Search for a command to run...
The final proteolytic step to generate the amyloid beta-protein (Abeta) of Alzheimer's disease (AD) from beta-amyloid precursor protein (APP) is achieved by presenilin (PS)-dependent gamma-secretase cleavage. AD-causing mutations in PS1 and PS2 result in a selective and significant increase in production of the more amyloidogenic Abeta42 peptide. PS1 and PS2 undergo endoproteolysis by an unknown enzyme termed presenilinase to generate the functional complex of N- and C-terminal fragments (NTF/CTF). To investigate the endoproteolytic activity that generates active PS, we used a mammalian cell-free system that allows de novo human PS NTF and CTF generation. PS NTF and CTF generation in vitro was observed in endoplasmic reticulum (ER)-enriched fractions of membrane vesicles and to a lesser extent in Golgi/trans-Golgi-network (TGN)-enriched fractions. AD-causing mutations in PS1 and PS2 did not alter de novo generation of PS fragments. Removal of peripheral membrane-associated and cytosolic proteins did not prevent de novo generation of fragments, indicating that presenilinase activity corresponds to an integral membrane protein. Among several general inhibitors of different protease classes that blocked the presenilinase activity, pepstatin A was the most potent inhibitor. Screening available transition state analogue gamma-secretase inhibitors led to the identification of two compounds that were able to prevent the de novo generation of PS fragments, with an expected inhibition of Abeta generation. Our studies provide a biochemical approach to characterize and identify this elusive presenilinase.