Alzheimer disease (AD) is the most common type of dementia worldwide. Several lines of evidence suggest that aberrant production, clearance and aggregation of amyloid-β peptide (Aβ) in brain are linked to the etiology of AD. Genome-wide association studies in late-onset AD patients have reported that genetic variation in phosphatidylinositol binding clathrin assembly protein (PICALM) gene confers a genetic risk for AD. PICALM gene encodes a protein called CALM that has a PtdIns(4,5)P2-binding AP180 N-Terminal Homology (ANTH) domain at its N terminus, along with several AP2/clathrin binding motifs in the C-terminal region. CALM functions in the initial step of clathrin-mediated endocytosis through the proper formation of clathrin coated pits on the cell surface. Previously, we showed that CALM impacts on the production ratio of the pathogenic Aβ species, Aβ42 through clathrin-mediated endocytosis of γ-secretase (Kanatsu et al., Nat Commun 2014). To further elucidate the role of PICALM/CALM in the pathogenesis of AD, we crossed Picalm+/- mice with AD model mice A7, in which human APP carrying familial AD mutations is overexpressed. We found that reduction of CALM expression in A7 mouse decreased the Aβ deposition as well as brain levels of insoluble Aβ42, suggesting that CALM facilitates amyloid deposition in vivo. Intriguingly, CALM expression was relatively high at piriform cortex, where amyloid plaques were predominantly and preferentially deposited. In addition, the binding capacity of the ANTH domain to PtdIns(4,5)P2 correlated with internalization efficacy of γ-secretase and Aβ42 production ratio. These results suggest that CALM modifies AD risk by regulation of the Aβ generation and pathology.