Accumulating evidence suggests at a link between amyloid beta (Aβ) accumulation and neurovascular dysfunction in early stages of Alzheimer’s disease (AD) (Iadecola, 2004). The ‘two-hit’ vascular model proposes that age and cardiovascular risk-factor related neurovascular deterioration, such as a reduction in neurovascular coupling and the breakdown of the blood brain barrier, may act synergistically with initial Aβ deposition creating a feedforward cycle of progressive change, ultimately promoting neuronal dysfunction and loss (Zlokovic, 2011). However, the factors that first establish such a vicious cycle have not yet been determined. We hypothesised that early Aβ-induced hyperactivity of excitatory neurons (Busche et al., 2012), combined with a reduced capacity of the neurovasculature to support this (Korte et al., 2020), emerge early in disease, which could subsequently create repeated transient hypoxic areas throughout the brain, promoting further Aβ deposition (Chen & Hassan, 2021).

To interrogate this we have used a novel mouse model combining humanised APOE and doxycycline-inducible APPSwe/Ind transgenes as well as the CaMKII driven GCaMP6f neuronal calcium reporter. In vivo 2Photon imaging and combined haemoglobin spectrometry/ laser doppler flowmetry allowed us to investigate neuron and vessel function over a time-course of Aβ accumulation. A surgically implanted cranial window placed over the visual cortex allowed us to record baseline measurements before APP expression is ‘switched on’ and at subsequent timepoints up to four months after.

A behavioural testing regime, on the same mice we imaged in vivo, indicates that four-months of Aβ accumulation represents an early AD model, with mild cognitive symptoms starting to appear. After four months of APP expression (and Aβ accumulation) mice performed around 35% worse than non-APP expressing controls during the acquisition trials of a simple, non-appetitive Barnes Maze paradigm (LMM interaction between APP expression and time, p = 0.005). Although these mice are still able to learn, this indicates a mild deficit in spatial working memory. Preliminary in vivo data from the visual cortex reveals an increase in neuronal calcium peak size (LMM interaction between APP expression and time, p = 0.04) and duration (LMM main effect of APP expression, p = 0.002) after four months of Aβ accumulation. Such an increase in neuronal activity should be met by a requisite increase in oxygen supply via neurovascular coupling, however oxygen saturation measurements from the same cortical region indicate a significant drop in tissue oxygenation from baseline in APP expressing mice (one-sided T-Test, p = 0.01).

Further data is being collected and analysed which will hopefully help inform our understanding of how neuronal and neurovascular changes induced by Aβ accumulation contribute to early AD pathological changes, preceding significant cognitive decline.