This is a very interesting study from a leading cell biology lab that tries to define the site of Aβ generation. Their approach of knocking down different components of trafficking steps in endosome and Golgi pathways is very nice. The authors provide more evidence in favor of the importance of APP internalization from the plasma membrane in Aβ generation, here using HEK cells transfected with APP. Interestingly, knockdown of early components of the ESCRT pathway led to APP accumulation in organelles that label for an early endocytic marker and also caused reduced Aβ40 secretion. These new data suggest that Aβ40 generation might occur distal to early endosomes.

Intriguingly, knockdown of different components involved in ESCRTIII, thought to be part of late endosomes/multivesicular bodies (based on an electron microscopy definition of more than one inner vesicle), elevated Aβ40 secretion. The authors describe that these knockdowns elevate APP localization to the trans-Golgi network (TGN), although looking at these images, we see that CHMP6 and VPS4 knockdowns give markedly...  Read more

This is a very interesting study from a leading cell biology lab that tries to define the site of Aβ generation. Their approach of knocking down different components of trafficking steps in endosome and Golgi pathways is very nice. The authors provide more evidence in favor of the importance of APP internalization from the plasma membrane in Aβ generation, here using HEK cells transfected with APP. Interestingly, knockdown of early components of the ESCRT pathway led to APP accumulation in organelles that label for an early endocytic marker and also caused reduced Aβ40 secretion. These new data suggest that Aβ40 generation might occur distal to early endosomes.

Intriguingly, knockdown of different components involved in ESCRTIII, thought to be part of late endosomes/multivesicular bodies (based on an electron microscopy definition of more than one inner vesicle), elevated Aβ40 secretion. The authors describe that these knockdowns elevate APP localization to the trans-Golgi network (TGN), although looking at these images, we see that CHMP6 and VPS4 knockdowns give markedly different APP labeling patterns. The authors hypothesize that these different knockdowns in ESCRTIII-related components lead to a rerouting of APP to the TGN. Knockdown of Rab9A provided some support for this conclusion. Remarkably, VPS35/retromer knockdown reduced Aβ40 secretion, which goes against the consensus that has been forming in the retromer/SorLA/APP field.

Caveats to this study include that it would have been nice to study APP processing more thoroughly rather than only with the readout of Aβ40 secretion by ELISA and APP localization by immunofluorescence using a C-terminal APP antibody. For example, changes in Aβ40 and 42 do not always mirror each other, and changes in secreted Aβ do not necessarily follow intracellular Aβ. It is also important to study primary neurons; for example, both BACE and presenilin components have been localized to endosomes in distal processes and synaptic compartments. For excellent recent reviews on APP trafficking, see Haass et al., 2012, and Rajendran and Annaert, 2012.

View all comments by Gunnar K. Gouras