Amyloid-β and chronic cerebral hypoperfusion in the early pathogenesis of Alzheimer’s disease
Salvadores Bersezio, Natalia
Alzheimer’s disease (AD) is a severe age-related neurodegenerative disorder and is the most common form of dementia. Although the pathogenesis of AD remains unknown, the deterioration of the cerebrovascular system constitutes a risk factor associated with the development of the disease. Notably, brain hypoperfusion, a feature of healthy ageing brain and AD, occurs prior to the onset of cognitive decline in AD and correlates with the severity of dementia. Although there is a clear link between hypoperfusion and cognitive alterations in AD, a causal relationship remains to be established. It was hypothesised that chronic cerebral hypoperfusion leads to the accumulation of parenchymal and vascular amyloid-β (Aβ), triggering the development of vascular lesion (microinfarcts (MIs) and haemorrhages) and altering the neurovascular unit (NVU) integrity. Second to this, it was hypothesised that reductions in Aβ levels by immunotherapy targeted to amyloid in young mice, reduce amyloid levels, and prevent vascular lesions improving cognitive performance. Three studies were conducted to test these hypotheses. In the first study, the aim was to characterise age-dependent changes in amyloidrelated pathology in a transgenic mouse model (Tg-SwDI). The temporal amyloid precursor protein (APP) expression, accumulation of parenchymal and cerebrovascular Aβ and Aβ-related microglial and astrocytic activation in the cortex, hippocampus and thalamus of the Tg-SwDI mice at 3, 6 and 9 months of age was compared to wild-type controls. Significantly higher APP expression (p < 0.05), as well as Aβ aggregation (p < 0.001) as the animals aged was found in the Tg-SwDI mice in all the brain regions analysed, which was accompanied by extensive and progressive activation of microglial (p < 0.001) and astrocytic (p < 0.01) cells. These data provided a basis to design the next studies, as it was planned to induce hypoperfusion in these mice before significant Aβ deposition occurs. In the second study, the aim was to investigate the effect of hypoperfusion on Aβ dynamics and subsequently, to study the contribution of hypoperfusion and Aβ pathology to the development of MIs and haemorrhages, and to the potential alteration of astrocyte and tight junction (TJ) integrity. To address this, mild chronic cerebral hypoperfusion was induced in Tg-SwDI and wild-type mice by bilateral common carotid stenosis for 1 and 3 months. A significant increase in soluble Aβ40/42 levels was initially found after 1 month of hypoperfusion in the parenchyma (Aβ40, p = 0.0239; Aβ42 p = 0.0198) in parallel with elevated APP levels and APP proteolytic cleavage products (p < 0.05). Thereafter, following 3 months, a significant increase in insoluble Aβ40/42 levels was determined in the parenchyma (Aβ40, p = 0.0024; Aβ42 p = 0.008) and vasculature (Aβ40, p = 0.0046; Aβ42 p = 0.0118) of Tg-SwDI mice. There was no change in the levels of Aβ co-localised to vessels following 1 month of hypoperfusion; however Aβ levels were significantly increased in cerebral vessels after 3 months (p = 0.0483). The proportion of Aβ containing vessels was significantly higher in the small vessels of the hypoperfused animals compared to sham mice (p < 0.05). MIs associated with microglial proliferation were present in the Tg-SwDI mice and the burden was exacerbated by hypoperfusion at 1 and 3 months (p < 0.05). Significantly higher levels of NADPH Oxidase-2 (NOX2) were found in the transgenic mice compared to the wild-type controls at both time-points analysed (p < 0.05), and this was exacerbated after 1 month of hypoperfusion in the Tg-SwDI mice (p < 0.05). There was a positive correlation between NOX2 and soluble parenchymal Aβ levels (r = 0.6643, p = 0.0019). A minimal effect on the development of haemorrhages at these time-points was observed. In parallel to this, astrocyte activation was significantly higher in the Tg-SwDI mice compared to the wild-type controls at both time-points studied (p < 0.05); however, no effect of hypoperfusion was observed. Also, significantly higher levels of aquaporin-4 (AQP4) in the Tg-SwDI mice compared to the wild-type controls following 1 month of hypoperfusion were found (p < 0.001). There was a positive correlation between AQP4 and soluble parenchymal Aβ levels (r = 0.4735, p = 0.0095). Claudin-5 levels were significantly higher in the Tg-SwDI mice compared to the wild-type controls at both time-points analysed (p < 0.0001), and this was exacerbated following 1 month of hypoperfusion in the transgenic model (p < 0.05). A positive correlation between claudin-5 and vascular Aβ levels was observed (r = 0.6113, p = 0.0004). Together, these data suggest a synergistic contribution of amyloid and hypoperfusion pathologies to the tissue damage and implicate a role of oxidative stress and inflammation. In the third study, the aim was to determine the effects of passive amyloid immunisation on Aβ levels, development of MIs and haemorrhages and behavioural performance in the Tg-SwDI mice. To address this, the mice underwent weekly intraperitoneal injections with either 3D6 or 10D5 antibodies during 3 months. Although there were no significant changes between control and 10D5/3D6 treated mice in amyloid levels, appearance of MIs and cognitive performance, it was noted that there was a trend towards a reduction in amyloid levels and MI area in the 10D5/3D6 treated mice compared to the control animals. Furthermore, there was no evidence of microhaemorrhages in response to the immunisation. These results demonstrate that Aβ immunotherapy with the antibodies 3D6 and 10D5 may potentially decrease parenchymal and vascular amyloid accumulation, reducing the appearance of MIs and notably without triggering the development of microhaemorrhages. Collectively, the findings presented in the current thesis demonstrate that chronic cerebral hypoperfusion increases parenchymal and vascular Aβ levels and point towards a mechanism in which the cascade of events including inflammation and oxidative stress, triggered synergistically by hypoperfusion and Aβ, resulted in the widespread development of MIs and NVU changes which may further induce the alteration of cognition networks. A mixed therapy, aimed at improving cerebrovascular health and targeting the accumulation of Aβ, represents a promising strategy to prevent neurodegenerative processes and further cognitive decline in AD.