|Home | About | Journals | Submit | Contact Us | Français|
Fukuyama type congenital muscular dystrophy accompanies central nervous system and ocular lesions. Morphological findings suggest that major central nervous system lesions, such as cortical dysplasia, are caused by the abnormal glia limitans due to an impairment of astrocytes. Increase of corpora amylacea and neurofibrillary tangles suggests acceleration of the aging process in the Fukuyama type congenital muscular dystrophy brain. Glycosylation of α-dystroglycan is decreased in the central nervous system of Fukuyama type congenital muscular dystrophy in a similar manner to the skeletal muscle, but dystroglycan mRNA levels appear to be increased. Glycosylated α-dystroglycan is reduced in the glia limitans formed by astrocytic endfeet. Slight accumulation of Nε-(carboxymethyl)lysine, an oxidative modification product, is observed in astrocytes of Fukuyama type congenital muscular dystrophy and in an astrocytoma cell line with suppressed fukutin expression. Cerebral cortical neurons of Fukuyama type congenital muscular dystrophy and controls react with an antibody for core α-dystroglycan but not with an antibody for glycosylated α-dystroglycan. Carboxymethyl lysine is accumulated in cortical neurons of a severe case of Fukuyama type congenital muscular dystrophy. Both astrocytes and neurons appear to be sensitive to oxidative stress when fukutin is suppressed. However, it is still unclear how the loss of fukutin causes astrocytic and neuronal dysfunction. Since the central nervous system is composed of several components that are closely related to each other, more investigations are needed for thorough understanding of the Fukuyama type congenital muscular dystrophy brain. Moreover, since astrocytes and epithelial cells may show different cellular responses to fukutin suppression, it seems important to evaluate the functions of fukutin in each type of cell or tissue, not only to prove the pathogenesis of Fukuyama type congenital muscular dystrophy, but also for applying appropriate therapies, especially those at molecular level.
Fukuyama type congenital muscular dystrophy (FCMD) is an autosomal recessive disease, found exclusively in Japan (1, 2). The gene responsible is fukutin coding a protein of 461 amino acids (3). FCMD is classified into a group of congenital muscular dystrophies accompanying central nervous system (CNS) and ocular lesions. Muscle-eye-brain disease (MEB) and Walker-Warburg syndrome (WWS) are also included in this group (1). The CNS lesions are generally characterized by cobblestone lissencephaly, in other words, type II lissencephaly or polymicrogyria.
The skeletal muscle of FCMD patients shows decreased glycosylation of α-dystroglycan (α-DG) (4), one of the components of the dystrophin-glycoprotein complex linking intracellular and extracellular proteins (1, 5). Muscular dystrophies showing a decrease of glycosylated α-DG are called α-dystroglycanopathy, in which FCMD, MEB, WWS, congenital muscular dystrophy IC (MDC1C), limb girdle muscular dystrophy 2I (LGMD2I) and MDC1D are included (5). Besides fukutin, other genes responsible for α-dystroglycanopathy have been found, e.g., protein O-linked mannose β1,2-N-acetylglucosaminyltransferase (POMGnT1) in MEB, protein-O-mannosyltransferase 1 (POMT1) and POMT2 in WWS, fukutin-related protein (FKRP) in MDC1C and LGMD2I, and LARGE in MDC1D (5). POMGnT1 has an enzymatic activity for the glycosylation of α-DG (6, 7). Co-expression of POMT1 and POMT2 is required for enzymatic activity (8). Fukutin seems to relate to the glycosylation of α-DG, but its function has not been proven directly.
The decreased glycosylation of α-DG has been observed in the CNS of FCMD, as well as in the skeletal muscle, by western blotting and immunohistochemistry (9). In contrast, the expression of DG mRNA appears to be increased in the cerebrum of FCMD (Fig. (Fig.1).1). Since the CNS is composed of several components such as neurons, glial cells, capillaries and leptomeninges, it is necessary to study which component is involved in the formation of CNS lesions.
Both in fetal and post-natal FCMD cases, CNS lesions are predominantly observed in the surface areas, represented by cortical dysplasia of the cerebrum and cerebellum (9). Basic structures of the CNS look normal. Maturation of the brain in fetal cases appears to correspond to the gestational age. Distribution and severity of the dysplasia are different from case to case.
In post-natal cases, the cerebral cortex exhibits disorganization of neurons with focal fusion of the surface. Vague protrusions of cortical tissues called verrucous dysplasia can be seen in some areas. Heterotopic neurons are scattered in the white matter. In the cerebrum of fetal cases, the glia limitans formed by astrocytic endfeet is disrupted, and varying degrees of glioneuronal tissues overmigrate through the defects depending on the severity of the defects. The verrucous dysplasia in post-natal cases resembles an over-migrated lesion observed in fetal cases (9). The cerebellum and brainstem are also affected from fetal to adult cases. The cerebellar cortex is dysplastic, usually focal in the dorsal part of the hemisphere. In the brainstem, heterotopic glioneuronal tissues are observed. The pyramidal tract and brainstem nuclei are abnormal in some cases (10). The spinal cord generally exhibits a normal configuration, but focal disruptions of the glia limitans are found in severe cases. The glia limitans is formed by closely apposed astrocytic endfeet. Normally, the structure is detected as a linear contour of the astrocytic cell membrane and a linear double layer, lamina lucida and lamina densa, of the basement membrane, by electron microscopy. In FCMD cases, both the basement membrane and astrocytic cell membrane show abnormal configurations (9). Even in areas where disruptions are not detected by light microscopy, the three-layered structure of the cell membrane and basement membrane are discontinuously ambiguous, and minute defects less than several µm in size can be seen. Morphological findings suggest that major malformative lesions such as cortical dysplasia are caused by the abnormal glia limitans due to an impairment of astrocytes. This hypothesis is consistent with the observations in embryos of fukutin-deficient chimeric mice (11). The glia limitans is disrupted with the reduction of glycosylated α-DG, but neither neuronal migration nor extension of radial glial fibers is affected in these chimeric mice (11). However, an impairment of immature neurons could be speculated from some minor findings observed in FCMD cases, such as heterotopic neurons in the cerebral white matter, which may indicate neuronal migration arrest (9). Astrocytes and neurons can be involved in the CNS malformation of FCMD, but the magnitude of involvement is probably more in astrocytes (Fig. (Fig.2).2). In addition to malformative lesions, there are some other lesions such as corpora amylacea and neurofibrillay tangles in post-natal FCMD, especially in patients surviving for a long time. These structures can generally be seen in normal aged people, but are exceptional in children and young adults. The aging process seems to be accelerated in FCMD. To consider the genesis of these structures, both primary and secondary events should be borne in mind, since astrocytes and neurons are closely related to each other. Dysfunction of astrocytes might cause neuronal dysfunction, and vice versa. Loss of fukutin might be able to induce cellular dysfunction directly, or indirectly via reduced glycosylation of α-DG (Fig. (Fig.22).
The expression of fukutin has been proved in primary cultured rat astrocytes and an astrocytoma cell line by reverse transcriptase-polymerase chain reaction (RT-PCR). The expression is also seen immunohistochemically in normal human CNS tissues (12, 13). In immunohistochemistry using an antibody for glycosylated α-DG, immunoreaction is reduced in the cerebral glia limitans of FCMD (9), although the reduction is not uniform. In contrast, the positive reaction with an antibody for the core peptides of α-DG is preserved (Fig. (Fig.1).1). To investigate whether the loss of fukutin alters the glycosylation of α-DG in astrocytes, a knock down of fukutin by RNAi interference was performed in a human astrocytoma cell line (1321N1). Stealth RNAi duplex for fukutin designed by Invitrogen (Carlsbad, CA, USA) was transfected using lipofectamin2000, according to the manufacturer’s instructions (Invitrogen). In this cell line, it was difficult to prove the decrease of glycosylation by immunohistochemistry and western blotting, because the cells only contain a small amount of glycosylated α-DG. However, the cells lost the ability to attach to laminin-coated surfaces after fukutin-suppression without significant difference in DG mRNA expression (data not shown). Since the sugar chain of α-DG is a receptor of laminin (5), it is possible that the core α-DG is expressed but the glycosylation is reduced. At light microscopy, the cerebral glia limitans is disrupted in fetal FCMD cases, but continuous with severe superficial gliosis in post-natal cases. Astrocytes are markedly increased in number and also elongate their cytoplasmic processes in the area of superficial gliosis (12). This may be a compensation for the fragility of the glia limitans. Because the fragility continues after birth, the metabolism of astrocytes, especially those involved in the superficial gliosis, may be altered. Nε-(carboxymethyl)lysine (CML), an oxidative modification product, accumulated slightly in astrocytes of the cerebrum of FCMD (Fig. (Fig.1)1) (14). In immunohistochemisty using cell-blocks, a slight increase of CML was found in fukutin-suppressed astrocytoma cells (data not shown). Although this is a result from tumor cells in a short experimental period in vitro, it is not contradictory that astrocytes may be sensitive to oxidative stress when fukutin is suppressed.
In the control fetal CNS, fukutin is expressed in immature neurons of the cerebral cortex and germinal matrix (12, 15). Purkinje cells and external and internal granular layer cells of the cerebellum also express fukutin. The expression of fukutin in mature neurons is somewhat controversial. There are reports showing the expression in many mature neurons of the cerebral cortex and in Purkinje cells (15). On the other hand, in our experiments, clearly positive reactions for fukutin are observed in a few of these cells (12). The expression is retained in many internal granular layer cells of the adult cerebellum (12, 15). These contradictory findings might be derived from differences in experimental procedures including the probes and antibodies used. However, it appears that the expression of fukutin tends to be low after the maturation of neurons in humans, although it depends on the type of neuron. In immunohistochemistry using the antibody for glycosylated α-DG, cerebral cortical neurons and neuropils are negative both in FCMD and control cases from fetuses to adults (Fig. (Fig.1).1). With an antibody for core α-DG, immature neurons of the cerebral cortex and germinal matrix are stained positively, and no apparent difference can be found between fetal FCMD and control cases (Fig. (Fig.1).1). α-DG is considered to play a key role for proper proliferation and differentiation in immature neuroepithelial cells (16). Since both fukutin and α-DG are expressed in immature neurons, fukutin might work via α-DG for the proper development of immature neurons. Heterotopic neurons in the cerebral white matter of FCMD patients support this speculation. In post-natal FCMD and control cases, neuronal cytoplasm and neuropils give positive reactions with the antibody for core α-DG. More dendrites appear to be stained in FCMD cases. This result appears to be compatible with a post-synaptic role of α-DG (17), but it is unclear whether glycosylated α-DG is required for this function or not. Similarly, there is no clear evidence of how fukutin is involved in the function of mature cerebral cortical neurons at present. On the other hand, altered glycosylation of α-DG has been observed in hippocampal neurons of FCMD (18). With immuhohistochemistry to detect oxidative modification products, there was a slight accumulation of CML in the neurons of a severe 2-year-old case (Fig. (Fig.1).1). Although there was no significant CML accumulation in common and mild cases from 14-27 years, there were more positive reactions for Mn superoxide dismutase, an enzyme against oxidative stress existing in mitochondria, compared to controls. In common and mild cases, more active participation of anti-oxidants may prevent the accumulation of CML. Neurons also appear to be sensitive to oxidative stress, and the accumulation of CML may be greater when gene impairment is severe.
The characteristics of neurons and astrocytes have gradually been elucidated in the CNS of FCMD. However, there are still many unresolved aspects. Even in neurons, it has still not been proven that fukutin works toward neuronal migration or against it, or has other roles. Increased sensitivity to oxidative stress in astrocytes and neurons may be related to the increase of corpora amylacea and neurofibrillay tangles, but the mechanism is unknown. Moreover, the roles of other components of the brain are unclear. Recently, defective peripheral nerve myelination has been found in fukutin-deficient chimeric mice (19). Oligodendroglia express and use dystroglycan as a laminin receptor to regulate myelination (20). Dysmyelination of cerebral white matter has been suggested by diffuse white matter hyperlucency or myelin pallor in FCMD infants (2). This white matter abnormality of FCMD infants might be derived from an impairment of oligodendroglia, or might be a secondary event due to cortical dysplasia, or both (Fig. (Fig.2).2). The basement membrane is formed around capillaries as well as in the glia limitans of the CNS, but glycosylated α-DG appears to be maintained in capillaries of FCMD patients when examined by immunohistochemistry (Fig. (Fig.1).1). It is unknown why this difference occurs. The dystrophin-glycoprotein complex exists in vascular endothelial cells (21). The structure of DGC might be different between capillaries and the glia limitans. Because the blood-brain barrier is altered in mdx-mice (22), it might be functionally abnormal also in FCMD patients. More investigations are needed for thorough understanding of the CNS lesion, because each component is closely related to one another for the proper function of the CNS.
Another unsolved point is that there are no significant lesions in various somatic organs in FCMD, either clinically or pathologically (2), despite the presence of fukutin expression in these organs (15). Little has been learned concerning the roles of fukutin in these somatic organs. In our experiments using RNAi, cellular responses after fukutin-suppression were completely different between astrocytoma and cancer cell lines (data not shown). In fukutin-suppressed astrocytoma cells, positive nuclear reactions for Ki-67 and cyclin D1 were reduced. In contrast, Ki-67-positive cells were increased and there were more positive reactions for phosphorylated c-jun in fukutin-suppressed HeLa cells. Although the result is in tumor cells, similar tendencies may be presumed in normal astrocytes and epithelial cells. It seems important to evaluate the functions of fukutin in each type of cell or tissue, not only to prove the pathogenesis, but also for applying appropriate therapies, especially those at molecular level.