The progression of β-amyloid deposition in the frontal cortex of the aged canine

Abstract

Brains from 41 aged canines (≥10 years of age) were examined immunohistochemically to characterize the laminar distribution and age-related progression of β-amyloid (Aβ) in frontal cortex. We classified the Aβ patterns into four distinct types. Type I was characterized by small, faint deposits of Aβ in deep cortical layers. Type II consisted of diffuse deposits of Aβ mainly in layers V and VI. Type III had both dense plaques in superficial layers, and diffuse deposits in deep layers. Finally, Type IV had solely dense plaques throughout all layers of cortex. We compared the Aβ distribution pattern between the Old canines (10–15 years, n=22) and the Very Old canines (>15 years, n=19). The Old group primarily had negative staining, or Type I and Type II patterns of amyloid deposition (73%). Conversely, the Very Old group had predominantly Types II, III and IV deposits (89.5%), a difference that was significant (P<0.05). We suggest that Aβ deposition in canine frontal cortex is a progressive age-related process beginning with diffuse deposits in the deep cortical layers followed by the development of deposits in outer layers. In support of this hypothesis, the deeper layer diffuse plaques in the Very Old group of dogs also contain the largest proportion of β-amyloid with an isomerized aspartic acid residue at position 7, indicating that these deposits had been present for some time. We also observed fiber-like Aβ immunoreactivity within regions of diffuse Aβ deposits. These fibers appeared to be degenerating neurites, which were negative for hyperphosphorylated tau. Therefore, these fibers may represent a very early form of neuritic change that precede tau hyperphosphorylation or develop by an alternative pathway.

Introduction

The classical neuropathological hallmarks of Alzheimer's disease (AD) are the progressive accumulation of extracellular β-amyloid (Aβ) within the parenchyma that form senile plaques, and intraneuronal cytoskeletal changes, which result in neurofibrillary tangles (NFTs) [3]. It has become clear that different plaque subtypes are present within the brains of aged non-demented individuals and AD patients. These plaque subtypes appear to progress through specific, identifiable stages beginning with diffuse, non-β-pleated structures followed by primitive plaques and then neuritic plaques, which are thioflavine- and Congo-red-positive [7]. Characterizing the development of these lesions and their relation to NFTs is critical to understanding the pathogenesis of AD. However, the study of these processes is exceptionally difficult using human tissue because they occur over the course of years.

Previously, we suggested that the aged canine is a model system particularly well-suited to the investigation of the initial stages of plaque formation 6, 8. The aged canine brain contains predominantly diffuse plaques. Furthermore, the incidence of plaque formation in the aged canine population is relatively high, without showing the classical feature of neurofibrillary tangle formation. Thus, this model may be useful for delineating the mechanisms involved in the initial stages of Aβ deposition and for studying the processes that promote plaque development into degenerative loci in the absence of tangles [8]. Such an animal model is necessary, since it is difficult to study early AD pathology in the human brain due to the reserve capacity of the brain to absorb damage [25], and the rarity of early-stage autopsy tissue.

In addition to early neuropathological changes being present in many aged canine brains, it has been reported that canines experience age-related cognitive dysfunction. Aged canines are impaired on a variety of tasks, including delayed non-matching-to-sample recognition learning and spatial learning 15, 22. Further, it has been reported that various kinds of cognitive dysfunction in the aged canine correlate with the extent of Aβ deposition in hippocampus and frontal cortex [6]. Accordingly, study of the aged canine brain may also aid in our understanding of the morphological and cognitive changes that occur in aging.

It is generally assumed that NFTs and Aβ deposits in the human brain are not distributed randomly, but rather they have a characteristic regional and laminar pattern [3]. The research literature regarding the laminar distribution of senile plaques in AD brain, however, is less than definitive 20, 24, 27. For example, Braak et al. reported that neuritic plaques are predominantly found in layers II and III of occipital isocortex [4], while Lewis and co-workers reported that neuritic plaques are most numerous in layers III and IV of the visual and auditory cortices [20]. In fact, several investigators suggest that plaques are more common in the superficial cortical layers compared with deeper layers 10, 11, 26, while others suggest just the opposite 30, 31. The apparent variability in the laminar specificity of plaque distribution in AD may be due in part to the variability in the disease stage of the cases examined in each study. As a result, characterizing the laminar distribution of Aβ deposits with advancing age in the canine cortex could help to clarify the distribution and progression of Aβ in AD brain.

Therefore, in the present study we sought to characterize the distribution and progression of pathological changes in AD by use of an animal model of senile plaque formation. Specifically, we sought to clarify the laminar distribution of Aβ within the frontal cortex of aged canines and to delineate the progression of Aβ deposition over time using Aβ immunocytochemistry and a large sample of aged canines. Because one consequence of amyloid deposition over a long period of time is the spontaneous isomerization of aspartic acid residues present at positions 1 and 7 of the β-amyloid peptide [28], we also used an affinity purified antibody specific for the isomerized form of β-amyloid to further investigate a progression hypothesis. We chose frontal cortex because it has been suggested that the distribution of Aβ is more consistent in frontal cortex than in the hippocampus, based on analyses of human brain [25].