Identification of candidate genes at the corticoseptal boundary during development.

Abstract

Cortical midline glia are critical to the formation of the corpus callosum during development. The glial wedge is a population of midline glia that is located at the corticoseptal boundary and expresses repulsive/growth-inhibitory molecules that guide callosal axons as they cross the midline. The glial wedge are the first cells within the cortex to express GFAP and thus may express molecules specific for glial maturation. The corticoseptal boundary is a genetically defined boundary between the cingulate cortex (dorsal telencephalon) and the septum (ventral telencephalon). The correct dorso-ventral position of this boundary is vital to the formation of both the glial wedge and the corpus callosum. Our aim was to identify genes expressed specifically within the glial wedge that might be involved in either glial differentiation, formation of the corticoseptal boundary or development of the corpus callosum. To identify such genes we have performed a differential display PCR screen comparing RNA isolated from the glial wedge with RNA isolated from control tissues such as the neocortex and septum, of embryonic day 17 mouse brains. Using 200 different combinations of primers, we identified and cloned 67 distinct gene fragments. In situ hybridization analysis confirmed the differential expression of many of the genes, and showed that clones G24F3, G39F8 and transcription factor LZIP have specific expression patterns in the telencephalon of embryonic and postnatal brains. An RNase Protection Assay (RPA) revealed that the expression of G39F8, G24F3 and LZIP increase markedly in the telencephalon at E16 and continue to be expressed until at least P0, during the period when the corpus callosum is forming.

Full text

Page 1

Identification of candidate genes at the corticoseptal boundary
during development
Wei-Bin Shena, Ce ´line Placheza, Aika S. Mongia, Linda J. Richardsa,b,*
aDepartment of Anatomy and Neurobiology and The Program in Neuroscience, The University of Maryland School of Medicine, Baltimore, MD 21201, USA
bSchool of Biomedical Sciences and The Queensland Brain Institute, The University of Queensland, St Lucia, Qld 4067, Australia
Received 20 May 2005; received in revised form 8 November 2005; accepted 8 November 2005
Available online 2 February 2006
Abstract
Cortical midline glia are critical to the formation of the corpus callosum during development. The glial wedge is a population of midline glia
that is located at the corticoseptal boundary and expresses repulsive/growth-inhibitory molecules that guide callosal axons as they cross the
midline. The glial wedge are the first cells within the cortex to express GFAP and thus may express molecules specific for glial maturation. The
corticoseptal boundary is a genetically defined boundary between the cingulate cortex (dorsal telencephalon) and the septum (ventral
telencephalon). The correct dorso-ventral position of this boundary is vital to the formation of both the glial wedge and the corpus callosum. Our
aim was to identify genes expressed specifically within the glial wedge that might be involved in either glial differentiation, formation of the
corticoseptal boundary or development of the corpus callosum. To identify such genes we have performed a differential display PCR screen
comparing RNA isolated from the glial wedge with RNA isolated from control tissues such as the neocortex and septum, of embryonic day 17
mouse brains. Using 200 different combinations of primers, we identified and cloned 67 distinct gene fragments. In situ hybridization analysis
confirmed the differential expression of many of the genes, and showed that clones G24F3, G39F8 and transcription factor LZIP have specific
expression patterns in the telencephalon of embryonic and postnatal brains. An RNase Protection Assay (RPA) revealed that the expression of
G39F8, G24F3 and LZIP increase markedly in the telencephalon at E16 and continue to be expressed until at least P0, during the period when the
corpus callosum is forming.
q 2005 Elsevier B.V. All rights reserved.
Keywords: Midline glia; Glial wedge; Axon guidance; Differentiation; Patterning; Differential display PCR; tcf4; Zinc finger protein 288
The corpus callosum is the largest commissure in the brain,
connecting neurons in the right and left cerebral hemispheres.
Callosal axons first cross the midline at embryonic day 15.5
(E15.5) in mouse (Ozaki and Wahlsten, 1992; Rash and
Richards, 2001) with fibers continually added from E16 to P0.
Cortical midline glia are critical to the formation of the corpus
callosum (Silver et al., 1982; 1993; Shu and Richards, 2001).
Such glia form at the boundary between the septum and the
cingulate cortex in a region known as the corticoseptal
boundary. The corticoseptal boundary is both a morphological
and genetic boundary between dorsally expressed genes such
as Emx1 and 2 and ventrally expressed genes such as vax2 and
Dlx. One glial population at the corticoseptal boundary is the
glial wedge (GW) which is part of the radial glial scaffold
of the cortex but expresses GFAP prior to any other region of
cortex (Shu et al., 2003a). Furthermore, the differentiation of
the GW from radial glia into glial fibrillary acidic protein
positive astrocytes suggests that these cells may be among the
first cells in the cortex to differentiate into mature glia and thus
we wanted to learn more about the molecular profile of these
specialized midline glia. The glial wedge also expresses the
chemorepulsive molecule slit2, required for callosal axon
guidance (Shu and Richards, 2001; Shu et al., 2003a,b).
However, in a number of different mouse mutants that display
agenesis of the corpus callosum and disruption of the glial
wedge, Slit2 is still expressed and repels cortical axons in vitro
(Shu et al., 2003c; Shen et al., 2002) indicating that additional
molecules may be expressed by the glial wedge that could
guide callosal axons at the midline.
To understand what makes these GW cells unique, whether
they express additional guidance molecules, and how the
corticoseptal boundary forms, we undertook a differential
Gene Expression Patterns 6 (2006) 471–481
www.elsevier.com/locate/modgep
1567-133X/$ - see front matter q 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.modgep.2005.11.004
*Corresponding author. Address: School of Biomedical Sciences and The
Queensland Brain Institute, The University of Queensland, Otto Hirschfeld
Building (81), Brisbane,Qld 4072, Australia. Tel.: C61 7 3365 2661/61 7 3365
2882; fax: C61 7 3365 1299.
E-mail address: richards@uq.edu.au (L.J. Richards).

Page 2

display PCR (DD–PCR) screen to identify genes specifically
expressed within this region. The results of this screen
identified genes previously known or predicted to be expressed
in the glial wedge such as the glial-expressed gene vimentin
and the transcription factor Nfib (Steele-Perkins et al., 2005),
however we did not identify any genes specific to the boundary
between the septum and the cingulate cortex nor any secreted
molecules (potentially axon guidance ligands) expressed by the
GW using this method.
1. Results and discussion
1.1. Identification of candidate genes expressed
by the glial wedge
We used DD–PCR to isolate genes from E17 brains that
were expressed at high levels in the GW compared with control
tissues (septum and neocortex) which were chosen because
they lie anatomically close to the GW but do not form part of
the corticoseptal boundary (Fig. 1(A)). Genes selected for
further analysis were highly expressed in the GW but not in the
septum, and either not expressed or expressed at very low
levels in the cortex of E17 brains. To perform the DD–PCR, ten
base pair primers of arbitrary sequence were used to generate
first strand cDNA to be used for the PCR reaction.32P-labeled
DD–PCR products generated from the GW, neocortex and
septum were separated on a 6% denaturing polyacrylamide gel.
Though many differential display bands were found to be
highly expressed in either the cortex or septum samples
(Fig. 1(B)), we did not pursue these genes further because they
were not the focus of our study. Differential bands that were
expressed at a higher level in the GW were excised from the
polyacrylamide gel (Fig. 1(B)), re-amplified and subcloned
into the pBluescript SK vector. The sequences of the inserts
were compared with the Genbank database using the BLAST
search engine. Using 200 primer combinations, we identified
67 differential display fragments with sizes ranging from 120 to
800 bp. These cDNA clones could be grouped according to
their expression patterns. The first group had a widespread
pattern of expression in the brain but with higher levels in the
midline region and ventricular zone (VZ)/subventricular zone
(SVZ) of the telencephalon (data not shown). Preliminary
analysis of these genes using BLAST suggested that they likely
represent genes with more general cell transcription or ‘house-
keeping’ roles. The second group had restricted expression
patterns with high levels at the midline, VZ/SVZ, and/or other
specific regions. This category of genes encoded different types
of proteins including cytoskeletal/process outgrowth related
proteins, extracellular matrix and adhesion molecules, tran-
scription factors, cell surface receptors, intracellular signaling
Fig. 1. Isolation of differential expressed bands. (A) Schematic of a coronal section of E17 brain, showing the regions (hatched boxes) where tissues were collected
for DD-PCR. CC, corpus callosum; GW, glial wedge; GS, glial sling; IGG, indusium griseum glia. (B) A sequencing gel displaying differentially expressed bands.
Total RNA for DD-PCR was isolated from the GW, septum and neocortex of E17 mouse brains.32P-labeled PCR products were separated on 6% denaturing
polyacrylamide gels. To minimize false positives, each reaction was performed in triplicate from three independent preparations of total RNA. Examples of bands
expressed only in the GW (1), septum (2) or neocortex (3) are shownas well as bands expressedin both the GWand the neocortex(4, 5). Onlybands that were highly
expressed in the GW (1, 4) were excised, reamplified and ligated into the pBluescript vector.
Table 1
Overview of genes identified by DD–PCR
Number of DD–PCR primer combinations
Number of sequencing gels
Number of identified genes
Cytoskeletal and related molecules
Extracellular matrix and adhesion molecules
Transcription factors
Cell surface receptors
Intracellular signaling molecules
DNA-binding/chromosome structure genes
Miscellaneous genes
Genes without significant homologues
Total
200
44
67
4
3
12
5
12
8
5
12
67
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481472

Page 3

Table 2
Differential expression of cDNA fragments and homologs identified
DD–PCR fragmentIdentityGenebank accession number
Extracellular matrix and adhesion molecules (3)
G4F1
G4F7
G30F6
Cytoskeleton and related molecules (4)
G4F3, G18F1
G6F2, G18F3
G9F8
G30F5
Transcription factors (12)
G1F3
G1F7
G2F12
G3F11
G7F13
G9F1, G29F4
G11F3
G15F3
Tenascin C gene, hexamarchion (HXB)
Mouse fat 1 cadherin; protocadherin
Mus musculus mRNA for L1 protein
XM_005348, X56160
AJ250768, AF100960
NM_007697, X94310
Vimentin
Acid calponin 3
Mouse ankyrin (epithelia specific Ank-3)
Mouse actin-related protein 1 homolog A, ARP1 (centractin alpha)
NM_011701, X56397
NM_019359, NM_028044
NM_009670
NM_016860
Jumonji
Nuclear factor I/B (NFI-B)
Similar to mouse transcription factor 4 (tcf-4)
T-box brain gene 1 (Tbr1)
Sall-like 1 protein
Mouse N-myc downstream regulated (Ndr2)
CREB, cAMP responsive element binging protein
LZIP protein gene, 50-end, transcription factor, recognizing CRE and AP-1
element
Similar to mesoderm induction early response 1 gene (MI-ER1);
immediately early gene, transcription factor
NFI-X3 transcription factor
Mus musculus nuclear factor, erythroid drived 2; p45 NF-E2 related factor 2
(NRF 2)
Mus musculus ring finger protein 23 (Rnf23), tripartite motif protein 39
(trim39). Transcription factor
NM_021878, D31967
NM_008687, Y07687
NM_013685
NM_009322, U16322
NM_021390, AJ271915
NM_013864
MMU46027
L22167
G23F9AXM_233245, NM_020948
G30F11
G34F13
S81451, Y07688
NM010902, U20532
G40F10, G40F11NM_024468
Cell surface receptors (5)
G1F1, G16F3, G25F7, G42F8
G22F6
G32F7
G25F9, G34F3, G39F10, G40F4
G32F2
Signaling pathway molecules (12)
G1F10
G16F8
G25F8
G16F9
G17F4, G30F9
G17F5
G32F5
G8F29
G29F8A
Eph receptor A4
Mus musculus metabotropic glutamate receptor 8
GABA-A receptor-associated protein-like protein 1
Mus musculus fibroblast growth factor (FGF) receptor-1
Mus musculus notch B, mouse notch 2
NM_007936
U17252.1
AF_180518
U22324.1, NM_010206
X68279, D32210
NELL2 (protein kinase C-binding protein)
Calcium binding protein, intestinal (Cai)
Calcineurin-binding protein calsarcin-1
Protein kinase piccolo
Rattus norvegicus protein kinase WNK1 (WNK1)
Mus musculus Dlxin-1 (Dlxin1)
Mus musculus Deltex 1 (DTX1)
Smad-interacting protein 1 (SIP 1)
Mus musculus similar to kinase D-interacting substance of 220 kDa; ankyrin
repeat-rich membrane-spanning protein
Homo sapiens protein tyrosine phosphatase, receptor type, F polypeptide
(PTPRF), interacting protein (liprin), alpha 1 (PPFIAL)
Mouse mRNA for Drctnnbla, down regulated by Ctnnbl
Mouse WTAP protein, wilms’ tumor 1-associating protein
NM_016743, U59230
NM_009787
NM_016599
AF_128892
AF_227741
NM_019791
AB015422
AF_033116
NM_053795
G32F8 XM_028889, NM_003626
G39F2
G34F7
DNA binding/chromosome structure (8)
G9F7
G9F9
G14F1
G20F3
AB_030242, XM_033623
AJ276707, XM_016021.2
Heterogeneous nuclear ribonucleo-protein A1 (hnrpa1)
HMG1-related DNA-binding protein
Apoptotic chromatin condensation inducer (Acinus)
Mus musculus Swi/SNF related matrix associated, actin regulator of
chromatin, subfamily a-like 1 (Snarcak1)
Homo sapiens TATA box binding protein (TBP)-associated RNA
polymerase II, A, 250KD (TAF2A)
Homo sapiens translocated promoter region (TPR)
Mouse gene for 18S rRNA
Snf2-related CBP activator protein (SRCAP)
NM_010447
S50213
NM_019567
NM_018817
G20F4XM_010137, NM004606
G22F4
G26F2
G29F7
Genes similar to clones in the RIKEN full-length library (6)
G1F4J
XM_001737
X00686
XM034367
Mus musculus 13 day embryo head cDNA, RIKEN full-length enriched
library, clone:3110037C01:homolog to hypothetical 22.1 Kda protein; Mus
musculus X2CR1 mRNA
XM001863 AK014130
(continued on next page)
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481 473

Page 4

molecules, DNA-binding/chromosome structure-related pro-
teins, and miscellaneous genes (Tables 1 and 2). Among these
genes, many have been reported to function in developmental
processes such as axonal pathfinding or neurite outgrowth (e.g.
EphA4, L1-like protein, ankyrin, tenascin-C) (Demyanenko
et al., 1999; Meiners et al., 1999; Kullander, et al., 2001;
Takemoto et al., 2002), and neurogenesis/differentiation of
neural progenitors (e.g. notch2, FGF receptor 1) (Morrison
et al., 2000; Chambers et al., 2001). Some of them may be
involved in the patterning of the dorsal telencephalon such as
sip1 (smad-interacting protein 1) that may interact with smad, a
BMP signaling pathway molecule (Verschueren et al., 1999),
and tcf4 (transcription factor 4), a downstream target gene of
the Wnt signaling cascade (Kolligs et al., 2002). In addition, we
found 12 gene fragments that encode Expressed Sequence Tag
(EST) genes or novel genes without significant homologs in
Genbank (Table 3).
1.2. Expression analysis of genes identified by DD–PCR
in mouse brain
The cell bodies of the GW are located in the medial wall of
the telencephalic ventricle at the corticoseptal boundary.
Thus, we expected that genes isolated from the GW
would be expressed at higher levels at the midline and
particularly within the VZ/SVZ. We analyzed the expression
patterns of the DD–PCR fragments by in situ hybridization
using35S-cRNA probes. These analyses confirmed that many
of the genes were highly expressed in the GW as well as other
regions of the telencephalon, although none were expressed in
the septum (data not shown). Fig. 2 shows examples of (1)
genes expressed at high levels in the VZ/SVZ (e.g. EphA4,
LZIP and G39F8) as well as other regions (Fig. 2(A)); (2)
genes whose expression was primarily in the dorsal
telencephalon such as G24F3, tcf4, NFI-B and smad-
interacting protein 1 (sip1) (Fig. 2(B)); and (3) genes whose
expression was limited to the neuroepithelium and SVZ such
as notch2 and Ndr2 (Fig. 2(C)).
Most of the genes isolated were known genes with
functions in brain development. For instance, the nuclear
factor I (NFI) gene family consists of NFI-A, -B, -C and -X
(Gronostajski, 2000) We identified Nfib and Nfix in the DD–
Table 2 (continued)
DD–PCR fragment IdentityGenebank accession number
G16F1Mus musculus RIKEN cDNA 2610507O21 gene (2610507O21Rik). Mus
musculus 15 day embryo head cDNA, RIKEN full-length enriched library,
clone:4021402N16; product:unnamed protein, putative similar to trans-
membrane protein quicken [Xenopus laevis]
Mus musculus 11 day embryo whole body cDNA, RIKEN full-length
enriched library, clone:2700046A07; products: hypothetical protein
Homo sapiens mRNA for KIAA0467 protein, partial cds
Mus musculus adult male medulla oblongata cDNA, RIKEN full-length
enriched library, clone:6330407J23, products: hypothetical glycine-rich
region containing protein
Mus musculus 16 day embryo head cDNA, RIKEN full-length enriched
library, clone:C130021H13 product: hypothetical Arginine-rich region/
Serine-rich region/Lysine-rich region containing protein
NM_028123 AK076213
G24F2AAK012381
G33F11D
G33F4A
AB007936
AK018139
G34F6
AK081496
Miscellaneous (3)
G8F13
G16F10A
G30F16
G31F12
G34F8
Mus musculus golgi reassembly stacking protein 2 (Grs2-pending)
Similar to mouse reticulon 1A (RTN-1A), a neuron-specific protein
Mus musculus lyosomal membrane glycoprotein 1 (Lamp1)
Mus musculus transgelin (TAGLN) mRNA
Mus musculus dolicho-phosphate (beta-D) mannosyltransferase (DPM1)
NM_027352, XM_130256
BC030455, AB074899
NM_010684
AF149291
NM_010072
Novel, without significant homology 12 addition clones-see Table 3.
Table 3
Summary of novel genes’ expression patterns
Clones Intensity of DD-PCR bandsIdentity
Glial
wedge
SeptumCortex
G4F10
CCKG
Predicted novel gene on
mouse chromosome 11
EST gene: U1-M-BZ1-bke-
d-07-O-UI-30
XM_222179 Rattus norve-
gicus similar to CG17396
gene product [Drosophila
melanogaster]. Predicted
novel gene
AL670462, Mouse DNA
sequence from clone RP23-
225F2 on chromosome X
EST gene on mouse
chromosome 10
Genscan predicted protein
EST gene on mouse
chromosome 11
Predicted gene: Transcrip-
tion factor-4 (tcf4)
Predicted gene: NULP1, a
helix-loop-helix protein
Predicted gene: Similar to
C1-tetrahydrofolate synthase
Predicted gene: Zinc finger
protein 288
Mus Musculus chromosome
9 BAC clone MGS1-117K9
ES cell line
G7F1
CKK
G8F14
CKG
G16F2
CKK
G16F11
CKG
G16F12
G22F3
C
C
K
K
G
K
G24F3
CCKG
G30F10
CCKG
G39F3
CKK
G39F8,
G39F12
G43F3
CKK
CKK
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481474

Page 5

PCR screen (Table 2), which are highly expressed in the
cortex and at the cortical midline (Chaudhry et al., 1997;
Gronostajski, 2000; Shu et al., 2003c; Steele-Perkins et al.,
2005). Furthermore, mutations in either gene causes agenesis
of corpus callosum and abnormalities in midline glial
formation (Shu et al., 2003c; Steele-Perkins et al., 2005).
Several genes such as EphA4, L1-like protein, and tenascin-C
have been reported to be involved in axonal pathfinding or
neurite outgrowth (Demyanenko et al., 1999; Meiners et al.,
1999; Kullander et al., 2001; Takemoto et al., 2002;
Goldshmit et al., 2004). In the EphA4 knockout mouse
corticospinal axons display abnormal growth and pathfinding
upon entering the gray matter of the spinal cord (Coonan
et al., 2001). In situ hybridization analyses of EphA4
demonstrate that it is expressed in the GW and the indusium
griseum at E17 (Fig. 2(A), and data not shown). Other genes
identified in the screen, such as L1 and ankyrin, cause a
reduction or complete agenesis of corpus callosum when
mutated in mice (Scotland et al., 1998; Demyanenko et al.,
1999). Interestingly, no secreted guidance factors were
isolated from the screen. This included factors known to be
expressed by the GW such as Slit2 nor any novel putative
guidance factors and could suggest that DD–PCR is not the
optimal method to use for isolating such genes.
Another group of genes we were interested in identifying
were those that may be involved in patterning the dorso-ventral
axis at the corticoseptal boundary. Our analysis of some of the
isolated genes is continuing but several candidate transcription
factors, including tcf4, sip1, and Nfib, were identified. These
genes demonstrated restricted expression patterns in the dorsal
telencephalon (Fig. 2(B)) and are known to interact with
patterning molecules such as members of the BMP and Wnt
gene families (Verschueren et al., 1999; Kolligs et al., 2002).
Finally, some genes identified may be involved in regulating
the proliferation and differentiation of neuronal and glial
progenitors.Forinstance,notch2isareceptorgenethatishighly
expressedintheventricularneuroepithelium(Fig.2(C))andmay
promote glial cell fate determination (Morrison et al., 2000;
Chambers et al., 2001). Furthermore, glial markers such as
Vimentin were present in the clones derived from the screen
(Fig. 2(D)).
1.3. Expression of novel genes isolated from
the DD–PCR screen
We expected that if a gene was critical for the development
of the corpus callosum, its expression levels would be up-
regulated at a developmental stage when the majority of the
callosal axons cross the midline. We chose several gene
candidates that were most specifically expressed in the GW to
examine their expression throughout brain development using
an RNase Protection Assay (RPA). RPA was used to determine
the mRNA levels of five clones that were previously known
genes (LZIP, EphA4, Calponin, Acinus and Jumonji) and five
clones that did not have significant homologs in the Genbank
database. Total RNA used in the RPA was isolated from mouse
brains at E12, E14, E16, E17 and P0. We found that four of the
five potentially novel genes,and one of the known genes, LZIP,
Fig. 2. In situ hybridization analysis of genes identified by DD-PCR in mouse brain. Bright-field images of film autoradiographs of the clones shown. Coronal
sections of E17 mouse brain were hybridized with35S-labeled antisense probes. A schematic in the lower right corner describes the anatomical location of structures
of interest labeled by the different probes. The key to the schematic is below; CP, cortical plate, VZ/SVZ, subventricular/ventricular zone and GE, ganglionic
eminence. Genes identified by DD-PCR were grouped according to their expression patters. Panel A shows genes highly expressed in the VZ/SVZ. Genes restricted
to the dorsal telencephalon are shown in Panel B, and panel C shows genes limited to VZ/SVZ. Scale bar in AZ1 mm for all sections in (A)–(D).
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481475

Page 6

were up-regulated at E16 (Fig. 3(A)). Semi-quantification of
this data showed that the transcription levels of clones G24F3
and G39F8 increased by 25–30-fold from E12/E14 to E16 and
then remained stable through P0 (Fig. 3(B)). RT-PCR
confirmed the expression of both G24F3 and G39F8 in both
the brain and spinal cord (Fig. 3(C)). G39F8 is expressed
predominantly in brain and at a very low level in the heart, but
not expressed in liver and limb (data not shown).
1.4. Expression of G24F3 in the developing telencephalon,
a new member of the tcf4 gene family
We have shown that the expression levels of G24F3 increase
significantly from E16 through P0, a period during which the
majority of callosal axons cross the midline, suggesting that
up-regulation of these genes may be involved in callosal
formation. To investigate their roles in callosal development,
we performed in situ hybridization to obtain a detailed
expression analysis in the developing and adult brain.
In situ hybridization analysis with the G24F3 probe
demonstrated that G24F3 expression is restricted in the dorsal
telencephalon from E12 through P10 (Fig. 4(A)). G24F3 is first
expressed at high levels in the neuroepithelium as early as at
E12 but is not detectable at E10 (data not shown). At E14, the
highest level of G24F3 expression is in the VZ of the dorsal
telencephalon and in the hippocampus, with lower levels of
expression in ventral regions such as the VZ of ganglionic
eminence (GE). As neurons generated in VZ/SVZ migrate
radially to form the cortical plate, G24F3 mRNA is expressed
in both the VZ/SVZ and the cortical plate at E15 through E17.
At the cortical midline G24F3 is expressed in the GW (Fig. 6).
In situ hybridization analysis on sagittal and horizontal sections
of E17 brains indicated that G24F3 is expressed in a high
caudal to low rostral gradient in the cortex (Fig. 4(A), E17
Fig. 3. Analysis of genes identified by DD-PCR. (A) RNase protection assay showing gene expression levels in whole mouse brain during development from E12 to
P0 for each gene. G39F8, G39F12, G24F3 and LZIP expression levels increased significantly from E16 to P0 (left panel) whereas genes such as G39F3, G7F1,
Acinus,jimonjiandcyclophilinwhereexpressedat similarlevelsthroughoutdevelopment. (B)Semiquantitativeanalysisoftheprotectedbandsshownin panel Afor
G24F3 and G39F8 clones indicates that their transcription levels increased by 25–30-fold from E16 through P0 compared with their expression levels at E12. (C)
Confirmation of G24F3 and G39F8 expressionin the brain from E13 to E17by RT-PCR, showing bothgenes are expressed in brain and spinal cord. (Note that due to
the non-quantitativenature ofthe RT-PCR method,althoughthe band in the E16lane appearslighter,we do notinterpretthis asreflectinga decreasein mRNA levels
at this stage of development). The expected PCR products are 483 bp for G24F3 and 300 bp for G39F8. (D) Northern blot analysis indicated a transcript size of
9.5 kb for both G24F3 and G39F8, and 7 and 4 kb for G2F12.
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481 476

Page 7

sagittal (E17 Sag) and horizontal (E17 Hor) sections). Sagittal
sections also showed a low level of expression in the pons and
the olfactory bulb at E17. G24F3 gene expression decreased in
the cortex of postnatal brains from P0 and thereafter, but high
levels of expression were maintained in the hippocampus and
dentate gyrus. In the adult, G24F3 continued to be expressed
only in the hippocampus and in granule cells of the cerebellum.
Further analysis of the sequence of G24F3 suggested that it
was either a novel gene or a new member of the transcription
factor-4 (tcf4) gene family. Evidence for this is that first, a
BLAST search using the G24F3 fragment sequence as a probe
did not match any cDNA in the Genbank database. However,
analysis of the genomic DNA database indicated that G24F3 is
located on mouse chromosome 18 and overlaps with the gene
encoding tcf4 (Fig. 4(B)). Tcf4 has been reported to be a
downstream target of the Wnt signaling pathway (Kolligs et al.,
2002) and Wnt has been shown to act in the dorso-ventral
patterning (Altmann & Brivanlou, 2001). Second, G24F3 has a
Fig. 4. G24F3expressioninmousebrainduringdevelopment.(A)Bright-fieldimagesoffilmautoradiographsoftheG24F3clone.35S-labeledprobeswerehybridized
to coronal sections (embryonic and/or adult brains); sagittal sections (E17 sag) and horizontal sections (E17 Hor). Expression of G24F3 is restricted to the dorsal
telencephalon of developing brains. High expression levels between E14-P0 are shown in both the dorsal VZ, the cortical plate (CP) and the hippocampus (HP, and
black arrow onsagittaland horizontalsections).G24F3is also expressedin the pons(PN),olfactorybulb(OB) atE17and in the cerebellum(Crb)in adultmouse.(B)
SchematicdemonstratingtheoverlapofG24F3andG2F12withthetcf4genein mousechromosome18.TheG24F3fragmentdoesnotmatchthecDNAsequencesin
theGenbankdatabase,buttheG2F12fragmentisidenticaltothetcf4genecDNA(AccessionNumberBC043050.1).(C)InsituhybridizationanalysiswithG24F3and
G2F12 (tcf4) probes comparing the different expression patterns at E17. A schematic on the right shows the regions of interest on the sections and can be used for
identification of these structures. G24F3 is highly expressed in both the VZ and the cortical plate, whereas G2F12 is highly expressed only in the cortical plate. Both
clones are highly expressed in the hippocampus, but the expression of G2F12 is particularly high in the dentate gyrus/CA3 region compared to G24F3 which is
uniformly expressed throughout the hippocampus (compare regions indicated by the arrows in (C). Scale bars in AZ500 mm (E12); 625 mm (E14), 750 mm (E15),
875 mm (E16), 975 mm (E17 coronal), 2.25 mm (E17 sagittal), 500 mm (P0 and E17 horizontal), 875 mm (P10),1.5 mm (Adult). Scale bar in CZ1 mm.
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481477

Page 8

differentexpressionpatternfromG2F12,afragmentidenticalto
tcf4 cDNA (Accession Number BC043050.1). Third, although
both genes overlap within the tcf4 gene on the chromosome 18,
andarehighlyexpressedinthedorsaltelencephalon(Fig.4(C)),
the expression patterns in the dorsal cortex appeared different.
G24F3 is expressed in both the VZ/SVZ and the cortical plate,
whereas G2F12 is expressed at a high level in the cortical plate
only (Fig. 4(A) and (C)). Furthermore, RT-PCR confirmed
G24F3 expression in the embryonic brains (Fig. 3(C)) and
Northern blot analysis confirmed that G24F3 has a larger
transcript size (9.5 kb) than that of G2F12/tcf4 (7 and 4 kb)
(Fig. 3(D)).
1.5. Expression of the novel gene, G39F8, during
forebrain development
In situ hybridization with the G39F8 probe demonstrated
that in the embryonic mouse brain, G39F8 is expressed in the
telencephalon (Fig. 5(A)). G39F8 expression appeared in the
neuroepithelium of telencephalon as early as E12. From E14
and thereafter, G39F8 is expressed in the VZ of the cortex and
GE. From E16 to P10, the highest level of expression was in
the indusium griseum and hippocampus, particularly the
dentate gyrus. From P10 to adult, G39F8 expression decreases
in the rostral forebrain in all areas except the hippocampus,
dentate gyrus, and external granular layer of the cerebellum
where the expression remains high.
Sequencing analysis showed that the G39F8 fragment
overlapped with the zinc finger protein 288 (zfp288) gene on
mouse chromosome 16 (Fig. 5(B)). However, it did not match
any known cDNAs in the Genbank database. A possibility is
that the G39F8 fragment may represent part of the 3prime;-
untranslated region (UTR) of the zfp288 gene. To test this
hypothesis, we screened a rat P1 brain cDNA library with the
G39F8 fragment as probe and obtained a 4.7 kb clone (data not
shown). A search of the Genbank and Genome databases
Fig. 5. The G39F8 clone is highly expressed in the dorsal telencephalon of developing brain. Bright-field images of film autoradiographs of the G39F8 clone.35S-
labeled probes were hybridized to coronal sections (embryonic and/or adult brains), sagittal sections (E17 sag), and horizontal sections (E17 Hor). The schematic on
the right demonstrates the location of structures of particular interest. G39F8 is expressed in the VZ/SVZ, ganglionic eminence (GE), olfactory bulb (OB), midline
cell populations including the IG, the hippocampus (black arrow) and the striatum (Str; shown in the E17 horizontal section). In adult brains G39F8 is expressed in
the hippocampus (arrow) and cerebellum (Crb). (B) Schematic demonstrating the overlap of the G39F8 fragment and the G39F8 library clone with the zfp288 gene
on mouse chromosome 16. (C) The G39F8 fragment does not match any cDNA sequences in the Genbank database, but the G39F8 library clone (4.7 kb) is similar to
clones in the Riken full-length enriched library (Accession number: AK036125, AK038731 and AK083191, respectively). Shown are overlaps in sequence
homology between the G39F8 library clone and each clone from the Riken library (listed above). Scale bars in AZ625 mm (E14), 750 mm (E15), 875 mm (E16),
975 mm (E17 coronal), 2.25 mm (E17 sagittal), 500 mm (P0 and E17 horizontal), 875 mm (P10),1.5 mm (Adult).
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481478

Page 9

indicated that this 4.7 kb clone is similar to clones in the
RIKEN full-length enriched library (Accession numbers
ak038731, ak083191, ak036125) (Fig. 5(C)), does not match
completely any member of the zinc finger gene family,
but overlaps with the zfp288 gene on mouse chromosome 16
(Fig. 5(B)). This evidence suggests that G39F8 is a
new member or a new isoform of this zinc finger protein
family.
Since our primary purpose of the screen was to identify
genes expressed by the glial wedge we wanted to confirm the
expression of G24F3, G39F12 and LZIP in the glial wedge
(Fig. 6(A)–(G)). To do this we compared the expression of
these genes with the known glial markers, glial fibrillary acidic
protein (GFAP; Fig. 6(I))) and vimentin (Fig. 6(H)) expressed
by the glial wedge. We examined the expression of G39F12
and G24F3 at E16, E17 and P0 and found that G39F12 was
expressed only in the dorsal region of the VZ/glial wedge
region whereas G24F3 and LZIP were expressed throughout
the VZ/glial wedge region. LZIP is a transcription factor that
has been shown to bind the chemokine receptor CCR1 and
modulate its response to the chemokine leukotactin-1 (Ko,
Jang, Kim, Kim, Sung and Kim, 2004). It is generally
ubiquitously expressed and thus its role, if any, in glial
development or callosal axon guidance is unclear. Finally,
G24F3 was highly expressed within the indusium griseum
throughout development (Fig. 6(D)–(F); IG). Both neurons and
glia reside within this region and thus at present we do not
know whether this expression is cell type specific or expressed
by both populations.
In conclusion, genes isolated from the GW by DD-PCR may
function in brain developmental processes such as axon
guidance, dorsal patterning and neurogenesis/differentiation
of neural progenitors. Our data suggested that clone G24F3 is
either a novel gene or a new member of the transcription factor-
4 (tcf4) gene family which may be involved in the dorsal
patterning of the telencephalon, and that clone G39F8 is a new
Fig. 6. Developmental expression of the G39F12, G24F3, LZIP and Vimentin clones obtained from the DD–PCR screen. Coronal sections were labeled by in situ
hybridization with35S-labeled probes of either the G39F12 (A to C, from E16 to P0, respectively) or G24F3 clones (D to F, from E16 to P0 coronal sections,
respectively). Shown are high power views of sections at the corticoseptal boundary to demonstrate expression within the region of the midline glial populations.
Sections were dipped in radiographic emulsion and then developed to display the expression patterns. For orientation and comparison purposes, panel I shows a
similar section stained with GFAP to label the midline glial populations (the indusium griseum glia (IGG) and the glial wedge (GW)). Both clones were highly
expressed in the IGG and GW with clone G24F3 being most highly expressed in the indusium griseum (IG, E and F). Panels G and H are radioactive in situ
hybridizations at E17 showing the expression of LZIP and Vimentin, respectively. Sections A to H were counterstained with cresyl violet. Scale bar in IZ200 mm in
all panels.
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481479

Page 10

member or an isoform of zinc finger protein 288 (zfp288) gene
family. Further analysis is required to identify the functions of
these novel isoforms in embryonic brain development.
2. Experimental procedures
2.1. Animals
C57BL/6J mice (The Jackson Laboratory, Bar Harbor, ME) were bred on
site at The University of Maryland animal facility under the care and approval
of the accredited University of Maryland, Baltimore Institutional Animal Care
and Use Committee. The date of vaginal plug detection is designated as
embryonic day (E) 0 and the date of birth as postnatal day (P) 0.
2.2. RNA Isolation
Living E17 C57BL/6J mouse embryonic brains were blocked and
embedded in 3% low melting point agar (Sea plaque; FMC Bioproducts,
Rockland, ME) and vibratome-sectioned at 350 mm. Glial wedge, cortex and
septum were dissected from the regions shown in Fig. 1(A). Three independent
samplesforeach tissue werepreparedfromE17brains.Total RNAwasisolated
using an RNeasy mini kit (Qiagen), treated with RNase-free DNase I (Gibco
BRL) and re-purified over an RNeasy column (Qiagen).
2.3. Differential display PCR
Differential display PCR (DD–PCR) was carried out following the
procedure as described (Gesemann et al., 2001). Briefly, reverse transcription
reactions were performed using the Superscript reverse-transcriptase II (Gibco
BRL) and 10 base pair primers of arbitrary sequence (Operon Technologies).
cDNA from this reaction was re-purified and used as templates for DD–PCR.
32P-labelled PCR products were separated on 6% denaturing polyacrylamide
gels. Gels were exposed to X-ray film overnight. Differences in the intensity of
the bands were determined by visual inspection of films and they were aligned
with the gels so that the differentially expressed bands could be excised. To
minimize false positives, each reaction was performed in triplicate from three
separate preparations of total RNA. Differentially expressed DNA fragments
were re-amplified and cloned into pBluescript vectors. Clones containing the
correct inserts were sequenced and compared with sequences in the NIH
Genbank Database. DD–PCR fragments were designated as G##F## (G##
indicated the polyacrylamide gel number and F## the excised fragment
number).
2.4. In situ hybridization
35S-UTP labeled DD–PCR fragments were used as probes for in situ
hybridization. Probe labeling performed as previously described (Shu et al.,
2003c). Briefly, 20 mm cryostat sections of embryonic mouse brains were
mounted on cold RNase-free gelatin-coated microscope slides and prefixed in
4% paraformaldehyde,treated with proteinaseK (Ambion,20 mg/ml), re-fixed,
and acetylated in 100 mM triethanolamine with acetic anhydride (TEA), then
dehydrated through a graded series of ethanols.
The probes were diluted with hybridization buffer which contains 50%
deionized formamide, 10% dextran sulfate, 1!Denhardts solution, 0.3 M
NaCl, 10 mM Trirs pH7.5, 10 mM sodium phosphate pH6.8, 5 mM EDTA,
25 mM DTT and 50 mM b-mercaptoethanol (b-ME) to a concentration of
50,000 cpm/ul. Sections were hybridized with 80–90 ml of the denatured
probe mixture in a humidified box and incubated overnight at 55 8C. Sections
were washed at 55 8C for 30 min in 5!SSC and 20 mM b-ME, 45 minutes at
65 8C and 2–3 h at 37 8C in 2!SSC, 50% formamide, 20 mM b-ME, 15 min
at 37 8C in 0.5 M NaCl in TE. Sections were treated with 20 mg/ml of RNase
A for 15 min at 37 8C, and washed for 5 min each in 2X then 0.2X SSC at
room temperature. Slides were dehydrated through a graded series of
ethanols. Air-dried slides were then exposed to X-ray film (Kodak) for
5 days. Following this sections were dipped in autoradiographic emulsion
(NTB2, Kodak), air-dried and stored in light-proof boxes at room
temperature for 4 weeks before being developed in D-19 and fixer
(Kodak). After developping, sections were counterstained with cresyl violet,
and dehydrated through a series of alcohols before being coverslipped with
DPX (Electron Microscopy Sciences) mounting medium.
2.5. Immunohistochemistry
Brains were embedded in 3% agarose (Noble Agar, Difco, Detroit, MI)
and all sections were cut at 45 mm on a Vibratome (Leica, Deerfield, IL).
Immunostaing was performed as previously described (Shu et al., 2003c).
Briefly sections were blocked in 0.2% Triton X-100 (Sigma) and 2% normal
goat serum (Vector Laboratories, Burlingame, CA) in PBS for 2 h. The
sections were then incubated overnight in rabbit anti-glial fibrillary acidic
protein (GFAP, 1/30K; Dako, Carpinteria, CA). Sections were washed with
PBS and incubated for 1 h with secondary antibody (biotinylated goat anti-
rabbit, 1/500; Vector Laboratories). The signal was amplifiyed using AB
solution (Vectastain elite kit, Vector Laboratories) for 1 h. Then staining was
revealed using Nickel DAB (3,3-diaminobenzidine tetrahydrochloride;
Sigma). Sections were mounted on 2% gelatin-coated glass slides, dehydrated
through a series of alcohols and coverslipped with DPX (Electron
Microscopy Sciences) mounting medium.
2.6. RNase protection assay
Plasmid DNAs containing DD–PCR fragments were linearized. Radio-
labelled antisense riboprobes were transcribed using T7 or T3 RNA
polymerases (Promega), and [a-32P]UTP (ICN, 800 and 10 mCi/ml). The
labeled probes had a specific activity of 7!107cpm/mg or higher. As an
internal control, an antisense riboprobe specific for cyclophilin was also
prepared. Approximately, 1.5!105cpm of labeled probe was hybridized to
10 mg of total RNA isolated from E12, E14, E16, E17 and P0 mouse brains.
Non-hybridized RNA was removed using RNase A (Roche, 10 mg/ml) and
RNase T1 (Gibco BRL, 1359 U/ml), and samples were then treated with
proteinaseK(Ambion,20 mg/ml).The protectedhybridswere separatedon 6%
denaturing polyacrylamide gels. The density of the cyclophilin signal was
roughly equivalent in all of the samples, suggesting that cyclophilin expression
is similar across these tissues. Semi-quantitative analysis was performed by
using QuantiScan software on RNase protection assay (RPA) bands (Biosoft).
2.7. RT-PCR
All PCR reagents were purchased from Applied Biosystems (ABI, Foster
City, USA).
Total RNA was isolated from the telencephalon of embryonic brains
with Trizol reagent (Invitrogen), and reverse transcribed into cDNA with
Superscript II (Invitrogen). Reverse transcription without Superscript II was
used as a negative control. The PCR conditions were as follows: initial
activation at 94 8C for 2 min; denaturation at 94 8C for 30 s, annealing at 53 8C
for 1 min, and extension at 72 8C for 1 min 30 cycles; followed by extension at
72 8C for 10 min and then a hold cycle at 4 8C. Primers for PCR were designed
based on the DD–PCR fragments. PCR primers for G24F3 were: forward
CCCTACACAACAACAGTCTGC, reverse GGGTCAAACCTTTGACAGC,
with an expected PCR product of 483 bp. Primers for G39F8 were: forward
TCCTAGCTTTGTGATTGG, reverse 1: TGCTAACATCACGTCAGG, with
an expected PCR product of 300 bp; reverse 2: GCTAAGGGAA-
CAAAAAGG,with an expected PCR product of 254 bp.
References
Altmann, C.R., Brivanlou, A.H., 2001. Neural patterning in the vertebrate
embryo. Int. Rev. Cytol. 203, 447–482.
Chambers, C.B., Peng, Y., Nguyen, H., Gaiano, N., Fishell, G., Nye, J.S., 2001.
Spatiotemporal selectivity of response to Notch1 signals in mammalian
forebrain precursors. Development 128, 689–702.
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481480

Page 11

Chaudhry, A.Z., Lyons, G.E., Gronostajski, R.M., 1997. Expression patterns of
the four nuclear factor I genes during mouse embryogenesis indicate a
potential role in development. Dev. Dyn. 208, 313–325.
Coonan, J.R., Greferath, U., Messenger, J., Hartley, L., Murphy, M.,
Boyd, A.W., et al., 2001. Development and reorganization of corticospinal
projections in EphA4 deficient mice. J. Comp. Neurol. 436, 248–262.
Demyanenko, G.P., Tsai, A.Y., Maness, P.F., 1999. Abnormalities in neuronal
process extension, hippocampal development, and the ventricular system of
L1 knockout mice. J. Neurosci. 19, 4907–4920.
Gesemann, M., Litwack, E.D., Yee, K.T., Christen, U., O’Leary, D.D., 2001.
Identification of candidate genes for controlling development of the basilar
pons by differential display PCR. Mol. Cell. Neurosci. 18, 1–12.
Goldshmit, Y., Galea, M.P., Wise, G., Bartlett, P.F., Turnley, A.M., 2004.
Axonal regeneration and lack of astrocytic gliosis in EphA4-deficient mice.
J. Neurosci. 24, 10064–10073.
Gronostajski, R.M., 2000. Roles of the NFI/CTF gene family in transcription
and development. Gene 249, 31–45.
Ko, J., Jang, S.-W., Kim, Y.-S., Kim, I.-S., Sung, H.J., Kim, H.-H., et al., 2004.
Human LZIP binds to CCR1 and differentially affects the chemotactic
activities of CCR1-dependent chemokines. Fed. Am. Soc. Exp. Biol. J. 18
(7), 890–892. doi:10.1096/fj.03-0867fje (full length article).
Kolligs, F.T., Nieman, M.T., Winer, I., Hu, G., Van Mater, D., Feng, Y., et al.,
2002. ITF-2, a downstream target of the Wnt/TCF pathway, is activated in
human cancers with beta-catenin defects and promotes neoplastic
transformation. Cancer Cell 1, 145–155.
Kullander, K., Croll, S.D., Zimmer, M., Pan, L., McClain, J., Hughes, V., et al.,
2001. Ephrin-B3 is the midline barrier that prevents corticospinal tract
axons from recrossing, allowing for unilateral motor control. Genes Dev.
15, 877–888.
Meiners,S.,Mercado,M.L.,Nur-e-Kamal, M.S.,Geller, H.M.,1999. Tenascin-
C contains domains that independently regulate neurite outgrowth and
neurite guidance. J. Neurosci. 19, 8443–8453.
Morrison, S.J., Perez, S.E., Qiao, Z., Verdi, J.M., Hicks, C., Weinmaster, G.,
et al., 2000. Transient notch activation initiates an irreversible switch from
neurogenesis to gliogenesis by neural crest stem cells. Cell 101, 499–510.
Ozaki, H.S., Wahlsten, D., 1992. Prenatal formation of the normal mouse
corpus callosum: a quantitative study with carbocyanine dyes. J. Comp.
Neurol. 323, 81–90.
Rash, B.G., Richards, L.J., 2001. A role for cingulate pioneering axons in the
development of the corpus callosum. J. Comp. Neurol. 434, 147–157.
Scotland, P., Zhou, D., Benveniste, H., Bennett, V., 1998. Nervous system
defects of AnkyrinB (K/K) mice suggest functional overlap between the
cell adhesion molecule L1 and 440-kD AnkyrinB in premyelinated axons.
J. Cell. Biol. 143, 1305–1315.
Shen, Y., Mani, S., Donovan, S.L., Schwob, J.E., Meiri, K.F., 2002. Growth-
associated protein-43 is required for commissural axon guidance in the
developing vertebrate nervous system. J. Neurosci. 22 (1), 239–247.
Shu, T., Richards, L.J., 2001. Cortical axon guidance by the glial wedge during
the development of the corpus callosum. J. Neurosci. 21, 2749–2758.
Shu, T., Puche, A.C., Richards, L.J., 2003a. Development of midline glial
populations at the corticoseptal boundary. J. Neurobiol. 57, 81–94.
Shu, T., Sundaresan, V., McCarthy, M.M., Richards, L.J., et al., 2003b. Slit2
guides both precrossing and postcrossing callosal axons at the midline
in vivo. J. Neurosci. 23, 8176–8184.
Shu, T., Butz, K.G., Plachez, C., Gronostajski, R.M., Richards, L.J., 2003c.
Abnormal development of forebrain midline glia and commissural
projections in Nfia knock-out mice. J. Neurosci. 23, 203–212.
Silver, J., Lorenz, S., Wahlsten, D., Coughlin, J., 1982. Axonal guidance during
development of the great cerebral commissures: descriptive and exper-
imental studies, in vivo, on the role of performed glial pathways. J. Comp.
Neurol. 210, 10–29.
Silver, J., Edwards, M.A., Levitt, P., 1993. Immunocytochemical demon-
stration of early appearing astroglial structures that form boundaries and
pathways along axon tracts in the fetal brain. J. Comp. Neurol. 328, 415–
436.
Steele-Perkins, G., Plachez, C., Butz, K.G., Yang, G., Bachurski, C.J.,
Kinsman, S.L., et al., 2005. The transcription factor gene Nfib is essential
for both lung maturation and brain development. Mol. Cell. Biol. 25, 685–
698.
Takemoto, M., Fukuda, T., Sonoda, R., Murakami, F., Tanaka, H.,
Yamamoto, N., 2002. Ephrin-B3-EphA4 interactions regulate the growth
of specific thalamocortical axon populations in vitro. Eur. J. Neurosci. 16,
1168–1172.
Verschueren, K., Remacle, J.E., Collart, C., Kraft, H., Baker, B.S.,
Tylzanowski, P., et al., 1999. SIP1, a novel zinc finger/homeodomain
repressor, interacts with Smad proteins and binds to 50-CACCT sequences
in candidate target genes. J. Biol. Chem. 274, 20489–20498.
W.-B. Shen et al. / Gene Expression Patterns 6 (2006) 471–481481