The amyloid-β precursor protein (APP) has been widely studied due to its role in Alzheimer´s disease (AD). When APP is sequentially cleaved by β- and γ-secretase, amyloid-β (Aβ) is formed. Aβ is prone to aggregate and is toxic to neurons. However, the main processing pathway for APP involves initial cleavage at the α-site, within the Aβ region, instead generating a neuroprotective soluble fragment, sAPPα. APP is a member of a protein family, also including the proteins APLP1 and APLP2, which are processed in a similar way as APP. In addition, K/O studies in mice have shown that the three proteins have overlapping functions where APLP2 play a key physiological role. The aim of this thesis was to study mechanisms underlying the α-secretase processing of APP and APLP2. We have used the human neuroblastoma cell-line SH-SY5Y as a model system and stimulated α-secretase processing with insulin-like growth factor-1 (IGF-1) or retinoic acid (RA). Our results show that the stimulated α-site cleavage of APP and APLP2 is regulated by different signaling pathways and that the cleavage is mediated by different enzymes. APP was shown to be cleaved by ADAM10 in a PI3K-dependent manner, whereas APLP2 was cleaved by TACE in a PKC-dependent manner. We further show that protein levels and maturation of ADAM10 and TACE is increased in response to RA, mediated by a PI3K- or PKC-dependent signaling pathway, respectively. Another focus of our research has been O-GlcNAcylation, a dynamic post-translational modification regulated by the enzymes O-GlcNAc transferase and O-GlcNAcase (OGA). We show that decreased OGA activity stimulates sAPPα secretion, without affecting APLP2 processing. We further show that ADAM10 is O-GlcNAcylated. Lastly, we show that APP can be manipulated to be cleaved in a similar way as APLP2 during IGF-1 stimulation by substituting the E1 domain in APP with the E1 domain in APLP2. Together our results show distinct α-site processing mechanisms of APP and APLP2.

The amyloid precursor protein (APP) protein has been in the limelight of research on Alzheimer´s disease (AD) pathogenesis because its proteolytic processing gives rise to the neurotoxic amyloid β (Aβ) peptide, the main constituent of amyloid plaques in the brains of AD patients. APP is sequentially processed by at least three different proteases termed α-, β-, and γ-secretases. The proteolytic processing of APP can be divided into two different pathways, the non-amyloidogenic and the amyloidogenic. Whether APP is processed by the non-amyloidogenic or the amyloidogenic pathway is highly dependent on colocalization of APP with the different processing enzymes. Hence, understanding the mechanism underlying regulation of APP trafficking and its related secretases is of great importance in our understanding of AD and AD pathogenesis. The aim of this thesis was to study the processing and trafficking of APP, how it may be regulated by the interaction with the adaptor protein, Fe65, and by a novel type of posttranslational modification, O-GlcNAcylation. We have used the human neuroblastoma cell line SH-SY5Y as a modell system to investigate the effect of Fe65 knock-down on APP processing. Our results showed that Fe65 knockdown did not have any effect on sAPPα secretion. However, decreased levels of C83 and C99 were observed, suggesting that Fe65 has a stabilizing effect on the C-terminal fragments. Furthermore, we investigated the effects of RA-induced neronal differentiation on Fe65 expression. We observed increased protein levels of Fe65 and an electrophoretic mobility shift due to increased phosphorylation of Fe65. O-GlcNAcylation is a dynamic posttranslational modification regulated by O-GlcNAc transferase (OGT) and O-GlcNAcase (OGA). To investigate the effect of O-GlcNAcylation on APP trafficking and processing, SH-SY5Y cells were treated with PUGNAc, an OGA inhibitor, to increase the cellular levels of O-GlcNAc. The results revealed that cell surface localization of mature APP was significantly enhanced without any affect on the total levels of APP. We further show evidence that ADAM10 is O-GlcNAcylated and that the effect of O-GlcNAcylation on APP processing is neuron-specific.