Impact of amyloid-β reduction on secondary pathological changes in transgenic mouse lines modelling Alzheimer-like pathology

Abstract

Alzheimer’s disease (AD) has grown to a global health crisis which comes with an urgent need for disease-modifying and preventative therapies. To advance current therapeutic approaches, it is crucial to identify the first pathogenic event (presumably amyloid-β misfolding) as well as mechanistically defined biomarkers along AD disease progression. In the first part of this thesis, we generated biomarker trajectories in an amyloid-β precursor protein (APP) transgenic mouse model that is widely used in the AD research field. We were then able to mimic in this mouse model the discrepancy in clinical settings between a therapeutic reduction of amyloid-β (Aβ) deposition and a lack of cognitive improvement, as assessed by neurofilament levels in the cerebrospinal fluid. Our data indicate an Aβ-dependent disease phase at which Aβ reduction prevented disease-associated neurodegeneration and an Aβ-independent phase characterized by proceeding neurodegeneration despite a reduction of brain Aβ. Interestingly, robust neurodegeneration was mainly associated with a saturated Aβ seeding activity, which in turn was not proportional to the level of brain Aβ deposition. Our data raise the hypothesis that seeding-active Aβ species are an important molecular link between Aβ deposition and neurodegeneration. The evaluation of this concept and its application in humans awaits further investigations. In the second part of this thesis, we used an APP transgenic mouse model with a prolonged Aβ aggregation lag phase. We then targeted the first Aβ nucleation event in vivo and characterized such early Aβ assemblies. Our results revealed that even before Aβ-deposition became histologically detectable, pre-amyloid Aβ species were clearly present. Further, immunotherapeutic targeting of such pre-amyloid seeds led to a substantial reduction in Aβ seeding activity and long-lasting beneficial effect on secondary pathologies. We thereby expand the therapeutic window for a promising Aβ-targeting therapy to a much earlier time point as previously assumed. Moreover, our data indicate that specific binding features of Aβ targeting antibodies determine whether an antibody recognized such early seeds. Since specific amyloid structures are associated with distinct toxicities, this knowledge offers potential implications for future antibody design for immunotherapies. In conclusion, the results of this thesis demonstrate the presence of pre-amyloid Aβ seeds. Direct targeting of such early Aβ seeds, or an early reduction in brain Aβ, have beneficial effects on secondary pathologies and neurodegeneration. Thus, our results expand the current therapeutic window to a new established disease phase characterized by the presence of pre-amyloid seeds and low Aβ seeding activity with high potential for disease prevention.