Sieger 2009
Sieger 2009
Sieger 2009
003509
Published by The Company of Biologists 2009
RESEARCH ARTICLE
SUMMARY
The zebrafish genome contains ten genes that encode class II cytokine-like peptides, of which the two that are related most closely to mammalian
interferon gamma (IFN-) were named IFN-1 and IFN-2. Although the zebrafish has become a popular model system to study immune mechanisms,
and although interferons are central regulators of immunity, which zebrafish cytokines correspond functionally to mammalian IFN- has not been
established. We used zebrafish embryos to assay the functions of IFN-1 and IFN-2, and have identified a subset of zebrafish homologs of the
mammalian IFN-responsive genes as IFN- targets in the zebrafish embryo: these genes are upregulated in response to raised levels of either IFN-
1 or IFN-2. Infection studies using two different pathogens show that IFN- signalling is required for resistance against bacterial infections in the
young embryo and that the levels of IFN- need to be regulated tightly: raising IFN- levels sensitizes fish embryos against bacterial infection. Embryos
injected with high doses of Escherichia coli are able to clear the bacteria within a day, and the -interferons are necessary for this defence reaction.
The protective response to Yersinia ruckeri, a natural fish pathogen that is lethal at low doses, also requires IFN-. As in the induction of target genes,
the two interferons act at least partly redundantly. Together with the previously demonstrated type III interferon response, these results show that
the counterparts of the mammalian viral and bacterial interferon-dependent defence functions are in place in zebrafish embryos, and suggest that
Disease Models & Mechanisms DMM
mammals, but significant divergence has been observed as well, Functional studies have been carried out in various fish and in
both in the way the adaptive immune system creates diversity, and cultured fish cell lines. The zebrafish class II cytokine genes whose
among the innate immune recognition and effector components functions have been analysed in vivo are ifn-j1, ifn-j2 (ifnphi2)
that are involved in direct interactions with pathogens (Traver et and ifn-j3 (ifnphi3). ifn-j1 is involved in the activation of virus-
al., 2003; Meijer et al., 2004; Jault et al., 2004; Litman et al., 2001; induced target genes and is necessary for resistance to viral
Yoder et al., 2004; van der Aa et al., 2009; Stein et al., 2007). One infections in the zebrafish embryo (Levraud et al., 2007); it is also
set of fast-evolving components that have a central role in immune involved in resistance against viral and bacterial infections in adult
defence are the interferons and their receptors, of which only a zebrafish (Lopez-Munoz et al., 2009). ifn-j2 and ifn-j3 have been
small subset has been studied in the zebrafish.
Interferons, which are classified into three groups (type I, II and
III) in mammals, belong to the family of class II cytokines, which
mediate immune and inflammatory responses. Apart from the
interferons [namely, the single IFN- (type II), three IFN-s (type
III), the large IFN-a (type I) expansions, and several others], this
family includes the interleukins IL-10, IL-19, IL-20, IL-22, IL-24
and IL-26. Systematic bioinformatic searches and the isolation of
cDNAs encoding interferons or interferon-related peptides from
various fish species have shown that there is a large number of class
II cytokine genes in teleosts. The only clear relationships between
fish and mammalian genes are seen in the interleukin group,
Disease Models & Mechanisms DMM
whereas the interferons are much more divergent (Fig. 1A). None
of the predicted or experimentally characterised fish interferons
closely resemble the mammalian type I or type III interferons.
However, genes have been found in several fish species that are
slightly more similar to IFN- than to any of the other mammalian
interferons, and that have therefore also been named IFN-, even
though the relationship to mammalian interferons is not supported
by high bootstrap values (Stein et al., 2007). Some fish species,
including the zebrafish, have two such genes (Igawa et al., 2006;
Milev-Milovanovic et al., 2006; Stolte et al., 2008), whereas in other
species of fish, only one gene has been found (Zou et al., 2005).
The apparent ifn- homologs are expressed in immune organs such
as the spleen and kidney, as well as in specific immune cells like
natural killer (NK) and T cells (Milev-Milovanovic et al., 2006; Stolte
et al., 2008; Zou et al., 2005). Zou et al. showed that, in cell culture
experiments, the rainbow trout (Oncorhynchus mykiss) IFN- is
functional and leads to the activation of the target gene Ip-10 and
enhanced respiratory burst activity in macrophages (Zou et al.,
2005).
Finally, fish have one or more interferons that are not related
closely to any of the mammalian interferons: these were named
IFN-js (Stein et al., 2007). This group includes an interferon that
has been referred to variously as ‘zebrafish interferon’, ‘IFN A/B’,
‘type I IFN’ and ‘IFN-’ (Altmann et al., 2003; Lutfalla et al., 2003;
Phelan et al., 2005; Robertsen, 2006; Wang et al., 2006; Levraud et
al., 2007), but the phylogenetic distance of this protein from the
mammalian type I and type III interferons does not support this
Fig. 1. Interferons in the zebrafish. (A)Phylogram of the class II cytokines
unambiguously, and it was therefore renamed IFNj1 (ZFIN ID: showing the different members in the mouse, human, Takifugu rubripes (fugu,
ZDB-GENE-030721-3). a species of pufferfish) and zebrafish. This is a highly simplified version of a
In summary, from the sequence comparison alone, it is not phylogram from Stein et al. (Stein et al., 2007). Branches with orthologous
possible to judge which of the fish genes correspond to which genes for the mouse and human are collapsed into one branch and marked in
mammalian interferon genes, or which of the predicted fish orange (e.g. IFN-, IL-19, IL-20, IL-22, IL-24, IFN-, IFN-k, IFN-); branches with
cytokines may fulfil functions similar to those known for orthologous genes for the zebrafish and pufferfish are collapsed into one
branch and marked in turquoise (IL-34, IL-35, IFN-j1), except in the case of
mammalian interferons. However, if the zebrafish is to be used as
IFN-/2 where the branches are not collapsed owing to the difference in
an experimental system for the study of immunity, inflammation names. Species-specific expansions of the a and interferons are also
and disease, then an understanding of the interferon signalling collapsed into one branch. (B)RT-PCR expression analysis of ifn-j1, ifn-1 and
system is essential, and functional studies are necessary to ifn-2 expression during zebrafish development. -actin was used as a control.
determine the roles of the interferons. hpfhours postfertilisation.
572 dmm.biologists.org
Gamma interferon zebrafish RESEARCH ARTICLE
shown recently to mediate resistance against viral infections in adult compared their expression in untreated fish embryos with their
zebrafish (Lopez-Munoz et al., 2009). However, ifn-j1 is the only expression in embryos injected with in vitro transcribed ifn-1 or
zebrafish interferon for which the corresponding receptor could ifn-2 mRNA. The mRNA was injected into 1–2-cell stage embryos
be identified (Levraud et al., 2007). It was shown to act through a and mRNA from pooled embryos was isolated for RT-PCR at 28
receptor composed of the cytokine receptor family member B1 hpf.
(CRFB1) and CRFB5 receptor chains. In view of its antiviral The genes we tested were ifn-j1, nramp, adar, mx, ifi30,
activity, and because of its genomic organization, Levraud and cathepsin D (ctsd), p27 (cdkn1b), cr3, gp91, lmp2 (also known as
colleagues concluded that IFN-j1 is most similar to the mammalian psmb9a) and members of two interferon-dependent immune
type III interferon, IFN- (Levraud et al., 2007). This further GTPases, the gbp and irg families of genes. These potential target
illustrates that the phylogenetic relationships suggested by sequence genes responded in different ways. The majority showed a high
comparisons are not a reliable basis for deducing functional basal expression, which was not influenced by overexpression of
analogies. ifn-1 or ifn-2 (Fig. 2). Except for some irg family members, which
The two ifn- genes in the zebrafish are inducible by both are discussed below, the only genes that responded to IFN- were
lipopolysaccharide (LPS) and polyriboinosinic-polyribocytidylic ifn-j1 and lmp2. Both showed low levels of basal expression and
acid (polyIC) in tissue culture cells (Igawa et al., 2006). One of the were upregulated strongly following injection of ifn-1 or ifn-2
genes, ifn-2 (also known as ifng1-2), has been shown to induce mRNA (Fig. 2).
mxb and mxc in ZF4 cells and mediates resistance against viral The immunity-related GTPase (IRG, p47) family in the zebrafish
infections in these cells (Lopez-Munoz et al., 2009). However, consists of 11 IRGs and three ‘quasi GTPases’ (Bekpen et al., 2005).
treatment of adult zebrafish with IFN-2 failed to mediate resistance Five of the irg genes (irge3, irge4, irgf1, irgf3 and irgf4) were induced
against viral and bacterial infections (Lopez-Munoz et al., 2009). strongly by both IFN-s, and two (irge2 and irgg1) were induced
Disease Models & Mechanisms DMM
To date, the function of ifn-1 (also known as ifng1-1) is completely weakly (Fig. 2). One of the quasi GTPases (irgq1) also showed a
unknown and nothing is known about the expression, induction weak induction, whereas the other two (irgq2 and irgq3) showed
and function of ifn-1 or ifn-2 in the zebrafish embryo. high levels of basal expression that was not altered by ifn-
To determine whether the two predicted zebrafish ifn- genes overexpression (Fig. 2). All of the genes that reacted to interferon
correspond to mammalian gamma interferons in their innate
immune functions, we used the zebrafish embryo to test their roles
in activating target genes and defending the fish against pathogens.
RESULTS
Expression of ifn- during zebrafish development
The zebrafish is able to respond to bacterial infections at an early
stage of development. To test whether interferon might be involved
in the response, we first analysed the stages during development
at which the gamma interferon genes ifn-1 and ifn-2 are
expressed. Real-time PCR (RT-PCR) shows high levels of ifn-1
mRNA in the eggs immediately after they are laid: the mRNA must
be supplied maternally, since zygotic transcription has not started
at this point. ifn-1 mRNA continues to be expressed, although at
lower levels, throughout embryonic development (Fig. 1B). ifn-2
mRNA was not found during the early stages of development, but
was sometimes seen at very low levels at later stages (Fig. 1B).
The expression of interferon in the absence of infection was
unexpected. We wanted to determine whether the expression of
the gamma interferons was truly constitutive or whether it might,
perhaps, be induced by pathogens present in the fish tanks. Thus,
we compared ifn expression in regularly obtained embryos with
expression in germfree embryos (see Methods). We did not detect
any differences in expression levels, suggesting that basal ifn
expression is independent of the presence of microorganisms (data
not shown).
IFN- target genes in the early zebrafish embryo Fig. 2. RT-PCR to identify IFN- targets in zebrafish. Expression of the
The finding that interferons are expressed in the early embryo raises potential target genes was assayed at 28 hpf in untreated control embryos
and in embryos injected with ifn-1 and ifn-2 mRNA, as indicated by the ‘+’
the question of whether the signalling pathway might be active.
signs at the top of the lanes. Injections of mRNA were performed in 1–2-cell
We used an overexpression approach to test whether ifn-1 or ifn- stage embryos. To verify successful injections, the levels of ifn-1 and ifn-2
2 are able to elicit a response. First, we searched for target genes were also tested by RT-PCR. All of the primer pairs used for RT-PCR were able
as a readout for IFN- signalling. We selected a set of zebrafish to amplify the predicted products when tested on genomic DNA. -actin was
homologs of known mammalian IFN-inducible genes and used as a control.
mRNA were induced equally well by ifn-1 and ifn-2, showing that
the two interferons appear to be able to trigger the same responses.
In some cases we saw a stronger response to the combination of
both interferons, which hints at some non-overlap in their
functions, although a particular sensitivity of these genes to the
total IFN levels cannot be excluded.
Since both gamma interferons strongly upregulated ifn-j1
expression, it was possible that their effect on some of the supposed
target genes might, in fact, be indirect, i.e. mediated through the
induction of ifn-j1. To test this, we overexpressed ifn-j1 and
measured the effect on the expression of the postulated gamma
interferon targets (lmp2 and the irg genes). As a control for the
function of ifn-j1 signalling, we tested the expression of the
previously identified ifn-j1 target gene viperin (Levraud et al.,
2007). As expected, we found clear viperin induction upon ifn-j1
overexpression (Fig. 3A). Expression of ifn-1 and ifn-2 was not
inducible by ifn-j1, showing that there is no feedback loop from
ifn-j1 on ifn- expression (Fig. 3A). The irg genes that were not
inducible by ifn- were also not influenced by ifn-j1. Of those genes
that responded to ifn-, only irge3 and irge4 were also inducible by
Disease Models & Mechanisms DMM
ifn-j1 (Fig. 3A), whereas lmp2, irgf1, irgf3 and irgf4 were not. Thus,
the induction of the irg genes in the zebrafish is comparable to the
situation in the mouse, where most irg genes are specific IFN-
targets and only a few are inducible by other interferons (Sorace
et al., 1995; Bafica et al., 2007; Carlow et al., 1998).
The finding that ifn-j1 is able to trigger the expression of irge3
and irge4 raises the possibility that the reaction of irge3 and irge4
to ifn- overexpression is mediated through ifn-j1. To test this, we
overexpressed ifn-1 and ifn-2 and, in parallel, blocked the ifn-j1
signalling pathway using a morpholino (crfb1-MO) that was specific
for the IFN-j1 receptor chain CRFB1, which has previously been
used successfully (Levraud et al., 2007). In this situation we still Fig. 3. Influence of IFN-j1 on IFN- target genes. (A)RT-PCR expression
observed the activation of irge3 and irge4, but not of the ifn-j1 analysis to monitor the effects of ifn-j1 overexpression on IFN- target genes.
response marker viperin (Fig. 3A,B). This shows that irge3 and irge4 The expression was assayed at 30 hpf in control (untreated) embryos, ifn-j1
can be activated by ifn- independently of ifn-j1. It also shows that RNA-injected embryos and embryos injected with ifn-j1 RNA plus a crfb1-
the CRFB1 receptor chain is not involved in IFN- signalling (Fig. specific morpholino (crfb1-MO). -actin was used as a control. (B)RT-PCR to
test the ifn-j1 dependence of irge3 and irge4 induction by ifn-1 plus ifn-2.
3B,C).
IFN-j1 signalling was blocked using a morpholino that was specific for the
Since extrinsic RNA can act as a strong immune stimulus, it was IFN-j1 receptor chain gene crfb1. Expression was tested in control (untreated)
possible that activation of the target genes was not the result of embryos, ifn-1 plus ifn-2 RNA-injected embryos, and ifn-1 plus ifn-2 RNA-
IFN- production from the injected mRNA, but rather the result injected embryos combined with a crfb1 morpholino injection, as indicated. -
of an immune reaction to the injected RNA. To test this, we injected actin was used as a control. (C)Scheme showing the hierarchy of IFN-, IFN-j1
enhanced green fluorescent protein (eGFP) mRNA and monitored and the identified target genes. (D)Response of target genes to control (eGFP)
the effect on gene expression. eGFP mRNA induced the expression mRNA injection.
of ifn-j1, irge3 and irge4, whereas the other genes that were
upregulated by IFN-1 and IFN-2 (irgf1, irgf3, irgf4 and lmp2) did we wanted to determine whether bacterial infections were able to
not respond to eGFP mRNA injection (Fig. 3D). Thus, the activation stimulate interferon signalling. We used two different bacterial
of ifn-j1, irge3 and irge4 may be part of an antiviral response, which strains, E. coli (DH5a) and Yersinia ruckeri, a natural fish pathogen
fits well with the finding that IFN-j1 is necessary for an antiviral (for a review, see Tobback et al., 2007), for this purpose. Both
defence (Levraud et al., 2007). By contrast, irgf1, irgf3, irgf4 and bacterial strains carried plasmids that constitutively expressed the
lmp2 appear to be true IFN- targets. red fluorescent protein dsRED so that their presence in the infected
In conclusion, these data show that the IFN- signalling pathway fish could be monitored easily (Fig. 4C,D). We initially infected the
can be activated in the early zebrafish embryo and that all signalling animals by incubation with bacteria. The bacteria were taken up
components needed to induce specific IFN- target genes must be and were detectable in the intestine, but did not spread from there
present. and did not kill the animals. We therefore turned to injecting bacteria
into the posterior blood island at 28 hpf. Embryos survived the
Response of interferon genes to bacterial infections injection of doses of up to ~3000 colony-forming units (c.f.u.) of E.
Having found that ifn- mRNA is present at low levels in the embryo, coli (DH5a), and were able to clear the infection within 2-3 days
and that the system necessary to respond to interferon is functional, (Fig. 4A). Thus, we used high doses of E. coli (2000-3000 c.f.u.) for
574 dmm.biologists.org
Gamma interferon zebrafish RESEARCH ARTICLE
MO-injected embryos) survived the E. coli infection (Fig. 5B). This widespread (Fig. 5C). The ifn-1+ifn-2 MO-injected embryos
was also reflected in the extent of bacterial clearance in the developed completely normally and did not show any signs of
surviving embryos at 24 hpi. Whereas the infection was moderate retardation before infection (supplementary material Fig. S1). This
in untreated embryos at 24 hpi (Fig. 5C), the ifn-1+ifn-2 MO- is also reflected by the high survival rate of their non-infected
injected embryos showed a strong infection that was more siblings (Fig. 5B). In order to further strengthen the conclusion that
576 dmm.biologists.org
Gamma interferon zebrafish RESEARCH ARTICLE
the failure in the defence of the embryos against bacteria is the produced from only one of the interferon genes is simply not
result of interferon downregulation, it would be desirable to rescue sufficient to protect against highly virulent pathogens. To test
the defects in the morpholino-treated embryos by co-injection of whether higher levels of IFN- can protect the embryo against the
a morpholino-insensitive interferon mRNA. This is impossible lethal effect of Y. ruckeri infection, we overexpressed ifn-1 and
because the expression of interferon from injected mRNA also ifn-2 and assayed the survival of the embryos. However, rather
increased sensitivity to infection (described in more detail below). than protection, we observed a higher death rate in the ifn--
However, the observed redundancy, that is, the fact that resistance overexpressing embryos. Whereas the infected control embryos
was lost only when both morpholinos were used together, makes showed survival rates of ~95% at 1 dpi, only 15% of the ifn- RNA-
it unlikely that the reduced resistance was the result of unspecific injected embryos survived to this time point when infected with
effects of one or both morpholinos. 15 c.f.u. of Y. ruckeri (Fig. 6A). A comparable effect on survival
Knockdown of both ifn-1 and ifn-2 also increased the mortality after infection has been observed for tumour necrosis factor a
caused by infection with Y. ruckeri (15 c.f.u.). Only 44% of these (Tnfa). Roca et al. found that zebrafish show a higher susceptibility
embryos survived the first 24 hours after infection compared with to viral (spring viremia of carp virus) and bacterial (Streptococcus
80% of untreated embryos (Fig. 5D). In contrast to E. coli infections, iniae) infection after Tnfa pre-treatment (Roca et al., 2008). Hence,
Y. ruckeri infection also led to slightly increased death rates in we were interested to see whether the ifn- overexpression
embryos in which ifn-1, ifn-2 or crfb1 had been knocked down influences tnfa expression in the early embryo. When we compared
individually (~65% survival at 24 hpi) (Fig. 5D). At 48 hpi, the effect tnfa expression in control embryos with ifn--overexpressing
of the ifn- single knockdowns was similar to that seen in embryos embryos, we found an upregulation of tnfa upon ifn-
with the double knockdown. overexpression (Fig. 6B). Furthermore, the level of tnfa in ifn--
These experiments show that IFN-1 and IFN-2 can protect overexpressing embryos was enhanced further when the embryos
were infected with bacteria (Fig. 6B). A similar effect was seen for
Disease Models & Mechanisms DMM
irge and irgf group are not (or only minimally) expressed in Thus, the reduction in overall interferon levels might lead to a more
untreated fish embryos. Based on the absence of a functional severe loss in resistance against Y. ruckeri than against E. coli. The
nucleotide-binding motif, it has previously been suggested that the functional equivalence of the two interferons in the response to Y.
function of the Irgq group members has diverged from that of the ruckeri is also consistent with the finding that knockdown of each
other IRGs (hence the name q or quasi GTPases) (Bekpen et al., interferon results in a similar reduction in survival rates, showing
2005). The difference in their regulation in the zebrafish embryo that, in each case, the remaining interferon is equally effective at
supports this notion. inducing a low-level defence.
irge3, irge4, irgf1, irgf3 and irgf4 were induced strongly by One detail in which the two zebrafish interferon sequences differ
interferon; irge2 was activated weakly. Since no cDNAs had been from each other is that IFN-2 contains a nuclear localisation signal
reported previously for irgf4 and irge2, this is the first experimental (NLS) in the same position as in mouse and human IFN-, whereas
support for the prediction of these two genes. irgg1 is activated IFN-1 does not. Nuclear localisation of mammalian IFN- by its
very weakly by interferon; this gene lies in tandem with the NLS has been proposed to be necessary for full biological activity
constitutively expressed irgq1, which showed some activation (Subramaniam et al., 2000), although a mechanism by which IFN
above its basal level upon interferon stimulation. These two genes crosses the plasma membrane to reach the cytosol has not been
have in fact been annotated as a fusion gene where the sequences described. Our finding that IFN-1, without an NLS, is able to
encoding the irgg1 exon are spliced to the downstream irgq1 exon. trigger a signalling cascade as efficiently as the NLS-containing IFN-
The different expression and regulation of the sequences amplified 2 indicates that an NLS is not a general requirement for IFN-
by our two exon-specific primer sets indicates that independent signalling.
transcripts must be made. Furthermore, RT-PCR using primer pairs Overexpression of ifn-1 and ifn-2 in the embryo leads to a
that should amplify transcripts containing both irgg1 and irgq1 did dramatically increased susceptibility to bacterial infection [a
Disease Models & Mechanisms DMM
not yield any such products, suggesting that there are two recent report suggests that this reaction is less pronounced in
independent genes. Nevertheless, it is possible that the response adult fish (Lopez-Munoz et al., 2009)], in a similar way to that
we see for irgq1 after interferon treatment may be because of the shown to occur in mice (Kohler et al., 1993). The increased
promoter activity of the upstream irgg1 gene. sensitivity to E. coli in mice treated with IFN- has been ascribed
The remaining IRG family members, irge1, irge5, irge6 and irgf2 to a septic response. Sepsis is a dysregulation of inflammation in
are not expressed in either the absence or presence of interferon. which an initial systemic inflammatory response, which includes
For irgf2, this finding is consistent with the fact that no expressed the upregulation of inflammatory cytokines such as TNF-a and
sequence tags (ESTs) have been found. An EST does exist for irge1 IL-1, is not balanced properly by a compensatory anti-
and we also detect expression in adult tissues (unpublished data) inflammatory response (reviewed by Buras et al., 2005). It is
showing that it can be expressed, but apparently it does not respond possible that the effects we observe also resemble sepsis. We find
to interferon. that ifn- does indeed induce the expression of the tnfa and il1b
The genes of the irge and irgf groups are the strongest interferon genes, which are upregulated further upon infection, indicating
responders that we have found in the zebrafish embryo. This was an overactivation of the systemic inflammatory response. This
possibly unexpected for the members of the irgf group. The overactivation may also be the cause of the high mortality effects
zebrafish irgf genes resemble the ‘irgc-like’ genes in mammals in that have been described in adult zebrafish treated with Tnfa,
that they have an intron in the same place in the middle of the which show enhanced susceptibility to viral and bacterial infection
main exon that encodes the G-domain (Bekpen et al., 2005) (Roca et al., 2008). However, in contrast to the situation in
(Christoph Rohde and Jonathan C. Howard, unpublished results), mammals, injection of high numbers of E. coli alone does not
and because the mammalian genes in this class are not inducible induce a septic shock, and fish show a high tolerance to LPS
by interferon. (Novoa et al., 2009). This may be related to the fact that fish
recognise bacterial surface components, such as LPS, in a different
Redundancy of ifn-1 and ifn-2 manner from mammals (Sepulcre et al., 2009). The extent to
The zebrafish genome and many other teleost genomes contain which a septic response in fish is related to sepsis in mammals
two ifn- genes that show only a low degree of identity to each will need to be investigated in the future.
other (18.8% amino acid identity between zebrafish ifn-1 and ifn-
2) (Zou et al., 2005; Igawa et al., 2006; Milev-Milovanovic et al., METHODS
2006; Stolte et al., 2008). Our results indicate that ifn-1 and ifn- Morpholino oligonucleotide design, overexpression constructs
2 act redundantly despite their low degree of sequence similarity. and injections
We did not find any target genes that were inducible by one Morpholino oligonucleotides (MO) were synthesised by Gene Tools
interferon and not the other, and knockdown of either interferon and injected into 1–2-cell stage zebrafish embryos. The injection
alone did not interfere with the ability of the embryos to clear E. solution consisted of 0.6 mM of the respective MO, 0.2% Phenol
coli that were injected into the posterior blood island. Red and 0.1 M KCl of which, on average, 2 nl were injected per
One finding that might argue against a complete functional embryo. Injections were carried out using an Eppendorf FemtoJet
equivalence of ifn-1 and ifn-2 is the increased mortality of Y. and micromanipulator.
ruckeri-infected embryos when expression of only one of the two The MO sequences used were: ifn-1, 5⬘-TTTCTGTGCTGT-
interferons is blocked. However, this may simply be the result of GAACCAAGTGATG-3⬘; ifn-2, 5⬘-TGAAGGCGTTCGC-
halving the total interferon levels. Clearly, the embryos are more TAAAGTTAGAGT-3⬘; crfb1, 5⬘-GAGTCACACTTTAGCAAT-
sensitive to Y. ruckeri (it is lethal even at low doses) than to E. coli. GATGAAG-3⬘.
578 dmm.biologists.org
Gamma interferon zebrafish RESEARCH ARTICLE
at 28.5°C in 1/3 zebrafish Ringer solution to score survival or for Lutfalla, G., Roest Crollius, H., Stange-Thomann, N., Jaillon, O., Mogensen, K. and
RNA isolation. The infection in untreated and ifn- knockdown Monneron, D. (2003). Comparative genomic analysis reveals independent
expansion of a lineage-specific gene family in vertebrates: the class II cytokine
embryos was monitored by visualizing dsRed fluorescence using a receptors and their ligands in mammals and fish. BMC Genomics 4, 29.
Leica MZ 16FA stereomicroscope. Meeker, N. D. and Trede, N. S. (2008). Immunology and zebrafish: spawning new
models of human disease. Dev. Comp. Immunol. 32, 745-757.
ACKNOWLEDGEMENTS Meijer, A. H., Gabby Krens, S. F., Medina Rodriguez, I. A., He, S., Bitter, W., Ewa
We are grateful to Juliane Hancke for technical assistance. We thank Julia Hunn, Snaar-Jagalska, B. and Spaink, H. P. (2004). Expression analysis of the Toll-like
Georges Lutfalla and Jonathan Howard for critical reading of the manuscript. We receptor and TIR domain adaptor families of zebrafish. Mol. Immunol. 40, 773-783.
are especially grateful to Röbbe Wünschiers for the statistical evaluation of the Milev-Milovanovic, I., Long, S., Wilson, M., Bengten, E., Miller, N. W. and Chinchar,
survival data. This article is freely accessible online from the date of publication. V. G. (2006). Identification and expression analysis of interferon gamma genes in
COMPETING INTERESTS channel catfish. Immunogenetics 58, 70-80.
The authors declare no competing financial interests. Novoa, B., Romero, A., Mulero, V., Rodriguez, I., Fernandez, I. and Figueras, A.
(2006). Zebrafish (Danio rerio) as a model for the study of vaccination against viral
AUTHOR CONTRIBUTIONS haemorrhagic septicemia virus (VHSV). Vaccine 24, 5806-5816.
D.S. and M.L. conceived and designed the experiments; D.S., C.S. and D.N. Novoa, B., Bowman, T. V., Zon, L. and Figueras, A. (2009). LPS response and
performed the experiments; A.M.v.d.S. constructed fluorescent bacteria; D.S., C.S. tolerance in the zebrafish (Danio rerio). Fish Shellfish Immunol. 26, 326-331.
and M.L. wrote the paper. O’Toole, R., Von Hofsten, J., Rosqvist, R., Olsson, P. E. and Wolf-Watz, H. (2004).
Visualisation of zebrafish infection by GFP-labelled Vibrio anguillarum. Microb.
SUPPLEMENTARY MATERIAL Pathog. 37, 41-46.
Supplementary material for this article is available at Phelan, P. E., Pressley, M. E., Witten, P. E., Mellon, M. T., Blake, S. and Kim, C. H.
http://dmm.biologists.org/lookup/suppl/doi:10.1242/dmm.003509/-/DC1 (2005). Characterization of snakehead rhabdovirus infection in zebrafish (Danio
rerio). J. Virol. 79, 1842-1852.
Received 17 April 2009; Accepted 22 June 2009.
Pozos, T. C. and Ramakrishnan, L. (2004). New models for the study of
REFERENCES Mycobacterium-host interactions. Curr. Opin. Immunol. 16, 499-505.
Altmann, S. M., Mellon, M. T., Distel, D. L. and Kim, C. H. (2003). Molecular and Prajsnar, T. K., Cunliffe, V. T., Foster, S. J. and Renshaw, S. A. (2008). A novel
Disease Models & Mechanisms DMM
functional analysis of an interferon gene from the zebrafish, Danio rerio. J. Virol. 77, vertebrate model of Staphylococcus aureus infection reveals phagocyte-dependent
1992-2002. resistance of zebrafish to non-host specialized pathogens. Cell Microbiol. 10, 2312-
Bafica, A., Feng, C. G., Santiago, H. C., Aliberti, J., Cheever, A., Thomas, K. E., 2325.
Taylor, G. A., Vogel, S. N. and Sher, A. (2007). The IFN-inducible GTPase LRG47 Pressley, M. E., Phelan, P. E., 3rd, Witten, P. E., Mellon, M. T. and Kim, C. H. (2005).
(Irgm1) negatively regulates TLR4-triggered proinflammatory cytokine production Pathogenesis and inflammatory response to Edwardsiella tarda infection in the
and prevents endotoxemia. J. Immunol. 179, 5514-5522. zebrafish. Dev. Comp. Immunol. 29, 501-513.
Bekpen, C., Hunn, J. P., Rohde, C., Parvanova, I., Guethlein, L., Dunn, D. M., Robertsen, B. (2006). The interferon system of teleost fish. Fish Shellfish Immunol. 20,
Glowalla, E., Leptin, M. and Howard, J. C. (2005). The interferon-inducible p47 172-191.
(IRG) GTPases in vertebrates: loss of the cell autonomous resistance mechanism in Roca, F. J., Mulero, I., Lopez-Munoz, A., Sepulcre, M. P., Renshaw, S. A., Meseguer,
the human lineage. Genome Biol. 6, R92. J. and Mulero, V. (2008). Evolution of the inflammatory response in vertebrates: fish
Buras, J. A., Holzmann, B. and Sitkovsky, M. (2005). Animal models of sepsis: setting TNF-alpha is a powerful activator of endothelial cells but hardly activates
the stage. Nat. Rev. Drug Discov. 4, 854-865. phagocytes. J. Immunol. 181, 5071-5081.
Burns, C. E. and Zon, L. I. (2006). Homing sweet homing: odyssey of hematopoietic Sanders, G. E., Batts, W. N. and Winton, J. R. (2003). Susceptibility of zebrafish (Danio
stem cells. Immunity 25, 859-862. rerio) to a model pathogen, spring viremia of carp virus. Comp. Med. 53, 514-521.
Carlow, D. A., Teh, S. J. and Teh, H. S. (1998). Specific antiviral activity demonstrated Sepulcre, M. P., Alcaraz-Perez, F., Lopez-Munoz, A., Roca, F. J., Meseguer, J.,
Cayuela, M. L. and Mulero, V. (2009). Evolution of lipopolysaccharide (LPS)
by TGTP, a member of a new family of interferon-induced GTPases. J. Immunol. 161,
recognition and signaling: fish TLR4 does not recognize LPS and negatively regulates
2348-2355.
NF-kappaB activation. J. Immunol. 182, 1836-1845.
Davis, J. M., Clay, H., Lewis, J. L., Ghori, N., Herbomel, P. and Ramakrishnan, L.
Sorace, J. M., Johnson, R. J., Howard, D. L. and Drysdale, B. E. (1995). Identification
(2002). Real-time visualization of mycobacterium-macrophage interactions
of an endotoxin and IFN-inducible cDNA: possible identification of a novel protein
leading to initiation of granuloma formation in zebrafish embryos. Immunity 17,
family. J. Leukoc. Biol. 58, 477-484.
693-702.
Stein, C., Caccamo, M., Laird, G. and Leptin, M. (2007). Conservation and divergence
Herbomel, P., Thisse, B. and Thisse, C. (1999). Ontogeny and behaviour of early
of gene families encoding components of innate immune response systems in
macrophages in the zebrafish embryo. Development 126, 3735-3745.
zebrafish. Genome Biol. 8, R251.
Igawa, D., Sakai, M. and Savan, R. (2006). An unexpected discovery of two interferon
Stolte, E. H., Savelkoul, H. F., Wiegertjes, G., Flik, G. and Lidy Verburg-van
gamma-like genes along with interleukin (IL)-22 and -26 from teleost: IL-22 and -26 Kemenade, B. M. (2008). Differential expression of two interferon-gamma genes in
genes have been described for the first time outside mammals. Mol. Immunol. 43, common carp (Cyprinus carpio L.). Dev. Comp. Immunol. 32, 1467-1481.
999-1009. Subramaniam, P. S., Larkin, J., 3rd, Mujtaba, M. G., Walter, M. R. and Johnson, H.
Jault, C., Pichon, L. and Chluba, J. (2004). Toll-like receptor gene family and TIR- M. (2000). The COOH-terminal nuclear localization sequence of interferon gamma
domain adapters in Danio rerio. Mol. Immunol. 40, 759-771. regulates STAT1 alpha nuclear translocation at an intracellular site. J . Cell Sci. 113,
Kimmel, C. B., Ballard, W. W., Kimmel, S. R., Ullmann, B. and Schilling, T. F. (1995). 2771-2781.
Stages of embryonic development of the zebrafish. Dev. Dyn. 203, 253-310. Sullivan, C. and Kim, C. H. (2008). Zebrafish as a model for infectious disease and
Kohler, J., Heumann, D., Garotta, G., LeRoy, D., Bailat, S., Barras, C., Baumgartner, immune function. Fish Shellfish Immunol. 25, 341-350.
J. D. and Glauser, M. P. (1993). IFN-gamma involvement in the severity of gram- Tobback, E., Decostere, A., Hermans, K., Haesebrouck, F. and Chiers, K. (2007).
negative infections in mice. J. Immunol. 151, 916-921. Yersinia ruckeri infections in salmonid fish. J. Fish Dis. 30, 257-268.
Levraud, J. P., Boudinot, P., Colin, I., Benmansour, A., Peyrieras, N., Herbomel, P. Traver, D., Herbomel, P., Patton, E. E., Murphey, R. D., Yoder, J. A., Litman, G. W.,
and Lutfalla, G. (2007). Identification of the zebrafish IFN receptor: implications for Catic, A., Amemiya, C. T., Zon, L. I. and Trede, N. S. (2003). The zebrafish as a
the origin of the vertebrate IFN system. J. Immunol. 178, 4385-4394. model organism to study development of the immune system. Adv. Immunol. 81,
Litman, G. W., Hawke, N. A. and Yoder, J. A. (2001). Novel immune-type receptor 253-330.
genes. Immunol. Rev. 181, 250-259. Trede, N. S., Langenau, D. M., Traver, D., Look, A. T. and Zon, L. I. (2004). The use of
Lopez-Munoz, A., Roca, F. J., Meseguer, J. and Mulero, V. (2009). New insights into zebrafish to understand immunity. Immunity 20, 367-379.
the evolution of IFNs: zebrafish Group II IFNs induce a rapid and transient expression Turner, D. L. and Weintraub, H. (1994). Expression of achaete-scute homolog 3 in
of IFN-dependent genes and display powerful antiviral activities. J. Immunol. 182, Xenopus embryos converts ectodermal cells to a neural fate. Genes Dev. 8, 1434-1447.
3440-3449. van der Aa, L. M., Levraud, J. P., Yahmi, M., Lauret, E., Briolat, V., Herbomel, P.,
Lu, M. W., Chao, Y. M., Guo, T. C., Santi, N., Evensen, O., Kasani, S. K., Hong, J. R. Benmansour, A. and Boudinot, P. (2009). A large new subset of TRIM genes highly
and Wu, J. L. (2008). The interferon response is involved in nervous necrosis virus diversified by duplication and positive selection in teleost fish. BMC Biol. 7, 7.
acute and persistent infection in zebrafish infection model. Mol. Immunol. 45, 1146- van der Sar, A. M., Musters, R. J., van Eeden, F. J., Appelmelk, B. J.,
1152. Vandenbroucke-Grauls, C. M. and Bitter, W. (2003). Zebrafish embryos as a model
580 dmm.biologists.org
Gamma interferon zebrafish RESEARCH ARTICLE
host for the real time analysis of Salmonella typhimurium infections. Cell Microbiol. 5, Westerfield, M. (1995). The Zebrafish Book. Eugene, OR: University of Oregon Press.
601-611. Yoder, J. A., Nielsen, M. E., Amemiya, C. T. and Litman, G. W. (2002). Zebrafish as an
van der Sar, A. M., Appelmelk, B. J., Vandenbroucke-Grauls, C. M. and Bitter, W. immunological model system. Microbes Infect. 4, 1469-1478.
(2004). A star with stripes: zebrafish as an infection model. Trends Microbiol. 12, 451- Yoder, J. A., Litman, R. T., Mueller, M. G., Desai, S., Dobrinski, K. P., Montgomery, J.
457. S., Buzzeo, M. P., Ota, T., Amemiya, C. T., Trede, N. S. et al. (2004). Resolution of
van der Sar, A. M., Stockhammer, O. W., van der Laan, C., Spaink, H. P., Bitter, W. the novel immune-type receptor gene cluster in zebrafish. Proc. Natl. Acad. Sci. USA
and Meijer, A. H. (2006). MyD88 innate immune function in a zebrafish embryo 101, 15706-15711.
infection model. Infect. Immun. 74, 2436-2441. Zou, J., Carrington, A., Collet, B., Dijkstra, J. M., Yoshiura, Y., Bols, N. and
Wang, L., Wang, L., Zhang, H. X., Zhang, J. H., Chen, W. H., Ruan, X. F. and Xia, C. Secombes, C. (2005). Identification and bioactivities of IFN-gamma in rainbow trout
(2006). In vitro effects of recombinant zebrafish IFN on spring viremia of carp virus Oncorhynchus mykiss: the first Th1-type cytokine characterized functionally in fish. J.
and infectious hematopoietic necrosis virus. J. Interferon Cytokine Res. 26, 256-259. Immunol. 175, 2484-2494.
Disease Models & Mechanisms DMM