Cerebral amyloidosis in a transgenic mouse model of Alzheimer's disease : impact and therapy

Senile dementia is a diagnostic category that includes all types of cognitive and memory
impairments that occur in the elderly. Alzheimer’s disease (AD) is the most common
form of dementia, which is characterized by progressive impairments in memory,
cognition, praxis, language, and behavior. Neuropathologically, major features of AD
include extracellular accumulation of amyloid β (Aβ) peptide in form of plaques,
intracellular tangles composed of hyperphosphorylated tau protein, selective
neurodegeneration, and synapse loss. More specifically, the cholinergic system is
compromised in AD. Cholinergic disruption in the neocortex and the hippocampus is
exemplified by diminished density of cholinergic terminals and fibers, reduction in
cholinergic receptors, and decreased choline acetyltransferase (ChAT) and
acetylcholinesterase (AChE) enzyme activities. In addition, loss of cholinergic neurons
in the basal forebrain, the major input to cortex and hippocampus, has been reported. The principal risk factors for AD include aging, mutations in the genes of amyloid
precursor protein (APP), presenilin (PS) 1 and PS2, and the presence of the ε4 allele of
the apolipoprotein E (apoE) gene. Mutations in APP, PS1 and PS2 genes are inherited
in an autosomal dominant fashion and are the cause of early-onset familial AD (FAD).
FAD accounts only for a minor percentage of all AD cases. It is, however, clinically and
pathologically indistinguishable from the much more common sporadic AD found in the
elderly. FAD mutations result in alteration of APP processing, leading to
overproduction of Aβ and thus formation of plaques. Based on the knowledge on the
genetic factors leading to AD, transgenic mice carrying FAD mutations were generated
and used to elucidate the role of Aβ in AD pathogenesis. The studies presented herein
make use of the well-established APP23 mouse model of cerebral amyloidosis. APP23
mice bear the FAD Swedish mutation on the APP gene, which results in increased
production of amyloidogenic Aβ and develop amyloid deposits progressively with age.
Moreover, linked to extensive amyloid deposition, these mice show additional AD-like
characteristics such as cerebral amyloid angiopathy, microglial activation, selective
neuron loss, and cognitive impairment. The purpose of the first part of the present research was to study the interaction of
amyloidosis and the cholinergic system. Therefore, we elucidated the extent of the
cholinergic changes in APP23 mice to deduce the contribution of amyloidosis to the
cholinergic deficit seen in AD. Stereological quantification of cortical cholinergic fibers
and measurement of ChAT and AChE enzyme activities implied a cholinergic
disruption in the neocortex. To establish whether this deficit was due to a loss of
cholinergic source neurons in the basal forebrain, ChAT-positive neurons in different
nuclei such as the medial septum and the nucleus basalis of Meynert (NBM) were
quantified with stereological methods. No decrease in cholinergic neuron number could
be found, suggesting that cortical cholinergic deficit is a local phenomenon and is solely
attributable to plaque formation. To further study the interaction between cholinergic
system and amyloid plaque formation and to test the hypothesis that cholinergic
depletion has an effect on amyloid plaque formation, the major cholinergic source to the
neocortex, the NBM, was experimentally lesioned. This lesion induced a cortical loss of
cholinergic fibers and enzyme activities, and subsequent plaque formation was
monitored. Results revealed a decreased plaque load in denervated areas. The outcome
of this experiment suggested that amyloid deposition and the cholinergic system are
somehow linked but the hypothesis that loss of cholinergic input does promote plaque
formation could not be confirmed. The aim of the second study was to clarify the impact of Aβ on neocortical synapses.
Although amyloid load has been shown to correlate with loss of synapses in AD, this
finding is complicated by the presence of neurofibrillary tangles and the loss of
subcortical input. Therefore, we followed neocortical synaptic changes throughout the
development of plaques. The presynaptic vesicular protein synaptophysin was analyzed
in different age groups of APP23 and wild-type mice, from 3 months of age with no
plaques, to 24 months with severe plaque load. Densitometric analysis of Western blots
did not reveal any differences between APP23 mice and wild-type controls. This finding
was further supported by stereological synapse counting, which showed that no
synapses are lost with aging in either genotype group. Moreover, APP23 mice did not
bear decreased synapse number compared with wild-types. Our results suggest that Aβ deposition is not sufficient to account for the synapse loss seen in AD. Alternatively, a
possible trophic effect of APP may prevent or delay a loss of synapses in our mouse
model. In the last part of the work presented here, we studied the effect of passive
immunization with antibodies against Aβ in APP23 mice. Vaccination holds great
potential in the fight against AD, and clinical trials with human patients have been
undertaken. However, we have shown that although very efficient in removing amyloid
deposits, immunization may bear the danger of leading to cerebral hemorrhages.
Although this finding might appear discouraging, it contributes to the understanding of
the mechanisms involved in plaque formation and thus eventually leads to new
therapeutical approaches. In summary, through the use of transgenic mice, the presented studies have brought
improved understanding of the pathogenesis of AD. We have shown that cerebral
amyloidosis is the cause for cortical cholinergic fiber and enzyme activity loss.
Interestingly, this fiber loss does not lead to retrograde degeneration of cholinergic basal
forebrain neuron cell bodies. Furthermore, overall synapse number was not changed in
the neocortex even with high plaque load. Together, these findings exemplify the
complexity of the impact of amyloid. The question remains open, whether the toxicity
of amyloid plaques has a direct effect or plays a more regulatory role in the complex
cascade leading to neurodegeneration. Clearance of amyloid by passive immunization
concomitantly induced microhemorrhages, further demonstrating the complex effects of
amyloid. Together, these studies are part of a better understanding of the
pathophysiological mechanisms leading to AD and are fundamental for new therapies
based on the causes of AD.