Cyclohexanehexol inhibitors of Ab aggregation prevent and reverse Alzheimer phenotype in a mouse model JoAnne McLaurin 1,2 , Meredith E Kierstead 1,2 , Mary E Brown 1 , Cheryl A Hawkes 1 , Mark H L Lambermon 1 , Amie L Phinney 1 , Audrey A Darabie 1 , Julian E Cousins 1 , Janet E French 1 , Melissa F Lan 1 , Fusheng Chen 1 , Sydney S N Wong 1 , Howard T J Mount 1,3 , Paul E Fraser 1,4 , David Westaway 1,2 & Peter St George-Hyslop 1,3 When given orally to a transgenic mouse model of Alzheimer disease, cyclohexanehexol stereoisomers inhibit aggregation of amyloid b peptide (Ab) into high-molecular-weight oligomers in the brain and ameliorate several Alzheimer disease–like phenotypes in these mice, including impaired cognition, altered synaptic physiology, cerebral Ab pathology and accelerated mortality. These therapeutic effects, which occur regardless of whether the compounds are given before or well after the onset of the Alzheimer disease–like phenotype, support the idea that the accumulation of Ab oligomers has a central role in the pathogenesis of Alzheimer disease. Multiple lines of evidence suggest that the accumulation of neurotoxic oligomeric aggregates of Ab may be a central event in the pathogenesis of Alzheimer disease 1,2 . If correct, this hypothesis predicts that inhibitors of Ab aggregation and toxicity may be effective in blocking this pathogenic cascade. We report here that orally administered cyclohexanehexol stereoisomers can block the accumulation of Ab oligomers in a dose-dependent manner, and reduce Alzheimer disease–like behavioral deficits, Alzheimer disease–like neuropatho- logy and accelerated mortality in a transgenic mouse model of Alzheimer disease. These in vivo experimental results both strongly support the notion that accumulation of Ab oligomers has a sig- nificant role in the pathogenesis of Alzheimer disease and suggest that inhibition of Ab aggregation may be a useful therapy for this disease. We previously showed that Ab oligomerization and fibril formation are strongly facilitated by phosphatidylinositol lipids 3–8 . Based on these observations, we hypothesized that derivatives of these phos- phatidylinositol moieties, particularly cyclohexanehexol stereoisomers (which also act as osmolyte modulators of protein folding 9 ), might compete with intact phosphatidylinositol for binding to Ab and be used to interfere with Ab fibril assembly. We recently showed that, in vitro, cyclohexanehexol stereoisomers can inhibit Ab fibril assembly, accelerate disassembly of preformed fibrils, preferentially stabilize Ab into nontoxic b-structured spherical micelle conformers and protect primary cultured neurons from Ab oligomer–induced toxicity 10,11 . We have also shown that these compounds have stereoisomer-specific differences in inhibition of Ab aggregation and cytotoxicity: scyllo-cyclohexanehexol and epi-cyclohexanehexol are more effective than myo-cyclohexanehexol 11 . To assess their effectiveness in vivo, we administered these com- pounds to a robust transgenic mouse model of Alzheimer disease 12 (TgCRND8). This model expresses a human amyloid precursor protein transgene (APP 695 ) bearing missense mutations that cause Alzheimer disease in humans (K670N, M671L and V717F). At about 3 months of age, these mice have progressive spatial learning deficits that are accompanied by rising cerebral Ab levels and by increasing numbers of cerebral amyloid plaques 12 . By 6 months of age, the levels of Ab and the morphology, density and distribution of amyloid plaques are similar to those seen in brains of people with well- established Alzheimer disease 12 . As in humans with Alzheimer disease, these biochemical, behavioral and neuropathological phenotypes are accompanied by accelerated mortality 12,13 . TgCRND8 mice and nontransgenic littermates were assigned to sex- and age-matched cohorts that were used to test the effectiveness of cyclohexanehexol stereoisomers in two different treatment paradigms. In the first paradigm, cyclohexanehexols were orally administered prophylactically, with treatment beginning at 6 weeks of age (that is, about 6 weeks before onset of phenotype) and continuing until either 4 or 6 months of age. In the second paradigm, compounds were given therapeutically beginning at 5 months of age (when the Alzheimer disease–like phenotype is already well established), and continuing until 6 months of age. Within each of these experimental arms, mice were randomly assigned to receive active compound (1,2,3,4,5/6- (epi) cyclohexanehexol or 1,3,5/2,4,6- (scyllo-) cyclohexanehexol), mock therapy (mannitol, a sugar of similar molecular weight) or no therapy. The endpoints of these studies were: cognitive function (as measured by spatial reference learning in the Morris water maze test 13,14 ), brain Ab levels, neuropathology and mortality. We did not Received 17 October 2005; accepted 26 April 2006; published online 11 June 2006; corrected online 19 June 2006 (details online); doi:10.1038/nm1423 1 Centre for Research in Neurodegenerative Diseases. 2 Department of Laboratory Medicine and Pathobiology. 3 Department of Medicine, and University Health Network, Toronto Western Hospital Research Institute and 4 Department of Medical Biophysics, University of Toronto, 6 Queen’s Park Crescent West, Toronto, Ontario M5S 3H2 Canada. Correspondence should be addressed to J.M. (j.mclaurin@utoronto.ca). NATURE MEDICINE VOLUME 12 [ NUMBER 7 [ JULY 2006 801 ARTICLES © 2006 Nature Publishing Group http://www.nature.com/naturemedicine