Neurodegeneration and neurogenesis in mouse models of aging and Alzheimer's disease

As the proportion of senior citizens gradually increases, the behavioral changes that occur with
normal aging and as a consequence of Alzheimer’s disease (AD) will afflict many of us in the
future. Aging is the major risk factor for AD, and pathological changes that occur in AD are
superimposed upon normal aging alterations. Thus, to understand etiologies and mechanisms of
AD it is important to distinguish normal aging from disease processes. In search of structural
parameters, which could correlate with the behavioral changes during normal aging and AD, the
discovery of neural progenitor cells and neurogenesis in the adult mammalian brain has received
much attention. Furthermore, advances in stem cell techniques have raised the possibility for
neuronal replacement strategies in neurodegenerative diseases such as AD. With progresses in
mouse genetics and the identification of genes linked to AD it has become possible to generate
transgenic mouse models that mimic key aspects of AD pathology. Studies involving such mouse
models have identified beta-amyloid peptide (Aβ), the main component of amyloid plaques, as
an important factor in the pathophysiology of AD. However, no general consensus exists about
the mechanism by which Aβ exerts its detrimental effects. The research described herein addresses
key questions regarding (i) neurogenesis and its modulation in the aging mouse brain, (ii) the
impact of cerebral amyloidosis on neurodegeneration and neurogenesis in a transgenic mouse
model of AD, and (iii) the application of a promising anti-Aβ immunotherapy in this transgenic
mouse model.
In a first study, we have examined the effect of aging on neurogenesis in the dentate gyrus of
C57BL/6 (B6) mice. We used the B6 line because it is one of the best characterized mouse
strains in neuroscience, and because it was shown to be relatively resistant to age-related structural
brain changes. Our results revealed a striking decrease in neurogenesis due to an age-related
reduction in neuronal proliferation. Interestingly, this decrease was observed until late adulthood
with no further decline with aging. Stimulated by recent findings that caloric restriction (CR)
might increase neurogenesis in young rodents, the potential of CR to postpone the age-related
decrease in neurogenesis was tested. However, results revealed no impact of CR on hippocampal
neurogenesis. Instead, a survival-promoting effect of CR on newborn glial cells in the hilar
region was observed. In a second study, the impact of cerebral amyloidosis on neurodegeneration was studied using a
recently generated murine model of AD, the APP23 mouse. This transgenic line overexpresses a
mutated human form of the amyloid precursor protein (APP), develops amyloid plaques, and
shows cognitive impairments with aging. Stereological estimation revealed a modest but
significant age-related neuron loss in the neocortex of APP23 mice. This observation is consistent
with the appearance of plaque-associated apoptotic and necrotic neurons in aged APP23 mice.
Encouraged by recent reports that demonstrated neocortical neurogenesis after targeted apoptosis,
we examined neurogenesis in the neocortex of APP23 mice with a high amyloid burden. However,
no evidence for neocortical neurogenesis, both in young and aged APP23 mice, was found.
In contrast, we found a fivefold increase in gliogenesis in aged transgenic mice when compared
to littermate controls.
During the last few years several therapeutic strategies have been proposed for treating AD, and
some of them have entered clinical trials. For example, it has been suggested that vaccination
with Aβ reduces cerebral amyloidosis and protects against cognitive deficits in different mouse
models of AD. Thus, in a third study, we investigated the effect of passive immunization in the
APP23 mouse, a model that exhibits amyloid plaques as well as cerebral amyloid angiopathy
(CAA), similar to that observed in the human AD brain. Our results showed significant clearance
of diffuse amyloid and reductions in the levels of the highly fibrillogenic Aβ42. However,
immunized mice exhibited a robust increase in the frequency and severity of CAA-associated
cerebral hemorrhages compared to non-vaccinated APP23 controls. Together with the
neuroinflammatory side effects recently observed in human trials, our results further stress the
need for a better understanding of the basic mechanisms involved in antibody-mediated
Aβ clearance.