Glucose can react nonenzymatically with amino groups of proteins to form covalent Amadori products. With time these adducts undergo further rearrangements to form irreversible advanced glycosylation endproducts (AGE), which accumulate with protein age. A specific AGE, 2-(2-furoyl)-4(5)-(2-furanyl)-1H-imidazole (FFI), has been identified on proteins in vivo. We have recently shown that a macrophage receptor specifically recognizes and internalizes proteins modified by AGE such as FFI, thus preferentially degrading senescent macromolecules. Reasoning that cellular turnover may be mediated by macrophage recognition of AGE-membrane proteins, we prepared human RBCs with FFI attached chemically. Human monocytes were incubated with either FFI-RBCs, IgG-opsonized RBCs, or PBS-treated RBCs. Erythrophagocytosis of FFI-RBCs was significantly higher than that of PBS-RBCs (55 vs. 4%; p less than 0.0025) and almost as high as that of IgG-RBCs (70%), and was competitively inhibited by AGE-BSA. AGE-RBCs were also prepared by incubating RBCs with various sugars. Human monocytes showed a 15% ingestion of glucose-RBCs, and a 26% ingestion of glucose-6-phosphate-RBCs, compared to 6% for PBS-RBCs. Similarly, diabetic mouse RBCs were phagocytosed by nearly three times more cells (21%) than normal mouse RBCs when exposed to syngeneic mouse macrophages. This phagocytosis was competitively inhibited (70%) by addition of excess AGE-BSA. The in vivo half-life of 51Cr-labeled mouse FFI-RBCs injected into syngeneic mice was reduced to 7 d, as compared to a half-life of 20 d for the controls. These data suggest that the macrophage receptor for the removal of glucose-modified proteins may also mediate the endocytosis of RBCs with AGE formed on their surface, and thus be responsible in part for the removal of some populations of aging cells.