A neurogenomics approach to gene expression analysis in the developing brain

Department of Developmental Neurobiology, St. Jude Children's Research Hospital, Memphis, TN 38105, United States.
Molecular Brain Research (Impact Factor: 2). 01/2005; 132(2):116-27. DOI: 10.1016/j.molbrainres.2004.10.002
Source: PubMed
Secreted and transmembrane proteins provide critical functions in the signaling networks essential for neurogenesis. We used a genetic signal sequence gene trap approach to isolate 189 genes expressed during development in e16.5 whole head, e16.5 hippocampus and e14.5 cerebellum. Gene ontology programs were used to classify the genes into respective biological processes. Four major classes of biological processes known to be important during development were identified: cell communication, cell physiology processes, metabolism and morphogenesis. We used in situ hybridization to determine the temporal and spatial patterns of gene expression in the developing brain using this set of probes. The results demonstrate that gene expression patterns can highlight potential gene functions in specific brain regions. We propose that combining bioinformatics with the gene expression pattern is an effective strategy to identify genes that may play critical roles during brain development.


Available from: John Mccoy
Research report
A neurogenomics approach to gene expression analysis
in the developing brain
Patricia Jensen
, Susan Magdaleno
, Karen M. Lehman
, Dennis S. Rice
Edward R. LaVallie
, Lisa Collins-Racie
, J.M. McCoy
, Tom Curran
Department of Developmental Neurobiology, St. Jude Children’s Research Hospital, Memphis, TN 38105, United States
Lexicon Genetics Inc., The Woodlands, TX 77381, United States
Department of Discovery Medicine, Wyeth Research, Cambridge, MA 02140, United States
Biogen-IDEC, Cambridge, MA 02142, United States
Accepted 5 October 2004
Available online 11 November 2004
Secreted and transmembrane proteins provide critical functions in the signaling networks essential for neurogenesis. We used a genetic signal
sequence gene trap approach to isolate 189 genes expressed during development in e16.5 whole head, e16.5 hippocampus and e14.5 cerebellum.
Gene ontology programs were used to classify the genes into respective biological processes. Four major classes of biological processes known
to be important during development were identified: cell communication, cell physiology processes, metabolism and morphogenesis. We used
in situ hybridization to determine the temporal and spatial patterns of gene expression in the developing brain using this set of probes. The results
demonstrate that gene expression patterns can highlight potential gene functions in specific brain regions. We propose that combining
bioinformatics with the gene expression pattern is an effective strategy to identify genes that may play critical roles during brain development.
D 2004 Elsevier B.V. All rights reserved.
Theme: Development and regeneration
Topic: Developmental genetics
Keywords: Signal sequence gene trap; In situ hybridization
1. Introduction
During neurogenesis, multiple biological proces ses func-
tion in concert to ensure that the diverse neuronal and glial
cell types that comprise the brain proliferate, differentiate,
migrate and form synapses at the appropriate time and place.
These processes rely on the precise control of temporal and
spatial expression of genes encoding secreted and membrane
associated molecules. Proteins destined for secretion or for
transport to locations within the membrane (e.g. neuro-
transmitters, growth factors, guidance cues, ion channels,
etc.) convey fundamental information necessary for the cell to
respond to the evolving intra- and extracellular environments
that occur during development. These molecules usually
contain a signal sequence that targets nacent proteins to the
secretory pathway via the endoplasmic reticulum and the
trans-Golgi network [10]. This feature permits selection in a
yeast genetic screen known as the signal sequence trap [11,5].
In this study, we used the signal sequence trap to identify
189 genes encoding signal peptide-containing proteins
expressed during development of the cerebral cortex, the
hippocampus and the cerebellum. We utilized bioinformatics
tools to characterize these genes based on gene ontology and
found that the majority are known to function in cell
communication, cell physiology processes, metabolism and
morphogenesis. Next, in situ hybridization analysis was used
to determine the temporal and spatial expres sion patterns of
this interesting subset of genes during brain development.
0169-328X/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
* Corresponding author. Tel.: +1 901 495 2253; fax: +1 901 495 2270.
E-mail address: (T. Curran).
Both authors contributed equally to this manuscript.
Molecular Brain Research 132 (2004) 116 127
Page 1
Combining gene expression pattern information with
bioinformatics establishes a framework on which scientific
hypothesis can be established to investigate gene function.
Information such as the biological function of a protein
product, the presence of conserved protein domains and
the location of the protein product in cellular compart-
ments are valuable resources when coupled with the gene
expression pattern. While the informatics revolution has
made a great impact on our ability to integrate gene
sequence, gene ontology and protein structure information,
it has proven much more challenging to combine this with
information on the temporal and spatial patterns of gene
2. Materials a nd methods
2.1. Library construction and yeast selection
Poly A+ RNA was isolated from e16.5 C57BL/6 mouse
whole brain, h ippocampus and e14.5 cerebellum and
reverse-transcribed to cDNA using SuperScript II reverse
transcriptase (Gibco/Life Technologies). The three different
random-primed cDNA libraries were cloned into the signal
sequence trap vector pSUC2T7M13ORI carrying a trun-
cated invertase gene (SUC2) for yeast selection as pre-
viously described [5,11]. The signal sequence trap selects
cDNAs that provide a functional signal peptide to the
truncated invertase gene. Plasmids were isolated from
colonies surviving the invertase selection. Subsequently,
surviving colonies were picked and grown indi vidually for
identification analysis. The cDNA inserts were amplified by
PCR using primers flanking the insert, sequenced and
analyzed by BLAST to determine the closest nucleotide
match to the clone. A second round of sequencing analysis
was performed on clones selected for further character-
ization based on the BLAST results.
2.2. Tissue collection
C57BL/6 mice (Harlan) were maintained at St. Jude
Children’s Research Hospital animal care facility in
accordance with the NIH Guidelines for the care and use
of laboratory animals. Mice were kept on a 12-h light cycle
with food and water ad libitum.
Timed matings were set up between adult C57BL/6 mice.
Females were inspected the following morning for presence
of vaginal plug and removed from the males. Noon of the
day plug was detected was designated as embryonic day 0.5
(e0.5). On e14.5 or 16.5, pregnant dams were deeply
anesthetized with isofluorane and killed by cervical dis-
location. Embryos were removed by laparotomy in ice-cold
phosphate buffered saline (PBS). Embryo heads were
immersion fixed overnight in 4% paraformaldehyde at 4
8C. The following day tissue was rinsed in PBS and brains
were dissected free of heads and cryo-p rotected in 30%
sucrose in 0.1 M PBS for 24 h. Brains were bisected in the
sagittal plane, embedded in a CryomoldR (Sakura Finetek
USA, Torrance, CA) containing TBS Tissue Freezing Media
(Triangle Biome dical Sciences, Durham, NC) and frozen on
dry ice. Cryo-sections, 16 Am, were collected on SuperFrost
Plus slides (Fisher Scientific).
2.3. In situ hybridization
Clones were amplified by PCR using primers flanking
the insert sequence and the vector T7 RNA polymerase
binding site. Following PCR purification and sequence
verification, P33-UTP riboprobes were generated by in vitro
transcription using T7 polymerase. In situ hybridization was
performed on E14.5 and 16.5 mouse brains as previously
described [22]. Images were collected using a Zeiss stereo-
scope and a Nikon Coolsnap ES Digital camera. Adobe
Photoshop version 7.0 was used to a djust brightness,
contrast and gamma levels of all images.
3. Results
3.1. Signal sequence trap selection of cDNA
To identify genes expressed during embryonic mouse
brain development, we used a signal sequence trap selection
in yeast. We generated three cDNA libraries from e16.5
whole brain, e16.5 hippoc ampus and e14.5 cerebellum.
Partial cDNAs from each library were subcloned into a
signal sequence trap vector pSUC2T7M13ORI carrying a
modified invertase gene that lacks its leader sequence
[5,11]. Invertase is a secreted enzyme that is required for
yeast growth on media containing sucrose or raffinose. Only
yeast containing cDNAs providing a functional signal
peptide to the truncated invertase gene survive under
specific growth conditions. Clones that survived the
selection screen were isolated and the cDNA insert was
amplified by PCR and nucleotide sequence was determined.
Nucleotide sequence was determined for 719 clones
representing the whole brain (428), cerebellum (221) and
hippocampus (70) libr aries (Table 1). BLAST analysis
revealed redundant clones, appearing multiple times within
and among the three libraries. In total, we isolated a
collection of 189 unique cDNA clones. Among the 189
cDNAs, 166 were unique to a single library and 23 were
common to two or more of the libraries. Only five of the
clones were common to all three libraries (cDNA sequence
BC004056, complement component factor h, latrophilin 1,
leucine rich repeat protein 3, neuronal and secretogranin II).
The majority of clones isolated, 163, had significant
sequence homology (N80%), to genes with an assigned
GeneID, i.e. sequences curated by the NCBI to represent a
gene locus. Three clones had significant homology to Riken
cDNAs with no assigned GeneID, and 20 clones were most
similar to genomic sequences. In addition, we identified
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127 117
Page 2
Table 1
List of cDNAs isolated by signal sequence trap
Description GeneID Accession Library
e16.5 whole
Procollagen, type IV, alpha 5 12,830 NM
007736 1
Protocadherin beta 4 93,875 NM
053129 1
Chondroitin sulfate proteoglycan 2 13,003 NM
019389 2
Integrin binding sialoprotein 15,891 NM
008318 1
Interleukin 2 receptor, gamma chain 16,186 NM
013563 6
Rho guanine nucleotide exchange factor (GEF) 1 16,801 NM
008488 7
Neural cell adhesion molecule 1 17,967 NM
010875 2 2
Neuropilin 2 18,187 NM
010939 4
Phosphodiesterase 3B, cGMP-inhibited 18,576 NM
011055 1
Regenerating islet-derived 3 gamma 19,695 NM
011260 1
Vascular cell adhesion molecule 1 22,329 NM
011693 1
Neurotrimin 235,106 NM
172290 1
Latrophilin 1 330,814 XP
134383 8 12 2
Metallothionein 2 17,750 NM
008630 1
Gamma-aminobutyric acid (GABA-A)
receptor, subunit gamma 2
14,406 NM
177408 2 1
Stathmin 1 16,765 NM
019641 1
Neuroblastoma ras oncogene 18,176 NM
010937 1
Neurexin II 18,190 XM
111780 1
Polycystic kidney disease 1 homolog 18,763 NM
013630 1
Transmembrane 4 superfamily member 9 56,224 NM
019571 1
Sorting nexin 5 69,178 NM
024225 4
Ras homolog gene family, member E 74,194 NM
028810 1
Insulin-like growth factor 1 16,000 NM
010512 6
Calcium/calmodulin-dependent protein kinase II alpha 12,322 NM
177407 1
Procollagen, type II, alpha 1 12,824 NM
031163 1
RAB18, member RAS oncogene family 19,330 NM
011225 1
Adenylate cyclase 6 11,512 NM
007405 1 2
Bone morphogenetic protein receptor, type 1A 12,166 NM
009758 1
N-acetylgalactosaminyltransferase 1
14,423 NM
013814 1
Leptin receptor 16,847 NM
010704 4
Mannosidase 1, beta 17,156 NM
010763 1
Protein tyrosine phosphatase, non-receptor type 13 19,249 NM
011204 1
Intestinal cell kinase 56,542 NM
019987 2
DNA segment, Chr 3, ERATO Doi 330, expressed 51,886 NM
057172 1
Microtubule-actin cross-linking factor 1 11,426 XM
110503 2
Apolipoprotein D 11,815 NM
007470 2
Calcium channel, voltage-dependent, L type,
alpha 1S subunit
12,292 XM
358335 1
Glycoprotein 38 14,726 NM
010329 1
Fatty acid binding protein 5, epidermal 16,592 NM
010634 7
ATP-binding cassette, subfamily B (MDR/TAP),
member 1A
18,671 NM
011076 4
18S RNA 19,791 X00686 2
Coronin, actin binding protein 1C 23,790 NM
011779 5
Roundabout homolog 2 (Drosophila) 268,902 XM
196160 4
Solute carrier organic anion transporter family,
member 1c1
58,807 NM
021471 2
Midkine 17,242 NM
010784 2
Prion protein 19,122 NM
011170 1
Apolipoprotein E 11,816 NM
009696 6 2
Heterogeneous nuclear ribonucleoprotein D 11,991 NM
007516 1
Heterogeneous nuclear ribonucleoprotein A1 15,382 U65316 1
Transthyretin 22,139 NM
013697 5 1
Bromodomain adjacent to zinc finger domain, 1B 22,385 NM
011714 1
Ribosomal protein L12 269,261 NM
009076 1 2
ATP synthase, H+ transporting, mitochondrial F1
complex, O subunit
28,080 NM
138597 1
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127118
Page 3
Description GeneID Accession Library
e16.5 whole
Proteasome (prosome, macropain) 26S subunit,
non-ATPase, 8
57,296 NM
026545 11
Glucuronyl C5-epimerase 93,683 NM
033320 1
MAP-kinase activating death domain 228,355 NM
145527 1
Leucine zipper domain protein 66,049 NM
133185 1
Transforming growth factor, beta 2 21,808 NM
009367 3
Insulin-like growth factor binding protein 5 16,011 NM
010518 2
Nephroblastoma overexpressed gene 18,133 NM
010930 1
Heterogeneous nuclear ribonucleoprotein
methyltransferase-like 1 (S. cerevisiae)
15,468 NM
133182 1
Zinc finger protein 451 98,403 NM
133817 3
5-Aminoimidazole-4-carboxamide ribonucleotide
formyltransferase/IMP cyclohydrolase
108,147 NM
026195 1
AT motif binding factor 1 11,906 AK083885 1
ATP synthase, H+ transporting mitochondrial
F1 complex, beta subunit
11,947 NM
016774 2
ATPase inhibitor 11,983 NM
007512 1
Cathepsin D 13,033 NM
009983 1
Cathepsin K 13,038 NM
007802 2
Ectodermal-neural cortex 1 13,803 NM
007930 8
FK506 binding protein 2 14,227 NM
008020 1
FK506 binding protein 7 14,231 NM
010222 5
GA repeat binding protein, beta 1 14,391 NM
010249 1
Heterogeneous nuclear ribonucleoprotein A/B 15,384 NM
010448 1
RNA-binding region (RNP1, RRM) containing 2 170,791 NM
133242 1
Pyruvate kinase, muscle 18,746 NM
011099 1
Proliferation-associated 2G4 18,813 XM
192681 1
SNF related kinase 20,623 NM
133741 1
Zinc finger homeobox 1a 21,417 NM
011546 1
Jumonji, AT rich interactive domain 1A (Rbp2 like) 214,899 XM
289906 1
Natriuretic peptide receptor 2 230,103 NM
173788 1
Nardilysin, N-arginine dibasic convertase,
NRD convertase 1
230,598 NM
146150 1
Thyroid hormone receptor associated protein 3 230,753 NM
146153 1
Protein kinase, lysine deficient 1 232,341 NM
198703 6 1
Fusion, derived from t(12;16) malignant
liposarcoma (human)
233,908 NM
139149 2
SREBP cleavage activating protein 235,623 XM
135190 2
Heterogeneous nuclear ribonucleoprotein D-like 50,926 NM
016690 3
Eukaryotic translation initiation factor 3, subunit 8 56,347 NM
146200 1
RNA (guanine-7-) methyltransferase 67,897 NM
026440 1
Serine caroboxypeptidase 1 74,617 NM
029023 3
Ubiquitin-conjugating enzyme E2N 93,765 NM
080560 1
CD68 antigen 12,514 NM
009853 6
RIKEN cDNA 1300002C08 gene 67,472 NM
026182 1
DNA segment, Chr 12, Wayne State
University 95, expressed
217,864 NM
198023 1
SLIT and NTRK-like family, member 2 245,450 XM
205324 2
Expressed sequence AI851790 268,354 NM
182807 1
TD and POZ domain containing 2 399,673 AF545858 1
Reticulon 1 104,001 NM
153457 3
Solute carrier family 33
(acetyl-CoA transporter), member 1
11,416 NM
015728 1
Actin, gamma, cytoplasmic 11,465 NM
009609 1
Angiopoietin 2 11,601 NM
007426 5 2
Calumenin 12,321 NM
007594 18
Serine (or cysteine) proteinase inhibitor,
clade H, member 1
12,406 NM
009825 1
CD86 antigen 12,524 NM
019388 1
Complement component factor h 12,628 NM
009888 101 53 23
Table 1 (continued)
(continued on next page)
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127 119
Page 4
Description GeneID Accession Library
e16.5 whole
Collagen, type III, alpha 1 (Ehlers–Danlos
syndrome type IV, autosomal dominant)
12,825 NM
009930 3
Cystatin C 13,010 NM
009976 1
Cathepsin L 13,039 NM
009984 1
Decorin 13,179 NM
007833 1
Dermatan sulphate proteoglycan 3 13,516 NM
007884 5
SPARC-like 1 (mast9, hevin) 13,602 NM
010097 1
Gene trap locus 6 14,897 NM
133975 1
Keratocan 16,545 NM
008438 7
Leukocyte cell derived chemotaxin 1 16,840 NM
010701 2
Leucine rich repeat protein 3, neuronal 16,981 XM
126894 14 6 2
Microfibrillar-associated protein 2 17,150 NM
008546 1
NADH dehydrogenase 6, mitochondrial mRNA 177,22 15 6
Neuron specific gene family member 1 18,196 NM
010942 1
Prothymosin alpha 19,231 NM
008972 1
Secretogranin II 20,254 NM
009129 2 30 1
Sema domain, immunoglobulin domain (Ig),
short basic domain, secreted, (semaphorin) 3C
20,348 NM
013657 1
Protocadherin 9 211,712 XM
139187 1
Ollistatin-like 5 213,262 NM
178673 1
RIKEN cDNA A630007B06 gene 213,993 NM
170757 1 1
Tissue factor pathway inhibitor 21,788 NM
011576 14 4
DNA segment, Chr 14, Abbott 1 expressed 218,850 XM
127656 2
Transmembrane 4 superfamily member 2 21,912 NM
019634 6
RIKEN cDNA 1810043M20 gene 226,562 XM
148990 4
Zinc finger protein, subfamily 1A, 4 (Eos) 22,781 NM
011772 1 1
Leukocyte receptor cluster (LRC) member 8 232,798 BC041775 2
cDNA sequence BC058674 233,210 XM
149933 1
cDNA sequence BC004056 234,463 NM
145599 3 12 3
cDNA sequence BC005561 243,171 XM
144450 1
Zinc finger protein 291 244,891 NM
175536 5
cDNA sequence BC060615 268,515 NM
198423 2
RIKEN cDNA A930034L06 gene 319,317 NM
175692 1
RIKEN cDNA A330076H08 gene 320,026 AK048029 1
RIKEN cDNA C730049O14 gene 320,117 AK039899 2
RIKEN cDNA D130059P03 gene 320,538 NM
177185 1
RIKEN cDNA D630045J12 gene 330,286 BC054075 3
Similar to KIAA1602 protein 380,969 AK048721 1
Hypothetical LOC414280 414,280 AL954858 1
Hypothetical LOC414282 414,282 AC132357 1
Reticulocalbin 3, EF-hand calcium
binding domain
52,377 NM
026555 1
DNA segment, Chr 5, Brigham and
Women’s Genetics 0860 expressed
52,822 BC049127 1
Axotrophin 57,438 NM
020575 2
Heterogeneous nuclear ribonucleoprotein H1 59,013 NM
021510 5
Endomucin 59,308 NM
016885 2 2
Tenomodulin 64,103 NM
022322 1
Neurotensin 67,405 AF304160 2
Niemann Pick type C2 67,963 NM
023409 6
Mitogen-activated protein kinase kinase
kinase 7 interacting protein 2
68,652 NM
138667 1
RIKEN cDNA 1810015C11 gene 69,102 AK007507 2
RIKEN cDNA 8430411H09 gene 71,474 NM
027805 1
RIKEN cDNA 2700083E18 gene 72,640 AK012557 1
RIKEN cDNA 2410091C18 gene 73,694 NM
028611 1
Neuropilin (NRP) and tolloid (TLL)-like 2 74,513 XM
134498 2
Signal peptide peptidase 3 74,585 AK014709 2
RIKEN cDNA 9430072K23 gene 77,292 AK020483 1
cDNA sequence BC005537 79,555 NM
024473 1
Table 1 (continued)
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127120
Page 5
three novel cDNA clones with no significant homology to
known genes as determined by BLAST analysis.
3.2. Classification of cDNAs
To classify the types of clones isolated in the signal
sequence trap, cDNAs with an assigned GeneID were
further catagorized by gene ontology (GO). GO is one of the
most important tools available for the processing of
information about gene products and their function. The
list of clones was imp orted into the GO Tree Machine
database, a bioinform atics tool designed to interpret func-
tional information from large sets of genes based on gene
ontology [28]. Gene ontology annota tions for bBiological
ProcessQ were determined for 101 of the cDNAs. The
majority of clones fell under four very broad GO terms: cell
communication, cellular physiological process, metabolism
and morphog enesis. To identify functional similarities
among these broad categories, the cDNAs in each group
were further analyzed by more refined GO annotation.
Thirty-three cDNAs annotated under the GO term cell
communication sorted into one or more of the following
categories: cell adhesion, cell–cell signaling and signal
transduction (Table 2). Twenty are known to function in
signal transduction processes. These included two small
GTPases, rab18 [15] and rhoe [18], and the Rho-family
guanine nucleotide exchange factor Arhgef1 [7] involved in
intracellular signaling cascades. Several clones involved in
cell surface receptor linked signal transduction including the
BMP receptor bmpr1a [6] , the cytokine receptor il2rg [14],
Table 1 (continued)
Description GeneID Accession Library
e16.5 whole
Lactamase, beta 80,907 NM
030717 2
RIKEN cDNA E430021N18 gene 96,935 NM
144796 1
Expressed sequence AI467484 98,376 XM
129584 1
Transmembrane 9 superfamily protein member 4 99,237 NM
133847 2
BAC clone RP23-323E20 from chromosome 3 AC127283 1
BAC clone RP23-413O13 from chromosome 8 AC105256 1
BAC clone RP23-67F22 from 18 AC125218 1
BAC clone RP24-117B21 from chromosome 5 AC124480 1
Chromosome 1, clone RP23-363M8 AC114588 3
Chromosome 1, clone RP24-499L20 AC125444 1
Chromosome 14 clone RP23-96F5 AC132470 2 1
Chromosome 18, clone RP23-478L7 AC101740 1
Chromosome 3 clone RP23-181C16 AC123075 1
Chromosome 6 clone RP24-69C19 AC124500 1
DNA sequence from clone RP23-233B9
on chromosome 4
AL669953 1
DNA sequence from clone RP23-426N4
on chromosome 4
AL671173 6
Mouse DNA sequence from clone
RP23-176N9 on chromosome 2
AL691489 3
Mouse DNA sequence from clone
RP23-190C17 on chromosome 2
AL806528 1
Mouse DNA sequence from clone
RP23-25A3 on chromosome X
AL662923 1
Mouse DNA sequence from clone
RP23-32L5 on chromosome 4
AL670673 1
Mouse DNA sequence from clone
RP23-366M19 on chromosome 11
AL592222 1
Mouse DNA sequence from clone
RP23-378I13 on chromosome 11
AL662838 1
Mouse DNA sequence from clone
RP23-41F14 on chromosome 11
AL713919 18
Cerebellum 175 1
Cerebellum 232 7
Cerebellum 40 1
RIKEN clone:A430046D13 AK040027 1
RIKEN clone:A630031M23 AK041714 1
RIKEN clone:D430004P15 AK084877 1
Strain C57BL6/J Chromosome
15 BAC, RP23-99A15
AC087113 1
428 221 70
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127 121
Page 6
and the neurotransmitter receptor gabrg2 [13] also fell into
this category.
Forty-one clones annotated under the GO term cellular
physiological process function in cell death, cell growth
and/or maintenance, cell motility and/or extracellular matrix
organization and biogenesis (Table 2). The majority of these
participate in growth and/or maintenance processes includ-
ing proliferation and transport. Many are known to be active
in cell organization and biogenesis, such as the actin binding
proteins coro1c [19] and macf1 [2].
Of the clones annotated under the GO term metabolism,
24 are involved in protein metabolism (Table 2). Nine
function in protein modification processes such as glyco-
sylation, dephosphor ylation and ph osphorylation. The
kinases isolated include two serine/threonine kinases snrk
[1,12] and prkwnk1 [4,27], camk2a [8] and ick [24].
Under the GO term morphogenesis, 17 clones are
involved in cellular morphogenesis and/or organogenesis
(Table 2). Not surprisingly, the majority of clones are
involved in neurogenesis, including several clones known to
function in axon guidance, such as the semaphorin sema3c
[21], the semaphorin receptor nrp2 [3] and the Slit receptor
robo2 [25].
Based on these analyses, we learned that the majority of
clones isolated function to regulate cell communication, cell
physiology, metabolism and morphogenesis. This is con-
sistent with the biological processes that are known to be
critical for brain development. Interestingly, many of the
clones have not yet been described to be active during
development of t he nervous system. We used in situ
hybridization to begin to characterize the function of these
genes in the brain.
3.3. ISH expression analysis
To further characterize the clones isolated in the signal
sequence trap, we determined their pattern of expression in
the developing brain. In situ hybridization analysis was
performed for all of the clones on sagittal sections from
e14.5 and e16.5 mouse brains. All of the clones analyzed
showed some level of expression in the brain at both time
points. However, the signal intensity and pattern of
Table 2
List of cDNAs sorted by GO annotation for biological process
GO term GO term ID Gene symbol
Cell communication GO:0007154
Cell adhesion GO:0007155 Col2a1 Galnt1 Lphn1 Pcdhb4 Tm4sf9
Col4a5 Hnt Ncam1 Reg3g Vcam1
Cspg2 Ibsp Nrp2 Rhoe
Cell–cell signaling GO:0007267 Camk2a Gabrg2 Nrxn2
Signal transduction GO:0007165 Adcy6 Ibsp Lepr Mt2 Pkd1 Rhoe
Arhgef1 Ick Lphn1 Nras Ptpn13 Snx5
Bmpr1a Igf1 Man1b Pde3b Rab18 Stmn1
Gabrg2 Il2rg
Cellular physiological process GO:0050875
Cell death GO:0008219 Igf1 Madd
Cell growth and/or maintenance GO:0008151 Abcb1a Camk2a Glce Macf1 Prnp Slco1c1
Apod Col2a1 Gp38 Mdk Psmd8 Snx5
Apoe Coro1c Hnrpa1 Mt2 Rab18 Stmn1
Atp5b D3Ertd330e Hnrpd Nov Rhoe Tgfb2
Atp5o Fabp5 Igfbp5 Nras Rn18s Tm4sf9
Baz1b Gabrg2 Lzf Pkd1 Rpl12 Ttr
Cell motility GO:0006928 Macf1 Robo2 Tm4sf9
Extracellular matrix organization
and biogenesis
GO:0030198 Col2a1 Nrxn2 Tgfb2
Morphogenesis GO:0009653
Cellular morphogenesis GO:0000902 Igfbp5 Macf1 Nov Tgfb2
Organogenesis GO:0009887 Col2a1 Igf1 Nrp2 Pkd1 Sema3c Stmn1
Gp38 Lzf Nrxn2 Robo2 Slitrk2 Tm4sf9
Metabolism GO:0008152
Protein metabolism GO:0019538 Apoe Ctsk Fkbp2 Ick Pa2g4 Scpep1
Bmpr1a Ctsl Fkbp7 Man1b Prkwnk1 Snrk
Camk2a Eif3s8 Galnt1 Nrd1 Psmd8 Ube2n
Col2a1 Enc1 Glce Rpl12 Ptpn13
List of cDNA clones were sorted by biological process using the GOTree Machine bioinformatics tool (
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127122
Page 7
expression varied greatly and was independent of known
GO biological process.
We grouped the expression patterns based on the level of
characterization in the literature to highlight the clones that
may have unexplored roles in the nervous system. Clones
were grouped as follows: (1) GO term annotated, expression
pattern available, (2) GO term annotated, expression pattern
in brain not available, (3) GO term not annotated and
expression pattern in brain not available and (4) novel
clones. Fig. 1 shows the expression pattern of many genes
that have already been shown to play a role in the nervous
system and an expression pattern is av ailable in the
literature. Another interesting group of genes could be
classified by biological process as defined by Gene
Ontology; however, to our knowledge, expression pattern
information is not available in the literature. Of genes in this
group, in situ hybridization analysis with probes for sorting
nexin 5 (snx5) and ras homolog gene family, member E
(rhoE) produced interesting expression patterns. snx5 is a
member of the SNX family of proteins that interact with
FANCA protein (Fanconi anemia complementation group
A). SNX proteins are thought to play important roles in
receptor trafficking between cellular organelles. snx5
expression can be detected in human cancer cell lines and
normal tissues [20]. In embryonic mouse brain, snx5 was
detected in the ventricular zones of the cerebral cortex,
midbrain and cerebellum, suggesting that snx5 may play a
specific role in receptor trafficking in cells that occupy these
regions (Fig. 2). rhoE is a member of the Rnd subfamily that
inhibits RhoA/ROCK signaling to promote actin stress fiber
and focal adhesion disassembly and inhibits cell cycle
progression in fibroblasts [26]. rhoE expression is high in
the basal ganglia, midbrain and cerebellum (Fig. 2),
suggesting that rhoE functions to inhibit cell cycle
progression in these structures. There were a number of
clones with minimal bioinform atics available, they were not
annotated for a role in biol ogical processes, and they have
not been characterized in the brain. We analyzed these genes
by in situ hybridization (Fig. 3). lect1 , neto2,
8430411H09Rik and tm4sf2 all displayed spatially restricted
expression patterns during brain development (Fig. 3). Of
particular note, lect1 (leukocyte cell derived chemotaxin 1)
encodes a protein with a putative tumor suppressor function,
and anti-angiogenic and bone remodeling activity [9,17].
lect1 has been extensively studied in chondrosarcomas and
cartilage tissue, but investigations into its function in the
Fig. 1. Gene trap clones with known biological function and brain expression pattern. Darkfield images of embryonic day e14.5 and e16.5 sagittal mouse brain
sections from in situ hybridization analysis using riboprobes specific for the indicated gene. Gene names of expression patterns presented in the left-hand
column: col2a, procollagen, type II, alpha 1; hnt, Neurotrimin, stmn1, Stathmin 1; ptpn13, Protein tyrosine phosphatase, non–receptor type 13. Gene names of
expression patterns presented in the right-hand column: encl, Ectodermal-neural cortex 1; igfbp5, Insulin–like growth factor binding protein 5; slitrk2, SLIT
and NTRK–like family, member 2, tgfb2, Transforming growth factor, beta 2. Calibration bar in lower right corner = 800 mm.
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127 123
Page 8
brain are lacking. The highly restricted pattern in the
developing brain and its putative biological functions raise
interesting questions about the role of lect1 in neuronal
development (Fig. 3).
Lastly, we isolated three cDNA sequences that showed no
significant homology with any sequences in GenBank. All
three were isolated from the e14.5 cerebellum library.
Cerebellum clone 175 and clone 232 displayed expression
throughout the brain at both e14.5 and e16.5 with cells in the
cortical plate in the cerebral cortex containing more robust
signal levels compared to other brain regions (Fig. 4).
Cerebellum clone 40 displayed ubiquitous low-level expres-
sion throughout the brain. We performed peptide alignment
analysis using the predicted amino acid sequence, but no
significant homology was found with any proteins in the
available databases.
4. Discussion
Understanding the function of genes during nervous
system development is challenging. To simplify the under-
taking, we employed a genetic screen to isolate genes that
encode proteins destined for the secretory pathway using the
yeast signa l sequence gene trap. The identity of hundreds of
clones was determined by nucleotide sequencing and classes
of genes wer e grou p ed tog et he r usin g gene o nt ol ogy
assignments with respect to biological processes. Several
major biological processes are represented by the clones
isolated in our screen. As an initial approach to assign
function to the isolated clones, we used in situ hybridization
analysis to determine spatial and temporal expression
patterns during brain development. We demonstrate that
knowledge of the gene expression pattern coupled with
bioinformatics can supply a solid framework to launch an
investigation of gene function.
The yeast libraries were constructed from nervous system
tissue where proli feration, migration and synaptogenesis are
all ongoing. This strategy proved lucrative given that over
700 clones representing 189 unique genes survived the yeast
screen. Various clones were represented numerous times
within each library and a handful of clones were represented
across all three libraries. The majority of genes we re
identical to gene entries in GenBank using a BLASTN
Fig. 2. Gene trap clones with known biological function but no brain expression pattern available in the literature. Darkfield images of embryonic day e14.5
and e16.5 sagittal mouse brain sections from in situ hybridization analysis using riboprobes specific for the indicated gene. Gene names of expression patterns
presented in the left-hand column: corolc, Coronin, actin binding protein 1C; lphn1, Latrophilin 1; snx5, Sorting nexin 5; lzf, Leucine zipper domain protein.
Gene names of expression patterns presented in the right-hand column: atic, 5–aminoimidazole–4–carboxamide ribonucleotide formyltransferase/IMP; rhoe,
Ras homolog gene family, member E; rab18, RAB18, member RAS oncogene family, cspg2, Chondroitin sulfate proteoglycan 2. Calibration bar in lower right
corner = 800 mm.
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127124
Page 9
analysis. Three clones encoded novel cDNA fragments that
showed very little similarity with known genes, thus further
analysis is needed to understand their role in development.
Following nu cleotide sequence de termination, we used
Gene Ontology to classify the known genes into biological
functional groups. Of the 189 genes, 101 encoded proteins
with entries in the Gene Ontology Consortium datatabase.
This database contains informat ion about the behavior of
known genes based on their associated molecular function,
cellular compo nents and biol ogical processes. Since our
goal is to begin to assign function to the isolated clones, all
known genes were sorted according to biological process.
Four major biological processes stood out among the rest.
Not surprising, the majority of clones sorted into processes
involved in cell communication, cell physiology processes
(cell growth and maintenance), metabolism and morpho-
genesis. All of these processes are known to be critical during
brain development. What was surprising is that many of the
genes isol ated in the yeast signal sequence gene trap were not
previously known to function in the nervous system. Gene
ontology analysis offered an unbiased classification of gene
function regardless of the species, tissue syst em, or cell type
that the function was originally published. Because biological
mechanisms are conserved between different tissue systems,
we can now propose similar functional roles for these genes
in the nervous system.
Gene function can be determined in technically challeng-
ing techniques, such as loss of function experiments. How-
ever, it has been estimated that 30% of these experiments
produce no effect making information on gene function
impossible to obtain [16]. When screening multiple genes
simultaneously, a more reliable but quick approach to execute
the objectives is desirable. To begin to investigate gene func-
tion in the developing nervous system, we chose to gather
temporal and spatial gene expression pattern information in
embryonic brain tissue. The expression pattern of a gene is
fundamental to its ability to exert its appropriate biological
function. We rationalized that knowing gene expression
patterns through in situ hybridization coupled with bioinfor-
matics could serve as a platform to begin to investigate gene
function in the brain. In situ hybridization protocols were
amended for use in a systematic fashion to obtain mRNA
distribution in embryonic day 14.5 and 16.5 brain.
A broad range of gene expression patterns was observed
after screening 189 clones by in situ hybridization analysis.
We were surprised that there was no expression pattern
information for many of the genes isolated from the gene trap
screen. Though the literature indicates a role for the gene in
the b rain, it lacks fundamental information such as pattern of
gene expression durin g development. The genes displayed in
Fig. 2 show genes that have been assigned to a biological
functional group but the pattern of gene expression in the
brain is currently unavailable. We attempted to sort the
expression patterns according to the four major biological
process groups that were highlighted by gene ontology
analysis. We discovered that regardless of the gene ontology,
the expression patterns were distributed randomly. Gene
expression patterns showing restriction to the ventricular
zone, the meninges, or other anatomical structures could be
observed in all ontology categories suggesting that ontology
Fig. 3. Gene trap clones with no known biological function and no brain expression pattern available in the literature. Darkfield images of embryonic day e14.5
and e16.5 sagittal mouse brain sections from in situ hybridization analysis using riboprobes specific for the indicated gene. Gene names of expression patterns
presented in the left-hand column: neto2, Neuropilin (NRP) and tolloid (TLL)-like 2; lect1, Leukocyte cell derived chemotaxin 1; tm4sf2, Transmembrane 4
superfamily member 2. Gene names of expression patterns presented in the right-hand column: 8430411H09Rik, RIKEN cDNA 8430411H09 gene;
1810015C11Rik, RIKEN cDNA 1810015C11 gene; BC005537, cDNA sequence BC005537. Calibration bar in lower right corner = 800 mm.
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127 125
Page 10
does not dictate expression pattern and vice versa. Rather, we
found that coupling gene ontology and other biological
platform to formulate new experiments to investigate gene
function in the nervous system.
We propose that expression pattern can function as a
central piece of information to highlight genes that may be
critical for specific events in brain development. For
example, gene expression patterns showing re stricted
expression in the cortical plate of the developing cerebral
cortex but not present in the ventricular zone suggest that
the gene plays a role in post-mitotic neurons but may not be
as necessary for proliferation (neto2, lphn, scpg2, enc1); see
Figs. 1–3. Combining the temporal and spatial expression
pattern with the available bioinformatics, chromosomal
location, gene ontology and predicted protein domains, we
could begin to hypothesize gene functions. For example,
neto2 (neuropillin and tolloid -like gene 2) is specifically
expressed in the develo ping cortical plate, and is undetect-
able in the ventricular zone of the cerebral cortex at
embryonic days 14.5 and 16.5. neto2 encodes a trans-
membrane protein with similarity to neuropilin-1 and
tolloid. It contains CUB domains and a low-density
lipoprotein class A module that is expressed in the retina
and brain [23]. Due to its homology with neuropilin and
tolloid-1, coupled with its expression pattern in the
developing cortical plate, we hypothesize that neto2 may
play a role in guidance of neuronal processes by interacting
with semaphorin-like molecules during formation of the
cerebral cortex. The appropriate experimental tests can now
be established to directly test this idea that was proposed
based on bioinformatics and gene expression pattern.
Our primary goal was to identify genes that function
during brain development. We utilized a primary screen to
identify genes th at encode secreted or transmembrane
proteins, then we used gene expression patterns to filter
those genes that may play a role in specific events in brain
development. Our analysis demonstrated that gene expres-
sion pattern may be applied as a primary filter to select for
genes with specific functions during development when
Fig. 4. Novel cDNA clones isolated from signal sequence gene trap. Three cDNA clones were isolated in the gene trap screen with no significant homology to
known genes. We used in situ hybridization analysis to determine the pattern of mRNA distribution in e14.5 and e16.5 mouse brain. Cerebellum clone 175 and
cerebellum clone 232 are ubiquitously expressed but more robust signal is detected in the cerebral cortex at both ages. Cerebellum clone 40 is expressed
ubiquitously at low levels at both ages. Calibration bar in lower right corner = 800 mm.
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127126
Page 11
pattern is combined with available bioinformatics from
Entrez gene at NCBI, Gene Ontology Consortium and other
databases available on the web. From our experience in this
project, we conclude that gene ontology functional groups
may provide a great source of candidate genes for screening
by in situ hybridization in the developing brain.
A database of expression patterns of all genes during
brain development would provide a valuable resource for the
neuroscience community. Academic and industrial research
projects would benefit from a central repository of gene
expression pattern information that would eliminate redun-
dant efforts to resolve such a primary characteristic of a gene
as its normal distribution pattern during development. Our
efforts in this project showed that expression pattern is a
fundamental component of gene function and therefore it is a
central piece of information that is critical for investigating
the role of the gene during normal development.
The authors would like to thank Shannon Dupuy-Davies
for helpful suggestions in the setup of this project and Anna
Seal, Andrew Asbury and Tony Cheung for help with in situ
hybridization processing. This work was supported in part
by NIH grant R37 NS36558 (T.C.) and the American
Lebanese Syrian Associated Charities (ALSAC).
[1] W. Becker, J. Heukelbach, H. Kentrup, H.G. Joost, Molecular cloning
and characterization of a novel mammalian protein kinase harboring a
homology domain that defines a subfamily of serine/threonine
kinases, Eur. J. Biochem. 235 (1996) 736 743.
[2] G. Bernier, M. Mathieu, Y. De Repentigny, S.M. Vidal, R. Kothary,
Cloning and characterization of mouse ACF7, a novel member of the
dystoninsubfamilyofactinbindingproteins,Genomics38(1996)19 29.
[3] H. Chen, A. Chedotal, Z. He, C.S. Goodman, M. Tessier-Lavigne,
Neuropilin-2, a novel member of the neuropilin family, is a high
affinity receptor for the semaphorins Sema E and Sema IV but not
Sema III, Neuron 19 (1997) 547 559.
[4] K.A. Choate, K.T. Kahle, F.H. Wilson, C. Nelson-Williams, R.P. Lifton,
WNK1, a kinase mutated in inherited hypertension with hyperkalemia,
localizes to diverse Cl
-transporting epithelia, Proc. Natl. Acad. Sci. U.
S. A. 100 (2003) 663 668 (Electronic publication 2003 Jan 8).
[5] P.S. Crosier, A. Bardsley, J.A. Horsfield, A.K. Krassowska, E.R.
Lavallie, L.A. Collins-Racie, J.H. Postlethwait, Y.L. Yan, J.M. McCoy,
K.E. Crosier, In situ hybridization screen in zebrafish for the selection
of genes encoding secreted proteins, Dev. Dyn. 222 (2001) 637 644.
[6] S. Davis, S. Miura, C. Hill, Y. Mishina, J. Klingensmith, BMP receptor
IA is required in the mammalian embryo for endodermal morpho-
genesis and ectodermal patterning, Dev. Biol. 270 (2004) 4763.
[7] T.M. Eisenhaure, S.A. Francis, L.D. Willison, S.R. Coughlin, D.J.
Lerner, The Rho guanine nucleotide exchange factor Lsc homo-
oligomerizes and is negatively regulated through domains in its
carboxyl terminus that are absent in novel splenic isoforms, J. Biol.
Chem. 278 (2003) 30975 30984.
[8] R.M. Hanley, M.E. Payne, F. Cruzalegui, M.A. Christenson, A.R.
Means, Sequence of the cDNA for the alpha subunit of calmodulin
kinase II from mouse brain, Nucleic Acids Res. 17 (1989) 3992.
[9] Y. Hiraki, T. Kono, M. Sato, C. Shukunami, J. Kondo, Inhibition of
DNA synthesis and tube morphogenesis of cultured vascular endothelial
cells by chondromodulin-I, FEBS Lett. 415 (1997) 321 324.
[10] A.C. Horton, M.D. Ehlers, Secretory trafficking in neuronal dendrites,
Nat. Cell Biol. 6 (2004) 585 591.
[11] K.A. Jacobs, L.A. Collins-Racie, M. Colbert, M. Duckett, M. Golden-
Fleet, K. Kelleher, R. Kriz, E.R. LaVallie, D. Merberg, V. Spaulding, J.
Stover, M.J. Williamson, J.M. McCoy, A genetic selection for isolating
cDNAs encoding secreted proteins, Gene 198 (1997) 289 296.
[12] N. Kertesz, J. Samson, C. Debacker, H. Wu, M.C. Labastie, Cloning
and characterization of human and mouse SNRK sucrose non-
fermenting protein (SNF-1)-related kinases, Gene 294 (2002) 13 24.
[13] P. Kofuji, J.B. Wang, S.J. Moss, R.L. Huganir, D.R. Burt, Generation of
two forms of the gamma-aminobutyric acidA receptor gamma 2-subunit
in mice by alternative splicing, J. Neurochem. 56 (1991) 713 715.
[14] S. Kumaki, M. Kondo, T. Takeshita, H. Asao, M. Nakamura, K.
Sugamura, Cloning of the mouse interleukin 2 receptor gamma chain:
demonstration of functional differences between the mouse and human
receptors, Biochem. Biophys. Res. Commun. 193 (1993) 356 363.
[15] E.B. McMurtrie, M.D. Barbosa, M. Zerial, S.F. Kingsmore, Rab17
and rab18, small GTPases with specificity for polarized epithelial
cells: genetic mapping in the mouse, Genomics 45 (1997) 623 625.
[16] G.L. Miklos, G.M. Rubin, The role of the genome project in determining
gene function: insights from model organisms, Cell 86 (1996) 521 529.
[17] Y. Nakamichi, C.Shukunami,T. Yamada, K. Aihara, H. Kawano, T. Sato,
Y. Nishizaki, Y. Yamamoto, M. Shindo, K. Yoshimura, T. Nakamura, N.
Takahashi,H.Kawaguchi, Y.Hiraki,S.Kato, ChondromodulinI is abone
remodeling factor, Mol. Cell. Biol. 23 (2003) 636 644.
[18] C.D.Nobes,I.Lauritzen,M.G.Mattei,S.Paris,A.Hall, P. Chardin,A new
member of the Rho family, Rnd1, promotes disassembly of actin filament
structures and loss of cell adhesion, J. Cell Biol. 141 (1998) 187 197.
[19] M. Okumura, C. Kung, S. Wong, M. Rodgers, M.L. Thomas,
Definition of family of coronin-related proteins conserved between
humans and mice: close genetic linkage between coronin-2 and
CD45-associated protein, DNA Cell Biol. 17 (1998) 779 787.
[20] T. Otsuki, S. Kajigaya, K. Ozawa, J.M. Liu, SNX5, a new member of the
sortingnexinfamily, bindstotheFanconianemiacomplementationgroup
A protein, Biochem. Biophys. Res. Commun. 265 (1999) 630 635.
[21] A.W. Puschel, R.H. Adams, H. Betz, Murine semaphorin D/collapsin
is a member of a diverse gene family and creates domains inhibitory
for axonal extension, Neuron 14 (1995) 941 948.
[22] D.S. Rice, M. Sheldon, G. D’Arcangelo, K. Nakajima, D. Goldowitz,
T. Curran, Disabled-1 acts downstream of Reelin in a signaling
pathway that controls laminar organization in the mammalian brain,
Development 125 (1998) 3719 3729.
[23] H. Stohr, C. Berger, S. Frohlich, B.H. Weber, A novel gene encoding a
putative transmembrane protein with two extracellular CUB domains
and a low-density lipoprotein class A module: isolation of alternatively
spliced isoforms in retina and brain, Gene 286 (2002) 223231.
[24] K. Togawa, Y.X. Yan, T. Inomoto, S. Slaugenhaupt, A.K. Rustgi,
Intestinal cell kinase (ICK) localizes to the crypt region and requires a
dual phosphorylation site found in map kinases, J. Cell. Physiol. 183
(2000) 129 139.
[25] N. Vargesson, V. Luria, I. Messina, L. Erskine, E. Laufer, Expression
patterns of Slit and Robo family members during vertebrate limb
development, Mech. Dev. 106 (2001) 175 180.
[26] P. Villalonga, R.M. Guasch, K. Riento, A.J. Ridley, RhoE inhibits cell
cycle progression and Ras-induced transformation, Mol. Cell. Biol. 24
(2004) 7829 7840.
[27] B. Xu, J.M. English, J.L. Wilsbacher, S. Stippec, E.J. Goldsmith,
M.H. Cobb, WNK1, a novel mammalian serine/threonine protein
kinase lacking the catalytic lysine in subdomain II, J. Biol. Chem. 275
(2000) 16795 16801.
[28] B. Zhang, D. Schmoyer, S. Kirov, J. Snoddy, GOTree Machine
(GOTM): a web-based platform for interpreting sets of interesting
genes using Gene Ontology hierarchies, BMC Bioinformatics 5
(2004) 16.
P. Jensen et al. / Molecular Brain Research 132 (2004) 116–127 127
Page 12
  • Source
    • "This type of experiment often leads to the identification of hundreds of differences in the expression of genes. Statistical analysis tools using GO annotations in the context of the GO hierarchy are used to cluster those genes into functional categories (Jensen et al., 2004; Xu et al., 2007; Hecht et al., 2007; Zhu et al., 2007; Matsuki et al., 2005; van Lunteren et al., 2006; Clemente et al., 2006; Agbemafle et al., 2005; Ivins et al., 2005; Choi et al., 2007; Baguma-Nibasheka et al., 2007; Vaes et al., 2006; Zhang et al., 2004; Li et al., 2006). An example of this type of experiment is the one carried out by James et al to study gene expression during chondrocyte differentiation (James et al., 2005). "
    [Show abstract] [Hide abstract] ABSTRACT: Developmental biology, like many other areas of biology, has undergone a dramatic shift in the perspective from which developmental processes are viewed. Instead of focusing on the actions of a handful of genes or functional RNAs, we now consider the interactions of large functional gene networks and study how these complex systems orchestrate the unfolding of an organism, from gametes to adult. Developmental biologists are beginning to realize that understanding ontogeny on this scale requires the utilization of computational methods to capture, store and represent the knowledge we have about the underlying processes. Here we review the use of the Gene Ontology (GO) to study developmental biology. We describe the organization and structure of the GO and illustrate some of the ways we use it to capture the current understanding of many common developmental processes. We also discuss ways in which gene product annotations using the GO have been used to ask and answer developmental questions in a variety of model developmental systems. We provide suggestions as to how the GO might be used in more powerful ways to address questions about development. Our goal is to provide developmental biologists with enough background about the GO that they can begin to think about how they might use the ontology efficiently and in the most powerful ways possible.
    Full-text · Article · Jan 2009 · Molecular Reproduction and Development
  • Source
    • "GDIR is known to be involved in regulation of the actin cytoskeleton [14]. CALU has been implicated in calcium signal modulation, and its mRNA has been reported to be expressed in whole head on embryonic day 16.5 [15,16]. These suggest a probable link between active signal transduction and changes in the cytoskeleton during the differentiation from ES cells to NS cells. "
    [Show abstract] [Hide abstract] ABSTRACT: Embryonic stem (ES) cells are pluripotent stem cells and give rise to a variety of differentiated cell types including neurons. To study a molecular basis for differentiation from ES cells to neural cells, we searched for proteins involved in mouse neurogenesis from ES cells to neural stem (NS) cells and neurons by two-dimensional gel electrophoresis (2-DE) and peptide mass fingerprinting, using highly homogeneous cells differentiated from ES cells in vitro. We newly identified seven proteins with increased expression and one protein with decreased expression from ES cells to NS cells, and eight proteins with decreased expression from NS cells to neurons. Western blot analysis confirmed that a tumor-specific transplantation antigen, HS90B, decreased, and an extracellular matrix and membrane glycoprotein (such as laminin)-binding protein, galectin 1 (LEG1), increased in NS cells, and LEG1 and a cell adhesion receptor, laminin receptor (RSSA), decreased in neurons. The results of RT-PCR showed that mRNA of LEG1 was also up-regulated in NS cells and down-regulated in neurons, implying an important role of LEG1 in regulating the differentiation. The differentially expressed proteins identified here provide insight into the molecular basis of neurogenesis from ES cells to NS cells and neurons.
    Full-text · Article · Jun 2008 · Biochimica et Biophysica Acta
  • Source
    • "Several large-scale, mouse brain-specific gene expression reference data sets are being constructed which are critical resources for atlas-based integrated analysis of neurodevelopment and neurodegenerative disease [12]. All of these data sets are anatomically-mapped, some by virtue of being assays performed directly in a tissue context both using standard in situ immunohistochemistry, either flu- orescence [13,14], alkaline-phosphatase [15] , digoxygenin-labeled antisense riboprobes (The Allen Brain Atlas)[12], or radioactivity (Brain Gene Expression Map or BGEM) [16], as well as GFP-reporter constructs used in the Gene Expression Nervous System Atlas (GENSAT) BAC Transgenics project [17] . For others, the anatomical context is defined only to a relatively coarse level of granularity by microdissecting mesoscopic brain regions181920. "
    [Show abstract] [Hide abstract] ABSTRACT: Three dimensional biomedical image sets are becoming ubiquitous, along with the canonical atlases providing the necessary spatial context for analysis. To make full use of these 3D image sets, one must be able to present views for 2D display, either surface renderings or 2D cross-sections through the data. Typical display software is limited to presentations along one of the three orthogonal anatomical axes (coronal, horizontal, or sagittal). However, data sets precisely oriented along the major axes are rare. To make fullest use of these datasets, one must reasonably match the atlas' orientation; this involves resampling the atlas in planes matched to the data set. Traditionally, this requires the atlas and browser reside on the user's desktop; unfortunately, in addition to being monolithic programs, these tools often require substantial local resources. In this article, we describe a network-capable, client-server framework to slice and visualize 3D atlases at off-axis angles, along with an open client architecture and development kit to support integration into complex data analysis environments. Here we describe the basic architecture of a client-server 3D visualization system, consisting of a thin Java client built on a development kit, and a computationally robust, high-performance server written in ANSI C++. The Java client components (NetOStat) support arbitrary-angle viewing and run on readily available desktop computers running Mac OS X, Windows XP, or Linux as a downloadable Java Application. Using the NeuroTerrain Software Development Kit (NT-SDK), sophisticated atlas browsing can be added to any Java-compatible application requiring as little as 50 lines of Java glue code, thus making it eminently re-useable and much more accessible to programmers building more complex, biomedical data analysis tools. The NT-SDK separates the interactive GUI components from the server control and monitoring, so as to support development of non-interactive applications. The server implementation takes full advantage of data center's high-performance hardware, where it can be co-localized with centrally-located, 3D dataset repositories, extending access to the researcher community throughout the Internet. The combination of an optimized server and modular, platform-independent client provides an ideal environment for viewing complex 3D biomedical datasets, taking full advantage of high-performance servers to prepare images and subsets of associated meta-data for viewing, as well as the graphical capabilities in Java to actually display the data.
    Full-text · Article · Feb 2007 · BMC Bioinformatics
Show more