Experimental approaches to study cerebral amyloidosis in a transgenic mouse model of Alzheimer's disease

Misfolding, aggregation and the accumulation of proteins in the brain are common
characteristics of diverse age-related neurodegenerative diseases. Each of these
neurodegenerative diseases is associated with abnormalities in the folding of a different
protein leading to protein aggregation and finally to neuronal death. Alzheimer’s
disease (AD) is one of these protein conformational diseases characterized by two major
neuropathological features: extracellular accumulation of amyloid-b (Ab) peptide in the
form of plaques and intracellular tangles consisting of hyperphosphorylated tau protein.
Although the majority of AD cases are sporadic, three genes have been described whose
mutations cause early-onset familial AD (FAD). The identification of mutations in these
genes has provided new opportunities to explore pathogenic mechanisms using
transgenic approaches. Based on the finding that mutations in these genes all lead to
elevated levels of Ab, new anti-amyloid therapies have been developed to either lower
the production of Ab or to clear the amyloid peptides. In the past few years, several groups have generated transgenic mouse models of
cerebral amyloidosis that exhibit age related Ab deposition similar to AD patients
through expression of mutated human amyloid precursor protein (APP). The studies
presented herein were done using such a transgenic mouse model, the APP23 mouse,
that overexpresses human APP with the Swedish mutation under the control of a neuron
specific Thy-1 promotor. APP23 transgenic mice develop cerebral amyloidosis in an
age- and region-dependent manner. Plaque formation starts early at 6 months of age and
is associated with the typical AD-like pathology including cerebral amyloid angiopathy,
neuron loss, glial activation and cognitive impairment. The purpose of this thesis was to study the mechanism and initiation of amyloid
formation as well as the spread of cerebral amyloidosis in vivo. The first series of
studies were conducted to define the role and contribution of extracellular versus
intracellular b-amyloid in plaque formation. To this end, we transplanted embryonic
wildtype (wt) and APP23 transgenic (tg) brain tissue into the hippocampus and cortex
of both APP23 and wt mice. We observed that APP23 grafts into wt hosts did not
develop amyloid deposits up to 20 moths post-grafting. In contrast, both tg and wt grafts into APP23 hosts developed amyloid plaques already 3 months post-grafting. The
amyloid deposits in wt grafts were surrounded by neuritic changes and gliosis similar to
the amyloid-associated pathology described in APP23 mice as well as in AD patients.
These results suggest that the phenotype of the transplanted tissue is strongly influenced
by the properties of the host. Moreover, these results provide evidence that diffusion of
Ab in the extracellular space is important for the spread of Ab pathology, that amyloid
formation starts extracellularly and that it is the extracellular amyloid that causes
neurodegeneration. The second set of experiments were performed to study the initiation of amyloid
deposition and to clarify which factors are involved in the seeding process in vivo. Since
seeded polymerization of Ab has already been demonstrated in vitro and in vivo, we
replicated and advanced these findings by intracerebral injection of diluted brain extract
from AD patients and brain extracts from aged APP23 transgenic mice into young predepositing
APP23 mice. AD and APP23 brain homogenate induced a similar amount of
seeded Ab deposits in the brain parenchyma and vessel walls four month post-infusion.
This seeding was time- and concentration-dependent. In contrast, no seeding was
observed when PBS was injected or when the same extract was injected into wt mice.
To address whether Ab itself is the seeding agent we injected synthetic Ab into young
APP23 mice. These synthetic Ab injections resulted in limited Ab deposition compared
to that obtained with Ab-rich brain extract. Our findings suggest that Ab-containing
human and mouse brain extracts can induce cerebral amyloidosis in vivo, and that Ab,
in combination with additional factors, initiates amyloid formation. The third part of the work presented here follows up on our previous finding that
diffusion of Ab in the extracellular space plays an important role in the spread of
cerebral amyloidosis. Therefore, we came up with the hypothesis that amyloid
deposition and the accompanied pathophysiology could influence extracellular space
(ECS) volume and interstitial fluid (ISF) diffusion properties. By using diffusion
weighted magnetic resonance imaging (DWI), we determined the diffusion properties in
the brains of young and aged APP23 transgenic mice and control littermates. Our results
indicate that fibrillar amyloid formation and the associated gliosis are accompanied by a
decrease in the apparent diffusion coefficient (ADC), suggesting that both build a barrier for interstitial fluid diffusion. Thus, in elderly people, ADC measurements and
the assessment of diffusion properties in the ECS could serve as a biomarker to detect
pathological events in the brain of AD patients.
In summary, the studies presented herein have increased our understanding of the
mechanisms leading to protein aggregation and finally to neurodegeneration in a
transgenic mouse model. We have shown that factors other than local Ab production,
such as diffusion in the extracellular space, are important in determining whether
amyloid pathology will occur. Moreover, the results highlight the relevance of
extracellular Ab to the pathogenesis of the disease. It still remains an open question
whether Ab itself is sufficient to initiate plaque formation, and if so, what
conformational form of Ab is required. Together, these studies provide insights into the
mechanisms and disease pathways which may lead to AD.