An Alzheimer's Tail

Amyloid protein remnant might regulate genes

Mitch Leslie

Key Words: N-CoR • NF-B • APP • AICD

Abstract: Scientists are buzzing about the -amyloid precursor protein (APP) because its midsection appears in the brain plaques of Alzheimer's disease. Although scientists have swarmed over this sticky protein fragment, a different one might also deserve attention. A new study reveals that the tail chopped from APP helps turn on a gene. The results uncover a function for the ephemeral tail in normal cells and suggest that some potential anti-Alzheimer's drugs might rob the brain of a necessary control molecule.

The -amyloid protein that gunks up the brains of Alzheimer's disease patients derives from APP. To release amyloid, enzymes snip off APP's head and a short tail known as AICD. A molecular ghost, AICD breaks down so quickly that scientists have never isolated it from brain cells--and only with great difficulty from other cell types, says molecular neurobiologist Sam Sisodia of the University of Chicago. Last year, researchers caught the first glimpse of AICD at work; it linked with two proteins to flick on a gene the researchers had inserted into mammalian cells. But no one had shown that the tail activates anything but an artificial gene.

Baek and colleagues were studying how cells control a protein corepressor that straddles a gene called KAI1 and helps shut it down. In cell culture experiments, they noticed that the AICD-carrying trio could muscle the corepressor away, presumably activating the gene. Mutating the tail of APP so that it can't link to its partners prevented the AICD team from alighting on KAI1, as did blocking the enzyme that releases AICD from APP. Together, these findings indicate that AICD enables the group to fasten to the gene. Mice that pump out abnormally large amounts of APP produce extra KAI1 protein and messenger RNA, which suggests that the AICD-containing crew turns on KAI1 in whole organisms.

Sisodia describes the work as a good start toward pinning a function on AICD. However, he wants stronger evidence linking the tail to KAI1 activation; no one has yet measured whether the extra APP in the mice that also manufacture extra KAI1 translates into additional AICD, for example. He also warns that researchers who try to implicate AICD in Alzheimer's disease will be chasing their tails. If the AICD-containing conglomerate activates genes, rising concentrations of the tail could cause potentially hazardous overproduction of some proteins, although no current evidence supports that idea, according to Sisodia and the paper's authors. But to get more AICD, APP production would also have to mount, and that doesn't happen in Alzheimer's patients, Sisodia says; although they accumulate amyloid, amounts of the precursor protein don't change. Edward Koo, a neuroscientist on the research team, says the results raise questions about new drugs that diminish the amount of amyloid in the brain by blocking one of the enzymes that cleaves APP. If AICD turns out to be part of an essential gene switch, this strategy could backfire. "It's a potential warning that if you inhibit AICD signaling, you may get unwanted side effects," he says. Only further work will reveal whether this tail packs a sting.

--Mitch Leslie

C. S. H. Baek, K. A. Oghi, D. W. Rose, E. H. Koo, C. K. Glass, M. G. Rosenfeld, Exchange of N-CoR corepressor and Tip60 coactivator complexes links gene expression by NF-B and -amyloid precursor protein. Cell 110, 56-67 (2002). [Abstract] [Full text]

Citation: M. Leslie, An Alzheimer's Tail. Science's SAGE KE (17 July 2002), http://sageke.sciencemag.org/cgi/content/abstract/sageke;2002/28/nw96