1. The effects of SDZ ENA 713, a novel acetyl cholinesterase inhibitor, on rat learning was studied using a step-down avoidance paradigm. 2. Injection of ibotenic acid into the caudolateral part of the basal forebrain (BF) innervating cholinergic neurons to the cerebral cortex, resulted in an increase in the number of trials required to obtain 300-second-latency, and also a decrease in the latency period after attaining 300-second-latency. 3. It is shown that the BF-lesioned rats are impaired in both acquisition and retention of learning. 4. Intraperitoneal injection of 0.10-0.05 mg/kg/day SDZ ENA 713 to the BF-lesioned rats showed amelioration of the learning impairment, with a decreased number of trials required to obtain 300-second-latency as well as an increase in the latency time after repeated training. 5. These results indicate that SDZ ENA 713 improves acquisition and retention impairment in BF-lesioned rats, and that this drug may be useful for demented patients with cholinergic dysfunction, such as Alzheimer's disease.

Clathrin, which constitutes coated vesicles and plays important roles in neuronal functions, has been reported to be involved in the pathology of Alzheimer's disease. In the brains of the patients with Pick's disease, distribution of clathrin was immunohistochemically investigated using monoclonal antibodies binding to different epitopes of clathrin light chain a and b. All the antibodies intensely labeled Pick's body and some perikarya of neurons, indicating impairment of slow axonal transport b (SCb). Antibodies against neurofilament, kinesin and synaptophysin also labeled Pick's body. These observations suggested impairment of axonal transport in the brains with Pick's disease, and might contribute to elucidating the pathology of Pick's body forming. It is implied that common pathological processes might lie in Alzheimer's disease and Pick's disease.

In aged rat brain, amyloid beta-protein precursor (APP) is accumulated in dendrites and cell bodies of Purkinje cells as full-length or truncated APP, because dendrites and cell bodies are positively stained by antibodies against both the amino- and carboxy-termini of APP. Western blot analysis of homogenates of brains of aged and young rats showed no apparent differences except for an increase in amino-terminal fragments in cerebrum and cerebellum of aged rat. These results indicate that the expression, transport or metabolism of APP in specific regions of brains may be affected by the aging process.

Clathrin, which constitutes coated vesicles, plays important roles in neuronal functions. In the brains of the patients with Alzheimer's disease, distribution of clathrin was immunohistochemically investigated using four monoclonal antibodies against clathrin light chains, LCB.1, LCB.2, X-16 and CON.1, to study the involvement of clathrin in the pathology of Alzheimer's disease. LCB.1, LCB.2, X-16, and CON.1 bind to the aminoterminus of the clathrin light chain b(LCb), to the neuron-specific insert of LCb, to the light chain a(LCa), and to LCa and LCb, respectively. In Alzheimer brains, granular staining of LCB.2 around neurons in the hippocampus was weaker or patchily defected in comparison with control brains. Some neurofibrillary tangles and neurons were intensely stained in Alzheimer brains by LCB.2, whereas neurons were weakly stained in control brains. Crowns of some senile plaques in the brains of early onset Alzheimer's disease were positively stained by LCB.2. LCB.1 supported the observations of LCB.2. Reactive astrocytes in Alzheimer brains were intensely stained by X-16. On the other hand, Western blot analysis using LCB.2 and X-16 demonstrated no apparent differences in protein amounts and molecular weights of LCa and LCb between control and Alzheimer brains. These observations demonstrated abnormal distribution of clathrin in Alzheimer brains, implying impairment of axonal transport in this disease.

Change in calcium response was studied to clarify the pathological process of alzheimer's disease (AD). Cultured fibroblasts from patients with familial Alzheimer's disease (FAD; n = 6), sporadic Alzheimer's disease (SAD); n = 4), and age-matched healthy control subjects (n = 4) were studied with an ACAS Interactive laser cytometer (ACAS-470). Fibroblasts from two independent families with FAD (OS-1, and OS-2 families) showed a suppressed calcium response after stimulation by 100 nM bradykinin (BK) 100 nM vasopressin (VP) or 10% FCS in CA(2+)-free condition compared with control fibroblasts at 48 h after plating. However, on the 7th day after plating, the abnormal calcium response was no longer observed. The height of the calcium peak showed periodic variation, indicating a relationship of calcium response with the cell cycle. When fibroblasts from OS-1 and OS-2 families were arrested in S phase, they showed a significantly suppressed calcium peak after BK stimulation. However, when those fibroblasts were arrested in other phases, they showed the same calcium peak as the other cells. The suppression of calcium response in S phase was indistinguishable from the calcium suppression induced by A23187 administration. Since Hardy type mutation on amyloid precursor protein gene is found in the OS-1 family, the observed abnormalities in calcium response might be related with pathological processing of amyloid precursor protein in AD. The reported abnormal calcium response, which is observed most obviously in fibroblasts in S phase, may indicate participation of the cell-cycle-dependent process in the pathology of AD.

Cyclosporin A, a calcineurin inhibitor, was administered into the rat hippocampus. Seven days after drug administration the effects of phosphatase activity suppression in the brain to changes in neuronal cytoskeletal proteins were studied. Rat brain homogenates injected with cyclosporin A showed 16-38% suppression of Ca/calmodulin-dependent phosphatase activity measured by 32P released from 32P-labeled histone, indicating that cyclosporin A acts as an inhibitor of calcineurin after binding with cyclophilin in the brain. Hematoxylin-eosin staining revealed many basophilic neurons in the pyramidal layer of hippocampus, cerebellum, thalamus and cerebral cortex. Immunohistochemical study with anti-phosphorylated neurofilament 200kDa antibody showed positive staining of neuronal perikarya of these basophilic neurons. The number of immunopositive neurons with anti-phosphorylated neurofilament 200KDa increased with the concentration of cyclosporin A injected into the brain, indicating a dose-dependent effect of the compound. Staining with anti-dephosphorylated neurofilament 200KDa (SMI-32) was decreased in neuronal perikarya of cyclosporin A injected brains compared with that of control brains injected with dimethyl sulfoxide alone, suggesting an increased phosphorylation of neurofilament 200KDa subunit protein. The perikarya of these basophilic neurons were not stained with anti-PHF (paired helical filaments) or anti-tau antibodies. Immunostaining with anti-ubiquitin was not increased in these cells. Immunostaining with anti-calcineurin A and anti-calcineurin B showed positive staining of neurons in hippocampus, cerebellum and caudate nucleus both in experimental and control brains. Calcineurin B immunoreactivity was more intense in pyramidal cells in hippocampus injected with cyclosporin A than in controls. Western blot study with antibody against calcineurin A revealed significantly more degradation products of calcineurin A in cyclosporin A injected brains. The present data suggest that cyclosporin A inhibits calcineurin activity in the brain, which results in increased phosphorylation of perikaryal neurofilaments. Since abnormal phosphorylation is speculated in the pathological process of Alzheimer's disease, the results will help elucidate the participation of phosphatase in the pathology of cytoskeletal proteins in Alzheimer's disease.

Functions of the tail region of neurofilament L have, to date, not been clearly elucidated. Bovine neurofilament L was cleaved into tail-less neurofilament L (50 kDa) and a tail fragment (19 kDa), by thrombin. Tail-less neurofilament L was deficit of the highly acidic domain of the tail region (approximately 77% of the entire region). Assembly of tail-less neurofilament L. was observed to be accelerated by both fluorometric and centrifugal measurements, compared with intact neurofilament L. The critical concentration of tail-less neurofilament L, which constitutes the constant unassembled pool, was approximately 0.25-times lower than that of neurofilament L. Under physiological conditions, tail-less neurofilament L formed a ribbon-like structure, whereas tail-less neurofilament L could form 10-nm filaments in an extremely low ionic-strength buffer in the presence of 1 mM MgCl2. An affinity-purified antibody directed against the tail fragment also accelerated neurofilament L assembly. The tail fragment neither coassembled with neurofilament L nor affect neurofilament L assembly. The acidic domain of the tail region may regulate neurofilament assembly and may be involved in 10-nm filament formation under physiological conditions.

Niigawa, H.; Tanimukai, S.; Takeda, M.; Hariguchi, S.; Nishimura, T. Author Information

Phosphorylation of glial fibrillary acidic protein (GFAP) induces disassembly of the filaments. An amino-terminal fragment of bovine GFAP (G-Hf) was produced by lysylendopeptidase digestion. G-Hf formed ribbon-like filaments in the presence of GFAP even in low ionic strength, whereas the fragment itself did not form any structures. Only one (PK3) of the five V8 protease fragments of G-Hf accelerated GFAP assembly to the same degree as G-Hf did, whereas the other fragments did not. When PK3 was cleaved into two fragments, it lost the assembly-accelerating property. The sequence of PK3 was determined as RRRVTSATRRSYVSSSE, which corresponded to residues 3-19 of porcine GFAP. It was concluded that PK3 contains a sequence indispensable for GFAP assembly and that neither PK1 (RRRVTS) nor PK2 (ATRRSYVSSSE) included all of the sequence. A single phosphorylation of PK3 by cyclic AMP-dependent protein kinase diminished its assembly-accelerating property. The phosphorylation site was determined as Ser-12 of porcine GFAP. It was shown that single phosphorylation of the amino-terminal head domain, which contains an indispensable sequence for GFAP assembly, might be sufficient for GFAP disassembly.

Ibotenic acid was injected into 3 parts of the lateral rat hippocampus. The animals were sacrificed 100 days after treatment, and studied immunohistochemically. The lesioned side of the hippocampus was highly atrophic with extensive neuronal loss and gliosis. Although silver staining revealed no particular structures such as neurofibrillary tangles or senile plaques, globular and granular depositions of amyloid beta-protein precursor (APP) immunoreactivity was observed by immunostaining in the lesion with the monoclonal antibody (clone 22C11). Increased immunoreactivities of glial fibrillary acidic protein (GFAP) and ubiquitin were found in the lesioned area, while the immunoreactivities of microtubule-associated protein 2 (MAP2) and 200 kDa neurofilament subunit protein (NF-H) were diminished. The results indicate that APP deposition is formed in the lesioned area where neuronal degeneration is produced by ibotenic acid in rat hippocampus.

Cytoskeletal proteins of the cultured fibroblasts obtained from Alzheimer's disease patients were studied. Western blotting studies of tubulin, actin, and vimentin showed no difference between Alzheimer and the control fibroblasts. Western blotting studies of vimentin revealed five partial degradation products in 50 K-57 K Da. molecular size region, but no difference in the degradation pattern was noticed between Alzheimer and the control fibroblasts. The size of fodrin molecule, however, was quite different between Alzheimer and the control fibroblasts. Comparing the molecular size of fodrin purified from the bovine brain, it is concluded that fodrin in Alzheimer fibroblasts is not degraded, while significant amount of fodrin in the control fibroblasts is partially degraded resulting in the smaller size of the 160 K and 200 K Da. molecular weight products.

Formalin-fixed paraffin-embedded hippocampal sections of brains with early-onset and late-onset Alzheimer's disease were studied immunohistochemically with antisera against cathepsin D and cathepsin B. In addition to the staining of neuronal perikarya, some of the senile plaques visualized by Bielshowsky silver staining and some of reactive astrocytes were positively stained with the antisera against cathepsin D and cathepsin B in brains with Alzheimer's disease. Abnormal localization of cathepsin D and cathepsin B immunoreactivity in neuronal perikarya was observed in brains with early-onset Alzheimer's disease. These findings demonstrate that the distribution of lysosomal proteases was altered in brains with Alzheimer's disease, suggesting the primary and/or secondary involvement of the lysosomal proteases in the pathological process of Alzheimer's disease.

The kinetics and dynamics of glial fibrillary acidic protein (GFAP) assembly was explored by a fluorescence energy transfer assay method. Purified GFAP was stoichiometrically labeled at a single cysteine residue with fluorescein-maleimide. Soluble labeled GFAP in a low ionic strength buffer was assembled into 10 nm filaments by rapidly increasing the ionic strength, and the kinetics of GFAP assembly was monitored by the reduction in fluorescence due to self-quenching of fluorescein. The extent of fluorescence quench correlated with both the formation of 10 nm filament morphology and the amount of protein pelleted at 12,000g. The assembly of GFAP is critically dependent upon both protein and magnesium ion concentration, and at the critical concentration for GFAP assembly is approximately 40 micrograms/ml. Disassembly of GFAP filaments was also observed as a relief of fluorescence quenching after dilution of labeled GFAP filaments. When labeled GFAP filaments were mixed with an excess of unlabeled filaments, a rapid increase of fluorescence was observed, which is due to an exchange of subunits between labeled and unlabeled GFAP filaments. These results indicate that GFAP filaments are dynamic structures and that a small pool of kinetically active unassembled GFAP subunits are in a dynamic equilibrium with assembled GFAP filaments. The ability of GFAP to assemble, disassemble, and undergo subunit exchange has important implications for the organization and dynamics of astroglia cell cytoskeleton during development and in response to injury.