No abstract

Alzheimer's disease (AD) is characterized by senile plaques caused by amyloid-β peptide (Aβ) accumulation. It has been reported that Aβ generation and accumulation occur in membrane microdomains, called lipid rafts, which are enriched in cholesterol and glycosphingolipids. Moreover, the ablation of cholesterol metabolism has been implicated in AD. Neprilysin (NEP), a neutral endopeptidase, is one of the major Aβ-degrading enzymes in the brain. Activation of NEP is a possible therapeutic target. However, it remains unknown whether the activity of NEP is regulated by its association with lipid rafts. Here we show that only the mature form of NEP, which has been glycosylated in the Golgi, exists in lipid rafts, where it is directly associated with phosphatidylserine. Moreover, the localization of NEP in lipid rafts is enhanced by its dimerization, as shown using the NEP E403C homodimerization mutant. However, the protease activities of the mature form of NEP, as assessed by in vitro peptide hydrolysis, did not differ between lipid rafts and nonlipid rafts. We conclude that cholesterol and other lipids regulate the localization of mature NEP to lipid rafts, where the substrate Aβ accumulates but does not modulate the protease activity of NEP.

BACE1 initiates processing of the amyloid precursor protein (APP) in the production of amyloid beta (Abeta) peptide. After beta-cleavage by BACE1, the C-terminal stub of the APP fragment is further processed by the gamma-secretase complex to produce Abeta. Because APP, Abeta, the gamma-secretase complex, and BACE1 are found in lipid raft membranes, Abeta production is widely accepted to occur in lipid rafts. However, whether BACE1 is activated within the rafts is unclear. To analyze the relationship between the activity and the localization of BACE1, we used a new BACE1 inhibitor, KMI-574, and separated raft membranes on sucrose density gradients. In the presence of KMI-574, the localization of BACE1 shifted from the rafts to nonraft membranes in HEK293 cells. We also analyzed the proteolytically inactive mutants, D93A, D289A, and D93A/D289A, of BACE1. These mutants also moved from rafts to nonrafts, and the D93A/D289A double-mutant localized exclusively to nonraft membranes. The mutants were defective in maturation by glynosylation and formed hyperoligomers, suggesting that the BACE1 oligomers could not exit from the ER and be transported to the Golgi apparatus. Our findings suggest that the activated conformation of BACE1 is important for protein transport and localization to lipid rafts.