␤-Secretase (BACE1) is the rate-limiting protease for the generation of the amyloid ␤-peptide (A␤) in Alzheimer disease. Mice in which the bace1 gene is inactivated are reported to be healthy. However, the presence of a homologous gene encoding BACE2 raises the possibility of compensatory mechanisms. Therefore, we have generated bace1, bace2, and double knockout mice. We report here that BACE1 mice display a complex phenotype. A variable but significant number of BACE1 offspring died in the first weeks after birth. The surviving mice remained smaller than their littermate controls and presented a hyperactive behavior. Electrophysiologically, subtle alterations in the steady-state inactivation of voltage-gated sodium channels in BACE1-deficient neurons were observed. In contrast, bace2 knockout mice displayed an overall healthy phenotype. However, a combined deficiency of BACE2 and BACE1 enhanced the bace1 ؊/؊ lethality phenotype. At the biochemical level, we have confirmed that BACE1 deficiency results in an almost complete block of A␤ generation in neurons, but not in glia. As glia are 10 times more abundant in brain compared with neurons, our data indicate that BACE2 could indeed contribute to A␤ generation in the brains of Alzheimer disease and, in particular, Down syndrome patients. In conclusion, our data challenge the general idea of BACE1 as a safe drug target and call for some caution when claiming that no major side effects should be expected from blocking BACE1 activity. Alzheimer disease (AD)1 is the most common cause of dementia for which neither a good diagnostic test nor an effective treatment is available yet. The most widely accepted hypothesis states that AD is initially triggered by the abnormal accumulation and possibly deposition of the small amyloid ␤-peptide (A␤) in different brain regions, which in turn initiates a pathogenic cascade that ultimately leads to neuronal death, AD pathology, and dementia. A␤ is cleaved from a long membranebound precursor, the amyloid precursor protein (APP), by two consecutive cleavages. ␤-and ␥-secretases are the enzymes that liberate the N and C termini of A␤, respectively, and are the subject of intense investigation because of their relevance as candidate therapeutic targets to treat AD.BACE1 and BACE2 are two highly homologous membranebound aspartyl proteases that can process APP at the ␤-secretase site (1-8). Although both enzymes exhibit many of the characteristics expected for ␤-secretase, it has been quite convincingly demonstrated that BACE1 is in fact the major ␤-secretase responsible for A␤ generation in brain (9 -11). Contrary to BACE1, BACE2 is more highly expressed in peripheral tissues, but also to some extent in brain (2,8,12,13), raising the question of whether BACE2 could contribute to the generation of the brain A␤ pool. Both BACE1 and BACE2 can cleave APP in vitro not only at Asp 1 (numbering considering the first amino acid of A␤ as position 1), but also at internal sites within the A␤ region. BACE1 cleaves between amino acids 10 and 11 o...

The aspartyl protease BACE1 cleaves the amyloid precursor protein and the sialyltransferase ST6Gal I and is important in the pathogenesis of Alzheimer's disease. The normal function of BACE1 and additional physiological substrates have not been identified. Here we show that BACE1 acts on the P-selectin glycoprotein ligand 1 (PSGL-1), which mediates leukocyte adhesion in inflammatory reactions. In human monocytic U937 and human embryonic kidney 293 cells expressing endogenous or transfected BACE1, PSGL-1 was cleaved by BACE1 to generate a soluble ectodomain and a C-terminal transmembrane fragment. No evidence of the cleavage fragment was seen in primary cells derived from mice deficient in BACE1. By using deletion constructs and enzymatic deglycosylation of the C-terminal PSGL-1 fragments, the cleavage site in PSGL-1 was mapped to the juxtamembrane region within the ectodomain. In an in vitro assay BACE1 catalyzed the formation of the PSGL-1 products seen in vivo. The cleavage occurred at a Leu-Ser peptide bond as identified by mass spectrometry using a synthetic peptide. We conclude that PSGL-1 is an additional substrate for BACE1.

The amyloid peptide is the main constituent of the amyloid plaques in brain of Alzheimer's disease patients. This peptide is generated from the amyloid precursor protein by two consecutive cleavages. Cleavage at the N terminus is performed by the recently discovered ␤-secretase (Bace). This aspartyl protease contains a propeptide that has to be removed to obtain mature Bace. Furin and other members of the furin family of prohormone convertases are involved in this process. Surprisingly, ␤-secretase activity, neither at the classical Asp 1 position nor at the Glu 11 position of amyloid precursor protein, seems to be controlled by this maturation step. Furthermore, we show that Glu 11 cleavage is a function of the expression level of Bace, that it depends on the membrane anchorage of Bace, and that Asp 1 cleavage can be followed by Glu 11 cleavage. Our data suggest that pro-Bace could be active as a ␤-secretase in the early biosynthetic compartments of the cell and could be involved in the generation of the intracellular pool of the amyloid peptide. We conclude that modulation of the conversion of pro-Bace to mature Bace is not a relevant drug target to treat Alzheimer's disease.