1974; Anonymous, 1977). In order to identify the A-T rich end fragment, Hind111 and BamHI digests of EAd was melted in 50% formamide for 15 min at different temperatures (49-54’) and then quickly chilled in ice and electrophoresed on an agarose gel (Zsak and Kisary, 1981). The terminal fragment that melts at a lower temperature than the other is designated as the right-terminal fragment. Southern hybridization. DNA fragments fractionated on an agarose gel by electrophoresis were transferred to nitrocellulose filters according to the method of Southern (1975). The probe for hybridization was prepared by digesting the plasmids containing BamHI-D or -D3 with SalI. The SalI site was filled in by repair synthesis catalyzed by DNA polymerase (Klenow fragment) in the presence of [a-32P]dCTP and unlabeled dTTP. The labeled DNA was treated with BamHI for pOHEAdBD or HpaII for pOHEAdBD3 plasmid. The filters containing DNA were hybridized to a =P-labeled probe at 40” for 20 hr, washed, dried, and exposed to x-ray films as described earlier (Fujinaga et al, 1979). The 32P-labeled probes were prepared by nick translation (Mackey et al, 1977). Molecular cloning of BamHI-terminal fragments of EAd The BamHI terminal fragments (D, Dl, and D3 in Fig. 2) of EAd DNA, prepared from infected MDBK cells
ET AL.
were purified by electrophoresis on agarose gels and treated with 0.5 M piperidine at 37” for 3 hr (Stillman et aL, 1982) to remove the residual peptides covalently bound to the original 5’ end of EAd DNA (Roninson and Padmanabhan, 1980). The DNA fragments were renatured in 1 M NaCl in 10 mm Tris-HCl, pH 7.5, and 1 mM EDTA at 65” for 6 hr. The vector plasmid, pBR322 DNA, was linearized by digestion with SalI and the 5’-protruding ends were made blunt by treatment with T4 DNA polymerase in the presence of 4 deoxynucleoside triphosphates. The DNA fragments were subsequently digested with BamHI and the 5’ ends were dephosphorylated by treatment with bacterial alkaline phosphatase. After three extractions with buffer-saturated phenol, the DNA was precipitated by the addition of two volumes of ethanol. The vector plasmid and the BamHI terminal fragments were mixed and ligated using T4 DNA ligase at 16” for 15 hr (Maniatis et al, 1982). For transformation, the Escherichia coli strain SGll, a ret A mutant of SG8 (Shinagawa et al, 1982) was used as the host (Cohen et al, 1973). Transformants were selected on heart infusion agar plates containing 25 pg/ml of ampicillin. The plasmid DNAs were isolated and analyzed by digestion with restriction endonucleases, BamHI and SalI. DNA sequence analysis. The plasmid DNA containing an EAd-BamHI terminal fragment (D, Dl, or D3) was digested with SalI and treated with bacterial alkaline phosphatase. The DNA was labeled at the 3’ end using DNA polymerase (Klenow) and [cy-32P]dTTP and digested with PstI. The two labeled fragments were fractionated by electrophoresis on an agarose gel. The labeled terminal fragment was subjected to the chemical degradation procedure of Maxam and Gilbert (1977) for DNA sequence analysis. RESULTS
Characterization of variant EAds possessing elongated terminal BamHI-D fragmerits. We discovered that EAd DNA prepared from the virions propagated in MDBK cells contained defective DNA mol-
PACKAGING-DEFECTIVE
ecules resulting from insertions of DNA sequences at the left terminus as shown in Fig. 1A (BarnHI-Dl). However, when the DNAs from 208 viral isolates from MDCK cells were examined by BumHI digestion, the patterns were all identical to each other in having only five fragments as shown in Fig. lB-a and the BumHI-Dl was not present. When the DNA from MDBK cells, in-
A
EAd
fected by 45 isolates of viral stocks, were analyzed by BumHI digestion, the pattern showed the presence of the terminal BumHI-D fragment and, in addition, the elongated fragments, Dl, D2, D3, and so on up to at least D’7 (Fig. lB-b). The sizes of these terminal fragments, BumHI-D, -Dl, -D2, and -D3 were estimated to be about 17’70,1870,1920, and 2030 bp, respectively.
B
C a' b'
a b
VA IL B ‘-c
69
DNA
o
b
A-
/A
BC-
=;
FIG. 1. Restriction endonuclease BamHI cleavage pattern of EAd DNA. (A) EAd DNA, prepared from purified virions which were propagated in MDBK cells, was digested with BarnHI and electrophoresed on an agarose gel. (B) EAd was cloned by limiting dilution technique in MDCK or MDBK cells as described in the text. The isolates were grown in the respective cells. The viral DNA extracted from the infected cells was digested with BamHI and electrophoresed on an agarose gel. DNA prepared from MDCK cells is shown in B-a, and from MDBK cells in B-b. The DNA fragments in the gel were examined by the Southern hybridization (1975) using q-labeled BarnHI-D of EAd DNA as a probe (B-a’, B-b’). (C) EAd DNAs prepared from virions and from infected cells were compared. MDBK cells were infected with EAd virus stock which was grown in MDBK cells. The infected cells were divided into two portions. The viral DNA was prepared directly from one portion of the infected cells, while the virions were purified from the other before the extraction of DNA. Both DNAs were digested with BarnHI and analyzed by agarose gel electrophoresis. Lane a contained DNA from the purified virions, and b, DNA from the infected cells.
70
ISHIYAMA
To confirm that the elongated fragments were derived from BumHI-D, the sequence homology between these fragments and BumHI-D was examined by Southern hybridization using 32P-labeled BamHI-D as a probe. As shown in Fig. lB-b’, the probe hydridized to all of these elongated DNA fragments, indicating that these fragments were derived from BamHI-D. Interestingly, the hybridization of the probe to the DNA isolated from MDCK infected cells showed only the presence of the terminal BumHI-D fragments (Fig. lB-a’). In Fig. lC, the EAd DNA was prepared from one portion of the infected cells, while the virions were purified from the other portion and subsequently, the viral DNA was extracted. Both DNAs were digested with BamHI and the digests were analyzed by electrophoresis on agarose gel. The results were the same as shown in Figs. 1A and B. These results indicate that the variant EAd DNAs containing BamHI-D2, -D3, and further elongated fragments were deficient in DNA packaging. E#ect of serial passage of EAd wild type virus in MDBK cells on the generation of elongated DNA molecules. We investigated the effect of serial passage of EAd in MDBK cells on the generation of defective genomes. Wild type EAd stock, obtained by cloning in MDCK cells was serially passaged in MDBK cells and the viral DNAs synthesized in the infected cells after second, third, sixth, and ninth passage were analyzed by BamHI digestion. As shown in Fig. 2A, BarnHI-Dl, -D2, -D3, and the further elongated fragments already appeared after second passage and the relative amounts seemed to be unaffected by subsequent serial passage of the virus. The results were reproducible as shown in Figs. 2B and C in which the virus was passaged three times. It is interesting to note that the replication of the virus in MDBK cells even during the first passage (Figs. 2B and C) produced the elongated DNA molecules. Location of BamHI-D fragment in the EAd genome. Although it was reported that BamHI-D fragment of EAd DNA was one of the terminal fragments (Shinagawa et al., 1983a), its location at the right or the
ET AL. A
B
C
FIG. 2. BumHI digestion of EAd DNA after serial passage in MDBK cells. Wild type EAd was serially paasaged in MDBK cells. The viral DNA was prepared from infected cells at the indicated passage level, digested with BumHI, and electrophoresed on an agarosegel. Passage numbers are shown above each lane. (B) and (C) represent duplicate experiments.
left molecular end of the genome was unknown. In order to determine the orientation of the BamHI terminal fragments of EAd, we made use of the observation of earlier investigators that the A-T rich half of the viral DNA is positioned to the right of the adenovirus genome in general (Doerfler and Kleinschmidt, 1970; Mulder et al, 1974). It would mean that the end fragment that melted at a lower temperature than the other was considered to be located at the right end of the DNA molecule (Anonymous, 1977). To find the A-T rich half of EAd DNA, BumHI digest of the DNA was heated in 50% formamide at various temperatures between 45 and 52”, chilled in ice rapidly, and then electrophoresed on an agarose gel. As shown in Fig. 3B, one of the terminal fragments, BumHIB melted at 51°, but the BamHI-D fragment was intact at that temperature, but melted at 52”, indicating that BumHI-B is A-T rich and hence, by convention, is located at the right end. Similar results were obtained when Hind111 digest was exam-
PACKAGING-DEFECTIVE
EAd
DNA
A DE
1 11
C
4:9-7.0 E
I
25.0 B
11 6.‘2
A
A
1 34.6’
I
I
c
BamHI
7911 A
0 53.;
II
I 51.4
,D
10
I I
G42’7C
37.1
B
EcoRI 60.2 E
F
B
100
I II 6f.6
71.9-77.2
Hind111
,E
10
FIG. 3. Restriction endonuclease cleavage maps of EAd DNA. (A) Cleavage maps of EAd DNA for BarnHI, EcoRI, and Hind111 were constructed as in the text. (B) To define the left-to-right orientation of the cleavage maps, BarnHI digest of EAd DNA in 50% formamide was heated at the indicated temperature for 15 min, chilled quickly, and electrophoresed on an agarose gel (Zsak and Kisary, 1981).
ined. HindIII-B fragment melted at 49”, whereas HindIII-A was intact at that temperature indicating that HindIII-B is A-T rich (data not shown). Therefore, BumHID fragment, which is a subfragment of HindIII-A is located at the left end and the elongation of EAd DNA occurred at the left end of the genome. Physical mapping and DNA sequence analysis of the elongated segments of EAd. In order to study these variants, arising due to the acquisition of additional sequences at the left end, at the molecular level, we cloned the BamHI-D, -Dl, and -D3 fragments into pBR322 vector as described under Materials and Methods. The recombinant plasmids containing the respective terminal fragments, -D, -Dl, and -D3 were designated as pOHEAdBD, pOHEAdBD1, and pOHEAdBD3, respectively (Fig. 4A). Since the regeneration of the Sal1 site was used for screening the re-
combinants resulting from the blunt end ligation of the left terminal BamHI fragments of EAd variants to the SalI-cut and filled-in site of the vector, only those molecules containing the dC at the 5’ left terminus such as Dl and D3 were cloned. D2 variant was not seen among the recombinants. The elongated segments of the left terminal BumHI-D, BamHI-Dl, and -D3 were analyzed by mapping the restriction endonuclease cleavage sites on the cloned fragments, pOHEAdBD, pOHEAdBD1, and pOHEAdBD3 for HincII, PstI, S’aZI, and SmaI (Fig. 4A). Among the enzymes used, H&c11 gave rise to the smallest fragment, 237 bp long that contained the original left end of EAd DNA in pOHEAdBD. In pOHEAdBD1 and pOHEAdBD3, the HincII fragments were 338 and 506 bp, respectively, in length due to the addition of DNA sequences at the left end during
ISHIYAMA
72
ET AL.
A POHEAdBD
Hincll cmtl
I
Him11 BMlHI
B D
5'-CATCATCMT MTATACAGG ACACACGGGCATGGGGCCM GAAAGGGGAGGAGTTGAGW GWGCGGGAG GCGGGGGCGG AGGCGGGGCG GCGGGCGG~ G ITR
Dl
amCTT---
5'4ATCATCAAT MTATACAGG ACACACGGGCATGGGGCCAAGMAGGGGAGGAGTTGAGGCGTGGCGGGAG TCATCAA IAATATACAG GACACACFCGCATGGGGGCA GCGGGGGCGG AGGCGGGGCG G4Y3XCGGG.A AGAAAGGGA GGAGTTGAGGCGTGGCGGGA GAGGCGGGGC GGCGGGCGGG A AAAAATTC---
D3
5'-cG4XCTCCCI TATGCGACTC CT~~~M~~C~~AGTAGGTTGILGGCCGTTGACCICCCCCG CCGCMGGAA TGGTGCATGCAAGGAGATGGCGGCCAACAGTCCCCCGGCCACGGGGCCTGCCACCATACC CACWXAAA CAAGCGCEA TGAGCCCGM GTGGCGAGCC CCXST’IWC CATCGGIGAT GTCGGCGATA TAGGCGCCAGCAACCGCACC TGTGGCGCCGGTGATGCCGGCACGATGCGTTCGGCTAG4G CAICAATAAT ATACAGGA--
hTR
FIG. 4. Restriction site mapping and nucleotide sequence analysis of the variants at the left end. (A) Restriction endonuclease cleavage maps of the cloned left-terminal fragments, BarnHI-D, -Dl, and -D3 of EAd DNA. The thick lines indicate the regions derived from EAd DNA and the thin lines indicate the region from pBR322. Fine maps of the extreme left-terminal H&c11 fragments for Hue111cleavage sites are shown in the expanded regions under each restriction map. The numbers indicate length (bp) of the fragments. (B) Nucleotide sequences at the extreme left end region of the cloned EAd DNAs, BornHI-D, -Dl, and -D3. The nucleotide sequences were determined using the chemical method of Maxam and Gilbert (1977) as described under Materials and Methods. The end point of ITR sequences are shown with bracketed arrow. In Dl, the second copy of ITR starts at nucleotide No. 3 and ends with a dG instead of the normal dC residue. In D3, the extra sequences of 270 nucleotides are fused to the 4th nucleotide of ITR.
the growth of EAd in MDBK cells (Figs. 4A and B). The fine mapping of H&c11 fragments of pOHEAdBD, pOHEAdBD1,
and pOHEAdBD3 was carried out using Hue111 (Fig. 4A). The H&II fragments of 86 and 115 bp were conserved in both
PACKAGING-DEFECTIVE
pOHEAdBD1 and pOHEAdBD3 in the same orientation as in pOHEAdBD (Fig. 4A) indicating that the additional sequences are added to the 36-bp Hue111 fragment at the original left end of the genome. We determined the nucleotide sequences of the wild type and the variant DNAs at the original left end of EAd (from SalI site toward the H&c11 site in Fig. 4A) and the sequence data are presented in Fig. 4B. The length of ITR is 103 bp in EAd D in Fig. 4B. In Dl, the sequence data showed that there is a tandem duplication of ITR at the left end except that there is a deletion of two nucleotides from the second copy of ITR. However, in D3, three nucleotides from the left end ITR of wild type were replaced by a stretch of 270 nucleotides and the nucleotide sequence data revealed that, unlike the case of Dl, the region was not a tandem duplication of ITR region. Nucleotide sequence analysis at the right terminal end using BumHI-B fragment of EAd DNA prepared from the infected MDBK cells indicated that there was no alteration to the EAd genome at the right terminus. Origin of the additional sequences at the left end of pOHEAdBD3. In order to determine whether the sequences added to the left end of the viral DNA in the molecular clone, pOHEAdBD3 is of cellular or viral origin, the DNA from the virions isolated from the infected MDCK cells was digested with BamHI (Fig. 5) and fractionated by electrophoresis on an agarose gel. The DNA fragments were denatured, neutralized, and transferred to a nitrocellulose filter. The probe DNA to be used for Southern hybridization should consist of sequences only from the newly acquired region at the left-hand end and not include ITR sequences which are common to both molecular ends. The probe was prepared from pOHEAdBD3 as described under Materials and Methods using HpuII as the second enzyme, which cleaves at a site 117 nucleotides away from the left-hand terminus of the variant, D3, (see Figs. 4A and B) within the newly acquired sequences, but 150 bp upstream from the ITR region. The ter-
EAd
DNA
73
FIG. 5. The origin of the extra sequences added to the left end of EAd. EAd DNA, purified from the virions isolated from the infected MDCK cells was digested with BamHI and electrophoresed on an agarose gel. The DNA fragments were transferred to a sheet of nitrocellulose filter and hybridized with the 120bp-long HpoII fragment as the probe which represents the left terminal region of the EAdBD3 DNA which contains the extra sequences added during replication. The labeled HpuII was isolated by digesting the plasmid pOHEAdBD3 with SalI and repairing the cohesive ends with DNA polymerase (Klenow fragment) in the presence of [o-“PJdCTP and unlabeled d’ITP. The labeled DNA was digested with H&I and fractionated by agarose gel electrophoresis. The hybridization was carried out according to Southern (1975). The internal BarnHI-E fragment ran out of the gel in this experiment.
minally labeled HpaII fragment was used as a probe for Southern hybridization of a BamHI digest of EAd DNA and the results are shown in Fig. 5. BumHI-B fragment hybridized with the probe indicating that these additional sequences are of viral origin and are derived from the right end of the viral genome.
ISHIYAMA
74
ET AL.
and different segments of the adjacent unique sequences. Growth of EAd in We have shown that EAd prepared di- MDBK cells differ from the growth of Ad12 rectly from the infected MDBK cells con- in Vero cells or human melanoma cells in tained defective DNA molecules in addition that the variants were generated even at to the wild type. These variants arose due the first passage due to the insertion of seto a duplication of ITR (EAdBDl) or due quences at the left end and the relative to the addition of sequences which showed amounts of these variants and the wild no homology to ITR (EAdBD3). Of all these type seemed to remain constant during sevariants, only EAdBDl was packaged into rial passage (Fig. 2). This observation inthe virions, as the increase in the length dicates that these variants do not seem to of the genome at the left end is only 101 have any improved growth properties, unbp. The other defective DNAs are found like Ad12 variants (Werner and zur Hauonly in the infected MDBK cells and are sen, 1978; Schwarz et al, 1982). In addition probably deficient in DNA packaging. The to the Ad12 variants, others that were deficiency of these molecules (EAdBD2, characterized (Hammerskjold and Win3, etc.) in DNA packaging is unlikely due berg, 1980; Garon et aL, 1982; Brusca and to their larger size of the genomes be- Chinnadurai, 1983) resulted from the recause the Ad capsid has the capacity to duplication of the left-end ITR and the adpackage Ad DNA which is up to 105% in jacent unique sequences at the right end. length (Hammarskjold and Winberg, 1980; However, when EAd replicates in MDBK Schwarz et al, 1982; Garon et al., 1982; cells, reduplications of ITR (EAdBDl) as Brusca and Chinnadurai, 1983). Daniel1 well as other sequences which map within (1976) and Tibbetts (19’7’7)reported that the right-hand terminal BumHI-B fragsequences from the left end of the genome ment (as in the case of EAdBD3) occurred are preferentially incorporated into in- at the left end of the genome. The reducomplete particles of Ad3 and Ad7. Anal- plication of 101 bp of ITR at the left end ysis of the DNA isolated from the incom- of EAd did not block the viral DNA to be plete particles prepared from the vari- packaged into the virions. However, the ants of Ad16 in which the sequences unique variant with a longer insertion of 270 bp to the left end were added to the right at the left end was not packaged. It is posend enabled Hammarskjold and Winberg sible that the distance from the left end of (1980) to conclude that there is a recogni- the genome to the DNA packaging signal tion signal for the encapsidation of viral can be varied up to approximately 101 bp DNA located about 390-400 bp from the for the DNA to get packaged but not up to left end and the reduplication of this signal 270 bp. at the right end enabled the packaging of The mechanism by which these aberrant the right terminal DNA into incomplete molecules of EAd are generated when EAd particles. undergoes replication in MDBK cells but A different type of variant was charac- not in MDCK cells is unknown at present. terized by Schwarz et al. (1982) who re- Ad DNA replicates in permissive cells ported that after serial passage of Ad12 in through a strand displacement model as cells of human melanoma cell line, variants originally proposed by J. Sambrook (see with enhanced growth properties were iso- Daniell, 1976). The elongation of the lated which carry additional sequences of daughter strands toward the opposite end regularly increasing size at the right end results in the progressive displacement of of the genome. Sequence analysis of these one of the parental strands (which were variants indicated that these extra se- seen as type I molecules under the elecquences were of viral origin. They origitron microscope by Lechner and Kelly, nated from either the extreme right or left 197’7). The self-complementary terminal end of the genome consisting of the ITR sequences of the completely displaced DISCUSSION
PACKAGING-DEFECTIVE
strands, probably hybridizes to each other to form the putative “panhandle” intermediate (Daniell, 1976) before replication to form the second daughter molecule in a normal permissive cell. The experimental evidence for the interaction of the two molecular ends during the replication of Ad DNA was reported by Stow (1982). It is interesting to note that replication of viral genome containing an additional 18 dG residues at the left end of Ad2 gave rise to the progeny viral genomes that contained perfectly normal ITR, suggesting that the interaction between the two ITRs and subsequent cellular repair might be involved in the restoration of normal ends (Stow, 1982). The model proposed by Daniel1 (1976) suggests that the displaced strand is susceptible to breakage and its 3’ end could fold back on itself at a region of partial homology and continue its synthesis until it reaches the 5’ end. The double-stranded DNA molecule could then be formed via a type II mechanism. If this model is correct, the length of reduplication of DNA sequences from one terminus to the other would depend on the location of the foldback region for the DNA synthesis. This model would require, for example, the displaced parental 1 strand of EAd which probably lost two or three nucleotides from the left end ITR could then transiently hybridize to a region adjacent to the righthand ITR and undergo DNA synthesis to generate an ITR reduplication variant. These molecules have the potential to undergo homologous recombination between the internal ITR and the terminal ITR either on the same or on a different molecule. The variants, EAdBD2,3, etc., might be the products of such recombination events, although a rather high frequency is required to produce these molecules which seems rather unlikely. According to this model, the reduplication of DNA sequences is not expected to be a specific event at either terminus since both displaced parental strands (r and 1) would participate in the generation of these mutants.._.. However, - our studies indicated that the addition of extra
EAd
DNA
75
sequences seems to be specific to the left end. The reason for this specificity is not clear. Further work is necessary to map the precise location of the DNA sequences within the right-terminal fragment which were duplicated at the left end during replication. ACKNOWLEDGMENT This work was partly supported by a grant-in-aid for Cancer Research from the Ministry of Education, Science and Culture, Japan, and by Grant 56480067, from the same Ministry (to M.S.) and from the National Cancer Institute (CA 33099 to R.P.). We thank Linda Hicks for typing this manuscript. REFERENCES ANONYMOUS (19’77). Adenovirus strand nomenclature: A proposal. J. V+oL 22,830~831. BRUSCA, J. S., and CHINNADURAI, G. (1983). Structure and reversion of an adenovirus type 2 mutant containing a duplication of the left end of the genome at the right end. virology 129,381-392. CHALLBERG, M. D., and ENGLUND, P. T. (1980). Specific labeling of 3’ termini with T4 DNA polymerase. In “Methods in Enzymology” (L. Grossman and K. Moldave, eds.), Vol. 65, pp. 39-43. Academic Press, Orlando, Fla. COHEN, S. N., CHANG, A. C. Y., BOYER, H. W., and HELLING, R. B. (1973). Construction of biologically functional bacterial plasmid in vitro. Proc Nat1 Acad Sci USA 79,3240-3244. DANIELL,, E. (1976). Genome structure of incomplete particles of adenovirus. J. viral 19.685-708. DOERFLER, W., and KLEINSCHMIDT (1970). Denaturation pattern of the DNA of adenovirus type 2 as determined by electron microscopy. J. Mel Bill 50, 579-593. ELLENS, D. J., SUSSENBACH, J. S., and JANSZ, H. S. (1974). Studies of the mechanism of replication of adenovirus DNA. III. Electron microscopy of replicating DNA. V+oEogy 61,427-442. FUJINAGA, K., SAWADA, Y., and SEKIKAWA, K. (1979). Three different classes of human adenovirus transforming DNA sequence. virdogy 93,578-581. GARON, C. F,, BERRY, K. W., and ROSE, J. A. (1972). A unique form of terminal redundancy in adenovirus DNA molecules. PTIX. Nat1 Acud Sci USA 69,24552459. GREEN, M., and PINA, M. (1964). Biochemical studies on adenovirus multiplication. VI. Properties of highly purified tumorigenic human adenoviruses and their DNA’s. Proc NatL Acad Sci USA 51.12511259.
76
ISHIYAMA
HAMMARSKJOLD,M. L., and WINBERG,G. (1980). Encapsidation of adenovirus 16 DNA is directed by a small DNA sequence at the left end of the genome. Cell 20,787-795. KAMADA, M., AKIYAMA, Y., SATO,K., and KODERA,S. (1977). Isolation of adenovirus from adult thoroughbred horses. Japan. J. Vet. Sci. 39,661-664. KRUCZE~,I., SCHWARZ,E., and ZURHAUSEN,H. (1981). Mutants of adenovirus type 12 after adaptation of growth in tumor cell lines. II. Reproducible acquisition of additional sequences after adaptation to a human cervical cancer line. Znt. J. Cancer 27,139143. LECHNER, R. L., and KEUY, T. J., JR. (1977). The structure of replicating adenovirus 2 DNA molecules. Cell 12,1007-1020. MACKEY,J. K., BRACKMANN,K. H., GREEN,M. R., and GREEN,M. (1977). Preparation and characterization of highly radioactive in vitro labeled adenovirus DNA restriction fragments. Biochemistry 16,44784483. MANIATIS,T., FRITSCH,E. F., and SAMBROOK, J. (1982). “Molecular Cloning: A Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. MAXAM, A. M., and GILBERT, W. (1977). A new method for sequencing DNA. Proc. Nat1 Acnd Sci USA 74, 560-564. MULDER,C., ARRAND, J. R., DELIUS, H., KELLER, W., PETTERSSON, U., ROBERTS,R. J., and SHARP,P. A. (1974). Cleavage mapping of DNA from adenovirus type 2 and 5 by restriction endonucleases, EcoRI and HpaI. Cold Spring Harbor Symp. Quad Bid
ET AL. SCHWARZ,E., REINKE, C., YAMAMOTO,N., and ZUR HAUSEN,H. (1982).Terminal rearrangements in the genome of adenovirus type 12 mutants adapted to growth in two human tumor cell lines. virology 116, 284-296. SHARP,P. A., MOORE,C., and HAVERTY,J. (1976). The infectivity of adenovirus 5-DNA protein complex. Virology 75,442-456.
SHINAGAWA,M., ISHIYAMA, T., PADMANABHAN,R., FUJINAGA,K., KAMADA, M., and SATO, G. (1983a). Comparative sequence analysis of the inverted terminal repetition in the genomes of animal and avian adenoviruses. Virology 125,4X-495. SHINAGAWA,M., MAKINO, S., HIRATO, T., ISHIGURO, N., and SATO, G. (1982). Comparison of DNA sequences required for the function of citrate utilization among different citrate utilization plasmids. J Bacterial. 151,1046-1050. SHINAGAWA,M., MATSUDA, A., ISHIYAMA, T., GOTO, H., and SATO,G. (1983b). A rapid and simple method for preparation of adenovirus DNA from infected cells. Mic~obiol. ZmmunoL 27,817-822. SHINAGAWA,M., and PADMANABHAN,R. (1979). Inhibition of a nuclease contaminant in the commercial preparations of Escherichia coli alkaline phosphatase. Anal. B&hem. 95,458-464. SHINAGAWA,M., PADMANABHAN,R. V., and PADMANABHAN,R. (1979). Isolation and characterization of a homogeneous DNA-protein complex from adenovirus type 2 virion. J. Gen. Viral. 45, 519-525. SOUTHERN, E. M. (1975).Detection of specific sequences among DNA fragments separated by gel electrophoresis. .Z Mol. Biol 98,503-517. STILLMAN,B. W., TOPP,W. C., and ENGLER,J. A. (1982). 39,397-400. Conserved sequences at the origin of adenovirus NORRBY,E., BARTHA, A., BOULANGER,P., DREIZIN, DNA replication. J. viral 44,530-537. R. S., GINSBERG,H. S., KALTER, S. S., KAWAMURA, STOW,N. D. (1982). The infectivity of adenovirus geH., ROWE,W. P., RUSSELL,W. C., SCHLESINGER, R. nomes lacking DNA sequences from their left-hand W., and WIGAND, R. (1976). Adenoviridae. Inter&termini. Nucleic Acids Rex 10, 5105-5119. rologv 7,117-125. TIBBETTS,C. (1977). Viral DNA sequences from inPADMANABHAN,R., and PADMANABHAN,R. V. (1977). complete particles of human adenovirus type 7. Cell Specific interaction of protein(s) at or near the ter12,243-249. mini of adenovirus 2 DNA. B&hem Biophys. Res. VAN DER EB, A. J., and VAN KESTEREN,L. W. (1966). Cmmun 75,955-%X. Structure and molecular weight of the DNA of adREKOSH,D. M. K., RUSSELL,W. C., BELLETT,A. J. D., enovirus type 5. B&him Biophys. Acta 129,441. and ROBINSON,A. J. (1977). Identification of a proWERNER,G., and ZURHAUSEN,H. (1978). Deletion and tein linked to the ends of adenovirus DNA. CelE11, insertion in adenovirus type 12 DNA after viral 283-295. replication in Vero cells. Virology 86, 66-77. ROBINSON,A. J., YOUNGHUSBAND, H. B., and BELLET~, WOLFSON,J., and DRESSLER,D. (1972). Adenovirus 2 A. J. D. (1973). A circular DNA-protein complex DNA contains an inverted terminal repetition. Pnw: from adenoviruses. Virology 56,54-69. Natl. Acad Sci. USA 69,3054-3057. RONINSON,I., and PADMANABHAN,R. (1980). Studies ZSAK, L., and KISARY, J. (1981). Studies on egg drop on the nature of the linkage between the terminal syndrome (EDS) and chick embryo lethal orphan protein and the adenovirus DNA. B&k Biophys. (CELO) avian adenovirus DNAs by restriction enRes. Commun 94,398-405. donucleases. J. Gen firol 57.87-95.