Báo cáo khoa học: Promoters of type I interferon genes from Atlantic salmon contain two main regulatory regions docx

Promoters of type I interferon genes from Atlantic salmon
contain two main regulatory regions
Veronica Bergan, Silje Steinsvik, Hao Xu, Øyvind Kileng and Børre Robertsen
Department of Marine Biotechnology, Norwegian College of Fishery Science, University of Tromsø, Norway
The type I interferon (IFN) system plays a critical role
in the innate immune defense against viruses in verte-
brates. Virus-infected cells synthesize and secrete type I
interferons (IFN-a ⁄ b), which circulate in the body and
protect other cells from viral infection. The antiviral
action is caused by binding of IFN-a ⁄ b to the type I
IFN receptor resulting in activation of transcription
of several hundred IFN-stimulated genes, some of
which encode proteins that inhibit viral replication.
The antiviral properties of at least three type I IFN-
induced proteins are well established. These comprise
dsRNA-activated protein kinase R (PKR), 2¢,5¢-oligo-
adenylate synthetase and Mx proteins [1].
Although the structures of the IFN-a and IFN-b
promoters from human and mouse have been long
known, the mechanisms involved in viral induction of
type I IFNs have only recently been uncovered [2,3].
Of great importance has been the discovery of IFN
super-producing blood cells called plasmacytoid
dendritic cells (pDCs) and the realization that the
Keywords
Atlantic salmon; interferon promoter;
interferon regulatory factor; nuclear factor
kappa B (NFjB); poly(I:C)
Correspondence
B. Robertsen, Department of Marine
Biotechnology, Norwegian College of

Fishery Science, University of Tromsø,
N-9037 Tromsø, Norway
Fax: +47 776 45110
Tel: +47 776 44487
E-mail:
(Received 24 April 2006, revised 9 June
2006, accepted 15 June 2006)
doi:10.1111/j.1742-4658.2006.05382.x
Recognition of viral nucleic acids by vertebrate host cells results in the syn-
thesis and secretion of type I interferons (IFN-a ⁄ b), which induce an anti-
viral state in neighboring cells. We have cloned the genes and promoters of
two type I IFNs from Atlantic salmon. Both genes have the potential to
encode IFN transcripts with either a short or a long 5¢-untranslated region,
apparently controlled by two distinct promoter regions, PR-I and PR-II,
respectively. PR-I is located within 116 nucleotides upstream of the short
transcript and contains a TATA-box, two interferon regulatory factor
(IRF)-binding motifs, and a putative nuclear factor kappa B (NFjB)-bind-
ing motif. PR-II is located 469–677 nucleotides upstream of the short tran-
script and contains three or four IRF-binding motifs and a putative
ATF-2 ⁄ c-Jun element. Complete and truncated versions of the promoters
were cloned in front of a luciferase reporter gene and analyzed for promo-
ter activity in salmonid cells. Constructs containing PR-I were highly
induced after treatment with the dsRNA poly(I:C), and promoter activity
appeared to be dependent on NF jB. In contrast, constructs containing
exclusively PR-II showed poor poly(I:C)-inducible activity. PR-I is thus the
main control region for IFN-a ⁄ b synthesis in salmon. Two pathogenic
RNA viruses, infectious pancreatic necrosis virus and infectious salmon
anemia virus, were tested for their ability to stimulate the minimal PR-I,
but only the latter was able to induce promoter activity. The established
IFN promoter-luciferase assay will be useful in studies of host–virus inter-

actions in Atlantic salmon, as many viruses are known to encode proteins
that prevent IFN synthesis by inhibition of promoter activation.
Abbreviations
2-AP, 2-aminopurine; EMEM, Eagle’s minimal essential medium; IFN, interferon; IPNV, infectious pancreatic necrosis virus; IRF, interferon
regulatory factor; IRF-E, interferon regulatory factor binding element; ISAV, infectious salmon anemia virus; LPS, lipopolysaccharide; NFjB,
nuclear factor kappa B; pDCs, plasmacytoid dendritic cells; PDTC, pyrrolidine dithiocarbamate; poly(I:C), polyinosinic polycytidylic acid;
PKR, dsRNA-activated protein kinase; PR, promoter region; PRD, positive regulatory domain; TLR, toll-like receptor.
FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS 3893
mechanism of virus-mediated induction of IFNs is dif-
ferent in pDCs and other body cells [4,5]. Most nucle-
ated cells of the body produce IFN-a ⁄ b in response to
recognition of dsRNA intermediates produced during
viral replication. The main sensors of dsRNA are two
intracellular RNA helicases (RIG-I and MDA5) [6–9],
which, on binding of dsRNA, interact with the mitoch-
ondrial protein MAVS (also called IPS-1) [10,11]. This
interaction leads to transcriptional induction of the
IFN-b gene through the co-ordinated activation of the
transcription factors interferon regulatory factor 3
(IRF-3), nuclear factor kappa B (NFjB) and ATF-
2 ⁄ c-Jun heterodimer [2]. Infected cells secrete mainly
IFN-b in the initial phase of infection, but switch to
IFN-a as a result of induction of IRF-7 synthesis dur-
ing the subsequent amplification phase of the IFN
response [12,13]. pDCs are specialized IFN producers
and represent a major source of IFN-a in humans
through activation of IRF-7 [14]. In pDCs, the main
sensors of viral infection are Toll-like receptors (TLRs)
expressed on the surface or in endosomes that recog-
nize viral RNA or DNA. Human pDCs mostly express

TLRs, which recognize ssRNA (TLR7 and TLR8) or
dsCpG-rich DNA (TLR9) [15]. Recognition of viral
nucleic acids by TLRs activates IRF-7, which tran-
scriptionally activates multiple IFN-a genes [16,17]. A
major difference between pDCs and other cell types is
their capacity to constitutively produce relatively high
concentrations of IRF-7 [18].
Virus-induced expression of IFN-a and IFN-b genes
is mediated by regulatory sequences located within
200 bp upstream of the transcription start site of their
promoters [19]. The IFN-b promoter contains four
positive regulatory domains (PRDs), which bind IRFs:
mainly IRF-3 and IRF-7 (PRDI and PRDIII), NF jB
(PRDII) and ATF-2 ⁄ c-Jun (PRDIV) [20]. The promot-
ers of IFN-a genes all contain DNA elements binding
IRF members, notably IRF-3, IRF-5 and IRF-7, but
they do not contain NFjB or ATF-2 ⁄ c-Jun binding
sites [21].
In mammals and birds, IFN-a ⁄ b genes are encoded
by intron-lacking genes whereas IFN-k genes possess a
4-intron ⁄ 5-exon structure [22,23]. Recently, type I IFN
genes of teleost fish were shown to possess a gene
structure similar to IFN-k genes, although their pro-
tein sequences are more similar to IFN-a than IFN-k
[24–28]. At present, little is known about the regula-
tion of fish IFN genes, although the promoter of the
zebrafish type I IFN gene was recently reported to
contain one IRF-binding site and one NFjB-binding
site [28]. The present work shows that type I IFN
genes of Atlantic salmon show a rather unique organ-

ization of the promoter in comparison with mammals,
birds and zebrafish. Atlantic salmon stimulated with
the dsRNA polyinosinic polycytidylic acid [poly(I:C)]
produces an IFN transcript with a short 5¢-UTR called
SasaIFN-a1, and another IFN transcript with a long
5¢-UTR called SasaIFN- a2. In this work, we cloned
two different Atlantic salmon IFN genes from genomic
DNA that encode putative transcripts similar in
sequence to SasaIFN-a1 and SasaIFN-a2. Surprisingly
both genes apparently have the potential to produce
both a short and long transcript because of the loca-
tion of two separate promoter regions, one of which is
present in the 5¢-UTR of the long transcript. To per-
form functional analysis of the Atlantic salmon IFN
promoter region, we fused the complete and truncated
versions of the promoter region to a luciferase reporter
gene and transfected it into Chinook salmon embryo
(CHSE-214) or Atlantic salmon head kidney TO cells.
Promoter activity was measured after stimulation with
poly(I:C) or virus infection.
Results
Cloning of full-length type I IFN genes from
genomic DNA
A genome walking approach was used to clone a 1281-
nucleotide sequence upstream of the SasaIFN- a1 tran-
scription start site. This allowed design of primers that
amplified genomic IFN sequences that expanded from
)1281 of the promoter region (PR) to the polyA signal
by PCR. Two full-length IFN genes, designated Sasa-
IFN-A1 (A1 for short) and SasaIFN-A2 (A2), were

identified in two different BAC clones (GenBank acces-
sion nos DQ354152 and DQ354153). A summary of
nucleotide data on the A1 and A2 gene is shown in
Table 1. Both genes possessed the five-exon ⁄ four-intron
structure found previously in fish type I IFN genes,
although the intron sizes were somewhat different from
those originally found in DNA from Atlantic salmon
[26]. The joined exon sequences of A1 and SasaIFN-a1
cDNA are completely identical, and the joined exon
sequences of A2 and SasaIFN-a2 cDNA have only
three nucleotide differences, possibly because they
represent different alleles (Table 2). In contrast, A1
diverges from SasaIFN-a2, with 10 mismatches, and
A2 and SasaIFN-a1 diverge by 12 mismatches. This
strongly suggests that A1 encodes the SasaIFN-a1
transcript and A2 the SasaIFN-a2 transcript. Overall
differences in A1 and A2, including differences in pro-
moter and intron regions (deletions, insertions and sub-
stitutions), confirm that they represent two different
genes rather than allele variants (Table 1). Two pseudo-
genes were also identified in the screening of BAC
Atlantic salmon interferon promoter V. Bergan et al.
3894 FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS
clones, one having a premature stop codon (accession
no. DQ354154) and one that appeared to be interrupted
by a transposase gene (accession no. DQ354155). The
pseudogenes were not investigated any further in this
work.
Analysis of the promoter regions
The alignment of the 765-bp sequence regions

upstream of the ORFs are very similar in the two
genes except for 10 nucleotide substitutions and two
insertions ⁄ deletions (Fig. 1). This was surprising
because SasaIFN-a1 was originally identified as a short
transcript (829 nucleotides) and SasaIFN-a2 as a long
transcript (1290 nucleotides). We thus expected that
the A1 gene would encode a short transcript and A2 a
long transcript. The present data indicate, however,
that both genes have the potential to encode both tran-
scripts.
A total of six (in A1) or seven (in A2) IRF-binding
elements (IRF-E) were identified in the 765-nucleotide
region upstream of the putative transcription start site
of SasaIFN-a1 (Fig. 1). The motifs conform to the
GAAA(G/C)GAAA(T/C) consensus sequence [29] and
were located at positions )63, )116, )376, )503,
)545, )639, and )669 relative to the putative Sasa-
IFN-a1 transcription start site. Interestingly, the
IRF-E sequences at positions )116 and )545 were
identical and probably bind the same IRF(s). In addi-
tion, we found two potential NFjB-binding sites, one
in close proximity to the SasaIFN-a1 transcriptional
start site ()80) and one more distant ()720) that
appeared to be truncated in the A2 promoter. An
ATF-2 ⁄ c-Jun element, which is essential for activity of
the human IFN-b promoter, was found in the distal
promoter region in close proximity to the IRF-E at
position )557. Moreover, an atypical TATA-box was
located at position )42 in both genes, and two
CCAAT-boxes at positions )296 and )579 in the A1

gene.
In summary, both genes appear to possess two
major regulatory regions:
(a) promoter region I (PR-I) located within 116 nucle-
otides upstream of the short transcript, containing a
noncanonical TATA-box, two IRF-binding motifs and
a putative NFjB-binding motif;
(b) promoter region II (PR-II), located 469–677 nucle-
otides upstream of the short transcript, containing
three to four IRF-binding motifs and an ATF-2 ⁄ c-Jun
element.
The putative salmon IFN promoters thus seem to
have a unique feature, as PR-I controls the synthesis
of a transcript with a short 5¢-UTR and PR-II controls
the synthesis of a transcript with a long 5¢-UTR.
Accordingly, the 5¢-UTR of the long transcript in fact
contains PR-I.
Activity of the A1 and A2 promoters
on poly(I:C) induction
To study the activity of the promoters, we cloned
full-length and deleted versions of the promoters in
front of a promoterless luciferase reporter gene
(Fig. 2A). From the A1 gene the following constructs
were made: pA1(
)135), pA1()202) and pA1()333)
containing only PR-I; and pA1()747) and pA1()1281)
containing both PR-I and PR-II. From the A2
gene, the construct pA2()275) containing only PR-II
was made. The constructs were transfected into
CHSE-214 cells or Atlantic salmon TO cells along with

a constitutively expressed b-gal standard (pJatLacZ)
and then stimulated with poly(I:C) to induce IFN
Table 1. Comparison of the Atlantic salmon genomic A1 and A2
IFN sequences.
Sequence compared
with SasaIFN-a1
Number of nucleotides
A1 A2 Differences
Upstream region 1281 1141 206
5¢-UTR 34 34
a
2
Exon 1 135 135 3
Intron 1 294 294 4
Exon 2 75 75 1
Intron 2 130 130 3
Exon 3 150 150 3
Intron 3 1999 2136 52
Exon 4 78 78 2
Intron 4 338 331 12
Exon 5 90 90 2
3¢-UTR 46
b
236 1
a
Putative 5¢-UTR of A2 was 501 nucleotides based on similarities
to SasaIFN-a2, but for comparison reasons the 5¢-UTR of A1 and
A2 was set to the same.
b
Only partial 3¢-UTR of A1 was cloned

and compared.
Table 2. Comparison of salmon IFN exon sequences to verify link-
age between genomic and cDNA clones. Percentage similarity is
shown in the upper triangle, and number of nucleotide differences
in the lower triangle. The most likely match is highlighted.
Pairwise percentage identity
SasaIFN-a1 SasaIFN-a2 Genomic A1 Genomic A2
Nucleotide differences
SasaIFN-a1 98.2 100.0 98.0
SasaIFN-a2 10 98.2 99.5
Genomic A1 0 10 98.0
Genomic A2 12 3 12
V. Bergan et al. Atlantic salmon interferon promoter
FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS 3895
promoter activity. Figure 2 shows the luciferase
activity from the different constructs relative to b-gal
measurements for poly(I:C)-stimulated or untreated
CHSE-214 (Fig. 2B) or TO cells (Fig. 2C). All con-
structs were induced by poly(I:C) in both cell types.
The overall higher promoter activity observed in
CHSE-214 cells compared with TO cells is probably
due to the fact that poly(I:C) was transfected
into CHSE-214 cells, whereas it was applied extra-
cellularly to TO cells. In CHSE-214 cells, the level of
induction was highest for pA1()1281), pA1()747)
and pA1()202), and lowest for pA2()275). This
indicates that PR-I is most important for poly(I:C)
induction in these cells. In TO cells, all constructs
showed similar levels of relative luciferase activity
after stimulation with poly(I:C). The main difference

from CHSE-214 cells was that pA1()1281), pA1()747)
and pA2()275) all showed relatively high basal
luciferase activity. Accordingly, the level of induc-
tion was highest for pA1()333), pA1()202) and
pA1()135). The minimal promoter showing highest
inducibility in TO cells was thus pA1()135), contain-
ing only PR-I.
The highly inducible minimal promoter construct,
pA1()202), and the full-length construct, pA1()1281),
were next compared for poly(I:C) induction in a time
course study in CHSE-214 and TO cells (Fig. 3). In
CHSE-214 cells, both promoter constructs were hardly
induced at all at 12 h, but showed increasing luciferase
activity at 24 h and 48 h after poly(I:C) treatment
(Fig. 3). At 48 h, the minimal IFN promoter was
induced more than 50-fold, whereas the full-length
promoter was induced only 13-fold (Fig. 3A). The
minimal promoter construct showed similar time kinet-
ics in TO cells, whereas the pA1()1281) construct
showed hardly any induction at any of the time points
(Fig. 3B).
A dose–response curve for poly(I:C) induction of
the minimal promoter construct was established. As
little as 50 ngÆmL
)1
was sufficient to induce the
Fig. 1. Promoter regions of SasaIFN-a1 (A1)
and SasaIFN-a2 (A2) genes. Potential tran-
scription factor binding sites and translation
start codon are boxed. The two putative

transcription start sites are indicated by bent
arrows. Bold lowercase letters or dashes
indicate nucleotides in A2 which are differ-
ent from A1. IRF-core binding motifs that
match the GAAANN consensus are highligh-
ted with bold letters. Putative promoter
regions (PR-I and PR-II) are shaded in grey.
Atlantic salmon interferon promoter V. Bergan et al.
3896 FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS
promoter significantly (14-fold), and 500 ngÆmL
)1
poly(I:C) was sufficient to give maximal induction
(50-fold) of the promoter (Fig. 4).
The optimal conditions to study the salmon inter-
feron promoter were to use CHSE-214 cells and
transfect them with the minimal promoter construct,
A
- 1 b k 1 8 2 .
c u L
c u L
c u L
c u L
c u L
c u L
c u L
E
- F
R I
F N κ κ
κ

B
T A u J - c / 2 - F n
B
/ c u l (

y t i v i t c a e s a r e f i c u l e v i t a l e R β ) l a g -
0
0
5
. 1 1 5 . 2 2 5
.
p G 3 L a b s i c
A p
(
2 -
2 5
7
)
A p
(
1 - 1
5
3
)
A p ( 1 - 2 2 0 )
A p
(
1 -
3
3 3 )

A p
(
1 - 7
7
4
)
A p ( 1
-
1 . 2 k b )
o
P
l
y
I :
C
U n
r
t e a
e
t d
C
1 0 2
3 5
4
c i
s
a b
3 L G
p
2

A
p (
5 7
2
- )
1
A p
(
5 3 1 - )
1 A
p
(
2
0
2
- )
1 A p ( 3 3 3 - )
1
A p
( 7
4
7
-
)
1
A
p (
1 - b k
2
, )

P
o
C :
I
y
l
n
U r t e a t e d
/
c u
l
(
y
t i v
i
t c a
e
s a r e f
i
c u l e v
i t
a
l e
R
β
) l a
g
-
Fig. 2. Analysis of IFN promoter activity in CHSE-214 and TO cells.
(A) Salmon IFN promoter–luciferase constructs including the posi-

tions of IRF-E, NFjB, and ATF-2 ⁄ c-Jun sites relative to the tran-
scription start site (+1). pA1 constructs are from the putative
SasaIFN-a1 promoter, and pA2 is the putative SasaIFN-a2 promo-
ter. (B) CHSE-214 or (C) TO cells were transiently transfected with
the promoter constructs plus a b-gal internal control vector in
24-well plates. At 24 h after transfection, triplicate wells of cells
were treated with 1 lgÆmL
)1
poly(I:C) (and Fugene) or left
untreated. Luciferase and b-gal activities were measured 48 h after
the stimulus using the dual-light luciferase kit. Luciferase activity is
expressed relative to b-gal (mean ± SD from three wells).
0
10
20
30
40
50
60
70
pA1(1.2) pA1(-202) pGL3basic
12 h
24 h
48 h
CHSE cells
A
Fold Induction
B
pA1(1.2) pA1(-202)
pGL3basic

12 h
24 h
48 h
TO cells
Fold Induction
0
2
4
6
8
10
12
14
16
18
Fig. 3. Induction of the minimal IFN promoter ()202) and the full-
length IFN promoter region ()1.2kb) over time in (A) CHSE-214 and
(B) TO cells. Reporter vectors: pGL3basic; absent promoter,
pA1()202); minimal IFN promoter, pA1()1.2kb); full-length IFN
upstream region. Cells were transiently transfected with the repor-
ter constructs plus a b-gal internal control vector in 24-well plates.
At 24 h after transfection, triplicate wells of cells were treated with
1 lgÆmL
)1
poly(I:C) (and Fugene) or left untreated (control). Lucif-
erase and b-gal activities were measured 12, 24 and 48 h after the
stimulus using the dual-light luciferase kit. Fold induction is lucif-
erase activity expressed relative to b-gal of poly(I:C)-treated cells
divided by nontreated control cells (mean ± SD from three wells).
V. Bergan et al. Atlantic salmon interferon promoter

FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS 3897
pA1()202), trigger the promoter with at least
500 ngÆmL
)1
poly(I:C) (complexed with the Fugene
transfection reagent), and read the luciferase values at
48 h after poly(I:C) treatment.
Effect of LPS and virus infection on the salmon
IFN promoter
As the salmon IFN promoter contained a putative
NFjB-binding motif, we wanted to test if lipopolysac-
charide (LPS) was able to induce the IFN promoter.
However, 50 lgÆmL
)1
LPS did not increase luciferase
activity from the minimal IFN promoter in neither
CHSE-214 (Fig. 5) or TO (not shown) cells. The cells
were also treated with poly(dG:dC) (complexed to Fu-
gene), to study whether dsDNA triggered the minimal
IFN promoter, but this was not the case (Fig. 5).
As viruses are known to induce IFN production
through dsRNA intermediates, the effect of virus infec-
tion on the IFN promoter was examined. For this pur-
pose, we used the two most common viral pathogens
of Atlantic salmon, the aquatic birnavirus infectious
pancreatic necrosis virus (IPNV) and the orthomyxo-
virus infectious salmon anemia virus (ISAV). No
increase in promoter activity was detected 48 h after
treatment of CHSE-214 cells with multiplicity of infec-
tion (moi) 5 of live IPNV (Fig. 5). Strong cytopathic

effects occurred in the cells 72 h after infection for
IPNV. As CHSE-214 cells are nonpermissive to most
ISAV strains, we used TO cells to study the effect of
ISAV infection on the IFN promoter. ISAV is strongly
detectable in these cells 48 h after infection and produ-
ces cytopathic effects after a period of 4–7 days [30].
ISAV (moi 5) was able to induce the minimal IFN
promoter 4–5-fold at 48 and 72 h, and more than nine-
fold 96 h after infection in TO cells (Fig. 6). Approxi-
mately 10–20% of the cells showed a cytopathic effect
at 96 h. These results show that, in nonimmune cells,
ISAV is able to turn on the IFN promoter, although
0
5
10
15
20
25
30
35
Untreated Poly(I:C) Poly(dG:dC) LPS IPNV
Fold Induction
Fig. 5. Effect of different stimulants on the minimal IFN promoter
in CHSE-214 cells. Cells were transiently transfected with the
pA1()202) construct plus a b-gal internal control vector in 24-well
plates. At 24 h after transfection, triplicate wells of cells were trea-
ted with 1 lgÆmL
)1
poly(I:C) (and Fugene), 1 lgÆmL
)1

poly(dG:dC)
(and Fugene), 50 lgÆmL
)1
LPS, moi 5 of IPNV, or left untreated. Lu-
ciferase and b-gal activities were measured 48 h after the stimulus
using the dual-light luciferase kit. Fold induction is luciferase activity
expressed relative to b-gal of stimulated cells divided by nontreated
control cells (mean ± SD from three wells).
0
10
20
30
40
50
60
70
80
5000 1000 500 100 50 10 5
Fold Induction
n
g
/ml Poly(I:C)
Fig. 4. Dose–response of poly(I:C) induction on the minimal IFN
promoter ()202). CHSE-214 cells were transiently transfected with
the pA1()202) construct plus a b-gal internal control vector in 24
well plates. At 24 h after transfection, triplicate wells of cells were
treated with different concentrations (5000–0 ngÆmL
)1
) of poly(I:C)
(and Fugene) or left untreated. Luciferase and b -gal activities were

measured 48 h after the stimulus using the dual-light luciferase kit.
Fold induction is luciferase activity expressed relative to b-gal of
poly(I:C)-treated cells divided by nontreated control cells (mean ±
SD from three wells).
0
2
4
6
8
10
12
14
12
24
48 72 96
Fold Induction
Time (h)
Fig. 6. Effect of ISAV infection on the minimal IFN promoter in TO
cells. Cells were transiently transfected with the pA1()202) con-
struct plus a b-gal internal control vector in 24-well plates. At 24 h
after transfection, triplicate wells of cells were treated with moi 5
of ISAV or left untreated. Luciferase and b-gal activities were meas-
ured at 12, 24, 48, 72 and 96 h after the stimulus using the dual-
light luciferase kit. Fold induction is mean luciferase activity
expressed relative to b-gal of stimulated cells divided by nontreated
control cells (mean ± SD from three wells).
Atlantic salmon interferon promoter V. Bergan et al.
3898 FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS
at a very late stage of infection, whereas IPNV is
apparently unable to trigger the IFN promoter.

Effect of 2-aminopurine (2-AP) and pyrrolidine
dithiocarbomate (PDTC) on the salmon IFN
promoter
The NFjB inhibitor, PDTC, and the kinase inhibitor,
2-AP, were used to study the involvement of NFjBin
the poly(I:C)-induced activation of the IFN promoter.
As shown in Fig. 7, PDTC produced  90% inhibition
of poly(I:C)-induced promoter activity at 1 lm and
 70% inhibition at 0.01 lm, which suggests that
NFjB is indeed involved in the poly(I:C)-induced acti-
vation of the salmon IFN promoter. About 55% inhi-
bition of promoter activity was observed with 0.01 and
0.1 mm 2-AP, which indicates that PKR or another
2-AP-sensitive kinase is involved in activation of
salmon IFN promoter.
Long IFN transcripts are produced at very low
levels in TO cells
Northern blot studies have previously shown that tran-
scripts with both short and long 5¢-UTRs are produced
in head kidney of poly(I:C)-treated Atlantic salmon
[26]. To examine whether both transcripts were pro-
duced in cultured TO cells after poly(I:C) induction,
a quantitative RT-PCR assay was designed. Primers
were designed from conserved regions within the ORF
to detect total IFN transcripts, and within the 5¢-UTR
of SasaIFN-a2 to detect long IFN transcripts. Total
IFN transcripts were gradually increased over time in
response to poly(I:C) stimulation, starting from basal
levels of about 4 · 10
4

transcript copies at 0 h, and
reaching almost 10
8
copies at 24 h (Fig. 8A). Tran-
scripts with a long 5¢-UTR were also produced in TO
cells (Fig. 8B), but at very low levels ranging from 14
copies at 0 h to 5044 24 h after stimulation; 2000–
20 000 times below the total IFN transcript quantity.
Total IFN was thus induced about 2300-fold at 24 h
compared with time point 0 h, whereas IFN2 was
induced only 360-fold. This confirms that the proximal
PR-I is most activated upon poly(I:C) induction, at
least in nonimmune cells, producing high amounts of
short IFN transcripts, while the distal PR-II is poorly
induced in TO cells.
Discussion
In this work we have identified the genes encoding
the two previously reported cDNAs SasaIFN- a1 and
0
20
40
60
80
100
120
140
(-) None 1 µ
M
PDTC 0.1 µ
M

PDTC
0.01 µ
M
PDTC
0.1 m
M
2-AP
0.01 m
M
2-AP
0.001 m
M
2-AP
Fold Induction
Fig. 7. Effect of the kinase inhibitor 2-AP and the NFjB inhibitor
PDTC on poly(I:C)-induced expression of the minimal IFN promoter.
CHSE-214 cells were transiently transfected with the pA1()202)
construct and a b-gal internal control vector in 24-well plates. At
24 h after transfection, triplicate wells of cells were treated with
different concentrations of inhibitors followed by poly(I:C) (and
Fugene) treatment or not (control). Luciferase and b-gal activities
were measured 48 h after the stimulus using the dual-light lucif-
erase kit. Fold induction is luciferase activity expressed relative to
b-gal of poly(I:C)-treated cells divided by nontreated control cells
(mean ± SD from three wells).
B
7.00E+03
6.00E+03
5.00E+03
4.00E+03

3.00E+03
2.00E+03
1.00E+03
0.00E+00
40 8 12 24
IFN2
Copy no/ng RNA
Time (h)
A
1.20E+08
1.00E+08
8.00E+07
6.00E+07
4.00E+07
2.00E+07
0.00E+00
40 8 12 24
Total IFN
Copy no/ng RNA
Time (h)
Fig. 8. Quantitative real-time PCR of total IFN transcripts (A) and
long IFN2 transcripts (B) in TO cells at different time points (hours)
after stimulation with poly(I:C) (and Fugene).
V. Bergan et al. Atlantic salmon interferon promoter
FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS 3899
SasaIFN-a2 [26], and studied their promoters. The
genes were previously thought to be distinguished by
their ability to produce different length transcripts.
However, the present data suggest that both genes
have the potential to encode transcripts with either a

short or a long 5¢-UTR. This can apparently be
explained by the presence of two main regulatory
regions in both genes: (a) a proximal promoter region
(PR-I) which includes position )202 to +26 from the
SasaIFN-A1 transcription start site, which controls
synthesis of a short transcript; (b) a distal region (PR-
II) corresponding to position )747 to )413, which
gives rise to a long transcript (Fig. 1). Luciferase
reporter gene assays in two different salmonid cell lines
showed that PR-I was strongly induced by the syn-
thetic dsRNA, poly(I:C), whereas PR-II of the A2
gene was hardly induced at all (Fig. 2A,B). This sug-
gests that PR-I is the main control region.
PR-I contains a putative NFjB-binding element
flanked by two IRF-Es and is thus most similar to the
human and mouse IFN-b and chicken IFN2 promoters
(Fig. 9). In contrast with IFN-b promoters, PR-I lacks
an ATF-2 ⁄ c-Jun element. Comparison of IFN promo-
ter sequences from different vertebrate species suggests
that the essential IRF-Es responsible for virus-induced
expression are located within the 170-nucleotide region
upstream from the ORF, and they all match either the
IRF-1 ⁄ 2 (AANNGAAA), the IRF-3 (
G
⁄
C
GAAANN)
or the IRF-7 (
T
⁄

C
GAAANN) consensus-binding motif
(Table 3) [31–33]. IRF-7 has recently been shown to be
the most important IRF controlling type I IFN expres-
sion, although IRF-3 also contributes substantially
Salmon IFNA1
Zebrafish IFN
-800
-800
-360
-800
-700
-750
-900
Human IFNB
Human IFNA1
Human IFNA4
Chicken IFN1-2
Chicken IFN2
ATF-2/c-Jun
NFκB
IRF-E
TATA-box
Fig. 9. Distribution of transcription factor
binding sites in IFN promoters from selec-
ted species.
Table 3. Comparison of proximal interferon regulatory elements within selected type I IFN promoters. The IRF binding cores are shaded
grey.
Interferon name Position from ORF Interferon regulatory element Accession no
Salmon IFNA1 )96 G

GAAAGTGAAAAC DQ354152
Salmon IFNA1 )149 G
GAAAATGAAAGT DQ354152
Zebrafish IFN )115 G
GAAAGGGAAAAC AJ544820
Zebrafish IFN )151 A
GAAAGTGAAAGC AJ544820
Fugu IFN )97 A
GAAAACGAAATC AJ583023
Fugu IFN )146 T
GAAAAGCAAAGG AJ583023
Tetraodon IFN )148 T
GAAATCCAAAAG AJ544889
Human IFN b )164 G
AAAACTGAAAGG X00973
Human IFN a 1 )125 A
GAAAGTGGAAAT AL353732
Human IFN a 1 )153 A
GAAATGGAAAGT AL353732
Human IFN a 1 )166 G
GAAAGCAAAAAA AL353732
Human IFN a 4b )123 G
AAAATGGAAATT X02955
Human IFN a 4b )164 A
GAAAGCAAAACA X02955
Chicken IFN1-2 )121 A
GGAAGGGAAAGA Y14968
Chicken IFN1-2 )166 C
AAAAGTGAAAGC Y14968
Chicken IFN2 )155 G

AAAAATGAAACA Y14969
Atlantic salmon interferon promoter V. Bergan et al.
3900 FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS
[16,34]. The role of IRFs in induction of fish IFNs is
as yet unknown. In salmonids, only IRF-1 and IRF-2
have been cloned [35].
One of the main differences in mammalian IFN pro-
moters is the presence or absence of an NFjB-binding
element. The putative NFjB-binding sequence in PR-I
(5¢-GGGAAATTCT-3¢) is only one nucleotide differ-
ent from the NFjB-binding site in the human IFN-b
promoter (5¢-GGGAAATTCC-3¢). The NFjB element
in the IFN-b promoter is believed to be essential for
an immediate early response to virus infection [36,37].
IFN-a promoters lack the NFjB element and are usu-
ally activated at a later time point, except for mouse
IFNA4, which also shows early expression in response
to virus infection [38]. The structure of PR-I thus sug-
gests that it controls an early response to virus infec-
tion in Atlantic salmon. A role for NFjB in the
activation of PR-I was further supported by two dif-
ferent inhibitor experiments. First, dose-dependent
inhibition of promoter activity by the NFjB inhibitor
PDTC was shown (Fig. 7). PDTC is a metal chelator
with antioxidant properties which specifically inhibits
NFjB-induced pathways [39]. Secondly, partial inhibi-
tion of promoter activity by 2-AP (Fig. 7), indicates a
role for PKR, a kinase known to be involved in NFjB
activation [40,41]. Although fish PKR has yet to be
reported, PKR-like sequences are present in the Gen-

Bank. On the other hand, it cannot be excluded that
2-AP inhibits another kinase in the IFN signaling
pathway.
The zebrafish IFN promoter, which is the only other
fish IFN promoter characterized so far, is claimed to
contain an NFjB element, but the putative binding
site does not conform with the NFjB consensus [28].
However, the salmon and zebrafish IFN promoters
both contain two IRF-Es at similar position and orien-
tation (Fig. 9). The IRF-E located at position )96 in
salmon and at position )115 in zebrafish differ by only
one nucleotide substitution and they are both present
in antisense orientation. The second IRF-E, at )149 in
salmon and )151 in zebrafish, differs in three nucleo-
tide positions (Table 3). Some species seem to have
IRF-Es and ATF-2 ⁄ c-Jun elements in the distal region
from the major transcription site, but only the salmon
IFN promoters, and perhaps also human IFNA1 pro-
moter, have the unique PR-I and PR-II organization
(Fig. 9). The PR-II has a somewhat different structure
in the two salmon IFN genes (Fig. 1). Both contain
three identical IRF-Es and an ATF-2 ⁄ c-Jun site. How-
ever, only PR-II of A1 contains an NF jB element and
a CCAAT-box. Furthermore, PR-II of A2 contains an
additional IRF-E. Whether the PR-IIs of the two
genes are regulated differently is not yet known.
Promoter constructs that have both PR-I and PR-II or
only PR-II showed a basal expression independent of
poly(I:C) induction (Fig. 2C). Basal expression is, how-
ever, a phenomenon often seen in promoter–reporter

assays containing long upstream regions from the tran-
scriptional start site. The leakiness of transcription is
probably due to lack of negative regulatory structures
such as chromatin packing and ⁄ or methylation. Basal
expression has also been observed for the zebrafish
IFN promoter distal 5¢-flanking regions ()2.2 to
)0.7 kb) in similar experiments [28].
PR-II probably controls synthesis of a transcript
with a long 5¢-UTR. Northern blotting showed that
both transcripts are present in head kidney of
poly(I:C)-treated Atlantic salmon, although the inten-
sity of the long transcript was about half of the short
transcript [26]. In TO cells, however, the long tran-
script was estimated to constitute only 10
-3
)10
-4
of the
total IFN mRNA (Fig. 8). The long transcript thus
appears to be mainly produced in cells of lymphoid
tissues.
The function of the long transcripts is interesting
since both transcripts from one gene are believed to
produce identical proteins. Long 5¢-UTRs are often
associated with genes related to cell growth and differ-
ent types of cellular stress [42,43]. Most of these genes
are poorly translated because of complex secondary
structure within the long 5¢-UTR or they contain small
upstream ORFs that are translated before the major
ORF [44]. Recently, promoter activity was found in

long 5¢-UTR sequences of genes that were believed to
have internal ribosome entry sites (IRES) as a mechan-
ism for translation [45–47]. These alternative promot-
ers were thought to be activated by certain types of
stressors to speed up transcription to smaller and more
efficiently translated mRNA; especially for genes that
were required in small amounts and which could be
toxic if over-produced [44]. This strengthens the idea
that the long 5¢-UTR may have a negative regulatory
function in salmon IFN production. In fact, alternative
promoter options usually have regulatory functions or
are associated with specific cell type expression [48]. As
salmonids have a tetraploid origin, the organization of
the IFN promoter in the two regions may be import-
ant for regulation of the expression levels of IFNs, to
prevent overproduction from the many IFN loci in the
salmon genome.
The alternative promoter found in the 5¢-UTR of
salmon IFN genes suggests an answer to another ques-
tion on the evolution of the intronless IFN genes of
birds and mammals. The intronless type I IFNs of
higher vertebrates most probably originated from a
retro-transposition event involving the transcript of an
V. Bergan et al. Atlantic salmon interferon promoter
FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS 3901
ancestral intron-containing IFN gene. This does, how-
ever, not immediately explain the origin of the IFN
promoters. The present observation of a promoter in
the 5¢-UTR of salmon IFN transcripts suggests that
the promoter of higher vertebrate IFN-a ⁄ b also origin-

ates from the same retroposition event that created the
first intronless IFN gene in vertebrates.
dsRNA is thought to activate the NFjB pathway by
binding to TLR3, RIG-I ⁄ MDA5 or PKR [6,41,49].
NFjB is involved in many different cellular stress
responses [50]. LPS is a well-known inducer of NFjB
activation, but is also thought to be central to TLR4-
mediated activation of IRF-3 to induce the IFN-b gene
in mice [51,52]. The minimal IFN promoter was not
triggered by LPS treatment in either CHSE-214 cells
(Fig. 5) or TO cells (data not shown). This indicates
that LPS alone is not sufficient to give an IFN
response or that the cell types used lack cell surface
receptors for LPS such as TLR4. Results suggest that
NFjB has a role in the activation of the salmon IFN
promoter, but it cannot act alone to initiate transcrip-
tion.
A hallmark of mammalian IFN-a ⁄ b is their rapid
induction by virus infection mainly because of the
recognition of viral dsRNA products [20]. Although
the dsRNA poly(I:C) strongly activated the salmon
IFN promoter, infection with neither IPNV nor ISAV
resulted in convincing activation of the promoter
(Figs 5 and 6). This suggests that both viruses have
developed mechanisms to avoid or inhibit the IFN
promoter in the early critical phases of infection. For
IPNV, this may explain why the virus does not trigger
Mx protein production in cultured cells [30,53]. IPNV
is very sensitive to IFN treatment, and the antiviral
mechanism is at least partly mediated by the Atlantic

salmon Mx1 protein [54]. The chicken birnavirus,
infectious bursal disease virus, has also been shown to
inhibit transcription of IFN genes [55], which suggests
a common immunosuppressive mechanism of this fam-
ily of dsRNA viruses. ISAV, on the other hand, was
able to induce the salmon IFN promoter 96 h after
infection (Fig. 6). However, a previous report has
shown that 5 moi of ISAV resulted in peak Mx protein
expression 24–48 h after infection [30]. This indicates
that ISAV may stimulate the Mx promoter independ-
ent of IFN. ISAV belongs to the same family as influ-
enza viruses, Orthomyxoviridae. The NS1 protein of
influenza virus is a well-known IFN antagonist which
is believed to act upstream of the IFN promoter,
through either NFjB or IRF-3 [56–58]. The ISAV NS
protein is thought to be encoded by segment 7, and
may also represent a candidate antagonist of the
salmon IFN promoter [59]. Taken together, our results
suggest that ISAV and IPNV are successful fish viruses
that have developed strategies to hinder IFN produc-
tion, at least in monocellular systems lacking signals
from the multicellular lymphoid system. However,
the establishment of a salmon IFN promoter reporter
assay gives the opportunity to search for viral proteins
that antagonize IFN production.
Experimental procedures
Cloning of genomic IFN sequences
Atlantic salmon genomic DNA was purified from full blood
of one individual fish by proteinase K digestion and phe-
nol ⁄ chloroform extraction as described [60]. A genomic

DNA library was prepared using the Universal Genome-
Walker kit (Clontech Laboratories Inc., Mountain View,
CA, USA). Nested-PCR of GenomeWalker library DNA
using the primer pairs BR23 and AP1 first and BR22 and
AP2 second was performed according to the kit manual
(primer details are listed in Table 4). The major PCR prod-
ucts were sequenced and new primers were designed to
clone full-length IFN genes from an Atlantic salmon BAC
library purchased from BACPAC Resources (http://bacpac.
chori.org/salmon214.htm). The genomic sequence for Sasa-
IFNa1 was obtained from the 409K8 BAC clone using the
primer pairs Band41AP2 ⁄ BR21, and SasaIFNa2 was found
in the 286F7 BAC using Band41AP2 ⁄ GAT2AAS (Table 4).
The sequences were obtained with long distance PCR using
the Dynazyme EXT PCR kit (Finnzymes Oy, Espoo,
Finland) and cloned into the pCR-XL-TOPO vector
(Invitrogen, Carlsbad, CA, USA).
Promoter constructs
Various constructs of the promoter region of SasaIFN-A1
and SasaIFN-A2 were PCR-cloned into the pGL3-basic
vector (Promega, Southampton, UK) using the primers
specified in Table 4. The full-length luciferase construct,
pA1()1.2 kb), was obtained with the primer pairs K5FBrev
and IFN1()1.2 kb); pA1()747) with K5FBrev and
IFN1()747); pA1()333) with K5FBrev and IFN1()333);
pA1()202) with K5FBrev and IFN1()202); pA1()135)
with K5FBrev and IFN1()135); and pA2()275) with
IFN2(+ 58) and IFN1()747) (Fig. 1A). The b-galactosi-
dase control vector (pJatLacZ) was a gift from J. Jørgensen
(Norwegian College of Fishery Science, Tromsø, Norway),

and contains the LacZ gene under the control of a rat
b-actin promoter [61].
Cells and viruses
TO cells originate from Atlantic salmon head kidney [62]
and were obtained from H. Wergeland (University of
Atlantic salmon interferon promoter V. Bergan et al.
3902 FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS
Bergen, Norway). TO and CHSE-214 cells were grown in
Eagle’s minimal essential medium (EMEM) supplemented
with 1% nonessential amino acids, 2 mml-glutamine,
5% fetal bovine serum, 100 lgÆmL
)1
streptomycin and
200 UÆmL
)1
penicillin G in 5% CO
2
at 20 ° C.
IPNV seriotype NVI-023 [63] was propagated in TO
cells. Virus was harvested from the medium supernatant
and titrated to 5 · 10
7
TCID
50
ÆmL
)1
and stored at )80 °C
until use. ISAV was likewise propagated in TO cells and
titrated to 1.9 · 10
8

TCID
50
ÆmL
)1
.
Chemicals used for promoter activation
or inhibition
Stock solutions were as follows: for synthetic dsRNA,
poly(I:C), 2 mgÆmL
)1
; for synthetic dsDNA, poly(dG:dC),
100 ngÆmL
)1
(both polymers from GE Healthcare, Amer-
sham Biosciences, Uppsala, Sweden); for LPS 2.5 mgÆmL
)1
;
for PDTC 10 mm; for 2-AP 10 mm. All chemicals were pur-
chased from Sigma-Aldrich Inc. (St Louis MO, USA), and
stock solutions were all prepared in NaCl ⁄ P
i
, divided into
aliquots, and stored at )80 °C before use.
Luciferase assays
For CHSE-214 cells, 1.8 · 10
5
cells were seeded for each
well in 24-well plates and transfected at 90% cell confluency
using the lipofectamine 2000 transfection reagent (Invitro-
gen). For each well, 300 ng of the luciferase vector con-

structs and 100 ng of the b-gal vector were complexed with
1 lL Lipofectamine 2000 in 100 lL EMEM without serum.
Transfections were performed in 500 lL minimal growth
medium (EMEM supplemented with 2% fetal bovine
serum) for 24 h before changing to 400 lL fresh minimal
growth medium. Poly(I:C), 1 lgÆmL
)1
, complexed with
3 lLÆmL
)1
Fugene 6 (Roche, Basel, Switzerland) in 100 lL
EMEM without serum was added to the cells for stimula-
tion or they were mock-treated with 100 lL EMEM with-
out serum (control). The poly(I:C)-containing medium was
replaced with fresh minimal medium 24 h after poly(I:C)
treatment.
For TO cells, transfections were performed using nucleo-
fection (Amaxa Biosystems, Cologne, Germany). Cells were
split 3 days before transfection to 70–90% confluency. The
cells were washed in NaCl ⁄ P
i
, trypsinated, and resuspended
in growth medium and centrifuged at 200 g for 10 min.
The cells were resuspended in serum-free medium and
counted. For each transfection, 5 · 10
6
cells were centri-
fuged at 200 g for 10 min and resuspended in 100 lL
nucleofector solution T (Amaxa Biosystems). A total of
15 lg plasmid was added, and the mixture was transferred

to a cuvette and transfected with program T-20 in the
nucleofector device (Amaxa Biosystems). After the transfec-
tion, 500 l L growth medium was added to the cells, left for
10 min, and subsequently seeded in 24-well plates. After
24 h, the cell medium was replaced with fresh medium to
remove unattached cells. The cells were left growing for at
least 7 days before stimulation with 1 lgÆmL
)1
poly(I:C) or
mock-treated with growth medium.
For both cell types, cells were harvested at various time
intervals after stimulation (12, 24 or 48 h) according to the
protocol for the dual-light luciferase kit (Applied Biosys-
tems, Foster City, CA, USA) using 50 lL lysis buffer for
each well. Samples were kept on ice during harvesting and
centrifuged at 14 000 g for 2 min to remove cell debris
before storage at )80 °C. Luciferase and b-gal activities
Table 4. Primer sequences used in the cloning procedures and real-time RT-PCR. The position is given in reference to SasaIFN-a1 (acces-
sion no. AY216594); R, reverse; F, forward.
Primer name Sequence 5¢fi 3¢ Position, orientation Purpose used
BR22 tgcaaataataagacaaatacacgt 56–80, R Genome walking
BR23-nested gctgtttgtttcgctgttagttttc 2–26, R Genome walking
Band41AP2 aaccaaggcctgtatttattaagcatctca )1281–1251, F BAC amplification
BR21 actttataaactggtaagggcgtagc 584–609, R BAC amplification
GAT2AAS cgtttttattcacattttcaatgttattttttcat 765–799, R BAC amplification
K5FBrev att
aagcttgctgtttgtttcgctgttagttttc 2–26, R Promoter cloning
IFN1()1,2kb) att
gctagcaaccaaggcctgtatttattaagca )1281–1251, F Promoter cloning
IFN1()747) att

gctagcagccctgtcaaaactattgactctg )747–722, F Promoter cloning
IFN1()333) att
gctagcgttcacgcgaagttattatcagttg )333–309, F Promoter cloning
IFN1()202) a
gctagcaaggagaatgtgtatagatttactgtga )202–176, F Promoter cloning
IFN1()135) att
gctagctgctgcatgtgctagtctggaaaatg )135–109, F Promoter cloning
IFN2(+ 58) att
aagcttgacattaatttagtgggtttcgttca )439–413, R Promoter cloning
SasaIFN1-F tgcagtatgcagagcgtgtg 77–96, F Real-time PCR
SasaIFN1-R tctcctcccatctggtccag 158–177, R Real-time PCR
SasaIFN2-F ttcgtccaggagaaggagca )171–152, F Real-time PCR
SasaIFN2-R ctgatcaacctaccggaggc )100–81, R Real-time PCR
AS18S-F tgtgccgctagaggtgaaatt – Real-time PCR
AS18S-R gcaaatgctttcgctttcg – Real-time PCR
V. Bergan et al. Atlantic salmon interferon promoter
FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS 3903
were measured using 20 lL of each sample following the
manual of the dual-light luciferase kit (Applied Biosystems).
Cell experiments with inhibitors (PTDC and 2-AP) were
performed as above, except that the inhibitors were applied
2 h before poly(I:C) treatment and left on the cells for
24 h. Then fresh minimal growth medium was added for
another 24 h before cells were harvested 48 h after
poly(I:C) treatment for luciferase measurements.
RNA isolation and cDNA synthesis
Atlantic salmon TO cells were seeded in six-well plates at
5 · 10
5
cells per well 24 h before poly(I:C) treatment. Cells

were then stimulated with 1 lgÆmL
)1
poly(I:C) complexed
with 3 lLÆmL
)1
Fugene 6 (Roche). Triplicate wells were
harvested at time points 0, 4, 8, 12 and 24 h after poly(I:C)
treatment and directly processed using the RNeasy mini kit
(Qiagen, Hilden, Germany). RNA samples were treated
with DNase I using the turbo DNA-free kit (Ambion Inc,
Austin, TX, USA) before measurement of RNA
concentrations with Nanodrop ND1000 (Nanodrop Tec.,
Wilmington, DE, USA). Total RNA (100 ng) was reverse-
transcribed using the TaqMan Reverse Transcription
Reagent kit (Applied Biosystems) in a 10-lL reaction for
1 h at 37 °C.
Quantitative RT-PCR
The SYBRÒ Green PCR kit (Applied Biosystems) was used
for real-time amplification of total IFN, IFN-a2 or 18S
rRNA (internal control). The SasaIFN1 primers were
designed to overlap exon–intron junctions and amplified
PCR products of both a1 and a2 (Table 4). The SasaIFN2
primers amplified regions in the 5¢-UTR of SasaIFN-a2 that
corresponded to the promoter region of SasaIFN-a1. Real-
time PCR was performed using a 9.5 lL cDNA (1 : 10 dilu-
tion) in 25 lL reaction mixture containing 12.5 lL2·
SYBR green PCR Mastermix (Applied Biosystems) and
300 nm of each primer. Real-time PCR was initiated at
95 °C for 10 min to activate the AmpliTaq GoldÒ DNA
polymerase, and then run at 40 cycles of 95 °C for 15 s fol-

lowed by 60 °C for 1 min. Each sample was run in triplicate,
and deviations more than 0.4 Ct between parallels were
rejected. 18S rRNA was used as reference gene and was
diluted 1 : 10 000 to avoid differences of more than 10 Ct
between reference and target amplicons. Quantitation of
IFN transcript per ng total RNA was determined by rela-
ting Ct values to a standard curve ranging from 10
5
to 10
9
copies of a plasmid containing the SasaIFN-a2 cDNA.
Acknowledgements
We thank Dr Jorunn Jørgensen at the Norwegian Col-
lege of Fishery Science, Tromsø, Norway for supplying
the pJatLacZ b-galactosidase vector. This work was
supported by The Research Council of Norway (grants
151938 ⁄ 150).
References
1 Samuel CE (2001) Antiviral actions of interferons. Clin
Microbiol Rev 14, 778–809.
2 Kawai T & Akira S (2006) Innate immune recognition
of viral infection. Nat Immunol 7, 131–137.
3 Haller O, Kochs G & Weber F (2006) The interferon
response circuit: induction and suppression by patho-
genic viruses. Virology 344, 119–130.
4 Magyarics Z & Rajnavolgyi E (2005) Professional type I
interferon-producing cells: a unique subpopulation of
dendritic cells. Acta Microbiol Immunol Hung 52, 443–
462.
5 Zhang Z & Wang FS (2005) Plasmacytoid dendritic cells

act as the most competent cell type in linking antiviral
innate and adaptive immune responses. Cell Mol Immu-
nol 2, 411–417.
6 Yoneyama M, Kikuchi M, Natsukawa T, Shinobu N,
Imaizumi T, Miyagishi M, Taira K, Akira S & Fujita T
(2004) The RNA helicase RIG-I has an essential func-
tion in double-stranded RNA-induced innate antiviral
responses. Nat Immunol 5, 730–737.
7 Andrejeva J, Childs KS, Young DF, Carlos TS, Stock
N, Goodbourn S & Randall RE (2004) The V proteins
of paramyxoviruses bind the IFN-inducible RNA heli-
case, mda-5, and inhibit its activation of the IFN-beta
promoter. Proc Natl Acad Sci USA 101, 17264–17269.
8 Kang DC, Gopalkrishnan RV, Wu Q, Jankowsky E,
Pyle AM & Fisher PB (2002) mda-5: An interferon-
inducible putative RNA helicase with double-stranded
RNA-dependent ATPase activity and melanoma
growth-suppressive properties. Proc Natl Acad Sci USA
99, 637–642.
9 Yoneyama M, Kikuchi M, Matsumoto K, Imaizumi T,
Miyagishi M, Taira K, Foy E, Loo YM, Gale M Jr,
Akira S, et al. (2005) Shared and unique functions of
the DExD ⁄ H-box helicases RIG-I, MDA5, and LGP2
in antiviral innate immunity. J Immunol 175, 2851–2858.
10 Kawai T, Takahashi K, Sato S, Coban C, Kumar H,
Kato H, Ishii KJ, Takeuchi O & Akira S (2005) IPS-1,
an adaptor triggering RIG-I- and Mda5-mediated type I
interferon induction. Nat Immunol 6, 981–988.
11 Seth RB, Sun L, Ea CK & Chen ZJ (2005) Identifica-
tion and characterization of MAVS, a mitochondrial

antiviral signaling protein that activates NF-kappaB
and IRF 3. Cell 122, 669–682.
12 Levy DE, Marie I, Smith E & Prakash A (2002)
Enhancement and diversification of IFN induction by
IRF-7-mediated positive feedback. J Interferon Cytokine
Res 22, 87–93.
Atlantic salmon interferon promoter V. Bergan et al.
3904 FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS
13 Sato M, Suemori H, Hata N, Asagiri M, Ogasawara
K, Nakao K, Nakaya T, Katsuki M, Noguchi S,
Tanaka N, et al. (2000) Distinct and essential roles of
transcription factors IRF-3 and IRF-7 in response to
viruses for IFN-alpha ⁄ beta gene induction. Immunity
13, 539–548.
14 Tailor P, Tamura T & Ozato K (2006) IRF family pro-
teins and type I interferon induction in dendritic cells.
Cell Res 16, 134–140.
15 Kadowaki N, Ho S, Antonenko S, Malefyt RW, Kaste-
lein RA, Bazan F & Liu YJ (2001) Subsets of human
dendritic cell precursors express different toll-like recep-
tors and respond to different microbial antigens. J Exp
Med 194, 863–869.
16 Honda K, Yanai H, Negishi H, Asagiri M, Sato M,
Mizutani T, Shimada N, Ohba Y, Takaoka A, Yoshida
N, et al. (2005) IRF-7 is the master regulator of type-I
interferon-dependent immune responses. Nature 434,
772–777.
17 Raftopoulou M (2005) IRF-7 triggers the interferon
alarm. Nat Cell Biol 7, 459.
18 Izaguirre A, Barnes BJ, Amrute S, Yeow WS, Megjug-

orac N, Dai J, Feng D, Chung E, Pitha PM & Fitzger-
ald-Bocarsly P (2003) Comparative analysis of IRF and
IFN-alpha expression in human plasmacytoid and
monocyte-derived dendritic cells. J Leukoc Biol 74,
1125–1138.
19 Doly J, Civas A, Navarro S & Uze G (1998) Type I
interferons: expression and signalization. Cell Mol Life
Sci 54, 1109–1121.
20 Honda K, Yanai H, Takaoka A & Taniguchi T (2005)
Regulation of the type I IFN induction: a current view.
Int Immunol 17, 1367–1378.
21 Malmgaard L (2004) Induction and regulation of IFNs
during viral infections. J Interferon Cytokine Res 24,
439–454.
22 Sheppard P, Kindsvogel W, Xu W, Henderson K,
Schlutsmeyer S, Whitmore TE, Kuestner R, Garrigues
U, Birks C, Roraback J, et al. (2003) IL-28, IL-29 and
their class II cytokine receptor IL-28R. Nat Immunol 4,
63–68.
23 Kotenko SV, Gallagher G, Baurin VV, Lewis-Antes A,
Shen M, Shah NK, Langer JA, Sheikh F, Dickensheets
H & Donnelly RP (2003) IFN-lambdas mediate anti-
viral protection through a distinct class II cytokine
receptor complex. Nat Immunol 4, 69–77.
24 Robertsen B (2006) The interferon system of teleost fish.
Fish Shellfish Immunol 20, 172–191.
25 Altmann SM, Mellon MT, Distel DL & Kim CH (2003)
Molecular and functional analysis of an interferon gene
from the zebrafish, Danio reri. J Virol 77, 1992–2002.
26 Robertsen B, Bergan V, Rokenes T, Larsen R & Albu-

querque A (2003) Atlantic salmon interferon genes:
cloning, sequence analysis, expression, and biological
activity. J Interferon Cytokine Res 23, 601–612.
27 Long S, Milev-Milovanovic I, Wilson M, Bengten E,
Clem LW, Miller NW & Chinchar VG (2006) Identifica-
tion and expression analysis of cDNAs encoding chan-
nel catfish type I interferons. Fish Shellfish Immunol 21,
42–59.
28 Chen JY, You YK, Chen JC, Huang TC & Kuo CM
(2005) Organization and promoter analysis of the zebra-
fish (Danio rerio) interferon gene. DNA Cell Biol 24,
641–650.
29 Tanaka N, Kawakami T & Taniguchi T (1993) Recog-
nition DNA sequences of interferon regulatory factor 1
(IRF-1) and IRF-2, regulators of cell growth and the
interferon system. Mol Cell Biol 13, 4531–4538.
30 Jensen I & Robertsen B (2002) Effect of double-
stranded RNA and interferon on the antiviral activity
of Atlantic salmon cells against infectious salmon ane-
mia virus and infectious pancreatic necrosis virus. Fish
Shellfish Immunol 13, 221–241.
31 Fujii Y, Shimizu T, Kusumoto M, Kyogoku Y, Tanigu-
chi T & Hakoshima T (1999) Crystal structure of an
IRF-DNA complex reveals novel DNA recognition and
cooperative binding to a tandem repeat of core
sequences. EMBO J 18, 5028–5041.
32 Escalante CR, Yie J, Thanos D & Aggarwal AK (1998)
Structure of IRF-1 with bound DNA reveals determi-
nants of interferon regulation. Nature 391, 103–106.
33 Lin R, Genin P, Mamane Y & Hiscott J (2000) Selective

DNA binding and association with the CREB binding
protein coactivator contribute to differential activation
of alpha ⁄ beta interferon genes by interferon regulatory
factors 3 and 7. Mol Cell Biol 20, 6342–6353.
34 Servant MJ, Tenoever B & Lin R (2002) Overlapping
and distinct mechanisms regulating IRF-3 and IRF-7
function. J Interferon Cytokine Res 22, 49–58.
35 Collet B, Hovens GC, Mazzoni D, Hirono I, Aoki T &
Secombes CJ (2003) Cloning and expression analysis of
rainbow trout Oncorhynchus mykiss interferon regula-
tory factor 1 and 2 (IRF-1 and IRF-2). Dev Comp
Immunol 27, 111–126.
36 Thanos D & Maniatis T (1995) Identification of the
rel family members required for virus induction of the
human beta interferon gene. Mol Cell Biol 15 , 152–
164.
37 Kirchhoff S, Wilhelm D, Angel P & Hauser H (1999)
NFkappaB activation is required for interferon regula-
tory factor-1-mediated interferon beta induction. Eur J
Biochem 261, 546–554.
38 Morin P, Genin P, Doly J & Civas A (2002) The
virus-induced factor VIF differentially recognizes the
virus-responsive modules of the mouse IFNA4 gene
promoter. J Interferon Cytokine Res 22, 77–86.
39 Schreck R, Meier B, Mannel DN, Droge W & Baeuerle
PA (1992) Dithiocarbamates as potent inhibitors of
nuclear factor kappa B activation in intact cells. J Exp
Med 175, 1181–1194.
V. Bergan et al. Atlantic salmon interferon promoter
FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS 3905

40 Fremont M, Vaeyens F, Herst CV, De Meirleir KL &
Englebienne P (2006) Double-stranded RNA-dependent
protein kinase (PKR) is a stress-responsive kinase that
induces NFkappaB-mediated resistance against mercury
cytotoxicity. Life Sci 78, 1845–1856.
41 Zamanian-Daryoush M, Mogensen TH, DiDonato JA
& Williams BR (2000) NF-kappaB activation by dou-
ble-stranded-RNA-activated protein kinase (PKR) is
mediated through NF-kappaB-inducing kinase and
IkappaB kinase. Mol Cell Biol 20, 1278–1290.
42 Kozak M (1987) An analysis of 5¢-noncoding sequences
from 699 vertebrate messenger RNAs. Nucleic Acids
Res 15, 8125–8148.
43 Kozak M (1988) Leader length and secondary structure
modulate mRNA function under conditions of stress.
Mol Cell Biol 8, 2737–2744.
44 Kozak M (2005) Regulation of translation via mRNA
structure in prokaryotes and eukaryotes. Gene 361,
13–37.
45 Han B & Zhang JT (2002) Regulation of gene expres-
sion by internal ribosome entry sites or cryptic promo-
ters: the eIF4G story. Mol Cell Biol 22, 7372–7384.
46 Wang Z, Weaver M & Magnuson NS (2005) Cryptic
promoter activity in the DNA sequence corresponding
to the pim-1 5¢-UTR. Nucleic Acids Res 33, 2248–2258.
47 Kozak M (2005) A second look at cellular mRNA
sequences said to function as internal ribosome entry
sites. Nucleic Acids Res 33, 6593–6602.
48 Landry JR, Mager DL & Wilhelm BT (2003) Complex
controls: the role of alternative promoters in mamma-

lian genomes. Trends Genet 19, 640–648.
49 Alexopoulou L, Holt AC, Medzhitov R & Flavell RA
(2001) Recognition of double-stranded RNA and activa-
tion of NF-kappaB by Toll-like receptor 3. Nature 413,
732–738.
50 Pahl HL (1999) Activators and target genes of Rel ⁄ NF-
kappaB transcription factors. Oncogene 18, 6853–6866.
51 Sankar S, Chan H, Romanow WJ, Li J & Bates RJ
(2005) IKK-i signals through IRF3 and NFkappaB to
mediate the production of inflammatory cytokines. Cell
Signal 18, 982–993.
52 Sakaguchi S, Negishi H, Asagiri M, Nakajima C, Mizu-
tani T, Takaoka A, Honda K & Taniguchi T (2003)
Essential role of IRF-3 in lipopolysaccharide-induced
interferon-beta gene expression and endotoxin shock.
Biochem Biophys Res Commun 306, 860–866.
53 Jensen V & Robertsen B (2000) Cloning of an Mx
cDNA from Atlantic halibut (Hippoglossus hippoglossus)
and characterization of Mx mRNA expression in
response to double-stranded RNA or infectious pancrea-
tic necrosis virus. J Interferon Cytokine Res 20, 701–710.
54 Larsen R, Rokenes TP & Robertsen B (2004) Inhibition
of infectious pancreatic necrosis virus replication by
atlantic salmon Mx1 protein. J Virol 78, 7938–7944.
55 Ragland WL, Novak R, El-Attrache J, Savic V & Ester
K (2002) Chicken anemia virus and infectious bursal
disease virus interfere with transcription of chicken
IFN-alpha and IFN-gamma mRNA. J Interferon Cyto-
kine Res 22, 437–441.
56 Talon J, Horvath CM, Polley R, Basler CF, Muster T,

Palese P & Garcia-Sastre A (2000) Activation of inter-
feron regulatory factor 3 is inhibited by the influenza A
virus NS1 protein. J Virol
74, 7989–7996.
57 Wang X, Li M, Zheng H, Muster T, Palese P, Beg AA
& Garcia-Sastre A (2000) Influenza A virus NS1 protein
prevents activation of NF-kappaB and induction of
alpha ⁄ beta interferon. J Virol 74, 11566–11573.
58 Marcus PI, Rojek JM & Sekellick MJ (2005) Interferon
induction and ⁄ or production and its suppression by
influenza A viruses. J Virol 79, 2880–2890.
59 McBeath AJ, Collet B, Paley R, Duraffour S, Aspehaug
V, Biering E, Secombes CJ & Snow M (2006) Identifica-
tion of an interferon antagonist protein encoded by
segment 7 of infectious salmon anaemia virus. Virus Res
115, 176–184.
60 Sambrook J, Fritsch E & Maniatis T (1989) Molecular
Cloning: a Laboratory Manual, 2nd edn. Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, NY.
61 Johansen A, Collet B, Sandaker E, Secombes CJ &
Jorgensen JB (2004) Quantification of Atlantic salmon
type-I interferon using an Mx1 promoter reporter gene
assay. Fish Shellfish Immunol 16, 173–184.
62 Wergeland HI & Jakobsen RA (2001) A salmonid cell
line (TO) for production of infectious salmon anaemia
virus (ISAV). Dis Aquat Organ 44, 183–190.
63 Santi N, Vakharia VN & Evensen O (2004) Identifica-
tion of putative motifs involved in the virulence of infec-
tious pancreatic necrosis virus. Virology 322, 31–40.
Atlantic salmon interferon promoter V. Bergan et al.

3906 FEBS Journal 273 (2006) 3893–3906 ª 2006 The Authors Journal compilation ª 2006 FEBS