22 January 2009. The precise mechanism and significance of APP ε cleavage has been debated since it was discovered some 10 years ago (see ARF related news story). One of the mysteries is that in theory, ε cleavage yields Aβ peptides of 49 amino acids, but in practice, Aβ49 is extremely rare. A paper in this week’s PNAS suggests why this might be. Using solid-state NMR to analyze APP structure, researchers led by Steven Smith at Stony Brook University, New York, confirm that APP exists as a dimer (see Scheuermann et al., 2001) and that it is held together by transmembrane helices. The structure suggests that a transition from helix to random coil right at the edge of the inner cytoplasmic membrane weakens the dimer and allows cleavage by γ-secretase at the ε position, which lies at the polar head of the inner membrane (see figure below).
Structural Data Supports Sequential Cleavage Model of APP Processing
NMR data indicates that APP exists as a dimer held, in part, by interactions between transmembrane helices. Helix-to-random coil transitions at the cytoplasmic side of the domains weakens binding sufficiently to allow ε cleavage by γ-secretase. That cleavage unravels the helical coils further, exposing the APP γ site to proteolysis. Extending the helices by adding three leucines prevents not only ε cleavage, but γ cleavage as well. Image credit: National Academy of Sciences
To test this idea, the researchers extended the helical dimer by adding three leucine residues. This not only eliminated the helix-to-random coil transition near the ε position, but also blocked ε and γ cleavage. The data support the sequential model for γ-secretase action proposed by Yasuo Ihara (see Gu et al., 2001). It posits that APP is first cleaved at the ε position, then at the γ position. According to Smith’s model, that first cleavage would weaken the helices, the protein would become locally unstructured, and the γ-site would become subject to proteolysis. Aβ49 would therefore be lost to Aβ40/42 as the protein is cleaved further.—Tom Fagan.
Sato T, Tanga T-C, Reubinsa G, Feia JZ, Fujimotob T, Kienlen-Campardc P, Constantinescud SN, Octavec J-N, Aimotob S, Smith SO. A helix-to-coil transition at the epsilon-cut site in the transmembrane dimer of the amyloid precursor protein is required for proteolysis. PNAS. 2009.