PS1 was the second familial EOAD gene to be identified. Sherrington et al (Sherrington et al 1995 identified five missense mutations from a novel gene that segregated within AD families. This gene was later called presenilin-1 and PS1 mutations account for the majority of familial EOAD forms, with estimates ranging from 18% to 50% (Campion et al 1995 Cruts et al 1997). PS1 mutations increase the amount of Aβ42 that is produced and these mutations are the most aggressive observed, with the AAO for PS1 patients normally within the fourth decade. A third locus for familial AD was localised to chromosome 1 by studying Volga-German (VG) kindreds. Further analysis identified a gene with sequence homology to PS1 and missense mutations in this homolog, PS2, were shown to segregate within the VG families (Levy-Lehad et al 1995 Rogaev et al 1995 PS2 mutations are found in relatively few EOAD families compared to PS1, although this may because individuals with PS2 mutations have a variable age of onset with many individuals developing AD at an age consistent with LOAD and hence these families may have been overlooked (Cruts et al 1997).
More than 100 mutations have been observed within PS1 and 6 from PS2. These mutations are mainly found within the TM regions of the proteins and in exon 8. The TM domains of the PSs are believed to be α-helical (Li and Greenwald 1998 and it has recently been observed that most pathogenic mutations appear to align along the helical faces (Hardy and Crook 2001 The exceptions to this are the mutations found in TM7 and TM8, but it is not entirely clear at present whether these domains form a conventional TM domain.
To date, only missense mutations or inframe deletions in the PSs have been associated with AD. While all PS1 mutations tested shift the ratio of Aβ40:Aβ42 to specifically favour Aβ42 in cultured cell models, constant Aβ40 levels are maintained (Janowsky et al 2004 whereas, PS2 mutations decrease Aβ40 levels suggesting partial loss-of-function (LOF) (Walker et al 2005). PS mutations are more aggressive with an earlier AAO than APP mutations, which supports PS loss-of-function (LOF). However, recent work has shown that PS1 mutations may alter PS1 function by placing a greater amount of C83/C99 substrate in contact with a γ-secretase that preferentially cleaves at residue 42 (Jankowsky et al 2004a 2004b PS1 is the major PS involved in γ-secretase activity (Herreman et al 1999 indicating that mutations in PS2 must have dramatic functional effects to have an observable affect in the presence of normal PS1. Indeed, 4 FAD PS2 mutations were shown to substantially change the Aβ40/42 ratio comparable with PS1 mutations (Walker et al 2005).
In addition to pathogenic coding mutations, PS1 promoter polymorphisms affect disease risk. Expression levels of PS1 modulate Aβ production and the C allele of the promoter polymorphism -22 C/T (also known as -48 C/T) increases risk for complex EOAD (Van Duijn et al 1999 Theuns et al 2000 Lambert et al 2001 This polymorphism has been shown to correlate with an increase in both Aβ40 and Aβ42 deposition (Lambert et al 2001). However, recent work observed a decrease in PS1 expression associated with the C allele (Theuns et al 2003
PS mutations are proposed to align along the α-helical faces of the PS protein and these α-helical TM domains of PS are proposed to encircle the C terminal stub of APP (Hardy and Crook 2001 The most severe of the PS1 mutations appear to alter conformation of the γ-secretase active site (Kornilova et al 2005 although to fully understand the mechanism by which PS mutations affect APP processing, whether it be LOF or gain-of-function (GOF), crystallisation of PS and PS with its APP substrate will be required.
See also the Presenilin mutations directory on the Alzheimer Research Forum website
Click to read this topic
6/6/2012 2:43:17 PM - 188.8.131.52