Alzheimer’s disease (AD) is a neurodegenerative disease characterized by the abnormal accumulation and aggregation of amyloid beta (Aβ) peptides within the brain. Generation of Aβ occur when the amyloid-beta precursor protein (APP) is proteolytically processed by β- and then γ-secretase in the amyloidogenic pathway. However, if APP instead is cleaved by α- and γ-secretase in the non-amyloidogenic pathway, Aβ formation is prevented and neuroprotective sAPPα is generated. In addition to these canonical processing pathways, APP can also be cleaved along non-canonical pathways by Δ, η, caspase or Meprinβ, resulting in numerous fragments that have different functional properties. The trafficking and processing of APP is a complex process and can be regulated by the adaptor protein Fe65. Following γ-secretase mediated cleavage of APP, the intracellular domain of APP and Fe65 can together translocate into the nucleus and regulate nuclear signaling. However, the exact mechanisms of how APP processing and APP/Fe65 nuclear signaling are regulated is still unclear.  

The aim of this thesis was to study different factors that may influence the regulation of APP processing and Fe65 nuclear localization. We found that phosphorylation of APP at Ser⁶⁷⁵ alters APP processing resulting in reduced levels of sAPPα and total sAPP, without affecting the plasma membrane level of APP. We could further observe an increased level of a slower migrating C99 like CTF, which was not generated by β-secretase cleavage of APP as there was no expression of BACE1 in the cell model used. Instead, generation of this CTF was blocked upon Meprinβ siRNA knockdown. Taken together these findings suggest that APP-Ser⁶⁷⁵ phosphorylation promotes Meprinβ processing of APP. In another study, we found that mutation of Ser²²⁸ at the Fe65 N-terminal dramatically increased the interaction between Fe65 and full-length APP. Moreover, this enhanced interaction resulted in decreased levels of non-amyloidogenic processing of APP and thus neuroprotective sAPPα. This suggest that the level of Fe65-APP interaction is important in regulating APP processing. Therefore, we also wanted to elucidate more about how the adaptor protein Fe65 is regulated. We found that Fe65 is likely phosphorylated on several residues in the N-terminus and that these phosphorylated forms preferentially localized in the cytoplasm. In addition, we could show that the nuclear level and nuclear/cytoplasmic ratio of Fe65 was decreased upon mutation of Fe65-Ser²²⁸ to glutamic acid, mimicking phosphorylation. Taken together this suggest that phosphorylation of Ser²²⁸ together with other residues in the N-terminus of Fe65 negatively regulate the Fe65 nuclear localization. In a third study, we could also show that the Fe65 PTB2 domain, rather than the WW domain, plays an important role in localizing Fe65 to the nucleus. Lastly, using different inhibitors, we found that blocking α-secretase processing decrease the Fe65 nuclear localization to the same extent as γ-secretase inhibition in both undifferentiated and RA or PMA differentiated cells. This suggest that α-secretase processing of APP or other Fe65 interacting transmembrane proteins play a more important role in regulation of Fe65 nuclear localization than previously thought. Interestingly, while ADAM10 was the most important α-secretase mediating this effect in undifferentiated cells, other α-secretases, likely ADAM17, played a more important role in RA or PMA differentiated neuroblastoma cells.

In summary, the results obtained in this thesis have increased the understanding of APP processing and how the adaptor protein Fe65 may act as a molecular switch altering APP cleavage.