The biology of the presenilin complexes

Abstract

APP proteolytic processing Alzheimer's Disease (AD) is characterized by the deposition of two kinds of abnormal protein aggregates, senile plaques and neurofibrillary tangles, and by neuronal dysfunction and cell loss in the brain. Senile plaques are primarily composed of extracellular deposits of hydrophobic 37-43 amino acid Aβ peptides. Aβ peptides are derived by successive enzymatic cleavages of the type I membrane protein, β-amyloid precursor protein (APP) (Haass and Selkoe 1993). APP is first cleaved close to the membrane in the extracellular domain by either α-or β-secretase, resulting in a release of soluble APP ectodomains, and residual membrane-tethered C-terminal protein stubs, termed C83 or C99, respectively (The numbers indicate the length of each carboxylterminal fragment). C83 and C99 are substrates for γ-secretase, an activity that generates p3 and Aβ peptides, respectively. γ-Secretase processes substrates at different positions within the membrane domain and thus, both Aβ and p3 have "ragged" termini. Aβ has been best studied in this regard and species between 37 and 43 amino acid residues have been identified. γ-Secretase cleavage of APP also releases the intracellular carboxy-terminal "APP intracellular domain" or "AICD". The function of both Aβ and AICD is the subject of intense investigations. Because Aβ42 is the primary constituent of the amyloid fibrils deposited in the AD brains, and mutations in APP and presenilin enhance the production of this peptide, γ-secretase cleavage of APP is a pivotal step in AD pathogenesis. It is striking that this proteolytic reaction occurs within the highly hydrophobic environment of the membrane. Identification of presenilin Genetic studies in familial AD (FAD) cases have identified disease-linked mutations in three genes that contribute to AD. The first pathogenic mutations in early-onset FAD families were found in the APP gene on chromosome 21 (Chartier-Harlin et al. 1991; Goate et al. 1991; Murrell et al. 1991). However, subsequent studies indicated that mutations in APP account only for a small fraction of FAD cases. Several genetic studies indicated a major locus for FAD on chromosome 14 in early onset autosomal dominant AD, and in 1995, the Presenilin1 (PS1) gene on chromosome 14 (14q24.3) was identified by positional cloning (Sherrington et al. 1995). Shortly thereafter, it was shown that mutations in the closely related PS2 gene on chromosome 1 (1q42.2) could cause FAD as well (Levy-Lahad et al. 1995; Rogaev et al. 1995). Studies in transgenic mice (Borchelt et al. 1996; Duff et al. 1996) and cultured cells (Citron et al. 1997; Scheuner et al. 1996; Tomita et al. 1997) have revealed that expression of FAD-linked PS variants elevates Aβ42/Aβ40 ratios. Moreover, transgenic mice that co-express FAD-mutant PS1 and APP develop amyloid plaques much earlier than age-matched mutant APP mice (Borchelt et al. 1997). Therefore, PS mutations cause a change in the Aβ42/40 ratio, but whether PS is directly involved in γ-secretase processing of APP was unclear. However, in PS-deficient neurons and fibroblasts, APP processing was greatly impaired, leading to the accumulation of the C83 and C99 APP fragments, the direct substrates of γ-secretase, and inhibition of Aβ (and p3) generation (De Strooper et al. 1998; Xia et al. 1998). Thus, PS are directly required for γ-secretase cleavage of APP. Overall, the findings imply that mutations in the substrate (APP) or in the proteolytic machinery (PS) result in similar changes in Aβ42 generation (Scheuner et al. 1996). This provides very strong support for the "amyloid cascade hypothesis". © 2007 Springer Science+Business Media, LLC. All rights reserved.