Our observation that squalestatin caused only small differen

Our observation that squalestatin caused only small differences in the production of Aβ40 and Aβ42 was in contrast to prior reports of the effects of other cholesterol synthesis inhibitors. This may be due to using squalestatin, a more specific cholesterol synthesis inhibitor that does not affect isoprenoid function [22], rather than conventional “statins” used in other reports. Furthermore, the reduction in cellular cholesterol concentrations in these studies was mild when compared to other studies [19], [37]. The key observation, that CM from squalestatin-treated saquinavir contained Aβ42 but did not cause synapse damage, showed that measuring concentrations of Aβ alone was a poor indicator of biological activity. That squalestatin selectively affected the production of soluble Aβ oligomers, thought to be major causes of synapse damage in AD [10], [16], [38], is consistent with a report that reduced cholesterol synthesis did not affect Aβ production but did extend lifespan in a mouse model of AD [39]. Cholesterol affects the formation and function of lipid rafts [40], [41] in which APP and many of the enzymes involved in the generation of Aβ are found [42], [43], [44]. Consequently, it seems likely that the squalestatin-induced reduction in cholesterol concentrations would affect the processing of APP to Aβ oligomers. Cholesterol concentrations within the ER were increased after activation of cPLA2[25]. This enzyme affects tubule formation and consequently intracellular trafficking of proteins [31], specifically within the trans-golgi network [45], a key area of APP metabolism [46]. Here we show that inhibition of cPLA2 in 7PA2 cells resulted in CM that did not cause synapse damage. Activation of cPLA2 leads to the production of PAF and PAF antagonists also reduced the release of Aβ oligomers from 7PA2 cells. As the addition of PAF reversed the effects of cPLA2 inhibitors we concluded that it was the key mediator involved in this pathway. 7PA2 cells treated with cPLA2 inhibitors or PAF antagonists, like cells treated with squalestatin, released less Aβ42 oligomers and more Aβ42 monomers. In contrast to squalestatin, neither cPLA2 inhibitors, nor PAF antagonists, had a significant effect on 7PA2 cellular cholesterol concentrations, and their effects upon Aβ production were not reversed by squalene indicating that they did not affect cholesterol synthesis. However, it could be argued that Aβ production might be affected by small changes in cholesterol concentrations in specific organelles which would not be seen in total cell membrane extracts. PAF receptors were concentrated in the ER and PAF increased cholesterol concentrations within isolated ER membranes. The cholesterol concentrations in cell membranes are controlled by the cholesterol ester cycle [47] and lipid droplets, containing high concentrations of cholesterol esters are concentrated in the ER [48]. The PAF-induced increase in cholesterol in ER membranes was accompanied by a reduction in cholesterol esters; there was an inverse correlation between concentrations of cholesterol and cholesterol esters, leading to the conclusion PAF activates CEHs. The inhibition of PAF-induced changes by CEH inhibitors supported this hypothesis. Notably CM from 7PA2 cells treated with CEH inhibitors were similar to CM from 7PA2 cells treated with PAF antagonists; they contained less Aβ42 oligomers, more Aβ42 monomers and did not cause synapse damage. Each class of drug tested, squalestatin, PAF antagonists or CEH inhibitors demonstrated an increased release of Aβ42 monomers that closely correlated with the reduction in Aβ42 oligomers. A key finding was that CM from cells treated with either squalestatin, a PAF antagonist or a CEH inhibitor reduced synapse damage caused by soluble Aβ derived from a brain extract. The neuroprotective effect was mediated by Aβ monomer preparations, consistent with reports that Aβ monomers are neuroprotective [23]. It is noteworthy that thee Aβ monomer preparations may contain other APP fragments in addition to Aβ40 and Aβ42 peptides as has been reported in cerebrospinal fluid [49]. Critically these results suggest that synapse damage is dependent upon the ratio of Aβ monomers and Aβ oligomers, rather than the total Aβ concentration. We can only speculate how squalestatin, PLA2 inhibitors, PAF antagonists and CEH inhibitors alter APP metabolism and subsequently the release of Aβ monomers/oligomers. All of the drugs affected cholesterol concentrations which is important for the formation and function of lipid rafts [41]. APP and many of the enzymes involved in the generation of Aβ are found in lipid rafts [42], [43] and Aβ42 is produced within cholesterol-sensitive lipid rafts [50]. Protein cargos in lipid rafts traffic via different pathways to those in the normal cell membrane [51] and consequently APP in lipid rafts may be targeted to different cell compartments (and consequently interacts with different enzymes) than APP found in the normal cell membrane. In addition, membrane cholesterol concentrations affect the process of exocytosis [52] and may alter the release of Aβ from cells. The possibility that changes in the concentrations of Aβ42 oligomers/monomers within cell supernatants was mediated by cholesterol-sensitive changes in the release of Aβ monomers/oligomers alone appears unlikely as these drugs also reduced concentrations of Aβ42 oligomers and increased Aβ42 monomers in cell extracts.