Adiponectin is an abundant adipose-specific protein, which acts as an anti-diabetic, anti-atherogenic, and anti-inflammatory adipokine. Although recent advances in the field of adiponectin have been made by the identification of adiponectin receptors and by the understanding about relationship between its multimerization and functions, detailed molecular background remains unclear. Our established anti-human adiponectin antibodies, ANOC 9103 and ANOC 9104, blocked some adiponectin functions such as the growth inhibition of B-lymphocytes on stromal cells and the inhibition of acetylated LDL uptake in macrophages, suggesting that they may recognize important functional regions of adiponectin. As a result of epitope mapping based on the ability to bind to the deleted adiponectin mutants, we identified that these antibodies recognize amino-terminal region of adiponectin before the beginning of the collagen-like domain. Notably, a peptide fragment (DQETTTQGPGVLLPLPKGACTGWMA) corresponding to amino acid residues 17-41 of human adiponectin could bind to restricted types of cells and block adiponectin-induced cyclooxygenase-2 gene expression and prostaglandin E 2 production in MS-5 stromal cells. Moreover, the deletion of its amino-terminal region reduced the abilities to inhibit not only collagen-induced platelet aggregation but also diet-induced hepatic steatosis. These data indicate that amino-terminal region of adiponectin is a physiologically functional domain and that a novel receptor, which recognizes amino-terminal region of adiponectin, may exist on some types of cells. Further investigations will contribute to the understanding of molecular mechanisms about adiponectin functions as well as to the designing of novel strategies for the treatment of patients with insulin-resistance, vascular dysfunction, and chronic inflammation.