JOURNAL OF VIROLOGY, Jan. 1988, p. 266-276 Vol. 62, No.

1
0022-538X/88/010266-11$02.00/0
Copyright C 1988, American Society for Microbiology

Nucleotide Sequence and Genome Organization of

Canine Parvovirus
A. PAUL REED, ELAINE V. JONES, AND TIMOTHY J. MILLER*
Department of Molecular Genetics, SmithKline Beckman Corporation, Swedeland, Pennsylvania 19406
Received 5 June 1987/Accepted 15 September 1987
The genome of a canine parvovirus isolate strain (CPV-N) was cloned, and the DNA sequence was
determined. The entire genome, including ends, was 5,323 nucleotides in length. The terminal repeat at the 3'
end of the genome shared similar structural characteristics but limited homology with the rodent parvoviruses.
The 5' terminal repeat was not detected in any of the clones. Instead, a region of DNA starting near the capsid
gene stop codon and extending 248 base pairs into the coding region had been duplicated and inserted 75 base
pairs downstream from the poly(A) addition site. Consensus sequences for the 5' donor and 3' acceptor sites
as well as promotors and poly(A) addition sites were identified and compared with the available information on
related parvoviruses. The genomic organization of CPV-N is similar to that of feline parvovirus (FPV) in that
there are two major open reading frames (668 and 722 amino acids) in the plus strand (mRNA polarity). Both
coding domains are in the same frame, and no significant open reading frames were apparent in any of the other
frames of both minus and plus DNA strands. The nucleotide and amino acid homologies of the capsid genes
between CPV-N and FPV were 98 and 99%, respectively. In contrast, the nucleotide and amino acid
homologies of the capsid genes for CPV-N and CPV-b (S. Rhode III, J. Virol. 54:630-633, 1985) were 95 and
98%, respectively. These results indicate that very few nucleotide or amino acid changes differentiate the
antigenic and host range specificity of FPV and CPV.

Canine parvovirus (CPV) is a member of the autono- stage but involve a mechanism occurring after synthesis of
mously replicating parvoviruses and is associated with en- replicative form (RF) DNA (3, 43, 44). Thus, the strain-
teritis and myocarditis in dogs. CPV-associated disease in specific DNA determinant expressed by the variant virus
dogs became predominant in 1978 (11) and is assumed to dictates the host range specificity, since MVMi and MVMp
have arisen as a variant of feline parvovirus (FPV) (47). are serologically identical and can bind to receptors on both
Information on the virus proteins (23) as well as on the DNA restrictive and productive cell types (43). Strain-specific
sequence (10, 41) verifies that there is a great deal of determinants can be observed in variant virus populations
homology between FPV and CPV. Despite the close simi- following high passage of the virus in transformed cell lines.
larity, the two viruses can be distinguished on the basis of For example, the 5' untranslated region of Hi parvoviruses
restriction enzyme mapping (28, 49), antigen cross-reactivity becomes reiterated when grown in NB cells (38, 39). A
(35), and host range specificity (49). However, high passage correlation between the DNA sequence variation observed
of CPV in feline and canine cells can induce mutations which in highly passaged parvovirus and changes in the virus DNA
can be identified as variants from the parental strain (34). In replication or gene expression patterns has not been demon-
addition, since 1978, new variants seem to have arisen in strated. Since high passage of parvoviruses induces changes
nature (35). Because these variants apparently arose so in virus specificity and antigenic characteristics (34, 38, 39),
quickly, it is of growing concern to understand more about it is of interest to compare the DNA sequence from a
the mechanisms of virus replication and gene expression. high-passage CPV isolate with sequences from related par-
Autonomous parvoviruses have a single-stranded linear voviruses. We have cloned and sequenced a strain of CPV
genome of approximately 5,000 nucleotides (nt). Nucleotide (isolated at Norden Laboratories, Lincoln, Nebr.) from a
sequence (3, 10, 15, 41, 42, 46) and transcriptional mapping high-passage infection in dog kidney cells. This data and the
data (13, 21, 24, 25, 36, 48) have revealed several common available data for CPV (strain b) (41) and FPV (10) are
features of parvoviruses. (i) There are two major open compared. In addition, except for the terminal 5' repeat, this
reading frames (ORFs). (ii) The ORF in the 3' half of the is the first report for the entire CPV genome.
genome encodes the nonstructural proteins, while the ORF
in the 5' half of the genome encodes the structural proteins.
(iii) The nonstructural and structural genes are initiated from MATERIALS AND METHODS
separate promotors. (vi) The mRNAs of the nonstructural Materials. All restriction enzymes and DNA modification
and structural proteins have coterminal poly(A) addition enzymes were purchased from New England BioLabs, Inc.,
sites (map position [mp] 94 to 96) and are spliced to allow Beverly, Mass., or Bethesda Research Laboratories, Inc.,
alternate templates for protein synthesis. Gaithersburg, Md. Isotopically labeled [o-32P]dATP (>3,000
The ability of parvoviruses to replicate in specific cells is Ci/mM and [a-35S]dATP (>1,000 Ci/mM) were purchased
dependent on both cellular and viral determinants. Host from Amersham Corp. Arlington Heights, Ill.
range- or strain-specific tropisms for both the fibrotrophic (p) M13 vectors were obtained from Joachim Messing and
and lymphotrophic (i) strains of the minute virus of mice were grown and prepared as described by Messing (30).
(MVM) do not seem to occur at the penetration or uncoating Plasmid DNAs were purified by using standard cleared-lysis
procedures and two bandings on CsCl (26). The plasmid
*
Corresponding author. DNAs used for cloning were pBR322, pHC624 (7; a gift from
266
VOL. 62, 1988
CPV SEQUENCE 267

P4 P38 ATO land, Maine) in TBE buffer or 0.6% agarose gels (LMT
V102 ATO AATAAA SeaPlaque; FMC Corp.) in TAE buffer (26). Bands were
0 + 10 20
i40 50I 60I 30 70 l
ProbsS
VbW I I 8 90 10 Virl mRNA electroeluted from the high-melting-point agarose and puri-
NSI II 4.9 Kb fied by extracting twice with butanol (saturated with 0.15 M
N82 3.3 Kb NaCl and 10 mM Tris, pH 7.6) and twice with phenol-
chloroform-isoamyl alcohol (25:25: 1). The DNA was etha-
3.0 Kb
nol precipitated and added to standard ligation reactions
VP2 HA|
.~'
en HPs 1|9 P 3.0 Kb (26). The ligated DNA was transformed into DH5 Esche-
II
richia coli cultures as described by Hanahan (22). Low-
pDO
E H On H Psang Pv p6
molecular-weight DNA or DNA in low concentrations was
ligated directly from the low-melting-temperature agarose
IH pNCPV17-SA _ (17). DNA bands were excised from the gel and melted at
65°C
I

PNCPV4-5 for 10 min. The volume was measured, and water was
Hg PNCPV5 added to give a 0.1% solution of melted agarose. Better
H H a19 Ni H results were obtained when the agarose did not gel during the
ligation or transformation procedures. Ligation reactions
were done in 200-pl aliquots by adding 178 1±1 of the melted
ATO agarose solution, 20 RI of 10 x ligation buffer (22), and 2,u
FIG. 1. Transcriptional and cloning map. The CPV genome is
of DNA ligase and incubating the reaction mixtures at 140C
represented in the 3'-.5' orientation and is divided by map for 16 h. Transformations were performed by adding 100 pul
of the ligation mixture to 200 pul of C600 competent cells
coordi-

nates. The major virion mRNAs and sizes are indicated on the right

in kilobases, while the proposed protein each encodes are presented prepared as described by Crouse et al. (17). All transforma-
on the left. The important restriction enzymes in relation to map tion mixtures were plated onto L plates containing 100 pig of
positions used in this study are indicated. E, EcoRI; H, HindIll; Bn, ampicillin per ml. Colonies were picked and screened for
BanI; Ps, PstI; Bg, BgllI; Pv, PvuII. The bottom part of the figure
CPV-specific inserts by dot blot hybridization by using
depicts the regions cloned into various plasmid vectors. The
hatched area indicates the region of CPV-specific DNA
cross-
32P-labeled purified viral DNA (26).
nant clones which displayed deletions or rearrangements.
in recombi-
Cloning strategy for the CPV genome. All designations for
NruI(-) site is the former NruI site used in ligation and is no longer
The

the virus genome orientation are drawn with the 3' end of the
a functional site. Cloned pD8 (blunt end to EcoRI of CPV RF DNA)
minus-strand DNA to the left (2).
was inserted between the EcoRI and PvuII sites of pGEM 2 (29).
The strategy for cloning the entire genome required digest-
Clone pH9 (EcoRI-HindIII of CPV RF DNA) was inserted ing purified CPV RF DNA with HindIll and EcoRI (mp 35
and 47.8 for HindIlI and mp 21 for EcoRI; Fig. 1). HindIll
between

the EcoRI and HindlIl sites of pHC624 (7).

cleaves the RF DNA into three fragments (28). One fragment
contains the major portion of the nonstructural-protein-
Imre Boros, Szeged, Hungary) and pGEM-2 (29; Promega coding region (NS1 and NS2) (mp 0 to 35). The central
Biotec, Madison, Wis.). fragment (mp 35 to 47.8) contains the proposed ATG (mp 44)
Virus and cells. The Norden dog kidney cell line NL-DK1 for the VP1 mRNA and the proposed major splice junctions
(9) was grown in basal medium containing Earle salts and 5% for VP1 and VP2 mRNAs. The third fragment contains the
fetal bovine serum. All virus passages were done on NL- major coding region of the VP1 and VP2 proteins (mp 47.8 to
DK1. The serum was previously irradiated with 2.5 Mrads of 100).
ionizing radiation to inactivate potential pathogens. CPV-N The 3' end (mp 0 to 35) fragment was cleaved with EcoRI
is a high passage of an isolate of CPV which displays a (mp 21) to generate an additional fragment. The 3'-most
lowered virulence. terminal fragment (mp 0 to 21) was cloned into the EcoRI-
DNA isolation. RF DNA was extracted from infected cells to-PvuII site of pGEM 2 (Fig. 1, pD8). The EcoRI-HindIII
by scraping the cells into medium and then centrifuging fragment (mp 21 to 35) was cloned in pHC624 (6) (Fig. 1,
(SS34, 1,500 rpm, 1 min). The cell pellet was suspended in pH9). The central HindIII fragment (mp 35 to 47.8) was
proteinase K (Boehringer Mannheim Biochemicals, India- easily cloned into the HindIII site of pBR322 (Fig. 1,
napolis,Ind.) solution (50 mM Tris hydrochloride [pH 7.5], pNCPV4-5). The 5'-end fragment was cloned into the
1% sodium dodecyl sulfate, 10 mM EDTA, 750 pug of HindIII-to-NruI (blunt-end site at mp 972) site of pBR322
proteinase K per ml) and incubated for 4 to h at 37°C. 5
(Fig. 1, pN67 and pNCPV17-5A).
Sodium chloride was added to a final concentration of 1 M, Sequencing CPV DNA mp 21 to 100. CPV-specific DNAs
and the mixture was incubated overnight at4°C. The chro- isolated from pH9, pNCPV4-5, and pNCPV5 (combination
mosomal DNA and particulate cellular debris were removed of noncrosshatched regions of pNCPV17-5A and pN67 (Fig.
by centrifugation (SW41, 40,000 rpm, 30 min, 4C). Low- 1) were digested with restriction enzymes (HindIlI, HincIl,
molecular-weight DNA and RNA were ethanol precipitated NcoI, AhaI, HpaII, RsaI, and AluI) to create random sets of
from the supernatant and then extracted with phenol-chloro- cloned fragments representing the entire region. The restric-
form-isoamylalcohol (25:25:1). The aqueous phase was lay- tion fragments were then cloned into M13mpl8 or M13mpl9
ered onto a 5 to 20% linear sucrose gradient and centrifuged double-stranded RF DNA by using combinations of appro-
(SW28, 24,000 rpm, 16 h,4°C). Fractions containing the RF priate restriction enzymes. Ligated DNA was transformed
DNA (monitored on a 1% agarose gel by electrophoresis in into JM101 cells, and white plaques were picked for single-
Tris borate buffer) were ethanol precipitated and suspended stranded template preparation. The template DNAs were
in deionized H20. then sequenced by using the dideoxy chain termination
Recombinant DNA and transformation. Purified RF DNA method (45) and a universal forward primer. Template
was digested with appropriate restriction enzymes, and the DNAs were isolated, and the sequence was determined on a
individual bands were separated on either 1% agarose gels minimum of two separate gels. Regions of overlap were
(HMT SeaKem; FMC Corp., Marine Colloids Div., Rock- determined by computer and direct analysis. Restriction
268 REED ET AL. J. VIROL.

enzyme sites at all junctions were identified from the se- progressively shorter regions of the repeated 5' untranslated
quencing gels. region. An AvaI site at nt 1425 of pBR322 was used to initiate
Sequencing of the CPV DNA mp 0 to 21. Plasmid pD8 the BAL 31 digestion. The AvaI site lies 453 bp downstream
contains mp 0 to 21 of CPV in a pGEM 2 plasmid. The CPV from the NruI site used as the blunt-end cloning site. XbaI
insert plus flanking pGEM 2 DNA was isolated on a TaqI linkers were added to generate a unique restriction enzyme
fragment and cloned in both orientations in M13mp19. The 3' site to aid in the manipulation of the pNCPV-2-coding
exonuclease activity of T4 DNA polymerase was then used region. Two clones (p7-1 and p14-14) were isolated from the
to produce a series of overlapping clones from single- transformation of BAL 31-treated DNA, which by size were
stranded DNA templates by using complementary DNA estimated to have the potential full-length, unreiterated 5'
oligomers to form specific restriction enzyme sites (18). The untranslated genomic DNA (Fig. 2). DNA sequence data of
DNA was then transformed into JM101, and template DNA p7-1 and p14-14 revealed that they differed in size by 255 bp
was prepared for sequencing. (Fig. 2). Clone p7-1 contains 58 nt immediately after the
Ambiguities arising in the sequence by using the dideoxy HaeIII site which was not vector DNA. Clone p14-14
chain termination method on the 3'-end palindromic se- contains an additional 255 bp, which begins 51 bp after the
quence were resolved by using Maxam and Gilbert sequenc- HaeIII site. This 255-bp region does not display any char-
ing (27). Complete resolution of the sequence required the acteristics of the 5' palindromic sequences contained within
incorporation of up to 40% (final) formamide in the sequenc- any of the other parvoviruses. Instead, this 255-bp region is
ing-gel mix. a repeat of part of the capsid-coding sequence upstream of
BAL 31 digestion. Plasmid DNA was digested with single- the stop codon (nt 4289 to 4543) (Fig. 2; see Fig. 4). The
site restriction enzymes and then purified by phenol-chloro- 255-bp region is a perfect repeat, except at nt position 4295,
form extraction and ethanol precipitation. The DNA was where a C is inserted to give the repeat at nt 5067 one extra
suspended in water, and 3 to 6 U of BAL 31 enzyme was base.
added in the appropriate buffer (26). The reactions were The untranslated region also displayed DNA sequence
terminated in 1% sodium dodecyl sulfate-10 mM EGTA reiteration observed in other parvoviruses (12). One 62-nt
[ethylene glycol-bis(P-aminoethyl ether)-N,N,N',N'-tetra- repeat begins at nt 4514 and ends at nt 4574 and flanks the
acetic acid] at 67°C. The DNA was phenol-chloroform stop codon at nt 4538. The second 62-nt repeat begins at nt
extracted, and XbaI linkers were added in the presence of 8 4575, ends at nt 4636, and contains a 1-base insertion at nt
U of T4 DNA ligase. After ligation, XbaI was added to digest 4593 and a transition (C to T) at nt 4603 (see Fig. 4). Another
the linker DNA, and the ends were ligated into covalently direct 62-nt repeat begins at nt 4701 and is repeated three
closed circles before transformation into E. coli. times, ending at nt 4886.
Computer program and analysis. The DNA sequence Sequence of the 3' end. The 3' end of the genome was
analysis was performed by using the Beckman Microgenie sequenced from clone pD8 which contains a 1.1-kilobase
on an IBM PC AT and the University of Wisconsin programs insert (mp 0 to 21) between the EcoRI and PvuII site of
with a DEC VAX computer. A sonic digitizer (Science pGEM 2 (Fig. 1). Since no convenient restriction sites were
Accessories Associates) was used with the Microgenie soft- available to allow easy subcloning into M13 vectors, the
ware to transfer sequences to a computer file. whole 1.1-kilobase insert plus some pGem 2-flanking regions
were cloned into M13mpl9. Nested sets of the single-
RESULTS stranded M13-cloned templates were made, starting from the
EcoRI site by using the T4 DNA polymerase 3' exonuclease
DNA sequence of the 5' untranslated region. By digesting technique (18). Two regions of the DNA in the T structure of
CPV RF DNA with EcoRI and HindIll, fragments were the palindrome could not be resolved by the normal dideoxy
generated which contained the blunt end of RF DNA and an chain termination method (Fig. 3). These two areas were
internal cohesive restriction enzyme site. The strategy was resolved by using a combination of Maxam and Gilbert
to directionally clone these fragments into pBR322. How- sequencing and increased denaturing conditions of the gel.
ever, very few of these clones were stable in various strains The 3' palindromic sequence resembles that of the rodent
of E. coli by using a single vector. Three separate cloning parvoviruses (12). The total length was 118 nt, with 92
vectors (pBR322, pGEM 2, and pHC624) were necessary to nucleotides making up the stem structure and 26 nt making
generate stable clones in transformation-competent E. coli. up the T structure (Fig. 3). Two regions of the stem structure
Two plasmids, pN67 and pNCPV17-5A (cloned in pBR322), were conserved between rodent parvoviruses and CPV-N:
were originally thought to contain an intact insert (mp 45.8 to nt 1 to 12 of the rodent and nt 1 to 12 of CPV-N (11 of 12
100) but in fact contained deletions or rearrangements when match) and nt 38 to 48 of CPV match nt 34 to 44 of rodent
analyzed by restriction enzymes (Fig. 1). The nonrearranged parvoviruses (10 of 10 match) (12). The T structure of
or nondeleted regions of pN67 and pNCPV17-5A were CPV-N displayed some divergence, with 18 of 26 nt match-
combined to form pNCPV5. Plasmid pNCPV5 was then ing sequence and location with 18 of 27 nt in rodent parvo-
combined with pNCPV4-5 to reconstruct the entire VP1- and viruses (nt 51 to 62, 65 to 67, and 69 to 71 of CPV-N match
VP2-coding region in pBR322 (Fig. 1, pNCPV-2). nt 49 to 60, 62 to 64, and 66 to 68 of rodent parvoviruses,
The region of the genome from mp 47.8 (HindIII) to the respectively) (12; Fig. 3).
end of the genome was sequenced from clone pNCPV5. This Sequencing mp 21 to 94. Recombinant DNA clones pH9,
clone contained a region larger than anticipated in the 5' pNCPV4-5, and pNCPV2 were used to determine the se-
untranslated region by restriction enzyme analysis. Partial quence of the major body of the genome (Fig. 1). The
sequence data and restriction mapping confirmed that about CPV-specific inserts were isolated from these recombinants
400 to 500 nt had been duplicated in the 5' untranslated and digested with multiple enzymes (see Materials and Meth-
region starting 51 base pairs (bp) from the HaeIII site (mp ods). The digested fragments were cloned into M13mpl8 and
94). By using the clone pNCPV2 which contains the entire M13mpl9 to generate random sets of clones. Computer
VP1- and VP2-coding sequences, BAL 31 enzyme treatment analysis was used to identify regions of overlap, and all
was used to generate a nested set of clones which contained restriction enzyme junctions were positively identified on
VOL. 62, 1988 CPV SEQUENCE 269

100 nt

4280 4380 4460 4580 4660 4780 4880 4980 5080 51610 5280 5323

1L
- Coding Region Untraslated Region

45AA AATAAA

255 5010 255

..
. ..
.
..
....
. .
-.--
4289 4543 5061 5316
62 62
4514 4636

62, 62 62
4701 486
2290 TAA Hoe m 5323
lw

2290 TAA Heem

Ho 5066

FIG. 2. Detailed map of the 5' untranslated region of genomic DNA. Clones p14-14 and p7-1 end in the 3'-terminal region. p14-14 and p7-1
were constructed by using BAL 31 as described in the text.

sequencing gel autoradiograms. Unlike the terminal regions, 1990) contained all the appropriate promoter components.
the internal sequence contained a high percentage of A and The P3.5 promoter of CPV-N is analogous in map position
T residues, which reduced the ambiguities and problems and sequence to other parvovirus promoters and probably
associated with compressions on denaturing gels. The se- initiates transcription of the large ORF A for the CPV
quence and single-letter amino acid translation of CPV-N nonstructural genes (15, 16, 42). The P38 promoter most
(mp 0 to 94) is shown in Fig. 4. likely initiates transcription of ORF B for the structural
Genome organization and assignment of CPV genes. The genes (VP1 and VP2) and is homologous to P38 promoters
potential coding domains for both the C strand (plus polarity) described for other parvoviruses (3, 25, 36, 40, 42). The
and the V strand (minus polarity) are shown in Fig. 5. There remaining six potential promoters were also analyzed for the
are two predominant ORFs, both occurring in frame 2 of the presence of E and A regions. Three of these promoters (P28,
C strand (Fig. 5). The first ORF, A (668 amino acids) is P45, and P94) contained all of the components, while three
probably the coding domain for the nonstructural proteins lacked the E or A region (Table 1).
(NS1 and NS2; 16, 24, 46). ORF B (722 amino acids) is most Two possible poly(A) addition sites were found by using
likely the coding domain for the capsid proteins (VP1 and AATAAA as the consensus sequence. One is in the major
VP2; 16, 24, 41, 46). There are several smaller ORFs in both coding body of ORF A (nt 1580), and the other is located 448
the C and V strands. Whether these small ORFs are of any bp downstream fromn the ORF B stop codon at nt 4538.
significance is not known. However, one small ORF (ORF S) The potential 5' donor sites (consensus sequence,
in frame 1 of the C strand is the end of the 255-nt duplication 'AGGT'AGT) and the 3' acceptor sites [consensus se-
from the coding domain ORF B (described above). This quence, (Py)6XCAGGc] used as mRNA splice junctions (14)
255-nt region is no longer open in frame 2 of the C strand, but are listed in Table 2 and mapped in Fig. 5. Six possible 5'
instead, 201 nt are now open in frame 1 (Fig. 4 and 5). An donor sites were identified. The sequence of these donor
ATG at nt 5110 and a TAA at nt 5311 would allow for a small sites represents regions with greater than 65% homology (six
peptide 34 amino acids long. In addition, a TATA box occurs of nine base pairs match) with the consensus sequence,
at nt 5026, placing a potential cap site at nucleotide position allowing only one mismatch. Three of the donor sites are
5056, 4 nt from the start of the 255-nt duplication. The TATA located in ORF A (nt 294, 306, and 508). One donor lies
at nt 5026 also has a potential E and A region (see below; immediately upstream of the proposed ATG start codon for
Table 1), making it a candidate for a eucaryotic promoter VP1 (nt 2277), while the remaining two (nt 2314 and 4375) are
sequence. By using comparative analysis of previously pub- in ORF B (Fig. 4 and Table 2).
lished data (10, 24, 25, 40, 41) with the data described here,
a more precise assignment of the CPV-coding region was
made. CPV DNA
A computer search for possible promoter regions was
done by using the information on eucaryotic promoters
described by Bensimhon et al. (5). These features include a lqw7.G-70 80 90 100 110 120

stable enabler region (E) approximately 100 bp upstream SO0 _>ACGTACG

EATGCATG A GAACCAGTCAA
T CTTGOTCAGTT
T
from the cap site, a G + C-rich activator region (A) approx- ,T_
O G Al
imately 50 to 75 bp upstream of the cap site, and the TATA
box which usually lies about 30 bp upstream from the cap _A40
30 20 10
site (5). Eight possible promoter sites were characterized by FIG. 3. Structure of the 3' palindromic DNA. Clone pD8 in
using a consensus search for the sequence TATAA (Table plasmid pGem 2 was used to determine the sequence of the 3' end of
1). Two promoters, one at mp 3.5 (P3.5) (Fig. 4, TATAAAA, the genome. Shaded areas are regions which are homologous to
nt 188) and a second at mp 38 (P38) (Fig. 4, TATAAAT, nt rodent parvovirus sequences.
270 REED ET AL. J. VIROL.

16 306 56 70 96 11 123
ATT CTTT ACAAC CAACTG AC CAAGTT CACGTAOCCTATCGACGTGATCACCCG CTGCGCGCGCCCTGCCCTACGG CAGT CACACGTCATACGTACC CTCCTTGGTCAGTTCCTTCTAAAGAATGA

121 136 166 170 19t 21223 246

TACCCCCTTTCTCTCTTTAAACTTCCCCCGCCAAAACGTCCCCCCCTAATTCTCCCCCTCCTTAAACCTATAAAACACAAACCATAC ACCCTTACTCACATTCCCTTCTTCTCTTrTCACAC
NSI
SS C N q Y T 3 8 V M B C V W L K K U A B N 8 A F S 7 V F
241 256 276' 2W 31g 336 36
ACTCAACCTCTCTTACTCTCACTAACCAACCATCTCTCCCAACCAGrATACTCACCAAGTTATCGACCCACTAAATTCCTTAtAAAAACATCCACAAAATCAACCATTTTCGTTTCTTTT
K C D P V Q L N C K D V I T N P Y T K P I Q N B8 L T S L I R C A Q T A M D Q T
361 370 396 410 436 466 476 436
TAAATCTCACAACGTCCAACTAAATGGAAACCATCTTCC CTCCAACAACTATACCAAACCAATTCAAAATCAACAACTAACATCTTTAATTACACCACCACAAACACCAATCCATCAAAC
B B B B M D W e s L V D S L A K 9 Q V Q T r D A L I K t C L P B V F V S K N I B
431
CCAACAAGAACAAATCGACTCCCAATCrCFSTGATACTCTCC CCAAAAAC CAACTACAAACTTTTGATGCATTAATTAAAAAATGTCTTTTTCAACTCTTTCTTTCTAAAAATATACA
P N B C V W r I Q B W C K D Q C W B C 3 V L L I S K N L Q Q A T C K W L I I Q
661 616 636 668 676 696 71? 736
ACCAAATCAATCTCTTTCCTTTATTCAACATCAATCCCCGAAAACATCAACCCTCCCATTGTCATCTTTTACTTCATACTAACAACTTACAACAACCAACTCCTAAATCCCTACCCACACA
M P M Y W S 2 W L V T L C S V N L T P T 9 K I K L * 9 I A B D S9 WiV T I L T Y
721 730 76 776 790 316 3a3 346
6060
AATCAATATCTATTCCACTAGATCCTTGG TGACTCTTTCTTCGCTAAACTTAACACCAACTCAAAACATTAsCrACMACAGATTCCACAACATAGTGAATGGGTGACTATATTAACATA

I 8 9 Q T K I D Y V 9 M V D F C N M I A Y Y r L T I K I I V 5 M T K B S G Y f L
841 356 876 396 910 936 960
CACACATAACCAAACAAAAACACTATGTTAAAATCGTTCATTTTGGAAATATCATACCATATTACTTTTTAACAAAGAAAAAAATTGTCCACATCACAAAACAAAGTGCCTATTTTTT
S T D S C W I r N r M I Y Q D I Q I V 9 T L Y T B Q M X P B T V B T T V T T A Q
961 976 996 1616 1636 165 1676
AACTACTCATTCTCCTTCCAAATTTAACTTTATCAACTATCAACACACACAAATTCTCACCACACTTTACACTCAACAAATCAAACCACAAACCCTTCAAACCACACTCACCACACCACA

R T I R C R I Q T K I 8 V S I K C T L R D L V 9 K I V T S P 9 D W M M L Q P D 9
Id1 1090 1116 1130 1160 1170 1196 1366
CCGAAACAAACCCCGCGCGAGMTTCAACTAAAAACCA ACTCTCAATCAAATGTACTTTCCCCCGACTTCCTTACTAAAACAC TAACATCAC CTCAAC ACTGCATCATCTTACAACCAGATAC
066@
Y I R M M A Q P C C 8 N L L I P T L B I C T L T L A R T I T A F B L I L e K A D
1201 1216 1236 126 1270
1316 1290 1336
TTATATTGAAATCATCCCACAACCAG eACCTGAAAATCTTTTAAAAAATACACTTGAAATTTWGTACTTTCACTTTACCAACAACAAAAACAG CATTTGAATTAATACTTCAAAAACCAGA
1636
11 T * L T PI D L A R 9 R T C Q I F R M 8 c W N W I K V C 11 A I A C V L 11 I Q G
1321 1336 1356 1396 1370 1416 1436
TAATACTAAACTAACTAACTTTCATCTTCCAAATTCrAGAACATCTCAAATTTTTACAATCCACGGATCCAATTCCATTAA.ACTTTCTCACCCTATACCATGTGTTTTAAATACACAAGC
3446

C I R 11 T V L r 8 c P A S T C K S T I A Q A T A Q A V C N1 V C C Y 11 A A 11 V 11 P
1441 1460 1476 1496 1161 1536 1556 1566
TGCTAAAACAAATACAG TTCTTTTTCATCCACCACCAACrACACCAAAATCTATCATTGCTCAACCCATACCACAsACCCTCCCTAATCTTCCTTGTTATAATCCACCAAATGTAAATTT

P r N D C T N I 11 L I W I e e A C 11 F C Q Q V N1 Q P K A 1,C S0¢ Q T I I I D Q K
1661 1676 1690 1610 163s 1650 1676 1630
TCCATTTAATCACTarIACCAATAAAAATTTAATMCCATTGAACAACsCCTAACTTTCCIWCACAACTTAATCAATTAAACCAATTTCTTCTCCACAAACAATTACAATTCATCAAAA
y&- I1{- lllnLcMI
C K C S 9 Q I e P T P v I 11 T T 11 1 N I T I V R I C C It B I P * 9 T Q P I I D R
1631 1696 1716 1736 1760 1776 176
ACCTAAACCAACTAACCAAATTCAACCAACTCCACTAATTATCACAACTAATCAAAATATAACAATTGTCACAATTCCATCTCAACAAACACCTCAACATACACAACCAATAACACACAC

M L 11 I I L V C K L P c D r c L V D I B R W P L I C A W L V KI 1 C Y 9 7T A N1
1361 1616 1836 1 36 1870 130 1916 1936
AATCTTGAACATTAACTTACTATC TAAGICCACCACACTTTCGTTTGGTTGATMAAAMACATCOOCTTAATATGTCCATGGTTACTTAAACATGGTTATGAATCAACCATCCCTAA
Alul
Y T R R W C K V P e W D 1C N W A B P K I Q B C I N S P C C K D L B T Q A A 9 N P
1921 1930 1956 1976 1996 2316 2636 3646
CTATACACAT CATTCCC CAAAACTACC ACAATC CC ATCAAAA CTCCCCCCAC CCTAAAATAC AA AACClHaA?TCACCACCTTC CAAACACTTACACACACAACCCCCAACCAATCC
Q S Q D Q V L T P L T P D V V D L A L R P W s T P D T P I A C T A N Q Q 9 N Q L
2041 2656 2676 2096 2116 2136 2160 3166
TCACACTCAACAC CMCGTTCT^AACTC CTCTCAC TCOCGACGTACTCC ACCTTC CACTCCAMC CC TCGCTACCTC CACATACC CCTATTC CACAAACTC CAAMTCAACAATCAAMCCAACT
H Ip-TI lIiT
C V T 3 K D V Q A S P T W S e I B A D L I A I F T S B Q L 3 3 D F R D D L D
2161 3170 ACAAACACCTC CA.AC 2190
TCCCCGTTACTC 2316 2330 32560 2276
CACTCCCACGTGCTC CCAAATACACCCACACCTCACACCCATCTTTACTTCTCAACAATTCGAAGACATTTTrCsaACGACTTCCATTAACG -
3336

VP1A TsT
M" A P P A K R A23!R G K C v L V K W C B G K D L I T M C Fr I G L V P P
2281 22Z90 231 230 2360 2387 2ON 3466
;zACGI~TCCCACCTCOOCCAAACACAC CCACCACACCTAACCCTCTCTTACTAAACTCCCGCCCACCCCAAACAM'TAATAACTTAACTAACTATCTCTTTTTmATACCACTCTGCCCTCC

C Y K Y L C P c N S L D Q C * P T N P S D A A A I 3 I D 9 A Y A A Y L I 9 G K N
2461 2410 24360 46 2470 2406 3616 2536
AGGTTATAAATATCTTGGGOCTGCCAACAGTCTTGACCAAGACACMCAACTAACCCTTCTCACCCCCCTCCAAACMACACCACGAAGITACCCTGCTTATCTTCCCTCTGCTAAAAA
I 1aeTI I TiadmII
Alul
P Y L Y Ir S P A D Q RP I D Q T I D A 9 D W C C I I C * Y P P I A I I A I A P V
2621 2630 2660 2670 2690 2610 2636 3646
CCCATACTTATATTTCTCCGCCAC C ACATCAACC CTTTATACATCAAA CTAACCACG CTAAACATTCCCCCCCCAAAATACCACATTATTTTTTTSACAarAAAAACCCAATTCCTCCACT
T XI-UT
L T D T P D 11 P s T 8 R P T 9 P T I R 8 K P P P * I P I N1 L A I I I 9 A G A a Q
3641
3656 3676
ATTAACTCATACACCACATCATCCATCAACATCAA CACCAACAA 2690
AACCAA CTAAA MAGTAAA 3710 3736 TCTTCCAAAAAAAAAOG;CCAGCACA
MCCACCACCTCATATTTTCATCAA 3766 3766

VP2
V I R D N L A PIM 8 D C A V Q P D C C Q P A V R N B I A T C 9 C N C S 00C C G C
2761 2776 4 2706 2616 33836 3360 3370 3336
ACTAAAAMAACAAMTCTTGCACC jATCAGTGATCCACCAGTTCAACCAGACCCTGCTCIAACCTCCTCTCAGAAATCJACACAwrcACATCTOCCAACCCCTCTOCACGOCGGGGGTG
a EiITT- fIT-U
C C S C C V C I S T C T F NN Q T B F Ir L 3 1 G W V 3 I T A N 3S R L V I L N
2881 2896 2916 2930 9560 2970 3906 S666
TCCTCCTTCTCCCCCTCTCCCCATTTCTACCCCACTTTCAATAFTCACACCC AATTTAAATTTTT4CAAAACCCATCCCTCCAAATCACACCAAACTCAACCAoACTTtrACAMiA
FIG. 4.-Continued.
VOL. 62, 1988 CPV SEQUENCE 271

N P B B NfY I I V V V Nf K T A Vf C A L, D A Q I V T P S L V
391 3616 3603 3060 3076 3 O9
S 3110 312
TATCCCACAAACTCAAAATTATACAACACTCGTTCTAAATAATATGCATAAAAsCIrGMTACCCAAACATCCCTTTACATCATATTCATCCACAAATTCTAACACCTT¢CTCATMT¢
K PutI K T V
linell
D A N A W C V * P N1F C D W Q L I V i T- 3S L 8 L V 9 F 3 Q 3 I F N V V L I T
3121 .3136 3160 3.176 3196 3210 3236 3246
TCATCCAAATCCTTCCCCACTTTCCTTTAATCCACCACATTCCCAACTAATTCTTAATACTATCACTCACTTCCATTTACTTACTTTTCAACAACAAJUTTTTAATCTTCTTTTAAACAC
v s e 9 A T Q P P T K V T N N D L T A S L I V A L D 9 11 N T 1S P 1 T P A A N I 9
3241 3266 3270 3296 3310 3336 3366 3366
TGTTTCACAATCTCCTACTCACCCACCAACTAAACTTTATAATAATGATTTAACTCCATCATTCATCC TTCCATTACATACTAATAATACTATCCCATTTACTCCACcMArATcAaA7lC

B T L C Ft Y P W 1 P T I i T P W 4 Y Y P q W D R T L I P S N T C T S C T P T N I
3631 3370 3396 3416 3436 3466 3476
TCACACATTCCCTTTTTATCCATCCAAACCAACCATACCAACTCCATCCACATATTATTTTCAATCCCATACAACATTAATACCATCTCATACTCCAACTACTCCCACACCAACAAAkTAT
3486
T B C T D P D D V q F Y T I B N S V P V N L L I T C D 9 t A T C T P t P D C I P
3481 34906 3516 3536 35660 357 3596 366
ATACCATcGIrAciCATCCACATCATCTTCAATTTTATACTATTCAAAATTCTCTC CCACTACACTTACTAACAACACCTCATCAATTCCTACACCAACATTTTTTmTCATTCTAAACC

C I L .T B T W Q T- I A L C L P P P L N S L P Q S 9 C A T N P C D I C V Q Q D I
S661 3616 3636 3656 3670 3696 3716 3739
ATGTAGACTAACACATACATCGCAAACAAATAGACCATTGCCCTTACCACCATTTCTAAATTCTTTCCCTCAATCTCAACCAWACTAACTTCGTCATATACCACTTCAACAACATAA

I I CV T Q S C N.T N Y I T I A T I B I F A I V C Y S A P T Y s P I A 3 T q C P
3721 3730 3756 3770 3796 3316 33a3
AACACCTCCTCTAACTCAAATCCCAAATACAAACTATATTACTcAACrACTATTATCACACCAaclCACCTTCCTTATACTCCACCATATTATTCTTTCACCCoTarAcAAAl:aO
3346
D P - luI I-we 1-TII
Alul
F K T P I A A C I C C A Q T Y1 B N1 q A A D C D P I Y A F C Q I C q A T T T T C
3541 336 3676 3696 3910 3936 3696 3966
ATTTA"AAACCCTATTCCACCACCACCCCCCCCACCCCAAACATATGCAAAATCAACCACCACATCCTCATCCAACATATCCATTTCMAGACAACATCCTCAAAAAACTACCACAACACC
D
s T P S I F T Y I A B Q D T C I T P 9 c D W I 4 N1 I 11 F N L P V T 11 D N V L L P
3061 3970 3996 4616 4063 4066 4670 4090
ACAAACACCTCAGAGATTTACATATATAC CACATCAACATACACCAAGATATCCAGAACCACATTCGATTCAAAATATTAiACTTTAACCTTCCTCTAAWCCAATCATAJATCTATTCCTACC
A
T D P I C C K T C I 11 Y T N1 I r N T Yt C P L T A L N N V P P v Y P N c Q I W D K
4631 4696 4116 4136 4166 4176 4196 4326
AACACATCCAATTCCACCTAAAACACCAATTAACTATACTAATATATTTAATACTTATCC?CCTTAACTCCATTAAATAATOTACCACCACTTTATCCAAATCCTCAAATTTCCCATAA

B F D t D L K r R 115 V N A P F V C Q N1 N C P C Q L F V K V A P 11 L T N B Y D P
4201 4310 4236 4260 4270 4296 4310 4336
AGAATTTGATACTCACTTAAAACCAACACTTCATGTAAATCCACCATTTCTTTCTCAAAATAATTCTCCTCCTCAATTATTTCTAAAACTTCCGCCCTAATTAACA AATCAATATCATCC
D A 9 A 11 11 S I I Vf T Yr S D F W W K C I L V P I A K L R A S 9 T W 11 P I q Q N 9
4321 4136 4366 4370 4396 4416 4430 4440
TCATCCATCTCCTAATATCTCAACAATTCTAACTTACTCACATTTTTCCTCCtAAACCTAAATTACTAl TTAAAwrAAACTAACACCCTCTCATACTTCCAATCCAATTCAACAAATCAC
flil
I 16 V D N 11 F N1 Y V P s P I G C V K I V Y B K S Q L A P R K L Yr
4441 4460 4476 4496 4616 4536 4566
TATTAATGTAGATAACCAATTTAACTATGTACCAACTAATATTGGACGTATCAAAATTGTATATCAAAAATCTCAACTACCACCTACAJUATTATATTAACATACTTACTATCCTTTTTA
46"
T& N A
L
4681 4670 4596 4616 4830 4668 4676 4683
TCTTtATTACATATCAACTACCACCTACAAAAATTATATTAATATACTTACTATCCii TiiATCTTTATTACATATTATTTTAACATTAATTAAATACA CCATAGAAATTCUTTAT

4681 4696 471t 4736 4756 4776 4796 4366

ATTTCATATACCATTTAGAAGTTTCTTAGATCCITATACAATAACTCTAACAAATACAAGAACATTTACATCATACT?ACTACCTTTTTACATCCATACATAACTCTAACAAATACAA

4831 4816 4833 486 4376 4396 4916 4920

CAACATTTACATCATACTTACTACCTTCTTATATCGTATACAATAACTCTAACAAATACAACAACATTTACATCATACTTACTACmTTTTTTATAAAATCTATTCTAAACCATTAATC
4921 4936 4960 4976 4996 5616 5636 9646
TATGTTCTTATGCTGTCGCTGCTTGCTTGCTTTCCCCTTACJAATATCTTAAGGACCAAAAAATCAATAAAACACATTTAAAACTAAAS C GCTAvTJACSTCTATAACCTCAACTAA
IitI AccI
6041 666 5676 696 511. 5a13 51560 5166
CCTTACCAtAACTATCAATCGTTCCGCCCCTAATTTAACAAATCAATATCATCCTCATCCATCTGCCtAATATCTCAACAATTCTAACTTACTCACATTSTCTCCTGAAACCTAAATTACTA
6161 5170 6196 0 6218 5230 626 6576 652
TTTAIAArAAACTAACACCCTCTCATACTTTAATCCAATTCAACAAATCACTATTAATCTATATAACCAATAACTATSACCAACTAATATT'CAQTATCAAAATTCTATATGAA
rFTw-
5281 6290 6531
AAATCTCAACTACCACCTACAAAATTATATTAACATCTCTAGA
Thul

FIG. 4. DNA sequence of the CPV complementary strand. The single-stranded nucleotide sequence of the C strand is shown with the
one-letter amino acid translation. Below the CPV nucleotide sequence is the one-letter amino acid letter indicating the changes that occur
between FPV (10) and CPV. Overlined regions indicate the proposed mRNA 5' donor splice junction, and the 3' acceptor splice junctions are
underlined.

A total of 16 possible 3' acceptor regions were identified (amino acids 352 to 516 of CPV have 90% or greater
(Table 2); however, only 5 of these sites demonstrated better homology with amino acids 350 to 514 of Hi and amino acids
than 65% homology (8 of 12 match) with the consensus 399 to 564 of MVM parvovirus). Incomplete sequence data
sequence (14) and retained the core CAGG region (Table 2). of the FPV genome did not allow us to compare amino acid
Two of the five high-homology 3' acceptor regions are in homologies of the nonstructural genes of CPV-N and FPV.
ORF A, and three are in ORF B (Fig. 5). However, the sequence for the structural genes was avail-
Sequence homology of CPV and other parvoviruses. The able and is compared in Table 4 and Fig. 6. By hydropathy
homology between CPV-N, CPV-b, FPV, and other parvo- analysis (University of Wisconsin computer programs),
viruses is shown in Table 3. CPV-N by both amino acid and CPV-N and FPV show very little difference between their
nucleotide homology is more analogous to FPV than to respective structural protein-coding regions.
CPV-b. Homology with MVM and Hi is lower, and very The amino acid differences among FPV, CPV-N, and
little homology exists between CPV-N and the human par- CPV-b are shown in Table 4. Only by close analysis can any
vovirus B19 (46). Certain stretches of DNA sequence within of these changes be readily apparent in the hydropathy chart
the nonstructural (NS1) coding region are highly conserved (Fig. 6). There are only 13 amino acid changes between the
between all parvoviruses except B19 human parvovirus CPV-N and FPV capsid genes (Table 4 and Fig. 4), while
272 REED ET AL. J. VIROL.

TATA
AATAAA
DNA were cloned by using an internal restriction enzyme
1
site and the blunt end of the RF DNA in a directional cloning
protocol. The 5' half of the genome was cloned into the
HindIII and NruI sites of pBR322, and the 3' half was cloned
in the EcoRI and PvuII sites of pGem 2. Restriction map and
11 s
EL--i
DNA sequence analysis of the 3' end sequence indicated that
the 3'-end palindromic sequence was cloned intact and
2j II A B ||||||| C resembles the T-stem structure of other rodent parvoviruses
(12). The 5'-end palindromic sequence was not found intact
3 I IIIIII II 11111 1 1 1 1 1 1 1 1 111I - in any of the clones sequenced. Instead, a 255-nt sequence
representing the carboxy terminus of the VP1-VP2-coding
region had been duplicated 51 nt downstream from the
HaeIII site. The duplication may have arisen as an anomaly
II1 of the cloning in E. coli, which was needed to stabilize the
clone. However, this sequence reiteration may be caused by
high passage of CPV-N in canine cells. At least two forms of
RF DNA (differing by approximately 500 bp) have been
4
observed in high-passage CPV-infected cells (data not
5
11111111I11I 11 1111111111 1I1
1 1 1 Ihl 111 1111111 111 1 shown). The 255-nt sequence contains 201 nt open in frame
1 (Fig. 5), and an upstream TATA sequence at 5026 (40 nt

6
1 1 1 1 1 1 11111111 1111 11111111111111 1 1 1 1 1111I 1 1 1VI downstream of the poly(A) addition sequence (Table 1)
could serve as a promoter for this small ORF S. The
significance of the 255-bp duplication and its potential coding
capacity were not examined for this report, but it is inter-
I I
1
.
2 3 4
I
esting that FPV also contains a similar TATA-type sequence
I KbpI 40 nt downstream from the poly(A) addition signal (10). By
comparison, Hi parvovirus contains a TATT sequence 34 nt
FIG. 5. Genomic organization of the viral (V) (minus-polarity) downstream of the poly(A) addition sequence, and MVM
and complementary (C) (plus-polarity) strand. Each line designates contains a TATT sequence 37 nt downstream of the poly(A)
the stop codon position of each frame in the viral and complemen- addition sequence (4, 42). A more striking homology is a
tary strands. At the top of the figure is the map of the consensus G-rich region which lies upstream of the poly(A) sequence.
sequence search for promoter-type elements (TATA) (upper) and
poly(A) addition signal (AATAAA) (lower). Arrowheads indicate In CPV-N the sequence GGGTGGTTGG occurs starting 49
the major TATA and AATAAA elements proposed for CPV mRNA nt upstream of the poly(A) addition sequence. The same
synthesis. The center part of the figure shows the map for the sequence occurs upstream of the poly(A) addition sequence
consensus sequence search for the 5' donor ('AGGT'AGT) (upper) for MVM, Hi, FPV, and CPV-b at a distance of 50, 48, 49,
and the 3' acceptor [(Py)6XCAGGc) (lower) splice junctions. Ar- and 49 nt, respectively (4, 10, 41, 42). This sequence is
rowheads indicate the proposed major mRNA splice sites. located approximately 80 to 100 nt upstream from the
potential TATA sequence described above and is similar to
there are 22 changes between FPV and CPV-b. A total of 10 the E region for eucaryotic promotors (5). Recombinant
of the amino acid changes are identical for both CPV-N and clone pNCPV-2 originally contained 400 to 500 additional nt
CPV-b (Table 4, footnote c). which were removed by BAL 31 treatment. Clones p7-1 and
The nucleotide differences among these viruses are also p14-14, generated in this manner, were chosen for examina-
very limited; only 47 nucleotides differ between FPV and tion because they approximated the expected length of the
CPV-N (mp 20 to 96 were compared). These results indicate full genome of CPV. It is now of interest to examine the
that even high-passage virus is homologous in sequence to sequence of the 400- to 500-bp duplicated region to deter-
FPV or CPV-b. The most prominent differences are located mine if any additional ORFs exist and specifically to deter-
in the 5' untranslated region and are changes in reiteration of mine if the poly(A) addition sequence is duplicated to
base sequence rather than changes in the sequence itself provide ORF S with a potential polyadenylation signal.
(Fig. 2 and 4). Although the complete genomic sequence has not been
DISCUSSION determined for CPV-b and FPV, the available data suggest
The entire sequence of the CPV-N genome was deter- that the coding domains for the nonstructural and structural
mined from clones made from RF DNA. The ends of the RF genes are in the same reading frame (46). Based on computer

TABLE 1. Analysis of potential CPV-N promoter sequences

Promoter' Enabling sequenceb Activatorb TATAc
P3.5 GGGCGGGA (145; -73) GGCGGGCA (159; -59) TATAAAA (188)
P28 GGACCAGC (1469; -99) GCTGTGGG (1517; -51) TAT AATG (1538)
P32 TAT AAC A (1738)
P38 GGGGAAAA (1935; -85) GGGCGGAG (1965; -55) TATAAAT (1990)
P45 GGGGGGAG (2337; -97) GTGCCTCC (2393; -41) TATAAAT (2404)
P63 GCCACCAA (3262; -44) TAT AAT A (3278)
P91 TAT AAAA (4894)
P94 GGTGTGGG (4932; -124) GGCCTCGT (5010; -46) TATAAGG (5026)
a The map position of the promoter is indicated.
b
The first number in parentheses is the position of the sequence. The second number is the position from the cap site.
c The TATA is assumed to be 30 bp upstream from the mRNA cap site. The number in parentheses is the position of the sequence.
VOL. 62, 1988 CPV SEQUENCE 273

TABLE 2. Location of consensus 3' acceptor and 5' donor sitesa

5' Donors 3' Acceptors

Sequence
Sequence
LoSequence
~~~~~~~~~~(nt)
Seune(nt)
Location Lation

AGG-AAGT ....................................... 294 GACAGCACAGGA ..1079

AGG-GAGT ....................................... 306 GCACAACCAGGA ..1224
GG-AAGT ......................................... 508 GCAAGTACAGGA ..1482
AAGGTACG ....................................... 2277 AAGCTTCCAGGA .1833
AGGTAAG ........................................ 2314 AATTCACCAGGT .2001
AAGGTAAA ....................................... 4375 AGAGAGCCAGGA.2308
GTGCCTCCAGGT .............. 2400
GCCGGTGCAGGA .............. 2754
AGAGCTACAGGA .............. 2847
TTTAATCCAGGA .............. 3153
ECTAAGAACAGGT ............... 3555
TTTGCTACAGGA ............... 3573
ATTGCAGCAGGA ............... 3861
ACCACAACAGGA ............... 3957
CAAGATACAGGA ............... 4002
lGGTAAAACAGGA ................. 104
"The
~~~~~~~~~A
A
The consensus sequence for the 5' donor was CAGGGTGAGT; the consensus sequence for the 3' acceptor was (Py)6XCAGGT2 (14). Of the nine possible
nucleotides in the 5' donor the computer searched for a minimum of one mismatch in at least six of nine nucleotides. No loop-outs were allowed. The computer
searched for a minimum of three mismatches with no loop-outs in at least 7 of the 12 possible nucleotides in the 3' acceptor. The regions underlined represent nu-
cleotide-matching consensus sequences.

analysis of the potential reading frames in both the C and V nt 2314 and a 3' receptor at nt 2400 (Table 2 and Fig. 5). This
strands of the virus, the coding domains do in fact lie within RNA would be analogous to MVM mRNA Rl, which is
the same frame of CPV-N (Fig. 5). ORF A contains a translated to make the NS1 protein. The 5' donor at nt 508
promoter element as expected at mp 3.5 (nt 188) and a (nt 508 showed greater homology to the 5' consensus se-
poly(A) addition site at nt 1580 (Fig. 5). The purpose of this
poly(A) site is not known; however, there were no new
mRNA transcripts detected on Northern gels (RNA blots) by TABLE 4. Comparison of amino acid charges between proposed
using CPV-specific probes (data not shown). It is predicted capsid protein sequence of FPV, CPV-N, and CPV-b
that all the CPV mRNAs have a coterminus at the poly(A) Amino acid FPV CPV-N CPV-b
site at nt 4985. ORF B (Fig. 5) contains a promoter element no.' 780929b
at nt 1990 (mp 38). These two promoters (mp 3.5 and 38) are 690 Asn Asn Lys
analogous to those observed in other parvoviruses (3, 15, 16, 741 Thr Lysc Lys
25, 36, 40, 42). The abundance of other promoter sites within 807 Gln Leuc Leu
the CPV sequence was not expected, although multiple 887 Asn Asn Ser
promoter sites do exist in bovine parvovirus (15). 888 Tyr Tyr Glu
The consensus sequence computer search for potential 5' 889 Lys Arg Lys
donor and 3' acceptor sites for CPV mRNA revealed sites 890 Deleted Deleted Asp
which are analogous to the locations mapped for MVM 891 Deleted Deleted Arg
mRNA splice junctions. Recent reports suggest that MVM 902 Lys Asnc Asn
910 Thr Ilec Ile
utilizes three separate splice junctions, using all three ORFs 912 Val Alac Ala
of the C strand to regulate synthesis of the four main mRNAs 927 Gly Gly Asp
(24, 33). By analogy to MVM, the predicted mRNA species 1118 Gln Gln Pro
for CPV-N may include an mRNA which uses a 5' donor at 1122 Arg Arg Lys
1132 Asp Asnc Asn
1164 Ile Thr Ile
TABLE 3. Percent nucleotide and amino acid homology between 1165 Pro Pro Deleted
CPV-N and other parvoviruses 1166 Ile Ile Deleted
% Homology with CPV-N 1176 Asp Tyr Asp
Virus strain Amino acid
1195 Gln Gln Glu
(relevant mp and reference) Nucleotide 1220 Ala Gluc Glu
NS1 and 2 VP1 and 2 1242 Thr Thr Ile
1311 Val Val Leu
CPV 780929 (mp 33-95; 41) 95 NDa 98 1371 Leu Valc Val
FPV (mp 20-96; 10) 98 99 99 1373 Asn Serc Ser
B19 human 22 23 20 1377 Val Glyc Gly
Hi 62 73b 53 1389 Pro Pro Gly
MVMi 47 73b 54
MVMp 47 73b 53 aAmino acid number is indicated in Fig. 6. The initiation codon for VP1 is
amino acid 671, and for VP2 it is amino acid 810. The nucleotide numbers for
a ND, Not determined in publication. VP1 and VP2 start codons are 2285 and 2786, respectively (Fig. 4).
bSpecific regions between mp 20 and 30 of the NS1 protein of Hi, MVMi, b The sequence of the CPV-b capsid genes published previously was that of
MVMp, and CPV-d are highly homologous (90%). This homology is not strain 780929 (Cornell University) (41).
observed with the human strain. cAmino acid changes common to both CPV strains.
274 REED ET AL. J. VIROL.

FPV
304 500 800 1100 1392
1 1 1 1 I 1 ---II-
I I I I a I I I a

-Aa.,Jv.v
A.- I - yaAAA- HPhobic

.v
...
-
-.
..
-.

old, V-7w v VYWIVAIV S/AIVVJI"%f

- -. . ..--
i
Lj. VNV,AmAvvv
' .
O - kj
....
m
..--. t - &d
v v VTWN
&- -A A-
-2
HPhilic

-A " A
HPhobic

A AI A , , AA- , IL h ..A - . &A L. .,

yA
a h AA

[
|,,, , , \ / / v W 1f, 2 HPhillc

304 Soo 1100 1392

Amino Acid Residue
CPV-N
FIG. 6. Hydropathy analysis of CPV and FPV. The amino acid sequence for CPV and the available sequence for FPV were compared for
hydrophobicity (HPhobic) and hyodrophilicity (HPhilic) by using University of Wisconsin computer programs. The available sequence for
FPV is equivalent to number 304 of the CPV NS1 gene. There is a 6-nt untranslated region between the stop codon for NS1 (amino acid 670)
and the start of VP1 (amino acid 672) (Fig. 4).

quence than did nt 532 of MVM [23]) and the 3' acceptor at available FPV sequence (10). The entire structural gene
nt 1833 for CPV-N (Table 2) are likely candidates for NS2 region is represented by amino acids 672 to 1392 (Fig. 6). The
mRNA coordinates. Unlike MVM, these messages must hydrophilic and hydrophobic character of the FPV and CPV
splice within the same frame since no other ORF exists on are almost identical. Within the capsid gene region of FPV
the C strand. The splice used to generate the mRNA and CPV-N (Fig. 6, amino acids 672 to 1392) there are only
expressing VP2 in CPV-N most likely utilizes the donor at nt 13 amino acid differences (Table 4). Some of these amino
2277 and the acceptor at nt 2400 (41). This would be acid differences could cause a charge difference, but only
analogous to mRNA R3 of MVM (24) and is designed to close analysis will reveal any change in hydropathy (Table 4
remove the ATG of VP1 at nt 2285 (Fig. 4). VP1 mRNA for and Fig. 6). There are 22 amino acid changes between FPV
CPV probably uses the same 3' acceptor at nt 2400 but uses and CPV-b in the capsid gene region (Table 4), and 10 of
an alternate 5' donor at nt 2314 to allow the VP1 ATG to these amino acids are shared with CPV-N. Four amino acid
remain intact (Fig. 4 and 5). The MVM VP1 mRNA (R3') changes occur between nt 3054 and 3783 (PstI-PvuII) of
uses splice junctions at similar map positions but also splices CPV-N, which has been indicated as an important region for
across two reading frames (24). These predicted splice determining antigen and host range specificity of the virus
junctions can be verified by Si mapping and cDNA sequenc- (34). CPV-b has seven amino acid changes in this region,
ing. The significance of additional donor and acceptor sites four of which are the same as those in CPV-N. Since the
(Table 2) is not known. However, since no new RNA species capsid genes are important for antigenic and infectivity
have been detected by using Northern blot analysis (data not characteristics (31, 32), these amino acid changes may be of
shown), these additional donor and acceptor sites are prob- great significance for determining a possible mechanism for
ably not major splice sites. capsid formation and species specificity (8, 19, 20).
The total DNA sequence data for CPV-b, CPV-N, and The homology comparison and genomic organization of
FPV are translated into percent homologies in Table 3. CPV-N and FPV demonstrate their close genetic identity.
CPV-N displays greater than 98% homology with FPV in However, the two viruses can be distinguished by antigenic
both nucleotide and amino acid sequence (Table 3 and Fig. and host range specificity. Therefore, it is evident that only
4). The rodent parvoviruses display various degrees of a few amino acid differences are critical for altering the
homology at the nucleotide level; however, the amino acids specificity of these viruses. The only other apparent differ-
match more closely (Table 1). Certain regions within the ence between CPV-N and FPV is in the reiteration of DNA
NS1 protein are highly conserved between FPV, CPV, and sequence within the 5' untranslated region of the genome.
the rodent parvoviruses. Amino acids 352 to 516 of CPV are There are two separate and unrelated 62-nt repeats in the 5'
90% homologous with amino acids 350 to 514 of Hi parvo- untranslated region of CPV-N. One flanks the stop codon at
virus and amino acids 399 to 564 of the MVM strains. Human nt 4538 and has been reported in other parvovirus sequences
B19 parvovirus displayed very little homology at the nucle- (3, 10, 41). The other 62-nt repeat (repeated three times) is
otide level or in genomic organization. However, a small located 65 nt downstream from the end of the first repeat
consensus region in the NS1 gene of B19, MVM, and (Fig. 2) and ends 100 bp upstream from the poly(A) addition
adeno-associated virus type 2 is also partially conserved in site (nt 4986). Beginning 66 nt downstream from the end of
CPV (amino acids 346 to 373) and is postulated to be the poly(A) addition site (nt 5061), a 255-bp duplication (nt
homologous to a domain in ATPase-like proteins of other 4289 to 4543) occurs. The purpose of this duplication is not
viruses (1). Homology comparisons for bovine parvovirus known, but details of the region are described above. Reit-
were not done, but probably bovine parvovirus is also eration of DNA sequence in the 5' untranslated region has
distantly related to CPV and FPV (15). been reported for variants of Hi parvovirus. These Hi
By using the available sequence data for FPV (11) and variants were isolated by high passage of the virus in a
CPV-N, a hydropathy chart was generated. The N-terminal transformed cell line (simian virus 40-transformed human
region of the FPV nonstructural gene region is not available, newborn kidney cells). A reiteration of a 55-nt region up-
but the C-terminal portion is represented by amino acids 304 stream of the poly(A) addition site (12, 38, 39) is comparable
to 670 (Fig. 6). Amino acid 304 is the position assigned to the in map positions to the reiterations observed in the 5'
proposed NS1 protein in ORF A of CPV-N (Fig. 5) and untranslated region of CPV-N. DNA sequence from CPV-b
represents the first available amino acid position from the (41) does not have the same reiteration as CPV-N does. This
VOL. 62, 1988 CPV SEQUENCE 275

may be the result of repeated passage, since CPV-N was and W. Hahn. 1985. Cloning and sequence of DNA encoding
sequenced from a high passage of a CPV isolate. As stated structural proteins of the autonomous parvovirus feline panleu-
above, these reiterations and duplications may have arisen kopenia virus. J. Virol. 55:574-582.
as a result of the cloning protocol. The significance of these 11. Carmichael, I. F., and L. N. Binn. 1981. New canine infections.
reiterations may be important in determining the events that Adv. Vet. Sci. 1981:1-37.
12. Carter, B. J. 1984. Variant on defective interfering parvovirus,
occur in the formation of altered or variant viruses. The p. 209-258. In K. I. Berns (ed.), The parvoviruses. Plenum
reiteration of the 5' untranslated region may be utilized Publishing Corp., New York.
differently during the packaging event. By using specific 13. Carter, B. J., C. A. Laughlin, and C. J. Marcuss-Sekura. 1984.
probes from cloned DNA, it will be possible to determine if Parvovirus transcription, p. 209-258. In K. I. Berns (ed.), The
high-passage RF DNA contains the same reiterations. parvoviruses. Plenum Publishing Corp., New York.
The strong homology of CPV-N with FPV supports the 14. Cech, T. R. 1983. RNA splicing: three themes with variations.
hypothesis that CPV arose as a variant of FPV. If it is Cell 34:713-716.
considered that CPV is an FPV variant replicating in dog 15. Chen, K. C., B. C. Shuli, E. A. Moses, M. Lederman, E. R.
cells (a semipermissive host), then the reiteration of the 5' Stout, and R. C. Bates. 1986. Complete nucleotide sequence and
genome organization of bovine parvovirus. J. Virol. 60:1085-
untranslated DNA of high-passage CPV is not unlike that 1097.
observed for variant Hi parvovirus. CPV, as a variant of 16. Cotmore, S. F., and P. Tattersall. 1986. Organization of non-
FPV, may undergo similar genomic changes after multiple structural genes of the autonomous parvovirus minute virus of
exposures to dogs in the natural population. Reports have mice. J. Virol. 58:724-732.
already been published indicating that variant forms of CPV 17. Crouse, G. F., A. Frischant, and H. Lehrach. 1983. An inte-
can be isolated from the natural dog population. Some grated and simplified approach to cloning into plasmid and
variants can be distinguished from previous isolates made in single stranded phages. Methods Enzymol. 101:78-89.
1978 (35). We have evidence (unpublished) which suggests 18. Dale, R. M. K., B. A. McClure, and J. P. Houchins. 1985. A
that high-passage CPV has a unique distribution of capsid rapid single-stranded cloning strategy for producing a sequential
series of overlapping clones for use in DNA sequencing: appli-
proteins in purified virions. Our laboratory is interested in cation to sequencing the corn mitochondrial 18S RDNA. Plas-
the dependence of virus capsid processing on DNA template mid 13:31-40.
determinants as a possible mechanism for inducing altered 19. Dietzschold, B., W. H. Wunner, T. J. Wiktor, A. D. Lopes, M.
antigenic or host range characteristics. Lafon, C. L. Smith, and H. Koprowski. 1983. Characterization
of an antigenic determinant of the glycoprotein that correlates
ACKNOWLEDGMENTS with pathogenicity of rabies virus. Proc. Natl. Acad. Sci. USA
80:70-74.
We thank Tom Kost for helpful comments on DNA sequencing. 20. Dorner, A. J., J. P. Stoye, and J. M. Coffin. 1985. Molecular
We thank Kaye Smith and Jim Waddle for technical assistance and basis of host range variation in avian retroviruses. J. Virol.
Cindy Rollins for typing the manuscript. We also thank Mike Gill for 53:32-39.
providing virus and cell culture. 21. Green, M. R., and R. G. Roeder. 1980. Transcripts of the
adeno-associated virus genome: mapping of the major RNAs. J.
LITERATURE CITED Virol. 36:79-92.
1. Anton, I. A., and D. P. Lane. 1986. Non-structural protein 1 of 22. Hanahan, D. 1983. Studies on transformation of Escherichia coli
parvoviruses: homology to purine nucleotide using proteins and with plasmids. J. Mol. Biol. 166:557-580.
early proteins of papovaviruses. Nucleic Acids Res. 14:7813. 23. Johnson, B. 1984. Parvovirus proteins, p. 259-295. In K. I.
2. Arnentrout, R., R. Bates, K. Berns, B. Carter, M. Chow, D. Berns (ed.), The parvoviruses. Plenum Publishing Corp., New
Dressler, K. Fife, W. Hauswirth, G. Hayward, G. Lavelle, S. York.
Rhode, S. Strauss, P. Tattersall, and D. Ward. 1978. A standard 24. Jongeneel, C. V., R. Sahli, G. K. McMaster, and B. Hirt. 1986.
nomenclature for restriction endonuclease fragments, p. A precise map of splice junctions in the mRNAs of minute virus
523-526. In D. Ward and P. Tattersall (ed.), Replication of of mice, an autonomous parvovirus. J. Virol. 59:564-573.
parvoviruses. Cold Spring Harbor Laboratory, Cold Spring 25. Labieniec-Pintel, L., and D. Pintel. 1986. The minute virus of
Harbor, N.Y. mice P39 transcription unit can encode both capsid proteins. J.
3. Astell, C. R., E. M. Gardiner, and P. Tattersall. 1986. DNA Virol. 57:1163-1167.
sequence of the lymphotropic variant of minute virus of mice, 26. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular
MVM(i), and comparison with the DNA sequence of the fibro- cloning: a laboratory manual. Cold Spring Harbor Laboratory,
tropic prototype strain. J. Virol. 57:656-669. Cold Spring Harbor, N.Y.
4. Astell, C. R., M. Thomson, M. Merchlinsky, and D. C. Ward. 27. Maxam, A. M., and W. Gilbert. 1980. Sequencing end-labeled
1983. The complete DNA sequence of minute virus of mice, an DNA with base-specific chemical cleavages. Methods Enzymol.
autonomous parvovirus. Nucleic Acids Res. 11:999-1018. 65:499-560.
5. Bensimhon, M., J. Gabarro-Arpa, and L. Reiss. 1983. Physical 28. McMaster, G. K., J.-D. Tratschin, and G. Siegl. 1981. Compar-
characteristics in eucaryotic promoters. Nucleic Acids Res. ison of canine parvovirus with mink enteritis virus by restriction
11:4521-4540. site mapping. J. Virol. 38:368-371.
6. Biggin, M. D., T. J. Gibson, and G. F. Hong. 1983. Buffer 29. Melton, D. A., P. A. Krieg, M. R. Rebagliati, T. Maniatis, K.
gradient gels and 35S label as an aid to rapid DNA sequence Zinn, and M. R. Green. 1984. Efficient in vitro synthesis of
determination. Proc. Natl. Acad. Sci. USA 80:3963-3965. biologically active RNA and RNA hybridization probes from
7. Boros, I., G. Posfai, and P. Venetianer. 1984. High-copy-number plasmids containing a bacteriophage SP6 promoter. Nucleic
derivatives of the plasmid cloning vector pBR322. Gene Acids Res. 12:7035-7056.
30:257-260. 30. Messing, J. 1983. New M13 vectors for cloning. Methods
8. Both, G. W., C. H. Shi, and E. D. Kilbourne. 1983. Hemagglu- Enzymol. 101:20-78.
tinin of swine influenza virus: a single amino acid change 31. Mitra, S., C. E. Snyder, R. C. Bates, and P. T. Banerjee. 1982.
pleiotropically affects viral antigenicity and replication. Proc. Comparative physiochemical and biological properties of two
Natl. Acad. Sci. USA 80:6996-7000. strains of Kilham rat virus, a non-defective parvovirus. J. Gen.
9. Brown, A. L., E. V. Davis, and W. H. Beckenhauer. 1968. Virol. 1:43-54.
Studies of an established canine kidney cell line. Cornell Vet. 32. Molitor, T. W., H. S. Joo, and M. S. Collett. 1985. KBSH
58:593-613. parvovirus: comparison with porcine parvovirus. J. Virol. 5:
10. Carlson, J., K. Rushlow, I. Maxwell, F. Maxwell, S. Winston, 257-263.
276 REED ET AL. J. VIROL.

33. Morgan, W. R., and D. C. Ward. 1986. Three splicing patterns nucleotide sequence of H-1 and mapping of its genes by hybrid-
are used to excise the small intron common to all minute virus arrested translation. J. Virol. 45:173-184.
of mice RNAs. J. Virol. 60:1170-1174. 43. Ron, D., P. Tattersall, and J. Tal. 1984. Formation of a host
34. Parrish, C. R., and L. E. Carmichael. 1986. Characterization range mutant of the lymphotropic strain of minute virus of mice
and recombination mapping of an antigenic and host range during persistent infection in mouse L cells. J. Virol. 52:63-
mutation of canine parvovirus. Virology 148:121-132. 69.
35. Parrish, C. R., P. H. O'Connell, J. F. Evermann, and L. E. 44. Sahli, R., G. K. McMaster, and B. Hirt. 1985. DNA sequence
Carmichael. 1985. Natural variation of canine parvovirus. Sci- comparison between two tissue-specific variants of the autono-
ence 230:1046-1048. mous parvovirus, minute virus of mice. Nucleic Acids Res. 13:
36. Pintel, D., D. Dadachanji, C. R. Astell, and D. C. Ward. 1983. 3617-3633.
The genome of minute virus of mice, an autonomous parvo- 45. Sanger, F., S. Nicklen, and A. R. Caulson. 1977. DNA sequenc-
virus, encodes two overlapping transcription units. Nucleic ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci.
Acids Res. 11:1019-1038. USA 74:5463-5467.
37. Pintel, D., M. J. Merchlinsky, and D. C. Ward. 1984. Expression 46. Shade, R. O., M. C. Blundell, S. F. Cotmore, P. Tattersall, and
of minute virus of mice structural proteins in murine cell lines C. R. Astell. 1986. Nucleotide sequence and genome organiza-
transformed by bovine papillomavirus-minute virus of mice tion of human parvovirus B19 isolated from the serum of a child
plasmid chimera. J. Virol. 52:320-327. during aplastic crisis. J. Virol. 58:921-936.
38. Rhode, S. L., III. 1978. Defective interfering particles of parvo- 47. Siegl, G. 1984. Canine parvovirus: origin and significance of a
virus H-1. J. Virol. 27:347-356. new pathogen, p. 259-295. In The parvoviruses. Plenum Pub-
39. Rhode, S. L., III. 1978. Replication process of the parvovirus, lishing Corp., New York.
H-1. X. Isolation of a mutant defective in replicative-form DNA 48. Tattersall, P., and A. J. Shatkin. 1977. Sequence homology
replication. J. Virol. 25:215-223. between the structural polypeptides of minute virus of mice. J.
40. Rhode, S. L., III. 1985. trans-Activation of parvovirus P38 Mol. Biol. 111:375-394.
promoter by the 76K noncapsid protein. J. Virol. 55:886-889. 49. Tratschin, J. D., G. K. McMaster, G. Kronauer, and G. Siegi.
41. Rhode, S. L., III. 1985. Nucleotide sequence of the coat protein 1982. Canine parvovirus: relationship to wild type and vaccine
gene of canine parvovirus. J. Virol. 54:630-633. strains of feline panleukopenia virus and mink enteritis virus. J.
42. Rhode, S. L., III, and P. R. Paradiso. 1983. Parvovirus genome: Gen. Virol. 61:33-41.