Deposition of the amyloid-beta protein (An) is an early and invariant event in the pathogenesis of Alzheimer's disease (AD), and the two proteases that cut Abeta from the amyloid precursor protein (APP), beta- and y-secretases, are considered important therapeutic targets. y-Secretase catalyzes an unusual proteolysis within the single transmembrane region of APP, and this cleavage requires the multi-transmembrane presenilins. More than 60 different missense mutations in presenilins cause AD, and these mutations alter y-secretase activity to increase levels of a highly fibrillogenic 42-residue variant of Abeta (Abeta42). Inhibitor profiling, molecular modeling, and mutagenesis studies reveal that y-secretase has properties of an aspartyl protease and may catalyze an unusual intramembranous proteolysis. Moreover, two conserved transmembrane aspartates in presenilins are required for y-secretase activity, and transition-state analogue inhibitors of y-secretase bind directly and specifically to presenilins. Taken together, these results strongly suggest that presenilins themselves are y-secretases, novel intramembrane-cleaving aspartyl proteases. The objectives of the present proposal are to understand the topography of the y-secretase active site, to identify presenilin residues that are in or near the active site, and to develop compounds for studying the normal role of y-secretase in vivo and its potential as a therapeutic target for AD. The steric limits of y-secretase pockets S3 through S4' will be probed by systematically varying the size of the corresponding substituents in transition-state analogue inhibitors. The discovery of potent compounds would allow the development of affinity matrices for the purification of this protease and identification of associated cofactors. Specific regions within presenilins that interact with such inhibitors (i.e., parts in or near the active site) will be identified by photoaffinity labeling. The specific labeling of transmembrane residues of presenilins would provide compelling evidence for an intramembranous active site. Finally, nonpeptide analogues of these transition-state mimics will be developed using combinatorial chemistry as part of a strategy to identify metabolically stable compounds that block y-secretase activity in whole organisms.
Showing the most recent 10 out of 34 publications
