Joint first authors Eric Schroeter and Ma Xenia Ilagan, tagged PS with two different epitopes to show that the protein dimerizes. Immunoprecipitating these chimeras with an antibody to either epitope also brought down the second chimera. Significantly, this coprecipitation only worked when the differently tagged proteins were expressed in the same hek293 cells, and not when lysates containing single chimeras were mixed. These experiments suggest that the dimers are formed and remain tightly bound in the cell.

The authors suggest that a dimer can explain how PS can catalyze cleavage in the middle and edge of the cell membrane, but emphasize that it may be necessary to have at least three aspartyl residues assembling at the dimer interface to form two catalytic sites. In support of this, Schroeter, Ilagan, and colleagues found that treating cells expressing PS with a photoaffinity label that interacts with the active site led to crosslinking of N-terminal and C-terminal ends of substrate molecules, as would be expected. However, they also found that N-terminal fragments were crosslinked together. The most likely explanation for the latter is that two catalytic sites exist in close proximity to each other.

Other evidence is also presented for a single catalytic site within γ-secretase. The authors found that a range of γ-secretase inhibitors, including MG132, calpeptin, and MD28170, have similar IC50s (the inhibitor concentration necessary to block half of the enzyme activity) for both Notch S3 cleavage (which occurs in the center of the membrane) and Aβ production (which occurs at the inner leaflet of the membrane). Furthermore, when the authors rank the potency of these compounds for inhibiting both cleavages, the orders are the same. And in a classic competition assay for proteolytic activity, Schroeter and Ilagan show that under saturating conditions, Notch and AβPP compete for cleavage, indicating they both sit in the same active site.

A PS dimer goes a long way to explain some of the perplexing observations on γ-secretase, the authors write. For example, with a dimer, allosteric effects may be evoked to explain why some inhibitors (ibuprofen, for one) can shift the preferred AβPP cleavage site to yield Aβ38 instead of Aβ42. Heterozygous dimers may also explain why some of the FAD mutations have partial activity.—Tom Fagan.

Reference:
Schroeter EH, Ilagan MX, Brunkan AL, Hecimovic S, Li YM, Xu M, Lewis HD, Saxena MT, De Strooper B, Coonrod A, Tomita T, Iwatsubo T, Moore CL, Goate A, Wolfe MS, Shearman M, Kopan R. A presenilin dimer at the core of the {gamma}-secretase enzyme: Insights from parallel analysis of Notch 1 and APP proteolysis. Proc Natl Acad Sci U S A. 2003 Oct 28;100(22):13075-13080. Epub 2003 Oct 17. Abstract