Abstract Body

Cholinergic deficiency is a characteristic of many neurodegenerative disorders including Alzheimer’s disease (AD). Decreased levels of the vesicular acetylcholine transporter (VAChT) have been detected in AD patients and imaging data suggest that VAChT levels can predict human pathology. However, whether changes in VAChT are associated with plaque formation is unknown. To test for a causal relationship between VAChT levels and Ab plaques, we crossed a humanized APP knock-in (KI) mouse, carrying 2 AD-associated familial mutations, with mice lacking VAChT in forebrain neurons. We found that mice with elimination of forebrain VAChT presented with increased plaque area and Abeta levels. In contrast, in mice in which VAChT is overexpressed, we found a decrease in plaque area and Abeta levels in younger double mutant mice when compared to APP KI mice. In both these lines, potential changes in microglia-associated with plaques was observed. To further understand how cholinergic signaling in microglia may influence neuroinflammation, we generated mice expressing hM3Dq (a muscarinic M3 DREADD) exclusively in microglia. Our results suggest that chronic activation of muscarinic Gq signalling in microglia can trigger an inflammatory-like response that preconditions microglia to decrease their response to further immunological challenges. Our results suggest that VAChT levels are casually linked to plaque formation in humanized APP KI mice and that muscarinic signalling in microglia can trigger microglial immunological memory in vivo, which may be applicable for manipulation of neuroinflammation in neurodegenerative diseases.

Aims

Epidemiological studies indicate that obstructive sleep apnoea is a strong risk factor for the development of Alzheimer’s disease but the mechanisms of the risk remain unclear.

Methods

We developed a method of modelling obstructive sleep apnoea in mice that replicates key features of human obstructive sleep apnoea: altered breathing during sleep, sleep disruption, moderate intermittent hypoxemia and cognitive impairment.

Results

When we induced obstructive sleep apnoea in a familial Alzheimer’s disease model, the mice displayed exacerbation of cognitive impairment and pathological features of Alzheimer’s disease, including increased levels of amyloid-beta and inflammatory markers, as well as selective degeneration of cholinergic basal forebrain neurons. These pathological features were not induced by chronic hypoxia or sleep disruption alone. Our results also revealed that the neurodegeneration was mediated by the oxygen-sensitive p75 neurotrophin receptor and hypoxia inducible factor 1 alpha activity. Furthermore, restoring blood oxygen levels during sleep to prevent intermittent hypoxia prevented the pathological changes induced by the OSA.

Conclusions

These findings provide a signalling mechanism by which obstructive sleep apnoea induces cholinergic basal forebrain degeneration and could thereby increase the risk of developing Alzheimer’s disease, as well as providing a rationale for testing a range of possible prophylactic treatment options for people with obstructive sleep apnoea and hypoxia including increased compliance of continuous positive airway pressure therapy.

Abstract Body

Neurons differ in their vulnerability to Alzheimer’s disease (AD). The basal forebrain cholinergic neurons are among the first to degenerate in AD. We have shown with longitudinal structural magnetic resonance imaging (sMRI) that abnormal basal forebrain degeneration (1) is detectable in cognitively normal older adults harboring preclinical cerebrospinal fluid concentrations of amyloid-β and phosphorylated tau, and (2) precedes and predicts cortical degeneration and cognitive impairment. While our initial findings hold promise for a novel preclinical AD biomarker, our current sMRI-based tools for measuring cholinergic neuronal integrity do not meet the clinical standard of sensitivity and specificity. The [18F] fluoroethoxy-benzovesamicol (FEOBV) radiotracer positron emission tomography (PET) holds promise for overcoming this obstacle. [18F] FEOBV binds to the vesicular acetylcholine transporter (VAChT), a glycoprotein involved in the transport of acetylcholine to synaptic vesicles in presynaptic terminals of cholinergic neurons. To date, the utility of [18F] FEOBV as a preclinical biomarker of AD pathology remains untested.

To accelerate development of the [18F] FEOBV radiotracer for routine use in clinical trials, drug discovery and clinical assessment, we have developed a novel mouse neuroimaging platform which integrates in vivo sMRI with and [18F] FEOBV PET to quantify the integrity of the cholinergic neurons in preclinical disease models. This platform provides critical benchmarks for calibrating the sensitivity and specificity of 18F] FEOBV PET to early disease-related changes in VAChT, enables high-throughput screening of preclinical pharmacological interventions, and is fully translatable to humans.

Abstract Body

Aims

Widespread projections from the cholinergic basal forebrain to the cortex play an important role in memory and other cognitive processes that are impaired in Alzheimer’s disease, dementia with Lewy bodies, and vascular dementia. We recently proposed an in vivo model of the human cholinergic system connectivity based on diffusion tensor imaging (DTI) and showed associations with cognition in a normal aging population. In the current study, we investigated this promising model in individuals along the AD spectrum.

Methods

N=405 participants (53 AD, 66 MCI, 174 SCD, 112 healthy controls) from the DELCODE study were included. We modeled several cholinergic tracts in each diagnostic group using an enhanced diffusion neuroimaging pipeline. A multivariate model was employed to show the role of these tracts in cognition. Additionally, the role of cerebrovascular disease was integrated with both global and regional information.

Results

We found specific loss in the white matter cholinergic projections with the progression of the disease. We were able to locate spatial patterns of substantial disruptions along these tracts between groups. The multivariate models showed different degree of contribution of the considered factors to cognition in each diagnostic group.

Conclusions

Here, we show the utility of an in vivo model of cholinergic pathways in (pre-)AD populations using DTI. Such a model may help to localize pathological changes and to evaluate the degree of macroscopical damage and, thus, contribute to unraveling pathological mechanisms involved in dementia.

Aims

In Alzheimer’s disease (AD) the enzyme acetylcholinesterase (AChE) co-localizes with phosphorylated tau (P-tau) within the neurofibrillary tangles. Previously we have demonstrated that AChE is increased in a mutant tau mouse model. Here we have studied whether modulation of glycogen synthase kinase-3β (GSK3β) phosphorylated tau influences AChE expression.

Methods

P-tau levels were increased in SH-SY5Y cells by over-expression of human wild- type tau and GSK3β, then AChE enzymatic activity, protein and transcript levels were measured

In aliquots of cerebrospinal fluid (CSF) from AD patients enrolled in a clinical trial of tideglusib, an irreversible GSK3β inhibitor AChE enzymatic activity, P-tau, tau and Aβ42 levels were analyzed.

Results

AChE enzymatic activity and protein levels were increased (20 ± 2 % and 440 ± 150 % respectively) in in overexpressing P-tau SH-SY5Y cells, corresponding this increase to the cholinergic ACHE-T variant. In addition, an imbalance in cholinergic activity with a decrease in cellular levels of the neurotransmitter acetylcholine (45 ± 10 %) was demonstrated in overexpressing P-tau SH-SY5Y cells differentiated to neurons. A direct interaction between P-tau and AChE was also demonstrated. Regarding CSF from Tideglusib clinical trial, AChE activity at the end of the treatment were higher (35 ± 16 %) in placebo-treated patients, probably as result of the treatment with AChE inhibitors. However this increment was not observed in tideglusib treated patients. Furthermore P-tau prior treatment correlated with AChE activity.

Conclusions

P-tau interacts with AChE and can modulate its expression suggesting a possible increment on AChE at initial phases of AD.

Aims

Methods

Results

Conclusions

Aims

Reduced cholinergic neurotransmission resulting from basal forebrain degeneration causes cognitive impairment in Alzheimer’s disease (AD). The underlying causes of cholinergic degeneration are poorly understood. We assessed whether AD-related reductions in the bioavailability of the lipid phosphatidylcholine potentiates vulnerability of cholinergic basal forebrain neurons. Cholinergic neurons may place high demand on phosphatidylcholine lipid pathways because they are large, highly plastic, and use phosphatidylcholine to synthesize acetylcholine. An AD-related bottleneck on phosphatidylcholine may therefore impact cholinergic neuronal functions earlier and more severely than other cell types.

Methods

We leveraged data from the Alzheimer’s Disease Neuroimaging Initiative. Participants were stratified according to cerebrospinal fluid (CSF) ratios of amyloid and tau into age-matched normal (n = 62) and abnormal (n = 161) CSF groups. Longitudinal structural magnetic resonance imaging data were used to quantify grey matter degeneration for each participant. Partial least squares analyses assessed the multivariate relationship between serum lipids, including phosphatidylcholine, and neuroimaging data.

Results

Of all serum lipids, phosphatidylcholine (p = 0.002 on 5000 permutations) and acylcarnitine (p = 0.003 on 5000 permutations) were the only lipid types to exhibit a relationship with grey matter degeneration which was modified by CSF-confirmed AD pathology. The relationship between phosphatidylcholine and longitudinal grey matter degeneration revealed a spatial pattern (Figure, blue) with high anatomical specificity to brain regions known to be cholinergic, such as the basal forebrain and striatum.

Conclusions

We provide in vivo evidence for a selective relationship between phosphatidylcholine and basal forebrain degeneration, suggesting that phosphatidylcholine levels may be a risk factor for preclinical AD.