Jonathan Aow1,2, Tzu-Rung Huang3, Gopal Thinakaran4, Edward H Koo5,6,7. 1. Agency for Science, Technology and Research (A*STAR), Genome Institute of Singapore, Singapore, Singapore. aowsj@gis.a-star.edu.sg. 2. Department of Medicine, National University of Singapore, Singapore, Singapore. aowsj@gis.a-star.edu.sg. 3. Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore, Singapore. 4. USF Health Byrd Alzheimer's Center and Research Institute and Department of Molecular Medicine, University of South Florida, Tampa, FL, USA. 5. Department of Medicine, National University of Singapore, Singapore, Singapore. edkoo@health.ucsd.edu. 6. Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore, Singapore. edkoo@health.ucsd.edu. 7. Department of Neurosciences, University of California San Diego, San Diego, CA, USA. edkoo@health.ucsd.edu.
Abstract
BACKGROUND: Alzheimer's disease amyloid-beta peptides (Aβ) are generated via sequential cleavage of the amyloid precursor protein (APP) by β-secretase (Bace1) and γ-secretase. Though the precise subcellular location(s) of Bace1-mediated APP cleavage remains unresolved, current models suggest APP internalization into Bace1-containing endosomes is a critical step. However, direct evidence for this model is lacking, and previous reports that probed the APP/Bace1 interaction (using co-expressed APP and Bace1 differentially labeled with fluorescent protein tags) did not determine if APP fluorescence originated from full-length APP (fl-APP) molecules that had internalized from the cell surface pool. METHODS: We adapted the bungarotoxin-ligand (BTX) system to label surface APP and track internalized fluorescent APP/BTX puncta in rodent primary neurons co-expressing fluorescently-tagged Bace1. Subsequently, we employed imaging and biochemical-based approaches to measure N- and C-terminal APP epitope levels in primary neurons, N2a neuroblastoma, and HeLa cell lines. RESULTS: We hypothesized that surface-labeled APP/BTX puncta would, upon internalization, colocalize with fluorescently-tagged Bace1. Unexpectedly, we observed a dramatic loss of internalized APP in co-transfected neurons and ~ 80-90% loss of surface-resident fl-APP, which we also observed in HeLa and N2a cells. Loss of surface fl-APP could be reversed by a Bace1 inhibitor, suggesting that enhanced Bace1-mediated APP cleavage was responsible for the altered processing and mis-sorting. Importantly, in a C-terminally-tagged APP construct, the majority of C-terminal fluorescence was preserved in HeLa cells despite the loss of N-terminal APP signal. This phenomenon was not only recapitulated in cultured neurons, but also showed a progressive disappearance of the APP N-terminal tag, reflecting continual cleavage of fl-APP by Bace1 away from the cell body. CONCLUSIONS: Our results strongly suggested that in APP/Bace1 co-expression approaches, there was significant early and aberrant Bace1-mediated APP cleavage that perturbed fl-APP trafficking from the secretory pathway onwards, resulting in a substantial loss of surface fl-APP, which in turn led to a marked reduction in APP internalization. In C-terminally-tagged APP constructs, a large fraction of the APP fluorescence signal therefore likely arose from fluorescently-tagged β-C-terminal-fragment (β-CTF) or downstream proteolytic derivatives instead of fl-APP. Thus, care is needed in interpreting results where APP is detected only with a C-terminal tag in the presence of Bace1 co-expression, and previous findings may need to be reinterpreted if it is unclear whether fl-APP is present in normal physiological levels.
BACKGROUND: Alzheimer's disease amyloid-beta peptides (Aβ) are generated via sequential cleavage of the amyloid precursor protein (APP) by β-secretase (Bace1) and γ-secretase. Though the precise subcellular location(s) of Bace1-mediated APP cleavage remains unresolved, current models suggest APP internalization into Bace1-containing endosomes is a critical step. However, direct evidence for this model is lacking, and previous reports that probed the APP/Bace1 interaction (using co-expressed APP and Bace1 differentially labeled with fluorescent protein tags) did not determine if APP fluorescence originated from full-length APP (fl-APP) molecules that had internalized from the cell surface pool. METHODS: We adapted the bungarotoxin-ligand (BTX) system to label surface APP and track internalized fluorescent APP/BTX puncta in rodent primary neurons co-expressing fluorescently-tagged Bace1. Subsequently, we employed imaging and biochemical-based approaches to measure N- and C-terminal APP epitope levels in primary neurons, N2a neuroblastoma, and HeLa cell lines. RESULTS: We hypothesized that surface-labeled APP/BTX puncta would, upon internalization, colocalize with fluorescently-tagged Bace1. Unexpectedly, we observed a dramatic loss of internalized APP in co-transfected neurons and ~ 80-90% loss of surface-resident fl-APP, which we also observed in HeLa and N2a cells. Loss of surface fl-APP could be reversed by a Bace1 inhibitor, suggesting that enhanced Bace1-mediated APP cleavage was responsible for the altered processing and mis-sorting. Importantly, in a C-terminally-tagged APP construct, the majority of C-terminal fluorescence was preserved in HeLa cells despite the loss of N-terminal APP signal. This phenomenon was not only recapitulated in cultured neurons, but also showed a progressive disappearance of the APP N-terminal tag, reflecting continual cleavage of fl-APP by Bace1 away from the cell body. CONCLUSIONS: Our results strongly suggested that in APP/Bace1 co-expression approaches, there was significant early and aberrant Bace1-mediated APP cleavage that perturbed fl-APP trafficking from the secretory pathway onwards, resulting in a substantial loss of surface fl-APP, which in turn led to a marked reduction in APP internalization. In C-terminally-tagged APP constructs, a large fraction of the APP fluorescence signal therefore likely arose from fluorescently-tagged β-C-terminal-fragment (β-CTF) or downstream proteolytic derivatives instead of fl-APP. Thus, care is needed in interpreting results where APP is detected only with a C-terminal tag in the presence of Bace1 co-expression, and previous findings may need to be reinterpreted if it is unclear whether fl-APP is present in normal physiological levels.
Authors: R Vassar; B D Bennett; S Babu-Khan; S Kahn; E A Mendiaz; P Denis; D B Teplow; S Ross; P Amarante; R Loeloff; Y Luo; S Fisher; J Fuller; S Edenson; J Lile; M A Jarosinski; A L Biere; E Curran; T Burgess; J C Louis; F Collins; J Treanor; G Rogers; M Citron Journal: Science Date: 1999-10-22 Impact factor: 47.728