Search

Transgenic mice carrying the Arctic (E693G) and Swedish (KM670/6701NL) amyloid-β precursor protein (AβPP) develop amyloid-beta (Aβ) deposits in the brain that resemble Alzheimer's disease neuropathology. Earlier studies of this model have documented morphologic features in selected parts of the cerebral cortex and hippocampus, but the spatial distribution within the brain and variance of Aβ deposits within a group of tg-ArcSwe mice is unknown. Using immunohistochemistry and brainwide microscopic analysis of 12-month-old tg-ArcSwe mice, we show that Aβx-40 plaque deposits are consistently present in the cerebral cortex, hippocampus, and thalamus and variably present in other regions. Using quantitative image analysis, we demonstrated that the average Aβ burden in the cortex and hippocampus is similar across animals, with coefficients of variance of 22% and 25%, respectively. This indicates that interventional studies of tg-ArcSwe mice are feasible using region-of-interest comparisons and that interventional trials require larger group sizes than commonly used. We also present an online atlas providing access to images showing the detailed characteristics and spatial distribution patterns of Aβx-40 labeling., (Copyright © 2014 Elsevier Inc. All rights reserved.)

Aquaporin-4 (AQP4) is the predominant water channel in brain and is selectively expressed in astrocytes. Astrocytic endfoot membranes exhibit tenfold higher densities of AQP4 than non-endfoot membranes, making AQP4 an excellent marker of astrocyte polarization. Loss of astrocyte polarization is known to compromise astrocytic function and to be associated with impaired water and K+ homeostasis. Here we investigate by a combination of light and electron microscopic immunocytochemistry whether amyloid deposition is associated with a loss of astrocyte polarization, using AQP4 as a marker. We used the tg-ArcSwe mouse model of Alzheimer's disease, as this model displays perivascular plaques as well as plaques confined to the neuropil. 3D reconstructions were done to establish the spatial relation between plaques and astrocytic endfeet, the latter known to contain the perivascular pool of AQP4. Changes in AQP4 expression emerge just after the appearance of the first plaques. Typically, there is a loss of AQP4 from endfoot membranes at sites of perivascular amyloid deposits, combined with an upregulation of AQP4 in the neuropil surrounding plaques. By electron microscopy it could be verified that the upregulation reflects an increased concentration of AQP4 in those delicate astrocytic processes that abound in synaptic regions. Thus, astrocytes exhibit a redistribution of AQP4 from endfoot membranes to non-endfoot membrane domains. The present data suggest that the development of amyloid deposits is associated with a loss of astrocyte polarization. The possible perturbation of water and K+ homeostasis could contribute to cognitive decline and seizure propensity in patients with Alzheimer's disease.

Little is known about how amyloid-beta (Abeta) is deposited in relation to the complex ultrastructure of the brain. Here we combined serial section immunoelectron microscopy with 3D reconstruction to elucidate the spatial relationship between Abeta deposits and ultrastructurally identified cellular compartments. The analysis was performed in a transgenic mouse model with mutant presenilin-1, and mutant amyloid-beta protein precursor (AbetaPP) and tau transgenes (3xTg-AD mice) and in aged dogs that develop Abeta plaques spontaneously. Reconstructions based on serial ultrathin sections of hippocampus (mice) or neocortex (dogs) that had been immunolabeled with Abeta (Abeta(1-42)) antibodies showed that the organization of extracellular Abeta deposits is more complex than anticipated from light microscopic analyses. In both species, deposits were tightly associated with plasma membranes of pyramidal cell bodies and major dendrites. The deposits typically consisted of thin sheets as well as slender tendrils that climbed along the large caliber dendritic stems of pyramidal neurons. No preferential association was observed between Abeta deposits and thin dendritic branches or spines, nor was there any evidence of preferential accumulation of Abeta around synaptic contacts or glial processes. Our data suggest that plaque formation is a precisely orchestrated process that involves specialized domains of dendrosomatic plasma membranes.

The two hallmark pathologies of Alzheimer's disease (AD) are amyloid plaques, composed of the small amyloid-beta (Abeta) peptide, and neurofibrillary tangles, comprised aggregates of the microtubule binding protein, tau. The molecular linkage between these two lesions, however, remains unknown. Based on human and mouse studies, it is clear that the development of Abeta pathology can trigger tau pathology, either directly or indirectly. However, it remains to be established if the interaction between Abeta and tau is bidirectional and whether the modulation of tau will influence Abeta pathology. To address this question, we used the 3xTg-AD mouse model, which is characterized by the age-dependent buildup of both plaques and tangles. Here we show that genetically augmenting tau levels and hyperphosphorylation in the 3xTg-AD mice has no effect on the onset and progression of Abeta pathology. These data suggest that the link between Abeta and tau is predominantly if not exclusively unidirectional, which is consistent with the Abeta cascade hypothesis and may explain why tauopathy-only disorders are devoid of any Abeta pathology.

Calcineurin is an abundant cytosolic protein that is implicated in the modulation of glutamate release. Here we show that the expression level of this enzyme is reduced in primary neuronal cultures treated with beta-amyloid. Parallel experiments in ETNA cell lines expressing SOD1 suggested that the effect of beta-amyloid on calcineurin expression is mediated by oxidative stress. The relevance of the in vitro experiments was assessed by analysis of tissue from patients with Alzheimer's disease (AD) and tissue from two strains of transgenic mice that mimic aspects of AD. The tissue from the AD brains displayed a pronounced downregulation of calcineurin immunoreactivity in profiles that were negative for glial fibrillary acidic protein (GFAP). In the hippocampus of the transgenic animals (which were analyzed in an early stage of the disease) the downregulation of calcineurin was restricted to mossy fiber terminals. A downregulation of the presynaptic pool of calcineurin may contribute to the dysregulation of glutamate release that is considered a hallmark of AD.

Alzheimer's disease (AD) is associated with the accumulation of extracellular deposits of the beta-amyloid protein (Abeta). Abeta is a result of misprocessing of the beta-amyloid precursor protein (APP). Gamma-secretase is involved in APP misprocessing and one hypothesis holds that this secretase is identical to PS1. We tested this hypothesis by determining whether PS is co-localised with Abeta in situ. Using confocal analyses and a sensitive immunogold procedure we show that PS and Abeta are co-localised within discrete microdomains of neuronal plasma membranes in AD patients and in aged dogs, an established model of human brain aging. Our data indicate that APP misprocessing occurs in discrete plasma membrane domains of neurons and provide evidence that PS1 is critically involved in Abeta formation.

In this review, we describe insights into beta-amyloid (Abeta) production using aged dogs as a model of human brain aging. The advantage of using dogs is that they naturally accumulate Abeta neuropathology with age. In parallel, dogs also develop age-associated learning and memory impairments. Thus, dogs can complement existing transgenic and nonhuman primate models typically used in aging studies. Dogs can live up to 18-19 years of age and companion dogs share the same environment as humans. Morphological brain changes as a function of age are clearly visible in vivo using magnetic image resonance scans. At the light microscopic level, dogs accumulate diffuse plaques with a distribution similar to that observed in human brain. Confocal studies suggest that Abeta accumulates on neuronal membranes in a segregated pattern. This pattern has been confirmed at the ultrastructural level using electron microscopy and provides insight into the deposition of Abeta into the extracellular space, possibly prior to overt plaque formation. Further, double immunogold labeling studies demonstrate that Abeta associated with the plasma membrane is colocalized with presenilin. These in vivo observations suggest a common site for both Abeta and presenilin supporting the hypothesis that the latter is involved with APP processing.

1. An electron microscopic study was undertaken to study beta-amyloid (Abeta) deposition and neuropathology in aged dogs. 2. A positive correlation between Abeta deposits and neuropathology was found in some dogs. Massive Abeta deposition was correlated to advanced lesions. 3. By use of immunocytochemistry Abeta fibers were identified within plaques, around vessels and in association with cell membranes.

beta-Amyloid (Abeta) is a constituent of senile plaques found with increasing age in individuals with Down syndrome (DS) and in the canine model of aging. Sections of DS and dog brain were immunostained using an affinity-purified polyclonal antibody for a posttranslationally modified Abeta with a racemized aspartate at position 7 (d7C16). The immunostaining characteristics of d7C16 Abeta in DS and dog brain indicate that it is present in all plaque subtypes, including the thioflavin-S-negative diffuse plaques that develop with age in dogs. The youngest DS case exhibited weak immunolabeling for d7C16 but the extent of d7C16-positive plaques increased with age. In addition, d7C16-positive plaques were initially found in clusters in the superficial layers of the frontal and entorhinal cortex but, with advancing age, increasing numbers appeared in deeper layers, suggesting a progression of Abeta deposition from superficial to deeper cortical layers. Ultrastructural studies in DS brain were confirmed using perfused dog brain and provided consistent results; thioflavin-S-negative diffuse plaques consist of fibrillar Abeta and racemized Abeta is associated with thicker and more highly interwoven fibrils than nonracemized Abeta. The use of antibodies to modified forms of the Abeta protein should provide insight into the progression of plaque pathology in DS and Alzheimer's disease brain., (Copyright 2000 Academic Press.)

The aged dog brain accumulates beta-amyloid in the form of diffuse senile plaques, which provides a potentially useful in vivo model system for studying the events surrounding the deposition of beta-amyloid. We used postembedding immunocytochemistry at the electron microscopic level to determine the subcellular distribution of beta-amyloid 1-40 and beta-amyloid 1-42 peptides in the prefrontal and parietal cortex of behaviorally characterized dogs ranging in age from one to 17 years. Immunogold particles signaling beta-amyloid 1-42 occurred over intracellular and extracellular fibrils that were approximately 8 nm in width. Intracellular beta-amyloid 1-42 fibrils were found in close proximity to glial fibrillary acidic protein fibers within astrocytes, but only in cells with signs of plasma membrane disruption. Neuronal labeling of beta-amyloid 1-42 appears to be associated with the plasma membrane. Membrane-bound beta-amyloid 1-42 occurs in the form of fine fibrils that are embedded in the dendritic membrane and appear to project into the extracellular space as determined by quantitative analysis of the immunogold particle distribution. Bundles of beta-amyloid 1-42 were also closely associated and/or integrated with degenerating myelin sheaths of axons. In one dog that was impaired on several cognitive tasks, extensive beta-amyloid 1-42 deposition was associated with microvacuolar changes and vascular pathology. The present findings suggest that beta-amyloid 1-42 may be generated at the dendritic plasma membrane as well as in intracellular compartments. The close association between beta-amyloid 1-42 and destroyed myelin suggests one possible new mechanism by which beta-amyloid 1-42 induces neurodegeneration.