Download Genomic organization of infectious salmon anaemia virus

Journal
of General Virology (2002), 83, 421–428. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
Genomic organization of infectious salmon anaemia virus
Sharon C. Clouthier,1 Trent Rector,2 Nathan E. C. Brown2 and Eric D. Anderson2
1
2
Microtek International, Ltd., Saanichton, British Columbia, Canada V8M 1Z8
Department of Biochemistry, Microbiology and Molecular Biology, University of Maine, Orono, Maine 04469, USA
The RNA genome segment order, nucleotide sequence and the putative encoded proteins were
determined for infectious salmon anaemia virus (ISAV). Eight segments of genomic viral RNA
between 1n0 and 2n4 kb in length were identified. RNA segments 1–6 each had a predicted single
open reading frame encoding the P1, PB1, NP, P2, P3 and HA proteins, respectively. Segment 7
encoded the P4/P5 proteins and segment 8 encoded the P6/P7 proteins. Seven virion proteins
with molecular masses between 25 and 72 kDa were found by SDS–PAGE analysis. The 72 and
42 kDa proteins were immunoreactive with ISAV antiserum from Atlantic salmon. The molecular
mass of the 72 kDa virion protein suggested that it was the NP protein encoded by segment 3. The
amino acid sequence was conserved, sharing 96n6 % identity with the NP protein of a Scottish ISAV
isolate. Comparison of the amino acid sequences obtained by N-terminal analyses and cDNA
nucleotide translation revealed that the 42 and 47 kDa proteins were the HA and P3 proteins
encoded by segments 6 and 5, respectively. In addition, analysis provided evidence for their
protein synthesis initiation sites. Like the HA protein, the signal sequence and potential
glycosylation sites of P3 suggested that it was a surface glycoprotein. The predicted amino acid
sequence shared 83n1, 84n0 and 99n6 % identity to the predicted P3 protein sequences for ISAV
isolates from Norway, Scotland and Maine, respectively. These results establish the specificity,
migration, number and nucleotide sequence of the eight RNA segments of the ISAV genome.
Introduction
Infectious salmon anaemia (ISA) is caused by an unclassified
virus that is related most closely to members of the family
Orthomyxoviridae (Krossøy et al., 1999). The primary tissue
tropism is endothelial cells of the vascular system of some
salmonid fish species (Hovland et al., 1994). Atlantic salmon
(Salmo salar) are particularly susceptible to the virus, with
infection resulting in overt signs of disease and mortalities
within farm populations ranging from 0 to 100 % after several
months. ISA virus (ISAV) particles are pleomorphic, enveloped
spheres with a diameter of 90–130 nm (Dannevig et al., 1995 ;
Nylund et al., 1995). They are covered evenly with surface
Author for correspondence : Eric Anderson.
Fax j1 207 581 2801. e-mail eandersn!maine.maine.edu
The GenBank accession numbers of the sequences reported in this
paper are : ISAV isolate CCBB – segments 1, AF404347 ; 2,
AF404346 ; 3, AF404345 ; 4, AF404344 ; 5, AF404343 ; 6,
AF404342 ; 7, AF404341 ; and 8, AF404340. ISAV isolate Norway –
segments 5, AF429987 ; and 7, AF429990. ISAV isolate Scotland –
segment 5, AF429988. ISAV isolate ME/01 – segments 5,
AF429986 ; 6, AY059402 ; and 7, AF429989.
0001-8175 # 2002 SGM
spikes of 13–15 nm in length and contain filamentous
nucleocapsid-structures that are released upon partial virion
disruption (Sommer & Mennen, 1996).
The virus haemagglutinates a variety of fish cells but not
erythrocytes from mammals and birds (Falk et al., 1997). The
virion contains an acetylesterase receptor-destroying activity
that does not affect influenza A or C virus haemagglutination,
suggesting that the receptors are different for the viruses.
Recent evidence shows that ISAV is similar to orthomyxoviruses in that it binds to sialic acid residues on host
cell surfaces and undergoes fusion with the cell in acidic
endosomes (Eliassen et al., 2000). The haemagglutinating and
acetylesterase activities seem to be carried out by two different
proteins in ISAV (Rimstad et al., 2001).
The ISAV genome is composed of eight segments of
single-stranded, negative-polarity RNA. The genes encoding
the putative PB1, NP, PA, HA and NS proteins have been
described. However, definitive correlation of each gene to their
corresponding genomic segment and to their encoded protein
product has not been made.
Analysis of the nucleotide sequences encoding the putative
NS and PB1 proteins shows that a minimum of two distinct
genomic strains of ISAV exist : the North American strain and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 06 May 2017 13:32:35
ECB
S. C. Clouthier and others
Virus and RNA purification. Virus was prepared from ISAVinfected CHSE-214 cell monolayers. Following complete cell lysis, the
cell culture supernatant was filtered, dialysed against polyethylene glycol
(PEG 8000 ; Sigma) and centrifuged for 2 h at 104 000 g using an SW28
rotor and a Beckman L8–70M ultracentrifuge. The virus pellet was
resuspended in TNE (10 mM Tris, 100 mM NaCl and 1 mM EDTA,
pH 7n5), layered onto a 25, 35 and 45 % discontinuous sucrose gradient
and centrifuged for 3 h at 132 000 g. Virus was collected from the
interface of the 35 and 45 % sucrose layers and centrifuged for 2 h at
104 000 g. Viral RNA (vRNA) was then isolated from the pelleted virus
using TRIzol reagent (Gibco BRL), as described by the manufacturer, and
used for the construction of cDNA libraries.
GCAAAGA. RNA (100 ng) isolated from purified ISAV or CHSE-214
cells (control) was mixed with the ISAV primer (20 pmol\µl), incubated
at 80 mC for 5 min and then combined with the following in a total of
20 µl : 4 µl 5i first-strand buffer (Gibco BRL), 2 µl 10 mM dNTP
mixture (Boehringer Mannheim), 1 µl 0n1 M DTT (Gibco BRL) and 1 µl
SuperScript II reverse transcriptase (15 U\µl Gibco BRL). The mixture
was incubated at 25 mC for 10 min and then at 42 mC for 1 h. The firststrand ISAV cDNA products synthesized by reverse transcription were
amplified by PCR using the ISAV primer and random hexamers. To the
first-strand reaction, the following components were added in a total of
100 µl : 1n5 µl 10 mM dNTP mixture (Boehringer Mannheim), 1n25 µl
ISAV primer (20 pmol\µl), 1 µl random hexamers (25 pmol\µl ; Gibco
BRL), 10 µl 10i PCR buffer with Mg#+ (Boehringer Mannheim) and 1 µl
Taq polymerase (5 U\µl ; Boehringer Mannheim). After 35 cycles of
94 mC for 30 s, 59 mC for 45 s and 72 mC for 1 min, the PCR products
were extended for 10 min at 72 mC. The amplified cDNA products were
then separated by agarose gel electrophoresis, purified from the gel and
cloned into the pGEM-T vector (Promega), according to the manufacturer’s instructions.
For the second approach, first-strand cDNA was synthesized from
ISAV vRNA by reverse transcription with random hexamer primers.
RNA (100 ng) isolated from purified ISAV or CHSE-214 cells (control)
was mixed with random hexamers (50 ng\µl ; Gibco BRL), incubated at
65 mC for 5 min, placed on ice for 2 min and then combined with the
following in a total of 20 µl : 4 µl 5i first-strand buffer (Gibco BRL), 2 µl
10 mM dNTP mixture (Boehringer Mannheim), 1 µl 0n1 M DTT (Gibco
BRL) and 1 µl SuperScript II reverse transcriptase (15 U\µl ; Gibco BRL).
The mixture was incubated at 25 mC for 10 min and then at 50 mC for
50 min. The TimeSaver cDNA Synthesis kit (Pharmacia) was then used
for second-strand cDNA synthesis. The first-strand reaction was added to
the second-strand reaction mixture, incubated at 12 mC for 30 min and
then at 22 mC for 1 h. After spin column purification, the bluntended, double-stranded cDNAs were cloned into dephosphorylated,
SmaI-digested pUC18 (Pharmacia), according to the manufacturer’s instructions.
For both cDNA libraries, Escherichia coli DH5α cells (Gibco BRL) were
transformed with the ligation reactions and the ampicillin-resistant
colonies containing either pGEM-T or pUC18 with cloned ISAV cDNA
were selected by blue\white screening. White colonies were transferred
onto 96-well plates containing 200 µl LB, 250 µg\ml ampicillin and 15 %
glycerol per well, grown overnight at 37 mC and stored at k20 mC.
cDNA library construction. Two different strategies were employed to clone the ISAV genome. For the first approach, first-strand
cDNA was synthesized from ISAV vRNA by reverse transcription with
an ISAV-specific primer, 5h AAGCAGTGGTAACAACGCAGAGTA-
RT–PCR amplification of segments 2, 6 and 8 from ISAV
isolate CCBB. First-strand cDNAs for segments 2, 6 and 8 were
synthesized from ISAV vRNA by reverse transcription using the primers
outlined in Table 1 and under the conditions described above. PCR
the European strain (Inglis et al., 2000). However, these gene
sequences cannot be used to differentiate between the various
European strains. Instead, the highly polymorphic region in the
putative HA protein is used to identify and separate closely
related species (Krossøy et al., 2001). Further resolution of the
ISAV genome and its organization is an important step toward
understanding the relationship between the individual virus
isolates. Towards this goal, we describe here the genome
structure of ISAV isolate CCBB. The antigenic variation of
ISAV has not been defined clearly and it is not known whether
a cross-neutralizing fish immune response can be elicited. To
address this question, we have identified ISAV proteins that
are immunoreactive in Atlantic salmon. Together, these results
are discussed with respect to their impact on rational vaccine
design.
Methods
Viruses and cell lines. ISAV strain CCBB was recovered by Micro
Technologies from infected Atlantic salmon in Back Bay, New Brunswick,
Canada. ISAV isolates from Norway, Scotland (390\98) and Maine
(ME\01) and a second Canadian isolate (NB-99) were also obtained from
Micro Technologies. ISAV was propagated in a Chinook salmon embryo
cell line, CHSE-214 (Lannan et al., 1984). Cell monolayers were infected
with ISAV and incubated at 15 mC in minimum essential medium
containing Hanks’ salts and supplemented with -glutamine (10 mM) and
5 % foetal bovine serum (MEM-H5 ; Gibco BRL).
Table 1. RT–PCR oligonucleotide DNA primers for RNA segments 2, 6 and 8 of ISAV
isolate CCBB
Segment
2
6
8
ECC
Primer name
Seg 2-5hF-mRNA
Seg 2-3hR-mRNA
HA forward
HA reverse
Class II
Seg 8-3hR-mRNA
Primer sequence (5h
3h)
GAACGCTCTTTAATAACCATG
TCAAACATGCTTTTTCTTC
AGCAAAGATGGCACGATTC
TGCACTTTTCTGTAAACGTACAAC
AAGCAGTGGTAACAACGCAGAGTCTATCTACCATG
TTATTGTACAGAGTCTTCC
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 06 May 2017 13:32:35
ISAV genomic organization
Table 2. Summary of groups formed from screening ISAV
cDNA libraries
The cDNA library from approach one had a total of 1364 clones and
that from approach two had a total of 768 clones.
Number of positive clones in
cDNA library
Probe
5 : E-6
Segment 2
1-1F2, 5-1F1
2 : C-5, 4 :D-8
5 : E-7
Segment 6
2 : B-10
Segment 8
Origin of
probe
Approach 1
Approach 2
Approach 2
RT–PCR
Approach 1
Approach 2
Approach 2
RT–PCR
Approach 2
RT–PCR
0
0
212, 1144
0
0
0
0
6
33
41
6, 43
5, 38
14
0
50
10
amplification was used for second-strand cDNA synthesis ; after 30 cycles
of 95 mC for 1 min, 50 mC for 1 min and 72 mC for 2 min, the PCR
products were extended for 10 min at 72 mC (see Table 1 for primers).
RT–PCR products were gel-purified according to the instructions of the
manufacturer (Qiagen).
Selection and identification of ISAV clones from the cDNA
libraries. The contents of one 96-well plate were transferred onto one
Hybond-N+ membrane (Amersham) and then placed on top of an LB agar
plate containing 250 µg\ml ampicillin. Clones were grown on the filters
at 37 mC overnight and the filters were processed for the times indicated
on soaking pads saturated with the following solutions : 0n5 N NaOH for
7 min ; 1 M Tris–HCl, pH 7n4, for 2 min ; 1 M Tris–HCl, pH 7n4, for
2 min ; 0n5 M Tris–HCl, pH 7n4, and 1n5 M NaCl for 4 min. The filters
were transferred to a bath of 2i SSC (1i SSC contains 0n15 M
NaCl and 0n015 M Na citrate) and 1 % SDS and then soaked in
$
2i SSC. After a brief wash in chloroform, the filters were air-dried and
then baked at 80 mC for 2 h. Prehybridization of the filters for 2 h in 6i
SSC, 0n5 % SDS, 5i Denhardt’s and 0n1 mg\ml E. coli tRNA (Sigma) was
followed by hybridization with a probe labelled with [α-$#P]dCTP (NEN)
by nick translation (Amersham). Nick translation was carried out as
outlined by the manufacturer. The cDNA libraries were screened initially
using cDNA (amplified by RT–PCR and gel purified) for segments 2, 6 or
8 of ISAV isolate CCBB. The remaining segments were identified using
probes consisting of gel-purified, restriction enzyme fragments digested
from the plasmids of library clones selected randomly. Library clones
were grouped based on the probe to which they hybridized (Table 2).
Plasmid DNA isolated from representative clones of each group using
Qiaprep columns (Qiagen) was sequenced at the University of Maine
Core Sequencing Facility, Maine, USA. Only those sequences that
matched other orthomyxovirus sequences or that did not match non-viral
sequences were analysed further.
Northern blot hybridization. Northern blot analysis was used to
correlate each representative sequence with a specific ISAV genomic
segment. Total RNA was isolated from CHSE-214 cell monolayers or
ISAV-infected CHSE-214 cell monolayers using TRIzol reagent. The
RNA was separated on a 2 % agarose gel containing formaldehyde and
transferred onto a Hybond-N+ membrane in 10i SSC by capillary
action, as described by Fourney et al. (1992). The probes used for
Northern blot analysis were gel-purified, restriction enzyme-digested
fragments from the plasmids of appropriate cDNA library clones. The
probes were labelled with [α-$#P]dCTP by nick translation and hybridized
to the blots at 42 mC for 18 h in ULTRAhyb (Ambion). The membranes
were washed twice for 5 min in 2i SSC and 0n1 % SDS at 42 mC and then
twice for 15 min in 0n1i SSC and 0n1 % SDS at 42 mC. The results were
recorded on Kodak X-OMAT AR film.
The probes used in the Northern blot hybridization experiments were
derived from four clones constructed using the second approach, one
clone constructed using the first approach and RT–PCR products of
the three known segments (Table 2). A single RNA blot was probed
consecutively with each of the eight individual probes. One probe was
hybridized to the Northern blot and the results were visualized by
autoradiography. The next probe was hybridized to the same Northern
blot, the results were visualized and compared with the results from the
previous hybridization. By repeating this process with each of the eight
probes, we were able to correlate each individual probe and its
corresponding nucleotide sequence with a specific RNA segment.
Construction of full-length clones of each ISAV genome
segment. The full-length cDNA sequences for each of the ISAV RNA
segments, with the exception of segment 1, was generated by rapid
amplification of cDNA ends (RACE) PCR using the RLM-RACE kit
(Ambion). The PCR products were cloned into either pCR2.1-TOPO
(Invitrogen) or pGEM-T, as directed by the manufacturers ’ instructions,
and then sequenced. AssemblyLIGN, version 1.0.9b (Oxford Molecular
Group), was used to order the overlapping sequenced DNA fragments
for construction of the full-length sequence.
PCR primers were designed from the consensus sequence obtained
for each ISAV RNA segment and used to amplify the full-length cDNA
sequence for each segment, with the exception of segment 1. The PCR
product for each segment was cloned into pGEM-T and DNA from three
representative clones was sequenced. The programs contained in
MacVector, version 6.5.3 (Oxford Molecular Group), were used to
identify open reading frames (ORFs) and regions of local similarity. The
nucleotide and predicted amino acid sequence for each ORF were
analysed by BLAST searches through the National Center for Biotechnology Information server (Altschul et al., 1990 ; Pearson & Lipman,
1988) or the Influenza database (Los Alamos National Laboratory). The
most likely cleavage sites for signal peptidase in HA and 5 : E-7 were
determined using SignalP, version 1.1 (Nielsen et al., 1997).
Generation of anti-ISAV immune sera. Anti-ISAV antibodies
were generated in Atlantic salmon injected with tissue culture supernatant
from ISAV-infected CHSE-214 cell monolayers (Opitz et al., 2000).
Mouse polyclonal and monoclonal antibodies (pAbs and mAbs, respectively) to ISAV were generated by Rob Beecroft (Immuno-Precise
Antibodies).
SDS–PAGE and Western blot analysis. Whole cell lysates of
naive and ISAV-infected CHSE-214 cells as well as purified ISAV were
screened for the presence of immunoreactive antigens with sera from
vaccinated and challenged Atlantic salmon. SDS–PAGE was carried out
as described by Laemmli (1970). Immunoreactive protein bands were
visualized by Western blot analysis using sera from ISAV-injected
Atlantic salmon, followed by incubation with mouse anti-salmonid
immunoglobulin 5F12 mAb (Immuno-Precise Antibodies). Alternatively,
the proteins on the membranes were screened with anti-ISAV mouse
pAbs and mAbs. Immunoreactive proteins were detected with goat antimouse immunoglobulin G–alkaline phosphatase conjugates (Southern
Biotechnology) and visualized with BCIP\NBT (Sigma).
N-terminal amino acid sequence analysis. The proteins of
purified ISAV were separated by SDS–PAGE, blotted onto PVDF
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 06 May 2017 13:32:35
ECD
S. C. Clouthier and others
membranes (Bio-Rad) and stained with 0n1 % Coomassie blue R-250 in
40 % methanol and 1 % acetic acid. The stained protein bands were
removed from the membrane and subjected to N-terminal amino acid
sequence analysis using an Applied Biosystems gas-phase sequencer,
model 470A, or a liquid-phase sequencer, model 473, with online
phenylthiohydantoin analysis.
Results
Genetic analysis
The RNA genome segment order, nucleotide sequence and
putative encoded proteins were determined for ISAV. To
accomplish this, two cDNA libraries were made using genomic
RNA from purified virus and screened using known sequences
from segments 2, 6 and 8 or randomly chosen cDNA sequences
from each library. Eight distinct cDNA hybridization groups
were identified. Of these, two groups were found in cDNA
library 1 and all but one segment of the ISAV genome in
cDNA library 2 (Table 2). Northern blot analysis of RNA from
naive or ISAV-infected cells was done using a probe made
from each cDNA hybridization group (Fig. 1). The RNA
segments were correlated to the ISAV cDNA by consecutively
hybridizing a Northern blot of mRNA from ISAV-infected
cells or purified virus with cDNA from each hybridization
group (Fig. 1). Eight RNA segments were identified ; segments
1 and 2 were both approximately 2400 nucleotides in length.
The ISAV RNA segment corresponding to each cDNA clone
is summarized in Table 3. The genome segments are numbered
with respect to their mobility in agarose gels, from the slowest
to the fastest, and comprise a genome of 14 500 nucleotides.
The length of each gene, the corresponding encoded
polypeptide(s) and the predicted molecular masses of the
translated proteins are summarized in Table 3. Only a partial
sequence from segment 1 was obtained. The cDNA sequences
of segments 1–6 were predicted to encode one ORF. Segments
7 and 8 were each predicted to encode two proteins.
Comparison of the cDNA nucleotide and predicted amino
acid sequences for the ISAV genome to those listed in the
Influenza database and in GenBank showed that RNA
segments 1 and 5 of ISAV isolate CCBB were unique. RNA
segments 2, 3, 4 and 6 were found to encode the putative
proteins PB1, NP, PA and HA, respectively. The predicted
sequences of the P6 and P7 proteins encoded on RNA segment
8 were similar to the sequences of the two ORFs on segment
8 from other ISAV isolates.
The protein sequence of the partial ORF encoded on
segment 1 was unique. The predicted amino acid sequence of
the PB1 protein, encoded by RNA segment 2, was 82n2–84n5 %
similar to the amino acid sequences of the PB1 proteins from
Norwegian (AJ002475) and Scottish (AF262392) ISAV isolates. The assignment of the NP protein to the ORF encoded
on RNA segment 3 was based on nucleotide sequence
similarity to the influenza A virus NP protein RNA-binding
region (Fig. 2) and to the putative NP protein sequence
described by Snow & Cunningham (2001). The sequence for
ECE
Fig. 1. Northern blot analysis of cellular RNA with eight ISAV-specific
probes. Purified cellular RNA was separated on a 2 % agarose gel and
transferred to a Hybond-N+ membrane. Lanes 1–7 and 10, cellular RNA
from CHSE-214 cells infected with ISAV isolate CCBB ; 8, cellular RNA
from ISAV isolate ME-01 ; 9, cellular RNA from CHSE-214 cells infected
with ISAV isolate NB-99 ; 11, cellular RNA from naive CHSE-214 cells. The
RNA blot was consecutively hybridized with radioactively labelled DNA
probes specific for one of the ISAV RNA segments. The results recorded
by autoradiography after the addition of each single probe to the same
RNA blot are shown in lanes 1–11. The probes are identified by segment :
lanes 1, segment 3 ; 2, segment 4 ; 3, segment 6 ; 4, segment 1 ; 5,
segment 5 ; 6, segment 7 ; 7, segment 8 ; 8–11, segment 2. Molecular
mass standards (kb) are indicate on the left. The RNA segments are
labelled on the right.
the NP protein of ISAV isolate CCBB was highly conserved,
sharing 96n6 % identity to that reported for the NP protein of
the Scottish ISAV isolate (AJ276858). The predicted protein
sequence of the P2 protein from RNA segment 4 shared 99 %
identity to the putative PA protein sequence (AF306548)
described by Ritchie et al. (2001). The nucleotide sequences for
segment 5 of the Scottish (AF429988), Norwegian (AF429987)
and Maine (AF429986) isolates of ISAV were 76n4, 76n0 and
99n7 % similar to the corresponding sequence of ISAV isolate
CCBB. The predicted translation of the ORF encoded by RNA
segment 6 shared 84n8–84n3 % similarity to the predicted HA
protein sequences for ISAV isolates from Norway (AF302799)
and Scotland (AJ276859) and 99n2 % similarity to the Maine
ISAV isolate (AY059402). The nucleotide sequence for ISAV
CCBB segment 7 shared 99n6 % identity with a reported ISAV
sequence (AX083264). The P4 and P5 proteins encoded on
segment 7 shared 99n2–99n3 % similarity to the translations
predicted for ORFs 1 and 2 from the reported sequence
(AX083264). The nucleotide sequence for segment 8 of the
Norwegian (AF429990) and ME\01 (AF429989) isolates of
ISAV shared 88n7–99n9 % identity to the corresponding
sequence from ISAV isolate CCBB. Our results confirmed that
segment 8 encoded two proteins, as reported previously by
Mjaaland et al. (1997). The amino acid sequence translated
from the largest ORF shared 75n6–97n9 % similarity to the
sequence reported previously for the Norwegian (AF262382),
Scottish (AJ242016) and Canadian (AJ242016) isolates of
ISAV.
ISAV proteins
N-terminal amino acid sequence analysis was used to
correlate viral proteins to the predicted translation of ORFs
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 06 May 2017 13:32:35
ISAV genomic organization
Table 3. RNA segments of ISAV isolate CCBB, their genes and encoded proteins
GenBank accession numbers for the segments are as follows : segments 1, AF404347 ; 2, AF404346 ; 3,
AF404345 ; 4, AF404344 ; 5, AF404343 ; 6, AF404342 ; 7, AF404341 ; and 8, AF404340.
Segment
Clone
Segment
length (kb)*
ORF length
(bp)
Encoded
protein
Polypeptide
length (aa)
Predicted
molecular
mass (kDa)
1749
2127
1851
1737
1335
1185
771
441
705
552
P1
PB1
NP
P2
P3
HA
P4
P5
P6
P7
–
709
617
579
445
395
257
147
235
184
–
80n5
68n0
65n3
48n8
43n1
28n6
16n3
26n5
20n3
1
2
3
4
5
6
7
5 :E-6
Segment 2
1-1F2\5-5F1
2 :C-5\4 :D-8
5 :E-7
HA
2 :B-10
2n4
2n4
2n2
1n9
1n6
1n5
1n3
8
Segment 8
1n0
* Based on the average length determined from Northern blot hybridization analyses with two to five
replicates per probe.
Fig. 2. Amino acid sequence alignment of the RNA-binding domain of the NP protein from influenza A and B viruses with the
putative NP protein RNA-binding domain from ISAV, as predicted using the CLUSTAL W system. ISAV NP, amino acids 189–307
(AF404345) ; influenza A virus (InfA) NP, amino acids 90–188 (P15675) ; InfB NP, amino acids 149–249 (P04666).
Identical amino acids and amino acid residues with similarity in physical and chemical properties are indicated as (*) and (,),
respectively. The NP protein RNA-binding domain from influenza A and B viruses was taken from Kobayashi et al. (1994).
encoded by RNA segments of ISAV isolate CCBB. Purified
ISAV proteins were separated by SDS–PAGE and visualized
by Coomassie blue staining (Fig. 3). Seven putative ISAV
proteins with estimated molecular masses of between 25 and
72 kDa were evident (Fig. 3). The four prominent proteins had
molecular masses of 72, 47, 42 and 25 kDa. N-terminal amino
acid sequence analysis was successful with three of the seven
proteins (Table 4). The amino acid sequence at the N terminus
of both the 40 and 42 kDa proteins was similar to amino
acids 17–41 in the predicted translation of the Norwegian
(AF302799), Scottish (AJ276859) and North American ISAV
HA genes. The region of overlap did not include the first 16
amino acids of the translation, indicating that these amino acids
were not present in the virion HA protein. The amino acid
sequence of the HA protein from ISAV isolate CCBB had a
signal peptidase cleavage site predicted to be between Ser"'
and Arg"(. Thus, after cleavage of the signal sequence, the
mature HA protein began at Arg"(. Analysis also showed that
the 40 kDa protein was a C-terminally truncated form of the
42 kDa HA protein. The 13 N-terminal amino acids from the
47 kDa protein were unique and identical to amino acids in the
predicted translation of ISAV RNA segment 5 (Table 4). The
translated sequence of the 47 kDa protein had a signal sequence
predicted to be 17 amino acids and the N-terminal amino acid
sequence results confirmed that the cleavage site was located
between Cys"( and Glu").
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 06 May 2017 13:32:35
ECF
S. C. Clouthier and others
and 42 kDa proteins of ISAV (Fig. 4A). Similar analyses were
performed with ISAV-specific mouse pAbs and mAbs. Of the
six immunoreactive proteins present in the cellular preparation
of ISAV and recognized by the mouse polyclonal sera, three
were present in the purified ISAV sample (42, 25 and
15 kDa ; Fig. 4B). Only the 42 kDa protein was recognized by
the mAb (Fig. 4C). For each serum tested, no reaction was
observed with the naive CHSE-214 sample, indicating that the
immunoreactive proteins were derived from ISAV.
Fig. 3. Analyses of proteins from purified ISAV isolate CCBB. Purified ISAV
was resuspended in SDS-sample buffer. The solubilized proteins were
separated by SDS–PAGE on a 5 % stacking gel and a 12 % resolving gel
and visualized by Coomassie blue staining. Lanes 1, 2n5 % ; 2, 5n0 % ; 3,
7n5 % total virus preparation. Molecular mass standards (kDa) are
indicated on the left. Arrows on the right point to four prominent protein
bands and are labelled with the corresponding molecular masses (kDa).
Table 4. N-terminal amino acid sequence analysis of
proteins from ISAV isolate CCBB
Protein
(kDa)
40
42
47
Sequence analysis
Similarity
analysis
RLXLRNHPDTTWIGDSRSDQSRXNQ
RLXLRNHPDTTWIGDSRSDQSRXNQ
EPXIXENPTXLAI
HA
HA
5 :E-7
Immunoreactive polypeptides encoded by the RNA segments were identified by Western blot analysis performed on
naive and ISAV-infected CHSE-214 cells (Fig. 4). Sera collected
from Atlantic salmon injected with ISAV reacted with the 72
Discussion
This paper reports the organization of the ISAV genome
segments as well as their genes and encoded proteins. The
genome size and number of segments correspond to those
reported previously for the European strain of ISAV (Mjaaland
et al., 1997). Analyses show that each of the eight segments of
the ISAV genome encodes one ORF, with the exception of
segments 7 and 8. The gene assignment for ISAV and the
proteins for which they encode are as follows : RNA segment
1 encodes P1, 2 encodes PB1, 3 encodes NP, 4 encodes P2, 5
encodes P3, 6 encodes HA, 7 encodes P4 and P5, and 8 encodes
P6 and P7. This organization differs from that reported for
other members of the Orthomyxoviridae but is most similar to
influenza A and B viruses. Phylogenetic distance calculations
also support a closer relationship of ISAV to influenza viruses
than to Thogoto and Dhori viruses (Krossøy et al., 1999). The
gene order for influenza A and B viruses and ISAV RNA
segment 2 is the same. However, the ISAV gene order does
have distinct differences relative to the following influenza A
and B viruses, RNA segments and the proteins for which they
encode : segments 3 (PA), 4 (HA), 5 (NP) and 6 (NA) (Lamb,
1989). RNA segments 7 and 8 for ISAV as well as influenza A
and B viruses are predicted to encode two proteins each. As
Fig. 4. Analyses of the immunoreactive antigens of ISAV isolate CCBB. Whole cell lysates were separated by SDS–PAGE,
transferred to a nitrocellulose membrane and the immunoreactive antigens were detected using Western blot analysis.
(A) Atlantic salmon anti-ISAV sera ; (B) ISAV pAb ; (C) ISAV mAb. Lanes 1, whole cell lysate of CHSE-214 cells ; 2, whole cell
lysate of CHSE-214 cells infected with ISAV isolate CCBB ; 3 (B and C only), semi-purified ISAV isolate CCBB. Molecular mass
standards (kDa) are indicated on the left.
ECG
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 06 May 2017 13:32:35
ISAV genomic organization
yet, which of these segments of ISAV encodes the NS and M
proteins is unknown.
Transcription of influenza virus RNA is primed by capped
RNA generated by an endonuclease that cleaves cellular RNA
10–15 nucleotides from their 5h ends. The heterogeneous cap
sequence is then followed by 12–16 conserved nucleotides
present in the vRNA (Lamb & Choppin, 1983). The 5h ends of
mRNAs corresponding to ISAV segments 2, 3, 6 and 8 contain
seven to eight conserved nucleotides related closely to the
conserved nucleotides in other members of the Orthomyxoviridae (Krossøy et al., 1999 ; 2001 ; Sandvik et al.,
2000 ; Rimstad et al., 2001 ; Snow & Cunningham, 2001). In this
study, segments 2, 3, 6, 7 and 8 contain at least the ‘ CAAAGA ’
portion of the consensus sequence. The mRNA polyadenylation signal for orthomyxoviruses is a stretch of five to
seven uridine residues, 15–22 nucleotides downstream from
the 5h end of the vRNA (Li & Palese, 1994). ISAV genomic
segments 7 and 8 each have a polyadenylation signal of four to
five uridine residues located 13–14 nucleotides downstream
from the 5h end of the vRNA (Sandvik et al., 2000). In this
study, segments 2, 4, 5, 6 and 7 contain similar signals at the
3h ends of their mRNA sequences, indicating that CCBB ISAV
contains the conserved orthomyxovirus polyadenylation signal.
ISAV segments 3 and 6 encode the NP (68 kDa) and HA
(43 kDa) proteins, respectively. Antibodies in serum from
Atlantic salmon infected with ISAV recognize a 72 and a
42 kDa protein. The assignment of the 72 kDa protein to NP
is based on size. The putative nucleocapsid protein from the
Scottish strain of ISAV shares 96n6 % identity to the NP
protein from the CCBB isolate. The high sequence conservation
suggests that the ISAV NP protein may be a type-specific
antigen, like the NP proteins of influenza viruses (Lamb &
Krug, 1996). A protective humoral immune response to
influenza viruses is made to the surface protein HA. The HAspecific antibodies in fish sera indicate that ISAV HA proteins
may play a similar role in protecting Atlantic salmon against
ISAV. These findings suggest that the ISAV NP and HA
proteins will be important antigens for ISA vaccine design and
typing of ISAV.
There are three predicted N-linked glycosylation sites in
the ISAV HA protein (Kornfeld & Kornfeld, 1985) : one is
unique to the North American isolates of ISAV (positions
155–157) and one is conserved among the European and North
American isolates of ISAV (positions 333–335). Also of interest
is the putative N-glycosylation site in the hypervariable
region of the HA protein from the Norwegian ISAV isolate
Glesvaer\2\90. The carbohydrates on the HA protein may
contribute to differences in immunogenicity and the ability
of various isolates of ISAV to haemagglutinate the red blood
cells of different species of vertebrates.
Segment 5 has a single ORF of 1335 nucleotides encoding
445 amino acids (48n8 kDa). N-terminal amino acid sequence
analysis confirms that the 47 kDa protein is encoded by
segment 5. The function of the protein remains unknown but
P3 is likely to be the 53 kDa protein reported for other ISAV
strains (Falk et al., 1997 ; Kibenge et al., 2000). In this study,
antibodies in serum from Atlantic salmon infected with ISAV
do not recognize the P3 protein. However, the hydrophobic
signal sequence of 17 amino acids and potential glycosylation
sites suggest that it is a surface glycoprotein. If this is the case,
then, like influenza A and B viruses, ISAV has two virion
surface glycoproteins.
RNA segment 7 of ISAV isolate CCBB contains two ORFs.
The largest ORF (j1) extends from the ATG codon at
nucleotides 47–49 to a termination codon at nucleotides
815–817 and encodes a 257 amino acid protein (28n6 kDa).
This segment contains a second ORF (j2) that could code for
147 amino acids (16n3 kDa). The eighth RNA segment of ISAV
isolate CCBB also encodes two ORFs : the first (j1) comprises
703 nucleotides and encodes a 235 amino acid protein
(26n5 kDa) ; the second (j2) is 552 nucleotides and encodes a
184 amino acid protein (20n3 kDa). RNA segments 7 and 8 of
influenza A virus encode two proteins each, the membrane
proteins M1 and M2 and the non-structural proteins NS1 and
NEP, respectively (Lamb, 1989). These proteins from influenza
A, B and C viruses are encoded by spliced transcripts. If ISAV
uses a similar coding strategy, the proteins on segment 7 may
also be made via alternative splicing. This is in contrast to the
ORF (j2) on segment 8, which begins one nucleotide after the
ATG of the second ORF (j1) and makes splicing unlikely.
The ISAV genome comprises eight segments of RNA.
There are ten predicted encoded proteins but more may be
found. The biological functions of the ISAV proteins remain
unknown, with the exception of the HA protein. The NP and
HA proteins are immunoreactive and may prove to be
important vaccine candidates or type-specific antigens. It may
be that the detection of anti-P3 antibodies in the Atlantic
salmon serum were masked by the antibody response to the
HA protein. It will be interesting to determine if antibodies to
the second glycoprotein, P3, can be generated in Atlantic
salmon in the absence of the HA protein. The North American
and European ISAV isolates are genetically different and may
represent distinct strains. Studies to determine if ISAV genetic
reassortment occurs and the consequences with respect to
pathogenicity are required. Reassortant studies with the
various strains of ISAV may also provide insight into the
biochemical structure\function of a given gene and its encoded
protein. These studies should provide the basis for rational
design of vaccines for ISAV.
The authors wish to thank Micro Technologies, Inc. for providing
the isolates of ISAV used in this study. We are very grateful to Sandy
Kielland (Protein Microsequencing Laboratory, University of Victoria)
for N-terminal protein sequence analysis. Financial support for this work
was provided by Microtek International (1998), Ltd. and by grants to
E. D. A. from Northeast Regional Aquaculture Consortium (NRAC) and
Sea Grant.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 06 May 2017 13:32:35
ECH
S. C. Clouthier and others
References
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J.
(1990). Basic local alignment search tool. Journal of Molecular Biology
215, 403–410.
Dannevig, B. H., Falk, K. & Namork, E. (1995). Isolation of the causal
virus of infectious salmon anaemia (ISA) in a long-term cell line from
Atlantic salmon head kidney. Journal of General Virology 76, 1353–1359.
Eliassen, T. M., Froystad, M. K., Dannevig, B. H., Jankowska, M.,
Brech, A., Falk, K., Romoren, K. & Gjoen, T. (2000). Initial events in
infectious salmon anemia virus infection : evidence for the requirement of
a low-pH step. Journal of Virology 74, 218–227.
Falk, K., Namork, E., Rimstad, E., Mjaaland, S. & Dannevig, B. H.
(1997). Characterization of infectious salmon anemia virus, an ortho-
myxo-like virus isolated from Atlantic salmon (Salmo salar L.). Journal of
Virology 71, 9016–9023.
Fourney, R. M., Miyakoshi, J., Kay, R. S., Jr & Paterson, M. C. (1992).
Northern blotting : efficient RNA staining and transfer. Focus 10, 5–7.
Hovland, T., Nylund, A., Watanabe, K. & Endresen, C. (1994).
Observation of infectious salmon anaemia virus in Atlantic salmon, Salmo
salar L. Journal of Fish Diseases 17, 291–296.
Inglis, J. A., Bruce, J. & Cunningham, C. O. (2000). Nucleotide sequence
variation in isolates of infectious salmon anaemia virus (ISAV) from
Atlantic salmon Salmo salar in Scotland and Norway. Diseases of Aquatic
Organisms 43, 71–76.
Kibenge, F. S. B., Lyaku, J. R., Rainnie, D. & Hammell, K. L. (2000).
Growth of infectious salmon anaemia virus in CHSE-214 cells and
evidence for phenotypic differences between virus strains. Journal of
General Virology 81, 143–150.
Kobayashi, M., Toyoda, T., Adyshev, D. M., Azuma, Y. & Ishihama, A.
(1994). Molecular dissection of influenza virus nucleoprotein : deletion
mapping of the RNA binding domain. Journal of Virology 68, 8433–8436.
Kornfeld, R. & Kornfeld, S. (1985). Assembly of asparagine-linked
Lamb, R. A. & Choppin, P. W. (1983). The gene structure and replication
of influenza virus. Annual Review of Biochemistry 52, 467–506.
Lamb, R. A. & Krug, R. M. (1996). Orthomyxoviridae : the viruses and
their replication. In Fields Virology, 3rd edn, pp. 1353–1395. Edited by B.
N. Fields, D. M. Knipe & P. M. Howley. Philadelphia : Lippincott–Raven.
Lannan, C. N., Winton, J. R. & Fryer, J. L. (1984). Fish cell lines :
establishment and characterization of nine cell lines from salmonids. In
Vitro 20, 671–676.
Li, X. & Palese, P. (1994). Characterization of the polyadenylation
signal of influenza virus RNA. Journal of Virology 68, 1245–1249.
Mjaaland, S., Rimstad, E., Falk, K. & Dannevig, B. H. (1997). Genomic
characterization of the virus causing infectious salmon anemia in Atlantic
salmon (Salmo salar L.) : an orthomyxo-like virus in a teleost. Journal of
Virology 71, 7681–7686.
Nielsen, H., Engelbrecht, J., Brunak, S. & von Heijne, G. (1997).
Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Engineering 10, 1–6.
Nylund, A., Hovland, T., Watanabe, K. & Endresen, C. (1995). Presence
of infectious salmon anaemia virus (ISAV) in tissues of Atlantic salmon,
Salmo salar L., collected during three separate outbreaks of the disease.
Journal of Fish Diseases 18, 135–145.
Opitz, H. M., Bouchard, D., Anderson, E., Blake, S., Nicholson, B. &
Keleher, W. (2000). A comparison of methods for the detection of
experimentally induced subclinical infectious salmon anaemia in Atlantic
salmon. Bulletin of the European Association of Fish Pathologists 20, 12–22.
Pearson, W. R. & Lipman, D. J. (1988). Improved tools for biological
sequence comparison. Proceedings of the National Academy of Sciences USA
85, 2444–2448.
Rimstad, E., Mjaaland, S., Snow, M., Mikalsen, A. B. & Cunningham,
C. O. (2001). Characterization of the infectious salmon anemia virus
genomic segment that encodes the putative hemagglutinin. Journal of
Virology 75, 5352–5356.
Ritchie, R. J., Heppell, J., Cook, M. B., Jones, S. & Griffiths, S. G.
(2001). Identification and characterization of segments 3 and 4 of the
oligosaccharides. Annual Review of Biochemistry 54, 631–664.
ISAV genome. Virus Genes 22, 289–297.
Krossøy, B., Hordvik, I., Nilsen, F., Nylund, A. & Endresen, C. (1999).
Sandvik, T., Rimstad, E. & Mjaaland, S. (2000). The viral RNA 3h- and
The putative polymerase sequence of infectious salmon anemia virus
suggests a new genus within the Orthomyxoviridae. Journal of Virology 73,
2136–2142.
Krossøy, B., Devold, M., Sanders, L., Knappskog, P. M., Aspehaug, V.,
Falk, K., Nylund, A., Koumans, S., Endresen, C. & Biering, E. (2001).
Cloning and identification of the infectious salmon anaemia virus
haemagglutinin. Journal of General Virology 82, 1757–1765.
Laemmli, U. K. (1970). Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680–685.
Lamb, R. A. (1989). Genes and proteins of the influenza viruses. In The
Influenza Viruses, pp. 1–87. Edited by R. M. Krug. New York : Plenum.
ECI
5h-end structure and mRNA transcription of infectious salmon anaemia
virus resemble those of influenza viruses. Archives of Virology 145,
1659–1669.
Snow, M. & Cunningham, C. O. (2001). Characterization of the putative
nucleoprotein gene of infectious salmon anaemia virus (ISAV). Virus
Research 74, 111–118.
Sommer, A.-I. & Mennen, S. (1996). Propagation of infectious salmon
anaemia virus in Atlantic salmon, Salmo salar L., head kidney macrophages. Journal of Fish Diseases 19, 179–183.
Received 18 October 2001 ; Accepted 26 October 2001
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 06 May 2017 13:32:35