Studies on amyloid beta (Ab|), the peptide thought to play a crucial role in the pathogenesis of Alzheimer's disease, have implicated mitochondria in Ab-mediated neurotoxicity. We used differentiated PC12 cells stably transfected with an inducible green fluorescent protein (GFP) fusion protein containing an N¢-terminal mitochondrial targeting sequence (mtGFP), to examine the effects of sub-lethal Ab on the import of nuclear-encoded proteins to mitochondria. Exposure to sub-lethal Ab 25-35 (10 lmol/L) for 48 h inhibited mtGFP import to mitochondria; average rates decreased by 20 ± 4%. Concomitant with the decline in mtGFP, cytoplasmic mtGFP increased significantly while mtGFP expression and intramitochondrial mtGFP turnover were unchanged. Sub-lethal Ab 1-42 inhibited mtGFP import and increased cytoplasmic mtGFP but only after 96 h. The import of two endogenous nuclear-encoded mitochondrial proteins, mortalin/mtHsp70 and Tom20 also declined. Prior to the decline in import, mitochondrial membrane potential (mmp), and reactive oxygen species levels were unchanged in Ab-treated cells versus reverse phase controls. Sustained periods of decreased import were associated with decreased mmp, increased reactive oxygen species, increased vulnerability to oxygen-glucose deprivation and altered mitochondrial morphology. These findings suggest that an Ab-mediated inhibition of mitochondrial protein import, and the consequent mitochondrial impairment, may contribute to Alzheimer's disease.

Background: Mitochondrial protein import is typically measured by adding radiolabeled precursor proteins to isolated mitochondria. We have developed a novel, highthroughput method for measuring protein import in live differentiated PC12 cells using a tetracycline (Tet) regulated, nuclear encoded, mitochondrially-targeted GFP fusion protein and flow cytometry. Methods: We generated a PC12 cell line stably transfected with an inducible GFP fusion protein (GFPmt) targeted to mitochondria. GFPmt PC12 cells were treated with NGF for one week to induce neuronal differentiation in the presence of Tet to silence GFP expression. On day seven GFPmt expression was induced by removal of Tet and these "GFP-on" cells were exposed to sublethal levels of CCCP (2 M) for 24 h. At 24 h, the cells were harvested in Ca ϩϩ-free PBS and the GFPmt signal in live intact cells was measured using flow cytometry. Since GFPmt is not fluorescent prior to being imported into mitochondria, the GFPmt signal reflected only GFPmt imported to mitochondria. PI was used to gate out contributions from dead cells. Turnover of GFPmt in mitochondria was also assessed; in this case, Tet was added to arrest GFPmt expression in GFP-on cells, and the subsequent decline of the fluorescent signal, in the absence of any new GFP synthesis, was measured by flow cytometry. Results: Exposure to 2 M CCCP for 24 h caused a 61% Ϯ 0.4 decline in GFPmt fluorescence compared to controls. This decline corresponded to a 30% Ϯ 7 decrease in GFPmt protein levels measured by Western blot of mitochondrial fractions, and a 72% Ϯ 5 decline in the import of newly synthesized GFPmt to mitochondria over a 1 h period 24-h after addition of 2 M CCCP measured by autoradiography. CCCP partially depolarized mitochondria but was not lethal for up to five days. Conclusions: This novel GFP-based flow cytometry assay is a rapid and sensitive technique for quantifying protein import to mitochondria in live neuronal cells. Cytometry Part A 56A:15-22, 2003.