Regional mosaic genomic heterogeneity in the elderly and in Alzheimer's disease as a correlate of neuronal vulnerability

Alzheimer's disease (AD) is a neurodegenerative disorder characterized by fibrillary aggregates of Aβ peptide and tau protein. The distribution of these pathological hallmarks throughout the brain is not random; it follows a predictive pattern that is used for pathological staging. However, most etiopathogenetic concepts, irrespective of whether they focus on Aβ or tau pathology, leave a key question unanswered: what is the explanation for the different vulnerabilities of brain regions in AD? The pattern of regional progression of neurofibrillary degeneration in AD to some extent inversely recapitulates ontogenetic and phylogenetic brain development. Accordingly, degeneration preferentially affects brain areas that have recently been acquired or restructured during anthropoid evolution, which means that the involvement of a neurodevelopmental mechanism is highly likely. Since evolutionary expansion of the neocortex is based on a substantial extension of the mitotic activity of progenitor cells, we propose a conceptual link between neurogenesis in anthropoid primates and a higher risk of accumulating mitotic errors that give rise to genomic aberrations commonly referred to as DNA content variation (DCV). If increased rates of DCV make neurons more vulnerable to AD-related pathology, one might expect there to be a higher rate of DCV in areas that are affected very early during the course of AD, as compared to areas which are hardly affected or are affected only during the most advanced stages. Therefore, in the present study, we comparatively analyzed the DCV in five different cortical areas that are affected during the early stage (entorhinal cortex), the intermediate stage (temporal, frontal, and parietal association cortex), and the late stage (primary sensory occipital cortex) of AD in both normal elderly subjects and AD patients. On average, we observed about 10 % neuronal mosaic DCV in the normal elderly and a two- to threefold increase in DCV in AD patients. We were able to demonstrate, moreover, that the neuronal DCV in the cerebral cortex of the normal elderly as well as the increased neuronal DCV in AD patients are not randomly distributed but instead show systematic regional differences which correspond to differences in vulnerability. These findings provide additional evidence that mosaic genomic heterogeneity may play a key role in AD pathology.