Synthesis and evaluation of bivalent ligands for binding to the human melanocortin-4 receptor

Abstract

Membrane proteins, especially G-protein coupled receptors (GPCRs), are interesting and important theragnostic targets since many of them serve in intracellular signaling critical for all aspects of health and disease. The potential utility of designed bivalent ligands as targeting agents for cancer diagnosis and/or therapy can be evaluated by determining their binding to the corresponding receptors. As proof of concept, GPCR cell surface proteins are shown to be targeted specifically using multivalent ligands. We designed, synthesized, and tested a series of bivalent ligands targeting the over-expressed human melanocortin 4 receptor (hMC4R) in human embryonic kidney (HEK) 293 cells. Based on our data suggesting an optimal linker length of 25±10Å inferred from the bivalent melanocyte stimulating hormone (MSH) agonist, the truncated heptapeptide, referred to as MSH(7): Ac-Ser-Nle-Glu-His-D-Phe-Arg-Trp-NH2 was used to construct a set of bivalent ligands incorporating a hMC4R antagonist, SHU9119: Ac-Nle-c[Asp-His-2'-D-Nal-Arg-Trp-Lys]-NH2 and another set of bivalent ligands containing the SHU9119 antagonist pharmacophore on both side of the optimized linkers. These two binding motifs within the bivalent constructs were conjoined by semi-rigid (Pro-Gly)3 units with or without the flexible poly(ethylene glycol) (PEGO) moieties. Lanthanide-based competitive binding assays showed bivalent ligands binds to the hMC4R with up to 240-fold higher affinity than the corresponding linked monovalent ligands.

Keywords: Bivalent ligands; Eu-binding assay; Linkers; Melanocortin receptors; Solid phase synthesis.

Figure 1

Preparation of monovalent and bivalent ligands. Reagents and conditions: (a) 1:1 or 1:4 piperidine in DMF; (b) Standard solid phase synthesis using Fmoc-chemistry; (c) PEGO attachment (Ref. 3); (d) Ac₂O/pyridine (50/50); (f) TFA/EDT/thioanisole/water (91/3/3/3).

Figure 2

Representative binding curves resulting from a typical competitive binding assay for monovalent ligand 4 (left), bivalent ligands 10 (middle) and 16 (right). Increasing concentration of ligands were competed using 10 nM Eu-α-NDP-MSH and hMC4R. Each data point represents the average of quadruplicate sample wells, with error bars indicating the standard error of means.

Figure 3. Comparison of the binding of monovalent ligand 2 and bivalent ligand 8 to hMC4R