RESEARCH ARTICLE

Molecular Reproduction & Development (2015)

A Comparison Between Egg Trancriptomes of Cod and

Salmon Reveals Species-Specific Traits in Eggs for
Each Species
ANNA WARGELIUS,1* TOMASZ FURMANEK,1 JER ^

OME

MONTFORT,2 AURELIE LE CAM,2 LENE KLEPPE,1
AMELIE JUANCHICH, ROLF B. EDVARDSEN, GEIR LASSE TARANGER , AND JULIEN BOBE2
1,2 1 1

1
Institute of Marine Research, Bergen, Norway
2
INRA, Campus de Beaulieu, Rennes, France

SUMMARY
Fish in use in aquaculture display large variation in gamete biology. To reach better
understanding around this issue, this study aims at identifying if species specific egg
life history traits can be hidden in the unfertilized egg. This was done by investigating
egg transcriptome differences between Atlantic salmon and Atlantic cod. Salmon and
cod eggs were selected due to their largely differencing phenotypes. An oligo
microarray analysis was performed on ovulated eggs from cod (n 8) and salmon
(n 7). The arrays were normalized to a similar spectrum for both arrays. Both arrays
were re-annotated with SWISS-Prot and KEGG genes to retrieve an official gene
symbol and an orthologous KEGG annotation, in salmon and cod arrays this
represented 14,009 and 7,437 genes respectively. The probe linked to the highest 
Corresponding author:
gene expression for that particular KEGG annotation was used to compare expres- Institute of Marine Research,
Nordnesgaten 50, P.O. Box 1870
sion between species. Differential expression was calculated for genes that had an Nordnes, Bergen N-5817, Norway.
annotation with score >300, resulting in a total of 2,457 KEGG annotations (genes) E-mail: anna.wargelius@imr.no
being differently expressed between the species (FD > 2). This analysis revealed that Grant sponsor: EU-Project LIFECYCLE;
immune, signal transduction and excretory related pathways were overrepresented Grant number: FP7222719
in salmon compared to cod. The most overrepresented pathways in cod were related
to regulation of genetic information processing and metabolism. To conclude this
analysis clearly point at some distinct transcriptome repertoires for cod and salmon
and that these differences may explain some of the species-specific biological
features for salmon and cod eggs.

Mol. Reprod. Dev. 2015. 2015 Wiley Periodicals, Inc. Published online in Wiley Online Library
(wileyonlinelibrary.com).
Received 6 February 2015; Accepted 2 April 2015 DOI 10.1002/mrd.22487

INTRODUCTION eggs from different species, through for example compar-

Fish eggs develop at a wide range of conditions depend-
ing on species. The conditions include variable levels of
salinity, temperature and depth. Eggs from different spe-
cies have probably evolved towards their prospective en-
vironments after fertilization, in terms of for example, size,
buoyancy (dermersal and floating eggs), robustness of egg
envelope, and immune defense. Comparing unfertilized
ative transcriptome or proteome studies may give some
Abbreviations: FD, fold change; GEO, gene expression omnibus; Id, iden-
tities; NARA, Norwegian Animal Research Authority; qPCR, quantitative PCR;
KEGG, kyoto encyclopedia of genes and genomes; RIN, RNA integrity; SWISS,
prot-protein sequence database

2015 WILEY PERIODICALS, INC.

Molecular Reproduction & Development
lead to which processes in the egg that can predict a
species specific feature such as egg size, buoyancy and
immune functions. Studies like this might reveal basic
biological features of fish eggs and possibly predict better
rearing protocols. This knowledge is of high significance
since early developmental steps are a known bottleneck for
production of fish in aquaculture, especially for marine eggs
(Kjorsvik et al., 1990; Bobe and Labbe, 2010).
Throughout oogenesis maternal RNAs are provided by
the mother to the fish egg. These RNAs control the first
developmental steps until the onset of zygotic transcription,
which in fish occurs at gastrulation (Schier, 2007). Maternal
RNAs provide key proteins for early stages and are there-
fore crucial for normal development of the embryo. Dis-
turbances of the regulation of these transcripts may have
severe consequences for the fish embryo such as arrested
development (Giraldez et al., 2006). A way to measure
developmental potential of the egg is to monitor the content
and diversity of maternal RNAs in the egg and such studies
have tried to identify egg quality markers in both rainbow
trout and Atlantic halibut (Aegerter et al., 2005; Bonnet
et al., 2007; Mommens et al., 2010).
Comparison of egg maternal RNA repertoires have
revealed conserved properties between unfertilized eggs
of human, frog, cow, and mouse eggs (Sylvestre et al.,
2013). Also another study has revealed high conservation
of oogenetic processes in between four vertebrates (Char-
lier et al., 2012). The above mentioned articles describe
studies where the aim of was to find evolutionary conserved
mechanisms between species. This type of comparisons
may also reveal differences in between species, such as
trait associated with egg quality. One such differing egg trait
has already been observed by our group in Atlantic cod and
by another group in Atlantic halibut (Bai et al., 2007; Kleppe
et al., 2012). Ovulated eggs from these species contains an
unusual high abundance of mitochondrial RNA transcripts,
at 3550%. It is further likely that this feature is linked to
high energy consumption during early developmental
stages in these species. It is however impossible to
Figure 1. Spectra of all array probes for cod (black line/spots) and salmon (gray line/spots) in (a) an un-normalized array and (b) a normalized array
between species. The number of probes on the array (x-axis) is plotted versus the intesity of the probe (y-axis).
2 Mol. Reprod. Dev. (2015)
WARGELIUS ET AL.
conclude if is a particular feature of marine eggs. But by
comparing the differences between transcriptomes of spe-
cies with different reproductive environments or egg char-
acteristicssuch as marine vs. fresh water eggs, large vs.
small eggs, demersal (sinking) vs. pelagic (floating) eggs
may reveal if the transcriptome contains species-specific
signals that confer these traits.
One such comparison could be between the large
(56 mm in diameter), demersal, fresh water egg of Atlantic
salmon (Gorodilov, 1996) a small (1.21.3 mm in diame-
ter), pelagic marine egg of Atlantic cod (Kjesbu et al., 1992).
These two species show a high degree of functional diver-
gence when it comes to aquaculture. The salmon egg is
robust and the fertilized egg displays high survival the cod
egg is fragile with low survival rate after fertilization under
aquaculture conditions.
The aim of the study was to find species specific traits in
the egg transcriptomes of Atlantic cod and Atlantic salmon.
In order to find potential differing traits, comparison of egg
transcriptomes of these two fish were conducted to resolve
differentially expressed genes that in turn were linked to
different pathways. Overrepresented pathways in either
salmon or cod eggs may be related to specific processes
contributing to their functional divergence.
RESULTS AND DISCUSSION
When comparing transcriptomes from different species
using microarray there were some technical issues to be
considered. The first obstacle was differences in the rela-
tive microarray signal intensities obtained from microarrays
run for cod and salmon. The strength of the signal was
stronger in the cod array (Fig. 1a). The microarray spectrum
was adjusted to a similar spectrum by multiplying the
intensities in the salmon array with 1.59 (Fig. 1b). In the
whole analysis the KEGG annotation in salmon repre-
sented 14,009 genes (44 K probe array) and in cod
7,437 genes (44 K probe array). Another problem with
 COMPARISON
the microarray design was that the oligos used were not
designed to be located in the same region of the individual
genes. Thereby the relative gene expression in between
the two species genes could have been biased by the
location of the probe in the two different species. However
the microarrays had been designed by the same company
pipeline (Agilent), which at least uses the same algorithm to
calculate the probes thereby possible reducing the variabil-
ity in the microarray comparison. For future comparative
studies it will therefore be more usable to use RNAseq as
the method measures the expression in the whole gene and
not only where the oligo binds (Marioni et al., 2008).
As the salmon genome has undergone a relatively
recent whole genome duplication (Berthelot et al., 2014)
this species displays many more duplicated genes com-
pared to cod. This displayed a general problem in the data
set as the KEGG annotation could have 10 probes associ-
ated to it in salmon while the same gene symbol in cod had
only 2 probes associated with this symbol. In addition, due
to the incompleteness of both genome sequences it is also
possible that the same gene had several probes associated
with it (Davidson et al., 2010; Star et al., 2011). To solve this
problem, the probe linked to the highest gene expression
for that orthologous KEGG annotation was used to com-
pare expression in between species. This was done since it
is most likely that the most pronounced expressed ortho-
logue has a significant function within the egg. It was also
the only option we had in a study where we compared the
cods diploid genome (Star et al., 2011) towards the partially
tetraploid salmonid genome (Berthelot et al., 2014). To be
more certain that orthologous genes were compared, dif-
ferential expression was only calculated for genes with an
annotation score above 300. To reassure the certainty of
gene expression five genes with interesting differential
expression level between cod and salmon were also as-
sayed by qPCR. Four out of five genes displayed similar
relative difference in gene expression between array
Figure 2. Normalized microarray signal intensities and qPCR results for five genes showing differential expression (y-axis) between cod and
salmon eggs. The x-axis shows either the (a) relative signal intensities obtained from the microarray for the selected five genes in cod (black bars)
and salmon (white bars) or (b) the fold-change differences detected by qPCR for the same genes between the two species. Data are presented as
mean  standard error.
Mol. Reprod. Dev. (2015)
OF EGG TRANCRIPTOMES FROM COD AND SALMON
intensities and qPCR (Fig. 2). One of them ppara showed
less relative correlation between microarray and qPCR.The
fifth gene carnitine O-palmitoyltransferase (cpt1) was de-
termined to be an invalid KEGG annotation based on that
the probe mapped to several places in the current gene
prediction in the salmon genome (Wang et al., 2014),
therefore representing an erroneous microarray spot in
the data set. To get rid of more incorrect data points, the
KEGG annotations shared between cod and salmon were
filtered according to a previously published protocol for
salmon (Wang et al., 2014) and all probes displaying
more than three hits in the salmon transcriptome was
filtered away. In the cod microarray similar filtering has
previously been performed, therefore none multiple tran-
scriptome mapping sites were found for cod (Kleppe et al.,
2014). This analysis left us with a total of 3,262 KEGG
annotations which could be compared between species.
2,457 of these KEGG IDs had a fold change >2 when
comparing cod and salmon (Supplementary Table S2).
In both species there were a number of genes that
showed differential gene expression between cod and
salmon using the orthologous KEGG annotation list (Fig.
3). It was observed that about 25% of the genes display a
similar expression level in between species (Fig. 3). Inter-
estingly there seems to be a higher fraction of genes in cod
that display much higher fold change difference compared
to salmon (5100 fold change, Fig. 3). KEGG pathway
analysis was performed on KEGG gene list with fold
changes >2 for cod and salmon (Supplementary Table
S2). Pathways represented in Figure 4, show those path-
ways that contained an overrepresentation of factors up-
regulated in cod (Fig. 4a) and salmon (Fig. 4b). In cod the
most differentially expressed pathways were related to
genetic information processing and metabolic processes.
This upregulation of metabolic pathways might be related to
the energy requirement to fulfill the coming fast numerous
cell divisions. This has already been suggested as the cod
3
Molecular Reproduction & Development
Figure 3. Fold-change difference of KEGG annotated genes between
salmon and cod. The fold-change differences (y-axis) and the number
of KEGG identities (x-axis) are shown for cod (black bars) and salmon
(white bars). Genes showing no difference between species are also
shown (grey bars).
egg transcriptome contains 50% mitochondrial transcript
(Kleppe et al., 2012). Likewise, other marine fish have also
shown to have a high level of mitochondrial transcripts in
the early embryo (2-cell stage) of halibut (Bai et al., 2007)
and in the oocyte of Senegalese sole (Tingaud-Sequeira
et al., 2009), while no information is available on this for
Figure 4. Most overrepresented pathways in cod and salmon eggs. The pathways overrepresented in (a) cod and (b) salmon are listed (y-axis),
and the number of transcripts with a fold change >2 for cod (black bars) and salmon (white bars) are plotted (x-axis). Unaffected transcripts are
those represented by a fold change <2 (gray bars).
4 Mol. Reprod. Dev. (2015)
WARGELIUS ET AL.
salmon (Mommens et al., 2010). Such metabolic activities
could be related to energy metabolism such as the utiliza-
tion of fat and amino acids in yolk (Fig. 4a). Also key
elements in the fatty acid metabolism cascade were over-
represented in cod including such as the peroxisome pro-
liferator-activated receptor alpha transcript (ppara, Fig. 2).
Whereas, ppara encodes a nuclear receptor known to
increase peroxisome activity in the cell leading to break-
down of long fatty acids into shorter which can be utilized to
create energy by the mitochondria (Qi et al., 2000). The
reverse expression was found for Prostaglandin G/H syn-
thase 2 (pghs1, Fig. 2). This is one of two genes that encode
an enzyme that catalyzes the conversion of arachinodate to
prostaglandin. This has further been linked to the matura-
tion of oocytes in fish and in other vertebrates. Whereas the
follicular production of prostaglandins (PGs) is required for
ovulation (Goetz, 1991) and in certain species of fish, PGs
mediate the resumption of meiosis and the maturation of
the oocyte. In salmon eggs the expression of pghs1 was
much higher than in cod pointing at dissimilar mechanism of
final maturation of eggs (Lister and Van Der Kraak, 2009).
Likewise there was detected an overrepresentation of
meiosis related transcripts in eggs of cod (Fig. 4a). This
may also be related to the general life history traits of eggs in
 COMPARISON
salmon and cod. Whereas final maturation is a longer
process in salmon that is a single spawner compared to
cod, which is a batch spawner, performing final maturation
every three days in the spawning season.
The pathways that were the most overrepresented in
cod were the spliceosome and RNA transport pathways
(Supplementary Fig. S1). These essential processes of
RNA maturation both in nucleus and in mitochondria re-
quire energy, which is provided among others by the RNA
helicases including the pre-mRNA-splicing factor ATP-de-
pendent RNA helicase prp2 (prp2, Fig. 2 and Supplemen-
tary Fig. S1). It may be possible that the vast amount of
mitochondrial RNA found in cod eggs needs immediate
maturation, and this fact may explain the big difference
between cod and salmon. Another explanation may lay in
the nature of regulation of RNA translation, it might be
possible that spliceosomal and RNA transport activities
in cod are at a higher level utilized in means of regulation
of translation of target maternal RNAs while other mech-
anisms is possibly applied to a greater degree in salmon.
For example in mouse it has been shown that translational
control of maternal RNAs can be regulated through splice
mechanisms (Seli et al., 2008). Further investigation is
however needed to confirm if any of these theories are true.
In salmon eggs the most overrepresented pathways
were related to immune, excretory, nervous system and
signal transduction pathways (Fig. 4b). The significant
differences in immune pathways represented in between
cod and salmon eggs (Fig. 4b), probably coincides with
what has already been observed for cod, namely that the
cod is lacking a major players in the immune system
including MHC II and CD4 (Star et al., 2011). Thereby,
since cod lacks these major players it has been suggested
that cod has an evolutionary different response to patho-
gens. Also in accordance with our results, immune gene
expression monitoring of cod eggs have revealed very few
RNAs encoding known defense mechanisms in the cod egg
(Seppola et al., 2009). The pathways overrepresented in
salmon were related to the innate and humoral immune
system. Even though the humoral system is probably not
active yet it can be remains from the mother immune
system that is reflected in for example T-cell receptor
signaling pathway. Otherwise in eggs an active immune
system against for example viruses can be mediated
through the innate immune pathway. One factor that for
example triggers the defense against dsRNA viruses is the
interferon regulatory factor 7 (irf7, Fig. 2). As expected from
the KEGG pathway analyses this gene displayed a higher
expression in salmon compared to cod, further strengthen-
ing the notion that salmon eggs seems to have a more
refined and effective immune defense compared to cod
eggs. This may also be related to the life history trait of cod
and salmon eggs as salmon eggs may be much more
exposed to viruses and bacteria. Since salmon eggs are
stationary and reside in the same environment every year
compared to cod eggs which are pelagic and therefore
dissimilar disease pressure is put on these eggs every year
which do not favor specific selection on a specific early
developmental immune system.
Mol. Reprod. Dev. (2015)
OF EGG TRANCRIPTOMES FROM COD AND SALMON
To conclude, the results obtained in this study suggest
that cod early development is highly divergent in relation to
genetic and metabolic processes compared to salmon.
Salmon eggs seem to contain an overexpresession of
immune, excretory, and signal transduction processes in
relation to cod. In the future, 287 more egg transcriptomes
will be characterized which will allow for further compar-
isons between species. Such multiple comparisons will with
a high probability narrow down the number of processes
found and almost certainly lead us towards key molecular
mechanisms behind egg life-history traits.
MATERIALS AND METHODS
Animals and Collection of Eggs
All samplings in this study were performed according to
the guidelines approved by the Norwegian Animal Re-
search Authority (NARA). The oxygen saturation was
>80% in the outlet for both cod and salmon broodstock.
Domesticated Atlantic cod (four generations in captivity)
were of different age (24 years), and had been kept in
netpens under ambient temperature (average 8.18C) and
light conditions, and fed daily using a commercially avail-
able dry feed. Ovulated Atlantic cod eggs were collected by
manual stripping in March 2009, according to Kleppe et al.
(2013). Briefly, unfertilized cod eggs from eight ovulated
females (1 ml eggs, n 5 per fish) were collected and
frozen in liquid nitrogen. Eggs from Atlantic salmon brood-
stock (Aquagen strain) were obtained at the Institute of
Marine Research (IMR), Matre (618 N) in October 2009.
The salmon females were exposed to ambient temperature
(average 10.18C) and photoperiod. Broodstock were
checked weekly for signs of ovulation (every 610
days). Unfertilized, ovulated salmon eggs from seven ovu-
lated females (10 ml eggs, n 5 batches per fish) were
collected and frozen in liquid nitrogen for microarray
analyses.
RNA Extraction for Microarray and qPCR
Analyses
For the microarray and qPCR analysis, total RNA was
extracted from 20 mg (about 10 eggs per sample) of Atlantic
cod eggs by applying RNeasy1 Mini Kit (QIAGEN, Oslo,
Norway). Genomic DNA was removed from RNA samples
using a Turbo DNA-freeTM Kit (Ambion, Austin, TX). Total
RNA was extracted from about 20 Atlantic salmon eggs for
microarray analysis, while only 5 eggs where used for
qPCR analysis. RNA was purified from salmon samples
using Tri-reagent, and further purified using a Nucleospin
RNA II kit, as previously described for rainbow trout (Bonnet
et al., 2007). A NanoDrop1 NP-1000 spectrophotometer
(NanoDrop Technologies, Wilmington, DE) was used to
measure the quantity and quality of the RNA samples. The
quality of the RNA samples was also checked using a 2100
Bioanalyzer (Agilent Technologies, Inc., Santa Clara, CA);
no samples with RNA integrity (RIN) values below 8.2 were
accepted for further analysis.
5
Molecular Reproduction & Development
Microarray Analysis
An Agilent oligo microarray experiment was performed
on labelled RNA extracted from cod (n 8) and salmon
eggs (n 7). A 44 K array consisting of about 21,000 genes
spotted in duplicate and an additional 2,000 genes spotted
once was used for cod (Kleppe et al., 2013). A 44 K array
with 42,527 genes spotted once was used for salmon
(Jantzen et al., 2011). Two hundred nanograms of total
RNA were used for each array. The efficiency of cDNA
synthesis and labeling was measured using a Nano
Drop spectrophotometer (NanoDrop Technologies). The
arrays were both hybridized using a one-color Agilent
microarray protocol, according to the manufacturers in-
structions (Low Input Quick Amp Labeling, version 6.5,
May 2010) (Agilent Technologies). An Agilent B Scanner
(G2305B) was used to scan the arrays. More details
on the array experiments are provided in a MIAME-compli-
ant description of the microarray studies at the
arrayexpress database (http://www.ebi.ac.uk/arrayex-
press), with accession number for cod (E-MTAB-2170)
and for salmon under Gene Expression Omnibus
(GEO) accession (http://www.ncbi.nlm.nih.gov/geo/que-
ry/acc.cgi, GSM1448872, GSM1448873, GSM1448874,
GSM1448876, GSM1448882, GSM1448883, and
GSM1448885).
After microarray scanning, the raw data were read and
processed by Feature Extraction software (Agilent Technol-
ogies) and imported into J-Express (Dysvik and Jonassen,
2001) for analysis. A total of 43,749 and 43,557 probes were
assessed for salmon and cod, respectively. To normalize
gene expression for each species, data were quantile nor-
malized (Bolstad et al., 2003), and missing-value replace-
ments were predicted using LS impute Adaptive (Bo et al.,
2004). The array data for each species were scaled to a
similar spectrum by multiplying all probe values in the salmon
microarray by 1.59 (Fig. 1), which was calculated from the
difference in median expression between the species-spe-
cific arrays. Both arrays were re-annotated to retrieve an
orthologous KEGG annotation (http://www.genome.jp/kegg/
ko.html) and a SWISS-Prot ID (http://web.expasy.org/docs/
swiss-prot_guideline.html) for all probes on the arrays.
Gene expression was grouped based on the ortholo-
gous KEGG annotation, which could be subsequently used
to link a species unique genes to specific pathways. The
BLAST score for a KEGG annotation was set above 300 to
avoid comparison of unrelated genes. To filter out errors in
the salmon data set, probes represented on the salmon
array were BLASTed against the first salmon genome
assembly (Acc. No. AGKD00000000.1) and the current,
predicted salmon transcriptome (Wang et al., 2014). As the
microarray array used in this study was produced based on
previously released genomic data from salmon (Jantzen
et al., 2011), many of the probes on this array hybridize to
several places in the salmon genome assembly rather than
being unique to the genome. Therefore, probes with more
than three hits in the genome/transcriptome were filtered
out of the final list. Probes representing the same gene were
also filtered out if their expression values differed by more
6 Mol. Reprod. Dev. (2015)
WARGELIUS ET AL.
than fourfold. Finally, genes were clustered based on gene
symbols for each species, and the expression correspond-
ing to the highest expression for that gene symbol was used
when analyzing differential expression between cod and
salmon eggs. The reference list of genes were further
selected using NOISeq (Tarazona et al., 2011) with a
threshold of 0.95. This analysis retrieved 3,262 KEGG
identities to compare.
All pathways containing genes of interest were visualized
by mapping the up- and down-regulated genes in different
colors on the KEGG pathway map (Kanehisa et al., 2012).
The KEGG-annotation BLAST score was manually checked
for each gene of interest; this score was set to be above 300
to avoid comparison of unrelated genes.
Quantitative PCR
Gene expression profiles were measured on selected
genes (Table 1) by qPCR. Egg from cod and salmon were
sampled in replicate from three different females (n 3) per
species. As internal reference is not applicable to eggs, the
external reference gene rabbit hemoglobin alpha (Hba),
was applied for the qPCR analysis (Kleppe et al., 2012).
One picogram of Hba mRNA (Sigma, Oslo, Norway) was
added to an equal mix of RNA obtained from cod and
salmon before cDNA synthesis (Kleppe et al., 2012). A
Superscript VILO cDNA synthesis kit was used for cDNA
production (Invitrogen, Carlsbad, Germany) from 100, 50,
25, 12.5, and 6.25 ng RNA per sample. To create a stan-
dard for qPCR, results from the 25 ng RNA samples were
used to calculate relative gene expression.
All primers were designed applying the software Primer
Express 3.0 (Applied Biosystems, Foster City, CA), and are
listed in Table 1 (Kleppe et al., 2012). qPCR was performed
on a SDS 7900HT Fast Real-Time PCR system (Applied
Biosystems, Oslo, Norway) with the following thermal cy-
cling conditions: 958C for 20 sec, and 40 cycles of 958C for
1 sec followed by 608C for 20 sec. Fast SYBR1 green PCR
Master Mix (Applied Biosystems) was used in PCR reac-
tions. All PCR reactions were run in duplicate. Using the
standard-curve method, PCR efficiencies were verified to
be approximately equal between target and reference
genes (using 100, 50, 25, 12.5, and 6.25 ng RNA). The
PCR efficiencies per primer pair are listed in Table 1.
Melting-curve analysis revealed that each primer pair pro-
duced only one product.
Twenty-five nanograms of RNA were used to calculate
relative abundance of transcripts. No-template controls
were run for each gene at each plate. The comparative
Ct (DCt) method was used to calculate relative expression
level (Applied Biosystems). All data were normalized to
rabbit Hba mRNA. For each gene, the data were calibrated
to the sample with the lowest value for DCt. One-way
ANOVA with the post test Tukeys multiple comparison
test was applied on log-converted qPCR values to reveal
significant differences between groups. GraphPad Prism
5.04 (GraphPad Software, Inc., La Jolla, CA) was used to
perform statistical tests and to make graphs. A P-value of
0.05 was considered significant.
 COMPARISON OF EGG TRANCRIPTOMES FROM COD AND SALMON

TABLE 1. Primers Used for qPCR

Accession no. Species Gene Primer sequence Efficiency

5685818 a
salmon ss prp2 (or ac dhx34) 50 -CCCAAAGACAAGGTGGTTCAC 105%
50 -TCCGGGCACCACAAAGTT
5678976a salmon ss ppara 50 -CGGTCACAGAGCTAACCGAGTT 99.6%
50 -GCAGAGTCACCTGGTCGTTCA
5629907 a
salmon ss cpt1 50 -CAGAGACGGTGCGTTCCTGTA 94.5%
50 -GGGTTGTCTTCCACCATGGA
5638067a salmon ss pghs1 50 -TTCTACCCTGCCCTCATGCT 90.1%
50 -AGCCCCCATCTCCACCAT
5621631 a
salmon ss irf7 50 -CCACATGTGCTTTGGAGAGAAGT 105%
50 -AGGGAACCACCTTGACAATGAT
ENSGMOT00000014710 cod ac prp2 (or ac dhx16) 50 -GCTTCCAGAGGTTCAGGACTAAAA 108%
50 -CCTCCCGTTTCCTGCTCTTA
ENSGMOT00000001139 cod ac ppara 50 -CGGCCCTTCAGCGACAT 108%
50 -CCAGGGCGTTGAACCTCAT
ENSGMOT00000014095 cod ac cpt1 50 -TCCCAACAAGCGCTCGTT 90.1%
50 CGTCGAGGGTGACGAAGAAG
ENSGMOT00000015968 cod ac pghs1 50 -TGCGTGTGCAACCGTTTAAC 107%
50 -GAAGCGTATGGCTTGAGTTCAA
ENSGMOT00000010511 cod ac irf7 50 -AAGGCCAAATGGAAGACCAA 99.2%
50 -TGGACATCCTGAAGCGTTTGT
a
The accession to the array sequences can be found at the CRISP site under cGRASP 44 K Salmonid Oligo Array; Release Date: June, 2008 (http://web.uvic.ca/grasp/
microarray/array.html).

ACKNOWLEDGMENTS
The authors would like to thank the Norwegian Micro-
array Consortium for providing guidance with analyses and
publication of microarray data. A special thanks to Dr rjan
Karlsen and Dr. Eva Andersson, for participating in sam-
plings and providing scientific advice, and to Stig Mhle for
performing microarray hybridizations. This study was
funded by the EU-project LIFECYCLE (FP7222719,
http://www.lifecycle.gu.se/, December, 19, 2013).
REFERENCES
Aegerter S,Jalabert B, Bobe J. 2005. Large scale real-time PCR
analysis of mRNA abundance in rainbow trout eggs in relation-
ship with egg quality and post-ovulatory ageing. Mol Reprod Dev
72:377385.
Bai J, Solberg C, Fernandes JM, Johnston IA. 2007. Profiling of
maternal and developmental-stage specific mRNA transcripts in
Atlantic halibut Hippoglossus hippoglossus. Gene 386:202210.
Berthelot C, Brunet F, Chalopin D, Juanchich A, Bernard M, Noel
B, Bento P, Da Silva C, Labadie K, Alberti A, Aury JM, Louis A,
Dehais P, Bardou P, Montfort J, Klopp C, Cabau C, Gaspin C,
Thorgaard GH, Boussaha M, Quillet E, Guyomard R, Galiana D,
Bobe J, Volff JN, Genet C, Wincker P, Jaillon O, Roest Crollius, .
2014. The rainbow trout genome provides novel insights into
evolution after whole-genome duplication in vertebrates. Nat
Commun 5:3657.
Bo TH, Dysvik B, Jonassen I. 2004. LSimpute: Accurate estimation
of missing values in microarray data with least squares meth-
ods. Nucleic Acids Res 32:e34.
Bobe J, Labbe C. 2010. Egg and sperm quality in fish. Gen Comp
Endocr 165:535548.
Bolstad BM, Irizarry RA, Astrand M, Speed TP. 2003. A compari-
son of normalization methods for high density oligonucleotide
array data based on variance and bias. Bioinformatics
19:185193.
Bonnet E, Fostier A, Bobe J. 2007. Microarray-based analysis of
fish egg quality after natural or controlled ovulation. BMC Ge-
nomics 8:55.
Charlier C, Montfort J, Chabrol O, Brisard D, Nguyen T, Le Cam A,
Richard-Parpaillon L, Moreews F, Pontarotti P, Uzbekova S,
Chesnel F, Bobe J. 2012. Oocyte-somatic cells interactions,
lessons from evolution. BMC Genomics 13:560.
Davidson WS, Koop BF, Jones SJ, Iturra P, Vidal R, Maass A,
Jonassen I, Lien S, Omholt SW. 2010. Sequencing the ge-
nome of the Atlantic salmon (Salmo salar). Genome Biol
11:403.
Dysvik B, Jonassen I. 2001. J-Express: Exploring gene expression
data using Java. Bioinformatics 17:369370.
Giraldez AJ, Mishima Y, Rihel J, Grocock RJ, Van Dongen S,
Inoue K, Enright AJ, Schier AF. 2006. Zebrafish MiR-430 pro-
motes deadenylation and clearance of maternal mRNAs. Sci-
ence 312:7579.
Goetz FW. 1991. Compartmentalization of prostaglandin synthe-
sis within the fish ovary. Am J Physiol 260:R862R865.
Gorodilov YN. 1996. Description of the early ontogeny of the
Atlantic salmon, Salmo salar, with a novel system of interval
(state) identification. Environ Biol Fish 47:109127.
Jantzen SG, Sanderson DS, von Schalburg KR, Yasuike M,
Marass F, Koop BF. 2011. A 44K microarray dataset of the
changing transcriptome in developing Atlantic salmon (Salmo
salar L.). BMC Res Notes 4:88.

Mol. Reprod. Dev. (2015) 7

Molecular Reproduction & Development
Kanehisa M, Goto S, Sato Y, Furumichi M, Tanabe M. 2012.
KEGG for integration and interpretation of large-scale molecular
data sets. Nucleic Acids Res 40:D109D114.
Kjesbu OS, Kryvi H, Sundby S, Solemdal P. 1992. Buoyancy
variations in eggs of Atlantic cod (Gadus morhua L.) in relation
to chorion thickness and egg size Theory and observations. J
Fish Biol 41:581599.
Kjorsvik E, Mangorjensen A, Holmefjord I. 1990. Egg quality in
fishes. Adv Mar Biol 26:71113.
Kleppe L, Edvardsen RB, Furmanek T, Taranger GL, Wargelius A.
2014. Global transcriptome analysis identifies regulated tran-
scripts and pathways activated during oogenesis and early
embryogenesis in atlantic cod. Mol Reprod Dev 81:619635.
Kleppe L, Edvardsen RB, Kuhl H, Malde K, Furmanek T, Drivenes
O, Reinhardt R, Taranger GL, Wargelius A. 2012. Maternal
30 UTRs: From egg to onset of zygotic transcription in Atlantic
cod. BMC Genomics 13:443.
Kleppe L, Karlsen O, Edvardsen RB, Norberg B, Andersson E,
Taranger GL, Wargelius A. 2013. Cortisol treatment of pre-
spawning female cod affects cytogenesis related factors in
eggs and embryos. Gen Comp Endocr 189:8495.
Lister AL, Van Der Kraak GJ. 2009. Regulation of prostaglandin
synthesis in ovaries of sexually-mature zebrafish (Danio rerio).
Mol Rep Dev 76:10641075.
Marioni JC, Mason CE, Mane SM, Stephens M, Gilad Y. 2008. RNA-
seq: An assessment of technical reproducibility and comparison
with gene expression arrays. Genome Res 18:15091517.
Mommens M, Fernandes JM, Bizuayehu TT, Bolla SL, Johnston
IA, Babiak I. 2010. Maternal gene expression in Atlantic halibut
(Hippoglossus hippoglossus L.) and its relation to egg quality.
BMC Res Notes 3:138.
Qi C, Zhu Y, Reddy JK. 2000. Peroxisome proliferator-activated
receptors, coactivators, and downstream targets. Cell Biochem
Biophys 32:187204.
Schier AF. 2007. The maternal-zygotic transition: Death and birth
of RNAs. Science 316:406407.
Seli E, Yaba A, Guzeloglu-Kayisli O, Lalioti MD. 2008. Alternative
splicing of the mouse embryonic poly(A) binding protein (Epab)
8 Mol. Reprod. Dev. (2015)
WARGELIUS ET AL.
mRNA is regulated by an exonic splicing enhancer: A model for
post-transcriptional control of gene expression in the oocyte.
Mol Hum Reprod 14:393398.
Seppola M, Johnsen H, Mennen S, Myrnes B, Tveiten H. 2009.
Maternal transfer and transcriptional onset of immune genes during
ontogenesis in Atlantic cod. Dev Comp Immunol 33:12051211.
Star B, Nederbragt AJ, Jentoft S, Grimholt U, Malmstrom M,
Gregers TF, Rounge TB, Paulsen J, Solbakken MH, Sharma
A, Wetten OF, Lanzen A, Winer R, Knight J, Vogel JH, Aken B,
Andersen O, Lagesen K, Tooming-Klunderud A, Edvardsen RB,
Tina KG, Espelund M, Nepal C, Previti C, Karlsen BO, Moum T,
Skage M, Berg PR, Gjoen T, Kuhl H, Thorsen J, Malde K,
Reinhardt R, Du L, Johansen SD, Searle S, Lien S, Nilsen F,
Jonassen I, Omholt SW, Stenseth NC, Jakobsen KS. 2011. The
genome sequence of Atlantic cod reveals a unique immune
system. Nature 477:207210.
Sylvestre EL, Robert C, Pennetier S, Labrecque R, Gilbert I, Dufort
I, Leveille MC, Sirard MA. 2013. Evolutionary conservation of
the oocyte transcriptome among vertebrates and its implications
for understanding human reproductive function. Mol Hum Re-
prod 19:369379.
Tarazona S, Garcia-Alcalde F, Dopazo J, Ferrer A, Conesa A.
2011. Differential expression in RNA-seq: A matter of depth.
Genome Res 21:22132223.
Tingaud-Sequeira A, Chauvigne F, Lozano J, Agulleiro MJ,
Asensio E, Cerda J. 2009. New insights into molecular path-
ways associated with flatfish ovarian development and atre-
sia revealed by transcriptional analysis. BMC Genomics
10:434.
Wang S, Furmanek T, Kryvi H, Krossoy C, Totland GK, Grotmol
S, Wargelius A. 2014. Transcriptome sequencing of Atlantic
salmon (Salmo salar L.) notochord prior to development of the
vertebrae provides clues to regulation of positional fate,
chordoblast lineage and mineralisation. BMC Genomics
15:141.
Supporting Information
Additional supporting information may be found in the
online version of this article at the publishers web-site.