The amyloid-β peptide (Aβ) is the major protein component of the characteristic cerebral plaques of Alzheimers disease (AD), and a large body of evidence supports a pathogenic role for this peptide. Thus, the proteases β- and γ-secretase that are responsible for carving Ab out of its precursor protein are considered prime targets for therapeutic design. β-Secretase is a membrane-anchored aspartyl protease of the pepsin family, while γ-secretase is much more complex. γ-Secretase requires presenilin, a multipass membrane protein that is the site of dozens of missense mutations that alter Ab formation and cause hereditary AD. Two conserved aspartates in presenilin are required for γ-secretase activity, and aspartyl protease transition-state analogue inhibitors of γ-secretase bind directly to presenilins, strong evidence that presenilin is the catalytic component of a novel membrane aspartyl protease. γ-Secretase appears to be a multi-component complex of integral membrane proteins, and so far presenilin and a single-pass membrane protein called nicastrin have been identified as members of this complex. A closely similar or identical protease activity is essential for a signaling pathway critical for embryogenesis and hematopoiesis, raising concerns about γ-secretase as a target. The development of potent and selective inhibitors with good pharmacokinetic properties may soon address these concerns.