The contribution of N-terminally modified amyloid beta to the etiology of Alzheimer's disease

A heterogeneou s mixture of amyloid beta (Aβ) isoforms exists in the brains of Alzheimer's disease (AD) patients. Despite decades of research, relatively little is known about the precise contribution of these variousAβ species to the development and progression of AD. Recent work has identified pyroglutamate-modified amyloid beta (Aβ pE) as a particularly abundant and toxic peptide. Transgenic mice designed to specifically overproduceAβ pE exhibit neuron loss and behavioral deficits. Unfortunately, general breeding issues and the severe pathology found in these models restrict their use for additional transgenic studies. In the first project of this thesis, the TBA42 mouse model was generated to overcome these problems and further explore the consequences ofAβ pE accumulation in vivo. Using immunohistochemistry, it was shown that TBA42 mice develop region-specific intraneuronal and extracellularAβ /Aβ pE deposits accompanied by progressive gliosis. Decreased anxiety and altered rearing behavior were the earliest and most persistent behavioral changes identified in this model. TBA42 mice also displayed age-dependent deficits in motor performance and working memory. The phenotype observed in the TBA42 model is comparable to otherAβ pE -generating transgenic mouse lines. However, the moderate pathology and behavioral impairments of TBA42 mice make them suitable for further transgenic experiments. Numerous studies have explored the therapeutic benefits of reducingAβ pE in AD mouse models. However, few have addressed whether elevatingAβ pE levels is sufficient to aggravate ongoing disease processes. Earlier attempts to answer this question relied on the ectopic overexpression of human glutaminyl cyclase (hQC) in an established AD mouse model. QC is the primary enzyme responsible for catalyzing the formation ofAβ pE. Since QC has multiple targets, it cannot be excluded that ectopic QC overexpression affected these other substrates, thereby influencing the results of previous experiments. To study how an exclusive increase inAβ pE affects AD pathology, the TBA42 and 5XFAD mouse models were crossed to produce FAD42 mice for the second project of this thesis. The 5XFAD mouse model is a well-characterized AD transgenic model with aggressive amyloid deposition. FAD42 mice exhibited aggravated behavioral deficits compared to 5XFAD and TBA42 mice. ELISA and plaque load measurements also revealed elevatedAβ pE in FAD42 mice. These results were accompanied by an increase in endogenous QC activity in FAD42 mouse brain. However, FAD42 mice displayed no changes inAβ x-42 or otherAβ isoforms, as determined by ELISA and mass spectrometry. In total, these observations support a key pathogenic role forAβ pE in AD and argue for its ability to seedAβ deposition. Aβ 4-42 is another majorAβ species in AD brain. Sedimentation studies suggested thatAβ 4-42 displays rapid aggregation kinetics, but nothing is known about the in vivo toxicity of this peptide. Most transgenic AD mouse models rely on artificial combinations of mutations to study amyloid pathology. However, the majority of AD patients do not possess mutations. Direct, moderate expression of a particularAβ species in vivo may therefore create a more physiologically relevant AD model. Given these considerations, the third project of this thesis focused on the creation of TBA83 mice, a transgenic model which exclusively expressesAβ 4-42. TBA83 mice exhibited sparse, region-specific intraneuronal and extracellularAβ deposits and mild gliosis. In addition, TBA83 mice displayed deficits in motor function and hippocampal-dependent memory. The lack of severeAβ deposition in TBA83 mice ultimately suggests a pathological function for solubleAβ 4-42. Taken together, the results of this thesis confirm the relevance ofAβ pE to AD progression. The pathogenic properties ofAβ 4-42 were also identified for the first time in vivo, warranting further studies of thisAβ isoform