A key question in understanding AD is whether extracellular Abeta deposition of parenchymal amyloid plaques or intraneuronal Abeta accumulation initiates the AD process. Amyloid precursor protein (APP) is endocytosed from the cell surface into endosomes where it is cleaved to produce soluble Abeta which is then released into the brain interstitial fluid. Intraneuronal Abeta accumulation is hypothesized to predominate from the neuronal uptake of this soluble extracellular Abeta rather than from ER/Golgi processing of APP. We demonstrate that substitution of the two adjacent histidine residues of Abeta40 results in a significant decrease in its binding with PC12 cells and mouse cortical/hippocampal neurons. These substitutions also result in a dramatic enhancement of both thioflavin-T positive fibril formation and binding to preformed Abeta fibrils while maintaining its plaque-binding ability in AD transgenic mice. Hence, alteration of the histidine domain of Abeta prevented neuronal binding and drove Abeta to enhanced fibril formation and subsequent amyloid plaque deposition--a potential mechanism for removing toxic species of Abeta. Substitution or even masking of these Abeta histidine residues might provide a new therapeutic direction for minimizing neuronal uptake and subsequent neuronal degeneration and maximizing targeting to amyloid plaques.