Inhibition of amyloid-beta (Abeta) aggregation is an attractive therapeutic strategy for Alzheimer's disease (AD). Certain phenolic compounds have been reported to have anti-Abeta aggregation effects in vitro. This study systematically investigated the effects of phenolic compounds on AD model transgenic mice (Tg2576). Mice were fed five phenolic compounds (curcumin, ferulic acid, myricetin, nordihydroguaiaretic acid (NDGA), and rosmarinic acid (RA)) for 10 months from the age of 5 months. Immunohistochemically, in both the NDGA- and RA-treated groups, Abeta deposition was significantly decreased in the brain (P < 0.05). In the RA-treated group, the level of Tris-buffered saline (TBS)-soluble Abeta monomers was increased (P < 0.01), whereas that of oligomers, as probed with the A11 antibody (A11-positive oligomers), was decreased (P < 0.001). However, in the NDGA-treated group, the abundance of A11-positive oligomers was increased (P < 0.05) without any change in the levels of TBS-soluble or TBS-insoluble Abeta. In the curcumin- and myricetin-treated groups, changes in the Abeta profile were similar to those in the RA-treated group, but Abeta plaque deposition was not significantly decreased. In the ferulic acid-treated group, there was no significant difference in the Abeta profile. These results showed that oral administration of phenolic compounds prevented the development of AD pathology by affecting different Abeta aggregation pathways in vivo. Clinical trials with these compounds are necessary to confirm the anti-AD effects and safety in humans.