ArticlePDF Available

Cloning and characterization of NE-dlg: A novel human homolog of the Drosophila discs large (dlg) tumor suppressor protein interacts with the APC protein


Abstract and Figures

We have cloned a cDNA for a novel human homolog of the Drosophila discs large (dig) tumor suppressor protein, termed NE-dlg (neuronal and endocrine dig). Northern blot analysis revealed that the gene is highly expressed in neuronal and endocrine tissues. Fluorescence in situ hybridization (FISH) and radiation hybrid mapping studies localized the NE-dlg gene to chromosome Xq13. We also found that the NE-dlg gene encoded a 100 kDa protein. Immunolocalization studies using an NE-dlg antibody showed that the protein tended to be expressed in non-proliferating cells, such as neurons, cells in Langerhans islets of the pancreas, myocytes of the heart muscles, and the prickle and functional layer cells of the esophageal epithelium. Proliferative cells, including various cultured cancer cell lines and basal cells in the esophageal epithelium, showed little expression of the NE-dlg protein. In addition, yeast two-hybrid screening and in vitro binding assays revealed that the NE-dlg interacted with the carboxyl-terminal region of the APC tumor suppressor protein. These data suggest that NE-dlg negatively regulates cell proliferation through its interaction with the APC protein.
Content may be subject to copyright.
Cloning and characterization of NE-dlg: a novel human homolog of the
Drosophila discs large (dlg) tumor suppressor protein interacts with the
APC protein
Keishi Makino
, Hiroaki Kuwahara
, Norio Masuko
, Yasuyuki Nishiyama
Tetsurou Morisaki
, Ji-ichiro Sasaki
, Mitsuyoshi Nakao
, Akira Kuwano
, Motomi Nakata
Yukitaka Ushio
and Hideyuki Saya
Department of Tumor Genetics and Biology;
Department of Neurosurgery, Kumamoto University School of Medicine, 2-2-1
Honjo Kumamoto 860;
Department of Hygiene Ehime, University School of Medicine, Ehime 791-02 and
Biomedical R&D
Department, Sumitomo Electric Industries, Ltd, Yokohama 244, Japan
We have cloned a cDNA for a novel human homolog of
the Drosophila discs large (dlg) tumor suppressor
protein, termed NE-dlg (neuronal and endocrine dlg).
Northern blot analysis revealed that the gene is highly
expressed in neuronal and endocrine tissues. Fluorescence
in situ hybridization (FISH) and radiation hybrid
mapping studies localized the NE-dlg gene to chromo-
some Xq13. We also found that the NE-dlg gene encoded
a 100 kDa protein. Immunolocalization studies using an
NE-dlg antibody showed that the protein tended to be
expressed in non-proliferating cells, such as neurons,
cells in Langerhans islets of the pancreas, myocytes of
the heart muscles, and the prickle and functional layer
cells of the esophageal epithelium. Proliferative cells,
including various cultured cancer cell lines and basal cells
in the esophageal epithelium, showed little expression of
the NE-dlg protein. In addition, yeast two-hybrid
screening and in vitro binding assays revealed that the
NE-dlg interacted with the carboxyl-terminal region of
the APC tumor suppressor protein. These data suggest
that NE-dlg negatively regulates cell proliferation
through its interaction with the APC protein.
Keywords: tumor suppressor genes; two-hybrid screen-
ing; chromosome Xq13; EST
About 50 tumor suppressor genes in Drosophila have
been identi®ed through genetic analysis. Germline
mutation in such genes behave as recessive lethals
(Gate et al., 1989). Germline mutations of the lethal
(1) discs large 1 (dlg) locus result in imaginal disc
neoplasia and a prolonged larval period followed by
death (Woods et al., 1989). The dlg protein includes
three copies of the DHR (discs-large homologous
region) domain, a SH3 motif, and a domain that is
homologous to the yeast guanylate kinase (GUK)
(Stehle et al., 1992). These structures suggest that dlg
plays a role in intracellular signal transduction, but
there has not yet been any demonstration of its
Recently, several genes which encode for polypep-
tides homologous to the Drosophila dlg have been
identi®ed in mammalian cells and categorized in a new
protein family termed MAGUK (membrane-associated
guanylate kinase homolog) (Woods et al., 1993). This
protein family consists of a human homolog of dlg
(hdlg-1) which is homologous to rat SAP-97 (Lue et
al., 1994; Muller et al., 1995), the rat synaptic protein
PSD-95/SAP-90 (Cho et al., 1992; Kistner et al., 1993),
the tight junction proteins ZO-1 (Willott et al., 1993)
and ZO-2 (Jesaitis et al., 1994; Itoh et al., 1993), and
human erythroid p55 (Ru et al., 1991). These
MAGUK proteins have been shown to localize at the
membrane-cytoskeleton interface, and some recently
have been reported to bind to membrane associated
proteins: hdlg-1/SAP97 interacts with the cytoskeletal
protein 4.1 (Lue et al., 1994) and the adenomatous
polyposis coli gene (APC) product (Matsumine et al.,
1996), and PSD-95/SAP90 binds to both NMDA
receptors (Kornau et al., 1995; Niethammer et al.,
1996) and Shaker-type potassium channels (Kim et al.,
1995). Therefore, it is reasonable to speculate that
MAGUK proteins have important roles in cellular
morphogenesis, adhesion, growth and regulation of ion
density by interacting with various key molecules.
In this study, we isolated a new human homolog of
the Drosophila dlg gene, which we designated NE-dlg,
by combination of computer analysis of the expressed
sequenced tag (EST) databases and two-step PCR
procedure. We report here the cDNA sequence and
chromosome location of this gene, and immunolocali-
zation of the encoded protein using an antibody raised
against a unique protein sequence in NE-dlg.
Furthermore, we found that NE-dlg interacted with
the APC tumor suppressor protein. The biological
signi®cance of this gene will be discussed.
Identi®cation of the human gene related to Drosophila
dlg and hdlg-1/SAP-97
To identify the human homolog of the Drosophila dlg,
we performed sequence database searching using the
Blast algorithm. A tblastn search of the Genbank
database using the dlg and hdlg/SAP-97 peptide
sequences revealed that seven EST clones with the
accession numbers T31002, M78186, R16282, F08659,
Correspondence: H Saya
Received 21 November 1996; revised 5 February 1997; accepted 5
February 1997
Oncogene (1997) 14, 2425 ± 2433
1997 Stockton Press All rights reserved 0950± 9232/97 $12.00
F07846, T71695, and T31204 have sequence similarity.
Four of these (T31002, M78186, F08659, T31204) were
found to be identical to the PSD-95/SAP-90A gene,
which had been previously identi®ed (Cho et al., 1992;
Kistner et al., 1993). Two (F07846, T71695) were parts
of the p55 gene (Ru et al., 1991). The sequence of the
R16282 EST clone was not found to be identical to any
previously reported gene. A 221-bp of R16282 cDNA
was ampli®ed by PCR using P1 and P2 primers and a
human fetal brain cDNA library. We performed two-
step PCR on the ampli®ed cDNA sequence bases to
clone the full length cDNA (Figure 1).
We identi®ed a 3100-bp cDNA which contains a
single large open reading frame encoding a polypeptide
of 817 amino acids (Figure 2). The nucleotide sequence
was con®rmed by sequencing several clones that were
generated by independent PCR in order to avoid PCR
artifacts. Any dierences in the nucleotide sequence of
the various clones were not found. The cDNA was
termed NE-dlg. The domain organization characteristic
of the other MAGUK family members was also
conserved in the NE-dlg polypeptide, with three
DHR segments in the N-terminal half of the protein,
a central SH3 motif, and a C-terminal GUK domain.
Moreover, NE-dlg was found to contain a pair of
XPPXY motifs in the amino-terminal region, which
shows a signi®cant level of speci®city in binding to a
subset of the WW domain that is a structured protein
module found in a wide range of regulatory,
cytoskeletal, and signaling molecules (Sudol et al.,
1995; Williamson, 1994). This protein showed high
similarity to hdlg-1/SAP-97 and Drosophila dlg with
75% and 60% identity, respectively. The three DHR
segments have 66% identity to analogous segments in
hdlg-1/SAP-97 and 73% to those of rat PSD-95. The
SH3 domain was the most similar region between NE-
dlg and hdlg-1/rat SAP-97 (81% identity). The GUK
domain was 39% identical to the protein sequence of
yeast GUK and 80% identical to analogous segments
in hdlg-1/SAP-97 (Figure 3).
Figure 1 Cloning strategy of the NE-dlg gene. PCR using P1 and
P2 primers was performed to amplify a core cDNA fragment
(221 bp) from a human fetal brain cDNA library. P3 and P4
primers were designed based on the sequence information of the
core cDNA. Ampli®cations of the 5'region (region A) and 3'
region (region C) were carried out by the two-step PCR procedure
using the fetal brain cDNA library. For ampli®cation of the
remaining 5'region (region B), P5 primer was used for the ®rst
PCR, and P6 and T3 primers were used for the second PCR.
Region A (1.1 kbp) spanned the DHR2 domain to the SH3 motif,
region B (1.2 kbp) included the initiating methionine site, and
region C (1.4 kbp) spanned the SH3 motif to the C terminal
noncoding sequence inclusive of the stop codon
Figure 3 Comparison of sequence identities in dlg-related
proteins. Protein sequences were compared by Multiple Sequence
Alignments at BCM Search Launcher (http://dot.imgen.bcm.tmc.
edu:9331/multi-align/multi-align.html). Identical residues are
indicated by an asterisk. Domains of DHR1-3, SH3 and GUK
are indicated by underlines. The NE-dlg protein showed high
similarity to PSD-95/SAP-90 and hdlg-1/SAP-97 with 80% and
75% identity, and moderate similarity to Drosophila dlg, with
60% identity, respectively
Figure 2 Predicted amino acid sequence of the NE-dlg. On the
basis of the deduced amino acid sequence, NE-dlg contains 817
amino acids. Homologous domains and motifs are indicated as
follows: DHR1 (ÐÐ), DHR2 (............), DHR3 (±-±), SH3
(............), GUK (±--±--±) and PY motif (boxes). The cDNA
sequence has been submitted to the GenBank database (accession
no. U49089)
Cloning and characterization of NE-dlg
KMakinoet al
Expression of NE-dlg mRNA in human tissues
To determine the expression pattern of NE-dlg mRNA,
we performed Northern blot analysis on various human
tissues. NE-dlg transcripts were detected in brain,
pancreas, thyroid, trachea and prostate (Figure 4a).
Since this gene was highly expressed in neuronal and
endocrine tissues, we designated it as NE-dlg.NE-dlg was
expressed in a tissue speci®c manner, whereas the hdlg-1/
SAP-97 was expressed in most of the adult tissues
examined. Brain was found to express two species (6.7
and 5.5 kb) of NE-dlg mRNA, and the other tissues
expressed two species of mRNA of 5.5 and 4.0 kb.
Although the size of the transcript in brain is dierent
from that in other tissues, a 2453 bp full-length ORF
cDNA fragment of NE-dlg gene could be ampli®ed from
RNAs extracted from thyroid and skeletal muscle as well
as from brain by RT ± PCR using NED forward and
NED reverse primers (Figure 4b).
Chromosomal localization of NE-dlg
To identify the chromosomal localization of the NE-dlg
gene, we performed ¯uorescence in situ hybridization
(FISH) and PCR mapping using somatic cell hybrid
analysis. In the FISH analysis (Figure 5a), 15 of 50
metaphases examined showed symmetrical double
spots on a single chromosome Xq13. All double spots
and 75% of the single spots detected were on
chromosome Xq13. No spots were found on the Y
chromosome. We therefore assigned the NE-dlg gene to
chromosome Xq13. Localization was further con®rmed
in sequential G-banded chromosomes. In somatic cell
hybrid analysis, PCR products of the expected sizes
(306 bp) were ampli®ed using S1 and AS1 primers
from seven of 24 hybrid cell lines in the monochro-
mosomal somatic cell hybrid DNA panel, and from 16
out of 93 hybrid cell lines in the Genebridge 4
radiation hybrid panel. Comparison with the human
chromosomal content of the hybrids, as determined by
the manufacturer and the Whitehead Institute/MIT
Center for Genome Research, localized the NE-dlg
gene to chromosome X and placed it 15.1 cR from
GATA64D08 (Figure 5b).
Detection of the NE-dlg protein
First, Western blot analysis of COS-7 cells, which were
transiently transfected by the HA-tagged NE-dlg
expression plasmid (pCGN-NE-dlg), was performed
using anti-HA and anti-NE-dlg antibodies. A band of
about 100 kDa was detected in lysates of pCGN-NE-
7.5 —
4.4 —
7.5 —
4.4 —
2.4 —
1.35 —
3054 —
1018 —
1112 1314
11 12 1314
1112 13 14
Figure 4 Expression of the NE-dlg transcript in various human tissues. (a) Northern blot analysis of the NE-dlg and hdlg-1/SAP-97
genes. Each lane contains 2 mg of human poly(A)
RNA. Lanes 1 ± 14: heart, brain, placenta, lung, liver, skeletal muscle, kidney,
pancreas, stomach, thyroid, spinal cord, lymph node, trachea and prostate. The blots were hybridized with a speci®c cDNA probe
of NE-dlg,hdlg-1 and a human b-actin. The size marker (kb) is indicated. (b) A 2453-bp fragment which corresponds to a full length
open reading frame of NE-dlg cDNA was ampli®ed from RNAs extracted from thyroid (lane 3) and skeletal muscle (lane 4) as well
as brain (lane 2) by RT ± PCR. The PCR was performed by using a set of primers (NED forward and NED reverse). The size
marker is indicated in lane 1
Figure 5 Chromosomal localization of the NE-dlg gene. (a)
Fluorescence in situ hybridization (FISH). Partial metaphases of
normal human chromosomes were hybridized with the NE-dlg
cDNA probe counterstained with propidium iodide. Hybridiza-
tion site is indicated by white arrow. (b) Radiation hybrid
mapping. The results of PCR using the radiation hybrid panel
were sent to the Whitehead Institute/MIT Center for Genome
Research via WWW and placed the NE-dlg gene at 15.1 cR from
GATA64D08 on chromosome Xq13
Cloning and characterization of NE-dlg
KMakinoet al
dlg transfected COS-7 cells with both anti-HA and
anti-NE-dlg antibodies (Figure 6a), which was very
similar to the calculated molecular mass of the full-
length protein encoded by the NE-dlg cDNA. By using
the NE-dlg antibody, we performed Western blot
analysis to detect the endogenous NE-dlg protein in
human and rat brain tissues and rat esophagus tissues.
The 100 kDa protein was detected in lysates from
human and rat brain tissues and rat esophagus tissues
(Figure 6a). Most of the NE-dlg protein expressed in
the transfected COS-7 cells was found in the 1% Triton
X-100 insoluble fraction but not in the soluble fraction
(Figure 6b, lane 1 and 2). However, the endogenous
protein was found in both the Triton X-100 soluble
and insoluble fractions (Figure 6b, lane 3 and 4).
Moreover, we performed Western blot analysis to
detect the endogenous NE-dlg protein in various
cancer cell lines as shown in Figure 6c. The
endogenous NE-dlg protein of 100 kDa was not
detected in the lysates from any cell lines examined
whereas b-tubulin was found in all samples.
Immunohistochemical analysis of NE-dlg in various
tissue sections
To determine the tissue distribution of NE-dlg protein,
immunohistochemical analysis using the anti-NE-dlg
antibody was performed in various frozen and
formalin-®xed paran-embedded tissue sections. Since
Northern blot analysis revealed speci®c expression of
the NE-dlg transcript in brain and endocrine tissues, we
®rst examined the protein expression in those tissues.
In brain, immunoperoxidase staining was seen along
the neuronal axon or dendrites in the molecular and
the outer granular layers of the cerebral cortex (Figure
7a). In pancreas, about half the cells in the Langerhans
islets demonstrated dense cytoplasmic staining (Figure
7b). However, double immunostaining using an anti-
insulin antibody with the anti-NE-dlg antibody
suggested that the distribution of NE-dlg positive
cells was not the same as that of band acells (data
not shown). We also examined protein expression in
various non-neuronal tissues, including thyroid, gastro-
intestinal tissues, skeletal muscle, liver, kidney, spleen
and lung. NE-dlg was highly expressed in the
cytoplasmic region of myocytes of the heart muscle
(Figure 7c), but it was hardly detected in smooth
muscle (data not shown). Cytoplasm of follicular cells
in the thyroid were moderately stained (data not
shown). Moreover, in strati®ed squamous epithelium,
such as the esophagus (Figure 7d) and tongue, the sites
of cell ± cell contact between epithelial cells in the
prickle and functional layers, which consist of non-
proliferating cells, were strongly stained. However, the
basal cell layer, which consisted of proliferative cells,
showed no staining.
Detection of an interaction between NE-dlg and APC
To learn about the role of the NE-dlg protein in cells,
we used the yeast two-hybrid system for identifying
proteins that bind to the NE-dlg protein. An
expression vector was constructed by fusing the
GAL4 DNA-binding domain to the three DHR
regions and SH3 motif of the NE-dlg protein. This
bait plasmid was cotransformed in yeast with a prey
plasmid containing a human fetal brain cDNA
expression library fused to the GAL4 activation
domain. We used a system that utilizes GAL4
recognition sites to regulate expression of both HIS3
and LacZ (Durfee et al., 1993). Colonies were selected
that grew on yeast drop-out media lacking Leu, Trp,
and His but contained 50 mM3-amino-1, 2, 4, triazole
and that were blue when assayed by X-gal ®lter assay.
Ten positive clones were obtained from 5610
transformants screened. To con®rm the speci®city of
the interactions, the plasmids expressing the activation
domain fusion proteins were recovered from the
positive clones and cotransformed with GAL4bd-NE-
137 —
79 —
Ab α-NE-dlg
137 —
79 —
12 34
α-NE-dlg Ab
1234 567
α-NE-dlg Ab
α-β-Tubulin Ab
79 —
79 —
42.9 —
Figure 6 Western blot analysis of NE-dlg. (a) Lysates prepared
from COS-7 cells, which were transfected with pCGN-NE-dlg
(lanes 1 and 3) and with pCGN (lanes 2 and 4), were detected by
anti-HA antibody (lanes 1 and 2) and by anti-NE-dlg antibody
(lanes 3 and 4). Lysates prepared from human brain (lane 5), rat
brain (lane 6) and rat esophagus (lane 7) were detected by anti-
NE-dlg antibody. The arrow indicates the 100-kDa full length
NE-dlg protein. (b) Proteins from Triton X-100 soluble fractions
(lane 1 and 3) and the insoluble fractions (lane 2 and 4) of the
pCGN-NE-dlg transfected COS-7 cells (lane 1 and 2) and rat
brain (lane 3 and 4) were detected by anti-NE-dlg antibody. (c)
Lysates prepared from rat brain (lane 1), various cancer cell lines
(HeLa, U251, ASPC1, SY5Y and HT1080: lane 2 ± 6) and pCGN-
NE-dlg transfected COS-7 cells (lane 7) were detected by anti-NE-
dlg antibody (upper panel) and by anti-b-tubulin antibody (lower
Cloning and characterization of NE-dlg
KMakinoet al
dlg and control baits (Figure 8a). Three clones were
positive in the secondary screening, and all of them
contained a 1200 bp insert whose sequence was
identical to a part of the adenomatous polyposis coli
gene (APC) cDNA sequence. The insert encoded the
carboxyl terminal 111-amino acid fragment of the APC
tumor suppressor protein (Figure 8b). To con®rm the
interaction between NE-dlg and APC observed in
yeast, a fusion protein that contained GST and the
three DHR regions and SH3 motif of NE-dlg was
expressed in bacteria using the pGEX-2TH expression
vector system. Whole cell lysates of human fibrosarco-
ma HT1080 cells were incubated with either GST or
the GST-NE-dlg bound GSH-beads. The binding
protein complex was analysed by Western blot using
anti-APC antibody, and it showed that APC eciently
associated with GST-NE-dlg, but not GST (Figure 8c).
We identi®ed and characterized NE-dlg gene, which is
a novel human homolog of Drosophila dlg tumor
suppressor gene. While the hdlg-1/SAP-97 transcript,
which had been previously identi®ed as another human
dlg homolog, was expressed in most human tissues,
NE-dlg transcript was abundantly detected in brain and
only moderately expressed in endocrine and glandular
tissues such as pancreas, thyroid, and trachea.
The NE-dlg protein sequence had high homology to
MAGUK protein families. The MAGUK protein
families have been shown to interact and/or co-
localize with various intracellular molecules, and to
be involved in membrane- and cytoskeleton-associated
signaling. In the present work, we performed the yeast
two-hybrid screening and found that NE-dlg interacted
with the carboxyl-terminal region of APC protein.
Matsumine and co-workers reported that the DHR
regions of hdlg-1/SAP-97 also interacts with the
carboxyl-terminal region of APC (Matsumine et al.,
1996). Since the amino acid sequence of DHR regions
in NE-dlg is highly conserved in that of hdlg-1/SAP-97,
APC protein may be able to interact with both human
dlg homologs. Rat postsynaptic protein (SAP102) was
identi®ed recently (Muller et al., 1996; Lau et al.,
1996), and it is likely to be the rat homolog of NE-dlg
based on the similarity of reported sequences. SAP102
has been demonstrated to be speci®cally expressed in
rat neuronal tissues and to interact with the carboxyl-
terminal tail of the NMDA receptor, to which hdlg-1/
SAP97 also binds. Both APC protein and the NMDA
receptor have a threonine (or serine)-X-valine (T/SXV)
motif at their carboxyl-terminal end, and this motif has
been shown to speci®cally bind to the DHR domains
(Kornau et al., 1995; Matsumine et al., 1996;
Niethammer et al., 1996). All these lines of evidence
suggest that NE-dlg/SAP102, as well as hdlg-1/SAP97,
plays a role in multiple cellular functions such as
Figure 7 Immunohistochemical analysis of NE-dlg by using anti-NE-dlg polyclonal antibody. (a) Immunoperoxidase staining
showed along the neuronal axon or dendrites on the cerebral cortex as indicated by a black arrow. (Frozen tissue section, 6400) (b)
Some cells in the Langerhans islets of the pancreas had dense cytoplasmic staining as pointed by a black arrow. (Frozen tissue
section, 6400) (c) The cytoplasmic regions of myocytes of the heart muscle were stained as pointed by black arrow. (Formalin-®xed
paran-embedded tissue section, 6200) (d) The sites of cell ± cell contact between epithelial cells in the prickle and functional layers
were strongly stained. However, the basal cells were not recognized by the antibody as indicated by a white arrow. (Formalin-®xed
paran-embedded tissue section, 6200)
Cloning and characterization of NE-dlg
KMakinoet al
clustering of NMDA receptors with submembraneous
cytomatrix at synaptic junctions in neuronal tissues
and regulating cell growth by interacting with APC
protein (Kornau et al., 1995; Matsumine et al., 1996;
Niethammer et al., 1996).
Immunolocalization studies using an anti-NE-dlg
antibody revealed that the NE-dlg protein is expressed
not only in neuronal and endocrine tissues, but also in
some non-neuronal tissues. Furthermore, intracellular
localization of the protein varied depending on cell
types. In the brain, NE-dlg protein was seen along
axons and dendrites but not found in cytoplasm of
cerebral cortex neurons. In glandular tissues, some
populations of islet cells in pancreas and follicular cells
in the thyroid were shown to have dense cytoplasmic
staining. These distributions of the NE-dlg protein
suggest that it may be associated with vesicular
transport and the secretion of neurotransmitters and
In strati®ed squamous epithelium, such as the
esophagus and tongue, characteristic expression
pattern of NE-dlg protein was observed by
immunohistochemical study. The protein was not
detected in basal cells, whereas the prickle and
functional layer cells highly expressed the protein at
sites of cell ± cell contact. Additionally, Western blot
analysis demonstrated that the 100 kDa NE-dlg
protein was not detected in cultured cancer cell
lines which have high proliferative activity. Further-
more, our recent observation revealed that the NE-
dlg protein was not detected any esophageal cancers,
while it was found in the adjacent esophageal
epithelium (Morisaki et al., in preparation). These
®ndings provide evidence of a tendency for lower
expression of NE-dlg protein in proliferative cells and
higher expression in non-proliferative cells, suggesting
that the NE-dlg expression negatively regulates cell
The molecular mechanism of the tumor suppressor
eect of APC is still obscure. However, recent reports
suggested that APC protein interacts with b-catenin
(Rubinfeld et al., 1993; Su et al., 1993) and this binding
may inhibit b-catenin-induced transcriptional activity,
resulting in growth suppression (Kinzler and Vogel-
stein, 1996). Furthermore, expression of wild-type APC
has been demonstrated to induce apoptosis in APC
mutated colorectal epithelial cells (Morin et al., 1996).
These observations indicate that APC could modulate
cadherin-catenin mediated cell adhesion signals and
consequently control cell growth and the cell death
process. Previous reports showed that most of the APC
gene mutations in tumor cells result in premature
translational termination and thereby produce trun-
cated APC proteins lacking their C-terminal regions.
Therefore, it is conceivable that the APC mutations
impair the APC-NE-dlg complex formation and may
consequently lead to development of tumors. This
could provide hypothesis that mutation of the NE-dlg
gene also impairs the complex formation without APC
mutation and results in tumorigenesis. Therefore, the
location of the NE-dlg gene on the chromosome Xq13
region can be a potential site for a tumor suppressor
gene. Although major genetic alterations including
large deletions, translocations or ampli®cation, of this
region have not been reported in any cancers, recent
observations and the pattern of familial transmission of
prostate cancer in men is consistent with an X-linked
or recessive genetic model of inheritance supports a
role of the X chromosome in the development of
certain cancers (Monroe et al., 1995). Further
experiments to detect alterations of the NE-dlg gene
in various inherited and sporadic cancers will be
required to demonstrate that it acts as a tumor
suppressor gene.
Drosophila has been analysed to have over 50
genes whose mutation results in hyperplasia or
neoplastic overgrowth in various tissues. More than
20 of them have been cloned and characterized, and
some have been identi®ed their mammalian homologs
(Watson et al., 1994). Our ®ndings demonstrated
that a human dlg homolog, like the Drosophila dlg,
is also involved in cell proliferative control and
tumor suppression. Accordingly, identifying and
characterizing human homologs of Drosophila tumor
suppressor genes potentially provides an alternative
approach to ®nding novel tumor suppressor genes
whose loss-of-function leads to development of
human tumors.
His(–) His(+)
200 kDa —
cell lysate ++
whole cell lysate
Detection: α-APC Ab
Figure 8 Interaction of NE-dlg with APC. (a) Yeast two-hybrid
assay of HIS3 reporter construct. The yeast transformants were
streaked onto synthetic medium plates lacking tryptophan, leucine
and histidine (left column) and plates lacking tryptophan and
leucine (right column). Interaction between NE-dlg and APC
resulted in activation of HIS3 reporter and enabled growth in the
absence of histidine. (b) Schematic representation of APC gene
product. The insert encodes the carboxyl terminal 111-amino acid
fragment of the APC protein. Underline indicates the consensus
motif (S/TXV) binding site for NE-dlg. (c) Protein-binding assay.
Lysates of HT1080 cells were incubated with either GST or with
the GST-NE-dlg fusion protein immobilized on GSH beads.
Bound cellular proteins were analysed by Western blotting and
the band speci®c for APC protein was detected by enhanced
chemiluminescene. APC associated speci®cally with GST-NE-dlg
(lane 1), but not with GST (lane 2)
Cloning and characterization of NE-dlg
KMakinoet al
Materials and methods
EST database screening and PCR-based full length cDNA
We searched for human homologs of the dlg gene by
scanning a database of human genes identi®ed by the
expressed sequence tag (EST) method using tblastn (Adams
et al., 1991; Altschul et al., 1990). The sequence
information of the clone R16282 obtained from EST
allowed the synthesis of a pair of speci®c primers, P1 (5'-
the P1 and P2 primers were performed to amplify a core
cDNA fragment from a lZAP human fetal brain cDNA
library (Stratagene). The core 221 bp sequence was
obtained from the PCR product and subcloned into a
pCRII TA cloning vector (Invitrogen). Two primers, P3
based on the sequence information of the core 221-bp
Ampli®cation of the 5'region was carried out by the two-
step PCR procedure using rTth DNA polymerase (Perkin
Elmer) which has proof reading activity. The ®rst PCR was
performed for 50 cycles using only P1 primer to amplify
single strand cDNA in a 15 ml reaction volume containing the
fetal brain cDNA library as a template. Each cycle consisted
of denaturation at 948C for 30 s, annealing at 608C for 30 s,
and extension at 728C for 90 s. The ®rst PCR product was
used as a template in the second run, where P3 and T3 (5'-
ATTAACCCTCACTAAAG-3') were utilized as primers to
amplify the 5'region cDNA. PCR was performed for 40
cycles of denaturation at 948C for 30 s, annealing at 608C for
30 s, and extension at 728C for 60 s. Ampli®cations of the 3'
region and the remaining 5'region were also performed by
the same two-step PCR as described above.
For the 3'region ampli®cation, P2 was used for the ®rst
round PCR, and P4 and T7 (5'-AATACGACTCACTATAG-
3') were used for the second round PCR. For ampli®cation of
the remaining 5'region, P5 (5'-GTTGTCTCCTGGGA-
TGTGCTGGTTG-3') was used for the ®rst PCR, and P6
were used for the second PCR. The PCR fragments were
ligated into a pCRII TA cloning vector and sequenced. DNA
sequencing of double-stranded plasmid DNAs was performed
with 373S DNA sequencer (Applied Biosystems) using
standard protocols. The nucleotide sequence was con®rmed
by sequencing several clones (more than three) that were
generated by independent PCR to avoid errors introduced
during the PCR reaction. DNA and protein sequence were
compared by the basic local alignment search tool (BLAST)
network service at the National Center for Biotechnology
Information (NCBI).
Northern blot analysis
Northern blot derived from multiple human tissues and
cancer cell lines containing 2 mgofpoly(A)
RNA per lane
were obtained from Clontech. The membranes were probed
with a 1.4 kbp cDNA fragment of NE-dlg,a1.1kbp
cDNA fragment of hdlg-1/SAP-97 and a 2.0 kbp human b-
actin cDNA that had been labeled with [a-
P]dCTP by
random-primed labeling. The membranes were exposed to
X-ray ®lm with an intensifying screen for 3 days at 7708C.
A 1.7 Kb cDNA fragment of human NE-dlg (n.t. 217 ±
1991) was used as a probe for FISH analysis. The probe was
biotin-labeled, ethanol precipitated with human placental
DNA and salmon sperm DNA, and resuspended in
hybridization solution (50% formamide/26SSC/10% dex-
tran sulfate). Chromosome slides were prepared from PHA-
stimulated female peripheral lymphocyte cultures treated
with thymidine synchronization and bromodeoxyuridine.
Direct R-banding FISH followed by G-banding with
Wright Giemsa staining was performed as described in
detail elsewhere (Kuwano et al., 1991). In the brief, slide
was stained with Hoechst 33258 mounted in 26SSC,
exposed to a black lamp at 758C, denatured, preassociated
for 30 min at 378C, and in situ hybridized. After detection
using ¯uorescein-labeled avidin, slides were mounted in
0.7 mg/ml propidium iodine and analysed with a Nikon
Opitiphoto-2 (Nikon, Japan) and Nikon ®lter combination
B-2A (excitation at 450 ± 490 nm). The photographs were
taken with Ektachrome ASA 100 ®lm (Kodak). The slide
was then desalted, immersed in borate buer, and stained in
a mixture of Wright's Giemsa solution and borate buer.
PCR mapping of NE-dlg gene
PCR was performed to detect NE-dlg sequences in the
Monochromosomal somatic cell hybrid DNA panel (UK
HGMP Resource Center) and in the Genebridge 4
Radiation Hybrid Screening Panel (Research Genetics,
Inc) using a set of primers (S-1 : 5'-TGAGTGGTGGCA-
GAAGTCCTGTTA-3') which were designed based on a
partial genomic sequence of the NE-dlg gene. The S-1 and
AS-1 primers were expected to amplify a 306 bp PCR
product from human genomic DNA as a template. PCR
was carried out in a 15ml reaction volume for 35 cycles.
Each cycle consisted of denaturation at 958Cfor30s,
annealing at 608C for 30 s, extension at 728Cfor30s,and
5ml of each reaction was analysed by electrophoresis
through a 2.0% agarose gel run in TAE buer.
The PCR results of the Radiation Hybrid Panel were sent
to the Whitehead Institute/MIT Center for Genome Research
via WWW ( for mapping of
the genes relative to the radiattion hybrid map of the human
genome (Hudson et al., 1995).
Production of anti-NE-dlg polyclonal antibody
Antibody against a unique protein sequence in NE-dlg was
raised by the subcutaneous immunization of rabbits with a
GAKN) which was coupled to KLH. The antibody was
anity-puri®ed on a column containing the peptide cross-
Construction of the NE-dlg expression vector
The full length ORF of the NE-dlg cDNA (2453 bp) was
ampli®ed by PCR from the lZAP human fetal brain cDNA
library by the primer NED forward (5'-CCGCGTCT-
containing a XbaI site (underlined) and the primer NED
TTCAGGGGATGGGAC-3') containing a KpnI site
(underlined). Cycling reactions were performed in a 25 ml
reaction volume for 40 cycles using rTth DNA polymerase.
Each cycle consisted of denaturation at 948Cfor1min,
annealing at 608C for 1 min, and extension at 728Cfor
2 min. The PCR fragments were digested with XbaI and
KpnI, and ligated into a pCGN expression vector. To
express HA epitope-tagged NE-dlg, the pCGN-NE-dlg was
transfected into COS-7 cells by the liposome-mediated gene
transfer method (Felgner et al., 1987).
Western blotting
For Western blotting, samples containing equal amounts of
protein (15 mg) from lysates of cultured cells and frozen
Cloning and characterization of NE-dlg
KMakinoet al
tissues were separated on an 8% polyacrylamide gel and
transferred to a nitrocellulose ®lter with a constant current
of 140 mA for 2 h. The ®lters were blocked overnight at
48C with PBS containing 10% skim milk and then
incubated with 1 : 1000 diluted anti-HA antibody
(12CA5), 1 : 500 diluted anti-NE-dlg antibody and
1 : 10000 diluted anti-b-tubulin antibody (CHEMICON)
in PBS containing 0.03% Tween 20 for 2 h, and washed
two times for 7 min each time with PBS containing 0.3%
Tween 20. The ®lters were then incubated with horseradish
peroxidase-conjugated anti-mouse IgG antibody for anti-
HA and anti-b-tubulin antibodies and anti-rabbit IgG
antibody for anti-NE-dlg antibody (Amersham) for
40 min, and speci®c proteins were detected using an
enhanced chemiluminescence system (Amersham).
Immunohistochemical analysis
Formalin-®xed, paran-embedded tissues were cut into
sections 4 mm thick, deparanized and rehydrated. Frozen
tissue sections were cut into sections 6 mm thick and ®xed
in 10% buered formalin for 15 min. After ®xation, the
slides were rinsed in PBS. Endogenous peroxidase was
inhibited by treatment with 0.3% H
in methanol for
15 min. Immunoperoxidase staining was performed with an
avidin-biotin kit (Vector, Burlingame, CA). Nonspeci®c
staining was blocked using normal goat serum. After
shaking o the serum, the slides were treated by primary
antibody (2 mg/ml) in PBS for 1 h. They were washed with
PBS and then incubated with secondary biotinylated anti-
rabbit immunoglobulin antibody for 30 min and avi-
din:biotin-conjugated horseradish peroxidase for 30 min,
with rinses in PBS between each step. The peroxidase
substrate, 3.3'-diaminobenzidine tetrahydrochloride (DAB:
3mg in 10ml of 0.05MTris-HCl, pH 7.6, in 0.01%
hydrogen peroxide), was added, and the peroxidase
reaction was allowed to continue for 5 min at room
temperature. Excess DAB was then washed o for 5 min,
and the slides were counterstained in hematoxylin, and
mounted with Crystal/Mount (Biomeda).
Yeast two-hybrid screening
The partial NE-dlg cDNA, which included three DHR
repeats and the SH3 region, was obtained by PCR using a
and BamHI cloning sites (underlined), respectively. The
PCR product was double digested with NdeI and BamHI
and subcloned into the pAS1-CYH2 vector as a fusion to
the CAL4 DNA-binding domain. The resultant plasmid,
pGAL4bd-NE-dlg was cotransformed with a prey plasmid
containing a human fetal brain cDNA library fused to the
GAL4 activation domain (GAL4ad) in the pGAD10 vector
(Clontech) by elecroporation. The transformants (5610
were screened with streaking on selection medium (SD-Trp,
-Leu and -His) followed by b-galactosidase ®lter assay. The
pGAD10 plasmids containing the inserted cDNA were
recovered from the positive clones and cotransformed with
pGAL4bd-NE-dlg or control baits to con®rm their
interaction. DNA sequencing of the cDNA inserted into
the positive plasmid was performed by the dideoxynucleo-
tide chain termination method.
Construction of GST fusion protein and in vitro binding assay
The NE-dlg cDNA fragment used in the pGAL4-bd-NE-
dlg plasmid was subcloned into a pGEX-2TH bacterial
expression vector. Expression and puri®cation of the GST-
NE-dlg protein were performed as described previously
(Takeshima et al., 1994). Human ®brosarcoma HT1080
cells were lysed with TNT buer (50 mMTris-HCl pH 7.4,
150 mMNaCl and 0.5% Triton X-100) containing 1 mM
phenylmethylsulfonyl ¯uoride (PMSF), 10 mg/ml leupeptin,
1.5 mMpepstatin, 3% aprotinin, and 1 mMsodium
orthovanadate, and the lysate was incubated with either
GST or the GST-NE-dlg fusion protein immobilized on
glutathion-agarose beads (GSH-beads) for 1 h at 48C. The
beads were then washed with TNT buer and heated in
Laemmlis' sample buer (Laemmli, 1970) at 1008Cfor
5 min. The samples were resolved by SDS ± PAGE and
transferred onto a nitrocellulose membrane, and detected
by anti-APC antibody (Ab-2: Oncogene Science, Inc).
We thank Drs K Tanaka and Y Takai, Department of
Molecular Biology and Biochemistry, Osaka University
Medical School, for technical advice revealed to yeast two-
hybrid screening; Dr Q Hu for providing the pCGN
expression vector; Dr H Maruta, Ludwig Institute for
Cancer Research, for providing the pGEX-2TH bacterial
expression vector; K Uriuda for secretarial assistance; and
Dr J Moon for editing the manuscript. This work was
supported by a grant for Cancer Research from the
Ministry of Education, Science and Culture of Japan
(HS). Genbank accession No. U49089.
Adams MD, Kelley JM, Gocayne JD, Dubnick M,
Polymeropoulos MH, Xiao H, Merril CR, Wu A, Olde
B, Moreno RF, Kerlavage AR, McCombie WR and
Venter JC. (1991). Science,252, 1651 ± 1656.
Altschul SF, Gish W, Miller W, Myers EW and Lipman DJ.
(1990). J. Mol. Biol.,215, 403 ± 410.
Cho KO, Hunt CA and Kennedy MB. (1992). Neuron.,9,
929 ± 942.
Durfee T, Becherer K, Chen PL, Yeh SH, Yang Y, Kilburn
AE, Lee WH and Elledge SJ. (1993). Genes Dev.,7, 555 ±
Felgner PL, Gadek TR, Holm M, Roman R, Chan HW,
Wenz M, Northrop JP, Ringold GM and Danielsen M.
(1987). Proc. Natl. Acad. Sci. USA,84, 7413 ± 7417.
Gate E and Mechler BM. (1989). Crit. Rev. Oncog.,1, 221 ±
Hudson TJ, Stein LD, Gerety SS, Ma J, Castle AB, Silva J,
Slonim DK, Baptista R, Kruglyak L, Xu SH, Hu X,
Hui L, Wu X, Vestergaard C, Wilson KM, Bae JS, Maitra
S, Ganiatsas S, Evans CA, DeAngelis MM, Ingalls KA,
Nahf RW, Horton LT Jr, Anderson MO, Collymore AJ,
Ye W, Kouyoumjian V, Zemsteva IS, Tam J, Devine R,
Courtney DF, Renaud MT, Nguyen H, O'Connor TJ,
Fizames C, Faure S, Gyapay G, Dib C, Morissette J, Orlin
JB, Birren BW, Goodman N, Weissenback J, Hawkins TL,
Foote S, Page DC and Lander ES. (1995). Science,270,
1945 ± 1954.
Itoh M, Nagafuchi A, Yonemura S, Kitani YT, Tsukita S
and Tsukita S. (1993). J. Cell. Biol.,121, 491 ± 502.
Jesaitis LA and Goodenough DA. (1994). J. Cell. Biol.,124,
949 ± 961.
Cloning and characterization of NE-dlg
KMakinoet al
Kim E, Niethammer M, Rothschild A, Jan YN and Sheng M.
(1995). Nature,378, 85 ± 88.
Kinzler KW and Vogelstein B. (1996). Cell,87, 159 ± 170.
Kistner U, Wenzel BM, Veh RW, Cases LC, Garner AM,
Appeltauer U, Voss B, Gundel®nger ED and Garner CC.
(1993). J. Biol. Chem.,268, 4580 ± 4583.
Kornau HC, Schenker LT, Kennedy MB and Seeburg PH.
(1995). Science,269, 1737 ± 1740.
Kuwano A, Ledbetter SA, Dobyns WB, Emanuel BS and
Ledbetter DH. (1991). Am. J. Hum. Genet.,49, 707 ± 714.
Laemmli UK. (1970). Nature,227, 680 ± 685.
Lau LF, Mammen A, Ehlers MD, Kindler S, Chung WJ,
Garner CC and Huganir RL. (1996). J. Biol. Chem.,271,
21622 ± 21628.
Lue RA, Marfatia SM, Branton D and Chishti AH. (1994).
Proc. Natl. Acad. Sci. USA,91, 9818 ± 9822.
Matsumine A, Ogai A, Senda T, Okumura N, Satoh K, Baeg
GH, Kawahara T, Kobayashi S, Okada M, Toyoshima K
and Akiyama T. (1996). Science,272, 1020 ± 1023.
Monroe KR, Yu MC, Kolonel LN, Coetzee GA, Wilkens
LR, Ross RK and Henderson BE. (1995). Nat. Med.,1,
827 ± 829.
Morin PJ, Vogelstein B and Kinzler KW. (1996). Proc. Natl.
Acad. Sci. USA,93, 7950 ± 7954.
Muller BM, Kistner U, Veh RW, Cases LC, Becker B,
Gundel®nger ED and Garner CC. (1995). J. Neurosci.,15,
2354 ± 2366.
Muller BM, Kistner U, Kindler S, Chung WJ, Kuhlendahl S,
Fenster SD, Lau LF, Veh RW, Huganir RL, Gundel®nger
ED and Garner CC. (1996). Neuron.,17, 255 ± 265.
Niethammer M, Kim E and Sheng M. (1996). J. Neurosci.,
16, 2157 ± 2163.
Rubinfeld B, Souza B, Albert I, Muller O, Chamberlain SH,
Masiarz FR, Munemitsu S and Polakis P. (1993). Science,
262, 1731 ± 1734.
Ru P, Speicher DW and Husain CA. (1991). Proc. Natl.
Acad. Sci. USA,88, 6595 ± 6599.
Stehle T and Schulz GE. (1992). J. Mol. Biol.,224, 1127 ±
Su LK, Vogelstein B and Kinzler KW. (1993). Science,262,
1734 ± 1737.
Sudol M, Chen HI, Bougeret C, Einbond A and Bork P.
(1995). FEBS. Lett.,369, 67 ± 71.
Takeshima H, Izawa I, Lee PS, Safdar N, Levin VA and Saya
H. (1994). Oncogene,9, 2135 ± 2144.
Watson KL, Justice RW and Bryant PJ. (1994). J. Cell. Sci.
Suppl.,18, 19 ± 33.
Williamson MP. (1994). Biochem. J.,297, 249 ± 260.
Willott E, Balda MS, Fanning AS, Jameson B, Van IC and
Anderson JM. (1993). Proc. Natl. Acad. Sci. USA,90,
7834 ± 7838.
Woods DF and Bryant PJ. (1989). Dev. Biol.,134, 222 ± 235.
Woods DF and Bryant PJ. (1993). Mech. Dev.,44, 85 ± 8 9.
Cloning and characterization of NE-dlg
KMakinoet al

Supplementary resource (1)

... 172,173 Like Dlg1, Dlg3 also associates with APC, although the effect of this interaction remains unclear. 174 While Dlg3 is less expressed in proliferative cells compared to nonproliferating ones such as neurons, 174,175 overexpression of Dlg3 results in the suppression of cell growth and downregulation of β-catenin as well as cell migration and invasion. 175,176 Consistent with a tumor suppressor function, altered localization and overall expression of Dlg is shown in cervical carcinomas, which are often related to human papillomavirus (HPV) infection. ...
... [177][178][179] In colorectal cancers that are related to APC mutation/inactivation, the expression and distribution of Dlg is also perturbed. 180 Moreover, somatic mutations in Dlg1 have been reported in breast ductal carcinoma, 181 and Dlg3 is downregulated in gastric, 182 esophageal, 174,175 and papillary thyroid carcinoma, 183 as well as glioblastoma. 176 In contrast, Dlg3 is reportedly upregulated in serous ovarian carcinomas. ...
Full-text available
Polarity protein complexes are crucial for epithelial apical-basal polarity and directed cell migration. Since alterations of these processes are common in cancer, polarity proteins have been proposed to function as tumor suppressors or oncogenic promoters. Here, we review the current understanding of polarity protein functions in epithelial homeostasis, as well as tumor formation and progression. As most previous studies focused on the function of single polarity proteins in simplified model systems, we used a genomics approach to systematically examine and identify the expression profiles of polarity genes in human cancer. The expression profiles of polarity genes were distinct in different human tissues and classified cancer types. Additionally, polarity expression profiles correlated with disease progression and aggressiveness, as well as with identified cancer types, where specific polarity genes were commonly altered. In the case of Scribble, gene expression analysis indicated its common amplification and upregulation in human cancer, suggesting a tumor promoting function.
... Both Scribble and Dlg contain multiple PDZ domains with high levels of conservation. These PDZ domains are vital to protein-protein interactions that include Scribble and Dlg within the polarity establishment pathway and others (Kallay et al., 2006;Takizawa et al., 2006;How et al., 2019;Bilder 2003;Bilder, Li, and Perrimon 2000;Bilder, Schober, and Perrimon 2003;Sotelo et al., 2012;Matsumine et al., 1996;Makino et al., 1997;Subbaiah et al., 2012). Lgl displays varying positions of its LGL domain, as well as different numbers of WD40 domains between species ( Figures 5C,D). ...
Full-text available
Epithelial tissues are vital to the function of most organs, providing critical functions such as secretion, protection, and absorption. Cells within an epithelial layer must coordinate to create functionally distinct apical, lateral, and basal surfaces in order to maintain proper organ function and organism viability. This is accomplished through the careful targeting of polarity factors to their respective locations within the cell, as well as the strategic placement of post-mitotic cells within the epithelium during tissue morphogenesis. The process of establishing and maintaining epithelial tissue integrity is conserved across many species, as important polarity factors and spindle orientation mechanisms can be found in many phyla. However, most of the information gathered about these processes and players has been investigated in bilaterian organisms such as C. elegans, Drosophila , and vertebrate species. This review discusses the advances made in the field of epithelial polarity establishment from more basal organisms, and the advantages to utilizing these simpler models. An increasing number of cnidarian model organisms have been sequenced in recent years, such as Hydra vulgaris and Nematostella vectensis . It is now feasible to investigate how polarity is established and maintained in basal organisms to gain an understanding of the most basal requirements for epithelial tissue morphogenesis.
... Indeed SAP97 also interacts with ezrin, an actin-binding protein crucial for morphogenesis of apical microvilli and basolateral infoldings in retinal pigment epithelial (RPE) cells. Through the PDZ2 domain the protein also interacts with the carboxyl-terminal S/TXV motif of the APC (Adenomatous polyposis coli) tumour suppressor protein and plays an important role in transducing the APC cell cycle blocking signal (Makino et al., 1997;Ishidate et al., 2000;Mimori-Kiyosue et al., 2007). Recent studies report that APC can bind to all the three PDZ domains from hDlg, whereas PTEN mainly binds to PDZ-2/hDlg. ...
... As extrasynaptic NR2B is the main factor that triggers excitotoxicity, then how is it that synaptic NR2B transfers to the extrasynapse? SAP102, as an important member of MAGUKs family, controls synapse formation by clustering neurotransmitter receptors in axons and dendrites of matured neurons [60]. Studies have shown that SAP102 localizes closely with NR2B by PDZ domin [27]. ...
Full-text available
Background: Chronic intermittent hypoxia-hypercapnia (CIHH) exposure leads to learnning and memory deficits in rats. Overactivation of N-methyl-D-aspartate receptors(NMDARs) can lead to the death of neurons through a process termed excitotoxicity, which is involved in CIHH-induced cognitive deficits. Excessively activated NR2B (GluN2B)-containing NMDARs was reported as the main cause of excitotoxicity. The ERK1/2 (extracellular signal-regulated kinase 1/2) signaling cascade acts as a key component in NMDARs-dependent neuronal plasticity and survival. Ca2+/calmodulin-dependent protein kinase II (CaMKII), synapse-associated protein 102 (SAP102) and Ras GTPase-activating protein (SynGAP) have been shown to be involved in the regulation of NMDAR-ERK signalling cascade. Recent studies revealed statins (the HMG-CoA reductase inhibitor) have effect on the expression of NMDARs. The present study intends to explore the potential effect of lovastatin on CIHH-induced cognitive deficits and the NR2B-ERK signaling pathway. Methods and findings: Eighty male Sprague Dawley rats were randomly divided into five groups. Except for those in the control group, the rats were exposed to chronic intermittent hypoxia-hypercapnia (CIHH) (9 ∼ 11%O2, 5.5 ∼ 6.5%CO2) for 4 weeks. After lovastatin administration, the rats performed better in the Morris water maze test. Electron microscopy showed alleviated hippocampal neuronal synaptic damage. Further observation suggested that either lovastatin or ifenprodil (a selective NR2B antagonist) administration similarly downregulated NR2B subunit expression leading to a suppression of CaMKII/SAP102/SynGAP signaling cascade, which in turn enhanced the phosphorylation of ERK1/2. The phosphorylated ERK1/2 induced signaling cascade involving cAMP-response element-binding protein (CREB) phosphorylation and brain-derived neurotrophic factor (BDNF) activation, which is responsible for neuroprotection. Conclusions: These findings suggest that the ameliorative cognitive deficits caused by lovastatin are due to the downregulation of excessive NR2B expression accompanied by increased expression of ERK signaling cascade. The effect of NR2B in upregulating pERK1/2 maybe due, at least in part, to inactivation of CaMKII/SAP102/SynGAP signaling cascade.
Full-text available
X-lined intellectual disability (XLID), formerly known as X-lined mental retardation, is defined as genetically heterogeneous disorders with remarkable cognitive impairment and abnormal adaptive behaviour skills. This study demonstrates the Disc-large homolog 3 (DLG3) gene impairment in 2 different unrelated male probands. The results detected two missense mutations in the DLG3 gene, c.2267 G > A (p.Arg756Gln) and c.2359G > A p. (Gly787Ser) using by NGS. Both mutations were run in the PolyPhen2 program for mutation sensitivity check and showed to have 0.709 and 1, respectively. The familial transmission pattern of MR detected both mothers to be heterozygote. The mutations were shown to have caused non-syndromic XLMR (NS-XLMR) as both males did not show any abnormal facial or physiological features. Based on the IQ measurement, proband 1 and 2’ IQs were measured 40 and 33, and they were diagnosed with moderate and severe XLMR, respectively. Both affected males showed significant deterioration in neural development and behaviour abilities, which indicates the significant impact of the mutation on neurotransmitters and maintenance of NMDA receptors in neural functions. However, further molecular and functional studies are necessary to provide more conclusive evidence of the detailed abnormalities caused by the reported mutations.
Familial adenomatous polyposis (FAP) is a dominantly inherited syndrome characterised by the development of hundreds to thousands of adenomatous polyps throughout the colon and rectum. These usually arise during the second or third decades of life, and in the absence of surgical intervention it is inevitable that one or more will progress from a benign to a malignant state. The underlying cause of FAP is mutation of the adenomatous polyposis coli (APC) gene, a tumour suppressor gene located on chromosome 5q21-22. Analysis of the APC gene undertaken during this study has provided data on the type and distribution of APC mutations in FAP patients. Fifty two unrelated patients were investigated and mutation was detected in 35 (67%) of them. Of the different types of mutation identified deletions and insertions were the most frequently noted accounting for 85% of all mutations. Nonsense and splice consensus sequence mutations were also detected, all of which are predicted to result in premature truncation of the APC protein. Mutations were identified in most regions of the APC gene assessed but showed a marked clustering within the 5' half of exon 15. Mutations at codon 1061 and codon 1309 were found to be responsible for 10% and 19% of FAP cases respectively. In addition to analysis of APC sequence variants chromosomal rearrangements and deletions in the vicinity of the APC gene were investigated using fluorescence in situ hybridisation (FISH). Aspects of the FAP phenotype were considered and compared to the site of mutation within the APC gene. A correlation between severity of phenotype, defined by number of polyps and age of disease onset, and location of APC mutation was identified. Mutation between codons 1309 and 1464 was found to be associated with early onset of cancer and the development of thousands rather than hundreds of colorectal polyps. Interestingly the majority of patients with no family history of FAP (new mutations) were found to have mutation within this 'severe' region of APC. Presymptomatic diagnosis of FAP was performed for members of FAP families that were of uncertain gene status. This was accomplished by direct mutation detection or by linkage analysis. Additionally a strategy for preimplantation genetic diagnosis of this disorder was also developed. This involved detection of APC mutation in single cells biopsied from human preimplantation embryos at the cleavage stage. After extensive preliminary testing this method reached clinical application, but no pregnancy was achieved on this occassion.
Full-text available
MicroRNAs (miRNAs) are important post-transcriptional regulators involved in hypoxia conditions; however, their roles in HepG2 cells remain poorly understood. Our previous study showed that hypoxia treatment modulated gene expression accompanied by with HepG2 cell proliferation arrest and increased cell death. To better understand the mechanism of phenotypic changes of HepG2 under hypoxia conditions; we conducted a comparative RNA sequencing to identify differentially expressed miRNAs between hypoxia treatment and control cells. In total, 165 differentially expressed miRNAs were identified, among which the expression of 114 miRNAs were up-regulated and that of 51 miRNAs were down-regulated in hypoxia treated HepG2 cells. Expression profiles of eleven randomly selected miRNAs were validated by qRT-PCR. Furthermore, 19 367 annotated target genes of differentially expressed miRNAs were predicted by bioinformatics tools. The Gene Ontology analysis indicated that the molecular function of target genes was primarily related to binding and catalytic activity, and that the Kyoto Encyclopedia of Genes and Genomes annotation for target genes were further classified into pathways involved in cellular processes, metabolism, organismal systems, genetic information processing, human disease and environmental information processing. Among the environmental information processing, certain pathways associated with cell proliferation and apoptosis, such as the hippo signalling pathway, wnt signalling pathway, MAPK signalling pathway and Jak-STAT signaling pathways, represented potential factors in the response to hypoxia treatment. In conclusion, the expression profiles of miRNA in HepG2 cells were significantly altered under hypoxia conditions; which were closely related to cell proliferation arrest and apoptosis. Our findings expand our understanding of miRNAs function in regulating cell fate under hypoxia conditions.
The MAGUKs (membrane-associated guanylate kinase homologs) are a family of proteins that act as molecular scaffolds for signaling pathway components at the plasma membrane of animal cells. They are localized in and required for the formation of several types of cell junctions, including epithelial tight and septate junctions as well as synaptic and neuromuscular junctions. They are also localized at the plasma membrane of other cell types, including erythrocytes, where they contribute to cell shape maintenance. MAGUKs function mainly by binding directly to the cytoplasmic termini of transmembrane proteins as well as to other signal transduction proteins. They appear to hold together elements of individual signaling pathways, thereby contributing to the efficiency and specificity of signaling interactions while simultaneously maintaining the structural specializations of the plasma membrane. BioEssays 1999;21:912–921.
Full-text available
DNA methylation of promoter-associated CpG islands is involved in the transcriptional repression of vertebrate genes. To investigate the mechanisms underlying gene inactivation by DNA methylation, we characterized a human MBD1 protein, one of the components of MeCP1, which possesses a methyl-CpG binding domain (MBD) and cysteine-rich (CXXC) domains. Four novel MBD1 isoforms (MBD1v1, MBD1v2, MBD1v3, and MBD1v4) were identified by the reverse transcription-PCR method. We found that these transcripts were alternatively spliced in the region of CXXC domains and the C terminus. Green fluorescent protein-fused MBD1 was localized to multiple foci on the human genome, mostly in the euchromatin regions, and particularly concentrated in the pericentromeric region of chromosome 1. Both the MBD sequence and genome methylation were required for proper localization of the MBD1 protein. We further investigated whether MBD1 isoforms are responsible for transcriptional repression of human genes. A bacterially expressed MBD1 protein bound preferentially to methylated DNA fragments containing CpG islands from the tumor suppressor genes p16 , VHL , and E-cadherin and from an imprinted SNRPN gene. All MBD1 isoforms inhibited promoter activities of these genes via methylation. Interestingly, MBD1 isoforms v1 and v2 containing three CXXC domains also suppressed unmethylated promoter activities in mammalian cells. These effects were further manifested in Drosophila melanogaster cells, which lack genome methylation. Sp1-activated transcription of methylated p16 and SNRPN promoters was inhibited by all of the MBD1 isoforms, whereas the isoforms v1 and v2 reduced Sp1-activated transcription from unmethylated promoters as well. These findings suggested that the MBD1 isoforms have different roles in methylation-mediated transcriptional silencing in euchromatin.
Multiple endocrine neoplasia type 1 (MEN1) arises from the oncogenic transformation of neuroendocrine cells due to inactivating mutations of the MEN1 gene, a putative tumor-suppressor gene (tsg). Except for interaction with the proto-oncogene JunD, we have ignored the function of its product, menin, because no homology with other proteins was reported. Starting from our observation of clusters of basic amino acids, we show here that they fit motifs of both binding sites for heparin-analogs of the extracellular matrix (ECM) and nuclear localization signals, and that the motifs are conserved in known tsg. We also demonstrate that each tsg product has about four regions of scrambled homology with certain regions of menin. Such commonality of domains qualifies menin as a tsg. To provide more precise indications for the functions of menin, we probed the Non-Redundant GenBank database with the previously unnoticed tandemly repeated pentamer QSQGT/S (a motif that is always spared by MEN1-associated mutations), or with segments of the tsg-dissimilar sequences of menin. A single copy of the pentamer was found in a number of neuron proteins, two of which are of particular relevance: ankyrin 2 and neuroendocrine-dlg (NE-dlg). The tsg-dissimilar sequences matched only 6 proteins (or 0.0017% of the stored sequences): early nodulin GRP3 (fava bean), SphI restriction endonuclease (Streptomyces phaechromogenes), brain micro glutamic acid-rich protein, Alzheimer's protein precursor, syndecan 2 and talin. The homology with talin is extremely important because it concerns an E-cadherin motif that is also possessed by the product (ret) of the gene causative of another MEN syndrome, MEN2. This motif is spared by mutations in both MEN1 and MEN2. In conclusion, menin has the characteristics of a multiple-domain tsg and, thus, has multiple functions, such as inactivation of an AP1 transcription factor (the proto-oncogene JunD), regulation of the bioavailability of ECM-stored growth factors, anchoring of the integral proteins of the plasma membrane to the cytoskeleton and cell adhesion. The molecular link between MEN1 and MEN2 that we are describing should explain not only the similarities between the two MEN syndromes but also the occurrence of mixed syndromes.
Full-text available
The complete amino acid sequence of a 55-kDa erythrocyte membrane protein was deduced from cDNA clones isolated from a human reticulocyte library. This protein, p55, is copurified during the isolation of dematin, an actin-bundling protein of the erythrocyte membrane cytoskeleton. Fractions enriched in p55 also contain protein kinase activity that completely abolishes the actin-bundling property of purified dematin in vitro. The predicted amino acid sequence of p55 does not contain any consensus sequence corresponding to the catalytic domains of protein kinases but does contain a conserved sequence found in the noncatalytic domains of oncogene-encoded tyrosine kinases. This conserved src homology 3 (SH-3) motif appears to suppress the tyrosine kinase activity of various oncoproteins and has also been found in several plasma membrane associated proteins involved in signal transduction. Northern blot analysis indicated that p55 mRNA was constitutively expressed during erythropoiesis and underwent 2-fold amplification after induction of K562 erythroleukemia cells toward the erythropoietic lineage. The abundant expression of p55 mRNA, along with protein 4.1 mRNA, was evident in terminally differentiated human reticulocytes. Although p55 has many features consistent with known peripheral membrane proteins, its tight association with the plasma membrane is reminiscent of an integral membrane protein. This fact may be partly explained by the observation that p55 is the most extensively palmitoylated protein of the erythrocyte membrane.
Full-text available
Automated partial DNA sequencing was conducted on more than 600 randomly selected human brain complementary DNA (cDNA) clones to generate expressed sequence tags (ESTs). ESTs have applications in the discovery of new human genes, mapping of the human genome, and identification of coding regions in genomic sequences. Of the sequences generated, 337 represent new genes, including 48 with significant similarity to genes from other organisms, such as a yeast RNA polymerase II subunit; Drosophila kinesin, Notch, and Enhancer of split; and a murine tyrosine kinase receptor. Forty-six ESTs were mapped to chromosomes after amplification by the polymerase chain reaction. This fast approach to cDNA characterization will facilitate the tagging of most human genes in a few years at a fraction of the cost of complete genomic sequencing, provide new genetic markers, and serve as a resource in diverse biological research fields.
Full-text available
A DNA-transfection protocol has been developed that makes use of a synthetic cationic lipid, N-[1-(2,3-dioleyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTMA). Small unilamellar liposomes containing DOTMA interact spontaneously with DNA to form lipid-DNA complexes with 100% entrapment of the DNA, DOTMA facilitates fusion of the complex with the plasma membrane of tissue culture cells, resulting in both uptake and expression of the DNA. The technique is simple, highly reproducible, and effective for both transient and stable expression of transfected DNA. Depending upon the cell line, lipofection is from 5- to greater than 100-fold more effective than either the calcium phosphate or the DEAE-dextran transfection technique.
The retinoblastoma protein (p110RB) interacts with many cellular proteins in complexes potentially important for its growth-suppressing function. We have developed and used an improved version of the yeast two-hybrid system to isolate human cDNAs encoding proteins able to bind p110RB. One clone encodes a novel type 1 protein phosphatase catalytic subunit (PP-1 alpha 2), which differs from the originally defined PP-1 alpha by an amino-terminal 11-amino-acid insert. In vitro-binding assays demonstrated that PP-1 alpha isoforms preferentially bind the hypophosphorylated form of p110RB. Moreover, similar p110RB sequences are required for binding PP-1 alpha 2 and SV40 large T antigen. Cell cycle synchrony experiments revealed that this association occurs from mitosis to early G1. The implications of these findings on the regulation of both proteins are discussed.
The crystal structure of guanylate kinase from Saccharomyces cerevisiae complexed with its substrate GMP has been refined at a resolution of 2.0 A. The final crystallographic R-factor is 17.3% in the resolution range 7.0 A to 2.0 A for all reflections of the 100% complete data set. The final model has standard geometry with root-mean-square deviations of 0.016 A in bond lengths and 3.0 in bond angles. It consists of all 186 amino acid residues, the N-terminal acetyl group, the substrate GMP, one sulfate ion and 174 water molecules. Guanylate kinase is structurally related to adenylate kinases and G-proteins with respect to its central beta-sheet with connecting helices and the giant anion hole that binds nucleoside triphosphates. These nucleotides are ATP and GTP for the kinases and GTP for the G-proteins. The chain segment binding the substrate GMP of guanylate kinase differs grossly from the respective part of the adenylate kinases; it has no counterpart in the G-proteins. The binding mode of GMP is described in detail. Probably, the observed structure represents one of several structurally quite different intermediate states of the catalytic cycle.
In CNS synapses, the synaptic junctional complex with associated postsynaptic density is presumed to contain proteins responsible for adhesion between pre- and postsynaptic membranes and for postsynaptic signal transduction. We have found that a prominent, brain-specific protein (PSD-95) enriched in the postsynaptic density fraction from rat brain is highly similar to the Drosophila lethal(1)discs-large-1 (dlg) tumor suppressor protein. The dlg protein is associated with septate junctions in developing flies and contains a guanylate kinase domain that is required for normal control of cell division. The sequence similarity between dlg and PSD-95 suggests that molecular mechanisms critical for growth control in developing organisms may also regulate synapse formation, stabilization, or function in the adult brain.
We present a genetic, developmental and molecular analysis of lethal(1)discs large-1[l(1)d.lg-1; Stewart et al., 1972], an oncogene of Drosophila. Mutations in this gene cause the imaginal discs to grow by cell proliferation beyond their normal final size, transform into solid tumors, fuse with one another and the brain, and lose their ability to differentiate. The oncogene represents the only known complementation group between two deficiency breakpoints, and 15 recessive lethal alleles are available. Cloning of the DNA between the two deficiency breakpoints defines a region of 45 +/- 2 kb. The l(1)d.lg-1 transcription unit is identified by both qualitative and quantitative effects of several l(1)d.lg-1 mutations on the RNA transcripts and by the presence of a DNA insert in one of the l(1)d.lg-1 alleles. It gives rise to at least five different transcripts ranging in size from 1.9 to 6.0 kb. Three other transcription units are present within this region, two 5' to the l(1)d.lg-1 gene and one at the 3' end. A near full-length cDNA from one of the larger transcripts of l(1)d.lg-1 has homology to genomic DNA spanning over 20 kb. A developmental profile of l(1)d.lg-1 transcription is presented. We discuss how mutations in this gene could disrupt epithelial structure and how this might be related to the excessive cell proliferation and interdisc fusion that is observed. We also compare this gene with another recessive oncogene of Drosophila, lethal(2)giant larvae, that has been cloned and characterized.
Using an improved method of gel electrophoresis, many hitherto unknown proteins have been found in bacteriophage T4 and some of these have been identified with specific gene products. Four major components of the head are cleaved during the process of assembly, apparently after the precursor proteins have assembled into some large intermediate structure.
ANCHORING of ion channels at specific subcellular sites is critical for neuronal signalling, but the mechanisms underlying channel localization and clustering are largely unknown (reviewed in ref. 1). Voltage-gated K+ channels are concentrated in various neuronal domains, including presynaptic terminals, nodes of Ranvier and dendrites, where they regulate local membrane excitability. Here we present functional and biochemical evidence that cell-surface clustering of Shaker-subfamily K+ channels is mediated by the PSD-95 family of membrane-associated putative guanylate kinases, as a result of direct binding of the carboxy-terminal cytoplasmic tails to the K+ channel subunits to two PDZ (also known as GLGF or DHR) domains in the PSD-95 protein. The ability of PDZ domains to function as independent modules for protein-protein interaction, and their presence in other junction-associated molecules (such as ZO-1 (ref. 3) and syntrophin), suggest that PDZ-domain-containing polypeptides may be widely involved in the organization of proteins at sites of membrane specialization.
The N-methyl-D-aspartate (NMDA) receptor subserves synaptic glutamate-induced transmission and plasticity in central neurons. The yeast two-hybrid system was used to show that the cytoplasmic tails of NMDA receptor subunits interact with a prominent postsynaptic density protein PSD-95. The second PDZ domain in PSD-95 binds to the seven-amino acid, COOH-terminal domain containing the terminal tSXV motif (where S is serine, X is any amino acid, and V is valine) common to NR2 subunits and certain NR1 splice forms. Transcripts encoding PSD-95 are expressed in a pattern similar to that of NMDA receptors, and the NR2B subunit co-localizes with PSD-95 in cultured rat hippocampal neurons. The interaction of these proteins may affect the plasticity of excitatory synapses.