Deregulation of oxidative phosphorylation system and energy homeostasis in Alzheimer's disease

Alzheimer’s disease (AD) is the most frequent form of dementia among the elderly
affecting dozens of million people worldwide. Post-mortem, the disease is characterized by
two main neuropathological hallmarks: extracellular amyloid plaques and intracellular
neurofibrillary tangles (NFTs). Amyloid plaques are composed of the amyloid-beta (Aβ)
protein, derived from its amyloid precursor protein (APP). NFTs are formed from paired
helical filaments composed of hyperphosphorylated tau, a microtubule-associated-protein.
Besides these well-characterized features, a growing body of evidence supports mitochondrial
dysfunctions as part of the spectrum of chronic oxidative stress occurring in AD. This energy
deficit may contribute finally to synaptic abnormality and neuronal degeneration observed in
selected brain areas of AD patients. Nevertheless, the specific mechanisms leading to
mitochondrial failure as well as the role of Aβ or/and tau within this process are only partly
understood.
The purpose of the thesis was therefore to elucidate the role of mitochondria in the
pathogenesis of AD. Specifically the thesis was designed to determine (1) the synergistic
effects of Aβ-tau interplay, (2) the impact of soluble Aβ forms and (3) effects of Ginkgo
biloba extract (GBE) on mitochondria in several models of AD.
(1) While many studies reported effects of amyloid plaques on energy metabolism, the
role of tau pathology was until recently unknown. In a previous study, our group has been the first to show that tau was also able to induce mitochondrial dysfunction and raise reactive
oxygen species (ROS) levels in brains of P301L mutant tau transgenic pR5 mice (pR5).
Moreover, we found an increased mitochondrial vulnerability of pR5 cortical cells towards
Aβ in vitro. Based on these findings, we hypothesized that Aβ and tau might share toxicity at
the mitochondrial level. To reveal proof in vivo, we investigated the brains of wild-type
control mice and three transgenic mouse models. Transgenic pR5 mice express P301L mutant
tau found in the frontotemporal dementia with Parkinsonism linked to chromosome 17
(FTDP-17), a dementia related to AD. These mice model the tangle pathology of AD but lack
Aβ plaques. Furthermore, they show a hippocampus- and amygdala-dependent behavioural
impairment related to AD. APP152 double-transgenic mice co-express the N141I mutant form
of PS2 together with the APPSwe mutant found in familial cases of AD (FAD). These mice
model the Aβ plaques pathology, but fail to form NFTs. In addition, the mice display agerelated cognitive deficits associated with discrete brain Aβ deposition and inflammation.
Finally, we crossed the two strains to generate tripleAD mice. In addition to be a robust model
mimicing both plaques and tangles, this new transgenic line presents both amyloidosis and
NFTs formation in an age-dependent manner. The progression of biochemical changes and
histopathological features in the mice is reminiscent of the course of AD pathogenesis. The
mice develop behavioral deficits before detection of protein aggregates correlating with the
early mitochondrial dysfunction hypothesis proposed in AD. We applied the optimized
quantitative mass-tag labelling proteomic technique, iTRAQ and nanoLC-ESI MS/MS mass
spectrometry, followed by sophisticated high-resolution assays for metabolic and energetic
functions. We demonstrated massive deregulation of 24 proteins of which one third were
mitochondrial proteins mainly related to complexes I and IV of the mitochondrial respiratory
system from the four strains of mice. Our functional analysis validated the proteomic
approach by confirming the strongest defects of the respiratory capacity mainly at complexes
I, IV and V in tripleAD mice. Taken together, we demonstrated for the first time stringent
mitochondrial respiratory capacity dysfunction and a failure to restore the energy metabolism
in presence of both Aβ and tau.
(2) However, how these lesions and their proteinaceous components impair
mitochondrial functions and ultimately lead to neuronal cell loss are unresolved so far.
Intriguingly, some recent studies suggest that oligomeric Aβ species may be the main culprit,
rather than fibrillar. This idea highlights the critical role of mitochondrial abnormalities in the
biochemical pathway by which intracellular Aβ can lead to neuronal dysfunction in AD. To
test this experimental paradigm, we examined in a second study the specific effects of soluble
Aβ on mitochondrial function under physiological conditions. To this end, human
neuroblastoma cells (SH-SY5Y) were stably transfected with cDNAs containing either a
vector alone (control cells) or the entire coding region of human wild-type APP (APP695).
APP cells led to a significantly increased Aβ secretion compared to control cells and mimiced
relevant conditions for AD patients as Aβ levels were within a picomolar range. We
established a novel high-resolution respiratory protocol to perform whole cell recording of
total cellular respiration and mitochondrial metabolic states. To ripen our analyses, individual
activity of mitochondrial respiratory enzymes (complex I to IV) and ATP levels were
measured. We concluded that chronic exposure to soluble Aβ results (i) in serious impairment
of mitochondrial respiratory machinery due to activity changes of complexes III and IV leading finally to (ii) a drop of ATP synthesis. This energy metabolism deficit may in turn
accelerate/lead to cell death commonly observed in AD.
(3) Finally, we resumed the previous work by investigating the potential protective
effect of standardized GBE (LI 1370) on Aβ-induced mitochondrial failure. Mainly, the
antioxidant properties of GBE have been proposed as dietary strategies for many years in agerelated
cognitive disorders including AD. We showed for the first time that under
physiological conditions GBE improves metabolic energy pathways by increasing the
coupling state of mitochondria per se, but with specific benefit in APP cells exhibiting Aβ-
induced mitochondrial failure. GBE effect on OXPHOS was even preserved in mitochondria
after isolation from treated cells. The GBE-induced amelioration of oxygen consumption most
likely arose from the modulation and respective normalization of the activity of mitochondrial
complexes I, III and IV that are markedly disturbed in APP cells finally yielding a rise in ATP
levels. Of note, these functional data were paralleled by an up-regulation of mitochondrial
DNA in GBE-treated cells.
In summary, the present thesis took aim to highlight the key role of mitochondria in
AD pathogenesis and the close inter-relationship of this organelle with the two main
pathological features of the disease. First, we demonstrated main defects of mitochondrial
respiratory capacity and the failure to restore energy homeostasis in mice with plaques and
tangles. Although, a molecular link between Aβ and tau is still missing, these in vivo results consolidate the idea that a synergistic effect of tau and Aβ augments the pathological
deterioration of mitochondria in AD by driving a vicious cycle. Secondly, we proved toxicity
of soluble Aβ forms, recently defined as the toxic correlate within the Aβ cascade, on the
mitochondrial function of vital cells. Finally, the critical role of mitochondria in early
pathogenesis of AD may make them into a preferential target for treatment strategies such as
antioxidants. Our work confirmed this idea and clearly showed stabilization and restoration of
energy metabolism in APP cells treated with GBE. In view of the increasing interest in
mitochondrial protection as treatment strategy in dementia, our findings of substantial
protection of mitochondria by GBE against Aβ-induced dysfunction deserves further
attention.