01 Aug 2002

In a wide-ranging and eloquent presentation at the 8th International Conference on Alzheimer’s Disease and Related Disorders held last week in Stockholm, Christian Haass (Abstract 1059) provided several experimental proofs that a small amount of presenilin is detected at the plasma membrane. Using a relatively new technique, total internal reflection microscopy (TIRM) Haass demonstrated that exogenous GFP-tagged PS could readily be detected in the endoplasmic reticulum and also on the plasma membrane. Furthermore, he was able to biotinylate mature Nicastrin on the plasma membrane and then in co-immunoprecipitation experiments he detected endogenous PS heterodimers associated with the biotinylated Nicastrin, thus indicating that two essential components of the γ-secretase complex, namely PS and Nicastrin, are detected together at the cell surface. Based on these experiments, Haass estimated that approximately one-thirtieth of total cellular PS was localized to the plasma membrane.

Haass also presented data demonstrating that Notch and CD44 are cleaved in a PS-dependent manner at a γ-site in the middle of the transmembrane domain at a position equivalent to cleavage after residue 40 of the A β sequence. This cleavage, which he termed site 4 (S4), gives rise to a p3-like, soluble peptide of both Notch and CD44. In addition, he also reported on the well-characterized site 3 (S3) cleavage of Notch detecting a secreted peptide Mass ~4358 consistent with proteolysis just 3 amino acid residues into the transmembrane sequence.

Finally, Haass reviewed data that AβPP was cleaved at a position similar to the Notch S3, aka the ε site, and that familial Alzheimer's disease mutations in PS cause an increase in γ-cleaved APP and Notch and a decrease in ε cleavage. This latter point was corroborated by another speaker, Peter St. George-Hyslop and further suggests that increased production of Aβ is the initiating event in AD pathogenesis as opposed to transcriptional deregulation mediated by an increase in functional AICD generation.

From the data presented it is not possible to determine whether cleavage at the γ and ε sites occur sequentially or simultaneously. In addition to the products discussed above, APPC59, AAβ49 and the 9 amino acid peptide spanning the γ and ε cleavage sites have yet to be detected. Only when these fragments and their temporal production is uncovered will it be possible to determine the order of the γ and ε cleavages.—Dominic Walsh