Recent studies on the familial Alzheimer's disease (FAD)-linked mutations in three independent genes have established the pathogenic role of beta-amyloid (Abeta) deposition as a common pathway leading to neurodegeneration. Most of these mutations seem to contribute to Abeta deposition by directly causing the overproduction of Abeta1-42, a form of Abeta with high insolubility attributed to its carboxyl-terminal structure, through secretory proteolysis. In contrast, the mechanism of Abeta deposition in sporadic Alzheimer's disease (SAD), which accounts for more than 90% of disease cases, is unclear. Because Abeta overproduction is rarely observed in SAD, a possible candidate mechanism is a decreased degradation, or dyscatabolism, of Abeta. It is notable that a reduction in catabolism of only 30-50% is estimated to exert an equivalent effect on Abeta metabolism as the overproduction seen in FAD. Identification of the in vivo catabolic processes responsible for Abeta disposition would provide a new basis for the development of preventive and therapeutic measures against the disease. I hypothesized recently that aminopeptidase-catalyzed proteolysis of Abeta may limit the rate of Abeta catabolism and that the reduction of a certain aminopeptidase activity would lead to Abeta dyscatabolism and thus to deposition (Aminopeptidase Hypothesis), based on the structural properties of Abeta deposited in human brain. Experimental and clinical observations supporting this hypothesis are accumulating although further work is necessary to fully evaluate its relevance. If the assumption proves to be true, both the familial and sporadic forms of AD may be referred to as "proteolytic disorders" in anabolic and catabolic terms, respectively.