| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Pediatrics,*
Jichi Medical
School, Tochigi; the Department of Mental Retardation and Birth Defect
Research,
National Institute of Neuroscience,
NCNP, Kodaira; the Department of Pathology,
Brain Research Institute, Niigata University, Niigata; and the
Department of Ultrastructure and
Histochemistry,§
Tokyo Insitute of Psychiatry,
Tokyo, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterals and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterals and MethodsResultsDiscussionReferences |
|---|
Recent advances in molecular genetics have revealed the gene defects causing many cases of lissencephaly. MDS is a contiguous gene syndrome that results from a microdeletion of chromosome 17p13.3 encompassing the LIS1 gene.5,6 This gene encodes the ß subunit of brain type platelet-activating factor acetylhydrolase (PAFAH1B1),7 which is present in abundance in normal brains but is deficient in those of patients with MDS.8,9 On the other hand, ILS consists of heterogeneous conditions. Approximately 40% of sporadic ILS patients have a deletion of 17p13.3 involving the LIS1 gene. Among those without such a deletion, 40% have a mutation or an intragenic deletion in the LIS1 gene, whereas the other 20% have a mutation in the XLIS (or DCX) gene located on chromosome Xq22.3-q23.10,11 Most of the latter patients with X-linked lissencephaly are male and hemizygous for the mutated XLIS gene. Morphological features of the brains of patients with an XLIS mutation resemble those with a LIS1 mutation except for several minor differences.11,12 X-linked female patients heterozygous for the same mutation usually show a milder phenotype, termed subcortical laminar heterotopia (SCLH), band heterotopia, or double cortex syndrome.13 The spectrum of these X-linked neuronal migration disorders is therefore called X-linked SCLH and lissencephaly syndrome.
The XLIS gene has recently been cloned.14,15 The predicted gene product, doublecortin (DC), is a novel protein with a molecular weight of about 40 kd. DC includes a potential Abl phosphorylation site at tyrosine residue 70 and other sites of potential phosphorylation. The absence of a signal peptide and a transmembrane domain suggests that DC is a hydrophilic, intracellular protein. DC shows high homology to KIAA0369 (KI), which has previously been isolated from a human fetal brain cDNA library.16 KI is an 80-kd protein encoded by a gene on chromosome 13q13-q14.1, consisting of an amino-terminal DC-like domain and a carboxy-terminal calcium calmodulin-dependent kinase-like domain.17 The DC and KI mRNAs are expressed at high levels in the fetal brains, which is compatible with their roles in brain morphogenesis.14,15,17 It has been speculated that the DC and KI proteins interact in a competitive fashion in a signaling pathway regulating neuronal migration.14,15,18 With regard to the protein expression, however, there has been no information available about the identification and characterization of these proteins.
In this study, we produced antibodies that specifically recognize DC and KI and used them to delineate the temporal and spatial patterns of expression of these proteins in normally developing brains of humans and rats.
|
Materals and Methods |
|---|
|
TopAbstractIntroductionMaterals and MethodsResultsDiscussionReferences |
|---|
Antisera were raised in rabbits against peptides synthesized according to the sequence deduced from human DC and KI cDNAs.14,15,16 The peptide fragments of DC consisted of the amino acid residues 1–10 plus a carboxy-terminal cysteine (peptide DC1; MELDFGHFDEC) 347–360 plus an amino-terminal cysteine (peptide DC2; CPLSLDDSDSLGDSM) and 271–284 (peptide DC3; CRVMKGNPSATAGP), and those of KI consisted of the residues 1–12 plus a carboxy-terminal cysteine (peptide KI1; MSFGRDMELEHFC) and 276–287 (peptide KI3; CRVVKSTSYTKI). The DC3 and KI3 peptides were mutually homologous: 5 of the 14 amino acids in the former and 6 of the 12 in the latter were shared by DC and KI.14-16 Peptide synthesis and immunization, as well as the collection and titration of the antisera, were performed as described previously.8 Some aliquots of sera were absorbed with the corresponding peptides and used for negative control experiments.8
Western Blotting
Human tissues for Western blotting were obtained from 10 patients ranging in age from 12 gestational weeks (GW) to 85 years. These patients died of non-neurological diseases, and necropsy was done within 12 hours postmortem. The tissues were kept frozen at -80°C until use.
Cerebral tissues were also removed from Wistar rats ranging from E14 to postnatal day 60 (P60) of age under anesthesia by inhalation of ether.
To study protein expression in tissue cultures, rat cerebral neurons and astrocytes were isolated as described previously19 and cultured for 3 days in vitro. We also used human neural cell lines, such as the neuronally differentiated NT2 teratocarcinoma cells,20 NB1 and TGW-III neuroblastoma cells, and I-23 and Bu17 glioma cells, as well as non-neural cells, such as the HeLa cervical carcinoma cells and U937 monocytic leukemia cells.
Protein extraction, sodium dodecyl sulfate polyacrylamide electrophoresis, and Western blotting were performed as described previously.8,21 Briefly, proteins were extracted with Tris/saline buffer containing 1% Triton X-100, and were assayed by Bradford’s method. They were then separated on a 10% polyacrylamide gel and blotted onto a polyvinylidene fluoride membrane. To detect DC and KI immunologically, the membrane was probed with one of the anti-DC/KI antibodies overnight at 4°C. The dilution used was 1:1200 for anti-DC1 and anti-DC2, 1:200 for anti-DC3 and anti-KI3, and 1:500 for anti-KI1. Successsive incubations were then done with biotinylated anti-rabbit IgG (Vector, Burlingame, CA; diluted 1:1,000) and a horseradish peroxidase-conjugated avidin-biotin complex (Vector), both at room temperature, for 1 hour. Diaminobenzidine HCl (DAB; Dojin, Osaka, Japan) was then used as the chromogen. To confirm that the samples loaded onto each lane contained an equal amount of protein (40 µg), unused portions of the transfer membranes were stained with Coomasie Brilliant Blue R-250.
Subcellular Fractionation
Crude subcellular fractionation was performed by differential centrifugation, as described elsewhere.8 Each fraction was cytochemically identified by Feulgen, methylgreen-pyronin, and Janus green stainings.
Immunohistochemistry
Formalin-fixed, paraffin-embedded sections were prepared from tissues of 12 control patients without a neurological disease, from 6 GW to 75 years of age. For experiments in rats, brain tissues of Wistar rats from E11 to P60 of age were fixed overnight in Bouin’s solution and embedded in paraffin.
Immunoperoxidase staining was performed by the biotin-streptoavidin method.22 Briefly, the slides were incubated with the first antibodies against DC and KI (diluted 1:200), and then processed using an S-HRP immunostaining kit (Seikagaku, Tokyo, Japan). The immunoproducts were visualized with DAB. No counterstaining was done.
|
Results |
|---|
|
TopAbstractIntroductionMaterals and MethodsResultsDiscussionReferences |
|---|
The specificity of the anti-DC/KI antibodies was first determined
by Western blotting of cerebral tissue proteins. On the blots of human
fetal cerebral homogenates, the anti-DC and anti-KI antibodies yielded
a doublet of bands at about 45 and 80 kd, respectively (Figure 1A)
. These proteins were identified as DC
and KI, because they were consistently recognized by the appropriate
antibodies and the molecular weights coincided with those predicted
previously.14-16
The anti-KI antibodies occasionally
yielded an additional, weak 45-kd band, which may have resulted from a
cross-reaction with DC. No band appeared on the blots probed with the
preimmune or preabsorbed antisera (Figure 1B)
.
|
Tissue distribution of DC and KI was studied by Western blotting
of various organs of human fetuses (Figure 2)
. Both antigens were present in the
central nervous system (CNS) and were most abundant in the cerebrum.
Other organs did not contain a detectable amount of these proteins,
although the adrenal glands showed a weak KI immunoreactivity.
|
. The anti-DC
and anti-KI antibodies cross-reacted with the rat antigens. In tissue
culture, the rat cerebral neurons and neuronally differentiated human
NT2 cells contained both DC and KI. DC was undetectable in the cultures
of non-neural cells, whereas a small amount of KI was present in the
astrocytic and neuroblastoma cells.
|
. Both proteins
were recovered in all of the fractions, but were most abundant in the
mitochondrion-rich and microsome-rich fractions.
|
. In humans, both the
proteins were present in fetuses, and were most abundant at about 20
GW. DC was undetetectable in the postnatal brains, whereas a low level
of KI expression persisted until adulthood (Figure 5A)
. Similar changes
were observed in rats. Both DC and KI were abundant in the fetal brains
and decreased in amount during the postnatal period. There was a
detectable amount of KI, but not of DC, in the adult brain (Figure 5B)
.
|
The anti-DC1, anti-DC2, anti-KI1, and anti-KI3 antibodies
positively labeled the CNS of rat (Figures 6–8)
and human (Figure 9)
fetuses.
Histologically, immunoreactivity for DC and KI was localized mainly in
the perikarya of cortical neurons and in the processes in the cortical
plate and intermediate zone. The staining patterns were essentially the
same with the anti-DC1 and anti-DC2 antisera. This was also the case
with the anti-KI1 and anti-KI3, although the intensity was stronger
with the former antibody (Figure 7)
. The
immunohistochemical distributions of DC and KI were similar (Figures 7–10)
. In negative control experiments with preimmune or preabsorbed
sera, there was no positive labeling (data not shown).
|
|
|
|
. The most
intense immunoreactivity was localized in the cell body of the
Cajal-Retzius cells, neurons that appear first in the cortical plate
(Figure 9, A and E)
. At E17, the
ventricular zone lost the immunoreactivity for DC, whereas the cortical
plate and intermediate zone remained strongly positive (Figure 8B)
.
Intensely labeled neuronal cell bodies were lined up in the innermost
layer of the cortex (Figure 9B)
. A small number of DC-positive cell
bodies were also present in the intermediate zone. Strongly
immunoreactive processes ran vertically in the cortex and horizontally
in the intermediate zone (Figures 8B and 9B)
. Immunoreactivity for KI
showed a similar distribution, although a low level of labeling
persisted in the ventricular zone (Figures 8F and 9F)
. At E20-P0,
neuronal perikarya became negative for DC and KI. The immunoreactive
processes broke up, resulting in a fine granular labeling that was most
prominent in the subpial layer (Figures 8C, 8G, and 9C)
. After P5, the
only DC-positive structure was a subset of astrocytes which were most
numerous in the subpial layer (Figures 8D and 9D)
. There was no
immunoreactivity for KI (Figure 8H)
.
We next performed a developmental study in the human brain. In the
cerebrum of a 6-GW-old fetus, there was no immunoreactivity for DC and
KI (data not shown). At 12 to 17 GW, the entire thickness of the
cerebrum was stained strongly for DC and moderately for KI. The
labeling was most intense in the cortex. These distribution patterns
were reminiscent of those in rat aged E14–17, although the labeling in
humans was mostly confined to the perikarya and proximal processes
(Figure 10, A
-D). DC and KI
immunoreactivities became restricted to the cortex at 22 to 26 GW
(Figure 10, C and G)
and was mostly lost from the brain at 33 GW and
thereafter (Figure 10, D and H)
, except for scattered astrocytes that
were moderately positive for DC (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterals and MethodsResultsDiscussionReferences |
|---|
KI mRNA is also abundant in the fetal brain. Unlike DC, the KI transcript is expressed at a low level in the adult brain.14,15,17 In a Northern blot analysis of adult human tissues, the expression of the KI mRNA was confined to the nervous system.17 The cell type-specific localization of KI has not been studied.
The present study further extended our knowledge about the developmental expression pattern of DC and KI by investigating their distibution at the protein level. We identified the DC and KI proteins for the first time by means of Western blotting using multiple antibodies that recognize different epitopes. The molecular weights of DC and KI were estimated to be 45 and 80 kd, respectively, values close to those which were expected based on the results of cDNA cloning.14-16 DC and KI were recognized on Western blots as doublet proteins, which may have resulted from alternative splicing14,15,17 or posttranslational modification. The reported splicing of the DC gene, however, does not affect the predicted open reading frame,14 and a Northern blot analysis detected only one message size.15 Studies are underway in our laboratory to reveal whether DC and KI undergo phosphorylation and other posttranslational modifications by producing additional antibodies, including those raised against phosphorylated and non-phosphorylated peptides corresponding to their potential phosphorylation sites.
Regarding the temporal pattern of expression, our Western blotting data on the DC and KI proteins showed patterns essentially identical to those reported previously for their mRNAs.14,15,17 These proteins resembled each other in that they were most abundant during the period of neuronal migration. The immunohistochemical distributions of DC and KI were also similar. In the early phase of neuronal dispersion (E14 in rats and 12–17 GW in humans), a robust expression of these proteins was noted in all of the zones: ventricular, intermediate, and cortical. During the subsequent stages (E17–20 in rats and 22–26 GW in humans), their immunoreactivities gradually became confined to the superficial layers. The strong DC and KI immunoreactivities of certain subsets of rat cortical neurons, such as the Cajal-Retzius cells at E14 and neurons in the innermost layer at E17, suggest a pivotal role of these cells in regulating cortical lamination. After the main surge of migrating neurons had arrested (after P5 in rats and after 33 GW in humans), the two proteins became hardly detectable on immunohistochemical sections. Thus, DC and KI tended to colocalize in the zones of ongoing neuronal migration, suggesting that they participate in the same signaling pathway regulating the entirety of this developmental process.
The immunohistochemical findings were somewhat different in the rat and
human cerebrum. In rats, many processes distant from the cell bodies
were positive for DC and KI (Figure 9, B and F)
, whereas in humans,
labeling tended to be restricted to the perisomal area (Figure 10, A and E)
. This difference may have resulted from agonal and postmortem
changes to which human specimens are liable. We also noted several
minor discrepancies between the Western blotting and immunostaining
results. First, immunohistochemistry failed to detect the KI protein in
the mature cerebrum (Figure 8H)
, where it was shown by Western blotting
to remain in a small amount (Figure 5)
. Second, Western blotting
detected DC and KI in the nucleus-rich fraction (Figure 4)
, whereas the
nuclei were unlabeled on the immunohistochemical sections (Figures 8–10)
. These discrepancies are probably due to the different
sensitivities of the two methods, although the second may have resulted
from impurity of the subcellular fraction. Third, a subset of
astrocytes disclosed a DC-like immunoreactivity on the
immunohistological sections of the cerebrum at the postmigratory
stages (Figure 9D)
, which was not detected by the Western blotting
(Figure 5)
. Although the astrocytic immunoreactivity could have
resulted from a cross-reactivity with a different antigen, it appears
worth examining the possibility that some astrocytes can indeed express
DC.
It is interesting that KI showed a somewhat wider distribution on
Western blots compared to DC, both temporally and spatially. KI
remained present in the brain until adulthood, whereas DC became
undetectable during early postnatal development (Figure 5)
. During the
fetal period, DC was expressed exclusively in the CNS, whereas KI was
also present in the adrenal gland (Figure 2)
. In tissue culture, the
expression of DC was confined to CNS neurons, whereas KI was produced
also by astrocytic and neuroblastoma cells (Figure 3)
. The presence of
small amounts of KI in these cells suggests that KI is involved in
their modest motility or in their nonmigratory activities during the
postnatal period. By contrast, the function of DC appears to be
specific to the migration of fetal neurons.
The mechanism by which DC and KI regulate neuronal migration will be the main subject for future studies. An attractive theory is that DC regulates the function of KI via a competition for binding to an effector protein, or via a modulation of the kinase activity of KI.14,15 The identification of this effector would be difficult, but PAFAH1B1 and other molecules involved in neuronal migration disorders could be candidates. A clue to this issue may be obtained by studying the subcellular localization of DC and KI and correlating the findings with those of the candidate proteins. It also appears important that DC is a potential substrate for Abl and a possible substrate for members of the mitogen-activated protein kinase family.14,15 To elucidate the precise roles of DC and KI in CNS development, it may be worth performing two lines of experiments: analysis of their Abl-dependent phophorylation and immunoelectron microscopic observations.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by a Grant-in-Aid for Scientific Research (10670753) from the Ministry of Education, Science and Culture, Japan, and by Grants for Research on Brain Science (H10-Brain-28), on Nervous and Mental Disorders (8A-10–37) and on Neurocutaneous Diseases (3310) from the Ministry of Health and Welfare, Japan.
Accepted for publication July 15, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterals and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Evangelisti, M. C. Florian, I. Massimi, C. Dominici, G. Giannini, S. Galardi, M. C. Bue, S. Massalini, H. P. McDowell, E. Messi, et al. MiR-128 up-regulation inhibits Reelin and DCX expression and reduces neuroblastoma cell motility and invasiveness FASEB J, December 1, 2009; 23(12): 4276 - 4287. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ohmae, S. Takemoto-Kimura, M. Okamura, A. Adachi-Morishima, M. Nonaka, T. Fuse, S. Kida, M. Tanji, T. Furuyashiki, Y. Arakawa, et al. Molecular Identification and Characterization of a Family of Kinases with Homology to Ca2+/Calmodulin-dependent Protein Kinases I/IV J. Biol. Chem., July 21, 2006; 281(29): 20427 - 20439. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Edelman, W.-Y. Kim, D. Higgins, E. G. Goldstein, M. Oberdoerster, and W. Sigurdson Doublecortin Kinase-2, a Novel Doublecortin-related Protein Kinase Associated with Terminal Segments of Axons and Dendrites J. Biol. Chem., March 4, 2005; 280(9): 8531 - 8543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mizumatsu, M. L. Monje, D. R. Morhardt, R. Rola, T. D. Palmer, and J. R. Fike Extreme Sensitivity of Adult Neurogenesis to Low Doses of X-Irradiation Cancer Res., July 15, 2003; 63(14): 4021 - 4027. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Friocourt, A. Koulakoff, P. Chafey, D. Boucher, F. Fauchereau, J. Chelly, and F. Francis Doublecortin Functions at the Extremities of Growing Neuronal Processes Cereb Cortex, June 1, 2003; 13(6): 620 - 626. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mizuguchi, S. Takashima, K. Ikeda, M. Kato, and A. Hori Loss of doublecortin in heterotopic gray matter of a fetus with subcortical laminar heterotopia Neurology, July 9, 2002; 59(1): 143 - 144. [Full Text] [PDF]
|
|
|
|
|
|
H. A. Burgess and O. Reiner Alternative Splice Variants of Doublecortin-like Kinase Are Differentially Expressed and Have Different Kinase Activities J. Biol. Chem., May 10, 2002; 277(20): 17696 - 17705. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Meyer Book Review: Human Neocortical Development: The Importance of Embryonic and Early Fetal Events Neuroscientist, August 1, 2001; 7(4): 303 - 314. [Abstract] [PDF]
|
|
|
|
 |
|
 P. T. Lin, J. G. Gleeson, J. C. Corbo, L. Flanagan, and C. A. Walsh DCAMKL1 Encodes a Protein Kinase with Homology to Doublecortin that Regulates Microtubule Polymerization J. Neurosci., December 15, 2000; 20(24): 9152 - 9161. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. Burgess and O. Reiner Cleavage of Doublecortin-like Kinase by Calpain Releases an Active Kinase Fragment from a Microtubule Anchorage Domain J. Biol. Chem., September 21, 2001; 276(39): 36397 - 36403. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
