вход по аккаунту


Enhanced expression of microtubule-associated protein 2 in large neurons of cortical dysplasia.

код для вставкиСкачать
Enhanced Expression of
Microtubule-associated Protein 2 in Large
Neurons of Cortical Dysplasia
Hideo Yamanouchi, MD, Weixian Zhang, MD, Venita Jay, MD, and Laurence E. Becker, MD
To evaluate neuronal cytoarchitectural changes in cortical dysplasia, we examined microtubule-associated protein 2
(MAP2) expression in surgically resected specimens obtained from 20 patients (age range, 3 months to 10 years) treated
for intractable epilepsy. Large neurons were investigated in the specimens from all patients and showed significantly
strong immunoreactivity with antibodies against MAP2 in the perikaryon and proximal portion of their processes. In
situ hybridization with MAP2 antitense riboprobe showed increased hybridization signal intensities in the large neurons,
which correlated with the pattern of immunoreactivity for MAP2. We conclude that MAP2 is strongly expressed in the
large neurons in cortical dysplasia. The results of preliminary immunoblotting in 1 patient with focal cortical dysplasia
showed that the low-molecular-weight form of MAP2 (MAP2c) was strongly expressed in the dysplastic cortex, suggesting
that MAP2c may be a major component contributing to the increased expression of MAP2 in the large neurons of
cortical dysplasia. Since it has been suggested that MAP2 plays a crucial role in the branching and remodeling of
neuronal processes, increased expression of MAP2 may reflect activated plasticity of the large neurons in cortical dyspasia.
Yamanouchi H, Zhang W, Jay V, Becker LE. Enhanced expression of microtubule-associated protein 2 in
large neurons of cortical dysplasia. Ann Neurol 1996;39:57-6 1
Improvements in diagnostic imaging have assisted in
identifying subtle cortical anomalies in patients with
intractable epilepsy. Pathologic examination of surgical
resections of those anomalies associated with epilepsy
has broadened our knowledge of cortical dysplasia. The
term cortical dysplasia is used to describe a malformed
cortex characterized by disorganization of the cortical
architecture with abnormalities of neuronal size, shape,
and lamination [I]. Cortical dysplasia falls under the
category of “neuronal migration disorders,” which can
encompass subtle neocortical abnormalities or more extensive cortical disorganization such as lissencephaly,
polymicrogyria, and hemimegalencephaly [I].
Large neurons, a common cytological feature in cortical dysplasia, measure usually greater than 200 bm2,
have enlarged nuclei with a single prominent nucleolus,
and are commonly located in the subcortical white
matter and cerebral cortex [I]. These neurons may
have numerous processes extending from the cell body
[2]. Morphologic studies have emphasized the cytoskeleta1 changes in large neurons. They have a coarse intracytoplasmic fibrillar structure and are strongly stained
with Bielschowsky silver technique [ 1, 21. They show
positive immunoreactivity with antibodies against
phosphorylated neurofilaments, T protein, and ubiqui-
tin [2]. However, they show negative immunoreactivity
with antibody against paired helical filament, which are
known to contain abnormally phosphorylated forms of
z protein, and thus are different from the neurofibrillary tangles of Alzheimer’s disease [2].
As another approach to the analysis of cytoskeletal
changes in cortical dysplasia, in this study, we focus
on microtubule-associated protein 2 (MAP2) expression, as MAP2 is one of the most abundant neuronal
microtubule-associated proteins [3].MAP2 promotes
tubulin polymerization into the microtubules, and microtubule assembly or reassembly. As a neuronal cytoskeletal protein, MAP2 is thought to be a crucial
regulator of outgrowth, plasticity, and maintenance of
neuronal processes [4-71.
High- (-280 kd) and low-molecular-weight (-70
kd) forms of MAP2 have been identified. They are
developmentally regulated and derived from separate
mRNAs, which are transcribed by alternate splicing
from a single MAP2 gene [3]. The high-molecularweight form of MAP2 is abundant in the adult brain,
whereas the low-molecular-weight form (MAP2c) is expressed during the embryonic and early postnatal period and its level in the adult brain is extremely low
[3].Thus, it has been speculated that MAP2c plays a
From the Department of Pathology, The Hospital for Sick Children, Toronto, Ontario, Canada
Address correspondence to Dr Becker, Department of Pathology,
The Hospital for Sick Children, 555 University Avenue, Toronto,
Ontario, Canada M 5 C 1x8.
Received Jul 7, 1995, and in revised form Aug 31. Accepted for
publication Aug 31, 1995.
Copyright 0 1996 by the American Neurological Association 57
role i n the growth a n d sprouting of neuronal processes
in immature neurons [8, 91.
These significant roles of MAP2 as a neuronal cytoskeletal protein prompted us to examine its expression in cortical dysplasia. T h e results of immunohistochemistry, in situ hybridization, a n d immunoblotting
demonstrated here show that a n abnormal expression
of MAP2 may exist i n the large neurons in cortical
Materials and Methods
Tissue Specimens
Surgical samples were retrospectively selected from 20 patients (14 boys, 6 girls; age range, 3 months to 10 years)
with a pathologic diagnosis of cortical dysplasia who were
treated for intractable epilepsy from 1985 to 1994 at The
Hospital for Sick Children, Toronto, Ontario, Canada. They
consisted of 6 patients with hernimegalencephaly who underwent hemispherectoniy and 14 patients with localization-related epilepsy whose resected specimens showed focal cortical
dysplasia. The surgical specimens were immediately fixed in
10% neutral buffered formalin for 24 hours, cut into three
to 25 pieces depending on the size of the specimen, then
embedded in paraffin blocks and sectioned serially at 5 pm.
These sections were used for iinmunohistochemistry and in
situ hybridization as well as routine stains with Lux01 fast
blue (LFB) and hematoxylin and eosin (H&E). Histologically normal cortex away from lesional tissue was available
in each case and served as normal controls.
The sections were pretreated with 1Yo H,O, in methanol for
30 minutes. Monoclonal antibody against MAP2 (HM-2,
Sigma, St Louis, M O ) was used as a primary antibody. H M 2 recognizes both the high-molecular-weight form of MAP2
and MAP2c. Blocking was accomplished with 2% normal
horse serum. The sections were incubated for 1 hour with
HM-2 (diluted 1 :80) and subsequently stained with the avidin-biotin peroxidase complex method (Vectastain ABC kit,
Vector Co, Burlingame, CA). The immunoreactivity was visualized with 3,3'-diaminobenzidine tetrahydrochloride as
chromogen, followed by counterstaining with hematoxylin.
Cloning of cDNA
According to the tnethods by Chomczynski and Sacchi [lo],
total RNA was extracted from the autopsied fresh human
cerebellum in a 21-month-old patient who had been healthy
until he died of accidental trauma. We chose a 785-base pair
segment of human MAP2 cDNA sequence 3,844 to 4,628
[ 111 for cloning. The chosen segment consisted of a sequence
specific to the high-molecular-weight form of MAP2 (from
sequence 3,844 ro 4,509) and a sequence common to :he
high-molecular-weight form of MAP2 and MAP2c (from sequence 4,5 10 to 4,628) [l ll. Two oligonucleotides, 5'ATGGAGGAAGGTCTTGGCAG 3' and 5' ACCATCGAGGATGATTTCAT 3', were synthesized as primers, and used for
polymerase chain reaction (PCR) according to the standard
methods [ 121. Each PCR cycle consisted of denaturation at
94°C for 1.5 minutes, annealing at 51°C for 2 minutes, and
58 Annals of Neurology
Vol 39
No 1 January 1996
polymerization at 72°C for 2 minutes. The I'CR product
was cloned into pBluescript I1 KS (Stratagene Cloning Systems, La Jolla, CA) at the site digested with EcoRV.
Preparation o f Riboprobe and In Situ Hybridization
The cloned vector was linearized and purified by phenol/
chloroform extraction and ethanol precipitation; then digoxigenin-labeled sense and antisense MAP2 riboprobes were
transcribed from the vector with T7 and T3 promoters, respectively. The methods for in situ hydridization were essentially according to the methods o f Simmons and co-workers
[ 131. Hybridization was carried out overnight (approximately
18 hours) at 40°C with the MAP2 riboprobes in hybridization buffer containing 2 X SSC (saline-sodium citrate) 50%
deionized formamide, 10% dextran sulfate, 1 X Denhardt
solution, 500 pglml single-stranded salmon sperm DNA,
and 10 mM dithiothreitol. After hybridization and wash,
the sections were incubated with anti-digoxigenin antibody
conjugated with alkaline phosphatase diluted to 1 : 500
(Boehringer Mannheim, Mannheim, Germany) at room
temperature for 1 hour. Nitro blue tetrazolium chloride was
used as chroniogen to visualize the hybridized signal.
Two frozen specimens (stored at -70°C) from a 7-monthold patient were available for immunoblotting. They were
taken from different sites of the resected temporal lobe. The
neuropathological diagnosis based on conventional stains was
focal cortical dysplasia. Frozen specimens were sectioned at
7 pm, and fixed with acetone and stained with H&E. One
specimen showed normal-appearing cortex and the other
demonstrated dysplastic cortex with large neurons. We used
the former specimen as a control for the latter. About 0.5
gm of each specimen was homogenized well in a 1-ml extraction buffer (140 m M NaCI, 1.0 m M EGI'A, 1 niM MgCI,,
10 m M Tris, p H 8.0, and 0.5 m M phenylmethylsulforiyl
fluoride), then placed in a boiling water bath for 5 minutes.
Samples were centrifuged at 2,000 g for 30 minutes a: 4"C,
and the supernatants were collected. Five mici-ogranis of protein from each sample were separated on a 6Yo sodium doclecyl sulfate-polyacrylamide gel and subsequently transferred to
polyvinylidene difluoride microporous membrane (Millipore,
Bedford, MA). After an incubation with HM-2 (diluted 1 :
500), the band detection was performed according to the
methods of Arias and colleagues [14]. The membrane was
subsequently stained with 0.1% Coomassie Blue to ensure
equal loading of the samples.
Microscopic observations of the specimens from 14 patients with focal cortical dysplasia fulfilled the pathologicaI criteria [I]. They showed poorly identified cortical lamination, blurred junction of the cortex and
white matter, a n d the presence of large neurons a n d
balloon cells. T h e specimens from G patients with
hemimegalencephaly also showed these characteristic
microscopic features. Large neurons had abundant cytoplasm with enlarged r o u n d nuclei containing a single
Fig 1. ~nlmunohistoc~iemistly(A-C) and in situ hybridiation (0-F) of niicrotubule-associated protein 2 (MAP.?) i n a surgirid
specimen .fioni a n 8-month-old boy with fical cortical dysplasia. (A) Low-power view (X 25 b ~ o r e30% reduction) of immunobistochernistrv. Intense(v immunoreactii~elarge neurons sprerld .from the deep layer of the cortex to the subcortical iuhite matter. (B)
Hig/~-powerviezo (X 200 before 30% reduction) of t h e boxed portion in A. (C) High-power view (X 400 before 30% reduction)
o f norma/-appearing cortex showing typical imn7ttnoreactivity of neurons f o r MAP2. This nricrophotogr~lpiJ u1as tlikrn at a d(firent location in the cortex specimen @om the same patient. (0)
Lou]-poiuer uietu (X 25 bpfore 30% rrductiori) of i n situ hybridization. Large neurons iuitll increased hybridized signal intensity IJave similar pattern to that shown i n A. (E) Highpower view
(X 300 before 30% reduction) of the boxed portion in D, showing that hrge neurons hniie selectively increrzsed I',llbridized signiil
intensity. (F) High-power view (X 300 before 30% reduction) of normal-appearing c0rte.y showing control signal intemity hybridization.
prominent nucleolus. They presented as clusters or single neurons in the cortex-white-niatter junction and
subcortical white matter, as well as in the cortex. They
sometimes had intracytoplasmic vacuoles. Balloon cells
had an abundant glassy eosinophilic cytoplasm with
usually one, occasionally two or more, irregularly
shaped, displaced nuclei. They commonly appeared in
clusters and were usually located in the white matter
and scattered within the cortex.
Im m uriohistoi-hemistry
In the normal-appearing cortex, iinmunoreactivity foiMAP2 was intense in the apical dendrites of neurons
and faint in the perikaryon. Almost all large neurons
had more intense immunoreactivity for MAP2 in their
perikaryon and neuronal processes compared with
those of the controls (see Fig 1, A-C). Some balloon
cells had positive immunoreactivity for h4AP2, but its
intensity was weaker than that in normal neurons.
In Situ Hybridizntion of MAP2
T w o different concentrations (1 pg/nil and 100 ngl
nil) of MAP2 riboprobe were applied for in situ hybridization. I n the 1 pg/ml concentration, MAP2 antisense riboprobe hybridized well in the proximal part of
the neuronal processes and perikaryon in the normalappearing neurons and large neurons. The hybridization intensity in the normal-appearing neurons was
weak when the concentration of the riboprobe was 100
ng/ml, whereas the hybridization signal intensity in the
large neurons was still strong in this low concentration
of the riboprobe (see Fig 1, D-F). MAP2 sense riboprobe always had a negative signal.
Yamanouchi et al: MAP? in Cortical Dysplasia
Fig 2. Immunoblotting of resected specimen homogenates
stained with antibod3, against microtubule-associatedprotein 2
(MAP2) (HM-2). Ajier the blots (lanes I , 2) were detected
on an XAR film, the membrane was stained with Coovnasie
Blue (hnes 3, 4). Lanes I and 3, control; lanes 2 and 4, cortical dysplasia with large neurons. Note that the MAP2c
band (-70 kd) is more intensely stained in lane 2 than in
lane 1, whereus the intensity of the high-molecular-weight
MAP2 band (-280 kd) in the two lanes is almost the same.
Molecular-mass markers corresponding to 205, 121, 86 and
50.7 kd are shown at the leji.
The control specimen had an intensely stained band of
high-molecular-weight MAP2 (-280 kd) and a faintly
stained band of MAP2c (-70 kd). O n the blot of the
specimen with large neurons in dysplastic cortex, the
high-molecular-weight MAP2 band demonstrated similar intensity to that of the control, whereas a band of
MAP2c was much more intensely stained than that of
the control (Fig 2).
In this study, both iinmunoreactivity for MAP2 and
hybridization signal intensity of the MAP2 antisense
riboprobe were increased in the Iarge neurons in cortical dysplasia. A preliminary immunoblotting study in
one case showed that the MAP2c band was much more
intensely stained on the blot of dysplastic cortex with
large neurons compared with that of the unaffected
normal cortex in the same patient. Our results suggest
that MAP2 expression is increased in the large neurons
60 Annals of Neurology Vol 39 No 1 January 1996
in cortical dysplasia, and we speculate that MAP2c may
be a major component in this increased expression of
MAP2 in large neurons.
Several studies [4-71 have suggested that MAP2
plays a crucial role in the sprouting, branching, arid
regeneration of neuronal processes, as well as in their
morphological maintenance. Ferreira and associates 141
have reported that selective MAP2 induction in neuroblastoma cell lines by ganglioside produced numerous,
long and highly branched neurites. Dinsmore and Solomon [5] have shown that generation of neuritic processes associated with MAP2 expression in embryonal
carcinoma cells was induced by retinoic acid, and that
transfection of these cells with antisense MAP2 sequence before retinoic acid treatment reduced the
growth of neuritic processes. Their results suggested
that the expression of MAP2 may be necessary for the
generation of neuritic processes. Chamak and collaborators [6] investigated dendritic branching of rat striatal
and mesencephalic neurons cultured under a variety of
conditions. They concluded that markedly branched
neurons were strongly immunoreactive for MAI’2,
whereas neuronal cells with few branches were only
weakly immunopositive for MAP2. Ciceres et al. [7]
found pathological evidence that suggested that MAP2
contributes to dendritic plasticity. They observed significant changes in the levels and distribution of MAP2
in the dendrites of granule cells in lesions of the dentate gyrus after unilateral destruction of the entorhinal
cortex. Regenerated dendrites showed increased immunoreactivity for MAP2, whereas a loss of MAP2 immunoreactivity was seen in the collapsed dendrites. Based
on all these observations and our results, we speculate
that an increased expression o f MAP2 may indicate the
presence of activated plasticity in large neurons in cortical dysplasia. Regeneration or remodeling of the neur o d processes may persistently occur in large neurons
in the setting of epilepsy.
Among the various isofornis of MAP2, MAP2c plays
an especially significant role in the generation and regeneration of neuronal processes. Matus’ group studied
the function of MAP2c by transfecting MAP2c cDNA
in nonneuronal cell lines [S, 91. Their study indicated
that MAP2c stabilizes microtubules without providing
an assembly initiation site and as a result produces relatively few long microtubule bundles [S]. In addition,
treatment with the actin-depolymerizing drug cytochalasin B in MAP2c-transfected cells led to the outgrowth
of microtubule-containing processes from the cell surface. This did not occur in transfected cells treated with
the microtubule-stabilizing agent taxol. These results
suggested that MAP2c confers properties on cellular
microtubules that are essential for outgrowth of neuronal processes, i.e., stability, bundling, and stiffness
Although MAP2c is abundant and widely distrib-
uted in the immature brain, it is present only in trace
levels in the adult brain and is distributed in limited
regions, including olfactory bulb [ 151 and retina [ 161.
Since z o n a l growth and dendritic reinnervation continue in the adult olfactory system, persistent expression of MAP2c in the adult olfactory bulb is thought
to reflect the role of MAP2c in neurite outgrowth and
plasticity [ 151. MAP2c expression is also concentrated
within the inner segments and cell bodies of photosensitive cells of adult retina 1161. Since photosensitive
cells are unique among retinal neurons in the constant
regeneration of their primary processes, MAP2c is suggested to be involved in the regeneration of adult neuronal processes [ 161. Based on these observations,
MAP2c may play an important role in regeneration or
remodeling of neuronal processes. These functional
roles of MAP2c may also be operative in pathological
conditions such as cortical dysplasia associated with epilepsy. The results of preliminary immunoblotting in
this study show that MAP2c may be a major component contributing to this increased expression of MAP2
in the large neurons, which strongly supports the hypothesis that regeneration and remodeling persistently
continues in the neuronal processes of the large neurons in cortical dysplasia.
This study was financially supported in part by a grant from Japan
Foundation for Aging and Health and by the Research Institute,
The Hospital for Sick Children, Toronto.
This study was prepared with the assistance of Editorial Services,
The Hospital for Sick Children, Toronto, Ontario, Canada.
1. Jay V, Becker LE. Surgical pathology of epilepsy: a ieview.
Pediatr Pathol 1994;14:731-750
2. Duong T, De Rosa MJ, Poukens V, et al. Neuronal cytoskeletal abnormalities in human cerebral cortical dysplasia. Acta
Neuropathol 1994;87:493-503
3. Tucker RI’. The role of microtubule-associated proteins in
brain morphogenesis: a review. Brain Res Rev 1990; 15: 101
Ferreira A, Busciglio J, Landa C, Ciceres A. Gangliosideenhanced neurite growth: evidence for a selective induction
of high-molecular-weight MAP-2. J Neurosci 1990; 10:293302
Dinsmore J H , Solomon F. Inhibition of MAP2 expression affects both morphological and cell division phenotypes of neuronal differentiation. Cell 1991;64:817-826
Chamak B, Fellous A, Glowinski J, Prochiantz A. MAP-2 expression and neuritic outgrowth and branching are coregulared
through region-specific neuro-astroglial interactions. J Neurosci
1987;7:3 163-3 170.
Ciceres A, Busciglio J, Ferreira A, Steward 0. An immunocytochemical and biochemical study of the microtubule-associated protein MAP-2 during postlesion dendriric remodeling in
the central nervous system of adult rats. Mol Brain Res 1988;
8. Weisshaar B, Doll T, Matus A. Reorganizntion of the microrw
bular cyroskeleton by embiyonic microtubule-associated protein 2 (MAP2c). Development 1992;116:1151-1161
9. Edson K, Weisshaar B, Matus A. Actin depolymerisation induces process formation on MAP2-transfected non-neuronal
cells. Developmenr 1993;117:689-700
0. Chomczynski P, Sacchi N . Single-step method of RNA isolation by acid guaiiidium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156-159
1. Albala JS, Kalcheva N , Shafit-Zagardo €3. Characterization of
the transcripts encoding two isoforms of human niicrotubuleassociated protein-2 (MAP-2). Gene 1993;136:377-378
2. Saiki RK, Gelfand D H , Stoffel S, et al. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1988;239:487-491
3. Simmons D M , Arriza JL, Swanson LW. A complete protocol
for in situ hybridization of messenger RNA in brain and other
tissues with radioactive single-stranded RNA probes. J Histotechno1 1989;12: 169- 18 1
14. Arias C, Sharma N , Davies P, Shafit-Zagardo B. Okadaic acid
induces early changes in microtubule-associated protein 2 and T
phosphorylation prior to neurodegeneration in cultured cortical
neurons. J Neurochem 1993;61:673-682
15. Viereck C, Tucker RP, Matus A. The adult rat olfactory system
expresses microtubule-associated proteins found in the developing brain. J Neurosci 1989;9:3547-3557
16. Tucker RP, Matus A. Microtubule-associated proteins characteristic of embryonic brain are found in the adult mammalian
retina. Dev Biol 1988;130:423-434
Yamanouchi et al: MAP2 in Cortical Dysplasia
Без категории
Размер файла
541 Кб
expressions, microtubule, associates, large, dysplasia, cortical, protein, enhance, neurons
Пожаловаться на содержимое документа