Mutations in the presenilin (PS) genes are the major cause of dominantly inherited Alzheimer's disease (AD). Therefore, a major goal in the quest to decipher the molecular mechanisms of AD is to understand the function (and dysfunction) of PS1 and PS2. Consistent with other reports, we have shown that AD-causing mutations in PS1 and 2 selectively enhance the cellular production and cerebral deposition of Abeta42. We have also obtained evidence that small amounts of APP and PS form complexes, and we have confirmed the observation of De Strooper et al. that the deletion of PS1 markedly reduces gamma-secretase processing of APP. Recently, we made the unexpected discovery that mutating either of 2 unusual and conserved transmembrane (TM) aspartates in PS1 blocked PS1 endoproteolysis and sharply reduced Abeta production. These findings lead us to hypothesize that presenilins have both autoproteolysis and gamma-secretase cleavage activities. This hypothesis will be tested by carrying out four Specific Aims: 1) examine the role of the TM Asp residues in PS1 endoproteolysis by introducing an Asp->Glu mutation or slightly shifting the """"""""active site"""""""" (the 2 aligned Asp residues) in PS1, and express human PS1 in E. coli to search for endoproteolytic products, 2) examine the role of the Asp residues in Abeta generation by studying the effect of multiple substitutions of the Asps on intracellular and extracellular Abeta levels, and by co-expressing PS1deltaE9 and C99 in E. coli to confirm the intrinsic gamma-secretase activity of PS1; 3) dissect the proteolytic activities of PS2 and examine if wt PS2 can rescue the biochemical phenotypes of Asp-mutant PS1 (reduced Abeta generation) and FAD mutant PSI (increased Abeta42 generation), in order to determine whether PS2 has a closely similar function to PS l; and 4) generate and characterize Asp-mutant PS1 tg mice to confirm the gamma-secretase activity of PS in vivo. After crossing APP tg mice with viable Asp-mutant PS1 tg mice, neuropathological effects in the offspring will be examined. All four Aims are based on unexpected but well documented and robust preliminary data. The results should provide novel information about the nature of the intramembranous proteolysis of transmembrane proteins (including APP and PS1) and should elucidate the unusual mechanism of Abeta generation, an emerging therapeutic target in AD.