alphaMSH has generally been accepted as the endogenous ligand for melanocortin 4 receptor (MC4R), which plays a major role in energy homeostasis. Targeting MC4R to develop antiobesity agents, many investigators have performed a structure-activity relationship (SAR) studies based on alphaMSH structure. In this report, we performed a SAR study using human betaMSH (5 - 22) (DEGPYRMEHFRWGSPPKD, peptide 1) as a lead sequence to develop potent and selective agonists for MC4R and MC3R. The SAR study was begun with a truncation of N terminus of betaMSH (5 - 22) together with acetylation of the N terminus and amidation of the C terminus of the peptide. Introduction of a cyclic disulfide constrain and replacement of L-Phe with D-Phe afforded a super potent agonist (peptide 5). Furthermore truncation at the C terminus generated a small and potent MC4R and MC3R agonist (Ac-YRcyclo[CEHdFRWC]amide, peptide 6), which exhibited no MC5R and greatly reduced MC1R activity. Molecular modeling of Ac-YRcyclo[CEHdFRWC]amide (peptide 6) revealed that Arg2 in the peptide formed a salt bridge with Glu4. Subcutaneous or intracerebroventricular administration of peptide 6 in rats showed potent in vivo efficacy as evidenced by its effects in reducing energy balance, increasing fat use, and decreasing weight gain in both acute and chronic rat metabolic studies. Furthermore, the antiobesity effect by peptide 6 was manifested only in wild-type but not MC4R-deficient mice, indicating that antiobesity effects of the peptide were attributed largely through MC4R but not MC3R agonist activity of the peptide.