 |
Previous Article | Next Article 
The Journal of Neuroscience, October 15, 2001, 21(20):8091-8107
RNA-Binding Protein Musashi2: Developmentally Regulated
Expression in Neural Precursor Cells and Subpopulations of Neurons in
Mammalian CNS
Shin-ichi
Sakakibara1, 2,
Yuki
Nakamura1, 3,
Hitoshi
Satoh4, and
Hideyuki
Okano1, 2, 3
1 Division of Neuroanatomy, Department of Neuroscience,
Biomedical Research Center, Osaka University Graduate School of
Medicine, Suita, Osaka 565-0871, Japan, 2 Core Research for
Evolutional Science and Technology, Japan Science and Technology
Corporation, Kawaguchi, Saitama 332-0012, Japan,
3 Department of Physiology, Keio University School of
Medicine, Shinjuku-ku, Tokyo 160-8582, Japan, and
4 Division of Pathology, Department of Cancer Research, The
Institute of Medical Science, The University of Tokyo, Minato-ku, Tokyo
108-8639, Japan
 |
ABSTRACT |
Musashi1 (Msi1) is a mammalian neural RNA-binding protein highly
enriched in neural precursor cells that are capable of generating both
neurons and glia during embryonic and postnatal CNS development. Here,
we identified Musashi2 (Msi2), a novel mammalian RNA-binding protein
that exhibits high sequence similarity to Msi1. The Msi2 transcript
appeared to be distributed ubiquitously in a wide variety of tissues,
consistent with the mRNA distribution of its Xenopus homolog, xrp1. However, the present study revealed cell
type-specific and developmentally regulated expression of Msi2 in the
mammalian CNS. Interestingly, Msi2 was expressed prominently in
precursor cells in the ventricular zone and subventricular zone with
the same pattern as Msi1 throughout CNS development. In the postnatal and adult CNS, this concurrent expression of Msi2 and Msi1 was seen in
cells of the astrocyte lineage, including ependymal cells, a possible
source for postnatal CNS stem cells. During neurogenesis, the
expression of both Msi2 and Msi1 was lost in most postmitotic neurons,
whereas Msi2 expression persisted in a subset of neuronal lineage
cells, such as parvalbumin-containing GABA neurons in the neocortex and
neurons in several nuclei of the basal ganglia. Msi2 may have a unique
role that is required for the generation and/or maintenance of specific
neuronal lineages. Furthermore, in vitro studies showed
that Msi2 and Msi1 have similar RNA-binding specificity. These two
RNA-binding proteins may exert common functions in neural precursor
cells by regulating gene expression at the post-transcriptional level.
Key words:
RNA-binding protein; Musashi2; Musashi1; CNS development; neural precursor cells; CNS stem cell; astrocyte; parvalbumin; GABAergic neuron
 |
INTRODUCTION |
A number of transcription factors
that function in the proliferation and differentiation of neural
precursor cells have been identified. However, the recent discovery of
neural-specific RNA-binding proteins raises the possibility that the
development of neural cells from the precursors may also be regulated
at the post-transcriptional level, including by mRNA stabilization or
translational control. We believe that two gene families are
represented by the neural RNA-binding proteins found in both
invertebrates and vertebrates (Okano, 1995 ). One, the Elav family,
possesses three sets of ribonucleoprotein (RNP)-type RNA recognition
motifs (RRMs) and includes the Drosophila elav gene (Yao et
al., 1993 ) and the vertebrate Hu genes (Okano and Darnell,
1997 ). Members of this family are expressed in postmitotic neurons and
are thought to function in the survival or differentiation of neurons
(Akamatsu et al., 1999 ). The other family, the Musashi (Msi) family, is
characterized by two RRMs and is conserved in ascidian (Kawashima et
al., 2000 ), nematode (Yoda et al., 2000 ), fruit fly (Nakamura et al.,
1994 ), frog (Richter et al., 1990 ), chick (Kaneko et al., 2000 ), mouse
(Sakakibara et al., 1996 ), and human (Good et al., 1998 ). In contrast
to the Elav family, the members of the Msi family are predominantly
expressed in neural precursor cells (Sakakibara et al., 1996 ; Pincus et
al., 1998 ; Kaneko et al., 2000 ). In Drosophila, Musashi
(d-Msi) is expressed in numerous precursor cells of the embryonic CNS
and in external sensory organ precursor cells (SOPs) (Nakamura et al.,
1994 ) and is required for the two successive asymmetric cell divisions
of the SOPs (Nakamura et al., 1994 ; Okabe et al., 2001 ). Although the
molecular functions of the Msi and Elav families remain obscure, their
different expression patterns suggest that the two families may play
distinct roles in the development and maintenance of neural cells.
During mammalian CNS development, neurons and glial cells are thought
to be generated from common neural precursor cells (CNS stem cells) in
the embryonic ventricular zone (VZ) (Davis and Temple, 1994 ; Johe et
al., 1996 ). We found previously that Msi1 is expressed predominantly in
proliferating multipotent neural precursors but not in newly generated
postmitotic neurons (Sakakibara et al., 1996 ; Kaneko et al., 2000 ). In
the perinatal stage, the VZ shrinks and becomes a single epithelial
layer, the ependyma. A second proliferative subventricular zone (SVZ)
then appears at the late embryonic stage and persists into adulthood,
in a considerably reduced form. It was believed for a long time that mammalian cortical neurogenesis occurs mainly in the embryonic VZ;
however, several recent studies demonstrated that the SVZ astrocytes
and/or ependymal cells correspond to a population of CNS stem cells
throughout postnatal and adult life (Scheffler et al., 1999 ).
Accordingly, substantial Msi1 protein expression persists in these
putative precursor cells into adulthood (Sakakibara and Okano,
1997 ).
In the present study we identified Msi2, a novel member of the Msi
family. Given their remarkable similarities in primary structure, RNA
binding specificity, and expression profile during CNS development,
Msi1 and Msi2 may play similar roles in neural precursor cells via
post-transcriptional regulation.
 |
MATERIALS AND METHODS |
Cloning of mouse msi2 cDNA. A gt 11 cDNA library
obtained from adult mouse cerebellum (Sakakibara et al., 1996 ) was
screened with both a 1.1 kb EcoRI fragment of the mouse
msi1 coding region (GenBank accession no. D4965) (Sakakibara
et al., 1996 ) and a 387 bp BamHI-NdeI fragment
containing the C terminus of the coding region of Xenopus
xrp1 cDNA (GenBank accession no. L02953) (Good et al., 1993 ).
Hybridization was performed against 1 × 107 plaques in a buffer (1 M
NaCl, 1% SDS, 10% dextran sulfate, 0.1 mg/ml salmon sperm DNA)
containing 5 × 105 cpm/ml of
32P-labeled random-primed probe for 18-24
hr at 60°C with the msi1 probe or at 55°C with the
xrp1 probe. The hybridized filters were washed twice for 20 min each at room temperature in 2× SSC and 0.1% SDS (low stringency).
Thirty-two positive clones that hybridized to both the msi1
and xpr1 probes were obtained. Among them, nine that
strongly hybridized to the xpr1 cDNA were selected,
subcloned into pBluescriptII (Stratagene, La Jolla, CA), and sequenced
by standard dideoxynucleotide sequencing using a dye-primer kit
(Amersham Pharmacia Biotech, Buckinghamshire, UK). Sequencing analysis
revealed multiple overlapping clones spanning a 0.5 kb 5'-UTR, a 0.8 kb 3'-UTR, and the 1.0 kb predicted open reading frame (ORF), which we
designated mouse musashi2 (msi2). A putative
alternatively spliced exon was found at the C terminus of the predicted
coding region with a 54 bp insertion, producing short- and long-form ORFs of msi2 cDNA. Expression of the short and long forms of
msi2 transcripts in vivo was confirmed by RT-PCR
analysis of RNA isolated from embryonic day (E) 12 and adult brains.
Homology searches and alignments were performed with BLAST and
FASTA algorithms on the National Center for Biotechnology
Information server. The GenBank accession numbers of the
proteins used for phylogenetic tree analysis (clustalW program
on the DNA Data Bank of Japan www server) are as follows: human (Hu)
U1snRNP70K (A25707), mouse (Mus) heterogeneous nuclear
ribonucleoprotein (hnRNP) A1 (NP 034577), Hu hnRNP A0 (Q13151), Mus
hnRNP A2/B1 (O88569), Hu hnRNP A3 (P51991), Hu hnRNP type A/B
(AAA36575), Rat AUF1 (BAB03468, BAB03466, BAB03467), Mus hnRNP C1/C2
(AAD03717), Hu hnRNP F (S43484), Mus hnRNP G (O35479), Hu hnRNP H
(I39358), Mus PTB (hnRNP I) (P17225), Mus brPTB (NP 062423), Hu
hnRNP L (P14866), Hu hnRNP M (P52272), Hu hnRNP R (T02673), Mus hnRNP U
(NP 058085), Mus TIA-1 (P52912), Mus TIAR (S72436), Mus HuR (NP
034615), Mus HuB (AAC52644), Mus HuC (Q60900), Mus HuD (JC2298),
Hu Brunol3 (AAB09040), Mus Lark (NP 033058), and Rat La/SS-B (JC1494).
Animals and tissue preparation. ICR (CD-1) mice, used for
the preparation of tissue protein extracts, RNA, or tissue sections, were obtained from Charles River Japan Inc. (Kanagawa, Japan). The date
of conception was established by the presence of a vaginal plug and
recorded as E0, and the day of birth was designated as postnatal day
(P) 0.
Northern blot analysis. Total RNA (20 µg) from each of the
mouse tissues and embryos was isolated using Trizol (Life
Technologies-BRL, Grand Island, NY) according to the manufacturer's
instructions, fractionated by electrophoresis on 1%
agarose-formaldehyde gels, and transferred onto Biodyne B nylon
membranes (Pall, Port Washington, NY).
32P-labeled probe was prepared using a
random-primed DNA labeling kit (Roche Diagnostics, Mannheim, Germany)
of a 750 bp 3'-UTR fragment of the mouse msi2 cDNA and
hybridized in 50% formamide, 6× SSPE, 5× Denhardt's solution, 0.5%
SDS, and 200 µg/ml salmon sperm DNA at 42°C for 16 hr. The
incubation was followed by stringent washes in 0.1× SSC, 0.1% SDS at
50°C, and filters were exposed for 48 hr on Kodak X-OMAT film (Kodak,
Rochester, NY). The integrity of the transferred RNA samples was
verified by reprobing each blot with a radiolabeled
-actin cDNA fragment (Clontech, Palo Alto, CA).
Production and purification of anti-Msi2 antibodies. A
peptide corresponding to the 14 amino acid terminal sequence of the Msi2 protein (MEANGSPGTSGSAN) followed by a cysteinamide residue was
synthesized for the purpose of antibody production. This peptide sequence does not overlap with that of the RNA-binding domain, RRM1
(see Fig. 1) and does not show any similarities to the corresponding N-terminal region of the Msi1 protein. Approximately 15 mg of the
peptide was coupled through the cysteinamide residue to
m-bromosuccinimide-treated keyhole limpet hemocyanin and
used to immunize New Zealand White rabbits. For the affinity
purification of the anti-Msi2 antisera, the synthesized peptide (5 mg)
was coupled covalently to activated 2-fluoro-1-methylpyridinium
toluene-4-sulfonate (FMP) Cellulofine (Seikagaku Kogyo, Tokyo,
Japan), according to the manufacturer's instructions. Ten
milliliters of the filtered (0.45 µm) whole antisera were
incubated with 3 ml of the peptide-FMP Cellulofine affinity resin,
pre-equilibrated with TBS buffer (0.15 M NaCl, 20 mM Tris-HCl, pH 7.5) overnight at 4°C. The
resin was then washed with 50 ml of 1 M NaCl, 1%
Triton X-100, 20 mM Tris-HCl, pH 7.5, followed by
20 ml of 0.15 M NaCl, eluted with 4 ml of 100 mM glycine-HCl, pH 2.0, at 4°C, and immediately
neutralized with 0.2 ml of 1 M Tris.
Recombinant Msi1 and Msi2 proteins. The 983 bp and 1072 bp
BamHI-EcoRI fragments corresponding to the
short- and long-form ORFs of the Msi2 protein were isolated by RT-PCR
of the RNAs of E12 embryonic and adult brains and subcloned in-frame
into the expression vector pRSET-A (Invitrogen, Carlsbad, CA) to
construct pRSET-Msi2S (short form) and pRSET-Msi2L (long form) to
generate fusion proteins containing six histidine residues at their
amino termini. Expression vectors pRSET-Msi1 (Sakakibara et al., 1996 ), pRSET-Msi2S, and pRSET-Msi2L were then introduced into
Escherichia coli strain BL21(DE3) pLysS, and the fusion
proteins were induced by incubation with 1 mM
isopropyl- -D-thiogalactopyranoside for 6 hr at
30°C. Recombinant fusion proteins (His6-Msi2S,
His6-Msi2L, and His6-Msi1)
were affinity purified using a Probond resin (Invitrogen) column as
described by the supplier. The purity and concentration of the fusion
proteins were verified by Coomassie Brilliant blue (Sigma, St. Louis,
MO) staining of an SDS-polyacrylamide (SDS-PAGE) gel of the
eluent, and by the Bradford assay (Bio-Rad, Hercules, CA).
Protein phosphatase treatment and immunoblotting. Tissue
lysate was prepared by homogenization in Buffer A [50
mM Tris-HCl, pH 7.6, 1 mM
potassium acetate, 1.5 mM magnesium acetate, 2 mM dithiothreitol (DTT), 100 µg/ml
phenylmethylsulfonyl fluoride (PMSF), 5 µg/ml aprotinin, 5 µg/ml
leupeptin], followed by centrifugation at 10,000 × g
for 10 min. Bacterially expressed, purified recombinant proteins (50 ng
per lane) or tissue lysate (30 µg of protein per lane) were resolved
on 10% SDS-PAGE gels that were then electroblotted onto Immobilon-P
membranes (Millipore, Bedford, MA) with a semidry transfer apparatus.
Equal loading of total protein from each tissue was determined by
standard Bradford assay and verified by Coomassie blue staining of a
duplicate gel. The chemiluminescent signals were detected by ECL
(Amersham Pharmacia Biotech) with Kodak X-OMAT film according to the
manufacturer's instructions. For the protein dephosphorylation assay,
endogenous Msi2 protein was partially purified from embryonic brain
lysate (E12.5). E12.5 brains (1.0 gm of wet weight) were homogenized in
5 ml of Buffer A and spun to pellet the nuclei (12,000 × g, 10 min, 4°C). The supernatant was removed and layered
gently over a 1.5 ml sucrose (30% w/v in Buffer A) cushion and spun at
130,000 × g, 4°C for 2 hr, in a Beckman SW55Ti
rotor. After removal of the S130 supernatant and sucrose cushion, the
pelleted polysome fraction was rinsed and resuspended in 500 µl of
Buffer A. A significant amount of Msi2 protein was recovered in this
polysome fraction. For protein phosphatase treatment, 10 µg of
protein from the cleared polysome fraction was incubated with 800 U of
recombinant protein phosphatase ( PPase) (New England BioLabs,
Beverly, MA) in 25 µl of 50 mM Tris-HCl, pH
7.5, 0.1 mM EDTA, 5 mM DTT,
0.01% polyoxyethylene (23) lauryl ether (Brij35), 2 mM MnCl2, 10 µg/ml PMSF,
5 µg/ml aprotinin, and 5 µg/ml leupeptin, at 30°C for 1 hr. The
PPase dephosphorylates phosphorylated serine, threonine, and
tyrosine residues. Control samples lacking PPase were incubated as
described above. Reactions were stopped with SDS-PAGE sample buffer,
and the samples were resolved on 10% SDS-PAGE gels for immunoblotting.
In vitro transcription/translation and RNA-binding
assays. The cDNA fragments corresponding to the coding region of
the Msi2 long form (nucleotides 524-1564; GenBank accession no.
AB056103) and Msi1 (nucleotides 64-1152; GenBank accession no. D49654) were isolated by PCR, using primers encoding a FLAG tag at the amino
termini, and subcloned into pCDNA3 (Invitrogen) to construct the
expression vectors pCDNA-msi2 and pCDNA-msi1. These plasmids were
transcribed and translated in rabbit reticulocyte lysate (TNT T7
quick-coupled transcription/translation system; Promega, Madison, WI)
in the presence of 0.4 mCi/ml of
35S-methionine (Amersham Pharmacia
Biotech), according to the manufacturer's suggested conditions.
Luciferase T7 control vector (Promega), which encodes a 61 kDa
Luciferase protein, was also translated in vitro as
described above. Binding of in vitro-translated proteins to
RNA-homopolymers was performed essentially as described previously (Swanson and Dreyfuss, 1988 ), with minor modifications. Briefly, 20 µl of each ribohomopolymer-agarose bead (Sigma), equilibrated with
the binding buffer (10 mM Tris-HCl, pH 7.4, 2.5 mM MgCl2, 0.5% Triton
X-100, 2 mg/ml pepstatin, 2 mg/ml leupeptin, 0.5% aprotinin, 1 mg/ml
heparin), was incubated with the
35S-labeled proteins (1 × 105 cpm) in 500 µl of binding buffer
containing either 100 or 250 mM NaCl for 15 min
on a rocking platform at 4°C. The beads were pelleted with a brief
spin and washed five times with 500 µl of binding buffer, before
resuspension in 50 µl of SDS-PAGE loading buffer. Bound proteins were
eluted by boiling, resolved on a 10% SDS-PAGE gel, and visualized by fluorography.
Immunohistochemistry. Pups and 8-week-old adults
(anesthetized by ether inhalation) were perfused through the left
ventricle with 4% paraformaldehyde (PFA) in 0.1 M PBS, pH 7.4. The embryos were removed by
cesarean section and immersed in the same fixative. Brains and other
tissues were dissected and post-fixed overnight at 4°C, cryoprotected
in 30% sucrose in PBS overnight at 4°C, and embedded in O.C.T.
compound (Tissue Tek; Miles, Elkhart, IN). Cryostat sections (12 µm)
were cut and affixed to 3-aminopropyltriethoxysilane-coated glass
slides (Matsunami Glass, Osaka, Japan). Sections were permeabilized in
0.4% Triton X-100 in PBS for 30 min, followed by blocking with 10%
normal serum of the species in which the secondary antibody was raised,
1% skim milk, and 0.1% Triton X-100 in PBS for 1 hr at room
temperature; subsequently, the sections were incubated in a mixture of
primary antibodies overnight at 4°C. The mixture always included the
affinity-purified rabbit anti-Msi2 polyclonal antibody, diluted 1:300
in 10% normal serum and 0.1% Triton X-100 in PBS, and one of the
following antibodies: anti-Msi1 (14H1, rat monoclonal
IgG2b, ascites) (Kaneko et al., 2000 ), diluted 1:500; anti-Hu proteins (mouse monoclonal IgG2b,
clone 16A11, which binds to Hu proteins including HuD, HuC, and HuB)
(Marusich et al., 1994 ), diluted 1:500; anti-nestin (Rat401, mouse
monoclonal IgG, Developmental Studies Hybridoma Bank of Iowa
University), diluted 1:250; anti-Ki67 (mouse monoclonal IgG,
Novocastra, Newcastle upon Tyne, UK), diluted 1:500; anti-glial
fibrillary acidic protein (GFAP) (mouse monoclonal
IgG1, Sigma), diluted 1:400; anti- 2', 3'-cyclic
nucleotide-3'-phosphohydrolase (CNPase) (mouse monoclonal IgG1, Sigma), diluted 1:100;
anti-microtuble-associated protein 2 (MAP2) (mouse monoclonal
IgG1, Sigma), diluted 1:500; anti- III tublin
(TuJ1) (Berkeley Antibody; mouse monoclonal
IgG2a/k), diluted 1:200; anti-parvalbumin (PV)
(mouse monoclonal IgG1, Chemicon, Temecula, CA),
diluted 1:500; anti-GABA (guinea pig polyclonal, Chemicon), diluted
1:400; anti-acetylcholine transferase (ChAT) (goat polyclonal antibody,
Chemicon), diluted 1:100; anti-Calbindin (CB) D28K (mouse monoclonal
IgG1, Sigma), diluted 1:1000; anti-tyrosine hydroxylase (TH) (mouse monoclonal IgG1, provided
by Dr. H. Hatanaka, Osaka University), diluted 1:1000; anti-Mac1 (mouse
monoclonal IgG1, Roche Diagnostics), diluted
1:100; or anti-bromodeoxyuridine (BrdU) (mouse monoclonal
IgG1, Sigma), diluted 1:5000. After four washes
with 0.1% Triton X-100 in PBS, the sections were incubated for 1 hr in
a mixture containing the following secondary antibodies diluted 1:500
in the same solution as the primary antibodies: Alexa Fluor
568-conjugated goat anti-rabbit IgG (Molecular Probes, Eugene, OR) and
Alexa Fluor 488-conjugated anti-mouse, rat, guinea pig, or goat IgG
(Molecular Probes). After being rinsed in PBS, the sections were
mounted and examined under a fluorescence microscope (Zeiss Axioplan 2)
equipped with the appropriate epifluorescent filters. Optical sections
were viewed using a scanning laser confocal imaging system (Zeiss
LSM510). For control sections, the anti-Msi2 antibody was omitted or
replaced with preimmune rabbit serum. The specificity of the anti-Msi2
antibody was examined by its preadsorption with the peptide (50 µg/ml) used for immunization, before its use for immunostaining
tissue sections.
Cell culture. Neurosphere cultures were prepared as
described previously (Nakamura et al., 2000 ). Briefly, cells from the frontal halves of E14.5 telencephalons were used for primary sphere formation (5 × 105 cells/5 ml per well, six-well plate) with the basic culture medium, DMEM/F-12 (1:1) (Life Technologies-BRL) containing 25 µg/ml insulin, 100 µg/ml transferrin, 20 nM progesterone, 60 µM putrescine, 30 nM sodium selenite, 20 ng/ml epidermal growth factor (EGF), and 10 ng/ml basic fibroblast growth factor (bFGF). After
cultivation for 10 d, the neurospheres were fixed in 4% PFA for
30 min at 4°C and embedded in O.C.T. compound for cryosectioning. NG108-15, a hybrid cell line of mouse neuroblastoma and rat glioma (Amano et al., 1974 ), was grown in DMEM (Life Technologies-BRL) with
10% (v/v) fetal bovine serum. Cells were grown to a density of 5 × 105
cells/cm2 on coverslips precoated with
0.01% poly-L-lysine (Sigma), fixed 4% PFA for
30 min at 4°C, and permeabilized in 0.05% Triton X-100 in PBS for 10 min. Immunoreactivity to Msi1 or Msi2 was visualized using
rhodamine-conjugated donkey anti-rabbit IgG or anti-rat IgG secondary
antibodies (Chemicon; 1:250 dilution). Cells were counterstained with
10 µM of Hoechst 33342 dye (Sigma) to identify nuclei. Optical images were obtained using a Zeiss LSM410 scanning laser confocal system.
Surgical procedures for brain lesions and bromodeoxyuridine
incorporation. Brain injury experiments were performed as
described previously (Sakakibara and Okano, 1997 ). Adult ICR mice, 2 months old, were anesthetized with Nembutal injections (10 mg/kg)
(Abbott Laboratories, North Chicago, IL), and a midline skin incision was made. A unilateral linear craniectomy was performed on the skull
using a drill, 3 mm left of the midline. The wound was made in the left
cerebral hemisphere; a 27 gauge stainless-steel needle was placed 3 mm
lateral to the sagittal fissure and 2 mm anterior to the lambda line,
and inserted through the pia to a depth of 1.5 mm. The brain was cut
sagittally, 4 mm anterior to the lambda line. The breaks in the skull
were filled with Bonewax, and the wound was closed with sutures. These
lesioned animals showed no obvious behavioral or motor deficits. At
4 d after surgery, the lesioned animals were given two
intraperitoneal injections of BrdU (100 mg/kg of body weight; Sigma)
over a 2 hr period and were again anesthetized and killed by perfusion
2 hr after the second injection. Immunostaining with anti-BrdU antibody
was performed as described previously (Sakakibara and Okano, 1997 ).
Chromosomal mapping of mouse msi1 and msi2. Mouse 129 sv/J
genomic DNA fragments of mouse msi1 (11.5 kb) and mouse
msi2 (11 kb), each containing the ATG translation initiation
codon, were subcloned into the pBluescriptII (Stratagene) and pZErO-1
(Invitrogen) plasmid vectors, respectively, and used for fluorescence
in situ hybridization (FISH) as described previously (Satoh
et al., 1993 ). Each of the plasmids was labeled with biotin-16-dUTP or
digoxygenin-11-dUTP using a nick translation kit (Roche Diagnostics)
and hybridized to metaphase chromosomes from mouse 129 embryonic stem
cells. On each slide, 1 µg of labeled DNA was applied with a 5- to
10-fold excess amount of mouse Cot-1 DNA (Life Technologies-BRL) to
block nonspecific hybridization to the repetitive sequences. The
hybridization signals were visualized with FITC-avidin (Roche
Diagnostics) or rhodamine-conjugated anti-digoxigenin (Vector Labs,
Burlingame, CA). Chromosomes were counterstained with
4',6-diamidine-2'-phenylindole dihydrochloride (DAPI) to identify
individual chromosomes. Photographs were taken of the FISH signals
superimposed on the images of DAPI-banded chromosomes.
GenBank accession numbers. The nucleotide
sequence data reported in this paper have been deposited with the GSDB,
DDBJ, EMBL, and NCBI under accession numbers AB056102 for
msi2S and AB056103 for msi2L.
 |
RESULTS |
Identification of msi2 and characterization of its
primary structure
To isolate the cDNA of a novel mammalian RNA-binding protein that
was homologous to Msi1, a mouse nervous system cDNA library was
screened using mouse msi1 and Xenopus
xrp1 cDNA probes at reduced stringency. The Xenopus
xrp1 gene (Good et al., 1993 ) encodes a protein that is related by
sequence to the NRP1 protein, a Xenopus homolog of Msi1
(Sakakibara et al., 1996 ). The largest DNA insert obtained was used as
a probe to rescreen the cDNA library at high stringency to obtain
full-length cDNA. The longest and single open reading frame, encoding a
346 amino acid protein with a predicted molecular mass of 37 kDa, was
identified from nine overlapping cDNAs (Fig.
1A). Sequence analysis
indicated that the gene product encoded by the cDNAs was a novel
RNA-binding protein. We designated this Msi1-related gene as mouse
musashi2. Multiple cDNA clones obtained from the library
screening and RT-PCR analysis of msi2 transcripts from RNA
isolated from E12 and adult mouse brains (data not shown) indicated
that there were two alternatively spliced transcripts depending on the
presence or absence of a short segment (18 amino acids) within the
C-terminal half of Msi2 (Fig. 1A, broken
underline). This alternative splicing could generate two Msi2
protein isoforms with predicted molecular masses of 36.9 and 35.7 kDa
(termed Msi2L and Msi2S, respectively).

View larger version (50K):
[in this window]
[in a new window]
|
Figure 1.
Primary structure and chromosomal localization of
Msi2 and Msi1. A, The deduced amino acid sequence of
mouse Msi2 protein. The two RNA recognition motifs are
underlined. Each RRM includes two highly conserved
sequences designated RNP-1 and RNP-2
(boxed sequences). The broken underlines
indicate the regions that are missing in the form arising from
alternative mRNA splicing. Asterisks denote the peptide
sequence used as an immunogen to generate anti-Msi2 antibody.
B, Domain structures of the mouse Msi2 and Msi1
proteins. The percentages of amino acid identity between Msi1 and Msi2
within each RRM and in the C-terminal half are shown. Bulged
small boxes represent the regions arising from alternative
splicing. C, Amino acid sequence alignment of mouse Msi2
with its known homologs, Xenopus RNA-binding protein
(XRP1), and mouse Msi1. The two RRMs are boxed, and gaps
in the alignment are indicated by dashes. Amino acid
residues conserved in more than two proteins are shaded.
D, Dendrogram showing a multiple sequence comparison of
various mammalian RRM-containing RNA-binding proteins. Sequences
derived from selected proteins were compared using the clustalW program
on the DDBJ www server. The branch lengths are proportional to the
differences between sequences. The accession numbers of each protein
are given in Materials and Methods. E, Chromosomal
localization of the mouse msi1 and msi2
genes by FISH. The left panels show metaphase
chromosomes stained with DAPI for the identification of individual
mouse chromosomes. Chromosomes 11 and 5
are numbered. The right panels represent
the same metaphase chromosomes with the hybridization signals from the
msi2 (top panels) and msi1
(bottom panels) genes (arrows indicate
each signal). The closed bars on the right side of each
ideogram indicate the possible location of the msi2 and
msi1 genes.
|
|
Many RNA-binding proteins contain a conserved sequence of 80-90 amino
acids, termed an RRM, that includes two short, highly conserved
core domains called RNP-1 and RNP-2 (Burd and Dreyfuss, 1994 ) (Fig.
1A). Msi2 contains two RRMs (RRM-1 and -2) that
contain well conserved RNP-1 consensus sequences,
(K/R)G(F/Y)(G/A)FVX(F/Y) (Burd and Dreyfuss, 1994 ). RNP-2 is less
conserved than RNP-1 but is characterized by its aliphatic and aromatic
nature and its relative position to RNP-1 (Burd and Dreyfuss, 1994 ).
Among a number of known RNA-binding proteins, a homology search against the expressed sequence tags (EST) and GenBank databases revealed that
Msi2 shared remarkable sequence similarity with Xenopus XRP1 and mouse Msi1, a neural RNA-binding protein (Fig.
1B-D). Msi2 and Msi1 showed significant
sequence similarity (75% amino acid identity in the overall
structure). In particular, the RRM regions of Msi2 showed 85 and 95%
identity to those of Msi1 and XRP1, respectively (Fig.
1B). The amino acid sequences of RNP-1 and RNP-2,
which are critical for binding to RNA substrates, were completely
conserved among these three proteins (Fig. 1C). This raises
the possibility that Msi2 may bind the same RNA molecules that are
recognized by Msi1 or XRP1. On the other hand, Msi2 did not show
extensive similarity to the RRMs of any other RNA-binding proteins,
including Drosophila Elav (Yao et al., 1993 ) and the vertebrate Hu proteins (HuB, HuC, HuD, and HuR) (Okano and Darnell, 1997 ). In the C-terminal tail of the protein, which does not contain the RRMs, Msi2 showed 67% amino acid identity with Msi1 (Fig. 1B). The major difference between Msi2 and XRP1 was a
58 amino acid segment in the C-terminal half of XRP1, resulting in a
length of 346 amino acids for Msi2 and 406 amino acids for XRP1 (Fig. 1C). Except for the two RRM domains, no other functional
domains or consensus sequences were identified in the putative Msi2
protein, as is also the case with Msi1 and XRP1. A short alanine-rich
stretch, the function of which is not known, is present in the
C-terminal tail of Msi2, Msi1, and XRP1 (Fig. 1C).
The relationship of the Msi2 and Msi1 proteins to other mammalian
RRM-containing proteins was demonstrated by the pairwise comparisons
presented in Figure 1D. Msi2 and Msi1 formed a unique subgroup within the hnRNP A/B class of proteins, the heterogeneous nuclear ribonucleoproteins that contain two copies of RRMs in addition
to auxiliary domains. Among this class, hnRNP type A/B protein (Khan et
al., 1991 ) and AUF1 (hnRNP D) (Zhang et al., 1993 ) were found to be
somewhat related to this Msi subgroup (Fig. 1D). AUF1
binds the U-rich elements of the 3'-UTR found in many proto-oncogene
and cytokine mRNAs and regulates the stability of these mRNAs (DeMaria
and Brewer, 1996 ; Blaxall et al., 2000 ).
To determine whether msi2 mRNA is expressed in a
tissue-specific manner, we performed Northern blotting of the RNA from
various tissues from adult mice, probing with a 750 bp 3'-UTR fragment of mouse msi2 cDNA (Fig.
2A). Substantial levels
of a major msi2 transcript of 7.1 kb was detected in all
tissues tested, with the highest expression observed in the cerebellum
and smaller amounts in the liver and small intestine. A 1.5 kb short
form of the msi2 transcript was detected only in the testis.
Northern blots of RNA from different developmental stages indicated the continuous expression of the msi2 transcript in the CNS
throughout development (Fig. 2B). This tissue
distribution and developmental profile of the msi2
transcripts paralleled the distribution profile reported for
Xenopus xrp1 mRNA (Good et al., 1993 ).

View larger version (64K):
[in this window]
[in a new window]
|
Figure 2.
Tissue distribution and developmental expression
of the msi2 transcripts. Total RNA (20 µg) from each
mouse tissue was prepared, and the levels of msi2 mRNA were
determined by Northern blot analysis. A, Tissue
distribution of msi2 mRNA in adults. Expression of a 7.0 kb transcript can be seen in most lanes. A short transcript of ~1.5
kb is observed only in the testis. B, Changes in
msi2 mRNA expression in the whole brain associated with
development. Msi2 mRNA expression can be seen at E10,
the earliest time point tested. The equal loading and quality of the
RNAs were tested by performing the hybridization with a
-actin probe, shown in each bottom
panel.
|
|
On the basis of the striking similarity of both the predicted protein
sequence and the expression pattern, we concluded that the
msi2 gene was a mouse homolog of Xenopus xrp1.
Msi1 and Msi2 seem to constitute a novel RNA-binding protein family
(Msi family) in rodents, although they show tissue distributions that
are different from each other.
Chromosomal localization of the msi1 and msi2
genes
Genomic Southern blot analysis using probes from msi1
and msi2 genomic fragments yielded the predicted single
bands for several appropriate restriction enzymes, indicating that both
are likely to be single-copy genes. The cytogenetic locations of the
msi1 and msi2 genes were determined by FISH. As
shown in Figure 1E, the msi1 and
msi2 genes were mapped to chromosome 5qE3-F and 11qB5-C, respectively, by DAPI bands. Our previous FISH and somatic cell hybrid
panel analyses (Good et al., 1998 ) localized the human msi1
gene to chromosome 12q24.1-24.31, which corresponds to the syntenic
region of the mouse msi1 locus. The msi2 gene was
predicted to exist on chromosome 17 q11-21 in humans by the
comparative map (Mouse Genomic Informatics database of the Jackson
Laboratory). Indeed, search of the human genome database revealed a
partial nucleotide sequence of the human msi2 gene on
chromosome 17 (covered by a 165 kb BAC clone; GenBank accession
no. AC005325), indicating the conservation of the msi2 and
msi1 genes across species, including Xenopus,
mouse, and human. In human, this msi2 syntenic region contained several loci of hereditary neurological disorders:
frontotemporal dementia with parkinsonism (FTDP17) (Spillantini and
Geodert, 1998 ), Pick disease (Munoz-Garcia and Ludwin, 1984 ), familial progressive subcortical gliosis (GPSC) (Petersen et al., 1997 ), and
neurofibromatosis type 1 (von Recklinghausen disease, NF1) (Mukonoweshuro et al., 1999 ).
RNA-binding properties of Msi2 protein
To test whether Msi2 could actually bind RNA, and if so, whether
Msi1 and Msi2 recognize common RNA targets, we performed RNA-binding
analysis in vitro. The full-length Msi2 and Msi1 proteins were prepared by in vitro transcription/translation, and
their ability to bind RNA-homopolymers was determined. Both the Msi2 and Msi1 proteins showed strong binding to poly(U), weaker
binding to poly(G), and no binding to poly(A) or poly(C), under normal stringency conditions in a buffer containing 100 mM NaCl (Fig. 3A). A 61 kDa luciferase
polypeptide showed no binding to any of the
homopolymers, confirming the RNA-binding activity of the Msi2 and
Msi1 proteins. Furthermore, binding experiments performed under various
degrees of stringency, up to 1.0 M NaCl, revealed that the Msi1 and Msi2 proteins both bound poly(U) and poly(G). The
binding to poly(U) was stable in 250 mM NaCl,
although the binding to poly(G) was only seen at 100 mM NaCl (Fig. 3B). The binding
properties of Msi1 observed in this study were slightly different from
our previous report, in which we found, using a Northwestern
assay, that Msi1 binds poly(G) more efficiently than poly(U)
(Sakakibara et al., 1996 ). This discrepancy might be attributable to
the different binding assays used in the two studies. Our previous Northwestern binding assay used a bacterially expressed, denatured, and
immobilized Msi1 polypeptide. We believe that the binding properties
observed in this report, in which soluble Msi1 protein produced from
in vitro transcription/translation was used, are more
representative of the in vivo binding capacity of Msi1.
Taken together, these observations indicated that Msi2 is an
RNA-binding protein with in vitro RNA-binding specificity
similar to that of Msi1. The Msi family proteins may share the same or
similar RNA(s) characterized by uridine-rich sequences as their
downstream target molecules in vivo.

View larger version (57K):
[in this window]
[in a new window]
|
Figure 3.
Similar RNA-binding properties of Msi2 and Msi1
proteins. A, Binding of the full-length Msi2 and Msi1
proteins to ribonucleotide homopolymers.
35S-methionine-labeled, in vitro-translated
Msi2, Msi1, and Luciferase (as a control protein) were
incubated with the indicated ribonucleotide homopolymer beads in buffer
containing 100 mM NaCl. Bound proteins were visualized by
SDS-PAGE and fluorography as described in Materials and Methods.
Ribonucleotide homopolymers and proteins used in the assays are
indicated at the top of each lane. Lanes
marked Input contain each 35S-labeled
protein used for binding reactions. The molecular sizes of markers are
indicated on the left. Both Msi2 and Msi1 bound strongly
to poly(U) and weakly to poly(G). Luciferase showed no binding to RNA
homopolymers. B, Binding profile of Msi2 and Msi1 to RNA
homopolymers under stringent conditions. Binding experiments of
in vitro-translated Msi2 and Msi1 with poly(U) and
poly(G) beads were performed as described above at the indicated NaCl
concentrations. In the buffer containing 0.25 M NaCl, Msi2
and Msi1 binding to poly(G) RNA was greatly diminished, whereas that to
poly(U) remained significant. Note that the binding properties of Msi2
to poly(U) and poly(G) in the given salt concentrations were almost
identical to those of Msi1.
|
|
Specificity of the Msi2 antibody
To localize the Msi2 protein in tissues, polyclonal antiserum
(anti-Msi2) was generated in rabbits using a peptide corresponding to
the amino-terminal 14 amino acids of mouse Msi2, in which no obvious
homology to Msi1 could be seen (Fig.
1A,C). As shown in Figure
4A, the anti-Msi2
antiserum recognized both recombinant full-length Msi2L and Msi2S
proteins but did not cross-react with Msi1 protein. Conversely, the
anti-Msi1 monoclonal antibody (14H1) (Kaneko et al., 2000 ) detected
Msi1 protein exclusively and not Msi2. In Western blots of embryonic
(E12) and postnatal (P3) brains, the anti-Msi1 antibody recognized a
doublet of ~37-40 kDa (Figs. 1B,
4B), which reflects alternative splicing products
(Sakakibara et al., 1996 ). The Msi2 protein was also detected as two
major polypeptides, with electrophoretic mobilities of 35 and 37 kDa (Fig. 4B, two arrows), that were likely to
correspond to Msi2S and Msi2L, respectively. Both the 35 and 37 kDa
proteins appeared to be composed of a closely spaced doublet (Fig.
4B). These doublets were probably caused by the
phosphorylation of Msi2L and Msi2S in vivo, because
treatment of the partially purified endogenous Msi2 with protein
phosphatase collapsed these doublets into single, faster-migrating
bands (Fig. 4C). The Msi2 sequence is rich in serine and
threonine residues (12%), many of which are potential phosphorylation
sites for casein kinase I and II, GSK3, or PKC. Immunoblot analysis of
msi2 knock-out mice verified that all these bands were
derived from the msi2 gene (our unpublished results). Together, these results established the specificity of the anti-Msi1 and anti-Msi2 antibodies.

View larger version (38K):
[in this window]
[in a new window]
|
Figure 4.
Western blot analysis and cytoplasmic localization
of Msi1 and Msi2 proteins. A, Specificity of the
anti-Msi2 and anti-Msi1 antibodies. Bacterially expressed full-length
Msi2S (lane 1), Msi2L (lane2), and Msi1
(lane 3) proteins were subjected to immunoblotting.
Membranes were probed with either affinity-purified rabbit polyclonal
anti-Msi2 or rat monoclonal anti-Msi1 (14H1) antibodies. The anti-Msi2
antibody specifically detected Msi2L and Msi2S and did not cross-react
with Msi1 protein. Conversely, the anti-Msi1 antibody recognized only
the Msi1 protein. The left panel shows the blot stained
with Amido black dye, confirming the integrity and equal loading of
each fusion protein. Lane M indicates protein size
markers. B, Changes in Msi2 and Msi1 protein expression
during brain development. A Western blot of protein extracts from E12
and P3 brains (30 µg protein per lane) was probed with anti-Msi1 or
anti-Msi2 antibodies. The anti-Msi1 antibody recognized a doublet at
37-40 kDa (arrows). Msi2 was detected as two major
proteins of 35 and 37 kDa (arrows), and each protein
consisted of a closely spaced doublet. C, Msi2 is a
phosphoprotein. Western blot analysis for Msi2 in the polysome fraction
from E12.5 brain was treated with (+) or without ( ) protein
phosphatase. Incubation with protein phosphatase produced a downward
shift of the slower migrating form in each doublet. D,
Confocal images of the neural cell line NG108 stained with anti-Msi1
(a) or anti-Msi2 (c),
showing their cytoplasmic localization. b, d, The same
fields, counterstained with Hoechst33342 dye to identify the individual
nuclei, are shown. Scale bar (shown in d), 8 µm.
|
|

View larger version (80K):
[in this window]
[in a new window]
|
Figure 5.
Msi2 expression in neural precursor cells residing
in the embryonic VZ and postnatal SVZ. Double-immunofluorescence
labeling of coronal sections through embryonic, postnatal, and adult
brains, with antibodies against Msi2 and Msi1, Ki67, Nestin, or MAP2.
A/B, C/D,
E/F, G/H,
I/J, K/L,
M/N, and
O/P represent each pair of double-stained
photomicrographs. The primary antibodies used are indicated in each
panel. A-F, Telencephalon at E10. Msi2
and Msi1 are uniformly distributed in the Ki67- and Nestin-positive
neuroepithelium. G/H, Cerebral cortex at E12.
Insets indicate higher magnifications of the VZ cells.
I-L, Cerebral cortex at E14.
Coincidental expression of Msi2 and Msi1 is seen in the VZ of the telencephalon and ganglionic eminences. Note the
absence of Msi2 expression in the intermediate zone and cortical plate,
which are occupied by MAP2-positive immature neurons.
M/N, The SVZ region surrounding the lateral ventricle in
P7 postnatal forebrain. O/P, Adult ependyma and SVZ
lining the lateral ventricle. Lateral is to the right,
and dorsal is up. Many densely packed cells that are
brightly immunostained with both Msi2 and Msi1 are observed in the P7
SVZ, adult SVZ, and ependymal layer. Small scattered astrocytes in the
parenchyma are also labeled with anti-Msi2 and anti-Msi1 antibodies.
The arrow in O/P indicates a
Msi2-positive but Msi1-negative neuron-like cell. Scale bars:
A-H, K, L,
50 µm; I, J, M,
N, 90 µm; O, P, 25 µm;
insets in G and H, 8 µm.
Asterisks indicate lateral ventricle. cp,
Cortical plate; vz, ventricular zone;
lge, lateral ganglionic eminence; cc,
corpus callosum; str, striatum.
|
|
Subcellular distribution of the Msi2 and Msi1 proteins
The Msi2 and Msi1 proteins were localized exclusively to the
cytoplasm of several neural cell lines, including NG108 (Fig. 4D) and PC12 (data not shown) cells. Cytoplasmic
expression of Msi2 and Msi1 was also observed in E10 neuroepithelial
cells (Figs. 5G,H,
insets, 6C). In
addition, Msi2 was present in the cytoplasm of cells in non-neural
tissues (data not shown) and in differentiated neurons, in which Msi2
protein was found in proximal regions of the somata rather than in the
dendritic arborizations or axons (see Figs. 8G,
9G). During subcellular fractionation of E12.5 brain
lysate, a significant amount of Msi2 was recovered in the polysome
fraction (Fig. 4C). Consistently, our observations by subcellular fractionation (Imai et al., 2001 ) as well as immunoelectron microscopy (our unpublished results) showed Msi1 protein to be associated with the free and membrane-bound polysomes in the embryonic VZ cells (our unpublished results), also supporting the notion that the
Msi family plays a role in regulating RNA metabolism in the cytoplasm.
However, it remains possible that Msi family proteins also participate
in nuclear events, such as mRNA splicing. In fact, we previously
observed the nuclear localization of Msi1 in a small population in
cultured astrocytes, although the significance of this observation
remains unclear (Kaneko et al., 2000 ).

View larger version (177K):
[in this window]
[in a new window]
|
Figure 6.
Msi2 expression in the spinal cord.
Double-immunofluorescence labeling of horizontal sections through the
embryonic and adult lumbar spinal cord, with antibodies against Msi2
and Msi1, MAP2, or Hu. A/B,
C/D, E/F,
G/H, I/J,
K/L, M/N, and
O/P represent each pair of double-stained
photomicrographs. The primary antibodies used are indicated in each
panel. A/B, E10 spinal cord. Msi2 and Msi1 are uniformly
distributed within the neural tube. C/D, Higher
magnification of the ventricular zone of E10 spinal cord, showing the
Msi2 expression in the Ki67-positive proliferating neuroepithelial
cells. E/F, E12 spinal cord.
G-L, E14 spinal cord.
K/L, Higher magnification of the bracketed area in
I/J. Msi2 and Msi1 are simultaneously expressed in the
embryonic VZ lining the central canal. Outside the VZ, most of the
MAP2-positive postmitotic neurons do not show immunoreactivity for
Msi2, whereas a subset of migrating neurons that are sparsely
distributed in the ventrolateral area of the mantle layer coexpress
Msi2 and MAP2 (bracketed area in G).
These Msi2-positive cells are negative for Msi1 and have a few
processes, indicating that they are migrating out of the VZ
(bracketed portion in I/J, and
arrows in K/L). M/N, Adult
spinal cord. O/P, Higher magnification of
M/N. In the adult spinal cord, Msi2 and Msi1 are highly
expressed in ependymal cells surrounding the central canal. Note the
Msi2 expression in some, but not all, populations of neurons scattered
within the parenchyma. Arrowheads in O
and P indicate Hu-positive but Msi2-negative neurons.
cc, Central canal. Scale bars: A,
B, 100 µm; C, D, 12 µm;
E-H, 50 µm; I, J, M, N,
45 µm; K, L, O, P, 10 µm.
|
|
Developmental expression of Msi2 in the CNS
Although the Msi1 protein was highly enriched in the embryonic
CNS, the Msi2 protein level appeared to stay at a relatively constant
level during brain development (Fig. 4B) in parallel with its mRNA levels (Fig. 2B). However, the
following immunohistochemical results clearly demonstrated that the
Msi2 expression was not ubiquitous, at least in various regions of the
embryonic and postnatal CNS. Rather, Msi2 was expressed in a cell
type-specific manner, and its expression appeared to be strongly
regulated spatially and temporally during the development of the mouse
brain and spinal cord. Notably, in the developing CNS, Msi2 was mainly
detected in proliferating precursor cells in the VZ and SVZ with a
pattern equivalent to that of Msi1.
VZ and SVZ
First, we examined the sites of Msi2 expression in the CNS at the
time when active neurogenesis and cell differentiation are known to
occur. The neural tube of the E10 mouse embryo is uniformly occupied
with mitotic and undifferentiated neuroepithelial cells. At this stage,
Msi2 expression was intense throughout the telencephalon and spinal
cord (Figs. 5, 6). Most of these Msi2-positive cells were Msi1-positive
(Figs. 5A,B,
6A,B), Ki67-positive (Figs.
5C,D, 6C,D), and
nestin-positive (Fig. 5E,F)
neuroepithelial cells. Ki67 is a marker for proliferating cells
(Nakamura et al., 2000 ), and nestin, an intermediate filament protein,
is a selective marker for neural progenitor cells, including CNS stem
cells (Lendahl et al., 1990 ). By E12-14, as CNS neurogenesis proceeds,
Msi1-positive mitotic neural precursors become restricted to the VZ in
the telencephalon and spinal cord; at this time, differentiated neurons
occupy the peripheral locations (Sakakibara et al., 1996 , Kaneko et
al., 2000 ). At the same time, simultaneous expression of Msi2
and Msi1 was detected in the VZ cells of most CNS regions, including
the telencephalon (Fig. 5G-L), spinal cord (Fig.
6C-L), and ganglionic eminence (Fig.
5I,J). Msi2 expression
disappeared in most of the MAP2-positive postmitotic neurons migrating
in the intermediate zone and cortical plate of the forebrain (Fig.
5I-L) and in the spinal cord (Fig.
6G,H).
In postnatal and adult CNS, cells in the ependyma and SVZ regions
showed significant levels of Msi2 and Msi1 expression. At P7, during
active gliogenesis, a large number of densely packed Msi2-positive
cells were found in the dorsolateral corner of the SVZ bordered by the
striatum and overlying the corpus callosum (Fig. 5M).
This region is known to be proliferative and to contain not only glial
precursors for astrocytes and oligodendrocytes (Levison and
Goldman, 1993 ; Sakakibara and Okano, 1997 ) but also neuronal precursors
(in the anterior part of the SVZ) for the interneurons of the olfactory
bulb (Lois and Alvarez-Buylla, 1993 ; Luskin, 1993 ). As gliogenesis
proceeds during postnatal development, the SVZ becomes thinner.
However, it persists throughout adulthood as a mitotically active layer
(Lewis, 1968 ). Many Msi2-positive cells were found in the ependyma and
SVZ cells in the adult forebrain (Fig. 5O). In the adult
spinal cord where the SVZ component had already disappeared, a high
expression of Msi2 remained in the ependymal cells lining the central
canal (Fig. 6M,O). Double
immunofluorescence revealed the coexpression of Msi2 and Msi1 in the
cells of the ependyma and SVZ throughout postnatal life. Recent
in vivo and in vitro studies have provided
evidence that the ependyma and/or SVZ of the postnatal mammalian
forebrain may be a source of CNS stem cells, which have the potential
to proliferate and differentiate into both neurons and glia (Morshead
et al., 1994 , 1998 ; Johe et al., 1996 ; Doetsch et al., 1999 ; Johansson
et al., 1999 ).
To examine the expression of Msi proteins in a CNS stem cell
population, we performed further immunostaining experiments using neurospheres. In the presence of mitogens such as EGF and bFGF, dissociated neural cells proliferate and form floating multicellular structures called "neurospheres' (Reynolds and Weiss, 1996 ;
Nakamura et al., 2000 ). Most of the cells in a neurosphere are clonally derived from a single CNS stem cell/progenitor and are thought to
possess the characteristics of CNS stem cells; i.e., they have self-renewing activity and are pluripotent, able to differentiate into
either neurons or glia (Reynolds and Weiss, 1996 ; Nakamura et
al., 2000 ). As shown in Figure
7A-C, Msi2 and
Msi1 were intensely coexpressed in cells of neurospheres derived from
E14.5 telencephalon. Coexpression of Msi2 and Ki67 (Fig. 7D,
inset) indicated that many Msi2-positive cells were endowed
with the proliferative activity in neurospheres. These neurospheres
were also uniformly immunolabeled with antibodies to nestin, and their
pluripotency was confirmed by differentiation assays using sister
neurospheres (data not shown).

View larger version (103K):
[in this window]
[in a new window]
|
Figure 7.
Msi2 expression in neurospheres and absence of
Msi2 in the immature neurons or oligodendrocyte precursor cells lying
in the SVZ. A-C, Double immunostaining
of neurosphere from E14.5 embryonic telencephalon, showing the
coexpression of Msi2 (B, red) and Msi1 (C,
green). Nuclei are counterstained with Hoechst dye (A,
blue). D, Msi2 expression (red)
in Ki67-positive proliferating cells (green) in a
neurosphere. Inset shows the higher view of the
neurosphere. Arrows point to the Msi2 and Ki67
double-positive cells. Note that Msi2 is mainly localized in the
cytoplasm of the cells in neurosphere, whereas the antibody to Ki67
only stains the nuclei. E-G, Scanning
confocal images of the coronal sections through the SVZ surrounding the
lateral ventricle of the P2 mouse forebrain. E, GFAP
(green) and Msi2 (red). Msi2 is
expressed in the GFAP-negative densely packed SVZ cells.
F, Hu (red) and Msi2
(green). Hu-positive immature neurons and
Msi2-positive cells distribute intermingled within the dorsolateral
area of the SVZ, but these two populations of cells hardly overlap with
each other. G, PDGFR- (red) and Msi2
(green). Msi2 expression is never found in
oligodendrocyte precursor cells that are positive for PDGFR- and lie
scattered in the SVZ. Scale bars: C, 50 µm;
D, 20 µm; E-G, 10 µm.
|
|
To address whether Msi2 expression is restricted to the neural
precursor cells/CNS stem cells in vivo, we double-stained
the sections through the SVZ regions with the antibodies to GFAP, Hu,
TuJ1, or PDGFR- . In the postnatal SVZ surrounding the lateral ventricles, the prominent expression of Msi2 was observed in the densely packed cells as described above and as shown in Figure 7E. We could not detect any cells reactive for GFAP within
the SVZ (Fig. 7E), whereas a few Msi2- and GFAP-positive
astrocytes that have multiple processes were present in the developing
corpus callosum (data not shown). Hu belongs to another RNA-binding
protein Elav family that is expressed in most immature and
differentiated neurons (Barami et al., 1995 ; Okano and Darnell, 1997 ;
Sakakibara and Okano 1997 ). The dorsolateral part of the SVZ region
contained many Hu-positive immature neurons that were generated from
the SVZ cells and appeared to be migrating into the lateral parenchyma. Although Msi2- and Hu-positive cells intermingled within these SVZ
regions, Msi2-positive cells were exclusively immunonegative for Hu,
and vice versa (Fig. 7F). Similarly,
TuJ1-positive immature neurons and their tangled neurites in these SVZ
regions were never labeled with Msi2 (data not shown). In
oligodendrocyte lineage, PDGFR- -positive cells residing in the
postnatal SVZ were proposed to be the earliest form of oligodendrocyte
progenitor cells (or O-2A progenitor cells), which divide in the SVZ to
give rise to slightly more differentiated cells of the oligodendrocyte
lineage in the nascent parenchyma, such as the corpus callosum and the neocortex (Nishiyama et al., 1996 ). Double immunofluorescence labeling
demonstrated that these PDGFR- -positive oligodendrocyte precursor
cells showed no immunoreactivity to Msi2 (Fig. 7G).
Taken together, these results suggested that Msi2 expression was mainly
associated with the proliferating precursor cells/CNS stem cells in the
VZ and SVZ and disappeared from the oligodendrocyte lineage and most
neurons early in their differentiation. This expression profile of Msi2
was quite similar to that of Msi1 during CNS development. In a
subset(s) of differentiated neurons, however, sustained Msi2 expression
was evident on the basis of the following immunohistochemical results.
Cerebrum
Neurons and glial cells, which originate in the embryonic VZ or
the postnatal SVZ, respectively, migrate into appropriate layers
through formative white matter, including the corpus callosum, to form
the stratified adult cerebral cortex. Msi1 is known to be expressed in
cells of the astrocyte lineage, which include both migrating
GFAP-negative (protoplasmic) and GFAP-positive differentiated
astrocytes (Miller and Raff, 1984 ; Sakakibara and Okano, 1997 ).
Msi2-expressing cells were found in the parenchymal regions of the
cerebrum throughout life, including the cerebral cortex, corpus
callosum, and striatum, although a lower immunoreactivity and sparser
distribution of Msi2 was seen in these regions than in the cells of the
SVZ. To identify the cell type that expressed Msi2 in the mature CNS,
double-label indirect immunostaining was performed using antibodies
against Msi2 and one of the following marker proteins: Hu for neurons,
CNPase for differentiated oligodendrocytes (Sprinkle 1989 ), and GFAP
and Msi1 for astrocytes. We observed the accumulation of Msi2-positive
cells throughout the cerebral cortex (Fig.
8A,B).
Some of these coexpressed Msi1 and possessed relatively thick and
branched processes (Fig.
8E,F), a feature typical of
differentiated astrocytes. Double labeling with antibodies to Msi2 and
GFAP confirmed the expression of Msi2 in astrocytes, as expected (Fig.
8I,J). Cells that were
immunoreactive for both Msi2 and GFAP were predominantly distributed in
the molecular layer (layer I) (Fig.
8I,J) and near the pial
surface. In addition, the deeper layers (layers II-VI) of the
cortex contained a considerable number of Msi2-positive but Hu-negative
astrocytic cells (Fig. 8G,H,
arrowhead). These cells coexpressed Msi1 (Fig.
8E,F) and thus might
represent protoplasmic astrocytes, which are known to be abundant in
cortical layers II-VI but are sparse in layer I (Sakakibara and Okano,
1997 ). Many of these astrocytes are not stained by GFAP
immunohistochemistry (Miller and Raff, 1984 ). In addition, astrocytes
that were closely associated with the endothelial cells of vascular
blood vessels were also immunoreactive for Msi2 (data not shown). In
oligodendrocytes, although Msi2- and CNPase-positive cells were found
commingled throughout the gray and white matter, CNPase-positive cells
were exclusively immunonegative for Msi2, and vice versa (Fig.
8K,L). This was consistent with the
observation indicating the absence of Msi2 from the oligodendrocyte
precursor cells residing in the postnatal SVZ, as described above.

View larger version (159K):
[in this window]
[in a new window]
|
Figure 8.
Distribution and cell type of Msi2-positive cells
in adult cerebrum. Double immunofluorescence labeling of the
somatosensory area of the adult neocortex, with antibodies against Msi2
and Msi1, or cell type-specific markers.
A/B, C/D,
E/F, G/H,
I/J, and
K/L represent each pair of double-stained
photomicrographs. Primary antibodies used are indicated in each panel.
A-D showing the expression of Msi2 in
astrocytes scattered throughout the gray (layers II-III) and white
matter (layer I) of the neocortex. Many small Msi2- and Msi1-positive
astrocytes with small cell bodies are seen throughout the cortex
(A, B). There is also a subset of neurons labeled with
Msi2 (A/B, arrows) with larger cell bodies, which are
found sparsely in layers II-III, and coexpress Hu antigens
(C/D, arrows) but are never labeled with Msi1 (compare
A/B with C/D). The pial surface is at the
top. E-L, High-power
photomicrographs of Msi2-expressing cells in the gray matter (layers
II-III, E-H) and white matter (molecular layer,
I/J; subcortical white matter,
K/L). In the gray matter, concurrent expression
of Msi2 and Hu is observed in a few round neuronal cells
(asterisks in G/H), whereas numerous
other Hu-positive neurons lack immunoreactivity for Msi2
(arrows in G/H). There is also a
considerable number of Hu-negative but Msi2-positive cells, which have
small oval cell bodies with elaborated multiple short processes
(arrowhead in G/H). These cells
may represent the population of GFAP-negative protoplasmic astrocytes
that are frequently observed in deep gray matter and are positive for
Msi1 (E/F). Fibrous astrocytes in the superficial
molecular layer and near the pial surface show colocalization of Msi2
and GFAP (I/J), whereas CNPase-positive
oligodendroglial cell bodies predominant in the subcortical white
matter are Msi2 negative (arrows in K/L).
Scale bars: A-D, 25 µm;
E-L, 8 µm.
|
|
On the other hand, the Msi2 protein was also expressed in a population
of large, round cells, distinct from the small, Msi1-positive astrocytes (Fig. 8A,B,
arrows). These Msi2-positive cells were found predominantly
within layers II-VI and overlapped with Hu immunoreactivity,
indicating that they were neurons (Fig. 8C,D, arrows, G,H, asterisk).
However, a comparison of the Hu and Msi2 expression clearly revealed
that the vast majority of Hu-positive neurons were negative for Msi2
(Fig. 8C,D, and arrows in
G,H). Consistent with the expression of Msi2 in a
subset of neurons in the cerebrum, Msi2-positive neuron-like cells were
seen in the embryonic spinal cord (bracketed area in Fig. 6G, H, arrows in Fig. 6K, L) and adult spinal cord (Fig. 6O,P)
and striatum (Fig. 5O,P), as mentioned above.
These data indicated that Msi2 expression was retained in a subset of
cells of the neuronal lineage, accompanied by its expression in the
astrocytes. Msi2 expression appeared to be absent or primarily
downregulated by the very early stage of the oligodendrocyte lineage,
as is the case for Msi1 (Sakakibara and Okano, 1997 ).
Msi2 expression in a subpopulation of GABA neurons
To determine the neuronal cell types that express Msi2 in the
adult cerebral cortex, we performed double immunofluorescence staining
with several markers for cholinergic (ChAT), dopaminergic (TH), and
GABAergic neurons (parvalbumin, calbindin-D28K, GABA). The cerebral
cortex of the adult mouse is composed of many types of neurons,
including large pyramidal neurons within layers II, III, IV, and V and
medium-sized nonpyramidal neurons distributed throughout the cortex.
Two calcium-binding proteins, PV and CB, are known to be
expressed in a small number of nonpyramidal neurons that contain GABA
as a neurotransmitter (Hendry et al., 1989 ; Celio 1990 ). Previous
studies have demonstrated that PV and CB localize to two almost
entirely distinct populations among these GABAergic interneurons
(Hendry et al., 1989 ; DeFelipe, 1997 ): PV is expressed in chandelier
and basket cells, whereas CB is found in double-bouquet cells
(DeFelipe, 1997 ). As shown in Figure 9A-D, expression
of Msi2 and PV coincided in a large number of cells, regardless of the
cortical area (i.e., parietal, frontal, or cingulate cortex) examined.
All of the PV-positive cells intensely expressed Msi2 and were
concentrated in layers II-V. We found very few or no cells that
stained positively for both Msi2 and CB (Fig.
9E-H). Consistent with these
observations, double immunofluorescence labeling with antibodies to
Msi2 and GABA indicated the expression of Msi2 in the subpopulations of
GABAergic neurons. Although many Msi2-positive cells were overlapped
with GABA-containing medium-sized neurons in layers II-III (Fig.
9I,J), there was also a
population of cells that were GABA positive but Msi2 negative (Fig.
9K,L, arrow). In
addition, Msi2 expression was absent from all the cortical pyramidal
neurons that could be easily distinguished from other cells by their
large cell body and typical morphology. In layer V, there were many
somata and proximal dendrites of large pyramidal neurons (presumptive
Betz cells) that were surrounded by PV-positive multi-terminal endings
typical of basket cell axon terminals (Fig. 9D,
asterisks), as observed previously (Hendry et al., 1989 ). Msi2 expression was absent from these PV-negative somata (Fig. 9C, asterisks). As shown in Figure 9,
M and N, ChAT-positive cholinergic cells with
small oval cell bodies and fine ascending processes were found sparsely
throughout the cortex, and they showed no Msi2 expression (Fig.
9M,N, arrows). These
observations indicated that during the development of the cerebral
cortex, Msi2 expression is maintained in the lineage of PV-containing
GABAergic neurons.

View larger version (147K):
[in this window]
[in a new window]
|
Figure 9.
Msi2 expression in
parvalbumin-positive GABAergic neurons in the adult neocortex.
Double-immunofluorescence labeling with antibodies against Msi2
and parvalbumin (PV), calbindin-D28K
(CB), GABA, choline acetyltransferase
(ChAT), or tyrosine hydroxylase
(TH) was performed on coronal sections of adult
cerebral cortex, basal ganglia, and hippocampus.
A/B, C/D,
E/F, G/H,
I/J, K/L,
M/N, O/P,
Q/R, and
S/T represent each pair of double-stained
photomicrographs. Primary antibodies used are indicated in each panel.
A/B, Neocortex layers II-III, showing the coincidental
expression of Msi2 and PV in GABAergic nonpyramidal neurons. Faint
staining of Msi2 is also seen in the fine-branched processes and small
cell bodies of astrocytes. C/D, Higher magnification of
the Msi2- and PV-positive neurons in the deeper layer V. The somata of
large pyramidal neurons surrounded by parvalbumin-positive
multi-terminal endings express neither Msi2 nor PV
(asterisks). E/F, Neocortex layers
II-III stained with Msi2 and CB. Note that these two populations of
neurons are intermingled but do not overlap with each other.
Long arrows indicate Msi2-negative CB-positive neurons.
Short arrows indicate the neurons that are Msi2 positive
but CB negative. G/H, Higher magnification of neurons
that express CB or Msi2 in layers II-III. I/J,
K/L, Layers II-III stained with antibodies to Msi2 and
GABA, showing the Msi2 expression in GABA-containing neurons.
(I/J and arrowhead in
K/L). Arrow in K/L points to a
GABA-containing but Msi2-negative neuron. M/N, Gray
matter of the neocortex, showing the absence of Msi2 expression from
ChAT-positive cholinergic neurons (arrows).
O/P, Hippocampal formation, showing the coexpression of
Msi2 and PV in a small number of interneurons in the CA1 subfield.
Numerous CA1 pyramidal neurons aligned in rows are not labeled with
either Msi2 or PV. Q/R, S/T, Nuclei of
the vertical limb of the diagonal band and the zona incerta,
respectively. In the basal ganglia, Msi2 immunoreactivity was
found in several ChAT-positive cholinergic neurons in the nuclei of
diagonal bands (Ch2 and Ch3), in addition to the PV-positive neurons.
Msi2 is also expressed in TH-positive dopaminergic neurons (A13) within
the zona incerta in the subthalamic region. Scale bars:
A, B, E, F,
M-T, 25 µm; C,
D, G, H,
I-L, 8 µm. Rad, Stratum
radiatum; Py, stratum pyramidale; Or,
stratum oriens.
|
|
Cerebellum
A dynamic pattern of Msi2 expression was also observed in the
developing cerebellum. In the cerebellar primordium at E14.5, Msi2 was
expressed in the small tightly packed cells of the external granule
cell layer (EGL) migrating out of the rhombic lip of the lateral recess
and also in the VZ of the fourth ventricle (data not shown). During
early postnatal development, the EGL covers the surface of the
developing cerebellum and is composed exclusively of proliferating
neuronal precursors. By P3, the expression of Msi2 could be detected in
the EGL, although to a lesser degree than that of Msi1 (Fig.
10A,B).
As neurogenesis progresses, two populations of cells appear in the EGL:
cells proliferating in the upper portion of the EGL (EGLa) and cells
undergoing the initial step of neuronal differentiation deeper in the
EGL (EGLb) (Altman, 1972 ; Kuhar et al., 1993 ). At P7, Msi1 expression
was predominantly observed within EGLa cells, as reported previously
(Sakakibara and Okano, 1997 ), whereas Hu was expressed in the
differentiating granule neurons within EGLb (Fig.
10E,G). At this time, only faint Msi2 expression was detected in EGLa (Fig.
10F,H).

View larger version (149K):
[in this window]
[in a new window]
|
Figure 10.
Msi2 expression in the cerebellum.
Double-immunofluorescence labeling of sagittal sections through the
early postnatal and adult cerebellum, with antibodies against Msi2 and
Msi1 or Hu. A/B,
C/D, E/F,
G/H, and
I/J represent each pair of double-stained
photomicrographs. Primary antibodies used are indicated in each panel.
A/B, P3 cerebellum. Intense Msi1 staining is observed in
neuronal precursor cells in the EGL (indicated by
arrow), in addition to the cells in the PCL
(arrowhead) and numerous glial precursor cells in the
folia white matter. Msi2 expression is prominent in the PCL and is
faint in EGL cells and glial precursor cells in the white matter. Deep
cerebellar nuclei (asterisk) are labeled only with Msi2.
E-H, P7 cerebellum. The EGL is toward
the top part of the panels. At P7, a low level of Msi2
expression is seen in the EGL (E/F). In the PCL,
Purkinje cells forming a single row show immunoreactivity for both Msi2
and Hu but not for Msi1, whereas many small cells residing in the
vicinity of the Purkinje cells express both Msi2 and Msi1 (compare
E/F with G/H). These cells are
likely to be the developing Bergmann glia. Msi2 expression is absent
from Hu-positive granule neurons in the IGL. C/D, Adult
cerebellum. In adults, Msi2 is predominantly expressed in the Purkinje
cells (arrows in C/D) and Bergmann glial
cells, which have cell bodies located adjacent to the large cell bodies
of Purkinje cells and extend their tangential processes into the ML.
Arrowhead in C/D represents the Golgi
neuron in the IGL. I/J, Higher magnifications of the
adult PCL. Arrows in I/J indicate the
Purkinje cells, and arrowhead in
I/J points to the soma of a Bergmann
glial cell. ML, Molecular layer; PCL,
Purkinje cell layer; IGL, internal granule layer;
EGL, external granule cell layer. Scale bars:
A, B, 100 µm; C-H, 30 µm; I, J, 8 µm.
|
|
The most striking difference between the Msi2 and Msi1 expression was
seen in the neuronal population within the Purkinje cell layer (PCL)
and deep cerebellar nuclei (Fig.
10A,B). In the developing and adult
PCL, Msi2 expression was conspicuous in the large Purkinje neurons,
which absolutely lacked Msi1 immunoreactivity (Fig.
10C,D,I,J,
arrows). Msi2 expression in differentiated neurons was also
observed within the deep cerebellar nuclei (Fig. 10B, asterisk) and in interneurons such as the stellate and
basket cells residing in the molecular layer (Fig.
10D). We noticed, however, that Msi2 was not
expressed in numerous granule neurons in the internal granule cell
layer (IGL), which were positive for Hu (Fig. 10G). In the
IGL, Msi2 was detectable only in Golgi interneurons (Fig.
10D, arrowhead). The sustained expression
of Msi2 in the GABA neurons was also observed in the adult cerebellar
cortex. Msi2 expression was confined to almost all the inhibitory GABA neurons, such as the stellate and basket cells in the molecular layer,
the Purkinje cells, and the Golgi cells in the IGL (Fig. 10).
Correspondingly, PV is known to be expressed in these cells (Celio,
1990 ).
In glial cells, Msi2 expression was detected in astrocytes within the
central white matter tracts and the IGL, albeit at a greatly reduced
level (Fig. 10B,D), compared with
Msi1, which is highly expressed in astrocytes in these regions (Fig.
10C) (Sakakibara and Okano, 1997 ). In the PCL, Msi2 and Msi1
expression were concomitantly observed in the radially aligned Bergmann
glial cells and their fibers coursing from the PCL through the
molecular layer to the pial surface (Fig.
10C-J). Indirect double labeling on
tissue sections of adult cerebellum with antibodies to GFAP and CNPase
indicated the colocalization of Msi2 and GFAP in various regions,
including the PCL (Bergmann glia) and foliar white matter tracts (data
not shown). CNPase was never coexpressed with Msi2 in these regions (data not shown). These results indicated the expression of Msi2 in
certain subpopulations of neurons and astrocytes in the cerebellum.
Other CNS regions
The persistent expression of Msi2 in certain subpopulations of
neurons was also clearly observed in other brain regions. In the adult
hippocampal formation, small, scattered Msi2-positive neurons were
detected in the stratum pyramidale, stratum oriens, radiatum, and
lacunosum-moleculare. A high-power view of the CA1 region showed
intense expression of Msi2 in a small number of neurons along with
diffuse expression in the astrocytes, which are uniformly distributed
within this region (Fig. 9O,P). In the CA1 and
CA3 subfields, ~20% of all the GABAergic neurons contain PV (Kosaka
et al., 1987 ). This PV-positive subpopulation of GABA neurons is
thought to represent fast-spiking cells, which are electrically and
metabolically more active than other local circuit neurons (Kawaguchi
et al., 1987 ). Msi2 was mainly localized to these PV-containing
GABAergic neurons in the CA1 region, although a few Msi2-positive
neurons were negative for PV (Fig. 9O,P). In
contrast, Msi2 immunoreactivity was absent from the CA1 pyramidal neurons (Fig. 9O), aligned in the stratum pyramidale, where
CB is known to be expressed (Celio, 1990 ). Many granule neurons of the
dentate gyrus did not express Msi2 either, whereas the anti-Msi2 antibody stained occasional cells in the subgranular region, the innermost region of the granular layer of the dentate gyrus (data not
shown). Our previous study reported that Msi1 shows the same expression
profile in this subgranular region (Kaneko et al., 2000 ). Neuron
production in most hippocampal regions is completed before birth in the
mouse; however, in the dentate gyrus, granule neurons continue to be
generated into adulthood (Eriksson et al., 1998 ). Given that the
subgranular region contains neural progenitor cells that are capable of
postnatal neurogenesis (Palmer et al., 1997 ), the Msi2-positive cells
in this region may represent a functionally distinct subset, i.e., a
neural progenitor population in the postnatal hippocampus.
On the other hand, in other restricted brain regions including the
basal ganglia and brainstem, Msi2 immunoreactivity was found in some
non-GABA neurons, which contained neither PV nor CB. Within the basal
ganglia, Msi2 was detected in subpopulations of neurons sparsely
distributed in the ventral pallidum, globus pallidus, and caudate
putamen (data not shown). Msi2 expression was seen in several
ChAT-positive cholinergic projection neurons in the striatum, nuclei of
diagonal bands (Ch2 and Ch3) (Fig. 9Q,R), medial
septal nucleus (Ch1), and medial habenular nucleus (Ch7) (data not
shown), in addition to the PV-positive neurons. Some dopaminergic
neurons, identified by the presence of TH, also expressed Msi2. Figure
9, S and T, shows the colocalization of Msi2 and
TH in zona incerta dopaminergic neurons (A13) in the thalamic area
surrounding the third ventricle. In the thalamus, epithalamus, and
hypothalamic regions, the neuronal expression of Msi2 was identified in
the medial preoptic nucleus, supraoptic nucleus, periventricular
nucleus, and medial and lateral habenular nuclei, whereas Msi1 was
expressed exclusively in astrocytes in these regions (data not shown).
Taken together, these observations substantially support the notion
that during the development of several different regions of the brain
(the cerebral cortex, hippocampus, and cerebellar cortex), Msi2
expression is maintained in the lineage of PV-containing GABAergic
neurons ranging from the precursor stage to the postmitotic and
differentiated stage, as well as in the astrocyte lineage. However, the
Msi2 expression observed in cholinergic and dopaminergic cells in other
brain regions (basal ganglia, brainstem) cannot be explained in this context.
Changes in Msi2 expression after injury
A previous report (Sakakibara and Okano, 1997 ) showed that Msi1
expression is drastically upregulated in reactive astrocytes after
brain injury. The fact that Msi2 was also expressed in cells of the
astrocyte lineage allowed us to examine the changes in Msi2 expression
during the formation of glial scars after CNS injury. Experimentally
induced damage to the CNS leads to the formation of a glial scar at the
site of injury and to multiple changes in the astrocytes surrounding
the damaged tissue. Numerous hypertrophic and hyperplastic reactive
astrocytes display increased levels of GFAP and become mitotically
active after the injury (Miyake et al., 1992 ). Immediately after the
injury (day 0), the expression levels of Msi2 and GFAP were unaffected
in the neighborhood of the injury site (cerebral cortex layers II-III)
(Fig.
11A,B). By day 2 after lesion, the number of Msi2-positive cells around the
lesion site was not significantly different from the number in
undamaged areas of the ipsilateral and contralateral hemispheres (data
not shown). At 4 d after lesion, there was a dramatic increase in
the number of Msi2-positive cells in the vicinity of the injury site
(the region within 100-200 µm from the injury site) (Fig. 11).
Immunohistochemical analysis of the area during glial scar formation
showed dramatic and coincident induction of the Msi2 and Msi1 proteins
at 3-4 d after lesion (Fig.
11E,F). In this region, numerous GFAP-positive astrocytes were present. These cells had long
and wavy processes extending in all directions and appeared to form a
thick bundle of glia; these are typical characteristics of reactive
astrocytes. These GFAP-positive cells were entirely immunopositive for
Msi2 (Fig. 11G,H, arrows). BrdU
administration experiments showed that these Msi2-positive reactive
astrocytes continued to proliferate even 4 d after lesion (~35%
of the BrdU-incorporated cells expressed Msi2) (Fig.
11C,D, arrows). This was consistent with previous studies showing the occurrence of cell division among
GFAP-expressing astrocytes induced by brain injury (Miyake et al.,
1992 ). On the other hand, there were many BrdU-positive but
Msi2-negative proliferating cells at the site of injury. Most of these
cells seemed to be activated ameboid microglia or infiltrated macrophages, which proliferated in response to the injury. The lack of
Msi2 induction in the microglia and macrophages was verified by the
double-staining with anti-Msi2 and anti-MacI antibodies (Thomas, 1992 )
(data not shown).

View larger version (94K):
[in this window]
[in a new window]
|
Figure 11.
Induced expression of Msi2 and Msi1 in reactive
astrocytes. Double-label fluorescence localization of Msi2 and Msi1,
BrdU, or GFAP in the injured region of adult cerebral cortex
immediately (day 0) (A, B) or 4 d
(C-H) after lesioning. All photomicrographs
correspond to cortex layers II-III, and the lesioned sites are toward
the top in all panels. A,
B, GFAP and Msi2 at day 0. The vicinity of the site of
injury shows no immunoreactivity for GFAP. C,
D, BrdU and Msi2 at 4 d after lesion. Msi2 is
expressed in a population of BrdU-positive proliferating cells
(arrows) that lie close to the injury site.
E, F, Msi1 and Msi2 at 4 d after
lesion. Simultaneous expression of Msi1 and Msi2 is evident in many
enlarged reactive cells. G, H, GFAP and
Msi2 at 4 d after lesion. Intense immunoreactivity of Msi2 is
observed in an increased number of GFAP-positive reactive astrocytes;
these cells exhibit enlarged, elongated cell bodies with multiple
processes (arrows). Arrowhead indicates a
rare Msi2-positive neuron-like cell that has an oval cell body and is
immunonegative for GFAP. Scale bar (shown in H):
A-H, 18 µm.
|
|
These observations demonstrated that Msi2 expression is induced at high
levels in reactive astrocytes in the course of scar formation, with a
profile similar to the expression of Msi1 (Sakakibara and Okano, 1997 ).
Expression of the msi2 and msi1 genes therefore might be controlled by a common regulatory mechanism in reactive astrocytes.
 |
DISCUSSION |
Msi2 is an evolutionarily conserved RNA-binding protein found in
different species of vertebrates (Xenopus, mouse, and
human). On the basis of their striking sequence homology and similar
RNA-binding preferences in vitro, Msi2 and Msi1 are likely
to constitute a unique RNA-binding protein family (the Msi family).
These two proteins are likely to play similar roles in the RNA
metabolism that occurs in the cytoplasm, such as in the localization or
stabilization of specific mRNAs, polyadenylation, or translational
regulation (for review, see MacCarthy and Kollmus, 1995 ).
Role of the Msi family in neural precursor cells/CNS
stem cells
In the present study we demonstrated the cell type-specific and
developmentally regulated expression of Msi2 in the mammalian CNS. In
particular, proliferative cells in the embryonic VZ and SVZ showed a
high level of Msi2 expression. As neurogenesis proceeded, Msi2
expression disappeared in most of the postmitotic or migrating neurons
within the intermediate zone or cortical plate. Interestingly, our
double-labeling results indicated that Msi2 expression was coincident
with the expression of Msi1 in proliferative cells. Previous in
vivo and in vitro analyses have shown that Msi1 is predominantly expressed in neural precursor cells/multipotent CNS stem
cells (Sakakibara et al., 1996 ; Kaneko et al., 2000 ); therefore, Msi2
and Msi1 are likely to be coexpressed in CNS stem cells during
embryonic CNS development. Marked Msi2 expression was seen in growing
neurospheres, further supporting its expression in CNS stem cells.
Postnatally, the concurrent expression of Msi2 and Msi1 was maintained
in proliferating neuronal and/or glial precursors in the SVZ
(Sakakibara and Okano, 1997 ; present study). The postnatal SVZ mainly
generates two glial lineages of cells: astrocytes and oligodendrocytes
(Levison and Goldman, 1993 ). Although the Msi2 and Msi1 proteins
disappeared from oligodendrocytes early in their differentiation, the
proteins continued to be expressed in cells of the astrocyte lineage,
including ependymal cells. The adult SVZ regions contain a population
of CNS stem cells (Morshead et al., 1994 ; Johe et al., 1996 ). However,
the SVZ might not be the only source of adult brain neurogenesis.
Johansson et al. (1999) presented evidence that ependymal cells might
be stem cells of the adult brain and spinal cord, whereas Doetsch et
al. (1999) provided compelling evidence that SVZ astrocytes, referred
to as "Type-B" cells, are precursors for neurogenesis and possibly stem cells. Type-B cells exhibit all of the characteristics of astrocytes, including GFAP expression. Consistent with this, recent heterotopic transplantation experiments (Leavitt et al., 1999 ) showed
that mature neocortical astrocytes retain the capacity to
dedifferentiate into radial glial-like cells, which are considered to
be the neuronal and glial precursor cells in embryonic corticogenesis (Malatesta et al., 2000 ; Noctor et al., 2001 ). So far, the cell type
that serves as the source of postnatal CNS stem cells in vivo is unresolved (Scheffler et al., 1999 ). In any case, the occurrence of Msi1 and Msi2 expression in the SVZ, ependymal cells, and
astrocytes during postnatal life as well as in embryonic
neuroepithelial cells of the VZ implies that Msi family proteins may
function in cells harboring stem cell characteristics. The upregulation of Msi2 and Msi1 in proliferating reactive astrocytes may also reflect
that these cells are endowed with the characteristics of glial
precursors or of neural stem cells capable of generating astrocytes.
In accordance with this idea, we have obtained results from preliminary
in vitro experiments using neurospheres lacking expression of both the msi1 and msi2 genes that indicate
that Msi1 and Msi2 are essential for the maintenance and/or
proliferation of CNS stem cells (S. Sakakibara, unpublished results).
The molecular basis for the function of the Msi family in CNS stem
cells might be elucidated by identifying target mRNAs of the Msi1 and
Msi2 proteins. We recently reported that Msi1 protein binds to the 3'-UTR of mouse numb mRNA, leading to repression of
Numb expression (Imai et al., 2001 ). The mouse Numb protein is known to
inhibit the Notch-signaling cascade (Wakamatsu et al., 1999 ), which is required for the self-renewing activity of CNS stem cells (Ohtsuka et
al., 1999 ; Gaiano et al., 2000 ; Nakamura et al., 2000 ). The present
sequence analysis and in vitro RNA-binding assay revealed a
significant conservation of the RRM sequences and the RNA-binding specificity between Msi1 and Msi2, supporting the idea that Msi family
proteins may share the same or similar target RNAs, which are
characterized by uridine-rich sequences, as their downstream target
molecules in vivo. Thus, Msi2 and Msi1 may exert common important functions in CNS stem cells by regulating the processing of
the same specialized mRNAs. Loss-of-function experiments may be able to
unequivocally reveal the function of the Msi proteins in embryonic and
postnatal mammalian CNS development. Phenotypic analyses of
msi2 and msi1 knock-out mice are currently in
progress in our laboratory.
Msi2 in a subset of GABA neurons
During the development of the cerebral cortex, Msi2 expression is
primarily downregulated in most of the cortical neurons except the GABA
interneurons. Interneurons containing the GABA neurotransmitter are the
major type of inhibitory neurons in the mammalian neocortex. These
cells comprise 25% of all the cortical neurons, and exert powerful
control over the local neuronal network by mediating both feedback and
feedforward inhibition. Are there any significant differences between
the developmental process of GABA interneurons and those of other types
of neurons? It is well known that most differentiating cortical
projection neurons migrate radially from the cortical VZ toward the
pial surface and form the cortical plate, where they temporarily settle
according to an inside-out gradient of positioning (Bayer and Altman,
1991 ). However, recent experiments using retroviral labeling or
chimeric embryos suggest that cortical projection neurons and GABA
interneurons are derived from distinct proliferative zones. Most, if
not all, cortical interneurons (GABAergic neurons) are born in the
VZ/SVZ of the lateral ganglionic eminence (LGE), the primordium of the striatum, and migrate tangentially from the LGE to the neocortex through the striatal-neocortical pathway (Tamamaki et al., 1997 ; Tan
et al., 1998 ). Expression of Msi2 in the embryonic LGE is compatible
with the idea that Msi2 is maintained by the GABAergic lineage
undergoing a tangential mode of cell movement. Furthermore, it is
possible that Msi2 expression is rapidly decreased in neurons that are
fated to migrate radially into the neocortex.
In adult neocortex, GABA interneurons are classified into two separate
subpopulations: the CB-positive double-bouquet cells and the
PV-positive chandelier and basket cells (Hendry et al., 1989 ; DeFelipe,
1997 ). Msi2 expression is retained in a subset of PV-containing
GABAergic neurons in the neocortex, hippocampus, and cerebellar cortex.
In these regions, other CB-positive GABAergic neurons or pyramidal
projection neurons showed little or no expression of Msi2.
PV-containing GABA interneurons show electrophysiological fast-firing
behavior and form a unique dense network connected by dendrodendritic
gap junctions (electrical synapses), presumably enabling the generation
of synchronous oscillations and the processing of spatiotemporal
information within brain regions (Fukuda and Kosaka, 2000 ). Proper
functioning of PV-positive GABA interneurons has been shown to require
selective expression or rapid turnover of metabotropic glutamate
receptor protein (mGluR5) (Whittington et al., 1995 ; Stinehelfer et
al., 2000 ; van Hooft et al., 2000 ) or the highly localized dendritic
expression of voltage-gated sodium and potassium ion channels (Martina
et al., 2000 ). Therefore, Msi2 may participate in the generation and/or
maintenance of these specific neuronal lineages by controlling the
local translation or stability of the mRNAs for these receptors and
channels, in addition to its role in neural precursor cells.
Intriguingly, the neuronal expression of Msi is not only a
characteristic of vertebrates but occurs in invertebrates as well. In
Drosophila, d-Msi is expressed in a small population of
differentiated neurons, including the photoreceptor cells (Hirota et
al., 1999 ), and in the precursor cells for the sensory organs and
embryonic CNS (Nakamura et al., 1994 ). In Caenorhabditis
elegans, Msi is expressed in certain populations of neurons,
including the GABAergic neurons in the ventral cord, although its
molecular significance in this organism remains unclear (Yoda et al.,
2000 ). Because we have failed to find any other Msi-related genes in
invertebrates, it is possible that mouse Msi2 has succeeded to the more
restricted neuronal expression of invertebrate Msi protein during evolution.
Msi2 and neurological diseases
Previous immunohistochemical studies have indicated a selective
reduction in or functional impairment of PV-containing GABA neurons in
the neocortex under certain pathological conditions, which include
schizophrenia (Beasley and Reynolds, 1997 ), hypothyroidism in rats
(Berbel et al., 1996 ), and an animal model of cortical dysplasia (Roper
et al., 1999 ). Msi2 may affect the pathogenesis of such diseases
through its function in PV-containing neurons. On the other hand, the
msi2 syntenic region contains several loci of hereditary
human neurological disorders: FTDP17, Pick disease, GPSC, and NF1.
FTDP17, GPSC, and Pick disease may be allelic and are accompanied by
varying degrees of dementia and atrophic degeneration of frontal lobe
(Lynch et al., 1994 ). NF1 is characterized by cafe-au-lait spots and
fibromatous tumors of the skin, but it is also frequently associated
with CNS tumors, including astrocytomas and ependymomas (Mukonoweshuro
et al., 1999 ). Although inherited FTDP17 and NF1 are caused by
mutations in the tau (Hutton et al., 1998 ) and
neurofibromin (Streubel et al., 1999 ) genes, respectively, the complexities of their clinical and pathologic symptoms raise the
possibility that additional loci may be involved in potentiating or
modifying each syndrome (Lynch et al., 1994 ; Streubel et al., 1999 ).
Indeed, a recent somatic mosaicism analysis of a patient with a complex
clinical phenotype of NF1 and a large deletion of the contiguous
genomic regions flanking the neurofibromin gene provides evidence for
the involvement of adjacent loci in the clinical manifestations of this
syndrome (Streubel et al., 1999 ). It will be interesting to determine
whether the msi2 gene is implicated in these neurological
disorders as a modifier.
 |
FOOTNOTES |
Received April 13, 2001; revised July 19, 2001; accepted July 27, 2001.
This work was supported by grants from the Japanese Ministry of
Education, Japan Science and Technology Corporation (Core Research for
Evolutional Science and Technology).
Correspondence should be addressed to Dr. Hideyuki Okano, Department of
Physiology, Keio University School of Medicine, 35 Shinanomachi,
Shinjuku-ku, Tokyo 160-8582, Japan. E-mail:
hidokano{at}sc.itc.keio.ac.jp.
S. Sakakibara's present address: Department of Histology and
Neurobiology, Dokkyo University School of Medicine, Mibu, Tochigi 321-0293, Japan.
 |
REFERENCES |
-
Akamatsu W,
Okano HJ,
Osumi N,
Inoue T,
Nakamura S,
Sakakibara S,
Miura M,
Matsuo N,
Darnell RB,
Okano H
(1999)
Mammalian ELAV-like neuronal RNA-binding proteins HuB and HuC promote neuronal development in both the central and peripheral nervous systems.
Proc Natl Acad Sci USA
96:9885-9890[Abstract/Free Full Text].
-
Altman J
(1972)
Postnatal development of the cerebellar cortex in the rat.
J Comp Neurol
145:353-398[Web of Science][Medline].
-
Amano T,
Hamprecht B,
Kemper W
(1974)
High activity of choline acetyltransferase induced in neuroblastoma × glioma hybrid cells.
Exp Cell Res
85:399-408[Web of Science][Medline].
-
Barami K,
Iversen K,
Furneaux H,
Goldman SA
(1995)
Early expression of Hu proteins by newly generated neurons of the adult avian forebrain.
J Neurobiol
28:82-101[Web of Science][Medline].
-
Bayer SA,
Altman J
(1991)
In: Neocortical development. New York: Raven.
-
Beasley CL,
Reynolds GP
(1997)
Parvalbumin-immunoreactive neurons are reduced in the prefrontal cortex of schizophrenics.
Schizophrenia Res
24:349-355[Web of Science][Medline].
-
Berbel P,
Marco P,
Cerezo JR,
DeFelipe J
(1996)
Distribution of parvalbumin immunoreactivity in the neocortex of hypothyroid adult rats.
Neurosci Lett
204:65-68[Medline].
-
Blaxall BC,
Pellett AC,
Wu SC,
Pende A,
Port JD
(2000)
Purification and characterization of
-adrenergic receptor mRNA-binding proteins.
J Biol Chem
274:4290-4297. -
Burd CG,
Dreyfuss G
(1994)
Conserved structures and diversity of functions of RNA-binding proteins.
Science
265:615-621[Abstract/Free Full Text].
-
Celio MR
(1990)
Calbindin D-28k and parvalbumin in the rat nervous system.
Neuroscience
35:375-475[Web of Science][Medline].
-
Davis AA,
Temple S
(1994)
A self-renewing multipotential stem cell in embryonic rat cerebral cortex.
Nature
372:263-266[Medline].
-
DeFelipe J
(1997)
Types of neurons, synaptic connections and chemical characteristics of cells immunoreactive for calbindin-D28K, parvalbumin and calretinin in the neocortex.
J Chem Neuroanat
14:1-19[Web of Science][Medline].
-
DeMaria CT,
Brewer G
(1996)
AUF1 binding affinity to A+U-rich elements correlates with rapid mRNA degradation.
J Biol Chem
271:12179-12184[Abstract/Free Full Text].
-
Doetsch F,
Caillé I,
Lim DA,
García-Verdugo JM,
Alvarez-Buylla A
(1999)
Subventricular zone astrocytes are neural stem cells in the adult mammalian brain.
Cell
97:703-716[Web of Science][Medline].
-
Eriksson PS,
Perfilieva E,
Bjork-Eriksson T,
Alborn AM,
Nordborg C,
Peterson DA,
Gage FH
(1998)
Neurogenesis in the adult human hippocampus.
Nat Med
4:1313-1317[Web of Science][Medline].
-
Fukuda T,
Kosaka T
(2000)
The dual network of GABAergic interneurons linked by both chemical and electrical synapses: a possible infrastructure of the cerebral cortex.
Neurosci Res
38:123-130[Web of Science][Medline].
-
Gaiano N,
Nye JS,
Fishell G
(2000)
Radial glial identity is promoted by Notch1 signaling in the murine forebrain.
Neuron
26:395-404[Web of Science][Medline].
-
Good PJ,
Rebbert ML,
Dawid IB
(1993)
Three new members of the RNP protein family in Xenopus.
Nucleic Acids Res
21:999-1006[Abstract/Free Full Text].
-
Good PJ,
Yoda A,
Sakakibara S,
Yamamoto A,
Imai T,
Sawa H,
Ikeuchi T,
Tsuji S,
Satoh H,
Okano H
(1998)
The human Musashi homolog 1 (MSI1) gene encoding the homologue of Musashi/Nrp-1, a neural RNA-binding protein putatively expressed in CNS stem cells and neural progenitor cells.
Genomics
52:382-384[Web of Science][Medline].
-
Hendry SHC,
Jones EG,
Emson PC,
Lawson DEM,
Heizmann CW,
Streit P
(1989)
Two classes of cortical GABA neurons defined by differential calcium protein immunoreactivities.
Exp Brain Res
76:467-472[Web of Science][Medline].
-
Hirota Y,
Okabe M,
Imai T,
Kurusu M,
Yamamoto A,
Miyao S,
Nakamura M,
Sawamoto K,
Okano H
(1999)
Musashi and Seven in absentia downregulate tramtrack through distinct mechanisms in Drosophila eye development.
Mech Dev
87:93-101[Web of Science][Medline].
-
Hutton M,
Lendon CL,
Rizzu P,
Baker M,
Froelich S,
Houlden H,
Pickering-Brown S,
Chakraverty S,
Isaacs A,
Grover A,
Hackett J,
Adamson J,
Lincoln S,
Dickson D,
Davies P,
Petersen RC,
Stevens M,
de Graaff E,
Wauters E,
van Baren J
(1998)
Association of misssense and 5'-splice-site mutations in tau with the inherited dementia FTDP-17.
Nature
393:702-705[Medline].
-
Imai T,
Tokunaga A,
Yoshida T,
Hashimoto M,
Mikoshiba K,
Weinmaster G,
Nakafuku M,
Okano H
(2001)
The neural RNA-binding protein Musashi1 translationally regulates the m-numb gene expression by interacting with its mRNA.
Mol Cell Biol
21:3888-3900[Abstract/Free Full Text].
-
Johansson CB,
Momma S,
Clarke DL,
Risling M,
Lendahl U,
Frisén J
(1999)
Identification of a neural stem cell in the adult mammalian central nervous system.
Cell
96:25-34[Web of Science][Medline].
-
Johe KK,
Hazel TG,
Muller T,
Dugich-Djordjevic MM,
McKay RD
(1996)
Single factors direct the differentiation of stem cells from the fetal and adult central nervous system.
Genes Dev
10:3129-3140[Abstract/Free Full Text].
-
Kaneko Y,
Sakakibara S,
Imai T,
Suzuki A,
Nakamura Y,
Sawamoto K,
Ogawa Y,
Toyama Y,
Miyata T,
Okano H
(2000)
Musashi1: an evolutionally conserved marker for CNS progenitor cells including neural stem cells.
Dev Neurosci
22:139-153[Web of Science][Medline].
-
Kawaguchi Y,
Katsumaru H,
Kosaka T,
Heizmann CW,
Hama K
(1987)
Fast spiking cells in rat hippocampus (CA1 region) contain the calcium-binding protein parvalbumin.
Brain Res
416:369-374[Web of Science][Medline].
-
Kawashima T,
Murakami AR,
Ogasawara M,
Tanaka KJ,
Isoda R,
Sasakura Y,
Nishikata T,
Okano H,
Makabe KW
(2000)
Expression patterns of musashi homologues of the ascidians, Halocynthia roretzi and Ciona intestinalis.
Dev Gene Evol
210:162-165[Web of Science][Medline].
-
Khan F,
Jaiswal AK,
Szer W
(1991)
Cloning and sequence analysis of a human type A/B hnRNP protein.
FEBS Lett
290:159-161[Web of Science][Medline].
-
Kosaka T,
Katsumaru H,
Hama K,
Wu J-Y,
Heizmann CW
(1987)
GABAergic neurons containing the Ca2+-binding protein parvalbumin in the rat hippocampus and dentate gyrus.
Brain Res
419:119-130[Web of Science][Medline].
-
Kuhar SG,
Feng L,
Vidan S,
Ross ME,
Hatten ME,
Heinz N
(1993)
Changing patterns of gene expression define four stages of cerebellar granule neuron differentiation.
Development
117:97-104[Abstract/Free Full Text].
-
Leavitt BR,
Hernit-Grant CS,
Macklis JD
(1999)
Mature astrocytes transform into transitional radial glia within adult mouse neocortex that supports directed migration of transplanted immature neurons.
Exp Neurol
157:43-57[Web of Science][Medline].
-
Lendahl U,
Zimmerman LB,
McKay R
(1990)
CNS stem cells express a new class of intermediate filament protein.
Cell
60:585-595[Web of Science][Medline].
-
Levison SW,
Goldman JE
(1993)
Both oligodendrocytes and astrocytes develop from progenitors in the subventricular zone of postnatal rat forebrain.
Neuron
10:201-212[Web of Science][Medline].
-
Lewis PD
(1968)
A quantitative study of cell proliferation in the subependymal layer of the adult rat brain.
Exp Neurol
20:203-207[Web of Science][Medline].
-
Lois C,
Alvarez-Buylla A
(1993)
Proliferating subventricular zone cells in the adult mammalian forebrain can differentiate into neurons and glia.
Proc Natl Acad Sci USA
9:2073-2077.
-
Luskin MB
(1993)
Restricted proliferation and migration of postnatally generated neurons derived from the forebrain subventricular zone.
Neuron
11:173-189[Web of Science][Medline].
-
Lynch T,
Sano M,
Marder KS,
Bell KL,
Foster NL,
Defendini RF,
Sima AAF,
Keohane C,
Nygaard TG,
Fahn S,
Mayeux R,
Rowland LP,
Wilhelmsen KC
(1994)
Clinical characteristics of a family with chromosome 17-linked disinhibition-dementia-parkinsonism-amyotrophy complex.
Neurology
44:1878-1884[Abstract/Free Full Text].
-
Malatesta P,
Hartfuss E,
Götz M
(2000)
Isolation of radial glial cells by fluorescent-activated cell sorting reveals a neuronal lineage.
Development
127:5253-5263[Abstract].
-
Martina M,
Vida I,
Jonas P
(2000)
Distal initiation and active propagation of action potentials in interneuron dendrites.
Science
287:295-300[Abstract/Free Full Text].
-
Marusich MF,
Furneaux HM,
Henion PD,
Weston JA
(1994)
Hu neuronal proteins are expressed in proliferating neurogenic cells.
J Neurobiol
25:143-155[Web of Science][Medline].
-
McCarthy JE,
Kollmus H
(1995)
Cytoplasmic mRNA-protein interactions in eukaryotic gene expression.
Trends Biochem Sci
20:191-197[Web of Science][Medline].
-
Miller RH,
Raff MC
(1984)
Fibrous and protoplasmic astrocytes are biochemically and developmentally distinct.
J Neurosci
4:585-592[Abstract].
-
Miyake T,
Okada M,
Kitamura T
(1992)
Reactive proliferation of astrocytes studied by immunohistochemistry for proliferating cell nuclear antigen.
Brain Res
590:300-302[Web of Science][Medline].
-
Morshead CM,
Reynolds BA,
Craig CG,
McBurney MW,
Staines WA,
Morassutti D,
Weiss S,
van der Kooy D
(1994)
Neural stem cells in the adult mammalian forebrain: a relatively quiescent subpopulation of subependymal cells.
Neuron
13:1071-1082[Web of Science][Medline].
-
Morshead CM,
Craig CG,
van der Kooy D
(1998)
In vivo clonal analyses reveal the properties of endogenous neural stem cell proliferation in the adult mammalian forebrain.
Development
125:2251-2261[Abstract].
-
Mukonoweshuro W,
Griffiths PD,
Blaser S
(1999)
Neurofibromatosis type 1: the role of neurology.
Neuropediatrics
30:111-119[Medline].
-
Munoz-Garcia D,
Ludwin SK
(1984)
Classic and generalized variants of Pick's disease: a clinicopathological, ultrastructural, and immunocytochemical comparative study.
Ann Neurol
16:467-480[Web of Science][Medline].
-
Nakamura M,
Okano H,
Blendy J,
Montell C
(1994)
Musashi, a neural RNA-binding protein required for Drosophila adult external sensory organ development.
Neuron
13:67-81[Web of Science][Medline].
-
Nakamura Y,
Sakakibara S,
Miyata T,
Ogawa M,
Shimazaki T,
Weiss S,
Kageyama R,
Okano H
(2000)
The bHLH gene Hes1 as a repressor of the neuronal commitment of CNS stem cells.
J Neurosci
20:283-293[Abstract/Free Full Text].
-
Nishiyama A,
Lin XH,
Giese N,
Helden CH,
Stallcup WB
(1996)
Co-localization of NG2 proteoglycan and PDGF
-receptor on O2A progenitor cells in the developing rat brain.
J Neurosci Res
43:299-314[Web of Science][Medline]. -
Noctor SC,
Flint AC,
Weissman TA,
Dammerman RS,
Kriegstein AR
(2001)
Neurons derived from radial glial cells establish radial units in neocortex.
Nature
409:714-720[Medline].
-
Ohtsuka T,
Ishibashi M,
Gradwohl G,
Nakanishi S,
Guillemot F,
Kageyama R
(1999)
Hes1 and Hes5 as Notch effectors in mammalian neuronal differentiation.
EMBO J
18:2196-2207[Web of Science][Medline].
-
Okabe M,
Imai T,
Kurusu M,
Hiromi Y,
Okano H
(2001)
Translational repression determines a neuronal potential in Drosophila asymmetric cell division.
Nature
411:94-98[Medline].
-
Okano H
(1995)
Two major mechanisms regulating cell-fate decisions in the developing nervous system.
Dev Growth Differ
37:619-629[Web of Science].
-
Okano HJ,
Darnell RB
(1997)
A hierarchy of Hu RNA binding proteins in developing and adult neurons.
J Neurosci
17:3024-3037[Abstract/Free Full Text].
-
Palmer TD,
Takahashi J,
Gage FH
(1997)
The adult rat hippocampus contains primordial neural stem cells.
Mol Cell Neurosci
8:389-404[Web of Science][Medline].
-
Petersen RB,
Tabaton M,
Chen SG,
Monari L,
Richardson SL,
Lynches T,
Manetto V,
Lanska DJ,
Markesbery WR,
Currier RD,
Autilio-Gambetti L,
Wilhelmsen KC,
Bambetti P
(1997)
Familial progressive subcortical gliosis: presence of prions and linkage to chromosome 17.
Neurology
45:1062-1067[Abstract/Free Full Text].
-
Pincus DW,
Keyoung H,
Restelli C,
Goodman RR,
Fraser RAR,
Edgar M,
Sakakibara S,
Okano H,
Nedergaad M,
Goldman SA
(1998)
FGF2/BDNF-responsive neuronal progenitor cells in the adult human subependyma.
Ann Neurol
43:576-585[Web of Science][Medline].
-
Reynolds BA,
Weiss S
(1996)
Clonal and population analyses demonstrate that an EGF-responsive mammalian embryonic CNS precursor is a stem cell.
Dev Biol
175:1-13[Web of Science][Medline].
-
Richter K,
Good PJ,
Dawid IB
(1990)
A developmentally regulated, nervous system-specific gene in Xenopus encodes a putative RNA-binding protein.
New Biol
2:556-565[Medline].
-
Roper SN,
Eisenschenk S,
King MA
(1999)
Reduced density of parvalbumin- and calbindin D28k-immunoreactive neurons in experimental cortical dysplasia.
Epilepsy Res
37:63-71[Web of Science][Medline].
-
Sakakibara S,
Okano H
(1997)
Expression of neural RNA-binding proteins in the postnatal CNS: implications of their roles in neuronal and glial cell development.
J Neurosci
17:8300-8312[Abstract/Free Full Text].
-
Sakakibara S,
Imai T,
Hamaguchi K,
Okabe M,
Aruga J,
Nakajima K,
Yasutomi Y,
Nagata T,
Kurihara Y,
Uesugi S,
Miyata T,
Ogawa M,
Mikoshiba K,
Okano H
(1996)
Mouse-Musashi-1, a neural RNA-binding protein highly enriched in the mammalian CNS stem cell.
Dev Biol
176:230-242[Web of Science][Medline].
-
Satoh H,
Nagai F,
Homma H,
Mori S,
Matsui M
(1993)
Regional assignment of rat androsterone UDP-glucuronosyltransferase gene (UGT2B2) to chromosome 14p21.2-p22.
Cytogenet Cell Genet
62:49-51[Web of Science][Medline].
-
Scheffler B,
Horn M,
Blumcke I,
Laywell ED,
Coomes D,
Kukekov VG,
Steindler DA
(1999)
Marrow-mindedness: a perspective on neuropoiesis.
Trends Neurosci
22:348-357[Web of Science][Medline].
-
Spillantini MG,
Geodert M
(1998)
Tau protein pathology in neurodegenerative diseases.
Trends Neurosci
21:428-433[Web of Science][Medline].
-
Sprinkle TJ
(1989)
2' 3'-Cyclic nucleotide 3'-phosphodiesterase: an oligodendrocyte-Schwann cell and myelin-associated enzyme of the nervous system.
Crit Rev Neurobiol
4:235-301[Web of Science][Medline].
-
Stinehelfer S,
Vruwink M,
Burette A
(2000)
Immunolocalization of mGluR1
in specific populations of local circuit neurons in the cerebral cortex.
Brain Res
861:37-44[Web of Science][Medline]. -
Streubel B,
Latta E,
Kehrer-Sawatzki H,
Hoffmann GF,
Fonatsch C,
Rehder H
(1999)
Somatic mosaicism of a greater than 1.7-Mb deletion of genomic DNA involving the entire NF1 gene as verified by FISH: further evidence for a contiguous gene syndrome in 17q11.2.
Am J Med Genet
87:12-16[Web of Science][Medline].
-
Swanson MS,
Dreyfuss G
(1988)
Classification and purification of proteins of heterogeneous nuclear ribonucleoprotein particles by RNA binding proteins.
Mol Cell Biol
3:2237-2241.
-
Tamamaki N,
Fujimori KE,
Takauji R
(1997)
Origin and route of tangentially migrating neurons in the developing neocortical intermediate zone.
J Neurosci
17:8313-8323[Abstract/Free Full Text].
-
Tan SS,
Kalloniatis M,
Sturm K,
Tam PPL,
Reese BE,
Faulkner-Jones B
(1998)
Separate progenitors for radial and tangential cell dispersion during development of the cerebral neocortex.
Neuron
21:295-304[Web of Science][Medline].
-
Thomas WE
(1992)
Brain macrophages: evaluation of microglia and their functions.
Brain Res Rev
17:61-74[Medline].
-
van Hooft JA, Giuffrida R, Blatow M, Monyer
H (2000) Differential expression of group I metabotropic
glutamate receptors in functionally distinct hippocampal interneurons.
20:3544-3551.
-
Wakamatsu Y,
Maynard TM,
Jones SU,
Weston JA
(1999)
NUMB localizes in the basal cortex of mitotic avian neuroepithelial cells and modulates neuronal differentiation by binding to NOTCH-1.
Neuron
23:71-81[Web of Science][Medline].
-
Whittington MA,
Traub RD,
Jefferys JG
(1995)
Synchronized oscillations in interneuron networks driven by metabotropic glutamate receptor activation.
Nature
373:612-615[Medline].
-
Yao K-M,
Samson M-L,
Reeves R,
White K
(1993)
Gene elav of Drosophila melanogaster: a prototype for neuron-specific RNA binding protein gene family that is conserved in flies and humans.
J Neurobiol
24:723-739[Web of Science][Medline].
-
Yoda A,
Sawa H,
Okano H
(2000)
MSI-1, a neural RNA-binding protein, is involved in male mating behavior in Caenorhabditis elegans.
Genes Cells
5:1-12[Abstract].
-
Zhang W,
Wagner BJ,
Ehrenman K,
Schaefer AW,
DeMaria CT,
Crater D,
DeHaven K,
Long L,
Brewer G
(1993)
Purification, characterization, and cDNA cloning of an AU-rich element RNA-binding protein, AUF1.
Mol Cell Biol
13:7652-7665[Abstract/Free Full Text].
Copyright © 2001 Society for Neuroscience 0270-6474/01/21208091-17$05.00/0
This article has been cited by other articles:

|
 |

|
 |
 
U Shivraj Sohur, J. G Emsley, B. D Mitchell, and J. D Macklis
Adult neurogenesis and cellular brain repair with neural progenitors, precursors and stem cells
Phil Trans R Soc B,
September 29, 2006;
361(1473):
1477 - 1497.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
N. A. Siddall, E. A. McLaughlin, N. L. Marriner, and G. R. Hime
The RNA-binding protein Musashi is required intrinsically to maintain stem cell identity
PNAS,
May 30, 2006;
103(22):
8402 - 8407.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
A. Ratti, C. Fallini, L. Cova, R. Fantozzi, C. Calzarossa, E. Zennaro, A. Pascale, A. Quattrone, and V. Silani
A role for the ELAV RNA-binding proteins in neural stem cells: stabilization of Msi1 mRNA.
J. Cell Sci.,
April 1, 2006;
119(Pt 7):
1442 - 1452.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
Y. Sugiyama-Nakagiri, M. Akiyama, S. Shibata, H. Okano, and H. Shimizu
Expression of RNA-Binding Protein Musashi in Hair Follicle Development and Hair Cycle Progression
Am. J. Pathol.,
January 1, 2006;
168(1):
80 - 92.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
D. N. Abrous, M. Koehl, and M. Le Moal
Adult Neurogenesis: From Precursors to Network and Physiology
Physiol Rev,
April 1, 2005;
85(2):
523 - 569.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
Y. Sasakura, S. Awazu, S. Chiba, and N. Satoh
Germ-line transgenesis of the Tc1/mariner superfamily transposon Minos in Ciona intestinalis
PNAS,
June 24, 2003;
100(13):
7726 - 7730.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
A. Barbouti, M. Hoglund, B. Johansson, C. Lassen, P.-G. Nilsson, A. Hagemeijer, F. Mitelman, and T. Fioretos
A Novel Gene, MSI2, Encoding a Putative RNA-binding Protein Is Recurrently Rearranged at Disease Progression of Chronic Myeloid Leukemia and Forms a Fusion Gene with HOXA9 as a Result of the Cryptic t(7;17)(p15;q23)
Cancer Res.,
March 15, 2003;
63(6):
1202 - 1206.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
S.-i. Sakakibara, Y. Nakamura, T. Yoshida, S. Shibata, M. Koike, H. Takano, S. Ueda, Y. Uchiyama, T. Noda, and H. Okano
RNA-binding protein Musashi family: Roles for CNS stem cells and a subpopulation of ependymal cells revealed by targeted disruption and antisense ablation
PNAS,
November 12, 2002;
99(23):
15194 - 15199.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
Y. Konishi, K. Lindholm, L.-B. Yang, R. Li, and Y. Shen
Isolation of Living Neurons from Human Elderly Brains Using the Immunomagnetic Sorting DNA-Linker System
Am. J. Pathol.,
November 1, 2002;
161(5):
1567 - 1576.
[Abstract]
[Full Text]
[PDF]
|
 |
|
|

|