The AD community’s overwhelming focus on β and γ-secretase has all but obscured a third enzyme that clips APP in an alternative, altogether more benign pathway. This ugly duckling is, of course, α-secretase. It has been snubbed partly because, unlike both β- and γ-secretase, its unique identity has remained stubbornly elusive, and partly because conventional wisdom holds that blocking a target enzyme is easier than boosting it, which is what a therapeutic strategy based on α-secretase would have to do. Despite all this, α-secretase fairly looks the part of a swan in a paper in the current issue of the Journal of Clinical Investigation.

Rolf Postina and Falk Fahrenholz at the University of Mainz in Germany, with colleagues there and in Belgium, present the first in-vivo evidence that manipulating the activity of one of the three major physiological α-secretase candidates—the disintegrin and metalloproteinase ADAM-10—can improve standard AD markers in a mouse model. In a clever experiment, the authors made mice that mildly over-express ADAM-10 in neurons and then bred them to transgenic mice that over-express, also in neurons, the APPV717I London mutation. They found that the double-transgenic progeny not only overproduced the requisite APPsα fragment (which is known to be neurotrophic), but also escaped the amyloid deposition and memory deficits that mark the APPV717I single-transgenic mice. Aβ reduction was moderately reduced in the double-transgenics.

This is interesting for several reasons, not least because the subtle ADAM-10 overexpression models the action of any future anti-amyloid drug on a secretase target more realistically than do more standard BACE or γ-secretase knockout experiments. The Postina et al. paper, along with a commentary by Stephan Lichtenthaler and Christian Haass discussing the therapeutic potential of promoting α-secretase cleavage, both are available in full online.—Gabrielle Strobel


  1. In this study, the authors provide the first in-
    vivo demonstration of an effect on plaque formation and neuronal function by stimulating the α-secretase pathway involved in APP processing. Presently, there are two known pathways by which APP is processed: The "bad" pathway involving the β- and γ-secretases, whose activation leads to over-accumulation of the β-amyloid peptide along with plaque formation, and the "good" pathway involving α-secretase, whose activation prevents β-amyloid from forming. Currently, all major therapeutic strategies involve inhibiting the β- or γ-secretases; however, an attractive alternative would be to stimulate the α-secretase pathways. Stimulation of this pathway may be a good alternative for several reasons, one of which is that inhibiting β-secretase activity may alter Notch signaling, possibly leading to untoward side effects. In this new study, Postina et al., demonstrate that by moderately overexpressing α-secretase, they prevented to a large extent plaque formation in a mouse model of Alzheimer's disease. In addition, the authors show that overexpression of this specific α-secretase (ADAM10) reverses impaired LTP and cognitive deficits in this animal model. This study suggests that pharmacological strategies designed to stimulate the α-secretase pathway may be beneficial in treating AD, and refutes the old saying that "good guys" always finish last.

  2. This group of German and Belgian investigators has produced what is apparently the first evidence that a member of the ADAM family of proteases functions as an alpha secretase for the beta amyloid precursor protein substrate in vivo. An attractive feature of the experimental approach was the evaluation of the impacts of different degrees of overexpression of the protease. A dominant negative construct was also employed. In crosses with APP V717I transgenic mice, morphological and functional studies demonstrated clear evidence of a protective effect of moderate overexpression of ADAM10. Enhanced production of neuroprotective APPsalpha and decreased levels of beta amyloid peptides were associated with such moderate overexpression. The results provide strong rationales for the further development of clinical interventions based upon enhancements of such alpha secretase activities. Such interventions will have to hit upon an appropriate level of enhancement, as we can assume that other products of APP processing, including beta amyloid peptides, have physiological functions that also require optimal concentrations.

    View all comments by George Martin

Make a Comment

To make a comment you must login or register.


Other Citations

  1. APPV717I London mutation

External Citations

  1. Postina et al. paper
  2. commentary

Further Reading

No Available Further Reading

Primary Papers

  1. . A disintegrin-metalloproteinase prevents amyloid plaque formation and hippocampal defects in an Alzheimer disease mouse model. J Clin Invest. 2004 May;113(10):1456-64. PubMed.