- Email: [email protected]
Tandemly repeated sequences are present at the ends of the DNA of raccoonpox virus
VIROLOGY
161,45-53
(1987)
Tandemly Repeated Sequences Are Present at the Ends of the DNA of Raccoonpox
Virus
BARBARA L. PARSONS AND DAVID J. PICKUP’ Department
of Microbiology
and Immunology,
Duke University
Received November
Medical
Center, Durham, North Carolina 277 10
7, 1986; accepted June 10, 1987
The DNA of raccoonpox virus (RCN) has been characterized by restriction enzyme analysis. DNA hybridization studies showed that all Hindill fragments of the 215kbp RCN DNA share some nucleotide sequence similarity with fragments of the DNA of cowpox virus (CPV). This information was used to construct a HindIll restriction map of the RCN DNA. The nucleotide sequence of the 2.2-kbp Sa/l end fragment of the RCN DNA has been determined from a cloned copy of the Hindlll 0 fragment. Of this 2.2-kbp region 75% consists of short, tandemly repeated sequences. It does not contain any open reading frames capable of encoding polypeptide chains of more than 62 amino acids. There are six related types of repeated sequence, and these are arranged into two separate sets, each flanked by nonrepeated sequences. The nucleotide sequences of both repeated and nonrepeated sequences within this Sal1 fragment are extremely similar to those of the Se/l-generated end fragments of the DNAs of CPV and vaccinia virus. The arrangements of the repeated and nonrepeated sequences are also similar in the DNAs of these three viruses. In contrast, the remainder of the RCN DNA is markedly different from the DNAs of other orthopoxviruses. The high degree of similarity between the ends of the RCN DNA and the ends of the other orthopoxvirus DNAs suggests that the complex arrays of repeated and nonrepeated sequences have been conserved because they have a role in virus tnukipkatiOn.
0 1987 Academic
Press,
Inc.
be present in the genomes of the other orthopoxviruses. Previous studies showed that the ends of the DNA of cowpox virus (CPV) contains two sets of short, tandemly repeated sequences; moreover, these repeated sequences and their flanking nonrepeated sequences are almost identical to the corresponding elements in the DNA of W (Baroudy and Moss, 1982; Baroudy et a/., 1982; Pickup et al., 1982). This similarity might reflect the conservation of these elements, or, because these viruses are closely related, it might simply reflect both a lack of sequence alteration and a lack of pressure to eliminate sequences that are of no value to the virus. A comparison of orthopoxvirus DNAs that are largely dissimilar would provide a better indication of regions that have been conserved. Most of the orthopoxvirus DNAs are remarkably similar with respect to the positions of their HindIll cleavage sites (Mackett and Archard, 1979); the DNA of raccoonpox virus (RCN) is the only known exception (Esposito and Knight, 1985). More extensive differences between RCN DNA and the other orthopoxvirus DNAs were detected by DNA hybridization experiments; of the DNAs of 38 viruses representative of all types of orthopoxviruses, RCN DNA was least similar to the others (Esposito and Knight, 1985). The extent of the differences between the DNA of RCN and the DNAs of the other orthopoxviruses suggests that the sequences which remain common to these DNAs are providing some functions that are advantageous to the virus.
INTRODUCTION The genomes of the orthopoxviruses are linear double-stranded DNAs, most of which are between 150 and 220 kbp in length (Joklik, 1962; Geshelin and Berns, 1974; Gangemi and Sharp, 1976; Wittek et a/., 1977; Mijller et a/., 1977; Mackett and Archard, 1979; Moyer and Rothe, 1980; Esposito and Knight, 1985). The most detailed studies on the structure of the ends of these viral DNAs have been done on the DNA of vaccinia virus (W),2 which is the prototype of the Orthopoxvirus genus. The W DNA contains inverted terminal repeats (ITRs) that are about 10 kbp in length (Garon et al,, 1978; Wittek et a/., 1978). At the two ends of the VV DNA the two strands of the DNA duplex are joined in a loop (Geshelin and Berns, 1974; Baroudy et al., 1982). The terminal 125 bp of VV DNA appear to be required for the resolution of unit length genomes from concatemers of viral DNA (DeLange et a/., 1986; Merchlinsky and Moss, 1986). The 3.5-kbp region next to the end loop consists largely of two sets of short, tandemly repeated sequences (Wittek and Moss, 1980; Baroudy and Moss, 1982; Baroudy et al., 1982). The significance of these repeated sequences and their flanking sequences is not clear. If these structural features are important for viral multiplication then it is likely that similar features will ’ To whom requests for reprints should be addressed. ’ Abbreviations used: RCN, raccoonpox virus; CPV, cowpoxvirus; VV. vaccinia virus; SFV, Shape fibroma virus; kbp, kilobase pairs.
45
0042.6622/87 $3.00 CopyrIght All nghts
0 1997 by Academic Press, Inc. of reproduction in any form reserved.
PARSONS
46
This paper describes the further characterization of the DNA of raccoonpox virus. The major objective of this study was to identify any conserved sequences within the end regions of the viral DNA. MATERIALS
AND
METHODS
Virus and viral DNA RCN, strain V71-I-85A (kindly provided by Dr. J. J. Esposito, Centers for Disease Control, Atlanta), was grown in human tk-143 cells. RCN produces A-type inclusions (ATI) that contain mature virus particles (unpublished observations). Virus was obtained from ATls that had been purified and then solubilized according to the methods of Pate1 et a/. (1986). This virus stock was used to infect monolayers of 143 cells. Virus was isolated from single plaques, and then the same procedure was repeated to obtain three-times plaque-purified virus. CPV, strain Brighton red, was grown in the chorioallantoic membrane of chick embryos. W, strain WR, was grown in mouse L cells. The viruses were purified as described by Joklik (1962). DNA was isolated from virus particles as described by Nevins and Joklik (1977). DNA hybridization
experiments
Standard methods were used to detect the hybridization of DNA fragments. DNA hybridization probes were labeled with 32P by nick-translation (Rigby et a/., 1977). The transfer of DNA fragments from agarose gels to nylon membranes (Biodyne A, Pall Ultrafine Filtration, Glen Cove, NY) and the conditions for DNADNA hybridization were as described by Southern (1975). The DNAs were hybridized at 65” for 16 hr in 6X SSC (0.9 M sodium chloride and 0.09 M sodium citrate), 2X Denhardt’s (0.04% Ficoll, 0.04% bovine serum albumin, and 0.04% polyvinylpyrrolidone), 0.1% SDS, 1 mM EDTA, and 20 pg/ml denatured salmon sperm DNA. The membranes were washed with 2x SSC, 0.1% SDS at 65” and then dried. Labeled DNAs on the membranes were visualized by autoradiography. Construction of the HindIll map of RCN DNA
restriction
The HindIll restriction map of RCN DNA was deduced largely from the extent of hybridization between the HindIll fragments of RCN DNA and mapped fragments of either W DNA or CPV DNA. The hybridization probes that were used are listed in Table 2. The positions of some fragments (HindIll R and Q) were determined by purifying these fragments and then using them as the hybridization probes. The RCN HindIll R fragment was used as a probe against mapped C/al
AND
PICKUP
restriction fragments of CPV Pstl C fragment (Pickup et a/., 1984). The RCN HindIll Q fragment was used as a probe against the mapped EcoRl fragments of the VV HindIll D fragment (Weinrich et al., 1985). Identification of the map positions of HindIll R and Q also established the order of these fragments relative to their respective adjacent fragments HindIll J and 0. The positions of the HindIll fragments derived from the two ends of the RCN genome were determined by the procedure of Smith and Birnstiel(l976) as follows. Nick-translated DNA probes (~177 and ~744) that were specific for the end regions of the viral DNA, were hybridized to blots of the resolved products of partial HindIll digestions of RCN DNA. ~177 is a plasmid that contains the 11.8-kbp terminal EcoRl fragment of the CPV genome; it hybridizes with both terminal HindIll fragments of RCN DNA. ~744 is a plasmid that contains a 3.3-kbp Pstl -Sal 1 fragment from the internal portion of the inverted terminal repeat of the CPV genome (Pickup et al., 1984); it hybridizes with only one of the terminal HindIll fragments of RCN DNA (HindIll H). The positions of HindIll cleavage sites within about 25 kbp of each end of the RCN DNA were deduced from the sizes of the digestion products that contained either probe-detectable end fragment. The sizes of these products of partial digestion were estimated from their electrophoretic mobilities (in a 0.6% agarose gel) relative to those of DNAs of known lengths. EcoRl, HindIll, Sail, andXba1 fragments of phage X DNA were used as size standards; their lengths have been determined from the nucleotide sequence of the DNA (Daniels eta/., 1983). Similarly, the lengths of the HindIll fragments of RCN, CPV, and VV DNAs were estimated from their electrophoretic mobilities relative to those of the EcoRl and HindIll fragments of phage X DNA.
TABLE 1 LENGTHSOF THE HindIll RESTRICTIONFRAGMENTSOF RCN DNA Fragment A B C D Eb F G H I J K
Length” 22.3 20.8 18.7 15.8 14.4 13.1 11.6 8.9 8.4 8.1 7.4
Fragment L M N Ob P Q R Sb T U V
’ Lengths are given in kilobase pairs. HindIll restriction shorter than 1.7 kbp in length were not sized. *Two molar fragments.
Length 7.0 5.4 5.1 4.9 4.5 2.9 2.8 2.4 2.2 2.1 1.7 fragments
STRUCTURE
OF RACCOONPOX
47
VIRUS DNA
TABLE 2 HYBRIDIZATIONOF CPV AND W DNA FRAGMENTSTO HindIll FRAGMENTSOF RCN DNA Hindlll fragments
of RCN DNA
ABCDEFGHIJKLMNOPQRSTU 1. CPV DNA probes Hindlll I Hindlll L Kpnl C Kpnl D Kpn,l G Kpnl K Kpnl L Pstl A Pstl c Pstl E Pstl F Psfl G Pstl I Pstl J Pstl M Pstl 0 p177* p7446 2. VV DNA probes Hindlll D Hindlll E Hindlll H Hindlll I Hindlll J Hindlll M Hindlll N Hindlll 0 p730*
V
+ +
+
+
+
+ +
+ +
+
+ f
+ + +
+
+ +
+
+
+
+ +
+
+
+
NT”
+
+
+
+
+
+
+ +
+ +
+ +
+ t
+ +
+
+ +
+ + + +
+ + +
+ +
a Not tested. b Plasmids containing viral DNA fragments that have not been designated by a single letter: ~177 contains the 11.8-kbp terminal EcoRl fragment of the CPV genome, ~744 contains a 3.3-kbp Sal1 -Pstl fragment from the internal portion of the CPV inverted terminal repeat, and ~730 contains a 3.5-kbp Xmal-Hindlll fragment from the W Hindlll A fragment.
Cloning of the HindIll-generated fragment of RCN DNA
4.9-kbp terminal
This end fragment was cloned according to the methods described by Pickup et a/. (1982). The terminal loops of the RCN DNA were removed by digestion with nuclease Sl 175 pg DNA and 1100 units of nuclease Sl in 375 ~1 of 30 mM sodium acetate, pH 4.5/0.3 n/r NaCI/l mM ZnSOJ5% (v/v) glycerol (57.5” for 5 min)]. The DNA was repaired with Escherichia co/i DNA polymerase I (Klenow fragment), and EcoRl linkers (GGAATTCC) were ligated to it. After the ligation the DNA was digested with EcoRl and HindIll. The terminal 4.9-kbp HindIll fragment (containing a cleaved EcoRl
recognition site in place of the terminal loop) was purified by agarose gel electrophoresis. This was then inserted between the EcoRl and HindIll sites in the DNA of plasmid pUC9 (Messing and Vieira, 1982). Standard methods were used to map the positions of restriction enzyme recognition sites within the cloned fragment. Nucleotide
sequence
analysis
Restriction enzyme-generated fragments of the cloned DNA corresponding to the terminal Sal1 fragment of the viral DNA were cloned into phage Ml 3 vectors mp18 and mpl9 (Norrander et al., 1983). These
PARSONS
48 0
RCN
CPV
vv
)
D
Nc( JVTP
E U
E
Ial
C
PCflG
F Qt.4 J NN K
E
A
EOIGLJH
D
A
C
NMK
F
S
A
LF
D
G
C
I
SSNKH
s
LH
B m
FIG. 1. A comparison of the HindIll restriction maps of RCN, CPV, and W DNAs. The maps of CPV DNA (Mackett and Archard, 1979) and W DNA (Mackett and Archard 1979; DeFilippes, 1982) are aligned with the map of the RCN DNA at the cross-hybridizing regions corresponding to CPV Hindlll E, W HindlIt D, and RCN Hindlll I, 0, and Q.
fragments were products of DNA cleavage by the following enzymes: EcoRl, Taql, Hirtfl, Ddel, and Sail. The orientations of the inserted fragments within the phage DNA were determined as described by Barnes (1980). The method of Henikoff (1984) was used to produce targeted breakpoints within the 1275-bp long innermost set of short, directly repeated sequences. First, to make unidirectional deletions extending from the Sal1 site into the set of repeats, the replicative form DNA of an Ml 3 mp18 clone containing the terminal Sal1 fragment (fcoR-Sal1 fragment) was cut with Sal1 and Sphl. The DNA was digested with exonuclease III at 37”, and aliquots were removed at 20-set intervals in order to obtain DNAs that had progressively longer singlestranded regions. The exonuclease III was inactivated by incubation at 70” for 10 min after the addition of 3 vol of 0.2 M NaCI, 5 mM EDTA, pH 8.0. Residual singlestranded regions were removed by nuclease Sl digestion, followed by repair with DNA polymerase I (Klenow fragment). fcoR1 linkers were ligated to the ends of the DNAs, and after subsequent fcoR1 digestion, the fragments were resolved by agarose gel electrophoresis. The sizes of the fragments were estimated from their electrophoretic mobilities; fragments larger than 800 bp and differing by increments of about 150 bp, were isolated and inserted into phage vector Ml 3 mp19. A similar procedure was used to make unidirectional deletions extending from the Doe1 site (at nucleotide 484) through the set of direct repeats, toward the Sal 1 site. For this purpose the 1663-bp Ddel fragment containing the second set of repeats was isolated, its ends were repaired with DNA polymerase I (Klenow fragment), and then it was inserted into the Smal site of the DNA of Ml 3 mp18. A clone was selected in which the orientation of the direct repeats (with respect to the Sphl and Sal1 sites of the vector DNA) was reversed relative to that in the clone used to generate the other set of unidirectional deletions. Otherwise similar methodology was used except that here only
AND
PICKUP
fragments greater than 300 bp long were inserted into the DNA of Ml 3 mpl9. These M 13 clones, which contained overlapping deletions, were used to determine the sequence of both strands of the entire fcoR 1-Sal 1 fragment. Their nucleotide sequences were determined according to the procedures of Sanger et al. (1980) and Biggin et al. (1983). RESULTS Characterization
of the DNA of RCN
HindIll digestion of the RCN DNA produced 25 fragments that were greater than 1600 bp long. The sum of the estimated sizes of these fragments (HindIll AV; Table 1) suggests that the intact RCN DNA is about 2 15 kbp long. Esposito and Knight (1985) reported that several HindIll fragments of RCN DNA did not hybridize with various cloned fragments of the DNAs of monkeypox virus, vaccinia virus, and variola virus. Our hybridization analyses showed that RCN HindIll fragments A through V all hybridize with fragments of CPV DNA (though fragments D, M, and R hybridize only weakly). Therefore it appears that RCN DNA and CPV DNA, two of the largest orthopoxvirus DNAs, contain some similar sequences that are not contained within the genomes of some of the other ot-thopoxviruses. Table 2 summarizes the results of hybridizations between RCN HindIll fragments and fragments of the DNAs of CPV and W. Most of the DNA probes hybridized with at least two fragments of RCN DNA. This identified fragments that were adjacent in the RCN DNA, thereby permitting construction of most of the HindIll map of this DNA. The procedure of Smith and Birnstiel (1976) was used to confirm the positions of HindIll sites close to the ends of the viral DNA. The completed Hindlll restriction map of RCN DNA is shown in Fig. 1. For comparison it is shown alongside the HindIll maps of the DNAs of VV and CPV (Mackett and Archard, 1979; DeFilippes, 1982). W is the pro.
----A
-c ----
E
D
T
.
T 1
T
-c-
--c--c-
c---c--b
DS I,,,
500
lob0
1&o
2&Q
FIG. 2. Restriction map of the cloned terminal Sal1 fragment of RCN DNA. Restriction enzyme cleavage sites are abbreviated as follows: E, EcoRl ; T, Taql ; D, Ddel ; S, Sal1 ; 1, Hinfl The arrows correspond to the extent of the nucleotide sequence (5’ to 3’ in the direction of the arrow) obtained from each template used. This Sal1 fragment contains 2195 base pairs.
STRUCTURE
Jc
OF RACCOONPOX
VIRUS DNA
120 maAGAGAcGAAA~-AG~cA~~AAAGA-
loo
ACA~CC#ACATTTlTAG&ATT
AAGAGAl&h?TFWAGACA+AAGAGAAATA 300
260 ----l-Y
320 ~UAAACTTTlTTA~AC&T
TATlTTlT$T&IllTTATGACd+AGA
---Y--
160 C
GAAATA+TG
i--
lUA%TTCTTAGGTCUAG
49
f.GAbAGAGACGA$ GACGA&ACAT>GAGACAT
500 540 520 Tl%AltXTTTMAATATGGACTAGAAIATGTCnXGUAAbACT
920 940 960 900 TlTTTTTATCACTCCA~AGMAGAGAMIA lTTTT$GTAAMCTITTlTATGACTCCA~A (5) *l-~~-T-c 1000 1020 1040 1060 1080 GUAGAGACGAAACATTTTATlSAGACAT’jAGAAAGAGMAGAGACGAAACA ~GTAAAACTTlTTTATGACTCCA~AGAAAG 1100 1120 ~CGA$CATl-lTTAG~TACA~~GMAGAGAAATA 1160 :'2f""--?-
1200 mGAcGAAAcA
1140 1160 IlTlTTGUAAAC~A~ACTCCA~~GAGA~C <5> 1220 1240 1260 TAMA-AmCTCCAm,AGAAAGAGAMGAGACGA
T-F1200 1300 A$AllllTAGTGAGACAT2pAMGAUAGAGAWAACA vG=-c
1600 1700 -TA~TAAAACTl-lTTTATGACTCC I 1720 1740 1760 1780 la00 A~,4GWAGATGMA~~AG~CA~~WGAGAAAnrrt""""rr lTTTT$CTAAAACTTllTTAEACTCCAlT~GAAAGAGATG u-cTcY
w7
1640 --%---=~
1320 1340 lTlTllTAn;A~CCA~AGMAGAGAAAGAGA
1660
1820 "7
1840 1860 GAWAAACA~TAAMCTTlTlTATGACTCCATT
1900 ~~--=Y--7
1920 -cGAAAcA* 2000 Yet-y--=-
2060 lTnTAAAmcAAAccccIII
1080 7""
1960 1960 TMAACllTlTAlGACTCCAlT~GAAAGAGATGA&ACAT 2020
2100 UAGTCTCCCGTACAC
2040 2060 CtTGTTACl'TAlTW2AGA~TCGACTlTMTGCGAATACA 2120
2140 2160 CGTATGTCCCM~CCACAA~TCCIGAGGAT'2ACAC~
2180 6lTUGACACACGAnUSnT?GTCTCGIWAC FIG. 3. Nucleotide sequence of the cloned terminal Sal1 fragment of RCN DNA. Nonrepeated sequences (nucleotides l-l 13, 479-732, and 2009-2195) are underlined. The extent and type of each repeated unit (see also Fig. 4) is indicated beneath the sequence. Incomplete repeats are indicated by an asterisk. Nucleotide 1 is adjacent to the EcoRl linker.
totype of the orthopoxvirus genus. CPV appears to have the largest genome of any orthopoxvirus; we estimate that it is 230 kbp long. Although there is little similarity between the HindIll map of RCN DNA and the HindIll maps of the CPV and W DNAs, there is extensive sim-
ilarity between the nucleotide sequences of these three DNAs. In these maps, most of the cross-hybridizing Hindlll fragments of RCN DNA and CPV DNA would have aligned well but for a region within the RCN HindIll E fragment, which appears to lack about 14 kbp
PARSONS
50
EOPVnunma
In!.& 1
&)J&AGAAAQA~-c&-
--------m-
-mum
1
AGAAAGAG----AcGAAA~----------rTrnAGr6AGAcAn
3
AGAMGAG------A-CA
~CCMACATTlWlTETAAAACTllTtTATGA~CCAlT
5
mQ,AA--rA I
CQmal
12
llllll
II
E
---- m-elTll7TETAAAACln-PPTATGA~CCA~
1 Tl-lTTA
This arrangement suggested that sets of short tandemly repeated sequences containing the Hinfl site were present at the ends of RCN DNA. Nucleotide sequence of the terminal Sdl fragment
9
-~rM~~A------~r-C~A~~C~n
AGAAAGAG
PICKUP
2
‘CSCCAn
b
6
9
AND
CAlT
FIG. 4. Composition of the repeated units in RCN DNA. The sequences are aligned to show maximum similarity. Vertical bars indicated mismatches. Nucleotides common to repeats of all six types are shown in the bottom line.
of DNA that is present in the HindIll G/F region of CPV DNA. Characterization of a cloned HindIll-generated terminal fragment of RCN DNA The terminal, cross-linked HindIll-generated fragments of this RCN DNA are the H and 0 fragments (Esposito and Knight, 1985). An RCN /-findIll 0 fragment, containing an EcoRl linker in the place of the terminal hairpin loop, was isolated, and then inserted into the DNA of plasmid pUC9. To estimate how much of the terminal loop of the viral DNA had been removed by the nuclease Sl treatment, the sizes of the cloned and uncloned terminal fragments were compared by agarose gel electrophoresis. The resolved fragments were transferred to a nylon membrane, and a nicktranslated probe of the cloned terminal Sal1 fragment was used to detect the band corresponding to the terminal Doe1 fragment of the viral DNA. This analysis indicated that about 30 bp of the terminal loop had been removed by the nuclease Sl treatment. A 2.2 kbp EcoRl-Sal1 fragment corresponding to the Sal1 fragment at the end of the viral DNA was subcloned from the HindIll 0 fragment. A restriction map of this Sal1 fragment is shown in Fig. 2. The fragment contains cleavage sites of only a few different restriction enzymes. However, there are multiple Hinfl sites, most of which are spaced at intervals of less than 100 bp.
The strategy employed to determine the sequence of the terminal Sa/l fragment is depicted in Fig. 2. The nucleotide sequence of this fragment is shown in Fig. 3. It does not contain any long open reading frames; its largest open reading frame (nucleotides 656-841; see Fig. 3) could encode a polypeptide chain of only 62 amino acids. The sequence contains three regions (nucleotides l-l 13,479-732, and 2009-2195) whose sequences are not repeated elsewhere within the Sal1 fragment. These three regions of nonrepeated sequences (designated NRl , NR2, and NR3) are separated by two sets of short, tandemly repeated sequences. The repeated sequences make up about 759/o of this terminal Sail fragment. The short repeated sequences are not all identical, but they are all similar in composition. There are six types of repeats, and these are shown in Fig. 4. All are AT rich, and all contain the core sequence
Interestingly, all of the short repeated sequences in the end regions of the DNAs of vaccinia virus (Baroudy and Moss, 1982; Baroudy et a/., 1982) and cowpox virus (Pickup et al., 1982) contain a sequence similar if not identical to this core sequence. DISCUSSION RCN appears to be the orthopoxvirus whose DNA has the least similarity with the DNAs of all the other orthopoxviruses. The nucleotide sequence of RCN DNA differs from those of the others even in regions that contain essential viral genes. Yet the end regions of RCN DNA, which do not appear to encode any viral proteins, are extremely similar to the end regions of the DNAs of VV and CPV.
TGTGTMCCCA-CACTGTAGAGMCTCTCTMTG~GTG~T~TCCC~MTCGAGA RCN ATTTTATTTAATGTCTAGAAAAAAA :: :: :: :: TGTGTMCCCA~ACTGTAGGAMCTCTM;AG~T~-~~~AT~ATC~CT~AT~GAGA CPV ATTTTATTTAGTGTCTAGAAMAM W
TGTGT~ACCCATGACTGTAGGAMCTCTAGAG~GT~-~~GATCGATCGC~ATAGAGA ATTTTATTTAGTGTCTAGAAAAAAA
FIG. 5. Comparison of the nucleotide sequence of the region between the terminal loop and the first set of repeats in the DNA of RCN with corresponding sequences of the NRl regions of VV DNA (Baroudy et al., 1982) and CPV DNA (Pickup et al., 1982). The first nucleotide shown in the sequence of the W DNA is that adjacent to the innermost nucleotide of the 1OCnucleotide terminal loop of the viral DNA. The nucleotide sequences of the DNAs of RCN and CPV are aligned to show the similarity among these sequences; mismatches are indicated by colons. Seventy-one nucleotides of W NRl region, and 71 nucleotides of CPV NRl region match with nucleotides in the 87nucleotide RCN NRl region (82% match), whereas 84 of the nucleotides in the VV NRl region match with nucleotides in the 88-nucleotide CPV NRl region (98% match).
STRUCTURE
PCN
OF RACCOONPOX
51
VIRUS DNA
G--AMTTTTGTACCI-CG~AG~GTATCCCCTA~~TTTTACTA~T~~ATGTM~
TTATATTATT :: :::: CAMAAATATTATACGA
CPV GTGIUATTTTGTACCMACCG-CAGTATCCCCTACAT~A~ACTA~TCA~AT~TA~ W
CAMAAATATTATACGA GTGAAATlTTGTACCAAAGAMAAA~GTGAGCAGTATCCCCTACATGGATTl'TACTAGATCATTTATATAC
PCN
CGTGTACG~-----------------------MTTAT~TATTG-TAT~TATATCGA-----~TC~~AG-----C~ .. .. .. .. : ::: .. .. .. .. --T~TACGTT~ATTATAT~~TTTTMCGTGT~TTAT~CA~ATT~AT~T~--~GTCT~T~CCTA~T~G~~A~
cm
RCN CG-TG~~TAGACMTTMTCGTGTGT~TGTMC~TGA~ATTACAGC----------:: :: ::::::::::::::: ::: : : :: ::: : CPV GG-ATGTT~TMGCTCTAC(TATATGTTGCACCT W
GGGATGTTGATAAGCTC&AGTATATGTTGTTGGACGTTAT&TTAiGAAATAGTTGA&
FIG. 6. Comparison of the nucleotide sequences of the NR2 regions (the regions between the two sets of short repeats) in the DNAs of RCN, CPV (Pickup et a/., 1982). and VV (Baroudy ef a/., 1982). The sequences are aligned to show maximum similarity; mismatches are indicated by colons. Of the 254 nucleotides in the RCN NR2 region 194 match with nucleotides in the 325nucleotide VV NR2 region, whereas 304 of the 312 nucleotides in the CPV NR2 region match with nucleotides in the W NR2 region. Gaps were introduced into the aligned sequences only if they would permit a match of at least 60% in a sequence of at least 10 nucleotides.
The most striking features of these end regions are the two sets of short, directly repeated sequences. The various short repeated sequences of RCN DNA are all similar (Fig. 4) and they possess a core sequence in common with those of the repeats in the DNAs of vaccinia virus and cowpox virus. Esposito and Knight (1985) have shown by DNA hybridization experiments that the ends of the DNAs of monkeypoxvirus, smallpox virus, cowpox virus, ectromelia virus, Tatera poxvirus, and camelpox virus all contain sequences that are similar to those of the repeats in W DNA. Therefore representative orthopoxviruses of all types appear to contain similar, repeated sequences at the ends of their
DNAs. However, the functions of these short repeated sequences are unknown. The nonrepeated sequences flanking the sets of repeats are also similar in the DNAs of RCN, CPV, and W. At least one function of the nonrepeated sequences at the ends of poxvirus DNA has been identified. DeLange et al. (1986) and Merchlinsky and Moss (1986) have shown that a dimer of the 125-bp terminal nonrepeated sequence of VV DNA is sufficient to promote the resolution of circular molecules into linear molecules containing ends similar to those of the viral DNA. The terminal 100 bp of the DNA of Shope fibroma virus (SFV) contains the signals necessary for the resolution
SAL RCN
mmd
SAL CPV
SAL
FIG. 7. Comparison of the organization of repeated sequences (open rectangles) and nonrepeated sequences (NRl, NR2, and NR3: black rectangles) in the terminal Se/l fragments of the DNAs of RCN, CPV (Pickup ef al., 1982). and VV. Data on the DNA of vaccinia virus are taken from Baroudy and Moss (1982) and Baroudy et al. (1982).
PARSONS
52
of dimers of this viral DNA (DeLange et a/., 1986). Furthermore, there is a high degree of similarity between the nucleotide sequences at the ends of the DNAs of VV and SFV. These results suggested that the resolution of poxvirus telomeres might occur by a mechanism common to poxviruses of various types. The similarity of the sequence at the end of RCN DNA with the sequences at the ends of the DNAs of other orthopoxviruses is consistent with this proposal. Figure 5 shows a comparison of the nucleotide sequences of the nonrepeated regions between the terminal loops and the first sets of repeats in the DNAs of RCN, CPV, and W. All three sequences are similar, particularly in the region including and flanking the 1 I-bp sequence CTAGAAAAAAA. This 1 1-bp sequence is also present 55 bp from each end of the DNA of SFV; it appears to be required for the resolution of telomeres of both VV and SVF (DeLange et a/., 1986). Functions of the other nonrepeated sequences have not yet been identified. Nevertheless, the high degree of similarity between the nucleotide sequences of the NR2 regions of RCN, W, and CPV (Fig. 6) suggests that these regions are required for some viral functions. The situation is less clear with respect to the NR3 region of RCN DNA. Although it possesses some similarity with the NR3 region of CPV DNA (97 of the 188 nucleotides in these sequences are identical), this similarity is not as marked as that between the corresponding sequences of either the NRl regions or the NR2 regions. The arrangement of the repeated and nonrepeated sequences at the ends of the RCN, CPV, and VV DNAs are shown in Fig. 7. Although these DNAs differ in the lengths of the sets of repeats, they each have the same arrangement of the repeated and nonrepeated sequences. There are variations of this basic arrangement. For example, the end regions of the DNAs of some strains of W contain more than two sets of repeats and extra copies of the NR2 sequence (Moss et a/., 1981). However, a minimum of one NRl, one NR2, and two sets of repeated sequences appears to be required, because viruses whose DNAs lack these features have not been isolated. In conclusion, these results suggest that the various repeated and nonrepeated sequences have been maintained in a specific arrangement because in this arrangement they provide some functions that are important in virus multiplication. ACKNOWLEDGMENTS This \jvork was supported by Program Project Grant 5 PO1 CA30246 from the National institutes of Health.
AND
PICKUP
REFERENCES BARNES,W. M. (1980). DNA cloning with single-stranded phage vectors. In “Genetic Engineering” (1. K. Setlow and A. Hollaender, Eds.), Vol. 2, pp. 185-200. Plenum, New York. BAROUDY,B. M., and Moss, B. (1982). Sequence homologies of diverse length tandem repetitions near ends of vaccinia virus genome suggest unequal crossing over. Nucleic Acids Res. 10, 5673-5679. BAROUDY,B. M., VENKATESAN,S., and Moss, B. (1982). Incompletely base-paired flip-flop terminal loops link the two DNA strands of the vaccinia virus genome into one uninterrupted polynucleotide chain. Ce//28,315-324. BIGGIN, M. D., GIBSON, T. J., and HONG, G. F. (1983). Buffer gradient gels and r5S] label as an aid to rapid DNA sequence determination. Proc. Natl. Acad. Sci. USA 80, 3963-3965. DANIELS, D. L., SCHROEDER,J. L., BLATTNER,F. R., SZYBALSKI,W., and SANGER,F. (1983). In “Lambda II: Appendix” (R. W. Hendrix, J. W. Roberts, F. W. Stahl, and R. A. Weinberg, Eds.). Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. DEFILIPPES, F. M. (1982). Restriction enzyme mapping of vaccinia virus DNA. J. tirol. 43, 136-l 49. DELANGE, A. M., REDDY, M., SCRABA, D., UPTON, C., and MCFADDEN, G. (1986). Replication and resolution of cloned poxvirus telomeres in viva generates linear minichromosomes with intact viral hairpin termini. 1. Viral. 59, 249-259. ESPOSITO,J. J., and KNIGHT,J. C. (1985). Orthopoxvirus DNA: A comparison of restriction profiles and maps. Virology 143, 230-251. GANGEMI, J. D., and SHARP, D. G. (1976). Use of a restriction endonuclease in analyzing the genomes from two different strains of vaccinia virus. 1. Viral. 20, 319-323. GARON, C. F., BARBOSA,E., and Moss, B. (1978). Visualization of an inverted terminal repetition in vaccinia virus DNA. Proc. Nat/. Acad. Sci. USA 75,4863-4867. GESHELIN,P., and BERNS,K. I. (1974). Characterization and localization of the naturally occurring cross-links in vaccinia virus DNA. J. Mol. Biol. 88, 785-796. HENIKOFF,S. (1984). Unidirectional digestion with exonuclease III creates targeted breakpoints for DNA sequencing. Gene 28, 351359. JOKLIK,W. K. (1962). The purification of four strains of poxviruses. Virology 18, 9-l 8. MACKET~, M., and ARCHARD, L. C. (1979). Conservation and variation in orthopoxvirus genome structure. 1. Gen. Viral. 45. 683-701. MESSING,J., and VIEW, J. (1982). A new pair of Ml 3 vectors for selecting either strand of double-digest restriction fragments. Gene 19,269-276. MERCHLINSKY,M., and Moss, B. (1986). Resolution of linear minichromosomes with hairpin ends from circular plasmids containing vaccinia virus concatamer junctions. Cell 45, 879-884. Moss, B., WINTERS,E., and COOPER,N. (1981). Instabilityand reiteration of DNA sequences within the vaccinia virus genome. Proc. Nat/. Acad. Sci. USA 78, 1614-1618. MOYER, R. W., and ROTHE, C. T. (1980). The white pock mutants of rabbit poxvirus. I. Spontaneous host range mutants contain deletions. Virology 102, 1 19-l 32. MOLLER. H. K., WITTEK, R., SCHAFFNER,W.. SCHUMPERLI,D., MENNA, A., and WYLER, R. (1977). Comparison of five poxvirus genomes by analysis with restriction endonucleases Hindlll, Baml, and EcoRI. 1. Gen. Viral. 38, 135-l 47. NEVINS.J. R.. and JOKLIK.W. K. (1977). Isolation and properties of the vaccinia virus DNA-dependent RNA polymerase. 1. Bio;. Chem. 252, 6930-6938. NORRANDER,J., KEMPE, T., and MESSING, J. (1983). Construction of
STRUCTURE
OF RACCOONPOX
improved M 13 vectors using oligonucleotide-directed mutagenesis. Gene 26, 101-106. PATEL,D. D., PICKUP,D. J., and JOKLIK,W. K. (1986). Isolation of cowpox virus A-type inclusions and characterization of their major protein component. Virology 149, 174-l 89. PICKUP, D. J., BASTIA, D.. STONE, H. O., and JOKLIK,W. K. (1982). Sequence of terminal regions of cowpox virus DNA: Arrangement of repeated and unique sequence elements. Proc. Nat/. Acad. Sci. USA79,7112-7116. PICKUP, D. J.. INK, 8. S., PARSONS, B. L.. Hu, W., and JOKLIK,W. K. (1984). Spontaneous deletions and duplications of sequences in the genome of cowpoxvirus. Proc. Nat/. Acad. Sci. USA 81,68176821. RIGBY, P. W. J., DIECKMANN, M., RHODES, C., and BERG, P. (1977). Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113, 237251. SANGER,F., COULSON. A. R., BARRELL,B. G., SMITH, A. J. H., and ROE,
VIRUS DNA
53
B. A. (1980). Cloning in single-stranded bacteriophage as an aid to rapid DNA sequencing. J. Mol. Biol. 143, 161-l 78. SMITH, H. O., and BIRNSTIEL,M. L. (1976). A simple method for DNA restriction site mapping. Nucleic Acids Res. 3, 2387-2398. SOUTHERN,E. M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. /. Mol. Biol. 98, 503517. WEINRICH,S. L., NILES, E. G., and HRUBY. D. E. (1985). Transcriptional and translational analysis of the vaccinia virus late gene L65. /. Virol. 55, 450-457. WITTEK, R., MENNA, A., M(~LLER, H. K., SCHUMPERLI, D., BOSELEY, P. G., and WYLER, R. (1978). Inverted terminal repeats in rabbit poxvirus and vaccinia virus DNA. 1. Viral. 28, 171-l 81, WITTEK, R., MENNA, A., SCHI)MPERLI, D., STOFFEL, S., MUELLER,H. K., and WYLER, R. (1977). HindIll and Ss?l restriction sites mapped on rabbit poxvirus and vaccinia virus DNA. 1. Viral. 23, 669-678. WITTEK, R., and Moss, B. (1980). Tandem repeats within the inverted terminal repetition of vaccinia virus DNA. Cell 21, 277-284.
161,45-53
(1987)
Tandemly Repeated Sequences Are Present at the Ends of the DNA of Raccoonpox
Virus
BARBARA L. PARSONS AND DAVID J. PICKUP’ Department
of Microbiology
and Immunology,
Duke University
Received November
Medical
Center, Durham, North Carolina 277 10
7, 1986; accepted June 10, 1987
The DNA of raccoonpox virus (RCN) has been characterized by restriction enzyme analysis. DNA hybridization studies showed that all Hindill fragments of the 215kbp RCN DNA share some nucleotide sequence similarity with fragments of the DNA of cowpox virus (CPV). This information was used to construct a HindIll restriction map of the RCN DNA. The nucleotide sequence of the 2.2-kbp Sa/l end fragment of the RCN DNA has been determined from a cloned copy of the Hindlll 0 fragment. Of this 2.2-kbp region 75% consists of short, tandemly repeated sequences. It does not contain any open reading frames capable of encoding polypeptide chains of more than 62 amino acids. There are six related types of repeated sequence, and these are arranged into two separate sets, each flanked by nonrepeated sequences. The nucleotide sequences of both repeated and nonrepeated sequences within this Sal1 fragment are extremely similar to those of the Se/l-generated end fragments of the DNAs of CPV and vaccinia virus. The arrangements of the repeated and nonrepeated sequences are also similar in the DNAs of these three viruses. In contrast, the remainder of the RCN DNA is markedly different from the DNAs of other orthopoxviruses. The high degree of similarity between the ends of the RCN DNA and the ends of the other orthopoxvirus DNAs suggests that the complex arrays of repeated and nonrepeated sequences have been conserved because they have a role in virus tnukipkatiOn.
0 1987 Academic
Press,
Inc.
be present in the genomes of the other orthopoxviruses. Previous studies showed that the ends of the DNA of cowpox virus (CPV) contains two sets of short, tandemly repeated sequences; moreover, these repeated sequences and their flanking nonrepeated sequences are almost identical to the corresponding elements in the DNA of W (Baroudy and Moss, 1982; Baroudy et a/., 1982; Pickup et al., 1982). This similarity might reflect the conservation of these elements, or, because these viruses are closely related, it might simply reflect both a lack of sequence alteration and a lack of pressure to eliminate sequences that are of no value to the virus. A comparison of orthopoxvirus DNAs that are largely dissimilar would provide a better indication of regions that have been conserved. Most of the orthopoxvirus DNAs are remarkably similar with respect to the positions of their HindIll cleavage sites (Mackett and Archard, 1979); the DNA of raccoonpox virus (RCN) is the only known exception (Esposito and Knight, 1985). More extensive differences between RCN DNA and the other orthopoxvirus DNAs were detected by DNA hybridization experiments; of the DNAs of 38 viruses representative of all types of orthopoxviruses, RCN DNA was least similar to the others (Esposito and Knight, 1985). The extent of the differences between the DNA of RCN and the DNAs of the other orthopoxviruses suggests that the sequences which remain common to these DNAs are providing some functions that are advantageous to the virus.
INTRODUCTION The genomes of the orthopoxviruses are linear double-stranded DNAs, most of which are between 150 and 220 kbp in length (Joklik, 1962; Geshelin and Berns, 1974; Gangemi and Sharp, 1976; Wittek et a/., 1977; Mijller et a/., 1977; Mackett and Archard, 1979; Moyer and Rothe, 1980; Esposito and Knight, 1985). The most detailed studies on the structure of the ends of these viral DNAs have been done on the DNA of vaccinia virus (W),2 which is the prototype of the Orthopoxvirus genus. The W DNA contains inverted terminal repeats (ITRs) that are about 10 kbp in length (Garon et al,, 1978; Wittek et a/., 1978). At the two ends of the VV DNA the two strands of the DNA duplex are joined in a loop (Geshelin and Berns, 1974; Baroudy et al., 1982). The terminal 125 bp of VV DNA appear to be required for the resolution of unit length genomes from concatemers of viral DNA (DeLange et a/., 1986; Merchlinsky and Moss, 1986). The 3.5-kbp region next to the end loop consists largely of two sets of short, tandemly repeated sequences (Wittek and Moss, 1980; Baroudy and Moss, 1982; Baroudy et al., 1982). The significance of these repeated sequences and their flanking sequences is not clear. If these structural features are important for viral multiplication then it is likely that similar features will ’ To whom requests for reprints should be addressed. ’ Abbreviations used: RCN, raccoonpox virus; CPV, cowpoxvirus; VV. vaccinia virus; SFV, Shape fibroma virus; kbp, kilobase pairs.
45
0042.6622/87 $3.00 CopyrIght All nghts
0 1997 by Academic Press, Inc. of reproduction in any form reserved.
PARSONS
46
This paper describes the further characterization of the DNA of raccoonpox virus. The major objective of this study was to identify any conserved sequences within the end regions of the viral DNA. MATERIALS
AND
METHODS
Virus and viral DNA RCN, strain V71-I-85A (kindly provided by Dr. J. J. Esposito, Centers for Disease Control, Atlanta), was grown in human tk-143 cells. RCN produces A-type inclusions (ATI) that contain mature virus particles (unpublished observations). Virus was obtained from ATls that had been purified and then solubilized according to the methods of Pate1 et a/. (1986). This virus stock was used to infect monolayers of 143 cells. Virus was isolated from single plaques, and then the same procedure was repeated to obtain three-times plaque-purified virus. CPV, strain Brighton red, was grown in the chorioallantoic membrane of chick embryos. W, strain WR, was grown in mouse L cells. The viruses were purified as described by Joklik (1962). DNA was isolated from virus particles as described by Nevins and Joklik (1977). DNA hybridization
experiments
Standard methods were used to detect the hybridization of DNA fragments. DNA hybridization probes were labeled with 32P by nick-translation (Rigby et a/., 1977). The transfer of DNA fragments from agarose gels to nylon membranes (Biodyne A, Pall Ultrafine Filtration, Glen Cove, NY) and the conditions for DNADNA hybridization were as described by Southern (1975). The DNAs were hybridized at 65” for 16 hr in 6X SSC (0.9 M sodium chloride and 0.09 M sodium citrate), 2X Denhardt’s (0.04% Ficoll, 0.04% bovine serum albumin, and 0.04% polyvinylpyrrolidone), 0.1% SDS, 1 mM EDTA, and 20 pg/ml denatured salmon sperm DNA. The membranes were washed with 2x SSC, 0.1% SDS at 65” and then dried. Labeled DNAs on the membranes were visualized by autoradiography. Construction of the HindIll map of RCN DNA
restriction
The HindIll restriction map of RCN DNA was deduced largely from the extent of hybridization between the HindIll fragments of RCN DNA and mapped fragments of either W DNA or CPV DNA. The hybridization probes that were used are listed in Table 2. The positions of some fragments (HindIll R and Q) were determined by purifying these fragments and then using them as the hybridization probes. The RCN HindIll R fragment was used as a probe against mapped C/al
AND
PICKUP
restriction fragments of CPV Pstl C fragment (Pickup et a/., 1984). The RCN HindIll Q fragment was used as a probe against the mapped EcoRl fragments of the VV HindIll D fragment (Weinrich et al., 1985). Identification of the map positions of HindIll R and Q also established the order of these fragments relative to their respective adjacent fragments HindIll J and 0. The positions of the HindIll fragments derived from the two ends of the RCN genome were determined by the procedure of Smith and Birnstiel(l976) as follows. Nick-translated DNA probes (~177 and ~744) that were specific for the end regions of the viral DNA, were hybridized to blots of the resolved products of partial HindIll digestions of RCN DNA. ~177 is a plasmid that contains the 11.8-kbp terminal EcoRl fragment of the CPV genome; it hybridizes with both terminal HindIll fragments of RCN DNA. ~744 is a plasmid that contains a 3.3-kbp Pstl -Sal 1 fragment from the internal portion of the inverted terminal repeat of the CPV genome (Pickup et al., 1984); it hybridizes with only one of the terminal HindIll fragments of RCN DNA (HindIll H). The positions of HindIll cleavage sites within about 25 kbp of each end of the RCN DNA were deduced from the sizes of the digestion products that contained either probe-detectable end fragment. The sizes of these products of partial digestion were estimated from their electrophoretic mobilities (in a 0.6% agarose gel) relative to those of DNAs of known lengths. EcoRl, HindIll, Sail, andXba1 fragments of phage X DNA were used as size standards; their lengths have been determined from the nucleotide sequence of the DNA (Daniels eta/., 1983). Similarly, the lengths of the HindIll fragments of RCN, CPV, and VV DNAs were estimated from their electrophoretic mobilities relative to those of the EcoRl and HindIll fragments of phage X DNA.
TABLE 1 LENGTHSOF THE HindIll RESTRICTIONFRAGMENTSOF RCN DNA Fragment A B C D Eb F G H I J K
Length” 22.3 20.8 18.7 15.8 14.4 13.1 11.6 8.9 8.4 8.1 7.4
Fragment L M N Ob P Q R Sb T U V
’ Lengths are given in kilobase pairs. HindIll restriction shorter than 1.7 kbp in length were not sized. *Two molar fragments.
Length 7.0 5.4 5.1 4.9 4.5 2.9 2.8 2.4 2.2 2.1 1.7 fragments
STRUCTURE
OF RACCOONPOX
47
VIRUS DNA
TABLE 2 HYBRIDIZATIONOF CPV AND W DNA FRAGMENTSTO HindIll FRAGMENTSOF RCN DNA Hindlll fragments
of RCN DNA
ABCDEFGHIJKLMNOPQRSTU 1. CPV DNA probes Hindlll I Hindlll L Kpnl C Kpnl D Kpn,l G Kpnl K Kpnl L Pstl A Pstl c Pstl E Pstl F Psfl G Pstl I Pstl J Pstl M Pstl 0 p177* p7446 2. VV DNA probes Hindlll D Hindlll E Hindlll H Hindlll I Hindlll J Hindlll M Hindlll N Hindlll 0 p730*
V
+ +
+
+
+
+ +
+ +
+
+ f
+ + +
+
+ +
+
+
+
+ +
+
+
+
NT”
+
+
+
+
+
+
+ +
+ +
+ +
+ t
+ +
+
+ +
+ + + +
+ + +
+ +
a Not tested. b Plasmids containing viral DNA fragments that have not been designated by a single letter: ~177 contains the 11.8-kbp terminal EcoRl fragment of the CPV genome, ~744 contains a 3.3-kbp Sal1 -Pstl fragment from the internal portion of the CPV inverted terminal repeat, and ~730 contains a 3.5-kbp Xmal-Hindlll fragment from the W Hindlll A fragment.
Cloning of the HindIll-generated fragment of RCN DNA
4.9-kbp terminal
This end fragment was cloned according to the methods described by Pickup et a/. (1982). The terminal loops of the RCN DNA were removed by digestion with nuclease Sl 175 pg DNA and 1100 units of nuclease Sl in 375 ~1 of 30 mM sodium acetate, pH 4.5/0.3 n/r NaCI/l mM ZnSOJ5% (v/v) glycerol (57.5” for 5 min)]. The DNA was repaired with Escherichia co/i DNA polymerase I (Klenow fragment), and EcoRl linkers (GGAATTCC) were ligated to it. After the ligation the DNA was digested with EcoRl and HindIll. The terminal 4.9-kbp HindIll fragment (containing a cleaved EcoRl
recognition site in place of the terminal loop) was purified by agarose gel electrophoresis. This was then inserted between the EcoRl and HindIll sites in the DNA of plasmid pUC9 (Messing and Vieira, 1982). Standard methods were used to map the positions of restriction enzyme recognition sites within the cloned fragment. Nucleotide
sequence
analysis
Restriction enzyme-generated fragments of the cloned DNA corresponding to the terminal Sal1 fragment of the viral DNA were cloned into phage Ml 3 vectors mp18 and mpl9 (Norrander et al., 1983). These
PARSONS
48 0
RCN
CPV
vv
)
D
Nc( JVTP
E U
E
Ial
C
PCflG
F Qt.4 J NN K
E
A
EOIGLJH
D
A
C
NMK
F
S
A
LF
D
G
C
I
SSNKH
s
LH
B m
FIG. 1. A comparison of the HindIll restriction maps of RCN, CPV, and W DNAs. The maps of CPV DNA (Mackett and Archard, 1979) and W DNA (Mackett and Archard 1979; DeFilippes, 1982) are aligned with the map of the RCN DNA at the cross-hybridizing regions corresponding to CPV Hindlll E, W HindlIt D, and RCN Hindlll I, 0, and Q.
fragments were products of DNA cleavage by the following enzymes: EcoRl, Taql, Hirtfl, Ddel, and Sail. The orientations of the inserted fragments within the phage DNA were determined as described by Barnes (1980). The method of Henikoff (1984) was used to produce targeted breakpoints within the 1275-bp long innermost set of short, directly repeated sequences. First, to make unidirectional deletions extending from the Sal1 site into the set of repeats, the replicative form DNA of an Ml 3 mp18 clone containing the terminal Sal1 fragment (fcoR-Sal1 fragment) was cut with Sal1 and Sphl. The DNA was digested with exonuclease III at 37”, and aliquots were removed at 20-set intervals in order to obtain DNAs that had progressively longer singlestranded regions. The exonuclease III was inactivated by incubation at 70” for 10 min after the addition of 3 vol of 0.2 M NaCI, 5 mM EDTA, pH 8.0. Residual singlestranded regions were removed by nuclease Sl digestion, followed by repair with DNA polymerase I (Klenow fragment). fcoR1 linkers were ligated to the ends of the DNAs, and after subsequent fcoR1 digestion, the fragments were resolved by agarose gel electrophoresis. The sizes of the fragments were estimated from their electrophoretic mobilities; fragments larger than 800 bp and differing by increments of about 150 bp, were isolated and inserted into phage vector Ml 3 mp19. A similar procedure was used to make unidirectional deletions extending from the Doe1 site (at nucleotide 484) through the set of direct repeats, toward the Sal 1 site. For this purpose the 1663-bp Ddel fragment containing the second set of repeats was isolated, its ends were repaired with DNA polymerase I (Klenow fragment), and then it was inserted into the Smal site of the DNA of Ml 3 mp18. A clone was selected in which the orientation of the direct repeats (with respect to the Sphl and Sal1 sites of the vector DNA) was reversed relative to that in the clone used to generate the other set of unidirectional deletions. Otherwise similar methodology was used except that here only
AND
PICKUP
fragments greater than 300 bp long were inserted into the DNA of Ml 3 mpl9. These M 13 clones, which contained overlapping deletions, were used to determine the sequence of both strands of the entire fcoR 1-Sal 1 fragment. Their nucleotide sequences were determined according to the procedures of Sanger et al. (1980) and Biggin et al. (1983). RESULTS Characterization
of the DNA of RCN
HindIll digestion of the RCN DNA produced 25 fragments that were greater than 1600 bp long. The sum of the estimated sizes of these fragments (HindIll AV; Table 1) suggests that the intact RCN DNA is about 2 15 kbp long. Esposito and Knight (1985) reported that several HindIll fragments of RCN DNA did not hybridize with various cloned fragments of the DNAs of monkeypox virus, vaccinia virus, and variola virus. Our hybridization analyses showed that RCN HindIll fragments A through V all hybridize with fragments of CPV DNA (though fragments D, M, and R hybridize only weakly). Therefore it appears that RCN DNA and CPV DNA, two of the largest orthopoxvirus DNAs, contain some similar sequences that are not contained within the genomes of some of the other ot-thopoxviruses. Table 2 summarizes the results of hybridizations between RCN HindIll fragments and fragments of the DNAs of CPV and W. Most of the DNA probes hybridized with at least two fragments of RCN DNA. This identified fragments that were adjacent in the RCN DNA, thereby permitting construction of most of the HindIll map of this DNA. The procedure of Smith and Birnstiel (1976) was used to confirm the positions of HindIll sites close to the ends of the viral DNA. The completed Hindlll restriction map of RCN DNA is shown in Fig. 1. For comparison it is shown alongside the HindIll maps of the DNAs of VV and CPV (Mackett and Archard, 1979; DeFilippes, 1982). W is the pro.
----A
-c ----
E
D
T
.
T 1
T
-c-
--c--c-
c---c--b
DS I,,,
500
lob0
1&o
2&Q
FIG. 2. Restriction map of the cloned terminal Sal1 fragment of RCN DNA. Restriction enzyme cleavage sites are abbreviated as follows: E, EcoRl ; T, Taql ; D, Ddel ; S, Sal1 ; 1, Hinfl The arrows correspond to the extent of the nucleotide sequence (5’ to 3’ in the direction of the arrow) obtained from each template used. This Sal1 fragment contains 2195 base pairs.
STRUCTURE
Jc
OF RACCOONPOX
VIRUS DNA
120 maAGAGAcGAAA~-AG~cA~~AAAGA-
loo
ACA~CC#ACATTTlTAG&ATT
AAGAGAl&h?TFWAGACA+AAGAGAAATA 300
260 ----l-Y
320 ~UAAACTTTlTTA~AC&T
TATlTTlT$T&IllTTATGACd+AGA
---Y--
160 C
GAAATA+TG
i--
lUA%TTCTTAGGTCUAG
49
f.GAbAGAGACGA$ GACGA&ACAT>GAGACAT
500 540 520 Tl%AltXTTTMAATATGGACTAGAAIATGTCnXGUAAbACT
920 940 960 900 TlTTTTTATCACTCCA~AGMAGAGAMIA lTTTT$GTAAMCTITTlTATGACTCCA~A (5) *l-~~-T-c 1000 1020 1040 1060 1080 GUAGAGACGAAACATTTTATlSAGACAT’jAGAAAGAGMAGAGACGAAACA ~GTAAAACTTlTTTATGACTCCA~AGAAAG 1100 1120 ~CGA$CATl-lTTAG~TACA~~GMAGAGAAATA 1160 :'2f""--?-
1200 mGAcGAAAcA
1140 1160 IlTlTTGUAAAC~A~ACTCCA~~GAGA~C <5> 1220 1240 1260 TAMA-AmCTCCAm,AGAAAGAGAMGAGACGA
T-F1200 1300 A$AllllTAGTGAGACAT2pAMGAUAGAGAWAACA vG=-c
1600 1700 -TA~TAAAACTl-lTTTATGACTCC I 1720 1740 1760 1780 la00 A~,4GWAGATGMA~~AG~CA~~WGAGAAAnrrt""""rr lTTTT$CTAAAACTTllTTAEACTCCAlT~GAAAGAGATG u-cTcY
w7
1640 --%---=~
1320 1340 lTlTllTAn;A~CCA~AGMAGAGAAAGAGA
1660
1820 "7
1840 1860 GAWAAACA~TAAMCTTlTlTATGACTCCATT
1900 ~~--=Y--7
1920 -cGAAAcA* 2000 Yet-y--=-
2060 lTnTAAAmcAAAccccIII
1080 7""
1960 1960 TMAACllTlTAlGACTCCAlT~GAAAGAGATGA&ACAT 2020
2100 UAGTCTCCCGTACAC
2040 2060 CtTGTTACl'TAlTW2AGA~TCGACTlTMTGCGAATACA 2120
2140 2160 CGTATGTCCCM~CCACAA~TCCIGAGGAT'2ACAC~
2180 6lTUGACACACGAnUSnT?GTCTCGIWAC FIG. 3. Nucleotide sequence of the cloned terminal Sal1 fragment of RCN DNA. Nonrepeated sequences (nucleotides l-l 13, 479-732, and 2009-2195) are underlined. The extent and type of each repeated unit (see also Fig. 4) is indicated beneath the sequence. Incomplete repeats are indicated by an asterisk. Nucleotide 1 is adjacent to the EcoRl linker.
totype of the orthopoxvirus genus. CPV appears to have the largest genome of any orthopoxvirus; we estimate that it is 230 kbp long. Although there is little similarity between the HindIll map of RCN DNA and the HindIll maps of the CPV and W DNAs, there is extensive sim-
ilarity between the nucleotide sequences of these three DNAs. In these maps, most of the cross-hybridizing Hindlll fragments of RCN DNA and CPV DNA would have aligned well but for a region within the RCN HindIll E fragment, which appears to lack about 14 kbp
PARSONS
50
EOPVnunma
In!.& 1
&)J&AGAAAQA~-c&-
--------m-
-mum
1
AGAAAGAG----AcGAAA~----------rTrnAGr6AGAcAn
3
AGAMGAG------A-CA
~CCMACATTlWlTETAAAACTllTtTATGA~CCAlT
5
mQ,AA--rA I
CQmal
12
llllll
II
E
---- m-elTll7TETAAAACln-PPTATGA~CCA~
1 Tl-lTTA
This arrangement suggested that sets of short tandemly repeated sequences containing the Hinfl site were present at the ends of RCN DNA. Nucleotide sequence of the terminal Sdl fragment
9
-~rM~~A------~r-C~A~~C~n
AGAAAGAG
PICKUP
2
‘CSCCAn
b
6
9
AND
CAlT
FIG. 4. Composition of the repeated units in RCN DNA. The sequences are aligned to show maximum similarity. Vertical bars indicated mismatches. Nucleotides common to repeats of all six types are shown in the bottom line.
of DNA that is present in the HindIll G/F region of CPV DNA. Characterization of a cloned HindIll-generated terminal fragment of RCN DNA The terminal, cross-linked HindIll-generated fragments of this RCN DNA are the H and 0 fragments (Esposito and Knight, 1985). An RCN /-findIll 0 fragment, containing an EcoRl linker in the place of the terminal hairpin loop, was isolated, and then inserted into the DNA of plasmid pUC9. To estimate how much of the terminal loop of the viral DNA had been removed by the nuclease Sl treatment, the sizes of the cloned and uncloned terminal fragments were compared by agarose gel electrophoresis. The resolved fragments were transferred to a nylon membrane, and a nicktranslated probe of the cloned terminal Sal1 fragment was used to detect the band corresponding to the terminal Doe1 fragment of the viral DNA. This analysis indicated that about 30 bp of the terminal loop had been removed by the nuclease Sl treatment. A 2.2 kbp EcoRl-Sal1 fragment corresponding to the Sal1 fragment at the end of the viral DNA was subcloned from the HindIll 0 fragment. A restriction map of this Sal1 fragment is shown in Fig. 2. The fragment contains cleavage sites of only a few different restriction enzymes. However, there are multiple Hinfl sites, most of which are spaced at intervals of less than 100 bp.
The strategy employed to determine the sequence of the terminal Sa/l fragment is depicted in Fig. 2. The nucleotide sequence of this fragment is shown in Fig. 3. It does not contain any long open reading frames; its largest open reading frame (nucleotides 656-841; see Fig. 3) could encode a polypeptide chain of only 62 amino acids. The sequence contains three regions (nucleotides l-l 13,479-732, and 2009-2195) whose sequences are not repeated elsewhere within the Sal1 fragment. These three regions of nonrepeated sequences (designated NRl , NR2, and NR3) are separated by two sets of short, tandemly repeated sequences. The repeated sequences make up about 759/o of this terminal Sail fragment. The short repeated sequences are not all identical, but they are all similar in composition. There are six types of repeats, and these are shown in Fig. 4. All are AT rich, and all contain the core sequence
Interestingly, all of the short repeated sequences in the end regions of the DNAs of vaccinia virus (Baroudy and Moss, 1982; Baroudy et a/., 1982) and cowpox virus (Pickup et al., 1982) contain a sequence similar if not identical to this core sequence. DISCUSSION RCN appears to be the orthopoxvirus whose DNA has the least similarity with the DNAs of all the other orthopoxviruses. The nucleotide sequence of RCN DNA differs from those of the others even in regions that contain essential viral genes. Yet the end regions of RCN DNA, which do not appear to encode any viral proteins, are extremely similar to the end regions of the DNAs of VV and CPV.
TGTGTMCCCA-CACTGTAGAGMCTCTCTMTG~GTG~T~TCCC~MTCGAGA RCN ATTTTATTTAATGTCTAGAAAAAAA :: :: :: :: TGTGTMCCCA~ACTGTAGGAMCTCTM;AG~T~-~~~AT~ATC~CT~AT~GAGA CPV ATTTTATTTAGTGTCTAGAAMAM W
TGTGT~ACCCATGACTGTAGGAMCTCTAGAG~GT~-~~GATCGATCGC~ATAGAGA ATTTTATTTAGTGTCTAGAAAAAAA
FIG. 5. Comparison of the nucleotide sequence of the region between the terminal loop and the first set of repeats in the DNA of RCN with corresponding sequences of the NRl regions of VV DNA (Baroudy et al., 1982) and CPV DNA (Pickup et al., 1982). The first nucleotide shown in the sequence of the W DNA is that adjacent to the innermost nucleotide of the 1OCnucleotide terminal loop of the viral DNA. The nucleotide sequences of the DNAs of RCN and CPV are aligned to show the similarity among these sequences; mismatches are indicated by colons. Seventy-one nucleotides of W NRl region, and 71 nucleotides of CPV NRl region match with nucleotides in the 87nucleotide RCN NRl region (82% match), whereas 84 of the nucleotides in the VV NRl region match with nucleotides in the 88-nucleotide CPV NRl region (98% match).
STRUCTURE
PCN
OF RACCOONPOX
51
VIRUS DNA
G--AMTTTTGTACCI-CG~AG~GTATCCCCTA~~TTTTACTA~T~~ATGTM~
TTATATTATT :: :::: CAMAAATATTATACGA
CPV GTGIUATTTTGTACCMACCG-CAGTATCCCCTACAT~A~ACTA~TCA~AT~TA~ W
CAMAAATATTATACGA GTGAAATlTTGTACCAAAGAMAAA~GTGAGCAGTATCCCCTACATGGATTl'TACTAGATCATTTATATAC
PCN
CGTGTACG~-----------------------MTTAT~TATTG-TAT~TATATCGA-----~TC~~AG-----C~ .. .. .. .. : ::: .. .. .. .. --T~TACGTT~ATTATAT~~TTTTMCGTGT~TTAT~CA~ATT~AT~T~--~GTCT~T~CCTA~T~G~~A~
cm
RCN CG-TG~~TAGACMTTMTCGTGTGT~TGTMC~TGA~ATTACAGC----------:: :: ::::::::::::::: ::: : : :: ::: : CPV GG-ATGTT~TMGCTCTAC(TATATGTTGCACCT W
GGGATGTTGATAAGCTC&AGTATATGTTGTTGGACGTTAT&TTAiGAAATAGTTGA&
FIG. 6. Comparison of the nucleotide sequences of the NR2 regions (the regions between the two sets of short repeats) in the DNAs of RCN, CPV (Pickup et a/., 1982). and VV (Baroudy ef a/., 1982). The sequences are aligned to show maximum similarity; mismatches are indicated by colons. Of the 254 nucleotides in the RCN NR2 region 194 match with nucleotides in the 325nucleotide VV NR2 region, whereas 304 of the 312 nucleotides in the CPV NR2 region match with nucleotides in the W NR2 region. Gaps were introduced into the aligned sequences only if they would permit a match of at least 60% in a sequence of at least 10 nucleotides.
The most striking features of these end regions are the two sets of short, directly repeated sequences. The various short repeated sequences of RCN DNA are all similar (Fig. 4) and they possess a core sequence in common with those of the repeats in the DNAs of vaccinia virus and cowpox virus. Esposito and Knight (1985) have shown by DNA hybridization experiments that the ends of the DNAs of monkeypoxvirus, smallpox virus, cowpox virus, ectromelia virus, Tatera poxvirus, and camelpox virus all contain sequences that are similar to those of the repeats in W DNA. Therefore representative orthopoxviruses of all types appear to contain similar, repeated sequences at the ends of their
DNAs. However, the functions of these short repeated sequences are unknown. The nonrepeated sequences flanking the sets of repeats are also similar in the DNAs of RCN, CPV, and W. At least one function of the nonrepeated sequences at the ends of poxvirus DNA has been identified. DeLange et al. (1986) and Merchlinsky and Moss (1986) have shown that a dimer of the 125-bp terminal nonrepeated sequence of VV DNA is sufficient to promote the resolution of circular molecules into linear molecules containing ends similar to those of the viral DNA. The terminal 100 bp of the DNA of Shope fibroma virus (SFV) contains the signals necessary for the resolution
SAL RCN
mmd
SAL CPV
SAL
FIG. 7. Comparison of the organization of repeated sequences (open rectangles) and nonrepeated sequences (NRl, NR2, and NR3: black rectangles) in the terminal Se/l fragments of the DNAs of RCN, CPV (Pickup ef al., 1982). and VV. Data on the DNA of vaccinia virus are taken from Baroudy and Moss (1982) and Baroudy et al. (1982).
PARSONS
52
of dimers of this viral DNA (DeLange et a/., 1986). Furthermore, there is a high degree of similarity between the nucleotide sequences at the ends of the DNAs of VV and SFV. These results suggested that the resolution of poxvirus telomeres might occur by a mechanism common to poxviruses of various types. The similarity of the sequence at the end of RCN DNA with the sequences at the ends of the DNAs of other orthopoxviruses is consistent with this proposal. Figure 5 shows a comparison of the nucleotide sequences of the nonrepeated regions between the terminal loops and the first sets of repeats in the DNAs of RCN, CPV, and W. All three sequences are similar, particularly in the region including and flanking the 1 I-bp sequence CTAGAAAAAAA. This 1 1-bp sequence is also present 55 bp from each end of the DNA of SFV; it appears to be required for the resolution of telomeres of both VV and SVF (DeLange et a/., 1986). Functions of the other nonrepeated sequences have not yet been identified. Nevertheless, the high degree of similarity between the nucleotide sequences of the NR2 regions of RCN, W, and CPV (Fig. 6) suggests that these regions are required for some viral functions. The situation is less clear with respect to the NR3 region of RCN DNA. Although it possesses some similarity with the NR3 region of CPV DNA (97 of the 188 nucleotides in these sequences are identical), this similarity is not as marked as that between the corresponding sequences of either the NRl regions or the NR2 regions. The arrangement of the repeated and nonrepeated sequences at the ends of the RCN, CPV, and VV DNAs are shown in Fig. 7. Although these DNAs differ in the lengths of the sets of repeats, they each have the same arrangement of the repeated and nonrepeated sequences. There are variations of this basic arrangement. For example, the end regions of the DNAs of some strains of W contain more than two sets of repeats and extra copies of the NR2 sequence (Moss et a/., 1981). However, a minimum of one NRl, one NR2, and two sets of repeated sequences appears to be required, because viruses whose DNAs lack these features have not been isolated. In conclusion, these results suggest that the various repeated and nonrepeated sequences have been maintained in a specific arrangement because in this arrangement they provide some functions that are important in virus multiplication. ACKNOWLEDGMENTS This \jvork was supported by Program Project Grant 5 PO1 CA30246 from the National institutes of Health.
AND
PICKUP
REFERENCES BARNES,W. M. (1980). DNA cloning with single-stranded phage vectors. In “Genetic Engineering” (1. K. Setlow and A. Hollaender, Eds.), Vol. 2, pp. 185-200. Plenum, New York. BAROUDY,B. M., and Moss, B. (1982). Sequence homologies of diverse length tandem repetitions near ends of vaccinia virus genome suggest unequal crossing over. Nucleic Acids Res. 10, 5673-5679. BAROUDY,B. M., VENKATESAN,S., and Moss, B. (1982). Incompletely base-paired flip-flop terminal loops link the two DNA strands of the vaccinia virus genome into one uninterrupted polynucleotide chain. Ce//28,315-324. BIGGIN, M. D., GIBSON, T. J., and HONG, G. F. (1983). Buffer gradient gels and r5S] label as an aid to rapid DNA sequence determination. Proc. Natl. Acad. Sci. USA 80, 3963-3965. DANIELS, D. L., SCHROEDER,J. L., BLATTNER,F. R., SZYBALSKI,W., and SANGER,F. (1983). In “Lambda II: Appendix” (R. W. Hendrix, J. W. Roberts, F. W. Stahl, and R. A. Weinberg, Eds.). Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. DEFILIPPES, F. M. (1982). Restriction enzyme mapping of vaccinia virus DNA. J. tirol. 43, 136-l 49. DELANGE, A. M., REDDY, M., SCRABA, D., UPTON, C., and MCFADDEN, G. (1986). Replication and resolution of cloned poxvirus telomeres in viva generates linear minichromosomes with intact viral hairpin termini. 1. Viral. 59, 249-259. ESPOSITO,J. J., and KNIGHT,J. C. (1985). Orthopoxvirus DNA: A comparison of restriction profiles and maps. Virology 143, 230-251. GANGEMI, J. D., and SHARP, D. G. (1976). Use of a restriction endonuclease in analyzing the genomes from two different strains of vaccinia virus. 1. Viral. 20, 319-323. GARON, C. F., BARBOSA,E., and Moss, B. (1978). Visualization of an inverted terminal repetition in vaccinia virus DNA. Proc. Nat/. Acad. Sci. USA 75,4863-4867. GESHELIN,P., and BERNS,K. I. (1974). Characterization and localization of the naturally occurring cross-links in vaccinia virus DNA. J. Mol. Biol. 88, 785-796. HENIKOFF,S. (1984). Unidirectional digestion with exonuclease III creates targeted breakpoints for DNA sequencing. Gene 28, 351359. JOKLIK,W. K. (1962). The purification of four strains of poxviruses. Virology 18, 9-l 8. MACKET~, M., and ARCHARD, L. C. (1979). Conservation and variation in orthopoxvirus genome structure. 1. Gen. Viral. 45. 683-701. MESSING,J., and VIEW, J. (1982). A new pair of Ml 3 vectors for selecting either strand of double-digest restriction fragments. Gene 19,269-276. MERCHLINSKY,M., and Moss, B. (1986). Resolution of linear minichromosomes with hairpin ends from circular plasmids containing vaccinia virus concatamer junctions. Cell 45, 879-884. Moss, B., WINTERS,E., and COOPER,N. (1981). Instabilityand reiteration of DNA sequences within the vaccinia virus genome. Proc. Nat/. Acad. Sci. USA 78, 1614-1618. MOYER, R. W., and ROTHE, C. T. (1980). The white pock mutants of rabbit poxvirus. I. Spontaneous host range mutants contain deletions. Virology 102, 1 19-l 32. MOLLER. H. K., WITTEK, R., SCHAFFNER,W.. SCHUMPERLI,D., MENNA, A., and WYLER, R. (1977). Comparison of five poxvirus genomes by analysis with restriction endonucleases Hindlll, Baml, and EcoRI. 1. Gen. Viral. 38, 135-l 47. NEVINS.J. R.. and JOKLIK.W. K. (1977). Isolation and properties of the vaccinia virus DNA-dependent RNA polymerase. 1. Bio;. Chem. 252, 6930-6938. NORRANDER,J., KEMPE, T., and MESSING, J. (1983). Construction of
STRUCTURE
OF RACCOONPOX
improved M 13 vectors using oligonucleotide-directed mutagenesis. Gene 26, 101-106. PATEL,D. D., PICKUP,D. J., and JOKLIK,W. K. (1986). Isolation of cowpox virus A-type inclusions and characterization of their major protein component. Virology 149, 174-l 89. PICKUP, D. J., BASTIA, D.. STONE, H. O., and JOKLIK,W. K. (1982). Sequence of terminal regions of cowpox virus DNA: Arrangement of repeated and unique sequence elements. Proc. Nat/. Acad. Sci. USA79,7112-7116. PICKUP, D. J.. INK, 8. S., PARSONS, B. L.. Hu, W., and JOKLIK,W. K. (1984). Spontaneous deletions and duplications of sequences in the genome of cowpoxvirus. Proc. Nat/. Acad. Sci. USA 81,68176821. RIGBY, P. W. J., DIECKMANN, M., RHODES, C., and BERG, P. (1977). Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113, 237251. SANGER,F., COULSON. A. R., BARRELL,B. G., SMITH, A. J. H., and ROE,
VIRUS DNA
53
B. A. (1980). Cloning in single-stranded bacteriophage as an aid to rapid DNA sequencing. J. Mol. Biol. 143, 161-l 78. SMITH, H. O., and BIRNSTIEL,M. L. (1976). A simple method for DNA restriction site mapping. Nucleic Acids Res. 3, 2387-2398. SOUTHERN,E. M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. /. Mol. Biol. 98, 503517. WEINRICH,S. L., NILES, E. G., and HRUBY. D. E. (1985). Transcriptional and translational analysis of the vaccinia virus late gene L65. /. Virol. 55, 450-457. WITTEK, R., MENNA, A., M(~LLER, H. K., SCHUMPERLI, D., BOSELEY, P. G., and WYLER, R. (1978). Inverted terminal repeats in rabbit poxvirus and vaccinia virus DNA. 1. Viral. 28, 171-l 81, WITTEK, R., MENNA, A., SCHI)MPERLI, D., STOFFEL, S., MUELLER,H. K., and WYLER, R. (1977). HindIll and Ss?l restriction sites mapped on rabbit poxvirus and vaccinia virus DNA. 1. Viral. 23, 669-678. WITTEK, R., and Moss, B. (1980). Tandem repeats within the inverted terminal repetition of vaccinia virus DNA. Cell 21, 277-284.