13 Apr 2007

At the 8th International Conference AD/PD, held last month in Salzburg, Austria, Dora Kovacs of the Massachusetts General Hospital presented new data advancing her story of ACAT inhibition as a potential new therapeutic approach to Alzheimer disease. Following an earlier report that the pharmacological compound (read: not a real human drug) CP113,818 reduces amyloid deposition in transgenic mice (Hutter-Paier et al., 2004), many observers have been curious where this approach would go next. In Salzburg, Kovacs reported that the compound avasimibe (a real drug) performs almost as well, validating the general approach.

Avasimibe is a small-molecule inhibitor of the enzyme ACAT-1. It counteracts the formation of cholesteryl esters from free cholesterol and fatty acids. The esters accumulate in the foam cells seen in atherosclerosis. Avasimibe had made it to phase 3 trials for cardiovascular disease, but the pharmaceutical giant Pfizer discontinued its development a few years ago. Internal drug interaction research apparently showed that avasimibe weakened the effect of lipitor, Pfizer’s leading cardiovascular drug, when it had instead been meant to boost lipitor and be taken alongside it (see Forbes.com; Sahi et al., 2003). Avasimibe also underwhelmed in the clinic (Tardif et al., 2004), and onto the dung heap it went.

AD aficionados will know that none other than Ronald Reagan was famous for quipping: “There might be a pony in there” when faced with bad news. He might have invoked his joke here, too, for when Kovacs tested two doses of avasimibe in APP/PS transgenic mice and wild-type mice, she found that the drug favorably changed amyloid endpoints. Working with scientists at the Austrian CRO JSW Research in Graz, Kovacs found that two months of treatment with 14.4 mg/kg of avasimibe per day reduced both serum and brain cholesteryl ester levels in the mice. In six-month old mice, it also reduced the brain’s plaque load by two thirds. To better model AD, where patients already have high plaque loads before they begin therapy, the scientists treated four 14 month-old transgenic mice whose brains were laden with amyloid. Avasimibe reduced diffuse deposits labeled by the antibody 6E10 but not core plaques stained with thioflavin S. Kovacs noted that she interprets her data as suggesting that the ACAT inhibitor interferes with Aβ production, allowing endogenous clearance mechanisms to remove diffuse deposits and reduce the net burden of Aβ. Kovacs’ lab attacked ACAT-1 by a third method, this time using siRNA in cultured cells. This, too, reduced total Aβ, Aβ42, and APP CTF generation. These experiments showed that a modest halving of ACAT-1 protein was sufficient to drive down cholesteryl ester levels by a fifth and Aβ secretion by 40 percent (Huttunen et al., 2007).

The gene for ACAT currently occupies rank 12 on the list of Top Alzgene Results, mainly thanks to a study that reported a common protective polymorphism associated with low brain amyloid load (Wollmer et al., 2003). How might it work? To get at the mechanistic underpinning of APP processing by ACAT-1, Kovacs, Henri Huttunen, and colleagues used the inhibitors as tools to find proteins that interact with APP in a way that is sensitive to ACAT inhibition. Once explored more deeply, this approach could reveal new pathways of APP processing. In Salzburg, Kovacs reported initial data that the ER chaperone GRP94, and the chaperone/protease HtrA2/Omi appear to bind APP in this way. Both proteins respond to ER stress, and both normally function somehow in the maturation and degradation of APP in the ER. While HtrA2 is primarily known as a mitochondrial protein, Kovacs and colleagues showed in Salzburg that it also occurs on the cytosolic side of the ER membrane and participates in ERAD (see related Salzburg story). This work is pointing to the endoplasmic reticulum as the cellular site at which ACAT inhibitors can influence the fate of APP.—Gabrielle Strobel.