Cell-culture studies have revealed some of the fundamental features of the interaction of amyloid Abeta with cells and the mechanism of amyloid accumulation and pathogenesis in vitro. A(beta)1-42, the longer isoform of amyloid that is preferentially concentrated in senile plaque (SP) amyloid deposits in Alzheimer's disease (AD), is resistant to degradation and accumulates as insoluble aggregates in late endosomes or lysosomes. Once these aggregates have nucleated inside the cell, they grow by the addition of aberrantly folded APP and amyloidgenic fragments of APP, that would otherwise be degraded, onto the amyloid lattice in a fashion analogous to prion replication. This accumulation of heterogeneous aggregated APP fragments and Abeta appears to mimic the pathophysiologyof dystrophic neurites, where the same spectrum of components has been identified by immunohistochemistry. In the brain, this residue appears to be released into the extracellular space, possibly by a partially apoptotic mechanism that is restricted to the distal compartments of the neuron. Ultimately, this insoluble residue may be further digested to the protease-resistant A(beta)n-42 core, perhaps by microglia, where it accumulates as senile plaques. Thus, the dystrophic neurites are likely to be the source of the immediate precursors of amyloid in the senile plaques. This is the opposite of the commonly held view that extracellular accumulation of amyloid induces dystrophic neurites. Many of the key pathological events of AD may also be directly related to the intracellular accumulation of this insoluble amyloid. The aggregated, intracellular amyloid induces the production of reactive oxygen species (ROS) and lipid peroxidation products and ultimately results in the leakage of the lysosomal membrane. The breakdown of the lysosomal membrane may be a key pathogenic event, leading to the release of heparan sulfate and lysosomal hydrolases into the cytosol. Together, these observations provide the novel view that amyloid deposits and some of the early events of amyloid pathogenesis initiate randomly within single cells in AD. This pathogenic mechanism can explain some of the more enigmatic features of Alzheimer's pathogenesis, like the focal nature of amyloid plaques, the relationship between amyloid, dystrophic neurites and neurofibrillary-tangle pathology, and the miscompartmentalization of extracellular and cytosolic components observed in AD brain.