26RFa is a new member of the RFamide peptide family that has been identified as the endogenous ligand of the orphan GPCR GPR103. As the C-terminal heptapeptide (26RFa((20-26))) mimics the action of the native peptide on food intake and gonadotropin secretion in rodents, we have synthesized a series of analogues of 26RFa((20-26)) and measured their potency to induce [Ca(2+)](i) mobilization in Gα(16)-hGPR103-transfected CHO cells. Systematic replacement of each residue by an alanine (Ala scan) and its D-enantiomer (D scan) showed that the last three C-terminal residues were very sensitive to the substitutions while position 23 tolerated rather well both modifications. Most importantly, replacement of Ser(23) by a norvaline led to an analogue, [Nva(23)]26RFa((20-26)), that was 3-fold more potent than the native heptapeptide. These new pharmacological data, by providing the first information regarding the structure-activity relationships of 26RFa analogues, should prove useful for the rational design of potent GPR103 receptor ligands with potential therapeutic application.

26RFa, a novel RFamide neuropeptide, is the endogenous ligand of the former orphan receptor GPR103. Intracerebroventricular injection of 26RFa and its C-terminal heptapeptide, 26RFa((20-26)), stimulates food intake in rodents. To develop potent, stable ligands of GPR103 with low molecular weight, we have designed a series of aza-β(3)-containing 26RFa((20-26)) analogues for their propensity to establish intramolecular hydrogen bonds, and we have evaluated their ability to increase [Ca(2+)](i) in GPR103-transfected cells. We have identified a compound, [Cmpi(21),aza-β(3)-Hht(23)]26RFa((21-26)), which was 8-fold more potent than 26RFa((20-26)) in mobilizing [Ca(2+)](i). This pseudopeptide was more stable in serum than 26RFa((20-26)) and exerted a longer lasting orexigenic effect in mice. This study constitutes an important step toward the development of 26RFa analogues that could prove useful for the treatment of feeding disorders.

Background and Purpose The neuropeptide 26RFa and its cognate receptor GPR103 are involved in the control of food intake and bone mineralization. Here, we have tested, experimentally, the predicted ligand‐receptor interactions by site‐directed mutagenesis of GPR103 and designed point‐substituted 26RFa analogues. Experimental Approach Using the X‐ray structure of the β2‐adrenoceptor, a 3‐D molecular model of GPR103 has been built. The bioactive C‐terminal octapeptide 26RFa(19–26), KGGFSFRF‐NH2, was docked in this GPR103 model and the ligand‐receptor complex was submitted to energy minimization. Key Results In the most stable complex, the Phe‐Arg‐Phe‐NH2 part was oriented inside the receptor cavity, whereas the N‐terminal Lys residue remained outside. A strong intermolecular interaction was predicted between the Arg25 residue of 26RFa and the Gln125 residue located in the third transmembrane helix of GPR103. To confirm this interaction experimentally, we tested the ability of 26RFa and Arg‐modified 26RFa analogues to activate the wild‐type and the Q125A mutant receptors transiently expressed in CHO cells. 26RFa (10−6 M) enhanced [Ca2+]i in wild‐type GPR103‐transfected cells, but failed to increase [Ca2+]i in Q125A mutant receptor‐expressing cells. Moreover, asymmetric dimethylation of the side chain of arginine led to a 26RFa analogue, [ADMA25]26RFa(20–26), that was unable to activate the wild‐type GPR103, but antagonized 26RFa‐evoked [Ca2+]i increase. Conclusion and Implications Altogether, these data provide strong evidence for a functional interaction between the Arg25 residue of 26RFa and the Gln125 residue of GPR103 upon ligand‐receptor activation, which can be exploited for the rational design of potent GPR103 agonists and antagonists.