Proteomic Profiling of Early Life Stages of European Grayling

(Thymallus thymallus)
Spiros Papakostas,† L. Asbjørn Vøllestad,‡ Craig R. Primmer,† and Erica H. Leder*,†

Division of Genetics and Physiology, Department of Biology, University of Turku, Finland, and Centre for
Ecological and Evolutionary Synthesis, Department of Biology, University of Oslo, Norway

Received May 20, 2010

Salmonids are teleost fish of profound research and economic interest. Embryonic development is a
key aspect of salmonid biology that can be critically affected by environmental parameters. Still, their
proteome during embryogenesis remains largely unexplored. This study investigates the proteome of
the eyed-egg and hatching stages of embryonic development of a salmonid species, European grayling
(Thymallus thymallus), using a shotgun proteomic approach. To deal with limited grayling protein
resources, the generated spectra were compared against an all-salmonid database using search and
multiple protein grouping algorithms to infer identifications. Functional enrichment analysis was carried
out at different levels (gene ontologies, pathways, networks) using zebrafish as a reference genome.
A total of 213 and 239 proteins were confidently detected in eyed and hatching stages, respectively.
Cell cycle, energy, and protein metabolism were the major processes common to both stages. Nuclear
activity and brain and eye development were the predominant functions in the eyed-stage proteome,
while central nervous system, skeletal muscle, and heart development prevailed in the hatching stage.
Overall, this research constitutes the first effort to describe the proteome during embryogenesis in
grayling or any salmonid species. It also presents a systematic approach by which existing resources
can enable proteome research in salmonids.

Keywords: teleosts • salmonids • grayling • embryonic development • mass spectrometry

Introduction Recent advances in mass spectrometry (MS)-based proteom-

Teleost fishes are the largest and most variable vertebrate
taxon.1 With about 27 000 recorded species, they comprise
more than 99% of all known ray-finned fish (actinopterygians),
which in turn include more than 95% of all living fish.2 Their
pronounced diversity is reflected in multiple aspects of their
biology such as morphology, ecology, behavior, genome orga-
nization, and development.1,2 Of particular interest is embry-
onic development in oviparous fishes since it can be profoundly
affected by a number of environmental factors such as water
temperature and dissolved oxygen concentrations.3 Adaptation
to environmental conditions has been suggested to be of great
importance during egg incubation and the critical phase that
follows fry emergence in teleost ectotherms.4,5 At the molecular
level, fast-evolving cis-regulatory and protein coding DNA
sequences are considered to play an important role in teleost
diversity.6 More generally, Carroll7 highlighted the evolutionary
significance of such molecular level phenomena, especially
when they affect proteins important for embryogenesis. In this
framework, studying the embryonic proteome of teleost fish
can be of fundamental and applied interest.
* To whom correspondence should be addressed. Mailing address:
Vesilinnantie 5, 20014 Turku, Finland. Phone: +358 2 333 7086. Fax: +358 2
333 6680. E-mail: erica.leder@utu.fi.
†
University of Turku.
‡
University of Oslo.
ics have simplified large scale identification and quantitative
profiling of organismal proteomes.8 Similarly, the ever-increas-
ing amount of information on protein and protein-protein
interactions allows for genes to be systematically reviewed and
classified according to their function(s), involvement in par-
ticular process(es) or subcellular location(s) into annotated
functional categories, pathways, and networks.9,10 This infor-
mation constitutes a significant resource for functional pro-
teomic studies especially when studying species for which only
limited molecular information is available.11,12 To date, the
proteome of the teleost embryo has been investigated in very
few species.13 In fact, most studies have focused on a single
species, the zebrafish (Danio rerio),13-16 which has been
extensively employed as an experimental model for embryo-
genesis, organogenesis, and general development in verte-
brates.17 Clearly, proteomic studies of a more diverse group of
teleost species are required in order to better understand the
molecular basis of teleost embryonic development.
Salmonid fishes have been the focus of considerable research
due to their economic importance, as well as their ecological
and evolutionary uniqueness.18 Particularly, the evolutionary
importance of early life history traits has been repeatedly
emphasized because they can affect juvenile competitive ability,
dispersal, foraging, and vulnerability to predation and climatic
conditions.4,19-23 However, the molecular basis underlying
ecologically relevant early life history traits has rarely, if ever,

4790 Journal of Proteome Research 2010, 9, 4790–4800 10.1021/pr100507s  2010 American Chemical Society
Published on Web 07/05/2010
Proteomic Profiling of Early Life Stages of European Grayling
been studied in any salmonid fish. In European grayling
(Thymallus thymallus), the evolutionary importance of embry-
onic development rate and hatching time has been demon-
strated in recently introduced populations inhabiting the
Lesjaskogsvatnet lake system in southern Norway.24,25 There-
fore, grayling individuals from this system can be useful
candidates for investigating early development at the protein
expression level in the context of evolutionary and ecological
adaptation. In this study, a MS-based approach was employed
to identify similarities and differences of the proteomes ex-
pressed at the eyed-egg and hatching larvae stages in grayling
from Lake Lesjaskogsvatnet. Overall, this work provides the first
systematic investigation of the early development proteome in
salmonids. Due to the limited number of available protein
sequences in grayling, this study also demonstrates a way by
which existing sequence resources can be used for the MS-
based protein detection in any salmonid species.
Materials and Methods
Sample Collection and Experimental Design. The samples
derived from two grayling spawning sites (Sandbekken and
Valåe) of the Lesjaskogsvatn lake system. They were the product
of the artificial fertilization between the pooled gametes of 30
individuals (15 sires, 15 dams) per spawning population. Eggs
from each population were reared in common garden condi-
tions at 6 °C. Embryos were sampled daily from fertilization to
when the yolk sac was absorbed. Individuals were selected from
a time point corresponding to the appearance of eye pigmenta-
tion (98-100 degree days, eyed-egg stage) and from when more
than 50% of the individuals are hatched (188 degree days,
hatching stage). Four individuals from each population for each
time point were used for protein analysis. All fish embryos used
in this study were collected according to animal experimenta-
tion guidelines at the sampling location in Norway.
Sample Preparation. Proteins were isolated from whole
embryos using Tri Reagent (Sigma), following the manufacturer
instructions. Protein pellets resulting from the extraction
protocol were dissolved in 1% SDS and quantified using a DC
protein assay (Bio-Rad). Because within-stage quantification
will be subsequently investigated, samples that were used in
this study for the MS analysis were labeled with iTRAQ reagents
following the manufacturer’s protocol (Applied Biosystems).
Briefly, 100 µg of each sample was acetone precipitated and
resuspended in 20 µL of 0.5 M triethyl ammonium bicarbonate
(TEAB) and 1 µL of 2% SDS (supplied in the iTRAQ kit). Samples
were reduced by the addition of 2 µL of 50 mM tris-carboxy-
ethyl phosphine hydrochloride (TCEP) and incubation for 1 h
at 60 °C, and then alkylated by the addition of 1 µL of 200 mM
methyl methane thiosulfonate (MMTS) and incubation for 10
min at room temperature (RT). Following this step, 10 µL of
trypsin (1 µg/µL) (Promega, Madison, WI) was added, and
samples were digested overnight at 37 °C. Ethanol was added
to the iTRAQ labels, and the reconstituted labels were added
to their respective samples. Samples were incubated for 1 h,
combined into their respective multiplexes, and then evapo-
rated. Samples were dissolved in 50 µL of 0.1% trifluoroacetic
acid (TFA) and desalted using MacroSpin C18 columns (The
Nest Group, Southborough, MA) according to the manufac-
turer’s directions. Samples were dissolved in isoelectric focusing
buffer consisting of 2 M urea and 2% IPG buffer, pH 3-10 (GE
Healthcare), and fractionated before mass spectrometry analy-
sis using isoelectric focusing on a 13 cm, pH 3-10 linear,
Immobiline DryStrip (GE Health Care). The strips were rehy-
research articles
drated for 12 h and focused on an IPGphor (Amersham
Pharmacia Biotech, Piscataway, NJ) using the following pro-
gram: hold at 500 V for 1 h, linear gradient from 500 to 1000 V
for 15 min, hold at 1000 V for 1 h, linear gradient from 1000 to
8000 V for 30 min, hold at 8000 V for 3.5 h. The gel strip was
cut into six fractions and peptides were extracted from the gel
strip by washing three times: first in 0.1% TFA, second in 0.1%
TFA, 50% acetonitrile (ACN), and third in 0.1% TFA, 100% ACN,
all for 15 min at 37 °C. The solution from the three washes
was combined and evaporated. The peptides were dissolved
in 0.1% TFA, 5% ACN and desalted using UltraMicroSpin C18
columns (The Nest Group) according to the manufacturer’s
directions.
Mass Spectrometry Analysis. LC-MS/MS analyses were
made using a system consisting of a Waters Cap-LC (Waters,
Milford, MA) coupled to a QSTAR Pulsar ESI-hybrid quadrupole
time-of-flight instrument (Applied Biosystems, Foster City, CA).
A 0.3 mm ×5 mm PepMap C18 µ-precolumn (LC Packings,
Dionex, Sunnyvale, CA) was used for sample loading and was
coupled to a 15 cm × 75 µm i.d. fused silica capillary column
packed with 5 µm Magic C18 (Michrom BioResources, Inc.,
Auburn, CA) for separation. A 90 min binary gradient from 2%
to 30% phase A was used at a flow rate of 200 µL/min (phase
A ) 5% ACN, 0.1% HCOOH; phase B ) 95% ACN, 0.1%
HCOOH). Data acquisition and instrument control was per-
formed using Analyst QS 1.1 software (Applied Biosystems,
Foster City, CA), with the mass spectrometer set to perform
1 s survey scans followed by two 2 s MS/MS scans of the two
most intense peaks (10 cps), with dynamic exclusion for 5 min.
Database Search and Protein Detection Parameters. Protein-
Pilot software (v.3, Applied Biosystems, ABI) was employed to
infer protein identifications. The detected protein threshold in
the Paragon algorithm was set to 1.3 (95% confidence level).
The search database was comprised of all the salmonid proteins
(14 885 in total) submitted in UniProt database (15.11 release).
The common contaminants file provided by ABI (ABSciex-
_ContaminantDB file) was also appended to these protein
sequences. The search effort was set to thorough ID, and both
biological modifications and amino acid substitutions were
included. The quality of the Paragon algorithm detection results
was assessed by performing a false discovery rate (FDR) analysis
at 1% global error rate using a decoy database as implemented
within the proteomics system performance evaluation pipeline
(PSEP) of the ProteinPilot software. The protein groups were
created by the Pro Group algorithm of the ProteinPilot software.
This algorithm helped reduce the protein redundancy of the
search database since it groups together proteins with the same
spectral evidence. Yet, these groups were further inspected on
the basis of the multidetected protein groups suggested by the
same algorithm. Multidetected groups that involved proteins
annotated under the same gene name or name entries in
UniProt were manually grouped together. For each protein, the
number of credible spectra (g95% confidence) uniquely as-
signed to them was counted, and for proteins detected only in
one of the stages, a one-tailed χ-square test with the Yates’
continuity correction was performed to examine whether the
spectra present were significantly greater than zero (p e 0.05).
Gene Ontology (GO) Enrichment Analysis. Hierarchical GO
overrepresentation tests were performed in Cytoscape 2.6.326
using the BiNGO 2.3 plugin.27 Four gene lists were investigated.
These involved all proteins that were detected in the eyed stage
(gene set a), all those detected in the hatching stage (set b),
those detected only in the eyed stage (set c), and those detected
Journal of Proteome Research • Vol. 9, No. 9, 2010 4791
research articles
only in the hatching stage (set d). The first two lists were used
to investigate the general functional properties of each stage,
whereas the two latter ones highlighted the functional differ-
ences between them. Since zebrafish is the most widely used
teleost species in developmental studies and has a well-
annotated genome, the D. rerio genome (Ensembl Zv8) was
employed as a gene reference set for the overrepresentation
tests. The D. rerio gene orthologs for the grayling proteins were
found using the BLAST tool in the UniProt database, whereas
the ontology files required for the interpretation and the
structured visualization of the overrepresented GO terms were
retrieved from the GO database (as of date 07-JAN-2010). The
hypergeometric overrepresentation tests were performed at
0.05 level of significance with the Benjamini-Hochberg FDR
multiple-testing correction.28
Functional relatedness of the overrepresented GO terms was
inferred using the ClueGO Cytoscape plugin.29 While BiNGO
reconstructs the hierarchical ontology trees, ClueGO uses kappa
statistics to create networks of functional GO clusters.29 As
above, the D. rerio genome was employed as a reference set,
and hypergeometric enrichment tests with Benjamini-Hochberg
correction were performed. The networks were generated by
setting the specificity to medium and the connectivity to equal
or greater than 0.1.29 The networks were visualized with the
GOlorize tool.30 Due to term interdependency in the GO
hierarchy, numerous GO terms of the same branch can be
significantly overrepresented.27 Therefore, both high-level (more
general) and low-level/terminal (more specific) GO terms were
used to describe each group.
Pathway and Network Analysis. Pathway and network
analysis were performed with the Ingenuity Pathway Analysis
(IPA) software, v.8.0. The Agilent zebrafish V2 microarray, which
fairly represents the whole zebrafish genome, was selected as
a reference set. Fisher exact tests were used to examine the
significance (p e 0.05) of the different pathways and networks
in the investigated gene lists. Since the Ingenuity Knowledge
Base is built on human (Homo sapiens), mouse (Mus musculus),
and rat (Rattus norvegicus) resources, pathways and network
functions that had to do with cancer, diseases, toxicity,
pathogen-influenced signals, and xenobiotic signals were not
considered in the analyses.
Results and Discussion
Proteins Detected. A total of 313 proteins were detected, 213
and 239 in the eyed and hatching stages, respectively, of which
139 were common to both stages (Supplementary Tables 1 and
2, Supporting Information). In addition to the minimum 95%
confidence threshold and the maximum 1% FDR, the confi-
dence of the reported detections can be assessed by the fact
that on average, at least three credible peptides (g95% confi-
dence) were assigned to each protein, namely, 3.1 and 4.3 for
the eyed and the hatching stage, respectively. These numbers
are without considering vitellogenins, the major components
of the egg yolk, which were identified by an average of more
than 100 credible peptides in each stage. The inclusion of
vitellogenins to these calculations would have caused an
upward bias of the average peptide numbers. Although a
number of proteins (65 eyed stage and 87 hatching stage) were
detected by a single credible peptide, these detections are still
highly confident, since almost all these proteins had an unused
score of greater than or equal to two, which corresponds to
99% detection confidence (Supplementary Table 1, Supporting
Information).
4792 Journal of Proteome Research • Vol. 9, No. 9, 2010
Papakostas et al.
Interestingly, the reported detections for the grayling spectra
were derived from a salmonid search database that included
less than 80 grayling proteins in total (data not shown). On
top of this, only three grayling entries were among the 313
reported proteins (Supplementary Table 1, Supporting Infor-
mation). Thus, almost all of the detections in grayling were
based on non-grayling protein sequence information. Given
that Thymallinae is one of the evolutionary most distant
subfamilies within salmonid phylogeny,31,32 it is likely that the
salmonid protein sequences can effectively be used together
as a search database for MS-based protein detections in any
salmonid species, following the procedure presented here. The
ProGroup algorithm can efficiently group together most of the
redundant protein entries. However, in a few cases, manual
classification can be necessary, based on the reported mul-
tidetected protein groups (Supplementary Table 1, Supporting
Information).
Functions in Common. Twenty functionally related groups
of GO terms were identified in both stages (Figure 1). Significant
terms, at p e 0.05, were detected in 17 of these groups, and 12
were practically the same in the two stages (Table 1, Supple-
mentary Tables 3 and 4, Supporting Information). These
common groups were largely involved in functions concerning
(a) protein synthesis and metabolism (translational elongation,
translation elongation factor activity, ribosomal subunits,
protein folding), (b) carbohydrate metabolism and energy
production (glycolysis, L-malate dehydrogenase activity, proton-
transporting ATP synthase complex), and (c) DNA packaging
and cell cycle regulation (nucleosome assembly, nucleosome,
regulation of cell cycle) (Table 1). Protein complexes concerning
cell cycle, protein synthesis, and energy metabolism have been
found to occupy central (core) positions in the yeast proteome
network,33 with the complexes themselves consisting of core,
module, and attachment proteins.34,35 In contrast to module
and attachment proteins, core ones were almost always present
in the complex.34 In zebrafish, using two-dimensional gel
electrophoresis to profile protein expression during embryonic
development, a number of constant gel spots were noticed
throughout developmental stages.14 An interesting hypothesis
follows from these findings, namely, that the GO-based func-
tional groups and proteins in common between the stages are
part of the predominant grayling core interactome during
embryonic development. However, more research is needed
before any clear conclusions can be drawn.
Evidence of pronounced activity involved in carbohydrate
metabolism in both the eyed and hatching stages was found
in the pathway analysis level as well. Nine out of 19 significantly
detected pathways were in common between the two stages,
out of which most of the metabolic pathways concerned
carbohydrate metabolism (Table 2, Supplementary Tables 5 and
6, Supporting Information). Carbohydrate metabolism has been
suggested to play a critical role in fish embryonic development,
with glucose serving primarily as an energy compound but also
as a substrate for the synthesis of both nucleic acids and
polysaccharides.36 Other pathways detected in both stages
involved the ILK, granzyme A, protein ubiquitination, and
regulation of actin-based motility by Rho pathways (Table 2).
The integrin-linked kinase (ILK) pathway is a highly conserved
functional complex from invertebrates to mammals with a
crucial role in embryonic development.37 It has been impli-
cated in cell proliferation, adhesion, migration, differentia-
tion, and brain and muscle development.37-39 There is also
evidence that integrins regulate cell spreading and migration
Proteomic Profiling of Early Life Stages of European Grayling research articles

Figure 1. Combined network view of the GO-based functional groups in the grayling eyed and hatching stages. A total of 20 groups
were detected. Dashed lines distinguish the generated clusters to the biological process, cellular component, or molecular functions
GO categories. For each category, every group has its own distinctive color. Multicolored nodes represent GO terms that belong to
more than one group. A description of each group on the basis of the high- (more general) and low-level significant GO terms is given
in Table 1. Since this is a combined view of the eyed and hatching stage networks, nodes were set to the same size, while node
abundances and connectivity are only indicative of the general network organization. More details can be found in Supplementary
Tables 3 and 4, Supporting Information.

Table 1. Description of the GO-Based Functional Groups Identified by ClueGO in the Eyed (eye) and Hatching (hatch) Stagesa
functional stage
group high-level GO detected terminal GO
Biological Process
G1 cellular homeostasis eye cellular iron-ion homeostasis
G2 cellular macromolecule metabolic process both regulation of cell cycle
G3 macromolecule biosynthetic process both translational elongation
G4 DNA metabolic process eye DNA replication initiation
G5 cellular macromolecular complex subunit organization both nucleosome assembly
G6 protein metabolic process both protein folding
G7 carbohydrate catabolic process both glycolysis
G8 lipid localization both lipid transport
G9 ns ns
G10 nucleobase, nucleoside, and nucleotide metabolic process eye nucleotide metabolic process
Cellular Component
G11 intracellular non-membrane-bound organelle both keratin filament
G12 mitochondrial membrane part both proton-transporting ATP synthase complex
G13 chromosomal part both nucleosome
G14 ribonucleoprotein complex both large/small ribosomal subunit
G15 proteasome complex hatch proteasome core complex
Molecular Function
G16 ns ns
G17 hydrolase activity, acting on acid anhydrites eye nucleoside triphosphatase activity
G18 ns ns
G19 oxidoreductase activity, acting on CH-OH group of donors both L-malate dehydrogenase activity
G20 translation factor activity, nucleic acid binding both translation elongation factor activity
a
The most significant high-level terms derived from ClueGO, whereas the low-level, terminal ones derived from BiNGO. Three of the groups (G9, G16,
and G18) had no significant terms (ns) at p e 0.05. All the significant GO terms for the BiNGO and ClueGO analyses are given in Supplementary Tables 3
and 4, Supporting Information, respectively.

through activation of Rho pathways.39 Lastly, the granzyme
A pathway is involved in apoptosis,40 while the protein
ubiquitination pathway is part of the protein catabolic
mechanism. The presence of these pathways in both pro-
teomes underlines the organogenesis-related processes of
cell proliferation, migration, and death among their common
functions. Previous studies have reported a number of
proteins associated with integrin and apoptosis signaling in
hatching stage zebrafish embryos.16 According to the results
of this work, the functioning of these pathways is likely to
be present earlier in development and presumably through-
out teleost embryogenesis.

Journal of Proteome Research • Vol. 9, No. 9, 2010 4793

research articles Papakostas et al.
Table 2. The Pathways That Were Found To Be Significantly Table 3. The Top Five Networks Identified by the IPA
Present (p e 0.05; in Bold When p e 0.01) in the Eyed (eye) Software and Their Five Most Significant Functions in the
and Hatching (hatch) Stagesa Eyed (NE1-NE5) and Hatching (NH1-NH5) Stagesa
p-value name network functions p value

pathway name eye hatch NE1 protein synthesis 2.00 × 10-8

molecular transport 6.91 × 10-4
Metabolic Pathways
nucleic acid metabolism 6.91 × 10-4
Amino Acid Metabolism small molecule biochemistry 6.91 × 10-4
arginine and proline metabolism b 0.014 cellular assembly and organization 3.36 × 10-3
glutathione metabolism 0.033 b NE2 cellular assembly and organization 3.44 × 10-5
phenylalanine, tyrosine, tryptophan biosynthesis 0.033 0.035 hepatic system development and 4.01 × 10-5
Carbohydrate Metabolism function
citrate cycle 0.012 0.013 cell death 2.63 × 10-4
fructose and mannose metabolism 0.033 b RNA post-transcriptional modification 3.05 × 10-4
glycolysis/gluconeogenesis 0.000 0.001 cell-to-cell signaling and interaction 5.11 × 10-4
glyoxylate and dicarboxylate metabolism 0.037 0.040 NE3 molecular transport 2.31 × 10-4
pentose phosphate pathway 0.012 b cell-to-cell signaling and interaction 2.91 × 10-4
pyruvate metabolism 0.002 0.014 lipid metabolism 5.97 × 10-4
cardiovascular system development 6.91 × 10-4
Energy Metabolism and function
methane metabolism b 0.016 visual system development and 6.91 × 10-4
Nucleotide Metabolism function
purine metabolism 0.000 b NE4 lipid metabolism 5.53 × 10-4
small molecule biochemistry 5.53 × 10-4
Signaling Pathways carbohydrate metabolism 1.07 × 10-3
Cellular Growth, Proliferation and Development molecular transport 1.07 × 10-3
ILK signaling 0.011 0.014 cell morphology 1.35 × 10-3
RAN signaling 0.002 b NE5 protein synthesis 3.43 × 10-6
molecular transport 2.02 × 10-3
Cellular Immune Response
nucleic acid metabolism 2.02 × 10-3
granzyme A signaling 0.000 0.001
small molecule biochemistry 2.02 × 10-3
Intracellular and Second Messenger Signaling cell cycle 2.70 × 10-3
Calcium Signaling b 0.010 NH1 protein synthesis 5.46 × 10-15
Protein Ubiquitination Pathway 0.004 0.018 cellular assembly and organization 3.44 × 10-5
RhoA Signaling b 0.004 hepatic system development and 4.01 × 10-5
function
Neurotransmitters and Other Nervous Systems Signaling
cell-to-cell signaling and interaction 5.11 × 10-4
regulation of actin-based motility by Rho 0.036 0.002
cellular function and maintenance 5.79 × 10-4
Organismal Growth and Development NH2 protein synthesis 3.66 × 10-5
actin cytoskeleton signaling b 0.009 post-translational modification 4.90 × 10-4
a
protein folding 4.90 × 10-4
The genes mapped onto the Agilent zebrafish V2 array platform are molecular transport 6.91 × 10-4
given in Supplementary Table 5. The genes involved in each of the
significant pathways can be found in Supplementary Table 6. b Pathways nucleic acid metabolism 6.91 × 10-4
not significantly detected. NH3 skeletal and muscular system 1.77 × 10-4
development and function
tissue morphology 1.77 × 10-4
Last, many network functions were the same between the cell death 2.63 × 10-4
two stages as well (Table 3, Supplementary Table 7, Supporting cell morphology 1.35 × 10-3
Information). As with GO functional categories and pathways, gene expression 2.91 × 10-3
the identified proteins grouped into networks that were in- NH4 lipid metabolism 4.80 × 10-6
volved with macromolecule/energy metabolism (carbohydrate, small molecule biochemistry 4.80 × 10-6
nucleic acid and lipid metabolism, protein synthesis), in vitamin and mineral metabolism 4.80 × 10-6
molecular transport 9.28 × 10-6
apoptosis (cell death), and in cell proliferation mechanisms
carbohydrate metabolism 2.78 × 10-5
(cellular assembly and organization), but also an additional
NH5 protein synthesis 1.81 × 10-5
category in the network analysis included developmental cell death 1.60 × 10-4
processes (hepatic system development and function) (Table cellular assembly and organization 1.90 × 10-4
3). These functions provide a general overview of the underlying cellular function and maintenance 1.90 × 10-4
commonalities between the two proteomes. Future studies, DNA replication, recombination and 4.78 × 10-4
investigating more development stages, will likely reveal whether repair
these common functions exist throughout embryonic development. a
Details concerning the molecules that participate in each of these
Eyed-Stage Specific Functions. Brain and eye development networks can be found in Supplementary Table 7, Supporting Informa-
were the predominant functions within the eyed stage. Based tion.
on spectra counts, 13 proteins were found to be significantly
present only in the eyed stage (Supplementary Table 8, Sup- and eye development. In neonatal mouse brains, MARCKS and
porting Information). Among them, the myristoylated alanine- MARCKS-like proteins were highly expressed in the retina of
rich C-kinase substrate (MARCKS) and the epsilon isoform of the developing eye,41 regulating retinal cell proliferation.42
14-3-3 proteins have been known to play a critical role in brain MARCKS-deficient mice displayed lethal brain deficiencies

4794 Journal of Proteome Research • Vol. 9, No. 9, 2010

Proteomic Profiling of Early Life Stages of European Grayling research articles

Figure 2. (A) Detailed view of the molecule interactions for the network NE3 identified in the eyed stage. Its functions include
“cardiovascular and visual system development”, which to some extent characterizes the specific functioning of the eyed-stage proteome.
(B) Detailed view of the hatching-stage network NH3, which is involved in “skeletal and muscular system development and function”.
More information about the interacting molecules is given in Supplementary Table 7, Supporting Information. Both networks were
drawn in Cytoscape 2.6.3 according to the results generated by the IPA software.

including lamination abnormalities of the cortex and the eye
retina.43 Similarly, the 14-3-3 protein has been implicated in
the eye and the early neural development in Drosophila and
mice.44,45 On a different functional level, five of the detected
pathways were also significant only in the eyed stage (Table
2). Again, the RAN pathway is involved in neural, brain and
eye development. During mouse embryogenesis Ran protein
expression was detected initially in neuroectoderm, neural
groove, neural folds, neural tube, and later within the encepha-
lic vesicle wall and spinal cord, especially during eye develop-

Journal of Proteome Research • Vol. 9, No. 9, 2010 4795

research articles Papakostas et al.
46
ment, including optic vesicle. In Drosophila embryos, Ran Table 4. All the Terminal Overrepresented GO Categories
expression was restricted to central nervous system neuroblasts Found in Gene Set c (Genes Derived from Proteins That Were
and was implicated in the eye/antenna disk development,47 Detected Only in the Eyed Stage) and in Set d (Proteins
whereas during Xenopus embryogenesis, Ran expression was Detected Only in Hatching Stage)a
again localized in the central nervous system as well as in corrected p-value
neural crest, mesenchyme, eyes, and otic vesicles.48 Lastly, at GO term eye (set c) hatch (set d)
the network functional level, one of the significant functions
Biological Process
of the eyed-stage detected network, NE3, is “visual system
DNA replication initiation 7.60 × 10-4
development and function” (Table 3). A number of genes heterocycle metabolic process 4.79 × 10-2
involved in the NE3 network concerned proteins detected only negative regulation of 2.12 × 10-2
in the eyed stage (Figure 2A). coagulation
Complement-controlled organogenesis processes were also nucleoside metabolic process 1.85 × 10-2
among the predominant properties of the eyed-stage proteome. response to hypoxia 4.57 × 10-2
Complement component C3 was significantly detected only in ribonucleoside monophosphate 3.47 × 10-2
the eyed stage (Supplementary Table 8, Supporting Informa- biosynthetic process
RNA interference 3.88 × 10-2
tion). Complement component 1Q and a serine proteinase
translational elongation 3.67 × 10-2
inhibitor (serpin), the alpha-1 antiprotease-like protein, were
ubiquitin-dependent protein 4.94 × 10-2
also detected only in the eyed stage (Supplementary Table 8, catabolic process
Supporting Information). Serpins are proteins that, among
Cellular Component
other functions, control complement activation, programmed
cytosol 1.45 × 10-2
cell death, and development.49 Complement-regulated apop-
integral to mitochondrial inner 3.88 × 10-2
totic pathways can have an important role in a number of early membrane
development and cell differentiation processes.50 In fact, PCNA complex 4.57 × 10-2
neurons of the inner nuclear layer of the vertebrate eye retina proteasome core complex 3.27 × 10-3
undergo considerable cell death during early embryonic ribonucleoside-diphosphate 3.88 × 10-2
stages.51 Lange et al.52 detected C3 transcripts throughout the reductase complex
embryonic development of Atlantic cod in a number of tissues, ribosome 4.38 × 10-21
including the eye. In general, apoptosis and phagocytosis can RNA-induced silencing complex 3.88 × 10-2
be pivotal to organogenesis as well as to the establishment of Molecular Function
the nervous and immune systems during the early stages of betaine-homocysteine 4.57 × 10-2
development.52,53 S-methyltransferase activity
Last, the comparison of the two stages at the GO level calcium-dependent 4.00 × 10-2
revealed, among other things, that terms related to nuclear phospholipid binding
calcium ion binding 2.17 × 10-2
activity (DNA replication, nuclear transport) were particularly
creatine kinase activity 4.57 × 10-2
overrepresented in the proteins detected at the eyed stage cytochrome c oxidase activity 4.57 × 10-2
(Table 4, Figure 3). In agreement with this, at the pathway level, DNA polymerase processivity 4.57 × 10-2
the purine metabolism pathway was significant only in the eyed factor activity
stage (Table 2). Furthermore, and in contrast to the apparent IMP cyclohydrolase activity 3.88 × 10-2
similarity of the most significant network functions (Table 3), nucleoside triphosphatase 5.89 × 10-3
in many cases different underlying processes were predominant activity
for each stage, and under the “cell morphology” network orotate 3.88 × 10-2
function, the most significant mechanism in the eyed stage was phosphoribosyltransferase
“morphology of nuclear matrix”, while in the hatching stage it activity
orotidine-5′-phosphate 3.88 × 10-2
was “morphology of microfilaments” (results not shown). It is
decarboxylase activity
likely that the increased nuclear activity at the eyed stage is phospholipase inhibitor activity 1.89 × 10-2
directly connected to the increased cell division required for phosphoribosylaminoinidazole- 3.88 × 10-2
embryonic growth and organogenesis during early embryo- carboxamide formyltransferase
genesis.54 activity
In teleost species, little is known about the proteome ribosome binding 2.25 × 10-3
functioning during the eye pigmentation of the embryo, yet structural constituent of 3.61 × 10-22
such early development stages can critically affect the whole ribosome
threonine-type endopeptidase 3.27 × 10-3
embryogenesis and adult growth processes.5 In zebrafish, eye
activity
pigmentation becomes apparent around 24 h postfertilization
transaldolase activity 4.57 × 10-2
(hpf) at the beginning of the pharyngula stage.55 By two- transketolase activity 3.88 × 10-2
dimensional gel electrophoresis, significant changes in the
a
protein profile were found to accompany the zebrafish embryo More details are given in Supplementary Table 3, Supporting
Information.
entering the pharyngula development period.14 From use of
MS on gel spots corresponding to abundant or stage-specific hpf zebrafish embryo.14 By investigating more detected proteins
proteins, a number of cytosolic, cytoskeletal, and nuclear and carrying out the functional analysis at different levels, the
protein identifications, many of which were also detected in present work has additionally revealed that cell proliferation and
grayling, were reported.14 Based on these identifications, pro- cell death, as well as early neural, brain, and eye development
cesses involved in metabolism, cytoskeleton, translation, and functions are among the major functions of the eyed-stage
protein degradation have been suggested to be present in the 24 proteome in grayling and likely in other teleosts as well.

4796 Journal of Proteome Research • Vol. 9, No. 9, 2010

Proteomic Profiling of Early Life Stages of European Grayling research articles

Figure 3. The result from the direct comparison of the eyed and hatching stages with ClueGO. Color gradient shows the proportion of
genes in the GO term associated with each of the two stages. Red coloring characterizes eyed-stage related GOs, while green the
hatching-stage ones. Equal proportions are shown in white. Node size is related to the statistical significance of the GO. Only terms
with a corrected p-value g 0.05 are effectively visible. Edge line width is proportional to the kappa scores, while node connectivity was
inferred using kappa statistics.22 Seven overrepresented GO terms and term clusters were related to the eyed stage (E1-E7) and eight
to the hatching stage (H1-H8). The most significant GOs for the clusters or terms are nucleotide biosynthetic process (E1), iron-ion
transport (E2), nuclear transport (E3), DNA replication initiation (E4), mitochondrial membrane part (E5), mitochondrial matrix (E6),
ribose phosphate diphosphokinase activity (E7), NADPH regeneration (H1), regulation of actin filament polymerization (H2), proton-
transporting ATP synthase complex (H3), proteasome core complex, alpha subunit complex (H4), ribosome (H5), inorganic cation
transmembrane transporter activity (H6), translation factor activity, nucleic acid binding (H7), threonine-peptidase activity (H8). More
details can be found in Supplementary Table 4, Supporting Information.

Figure 4. Scatter plot of the average number of unique credible (g95 confidence) spectra per technical replicate assigned to each of
the 313 detected proteins in the eyed and hatching stages. The names of those proteins with the most assigned spectra are noted on
the graph. A detailed view of these data can be found in Supplementary Table 8, Supporting Information.

Hatching-Specific Functions. A total of 19 proteins and five
pathways were significantly detected only in the hatching stage
(Table 2, Supplementary Table 8, Supporting Information).
Skeletal muscle development was found to be a major hatch-
ing-specific function. Creatine kinases 1 and 2, both among
the significant proteins, are involved in muscle-specific me-
tabolism.56 Troponin T isoform 1, another significant protein,
is involved in muscle maturation.57 In addition to the specific
presence of particular proteins, their relative abundance, as can
be inferred by the spectral evidence,58,59 also highlights the
muscle functioning of the hatching-stage detected proteins. A
number of skeletal muscle proteins, namely the myosin heavy
chain, actin alpha, and actin beta, had considerably higher
spectral counts in the hatching than in the eyed stage (101,
80, and 103 versus 2.6, 23, and 8, respectively; Figure 4,
Supplementary Table 8, Supporting Information). Evidence for
hatching-specific myoskeletal development functioning can
also be found in the GO, pathway, and network levels of
analysis. The GO terms of actin filament polymerization,
creatine kinase activity, and calcium ion binding were signifi-
cant only in the hatching stage (Figure 3, Table 4). The RhoA
and actin cytoskeleton signaling pathways, both involved in
muscle tissue development,60 were detected only in the hatch-
ing stage (Table 2). Lastly, one of the most significant functions
of the network NH3 was “skeletal and muscular system
development and function” (Table 3). A number of genes,
whose proteins were detected only in the hatching stage,
participate in this network (Figure 2B).
Other hatching-specific functions involve central nervous
system and heart development. In particular, the fatty acid
binding protein (FABP) brain type and heart type are two
proteins detected only in the hatching stage (Supplementary
Table 8, Supporting Information). The FABP brain type is an
intracellular protein expressed particularly in the central
nervous system with an important role during neurogenesis
in vertebrates.61,62 During zebrafish late embryogenesis, FABP

Journal of Proteome Research • Vol. 9, No. 9, 2010 4797

research articles
brain type mRNA transcripts were detected in the lateral floor
plate of the spinal cord, the lateral hindbrain, the ventricular
brain zone, the forebrain, and the retina.63 On the other hand,
the FABP heart type has been involved in regulating cardi-
omyocyte growth and differentiation of neonatal hearts.64
Furthermore, in addition to taking part in the developmental
processes of muscle system development, the Rho signaling
pathways have been found to play an important role in neural
system and heart development in zebrafish.65,66 RhoA has also
been implicated in the heart development of chick embryos.67
Last, the comparison of the two stages at the GO level
revealed terms related to (a) protein catabolism (ubiquitin-
dependent catabolic process, proteasome core complex, threo-
nine-peptidase activity), (b) hydrogen ion transmembrane
transporter activity (cytochrome c oxidase activity), and (c)
ribosome binding to be particularly overrepresented in the
hatching-detected proteins (Table 4, Figure 3). Ubiquitin-
mediated proteolysis can have a critical role in both somite
segmentation and differentiation processes by which the
growing embryo will eventually develop its skeletal muscle,
axial skeleton, and dermis.68 In agreement with these results,
proteasome and ubiquitin up-regulation has been detected at
both the transcriptome and the proteome level at later stages
of zebrafish development.16,69
Previous efforts investigating the zebrafish hatching pro-
teome (72 hpf) by employing two-dimensional gel electro-
phoresis, MS, and two-dimensional PAGE methodologies iden-
tified numbers of proteins largely associated with organ systems
such as central nervous system, heart, and skeletal muscle.16
At the cellular level, these proteins were related to structure,
transcription/translation, cell cycle, ion transport, and nucle-
otide, carbohydrate, lipid and energy metabolism.16 Pathways
with the most contributing proteins were related to calcium,
integrin, extracellular signal-regulated kinase/mitogen-activated
protein kinase, and vascular endothelial growth factor signal-
ing.16 These results are in good agreement with the findings of
the present work, suggesting similar functioning of the hatch-
ing-stage proteome between zebrafish and grayling. This agree-
ment additionally strengthens the validity of the methodology
presented here, highlighting a way by which the salmonid
embryonic proteome can be studied, even in species with
limited genomic/proteomic information.
Conclusions
This work is the first systematic investigation of the embry-
onic proteome functioning in European grayling and in salmo-
nids in general. Based on the proteins detected, the analyses
that were carried out at multiple levels (spectra counts, GOs,
pathways, and networks) largely agreed but also complemented
each other with respect to both the similarities and the
differences of the two proteomes. The common functions had
to do primarily with cell cycle, energy, and protein metabolism.
In both stages, glycolytic and cell proliferation pathways/
networks highlighted the energy and growth requirements of
the developing embryo. Cell migration, cell differentiation, and
apoptotic processes were related to organogenesis within the
two proteomes. Significant nuclear activity together with early
neural, brain, and eye development-related pathways and
network functions were predominant in the eyed-stage pro-
teome. On the other hand, central nervous system, heart, and
myoskeletal system developmental processes were among the
functions that characterized the hatching-stage proteome.
Given that MS-based proteomics deals with the detection of
4798 Journal of Proteome Research • Vol. 9, No. 9, 2010
Papakostas et al.
8
the most abundant proteins, this study likely describes the
major functions of the two proteomes. Yet, functional clas-
sification of the detected proteins could serve as a valuable
resource for future efforts targeting the grayling or salmonid
proteome.
Acknowledgment. This work has been financially
supported by the Research Council of Norway and the
Finnish Centre of Excellence in Genetics and Physiology. We
would like to thank Thrond Haugen, Gaute Thomassen, and
Nicola Barson for fieldwork assistance and sampling the
grayling embryos, the Turku Centre for Biotechnology
Proteomics Facility for technical assistance, particularly
Anne Rokka and Arttu Heinonen, and Robert Molder for
useful discussions concerning proteomic analyses.
Supporting Information Available: Tables containing
the ProteinPilot unused and total scores, percent coverage,
number of assigned peptides and UniProt identifiers for the
salmonid detected proteins; the names of the proteins detected
in each of the eyed and hatching stages; the results of the
BiNGO analysis for proteins detected in the eyed stage (set a),
the hatching stage (set b), the unique to eyed stage (set c) and
the unique to hatching stage (set d); the significant GO terms
(p e 0.05) for the combined and the comparison GO-based
functional grouping analysis performed with ClueGO; the
zebrafish ortholog genes mapped onto the Agilent zebrafish
V2 microarray by the IPA software for the eyed (set a) and the
hatching (set b) stage; the zebrafish V2 microarray mapped
genes associated with each of the significantly detected path-
ways by the IPA software in the eyed and the hatching stage;
the zebrafish V2 microarray genes associated with the IPA top
five detected networks in the eyed (networks NE1 to NE5) and
hatching (NH1 to NH5) stages; and the unique (at g 95
confidence) spectra assigned to each of the reported proteins
per pooled sample (iTRAQ run). This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Nelson, J. S. Fishes of the world, 4th ed.; John Wiley and Sons:
New York, 2006.
(2) Volff, J. N. Genome evolution and biodiversity in teleost fish.
Heredity 2005, 94, 280–294.
(3) Johnston, I. A. Environment and plasticity of myogenesis in teleost
fish. J. Exp. Biol. 2006, 209, 2249–2264.
(4) Jensen, L. F.; Hansen, M. M.; Pertoldi, C.; Holdensgaard, G.;
Mensberg, K. L. D.; Loeschcke, V. Local adaptation in brown trout
early life-history traits: implications for climate change adapt-
ability. Proc. R. Soc. B: Biol. Sci. 2008, 275, 2859–2868.
(5) Macqueen, D. J.; Robb, D. H. F.; Olsen, T.; Melstveit, L.; Paxton,
C. G. M.; Johnston, I. A. Temperature until the ‘eyed stage’ of
embryogenesis programmes the growth trajectory and muscle
phenotype of adult Atlantic salmon. Biol. Lett. 2008, 4, 294–298.
(6) Ravi, V.; Venkatesh, B. Rapidly evolving fish genomes and teleost
diversity. Curr. Opin. Genet. Dev. 2008, 18, 544–550.
(7) Carroll, S. B. Evo-devo and an expanding evolutionary synthesis:
A genetic theory of morphological evolution. Cell 2008, 134, 25–
36.
(8) Gstaiger, M.; Aebersold, R. Applying mass spectrometry-based
proteomics to genetics, genomics and network biology. Nat. Rev.
Genet. 2009, 10, 617–627.
(9) Okuda, S.; Yamada, T.; Hamajima, M.; Itoh, M.; Katayama, T.; Bork,
P.; Goto, S.; Kanehisa, M. KEGG Atlas mapping for global analysis
of metabolic pathways. Nucleic Acids Res. 2008, 36, W423–W426.
(10) Rhee, S. Y.; Wood, V.; Dolinski, K.; Draghici, S. Use and misuse of
the gene ontology annotations. Nat. Rev. Genet. 2008, 9, 509–515.
(11) Carpentier, S. C.; Panis, B.; Vertommen, A.; Swennen, R.; Sergeant,
K.; Renaut, J.; Laukens, K.; Witters, E.; Samyn, B.; Devreese, B.
Proteome analysis of non-model plants: A challenging but power-
ful approach. Mass Spectrom. Rev. 2008, 27, 354–377.
Proteomic Profiling of Early Life Stages of European Grayling
(12) Buggiotti, L.; Primmer, C. R.; Kouvonen, P.; Bures, S.; Corthals,
G. L.; Leder, E. H. Identification of differentially expressed proteins
in Ficedula flycatchers. Proteomics 2008, 8, 2150–2153.
(13) Forne, I.; Abian, J.; Cerda, J. Fish proteome analysis: Model
organisms and non-sequenced species. Proteomics 2010, 10, 858–
872.
(14) Tay, T. L.; Lin, Q. S.; Seow, T. K.; Tan, K. H.; Hew, C. L.; Gong,
Z. Y. Proteomic analysis of protein profiles during early develop-
ment of the zebrafish Danio rerio. Proteomics 2006, 6, 3176–3188.
(15) Link, V.; Carvalho, L.; Castanon, I.; Stockinger, P.; Shevchenko, A.;
Heisenberg, C. P. Identification of regulators of germ layer
morphogenesis using proteomics in zebrafish. J. Cell Sci. 2006,
119, 2073–2083.
(16) Lucitt, M. B.; Price, T. S.; Pizarro, A.; Wu, W.; Yocum, A. K.; Seiler,
C.; Pack, M. A.; Blair, I. A.; FitzGerald, G. A.; Grosser, T. Analysis
of the zebrafish proteome during embryonic development. Mol.
Cell. Proteomics 2008, 7, 981–994.
(17) Love, D. R.; Pichler, F. B.; Dodd, A.; Copp, B. R.; Greenwood, D. R.
Technology for high-throughput screens: the present and future
using zebrafish. Curr. Opin. Biotechnol. 2004, 15, 564–571.
(18) Schaffer, W. M. Life histories, evolution and salmonids. In Evolu-
tion Illuminated: Salmon and their relatives; Hendry, A. P., Stearns,
S. C., Eds.; Oxford University Press: New York, 2004; pp 20-51.
(19) Sundstrom, L. F.; Lohmus, M.; Devlin, R. H. Selection on increased
intrinsic growth rates in coho salmon Oncorhynchus kisutch.
Evolution 2005, 59, 1560–1569.
(20) Haugen, T. O.; Vollestad, L. A. Population differences in early life-
history traits in grayling. J. Evol. Biol. 2000, 13, 897–905.
(21) Einum, S.; Fleming, I. A. Selection against late emergence and small
offspring in Atlantic salmon (Salmo salar). Evolution 2000, 54, 628–
639.
(22) Hendry, A. P. Adaptive divergence and the evolution of reproduc-
tive isolation in the wild: an empirical demonstration using
introduced sockeye salmon. Genetica 2001, 112-113, 515–534.
(23) Vasemagi, A.; Gross, R.; Palm, D.; Paaver, T.; Primmer, C. R.
Discovery and application of insertion-deletion (INDEL) polymor-
phisms for QTL mapping of early life-history traits in Atlantic
salmon. BMC Genomics 2010, 11, 156.
(24) Koskinen, M. T.; Haugen, T. O.; Primmer, C. R. Contemporary
fisherian life-history evolution in small salmonid populations.
Nature 2002, 419, 826–830.
(25) Barson, N. J.; Haugen, T. O.; Vollestad, L. A.; Primmer, C. R.
Contemporary isolation-by-distance, but not isolation-by-time,
among demes of European grayling (Thymallus thymallus, Lin-
naeus) with recent common ancestors. Evolution 2009, 63, 549–
556.
(26) Shannon, P.; Markiel, A.; Ozier, O.; Baliga, N. S.; Wang, J. T.;
Ramage, D.; Amin, N.; Schwikowski, B.; Ideker, T. Cytoscape: A
software environment for integrated models of biomolecular
interaction networks. Genome Res. 2003, 13, 2498–2504.
(27) Maere, S.; Heymans, K.; Kuiper, M. BiNGO: A Cytoscape plugin to
assess overrepresentation of gene ontology categories in biological
networks. Bioinformatics 2005, 21, 3448–3449.
(28) Benjamini, Y.; Hochberg, Y. Controlling the false discovery rate -
A practical and powerful approach to multiple testing. J. R. Stat.
Soc., Ser. B: Methodol. 1995, 1, 289–300.
(29) Bindea, G.; Mlecnik, B.; Hackl, H.; Charoentong, P.; Tosolini, M.;
Kirilovsky, A.; Fridman, W. H.; Pages, F.; Trajanoski, Z.; Galon, J.
ClueGO: A Cytoscape plug-in to decipher functionally grouped
gene ontology and pathway annotation networks. Bioinformatics
2009, 25, 1091–1093.
(30) Garcia, O.; Saveanu, C.; Cline, M.; Fromont-Racine, M.; Jacquier,
A.; Schwikowski, B.; Aittokallio, T. GOlorize: A Cytoscape plug-in
for network visualization with gene ontology-based layout and
coloring. Bioinformatics 2007, 23, 394–396.
(31) Koop, B. F.; von Schalburg, K. R.; Leong, J.; Walker, N.; Lieph, R.;
Cooper, G. A.; Robb, A.; Beetz-Sargent, M.; Holt, R. A.; Moore, R.;
Brahmbhatt, S.; Rosner, J.; Rexroad, C. E.; McGowan, C. R.;
Davidson, W. S. A salmonid EST genomic study: Genes, duplica-
tions, phylogeny and microarrays. BMC Genomics 2008, 9, 545.
(32) Yasuike, M.; Jantzen, S.; Cooper, G. A.; Leder, E.; Davidson, W. S.;
Koop, B. F. Grayling (Thymallinae) phylogeny within salmonids:
Complete mitochondrial DNA sequences of Thymallus arcticus and
Thymallus thymallus. J. Fish Biol. 2010, 76, 395–400.
(33) Gavin, A. C.; Bosche, M.; Krause, R.; Grandi, P.; Marzioch, M.;
Bauer, A.; Schultz, J.; Rick, J. M.; Michon, A. M.; Cruciat, C. M.;
Remor, M.; Hofert, C.; Schelder, M.; Brajenovic, M.; Ruffner, H.;
Merino, A.; Klein, K.; Hudak, M.; Dickson, D.; Rudi, T.; Gnau, V.;
Bauch, A.; Bastuck, S.; Huhse, B.; Leutwein, C.; Heurtier, M. A.;
Copley, R. R.; Edelmann, A.; Querfurth, E.; Rybin, V.; Drewes, G.;
Raida, M.; Bouwmeester, T.; Bork, P.; Seraphin, B.; Kuster, B.;
research articles
Neubauer, G.; Superti-Furga, G. Functional organization of the
yeast proteome by systematic analysis of protein complexes.
Nature 2002, 415, 141–147.
(34) Gavin, A. C.; Aloy, P.; Grandi, P.; Krause, R.; Boesche, M.; Marzioch,
M.; Rau, C.; Jensen, L. J.; Bastuck, S.; Dumpelfeld, B.; Edelmann,
A.; Heurtier, M. A.; Hoffman, V.; Hoefert, C.; Klein, K.; Hudak, M.;
Michon, A. M.; Schelder, M.; Schirle, M.; Remor, M.; Rudi, T.;
Hooper, S.; Bauer, A.; Bouwmeester, T.; Casari, G.; Drewes, G.;
Neubauer, G.; Rick, J. M.; Kuster, B.; Bork, P.; Russell, R. B.; Superti-
Furga, G. Proteome survey reveals modularity of the yeast cell
machinery. Nature 2006, 440, 631–636.
(35) Pang, C. N. I.; Krycer, J. R.; Lek, A.; Wilkins, M. R. Are protein
complexes made of cores, modules and attachments? Proteomics
2008, 8, 425–434.
(36) Lahnsteiner, F. Carbohydrate metabolism of vitellogenic follicles
and eggs of Serranus cabrilla (Serranidae) and Mullus barbatus
(Mullidae) and of embryos of Sparus aurata (Sparidae). Fish
Physiol. Biochem. 2006, 32, 131–139.
(37) Liang, X. Q.; Zhou, Q.; Li, X. D.; Sun, Y. F.; Lu, M.; Dalton, N.; Ross,
J.; Chen, J. PINCH1 plays an essential role in early murine
embryonic development but is dispensable in ventricular cardi-
omyocytes. Mol. Cell. Biol. 2005, 25, 3056–3062.
(38) Niewmierzycka, A.; Mills, J.; St-Arnaud, R.; Dedhar, S.; Reichardt,
L. F. Integrin-linked kinase deletion from mouse cortex results in
cortical lamination defects resembling cobblestone lissencephaly.
J. Neurosci. 2005, 25, 7022–7031.
(39) Giancotti, F. G.; Ruoslahti, E. Transduction - Integrin signaling.
Science 1999, 285, 1028–1032.
(40) Elmore, S. Apoptosis: A review of programmed cell death. Toxicol.
Pathol. 2007, 35, 495–516.
(41) Stumpo, D. J.; Eddy, R. L.; Haley, L. L.; Sait, S.; Shows, T. B.; Lai,
W. S.; Young, W. S.; Speer, M. C.; Dehejia, A.; Polymeropoulos,
M.; Blackshear, P. J. Promoter sequence, expression, and fine
chromosomal mapping of the human gene (MLP) encoding the
MARCKS-like protein: Identification of neighboring and linked
polymorphic loci for MLP and MACS and use in the evaluation of
human neural tube defects. Genomics 1998, 49, 253–264.
(42) Zhao, J.; Izumi, T.; Nunomura, K.; Satoh, S.; Watanabe, S. MARCKS-
like protein, a membrane protein identified for its expression in
developing neural retina, plays a role in regulating retinal cell
proliferation. Biochem. J. 2007, 408, 51–59.
(43) Stumpo, D. J.; Bock, C. B.; Tuttle, J. S.; Blackshear, P. J. MARCKS
deficiency in mice leads to abnormal brain-development and
perinatal death. P. Natl. Acad. Sci. U.S.A. 1995, 92, 944–948.
(44) McConnell, J. E.; Armstrong, J. F.; Hodges, P. E.; Bard, J. B. L. The
mouse 14-3-3 epsilon isoform, a kinase regulator whose expression
pattern is modulated in mesenchyme and neuronal differentiation.
Dev. Biol. 1995, 169, 218–228.
(45) Su, T. T.; Parry, D. H.; Donahoe, B.; Chien, C. T.; O’Farrell, P. H.;
Purdy, A. Cell cycle roles for two 14-3-3 proteins during Drosophila
development. J. Cell Sci. 2001, 114, 3445–3454.
(46) Lopez-Casas, P. P.; Lopez-Fernandez, L. A.; Krimer, D. B.; del Mazo,
J. Ran GTPase expression during early development of the mouse
embryo. Mech. Dev. 2002, 113, 103–106.
(47) Koizumi, K.; Stivers, C.; Brody, T.; Zangeneh, S.; Mozer, B.;
Odenwald, W. F. A search for Drosophila neural precursor genes
identifies ran. Dev. Genes Evol. 2001, 211, 67–75.
(48) Onuma, Y.; Nishihara, R.; Takahashi, S.; Tanegashima, K.; Fukui,
A.; Asashima, M. Expression of the Xenopus GTP-binding protein
gene Ran during embryogenesis. Dev. Genes Evol. 2000, 210, 325–
327.
(49) Kalsheker, N.; Morley, S.; Morgan, K. Gene regulation of the serine
proteinase inhibitors alpha(1)-antitrypsin and alpha(1)-antichy-
motrypsin. Biochem. Soc. Trans. 2002, 30, 93–98.
(50) Mastellos, D.; Lambris, J. D. Complement: More than a ‘guard’
against invading pathogens. Trends Immunol. 2002, 23, 485–491.
(51) Cusato, K.; Stagg, S. B.; Reese, B. E. Two phases of increased cell
death in the inner retina following early elimination of the ganglion
cell population. J. Comp. Neurol. 2001, 439, 440–449.
(52) Lange, S.; Dodds, A. W.; Gudmundsdottir, S.; Bambir, S. H.;
Magnadottir, B. The ontogenic transcription of complement
component C3 and apolipoprotein A-I tRNA in Atlantic cod (Gadus
morhua L.) - a role in development and homeostasis. Dev. Comp.
Immunol. 2005, 29, 1065–1077.
(53) Zakeri, Z.; Lockshin, R. A. Cell death during development. J. Im-
munol. Methods 2002, 265, 3–20.
(54) Rodenas, E.; Klerkx, E. P. F.; Ayuso, C.; Audhya, A.; Askjaer, P. Early
embryonic requirement for nucleoporin Nup35/NPP-19 in nuclear
assembly. Dev. Biol. 2009, 327, 399–409.
Journal of Proteome Research • Vol. 9, No. 9, 2010 4799
research articles
(55) Kimmel, C. B.; Ballard, W. W.; Kimmel, S. R.; Ullmann, B.; Schilling,
T. F. Stages of embryonic development of the zebrafish. Dev. Dyn.
1995, 203, 253–310.
(56) Xu, Y. F.; He, J. Y.; Wang, X. K.; Lim, T. M.; Gong, Z. Y.
Asynchronous activation of 10 muscle-specific protein (MSP) genes
during zebrafish somitogenesis. Dev. Dyn. 2000, 219, 201–215.
(57) Lin, Y.; Chen, Y.; Yang, X. X.; Xu, D.; Liang, S. P. Proteome analysis
of a single zebrafish embryo using three different digestion
strategies coupled with liquid chromatography-tandem mass
spectrometry. Anal. Biochem. 2009, 394, 177–185.
(58) Liu, H. B.; Sadygov, R. G.; Yates, J. R. A model for random sampling
and estimation of relative protein abundance in shotgun pro-
teomics. Anal. Chem. 2004, 76, 4193–4201.
(59) Elliott, M. H.; Smith, D. S.; Parker, C. E.; Borchers, C. Current trends
in quantitative proteomics. J. Mass Sprectrom. 2009, 44, 1637–1660.
(60) Tsapara, A.; Luthert, P.; Greenwood, J.; Hill, C. S.; Matter, K.; Balda,
M. S. The RhoA activator GEF-H1/Lfc is a transforming growth
factor-beta target gene and effector that regulates alpha-smooth
muscle actin expression and cell migration. Mol. Biol. Cell 2010,
21, 860–870.
(61) Feng, L.; Hatten, M. E.; Heintz, N. Brain lipid-binding protein
(BLBP) - a novel signaling system in the developing mammalian
CNS. Neuron 1994, 12, 895–908.
(62) Kurtz, A.; Zimmer, A.; Schnutgen, F.; Bruning, G.; Spener, F.;
Muller, T. The expression pattern of a novel gene encoding brain
fatty-acid-binding protein correlates with neuronal and glial-cell
development. Development 1994, 120, 2637–2649.
(63) Liu, R. Z.; Denovan-Wright, E. M.; Degrave, A.; Thisse, C.; Thisse,
B.; Wright, J. M. Differential expression of duplicated genes for
4800 Journal of Proteome Research • Vol. 9, No. 9, 2010
Papakostas et al.
brain-type fatty acid-binding proteins (fabp7a and fabp7b) during
early development of the CNS in zebrafish (Danio rerio). Gene
Expression Patterns 2004, 4, 379–387.
(64) Tang, M. K.; Kindler, P. M.; Cai, D. Q.; Chow, P. H.; Li, M.; Lee,
K. K. H. Heart-type fatty acid binding proteins are upregulated
during terminal differentiation of mouse cardiomyocytes, as
revealed by proteomic analysis. Cell Tissue Res. 2004, 316, 339–
347.
(65) Weiser, D. C.; Row, R. H.; Kimelman, D. Rho-regulated myosin
phosphatase establishes the level of protrusive activity required
for cell movements during zebrafish gastrulation. Development
2009, 136, 2375–2384.
(66) Berndt, J. D.; Clay, M. R.; Langenberg, T.; Halloran, M. C. Rho-
kinase and myosin II affect dynamic neural crest cell behaviors
during epithelial to mesenchymal transition in vivo. Dev. Biol.
2008, 324, 236–244.
(67) Kaarbo, M.; Crane, D. I.; Murrell, W. G. RhoA is highly up-regulated
in the process of early heart development of the chick and
important for normal embryogenesis. Dev. Dyn. 2003, 227, 35–47.
(68) Rida, P. C. G.; Le Minh, N.; Jiang, Y. J. A Notch feeling of somite
segmentation and beyond. Dev. Biol. 2004, 265, 2–22.
(69) Mathavan, S.; Lee, S. G. P.; Mak, A.; Miller, L. D.; Murthy, K. R. K.;
Govindarajan, K. R.; Tong, Y.; Wu, Y. L.; Lam, S. H.; Yang, H.; Ruan,
Y. J.; Korzh, V.; Gong, Z. Y.; Liu, E. T.; Lufkin, T. Transcriptome
analysis of zebrafish embryogenesis using microarrays. PLOS
Genet. 2005, 1, 260–276.
PR100507S