page 953). With too little of this membrane lipid, neurons produce more of the Alzheimer's-associated amyloid peptide (Aβ). The results are a warning that the effects of cholesterol-lowering drugs may need to be confined to the liver, where circulating sterols and fatty acids are made.
Aβ production begins when the APP precursor protein is cleaved by the BACE1 β-secretase. The new results show that cholesterol limits this cleavage by keeping BACE1 from its substrate. Most APP was found in cholesterol-free detergent-soluble domains, in which only ∼10% of the total APP was found in contact with BACE1. Much of the BACE1 was instead found in separate cholesterol-rich rafts.
In cultured neurons treated with anti-cholesterol drugs, twice as much APP colocalized with BACE1, and the cells consequently secreted more Aβ. Similarly, more BACE1 leaked into nonraft, lower cholesterol regions in neurons from Alzheimer's patients.
Clinical studies correlated the use of statins, which reduce primarily circulating cholesterol, with a lower, not higher, risk of Alzheimer's. The authors suggest that this benefit may be a side effect of the cleared vessels' improved ability to oxygenate brain tissue.
Brain activity requires constant cholesterol levels in axons and the insulating myelin. Neurons therefore synthesize their own cholesterol, and the blood–brain barrier prevents its uptake from the blood stream. Others have shown that synaptic activity and remodeling is impaired by drug-induced cholesterol loss. Based on these results and their own, Abad-Rodriguez et al. warn that current efforts to reduce brain cholesterol with statins that cross the blood–brain barrier are unwarranted and may have dangerous consequences.