Vascular correlates in a mouse model of Alzheimer's disease

Abstract

Aims
During the progression of Alzheimer’s disease (AD), recognisable impairments in the vasculature are considered to be confined to the brain and occur solely as a consequence of the neuropathology associated with the disease. Here we use the microvascular corrosion casting technique with scanning electron microscopy (SEM) to recreate and examine the cerebral and systemic vascular ultrastructure of the APPswe/PS1dE9 (APP/PS1) murine model of AD. Our aims are to determine whether cerebral and systemic vascular ultrastructural impairments coincide with the neuropathology of 9-month old APP/PS1 mice and that liraglutide (Victoza®) treatment can restore vascular ultrastructure (chapter 3). In addition, we test whether cerebral and systemic vasculature ultrastructural impairments exist prior to cognitive dysfunction in 4-5 month old APP/PS1 mice (chapter 4) and that liraglutide treatment can slow onset of neuropathology in 6-7 month old APP/PS1 mice (chapter 5). The hypothesis tested is that cerebral and systemic vascular ultrastructural impairments precede the onset of cognitive dysfunction in APP/PS1 mice representing a therapeutic target to slow the development of disease progression in this AD model.

Methods
In chapter 3; seven-month old male wild-type and APP/PS1 mice (n=4) received once-daily intraperitoneal injections of either liraglutide [25 nmol/kg body weight (bw)] or saline (0.9% w/v NaCl) for eight weeks. Following treatment, each nine-month old animal received an intraperitoneal injection of heparin prior to non-recovery anaesthesia (ketamine [100 mg/kg bw]; xylazine [16 mg/kg bw]) and a sequential transcardiac perfusion of saline, 2.5% glutaraldehyde and microvascular casting solution. Following a 12-hour polymerisation, the casted organs were dissected free, macerated in 5-7.5% KOH, repeatedly washed with dd.H20, air-dried and sputter-coated in Pd/Au prior to evaluation by SEM. In chapter 4; the body weight of four-to-five month old wild-type and APP/PS1 mice (n=6) in addition to the recognition memory (n=11), brain weight (n=6), numbers of thioflavin-S positive plaques (n=6), fasting intraperitoneal glucose tolerance (n=11) and urinary albumin creatinine ratios (uACR; n=5) were determined as well as microvascular corrosion casting (n=6). In chapter 5; 4-5 month old wild-type and APP/PS1 transgenic mice received once-daily injections of either liraglutide [25 nmol/kg bw; n=6) or saline(0.9% w/v NaCl; n=5) for eight weeks prior to the assessment of body weight, food intake, recognition memory, fasting intraperitoneal glucose tolerance, serum albumin, brain weight and numbers of cerebral thioflavin-S positive plaques.

Results
The microvascular corrosion casting technique precisely replicated lumenal morphology, which permitted an accurate identification of (micro)vessel morphological features as well as highlighting pathological features such as vessel leakage. In chapter 3, we reveal the intricate networks of smoothly-contoured cerebral, hepatic and renal microvessels in wild-type mice treated with either liraglutide or saline. This was in contrast to the conspicuous extravasation of casting solution observed on the walls of the cerebral and renal microvasculature from saline-treated APP/PS1 transgenic mice, a morphological feature that was ameliorated following treatment with liraglutide. In chapter 4, we demonstrate that 4-5 month old wild-type mice are indistinguishable from their age-matched APP/PS1 littermates in terms of body weight, brain weight, recognition memory, as well as fasting intraperitoneal glucose tolerance, although APP/PS1 mice have 0.9 ± 0.2 thioflavin-S positive-amyloid plaques in the cerebral cortex. Yet these cognitively-healthy APP/PS1 mice exhibited conspicuous extravasations of casting solution from cerebral capillaries in addition to highly significant reductions in the number of hepatic microvessels (p=0.0002), an elevated number of splenic microvascular intussusceptive pillars (p=0.0231), low levels of proteinuria (p=0.0079) and reduced renal albumin clearance (p=0.0079), but with no change in creatinine excretion (p=0.4902). In chapter 5 we show that the prophylactic treatment of 4-5 month old APP/PS1 mice with liraglutide prior to neurodegeneration did not influence body weight, brain weight, food intake, fasting intraperitoneal glucose tolerance, serum albumin or reduce the numbers of thioflavin-S positive plaques in brain.

Conclusion
These results strongly support an initiatory and/or contributory role for the systemic vasculature in the pathological progression of neurodegeneration in APP/PS1 mice. These findings warrant an urgent examination of the vasculature as a diagnostic and/or therapeutic target for slowing the development of vascular-mediated neurodegeneration in common complex and progressive neurodegenerative diseases such as AD and Parkinson’s disease (PD) that are increasingly prevalent but lack disease-modifying therapies.

Date of Award	Mar 2016
Original language	English
Sponsors	Department for Employment and Learning (DEL)
Supervisor	Christopher Mitchell (Supervisor)