22 Jul 2002

Stockholm. Injecting a fresh set of data into the debate about what is toxic and why-fibrillar amyloid or early, oligomeric forms of Aβ-scientists at the 8th International Conference on Alzheimer’s Disease and Related Disorders today presented their work on clusterin, an Aβ binding protein also called ApoJ. The study appears today in the Proceedings of the National Academy of Sciences.

Ron DeMattos, David Holtzman and colleagues of Washington University, St. Louis, with others at Eli Lilly and Co. and at University of Cincinnati College of Medicine, took earlier in vitro work on interactions between clusterin and Aβ in vivo by breeding PDAPP mice with a clusterin knockout strain.

They found that mice overproducing Aβ but lacking clusterin accumulated and deposited as much amyloid as did mice that had clusterin. However, the amyloid was different. The clusterin-free PDAPP mice had fewer fibrillar deposits and altered pools of soluble Aβ. At the same time, they showed markedly less neuritic dystrophy around the sites of amyloid deposition in their brains. This finding uncouples amyloid deposition from neurotoxicity and points a finger at an intriguing protein that modifies the formation of toxic amyloid species. ApoE and ApoJ/clusterin are the two most abundant apolipoproteins in the brain.

With clusterin, a larger proportion of Aβ converted to thioflavine-S positive fibrillar amyloid than without clusterin. How could it work? This remains unclear, but DeMattos et al. suggest that clusterin might promote the formation of toxic oligomeric forms of Aβ, as previous in vitro studies have suggested.

It is unclear how clusterin’s observed effect of changing soluble Aβ pools in vivo relates to its other observed effect of modifying fibrillization in a way that increases toxicity. An emerging literature is beginning to describe clusterin as a secreted chaperone, which can solubilize different proteins that have in common that they expose hydrophobic sites. Perhaps chaperone-like interactions between clusterin and Aβ change the equilibrium between soluble and deposited Aβ, the authors speculate. This could unmask epitopes that increase toxicity to neurites, either on soluble Aβ or on fibrillar amyloid.—Gabrielle Strobel