Radiolabeled peptides hold an important place in both diagnostic imaging and targeted therapy, owing to their high affinity for specific receptors. Among the available radioisotopes, fluorine-18 is one of the most widely used in positron emission tomography (PET): its favorable physical properties, its straightforward cyclotron production, and its half-life of approximately 110 minutes—compatible with the biodistribution kinetics of peptides—make it an excellent candidate. Since peptides are sensitive molecules, however, their radiolabeling requires mild conditions. The preferred strategy remains indirect labeling, based on a fluorine-18-labeled prosthetic group, ideally grafted via a "click" chemistry reaction offering speed, high conversion, and complete chemoselectivity.
In this context, the authors compare two fully automated methods for the fluorine-18 radiolabeling of three peptides of therapeutic interest (apelin, a C8F17 derivative of apelin, and P140). The first relies on the copper-catalyzed azide-alkyne cycloaddition (CuAAC), widely used but associated with oxidation reactions of amino acid side chains linked to the formation of reactive oxygen species. The second is based on a hetero-Diels-Alder (HDA) reaction of dithioesters (thia-Diels-Alder), previously reported by the team, without catalyst or metal.
Both approaches lead to comparable yields and synthesis times: 39% decay-corrected yield in 120 minutes for CuAAC, and 37.6% in approximately 149 minutes for HDA, with radiochemical purities greater than 95% in both cases. The differences lie elsewhere. The CuAAC route requires a substantial load of copper salts (30 equivalents per peptide-azide) as well as several additives—sodium ascorbate, aminoguanidine, and sodium gentisate—but uses only a small quantity of peptide (0.3 to 0.5 mg). By-products were observed during or after the reaction, including one attributed to the Glaser coupling of propargyl groups. Conversely, HDA, described as a minimalist method, requires only an appropriate water/ethanol ratio (70/30) and no additives, with no by-product ever having been detected; it does, however, require a higher quantity of peptide (3 mg) to achieve equivalent kinetics.
The authors highlight an advantage specific to CuAAC: the possibility of interchanging the two partners (peptide-alkyne and fluoro-azide), a flexibility that HDA does not yet offer. They nonetheless conclude that the hetero-Diels-Alder of dithioesters, despite the quantity of peptide involved, constitutes a simple and efficient approach for the fluorine-18 labeling of peptides, and an attractive alternative for accessing radiolabeled biomolecules under mild conditions.