Imaging basal forebrain dysfunction in Alzheimer’s disease

Abstract

The basal forebrain, a grey matter region located in the medial and ventral aspects of the mammalian brain, consists of highly heterogeneous populations of cells including, but not limited to: cholinergic, GABAergic, glutamatergic and calbindin-positive neurons. The function of the basal forebrain has been linked to mediating reward-related behaviours, attention and, more recently, learning and memory in humans. In healthy aging, the basal forebrain undergoes slow, progressive degeneration over time. However in conditions such as Down-syndrome, Alzheimer’s disease and Parkinson’s disease this degenerative process is highly increased; why the basal forebrain is particularly vulnerable to neurodegeneration is unclear. Here we show that degeneration of the basal forebrain in Alzheimer’s disease occurs very early, even in patients diagnosed with a prodromal stage of the disease, called amnestic mild cognitive impairment. By imaging and measuring the basal forebrain volume noninvasively through magnetic resonance imaging (MRI) in humans, the state of an individual’s basal forebrain can be assessed and might be predictive of future disease progression. We found that subjects with a basal forebrain volume below a certain cut-off value were at a 7 times higher risk of having a worse diagnosis within ~18 months. In addition, basal forebrain degeneration correlates with the two major hallmarks of Alzheimer’s disease pathology: amyloid beta and tau protein burden in the brain in living subjects. During the development of Alzheimer’s disease, degeneration of the basal forebrain follows a distinct posterior-to-anterior pattern, and the volume of posterior basal forebrain nuclei is correlated to amyloid beta level in asymptomatic subjects, whereas the volume of anterior basal forebrain nuclei is correlated to amyloid beta level in subjects suffering from amnestic mild cognitive impairment and Alzheimer’s disease. Furthermore, a strong correlation exists between basal forebrain atrophy, particularly of the posterior volume, and tau burden in brain regions that are typically affected by tau pathology and which are also anatomically connected to the posterior basal forebrain. In addition, we found that a weaker correlation exists between basal forebrain atrophy and amyloid burden, as compared to the correlation between basal forebrain atrophy and tau burden. Also, tau and basal forebrain atrophy were associated with cognitive measures in Alzheimer’s disease subjects and controls, whereas amyloid burden was not. Our findings therefore support the idea that Aβ accumulation is a necessary but insufficient cause for progression to AD, with secondary, downstream or parallel events being required to trigger tau aggregation and BF degeneration, which then directly underpin cognitive decline. Based on these findings we suggest that an evaluation of basal forebrain atrophy should be included into routine MRI diagnostic procedures aimed at assessing the likelihood of development of dementia in elderly subjects. We further show, through selective ablation of a single population of cells (cholinergic neurons) of the basal forebrain in mice, that such subtle cell death in this region – similar to very early basal forebrain degeneration - does not affect the basal forebrain volume itself, but that MRI methods can still detect alterations in the diffusion of water molecules. This technique of diffusion MRI might offer an even more sensitive diagnostic method for Alzheimer’s disease risk. In order to conduct studies investigating degeneration of basal forebrain neurons in mice reliably and in greater detail in the future (through a combination of histological and MRI analysis), we are in the process of developing an average volume of the cholinergic nuclei of the basal forebrain from histological segmentation of three mice. The volumes are registered to MR images of the same mice and can therefore be used as a guide and reference by other researchers aiming to study basal forebrain pathology using mouse models recapitulating diseases such as Alzheimer’s disease and Parkinson’s disease. The work of this thesis will help to better understand the development of Alzheimer’s disease, further establish the role of basal forebrain dysfunction as a hallmark of Alzheimer’s disease pathology, and provide another important step towards an early diagnosis of this condition.