Regulation of Alzheimer’s disease-relevant protein processing in human neurons of the LUHMES cell line

Abstract

Alzheimer’s disease (AD) is a neurodegenerative disorder, whose major pathologies are the excessive generation of amyloid beta (Abeta) peptides and the aberrant phosphorylation of tau proteins. AD exists most commonly as sporadic form (SAD) without hereditary background (SAD). Interestingly, the level and/or activity of BACE, an enzyme crucially involved in Abeta generation, appears to be elevated in SAD brains. In order to replicate the disease states, transgenic mouse models and cell lines have been generated. Yet, the transfer of results from animals to humans is difficult and cell lines often lack neuronal properties or exhibit only weak endogenous expression of key proteins. As new approach in the framework of this doctoral thesis, we used human neurons, differentiated in vitro from the LUHMES cell line.Initially, we established a new 2-step differentiation protocol. We showed that by this procedure, the precursor cells irreversibly converted into post-mitotic neurons within 5 days, accompanied by an extensive outgrowth of neurites and the upregulation of synaptic proteins. It was possible to trigger neuronal differentiation even in the absence of the medium factors cAMP and GDNF, but the expression of some dopaminergic markers and the production of dopamine depended on the presence of cAMP. Thereby, we described for the first time LUHMES with a ‘non-dopaminergic’ phenotype, which are suitable for applications in various research fields. Next, we characterized LUHMES as human model for AD-relevant studies. We investigated how the expression and interaction of key proteins like amyloid precursor protein (APP) and BACE were regulated. During long-term cell culture (10 days), we observed that Abeta generation and tau phosphorylation continuously increased, and that both could be pharmacologically or biologically modulated like in primary neurons. We also revealed that the Abeta increase was induced through activation of the RET receptor by growth factors such as GDNF, and that the PI3K pathway downstream of RET was involved. These data indicate that a growth factor elevation, e.g. as defense mechanism in the AD brain, can lead to further augmentation of Abeta production. Finally, we established a stable BACE-overexpressing LUHMES cell line in order to mimic SAD conditions. Interestingly, BACE overexpression was not constitutive but progressively increased during differentiation of the cells, resulting in a correspondingly enhanced beta-cleavage of APP. However, while moderate BACE overexpression was linked to strong Abeta production as expected, BACE levels above a certain threshold prompted the cells to generate significantly less Abeta than wildtype LUHMES. In addition, the reduction of BACE activity in these cells, e.g. by treatment with low concentrations of BACE inhibitors, led to a strong rise in Abeta. These findings contribute to the understanding of the complex biology of APP processing and may have implications for the development of partial BACE inhibitors as AD therapeutic.