I recommend this paper, which uses clear-cut cell biological methods to provide evidence that γ-secretase activity occurs at or after APP delivery to the plasma membrane (PM) and/or early endosomes. As the authors point out, they used a “simple approach,” which in a way just strengthens their work. They employ well-established means to block trafficking, between ER to Golgi, within Golgi, Golgi to PM, and endocytosis from PM, in conjunction with tracking the cellular distribution of GFP-APP C99. High-resolution fluorescent microscopy provides visualization of cellular trafficking that cannot be matched by biochemical assays.

Their results support previous work from many other investigators, especially that from Eddie Koo and colleagues. Among indirect support, also in neurons of the brain that are known to transport APP to their synapses, is work from Jie Shen, Carlos Saura, and colleagues demonstrating APP CTF accumulation in synaptic compartments in PS conditional knockout mice (Saura et al., 2005).

Kaether et al. provide...  Read more

I recommend this paper, which uses clear-cut cell biological methods to provide evidence that γ-secretase activity occurs at or after APP delivery to the plasma membrane (PM) and/or early endosomes. As the authors point out, they used a “simple approach,” which in a way just strengthens their work. They employ well-established means to block trafficking, between ER to Golgi, within Golgi, Golgi to PM, and endocytosis from PM, in conjunction with tracking the cellular distribution of GFP-APP C99. High-resolution fluorescent microscopy provides visualization of cellular trafficking that cannot be matched by biochemical assays.

Their results support previous work from many other investigators, especially that from Eddie Koo and colleagues. Among indirect support, also in neurons of the brain that are known to transport APP to their synapses, is work from Jie Shen, Carlos Saura, and colleagues demonstrating APP CTF accumulation in synaptic compartments in PS conditional knockout mice (Saura et al., 2005).

Kaether et al. provide substantial evidence for the importance of ER-to-PM trafficking of their APP CTF construct prior to AICD formation, using several different cell lines and employing Western blot controls. Less developed in this excellent paper are the final data suggesting that endocytosis from the PM is not required for γ-secretase activity. Here only one cell line is used, although which one is not indicated, and no biochemical support is provided for the conclusions that appear to suggest that γ-secretase does not require endocytosis. This is an area that has seen differing results; a recent study using this dynaminK44A dominant negative construct to block endocytosis provided evidence for markedly reduced Aβ secretion after blocking endocytosis, although this effect appears to be cell type-specific (Carey et al., 2005). Nevertheless, Kaether et al. conclude that γ-secretase activity is localized at the PM "and/or early endosomes.” The authors were wise not to preclude the possibility for some γ-secretase activity that might have been overlooked in other compartments using their current approach.

In general, diverse data appear to support that endosomes are major sites for both β- and γ-secretase cleavages; see also an excellent recent review on APP processing by (Vetrivel and Thinakaran, 2006).

View all comments by Gunnar K. Gouras

This paper gives an answer to the long asked question regarding where in cells APP and Notch processing takes place. The next question to be answered is wether the same applies to the processes in neurons.

View all comments by Takaomi Saido