The recent discovery of resident neural stem cells (NSCs) and their restorative potential has reversed earlier notions of the brain as a static and unchangeable structure. In particular the process of adult neurogenesis, i.e. the generation of new neurons from NSCs located in the adult hippocampal dentate gyrus has attracted considerable interest. This process is modifiable, conserved across mammalian species and plays an important role in cognition and behaviour, as measured experimentally e.g. in specific spatial memory and pattern separation paradigms. Although much less is known about its role in humans, neurogenesis appears to be important for cognitive function and its maintenance  during aging. Since decreased neurogenesis parallels advancing age, it may represent an increased vulnerability to neuropathology and possibly be impaired during Alzheimer’s disease.  Research described in this thesis focused on proliferation and neurogenesis in the adult and aging brain, both in postmortem studies on the middle aged primate and Alzheimer brain but also in experimental studies on mice.

We studied changes in structural plasticity and neurogenesis in relation to social stress exposure in a primate model, and in relation to amyloid pathology and dementia in the presenile Alzheimer brain. We further tested in experimental studies whether activity and pharmacological methods can stimulate neurogenesis and behavior in inbred mice (young and old) and in a well known mouse model for Alzheimer's disease. These studies were performed in order to evaluate the therapeutic capactity of neurogenesis to modulate age-related memory problems and Alzheimer neuropathology.

In Chapter 3 we studied the consequences of psychosocial stress on neurogenesis in 2 brain regions of the middle-aged common marmoset, a new world primate. We established that neurogenesis in the hippocampus of marmosets is not sensitive to psychosocial stress exposure when measured after a two-week recovery period. Surprisingly, large numbers of neuroblasts were found to be additionally present in the basal and lateral nuclei of the amygdala, at a dramatically higher density compared to the hippocampal dentate gyrus in animals this age. We further showed that these DCX-positive, immature neurons were migratory as shown by PSA-NCAM co-expression. Similar cells were also seen in the entorhinal cortex of this species. Similar to the hippocampus, stress failed to change the density of neuroblasts in the amygdala in this species. Our findings show that an extensive population of immature neurons is present in the amygdala and entorhinal cortex of the middle aged marmoset. Whether these cells play a role in structural plasticity and behavior, possibly even emotional memory, has yet to be established.

In Chapter 4, we studied proliferation of neuroinflammatory cells in the aged
human hippocampus (>70 years age) in relation to amyloid pathology and dementia. Our work followed up on our previous publication showing that proliferation was increased in the hippocampus of senile AD patients, an increase that appeared to be largely due to proliferation in glia-rich regions of this brain region. However, the phenotype of these proliferating cells was not established.

In our study, we co-labeled the proliferation marker PCNA with GFAP expressing astrocytes, and Iba1-expressing microglia. While astrocytes failed to co-express the proliferation marker PCNA, we could demonstrate that it were particularly the Iba1 expressing microglia cells that proliferate in the AD brain. This phenomenon was observed across disease conditions and in the presence of Aβ plaques, indicating that Aβ plaques may spur microglial proliferation. This suggests that microglial proliferation occurs during the early stages of disease, and could make a so far unknown contribution to neuroinflammation in general and to the subsequent progression of cognitive
decline and dementia in particular.

Regarding pharmacology-dependent neurogenesis, previous work had established that antidepressants from the selective serotonin reuptake (SSRI) class increased neurogenesis and neuronal maturation, leading to higher neuronal cell survival in young inbred mice. In chapter 5, we first evaluated the ability of duloxetine hydrochloride, a dual-pharmacology SSRI/SNRI compound to induce neurogenesis. Duloxetine was compared with fluoxetine, a classic and commonly used SSRI, and with voluntary wheel running, to measure stimulation of neurogenesis in young female C57Bl6J mice.

Our findings indicate that neither drug was able to improve the survival of new neurons in the dentate, although increased neuronal differentiation was observed for fluoxetine. Behaviorally however, the animals treated with fluoxetine and duloxetine experienced higher levels of anxiety compared to control animals. Compared to the minor effect of fluoxetine on differentiation, wheel running produced a profound increase in neurogenesis that would be the basis for our follow up experiment in middle-aged mice.

In Chapter 6 we tested the ability of prolonged physical activity (> 6M) to stimulate neurogenesis and hippocampal function in female C57Bl6J mice from middle-age onwards. In addition, the were tested for the ability of this treatment to preserve spatial memory as tested in the Morris Water Maze.

Our result show that wheel running in middle aged mice was able to preserve spatial memory performance. The mice further showed elevated neurogenesis and increased levels of BDNF protein, a neurotropic factor with known neuroprotective properties. Mice allowed to exercise through long-term wheel running furher performed better in a spatial memory test. This indicates that prolonged wheel running reversed age-dependent deficits in spatial memory and in neurogenesis. They also show that BDNF likely plays an important role in mediating the effects of wheel running. These data sets reflect that prolonged exercise in mid-life has beneficial effects on both hippocampal structure and function.

In chapter 8 we report on a similar experimental paradigm as described in chapters 5 and 6 but then utilized it for the 3xTg Alzheimer mouse model that harbors 3 mutant transgenes (APP, tau and PS1; 3xTg) and recapitulate both Aβ and tau pathology, the main neuropathological hallmarks of AD. We explored the synergistic treatment of the 3xTg mice with fluoxetine and voluntary wheel running, with the expectation that combined intervention would have an additive effect in reversing deficits in neurogenesis and hippocampal function. Wheel running was shown to significantly increase neurogenesis in 3xTg mice but did not significantly elevate BDNF as was found before in the aging C57Bl6 mice. Surprisingly, the 3xTg mice never exhibited deficits in spatial learning and memory, disassociating the relationship between neurogenesis and spatial learning and memory performance. This finding is carefully reviewed, with special attention for the genetic background of the 3xTg mice, the phenotype of these animals, and the behavioral tests employed.

Our findings presented here provide insight into the complex regulation of neurogenesis and the difficulties encountered when attempting to stimulate stem cell proliferation in the brain for modifying age- and disease-associated behaviors. Our work further highlights the important role for glial cells during health and disease. It is hoped that improved therapies for those afflicted by neurodegenerative diseases will be found on the path to discovery.