Alzheimer disease associated Abeta and gamma-secretase : mitochondrial localization and involvement in cell death

Abstract

In several neurodegenerative disorders, including Alzheimer disease (AD) and Parkinson disease (PD), protein aggregation is accompanied by mitochondrial dysfunction. An increasing number of proteins found in mitochondria have multiple localizations. The Amyloid beta-peptide precursor protein (APP), a key protein in AD, has a chimeric targeting signal (endoplasmic reticulum (ER) + mitochondria). APP is subsequently cleaved by the beta- and gamma-secretase resulting in production of the Amyloid beta-peptide (Abeta) of different lengths (e.g. Abeta40 and Abeta42). The gamma-secretase complex consists of, presenilin (PS)-1 or -2, Nicastrin (Nct), Aph-1 and Pen-2. Interestingly, both PS and Abeta have been found in mitochondria. The aim of this thesis is to investigate if all the gamma-secretase components are present in mitochondria and furthermore, to study the mitochondrial uptake mechanisms for Abeta and to assess the stability and activity of the gamma-secretase complex during cell death.

In paper I we identified a chimeric targeting signal (ER+ mitochondria) in Nct and showed that active gamma-secretase complexes are localized to mitochondria.

In paper II Nicastrin was detected in mitochondria independently of PS, PS was detected in mitochondria independently of Nct, and Aph-1 was detected in mitochondria independently of PS or Nct. Further, Nct interacted with the mitochondrial import machinery and ~3% of all Nct and ~1.5% of the active gamma-secretase complexes were located to mitochondria. The topology of the mitochondrial gamma-secretase complex was similar to the reported topology of the gamma-secretase complex in ER.

In paper III Abeta40 and Abeta42 were found to be imported through the TOM-complex, translocase of the outer membrane, in the mitochondrial import machinery and found to be localized to the mitochondrial cristae. The same localization pattern was also detected in samples from human cortex with amyloid plaques.

In paper IV the stability and activity of gamma-secretase complexes during cell death were investigated. gamma-Secretase was shown to be active and to contain a caspase cleaved C-terminal of PS1 (PS1-caspCTF). Transient transfections with a caspCTF construct reconstituted the gamma- secretase activity in PS null cells (BD8) and apoptotic HEKAPPswe cells showed Abeta40 production.

In paper V the Abeta40 and Abeta42 production by gamma-secretase complexes containing PS1- caspCTF were investigated in cells stably expressing PS1-caspCTF or PS1-CTF on a PS null background (BD8 cells). Here we demonstrated a significant increase in the intracellular Abeta42/Abeta40 ratio in PS1-caspCTF cells as compared to PS1-wtCTF cells.

We here present data linking some of the central proteins of AD pathogenesis to mitochondria and indicating an altered Abeta production during cell death. Mitochondrial dysfunction may underlie the loss of synapses occurring in AD. Therefore, future drugs targeting mitochondria to strengthen their function may be a novel strategy for treatment of AD.