We hypothesized that the physical interaction between the amyloid precursor protein (APP) and Notch 1 (N1) may be mediating the reported cross-talk between the respective signaling pathways. Immunoprecipitation of mouse N1 (mN1) or extracellular domain truncated mN1 (mN1-TM, mimics TACE-produced membrane-bound C-terminal fragment) specifically coprecipitated APP(751). Conversely, immunoprecipitation of APP(751) specifically coprecipitated mN1, furin-generated membrane-bound mN1 C-terminal fragment (f.mN1-TM), or mN1-TM. The London mutation of APP did not affect the APP(751)/mN1 interaction. Coexpression of APP(751) and mN1 did not affect APP processing or production of mN1 intracellular domain (mNICD). The APP(751)/mN1 interaction was Numb-independent, insofar as it was observed in HEK293 cells that lack detectable levels of Numb and was unaffected by the expression of exogenous Numb or deletion of the APP cytoplasmic domain, including the Numb-binding YENPTY sequence. This interaction was unaffected even when the N-terminal 647 amino acids of APP were replaced by a sequence of secreted alkaline phosphatase. These data combined with data showing interaction between mN1-TM and APP(751) suggest that their transmebrane domains and short sequences around them are sufficient for the interaction and that APP(751) and mN1 interact in cis. Our results imply novel functions of APP and/or N1 that derive from their interaction.

Presenilin-1 (PS1) has gained intensive attention in relation to Alzheimer’s disease, since it has been shown that PS1 mutations are linked to familial Alzheimer’s disease (FAD), and that PS1 is a member of the high molecular weight complex of γ-secretase, which generates the carboxyl end of β-amyloid peptide (γ-cleavage). A parallel line of evidence suggests that upon formation of cell-cell contacts, presenilin colocalizes with cadherins at the cell surface and stabilizes the cadherin-based adhesion complex. Under conditions stimulating cell-cell dissociation, cadherins are processed by a PS1/γ-secretase activity, promoting disassembly of adherens junctions, and resulting in the increase of cytosolic β-catenin, which is an important regulator of the Wnt/Wingless signaling pathway. PS1 also controls the cleavage of a number of transmembrane proteins at the interface of their transmembrane and cytosolic domains (Ε-cleavage), producing intracellular fragments with a putative transcriptional role. Remarkably, cleavage of N-cadherin by PS1 produces an intracellular fragment that downregulates CREB-mediated transcription, indicating a role of PS1 in gene expression. PS1 mutations associated with FAD abolish production of the N-cadherin intracellular fragment and thus fail to suppress CREB-dependent transcription. These findings suggest an alternative explanation for FAD that is separate from the widely accepted ‘amyloid hypothesis’: dysfunction in transcription regulatory mechanisms.

No abstract

In order to study N1 processing, we expressed human N1 (hN1) in HEK293 cells (293-hN1). Following Western blot analysis of 293-hN1 extracts, we detected, in addition to full-length hN1 and the N1 extracellular domain truncated form (N1-TM), a novel extracellular domain truncated form of hN1 with a COOH-terminal deletion, designated hN1-TMΔCT. Treatment of cells with the γ-secretase inhibitor L-685,458 resulted in an accumulation of hN1-TMΔCT suggesting that this fragment is a γ-secretase substrate. To identify the proteolytic activity(ies) that generates hN1-TMΔCT, we treated 293-hN1 cells with inhibitors of proteasome, calpains, caspases, serine and cysteine proteases. Despite the presence of a caspase-3 cleavage site within hN1 intracellular domain, none of the caspase inhibitors inhibited hN1-TMΔCT production. The proteasomal inhibitors used had also no effect. Incubation of cells with the cysteine protease inhibitor E64d resulted in the accumulation of hN1-TM and the inhibition of hN1-TMΔCT production suggesting a precursor-product relationship and that a cysteine protease is involved. Similarly, treatment of cells expressing amyloid precursor protein or E-cadherin with E-64d resulted in the accumulation of COOH-terminal fragments suggesting that these proteins are also processed within their intracellular domain by a cysteine protease. Processing towards hN1-TMΔCT requires maturation and transport of hN1 to the cell surface since treatment with brefeldin A inhibited its production and resulted in accumulation of hN1. Processing of hN1 within its intracellular domain could generate fragments that can exert novel functions and/or interfere with the function of hN1 intracellular domain.